Search

Parkinson's disease is a neurodegenerative disorder where dopamine neurons in the substantia nigra of ventral mesencephalon undergo degeneration. In addition to the loss of dopamine neurons, noradrenaline neurons in the locus coeruleus degenerate, actually to a higher extent than the dopamine neurons. The interaction between these two nuclei is yet not fully known, hence this study was undertaken to investigate the role of locus coeruleus during development of dopamine neurons utilizing the intraocular grafting model. Fetal ventral mesencephalon and locus coeruleus were implanted either as single grafts or co-grafts, placed in direct contact or at a distance. The results revealed that the direct attachment of locus coeruleus to ventral mesencephalon enhanced graft volume and number of tyrosine hydroxylase (TH)-positive neurons in ventral mesencephalic grafts. Cell counts of subpopulations of TH-positive neurons also immunoreactive for aldehyde dehydrogenase 1-A1 (ALDH1) or calbindin, revealed improved survival of ALDH1/TH-positive neurons. However, the number of calbindin/TH-positive neurons was not affected. High density of dopamine-beta-hydroxylase (DBH)-positive innervation in the ventral mesencephalon placed adjacent to locus coeruleus was correlated to the improved survival. Ventral mesencephalic tissue, implanted at a distance to locus coeruleus, did not demonstrate improved survival, although DBH-positive nerve fibers were detected. In conclusion, the direct contact of locus coeruleus resulting in dense noradrenergic innervation of ventral mesencephalon is beneficial for the survival of ventral mesencephalic grafts. Thus, when trying to rescue dopamine neurons in Parkinson's disease, improving the noradrenergic input to the substantia nigra might be worth considering.

Neuroinflammation plays a critical role in loss of dopamine neurons during brain injury and in neurodegenerative diseases. Diets enriched in foods with antioxidant and anti-inflammatory actions may modulate this neuroinflammation. The model of 6-hydroxydopamine (6-OHDA) injected into the dorsal striatum of normal rats, causes a progressive loss of dopamine neurons in the ventral mesencephalon. In this study, we have investigated the inflammatory response following 6-OHDA injected into the striatum of adult rats treated with diet enriched in blueberry or spirulina. One week after the dopamine lesion, a similar size of dopamine degeneration was found in the striatum and in the globus pallidus in all lesioned animals. At 1 week, a significant increase in OX-6- (MHC class II) positive microglia was found in animals fed with blueberry- and spirulina-enriched diets in both the striatum and the globus pallidus. These OX-6-positive cells were located within the area of tyrosine hydroxylase (TH) -negativity. At 1 month after the lesion, the number of OX-6-positive cells was reduced in diet-treated animals while a significant increase beyond that observed at 1 week was now present in lesioned control animals. Dopamine recovery as revealed by TH-immunohistochemistry was significantly enhanced at 4 weeks postlesion in the striatum while in the globus pallidus the density of TH-positive nerve fibers was not different from control-fed lesioned animals. In conclusion, enhanced striatal dopamine recovery appeared in animals treated with diet enriched in antioxidants and anti-inflammatory phytochemicals and coincided with an early, transient increase in OX-6-positive microglia.

Transplantation of fetal ventral mesencephalic (VM) tissue shows great promise as an experimental therapy for patients with Parkinson's disease. However, cell survival in brain tissue grafts is poor, with survival rates of only 5-15%. We have utilized a combination of the caspase inhibitor bocaspartyl (OMe)-fluoromethylketone (BOC-ASP-CH2F) and glial cell line-derived neurotrophic factor (GDNF) to enhance survival of grafted dopamine neurons. The VM tissue was dissected from embryonic day 13-15 rat fetuses, incubated in different doses of BOC-ASP-CH2F and GDNF, and transplanted to the anterior chamber of the eye of adult rats. Growth of the tissue was assessed through the translucent cornea. Doses of 50 and 100 micromolar of the general caspase inhibitor appeared to have detrimental effects on mesencephalic tissue, while 20 micromolar had beneficial effects on overall transplant growth. A combination of the caspase inhibitor and GDNF appeared to have more prominent effects on cell survival as well as dopaminergic fiber density than either agent by itself. The transplants doubled in size when they were treated with a combination of BOC-ASP-CH2F and GDNF, and cell death markers were significantly reduced at both 48 h and 4-6 days postgrafting. This is, to our knowledge, the first combined approach using apoptotic blockers with trophic factors, and demonstrates a viable strategy for protection of developing neurons, since several different aspects of graft function may be addressed simultaneously., (Copyright 2001 Academic Press.)

