Functional and morphological analysis of the subretinal injection of retinal pigment epithelium cells.
ABSTRACT Replacement of retinal pigment epithelium (RPE) cells by transplantation is a potential treatment for some retinal degenerations. Here, we used a combination of invasive and noninvasive methods to characterize the structural and functional consequences of subretinal injection of RPE cells. Pigmented cells from primary cultures were injected into albino mice. Recovery was monitored over 8 weeks by fundus imaging, spectral domain optical coherence tomography (sdOCT), histology, and electroretinography (ERG). sdOCT showed that retinal reattachment was nearly complete by 1 week. ERG response amplitudes were reduced after injection, with cone-mediated function then recovering better than rod function. Photoreceptor cell loss was evident by sdOCT and histology, near the site of injection, and is likely to have been the main cause of incomplete recovery. With microscopy, injected cells were identified by the presence of apical melanosomes. They either established contact with Bruch's membrane, and thus became part of the RPE monolayer, or were located on the apical surface of the host's cells, resulting in apposition of the basal surface of the injected cell with the apical surface of the host cell and the formation of a series of desmosomal junctions. RPE cell density was not increased, indicating that the incorporation of an injected cell into the RPE monolayer was concomitant with the loss of a host cell. The transplanted and remaining host cells contained large vacuoles of ingested debris as well as lipofuscin-like granules, suggesting that they had scavenged the excess injected and host cells, and were stressed by the high digestive load. Therefore, although significant functional and structural recovery was observed, the consequences of this digestive stress may be a concern for longer-term health, especially where RPE cell transplantation is used to treat diseases that include lipofuscin accumulation as part of their pathology.
- SourceAvailable from: iovs.org[show abstract] [hide abstract]
ABSTRACT: To determine whether transplanting normal retinal pigment epithelium (RPE) into the subretinal space influences photoreceptor function and degeneration in Rpe65(-/-) mice. RPE cells were isolated from eyes of normal mice and transplanted to the subretinal space of one eye of Rpe65(-/-) mice. The other eye received a subretinal injection of saline or was not touched. Corneal electroretinograms (ERGs) from both eyes were monitored before and after surgery to follow progression of the degeneration. The width of the outer nuclear layer was measured in the area of transplantation and compared with a similar area in control retinas. Transplantation of RPE increased ERG amplitude maximally at 3.7 weeks after surgery. This rescue effect slowly diminished with time. Sham surgery had little effect on the ERG. The width of the outer nuclear layer in the area receiving RPE transplants was slightly greater than in control subjects. Evidence of the presence of RPE transplants in the subretinal space decreased with time after transplantation without signs of inflammation. Retinal degeneration in the Rpe65(-/-) mice is slowly progressive. Photoreceptor function can be transiently increased for several months and anatomic degeneration slightly reduced in Rpe65(-/-) mice by RPE cell transplantation. Loss of the rescue effect may be due to degeneration of the transplanted RPE.Investigative Ophthalmology & Visual Science 11/2002; 43(10):3307-11. · 3.44 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The first step toward light perception is 11-cis to all-trans photoisomerization of the retinaldehyde chromophore in a rod or cone opsin-pigment molecule. Light sensitivity of the opsin pigment is restored through a multistep pathway called the visual cycle, which effects all-trans to 11-cis re-isomerization of the retinoid chromophore. The maximum throughput of the known visual cycle, however, is too slow to explain sustained photosensitivity in bright light. Here, we demonstrate three novel enzymatic activities in cone-dominant ground-squirrel and chicken retinas: an all-trans-retinol isomerase, an 11-cis-retinyl-ester synthase, and an 11-cis-retinol dehydrogenase. Together these activities comprise a novel pathway that regenerates opsin photopigments at a rate 20-fold faster than the known visual cycle. We suggest that this pathway is responsible for sustained daylight vision in vertebrates.Neuron 10/2002; 36(1):69-80. · 15.77 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Melanosomes within the retinal pigment epithelium (RPE) of mammals have long been thought to exhibit no movement in response to light, unlike fish and amphibian RPE. Here we show that the distribution of melanosomes within the mouse RPE undergoes modest but significant changes with the light cycle. Two hours after light onset, there is a threefold increase in the number of melanosomes in the apical processes that surround adjacent photoreceptors. In skin melanocytes, melanosomes are motile and evenly distributed throughout the cell periphery. This distribution is due to the interaction with the cortical actin cytoskeleton mediated by a tripartite complex of Rab27a, melanophilin, and myosin Va. In ashen (Rab27a null) mice RPE, melanosomes are unable to move beyond the adherens junction axis and do not enter apical processes, suggesting that Rab27a regulates melanosome distribution in the RPE. Unlike skin melanocytes, the effects of Rab27a are mediated through myosin VIIa in the RPE, as evidenced by the similar melanosome distribution phenotype observed in shaker-1 mice, defective in myosin VIIa. Rab27a and myosin VIIa are likely to be required for association with and movement through the apical actin cytoskeleton, which is a prerequisite for entry into the apical processes.Molecular Biology of the Cell 06/2004; 15(5):2264-75. · 4.60 Impact Factor
