- Email: [email protected]
Basic fibroblast growth factor promotes the survival of embryonic ventral mesencephalic dopaminergic neurons—II. Effects on nigral transplantsin vivo
NeuroscienceVol. 56, No. 2, pp. 389-398, 1993 Printed in Great Britain
0306-4522/93$6.00 + 0.00 Pergamon Press Ltd IBRO
BASIC FIBROBLAST GROWTH FACTOR PROMOTES THE SURVIVAL OF EMBRYONIC VENTRAL MESENCEPHALIC DOPAMINERGIC NEURONS-II. EFFECTS ON NIGRAL TRANSPLANTS IN I/l’lrO E. MAYER,*? J. W. FAWCETT*$ and S. B. D~*~§ *MRC
Cambridge Centre for Brain Repair, and the Departments of tExperimenta1 Psychology and $Physioiogy, University of Cambridge, Downing Street, Cambridge, U.K.
At&raft--The clinical potential of transplants of fetal dopaminergic neurons is limited by the fact that the percentage of cells surviving in such grafts is in general quite low. This report investigates the use of basic fibroblast growth factor administration (given either as a pretreatment or by repeated intrastriatal infusions) to promote the survival and behavioural efficacy of embryonic dopamine-rich nigral transplants in rats. pretreatment of the graft tissue by brief incubation with basic fibroblast growth factor increased the survival of tyrosine hydroxyl~-immunoreactive (presumed do~~ner~c) neurons in the grafts in comparison to control grafts, and accelerated the recovery in the transplants animals in tests of drug-induced rotational asymmetry. However, the clear advantage seen in the rotation test conducted three weeks after transplantation had disappeared by nine weeks. The moderate effects of pretreatment were markedly enhanced by repeated intrastriatal infusion of basic fibroblast growth factor into the host animals over 20 days following transplantation. This resulted in > 100% increase in the number of dopaminergic neurons surviving in the grafts, and was accompanied by a significantly greater recovery of the rats’ rotational asymmetries which persisted over the full nine weeks of testing. However, the repeated intracerebral infusions induced an inflammatory reaction in the striatum, and the associated trauma both complicates the interpretation of the mechanism of observed recovery and compromises the utility of this route of basic fibroblast growth factor administration for promoting graft survivai.
One of the major
problems that is emerging from recent clinical trials of embryonic neural transplantation in Parkinson’s disease is the unreliabile sur-
vival of grafted dopaminergic neuronsi6*” Similarly, when dopamine neurons are grafted as a dissociated cell suspension in rats, the yield of surviving cells is frequently much lower (l-5%) than is typically found in many other types of graft.) It is increasingly apparent that the survival and growth of grafted neurons is dependent upon the availability of a variety of growth factors “,m~23 which raises the possibility that graft survival might be promoted by exogenous t~a~ent with selected growth factors. Several studies indicate that basic fibroblast growth factor (bFGFr can promote the survival of a variety of embryonic neurons, among them ventral mesencephalic dopaminergic neurons.Lo,isJ8J9 Recent evidence suggests that at least part of this effect in culture is mediated by astrocytes.9~‘8 When administered to adult dopaminergic ventral mesencephalic $To whom correspondence should be addressed at: Department of Experimental Psychology, University of Cambridge, Dow&g Street, CarnbridgeCBZ 3EB, U.K. Abbreuiations: bFGF, basic fibroblast growth factor: EDTA, ethyldiaminetetraacetate; MPT?, l-methyl-& phenyl-tetrahydropyridine; PB, phosphate buffer; PBS, phosphatebuffered saline; TH, tyrosine hydroxylase; &OHDA, dhydroxydopamine.
neurons that have been treated with l-methyl4 phenyl-tetrahydropyridine (MPTP), bFGF is able to rescue them from atrophy or death.24 Moreover, repeated infusions of bFGF have been shown to exert a potent neurotrophic effect on the growth of a variety of brain areas transplanted into the anterior chamber of the eye. I2 In the context of promoting the survival of dopaminergic grafts, Steinbusch and coworker@’ have found that bFGF pretreatment increases the survival and growth of dopaminergic neurons within transplants. However, they have cautioned that this effect may be attributabie to heparin which was co-administered, in their experiments, in order to stabilize bFGF. However, other observations suggest that whereas the effects of acidic FGF are strongly dependent on the presence of heparin, the efficacy of bFGF may be potentiated only slightly by it.’ Thus, bFGF may prove a useful tool for promoting the viability of dopamine neurons in embryonic transplants. In the a~mpanying paper’” we show that bFGF can indeed promote the survival of dopamine neurons over at least four weeks in vitro. Unilateral lesions of the nigrostriatal pathway deplete forebrain dopamine and have provided the most widely used animal model of Parkinsonism within which to study the viability of nigral grafts6 Therefore, in the present study, we have used this lesion
390
1:. lh,AYtK
and graft model to investigate the effect of administration of bFGF (either as a pretreatment or by repeated central injections into the host striatum) on the survival and functional efficacy of dopamine-rich embryonic nigral tissues transplants in ho. EXPERIMENTAL
PROCEDURES
Subjects
Young adult female rats of the Sprague-Dawley strain (Harlan-Olac, Bicester, Oxon, U.K.) were used as transplant recipients, weighing approximately 200 g at the time of first surgery. They were housed under standard lightdark conditions, with food and water supplied ad libitum. Transplant tissues were dissected from embryos (crown-rump length _ 10mm) collected by caesarean section from pregnant rats of the same outbred strain. Lesions
A total of 108 rats received lesions, 48 in Experiment I and 60 in Experiment II. Rats were anaesthetized with 3 ml/kg equithesin anaesthesia and mounted in a Kopf stereotaxic frame. The toxin 6-hydroxydopamine hydrobromine (dOHDA; Sigma) was dissolved to a concentration of 3 mg/ml (free base weight) in 0.02% ascorbate saline. Four microlitres of the toxin was infused at a rate of 1 pl/min via a 30-gauge stainless steel cannula which was stereotaxically implanted into the right ascending nigrostriatal bundle at coordinates: A = -4.4 mm (caudal to bregma), L = 1.Omm (lateral to the midline), and V = 7.8 mm (ventral to the dura mater). with the nose bar set -2.3 mm below the interaural line. A further 4 min was allowed for diffusion before withdrawing the cannula. The wound was then cleaned and sutured. No futher postoperative care was required. Rotation tests
The efficacy of lesions was checked using the amphetamine-induced rotation model of Ungerstedt2* Animals were tested in automated rotometer bowls, modelled after the design of Ungerstedt and Arbuthnott,29 and connected to a microcomputer for data collection. Approximately 15 days following the lesion surgery, all animals were injected with 2.5 mg/kg methamphetamine hydrochloride (Sigma) and full 360” turns in either direction were recorded over a 90-min test-period. Only animals making >400 net ipsilatera1 turns over 90min were considered to have effective lesions for inclusion in the subsequent experiments. Animals were allocated to counterbalanced groups on the basis of their rotation scores. Further 90-min rotation tests were carried out using the same dose of amphetamine at three, six and nine weeks after transplantation. Neural transplantation
Cell suspensions for transplantations were prepared from fetal ventral mesencephalic tissue according to the same protocol used in the parallel in vitro experiments.‘* Briefly, donor embryos were removed from pregnant female rats, under terminal barbiturate anaesthesia. The ventral mesencephalon, containing the dopamine cells of the developing substantia nigra, was dissected in Leibowitz L-15 medium (Imperial Laboratories), the meninges removed, and the pieces incubated in 0.1% trypsin (type V, Sigma) with 0.02% EDTA (Imperial Laboratories) for 6 min at 37°C. A solution of 0.04% deoxyribonuclease (Sigma) at 37°C was added, the pieces shaken gently and centrifuged for 1 min at 100 g. The supernatant was then removed and a triturating solution was added [3 mg of bovine serum albumin, 10 mg deoxyribonuclease, 0.5 mg soybean trypsin inhibitor and 0.02% EDTA per ml of phosphate-buffered saline (PBS)], and the cells triturated gently through a 23-g needle
<‘I id
mounted on a l-ml syringe. The resulting cell suspension was then centrifuged at 1OOg for 4 min and the triturating solution removed. The cells were then resuspended to ;I concentration of one embryonic piece per 2 ~1 of medium The graft tissue was implanted by stereotaxic injection using a lo-p1 glass microsyringe (Scientific Glass Engineering, Ringwood, Australia). A 3-pl volume of ihc cell suspension was injected directly into the right (lesioned) striatum of the host rats at a rate of 1pl/min, with a further 5 min allowed before slowly withdrawing the syringe needle. Since it was necessary to make more than one suspension to graft all of the animals involved in one experiment. each suspension was used to graft equal numbers of animals in vehicle- and bFGF-treated groups. In Experiment I, each cell suspension was divided into two equal parts, with 10 ng/5 ~1 of bFGF added to one aliquot and a similar volume of vehicle to the other. A single graft deposit cell suspension was injected into the right striatum at the following coordinates: A = 1.0 mm. L = 3.0mm and V = 5.0mm, with the nosebar set at -2.3 mm. In total, 18 rats received bFGF pre-treated grafts, 18 received control grafts, and 12 remained with lesions alone. In Experiment II, two graft deposits were placed at the following coordinates: A = 1.6 mm, L = 2 mm, V = 4.5 mm and A = -0.4 mm, L = 4 mm, V = 5 mm, with the nosebar set at -2.3 mm. bFGF or vehicle was infused into the host striatum in proximity to the grafts via a subsequently implanted cannula (see below). In total, 24 rats received bFGF treated grafts, 24 received control grafts. and 12 remained with lesions alone. Cannulation and infusion
