Interactions of neurotrophic factors GDNF and NT-3, but not BDNF, with the immune system following fetal spinal cord transplantation

BRAIN RESEARCH ELSEVIER

Brain Research 722 (1996) 153-167

Research report

Interactions of neurotrophic factors GDNF and NT-3, but not BDNF, with the immune system following fetal spinal cord transplantation Masaki Shinoda a, *, Barry J. Hoffer b, Lars Olson

a

a Department ofNeuroscience, Berzelius Laboratory, Karolinska Institute, S-171 77 Stockholm, Sweden b Department of Pharmacology, University of Colorado Medical School, Denver, CO, USA

Accepted 7 February 1996

Abstract Glial cell line-derived neurotrophic factor (GDNF) is known to stimulate survival of dopaminergic and spinal cord motor neurons. However, little is known of the possible immune sequelae of GDNF exposure, or that of other putative trophic factors. To address these questions we utilized in oculo grafts of spinal cord, wherein we could induce different levels of immune responses via allogeneic vs. syngeneic combinations. Adult female Sprague-Dawley and Fisher rats were used as hosts for allogeneic and syngeneic grafts, respectively. Embryonic age 14-15-day-old fetuses were taken from pregnant dams of each strain, and cervical spinal cords were removed and dissected. Pieces of the spinal cord were transplanted into the anterior chamber of the eye within each strain. At 5-day intervals, 0.5/zg of GDNF, br~dn-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3) or cytochrome c (CC) was injected into the anterior chamber of the eye and the sizes of the transplants were measured for the Sprague-Dawley rats. The same injections and measurements, but only for GDNF and CC, were carried out using Fisher rats. As expected, GDNF increased transplant survival and growth in both the Sprague-Detwley and Fisher animals. At day 41-42, all rats were sacrificed. Cameral graft appearance was evaluated by cresyl violet and immunohistochemically using antibodies against neurofilament (NF), calcitonin gene-related peptide (CGRP) and glial fibrillary acidic protein ((3FAP). To monitor immune responses, the following monoclonal antibodies were used: OX38 against CD4, OX18 against MHC class I (MHC1), OX8 against CD8, OX6 against MHC class II (MHCII), OX42 against CD1 lb, R73 against a and fl T cell receptor (TcR), and ED1. In the Sprague-Dawley grafts, significantly higher amounts of CD8 + , T lymphocyte + , MHCI + and MHCII + antigen-presenting cells (APC) were observed in GDNF-treated transplants. These markers were also increased in NT-3-treated groups. There were two types of OX-42 + cells, one was the ordinary ramified microglial cell, the other appeared to be a phagocytic cell, looking like the interstitial proliferating variety. Interestingly, the phagocytic OX-42 + cells had the same distribution as ED1 + and MHCII + cells. In contrast, th,ere were few immunoreactive cells after GDNF treatment in the inbred Fisher animals, similar to the CC control group. These results suggest that GDNF and to some extent NT-3, can activate the immune system in allogeneic graft combinations, but that these trophic factors do not produce overt rejection, and do not per se induce immune responses. Keywords: Transplantation Immunology; Glial cell line derived neurotrophic factor; Brain derived neurotrophic factor; Neurotrophin-3; Ramified microglia; Reactive microglia

1. Introduction Glial cell line-derived growth factor (GDNF) is a recently discovered member of the transforming growth factor-/3 (TGF-/3) superfanaily [29]. Recent papers have reported that G D N F has trophic effects, not only on ventral Abbreviations: GDNF, glial cell line-derived neurotrophic factor; BDNF, brain-derived neurotrophic factor; NT-3, neurotrophin-3; CC, cytochrome c; MHCI, major histocompatibility class I; MHCII, major histocompatibility class II; TcR, T cell receptor; APC, antigen presenting cell; TGF, transforming growth factor; IFN-% interferon-y; IL-1, interleukin-1; IL-2, interleukin-2; TNF-ct, tumor necrosis factor-a. * Corresponding author. Fax: (46) (8) 32-3742. 0006-8993/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved PII S0006- 8993(96)00208-9

mesencephalic dopaminergic neurons [4,50,52], but also on spinal cord motoneurons [39,61] and on spinal cord transplants [54]. Moreover, GDNF m R N A appears to be upregulated in pathological conditions such as experimental status epilepticus in hippocampus and striatum [45]. The TGF-/3 family proteins are well known as immunological modulators as well [59]. Amongst the suppressant functions of TGF-/3 in the immune systems are: 1, production and secretion by a variety of immune cells; 2, inhibition of proliferation of thymocytes, T cells, B cells, and natural killer cells; 3, inhibition of generation of cytotoxic macrophages; 4, inhibition of IgG and IgM production; 5, switching B cells from IgG to IgA production; 6, modula-

154

M. Shinoda et al. / Brain Research 722 (1996) 153-167 8

tion of cytokine production [interferon-7 (IFN-7), tumor necrosis factor-a ( T N F - a ) and interleukin- 1 [3 (IL- 1/3); and 7, chemoattraction for monocytes and neutrophils. Other authors, in contrast, have reported that TGF-/3 proteins can elicit strong immunological reactions (for reviews see [11,22,56,59]). However, there are no reports of the possible effects of GDNF on the immune system, either in terms of upregulation or suppression. Microglia are of particular interest in CNS immunology. These glial elements can be divided into several morphological subclasses. Ramified microglial cells are present in the normal adult CNS. Ameboid microglial cells are seen in the developing CNS. Reactive microglia is found in the injured CNS (for reviews see [26,40,51]). In cases of increased immunological reactivity, for example in conjunction with CNS infections, Rio-Hortega [43] found hypertrophic microglia a n d / o r ameboid microglia with phagocytic activities in human CNS tissue, and indicated that they were derived from the ramified type. Intraocular transplantation of fetal CNS tissue provides a unique method to examine in vivo effects of trophic factors. These peptides can be selectively administrated into the anterior chamber, which minimizes remote or indirect drug effects [7,14,50,54]. We have previously demonstrated that fetal spinal cord tissue survives grafting to the anterior chamber of the eye [19]. Such transplants receive functional inputs from cografts of brainstem nuclei containing serotonin [20,21] and noradrenaline neurons [18,25]. Intraocular spinal cord tissue also becomes functionally connected to cortex cerebri, spinal cord and skeletal muscle cografts [53]. Given these organotypic properties of spinal cord tissue transplants in oculo, we have utilized this model to study immune responses to trophic factor exposure. Of particular relevance for immunological studies is that allogeneic grafts, even if they manifest good survival and growth, contain significant numbers of im-

