ACKNOWLEDGEMENTS
In the first place I would like to thank Dr. Paulo Pereira for given me the opportunity of
being a part of his team during the last and for being patient with this stubborn student,
I hope I will rise up to your expectations.
I also have to thank João Ferreira for letting me in the “autophagy team”, for all the
knowledge, ideas, results, help and cigarretes you shared with me. And occasionally
for the beers and such – rock n’ roll!!!!
I also have to thank Carla Bento and Steve, for their infinite patient and will to reply to
so many silly questions and for listening to many strange ideas without laughing (most
of the times at least), and the rest of the Biology of Ageing group at IBILI for receiving
me so well. Thank you Vanda, for all the conversations and critical analysis of my work,
for teaching me and for believing that I will make it in science.
A special thanks to Filipa, my colleague for the last year. Thank you for your support
and for trusting in me despite all the “madness”. You are truly an inspiration.
To Marta, for all the support that you gave me, without you I would not be here.
To Ana, Diana and Patrícia, for all the wonderfull moments in 98 E.
Gostava também de agradecer aos meus pais e avós, por me terem apoiado sempre
nesta aventura em Coimbra. A todos os que não mencionei mas que foram
importantes para mim nestes últimos dois anos.
3
4
TABLE OF CONTENTS
ABBREVIATIONS ........................................................................................................ 8
RESUMO .................................................................................................................... 13
ABSTRACT ................................................................................................................ 15
CHAPTER 1 – INTRODUCTION ................................................................................ 17
1.1. A BRIEF STORY OF THE CONCEPT OF PROTEIN TURNOVER .................................... 19
1.2. W HY CONTINUOUS TURNOVER? .......................................................................... 21
1.3. PATHWAYS FOR ENDOGENOUS PROTEIN DEGRADATION ........................................ 23
1.3.1. Prevalence of different proteolytic pathways.......................................................... 23
1.3.2 There are several routes to intralysosomal protein degradation ....................... 24
1.3.3. Autophagy: cellular self-eating.................................................................................... 25
1.4. MACROAUTOPHAGY ........................................................................................... 27
1.4.1. Overview ............................................................................................................................ 27
1.4.2. Morphology of the macroautophagic process ........................................................ 28
1.4.3. The molecular machinery of macroautophagy ...................................................... 30
1.4.4. Signalling pathways regulating macroautophagy ................................................. 35
1.4.5. Selective macroautophagic processes and the Cvt pathway............................ 38
1.5. MICROAUTOPHAGY ............................................................................................ 41
1.5.1. Overview of the microautophagic process .............................................................. 41
1.5.2. Mechanisms of microautophagy ................................................................................ 42
1.5.3. Regulation of microautophagy .................................................................................... 43
1.5.4. Micropexophagy .............................................................................................................. 44
1.6. CHAPERONE-MEDIATED AUTOPHAGY (CMA) ....................................................... 45
1.6.1. Overview ............................................................................................................................ 45
1.6.2. The KFERQ-like sequences ........................................................................................ 46
1.6.3. Substrate proteins of CMA ........................................................................................... 49
1.6.4. The molecular chaperone complex in the cytosol and associated with the
lysosomal membrane. ............................................................................................................... 50
5
1.6.5. Hsc70 in the lysosomal lumen .................................................................................... 52
1.6.6. LAMP-2A ........................................................................................................................... 54
1.6.7. Regulation of CMA ......................................................................................................... 59
1.6.8. Physiological role of CMA ............................................................................................ 63
1.7. CROSSTALK BETWEEN DIFFERENT PROTEOLYTIC PATHWAYS................................. 66
1.7.1. Overview ............................................................................................................................ 66
1.7.2. Links between UPS and autophagy .......................................................................... 67
1.7.3. Mechanisms of UPS and macroautophagy crosstalk .......................................... 69
1.7.4. Crosstalk between CMA and macroautophagy ..................................................... 71
1.7.5 Coordinated function of CMA, UPS and macroautophagy.................................. 75
1.7.6. Lessons from α-synuclein ............................................................................................ 76
1.7.7. Regulation of the crosstalk between selective degradation pathways ........... 78
1.7.8. Concluding remarks ....................................................................................................... 79
1.8 HIF-1 IS A PROTOTYPICAL SUBSTRATE OF THE PROTEASOME THAT CAN BE
DEGRADED IN THE LYSOSOME .................................................................................... 80
1.8.1. Biology of Hypoxia-Inducible Factor 1 (HIF-1) ....................................................... 80
1.8.2. The HIF family of proteins ............................................................................................ 82
1.8.3. Oxygen-dependent regulation of HIF-1α ................................................................. 83
1.8.4. Oxygen-independent degradation of HIF-1α .......................................................... 86
CHAPTER 2 – MATERIAL & METHODS ................................................................... 87
2.1. CELL CULTURE & MEDIA ..................................................................................... 88
2.2. W ESTERN BLOT ................................................................................................. 90
2.3. MTT CELL VIABILITY ASSAY ................................................................................. 91
2.4. CYCLOHEXAMIDE-CHASE .................................................................................... 91
2.5. IMMUNOPRECIPITATION ...................................................................................... 92
2.6. PLASMIDS AND LENTIVIRAL VECTORS ................................................................... 93
2.7. TRANSDUCTION WITH LENTIVIRAL VECTORS ......................................................... 94
2.8. TRANSIENT TRANSFECTION ................................................................................. 94
2.9. IMMUNOCYTOCHEMISTRY.................................................................................... 95
2.10. STATISTICAL ANALYSIS ..................................................................................... 96
6
CHAPTER 3 – RESULTS ........................................................................................... 97
3.1. HIF-1 IS DEGRADED BOTH BY THE PROTEASOME AND THE LYSOSOME; ................ 98
3.1.1. Inhibition of lysosomal degradation increased HIF-1α protein levels ............. 99
3.1.2. Inhibitors of lysosomal degradation increase HIF-1α protein levels in a
dose-dependent manner......................................................................................................... 101
3.1.3. Stabilization of HIF-1α by chloroquine is independent of VHL-dependent
degradation in RCC4 cells ..................................................................................................... 104
3.1.4. Relative efficiency of HIF-1α degradation by lysosomal and proteasomal
pathways ...................................................................................................................................... 106
3.2. LYSOSOMAL DEGRADATION OF HIF-1 IS NOT MEDIATED BY MACROAUTOPHAGY.. 108
3.2.1. Inhibition of macroautophagy does not stabilize HIF-1α .................................. 108
3.3. HIF-1 UNDERGOES LYSOSOMAL DEGRADATION TROUGH CMA .......................... 111
3.3.1. HIF-1α interacts with LAMP-2 and Hsc70 in ARPE-19 cells ........................... 112
3.3.2. HIF-1α has a KFERQ-like motif (both mouse and human protein) ............... 114
3.3.3. Inhibition of CMA leads to accumulation of HIF-1α ............................................ 115
3.3.4. Mutant HIF-1α protein with an altered KFERQ domain accumulates under
basal conditions in ARPE-19 cells ....................................................................................... 116
CHAPTER 4 – DISCUSSION ................................................................................... 121
4.1. OVERVIEW ....................................................................................................... 122
4.2. LYOSOME INHIBITORS INDUCE HIF-1 STABILIZATION ......................................... 123
4.3. INTRACELLULAR ALKALINIZATION MIGHT IMPACT HIF-1 LEVELS, INDEPENDENTLY OF
LISOSOMAL-MEDIATED DEGRADATION ....................................................................... 124
4.4. CMA IS LIKELY TO BE THE PATHWAY FOR LYSOSOMAL DEGRADATION OF HIF-1.. 126
4.5. HIF-1 ACTIVITY WAS NOT ASSESSED UNDER LYSOSOME INHIBITION ..................... 127
4.6. W HAT IS THE ROLE OF HIF-1 DEGRADATION THROUGH CMA? .......................... 128
4.7. HIF-1 AS A MODEL SUBSTRATE IN THE ELLUCIDATION OF THE MOLECULAR BASIS OF
THE CROSSTALK BETWEEN CMA AND THE UPS ........................................................ 131
4.8. CONCLUDING REMARKS .................................................................................... 133
CHAPTER 5 – REFERENCES ................................................................................. 135
7
8
ABBREVIATIONS
3-MA: 3-methyadenine;
2-OG: 2-oxoglutarate;
ADP: adenosine diphosphate;
AMPK: 5’-AMP-activated protein kinase;
ATG: AuTophaGy-related genes;
ATP: adenosine triphosphate;
Bag-1: BCL2-associated athanogene-1;
bHLH: basic helix-loop-helix;
BOH: β-hydroxybutyrate;
cAMP: cyclic adenosine monophosphate;
CBP: CREB binding protein;
cGMP: cyclic guanosine monophosphate;
CHIP: carboxyl terminus of Hsc70-binding protein;
CHX: cyclohexamide;
CMA: chaperone-mediated autophagy;
COMMD1: copper metabolism MURR1 domain containing 1 protein;
Cvt: cytoplasm to vacuole targeting;
DAPk: death-associated protein kinases;
E1: ubiquitin-activating enzyme;
E2: ubiquitin-carrier enzyme;
E3: ubiquitin-protein ligase;
EGF: epidermal growth factor;
EGFR: epidermal growth factor receptor;
ER: endoplasmic reticulum;
GTP: guanosine triphosphate;
9
HAF: hypoxia-associated factor;
HDAC6: histone deacetylase 6;
HIF-1α: hypoxia-inducible factor-1 transcription factor;
Hip: Hsc70 interacting protein;
Hop: Hsc70-Hsp90 organizer protein;
HRE: hypoxia-response element;
Hsc70: heat-shock cognate protein of 70 kDa;
IκB: inhibitor of the nuclear factor κB;
IP3: inositol-trisphosphate;
IP3R: inositol-trisphosphate (IP3) receptor;
IPAS: inhibitory PAS domain protein;
LAMP-2A: lysosome-associated membrane protein type 2A;
LIR: LC3-interacting region;
Lys-Hsc70: isoform Hsc70 in the lysosomal lumen;
MGO: methylglyoxal;
MHC-II: class II major histocompatibility complex molecules;
MIPA: micropexophagic membrane apparatus;
MTOC: microtubule-organizing centre;
NFκB: nuclear factor κB;
NO: Nitric Oxide;
ODD: oxygen-dependent degradation domain;
PAS: phagophore-assembly site;
PAS family: PER-ARNT-SIM;
PE: phosphatidylethanolamine;
PHD: prolyl hydroxilase domain;
PI3K: phosphoinositidine 3-kinase;
PMN: piecemeal autophagy;
RACK1: activated protein kinase C 1;
10
RCAN1: regulator of calcineurin1;
RNA: ribonucleic acid;
RNase A: ribonuclease A;
TAD: transactivation domain;
TCA: tricarboxilic acid cycle;
TMP: 2,2,4-Trimethylpentane;
TOR: target of rapamycin;
UBA: ubiquitin-associated;
UBL: ubiquitin-like protein;
UPS: ubiquitin-proteasome pathway;
VEGF: vascular endothelial growth factor;
VHL: von Hippel-Lindau tumour suppressor protein;
VTC: vacuolar transport complex;
11
12
Resumo
HIF-1 (Hypoxia-Inducible Factor 1) é um factor de transcrição heterodimérico
responsável pela regulação da resposta celular à hipoxia, além de estar envolvido em
várias patologias, incluindo o cancro e doenças cardiovasculares. A subunidade lábil
HIF-1α é predominantemente degradada por uma via dependente de O2. Em
normoxia, o HIF-1α é hidroxilado e degradado pelo proteossoma, através da ligase de
ubiquitina VHL (Von Hippel Lindau). Nesta tese é descrito um novo mecanismo
molecular para a degradação do HIF-1α, independente da VHL e do proteossoma.
Com base nos resultados obtidos, sugere-se que o HIF-1α é degradado por autofagia
mediada por chaperones (AMC), uma nova via proteolítica selectiva, através da qual
os substratos são endereçados para o lisossoma. Para além de contribuir para a
compreensão dos mecanismos envolvidos na regulação desta proteína, este trabalho
apresenta o HIF-1α como um substrato “dual-pathway”, que pode ser degradado por
AMC ou no proteasoma. Deste modo, o HIF-1α poderá constituir um substrato modelo
para o estudo dos mechanismos moleculares envolvidos na intercomunicação entre
diferentes sistemas proteolíticos.
Palavras-Chave: HIF-1α, Autofagia Mediada por Chaperones (AMC), Via da
Ubiquitina-Proteassoma (VUP), substrato “dual-pathway”;
13
14
Abstract
Hypoxia-Inducible Factor 1 (HIF-1) is a heterodimeric transcription factor that
mediates cellular adaptive response to low O2. HIF-1 also plays a key role in several
pathological conditions, including cancer and ischemic cardiovascular disease. The
labile subunit HIF-1α is primarily regulated via O2-dependent proteolytic degradation.
Under normoxia, HIF-1α is hydroxylated and subsequently targeted for proteasomal
degradation by the ubiquitin ligase VHL (Von Hippel Lindau). The work presented in
this thesis supports the existence of a new molecular mechanism for degradation of
HIF-1α that is independent both on VHL and on the proteasome. The data, obtained
from several cell culture systems, suggests that HIF-1α is degraded by chaperonemediated autophagy (CMA), a new form of selective protein degradation that targets
substrates to the lysosome. In addition to contributing to the clarification of the
mechanisms mediating HIF-1α regulation, this work establishes HIF-1α as dualpathway substrate, capable of undergo proteolysis by CMA or by the proteasome.
Hence, HIF-1a is a potential model substrate in the investigation of the molecular basis
of the crosstalk between proteolytic systems.
Keywords: HIF-1α, Chaperone-Mediated Autophagy (CMA), Ubiquitin-proteasome
System (UPS), dual-pathway substrates;
15
16
CHAPTER 1
Introduction
17
18
1.1. A brief story of the concept of protein turnover
The concept of protein turnover is about 70 years old. It was in the late 1930s
that Rudolf Scheonheimer begun to challenge the concept that the body structural
proteins were essentially stable, subject only to “wear and tear”, while dietary proteins
functioned primarily as energy-providing fuel (Ciechanover, 2005b).
Until then the
paradigm was that protein synthesis was restricted to periods of growth and that urea
was the fate of dietary amino acids (Kirschner, 1999).
In his pioneer studies using
15
N-labelled amino acids, Scheonheimer was able
to show unequivocally that the body structural proteins are continuously synthesized
and degraded, and that even individual amino acids are in a state of dynamic
interconversion
(Ciechanover, 2005a). In the book, “The Dynamic State of Body
Constituents”, published after his death in 1941, his findings are summarized: “The
simile of the combustion engine pictured the steady-state flow of fuel into a fixed
system, and the conversion of this fuel into waste products. The new results imply that
not only the fuel, but the structural materials are in a steady state of flux. The classical
picture must thus be replaced by one which takes account of the dynamic state of body
structure (Ciechanover, 2005a). However, the idea of protein turnover was not widely
accepted, being challenged as late as the mid-1950s (e.g. see (Hogness et al., 1955))
(Ciechanover, 2005a). Nevertheless, other experiments carried out at the same time
demonstrated that cellular proteins are in a dynamic state (e.g. (Mandelstam, 1958;
Simpson, 1953)).
This concept was further strengthened by the discovery of the lysosome by
Christian de Duve in 1955 (De Duve et al., 1955; Gianetto and De Duve, 1955), an
organelle that could perform the degradation of intracellular proteins (Ciechanover,
2005a). Between the mid-1950s and the mid-1970s the lysosome was believed to
degrade intracellular proteins but several lines of experimental evidence indicated that
the degradation of at least certain classes of proteins must be nonlysosomal
19
(Ciechanover, 2005a). This evidence led to a new hypothesis in which most
intracellular proteins, under basal metabolic conditions, were degraded elsewhere in
the cell (Ciechanover, 2005b).
In the late 1970s, a cell-free proteolytic system was described in reticulocytes,
cells which lack lysosomes (Etlinger and Goldberg, 1977). Ensuing work by Avram
Hershko and Aaron Ciechanover on fractionation of crude reticulocytes extracts yielded
two fractions, I and II, both of which were necessary to reconstitute the ATP dependent
proteolytic activity of the crude extract. A small heat-stable protein, component of
fraction I - later showed to be ubiquitin (Wilkinson et al., 1980) - was hypothesized to
be an activator for a protease in fraction II (Ciechanover et al., 1978). Ciechanover,
Hershko and Irwin Rose soon discovered that several moieties of this protein were
found to be covalently conjugated to the target substrate, in the presence of ATP,
tagging it to proteolysis (Ciechanover et al., 1980; Hershko et al., 1980). This process
designated the ubiquitin-protein ligase system, required the sequential activities of
three enzymes: E1, the ubiquitin-activating enzyme, E2, the ubiquitin-carrier protein
and E3, the ubiquitin-protein ligase (Hershko et al., 1983). The E3 was a specific
substrate-binding component, indicating that different proteins might be specifically
recognized and targeted for degradation by different ligases (Ciechanover, 2005a),
although the exquisite specificity of the system was not anticipated at the time
(Kirschner, 1999).
Hence, by the early 1980s a new system for the degradation of intracellular
proteins was firmly established. In 1986 the eukaryotic ATP-dependent protease
involved in this route of protein degradation was identified (Hough et al., 1986) and
later revealed as a ≈ 2,5 MDa enzyme complex called the 26S proteasome (Kirschner,
1999). In the late 1980s and early 1990s, a number of studies addressed the role of
this system in the degradation of key regulatory proteins followed by the investigation
of mechanisms that underlie the recognition and regulation of degradation of specific
proteins (Ciechanover, 2005b). The advances in sequencing the human genome
20
allowed the identification of hundreds of E3s, highlighting the complexity and selectivity
of the system (Ciechanover, 2005b). The discovery of nonproteolytic functions for
ubiquitination, such as regulation of transcription and targeting of membrane protein to
the lysosome, and the discovery of covalent modification by ubiquitin-like proteins
(UBLs), also involved in nonproteolytic functions, led to the emerging role of covalent
protein conjugation in regulation of a broad array of cellular process (Ciechanover,
2005a).
Currently it is accepted that, not only the intracellular proteins are turning over
extensively but also that stability of many proteins is regulated individually and can vary
according to different conditions. Thus, from a seemingly unregulated and nonspecific
end process, proteolysis of cellular proteins has emerged has a highly complex,
temporally controlled and tightly regulated process, that plays major roles in several
important pathways (e.g. in cell cycle progression, regulation of transcription, antigen
presentation, receptor-mediated endocytosis, quality control of proteins, modulation of
metabolic pathways). Consequently, regulated proteolysis now has an equally
important position as transcriptional and translational control in the regulation of cellular
process (Ciechanover, 2005a).
1.2. Why continuous turnover?
With different dynamics, all proteins are continuously synthesized and degraded
(Cuervo, 2004b) but two general themes in protein degradation can be distinguished:
regulation and quality control (Hampton and Garza, 2009).
Regulated degradation controlled by physiological and developmental signals
that determine the selective degradation of a protein, allowing cells to adapt to a
changing environment or to respond to specific intracellular conditions (Hampton and
Garza, 2009). Examples of this process include the degradation of p53 (Fang et al.,
21
2000), the temporally controlled destruction of several proteins during cell cycle
progression (cyclins, transcription factors, spindle components and other cell cycle
regulators) (Pines, 2006) and the degradation of glucose-synthesizing enzymes in the
presence of high levels of glucose (Regelmann et al., 2003). This regulatory role has
been classically ascribed to proteasomal degradation, though the regulatory
importance of lysosomal degradation has begun to emerge (Cuervo, 2004b).
The second theme in the degradation of intracellular proteins is proteolysis to
provide cellular quality control. As the conditions of the intracellular environment favour
denaturation and constantly generate highly reactive molecules, a quality-control
system is needed to monitor mature proteins for postsynthetic denaturation or chemical
damage (Goldberg, 2003). In fact, for many proteins, a random event leading to
denaturation could be the critical event triggering degradation (Goldberg and Dice,
1974). The majority of the degraded proteins however, are still functional before
degradation, establishing protein degradation as a “preventive” mechanism that
ensures the replacement of cellular components before they lose functionality (Cuervo,
2004b). The quality-control degradative system is also responsible for the degradation
of newly synthesized proteins. Noticeably, as many as 30% of newly synthesized
proteins in eukaryotes might undergo degradation within minutes of synthesis. These
proteins are likely products of unsuccessful protein folding or errors in ribosomal
function (Goldberg, 2003).
This rapid and selective degradation of misfolded and damaged proteins has an
important homeostatic function, as it is essential to normal cell function and viability.
Indeed some authors believe that the cell’s degradative machinery must have initially
evolved to serve this quality-control system (Goldberg, 2003).
22
1.3. Pathways for endogenous protein degradation
1.3.1. Prevalence of different proteolytic pathways
Proteolysis is a process that is continuously occurring in both prokaryotic and
eukaryotic cells (with numerous purposes as discussed earlier – see the section “Why
continuous turnover”). Inside the cells proteins are in a dynamic equilibrium and
degradation is potentially as important as synthesis in the control of the protein levels
(Fuertes et al., 2003a). In order to carry out this task cells have several proteolytic
systems but in eukaryotic cells the two main cellular protein degradation systems are
the cytosolic ubiquitin-proteasome system (UPS) and the lysosomal system (Glickman
and Ciechanover, 2002; Goldberg, 2003). In fact, in human fibroblasts proteasomal
and lysosomal degradation equally account for 80% or more of the overall intracellular
protein degradation (Fuertes et al., 2003a). Other proteases operate in the cells, for
instance calpains, which participate in Ca2+ -mediated proteolysis (Croall and Ersfeld,
2007), mitochondrial and other organellar proteases (Koppen and Langer, 2007) and
caspases (Chowdhury et al., 2008). These proteases have more specialised roles and
their contribution to the overall cellular protein turnover is not well defined, although it is
estimated that they contribute to about 15-20% of total protein degradation (Fuertes et
al., 2003a; Fuertes et al., 2003b). However, the cell type and the cellular conditions are
factors that can determine the prevalence of one proteolytic system over the other
(Cuervo, 2004b).
According to the extension of their half-lives, intracellular proteins can be
classified into two groups: short-lived proteins, with half-lives that vary from 10-20
minutes, approximately and long-lived proteins (Hutson and Mortimore, 1982). Shortlived proteins constitute less than 1% of total proteins in hepatocytes, nevertheless
they can contribute to up to one third of the total protein degradation in any given
moment because of their rapid turnover (Mortimore and Poso, 1987). The prevailing
23
view is that degradation of short-lived proteins occurs through the UPS, whereas longlived proteins are mostly subjected to lysosomal degradation (Mizushima and Klionsky,
2007). However, lysosomes can also degrade short-lived proteins (Ahlberg et al., 1985)
and proteasomal degradation constitutes an important pathway for the degradation of
long-lived proteins (Fuertes et al., 2003a).
The following sections will attempt to describe the general mechanisms and
main characteristics of the autophagic pathways of lysosomal degradation, as the UPS
has been extensively reviewed in the literature. Special focus will be on chaperonemediated autophagy (CMA), the selective autophagic pathway of protein degradation.
1.3.2 There are several routes to intralysosomal protein degradation
The discovery of the lysosome as a distinct entity in 1955 by de Duve (De Duve
et al., 1955) helped to support the concept of dynamic turnover of proteins and
provided an initial explanation for controlled protein degradation since the proteases
were separated from their substrates by a membrane (Ciechanover, 2005a).
Initially defined as a vacuolar structure that contains several hydrolytic
enzymes, which function optimally at an acidic pH (Ciechanover, 2005a), lysosomes
are the organelles containing the highest concentration of proteases inside the cell (de
Duve, 1983) and comprise the site of degradation of various proteins, both intracellular
and extracellular (Cuervo and Dice, 1998). The unspecific degradation of long-lived
proteins is usually attributed to the lysosomes. However, lysosomes also degrade
short-lived proteins (e.g. (Ahlberg et al., 1985) and more recently selective processes
of lysosomal degradation have been described (discussed bellow). In lysosomal
degradation, the main regulatory step is the transport of the substrate to the lumen of
the organelle. Intralysosomal protein degradation can occur by different mechanisms,
which differ essentially in the way that the substrate is transported into the lytic
compartment (Cuervo, 2004b).
24
Degradation of both extracellular proteins and plasma membrane proteins
occurs after internalization by endocytosis (or pinocytosis in the case of the former)
(Cuervo and Dice, 1998). The resulting vesicular structures, the endosomes, follow a
discontinuous maturation, resulting in several vacuolar structures that ultimately fuse
with lysosomes. The content of the endosomes is then degraded by acidic hydrolases.
The targeting of the internalized proteins is highly selective and the membrane proteins
may be recycled back to the plasma membrane (Pryor and Luzio, 2009; Steinman et
al., 1983). Regulation of subcellular localization and turnover of membrane proteins is
regulated mainly by ubiquitination. It is well documented that monoubiquitination of
integral membrane proteins mediates the sorting of these proteins to the lysosome,
though Lys63 – linked polyubiquitination is emerging as an important sorting signal
(Miranda et al., 2007). Secretory proteins are also susceptible to intralysosomal
degradation by crinophagy (from the Greek “crin”, meaning “to secrete”), a process that
involves the direct fusion of secretory vesicles with lysosomes (Cuervo and Dice, 1998;
Klionsky et al., 2007a).
1.3.3. Autophagy: cellular self-eating
Lysosomes are also able to participate in the degradation of cytoplasmic
components (soluble proteins and organelles). This process is broadly classified as
autophagy (Rubinsztein et al., 2007), an evolutionary conserved process in eukaryotes
from yeasts to mammals (Yorimitsu and Klionsky, 2005). Christian de Duve coined the
term “autophagy” in 1963, in order to distinguish the lysosomal degradation, or “eating”
(phagy), of part of the cell’s self (auto) from the breakdown of extracellular material
(heterophagy) (Klionsky, 2007). The term intended to illustrate several observations
from electron microscopy studies showing parts of the cytoplasm, including organelles
in various degrees of disintegration, inside single- or double-membrane vesicles
(Ashford and Porter, 1962; Clark, 1957; De Duve and Wattiaux, 1966; Novikoff, 1959).
25
Over the years autophagy was thought to be mainly involved in maintaining the
cellular homeostasis, as it is the process responsible for the continuous turnover of
intracellular organelles and the major mechanism involved in adaptation to poor
nutritional conditions (Bandyopadhyay and Cuervo, 2007). Indeed, in yeast and in
higher eukaryotes nutrient starvation strongly induces autophagy, allowing the
degradation of unneeded proteins and the recycling of amino acids for the synthesis of
proteins that are essential for cell survival (Yorimitsu and Klionsky, 2005). In yeast cells
defective in autophagy, under nitrogen starvation, a reduction in the pool of amino
acids leads to the inability to synthesize proteins that are essential for survival
(Onodera and Ohsumi, 2005).
More recently, an essential role for autophagy was recognized in diverse
cellular functions, including pathogen-to-host defence, cellular differentiation, tissue
remodelling, growth control and mechanisms related to cell death/survival in response
to intra- and extracellular stresses (Maiuri et al., 2007b; Mizushima, 2005; Mizushima
and Klionsky, 2007). Autophagy also seems to act as an anti-aging mechanism
(Bergamini, 2006; Zhang and Kaufman, 2006). Additionally, autophagy may play a
protective role in several protein conformational disorders, including neurodegenerative
disorders, such as Huntington’s, Alzheimer’s and Parkinson’s diseases and several
myopathies. Nevertheless, the role of autophagy in these diseases as well as in cancer
is still poorly understood (Kundu and Thompson, 2008; Todde et al., 2009).
The final step of autophagic processes is the degradation of the substrates in
the lumen of the lysosome. However, depending on the mechanism that delivers the
substrate to the lysosomes, three different types of autophagy have been described in
mammalian
cells:
macroautophagy,
microautophagy
and
chaperone-mediated
autophagy (CMA) (Bandyopadhyay and Cuervo, 2007). The different types of
autophagy differ in their mechanisms and function (Mizushima et al., 2008). The
following sections try to be a brief overview of these pathways and their significance for
the cell functioning.
26
1.4. Macroautophagy
1.4.1. Overview
The term “autophagy” now refers broadly to any cellular degradative process
that involves the delivery of cytoplasmic cargo to the lysosome (Levine and Kroemer,
2008)., At the time of it’s first uses (De Duve and Wattiaux, 1966; Nedelsky et al.,
2008), however, “autophagy” in fact referred to the process of macroautophagy, even
though this term was only introduced later (Klionsky et al., 2007a; Nedelsky et al.,
2008).
Macroautophagy is a process by which cytoplasmic components are
sequestered inside double-membrane vesicles and delivered to the lysosome (or
vacuole in yeast and plants) It constitutes the major regulated catabolic mechanism
that eukaryotic cells use to degrade long-lived proteins and organelles (Geng and
Klionsky, 2008; Levine and Kroemer, 2008).
Unlike
proteasomal
degradation
(and
CMA
as
discussed
bellow),
macroautophagy is not limited by steric considerations, allowing for the sequestration
of entire organelles (Geng and Klionsky, 2008; Mizushima and Klionsky, 2007). In this
process however, there is engulfment of large portions of the cytoplasm along with the
substrate, including “bystanders”. Hence, macroautophagy has been considered a less
selective (or even nonselective) degradative pathway than the UPS (Nedelsky et al.,
2008).
This form of autophagy occurs at a basal level under normal growth conditions,
performing
homeostatic
functions
such
as
protein
and
organelle
turnover.
Macroautophagy can be rapidly upregulated when the cell needs to generate nutrients
and energy, as during starvation (discussed bellow) and when cells undergo structural
alterations, as during developmental transitions. Additionally, macroautophagy is also
upregulated to degrade cytoplasmic components resulting from oxidative stress, as
27
well as after infection by an intracellular pathogen and also in response to protein
aggregate accumulation (Levine and Kroemer, 2008). A detailed description of the
several physiologic roles of macroautophagy, as well as its role in diseases is beyond
the scope of this work. However, these aspects are subject of recent reviews (Kundu
and Thompson, 2008; Levine and Kroemer, 2008; Lum et al., 2005; Maiuri et al.,
2007b; Mizushima, 2005; Mizushima and Klionsky, 2007; Mizushima et al., 2008;
Todde et al., 2009; Yorimitsu and Klionsky, 2005).
1.4.2. Morphology of the macroautophagic process
Macroautophagy has been studied in mammalian cells since the 1950s (Geng
and Klionsky, 2008) and its progression was first characterized in this system. In the
early 1990s, it was demonstrated that the morphology of this process on yeast was
similar to that found in mammals (Klionsky, 2007). As this discovery would lead to
further studies that help uncover the molecular basis for macroautophagy (discussed
below), the main steps of this process for both yeast and higher eukaryotes are
described next.
Conceptually, macroautophagy can be broken down in several steps (see fig.1
for a schematic representation). The first step is designated vesicle nucleation and
comprises the formation of the double membrane structure that engulfs the cargo: the
isolation membrane, (also called the phagophore) (Levine and Kroemer, 2008). The
origin of this membrane is still uncertain, but there is strong evidence that supports its
de novo formation, as it does not seem to originate by a budding process from preexisting membranes (Noda et al., 2002). Nevertheless, several reports suggest that
the membrane origins from the early secretory pathway (Hamasaki et al., 2003;
Reggiori et al., 2004), the mitochondria or both organelles (Luo et al., 2009a).
Subsequently, the phagophore expands, in a process designated vesicle elongation, in
which additional membrane is delivered to the phagosome. In the next step, vesicle
28
completion, the two extremities of the phagophore then fuse, sequestering the
cytoplasmic material in a double-membrane vesicle called the autophagosome (Levine
and Kroemer, 2008).
In the yeast autophagosome, assembly is proposed to occur in a perivacuolar
location, termed pre-autophagosomal structure or phagophore-assembly site (PAS).
The PAS is where the core machinery for macroautophagy (discussed bellow) is
concentrated and contains lipids, organized as micelles or small vesicles (Kim et al.,
2002; Suzuki et al., 2001). In mammals, a PAS was not identified, as the phagophore
is the earliest morphologically detectable autophagy compartment in mammalian cells
(Mizushima et al., 2001). Like in yeast, there is co-localization of different proteins of
the core machinery for macroautophagy co-localize, but in mammalian cells, this has
been observed at multiple foci (Mizushima et al., 2003; Mizushima et al., 2001; Young
et al., 2006). These observations suggest the presence of multiple vesicle formation
sites (Geng and Klionsky, 2008), although the origin of autophagosomes in mammals
is still under debate (Reggiori and Klionsky, 2005).
After vesicle completion, the autophagosome fuses with endosomal vesicles
and/or with lysosomes (Eskelinen, 2005), (or with the vacuole in yeast), creating a
degradative vacuole termed autolysosome (Berg et al., 1998). This fusion process is
similar to that which mediates homotypic and heterotypic fusion events within the
endosomal pathway or Golgi (Kundu and Thompson, 2008). The vesicle resulting of
the fusion of an autophagosome with an endosome is called an amphisome
(Eskelinen, 2005). There is occasional ambiguity in distinguishing autophagosomes,
amphisomes and autolysosomes morphologically, thus the term “autophagic vacuoles”
can appear in the literature referring to all three structures (Nedelsky et al., 2008). As
the outer membrane of the autophagosome fuses with the lysosomal membrane, the
engulfed material, along with the inner membrane, is degraded by lysosomal
hydrolases (Todde et al., 2009). In the final step of macroautophagy, the amino acids
that result from the degradation process are exported to the cytosol, trough lysosomal
29
membrane permeases. This recycling function of autophagy plays a critical role in the
survival of yeast under starvation conditions (Kundu and Thompson, 2008).