Basic FGF does not cross the blood-brain barrier, and so must be administered centrally. Since the relatively short biological half-life of bFGF did not permit its continuous infusion from a reservoir in a minipump, as has been possible with nerve growth factor,‘2.” repeated injections were made via implanted cannulae in order to deliver exogenous bFGF to the transplants. An 8-mm-long 23gauge stainless steel guide cannula was stereotaxicahy implanted into the striatum halfway between the two graft deposits (A = 0.6 mm, L = 3 mm. V = 3 mm) and attached to the skull with stainless steel screws and dental cement (see Fig. 1). The cannula was maintained with an g-mm stylet in order to avoid blockages. To make the intracerebral injections, a IO-mm 30-gauge stainless steel infusion cannula was connected via thin polyethylene tubing to a IO-PI glass microsyringe mounted in a Harvard microinfusion pump. The stylet was removed from the guide cannula in the skull, the infusion cannula was inserted, and 20 ng human recombinant bFGF (Boehringer, Mannheim) was infused over 2 min in 1~1 of vehicle (0.1% bovine serum albumin in 0.1 M PBS), followed by a 3-min diffusion time. Control grafts received similar infusions of 1 ~1 vehicle alone. A total of 10 repeated infusions were made in each animal, once every two days over 20 days, commencing on the second day after transplantation. Histological analysis
Animals were killed either three weeks (Experiment I, n = 16, Experiment II, n = 28) or nine weeks (Experiment I, n = 32, Experiment II, n = 32) after implantation in a balanced manner between groups. Rats were perfused transcardially under terminal barbiturate anaesthesia with 200 ml of ice-cold 0.1 M PBS containing 0.02% sodium nitrite, followed by 500ml of 4% paraformaldehyde in 0.1 M nhosohate buffer (PB, uH 7.4). Brains were removed and p&fixed for 2 h and then placed in 30% sucrose in PBS, until they sank. Sections were cut at a thickness of 60 pm on a freezing sledge microtome. One in five sections was stained with Cresyl Violet and two in five sections through the transplant stained immunocytochemically for
Effects of bFGF on transplanted
dffi
391
dopaminergic neurons
Data were analysed by split-plot analyses of variance with the treatment groups as the between-subject factor and the different tests or survival times as the within-subjects factor.” Multiple comparisons were made according to the procedure based on Sidak’s multiplicative inequality.% RESULTS
Experiment I: pre-treated transplants
Fig. 1. Schematic representation of apparatus for repeated administration of bFGF or vehicle to transplants (T) within the host caudate-putamen (CPU). A permanent indwelling 8-mm guide cannula was implanted with the tip 3 mm below dura mater. The guide cannula was maintained in position by dental cement, fixed to the skull by screws, and was filled with a stylet at all times other than when the central injections were made. The infusion cannula extended 2 mm below the tip of the guide cannula and was inserted on 10 occasions, on alternate days over 20 days after transplantation.
tyrosine hydroxylase (TH). Sections were pretreated with 3% hydrogen peroxide and 10% methanol in PBS for 5 min and then washed thoroughly in PBS. Sections were then incubated for 20min in 5% normal goat serum in PBS containing 0.1% Triton X-100. Following this sections were incubated in primary antibody dissolved in PBS containing 1% normal goat serum and 0.1% Triton X-100 at +4”C for 36 h. Sections were thoroughly washed for 20min and incubated in biotinylated secondary antibody dissolved in the same solution for 90min at +4”C. In Experiment I, a monoclonal antibody against TH (Boehringer Mannheim) was used with a biotinylated anti-mouse IgG secondary antibody (Caltag, California). In Experiment II, a polyclonal anti-TH antibody (Jacques Boy Laboratories, France) was used with a biotinylated anti-rabbit IgG secondary antibody (Vector Laboratories, U.K.). The sections were washed for a further 20 min and then incubated with a streptavidin-biotinylated horseradish peroxidase complex (Sera-Lab, U.K.) for 30 min before being washed again and transferred to PB. Antibody complexes were visualized using the 3,3’-diaminobenzidine reaction. Sections were then mounted on to gelatinized slides and allowed to air-dry, followed by dehydration in graded alcohols, clearing in Xylene, and mounting in DPX.
Drug-induced rotation. The effect of bFGFpretreatment on transplant-derived changes in amphetamine-induced rotation is shown in Fig. 2. Whereas the lesion group showed a progressive increase in amphetamine-induced rotation scores over four tests, both treated and control graft groups showed a progressive compensation of rotation scores (Groups x Tests interaction, Fs,,ss = 9.72, P < 0.001). Although it is clear that the biggest difference was that between the lesion and the two
graft groups, it also appears that the bFGF-pretreated graft animals compensated more rapidly than the control graft animals. Indeed, multiple comparisons between the two graft groups at each time point indicated a significant difference between bFGF and control groups at the three week time point (tsS = 3.29, P < 0.01) but not after six or nine weeks (te5= 2.11 and 1.18, respectively, both n.s.). Tyrosine hydroxylase-positive neuron counts. The effect of bFGF-pretreatment on the survival and differentiation of TH-positive neurons within the transplants is illustrated in Fig. 3A. Pretreatment with bFGF produced an increase in the number of TH-positive neurons observed in the transplants at both time-points (Fi,31 = 7.86, P < 0.01). In addition,
there were significantly more cells after nine weeks than after three weeks (F,,,, = 7.17, P <0.05), although this increase could not be attributed with confidence to bFGF treatment (Groups x Time interaction, F,,j, = 0.68, n.s.). Transplant volume. The effect of bFGF-pretreatment on transplant volume is illustrated in Fig. 3B.
Pretreatment 1500
of transplants
T
,:Fts
+ bFGF T
I
1 -500
increased
r
Analysis of results
Sections were analysed on a Leitz Dialux 22 microscope outlines of the transplants were traced using a Wild drawing tube attachment. The areas of transplants were calculated using a Kurta drawing pad, supported by a microcomputer running the Jandel Sigma Scan software. The respective transplant volumes were calculated by multiplying the distance between stained sections by the sum of the plotted areas for each transplant. Cell counts derived from histological sections were corrected according to the formula derived by Abercrombie.’ For the pie-treated transplants (Experiment I) the data are presented for a single graft placement, whereas in the case of the infused transplants (Experiment II) the data represent the sum of two transplant placements.
with bFGF
’ pre-op
grafts
.. ... i
I 3 weeks Rotation
/ 6 weeks
9 weeks
test
Fig. 2. Amphetamine-induced rotation scores (means + S.E.M.) in Experiment I, on the four tests prior to, and three, six and nine weeks following transplantation. In the two transplantation groups, the grafts were either pretreated with bFGF (bFGF grafts) or without the growth factor (control grafts). The third group received the lesion alone and no graft.
E. MAYERet al.
392
1600 1500 m
[~ ~
Control p r e - t r e a t m e n t b£GF ~ pre treatment ~
~
e
1200
E
800
i o o o [-
• Control grafts ~ bFGF grafts
~, E
..... 400
...... i
i ..........
1 ....
500
o
z
........... { .......
i I
-400 3 week
9 week
I )re--op
r 3 weeks
I 6 weeks
Rotation
test
Fig. 4. Amphetamine-induced rotation scores (means __+S.E.M.) in Experiment II, on the four tests prior to, and three, six and nine weeks following transplantation. In the two transplantation groups, the animals received repeated infusions of either bFGF (bFGF grafts) or the vehicle solution (control grafts) injected via implanted cannulae positioned between the two graft deposits. The third group received the lesion alone and no graft.
z
t3
I _ 9 weeks
1
3 week
9 week Survival time
Fig. 3. The effect of bFGF-pretreatment in Experiment I on the numbers of TH-positive neurons observed in transplants (A) and the total volumes of the transplants (B). Subgroups of animals were killed three and nine weeks following transplantation. Vertical bars indicate the standard errors of measurement of each value. transplant volume (F1,29 = 7.25, P < 0.05) and in addition grafts were greater in volume after nine weeks than after three weeks (Fi,m=6.85, P <0.05). Although it appears that the expansion of graft volume between three and nine weeks was greater in the control group, again this difference did not achieve statistical significance (Groups x Time i n t e r a c t i o n , FI,29 = 2.25, n.s.).
