[1] 17w Sertoli cells decrease microglial response and increase engraftment of human hNT neurons in the hemiparkinsonian rat striatum

[1] 202w Sertoli cells (SCs) provide immune protection and nutritive support to the developing germ cells in the testis. Sertoli cells have also been shown to provide immune protection to islets transplanted outside the testes. In this study, the ability of these cells to diminish the infiltration/activation of microglia into a neural graft implanted in the lesioned striatum of a hemiparkinsonian rat was investigated. Human neuron-like cells (hNT neurons) were implanted either alone or in combination with rat SCs. Three months later, the animals were sacrificed and immunohistochemistry was performed to determine the survival of the xenografted neurons as well as microglial infiltration/activation. Cotransplantation of the SCs with the hNT neurons increased graft survival and was associated with an increase in graft size. Furthermore, there were fewer microglia present in the grafted tissue of the cotransplantation groups. These results show that SCs retain their immunosuppressive ability even within the brain. As immune responses to grafted neural tissue within the central nervous system become better understood, this ability of the SCs to provide localized immunosuppression to the transplanted tissue may become more important. This is particularly true as the search for alternative sources of neural tissue to treat neurodegenerative diseases expands to encompass other species.

[1] 246w The Sertoli cell (SC) is intrinsic to the testis where its function is to provide an appropriate environment for the development of the antigenic germ cells. In the testis, the SC provides an immunological barrier for the germ cell by both forming the blood-testis barrier and also producing a paracrine factor that can inhibit interleukin-2 (IL-2) production and T-lymphocyte responsiveness to IL-2 [22]. These cells also produce CD-95 ligand (FasL), which when bound to the CD-95 receptor (Fas) on activated lymphocytes induces apoptosis and the subsequent down-regulation of the im-mune response. Bellgrau et al. [2] showed that mouse SC express FasL constituitively and survive indefinitely after being transplanted under the kidney capsule, while SC from FasL deficient gld mice do not survive. The FasL mechanism, however, is controversial because if you transfect islet cells to express FasL, these cells appear to be preferentially destroyed upon transplantation [7]. Even though the mechanism of immune protection is not yet fully understood, SCs have been shown to increase the survival of co-transplanted islets in a diabetic rat model [9,20,21,23,24]. Further, lectin histochemistry has been used to show that the microglial infiltration of adrenal chromaffin cell grafts in the striatum is reduced when SCs are cografted with these dopamine (DA)producing cells [16]. We have also obtained evidence that the SC may reaggregate in the striatum [16]. In this way, the cells may segregate the grafted neurons to prevent their immune detection or the migration of immune cells into the graft.

[2] 78w The purpose of the current in vivo study was to determine whether allografts of SCs cotransplanted with dopaminergic neurons obtained from an immortalized cell line of human neuron-like cells (hNT neurons) in the 6-hydroxydopamine (6-OHDA) lesioned rat striatum would enhance the graft survival. By transplanting the hNT neurons, it was possible to examine whether the SC could decrease the host immune response in a xenogeneic situation in which graft rejection would occur without some form of immune suppression.

[1] 124w The grafted hNT neurons were localized within the host striatum through immunostaining for human neural cell adhesion molecule with the MOC-1 antibody. In the hNT alone group, only 50% of the grafts survived, while in the two hNT ϩ SC groups, 100% of the grafts were still present 3 months later. As can be seen in Fig. 1A-H, the grafts in the hNT alone group were smaller by one-half than in the cocultured hNT ϩ SC (H ϭ 5.44, p Ͻ .06; see Fig. 1A and B). There was also a tendency for the grafts of the hNT ϩ SC group that were not cocultured prior to transplantation to be bigger than those of the hNT group alone, but smaller than the cocultured group.

[2] 69w Sections were also stained with the monoclonal antibody OX-42, which is directed against the C3 complement receptor on microglia. From Fig. 1C and D, it is clear that there is more OX-42 reaction product visible in the hNT alone group compared to the cocultured hNT ϩ SC group. The amount of OX-42 present in the graft tissue and the surrounding host striatum was quantified by 442 WILLING ET AL.

[3] 139w measuring the intensity of reaction product labeling from all sections with graft present. These results were then expressed as the intensity of staining per vol. of graft in cubic mm. When standardized for graft vol., there were significantly fewer microglia present in the cocultured hNT ϩ SC group (H ϭ 6.47, p Ͻ .04; see Table 1) while the difference between the two hNT ϩ SC groups was not significant (U ϭ Ϫ1.55, p Ͻ .12). Conversely, the fewer microglia present, the larger the graft size. When sections were labeled for MHC-II to indicate microglial activation (OX-6), MHC-II positive microglia were present in all groups (Fig. 1E-H). When expressed as the number of positive cells per cubic mm; there were fewer activated microglia in the hNT ϩ SC cocultured group (see Table 1) than in the hNT alone grafts.