Functional and morphological analysis of the subretinal
injection of retinal pigment epithelium cells
MAREN ENGELHARDT,1,2,* CHINATSU TOSHA,1,* VANDA S. LOPES,1,3BRYAN CHEN,1
LISA NGUYEN,1STEVEN NUSINOWITZ,1AND DAVID S. WILLIAMS1,4
1Department of Ophthalmology, Jules Stein Eye Institute, David Geffen School of Medicine, University of California, Los Angeles,
2Department of Anatomy, Institute of Neuroanatomy, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany
3Centre of Ophthalmology, Faculty of Medicine, University of Coimbra, Coimbra, Portugal
4Department of Neurobiology, David Geffen School of Medicine, University of California, Los Angeles, California
(RECEIVED March 28, 2011; ACCEPTED November 18, 2011; FIRST PUBLISHED ONLINE March 1, 2012)
Replacement of retinal pigment epithelium (RPE) cells by transplantation is a potential treatment for some retinal
degenerations. Here, we used a combination of invasive and noninvasive methods to characterize the structural and
functional consequences of subretinal injection of RPE cells. Pigmented cells from primary cultures were injected into
albino mice. Recovery was monitored over 8 weeks by fundus imaging, spectral domain optical coherence tomography
(sdOCT), histology, and electroretinography (ERG). sdOCT showed that retinal reattachment was nearly complete by 1
week. ERG response amplitudes were reduced after injection, with cone-mediated function then recovering better than
rod function. Photoreceptor cell loss was evident by sdOCT and histology, near the site of injection, and is likely to have
been the main cause of incomplete recovery. With microscopy, injected cells were identified by the presence of apical
melanosomes. They either established contact with Bruch’s membrane, and thus became part of the RPE monolayer, or
were located on the apical surface of the host’s cells, resulting in apposition of the basal surface of the injected cell with the
apical surface of the host cell and the formation of a series of desmosomal junctions. RPE cell density was not increased,
transplanted and remaining host cells contained large vacuoles of ingested debris as well as lipofuscin-like granules,
suggesting that they had scavenged the excess injected and host cells, and were stressed by the high digestive load.
Therefore, although significant functional and structural recovery was observed, the consequences of this digestive stress
may be a concern for longer-term health, especially where RPE cell transplantation is used to treat diseases that include
lipofuscin accumulation as part of their pathology.
Keywords: Retina, Spectral domain optical coherence tomography, Cell transplantation, Lipofuscin,
The transplantation of healthy cells to replace diseased or lost cells
offers a potential treatment for a variety of forms of retinal degen-
eration (reviewed in Lamba et al., 2009).Toreplacephotoreceptoror
retinal pigment epithelium (RPE) cells, cells need to be transplanted
into the subretinal space; that is, the space between these two cell
layers. Subretinal injections result in retinal detachment, which must
be followed by successful reattachment. Compared with the injection
of a solution, after which the retina can be immediately reattached
of cells is a greater challenge. Experiments with the injection of RPE
cells for treating rodent models of retinal degeneration have been
reported, describing moderate success (Li & Turner, 1988a; Sheedlo
et al., 1989; Lund et al., 2001; Gouras et al., 2002; Arai et al., 2004;
Lu et al., 2009). However, reports of the consequences of the cell
injection procedure on retinal structure and function have been
In the present study, a major goal was to assess the extent of
initial retinal damage and the time course of recovery following
injection of RPE cells into the subretinal space of mouse eyes.
Structural and functional recovery was assessed noninvasively using
spectral domain optical coherence tomography (sdOCT), which has
only recently been adapted for use in the mouse (Ruggeri et al.,
2007; Fischer et al., 2009; Huber et al., 2009; Cebulla et al., 2010;
Gabriele et al., 2010), and electroretinography (ERG). Microscopy
wasalsoperformed.Asecond goalof thisstudy was tostudy howthe
injected RPE cells are incorporated into the host retina. By electron
microscopy, we provide new insight into interactions between trans-
planted and host RPE cells.
Address correspondence and reprint requests to: Dr. David S. Williams
or Dr. Steven Nusinowitz, Department of Ophthalmology, Jules Stein
Eye Institute, David Geffen School of Medicine, University of California,
Los Angeles, CA 90095.
E-mail: firstname.lastname@example.org or email@example.com
*These authors contributed equally.
Visual Neuroscience (2012), 29, 83–93.
Copyright ? Cambridge University Press, 2012 0952-5238/12 $25.00
Materials and methods
All experiments were performed in accordance with the guidelines
established by the Animal Research Committee of the University
of California in Los Angeles. Mice were maintained on a 12-h
light/12-h dark cycle with 30–80 lux fluorescence lighting during
used for the production of primary RPE cell cultures, as described
previously (Gibbs & Williams, 2003). BALB/c albino wild-type
mice (32 or 65 days old) were used as recipients for subretinal
injections. Injected mice were immunosuppressed with cyclospor-
weight in their drinking water.