7 6

- o - CC

s .~

4

.-43- NT3

'~

s

~

~

2

BDNF

1 0 0

5

10

15 Days

20

25

30

35

40

postgra~g

Fig. 2. Growth curve of neurotrophin-treated intraocular transplants. Neurotrophin-treated groups showed somewhat more growth. mune elements. Some of these cells are reactive microglia [46]. In this study we sought to use the in oculo graft model, and spinal cord transplants known to be appropriate targets for GDNF activity, to determine if this TGF-/3 superfamily trophic molecule might cause or interact with immune mechanisms. We utilized both allogeneic and syngeneic grafts to vary the basal immune status of the CNS tissue. Finally, we tested additional spinal cord trophic factors, in the neurotrophin family, for purposes of comparison. These factors include BDNF and NT-3, both of which are reported to have survival-promoting activity on spinal cord tissue [54].

2. M a t e r i a l s a n d m e t h o d s

2.1. Intraocular grafting procedures

Thirty-nine young adult (150 g) male Sprague-Dawley rats (SD rats; B and K, Sweden) and 12 young adult

a

b

Sprague-Dawley strain

Fisher strain *

9

12

8 7

i6

--o- CC -O-GDNF

.~ s

2

2

1 0 0

5

10

15

20

25

Days postgrafting

30

35

40

0

5

10

15

20

25

30

35

40

Days p o s t g r a f t i n g

Fig. 1. a: growth curves of intraocular transplants given GDNF (squares) or cytochrome c (circles) in Sprague-Dawley rats. The transplants with GDNF treatment show increased growth (mean + S.E.M., * P < 0.05, Mann-Whitney). b: growth curves of intraocular transplants with GDNF or cytochrome c treatment in Fisher rats. The transplants are larger in GDNF-treated group (mean + S.E.M., * P < 0.05, Mann-Whitney).

M. Shinoda et al. / Brain Research 722 (1996) 153-167

(150-200 g) male Fisher rats (Charles River, Germany) were used as recipients of intraocular grafts. Fetuses from the pregnant rats of the same strain were used as donors of spinal cord tissue grafts. Pieces of spinal cord 1-2 mm 3 in size, from fetuses at embryonic day 15 (El5), were dissected out and bilaterally grafted to the anterior chamber of the eye of adult hosts. The graft surgery was performed under ether anesthesia and eyes were pretreated with a drop of 1% atropine solution to prevent prolapse of the iris. Grafts were placed on the anterior surface of the iris through a small opening in the cornea as previously described [38]. Every fifth day after grafting, up to 40 days, the volume of each transplant was estimated by stereomicroscopic observations, measuring the longest diameter multiplied by the diameter perpendicular to it. Eyes were injected with 5 /zl solutions containing 0.5 /zg of GDNF (100 ;zg/ml) [54], cytochrome c (CC, 100/zg/ml), brain derived neurotrophic factor (BDNF, 100 ;zg/ml), or neurotrophin-3 (NT-3, 1 0 0 / z g / m l ) every fifth day for the SD grafts. For the Fisher animals, only GDNF and CC treatment groups were studied. Donor material from any one dam was equally distributed amongst all treatment groups. The measured sizes of intraocular grafts have been shown to correlate well with the actual weight of such grafts at sacrifice [5].

Immunol'ogical Total Score

155

2.2. Immunohistochemistry At 41-42 days after the transplantation, host animals were deeply anesthetized with sodium pentobarbital (40 m g / k g i.p.) and perfused via the ascending aorta with 50 ml of 37°C calcium-free Tyrode solution followed by 300 ml of ice-cold fixative (4% paraformaldehyde in phosphate buffer). Transplants attached to the host irises were postfixed by immersion in the same fixative for 30-60 min, rinsed in 10% sucrose, frozen and sectioned on a cryostat at 14 /zm. Sections, at periodicities of eight to twelve, were used for glial fibrillary acid protein (GFAP) (1:100), neurofilament (NF) (1:400), and CGRP immunohistochemistry (1:400), as well as cresyl violet and toluidine blue staining. The following mouse monoclonal antibodies were used to monitor various elements in the immune system: rat CD4 antigen (clone OX-38), rat CD8 antigen (clone OX-8), rat CD1 lb antigen (clone OX-42), rat MHC class I antigen (clone OX-18), rat MHC class II antigen (clone OX-6), rat c~ and fl T cell receptor antigen (clone R73) and rat ED1 antigen against monocytes and macrophages [9,49]. Primary antibodies were diluted 1:1000 except ED1 which was diluted 1:200. All tissues were stored in a humid chamber at 4°C for 48 h. Secondary antibodies were fluorescein linked anti-mouse IgG, for OX-38, OX-8, OX-

OX42 points 4-

points

3.5 ~

8-

1 -'-I

7_"

32.5

f r

2 1.5

1 0

ii1 &

GDNF

NT3

1 .5 0

'C'C" GI~NF NT3

ED1

BDNF I~C

r r

points

l~GbrqF i

t

43

Bfr~ Fdc FGO~

4 3.5 3 2,5 2 1.5 1 5 0 CC

GIgNF

N;F3

BE)NF

FCC

FGDNF

Fig. 3. Immunological scores for th,~ Sprague-Dawley grafts. Here and in Fig. 4 the protein administration is indicated under the corresponding bar and for the Fisher grafts, the protein administrated is proceeded by an 'F'. Left: total immunological scores for each trophic factor in each transplant group. The strongest immunological reactions were found after GDNF treatment in the S-D group (mean + S.E.M., * P < 0.05, * * P < 0.01, * * * P < 0.001, Mann-Whitney). Upper right: OX-42- (CDl lb) positive interstitial elements in each treatment group. GDNF-treated allogeneic grafts had the largest amount of interstitial elements. Lower right: EDl-positive reactive glial cells in each treatment group. As with OX-42, GDNF-treated Sprague-Dawley grafts had the largest score (mean -_- S.E.M., * P < 0.05, * * P < 0.01, * * * P < 0.001, Mann-Whitney).

