by
B.S. Chemistry and Mathematics (2006)
Stanford University
Submitted to the Department of Biology
In Partial Fulfillment of the Requirements for the Degree of
Doctor in Philosophy in Biology at the
Massachusetts Institute of Technology
May 2011
!
2011 Massachusetts Institute of Technology
All rights reserved.
Signature of Author………………………………………………………………………………..
Arvind Ravi
Biology
May 20, 2011
Certified by…………………………………………………………………………………………
Phillip A. Sharp
Institute Professor of Biology
Thesis Supervisor
Accepted by……………………………………………………………………………….............
Robert Sauer
Professor of Biology
Chair, Biology Graduate Committee
1

by
MicroRNAs (miRNAs) constitute a class of ~22 nucleotide RNAs with broad regulatory roles in gene expression. Dependent largely on the enzyme Dicer for their generation from longer precursor transcripts, mammalian miRNAs direct posttranscriptional repression of mRNAs based on complementarity to sites in their 3´ UTRs.
To better understand how these regulators impact fundamental processes such as development and cancer, we explored the functional consequences of Dicer loss and subsequent miRNA loss across a range of embryonic and somatic tissues.
In embryonic stem (ES) cells, we identified a latent susceptibility to genotoxic stress following deletion of Dicer. Re-expression of the abundant miR-290-295 cluster or knockdown of two novel targets, Caspase 2 and Ei24, could partially restore cell survival after DNA damage, implicating them as important players in a larger stress-responsive
ES network under miRNA control.
To better understand changes in miRNA and target repertoires at a global scale, we applied a novel evolutionary analysis to the mouse genome designed to speciesspecific innovations. Using this method, we uncovered the genome-wide signature of miRNAs functionally related to the miR-290-295 cluster that we term the Sfmbt2 cluster.
In addition to ES cells, placental tissues express these miRNAs at high levels suggesting that mice have co-opted an existing proliferative network to support rapid placental growth.
Finally, we evaluated Dicer loss in two transformed somatic cell types, namely sarcoma cells and mesenchymal stem cells. Surprisingly, these cells tolerated Dicer deletion without loss of viability and retained several properties of their Dicer intact counterparts including surface marker expression and tumorigenicity. Comparison of expression data in these cells and ES cells revealed that while many miRNA targets show relatively little change before and after Dicer loss, a subset of genes that differ between embryonic and somatic cells may be controlled in large part by cell type specific miRNAs.
In summary, these data shed light on many fundamental aspects of miRNA function in mammalian cells, expanding our understanding of molecular targets as well their downstream cellular roles. As our knowledge about short RNA regulation grows, we are sure to continue uncovering important connections between post-transcriptional regulation and the underlying biology of human development and disease.
Thesis Supervisor:
Phillip A. Sharp, Institute Professor of Biology
2

Abstract ................................................................................................................2
Table of Contents ................................................................................................3
Acknowledgments ...............................................................................................4
CHAPTER 1: Introduction ...................................................................................6
Discovery of miRNAs .........................................................................................8
MiRNA biogenesis and the role of Dicer ............................................................9
Mechanisms of miRNA function.......................................................................11
Evolution of miRNAs and targets .....................................................................13
MiRNAs and cell identity ..................................................................................15
MiRNA regulation of the embryonic stem cell state .........................................16
MiRNAs and cancer .........................................................................................19
CHAPTER 2: A Latent Pro-survival Function for the Mir-290-295 Cluster in
Mouse Embryonic Stem Cells ..........................................................................37
CHAPTER 3: Genome-Wide Impact of a Novel Rapidly Expanded MicroRNA
Cluster in Mouse................................................................................................90
CHAPTER 4: Viability of Transformed Somatic Cells in the Absence of Dicer
...........................................................................................................................155
CHAPTER 5: The MicroRNA Landscape of the Somatic Mesenchymal State
...........................................................................................................................181
CHAPTER 6: Conclusions...............................................................................232
3

First and foremost, to Phil for giving me the incredible opportunity to work in your lab. Your guidance has been invaluable to my growth as a scientist and person, and I am incredibly fortunate to count you as a mentor.
To my committee members, Professors David Bartel and Rudolf Jaenisch, for your countless suggestions and helpful direction throughout the course of this work.
To my collaborators and friends in the Burge and Jacks labs, who have helped me achieve a balance between the thinking and doing required to make progress in research.
To the entire Sharp lab, who have made it a truly wonderful place to do science, and in particular: to Grace, for your generosity and mentorship during my first few years as a molecular biologist, to Allan, for your willingness to discuss an idea any time day or night, and for being a true colleague and friend throughout my graduate work, and to Andrew, for helping me keep a healthy perspective on work and life, and reminding me from time to time that there is a world outside the lab.
I would also like to thank the Hertz Foundation for supporting my graduate work, and for connecting me to a community of talented and creative individuals whom
I am honored to call friends.
Finally, to my family, who have worked so hard to give me the incredible opportunities I have before me.
4

To my family
5

6

Short RNAs provide a critical mode of gene expression control in a wide variety of organisms. In this introduction to regulation by short RNAs and specifically microRNAs (miRNAs), we examine some of the core principles of their structure, biogenesis, and function. In addition, we explore their numerous roles in mammalian physiology and disease, including reinforcement of cell identity and control of oncogenic transformation.
7

The earliest miRNAs were identified through genetic screens in C. elegans to identify regulators of developmental timing (Lee et al., 1993; Wightman et al.,
1993). In these studies, the gene lin-4 was necessary for development beyond an early larval stage (L1), yet closer examination of its genomic sequence revealed the absence of an open reading frame. Taken together with the complementarity between the RNA products of the lin-4 locus and its negatively regulated protein target, lin-14 , these studies suggested a novel model in which antisense RNA-RNA interactions guide repression of a gene. While numerous subsequent studies have broadened and refined this initial picture of miRNA regulation, the interaction between regulatory elements at the RNA level remains a uniting principle of this field.
After these initial characterizations, further studies began to generalize these observations, suggesting that miRNA regulation was in fact more extensive than previously thought. A second noncoding RNA gene, let-7 , also involved in the regulation of developmental timing, was identified through further genetic screens (Reinhart et al., 2000) and found to be well conserved across a number of species (Pasquinelli et al., 2000).
Shortly thereafter, biochemical approaches helped establish these genes as members of a broader class. In particular, use of a novel cloning strategy specific to the end chemistry of miRNAs (i.e., a 5 ´ terminal phosphate and a 3 ´ terminal hydroxyl group) enabled simultaneous discovery of many novel miRNAs in relatively short order, with conservation in some cases extending to a number
8

of other species, including humans (Lagos-Quintana et al., 2001; Lau et al.,
2001; Lee and Ambros, 2001).
Since these early discoveries, a number of studies have attempted to build comprehensive miRNA catalogues, often using a combination of sequencing and informatic approaches. As sequencing technology has advanced, profiling datasets have increased in depth allowing identification of lower abundance species. For instance, initial studies with use of massively parallel signature sequencing (MPSS) technology allowed the detection of 390 miRNA species from different stages of embryonic development in mice (Mineno et al., 2006).
More recently, similar data has been revisited with greater library complexity and tissue coverage, improving this count to over 500 miRNAs (Chiang et al., 2010).
Further decreases in the cost of sequencing and increases in the diversity of tissues profiled will likely continue to drive miRNA discovery.
Despite the diversity of miRNA genes known, the processing steps required are largely common to most miRNAs (Kim, 2005; Winter et al., 2009).
First, a primary miRNA transcript (pri-miRNA) is transcribed by PolII. This initial transcript can by polycistronic, capable of generating multiple independent miRNAs through subsequent processing steps. After transcription the pri-miRNA undergoes a first cleavage step by the RNAse III enzyme Drosha (Lee et al.,
2003), which recognizes a roughly 33 bp double-stranded stem as well as the
9

adjacent transition from double- to single-stranded RNA in complex with Dgcr8
(Han et al., 2006).
The resulting sequence, known as the pre-miRNA, is exported from the nucleus by Exportin-5 (Lund et al., 2004; Yi et al., 2003), after which it undergoes additional processing in the cytoplasm by Dicer, the final enzyme required for the maturation of nearly all mammalian miRNAs (Bernstein et al., 2001; Grishok et al., 2001; Hutvagner et al., 2001; Ketting et al., 2001). This cleavage generates a characteristic roughly 22-nucleotide RNA duplex with 2 nucleotide overhangs at the 3 ´ ends. Dicer works in complex with other proteins including Tar RNA
Binding Protein (TRBP) and Protein Activator of PKR (PACT), which promote efficient pre-miRNA cleavage but are not required (Chendrimada et al., 2005;
Gregory et al., 2005; Haase et al., 2005).
Following cleavage, the RNA duplex is unwound to promote loading of the mature miRNA strand, named the guide strand, into the RNA-induced Silencing
Complex (RISC), with strand selection driven by thermodynamic instability at the
5´ end (Khvorova et al., 2003; Schwarz et al., 2003). The guide strand is loaded directly into one of the Argonaute (Ago) family proteins, Ago1-4, which are the key effectors of miRNA-mediated gene silencing (Hutvagner and Zamore, 2002;
Liu et al., 2004; Meister et al., 2004; Pillai et al., 2004).
While most miRNAs follow these canonical processing steps, there are notable exceptions. A class of miRNAs known as mirtrons do not undergo
Drosha cleavage, instead generating pre-miRNAs directly from a splicing event
(Babiarz et al., 2008; Berezikov et al., 2007; Okamura et al., 2007; Ruby et al.,
10

2007). In addition, miR-451 has recently been reported to be Dicer independent, retaining sufficient complementarity in the stem loop to undergo cleavage by
Argonaute 2, a downstream effector of miRNA mediated posttranscriptional repression (Cheloufi et al., 2010; Cifuentes et al., 2010).
Once a mature miRNA has been generated and loaded into the RISC complex, it is capable of directing repression of potential mRNA targets.
Targeting by miRNAs is largely dictated by bases 2-7 of the mature miRNA, known as the miRNA seed, which have complementarity to target sites in the 3 ´
UTRs of transcripts (Brennecke et al., 2005; Doench and Sharp, 2004;
Kloosterman et al., 2004; Lai et al., 2005; Lewis et al., 2003; Lewis et al., 2005).
Such targeting allows a single miRNA sequence to theoretically have hundreds of genomic targets, a prediction confirmed by direct transfection or knockout studies followed by global expression profiling (Giraldez et al., 2006; Lim et al.,
2005; Rodriguez et al., 2007).
Although seed match complementarity appears to be the dominant mode of target identification, there exist alternative means by which targets can be recognized. In some cases, additional pairing at the 3 ´ end (usually with bases
13-16 of the mature miRNA) can compensate for mismatches in the seed region, or can augment seed matches (Grimson et al., 2007). Another model of potentially non-seed matches involves the recently reported centered site pairing, which features pairing to positions 4-15 of the mature miRNA (Shin et al., 2010).
11

Finally, sites with near perfect complementarity between a miRNA and its corresponding target can direct Argonaute 2 mediated target cleavage opposite positions 10 and 11 of the mature miRNA, an exception to the more commonly observed mechanisms of destabilization and translational repression (described below) (Elbashir et al., 2001; Hutvagner and Zamore, 2002; Yekta et al., 2004).
However, the frequency of these additional site types appears to be relatively low compared to canonical seed targeting (Bartel, 2009).
Following identification of an mRNA target by a loaded Argonaute protein,
(provided there is not extensive complementarity between the RNA sequences) there are two basic models for how repression is achieved: 1) destabilization of the mRNA transcript, and 2) inhibition of translation. A number of studies evaluating either specific targets in vitro or employing more global profiling approaches have lent support to both of these models.
Data from cell culture systems examining endogenous miRNA targets as well as artificial systems with miRNA reporters have suggested that in some cases notable repression occurs at the translational step, with relatively less repression at the level of the mRNA target (Behm-Ansmant et al., 2006;
Brennecke et al., 2003; Petersen et al., 2006; Pillai et al., 2005). In particular, several studies point to repression at the pre-initiation stage, i.e., prior to ribosome assembly and productive translation (Bhattacharyya et al., 2006; Ding and Grosshans, 2009; Pillai et al., 2005).
Other studies point towards repression being mediated at the level of mRNA stability. Studies of miRNA reporter constructs have shown loss of gene
12

expression concordant with loss of mRNA abundance (Behm-Ansmant et al.,
2006; Eulalio et al., 2009). The mechanism of this destabilization involves the recruitment of both deadenylases CCR4:NOT1 and decapping enzymes
DCP1:DCP2, in part through mediators such as GW182 (Behm-Ansmant et al.,
2006; Fabian et al., 2009; Zekri et al., 2009).
Recent experiments at the level of global gene expression have also attempted to discern the relative extent of mRNA destabilization vs. translational repression. Using combined quantitative mass spectrometry and mRNA expression arrays, two groups have attempted to quantify the relative repression at each of these levels. Though in both cases, they observe that repression is often detectable at the mRNA level, Selbach et al report a more prominent role for translational repression in several target genes (Baek et al., 2008; Selbach et al., 2008). However, further experiments using ribosome profiling suggest that for most genes, the majority of decreased protein expression can be detected at the mRNA level (Guo et al., 2010).
Given the nucleotide complementarity required between miRNAs and target transcripts, direct analysis of genomic sequences has provided a rich resource for examining the conservation of miRNA targeting across different lineages. For a single highly conserved miRNA, the number of corresponding conserved targets can be quite extensive, averaging roughly 400 genes in mammals (Friedman et al., 2009). Considering all miRNA targets combined,
13

over half of mammalian proteins are predicted to contain conserved miRNA target sites (Friedman et al., 2009).
While the sequence of a miRNA can be conserved to a large degree over evolutionary time, the exact network elements being targeted can be relatively dynamic. For instance, the highly conserved miR-1 targets the transcription factor Hand-2 in mammals in order to promote cardiac differentiation (Zhao et al.,
2005). While its homolog in Drosophila differs by only 2 nucleotides and is induced by similar transcription factors (Sokol and Ambros, 2005), it no longer targets Hand-2, instead mediated differentiation through downregulation of the
Notch ligand, Delta (Kwon et al., 2005).
Another example of network level conservation involves the miR-295 family (described in more detail below). Members of the miR-295 family, along with its vertebrate orthologs miR-430/427, target multiple members of the Nodal pathway (Rosa et al., 2009). However, the particular elements targeted appear to be lineage-specific, suggesting that although this miRNA family retains a role in vertebrate germ layer specification, its precise regulatory network, including both miRNAs and targets, has been dynamic across evolution. Given that many miRNAs appear to act on related elements of gene networks (Tsang et al., 2010), it is likely that many more cases of network rewiring will be revealed as this area progresses.
14

As is the case with transcription factors, miRNAs have a variety of expression patterns, from those that are ubiquitously expressed to those with cell type specificity. On the whole, it appears that miRNAs may be more cell-type specific than protein coding genes, as they are superior classifiers of cell identity
(Lu et al., 2005). Indeed, careful examination of miRNA conservation from the earliest bilaterian species has suggested a close link between miRNA evolution and cell identity, particularly in tissues from the central nervous system, sensory organs, muscle, and gut (Christodoulou et al., 2010).
Global expression studies comparing mRNA and miRNA expression across cell types or following transfection of exogenous miRNAs support the notion that miRNA regulation can reinforce cell identity. In the case of miR-124 and miR-1, two highly tissue-specific miRNAs expressed in neurons and muscle, respectively, transfection into HeLa cells led to widespread expression changes towards the profile of the tissue in which the miRNAs are normally found (Lim et al., 2005). In particular, these changes are evident among genes expressed at low levels in the miRNA-containing tissue, suggesting that their repression is in part direct miRNA-mediated.
In support of this finding, genes that are highly coexpressed with a specific miRNA show a strong depletion of miRNA target sites in their 3 ´ UTRs, a phenomenon termed target avoidance (Farh et al., 2005; Stark et al., 2005). The mRNAs that do have targets for these tissue-specific miRNAs are most likely to be expressed in adjacent tissues (Stark et al., 2005). These findings suggest a
15

model by which miRNAs can promote developmental decisions by more precisely delineating gene expression states from neighboring cell states.
Like a number of other tissues, embryonic stem (ES) cells express certain cell type specific miRNAs, such as the miR-290-295 cluster in mice (or its human homolog, the miR-371-373 cluster) (Houbaviy et al., 2003; Suh et al., 2004).
Sequence analysis suggests that this cluster is specific to placental mammals, as clear homologs could not be found in marsupial and more distant genomes.
Although not specific to ES cells, the miR-17-92 cluster, which contains a similar seed sequence with a one base shift, is also observed to decrease upon differentiation of ES cells, suggesting that it too may contribute to the ES state
(Chen et al., 2007). Notably, expression of the miR-290-295 cluster is also observed in primordial germ cells and spermatogonia, leading to the notion that these miRNAs may more generally characterize a highly proliferative developmental state (Hayashi et al., 2008).
Roles of the miR-290-295 cluster
Consistent with their high embryonic expression, miR-295 family miRNAs regulate some of the core features of ES cell physiology. For instance, these miRNAs directly target Rbl2, a negative regulator of Dnmt expression, including de novo methyltransferases Dnmt3a and Dnmt3b (Benetti et al., 2008; Sinkkonen et al., 2008). This decrease in the capacity for de novo methylation correlates
16

with an inability to stably repress core pluripotency genes upon differentiation, although is likely to cooperate with other perturbations given the viability of
Dnmt3 deficiency. Thus, the miR-295 family may have an important role in maintaining the ability of pluripotent cells to transition into somatic lineages.
Another important role for the miR-290-295 cluster is in facilitating rapid proliferation, a hallmark of the ES state. Upon global miRNA loss following Dicer or Dgcr8 deletion, ES cells undergo a marked decrease in proliferation rate with a concomitant increase in the G1 cell population (Kanellopoulou et al., 2005;
Murchison et al., 2005; Wang et al., 2007). In a screen of miRNAs that could rescue this defect, the miR-295 and miR-17 families were identified, capable of restoring cells towards wild type cycling (Wang et al., 2008). The cell cycle regulator p21 is directly targeted by miR-295, and knockdown of this target alone is able to partially rescue cell growth, although additional targets are likely to contribute.
Finally, members of the miR-295 network appear to be intimately linked to the core ES identity. First, the central pluripotency factors Oct4, Sox2, Nanog, and Tcf3 drive transcription of the cluster at the level of its promoter (Marson et al., 2008). In addition, re-expression of miR-295 in miRNA deficient ES cells is able to restore expression of the core ES gene Lin28 (described further below)
(Melton et al., 2010). Finally, reprogramming of mouse embryonic fibroblasts
(MEFs) to induced pluripotent stem cells (iPS) is enhanced by transfection of miR-295 family miRNAs (Judson et al., 2009), and in fact, can be achieved by ES miRNAs alone (Anokye-Danso et al., 2011; Subramanyam et al., 2011).
17

Importantly, given the expression of other miRNAs with identical as well as related seeds in ES cells, it is likely that some of these identified roles are redundant with those of other miRNAs in this cell type. Therefore, these attributed functions may be more generally characterized as those of ‘AAGUGC’ and related seeds, rather than strictly the miR-290-295 cluster alone.
Blockade of let-7 in the embryonic state
In maintaining the ES identity, it appears that the miR-295 family acts in opposition to the let-7 family, which characterizes the somatic state. A key mediator of this shift upon differentiation is Lin28, which can inhibit production of mature let-7 (Viswanathan et al., 2008), and has also been shown to promote reprogramming (Yu et al., 2007). The blockade of let-7 production is mediated by recruitment of a terminal uridylyl transferase (TUT4/zcchc11) that leads to poly
U-tailing and subsequent degradation of pre-let-7 (Hagan et al., 2009; Heo et al.,
2009). As added support for the antagonism between let-7 and the ES state, inhibition of let-7 in MEFs leads to improved rates of reprogramming (Melton et al., 2010). Thus, miRNA maintenance of the ES state depends upon both high levels of miR-290-295 expression and low levels of let-7 (Figure 1).
The sequence of let-7 as well as its upregulation in later developmental stages is a conserved feature across many organisms (Lagos-Quintana et al.,
2003; Pasquinelli et al., 2003; Pasquinelli et al., 2000). Consistent with this observation, many of the conserved let-7 targets have important functions in the embryonic state. One of the earliest targets identified for let-7 was the C. elegans
18

gene lin-41 (Reinhart et al., 2000), a relationship that has been preserved in mice, where lin-41 acts as a stem-cell specific ubiquitin ligase (Rybak et al.,
2009). Another example of a shared target across species are Ras family proteins, which have essential roles in driving proliferation in early development
(Koera et al., 1997), and which are targeted by let-7 from worms to humans
(Johnson et al., 2005). These and other examples pinpoint let-7 as a key switch in shutting off the embryonic state (Roush and Slack, 2009).
Given the extensive networks regulated by miRNAs, it is perhaps not surprising that they have diverse roles in the initiation and maintenance of cancer. Depending on their specific cellular targets, miRNAs can act as both tumor suppressors and oncogenes, much like traditional protein-coding genes
(summarized in Table 1). In addition, mutations that alter the ability of miRNAs to target particular transcripts can also modulate tumor formation.
MiRNAs as oncogenes
Among the set of oncogenic miRNAs is miR-155, originally found to be over-expressed in certain types of lymphomas (Kluiver et al., 2005; Metzler et al.,
2004). Expression of Bic, the noncoding RNA in which miR-155 resides, was shown to cooperate with c-Myc expression in promoting growth of chicken embryo fibroblasts (Tam et al., 2002). Later, transgenic models demonstrated
19

that overexpression of miR-155 in B cells induced excessive pre-B cell proliferation culminating in lymphomagenesis (Costinean et al., 2006).
A second miRNA that is often overexpressed in various cancers is miR-
21, found across a range of hematological as well as solid tumors (Volinia et al.,
2006). A negative regulator of the tumor suppressors Pten and Pdcd4 (Frankel et al., 2008; Meng et al., 2007), miR-21 can prevent apoptosis in human glioblastoma cells (Chan et al., 2005). Conversely, loss of miR-21 mediated target repression has the opposing effect, leading to caspase activation and induction of cell death (Chan et al., 2005; Meng et al., 2007).
Finally, numerous studies have identified overexpression of the miR-17-92 cluster as a common feature of many human cancers (van Haaften and Agami,
2010). Expression of the cluster is driven by c-Myc (O'Donnell et al., 2005), and members of the cluster are known to target a number of tumor suppressors, including Bim, Pten, and p21 (Petrocca et al., 2008; Ventura et al., 2008; Xiao et al., 2008). The related miRNAs, miR-372 and miR-373, which contain a shifted seed relative the miR-17-92 cluster, have also been implicated as oncogenic via their inhibition of the tumor suppressor Lats2 (Voorhoeve et al., 2007).
MiRNAs as tumor suppressors
In addition to promoting transformation, a number of miRNAs have been predicted or demonstrated to prevent tumor growth. One miRNA family classically thought to act as a tumor suppressor is let-7, which shows widespread downregulation across a number of tumor types (Johnson et al., 2005). Given the
20

finding that cancers recapitulate aspects of the embryonic state (Ben-Porath et al., 2008), this role for let-7 fits well with its reinforcement of the somatic state described earlier. In addition to directly targeting oncogenes such as Ras,
Hmga2, c-Myc, and Igf2bp1 (Johnson et al., 2005; Lee and Dutta, 2007; Mayr et al. 2007; Mayr and Bartel, 2009; Sampson et al., 2007), let-7 has been shown in mouse models to inhibit tumor growth in vivo (Esquela-Kerscher et al., 2008;
Kumar et al., 2008).
Another miRNA family with tumor-suppressive roles is the miR-15a/16-1 cluster, found in a genomic region frequently deleted in chronic lymphocytic leukemia (Calin et al., 2002). Given that it targets Bcl-2, an important pro-survival gene, it is thought that loss of this cluster represents a key step in oncogenesis
(Cimmino et al., 2005).
Finally, expression of miR-29 family members appears to be decreased in a variety of tumor types (Garzon et al., 2008; Mott et al., 2007), and is also found in a genomic region that frequently undergoes loss in specific hematological malignancies such as acute myeloid leukemia (Pedersen-Bjergaard et al., 1995).
Both in vitro and in vivo studies suggest that miR-29 re-expression in cancer cells can lead to direct targeting of pro-survival genes such as Mcl-1 and Tcl-1, acting through these and other targets to induce apoptosis (Fabbri et al., 2007; Mott et al., 2007).
21

MiRNA target site mutations in cancer
Just as mutations causing gain or loss of miRNAs can lead to signaling dysregulation and ultimately cancer, so can mutations in the target sites of specific genes. For instance, translocations of Hmga2 have been described that abolish the multiple let-7 target sites in its 3 ´ UTR, leading to loss of repression and promotion of tumorigenic properties (Mayr et al., 2007). A more subtle variation in the 3 ´ UTR of Kras, namely a SNP interfering with let-7 repression, has been linked to a 2-fold increased risk of non-small cell lung cancer (Chin et al., 2008). In addition, mutations introducing novel sites can lead to aberrant targeting of tumor suppressors. One example of such a model is Mdm4 targeting by miR-191 due to introduction of a novel target site from a rare SNP, which has been linked to enhanced progression and decreased chemosensitivity in human ovarian cancer (Wynendaele et al., 2010).
Role of global miRNA levels and Dicer in oncogenic transformation
Although individual miRNAs can act as tumor suppressors or oncogenes based on their cellular targets, human cancers appear to feature a global downregulation of miRNA levels (Gaur et al., 2007; Lu et al., 2005). In part, this reduction may be a result of c-Myc driven transcriptional repression, as chromatin immunoprecipitation experiments have revealed widespread binding of
Myc at miRNA promoters (Chang et al., 2008). Notably, enforced expression of these repressed miRNAs inhibits tumor development, suggesting a causal role for reduced miRNA expression in oncogenesis (Chang et al., 2008).
22

In addition to transcriptional repression, processing defects likely contribute to reduced miRNA levels. Originally suggested by the observation that some mature miRNA levels are reduced despite little expression change in their primary transcripts (Thomson et al., 2006), instances of defective processing have since been identified, such as mutations in Tarbp2 that lead to destabilization of its binding partner Dicer, and thus defective miRNA processing
(Melo et al., 2009).
To more precisely address the role of global miRNA levels in transformation, studies have directly manipulated expression of Dicer and studied subsequent effects on tumor formation and growth. Short hairpins directed against Dicer as well as Drosha and Dgcr8 induced more rapid tumor growth and a more invasive phenotype in mouse lung adenocarcinoma cells upon subcutaneous injection (Kumar et al., 2007). Furthermore, loss of a single copy of Dicer upon oncogene activation promoted tumor formation in both lung adenocarcinoma and soft tissue sarcoma models (Kumar et al., 2007; Kumar et al., 2009). However, such enhancement of oncogenesis was not observed in an
E !
-myc model of lymphomagenesis, as heterozygous Dicer loss showed led to similar tumor kinetics as homozygous wild type Dicer controls (Arrate et al.,
2010). Interestingly, in all three of these models, complete Dicer loss appeared to be strongly disfavored, as Dicer null tumor cells could not be recovered (Arrate et al., 2010; Kumar et al., 2009).
The nature of heterozygous loss being favored but homozygous loss being disfavored is consistent with data gathered from human cancers. Examination of
23

Dicer copy number from sequencing data of patient samples identified numerous cases of hemizygous loss, but failed to identify examples of homozygous loss
(Kumar et al., 2009). In addition, in a rare familial pediatric lung tumor known as pleuropulmonary blastoma in which patients have a germline mutation inactivating one allele of Dicer, immunohistochemistry from tumors suggests that
Dicer expression is retained in cancer tissue (Hill et al., 2009). Thus, these data point to a complex interaction between Dicer gene dosage, miRNA levels, and tumor propensity wherein partial decreases are tolerated and/or favored, while global loss is detrimental.
24

MicroRNA expression of embryonic stem (ES) cells is dominated by the miR-295 family of miRNAs, which direct a number of functions essential to the
ES state, including proper regulation of methylation and rapid proliferation
(Benetti et al., 2008; Sinkkonen et al., 2008; Wang et al., 2008). The transition to the differentiated state is marked by a significant downregulation of the miR-290-
295 cluster, and concomitant upregulation of members of the let-7 family of miRNAs.
25

Depending on the specific genes targeted, miRNAs can act to either promote or inhibit tumor formation. Abbreviations: CLL, chronic lymphocytic leukemia; AML, acute myeloid leukemia; DLBCL, diffuse large B cell lymphoma;
FLT3-ITD, FMS-like tyrosine kinase 3 in tandem duplication mutations; BL,
Burkitt lymphoma; TS, tumor suppressor; OG, oncogene. Adapted from Garzon et al (Garzon et al., 2009).
26

Anokye-Danso, F., Trivedi, C. M., Juhr, D., Gupta, M., Cui, Z., Tian, Y., Zhang,
Y., Yang, W., Gruber, P. J., Epstein, J. A., and Morrisey, E. E. (2011).
Highly Efficient miRNA-Mediated Reprogramming of Mouse and Human
Somatic Cells to Pluripotency. Cell Stem Cell 8 , 376-388.
Arrate, M. P., Vincent, T., Odvody, J., Kar, R., Jones, S. N., and Eischen, C. M.
(2010). MicroRNA biogenesis is required for Myc-induced B-cell lymphoma development and survival. Cancer Res 70 , 6083-6092.
Babiarz, J. E., Ruby, J. G., Wang, Y., Bartel, D. P., and Blelloch, R. (2008).
Mouse ES cells express endogenous shRNAs, siRNAs, and other
Microprocessor-independent, Dicer-dependent small RNAs. Genes Dev
22 , 2773-2785.
Baek, D., Villen, J., Shin, C., Camargo, F. D., Gygi, S. P., and Bartel, D. P.
(2008). The impact of microRNAs on protein output. Nature 455 , 64-71.
Bartel, D. P. (2009). MicroRNAs: target recognition and regulatory functions. Cell
136 , 215-233.
Behm-Ansmant, I., Rehwinkel, J., Doerks, T., Stark, A., Bork, P., and Izaurralde,
E. (2006). mRNA degradation by miRNAs and GW182 requires both
CCR4:NOT deadenylase and DCP1:DCP2 decapping complexes. Genes
Dev 20 , 1885-1898.
Ben-Porath, I., Thomson, M. W., Carey, V. J., Ge, R., Bell, G. W., Regev, A., and
Weinberg, R. A. (2008). An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet 40 ,
499-507.
Benetti, R., Gonzalo, S., Jaco, I., Munoz, P., Gonzalez, S., Schoeftner, S.,
Murchison, E., Andl, T., Chen, T., Klatt, P.
, et al.
(2008). A mammalian microRNA cluster controls DNA methylation and telomere recombination via Rbl2-dependent regulation of DNA methyltransferases. Nat Struct Mol
Biol 15 , 998.
Berezikov, E., Chung, W. J., Willis, J., Cuppen, E., and Lai, E. C. (2007).
Mammalian mirtron genes. Mol Cell 28 , 328-336.
Bernstein, E., Caudy, A. A., Hammond, S. M., and Hannon, G. J. (2001). Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature
409 , 363-366.
Bhattacharyya, S. N., Habermacher, R., Martine, U., Closs, E. I., and Filipowicz,
W. (2006). Relief of microRNA-mediated translational repression in human cells subjected to stress. Cell 125 , 1111-1124.
Brennecke, J., Hipfner, D. R., Stark, A., Russell, R. B., and Cohen, S. M. (2003). bantam encodes a developmentally regulated microRNA that controls cell proliferation and regulates the proapoptotic gene hid in Drosophila. Cell
113 , 25-36.
Brennecke, J., Stark, A., Russell, R. B., and Cohen, S. M. (2005). Principles of microRNA-target recognition. PLoS Biol 3 , e85.
27