Presumably, this recycling function is conserved in higher eukaryotes, but at this time,
there is no direct data proving this concept (Levine and Kroemer, 2008).
Figure 1. Simplified schematic representation of macroautophagic process
morphology. Sequestration begins with the formation of a phagophore that
expands into a double-membrane autophagosome while surrounding a portion of
the cytoplasm. The autophagosome may fuse with an endosome (the product of
endocytosis). The product of endosome–autophagosome fusion is known as an
amphisome. The completed autophagosome or amphisome fuses with a
lysosome, which supplies acid hydrolases. The enzymes in the resulting
compartment, an autolysosome, break down the inner membrane from the
autophagosome and degrade the cargo. The resulting macromolecules are
released through permeases and recycled in the cytosol. Adapted from (Klionsky,
2007).
1.4.3. The molecular machinery of macroautophagy
Although macroautophagy has its roots in the mammalian system, the major
breakthroughs in the unravelling the molecular mechanisms of this process have come
from genetic screens using mutant yeast, starting in the early 1990s (Klionsky, 2007).
30
Since then, there were identified approximately 30 genes specifically required for this
pathway (Xie and Klionsky, 2007). These are designated AuTophaGy-related genes
(ATG) (Klionsky et al., 2003). In this work, ATG gene products will be designated Atg.
Orthologues of these genes have been characterized in mammals, implying a
conserved mechanism (Eskelinen and Saftig, 2009; Geng and Klionsky, 2008). Most of
the proteins required for macroautophagy seem to function at the autophagosome
formation (Yorimitsu and Klionsky, 2005); including proteins involved in two ubiquitinlike conjugation systems.
Vesicle nucleation depends on stepwise recruitment of several Atg proteins and
other components starting with Atg17. This protein recruits Atg13 and Atg9, both of
which are necessary to activate Vps34, a class III phosphoinositidine 3-kinase (PI3K)
and of its associated proteins (Suzuki et al., 2007). Atg9, a transmembrane protein,
might mark the initial nucleating membrane (Young et al., 2006).Vps34 is associated
with Atg6 (whose mammalian homolog is termed Beclin 1) (Kihara et al., 2001a; Tassa
et al., 2003) and the serine/threonine kinase Vps15 (Panaretou et al., 1997). This
complex acts as lipid kinase that mediates nucleation (Levine and Kroemer, 2008) and
is responsible for localize the conjugation systems to the forming unit (Kihara et al.,
2001b). There are other proteins regulatory proteins, such as the UVARAG (Liang et
al., 2006), Ambra 1 (Fimia et al., 2007) and Bif-1 (Takahashi et al., 2007), that
associate with the Beclin 1-Vps34-Vps15 complex. These proteins respectively
enhance autophagy and positively modulate Vps34 activity and autophagosome
formation. Conversely, the proto-oncogenic proteins Bcl-2 and Bcl-XL also can bond
the complex, through Beclin-1, decreasing the PI3K activity and thus inhibiting the
autophagosome formation (Maiuri et al., 2007a; Pattingre et al., 2005).
31
Figure 2. Schematic representation of ubiquitin-like conjugation cascades in
vesicle elongation during macroautophagy. Assembly and elongation of
autophagic membranes are accomplished via sequential action of UPS-like E1E2-E3 cascades. In each case, an E1 enzyme activates a ubiquitin-like protein
(UBL) such as ubiquitin, Atg12, or Atg8. The UBL is then transferred to an E2
conjugating enzyme, followed by an association with an E3 ligase that promotes
association of the UBL and its target. (a) In the UPS, ubiquitination of substrates
is accomplished by an E1-activating enzyme, E2-conjugating enzyme, and an E3ligase. (b) In the first arm of the Atg conjugation pathway, Atg12 associates with
the E1-like Atg7, is transferred to the E2-like Atg10, and is subsequently
conjugated to Atg5. No E3-like protein has been identified in this pathway. (c) In
the second arm of the Atg conjugation pathway, Atg8 associates with the E1-like
Atg7, is transferred to the E2-like Atg3, and is conjugated to PE via the E3-like
action of the Atg12-Atg5 complex. Finally, the protease Atg4 removes Atg8 from
the membrane upon autophagosome completion. Adapted from (Geng and
Klionsky, 2008) and (Nedelsky et al., 2008).
Another Atg protein, Atg1 mediates induction of macroautophagy (Kamada et
al., 2000; Young et al., 2006). The mammalian homolog of Atg1 is ULK1 (Chan et al.,
2007; Young et al., 2006). Atg1 is a serine/threonine kinase that interacts with Atg13
and Atg17, forming complex that responds to upstream signals (Levine and Kroemer,
32
2008) (discussed bellow). The hiperphosphorylation of Ag13 leads to its dissociation
from the complex, attenuating the Atg1 activity (Maiuri et al., 2007b).
Vesicle elongation is mediated by two ubiquitin-like systems, the Atg12 and
Atg8 systems (see fig.2 for a schematic representation) (Levine and Kroemer, 2008).
Although they contain a conserved ubiquitin-fold region, neither protein is a homolog of
ubiquitin, (Sugawara et al., 2004; Suzuki et al., 2005). Also, they are covalently
attached to their substrate in an enzymatic pathway similar to the ubiquitin system
(Geng and Klionsky, 2008). The Atg12 conjugation system is triggered by Atg7, an E1
homolog (Komatsu et al., 2001; Mizushima et al., 1998), which activates Atg12. After
activation, Atg12 is transferred to Atg10, an E2-like enzyme (Shintani et al., 1999), and
eventually conjugated to the target protein Atg5, in a reaction which does not seem to
involve a typical E3 (Geng and Klionsky, 2008). Atg5 further interacts with Atg16,
forming a high molecular weight complex, which probably represents a tetramer of
Atg2-Atg5-Atg16. The Atg2-Atg5-Atg16 complex is essential to macroautophagy
(Kuma et al., 2002). This conjugation system seems conserved in mammals (Geng and
Klionsky, 2008).
In contrast, in the other conjugating system the Atg8 protein is attached to a
lipid, phosphatidylethanolamine (PE) (Geng and Klionsky, 2008). Atg8 is initially
cleaved by the protease Atg4 (Kirisako et al., 2000). Atg7 (the E1 homolog that is
involved in both conjugation cascades) forms a thioester bond with the processed Atg8
resulting in its activation (Ichimura et al., 2000). The activated Atg8 is subsequently
transferred to Atg3, an E2-like protein (Ichimura et al., 2000), that some authors
consider to have homology whit E2 proteins (Tanida et al., 2002). Atg8 is then
conjugated with PE, in a process that is accelerated by the Atg12-Atg5 conjugate
(Hanada et al., 2007), which is also required for the correct localization of Atg8 to the
PAS (Suzuki et al., 2007). There are four metazoan Atg8 homologs, MAP-LC3
(typically abbreviated LC3), GABARAP, GATE-16, and Atg8L, only the first of which is
extensively characterized (Nedelsky et al., 2008). As with the yeast Atg8, PE
33
conjugation with LC3 during autophagy results in a nonsolube form of the protein
(Atg8-PE and LC3-II), that stably associates with the both sides of the growing
membrane (Levine and Kroemer, 2008).
The Atg12-Atg5 conjugate is likely to have an E3-like activity in Atg8/LC3
lipidation (Fujita et al., 2008; Hanada et al., 2007), although it lacks the domains
conserved in other known E3s (Geng and Klionsky, 2008). Atg12-Atg5-Atg16 is mostly
localized in the outer surface of the expanding phagophore, dissociating immediately
before or after autophagosome completion (Mizushima et al., 2003; Mizushima et al.,
2001). This conjugate was purposed to function as a coatomer, involved in vesicle
formation (Ichimura et al., 2000), a hypothesis not supported by recent data (Geng et
al., 2008). On the other hand, Atg8-PE/LC3II is a candidate to act as scaffold, as part
of its population remains inside the autophagosome following completion as there
seems to be a quantitative relationship between the amount of At8 and the vesicle size
(Geng and Klionsky, 2008). Atg8-PE was also reported to mediate membrane tethering
and hemifusion in vitro (Nakatogawa et al., 2007). Atg8 retrieval from the surface of
autophagosome is mediated by the protease Atg4. Atg9 is also recycled to the PAS, in
a process involving Atg1, Atg2 and Atg18 (Reggiori et al., 2004).After autophagosome
completion, it docks and fuses with an endosomal vesicle/lysosome. This fusion
involves identical fusion machinery to that which mediate homotypic and heterotypic
fusion events in the endosomal pathway or Golgi (Kundu and Thompson, 2008). In
mammals, the fusion with the autophagosome requires also other proteins that act
more generally in lysosomal function, such as the lysosomal membrane proteins
LAMP-2 and CLN3 (Levine and Kroemer, 2008).
The degradation of autophagic bodies, along with their cargo, involves several
lysosomal hydrolases, including the proteases cathepsin B, D and L and in yeast the
lipase Atg15 (Levine and Kroemer, 2008) (Kundu and Thompson, 2008). Atg proteins
however, are also involved in the export of amino acids, which result from autophagic
degradation. In the yeast, transport from the vacuolar lumen (which corresponds to the
34
mammalian lysosomal lumen) to the cytoplasm occurs through partially redundant
permeases, including Atg22 and Avt3 and Avt4 (in mammals only orthologs of Avt3
and Avt4 were identified) (Kundu and Thompson, 2008). The recycled amino acids
can be used on protein synthesis or be further processed and, together with the fatty
acids, used by the tricarboxilic acid cycle (TCA) to maintain cellular ATP production
(Levine and Kroemer, 2008).
1.4.4. Signalling pathways regulating macroautophagy
As macroautophagy is involved in various physiological processes (see above),
its tight regulation is of vital importance. Macroautophagy occurs at a basal level in
most or all cells, and it can be induced by a variety of cues (Mizushima and Klionsky,
2007).
The most generalized stimuli for macroautophagy is nutrient deprivation
(Cuervo, 2004b). Other events that stimulate macroautophagy include the decrease of
specific regulator amino acids (leucine in particular) (Meijer and Codogno, 2006) and
increase in levels of glucocorticosteroids and thyroid hormone. Some apoptotic stimuli
also lead to the activation of macroautophagy. Conversely, insulin, cAMP and cGMP
various growth factors are recognized physiological inhibitors (Cuervo, 2004b). Other
cues such as temperature, oxygen concentrations and cell density are important
macroautophagy regulators. However, the effects of some regulators depends on the
tissue analyzed (e.g. glucagon, a general activator, inhibits macroautophagy in cardiac
and skeletal muscle) (Cuervo, 2004b). Nevertheless, many of the molecular
mechanisms and signalling cascades involved in the regulatory process remain unclear
(Kundu and Thompson, 2008). Hence, this work focus mainly in the regulation of
macroautophagy by the nutritional status, for which the mechanisms are better defined.
Macroautophagy is a cell defence mechanism when the supply of nutrients is
limited. For instance, in mice, starvation-induced autophagy is of vital importance
35
immediately after birth, when the maternal supply is suddenly interrupted (Kuma et al.,
2004). Availability of amino acids, in particular, is an important regulator of
macroautophagy. Amino acid starvation induces autophagy, while amino acids
generated by lysosomal degradation act as a feedback inhibitor of autophagosome
formation (Eskelinen and Saftig, 2009). The inhibitory role of amino acids on
macroautophagy was demonstrated in 1977 (Klionsky, 2007).
However, the critical breakthrough in unravelling the regulatory mechanism
occurred only in 1995, when it was observed that rapamycin, an inhibitor of TOR
kinase, acts as an macroautophagy inducer (Klionsky, 2007). TOR (target of
rapamycin) is a conserved serine/threonine protein kinase that is part of two
multiprotein complexes, TORC1 and TORC2, which regulate cell growth and
differentiation and metabolism (Wullschleger et al., 2006). TOR also plays a conserved
role in nutrient sensing. Amino acids upregulate TORC1 signalling, (specially the
presence of leucine) (Wullschleger et al., 2006).
At normal growth conditions, TOR is active and phosphorylates Atg13 (Kamada
et al., 2000). Atg13 association with Atg1, in a complex that also include Atg17, is
required for Atg1 mediated macroautophagy induction (Maiuri et al., 2007b).
Phosphorylation of Atg13 leads to its dissociation from the complex, resulting in a
decrease in Atg1, and macroautophagy inhibition.
Conversely, upon nitrogen
starvation TOR is inactivated, resulting in Atg13 hypophosphorylation and reassociation with Atg1, thus inducing macroautophagy (Kamada et al., 2000; Maiuri et
al., 2007b; Young et al., 2006).
The mechanism(s) by which the amino acid status is communicated to TOR is
not clear (Wullschleger et al., 2006). It was proposed that TOR acts downstream a
putative amino acid receptor, located on the plasma membrane (fig.3), which senses
extracellular amino acids (Kanazawa et al., 2004). The signalling pathway tough, has
not know been elucidated. Recent studies have show that amino acids can induce
TOR signalling through activation of Vps34-Beclin 1, a class III PI3K complex
36
(Nobukuni et al., 2005). However, these findings are difficult to reconcile with the
stimulatory role of the Vps34-Beclin 1 complex in macroautophagy induction
(discussed above) (Pattingre et al., 2008). On the other hand, a meal causes a rise in
insulin levels, a recognized macroautophagy inhibitor (Pfeifer, 1978). Rising insulin
levels lead to the activation of the membrane insulin receptors, which in turn activate
the class I PI3K signalling pathway resulting in activation of TOR (fig.3) in a Akt/protein
kinase dependent manner (Meijer and Codogno, 2006).
Although TOR has been considered the central regulator of macroautophagy,
there is also TOR-independent regulation. For example, the protein kinase Gcn2 is
proposed to act like an intracellular amino acid levels sensor involved in
macroautophagy (Talloczy et al., 2002). Signals from Gcn2 are mediated by the
eIF2alpha kinase signalling pathway, including the eukaryotic initiation factor eIFα,
which supports autophagy when phosphorylated at Ser51 (Talloczy et al., 2002). Also,
a recent study showed that amino acid regulates the binding of Bcl-2 to Beclin 1 (which
inhibits macroautophagy as described above) (fig.3). Starvation leads to c-Jun Nterminal kinase 1 mediated phosphorylation of Bcl-2, which inhibits Bcl-2 association
with Beclin 1, resulting in macroautophagy activation (Eskelinen and Saftig, 2009).
Some of the other proteins involved in macroautophagy regulation include,
among others, 5’-AMP-activated protein kinase (AMPK) which responds to low energy,
double-stranded RNA, p53, death-associated protein kinases (DAPk), the ERmembrane-associated protein Ire-1, GTPases, Erk1/2, ceramide, calcium, the inositoltrisphosphate (IP3) receptor (IP3R). Detailed information about pathways regulating
macroautophagy is available in several recent reviews (Criollo et al., 2007; Eskelinen
and Saftig, 2009; Meijer and Codogno, 2004, 2006; Mizushima and Klionsky, 2007;
Nair and Klionsky, 2005; Rubinsztein et al., 2007).
37
1.4.5. Selective macroautophagic processes and the Cvt pathway
Albeit being generally considered as a “nonselective” degradative pathway,
several findings over the years suggest that macroautophagy can be “selective” in
target protein complexes, organelles and microbes (Kirkin et al., 2009).
A
key
difference
between
“selective
and
“nonselective”
forms
of
macroautophagy seems to be that the signal for the trapping membrane biogenesis
comes, in the “selective” process, from the targeted cellular component, whereas the
Figure 3. Simplified model of autophagy regulation. Macroautophagy occurs
at a basal level and can be induced in response to environmental signals including
nutrients and hormones. The best-characterized regulatory pathway includes a
class I PI3K and TOR, which act to inhibit autophagy. Bcl-2 inhibits
macroautophagy by decreasing the activity of a class III PI3K, which is needed
for activation of autophagy. Dashed lines indicate possible regulatory effects.
Adapted from (Mizushima et al., 2008).
“nonselective” process results from direct activation of the autophagic machinery
(Kundu and Thompson, 2008). This targeted degradation is presumed to require both
components from the basal macroautophagic machinery and signals from the cargo.
However little is known about this mechanism, but the yeast Cvt pathway may provide
some insight (Kundu and Thompson, 2008).
38
The Cvt (cytoplasm to vacuole targeting) pathway is a biosynthetic pathway,
identified only in the yeast species Saccharomyces cerevisiae (Meijer et al., 2007) and
Pichia pastoris (Farre et al., 2007).The Cvt is a route to deliver two hydrolases,
synthesized in the cytoplasm to the vacuole. As the morphology and molecular
mechanisms of this pathway extensively overlap with those of macroautophagy, the
Cvt pathway is frequently described as selective form of autophagy in the literature
(Kirkin et al., 2009; Klionsky, 2007; Kundu and Thompson, 2008). In this pathway the
enzymes, that form large complexes in the cytosol, are recognized by the cargo
receptor protein Atg19 (Scott et al., 2001). Subsequently, Atg19 interacts with Atg11,
an Atg protein involved in the transport of the complex to the PAS (Shintani et al.,
2002). Here, Atg19 interacts with Atg8 and the Cvt vesicle forms (similar to the
autophagosome, as described above), involving the cargo (Noda et al., 2008; Shintani
et al., 2002). The Cvt vesicles ultimately fuse with the vacuole delivering the enzymes.
Although this pathway is not conserved in higher eukaryotes, it has some similarities
with macroautophagic processes considered “selective”.
Since the early 1970s, that several reports suggest that there is specific
sequestration of organelles, such as peroxisomes, mitochondria and portions of the
endoplasmic reticulum (ER) membrane, by macroautophagy, both in yeast and higher
eukaryotes (Klionsky, 2007). This mechanism was proposed to play a role in cellular
remodelling, by allowing the removal of superfluous organelles (Klionsky, 2007) (as
opposed to the “bulk” degradation induced for example by starvation). In the late
1990s, the concept of selective degradation of cargo by macroautophagy was
expanded to include degradation of damaged organelles, after it was demonstrated
that depolarization of mitochondria, during mitochondrial permeability transition, could
lead to selective degradation of this organelle by macroautophagy (Elmore et al., 2001;
Lemasters et al., 1998; Xue et al., 2001). This process of selective mitochondria
removal was subsequently termed mitophagy (Klionsky et al., 2007a). The continuous
39
turnover of mitochondria is well established and takes place by autophagosomal
sequestration of mitochondria along with other cytosolic components, by the
mechanism of macroautophagy described earlier (Stromhaug et al., 1998). Similarly,
Mitophagy appears to rely on the presence of the key effectors of macroautophagy,
although the regulation of this process is not well understood (Kundu and Thompson,
2008). In yeast, on two mitochondrial proteins, Uth1 (Camougrand et al., 2004) and
Auo1 (Tal et al., 2007) are involved in this process. The signalling cascades in which
these proteins participate however, are unknown and Uth1 is not conserved in higher
eukaryotes.
The selective degradation of peroxisomes, termed pexophagy (Bormann and
Sahm, 1978; Veenhuis et al., 1983; Veenhuis et al., 1978), on the other hand, shares
some similarities with the Cvt pathway. Like in the Cvt pathway, in pexophagy Atg11 is
important for selection of perixosomes as cargo. However, in pexophagy an Atg19
protein is not required (Meijer et al., 2007). Still two organelle membrane proteins,
Pex14 and Pex13, appear to be crucial to this process (Bellu et al., 2001; Bellu et al.,
2002).
Additionally, selective autophagic sequestration of ER membranes, termed
reticulophagy, has been described (Klionsky et al., 2007a). Under ER stress
conditions, such as the accumulation of aberrant proteins in the ER lumen, an ER
sensor signals the induction of macroautophagy and triggers the dynamic membrane
events that are required to form an autophagosome. This process leads to the
selective sequestration of portions of the ER (Yorimitsu and Klionsky, 2007). Although
the mechanisms underlying the selective sequestration of particular portions of the ER
membrane remain unclear, the phagophore possibly recognizes membrane fragments
containing misfolded proteins (Yorimitsu and Klionsky, 2007). A signalling pathway
involving the proteins Ire1, Hac1 is proposed to be involved in this type of
macroautophagy in yeast. In mammalian cells, PERK and ATF6 appear to be two other
ER stress sensors, however, the signalling pathway involved in macroautophagy
40
induce by ER stress are not conclusive (Bernales et al., 2006; Yorimitsu and Klionsky,
2007).
More recently, it was described in mammals a role for macroautophagy in
degradation of oligomerized misfolded proteins that are ubiquitinated (Kirkin et al.,
2009). This process involves proteins p62/SQSTM1 and NBR1, which are able to bind
polyubiquitin chains (preferably K63 chains), as well as LC3 (the mammalian homolog
of Atg8). Much like the mechanism of the Cvt pathway, these proteins are proposed to
act as “autophagic receptors”, signalling the ubiquitinated cargo for macroautophagic
degradation (Kirkin et al., 2009). In a related mechanism, HDAC6 binds an
ubiquitinated aggregate of misfolded proteins and it is required for their transport them
along to aggresomes (Kawaguchi et al., 2003; Pandey et al., 2007b). Aggresomes are
higher order protein aggregates, in which toxic misfolded proteins are sequestered,
and usually found at the microtubule-organizing centre (MTOC) (Bjorkoy et al., 2005;
Kopito, 2000). As lysosomes are enriched at the MTOC and mature autophagosomes
are transported along the microtubules for they fusion with the lysosomes (Fass et al.,
2006), aggresomes might allow concentration of all the necessary components for
autophagic clearance of aggregated proteins (Kirkin et al., 2009). The significance of
p62 and HDAC6 in protein degradation will be further discussed in the section
“Mechanisms of UPS and macroautophagy crosstalk”.
1.5. Microautophagy
1.5.1. Overview of the microautophagic process
Microautophagy is another autophagic process by which whole regions of the
cytosol can be delivered to the lysosome. The main difference when compared with
macroautophagy is that, in microautophagy, the components to be degraded are
41
sequestered by the lysosome/vacuole membrane itself, instead of being engulfed by an
autophagosome (Marzella et al., 1981; Todde et al., 2009). Microautophagy is a
conserved mechanism (Huang and Klionsky, 2002). It was initially described in
mammalian cells (Cuervo, 2004b) but has been more extensively studied in yeast
(Cuervo, 2004b; Todde et al., 2009).
This form of autophagy is responsible for the continuous slow degradation of
cytosolic proteins, in normal nutritional conditions (Cuervo and Dice, 1998). As with
macroautophagy, complete organelles are also substrates for microautophagy (Cuervo,
2004b; Dubouloz et al., 2005). Even fractions of the vacuolar membrane are reported
to be degraded by microautophagy in yeast (Mijaljica et al., 2007), possibly as a
mechanism to reduce the organellar size. Such mechanism could play an important
role, by resizing the vacuole after fusion with large autophagosomes during
macroautophagic processes (Mijaljica et al., 2007).
1.5.2. Mechanisms of microautophagy
During microautophagy process, the lysosome/vacuole can adopt very different
shapes, such as invaginations (which can have multiple ramifications) and tubulations
(fig.4), to trap the substrates. The invagination and pinching of these structures
requires an intact membrane potential and participation of GTPases (Cuervo, 2004b).
In this process, small regions of the cytosol are internalized (Cuervo and Dice, 1998).
Conversely, to take up larger structures, the membrane can also form finger like
protrusions, which surround the cellular components to be degraded. In either case,
after homotypic fusion of the involving membranes, the substrates are transported to
the lumen of the lysosome/ vacuole (Todde et al., 2009).
42
1 μm
Figure 4. Electron micrograph of microautophagic invaginations in vivo.
Microautophagic invaginations in the yeast Saccharomyces cerevisiae in vivo,
seen by ultrathin sectioning electron microscopy. Adapted from (Uttenweiler and
Mayer, 2008).
Although the morphologic aspects of morphology are reasonably well
described, only recently the dissection of molecular has begun to see significant
advances (Uttenweiler and Mayer, 2008). An important protein complex for
microautophagy, the Vacuolar Transport Complex (VTC), was recently identified in
yeast (Uttenweiler et al., 2007). The complex is composed of four proteins, Vtc1,2,3
and 4, which localize to the vacuolar membrane (but also to other cellular membranes),
and are proposed to act in autophagic tube organization or vesicle scission. In vitro,
Atg proteins are apparently not required for the microautophagic process described
above (Uttenweiler et al., 2007).
However, the understanding of the regulation of
microautophagy in mammals is still very limited (Bandyopadhyay and Cuervo, 2007).
1.5.3. Regulation of microautophagy
In spite of being poorly understood, microautophagy is generally considered a
type of autophagy that is constitutively active, involved mainly in the constitutive
degradation of soluble long-lived proteins (Bandyopadhyay and Cuervo, 2007;
Mortimore et al., 1988). Accordingly, microautophagy has been considered
43
unresponsive to classic macroautophagic stimuli such as nutrient depravation,
glucagon or regulatory amino acids (Cuervo, 2004b; Mortimore et al., 1988).
More recently, induction of microautophagy was demonstrated in yeast cells
that experience nitrogen starvation (Dubouloz et al., 2005). Like in macroautophagy,
the TOR signalling complex (described above, in the macroautophagy section)
mediates this induction. In addition to TOR, a second regulatory complex, the EGO
complex, which consists of three proteins, Ergo 1, Ergo 3 and the GTPase Gtr2,
controlled the induction of microautophagy. Again, this process is proposed to
counterbalances the massive macroautophagy-mediated membrane influx toward the
vacuolar membrane (Dubouloz et al., 2005).
1.5.4. Micropexophagy
In yeast, a particular form of microautophagy that leads to the preferential
degradation of paroxysms, through direct engulfment by the vacuole, was identified. It
was termed micropexophagy (Mukaiyama et al., 2002; Veenhuis et al., 2000).
However, this process has some particularities that set it apart from the classical
microautophagy.
For fusion of the sequestering membranes the involvement of a second
membrane structure, the micropexophagic membrane apparatus (MIPA) seems to be
important (Sakai et al., 2006). Moreover, unlike general autophagy, various Atg genes
have shown to be require for micropexophagy, most likely for the recognition and
sequestration of the organelles (Sakai et al., 2006), in a process similar to
macropexophagy. Specifically, localization of lipidated At8 (Atg-PE) is on the MIPA
membrane is required for the process, as well as the activities of Atg4 and Atg7.
A few other proteins required for micropexophagy were identified (Todde et al.,
2009), among them is another Atg protein, Atg30, which interacts with the peroxisomal
membrane proteins Pex14 and Pex2. This is important to recognitions of the
44
peroxisome be degraded in both micro- and macroautophagic processes (Farre et al.,
2008).
Additionally, a selective process of mitochondrial degradation, morphologically
similar to microautophagy, was described recently in yeast upon nitrogen starvation
(Kissova et al., 2007). A process by which small portions of the nucleus are degraded,
called piecemeal autophagy (PMN), also resembles microautophagy, although the
general autophagic machinery does not seem to be involved (Roberts et al., 2003).
1.6. Chaperone-Mediated Autophagy (CMA)
1.6.1. Overview
The third type of autophagy described in mammalian cells is chaperonemediated autophagy (CMA). In addition to the UPS, CMA is the major form of selective
protein degradation in the cell (Cuervo and Dice, 1996). Contrasting with the other
forms of autophagy, macroautophagy and microautophagy, which generally involve
non-selective “in bulk” engulfment of complete cytosolic regions, in CMA the substrate
proteins are directly translocated through the lysosomal membrane in to the lumen,
without prior formation of intermediate vesicles (Cuervo, 2004b). In this form of
autophagy the a particular pool of proteins are selectively targeted and translocated
into the lysosomal lumen, one at the time (Cuervo and Dice, 1998). Thus, only soluble
proteins but not organelles are degraded through CMA.
Substrate proteins for CMA contain in their amino acid sequence a lysosomal
targeting motif specific for this pathway, which is biochemically related to the
pentapeptide KFERQ (Cuervo and Dice, 1998). This CMA-targeting motif on the
substrate proteins is recognized by a cytosolic chaperone, the heat-shock cognate
protein of 70 kDa (Hsc70) and its co-chaperones (Kaushik and Cuervo, 2009).
45
Subsequently, this complex is targeted to the lysosome, where the substrate binds to
the lysosome-associated membrane protein type 2A (LAMP-2A) (Cuervo and Dice,
1996). The substrate interacts directly with the cytosolic tail of LAMP-2A, by a region
other than the KFERQ-related motif (which is bound to Hsc70). Additionally,
multimerization of LAMP-2A is necessary to substrate translocation (Cuervo, 2004b).
After undergoing unfolding, the substrate is translocated into the lysosomal lumen, in
an ATP-dependent manner. The translocation of the substrate protein is assisted by a
resident by a resident lysosomal Hsc70 (lys-Hcs70) and is rapidly degraded
(Agarraberes et al., 1997). This mechanism is schematically represented in figure 5. It
would be plausible to think that CMA could also be a biogenic pathway for the delivery
of enzymes into lysosomes, however, none of the known lysosomal hydrolases contain
a KFERQ-related motif (Kaushik and Cuervo, 2009).
CMA is present in most types of mammalian cells, although its activity varies
according to the cell type and with the cellular conditions (Massey et al., 2004). In
almost all cells, CMA a basal level of CMA can be detected, but maximal activation of
this type of autophagy occurs under stress conditions. Prolonged nutrient deprivation
(serum removal in cultured cells or starvation in rodents) (Wing et al., 1991), mild
oxidative stress (Kiffin et al., 2004) and exposure to toxic compounds that induce
protein damage (Bandyopadhyay et al., 2008; Cuervo et al., 1999) are conditions
known to activate CMA.
In the following sections, some of the physiological roles of CMA as well as
critical aspects of its mechanism and regulation will be discussed in more detail.
1.6.2. The KFERQ-like sequences
As mentioned above, most known substrates of CMA contain a peptide
sequence biochemically related to KFERQ (Dice, 2007). This CMA-target motif was
first identified in ribonuclease A (RNase A) (Dice et al., 1986; Majeski and Dice, 2004).
46
0
0
Figure 5. Hypothetical model for the transport of cytosolic proteins to
lysosomes by CMA. In the amino acid sequence of specific cytosolic proteins a
motif (KFERQ) is recognized by cytosolic Hsc70 and its cochaperones. The
substrate/ chaperones complex binds to a Lamp2A and the unfolded substrate,
along with the membrane receptor, are translocated into the lysosomal lumen
assisted by lys-Hsc70. Unknown mechanisms dissociate chaperone and receptor
from the substrate that is rapidly degraded by the lysosomal proteases. The
internalized receptor, probably assisted by the luminal chaperone, is reinserted
back into the lysosomal membrane. Inset shows in a transversal section (top
view) a theoretical model of how molecules of the receptor organize in the
lysosomal membrane to accommodate the multimeric substrate/chaperone
complexes. It is hypothesized that the multi-spanning structure might act as a
transporter. Adapted from (Cuervo, 2004b).
The sequence of the CMA-targeting motif consists of glutamine (Q) preceded or
followed by four amino acids consisting of a basic, lysine (K), arginine (R), an acidic,
aspartatic acid (D) or glutamic acid (E), a bulky hydrophobic, phenylalanine (F),
isoleucine (I), leucine (L) or valine (V), and a repeated basic or bulky hydrophobic
amino acid (Majeski and Dice, 2004). In addition, it has been hypothesized that a small
47
subset of these peptides allows substitution of Q by the related residue asparagine (N)
(Majeski and Dice, 2004). KFERQ-like motifs are present in 30% of cytosolic proteins,
as revealed in immunoprecipation experiments with antibodies raised against KFERQ
(Chiang and Dice, 1988; Wing et al., 1991). In most of these proteins (>80%) the
KFERQ motif appears to be exposed, as they could also be immunoprecipitated
without denaturation (Chiang and Dice, 1988).
Nevertheless, the presence of this motif in a protein sequence does not
necessarily imply that the protein is undergoing degradation through CMA. Instead, it is
an indication that it can be degraded through this pathway, as often the targeting motifs
are only exposed on the surface of the protein after conformational modifications
(Kaushik and Cuervo, 2009). For instance, multimeric proteins can have hidden
KFERQ motifs, as proposed for rat liver aldolase B (Susan and Dunn, 2001). In this
case, the aldolase tetramer may have to be dissociated by ubiquitination in order to the
KFERQ-like sequence to be exposed (Susan and Dunn, 2001). Furthermore, the
existence of three-dimensional KFERQ-like motifs is considered to be possible, in
theory, although only linear motifs have been identified (Cuervo, 2004b).
The determinants that drive the recognition of the KFERQ motif by Hsc70 are
not clarified (Cuervo, 2004b). It is known that some modifications of the substrates,
such as oxidation and denaturation, increase their degradation via CMA (Cuervo et al.,
1999; Cuervo et al., 1998), although the mechanisms by which this increase occurs
remain illusive. A possible explanation is that CMA stimulating factors induce small
conformational changes in the substrate, making the KFERQ-like motif more
accessible to the chaperone. On the other hand, it could be the ability of Hsc70 to
recognize the motif that is modified (Cuervo, 2004b).
It is important to emphasize that targeting motifs for different proteolytic systems
often coexist in the same protein (Massey et al., 2004). Hence, a single protein can be
degraded through different pathways in the same cell, depending on changes in the
protein itself (posttranslational modifications) or in the cellular conditions that result in
48
activation/inhibition of particular proteolytic pathways (Cuervo, 2004b). In conclusion,
the presence of a scoring KFERQ-like sequence in the primary structure of a protein is
not sufficient to identify substrates of CMA. This stresses the importance of proper
experimental determination of a protein as a CMA substrate, according to the multiple
criteria established in the literature (Kaushik and Cuervo, 2009).