M. Shinoda et al. / Brain Research 722 (1996) 153-167

156 MHCI

MHCII

r

TcR I

points

,

points

~.

4, 3.5,

3"

f

points

1

2

1.6,

1.8

1.4

I

, ,*

1.6

1.2 1.4

2 1,5,

1.2

I

1

.8

.8

.6

.6

,4

1, .4

.5,

.2

.2

0

0

CC GDNF NTa

B6NFFCC FGbNF

_L. c c GI)NF NT3 BDNF FCC FGDNF

0 CC

G D N F NT3

B D N F FCC F G D N F

CD8

CD4

I~knts

points

1.8.

.18

1.6,

.16

1.4.

.14

1.2.

.12 .1

1.

.06

,S.

,06,

.6,

.04.

.4.

,02.

.2.

_[_ _L _L ,, , ,,, ,, , CC GDNFNT3 BDNFFCC FGDNF

_I_ K~ . , ,,

_I_ ,,,

0

,

TI

CC

GDNF

NT3 B D N F F C C

FGDNF

Fig. 4. Scores for other immunological elements. There were many MHC loci-positive APCs and significant numbers of CD4- and CD8-positive lymphocytes (mean + S.E.M., * P < 0.05, * * P < 0.01, * * * P < 0.001, Mann-Whitney).

42, OX-18 and R-73, and anti-rabbit IgG fluorescein-isothiocyanate (FITC) for GFAP, NF and CGRP [46,47]. Immunological scores were based on cell counts as follows: scores of 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5 or 5 were given to grafts where average numbers of immunoreactive cells/field of view using a 20 X lens were 0-2,

areas measured using the field of view as a measuring unit: scores of 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5 or 5 were given to grafts where these elements were found in 0-4, 5-14, 15-24, 25-34, 35-44, 45-54, 55-64, 65-74, 7 5 84, 85-94 and > 95% of each graft.

3-7, 8-12, 13-17, 18-22, 23-27, 28-32, 33-37, 38-42, 4 3 - 4 7 and > 48, respectively. When, in addition to cells, immunoreactive glial and or fibrous processes were also noted, the scores were incremented by 0.5, 1 or 2 for small, moderate or large amounts of such material seen with MHCI and II stains. A 'total immunological score' for each graft was calculated by summation of the 4 different antibody-specific scores for CD4, CD8, MHC class I and II. Scores of OX-42 (CD1 lb) and ED1 positivity were obtained by estimating the density of OX-42 calculating the percentage of the graft volume containing the markers. For each graft, three representative sections were analysed, and immunoreactive areas and total section

3. Results

3.1. Effects of trophic factors on graft growth Fig. 1a and b show the growth of intraocular spinal cord grafts with or without GDNF. In both strains GDNF led to the expected [54] larger graft volumes (Mann-Whitney, P < 0.05). As illustrated in Fig. 2, BDNF and NT-3 treatment of grafts in the outbred strain led to small, but nonsignificant increases in final sizes of grafts as compared to cytochrome c-treated controls.

Fig. 5. Typical appearance of transplants in micrographs after GDNF treatment in Sprague-Dawley strain ( x 96). (a) Cresyl violet, (b) NF, (c) MHCI (OX-18), (d) MHCII (OX-6), (e) TcR (R-73), (f) CD1 lb (OX-42), (g) ED1. There are many APCs with MHCI and II immunoreactivity. MHCI reactivity located at the graft surface, but not in CNS cells. MHCII reactivity is found throughout the transplants, but is especially related to pericytes. Interstitial appearance of microglia is seen with OX-42 staining. These are aggregated around vascular structures, and thought to be related to MHCII-positive APC. Neurofilament-positive fibers are sparse. There are many EDl-positive cells at the graft-iris interface, the surrounding vessels, and the outer surface (glia limitans membrane).

M. Shinoda et al. / Brain Research 722 (1996) 153-167

157

~r

-

?

#

'~'='~'° ~'~- ~ ~ " ~

l'~',.,

~

~

158

M. Shinoda et al. / Brain Research 722 (1996) 153-167

M. Shinodaet aL/ Brain Research 722 (1996)153-167 3.2. Overall immunological response to growth factors

Fig. 3 (left) summarizes the total immunological scores. GDNF-treated allogeneic grafts in the outbred SD strain contained large numbers of immune-related ceils. Not only lymphocytes were present, but also antigen-presenting cells (Fig. 4). NT-3 treatment also led to large amounts of immunoreactive material. GDNF treatment of syngeneic grafts, using the inbred Fisher strain, did not cause any detectable immunological reaction in comparison to the CC-treated syngeneic control grafts. Interestingly, CCtreated 'control' allogeneic grafts had higher immunological scores than CC 'control' syngeneic grafts (MannWhitney, P < 0.05). 3.3. Immunological patterns for individual growth factors

