Microencapsulated Dopamine (DA)-Induced Restitution of Function in 6-OHDA-Denervated Rat Striatum in vivo: Comparison Between Two Microsphere Excipients

Biodegradable controlled-release microsphere systems made with the biocompatible biodegradable polyester excipient poly [DL lactide-co-glycolide] constitute an exciting new technology for drug delivery to the central nervous system (CNS). The present study describes functional observations indicating that implantation of dopamine (DA) microspheres encapsulated within two different polymer excipients into denervated- striatal tissue assures a prolonged release of the transmitter in vivo. Moreover, in this regard, the results show that there were clear cut temporal differences in the effect of the two DA microsphere formulations compared in this study, probably reflecting variations in the actual composition (i.e., lactide to glycolide ratio) of the two copolymer excipients examined. This technology has considerable potential for basic research with possible clinical application.


INTRODUCTION
The main neuroehemieal characteristic of Parkinson's disease is a marked lesion of the nigro-striatal dopamine (DA) pathway. Current clinical management of Parkinson's disease relies strongly on substitution treatment with the DA prec.ursor L-DOPA and/or direct DA agonists, administered via the parenteral route. In recent years several attempts have been made to reverse neurologically debilitating nigrostriatal DA deficiency by different alternative means of replacement therapy /1L Although some advances have been reported with neuronal and adrenal medullary tissue transplants, this approach has not yet proven simple and effective to compensate for the underlying DA neuronal loss in humans with Parkinson's disease/9-11/. Intracerebroventrieular infusion is another method currently being tested in experimental Parkinsonism ]4/and in attempted replacement therapy of certain other neurodegenerative diseases /6/. Ideally, however, the delivery of substitution agents should be restricted specifically to the relevant damaged regions of the central nervous system (CNS).
Injectable drug delivery formulations socalled mierospheres have recently been developed, in which drugs are microencapsulated within bioeompatible and biodegradable copolymer exeipients like poly [DL-lactide-eoglyeolide] [3L These co-polymers are from the same class of material which has been used for nearly two decades in resorbable surgical sutures. The eopolymer microsphere formulation VOLUME 2, NO. 3-4, 1991  protects drugs from degradation and allows release of the drug from its excipient at a controlled rate for prolonged periods of time (for weeks, or even months)/3,20/. In view of its potential therapeutic usefulness, we have studied the mierosphere methodology as a means of delivering DA to the rat central nervous system (CNS)  Another interesting advantage of the mierospheres is that the duration of drug release can be modified by manipulating the "shell" constituent copolymer biodegradation kinetics, e.g., by changing the ratio of lactide to glycolide in the formulation /20/. To this end, DA was mieroeneapsulated in two copolymer excipients having different lactide to glycolide molar ratios. The object of this report is to compare these 2 different copolymer formulations. This is accomplished by comparing the duration of time that these intrastriatally implanted DA microspheres are able to reduce the extent of DA agonist-induced rotational behavior in unilaterally 6-OHDA-lesioned rats, used as a functional index of postsynaptic DA receptor denervation supersensitivity.