Calin, G. A., Dumitru, C. D., Shimizu, M., Bichi, R., Zupo, S., Noch, E., Aldler, H.,
Rattan, S., Keating, M., Rai, K.
, et al.
(2002). Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A 99 , 15524-15529.
Chan, J. A., Krichevsky, A. M., and Kosik, K. S. (2005). MicroRNA-21 is an antiapoptotic factor in human glioblastoma cells. Cancer Res 65 , 6029-
6033.
Chang, T. C., Yu, D., Lee, Y. S., Wentzel, E. A., Arking, D. E., West, K. M., Dang,
C. V., Thomas-Tikhonenko, A., and Mendell, J. T. (2008). Widespread microRNA repression by Myc contributes to tumorigenesis. Nat Genet 40 ,
43-50.
Cheloufi, S., Dos Santos, C. O., Chong, M. M., and Hannon, G. J. (2010). A dicer-independent miRNA biogenesis pathway that requires Ago catalysis.
Nature 465 , 584-589.
Chen, C., Ridzon, D., Lee, C. T., Blake, J., Sun, Y., and Strauss, W. M. (2007).
Defining embryonic stem cell identity using differentiation-related microRNAs and their potential targets. Mamm Genome 18 , 316-327.
Chendrimada, T. P., Gregory, R. I., Kumaraswamy, E., Norman, J., Cooch, N.,
Nishikura, K., and Shiekhattar, R. (2005). TRBP recruits the Dicer complex to Ago2 for microRNA processing and gene silencing. Nature
436 , 740-744.
Chiang, H. R., Schoenfeld, L. W., Ruby, J. G., Auyeung, V. C., Spies, N., Baek,
D., Johnston, W. K., Russ, C., Luo, S., Babiarz, J. E.
, et al.
(2010).
Mammalian microRNAs: experimental evaluation of novel and previously annotated genes. Genes Dev 24 , 992-1009.
Chin, L. J., Ratner, E., Leng, S., Zhai, R., Nallur, S., Babar, I., Muller, R. U.,
Straka, E., Su, L., Burki, E. A.
, et al.
(2008). A SNP in a let-7 microRNA complementary site in the KRAS 3' untranslated region increases nonsmall cell lung cancer risk. Cancer Res 68 , 8535-8540.
Christodoulou, F., Raible, F., Tomer, R., Simakov, O., Trachana, K., Klaus, S.,
Snyman, H., Hannon, G. J., Bork, P., and Arendt, D. (2010). Ancient animal microRNAs and the evolution of tissue identity. Nature 463 , 1084-
1088.
Cifuentes, D., Xue, H., Taylor, D. W., Patnode, H., Mishima, Y., Cheloufi, S., Ma,
E., Mane, S., Hannon, G. J., Lawson, N. D.
, et al.
(2010). A novel miRNA processing pathway independent of Dicer requires Argonaute2 catalytic activity. Science 328 , 1694-1698.
Cimmino, A., Calin, G. A., Fabbri, M., Iorio, M. V., Ferracin, M., Shimizu, M.,
Wojcik, S. E., Aqeilan, R. I., Zupo, S., Dono, M.
, et al.
(2005). miR-15 and miR-16 induce apoptosis by targeting BCL2. Proc Natl Acad Sci U S A
102 , 13944-13949.
Costinean, S., Zanesi, N., Pekarsky, Y., Tili, E., Volinia, S., Heerema, N., and
Croce, C. M. (2006). Pre-B cell proliferation and lymphoblastic leukemia/high-grade lymphoma in E(mu)-miR155 transgenic mice. Proc
Natl Acad Sci U S A 103 , 7024-7029.
28

Ding, X. C., and Grosshans, H. (2009). Repression of C. elegans microRNA targets at the initiation level of translation requires GW182 proteins.
EMBO J 28 , 213-222.
Doench, J. G., and Sharp, P. A. (2004). Specificity of microRNA target selection in translational repression. Genes Dev 18 , 504-511.
Elbashir, S. M., Lendeckel, W., and Tuschl, T. (2001). RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev 15 , 188-200.
Esquela-Kerscher, A., Trang, P., Wiggins, J. F., Patrawala, L., Cheng, A., Ford,
L., Weidhaas, J. B., Brown, D., Bader, A. G., and Slack, F. J. (2008). The let-7 microRNA reduces tumor growth in mouse models of lung cancer.
Cell Cycle 7 , 759-764.
Eulalio, A., Huntzinger, E., Nishihara, T., Rehwinkel, J., Fauser, M., and
Izaurralde, E. (2009). Deadenylation is a widespread effect of miRNA regulation. RNA 15 , 21-32.
Fabbri, M., Garzon, R., Cimmino, A., Liu, Z., Zanesi, N., Callegari, E., Liu, S.,
Alder, H., Costinean, S., Fernandez-Cymering, C.
, et al.
(2007).
MicroRNA-29 family reverts aberrant methylation in lung cancer by targeting DNA methyltransferases 3A and 3B. Proc Natl Acad Sci U S A
104 , 15805-15810.
Fabian, M. R., Mathonnet, G., Sundermeier, T., Mathys, H., Zipprich, J. T.,
Svitkin, Y. V., Rivas, F., Jinek, M., Wohlschlegel, J., Doudna, J. A.
, et al.
(2009). Mammalian miRNA RISC recruits CAF1 and PABP to affect
PABP-dependent deadenylation. Mol Cell 35 , 868-880.
Farh, K. K., Grimson, A., Jan, C., Lewis, B. P., Johnston, W. K., Lim, L. P.,
Burge, C. B., and Bartel, D. P. (2005). The widespread impact of mammalian MicroRNAs on mRNA repression and evolution. Science 310 ,
1817-1821.
Frankel, L. B., Christoffersen, N. R., Jacobsen, A., Lindow, M., Krogh, A., and
Lund, A. H. (2008). Programmed cell death 4 (PDCD4) is an important functional target of the microRNA miR-21 in breast cancer cells. J Biol
Chem 283 , 1026-1033.
Friedman, R. C., Farh, K. K., Burge, C. B., and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res
19 , 92-105.
Garzon, R., Calin, G. A., and Croce, C. M. (2009). MicroRNAs in Cancer. Annu
Rev Med 60 , 167-179.
Garzon, R., Volinia, S., Liu, C. G., Fernandez-Cymering, C., Palumbo, T.,
Pichiorri, F., Fabbri, M., Coombes, K., Alder, H., Nakamura, T.
, et al.
(2008). MicroRNA signatures associated with cytogenetics and prognosis in acute myeloid leukemia. Blood 111 , 3183-3189.
Gaur, A., Jewell, D. A., Liang, Y., Ridzon, D., Moore, J. H., Chen, C., Ambros, V.
R., and Israel, M. A. (2007). Characterization of microRNA expression levels and their biological correlates in human cancer cell lines. Cancer
Res 67 , 2456-2468.
29

Giraldez, A. J., Mishima, Y., Rihel, J., Grocock, R. J., Van Dongen, S., Inoue, K.,
Enright, A. J., and Schier, A. F. (2006). Zebrafish MiR-430 promotes deadenylation and clearance of maternal mRNAs. Science 312 , 75-79.
Gregory, R. I., Chendrimada, T. P., Cooch, N., and Shiekhattar, R. (2005).
Human RISC couples microRNA biogenesis and posttranscriptional gene silencing. Cell 123 , 631-640.
Grimson, A., Farh, K. K., Johnston, W. K., Garrett-Engele, P., Lim, L. P., and
Bartel, D. P. (2007). MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell 27 , 91-105.
Grishok, A., Pasquinelli, A. E., Conte, D., Li, N., Parrish, S., Ha, I., Baillie, D. L.,
Fire, A., Ruvkun, G., and Mello, C. C. (2001). Genes and mechanisms related to RNA interference regulate expression of the small temporal
RNAs that control C. elegans developmental timing. Cell 106 , 23-34.
Guo, H., Ingolia, N. T., Weissman, J. S., and Bartel, D. P. (2010). Mammalian microRNAs predominantly act to decrease target mRNA levels. Nature
466 , 835-840.
Haase, A. D., Jaskiewicz, L., Zhang, H., Laine, S., Sack, R., Gatignol, A., and
Filipowicz, W. (2005). TRBP, a regulator of cellular PKR and HIV-1 virus expression, interacts with Dicer and functions in RNA silencing. EMBO
Rep 6 , 961-967.
Hagan, J. P., Piskounova, E., and Gregory, R. I. (2009). Lin28 recruits the
TUTase Zcchc11 to inhibit let-7 maturation in mouse embryonic stem cells. Nat Struct Mol Biol 16 , 1021-1025.
Han, J., Lee, Y., Yeom, K. H., Nam, J. W., Heo, I., Rhee, J. K., Sohn, S. Y., Cho,
Y., Zhang, B. T., and Kim, V. N. (2006). Molecular basis for the recognition of primary microRNAs by the Drosha-DGCR8 complex. Cell 125 , 887-901.
Hayashi, K., Chuva de Sousa Lopes, S. M., Kaneda, M., Tang, F., Hajkova, P.,
Lao, K., O'Carroll, D., Das, P. P., Tarakhovsky, A., Miska, E. A., and
Surani, M. A. (2008). MicroRNA biogenesis is required for mouse primordial germ cell development and spermatogenesis. PLoS One 3 , e1738.
Heo, I., Joo, C., Kim, Y. K., Ha, M., Yoon, M. J., Cho, J., Yeom, K. H., Han, J., and Kim, V. N. (2009). TUT4 in concert with Lin28 suppresses microRNA biogenesis through pre-microRNA uridylation. Cell 138 , 696-708.
Hill, D. A., Ivanovich, J., Priest, J. R., Gurnett, C. A., Dehner, L. P., Desruisseau,
D., Jarzembowski, J. A., Wikenheiser-Brokamp, K. A., Suarez, B. K.,
Whelan, A. J.
, et al.
(2009). DICER1 mutations in familial pleuropulmonary blastoma. Science 325 , 965.
Houbaviy, H. B., Murray, M. F., and Sharp, P. A. (2003). Embryonic stem cellspecific MicroRNAs. Dev Cell 5 , 351-358.
Hutvagner, G., McLachlan, J., Pasquinelli, A. E., Balint, E., Tuschl, T., and
Zamore, P. D. (2001). A cellular function for the RNA-interference enzyme
Dicer in the maturation of the let-7 small temporal RNA. Science 293 , 834-
838.
Hutvagner, G., and Zamore, P. D. (2002). A microRNA in a multiple-turnover
RNAi enzyme complex. Science 297 , 2056-2060.
30

Johnson, S. M., Grosshans, H., Shingara, J., Byrom, M., Jarvis, R., Cheng, A.,
Labourier, E., Reinert, K. L., Brown, D., and Slack, F. J. (2005). RAS is regulated by the let-7 microRNA family. Cell 120 , 635-647.
Judson, R. L., Babiarz, J. E., Venere, M., and Blelloch, R. (2009). Embryonic stem cell-specific microRNAs promote induced pluripotency. Nat
Biotechnol 27 , 459-461.
Kanellopoulou, C., Muljo, S. A., Kung, A. L., Ganesan, S., Drapkin, R., Jenuwein,
T., Livingston, D. M., and Rajewsky, K. (2005). Dicer-deficient mouse embryonic stem cells are defective in differentiation and centromeric silencing. Genes Dev 19 , 489-501.
Ketting, R. F., Fischer, S. E., Bernstein, E., Sijen, T., Hannon, G. J., and
Plasterk, R. H. (2001). Dicer functions in RNA interference and in synthesis of small RNA involved in developmental timing in C. elegans.
Genes Dev 15 , 2654-2659.
Khvorova, A., Reynolds, A., and Jayasena, S. D. (2003). Functional siRNAs and miRNAs exhibit strand bias. Cell 115 , 209-216.
Kim, V. N. (2005). MicroRNA biogenesis: coordinated cropping and dicing. Nat
Rev Mol Cell Biol 6 , 376-385.
Kloosterman, W. P., Wienholds, E., Ketting, R. F., and Plasterk, R. H. (2004).
Substrate requirements for let-7 function in the developing zebrafish embryo. Nucleic Acids Res 32 , 6284-6291.
Kluiver, J., Poppema, S., de Jong, D., Blokzijl, T., Harms, G., Jacobs, S.,
Kroesen, B. J., and van den Berg, A. (2005). BIC and miR-155 are highly expressed in Hodgkin, primary mediastinal and diffuse large B cell lymphomas. J Pathol 207 , 243-249.
Koera, K., Nakamura, K., Nakao, K., Miyoshi, J., Toyoshima, K., Hatta, T., Otani,
H., Aiba, A., and Katsuki, M. (1997). K-ras is essential for the development of the mouse embryo. Oncogene 15 , 1151-1159.
Kumar, M. S., Erkeland, S. J., Pester, R. E., Chen, C. Y., Ebert, M. S., Sharp, P.
A., and Jacks, T. (2008). Suppression of non-small cell lung tumor development by the let-7 microRNA family. Proc Natl Acad Sci U S A 105 ,
3903-3908.
Kumar, M. S., Lu, J., Mercer, K. L., Golub, T. R., and Jacks, T. (2007). Impaired microRNA processing enhances cellular transformation and tumorigenesis. Nat Genet 39 , 673-677.
Kumar, M. S., Pester, R. E., Chen, C. Y., Lane, K., Chin, C., Lu, J., Kirsch, D. G.,
Golub, T. R., and Jacks, T. (2009). Dicer1 functions as a haploinsufficient tumor suppressor. Genes Dev 23 , 2700-2704.
Kwon, C., Han, Z., Olson, E. N., and Srivastava, D. (2005). MicroRNA1 influences cardiac differentiation in Drosophila and regulates Notch signaling. Proc Natl Acad Sci U S A 102 , 18986-18991.
Lagos-Quintana, M., Rauhut, R., Lendeckel, W., and Tuschl, T. (2001).
Identification of novel genes coding for small expressed RNAs. Science
294 , 853-858.
Lagos-Quintana, M., Rauhut, R., Meyer, J., Borkhardt, A., and Tuschl, T. (2003).
New microRNAs from mouse and human. RNA 9 , 175-179.
31

Lai, E. C., Tam, B., and Rubin, G. M. (2005). Pervasive regulation of Drosophila
Notch target genes by GY-box-, Brd-box-, and K-box-class microRNAs.
Genes Dev 19 , 1067-1080.
Lau, N. C., Lim, L. P., Weinstein, E. G., and Bartel, D. P. (2001). An abundant class of tiny RNAs with probable regulatory roles in Caenorhabditis elegans. Science 294 , 858-862.
Lee, R. C., and Ambros, V. (2001). An extensive class of small RNAs in
Caenorhabditis elegans. Science 294 , 862-864.
Lee, R. C., Feinbaum, R. L., and Ambros, V. (1993). The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 75 , 843-854.
Lee, Y., Ahn, C., Han, J., Choi, H., Kim, J., Yim, J., Lee, J., Provost, P.,
Radmark, O., Kim, S., and Kim, V. N. (2003). The nuclear RNase III
Drosha initiates microRNA processing. Nature 425 , 415-419.
Lee, Y. S., and Dutta, A. (2007). The tumor suppressor microRNA let-7 represses the HMGA2 oncogene. Genes Dev 21 , 1025-1030.
Lewis, B. P., Shih, I. H., Jones-Rhoades, M. W., Bartel, D. P., and Burge, C. B.
(2003). Prediction of mammalian microRNA targets. Cell 115 , 787-798.
Lewis, B. P., Burge, C. B., and Bartel, D. P. (2005). Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 120 , 15-20.
Lim, L. P., Lau, N. C., Garrett-Engele, P., Grimson, A., Schelter, J. M., Castle, J.,
Bartel, D. P., Linsley, P. S., and Johnson, J. M. (2005). Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 433 , 769-773.
Liu, J., Carmell, M. A., Rivas, F. V., Marsden, C. G., Thomson, J. M., Song, J. J.,
Hammond, S. M., Joshua-Tor, L., and Hannon, G. J. (2004). Argonaute2 is the catalytic engine of mammalian RNAi. Science 305 , 1437-1441.
Lu, J., Getz, G., Miska, E. A., Alvarez-Saavedra, E., Lamb, J., Peck, D., Sweet-
Cordero, A., Ebert, B. L., Mak, R. H., Ferrando, A. A.
, et al.
(2005).
MicroRNA expression profiles classify human cancers. Nature 435 , 834-
838.
Lund, E., Guttinger, S., Calado, A., Dahlberg, J. E., and Kutay, U. (2004).
Nuclear export of microRNA precursors. Science 303 , 95-98.
Marson, A., Levine, S. S., Cole, M. F., Frampton, G. M., Brambrink, T.,
Johnstone, S., Guenther, M. G., Johnston, W. K., Wernig, M., Newman, J.
, et al.
(2008). Connecting microRNA genes to the core transcriptional regulatory circuitry of embryonic stem cells. Cell 134 , 521-533.
Mayr, C., and Bartel, D. P. (2009). Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells. Cell 138 , 673-684.
Mayr, C., Hemann, M. T., and Bartel, D. P. (2007). Disrupting the pairing between let-7 and Hmga2 enhances oncogenic transformation. Science
315 , 1576-1579.
32

Meister, G., Landthaler, M., Patkaniowska, A., Dorsett, Y., Teng, G., and Tuschl,
T. (2004). Human Argonaute2 mediates RNA cleavage targeted by miRNAs and siRNAs. Mol Cell 15 , 185-197.
Melo, S. A., Ropero, S., Moutinho, C., Aaltonen, L. A., Yamamoto, H., Calin, G.
A., Rossi, S., Fernandez, A. F., Carneiro, F., Oliveira, C.
, et al.
(2009). A
TARBP2 mutation in human cancer impairs microRNA processing and
DICER1 function. Nat Genet 41 , 365-370.
Melton, C., Judson, R. L., and Blelloch, R. (2010). Opposing microRNA families regulate self-renewal in mouse embryonic stem cells. Nature 463 , 621-
626.
Meng, F., Henson, R., Wehbe-Janek, H., Ghoshal, K., Jacob, S. T., and Patel, T.
(2007). MicroRNA-21 regulates expression of the PTEN tumor suppressor gene in human hepatocellular cancer. Gastroenterology 133 , 647-658.
Metzler, M., Wilda, M., Busch, K., Viehmann, S., and Borkhardt, A. (2004). High expression of precursor microRNA-155/BIC RNA in children with Burkitt lymphoma. Genes Chromosomes Cancer 39 , 167-169.
Mineno, J., Okamoto, S., Ando, T., Sato, M., Chono, H., Izu, H., Takayama, M.,
Asada, K., Mirochnitchenko, O., Inouye, M., and Kato, I. (2006). The expression profile of microRNAs in mouse embryos. Nucleic Acids Res
34 , 1765-1771.
Mott, J. L., Kobayashi, S., Bronk, S. F., and Gores, G. J. (2007). mir-29 regulates
Mcl-1 protein expression and apoptosis. Oncogene 26 , 6133-6140.
Murchison, E. P., Partridge, J. F., Tam, O. H., Cheloufi, S., and Hannon, G. J.
(2005). Characterization of Dicer-deficient murine embryonic stem cells.
Proc Natl Acad Sci U S A 102 , 12135-12140.
O'Donnell, K. A., Wentzel, E. A., Zeller, K. I., Dang, C. V., and Mendell, J. T.
(2005). c-Myc-regulated microRNAs modulate E2F1 expression. Nature
435 , 839-843.
Okamura, K., Hagen, J. W., Duan, H., Tyler, D. M., and Lai, E. C. (2007). The mirtron pathway generates microRNA-class regulatory RNAs in
Drosophila. Cell 130 , 89-100.
Pasquinelli, A. E., McCoy, A., Jimenez, E., Salo, E., Ruvkun, G., Martindale, M.
Q., and Baguna, J. (2003). Expression of the 22 nucleotide let-7 heterochronic RNA throughout the Metazoa: a role in life history evolution? Evol Dev 5 , 372-378.
Pasquinelli, A. E., Reinhart, B. J., Slack, F., Martindale, M. Q., Kuroda, M. I.,
Maller, B., Hayward, D. C., Ball, E. E., Degnan, B., Muller, P.
, et al.
(2000). Conservation of the sequence and temporal expression of let-7 heterochronic regulatory RNA. Nature 408 , 86-89.
Pedersen-Bjergaard, J., Pedersen, M., Roulston, D., and Philip, P. (1995).
Different genetic pathways in leukemogenesis for patients presenting with therapy-related myelodysplasia and therapy-related acute myeloid leukemia. Blood 86 , 3542-3552.
Petersen, C. P., Bordeleau, M. E., Pelletier, J., and Sharp, P. A. (2006). Short
RNAs repress translation after initiation in mammalian cells. Mol Cell 21 ,
533-542.
33

Petrocca, F., Visone, R., Onelli, M. R., Shah, M. H., Nicoloso, M. S., de Martino,
I., Iliopoulos, D., Pilozzi, E., Liu, C. G., Negrini, M.
, et al.
(2008). E2F1regulated microRNAs impair TGFbeta-dependent cell-cycle arrest and apoptosis in gastric cancer. Cancer Cell 13 , 272-286.
Pillai, R. S., Artus, C. G., and Filipowicz, W. (2004). Tethering of human Ago proteins to mRNA mimics the miRNA-mediated repression of protein synthesis. RNA 10 , 1518-1525.
Pillai, R. S., Bhattacharyya, S. N., Artus, C. G., Zoller, T., Cougot, N., Basyuk, E.,
Bertrand, E., and Filipowicz, W. (2005). Inhibition of translational initiation by Let-7 MicroRNA in human cells. Science 309 , 1573-1576.
Reinhart, B. J., Slack, F. J., Basson, M., Pasquinelli, A. E., Bettinger, J. C.,
Rougvie, A. E., Horvitz, H. R., and Ruvkun, G. (2000). The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans.
Nature 403 , 901-906.
Rodriguez, A., Vigorito, E., Clare, S., Warren, M. V., Couttet, P., Soond, D. R., van Dongen, S., Grocock, R. J., Das, P. P., Miska, E. A.
, et al.
(2007).
Requirement of bic/microRNA-155 for normal immune function. Science
316 , 608-611.
Rosa, A., Spagnoli, F. M., and Brivanlou, A. H. (2009). The miR-430/427/302 family controls mesendodermal fate specification via species-specific target selection. Dev Cell 16 , 517-527.
Roush, S. F., and Slack, F. J. (2009). Transcription of the C. elegans let-7 microRNA is temporally regulated by one of its targets, hbl-1. Dev Biol
334 , 523-534.
Ruby, J. G., Jan, C. H., and Bartel, D. P. (2007). Intronic microRNA precursors that bypass Drosha processing. Nature 448 , 83-86.
Rybak, A., Fuchs, H., Hadian, K., Smirnova, L., Wulczyn, E. A., Michel, G.,
Nitsch, R., Krappmann, D., and Wulczyn, F. G. (2009). The let-7 target gene mouse lin-41 is a stem cell specific E3 ubiquitin ligase for the miRNA pathway protein Ago2. Nat Cell Biol 11 , 1411-1420.
Sampson, V. B., Rong, N. H., Han, J., Yang, Q., Aris, V., Soteropoulos, P.,
Petrelli, N. J., Dunn, S. P., and Krueger, L. J. (2007). MicroRNA let-7a down-regulates MYC and reverts MYC-induced growth in Burkitt lymphoma cells. Cancer Res 67 , 9762-9770.
Schwarz, D. S., Hutvagner, G., Du, T., Xu, Z., Aronin, N., and Zamore, P. D.
(2003). Asymmetry in the assembly of the RNAi enzyme complex. Cell
115 , 199-208.
Selbach, M., Schwanhausser, B., Thierfelder, N., Fang, Z., Khanin, R., and
Rajewsky, N. (2008). Widespread changes in protein synthesis induced by microRNAs. Nature 455 , 58-63.
Shin, C., Nam, J. W., Farh, K. K., Chiang, H. R., Shkumatava, A., and Bartel, D.
P. (2010). Expanding the microRNA targeting code: functional sites with centered pairing. Mol Cell 38 , 789-802.
Sinkkonen, L., Hugenschmidt, T., Berninger, P., Gaidatzis, D., Mohn, F., Artus-
Revel, C. G., Zavolan, M., Svoboda, P., and Filipowicz, W. (2008).
MicroRNAs control de novo DNA methylation through regulation of
34

transcriptional repressors in mouse embryonic stem cells. Nat Struct Mol
Biol 15 , 259-267.
Sokol, N. S., and Ambros, V. (2005). Mesodermally expressed Drosophila microRNA-1 is regulated by Twist and is required in muscles during larval growth. Genes Dev 19 , 2343-2354.
Stark, A., Brennecke, J., Bushati, N., Russell, R. B., and Cohen, S. M. (2005).
Animal MicroRNAs confer robustness to gene expression and have a significant impact on 3'UTR evolution. Cell 123 , 1133-1146.
Subramanyam, D., Lamouille, S., Judson, R. L., Liu, J. Y., Bucay, N., Derynck,
R., and Blelloch, R. (2011). Multiple targets of miR-302 and miR-372 promote reprogramming of human fibroblasts to induced pluripotent stem cells. Nat Biotechnol.
Suh, M. R., Lee, Y., Kim, J. Y., Kim, S. K., Moon, S. H., Lee, J. Y., Cha, K. Y.,
Chung, H. M., Yoon, H. S., Moon, S. Y.
, et al.
(2004). Human embryonic stem cells express a unique set of microRNAs. Dev Biol 270 , 488-498.
Tam, W., Hughes, S. H., Hayward, W. S., and Besmer, P. (2002). Avian bic, a gene isolated from a common retroviral site in avian leukosis virusinduced lymphomas that encodes a noncoding RNA, cooperates with cmyc in lymphomagenesis and erythroleukemogenesis. J Virol 76 , 4275-
4286.
Thomson, J. M., Newman, M., Parker, J. S., Morin-Kensicki, E. M., Wright, T., and Hammond, S. M. (2006). Extensive post-transcriptional regulation of microRNAs and its implications for cancer. Genes Dev 20 , 2202-2207.
Tsang, J. S., Ebert, M. S., and van Oudenaarden, A. (2010). Genome-wide dissection of microRNA functions and cotargeting networks using gene set signatures. Mol Cell 38 , 140-153. van Haaften, G., and Agami, R. (2010). Tumorigenicity of the miR-17-92 cluster distilled. Genes Dev 24 , 1-4.
Ventura, A., Young, A. G., Winslow, M. M., Lintault, L., Meissner, A., Erkeland, S.
J., Newman, J., Bronson, R. T., Crowley, D., Stone, J. R.
, et al.
(2008).
Targeted deletion reveals essential and overlapping functions of the miR-
17 through 92 family of miRNA clusters. Cell 132 , 875-886.
Viswanathan, S. R., Daley, G. Q., and Gregory, R. I. (2008). Selective blockade of microRNA processing by Lin28. Science 320 , 97-100.
Volinia, S., Calin, G. A., Liu, C. G., Ambs, S., Cimmino, A., Petrocca, F., Visone,
R., Iorio, M., Roldo, C., Ferracin, M.
, et al.
(2006). A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl
Acad Sci U S A 103 , 2257-2261.
Voorhoeve, P. M., le Sage, C., Schrier, M., Gillis, A. J., Stoop, H., Nagel, R., Liu,
Y. P., van Duijse, J., Drost, J., Griekspoor, A.
, et al.
(2007). A genetic screen implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Adv Exp Med Biol 604 , 17-46.
Wang, Y., Baskerville, S., Shenoy, A., Babiarz, J. E., Baehner, L., and Blelloch,
R. (2008). Embryonic stem cell-specific microRNAs regulate the G1-S transition and promote rapid proliferation. Nat Genet 40 , 1478-1483.
35

Wang, Y., Medvid, R., Melton, C., Jaenisch, R., and Blelloch, R. (2007). DGCR8 is essential for microRNA biogenesis and silencing of embryonic stem cell self-renewal. Nat Genet 39 , 380-385.
Wightman, B., Ha, I., and Ruvkun, G. (1993). Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in
C. elegans. Cell 75 , 855-862.
Winter, J., Jung, S., Keller, S., Gregory, R. I., and Diederichs, S. (2009). Many roads to maturity: microRNA biogenesis pathways and their regulation.
Nat Cell Biol 11 , 228-234.
Wynendaele, J., Bohnke, A., Leucci, E., Nielsen, S. J., Lambertz, I., Hammer, S.,
Sbrzesny, N., Kubitza, D., Wolf, A., Gradhand, E.
, et al.
(2010). An illegitimate microRNA target site within the 3' UTR of MDM4 affects ovarian cancer progression and chemosensitivity. Cancer Res 70 , 9641-
9649.
Xiao, C., Srinivasan, L., Calado, D. P., Patterson, H. C., Zhang, B., Wang, J.,
Henderson, J. M., Kutok, J. L., and Rajewsky, K. (2008).
Lymphoproliferative disease and autoimmunity in mice with increased miR-17-92 expression in lymphocytes. Nat Immunol 9 , 405-414.
Yekta, S., Shih, I. H., and Bartel, D. P. (2004). MicroRNA-directed cleavage of
HOXB8 mRNA. Science 304 , 594-596.
Yi, R., Qin, Y., Macara, I. G., and Cullen, B. R. (2003). Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes Dev 17 ,
3011-3016.
Yu, J., Vodyanik, M. A., Smuga-Otto, K., Antosiewicz-Bourget, J., Frane, J. L.,
Tian, S., Nie, J., Jonsdottir, G. A., Ruotti, V., Stewart, R.
, et al.
(2007).
Induced pluripotent stem cell lines derived from human somatic cells.
Science 318 , 1917-1920.
Zekri, L., Huntzinger, E., Heimstadt, S., and Izaurralde, E. (2009). The silencing domain of GW182 interacts with PABPC1 to promote translational repression and degradation of microRNA targets and is required for target release. Mol Cell Biol 29 , 6220-6231.
Zhao, Y., Samal, E., and Srivastava, D. (2005). Serum response factor regulates a muscle-specific microRNA that targets Hand2 during cardiogenesis.
Nature 436 , 214-220.
36

The material in this chapter was adapted with permission from the following publication:
Grace X.Y. Zheng, Arvind Ravi, J. Mauro Calabrese, Lea A. Medeiros, Oktay Kirak,
Lucas M. Dennis, Rudolf Jaenisch, Christopher B. Burge, Phillip A. Sharp (2011). A latent pro-survival function for the miR-290-295 cluster in mouse embryonic stem cells.
PLoS Genet. (in press)
Experimental contributions:
This work represents an equal collaboration between Grace Zheng and Arvind Ravi.
Arvind Ravi performed the Dicer null expression profiling and target validation.
Grace Zheng performed the informatics and apoptosis experiments.
37

MicroRNAs (miRNAs) post-transcriptionally regulate the expression of thousands of distinct mRNAs. While some regulatory interactions help to maintain basal cellular functions, others are likely relevant in more specific settings, such as response to stress. Here we describe such a role for the mir-
290-295 cluster, the dominant miRNA cluster in mouse embryonic stem cells
(mESCs). Examination of a target list generated from bioinformatic prediction as well as expression data following miRNA loss revealed strong enrichment for apoptotic regulators, two of which we validated directly: Caspase 2, the most highly conserved mammalian caspase, and Ei24, a p53 transcriptional target.
Consistent with these predictions, mESCs lacking miRNAs were more likely to initiate apoptosis following genotoxic exposure to gamma irradiation or doxorubicin. Knockdown of either candidate partially rescued this pro-apoptotic phenotype, as did transfection of members of the mir-290-295 cluster. These findings were recapitulated in a specific mir-290-295 deletion line, confirming that they reflect miRNA functions at physiological levels. In contrast to the basal regulatory roles previously identified, the pro-survival phenotype shown here may be most relevant to stressful pregnancies, where pro-oxidant metabolic states induce DNA damage. Similarly, this cluster may mediate chemotherapeutic resistance in a neoplastic context, making it a useful clinical target.
38

MicroRNAs (miRNAs) are endogenous ~22nt RNAs that regulate gene expression post-transcriptionally. In animals, the ability of miRNAs to accomplish this regulation depends on complementarity between mature miRNA sequences and their mRNA targets. Most commonly, partial binding of miRNAs leads to destabilization of mRNA transcripts and/or inhibition of productive translation, and in rare cases perfect complementarity instead causes target cleavage. Both in vitro experiments and bioinformatics have shown that matches to positions 2-7 of the miRNA, referred to as the miRNA “seed,” are generally required for effective miRNA-directed mRNA downregulation (Grimson et al., 2007; Nielsen et al., 2007).
The roles of miRNAs in mouse embryonic stem cells (mESCs) have been of particular interest, as this knowledge may shed light on key aspects of mammalian development and generate useful insights into reprogramming and cancer, both of which recapitulate aspects of an ESC expression state (Ben-
Porath et al., 2008; Wernig et al., 2007). In addition, the survival of mESCs in the absence of Dicer (Dcr), the key RNase III enzyme that generates mature miRNAs, makes them a unique model system for dissecting miRNA function
(Kanellopoulou et al., 2005; Murchison et al., 2005). Several large-scale sequencing datasets (Babiarz et al., 2008; Ciaudo et al., 2009; Leung, 2010) have revealed that the mir-290-295 cluster constitutes the dominant miRNA population in mESCs, giving rise to about 50% of all reads in these cells (Table
39

