CA Patent 2135800 - COMPOSITIONS USEFUL IN DIAGNOSIS AND PROPHYLAXIS OF LYME DISEASE

Description

COMPOSITIONS USEFUL IN DIAGNOSIS AND

PROPHYLAXIS OF LYME DISEASE

Cross-Reference to RelatedAsplication

This is a continuation-in-part of pending U. S.

Patent Application SN 944,464, filed September 14, 1992, which is a continuation-in-part of U. S. Patent

Application SN 07/889,015, filed May 26, 1992.

Field of the Invention

The present invention relates generally to the field of pharmaceutical and diagnostic compositions useful in the diagnosis, treatment and prophylaxis of

Lyme borreliosis. More specifically, the invention provides isolated natural and recombinant proteins, polypeptides and macromolecular antigens that are derived from Borrelia burgdorferi and differentiated in the tick vector.

Background of the Invention

The causative agent of Lyme borreliosis, i.e.,

Lyme disease, the bacterium Borrelia burgdorferi, is transmitted by the bite of various species of Ixodes ticks carrying the spirochete. The reservoir of the infection is most likely the white footed mouse,Peromyscus leucopus, and the disease can be transmitted to many mammalian species including dogs, cats, and man [J. G. Donahue, et al., Am.S. Troop. Med.His.; 36:92-96 (1987); R. T. Green, petal., J. Clin. Micro., 26:648-653(1988)1. The B. burgdorferi strain B31 is believed to be responsible for about 90% of the known isolates of

Borrelosis in the United States.The diagnosis of Lyme disease in humans and animals has been compromised by the lack of definitive serology leading to rapid and accurate testing and the lack of suitable bacterial antigens capable of eliciting an immune response in vaccinates.

Cultured preparations of Borrelia burgdorferi have been used to generate whole cell sonicate for ELISA and western blot analysis, which has yielded only marginal results [M. Karlsson et al., Eur. J. Clin.

Microbiol. Infect. Dis., 8:871-877 (1990), S. W. Luger et al., Arch. Intern. Med.,15:761-763 (1990), I. Olsson et al., Acta. Derm. Venereol. (Stockh.), 71:127-133(1991)g.

A number of bacterial antigens have been identified that may be more useful in serodiagnosis, including outer surface proteins A and B (OspA and OspB), flagellin, and other proteins designated P21, P39, P66, and P83 according to their estimated molecular weights [A. G.

Barbour et al., Infect. Immun., 45:94-100 (1984); W. J.

Simpson et al., J. Clin. Microbiol.. 28:1329-1337 (1990);

K. Hansen et al., Infect. Immun., 56:2047-2053 (1988); K.

Hansen et al., Infect. J. Clin. Microbiol., 26:338-346 (1988); B. Wilske et al.,Zentral. Bakteriol.

Parasitenkd. Infektionshkr.Hvs. Abt. 1 Oria.Peihe. A.

263:92-102 (1986); D. W. Dorward et al., J. Clin.

Microbiol., 29:1162-1170 (1991)].

There exist a wealth of publications relating to proteins and polypeptides of Borrelia burgdorferi which suggest their use as diagnostic or pharmaceutical agents. These documents predominantly produce the B.

burgdorferi proteins and polypeptides in in vitro culture, primarily in Barbour-Stoenner-Kelley (BSK) modified medium, and identify the polypeptides and proteins on the bases of their molecular weights using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) techniques. For example, Published NTIS US patent application No. 485,551 refers to B. burgdorferi proteins of molecular weights of 11, 13, 22, 13, 34 and 83 kDa. Published European patent application No.

465,204, published January 8, 1992 refers to antigens spanning the molecular weight range from 25 through 93kDa. International Patent Application No.

PCT/US91/01500, published September 19, 1991 refers to B.

burgdorferi antigens having molecular weights of 28 and 39 kDa. International Patent Application No.

PCT/EP90/02282, published July 11, 1991 refers to B.

burgdorferi antigens having molecular weights of 17, 22, 31, 41 and 100 kDa. International Patent Application No.

PCT/DK89/00248, published May 3, 1990 refers to B.

burgdorferi antigens having molecular weights spanning the range of 20 through 85 kDa.

PCT patent application No.W092/00055, published January 9, 1992, discloses OspA and OspB polypeptides, fusion proteins and derivatives of B.

burgdorferi strains N40 and 25015. This publication refers to any DNA sequence or polypeptide that elicits in a treated mammalian host an immune response effective to protect against Lyme Disease caused by infection with

Borrelia burgdorferi. In fact, the polypeptides are only identified by immunizing a mouse of strain C3H/He with polypeptide derived from culture grown Borrelia burgdorferi, needle inoculating an immunized animal with

Borrelia burgdorferi, and selecting a polypeptide which protects the immunized animal against the inoculated

Borrelia burgdorferi.

Few of the art-described B. burgdorferi antigens have been characterized by immunological criteria, e.g., immune response assay. The vast majority of the published data analyzing the utility of these bacterial antigens in serology testing for Lyme disease have used a system of needle inoculation of cultured spirochetes or purified proteins to assess immune responsiveness [U. E. Schaible, et al., Eur. J. Immunol., 21:2397-2405 (1991); S. W. Barthold, et al., Am. J. of

Path., 139:263-273(1991)). To date there have been no observations about whether or not a difference exists in the course of the infection or the nature of the immune response when the B. burgdorferi bacterial antigens are transmitted to the animal by tick bite as compared to experimental infection by needle inoculation. See, also,

J. A. Nelson et al., J. Infect. Dis., 161:1034-1035 (1990); W.F. Schubach et al., J. Clin. Microbiol., 59:1911-1915 (1991); S. W. Barthold et al., J. Infect.

Dis., 162:133-138 (1990); J. Lindenmayer et al., J. Clin.

Microbiol., 28:92-96 (1990); and U. E. Schaible et al.,

Proc. Natl. Acad. Sci., 87:3768-3772 (1990).

There is a need in the art for methods and compositions for accurate and early detection of Lyme disease, as well as treatment and prophylaxis of natural infection of animals with B. burgdorferi.

Summarv of the Invention

In one aspect, the invention provides isolated

B. burgdorferi antigens which are regulated and differentiated by growth of the B. burgdorferi in a tick vector. Novel antigens of the invention are listed below in Table I.

Certain of these antigens are characterized as being B. burgdorferi B31 strain specific and major histocompatibility complex(MHC) nonrestricted. Certain other of these antigens are characterized as being MHCrestricted. Sera generated to these antigens (B31 MHC nonrestricted and B31 MHC restricted) are further characterized by the ability or lack of ability to react with B. burgdorferi JD-1 strain; the antigens themselves (B31 MHC nonrestricted and B31 MHC restricted) are further characterized by being homologous or heterologous with B. burgdorferi JD-1 strain antigens. The most preferred antigens of this invention, because of their ability to induce cross-strain immunity to B. burgdorferi in different animal haplotypes, are characterized by being B31 MHC nonrestricted, JD-1 crossreactive, and JD-1 nonrestricted.Other antigens are also useful in vaccine compositions and as diagnostics.

In another aspect, the invention provides a B.

burgdorferi antigen obtained from the microorganism which is isolated directly from a tick which has optionally been infested (feeding) on infected host animals.

Optionally the B. burgdorferi has been passaged only a minimum number of times in growth media to retain its virulence and antigenicity before the antigen is isolated. A desirable minimum number of in vitro passages is between 1 and 6.

In yet another aspect, the invention provides a

B. burgdorferi antigen which is prepared from B.

burgdorferi DNA isolated from a tick vector or from low passage culture of the microorganism. Still a further aspect provides a B. burgdorferi antigen which is prepared from PCR gene banks derived from ticks infected with B. burgdorferi or from low passage culture.

In another aspect, the present invention provides antibodies to the novel B. burgdorferi antigens of the invention. Such antibodies may be monoclonal, polyclonal or recombinant.

In yet another aspect, the present invention provides diagnostic reagents, therapeutic compositions, and vaccinal compositions which comprise at least one of the novel antigens or antibodies of the invention.

In a further aspect, the invention provides diagnostic assays and kits, methods of treating Lyme disease, and methods of preventing Lyme disease which utilize the reagents and compositions of the invention.