RPE cells that had been in primary culture for 1 week (Gibbs &
Williams, 2003) were used for subretinal injections. To harvest cells
from culture, the medium was removed; the cells were washed briefly
at 37°C. Cells were removed from the plastic culture dish by adding 1
ml of growth medium [Dulbecco’s Modified Eagle’s Medium with
4.5 g/l glucose (Invitrogen, Grand Island, NY)], plus 0.1 g/l penicillin/
streptomycin, 2 mM l-glutamine, 110 mg/l sodium pyruvate, 10%
bovine fetal calf serum (FCS), and 1x Minimum Essential Medium
Eagle (MEM) nonessential amino acids per well (Gibbs & Williams,
2003), pooled and collected in a pellet by centrifugation for 5 min at
1000 rpm. Cells were washed three times in PBS to remove all residue
in PBS to a final concentration of 30,000 cells/ll. Cells were injected
using a NanoFilTMsubmicroliter injection system (World Precision
Instruments, Sarasota, FL), while a surgical microscope was used for
was administered. A sclerotomy was performed approximately 1 mm
posterior from the limbus with a 27G beveled needle. This needle was
inserted to approximately 75% of its full length, avoiding the lens.
Upon its withdrawal,a 33Gblunt-end injection cannula attached to the
microsyringe pump was inserted through the sclerotomy and tilted at
approximately 45 deg in a tangential direction, pointing towards the
posterior pole of the eye. Full insertion of the blunt-end cannula
optic nerve head. One microliter of PBS, containing ;30,000 RPE
cells, was injected into the subretinal space.
Fundus images were obtained using the Micron II retinal imaging
microscope (Phoenix Research Laboratories, Inc., Pleasanton, CA).
solution containing ketamine (15 mg/g) and xylazine (7 mg/g body
weight). Pupil dilation was accomplished by adding a drop of 1%
atropine sulfate. The mouse was placed on a movable platform so that
the eye could be aligned with the axis of the camera and the objective
lens set was positioned so that it touched the corneal surface (corneal
applanation). A drop of methylcellulose (2.5%) was placed on the
corneal surface. The illuminating light source was broadband but
filtered to produce light between 400 and 700 nm. Serial images were
recorded to document the change in retinal and ocular appearance over
the course of the study. Fundus images were exported as 1924 3 768
pixels 24-bit colored portable network graphics files.
Spectral domain optical coherence tomography
Ultra high-resolution sdOCT imaging was performed with a com-
mercially available sdOCT system (Bioptigen, Research Triangle
Park, NC). Mice were anesthetized and pupils dilated as described
above. Recordings were made with a 50 deg field of view, yielding
an image 1.5 mm in diameter. En face view C-scans were recorded,
recorded immediately after injection, at 1 and 4 days, and at 1, 2, 3,
4, and 8 weeks after injection in the same mouse. Resulting images
were exported as 640 3 480 pixel 8-bit gray bitmap files and
processed in Adobe Photoshop CS3.
Retinal layer thickness measurements were made with an
on-screen caliper supplied by the manufacturer of the sdOCT
and calibrated for the mouse eye. Retinal thickness was measured
from the outer edge of the nerve fiber layer to the band identified as
the RPE, a measurement hereafter referred to as the total retinal
thickness. Thickness measurements were made in the region showing
the largest retinal detachment following injection; subsequent meas-
urements in the same mouse were made in exactly the same location,
using the distance from the optic nerve head as a reference and using
the on-screen caliper as the measuring device.
Electroretinography was performed as previously described (Nusi-
nowitz et al., 2007). Briefly, after overnight dark adaptation, ERGs
were recorded from the corneal surface of the injected eye using
a gold loop electrode referenced to a similar gold wire in the mouth.
A needle electrode in the tail served as the ground. All stimuli were
presented in a large integrating sphere coated with highly reflective
white matte paint (#6080; Eastman Kodak Corporation, Rochester,
NY). A photic stimulator (Model PS33 Plus; Grass-Telefactor, West
Warwick, RI) affixed to the outside of the sphere illuminated its
interior with brief flashes of light. Responses were amplified 10,000
times (Grass P511 High Performance AC Amplifier), band-pass
filtered (0.1–300 Hz), digitized using an I/O board (PCI-6221;
National Instruments, Austin, TX) in a personal computer, and
averaged. Rod-mediated responses were recorded to blue flashes
(Wratten 47A; kmax5 470 nm) with an intensity held constant at
flash (5.00 log cd-s/m2) on a rod-saturating background (32 cd/m2). All
presentation rate was slowed to 0.2 Hz.
Eyes were enucleated at 4 or 8 weeks after injection and processed
for histology. Whole eyes were fixed in a mixture of 2% gluta-
raldeyhde, 2% paraformaldehyde in 0.1 m cacodylatebuffer(pH7.4).
After dissection to generate posterior eyecups, tissues were postfixed
in 2% osmium tetroxide in 0.1 m cacodylate buffer and processed for
embedment in Epon. Semithin sections were stained with toluidine
blue for light microscopy, and ultrathin sections were stained with
uranyl acetate and lead citrate for transmission electron microscopy.
