TISSUE-SPECIFIC STEM CELLS
Progenitor Cells from the Porcine Neural Retina Express
Photoreceptor Markers After Transplantation to the Subretinal
Space of Allorecipients
HENRY KLASSEN,a,b JENS FOLKE KIILGAARD,c TASNEEM ZAHIR,b BOBACK ZIAEIAN,a IVAN KIROV,a
ERIK SCHERFIG,c KARIN WARFVINGE,d MICHAEL J. YOUNGb
a
Stem Cell Research, Children’s Hospital of Orange County, Orange, California, USA; bSchepens Eye Research
Institute, Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, USA; cEye Department,
Rigshospitalet and Eye Pathology Institute, Copenhagen University, Copenhagen, Denmark; dWallenberg Retina
Center, Department of Ophthalmology, Lund University, Lund, Sweden
Key Words. Stem cells • Photoreceptor • Rhodopsin • Recoverin • Transducin • Swine
ABSTRACT
Work in rodents has shown that cultured retinal progenitor
cells (RPCs) integrate into the degenerating retina, thus
suggesting a potential strategy for treatment of similar degenerative conditions in humans. To demonstrate the relevance of the rodent work to large animals, we derived
progenitor cells from the neural retina of the domestic pig
and transplanted them to the laser-injured retina of allorecipients. Prior to grafting, immunocytochemical analysis
showed that cultured porcine RPCs widely expressed neural
cell adhesion molecule, as well as markers consistent with
immature neural cells, including nestin, Sox2, and vimentin.
Subpopulations expressed the neurodevelopmental markers
CD-15, doublecortin, ␤-III tubulin, and glial fibrillary acidic
protein. Retina-specific markers expressed included the bipolar marker protein kinase C␣ and the photoreceptorassociated markers recoverin and rhodopsin. In addition,
reverse transcription-polymerase chain reaction showed expression of the transcription factors Dach1, Hes1, Lhx2,
Pax6, Six3, and Six6. Progenitor cells prelabeled with vital
dyes survived as allografts in the subretinal space for up to
5 weeks (11 of 12 recipients) without exogenous immune
suppression. Grafted cells expressed transducin, recoverin,
and rhodopsin in the pig subretinal space, suggestive of
differentiation into photoreceptors or, in a few cases, migrated into the neural retina and extended processes, the
latter typically showing radial orientation. These results
demonstrate that many of the findings seen with rodent
RPCs can be duplicated in a large mammal. The pig offers
a number of advantages over mice and rats, particularly in
terms of functional testing and evaluation of the potential
for clinical translation to human subjects. STEM CELLS
2007;25:1222–1230
Disclosure of potential conflicts of interest is found at the end of this article.
INTRODUCTION
There has been recent interest in identifying and culturing stem
and progenitor cells from the central nervous system (CNS). On
the one hand, studies of this type provide insights into the
cellular mechanism underlying CNS development [1–3],
whereas on the other hand, considerable enthusiasm has been
generated by the demonstrated potential of these cells for CNS
repair following transplantation to the brain [4, 5] and spinal
cord [6, 7] of rodents. Work has shown equal promise in the
retina, both in terms of developmental neurobiology [8] and
regeneration [9 –12]. Encouraging results such as these have
raised the possibility of applying CNS progenitor cell transplantation to patients with retinal degenerative disorders.
A major challenge facing translational efforts with respect
to transplantation of stem and progenitor cells to the CNS is
the need for more predictive animal models. Although rodents
continue to play a pivotal role in basic science research, the
further development of scientific breakthroughs into new cell-
based therapeutic modalities is hampered by the disparity between rodents and humans in terms of both the underlying
biology and the surgical anatomy. This is particularly true in the
eye.
One animal that has proven particularly useful for modeling
human subretinal surgery is the pig [13–15]. In addition to
anatomical considerations, we have recently reported that the
gene expression profile of cultured porcine forebrain progenitor
cells more closely resembles that of the analogous human cells
than that of the mouse [16]. This raises the question of whether
progenitor cells can also be cultured from the pig retina and, if
so, whether the gene expression of these cells resembles their
human counterparts [17, 18] more than the analogous mouse
cells [10].
Here we show that progenitor cells can be propagated from
fetal porcine retina (porcine retinal progenitor cells [pRPCs])
and that these cells express a range of well-established immature
neurodevelopmental and retinal markers in culture. The pRPCs
also express mature markers after differentiation, including the
rod photoreceptor marker rhodopsin, both in culture and follow-
Correspondence: Henry J. Klassen, M.D., Ph.D., Department of Ophthalmology, School of Medicine, University of California, Irvine, 101
The City Drive, Building 55, Orange, California 92868-4380, USA. Telephone: 714-456-7370; Fax: 714-456-5073; e-mail:
hklassen@uci.edu Received August 29, 2006; accepted for publication January 4, 2007; first published online in STEM CELLS EXPRESS
January 11, 2007. ©AlphaMed Press 1066-0599/2007/$30.00/0 doi: 10.1634/stemcells.2006-0541
STEM CELLS 2007;25:1222–1230 www.StemCells.com
Klassen, Kiilgaard, Zahir et al.
1223
Table 1. Primary antibodies for immunocytochemistry
Antigen
CD15
DCX
GFAP
NCAM
Nestin
PKC␣
Recoverin
Rhodopsin
Sox-2
Transducin
␤-III Tubulin
Vimentin
Species
Supplier
Product code
Dilution
Mouse
Goat
Guinea pig
Rabbit
Mouse
Goat
Rabbit
Mouse
Goat
Rabbit
Mouse
Mouse
BD Pharmingena
Santa Cruz Biotechnologyb
Chemiconc
Chemicon
BD Pharmingen
Santa Cruz Biotechnology
Chemicon
R. Moldayd
Santa Cruz Biotechnology
CytoSignale
Chemicon
Sigma-Aldrichf
559045
sc-8066
AB1540
AB5032
611658
sc-12356
AB5431P
4D2
sc-17320
TF15
MAB1637
V 6630
1:100
1:100
1:200
1:100
1:400
1:1,000
1:100–200
1:500
1:50
1:1,000
1:100
1:200
a
San Diego, http://www.bdbiosciences.com/pharmingen.