than did the control grafts (FI,36 = 15.49, P < 0.001). Although it appears that the stimulatory effect of b F G F on the number o f surviving TH-positive cells was greater at the longer survival time, this change was not significant (Groups x Time interaction, FI,36 = 2.79, n.s.). Transplant volume. The effect of bFGF-infusion on graft volume is shown in Fig. 5B. Infusion of b F G F increased the volume of the transplants (FI,36 = 21.81, P < 0.001), which remained stable between three and nine weeks (FI,36 = 0.14, n.s.). The difference between a. 1500
[[[[[] Control infusio bFGF infusion
1000
500 Z
Experiment H: infused transplants 0
Drug-induced rotation. The effect of central infusion of b F G F on amphetamine-induced rotation is shown in Fig. 4. The lesion group showed a progressive increase in amphetamine-induced net ipsilateral turns over the four test sessions, whereas both control- and bFGF-infused transplant groups showed a progressive compensation in their rotational scores (Groups x Test interaction, F6,105 = 11.85, P < 0.001). Subsequent multiple comparisons between b F G F and control infused graft groups at each time point indicated that the smaller differences between b F G F and control graft groups were nevertheless significant at both the six and nine week tests (ts5 = 2.71 and 2.59, respectively, both P < 0.05). Tyrosine hydroxylase-positive neuron counts. The effect of bFGF-infusion on the survival of THpositive neurons in the grafts is illustrated in Fig. 5A. At both time points, the b F G F - t r e a t e d grafts contained more TH-positive neurons in their transplants
B,
5
~"
4
E E
9 week
3 week
3
z
0
3 week
9 week Survival time
Fig. 5. The effect of repeated infusion of bFGF in Experiment II on the numbers of TH-positive neurons observed in transplants (A) and the total volumes of the transplants (B). Subgroups of animals were killed three and nine weeks following transplantation. Vertical bars indicate the standard errors of measurement of each value.
Effects of bFGF on transplanted dopaminergic neurons
the two groups was similar at the two time points (Groups x Time interaction, F, M= 0.20, ns.). Other h~tolog~c~l obser~=tjo~s. Light microscopic analysis of TH-immunos~in~ sections of ventral mesencephalon verified the efficacy of the lesions in all animals in both Experiment I (pretreatment of transplants) and Experiment II (the effects of repeated infusion). Transplants were found to contain TH-positive neurons in all grafted animals (illustrated in Fig. 6). In the first experiment, initial staining using a biotinylated anti-mouse IgG secondary antibody to label the monoclonal primary antibody produced a very high level of background staining over a wide area around the sites of the central infusions. The response was as strong in the vehicle-treated as in the bFGF-treated animals, and was reproduced by reaction with the secondary antibody alone. This suggested that repeated central infusions stimulated a mild but signifi~nt ongoing in~~matory response in the host animals. In contrast, no similar baekground reaction was obtained using the anti-rabbit IgG secondary antibody, which was the reason for changing to the polyclonal primary antibodies in the second experiment. DIscussICWl The present results indicate that bFGF can promote the survival and functional efficacy of nigral transplants in the unilateral 6-OHDA rat model of Parkinsonism. The effects were only slight when the growth factor was administered as a brief pretreatment at the time of graft implantation, but had a more pronounced effect when administered centrally over 20 days follo~ng transplan~tion. Basic fibroblast growth factor effects on the lesions The present results appear at first to contrast with the observations of Otto and Unsicker,24 who found that bFGF could promote recovery from MPTP lesion-induced deficits. By contrast, bFGF treatment yielded no apparent sparing of the lesions when made with the toxin B-OHDA in the present experiment, nor prompted any apparent recovery in the lesioned animals. It is likely that the major reason for this difference relates to both the timing and extent of the lesions. Thus, MPTP lesions leave the more medial ventral tegmental area neurons in the ventral mesencephalon largely intact and spare mesolimbic dopamine projections to the ventral striatum. It is known
393
that a variety of trophic influences, such as adrenal graft implants or additional lesions alone, can stimulate recovery from MPTP lesions by promoting sprouting of the ventral dopaminergic system into dorsal striatal areas.’ In the present experiments, all lesions were verified histologically and were found to be virtually complete, and previous experience indicates that the neurons of the ventral mesencephalon are lost within seven days of 6-OHDA injection. Thus host dopamine neurons would no longer be available to be rescued at the time of bFGF treatment in the host brain three weeks after lesion surgery. The efficacy of bFGF to promote the survival of embryonic TH-positive neurons within the grafts is fully compatible with the observations of Otto and Unsickerz4 of the capacity of this growth factor to protect the same population of neurons in culture from the oxidative stress induced by MPTP. Pretreatment The results of Experiment I, suggested that bFGF pretreatment could enhance the initial recovery of lesion-induced asymmetry by implants of nigral grafts. Thus, the differences between the treated and control graft groups were significant three weeks after implantation but were no longer so by the six- and nine-week tests. This pattern of resuits could be explained in several ways. First, bFGF may have had an independent pharmacological effect on the behavioural asymmetry, but after acute treatment was not long lasting. This would require that the single bolus of bFGF should be available for a longer period than would be expected from its rather transient half life in vitro, but it is possible that prolonged availability of the present bFGF pre-treatments may have been promoted by stabilization in the extracellular matrix,*’ from which it could have been released by the action of proteases. Second, a perhaps more plausible explanation is that bFGF accelerated the maturation of TH-positive neurones or facilitated early outgrowth from the transplant into the host striatum. The fact that the control transplants did eventually attain a similar level of recovery suggests that bFGF-pretreatment increased the rate at which early transplant maturation and/or graft-host interactions occurred. Nevertheless, bFGF did increase the number of TH-positive neurons surviving within transplants. This may reflect (i) an increased resistance to the trauma associated with the production of a cell suspension, (ii) a direct trophic effect of bFGF on the
Fig. 6. TH immunocytochemistry. Representative examples are shown at both low and high magnifications of transplants treated with bFGF or control treatments in Experiments I and II. (A,B) Control transplant of a single 3941 deposit of the embryonic nigral cell suspension from Experiment I. (CD) Transplant pretreated with bFGF from Experiment 1. (E,F) Control transplant of one of the two 39.41deposits of the embryonic nigral cell suspension from Experiment II. (G,H) Transplant which has received repeated infusions of bFGF from experiment II. All examples are based on animals surviving nine weeks following transplantation. Scale bars = 500 ,um (A-D), 100 grn (E-H).
I
ISC 56iZ--F
Fig. 6E-H.
survival of grafted dopmaine neurons, or (iii) an increased division of TH-positive neurons in the grafts.lY The fact that bFGF-pretreatment increased transplant volume to a similar extent suggests that any trophic or mitogenic influences are not restricted to the dopaminergic TH-positive neurons but apply to many if not all cell populat~o~~s within the grafts. Whereas this may reflect an increase in the number of glial ceils within transplants,* recent evidence points to the fact that bFGF does have a direct survival-promoting effect on neurons.?” The fact that this greater cell survival was not reflected by larger long-term functional effects may reflect the suggestion that the amphetamine rotation test is exquisitely sensitive to dopaminergic reinnervation, such that as few as 200-300 cells need survive to yield complete recovery by this measure3 Thus, all grafts in the present experiment, bFGF-treated and control alike, exceeded this level of cell survival and consequently any differences in the functional changes were masked by the ceiling level of recovery.
The resutts of Experiment II suggested that repeated intracerebral infusions of bFGF into the vicinity of transplants produced a greater increase in TH-positive neuron survival and more prolonged functional benefit than had been achieved with the single pretreatment. In this case the magnitude of the effects were sufficient to yield clearly signi~~nt benefits in terms of both cell survival and of functional recovery that became progressively greater at longer survival times, up to the nine-week tests, a full six weeks after the termination of the actual bFGF treatment. However, the presence of a marked inflammatory reaction in response to repeated intracerebral infusions of bFGF compIicates both the interpretation of the present results and their potential applications. At the practical level, the resulting cross-reactivity with the mouse IgG secondary antibodies restricted the range of primary antibody reactions available to characterize the grafts. More fundamentally, the observed reaction implies the induction of a variety of inflammatory mediators, which may exert neurotrophic actions in their own right. For example, interleukin-6 has been found to promote the survival of postnatal dopaminergic neurons in z&o,‘” and implants of interleukin-1 pellets can promote recovery from partial nigrostriatal lesions,3i Indeed, this process may in fact be accompanied by the induction of wound-derived neurotrophic activity,20 which may itself be attributable to an FGF-like molecule. Thus, the presence of an inflammatory response (produced in relation to the repeated trauma associated with the infusion process) may provide a neurotrophic influence in its own right or may complement the action of bFGF on the development of the transplants. Nevertheless, a similar degree of inflammatory
response was seen in the animals receiving vehicle infusions. Consequently, the larger number 01 surviving TH-positive neurons in the bFGF-treated grafts and the substantially greater functional recovery in these animals, was not simply attributable to inflammatory mediators, even if they may have contributed to the efficacy of the bFGF treatment.
Whereas some tissues (e.g. neocortex, striatun~) survive and grow well following transplantation, grafts of brainstem catecholamine neurons have generally been found to provide relatively low yields of surviving cells. The present results endorse the general approach of att~pting to enhance the viability of embryonic ventral mesencephalic transplants with potential neurotrophic molecules. Thus, Nishino and coileagues2? have found that tGS gangliosides can promote graft survival and recovery in transplanted hemiparkinsonian rats on similar motor tests to those employed here. Although there are still only a few studies addressing the problem of promoting dopamine neuron survival following transpbnthe tation, initial studies suggest that bFGF, insulin-like growth factors, brain-derived neurotrophic factor, platelet-derived growth factor, and interleukin-6 can all prove effective in promoting the viability of dopamine neurons in ~*itro.‘“‘.“’ ‘5.iq.11 So far, only the first of these has been investigated in any detail in oiw.5~27 The present data provide further evidence that bFGF can increase the survival of transplanted fetal dopaminergic neurons, and indicates that the mild benefit reported by Steinbusch and colleagues is not simply attributable to their use of heparin, which was not used in the present experiments.