1.6.3. Substrate proteins of CMA
The substrates of CMA constitute a heterogeneous pool of proteins, without
functional or structural similarities, apart from for the presence of a KFERK-like motif
(Cuervo, 2004b).
Identified substrates include, among others, RNase A (Backer et al., 1983),
several glycolytic enzymes (e.g. glyceraldehyde-3-phosphate dehydrogenase (Cuervo
et al., 1994), aldolase B (Susan and Dunn, 2001), pyruvate kinase (Majeski and Dice,
2004)), particular transcription factors (e.g. c-Fos (Aniento et al., 1996), HSF1 (Yang
et al., 2008) and the neuronal survival factor MEF2D (Yang et al., 2009)), inhibitors of
transcription factors ( the inhibitor of NFκB, IκB (Cuervo et al., 1998)), some subunits
of the 20S proteasome (Cuervo et al., 1995), some members of the annexin family
(Cuervo et al., 2000),a group of calcium-lipid binding proteins, cytosolic forms of
secretory proteins (α-2-microglobulin (Cuervo et al., 1999)), regulator of calcineurin1
(RCAN1) (Liu et al., 2009b), and α-synuclein (Cuervo et al., 2004).
So far, it was assumed that CMA was only responsible for the degradation of
soluble, cytosolic proteins. However, it was recently demonstrated the degradation of
membrane receptors, EGFR and ErbB2, by CMA (Shen et al., 2009). In addition,
unpublished studies by J. Fred Dice, one of the pioneers of the field of CMA, show that
KFERQ-like motifs occur in 80% of aminoacyl-tRNA synthases, suggesting that theses
enzymes could be substrates for CMA (Dice, 2007). For a more comprehensive list of
CMA substrates please see (Cuervo, 2004b; Kaushik and Cuervo, 2009; Majeski and
49
Dice, 2004). CMA is involved in the degradation of both functional and damaged
proteins.
Taking into account the broad nature of CMA substrates, and their participation
in different cellular processes, it is reasonable to predict that changes in the activity of
this pathway may have major consequences on the cell functioning (Kaushik and
Cuervo, 2009). Strikingly, both the UPS pathway and CMA selectively degrade some of
these substrate proteins, such as IκB, α-synuclein, RCAN1 and EGFR. This apparent
redundancy will be discussed in greater detail in the “Crosstalk between different
proteolytic pathways” section.
1.6.4. The molecular chaperone complex in the cytosol and associated with the
lysosomal membrane.
As described above, the targeting motif for CMA, is recognized in the cytosol by
Hsc70, a constitutive member of the Hsp70 family of chaperones (Chiang et al., 1989).
This molecular chaperone is involved in several cellular processes, such as
dissociation of clathrin and assembly proteins from coated vesicle (Newmyer et al.,
2003), the presentation of antigenic peptides to major histocompatibility complex type II
molecules (Milani et al., 2002) and significantly, the translocation of proteins across
membranes (Agarraberes and Dice, 2001b; Dice, 2007; Lazdunski and Benedetti,
1990). For instance, Hsc70 is involved in the translocation of newly synthesized
proteins across the endoplasmic reticulum and the mitochondrial and chloroplast
membranes (Chen and Schnell, 1999; Majeski and Dice, 2004; Vogel et al., 1990).
The binding of Hsc70 to substrate proteins is regulated by ATP binding and
hydrolysis. The ADP-bound form of this chaperone has the highest affinity for protein
substrates (Agarraberes and Dice, 2001b; Lazdunski and Benedetti, 1990). In CMA,
Hsc70 recognizes a peptide sequence containing the KFERQ-like motif (determined in
vitro to have at least 20 amino acids) (Terlecky et al., 1992) and aids in the delivery of
50
the substrates to the lysosomal membrane(Chiang et al., 1989). The delivery
mechanism however, is still poorly understood. Furthermore, there is evidence that
cytosolic Hsc70, or that docked to the cytosolic face of the lysosomal membrane, helps
to unfold the substrate proteins. This unfolding is required for substrate entry into
lysosomes, as folded proteins cannot be translocated across the membranes
(Salvador et al., 2000).
The interaction between Hsc70 and the substrates is modulated by nucleotides
and by interaction with other cytosolic chaperones and cochaperones. Other proteins,
identified as being part of the complex at the lysosome membrane required for protein
translocation, include Hsp90, hsp40, hip, hop and Bag-1
(Agarraberes and Dice,
2001a). The specific role of each of the cochaperones described in CMA remains
unknown (Kaushik and Cuervo, 2006). By extension of what is known about their
participation
in
other
cellular
processes,
it
is
likely
that
they
modulate
substrate/chaperone interaction and unfolding of the substrate protein at the lysosomal
membrane by regulating the ATP/ADP hydrolysis cycles of Hsc70 (Kaushik and
Cuervo, 2006). Additionally, it was hypothesized that putative resident unfoldases in
the lysosomal membrane could be involved in this process (Agarraberes and Dice,
2001a). The cochaperones themselves could also act as chaperones (Dice, 2007).
Specifically, the heat shock protein of 40 kDa (Hsp40) has been shown to stimulate the
ATPase activity of Hsc70. Consequently, association of Hsp40 with Hsc70 leads to
increased binding and release of substrate proteins by the latter (Suh et al., 1999).
The Hsc70 interacting protein (Hip) stimulates the interaction of Hsc70, Hsc40 and the
substrates (Hohfeld et al., 1995). In contrast, the heat shock protein of 90 kDa (Hsp90)
is involved in the refolding of unfolded proteins and/or can prevent unfolded proteins
from aggregating (Richter and Buchner, 2006). Hsp90 inhibitors are reported to lead to
degradation of Hsc70 substrate proteins (Ibrahim et al., 2005; Isaacs et al., 2002)
Other lines of evidence indicate that chemical inhibitors of Hsp90 stimulate CMA
degradation (Finn et al., 2005; Li, 2006; Shen et al., 2009). This suggests that Hsp90
51
could have an inhibitory role in the molecular chaperone complex involved in CMA
substrates translocation across the lysosomal membrane. Accordingly, it has been
proposed that Hsp90 maybe acting by refolding protein substrates before they can be
translocated into the lysosome for degradation (Finn et al., 2005). The Hsc70-Hsp90
organizer protein (hop) links Hsc70 and Hsp90 and (Demand et al., 1998) and may so
act as a nucleotide exchanger (Agarraberes and Dice, 2001a). The BCL2-associated
athanogene-1 (Bag-1) consists of isoforms that can be positive or negative regulators
of Hsc70 (Luders et al., 2000; Terada and Mori, 2000).
The substrate protein has different sites of interaction for the chaperone
complex and LAMP-2A, allowing the association between the complex and LAMP-2A
at the lysosomal membrane. Hsc70, Hsp40, Hip and Hop are shown to be necessary
for substrate protein translocation (Agarraberes and Dice, 2001a). In their 2004 review
about CMA mechanism, Majesky and Dice predicted that other chaperones and cochaperones could also be part of this complex, particularly proteins that are know to
interact with Hsc70 and Hsp90 in different cellular processes (Majeski and Dice, 2004).
The authors include propose several proteins, including the carboxyl terminus of
Hsc70-binding protein (CHIP). A possible role for CHIP in CMA will be discussed in this
work, in the section “Crosstalk between different proteolytic pathways”.
1.6.5. Hsc70 in the lysosomal lumen
In CMA, the presence of an isoform Hsc70 in the lysosomal lumen (lys-Hsc70)
(see fig.5) is absolutely necessary for the translocation of the substrates proteins
across the lysosomal membrane (Agarraberes et al., 1997). In contrast with its
cytosolic counterpart, the lys-Hsc70 is present in the lysosomal lumen without any of
cochaperones of the complex described earlier (Agarraberes and Dice, 2001a;
Agarraberes et al., 1997). Lys-Hsc70 corresponds to the more acidic of the several
cytosolic isoforms of Hsc70 (Agarraberes et al., 1997).
52
Based on their lys-Hsc70 content, rat liver lysosomes can be separated in two
subpopulations, with slightly different densities, that differ in their ability to carry out
CMA, (Cuervo et al., 1997). The more active population contains abundant lys-Hsc70
while the less active contains little lys-Hsc70. If the levels of Hsc70 are increased in
this latter population of lysosomes, it can become more active for CMA (Cuervo et al.,
1997). Similarly, in response to prolonged starvation, there is an increase in lys-Hsc70
levels. This increase corresponds both to a greater percentage of lysosomes
containing lys-Hsc70 and a greater number of lys-Hsc70 molecules per lysosome
(Majeski and Dice, 2004). Thus, this suggests that lys-Hsc70 can contribute to
activation of CMA (Majeski and Dice, 2004).
The mechanism by which this isoform of Hsc70 is translocated to the lysosomal
lumen is unknown. It has been proposed that all isoforms of Hsc70, and also the other
chaperones
in
the
complex,
enter
the
lysosome
by
macroautophagy
or
microautophagy, but that the only the most acidic isoform of Hsc70 is stable in that
environment (Agarraberes et al., 1997). Instead, the more basic isoforms may be
modified within the lysosome to become more acid. Conversely, as Hsc70 contains two
KFERQ sequences and is a known substrate for CMA, Hsc70 is also proposed to enter
lysosomes by CMA (Cuervo and Dice, 2000a; Cuervo et al., 1997). In this case, the
translocation would require unfolding of Hsc70, so the other chaperones of the
complex likely would not be translocated along Hsc70.
The mechanism by which lys-Hsc70 mediates substrate protein translocation
into the lysosomal lumen is also unknown. It was speculated that lys-Hsc70 was
required to pull proteins into the lysosomal lumen because of analogous roles of
Hsp70s in the import of proteins into the endoplasmic reticulum, mitochondria and
chloroplasts (Agarraberes and Dice, 2001b; Artigues et al., 2002; Brodsky et al., 1995;
McClellan et al., 1998; Terada et al., 1996). However, in both the endoplasmic
reticulum and mitochondria, the organellar Hsp70 requires ATP in order to complete
the repeated cycles of peptide binding and release required to pull proteins across the
53
membrane (Agarraberes and Dice, 2001b; Majeski and Dice, 2004; McClellan et al.,
1998). As mammalian lysosomes are not known to contain ATP, such mechanism
would imply a novel action of Hsc70 on the substrates, at least under acidic conditions
(Majeski and Dice, 2004). Adding to the complexity of the role of lys-Hsc70 in CMA,
other studies suggest that this form of Hsc70 is also important to reinsertion of LAMP2A from the lysosomal lumen into the lysosomal membrane (Cuervo and Dice, 2000b)
(fig.5).
1.6.6. LAMP-2A
The selective degradation of proteins in lysosomes (by the process latter named
CMA) was first described in the early 1980s (Neff et al., 1981). Although mechanistic
features of the pathway resembled the import of precursor proteins in other organelles,
such as mitochondria and endoplasmic reticulum (as discussed earlier), and suggested
a presence of a specific receptor it was not until the mid 1990s that it was identified
(Cuervo and Dice, 1996). Several lines of evidence suggest the presence of a receptor
for CMA in the lysosomal membrane. In isolated lysosomes, substrate proteins for
CMA show saturable binding to the lysosomal membrane and this binding can be
partially inhibited by prior mild protease treatment of the lysosomes (Terlecky and Dice,
1993). In addition, binding of the substrate protein can be competed by addition of
other substrate proteins but not by proteins that are not substrates (Cuervo et al., 2000;
Cuervo et al., 1999; Cuervo et al., 1994).
Screening the lysosomal membrane proteins with known substrates of CMA led
to the identification of a 96 kDa protein (Cuervo and Dice, 1996), subsequently shown
to be the lysosome-associated membrane protein type 2A (LAMP-2A) by sequence
analysis (formerly Lgp96). LAMP-2A was a known lysosomal membrane protein with
previously unknown function. The major domain of LAMP-2A is in the lysosomal matrix
and it is highly glycosylated (the protein has a predicted molecular weight of 45 kDa).
54
Additionally, it has a single transmembrane region and a short 12 amino acid cytosolic
tail (fig.6), corresponding to the sequence GLKRHHTGYEQF (human), which binds the
substrate proteins (Cuervo and Dice, 1996). Particularly, the four basic amino acids
KRHH, are reported to be required for binding to the protein substrates (Cuervo and
Dice, 2000c). The amino acids in the substrate that interact with LAMP-2A are not
defined (Majeski and Dice, 2004).
LAMP-2A is one of the three isoforms that can arise from alternative splicing of
the LAMP-2 gene. The amino acid sequence differences between LAMP-2A, LAMP-2B
and LAMP-2C are restricted to the transmembrane region and the cytosolic tail
(Eskelinen et al., 2005). However, only LAMP-2A acts as receptor for CMA (Cuervo
and Dice, 1996). The functions of LAMP-2B and LAMP-2C are uncertain but they are
expected to be involved in cholesterol trafficking, lysosome biogenesis and
macroautophagy, based on the phenotype of a complete LAMP-2 knockout mouse
(Eskelinen et al., 2004; Huynh et al., 2007; Tanaka et al., 2000).
Binding of the substrate proteins to LAMP-2A is generally rate limiting for CMA.
In fact, levels of LAMP-2A are tightly regulated and directly correlate with CMA activity
(Cuervo and Dice, 2000b, c). Regulation of CMA will be addressed in the following
section of this work.
Lysosomal levels of LAMP-2A are regulated by at least three mechanisms:
dynamic distribution between the lysosomal lumen and the membrane, regulated
cleavage at the membrane in discrete lipid microdomains, and transcriptional regulation
of LAMP-2 gene (Cuervo and Dice, 2000b; Cuervo et al., 2003; Kiffin et al., 2004).
A portion of full-sized LAMP-2A resides within the lysosomal lumen (Cuervo and
Dice, 2000b), perhaps complexed with lipids (Jadot et al., 1997). Upon CMA activation
by prolonged starvation, these molecules reinsert into the membrane, thus reducing
the amount of LAMP-2A in the lumen and increasing its quantity at the lysosomal
membrane (Cuervo and Dice, 2000b; Dice, 2007). Furthermore, it was also suggested
that molecules of LAMP-2A could be internalized in the lumen, associated to the
55
substrate protein (fig.5) and be reinserted into the membrane. In this case, LAMP-2A
would undergo substrate-coupled cycles of insertion/deinsertion from the membrane
(Cuervo, 2004b; Cuervo and Dice, 2000b). The reinsertion process is independent of
the luminal pH but requires an intact membrane potential (Cuervo and Dice, 2000b). In
contrast, it is not known if a membrane potential or lysosomal acidification is required
for transport of the substrate proteins across the lysosomal membrane.
Additionally, the degradation rate of LAMP-2A can also be regulated. LAMP-2A
is normally turned over in the lysosome by a three-step process. The first step is the
cleavage of its cytosolic tail by a still unidentified membrane metalloprotease.
Subsequently, Cathepsin A cleaves the peptide chain at the junction between the
transmembrane and luminal regions of LAMP-2A (Cuervo et al., 2003), releasing a
truncated form of LAMP-2A, that is rapidly degraded by luminal proteases. Upon CMA
Lysosomal membrane
Figure 6. Schematic representation of LAMP-2A. LAMP-2A acts as a
receptor for substrate proteins of CMA. It consists of a large domain within the
lumen, which is highly glycosylated and a single membrane-spanning region,
and a 12 amino acid cytosolic tail, which binds the protein substrate. Adapted
from (Majeski and Dice, 2004).
activation, degradation of LAMP-2A is reported to decrease, leading to stabilization of
LAMP-2A at the lysosomal membrane and thus higher levels of receptor available for
substrate binding/translocation (Cuervo and Dice, 2000b). LAMP-2A is mainly localized
56
in detergent-resistant microdomains, rich in cholesterol and glycosphingolipids, of the
lysosomal membrane when CMA activity is low but move out of these regions when
CAM is activated, such as during prolonged starvation and exposure to oxidants
(Kaushik et al., 2006). Only LAMP-2A outside the cholesterol-rich microdomains is able
to multimerize (which is required for substrate translocation), whereas LAMP-2A
located within these regions is susceptible to proteolytic cleavage and degradation
(Kaushik et al., 2006).
Finally, upon mild oxidative stress the levels of both membrane and luminal
LAMP-2A are increased (Kiffin et al., 2004). The increase of LAMP-2A in these
conditions is attained through a transcriptional up-regulation of this protein, contrasting
with the increase in LAMP-2A resulting from nutritional stress, which does not requires
de novo synthesis (Kiffin et al., 2004).
In a recent study, it was shown that, in the lysosomal membrane, LAMP-2A
organization ranges from monomers to higher-molecular-mass complexes, up to > 800
kDa (Bandyopadhyay et al., 2008). These defined protein multimers dynamically
assemble and disassemble and different steps in CMA depend on the formation of
different LAMP-2A-containing complexes at the lysosomal membrane (Bandyopadhyay
et al., 2008). In the same study, it was also shown that CMA substrates bind
preferentially to LAMP-2A monomers, while the efficient translocation of substrate
proteins across the lysosomal membrane requires the formation of a particular 700-kDa
LAMP-2A-containing complex (Bandyopadhyay et al., 2008).
Moreover, two chaperones related to CMA, Hsc70 and Hsp90, play distinctive
roles in this process: while Hsc70 promotes the organization of LAMP-2A into
monomers or smaller complexes, the interaction of Hsp90 with LAMP-2A (in the
lysosomal lumen) is likely to be required to maintain the stability of the receptor through
different levels of multimerization (fig.7) (Bandyopadhyay et al., 2008). These novel
roles for the chaperones raised a new hypothesis for the increased bending of CMA
substrates to lysosomes, when exogenous Hsc70 is added. According to the new
57
Figure 7. Hypothetical model for the dynamic assembly and disassembly of
LAMP-2A (L-2A) into multimeric complexes at the lysosomal membrane
(L. Mb). CMA substrates are delivered by cytosolic hsc70 bind preferentially to
monomers of LAMP-2A at the lysosomal membrane. The binding of substrates
promotes the multistep organization of LAMP-2A into higher-order multimeric
complexes. The translocation of substrates requires the formation of an
approximately 700-kDa LAMP-2A complex that is disassembled into smaller
complexes in an hsc70-dependent manner when substrates are no longer
present. The interaction of hsp90 with LAMP-2A at the luminal side of the
lysosomal membrane stabilizes this receptor while transitioning between the
multimeric membrane complexes. L. Mtx, lysosomal matrix. Adapted from
(Bandyopadhyay et al., 2008).
findings, this may be not only due to targeting of the substrate (through the interaction
of the chaperone with the KFERQ motif), but also to the Hsc70-mediated disassembly
of LAMP-2A into monomers (the form preferably bound by the substrate)
(Bandyopadhyay et al., 2008).
Importantly, besides LAMP-2A, three other proteins, that remain undindentified,
were found to be part of the translocation complex for CMA (Bandyopadhyay et al.,
2008).
58
1.6.7. Regulation of CMA
CMA has only been identified in mammalian cells so far, but in this group, it
appears to be a generalized form of autophagy, described in many cellular types and
tissues (Cuervo and Dice, 1998; Kaushik and Cuervo, 2009). The described CMA
components - Hsc70 and cochaperones, LAMP-2A and lys-Hsc70- also seem to have
a ubiquitous distribution in mammals, although their levels vary according to the cell
type, thus resulting in differences in CMA activity (Gough and Fambrough, 1997;
Massey et al., 2004).
Although at different levels, CMA activity has been detected in several cell lines
and in primary cells in culture, including human skin fibroblasts (Kaushik and Cuervo,
2009). In rodents, CMA as been detected in liver, kidney, heart, spleen and lung,
mainly based on the changes in the levels of KFERQ-containing proteins in response
to starvation, although CMA can be detected in this organs under even basal
conditions (Chiang and Dice, 1988; Kaushik and Cuervo, 2009; Massey et al., 2004;
Wing et al., 1991). In the liver, an upregulation of CMA is also observed as result of
exposure of the animals to pro-oxidants and toxic compounds (Cuervo et al., 1999;
Kiffin et al., 2004). Conversely, in other tissues, such as testis, brain and skeletal
muscle CMA is not upregulated in response to prolonged fasting (Cuervo and Dice,
1998; Wing et al., 1991). These differences in CMA activation could be interpreted as
related to the different roles of each tissue during nutritional stress (Cuervo and Dice,
1998). Nevertheless, CMA activation in tissues unresponsive to nutritional stress is
possible under other circumstances, as the basic components for this pathway are
present (Massey et al., 2004). In fact, studies recent studies with mouse neurons
support the presence of CMA activity in these cells, despite the unresponsiveness o
changes in the nutritional status in the brain (Martinez-Vicente et al., 2008). Even
though it is known that CMA is activated in response to nutrient deprivation, exposure
mild-oxidative stress and toxins that modify proteins, the mechanisms that regulate
CMA are mostly unknown.
59
Most of the regulation of this pathway takes place at the lysosomal
compartment, were the binding of substrates to the receptor are the rate limiting step of
the pathway (Bandyopadhyay et al., 2008; Cuervo and Dice, 2000b, c; Kaushik et al.,
2006). The activation of CMA during starvation is parallel with an increase in lys-Hsc70
inside the lysosomes, although the mechanisms by which increase occurs are not
known (Agarraberes et al., 1997), as stated above. Hypothetical models for the
regulation of the uptake of lys-Hsc70 will be discussed in the section “Crosstalk
between different proteolytic pathways”. Once the levels of lys-Hsc70 are enough to
guarantee substrate translocation, regulation of CMA commence to depend on the
levels of LAMP-2A at the lysosomal membrane (Cuervo and Dice, 2000b; Massey et
al., 2004).
As discussed in the previous section, the levels of LAMP-2A in the lysosomal
membrane increase with CMA activation. Upon starvation, this increase does not
require de novo synthesis of protein. Under these conditions, the increase of LAMP-2A
in the lysosomal membrane is initially attained first by reducing the normal turnover of
the receptor at the membrane. If the starvation persists, this increase is attained by
reinsertion of the pool of LAMP-2A located at the lysosomal lumen to the lysosomal
membrane (Cuervo and Dice, 2000b; Massey et al., 2004). Additionally, in rodents, if
starvation persist beyond three days, part of the pool with reduced CMA ability acquire
more lys-Hsc70 and become competent for this pathway (Cuervo and Dice, 2000b).
In contrast, acute conditions leading to CMA activation, such as exposure to
pro-oxidant agents, increase the levels of LAMP-2A at the lysosomal membrane (and
lumen) predominantly by de novo synthesis of the receptor (Kiffin et al., 2004; Massey
et al., 2004). It has been proposed that the higher affluence of LAMP-2A molecules to
lysosomes saturates the capability of the lipid in the microdomains, leaving the most of
the receptor outside these regions, where it has been shown to bind the protein
substrate (Kaushik et al., 2006). Part of the enhanced CMA in such conditions could
also result directly from the oxidative modifications of the protein substrates (Kaushik
60
and Cuervo, 2006; Kiffin et al., 2004). It is possible that partial unfolding, typically
associated with oxidative damage could accelerate the translocation across the
lysosomal membrane by eliminating or facilitating the unfolding step. On the other
hand, the partial unfolding could expose hidden KFERQ-like motifs on the substrate,
facilitating their recognition by the cytosolic chaperone complex (Kaushik and Cuervo,
2006) (fig.8).
Figure 8. Activation of CMA as part of the oxidative stress response.
Different mechanisms can contribute to the enhanced degradation of proteins
via CMA during mild oxidative stress. (A) Effect on the substrates: exposure of
hidden CMA-targeting motifs, partial unfolding and generation of CMAtargeting motifs in usually non-substrate proteins, could all contribute to
facilitate substrate delivery and translocation into lysosomes. (B) Effect on the
lysosomal system: mild-oxidative stress results in an increase in the lysosomal
levels of the major components of the CMA-translocation machinery at the
lysosomal membrane. Adapted from (Kaushik and Cuervo, 2006).
It is anticipated that CMA activation leads to the increased removal of proteins
bearing KFERQ-like motifs. However, considering that the CMA-target motif depends
primarily on the biochemical properties of the constituent amino acids, it as been
61
proposed that oxidation could create a targeting motive to this pathway, by modifying
one or more amino acids residues (e.g. by deamination) (Gracy et al., 1998). Yet, this
latter hypothesis is missing experimental support (Kaushik and Cuervo, 2006) (fig.8).
Similarly, exposure to toxic compounds can target specific proteins, leading to
alterations that facilitate their degradation by CMA. It has been reported that exposure
to a gasoline derivative (e.g. 2,2,4-Trimethylpentane – TMP, an octane isomer), leads
to increased α2-migrobulin (a lipid carrier) degradation by CMA in the rat kidney
(Cuervo et al., 1999).
The different demands of the cell upon of the stimuli that activate CMA may
explain some details of the mechanism of activation. During starvation, CMA is
activated to provide essential components for the synthesis of essential proteins.
Under these conditions, the de novo synthesis of LAMP-2A to increase CMA activity
possibly would not be beneficial, due to the shortage of amino acids. Therefore, more
conservative mechanisms are adopted, the down regulation of LAMP-2A degradation
and intralysosomal reinsertion (Cuervo and Dice, 2000b). In contrast, the de novo
synthesis of LAMP-2A upon oxidative damage might be advantageous, because it
provides a faster mechanism of CMA activation (in response to starvation CMA activity
peaks only after several hours). However, this is true only if the damage occurs under
normal nutritional conditions (Kiffin et al., 2004).
Nevertheless, the signalling mechanisms leading to CMA activation/inactivation
remain unknown. During nutrient deprivation, there are increases in the levels of
ketone bodies circulating in the blood with a time course reminiscent of the time course
for activation of CMA (Dice, 2007; Finn and Dice, 2005). Physiological concentrations
of β-hydroxybutyrate (BOH, one of such ketone bodies) increased CMA in human
fibroblasts, but this increase appears to result of oxidation of the substrate proteins by
the BOH (Finn and Dice, 2005). Epidermal growth factor (EGF) has been shown to
reduce CMA activity in kidney epithelial cells (Franch et al., 2001) . Functional Ras and
class 1 PI3-kinases are required for this inhibition; however, secondary messengers
62
have not been identified (Franch et al., 2002). In contrast with this extracellular signal,
activation of CMA during oxidative stress or selective protein damage is likely to result
from an intracellular signal (Massey et al., 2004).
1.6.8. Physiological role of CMA
The first described role of CMA, and the best-studied role, is during the
response to nutritional stress (Massey et al., 2004; Terlecky et al., 1992). In cultured
cells, as well as in rodent tissues, the nutritional sate correlates with activation of CMA,
given that removal of serum (in cells in culture) or prolonged starvation (in rodents)
results in increased rates of CMA (Backer et al., 1983; Neff et al., 1981; Wing et al.,
1991). However, most cells respond initially to starvation by activating macroautophagy
(Massey et al., 2006a). Conversely, in rat liver, the activity of this pathway declines
after 6h after nutrient removal and it becomes practically undetectable beyond 10-12 h
of starvation (Massey et al., 2004; Ueno et al., 1991). If starvation persists, CMA
activity increases progressively up to 88 h (Massey et al., 2004). activated and
becomes the new source of amino acid by selective degradation of proteins that
contain a KFERQ-motif (Cuervo, 2004b).
This pattern is also evident in confluent fibroblasts in cultures after serum
removal (fig.9). In these experiments, after 4h of serum deprivation macroautophagy
becomes the predominant form of protein degradation, but after a prolonged starvation
(24-28 h) it becomes greatly reduced, whereas other forms of lysosomal degradation
are increased (presumably CMA, as microautophagy is considered not to be inducible
by nutrient deprivation) (Fuertes et al., 2003a).
The coordinated shift in protein degradation, from macroautophagy to CMA,
probably obeys the need of a more selective degradation as starvation progresses
(Massey et al., 2004). As the first response to nutritional stress, the large capacity of
macroautophagy could accommodate better the substantial demand for essential
63
components (amino acids, lipids, glycidic groups and nucleotides) required to continue
to synthesis of essential macromolecules for the cell, as well as new proteins in
involved the response to starvation conditions (Massey et al., 2004; Massey et al.,
2006a). However, as the starvation persists, nonselective, “bulk” degradation of
cytosolic components could lead to elimination of essential components, or newly
synthesized components. Under this conditions some selectivity of substrates would be
beneficial (Massey et al., 2004).
Regarding proteins, CMA can provide that selectivity, as some of the proteins
contain the KFERQ motif could be nonessential under those nutritional conditions.
Since glycolysis is reduced during starvation, glycolytic enzymes such as aldolase B,
GAPDH and pyruvate kinase (all shown to be CMA substrates) could be degraded
(Majeski and Dice, 2004; Massey et al., 2004). Moreover, it is also possible that, under
these conditions, the degradation of proteins with CMA-targeting motif could have a
Figure 9. Contribution of various proteolytic pathways to the degradation of
long-lived proteins in human fibroblasts. The contribution of proteasomes,
macroautophagy, other lysosomal pathways (“other lysosomal”) and other nonlysosomal pathways different from proteasomes (“other pathways”), for total
protein degradation, in confluent cultures of human fibroblasts. Conditions:
Confluent (confluent, 4 h incubation in medium containing serum) Confluent/SW
(confluent, 4 h incubation in medium without serum) and prolonged starvation
(confluent, 24-28 h incubation in medium without serum). Adapted from (Fuertes
et al., 2003a).
64
regulatory function. For example, the degradation of some of the subunits of the 20S
proteasome could be responsible for the decrease in proteasomal activity reported
during starvation (Cuervo et al., 1995). In theory, this could prolong the half-live of
many stress-related proteins, which are normally degraded by the proteasome (Massey
et al., 2004). Other regulatory factor that could be subject to a similar type of regulation
is the inhibitor of the nuclear factor κB (IκB), that given the reduction of the
proteasomal activity would be degraded mainly by CMA. Thus, CMA would become
responsible for maintaining NFκB signalling (Massey et al., 2004).
The activation of CMA by mild-oxidative stress and exposure compounds
(discussed above) could also be beneficial, by providing a selective degradation of the
altered proteins, without degrading the functional proteins nearby (as it would be
expected, if these proteins were to be degraded by macroautophagy) (Massey et al.,
2006a). As the role of the UPS in the removal of altered proteins is also been proposed
(Goldberg, 2003; Grune et al., 2003; Imai et al., 2003), it is possible that CMA assists
the former proteolytic pathway in this function, given that the same protein can be
degraded by different proteolytic systems (Massey et al., 2004).
In addition to response to starvation and removal of modified proteins, CMA has
also an important role in aging (Kaushik et al., 2007) and has recently been described
as a key player in antigen presentation via class II major histocompatibility complex
molecules (MHC-II) (Zhou et al., 2005). CMA is also involved in several diseases.
These aspects of the biological role of CMA in mammalian cells will not be addressed
this work. However, detailed information about these subjects is available through the
following references (Bandyopadhyay and Cuervo, 2007; Cuervo, 2004a, 2008;
Kaushik et al., 2007; Massey et al., 2004; Massey et al., 2006c; Massey et al., 2006d;
Zhang and Cuervo, 2008; Zhou et al., 2005).
65
1.7. Crosstalk between different proteolytic pathways
1.7.1. Overview
Over the last 70 years, the concept of protein turnover has covered a long way.
From a seemingly unregulated and nonspecific end process, degradation of
endogenous proteins has emerged has a highly complex, temporally controlled and
tightly regulated process, that plays a key role in the cellular regulation, both under
normal conditions and in times of cellular stress. Much has been learned about the
mechanisms of proteolysis since the pioneer work of Rudolf Scheonheimer, in the
1930s. In nowadays, the major pathways responsible for protein degradation are fairly
well described and important molecular determinants that regulate this process were
identified. However, much remains to be learned about the role of protein degradation
in the regulation of the cellular processes.
In such an example, the prevailing assumption that each protein follows a
particular proteolytic pathway for its degradation is being increasingly challenged.
Instead, accumulating lines of evidence suggest that a protein can be simultaneously
degraded more than one proteolytic systems, or that different pathways alternate in
protein degradation depending on the cellular conditions (Massey et al., 2006c). For
instance, IκB is usually degraded via UPS although it can also be degraded by
calpains, and upon nutrient deprivation, it becomes a substrate for CMA (as discussed
earlier) (Cuervo, 2004b; Cuervo et al., 1998). It is thus reasonable to expect that some
level of communication exist among different protein degradation system to orchestrate
these coordinated activities (Massey et al., 2006a).
The key players and molecular mechanisms involved in this unanticipated
crosstalk however, remain largely unknown. Similarly, the significance of this
coordinated intercommunication between the different proteolytic systems to cellular
regulation is poorly understood. In part, redundant mechanisms of protein degradation
66
are probably aimed to compensate for the failure of one proteolytic system.
Additionally, this redundancy could also serve a regulatory function (Cuervo, 2004b),
allowing the cell to respond adequately upon different situations, which involve different
demands (e.g. the sequential response to starvation and the response to oxidative
damage discussed earlier).
In the following sections, it will be presented several evidences supporting the
concept of crosstalk between the different proteolytic pathways and some of the
proposed mechanisms for this intercommunication will be discussed.
1.7.2. Links between UPS and autophagy
The view of UPS and autophagy as independent, parallel, degradation systems
was first challenged by the observation that monoubiquitination functions as a key
signal for endocytosis. Endossomal sorting often leads to degradation of ubiquitinated
membrane proteins in lysosomes (Pryor and Luzio, 2009). Additionally, the vesicular
structures of the endocytic pathway can interact and fuse with autophagosomes,
forming an amphisome, that subsequently fuses with the lysosome (as previously
discussed) (Eskelinen, 2005).
