Biodegradable Microparticles with an Entrapped Branched Octameric Peptideas a Controlled-Release HIV-1 Vaccine

MANMOHAN SINGH*‡X, J. P. MCGEE§, XUAN-MAO LI*, WAYNE KOFF*, TIM ZAMB*, C. Y. WANG*, AND D. T. O’HAGAN*‡

Received April 29, 1997, from the *United Biomedical, Inc., 25 Davids Drive, Hauppauge, NY 11788, and §Core Technologies Ltd., HannahResearch Institute, Kirkhill, Ayr KA6 5HL, U.K. Accepted for publication August 3, 1997X. ‡Present address: Chiron Corporation,4560 Horton Street, Emeryville, CA 94608.

Abstract 0 Polyactide-co-glycolide microparticles, with an entrappedbranched octameric peptide from human immunodeficiency virus (HIV-1), were prepared by a solvent evaporation method. The microparticleswere characterized for size distribution, antigen loading level, and integrity.Mice in one group were each immunized with a single dose of a controlled-release microparticle formulation containing 300 µg of peptide and theserum IgG responses to the antigen were compared with those of micefrom a second group that were immunized at 0, 4, and 26 weeks with100-µg doses of the same peptide immunogen adsorbed to alum. Thecontrolled-release microparticles induced an antibody response comparableto that from the alum-immunized group. The subcutaneous and theintramuscular routes of administration were compared in additional groupsof mice for the microparticles, and both routes induced similar responses.A suspending vehicle for the microparticles was also evaluated and didnot affect the immunogenicity of the controlled-release formulationcontaining both small and large microparticles, although the immunoge-nicity of smaller microparticles immunized alone was affected.

Introduction

The number of reported cases of human immunodeficiencyvirus (HIV) infection by the turn of the century is expected tototal ∼40 million, according to the World Health Organization(WHO).1 Hence, the development of an effective HIV vaccineis an important goal towards which many groups are working.However, this goal is a formidable task because of thevariability of HIV, the various transmission routes of thevirus, and a lack of understanding of the possible mechanismsof immune protection.2,3 Several studies recently performedwith the HIV envelope glycoprotein (gp120) have demon-strated that neutralizing antibodies directed to the principalneutralizing determinant (PND) in the V3 region of gp120 canprotect chimpanzees from virus challenge.4-6 Although neu-tralizing antibodies may be responsible for elimination of cell-free virus particles, cell-mediated immune responses, par-ticularly cytotoxic T lymphocytes (CTL), are responsible forclearance of virus-infected cells and are likely to be animportant component of protective immunity.7

Controlled-release microparticles prepared from polyactide-co-glycolide (PLG) have been extensively evaluated in recentyears,8,9 with the objective of extending the duration of drugor antigen release. Controlled release would allow a singleinjection of the microparticles to suffice for long-term therapyand would obviate the need for frequent injections. The choiceof PLG as a polymer for these formulations is based on itslong-term safety in humans, its biodegradability, and thecommercial availability of a variety of polymers of differentmolecular weights and monomer ratios.10-15 Controlled-release vaccines comprising PLGmicroparticles have recentlybeen shown to be effective for the induction of long-term

immune responses following a single dose.16-25 These formu-lations offer the potential advantage of inducing protectiveimmunity following a single injection. The development of acontrolled-release vaccine for HIV would increase patientcompliance, reduce the number of drop-outs, and would reducethe overall vaccine immunization costs by providing savingson delivery, storage, and number of administrations.22-25

These issues are crucial for the development of a HIV vaccinethat could be used throughout the world, including thedeveloping world.Vaccine studies reported to date involving controlled-release

microparticles have been concerned exclusively with proteinsand not with peptide-based vaccines. For example, earlierreports evaluated the effect of the adjuvant QS21 entrappedin PLG microparticles, along with recombinant gp120.27-30