GDNF-treated transplants in Sprague-Dawley rats showed excellent neuronal survival (Fig. 5a and b). However, there were also many lymphocytes present (Fig. 5e). In addition, proliferation of MHCII-positive pericytes, probably APCs related to the vascular structures, and parenchymal MHCII-positive glial components (Fig. 5d), as well as OX-42-positive elements, and EDl-positive elements were found. The OX-42 and EDl-positive cells had the same distributions as the MHCII-positive components (Fig. 5f and g). NT-3-treated transplants in Sprague-Dawley rats also showed excellent neuronal survival (Fig. 6a and b). As for the GDNF-treated group from the same strain, proliferation of MHCII-positive elements and lymphocytic infiltration were also found (Fig. 6d and e). However, OX-42-positive interstitial elements and EDl-positive reactive microglial cells were much less prominent (Fig. 6f and g). In addition, both GDNF- and NT-3-treated transplants in Sprague-Dawley showed MHCI-positive structures (Fig. 5c and Fig. 6c). BDNF-treated transplants in Sprague-Dawley rats also showed good neuronal survival (Fig. 7a and b). Fine ramified microglial cells were seen using OX-42 antibodies (Fig. 7f) and these cells were also positive for MHCI; however, almost no ED1 positivity was found (Fig. 7g). In addition, there were occasional MHCII-positive cells, particularly in glia limitans at the graft surface, capillary pericytes (Fig. 7d) and a few lymphocytes (Fig. 7e). Cytochrome c-treated transplants in Sprague-Dawley rats showed similarly good neuronal morphology (Fig. 8a and b), resembling BDNF in SD rats. As in those transplants, fine ramified microglial cells could be seen with

159

OX-42 antibodies (Fig. 8c) and these were also MHCIpositive (Fig. 8f). Although some EDl-positive cells were identified (Fig. 8g), MHCII-positive cells and lymphocytes were quite sparse (Fig. 8d and e). GDNF-treated transplants in Fisher rats showed morphologically well-defined structures (Fig. 9a and b). Ramified microglial components could be identified with OX-42 (Fig. 9f) and MHCI antibodies (Fig. 9c). There were also a few MHCII-positive cells, lymphocytes and EDl-positive reactive microglial components (Fig. 9d, e and g). Cytochrome c-treated transplants in Fisher rats also showed morphologically well-defined structures (Fig. 10a and b). Ramified microglial cells were seen using OX-42 and MHCI antibodies (Fig. 10c and f). As with GDNFtreated Fisher rats, there were a few MHCII-positive cells, lymphocytes and EDl-positive reactive microglial components (Fig. 10d, e and g). 3.4. Morphological differences in microglial cells

Two types of microglial cells were distinguished, ramified and reactive microglia. Ramified microglial cells were identified using OX-42 antibodies (Fig. 7f/Fig. 10f and Fig. l la) and, in some cases, MHCI antibodies (Fig. 7c/Fig. 10c). On the other hand, reactive microglial cells were identified using ED1 antibodies (Fig. 5g/Fig. 6g and Fig. l lb). Interestingly, the distribution of reactive microglia matched that of MHCII-positive (Fig. 5e/Fig. 6e) and OX-42-positive (Fig. 5f/Fig. 6f) elements. In addition, every transplant with proliferation of EDl-positive cells also showed lymphocytic infiltration. 3.5. CGRP-positive motoneurons in transplants

In some cases, well-defined motoneurons were detected using CGRP immunohistochemistry (Fig. 10b, d and e). CGRP-positive motoneurons were particularly easy to recognize in less immunoreactive transplants.

4. Discussion

Recently, several authors have reported that GDNF exerts trophic support for dopaminergic systems [4,50,52]. GDNF also has potent trophic effects on other CNS tissues [39,61]. Trok et al. [54] reported that GDNF has positive effects on postnatal spinal cord in intraocular allogeneic and syngeneic transplants [54]. In our study, many more MHC loci-positive APC and lymphocytes were found in

Fig. 6. Typicalappearanceof transplantsin micrographsafterNT3 treatmentin the Sprague-Dawleyrats (× 96). (a) Cresylviolet,(b) MHCI(OX-18), (c) MHCII (OX-6), (d) TcR (R-73), (e) CD1lb (OX-42), (f) ED1, (g) NF. There are marked immunecell infiltrates, similar to GDNF-treatedgroups. However, MHCII-positivecells in the graft are much less numerousthan with GDNF. In addition,the interstitialmicroglialproliferationis less than in GDNF-treatedgroups. Densityof neurofilament-positivefibers is much greaterthan with GDNF.

160

M. Shinoda et al. / Brain Research 722 (1996) 153-167

M. Shinoda et al. / Brain Research 722 (1996) 153-167

allogeneic spinal cord grafts to the anterior chamber of the eye of Sprague-Dawley rats after GDNF treatment. However, in the Fisher strain, few MHC loci-positive APC could be seen. This suggests that GDNF upregulates MHC loci in allogeneic but not syngeneic transplants. GDNF is a distant member of the transforming growth factor-/3 (TGF-fl) superfamily [29]. While there have been no specific studies of GDNF on immunological elements, TGF-/3 molecules are known for having various immunomodulatory effects. TGF-/3 is produced by a large number of different cell types, among them various cells of the immune system, including macrophages, peritoneal monocytes and neutrophils, T lymphocytes and B cells. In a similar fashion, many immune cells show specific responses after TGF-fl stimulation; TGF-fl inhibits the IL2-dependent T-cell proliferation of both CD4 + and CD8 + subsets and the activation of T lymphocytes. TGF-fl also suppresses differentiation and proliferation of natural killer cells, inhibits the IL-2-induced lymphokine-triggered killer cell activation and reduces ~theIL-2 dependent proliferation of B cells as well as the IL-2 or B-cell differentiation factor-dependent immunog]lobulin secretion by B cells. In addition, TGF-fl is able to switch B cells from IgG to IgA production. Furthermore, TGF-/3 inhibits the action of IL-1 by down-regulation o1! IL-1 receptor expression. It is the most potent chemoattractant known for monocytes as well as for human peripheral blood neutrophils. With regard to the immune system, the in vitro effects of TGF-/3 are diverse and primarily result in immunosuppression [59]. McNeill et al. reported that in cerebral ischemic models, exogenous TGF-/31 given soon after hypoxic-ischemic brain injury may have therapeutic potential via inhibition of the microglial reaction [36]. Lindholm et al. reported TGF-/31 stimulates expres,;ion of nerve growth factor in cultured rat astroglia [30]. Prehm and Krieglstein reported that TGF/31 acts as a neuroprotectant following short-term exposure to glutamate, but exacerbates the effects of chronic glutamate exposure [42]. On the other hand, Wahl et al. reported that TGF-/3 antagonizes IL-1, 2 and 3, although TGF-/3 also induced the secretion of IL-1 by purified macrophages [55,57,58]. Interestingly, in the CNS, activated microglial cells can secrete high levels of IL-1, compared with other CNS cells [15]. Activated microglia functionally act as macrophages and have the same origin as other macrophages [8,31,33,41,43]. In addition, TGF-/3 is also well known as an accelerator of angiogenesis and wound healing [57]. In our study, many MHCII and ED 1positive cells, and interstitial OX-42-positive elements, were noted near vascular components. These findings sug-

