articles a chlamydial major outer membrane protein extract as a

Articles

A Chlamydial Major Outer Membrane Protein Extract as aTrachoma Vaccine Candidate

Mauro Campos,* Sukumar Pal,^ Terrence P. O'Brien,% Hugh R. Taylor,%Robert A. Prendergast,% and Judith A. Whittum-HudsonX

Purpose. As shown in infected humans and in animal models of chlamydial infection, themajor outer membrane protein (MOMP) of Chlamydia trachomatis is immunogenically potent.The purpose of this investigation was to test in the cynomolgus monkey model of trachomaa new extract of MOMP as a candidate vaccine against ocular chlamydial infection.

Method. The nonionic detergent octyl-/?-D glucopyranoside (OGP) was used to extract MOMPfrom purified C. trachomatis (serovar C) elementary bodies. Protective immunization withOGP-MOMP by mucosal and systemic routes was compared in the cynomolgus monkey modelof trachoma. All control and immunized monkeys were challenged by topical application ofinfectious C. trachomatis to the conjunctivae 35 days after the initiation of immunization.

Results. Immunization with OGP-extracted MOMP successfully induced chlamydia-specificlocal and systemic immunity to MOMP and to whole organism before challenge and earlyclearance of infection by systemically immunized monkeys. Although ocular disease was notsignificandy reduced in either immunized group compared to control animals, the lowestclinical and microbiologic disease scores developed in two animals in the mucosal group withthe highest immunoglobulin A tear antibody titers at days 0 to 14, whereas higher tear andserum immunoglobulin G correlated with reduced disease' in the systemically immunizedgroup.

Conclusions. These data demonstrate that despite evidence of vigorous MOMP-specific andother chlamydia-specific serologic and cell-mediated immunity, as well as anamnestic serologicresponses to chlamydia, vaccination with OGP-MOMP was only partially protective againstchlamydial ocular disease. The partial protection correlated best with tear immunoglobulinA responses after mucosal immunization and with local and systemic immunoglobulin Gresponses after peripheral immunization, suggesting that alternative chlamydial antigens mayhave to be considered in future vaccine development to induce more effective protectiveimmunity and that evaluation of efficacy must be appropriate to route of immunization. InvestOphthalmol Vis Sci. 1995;36:1477-1491.

From the * Conselho Brasileiro de Oftalmologia, Sao Paulo, Brazil; the ^Departmentof Pathology, University of California at Irvine, Irvine, California; %The WilmerOphthalmological Institute, Johns Hopkins University School of Medicine, Baltimore,Maryland; and the ^Department of Ophthalmology, University of Melbourne, TheRoyal Victorian Eye and Ear Hospital, East Melbourne, Victoria, Australia.Presented in part as a poster at the annual meeting of the Association for Researchin Vision and Ophthalmology, Sarasota, Florida, May 1992.Supported in part by grants from the National Institutes of Health/National EyeInstitute (EY03324) and The Edna McConnell Clark Foundation, New York, NewYork (JWH). SP was supported by a postdoctoral fellowship from Fight for Sight,Inc., Research Division of Prevent Blindness America.Submitted for publication July 28, 1994; revised January 19, 1995; acceptedJanuary 27, 1995.Proprietary interest category: N.Reprint requests: Judith A. Whittum-Hudson, The Wilmer Institute, Johns HopkinsHospital, 600 N. Wolfe Street, Baltimore, MD 21287-9142.

X rachoma is one of the most common infectiousdiseases in humans, among the most prevalent of ocu-lar infections, and the leading cause of infectiousblindness.1"5 Chlamydial genital tract infection, whichis often accompanied by conjunctivitis, is three timesmore common than gonorrhea,67 and Chlamydia tra-chomatis is the leading cause of tubal infertility andpelvic inflammatory disease.6'8'9 A successful antichla-mydial vaccine not only would prevent these sequelaeof infection, it also would substantially reduce healthcare costs accompanying these infections.

Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8Copyright © Association for Research in Vision and Ophthalmology 1477

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933190/ on 02/14/2018

1478 Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8

The chlamydial major outer membrane protein(MOMP) has been favored as an antichlamydial vac-cine candidate because vigorous anti-MOMP re-sponses are seen in naturally infected patients andanimal models.10"12 The major outer membrane pro-tein is a 394-amino acid residue protein with four vari-able domains that have been immunologically identi-fied.13"16 Monoclonal antibodies that recognize sur-face-exposed epitopes of the MOMP also neutralizeelementary body (EB) infectivity if EB are exposed toantibodies before challenge.1718 Several methods havebeen used to obtain purified whole MOMP or individ-ual MOMP variable domains for vaccine testing: high-pressure liquid chromatography purification of so-dium dodecyl sulfate-extracted proteins19 and recom-binant MOMP subunits obtained either from expres-sion vectors2021 or as fusion protein products13 (alsoWhittum-Hudson et al, manuscript in preparation).Sodium dodecyl sulfate-extracted MOMP was poorlyimmunogenic in monkeys, possibly because importantconformational epitopes were destroyed during ex-traction.19 Recombinant MOMP subunits of one ormore variable domains were only weakly immuno-genic,20'21 even when these were conjugated to choleratoxin.20

Recent studies by Batteiger et al showed that anMOMP preparation, obtained from C. psittaci by se-quential extraction with Sarkosyl22 and the nonionicdetergent octyl-/3-D-glucopyranoside (OGP), inducedserologic and cell-mediated immune responses againstwhole organisms and was partially protective in theguinea pig model of genital infection.23'24 Similar re-sults were seen with an OGP-extracted MOMP (OGP-MOMP) vaccine against the sheep abortion strain ofC. psittaci,25 for which protection was measured by anincrease in live births. These successful results havebeen attributed to preservation of more of the nativethree-dimensional structure of the MOMP moleculeduring extraction with the nonionic OGP detergent.22

Our studies were undertaken to test the ability ofOGP-MOMP prepared from a human biovar of C.trachomatis (serovar C) to induce a protective immuneresponse (s) in the cynomolgus monkey model of tra-choma. Two routes of immunization were tested, mu-cosal (oral and topical ocular) and systemic becauseC. trachomatis infects mucosal tissues and inductionof mucosal immunity is thought to be important forprotective immunity11'26 and because MOMP has beenshown to be highly immunogenic when delivered sys-temically.18'24

MATERIALS AND METHODS

AnimalsTen wild-caught young adult cynomolgus monkeys(Macaca fascicularis) were used in these studies

(Charles River Primates, Port Washington, NY). Allprocedures were performed on deeply anesthetizedmonkeys (ketamine hydrochloride), and those proce-dures adhered to National Institutes of Health guide-lines and the ARVO Statement for the Use of Animalsin Ophthalmic and Vision Research.