More recently, it became apparent that these degradation systems share certain
substrates and regulatory molecules and show a compensatory function in some
contexts. Although it is traditionally assumed that the short-lived proteins are normally
degraded via the UPS, it was found that lysosomes also have an important
participation in the degradation of those proteins. In fact, the lysosomal degradation of
short-lived proteins becomes the predominant pathway for degradation of these
proteins in conditions including nutrient deprivation (Fuertes et al., 2003b).
Conversely, degradation of long-lived proteins, usually assumed to be degraded
in the lysosome, was shown to occur also via UPS (Fuertes et al., 2003a).
Furthermore, in the nervous system of mice with conditional knockout of the essential
67
macroautophagy genes Atg5 or Atg7, there was observable accumulation of ubiquitinpositive protein aggregates, resulting in neurodegeneration (Hara et al., 2006; Komatsu
et al., 2006). As there was no detectable defect in UPS function, these results suggest
that some ubiquitinated proteins are in fact normally degraded by autophagy.
Other lines of evidence supporting the relation between the UPS and
macroautophagy come from a series of in vitro studies, in which the dysfunction of the
UPS was induced. When cultured cells are treated with proteasome inhibitors or are
challenged with excess misfolded protein that overwhelms the UPS (the role of UPS in
the removal of altered proteins is well established (Goldberg, 2003; Grune et al., 2003;
Imai et al., 2003)), ubiquitinated misfolded proteins are transported to aggresomes.
These juxtanuclear structures are higher order protein aggregates, in which toxic
misfolded proteins are sequestered, that apparently have a cytoprotective role (Taylor
et al., 2003). The clearance of misfolded proteins from the aggresomes was found to
be mediated by macroautophagy, implicating this pathway as a compensatory
mechanism for degrading misfolded proteins when the proteasomal degradation is
impaired (Iwata et al., 2005a; Iwata et al., 2005b; Taylor et al., 2003; Yamamoto et al.,
2006).
In addition, the inhibition of the UPS in vitro has been found to induce
autophagy (Iwata et al., 2005b; Pan et al., 2008; Rideout et al., 2004). Similar induction
of autophagy is observed in response to the genetic impairment of the proteasome in
Drosophila, and it appears to have a protective role (fig.10).
On the other hand, while the acute inhibition of the UPS described above leads
to activation of macroautophagy, chronic moderate inhibition of this pathway (which
probably resembles more the impairment of the UPS in Alzheimer’s disease and
Parkinson’s disease) results in macroautophagy deregulation (Ding et al., 2003).
Neural cells, subjected to prolonged exposure to low concentrations of proteasome
inhibitors, display both constitutive activation of macroautophagy and the inability to
upregulate this process upon exposure to stressors (Ding et al., 2003).
68
1.7.3. Mechanisms of UPS and macroautophagy crosstalk
In spite of the evidences strongly supporting the intercommunication between
UPS and macroautophagy, the mechanisms involved in the coordination of these
pathways are only beginning to be clarified.
Figure 10. A Drosophila model of proteasome impairment is modified by
manipulation of autophagic activity. (a, b) The temperature-sensitive DTS7 ( a
subunit of the 20S proteasome) mutant shows a normal eye phenotype at the
permissive temperature of 22 °C and a significant degenerative phenotype at the
restrictive temperature of 28 °C. (c) RNAi knockdown of the autophagy gene
atg12 results in an enhancement of the DTS7 degenerative phenotype, suggesting
that the autophagic activity that is induced in response to proteasome impairment
is compensatory. (d) Treatment of DTS7 flies with rapamycin suppresses the
degenerative phenotype, demonstrating that induction of autophagy can
compensate for impaired proteasome function. Adapted from Nedelsky et al.,
2008. Experiments described in Pandey et al., 2007b.
It is reasonable to consider that macroautophagy could be indirectly activated
by the accumulation of altered proteins. In several studies the activation of
macroautophagy results from expression of mutant proteins containing PolyQ
expansion (huntingtin and ataxin-1), that leads accumulation of aggregates inside the
cells, which cannot be degraded by the proteasome (Iwata et al., 2005a; Yamamoto et
69
al., 2006). Similarly, conditions leading to protein aggregation in the endoplasmic
reticulum (ER) have been shown sufficient to upregulate the autophagic degradation of
portions of the membrane of this organelle (Kamimoto et al., 2006).
Direct inhibition of the proteasome is also shown to activate macroautophagy
(Iwata et al., 2005b; Pan et al., 2008; Rideout et al., 2004). Still, this activation could
result of the associated accumulation of undegrated substrate proteins. However,
several regulators have emerged as important players in mediating this crosstalk,
including histone deacetylase 6 (HDAC6) (Iwata et al., 2005b; Pandey et al., 2007a;
Pandey et al., 2007b), p62/sequestosome 1 (p62) (Bjorkoy et al., 2005) and the FYVEdomain containing protein Alfy (Nedelsky et al., 2008) (which will not be discussed in
this work), all of which have been found to regulate or be essential to aggresome
formation (Nedelsky et al., 2008).
HDAC6 is a cytoplasmic microtubule-associated deacetylase that interacts with
polyubiquitinated proteins, possibly providing a link between ubiquitinated cargo and
transport proteins and the transport machinery (Kawaguchi et al., 2003). HDAC6
activity appears to be important for trafficking ubiquitinated proteins and lysosomes in
vitro, leading to the suggestion that this protein coordinates the delivery of substrates
to autophagic machinery (as previously discussed in the section “Selective
macroautophagic processes and the Cvt pathway”) (Iwata et al., 2005b; Kawaguchi et
al., 2003; Kopito, 2003). Importantly, HDAC6 overexpression was demonstrated to
suppress degeneration caused both by impaired UPS activity and by toxic
polyglutamine expression, in a macroautophagy dependent way (Pandey et al.,
2007b). This is consistent with a role for HDAC6 in linking UPS and the compensatory
activation of macroautophagy (Nedelsky et al., 2008; Pandey et al., 2007b).
p62 is a cytosolic protein found to interact with ubiquitinated proteins, through
an ubiquitin-associated (UBA) domain (Geetha and Wooten, 2002; Seibenhener et al.,
2004) and with LC3 through an LC3-interacting region (LIR) (Pankiv et al., 2007).
Moreover, p62 was found to localize to a variety of ubiquitin-positive neuropathological
70
inclusions, including polyQ expanded huntingtin aggregates in Huntington’s disease
and Lewy bodies in Parkinson’s disease (Kuusisto et al., 2002; Nedelsky et al., 2008;
Zatloukal et al., 2002). Several studies have shown that p62 is essential to the
formation of ubiquitin-positive protein inclusions (which have a protective role, by
allowing the autophagic clearance of the altered proteins) (Bjorkoy et al., 2005;
Komatsu et al., 2007; Ramesh Babu et al., 2008). Thus, it has been proposed that p62
acts as “autophagic receptor”, signalling the ubiquitinated cargo for macroautophagic
degradation.
Its has been assumed the signal for autophagic degradation, responsible for
recruiting p62 and HDAC6 is provided by polyubiquitination of the substrates with K63
chains (Olzmann et al., 2007; Tan et al., 2007).
1.7.4. Crosstalk between CMA and macroautophagy
Early evidence of intercommunication between CMA and macroautophagy
came from the temporal relation between both pathways in response to nutrient
starvation (Massey et al., 2006c). As discussed in the section “Physiological role of
CMA”, upon nutrient deprivation the primary response in most tissues is the activation,
however if the starvation persists beyond 4-6 h there is an increase in CMA. Although
this synchronized switch was identified in the early 1990s, the molecular mechanisms
beyond this coordination remain illusive.
It has been proposed that regulators of each of these pathways are natural
substrates for the other (Massey et al., 2006a). Accordingly, it was hypothesized that
endogenous
repressors
of
CMA
(yet
to
be
identified)
are
degraded
by
macroautophagy. Thus, the activation of macroautophagy upon nutrient deprivation
would result in the degradation of the repressor and as a result, in progressive
activation of CMA. In a similar way, CMA could degrade effectors or activators of
macroautophagy. Strikingly, several Atgs contain in their sequence the CMA-targeting
71
motif (Massey et al., 2006c). Upon activation of CMA, these components would be
degraded, limiting their availability to macroautophagy, while inhibition of CMA should
increase the pool of these proteins.
Considering that CMA and macroautophagy share a common terminal
compartment, another nonexclusive possibility is that the activation of each pathway
delivers to the lysosomes components required for the activation of the other (Massey
et al., 2006a). The cytosolic Hsc70 could be a good candidate as crosstalking
mediator, as only lysosomes containing Hsc70 in their lumen are competent for CMA
(as discussed earlier). Macroautophagy is responsible for “in bulk” degradation of
whole regions of the cytosol. Hence, the delivery to lysosomes of Hsc70, along with
many other cytosolic proteins, when macroautophagy is activated would progressively
increase the levels oh the chaperone inside the lysosomes, making them CMA
competent (Massey et al., 2006a). For a schematic representation of these teorical,
models please see figure 11.
Recently, a proteomic analysis of autophagosomes has revealed that Hsc70 is
highly enriched in autophagosomes (Overbye et al., 2007). These results thus,
apparently support the abovementioned hypothesis. Another set of experiments,
however, demonstrated the presence of Hsc70 in lysosomes, in cells with impaired
macroautophagy (Kaushik et al., 2008), which seem to discard the role of
macroautophagy as the main mechanism for delivery of Hsc70 to lysosomes.
This work provides evidence of a more direct crosstalk between CMA and
macroautophagy. It demonstrates that a block of macroautophagy results in
constitutive activation, in basal nutritional conditions and in response to nutrient
deprivation (Kaushik et al., 2008). Although in both conditions the compensatory
activation of CMA involves the increase in the lysosomal levels of LAMP-2A and lysHsc70 and the total amount of lysosomes competent for CMA, this increase seems to
obey different mechanisms.
Specifically, the increase in LAMP-2A under basal
conditions results from de novo synthesis of the receptor protein, whereas during
72
nutrient deprivation, decreased degradation of the LAMP-2A molecules at the
membrane ( a more conservative mechanism) seems the main mechanism for elevated
levels of the receptor at the membrane (Kaushik et al., 2008).
Moreover, other results of this work suggest that autophagosome fusion with
the lysosome dissipates the pH of the latter organelle. The authors propose that the
raise in lysosomal pH when macroautophagy is activated destabilize essential
components for CMA such as lys-Hsc70 (predicted to be more stable under lower pH)
(Kaushik et al., 2008). Thus, the decreased pH upon macroautophagy blockage could
drive CMA upregulation.
Besides nutritional stress, mild-oxidative stress and exposure to toxic
compounds, which activate CMA, also activate macroautophagy. Again, there seems to
be some level of coordination between the two forms of autophagy (Massey et al.,
2006a).
Activation of CMA upon alteration of proteins (discussed in the section
“Regulation of CMA”) may facilitate the selective removal of altered soluble proteins.
However, once misfolded proteins organize into oligomeric, pro-aggregating complexes
they cannot be removed via CMA, as protein unfolding is required for translocation
across the lysosomal membrane. In these conditions, macroautophagy is activated, in
order to clear the aggregates (Iwata et al., 2005b; Ravikumar et al., 2002). As proteins
in these aggregates are often ubiquitinated, macroautophagy activation could depend
on one of the mechanisms discussed above. However, a more indirect mechanism of
macroautophagy activation was also proposed (discussed in the following section).
Other evidence that suggests a crosstalk between CMA and macroautophagy is
that selective inhibition of CMA leads to an upregulation of the later (Massey et al.,
2006b). This work shows that under basal conditions upregulation of macroautophagy
maybe enough to compensate for CMA inhibition. However, upon exposure to
particular stress conditions, specifically conditions that lead to protein oxidation,
activation of macroautophagy fails to compensate the blockage of CMA. On the other
73
hand, this activation of macroautophagy upon CMA blockage can compensate for the
response to heat-shock and nutrient deprivation (Massey et al., 2006b). Therefore,
these results suggest that requirement for one or other proteolytic pathway may be
related to the type of protein damage. Accordingly, it has been proposed that, if protein
damage leads to aggregation, the activation of macroautophagy is more favourable.
However, if protein damage results mostly in protein unfolding and not aggregation, the
selectivity offered by CMA may be more beneficial (Massey et al., 2006b).
Figure 11. Hypothetical mechanisms that can contribute to regulate the
crosstalk between macroautophagy and CMA.
Sequential activation of
macroautophagy and CMA occurs during nutritional stress. The increase in CMA
coinciding with a decrease in macroautophagy could result from the degradation
of an endogenous CMA repressor via macroautophagy (A) and/or the degradation
of macroautophagy effectors by CMA (B). It is also possible that the delivery of
cytosolic regions to lysosomes by macroautophagy augments the lysosomal
content of hsc70, making these organelles competent for CMA activity (C).
Adapted from Massey et al., 2006a.
In conclusion, it seems that compensation of an autophagic pathway by the
other is never complete. Although macroautophagy could degrade CMA substrates, it
lacks the selectivity of the latter pathway, which could be important when specific
proteins have to be degraded (as described above). Conversely, CMA cannot degrade
protein aggregates or organelles
74
(Kaushik et al., 2008). However, despite these
limitations the crosstalk between CMA and macroautophagy seems beneficial for the
cells, under basal conditions.
1.7.5 Coordinated function of CMA, UPS and macroautophagy
CMA does not seem to be limited to intercommunicate with macroautophagy.
Instead, the evidences suggest a coordinated functioning of all the three proteolytic
systems here discussed.
Albeit less understood than the intercommunication between macroautophagy
and the UPS, there are some evidences that suggest a crosstalk between CMA and
the UPS. The selective degradation of certain subunits of the 20S proteasome bearing
a KFERQ-like motif via CMA upon starvation (Cuervo et al., 1995) appears to support
this concept. Importantly, in a recent study it was reported that acute blockage of CMA
leads to a transient impairment in macroautophagy and in the UPS (Massey et al.,
2008). Additionally, the authors report that, although there is subsequent activation of
macroautophagy and restoration of normal proteasomal activity, these compensatory
mechanisms are unable to eliminate and/or prevent the accumulation of intracellular
altered proteins (Massey et al., 2008).
The reported activation of macroautophagy following prolonged CMA
inactivation is consistent with the evidences presented earlier for the crosstalk between
the two autophagic pathways. However, it seems to be a bi-phasic effect of blockage of
CMA in macroautophagy. This effect could be related to alterations on the relative
levels of proteins that interact with TOR, a key negative regulator of macroautophagy
(discussed in the section “Signalling pathways regulating macroautophagy”). TOR
exists at least at two different complexes, each with different associated proteins and
involved in different cellular processes (Wullschleger et al., 2006). The TORC1
complex results from association of TOR with the proteins GβL and Raptor, among
other proteins. This complex is sensitive to rapamycin and its activation inhibits
75
macroautophagy (Wullschleger et al., 2006). The other TOR complex, TORC2 forms
by association of TOR to GβL and Rictor, and its activation stimulates cell proliferation
and survival (Wullschleger et al., 2006). As some results indicate that upon acute CMA
inhibition there is a greater reduction of the levels of Rictor, it was proposed that
formation of TORC1 is promoted, resulting in reduced rates of macroautophagy.
However, as the blockage progresses there is accumulation of aggregated proteins
that would sequester TOR, as described elsewhere (Ravikumar et al., 2004), leading to
decreased levels of active TOR and constitutive activation of macroautophagy (Massey
et al., 2008). This unspecific mechanism contrasts with the more selective mechanisms
of macroautophagy activation in response to protein aggregation, involving specific
signals and receptor proteins described earlier.
Regarding the transient decrease in the proteasomal activity upon acute CMA
inhibition, it was proposed that inability to degrade specific subunits of the proteasome
bearing a KFERQ-like motif when CMA is blocked. Thus, this would lead to a relative
increase in of those particular subunits among the pool of proteasomal constituents,
resulting in qualitative changes in proteasome composition (Massey et al., 2008).
However, results that support this hypothesis have not been published.
1.7.6. Lessons from α-synuclein
A good model to illustrate the compensatory mechanisms among the different
proteolytic systems would be α-synuclein. This neuronal protein is the main component
of the inclusion bodies present in almost all forms of Parkinson’s disease (Cuervo et
al., 2004). α-synuclein can be degraded by the UPS, macroautophagy and CMA
(Cuervo et al., 2004; Webb et al., 2003).
Soluble, unmodified forms of α-synuclein are normally degraded by both UPS
and CMA (Cuervo et al., 2004; Webb et al., 2003) but α-synuclein oligomers and fibres,
toxic structures, are not degraded by these pathways. Oligomers and fibres are shown
76
Figure 12. Hypothetical model for the compensatory mechanisms among
different proteolytic systems elicited at different stages of Parkinson’s
Disease. (Left) Under normal conditions or very early stages of Parkinson’s
disease, soluble forms of α-synuclein are degraded via CMA and/or the UPS.
(Middle) Mutations or toxic post-translational modifications of a-synuclein
interfere with normal functioning of the UPS and of CMA. Blockage of these
systems induces upregulation of macroautophagy, the only system able to remove
large protein aggregates. (Right) As the disease progresses, overloading of the
macroautophagic system along with precipitating factors, such as the agedependent decline in activity of these three proteolytic systems, results in poor
clearance of autophagic vacuoles (AVs), accumulation of protein aggregates and
toxic forms of the protein and eventually cell death. Adapted from Massey et al.,
2006c.
to block proteasome activity (Massey et al., 2006c). Conversely, mutant or dopaminemodified α-synuclein blocks CMA (Cuervo et al., 2004; Martinez-Vicente et al., 2008).
Blockage of both pathways has been shown to activate macroautophagy (Iwata et al.,
2005a; Massey et al., 2006b), which under this conditions would be particularly
beneficial to the cells, possibly by allowing selective clearance of the protein
aggregates, possibly mediated by HDAC6 (Iwata et al., 2005b; Webb et al., 2003).
However, as this compensation mechanism is incomplete the cells would be
more susceptible to stress, thus providing a possible explanation of the aggravating
effect of conditions that lead to oxidative stress in the course of Parkinson’s disease
77
(Massey et al., 2006c). Furthermore, according to this model, the decrease of the
activity of all three proteolytic systems with age would contribute to the overload of the
compensatory mechanism. This age-dependent decline in proteolytic degradation
could explain the aggravated symptoms in older patients (Massey et al., 2006c). This
theorical model is schematically represented in the figure 12.
1.7.7. Regulation of the crosstalk between selective degradation pathways
So far, the known mechanisms of crosstalk are probably aimed to compensate
for the failure of one proteolytic system. Nevertheless, the degradation of specific
proteins is often a complex, temporally controlled and tightly regulated process.
Accordingly, it seems reasonable to expect that additional mechanisms for the
coordination of the communication between the proteolytic pathways might exist
besides these extensive changes in the activity of these systems.
In such an example, activation of CMA during oxidative stress seems to obey
the need for selective removal of damaged proteins without affecting the functional
neighbouring ones. Additionally, degradation of oxidized proteins also can occur
through the UPS (Goldberg, 2003; Grune et al., 2003; Imai et al., 2003). Consequently,
it could be expected that the modified proteins would be more rapidly degraded,
following the same pathways that usually degrade their unmodified counterparts (Kiffin
et al., 2004). However, across this work several evidences that support the concept
that the same protein can be degraded by different proteolytic systems, depending on
the cellular conditions, were discussed. Thus, a model can be envisioned in which the
activation of the UPS upon oxidative stress would provide for the rapid removal of
oxidized proteins, but upon prolong exposure to such conditions oxidized proteins
would instead be delivered to the lysosome via CMA. This switch in the degradation
pathway would allow the cell to devote the UPS to critical regulatory tasks (Kiffin et al.,
2004).
78
Yet another possibility is that both systems would be simultaneously active
during oxidation, and depending on the cell type and particular conditions different
percentages of oxidized proteins would be delivered to one system or the other (Kiffin
et al., 2004). In fact, in different contexts, several protein substrates have been shown
to be degraded by both CMA and UPS, namely α-synuclein (Cuervo et al., 2004; Webb
et al., 2003), IκB (Cuervo, 2004b; Cuervo et al., 1998), the regulator of calcineurin1
(RCAN1) (Liu et al., 2009b) and the membrane receptor EGFR (Shen et al., 2009).
Among these few dual-pathway substrates identified, to distinct behaviors can
distinguished. For example, while RCAN 1 appears to be degraded simultaneously by
both pathways under basal conditions (Liu et al., 2009b), EGFR is usually a substrate
of the UPS and its diverted to lysosomal degradation upon exposure to an Hsp90
inhibitor (Shen et al., 2009).
1.7.8. Concluding remarks
The different lines of evidence discussed across this work apparently support
the existence of a coordinated communication between the different proteolytic
systems. This crosstalk seems mainly involved in eliciting a compensatory response
upon dysfunction of one or more of protein degradation pathways. Taking into account
what is known, it seems that the compensatory mechanisms resulting from this
crosstalk, are able to maintain the cell viability under basal conditions but proper
functioning of all proteolytic systems is required for an effective response to stressors.
Furthermore, this broad intercommunication suggests a model in which the route of
degradation maybe linked to which system is most capable of efficiently degrading it.
However, much is still unknown. Future efforts directed to a better
understanding of the molecular mechanisms regulating the crosstalk among proteolytic
systems, would be beneficial. Failure of one or more protein degradation systems is a
feature of several pathologic conditions including neurodegenerative disorders. Activity
79
of the proteolytic pathways also decreases with aging. A deeper understanding of the
coordination between these systems could enable manipulations to enhance and
maintain compensatory mechanisms or help preventing the initial failure in protein
degradation.
1.8. HIF-1α is a prototypical substrate of the proteasome that can be degraded in
the lysosome
Dual-pathway substrates could be valuable models for exploring the molecular
bases of the crosstalk between proteolytic systems. As mentioned above, a number of
substrates have already been reported to be degraded both by the UPS and by CMA.
Predictably, other substrates other substrates that might undergo proteolysis through
different systems will be described. Indeed, in a recent work by Olmos et al. it is
reported that the –α of the Hypoxia-Inducible Factor 1 transcription factor (HIF-1α) can
be degraded in the lysosome (Olmos et al., 2009). However, the lysosomal pathway for
HIF-1α was not elucidated. The work described in this thesis will address this question.
The following sections seek to provide a general overview of HIF-1α biology,
emphasizing on oxygen dependent degradation of this protein.
1.8.1. Biology of Hypoxia-Inducible Factor 1 (HIF-1)
Hypoxia-Inducible factor 1 (HIF-1) is a transcription factor that mediates the
cellular response to reduced O2 levels, regulating, directly or indirectly, over 100 genes
crucial for adaptation to hypoxic conditions (<5% O2) (Weidemann and Johnson, 2008).
HIF activity is a critical factor in development and physiology of metazoans, impacting a
broad range of cellular processes, such as metabolic adaptation, erythropoiesis,
angiogenesis and vascular tone, cell growth and differentiation, survival and apoptosis
80
(Fig. 13) (for reviews see Maxwell et al., 2001; Weidemann and Johnson, 2008). HIF-1
Figure 13. Target genes of HIF-1. ADM, adrenomedullin; ALDA, aldolase
A; ALDC, aldolase C; AMF, autocrine motility factor; CATHD, cathepsin
D; EG-VEGF, endocrinegland-derived VEGF; ENG, endoglin; ET1,
endothelin-1; ENO1, enolase 1; EPO, erythropoietin; FN1, fibronectin 1;
GLUT1, glucose transporter 1; GLUT3, glucose transporter 3; GAPDH,
glyceraldehyde-3-P-dehydrogenase; HK1, hexokinase 1; HK2, hexokinase
2; IGF2, insulin-like growth-factor 2; IGF-BP1, IGF-factor-binding-protein
1; IGF-BP2, IGF-factor-binding-protein 2; IGF-BP3, IGF-factor-bindingprotein 3; KRT14, keratin 14; KRT18, keratin 18; KRT19, keratin 19;
LDHA, lactate dehydrogenase A; LEP, leptin; LRP1, LDL-receptor-related
protein 1; MDR1, multidrug resistance 1; MMP2, matrix metalloproteinase
2;
NOS2,
nitric
oxide
synthase
2;
PFKBF3,
6-phosphofructo-2-
kinase/fructose-2,6-biphosphatase-3; PFKL, phosphofructokinase L; PGK1,
phosphoglycerate kinase 1; PAI1, plasminogen-activator inhibitor 1; PKM,
pyruvate kinase M; TGF-α, transforming growth factor-α; TGF-β3,
transforming growth factor-β3; VEGF, vascular endothelial growth factor;
UPAR, urokinase plasminogen activator receptor; VEGFR2, VEGF
receptor-2; VIM, vimentin. Adapted from Semenza, 2003.
81
also plays a key role in several pathological conditions, including cancer, ischemic
cardiovascular diseases and neovascularization of the retina (Arjamaa and Nikinmaa,
2006; Melillo, 2004; Rankin and Giaccia, 2008), making HIF-1 regulation a major issue
in cell biology.
Under hypoxic conditions HIF-1 interacts with co-activators and binds to
hypoxia-response elements (HREs) in the regulatory regions of target genes. Several
target genes of HIF-1, such as vascular endothelial growth factor (VEGF), are activated
in response to hypoxia in most cell types. However, for the majority of HIF-1 target
genes, expression induced by hypoxia is cell-type dependent. This discrepancy results
from functional interactions of HIF-1 with other transcription factors, which determines
the subgroup of HIF-1 target genes that is activated in any particular hypoxic cell
(Semenza, 2003).
1.8.2. The HIF family of proteins
HIF is a heterodimeric transcription factor, member of the PAS family (PERARNT-SIM) of basic helix-loop-helix (bHLH) transcription factors, composed of a
constitutive HIF-1β subunit and inducible HIF-α subunit. Three principal HIF-α isoforms
have been described thus far, HIF-1α, HIF-2α and HIF-3α, all encoded by different
gene loci (reviewed in Weidemann and Johnson, 2008). The HIF-α subunits have
related architecture, with a bHLH domain for DNA binding and dimerization, a PAS
domain for dimerization and target gene specificity, an oxygen-dependent degradation
domain (ODD), involved in degradation by the UPS, and one or two transactivation
domains (TADs) (Fig.14) (see Rankin and Giaccia, 2008 for references). HIF-1α is
ubiquitously expressed, while HIF-2α displays tissue specific-expression. HIF-3α is not
well characterized, but one splicing variant of the HIF-3α is an inhibitory PAS domain
protein (IPAS), which inhibits HIF response, by forming transcriptionally inactive
heterodimers, with HIF-1α (Makino et al., 2001).
82
Figure 14. Schematic representation of HIF-α family member protein
domains. HIFs are members of the basic helix-loop-helix (bHLH)/PERARNT-SIM (PAS) domain family of transcription factors that mediate
transcriptional responses to oxygen deprivation. They bind to DNA as
heterodimers composed of an oxygen-sensitive HIF-α subunit (HIF-1α, -2α,
or -3α) and a constitutive HIF-β subunit. The bHLH and PAS domains
found in all HIF family members mediate DNA binding and dimerization,
respectively.
HIF-α
subunits
contain
a
unique
oxygen-dependent
degradation domain (ODD) that controls HIF-α stability in an oxygendependent manner. In addition, HIF family members contain transactivation
domains (TADs) that mediate target gene activation. HIF-1a and HIF-2a
contain two TADs that contribute to target gene activation. Adapted from
Rankin and Giaccia, 2008.
1.8.3. Oxygen-dependent regulation of HIF-1α
Several factors contribute to the induction of HIF-1. HIF-1 activity is responsive to a
number of different stimuli, including cytokines and hormones and the expression of
the HIF-α subunits is tightly regulated at the transcriptional/synthesis level and by posttranslational modifications. Nevertheless, the best-understood mechanisms regulating
HIF-α expression act mainly at the level of protein stability (please see Galban and
Gorospe, 2009; Semenza, 2003; Yee Koh et al., 2008; Zhou and Brune, 2006) for a
more detailed review]. This section will explore the mechanisms of HIF-1α regulation
by oxygen, particularly O2-dependent degradation, which is the canonical pathway for
HIF-1α degradation.
83
Cells continuously synthesize and degrade HIF-1α, but in conditions of normal O2
availability, HIF-1α has a half-life of < 5 minutes (Huang et al., 1996; Yu et al., 1998).
However, under hypoxia HIF-1α degradation is inhibited the levels of this subunit are
stabilized (Jiang et al., 1996). The interface between O2 and HIF-1α is provided by the
oxygen-dependent hydroxylation of two prolyl residues (P402 and P564 in the human
protein) in the ODD domain of this subunit (Fig.15) (Ivan et al., 2001; Jaakkola et al.,
2001). HIF-2α is subject to a similar regulation.
Hydroxylation in the ODD of HIF-α subunits is catalysed by specific prolyl
hydroxilase domain (PHD) enzymes (PHD1, PHD2 and PHD3). PHD2 preferentially
hydroxylitates HIF-1α whereas HIF-2α is a preferential target of PHD1 and PHD3 (see
Schofield and Ratcliffe, 2005 for a review). PHDs require O2 (thus providing the link
with oxygen levels in the cell) and 2-oxoglutarate (2-OG) as substrates and Fe (II) and
ascorbate as co-factors. One oxygen atom is inserted into the HIF-1α peptide at the
prolyl residue while the other reacts with 2-OG, yielding succinate and CO2. The
requirement for iron of PHDs explains why iron chelators (e.g. desferrioxamine) and
iron antagonists (e.g. cobalt chloride) are utilized as hypoxia mimetic compounds in
experimental systems (Salnikow et al., 2004; Wang and Semenza, 1993).
Hydroxylation of at least one of the residues allows the interaction with the von
Hippel-Lindau tumour suppressor protein (VHL), which is the recognition component of
an E3 ubiquitin ligase complex that targets HIF-1α for degradation by the UPS (Fig.15)
(Masson et al., 2001; Maxwell et al., 1999; Ohh et al., 2000). This complex also
includes other proteins, namely elongin B, elongin C, Rbx1 and Cul2, which also
participate in other E3 ubiquitin ligase complexes. Under hypoxia HIF-1α hydroxylation
is suppressed, allowing the subunit to be translocated to the nucleus. There it interacts
with HIF-1β and binds to the HREs in promoter or enhancer regions of target genes
(Kimura et al., 2001; Semenza and Wang, 1992). In addition to regulating HIF-1α
stability, HIF-1 transcriptional activity is also determined by an oxygen-dependent
hydroxylation. This reaction is carried out by an asparaginyl hydroxylase termed as
84
factor-inhibiting HIF (FIH), which hydroxyltates an asparagin residue in the C-terminal
TAD of
HIF-1α (N803 in the human protein)under normoxic conditions. This
hydroxylation inhibits the interaction between HIF-1α and the co-activators p300/CBP
(CREB binding protein), thereby preventing HIF-1 induction of target genes (Fig.15)
(Schofield and Ratcliffe, 2005).
Figure 15. Schematic representation of regulation of HIF-1a protein by
prolyl hydroxylation and proteasomal degradation. There are three
hydroxylation sites in the HIF-1a subunit: two prolyl residues in the oxygendependent degradation domain (ODD) and one asparaginyl residue in the Cterminal transactivation domain (C-TAD). In the presence of oxygen, prolyl
hydroxylation is catalyzed by the Fe(II)-, oxygen- and 2-oxoglutaratedependent PHDs. The hydroxylated prolyl residues allow capture of HIF-1a
by the von Hippel–Lindau protein (VHL), leading to ubiquitination and
subsequent proteasomal degradation. Asparaginyl hydroxylation is catalyzed
by an enzyme termed as factor-inhibiting HIF (FIH) at a single site in the CTAD. This hydroxylation prevents cofactor recruitment. In the absence of
hydroxylation due to hypoxia or PHD inhibition, HIF-1a translocates to the
nucleus, heterodimerizes with HIF-1b and binds to hypoxia-response
elements (HREs) in the regulatory regions of target genes. Adapted from
Weidemann and Johnson 2008.
85
1.8.4. Oxygen-independent degradation of HIF-1α
Several other mechanisms leading to HIF-1α degradation, other than the well
established O2-dependent pathway, have been described. However, the physiological
relevance of these mechanisms remains largely undetermined in most cases. In
addition to the VHL ubiquitin ligase complex, other molecular players were shown to be
able to direct HIF-1α for proteasomal degradation. The receptor for activated protein
kinase C 1 (RACK1) – E3 ubiquitin ligase complex, the Hypoxia-Associated Factor
(HAF;, a putative ubiquitin ligase), Copper Metabolism MURR1 Domain containing 1
protein (COMMD1) have all been reported to mediate O2-independent proteasomal
degradation of HIF-1α, that does not exclude the VHL-mediated pathway (Koh et al.,
2008; Liu et al., 2007; van de Sluis et al., 2009). The ubiquitin ligase that degrades
p53, MDM2, has also been implicated in HIF-1α degradation; its exact role however,
remains controversial (Chen et al., 2003; Hansson et al., 2002; Ravi et al., 2000).
More recently, data from our laboratory as well as an independent report, have
established that the E3 ubiquitin ligase CHIP is likely to be involved in the ubiquitination
and proteasome degradation of HIF-1α. CHIP seems to mediate HIF-1α degradation in
response to the accumulation of intracellular methylglyoxal (MGO) (Bento et al.,
submitted) and during prolonged hypoxia (Luo et al., 2009b). In addition to proteasomal
degradation, HIF-1α has been reported to undergo degradation by calpains (Zhou et
al., 2006) and by the lysosome (Olmos et al., 2009) as mentioned above.
Notwithstanding the diversity of molecular players reported to be involved in HIF-1α
degradation, the physiological relevance and the regulation of these O2-independent
pathways for HIF-1α has remained largely unexplored. A number of hypotheses
regarding the role of O2-independent degradation of HIF-1α will be discussed in
Chapter 4.
