Minor Immunoreactivity in GDNF-, BDNF-, or NT-3-Treated Substantia Nigra Allografts

Glial-cell-line-derived neurotrophic factor (GDNF) stimulates the survival of dopaminergic neurons. Little is known, however, about the possible immune sequelae of GDNF exposure or of exposure to other putative trophic factors. To address these questions, pieces of mesencephalic tissue, substantia nigra, from 15-day-old donor embryos were transplanted into the anterior chamber of the eye of adult male Sprague- Dawley recipient rats. At 5-day intervals, an aliquot (0.5 μg) of GDNF, brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), or cytochrome-C (CC) was injected into the anterior chamber of the eye of the recipients, and the sizes of the transplants were measured. GDNF increased transplant survival and growth. On day 42, all rats were sacrificed, and the grafts were evaluated by cresyl-violet staining and by immunohistochemistry using antibodies raised against neurofilament (NF), tyrosine hydroxylase, or glial fibrillary acidic protein (GFAP), as well as the following monoclonai antibodies: OX-38 anti-CD4, OX-8 anti-CD8, OX-18 anti-MHC class I, OX-6 anti- MHC class II, OX-42 anti-CD11b, R-73 anti-α and anti-β T-cell receptor, and EDI raised against monocytes/macrophages. BDNF-treated grafts showed only weak immunoreactivity, and even weaker reactions were seen in grafts treated with NT-3, GDNF, or CC. No single immune system marker was significantly elevated in grafts from any treatment group. We used OX-42 and EDI to study possible alterations of microglial components. Ramified microglial cells were found in GDNF-treated grafts and to a lesser extent in NT-3 and BDNF-treated grafts. EDl-labeled reactive microglial components were found in NT-3- and BDNF-treated grafts. Additionally, large and rounded OX-42-positive phagocytic cells were found in NT-3-treated grafts. Together with our previous finding that GDNF treatment of spinal cord transplants activates immune responses and leads to microglial activation, our data dempnstrate that although treatment with GDNF and to some degree with BDNF can enhance immune responses to immunogenic grafts, such as fetal spinal cord grafts, but the trophic factors per se do not elicit any marked response in non-immunogenic grafts like substantia nigra.


INTRODUCTION
Glial-cell-line-derived neurotrophic factor (GDNF) shows trophic effects not only on ventral mesencephalic dopaminergic neurons /4,6,11,18, 37,44,48/ but also on spinal cord motoneurons VOLUME 6, NO. 2,1997 /30,54/, but little is known about the possible immune sequelae of host and graft exposure to GDNF or to other putative trophic factors. In allogeneic but not syngeneic intraocular fetal spinal cord transplants, GDNF upregulates the expression of major histocompatibility (MHC) loci antigens and activates graft-derived microglial cells/40/. GDNF mRNA in the cortex and hippocampus is upregulated in such pathological conditions as experimentally induced status epilepticus in the hippocampus and striatum/38/. GDNF is a member of the transforming growth factor-13  superfamily /24/  Such proteins can be selectively administered into the anterior chamber, thus minimizing remote or indirect drug effects/6,18,44,49/. For example, fetal substantia nigra tissue grafted to the anterior chamber of the eye responds to GDNF /44/. Of particular relevance for immunological studies is that allogeneic grafts, even when manifesting good survival and growth, contain significant amounts of immunoreactive elements. In addition, area-specific differential immune responses are elicited aRer grafting/39/. In the study presented here, we used embryonic substantia nigra transplants and an in oculo graft model as appropriate targets for GDNF activity to determine whether the TGF-13 superfamily trophic molecule, in addition its stimulatory effects, might induce or interact with immune elements. To monitor the various elements of the immune system after GDNF administration, we used monoclonal antibodies (mAbs) raised against the following T-cell, B-cell, or monocytehnacrophage cell-surface antigens: rat CD4 (OX-38), rat CD8 (OX-8), rat MHC class I (OX-18); rat MHC class II (OX-6). MAb (OX-42) raised against rat CDllb identifies cellular complement-receptor type 3 (CR3), which is found on the surface of classical (ramified) and ameboid and/or reactive microglial subtypes /14,36,/45/. MAb ED-1, raised against rat monocytes and macrophages, is a cytoplasmic marker of macrophages or activated microglia or both/8/that localizes only to cells of the monocytic lineage /9,45/. We used allogeneic grafts to vary the basal immune status of the CNS tissue. For purposes of comparison, we tested other trophic factors from the neurotrophin family, including brain-derived neurotrophic factor (BDNF), which has survivalpromoting activity on substantia nigra/17,21/, and neurotrophin-3 (NT-3), which influences immune reactions in grafted CNS tissue/40/.