Although QS21 have been used in humans in a few Phase 1trials, it has yet to establish itself as a nontoxic adjuvant formass immunization. The immunogen used in the currentstudies consisted of a sequence of 30 amino acids from the V3loop of gp120, representing the PND of HIV-1, which wassynthesized as a branched octameric peptide.31,32 No ad-ditional adjuvants were included in the microparticle formula-tions. The octameric peptide has previously been shown tobe highly immunogenic and to induce significant titers ofneutralizing antibodies in an animal model.31 Moreover, thisimmunogen has undergone Phase I clinical evaluation inhuman volunteers and has been shown to induce neutralizingantibodies and T cell responses.33The objective of the current study was to evaluate the

immunogenicity of the branched octameric peptide (p200M)as a controlled-release formulation in mice. Additional pep-tides have already been synthesized to represent the majorclasses of HIV-1 present worldwide. These peptides will becombined in future studies as peptide “cocktail” to attempt todevelop a HIV-1 vaccine that could be used globally.

Materials and MethodsMaterialssThe polyactide-co-glycolide (PLG) polymers were ob-

tained from Boehringer Ingelheim (Montvale, NJ). The polymers usedin this study were RG505 (PLG 50/50; MW, 65 KDa), RG858 (PLG85/15; MW, 108 KDa), R208 (PLG 100/0; MW, 138 KDa). Thebranched octameric peptide (p200M), representing the gp120 V3sequence of HIV-1MN (amino acid residues 295-325) was synthesizedin the Protein Chemistry Department at United Biomedical by themethod described by Wang et al.32 and was attached to the heptalysylcore via modification of the Merrifield solid-phase synthesis. The corebears eight reactive amino termini onto which peptide chains aresimultaneously attached. The V3 MN branched octameric peptidewas purified to >95% homogeneity by HPLC fractionation to yield asingle peak. All other reagents were obtained from Sigma Chemicals(St. Louis, MO) and were used as shipped. Aluminum hydroxide gel(Alhydrogel) was purchased from Superfos, Denmark.Peptide CharacterizationsPeptide concentration was deter-

mined by bicinchoninic assay (BCA) and confirmed by amino acidanalysis (AAA). The molecular weight was estimated by sodiumdodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) withstandard pre-stained markers. A 10% gel was used for the electro-X Abstract published in Advance ACS Abstracts, September 15, 1997.

© 1997, American Chemical Society and S0022-3549(97)00174-3 CCC: $14.00 Journal of Pharmaceutical Sciences / 1229American Pharmaceutical Association Vol. 86, No. 11, November 1997

phoresis under standard operating conditions. These parameterswere also used later for evaluating the antigen integrity and bydetermining the extent of aggregation of the peptide after microen-capsulation.Preparation of MicroparticlessMicroparticles were prepared

by a previously reported solvent evaporation technique.34,35 Briefly,the larger microparticles with a 2% (w/w) loading level were preparedby diluting the antigen solution to 2 mL and emulsifying with 10 mLof a 10% (w/v) polymer solution in methylene chloride at high speedwith a Silverson homogenizer. The primary emulsion was then addedto 50 mL of distilled water containing polyvinyl alcohol (10%, w/v).This procedure resulted in the formation of a w/o/w emulsion thatwas stirred at 1000 rpm for 12 h at room temperature, and themethylene chloride was allowed to evaporate. The resulting micro-particles were filtered, washed twice in distilled water, and dried ina desiccator.Smaller sized microparticles were prepared in a similar manner,

but the polymer concentration and stirring speeds were varied toobtain uniform microparticles of <10 µm. The preparation methodsfor both small and large microparticles had earlier been standardizedusing ovalbumin (OVA).35

Particle Size and Surface MorphologysThe size distributionof the microparticles was determined with a particle size analyzer(Malvern Instruments, U.K.). The surface morphology was deter-mined by scanning electron microscopy (35 JEOL SEM) with a 100°gold-palladium coating.Antigen Loading LevelssThe loading level of the antigen within

the microparticles was determined by dissolving 20 mg of themicroparticles in 2 mL of 5% SDS-0.1 M sodium hydroxide solutionat room temperature. The amount of immunogen was determinedby BCA protein assay.26