S1). Many of the miRNAs in this cluster share the hexamer seed ‘AAGUGC,’ which is also expressed at much lower levels by the mir-302 and mir-467 clusters, contributing less than 5% of total reads (Table S2). A similar percent contribution to total miRNA levels comes from the miR-17-92 family, which contains the shifted seed ‘AAAGUG,’ and therefore may share some common targets (Table S2) (Babiarz et al., 2008; Ciaudo et al., 2009; Leung, 2010). Given the abundance of the mir-290-295 cluster and these related sequences, much of mESC miRNA physiology is likely to be a function of this dominant seed sequence.
Within the mir-290-295 cluster, the ‘AAGUGC’ seed is found in miR-290-
3p, miR-291a-3p, miR-291b-3p, miR-292-3p, miR-294, and miR-295. Consistent with their high expression, these miRNAs (which we shall refer to as the mir-295 cluster) have been linked to a number of functions in ES cells including maintenance of pluripotency and proliferation. For instance, miR-290-295 miRNAs have been shown to target Rbl2, which controls the expression of
Dnmt3a and Dnmt3b (Benetti et al., 2008; Sinkkonen et al., 2008), suggesting a role for this miRNA cluster in regulating de novo DNA methylation. In addition, miR-290-295 miRNAs have been found to accelerate cell proliferation by promoting the G1 to S phase transition through targets such as p21 and Lats2
(Wang et al., 2008). However, additional roles for this cluster remain to be elucidated.
Using a combination of target prediction data with microarrays of mESCs before (Dcr WT) and after (Dcr KO) miRNA loss, as well as before (295 WT) and
40

after (295 KO) specific deletion of the mir-295 cluster (Medeiros et. al.
, manuscript in preparation), we have identified novel targets of the mir-295 cluster in ES cells. Initial analysis suggested strong enrichment of targets involved in apoptosis, a function that to date has not been linked to ES-cell specific miRNAs.
Through gain- and loss-of-function studies, we show that miR-290-295 miRNAs indeed serve a protective function in preventing mESC apoptosis during exposure to genotoxic stress. This protective effect appears to be mediated in part by direct repression of two novel targets, Caspase 2 and Ei24. As activation of these targets is dependent on DNA damage, we propose that their regulation may be particularly relevant during physiological stress in embryonic development. In addition, given prior indications that these two genes act as tumor suppressors, misexpression of this cluster in the context of cancer may promote resistance to standard genotoxic therapeutics.
In order to identify additional endogenous targets of mESC miRNAs, we performed expression profiling of mESCs following Cre recombinase-mediated
Dcr deletion using a previously characterized floxed Dcr mESC line (Calabrese and Sharp, 2006; Harfe et al., 2005; Leung et al., 2006). As Dcr deletion leads to slower proliferation (Murchison et al., 2005; Wang et al., 2008), acute loss was examined in a polyclonal population, averaging over potential clonal variants and
41

enriching for initial miRNA-mediated derepression rather than subsequent compensatory changes. Indeed, expression profiling from 3 biological replicates taken 5 days following deletion, a time point by which cells were predominantly
Dcr null and a majority of miRNAs were lost (Figure 1A and 1B), showed better clustering than 3 chronic deletion cell lines, as indicated by higher Pearson correlation coefficients (Figure S1). To confirm that targets of the mir-295 cluster show a transcriptome-wide signature in this dataset, we calculated a cumulative density function (cdf) plot comparing expression differences for the set of all mir-
295 cluster targets as determined by Targetscan 5.1 (Friedman et al., 2009).
Relative to a control set of genes ( control ) matched for 3' UTR length, dinucleotide composition, and expression level, the mir-295 cluster target set
( targets ) was more derepressed upon Dcr loss (Figure 1C). An even larger derepression was seen for conserved mir-295 cluster targets ( conserved targets ), suggesting further enrichment of genuine targets in this set (Figure 1C).
This observation supports the utility of these expression data for target discovery.
To better understand the global effects of miRNA loss in ESCs, we next performed Gene Ontology (GO) analysis on an initial candidate set. Enrichment in specific GO categories was tested for all genes that increased on Dcr loss
(defined as " 1.2 fold up-regulation). The top statistically significant categories included “Regulators of Apoptosis” and “Cell Cycle” ( p = 2.1e
-8 and p = 5.6e
-5 , respectively). We further refined our candidate list using available array data from the 295 KO line, which also showed cdf plot signature changes for mir-295 cluster targets (Figure 1C) (Medeiros et. al ., manuscript in preparation). In all,
42

807 candidates were identified as Targetscan-predicted targets of the cluster that showed at least a 1.2 fold up-regulation in knockout populations from both datasets (Figure 2A, Table S3). Over 40% of upregulated transcripts were shared between the Dcr KO and 295 KO lines, consistent with the finding that the mir-
295 cluster contributes around half of all miRNAs in ES cells. The fact that this overlap is not even greater may be due to direct and indirect effects of non-
AAGUGC seeds, as there is significantly more overlap – in fact, closer to 70% – between the two data sets when considering only those genes that are
Targetscan-predicted AAGUGC targets ( p < 0.001, Fischer’s exact test).
Several candidate target genes were selected for further examination based on their degree of upregulation in Dcr KO and 295 KO cells, as well as their functional annotations. The tested targets span a range of biological functions and processes, from cell cycle regulators Lats2 and p21 to immunological signal transduction components Irf9 and Irak3. Their 3' UTRs were cloned into luciferase constructs, and expression levels between Dcr WT and Dcr KO cells were evaluated (Figure 2B). All candidates tested displayed at least mild repression relative to a control construct lacking miRNA target sites, ranging from strong (~5-fold) to modest (~30%) down-regulation. The magnitude of repression for the previously identified miR-295 targets Lats2 and p21 was comparable to that observed previously (Wang et al., 2008). Additional transfection studies confirmed that repression could be conferred specifically by miR-295 in a Dcr KO background (Figure S2). These in vitro results support the enrichment of our candidate list for true miR-295 targets.
43

We chose to more closely examine one of the most strongly downregulated reporter targets, Caspase 2 (Casp2), along with Ei24, as these targets could provide a novel link between ESC-specific microRNAs and cell survival.
Casp2, an initiator of apoptosis in response to genotoxic stress (Li and Yuan,
2008), has four AAGUGC binding sites in its 3' UTR. Quantitative RT-PCR analysis demonstrated an approximately 5-fold increase in Casp2 transcript levels in Dcr KO cells, consistent with the degree of derepression observed with the luciferase reporter assay (Figure 3A). This observation indicates that the majority of miRNA repression likely occurs at the level of transcript stability. In support of the reporter assay, Dcr KO cells showed a comparable increase in
Casp2 at the protein level, which could be partially rescued by transfection of either miR-295, miR-467a (which shares the same hexamer seed), or a Casp2 siRNA, but not by siRNAs against other unrelated targets (Figure 3B).
Transfection of miR-295 also strongly repressed an intact Casp2 reporter in these cells, but not a reporter in which the four target sites were mutated (Figure
3C). Combinatorial mutagenesis revealed that repression was not conferred equally by these four sites, as much of the repression was lost by mutation of the first two sites alone (Figure 3D). Taken together, these data suggest that direct miRNA-mediated repression of Casp2 leads to approximately 5-fold repression,
44

making it one of the most potently repressed mir-295 cluster targets identified to date.
We additionally characterized the novel target Ei24, which has also been implicated in apoptosis. Originally identified as a direct p53 transcriptional target that binds Bcl2 (Gu et al., 2000; Zhao et al., 2005), the Ei24 3' UTR contains one
7mer miR-295 site. The 3' UTR of Ei24 fused to a luciferase reporter conferred approximately 2-fold repression in Dcr WT cells relative to Dcr KO cells, an effect that could be restored following transfection of miR-295 (Figure 3C). Notably, repression was lost upon mutation of the seed site, confirming that Ei24 is a direct target.
Based on these repression data as well as the earlier informatic predictions, we tested whether mir-295 cluster miRNAs could modulate apoptosis in mESCs. The basal apoptosis rates of Dcr WT and KO ES cells in a 24 h period were compared by staining them with antibodies against cleaved Caspase 3
(Casp3) and then analyzing cells by flow cytometry. Under these conditions, only modest basal apoptotic rates were observed, with Dcr KO ES cells showing a slightly higher apoptosis rate than Dcr WT cells (Figure 4A, Figure S3A). Given that ESCs are highly sensitive to DNA damage (Tichy and Stambrook, 2008) and both validated targets have been implicated in the DNA damage response, we
45

hypothesized that the mir-295 cluster may be specifically protective in the context of genotoxic stress. To test this, we first examined the effect of exposing WT and
Dcr KO cells to gamma irradiation or doxorubicin. Gamma irradiation induces
DNA damage and activates ATM and p53, as does doxorubicin, a topoisomerase
II inhibitor (Nitiss, 2009). These signals lead to activation of the intrinsic apoptosis pathway and result in the cleavage of Casp3 (Fulda and Debatin,
2006). We were able to confirm this cleavage product by Western blot in our cell culture system, as well as cleavage of Nanog, a previously reported Casp3 target
(Fujita et al., 2008) (Figure S5A). We also observed a decrease in Casp2 levels and the appearance of the previously described 35kD cleavage product (Upton et al., 2008), confirming its activation in our system (Figure S5B). Because this band was specific to DNA damage induction, the upregulation of Casp2 in Dcr
KO cells appears to be insufficient to generate autocleavage. Both Dcr WT and
KO ES cells showed minimal Casp3 activation immediately after 5-Gy gammairradiation, in line with previous descriptions of a 1-2 h lag phase in its activation
(Figure 4A) (Tyas et al., 2000). However, there was a notable difference in their responses 24 h after the treatment (and to a lesser extent 10 h after treatment,
Figure S4A); while 10% of WT cells became apoptotic, more than 30% of the Dcr
KO cells exhibited Casp3 activity (Figure 4A). Similar results were seen using
Annexin V staining, a complementary assay for detecting early apoptosis (Figure
S3C). Importantly, mature miRNA levels from the mir-295 cluster were unchanged by these stressors (Figure S3D). Therefore, it appears that loss of
46

mature miRNAs leads to an enhancement of apoptosis in the presence of DNA damage.
To examine whether the miR-295 targets modulated apoptosis, we transfected a series of siRNAs into Dcr KO cells and evaluated cell death following irradiation. The difference in Casp3 activation between 0 and 24 h timepoints was calculated in order to account for differences in transfectionspecific toxicity (Figure S6A and S6B). Relative to control siRNAs, transfection of miR-290-3p or miR-295 drastically decreased the apoptosis response of Dcr KO cells to gamma irradiation (Figure 4B, Figure S6A and S6B). The reduction in apoptosis is specific to the AAGUGC seed, as seed mutants failed to rescue Dcr
KO ES cells from apoptosis. When we applied siRNAs specific to each validated target, or to Bim, a well-characterized proapoptotic factor, cells exhibited a decrease of 5-10% in Casp3 activation 24 h after irradiation, a level similar to mir-295 cluster miRNA overexpression (Figure 4B, Figure S6A and S6B). Similar findings were obtained when cells were treated with 100 nM - 300 nM doxorubicin, suggesting that the identified pathways are relevant to DNA damage in general (Figure 4C and 4D, Figure S4B, Figure S6C and S6D).
47

Because deletion of Dcr involves global miRNA loss, and three additional clusters containing the same or similar hexamer seed, mir-302, mir-467, and mir-
17-92, are expressed in ESCs (Table S2), we examined the 295 KO line to determine the specific contribution of the mir-295 cluster to cell survival. Genetic deletion offers the best insight into physiological function as it avoids overexpression artifacts of exogenous miRNAs or toxicity effects of miRNA inhibitors. This system also avoids confounding by alternative miRNAindependent roles for Dcr itself in cell survival, as have been recently reported in
C. elegans ( Nakagawa et al., 2010 ). We first re-examined the reporter constructs for Casp2 and Ei24 in the 295 KO ESC line relative to its wild-type counterpart. In this context, the Casp2 reporter was derepressed approximately half as strongly as it was in Dcr KO cells, suggesting that the miR-302 and miR-467a families of miRNAs incompletely compensate for loss of the mir-295 cluster (Figure 5A).
This partial derepression in the mir-295 cluster deletion probably reflects the quantitative change in the total level of AAGUGC seed miRNAs, as exogenous miR-295 could further repress Casp2 protein levels (Figure 5B). We next exploited 295 KO ES cells to determine whether these cells had an increase in apoptosis upon exposure to DNA damaging agents. 295 WT and KO ES cells were irradiated and the level of cleaved Casp3 activity was measured 0 and 24 h after treatment (Figure 5C, Figure S3B). As expected, 295 KO cells were much more sensitive to irradiation than their WT counterparts (Figure 5C). Again, overexpression of two miRNAs in the cluster, miR-290-3p and miR-295, reduced the rate of apoptosis (Figure 5D, Figure S7A and S7B). In addition, knockdown of
48

the validated targets Casp2 or Ei24, or the pathway component Bim, partially rescued cells from apoptosis caused by irradiation (Figure 5D, Figure S7A and
S7B). We repeated these experiments with 100 nM doxorubicin as before, obtaining similar results (Figure 5E and 5F, Figure S7C and S7D). Therefore, deletion and restoration of mir-295 cluster miRNAs recapitulate the modulation of apoptosis rates seen in a Dcr null context.
Here, we provide the first demonstration that the mir-295 cluster can suppress apoptosis in mESCs following exposure to the genotoxic stressors ionizing irradiation and doxorubicin. Initially suggested by an informatic comparison of global expression data following Dicer loss, the link between ESC miRNAs and cell death may act in part through the novel targets Casp2 and
Ei24. In the case of Casp2, this appears to occur through multiple seed match sites in the 3' UTR leading to a roughly 5 fold reduction in expression, while for
Ei24, targeting is achieved through just a single complementary site conferring approximately a 2 fold repression.
Although the exact functions of these two mediators are still emerging, multiple lines of evidence suggest that they are important in cell survival. Initial studies of Casp2 knockout mice showed increased numbers of oocytes suggesting resistance to cell death, which was confirmed by their decreased sensitivity to doxorubicin (Bergeron et al., 1998). Subsequent studies have extended this pro-survival phenotype of Casp2 loss to include a number of
49

tissues and DNA damaging agents (Kumar, 2009). Ei24, which was originally identified in a screen for etoposide-induced transcripts, has been shown to promote cell death by binding and sequestering Bcl2 (Zhao et al., 2005).
Interestingly, these genes as well as several previously identified miR-295 family targets are known to be directly or indirectly associated with p53. Indeed,
Pathway Analysis of well-characterized miR-295 targets brought up a single significant network ( p = 10 -14 ), “Cell Death, Cell Cycle, Cellular Function and
Maintenance,” which prominently featured p53 (Figure 6). Activation of Casp2 can occur through a protein complex in which it associates with the p53 target
PIDD (Kumar, 2009). Ei24 itself is a p53 transcriptional target, identified as one of 14 genes induced by adenoviral transfection of p53 into a p53-null colon cancer cell line (Zhao et al., 2005). We additionally confirmed two previously identified miR-295 targets, p21 (also known as Cdkn1a) and Lats2 (Wang and
Blelloch, 2009). In the case of p21, direct activation by p53 promotes cell cycle arrest at the G1/S phase (Abbas and Dutta, 2009). Lats2, while also induced by p53, exists in a positive feedback loop with p53 in which it binds and inhibits
Mdm2, thereby activating p53 (Aylon et al., 2006). Thus, miR-295 family miRNAs target a number of p53 associated genes and in all cases antagonizing p53 activation, consistent with the protective effect we have identified here.
Like p53, the mir-295 cluster affects both arms of cellular proliferation, namely cell death and cell cycle progression (Wang et al., 2008; Wang and
Blelloch, 2009). Unlike cell cycle progression, however, the anti-apoptotic role is likely to have the greatest developmental consequences following DNA damage,
50

induced physiologically by oxidative stress or metabolites. Interestingly, even simply ex vivo cell passage may be sufficient to induce a low level of stress, as evidenced by the slightly higher apoptotic rate of Dicer null cells under basal culture. In their pro-survival capacity, these miRNAs may confer robustness during embryonic development, as has been demonstrated in Drosophila . For instance, miR-7 has been shown to participate in a complex network of feedback loops to ensure proper photoreceptor cell development despite temperature fluctuations in development (Li et al., 2009). Similarly, miR-263a/b appear to prevent patterning defects in bristle formation, again consistent with the notion that they promote the fidelity of developmental trajectories ( Hilgers et al., 2010; Li et al., 2009). Early phenotypic data from mir-290-295 KO mice suggests an incompletely penetrant gestational phenotype (Medeiros et. al ., manuscript in preparation), supporting the model that loss of this cluster is tolerated in certain developmental scenarios, perhaps including those with limited stressors during gestation.
Beyond regulating development, the miRNAs described here may also have important consequences for cancer, as both Casp2 and Ei24 are considered tumor suppressors. In the case of Casp2, this has been best demonstrated in the Eu-myc lymphoma model, where loss of even a single copy of Casp2 can accelerate malignant transformation (Ho et al., 2009). Similarly,
Ei24 is found in a region that shows frequent loss-of-heterozygosity in solid tumors, and its loss has been associated with increased breast cancer invasiveness (Zhao et al., 2005). In addition, knockdown of Ei24 in mouse
51

fibroblasts or human breast cancer cell lines leads to increased resistance against etoposide-induced apoptosis (Mork et al., 2007; Zhao et al., 2005).
Consistent with these findings, the mir-295 cluster itself has been speculated to be an “oncomir” cluster, as overexpression of its human homolog, the mir-371-
373 cluster, has been found in various human tumors (Lee et al., 2009; Palmer et al., 2010; Rippe et al., 2010 ) and may promote malignant transformation
(Voorhoeve et al., 2006). Given our findings, we hypothesize that these miRNAs may have a survival promoting function with dual effects, helping cells navigate physiological stresses during development, and helping cancer cells maintain viability in the face of genotoxic chemotherapeutic agents.
In conclusion, these data expand our understanding of ESC miRNA function, linking the ES cell specific miR-295 family to key players in cell death.
Further, this analysis reveals a complex relationship between embryonic stem cell miRNA regulation and p53 activation.
Feeder-free Dicer1 flox/flox and Dicer1 -/ mouse embryonic stem cells
(mESCs) were generated and maintained on gelatin as described previously
(Calabrese et al., 2007) . mESCs cells containing a floxed and excised mir-295 cluster were generated in a similar manner and will be described in an upcoming publication (Medeiros et. al ., manuscript in preparation).
52

See Table S4.
MicroRNA mediated repression of each candidate gene was tested by cloning PCR amplified products corresponding to the entire 3' UTR downstream of a pRL-CMV Renilla luciferase reporter as described previously (Doench and
Sharp, 2004). Nucleotides 5-7 of Casp2, Bim, and Ei24 binding sites were mutated by Quickchange site-directed mutagenesis. Digests were performed using either XhoI or SalI to give the 5' site and ApaI or NotI to give the 3' site.
Firefly luciferase (pGL3) was used as a transfection control. Data shown are summaries of three or more independent trials. 24 hours before transfection 1e 5 mESC cells were plated/well of gelatinized 24-well plate. Cells were transfected with 2
µ l Lipofectamine 2000 (Invitrogen), 0.1
µ g of CMV-GFP plasmid
(Invitrogen), 0.7 µ g of pWS (carrier plasmid), and 50 nM siRNAs in 300 µ l of
Opti-MEM (Invitrogen). 4 hours after transfection, transfection mix was removed from cells and replaced with ESC media. 24 hours after transfection, cells were lysed with 1X Passive Lysis Buffer (Promega) and Dual luciferase was measured using Dual Luciferase reporter assay system (Promega) according to manufacturer’s instructions.
53

Total RNA was isolated from ES cells with or without acute Dicer deletion using Trizol (Invitrogen), following the standard protocol. Approximately 50 µ g of each RNA was loaded onto a 15% denaturing MOPS gel, according to the
Northern Blot protocol outlined previously (Seila et al., 2008). Membranes were probed for miR-292 and exposed to a phosphoimager before scanning. Prior to hybridizing with a different probe, membranes were stripped by incubating the membrane in boiling 0.1% SDS for 30 minutes and loss of signal was confirmed prior to rehybridization.
24 hours after transfection with short RNAs, Dicer1 -/, Dicer flox/flox , miR-290-
295 -/, or miR-290-295 flox/flox cells were lysed in RIPA buffer (1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, in pH 7.4 PBS) containing protease inhibitors.
30–50 µ g lysate was loaded onto 8-12% Bis-Tris gels (Invitrogen) and wettransferred at 4
°
C to Westran PVDF membranes for 2 h at 70V. After 1 h blocking at room temperature in 5% milk-TBST, membranes were probed overnight at 4 ° C with 1:2000 mouse anti-vinculin (Santa Cruz Biotechnology) or
1:200 rat anti-Caspase 2 (Millipore, 10C6). After 2x 10 min. TBST washes, membranes were probed for 1 h at room temperature with 1:2000 corresponding hRP-conjugated secondary, washed an additional 2x 10 min. in TBST, and visualized using Western Lightning Plus ECL (PerkinElmer).
54

Trizol (Qiagen) was used to extract RNA from Dicer1 flox/flox and Dicer1 -/- cells. A Superscript III kit (Invitrogen) was used to reverse transcribe 1 µ g RNA following DNAse treatment with the Turbo-DNA free kit (Ambion), and real time
PCR was performed with the primer sequences listed, using beta actin for normalization.
24 hours before transfection 2e 5 mESC cells were plated/well of gelatinized 12-well plates. Cells were transfected with 4
µ l Lipofectamine 2000
(Invitrogen), 0.2
µ g pCAGGS-mCherry plasmid, 1.4
µ g of pWS, and 50 nM of siRNA in 600 µ l of Opti-MEM (Invitrogen). 4 hours after transfection, transfection mix was removed from cells and replaced with ESC media. 24 hours after transfection, cells were exposed to 5-Gy gamma radiation or 100 nM doxorubicin.
Immediately after exposure, one plate of cells were trypsinized and fixed with 1x
BD Perm buffer. Cells were stained with Rabbit Anti-Casp3 antibody (BD
Biosciences) at 1:100 for 20 min. at room temperature. Following washing, cells were incubated with Alexa-488-conjugated secondary antibody (diluted 1:250)
(Invitrogen) for 60 min. at room temperature, washed, and resuspended in BD
FACS buffer containing 1:5000 Hoechst stain. 24 hours after the treatment, another plate of cells was trypsinized and treated with the same protocol for
FACS analysis.
55

Casp3 assays were also performed on Dcr KO and WT mESCs without transfection. 24 hours before collecting cells for the 0 h time point for Casp3 assay, 2e 5 mESC were plated/well of gelatinized 6-well plates. In the context of genotoxic stress, 4e 5 mESCs were plated/well of gelatinized 6-well plates. 24 hours after plating, cells were treated with 5-Gy radiation or 100 nM doxorubicin.
Casp3 assays were performed at 0 h and 24 h after the treatment following the same protocol described above.
4e 5 mESCs were plated/well of gelatinized 6-well plates. 24 h after plating, cells were exposed to 100 nM doxorubicin. Cells were trypsinized 0 h and 24 h after the treatment for Annexin V detection, following Annexin V-FITC apoptosis detection kit (BD Biosciences).
Microarray analysis was performed 5 days following transfection of
Dicer flox/flox wild-type cells with either GFP alone or GFP and Cre recombinase, and data were analyzed using biological triplicates. Microarrays for the mir-295 cluster deletion were performed on two deletion and two wild-type lines independently derived. Spot replicates were condensed using geometric means.
The log fold change (LFC) value for Dcr WT/Dcr KO was defined as the difference between the mean log expression in Dcr WT cells and the mean log
56

expression in Dcr KO cells. The conserved set of targets were downloaded from
TargetScanMouse5.1 website (http://www.targetscan.org/mmu_50/). To identify targets predicted for the AAGUGC seed family, we looked at all miRNAs that contain AAGUGC in their seed region. More specifically, they include “miR-291b-
3p/519a/519b-3p/519c-3p”, “miR-290-3p/292-3p/467a”, “miR-467cd”, “miR-
106/302”, and “miR-467b”. We excluded all the targets of “miR-302ac/520f”, as well as T1A 7mer targets of “miR-467b”, as they do not contain the 6-mer match to AAGUGC. Targets with top 10% of branch length scores were considered
“conserved”.
Gene Set Analysis Toolkit (http://bioinfo.vanderbilt.edu/webgestalt/) was used to perform GO analysis. Targets and controls were generated as described in the text. Network data were analyzed through the use of Ingenuity Pathways
Analysis (Ingenuity® Systems, www.ingenuity.com). Ingenuity Pathway Analysis
(IPA) was performed on the set of validated miR-295 targets to identify the most strongly associated canonical pathways.
All test statistics were calculated using R (http://www.r-project.org). The
Wilcoxon rank sum test was used because it does not assume normality of the underlying distributions. T-tests and Kolmogorov–Smirnov (KS) test using these data gave generally similar results.
57

We thank J.R. Neilson and A. Seila for experimental assistance and reagents, and members of the Jaenisch, Burge, and Sharp laboratories for helpful discussions. AR was funded by a Fannie and John Hertz Foundation
Fellowship. This work was supported by NIH grants RO1-GM34277, NCI grant
PO1-CA42063, and the NCI Cancer Center Support (core) grant P30-CA14051.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
58

Figure 1. Acute Dicer deletion and specific mir-295 cluster deletion show global target derepression signatures. A.
PCR for confirmation of Dicer loss on acute deletion. Dicer floxed and floxed-deleted bands were amplified over a nine day time course to determine the point of maximal enrichment for Dicer null cells. B.
Northern blot for mature miRNAs following Dicer deletion. A DNA probe for the abundant miR-292 was assessed at 5, 7, and 9 days following Dicer deletion, with a probe for U6 to control for loading. C.
Cumulative distribution functions (cdfs) of log
2
fold change (LFC) in mRNA expression between Dcr WT and Dcr KO ES cells (top panel) and 295 WT and 295 KO ES cells (bottom panel) are plotted. Plots include conserved targets (red line), all predicted miR-
295 targets (blue line), and control mRNAs (grey line). targets include ~3000 predicted TargetScan targets of the miR-295 cluster. Conserved targets contains the ~300 genes in the top 10% of targets ranked by TargetScan 5.1 branch length scores. The control mRNA set was selected to match the predicted targets in expression, 3' UTR length and composition. Targets are derepressed in both
Dcr KO as well as 295 KO mESCs ( p
"
2.2e
-16 and p = 9.7e
-9 by rank sum test, respectively). Similarly, conserved targets are derepressed in both Dcr KO as well as 295 KO mESCs ( p " 2.2e
-16 and p = 3.1e
-14 by rank sum test, respectively).
59

Figure 2. Microarray data and Targetscan predictions identify candidate miR-295 targets in mESCs. A. Venn diagram of microarray and target prediction data used to generate mir-295 cluster target candidates. “mir-295 cluster predicted targets” = predicted TargetScan targets of the miR-295 cluster;
“295 KO” = genes that showed a 1.2 fold upregulation on mir-295 cluster loss;
“Dcr KO” = genes that showed a 1.2 fold upregulation on Dicer loss. Only ESexpressed genes (i.e. genes with an expression of at least 16 on the wild type arrays) were considered for analysis. B.
Activity of luciferase reporters of predicted targets of the mir-295 cluster were assayed in Dcr WT and Dcr KO ES cells. Luciferase reporters contain full length 3' UTRs of predicted targets.
Relative luciferase activity is the ratio of the reporter’s activity in WT ES cells and
Dcr KO ES cells.
Figure 3. The pro-apoptotic genes Caspase 2 (Casp2) and Ei24 are direct targets of the mir-295 cluster. A. RT-PCR of Casp2 mRNA in Dcr KO cells. B.
Western blot of Casp2 in Dcr WT and Dcr KO ES cells. 50 nM miR-295, miR-
467a, Bim siRNA, Casp2 siRNA and Ei24 siRNA were transfected into Dcr KO
ES cells, and Casp2 protein expression was assayed 24 hours later. miR-467a shares the same hexamer seed with miR-295, and Bim siRNA and Ei24 siRNA served as negative controls. C.
Luciferase reporters with a full length Casp2 3'
UTR or Ei24 3' UTR, as well as their seed mutant versions were assayed by comparing their activities in Dcr WT and Dcr KO ES cells. Casp2:4M has all 4
AAGUGC seed binding sites mutated, and Ei24:1M has its single AAGUGC seed
60

binding site mutated. 20 nM miR-295 was transfected into Dcr KO ES cells and the relative activity measured versus non-transfected cells to test if the repression of luciferase reporters is specifically due to AAGUGC miRNAs. D.
Casp2 luciferase reporters bearing different combinations of AAGUGC seed binding site mutations were tested in Dcr WT and Dcr KO ES cells. Casp2:2 , 2nd
AAGUGC binding site was mutated; Casp2:3,4 , 3rd and 4th binding sites were mutated; Casp2: 2,3,4 , 2nd, 3rd, and 4th binding sites were mutated; Casp2:1,2 ,
1st and 2nd binding sites were mutated; Casp2: 1,2,3,4 , all binding sites were mutated. n # 3 for all experiments, and results are shown as mean ± S.E.M.
(standard error of the mean). P-values were obtained by t-tests, * denotes p "
0.05, and ** denotes p " 0.01.
Figure 4. Downregulation of Casp2 and Ei24 partially rescues the increased apoptotic rate of Dcr KO cells following genotoxic stress. A. The percentage of cells expressing cleaved Caspase 3 (Casp3) in Dcr WT and Dcr KO ES cells after exposure to 5-Gy radiation (0 and 24 hours after radiation treatment).
Cleaved Casp3 was assayed by flow cytometry, and was used to estimate apoptotic response. Apoptosis rate of Dcr KO cells is shown in black bars, and that of WT cells is shown in white bars. B. Dcr KO cells were treated with 5-Gy radiation 24 hours after transfection of 50 nM miR-295, miR-290-3p, Bim siRNA,
Casp2 siRNA or Ei24 siRNA. Casp3 activity was assayed 0 and 24 h after the treatment, and the difference in apoptosis rate is shown. Transfection of seed mutants and control siRNAs into Dcr KO cells, and transfection of control siRNAs
61

into WT cells served as controls. C,D. A similar series of experiments was performed in Dcr WT and Dcr KO cells using 100 nM doxorubicin. n
#
3 for all experiments. Results are shown as mean
±
S.E.M. P-values were obtained by
Mann-Whitney tests, * denotes p
"
0.05, and ** denotes p
"
0.01.
Figure 5. Loss of the mir-295 cluster derepresses Casp 2 and Ei24 3' UTRs and enhances sensitivity to DNA damaging agents. A. Luciferase reporters with full length versions of the Casp2 3' UTR or Ei24 3' UTR, as well as their seed mutant versions were assayed in 295 WT and 295 KO ES cells.
B. Western blot of Casp2 in 295 WT and 295 KO ES cells. 50 nM miR-295, miR-467a, and miR-20a were transfected into 295 KO ES cells, and Casp2 protein expression was assayed 24 hours after the transfection. C. The percentage of cells with cleaved Casp3 in 295 WT and 295 KO ES cells after exposure to 5-Gy radiation
(0 and 24 hours after radiation treatment). Cleaved Casp3 was assayed by flow cytometry, and was used to estimate apoptotic response. Apoptosis of 295 KO cells is shown in black bars, and that of 295 WT cells is shown in white bars.
D.
Cells were transfected with 50 nM siRNAs as shown, and difference in apoptotic response of 295 WT and 295 KO ES cells 24 hours after exposure to 5-Gy radiation was plotted.
E,F. A similar series of experiments was performed on 295
WT and 295 KO cells after exposure to 100 nM doxorubicin. n
#
3 for all experiments. Results are shown as mean
±
S.E.M. (standard error of the mean).
P-values were results of Mann-Whitney tests, and * denotes p " 0.05, and ** denotes p
"
0.01.
62

Figure 6. Ingenuity Pathway Analysis (IPA) of miR-295 targets. IPA was performed for validated targets of miR-295 from this and prior studies (Benetti et al., 2008; Sinkkonen et al., 2008; Wang et al., 2008), and identified the network,
“Cell Death, Cell Cycle, Cellular Function and Maintenance,” which centers around p53. Solid lines indicate direct interactions and dashed lines indicate indirect interactions. Validated targets of miR-295 are shown in red and computationally predicted targets are shown in orange.
Table S1. Sequences and expression level of the mir-295 cluster in ES cells. The 6-mer seed is highlighted in bold. The cloning statistics were taken from previously published studies (Babiarz et al., 2008; Ciaudo et al., 2009;
Leung , 2010 ).
Table S2. Sequences and expression level of the mir-302, mir-467, and mir-
17-92 clusters in ES cells. The 6-mer seed is highlighted in bold. The cloning statistics were taken from previously published studies (Babiarz et al., 2008;
Ciaudo et al., 2009; Leung , 2010 ).
Table S3. Predicted targets of the mir-295 cluster.
Table S4. Oligos and siRNAs used in all experiments.
63

Figure S1. Microarray data from polyclonal acute dicer deletion samples show better inter-sample correlation than data from clonal chronic dicer deletion lines. Biological triplicates were obtained for both 5 days post Dicer deletion (Acute 1-3) and over 1 month following Dicer deletion (Chronic 1-3), after which samples were normalized together. Pearson correlations for all pairwise comparisons within the two conditions are shown here.
Figure S2. Repression of predicted targets of the miR-295 cluster in Dcr WT and Dcr KO ES cells. Activity of luciferase reporters of predicted targets were assayed in WT, Dcr KO ES cells, as well as in Dcr KO ES cells after over expression of 20 nM miR-295. n " 2, and results are shown as mean ± S.E.M.
Figure S3. Comparison of WT and KO cells’ apoptosis response to stress.
A. The percentage of cells expressing cleaved Caspase 3 (Casp3) in WT and
Dcr KO ES cells under normal culturing conditions (0 and 24 hours after plating).
B. The percentage of cells expressing cleaved Caspase 3 (Casp3) in 295 WT and 295 KO ES cells under normal culturing conditions (0 and 24 hours after plating). Cleaved Casp3 was assayed by flow cytometry, and was used to estimate apoptosis response. Apoptosis of KO cells is shown in black bars, and that of WT cells is shown in white bars.
C. Annexin V positive cells were assayed by flow cytometry immediately or 24 h after exposure to 100 nM doxorubicin. n =
2, and results are shown as mean
±
S.E.M. (standard error of the mean).
D.
64