CONCIAJSION
Pretreatment of a cell s~s~ns~on provides a moderate and relatively short-lived benefit on transplant eficacy, whereas repeated infusions of bFGF produce a long-term increase in cell survival and the functional efficacy of the transplants. However, repeated infusions, at least with the procedures adopted in the present experiment, produce an inflammato~ reaction which makes it difficult to attribute the functional effects to bFGF only, and complicates the use of similar treatments in any clinical application. Future work must be directed both to improving modes of growth factor administration, and to investigating the potential of other neurotrophic molecules, singly and in combination with members of the FGF family, to promote graft survival. ~c~nou,ledge~nts-ibis work was supported by the Medical Research Council.
Effects of bFGF on transplanted
dopaminergic neurons
391
REFERENCES
1. Abercrombie M. (1946) Estimation of the nuclear population from microtome sections. Anat. Rec. 94, 239-247. 2. Bohn M., Bohn M. C., Cupit L., Marciano F. and Gash D. M. (1987) Adrenal medulla grafts enhance recovery of striatal dopaminergic fibers. Science 237, 913-916. 3. Brundin P., Isacson 0. and Bjiirkhmd A. (1985) Monitoring cell viability in suspensions of embryonic CNS tissue and its use as a criterion for intracerebral graft survival. Brain Res. 331, 251-259. 4. Burgess W. and Macaig T. (1989) The heparin binding (fibroblast) growth factor family of proteins. A. Rev. Biochem. 58, 575-606. 5. Dolleman-Van Der Wee1 M. J., Nijssen A., De Vente J., Ramaekers F. C. S., Cordell B., Fuller F. and Steinbusch
H. W. M. (1991) Morphological and behavioural effects of basic fibroblast growth factor and heparin on transplanted fetal dopaminergic neurones and astrocytes in the denervated rat caudate-putamen. In Growth Factors in Alzheimer’s Disease (eds Hefti F., Brachet Ph., Will B. and Christen Y.), pp. 149-164. Springer, Berlin. 6. Dunnett S. B. (1991) Transplantation of embryonic dopamine neurons: what we know from rats. J. Neural. 238, 65-14. 7. Eckenstein F. P., Shipley G. D. and Nishi R. (1991) Acidic and basic fibroblast growth factors in the nervous system: distribution and differential alteration of levels after injury of central versus peripheral nerve. J. Neurosci. 11,
412419. 8. Eclancher F., Perraud F., Faltin J., Labourdette 9.
10. 11. 12. 13.
14. 15.
16.
G. and Sensenbrenner M. (1990) Reactive astrogliosis after basic fibroblast growth factor (bFGF) injection in injured neonatal rat brain. GIia’3, 5022509. Engele J. and Bohn M. C. (1991) The neurotrophic effects of fibroblast growth factors on dopaminergic neurons in vitro are mediated by mesencephalic glia. J. Neurosci. 11, 3070-3078. Ferrari G., Minozzi M. C., Toffano G., Leon A. and Skaper S. D. (1989) Basic fibroblast growth factor promotes the survival and development of mesencephalic neurones in culture. Deul Biol. 133, l&147. Gage F. H., Fisher L. J., Jinnah J. H. A., Rosenberg M. B., Tuszynski M. and Friedmann T. (1990) Grafting genetically modified cells to the brain: conceptual and technical issues. Prog. Brain Res, 82, l-10. Giacobini M. M. J., Hoffer B. J., Zerbe G. and Olson L. (1991) Acidic and basic fibroblast growth factors augment growth of fetal brain tissue grafts. Expl Brain Res. 86, 73-81. Hama T., Kushima Y., Miyamoto M., Kubota M., Takei N. and Hatanaka H. (1991) Interleukin-6 improves the survival of mesencephalic catecholaminergic and septal cholinergic neurones from postnatal, two-week-old rats in cultures. Neuroscience 40, 445-452. Hyman C., Hofer M., Barde Y. A., Juhasz M., Yancopoulos G. D., Squint0 S. P. and Lindsay R. M. (1991) BDNF is-a neurotrophic factor for dopaminergic neurones of the substantia nigra. Nature 350, 230-232. Knusel B.. Michel P. P.. Schwaber J. S. and Hefti F. (1990) Selective and non-selective stimulation of central cholinereic and dopaminergic development in vitro by nerve growth factor, basic fibroblast growth factor, epidermal growth factor, insulin and the insulin-like growth factors I and II. J. Neurosci. 10, 558-570. Lindvall 0. (1989) Transplantation into the human brain: present status and future possibilities. J. Neural. Neurosurg. Psychiat., Spec. Suppl. 39-54.
17. Lindvall
0.
(1991) Prospects
of transplantation
in human
neurodegenerative
diseases.
Trenak Neurosci.
14,
376-384.
18. Mayer E., Fawcett J. W. and Dunnett S. B. (1993) Basic FGF promotes the survival of embryonic ventral mesencephalic dopamine neurones. I. Effects in vitro. Neuroscience 56, 379-388. 19. Mayer E., Dunnett S. B. and Fawcett J. W. (1993) Mitogenic effect of basic fibroblast growth factor on embryonic ventral mesencephalic dopaminergic neurone precursors. Deul Brain Res. (in press). 20. Nieto-Sampedro M., Manthorpe M., Barbin G., Varon S. and Cotman C. W. (1983) Injury induced neuronotrophic activity in adult rat brain: correlation with survival of delayed implants in the wound cavity. J. Neurosci. 3, 2219-2229.
21. Nikkhah G., Odin P., Smits A., Tingstrijm A., Othberg A., Brundin P., Funa K., and Lindvall 0. (1993) Platelet-derived growth factor promotes survival of rat and human mesencephalic dopaminergic neurons in culture. Expl Brain Res. 92, 516523.
22. Nishino H., Hashitani T., Isobe Y., Furayama F., Sato H., Kumazaki M., Hirokomi K. and Awaya A. (1990) tGS ganglioside induces peculiar morphological features in grafted dopaminergic cells and promotes motor recovery in rats with unilateral lesions in the nigrostriatal dopamine pathway. Brain Res. 534, 73-82. 23. Olson L., Ayer-LeLievre C., Ebendal T., Eriksdotter-Nilsson M., Emfors P., Henschen A., Hoffer B., Giacobini M. B., Mouton P., Palmer M., Persson H., Sara V., Stromberg I. and Wetmore C. (1990) Grafts, growth factors, and grafts that make growth factors. Prog. Brain Res. 82, 5546. 24. Otto D. and Unsicker K. (1990) Basic FGF reverses chemical and morphological deficits in the nigrostriatal system of MPTP-treated mice. J. Neurosci. 10, 1912-1921. 25. Rifkin D. B. and Moscatelli D. (1989) Recent developments in the cell biology of basic fibroblast growth factor. J. Cell Biol. 109, Id. 26. Rohlf F. J. and Sokal R. R. (1981) Statistical Tables. Freeman and Co., New York. 27. Steinbursch H. W. M., Vermeulen R. J. and Tonnaer J. A. D. M. (1990) Basic fibroblast growth factor enhances the survival and sprouting of fetal dopaminergic cells implanted in the denervated rat caudate-putamen: preliminary observations. Prog. Brain Res. 82, 81-86. 28. Ungerstedt U. (1971) Striatal dopamine release after amphetamine or nerve degeneration revealed by rotational behaviour. Acta physiol. stand., Suppl. 367, 4968. 29. Ungerstedt U. and Arbuthnott G. W. (1970) Quantitative recording of rotational behaviour of rats after 6hydroxydopamine lesions of the nigrostriatal dopamine system. Brain Res. 24, 485493. 30. Unsicker K., Reichert-Preibsch H. and Wewetzer K. (1992) Stimulation of neuron survival by basic FGF and CNTF is a direct effect and not mediated by non-neuronal cells: evidence from single cell cultures. Devl Brain Res. 65,285-288. 31. Wang J., Plunkett R. J., Sheng J. G., Oldfield E. H. and Bankiewicz K. S. (1991) Recovery in hemiparkinsonian rats after intracaudal implantation of IL-1 pellets. Sot. Neurosci. Abstr. 17, 348. 32. Williams L. R., Vahlsing H. L., Lindamood T., Varon S., Gage F. H. and Manthorpe M. (1987) A small gauge cannula device for continuous infusion of exogenous agents into the brain. Expl Neural. 95, 734754.