Protein Purification

Elementary bodies were purified from C. trachomatisserovar C/TW-3 by density gradient centrifugation inPercoll (Pharmacia, Uppsala, Sweden), as previouslydescribed.2728The major outer membrane protein waspurified from EB by extraction using the nonionicdetergent, octyl-/? D-glucopyranoside (OGP; SigmaChemical, St. Louis, MO) and dithiothreitol (DTT;Sigma) according to the method of Batteiger et al.24

For each extraction, 10 ml of EB (~10 mg total pro-tein) in HEPES saline buffer (pH 7.4) were suspendedin an equal volume of 4% Sarkosyl (Sigma).22 Thesolution was incubated for 30 minutes at 37°C, soni-cated briefly, and centrifuged at 100,000g for 1 hourat 18°C. The pellet was dissolved in 20 ml of 2% Sarko-syl and 40 mM DTT in HEPES buffer, incubated 30minutes at 37°C, and centrifuged again at 100,000gfor 1 hour at 18°C. The pellet was resuspended in 2.5ml of 2% OGP-40 mM DTT. After 1 hour at 37°C,the solution was centrifuged at 100,000g"for 1 hour.The supernatant contained primarily MOMP and wasdesignated OGP-MOMP (Fig. 1). Yields of outermembrane proteins comprised predominantly of theMOMP were approximately 2% of the total startingproteins. Because MOMP was lost during each extrac-tion step, five separate OGP extraction procedures (80mg of total starting protein) were required to immu-nize eight monkeys with a total of 360 fj,g of chlamydialprotein over 5 weeks. Fractions of OGP-MOMP fromthe individual extractions were pooled before immu-nization, concentrated on a Speed Vac (Savant Instru-ments, Farmingdale, NY) without heat, and dialyzedto remove residual detergent.

Immunization

Monkeys were immunized either mucosally or systemi-cally over a period of 5 weeks using the dosage proto-cols shown in Table 1. The mucosally immunizedgroup of four monkeys (animal numbers 641, 767,353, and 409) received oral and topical ocular applica-tions of OGP-MOMP mixed with cholera toxin(CTox) (List Biological Laboratories, Campbell, CA)as adjuvant at days —35 and —21. A final ocular boostof 20 fig OGP-MOMP alone was administered 14 daysbefore challenge. The systemically immunized groupof four monkeys (animals 907, 635, 600, and 646)received OGP-MOMP with complete Freund's adju-vant intramuscularly (100 /xg of each) at —35 days and


Test of OGP-MOMP as a Chlamydial Vaccine

M W 1 2 3 4 5 6

1479

M W 1 2 3 4 5 6io 3

97.4 -

66.2 -

45.0 -

31.0 -

MOMP

97.4 -

66.2 -

45.0 -

31.0 - .

! < • • * A

21.5 -

14.4 -

A «

21.5 -

14.4 —

B

nGURE l. Analysis of the octyl-/?-d glucopyranoside-extractea major outer memorane pro-tein (OGP-MOMP) vaccine preparations by sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE) and by immunoblotting. (A) SDS-PAGE profile demonstra-ting enrichment for the MOMP during a typical sequential purification; (1) solubilizedhomologous C serovar elementary body (parent sample); (2) Sarkosyl total; (3) Sarkosyl-soluble fraction (supernatant); (4) Sarkosyl + dithiothreitol total; (5) Sarkosyl + dithiothrei-tol supernatant; and (6) OGP supernatant (OGP-MOMP). Molecular mass markers are atleft. Gel was stained with Coomassie blue. (B) Enrichment for MOMP during the sequentialextraction procedure as visualized by immunoblotting from a duplicate of the gel in A. Seelegend to A for lane designations. The positions of MOMP and low molecular weight markersare indicated. A high-titer polyclonal anti-chlamydial antibody was used to detect the pres-ence of MOMP in each fraction.

additional boosts of 100 ng each of OGP-MOMP inincomplete Freund's adjuvant at days —21 and —14.Two control monkeys (animals 78 and 657) receivedOGP-buffer mixed with CTox or Freund's adjuvant attimes and by routes in parallel with the respective testgroups. The last immunizations were 14 days beforeinfectious challenge. Additional historic control mon-keys receiving only adjuvants and infectious ocularchallenges with C serovar were included in analysesof clinical and microbiologic disease.29'30 These lattercontrol animals were graded clinically by one of thesame ophthalmologists involved in the current experi-ments.

Infectious Ocular Challenge

All monkeys were challenged topically in each eye withinfectious C. trachomatis serovar C (TW-3) EB on day0, as previously described.29'30 Approximately 5 X 103

inclusion-forming units (IFU) in 20 [A of sucrose-

phosphate-glutamic acid buffer were applied topi-cally to each conjunctival sac. Inoculum concentrationwas verified by retttration on McCoy cells using stan-dard methods.

Clinical Examination and Specimen Collection

The clinical response of each eye was graded ac-cording to degree of conjunctival inflammation andfollicular response, as described previously. l9l30~32 Allmonkeys were examined weekly for 6 weeks and thenbiweekly in random order without examiner knowl-edge of group allocation. In addition, photodocumen-tation of each examination time for each monkey wasread independently by two readers; discrepancies be-tween grading scores were resolved by a third com-bined reading.3*'33 The clinical response was gradedon a scale of 0 to 3 for each of 10 signs of conjunctivalinflammation to obtain a total inflammatory score foreach monkey.293033 A total clinical disease score



TABLE l. Immunization Protocol for Groups of Cynomolgus Monkeys Immunized During a5-Week Period With OGP-Extracted MOMP and Adjuvants

Day ofExperiment

-35

-21

-14

0

Route(n = 4)

OCPOIMOCPOIMOCPOIM

Mucosal(n = 2)

OGP-MOMP + CToxOGP-MOMP + CTox

0OGP-MOMP + CToxOGP-MOMP + CTox

0OGP-MOMP

00

Ocular challenge with 5000 IFU of C serovar,

Immunizations

Systemic(n = 4)

00

OGP-MOMP + CFA00

OGP-MOMP + IFA00

OGP-MOMP + IFA

Chlamydia trachomatis

Control(n = 2)

OGP buffer + CToxOGP buffer + CToxOGP buffer + CFAOGP buffer + CToxOGP buffer + CToxOGP buffer + IFAOGP buffer

0OGP buffer + IFA

OC = ocular, 20 pg, 1:1 ratio; PO = oral, 100 /xg, 1:1 ratio; IM = intramuscular, 100 fig, 1:1 ratio; OGP = octylglucopyranoside; MOMP= major outer membrane protein; CTox = cholera toxin; CFA = complete Freund's adjuvant; IFA = incomplete Freund's adjuvant.Controls received equal volumes of adjuvants or buffers.

(TCDS) was obtained by adding the scores of all eyes,and the mean of these scores was calculated for eachtreatment group for each time point.30 Conjunctivalswabs were taken after clinical examination at eachtime point for chlamydial culture and for direct fluo-rescent antibody (DFA) cytology.34 Serum and tearswere collected as previously reported.19

Proliferative Assay of Lymphocyte Function

Peripheral blood mononuclear cells, purified fromheparinized blood as previously described,27 weretested for chlamydia-specific lymphoproliferative re-sponses. Peripheral blood mononuclear cells (105

cells/100 /A) were cultured in triplicate, round-bot-tom microwells with 100 fA of homologous C serovarEB (C-EB; 1 X 104-l X 106IFU/ml), concanavalin A(1 /ug/well), or pokeweed mitogen (1 //g/well). Con-trol wells received only cells and medium. After 72hours of culture, cells were pulsed with 1 /uCi of 3H-thymidine and then harvested 16 to 20 hours lateronto glass fiber filters for scintillation counting. Thestimulation index was determined by the standard for-mula (experimental cpm) + (medium control cpm).A stimulation index of 4 or more was considered posi-tive for monkey peripheral blood mononuclear cells(PBMC).27