Northern analysis for miR-295 in Dcr KO ES cells, WT ES cells, and WT ES cells
6 hours after 2
µ
M doxorubicin treatment.
Figure S4. Time course of WT and KO cells’ apoptosis response to stress.
A. WT and Dcr KO ES cells were treated with 5-Gy radiation. Casp3 activity was assayed 0 h, 10 h, and 24 h after the treatment, and the differences in apoptosis rate (between 10 h and 0 h, and between 24 h and 0 h) are shown. Cleaved
Casp3 was assayed by flow cytometry, and was used to estimate apoptosis response. Apoptosis rate of Dcr KO cells is shown in black bars, and that of WT cells is shown in white bars. n = 2, and results are shown as mean ± S.E.M.
B.
WT and Dcr KO cells were treated with 2 doses of doxorubicin (200 nM and 300 nM). Casp3 activity was assayed 0 h, 10 h, and 24 h after the treatment, and the differences in apoptosis rate (between 10 h and 0 h, and between 24 h and 0 h) are shown. n = 1 with 3 technical replicates, and results are shown as mean
± standard deviation.
Figure S5. Analysis of Caspase 2 and Caspase 3 cleavage upon DNA damage induction. A. Dcr KO cells treated with 2 µ M doxorubicin for 6 hours.
Blot was probed with an antibody specific for cleaved Caspase 3 and for its target, Nanog, before and after treatment.
B. Dcr KO and Dcr WT cells following
2
µ
M doxorubicin treatment for 6 hours. Intact Caspase 2 (49 kD) and cleaved
Caspase 2 (35 kD) are both shown.
65

Figure S6. Raw data of Dcr KO and WT cells’ apoptosis response after treatment of 5-Gy radiation or 100 nM doxorubicin. A. Dcr KO cells were treated with 5-Gy radiation 24 hours after transfection of 50 nM miR-295 or miR-
290-3p. Caspase 3 activity was assayed 0 and 24 h after the treatment.
Transfection of seed mutants and control siRNAs (50 nM) into Dcr KO cells, and overexpression of control siRNAs (50 nM) into WT cells served as controls.
B.
Dcr KO cells were treated with 5-Gy radiation 24 hours after transfection of 50 nM siRNAs against Bim, Casp2, and Ei24. Caspase-3 activity was assayed 0 and 24 h after the treatment. C,D. A similar series of experiments was performed in Dcr WT and Dcr KO cells using 100 nM doxorubicin. n
#
3 for all experiments.
Results are shown as mean
±
S.E.M. (standard error of the mean).
Figure S7. Raw data of 295 KO and 295 WT cells’ apoptosis response after treatment of 5-Gy radiation or 100 nM doxorubicin. A. 295 KO cells were treated with 5-Gy radiation 24 hours after transfection of 50 nM of miR-295 or miR-290-3p. Caspase 3 activity was assayed 0 and 24 h after the treatment.
Transfection of seed mutants and control siRNAs (50 nM) into 295 KO cells, and overexpression of control siRNAs (50 nM) into 295 WT cells served as controls.
B. 295 KO cells were treated with 5-Gy radiation 24 hours after transfection of 50 nM siRNAs against Bim, Casp2, and Ei24. Caspase-3 activity was assayed 0 and 24 h after the treatment. C,D. A similar series of experiments was performed in 295 WT and 295 KO cells using 100 nM doxorubicin. n # 3 for all experiments.
Results are shown as mean ± S.E.M. (standard error of the mean).
66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

Abbas, T., and Dutta, A. (2009). p21 in cancer: intricate networks and multiple activities. Nat Rev Cancer 9, 400-414.
Aylon, Y., Michael, D., Shmueli, A., Yabuta, N., Nojima, H., and Oren, M. (2006).
A positive feedback loop between the p53 and Lats2 tumor suppressors prevents tetraploidization. Genes Dev 20, 2687-2700.
Babiarz, J. E., Ruby, J. G., Wang, Y., Bartel, D. P., and Blelloch, R. (2008).
Mouse ES cells express endogenous shRNAs, siRNAs, and other
Microprocessor-independent, Dicer-dependent small RNAs. Genes Dev
22, 2773-2785.
Ben-Porath, I., Thomson, M. W., Carey, V. J., Ge, R., Bell, G. W., Regev, A., and
Weinberg, R. A. (2008). An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet 40,
499-507.
Benetti, R., Gonzalo, S., Jaco, I., Munoz, P., Gonzalez, S., Schoeftner, S.,
Murchison, E., Andl, T., Chen, T., Klatt, P., et al. (2008). A mammalian microRNA cluster controls DNA methylation and telomere recombination via Rbl2-dependent regulation of DNA methyltransferases. Nat Struct Mol
Biol 15, 268-279.
Bergeron, L., Perez, G. I., Macdonald, G., Shi, L., Sun, Y., Jurisicova, A.,
Varmuza, S., Latham, K. E., Flaws, J. A., Salter, J. C., et al. (1998).
Defects in regulation of apoptosis in caspase-2-deficient mice. Genes Dev
12, 1304-1314.
Calabrese, J. M., Seila, A. C., Yeo, G. W., and Sharp, P. A. (2007). RNA sequence analysis defines Dicer's role in mouse embryonic stem cells.
Proc Natl Acad Sci U S A 104, 18097-18102.
Calabrese, J. M., and Sharp, P. A. (2006). Characterization of the short RNAs bound by the P19 suppressor of RNA silencing in mouse embryonic stem cells. RNA 12, 2092-2102.
Ciaudo, C., Servant, N., Cognat, V., Sarazin, A., Kieffer, E., Viville, S., Colot, V.,
Barillot, E., Heard, E., and Voinnet, O. (2009). Highly dynamic and sexspecific expression of microRNAs during early ES cell differentiation.
PLoS Genet 5, e1000620.
Doench, J. G., and Sharp, P. A. (2004). Specificity of microRNA target selection in translational repression. Genes Dev 18, 504-511.
Friedman, R. C., Farh, K. K., Burge, C. B., and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res
19, 92-105.
Fujita, J., Crane, A. M., Souza, M. K., Dejosez, M., Kyba, M., Flavell, R. A.,
Thomson, J. A., and Zwaka, T. P. (2008). Caspase activity mediates the differentiation of embryonic stem cells. Cell Stem Cell 2, 595-601.
Fulda, S., and Debatin, K. M. (2006). Extrinsic versus intrinsic apoptosis pathways in anticancer chemotherapy. Oncogene 25, 4798-4811.
86

Grimson, A., Farh, K. K., Johnston, W. K., Garrett-Engele, P., Lim, L. P., and
Bartel, D. P. (2007). MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell 27, 91-105.
Gu, Z., Flemington, C., Chittenden, T., and Zambetti, G. P. (2000). ei24, a p53 response gene involved in growth suppression and apoptosis. Mol Cell
Biol 20, 233-241.
Harfe, B. D., McManus, M. T., Mansfield, J. H., Hornstein, E., and Tabin, C. J.
(2005). The RNaseIII enzyme Dicer is required for morphogenesis but not patterning of the vertebrate limb. Proc Natl Acad Sci U S A 102, 10898-
10903.
Hilgers, V., Bushati, N., and Cohen, S. M. (2010). Drosophila microRNAs 263a/b confer robustness during development by protecting nascent sense organs from apoptosis. PLoS Biol 8, e1000396.
Ho, L. H., Taylor, R., Dorstyn, L., Cakouros, D., Bouillet, P., and Kumar, S.
(2009). A tumor suppressor function for caspase-2. Proc Natl Acad Sci U
S A 106, 5336-5341.
Kanellopoulou, C., Muljo, S. A., Kung, A. L., Ganesan, S., Drapkin, R., Jenuwein,
T., Livingston, D. M., and Rajewsky, K. (2005). Dicer-deficient mouse embryonic stem cells are defective in differentiation and centromeric silencing. Genes Dev 19, 489-501.
Kumar, S. (2009). Caspase 2 in apoptosis, the DNA damage response and tumour suppression: enigma no more? Nat Rev Cancer 9, 897-903.
Lee, K. H., Goan, Y. G., Hsiao, M., Lee, C. H., Jian, S. H., Lin, J. T., Chen, Y. L., and Lu, P. J. (2009). MicroRNA-373 (miR-373) post-transcriptionally regulates large tumor suppressor, homolog 2 (LATS2) and stimulates proliferation in human esophageal cancer. Exp Cell Res 315, 2529-2538.
Leung, A. G. Y., Arjun Bhutkar, Grace X Zheng, Andrew D Bosson, Cydney B
Nielsen, Phillip A Sharp (2010). Genome-wide identification of Ago2 binding sites from mouse embryonic stem cells with and without mature microRNAs. Nat Struct Mol Bio.
Leung, A. K., Calabrese, J. M., and Sharp, P. A. (2006). Quantitative analysis of
Argonaute protein reveals microRNA-dependent localization to stress granules. Proc Natl Acad Sci U S A 103, 18125-18130.
Li, J., and Yuan, J. (2008). Caspases in apoptosis and beyond. Oncogene 27,
6194-6206.
Li, X., Cassidy, J. J., Reinke, C. A., Fischboeck, S., and Carthew, R. W. (2009).
A microRNA imparts robustness against environmental fluctuation during development. Cell 137, 273-282.
Mork, C. N., Faller, D. V., and Spanjaard, R. A. (2007). Loss of putative tumor suppressor EI24/PIG8 confers resistance to etoposide. FEBS Lett 581,
5440-5444.
Murchison, E. P., Partridge, J. F., Tam, O. H., Cheloufi, S., and Hannon, G. J.
(2005). Characterization of Dicer-deficient murine embryonic stem cells.
Proc Natl Acad Sci U S A 102, 12135-12140.
87

Nakagawa, A., Shi, Y., Kage-Nakadai, E., Mitani, S., and Xue, D. (2010).
Caspase-dependent conversion of Dicer ribonuclease into a deathpromoting deoxyribonuclease. Science 328, 327-334.
Nielsen, C. B., Shomron, N., Sandberg, R., Hornstein, E., Kitzman, J., and
Burge, C. B. (2007). Determinants of targeting by endogenous and exogenous microRNAs and siRNAs. RNA 13, 1894-1910.
Nitiss, J. L. (2009). Targeting DNA topoisomerase II in cancer chemotherapy. Nat
Rev Cancer 9, 338-350.
Palmer, R. D., Murray, M. J., Saini, H. K., van Dongen, S., Abreu-Goodger, C.,
Muralidhar, B., Pett, M. R., Thornton, C. M., Nicholson, J. C., Enright, A.
J., and Coleman, N. (2010). Malignant germ cell tumors display common microRNA profiles resulting in global changes in expression of messenger
RNA targets. Cancer Res 70, 2911-2923.
Rippe, V., Dittberner, L., Lorenz, V. N., Drieschner, N., Nimzyk, R., Sendt, W.,
Junker, K., Belge, G., and Bullerdiek, J. (2010). The two stem cell microRNA gene clusters C19MC and miR-371-3 are activated by specific chromosomal rearrangements in a subgroup of thyroid adenomas. PLoS
One 5, e9485.
Seila, A. C., Calabrese, J. M., Levine, S. S., Yeo, G. W., Rahl, P. B., Flynn, R. A.,
Young, R. A., and Sharp, P. A. (2008). Divergent transcription from active promoters. Science 322, 1849-1851.
Sinkkonen, L., Hugenschmidt, T., Berninger, P., Gaidatzis, D., Mohn, F., Artus-
Revel, C. G., Zavolan, M., Svoboda, P., and Filipowicz, W. (2008).
MicroRNAs control de novo DNA methylation through regulation of transcriptional repressors in mouse embryonic stem cells. Nat Struct Mol
Biol 15, 259-267.
Tichy, E. D., and Stambrook, P. J. (2008). DNA repair in murine embryonic stem cells and differentiated cells. Exp Cell Res 314, 1929-1936.
Tyas, L., Brophy, V. A., Pope, A., Rivett, A. J., and Tavare, J. M. (2000). Rapid caspase-3 activation during apoptosis revealed using fluorescenceresonance energy transfer. EMBO Rep 1, 266-270.
Upton, J. P., Austgen, K., Nishino, M., Coakley, K. M., Hagen, A., Han, D., Papa,
F. R., and Oakes, S. A. (2008). Caspase-2 cleavage of BID is a critical apoptotic signal downstream of endoplasmic reticulum stress. Mol Cell
Biol 28, 3943-3951.
Voorhoeve, P. M., le Sage, C., Schrier, M., Gillis, A. J., Stoop, H., Nagel, R., Liu,
Y. P., van Duijse, J., Drost, J., Griekspoor, A., et al. (2006). A genetic screen implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Cell 124, 1169-1181.
Wang, Y., Baskerville, S., Shenoy, A., Babiarz, J. E., Baehner, L., and Blelloch,
R. (2008). Embryonic stem cell-specific microRNAs regulate the G1-S transition and promote rapid proliferation. Nat Genet 40, 1478-1483.
Wang, Y., and Blelloch, R. (2009). Cell cycle regulation by MicroRNAs in embryonic stem cells. Cancer Res 69, 4093-4096.
88

Wernig, M., Meissner, A., Foreman, R., Brambrink, T., Ku, M., Hochedlinger, K.,
Bernstein, B. E., and Jaenisch, R. (2007). In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature 448, 318-324.
Zhao, X., Ayer, R. E., Davis, S. L., Ames, S. J., Florence, B., Torchinsky, C.,
Liou, J. S., Shen, L., and Spanjaard, R. A. (2005). Apoptosis factor
EI24/PIG8 is a novel endoplasmic reticulum-localized Bcl-2-binding protein which is associated with suppression of breast cancer invasiveness. Cancer Res 65, 2125-2129.
89

The material in this chapter was adapted from the following publication:
Grace X.Y. Zheng, Arvind Ravi, Genevieve M. Gould, Christopher B. Burge, Phillip A.
(2011). Genome-wide impact of a novel rapidly expanded microRNA cluster in mouse.
(in review)
Experimental contributions:
This work represents an equal collaboration between Grace Zheng and Arvind Ravi.
Arvind Ravi and Grace Zheng designed the evolutionary analysis.
Arvind Ravi performed cluster reconstruction and target validation.
Grace Zheng performed the informatic targeting analysis and cell culture assays.
90

Variations in microRNA (miRNA) gene and/or target repertoire are likely to be key drivers of phenotypic differences between species. To better understand these changes, we developed a computational method that identifies signatures of species-specific target site gain and loss associated with miRNA acquisition.
Interestingly, several of the miRNAs that show corresponding evolutionary signals in mouse 3' UTRs derive from a single rapidly expanded rodent-specific miRNA cluster. Located in the intron of Sfmbt2 , a maternally imprinted polycomb gene, these miRNAs (referred to as the Sfmbt2 cluster) are expressed in both embryonic stem cells (ESCs) and the placenta. One abundant miRNA from the cluster, miR-467a, functionally overlaps with the mir-290-295 cluster in promoting growth and survival of mouse ESCs (mESCs). Predicted novel targets of the remaining cluster members are enriched in pathways regulating cell survival.
Two relevant species-specific target candidates, Lats2 and Dedd2, were validated in cultured cells. We suggest that the rapid evolution of the Sfmbt2 cluster may be a result of intersex conflict for growth regulation in early mammalian development, and could provide a general model for the genomic response to acquisition of miRNAs and similar regulatory factors.
91

The emergence of novel regulatory interactions provides a critical means of evolutionary change (King and Wilson, 1975). By introducing new regulatory elements, or simply rewiring existing elements, organisms can adapt to alterations in their environment. In the case of protein coding genes, a number of precedents for these principles have been established. For instance, the transcriptional cofactor TAFII105 emerged in mammals to specifically direct expression of a subset of genes in ovarian follicle cells (Levine and Tjian, 2003).
Similarly, the remapping of existing transcription-factor networks through promoter evolution is thought to be widespread, even between similar species
(Bradley et al., 2010; Chen and Rajewsky, 2007; Ruby et al., 2007). However, given the relatively small increases in protein-coding genes across more complex organisms, many have turned their attention to the roles of noncoding RNAs in explaining such evolutionary changes (Mattick and Makunin, 2006).
Among noncoding RNAs, miRNAs are thought to be particularly relevant to phenotypic differences between species, with some claiming that miRNA gene number scales roughly with organismal complexity (Niwa and Slack, 2007).
Although only ~22 nucleotides in length, miRNAs can repress the expression of hundreds of genes post-transcriptionally, making them ideal candidates for the establishment or alteration of large regulatory networks. Indeed, it has been suggested that miRNAs are in fact more “evolvable” elements than transcription factors because targeting of a novel sequence requires changing only one or a
92

few bases rather than a complex set of amino acid changes (Chen and
Rajewsky, 2007). However, the constraints of processing require that precursors be present as hairpin structures in the genome, therefore favoring their emergence via certain evolutionary routes. Three mechanisms in particular have been hypothesized for miRNA generation: 1) duplications leading to additional copies of existing miRNAs; 2) processing of transposable elements that contain terminal inverted repeats; or 3) processing of hairpin structures generated by mutational processes ( Nozawa et al., 2010 ). Following gene duplication in the first mechanism, subsequent base substitutions could produce changes in targeting, particularly those corresponding to positions 2-7 from the 5' end of the miRNA, known as the miRNA “seed” (Bartel, 2009).
In contrast to miRNA evolution, which simultaneously introduces a multitude of novel target interactions, changes in single target sites can provide more modest increments of evolutionary change. Nevertheless, these changes could also be important drivers of organismal differences. For instance, variations in Nodal family targeting by the miR-430/427/302 family guide differences in germ layer specification during development across a range of vertebrates (Rosa et al., 2009). Even within a species, presence or absence of even a single target interaction may have notable effects. For instance, in the case of Texel sheep, a forward genetic screen identified a single base change in the myostatin 3' UTR that creates a miRNA target site and confers muscular hypertrophy (Clop et al., 2006). In humans, a similar case has been reported for
Tourette’s Syndrome, where changes in the SLITRK1 transcript enhance
93

repression by hsa-miR-189 (Abelson et al., 2005). A more global assessment using human SNP datasets identified a number of variants that may alter miRNA binding (Chen and Rajewsky, 2006). This study suggested that many target sites in the human genome are under purifying selection to maintain the presence or absence of a miRNA target site, but only identified a single instance of positive selection, presumably because few SNPs passed the high heterozygosity thresholds required for informative intraspecies analysis.
Here we describe a novel method for identifying species-specific changes in miRNA-target relationships, and apply it to identify miRNA innovations in the mouse genome. We find that many of the mouse-specific changes correspond to a single genomic locus, located in the intron of an imprinted polycomb group gene (Klymenko et al., 2006). The miRNAs in this cluster, which we refer to as the Sfmbt2 cluster, appear to have expanded through a duplication–divergence mechanism, generating both novel seeds and seeds corresponding to earlier miRNA families. Finally, predictions suggest that these miRNAs may in part regulate targets involved in growth and survival, in line with predicted roles for the mir-290-295 cluster, the dominant cluster in mouse ES cells. As the expression patterns of these two clusters appear to mirror one another, we suggest that the Sfmbt2 cluster may promote proliferation in extra-embryonic tissue, serving as a counterpart to the mir-290-295 cluster in early murine development.
94

To better understand how gene networks respond to introduction of a novel miRNA, we examined the 3' UTR landscape of the mouse genome in search of signatures that represent species-specific responses to different miRNA repertoires, using relative sequence conservation with humans as a background for comparison. We hypothesized that comparative analysis of 3'
UTRs would reveal two groups of genes associated with species-specific miRNAs: 1) genes that gained binding sites to the miRNAs in their 3' UTRs, and whose downregulation improved fitness; and 2) genes that were inadvertently targeted by novel miRNAs, and that subsequently lost target sites to maintain expression levels in certain cellular states.
To test for the presence of each group of genes, aligned 3' UTRs from the mouse and human genomes were analyzed for signatures of target site gain and loss (Figure 1A). In the case of target site gain, we considered seed-match sites present in mouse 3' UTRs, whereas for target site loss, we considered the reverse scenario, requiring a seed match site in human but not necessarily in mouse. Functional species-specific miRNAs would be expected to generate both gain and loss signatures. Although sites with imperfect seed matches have been described, including compensatory downstream matches (Bartel, 2009) and centered pairing sites (Shin et al., 2010), we defined predicted targets using the
95

traditional method of seed complementarity (7mer-M8 type sites), as it is the most common type of target site and the best characterized to date (Bartel,
2009). For miRNA targets that have been positively selected to gain a target site, we would expect higher nucleotide variation between species at the miRNA binding site than at neighboring sequences in the same 3' UTR (Figure 1A, B, left panels). To measure this difference, we defined the statistic d as the difference in average per base mutation frequency between the miRNA target site and the frequency in the adjacent 42 nt upstream and 42 nt downstream of the target site
(Figure 1A, B, middle panels). Significance of selection was then assessed by comparing d for a given seed to that of control oligonucleotides with similar abundance in mouse and human 3' UTRs (Figure 1A, B, right panels). Using this approach, 25 of 201 mouse-specific miRNA seeds (12.4%) showed a signal for gain of target sites (Figure 2A, Figure S1, Table 1), roughly twice the rate observed for all heptamers (6.02%, p = 0.004, Chi-square test). Thus, our analysis detected evidence of selection for acquisition of seed matches to these mouse-specific miRNAs in the mouse genome.
We additionally tested for site loss. Again, we found a number of miRNAs
(16/201 = 7.96%) whose target sites were lost at greater frequencies in mouse 3'
UTRs than controls, but this fraction was not significantly higher than the fraction observed for control heptamers (6.02%) (Figure 2A, Table 1). Of these 16, a significant fraction, 13, overlapped with those seeds showing site gain signatures above (13/201, p < 0.0001, Fisher’s exact test). Target sites of the 7mer-M8 type generally confer somewhat greater repression at the mRNA level than 7mer-A1
96

sites. Similar results were obtained when considering 7mer-A1 miRNA targets, albeit with a weaker signal (Table S1). Human-specific gain/loss events were not statistically enriched over the frequency seen for control heptamers, which may reflect the less refined miRNA annotations presently available in humans (Table
S2).
As a validation of our method, we tested the behavior of target sites for miRNAs known to be conserved between mouse and human, predicting that these would show trends opposite those observed for nonconserved miRNAs, namely a decreased mutation rate at the seed binding site relative to adjacent sequence (negative d ) (Figure 1C). Using these criteria, of 218 miRNA seeds that are shared between human and mouse, target sites of 44 (20.18%) seeds showed a significant purifying selection signal (Figure S1, Table S3 and Table
S4), a ratio substantially higher than that observed for all heptamers (4.79%, p =
0.001, Chi-square test). These 44 miRNAs have an average signal-tobackground ratio for site conservation of 3.39:1 based on previous calculations
(Friedman et al., 2009), confirming the ability of our method to recognize genomic signatures of selection. As expected, a comparison of the distribution of d for mouse miRNAs showing a target conservation signature between mouse and human showed a downward skew in seed match mutation rate relative to those with targets undergoing positive selection (Figure S1).
97

Sfmbt2
We next evaluated whether the above 28 seeds (the unique set showing either gain or loss signals above) had any interesting contextual relationships in the mouse genome, so that we could better discern evolutionary “hot spots” for miRNA diversity. Interestingly, when the 28 miRNAs with significant target site selection were plotted on a chromosome map of the mouse genome, a prominent clustering was apparent on chromosome 2. Five additional regions on chromosomes 2, 3, 4, 12, and X also contained miRNAs significant for both gain and loss of target sites, suggesting that these miRNAs may be functional (Figure
2B). The chromosome 12 miRNAs are part of a cluster of 40 miRNAs near the imprinted Dlk-Gtl2 region (Seitz et al., 2004). These miRNAs show preferential expression in the placenta and brain, and have been suggested to have key roles in embryonic development and neurogenesis (Glazov et al., 2008). Examination of miRNA expression data revealed that the remaining miRNAs outside of chromosome 12 with both gain and loss signals are expressed in germ tissue
(miR-511) or during early embryonic development (miR-1198-5p, miR-3094, miR-
380, miR-466k) (Chiang et al., 2010), although detailed characterization of their expression and functions is minimal.
Closer inspection of the largest cluster of hits, located proximally on chromosome 2, revealed that they were derived from 9 th intron of Sfmbt2 , a polycomb group gene (Kuzmin et al., 2008). In all, this intron contains 36 distinct miRNAs, which we refer to as the Sfmbt2 cluster miRNAs (Figure 3A and Table
98

S5). Although the coding region of the Sfmbt2 gene is highly conserved among vertebrates, the intron that harbors the miRNA cluster bears little similarity to those outside of rodent species. Two of the 36 mouse Sfmbt2 miRNAs can be mapped to the homologous intron in rat, but none can be aligned to the corresponding intron in human. Consistent with these findings, expression of the
Sfmbt2 cluster – cloned from both T cells (Neilson et al., 2007) and embryonic stem cells (Calabrese et al., 2007) – has only been detected in mouse tissue.
To better understand the nature of the expansion, we more closely examined the 50 kb intronic sequence containing these miRNA genes. Largely consisting of simple repeats and B4 repeats - one of four SINE families whose expansion marks the mouse genome (Waterston et al., 2002) – the cluster is likely to have derived from a seeding transposition event, followed by expansion through further segmental duplications. The retention of the flanking regions of the intronic expansion in rat suggests a two-step expansion, with the first phase occurring in a common rodent ancestor, and the second involving mouse-specific expansions of an internal 2 kb tiling unit. Consistent with this model, the miRNAs contained within the cluster can be classified into a handful of broad groups based on sequence similarity: miR-297s, miR-466s, miR-467s, and miR-669s. Of these four classes, members of the miR-297s and miR-467s can be considered miRNA “families” as they largely consist of the seeds ‘UGUAUG’ and ‘AAGUGC’, respectively, whereas the miR-466s are largely a “superfamily” of the shifted seeds ‘GAUGUG’, ‘AUGUGU’, and ‘UGUGUG’ and the miR-669s contain a diversity of seeds. While miR-297s map only to the ends of the intron, the miR-
99

466s, miR-467s and miR-669s are also part of a 2 kb region that tiles across the intron 13 times (Figure 3A). Sequence variation within these four classes in the
Sfmbt2 cluster is sufficient to generate 23 distinct miRNA seeds, the majority of which are not found in humans.
Given the sequence heterogeneity of the entire expansion, reconstruction of the flanking region’s evolution was challenging. However, alignment of the tiling 2 kb regions in the mouse specific expansion revealed shared mutations accumulating in different repeat units, allowing reconstruction of a plausible duplication history (Figure 3B, Figure S2A). As few as ten steps appear to have generated the 13 repeated units, nearly all of which are tandem duplications.
Assuming that the sequence differences among tiling units resulted primarily from neutral evolution, the pair with the greatest sequence dissimilarity (having a base substitution rate of roughly 3% over their sequence length) appears to have emerged approximately 6 million years ago (Mya) based on the estimated murine base substitution rate of 4 $ 10 -9 per site per year (Waterston et al., 2002). Thus, this final expansion of the cluster is predicted to have occurred after the mouserat divergence (which occurred ~20-25 Mya), reinforcing that this was a mousespecific event (Waterston et al., 2002). Clustering of individual miRNA precursors suggests that miR-669 elements were most similar to the ancestral sequence, and that subsequent duplications gave rise to a miR-466 precursor, from which both the miR-467s and miR-297s derived (Figure 3C, Figure S2B).
While the seed regions of these latter two miRNA classes are well conserved within each group, there is relatively less similarity among the 5' ends of
100

sequences in the miR-466s and miR-669s, which emerged earlier (Figure S3). In spite of this seed divergence, a number of miR-466 and miR-669 class miRNAs were detected as having gain and loss signatures calculated above, suggesting that they retained function as they diverged.
As the Sfmbt2 miRNA cluster has not been characterized in depth, we first explored whether these miRNAs are Dicer-dependent and function like canonical miRNAs. Northern Blot analysis showed that while mature forms of these miRNAs are present in wild type (WT) ES cells, only their precursors could be detected in Dicer knockout ( Dcr KO) cells (Figure S4A). This is consistent with previous cloning results, which failed to detect mature Sfmbt2 miRNAs in Dcr KO mESCs (Calabrese et al., 2007). In addition, miRNAs from this cluster can repress luciferase targets that have perfect miRNA binding sites in their 3' UTRs
(Figure S4B), suggesting that they can participate in canonical RNA-induced silencing pathways. Furthermore, we used available microarray data to assess whether Sfmbt2 miRNAs can destabilize mRNA targets containing seed complementarity, which appears to be the dominant means of miRNA function in mammals (Guo et al., 2010). When comparing the gene expression profiles of
WT and Dcr KO mESCs from microarray data (Zheng et al, manuscript in preparation), predicted targets of each Sfmbt2 miRNA (defined as those mRNAs with at least a 7mer-A1 or 7mer-M8 match) showed a significant increase in expression in Dcr KO cells relative to control genes matched for 3' UTR length, dinucleotide composition, and expression in WT mESCs (Table S6, Figure S5).
Interestingly, these expression shifts were evident even for miRNAs without
101

significant evolutionary signatures, although those miRNAs that were identified in our earlier analysis with greater seed conservation (Figure S3) showed increased significance, suggesting they may have a greater number of functional targets.
Sfmbt2
To dissect the functions of the Sfmbt2 cluster, we began by examining the represented seeds and their potential targets. Notably, the miR-467a family of miRNAs shares the hexamer seed ‘AAGUGC’ with other conserved miRNA clusters, such as the miR-290-295 and miR-302 clusters, which are expressed predominantly in ES cells (Figure 4A). However, the remainder of its mature sequence differs from the latter two clusters and its mature sequence derives from the 5’ rather than 3’ arm, consistent with its independent derivation from an ancestral miR-669 sequence. Given the previously described central roles for the miR-290-295 cluster and its human counterpart, the miR-371-373 cluster, in targeting negative regulators of ES cell cycle progression (Qi et al., 2009; Wang et al., 2008), we tested whether miR-467a family members might have similar functions. Indeed, when tested on a series of previously validated miR-295 targets in Dicer null mouse ES cells, such as p21 and Lats2, miR-467a was capable of reporter repression (Figure S6). In addition, when we tested its ability to rescue the G1/S delay in Dicer null cells, the effect of miR-467a was comparable to that of miR-295 in restoring wild type cycling (Figure 4B). We
102

have recently identified a novel anti-apoptotic phenotype conferred by the miR-
295 family following genotoxic stress (Zheng et al, manuscript in preparation), and found that miR-467a recapitulated this protection following gamma radiation
(Figure 4C) or doxorubicin treatment (Figure 4D) when transfected into Dicer null cells, as compared to a seed-mutant version. Taken together, these results suggest that miR-467a family members could reinforce proliferative cellular programs in parallel with miR-295.
Examination of available global miRNA expression data suggested that in addition to ES cells, placental tissue shows high expression of Sfmbt2 miRNAs
(Landgraf et al., 2007). Comparing the relative expression of miR-467a by
Northern blot analysis, there appeared to be a 4-fold upregulation in placental tissue relative to ES cells (Figure 5A, Figure S7A). Similar expression trends were observed with representatives of the miR-297 (7-fold increase) and miR-
466 (2.5-fold increase) classes (Figure 5A, Figure S7B, C). These findings are consistent with mRNA expression data for Sfmbt2 (Lattin et al., 2007), which has been shown to be maternally imprinted (and therefore paternally expressed) in the placenta (Kuzmin et al., 2008). Therefore, it is likely that expression of the intronic miRNA cluster is dependent on transcription of the entire gene, as has been suggested for many miRNAs similarly positioned within host genes
(Baskerville and Bartel, 2005). In contrast to Sfmbt2 miRNAs, miR-295, which shares the AAGUGC seed with miR-467a and is highly expressed in ES cells, was barely detectable in the placenta (Figure 5A, Figure S7D). The differing expression patterns of these miRNAs, also confirmed by a global miRNA profiling
103

dataset (Figure S7E) (Landgraf et al., 2007), suggests that the miR-295 family may be principally responsible for the proliferation of ES cells while some members of the Sfmbt2 miRNAs may regulate placental growth.
Sfmbt2
Because the remaining members of the Sfmbt2 cluster that are associated with genome-wide signals of selection have seeds unique to mice, we were able to predict potential mouse-specific target transcripts. As defined above, these target sites should have significantly higher mutation rate than adjacent sequences. We assessed their significance with the binomial test, and identified
511 placentally expressed genes whose target sites have probably been specifically selected for in mouse (p " 0.05, FDR < 0.40) (Figure S8). Enrichment in annotated pathways and functions was tested for all 511 genes using Ingenuity
Pathway Analysis, with the top statistically significant categories including “Cell
Death” (p = 6.7 $ 10 -5 to 4.6 $ 10 -2 ) and “Cell Cycle” (p = 2.1 $ 10 -4 to 4.7 $ 10 -2 ).
Given the similarity between these annotated functions and known functions of the miR-290-295 cluster, we compared this list of non-AAGUGC Sfmbt2 seeds to predicted AAGUGC targets to determine whether significant co-targeting was evident. Indeed, a statistically significant overlap of 157 common genes (p = 3.7
$ 10 -5 , hypergeometric test) was present, supporting the notion that these two miRNA clusters were selected in part for similar functional roles (Figure 5B).
104