Transvitreal injection of RPE cells into the subretinal space
RPE cells were obtained from 1-week old primary cultures, in which
the cells were almost confluent and contained robust pigmentation
(Fig. 1). They were washed and resuspended in saline buffer for
Engelhardt et al.
injection. The viability of the dissociated cells was confirmed by
seeding some in a new well and observing that they attached and
survived. Fig. 2A and 2B shows fundus images of a representative
retina before and immediately after injection. The postinjection
fundus image shows scattered patches of pigmented cells throughout
the retina. Following dissection up to 8 weeks after injection, pig-
mented cells were observed scattered throughout the eyecups (Fig. 2C
Fig. 2B), possibly due to reflux from the subretinal space as the needle
Noninvasive monitoring of initial detachment with sdOCT
The subretinal injection of RPE cells resulted in a local retinal
detachment that was evident by sdOCT. Fig. 3 shows a series of
sdOCT images, illustrating the injection site in two different retinas,
as examples. Immediately after injection, a significant bleb was
(Fig. 3A and 3B, asterisks). One week later, the original bleb was
significantly reduced, although small regions of detachment
remained (Fig. 3C and 3D). At 4 weeks, most mice showed no
evidence of retinal detachment (Fig. 3E and 3F). By 8 weeks
postinjection, the retinal detachment was completely resolved in
the initial bleb varied, ranging from approximately 10 to 50% of the
retina, even though the same injection parameters (total volume,
number of cells, flow rate of injection) were maintained. No
correlation between bleb size and age (P32 or p65) was evident.
A comparison of sdOCT images before and after injection
suggested that many retinas became slightly thinner at the injection
site. Retinal thickness measurements, including all retinal layers
from the vitreal surface to the basal RPE, were made at the same
locations within and outside of the injection site before and at various
times after injection. They showed that, 8 weeks after injection, total
retinal thickness was reduced by an average of ;13%within0.1 mm
of the injection site, but itremained unchanged in other areas.In a few
extreme examples, the thinning was associated with disturbances in
thereflectance bandsassociated withall retinallayers.Fig.4 showsan
example of an sdOCT retinal cross-section image taken prior to
injection and one taken at 8 weeks after injection. In this case, severe
thinning of the nerve fiber layer and the inner plexiform layer is
evident in a very focused area (Fig. 4B).
Full-field ERG analysis
Retinal function was evaluated by ERG and summarized in Fig. 5.
The mean ERG response elicited by our brightest flash intensity
under dark- and light-adapted conditions is shown at baseline
(prior to injection) and at 1 and 8 weeks after injection. Response
amplitudes under dark-adapted conditions were reduced significantly
1 week after injection compared to baseline. Unlike the observations
made by sdOCT, we observed a difference between the two age
groups, with the younger group (P32) showing a larger decline in
mean amplitude (z 5 ?2.52, P 5 0.012 and z 5 ?1.83, P 5 0.068,
for P32 and P65, respectively). The dark-adapted ERG tended to
improve over time, but even 8 weeks postinjection, ERG amplitudes
remained significantly smaller than the preinjection baseline meas-
ures (z 5 ?2.52, P 5 0.012 and z 5 ?1.83, P 5 0.068, for P32 and
P65, respectively.) The light-adapted ERG, mediated by cones, was
much less affected by the injection. While ERG amplitudes were
modestly reduced at 1-week postinjection (z 5 ?2.52, P 5 0.012
and z 5 ?2.02,P 5 0.043 for the P32 and P65 groups,respectively),
cone ERG function showed complete recovery after 8 weeks
(z 5 ?1.01, P 5 0.31 and z 5 0.944, P 5 0.345, for P32 and
Semithin sections of retinas from mice 8 weeks after injection were
examined by light microscopy. Retinas were found to contain
extensive regions of pigmented cells in their RPE layer, suggesting
that the transplanted cells (originating from pigmented mice) had
spread beyond the site of injection and had integrated into the host
RPE over relatively large regions (Fig. 6A–6E). Near the site of
injection, some retinal damage was evident. Photoreceptor cell loss,
as indicated by fewer rows of cells in the photoreceptor nuclear
layer, was somewhat more widespread than the focused thinning
observed by sdOCT but still relatively localized to the site of
injection. In this region of cell loss, the RPE was typically thicker
of the retina, there were no pigmented RPE cells (Fig. 6G).
To determine the extent of transplantation across the retina, we
scored cells along complete retinal sections, passing near the center
of the retina, as to whether they were pigmented, and whether
melanosomes could be identified in the apical processes. The rationale
for identifying melanosomes in the apical processes is that these
melanosomes are a better indicator of endogenous melanosomes.
Melanosomes in the cell body could have originated from ingestion of
an injected cell or its debris by a host cell and be contained within
a phagocytic vacuole (see below). However, to be located in the apical
region of the RPE, a melanosome must be subject to a dynamic
motility process that requires its membrane to be exposed to the
cytosol. To be moved into the apical RPE, RAB27A must first
associate with the membrane of the melansosome. RAB27A links the
melanosome (via the protein, MYRIP) to MYO7A, an actin-based
motor that drives the movement of melanosomes into the apical region
through the apical meshwork of actin filaments (Liu et al., 1998; Futter
et al., 2004; Gibbs et al., 2004).
The result of scoring individual RPE cells in sections of one
injected retina, 8 weeks after injection, is depicted in Fig. 7. No
melanosomes were detected in the far periphery of the injected
Fig. 1. RPE cells cultured under growth conditions for 1 week. Image shows
a typical cell culture dish with almost confluent heavily pigmented 129SV-
derived RPE cells after 1 week in culture. Cells at this stage were harvested
for injection. Scale bar 5 100 lm.
RPE cell transplantation
retina, indicating that this region of the retina was unaffected (see
also Fig. 6G). A mean of 64% of the cells counted were found to
contain melanosomes (black and gray cells), with 50% of these
melanosome-containing cells (i.e., 32% of the total cells) demon-
strating melanosomes in the apical processes (gray cells).