Efforts have been made to counteract the symptoms of Parkinson's disease by substituting the loss of dopaminergic neurons with fetal ventral mesencephalic grafts. One of the postulated limiting factors in this treatment is the relatively poor cell survival and limited graft-derived fiber outgrowth. Recent results documenting enhanced survival of grafted dopaminergic neurons showed no positive correlation to enhanced innervation of the striatal target. Therefore this study was undertaken to investigate whether all surviving grafted dopaminergic neurons projected to the striatal target. Hence, fetal ventral mesencephalic tissue was implanted adjacent to mature versus immature striatal tissue using in oculo and intraventricular grafting techniques. In in oculo grafting, fetal ventral mesencephalon was implanted simultaneously with fetal lateral ganglionic eminence (immature striatal target) or to already matured striatal in oculo grafts (mature striatal target). Furthermore, fetal ventral mesencephalon was implanted into the lateral ventricle adjacent to mature dopamine-depleted striatum. The retrograde tracer fluorogold was injected into the striatal portion of the in oculo cografts and into reinnervated areas of the adult brain. Immunohistochemistry revealed that a significantly larger proportion of tyrosine hydroxylase-positive neurons in the ventral mesencephalic graft was innervating in oculo immature striatal tissue, and hence was fluorogold-positive, in comparison with the number of tyrosine hydroxylase-positive neurons innervating mature striatal tissue. Moreover, intracranial transplantations showed that tyrosine hydroxylase-positive neurons were distributed within the grafts in dense clusters of cells. In most clusters tyrosine hydroxylase-positive cells were fluorogold-negative but calbindin-positive. In a few tyrosine hydroxylase-positive cell clusters, neurons were coexpressing fluorogold but were calbindin-negative. In conclusion, significantly more dopamine neurons projected to immature than to mature striatal tissue and thus, a subpopulation of grafted dopaminergic neurons was not projecting into adult striatum. Thus, the results from this study show that further attempts to enhance survival of grafted dopamine neurons in purpose to enhance graft-derived fiber outgrowth and efficacy should also consider different subtypes of dopamine neurons., (Copyright 2001 Academic Press.)

Chromaffin cells have been recognized for their ability to transform into sympathetic ganglion-like cells in response to nerve growth factor (NGF) or to stimulation of other neurotrophic factors. Transforming growth factor beta (TGFbeta) family members have been shown to potentiate the effect of different trophic factors. The aim of this study was to investigate if TGFbeta may influence NGF-induced neuronal transformation and regulation of NGF, TGFbeta1, and their receptors in the adult rat chromaffin tissue after grafting. Intraocular transplantation of adult chromaffin tissue was employed and grafts were treated with TGFbeta1 and/or NGF. Graft survival time was 18 days after which the grafts were processed for TGFbeta luciferase detection assay, NGF enzyme immunoassay, or in situ hybridization. In grafts stimulated with NGF, increased levels of TGFbeta1 and TGFbeta1 mRNA were detected. When grafts instead were treated with TGFbeta1, enhanced levels of NGF protein were found. Furthermore, a positive mRNA signal corresponding to the transforming growth factor II receptor (TbetaRII) was found in the chromaffin cells of the normal adrenal medulla as well as after grafting. No increase of TbetaRII mRNA levels was detected after transplantation or after TGFbeta1 treatment. Instead a reduction of TbetaRII mRNA expression was noted after NGF treatment. NGF stimulation of grafts increased the message for NGF receptors p75 and trkA in the chromaffin transplants. Grafts processed for evaluations of neurite outgrowth were allowed to survive for 28 days and were injected weekly with NGF and/or TGFbeta1. NGF treatment resulted in a robust innervation of the host irides. TGFbeta1 had no additive effect on nerve fiber formation when combined with NGF. Combined treatment of NGF and anti-TGFbeta1 resulted in a significantly larger area of reinnervation. In conclusion, it was found that NGF and TGFbeta1 may regulate the expression of each other's protein in adult chromaffin grafts. Furthermore, TbetaRII mRNA was present in the adult rat chromaffin cells and became downregulated as a result of NGF stimulation. Although no synergistic effects of TGFbeta1 were found on NGF-induced neurite outgrowth, it was found that TGFbeta1 and NGF signaling are closely linked in the chromaffin cells of the adrenal medulla., (Copyright 2000 Academic Press.)

