In stem cell line has properties similar to those

In order to develop new therapies for retinal degenerative diseases,
replacing damaged tissue with functional retinal cells becomes a crucial step.
Two main sources for such cells include the RPE cells to prevent degeneration
of photoreceptors, and the photoreceptor precursors to repair the degenerating
neural retina. Approaches to use retinal pigment epithelium as a therapeutic tool
for degenerating retina began somewhere in the 1980s when Parysek et al. (1983)
transplanted strips of retinal tissue with the intact RPE still attached. The
first successful transplantation of RPE cells into the sub-retinal space in
monkeys was accomplished by Gouras
et al. (1985). RPE cell replacement in AMD has been pursued since then
and its beneficial effects have been demonstrated for dry AMD (Schwartz et al.,

In this study, we have used hESC line, Relicell®hES1
which was the first hESC line derived and characterized from the Indian subcontinent
(Mandal et al., 2006).
This blastocyst-derived pluripotent stem cell line has properties similar to
those described by some other groups (Heins et al., 2004; Simón et al., 2005; Oh et al.,
2005). Relicell®hES1 was derived and cultured on MEFs
and extensively characterized for their pluripotency (Mandal et al., 2006). Later, these hESC
lines were adapted to feeder-free culture systems and subsequently
re-characterized (data not shown). The change in culturing practice did not
alter the pluripotency of these cells; which has also been reported previously (Mehta et al., 2010).

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

One of the first reports of in
vitro differentiation of RPE-like cells, published in 2002, was by
co-culturing monkey ES cells with the stromal cell line PA6 (Kawasaki et al.,
2002). In the last few years, various studies have shown successful
generation of RPE cells in vitro. In
addition, different culture conditions were employed to improve the efficiency
from about 1% to between 38-70% of RPE-like cells derived in vitro from pluripotent stem cells (Idelson et al., 2009; Meyer et al., 2009; Osakada et al.,
2009a; Klimanskaya et al., 2004; Osakada et al., 2009b). In previously
published studies, RPE differentiation of hESCs required incubation with
undefined components and/ or FBS (Meyer et al., 2009; Hirami et al., 2009). A drawback of using
serum in culture media is the introduction of ambiguous signalling molecules
including hormones and cytokines, and other known and unknown factors (Tian et al., 2005; Tezel and Del Priore, 1998; Shah, 1999). Efforts have
been made with limited success to move towards xeno-free defined protocols for
RPE differentiation of pluripotent stem cell lines (Vaajasaari et al., 2011).

The present study establishes for the first time that it is possible
to differentiate the Relicell®hES1 into RPE-like cells in vitro using simple feeder-free
conditions employing only defined and/ or recombinant DNA-derived products. Our
protocol for RPE derivation in feeder-free conditions differs from previously
published procedures that show spontaneous differentiation into RPE fate after withdrawal
of bFGF (Klimanskaya
et al., 2004). It is interesting to see this occurrence but difficult to
understand why this happens. In fact, we too have observed this typical
phenomenon during maturation of ExMat-RPE
cells from expanded hESC-RPE.

For effective differentiation many growth supplements have been used
either alone or in combination. Two such growth supplements, B27 and N2, have
shown to promote expansion of embryonic and adult neural stem cells in
serum-free neurosphere cultures (Wachs et al., 2003; Svendsen et al., 1995).
Further, various investigators have used different culture supplements such as
Dkk-1, noggin, NIC, activin A or transforming growth factor beta (TGF-?) to
promote RPE differentiation. However, their reported yield of RPE cells after
4–8 weeks of differentiation was around 1%, and very few EBs contained
pigmented cells (Klimanskaya
et al., 2004). Recent studies with Wnt and Nodal antagonists have
certainly improved the yield of pigmented RPE cells within the hESC colonies
following 8 weeks of differentiation (Osakada et al., 2008). Recently, Idelson et al. (2009)
reported that directed differentiation of hESCs, grown on human foreskin
fibroblasts feeders, to RPE cells, under defined conditions improved efficiency
of cell clusters containing pigmented areas after 6 to 8 weeks of