The pigmentation evident by light microscopy suggested that the
RPE cell transplantation was quite extensive, and except for near
the site of injection, where more than one layer of RPE cells was
clearly evident, the RPE appeared quite normal. By electron
microscopy, we first determined the density of the resulting RPE
cell layer. We found that the injected retinas contained 7.8 6 0.6
cells per 100 lm, which compares to the 8.4 6 0.2 cells per 100 lm
in noninjected retinas. Only cells that made contact with Bruch’s
membrane were counted. For this result to be consistent with the
observations of light microscopy, either many of the host RPE cells
were replaced by injected RPE cells, the injected cells did not
integrate fully into the RPE monolayer (and thus make contact with
Bruch’s membrane), or host cells internalized melanosomes by
ingesting injected cells or their debris. Our observations by electron
microscopy suggested that each of these three occurred.
Some cells that extended from Bruch’s membrane to the outer
segments contained individual melanosomes, which, in some cases,
were present in the apical processes (Fig. 8A), suggesting that they
originated from the injected cells. These cells may have replaced
host RPE cells that were dislodged from Bruch’s membrane during
the retinal detachment, as shown previously with injections of cells
into the subretinal space (Lopez et al., 1987; Li & Turner, 1988a).
However, some of the transplanted cells formed a second layer
between the apical surface of the host RPE and the outer segments
apical surface of the host cell became flattened, and the two
junctions (Fig. 9A and 9B). In both cases, whether the transplanted
cell established itself on Bruch’s membrane, or on the apical surface
of a host cell, its apical surface showed typical RPE processes that
extended between the outer segments (Figs. 8A, 8B, 9A, and 9C),
and the cells appeared to be polarized normally. Although the cells
Fig. 2. Visualization of injected cells. Representative fundus image of an adult Balb/c albino mouse before (A) and immediately after
(B) cell injection. The postinjection fundus image (B) shows scattered patches of pigmented cells distributed throughout the posterior pole
including pigmented clusters of cells (arrows) that appear to lie at the retina–vitreous interface, and a large clump (arrowhead) that appears
to have refluxed back into the vitreous when the injection needle was retracted. Enucleated eyecups are shown at 4 weeks (C, back of the
eyecup) and 8 weeks (D, interior of the eyecup, with lens removed) after injection. Pigmented cells can be seen scattered throughout the
eyecups. In each of (A–D), the optic nerve head is identified by an asterisk.
Engelhardt et al.
that formed a second layer did not make contact with a basement
membrane and did not develop basal infoldings, their mitochondria
were localized basally, and some of their melanosomes were present
in apical processes (Figs. 8B, 9A, and 9B).
Cells that contained clusters of melanosomes and cellular debris
were also observed (Fig. 9D). Light microscopy suggested that
these cells were normally pigmented transplanted cells. However,
electron microscopy suggests that they were likely host cells that
had ingested injected pigmented cells or their debris. In Fig. 8B, for
example, the melanosomes indicated by large arrowheads appear to
be nearly all clustered within a phagocytic vacuole. Interestingly,
most of the transplanted cells and original host cells contained
Fig. 3. sdOCTimages of two injected eyesover a time course of 8weeks. Eyes wereimaged with five individual scans of the central retina
the left, and those of the other eye are on the right. Scans were made immediately after cell injection (A, B), with follow-up scans 1 week
(C, D), 4 weeks (E, F), and 8 weeks (G, H) later. Immediately after cell injection, a distinct bleb with substantial retinal detachment
appears. Injected RPE cells are visible as lighter areas within the bleb (asterisks in A and B). One week after injection, the initial bleb has
dissolved and retinal layers have settled back onto the RPE layer; however, areas of retinal detachment are still evident (arrows in C, D).
Four weeks after injection, the retina appears normal, without evident retinal detachment. However, the somewhat distorted image quality
in (E) could be a sign of obstructing vitreal accumulation of cells or fluid. Eight weeks postinjection, both retinas appear normal with no
persisting sign of damage caused by the procedure (G, H). RPE, retinal pigment epithelium; IS/OS, photoreceptor inner and outer
segments, respectively; OLM, outer limiting membrane; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer;
IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar 5 300 lm.
RPE cell transplantation
lipofuscin-like granules (e.g., arrows in Fig. 8A–8D), suggesting
that they had been challenged with excessive demands for degra-
dation of internalized products.
Using a combination of noninvasive and invasive methods, we
have monitored retinal structure and function following the
Fig. 4. sdOCT comparison of an injection site after 8 weeks. Representative sdOCT retinal cross-sections at baseline(left panel) and at the
same location 8 weeks after injection (right panel). Note the structural changes at the injection site and the apparent thinning of the retina.
RPE, retinal pigment epithelium; OS, photoreceptor outer segments; IS, photoreceptor inner segments; OLM, outer limiting membrane;
ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer.
Fig. 5. Retinal function as assessed by electroretinography (ERG). (A) Dark-adapted ERG responses evoked by a short-wavelength flash
(0.337 cd-s/m2) presumed to reflect mainly rod-mediated function. (B) Light-adapted (cone-mediated) ERG responses evoked by a white
flash (5.00 cd-s/m2) on a rod-saturating background (32.0 cd/m2). Electroretinography was performed serially on mice that received
injections at either postnatal day 32 (P32) or at postnatal day 65 (P65). ERGs were recorded prior to injections (time indicated by the
vertical dashed line) and then again at 1, 4, and 8 weeks after injection. The same mice were used for both dark- and light-adapted ERG
recordings. Error bars indicate 6s.e.m.