39x
I’.. MAYER YI ul
33. Williams L. R., Varon S., Peterson G. M.. Wictorin K., Fischer W., Bjorklund A. and Gage F. H. (IY86) Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbria fornix transection. Prf~ nrun. Acad. Sci. U.S.A. 83, 9231-9235. 34. Winer B. J. (1971) Sfatistical Principles in E.rperimental Design, 2nd edn. McGraw--Hill. New York. (Accepted
IO May
1993)
0306-4522/93$6.00 + 0.00 Pergamon Press Ltd IBRO
BASIC FIBROBLAST GROWTH FACTOR PROMOTES THE SURVIVAL OF EMBRYONIC VENTRAL MESENCEPHALIC DOPAMINERGIC NEURONS-II. EFFECTS ON NIGRAL TRANSPLANTS IN I/l’lrO E. MAYER,*? J. W. FAWCETT*$ and S. B. D~*~§ *MRC
Cambridge Centre for Brain Repair, and the Departments of tExperimenta1 Psychology and $Physioiogy, University of Cambridge, Downing Street, Cambridge, U.K.
At&raft--The clinical potential of transplants of fetal dopaminergic neurons is limited by the fact that the percentage of cells surviving in such grafts is in general quite low. This report investigates the use of basic fibroblast growth factor administration (given either as a pretreatment or by repeated intrastriatal infusions) to promote the survival and behavioural efficacy of embryonic dopamine-rich nigral transplants in rats. pretreatment of the graft tissue by brief incubation with basic fibroblast growth factor increased the survival of tyrosine hydroxyl~-immunoreactive (presumed do~~ner~c) neurons in the grafts in comparison to control grafts, and accelerated the recovery in the transplants animals in tests of drug-induced rotational asymmetry. However, the clear advantage seen in the rotation test conducted three weeks after transplantation had disappeared by nine weeks. The moderate effects of pretreatment were markedly enhanced by repeated intrastriatal infusion of basic fibroblast growth factor into the host animals over 20 days following transplantation. This resulted in > 100% increase in the number of dopaminergic neurons surviving in the grafts, and was accompanied by a significantly greater recovery of the rats’ rotational asymmetries which persisted over the full nine weeks of testing. However, the repeated intracerebral infusions induced an inflammatory reaction in the striatum, and the associated trauma both complicates the interpretation of the mechanism of observed recovery and compromises the utility of this route of basic fibroblast growth factor administration for promoting graft survivai.
One of the major
problems that is emerging from recent clinical trials of embryonic neural transplantation in Parkinson’s disease is the unreliabile sur-
vival of grafted dopaminergic neuronsi6*” Similarly, when dopamine neurons are grafted as a dissociated cell suspension in rats, the yield of surviving cells is frequently much lower (l-5%) than is typically found in many other types of graft.) It is increasingly apparent that the survival and growth of grafted neurons is dependent upon the availability of a variety of growth factors “,m~23 which raises the possibility that graft survival might be promoted by exogenous t~a~ent with selected growth factors. Several studies indicate that basic fibroblast growth factor (bFGFr can promote the survival of a variety of embryonic neurons, among them ventral mesencephalic dopaminergic neurons.Lo,isJ8J9 Recent evidence suggests that at least part of this effect in culture is mediated by astrocytes.9~‘8 When administered to adult dopaminergic ventral mesencephalic $To whom correspondence should be addressed at: Department of Experimental Psychology, University of Cambridge, Dow&g Street, CarnbridgeCBZ 3EB, U.K. Abbreuiations: bFGF, basic fibroblast growth factor: EDTA, ethyldiaminetetraacetate; MPT?, l-methyl-& phenyl-tetrahydropyridine; PB, phosphate buffer; PBS, phosphatebuffered saline; TH, tyrosine hydroxylase; &OHDA, dhydroxydopamine.
neurons that have been treated with l-methyl4 phenyl-tetrahydropyridine (MPTP), bFGF is able to rescue them from atrophy or death.24 Moreover, repeated infusions of bFGF have been shown to exert a potent neurotrophic effect on the growth of a variety of brain areas transplanted into the anterior chamber of the eye. I2 In the context of promoting the survival of dopaminergic grafts, Steinbusch and coworker@’ have found that bFGF pretreatment increases the survival and growth of dopaminergic neurons within transplants. However, they have cautioned that this effect may be attributabie to heparin which was co-administered, in their experiments, in order to stabilize bFGF. However, other observations suggest that whereas the effects of acidic FGF are strongly dependent on the presence of heparin, the efficacy of bFGF may be potentiated only slightly by it.’ Thus, bFGF may prove a useful tool for promoting the viability of dopamine neurons in embryonic transplants. In the a~mpanying paper’” we show that bFGF can indeed promote the survival of dopamine neurons over at least four weeks in vitro. Unilateral lesions of the nigrostriatal pathway deplete forebrain dopamine and have provided the most widely used animal model of Parkinsonism within which to study the viability of nigral grafts6 Therefore, in the present study, we have used this lesion
390
1:. lh,AYtK
and graft model to investigate the effect of administration of bFGF (either as a pretreatment or by repeated central injections into the host striatum) on the survival and functional efficacy of dopamine-rich embryonic nigral tissues transplants in ho. EXPERIMENTAL
PROCEDURES
Subjects
Young adult female rats of the Sprague-Dawley strain (Harlan-Olac, Bicester, Oxon, U.K.) were used as transplant recipients, weighing approximately 200 g at the time of first surgery. They were housed under standard lightdark conditions, with food and water supplied ad libitum. Transplant tissues were dissected from embryos (crown-rump length _ 10mm) collected by caesarean section from pregnant rats of the same outbred strain. Lesions
A total of 108 rats received lesions, 48 in Experiment I and 60 in Experiment II. Rats were anaesthetized with 3 ml/kg equithesin anaesthesia and mounted in a Kopf stereotaxic frame. The toxin 6-hydroxydopamine hydrobromine (dOHDA; Sigma) was dissolved to a concentration of 3 mg/ml (free base weight) in 0.02% ascorbate saline. Four microlitres of the toxin was infused at a rate of 1 pl/min via a 30-gauge stainless steel cannula which was stereotaxically implanted into the right ascending nigrostriatal bundle at coordinates: A = -4.4 mm (caudal to bregma), L = 1.Omm (lateral to the midline), and V = 7.8 mm (ventral to the dura mater). with the nose bar set -2.3 mm below the interaural line. A further 4 min was allowed for diffusion before withdrawing the cannula. The wound was then cleaned and sutured. No futher postoperative care was required. Rotation tests
The efficacy of lesions was checked using the amphetamine-induced rotation model of Ungerstedt2* Animals were tested in automated rotometer bowls, modelled after the design of Ungerstedt and Arbuthnott,29 and connected to a microcomputer for data collection. Approximately 15 days following the lesion surgery, all animals were injected with 2.5 mg/kg methamphetamine hydrochloride (Sigma) and full 360” turns in either direction were recorded over a 90-min test-period. Only animals making >400 net ipsilatera1 turns over 90min were considered to have effective lesions for inclusion in the subsequent experiments. Animals were allocated to counterbalanced groups on the basis of their rotation scores. Further 90-min rotation tests were carried out using the same dose of amphetamine at three, six and nine weeks after transplantation. Neural transplantation
Cell suspensions for transplantations were prepared from fetal ventral mesencephalic tissue according to the same protocol used in the parallel in vitro experiments.‘* Briefly, donor embryos were removed from pregnant female rats, under terminal barbiturate anaesthesia. The ventral mesencephalon, containing the dopamine cells of the developing substantia nigra, was dissected in Leibowitz L-15 medium (Imperial Laboratories), the meninges removed, and the pieces incubated in 0.1% trypsin (type V, Sigma) with 0.02% EDTA (Imperial Laboratories) for 6 min at 37°C. A solution of 0.04% deoxyribonuclease (Sigma) at 37°C was added, the pieces shaken gently and centrifuged for 1 min at 100 g. The supernatant was then removed and a triturating solution was added [3 mg of bovine serum albumin, 10 mg deoxyribonuclease, 0.5 mg soybean trypsin inhibitor and 0.02% EDTA per ml of phosphate-buffered saline (PBS)], and the cells triturated gently through a 23-g needle
<‘I id
mounted on a l-ml syringe. The resulting cell suspension was then centrifuged at 1OOg for 4 min and the triturating solution removed. The cells were then resuspended to ;I concentration of one embryonic piece per 2 ~1 of medium The graft tissue was implanted by stereotaxic injection using a lo-p1 glass microsyringe (Scientific Glass Engineering, Ringwood, Australia). A 3-pl volume of ihc cell suspension was injected directly into the right (lesioned) striatum of the host rats at a rate of 1pl/min, with a further 5 min allowed before slowly withdrawing the syringe needle. Since it was necessary to make more than one suspension to graft all of the animals involved in one experiment. each suspension was used to graft equal numbers of animals in vehicle- and bFGF-treated groups. In Experiment I, each cell suspension was divided into two equal parts, with 10 ng/5 ~1 of bFGF added to one aliquot and a similar volume of vehicle to the other. A single graft deposit cell suspension was injected into the right striatum at the following coordinates: A = 1.0 mm. L = 3.0mm and V = 5.0mm, with the nosebar set at -2.3 mm. In total, 18 rats received bFGF pre-treated grafts, 18 received control grafts, and 12 remained with lesions alone. In Experiment II, two graft deposits were placed at the following coordinates: A = 1.6 mm, L = 2 mm, V = 4.5 mm and A = -0.4 mm, L = 4 mm, V = 5 mm, with the nosebar set at -2.3 mm. bFGF or vehicle was infused into the host striatum in proximity to the grafts via a subsequently implanted cannula (see below). In total, 24 rats received bFGF treated grafts, 24 received control grafts. and 12 remained with lesions alone. Cannulation and infusion