Santa Cruz, CA, http://www.scbt.com.
Temecula, CA, http://www.chemicon.com.
d
University of British Columbia, Vancouver, BC, Canada.
e
Irvine, CA, http://www.cytosignal.com
f
St. Louis, http://www.sigmaaldrich.com.
Abbreviations: DCX, doublecortin; GFAP, glial fibrillary acidic protein; PKC␣, protein kinase C␣.
b
c
ing transplantation to the subretinal space of allogeneic recipients.
MATERIALS
AND
METHODS
washed with Ca2⫹- and Mg2⫹-free Hanks’ balanced salt solution
(HBSS) (Invitrogen, Carlsbad, CA, http://www.invitrogen.com).
Fresh medium containing CNTF (20 ng/ml; R&D Systems) or 10%
FBS, but no EGF or bFGF, was added to the experimental wells/
flasks. The differentiation medium was changed every 3 days and
the cells maintained for up to 7 days.
Donor Animals
Immunocytochemistry
A pregnant sow was placed under general anesthesia, the uterine
horns and fetuses were removed, and the sow was terminated prior
to waking. Work was performed according to an Institutional Animal Care and Use Committee-approved protocol. All tissues were
obtained in compliance with NIH and institutional guidelines.
Live cells grown on chamber well slides were fixed for 10 minutes
in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS)
(Irvine Scientific). The fixed cells were washed with PBS containing 0.05% (wt/vol) sodium azide. A blocking solution consisting of
Tris-buffered saline (TBS) ⫹ 0.3% Triton X-100 ⫹ 3% donkey
serum (Jackson Immunoresearch Laboratories, West Grove, PA,
http://www.jacksonimmuno.com) was then applied for 15 minutes.
Cells were subsequently rinsed twice in 0.1 M TBS buffer. Primary
antibodies were diluted in 250 ␮l of antibody buffer (TBS ⫹ 0.3%
Triton X-100 ⫹ 1.0% donkey serum) at concentrations determined
through usage in the laboratory (Table 1). Primary antibodies were
applied to the samples and left at 5°C overnight and then rinsed
twice with TBS the next day. Secondary antibodies were donkeyderived and diluted 1:100 in antibody buffer or goat-derived and
diluted 1:800 (Jackson Immunoresearch Laboratories). Secondary
antibodies were applied and left at 5°C overnight. The following
day, samples were rinsed for 5 minutes with TBS three times. Slides
were mounted with Prolong Antifade Kit (Molecular Probes Inc.,
Eugene, OR, http://probes.invitrogen.com). Digital images obtained
with an Olympus Ix70 Microscope and Optronics Quantifire CCD
camera (Tokyo, http://www.olympus-global.com). Electronic image
files were managed using Image Pro Plus 4.0 software with AFA
plugin 4.5 (Media Cybernetics, Bethesda, MD, http://www.
mediacy.com).
Cell Isolation and Culture
The techniques used for isolation of human RPCs were described
previously [18]. The present isolation followed a similar protocol, in
this case using fetal pigs collected at 60 days of gestation (typical
gestational period in pig being 114 days). Briefly, the eyes were
removed, the sclera and choroid were incised and reflected, the
retinal pigment epithelium (RPE) was opened by tangential traction,
and the neural retina was extruded from the globe and cut free from
attachments along its periphery and at the optic nerve head. Pooled
retinal tissue was minced and enzymatically digested, and the liberated cells were washed and cultured at high density in fibronectincoated flasks containing Dulbecco’s modified Eagle’s medium/
Ham’s F-12 medium with high glucose (Irvine Scientific, Irvine,
CA, http://www.irvinesci.com), L-glutamine (200 mM), BIT9500
(10% by volume; Stem Cell Technologies, Vancouver, BC, Canada,
http://www.stemcell.com), epidermal growth factor (EGF) (20 ng/
ml), basic fibroblast growth factor (bFGF) (40 ng/ml), and antibiotics. Fetal bovine serum (FBS) (10% by volume) was included
overnight, and the medium was completely changed the next day
with growth factors reduced to 20 ng/ml. Subsequently, cells were
fed by 50% medium exchange every 2–3 days and passaged at
confluence using Cell Dissociation Buffer (Gibco, Grand Island,
NY, http://www.invitrogen.com), centrifugation, and gentle trituration.