161

gest that GDNF may also be important for angiogenic functions in nervous tissues. Although the CNS has been described as an 'immunologically privileged site' [3,6,13,17,23], there are a number of immunologically related cells one can identify. Amongst the most important elements are microglia. Although the origin of microglial cells is somewhat controversial [24,27,43,48], many researchers assume that microglia come from embryonal circulatory macrophages [8,31,33,41]. There are three types of microglial cells in the CNS: ramified microglia are prominent in the mature CNS and are derived from ameboid microglia; ameboid microglia are present during late prenatal and early postnatal ages; reactive microglia appear primarily in cases of CNS injury (for reviews see [26,33,40,51]). These three microglial cells types have different properties which are manifest immunocytochemically, OX-42 recognizes cellular complement receptor type 3 (CR3) found in classical and ameboid a n d / o r reactive microglial subtypes [16,44,51]. ED1 is a cytoplasmic marker of macrophages [9], and only localizes to cells with monocytic lineage [10,51]. In addition, in cases of neuronal damage, microglial cells can express MHC loci [1,2]. In our study, the distribution of EDl-positive reactive microglia was well correlated with MHC loci- (especially class II) positive APC. We allowed grafts to remain in oculo for 6 weeks for two reasons. First, since grafts were made from E14-15 donors and there are some EDl-positive microglia normally found in CNS for up to 3 weeks after birth [8], this 6-week period of graft maturation should theoretically maximize exposure of the transplanted tissue to its own, graft-derived microglial cells. The second reason is that we have previously reported that longer term intracerebral grafts show more immune cells after peripheral alloimmunization [23]; hence, a longer in oculo maturation period might augment immunoreactivity. We used spinal cord transplants, as opposed to grafts of other CNS regions, in this study also for two reasons. First, spinal cord transplants in the anterior chamber of the eye show a larger number of immunoreactive cells at the iris-transplants interface [19,46] than brain tissue grafts in the same location. Second, the spinal cord also has a large proportion of white matter, which contains more ameboid microglial cells during development [24]. Ameboid a n d / o r activated microglial cells can express MHC loci and IL-1 [15,32], and thus may have more immunomodulatory potential. The origins of the immune cells seen here are still unclear. All lymphocytes are probably of host origin;

Fig. 7. Typicalappearance of trmlsplants in micrographs after BDNF treatment in Sprague-Dawleyrats (X 96). (a) Cresyl violet, (b) NF, (c) MHCI (OX-18), (d) MHCII (OX-6), (e)TcR (R-73), (f) CDIlb (OX-42), (g) ED1. There are few immune-relatedcomponentslocalized with each antibody. Extensive dendritic network (ramified microglia)can be seen with MHCI and OX-42 staining. Well definedneurofilamentimmunoreactivityis present.

162

M. Shinoda et al. / Brain Research 722 (1996) 153-167

Fig, 8. Typical appearance of transplants in micrographs after cytochrome c treatment in Sprague-Dawley rats ( × 96). (a) Cresyl violet, (b) NF, (c) MHCI (OX-18), (d) MHCII (OX-6), (e) TcR (R-73), (f) CD1 lb (OX-42), (g) ED1. Some dendritic network (ramified microglia) can be seen with MHCI and OX-42 stainings. Also a few immunologically related cells can be seen.

163

M. Shinoda et a l . / Brain Research 722 (1996) 153-167

m

i!i~

Fig. 9. Typical appearance of trmlsplants in micrographs after GDNF treatment in Fisher rats ( × 96). (a) Toluidine blue, (b) NF, (c) MHCI (OX-18), (d) MHCII (OX-6), (e) TcR (R-73), (f) C D l l b (OX-42), (g) ED1. Little immunological reactivity is demonstrated except for positive structures on the endothelial walls with MHCI staiaing.

164

M. Shinoda et al. / Brain Research 722 (1996) 153-167

i

M. Shinoda et al. / Brain Research 722 (1996) 153-167

165

i i! i ¸¸ !