Engelhardt et al.
subretinal injection of RPE cells. Recovery from the initial retinal
detachment was observed in live animals, and the resulting trans-
plantation was analyzed by light and electron microscopy.
We injected cultured RPE cells to mimic biological conditions
currently under investigation for cell replacement strategies (e.g.,
Lund et al., 2001;Seileretal.,2010).Inthefuture,stemcell-derived
RPE cells will be a common source for transplantation, and these
the cells used in the present study. After each injection, a bleb of
retinal detachment was evident by sdOCT. This noninvasive
approach, which has relatively recently been adapted for use with
mouse eyes (Ruggeri et al., 2007; Fischer et al., 2009; Huber et al.,
2009; Cebulla et al., 2010; Gabriele et al., 2010), allowed us to
monitor the reattachment process in each eye. Reattachment took
significantly longer than the ;1 day reported previously for fluid
injections (Timmers et al., 2001). The time taken for reattachment is
likely to be an important parameter in the recovery of retinal
structure and function and is likely to be responsible for the
permanent structural and functional losses we observed in many
Comprehensive data from the cat retina show that the RPE–
photoreceptor interface and the inner retina undergo significant
structural changes after retinal detachment, and that the severity of
these changes depends on the extent of the detachment (Anderson
et al., 1983). In particular, proliferation of the RPE has been
observed within the detachment bleb during the first days of
detachment (Anderson et al., 1981), and areas where this pro-
liferation has occurred are evident after reattachment (Anderson
et al., 1986). In the present study, RPE proliferation may have
contributed to the perturbed organization of the RPE and limited
outer segment recovery that was evident near the site of injection
(as in Fig. 6F), although, even in these regions, we observed cells
with apical melanosomes, indicating that they had not resulted from
proliferation of the host RPE, which was albino.
Consistent with the structural changes that were observed, rod-
and cone-mediated retinal functions, as assayed with full-field
ERGs, were also compromised after injection. Disruption of the
circulating current in the outer retina due to the separation of
the photoreceptors from the RPE is likely to have contributed to the
decrease in function shortly after injection. However, the lack of
full recovery of rod photoreceptor function by 8 weeks post-
injection is likely due to photoreceptor cell loss, especially near the
site of injection.
In contrast to rod-mediated function, cone-mediated function
recovered almost completely. Following transplantation of RPE
cells into the Royal College of Surgeons (RCS)ratretina,aprevious
Fig. 6. Histological analysis of retinal morphology after injection. Micrographs of semithin sections from eyes, 8 weeks after injection.
retina. Atleast someof the cellscontainnormal functioningmelanosomes,evident by the presenceof melanosomesin the apicalprocesses
(andadjacentto thetips ofthe outersegments).(F) Regionof a retinanearthe siteof injectionshowingperturbedoutersegmentsandmore
a normal albino retina. RPE, retinal pigment epithelium; OS, photoreceptor outer segments; IS, photoreceptor inner segments; ONL, outer
(i.e., photoreceptor) nuclear layer. Scale bar 5 25 lm.
RPE cell transplantation
study also observed different effects on cone and rod function (Pinilla
et al., 2005). Cone function may be less affected since their visual
pigment can be cycled through Müller cells, and thus, they are less
In studying the fate of the injected cells by light and electron
microscopy, we determined that some of the injected cells
appeared to have established contact with Bruch’s membrane,
while others formed a second layer by resting on the apical surface
of the host’s cells, thus confirming interpretations from some earlier
studies (Li & Turner, 1988b, 1991; Sheedlo et al., 1991; Zhang &
Bok, 1998). Since cell density was not altered, cells that were
transplanted on to Bruch’s membrane are likely to have filled spaces
left by host cells that were dislodged during the injection and
detachment of the retina. Images of patches of denuded Bruch’s
(Lopez et al., 1987; Li & Turner, 1988a).
Electron microscopy also indicated that the cells of the “new”
RPE had undergone a heavy digestive load. Both host and trans-
in particular, had large phagocytic vacuoles containing debris,
including melanosomes, which were probably from injected cells
(some of which may have become lysed during preparation or
injection). Presumably, ingestion of dislodged host cells also
contributed to this digestive load. The ingestion of melanosomes
by host cells resulted in large areas of what appeared, by light
microscopy, to be contiguous pigmented (and therefore trans-
planted) RPE cells; in fact, these areas were a mixture of trans-
planted and host cells.
The observation of abnormal numbers of lipofuscin-like gran-
ules indicates that the demand for degradation of the large amount
of internalized products was deleterious. This may be of particular
relevance to proposed treatments by RPE cell transplantation of
retinal disease that is characterized by an abnormal accumulation
of lipofuscin, such as Stargardt macular degeneration (Weng et al.,
1999; Lu et al., 2009).