Differentiation in Culture
To differentiate cultured cells, growth medium was replaced with
medium without mitogens but containing either FBS or ciliary
neurotrophic factor. Specifically, neurospheres were grown for 24
hours in complete growth medium containing EGF and bFGF on
tissue culture slides coated with either poly-D-lysine and laminin or
fibronectin to generate adherent retinal progenitor cells. The complete cell culture medium was then removed, and the cells were
www.StemCells.com
Immunoblot Analysis
Retinal progenitor cells or fetal porcine retinas (60 days gestation)
were homogenized in lysis buffer (1% Triton X-100, 10 mM TrisHCl, pH 7.4, 5 mM EDTA, 50 mM NaCl, 50 mM NaF) containing
protease inhibitor cocktail (1:100 dilution; Sigma-Aldrich) and
phosphatase inhibitor (1:100 dilution; Sigma-Aldrich). Protein levels of total cell lysates were quantified with a protein assay kit
(Bio-Rad, Hercules, CA, http://www.bio-rad.com). The protein
samples (20 ␮g) were separated on polyacrylamide gels (NuPage;
Invitrogen) for 40 minutes at 160 V and transferred to polyvinylidene difluoride membranes (Invitrolon; Invitrogen) for 60 minutes
at 30 V. After transfer, the membranes were blocked in 5% nonfat
dry milk in TBS-T (10 mM Tris HCl, pH 7.4, 150 mM NaCl, 0.1%
Tween 20) for 30 minutes. The blots were incubated with the
following primary antibodies: glial fibrillary acidic protein (GFAP)
Porcine Retinal Progenitor Cells
1224
Table 2. Reverse transcription-polymerase chain reaction primers and conditions
Gene
Dach1
DCX
GFAP
Hes1
Lhx2
Nestin
Pax6
Six3
Six6
Sox2
␤-Actin
5ⴕ Primer
3ⴕ Primer
Annealing
temperature (°C)
Size (base
pairs)
AGGCTTTCG ACCTGTTCCTGAA
AATCCCAACTGGTCTGTCAAC
ACATCGAGATCGCCACCTAC
CAGCCAGTGTCAACACGACAC
CGGTGGACAAGCAGTGGCACAT
GGCAGCGTTGGAACAGAGGTTGGA
CCAGCCAGAGCCAGCATGCAGAACA
AGCGGACTCGGAGCCTGTTG
GGTGGGCAACTGGTTCAAAAACC
GGCAGCTACAGCATGATGCAGGAGC
CGTGCTGCTGACCGAGGCC
GCTGTCAGACCTGTTGGTGGAA
GTTTCCCTTCATGACTCGGCA
ACATCACATCCTTGTGCTCC
TCGTTCATGCACTCGCTGA
TCCTTCATGCCGAAG TGGTCGC
CTCTAAACTGGAGTGGTCAGGGCT
GGTTGGTAGACACTGGTGCTGAAACT
AGCGCATGCCGCTCGGTCCA
TGTCGCTGGACGTGATGGAGATG
CTGGTCATGGAGTTGTACTGCAGG
TTCGTGGATGCCACAGGAC
54
57
64
56
54
65
73
66
66
73
68
336
405
219
307
274
718
950
202
212
131
522
Primer designs were based on human gene sequence information. When possible, primers were chosen to flank at least 1 intron.
(1:1000 dilution; Chemicon), protein kinase C␣ (PKC␣) (1:200
dilution; Santa Cruz Biotechnology Inc.), and Recoverin (1:1,000
dilution; Chemicon). Subsequently, blots were incubated with
horseradish peroxidase-conjugated species-specific secondary antibodies, and the signals were visualized with the enhanced chemiluminescence Western blotting detection system (Amersham Biosciences, Piscataway, NJ, http://www.amersham.com).
Reverse Transcription-Polymerase Chain Reaction
Total RNA was extracted from progenitor cells at 3 weeks in culture
(passage 4) using Purescript RNA Isolation Kit (Gentra, Valencia,
CA, http://www.gentra.com), according to the manufacturer’s protocol. Residual genomic DNA was eliminated with DNase (DNAfree; Ambion, Austin, TX, http://www.ambion.com). Extracted
RNA was then reverse transcribed using Moloney murine leukemia
virus (M-MLV) reverse transcriptase (Invitrogen). Negative controls were performed that contained RNA, but no M-MLV reverse
transcriptase, to further eliminate the possibility that polymerase
chain reaction (PCR) product resulted from amplified genomic
DNA. Automated PCR was carried out in a final volume of 50 ␮l
with 3 ␮l of cDNA template, 0.75 ␮l of forward and reverse primers
(0.5 ␮g/␮l) (Qiagen, Hilden, Germany, http://www1.qiagen.com)
(Table 2), and 1.25 units of Taq DNA polymerase (Amersham
Biosciences) in a Techne Genius thermocycler. Primer designs were
based on human gene sequences. Initial denaturation for 4 minutes
at 94°C was followed by 30 cycles of 1 minute at 94°C, 1 minute at
the corresponding annealing temperature, and 1 minute at 72°C. The
final step consisted of 7 minutes of extension at 72°C. Products
were run on 2% agarose gels and visualized with ethidium bromide
against a 100-base pair ladder.
Recipient Animals and Surgery
Twelve female domestic pigs (Danish Landrace breed; age, 4
months; approximate weight, 30 kg) were used as recipients. Prior
to surgery, animals were given intramuscular injections of 15 mg/ml
midazolam (DormicumA; Roche, Hvidovre, Denmark, http://www.
roche.com) and a 3-ml composition of 11.9 mg of zolazepam
(Zoletin 50 Vet; Virbac SA, Carros, France, http://www.virbaccorp.
com) and 11.9 mg of tiletamine (Zoletin) mixed with 12.38 mg/ml
xylazine (Intervet, Skovlunde, Denmark, http://www.intervet.com),
14.29 mg/ml ketamine (Intervet), and 2.38 mg/ml methadone (Nycomed, Roskilde, Denmark, http://www.nycomed.com). Endotracheal intubation was performed, and the pigs were artificially ventilated and anesthetized with 2%–3% isoflurane (Abbott, Solna,
Sweden, http://www.abbott.com) in combination with oxygen.
Stroke volume (300 ml/stroke) and respiratory frequency (12
breaths per minutes) were held constant throughout the duration of
the surgery. The left pupil was dilated, and the cornea was anesthetized with topical drops consisting of 0.4% oxybuprocain (SAD,
Copenhagen, Denmark), 10% Metaoxedrin (SAD), 0.5% Mydriacyl
(Alcon, Belgium, http://www.alconlabs.com), 1% atropine (SAD),
and 5% povidone-iodine (SAD). Surgery was restricted to the left
eye in all animals.
Table 3. Parameters of the laser lesions
Laser parameters
Pig no.
93
94
95
96
97
98
101
102
103
104
105
106
Spot size
a
(mm)
Duration
(seconds)
Power
(W)
Spot no.