Serology

Enzyme-linked immunoadsorbent assays (ELISA)were performed as previously described.27 Plates werecoated overnight at 4°C with freshly thawed aliquotsof EB (106 IFU/well) diluted in phosphate-bufferedsaline (PBS; pH 7.4). Tears were eluted from cellulosesponges (Edward Week, Research Triangle Park, NC)

in 0.22 ml of PBS containing 1% bovine serum albu-min (PBS-BSA) to yield an initial dilution of approxi-mately 1:10. Serum or tears were diluted serially (1:20to 1:2560) in the plates, and plates were incubatedfor 90 minutes at 37°C. After washes with PBS-0.5%Tween, alkaline phosphatase (AP)-conjugated antihu-man immunoglobulin G (IgG; 1:1000) or antihumanimmunoglobulin A (IgA; 1:500) (Kirkegaard andPerry, Gaithersburg, MD) diluted in PBS-BSA-0.05%Tween was added to each well, and plates were incu-bated for an additional 60 minutes at 37°C. The reac-tion was developed with paranitrophenylphosphate indiethanolamine buffer. Control wells contained no an-tigen but received all other reagents. When CTox wasused as the ELISA test antigen, it was diluted to deliver1 /zg/well in 0.05 M carbonate-bicarbonate buffer(pH 9.6). Known positive and negative serum or tearsamples for each antigen were included in each assay.The pretreatment samples from each monkey wereused to determine the negative cutoff OD405 for eachgroup of longitudinal specimens. All serum or tearsamples from an individual monkey were assayed onthe same plate. The entire plate was read on a Biotek(Winooski, VT) ELISA reader at OD405 and was ana-lyzed using software provided by Dr. A. Donnenberg(University of Pittsburgh, Pittsburgh, PA).35 Resultsare expressed as endpoint titers.

Sodium Dodecyl Sulfate-Polyacrylamide GelElectrophoresis and Immunoblotting

Chlamydial proteins obtained from viable C-EB wereseparated under reducing conditions by sodium dode-cyl sulfate-polyacrylamide gel electrophoresis on12.5% preparative gels using standard meth-


Test of OGP-MOMP as a Chlamydial Vaccine 1481

ods.l9l22>23'36'37 Biotinylated low molecular weight stan-dards (BioRad, Richmond, CA) were included in eachgel. After electrophoresis, polypeptides were electro-phoretically transferred to nitrocellulose paper (NCP)(Schleicher & Schuell, Keene, NH) using an LKBMultiphor II Transfer System (Pharmacia, Piscataway,NJ). Unoccupied binding sites on the NCP wereblocked with BLOTTO (5% dry milk in 150 mMNaCl-10 mM Tris, 0.5% Tween) at room temperaturefor 1 hour.38 The NCP was then cut into 2.5-mm widestrips, and these were incubated overnight at 4°C withtears (1:50) or serum (1:100). All dilutions were madewith BLOTTO. The next day, strips were washed andincubated with alkaline phosphatase-conjugated anti-human IgG (1:500) or IgA (1:250) for 60 minutes atroom temperature; control strips containing biotinyl-ated molecular weight protein standards were incu-bated with alkaline phosphatase-conjugated avidin(1:1000) (BioRad). After washing, strips were devel-oped with nitroblue tetrazolium and 5-bromo-4 chlo-rol-3-indolylphosphate (Sigma) in 100 mM Tris-HClwith 100 mM NaCl and 5 mM MgCl2 (pH 9.5). Thereaction was stopped by adding 20 mM Tris-HCl with5 mM ethylenediamine tetraacetic acid (pH 8.0) ,39 Wealso tested the effects of immunization and ocularchallenge on serologic responses to two different ex-tracts of chlamydial proteins obtained from solubi-lized elementary bodies—a Triton X-100 extract fromC. trachomatis serovar C prepared as described pre-viously40'41 to yield primarily the 57-kd heat shock pro-tein (hsp60) and a Sarkosyl extract of C. trachomatisserovar C to yield the cysteine-rich 60 kd (omp2) pro-tein in the Sarkosyl-insoluble fraction.22 Approxi-mately 100 /Ltg of chlamydial proteins obtained fromthese two extracts were resolved on separate 12.5%preparative gels and transferred to NCP membranesas described above for whole EB. Control strips wereincubated with an anti-hsp60 monoclonal antibody(anti-GroEL, a gift of Dr. P. Bavoil, University of Roch-ester, Rochester, NY),42 or an anti-omp2 monoclonalantibody (a gift of W. J. Newhall V, Centers for DiseaseControl, Atlanta) to confirm the identity of the anti-gens. Duplicate serum samples were incubated withblots containing either hsp60 or omp2. Protein bandson all NCP strips were quantitated by densitometry(Hoeffer Scientific Instruments, San Francisco, CA)and analyzed using software purchased from Hoeffer.Densitometric results are expressed as mean peakheights ± SE in arbitrary units. Protein bands wereidentified by comparison with molecular weight stan-dards run on the same gel. Twenty test strips wereobtained per blotted minigel, and all samples from anindividual monkey were tested on NCP strips preparedfrom the same gel.

Statistical Analyses

Clinical disease scores and microbiologic results foreach group were compared by nonparametric Mann-Whitney rank sum tests. Unpaired Student's /-testswere used to compare the results of proliferativeassays.

RESULTS

Biochemical Analyses of the Vaccine Candidate

The purity of OGP-MOMP was analyzed by compar-ing samples from each extraction step on a Coomassieblue-stained sodium dodecyl sulfate-polyacrylamidegel electrophoresis gel (Fig. 1A) and by immunoblot-ting with an antichlamydial antibody (Fig. IB). Resultsfor a typical sequential purification of MOMP areshown in Figure 1A. Lane 6 shows the final purifiedMOMP (OGP-MOMP) used to vaccinate the mon-keys. Although other proteins were seen in this finalfraction, densitometric analyses of the immunoblotprepared from the same fractions shown in Figure IBdemonstrate that significant solubilization of EB wasachieved with OGP. Approximately 30% to 50% of theMOMP in the starting whole EB solution was recov-ered while many of the chlamydial proteins were re-moved.

Chlamydia-Specific Antibody Production AfterImmunization

Sera and tears collected longitudinally during theOGP-MOMP immunization period and after infec-tious challenge were tested by ELISA against intact EBfor the presence of chlamydia-specific IgA (tears) andIgG (serum and tears) antibodies (Fig. 2). High titersof antichlamydial serum IgG developed in 7 of 8OGP-MOMP immunized monkeys before infectiousocular challenge on day 0 (1:100 - > 1:5000) (Fig.2A). Maximum serum IgG responses were reached 2weeks after infectious challenge, which is consistentwith a secondary response to infectious challenge.These high titers persisted throughout the course ofthe experiment. In the control group, positive re-sponses were first seen at day 21, and they remainedlower than those in the immunized groups throughday 84. By day 98, the control monkey's serum IgGtiters were no longer significandy different from thosein the mucosally immunized group (Fig. 2A). Anti-chlamydial antibody titers of systemically immunizedmonkeys at day 0 were equal to the maximum titer ofthe control group at day 84. The pattern of tear IgGresponses was similar to that observed for sera, exceptthat titers were lower at day 0 (1:36 to 1:855). Antichla-mydial tear IgG increased rapidly in both immunizedgroups after ocular challenge, reached peak levels at


Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8

o-o

DAYS AFTER CHALLENGE

6000 -i

A

0 MUCOSAL

O . SYSTEMIC

A CONTROL

B-56 -42 -28 -14 0 14 28 42 56 70 84 98


1400 -

•oca

Q.