86
CHAPTER 2
Material & Methods
87
88
2.1. Cell culture & Media
The retinal pigment epithelium cell line ARPE-19 (LGC Promochem,
Teddington, UK) was cultured in Dulbecco’s modified Eagle’s medium/Ham’s F12
(DMEM:F12) (1:1) supplemented with 10% fetal bovine serum (FBS), antibiotics (100
U/ml penicillin, 100 μg/ml streptomycin and 250 ng/ml amphotericin B) and GlutaMax
(1x). The mouse fibroblast cell line NIH-3T3 (kindly provided by Professor Ana Maria
Cuervo, Department of Developmental and Molecular Biology, Marion Bessin Liver
Research Center, Albert Einstein College of Medicine; Bronx, New York USA) were
grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal
bovine serum, antibiotics and glutamine. The renal cell carcinoma cell line RCC4
VHL -/-
(ECACC, Salisbury, UK) was also grown in DMEM supplemented with 10% fetal bovine
serum, antibiotics and glutamine. All cell lines were maintained at 37ºC under 5% CO 2.
All media, GlutaMax and antibiotics were purchased from Invitrogen (Carlsbad, CA,
USA).
For all experiments, cells were grown in 75 ml flasks until they reached
confluence. The medium was removed and cells were then detached, by adding 3 ml
of GIBCO® TrypLE™ cell dissociation enzyme Invitrogen (Invitrogen, Carlsbad, CA,
USA) to the flask and incubating for 3-5 minutes at 37ºC. Dissociation was stopped by
adding 5 ml of the respective culture medium, the suspension was homogenized and
transferred to a falcon tube. Subsequently, a sample of the cell suspension was taken
and cells were counted in a hemocytometer (improved Neubauer), using an inverted
light microscope. The cell suspension was centrifuged for 5 minutes at 1214 gs (using
a KUBOTA 5400 table centrifuge, with a RS-720M rotor), the supernatant was then
removed and the cells were ressuspendended in the apprpriate volume of culture
medium. Cells were platted in 60 x 15 mm plates or 6-well culture plates. Respectively,
1x106 and 4.5x105 NIH-3T3 cells and 1.25x106 and 5.25x105 ARPE-19 and RCC4
cells wee platted. When appropriated, cells were treated with the following agents: 5
89
mM, 10 mM, 20 mM, 30 mM of ammonium chloride (NH4Cl), for 2,4 or 8 hours; 25 μM,
50 μM, 100 μM, 200 μM of chloroquine for 2, 4, 8 and 12 hours; 10 mM of 3methyadenine (3- MA) for 4 hours; 20 μM of leupeptin for 8 hours (all the abovementioned reagents were purchased from Sigma-Aldrich, St. Louis, MO, USA); 10 μM
or 20 μM MG132 (Calbiochem, San Diego, CA, USA) for 2,4,8 and 12 hours.
Treatments were performed after cells reached confluence. Before performing
the treatments the culture medium was replaced by 2 ml of fresh medium and cells
were left at 37ºC in a CO2 incubator during the incubation period.
2.2. Western Blot
After appropriate treatments, the medium was removed, cells were washed
twice in phosphate-buffered saline (PBS) solution, and cells were harvested in 200-150
μl (60 x 15 mm petri dishes or 6-well culture plates respectively) in 2x Laemmli buffer
(10% SDS), using a cell-scraper. The cell extracts were boiled at 100ºC for 5 minutes
and
sonicated.
Whole
cell
extracts
were
resolved
by
SDS-PAGE
and
electrophoretically transferred onto nitrocellulose (Bio-Rad Laboartories, Hercules, CA,
USA) or polyvinylidene fluoride (PVDF) membranes. The membranes were blocked
with 5% nonfat milk in TBS-T (20 mM Tris, 150 mM NaCl, 0,2% Tween 20, pH 7.6) and
probed for several proteins, using specific primary and horseradish peroxidise (HRP)conjugated secondary antibodies. The antibodies used for this work are listed in TableI.
Immunoreactive
bands
were
visualized
with
an
ECL
(enhanced
chemioluminescence) system (GE Healthcare Bio-Sciences, Uppsala, Sweden) for
detection using film or Immun-Star WesternCTM chemiluminescent kit (Bio-Rad
Laboratories, Hercules, CA, USA), for CCD imaging.
90
2.3. MTT cell viability assay
NIH-3T3 cells were seeded onto 24-well plates (3x105 cells/ well) and allowed
to grow until confluence was reached. Cells were treated with 5 mM, 10 mM, 20 mM,
30 mM of NH4Cl or 25 μM, 50 μM, 100 μM, 200 μM of chloroquine for 2-4 hours.
After the treatments, the cells were washed twice with PBS and incubated with 0.5
mg/ml MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Invitrogen,
Carlsbad, CA, USA) in DMEM medium, without serum or antibiotics, for 2 hours at
37°C in a cell culture incubator. Subsequently, supernatants were removed and the
precipitated dye was dissolved in 300 μl 0.04 M HCl (in isopropanol) and absorbance
was quantified at a wavelength of 570 nm, with wavelength correction at 620 nm, using
a Biotek Synergy HT spectrophotometer (Biotek, Winooski, VT, USA)
2.4. Cyclohexamide-Chase
NIH-3T3 cells cultured in 6-well plates were treated with 50 μM of chloroquine
or 20 μM of MG132 for 2 hours. 25 μM of cyclohexamide (CHX; Sigma-Aldrich, St.
Louis, MO, USA) where then added and cells were harvested at different time points in
2x Laemmli buffer, as described above.
The whole cell extracts were prepared as described and separated by SDSPAGE. Western blot analysis was performed after transference to nitrocellulose
membrane.
91
2.5. Immunoprecipitation
ARPE-19 cells cultured in 60 x 15 mm plates were washed twice with PBS,
scraped off the dishes and collected in ice-cold PBS. Pellets were resuspended in 100
μl of lysis buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 10 mM IOD, 2 mM PMSF, 20
mM Na3MoO4, 0.5% NP-40 and complete-mini protease inhibitor cocktail; Roche
Applied Science, Indianapolis, USA), incubated for 30 minutes on ice and briefly
sonicated.
Following centrifugation at 16,000 g for 10 minutes, 20 μl of the supernatants
were denaturated with 2x Laemmli buffer (input). The remaining supernatants (80 μl)
were transferred to new tubes and 2.5 μg of anti-HIF-1α monoclonal antibodies were
added. Subsequently, 300 μl of lysis buffer, with no NP-40, were added to the mixture
and the samples were incubated overnight at 4ºC with gentle agitation. Thereafter, 50
μl of protein G-Sepharose (GE Healthcare Bio-Sciences, Uppsala, Sweden) were
added to the samples and incubations proceeded at 4ºC for 2 hours. Beads were
washed 3 times with lysis buffer and the immunoprecipitated proteins were eluted with
2x Laemmli buffer and boiled at 100ºC.
Eluted samples were loaded on SDS-PAGE and western blot analyses were
performed using anti-ubiquitin, anti-HIF-1α, anti-Hsc70, anti-LAMP-2 antibodies.
Table I – List of primary and secondary antibodies used for Western Blot.
Antibody
Host
Clone/Cat.#
~MW (kDa)
Dilution
Anti-HIF-1α
Mouse
MA1-516
120 kDA
1:1000
Company supplier
Affinity Reagents,
Rockford, IL, USA
Pierce Antibodies,
Anti-HIF-1α
Rabbit
PA1-16609
120 kDA
1:2000
Thermo
SCIENTIFIC,
Rockford, IL, USA
92
Anti-LAMP2
Rabbit
512200
≥ 100 kDa
1:2000
Zymed-Invitrogen,
Carlsbad, CA, USA
Stressgen,
Anti-Hsc70
Rat
1B5
73
1:2000
Farmingdale, NY,
USA
Anti-V5 tag
Mouse
R960-25
PMW + 2.8
1:2000
Rabbit
PA1-16931
(LC3-I)/ ≈ 17
1:2000
kDA(LC3-II)
Anti-Actin
Anti-MouseHRP
Anti-RabbitHRP
Anti-RatHRP
Mouse
C4
Goat
626520 #
Goat
656120 #
-
Goat
#81-9520
-
CA, USA
Pierce Antibodies,
≈ 19 kDa
Anti-LC3
Invitrogen, Carlsbad,
Thermo
SCIENTIFIC,
Rockford, IL, USA
Millipore-Chemicon,
43
1:100000
-
1:5000/
Bio-Rad, Hercules,
1:20000
CA, USA
1:5000/
Bio-Rad, Hercules,
1:10000
CA, USA
1:8000
Billerica, MA, USA
Zymed-Invitrogen,
Carlsbad, CA, USA
PMW – Protein molecular weight.
Clone/Cat. # – Clone designation or catalogue number of the
antibodies.
2.6. Plasmids and lentiviral vectors
In this work, the pcDNA3 hHIF-1α wt-V5 plasmid used was kindly provided by
Dr. Thilo Hagen, University of Nottingham UK (Hagen et al., 2003). The plasmid
pcDNA3 hHIF-1α (N529A and E530A)-V5 was used with the permission of Dr. João
Ferreira (Ferreira et al., manuscript under preparation). Lentiviral vectors containing
the short hairpin RNA (shRNA) against ATG/ and LAMP-2A, as well as the respective
empty vector, were kindly provided by Dr. Ana Maria Cuervo, Albert Einstein College of
93
Medicine USA (Massey et al., 2008). The promoter along with the hairpin sequences
have been cloned into the pCCL.PPT.hPGK.GFP.Wpre lentiviral transfer construct.
2.7. Transduction with lentiviral vectors
One day before transduction, NIH-3T3 mouse fibroblasts were plated onto 60 x
15mm plates (5x105cells/plate). Cells were allowed to grow until they reached about
40% confluence and the culture medium was replaced with fresh DMEM with 10% FBS
and antibiotics (2 ml/plate). Prior to adding the lentivirus (1x107 TU), 6μg/ml of
polybrene Sigma-Aldrich, St. Louis, MO, USA) was added to the medium (to a final
volume of 2 ml). Cells were incubated with the lentiviral particles for 2 days at 37ºC and
then the medium was replaced. Culture medium was replaced every 2 days and at the
seventh day cells were washed with PBS 1X and the efficiency of transduction was
verified by fluorescent microscopy for GFP expression.
Plates with 90-95% of transduced cells were then treated with 200 μM of
chloroquine or 20 μM of MG-132 for 8 hours. Cells were harvested in PBS and the
pellet has solubilized in Tris-HCl 50 mM, NaCl 150 mM and 0.5% of NP-40. 20 μl of
whole lysate were denatured in Laemmli buffer 4X, boiled for 5 minutes and Western
blot was performed as described above.
2.8. Transient transfection
One day before transfection, ARPE-19 cells were seeded in 60 x 15mm plates
so that the cells were 90% confluent at the time of transfection. For each transfection
sample, 4 μg of plasmid DNA were diluted in 100 μl of Opti-MEM I Reduced Serum
Medium (Invitrogen, Carlsbad, CA, USA). Subsequently, 10 μl of Lipofectamine 2000
94
(Invitrogen, Carlsbad, CA, USA) transfection reagent were diluted in 100 μl of OptiMEM I Medium and incubated for 5 minutes at room temperature. After 5 minutes of
incubation, diluted DNA and diluted Lipofectamine were combined and incubated
together for 20 minutes at room temperature. Meanwhile, the culture medium of each
well was removed and cells were washed twice with PBS and cultured in 2 ml of OptiMEM I Medium.
After the incubation, the DNA/Lipofectamine complexes (total volume 200 μl)
were added to each plate. The plates were mixed gently and cells were incubated at
37ºC in a CO2 incubator for ≈16 hours prior to test for transgene expression. Opti-MEM
I Medium was replaced by full DMEM:F12 medium (with FBS) after 5 hours and 15
minutes of DNA/Lipofectamine incubation.
2.9. Immunocytochemistry
ARPE-19 cells were grown o coverslips in a 12-well plate (5x105 cells/ well were
seeded). When appropriate, the culture medium was removed and cells were washed
in PBS 1X and fixed with 4% paraformaldehyde (PFA) for 15 minutes at room
temperature. The cells were washed with PBS 1X and 500 μl of 20 mM NH4Cl (in PBS
1X) was added to each well. After 10 minutes the cells were washed again and then
incubated in 500 μl of a permeabilization/blocking solution (0,05% saponin, 0,5%
bovine serum albumin,BSA, in PBS 1X) for 15 minutes. The fixed cells were again
washed with PBS 1X, prior to incubation with 1:200 anti-V5 tag antibody (in
permeabilization/blocking solution) (Table II) for 1,5 hours at room temperature. The
coverslips were then rinsed three times in the permeabilizing/blocking solution and the
cells were incubated with 1:200 Alexa 568-conjugated goat anti-mouse antibody
(TableII) (in permeabilization/blocking solution) for 1 hour at room temperature.
Subsequently, the coverslips were washed with permeabilization/blocking solution and
95
mounted with Fluoro-Gel II (Electron Microscopy Sciences, Hatfield, PA, USA)
mounting medium.
The cells were imaged by fluorescence microscopy using a Leica DFC350 FX
fluorescence microscope (Leica-Microsystems, Bannockburn, IL, USA).
Table II – List of primary and secondary antibodies used for Immunocytochemistry
.
Antibody
Host
Clone/Cat.#
Dilution
Anti-V5 tag
Mouse
R960-25
1:200
Anti-mouse- Alexa
Fluor® 568
Company supplier
Invitrogen, Carlsbad,
CA, USA
Molecular Probes -
Goat
#A11004
1:200
Invitrogen, Carlsbad,
CA, USA
Clone/Cat. # – Clone designation or catalog number of the antibodies.
2.10. Statistical Analysis
Data are reported as the mean ± standard deviation (SD) of at least three
independent experiments except when it is stated otherwise. Comparisons between
multiple groups were performed by one-way analysis of variance test (ANOVA) with the
Dunnet’s multiple comparison tests, while comparisons between two groups were
performed using t test analysis [GraphPad Prism 5.0 software (GraphPad Software, La
Jolla, CA, USA)]. In all cases, p < 0.05 was considered significant.
96
CHAPTER 3
Results
97
98
3.1. HIF-1α is degraded both by the proteasome and the lysosome;
As described in the previous section, one of the most important determinants of HIF-1
activity is the regulation the subunit HIF-1α protein levels. As cellular adaptive
responses to hypoxia are mediated by HIF-1, the oxygen-dependent degradation of
HIF-1α, mediated by the tumour suppressor VHL, is of particular importance (Maxwell
et al., 1999). However several other UPS-dependent mechanisms regulating the
protein stability have been identified (Koh et al., 2008; Liu et al., 2007; Luo et al.,
2009b; van de Sluis et al., 2009; Zhou et al., 2006).
The canonical view is that HIF-1α levels are mainly regulated by ubiquitination
and proteasomal degradation. Interestingly, recent data from our laboratory have
further show that HIF-1α is targeted for proteasomal degradation by the E3 carboxyl
Terminus of the Hsc70-Interacting protein (CHIP) in the presence of methylglyoxal
(MGO) in ARPE-19 cells (Bento et al., under review). During these experiments
however, it was noted that the use of proteasome inhibitors did not fully prevent HIF-1α
degradation, thus suggesting that another degradation pathway could be involved in
this subunit degradation.
3.1.1. Inhibition of lysosomal degradation increased HIF-1α protein levels
Taking into account that lysosomal degradation is the other major pathway for
protein degradation in the cell, we treated ARPE-19 cells with several lysosome
inhibitors and the levels of HIF-1α protein were assessed by Western blot after SDSPAGE. Surprisingly the data reveals that this treatment led to accumulation of HIF-1α
(Fig.16), even under normoxia.
Three different well established inhibitors of the lysosome, which act by different
mechanisms, were used in order to eliminate possible off-target effects of the
treatments on HIF-1α levels. NH4Cl and chloroquine (CQ) are acidotropic drugs that
99
function as intralysosomal proton trapping agents that raise the lysosomal pH,
suppressing the activity of lysosomal acid hydrolases (see Klionsky et al., 2007b fore
references). Additionally, chloroquine also acts by inhibiting specifically the lysosomal
protease cathepsin B (Wibo and Poole, 1974). However both drugs have been report
to affect to some degree protein synthesis (Seglen et al., 1979) and NH4Cl can disturb
the nitrogen metabolism. Also, these lysosomotropic compounds can induce swelling of
lysosomes and other acidic comparts, interfering with endocytosis, secretion and
mobility (see Klionsky et al., 2007b). Nevertheless, both drugs are still widely used to
inhibit lysosomal degradation. Leupeptin is a protease inhibitor of small molecular
weight that has been shown to inhibit the lysosomal cysteine proteases cathepsins B,
H and L (see Seglen et al., 1979), although it can also inhibit some non-lysosomal
proteases (e.g. calpains Fuertes et al., 2003a; Lee and Goldberg, 1998)
Figure 16. Treatment with inhibitors of lysosomal degradation leads to
HIF-1α accumulation in ARPE-19 cells. ARPE-19 cells were incubated
with 100 μM of leupeptin, 20 mM of NH4Cl, 200 μM of chloroquine (CQ) or
20 uM of MG-132 during 8 hours. Control cells were left untreated. Cells
were harvested in PBS and the pellet was homogenized in Tris-HCl 50 mM,
NaCl150 mM and 0.5% of NP-40. 20 ul of total lysate were denatured in
Laemmli buffer. Samples were separated by SDS-PAGE and transferred to
PVDF membranes. The membranes were probed with specific antibodies
against HIF-1
and β-actin. β-actin was used as a loading control.
The control experiment (without any inhibitor) shows very low levels of HIF-1α protein,
in agreement with the very short half-life described for this protein under normoxic
100
conditions (Huang et al., 1996; Yu et al., 1998). Conversely, inhibition of the
proteasome, using MG132, led to a dramatic increase in HIF-1α levels, since under
these conditions HIF-1α is expected to be degraded mainly through the canonical VHLdependent pathway. In comparison with MG132, treatments with lysosome inhibitors
induced only a small increase in HIF-1α. This suggests that the lysosome is
responsible for the degradation of a small fraction of HIF-1α. Among the different
inhibitors, leupeptin induced the lowest accumulation of HIF-1α, consistent with other
reports that this drug is less efficient in inhibiting the lysosomal activity than acidotropic
compounds (e.g. (Fuertes et al., 2003a; Seglen et al., 1979).
Altogether, experiments using three different lysosome inhibitors suggest that HIF1α is beeing degraded in the lysosome.
3.1.2. Inhibitors of lysosomal degradation increase HIF-1α protein levels in a
dose-dependent manner
In order to further confirm if the increase in HIF-1α could be ascribed to
lysosome inhibitors, NIH-3T3 cells were exposed to increasing concentrations of NH4Cl
and chloroquine for different incubation periods and the levels of Hif1α protein were
assessed by Western blot. NIH-3T3 is an immortalized mouse fibroblast cell line, in
which HIF-1α homolog, Hif1α, undergoes similar oxygen dependent regulation (e.g.
Botusan et al., 2008; Laughner et al., 2001; Li et al., 2005).
Both macroautophagy
and CMA are well characterized in this cell line, thereby facilitating the identification of
the pathway involved in HIF-1α degradation. Autophagy is well characterized in this cell
line, particularly macroautophagy and CMA (Kaushik and Cuervo, 2006; Massey et al.,
2008; Massey et al., 2006d), thereby facilitating the determination of which pathway of
lysosomal degradation is involved in the regulation of HIF-1α protein levels.
Additionally, the use of an additional cell line will allow establishing the regulation of
101
HIF-1α as a broader mechanism, not limited to ARPE-19 (human epithelial cells). The
effect of the treatments on cell viability was assessed by the MTT assay.
Cell lysates were separated on SDS-PAGE gel and Western blot of the
endogenous Hif1α showed that the NH4Cl treatment led to a dose-dependent increase
in Hif1α protein levels, both at 2h and 4h of incubation (Fig. 17, A and B respectively).
15
10
5
M
10
M
m
13
2
30
m
G
N
H
4
20
H
4
N
N
H
4
10
m
M
M
5m
C
H
4
N
M
150
100
50
10
M
M
m
13
2
30
m
G
M
N
H
4
20
H
4
N
N
H
4
10
m
M
M
5m
C
H
4
N
M
0
T
HIF-1 relative protein levels
B
M
0
T
A
HIF-1 relative protein levels
However, the sample size is not sufficient for solid statistical analysis.
Figure 17. Treatment with ammonium chloride (NH4Cl) leads to a dosedependent accumulation of HIF-1α in NIH-3T3 cells. A,B. NIH-3T3 cells
were incubated with the depicted concentrations of NH4Cl or 10 uM of MG132 for 2 hours A) or 4 hours B). Control (CT) cells were left untreated.
Cells were harvested in 2x Laemmli buffer and the whole extract was boiled
at 100º for about 5 minutes and then sonicated. Samples were separated by
SDS-PAGE and transferred to nitrocellulose membranes. The membranes
were probed with specific antibodies against HIF-1
and β-actin. β-actin
was used as a loading control. The graph data represents the mean ± SD of
three independent experiments. Significance was set for differences from the
control with p≤ 0,05 (one way ANOVA with the Dunnet’s comparison test).
102
In a separate experiment, after being exposed to NH4Cl, 3T3 cells were
incubated with MTT for 2 or 4 hours in culture medium and the cell viability was inferred
following spectrophotometric analysis (Fig. 18). At the abovementioned times of
incubation, the treatments with NH4Cl did not appear to compromise cell viability.
A
B
2.0
Relative cell viability
2.0
1.5
1.0
0.5
**
**
1.5
1.0
0.5
M
10
m
M
13
2
30
m
M
M
G
H
4
20
N
H
4
N
H
4
10
m
M
M
C
5m
N
G
M
H
4
M
10
m
M
13
2
30
m
M
20
H
4
N
H
4
N
N
H
4
10
5m
m
M
M
T
C
H
4
N
T
0.0
0.0
N
Relative cell viability
2.5
Figure 18. Treatment with ammonium chloride (NH4Cl) does not
compromise cell viability in NIH-3T3 cells. A,B. NIH-3T3 cells were
plated in 24-well plates and were treated with the depicted concentrations of
NH4Cl or 10 uM of MG-132 for 2 hours A) or 4 hours B). CT cells were left
untreated. Cells were subsequently incubated with 0,5 mg/ml MTT in
DMEM culture medium (without serum or antibiotics) at 37ºC for 2 hours.
The precipitate was obtained, dissolved in isopropanol and absorvance was
quantified at 570 nm and 620nm.The results represent the mean + SD of at
least three independent experiments. Significance was set for differences
from the control with * p≤ 0,05, ** p≤ 0,01 (one way ANOVA with the
Dunnet’s comparison test).
Conversely, treatments with chloroquine led to increased expression of Hif1α
only in the lower range of concentrations used (Fig.19). Indeed, incubation with higher
concentrations of chloroquine decreased Hif1α levels at either incubation time.
Chloroquine has been reported to decrease cell viability, even at concentrations which
are used for inhibiting the lysosome (Seglen et al., 1979). However, in this case, the
treatments have not altered significantly the cell viability (Fig.20). Nevertheless, the
103
incubation with chloroquine induced severe morphological changes on the cells, which
HIF-1 relative protein levels
5
2.5
4
3
2
1
M
10
M
M
13
2
20
0
M
G
C
Q
10
0
C
C
2.0
1.5
1.0
0.5
M
10
M
13
2
20
0
M
G
M
C
Q
10
0
C
C
Q
Q
50
M
C
C
Q
25
M
0.0
T
B
Q
Q
50
M
C
C
Q
25
M
0
T
A
HIF-1 relative protein levels
could be affecting their functioning.
Figure 19. Treatment with chloroquine (CQ) leads to a dose-dependent
accumulation of HIF-1α in NIH-3T3 cells. A,B. NIH-3T3 cells were
incubated with the depicted concentrations of chloroquine (CQ) or 10 uM of
MG-132 for 2 A) or 4 hours B). Control (CT) cells were left untreated.
Western blot was performed as described in figure 17, and the membranes
were probed for HIF-1
and β-actin. β-actin was used as a loading control.
The graph data represents the mean ± SD of three independent experiments,
except for A. Significance was set for differences from the control with p≤
0,05 (one way ANOVA with the Dunnet’s comparison test).
3.1.3. Stabilization of HIF-1α by chloroquine is independent of VHL-dependent
degradation in RCC4 cells
The best described mechanism for HIF-1α regulation is degradation by the
proteasome under normoxia. In the presence of physiological O2 tension, two proline
residues in the ODD domain of HIF-1α (Pro402 and Pro564 in the human protein) are
hydroxylated
104
by
specific
prolyl
hydroxilases
(PHD).
This
oxygen-dependent
A
B
1.5
Relative cell viability
1.0
0.5
0.0
1.0
0.5
M
10
M
M
G
13
2
M
20
0
Q
10
0
C
Q
50
M
Q
C
C
T
25
M
C
Q
M
10
M
M
G
13
2
M
20
0
Q
10
0
C
50
M
Q
C
Q
C
C
Q
C
25
M
T
0.0
C
Relative cell viability
1.5
Figure 20. Treatment with chloroquine (CQ) does not compromise cell
viability in NIH-3T3 cells. A,B. NIH-3T3 cells were treated with the
depicted concentrations of chloroquine (CQ) or 10 uM of MG-132 for 2 A)
or 4 hours B). Control (CT) cells were left untreated. MTT cell viability
assay was performed as described in figure 18. Results represent the mean ±
SD of at least three independent experiments. Significance was set for
differences from the control with p≤ 0,05, (one way ANOVA with the
Dunnet’s comparison test).
hydroxylation mediates the interaction of HIF-1α with VHL and subsequent degradation
of HIF-1α by the UPS (Maxwell 1999, Ohh 2000). PHD activity requires O2 and 2-OG
as co-substrates and Fe (II) and ascorbate as co-factors (Salnikow et al., 2004; Wang
and Semenza, 1993; Yuan et al., 2003). However, the increase of the intracellular pH
might affect the availability or the binding of these compounds to the PHDs (hypothesis
discussed in greater detail in the following chapter). NH4Cl and lysosomotropic amines
such as chloroquine are known to lead to small increases of the pH of the cytoplasm.
To address the possibility that the lysosome inhibitors used led to an increase in HIF1α levels by disrupting PHD-dependent hydroxylation of HIF-1α (and thus VHLdependent degradation) RCC4 cells, a renal carcinoma cell line, deficient for VHL was
used (Maxwell et al., 1999).
RCC4 cells were treated with chloroquine for 12h. Cell extracts were prepared and
samples were separated by SDS-PAGE. Levels of HIF-1α protein were assessed by
Western blot. As a control for proteasome-dependent degradation of HIF-1α, RCC4
105
were treated with MG132 for 12h (Fig. 21). The chloroquine treatment increased HIF1α levels in a dose-dependent. Because these cells lack VHL, this data suggests that
the lysosome inhibitors are not increasing HIF-1α by interfering with the PHD-
10
8
6
4
2
10
M
13
2
M
G
C
Q
25
M
C
Q
C
50
M
0
T
HIF-1 relative protein levels
dependent degradation.
Figure 21 – Treatment with chloroquine leads to a dose-dependent
accumulation of HIF-1α in RCC4VHL-/- cells. RCC4VHL-/-cells were
incubated with the depicted concentrations of chloroquine (CQ) or 10 μM of
MG-132 for 12 hours. Control (CT) cells were left untreated. Western blot
was performed as described in figure17, and the membranes were probed for
HIF-1
and β-actin. β-actin was used as a loading control. The graph data
represents quantification of relative bans intensities of one experiment.
3.1.4. Relative efficiency of HIF-1α degradation by lysosomal and proteasomal
pathways
Because both the proteasome and the lysosome might be implicated in
degradation of HIF-1α, the relative effenciency of the two pathways was addressed.
In order to do so, 3T3 cells were treated with chloroquine or MG132 for 2h, and
subsequently incubated with the protein translator inhibitor CHX up to 5 hours (in the
presence of the inhibitors). Cell extracts were obtained at different times and Hif1α
levels were measured after Western blot (Fig.22). Due to an experimental error, 0’ time
for the MG132 treatment could not be used, so the densiometric analysis was
performed starting at 30 minutes for both treatments blot (Fig.22).
106
Despite this error, in cells treated with proteasome inhibitor MG132 a
progressive reduction in the levels of Hif1α is apparent, after addition of CHX blot
(Fig.22). This result indicates that proteolytic pathways other than the UPS, contribute
considerably for regulation of Hif1α levels. Conversely, the CHX-chase revealed a
reduction in Hif1α half-live in cells treated with chloroquine blot (Fig.22). Thus, in
addition to degradation in the lysosome, Hif1α undergoes rapid proteolysis by other
pathways, presumably by the UPS.
C
Relative HIF-1 protein levels
A
B
MG132
CQ
1.0
0.5
0.0
0
2
4
6
CHX (h)
Figure 22 – The proteasomal pathway of Hif1α degradation is more
efficient than the lysosomal pathway in NIH-3T3 cells. A,B. NIH-3T3
cells were incubated with 50 μM chloroquine (CQ) or 20 μM of MG-132 for
2 hours A) and B) respectively. Subsequently, 25 μM was added to each
sample in the presence of the proteolytic inhibitors. Cells were harvested in
2x Laemmli buffer after the periods depicted. Western blot was performed as
described in figure 17, and the membranes were probed for HIF-1
and β-
actin. β-actin was used as a loading control. C. The graph data represents the
densiometric analysis of each set of samples.
Altogether, the data suggests that the lysosomal pathway could contribute
significantly for degradation of Hif1α in 3T3, although this protein is more efficiently
107
degraded by the proteasome, consistent with what has been showed in the literature
for HIF-1α.
3.2. Lysosomal degradation of HIF-1α is not mediated by macroautophagy
The main regulatory step in the lysosomal degradation is the transport of the
substrates into the lytic compartment. Cytosolic substrates have been described to
enter the lysosome by different pathways, including macroautophagy, CMA and
microautophagy (Cuervo, 2004b).
The data obtained so far supports the view that HIF-1α can be degraded by the
lysosome. In fact, during the completion of this work, Olmos and colleagues have
shown that HIF-1α also undergoes degradation by a lysosomal-dependent mechanism
in HK2 human kidney cells (Olmos et al., 2009), supporting the results obtained in our
lab. However, the pathway for delivery of HIF-1α to the lysosome was not identified.
Subsequently, further experiments were made, in order to identify the lysosomal
pathway responsible for directing HIF-1α for the lysosome.
3.2.1. Inhibition of macroautophagy does not stabilize HIF-1α
Macroautophagy is the best-characterized type of autophagy. Macroautophagy
was traditionally regarded as a form of bulk degradation of portions of the cytoplasm,
including proteins and organelles. However, an increasing body of recent evidence
suggests that macroautophagy can be a highly selective process.
There is now ample evidence for selective autophagic degradation of various cellular
structures, including protein aggregates, mitochondria, and microorganisms (Xie and
Klionsky, 2007). Surprisingly, not only protein aggregates but also soluble cytosolic
proteins can be selectively degraded by macroautophagy, including ribosomes and
108
regulators of the NF-κB pathway (Kim et al., 2008; Kraft et al., 2008; Tung et al., 2010;
Xiao, 2007). Indeed, studies have shown that proteins prone to aggregate are primarly
targets of macroautophagy in the soluble form, before aggregation (Hara et al., 2006;
Komatsu et al., 2006). Selective macroautophagy involves both ubiquitination of the
substrate and recognition by an ubiquitin-binding cargo receptor (e.g p62) that
simultaneously interacts with macroautophagic machinery (for a recent review
4
3
2
1
3M
l
4C
N
H
C
A
0
T
HIF-1 relative protein levels
see(Lamark and Johansen, 2010)).
Figure 23. Inhibition of macroautophagy does not lead to Hif1α
accumulation in NIH-3T3 cells. NIH-3T3 cells were incubated with 10mM
of NH4Cl or 10 mM of 3-methyladenine (3-MA) for 4 hours. Control (CT)
cells were left untreated. Western blot was performed as described in figure
17, and the membranes were probed for HIF-1
and β-actin. β-actin was
used as a loading control. The graph data represents quantification of relative
bans intensities of only one experiment.
Thus, in order to assess the possible role of macroautophagy in HIF-1α
degradation, 3T3 cells were treated for 4 hours with 3-methyladenine (3-MA), a class III
PI3K inhibitor, that blocks the autophagosome formation (Klionsky et al., 2007b). In the
same experiment, 3T3 cells were treated with NH4Cl as a positive control of Hif1α
stabilization (Fig. 23). Levels of Hif1α were analysed after Western blot and, in the cells
treated with 3-MA there was a marked decrease of Hif1α even when compared to the
untreated cells.
Reduction in Hif1α after inhibition of macroautophagy indicates that the protein
is being degraded in the lysosome by a process that is not macroautophagy. 3-MA has
109
been reported to inhibit autophagy without impairing protein synthesis or cell viability
(Klionsky et al., 2007b; Seglen and Gordon, 1982). However, this compound can also
inhibit class I PI3-kinase activity, and because it used at very high concentrations (10
mM), it can target other kinases and affect other cellular processes (Klionsky et al.,
2007b; Mizushima et al., 2010).
To avoid the possible off-target effects of 3-MA, shRNA were used to knock
down ATG7, thus preventing macroautophagy in 3T3 cells. ATG7 is essential for
expansion of the phagophore, and silencing or knocking down of ATG7 disrupts
macroautophagy (Komatsu et al., 2005; Liu et al., 2009a).