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.1
0.1
0.1
0.1
Not recorded
0.1
0.5
0.5
0.5
0.5
0.5
0.5
0.09
0.09
0.09
0.09
Not recorded
0.09
1.2
1.2
1.2
1.2
1.2
0.3
9
9
9
9
9
10
10
10
9
9
9
9
a
Spot size was estimated by comparison with optic disc diameter
using fundus photography.
Using a localized three-port pars plana vitrectomy, the central
and posterior vitreous was removed, together with the posterior
hyaloid membrane. To promote integration of grafted pRPCs into
the host retina, focal damage was induced in all 12 animals via
application of green argon laser burns to the area centralis, as
described previously [19]. Laser parameters and number of burns
applied are shown for each animal in Table 3. A bleb was then
elevated in the area of laser burns by subretinal injection of 0.25–
0.5 ml of 0.9% NaCl through a 41-gauge needle. Endodiathermy
was applied to the detached retina prior to enlargement of the
retinotomy for transplantation.
To prelabel the cells prior to transplantation, cultured pRPCs
were incubated for 5 minutes with one or more vital dyes, followed
by three washes in HBSS with gentle centrifugation. These included
a non-nuclear vital dye used to visualize donor cell morphology; a
lipophilic dye, either PKH26 (red fluorochrome; 2 ⫻ 106 M; SigmaAldrich) or PKH67 (green fluorochrome; 2 ⫻ 106 M; SigmaAldrich); and, in some cases, the nuclear dye 4⬘,6-diamidino-2phenylindole (DAPI) (10 ␮g/ml; Sigma-Aldrich) as well. Prelabeled
pRPCs were injected as a single cell suspension containing approximately 2 ⫻ 107 cells into the retinal bleb using a 20-, 25-, or
27-gauge metal cannula, with or without silicon tip. Immediate
reflux of some cells into the vitreous cavity was frequently observed. An air bubble was placed under the retinotomy to prevent
additional reflux after withdrawal of the cannula. Chloramphenicol
(SAD) was applied to the conjunctiva and ocular surface after
completion of surgery. The pigs were examined weekly by ophthalmoscopy to clinically evaluate all operated eyes, with particular
attention to the vitreous, retina, and transplantation site.
The research protocol used here was approved by the Danish
Animal Experiment Inspectorate and is also in accordance with the
Klassen, Kiilgaard, Zahir et al.
1225
Table 4. Transplantation of porcine retinal progenitor cells to the
subretinal space of allogeneic recipients
Pig no.
97
94
98
101
102
93
95
103
104
96
105
106
Fluorochrome
prelabeling
PKH67
PKH67
PKH67
PKH67
PKH26
PKH67
PKH67
PKH67
PKH26
PKH67
PKH67
PKH26
Survival
time
(green)
30 Minutes
(green)
1 Week
(green)
1 Week
(green) ⫹ DAPI 1 Week
(red) ⫹ DAPI
1 Week
(green)
2 Weeks
(green)
2 Weeks
(green) ⫹ DAPI 2 Weeks
(red) ⫹ DAPI
2 Weeks
(green)
5 Weeks
(green) ⫹ DAPI 5 Weeks
(red) ⫹ DAPI
5 Weeks
Retinal
treatment
Cells
Laser
Laser
Laser
Laser
Laser
Laser
Laser
Laser
Laser
Laser
Laser
Laser
⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹
⫹⫹
⫺
Figure 1. Cell cultures from the fetal porcine neural retina. Cells grew
as suspended aggregates (spheres) in uncoated tissue culture flasks (A)
or as an adherent monolayer on substrates such as fibronectin (B) or
laminin. Magnification, ⫻100 (A), ⫻200 (B).
Abbreviation: DAPI, 4⬘,6-diamidino-2-phenylindole.
Association for Research in Vision and Ophthalmology statement
for the Use of Animals in Ophthalmic and Vision Research.
Tissue Processing
Eyes were enucleated under anesthesia at 30 minutes (n ⫽ 1), 1
week (n ⫽ 4), 2 weeks (n ⫽ 4), and 5 weeks (n ⫽ 3) posttransplantation (Table 4). Following removal of globes, pigs were
sacrificed by i.v. injection of 2– 4 g of pentobarbital (200 mg/ml;
KVL, Copenhagen, Denmark, http://www.kvl.dk). Intact globes
were placed in 4% paraformaldehyde (PFA) for 10 –20 minutes. For
each eye, the anterior segment, with lens, was then removed, and the
posterior segment was postfixed for 2 hours in 4% PFA, followed
by rinsing in increasing concentrations of sucrose containing Sörensen’s phosphate buffer. A horizontal slice was made from the
temporal retinal margin to 2–3 mm nasal to the optic disc, thus
comprising the temporal ciliary margin, the area centralis, and the
optic disc. These tissues were embedded in a gelatin medium, and
a series of 12-␮m sections were cut on a cryostat. Every 10th slide
was stained with H&E.
Immunohistochemistry
Ocular tissue sections were exposed to primary anti-sera (Table 1)
for 16 –18 hours in a moist chamber at 4°C, followed by rinsing in
0.1 M PBS with 0.25% Triton X-100. Tissue sections were then
incubated with secondary fluorescein isothiocyanate (FITC) or
Texas Red-conjugated antibodies (1:200; Jackson Immunoresearch
Laboratories) for 1–2 hours in the dark at room temperature. The
nonoperated contralateral eyes served as normal controls. There
were additional negative controls from the operated eyes in which
the primary antisera were omitted. Specimens were examined using
an epifluorescence microscope. Colocalization of FITC or Texas
Red-labeled primary antibodies and DAPI⫹ cells was assessed by
superimposition of separate digital images of each fluorochrome.
RESULTS
Cell Isolation and Culture
With short-term exposure to serum, isolated fetal porcine
retinal cells quickly adhered to fibronectin or laminin substrates and remained adherent after FBS was removed. Adherent cultures showed high short-term viability, as evidenced by a relative paucity of suspended cells or debris.