Q>

• — '

onLJ—h-

flBO

DY

1200 -

1000 -

800 -

600 -

400 -

200 -

£ MUCOSAL

O SYSTEMIC

A CONTROL

FIGURE 2. Anti-chlamydialantibody responses in se-rum and tears after octyl-/?-D glucopyranoside-majorouter membrane proteinimmunization. Values rep-resent group mean titer ±SEM for each time point asdetermined in an enzyme-linked immunoadsorbentassay with whole homolo-gous C serovar elementarybodies. One control mon-key died 35 days after chal-lenge. (A) Anti-chlamydialserum immunoglobulin G;(B) antichlamydial tear im-muoglobulin G; and (C)antichlamydial tear immu-noglobulin A.

-56 -42 -28 -14 14 42 56 70 84 98




days 14 to 28, and then gradually decreased. Controlgroup tear IgG was not detected until 4 weeks afterocular challenge and increased through day 98, whichis typical of a primary infection30 (Fig. 2B). Antichla-mydial tear IgA responses were lower than IgG butwere positive (> 1:100) at day 0 in 3 of 8 immunizedmonkeys (one mucosally immunized and two systemi-cally immunized monkeys). All eight immunized mon-keys had peak IgA titers by day 14 and remained posi-tive through day 98 (Fig. 2C). Control monkeys werenegative for chlamydia-specific tear IgA until afterday 28.

To verify that the mucosal immunization protocolwas adequate, antibody responses in serum and tearsto the CTox used as mucosal adjuvant were also deter-mined by ELISA for several time points. All recipientsof CTox showed increasing titers of CTox-specific se-rum antibody after CTox-immunization. On day —7,after a primary and boosting immunization, anti-CToxserum IgG titers reached their maximum (> 1:5000)and remained high at least through day 28 (later timeswere not tested). Tear anti-CTox IgG responses alsoincreased during the immunization period and werehighly positive by day 0 (1:500 to 1:2750) in all CToxrecipients (data not shown). Tear IgA responses toCTox were lower, and only 2 of 6 monkeys showedpositive anti-CTox IgA titers at day 0 (data not shown).However, anti-CTox tear IgA was detected 2 weeksafter infection in 5 of 6 CTox-immunized monkeys,indicating successful mucosal priming by this adjuvantand demonstrating that no monkeys were nonre-sponders to exogenous antigens. Similar local anti-CTox B cell responses after oral priming, ocular prim-ing, or both have been reported.39'43

Immunoblot Analysis of the Fine AntigenicSpecificities of Humoral Responses

Before infectious ocular challenge, ELISA and lym-phoproliferative responses to whole EB were positivein both immunized groups. These results differ fromprevious studies with purified MOMP or MOMP sub-unit vaccines in which such responses were not ob-served, and they suggest that at least some of the nativeconformation of MOMP (and other surface proteins)was preserved during OGP extraction. Immunoblot-ting was used to compare the extent to which monkeyswere specifically immunized against MOMP beforeand after challenge because ELISA titers would reflecta cumulative response to all surface molecules. Alleight OGP-MOMP immunized monkeys had detect-able anti-MOMP IgG responses in tears before infec-tious challenge, and 7 of 8 monkeys had anti-MOMPserum IgG. Serum IgG responses increased after chal-lenge, suggesting an anamnestic response to infec-tious challenge. Only 1 of 4 mucosally primed mon-

keys had weakly detectable anti-MOMP tear IgA at day0, whereas 2 of 4 systemically immunized monkeyswere immunoblot positive for IgA before challenge.The one mucosally immunized monkey that never ex-hibited anti-MOMP tear antibody by immunoblottingwas positive by ELISA against whole organism on day0 and thereafter. Control monkeys exhibited no de-tectable anti-MOMP responses before challenge, andnone had anti-MOMP IgA in tears before day 28 afterchallenge as measured by either blotting or ELISA.Serum and tear responses to other chlamydial pro-teins also were detected in several animals at day 0,presumably as a result of additional membrane pro-teins contained in the OGP extract (Fig. 3).

Serum antibody responses to the two 60-kd chla-mydial proteins appeared to develop differently in thetwo immunized groups: The mean serum IgG re-sponses against the chlamydial hsp60 were highest inthe mucosally immunized group at all days tested(days —35, 0, 28, and 84); weak anti-hsp60 responsesdeveloped in systemically immunized animals (Fig.4A). In each group, only two monkeys exhibited posi-tive anti-hsp60 responses at any of the times tested.Responses to the cysteine-rich 60-kd protein (omp2)were weakly positive in the systemically immunizedgroup at days 0, 28, and 84 after ocular challenge,and the mean responses by this group did not changesignificantly over the period of the experiment. Incontrast, the anti-omp2 responses in the mucosallyimmunized group were negative at day 0 but increasedbetween days 28 and 84, by which time 3 of 4 mucosallyimmunized animals exhibited positive responses (Fig.4B). Individual clinical disease scores correlated morewith the antibodies to omp2 than those to hsp60 atday 84 (data not shown). Control animals had no de-tectable responses against either the hsp60 or omp2proteins at any of the times tested and probably relatesto receipt of only a single antigenic challenge by infec-tion (Fig. 4).

Cell-Mediated Immunity After Immunization

Peripheral blood mononuclear cells were tested inproliferative assays to determine whether chlamydia-specific cell-mediated immunity to whole organismwas induced by immunization. Proliferative responsesof both OGP-MOMP immunized groups to severaltenfold dilutions of EB were similar regardless of im-munization protocol, and 5 of 8 immunized monkeysexhibited vigorous chlamydia-specific responses 14days before infectious challenge. Mucosally primedmonkeys exhibited a mean SI of 10, whereas PBMCcells from systemically primed monkeys responded toan optimal concentration of 106 IFU/ml with a meanSI of 7 (Fig. 5). These values were not significantlydifferent from each other, though both were signifi-


1484

66 -

45 -

31 -

22 •

14

66 -

45 -

31 -

22 -

14 -

Investigative Ophthalmology & Visual Science, July 1995, Vol. 36, No. 8

A A B B C

66 -

MOMP 45 -

31 -

2 2 -

14 -

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

A A B B C

MOMP

i: V.

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

A A B B C

MOMP

66 -

45 -

3 1 -

2 2 -

1 4 -

L 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

cantly higher than responses of control monkeys byStudent's (-tests (P < 0.05). Proliferative responseswere apparently short lived because they returned tobaseline levels by the day of challenge. Later, after theboosting effect of ocular infection, some animals ineach vaccine group again exhibited positive cell-medi-ated immunity. Cell-mediated immunity for monkeys907, 635, and 600 was positive more frequently afterinfectious challenge; in the mucosal group, positivecell-mediated responses by PBMC at day 0 and at 2

66 -

45 -

31 -

2 2 -

14 -

M O M P

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

A A B B C

MOMP

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

A A B B C

L

I M O M P

6 6 -

4 5 -

3 1 -

2 2 -

14 -

I " 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

weeks after challenge correlated with reduced diseasefor one monkey (monkey 767). No animals in themucosal group exhibited positive PBMC proliferationto whole organism after this time, although high fre-quencies of chlamydia-specific T cells were detectedin conjunctivae and draining lymph nodes of onemonkey-vaccine group each at 5 and 9 weeks afterchallenge. As expected, high frequencies of chla-mydia-specific conjunctival lymphocytes were presentat 5 weeks in more diseased conjunctivae (monkey



