| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Wallenberg Retina Center, Department of Ophthalmology, Lund University Hospital, Lund, Sweden;
b Wallenberg Neuroscience Center, Division of Neurobiology, University of Lund, Lund, Sweden
Key Words. Brain-derived precursor cells • Subretinal transplantation • Normal rats • Long-term survival • Glial differentiation
Correspondence:
Karin Warfvinge, Ph.D., Wallenberg Retina Center, Department of Ophthalmology, Lund University Hospital, S-221 84 Lund, Sweden. Telephone: 46-46-2220772; Fax: 46-46-2220774; e-mail: karin.warfvinge{at}oft.lu.se
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Currently, we have therefore investigated the long-term survival capacity of the RN33B cell line after subretinal transplantation to the adult normal rat. Our results demonstrated that these brain-derived precursor cells survive in the retina for at least 4 months posttransplantation in immunosuppressed rats. Immunohistochemistry with several glial markers indicated that the majority of the grafted cells differentiated into glial cells, found in different maturity stages. In addition, these cells displayed extensive migration over long distances, and a major fraction of the cells expressed proliferation markers at 4 months posttransplantation.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Animals and Surgery
Adult, female Sprague-Dawley rats were obtained from BK Universal (Uppsala, Sweden). Thirty-one rats were subretinally (i.e., between the neuroretina and RPE) and unilaterally transplanted with 200,000 RN33B cells, trypsinized and prepared as a single cell suspension at 100,000 cells/µl in Hanks' balanced salt solution (HBSS; GIBCO; Life Technologies). In addition, eight control rats were subretinally injected with the HBSS medium only. For surgery, the animals were anesthetized using Equithesin (0.3 ml/100 g body weight i.p.).
The technique for subretinal transplantation was based on the method described by Kwan et al. [25] and has previously been described in detail [21]. In brief, a nose bar secured the head, and two sutures (6/0 Vicryl Ethicon; Johnson-Johnson; Brussels, Belgium) were attached to the upper left eyelid to expose the conjunctiva. Two further sutures were then applied directly to the conjunctiva. By gently pulling the sutures and securing them to the table, good exposure of the intended injection site and stabilization of the globe were achieved. Under the operating microscope, the superior part of the conjunctiva was temporarily removed. A 23-gauge needle (Becton Dickinson; Franklin Lakes, NJ; http://www.bd.com) was used to gently remove the sclera, behind the limbus, at the intended site of injection. A small amount of air was drawn into a 30-mm-thin tube on a plastic tip (1-10 µl GELoader Tip; Eppendorf, Hamburg, Germany; http://www.brinkmann.com) mounted on a Hamilton microliter syringe, followed by 2 µl of the cell suspension. The plastic tube was then gently put into the subretinal space through the opening in the sclera and the infusion started. When a retinal bleb was visible through the thin sclera, the tube was slowly moved toward the optic nerve while the infusion continued. The infusion caused the retina to detach from the RPE. Once the cell suspension was delivered, the tube was slowly retracted. Prior to removal from the subretinal space, a small amount of air was delivered to "lock" the suspension at the site of injection. After removal of the securing sutures, the detached piece of the conjunctiva was reattached to the incision site.
Eleven graft recipients were immunosuppressed (from 1 day prior to surgery) with cyclosporin (SandimmuneTM; Novartis; Täby, Sweden; http://www.novartis.com; daily i.p. injections of 10 mg/kg) given in combination with cortisone (Prednisolone; Pharmacia Upjohn; Uppsala, Sweden; http://www.pharmacia.se; daily s.c. injections for 5 days, 10 mg/kg). In addition, three immunosuppressed rats were injected with the medium only. The immunosuppressed rats were also given antibiotics (Borgal; Hoechst Roussel Vet; Münich, Germany; 2 ml/l H2O) in their drinking water during the first week. A group of 25 rats (20 rats were transplanted with the RN33B cells and 5 rats with the medium only) received no immunosuppression at all.
All animal-related work was carried out according to local ethical guidelines and approved animal care protocols.
Tissue Processing
At 4 months posttransplantation, the rats were sacrificed with CO2, and the eyes enucleated and directly placed in 4% paraformaldehyde (PFA) for 5-10 minutes. The anterior segment and the lens were subsequently removed, and the posterior eyecup postfixed for 2-4 hours in 4% PFA. Retinas were then processed in rising concentrations of sucrose-containing Sørensen's phosphate buffer (pH 7.2), embedded in gelatin medium, and cryostat sectioned at 12 µm. Every tenth slide was stained with hematoxylin-eosin (Htx-Eosin).