To explore targets relevant to our pathway analysis results, we tested whether two previously well characterized proliferation-related genes contain functional mouse-specific target sites: Dedd2 (p = 0.16, FDR < 65%) and Lats2
(p = 0.23, FDR < 57%). Dedd2 is a well conserved inducer of apoptosis across a variety of cell types, known to associate with Caspase 8 and Caspase 10 via its
Death Effector Domain (DED) (Alcivar et al., 2003). Lats2, a target of the miR-
290-295 cluster (Wang et al., 2008) and its homologous cluster in human
(Voorhoeve et al., 2006), participates in both cell cycle progression (Li et al.,
2003) and apoptosis (Ke et al., 2004), in part through a positive feedback loop with p53 (Aylon et al., 2006). Both genes show acquisition of target sites for
Sfmbt2 cluster miRNAs specifically in mouse with a mutation profile higher than surrounding 3' UTR regions (Figure 5C). Reporters containing the 3' UTRs of these genes fused to luciferase were transfected into Dicer KO cells with and without miRNAs predicted to have gained target sites in mice. Of the two potential target sites acquired by Lats2, the miR-466f-3p site showed a ~25% repression relative to a site-mutant construct, while the miR-669-3p site appeared to be nonfunctional (Figure 5D). In the case of Dedd2, miR-297a led to a ~33% repression (Figure 5D). Thus, mouse-specific sequence variations in the
3' UTRs of these genes gave rise to functional target sites.
105

Here we present a novel method of studying species-specific changes in miRNA-target relationships and apply it to uncover the genome-wide response to miRNA acquisition. In contrast to many current methods of target prediction, which detect conserved relationships (Betel et al., 2008; Krek et al., 2005; Lewis et al., 2005), our method emphasizes relative divergence to identify positive selection for site gain and loss. Our results indicate that introduction of a novel miRNA cluster can be associated with genome-wide adaptation.
In the case of the Sfmbt2 cluster, 23 distinct seeds were created, representing one of the largest miRNA expansions described (Hertel et al.,
2006). This expansion was associated with significant genome-wide 3' UTR alterations in the mouse lineage. In some cases, these corresponded to avoidance of targeting, suggesting that a subset of novel interactions are disadvantageous. Such a model has been suggested by intragenomic analyses in which target site depletion is observed in highly expressed transcripts of the same tissue as the miRNA (Farh et al., 2005), or target site enrichment is observed in transcripts of adjacent tissues (Stark et al., 2005). Intriguingly, a larger number of cases of site gain were observed, suggesting that this cluster was positively selected for functional targeting.
Taking advantage of interspecies comparison, a genome-wide response to the presence of the Sfmbt2 cluster was detected. While many of the identified target interactions are likely to be novel, a subset may functionally overlap with
106

those of the miR-290-295 cluster, which promotes proliferation (Wang et al.,
2008) and survival (Zheng et al, manuscript in press). In some cases, single 3'
UTRs may be targeted by multiple cluster members, as was observed for Lats2
(a target of miR-467a and miR-466f-3p), a gene known to oppose proliferation and growth (Alcivar et al., 2003; Ke et al., 2004; Li et al., 2003). Inhibition of
Lats2 and our other validated target, Dedd2, fits well with the position of the
Sfmbt2 cluster in a paternally expressed placental gene (Kuzmin et al., 2008), as such genes are thought to be commonly involved in redistributing resources from mother to offspring under the parental sex conflict model (Moore and Haig,
1991). Consistent with this idea, paternal duplication of proximal chromosome 2
(which includes the Sfmbt2 gene) results in placental growth enhancement, whereas maternal disomy results in fetal and placental growth reduction
(Cattanach et al., 2004). Our data suggest that some of these growth effects may result from loss of the miRNA cluster in addition to the coding gene.
The Sfmbt2 cluster expansion in mouse may additionally provide a useful model for other species- and lineage-specific miRNA expansions. For instance, in humans, an analogous primate-specific miRNA expansion is present on
Chromosome 19 (C19MC), with a total of 46 miRNAs within a 100 kb interval being processed from the intron of a noncoding RNA (Bortolin-Cavaille et al.,
2009). Also found in a novel genomic expansion highly enriched for repetitive elements – in this case, Alu elements, the most significantly expanded transposons in humans (Cordaux and Batzer, 2009) – this host gene is similarly maternally imprinted in the placenta (Noguer-Dance et al., 2010), where it is
107

highly expressed relative to other tissues (Bortolin-Cavaille et al., 2009). In addition to a number of novel, primate-specific seeds, this cluster also includes miRNAs with the seed ‘AAGUGC,’ found in miR-371-373 (the human counterpart of the miR-290-295 cluster) (Lin et al.). Indeed, recent studies have identified aberrant expression of this cluster in human cancers, where it is thought to enhance oncogenicity by promoting cell survival and growth (Li et al., 2009;
Rippe et al., 2010). These observations parallel the results presented here, suggesting that the same pro-survival functions that are advantageous to cancer cells may have spurred the emergence and fixation of these two clusters for their contributions to inter-sex conflict.
Feeder-free Dcr flox/flox and Dcr -/ mouse embryonic stem cells (mESCs) were generated and maintained as described previously (Calabrese et al., 2007). mESCs cells containing a floxed and excised mir-290-295 cluster were similarly generated, as described in an upcoming publication (Medeiros et al. manuscript in preparation).
siRNA miR-297a
Sequence (5’ % 3’ unless otherwise noted)
5’- AUGUAUGUGUGCAUGUGCAUGU -3’
3’- UGUACAUACACACGUACACGUA -3’
108

miR-466a-5p miR-466f-3p miR-295
5’- UAUGUGUGUGUACAUGUACAUAU -3’
3’- UAAUACACACACAUGUACAUGUA -3’
5’- CAUACACACACACAUACACAC -3’
3’- GUGUAUGUGUGUGUGUAUGAC -3’
5’- AAAGUGCUACUACUUUUGAGUCU -3’
3’- UCUUUCACGAUGAUGAAAACUCA -3’ miR-295 mutant
5’- AAAGACGUACUACUUUUGAGUCU -3’
3’- UCUUUCUGCAUGAUGAAAACUCA -3’ miR-467a
5’- UAAGUGCCUGCAUGUAUAUGCG -3’
3’- GCAUUCACGGACGUACAUAUAC -5’ miR-467a mutant
5’- UAAGACGCUGCAUGUAUAUGCG -3’
3’- GCAUUCUGCGACGUACAUAUAC -5’ miR-669f
5’- CAUAUACAUACACACACACGUAU -3’
3’- ACGUGUGUGUGUAUGUAUAUGAU -5’ si-p21
Control siRNA
3 ' UTR
Primers
Dedd2
Forward
Dedd2
Reverse
Mutagenic primers
Dedd2
Lats2
Northern
LNA Oligos
(from Dharmacon, Smartpool)
(from Dharmacon, Accell Non-targeting pool)
5’- AATAACTCGAGGGGAGGCATAACCCCCTGC -3’
5’- AATAAGGGCCCCCCACCTGTGCCCTTTCCA -3’
5’- GCTCTGCCTGCCGCAGTAACAGATATCCCACTC -3’
5’- CTTAGACATATAGGTGTGTTATTAACTATAGATAAACACACA -3’ miR-295 LNA 5’- AGACTCAAAAGTAGTAGCACTTT -3’ miR-297a 5’- ACATGCACATGCACACATACAT -3’ miR-466a-5p 5’- ATGTACATGTACACACACATA -3’ miR-466d-3p 5’- TCTATGTGTGCGTGTATGTATA -3’ miR-467a 5’- CGCATATACATGCAGGCACTTA -3’ miR-669a-3p 5’- ACGTGTGTGTGTATGTTATGT -3’
Gln-tRNA 5’- TGGAGGTTCCACCGAGAT -3’
MicroRNA mediated repression of candidate genes was tested by cloning
PCR amplified products corresponding to the entire 3' UTR downstream of a
109

pRL-CMV Renilla luciferase reporter as described previously (Doench and
Sharp, 2004).
Total placental RNA from d11.5 and d13.5 was obtained from the Lees lab as described previously (Lee et al., 2009). Approximately 36 µ g of each RNA was loaded onto a 12% denaturing UREA gel, according to the Northern Blot protocol outlined previously (Calabrese et al., 2007). Membrane probed with Gln-tRNA was exposed to a phosphoimager for 3 hours before being scanned; miR-297a-
5p membrane was exposed for 16 hours; miR-295 for 5 hours; miR-466a-5p for
24 hours; and miR-467a for 24 hours. Prior to hybridizing with a different probe, membranes were stripped by incubating the membrane in boiling 0.1% SDS for
30 minutes and loss of signal was confirmed prior to rehybridization.
Microarray data was obtained from and processed according to Zheng et al (manuscript in preparation). Briefly, Dcr flox/flox embryonic stem cells were profiled in triplicate 5 days following either cre-mediated Dcr deletion or transfection of a control GFP plasmid. Affymetrix Mouse 430_2 array data were normalized using GCRMA, The log2 fold change (LFC) value for WT/KO was defined as the difference between the mean log expression in WT cells and the mean log expression in Dcr KO cells. Targets of a given Sfmbt2 miRNA were selected based on at least one match to a T1-A or M2-8 7mer of the mature miRNA sequence.
110

Human, dog and horse mature miRNA sequences were obtained from miRBase Release 15 (Griffiths-Jones et al., 2008). Mouse mature miRNA sequences were obtained from Chiang et al (Chiang et al., 2010). miRNA seeds were defined as nucleotides 2-8 of the mature sequence (unless otherwise specified). A miRNA seed was considered shared between mouse and human if it was found in both human and mouse. However, a miRNA seed was defined as mouse-specific if it was found in mouse, but not in human, dog, or horse.
Likewise, a miRNA seed was defined as human-specific if it was found in human, but not in mouse, dog, or horse.
Sfmbt2 miRNA precursor coordinates and sequences were obtained from
Chiang et al (Chiang et al., 2010). ClustalW was used to obtain multiple sequence alignments among precursors (Larkin et al., 2007). For alignment of tiling regions, the entire pre-miR sequence of the three miRNAs in each tiling unit as well as the intervening bases were used for phylogenetic reconstruction.
Aligned human, mouse, and dog 3' UTRs were obtained from TargetScan
5.1 (Friedman et al., 2009). A total of 17840 3' UTRs were used for the analysis.
First, heptamer seed matches were identified in mouse 3' UTRs and their aligning human sequences. Second, the difference, d , between the mutation rate at targets sites and adjacent sequences was calculated. The mutation rate was
111

defined as the fraction of mismatched nucleotides in a 7-nt window. The mutation rate at the target site was calculated as the mutation rate at the heptamer seed.
The mutation rate of adjacent sequences was calculated as the average of the mutation rates of the remaining non-overlapping heptamers in the 84-nt window.
Lastly, the d of the heptamer of interest was compared to those of control heptamers to calculate p-values. The control heptamers were generated with sequence composition matching the heptamer of interest.
Mouse-specific Sfmbt2 cluster targets were mRNAs whose 3' UTRs have at least one Sfmbt2 miRNA target site in mouse, but no aligned site in human or dog. The mutation rate at the target site should be higher than expected from the neighboring regions if the site was positively selected. The p-values were generated using a binomial test with the following formula,
P = P(Xs # Ns|Nt) = 1-P(Xs<Ns|Nt) where
P(Xs=k|Nt) has a binomial distribution with parameters Nt and (7/84);
Xs = number of mutated nucleotides in the seed region;
Nt = total number of mutated nucleotides in the region that covers the seed region and the 2 flanking regions each 42 nucleotides long;
Ns = number of mutated nucleotides in the seed region observed for a target site
!
112

Data were analyzed through the use of Ingenuity Pathways Analysis
(Ingenuity® Systems, www.ingenuity.com). Ingenuity Pathway Analysis (IPA) was performed on the set of 511 genes with evidence of positive selection in the mouse placenta. Genes were compared to a reference set of all placentally expressed genes defined by an array expression greater than 4 from Lee et al
(Lee et al., 2009). All available data sources were used, and species information was restricted to mouse.
All test statistics were calculated using R (http://www.r-project.org). The
Wilcoxon rank sum test was used to assess the significance of CDF’s because it does not assume normality of the underlying distributions. T-tests and
Kolmogorov–Smirnov (KS) test using these data gave generally similar results.
Chi-square test was used in positive and purifying selection analyses to assess over-representation of specific categories relative to the control.
We thank Joel Neilson for reading of the manuscript and the Sharp and
Burge labs for helpful advice and discussions. AR was funded by a Fannie and
John Hertz Foundation Fellowship. This work was supported by NIH grants RO1-
GM34277, NCI grant PO1-CA42063, and the NCI Cancer Center Support (core) grant P30-CA14051. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
113

Table 1. Target sites of 28 mouse-specific miRNA seeds are under positive selection.
A.
Target sites of 25 seeds show a significant gain signal. B.
Target sites of 16 seeds show a significant loss signal. Sfmbt2 miRNAs are shaded in blue.
Figure 1. Positive and purifying selection analysis for miRNAs in mouse and human.
A.
A flow chart of the analysis. First, mouse and human 3' UTRs are aligned and seed match sites are identified. Then the mutation rate at the miRNA binding site was compared to those of adjacent sequences. The difference in average per base mutation rate was noted as d . Lastly, d was compared to that of control heptamers. B.
Illustration of the analysis of miR-297a-5p, an example of a species-specific miRNA. Left panel, the alignment of mouse and human 3'
UTRs of a target of miR-297a-5p. Seed binding site is colored in dark red. Middle panel, mutation rate of an 84-nt sequence spanning the miR-297a-5p match site.
Right panel, distribution of d ’s from control heptamers. The black arrow points at the d for miR-297a-5p match sites. C.
Illustration of the analysis of miR-1, a conserved miRNA between mouse and human. Left panel, the alignment of mouse and human 3' UTRs of a target of miR-1. Seed binding site is colored in dark red. Middle panel, mutation rate of an 84-nt sequence spanning the miR-1 match site. Right panel, distribution of d ’s from control heptamers. The black arrow points at the d from miR-1 match sites. Frequency of mutation denotes the
114

fraction of mismatched nucleotides in the 7-nt window. “0” indicates the miRNA binding site, “42” refers to 42-nt downstream of the miRNA binding site, “-42” refers to 42-nt upstream of the miRNA binding site.
Figure 2. Target sites of 28 mouse miRNAs show significant positive selection signal. A.
Summary statistics of positive selection signals detected from mouse miRNA M8 heptamer binding sites in mouse and human 3' UTRs.
“count” represents the number of heptamers that were significant in the analysis.
“total” represents the total number of possible heptamers in the specific test category. P-values were the result of Chi-square tests. There are 16384 heptamers in total. There are 201 mouse-specific miRNA seeds (dog and horse miRNAs were used as an outgroup to determine the species specificity).
B.
Distribution along the 21 chromosomes (chromosomes 1-19, X and Y) of mouse miRNA seeds that showed a significant positive selection signal. Each dot represents a miRNA (blue, site gain; red, site loss).
Figure 3. Genomic structure of the Sfmbt2 miRNA cluster and phylogenetic relationship among members of the miRNA cluster. A.
Precursors of Sfmbt2 miRNAs were mapped to the 9th intron of the Sfmbt2 gene based on the coordinates of precursors. The intron spans a region from roughly 10.39Mbp to
10.44Mbp on Chromosome 2. The Sfmbt2 miRNAs are color coded as follows: miR-297s, green; miR-466s, black; miR-467s, blue; miR-669s, magenta. B.
Plausible reconstruction of the internal Sfmbt2 cluster expansion. The
115

duplication series shown is consistent with predicted ancestral relationships from
ClustalW alignments of the 13 three-miRNA tiling units. *Steps IV and V may have occurred as a single duplication event. C.
Emergence order of the four primary miRNA classes of the Sfmbt2 cluster. Branches reflect evolutionary distance to most ancestral element in a given class, based on ClustalW alignments of all miRNA precursors in the cluster (Figure S3B).
Figure 4. miR-467 can promote growth and survival of mouse ES cells. A.
seed sequences of miR-295, miR-302, and miR-467 families. B.
Dcr KO ES cells were transfected with 50 nM of miR-467, siRNA against p21, and other control siRNAs. 24 hours after transfection, cells were incubated with BrdU for 10 min, and BrdU positive cells were analyzed with flow cytometry. Assays with miR-
467a seed mutant (467aMut), miR-295 seed mutant (295Mut) and control siRNAs served as negative controls. Results are percentages in each stage of the cell cycle and are shown as mean
±
S.E.M. (standard error of the mean). n =
2 for miR-467a seed mutant and p21 siRNA transfections. n # 3 for remaining transfections. P-values were results of Mann-Whitney tests, and * denotes p "
0.05. C.
ES cells were treated with 5-Gy radiation 24 hours after transfection of
50nM miR-467a or a control siRNA. Casp3 activity was assayed 0 and 24 h after treatment, and the difference in apoptosis rate is shown. Apoptosis rate of Dcr
KO cells is shown in black bars, and that of WT cells is shown in white bars. D.
A similar series of experiments as C was performed in Dcr KO and WT cells using
116

100 nM doxorubicin. n # 3 for all experiments. Results are shown as mean ±
S.E.M. P-values were results of Mann-Whitney tests, ** denotes p " 0.01.
Figure 5. Expression of Sfmbt2 miRNAs in ES cells and the placenta, and two validated targets of the Sfmbt2 miRNAs. A.
Northern blot analysis of miR-
467a, miR-297a-5p, and miR-466a-5p from the Sfmbt2 cluster as well as miR-
295 in placental and ES cells. Gln tRNA was probed as a loading control. B.
Intersection of predicted Sfmbt2 mouse-specific targets and miR-295 targets. Pvalue represents the significance of the intersection, and was calculated by the hypergeometric distribution. A total of 11922 placentally expressed genes were considered for the analysis. C.
Comparison of mouse and human 3' UTRs flanking gained miRNA target sites in Lats2 and Dedd2. Bases that differ are marked by X’s, and the seed and seed complement are boxed. D.
Luciferase reporters with full length Lats2 3' UTR, Dedd2 3' UTR, as well as their seed mutant versions were assayed in Dcr KO ES cells. 20 nM of miR-297a-5p, miR-
466f-3p, and miR-669f-3p were transfected in Dcr KO cells. n = 3 and results are shown as mean ± S.E.M. P-values were results of Mann-Whitney tests, * denotes p
"
0.05.
Supplemental Table 1. Summary statistics of positive selection signals detected from mouse miRNA A1 heptamer binding sites. Top panel, gain signal. Bottom panel, loss signal. “count” represents the number of heptamers that were significant in the analysis. “total” represents the total number of
117

possible heptamers in the specific test category. P-values were the result of Chisquare tests. There are 16384 total heptamers. There are 151 mouse specific A1 miRNA seeds (dog and horse miRNAs were used as an outgroup to decide on the species specificity).
Supplemental Table 2. Summary statistics for human specific miRNAs. Top panel, gain signal. Bottom panel, loss signal. “count” represents the number of heptamers that were significant in the analysis. “total” represents the total number of possible heptamers in the specific test category. P-values were the result of Chi-square tests. There are 16384 total heptamers. There are 595 human specific M8 miRNA seeds, and 470 A1 miRNA seeds (dog and horse miRNAs were used as an outgroup to decide on the species specificity).
Supplemental Table 3. Summary statistics of purifying selection signals detected from heptamer binding sites in mouse and human 3' UTRs. “count” represents the number of heptamers that were significant in the analysis. “total” represents the total number of possible heptamers in the specific test category.
P-values were the result of Chi-square tests. There are 16384 total heptamers.
There are 218 M8 and 234 A1 miRNA seeds that are shared between human and mouse miRNAs.
Supplemental Table 4. Target sites of 44 M8 miRNA seeds are under purifying selection.
118

Supplemental Table 5. Description of the Sfmbt2 miRNA cluster.
Sequences, seeds, and coordinates of Sfmbt2 miRNAs in mouse ES cells are listed. The sequences and coordinates were taken from Chiang et al ( Chiang et al., 2010 ).
Supplemental Table 6. Target statistics of Sfmbt2 miRNAs. LFC of predicted
6-mer and 7-mer targets (between WT and Dcr KO ES cells) of each Sfmbt2 miRNA was compared to that of controls matched for dinucleotide content, 3'
UTR length, and WT expression by Wilcoxon test.
Supplemental Table 7. Predicted mouse-specific targets of the Sfmbt2 miRNAs.
Supplemental Figure 1. Distribution of d for conserved or species-specific miRNAs is different from that of all miRNAs. A. top panel, distribution of d for all mouse and human miRNA seeds (gray); bottom panel, distribution of d for conserved (red) and species-specific miRNAs (blue) with significant mutation profiles. B.
cumulative distribution functions of d values for conserved (red) and species-specific (blue) miRNAs are significantly different from that for all miRNAs
(gray), p-values " 2.2 $ 10 -16 and 5.3 $ 10 -14 by Kolmogorov-Smirnov test, respectively.
119

Supplemental Figure 2. Multiple sequence alignments of Sfmbt2 cluster sequences. A. 13 three-miRNA tiling units were aligned with ClustalW to identify relative ancestry. B.
Phylogenetic tree of Sfmbt2 miRNAs based on ClustalW alignment of precursor sequences.
Supplemental Figure 3. Precursors of Sfmbt2 miRNAs were aligned with
ClustalW. Alignments of mature sequences and percent conservation (% cons) calculated along the mature sequences were shown for each miRNA class found in the Sfmbt2 miRNA cluster. A. miR-297s.
B.
miR-467s. C, miR-
466s. D, miR-669s.
Supplemental Figure 4. The expression of Sfmbt2 miRNAs is dependent on
Dicer, and Sfmbt2 miRNAs can repress luciferase targets with complementary binding sites in the 3' UTRs. A. Northern analysis for two of the Sfmbt2 miRNAs in WT and Dcr KO mESCs. miR-669a-3p LNA probe for lanes 1 and 2, and miR-466d-3p LNA probe for lanes 3 and 4. 10-base pair DNA ladder for lane 5.
B. Luciferase reporters with 2 perfect binding sites or 2 bulged binding sites were transfected into WT (white bar) and Dcr KO (black bar). 100 nM miR-467a was added back to Dcr KO cells, and the luciferase activity of respective reporters was plotted (gray bar). Luciferase reporters with 2 mutated bulged binding sites (mir-467aMut 2x-bulged) were used as negative controls. n
= 3 and results are shown as mean ± S.E.M. P-values were results of Mann-
Whitney tests, * denotes p " 0.05.
120

Supplemental Figure 5. CDFs (cumulative distribution functions) of log2 fold change (LFC) in mRNA expression between WT and Dcr KO ES cells are plotted for target sets of each Sfmbt2 miRNA seed. 6-mer targets (red), predicted targets with a hexamer match to the Sfmbt2 miRNA seed in their 3'
UTRs. 7-mer targets (green), predicted targets with a heptamer match (M8 or A1) to the Sfmbt2 miRNA seed in their 3' UTRs. control (blue) was selected to match the 6-mer targets in dinucleotide composition, 3' UTR length, and expression in
WT ES cells.
Supplemental Figure 6. Activity of luciferase reporters of predicted targets of the miR-467a cluster were assayed in WT and Dcr KO ES cells. Luciferase reporters contain full length 3' UTRs of predicted targets. Relative luciferase activity is the ratio of the reporter’s activity in WT ES cells and Dcr KO ES cells. n
#
2 for all experiments, and results are shown as mean
±
S.E.M. P-values were results of t-tests, * denotes p
"
0.05 and ** denotes p
"
0.01.
Supplemental Figure 7. Quantitation of relative Sfmbt2 miRNA expression in ES cells and the placenta (d11.5 and d13.5). All the expression is normalized to the miRNA expression in WT ES cells. A. miR-467a.
B. miR-
297a-5p.
C. miR-466a-5p.
D. miR-295.
E. Expression of the Sfmbt2 miRNA cluster and the miR-295 families in ES cells and the placenta from miRAtlas
(Landgraf et al., 2007).
121

Supplemental Figure 8. P-value distribution of mouse-specific Sfmbt2 cluster target genes. Dashed line indicates threshold for false discovery.
122

Table 1
123

124

125

Class (includes multiple seeds)
126

127

(Non AAGUGC) (AAGUGC)
128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

Abelson, J. F., Kwan, K. Y., O'Roak, B. J., Baek, D. Y., Stillman, A. A., Morgan,
T. M., Mathews, C. A., Pauls, D. L., Rasin, M. R., Gunel, M.
, et al.
(2005).
Sequence variants in SLITRK1 are associated with Tourette's syndrome.
Science 310 , 317-320.
Alcivar, A., Hu, S., Tang, J., and Yang, X. (2003). DEDD and DEDD2 associate with caspase-8/10 and signal cell death. Oncogene 22 , 291-297.
Aylon, Y., Michael, D., Shmueli, A., Yabuta, N., Nojima, H., and Oren, M. (2006).
A positive feedback loop between the p53 and Lats2 tumor suppressors prevents tetraploidization. Genes Dev 20 , 2687-2700.
Bartel, D. P. (2009). MicroRNAs: target recognition and regulatory functions. Cell
136 , 215-233.
Baskerville, S., and Bartel, D. P. (2005). Microarray profiling of microRNAs reveals frequent coexpression with neighboring miRNAs and host genes.
RNA 11 , 241-247.
Betel, D., Wilson, M., Gabow, A., Marks, D. S., and Sander, C. (2008). The microRNA.org resource: targets and expression. Nucleic Acids Res 36 ,
D149-153.
Bortolin-Cavaille, M. L., Dance, M., Weber, M., and Cavaille, J. (2009). C19MC microRNAs are processed from introns of large Pol-II, non-protein-coding transcripts. Nucleic Acids Res 37 , 3464-3473.
Bradley, R. K., Li, X. Y., Trapnell, C., Davidson, S., Pachter, L., Chu, H. C.,
Tonkin, L. A., Biggin, M. D., and Eisen, M. B. (2010). Binding site turnover produces pervasive quantitative changes in transcription factor binding between closely related Drosophila species. PLoS Biol 8 , e1000343.
Calabrese, J. M., Seila, A. C., Yeo, G. W., and Sharp, P. A. (2007). RNA sequence analysis defines Dicer's role in mouse embryonic stem cells.
Proc Natl Acad Sci U S A 104 , 18097-18102.
Cattanach, B. M., Beechey, C. V., and Peters, J. (2004). Interactions between imprinting effects in the mouse. Genetics 168 , 397-413.
Chen, K., and Rajewsky, N. (2006). Natural selection on human microRNA binding sites inferred from SNP data. Nat Genet 38 , 1452-1456.
Chen, K., and Rajewsky, N. (2007). The evolution of gene regulation by transcription factors and microRNAs. Nat Rev Genet 8 , 93-103.
Chiang, H. R., Schoenfeld, L. W., Ruby, J. G., Auyeung, V. C., Spies, N., Baek,
D., Johnston, W. K., Russ, C., Luo, S., Babiarz, J. E.
, et al.
(2010).
Mammalian microRNAs: experimental evaluation of novel and previously annotated genes. Genes Dev 24 , 992-1009.
Clop, A., Marcq, F., Takeda, H., Pirottin, D., Tordoir, X., Bibe, B., Bouix, J.,
Caiment, F., Elsen, J. M., Eychenne, F.
, et al.
(2006). A mutation creating a potential illegitimate microRNA target site in the myostatin gene affects muscularity in sheep. Nat Genet 38 , 813-818.
151

Cordaux, R., and Batzer, M. A. (2009). The impact of retrotransposons on human genome evolution. Nat Rev Genet 10 , 691-703.
Doench, J. G., and Sharp, P. A. (2004). Specificity of microRNA target selection in translational repression. Genes Dev 18 , 504-511.
Farh, K. K., Grimson, A., Jan, C., Lewis, B. P., Johnston, W. K., Lim, L. P.,
Burge, C. B., and Bartel, D. P. (2005). The widespread impact of mammalian MicroRNAs on mRNA repression and evolution. Science 310 ,
1817-1821.
Friedman, R. C., Farh, K. K., Burge, C. B., and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res
19 , 92-105.
Glazov, E. A., McWilliam, S., Barris, W. C., and Dalrymple, B. P. (2008). Origin, evolution, and biological role of miRNA cluster in DLK-DIO3 genomic region in placental mammals. Mol Biol Evol 25 , 939-948.
Griffiths-Jones, S., Saini, H. K., van Dongen, S., and Enright, A. J. (2008). miRBase: tools for microRNA genomics. Nucleic Acids Res 36 , D154-158.
Guo, H.
, Ingolia, N. T.
, Weissman, J. S.
, and Bartel, D. P.
( 2010 ). Mammalian microRNAs predominantly act to decrease target mRNA levels . Nature
466 , 835-840 .
Hertel, J., Lindemeyer, M., Missal, K., Fried, C., Tanzer, A., Flamm, C., Hofacker,
I. L., and Stadler, P. F. (2006). The expansion of the metazoan microRNA repertoire. BMC Genomics 7 , 25.
Ke, H., Pei, J., Ni, Z., Xia, H., Qi, H., Woods, T., Kelekar, A., and Tao, W. (2004).
Putative tumor suppressor Lats2 induces apoptosis through downregulation of Bcl-2 and Bcl-x(L). Exp Cell Res 298 , 329-338.
King, M. C., and Wilson, A. C. (1975). Evolution at two levels in humans and chimpanzees. Science 188 , 107-116.
Klymenko, T., Papp, B., Fischle, W., Kocher, T., Schelder, M., Fritsch, C., Wild,
B., Wilm, M., and Muller, J. (2006). A Polycomb group protein complex with sequence-specific DNA-binding and selective methyl-lysine-binding activities. Genes Dev 20 , 1110-1122.
Krek, A., Grun, D., Poy, M. N., Wolf, R., Rosenberg, L., Epstein, E. J.,
MacMenamin, P., da Piedade, I., Gunsalus, K. C., Stoffel, M., and
Rajewsky, N. (2005). Combinatorial microRNA target predictions. Nat
Genet 37 , 495-500.
Kuzmin, A., Han, Z., Golding, M. C., Mann, M. R., Latham, K. E., and Varmuza,
S. (2008). The PcG gene Sfmbt2 is paternally expressed in extraembryonic tissues. Gene Expr Patterns 8 , 107-116.
Landgraf, P., Rusu, M., Sheridan, R., Sewer, A., Iovino, N., Aravin, A., Pfeffer, S.,
Rice, A., Kamphorst, A. O., Landthaler, M.
, et al.
(2007). A mammalian microRNA expression atlas based on small RNA library sequencing. Cell
129 , 1401-1414.
Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A.,
McWilliam, H., Valentin, F., Wallace, I. M., Wilm, A., Lopez, R.
, et al.
(2007). Clustal W and Clustal X version 2.0. Bioinformatics 23 , 2947-2948.
152

Lattin, J., Zidar, D. A., Schroder, K., Kellie, S., Hume, D. A., and Sweet, M. J.
(2007). G-protein-coupled receptor expression, function, and signaling in macrophages. J Leukoc Biol 82 , 16-32.
Lee, E. Y., Yuan, T. L., Danielian, P. S., West, J. C., and Lees, J. A. (2009).
E2F4 cooperates with pRB in the development of extra-embryonic tissues.
Dev Biol 332 , 104-115.
Levine, M., and Tjian, R. (2003). Transcription regulation and animal diversity.
Nature 424 , 147-151.
Lewis, B. P., Burge, C. B., and Bartel, D. P. (2005). Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 120 , 15-20.
Li, M., Lee, K. F., Lu, Y., Clarke, I., Shih, D., Eberhart, C., Collins, V. P., Van
Meter, T., Picard, D., Zhou, L.
, et al.
(2009). Frequent amplification of a chr19q13.41 microRNA polycistron in aggressive primitive neuroectodermal brain tumors. Cancer Cell 16 , 533-546.
Li, Y., Pei, J., Xia, H., Ke, H., Wang, H., and Tao, W. (2003). Lats2, a putative tumor suppressor, inhibits G1/S transition. Oncogene 22 , 4398-4405.
Lin, S., Cheung, W. K., Chen, S., Lu, G., Wang, Z., Xie, D., Li, K., Lin, M. C., and
Kung, H. F. Computational identification and characterization of primatespecific microRNAs in human genome. Comput Biol Chem.
Mattick, J. S., and Makunin, I. V. (2006). Non-coding RNA. Hum Mol Genet 15
Spec No 1 , R17-29.
Moore, T., and Haig, D. (1991). Genomic imprinting in mammalian development: a parental tug-of-war. Trends Genet 7 , 45-49.
Neilson, J. R., Zheng, G. X., Burge, C. B., and Sharp, P. A. (2007). Dynamic regulation of miRNA expression in ordered stages of cellular development.
Genes Dev 21 , 578-589.
Niwa, R., and Slack, F. J. (2007). The evolution of animal microRNA function.
Curr Opin Genet Dev 17 , 145-150.
Noguer-Dance, M.
, Abu-Amero, S.
, Al-Khtib, M.
, Lefevre, A.
, Coullin, P.
, Moore,
G. E.
, and Cavaille, J.
( 2010 ). The primate-specific microRNA gene cluster
(C19MC) is imprinted in the placenta . Hum Mol Genet 19 , 3566-3582 .
Nozawa, M.
, Miura, S.
, and Nei, M.
( 2010 ). Origins and evolution of microRNA genes in Drosophila species . Genome Biol Evol 2 , 180-189 .
Qi, J., Yu, J. Y., Shcherbata, H. R., Mathieu, J., Wang, A. J., Seal, S., Zhou, W.,
Stadler, B. M., Bourgin, D., Wang, L.
, et al.
(2009). microRNAs regulate human embryonic stem cell division. Cell Cycle 8 , 3729-3741.
Rippe, V.
, Dittberner, L.
, Lorenz, V. N.
, Drieschner, N.
, Nimzyk, R.
, Sendt, W.
,
Junker, K.
, Belge, G.
, and Bullerdiek, J.
( 2010 ). The two stem cell microRNA gene clusters C19MC and miR-371-3 are activated by specific chromosomal rearrangements in a subgroup of thyroid adenomas . PLoS
One 5 , e9485 .
Rosa, A., Spagnoli, F. M., and Brivanlou, A. H. (2009). The miR-430/427/302 family controls mesendodermal fate specification via species-specific target selection. Dev Cell 16 , 517-527.
153