FIGURE 3. Immunoblot analysis of antichlamydial antibody in serum and tears from octyl-/?-D glucopyranoside-major outer membrane protein (OGP-MOMP)-immunized and controlmonkeys. Elementary bodies were solubilized in Laemli buffer and separated by sodiumdodecyl sulfate-polyacrylamide gel electrophoresis for transfer to nitrocellulose. Boundimmunoglobulin G and immunoglobulin A were detected with alkaline phosphatase-conju-gated antihuman reagents. (A,B) Antichlamydial serum immunoglobulin G; (C,D) antichla-mydial tear immunoglobulin G; and (E,F) antichlamydial tear immunoglobulin A. Resultsfor each group are shown for the three antibody assays. In each figure, lanes labeled Arepresent monkeys from the mucosal group; lanes labeled B represent monkeys from thesystemic group; and lanes labeled C represent control monkeys. Blots for the same fivemonkeys are aligned in Figures 3A, 3C, and 3E (animal numbers: 641, 767, 967, 635, and78); and the second five in Figures 3B, 3D, and 3F (animal numbers: 353, 409, 646, 600,and 657). Lanes 1 to 4 correspond, respectively, to preimmune (day —42), day 0, day +7,and day +28 of the experiment, where day 0 was day of infectious ocular challenge. Thelow molecular weight markers are indicated on the left margin. MOMP is indicated by anarrow.

641; 1.1 chlamydia-specific cells per 100 cells), com-pared to a systemically immunized and partially pro-tected animal (monkey 635; 1.7 chlamydia-specificcells per 1000). The frequencies of chlamydia-specificcells in conjunctiva had fallen by 40- to 50-fold 4 weekslater when an additional monkey in each group wastested. Because of limited numbers of control mon-keys, conjunctivae were not collected from unimmu-nized animals in this experiment, but results for lymphnode cells were similar to both vaccine groups at 5and 9 weeks (40/106 cells to 50/106 cells; not shown).These data are consistent with our previous report onchlamydia-specific T cells in blood and conjunctiva.44

Taken together, the results demonstrate that longer-lived cell-mediated immunity was induced by systemicimmunization with OGP-MOMP. Cell-mediated im-munity may correlate more with clearance of organismthan with preventing initial infection and could ex-plain why systemically immunized animals became cul-ture negative earlier than other groups. This is consis-tent with cell-mediated immunity to other intracellu-lar pathogens.

Clinical Responses to Immunization

Clinical disease scores and microbiologic effects ofimmunization were determined for the three groupsof monkeys after infectious ocular challenge. Totalclinical disease score was calculated for each animal,and a mean TCDS was determined for each group asshown in Figure 6. Conjunctival inflammation, includ-ing follicles, developed in mucosally immunized andmock mucosally primed control groups after the appli-cation of cholera toxin and OGP-containing samplesat boosting immunization on day —14. This responseresolved largely in the 14 days before infectious ocular

challenge. However, residual conjunctival sensitivitymight have influenced heightened clinical diseaseseen in the mucosally primed group after infectiouschallenge, similar to persistent sensitivity seen afterthe development of contact lens hypersensitivity.

After live chlamydial challenge, the mean TCDSof the two immunized groups did not differ signifi-cantly from that of the control group. The disease inthe mucosally immunized group was marginally worsethan that of systemically immunized animals duringthe first two weeks after challenge. One monkey fromthe control group died 35 days after challenge, andclinical disease worsened in the surviving control mon-key during the last 3 weeks of the experiment. Thisincreased inflammation was coincident with a culture-confirmed secondary bacterial infection with S. epider-midis.45 Control animal data shown in Figure 6 include15 additional monkeys challenged with 5000 IFU ofC. trachomatis (C-TW/3) from numerous experimentsduring the last 4 years.

Microbiologic Effects of Immunization

The frequency of positive chlamydial isolation culturesfrom the conjunctiva did not differ significantly forthe two immunized groups, but the frequency of infec-tious chlamydia was lower after systemic immunizationthan in the other two groups. Systemic immunizationwith OGP-MOMP clearly reduced infection beforeday 21 (Fig. 7A). The surviving control monkey hadpeak positive culture results at 56 days, several weeksafter both immunized groups had declining positiveculture isolations. Because the infection in 15 historiccontrol monkeys peaked by culture assay between 28and 35 days after challenge, we think this late positiveresult is accounted for by complications of secondaryconjunctival infection.45 Direct fluorescent antibody



100 n

80 -

60 -

20 -

Q MucosalO SystemicA Controls

-28 0 28 56


100 -i

0 MucosalO SystemicA Controls

40 -

20 -

B -28 0 28 56


84

FIGURE 4. Limited development of anti-hsp60 and anti-omp2immunoglobulin G antibodies in serum. Values representthe mean peak height (±SEM) units obtained by densito-metric analyses of immunoblots that which contained either(A) chlamydial hsp604041 or (B) omp2.22 Serum collectedbefore (day —35 and 0) or after (days +28 and +84) ocularchallenge were tested for reactivity to the two 60-kd chlamy-dial proteins.

results were similar to culture results, except at day 7after challenge when systemically immunized monkeyshad lower 15-field EB counts than the other groups(Fig. 7B). However, in comparison to historic controls,no statistically significant reduction in DFA scores wasseen, and mucosal immunization appeared to resultin higher DFA scores (Fig. 7C). Thus, systemic immu-nization induced an earlier culture-negative, but nota DFA-negative, state and may reflect nonviable organ-isms detectable by.DFA.

When individual animals in each immunizationgroup were examined for correlations between lowerclinical and microbiologic disease scores and higherantibody responses, two animals in each group showedpartial protective immunity: animals 767 and 409 in

the mucosally immunized group and systemically im-munized animals 907 and 600. Monkeys 767 and 907consistently exhibited less disease by any criterion andhad the highest antibody responses for their respectivegroups. Associations of other animals in each groupvaried, except that all animals in., the systemic grouphad negative culture results 4 weeks earlier than thosein the .mucosal group. Of possible significance for fu-ture vaccine strategy, individual tear IgA titers and day0 blots (rather than serum or tear IgG) correlated withpartial protection for monkey 767 in the mucosalgroup; this animal was the one of two animals in thisgroup that exhibited positive proliferative responses(stimulation indices > 4) at day 14 after challenge. Incontrast, tear and serum IgG titers and blots (ratherthan tear IgA) correlated with partial protection formonkey 907 in die systemic group; proliferative re-sponses for monkey 907 returned to baseline by day 0.

DISCUSSION

Failure of previous attempts to vaccinate against chla-mydial infections has usually been attributed to theinability to induce an adequate antichlamydial im-mune response to wholeorganismbetore challenge. Ourdata demonstrate that vaccination with an OGP ex-tract of MOMP induced strong humoral and cell-medi-ated immune responses to intact or solubilized organ-ism before infectious challenge. These immune re-sponses did not significandy protect either group

C

E

in

20

15

10

•

oA•

MucosalSystemicControlTotal d . -35

- 4 2 - 3 5 - 2 8 - 2 1 - 1 4 - 7 0

Days before Ocular ChallengeFIGURE 5. Cell-mediated immunity after OGP-MOMP im-munization as measured by lymphoproliferative responsesto whole homologous C serovar elementary bodies. Eachpoint represents the mean stimulation index (±SEM) fortriplicate samples from each monkey per group. Responsesby both immunized groups were significantly higher thanthose of controls on day —7 (P < 0.05). Responses to theoptimal elementary body concentration of 106 inclusion-forming units/ml are shown.