Immunohistochemistry
Frozen sections were thawed and washed in 0.1 M phosphate buffered saline with 0.25% Triton X-100 (pH 7.2). The sections were exposed to a polyclonal rabbit primary antiserum against ß-galactosidase or chicken primary antiserum against GFP (Table 1
) in a moist chamber (16-18 hours, +4°C) followed by rinsing in PBS/Triton X-100 and incubation with antirabbit Texas Red or FITC (fluorescein isothiocyanate), or antichicken FITC-conjugated secondary antibodies (Jackson; West Grove, PA; 1:80, 30 minutes room temperature in darkness). For double labeling of the ß-galactosidase- or GFP-immunoreacted material, the sections were then processed with either of the primary mouse antisera listed in Table 1 and with antimouse Texas Red (Jackson) secondary antibodies. In cases where the primary antibodies were made in rabbit, only GFP immunoreacted material was used.
|
Normal or sham operated eyes, processed in parallel, were used as controls. In addition, negative controls with omission of the primary antisera were performed.
Microscopy
The retinas were examined by light and epifluorescence microscopy.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
In the animals with surviving cells, implanted cells were found integrated into the RPE (Fig. 1A), the outer plexiform layer (OPL) (Figs. 1A, 1C, 1D, and 2D), the inner nuclear layer (INL) (Figs. 1C and 2D), (mainly) the inner plexiform layer (IPL) (Figs. 1C, 1D, 2A, 2D, and 3), and the nerve fiber layer (NF) (Figs. 1B, 1C, 1D, and 2D), and the optic disc/nerve (Figs. 1B, 1D, 2C, and 2E). A few cells were found within the ganglion cell layer (GCL) (Figs 2A and 2C). There existed minor subpopulations of implanted cells that expressed either ß-galactosidase or GFP. The distribution of these subpopulations varied within the same section; areas were found with many GFP but few ß-galactosidase-positive cells and vice versa, which indicated a downregulation of either of the two markers. However, most often, both antigens were expressed in the majority of the grafted cells. Thus, the two methods of transplant identification showed the same distribution pattern of implanted cells.
|
|
|
); in the fifth, almost the entire retina contained grafted cells. Although the host retinas harbored a vast amount of implanted cells, the normal cyto-architecture of the host seemed unaffected. Cells that had migrated longer distances from the transplant site showed the same immunohistochemical ß-galactosidase (Fig. 1D) or GFP intensity as those close to the transplant site. In addition, the grafted cells displayed similar morphologies throughout the entire retina.
Judging from the PCNA and SV40 large T-antigen immunohistochemistry, many cells in the host retina were in a proliferative stage in the RPE, OPL, INL, IPL, and NF (Figs. 4A-4D). These cells outnumbered those that were stained with either ß-galactosidase or GFP antibodies (Figs. 4B and 4D), a finding that may be attributed to downregulation of the transplant markers. Thus, an additional subpopulation of grafted cells might exist that lacks the expression of either of the two identification markers, but expresses the proliferation markers. However, most of the ß-galactosidase- and GFP-positive cells were also PCNA or SV40 large T-antigen positive. No PCNA immunoreactive cells were found in the HBSS-injected retinas, in the right untreated control eyes, or in the retinas with no surviving RN33B cells.
|
All controls with normal or sham operated eyes gave negative results with the appropriate detection methods.
Morphology and Phenotypic Characteristics of the Implanted Cells
Visualization of the implanted cells with ß-galactosidase or GFP immunohistochemistry revealed that the transplanted cells had acquired morphologies different from those observed in vitro. Many of the cells displayed retinal-cell-like morphologies, with formation of processes often spreading horizontally (Figs. 1A and 1C). A few cells with bipolar-like morphology (Fig. 1C) were found in the INL.
A significant number of the transplanted cells were immunoreactive to the glial-associated markers used. In the OPL and IPL, the presence of a few glial progenitors was determined by colocalization of GFP and the proteoglycan marker NG2 (Fig. 2B). Many of the implanted cells, found in mainly the OPL and IPL, were vimentin immunoreactive (Fig. 2A). Expression of the oligodendroglial markers, Rip (Figs. 2C and 2D) or MBP (Fig. 2E), in many grafted cells in NF and optic disc/nerve indicated a differentiation of the cells into mature oligodendrocytes.
On the other hand, we were not able to reveal any transplanted cells coexpressing ß-galactosidase or GFP and any of the retinal specific markers used, i.e., cytokeratin, RPE65, parvalbumin, calbindin, Rho 4D2, or recoverin.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Survival, Integration, and Proliferation of the Implanted RN33B Cell Line
We have recently described the short-term survival, integration, and differentiation of two brain-derived precursor cell lines (C17-2 and RN33B) after subretinal transplantation to the normal adult rat [21]. It was shown that these cell lines, including the RN33B cell line, survived for 3-4 weeks posttransplantation. The present study shows that the RN33B cell line survives up to 4 months postgrafting in the adult immunosuppressed rat retina. We were not able to demonstrate any surviving cells in the nonimmunosuppressed rats. Recent results [26] on RN33B cell transplantation to immunosuppressed normal and dystrophic rats of different ages showed that the number of animals with surviving grafted cells varied between the different groups of animals. The reason for the discrepancy between groups is not known. It might be due to technical difficulties, since the surgical procedure is technically demanding, or to the use of different batches of the precursor cells. Furthermore, unpublished observations from our laboratory indicate that survival times longer than 3-4 weeks will decrease the possibility of detecting surviving RN33B cells in the adult rat retina.