One of the drawbacks with fetal ventral mesencephalic (VM) grafts in Parkinson's disease is the limited outgrowth into the host striatum. In order to enhance graft outgrowth, epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) were administered by implantation of bioactive rods to the lateral part of the striatum to support grafted fetal VM implanted to the medial portion of the striatum. The polymer-based bioactive rods allow for a local secretion of neurotrophic factors over a time period of approximately 2 weeks. Moreover, glial cell line-derived neurotrophic factor (GDNF) and transforming growth factor-beta1 (TGFbeta1) were administered using the same technique. Concomitant administration of GDNF and TGFbeta1 was achieved by insertion of one GDNF and one TGFbeta1 rod. This was performed to investigate possible additive effects between GDNF and TGFbeta1. Rotational behavior, outgrowth from and nerve fiber density within the VM graft, and the number of TH-positive cells were studied. Functional compensation by reduction of rotational behavior was significantly enhanced in animals carrying bFGF and GDNF rods in comparison with animals carrying only VM graft. EGF and bFGF significantly increased the innervation density. Moreover, the nerve fiber density within the grafts was significantly enhanced by bFGF. Cell counts showed that a significantly higher number of TH-positive neurons was found in grafts treated with bFGF than that found in GDNF-treated grafts. An additive effect of TGFbeta1 and GDNF was not detectable. These results suggest that bioactive rods is a useful tool to deliver neurotrophic factors into the brain, and since bFGF was a potent factor concerning both functional, immunohistochemical and cell survival results, it might be of interest to use bFGF-secreting rods for enhancing the overall outcome of VM grafts into patients suffering from Parkinson's disease., (Copyright 2000 Academic Press.)

Grafting of ventral mesencephalic tissue in Parkinson's disease results in a partial dopaminergic reinnervation of host brain and dopamine agonist-induced rotational behavior is not completely reversed. To study a possible malfunction of the grafts, extracellular recordings with local applications of quinpirole were utilized and the neurophysiological results showed that a normalization of the upregulated dopamine D2 receptor supersensitivity occurred in reinnervated areas of the host striatum as well as in noninnervated areas remote from the graft innervation. Furthermore, the inhibitory effects on striatal nerve cell firing rate by the D1 receptor agonist SKF 81297 were not different in noninnervated or reinnervated areas of the striatum compared to the control side as seen from the dose-response curves. However, spontaneous striatal neuronal firing was significantly upregulated in noninnervated areas, while it was normalized in areas reached by graft-derived nerve fibers. Dual-probe microdialysis studying potassium-evoked glutamate release revealed that there was no difference in extracellular glutamate levels measured within or lateral to graft dopamine reinnervation. Thus, the upregulated spontaneous activity was not due to a difference in extracellular glutamate levels. The remaining rotational behavior seen after grafting was studied and recordings were performed in the striatum following systemic injection of the D1/D2 agonist apomorphine. The results revealed that apomorphine at the dose used to elicit turning behavior (0.05 mg/kg) still affected striatal neurons in noninnervated areas, while no effect was detected in reinnervated areas and in the intact side. However, a lower dose of apomorphine (0.005 mg/kg) showed no effects on striatal firing in graft reinnervated striata but only after dopamine depletion. In conclusion, the D2 supersensitivity is downregulated in graft-reinnervated striatum as well as in striatal areas lateral to the reinnervation when using selective D2 agonists, but the downregulation is not completely normalized when studying combined effects of D1/D2 agonists. Furthermore, the striatal neurons were firing significantly faster in noninnervated areas compared to reinnervated areas of graft-reinnervated striatum, which was most likely not due to changes in the glutamatergic input., (Copyright 2000 Academic Press.)