There was morphological evidence of active proliferation,
seen as numerous dividing profiles and rapid increase in the
density of the monolayer. Cellular morphologies were similar
to those observed in other neural progenitor cultures (Fig.
1A, 1B) and showed features in common with cultured progenitors from immature porcine forebrain [16] and human
www.StemCells.com
Figure 2. Marker expression by cultured porcine retinal progenitor
cells (pRPCs). (A): Cultured pRPCs showed widespread expression of
the surface marker NCAM (green) and more restricted expression of the
surface marker CD15 (red). (B): Expression of the neuroblast marker
doublecortin (red) was restricted to a subset of small round cells, seen
here against low-level labeling for glial fibrillary acidic protein (GFAP)
(green). (C): Other markers showing restricted expression included the
early neuronal marker ␤-III tubulin (red) and the retinal marker recoverin (green). There was widespread nuclear expression of the transcription factor Sox2 (red), as well as widespread cytoplasmic expression the
intermediate filament vimentin (blue), whereas the intermediate filament GFAP (green) (D) showed limited expression. (E): The bipolar
cell marker protein kinase C␣ (green) was also expressed by a subset of
cells that differed from those expressing the photoreceptor marker
recoverin (red). (F): The intermediate filament nestin (red) showed
cytoplasmic localization in cells counterstained with the nuclear marker
4⬘,6-diamidino-2-phenylindole (green). Magnification, ⫻400 (A–E),
⫻200 (F).
retina [18]. In the absence of substrate, many cells remained
suspended and via proliferation quickly generated the classic
cellular clusters commonly referred to as neurospheres (Fig.
1A). Qualitative comparison of porcine retinal cultures suggested similar short-term survival when grown in growth
media containing EGF, bFGF, or both.
1226
Porcine Retinal Progenitor Cells
Figure 4. Immunoblot analysis of E60 fetal porcine retina versus
cultured porcine RPCs. GFAP was clearly expressed in developing
retinal tissue but only faintly detected in RPCs, whereas recoverin and
PKC␣ were expressed in both. Abbreviations: GFAP, glial fibrillary
acidic protein; PKC, protein kinase C; RPC, retinal progenitor cell.
Figure 3. Colocalization of rhodopsin and recoverin in porcine retinal
progenitor cells (pRPCs). Differentiating pRPCs expressed the photoreceptor markers recoverin (green) (A) and rhodopsin (red) (B), with
co-expression of the two markers seen in a subset of cells (C). Cells
co-expressing rhodopsin and recoverin frequently exhibited morphologies suggestive of rod photoreceptors.
Marker Expression in Culture
Porcine RPC cultures exhibited widespread expression of a
number of immature markers by immunocytochemistry (Fig.
2), including the nuclear transcription factor Sox2 (Fig. 2D),
as well as the proliferation marker Ki67 (data not shown),
along with subpopulations of cells expressing the surface
carbohydrate moiety CD15 (LeX) (Fig. 2A) and the cytoskeletal-associated proteins doublecortin (DCX) (Fig. 2A), GFAP
(Fig. 2B), ␤-III tubulin (Fig. 2C), and vimentin (Fig. 2D). We
previously reported difficulties demonstrating nestin expression in porcine forebrain progenitors using a number of
antibodies [16]. We now demonstrate widespread nestin labeling using a different anti-nestin antibody (BD 611658;
Fig. 2F). Cells within the cultures were broadly positive for
more mature markers of neural lineage, particularly the adhesion molecule NCAM (Fig. 2A), and many cells were
positive for the cytoskeletal protein GFAP (Fig. 2B), as we
have seen previously in progenitor cultures from human
forebrain [20], human retina [18], and porcine forebrain [16].
Distinct clusters of cells expressed the retinal markers recoverin, as seen previously with human RPCs [18]. Furthermore,
although some cells expressed PKC␣, consistent with bipolar
cell differentiation, there was no evidence of double labeling
for recoverin and PKC␣, as would be expected if the recoverin labeling was from bipolar cells (Fig. 2E). Estimated
marker expression as a percentage of cells in culture was as
follows: CD15, 20%; NCAM, 90%; DCX, 11%; GFAP, 27%;
␤-III tubulin, 9%; recoverin, 36%; Sox2, ⬎95%; vimentin,
⬎95%; PKC␣, 12%; nestin, ⬎95%.
Although recoverin is known to be expressed in vivo by a
subset of rod bipolar cells, the double labeling for recoverin and
Figure 5. RT-PCR analysis of gene expression by retinal progenitor
cells (RPCs). Primers designed for human genes were used to probe
RNA extracted from cultured porcine RPCs. Evidence was found for
expression of nestin (faint), GFAP, Lhx2, Hes1, Dach1, DCX, Six3,
Six6, Sox2, and Pax6 (faint). Alternating lanes contain sample with (⫹)
and without (⫺) reverse transcriptase, the latter serving as negative
control. Ladders (100 base pairs) provided for reference. Abbreviations:
DCX, doublecortin; GFAP, glial fibrillary acidic protein; RT-PCR,
reverse transcription-polymerase chain reaction.
rhodopsin shown here under differentiation conditions is consistent with rod photoreceptors, and the morphology of these
cells is suggestive of photoreceptors as well (Fig. 3).
Immunoblot data comparing porcine retinal tissue to cultured porcine RPCs confirmed the expression of recoverin and
PKC␣ by cells from both populations, (Fig. 4). GFAP was
clearly expressed in retinal tissue but only faintly detected in
RPCs.
Klassen, Kiilgaard, Zahir et al.