In Vitro Release AssaysSelected batches of the microparticleswere evaluated for in vitro release. Briefly, 10-mg microparticles wereweighed in polypropylene tubes and suspended in 1 mL of PBS. Alltubes were placed on a end-over-end shaker set at 37 °C. At eachtime point, three tubes were withdrawn, and the supernatant wasassayed for amount of antigen released. The mean of three valueswas calculated. The pH of the release buffer was monitored for 4weeks and varied between 6.0 and 7.4.Assessment of Antigen IntegritysThe antigen integrity follow-

ing microencapsulation was evaluated by SDS-PAGE and immuno-blotting. The encapsulated immunogen was released from the PLGmicroparticles over a period of 3 weeks (partial release of total antigenload) at 37 °C in PBS (50 mM, pH 7.4) and concentrated by Amiconultrafiltration before being evaluated by Western blot. Hyperimmunesera to p200M raised in Balb/C mice was used to bind to theimmunogen. Immunogen samples were analyzed before microencap-sulation and after release from the microparticles with a 12% gel inthe Mini-Protean system from Bio-Rad. The standard immunoblotassay kit (Bio-Rad; 170-6463) was used. The blots were visualizedby 4-chloro-1-naphthol. The analysis revealed that the antigen waslargely unaltered by the microencapsulation process as previouslyreported.32

Preparation of the Suspending VehiclesThe suspendingvehicle was prepared by dissolving 0.2% Tween 20, 0.1% carboxymethyl cellulose, and 1% sorbitol in distilled water and passing theresultant solution through a 0.22-µm filter. This vehicle was formu-lated to aid microparticle wetting and dispersion and to allow doseuniformity during administration.DosesThe dose of P200M was 100 µg at 0, 4, and 26 week time

points on alum, and 300 µg at day 0 in PLG microparticles.AnimalssFemale CD1 mice (6-8 weeks of age; 20-25 g) were

used for the study and were maintained in standard housing with anormal diet at HRP Inc. (Denver, PA). Each group consisted of eightanimals in all of the studies.Evaluation of the Immunogenicity of Controlled-Release

Microparticles in MicesThe objective of this study was to evaluatethe immunogenicity of p200M entrapped in microparticles and tocompare the responses obtained with those obtained with three dosesof p200M on alum at 0, 1, and 6 months. The controlled-releaseformulation of p200M consisted of a combination of microparticlesprepared from three different polymers (PLG5 0/50, 85/15, and 100/0). The PLG 50/50 microparticles were prepared as small micropar-ticles (<10 µm) to provide the initial boosting of the immune response,and the PLG85/15 and the PLG100/0 microparticles were preparedwith a larger size (>10 µm) to provide later booster responses. This

approach to the development of controlled-release vaccines based onmicroparticles has previously been shown to be successful with proteinimmunogens.40A group of mice was immunized with p200M microparticles

suspended in normal saline by the intramuscular route. The injection(100 µL) was administered at one site in the hind legs. A second groupwas immunized with p200M on alum (100 µg) in the same injectionvolume at 0, 4, and 26 weeks.Effect of Route of Administration on the Immunogenicity

of MicroparticlessThe objective of this study was to compareimmunization in mice with microparticles by two different adminis-tration routes; that is, the subcutaneous (sc) and the intramuscular(im) routes. PLG 50/50 microparticles (<10 µm) containing 300 µgof p200M were administered to groups of mice by the sc and im routesin the hind legs.Effect of a Suspending Vehicle on the Immunogenicity of

MicroparticlessThe objective of this study was to determine if theimmunogenicity of microparticles was adversely affected by thedispersion of the formulation in a suspending vehicle prior toadministration. Two groups of mice were immunized with PLG50/50 containing 300 µg of p200M suspended either in normal saline orin the suspending vehicle.The controlled-release microparticle formulation used in the initial

study (combination of three polymers, 300 µg of p200M), eithersuspended in normal saline or in the suspending vehicle, was alsoevaluated in two groups of mice. The antibody responses in both ofthese studies were evaluated for 28 weeks.Estimation of SerumAntibodies by ELISAsThe ELISA for the

serum antibodies was performed essentially as previously de-scribed,31,32 with monomeric peptides adsorbed to the ELISA plates(1 µg/mL), followed by incubation with the sera from the immunizedanimals. The levels of IgG antibodies in the sera was quantified withhorse-radish-peroxidase-conjugated goat antiserum to mice igG. Thedata are expressed as antibody titers, which are the log 10 of thereciprocal of the serum dilution that gave an absorbance reading of0.5 at 492 nm.