, ":

,r

'i

~i

i~Ii~!¸ ~i~ii

Fig. 11. Upper panels: microglial morphologies.(a) Well-developed ramified microgliacan be seen after OX-42 staining in BDNF-treated Sprague-Dawley rats. X400. (b) Proliferation of reactive microglial cells are seen with ED1 staining in GDNF-treated Sprague-Dawley rats. ×400. Lower panels: ot-motoneuronscan be seen in those transplants without marked immunological reactions. (c) CGRP staining from same treatment group shown in Fig. 7. Well-developed a-motoneuron can be seen (arrow). × 100. (d) CGRP staining from cytochrome c-treated Sprague-Dawleyrats. Some a-motoneuronscan be seen (asterisks). × 200. (e) Higher magnificationof (d). × 400.

however, the source of the APCs is equivocal. In grafts with marked immunological reactions, almost all of the MHC class II-positive cells were localized close to vascular structures, but a few were not and resembled microglial or astroglial cells. In transplants with more modest immune reactions, many MHCI-positive ramified microglial cells elaborated fine dendritic networks. Moreover, amotoneurons could be more rapidly detected in these less affected grafts. We have already reported that in fetal mesencephalic transplants in situ, fewer numbers of tyrosine hydroxylase-positive cells could be observed in grafts showing a stronger immune response after peripheral alloimmunization [47]. Thi~ suggests that development of normal neuronal elements in grafts of CNS tissue is not compatible with marked immune responses. In our study, treatment with NT-3 also led to significant aggregations of immune cells and MHC loci-positive APCs.

Recently, Laurenzi et al. [28] reported that neurotrophin mRNAs are present in the immune system. NT-3 and NT-4 mRNA, as well as trkB and trkC transcripts, were found in thymus, thymic stroma (tissue depleted of mononuclear cells), spleen and splenic stroma. Concanavalin A-treated thymic and lipopolysaccharide-treated splenic mononuclear cells expressed a twofold increase in NT-4 but not NT-3 [28]. A role for NT-3 in immune system function is compatible with the findings reported here. It has long been recognized that the brain enjoys an immunological privilege with respect to the prolonged survival of transplanted allogeneic and some xenogeneic tissues, in comparison with the survival of similar tissue grafted to peripheral sites [3,17,37,60]. This does not mean, however, that immune reactions cannot take place in the brain, as evidenced by the fact that a variety of experimental immune-mediated diseases, and a number of human

Fig. 10. Typical appearance of transplants in micrographs after cytochrome c in Fisher rats (× 96). (a) Toluidine blue, (b) NF, (c) MHCI (OX-18), (d) MHCII (OX-6), (e) TcR (R-73), (f) CD1 lb (OX-42), (g) ED1. Some dendritic network (ramified microglia) can be seen with MHCI and OX-42 staining. No significant immunologicalreactions are exhibited in this treatment group.

166

M. Shinoda et al. / Brain Research 722 (1996) 153-167

i n f l a m m a t o r y disorders with a u t o i m m u n e characteristics (e.g., multiple sclerosis [35]) affect the CNS. Furthermore, allografts in the brain are usually p r o m p t l y rejected w h e n the a n i m a l is exposed to the same alloantigens in the periphery [12,23,34]. Thus, the privilege is relative, not absolute, and m a y simply represent quantitative differences b e t w e e n tissue grafted in the brain, compared to peripheral sites, with respect to its interaction with the i m m u n e system. In conclusion, G D N F and NT-3, but not B D N F , enhance i m m u n e responses in allogeneic spinal cord grafts, while m u c h smaller effects of G D N F are found in syngeneic grafts. Since G D N F did not elicit strong i m m u n e responses in the syngeneic grafts, the upregulations in the allogeneic transplants c a n n o t be explained by the graft surgery or post-surgical injury. Future studies are needed to determine if these trophic factors also upregulate imm u n e responses in the brain in situ after intracerebral administration.

macrophage subpopulations in the rat recognized by monoclonal antibodies ED-1, ED-2 and ED-3, Immunology, 54 (1985) 589-599. [10] Ellison, J.A. and De Vellis, J., Ameboid microglia expressing GD3 ganglioside are concentrated in regions of oligodendrogenesis during development of the rat corpus callosum, Glia, 14 (1995) 123-132. [11] Fontana, A., Constam, D.B., Frei, K., Malipiero, U. and Pfister, H.W., Modulation of the immune response by transforming growth factor beta, Int. Arch. Allergy Immunol., 99 (1992) 1-7. [12] Freed, W.J., Dymecki, J., Poltorak, M. and Rodgers, C.R., Intraventricular brain allografts and xenografts: studies of survival and rejection with and without systemic sensitization, Prog. Brain Res., 78 (1988) 233-241. [13] Fuchs, H.E. and Bullard, D.E., Immunology of transplantation in the central nervous system, Appl. Neurophysiol., 51 (1988) 278-296. [14] Giacobini, M.M.J., Hoffer, B. and Olson, L., Acidic and basic fibroblast growth factor can enhance growth and development of transplanted fetal brain tissue. Evidence from in oculo grafting, Exp. Brain Res., 86 (1991) 73-81. [15] Giulian, D., Baker, T.J., Shih, L.C. and Lachman, L.B., Interleukin 1 of the central nervous system is produced by ameboid microglia, J. Exp. Med., 164 (1986) 594-604. [16] Graeber, M.B., Streit, W.J. and Kreutzberg, G.W., Axotomy of the rat facial nerve leads to increased CR3 complement receptor expression by activated microglial cells, J. Neurosci. Res., 21 (1988)

18-24. Acknowledgements This project was supported b y the Swedish M R C and U S P H S . The authors thank S u s a n n e Almstrtim, Servet Eken, M o n i c a Casserltiv and Karin LundstriSmer for expert technical assistance, and Ida Engqvist for editorial support. G D N F was a generous gift from A m g e n Inc.