Animals, surgery and behavioral testing
Twenty five male Sprague-Dawley rats (200-250 g; AB Laboratorietjtinst, Sollentuna, Sweden) were unilaterally lesioned in the ascending median forebrain bundle (MFB) of monoamine neurons (coordinates A-P -4.3, L+ 1.4, D-V-8.7 from bregma, midline, and top of skull, respectively)/16/. The neurotoxin used for the purpose was 6-hydroxydopamine x HCI (6-OHDA; Sigma, St Louis, MO, USA), administered in a dose of 8 /xg/4 1 saline vehicle (containing 0.1% ascorbate). Two weeks after lesion, the rats were challenged with the classical DA agonist apomorphine (0.1 mg/kg s.c.) and rotational responses were monitored in a computerized rotometer set-up /13/. Only rats in which the DA denervation has been successful (_> 90% striatal DA depletion) will display strong contralateral rotation to apomorphine challenge in this model /8,21L Rats responding to apomorphine with less than 400 contralateral rotations per 60 rain within the first two weeks of testing were eliminated from the study (n 7).
Testing of the remaining 18 positive responders was then continued on a weekly basis until a stable level of rotation was established (max 10% difference between three consecutive sessions). This criterion was typically met about 6 weeks after the initial testing session (data not shown).
Polymers employed for encapsulation of DA DA was encapsulated in two types of copolymer excipient. One copolymer had a 50:50 molar ratio of lactide to glycolide (referred to as DA 50:50). The other copolymer had a 65:35 molar ratio of lactide to glyeolide (DA 65:35). It was predicted that, because of its higher lactide content, the 65:35 copolymer will take longer to biodegrade than the 50:50 eopolymer, thus potentially affording a longer duration of delivery of DA in vivo. The injections were performed with a 10 microliter glass capillary tube (calibrated at 3 microliter intervals) connected to a standard Hamilton syringe (50/xl) via polyethylene tubing.
The entire duration of injection was 3 min/site.
Upon completion of the second infusion, the injection cannula was left in situ for an additional 60 seconds before being slowly retracted. The skin wound was closed with surgical dips and the animals allowed to recover from anesthesia (about 30-45 minutes thereafter). They were placed into the rotometers and their behavior recorded for the subsequent 2-3 hours. Starting one week after implantation of the DA mierospheres, the rats were repeatedly tested for DA agonist-induced rotation on a weekly basis for 6-8 weeks.

Histology
At the termination of the 6-8 week schedule the rats were perfused with 5% glutaraldehyde and a series of consecutive sections (10/m) of the striatum was incubated with anti-DA antiserum/7,12/. The immunoeytoehernistry was performed with the standard avidin-biotin complex method, thus allowing visualization of tissue DA immunoreactivity.

Statistics
The data are presented as the total number of apomorphine-induced contralateral rotations summed over 60 minutes, means +_ SEM. The results were expressed as the mean and standard error of the mean (SEM). Probability levels of 5% or less were considered statistically significant.

RFULTS
Acute rotational response to DA microsphere implantation The 18 rats selected for microsphere implantation experiments had responded with the typical "two-peak" contralateral rotation pattern when initially challenged with apomorphine (0.1 mg/kg s.c.), a reliable indicator of > 90% DA lesion success/8,21L As previously reported /14/, upon recovery from ether anesthesia (= 30-45 min after intrastriatal implantation of DA microspheres) rats exhibited immediate contralateral rotational behavior with an amplitude comparable to that of apomorphine (0.1 mg/kg s.c.) but with longer duration.
Rats receiving DA 65:35 mierospheres displayed a somewhat more protracted response to the microsphere implantation but had a peak rotation amplitude similar to the DA 50:50 microsphere group /14/. Animals implanted with empty mierospheres did not display rotational behavior.
Rotational responses to apomorphine in microsphere-implanted rats DA DA 50:50 implants The total number of pre-implantation contralateral rotations for this group of rats was 553 +_. 58 (n 5). When compared to this value, the central striatal DA 50:50 microsphere implants resulted in a reduction to 397 +_ 19, a 29% decrease in the total number of apomorphineinduced contralateral rotations 4 weeks postimplantation. During the subsequent two weeks the response to apomorphine challenge slowly approached the pre-microsphere baseline ( Fig.   1). At 6 weeks there was a reduction to 497 _+ 48, representing a non-significant 11% decrease in the total number of apomorphine-induced contralateral rotations.