ResultsBoth the small and large p200M microparticles had a good

loading efficiency of the immunogen ranging from 78 to 94%.The mean size for the small microparticles was ∼500 nm andthat for the large microparticles was ∼28 µm. The SEManalysis of the microparticle batches exhibited smooth sur-faces without any cracks or craters (results not shown). TheWestern blots of the entrapped antigen demonstrated antigenintegrity during the microencapsulation process. The cumu-lative in vitro antigen release profiles from PLG50/50 (smallmicroparticles), PLG85/15, and PLG100/0 (large micropar-ticles) are shown in Figure 1.In the initial study, the antibody response generated by a

single injection of the controlled-release microparticles wascomparable to those obtained with three injections with thepeptide on alum (Figure 2). At 4 weeks, the antibody titerfrom the microparticles was higher than the alum-immunized

Figure 1sCumulative in vitro release profile of p200M from three differentmicroparticles of two sizes. Key: (0) PLG50/50 (<10 µm); (O) PLG85/15 (>10µm); (4) PLG100/0 (>10 µm).

1230 / Journal of Pharmaceutical SciencesVol. 86, No. 11, November 1997

group. But, after the boost dose, the antibody response in thealum group increased and was higher than the microparticlesuntil week 20. The response in the alum group once againincreased after the second boost at week 26, but then declined.In contrast, the response to the microparticles was maintainedfor at least 48 weeks.The second study indicated that the sc and im routes of

immunization were comparable for the induction of immuneresponses to a microencapsulated peptide (Figure 3). Thisresult is an important observation because it indicates thatthe sc route can be used to evaluate the performance ofcontrolled-release vaccines in small animal models. It is oftenvery difficult to immunize by the im route in small animals,and the sc route is preferred for ease of administration, despitethe fact that most vaccine products are administered by theim route.The third study (Figure 4) showed that single polymer (PLG

50/50) microparticles (<10 µm) in normal saline gave a betterantibody response than the same microparticles in a suspend-ing vehicle. However, a mixture of the controlled-releasemicroparticles prepared from the three different polymers(containing both <10-µm and >10-µmmicroparticles) induced

comparable immune responses in saline or the suspendingvehicle (Figure 5).

DiscussionReproducible batches of both small and large microparticles

containing entrapped p200M were prepared by a solventevaporation method. The mean size for the smaller micro-particles was ∼500 nm, the ideal size for phagocytosis byantigen presenting cells.11 The larger microparticles had anaverage size of ∼28 µm, which would make them too large tobe phagocytosed. Therefore, these microparticles were idealfor the controlled release of antigens over prolonged periods.The Western blots of p200M before and after microencapsu-lation were comparable (results not shown; previously re-ported32), suggesting that the antigen was largely unalteredby the process of microencapsulation. This result is animportant finding especially because it has been reported thatseveral antigens are partially denatured or altered uponexposure to moisture and organic solvents and lose theirantigenicity.36,37The in vitro release profiles show that the antigen release

is faster for the PLG 50/50 microparticles than PLG 85/15 or

Figure 2sAntibody responses in CD1 mice following immunization with p200Min controlled-release microparticles as (0) a single injection at day 0 or (9) onalum at 0-, 4-, and 26-week intervals. The titers are in log10 scale representingmean ± SEM.

Figure 3sAntibody responses in CD1 mice following immunization with p200Min PLG 50/50 microparticles as a single injection by the (0) sc and (9) im routes.The titers are in log10 scale representing mean ± SEM.

Figure 4sAntibody responses in CD1 mice following immunization with p200Min PLG 50/50 microparticles (<10 µm) as a single injection in (0) normal salineor in (9) the suspending vehicle. The titers are in log10 scale representing mean± SEM.

Figure 5sAntibody responses in CD1 mice following immunization with p200Min controlled-release microparticles (three polymers, small and large microparticles)as a single injection in (0) normal saline or in (9) a suspending vehicle. Thetiters are in log10 scale representing mean ± SEM.