Ruby, J.G., Stark A, Johnston W.K., Kellis M., Bartel D.P., Lai E.C. (2007).
Evolution, biogenesis, expression, and target prediction of a substantially expanded set of Drosophila microRNAs. Genome Res 17 , 1850-1864.
Seitz, H., Royo, H., Bortolin, M. L., Lin, S. P., Ferguson-Smith, A. C., and
Cavaille, J. (2004). A large imprinted microRNA gene cluster at the mouse
Dlk1-Gtl2 domain. Genome Res 14 , 1741-1748.
Shin, C.
, Nam, J. W.
, Farh, K. K.
, Chiang, H. R.
, Shkumatava, A.
, and Bartel, D.
P.
( 2010 ). Expanding the microRNA targeting code: functional sites with centered pairing . Mol Cell 38 , 789-802 .
Stark, A., Brennecke, J., Bushati, N., Russell, R. B., and Cohen, S. M. (2005).
Animal MicroRNAs confer robustness to gene expression and have a significant impact on 3'UTR evolution. Cell 123 , 1133-1146.
Voorhoeve, P. M., le Sage, C., Schrier, M., Gillis, A. J., Stoop, H., Nagel, R., Liu,
Y. P., van Duijse, J., Drost, J., Griekspoor, A.
, et al.
(2006). A genetic screen implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Cell 124 , 1169-1181.
Wang, Y., Baskerville, S., Shenoy, A., Babiarz, J. E., Baehner, L., and Blelloch,
R. (2008). Embryonic stem cell-specific microRNAs regulate the G1-S transition and promote rapid proliferation. Nat Genet 40 , 1478-1483.
Waterston, R. H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J. F., Agarwal,
P., Agarwala, R., Ainscough, R., Alexandersson, M., An, P.
, et al.
(2002).
Initial sequencing and comparative analysis of the mouse genome. Nature
420 , 520-562.
154

The material in this chapter was adapted from the following publication:
Arvind Ravi, Allan M. Gurtan, Madhu S. Kumar, Christine Chin, Jacqueline A. Lees,
Tyler Jacks, Phillip A. Sharp (2011). Viability of transformed somatic cells in the absence of Dicer. (in prep)
Experimental contributions:
This work represents an equal collaboration between Arvind Ravi and Allan Gurtan.
Arvind Ravi characterized the sarcoma model.
Allan Gurtan characterized the mesenchymal stem cell model.
155

MicroRNA downregulation has been shown to enhance tumor formation, yet tumors from patient samples and mouse models appear to retain at least one copy of Dicer, suggesting that complete microRNA loss may be incompatible with tumor formation. Here we demonstrate that Dicer null cells can remain stably proliferative, albeit at a lower rate, both in vitro and in vivo , challenging an essential role for short RNA regulation in tumorigenesis.
MicroRNAs are a class of short ~22 nucleotide RNAs predicted to regulate nearly half of all protein coding genes (Bartel, 2009; Friedman et al., 2009). The breadth of this regulation is supported by the early embryonic lethality of mice lacking Dicer, the final enzyme required for the maturation of nearly all microRNAs (Bernstein et al., 2003). Even in the case of tumors, which frequently feature a global reduction in miRNA levels (Gaur et al., 2007; Lu et al., 2005), complete loss of Dicer is not observed in a number of human patient samples or in mouse models explicitly testing the role of Dicer loss in tumorigenesis ( Arrate et al., 2010; Kumar et al., 2009). These results have supported the hypothesis that widespread deregulation downstream of complete microRNA loss is simply
156

incompatible with an essential feature of tumorigenesis, be it stable proliferation or in vivo growth.
In order to better understand the seemingly essential role of microRNAs in cancer, we enforced in vitro deletion of Dicer in sarcomas cell lines generated by hindlimb injection of Adeno-Cre virus in KrasLSL-G12D , p53 f/f , Dicer f/f mice. As described earlier, the resultant tumors never display homozygous loss, instead retaining at least one conditional Dicer allele (Kumar et al., 2009). To delete this remaining allele, we transduced Dicer heterozygous cells with a retroviral construct encoding Puro.CreER
T2 and activated recombination in vitro with tamoxifen treatment (Figure 1A). A genotyping time course indicated efficient homozygous recombination shortly after treatment (Figure S1). After multiple passages, however, only heterozygous cells could be isolated, consistent with previous findings in vivo in both this sarcoma model and an E !
-myc/Dicer lymphoma model ( Arrate et al., 2010; Kumar et al., 2009).
This result suggested two possibilities: (1) death of cells that have lost both copies of Dicer, or (2) a selective growth advantage of cells retaining a functional Dicer allele. To address these possibilities and prevent the preferential outgrowth of Dicer retaining cells, we isolated clonal populations of cells via low density plating shortly after tamoxifen treatment. Little cell death was observed and all of the clones appeared morphologically similar to the parental cell line but genotyping indicated that the majority of isolated clones had deleted the second
157

allele of Dicer (Figure 1A). This result suggests that sarcoma cells survive after homozygous Dicer deletion but have a growth disadvantage relative to cells retaining Dicer expression. Hereafter, we will refer to homozygous Dicerdeletions as Dicer -/ cells and parental heterozygous Dicer cell line as Dicer f/cells.
To determine whether Dicer -/ cells were truly devoid of miRNA regulation, we first ruled out the possibility of compensatory production of mature miRNAs by Northern analysis for abundant miRNAs. Notably, let-7, miR-16, and miR-17 were all detected abundantly in our Dicer f/ cells, whereas Dicer -/ cells showed an absence of mature miRNAs and concomitant accumulation of precursors (Figure
1B). Furthermore, luciferase reporters containing target sites for any of these three miRNAs were also derepressed 3-6 fold, consistent with functional loss
(Figure 1C). Finally, to evaluate proliferate changes accompanying Dicer loss, we measured the proliferation for each genotype in vitro . Consistent with a model of loss by competition, Dicer -/ cells divided more slowly, every ~15 hours, than the Dicer f/ controls, every ~12 hours (Figure 1D), but without obvious senescence or onset of crisis.
Taken together, these findings indicate that genetic ablation of Dicer and the resultant misregulation of gene expression can be readily tolerated in this particular mouse sarcoma model. However, related in vivo mouse models as well as genotyping data from human tumors suggest that complete Dicer loss tumors may not be tolerated in vivo . Several possibilities may explain this apparent requirement for Dicer. Loss of Dicer may confer (1) greater sensitivity to stress
158

induced cell death due to cell-autonomous or nonautonomous conditions within the organism (e.g., immune clearance), (2) gene expression changes that lead to loss of properties such as invasiveness that are required for tumor development, or (3) proliferative defects and subsequent loss through competition in vivo by
Dicer-expressing cells.
To address these possibilities, we carried out tumor formation assays in vivo with Dicer f/ or Dicer -/ sarcoma cells. Upon subcutaneous injection of 1 #
10 6 cells into the flanks of immune-compromised mice, Dicer -/ cells were indeed tumorigenic, forming tumors at 7/18 sites within 24 days, as compared to 4/8 sites by day 14 for the original Dicer f/ strain. To better evaluate the difference in tumor formation kinetics, we repeated this injection experiment with 2.5 # 10 4 cells. At this lower cell number, Dicer -/ sarcoma cells began to develop tumors in
~45 days, as compared to 22 days for the parental Dicer f/ sarcoma cell line
(Figure 2A). Pathologic analysis of either Dicer -/ or Dicer f/ tumors identified both as high grade sarcomas with pleomorphic nuclei and abnormal mitoses, consistent with previous reports of Kras driven sarcoma models (Figure 2B)
(Kirsch et al., 2007; Kumar et al., 2007). No obvious genotype-dependent differences were seen, and samples could not be readily distinguished in a blinded analysis.
Published studies examining the role of Dicer in tumor formation typically use immunocompetent animals, raising the possibility that Dicer -/ cells are targeted for clearance by an immune response. To address this caveat, we performed syngeneic injections into C57Bl6/SV129 F1 mice, and again observed
159

the formation of Dicer -/ tumors. As before, the rate was slower than the parental
Dicer f/ line, with the first tumors appearing 7 days after injection of the Dicer f/cells and 21 days after injection of the Dicer -/ cells (Figure S2). Thus, the absence of Dicer impairs but does not preclude tumor formation, even in an immunocompetent background.
Although the cells at the time of the injection lacked miRNAs, it is possible that in vivo growth led to a strong selection to regain miRNA processing, or that low numbers of contaminating Dicer f/ cells had grown out. Therefore, we genotyped DNA prepared from primary tumor tissue and confirmed a significant recombined band corresponding to the injected Dicer -/ cells (Fig 2C). Northern analysis of primary tumors revealed accumulation of precursors as well as significant but incomplete depletion of mature miRNAs (Fig 2D). The residual mature miRNA observed is likely due to host tissue contamination, as evidenced by the greatest miRNA signal in the tumor section showing the greatest wild-type contaminant band by PCR (Fig 2E, 2F).
To test whether the residual miRNAs were a result of contaminating wildtype tissue in the tumor, rather than restoration of miRNA processing ability by the injected cells, we generated cell lines from these tumors. PCR genotyping confirmed a depletion of the wild type tissue during this process (Figure 2E). By
Northern blotting, the level of mature miRNAs was lower than the detection limit of the blot, and this was again accompanied by an enrichment in the pre-miRNA
(Figure 2F). These trends support the notion that the injected Dicer -/ cells were able to survive and proliferate in vivo without recovery of miRNA processing.
160

Thus, the earlier in vitro results appear to extend to an in vivo setting, with sarcoma cells retaining the capacity to form phenotypically similar tumors in vivo , albeit more slowly, in the absence of Dicer and miRNAs.
Thus far, these results demonstrate that despite resultant misregulation of gene expression, homozygous loss of Dicer can be tolerated by murine sarcoma cells lacking p53 and expressing oncogenic K-ras. The viability of these cells may be a function of the strong oncogenic background required for rapid in vivo growth or may require additional genetic alterations that occur during tumor formation. Therefore, we tested whether Dicer loss could be tolerated in a second, related in vitro model. Since sarcomas are thought to be mesenchymal in origin (Clark et al., 2005), we turned to mesenchymal stem cells (MSCs), a multipotent population of cells that can differentiate into osteoblasts, chondrocytes, adipocytes, or myocytes (Pittenger et al., 1999).
From a one year-old adult Dicer f/f mouse, we prepared a primary culture of
MSCs that was then immortalized with a retroviral vector encoding SV40 large Tantigen (Figure 3A). Individual clones were isolated and the resulting cell lines analyzed by flow cytometry to confirm the expression of CD49e and CD106 (Fig
3B, left panels), surface markers associated with MSCs (Pittenger, 2008), and the absence of CD31, specific to endothelial cells, and CD45, a marker of hematopoietic stem cells (data not shown).
To delete Dicer, we carried out Adeno-Cre-GFP infection and sorted the cells for GFP to enrich for infected cells. A vast majority of cells generated by this protocol were indeed Dicer -/, as evidenced by the predominance of the
161

deletion-specific PCR product 6 days after sorting (Figure S3A). The efficiency of recombination early in this time course was supported by loss of Dicer protein
(Figure S3B), as well as decreases in mature miRNA levels by qPCR (Figure
S3C) and Northern at day 7 (Figure S3D). However, as observed for the sarcoma cells, additional passage led to a decrease in the deletion-specific PCR band (Figure S3A), suggesting outgrowth of Dicer-expressing cells in mixed culture.
To prevent competition by Dicer retaining cells, we repeated the strategy used for the sarcoma line and isolated clones following Adeno-Cre-GFP-infection and GFP-sorting. Of these clones, a majority (55%) had undergone homozygous deletion of Dicer, indicating that immortalized MSCs can readily tolerate Dicer loss. As before, Dicer -/ clones were stably proliferative and remained Dicer-null after multiple passages (>3 months) as determined by PCR genotyping (data not shown). To confirm mature miRNA loss, we carried out qPCR analysis of several ubiquitously expressed, abundant miRNAs (let-7, miR-24, -26, and -31) and observed a ~100-fold reduction of miRNA expression following Dicer loss (Figure
3D). Confirming functional loss, we observed a ~5-fold derepression of a let-7 luciferase reporter in Dicer -/ cells (Figure 3E). Finally, a comparison of growth rates revealed a proliferative lag in Dicer -/ MSCs, which have a doubling time of
~20 hours relative to ~14 hours in their Dicer f/f counterparts (Figure 3F). Notably,
Dicer -/ MSCs remained positive for CD49e and CD106 (Figure 3B, right panels) and negative for CD31 and CD45 (data not shown), suggesting a retention of cell identity in the absence of miRNAs.
162

Taken together, these studies demonstrate that Dicer loss and subsequent global microRNA loss are in fact tolerable in specific transformed states. The high frequency with which we isolated Dicer -/ clones from two different cell types suggests that miRNAs are not required for viability under basal conditions in vitro and argues against the possibility that proliferation of
Dicer -/ cells follows from additional compensatory mutations or events. The viability of these transformed cells leads to the intriguing conclusion that total miRNA loss itself, and resultant genetic misregulation, is not intrinsically lethal, but rather triggers secondary signals that actively initiate cell death pathways.
Notably, inactivation of p53 activity is a common feature of these models and may either facilitate or be required for viability in the absence of Dicer. This possibility is consistent with the observation that p53 loss allows primary MEFs to bypass an immediate senescence phenotype induced by Dicer loss (Mudhasani et al., 2008). However, p53 loss alone only prolongs proliferative capacity of
Dicer -/ MEFs for a few additional passages (personal communication with S.N.
Jones), suggesting that other events, such as activation of oncogenes or inactivation of additional tumor suppressor genes, are required.
Recent observations also suggest that the transformed state in general may be less dependent on miRNA regulation. In addition to the finding that global miRNA levels are reduced in tumors (Gaur et al., 2007; Lu et al., 2005), specific oncogenes such as myc are known to directly repress expression of numerous miRNAs (Chang et al., 2008). Furthermore, it has been reported that
163

rapid proliferation and oncogenic transformation promote genome-wide usage of shorter alternative 3’UTRs (Mayr and Bartel, 2009; Sandberg et al., 2008) resulting in a decrease in the targets of miRNA regulation. Our data unite these observations, suggesting that the unique convergence of globally decreased expression and diminished opportunity for targeting may collaborate to render the entire short RNA regulatory layer less important in the transformed state.
Sarcoma cells
The generation of Kras G12D , Trp53 -/, Dicer f/ sarcoma cells is detailed in a previous publication (Kumar et al., 2009). In brief, sarcomas were generated from hindlimb injection of Adeno-Cre into Kras LSL-G12D ; Trp53 f/f ; Dicer f/f mice as described previously (Kirsch et al., 2007). Tumor tissue was trypsinized and plated under standard conditions to give sarcoma cells. These cells were subsequently infected with MSCV.CreER
T2 .puro. Cells were selected with 2.5
!
g/ml puromycin for 3 days.
Cells were then treated with 250 nM 4-hydroxy tamoxifen for 24 hours to recombine the floxed Dicer allele and plated at a 1:5000 dilution in a 10 cm tissue culture dish. Colonies were selected after 7 days and genotyped as described by
Calabrese et al (Calabrese et al., 2007) once they reached confluence.
164

Mesenchymal stem cells
Primary MSC cultures were prepared similarly as previously described(Mukherjee et al., 2008) but with some differences as noted here. A one year-old Dicer f/f mouse(Harfe et al., 2005) was sacrificed by CO
2 asphyxiation. Tibia and femurs were isolated, cleaned of excess soft tissue, and crushed with mortar and pestle in PBS supplemented with 0.5% FBS (PBS/FBS).
The suspension was then filtered at 70-
µ m and centrifuged at 1000 rpm for 5 minutes. After resuspending the pellet in ACK lysis buffer for 4 minutes,
PBS/FBS was added to neutralize the lysis buffer. The sample was centrifuged again at 1000 rpm, washed once more with PBS/FBS, and centrifuged again.
The pellet was resuspended into 2 ml of alpha-MEM supplemented with pen/strep and 15-20% serum (Alpha-MEM Primary). The cells were plated into two T75 flasks in 10 ml each of Alpha-MEM Primary. The media was replaced
24-48 hours after plating.
Approximately 2 weeks after plating, primary MSCs were infected with a retroviral construct encoding SV40 large T-antigen. Multiple colonies grew out following infection and were passaged together as a polyclonal population in alpha-MEM supplemented with pen/strep and 10% FBS (Alpha-MEM Complete).
Individual clones were then isolated from this polyclonal population of cells, infected with Adeno-Cre-GFP, and sorted by FACS to isolate GFP-positive cells.
Sorted cells were then plated at low density and individual clones were grown out, passaged, and genotyped to determine Dicer status.
165

Tumor-bearing animals were sacrificed with CO
2
asphyxiation. Tumors were isolated, fixed in 4% paraformaldehyde, transferred to 70% ethanol, and embedded in paraffin. Tumors were then sectioned and stained with hematoxylin and eosin.
Cells were lysed with 2X sample buffer (Bio-Rad) containing 5% & mercaptoethanol, boiled for 5 minutes, and subjected to SDS–polyacrylamide gel electrophoresis (SDS-PAGE). After electrophoresis, proteins were transferred to
PVDF membranes using a submerged transfer apparatus (BioRad, Hercules,
CA). After blocking with 5% nonfat dried milk in TBS-T (50 mM Tris-HCl, pH 8.0,
150 mM NaCl, 0.1% Tween-20), the membrane was incubated with the primary antibody diluted in TBS-T (1:250 dilution for Dicer; 1:1000 for tubulin), washed extensively, and incubated with the appropriate horseradish peroxidase–linked secondary antibody (Amersham, Piscataway, NJ). Chemiluminescence was used for detection. Rabbit anti-Dicer and mouse anti-Tubulin antibodies (Sigma) were used.
RNA was prepared from 10 6 cells using Qiazol Reagent according to the manufacturer’s protocol (Qiagen). 20 !
g RNA was mixed with an equal volume of formamide loading buffer, denatured at 95º C for 5 minutes, and run for 1 hour at 35W on a 15% denaturing polyacrylamide gel (Sequagel, National
166

Diagnostics) after 30 minutes of pre-running. A semi-dry transfer apparatus set to 18V was used to transfer the RNA to a Hybond-N+ nylon membrane (GE
Healthcare Life Sciences) for 1.5 hours at 4°C. RNA was then UV crosslinked at
1.2 E6 uJoules in a Crosslinker 2400 (Stratagene) on top of Whatman paper.
The membrane was prehybridized with Ultrahyb oligo (Ambion) for 1 hour, and then probed overnight at 37°C with a 5’ end labeled locked nucleic acid (LNA) probe for let-7g. The membrane was washed twice for 30 minutes in
2xSSC/0.1% SDS buffer and then imaged on a Storm scanner (Molecular
Dynamics) for 24 hours. The membrane was then stripped and exposed to confirm probe removal prior to reprobing with LNAs for let-7g, mir-16, or mir-17.
Loading was confirmed with a DNA oligo probe for glutamine tRNA with 1 hour exposure.
Mesenchymal stem cells were washed with PBS and incubated with either a negative control isotype antibody or primary antibodies against CD49e and
CD106 (BD Biosciences), washed, stained with fluorescence-conjugated secondary antibodies, and sorted on a BD FACSCalibur flow cytometer
(Swanson Biotechnology Center). GFP positive cell sorting was similarly performed on live cells following Adeno-Cre-GFP infection to enrich for infected cells.
167

2.5 x 10 4 , 10 5 , or 10 6 Kras G12D ; Trp53 -/sarcoma cells of Dicer fl/- or Dicer -/genotype were suspended in PBS and subcutaneously injected into the flanks of
Rag2 -/ or C57Bl6/SV129 F1 mice. Tumors were measured over time by calipers and volumes were assessed as described previously (Sage et al., 2000). DNA,
RNA, and histological sections were prepared from tumors. Some tumors were trypsinized and replated for the development of secondary cell lines as described above. Protocols were performed with the approval of the Massachusetts
Institute of Technology Committee for Animal Care, and was consistent with the
Guide for Care and Use of Laboratory Animals, National Research Council, 1996
(institutional animal welfare assurance no. A-3125-01).
We thank members of the Sharp, Jacks, and Lees Labs for many helpful discussions. We are also specifically grateful to Margaret Ebert, Amy Seila, and
Eliezer Calo for their generous contribution of reagents. This work was supported by a Fannie and John Hertz Foundation Fellowship (A.R.), a Leukemia and
Lymphoma Society Grant (5198-09), an NIH grants RO1-GM34277 (P.A.S.), NCI grant PO1-CA42063 (P.A.S.), and the NCI Cancer Center Support (core) grant
P30-CA14051.
168

169

Figure 1. A. Derivation scheme for Dicer -/ sarcoma cells. Hindlimb injection of
Adeno-cre generates Kras G12D , p53 -/, Dicer f/ tumors. Clones isolated following
Cre-ER integration and tamoxifen treatment were genotyped by PCR to identify
Dicer -/ clones.
B. Northern analysis for precursor and mature miRNAs.
Glutamine tRNA was used to control for loading, and a dilutions series of Dicer f/-
RNA (1:1 to 1:16) is provided for quantitation. C. Luciferase reporter assays for abundant miRNAs. The renilla luciferase reporter contains six bulged sites for let-7, and two perfect sites for miR-16 and miR-17. Targeted renilla luciferase reporters were normalized to nontargeted firefly luciferase reporters.
Renilla/firefly luciferase expression was normalized to expression in the Dicer f/sarcoma cell line. D. Proliferation assay comparing Dicer f/ and Dicer -/ sarcoma cell lines.
Figure 2. A. Injection of 2.5 x 10 4 Dicer f/ and Dicer -/ sarcoma cells into the flanks of nude mice. Dicer -/ tumors were first apparent at 45 days, compared to 22 days for Dicer f/ cells.
B. Hematoxylin and eosin section of Dicer f/ and Dicer -/ tumors.
In C-F, each lane represents an independent tumor derived from one injection of the indicated Dicer -/ sarcoma cell line.
C. PCR genotyping of Dicer -/ tumors.
Recombined and floxed bands are derived from the injected tumor cells, while wild type bands derive from host tissue.
D. Northern analysis of tumor tissue derived from sarcoma injections.
E. PCR and F. Northern analysis following one
170

round of in vitro passage of secondary tumors. *denotes corresponding DNA and
RNA from the tumor sample with the greatest wild type contamination.
Figure 3. A. Schematic of MSC preparation. Primary MSC cultures were prepared from the tibia, femur, and pelvic bones of a one-year old Dicer f/f mouse.
The primary cells were then infected with retrovirus encoding SV40 large Tantigen. Monoclonal cultures were then isolated, infected with Adeno-Cre-GFP, sorted by FACS for GFP-positive cells, and plated at low density to isolate Dicerrecombined clones.
B. Cell surface marker expression in Dicer f/f (left) and Dicer -/-
(right) MSCs. Cells were analyzed by flow cytometry with antibodies against
CD49e and CD106.
C. PCR genotyping of clonally isolated Dicer f/f or Dicer -/-
MSCs. Clones 6.8 and 6.9 (lanes 3, 4) were derived from parental clone 6 (lane
2), and clones 12.2 and 12.4 (lanes 7, 8) were derived from parental clone 12
(lane 5). PCR genotyping of Dicer f/ sarcoma cell line was used as a heterozygous control (lane 1).
D. Expression of miRNAs in Dicer f/f and Dicer -/-
MSCs. Total RNA was analyzed with a Qiagen miScript qPCR assay for let-7, miR-24, -26, and -31). E. Luciferase reporter assay for let-7. The reporter contains 6 bulged sites. Targeted renilla luciferase reporters were normalized to nontargeted firefly luciferase reporters. Renilla/firefly luciferase expression was normalized to expression in the Dicer f/f MSC line.
F. Proliferation assay in Dicer f/f and Dicer -/ MSC lines.
171

Figure S1. Time course of acute Dicer deletion in Dicer f/ sarcomas. Dicer f/sarcoma cells containing Cre-ER were treated with tamoxifen to induce excision of the remaining Dicer allele, and then genotyped daily for 16 days.
Figure S2. Tumor injection time course for Dicer f/ and Dicer -/ cells in immune competent C57Bl6/SV129 F1 mice.
First detectable tumors appeared at 7 and 21 d, respectively.
Figure S3. A. PCR genotyping of Adeno-Cre-GFP-infected MSCs after FACS for
GFP. MSCs were analyzed 7 days post-sort by B. Dicer Western blot, C. qPCR, and D. Northern blot.
172

173

174

175

176

177

178

Arrate, M. P., Vincent, T., Odvody, J., Kar, R., Jones, S. N., and Eischen, C. M.
(2010). MicroRNA biogenesis is required for Myc-induced B-cell lymphoma development and survival. Cancer Res 70 , 6083-6092.
Bartel, D. P. (2009). MicroRNAs: target recognition and regulatory functions. Cell
136 , 215-233.
Bernstein, E., Kim, S. Y., Carmell, M. A., Murchison, E. P., Alcorn, H., Li, M. Z.,
Mills, A. A., Elledge, S. J., Anderson, K. V., and Hannon, G. J. (2003).
Dicer is essential for mouse development. Nat Genet 35 , 215-217.
Calabrese, J. M., Seila, A. C., Yeo, G. W., and Sharp, P. A. (2007). RNA sequence analysis defines Dicer's role in mouse embryonic stem cells.
Proc Natl Acad Sci U S A 104 , 18097-18102.
Chang, T. C., Yu, D., Lee, Y. S., Wentzel, E. A., Arking, D. E., West, K. M., Dang,
C. V., Thomas-Tikhonenko, A., and Mendell, J. T. (2008). Widespread microRNA repression by Myc contributes to tumorigenesis. Nat Genet 40 ,
43-50.
Clark, M. A., Fisher, C., Judson, I., and Thomas, J. M. (2005). Soft-tissue sarcomas in adults. N Engl J Med 353 , 701-711.
Friedman, R. C., Farh, K. K., Burge, C. B., and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res
19 , 92-105.
Gaur, A., Jewell, D. A., Liang, Y., Ridzon, D., Moore, J. H., Chen, C., Ambros, V.
R., and Israel, M. A. (2007). Characterization of microRNA expression levels and their biological correlates in human cancer cell lines. Cancer
Res 67 , 2456-2468.
Harfe, B. D., McManus, M. T., Mansfield, J. H., Hornstein, E., and Tabin, C. J.
(2005). The RNaseIII enzyme Dicer is required for morphogenesis but not patterning of the vertebrate limb. Proc Natl Acad Sci U S A 102 , 10898-
10903.
Kirsch, D. G., Dinulescu, D. M., Miller, J. B., Grimm, J., Santiago, P. M., Young,
N. P., Nielsen, G. P., Quade, B. J., Chaber, C. J., Schultz, C. P.
, et al.
(2007). A spatially and temporally restricted mouse model of soft tissue sarcoma. Nat Med 13 , 992-997.
Kumar, M. S., Lu, J., Mercer, K. L., Golub, T. R., and Jacks, T. (2007). Impaired microRNA processing enhances cellular transformation and tumorigenesis. Nat Genet 39 , 673-677.
Kumar, M. S., Pester, R. E., Chen, C. Y., Lane, K., Chin, C., Lu, J., Kirsch, D. G.,
Golub, T. R., and Jacks, T. (2009). Dicer1 functions as a haploinsufficient tumor suppressor. Genes Dev 23 , 2700-2704.
Lu, J., Getz, G., Miska, E. A., Alvarez-Saavedra, E., Lamb, J., Peck, D., Sweet-
Cordero, A., Ebert, B. L., Mak, R. H., Ferrando, A. A.
, et al.
(2005).
MicroRNA expression profiles classify human cancers. Nature 435 , 834-
838.
179

Mayr, C., and Bartel, D. P. (2009). Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells. Cell 138 , 673-684.
Mudhasani, R., Zhu, Z., Hutvagner, G., Eischen, C. M., Lyle, S., Hall, L. L.,
Lawrence, J. B., Imbalzano, A. N., and Jones, S. N. (2008). Loss of miRNA biogenesis induces p19Arf-p53 signaling and senescence in primary cells. J Cell Biol 181 , 1055-1063.
Mukherjee, S., Raje, N., Schoonmaker, J. A., Liu, J. C., Hideshima, T., Wein, M.
N., Jones, D. C., Vallet, S., Bouxsein, M. L., Pozzi, S.
, et al.
(2008).
Pharmacologic targeting of a stem/progenitor population in vivo is associated with enhanced bone regeneration in mice. J Clin Invest 118 ,
491-504.
Pittenger, M. F. (2008). Mesenchymal stem cells from adult bone marrow.
Methods Mol Biol 449 , 27-44.
Pittenger, M. F., Mackay, A. M., Beck, S. C., Jaiswal, R. K., Douglas, R., Mosca,
J. D., Moorman, M. A., Simonetti, D. W., Craig, S., and Marshak, D. R.
(1999). Multilineage potential of adult human mesenchymal stem cells.
Science 284 , 143-147.
Sage, J., Mulligan, G. J., Attardi, L. D., Miller, A., Chen, S., Williams, B.,
Theodorou, E., and Jacks, T. (2000). Targeted disruption of the three Rbrelated genes leads to loss of G(1) control and immortalization. Genes
Dev 14 , 3037-3050.
Sandberg, R., Neilson, J. R., Sarma, A., Sharp, P. A., and Burge, C. B. (2008).
Proliferating cells express mRNAs with shortened 3' untranslated regions and fewer microRNA target sites. Science 320 , 1643-1647.
180

The work included here represents an ongoing analysis of the changes accompanying
Dicer loss in somatic cells. As such, further experiments will be undertaken in order to confirm the findings described here.
Experimental contributions:
This work represents an equal collaboration between Arvind Ravi and Allan Gurtan.
Arvind Ravi performed experiments for the sarcoma cells as well as the crossdevelopment analysis.
Allan Gurtan performed experiments for the mesenchymal stem cells.
Additional contributions in analysis were made by Arjun Bhutkar and Charlie Whittaker.
181

Here, we present both global microarray expression profiling and highthroughput short RNA sequencing data from two somatic cell lines with and without Dicer, the critical enzyme required for the final maturation of nearly all mammalian microRNAs. Comparing data from a mouse sarcoma as well as a mouse mesenchymal stem cell line, we identify a shared microRNA profile of the mesenchymal state characterized by let-7, miR-22, miR-21, and miR-15/16. The majority of conserved targets show subtle changes on Dicer loss, consistent with a fine-tuning model of microRNA function. However, a distinct set of “switch-like” cellular targets (i.e., those with greater than 4-fold change) can be observed, and in the majority of cases, restore gene expression towards an embryonic stem cell
(ESC) state. Notably, a similar comparison from mouse ESCs with and without
Dicer identified an analogous set of “switch” targets in ESCs, suggesting that for a subset of mammalian genes, changes downstream of microRNA regulation are sufficient to explain a majority if not all of the expression change between the embryonic and somatic states. Given the ability of microRNAs to modulate both differentiation as well as reprogramming, these genes may represent key intermediates in the transcriptional landscape of the embryonic-somatic transition.
182