Based on published literature, we attempted to develop a culturing
system that was free from serum and feeder cells. In our protocol, medium was
supplemented with serum substitute such as B27. Additionally, instead of
providing the cells with more growth supplements, we cultured the cells in
conditioned medium from ARPE?19 cells and observed that about 30-40% of the
culture dish area was covered with pigmented RPE-like cell clusters after 6-8
weeks of differentiation. This was confirmed by the expression of markers
during the various stages of differentiation (neuroectoderm, eye-field,
optic-vesicle, early and late RPE) at molecular level (Fig. 3). Human ESC
differentiation on MatrigelTM initially and then on Laminin or on
MatrigelTM alone resulted in similar numbers of pigmented clusters (Fig.
2), which is also reported recently (Rowland et al., 2012). Similar results were obtained when we used
another hESC line, Relicell®hES4, in our study (data not shown).
Thus, this approach of using ARPE-19-conditioned medium has given us an
efficient method of RPE differentiation in
vitro. This simple protocol has consistently given us the desired outcome.
We show that the pigmented clusters of RPE cells at an early (day 59)
differentiation stage acquired characteristic RPE polygonal morphology and
dark-brown pigmentation (Fig. 2A), which is very similar to previously
published hESC?RPE cells (Klimanskaya et al., 2004). Our results
demonstrated that the appearance of the first pigmented cells was seen, between
day 45 and day 49 in culture, and pigmentation increased by day 100 of
differentiation (Fig. 2B). Most of the published reports, in comparison,
describe the appearance of the first pigmented cells normally around 4 to 8
weeks (Carr et al., 2009; Klimanskaya et al., 2004). During the course
of differentiation, the cells expressed early eye-field marker, Pax6, and
maintained this expression in the cell population over extended period while
early neural-retina marker, Rx, showed very faint expression by day 14 and
disappeared thereafter. The pluripotent stem cell marker, Oct-4 was expressed
in the undifferentiated hESCs which disappeared by day 14.

During embryonic development, once the optic vesicles evaginate from
the paired eye fields, expression of MiTF occurs in all the cells fated to
become the retina (Bharti et al., 2008; Chow et al., 2001). However, the switch to
differentiate into either RPE or neural retina is made during the late optic
vesicle and optic cup stages through differential expression of Chx10 (Horsford et al.,
2005; Rowan et al.,
2004). Neural retinal progenitors envisioned for the inner layer of the
optic cup remain Chx10+ and MiTF- (Rowan et al., 2004). On the
contrary, cells intended for the outer layer of the optic cup express MiTF and
down-regulate Chx10 and subsequently differentiate into RPE cell fate. In our experiments,
expression of Chx10 was seen by day 14 and up to day 28, after which it
decreased. Early RPE marker, MiTF, was detected by day 14 and the expression
increased during differentiation suggesting that the cells were directed
towards RPE cell fate (Fig. 3A). Further, it has been shown that the
expression of mature-RPE marker, RPE65, essential for the regeneration of the
visual pigment, is typically lost in cultured RPE cells (Idelson et al., 2009; Vugler et al., 2008). In our study, the
mature RPE cell-specific markers, RPE65, CRALBP1, and Bestrophin, were detected
for the first time at day 60 and the expression of RPE65 increased by day 100.
In addition, other RPE cell markers, Otx2, PEDF, PMEL17, MERTK, and VEGF-A,
were also expressed relatively early during the differentiation, with TYRP1
being expressed by day 60. These collectively suggest that the differentiated
cells closely resemble the native RPE cells. The mesodermal marker, Brachyury,
and the mesendodermal marker, GATA-4, were not expressed during the stages of
differentiation. In addition, the early and late markers of photoreceptor
differentiation, CRX, Recoverin, Opsin, and Rhodopsin, were absent in cells
derived from Relicell®hES1 strongly indicating the RPE cell
fate of the derived cells (Fig. 3B).

In addition, transmission electron microscopy was carried out to
confirm the ultra-structural characteristics of RPE cells. Human ESC-derived
RPE clusters showed prominent nuclei, tight-junction complexes, presence of
apical microvilli, and abundant melanin-producing granules in the cytoplasm (Fig.
4) – all characteristics of retinal morphology.