Journal of Pharmaceutical Sciences / 1231Vol. 86, No. 11, November 1997

100/0 microparticles. This result is consistent with earlierresults with different antigens where PLG 50/50 polymerdemonstrated a faster release over a 4-week period thanpolymers with higher lactide monomer ratios.35 The controlled-release microparticle formulation prepared containing en-trapped p200M immunogen demonstrated a sustained anti-body response with a single injection of the immunogen inmicroparticles. An antibody response to the HIV immunogenwith a single microparticle injection was observed for 48 weeksfollowing a single injection. Moreover, the antibody responsewas comparable to the same dose of peptide administered onthree occasions at 0, 4, and 26 weeks. This observationsuggests that the entrapped antigen is released slowly overseveral months and that a single-dose microparticle vaccinemay replace the multiple-dose vaccine schedule used in initialclinical trials.32 Similar results have been previously reportedwith a subunit HIV-1 gp120 vaccine.27-30 However, thepreviously reported microparticle formulations also containedan entrapped adjuvant (QS21) in the microparticles. Theinclusion of adjuvants in controlled-release microparticles isundesirable for a number of reasons, including cost, difficultyof manufacture, and the possibility of sustained exposure toa potentially toxic agent. The results from the current studiesindicate that PLG microparticles are capable of inducingsustained antibody titers in mice in the absence of additionaladjuvants. Furthermore, the PLG microparticles also engen-der a Th1-type response similar to alum,41,42 and thus provideconstant stimulation of the immune system for long-termprotection. This result is probably a consequence of theinclusion of small microparticles (<10 µm) for an adjuvanteffect20,21 and also the optimal selection of particle size andpolymer for controlled-release characteristics.The antibody response with PLG 50/50 microparticles with

entrapped p200M was similar for both sc and im administra-tion. This similarity is important because it indicates thatinitial formulations may be evaluated by the sc route in smallanimal models and later switched to the more commonly usedim route in larger animal models. The im route is difficult touse in small animal models, particularly for microparticleformulations, which often need to be dispersed in an ap-preciable volume of buffer for easy administration.The PLGmicroparticle products that are currently available

on the market as drug delivery systems use a suspendingvehicle for optimum dispersion and wetting of the micropar-ticles. However, the majority of vaccine-related studies withPLG microparticles to date have involved the administrationof microparticles in saline.20-25 There was a concern that theimmunogenicity of the microparticle formulations might beadversely affected by an administration vehicle, which wouldbe needed in a final formulation of a controlled-release vaccinefor administration to humans. Therefore, the immunogenicityof two microparticle formulations were evaluated followingdispersion of the particles in an appropriate suspendingvehicle. The suspending vehicle inhibited the immunogenicityof the single polymer (PLG 50/50) small microparticles (<10µm) in comparison with administration in saline. Thisdifference may be a consequence of a component of the vehicle(e.g., Tween 20) adsorbing to the surface of the microparticlesand inhibiting their uptake by antigen-presenting cells.Alternatively, the high viscosity of the suspending vehicle mayhave prevented the efficient uptake of the microparticles.However, encouragingly, the immunogenicity of the combinedcontrolled-release formulation, containing both small andlarge microparticles, was not affected by the administrationvehicle, indicating that the immunogenicity of the combinedmicroparticle formulation is less dependent on the initialuptake of the microparticles by antigen-presenting cells.Hence, in future studies involving controlled-release micro-particles, the formulation may be administered in an ap-

propriate suspending agent without impairing the immuno-genicity of the formulation.We have demonstrated that a branched octameric peptide

immunogen fromHIV-1 can be entrapped in controlled-releasemicroparticles and converted into a single-shot vaccine. Mi-croparticles prepared by a similar process, but with a differentantigen, have recently undergone detailed toxicological evalu-ation and appear to be safe for human use.38,39 In a recentreport, we have shown that combining a portion of the totaldose of an antigen on alum and entrapping the remainder inmicroparticles gives a better initial antibody response withmicroparticles alone.40 Hence, the current formulation forp200M may be improved by the addition of alum to themicroparticles. However, splitting the two components of theformulation, alum and microparticles, at two injection siteson one animal does not improve the immune response anyfurther.40