Grafting of fetal ventral mesencephalon in Parkinson's disease has been extensively studied. A crucial draw back of this technique is the low survival rate of the dopaminergic neurons. It has been documented that only 5-20% of the grafted neurons survive, and to enhance graft efficacy to a satisfying level, increased cell survival is of utmost desire. In this study we have used the antioxidant tiriliazad mesylate (U-74006F) to study the effect on the survival of dopaminergic neurons after grafting. The in oculo grafting model was used and ventral mesencephalon was dissected from E14-E15 rat fetuses in Hanks' balanced salt solution (HBSS), in Dulbecco's modified Eagle medium (DMEM), or in 0.3, 3.0, or 30 microM U-74006F diluted in DMEM. The tissue was then inserted into the anterior chamber of the eye. Some of the transplants were further treated with intraocular injections of 3 or 30 microM U-74006F (5 microliters) weekly for 2 weeks. Quantification of tyrosine hydroxylase (TH)-immunoreactive profiles revealed that in transplants treated with U-74006F at dissection only, no change in the number of TH-positive neurons was found. Pretreatment of 0.3 microM U-74006F during dissection combined with intraocular injections of U-74006F after grafting, on the other hand, resulted in a dose-dependent enhancement of survival of TH-positive neurons. Dissection in, and intraocular treatment with, 3 microM U-74006F resulted in a significantly enhanced survival of TH-positive neurons whereas using U-74006F at a concentration of 30 microM did not change the cell survival compared to solely DMEM-treated grafts. Thus, 30 microM was interpreted to be an overdose. Comparing cell survival when dissected in DMEM with that dissected in HBSS showed that DMEM was clearly superior. Nerve fiber formation was most pronounced in grafts treated with 3 microM U-74006F. In conclusion, survival of TH-positive neurons is enhanced by U-74006F, which is readily available for clinical use and thus could be employed to enhance graft survival when transplanting patients suffering from Parkinson's disease.

Microglia express cytokines, major histocompatibility (MHC) loci, and several other immunologically important constituents. The aim of this study was to detect immunological responses of microglial cells following allogeneic dopaminergic transplantation using active and adoptive immunizations. Adult inbred Fisher 344 (F344 RT1) rats were unilaterally dopamine (DA) depleted in striatum by injection of 6-hydroxydopamine. The degree of degeneration was assessed by recording the rotational response to apomorphine. Fetal ventral mesencephalic tissue containing DA neuroblasts from Wistar-Furth (WF, RT1u) rat donors (9-12 mm CRL) were later implanted in striatum on the lesioned side. Lymph nodes and spleen cells were collected aseptically, resuspended, and diluted for isovolumetric injections. Animals selected for active immunization were injected intraperitoneally with varying amounts of WF lymphocytes. Animals selected for adoptive immunization (transferred immunity) were intraperitoneally injected with 10(8) F344 lymphocytes prepared from animals actively immunized 3 weeks previously. Monoclonal antibodies against CD4 (OX38), CD8 (OX8), CD11b (OX42), MHC class I (OX18), monomorphic MHC class II (OX-6), and ED1 and polyclonal antibodies against tyrosine hydroxylase (TH) were used for immunohistochemistry. We found that the degree of ED1-positive cell proliferation was well correlated to the immunization patterns. Groups that were actively immunized with or without prior adoptive immunization had a larger amount of reactive microglial proliferation. ED1 immunohistochemistry revealed patterns of immunolabeling of engrafted areas: 8-12 weeks after grafting in nonimmunized and adoptively immunized groups reactive microglial proliferation occurred only at the graft periphery. Active and adoptive + active immunization led to ED1-IR within the grafts themselves. At early stages nonimmunized groups had an ED1 pattern which was partially inside the grafts. At early time points nonimmunized groups contained ameboid microglial cells within the grafts which disappeared at later stages and were absent in the immunized groups. ED1-positive ameboid microglial cells within the grafts may be of graft origin and constitute a part of a continued normal development of the fetal tissue.