In an additional observation, we found where injected cells
established themselves on top of the host RPE cells, rather than on
Bruch’s membrane, the interacting surface membranes change their
shape—the basal infoldings of the injected cell and the apical
microvilli of the host cell are lost—and they appose each other with
desmosome junctions. The resulting bilayer interfaces with Bruch’s
membrane and interdigitates with the outer segments in the same
manner as a normal RPE, which is a monolayer. However, the effect
of a bilayer on RPE physiology is unclear.
In conclusion, we show by noninvasive structural and func-
tional monitoring as well as conventional microscopy that the
injection of RPE cells into the subretinal space can be quite
disruptive. The retina remains detached for a significantly longer
period than that described for the injection of fluid. This detachment
was monitored by sdOCT and shown to affect retinal electrophys-
iology adversely. However, retinas showed structural and functional
recovery after the injection, as well as the successful integration of
Fig. 7. Pigmentation of RPE cells. Each strip represents the RPE cells in one
section, extending from ora serrata to ora serrata and passing near the center
of a retina. Cells were determined to contain melanosomes in the cell body
only (black), in the apical processes (gray), or not at all (white). The left strip
represents a section from an albino retina that was not injected. The other
strips represent sections from a single injected retina; the sections are spaced
at least 10 lm from each other.
Engelhardt et al.
outer segments (OS) and appears to be a transplanted cell,based on the presence of individual melanosomes that are scattered throughout the
cell, including the apical region. (B) A region where injected cells have been transplanted on to the apical surface of a host RPE cells. Small
melanosomes that appear to have been ingested by host RPE cells. Note that the transplanted cells are well polarized, with melanosomes in
cells (and contains transplanted cells elsewhere). Lipofuscin-like granules (arrows) are evident in the transplanted cells in (A) and (B) and in
the host cells shown in (C) and (D). Scale bar 5 2 lm.
RPE cell transplantation
injected cells into the host RPE layer. A concern is raised over the
presence of lipofuscin granules in the resulting RPE cells, particularly
The authors thank Douglas Yasumura and Michael Matthes (Beckman
Vision Center, University of California, San Francisco) for expert technical
advice. We also thank Gabriel Travis, Dean Bok, Xanjie Yang (Jules Stein
Eye Institute, UCLA), and Guoping Fan (Department of Human Genetics,
UCLA) for valuable discussions. This study was funded by the California
Institute of Regenerative Medicine (CIRM TR1-01272), NIH Core Grant
2P30EY000331, and NIH Grant 5R01EY007042. D.S.W. is a Jules and
Doris Stein RPB Professor.
Anderson, D.H., Guerin, C.J., Erickson, P.A., Stern, W.H. & Fisher, S.K.
(1986). Morphological recovery in the reattached retina. Investigative
Ophthalmology & Visual Science 27, 168–183.
Anderson, D.H., Stern, W.H., Fisher, S.K., Erickson, P.A. &
Borgula, G.A. (1981). The onset of pigment epithelial proliferation
after retinal detachment. Investigative Ophthalmology & Visual Science
Anderson, D.H., Stern, W.H., Fisher, S.K., Erickson, P.A. &
Borgula, G.A. (1983). Retinal detachment in the cat: The pigment
Fig. 9. Electron micrographs of the RPE layer after injection. (A) The boundary between two RPE cells, one (lower) that possesses apical
processes that interdigitate aroundthe tips of photoreceptorouter segmentsand one (upper)whose apical surface abuts the basal surface of
the lower cell. The lower cell appears to have originated from the population of injected cells (note melanosome in lower right).
Desmosomes (arrowheads) are evident along the junction of the two cells. Scale bar 5 2 lm. (B) Higher magnifications of boundary
between two RPE cells, where one cell resides on top of the other. Arrowheads indicated desmosomes. Scale bar 5 200 nm. (C) Apical
population of injected cells. AP, apical processes. Scale bar 5 1 lm. (D) Melanosomes and cellular debris that appear to be clustered in
a phagocytic vacuole within a host RPE cell. Scale bar 5 500 nm.
Engelhardt et al.
epithelial-photoreceptor interface. Investigative Ophthalmology &
Visual Science 24, 906–926.
Arai, S., Thomas, B.B., Seiler, M.J., Aramant, R.B., Qiu, G., Mui, C.,
de Juan, E. & Sadda, S.R. (2004). Restoration of visual responses
followingtransplantation of intactretinalsheetsin rdmice.Experimental
Eye Research 79, 331–341.
(2010). Spectral domain optical coherence tomography in a murine retinal
detachment model. Experimental Eye Research 90, 521–527.
Fischer, M.D., Huber, G., Beck, S.C., Tanimoto, N., Muehlfriedel, R.,
Fahl, E., Grimm, C., Wenzel, A., Reme, C.E., van de Pavert, S.A.,
Wijnholds, J., Pacal, M., Bremner, R. & Seeliger, M.W. (2009).
Noninvasive, in vivo assessment of mouse retinal structure using optical
coherence tomography. PLoS One 4, e7507.
(2004). The role of Rab27a in the regulation of melanosome distribution
within retinal pigment epithelial cells. Molecular Biology of the Cell 15,
Gabriele, M.L., Ishikawa, H., Schuman, J.S., Bilonick, R.A., Kim, J.S.,
Kagemann, L. & Wollstein, G. (2010). Reproducibility of spectral-
domain optical coherence tomography total retinal thickness measurements
Libby, R.T. & Williams, D.S. (2004). Role of myosin VIIa and Rab27a in
the motility and localization of RPE melanosomes. Journal of Cell Science
Gibbs, D. & Williams, D.S. (2003). Isolation and culture of primary mouse
retinal pigmented epithelial cells. Advances in Experimental Medicine &
Biology 533, 347–352.