1227
Figure 6. Transplantation of porcine retinal progenitor cells (pRPCs) to the subretinal space of allogeneic recipients. Cultured pRPCs were prelabeled
with 4⬘,6-diamidino-2-phenylindole (DAPI) (blue) prior to transplantation. DAPI is seen in the left panel of each row. The middle panel in each row
shows immunoreactivity for a particular marker. The right panel in each row is a merged image of the two panels to the left. Following transplantation,
donor cells (blue) co-expressed the photoreceptor marker transducin (red) ([A]; higher power, [B]). Similarly, donor cells co-expressed the rod
photoreceptor marker rhodopsin (red) ([C]; higher power, [D]), as well as recoverin (green) ([E]; higher power, [F]). A subset of cells expressed the
intermediate filament GFAP (green) ([G]; higher power, [H]). Superficial as well as deep, radially oriented GFAP⫹ host profiles were evident (G),
and colocalization of DAPI and GFAP was also seen (arrows) (H). Scale bars ⫽ 100 ␮m. Abbreviations: GCL, ganglion cell layer; INL, inner nuclear
layer; IPL, inner plexiform layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.
To further characterize porcine RPCs, extracted RNA was
analyzed using reverse transcription-PCR (Fig. 5). The product
for the human nestin transcript was faint, as previously reported
www.StemCells.com
for porcine forebrain progenitors using these primers [16]. Other
genes associated with the developing retina were also examined,
with particular emphasis on nuclear transcription factors
Porcine Retinal Progenitor Cells
1228
Figure 6. (Continued)
(NTFs). The finding of Sox2 expression, seen above using
immunocytochemistry (ICC), was replicated at the transcript
level. Hes1, another NTF expressed during neural development,
was also present. In addition, a number of NTFs previously
associated with eye specification and retinal development in
vertebrates were identified in porcine RPCs. These include
Lhx2, Six3, Six6, Dach1, and Pax6, with the signal for Pax6
being the weakest among these (Fig. 5). In terms of non-NTF
genes, transcripts for the intermediate filament GFAP and the
neuroblast-associated protein DCX were evident as well (Fig.
4), again replicating findings seen by ICC.
Transplantation
Prelabeled donor cells were identified in 11 of 12 allograft
recipients (Table 4). There was evidence of graft survival at all
time points from 30 minutes to 5 weeks. Survival was quite
substantial in all animals at 1 and 2 weeks but variable at 5
weeks. These cells were frequently located in the subretinal
space (Fig. 6A– 6H), although there was also evidence of labeled cells and processes within the neural retina (supplemental
online Fig. 7). In terms of evidence for migration, the lateral
limit for labeled cells was 4 mm from the injection bleb. Despite
evidence of some diffusion of DAPI and uptake by host retinal
cells, the bulk of cells in the subretinal space exhibited brighter
DAPI labeling and could be readily distinguished from the
well-organized cells of the host. Many of these subretinal (presumed donor) cells double-labeled for DAPI and photoreceptorassociated markers, including transducin (Fig. 6A, 6B), rhodopsin (Fig. 6C, 6D), and recoverin (Fig. 6E, 6F). A subset of cells
were found to express GFAP (Fig. 6G, 6H). No evidence of
perivascular cuffing or other signs of PMN or mononuclear cell
infiltration were observed in H&E-stained sections (data not
shown).
DISCUSSION
In this study, we show that progenitor cells can be cultured from
the retina of the pig, the first demonstration of ex vivo cultivation of retinal progenitor cells from a nonhuman large animal.
The pRPCs proliferate in vitro and express NTFs implicated in
retinal specification, as well as other markers generally associated with CNS development. In the presence of serum, a subset
of pRPCs co-express recoverin and rhodopsin, evidence of
differentiation into cells of retinal lineage, namely, rod photoreceptors. The availability of porcine RPCs allows us to compare the characteristic of these cells to RPCs from rodents and
Figure 7. Donor cells prelabeled with both 4⬘,6-diamidino-2-phenylindole (DAPI, blue) and PKH26 (red) showed extension of donor-derived
processes into the inner retina (arrows). Abbreviations: INL, inner
nuclear layer; IPL, inner plexiform layer.
humans. Potential points of comparison include isolation and
culture techniques, cellular morphology, and gene expression.
The paucity of porcine-specific reagents currently limits our
ability to evaluate gene expression in cells from this species. A
gene not well detected here, but of interest to the current study,
is the intermediate filament nestin, a marker highly expressed by
neuroepithelial cells during development. Nevertheless, the expression of many porcine genes can be detected with available
reagents, presumably as a result of sufficient homology with
other mammalian species, particularly human, as we have
shown previously [16] and both confirm and extend here. Therefore, despite difficulties with detection of nestin in porcine cells,
we have been able to collect a range of markers that highlight
the similarity of these cells to retinal progenitors from other
species.
The porcine RPCs exhibit a number of similarities to forebrain progenitors from the pig, as well as forebrain and retinal
progenitors from other species. The proliferation marker Ki-67
and adhesion molecule NCAM are heavily expressed across
species by a broad range of CNS progenitors, as well as other
cell types. The surface carbohydrate epitope CD15 (LeX) and
highly conserved transcription factors Hes1 and Sox2 have been
more specifically associated with immature CNS progenitors in
Klassen, Kiilgaard, Zahir et al.
a number of mammalian species, including mice, humans, and
pigs [16, 21–24].
Markers of neuronal development were expressed by
subpopulations of porcine progenitor cells, including DCX
and ␤-III tubulin. We have previously identified both of these
cytoplasmic markers in subpopulations of cultured CNS progenitors from mouse [10], human [18, 20], and pig [16].
Here, we also find that subpopulations of pRPCs also express
the glial-associated markers vimentin and GFAP. Although
we have not seen substantial GFAP expression in mouse
retinal progenitor cultures maintained under proliferation
conditions [10], the present data are consistent with our
previous findings of GFAP and vimentin in human forebrain
[20] and retinal [18] progenitors, as well as progenitors from
the pig forebrain [16]. Hence, although there is considerable
overlap in the genes expressed by CNS progenitor cells, the
evidence to date supports the concept that the expression
profiles of porcine CNS progenitors more closely resemble
those of the human than of mouse.