[1] 172w Cotransplantation of the SC with hNT neurons increased the number of surviving grafts as well as the size of the surviving grafts. The incidence of graft survival was 100% in the cotransplanted groups, but only 50% in the hNT alone transplants. Fur-thermore, as determined by the immunohistochemical labeling of the grafts with the hNT neuron-specific antibody, MOC-1, the grafts in the hNT ϩ SC coculture group were almost twice the size of the grafts in the hNT alone group. Because an hNT specific label was used, this doubling of graft size is likely to be a result of an increase in the number of surviving hNT neurons. However, without a specific marker that allows us to unequivocally label every hNT neuron, the alternate hypothesis, that the hNT neurons are interspersed among the cotransplanted SC and, therefore, occupy a greater vol. within the striatum cannot be dismissed. If this is the case, then the density of the MOC-1 staining would be less in the cotransplanted groups. This is not observed in these grafts.

[2] 200w The increased graft survival is consistent with the earlier demonstrations of Selawry et al. [20,21,24] and Korbutt et al. [9] that SC could enhance the survival of islet cells when they were cotransplanted under the kidney capsule of diabetic rats. With little or no immunosuppression with cyclosporin, normoglycemia was reestablished. Initial verification of these early studies in diabetic rats demonstrated that the SC can survive in the striatum without immunosuppression [19]. Furthermore, these data expand on the earlier demonstrations of the possible immune-modulatory capabilities of the SC in vitro [16 -18]. The ability of the SC to decrease the infiltration and activation of microglia around the graft site is the most powerful demonstration yet that the SC can provide extratesticular immune protection. In the hNT alone group, the degree of microglial infiltration/activation was much greater than in the cocultured group; this greater activation was accompanied by the survival of fewer grafts and these grafts were smaller. Given that the SC can decrease the microglial response to these xenografts, it is unlikely that the SC themselves would elicit an increase in microglial infiltration. However, further studies to determine the microglial response to SC transplants alone are necessary to address this issue.

[3] 132w Until recently, the central nervous system (CNS) has been considered an immunologically privileged site within the body. The presence of the blood-brain barrier (BBB) normally prevents the infiltration of peripheral cells into the CNS. The normal inflammatory response in the brain includes the activation of astrocytes, macrophages and microglia, and the infiltration of macrophages. However, with transplantation, the BBB is disrupted, allowing graft MHC antigens to be recognized, thereby activating a systemic immune response and lymphocyte infiltration into the graft. Because the BBB in rodent models can take as long as 1 month to close and then may still remain permeable to larger proteins [15], there may be unhindered surveillance of the transplanted tissue and unhindered T and B cell infiltration of the graft [5,8,25]. This will be examined in future studies.

[4] 202w In clinical studies, it is common to remove patients from cyclosporin treatment 6 -18 months postsurgery. In patients that have died of unrelated causes, it is possible to find viable allografts even without long-term immune suppression [10,11,14,26]. However, it is not known if the initial immunosuppression is critical to graft survival or not. Furthermore, in those patients in whom immune markers have been examined, there were immune cells present in the grafts, even though these were allografts where activation of the immune response is thought to be minimal [10]. The significance of these findings is not currently known. As researchers look to other species to provide an abundant source of DA neurons, the issue of providing localized immune suppression will become more important [6]. Even in a xenograft situation, where graft rejection is likely, the SC were able to protect the grafts from rejection. However, a more thorough examination of the SCs' ability to provide immune protection in the CNS is planned and must entail the identification of T cells, B cells, and macrophages in and around the graft site. In conclusion, we have demonstrated that the SC can be a powerful new tool to enhance the survival of transplanted DA neurons.

[1] 71w Fifteen male Sprague Dawley rats weighing 275-325 g at the time of lesioning were assigned to either hNT neurons alone, hNT neurons ϩ SC cografts, and hNT ϩ SC, cocultured and cografted. Three animals died prematurely and were excluded from the analysis. The animals were group housed with ad lib access to food and water in a temperature controlled room on a 12:12 h light cycle (lights on at 0600 h).

[2] 186w The SCs were isolated from prepubertal male rats (16 -19 days old) as previously described [4,12,16]. The tunica albugine was removed from the individual testis and the tissue was sequentially digested first with trypsin (0.25%) and then with collagenase (0.1%) at 37°C. The resulting SC aggregates were plated in 1.0 ml of media (DMEM:F12, Gibco, BRL, Grand Island, NY, USA), 0.1% insulin-transferrin-selenium (ITS; Sigma, St. Louis, MO, USA), 50 ng/ml retinol acetate (Sigma), and 50 g/ml gentamicin sulfate (Sigma) per 16 mm tissue culture wells and incubated at 39°C in 5% CO 2 -95% air for 48 h. The SC cultures were then washed twice and returned to the CO 2 -injected incubator at 37°C in 5% CO 2 , 95% air for another 48 h. The resulting pretreated Sertoli-enriched monocultures contained greater than 95% SC as described previously [3,13]. The cells were gently trypsinized and centrifuged twice at 800 rpm for 2-3 min in DMEM:F12 and viability assessed using trypan blue dye exclusion. The cell concentration was adjusted and the cells were replated either alone or 1:1 in coculture with hNT neurons for 72 h.