Basic FGF does not cross the blood-brain barrier, and so must be administered centrally. Since the relatively short biological half-life of bFGF did not permit its continuous infusion from a reservoir in a minipump, as has been possible with nerve growth factor,‘2.” repeated injections were made via implanted cannulae in order to deliver exogenous bFGF to the transplants. An 8-mm-long 23gauge stainless steel guide cannula was stereotaxicahy implanted into the striatum halfway between the two graft deposits (A = 0.6 mm, L = 3 mm. V = 3 mm) and attached to the skull with stainless steel screws and dental cement (see Fig. 1). The cannula was maintained with an g-mm stylet in order to avoid blockages. To make the intracerebral injections, a IO-mm 30-gauge stainless steel infusion cannula was connected via thin polyethylene tubing to a IO-PI glass microsyringe mounted in a Harvard microinfusion pump. The stylet was removed from the guide cannula in the skull, the infusion cannula was inserted, and 20 ng human recombinant bFGF (Boehringer, Mannheim) was infused over 2 min in 1~1 of vehicle (0.1% bovine serum albumin in 0.1 M PBS), followed by a 3-min diffusion time. Control grafts received similar infusions of 1 ~1 vehicle alone. A total of 10 repeated infusions were made in each animal, once every two days over 20 days, commencing on the second day after transplantation. Histological analysis
Animals were killed either three weeks (Experiment I, n = 16, Experiment II, n = 28) or nine weeks (Experiment I, n = 32, Experiment II, n = 32) after implantation in a balanced manner between groups. Rats were perfused transcardially under terminal barbiturate anaesthesia with 200 ml of ice-cold 0.1 M PBS containing 0.02% sodium nitrite, followed by 500ml of 4% paraformaldehyde in 0.1 M nhosohate buffer (PB, uH 7.4). Brains were removed and p&fixed for 2 h and then placed in 30% sucrose in PBS, until they sank. Sections were cut at a thickness of 60 pm on a freezing sledge microtome. One in five sections was stained with Cresyl Violet and two in five sections through the transplant stained immunocytochemically for
Effects of bFGF on transplanted
dffi
391
dopaminergic neurons
Data were analysed by split-plot analyses of variance with the treatment groups as the between-subject factor and the different tests or survival times as the within-subjects factor.” Multiple comparisons were made according to the procedure based on Sidak’s multiplicative inequality.% RESULTS
Experiment I: pre-treated transplants
Fig. 1. Schematic representation of apparatus for repeated administration of bFGF or vehicle to transplants (T) within the host caudate-putamen (CPU). A permanent indwelling 8-mm guide cannula was implanted with the tip 3 mm below dura mater. The guide cannula was maintained in position by dental cement, fixed to the skull by screws, and was filled with a stylet at all times other than when the central injections were made. The infusion cannula extended 2 mm below the tip of the guide cannula and was inserted on 10 occasions, on alternate days over 20 days after transplantation.
tyrosine hydroxylase (TH). Sections were pretreated with 3% hydrogen peroxide and 10% methanol in PBS for 5 min and then washed thoroughly in PBS. Sections were then incubated for 20min in 5% normal goat serum in PBS containing 0.1% Triton X-100. Following this sections were incubated in primary antibody dissolved in PBS containing 1% normal goat serum and 0.1% Triton X-100 at +4”C for 36 h. Sections were thoroughly washed for 20min and incubated in biotinylated secondary antibody dissolved in the same solution for 90min at +4”C. In Experiment I, a monoclonal antibody against TH (Boehringer Mannheim) was used with a biotinylated anti-mouse IgG secondary antibody (Caltag, California). In Experiment II, a polyclonal anti-TH antibody (Jacques Boy Laboratories, France) was used with a biotinylated anti-rabbit IgG secondary antibody (Vector Laboratories, U.K.). The sections were washed for a further 20 min and then incubated with a streptavidin-biotinylated horseradish peroxidase complex (Sera-Lab, U.K.) for 30 min before being washed again and transferred to PB. Antibody complexes were visualized using the 3,3’-diaminobenzidine reaction. Sections were then mounted on to gelatinized slides and allowed to air-dry, followed by dehydration in graded alcohols, clearing in Xylene, and mounting in DPX.
Drug-induced rotation. The effect of bFGFpretreatment on transplant-derived changes in amphetamine-induced rotation is shown in Fig. 2. Whereas the lesion group showed a progressive increase in amphetamine-induced rotation scores over four tests, both treated and control graft groups showed a progressive compensation of rotation scores (Groups x Tests interaction, Fs,,ss = 9.72, P < 0.001). Although it is clear that the biggest difference was that between the lesion and the two
graft groups, it also appears that the bFGF-pretreated graft animals compensated more rapidly than the control graft animals. Indeed, multiple comparisons between the two graft groups at each time point indicated a significant difference between bFGF and control groups at the three week time point (tsS = 3.29, P < 0.01) but not after six or nine weeks (te5= 2.11 and 1.18, respectively, both n.s.). Tyrosine hydroxylase-positive neuron counts. The effect of bFGF-pretreatment on the survival and differentiation of TH-positive neurons within the transplants is illustrated in Fig. 3A. Pretreatment with bFGF produced an increase in the number of TH-positive neurons observed in the transplants at both time-points (Fi,31 = 7.86, P < 0.01). In addition,
there were significantly more cells after nine weeks than after three weeks (F,,,, = 7.17, P <0.05), although this increase could not be attributed with confidence to bFGF treatment (Groups x Time interaction, F,,j, = 0.68, n.s.). Transplant volume. The effect of bFGF-pretreatment on transplant volume is illustrated in Fig. 3B.
Pretreatment 1500
of transplants
T
,:Fts
+ bFGF T
I
1 -500
increased
r
Analysis of results
Sections were analysed on a Leitz Dialux 22 microscope outlines of the transplants were traced using a Wild drawing tube attachment. The areas of transplants were calculated using a Kurta drawing pad, supported by a microcomputer running the Jandel Sigma Scan software. The respective transplant volumes were calculated by multiplying the distance between stained sections by the sum of the plotted areas for each transplant. Cell counts derived from histological sections were corrected according to the formula derived by Abercrombie.’ For the pie-treated transplants (Experiment I) the data are presented for a single graft placement, whereas in the case of the infused transplants (Experiment II) the data represent the sum of two transplant placements.
with bFGF
’ pre-op
grafts
.. ... i
I 3 weeks Rotation
/ 6 weeks
9 weeks
test
Fig. 2. Amphetamine-induced rotation scores (means + S.E.M.) in Experiment I, on the four tests prior to, and three, six and nine weeks following transplantation. In the two transplantation groups, the grafts were either pretreated with bFGF (bFGF grafts) or without the growth factor (control grafts). The third group received the lesion alone and no graft.
E. MAYERet al.
392
1600 1500 m
[~ ~
Control p r e - t r e a t m e n t b£GF ~ pre treatment ~
~
e
1200
E
800
i o o o [-
• Control grafts ~ bFGF grafts
~, E
..... 400
...... i
i ..........
1 ....
500
o
z
........... { .......
i I
-400 3 week
9 week
I )re--op
r 3 weeks
I 6 weeks
Rotation
test
Fig. 4. Amphetamine-induced rotation scores (means __+S.E.M.) in Experiment II, on the four tests prior to, and three, six and nine weeks following transplantation. In the two transplantation groups, the animals received repeated infusions of either bFGF (bFGF grafts) or the vehicle solution (control grafts) injected via implanted cannulae positioned between the two graft deposits. The third group received the lesion alone and no graft.
z
t3
I _ 9 weeks
1
3 week
9 week Survival time
Fig. 3. The effect of bFGF-pretreatment in Experiment I on the numbers of TH-positive neurons observed in transplants (A) and the total volumes of the transplants (B). Subgroups of animals were killed three and nine weeks following transplantation. Vertical bars indicate the standard errors of measurement of each value. transplant volume (F1,29 = 7.25, P < 0.05) and in addition grafts were greater in volume after nine weeks than after three weeks (Fi,m=6.85, P <0.05). Although it appears that the expansion of graft volume between three and nine weeks was greater in the control group, again this difference did not achieve statistical significance (Groups x Time i n t e r a c t i o n , FI,29 = 2.25, n.s.).
than did the control grafts (FI,36 = 15.49, P < 0.001). Although it appears that the stimulatory effect of b F G F on the number o f surviving TH-positive cells was greater at the longer survival time, this change was not significant (Groups x Time interaction, FI,36 = 2.79, n.s.). Transplant volume. The effect of bFGF-infusion on graft volume is shown in Fig. 5B. Infusion of b F G F increased the volume of the transplants (FI,36 = 21.81, P < 0.001), which remained stable between three and nine weeks (FI,36 = 0.14, n.s.). The difference between a. 1500
[[[[[] Control infusio bFGF infusion
1000
500 Z
Experiment H: infused transplants 0
Drug-induced rotation. The effect of central infusion of b F G F on amphetamine-induced rotation is shown in Fig. 4. The lesion group showed a progressive increase in amphetamine-induced net ipsilateral turns over the four test sessions, whereas both control- and bFGF-infused transplant groups showed a progressive compensation in their rotational scores (Groups x Test interaction, F6,105 = 11.85, P < 0.001). Subsequent multiple comparisons between b F G F and control infused graft groups at each time point indicated that the smaller differences between b F G F and control graft groups were nevertheless significant at both the six and nine week tests (ts5 = 2.71 and 2.59, respectively, both P < 0.05). Tyrosine hydroxylase-positive neuron counts. The effect of bFGF-infusion on the survival of THpositive neurons in the grafts is illustrated in Fig. 5A. At both time points, the b F G F - t r e a t e d grafts contained more TH-positive neurons in their transplants
B,
5
~"
4
E E
9 week
3 week
3
z
0
3 week
9 week Survival time
Fig. 5. The effect of repeated infusion of bFGF in Experiment II on the numbers of TH-positive neurons observed in transplants (A) and the total volumes of the transplants (B). Subgroups of animals were killed three and nine weeks following transplantation. Vertical bars indicate the standard errors of measurement of each value.