It has been suggested that survival without immunosuppression is associated with low levels of major histocompatibility complex (MHC) expression of the precursor cells at baseline, i.e., prior to transplantation [27]. It is not known whether the cells used in the present study show low levels of MHC expression in vitro, although preliminary in vivo results from our laboratory indicate an upregulation of MHC class I in the surviving RN33B cells. Thus, in our hands, an alloimmune response is present with the cells and the methods used. This has to be further evaluated.
The reports on long-term survival of precursor cell grafts to the retina are sparse. However, it has been shown that at 4 months posttransplantation EGF-propagated rat brain-derived striatal and spinal precursor cells had survived, integrated into the mouse retina, and differentiated into oligodendrocytes [20]. Also, bFGF-propagated adult rat hippocampal progenitor cells survived, integrated, and differentiated for 4 months after intravitreal transplantation to young diseased rats [19]. In addition, investigations on transplantation to the brain suggest, for example, that bFGF-propagated rat embryonic forebrain cells survive for at least 7 months in different brain regions [28]. The SV40 large T-antigen immortalized HiB5 (embryonic rat hippocampal cells) cell line survived for 6 months in the adult rat striatum [29], and the RN33B for up to 6 months in the adult and neonatal rat hippocampus and cerebral cortex [30]. Thus, immortalized precursor cell lines, including the RN33B cell line, and, for example, EGF- or bFGF-propagated rat precursor cells, have the ability to survive over a long period of time when transplanted to different regions of the CNS (including the retina).
In the present study, the grafted RN33B cells integrated into the RPE, inner layers of the retina, and the anterior part of the optic nerve, but not into the photoreceptor layer. Takahashi [18] and Young [19] and coworkers injected adult rat hippocampus-derived neural progenitor cells into the vitreous of newborn healthy rats and rats with retinal degeneration, and reported integration of the implanted cells into all retinal layers, including the outer nuclear layer, with expression of mature neural markers such as NF-200, MAP-5, and calbindin. It might be speculated that the outer nuclear layer differs from the other layers in terms of environmental signals facilitating integration of certain grafted precursor cells, which do not include the cells used in the present study.
The current study also demonstrates that these precursor cells are able to migrate over long distances within the host retina, since grafted cells were found in the entire retina (from one eccentricity to the other) in four out of five rats 4 months posttransplantation. In addition, grafted cells were distributed in an even pattern mainly within the OPL and IPL, but also within INL. Earlier results suggested that certain migratory cells appear to be glial precursors in a proliferative, migratory stage of their development [20,31]. To investigate whether this was the case with the precursor cells used in the present study, we used immunohistochemistry with several glial-associated markers to determine the fate of the implanted cells. Indeed, many of the grafted cells expressed glial markers with a preference for oligodendrocyte differentiation, which is in contrast to evaluation of the cell line in culture [22] or following grafting into adult and neonatal rat hippocampus and cerebral cortex [30,32], where the RN33B cells differentiated into neurons, but not glial cells. However, results from our laboratory indicate that the RN33B cells differentiate into neurons, astrocytes, and oligodendrocytes when transplanted to different regions of the neonatal rat brain (Lundberg et al. manuscript in preparation). Ader and coworkers reported that a portion of striatal and spinal precursor cells transplanted to the retina of young mice differentiated into oligodendrocytes [20]. In addition, these oligodendrocytes were found in virtually the entire nerve fiber layer of the host retina 4 months posttransplantation. Thus, it seems that certain precursor cells with intrinsic and/or extrinsic glial cell lineage signals have the capability of migrating over long distances within the retina.