Organisms use many layers of regulation to ensure proper gene expression. From chromatin remodeling to ubiquitination, each step in gene expression is tightly controlled to achieve normal cellular function. Disruption of these regulatory layers can lead to severe phenotypic consequences at both the cellular and organismal levels. MicroRNAs (miRNAs), a class of short ~22 nucleotide RNAs, guide one such critical mode of regulation in mammalian cells.
Predicted to repress over half of all genes post-transcriptionally (Friedman et al.,
2009), alterations in the function of these short RNAs can disrupt a variety of fundamental processes including cell cycle control (Wang et al., 2008), apoptosis
(Chivukula and Mendell, 2008), and differentiation (Herranz and Cohen, 2010;
Stefani and Slack, 2008). In addition, the miRNA pathway has been conserved through much of animal evolution (Grimson et al., 2008), as have a number of critical target interactions (Chen and Rajewsky, 2006), underscoring the importance of their regulatory role.
Supporting the notion of an essential role for miRNAs, mouse knockouts of Dgcr8 and Dicer, key components of miRNA biogenesis, lead to lethality midgestation and exhibit severe differentiation defects (Bernstein et al., 2003;
Wang et al., 2007). This block in differentiation has been recapitulated in vitro with Dicer knockout embryonic stem (ES) cells (Kanellopoulou et al., 2005;
Murchison et al., 2005), which are unable to stably downregulate pluripotency genes, potentially in part through dysregulation of methylation (Benetti et al.,
2008; Sinkkonen et al., 2008). While the ES Dicer deletion model has been
183

valuable in understanding microRNA roles in early development, the inability to differentiate these cells complicates the study of global miRNA roles in the somatic state. As a result, most studies to date have utilized indirect means to infer the extent to which miRNA regulation contribute to state change in more differentiated tissues, such as global expression correlations between short
RNAs and mRNAs (Farh et al., 2005; Shkumatava et al., 2009; Tsang et al.,
2007). Therefore, the extent to which tissue-specific expression is achieved by miRNA-dependent vs. independent mechanisms remains unclear.
The characterization of stable Dicer null somatic cell lines would provide an alternate avenue towards the study of global miRNA roles later in development. Towards this end, we have recently generated two novel models that permit the study of complete Dicer loss in the somatic state: 1) a Kras G12D , p53 -/, Dicer f/ sarcoma cell line, and 2) an SV40 large T-antigen immortalized
Dicer f/f mouse mesenchymal stem cell (MSC) line (Ravi et al, manuscript in preparation). In both of these models, complete Cre-mediated recombination at the Dicer locus allows for the generation of stably proliferative Dicer null cells at high frequency, enabling study of the functional correlates of miRNA loss in homogenous populations. For simplicity, these cells and their corresponding
Dicer intact cell lines will be referred to by their Dicer status, ie., as Dicer f/f ,
Dicer f/, or Dicer -/ cells. Taking advantage of high throughput short RNA and expression array profiling from these novel cell lines as well as embryonic stem
(ES) cells, we provide a complete characterization in two independent systems of the consequences of miRNA loss at a genome-wide scale. These data provide a
184

valuable basis for the global understanding of regulation by short RNAs in mammalian cells.
To characterize the basal miRNA expression of mesenchymal cell types, we carried out massively parallel sequencing of small RNAs ranging from ~15-50 nucleotides in length. Both Dicer f/ sarcoma and Dicer f/f MSC libraries contained comparable sequencing depths at 9.3 and 9.9 million reads, respectively.
Mapping of these sequences to the mouse genome allowing 1 mismatch identified nearly half of the reads in each library as mature miRNAs (Figure 1A and 1B, upper panels). Notably, relaxing the mapping criteria to include up to 2 bases of variation in the 3´ end allowed an additional 20% of reads to be mapped as mature miRNAs (Figure S1). Thus, miRNA derived sequences represent the dominant fraction of short RNA reads in these datasets.
Of reads mapping to mature miRNAs, the six most abundant families in each library account for the majority of sequenced miRNAs. Four of these six families overlap between the sarcoma and MSC libraries, namely let-7 (32-37%), miR-22 (13-17%), miR-21 (3-6%), and miR-15/16 (4-8%) (Figure 1A and B, lower panels, Table S1). The let-7 dominance observed here is characteristic of a number of somatic tissues, including embryonic fibroblasts and neural precursors
(Marson et al., 2008). The similarity in the sarcoma and MSC profiles was
185

evident beyond simply the most abundant sequences, as the two samples showed a Pearson correlation of 0.8 when considering all miRNAs (Figure 1C).
Despite these similarities, we also observed modest expression differences between the two cell libraries. For instance, miR-21 expression was 3-fold lower in MSCs relative to the sarcoma. This expression change may be attributable to the Kras G12D allele in the sarcoma, as Kras has been reported to drive miR-21 expression in rat (Landgraf et al., 2007). Notably, reads from the miR-290-295 cluster, specific to embryonic stem cells, were negligible in number, and the global miRNA correlation was much lower with ES cell sequencing data ( r =
0.02), further distinguishing these cells as somatic (Figure 1D).
As miRNA profiles have been shown to be relatively faithful classifiers of cell of origin in human tissues (Lu et al., 2005), we additionally compared the global miRNA signatures of Dicer f/ sarcoma and Dicer f/f MSC lines to those from various somatic cell lines (Landgraf et al., 2007). A cross-correlation heatmap based on unsupervised hierarchical clustering was used to identify similarities in miRNA profiles across cell types (Figure 1E). The sarcoma and MSC datasets correlated most highly with one another, and belonged to a supercluster that included adipocytes ( r = 0.75 vs. sarcoma , r = 0.67 vs. MSC), a cell type into which MSCs are known to differentiate (Pittenger et al., 1999). Notably, data from our recent deep sequencing of ES cells (Leung et al., 2011) clustered distinctly from the somatic cell types reinforcing the distinctness of the embryonic and somatic states. Taken together, these results demonstrate that our sarcoma and
MSC lines have a shared miRNA expression profile and support the longstanding
186

hypothesis that these two cell types are derived from a common origin (Mohseny and Hogendoorn, 2011).
In order to evaluate the extent of miRNA loss in our Dicer null systems, we additionally performed small RNA sequencing in corresponding Dicer null cell lines (Ravi et al, manuscript in preparation). In contrast to the roughly half of all reads that mapped to mature miRNAs with 0 or 1 mismatches in Dicer intact libraries, fewer than 1% of reads mapped to mature miRNAs in Dicer -/ cells, with a commensurate increase in the remaining mapping categories (Figure 2A and
B, S1). In addition, a number of correlates of miRNA loss could be observed in this dataset, such as loss of a distinct 21-23 nt peak (Figure S3), and loss of the corresponding U bias in position 1 of this size range (Figure S4). Approximately
50% of mature miRNAs detected in Dicer intact cells became undetectable in
Dicer -/ cells, while the remainder of miRNAs underwent a median decrease of greater than 100-fold, confirming the global and dramatic loss of mature miRNAs with homozygous Dicer loss. For the most abundant miRNA seed families such as let-7 and miR-22, discussed in more detail below, we observed near-complete loss of miRNAs, with ~1,000 to ~1,000,000 fold reductions across the datasets
(Table S1).
Recently, it has been reported that miR-451 undergoes noncanonical
Dicer-independent, Ago2-dependent processing (Cheloufi et al., 2010; Cifuentes
187

et al., 2010). Therefore, we additionally looked for signatures of Dicer-dependent miRNAs. Combining data from the two datasets, we found little evidence of abundant miRNAs (i.e., having over 1000 reads, or 0.3% of the total miRNA pool) that show comparable abundance between Dicer intact and null libraries (Figure
2C and 2D). To better understand the nature of the residual miRNA reads in
Dicer -/ cells, we analyzed the cleavage site profiles of reads terminating within a
15 base window of the annotated Dicer cleavage site (Figure 2E). The diffuse distributions following Dicer loss, particularly at the 3´ end, suggests a model of inefficient processing by an alternative RNAse such as Drosha or Argonaute 2, possibly in conjunction with exonucleolytic processing. Notwithstanding these heterogeneous low abundance reads, the vast majority of the cellular miRNA pool appears to be eliminated on Dicer loss.
-/-
Computational predictions estimate that half of all protein-coding genes are regulated by miRNAs (Friedman et al., 2009). Therefore, to determine the effect of miRNA loss on gene expression in our Dicer null somatic models, we performed microarray expression profiling in both our sarcoma and MSC models.
Hierarchical clustering of the samples classified them primarily by Dicer status
(Figure 3A, S5) indicating that the changes observed in gene expression are sufficiently consistent and significant to distinguish Dicer intact and deficient cells. In addition, examination of the most significantly changed genes revealed
188

a number with conserved target sites for abundant microRNAs identified earlier, suggesting their upregulation may be due in part to loss of direct miRNA repression (Figure 3A).
As many miRNA-target interactions can have relatively subtle effects on gene expression, we additionally plotted cumulative distribution function (CDF) curves to determine if predicted targets of individual miRNAs show a general increase in expression upon Dicer loss (Figure 3B). Compared to control transcripts matched for 3´ UTR length, GC-content, and gene expression level, statistically significant shifts were observed within the sarcoma dataset for targets of abundant miRNAs such as let-7 (p=2x10 -13 ), miR-22 (p=1x10 -9 ), and miR-15
(p=1x10 -16 ). In contrast, shifts were of much lower statistical significance for lowly and non-expressed miRNAs, such as miR-138 (p=.17), miR-331 (p=.04), miR-122 (p=.005), and miR-204 (p=.00014). Similar results were observed with the MSC dataset (data not shown). Finally, Ingenuity Pathway Analysis identified altered gene expression networks consistent with known and predicted targeting functions of expressed miRNAs. For instance, alterations in ‘Cell Cycle,’ ‘Cell
Death,’ and ‘Cellular Compromise’ were detected, which have been previously linked to miR-15/16, miR-24, let-7, miR-221/222, and miR-17 (Chivukula and
Mendell, 2008; Subramanian and Steer, 2010). These results suggest that derepression of direct miRNA targets contributes to global alterations in gene expression in Dicer -/ cells relative to their Dicer f/ counterparts.
To carry out a more direct comparison of Dicer loss in the sarcoma and
MSC lines (which could not be directly correlated due to use of different platforms), we compared the expression array data using Gene Set Enrichment
189

Analysis (GSEA), which can sensitively identify concerted changes in a group of genes across an experimental condition (Subramanian et al., 2005). For this analysis, we constructed two custom gene sets based on our sarcoma data: (1)
“Arv_High-in-null”, which consists of genes upregulated in the sarcoma following
Dicer loss and likely encompasses direct miRNA targets, and (2) “Arv_High-inhet”, which consists of genes downregulated in the sarcoma line upon Dicer loss and therefore captures changes secondary to derepression of miRNA targets.
When compared to the MSC expression data, both gene sets were significantly enriched in their respective Dicer f/f or Dicer -/ arrays (p < 0.001), suggesting that not only do the two models share a number of direct miRNA targets, as would be predicted from their similar miRNA expression, but they additionally share more global perturbations of downstream networks.
A number of studies have suggested that miRNA regulation plays a critical role in defining cell identity (Christodoulou et al., 2010; Farh et al., 2005; Lim et al., 2005). In order to identify how global miRNA loss impacts somatic identity, we used supervised Principle Component Analysis (PCA) to compare the expression between ES cells and somatic cells with and without Dicer. In this analysis, genes that change most significantly between the embryonic and somatic states (using available data from MEFs and iPS cells) are used to define
190

a single axis on which each array data set can be plotted. Projection of Dicer null cell lines onto this axis revealed that they clustered most closely with their parental Dicer intact counterparts (Figure 4A and 4B, upper panels), suggesting that miRNA-independent regulation is sufficient to grossly maintain cell identity.
To confirm that this method is sensitive to transitions along the embryonicsomatic axis, we additionally plotted available reprogramming datasets in which mouse embryonic fibroblasts containing inducible four-factor constructs (Oct-4,
Sox-2, Klf-4, c-Myc) were converted to stable induced pluripotent stem (iPS) cells
(Mikkelsen et al., 2008; Samavarchi-Tehrani et al., 2010). In both cases, cells show a clear trajectory in the embryonic direction, confirming the ability of this analysis to detect biologically relevant changes in cell state (Figure 4A and 4B, lower panels).
While the general identity of cells appears similar regardless of genotype, we attempted to determine whether a more subtle shift in cell state was induced by Dicer loss. To this end, we first examined the transcription factors that most strongly define each cell types. Comparing expression profiles from Dicer-intact cells, we identified the 10 most differentially expressed transcription factors between the embryonic and mesenchymal states based on each array platform
(Table S4). Overlapping these sets, we re-identified a number of the core ES transcription factors, including Pou5f1 (Oct4), Nanog, and Sox2. Although a comparable set of core factors is not presently known for the mesenchymal state, we identified a candidate set of factors in an analogous fashion, namely Prrx1,
Runx2, Tbx15, and Nfix. Comparison of expression between these factors
191

showed that none had reproducible changes with respect to Dicer status (Figure
S6), consistent with the gross retention of cell identity in the absence of Dicer.
To determine more specifically whether miRNA loss may drive changes relevant to cell identity, we looked at the set of conserved targets for miR-295 and let-7, the most abundant miRNAs in ES and mesenchymal cells, comparing their expression change on Dicer loss to their change across cell identity. The majority of targets showed relatively little change on Dicer loss (< 2-fold), consistent with a model of miRNAs as fine-tuners of gene expression (Figure 4C and D, Figure S7A and S7B). However, a subset of genes appeared to be more switch-like, showing greater than 4-fold expression changes following deletion of
Dicer. Indeed, among these genes with large expression changes, the majority showed the same direction of change on Dicer loss as they did in the opposing cell state in which their targeting miRNA is not expressed (Figure 4E and F,
Figure S8A and S8B). For instance, Lats2, a well-characterized miR-295 target
(Wang et al., 2008), is expressed roughly 12-fold higher following miRNA loss in
ES cells, and roughly 50-fold higher in somatic cells. Interestingly, even for the conserved targets that underwent a decrease in expression following Dicer loss – suggestive of indirect regulation – the expression change across cell types often mirrored the downregulation observed. Taken together, these findings identify a number of genes for which regulation downstream of miRNAs may significantly contribute to cell type specific gene expression.
192

Here, we present global expression profiling of the small RNA and mRNA populations of two independent models of complete Dicer loss, providing a unique window into global miRNA function in the somatic state. Informatic analysis of this profile data along with existing data in the literature allowed us to define a number of features that may broadly define the mesenchymal state, including predominant expression of let-7 and miR-22, as well as expression of key transcription factors such as Prrx1 and Runx2. Profiling data from matched
Dicer -/ datasets confirmed widespread miRNA loss, supporting the value of this dataset for the dissociation of miRNA-dependent and miRNA-independent celltype specific gene expression.
Changes in cell state are of particular interest in this system as a number of studies suggest that miRNAs are key enforcers of cell identity: 1) microRNA expression is a superior classifier relative to mRNA expression for the identification of cell type (Lu et al., 2005), 2) many conserved miRNAs also retain tissue specificity across a range of species, suggesting that they play conserved functions in cell type definition (Christodoulou et al., 2010), and 3) modulation of miRNAs has been shown to promote cell identity changes, such as somatic cell reprogramming, either in concert with other core transcription factors (Judson et al., 2009; Melton et al., 2010), or on its own (Anokye-Danso et al., 2011; Liao et al., 2011). Despite these strong relationships between known miRNAs and cellular identity, our data suggest that miRNA-independent mechanisms are sufficient to maintain many of the features of the ES and mesenchymal states.
193

Although the global transcriptional profiles of Dicer null cells most resemble their Dicer intact counterparts, a number of changes downstream of miRNA regulation are evident. First, significant CDF shifts can be observed for a number of highly expressed miRNAs, suggesting that many targets are derepressed in response to miRNA loss. In addition, IPA analysis suggests that these alterations have functional consequences downstream such as proliferative dysregulation, which is indeed observed in culture in these models (Ravi et al, manuscript in preparation).
Upon examination of individual miRNA targets, expression changes in large part support a model of miRNAs as fine-tuners of gene expression (Bartel and Chen, 2004), with roughly 80% of conserved targets changing by less than
2-fold on Dicer loss. Although this is consistent with the magnitude of repression conferred by single target sites, because these systems involves simultaneous loss of all miRNAs, it additionally suggests that the sum of cooperative interactions still leads to only modest expression changes within a given cell type in many cases.
In spite of the dominant trends suggesting potentially modest expression changes in the absence of miRNAs, we were able to identify a number of “switchlike” targets (~5% of conserved targets) undergoing greater than 4-fold changes in expression. Because this study examines miRNAs in an endogenous context, some of these expression changes are likely to be mediated by indirect pathways, particularly those involving decreases upon miRNA loss. Whether direct or indirect, the relatively strong concordance between these changes and
194

those observed between cell types suggests that these interactions may be critical drivers of large expression changes physiologically.
An important addition to this work in the near term will be the validation of these expression changes using quantitative PCR, and the evaluation of direct miRNA-mediated repression extent using luciferase constructs. These additional assays will be necessary to confirm that the expression changes reported here are downstream of specific miRNAs such as miR-295 and let-7.
Although many strong miRNA targets appeared to be re-expressed, there are a number of notable exceptions. Genes such as Lin28, known to be involved in let-7 processing and itself a characterized let-7 target (Rybak et al., 2008;
Viswanathan et al., 2008), failed to be detected upon somatic Dicer loss, suggesting that alternative epigenetic mechanisms reinforce miRNA repression in somatic cells. Further studies uncovering relationships between posttranscriptional regulation, chromatin changes, and DNA methylation will help uncover the extent to which these layers of control collaborate to reinforce expression programs.
For those genes that do show switch-like behavior, the large degree of reversibility in expression raises the question of whether certain physiological conditions might recapitulate Dicer or miRNA loss in vivo . In the case of ES cells, downregulation of the miR-290-295 cluster occurs concomitantly with differentiation (Houbaviy et al., 2003), suggesting a motivation for active transcription and simultaneous repression of a gene. However, instances in which somatic miRNAs such as let-7 and miR-22 are downregulated are relatively less well characterized, as studies to date suggest that during normal
195

development and differentiation, miRNA expression is enhanced (Hwang et al.,
2009; Lu et al., 2005).
One potential driver of such decreases in miRNA repression may be cellular stress. For instance, NF-kB expression has been shown to participate in a complex negative feedback loop with let-7, regulating cellular transformation in mammary epithelial cells (Iliopoulos et al., 2009). In addition, lipopolysaccharide
(LPS) stimulation, an indirect inducer of NF-kB signaling, has been shown to downregulate let-7 in cholangiocytes (Hu et al., 2009). Thus, rather than being essential for linear developmental trajectories, re-expression of these targets may accompany a reversible network of dedifferentiation that accompanies the stress response and/or oncogenic transformation.
In support of such a model, a number of the somatic switch targets identified here have been implicated in oncogenic transformation. For instance,
Igf2bp family members, which act to stabilize the oncogene Myc among other cellular targets, are known to be overexpressed in a wide variety of human tumors (Yisraeli, 2005). Nr6a1 (GCNF) has been identified as a marker of germ cell tumors, and in particular seminomas (Juric et al., 2005). Given their ability to undergo sizeable expression changes in response to altered post-transcriptional regulation, these switch genes may be critical mediators that directly link the widespread downregulation of miRNAs in cancer to the transformed state.
196

The generation of Kras G12D , Trp53 -/, Dicer f/ sarcoma cells is detailed in a previous publication (Kumar et al., 2009), and subsequent infection with
MSCV.CreER
T2 .puro followed by tamoxifen treatment were performed to generate subsequent Dicer -/ sarcoma cells (Ravi et al., manuscript in preparation). Dicerf/f and Dicer-/- were generated as previously described
(Mukherjee et al., 2008) using SV40 Large T-antigen for immortalization (Ravi et al., manuscript in preparation).
For sarcoma samples, Dicer f/ and Dicer -/ cells were grown to confluency in T25s and total RNA was prepared using a Qiagen RNeasy kit (with manufacturer protocol). For MSCs, three clones of Dicer f/f and four clones of
Dicer -/ cells were grown to confluence in 6-well plates, after which total RNA was prepared with Qiazol reagent (with manufacturer protocol). Sarcoma and MSC samples were analyzed on an Affymetrix GeneChip Mouse Genome 430 2.0 or
GeneChip Mouse Exon 1.0 ST arrays, respectively, by the BioMicro Center core facility. The microarray data were processed in the Swanson Biotechnology
Center. CEL files were processed with Partek software using the GCRMA method. Hierarchical clustering for datasets was performed based on average linkage distance of the cross-correlation matrix.
197

GSEA (Mootha et al., 2003; Subramanian et al., 2005) was performed on both sarcoma and MSC expression data with redundant probes collapsed by maximum value and highest confidence criteria. GSEA was performed with 1000 permutations of the gene sets, a weighted enrichment statistic, and a signal-tonoise metric.
Small RNA sequencing from sarcoma and MSC lines was carried out as described previously (Leung et al., in press, NSMB). Raw sequence reads (after linker stripping) were mapped to the mouse genome (build mm9) using the
Bowtie short read alignment tool (Langmead et al., 2009), allowing for a single base pair mismatch per unique alignment (U01) to accommodate sequencing errors. All such unique matches to the genome were first processed to identify hits to microRNA annotations (from miRbase release 14). Reads mapping to fulllength mature miRNA or miRNA* annotations with exact ends were categorized as mature miRNAs. These were subsequently used to determine the level of mature miRNAs in various datasets under study. Reads that mapped to the pre- miRNA hairpin but did not meet the full-length mature miRNA criterion were categorized as pre-miRNA overlaps.
UCSC mouse genome (mm9) annotation sets (http://genome.ucsc.edu) for coding transcripts were then used to identify reads mapping to exonic CDS,
3’UTR, and 5’UTR regions. The remaining reads were then processed to identify hits to non-coding RNA annotations using: ncRNA from fRNAdb ver 3.4
198

(http://www.ncrna.org/frnadb/), Genomic tRNA Database
(http://gtrnadb.ucsc.edu/), and Rfam ver. 9.1 (http://rfam.sanger.ac.uk/).
Following this, reads mapping to intronic regions (http://genome.ucsc.edu) were identified. The remaining reads were classified as “intergenic”.
Reads classified as mature miRNA and pre-miRNA hits were further analyzed using custom software to determine the distribution of read-ends relative to the 5’ or 3’ ends of mature-miRNA annotations. Additionally, read-end distributions for reads anchored at either annotated end of mature-miRNAs were also obtained. Using mature miRNA sequence data from Chiang et al (Chiang et al., 2010), the prevalence of known miRNA seeds within reads mapped to mature miRNAs was also determined for various datasets. Libraries were normalized to account for the variable sequencing depth obtained. The mean of the median fold-change between libraries was used to normalize reads in the Dicer -/datasets, and corresponded to a correction factor of 2.2 in the sarcoma dataset and 2.3 in the MSCs.
MicroRNA profiles corresponding to cell lines were obtained from a previous miRNA expression atlas (Landgraf et al., 2007), as well as recently published data from our laboratory for ES cells (Leung et al., 2011).
Unsupervised hierarchical clustering based on profile correlations was used with linkage determined by average distance. Clustering order was then used to produce a heatmap of cross-correlation data.
199

PCA was performed on global expression data, with dicer intact mesenchymal or ES cell types used to define the first weights used for the first principle component axis, PC1. PC1 therefore captures the maximium variance observed across the embryonic-somatic axis in the data. The gene weights were then used to project the remaining expression datasets onto PC1 in order to compare changes on Dicer loss, or in the case of the reprogramming timecourses, progressive transitions between the somatic and embryonic states.
We thank members of the Sharp, Jacks, and Lees Labs for many helpful discussions. This work was supported by a Fannie and John Hertz Foundation
Fellowship (A.R.), a Leukemia and Lymphoma Society Grant (5198-09), an NIH grants RO1-GM34277 (P.A.S.), NCI grant PO1-CA42063 (P.A.S.), and the NCI
Cancer Center Support (core) grant P30-CA14051.
200

Figure 1.
A.
Distribution of reads following mapping to mouse genome with 0 or
1 mismatches (upper panel), and subdivision of miRNA mapped reads to annotated miRNAs (lower panel). B.
Mapping and miRNA distribution data for mesenchymal stem cell (MSC) library. Size distribution of reads in Dicer f/ and
Dicer -/ libraries. C.
Scatterplot of mature miRNA reads between sarcoma and
MSC libraries, or D.
between ES and MSC libraries. E.
Hierarchical clustering and cross-correlation heatmap of Sarcoma and MSC miRNA profiles along with those of various somatic cells and embryonic stem cells. MSC and sarcoma datasets clustered most similarly with each other ( r = 0.8), while ES cells clustered distinctly from all somatic cell types.
Figure 2. A.
Distribution of small RNA reads following Dicer loss in sarcoma and
B.
MSC datasets. Remaining reads mapping to mature miRNAs constituted less than 1% of the total reads in each library. C.
Dicer dependence plot showing relative reads in Dicer -/ and Dicer f/ libraries for the sarcoma cell line, or D.
Dicer -
/ and Dicer f/f libraries for the MSC cell line. E.
Distribution of 5´ and 3´ ends of reads in the sarcoma libraries near predicted Dicer cleavage sites. Only reads containing the precise Drosha predicted termini were considered.
Figure 3.
A.
Heatmap of expression data from Dicer f/ and Dicer -/ sarcoma cells.
Unsupervised hierarchical clustering classified samples by Dicer status. Top 20 significantly changed genes are plotted for each genotype. Several genes
201

enriched in Dicer -/ cells contain conserved target sites for abundant miRNA identified by sequencing. B. Cumulative density function plots in sarcoma cell lines of conserved Targetscan sites for highly expressed miRNAs. Log fold change in expression between Dicer f/ and Dicer -/ cells is shown. Nontargeted controls (blue) were selected to have similar length and base composition as predicted targets (green). Left-shifts of targets relative to controls indicate target derepression in Dicer -/ sarcoma cells.
Figure 4. A. Principal components analysis (PCA) of global 3´ UTR expression data from Dicer f/ sarcoma and ES cells, and their Dicer -/ counterparts (upper). 3´
UTR data from a fibroblast reprogramming experiment are included for reference, and were used for training of the axis (lower) (Mikkelsen et al., 2008). MEF, d4 =
4 days into induction timecourse of Oct4, Sox2, Klf4, and c-Myc; MCV6, MCV8 = partially reprogrammed iPS lines; MCV8.1 – fully reprogrammed iPS line. B.
PCA comparing exon array data from MSC and ES cells in the presence and absence of Dicer (upper). Exon array data from a reprogramming timecourse are included for reference and were used for training of the axis (lower) (Samavarchi-
Tehrani et al., 2010). MEF = parental MEF line; MEF d0-d21 = MEF line with inducible expression of Oct4, Sox2, Klf4, and c-Myc; 6CP and 6CS = fully reprogrammed iPS cell lines. C.
Scatterplot of conserved miR-295 targets and their expression difference between Dicer f/ ES and sarcoma cell types versus their expression change on Dicer loss in ES cells. Switch targets (those that change over 4-fold on Dicer loss) are shown in red. D.
Scatterplot of fold change of conserved let-7 targets and their expression difference between Dicer f/ ES
202

and sarcoma cells versus their expression change on Dicer loss in sarcoma cells.
E.
Heatmap of switch targets (genes undergoing a greater than 4-fold change on
Dicer loss) in ES cells for miR-295 and their relative expression in Dicer -/ ES cells and in Dicer f/ sarcoma cells. F.
A similar heatmap for somatic switch targets of let-7. In both cases, the majority of expression changes are coherent with the change seen across cell types.
Table S1.
MicroRNA profiling of Dicer intact and null sarcoma and MSC cells.
Sequences are sorted by maximum number of reads in Dicer f/ sarcoma cells, and were identified as having an exact sequence alignment to the annotated mature miRNA with up to 1 mismatch.
Table S2. Embryonic and mesenchymal transcription factors with the greatest cell type specificity. The most differentially expressed transcription factors were identified from ES-sarcoma comparisons in 3´ UTR array data and ES-MSC comparisons in exon array data are shown, along with the intersection of these lists (in bold).
Figure S1.
Mapped annotations of all reads sequenced in Dicer intact sarcoma and MSC libraries. ‘Mature miRNA +/- 2bp 3´ slip’ indicates reads that terminate within 2 bases of the annotated 3´ end of a mature miRNA, but an exact 5´ end.
‘Non-template/Edited mature miRNA (R01)’ indicates reads that contain the central 15 bases of a mature miRNA but have at least 2 mismatches to the full annotated miRNA sequence. ‘< 15 bp Repeat mapped (R01)’ includes
203

sequences under 15 bases in total length that mapped repetitively. ncRNA = noncoding RNA, CDS = coding sequence.
Figure S2.
Full annotations of all reads in Dicer -/ sarcoma and MSC libraries.
‘Mature miRNA +/- 2bp 3´ slip’ indicates reads that terminate within 2 bases of the annotated 3´ end of a mature miRNA, but an exact 5´ end. ‘Nontemplate/Edited mature miRNA (R01)’ indicates reads that contain the central 15 bases of a mature miRNA but have at least 2 mismatches to the full annotated miRNA sequence. ‘< 15 bp Repeat mapped (R01)’ includes sequences under 15 bases in total length that mapped repetitively. ncRNA = noncoding RNA, CDS = coding sequence.
Figure S3. Size distribution of reads from small RNA sequencing in sarcoma
(upper) and MSC (lower) libraries. Dicer loss is accompanied by loss of the dominant 22-23 base peak seen in Dicer intact libraries.
Figure S4. First base distribution by sequence length for Dicer f/ and Dicer -/sarcoma libraries. Total reads from small RNA cloning were binned by sequence length, and the frequency of a given base being the first in the sequence is shown.
Figure S5. Heatmap of expression profiling data from Dicer f/f and Dicer -/ MSCs.
Unsupervised hierarchical clustering classified samples primarily by Dicer status.
204

Figure S6. A.
Expression profiling data from ES cells and sarcoma cells with and without Dicer, comparing the most strongly ES-specific transcription factors
(upper) and somatic transcription factors (lower). B.
A similar comparison was performed for ES and MSC cells.
Figure S7.
A.
Scatterplot of conserved miR-295 targets and their expression difference between Dicer f/f ES and MSC cell types versus their expression change on Dicer loss in ES cells. Switch targets (those that change over 4-fold on Dicer loss) are shown in red. B.
Scatterplot of fold change of conserved let-7 targets and their expression difference between Dicer f/f ES and MSC cells versus their expression change on Dicer loss in sarcoma cells.
Figure S8. A.
Heatmap of switch targets (genes undergoing a greater than 4-fold change on Dicer loss) in ES cells for miR-295 and their relative expression in
Dicer -/ ES cells and in Dicer f/f MSCs. B.
A similar heatmap for somatic switch targets of let-7. In both cases, the majority of expression changes are coherent with the change seen across cell types.
205