DA 65:35 implants
The total number of pre-implantation contralateral rotations for this group of rats was 506 +_. 26 (n 9). The DA 65:35 microspheres resulted in a reduction to 319 +/-43, representing a 37% VOLUME 2, NO. 3-4, 1991  implantation. From then on and up to 8 weeks following implantation there was an average reduction to 375 _ _ _ 50, corresponding to a 26.% decrease in the total number, of apomorphineinduced contralateral rotations (Fig. 2). Since, as indicated above, the DA 65:35 polymer excipient was predicted to outlast the DA 50:50, it is interesting to note that the reduction in the number of apomorphine rotations from week 3 through week 8 remained rather stable with the former formulation whereas the apomorphine response in the DA 50:50 group was no longer significantly suppressed after week 5.

Sham implants
During the 6 week testing period postimplantation,, rats that received empty microspheres did not display any significant change in the number of apomorphine-induced contralateral rotations when compared to the pre-implantation baseline (Fig. 3).

Immunocytochemistry
Micrographs of DA 50:50 are not presented in this report. However, previously we demonstrated that DA was present in DA 50:50 microspheres 3 weeks after implantation /14/.-Our immunocytochemical observation of these microspheres 6 weeks after implantation revealed that the majority of the microspheres had biodegraded or were in the process of doing so. There were, however, sparse traces of DA immunoreactivity in some microspheres that had not fully biodegraded. In no case was the immunocytochemical reaction comparable to that of DA 65:35. In the DA 65:35 implant group, however, there were still mierospheres containing respectable amounts of DA at 8 weeks (Fig. 4A-B). Examination of the micrographs in this latter group also reveals microspheres in the process of biodegrading at 8 weeks post-implantation (Fig. 4B) microsphere-implanted rats, thus equally upporting the specificity of the DA anffoody action and confirming the efficiency of the 6-OHDA lesion. Judging from routine microscopic. examination of the injection areas in DA 50:50, DA 65:35 and "sham"-treatcd animals, the implantation of microsphcrcs into the striatum did not appear to induce inflammatory reactions nor to result in damage to the surrounding tissue (cf. Fig. 4A  microsphere mixtures elicited strong contralateral rotation, considered to be mediated by postsynaptic DA receptor stimulation. However, this behavioral response vanished over a period of about 3-4 h, tentatively due to rapid receptor desensitization/13,14/. In the subsequent weekly test sessions the DA 50:50 microsphere group displayed a significantly reduced total number of apomorphine-induced contralateral rotations up to and including the 4th testing week postimplantation. However, in the DA 65:35 experiment the apomorphine response was reduced for at least 8 weeks. Even at this time the response to the DA agonist challenge was significantly lower than the pre-implantation baseline, suggesting that there was still enough DA remaining to maintain a reduction of the 6-OHDA-induced denervation supersensitivity. This interpretation is further supported by the immunocytochemical analysis, revealing the persistence of microspheres with intact or nearintact DA contents also at 8 weeks after implantation (el. Fig. 4A-B). It is also interesting to note in this context that the time-course and extent of reduction in denervafion supersensitivity produced by implantation of DA 65:35 microspheres is comparable with that reported after striatal grafting of adrenal medullary chromaffin tissue/19/.
Our results support the idea that the altered duration ofin vivo release of encapsulated drug from resorbable polymer excipients is a function of the actual copolymer composition. There is convincing evidence from in vitro experiments that the rate of drug release from microspheres is very sensitive to variations in the excipient constituents /2,20/. Interestingly, this seems to apply also in vivo, the major difference being that in this case the time scale for microsphere biodegradation is now weeks/months rather than minutes/hours /20/. In the present study, the implanted 65:35 DA microspheres resulted in a IMPLANTATION OF MICROENCAPSULATED DA 1"/1 Fig. 4: Photomicrographs of frontal cryostat sections (10/m) o1' the denervated stdatum after Implantation of DA 65:35 .B) or empty (C) microspheres. All the sections were incubated with anti-DA antiserum (1/500). The sections were processed for immunocytochemistry using the standard avidin-biotin-peroxidase complex method.