A
B
Figure 24. Silencing of ATG7 disrupts macroautophagy but does not
lead to stabilization of Hif1α in NIH-3T3 cells. NIH- 3T3 Cells were
transduced with empty vector or ATG7 shRNA and left for 5 days at 37ºC.
The cells were then harvested in PBS 1X and the pellet was solubilized in
Tris-HCl 50 mM, NaCl 150 mM and 0.5% of NP-40. 20 ul of total lysate
were denatured in Laemmli buffer 4X. Samples were separated by SDSPAGE and transferred to PVDF membranes. The membranes were probed
with specific antibodies against HIF-1α and β-actin A) or LC3 and β-actin
B). β-actin was used as a loading control. Vectors were kindly provided by
Dr. Ana Maria Cuervo.
NIH-3T3 cells were transduced with a lentiviral vector containing shRNA against
ATG7. Cells were left for 5 days expressing the shRNA and then harvested. The
lysates were separated by SDS-PAGE, followed by Western blot analysis (Fig. 24). As
a control experiment cells were transduced with an empty vector. Cells expressing
110
shRNA against ATG-7 show a decrease in the conversion of LC3-I to LC3-II. Lipidation
of LC3-I to LC3-II is necessary for the elongation of the phagophore membrane and is
a widely used autophagy assay (Klionsky et al., 2007). The decrease in LC3 lipidation
described in figure 24 indicates a disruption of macroautophagy. However, there was
no detectable increase in Hif1α levels in these cells comparing with the cells
transduced with the empty vector.
Thus, inhibition of macroautophagy by expressing shRNA against ATG7 further
confirms the results obtained with the pharmacological approach and suggested that
another pathway mediates lysosomal degradation of HIF-1α.
3.3. HIF-1α undergoes lysosomal degradation trough CMA
Apart from macroautophagy, the other main autophagic pathways that have
been shown to coexist in mammalian cells are microautophagy and CMA. CMA is a
selective type of autophagy that provides an alternative source of aminoacids under
conditions of prolonged starvation (Cuervo, 2004b) and cellular quality control
(Bandyopadhyay and Cuervo, 2007; Kiffin et al., 2004). However, specialized functions
for CMA in particular cell types, related with degradation of a specific protein or group
of proteins (including transcription factors), has been described (Cuervo, 2009).
Particularly, in neurons, CMA modulates the intracellular levels of the transcription
factor myocyte enhancer factor D (MEF2D), promoting cell survival (Yang et al., 2009).
In renal tubular cells, degradation of the transcription factor Pax2 trough CMA
modulates kidney growth (Franch et al., 2001).
Because of its specificity, CMA is likely to be the pathway that mediates
degradation of HIF-1α in the lysosome. Consistent with this hypothesis, Olmos and
colleagues showed that, in HK-2 cells, HIF-1α interacts with the lysosome-associated
membrane protein type 2A (LAMP-2A), which acts as a receptor for substrates of CMA
111
(Cuervo and Dice, 1996, 2000a, b). However, to the best of our understanding, there
are no pharmacological inhibitors of CMA. Thus, alternative approaches had to be
used to show the involvement of CMA in degradation of HIF-1α.
3.3.1. HIF-1α interacts with LAMP-2 and Hsc70 in ARPE-19 cells
All known CMA substrates interact with cytosolic Hsc70 and LAMP-2A (Kaushik
and Cuervo, 2009). Hsc70 is involved in several processes in the cell and for most
functions, Hsc70 binds to stretches of hydrophobic residues in the substrates (Cuervo,
2009). However, in the CMA pathway, Hsc70 selectively recognizes the KFERQ motif
of the substrates. In fact, Hsc70 is the only chaperone able to recognize the KFERQ
motif and recognition of the KFERQ motif by Hsc70 leads to the delivery of the
substrate to the lysosomes (Cuervo, 2009).
For the substrate to be translocated to the lysosomal lumen it must first bind to
the CMA receptor LAMP-2A at the lysosomal membrane (Bandyopadhyay et al., 2008;
Cuervo and Dice, 1996). LAMP-2A is one of three spice variants of LAMP-2 protein,
but it is the only one involved in CMA and LAMP-2A levels at the lysosomal membrane
directly correlate with CMA activity (Cuervo and Dice, 1996Cuervo and Dice, 2000b).
To confirm that in ARPE-19 cells HIF-1α interacts with LAMP-2A and Hsc70,
co-immunoprecipitation assays were performed in cells treated with chloroquine or
MG132 for 8 hours. Control cells were left untreated and in the mock experiment cells
were treated with MG132 in order to ensure accumulation of HIF-1α would be present.
112
A
B
Figure 25. HIF-1α interacts with LAMP2 and HSC70 after lysosome or
proteasome inhibition, in ARPE-19 cells. ARPE-19 cells were incubated
with 200 μM of chloroquine (CQ) or 20 μM of MG-132 for 8 hours. Control
cells (CT) were left untreated and mock cells were also treated with 20 μM
of MG-132 for 8 hours. Cells were harvested in PBS and the pellet has
solubilized in Tris-HCl 50 mM, NaCl 150 mM and 0.5% of NP-40. A
(Input). 20 μl of whole lysate were denatured in Laemmli buffer 4X. B (IP:
HIF-1α). The remaining sample was incubated with antibodies against HIF1α overnight at 4ºC, except for mock cells. Subsequently the lysates were
incubated with protein G sepharose beads for two hours. The beads were
washed 3 times with lysis buffer 0.15 % of NP-40 and denatured with
Laemmli buffer 4X (IP: HIF-1α). All samples were separated by SDS-PAGE
and transferred to PVDF membranes. The membranes were probed with
specific antibodies against LAMP2, HSC70 and HIF-1α.
Cells were harvested and part of the samples was incubated with antibody against HIF1α. In the mock experiment no antibody as used. HIF-1α was then immunoprecipitated
with protein G sepharose beads. Subsequently, total lysates and immunoprecipitates
were separated by SDS-PAGE and the levels of HIF-1α, LAMP-2 and Hsc70 were
assessed by Western blot (Fig. 25). Because at the time that the experiment was
performed no antibody able to recognize the human LAMP-2A was available in the
laboratory, an antibody against LAMP-2 was used instead. This antibody also
recognizes the isoforms –B and –C of LAMP-2.
113
As expected, treatment with MG132 led to a greater accumulation of HIF-1α
and as a consequence, a higher amount of HIF-1α was immunoprecipitated under that
condition. In the mock experiment (without antibody) only a very faint band was visible,
indicating that there is virtually no unspecific binding of HIF-1α to the beads. The
treatments did not alter the levels of either Hsc70 or LAMP-2. In the control coimmunoprecipitation experiment, very low levels of Hsc70 or LAMP-2 were detected,
showing that HIF-1α is being degraded in the proteasome. Treatment with chloroquine
or MG132 resulted in an increase in interaction between HIF-1α and Hsc70, as well as
with LAMP-2 (presumably LAMP-2A). However, levels of LAMP-2 that interact HIF-1α
are markedly higher after chloroquine treatment when compared with MG132 treatment
(Fig. 25).
3.3.2. HIF-1α has a KFERQ-like motif (both mouse and human protein)
In addition to interaction with LAMP-2A and Hsc70, other requirements for a
protein to be stablished as a CMA substrate: the presence of a targeting pentapeptide
motif, biochemically related to KFERQ (hereafter referred to as “KFERQ motif”), in its
amino acid sequence (Dice, 1990).
The KFERQ motif is degenerated and allows for a series of different amino acid
combinations. Indeed, a “pure” KFERQ sequence is only found in ribonuclease A. The
characteristics that define a KFERQ motif are however well defined: the sequence
should contain Q as a flanking amino acid, either at its beginning or at the end; it could
contain up to two of the allowed hydrophobic residues (I, F, L or V) or two of the
allowed positive residues (R or K) but only one negative charged residue (E or D)
(Cuervo, 2009). In certain proteins Q could be substituted by N (Cuervo et al., 1994),
forming a non-canonical KFERQ motif. This substitution however, does not work for all
proteins.
114
In 2009, Olmos et al. reported a failure in identifying a KFERQ motif in HIF-1α
(Olmos et al., 2009). However, we were able to identify a non-canonical KFERQ in the
human HIF-1α amino acid sequence and two canonical KFERQ motifs in the mouse
protein sequence (Fig. 26). Importantly, the presence of a KFERQ motif in the
sequence of a protein only indicates that it can be potentially degraded by CMA, but is
not sufficient to ensure that the protein is a CMA substrate. Thus, in order to establish a
protein as a substrate for CMA, a number of experimental evidence is required.
Figure 26. Both human HIF-1α and the mouse homolog Hif1α have KFERQlike motifs in their sequence. Detail of the sequence alignment of the human HIF1α and the mouse homolog Hif1α. The KFREQ-like motifs are highlighted in pink in
the human sequence (529NEFKL533) and in yellow in the mouse sequence
(512ERLLQ516 and
686RVIEQ690).
The KFERQ-like motif of human HIF-1α is non-
canonical, with a N in substitution of a Q, more commonly found in these motifs.
3.3.3. Inhibition of CMA leads to accumulation of HIF-1α
The rate limiting step for CMA is the binding of the substrate proteins to LAMP2A. In fact, knockdown of LAMP-2A was shown to lead to the selective blockade of
CMA (Kaushik and Cuervo, 2009 Massey et al., 2006a).
To confirm if degradation of HIF-1α in the lysosome is indeed mediated by
CMA, 3T3 cells were infected with lentivirus and transduced with a vector constructed
to express shRNA for LAMP-2A (the vector was kindly provided by Ana Maria Cuervo)
(Massey et al., 2008). Briefly, cells were transduced with the lentiviral constructs and
harvested 5 days post infection. Lysates were separated by SDS-PAGE and levels of
115
HIF-1α were assessed by Western blot (Fig. 27). As a control experiment, cells were
transduced with an empty vector.
The results show an increase in the levels of HIF-1α after knockdown of LAMP2A. Hence, blockade of CMA leads to stabilization of Hif1α, supporting the hypothesis
that Hif1α undergoes CMA-mediated lysosomal degradation (Fig. 27).
Figure 27. Silencing of LAMP-2A leads to stabilization of Hif1α in NIH3T3 cells. NIH- 3T3 Cells were transduced with empty vector or LAMP-2A
shRNA and left for 5 days at 37ºC. The cells were then harvested in PBS 1X
and the pellet was solubilized in Tris-HCl 50 mM, NaCl 150 mM and 0.5%
of NP-40. 20 ul of total lysate were denatured in Laemmli buffer 4X.
Samples were separated by SDS-PAGE and transferred to PVDF
membranes. The membranes were probed with specific antibodies against
HIF-1α, LAMP-2A and β-actin. β-actin was used as a loading control.
Vectors were kindly provided by Dr. Ana Maria Cuervo.
3.3.4. Mutant HIF-1α protein with an altered KFERQ domain accumulates under
basal conditions in ARPE-19 cells
Evidence reported here suggests that the mouse HIF-1α is indeed degraded
trough CMA. However, while the mouse protein has two classical KFERQ motifs, the
human HIF-1α has only one non-canonical KFERQ motif, which might suggest
differences in the regulation of HIF-1α protein between the two species. The specific
recognition of this motif by the cytosolic chaperone Hsc70 leads to the delivery of the
116
substrate to the lysosome and subsequent degradation. Mutation of this motif is
predicted to prevent the CMA-mediated degradation of the substrates (Cuervo, 2009).
To address this, ARPE-19 cells were transfected with plasmids containing wild-type
HIF-1α (HIF-1αwt) or HIF-1α in which mutations were introduced, in the KFERQ
sequence (the sequence
529NEFKL533
was replaced by
529AAFKL533,
hereafter referred
to as HIF-1αAA), turning it non-functional. Both proteins had a V5 tag attached to their
carboxylic terminal. Cells were allowed to overexpress the mutant proteins for ≈16
hours and then cells were treated with chloroquine or MG132. To generate appropriate
controls, some cells were not treated with proteolytic inhibitors. After 8 hours of
incubation with the inhibitors cells were harvested. An aliquot was used for
immunoprecipitation experiments and incubated overnight with antibody against LAMP2. Both total lysates and immunoprecipitates were separated by SDS-PAGE. LAMP-2
and V5 tags were detected by Western blot (Fig. 28).
In the controls (without treatment with inhibitors), higher levels of HIF-1αAA were
detected, as compared to HIF-1αwt, indicating that under basal conditions the mutated
protein is more stable. Treatment with chloroquine resulted in a marked accumulation
of the wild type protein, but only in a minor increase in HIF-1αAA, implying that the
protein is not undergoing degradation in the lysosome (Fig.28).
These results provide evidence that the KFERQ motif present in human HIF-1α
is sufficient and necessary to target the transcription factor for lysosome degradation,
presumably by CMA. Inhibition of the proteasome following treatment with MG132 led
to a greater accumulation of both wild type and mutant HIF-1α. Under these conditions
HIF-1αwt that co-precipitated with LAMP-2 suggests that more HIF-1α is being directed
to the lysosome for degradation, presumably through CMA. The increase in the
interaction of HIF-1α with LAMP-2 might simply be a result of the higher levels of HIF1α present in this condition (Fig. 28). Alternatively, a regulated and a still unknown
event can direct HIF-1α from the UPS to the lysosome. Importantly, HIF-1αAA does not
117
co-precipitate with LAMP-2, although treatment with MG132 also leads to a marked
increase in HIF-1αAA levels.
Thus, a functional KFERQ motif seems to be required for interaction between
HIF-1α and LAMP-2, supporting the hypothesis of a shuttling of HIF-1α between the
UPS and CMA.
A
B
Figure 28. A mutated HIF-1α protein, with an altered KFERQ-like
domain, is more stable than the wild-type form and does not interacts
with LAMP2 after lysosome or proteasome inhibition, in ARPE-19 cells.
ARPE-19 cells were transiently transfected with a vector containing a HIF1αwt-V5 or with a vector containing HIF-1αAA-V5 and left for ≈16 hours at
37ºC. Cells were then incubated with 200 μM of chloroquine (CQ) or 20 μM
of MG-132 for 8 hours. Control (CT) cells were left untreated. Cells were
harvested in PBS and the pellet has solubilized in Tris-HCl 50 mM, NaCl
150 mM and 0.5% of NP-40. A (Input). 20 μl of whole lysate were
denatured in Laemmli buffer 4X. B (IP: HIF-1α). The remaining sample
was incubated with antibodies against HIF-1α overnight at 4ºC.
Subsequently the lysates were incubated with protein G sepharose beads for
two hours. The beads were washed 3 times with lysis buffer 0.15 % of NP-40
and denatured with Laemmli buffer 4X (IP: HIF-1α).
All samples were separated by SDS-PAGE and transferred to PVDF
membranes. The membranes were probed with specific antibodies against
LAMP2, HIF-1α and β-actin. β-actin was used as a loading control. The
HIF-1αwt-V5 vector was kindly provided by Dr. Thilo Hagen.
118
Immunocytochemistry data shows that HIF-1αAA protein does not form aggregates and
has behaviour similar to the wild type protein (Fig. 29). Indeed, both HIF-1αwt and HIF1αAA are translocated to the nucleus, although there seems to be differences in the
level of expression between the two proteins.
A
B
Figure 29. The mutated HIF-1αAA protein, is translocated to the nucleus,
similar to the wild-type form, in ARPE-19 cells. ARPE-19 cells were
plated in a 24-well plate with glass coverslips and were transiently
transfected with a vector containing a A) HIF-1αwt-V5 or with B) a vector
containing HIF-1αAA-V5 and left for ≈16 hours at 37ºC. Cells were labelled
with anti-V5 antibody and DAPI as a marker for the nuclei. The HIF-1αwtV5 vector was kindly provided by Dr. Thilo Hagen. Images were taken under
magnification of 630x.
119
120
CHAPTER 4
Discussion
121
122
4.1. Overview
Hypoxia–inducible factor 1 (HIF-1) is a transcription factor that mediates the
cellular response to low O2 levels in several physiological contexts (Liu et al. 2007).
HIF-1 also plays a key role in several pathological conditions, including cancer and
ischemic cardiovascular disease (Arjamaa and Nikinmaa, 2006; Melillo, 2004; Rankin
and Giaccia, 2008), thus making the regulation of HIF-1, a major issue in molecular
and cellular biology. HIF-1 is a heterodimer composed of one of three labile α subunits
(-1α, -2α and -3α) and a constitutive β subunit. The regulation of HIF-1α subunit is a
critical determinant of HIF-1 activity in most cell types. This work provides evidences of
degradation of HIF-1α in the lysosome, through CMA, in addition to the wellestablished degradation of this protein by the UPS.
CMA is a selective pathway for protein degradation. Although it was first though
to provide an alternative source of amino acids, CMA also participates in cellular quality
control. Additionally, CMA plays specialized cell-type specific functions, often related to
degradation of specific proteins (Cuervo, 2009). Several types of proteins undergo
proteolysis through CMA, including, among others, enzymes involved in glycolysis (e.g.
GAPDH) (Cuervo et al., 1994), calcium-binding proteins (e.g. annexins) (Cuervo et al.,
2000), α-synuclein (Cuervo et al., 2004), transcription factors (Pax2 and MEF2D)
(Franch et al., 2001; Yang et al., 2009) and unexpectedly a growth factor receptor
(EGFR) (Shen et al., 2009).
4.2. Lyosome inhibitors induce HIF-1α stabilization
In this work, we show that inhibition of the lysosome leads to an increase in
HIF-1α levels, under normoxic conditions, consistent with a recent report from Olmos
and colleagues (Olmos et al., 2009). Stabilization of HIF-1α was observed after
123
treatments with several lysosome inhibitors, used during different times and at different
concentrations (Figs. 16, 17, 19).
These results reinforce the hypothesis that the described increase in HIF-1α
levels is due to inhibition of lysosome function and are not to off-target effects.
Furthermore, different cell lines were used, both human and mouse, suggesting that
the degradation of HIF-1α in the lysosome is indeed a widespread mechanism (Figs.
16, 17, 19, 21).
In most the experiments described in this work, treatments with inhibitors of
lysosome proteolysis for 2-8 hours were sufficient to promote an increase in HIF-1α
levels. However, in RCC4 cells 12 hour treatments were required (Fig. 21). Similarly, in
the report by Olmos et al. 24 hours treatments with chloroquine were used. This
discrepancy probably reflects cell type differences in the activity of the lysosomal
pathway in HIF-1α degradation, as well as the number of lysosomes per cell.
In addition, in the experiments described in this thesis MG132 was used to
inhibit the proteasome, however this compound was shown to inhibit other proteases to
some degree, including cathepsin B (Fuertes et al., 2003; Lee and Goldberg, 1998).
Thus, the contribution of the lysosome to HIF-1α degradation in the cell culture
systems used is likely to be underestimated.
4.3. Intracellular alkalinization might impact HIF-1α levels, independently of
lisosomal-mediated degradation
On the other hand, López-Lazaro has predicted that intracellular alkalinization
can potentially increase the levels of HIF-1, independently of lysosome inhibition
(Lopez-Lazaro, 2006). In fact, treatment with lysosomotropic amines leads to a small
increase in intracellular pH. However, another report suggests that normoxic acidosis
results in nucleolar sequestration of VHL and estabilization HIF-1α (Mekhail et al.,
124
2004). Olmos et al. raised the hypothesis that inhibition of the lysosome could lower
the availability of Fe (II), thus inhibiting the PHD activity and disrupting the VHLdependent degradation of HIF-1α (Olmos et al., 2009).
Previous reports have shown that PHD activity is regulated by the availability of
Fe (II) (Gerald et al., 2004) and treatments with iron chelators and iron antagonists,
such as desferrioxiamine and cobalt chloride, are widely used to mimic hypoxia in cell
culture (Salnikow et al., 2004; Wang & Semenza, 1993; Yuan et al., 2003).
Consistently, in hepatoma cells, chloroquine was reported to inhibit the degragadation
of ferritin in the lysosome, causing more iron to be retained in this protein and leading
to a decrease in the availability of Fe (II) (Kidane et al., 2006).
To assess this hypothesis it would be necessary to measure intracellular Fe (II)
after lysosome inhibition. This hypothesis would predict that lower levels of Fe (II)
would inhibit the hydroxylation of HIF-1α and would subsequently abrogate the VHLmediated degradation of HIF-1α in the UPS. Nevertheless, in VHL-deficient RCC4
cells, treatments with lysosomal inhibitors increased HIF-1α levels (Fig. 21), suggesting
that the putative effects of the treatments on Fe (II) availability are not leading to HIF1α stabilization by impairing its degradation through VHL-UPS pathway.
However, several other factors that lead to activation of HIF-1 under normoxic
conditions have been described, including glucose metabolites and tricarboxylic cycle
(TCA) products (Dalgard et al., 2004; Lu et al., 2005; Lu et al., 2002). Particularly, it
has been proposed that pyruvate and oxaloacetate might prevent HIF-1α degradation
by inhibiting the PHDs (Dalgard et al., 2004). Succinate and fumarate have also been
reported to inhibit PHD-dependent degradation of HIF-1α by directly competing with 2oxoglutarate for the binding site of this co-substrate on the PHD (Isaacs et al., 2005).
Given that glycolysis metabolites can inhibit HIF-1α degradation, López-Lazaro
suggested that any factor capable of activating glycolysis can potentially increase HIF1α protein levels (Lopez-Lazaro, 2006). Because intracellular alkalinization activates
glycolysis (Erecinska et al., 1995; Trivedi and Danforth, 1966), the same hypothesis
125
predicts that even small increases in the intracellular pH could lead to HIF-1α
stabilization. Therefore, an argument can be made that NH4Cl and chloroquine could
be affecting HIF-1α levels by increasing the intracellular pH.
4.4. CMA is likely to be the pathway for lysosomal degradation of HIF-1α
Nevertheless, this work also provides strong evidence that HIF-1α is directed to
the lysosome through CMA. We have shown that HIF-1α has been shown to have a
KFERQ motif (a non-canonical KFERQ in the human protein) (Fig. 26) and that HIF-1α
can interact both with LAMP-2A and Hsc70 (Fig. 25). More importantly, silencing of
CMA receptor LAMP-2A leads to HIF-1α stabilization (Fig. 27).
However compelling, our data has not provide direct evidence of Hsc70 binding
to the KFERQ motif of HIF-1α. Thus, to this point, we cannot discard the possibility that
an ancillary protein, involved in regulation of HIF-1α stability, is the substrate for CMA.
If this is the case, then the inhibition of degradation of this putative protein would lead
indirectly to the accumulation of HIF-1α. In fact, Olmos et al. have identified a KFERQ
motif in HSP90, a chaperone that is required for stabilizing HIF-1α (Isaacs et al., 2002;
Olmos et al., 2009)
The experiments in which the mutant HIF-1α in the KFERQ motif (HIF-1αAA)
was overexpressed do suggest that the HIF-1αAA is indeed more stable than the wild
type form (Fig. 28), further supporting the hypothesis that HIF-1α is indeed a substrate
for CMA. However, because the amino acid substitutions occur within the ODD domain
of HIF-1α, there is the possibility that they could be interfering with VHL binding to HIF1α and impairing its subsequent degradation by the proteasome.
In order to establish HIF-1αAA as a CMA substrate, its ability to be translocated
into isolated lysosomes in an ATP- and Hsc70-dependent manner will have to be
126
assessed, as this procedure provides the definitive experimental evidence to establish
a protein as a CMA substrate (Cuervo, 2009).
4.5. HIF-1 activity was not assessed under lysosome inhibition
One major caveat of this work is the absence of sufficient experimental
evidence to adress HIF-1 transcriptional activity in the presence of lysosome inhibitors.
Immunofluoresce data shows that overexpressed HIF-1αAA is indeed translocated to
the nucleus, even under normoxia (Fig. 29), although we have not further address its
transcriptional activity Presumably, overexpressing high levels of HIF-1α (both mutant
and wild type), would be sufficient for some of to the protein to escape the proteolytic
systems and be translocated to the nucleus (Fig. 29). Hence, the information about
HIF-1α transcriptional activity that can be derived from these experiments is very
limited.
To assess activity of HIF-1α stabilized as a result of inhibition of CMA further
experiments need to be performed. To assess nuclear translocation of the stabilized
HIF-1α, subcellular fractionation experiments could be performed, for the experimental
conditions described earlier, and the levels of the HIF-1α in the nuclear and cytosolic
fractions would be analyzed. Binding of HIF-1α transcription factor to the HREs could
be assed by electrophoretic mobility gel shift assay (EMSA). A gene reporter assays,
using a plasmid containing the luciferase gene under a promoter regulated by a HRE
could be used to assess HIF activity (Catrina et al., 2004). Additionally, the levels of
typical HIF-1 target genes, such as VEGF, could be quantified by real time RT-PCR.
127
4.6. What is the role of HIF-1α degradation through CMA?
The experiments described in this work were performed in normoxic conditions,
suggesting that HIF-1α degradation in the lysosome through CMA occurs under basal
conditions. Moreover, treatment with lysosomal inhibitors led to the accumulation of
HIF-1α in VHL-deficient RCC4 cells, indicating that it is a VHL-independent process.
However, a number of important questions remain unanswered.
For example, does the CMA-mediated degradation of HIF-1α occur under
hypoxic conditions? If so, what is its biological significance? It would be important to
assess this mechanism for degradation under hypoxia, in order to gain further insight
into the physiological relevance in the cellular response to low availability of O2.
Several other mechanisms leading to HIF-1α degradation, other than the
wellestablished O2-dependent
pathway,
have been described. However,
the
physiological relevance of these mechanisms remains largely undetermined in most
cases. In addition to the VHL ubiquitin ligase complex, other molecular players were
shown to be able to direct HIF-1α for proteasomal degradation. The receptor for
activated protein kinase C 1 (RACK1) – E3 ubiquitin ligase complex, the HypoxiaAssociated Factor (HAF, a putative ubiquitin ligase), Copper Metabolism MURR1
Domain containing 1 protein (COMMD1) have all been reported to mediate O2independent proteasomal degradation of HIF-1α, that does not exclude the VHLmediated pathway (Koh et al., 2008; Liu et al., 2007; van de Sluis et al., 2009). The
ubiquitin ligase that degrades p53, MDM2, has also been implicated in HIF-1α
degradation. Its exact role, however, remains controversial (Chen et al., 2003; Hansson
et al., 2002; Ravi et al., 2000).
The significance of these pathways for HIF-1α has not been explored, however
Liu and colleagues raised the possibility that RACK1 activity might contribute for the
basal levels of HIF-1α, which have been shown to vary in vivo between the different
128
tissues and cell types (Stroka et al., 2001). Responses to O2 availability would thus be
superimposed upon this basal rate (Liu et al., 2007).
More recently, data from our laboratory as well as an independent report, have
established that the E3 ubiquitin ligase CHIP is likely to be involved in the ubiquitination
and proteasome degradation of HIF-1α. CHIP seems to mediate HIF-1α degradation in
response to the accumulation of intracellular methylglyoxal (MGO) (Bento et al.,
submitted) and during prolonged hypoxia (Lou et al., 2009). Thus, although it is
tempting to speculate that the alternative mechanisms for HIF-1α might have a
compensatory role, by maintaining the levels of the protein tightly regulated in case of
dysfunction of the primary pathway, these mechanisms are likely to function under
different cellular and environmental contexts.
Besides proteasomal degradation, HIF-1α has been reported to undergo
degradation by the calpains (Zhou et al., 2006). This O2- independent mechanism is
thought to respond to nitric oxide (NO) signalling and calcium fluctuations during
hypoxia, representing an alternative proteolytic pathway to the UPS for HIF-1α
degradation. Again, this system is likely to participate in the regulation of the levels of
HIF-1α in a context-specific manner, not functioning merely as a compensatory system.
In this work we provide evidences of HIF-1α degradation in the lysosome
through CMA. The data further suggests that this is a constitutively operating
mechanism under normoxia. However, the contribution of CMA to HIF-1α proteolysis
appears to be relatively modest when compared with the highly efficient VHL-mediated
degradation oh the protein by the UPS. In their report, Olmos et al. hypothesize that
HIF-1α degradation in the lysosome would help to avoid excessive accumulation of the
protein, when there is a dysfunction in the UPS (Olmos et al., 2009). Indeed, mild
oxidative stress has been shown to lead to a partial inhibition of the proteasome
(Fernandes et al., 2008; Zhang et al., 2008). Decreases in proteasomal activity are
also reported with aging. Additionally, alterations in the UPS have been implicated in
the pathogenesis of many diseases including cancer and neurodegenerative disorders
129
(see Martinez-Vicente et al., 2005 for references). Thus, CMA-mediated degradation of
HIF-1α would function primarily as a compensatory system. A role for this mechanism
in determining the basal levels of HIF-1α in different cell types, similar to what was
proposed for RACK1-mediated degradation of this protein, should also be considered.
In addition, some authors argue that sustained transcription, mediated by
specific transcription factors, requires proteolytic activity to remove “spent” activators,
in order to reset the promoter (Kodadek et al., 2006; Lipford et al., 2005) and in a
number of cases this role has been attributed to the proteasomal system. It is
conceivable that CMA could perform a similar function, although this hypothesis has
not been explored thus far. Indeed, several transcription factors, namely Pax2, MEF2D
an HSF1 have been shown to be degraded by CMA (Franch et al., 2001; Yang et al.,
2008; Yang et al., 2009). In the case of HIF-1, this hypothetical role of CMA in the
regulation of HIF-1α levels would be especially significant under conditions of limited
O2 availability.
This work did not assess the CMA-mediated degradation of HIF-1α under
hypoxia. However, if CMA-dependent degradation of HIF-1α does indeed occur in
these conditions, it could be hypothesized that this mechanism could help to modulate
the cell response to low O2, by preventing the excessive accumulation of HIF-1α,
during prolonged hypoxia. In fact, HIF-1α levels are attenuated during prolonged
hypoxia, in spite of the striking increase in the levels of the protein in response to acute
hypoxia (Uchida et al., 2004). This decrease was shown to be regulated at the level of
HIF-1α mRNA stability. Conversely, Luo et al. suggest in a recent report that the E3
ubiquitin ligase CHIP ubiquitinates and targets HIF-1α for proteasomal degradation
during prolonged hypoxia, thus contributing for the decrease in the levels of this protein
(Luo et al., 2009). Interestingly, results from our laboratory show that CHIP
coimmunoprecipitates both with HIF-1α and LAMP-2, which suggests an interaction
between HIF-1α and the CMA receptor present at the lysosomal membrane (LAMP-2A)
130
and that dominant negative mutants of CHIP attenuate the stabilization of HIF-1α
mediated by lysosome inhibition (data not shown).
These results raise the possibility that CHIP could be involved in an unexpected
crosstalk between CMA and the UPS, by which the two proteolytic pathways are
concurring for HIF-1α degradation. Consistent with this hypothesis, CHIP has been
shown to be able to direct α-synuclein both to lysosomal- and proteasomal-mediated
degradation, in a cell culture model (Shin et al., 2005). However, the lysosomal
pathway involved in this process was not investigated. Additionally, in a recent report,
Arnt et al. showed that ubiquitination of damaged proteins in the Z disks of muscle cells
by CHIP triggers their sorting towards lysososmal degradation. In this case, though, the
target proteins are degraded through macroautophagy (Arndt et al., 2009) .
4.7. HIF-1α as a model substrate in the ellucidation of the molecular basis of the
crosstalk between CMA and the UPS
In the last few years, accumulating lines of evidence suggest that a protein can
be simultaneously degraded by more than one proteolytic system, or that different
pathways compete for protein degradation, depending on specific cell conditions. The
crosstalk between the different proteolytic systems allows for a compensatory
mechanism, in which the failure of one system leads to the up-regulation of the others,
in order to maintain protein homeostasis (Martinez-Vicente et al., 2005). Thus, the
understanding of the crosstalk among the differentent proteolytic systems could impact
our understanding of several protein conformational disorders, which have as a
common feature an impairment of the clearance of normal and modified proteins, such
as neurodegenerative disorders (Bandyopadhyay and Cuervo, 2007).
The failure of the proteolytic systems with aging has been propose as key in
numerous age-related pathologies.
Understanding how the systems balance their
131
activity could be beneficial to envision future therapeutic approaches directed at the
protein quality control (Martinez-Vicente et al., 2005).
The mechanisms driving the crosstalk between UPS and macroautophagy have
begun to be explored on the last few years. Some of these mechanisms were
discussed in Chapter 1 (see also Korolchuk et al., 2010; Lamark and Johansen, 2010
for recent review and references). Additionally, there are several evidences of
intercommunication between macroautophagy and CMA. A striking example is the
sequential activation of the two pathways during the response to nutrient deprivation in
confluent fibroblasts in culture and in the rat liver during starvation (Cuervo et al.,
1995a).
Furthermore,
chronic
blockage
of
CMA
leads
to
upregulation
of
macroautophagy and cells with impaired macroautophagy display constitutive
activation of CMA, supporting a direct crosstalk between these two autophagic
pathways to maintain protein homeostasis (Kaushik et al., 2008; Massey et al., 2008;
Massey et al., 2006a).
On the other hand, there is limited evidence of a direct crosstalk between CMA
and the UPS. Certain subunits of the proteasome are reported to be selectively
degraded by CMA(Cuervo et al., 1995b). In addition, transient blockage of CMA led to
a temporary dysfunction of the proteasome, besides of impacting on macroautophagy
(Massey et al., 2006a). Another, more indirect, evidence for a crosstalk between these
two pathways comes from the finding that several proteins are able to undergo
proteolysis both by CMA and UPS, namely α-synuclein, IkB, RCAN1 and the
membrane receptor EGFR (Cuervo et al., 1998; Cuervo et al., 2004; Liu et al., 2009a;
Shen et al., 2009).