Effects of bFGF on transplanted dopaminergic neurons
the two groups was similar at the two time points (Groups x Time interaction, F, M= 0.20, ns.). Other h~tolog~c~l obser~=tjo~s. Light microscopic analysis of TH-immunos~in~ sections of ventral mesencephalon verified the efficacy of the lesions in all animals in both Experiment I (pretreatment of transplants) and Experiment II (the effects of repeated infusion). Transplants were found to contain TH-positive neurons in all grafted animals (illustrated in Fig. 6). In the first experiment, initial staining using a biotinylated anti-mouse IgG secondary antibody to label the monoclonal primary antibody produced a very high level of background staining over a wide area around the sites of the central infusions. The response was as strong in the vehicle-treated as in the bFGF-treated animals, and was reproduced by reaction with the secondary antibody alone. This suggested that repeated central infusions stimulated a mild but signifi~nt ongoing in~~matory response in the host animals. In contrast, no similar baekground reaction was obtained using the anti-rabbit IgG secondary antibody, which was the reason for changing to the polyclonal primary antibodies in the second experiment. DIscussICWl The present results indicate that bFGF can promote the survival and functional efficacy of nigral transplants in the unilateral 6-OHDA rat model of Parkinsonism. The effects were only slight when the growth factor was administered as a brief pretreatment at the time of graft implantation, but had a more pronounced effect when administered centrally over 20 days follo~ng transplan~tion. Basic fibroblast growth factor effects on the lesions The present results appear at first to contrast with the observations of Otto and Unsicker,24 who found that bFGF could promote recovery from MPTP lesion-induced deficits. By contrast, bFGF treatment yielded no apparent sparing of the lesions when made with the toxin B-OHDA in the present experiment, nor prompted any apparent recovery in the lesioned animals. It is likely that the major reason for this difference relates to both the timing and extent of the lesions. Thus, MPTP lesions leave the more medial ventral tegmental area neurons in the ventral mesencephalon largely intact and spare mesolimbic dopamine projections to the ventral striatum. It is known
393
that a variety of trophic influences, such as adrenal graft implants or additional lesions alone, can stimulate recovery from MPTP lesions by promoting sprouting of the ventral dopaminergic system into dorsal striatal areas.’ In the present experiments, all lesions were verified histologically and were found to be virtually complete, and previous experience indicates that the neurons of the ventral mesencephalon are lost within seven days of 6-OHDA injection. Thus host dopamine neurons would no longer be available to be rescued at the time of bFGF treatment in the host brain three weeks after lesion surgery. The efficacy of bFGF to promote the survival of embryonic TH-positive neurons within the grafts is fully compatible with the observations of Otto and Unsickerz4 of the capacity of this growth factor to protect the same population of neurons in culture from the oxidative stress induced by MPTP. Pretreatment The results of Experiment I, suggested that bFGF pretreatment could enhance the initial recovery of lesion-induced asymmetry by implants of nigral grafts. Thus, the differences between the treated and control graft groups were significant three weeks after implantation but were no longer so by the six- and nine-week tests. This pattern of resuits could be explained in several ways. First, bFGF may have had an independent pharmacological effect on the behavioural asymmetry, but after acute treatment was not long lasting. This would require that the single bolus of bFGF should be available for a longer period than would be expected from its rather transient half life in vitro, but it is possible that prolonged availability of the present bFGF pre-treatments may have been promoted by stabilization in the extracellular matrix,*’ from which it could have been released by the action of proteases. Second, a perhaps more plausible explanation is that bFGF accelerated the maturation of TH-positive neurones or facilitated early outgrowth from the transplant into the host striatum. The fact that the control transplants did eventually attain a similar level of recovery suggests that bFGF-pretreatment increased the rate at which early transplant maturation and/or graft-host interactions occurred. Nevertheless, bFGF did increase the number of TH-positive neurons surviving within transplants. This may reflect (i) an increased resistance to the trauma associated with the production of a cell suspension, (ii) a direct trophic effect of bFGF on the
Fig. 6. TH immunocytochemistry. Representative examples are shown at both low and high magnifications of transplants treated with bFGF or control treatments in Experiments I and II. (A,B) Control transplant of a single 3941 deposit of the embryonic nigral cell suspension from Experiment I. (CD) Transplant pretreated with bFGF from Experiment 1. (E,F) Control transplant of one of the two 39.41deposits of the embryonic nigral cell suspension from Experiment II. (G,H) Transplant which has received repeated infusions of bFGF from experiment II. All examples are based on animals surviving nine weeks following transplantation. Scale bars = 500 ,um (A-D), 100 grn (E-H).
I
ISC 56iZ--F
Fig. 6E-H.
survival of grafted dopmaine neurons, or (iii) an increased division of TH-positive neurons in the grafts.lY The fact that bFGF-pretreatment increased transplant volume to a similar extent suggests that any trophic or mitogenic influences are not restricted to the dopaminergic TH-positive neurons but apply to many if not all cell populat~o~~s within the grafts. Whereas this may reflect an increase in the number of glial ceils within transplants,* recent evidence points to the fact that bFGF does have a direct survival-promoting effect on neurons.?” The fact that this greater cell survival was not reflected by larger long-term functional effects may reflect the suggestion that the amphetamine rotation test is exquisitely sensitive to dopaminergic reinnervation, such that as few as 200-300 cells need survive to yield complete recovery by this measure3 Thus, all grafts in the present experiment, bFGF-treated and control alike, exceeded this level of cell survival and consequently any differences in the functional changes were masked by the ceiling level of recovery.
The resutts of Experiment II suggested that repeated intracerebral infusions of bFGF into the vicinity of transplants produced a greater increase in TH-positive neuron survival and more prolonged functional benefit than had been achieved with the single pretreatment. In this case the magnitude of the effects were sufficient to yield clearly signi~~nt benefits in terms of both cell survival and of functional recovery that became progressively greater at longer survival times, up to the nine-week tests, a full six weeks after the termination of the actual bFGF treatment. However, the presence of a marked inflammatory reaction in response to repeated intracerebral infusions of bFGF compIicates both the interpretation of the present results and their potential applications. At the practical level, the resulting cross-reactivity with the mouse IgG secondary antibodies restricted the range of primary antibody reactions available to characterize the grafts. More fundamentally, the observed reaction implies the induction of a variety of inflammatory mediators, which may exert neurotrophic actions in their own right. For example, interleukin-6 has been found to promote the survival of postnatal dopaminergic neurons in z&o,‘” and implants of interleukin-1 pellets can promote recovery from partial nigrostriatal lesions,3i Indeed, this process may in fact be accompanied by the induction of wound-derived neurotrophic activity,20 which may itself be attributable to an FGF-like molecule. Thus, the presence of an inflammatory response (produced in relation to the repeated trauma associated with the infusion process) may provide a neurotrophic influence in its own right or may complement the action of bFGF on the development of the transplants. Nevertheless, a similar degree of inflammatory
response was seen in the animals receiving vehicle infusions. Consequently, the larger number 01 surviving TH-positive neurons in the bFGF-treated grafts and the substantially greater functional recovery in these animals, was not simply attributable to inflammatory mediators, even if they may have contributed to the efficacy of the bFGF treatment.
Whereas some tissues (e.g. neocortex, striatun~) survive and grow well following transplantation, grafts of brainstem catecholamine neurons have generally been found to provide relatively low yields of surviving cells. The present results endorse the general approach of att~pting to enhance the viability of embryonic ventral mesencephalic transplants with potential neurotrophic molecules. Thus, Nishino and coileagues2? have found that tGS gangliosides can promote graft survival and recovery in transplanted hemiparkinsonian rats on similar motor tests to those employed here. Although there are still only a few studies addressing the problem of promoting dopamine neuron survival following transpbnthe tation, initial studies suggest that bFGF, insulin-like growth factors, brain-derived neurotrophic factor, platelet-derived growth factor, and interleukin-6 can all prove effective in promoting the viability of dopamine neurons in ~*itro.‘“‘.“’ ‘5.iq.11 So far, only the first of these has been investigated in any detail in oiw.5~27 The present data provide further evidence that bFGF can increase the survival of transplanted fetal dopaminergic neurons, and indicates that the mild benefit reported by Steinbusch and colleagues is not simply attributable to their use of heparin, which was not used in the present experiments.