A DA 65:35 microcapules 8 weeks following implantation. Notice that the tissue on either side of the injection tract is intact. DA microspheres depicted by the immunocytochemical reaction are still within the striatum, x 160. B This photomicrograph is a higher magnification of A. A number of immunoreactive DA microspheres can be observed (arrows). There are some microcapules in the process of degrading (arrow heads), x 640. C Empty microspheres (arrows) 6 weeks after implantation. Notice that these microspheres are totally devoid o1 immunoreactivity, x 640. rather constant reduction in the apomorphine response from week 3 through week 8. Thus, it appears that even though the liberation of DA from these polymers relies on hydrolytic surface erosion in vivo, the rate of this process is dependent on the actual exeipient composition so as to allow reasonable control of the release in the manufacturing step by, e.g., altering the laetide to glycolide ratio in the copolymer. This concept is also supported by the results with the 50:50 DA microspheres. Importantly, the present data equally support the biocompatibility of the mierosphere formulations. It might be suggested that biodegradation through hydrolysis of the exeipients used would be harmful to the brain tissue. If so, it could be expected that the striatal tissue would be damaged during the process and thus the rats would no longer be able to respond to apomorphine after termination of the biodegradation period. Our results dearly demonstrate that this is not the case.
First, the immunocytoehemieal study revealed that introduction of the mierospheres produced minimal (if any) tissue damage (of. Fig. 4).
Secondly, the DA 50:50 animals returned to preimplantation baseline values, and thirdly, animals implanted with "sham" microspheres (without DA, but of the same shell composition) did not display any reduction in their apomorphine-induced rotational behavior over the entire 6-week period studied.
The present copolymer microsphere preparations appear to compare favorably with other reported polymeric excipient devices used in studies aiming for sustained release of DA. Other researchers have recently reported functional results similar to ours, using single, ram-thick ethylene-vinyl acetate (EV Ae) DAcontaining copolymer matrix rods or discs/5,22/. As compared to the microsphere preparations, these latter polymer designs inevitably cause much more mechanical tissue damage by virtue of their size when implanted and removed/22/; indeed, aspiration of parts of the overlying neoeortex was required in order to put the discshaped DA-entrapping polymer in position for release into the striatum/5/. Moreover, the EV Ae devices are non-resorbable and while, in one of the studies, only restricted gliosis was reported /5/, the chronic biocompatibility also remains to be established. Although the mierospheres used in our studies cannot be removed once implanted, they cause minimal tissue damage and no additional manipulation of the experimental animal is necessary to get the DA within reach of the striatal tissue.
As referred to above, the DA 50:50 microsphere implants into the central portion of the striatum resulted in a clear-cut, _> 4-week long, blunting of the response to apomorphine challenge. In contrast, in a previous study, we found no attenuation in the rotational response to apomorphine 3 40 days after lateral striatal DA 50:50 mierosphere implants /14/. The present results therefore represent another addition to the growing list of examples of functional heterogeneity of the striatum/15,17,18,23/.
Injectable mierosphere formulations have earlier been demonstrated as a feasible means of delivering drugs at intended sites of action, for prolonged periods of time, at required rates and in proper therapeutic doses, to targets outside the blood brain barrier/20/. The present results indicate that DA microsphere preparations have the potential of being employed as a source of transmitter replacement also within brain tissue in vivo. These formulations allow sustained diffusion of the microeneapsulated DA into the CNS at a controlled rate for predetermined periods of time, assure functional significance and at the same time appear to remain compatible with the host tissue. This new "slowrelease", target-directed approach provides not only a useful tool in basic neuroseienee but conceivably, after completing similar experiments in non-human primates, it may also be employed in the clinical management of neurodegenerative illness such as Parkinson's disease. to A.D., The Medical Faculty of the University of G6teborg and from the Swedish MRC (no. 7486 to S.H. and nos. 2207 and 8168 to A.D.) is gratefully acknowledged. We thank Mr. George McFarlane and Ms. Anne Kling-Pedersen for skillful technical assistance.