Intraocular grafting procedures
Young adult male Sprague-Dawley rats (SD rats, B & K, Sweden) weighing 150 g each were used as recipients of intraocular grafts. Fetuses from pregnant female rats were used as donors of embryonic mesencephalic tissue grafts. On the 15th embryonic-day (El5), pieces (1-3 mm Wide) of substantia nigra were dissected out from the fetuses and bilaterally grafted, under ether anesthesia, to the anterior eye chamber of the eye of the adult hosts. To prevent prolapse of the iris, the eyes were pretreated with a drop of 1% atropine solution. The grafts were placed on the anterior surface of the iris through a small opening in the cornea as previously described/29/. Every 5th day aer grafting (up to day 40), the volume of each transplant was estimated by stereomicroscopic observation, measuring the longest diameter multiplied by the diameter perpendicular to it. Every 5th day, each eye was injected with 5 L of a solution containing 0.5 lag of one of the following factors: (a) GDNF (100 g/mL), (b) cytochrome C (CC, 100 lag/mL), (c) BDNF (100 lag/mL), (d) NT-3 (100 lag/mL) /49/. Donor material from any one dam was equally distributed among all treatment groups. The measured sizes of the gratis correlate well with the actual weight ofthe grats at sacrifice/5/. Immunohistochemistry Forty-two days after transplantation, the host animals were deeply anesthetized with sodium pentobarbital (60 mg/kg, i.p.) and perfused via the ascending aorta with 50 mL of calcium-free tyrode solution (37 C) and then with 300 mL of ice-cold fixative (4% paraformaldehyde in phosphate buffer).

DISCUSSION
The combined data from this and other studies performed in our laboratories demonstrate that although treatment with GDNF and to some degree with BDNF can enhance immune responses to such immunogenic tissue as fetal spinal cord grafts, the trophic factors per se do not elicit a significant response in non-immunogenic grafs like substantia nigra. In the present study, although GDNF treatment led to significantly larger substantia nigra graft volumes than CC treatment, both GDNF and CC showed almost the same level of immune reactivity, except for MHC class I positive elements. GDNF-treated grafts also had the highest values of normal ramified microglial distribution, as well as lower numbers of immunoreactive microglial components and OX-42 positive round and large microglial cells.
The origin of microglial cells is controversial /20,35/42/, although most studies have suggested a mesodermal origin /7,25,27,33/. Three types of microglial cells are found in the CNS. Ramified microglia, prominent in the mature CNS, are derived from ameboid microglia, which, in turn, are present during late prenatal and early post-natal ages. Reactive microglia appear primarily in cases of CNS injury (reviewed in /19,27,33,45/). After 6-OHDA-induced dopaminergic denervation, microglial cells in substantia nigra have been shown to express MHC loci /1/, and ED1 and OX-42positive microglial cells have also been found in the substantia nigra area (Shinoda, Lindqvist, and Olson, unpublished).
As mentioned before, GDNF belongs to the TGF-13 superfamily /24/ of immunomodulators. VOLUME 6, NO. 2,1997 TGF-13 is produced by a large number of different cell types, including various cells of the immune system, such as macrophages, peritoneal monocytes and neutrophils, as well as by Tand Blymphocytes. In cases of injury (localized inflammation), platelet aggregation and degranulation, one of the earliest events in the inflammatory response, might occur and trigger the release of TGFI3. Activated macrophages release TGF-13 and other cytokines (for example, intedeukin-1 (IL-1), tumor necrosis factor (TNF), fibroblast growth factor (FGF), and platelet-derived growth factor (PDGF). procedures, in which 0.1, 1 gtg, and 1 mg/eye GDNF were given to stimulate neuronal growth. In the present study, the final sizes of GDNF-treated substantia nigra transplants were more than 1.5-fold larger than those in Stromberg's experiments. The reasons for such differences are as follows: (a) We used 8 injections of GDNF at 5-day intervals, and the final period was 42 days after the first injection, whereas in the former experiments the final period was only 33 days.
Nerve growth factor (NGF) and other neurotrophins have been shown to produce several immunomodulatory effects, such as in mast cells /2,3,32/, B-cell stimulation /31,43,46/ and T-lymphocyte stimulation/47/. In the present study, NT-3, which had no effect on graft growth, elicited small immunomodulatory effects, particularly microglial activation. The significantly increased numbers of OXo42-positive large and round cells in NT-3-treated transplants suggests that NTo3 stimulates macrophage-related cytotoxic immunity in neuronal cells. NTo3 and NT-4 mRNA, as well as truncated trkB and trkC transcripts, have been found in rat thymus, thymic stroma (tissue depleted of mononuclear cells), spleen, and splenic stroma /23/. Concanavalin A-treated rat-thymus cells and lipopolysaccharide-treated rat-spleen mononuclear cells express a two-fold increase in NT-4 but not in NTo3 or BDNF. We have previously shown that BDNF has almost no immunomodulatory effect on fetal spinal cord grafts in oculo/40/. Thus it appears that BDNF does not elicit a strong immunoreaction in vivo.
In conclusion, neither GDNF nor BDNF enhanced immune reactivity in allogeneic embryonic substantia nigra grafts. GDNF may act to upregulate immune markers in allogeneic CNS tissue grats where MHC Class I is strongly expressed, such as in spinal cord, but not in allogeneic CNS tissue in which MHC Class I is weakly expressed, such as in substantia nigra.