Other aspects and advantages of the present invention are described further in the following detailed description of the preferred embodiments thereof.

Brief Description of the Drawings

Fig. 1 illustrates the antibody response to whole-cell sonicate, purified flagella, and recombinant

OspA in tick-infected hamsters. Antibody response was determined by ELISA on whole-cell sonicate(O), purified flagella(s), or purified recombinant OspA(A).

Geometric mean (GMT)log10 ELISA titers are reported (n = 6). Standard error of the mean for all points was < 0.1.

Fig. 2 illustrates the antigenic dose-response analysis of hamsters inoculated with various concentrations of B. burgdorferi. Hamsters were bled at 2(3) and 4 (0) weeks post-infection(P.I.). Antibody response was determined by ELISA on whole-cell sonicate (solid lines) or recombinant OspA (dotted lines).

Geometric mean (GMT) logs0 ELISA titers are reported (n = 2 at 14 days, n = 4 or 5 at 28 days).

Detailed Description of the Invention

Antigens

The present invention identifies novel isolated

Borrelia burgdorferi antigens which are regulated and/or differentiated by transmission of the B. burgdorferi through a tick vector. As used herein, a tick vector refers to an arthropod organism of the Insect Family, which is capable of vectoring B. burgdorferi. The tick vector is optimally of the Ixodes genus, and of the

Ixodes complex I.dammini, I. scapularus, I. pacificus, or I. ricinus or related species or genera that can transmit the microorganism which causes Lyme disease.

Because growth of B. burgdorferi in growth media in vitro causes the loss of virulence associated with Lyme disease, this invention uses response in genetically defined MHC haplotype mice to serologically identify the differences between tick borne B.

burgdorferi antigens and those produced by culture grown

B. burgdorferi. One of the best indicators of differentiation and/or regulation is the appearance or loss of antigens as measured by antibody specificity produced when the tissue or cells are inoculated into a human or an animal. Antigens produced and vectored by ticks are distinct from prior art B. burgdorferi antigens which have been multiply passaged in vitro in growth media.

Furthermore, B. burgdorferi that truly are the causative pathogens of Lyme disease in humans and animals are believed to share a common group of molecules that induce the clinical symptoms associated with Lyme disease. Because B. burgdorferi passed in vitro lose virulence, the proteins or other molecules associated with virulence factors are considered to be provided by the tick since it provides the only viable way to transmit disease. The present inventors have identified antigens useful in vaccines by determining if the antigen is major histocompatibility complex (MHC) restricted.

An MHC is a group of proteins (receptors) on the surface of all cells in an animal or man. See, P.J.

Bjorkman et al, Nature, 329:506-518 (1987). Typically, proteins of all cells of an animal or man degrade or turn over as a natural progression of events. To prevent auto-reactivity, the immune surveillance looks at each peptide removed from the cell. Typically, it is displayed through the MHC as it leaves the cell or is internalized and degraded to individual amino acids to re-enter the metabolite pool. Because an animal does not want to react to its own proteins, that animal's "self" proteins are restricted to response. If a protein looks like a self protein then the immune system of that animal will not react. If the immune system sees a protein or antigen as foreign, the MHC does not restrict the response.

Certain antigens of this invention are significantly characterized as not restricted to any major histocompatibility complex (MHC) haplotype. By the phrase "not MHC restricted" is meant that an antigen of this invention is capable of eliciting an immune response to Lyme disease in humans and animals regardless of their immunogenetic background (the human or animal's MHC does not restrict the immune response). Because the antigens of this invention describe those serological responses that occur in mice early after infestation with infected ticks, they may also represent the virulence (common) components associated with virulent B. burgdorferi infection.

Certain preferred antigens of this invention have been shown to elicit an immune response across three different MHC backgrounds in mice. Preferably, the antigens are further characterized as crossreactive with more than one B. burgdorferi strain that is capable of inducing clinical symptomology in humans and/or animals associated with Lyme disease, as demonstrated in the examples below. As used herein when referring to antigens, the term "crossreactive" indicates that sera generated to antigens derived from one strain is capable of reacting with antigens of another strain and that the antigens of the two strains are homologous. Sera generated to antigens derived from one strain which is not capable of reacting with antigens of another strain indicates that the antigens are heterologous.

Antigens of B. burgdorferi are identified below with reference to their molecular weight in kilodaltons.

For example, a B. burgdorferi antigen of about 17 kd in molecular weight is identified as P17; an antigen of about 22 kd in molecular weight is identified as P22 and so on. P41 (fla), as used in Table I, refers to a flagella-associated antigen of molecular weight 41 kd.

The other antigens of Table I are not associated with the flagella, outer surface protein (Osp) A, or Osp B.

The characteristics of B. burgdorferi lowpassage antigens (i.e., antigens obtained from the

Borrelia microorganism which has been passaged in culture less than 6 times) of this invention are identified below in the table and by reference to several examples presented below. Table I summarizes the data from

Examples 11-13, in which western blot analysis was performed using sera from a panel of inbred strains of mice infected with B31 strain B. burgdorferi by exposure to Ixodesdammini ticks infected with the B31 isolate of the spirochete on proteins from B31 and JD-1 strains.

Table I shows that certain antigens are characterized by being B31 MHC nonrestricted, i.e. these antigens are capable of eliciting an immune response in all haplotypes. These B31 MHC nonrestricted antigens are further characterized by being (a) JD-1 crossreactive and

JD-1 MHC nonrestricted, (b) JD-1 crossreactive and JD-1

MHC restricted, or (c) JD-1 nonreactive. The antigens of group (a), are capable of eliciting the broadest immune response in animals. Antigens which fall into categories (b) and (c) are also useful in vaccine compositions of this invention.

Table I also shows that the antibody response to a number of the novel protein antigens derived from B.

burgdorferi strain B31 were MHC restricted. For example, until late in the infection, only the mouse strain B10 animals respond to the P83 protein. Since mice of the strains B1O.BR and B1O.D2 are perfect genetic matches except for the MHC locus, apparently, MHC haplotypesH-2d and H-2k are relatively inefficient at presenting this protein when compared to H-2b. See, particularly, Examples 10 and 12 below.

As described above for the nonrestricted antigens, these MHC-restricted antigens are also further characterized by being (a) JD-1 crossreactive and JD-1

MHC nonrestricted, (b) JD-1 crossreactive and JD-1 MHC restricted, or (c) JD-1 nonreactive. Although the immune response elicited by these novel B. burgdorferi antigens is B31MMC restricted, and thus, do not elicit an immune response in all animal haplotypes, they are also useful in vaccines to prevent Lyme Disease in humans and other animals and in diagnostic assays.

A summary of the characteristics of the novel

B. burgdorferi antigens of the invention, obtained in

Examples 9-13, is shown in Table I below.

Table I

Summary of antibody binding data

B31 MHC B31 MHC

restricted nonrestricted

JD-1 P29 P24 nonreactive P22

P32 P14

JD-1 P83 P17 crossreactive P52-P65

MHC restricted P41 (fla)

P28

P34

JD-1 P30 P43 crossreactive P39

MHC nonrestricted P18

Another preferred characteristic of the novel antigens of this invention lies in their ability to produce antibodies in an infected human or animal early in the B. burgdorferi infection period,e.g., within the first two weeks post infection(P.I.). Still another preferred characteristic of antigens of this invention is that the antibodies directed thereto are stably produced in the human or animal throughout the infection period.

By the phrase "stably produced throughout the infection period" is meant detectable production of antibody in the period post infection spanning from about two weeks P.I.

through and beyond 60 days P. I. It is anticipated that such antibodies will remain detectable in a human or an animal for up to one year P.I. or longer because in most cases the spirochete will persist in the host for long periods of time. However, antigens isolated from the organism and formulated with an appropriate delivery system are anticipated to provide an immune response that would predispose the animal to infection.

Thus, these antigens of this invention are preferably characterized by the ability to prevent or protect a human or an animal against Lyme Disease by eliciting a stable protective antibody response in a human or animal early during Borrelia burgdorferi infection. Given the above characteristics, these novel

B. burgdorferi antigens are useful in diagnosing Lyme disease, and in vaccinal and therapeutic compositions.

The antigens of the invention may be isolated from any selected strain, variant, group or subunit of

Borrelia. Preferably, antigens are isolated from B.

burgdorferi, particularly from subgroups A and B which are associated most frequently with Lyme borreliosis.