The work described in this thesis supports the hypothesis that HIF-1α, a
prototypical substrate of the UPS, is indeed a dual-pathway substrate. The elucidation
of the mechanisms driving the shuttling of the substrates between CMA and the UPS
could provide new insights on the regulation of the crosstalk between the different
132
proteolytic pathways, in addition to impacting the understanding of the regulation of the
substrates per se.
4.8. Concluding remarks
In summary, the data presented in this thesis strongly suggests that HIF-1α is a
substrate for CMA, a selective pathway of lysosomal degradation. This novel
proteolytic pathway maybe a part of a general mechanism for regulation of the levels of
HIF-1α, as it was found to be active in several types of cells. The degradation of HIF-1α
through CMA is active under normal O2 levels and seems to be independent of VHL.
Although the physiological function of this mechanism was not addressed in this work,
it is tempting to speculate that it may regulate the levels of HIF-1α under hypoxia, in
addition to contribute to HIF-1α degradation in normoxia.
Thus, the results described add further complexity to the already intricate
picture of the regulation of HIF-1α, which is a critical determinant of HIF-1 activity.
133
134
CHAPTER 5
References
135
136
Agarraberes, F.A., and Dice, J.F. (2001a). A molecular chaperone complex at the
lysosomal membrane is required for protein translocation. J Cell Sci 114, 2491-2499.
Agarraberes, F.A., and Dice, J.F. (2001b). Protein translocation across membranes.
Biochim Biophys Acta 1513, 1-24.
Agarraberes, F.A., Terlecky, S.R., and Dice, J.F. (1997). An intralysosomal hsp70 is
required for a selective pathway of lysosomal protein degradation. J Cell Biol 137, 825834.
Ahlberg, J., Berkenstam, A., Henell, F., and Glaumann, H. (1985). Degradation of short
and long lived proteins in isolated rat liver lysosomes. Effects of pH, temperature, and
proteolytic inhibitors. J Biol Chem 260, 5847-5854.
Aniento, F., Papavassiliou, A.G., Knecht, E., and Roche, E. (1996). Selective uptake
and degradation of c-Fos and v-Fos by rat liver lysosomes. FEBS Lett 390, 47-52.
Arjamaa, O., and Nikinmaa, M. (2006). Oxygen-dependent diseases in the retina: role
of hypoxia-inducible factors. Exp Eye Res 83, 473-483.
Artigues, A., Iriarte, A., and Martinez-Carrion, M. (2002). Binding to chaperones allows
import of a purified mitochondrial precursor into mitochondria. J Biol Chem 277, 2504725055.
Ashford, T.P., and Porter, K.R. (1962). Cytoplasmic components in hepatic cell
lysosomes. J Cell Biol 12, 198-202.
Backer, J.M., Bourret, L., and Dice, J.F. (1983). Regulation of catabolism of
microinjected ribonuclease A requires the amino-terminal 20 amino acids. Proc Natl
Acad Sci U S A 80, 2166-2170.
Bandyopadhyay, U., and Cuervo, A.M. (2007). Chaperone-mediated autophagy in
aging and neurodegeneration: lessons from alpha-synuclein. Exp Gerontol 42, 120128.
Bandyopadhyay, U., Kaushik, S., Varticovski, L., and Cuervo, A.M. (2008). The
chaperone-mediated autophagy receptor organizes in dynamic protein complexes at
the lysosomal membrane. Mol Cell Biol 28, 5747-5763.
Bellu, A.R., Komori, M., van der Klei, I.J., Kiel, J.A., and Veenhuis, M. (2001).
Peroxisome biogenesis and selective degradation converge at Pex14p. J Biol Chem
276, 44570-44574.
Bellu, A.R., Salomons, F.A., Kiel, J.A., Veenhuis, M., and Van Der Klei, I.J. (2002).
Removal of Pex3p is an important initial stage in selective peroxisome degradation in
Hansenula polymorpha. J Biol Chem 277, 42875-42880.
Berg, T.O., Fengsrud, M., Stromhaug, P.E., Berg, T., and Seglen, P.O. (1998).
Isolation and characterization of rat liver amphisomes. Evidence for fusion of
autophagosomes with both early and late endosomes. J Biol Chem 273, 21883-21892.
137
Bergamini, E. (2006). Autophagy: a cell repair mechanism that retards ageing and ageassociated diseases and can be intensified pharmacologically. Mol Aspects Med 27,
403-410.
Bernales, S., Papa, F.R., and Walter, P. (2006). Intracellular signaling by the unfolded
protein response. Annu Rev Cell Dev Biol 22, 487-508.
Bjorkoy, G., Lamark, T., Brech, A., Outzen, H., Perander, M., Overvatn, A., Stenmark,
H., and Johansen, T. (2005). p62/SQSTM1 forms protein aggregates degraded by
autophagy and has a protective effect on huntingtin-induced cell death. J Cell Biol 171,
603-614.
Bormann, C., and Sahm, H. (1978). Degradation of microbodies in relation to activities
of alcohol oxidase and catalase in Candida boidinii. Arch Microbiol 117, 67-72.
Botusan, I.R., Sunkari, V.G., Savu, O., Catrina, A.I., Grunler, J., Lindberg, S., Pereira,
T., Yla-Herttuala, S., Poellinger, L., Brismar, K., et al. (2008). Stabilization of HIF1alpha is critical to improve wound healing in diabetic mice. Proc Natl Acad Sci U S A
105, 19426-19431.
Brodsky, J.L., Goeckeler, J., and Schekman, R. (1995). BiP and Sec63p are required
for both co- and posttranslational protein translocation into the yeast endoplasmic
reticulum. Proc Natl Acad Sci U S A 92, 9643-9646.
Camougrand, N., Kissova, I., Velours, G., and Manon, S. (2004). Uth1p: a yeast
mitochondrial protein at the crossroads of stress, degradation and cell death. FEMS
Yeast Res 5, 133-140.
Chan, E.Y., Kir, S., and Tooze, S.A. (2007). siRNA screening of the kinome identifies
ULK1 as a multidomain modulator of autophagy. J Biol Chem 282, 25464-25474.
Chen, D., Li, M., Luo, J., and Gu, W. (2003). Direct interactions between HIF-1 alpha
and Mdm2 modulate p53 function. J Biol Chem 278, 13595-13598.
Chen, X., and Schnell, D.J. (1999). Protein import into chloroplasts. Trends Cell Biol 9,
222-227.
Chiang, H.L., and Dice, J.F. (1988). Peptide sequences that target proteins for
enhanced degradation during serum withdrawal. J Biol Chem 263, 6797-6805.
Chiang, H.L., Terlecky, S.R., Plant, C.P., and Dice, J.F. (1989). A role for a 70kilodalton heat shock protein in lysosomal degradation of intracellular proteins. Science
246, 382-385.
Chowdhury, I., Tharakan, B., and Bhat, G.K. (2008). Caspases - an update. Comp
Biochem Physiol B Biochem Mol Biol 151, 10-27.
Ciechanover, A. (2005a). Intracellular protein degradation: from a vague idea thru the
lysosome and the ubiquitin-proteasome system and onto human diseases and drug
targeting. Cell Death Differ 12, 1178-1190.
138
Ciechanover, A. (2005b). Proteolysis: from the lysosome to ubiquitin and the
proteasome. Nat Rev Mol Cell Biol 6, 79-87.
Ciechanover, A., Heller, H., Elias, S., Haas, A.L., and Hershko, A. (1980). ATPdependent conjugation of reticulocyte proteins with the polypeptide required for protein
degradation. Proc Natl Acad Sci U S A 77, 1365-1368.
Ciechanover, A., Hod, Y., and Hershko, A. (1978). A heat-stable polypeptide
component of an ATP-dependent proteolytic system from reticulocytes. Biochem
Biophys Res Commun 81, 1100-1105.
Clark, S.L., Jr. (1957). Cellular differentiation in the kidneys of newborn mice studies
with the electron microscope. J Biophys Biochem Cytol 3, 349-362.
Croall, D.E., and Ersfeld, K. (2007). The calpains: modular designs and functional
diversity. Genome Biol 8, 218.
Cuervo, A.M. (2004a). Autophagy: in sickness and in health. Trends Cell Biol 14, 7077.
Cuervo, A.M. (2004b). Autophagy: many paths to the same end. Mol Cell Biochem 263,
55-72.
Cuervo, A.M. (2008). Autophagy and aging: keeping that old broom working. Trends
Genet 24, 604-612.
Cuervo, A.M. (2009). Chaperone-mediated autophagy: selectivity pays off. Trends
Endocrinol Metab 21, 142-150.
Cuervo, A.M., and Dice, J.F. (1996). A receptor for the selective uptake and
degradation of proteins by lysosomes. Science 273, 501-503.
Cuervo, A.M., and Dice, J.F. (1998). Lysosomes, a meeting point of proteins,
chaperones, and proteases. J Mol Med 76, 6-12.
Cuervo, A.M., and Dice, J.F. (2000a). Age-related decline in chaperone-mediated
autophagy. J Biol Chem 275, 31505-31513.
Cuervo, A.M., and Dice, J.F. (2000b). Regulation of lamp2a levels in the lysosomal
membrane. Traffic 1, 570-583.
Cuervo, A.M., and Dice, J.F. (2000c). Unique properties of lamp2a compared to other
lamp2 isoforms. J Cell Sci 113 Pt 24, 4441-4450.
Cuervo, A.M., Dice, J.F., and Knecht, E. (1997). A population of rat liver lysosomes
responsible for the selective uptake and degradation of cytosolic proteins. J Biol Chem
272, 5606-5615.
Cuervo, A.M., Gomes, A.V., Barnes, J.A., and Dice, J.F. (2000). Selective degradation
of annexins by chaperone-mediated autophagy. J Biol Chem 275, 33329-33335.
139
Cuervo, A.M., Hildebrand, H., Bomhard, E.M., and Dice, J.F. (1999). Direct lysosomal
uptake of alpha 2-microglobulin contributes to chemically induced nephropathy. Kidney
Int 55, 529-545.
Cuervo, A.M., Hu, W., Lim, B., and Dice, J.F. (1998). IkappaB is a substrate for a
selective pathway of lysosomal proteolysis. Mol Biol Cell 9, 1995-2010.
Cuervo, A.M., Mann, L., Bonten, E.J., d'Azzo, A., and Dice, J.F. (2003). Cathepsin A
regulates chaperone-mediated autophagy through cleavage of the lysosomal receptor.
EMBO J 22, 47-59.
Cuervo, A.M., Palmer, A., Rivett, A.J., and Knecht, E. (1995). Degradation of
proteasomes by lysosomes in rat liver. Eur J Biochem 227, 792-800.
Cuervo, A.M., Stefanis, L., Fredenburg, R., Lansbury, P.T., and Sulzer, D. (2004).
Impaired degradation of mutant alpha-synuclein by chaperone-mediated autophagy.
Science 305, 1292-1295.
Cuervo, A.M., Terlecky, S.R., Dice, J.F., and Knecht, E. (1994). Selective binding and
uptake of ribonuclease A and glyceraldehyde-3-phosphate dehydrogenase by isolated
rat liver lysosomes. J Biol Chem 269, 26374-26380.
de Duve, C. (1983). Lysosomes revisited. Eur J Biochem 137, 391-397.
De Duve, C., Pressman, B.C., Gianetto, R., Wattiaux, R., and Appelmans, F. (1955).
Tissue fractionation studies. 6. Intracellular distribution patterns of enzymes in rat-liver
tissue. Biochem J 60, 604-617.
De Duve, C., and Wattiaux, R. (1966). Functions of lysosomes. Annu Rev Physiol 28,
435-492.
Demand, J., Luders, J., and Hohfeld, J. (1998). The carboxy-terminal domain of Hsc70
provides binding sites for a distinct set of chaperone cofactors. Mol Cell Biol 18, 20232028.
Dice, J.F. (2007). Chaperone-mediated autophagy. Autophagy 3, 295-299.
Dice, J.F., Chiang, H.L., Spencer, E.P., and Backer, J.M. (1986). Regulation of
catabolism of microinjected ribonuclease A. Identification of residues 7-11 as the
essential pentapeptide. J Biol Chem 261, 6853-6859.
Ding, Q., Dimayuga, E., Martin, S., Bruce-Keller, A.J., Nukala, V., Cuervo, A.M., and
Keller, J.N. (2003). Characterization of chronic low-level proteasome inhibition on
neural homeostasis. J Neurochem 86, 489-497.
Dubouloz, F., Deloche, O., Wanke, V., Cameroni, E., and De Virgilio, C. (2005). The
TOR and EGO protein complexes orchestrate microautophagy in yeast. Mol Cell 19,
15-26.
Eskelinen, E.L. (2005). Maturation of autophagic vacuoles in Mammalian cells.
Autophagy 1, 1-10.
140
Eskelinen, E.L., Cuervo, A.M., Taylor, M.R., Nishino, I., Blum, J.S., Dice, J.F.,
Sandoval, I.V., Lippincott-Schwartz, J., August, J.T., and Saftig, P. (2005). Unifying
nomenclature for the isoforms of the lysosomal membrane protein LAMP-2. Traffic 6,
1058-1061.
Eskelinen, E.L., and Saftig, P. (2009). Autophagy: a lysosomal degradation pathway
with a central role in health and disease. Biochim Biophys Acta 1793, 664-673.
Eskelinen, E.L., Schmidt, C.K., Neu, S., Willenborg, M., Fuertes, G., Salvador, N.,
Tanaka, Y., Lullmann-Rauch, R., Hartmann, D., Heeren, J., et al. (2004). Disturbed
cholesterol traffic but normal proteolytic function in LAMP-1/LAMP-2 double-deficient
fibroblasts. Mol Biol Cell 15, 3132-3145.
Etlinger, J.D., and Goldberg, A.L. (1977). A soluble ATP-dependent proteolytic system
responsible for the degradation of abnormal proteins in reticulocytes. Proc Natl Acad
Sci U S A 74, 54-58.
Fang, S., Jensen, J.P., Ludwig, R.L., Vousden, K.H., and Weissman, A.M. (2000).
Mdm2 is a RING finger-dependent ubiquitin protein ligase for itself and p53. J Biol
Chem 275, 8945-8951.
Farre, J.C., Manjithaya, R., Mathewson, R.D., and Subramani, S. (2008). PpAtg30 tags
peroxisomes for turnover by selective autophagy. Dev Cell 14, 365-376.
Farre, J.C., Vidal, J., and Subramani, S. (2007). A cytoplasm to vacuole targeting
pathway in P. pastoris. Autophagy 3, 230-234.
Fass, E., Shvets, E., Degani, I., Hirschberg, K., and Elazar, Z. (2006). Microtubules
support production of starvation-induced autophagosomes but not their targeting and
fusion with lysosomes. J Biol Chem 281, 36303-36316.
Fimia, G.M., Stoykova, A., Romagnoli, A., Giunta, L., Di Bartolomeo, S., Nardacci, R.,
Corazzari, M., Fuoco, C., Ucar, A., Schwartz, P., et al. (2007). Ambra1 regulates
autophagy and development of the nervous system. Nature 447, 1121-1125.
Finn, P.F., and Dice, J.F. (2005). Ketone bodies stimulate chaperone-mediated
autophagy. J Biol Chem 280, 25864-25870.
Finn, P.F., Mesires, N.T., Vine, M., and Dice, J.F. (2005). Effects of small molecules on
chaperone-mediated autophagy. Autophagy 1, 141-145.
Franch, H.A., Sooparb, S., Du, J., and Brown, N.S. (2001). A mechanism regulating
proteolysis of specific proteins during renal tubular cell growth. J Biol Chem 276,
19126-19131.
Franch,
H.A., Wang,
X.,
Sooparb,
S.,
Brown,
N.S.,
and
Du,
J.
(2002).
Phosphatidylinositol 3-kinase activity is required for epidermal growth factor to
suppress proteolysis. J Am Soc Nephrol 13, 903-909.
141
Fuertes, G., Martin De Llano, J.J., Villarroya, A., Rivett, A.J., and Knecht, E. (2003a).
Changes in the proteolytic activities of proteasomes and lysosomes in human
fibroblasts produced by serum withdrawal, amino-acid deprivation and confluent
conditions. Biochem J 375, 75-86.
Fuertes, G., Villarroya, A., and Knecht, E. (2003b). Role of proteasomes in the
degradation of short-lived proteins in human fibroblasts under various growth
conditions. Int J Biochem Cell Biol 35, 651-664.
Fujita, N., Itoh, T., Omori, H., Fukuda, M., Noda, T., and Yoshimori, T. (2008). The
Atg16L complex specifies the site of LC3 lipidation for membrane biogenesis in
autophagy. Mol Biol Cell 19, 2092-2100.
Galban, S., and Gorospe, M. (2009). Factors interacting with HIF-1alpha mRNA: novel
therapeutic targets. Curr Pharm Des 15, 3853-3860.
Geetha, T., and Wooten, M.W. (2002). Structure and functional properties of the
ubiquitin binding protein p62. FEBS Lett 512, 19-24.
Geng, J., Baba, M., Nair, U., and Klionsky, D.J. (2008). Quantitative analysis of
autophagy-related protein stoichiometry by fluorescence microscopy. J Cell Biol 182,
129-140.
Geng, J., and Klionsky, D.J. (2008). The Atg8 and Atg12 ubiquitin-like conjugation
systems in macroautophagy. 'Protein modifications: beyond the usual suspects' review
series. EMBO Rep 9, 859-864.
Gianetto, R., and De Duve, C. (1955). Tissue fractionation studies. 4. Comparative
study of the binding of acid phosphatase, beta-glucuronidase and cathepsin by rat-liver
particles. Biochem J 59, 433-438.
Glickman, M.H., and Ciechanover, A. (2002). The ubiquitin-proteasome proteolytic
pathway: destruction for the sake of construction. Physiol Rev 82, 373-428.
Goldberg, A.L. (2003). Protein degradation and protection against misfolded or
damaged proteins. Nature 426, 895-899.
Goldberg, A.L., and Dice, J.F. (1974). Intracellular protein degradation in mammalian
and bacterial cells. Annu Rev Biochem 43, 835-869.
Gough, N.R., and Fambrough, D.M. (1997). Different steady state subcellular
distributions of the three splice variants of lysosome-associated membrane protein
LAMP-2 are determined largely by the COOH-terminal amino acid residue. J Cell Biol
137, 1161-1169.
Gracy, R.W., Talent, J.M., and Zvaigzne, A.I. (1998). Molecular wear and tear leads to
terminal marking and the unstable isoforms of aging. J Exp Zool 282, 18-27.
142
Grune, T., Merker, K., Sandig, G., and Davies, K.J. (2003). Selective degradation of
oxidatively modified protein substrates by the proteasome. Biochem Biophys Res
Commun 305, 709-718.
Hampton, R.Y., and Garza, R.M. (2009). Protein quality control as a strategy for
cellular regulation: lessons from ubiquitin-mediated regulation of the sterol pathway.
Chem Rev 109, 1561-1574.
Hanada, T., Noda, N.N., Satomi, Y., Ichimura, Y., Fujioka, Y., Takao, T., Inagaki, F.,
and Ohsumi, Y. (2007). The Atg12-Atg5 conjugate has a novel E3-like activity for
protein lipidation in autophagy. J Biol Chem 282, 37298-37302.
Hansson, L.O., Friedler, A., Freund, S., Rudiger, S., and Fersht, A.R. (2002). Two
sequence motifs from HIF-1alpha bind to the DNA-binding site of p53. Proc Natl Acad
Sci U S A 99, 10305-10309.
Hara, T., Nakamura, K., Matsui, M., Yamamoto, A., Nakahara, Y., Suzuki-Migishima,
R., Yokoyama, M., Mishima, K., Saito, I., Okano, H., et al. (2006). Suppression of basal
autophagy in neural cells causes neurodegenerative disease in mice. Nature 441, 885889.
Hershko, A., Ciechanover, A., Heller, H., Haas, A.L., and Rose, I.A. (1980). Proposed
role of ATP in protein breakdown: conjugation of protein with multiple chains of the
polypeptide of ATP-dependent proteolysis. Proc Natl Acad Sci U S A 77, 1783-1786.
Hershko, A., Heller, H., Elias, S., and Ciechanover, A. (1983). Components of ubiquitinprotein ligase system. Resolution, affinity purification, and role in protein breakdown. J
Biol Chem 258, 8206-8214.
Hogness, D.S., Cohn, M., and Monod, J. (1955). Studies on the induced synthesis of
beta-galactosidase in Escherichia coli: the kinetics and mechanism of sulfur
incorporation. Biochim Biophys Acta 16, 99-116.
Hohfeld, J., Minami, Y., and Hartl, F.U. (1995). Hip, a novel cochaperone involved in
the eukaryotic Hsc70/Hsp40 reaction cycle. Cell 83, 589-598.
Hough, R., Pratt, G., and Rechsteiner, M. (1986). Ubiquitin-lysozyme conjugates.
Identification and characterization of an ATP-dependent protease from rabbit
reticulocyte lysates. J Biol Chem 261, 2400-2408.
Huang, L.E., Arany, Z., Livingston, D.M., and Bunn, H.F. (1996). Activation of hypoxiainducible transcription factor depends primarily upon redox-sensitive stabilization of its
alpha subunit. J Biol Chem 271, 32253-32259.
Huang, W.P., and Klionsky, D.J. (2002). Autophagy in yeast: a review of the molecular
machinery. Cell Struct Funct 27, 409-420.
143
Hutson, N.J., and Mortimore, G.E. (1982). Suppression of cytoplasmic protein uptake
by lysosomes as the mechanism of protein regain in livers of starved-refed mice. J Biol
Chem 257, 9548-9554.
Huynh, K.K., Eskelinen, E.L., Scott, C.C., Malevanets, A., Saftig, P., and Grinstein, S.
(2007). LAMP proteins are required for fusion of lysosomes with phagosomes. EMBO J
26, 313-324.
Ibrahim, N.O., Hahn, T., Franke, C., Stiehl, D.P., Wirthner, R., Wenger, R.H., and
Katschinski, D.M. (2005). Induction of the hypoxia-inducible factor system by low levels
of heat shock protein 90 inhibitors. Cancer Res 65, 11094-11100.
Ichimura, Y., Kirisako, T., Takao, T., Satomi, Y., Shimonishi, Y., Ishihara, N.,
Mizushima, N., Tanida, I., Kominami, E., Ohsumi, M., et al. (2000). A ubiquitin-like
system mediates protein lipidation. Nature 408, 488-492.
Imai, J., Yashiroda, H., Maruya, M., Yahara, I., and Tanaka, K. (2003). Proteasomes
and molecular chaperones: cellular machinery responsible for folding and destruction
of unfolded proteins. Cell Cycle 2, 585-590.
Isaacs, J.S., Jung, Y.J., Mimnaugh, E.G., Martinez, A., Cuttitta, F., and Neckers, L.M.
(2002). Hsp90 regulates a von Hippel Lindau-independent hypoxia-inducible factor-1
alpha-degradative pathway. J Biol Chem 277, 29936-29944.
Ivan, M., Kondo, K., Yang, H., Kim, W., Valiando, J., Ohh, M., Salic, A., Asara, J.M.,
Lane, W.S., and Kaelin, W.G., Jr. (2001). HIFalpha targeted for VHL-mediated
destruction by proline hydroxylation: implications for O2 sensing. Science 292, 464468.
Iwata, A., Christianson, J.C., Bucci, M., Ellerby, L.M., Nukina, N., Forno, L.S., and
Kopito,
R.R.
(2005a).
Increased
susceptibility
of
cytoplasmic
over
nuclear
polyglutamine aggregates to autophagic degradation. Proc Natl Acad Sci U S A 102,
13135-13140.
Iwata, A., Riley, B.E., Johnston, J.A., and Kopito, R.R. (2005b). HDAC6 and
microtubules are required for autophagic degradation of aggregated huntingtin. J Biol
Chem 280, 40282-40292.
Jaakkola, P., Mole, D.R., Tian, Y.M., Wilson, M.I., Gielbert, J., Gaskell, S.J.,
Kriegsheim, A., Hebestreit, H.F., Mukherji, M., Schofield, C.J., et al. (2001). Targeting
of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl
hydroxylation. Science 292, 468-472.
Jadot, M., Dubois, F., Wattiaux-De Coninck, S., and Wattiaux, R. (1997).
Supramolecular assemblies from lysosomal matrix proteins and complex lipids. Eur J
Biochem 249, 862-869.
144
Jiang, B.H., Semenza, G.L., Bauer, C., and Marti, H.H. (1996). Hypoxia-inducible factor
1 levels vary exponentially over a physiologically relevant range of O2 tension. Am J
Physiol 271, C1172-1180.
Kamada, Y., Funakoshi, T., Shintani, T., Nagano, K., Ohsumi, M., and Ohsumi, Y.
(2000). Tor-mediated induction of autophagy via an Apg1 protein kinase complex. J
Cell Biol 150, 1507-1513.
Kamimoto, T., Shoji, S., Hidvegi, T., Mizushima, N., Umebayashi, K., Perlmutter, D.H.,
and Yoshimori, T. (2006). Intracellular inclusions containing mutant alpha1-antitrypsin Z
are propagated in the absence of autophagic activity. J Biol Chem 281, 4467-4476.
Kanazawa, T., Taneike, I., Akaishi, R., Yoshizawa, F., Furuya, N., Fujimura, S., and
Kadowaki, M. (2004). Amino acids and insulin control autophagic proteolysis through
different signaling pathways in relation to mTOR in isolated rat hepatocytes. J Biol
Chem 279, 8452-8459.
Kaushik, S., and Cuervo, A.M. (2006). Autophagy as a cell-repair mechanism:
activation of chaperone-mediated autophagy during oxidative stress. Mol Aspects Med
27, 444-454.
Kaushik, S., and Cuervo, A.M. (2009). Methods to monitor chaperone-mediated
autophagy. Methods Enzymol 452, 297-324.
Kaushik, S., Kiffin, R., and Cuervo, A.M. (2007). Chaperone-mediated autophagy and
aging: a novel regulatory role of lipids revealed. Autophagy 3, 387-389.
Kaushik, S., Massey, A.C., and Cuervo, A.M. (2006). Lysosome membrane lipid
microdomains: novel regulators of chaperone-mediated autophagy. EMBO J 25, 39213933.
Kaushik, S., Massey, A.C., Mizushima, N., and Cuervo, A.M. (2008). Constitutive
activation of chaperone-mediated autophagy in cells with impaired macroautophagy.
Mol Biol Cell 19, 2179-2192.
Kawaguchi, Y., Kovacs, J.J., McLaurin, A., Vance, J.M., Ito, A., and Yao, T.P. (2003).
The deacetylase HDAC6 regulates aggresome formation and cell viability in response
to misfolded protein stress. Cell 115, 727-738.
Kiffin, R., Christian, C., Knecht, E., and Cuervo, A.M. (2004). Activation of chaperonemediated autophagy during oxidative stress. Mol Biol Cell 15, 4829-4840.
Kihara,
A.,
Kabeya,
Y.,
Ohsumi,
Y.,
and
Yoshimori,
T.
(2001a).
Beclin-
phosphatidylinositol 3-kinase complex functions at the trans-Golgi network. EMBO Rep
2, 330-335.
Kihara, A., Noda, T., Ishihara, N., and Ohsumi, Y. (2001b). Two distinct Vps34
phosphatidylinositol 3-kinase complexes function in autophagy and carboxypeptidase
Y sorting in Saccharomyces cerevisiae. J Cell Biol 152, 519-530.
145
Kim, J., Huang, W.P., Stromhaug, P.E., and Klionsky, D.J. (2002). Convergence of
multiple autophagy and cytoplasm to vacuole targeting components to a perivacuolar
membrane compartment prior to de novo vesicle formation. J Biol Chem 277, 763-773.
Kim, P.K., Hailey, D.W., Mullen, R.T., and Lippincott-Schwartz, J. (2008). Ubiquitin
signals autophagic degradation of cytosolic proteins and peroxisomes. Proc Natl Acad
Sci U S A 105, 20567-20574.
Kimura, H., Weisz, A., Ogura, T., Hitomi, Y., Kurashima, Y., Hashimoto, K., D'Acquisto,
F., Makuuchi, M., and Esumi, H. (2001). Identification of hypoxia-inducible factor 1
ancillary sequence and its function in vascular endothelial growth factor gene induction
by hypoxia and nitric oxide. J Biol Chem 276, 2292-2298.
Kirisako, T., Ichimura, Y., Okada, H., Kabeya, Y., Mizushima, N., Yoshimori, T.,
Ohsumi, M., Takao, T., Noda, T., and Ohsumi, Y. (2000). The reversible modification
regulates the membrane-binding state of Apg8/Aut7 essential for autophagy and the
cytoplasm to vacuole targeting pathway. J Cell Biol 151, 263-276.
Kirkin, V., McEwan, D.G., Novak, I., and Dikic, I. (2009). A role for ubiquitin in selective
autophagy. Mol Cell 34, 259-269.
Kirschner, M. (1999). Intracellular proteolysis. Trends Cell Biol 9, M42-45.
Kissova, I., Salin, B., Schaeffer, J., Bhatia, S., Manon, S., and Camougrand, N. (2007).
Selective and non-selective autophagic degradation of mitochondria in yeast.
Autophagy 3, 329-336.
Klionsky, D.J. (2007). Autophagy: from phenomenology to molecular understanding in
less than a decade. Nat Rev Mol Cell Biol 8, 931-937.
Klionsky, D.J., Cregg, J.M., Dunn, W.A., Jr., Emr, S.D., Sakai, Y., Sandoval, I.V.,
Sibirny, A., Subramani, S., Thumm, M., Veenhuis, M., et al. (2003). A unified
nomenclature for yeast autophagy-related genes. Dev Cell 5, 539-545.
Klionsky, D.J., Cuervo, A.M., Dunn, W.A., Jr., Levine, B., van der Klei, I., and Seglen,
P.O. (2007a). How shall I eat thee? Autophagy 3, 413-416.
Klionsky, D.J., Cuervo, A.M., and Seglen, P.O. (2007b). Methods for monitoring
autophagy from yeast to human. Autophagy 3, 181-206.
Koh, M.Y., Darnay, B.G., and Powis, G. (2008). Hypoxia-associated factor, a novel E3ubiquitin ligase, binds and ubiquitinates hypoxia-inducible factor 1alpha, leading to its
oxygen-independent degradation. Mol Cell Biol 28, 7081-7095.
Komatsu, M., Tanida, I., Ueno, T., Ohsumi, M., Ohsumi, Y., and Kominami, E. (2001).
The C-terminal region of an Apg7p/Cvt2p is required for homodimerization and is
essential for its E1 activity and E1-E2 complex formation. J Biol Chem 276, 9846-9854.
146
Komatsu, M., Waguri, S., Chiba, T., Murata, S., Iwata, J., Tanida, I., Ueno, T., Koike,
M., Uchiyama, Y., Kominami, E., et al. (2006). Loss of autophagy in the central nervous
system causes neurodegeneration in mice. Nature 441, 880-884.
Komatsu, M., Waguri, S., Koike, M., Sou, Y.S., Ueno, T., Hara, T., Mizushima, N.,
Iwata, J., Ezaki, J., Murata, S., et al. (2007). Homeostatic levels of p62 control
cytoplasmic inclusion body formation in autophagy-deficient mice. Cell 131, 1149-1163.
Komatsu, M., Waguri, S., Ueno, T., Iwata, J., Murata, S., Tanida, I., Ezaki, J.,
Mizushima, N., Ohsumi, Y., Uchiyama, Y., et al. (2005). Impairment of starvationinduced and constitutive autophagy in Atg7-deficient mice. J Cell Biol 169, 425-434.
Kopito, R.R. (2000). Aggresomes, inclusion bodies and protein aggregation. Trends
Cell Biol 10, 524-530.
Kopito, R.R. (2003). The missing linker: an unexpected role for a histone deacetylase.
Mol Cell 12, 1349-1351.
Koppen, M., and Langer, T. (2007). Protein degradation within mitochondria: versatile
activities of AAA proteases and other peptidases. Crit Rev Biochem Mol Biol 42, 221242.
Kraft, C., Deplazes, A., Sohrmann, M., and Peter, M. (2008). Mature ribosomes are
selectively degraded upon starvation by an autophagy pathway requiring the
Ubp3p/Bre5p ubiquitin protease. Nat Cell Biol 10, 602-610.
Kuma, A., Hatano, M., Matsui, M., Yamamoto, A., Nakaya, H., Yoshimori, T., Ohsumi,
Y., Tokuhisa, T., and Mizushima, N. (2004). The role of autophagy during the early
neonatal starvation period. Nature 432, 1032-1036.
Kuma, A., Mizushima, N., Ishihara, N., and Ohsumi, Y. (2002). Formation of the
approximately 350-kDa Apg12-Apg5.Apg16 multimeric complex, mediated by Apg16
oligomerization, is essential for autophagy in yeast. J Biol Chem 277, 18619-18625.
Kundu, M., and Thompson, C.B. (2008). Autophagy: basic principles and relevance to
disease. Annu Rev Pathol 3, 427-455.
Lamark, T., and Johansen, T. (2010). Autophagy: links with the proteasome. Curr Opin
Cell Biol 22, 192-198.
Laughner, E., Taghavi, P., Chiles, K., Mahon, P.C., and Semenza, G.L. (2001). HER2
(neu) signaling increases the rate of hypoxia-inducible factor 1alpha (HIF-1alpha)
synthesis: novel mechanism for HIF-1-mediated vascular endothelial growth factor
expression. Mol Cell Biol 21, 3995-4004.