Compared with cells from the developing pig forebrain
[16], progenitor cells from the pig retina differ by way of
their expression of the photoreceptor-associated gene recoverin. After differentiation or transplantation, the additional
photoreceptor markers rhodopsin and transducin are upregulated as well. Furthermore, pRPCs express highly conserved
genes associated with specification of the eye and retina in a
wide range of metazoan species, from fly to human. Here,
these were found to include Dach1, Lhx2, Pax6, Six3, and
Six6, both confirming and extending our previous findings
with respect to these genes in analogous cells from human
forebrain and retina [18]. New to the present study is the
identification of Lhx2 transcripts in cultured RPCs. Lhx2 is a
transcription factor of the Lim family that has been implicated in eye specification in Xenopus [25, 26] and is expressed in the forebrain, optic vesicle, and developing neural
retina of the mouse [27, 28]. Here, we show the expression of
this gene by cultured RPCs from a large mammal. Although
Lhx2-deficient mouse embryos are anophthalmic, the role
played by this gene later in retinal development remains to be
determined. Since this gene might be involved in the phenotypic specification of retinal cells, it would be of interest to
know the relative co-expression of Lhx2 and other NTFs, as
well as non-nuclear genes, within cultured RPC populations.
In previous work in porcine recipients, we showed that
cultured murine RPCs can be transplanted to the subretinal
space and that these cells show evidence of morphological
integration into the neural retina and RPE layers [19], before
being destroyed by a vigorous immune response, characterized
by dense choroidal infiltrates localized to the region of the
xenogeneic grafts [29]. These findings serve as a point of
comparison with respect to the present study, in which we again
observed integration of grafted RPCs into the retina, but also
evidence of photoreceptor differentiation and improved graft
survival. Here, characterization of donor cell morphology demonstrated extension of processes into the host retina but was
limited by the use of vital dyes, which yielded notably inferior
results compared with endogenous green fluorescent protein
(GFP) expression. The development of a GFP-transgenic pig
provides a potential solution to this problem [30] and we are
currently developing GFP⫹ donor cells from this source (H.
Klassen, M. Young, R. Prather, unpublished data). Our current
finding of widespread expression of recoverin, rhodopsin, and
transducin by grafted cells suggests a relatively advanced stage
of photoreceptor development. Whether these cells possess
outer segments or are capable of responding to light remains to
be determined.
www.StemCells.com
1229
The rod-like morphology seen in the present study, but not
in previous work by Banin et al. [31], could relate to the use of
embryonic stem (ES) cells by the latter investigators. ES cells
require greater epigenetic specification than RPCs to differentiate into mature photoreceptors. In addition, those investigators
used a xenogeneic transplantation model (human to rat) that
likely provides fewer instructive cues compared with the allogeneic porcine model used here.
In terms of donor cell survival, the porcine RPCs used here
proved much superior to the murine cells used previously. This
can in large part be attributed to the improved immunological
compatibility of the allogeneic pRPCs over the xenogeneic
murine cells. Yet even in the absence of an obvious immune
response, survival of grafted cells can be low following progenitor cell transplantation [32], and therefore the degree of survival
seen here is encouraging. The reason for the variable survival
seen at 5 weeks, the longest time point examined, is not clear,
and the possibility of immunological attack has not been ruled
out.
CONCLUSION
Further development of the porcine RPC allograft model will
benefit from the use of cells constitutively expressing a
reporter gene to positively identify cells of donor origin and
to better evaluate the morphology of cells following engraftment. Also of interest is the application of such cells in
porcine models of retinal disease, such as the pig that develops photoreceptor degeneration secondary to expression of a
mutant human rhodopsin transgene [33]. It has been proposed
that porcine cells might provide a source of xenograft material for human application; however, it is evident that substantial biological challenges, including immunological incompatibility, would first need to be overcome to make this
a realistic option. Nevertheless, the porcine RPC allograft
model represents a useful tool for evaluating issues likely to
be faced if intraocular progenitor transplantation is to find
application in humans.
ACKNOWLEDGMENTS
We thank Dr. Marie Shatos, Jaqueline Doherty, Caijui Zhang,
and Hubert Nethercott for technical assistance; Prof. Robert
Molday for the kind gift of anti-rhodopsin antibody; and Prof.
Berndt Ehinger, Dr. Morten la Cour, and Dr. Phil Schwartz
for intellectual input related to this project. This work was
supported by the Gail and Richard Siegal Foundation, National Institute of Neurological Disorders and Stroke
NS044060 (H.K.), National Eye Institute EY09595 (M.J.Y.),
Larry Hoag Foundation (H.K.), BMRC Grant 05/1/35/19/421
(H.K.), the Crown Princess Margareta’s Committee for the
Blind (K.W.), the Swedish Association of the Visually Impaired (K.W.), the Swedish Science Council (Medicine)
(K.W.), and the Minda de Gunzburg Research Center for
Retinal Transplantation (M.J.Y.).
DISCLOSURES
OF
OF
POTENTIAL CONFLICTS
INTEREST
The authors indicate no potential conflicts of interest.
Porcine Retinal Progenitor Cells
1230
REFERENCES
18
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Alvarez-Buylla A, Herrera D, Wichterle H. The subventricular zone:
Source of neuronal precursors for brain repair. Prog Brain Res 2000;127:
1–11.
Anthony TE. Radial glia serve as neuronal progenitors in all regions of
the central nervous system. Neuron 2004;41:881– 890.
Yoon K, Nery S, Rutlin M et al. Fibroblast growth factor receptor
signaling promotes radial glial identity and interacts with Notch1 signaling in telencephalic progenitors. J Neurosci 2004;24:9497–9506.