However, the spirochete isolates may also be from subgroups C through F [A. G. Barbour et al., J. Infect.

Dis., 152:478-484(1985)), as well as other subgroups yet to be identified.

The antigens of this invention are obtained and isolated from the microorganism directly from the tick vector or from sera of a human or animal infected with B.

burgdorferi via a tick bite. The antigens may also be obtained by serial tick to host animal infections and then obtained from the tick or the serum of a tick infested, B. burgdorferi infected animal. None of the prior art references describe the need to maintain the B.

burgdorferi by tick to host passage. While passaging in growth media is preferably avoided, optionally, the microorganism can be passaged at least once in BSK media simply to amplify the microorganism to provide the antigenic material. For example, where the B.

burgdorferi strain JD-1 is employed, one passage supplies sufficient amplification. Where the B31 strain is employed, passages up to 6 may be employed. However, to obtain the antigens of this invention, no high passage in vitro culturing is employed. For the purposes of this invention, the term "high passage" is defined as greater than 6 passages for strain B31, and more than one passage for strainJD1. Generally, one of skill in the art can determine the number of passages which constitute "high passage" for other strains of B. burgdorferi by assaying the passaged microorganism in hamsters, as described in

Examples 5 through 8 below. Failure of the microorganism to cause clinical symptoms of Lyme Disease, or to illustrate the growth characteristics of B. burgdorferi in the hamsters is indicative of "high passage" for that strain.In other words, the microorganism has been cultured in vitro sufficiently to lose most of its virulence and thus the antigens obtained therefrom differ from the antigens of the present invention.

The antigens of the invention are preferably isolated by immunoblot procedures according to their respective molecular weights, as described below in

Example 4. Such isolation may provide the antigens in a form substantially free from other proteinaceous and nonproteinaceous materials of the microorganism and the tick vector. The molecules comprising the B. burgdorferi polypeptides and antigens of this invention may beisolated from the tick and further purified using any of a variety of conventional methods including: liquid chromatography such as normal or reversed phase, using

HPLC, FPLC and the like; affinity chromatography (such as with inorganic ligands or monoclonal antibodies); size exclusion chromatography; immobilized metal chelate chromatography; gel electrophoresis; and the like. One of skill in the art may select the most appropriate isolation and purification techniques without departing from the scope of this invention.

Additionally, the polynucleotide and amino acid sequences of these antigens may be obtained by resort to conventional genetic engineering techniques, PCR, and the like. [See, e.g., Sambrook et al., Molecular Cloning. 2d edition, Cold Spring Harbor Laboratory, NY(1989)1. For example, the antigen of this invention may be prepared or isolated from B. burgdorferi DNA isolated from the tick vector using DNA probes and PCR techniques.

Alternatively, the antigen may be obtained from PCR gene banks derived from B. burgdorferi-infected ticks or from low passage of the microorganism in growth media.

While not wishing to be bound by theory, the inventors believe that the tick vectors modify the B.

burgdorferi antigens through tick regulatory or genetic elements. B. burgdorferi is believed to need the environment of its vector, the tick, in order to differentiate as an organism or to process certain of its proteins and antigens into virulentand infective states.

Thus, the B. burgdorferi proteins or antigens of this invention which are obtained from the tick are either modified by the tick environment, or are completely different proteins from those obtained by repeated culturing of B. burgdorferi in growth media.

Borreliaburgdorferi ;rnay lose its virulence after only two or more passages in culture media. The

B31 prototype strain obtained from the American Type

Culture Collection, Rockville, Maryland is distributed at the extremely high passage level of at least sixty.

Therefore, the inventors speculate that this strain is only weakly virulent by the time it reaches the requestors of the strain. It is to be noted that most of the reports in the art deal with this strain and indicate that it is obtained from the ATCC.

As illustrated in the examples below, long term passage of strains of B. burgdorferi induced a significant change in the expression of a number of bacterial proteins. See, particularly, Examples 12 and 13. Western blot analysis using the same sera on low passage and high passage isolates of the same strain of bacterium gave remarkably different banding patterns.

Most notably, a number of proteins were not detected in the high passage isolates and the majority of bands are of less intensity. A number of these protein antigens were lost or modified by multiple passages. In the case of either strain of bacteria used in the examples, the reactivity of the mouse sera with lysates of these high passage derivatives of B31 and JD-1 was remarkably less than with the low passage derivative of the same strain (Example 13). Protein bands which were no longer detected at all included P14, P18, P22, P24, P28, a number of bands in the 31 to 38 kD range, P43, P52, and some of the 55 to 65 kD bands. The detection of the P39 bands is consistent throughout the screen of all these different sources of antigen with all of the mouse sera.

These results indicated a significant difference in the antigenic profile of the low and high passage bacterial isolates.

The differences in the antigens of this invention versus those of the art is further evidenced by the differential expression of B. burgdorferi antigens detected in hamsters and mice, when the route and type of infection was altered. Also demonstrated by the examples, is the observation that the pattern of antibody response from inbred strains of mice naturally infected with Borrelia burgdorferi transmitted by the feeding of infected Ixodes ticks differed significantly from the antibody response pattern in mice equivalently infected with needle inoculation of cultured Borrelia burgdorferi spirochetes. See, particularly, Examples 7, 10, and 11.

Needle-inoculated, cultured B. burgdorferi produced different antigenic responses than did needle inoculated homogenates of infested tick vectors, and tick-vectored infection.

Antigens of this invention are also preferably characterized by recognition by sera or T cells obtained from tick infested, B. burgdorferi infected, humans or animals. In experiments, sera from animals treated with

Borrelia burgdorferi infected ticks failed to recognize a family of Borrelia burgdorferi proteins grown in culture media. See, for example, Examples 7 and 10 below.

The prior art fails to positively identify any of the Borrelia burgdorferi antigens on the basis of criteria which are able to discriminate between various reported Borrelia burgdorferi antigens, because the most frequently reported characteristic is molecular weight.

However, despite similar molecular weights, the present inventors are able to discriminate between:

(a) Borrelia burgdorferi antigens of this invention defined or regulated by transmission through the tick vector from culture-grown Borrelia burgdorferi described in the art (see Example 12);

(b) Borrelia burgdorferi antigens identified by antibody from individuals infested with B. burgdorferi infected ticks, but screened using high passage (as in the prior art) versus low passage culture grown Borrelia burgdorferi extracts;

(c) Borrelia burgdorferi antigens identified by antibody from individuals vaccinated with high-passage culture grown Borrelia burgdorferi extracts or subunits and screened with tick derived or low-passage Borrelia burgdorferi extracts or subunits;;

(d) antigens from Borrelia burgdorferi that initiate immune pathology or have virulence factors, i.e., adhesions factors that chelate tissue in the tick or host. The high passage cultured B. burgdorferi are not virulent and have visibly different antigenic profiles, as described below in Example 12; and

(e) Borrelia burgdorferi antigens that are not restricted to MHC haplotypes (see Table I).

The antigens of this invention may be proteinaceous, such as a protein, a polypeptide or fragment thereof. These antigens may be isolated generally free from contamination with proteinaceous and other materials with which the antigens are associated in nature.

Thus, antigens of this invention may be characterized by immunological measurements against the vector borne pathogen using MHC-restricted response in genetically inbred animals. These measurements include, without limitation, western blot, macromolecular weight determinations by biophysical determinations, such as

SDS-PAGE/staining, HPLC and the like, antibody recognition assays, T-cell recognition assays, MHC binding assays, and assays to infer immune protection or immune pathologybyvadoptive transfer of cells, proteins or antibodies.

It should also be understood that the B.

burgdorferi antigens, proteins and polypeptides of this invention may be part of a largerand/or multimeric protein. Antigens of this invention may be in combination with B. burgdorferi outer surface proteins, such as OspA and OspB. In such combination, the antigen may be in the form of a fusion protein. For example, an antigen or polypeptide of this invention may be fused at its N-terminus or C-terminus to OspA polypeptide, or OspB polypeptide or to a non-OspA non-OspB polypeptide or combinations thereof. OspA and OspB polypeptides which may be useful for this purpose include polypeptides identified by the prior art [see, e.g.PCT/US91/04056] and variants thereof.Non-OspA, non-OspB polypeptides which may be useful for this purpose include polypeptides of the invention and those identified by the prior art, including, the B. burgdorferi, flagella-associated protein and fragments thereof, other B. burgdorferi proteins and fragments thereof, and non-B. burgdorferi proteins and fragments thereof.