CONCIAJSION
Pretreatment of a cell s~s~ns~on provides a moderate and relatively short-lived benefit on transplant eficacy, whereas repeated infusions of bFGF produce a long-term increase in cell survival and the functional efficacy of the transplants. However, repeated infusions, at least with the procedures adopted in the present experiment, produce an inflammato~ reaction which makes it difficult to attribute the functional effects to bFGF only, and complicates the use of similar treatments in any clinical application. Future work must be directed both to improving modes of growth factor administration, and to investigating the potential of other neurotrophic molecules, singly and in combination with members of the FGF family, to promote graft survival. ~c~nou,ledge~nts-ibis work was supported by the Medical Research Council.
Effects of bFGF on transplanted
dopaminergic neurons
391
REFERENCES
1. Abercrombie M. (1946) Estimation of the nuclear population from microtome sections. Anat. Rec. 94, 239-247. 2. Bohn M., Bohn M. C., Cupit L., Marciano F. and Gash D. M. (1987) Adrenal medulla grafts enhance recovery of striatal dopaminergic fibers. Science 237, 913-916. 3. Brundin P., Isacson 0. and Bjiirkhmd A. (1985) Monitoring cell viability in suspensions of embryonic CNS tissue and its use as a criterion for intracerebral graft survival. Brain Res. 331, 251-259. 4. Burgess W. and Macaig T. (1989) The heparin binding (fibroblast) growth factor family of proteins. A. Rev. Biochem. 58, 575-606. 5. Dolleman-Van Der Wee1 M. J., Nijssen A., De Vente J., Ramaekers F. C. S., Cordell B., Fuller F. and Steinbusch
H. W. M. (1991) Morphological and behavioural effects of basic fibroblast growth factor and heparin on transplanted fetal dopaminergic neurones and astrocytes in the denervated rat caudate-putamen. In Growth Factors in Alzheimer’s Disease (eds Hefti F., Brachet Ph., Will B. and Christen Y.), pp. 149-164. Springer, Berlin. 6. Dunnett S. B. (1991) Transplantation of embryonic dopamine neurons: what we know from rats. J. Neural. 238, 65-14. 7. Eckenstein F. P., Shipley G. D. and Nishi R. (1991) Acidic and basic fibroblast growth factors in the nervous system: distribution and differential alteration of levels after injury of central versus peripheral nerve. J. Neurosci. 11,
412419. 8. Eclancher F., Perraud F., Faltin J., Labourdette 9.
10. 11. 12. 13.
14. 15.
16.
G. and Sensenbrenner M. (1990) Reactive astrogliosis after basic fibroblast growth factor (bFGF) injection in injured neonatal rat brain. GIia’3, 5022509. Engele J. and Bohn M. C. (1991) The neurotrophic effects of fibroblast growth factors on dopaminergic neurons in vitro are mediated by mesencephalic glia. J. Neurosci. 11, 3070-3078. Ferrari G., Minozzi M. C., Toffano G., Leon A. and Skaper S. D. (1989) Basic fibroblast growth factor promotes the survival and development of mesencephalic neurones in culture. Deul Biol. 133, l&147. Gage F. H., Fisher L. J., Jinnah J. H. A., Rosenberg M. B., Tuszynski M. and Friedmann T. (1990) Grafting genetically modified cells to the brain: conceptual and technical issues. Prog. Brain Res, 82, l-10. Giacobini M. M. J., Hoffer B. J., Zerbe G. and Olson L. (1991) Acidic and basic fibroblast growth factors augment growth of fetal brain tissue grafts. Expl Brain Res. 86, 73-81. Hama T., Kushima Y., Miyamoto M., Kubota M., Takei N. and Hatanaka H. (1991) Interleukin-6 improves the survival of mesencephalic catecholaminergic and septal cholinergic neurones from postnatal, two-week-old rats in cultures. Neuroscience 40, 445-452. Hyman C., Hofer M., Barde Y. A., Juhasz M., Yancopoulos G. D., Squint0 S. P. and Lindsay R. M. (1991) BDNF is-a neurotrophic factor for dopaminergic neurones of the substantia nigra. Nature 350, 230-232. Knusel B.. Michel P. P.. Schwaber J. S. and Hefti F. (1990) Selective and non-selective stimulation of central cholinereic and dopaminergic development in vitro by nerve growth factor, basic fibroblast growth factor, epidermal growth factor, insulin and the insulin-like growth factors I and II. J. Neurosci. 10, 558-570. Lindvall 0. (1989) Transplantation into the human brain: present status and future possibilities. J. Neural. Neurosurg. Psychiat., Spec. Suppl. 39-54.
17. Lindvall
0.
(1991) Prospects
of transplantation
in human
neurodegenerative
diseases.
Trenak Neurosci.
14,
376-384.
18. Mayer E., Fawcett J. W. and Dunnett S. B. (1993) Basic FGF promotes the survival of embryonic ventral mesencephalic dopamine neurones. I. Effects in vitro. Neuroscience 56, 379-388. 19. Mayer E., Dunnett S. B. and Fawcett J. W. (1993) Mitogenic effect of basic fibroblast growth factor on embryonic ventral mesencephalic dopaminergic neurone precursors. Deul Brain Res. (in press). 20. Nieto-Sampedro M., Manthorpe M., Barbin G., Varon S. and Cotman C. W. (1983) Injury induced neuronotrophic activity in adult rat brain: correlation with survival of delayed implants in the wound cavity. J. Neurosci. 3, 2219-2229.
21. Nikkhah G., Odin P., Smits A., Tingstrijm A., Othberg A., Brundin P., Funa K., and Lindvall 0. (1993) Platelet-derived growth factor promotes survival of rat and human mesencephalic dopaminergic neurons in culture. Expl Brain Res. 92, 516523.
22. Nishino H., Hashitani T., Isobe Y., Furayama F., Sato H., Kumazaki M., Hirokomi K. and Awaya A. (1990) tGS ganglioside induces peculiar morphological features in grafted dopaminergic cells and promotes motor recovery in rats with unilateral lesions in the nigrostriatal dopamine pathway. Brain Res. 534, 73-82. 23. Olson L., Ayer-LeLievre C., Ebendal T., Eriksdotter-Nilsson M., Emfors P., Henschen A., Hoffer B., Giacobini M. B., Mouton P., Palmer M., Persson H., Sara V., Stromberg I. and Wetmore C. (1990) Grafts, growth factors, and grafts that make growth factors. Prog. Brain Res. 82, 5546. 24. Otto D. and Unsicker K. (1990) Basic FGF reverses chemical and morphological deficits in the nigrostriatal system of MPTP-treated mice. J. Neurosci. 10, 1912-1921. 25. Rifkin D. B. and Moscatelli D. (1989) Recent developments in the cell biology of basic fibroblast growth factor. J. Cell Biol. 109, Id. 26. Rohlf F. J. and Sokal R. R. (1981) Statistical Tables. Freeman and Co., New York. 27. Steinbursch H. W. M., Vermeulen R. J. and Tonnaer J. A. D. M. (1990) Basic fibroblast growth factor enhances the survival and sprouting of fetal dopaminergic cells implanted in the denervated rat caudate-putamen: preliminary observations. Prog. Brain Res. 82, 81-86. 28. Ungerstedt U. (1971) Striatal dopamine release after amphetamine or nerve degeneration revealed by rotational behaviour. Acta physiol. stand., Suppl. 367, 4968. 29. Ungerstedt U. and Arbuthnott G. W. (1970) Quantitative recording of rotational behaviour of rats after 6hydroxydopamine lesions of the nigrostriatal dopamine system. Brain Res. 24, 485493. 30. Unsicker K., Reichert-Preibsch H. and Wewetzer K. (1992) Stimulation of neuron survival by basic FGF and CNTF is a direct effect and not mediated by non-neuronal cells: evidence from single cell cultures. Devl Brain Res. 65,285-288. 31. Wang J., Plunkett R. J., Sheng J. G., Oldfield E. H. and Bankiewicz K. S. (1991) Recovery in hemiparkinsonian rats after intracaudal implantation of IL-1 pellets. Sot. Neurosci. Abstr. 17, 348. 32. Williams L. R., Vahlsing H. L., Lindamood T., Varon S., Gage F. H. and Manthorpe M. (1987) A small gauge cannula device for continuous infusion of exogenous agents into the brain. Expl Neural. 95, 734754.
39x
I’.. MAYER YI ul
33. Williams L. R., Varon S., Peterson G. M.. Wictorin K., Fischer W., Bjorklund A. and Gage F. H. (IY86) Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbria fornix transection. Prf~ nrun. Acad. Sci. U.S.A. 83, 9231-9235. 34. Winer B. J. (1971) Sfatistical Principles in E.rperimental Design, 2nd edn. McGraw--Hill. New York. (Accepted
IO May
1993)