References and Notes1. Center for Disease Control. MMWR 1991, 40, 357-369.2. Koff, W. C.; Hoth, D. F. Science 1988, 241, 426-431.3. Theoretical Biology and Biophysics; Myers, G.; Korber, B.;

Berzofsky, J. A.; Smith, R. F.; Pavlakis, G. N., Eds.; 1992.4. Berman, P. W.; Gregory, T. J.; Riddle, L.; Nakamura, G. R.;

Champe, M. A.; Porter, J. P.; Wurm, F. M.; Herschberg, R. D.;Cobb, E. K.; Eichberg, J. W. Nature 1990, 345, 622-625.

5. Girad, M.; Kieny, M. P.; Pinter, A.; Barre-Sinoussi, F.; Nara,P.; Kolbe, H.; Kusumi, K.; Chaput, A.; Reinhart, T.; Muchmore,E. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 542-546.

6. Emini, E. A.; Schlief, W. A.; Nunberg, J. H.; Conley, A. J.; Eda,Y.; Tokiyoshi, S.; Putney, S. D.; Matsushita, S.; Cobb, K. E.; Jett,C. M. Nature 1992, 355, 728-730.

7. Zinkernagel, R.; Haenseler, E.; Leist, T.; Cerny, A.; Hentgartner,H.; Althage, A. J. Exp. Med. 1986, 164, 1075-1092.

8. Furr, B. J. A. Royal Society of Medicine International Congressand Symposium Series 1987, 125, 1-5.

9. Burns, R. A.; Vitale, J.; Sanders, L. M. J. Microencapsulation1990, 7, 397-413.

10. Cutwright, D. E.; Perez, E.; Beasley, J. D.; Larson, W. G.; Posey,W. R. J. Oral Surg. 1974, 37, 132-142.

11. Tabata, Y.; Ikada, Y. Biomaterials 1988, 9, 356-362.12. Wise, D. L.; Fellman, T. D.; Sanderson, J. E.; Wentworth, R. L.

In Drug Carriers in Medicine, Gregoriadis, G., Ed.; Academic:London, 1979; p 237.

13. Hutchinson, F. G.; Furr, B. J. A. In Delivery Systems for PeptideDrugs; S. S. Davis, L. Illum and E. Tomlinson, Eds.; Plenum:New York, 1986; pp 115-124.

14. Yamaguchi, K.; Anderson, J. M. J. Controlled Release 1993, 24,81.

15. Pitt, C. G.; Gratzl, M. M.; Kummel, G. L.; Surles, J.; Schindler,A. Biomaterials 1981, 2, 215.

16. Aguado, M. T. Vaccine 1983, 11, 596.17. Aguado, M. T.; Lambert, P. H. Immunobiology 1992, 184, 113-

125.18. Alonso, M. J.; Gupta, R. K.; Caroline, M.; Siber, G. R.; Langer,

R. Vaccine 1994, 12, 299.19. O’Hagan, D. T. Adv. Drug. Del. Rev. 1990, 5, 265-285.20. O’Hagan, D. T.; Rahman, D.; McGee, J. P.; Jeffery, H.; Davies,

M. C.; Williams, P.; Davis, S. S. Immunology 1991, 73, 239.21. Eldridge, J. H.; Stass, J. K.; Meulbroek, J. A.; McGhee, J. R.;

Tice, T. R.; Gilley, R. Mol. Immunol. 1991, 28, 287.22. Singh, M.; Singh, O.; Talwar, G. P. Pharm. Res. 1995, 12, 1796.23. Men, Y.; Thomasin, C.; Merkle, H. P.; Gander, B.; Corradin, G.

Vaccine 1995, 13, 683-689.24. Singh, M.; Singh, O.; Singh, A.; Talwar, G. P. Pharm. Res. 1991,

7, 958-961.25. Singh, M.; Singh, O.; Singh, A.; Talwar, G. P. Int. J. Pharm.