In one embodiment of this invention, fusion proteins comprising multiple polypeptides of this invention are constructed for use in the methods and compositions of this invention. These fusion proteins or multimeric proteins may be produced recombinantly, or may be synthesized chemically. They also may include the polypeptides of this invention fused or coupled to moieties other than amino acids, including lipids and carbohydrates. Further, antigens of this invention may be employed in combination with other Borrelia vaccinal agents described by the prior art, as well as with other species of vaccinal agents, e.g, derived from parvovirus, coronavirus,leptosporidia,,rand rabies. Such proteins are effective in the prevention, treatment and diagnosis of Lyme disease as caused by a wide spectrum of B.

burgdorferi isolates.

Antigens of this invention are non-OspA, non

OspB B. burgdorferi antigens that can elicit a protective response in vaccinated humans or animals that are exposed to infestation with Borrelia burgdorferi infected ticks where any of the following circumstances apply:

(a) where preferred antigens are recognized by sera or T-cells taken from humans or animals infested with B. burgdorferi infected ticks (see Examples 11-13);

(b) where preferred antigens are not restricted to preferred MHC haplotypes in mice (see Table

I, and Examples 11-13);

(c) where preferred antigens are cross reactive with more than one common form of Borrelia burgdorferi in nature, e.g., B31 or JD1 (see Table I, and

Examples 11-13);;

(d) where antigens are derived from Borrelia burgdorferi taken from the host or tick without intervention of in vitro cultivation or with minimum numbers of passages in growth media (see Table I, and

Example 9);

(e) where the non-OspA and non-OspB antigen can elicit in a vaccinated human or animal antibody or a cell mediated immune (CMI) response that can eliminate or greatly reduce the ability to cultivate viable Borrelia burgdorferi from infected ticks used to infect host humans or animals (see Example 15).

Examples of antigens of this invention which induce a protective immune response in humans or animals and are not restricted to MHC include the following: The antigen P18 is characterized by a molecular weight of about 18 kd as measured by immunoblot (See Example 4).

This antigen-is capable of eliciting an antibody response by one month post-infection and stably throughout infection in humans or animals infected with B.

burgdorferi by a tick vector. It also elicits a response for mice of the H-2k, H-2d, and H-2b haplotypes of the MHC, i.e., P18 is not MHC-restricted. This antigen also demonstrates the ability to elicit a crossreactive response between strains of B. burgdorferi, e.g., strains

B31 and JD-1.

The antigen P43 is characterized by a molecular weight of about 43 kD as measured by immunoblot (See

Example 4). This antigen is capable of eliciting an antibody response in animals infected with B. burgdorferi by a tick vector by 12 days after infection, which response remains stable throughout infection. P43 is capable of eliciting a response from hamsters and mice of the H-2k, H-2b, and H-2d haplotypes of the MHC, i.e., P43 is not MHC restricted. This antigen is expressed by at least 2 different strains of B. burgdorferi, B31 and JD1, low passage isolates, but not by high passage (long term culture) isolates of the same strains, and elicits a cross-reactive response between strains B31 and JD-1.

For example, a small panel of human patients that were demonstrated to be infected with Borrelia burgdorferi(by culturing the spirochetes from skin biopsies at the leading edge of the characteristic erythema migrans rash) were screened. Plasma from these patients were reactive with P43 in 10/20 patients using JD-1 low passage as antigen and 11/20 patients using B31 low passage as antigen.

Another exemplary antigen of this invention is

P39, which is characterized by a molecular weight of about 39 kd as measured by immunoblot. Like the other antigens of this invention, it elicits an antibody response in humans and animals infected with B.

burgdorferi by a tick vector, which response is stable throughout infection. It is not MHC-restricted, and it can elicit a cross-reactive response between a variety of

B. burgdorferi strains.

Still another B. burgdorferi antigen of the invention which is of interest isP17. The isolated antigen P17 is characterized by a molecular weight of about 17 kD as measured by immunoblot (See Example 4).

This antigen is also characterized by the ability to elicit an antibody response by one month post-infection and stably throughout infection. P17 is also capable of eliciting an MHC-nonrestricted response when assayed on homologous strain B31, but responds to the heterologous strain JD-1 in a restricted fashion. (see Example 13).

P17 is not MHC restricted with respect to B31, but restricted with respect to JD-1. This antigen also demonstrates the ability to elicit a crossreactive response between strains of B. burgdorferi, e.g., strains

B31 and JD-1. Other B31 MHC nonrestricted antigens are

P24, P22 andP14, which are non-reactive with JD-1.

Still other antigens of interest in the present invention are those which are B31 MHC restricted, but JD1 non-reactive, i.e., P29 and P32. Other antigens of interest are those which are B31 MHC restricted, JD-l crossreactive and non-MHC restricted, i.e., P30. B.

burgdorferi antigens, P83, P52-65, P41 (fla), P28 and P34 are B31 non-MHC restricted, JD-1 crossreactive but restricted.

The antigens of the present invention may also be modified to increase their immunogenicity. For example, the antigens may be coupled to chemical compounds or immunogeniccarriers,:provided that the coupling does not interfere with the desired biological activity of either the antigen or the carrier. For a review of some general considerations in coupling strategies, see Antibodies. A Laboratorv Manual, Cold

Spring Harbor Laboratory, ed. E. Harlow and D.Lane (1988). Useful immunogenic carriers known in the art, include, without limitation, keyhole limpet hemocyanin (KLH); bovine serum albumin (BSA), ovalbumin, PPD (purified protein derivative of tuberculin); red blood cells; tetanus toxoid; cholera toxoid; agarose beads; activated carbon; or bentonite.Useful chemical compounds for coupling include, without limitation, dinitrophenol groups and arsonilic acid. The antigens may also be modified by other techniques, such as denaturation with heat and/or SDS. Further, the antigens of the invention may be optionally fused to a selected polypeptide or protein, e.g. Borrelia antigens OspA and OspB, other

Borrelia antigens, and proteins or polypeptides derived from other microorganisms.

Any antigen of the present invention may be used in the form of a pharmaceutically acceptable salt.

Suitable acids and bases which are capable of forming salts with the polypeptides of the present invention are well known to those of skill in the art, and include inorganic and organic acids and bases.

Antibodies

The present invention also provides antibodies capable of recognizing and binding naturally-occurring antigens of this invention from the B. burgdorferi pathogen, when it is present in a biological fluid. For purposes of this invention, the term "biological fluid" includes blood, plasma, serum, tears, saliva, urine, vaginal secretions, synovial fluid bladder wall, or ear puncture. These antibodies are useful in diagnosis of

Lyme disease and in therapeutic compositions for treating humans and/or animals that test positive for, or, prior to testing, exhibit symptoms of, Lyme Disease. The antibodies are useful in diagnosis alone or in combination with antibodies to other antigens of this invention as well as antibodies to other known B.

burgdorferi antigens. These antibodies may be generated by conventional means utilizing the isolated antigens of this invention. For example, polyclonal antibodies may be generated by conventionally stimulating the immune system of a selected animal or human with one or more of the isolated antigens, allowing the immune system to produce natural antibodies thereto, and collecting these antibodies from the animal or human's blood or other biological fluid.

Table VI below provides a list of monoclonal antibodies of the invention. However, it is also desirable to obtain and utilize other monoclonal antibodies (MAb) for the practice of the methods of this invention.

Therefore, hybridoma cell lines expressing desirable MAbs may be generated by well-known conventional techniques, e.g. Kohler and Milstein and the manner known modifications thereof. Similarly desirable high titer antibodies may also be generated by applying known recombinant techniques to the monoclonal or polyclonal antibodies developed to these antigens [see, e.g., PCT Patent Application No. PCT/GB85/00392; British

Patent Application Publication No.Go2188638; Amit et al., Science, 233:747-753 (1986); Queen et al., Proc.