The search for trophic factors that can support injured dopaminergic neurons and can enhance dopaminergic graft survival and outgrowth for therapeutic uses in Parkinson's disease has lately focused on members of the transforming growth factor (TGF) beta super-family. In this paper we have studied the effects of a member of the TGB beta family, glial cell line-derived neurotrophic factor (GDNF), on immature and mature ventral mesencephalic tissue grafted to the anterior chamber of the eye. The results confirm that GDNF increases survival of TH-positive neurons and enhances TH-immunoreactive nerve fiber formation when the grafts are treated during their development. The distribution of nerve terminals is densest within the area of TH-immunoreactive neurons and at the surface of the grafts. However, there is no change in the number of calcium-binding protein (CaBP)-positive neurons, suggesting that the subpopulation of TH-positive neurons that is increased are the CaBP-negative neurons of the ventral tier of pars compacta. Terminals from those neurons form the striatal patches during normal development. When the grafts are treated with GDNF after maturation, no change in TH-positive cell survival is seen but an increase of nerve terminals is still found within the cell dense area of the graft. Potassium-evoked dopamine release, measured using in vivo chronoamperometry, revealed significantly increased extracellular overflow in transplants treated with GDNF during development. The dopamine uptake blocker nomifensine significantly increased the time for clearance of the released dopamine. These data suggest that GDNF treatment of immature grafts enhances survival of TH-positive neurons, which would have innervated the striatal patches, and also increases TH-immunoreactive nerve fiber formation and dopamine release. Furthermore, GDNF treatment of mature grafts also increases dopamine fiber formation within the TH-positive neuronal area, indicating that adult dopaminergic neurons are also responsive to this agent.

It has previously been shown that the chromaffin cells of the adrenal medulla respond to nerve growth factor (NGF) with neurite outgrowth and increased cell survival in tissue culture or after grafting. In the present study we evaluated the dose dependency in neurite outgrowth from chromaffin tissue to recombinant human NGF (rhNGF). Therefore, pieces of adrenal medullary tissue from adult rat were grafted into the anterior chamber of the eye of previously sympathectomized recepients. Survival time was 4 weeks. At grafting and at Days 7, 14, and 21 postgrafting, the eyes were injected with 5 microliters of rhNGF at concentrations of 10, 30, 60, 100, 150, and 200 micrograms/ml, or with a control solution. All grafts, including the controls, survived well and became vascularized. At the low doses of rhNGF, 10 and 30 micrograms/ml, a small area of the irides was reinnervated and the density of the nerve fiber network was low. The maximal response was obtained at 100 micrograms rhNGF/ml. Using larger concentrations of 150 and 200 micrograms rhNGF/ml, the density of the nerve fiber network did not change, but the reinnervated area of the irides was significantly decreased compared to the outgrowth seen in irides treated with 100 micrograms/ml. In conclusion, adult rat chromaffin tissue responds to rhNGF in a dose-dependent manner. However, at the highest doses used, the outgrowth area was suboptimal, although nerve fiber density was maximal. These results indicate that to obtain maximal effects, the dose of NGF is critical.