We have so far been successful in deriving clusters of RPE cells in
culture. Now the major goal was to generate sufficient number of cells, as
required, for clinical use. For this, we needed to expand these cells. We tried
different methods to dissociate the RPE clusters and expand these cells in vitro (Supplementary
information, Table S2). After mechanical transfer, the cells did not grow as
explant cultures. When treated with EDTA, trypsin, TrypLETM, the
cells did not detach at all. In the case of AccutaseTM treatment,
they detached and dispersed, but they did not form the typical hexagonal
morphology, nor did they divide. Not many investigators have sought to expand
differentiated RPE cells for clinical use. In our attempts to dissociate the
RPE clusters, it was observed that after trypsinization with either trypsin or
TrypLETM, the RPE clusters remained attached while the surrounding
cells lifted off. These RPE clusters when maintained in culture for periods
ranging from 7days to 1 year remained quiescent and retained their typical
hexagonal morphology. Treatment of these clusters with serum and cocktail of
growth factors triggered budding of cells and expansion to form monolayer of
RPE cells. This was the first time we could expand RPE cells from
differentiated clusters. However, these expanded cells had lost the typical RPE
morphology and pigmentation as well. The phenomenon of loss of hexagonal cell morphology
and pigmentation when adult RPE cells are grown in culture has also been
reported in literature (Opas, 1994). One of the major concerns after in vitro expansion was the identity
and stability of the expanded cells. Karyotype analysis of the expanded
hESC-derived RPE cells confirmed that there was no chromosomal abnormality
induced during expansion, showing genetic stability (Supplementary information,
Fig. S2).

While we could generate hESC-derived RPE cells, the question of
their authenticity was still debatable. RPE cells are characterized by the four
‘P’s: Polygonal, Pigmented, Phagocytic, and Polarized; we carried out tests to
ascertain that our cells exhibited all these characteristics. At this stage,
the question of their identity was debatable as the characteristic hexagonal
morphology and pigmentation associated with RPE cells was not seen in the
expanded cells even after becoming confluent. It is known that the growth
factors in the medium could be preventing the progression towards mature RPE
stage. Indeed, subsequent removal of growth factors resulted in the cells
regaining the typical hexagonal morphology as early as day 7 (Fig. 5) and
pigmentation over a period of 15-30 days. In addition, comparison of the
pigmentation in hESC-derived RPE cells cultured on plastic surface and MatrigelTM
showed a marked increase in the levels of pigmentation in cells cultured on
MatrigelTM at day 21 (Fig. 6). This confirmed that we were able to
induce expanded cells to re-establish two typical cellular characteristics of

ExMat-RPE cells were first characterized at the molecular level. Gene
expression profile of these cells showed expression of RPE-specific markers,
Pax6, MiTF, RPE65, and CRALBP1. In the present study, the mRNA expression of
bestrophin was clearly upregulated in ExMat-RPE
cells compared to the expanded RPE cells. In addition, we also observed
expression of melanogenic marker, TYRP1, in ExMat?RPE
cells. Other RPE cell markers such as MERTK and PMEL17 showed higher expression
in ExMat?RPE cells. In addition, the
pluripotent stem cell marker Oct-4 was not detected in the ExMat-RPE cells and was confirmed at the molecular and cellular level
(Fig. 7, 9). The early and late markers of photoreceptor differentiation, CRX,
Recoverin, Opsin, and Rhodopsin, were not expressed in these cells (Fig. 7). There
was high protein expression of the early neural and eye-field markers, Nestin,
Pax6, and MiTF; mature-RPE markers,
RPE65, CRALBP1, Bestrophin, CK18, and the tight-junction protein, ZO-1 (Fig.
8). Importantly, pigmented ExMat?RPE
cells expressed weak levels of pluripotent stem cell marker, SSEA?4 (Fig. 9).

Functional analysis of the ExMat-RPE
cells was confirmed using transmission electron microscopy. We observed a
monolayer of RPE cells in which cell-cell contacts were noticed. We also
observed that ExMat-RPE cells
exhibited ultrastructural epithelial features such as presence of microvilli at
the apical membrane, abundant melanosomes, and tight junctions between the
cells suggesting typical features of native RPE (Fig. 10).