• MUCOSAL

O SYSTEMIC

A CONTROL


FIGURE 6. Total clinical disease scores (TCDS) for OGP-MOMP-immunized and control monkeys. The mean TCDSis shown for each group. Closed circles = mucosally immu-nized (n = 4); open circles — systemically immunized (n =4). The TCDS for two concurrent control monkeys plus 15historical controls receiving the same dosage of TW/3 (5000inclusion-forming units) are included (n = 17, open trian-gles). Differences between groups were not statistically sig-nificant.

against infection or clinical disease after ocular chal-lenge with infectious chlamydia, although they wereassociated with a modest reduction in the recovery ofviable C. trachomatis in cultures after challenge and,on an individual basis, with reduced clinical disease.

Vaccine candidates tested in the cynomolgus mon-key model of trachoma before the present studies in-cluded oral immunization with viable or killed C. tra-chomatis whole EB of various serovars,29"31 recombi-nant chlamydial lipopolysaccharide,46 a sodiumdodecyl sulfate extract of MOMP,19 MOMP subunits,20

and chlamydial hsp70.47 Other tests of whole chla-mydia as protective immunogens also generally failedto protect against disease.48"52 Although subunit vac-cines prepared from one or more C. trachomatis anti-gens have been suggested as the most promising vac-cine candidates,2021 in the cynomolgus monkey modelpartial resistance to ocular challenge was induced pre-viously only by oral administration of viable serovarL2 EB.31'53 Serologic responses before challenge wereinduced poorly by these latter vaccine candidates, andafter challenge immune responses did not differ sub-stantially, regardless of immunization status.20 No can-didate vaccine to date has reduced clinical or microbi-ologic disease better than a previous ocular infec-tion.29'30

In contrast to these previous attempts, vaccinationwith OGP extract containing primarily MOMP(ompl) induced strong chlamydia-specific humoraland cell-mediated immune responses to whole organ-

ism in naive monkeys before infectious ocular chal-lenge. However, only serologic responses were sus-tained in both immunization groups. In the currentexperiments, the immune responses induced by mu-cosal and systemic administrations of OGP-MOMPwere impressive. It is generally thought that an effec-tive vaccine against trachoma would require stimula-tion of local immunity at the mucosal surface of theconjunctiva.11'54 In humans, a significant inverse corre-lation between the titer of secretory IgA antibodiesin genital secretions and the presence of chlamydiaisolated from the cervix has been shown.12 Also, asecretory IgA antibody response in the genital tractof BALB/c mice is associated with protection againstsubsequent C. trachomatis infection.10 In the currentstudy, low titers of IgA were detected in tears from thesystemically primed animals before ocular infection.These eyes were uninflamed before challenge, sug-gesting that the IgA in tears before challenge hadnot leaked from serum. Of interest after infectiouschallenge, 2 of 4 mucosally immunized animals hadhigher levels of tear IgA that, although still of lowtiter, correlated with greater reduction in microbio-logic and clinical disease. The neutralizing capabilityof these tears was not tested, and an absence of neu-tralization or inadequate IgA titers could explain thefailure to protect eyes better from infectious chal-lenge. Other studies have demonstrated the presenceof protective chlamydial-neutralizing antibodies intears.48'54"56

Either mucosal or systemic immunization withOGP-MOMP induced chlamydia-specific cell-medi-ated immunity as measured by lymphoproliferative re-sponses by peripheral blood mononuclear cells towhole EB. Cell-mediated immunity to purified MOMPor other purified chlamydial antigens was not tested.The cell-mediated immune responses by PBMC wereshort lived before infectious challenge, although puri-fied protein derivative of tuberculin skin test resultsfor recipients of complete Freund's adjuvant were pos-itive for several months after challenge (data notshown). Ocular infection appeared to boost PBMCresponses in the systemic group, but responses variedamong the individual animals. Given the high anti-body titers, particularly in systemically immunized ani-mals, it is possible that peripheral blood T-cell re-sponses were downregulated by immunization as re-sponding lymphocytes localized to mucosal tissues ordraining regional lymphoid tissues. The latter re-sponses were not tested in all animals.

There is no firm evidence that anti-MOMP, cell-mediated immunity itself plays a major protective rolein this disease, although CD4+T cells are probablyrequired for antibody and memory responses. Thereis evidence that local anti-hsp60 T-cell responses are


1488 Investigative Ophthalmology 8c Visual Science, July 1995, Vol. 36, No. 8

250 -i

Q MUCOSAL

O SYSTEMIC

A CONTROL

84 98


# Mucosal

O Systemic

A Control

B28 42 56 70


84

•

oA

MUCOSAL

SYSTEMIC

CONTROLS

(n = 4)

(n = 4)

(n=17)

28 42 56 70


FIGURE 7. Microbiologic evi-dence of partial protectiveimmunity. (A) Culture de-tection of chlamydial inclu-sion bodies in swabs of theconjunctiva for the threegroups of monkeys. Meannumbers of inclusions de-tected in 15 high powerfields (original magnifica-tion, X4.00). (B) Directfluorescence antibody assayresults of 15 high-powerfield counts on conjunctivalsmears. (C) The direct flu-orescent antibody scoresbased on a score of 1 to 4for these groups. The con-trol group includes 15 his-torical control monkeys.



deleterious because they are manifested as local de-layed hypersensitivity reactions.40'41'57 Chlamydialhsp70 does not elicit such reactions in immune ani-mals.47 The role for CD8+T cells in either protectionor pathogenesis is unknown, although many CD8+

cells are present in the inflamed conjunctiva duringinfection.53'58 The immunoblotting results against thetwo chlamydial 60-kd proteins suggested that bothroutes of immunization influenced the immunogenic-ity of additional non-MOMP chlamydial antigens. Be-cause Sarkosyl detergent eliminates the hsp60 antigenfrom preparations, it is not unexpected that antibodyto this antigen did not develop until after infectiouschallenge with whole chlamydia. Similarly, theamount of 60-kd omp2 protein in the OGP-extractmay have been suboptimal to induce an antibody re-sponse after mucosal immunization. However, ocularinfection after immunization was sufficient to boosta previously undetected response because detectableresponses to either antigen did not develop in controlanimals through day 84.

Despite vigorous serologic responses before andafter challenge, transiently positive cellular immuneresponses, and some microbiologic evidence for re-duction in shedding of organism, there were no sig-nificant differences in clinical disease scores for immu-nized and nonimmunized groups of monkeys. Thiscan be attributed in part to the individual variability ofimmune responses to chlamydia by outbred monkeys.However, when serologic responses by individual ani-mals were compared to disease parameters, highertear IgA titers corresponded to best reduction in mi-crobiologic and clinical scores after mucosal immuni-zation. Tear and serum IgG corresponded best to par-tial protection in the systemically immunized groupof monkeys. Cellular immunity persisted through day14 after ocular challenge in two of the mucosally im-munized animals, whereas responses in the systemicgroup remained low for three of four animals at thistime. At 4, 9, and 14 weeks, PBMC from one to threesystemically immunized animals had positive stimula-tion indices, with monkey 907 showing positive resultsat all three times. Mucosally immunized monkeys hadnegative stimulation indices of < 4 after 2 weeks. Al-though extrapolations to requirements for a humanvaccine cannot be made from such small numbers ofanimals, these results suggest that more than oneroute of immunization and mechanism may be protec-tive against chlamydial ocular infection and that acocktail vaccine containing MOMP and additional an-tigen (s) may be required even to protect partially100% of test subjects.