The potential role of glial cell line-derived neurotrophic factor (GDNF) as a trophic molecule for midbrain dopamine neurons was examined using two different approaches: in situ hybridization and intraocular transplantation. The presence of mRNA for GDNF was noted in striatal and ventral limbic dopaminergic target areas in the developing (E20-P7) rat, but not the adult rat. Signals were also found in nondopaminergic areas during maturation, such as the cerebellar anlage, spinal cord, and thalamus. Lesions of the nigrostriatal pathway in neonatal or adult rats, using 6-hydroxydopamine injected into the medial forebrain bundle, did not elicit upregulation of mRNA for GDNF. Grafts of fetal ventral mesencephalon in the anterior eye chamber were exposed to repeated injections of GDNF, which elicited a marked and dose-dependent increase in transplant volume. A low (0.1 microgram/eye) and high (1 microgram/eye) dose of GDNF both led to a somewhat larger mean area of dopamine fiber outgrowth into host irides. In the transplants, cell counts of tyrosine hydroxylase (TH)-immunoreactive neurons revealed a doubling of cell numbers in the low-dose group and about four times as many cells in the high-GDNF-dose group compared to controls. Moreover, the density of TH-immunoreactive nerve fibers was markedly and significantly higher in transplants treated with the high GDNF dose. Since the volumes of these transplants were also larger, the total amount of both TH-positive cells and TH-positive nerve fibers was many-fold greater in the high-GDNF group than that in the controls. Taken together, these data support the concept that GDNF functions as a dopaminotrophic factor in vivo.

Two patients with advanced Parkinson's disease were followed for 6 months before, and 18 months after, receiving stereotaxic grafts of fetal mesencephalic tissue from aborted human fetuses. Parameters studied included a series of standardized tests of movement, response to levodopa, electrophysiological recording of the motor readiness potential, and positron emission tomography (PET) with ligands based upon levodopa and upon the dopamine reuptake inhibitor nomifensine. The patients each received stereotaxic implantation of ventral mesencephalic tissue containing midbrain dopamine neurons from aborted human fetuses of 8 to 10 weeks gestational age into the caudate and putamen of one hemisphere. Throughout their 18-month course, the patients were treated with cyclosporine, azathioprine, and glucocorticoids to minimize the risk of graft rejection. There were no significant complications from the procedure, but there was also no major change in their assessment of impairment on the Hoehn and Yahr scale. However, significant changes were observed in clinical, electrophysiological, and PET measures. Changes in these parameters, apparent at 6 months postoperatively, were described in detail in a previous report. The purpose of this present report is to provide follow-up data from the subsequent year with an emphasis on longitudinal evaluation methodology. Standardized clinical testing showed a small but long-term improvement in the first of the two patients. Following the operation, she was able to walk in "off" periods, which she had not been able to do preoperatively. This improvement was accompanied by increased walking speed and reduction in the time necessary to perform a series of pronation and supination movements using both hands. Although these improvements have continued throughout the postoperative period, they have not alleviated her basic neurological impairment. The second patient showed similar improvement during the first 6 months; she then reverted to her preoperative status at the end of the 18-month follow-up period. The electrophysiological recordings were consistent with the clinical findings. Both patients had significant changes in the motor readiness (bereitschafts) potential amplitude, which was greatest 5 to 7 months postoperation. The amplitude of the potential declined subsequently for both patients, but remained significantly elevated over the preoperative baseline for patient 1. The analysis of the PET scans was somewhat compromised by technical problems in the preoperative scans. However, they are also consistent with the clinical data. In comparisons of the operated and the unoperated sides, fluoro-dopa showed increased uptake in the caudate nucleus of patient 1 at 6 months and at 13 months.(ABSTRACT TRUNCATED AT 400 WORDS)

Peripheral nerve "bridges" demonstrate the ability to facilitate axonal growth and regenerate adult and fetal central nervous system tissue. The purpose of this study was to determine if co-grafted peripheral nerve tissue could enhance the ability of fetal dopamine (DA) cell transplants to reinnervate host striatum that had been denervated unilaterally. Male Fisher-344 rats were unilaterally lesioned with 6-hydroxydopamine to eliminate the nigrostriatal DA pathway. A total of 31 rats demonstrated a pattern of rotation indicative of a greater than 98% depletion in DA. Rats were kept as nongrafted controls (n = 6), grafted with sciatic nerve (PN) minces (n = 6), grafted with fetal ventral mesencephalon (VM; n = 10), or co-grafted with VM and PN minces (n = 9). All groups were then tested for changes in apomorphine-induced rotational behavior. The PN control group showed no significant differences in rotation when compared to pregrafting levels and to the lesioned nongrafted group. Both the VM-grafted group and the VM-PN co-grafted group showed significant (P less than 0.01, one-way ANOVA) decreases in rotations beginning at 1.5 weeks postgrafting. There was a progressive decrease in rotations up to 12 weeks, the last test point examined. Interestingly, the co-graft group revealed a significantly greater decrease in rotation (P less than 0.05) than the VM group beginning at 5 weeks and continuing out to the 12-week test point. Histological and immunocytochemical studies showed good survival of both PN and VM grafts. The augmented recovery could not be accounted for by increased DA cell survival or host brain DA reinnervation in the co-graft group. Taken together, these findings suggest that PN tissue enhances the ability of fetal VM grafts to reinnervate host brain.