Of particular relevance to the pathogenesis of macular degeneration
is the ability of RPE cells to engulf and degrade the photoreceptor outer
segments that are shed every day (Rattner and Nathans, 2006). Photoreceptor cells go
through the renewal process every day and RPE cells take charge of disposing
the generated waste by phagocytosing photoreceptor outer segments in vivo. Therefore, it is crucial
for ExMat-RPE cells to exhibit
phagocytic activity in vitro to
confirm that these cells have not lost this functionality in the process of
expansion and maturation. Recently, Schwartz et al. (2012) reported the in vitro phagocytic properties of hESC-RPE cells by using
fluorescently labeled pHrodoTM BioParticles®. In our
study, ExMat?RPE cells were able to
phagocytose the fluorescently labeled BioParticles® in vitro, thus demonstrating their
functionality in vitro (Fig. 11A).
In the same study, we observed no evidence of engulfment of BioParticles®
by UCMSCs (Fig. 11B).

Polarization is a critical feature of the RPE monolayer. Normally,
basal secretion of VEGF-A from the RPE provides constitutive support for the
maintenance of the choriocapillaris, while secretion of PEDF into the
interphotoreceptor matrix provides neurotrophic support for the photoreceptors.
We examined the ability of ExMat-RPE
cells to preferentially secrete VEGF?A to the basolateral side and PEDF to the
apical side. We found notable difference in the secretion of VEGF-A at the
basal side, while no significant difference in the secretion pattern of PEDF at
either side was noted (Fig. 12).

Human ESCs offer the advantage of providing an unlimited number of
healthy young RPE cells with potentially reduced immunogenicity (Drukker et al., 2006).
The preliminary report of the first clinical transplantation of hESC-derived RPE
cells demonstrated that these cells showed no signs of rejection or tumour
formation four months post-transplantation. This study was a major step in
retinal regenerative medicine describing the safety of injecting hESC-derived RPE
cells in patients with Stargardt’s macular dystrophy and dry age-related
macular degeneration (Schwartz et al., 2012). Recently, follow-up study by Schwartz et al. (2015) provided the
first evidence of long term safety of transplanted hESC-derived RPE cells in 18
patients with atrophic AMD and Stargardt’s macular dystrophy. Based on the above study, it can be concluded that hESC-derived RPE
cells can be safely used as a source for allogeneic therapy in humans. However,
efficacy outcome of such studies are still awaited. While the above study
demonstrated safety, no immunological characterization of the derived cells was

We are the first to determine the immunological properties of ExMat?RPE cells. Usui et al. (2008) reported that B7?H1, a
negative immunoregulator, was expressed on the surface of human RPE cells,
while there was no expression of B7?H1 seen on murine RPE cells. We have shown
that ExMat-RPE cells did not express
B7?H1, but there was induced expression of B7-H1 (78% ± 3.07%) following IFN?? treatment
(Fig. 13C). Further, expression of another very strong negative
immunoregulator, IDO, by various cell types such as placental cytotrophoblasts (King and Thomas, 2007)
and mesenchymal stem cells (Tipnis
et al., 2010; Ryan et al., 2007; Meisel et al., 2004) has been causally
linked with protection of these cells from allogeneic T cell attack in vivo. We also observed high
levels of IDO expression in untreated and IFN-?-treated ExMat-RPE cells (Fig. 13C). In contrast, B7?DC, a positive
co-stimulatory protein, was not detected in ExMat-RPE
cells, although B7-DC mRNA was detected in both untreated and IFN??-treated
cells. Expression of B7-DC at mRNA level and not translating to surface protein
expression has been reported in several primary and telomerase-immortalized
ocular cell types (Usui et
al., 2008). The ExMat-RPE
cells also expressed HLA?DR antigen when stimulated with IFN-? for 5 days in
culture (Fig. 13B). While HLA?DR upregulation increases the chances of graft-rejection
by the MHC Class II antigen, ExMat-RPE
cells showed induced expression of B7?H1 and constitutive expression of IDO.
Expression of these negative immunoregulatory molecules together by ExMat?RPE cells may have prevented
T-cell proliferation in vitro in
the MLR assay in the absence and presence of IFN-? (Fig. 14), a probable
mechanism by which they might exhibit immune suppression. Our preliminary
results of the immunological investigation are promising, and suggest that ExMat-RPE cells do not elicit an immune
response in vitro. However, a more
detailed MLR study is needed to confirm the suitability of using these cells in
clinical applications.

This study suggests that it might be possible to utilize ExMat-RPE cells in regenerative
cell-based therapies. This may provide a significant advancement in generating
large number of cells for developmental studies, drug screening, and cell
therapy in future.