Gouras, P., Kong, J. & Tsang, S.H. (2002). Retinal degeneration and RPE
transplantation in Rpe65(?/?) mice. Investigative Ophthalmology &
Visual Science 43, 3307–3311.
Paquet-Durand, F., Wenzel, A., Humphries, P., Redmond, T.M.,
Seeliger, M.W. & Fischer, M.D. (2009). Spectral domain optical
coherence tomography in mouse models of retinal degeneration.
Investigative Ophthalmology & Visual Science 50, 5888–5895.
Lamba, D.A., Karl, M.O. & Reh, T.A. (2009). Strategies for retinal repair:
Cell replacement and regeneration. Progress in Brain Research 175,
Li, L.X. & Turner, J.E. (1988a). Transplantation of retinal pigment
epithelial cells to immature and adult rat hosts: Short- and long-term
survival characteristics. Experimental Eye Research 47, 771–785.
Li, L.X. & Turner, J.E. (1988b). Transplantation of retinal pigment
epithelial cells to immature and adult rat hosts: Short- and long-term
survival characteristics. Experimental Eye Research 47, 771–785.
Li, L. & Turner, J.E. (1991). Optimal conditions for long-term photore-
ceptor cell rescuein RCS rats: The necessityfor healthy RPEtransplants.
Experimental Eye Research 52, 669–679.
Liu, X., Ondek, B. & Williams, D.S. (1998). Mutant myosin VIIa causes
defective melanosome distribution in the RPE of shaker-1 mice. Nature
Genetics 19, 117–118.
Lopez, R., Gouras, P., Brittis, M. & Kjeldbye, H. (1987). Trans-
plantation of cultured rabbit retinal epithelium to rabbit retina using
a closed-eye method. Investigative Ophthalmology & Visual Science
Lu, B., Malcuit, C., Wang, S., Girman, S., Francis, P., Lemieux, L.,
Lanza, R. & Lund, R. (2009). Long-term safety and function of RPE
from human embryonic stem cells in preclinical models of macular
degeneration. Stem Cells 27, 2126–2135.
Lund, R.D., Adamson, P., Sauve, Y., Keegan, D.J., Girman, S.V.,
Wang, S., Winton, H., Kanuga, N., Kwan, A.S., Beauchene, L.,
Zerbib, A., Hetherington, L., Couraud, P.O., Coffey, P. &
Greenwood, J. (2001). Subretinal transplantation of genetically mod-
ifiedhumancell lines attenuatesloss ofvisualfunctionin dystrophicrats.
Proceedings of the National Academy of Sciences of the United States of
America 98, 9942–9947.
Mata, N.L., Radu, R.A., Clemmons, R.C. & Travis, G.H. (2002).
Isomerization and oxidation of vitamin a in cone-dominant retinas: A
novel pathway for visual-pigment regeneration in daylight. Neuron
Nusinowitz, S., Ridder, W.H. III. & Ramirez, J. (2007). Temporal
Pinilla, I., Lund, R.D., Lu, B. & Sauve, Y. (2005). Measuring the cone
rescue in RCS rats. Vision Research 45, 635–641.
Ruggeri, M., Wehbe, H., Jiao, S., Gregori, G., Jockovich, M.E.,
Hackam, A., Duan, Y. & Puliafito, C.A. (2007). In vivo three-
dimensional high-resolution imaging of rodent retina with spectral-
domain optical coherence tomography. Investigative Ophthalmology &
Visual Science 48, 1808–1814.
Seiler, M.J., Rao, B., Aramant, R.B., Yu, L., Wang, Q., Kitayama, E.,
Pham, S., Yan, F., Chen, Z. & Keirstead, H.S. (2010). Three-
dimensional optical coherence tomography imaging of retinal sheet
implants in live rats. Journal of Neuroscience Methods 188, 250–257.
Sheedlo, H.J., Li, L.X. & Turner, J.E. (1989). Functional and structural
characteristics of photoreceptor cells rescued in RPE-cell grafted retinas
of RCS dystrophic rats. Experimental Eye Research 48, 841–854.
Sheedlo, H.J., Li, L. & Turner, J.E. (1991). Photoreceptor cell rescue at
early and late RPE-cell transplantation periods during retinal disease in
RCS dystrophic rats. Journal of Neural Transplantation & Plasticity
Timmers, A.M., Zhang, H., Squitieri, A. & Gonzalez-Pola, C. (2001).
Subretinal injections in rodent eyes: Effects on electrophysiology and
histology of rat retina. Molecular Vision 7, 131–137.
Weng, J., Mata, N.L., Azarian, S.M., Tzekov, R.T., Birch, D.G. &
Travis, G.H. (1999). Insights into the function of Rim protein in
photoreceptors and etiology of Stargardt’s disease from the phenotype
in abcr knockout mice. Cell 98, 13–23.
Zhang, X. & Bok, D. (1998). Transplantation of retinal pigment epithelial
cells and immune response in the subretinal space. Investigative
Ophthalmology & Visual Science 39, 1021–1027.
RPE cell transplantation