References [1] Akiyama, H., Itagaki, S. and McGeer, P.L., Major histocompatibility complex antigen expression on rat microglia following epidural kainic acid lesions, Z Neurosci. Res., 20 (1988) 147-157. [2] Akiyama, H. and McGeer, P.L., Microglial response to 6-hydroxydopamine-induced substantia nigra lesions, Brain Res., 489 (1989) 247-253. [3] Barker, C.F. and Billingham, R.E., lmmunologically privileged sites, Adv. Immunol., 25 (1977) 1-54. [4] Beck, K.D., Valverde, J., Alexl, T., Poulsen, K., Moffat, B., Vandlen, R.A., Rosenthal, A. and Hefti, F., Mesencephalic dopaminergic neurons protected by GDNF from axotomy-induced degeneration in the adults brain, Nature, 373 (1995) 339-341. [5] Bjtirklund, H., Olson, L., Seiger, A,. and Hoffer, B., Chronic lead and brain development: intraocular brain grafts as a method to reveal regional and temporal effects in the cerebral nervous system, Environ. Res., (1980) 224-236. [6] Brent, L., Immunologically privileged site. In B. Johansson, C. Owman and H. Widner (Eds.), Pathophysiology of the Blood-brain Barrier, Elsevier, 1990, pp. 172-177. [7] Cheng, H., Hoffer, B., Strtimberg, I., Russell, D. and Olson, L., The effect of glial cell line-derived neurotrophic factor in fibrin glue on developing dopamine neurons, Exp. Brain Res., 104 (1995) 199-206. [8] Chugani, D.C., Kedersha, N.L. and Rome, L.H., Vault immunofluorescence in the brain: new insights regarding the origin of microglia, J. Neurosci., 11 (199l) 256-268. [9] Dijkstra, C.D., D~ipp, E.A., Joling, P. and Kraal, G., The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct

[17] Head, J.R. and Billingham, R.E., Immunologically privileged sites in transplantation immunology and oncology, Perspect. BioL Med., 29 (1985) 115-131. [18] Henschen, A., Goldstein, M. and Olson, L., The innervation of intraocular spinal cord transplants by cografts of locus ceruleus and substantia nigra neurons, Brain Res., 433 (1987) 237-247. [19] Henschen, A., Hoffer, B. and Olson, L., Spinal cord grafts in oculo: survival, growth, histological organization and electrophysiological characteristics, Exp. Brain Res., 60 (1985) 38-60. [20] Henschen, A., Palmer, M.R. and Olson, L., Raphe dorsalis-spinal cord cografts in oculo: electrophysiological evidence for an excitatory serotonergic innervation of transplanted spinal neurons, Brain Res. Bull., 17 (1986) 801-808. [21] Henschen, A., Verhofstad, A.A. and Olson, L., Intraocular grafts of nucleus raphe dorsalis provide cografted spinal cord with a serotonergic innervation, Brain Res. Bull., 15 (1985) 335-342. [22] Hooper, W.C., The role of transforming growth factor-beta in hematopoiesis. A review, Leukoc. Res., 15 (1991) 179-184. [23] Hudson, J.L., Hoffman, A., Strtimberg, I., Hoffer, B.J. and Moorhead, J.W., Allogeneic grafts of fetal dopamine neurons: behavioral indices of immunological interactions, Neurosci. Lett., 171 (1994) 32-36. [24] Hutchins, K.D., Dickson, D.W., Rashbaum, W.K. and Lyman, W.D., Localization of microglia in the human fetal cervical spinal cord, Dev. Brain Res., 66 (1992) 270-273. [25] Inoue, H.K., Henschen, A. and Olson, L., Ultrastructure of spinal cord grafts with and without cografts of locus coeruleus in oculo, Exp. Neurol., 102 (1988) 109-120. [26] Jordan, F.L. and Thomas, W.E., Brain macrophages: question of origin and interrelationship, Brain Res. Rev., 13 (1988) 165-178. [27] Kitamura, T., Miyake, T. and Fujita, S., Genesis of resting microglia in the gray matter of mouse hippocampus, J. Comp. Neurol., 226 (1984) 421-433. [28] Laurenzi, M.A., Barbany, G., Timmusk, T., Lindgren, J.A. and Persson, H., Expression of mRNA encoding neurotrophins and neurotrophin receptors in rat thymus, spleen tissue and immunocompetent cells. Regulation of neurotrophin-4 mRNA expression by mitogens and leukotriene B4, Eur. J. Biochem., 223 (1994) 733-74l. [29] Lin, L.F., Doherty, D.H., Lile, J.D., Bektesh, S. and Collins, F., GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons, Science, 260 (1993) 1130-1132.