Lazdunski, C.J., and Benedetti, H. (1990). Insertion and translocation of proteins into
and through membranes. FEBS Lett 268, 408-414.
Lee, D.H., and Goldberg, A.L. (1998). Proteasome inhibitors: valuable new tools for cell
biologists. Trends Cell Biol 8, 397-403.
147
Levine, B., and Kroemer, G. (2008). Autophagy in the pathogenesis of disease. Cell
132, 27-42.
Li, D. (2006). Selective degradation of the IkappaB kinase (IKK) by autophagy. Cell
Res 16, 855-856.
Li, Y.M., Zhou, B.P., Deng, J., Pan, Y., Hay, N., and Hung, M.C. (2005). A hypoxiaindependent
hypoxia-inducible
factor-1
activation
pathway
induced
by
phosphatidylinositol-3 kinase/Akt in HER2 overexpressing cells. Cancer Res 65, 32573263.
Liang, C., Feng, P., Ku, B., Dotan, I., Canaani, D., Oh, B.H., and Jung, J.U. (2006).
Autophagic and tumour suppressor activity of a novel Beclin1-binding protein UVRAG.
Nat Cell Biol 8, 688-699.
Liu, H., Wang, P., Song, W., and Sun, X. (2009a). Degradation of regulator of
calcineurin 1 (RCAN1) is mediated by both chaperone-mediated autophagy and
ubiquitin proteasome pathways. FASEB J 23, 3383-3392.
Liu, H., Wang, P., Song, W., and Sun, X. (2009b). Degradation of regulator of
calcineurin 1 (RCAN1) is mediated by both chaperone-mediated autophagy and
ubiquitin proteasome pathways. FASEB J.
Liu, Y.V., Baek, J.H., Zhang, H., Diez, R., Cole, R.N., and Semenza, G.L. (2007).
RACK1 competes with HSP90 for binding to HIF-1alpha and is required for O(2)independent and HSP90 inhibitor-induced degradation of HIF-1alpha. Mol Cell 25, 207217.
Luders, J., Demand, J., Papp, O., and Hohfeld, J. (2000). Distinct isoforms of the
cofactor BAG-1 differentially affect Hsc70 chaperone function. J Biol Chem 275,
14817-14823.
Lum, J.J., DeBerardinis, R.J., and Thompson, C.B. (2005). Autophagy in metazoans:
cell survival in the land of plenty. Nat Rev Mol Cell Biol 6, 439-448.
Luo, S., Chen, Q., Cebollero, E., and Xing, D. (2009a). Mitochondria: one of the origins
for autophagosomal membranes? Mitochondrion 9, 227-231.
Luo, W., Zhong, J., Chang, R., Hu, H., Pandey, A., and Semenza, G.L. (2009b). Hsp70
and CHIP selectively mediate ubiquitination and degradation of hypoxia-inducible
factor (HIF)-1alpha but Not HIF-2alpha. J Biol Chem 285, 3651-3663.
Maiuri, M.C., Le Toumelin, G., Criollo, A., Rain, J.C., Gautier, F., Juin, P., Tasdemir, E.,
Pierron, G., Troulinaki, K., Tavernarakis, N., et al. (2007a). Functional and physical
interaction between Bcl-X(L) and a BH3-like domain in Beclin-1. EMBO J 26, 25272539.
Maiuri, M.C., Zalckvar, E., Kimchi, A., and Kroemer, G. (2007b). Self-eating and selfkilling: crosstalk between autophagy and apoptosis. Nat Rev Mol Cell Biol 8, 741-752.
148
Majeski, A.E., and Dice, J.F. (2004). Mechanisms of chaperone-mediated autophagy.
Int J Biochem Cell Biol 36, 2435-2444.
Makino, Y., Cao, R., Svensson, K., Bertilsson, G., Asman, M., Tanaka, H., Cao, Y.,
Berkenstam, A., and Poellinger, L. (2001). Inhibitory PAS domain protein is a negative
regulator of hypoxia-inducible gene expression. Nature 414, 550-554.
Mandelstam, J. (1958). Turnover of protein in growing and non-growing populations of
Escherichia coli. Biochem J 69, 110-119.
Martinez-Vicente, M., Talloczy, Z., Kaushik, S., Massey, A.C., Mazzulli, J., Mosharov,
E.V., Hodara, R., Fredenburg, R., Wu, D.C., Follenzi, A., et al. (2008). Dopaminemodified alpha-synuclein blocks chaperone-mediated autophagy. J Clin Invest 118,
777-788.
Marzella, L., Ahlberg, J., and Glaumann, H. (1981). Autophagy, heterophagy,
microautophagy and crinophagy as the means for intracellular degradation. Virchows
Arch B Cell Pathol Incl Mol Pathol 36, 219-234.
Massey, A., Kiffin, R., and Cuervo, A.M. (2004). Pathophysiology of chaperonemediated autophagy. Int J Biochem Cell Biol 36, 2420-2434.
Massey, A.C., Follenzi, A., Kiffin, R., Zhang, C., and Cuervo, A.M. (2008). Early cellular
changes after blockage of chaperone-mediated autophagy. Autophagy 4, 442-456.
Massey, A.C., Kaushik, S., and Cuervo, A.M. (2006a). Lysosomal chat maintains the
balance. Autophagy 2, 325-327.
Massey, A.C., Kaushik, S., Sovak, G., Kiffin, R., and Cuervo, A.M. (2006b).
Consequences of the selective blockage of chaperone-mediated autophagy. Proc Natl
Acad Sci U S A 103, 5805-5810.
Massey, A.C., Kiffin, R., and Cuervo, A.M. (2006c). Autophagic defects in aging:
looking for an "emergency exit"? Cell Cycle 5, 1292-1296.
Massey, A.C., Zhang, C., and Cuervo, A.M. (2006d). Chaperone-mediated autophagy
in aging and disease. Curr Top Dev Biol 73, 205-235.
Masson, N., Willam, C., Maxwell, P.H., Pugh, C.W., and Ratcliffe, P.J. (2001).
Independent function of two destruction domains in hypoxia-inducible factor-alpha
chains activated by prolyl hydroxylation. EMBO J 20, 5197-5206.
Maxwell, P.H., Pugh, C.W., and Ratcliffe, P.J. (2001). Activation of the HIF pathway in
cancer. Curr Opin Genet Dev 11, 293-299.
Maxwell, P.H., Wiesener, M.S., Chang, G.W., Clifford, S.C., Vaux, E.C., Cockman,
M.E., Wykoff, C.C., Pugh, C.W., Maher, E.R., and Ratcliffe, P.J. (1999). The tumour
suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent
proteolysis. Nature 399, 271-275.
149
McClellan, A.J., Endres, J.B., Vogel, J.P., Palazzi, D., Rose, M.D., and Brodsky, J.L.
(1998).
Specific
molecular
chaperone
interactions
and
an
ATP-dependent
conformational change are required during posttranslational protein translocation into
the yeast ER. Mol Biol Cell 9, 3533-3545.
Meijer, A.J., and Codogno, P. (2006). Signalling and autophagy regulation in health,
aging and disease. Mol Aspects Med 27, 411-425.
Meijer, W.H., van der Klei, I.J., Veenhuis, M., and Kiel, J.A. (2007). ATG genes
involved in non-selective autophagy are conserved from yeast to man, but the selective
Cvt and pexophagy pathways also require organism-specific genes. Autophagy 3, 106116.
Melillo, G. (2004). HIF-1: a target for cancer, ischemia and inflammation--too good to
be true? Cell Cycle 3, 154-155.
Mijaljica, D., Prescott, M., Klionsky, D.J., and Devenish, R.J. (2007). Autophagy and
vacuole homeostasis: a case for self-degradation? Autophagy 3, 417-421.
Milani, V., Noessner, E., Ghose, S., Kuppner, M., Ahrens, B., Scharner, A., Gastpar,
R., and Issels, R.D. (2002). Heat shock protein 70: role in antigen presentation and
immune stimulation. Int J Hyperthermia 18, 563-575.
Miranda, M., Dionne, K.R., Sorkina, T., and Sorkin, A. (2007). Three ubiquitin
conjugation sites in the amino terminus of the dopamine transporter mediate protein
kinase C-dependent endocytosis of the transporter. Mol Biol Cell 18, 313-323.
Mizushima, N. (2005). The pleiotropic role of autophagy: from protein metabolism to
bactericide. Cell Death Differ 12 Suppl 2, 1535-1541.
Mizushima, N., and Klionsky, D.J. (2007). Protein turnover via autophagy: implications
for metabolism. Annu Rev Nutr 27, 19-40.
Mizushima, N., Kuma, A., Kobayashi, Y., Yamamoto, A., Matsubae, M., Takao, T.,
Natsume, T., Ohsumi, Y., and Yoshimori, T. (2003). Mouse Apg16L, a novel WDrepeat protein, targets to the autophagic isolation membrane with the Apg12-Apg5
conjugate. J Cell Sci 116, 1679-1688.
Mizushima, N., Levine, B., Cuervo, A.M., and Klionsky, D.J. (2008). Autophagy fights
disease through cellular self-digestion. Nature 451, 1069-1075.
Mizushima, N., Noda, T., Yoshimori, T., Tanaka, Y., Ishii, T., George, M.D., Klionsky,
D.J., Ohsumi, M., and Ohsumi, Y. (1998). A protein conjugation system essential for
autophagy. Nature 395, 395-398.
Mizushima, N., Yamamoto, A., Hatano, M., Kobayashi, Y., Kabeya, Y., Suzuki, K.,
Tokuhisa, T., Ohsumi, Y., and Yoshimori, T. (2001). Dissection of autophagosome
formation using Apg5-deficient mouse embryonic stem cells. J Cell Biol 152, 657-668.
150
Mizushima, N., Yoshimori, T., and Levine, B. (2010). Methods in mammalian
autophagy research. Cell 140, 313-326.
Mortimore,
G.E.,
Lardeux,
B.R.,
and
Adams,
C.E.
(1988).
Regulation
of
microautophagy and basal protein turnover in rat liver. Effects of short-term starvation.
J Biol Chem 263, 2506-2512.
Mortimore, G.E., and Poso, A.R. (1987). Intracellular protein catabolism and its control
during nutrient deprivation and supply. Annu Rev Nutr 7, 539-564.
Mukaiyama, H., Oku, M., Baba, M., Samizo, T., Hammond, A.T., Glick, B.S., Kato, N.,
and Sakai, Y. (2002). Paz2 and 13 other PAZ gene products regulate vacuolar
engulfment of peroxisomes during micropexophagy. Genes Cells 7, 75-90.
Nakatogawa, H., Ichimura, Y., and Ohsumi, Y. (2007). Atg8, a ubiquitin-like protein
required for autophagosome formation, mediates membrane tethering and hemifusion.
Cell 130, 165-178.
Nedelsky, N.B., Todd, P.K., and Taylor, J.P. (2008). Autophagy and the ubiquitinproteasome system: collaborators in neuroprotection. Biochim Biophys Acta 1782, 691699.
Neff, N.T., Bourret, L., Miao, P., and Dice, J.F. (1981). Degradation of proteins
microinjected into IMR-90 human diploid fibroblasts. J Cell Biol 91, 184-194.
Newmyer, S.L., Christensen, A., and Sever, S. (2003). Auxilin-dynamin interactions link
the uncoating ATPase chaperone machinery with vesicle formation. Dev Cell 4, 929940.
Nobukuni, T., Joaquin, M., Roccio, M., Dann, S.G., Kim, S.Y., Gulati, P., Byfield, M.P.,
Backer, J.M., Natt, F., Bos, J.L., et al. (2005). Amino acids mediate mTOR/raptor
signaling through activation of class 3 phosphatidylinositol 3OH-kinase. Proc Natl Acad
Sci U S A 102, 14238-14243.
Noda, N.N., Kumeta, H., Nakatogawa, H., Satoo, K., Adachi, W., Ishii, J., Fujioka, Y.,
Ohsumi, Y., and Inagaki, F. (2008). Structural basis of target recognition by Atg8/LC3
during selective autophagy. Genes Cells 13, 1211-1218.
Noda, T., Suzuki, K., and Ohsumi, Y. (2002). Yeast autophagosomes: de novo
formation of a membrane structure. Trends Cell Biol 12, 231-235.
Novikoff, A.B. (1959). The proximal tubule cell in experimental hydronephrosis. J
Biophys Biochem Cytol 6, 136-138.
Ohh, M., Park, C.W., Ivan, M., Hoffman, M.A., Kim, T.Y., Huang, L.E., Pavletich, N.,
Chau, V., and Kaelin, W.G. (2000). Ubiquitination of hypoxia-inducible factor requires
direct binding to the beta-domain of the von Hippel-Lindau protein. Nat Cell Biol 2, 423427.
151
Olmos, G., Arenas, M.I., Bienes, R., Calzada, M.J., Aragones, J., Garcia-Bermejo,
M.L., Landazuri, M.O., and Lucio-Cazana, J. (2009). 15-Deoxy-Delta(12,14)prostaglandin-J(2) reveals a new pVHL-independent, lysosomal-dependent mechanism
of HIF-1alpha degradation. Cell Mol Life Sci 66, 2167-2180.
Onodera, J., and Ohsumi, Y. (2005). Autophagy is required for maintenance of amino
acid levels and protein synthesis under nitrogen starvation. J Biol Chem 280, 3158231586.
Overbye, A., Fengsrud, M., and Seglen, P.O. (2007). Proteomic analysis of membraneassociated proteins from rat liver autophagosomes. Autophagy 3, 300-322.
Pan, T., Kondo, S., Zhu, W., Xie, W., Jankovic, J., and Le, W. (2008). Neuroprotection
of rapamycin in lactacystin-induced neurodegeneration via autophagy enhancement.
Neurobiol Dis 32, 16-25.
Panaretou, C., Domin, J., Cockcroft, S., and Waterfield, M.D. (1997). Characterization
of p150, an adaptor protein for the human phosphatidylinositol (PtdIns) 3-kinase.
Substrate presentation by phosphatidylinositol transfer protein to the p150.Ptdins 3kinase complex. J Biol Chem 272, 2477-2485.
Pandey, U.B., Batlevi, Y., Baehrecke, E.H., and Taylor, J.P. (2007a). HDAC6 at the
intersection of autophagy, the ubiquitin-proteasome system and neurodegeneration.
Autophagy 3, 643-645.
Pandey, U.B., Nie, Z., Batlevi, Y., McCray, B.A., Ritson, G.P., Nedelsky, N.B.,
Schwartz, S.L., DiProspero, N.A., Knight, M.A., Schuldiner, O., et al. (2007b). HDAC6
rescues neurodegeneration and provides an essential link between autophagy and the
UPS. Nature 447, 859-863.
Pankiv, S., Clausen, T.H., Lamark, T., Brech, A., Bruun, J.A., Outzen, H., Overvatn, A.,
Bjorkoy, G., and Johansen, T. (2007). p62/SQSTM1 binds directly to Atg8/LC3 to
facilitate degradation of ubiquitinated protein aggregates by autophagy. J Biol Chem
282, 24131-24145.
Pattingre, S., Espert, L., Biard-Piechaczyk, M., and Codogno, P. (2008). Regulation of
macroautophagy by mTOR and Beclin 1 complexes. Biochimie 90, 313-323.
Pattingre, S., Tassa, A., Qu, X., Garuti, R., Liang, X.H., Mizushima, N., Packer, M.,
Schneider, M.D., and Levine, B. (2005). Bcl-2 antiapoptotic proteins inhibit Beclin 1dependent autophagy. Cell 122, 927-939.
Pfeifer, U. (1978). Inhibition by insulin of the formation of autophagic vacuoles in rat
liver. A morphometric approach to the kinetics of intracellular degradation by
autophagy. J Cell Biol 78, 152-167.
Pines, J. (2006). Mitosis: a matter of getting rid of the right protein at the right time.
Trends Cell Biol 16, 55-63.
152
Pryor, P.R., and Luzio, J.P. (2009). Delivery of endocytosed membrane proteins to the
lysosome. Biochim Biophys Acta 1793, 615-624.
Ramesh Babu, J., Lamar Seibenhener, M., Peng, J., Strom, A.L., Kemppainen, R.,
Cox, N., Zhu, H., Wooten, M.C., Diaz-Meco, M.T., Moscat, J., et al. (2008). Genetic
inactivation of p62 leads to accumulation of hyperphosphorylated tau and
neurodegeneration. J Neurochem 106, 107-120.
Rankin, E.B., and Giaccia, A.J. (2008). The role of hypoxia-inducible factors in
tumorigenesis. Cell Death Differ 15, 678-685.
Ravi, R., Mookerjee, B., Bhujwalla, Z.M., Sutter, C.H., Artemov, D., Zeng, Q., Dillehay,
L.E., Madan, A., Semenza, G.L., and Bedi, A. (2000). Regulation of tumor
angiogenesis by p53-induced degradation of hypoxia-inducible factor 1alpha. Genes
Dev 14, 34-44.
Ravikumar, B., Vacher, C., Berger, Z., Davies, J.E., Luo, S., Oroz, L.G., Scaravilli, F.,
Easton, D.F., Duden, R., O'Kane, C.J., et al. (2004). Inhibition of mTOR induces
autophagy and reduces toxicity of polyglutamine expansions in fly and mouse models
of Huntington disease. Nat Genet 36, 585-595.
Regelmann, J., Schule, T., Josupeit, F.S., Horak, J., Rose, M., Entian, K.D., Thumm,
M., and Wolf, D.H. (2003). Catabolite degradation of fructose-1,6-bisphosphatase in
the yeast Saccharomyces cerevisiae: a genome-wide screen identifies eight novel GID
genes and indicates the existence of two degradation pathways. Mol Biol Cell 14,
1652-1663.
Reggiori, F., and Klionsky, D.J. (2005). Autophagosomes: biogenesis from scratch?
Curr Opin Cell Biol 17, 415-422.
Reggiori, F., Tucker, K.A., Stromhaug, P.E., and Klionsky, D.J. (2004). The Atg1-Atg13
complex regulates Atg9 and Atg23 retrieval transport from the pre-autophagosomal
structure. Dev Cell 6, 79-90.
Richter, K., and Buchner, J. (2006). hsp90: twist and fold. Cell 127, 251-253.
Rideout, H.J., Lang-Rollin, I., and Stefanis, L. (2004). Involvement of macroautophagy
in the dissolution of neuronal inclusions. Int J Biochem Cell Biol 36, 2551-2562.
Roberts, P., Moshitch-Moshkovitz, S., Kvam, E., O'Toole, E., Winey, M., and Goldfarb,
D.S. (2003). Piecemeal microautophagy of nucleus in Saccharomyces cerevisiae. Mol
Biol Cell 14, 129-141.
Rubinsztein, D.C., Gestwicki, J.E., Murphy, L.O., and Klionsky, D.J. (2007). Potential
therapeutic applications of autophagy. Nat Rev Drug Discov 6, 304-312.
Sakai, Y., Oku, M., van der Klei, I.J., and Kiel, J.A. (2006). Pexophagy: autophagic
degradation of peroxisomes. Biochim Biophys Acta 1763, 1767-1775.
153
Salnikow, K., Donald, S.P., Bruick, R.K., Zhitkovich, A., Phang, J.M., and Kasprzak,
K.S. (2004). Depletion of intracellular ascorbate by the carcinogenic metals nickel and
cobalt results in the induction of hypoxic stress. J Biol Chem 279, 40337-40344.
Salvador, N., Aguado, C., Horst, M., and Knecht, E. (2000). Import of a cytosolic
protein into lysosomes by chaperone-mediated autophagy depends on its folding state.
J Biol Chem 275, 27447-27456.
Schofield, C.J., and Ratcliffe, P.J. (2005). Signalling hypoxia by HIF hydroxylases.
Biochem Biophys Res Commun 338, 617-626.
Scott, S.V., Guan, J., Hutchins, M.U., Kim, J., and Klionsky, D.J. (2001). Cvt19 is a
receptor for the cytoplasm-to-vacuole targeting pathway. Mol Cell 7, 1131-1141.
Seglen, P.O., and Gordon, P.B. (1982). 3-Methyladenine: specific inhibitor of
autophagic/lysosomal protein degradation in isolated rat hepatocytes. Proc Natl Acad
Sci U S A 79, 1889-1892.
Seglen, P.O., Grinde, B., and Solheim, A.E. (1979). Inhibition of the lysosomal pathway
of protein degradation in isolated rat hepatocytes by ammonia, methylamine,
chloroquine and leupeptin. Eur J Biochem 95, 215-225.
Seibenhener, M.L., Babu, J.R., Geetha, T., Wong, H.C., Krishna, N.R., and Wooten,
M.W. (2004). Sequestosome 1/p62 is a polyubiquitin chain binding protein involved in
ubiquitin proteasome degradation. Mol Cell Biol 24, 8055-8068.
Semenza, G.L. (2003). Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3, 721732.
Semenza, G.L., and Wang, G.L. (1992). A nuclear factor induced by hypoxia via de
novo protein synthesis binds to the human erythropoietin gene enhancer at a site
required for transcriptional activation. Mol Cell Biol 12, 5447-5454.
Shen, S., Zhang, P., Lovchik, M.A., Li, Y., Tang, L., Chen, Z., Zeng, R., Ma, D., Yuan,
J., and Yu, Q. (2009). Cyclodepsipeptide toxin promotes the degradation of Hsp90
client proteins through chaperone-mediated autophagy. J Cell Biol 185, 629-639.
Shintani, T., Huang, W.P., Stromhaug, P.E., and Klionsky, D.J. (2002). Mechanism of
cargo selection in the cytoplasm to vacuole targeting pathway. Dev Cell 3, 825-837.
Shintani, T., Mizushima, N., Ogawa, Y., Matsuura, A., Noda, T., and Ohsumi, Y.
(1999). Apg10p, a novel protein-conjugating enzyme essential for autophagy in yeast.
EMBO J 18, 5234-5241.
Simpson, M.V. (1953). The release of labeled amino acids from the proteins of rat liver
slices. J Biol Chem 201, 143-154.
Steinman, R.M., Mellman, I.S., Muller, W.A., and Cohn, Z.A. (1983). Endocytosis and
the recycling of plasma membrane. J Cell Biol 96, 1-27.
154
Stromhaug, P.E., Berg, T.O., Fengsrud, M., and Seglen, P.O. (1998). Purification and
characterization of autophagosomes from rat hepatocytes. Biochem J 335 ( Pt 2), 217224.
Sugawara, K., Suzuki, N.N., Fujioka, Y., Mizushima, N., Ohsumi, Y., and Inagaki, F.
(2004). The crystal structure of microtubule-associated protein light chain 3, a
mammalian homologue of Saccharomyces cerevisiae Atg8. Genes Cells 9, 611-618.
Suh, W.C., Lu, C.Z., and Gross, C.A. (1999). Structural features required for the
interaction of the Hsp70 molecular chaperone DnaK with its cochaperone DnaJ. J Biol
Chem 274, 30534-30539.
Susan, P.P., and Dunn, W.A., Jr. (2001). Starvation-induced lysosomal degradation of
aldolase B requires glutamine 111 in a signal sequence for chaperone-mediated
transport. J Cell Physiol 187, 48-58.
Suzuki, K., Kirisako, T., Kamada, Y., Mizushima, N., Noda, T., and Ohsumi, Y. (2001).
The pre-autophagosomal structure organized by concerted functions of APG genes is
essential for autophagosome formation. EMBO J 20, 5971-5981.
Suzuki, K., Kubota, Y., Sekito, T., and Ohsumi, Y. (2007). Hierarchy of Atg proteins in
pre-autophagosomal structure organization. Genes Cells 12, 209-218.
Suzuki, N.N., Yoshimoto, K., Fujioka, Y., Ohsumi, Y., and Inagaki, F. (2005). The
crystal structure of plant ATG12 and its biological implication in autophagy. Autophagy
1, 119-126.
Takahashi, Y., Coppola, D., Matsushita, N., Cualing, H.D., Sun, M., Sato, Y., Liang, C.,
Jung, J.U., Cheng, J.Q., Mule, J.J., et al. (2007). Bif-1 interacts with Beclin 1 through
UVRAG and regulates autophagy and tumorigenesis. Nat Cell Biol 9, 1142-1151.
Tal, R., Winter, G., Ecker, N., Klionsky, D.J., and Abeliovich, H. (2007). Aup1p, a yeast
mitochondrial protein phosphatase homolog, is required for efficient stationary phase
mitophagy and cell survival. J Biol Chem 282, 5617-5624.
Tanaka, Y., Guhde, G., Suter, A., Eskelinen, E.L., Hartmann, D., Lullmann-Rauch, R.,
Janssen, P.M., Blanz, J., von Figura, K., and Saftig, P. (2000). Accumulation of
autophagic vacuoles and cardiomyopathy in LAMP-2-deficient mice. Nature 406, 902906.
Tanida, I., Tanida-Miyake, E., Komatsu, M., Ueno, T., and Kominami, E. (2002).
Human Apg3p/Aut1p homologue is an authentic E2 enzyme for multiple substrates,
GATE-16, GABARAP, and MAP-LC3, and facilitates the conjugation of hApg12p to
hApg5p. J Biol Chem 277, 13739-13744.
Tassa, A., Roux, M.P., Attaix, D., and Bechet, D.M. (2003). Class III phosphoinositide
3-kinase--Beclin1 complex mediates the amino acid-dependent regulation of
autophagy in C2C12 myotubes. Biochem J 376, 577-586.
155
Taylor, J.P., Tanaka, F., Robitschek, J., Sandoval, C.M., Taye, A., Markovic-Plese, S.,
and Fischbeck, K.H. (2003). Aggresomes protect cells by enhancing the degradation of
toxic polyglutamine-containing protein. Hum Mol Genet 12, 749-757.
Terada, K., and Mori, M. (2000). Human DnaJ homologs dj2 and dj3, and bag-1 are
positive cochaperones of hsc70. J Biol Chem 275, 24728-24734.
Terada, K., Ueda, I., Ohtsuka, K., Oda, T., Ichiyama, A., and Mori, M. (1996). The
requirement of heat shock cognate 70 protein for mitochondrial import varies among
precursor proteins and depends on precursor length. Mol Cell Biol 16, 6103-6109.
Terlecky, S.R., Chiang, H.L., Olson, T.S., and Dice, J.F. (1992). Protein and peptide
binding and stimulation of in vitro lysosomal proteolysis by the 73-kDa heat shock
cognate protein. J Biol Chem 267, 9202-9209.
Terlecky, S.R., and Dice, J.F. (1993). Polypeptide import and degradation by isolated
lysosomes. J Biol Chem 268, 23490-23495.
Todde, V., Veenhuis, M., and van der Klei, I.J. (2009). Autophagy: principles and
significance in health and disease. Biochim Biophys Acta 1792, 3-13.
Tung, Y.T., Hsu, W.M., Lee, H., Huang, W.P., and Liao, Y.F. (2010). The evolutionarily
conserved interaction between LC3 and p62 selectively mediates autophagydependent degradation of mutant huntingtin. Cell Mol Neurobiol 30, 795-806.
Ueno, T., Muno, D., and Kominami, E. (1991). Membrane markers of endoplasmic
reticulum preserved in autophagic vacuolar membranes isolated from leupeptinadministered rat liver. J Biol Chem 266, 18995-18999.
Uttenweiler, A., and Mayer, A. (2008). Microautophagy in the yeast Saccharomyces
cerevisiae. Methods Mol Biol 445, 245-259.
Uttenweiler, A., Schwarz, H., Neumann, H., and Mayer, A. (2007). The vacuolar
transporter chaperone (VTC) complex is required for microautophagy. Mol Biol Cell 18,
166-175.
van de Sluis, B., Groot, A.J., Vermeulen, J., van der Wall, E., van Diest, P.J.,
Wijmenga, C., Klomp, L.W., and Vooijs, M. (2009). COMMD1 Promotes pVHL and O2Independent Proteolysis of HIF-1alpha via HSP90/70. PLoS One 4, e7332.
Veenhuis, M., Douma, A., Harder, W., and Osumi, M. (1983). Degradation and turnover
of peroxisomes in the yeast Hansenula polymorpha induced by selective inactivation of
peroxisomal enzymes. Arch Microbiol 134, 193-203.
Veenhuis, M., Salomons, F.A., and Van Der Klei, I.J. (2000). Peroxisome biogenesis
and degradation in yeast: a structure/function analysis. Microsc Res Tech 51, 584-600.
Veenhuis, M., van Dijken, J.P., Pilon, S.A., and Harder, W. (1978). Development of
crystalline peroxisomes in methanol-grown cells of the yeast Hansenula polymorpha
and its relation to environmental conditions. Arch Microbiol 117, 153-163.
156
Vogel, J.P., Misra, L.M., and Rose, M.D. (1990). Loss of BiP/GRP78 function blocks
translocation of secretory proteins in yeast. J Cell Biol 110, 1885-1895.
Wang, G.L., and Semenza, G.L. (1993). Desferrioxamine induces erythropoietin gene
expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models
of hypoxia signal transduction. Blood 82, 3610-3615.
Webb, J.L., Ravikumar, B., Atkins, J., Skepper, J.N., and Rubinsztein, D.C. (2003).
Alpha-Synuclein is degraded by both autophagy and the proteasome. J Biol Chem 278,
25009-25013.
Weidemann, A., and Johnson, R.S. (2008). Biology of HIF-1alpha. Cell Death Differ 15,
621-627.
Wibo, M., and Poole, B. (1974). Protein degradation in cultured cells. II. The uptake of
chloroquine by rat fibroblasts and the inhibition of cellular protein degradation and
cathepsin B1. J Cell Biol 63, 430-440.
Wilkinson, K.D., Urban, M.K., and Haas, A.L. (1980). Ubiquitin is the ATP-dependent
proteolysis factor I of rabbit reticulocytes. J Biol Chem 255, 7529-7532.
Wing, S.S., Chiang, H.L., Goldberg, A.L., and Dice, J.F. (1991). Proteins containing
peptide sequences related to Lys-Phe-Glu-Arg-Gln are selectively depleted in liver and
heart, but not skeletal muscle, of fasted rats. Biochem J 275 ( Pt 1), 165-169.
Wullschleger, S., Loewith, R., and Hall, M.N. (2006). TOR signaling in growth and
metabolism. Cell 124, 471-484.
Xiao, G. (2007). Autophagy and NF-kappaB: fight for fate. Cytokine Growth Factor Rev
18, 233-243.
Xie, Z., and Klionsky, D.J. (2007). Autophagosome formation: core machinery and
adaptations. Nat Cell Biol 9, 1102-1109.
Yamamoto, A., Cremona, M.L., and Rothman, J.E. (2006). Autophagy-mediated
clearance of huntingtin aggregates triggered by the insulin-signaling pathway. J Cell
Biol 172, 719-731.
Yang, J., Bridges, K., Chen, K.Y., and Liu, A.Y. (2008). Riluzole increases the amount
of latent HSF1 for an amplified heat shock response and cytoprotection. PLoS One 3,
e2864.
Yang, Q., She, H., Gearing, M., Colla, E., Lee, M., Shacka, J.J., and Mao, Z. (2009).
Regulation of neuronal survival factor MEF2D by chaperone-mediated autophagy.
Science 323, 124-127.
Yee Koh, M., Spivak-Kroizman, T.R., and Powis, G. (2008). HIF-1 regulation: not so
easy come, easy go. Trends Biochem Sci 33, 526-534.
Yorimitsu, T., and Klionsky, D.J. (2005). Autophagy: molecular machinery for selfeating. Cell Death Differ 12 Suppl 2, 1542-1552.
157
Yorimitsu, T., and Klionsky, D.J. (2007). Eating the endoplasmic reticulum: quality
control by autophagy. Trends Cell Biol 17, 279-285.
Young, A.R., Chan, E.Y., Hu, X.W., Kochl, R., Crawshaw, S.G., High, S., Hailey, D.W.,
Lippincott-Schwartz, J., and Tooze, S.A. (2006). Starvation and ULK1-dependent
cycling of mammalian Atg9 between the TGN and endosomes. J Cell Sci 119, 38883900.
Yu, A.Y., Frid, M.G., Shimoda, L.A., Wiener, C.M., Stenmark, K., and Semenza, G.L.
(1998). Temporal, spatial, and oxygen-regulated expression of hypoxia-inducible
factor-1 in the lung. Am J Physiol 275, L818-826.
Yuan, H.T., Li, X.Z., Pitera, J.E., Long, D.A., and Woolf, A.S. (2003). Peritubular
capillary loss after mouse acute nephrotoxicity correlates with down-regulation of
vascular endothelial growth factor-A and hypoxia-inducible factor-1 alpha. Am J Pathol
163, 2289-2301.
Zhang, C., and Cuervo, A.M. (2008). Restoration of chaperone-mediated autophagy in
aging liver improves cellular maintenance and hepatic function. Nat Med 14, 959-965.
Zhang, K., and Kaufman, R.J. (2006). The unfolded protein response: a stress
signaling pathway critical for health and disease. Neurology 66, S102-109.
Zhou, D., Li, P., Lin, Y., Lott, J.M., Hislop, A.D., Canaday, D.H., Brutkiewicz, R.R., and
Blum, J.S. (2005). Lamp-2a facilitates MHC class II presentation of cytoplasmic
antigens. Immunity 22, 571-581.
Zhou, J., and Brune, B. (2006). Cytokines and hormones in the regulation of hypoxia
inducible factor-1alpha (HIF-1alpha). Cardiovasc Hematol Agents Med Chem 4, 189197.
Zhou, J., Kohl, R., Herr, B., Frank, R., and Brune, B. (2006). Calpain mediates a von
Hippel-Lindau protein-independent destruction of hypoxia-inducible factor-1alpha. Mol
Biol Cell 17, 1549-1558.
158