Brüstle O, Jones K, Learish R et al. Embryonic stem cell-derived glial
precursors: A source of myelinating transplants. Science 1999;285:
754 –756.
Lie D, Song J, Colamarino S et al. Neurogenesis in the adult brain: New
strategies for central nervous system diseases. Annu Rev Pharmacol
Toxicol 2004;44:399 – 421.
McDonald J, Becker D, Holekamp T et al. Repair of the injured spinal
cord and the potential of embryonic stem cell transplantation. J Neurotrauma 2004;21:383–393.
Davies J, Huang C, Proschel C et al. Astrocytes derived from glialrestricted precursors promote spinal cord repair. J Biol 2006;5:7.
Kim J, Wu H, Lander A et al. GDF11 controls the timing of progenitor
cell competence in developing retina. Science 2005;308:1927–1930.
Young M, Ray J, Whiteley S et al. Neuronal differentiation and morphological integration of hippocampal progenitor cells transplanted to the
retina of immature and mature dystrophic rats. Mol Cell Neurosci 2000;
16:197–205.
Klassen H, Ng T, Kurimoto Y et al. Multipotent retinal progenitors
express developmental markers, differentiate into retinal neurons, and
preserve light-mediated behavior. Invest Ophthalmol Vis Sci 2004;45:
4167– 4173.
Wojciechowski A, Englund U, Lundberg C et al. Survival and long
distance migration of brain-derived precursor cells transplanted to adult
rat retina. STEM CELLS 2004;22:23–38.
Lavik E, Klassen H, Warfvinge K et al. Fabrication of degradable
polymer scaffolds to direct the integration and differentiation of retinal
progenitors. Biomaterials 2005;26:3187–3196.
Del Priore L, Kaplan H, Hornbeck R et al. Retinal pigment epithelial
debridement as a model for the pathogenesis and treatment of macular
degeneration. Am J Ophthalmol 1996;122:629 – 643.
Nicolini J, Kiilgaard J, Wiencke A et al. The anterior lens capsule used
as support material in RPE cell-transplantation. Acta Ophthalmol Scand
2000;78:527–531.
Ghosh F, Wong F, Johansson K et al. Transplantation of full-thickness
retina in the rhodopsin transgenic pig. Retina 2004;24:98 –109.
Schwartz P, Nethercott H, Kirov I et al. Expression of neurodevelopmental markers by cultured porcine neural precursor cells. STEM CELLS
2005;23:1286 –1294.
Kelley M, Turner J, Reh T. Regulation of proliferation and photoreceptor
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
differentiation in fetal human retinal cell cultures. Invest Ophthalmol Vis
Sci 1995;36:1280 –1289.
Klassen H, Ziaeian B, Kirov I et al. Isolation of retinal progenitor cells
from post-mortem human tissue and comparison with autologous brain
progenitors. J Neurosci Res 2004;77:334 –343.
Warfvinge K, Kiilgaard J, Lavik E et al. Retinal progenitor cell xenografts to the pig retina: Morphologic integration and cytochemical differentiation. Arch Ophthalmol 2005;123:1385–1393.
Schwartz P, Bryant P, Fuja T et al. Isolation and characterization of
neural progenitor cells from post-mortem human cortex. J Neurosci Res
2003;74:838 – 851.
Allendoerfer K, Durairaj A, Matthews G et al. Morphological domains of
Lewis-X/FORSE-1 immunolabeling in the embryonic neural tube are due
to developmental regulation of cell surface carbohydrate expression. Dev
Biol 1999;211:208 –219.
Klassen H, Schwartz M, Bailey A et al. Surface markers expressed by
multipotent human and mouse neural progenitor cells include tetraspanins and non-protein epitopes. Neurosci Lett 2001;312:180 –182.
Cai J, Wu Y, Mirua T et al. Properties of a fetal multipotent neural stem
cell (NEP cell). Dev Biol 2002;251:221–240.
Capela A, Temple S. LeX is expressed by principle progenitor cells in the
embryonic nervous system, is secreted into their environment and binds
Wnt-1. Dev Biol 2006;291:300 –313.
Zuber M, Gestri G, Viczian A et al. Specification of the vertebrate eye by
a network of eye field transcription factors. Development 2003;130:
5155–5167.
Viczian A, Bang A, Harris W et al. Expression of Xenopus laevis Lhx2
during eye development and evidence for divergent expression among
vertebrates. Dev Dyn 2006;235:1133–1141.
Porter F, Drago J, Xu Y et al. Lhx2, a LIM homeobox gene, is required
for eye, forebrain, and definitive erythrocyte development. Development
1997;124:2935–2944.
Bernier G, Vukovich W, Neidhardt L et al. Isolation and characterization
of a downstream target of Pax6 in the mammalian retinal primordium.
Development 2001;128:3987–3994.
Warfvinge K, Kiilgaard J, Klassen H et al. Retinal progenitor cell
xenografts to the pig retina: immunological reactions. Cell Transplant
2006;15:603– 612.
Lai L, Park K, Cheong H et al. Transgenic pig expressing the enhanced
green fluorescent protein produced by nuclear transfer using colchicine-treated fibroblasts as donor cells. Mol Reprod Dev 2002;62:
300 –306.
Banin E, Obolensky A, Idelson M et al. Retinal incorporation and
differentiation of neural precursors derived from human embryonic stem
cells. STEM CELLS 2006;24:246 –257.
Tomita M, Lavik E, Klassen H et al. Biodegradable polymer composite
grafts promote the survival and differentiation of retinal progenitor cells.
STEM CELLS 2005;23:1579 –1588.
Petters R, Alexander C, Wells K et al. Genetically engineered large
animal model for studying cone photoreceptor survival and degeneration
in retinitis pigmentosa. Nat Biotechnol 1997;15:965–970.
See www.StemCells.com for supplemental material available online.