Nat'l. Acad. Sci. USA, 86:10029-10033 (1989); PCT Patent

Application No.PCT/WO9007861;-and Riechmann et al.,

Nature, 332:323-327(1988)). Alternatively, the antigens may be assembled as multi-antigenic complexes [see, e.g.,

European Patent Application 0339695, published November 2, 19893 and employed to elicit high titer antibodies capable of binding the selected antigen as it appears in the biological fluids of an infected animal or human.

For use in diagnostic assays, the antibodies may be associated with conventional labels which are capable, alone or in concert with other compositions or compounds, of providing a detectable signal. Where more than one antibody is employed in a diagnostic method, the labels are desirably interactive to produce a detectable signal. Most desirably, the label is detectable visually, e.g. colorimetrically. A variety of enzymesystems-have been described in the art which will operate to reveal a colorimetric signal in an assay. As one example, glucose oxidase (which uses glucose as a substrate) releases peroxide as a product. Peroxidase, which reacts with peroxide and a hydrogen donor such as tetramethyl benzidine (TMB) produces an oxidized TMB that is seen as a blue color.Other examples include horseradish peroxidase (HRP) or alkaline phosphatase (AP), and hexokinase in conjunction with glucose-6phosphate dehydrogenase which reacts with ATP, glucose, and NAD+ to yield, among other products, NADH that is detected as increased absorbance at 340 nm wavelength.

Other label systems that may be utilized in the methods of this invention are detectable by other means, e.g., colored latex microparticles [Bangs Laboratories,

Indiana] in which a dye is embedded may be used in place of enzymes to form conjugates with the antibodies and provide a visual signal indicative of the presence of the resulting complex inapp1lic�ble assays, fiuorescent compounds, radioactive compounds or elements, or immunoelectrodes. Detectable labels for attachment to antibodies useful in diagnostic assays of this invention may be easily selected from among numerous compositions known and readily available to one skilled in the art of diagnostic assays. The methods and antibodies of this invention are not limited by the particular detectable label or label system employed.

Diagnostic Methods and Assays

The present invention also provides methods of diagnosing Lyme disease. These diagnostic methods may be useful in treating humans or animals exhibiting the clinical symptoms of, or suspected of having, Lyme disease.

In one embodiment, this diagnostic method involves detecting the presence of naturally occurring anti-B. burgdorferi antibodies which are produced by the infected human or animal patient's immune system in its biological fluids, and which are capable of binding to the antigens of this invention or combinations thereof, optionally including P39 protein. This method comprises the steps of incubating at least one B. burgdorferi antigen of this invention and optionally at least one other antigen from B. burgdorferi with a sample of biological fluids from the patient. Antibodies present in the fluids as a result of B. burgdorferi infection will form an antibody-antigen complex with the antigen.

Subsequently the reaction mixture is analyzed to determine the presence or absence of these antigenantibody complexes. The step of analyzing the reaction mixture comprises contacting the reaction mixture with a labeled specific binding partner for the antibody. In a preferred embodiment, the method entails conjugating latex beads to the isolated antigens of this invention.

Subsequently, the biological fluid is incubated with the bead/protein conjugate, thereby forming a reaction mixture. The reaction mixture is then analyzed to determine the presence of the antibodies.

In another embodiment, the diagnostic method of the invention involves detecting the presence of the naturally occurring antigen itself in its association with the B. burgdorferi pathogen in the biological fluids of an animal or human infected by the pathogen. This method includes the steps of incubating an antibody of this invention (e.g. produced by administering to a suitable human and/or animal at least one Borrelia burgdorferi antigen of this invention and optionally at least one other B. burgdorferi antigen), preferably conventionally labelled for detection, with a sample of biological fluids from a human or an animal to be diagnosed. In the presence of B. burgdorferi infection of the human or animal patient, an antigen-antibody complex is formed (specific binding occurs).

Subsequently, excess labelled antibody is optionally removed, and the reaction mixture is analyzed to determine the presence or absence of the antigen-antibody complex and the amount of label associated with the antibody.

Assays employing a protein antigen of the invention can be heterogenous (i.e., requiring a separation step) or homogenous. If the assay is heterogenous, a variety of separation means can be employed, including centrifugation, filtration, chromatography, or magnetism.

One preferred assay for the screening of blood products or other physiological or biological fluids is an enzyme linked immunosorbent assay, i.e., an ELISA.

Preferably, the ELISA is an antigen-capture two-site

ELISA capable of recognizing many strains of B.

burgdorferi and capable of distinguishing B. burgdorferiinfected from uninfected I. dammini (larvae, nymphs, or adults), using a third of a tick extract, thereby leaving adequate materials for confirmational testing.

However, typically in an ELISA, the isolated antigen of the invention is adsorbed to the surface of a microtiter well directly or through a capture matrix (i.e., antibody). Residual protein-binding sites on the surface are then blocked with an appropriate agent, such as bovine serum albumin (BSA), heat-inactivated normal goat serum (NGS), or BLOTTO (a buffered solution of nonfat dry milk which also contains a preservative, salts, and an antifoaming agent). The well is then incubated with a biological sample suspected of containing specific anti-B. burgdorferi antibody. The sample can be applied neat, or more often, it can be diluted, usually in a buffered solution which contains a small amount (0.1-5.0% by weight) of protein, such as

BSA, NGS, or BLOTTO.After incubating for a sufficient length of time to allow specific binding to occur, the well is washed to remove unbound protein and then incubated with labeled anti-human immunoglobulin (aHuIg). The label can be chosen from a variety ofenzyr-es, including horseradish peroxidase (HRP), Pgalactosidase, alkaline phosphatase, and glucose oxidase, as described above. Sufficient time is allowed for specific binding to occur again, then the well is washed again to remove unbound conjugate, and the substrate for the enzyme is added. Color is allowed to develop and the optical density of the contents oft well is determined visually or instrumentally.

Further, the MAbs or other antibodies of this invention which are capable of binding to the antigen(s) can be bound to ELISA plates. In another diagnostic method, the biological fluid is incubated on the antibody-bound plate and washed. Detection of any antigen-antibody complex, and qualitative measurement of the labelled MAb is performed conventionally, as described above. Such an "antigen capture" assay has been developed for the detection of B. burgdorferi in ticks using monoclonal antibodies specific for OspA and is currently the preferred method of detecting the presence of B. burgdorferi in biological samples according to this invention. This assay is described in more detail in Example 14 below. Briefly, the assay uses monoclonal antibodies for antigen capture and labelled rabbit polyclonal anti-B. burgdorferi antibodies for signal generation.The assay has a sensitivity of between about 50 to about 200 B. burgdorferi spirochetes, more particularly about 100 spirochetes. Infections in nymphs and adults can be determined using less than25% of a tick, thereby allowing confirmational testing.

Other useful assay formats include the filter cup and dipstick. In the former assay, an antibody of this invention is fixed to a sinter glass filter to the opening of a small cap. The biological fluid or sample (5 mL) is worked through the filter. If the antigen is present (i.e., B. burgdorferi infection), it will bind to the filter which is then visualized through a second antibody/detector. The dipstick assay involves fixing an antigen or antibody to a filter, which is then dipped in the biological fluid, dried and screened with a detector molecule.

It should be understood by one of skill in the art that any number of conventional assay formats, particularly immunoassay formats, may be designed to utilize the isolated antigens and antibodies of this invention for the detection of B. burgdorferi infection in animals and humans. This invention is thus not limited by the selection of the particular assay format, and is believed to encompass assay formats which are known to those of skill in the art.

Diagnostic Kits

For convenience, reagents for ELISA or other assays according to this invention may be provided in the form of kits. Such kits are useful for diagnosing infection with B. burgdorferi in a human or an animal sample which contains at least one B. burgdorferi antigen of this invention and/or at least one antibody capable of binding at least one B. burgdorferi antigen of this invention.

These kits can includemicroliter plates to which the B. burgdorferi antigen proteins or antibodies of the invention have been pre-adsorbed, various diluents and buffers, labeled conjugates for the detection of specifically bound antigens or antibodies, and other signal-generating reagents, such as enzyme substrates,factors and chromogens. Other components of these kits can easily be determined by one of skill in the art.