The conjunctival inflammation, induced in themucosally immunized group by concomitant topicalinoculation of CTox with OGP or buffer may have

influenced the clinical results. Instillation of OGP orCTox alone did not induce inflammation. It is possiblethat residual detergent altered the permeability of theconjunctivae so that the cholera toxin could penetrateand cause an inflammatory response. Similar observa-tions have been made with Triton X-100 buffer prepa-rations.58

It is generally accepted that the ideal vaccineshould completely prevent infection or, failing that, atleast decrease the intensity and duration of infection.Recently, Allen and Stephens59 demonstrated differ-ences in the ability of the MOMP variable segmentsto elicit T-cell help for antibody responses in mice.More information regarding such T-helper cell re-quirements may aid efforts to optimize an anti-tra-choma vaccine candidate. The current study showsthat the OGP-MOMP vaccine induced specific anti-body and cell-mediated immunity responses beforechallenge, but this vaccine only partially protectedagainst infectious challenge. Partial protection ap-peared to correlate with serologic and cellular im-mune responses, but antibody class and timing of cell-mediated immunity responses reflected the route ofimmunization. Additional study of vaccine candidatesretaining sufficient native conformation to induce im-mune responses to whole organism will help clarify therequirements for protective immunity to chlamydia.

Key Words

Chlamydia trachomatis, conjunctiva, major outer membraneprotein, trachoma, vaccine

Acknowledgments

The authors thank Dr. W. J. Newhall V and Dr. P. Bavoilfor their gifts of anti-60 kd protein antibodies, Dr. B. Bat-teiger and colleagues for their extraction protocol and help-ful discussions, and Ms. B. Munoz for performing statisticalanalyses. The authors also thank Ms. Vivian Velez and Ms.Lori Dejong for their expert technical assistance.

References

1. Taylor HR, Siler JA, Mkocha HA, Munoz B, West S.The natural history of endemic trachoma: A longitudi-nal study. Amf Trop Med Hyg. 1992;46:552-559.

2. Taylor HR, Siler JA, Mkocha HA, et al. Longitudinalstudy of the microbiology of endemic trachoma, f ClinMicrobiol. 1991;29:1593-1595.

3. West SK, Rapoza P, Munoz B, Katala S, Taylor HR.Epidemiology of ocular chlamydial infection in a tra-choma-hyperendemic area./Infect Dis. 1991; 163:752-756.

4. Taylor HR. Trachoma. Int Ophthalmol. 1990; 14:201-204.

5. Taylor HR, Bell T. Chlamydial infections. In: Strick-land G, ed. Hunter's Tropical Medicine. Philadelphia:WB Saunders; 1984:233-238.


1490 Investigative Ophthalmology 8c Visual Science, July 1995, Vol. 36, No. 8

6. Schachter J, Grossman M, Sweet RL, et al. Prospectivestudy of perinatal transmission of Chlamydia tracho-matis.J Am Med Assoc. 1986; 255:3374-3377.

7. Schachter J, Dawson CR. The epidemiology of tra-choma predicts more blindness in the future. ScandJInfect Dis. 1990;69(suppl):55-62.

8. Ledger WJ. The next 10 years in the treatment ofpelvic infections. Am J Med. 1985; 78:194-203.

9. Westrom L, Wolner-Hanssen P. Pathogenesis of pel-vic inflammatory disease. Genitourin Med. 1993; 69:9-17.

10- Cui ZD, Tristram D, LaScolea LJ, Kwiatkowski T Jr,Kopti S, Ogra PL. Induction of antibody response toChlamydia trachomatis in the genital tract by oral immu-nization. Infect Immun. 1991;59:1465-1469.

11. Schachter J. Overview of Chlamydia trachomatis infec-tion and the requirements for a vaccine. Rev Infect Dis.1985;7:713-716.

12. Brunham RC, Peeling R, Maclean I, McDowell J,Persson K, Osser S. Post-abortal Chlamydia trachomatissalpingitis: Correlating risk with antigen-specific re-sponses and with neutralization. / Infect Dis.1987; 155:749-755.

13. Toye B, Zhong GM, Peeling R, Brunham RC. Immu-nologic characterization of a cloned fragment con-taining the species-specific epitope from the majorouter membrane protein of Chlamydia trachomatis. In-fect Immun. 1990; 58:3909-3913.

14. Baehr W, Zhang Y, Joseph T, et al. Mapping antigenicdomains expressed by Chlamydia trachomatis majorouter membrane protein genes. ProcNatlAcadSci USA.1988; 85:4000-4004.

1$. Stephens RS, Wagar E, Schoolnik G. High resolutionmapping of serovar-specific and common antigenicdeterminants of the major outer membrane proteinof Chlamydia trachomatis. JExp Med. 1988; 167:817-831.

16. Stephens RS, Sanches-Pescador R, Wagar E, InouyeC, Urdea M. Diversity of Chlamydia trachomatis majorouter membrane protein genes. / Bacteriol. 1987; 169:3879-3885.

17. Caldwell HD, Perry LJ. Neutralization of Chlamydiatrachomatis infectivity with antibodies to the majorouter membrane protein. Infect Immun. 1982; 38:745-754.

18. Maclean IW, Peeling RW, Brunham RC. Characteriza-tion of Chlamydia trachomatis antigens with mono-clonal and polyclonal antibodies. Can J Microbiol.1988;34:141-147.

19. Taylor HR, Whittum-Hudson J, Schachter J, CaldwellHD, Prendergast RA. Oral immunization with chla-mydial major outer membrane protein (MOMP). In-vest Ophthalmol Vis Sci. 1988;29:1847-1853.

20. Taylor HR, Stephens RS, Whittum - Hudson JA, Pren-dergast RA. Initial evaluation of trachoma subunit vac-cines. ARVO Abstracts. Invest Ophthalmol Vis Sci.1989; 30:381.

2J. Wong E, Pribnow JA, Stephens R. Topical ocular im-munization with defined Chlamydia trachomatis majorouter membrane epitopes. ARVO Abstracts. InvestOphthalmol Vis Sci. 1990; 31:571.

22. Bavoil P, Ohlin A, Schachter J. Role of disulfide bond-ing in outer membrane structure and permeability inChlamydia trachomatis. Infect Immun. 1984; 44:479-485.

23. Batteiger BE, Rank RG, Soderberg LSF. Immunizationof guinea pigs with isolated chlamydial outer mem-brane proteins. In: Bowie WR, Caldwell HD, Jones RP,etal, eds. Chlamydial Infections. Cambridge: CambridgeUniversity Press; 1990:265-268.

24. Batteiger BE, Rank RG, Bavoil PM, Soderberg LSF.Partial protection against genital reinfection by immu-nization of guinea pigs with isolated outer membraneproteins of the chlamydial agent guinea pig inclusionconjunctivitis. / Gen Microbiol. 1993; 139:2965-2972.

25. Tan TW, Herring AJ, Anderson IE, Jones GE. Protec-tion of sheep against Chlamydia psittaci infection with,a subcellular vaccine containing the major outer mem-brane protein. Infect Immun. 1990;58:3101-3108.