Solid pieces of human fetal mesencephalic tissue were grafted to the lateral ventricle adjacent to dopamine-depleted striata of rats immunosuppressed with cyclosporin A. Apomorphine-induced rotations were performed before and at monthly intervals after grafting. Reductions in rotations were seen at 2 months post-grafting and these reductions progressively increased. Spontaneously active dopaminergic cells were found within the grafts using extracellular single-unit recording techniques. Recordings of striatal cells ipsilateral to the graft revealed "normal" firing rates compared to those of neurons in the control striatum. In response to the local application of the dopamine antagonist cis-flupenthixol, both the dopaminergic and striatal neurons showed dose-dependent excitations. Potassium-evoked releases of electroactive species ipsilateral to the fetal human graft, measured using high-speed in vivo electrochemistry, revealed response amplitudes that were similar to control striatum when an electrode was placed adjacent to the graft; distal to the graft the responses showed smaller amplitudes but prolonged time courses. Much greater levels of dopamine and serotonin were detected in the grafts, compared to in normal rat substantia nigra, as measured with HPLC coupled to a 16-channel electrochemical array detector. Immunocytochemical studies using antibodies against tyrosine hydroxylase (TH), revealed not only TH-positive cells within the graft, but also a few positive neurons that migrated into the host striatum. Numerous TH-immunoreactive fibers penetrated into striatum and reinnervated its total volume. Taken together, these data suggest that intraventricular graft placement may be a highly efficacious technique for studying fetal brain tissues in terms of maturation, reinnervation, and function.

Human fetal hippocampal tissue from normal women was obtained following elective abortion in the 8th to the 11th week of gestation. The hippocampal tissue was transplanted to the anterior chamber of the eye of adult athymic nude rats, where it was allowed to develop for up to 9 months before histological and electrophysiological evaluation. The transplants were revascularized from the host iris and many grew extensively in oculo. Large neurons were present in all transplants. Immunohistochemical studies revealed glutamic acid decarboxylase-containing terminals and clusters of gamma-aminobutyric acid-positive nerve cell bodies within the transplants, as well as scattered tyrosine hydroxylase-positive and acetylcholinesterase-containing fibers. Single neurons recorded extracellularly from transplants 4-9 months in oculo showed a slow spontaneous discharge, with both complex and single action potentials. Stimulation of the transplant surface evoked a small initial wave followed by a larger and longer-lasting field potential, similar to that seen in hippocampus in situ. A conditioning-testing paradigm was used to evaluate the presence of inhibitory circuitry in the hippocampal transplants. Significant suppression of the evoked test response was seen with interstimulus intervals ranging from 20 to 500 ms. Superfusion of enkephalin (100-300 nM) or penicillin (1600 U/ml) increased slow-wave activity, as did tetanic electrical stimulation. These treatments appeared to generate ictal-like activity, which in some cases persisted as interictal spikes. Illumination of the retina also increased neuronal activity, presumably by reflex activation of cholinergic afferents from the parasympathetic innervation of the iris. Taken together, our data suggest that fragments of hippocampus from aborted first trimester human fetuses, grafted to the eye chamber of rodent hosts, develop many organotypic histological and physiological features. This preparation may provide a unique means for the study of neurobiological properties of human brain in both normal and disease states.