Such components may include polyclonal or monoclonal capture antibodies to a B31 non-MHC-restricted, JD-1 crossreactive and non-MHC restricted, antigen of this invention, e.g.,P18, P39 or P43; to a B31 non-MHC restricted, JD-1 MHC-restricted antigen,P17; or a cocktail of two or more of the antibodies, purified or semi-purified extracts of these antigens as standards,

MAb detector antibodies, an anti-mouse or anti-human antibody with indicator molecule conjugated thereto, an

ELISA plate prepared for absorption, indicator charts for colorimetric comparisons, disposable gloves, decontamination instructions, applicator sticks or containers, and a sample preparator cup. B31 MHC restricted antigens, such as those listed in Table I above, may also form part of this kit.Suitable exemplary monoclonal antibodies of the invention are provided in Table VI. These kits provide a convenient, efficient way for a clinical laboratory to diagnose B.

burgdorferi infection.

Therapeutic Compositions

The antibodies of the invention, alone or in combination with other Borrelia antigens such as anti-39 kD, anti-OspA or anti-OspB, may further be used in therapeutic compositions and in methods for treating humans and/or animals infected with B. burgdorferi. Such a therapeutic composition may be formulated to contain a carrier or diluent and one or more of the antibodies of the invention (see Table VI). Suitable pharmaceutically acceptable carriers facilitate administration of the proteins but are physiologically inert and/or nonharmful.

Carriers may be selected by one of skill in the art. Exemplary carriers include sterile saline, lactose, sucrose, calcium phosphate,. gelatin, dextrin, agar, pectin, peanut oil, olive oil, sesame oil, and water.

Additionally, the carrier or diluent may include a time delay material, such as glycerol monostearate or glycerol distearate alone or with a wax. -In addition; slow release polymer formulations can be used.

Optionally, this composition may also contain conventional pharmaceutical ingredients, such as preservatives, or chemical stabilizers. Suitable ingredients which may be used in a therapeutic composition in conjunction with the antibodies include, for example, casamino acids, sucrose, gelatin, phenol red, N-Z amine, monopotassium diphosphate, lactose, lactalbumin hydrolysate, and dried milk.

Alternatively, or in addition to the antibodies of the invention, other therapeutic agents useful in treating Lyme disease, e.g., antibiotics or immunostimulatory agents, antagonists to receptor mediated activity of antigens of this invention and cytokine regulation'elements are expected to be useful in reducing and eliminating disease symptoms. Agents which can be used to suppress or counteract the immune suppressants released by the tick vector should act to assist the natural immunity of the infected human or animal. Thus, such agents operate in concert with the therapeutic compositions of this invention. The development of therapeutic compositions containing these agents is within the skill of one in the art in view of the teaching of this invention.

According to the method of the invention, a human or an animal may be treated for B. burgdorferi infection by administering an effective amount of such a therapeutic composition. Preferably, such a composition is administered parenterally, preferably intramuscularly or subcutaneously. However, it may also be formulated to be administered by any other suitable route, including orally or topologically.

A therapeutic composition of the invention may contain between about 0.05Ag/mL to about 1000 g/mL of an antibody of the invention. Such a composition may be administered 1 - 3 times per day over a 1 day to 12 week period. However, suitable dosage adjustments may be made by the attending physician or veterinarian depending upon the age, sex, weight and general health of the human or animal patient.

Vaccine Compositions

In another aspect, the present invention provides a vaccine composition useful in protecting against Lyme disease associated with B. burgdorferi and a prophylactic method entailing administering to an animal or human an effective amount of such a composition. This vaccine composition may contain one or more of the antigens of the invention, or mixtures of two or more of these isolated proteins, or combinations of these antigens with other antigens of B. burgdorferi, such as the 39kD protein, the OspA and OspB proteins and a pharmaceutically acceptable carrier or diluent.

Exemplary carriers are as described above.

Optionally, the vaccine composition may further contain adjuvants, preservatives, chemical stabilizers, or other antigenic proteins. Typically, stabilizers, adjuvants, and preservatives are optimized to determine the best formulation for efficacy in the target human or animal. Suitable exemplary preservatives include chlorobutanol, potassium sorbate, sorbic acid, sulfur dioxide, propyl gallade, the parabens, ethyl vanillin, glycerin, phenol, and parachlorophenol.

One or more of the above described vaccine components may be admixed or adsorbed-with a conventional adjuvant. The adjuvant is used as a non-specific irritant to attract leukocytes or enhance an immune response. Such adjuvants include, among others, mineral oil and water, aluminum hydroxide, Amphigen, Avridine,

Ll2l/squalene, D-lactide-po D-lactide-polylactide/glycoside, pluronic plyois, muramyl dipeptide, killedBordetella, and saponins, such as Quil A.

Suitable amounts of the active ingredient can be determined by one of skill in the art based upon the level of immune response desired. In general, however, the vaccine composition contains between 1 ng to 1000 mg antigen, and more preferably, 0.05pg to 1 mg per mL of antigen. In addition, a vaccine composition of the invention may further comprise other, non-B. burgdorferi antigens, including, Bordetella bronchiseptica, canine parvovirus, canine distemper, rabies, Leptosporidia, canine coronavirus, and canine adenovirus. Other vaccinal antigens originating from other species animals may also be included in these compositions, e.g., feline coronavirus, etc.

Suitable doses of the vaccine composition of the invention can be readily determined by one of skill in the art. Generally, a suitable dose is between 0.1 to 5 mL of the vaccine composition. Further, depending upon the human patient or the animal species being treated, i.e. its weight, age, and general health, the dosage can also be determined readily by one of skill in the art.

In general, the vaccine will be administered once on a seasonal basis. Each tick season, usually in the spring, a booster should be administered. The vaccine may beadmiriistered by any suitable route.

However, parenteral administration, particularly intramuscular, and subcutaneous, is the preferred route.

Also preferred is the oral route of administration.

The following examples illustrate the preferred methods for obtaining protein antigens of the invention and preparing the assays and compositions of the invention. Significantly, these examples indicate that the antigens of this invention have the potential to be more accurate diagnostic reagents than those presently in use. The P17, P18, P39, and P43 proteins induced a non

MHC restricted response (with respect to B31 at least) in these examples and were crossreactive between at least two strains of Borrelia burgdorferi. A P39 protein has already been cloned and a monoclonal antibody specific for P39 has been generated (Simpson and Schwan, unpublished). The results of the following examples show that isolated antigens of this invention, P17, P18 and

P43, as well as P39, have diagnostic potential for Lyme disease and may improve Lyme serology.These examples are illustrative only and do not limit the scope of the invention.

Example 1 - Bacterial Strains and Their Growth

The JD1 strain of B. burgdorferi [J. Piesman et al., J. Clin. Microbiol., 25:557-558 (1987) and T. G.

Schwan et al., J. Clin. Microbiol., 27:1734-1738 (1989)] was maintained by alternating B. burgdorferi-infected I.

dammini tick to hamster to tick passage for four repetitions in golden Syrian hamsters (SASCO, Omaha, NE).

B. burgdorferi organisms were cultured at 340C in a modified BSK II culture media [R. J. Sinsky et al..,

J. Clin.Microbiol.-, 27:1723-1727 (1989)]. This BSK II broth was further modified by adding 20 mg of cycloheximide per liter to selectively inhibit fungal contamination of cultured hamster ear biopsies.

The outer tick surfaces were first decontaminated with soaks in Wescodyne (Amsco Medical

Products, Erie, PA) and 70% ethanol, followed by soaking in BSKII medium. Homogenates of infected Ixodes dammini nymphs, 2 to 3 weeks post-repletion, were then prepared by grinding 10 to 20 nymphs in 1 to 2 ml of phosphatebuffered saline (PBS) in a Ten Broeck homogenizer.

Hamsters were inoculated i.p. or i.d., at multiple sites, with 0.4 ml of homogenate.

Example 2 - Hamster Infection

For the studies with hamsters, the animals were exposed by inoculation to homogenates of cultured B.

burgdorferi-infected ticks, or exposed to B. burgdorferi by tick feeding. Hamsters were infected by tick feeding with B. burgdorferi-infected I. dammini as previously described [J. Piesman et al., J. Clin. Microbiol., 25:557-558 (1987)]. For needle inoculations, cultured organisms were suspended by gentle vortexing, enumerated by dark-field microscopy, and diluted appropriately in

BSK II medium. For initial studies, hamsters were inoculated i.p. or i.d., at multiple sites, with 0.7 ml of BSK II containing a standardized inoculum. For the dose-response study, hamsters were inoculated i.d. with 0.1 ml of varying concentrations of organisms (1 x104 to 1 x 108cells/ml).