26. Montgomery PC, RockeyJC, Majumdar AS, Lemaitre-Coelho IM, Vaerman J-P, Ayyildiz H. Parameters in-fluencing the expression of IgA antibodies in tears.Invest Ophthalmol Vis Sci. 1984; 25:369-373.

27. Whittum-Hudson JA, Taylor HR. Antichlamydialspecificity of conjunctival lymphocytes during experi-

Triental ocular infection. Infect Immun. 1989; 57:2977-2983.

28. Newhall WJ, Terko P, Wilde CE, Batteiger BE, JonesRB. Serovar determination of Chlamydia trachomatisisolates by using type-specific monoclonal antibodies./ Clin Microbiol. 1986;23:333-338.

29. Taylor HR. Development of immunity to ocular chla-mydial infection. Am f Trop Med Hyg. 1990; 42:358-364.

30. Taylor HR. Ocular models of chlamydial infection.Rev Infect Dis. 1985; 7:737-740.

31. Taylor HR, Young E, MacDonald AB, Schachter J,Prendergast RA. Oral immunization against chlamyd-ial eye infection. Invest Ophthalmol Vis Sci.1987; 28:249-258.

32. West SK, Taylor HR. Reliability of photographs forgrading trachoma in field studies. Br f Ophthalmol.1990;74:12-13.

33. Thylefors B, Dawson CR, Jones BR, West SK, TaylorHR. A simple system for the assessment of trachomaand its complications. Bull World Health Organ.1987;65:477-483.

34. Taylor HR, Agarwala N, Johnson SL. Detection of ex-perimental Chlamydia trachomatis eye infection in con-junctival smears and in tissue culture by use of fluo-rescein-conjugated monoclonal antibody. / Clin Micro-biol. 1984; 20:391-395.

35. Drummond J, Shah K, Donnenberg A. Cell-mediatedimmune responses to BK virus in normal individuals.fMed Virol. 1985; 17:237-247.

36. Batteiger BE, Rank RG. Analysis of the humoral im-mune response to chlamydial genital infection inguinea pigs. Infect Immun. 1987; 55:1767-1773.

37. Newhall WJ V, Batteiger B, Jones RB. Analysis of hu-man serological responses to proteins of Chlamydiatrachomatis. Infect Immun. 1982; 38:1181 -1189.

38. Pal S, Taylor HR, Whittum - Hudson JA. Identification



of fine antigenic specificities of antibody responsesin experimental Chlamydia trachomatis infection. InvestOphthalmol Vis Sri. ARVO Abstracts. 1990; 31:448.

39. Pal S, Taylor HR, Huneke RB, Prendergast RA, Whit-tum-Hudson JA. Frequency of antigen-specific B cellsduring experimental ocular Chlamydia trachomatis in-fection. Infect Immun. 1992;60:5294-5297.

40. Morrison RP, Lyng K, Caldwell HD. Chlamydial dis-ease pathogenesis: Ocular hypersensitivity elicited bya genus-specific 57 kd protein. J Exp Med. 1989; 69:663-675.

41. Watkins NG, Hadlow WJ, Moos AB, Caldwell HD. Ocu-lar delayed hypersensitivity: A pathogenic mechanismof chlamydial conjunctivitis in guinea pigs. Proc NatlAcadSri USA. 1986;83:7480-7484.

42. Bavoil P, Stephens RS, Falkow S. A soluble 60 kilodal-ton antigen of Chlamydia spp. is a homologue of Esche-richia coli GroEL. Mol Microbiol. 1990;4:461-469.

43. Taylor HR, Pierce NF, Zhang P, Schachter J, Sil-verstein AM, Prendergast RA. Secretory immune cellu-lar traffic between the gut and the eye. In: O'ConnorGR, Chandler JW, eds., Advances in Immunology andImmunopathology of the Eye. New York: Masson Publish-ing; 1985:208-211.

44. Pal S, Pu Z, Huneke RB, Taylor HR, Whittum-Hud-son JA. Chlamydia-specific lymphocytes in conjunctivaduring ocular infection: Limiting dilution analysis. RegImmunol. 1990; 3:171-176.

45. Taylor HR, Kolarczyk RA, Johnson SL, Schachter J,Prendergast RA. Effect of bacterial secondary infec-tion in an animal model of trachoma. Infect Immun.1984; 44:614-616.

46. Taylor HR, Prendergast RA. Attempted oral immuni-zation with chlamydial lipopolysaccharide subunit vac-cine. Invest Ophthalmol Vis Sri. 1987;28:1722-1726.

47. Taylor HR, Maclean IW, Brunham RC, Pal S, Whit-tum-Hudson J. Chlamydial heat shock proteins andtrachoma. Infect Immun. 1990; 58:3061-3063.

48. Nichols RL, Oerdey RE, Fraser CEO, MacDonald AB,McComb DE. Immunity to chlamydial infections ofthe eye: VI: Homologous neutralization of trachomainfectivity for the owl monkey conjunctivae by eye se-

cretions from humans with trachoma. J Infect Dis.1973; 127:429-431.

49. Grayston JT, Wang S. New knowledge of chlamydiaeand the disease they cause. / Infect Dis. 1975; 132:87-104.

50. Fraser CEO, McComb DE, Murray ES, MacDonald AB.Immunity to chlamydial infections of the eye: IV: Im-munity in owl monkeys to reinfection with trachoma.Arch Ophthalmol. 1975;93:518-521.

51. Thygeson P, Dawson C, Hanna L, Jawetz E, OkumotoM. Observations on experimental trachoma in mon-keys produced by strains of virus propagated in yolksac. Am J Ophthalmol. 1960; 50:907-917.

52. Jawetz E, Rose L, Hanna L, Thygeson P. Experimentalinclusion conjunctivitis in man: Measurements of in-fectivity and resistance. JAm Med Assoc. 1965; 194:620-632.

53. Whittum-Hudson JA, Prendergast RA, Taylor HR.Changes in conjunctival lymphocyte populations in-duced by oral immunization with Chlamydia tracho-matis. CurrEyeRes. 1986; 5:973-979.

54. Murray ES, Charbonnet LT, MacDonald AB. Immu-nity to chlamydial infections of the eye: The role ofcirculatory and secretory antibodies in resistance tore-infection with guinea pig inclusion conjunctivitis. /Immunol. 1973; 110:1518-1525.

55. Rank RG, White HJ, Barron AL. Humoral immunityin the resolution of genital infection in female guineapigs infected with the agent of guinea pig inclusionconjunctivitis. Infect Immun. 1979;26:573-579.

56. Barenfanger J, MacDonald AB. The role of immuno-globulin in the neutralization of trachoma infectivity.JImmunol. 1974; 113:1607-1617.

57. Taylor HR, Johnson SL, Schachter J, Caldwell HD,Prendergast RA. Pathogenesis of trachoma: The stim-ulus for inflammation. / Immunol. 1987; 138:3023-3027.

58. Whittum-Hudson JA, Taylor HR, Farazdaghi M,Prendergast RA. Immunohistochemical study of thelocal inflammatory response to chlamydial ocular in-fection. Invest Ophthalmol Vis Sri. 1986; 27:64-69.

59. Allen JE, Stephens RS. An intermolecular mechanismof T cell help for the production of antibodies tothe bacterial pathogen, Chlamydia trachomatis. Eur fImmunol. 1993; 23:1169-1172.


articles a chlamydial major outer membrane protein extract as a

Documents