M. Shinoda et al. / Brain Research 722 (1996) 153-167 [30] Lindholm, D., Hengerer, B., Zafra, F. and Thoenen, H., Transforming growth factor-fl 1 stimulates expression of nerve growth factor in the rat CNS, Neuroreport, 1 (1990) 9-12. [31] Ling, E.A., Transformation of monocytes into amoeboid microglia in the corpus callosum of postnatal rats, as shown by labelling monocytes by carbon particles, J. Anat., 128 (1979) 847-858. [32] Ling, E.A., Kaur, C. and Wong, W.C., Expression of major histocompatibility complex and leukocyte common antigens in ameboid microglia in postnatal rats, J. Anat., 177 (1991) 117-126. [33] Ling, E.A. and Wong, W.C., The origin and nature of ramified and amoeboid microglia: a historical review and current concepts, Glia, 7 (1993) 9-18. [34] Mason, D.W., Charlton, H.M., Jones, A.J., Lavy, C.B., Puklavec, M. and Simmonds, S.J., The fate: of allogeneic and xenogeneic neuronal tissue transplanted into the third ventricle of rodents, Neuroscience, 19 (1986) 685-694. [35] McFarland, H.F., Immunology of multiple sclerosis, Ann. NYAcad. Sci., 540 (1988) 99-105. [36] McNeill, H., Williams, C., Guan, J., Dragunow, M., Lawlor, P., Sirimanne, E., Nikolics, K. and Gluckman, P., Neuronal rescue with transforming growth factor-beta 1 after hypoxic-ischemic brain injury, Neuroreport, 5 (1994) 901-904. [37] Nicholas, M.K. and Amason, B.G.W., Immunologic considerations in transplantation to the CNS. In F. Seil (Ed.),, Neural Regeneration and Transplantation, A.R. Liss, 1989, pp. 239-284. [38] Olson, L., Seiger, .~. and SIx/Smberg, I., Intraocular transplantation in rodents. A detailed accou:at of the procedure and examples of its use in neurobiology with special reference to brain tissue grafting. In S. Fedoroff (Ed.),, Advances in Cellular Neurobiology, Vol 4, Academic Press, 1983, pp. 407-442. [39] Oppenheim, R.W., Houenott, L.J., Johnson, J.E., Lin, L.-F.H., Li, L., Lo, A.C., Newsome, A.I_,., Prevette, D.M. and Wang, S., Developing motor neurons rescued from programmed and axotomy-induced cell death by GDNF, Nature, 373 (1995) 344-346. [40] Perry, V.H. and Gordon, &, Macrophage and microglia in the nervous system, Trends Neurosci., 11 (1988) 273-277. [41] Perry, V.H., Hume, D.A. and Gordon, S., Immunohistochemical localization of macrophages and microglia in the adult and developing mouse brain, Neuroscience, 15 (1985) 313-326. [42] Prehn, J.H.M. and Krieglstein, J., Opposing effects of transforming growth factor-ill on glutamate neurotoxicity, Neuroscience, 60 (1994) 7-10. [43] Rio-Hortega, P., Microglia. In Cytology and Cellular Pathology of the Nervous System, Paul B. Hoeber Inc., 1932, pp. 481-584. [44] Robinson, A.P., White, T.M and Mason, D.W., Macrophage heterogeneity in the rats as delineated by two monoclonal antibodies MRC Ox-41 and MRC Ox-42, the latter recognizing complement receptor type 3, Immunology, 57 (1986) 239-247. [45] Schmidt-Kastner, R., Tomac, A., Hoffer, B.J., Bektesh, S., Rosenzweig, B. and Olson, L., G]ial cell-line derived neurotrophic factor (GDNF) mRNA upregulation in striatum and cortical areas after pilocarpine-induced status epilepticus in rats, Mol. Brain Res., 26 (1994) 325-330. [46] Shinoda, M., Giacobini, I~[., Schmidt-Kastner, R., Trok, K. and

[47]

[48]

[49]

[50]

[51] [52]

[53]

[54]

[55] [56]

[57]

[58]

[59]

[60]

[61]

167

Olson, L., Differential immune responses to fetal intracameral spinal cord and cortex cerebri grafts, Exp. Brain Res., 110 (1996) in press. Shinoda, M., Hudson, J.L., StrSmberg, I., Hoffer, B.J., Moorhead, J.W. and Olson, L., Allogeneic grafts of fetal dopamine neurons: immunological reactions following active and adoptive immunizations, Brain Res., 680 (1995) 180-195. Sminia, T., De Groot, C.J.A., Dijkstra, C.D., Koetsier, J.C. and Polman, C.H., Macrophage in the central nervous system of the rat, Immunobiology, 174 (1987) 43-50. Stichel, C.C. and Muller, H.W., Extensive and long-lasting changes of glial ceils following transection of the postcommissural fornix in the adult rat, Glia, 10 (1994) 89-100. Str~Smberg, I., Bj~irklund, L., Johansson, M., Tomac, A., Collins, F., Olson, L., Hoffer, B. and Humpel, C., Glial cell line-derived neurotrophic factor is expressed in the developing but not adult striatum and stimulates developing dopamine neurons in vivo, Exp. Neurol., 124 (1993) 401-412. Thomas, W.E., Brain macrophages: evaluation of microglia and their functions, Brain Res. Rev., 17 (1992) 61-74. Tomac, A., Lindqvist, E., Lin, L.-F.H., Ogren, S.O., Young, D., Hoffer, B.J. and Olson, L., Protection and repair of the nigrostriatal dopaminergic system by GDNF in vivo, Nature, 373 (1995) 335339. Trok, K., Freund, R.K., Palmer, M.R. and Olson, L., Spinal cordskeletal cografls: trophic and functional interactions, Brain Res., 659 (1994) 138-146. Trok, K., Hoffer, B.J. and Olson, L., Glial cell line-derived neurotrophic factor enhances survival and growth of pre- and postnatal spinal cord transplants, Neuroscience, 71 (1996) 231-241. Wahl, S.M, The role of transforming growth factor-beta in inflammatory processes, Immunol. Res., 10 (1991) 249-254. Wahl, S.M., Costa, G.L., Mizel, D.E., Allen, J.B., Skaleric, U. and Mangan, D.F., Role of transforming growth factor beta in the pathophysiology of chronic inflammation, J. Periodontol., 64 (1993) 450-455. Wahl, S.M., Hunt, D.A., Wakefield, L.M., McCartney-Francis, N., Wahl, L.M., Roberts, A.B. and Spore, M.B., Transforming growth factor type /3 induces monocyte chemotaxis and growth factor production, Proc. Natl. Aead. Sci. USA, 84 (1987) 5788-5792. Wahl, S.M., Hunt, D.A., Wong, H.L., Dougherty, S., McCartneyFrancis, N., Wahl, L.M., Ellingsworth, L., Schmidt, J.A., Hall, G., Roberts, A.B. and Sporn, M.B., Transforming growth factor/3 is a potent immunosuppressive agent that inhibits IL-1 dependent lymphocyte proliferation, J. Immunol., 140 (1988) 3026-3032. Waltenberger, J., Miyazono, K., Funa, K., Wanders, A., Fellstrom, B. and Heldin, C.H., Transforming growth factor-beta and organ transplantation, Transplant. Proc., 25 (1993) 2038-2040. Widner, H. and Brundin, P., Immunological aspects of grafting in the mammalian central nervous system. A review and speculative synthesis, Brain Res., 472 (1988) 287-324. Yan, Q., Matheson, C. and Lopez, O.T., In vivo neurotrophic effects of GDNF on neonatal and adult facial motor neurons, Nature, 373 (1995) 341-344.