Example 3 - Enzvme-Linked Immunosorbent Assays (ELISA)

The following assay was used with both the hamster studies of Examples 5 through 8 and the mice studies of Examples 9 through 13.

The ELISA assay was performed essentially as described in J. T. Roehrig et al.,Viroloav, 118:269-278 (1982) Briefly, antigens were coated on Immulon 2 plates (Dynatech, Inc., Kensington, MD) overnight at4 C at a concentration of 2pg/well for whole-cell sonicate, 60 ng/well for purified flagella, or 1pg/well for recombinant OspA. The whole-cell sonicate was prepared according to H. Russell et al., J. Infect. Dis., 149:465 470 (1984), and periplasmic flagella were prepared using the procedure of A. G. Barbour et al., Infect. Immun., 52:549-554 (1986). The purified recombinant OspA was a gift from Dr. John Dunn, Brookhaven National Laboratory [J. J. Dunn et al., Protein Expression and Purification, 1:159-168 (1991)].

ELISA antigen concentrations were determined in a standard box titration of antigen with positive or negative control sera. The antigen concentrations used were the minimum amounts demonstrating maximal binding of positive control sera throughout the antisera dilution range. After coating with antigen, wells were then blocked for 1 hour at 370C with 3% goat serum in PBS.

Plates were rinsed with PBS with 0.5% Tween 20 (rinse buffer). Bound hamster antibody was detected with goat anti-hamster IgG (H+L) peroxidase diluted 1:20,000 or 1:40,000 (Cappel Laboratories, Malvern, PA). Substrate used to detect bound conjugate was 3,3',5,5'tetramethylbenzidine.(TMB)/H2O2 [T. F. Tsai et al., J.

Clin. Microbiol., 26:2620-2625 (1988)]. Reactivity was stopped after 30rin by adding 50pl/well of 2N H2SO4.

Absorbance was measured at 450 nm in a Titertek Multiskan

MC spectrophotometer (ICN Flow Laboratories, Costa Mesa,

CA).

Example 4 - Immunoblots

The following procedure was used with the hamster studies of Examples 5 through 8 and the mice studies of Examples 9 through 13.

Immunoblots were modified from previously published procedures [J. T. Roehrig et al.,Virology, 142:347-356 (1985) and H. Towbin et al., Proc. Natl.

Acad. Sci. USA, 76:4350-4354 (1979)]. B. burgdorferi, strainJD1, proteins (1 x 107 cells/gel) were separated on 12.5% polyacrylamide preparative minigels [U. K. Laemmli,

Nature (London), 277:680-685 (1970)] using the highmolecular-weight BioRad Rainbow marker as standards (BioRad, Richmond, VA). The gels were stopped when the 21.5-kDa marker migrated to the bottom of the resolving gel. Molecular weight determinations were based on average gel mobilities (n = 5).

Proteins were electroblotted onto Immobilon-P membranes (Millipore Corporation, Bedford, MA) for 1 hour at 100 mA in a Semiphor blotter (Hoeffer Scientific

Instruments, San Francisco, CA). The end of the membrane was cut and used as marker following staining with Gold

Blot protein dye (Integrated Separation Systems, Hyde

Park, MA). The rest of the membrane was blocked for 1 hour at 370C or overnight at 40C with 3% goat serum in

PBS, rinsed in rinse buffer, and cut into strips. Strips were incubated for 2 hours at room temperature with rocking with hamster sera diluted in rinse buffer.

Murine monoclonal antibody (MAb) tissue culture supernatants specific for p41, H9724; OspA, H5332; and

OspB, H6831 (Symbicom, Umea, Sweden) were used at a 1:5 or 1:20 dilutions. Bound hamster antibody was detected with a goat anti-hamster IgG(H+L) peroxidase-labeled antibody (Cappel Laboratories, Malvern, PA). Bound murine MAb was detected with goat anti-mouse IgG(F) peroxidase-labeled antibody (Jackson ImmunoResearch

Laboratories, Inc., West Grove, PA). Substrate used to detect bound peroxidase conjugate was a modified histochemicalTMB/H2O2 substrate (Kirkegaard and Perry

Laboratories, Gaithersburg, MD) or a chemiluminescent substrate (Amersham International, Arlington Heights,

IL). Stained blots were allowed to develop for 1 to 5 minutes and the reaction was stopped by rinsing in water.

For the chemiluminescent substrate, the blot image was recorded by fluorography usingHyperfilmTM-ECL (Amersham

International, Arlington Heights, IL).

Example 5 - The Hamster Immune Response Followina Various

Routes of Inoculation with Various Antigen Sources

To compare the effect that route of inoculation and antigen source had on the hamster antibody response to B. burgdorferi, animals were infected either by needle inoculation or by tick feeding as described in Examples 1 and 2 above. Both the type and quantity of antigens used for infection were varied. Animals were inoculated i.p.

or i.d. with culture-grown bacteria or with homogenates of B. burgdorferi-infected I. dammini ticks. The number of infected ticks used for hamster infection was also varied to simulate a natural dose response.

One of the animals, number 3 in Table II below, was immunized with an antigen enriched for flagella prepared by the procedure of Barbour et al., Infect.

Immun., 52:549-554 (1986) prior to being fed on by B.

burgdorferi infected nymphs. The immunization protocol for this animal consisted of subcutaneous inoculation in two locations with 50 g antigen in Freund's completeaduvant. Two weeks later animals were boosted with 50ssg antigen in Freund's incomplete adjuvant. The preinfection titer of this animal as measured in wholecell sonicate ELISA was 1:320,000.

Hamsters were bled at various times postchallenge, and their antibody titers to whole-cell sonicate or to purified flagella were determined by ELISA (Table II). Efficiency of infection was monitored by isolating B. burgdorferi organisms from ear biopsies as described in Example 6 below. All hamsters were culture positive at one month post-infection(P.I.) except hamster number 27.

In Table II below, column 3 indicates whether the hamsters were infected either by feeding with B.

burgdorferi-infected I. dammini nymphs or larvae, or needle-inoculated with tick homogenates or culture-grown spirochetes; column 4 indicates number of infected ticks used for feeding or homogenate production, or concentration (log,0) of pure spirochetes; column 5 indicates number of days postinfection when serum samples were taken; and column 6 lists the reciprocal end-point

ELISA titers on whole-cell sonicate (Son) or purified flagella antigen.

Table II

Hamster Antibody Response to Infection with

B. burgdorferi by Various Routes

ELISA Titers

Route of Days ~~~~~~~~~~~~~~~

Hamster Infection Source Dose P.I. Son Flagella 3 Fed Nymph 12 29 64000 64000 30 Fed Nymph 10 93 2560 10 31 Fed Nymph 10 37 1280 40 25 Fed Nymph 12 42 640 40 32 Fed Nymph 12 43 2560 20 28 Fed Nymph 12 33 320 20 29 Fed Nymph 12 42 320 20 5 Fed Nymph 13 41 640 40 33 Fed Nymph 16 49 1280 < 10 26 Fed Nymph 17 42 320 20 34 Fed Nymph 20 49 320 20 27 Fed Larva 40 49 10 10 10 i.d. Nymph 14 35 1280 320 6 i.d. Nymph 11 39 1280 160 18 i.d. Nymph 17 33 1280 40 20 i.d. Nymph 17 33 640 40 23 i.d. Nymph 19 36 640 80 11 i.p. Nymph 14 35 1280 640 7 i.p. Nymph 11 39 2560 640 19 i.p. Nymph 17 33 1280 20 21 i.p. Nymph 17 33 2560 160 22 i.p. Nymph 19 36 80 20 24 i.p. Nymph 19 36 1280 160 13 i.d. Culture 7.0 14 2560 160 14 i.d.Culture 7.0 40 20480 3200 15 i.d. Culture 7.0 40 20480 640 9 i.d. Culture 8.0 37 10240 640 12i.p Culture 7.0 14 5120 460 16 i.p. Culture 7.0 40 20480 320 17 i.p. Culture 7.0 40 20480 800 8 i.p. Culture 8.0 37 10240 640

C-1 Control < 10 10

C-2 Control 10 10

C-3 Control < 10 < 10

C-4 Control < 10 10

Pre-immunization titers for all animals were < 40 on whole-cell sonicate, < 20 on OspA, and < 10 on flagella antigen. The serum specimens from hamsters having high whole-cell sonicate ELISA titers (3, 8, 9, 14, 15, 16 and 17) were diluted 1:100, while all other sera were diluted 1:50 to better compare antibody activity in immunoblots, which were performed as described in Example 4. Bound antibodies were detected with theTMB/H2O2 histochemical substrate. Tick-infected hamsters were bled at weekly intervals.