206

207

208

209

2
210
2

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

Anokye-Danso, F., Trivedi, C. M., Juhr, D., Gupta, M., Cui, Z., Tian, Y., Zhang,
Y., Yang, W., Gruber, P. J., Epstein, J. A., and Morrisey, E. E. (2011).
Highly Efficient miRNA-Mediated Reprogramming of Mouse and Human
Somatic Cells to Pluripotency. Cell Stem Cell 8 , 376-388.
Bartel, D. P., and Chen, C. Z. (2004). Micromanagers of gene expression: the potentially widespread influence of metazoan microRNAs. Nat Rev Genet
5 , 396-400.
Benetti, R., Gonzalo, S., Jaco, I., Munoz, P., Gonzalez, S., Schoeftner, S.,
Murchison, E., Andl, T., Chen, T., Klatt, P.
, et al.
(2008). A mammalian microRNA cluster controls DNA methylation and telomere recombination via Rbl2-dependent regulation of DNA methyltransferases. Nat Struct Mol
Biol 15 , 998.
Bernstein, E., Kim, S. Y., Carmell, M. A., Murchison, E. P., Alcorn, H., Li, M. Z.,
Mills, A. A., Elledge, S. J., Anderson, K. V., and Hannon, G. J. (2003).
Dicer is essential for mouse development. Nat Genet 35 , 215-217.
Cheloufi, S.
, Dos Santos, C. O.
, Chong, M. M.
, and Hannon, G. J.
( 2010 ). A dicer-independent miRNA biogenesis pathway that requires Ago catalysis .
Nature 465 , 584-589 .
Chen, K., and Rajewsky, N. (2006). Deep conservation of microRNA-target relationships and 3'UTR motifs in vertebrates, flies, and nematodes. Cold
Spring Harb Symp Quant Biol 71 , 149-156.
Chiang, H. R., Schoenfeld, L. W., Ruby, J. G., Auyeung, V. C., Spies, N., Baek,
D., Johnston, W. K., Russ, C., Luo, S., Babiarz, J. E.
, et al.
(2010).
Mammalian microRNAs: experimental evaluation of novel and previously annotated genes. Genes Dev 24 , 992-1009.
Chivukula, R. R., and Mendell, J. T. (2008). Circular reasoning: microRNAs and cell-cycle control. Trends Biochem Sci 33 , 474-481.
Christodoulou, F., Raible, F., Tomer, R., Simakov, O., Trachana, K., Klaus, S.,
Snyman, H., Hannon, G. J., Bork, P., and Arendt, D. (2010). Ancient animal microRNAs and the evolution of tissue identity. Nature 463 , 1084-
1088.
Cifuentes, D.
, Xue, H.
, Taylor, D. W.
, Patnode, H.
, Mishima, Y.
, Cheloufi, S.
, Ma,
E.
, Mane, S.
, Hannon, G. J.
, Lawson, N. D.
, et al.
( 2010 ). A novel miRNA processing pathway independent of Dicer requires Argonaute2 catalytic activity . Science 328 , 1694-1698 .
Farh, K. K., Grimson, A., Jan, C., Lewis, B. P., Johnston, W. K., Lim, L. P.,
Burge, C. B., and Bartel, D. P. (2005). The widespread impact of mammalian MicroRNAs on mRNA repression and evolution. Science 310 ,
1817-1821.
Friedman, R. C., Farh, K. K., Burge, C. B., and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res
19 , 92-105.
Grimson, A., Srivastava, M., Fahey, B., Woodcroft, B. J., Chiang, H. R., King, N.,
Degnan, B. M., Rokhsar, D. S., and Bartel, D. P. (2008). Early origins and
228

evolution of microRNAs and Piwi-interacting RNAs in animals. Nature 455 ,
1193-1197.
Herranz, H., and Cohen, S. M. (2010). MicroRNAs and gene regulatory networks: managing the impact of noise in biological systems. Genes Dev 24 , 1339-
1344.
Houbaviy, H. B., Murray, M. F., and Sharp, P. A. (2003). Embryonic stem cellspecific MicroRNAs. Dev Cell 5 , 351-358.
Hu, G., Zhou, R., Liu, J., Gong, A. Y., Eischeid, A. N., Dittman, J. W., and Chen,
X. M. (2009). MicroRNA-98 and let-7 confer cholangiocyte expression of cytokine-inducible Src homology 2-containing protein in response to microbial challenge. J Immunol 183 , 1617-1624.
Hwang, H. W., Wentzel, E. A., and Mendell, J. T. (2009). Cell-cell contact globally activates microRNA biogenesis. Proc Natl Acad Sci U S A 106 ,
7016-7021.
Iliopoulos, D., Hirsch, H. A., and Struhl, K. (2009). An epigenetic switch involving
NF-kappaB, Lin28, Let-7 MicroRNA, and IL6 links inflammation to cell transformation. Cell 139 , 693-706.
Judson, R. L., Babiarz, J. E., Venere, M., and Blelloch, R. (2009). Embryonic stem cell-specific microRNAs promote induced pluripotency. Nat
Biotechnol 27 , 459-461.
Juric, D., Sale, S., Hromas, R. A., Yu, R., Wang, Y., Duran, G. E., Tibshirani, R.,
Einhorn, L. H., and Sikic, B. I. (2005). Gene expression profiling differentiates germ cell tumors from other cancers and defines subtypespecific signatures. Proc Natl Acad Sci U S A 102 , 17763-17768.
Kanellopoulou, C., Muljo, S. A., Kung, A. L., Ganesan, S., Drapkin, R., Jenuwein,
T., Livingston, D. M., and Rajewsky, K. (2005). Dicer-deficient mouse embryonic stem cells are defective in differentiation and centromeric silencing. Genes Dev 19 , 489-501.
Kumar, M. S., Pester, R. E., Chen, C. Y., Lane, K., Chin, C., Lu, J., Kirsch, D. G.,
Golub, T. R., and Jacks, T. (2009). Dicer1 functions as a haploinsufficient tumor suppressor. Genes Dev 23 , 2700-2704.
Landgraf, P., Rusu, M., Sheridan, R., Sewer, A., Iovino, N., Aravin, A., Pfeffer, S.,
Rice, A., Kamphorst, A. O., Landthaler, M.
, et al.
(2007). A mammalian microRNA expression atlas based on small RNA library sequencing. Cell
129 , 1401-1414.
Langmead, B., Trapnell, C., Pop, M., and Salzberg, S. L. (2009). Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10 , R25.
Leung, A. K., Young, A. G., Bhutkar, A., Zheng, G. X., Bosson, A. D., Nielsen, C.
B., and Sharp, P. A. (2011). Genome-wide identification of Ago2 binding sites from mouse embryonic stem cells with and without mature microRNAs. Nat Struct Mol Biol 18 , 237-244.
Liao, B., Bao, X., Liu, L., Feng, S., Zovoilis, A., Liu, W., Xue, Y., Cai, J., Guo, X.,
Qin, B.
, et al.
(2011). Microrna cluster 302-367 enhances somatic cell reprogramming by accelerating a mesenchymal-to-epithelial transition. J
Biol Chem.
Lim, L. P., Lau, N. C., Garrett-Engele, P., Grimson, A., Schelter, J. M., Castle, J.,
Bartel, D. P., Linsley, P. S., and Johnson, J. M. (2005). Microarray
229

analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 433 , 769-773.
Lu, J., Getz, G., Miska, E. A., Alvarez-Saavedra, E., Lamb, J., Peck, D., Sweet-
Cordero, A., Ebert, B. L., Mak, R. H., Ferrando, A. A.
, et al.
(2005).
MicroRNA expression profiles classify human cancers. Nature 435 , 834-
838.
Marson, A., Levine, S. S., Cole, M. F., Frampton, G. M., Brambrink, T.,
Johnstone, S., Guenther, M. G., Johnston, W. K., Wernig, M., Newman, J.
, et al.
(2008). Connecting microRNA genes to the core transcriptional regulatory circuitry of embryonic stem cells. Cell 134 , 521-533.
Melton, C., Judson, R. L., and Blelloch, R. (2010). Opposing microRNA families regulate self-renewal in mouse embryonic stem cells. Nature 463 , 621-
626.
Mikkelsen, T. S., Hanna, J., Zhang, X., Ku, M., Wernig, M., Schorderet, P.,
Bernstein, B. E., Jaenisch, R., Lander, E. S., and Meissner, A. (2008).
Dissecting direct reprogramming through integrative genomic analysis.
Nature 454 , 49-55.
Mohseny, A. B., and Hogendoorn, P. C. (2011). Mesenchymal Tumors: When
Stem Cells Go Mad. Stem Cells.
Mootha, V. K., Lindgren, C. M., Eriksson, K. F., Subramanian, A., Sihag, S.,
Lehar, J., Puigserver, P., Carlsson, E., Ridderstrale, M., Laurila, E.
, et al.
(2003). PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat
Genet 34 , 267-273.
Mukherjee, S., Raje, N., Schoonmaker, J. A., Liu, J. C., Hideshima, T., Wein, M.
N., Jones, D. C., Vallet, S., Bouxsein, M. L., Pozzi, S.
, et al.
(2008).
Pharmacologic targeting of a stem/progenitor population in vivo is associated with enhanced bone regeneration in mice. J Clin Invest 118 ,
491-504.
Murchison, E. P., Partridge, J. F., Tam, O. H., Cheloufi, S., and Hannon, G. J.
(2005). Characterization of Dicer-deficient murine embryonic stem cells.
Proc Natl Acad Sci U S A 102 , 12135-12140.
Pittenger, M. F., Mackay, A. M., Beck, S. C., Jaiswal, R. K., Douglas, R., Mosca,
J. D., Moorman, M. A., Simonetti, D. W., Craig, S., and Marshak, D. R.
(1999). Multilineage potential of adult human mesenchymal stem cells.
Science 284 , 143-147.
Rybak, A., Fuchs, H., Smirnova, L., Brandt, C., Pohl, E. E., Nitsch, R., and
Wulczyn, F. G. (2008). A feedback loop comprising lin-28 and let-7 controls pre-let-7 maturation during neural stem-cell commitment. Nat Cell
Biol 10 , 987-993.
Samavarchi-Tehrani, P., Golipour, A., David, L., Sung, H. K., Beyer, T. A., Datti,
A., Woltjen, K., Nagy, A., and Wrana, J. L. (2010). Functional genomics reveals a BMP-driven mesenchymal-to-epithelial transition in the initiation of somatic cell reprogramming. Cell Stem Cell 7 , 64-77.
Shkumatava, A., Stark, A., Sive, H., and Bartel, D. P. (2009). Coherent but overlapping expression of microRNAs and their targets during vertebrate development. Genes Dev 23 , 466-481.
230

Sinkkonen, L., Hugenschmidt, T., Berninger, P., Gaidatzis, D., Mohn, F., Artus-
Revel, C. G., Zavolan, M., Svoboda, P., and Filipowicz, W. (2008).
MicroRNAs control de novo DNA methylation through regulation of transcriptional repressors in mouse embryonic stem cells. Nat Struct Mol
Biol 15 , 259-267.
Stefani, G., and Slack, F. J. (2008). Small non-coding RNAs in animal development. Nat Rev Mol Cell Biol 9 , 219-230.
Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette,
M. A., Paulovich, A., Pomeroy, S. L., Golub, T. R., Lander, E. S., and
Mesirov, J. P. (2005). Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl
Acad Sci U S A 102 , 15545-15550.
Subramanian, S., and Steer, C. J. (2010). MicroRNAs as gatekeepers of apoptosis. J Cell Physiol 223 , 289-298.
Tsang, J., Zhu, J., and van Oudenaarden, A. (2007). MicroRNA-mediated feedback and feedforward loops are recurrent network motifs in mammals.
Mol Cell 26 , 753-767.
Viswanathan, S. R., Daley, G. Q., and Gregory, R. I. (2008). Selective blockade of microRNA processing by Lin28. Science 320 , 97-100.
Wang, Y., Baskerville, S., Shenoy, A., Babiarz, J. E., Baehner, L., and Blelloch,
R. (2008). Embryonic stem cell-specific microRNAs regulate the G1-S transition and promote rapid proliferation. Nat Genet 40 , 1478-1483.
Wang, Y., Medvid, R., Melton, C., Jaenisch, R., and Blelloch, R. (2007). DGCR8 is essential for microRNA biogenesis and silencing of embryonic stem cell self-renewal. Nat Genet 39 , 380-385.
Yisraeli, J. K. (2005). VICKZ proteins: a multi-talented family of regulatory RNAbinding proteins. Biol Cell 97 , 87-96.
231

232

In this thesis, we broaden our understanding of miRNA regulation in relation to various developmental and pathological states. We begin in Chapter 2 with a novel stress-responsive survival phenotype in ES cells regulated by the miR-290-295 cluster, the most abundant miRNA family in ES cells. In conjunction with existing findings that these miRNAs promote cell cycle progression (Wang et al., 2008) as well as results from Chapter 3 on overlapping functions of the placental Sfmbt2 cluster, these data suggest a broader role for miRNAs in maintaining the rapid coordinated growth essential to early mammalian development.
Having identified functional relationships between embryonic miRNAs and properties of early developmental cell types, we turn our attention to transformed mesenchymal cells in Chapters 4 and 5 to extend our knowledge of miRNA regulation across the embryonic-somatic axis. While key features of gene expression are maintained in the absence of Dicer (e.g., core transcription factor expression, cellular morphology), we nonetheless identify a class of tissuespecific genes that show miRNA-dependent expression patterns. In certain cases, these genes have been linked to core properties of their respective states, such as Lats2, an ES miRNA target that inhibits proliferation (Voorhoeve et al.,
2007; Wang et al., 2008), and Igf2bp1, a somatic miRNA target that confers stem-like growth properties (Mayr and Bartel, 2009). However, for other genes, the downstream consequences of repression are less clear, as in the case
233

Zfp827, a poorly characterized zinc finger protein that shows strong upregulation in both Dicer -/ ES and somatic cells relative to wild type ES cells. Given the central roles of those targets whose functions are known, we believe these remaining genes will be fruitful candidates for further study regarding the maintenance of cell state.
Because our studies involve deletion of Dicer, we have been able to study essentially complete loss of the miRNA regulatory layer, allowing us to reach general conclusions about its role in mammalian systems. On the one hand, we find that loss of this layer can be tolerated with relatively little consequence at the cellular level in certain transformed states. Indeed, not only were miRNA-free cells viable in culture, but they also retained the capacity to form tumors in vivo .
On the other hand, we describe compelling evolutionary signatures of selection at both the miRNA and 3 ´ UTR levels, suggesting that miRNA networks may be critical determinants of organismal fitness. These results support observations made by others regarding the occassionally extensive conservation of certain miRNAs and their targets across distant species (Christodoulou et al., 2010;
Friedman et al., 2009).
Taken together, these findings position miRNAs within the general set of regulators that are non-essential for basic properties of in vitro proliferation, but whose function may be important at the organismal scale. One model of how they might play such a role is by fine-tuning expression of target genes whose levels are sensitive determinants of broader biological phenotype (Bartel and
234

Chen, 2004). For instance, it has been recently reported that even a 20% decrease in the expression of tumor suppressor Pten can uncover a broad spectrum of cancers in mice (Alimonti et al., 2010). Within the miRNA field, such a model has been proposed for miR-9a and its target gene senseless , wherein miR-9a levels tailor the number of sensory organ precursors produced during
Drosophila development (Li et al., 2006). Even though the majority of expression changes observed in our ES and somatic Dicer deletions are modest, it is possible that many of them similarly maintain expression levels within an optimal scale for organismal function.
One interesting subset of such interactions that may be particulary difficult to notice in culture are non-cell autonomous effects. The highly expressed muscle-specific miRNA, miR-206, has been reported to participate in such pathways (Chen et al., 2006; Rao et al., 2006). Loss of miR-206 impairs neuromuscular junction formation following injury, acting in part through regulation of FGFBP1, which has known roles in retrograde signaling during reinnervation (Williams et al., 2009). Similar non-cell autonomous phenotypes have been reported for the miR-17-92 cluster, as its expression enhances Mycinduced neovascularization (Dews et al., 2006).
Another potentially distinct class of functions consistent with our findings here involve robustness to stress, which may be critical in tolerating the environmental perturbations that characterize natural systems. Indeed, our data in Chapter 2 describe just such an interaction, wherein gene expression changes caused by miR-295 calibrate a system’s response to stress with relatively minor impact in the unstressed state. A similar model has been described in
235

Drosophila for miR-7, which targets networks found in a number of sensory organs and whose loss of function phenotype is only revealed upon developmental temperature fluctuation (Li et al., 2009).
Given the vast signaling networks in which miRNAs are predicted to participate (Tsang et al., 2010), future studies are likely to elaborate on each of these models, bridging the gap between in vitro observations and their ultimate consequences in vivo .
In addition to conceptual advances, these studies introduce a set of reagents and methods that will facilitate future study of miRNA biology. First, the use of genome-wide approaches for assessing miRNA evolution expand our ability to map miRNA networks in different species. As much of miRNA biology to date has emphasized interactions that are conserved, we hope that this method will enable discoveries in the relatively less explored domain of species-specific regulation. Such results could also be combined with similar analyses of other trans-acting factors such as transcription factors and RNA-binding proteins. In this way, larger coevolving regulatory networks connecting both posttranscriptional and transcriptional elements could be identified, building a more sophisticated model of evolutionary regulatory change.
We also describe the creation of stable Dicer deleted somatic cell lines in sarcoma and mesenchymal stem cell backgrounds. Such models may be advantageous alternatives to acute Dicer deletion in vitro or to in vivo deletions
236

within complex tissue architectures as they allow relatively simple manipulation of homogenously deleted populations. In addition, they enable study of loss of entire miRNA families such as let-7, whose multiple genomic loci complicate derivation of targeted knockouts. Finally, the viability of Dicer null cells in vivo provides a unique opportunity to perform manipulations ex vivo and then evaluate their consequences in an endogenous setting.
Examination of global gene expression changes following Dicer loss in ES cells (Chapter 2) as well as in sarcoma and MSC cell lines (Chapter 5) revealed widespread alterations in the transcriptome among predicted miRNA targets, consistent with previous observations from miRNA gain and loss studies (Baek et al., 2008; Benetti et al., 2008; Lim et al., 2005; Selbach et al., 2008; Sinkkonen et al., 2008). However, some known miRNA targets failed to be re-expressed upon
Dicer deletion. For instance, Oct4 is targeted by several somatic miRNAs including miR-134 and miR-145, but is not readily detected in Dicer -/ somatic cells (Tay et al., 2008; Xu et al., 2009). Although in this case, silencing is reinforced by promoter methylation upon differentiation (Yeo et al., 2007), the full extent to which post-transcriptional downregulation collaborates with other modes of repression in mammals is currently unclear.
Performing global epigenetic mapping of the somatic cell lines described here may be a valuable approach to answering this question. In other
237

organisms, such as fission yeast and plants, RNAi can mediate transcriptional gene silencing (Henderson and Jacobsen, 2007; Matzke and Birchler, 2005).
However, there have been only a handful of examples in mammalian cells reporting similar phenomena (Janowski et al., 2005; Janowski et al., 2006;
Janowski et al., 2007; Kim et al., 2006; Morris et al., 2004). As an obvious mechanistic link between post-transcriptional and transcriptional gene regulation has not been established in mammals, detailed epigenetic profiling of the somatic
Dicer -/ lines may provide both insight into the extent of such regulatory feedback as well as candidate loci for in depth study.
Finally, even in the absence of a direct relationship between miRNA targeting and epigenetic control, there are precedents for miRNA regulation of epigenetic state through indirect means. Global DNA demethylation is observed in Dicer -/ ES cells, mediated in part by loss of miRNA repression of Rbl2, a negative regulator of DNA methyltransferase expression (Sinkkonen et al., 2008)
(Benetti et al., 2008). The genes Bmi1 and Suz12, components of polycomb repressive complexes PRC1 and PRC2, respectively, are known targets of miR-
200 family miRNAs (Iliopoulos et al., 2010; Shimono et al., 2009; Wellner et al.,
2009). Thus, while such data may fail to identify a direct function of short RNAs in transcriptional silencing, they can shed light on further epigenetic control downstream.
In addition to exploring epigenetic relationships with these novel Dicer deleted cell lines, one could perform global screening with a miRNA library to
238

explore specific phenotypes of interest. Such a screen was used to first identify the proliferative function of the miR-295 cluster in Dicer null ES cells (Wang et al., 2008), and could be similarly used to identify the cause of the growth delay in our sarcoma and MSC cells. Attractive candidates from this early screen could be followed up in vivo using competitive tumor formation assays with stable
Dicer-independent miRNAs modeled on the structure of miR-451 (Cheloufi et al.,
2010; Cifuentes et al., 2010). Thus, these models may serve a valuable role in identifying novel oncogenic miRNAs whose targets deserve further study.
Ultimately, the findings presented here suggest a variety of avenues to approach the study of global miRNA regulatory roles. In addition, the tools and reagents developed in the course of this work will facilitate novel approaches to exploring miRNA biology at a range of scales. As research in this field continues to progress, connecting short RNAs to cellular and organismal function, we hope to build a more sophisticated understanding of mammalian physiology, and in doing so, enable an increasingly advanced arsenal to approach human disease.
239

Alimonti, A., Carracedo, A., Clohessy, J. G., Trotman, L. C., Nardella, C., Egia,
A., Salmena, L., Sampieri, K., Haveman, W. J., Brogi, E.
, et al.
(2010).
Subtle variations in Pten dose determine cancer susceptibility. Nat Genet
42 , 454-458.
Baek, D., Villen, J., Shin, C., Camargo, F. D., Gygi, S. P., and Bartel, D. P.
(2008). The impact of microRNAs on protein output. Nature 455 , 64-71.
Bartel, D. P., and Chen, C. Z. (2004). Micromanagers of gene expression: the potentially widespread influence of metazoan microRNAs. Nat Rev Genet
5 , 396-400.
Benetti, R., Gonzalo, S., Jaco, I., Munoz, P., Gonzalez, S., Schoeftner, S.,
Murchison, E., Andl, T., Chen, T., Klatt, P.
, et al.
(2008). A mammalian microRNA cluster controls DNA methylation and telomere recombination via Rbl2-dependent regulation of DNA methyltransferases. Nat Struct Mol
Biol 15 , 268-279.
Cheloufi, S., Dos Santos, C. O., Chong, M. M., and Hannon, G. J. (2010). A dicer-independent miRNA biogenesis pathway that requires Ago catalysis.
Nature 465 , 584-589.
Chen, J. F., Mandel, E. M., Thomson, J. M., Wu, Q., Callis, T. E., Hammond, S.
M., Conlon, F. L., and Wang, D. Z. (2006). The role of microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation. Nat
Genet 38 , 228-233.
Chen, J. F., Murchison, E. P., Tang, R., Callis, T. E., Tatsuguchi, M., Deng, Z.,
Rojas, M., Hammond, S. M., Schneider, M. D., Selzman, C. H.
, et al.
(2008). Targeted deletion of Dicer in the heart leads to dilated cardiomyopathy and heart failure. Proc Natl Acad Sci U S A 105 , 2111-
2116.
Christodoulou, F., Raible, F., Tomer, R., Simakov, O., Trachana, K., Klaus, S.,
Snyman, H., Hannon, G. J., Bork, P., and Arendt, D. (2010). Ancient animal microRNAs and the evolution of tissue identity. Nature 463 , 1084-
1088.
Cifuentes, D., Xue, H., Taylor, D. W., Patnode, H., Mishima, Y., Cheloufi, S., Ma,
E., Mane, S., Hannon, G. J., Lawson, N. D.
, et al.
(2010). A novel miRNA processing pathway independent of Dicer requires Argonaute2 catalytic activity. Science 328 , 1694-1698.
Damiani, D., Alexander, J. J., O'Rourke, J. R., McManus, M., Jadhav, A. P.,
Cepko, C. L., Hauswirth, W. W., Harfe, B. D., and Strettoi, E. (2008). Dicer inactivation leads to progressive functional and structural degeneration of the mouse retina. J Neurosci 28 , 4878-4887.
Dews, M., Homayouni, A., Yu, D., Murphy, D., Sevignani, C., Wentzel, E., Furth,
E. E., Lee, W. M., Enders, G. H., Mendell, J. T., and Thomas-Tikhonenko,
A. (2006). Augmentation of tumor angiogenesis by a Myc-activated microRNA cluster. Nat Genet 38 , 1060-1065.
240

Friedman, R. C., Farh, K. K., Burge, C. B., and Bartel, D. P. (2009). Most mammalian mRNAs are conserved targets of microRNAs. Genome Res
19 , 92-105.
Harvey, S. J., Jarad, G., Cunningham, J., Goldberg, S., Schermer, B., Harfe, B.
D., McManus, M. T., Benzing, T., and Miner, J. H. (2008). Podocytespecific deletion of dicer alters cytoskeletal dynamics and causes glomerular disease. J Am Soc Nephrol 19 , 2150-2158.
Henderson, I. R., and Jacobsen, S. E. (2007). Epigenetic inheritance in plants.
Nature 447 , 418-424.
Iliopoulos, D., Lindahl-Allen, M., Polytarchou, C., Hirsch, H. A., Tsichlis, P. N., and Struhl, K. (2010). Loss of miR-200 inhibition of Suz12 leads to polycomb-mediated repression required for the formation and maintenance of cancer stem cells. Mol Cell 39 , 761-772.
Janowski, B. A., Huffman, K. E., Schwartz, J. C., Ram, R., Hardy, D., Shames, D.
S., Minna, J. D., and Corey, D. R. (2005). Inhibiting gene expression at transcription start sites in chromosomal DNA with antigene RNAs. Nat
Chem Biol 1 , 216-222.
Janowski, B. A., Huffman, K. E., Schwartz, J. C., Ram, R., Nordsell, R., Shames,
D. S., Minna, J. D., and Corey, D. R. (2006). Involvement of AGO1 and
AGO2 in mammalian transcriptional silencing. Nat Struct Mol Biol 13 , 787-
792.
Janowski, B. A., Younger, S. T., Hardy, D. B., Ram, R., Huffman, K. E., and
Corey, D. R. (2007). Activating gene expression in mammalian cells with promoter-targeted duplex RNAs. Nat Chem Biol 3 , 166-173.
Kim, D. H., Villeneuve, L. M., Morris, K. V., and Rossi, J. J. (2006). Argonaute-1 directs siRNA-mediated transcriptional gene silencing in human cells. Nat
Struct Mol Biol 13 , 793-797.
Li, X., Cassidy, J. J., Reinke, C. A., Fischboeck, S., and Carthew, R. W. (2009).
A microRNA imparts robustness against environmental fluctuation during development. Cell 137 , 273-282.
Li, Y., Wang, F., Lee, J. A., and Gao, F. B. (2006). MicroRNA-9a ensures the precise specification of sensory organ precursors in Drosophila. Genes
Dev 20 , 2793-2805.
Lim, L. P., Lau, N. C., Garrett-Engele, P., Grimson, A., Schelter, J. M., Castle, J.,
Bartel, D. P., Linsley, P. S., and Johnson, J. M. (2005). Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature 433 , 769-773.
Matzke, M. A., and Birchler, J. A. (2005). RNAi-mediated pathways in the nucleus. Nat Rev Genet 6 , 24-35.
Mayr, C., and Bartel, D. P. (2009). Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells. Cell 138 , 673-684.
Morris, K. V., Chan, S. W., Jacobsen, S. E., and Looney, D. J. (2004). Small interfering RNA-induced transcriptional gene silencing in human cells.
Science 305 , 1289-1292.
Rao, P. K., Kumar, R. M., Farkhondeh, M., Baskerville, S., and Lodish, H. F.
(2006). Myogenic factors that regulate expression of muscle-specific microRNAs. Proc Natl Acad Sci U S A 103 , 8721-8726.
241

Selbach, M., Schwanhausser, B., Thierfelder, N., Fang, Z., Khanin, R., and
Rajewsky, N. (2008). Widespread changes in protein synthesis induced by microRNAs. Nature 455 , 58-63.
Shimono, Y., Zabala, M., Cho, R. W., Lobo, N., Dalerba, P., Qian, D., Diehn, M.,
Liu, H., Panula, S. P., Chiao, E.
, et al.
(2009). Downregulation of miRNA-
200c links breast cancer stem cells with normal stem cells. Cell 138 , 592-
603.
Sinkkonen, L., Hugenschmidt, T., Berninger, P., Gaidatzis, D., Mohn, F., Artus-
Revel, C. G., Zavolan, M., Svoboda, P., and Filipowicz, W. (2008).
MicroRNAs control de novo DNA methylation through regulation of transcriptional repressors in mouse embryonic stem cells. Nat Struct Mol
Biol 15 , 259-267.
Tay, Y., Zhang, J., Thomson, A. M., Lim, B., and Rigoutsos, I. (2008).
MicroRNAs to Nanog, Oct4 and Sox2 coding regions modulate embryonic stem cell differentiation. Nature 455 , 1124-1128.
Tsang, J. S., Ebert, M. S., and van Oudenaarden, A. (2010). Genome-wide dissection of microRNA functions and cotargeting networks using gene set signatures. Mol Cell 38 , 140-153.
Voorhoeve, P. M., le Sage, C., Schrier, M., Gillis, A. J., Stoop, H., Nagel, R., Liu,
Y. P., van Duijse, J., Drost, J., Griekspoor, A.
, et al.
(2007). A genetic screen implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Adv Exp Med Biol 604 , 17-46.
Wang, Y., Baskerville, S., Shenoy, A., Babiarz, J. E., Baehner, L., and Blelloch,
R. (2008). Embryonic stem cell-specific microRNAs regulate the G1-S transition and promote rapid proliferation. Nat Genet 40 , 1478-1483.
Wellner, U., Schubert, J., Burk, U. C., Schmalhofer, O., Zhu, F., Sonntag, A.,
Waldvogel, B., Vannier, C., Darling, D., zur Hausen, A.
, et al.
(2009). The
EMT-activator ZEB1 promotes tumorigenicity by repressing stemnessinhibiting microRNAs. Nat Cell Biol 11 , 1487-1495.
Williams, A. H., Valdez, G., Moresi, V., Qi, X., McAnally, J., Elliott, J. L., Bassel-
Duby, R., Sanes, J. R., and Olson, E. N. (2009). MicroRNA-206 delays
ALS progression and promotes regeneration of neuromuscular synapses in mice. Science 326 , 1549-1554.
Xu, N., Papagiannakopoulos, T., Pan, G., Thomson, J. A., and Kosik, K. S.
(2009). MicroRNA-145 regulates OCT4, SOX2, and KLF4 and represses pluripotency in human embryonic stem cells. Cell 137 , 647-658.
Yeo, S., Jeong, S., Kim, J., Han, J. S., Han, Y. M., and Kang, Y. K. (2007).
Characterization of DNA methylation change in stem cell marker genes during differentiation of human embryonic stem cells. Biochem Biophys
Res Commun 359 , 536-542.
242

15509 Hitchcock Road
Chesterfield, MO 63017
(636) 328-1594
aravi@mit.edu
EDUCATION
Harvard-MIT Health, Sciences, and Technology Program
M.D. Candidate
Massachusetts Institute of Technology
Ph.D. Candidate in Biology
Stanford University
B.S. Chemistry (honors) and Mathematics
HONORS AND AWARDS
2006-present
2008-present
2002-2006
Fannie and John Hertz Foundation Graduate Fellowship
A. Stone Freedberg Fellowship
Marsden Memorial Prize
Barry M. Goldwater Scholarship
High Honors, US Chemistry Olympiad
Honorable Mention, 2002 Statistics in Chemistry Award, American Statistical Association
RESEARCH EXPERIENCE
Massachusetts Institute of Technology, Graduate Student
Advisor: Institute Professor Phillip A. Sharp
Research: Characterization of microRNA functions in early development and cancer.
2008-present
Harvard-MIT Health, Sciences, Technology Program, Medical Student 2006-present
Advisor: Professor Donald Ingber
Research: Identifying the endothelial signaling pathways that respond to mechanical stretch.
Stanford University, Undergraduate Researcher
Advisor: Professor Chaitan Khosla
Research: Developing novel anti-tubercular drugs via polyketide synthases.
National Institutes of Health, Research Assistant
2004-2006
2004
Advisor: Deputy Lab Chief Ronald Germain
Research: Modeling lymph node traffic using two-photon microscopy data.
Washington University, Research Assistant
Advisor: Professor Larry Taber
Research: Studying the biomechanical forces that govern cardiac development using chick embryos.
2003
243

Washington University, Research Assistant
Advisor: Professor Nathan Ravi
Research: Designing quantitative structure-activity relationships for intraocular lens polymers.
1999-2002
PUBLICATIONS
“The MicroRNA Landscape of the Somatic Mesenchymal State.” Ravi A* , Gurtan A*,
Bhutkar A, Whittaker C, Sharp PA. (in prep)
“Viability of Transformed Somatic Cells in the Absence of Dicer.” Ravi A* , Gurtan A*,
Kumar MS, Chin C, Lees J, Jacks T, Sharp PA. (in prep)
“Genome-Wide Impact of a Novel Rapidly Expanded MicroRNA Cluster in Mouse.”
Zheng GXY*, Ravi A* , Gould GM, Burge CB, Sharp PA. (in review)
“A Latent Pro-survival Function for the Mir-290-295 Cluster in Mouse Embryonic Stem
Cells.” Zheng GXY*, Ravi A* , Calabrese JM, Medeiros LA, Kirak O, Dennis LM,
Jaenisch R, Burge CB, Sharp PA. PLoS Genet. 2011 (in press).
“TRPV4 Channels Mediate Cyclic Strain-Induced Endothelial Cell Reorientation Through
Integrin-to-Integrin Signaling.” Thodeti CK, Matthews B, Ravi A , Mammoto A,
Ghosh K., Bracha AL, Ingber DE. Circ Res . 2009; 104(9): 1123-30.
“Computational Model for Early Cardiac Looping” Ramasubramaniam A, Latacha KS,
Benjamin JM, Voronov DA, Ravi A , Taber LA. Ann Biomed Eng . 2006; 34(8):
1655-69.
“ Quantitative Structure-Activity Relationship of Acrylate and Methacrylate Derivatives”
Ravi A , Spitznagel E, Ravi N. Polymer Preprints . Vol. 45, no. 2, pp. 512-513.
Aug. 2004.
*denotes equal contribution
POSTERS/PRESENTATIONS
“Roles of MicroRNAs in Early Development.” A. Ravi. Tsing Hua University, Beijing,
China, October 2010.
“Viability of Somatic Cells in the Absence of Dicer.” A. Ravi. Oral presentation at Koch
Institute for Integrative Cancer Research Fall Meeting. 2010.
“A MicroRNA-Free Sarcoma Model.” A. Ravi, MS Kumar, C Chin, T Jacks, PA Sharp.
Presented as a poster at 2010 RNA-UNY Structure, Function, and Application
Conference, 2010 MIT Center for Cancer Research Symposium, 2009 MIT
Integrative Cancer Biology Program Retreat, 2009 MIT.
“MicroRNA Function in Mouse ES Cells.” A. Ravi. Oral presentation at MIT Floor
Meeting, 2009.
“Characterization of Biochemical Events Following Acute Dicer Loss in Mouse
Embryonic Stem Cells.” A. Ravi, M. Calabrese, P. Sharp. Presented at Soma
Weiss Poster Session, Harvard Medical School, 2008.
“Precursor-Directed Synthesis of Rifamycin Derivatives” A. Ravi, N. Schnarr, C. Khosla.
Presented at Undergraduate Research Programs Fall Poster Session, 2005.
“Computer Algorithms for Quantitative Tracking of Immune Cell Interactions Imaged in
Situ.” A. Ravi, H. Qi, A. Huang, R.N. Germain. Presented at NIH Summer
Interns Poster Day. 2004.
“Quantitative Structure-Activity Relationship of Acrylate and Methacrylate Derivatives.”
A. Ravi, E. Spitznagel, N. Ravi. International Conference of the Association for
Research in Vision and Ophthalmology. 2001.
244

MISCELLANEOUS
Co-founder and Director, Our Golden Lotus, Inc. (501c3 nonprofit). Charitable organization to provide bridge funds to immigrant survivors of domestic violence.
Staff writer for the Harvard-MIT HST Connector (departmental publication).
Former President and Founder of Students for Healthy Youth, a non-profit student group working with underprivileged children in East Palo Alto in weekly sessions to combat epidemic of childhood obesity. Organized and trained fellow leaders and members, designed curriculum, garnered support from school administrators.
Endurance Events.
-Stanford Triathlon member (competed at Collegiate Nationals and San Jose
International Competitions),
-Raised over $1000 for American Diabetes Association as runner in 2007
Phoenix Marathon and 2005 Atlanta Marathon
-Completed 2003, 2004 St. Louis Half-Marathons
-Youngest finisher, 1997 St. Louis Marathon
245