To reveal whether proliferation occurred within the retina, sections were immunohistochemically processed with antibodies against the SV40 large T-antigen or the PCNA antigen. All transplanted retinas with surviving cells showed immunoreactivity to both antigens. Approximately 200,000 cells were grafted, and preliminary results on whole mount preparations have shown that only a small fraction (less than 5%) survives for 3 weeks. It was not possible to estimate, in sections, the number of grafted cells residing within the retina in the present material. However, it clearly exceeded the amount present after 3 weeks. It is therefore obvious that a long-lasting proliferation occurs within the retina. In addition, the proliferation seemed to have been slow, since the host retinas showed no sign of damage, although no attempt was made to thoroughly scrutinize the effect on the host, except for the examination of the Htx-Eosin- and immunohistochemical-stained sections using antibodies described in Table 1. It cannot be excluded that some of the host cells also were in a proliferative stage, since not all the SV40 large T-antigen or the PCNA antigen immunoreactive cells were ß-galactosidase or GFP positive. It is known that retinal trauma is associated with a Müller cell gliosis [33]. This might indicate that some of the proliferative cells found were Müller cells, induced to proliferate by the transplantation per se. However, the marker genes for the RN33B cells might be downregulated, which is shown in the present study by the presence of minor subpopulations of implanted cells that expressed either ß-galactosidase or GFP. It is not known whether the proliferation of the ß-galactosidase- or GFP-immunoreactive cells reflects an unregulated expression of the transforming oncogene (SV40 large T-antigen) used or if it is a result of proliferating signals from the host. Our observations of proliferating cells 4 months posttransplantation are not consistent with the long-term in vivo study on HiB5 cells also immortalized with the SV40 large T-antigen [28], and in vitro data of Frederiksen et al. [34] and Whittemore and White [22]. Nor is it consistent with the finding that the temperature-sensitive Large-T oncoprotein is rapidly degraded soon after transplantation, shown by Cattaneo et al. [35]. On the other hand, our results are in accordance with Ader and coworkers [20], who have shown long-term survival and migration of mitogen-propagated precursor cells, observations that were indicative of long-lasting proliferation.
Morphology and Phenotypic Characteristics of the Implanted Cells
In our previous study [21], grafted cells integrated in a similar way as in the present study, and some of the implanted cells expressed vimentin, synaptophysin, or P75 (low affinity nerve growth factor receptor). In the present, long-term study, many grafted cells expressed oligodendrocyte markers. However, revealed by the morphology, some grafted cells resembled various retina-like phenotypes. Several different brain-derived and immortalized cell lines have previously been shown to integrate and differentiate in a site-specific manner after implantation into various regions of the neonatal and adult brain or spinal cord [11]. Indeed, the RN33B cell line has previously been shown to differentiate region specifically into cortical pyramidal-like cells in the neocortex and into neurons closely resembling hippocampal neurons when placed directly into that structure [11,30,32]. It is also known that the adult CNS can express signals inducing neuronal differentiation of implanted progenitors, such as RN33B cells [30,32,36]. The current observations also suggest that the adult, normal retina retains signals that allow for integration of the RN33B cells. However, the differentiation into retinal neurons is absent.
Oligodendrocytes are normally absent from the retina and the anterior part of the optic nerve. Thus, the ganglion cell axons are devoid of myelin within these regions. However, the ability for the ganglion cells to become myelinated exists, since the axons are covered with myelin in the posterior part of the optic nerve, that is, posterior to the scleral border (lamina cribrosa). In addition, myelination of the intraretinal part of the ganglion cell axon appears under various pathological conditions [37,38] or after transplantation of oligodendrocyte precursor cells to the retina [39]. The reason why the myelination process, under normal conditions, is inhibited anterior to the scleral border is not known. However, it has been speculated that this border exhibits some inhibitory factor(s) that prevents the migration of oligodendrocytes into the retina [40,41]. In the present study, transplanted RN33B cells migrated into the nerve fiber layer and the anterior part of the optic nerve, but never further than to the lamina cribrosa. In addition, many of these cells expressed Rip or MBP antigens, which indicates that these cells had the potential to produce myelin. The normal sequence of differentiation, in terms of chronology of differentiation markers, indicates that Rip and MBP expression stand for maturation [42]. Ader and coworkers [20] reported that a portion of striatal and spinal precursor cells transplanted to the retina of young mice differentiated into oligodendrocytes ensheathing the normally unmyelinated segments of the ganglion cell axons. In addition, these oligodendrocytes were found in virtually the entire nerve fiber layer of the host retina. The adequate cues for oligodendrocyte differentiation during development might be lacking and/or some inhibitory factor(s) might exist. The present study supports the view that these cues do exist, but at a level that would only permit differentiation under extreme circumstances, i.e., subretinal transplantation and/or the use of certain precursor cells, and only in the vicinity of the nerve fiber layer and the optic disc/nerve.
|
CONCLUSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
ACKNOWLEDGMENT |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
This work is dedicated to Professor Berndt Ehinger on the occasion of his 65th birthday in September, 2002.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
. Invest Ophthalmol Vis Sci
2001;42:198a.This article has been cited by other articles:
|
|
 |
|
  A. B. Wojciechowski, U. Englund, C. Lundberg, and K. Warfvinge Survival and Long Distance Migration of Brain-Derived Precursor Cells Transplanted to Adult Rat Retina Stem Cells, January 1, 2004; 22(1): 27 - 38. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