1992, 85, R5-R8.26. Sharif, S.; O’Hagan, D. T. Int. J. Pharm. 1995, 115, 259.27. Cleland, J. L.; Powell, M. F.; Lim, A.; Barron, L.; Berman, P.

W.; Eastman, D. J.; Nunberg, J. H.; Wrin, T.; Vennari, J. C.AIDS Res. Human Retroviruses 1994, 10, 821-826.

28. Cleland, J. L.; Barron, L.; Berman, P. W.; Daugherty, A.;Gregory, T. J.; Lim, A.; Vennari, J. C.; Wrin, T.; Powell, M. F.J. Pharm. Sci. 1996, 85, 1346-1349.

29. Cleland, J. L.; Barron, L.; Daugherty, A.; Eastman, D.; Kensil,C.; Lim, A.; Weissburg, R. P.; Wrin, T.; Vennari, J. C.; Powell,M. F. J. Pharm. Sci. 1996, 85, 1350-1354.

1232 / Journal of Pharmaceutical SciencesVol. 86, No. 11, November 1997

30. Cleland, J. L.; Lim, A.; Barron, L.; Duenas, E. T.; Powell, M. F.J. Controlled Release 1997, 47, 135-150.

31. Wang, C. Y.; Looney, D. J.; Li, M. L.; Walfield, A. M.; Ye, J.;Hosein, B.; Tam, J. P.; Wong-Staal, F. Science 1991, 254, 285-288.

32. O’Hagan, D. T.; McGee, J. P.; Boyle, R.; Gumaer, D.; Li, X. M.;Potts, B.; Wang, C. Y.; Koff, W. C. J. Controlled Release 1995,36, 75-84.

33. Gorse, G. J.; Keefer, M. C.; Belshe, R. B.; Mathews, T. J.; Forrest,B. D.; Hsieh, R. H.; Koff, W. C.; Hanson, C. V.; Dolin, R.;Weinhold, K. J.; Frey, S. E.; Ketter, N.; Fast, P. E. J. Infect.Dis. 1996, 173, 330-339.

34. Jeffery, H.; Davis, S. S.; O’Hagan, D. T. Pharm. Res. 1993, 10,362.

35. McGee, J. P.; Singh, M.; Li, X. M.; Qiu, H.; O’Hagan, D. T. J.Microencapsulation 1997, 14, 197-210.

36. Schwendeman, S. P.; Costantino, H. P.; Gupta, R. K.; Siber, G.R.; Kibanov, A. M.; Langer, R. Proc.Natl. Acad. Sci.U.S.A. 1995,92, 11234.

37. Schwendeman, S. P.; Costantino, H. P.; Gupta, R. K.; Siber, G.R.; Kibanov, A. M.; Langer, R. Proc. Int. Symp. ControlledRelease Bioact. Mater. 1995, 22, 41-42.

38. Chandrasekaran, R.; Giri, D. K.; Chaudhury, M. R.Human Exp.Toxicol. 1996, 15, 349-351.

39. Chaudhury, M. R.; Sharma, K.; Giri, D. K. Human Exp. Toxicol.1996, 15, 205-207.

40. Singh, M.; Li, X. M.; McGee, J. P.; Zamb, T.; Koff, W.; Wang, C.Y.; O’Hagan, D. T. Vaccine 1997, 15, 475-481.

41. Moore, A.; McGuirk, P.; Adams, S.; Jones, W. C.; McGee, J. P.;O’Hagan, D. T.; Mills, K. H. G. Vaccine 1995, 13, 1741-1749.

42. Vordermeier, H. M.; Coombes, A. G. A.; Jenkins, P.; McGee, J.P.; O’Hagan, D. T.; Davis, S. S.; Singh, M. Vaccine 1995, 13,1576-1582.

AcknowledgmentsWe acknowledge the help rendered by Joanne DeSteffano, Ritu

Singh, and Hanying Wang in the animal experiments. We also thankthe Peptide Chemistry Lab at United Biomedical, Inc., for providingsufficient amounts of p200M for the experiments.

JS970174N

Journal of Pharmaceutical Sciences / 1233Vol. 86, No. 11, November 1997