[3] 68w The hNT neurons were obtained from Layton Bioscience, Inc. (Atherton, CA, USA). They were thawed quickly at 37°C and transferred to a 15 cc centrifuge tube with DMEM:F12. They were then centrifuged for 7 min at 1000 rpm, the supernatant removed, and the cells resuspended in media. The hNT neurons were then cultured for 72 h prior to transplantation either alone or in a 1:1 coculture with SC.

[4] 69w Just before transplantation, the cells in culture were lightly trypsinized, centrifuged three times at 700 rpm for 3 min, and then resuspended in HBSS/HEPES (Gibco, BRL). Viability and cell number were assessed using the trypan blue dye exclusion method. For the hNT alone group, the cell concentration was adjusted to 20,000 cells/l. For the cotransplants, the cell concentration was adjusted to 40,000 cells/l (20,000 hNT cells/l plus 20,000 SC/l).

[5] 151w The rats were anesthetized with Equisthesin (3.5 ml/kg) and placed in a stereotaxic frame. Using a 22-gauge needle on a 10 l Hamilton syringe, the transplants were placed in the striatum (1.2 mm anterior to bregma, ϩ2.6 mm laterally and Ϫ5.2 and Ϫ4.7 mm ventral to dura with the toothbar set at zero). For the hNT ϩ SC cograft groups, the two cell types were mixed prior to the first surgery while for the hNT ϩ SC coculture and cograft group, the cell suspensions were mixed at the time of coculturing. Both of these cosuspensions were remixed at transplantation. The cell suspensions were delivered at a rate of 1 l/min, with 2 l being deposited at each injection site. The needle was left in place for 5 min after the last injection and then withdrawn. Cyclosporin (6 mg/kg) was administered the day of surgery, the day following surgery, and then discontinued.

[6] 201w Three months after transplantation, the rats were anesthetized with 10% chloral hydrate and transcardially perfused with 50 ml of 0.1 M phosphate buffer followed by 250 ml of 4% paraformaldehyde in 0.1 M phosphate buffer. The brains were removed, postfixed 24 h in paraformaldehyde, cryoprotected with 20% sucrose, and then sectioned at 30 in a cryostat. All the immunohistochemistry was performed free-floating using the ABC method. Endogenous peroxidase activity was quenched by incubating the tissue with 3% hydrogen peroxide and 10% methanol. After blocking the tissue with 1% NHS, the sections were then incubated overnight at 4°C in primary antibody. The primary antibodies employed were the MOC-1 (CD56 clone 1:200, Dako, Carpinteria, CA, USA) labeling human neuron cellular adhesion molecule, which is expressed by the hNT neurons, OX-42 (1:300, Serotec, Oxford, UK), which labels microglia, and OX-6 (1:900, Serotec), which labels MHC-II antigens. The sections were rinsed and then incubated in biotinylated rat adsorbed horse antimouse secondary antibody (1:300, Vector, Burlingame, CA, USA) for 60 min. Following rinsing, the sections were placed in ABC for 60 min and visualized with DAB (2 min). The sections were coverslipped with Permount and the number of TH-positive cells was counted using unbiased counting techniques.

[7] 172w Data are reported as means Ϯ SEM. With the OX-42 and MOC-1 staining, the density of the reaction product was too great to make individual counts. In these cases, measurement of the intensity of immunohistochemical labeling was performed using the Image Pro computer analysis system. An image of the graft within the striatum and of the contralateral striatum were captured digitally at 25ϫ magnification. The optical density (intensity) of the immunohistochemical reaction product within the graft was corrected for background staining by subtracting the optical density of the contralateral striatum. The graft was outlined on the computer screen and the optical density within this border was measured. In the case of the OX-6 staining, it was possible to identify individual microglia. Therefore, the cells were counted using the computer analysis system and the counts corrected using Abercrombie's formula [1]. The results were not normally distributed, making an analysis of variance (ANOVA) inappropriate. These results were tested using a Kruskal-Wallis nonparametric test. Post hoc analysis was done using the Mann-Whitney test where appropriate.

[1] 125w The animals were anesthetized with Equithesin (3.5 ml/kg i.p.) and placed in a Kopf stereotaxic frame. A single injection of 6-OHDA (Sigma; 2.5 l, 3.6 g/l for a total dose of 9 g in 0.2% ascorbic acid) was delivered at a rate of 1 l/min at 4.4 mm posterior to bregma, Ϫ1.2 mm laterally, and Ϫ7.8 mm ventral to dura with the toothbar set at Ϫ2.4 mm. Five min after the injection, the needle was slowly withdrawn and the incision was sutured with wound clips. Three weeks postlesion, the animals were tested in an automated rotometer for apomorphine-induced rotation (0.2 mg/ kg, s.c.). Those animals that failed to rotate at least eight turns/min over the 30-min test on three occasions were excluded from the study.