All hamsters inoculated with 1 x 107 or 1 x 108 cells of culture-grown JD1 produced antibody to both whole-cell sonicate and flagella in ELISA, and OspA and

OspB in immunoblot, regardless of route of inoculation (Table II). As positive antibody controls for the OspA

ELISA, hamsters 15 and 16 were also tested for antibody to recombinant OspA antigen. These hamsters had anti

OspA titers of 1:320. Low levels of anti-OspA (but no anti-OspB antibody) could occasionally be demonstrated by immunoblot in some animals inoculated with tick homogenate.

Following tick infection, however, no or very low anti-OspA or anti-OspB reactivity could be identified, regardless -of the tick dose. All naturally infected animals, and many animals inoculated with tick homogenate, failed to demonstrate anti-flagella or antip41 antibody by either ELISA or immunoblot, respectively (Table II). An antibody response to 47-, 43- (P43), 21-, and 20-kDa proteins was observed in the hamsters, regardless of the route of inoculation or the type of immunogen. The P41 and P43 could be differentiated by the selective P41 reactivity of the anti-flagellin MAb

H9274. A similar differentiation could be observed in hamster 3 which had been immunized with partially purified flagellin prior to tick challenge. In this hamster a more intense P41 band is observed in the immunoblot consistent with this animal's pre-existing immunity to flagellin.

Example 6 - Kinetics of the Hamster Immune Response

Following Infection by Tick Bite

Tick-infected hamsters failed to respond well to OspA and OspB. To better characterize the kinetics of the hamster antibody response following natural infection, the production of antibodies over time (Fig.

1) was analyzed. The efficiency of natural infection was measured by isolating spirochetes from ear biopsies [R.

J. Sinsky et al., J. Clin. Microbiol., 27:1723-1727(1989)].

Hamster ears were first disinfected with 70% ethanol, and ear wedges were cut. Ear biopsies were disinfected by soaking sequentially in Wescodyne and 70% ethanol, for 15 minutes each. Wedges (25 mm2) were minced with a scalpel, and all pieces were inoculated into 4-ml tubes containing BSK II media, and incubated at 340C.

Cultures were observed until positive, or for 28 days.

All hamsters experienced sustained infection, as demonstrated by cultural isolation of B. burgdorferi from ear biopsies sampled one month after inoculation or tick feeding, unless otherwise noted. The hamster antibody response was rapid: high antibody titers occurred one week P.I. and remained stable throughout the experiment. After peaking at one week, antibody titer to the flagellin declined and remained stable at a lower titer (Fig. 1). This result was confirmed by immunoblot reactivity with P41.

This assay pooled hamster serum samples (n=6 for weeks 1-5, n=3 for week 6, n=2 for week 19, n=l for week 42), which were diluted 1:40 prior to blotting. The chemiluminescent substrate was used to identify antibody binding.

Anti-P43 antibody was detected at one week by immunoblot and appeared to remain stable throughout the infection. Very little anti-OspA antibody and no anti

OspB antibody could be detected by immunoblot through 19 weeks P.I. At 42 weeksP.I., the single surviving hamster had very little demonstrable antibody to either

OspA or OspB. This hamster was, however, still culture positive. Antibody to the P39 antigen was not detected until 2-4 weeks P.I. Anti-P39 was still present late in infection.

Example 7 - AntibodyDose-Response to Various Amounts of

Needle Inoculated Cultured B.buradorferi

To determine whether the differences in antibody responses between tick-infected or needleinoculated hamsters were due to introduction of more antigen by needle inoculation, an antigen dose-response analysis was performed using needle inoculation of varying amounts of culture grown B. burgdorferi (3.0 to 7.0 logs0 cells). Animals were bled at 2 and 4 weeks post-immunization. Pre-immunization titers for all animals were < 40 on whole-cell sonicate and < 20 on OspA antigen.

-Maximal antibody reactivity to whole-cell sonicate antigen was detected following inoculation of doses greater than 1 x 106 of B. burgdorferi (Fig. 2).

All animals inoculated with 1 x 106 or 1 x 107 cells, were both culture positive for B. burgdorferi and seropositive in whole-cell sonicate ELISA. One of five animals inoculated with 1 x 105 cells was culture positive, but all five were seropositive in whole-cell sonicate ELISA.

Two of five animals inoculated with 1 x 104 cells were culture positive, but three animals were seropositive in whole-cell sonicate ELISA. All animals inoculated with 1 x 103 bacteria were culture negative and failed to raise detectable antibody when measured in whole-cell sonicate

ELISA. The LD50 of this experiment was 1.3 x 105 cells/animal as determined by ear biopsy isolation.

In order to illustrate the antigen specificity of B. burgdorferi-inoculated hamsters as a function of dose, hamsters were needle inoculated with 1 x10S cells or more. Because ELISA titers of individual hamsters on whole-cell sonicate were essentially identical, 4 week serum samples were pooled (n=4 or 5 animals) and diluted 1:40 before use in immunoblots. The chemiluminescent substrate was used to identify antibody binding.

These needle-inoculated animals produced high levels of anti-OspA and anti-OspB at 4 weeks P.I. Unlike the results of the immunoblot, OspA antibody as measured in OspA ELISA could be detected only after inoculation of 1 x 106 or 1 x 107 cells (Fig. 2). The immunoblot, therefore, appeared to be more sensitive in detecting

OspA antibodies than the ELISA.

Example 8 - Comparison of Antibodv Response by quantitative Densitometrv

A direct comparison of antibody responses to

P43, P41, P39, OspB and OspA was performed by comparing the immunoblot reactivities observed by quantitative densitometry. Amounts of antibody bound in immunoblots were determined by integration of peak areas from

densitometer scans using a Gilford 2600

spectrophotometer. The amount of detected anti-P43 was

standardized to 1.0. For comparison to anti-P43

antibody, the amounts of other protein specific

antibodies were normalized to the P43 value of 1.0. All

sera were used at a 1:40 dilution.

These results are illustrated below in Table

III. For animals challenged by needle inoculation the

doses shown are reported aslog,0 cells/challenge.

Table III

Amount of Antibody Detected Following Various

Routes of Inoculation

Relative amount of bound antibody to:

Weeks

Route P.I. Dose P43 P41 P39 OspB OspA

Tick 1n.a. 1.0 0.6 0 0 0.4

4 n.a. 1.0 0 0.2 0.1 0.8

42 n.a. 1.0 0 0.3 0.1 0.1

Needle 4 5.0 1.0 0 0.5 3.3 3.2

6.0 1.0 0 0.6 3.8 3.5

7.0 1.0 0 0.9 5.0 4.0

n.a. = not applicable.

The densitometry results confirmed visual

observations. At least 4-fold more anti-OspA and 30-fold

more anti-OspB were produced following needle inoculation

at 28 days P.I. The anti-P41 response at 7 daysP.1.

following natural infection was also observed, although

this response was not as high as that for anti-P43.

Similarly, the anti-P43 response was greater than the

anti-P39 response.

Example 9 - Infected Tick Colonies and Antigen

Preparation for Mice Studies

A. Maintenance of Infected Tick Colonies

Freshly hatched Ixodes dammini larvae were fed on ICR outbred mice infected with B31 strain of Borrelia burgdorferi and subsequently maintained at 40C with high humidity. Molted nymphs were used to infect mice by the natural route of tick exposure. All infected ticks were maintained by a cycle of infected ticks transmitting the spirochete to mice and these mice were used to infect the next hatch of tick larvae.

B. Preparation of Borrelia antigens

Strains B31 and JD-1, both originally isolated from I. dammini [available upon request from CDC, Fort

Collins, CO], were inoculated into culture from either low passage or high passage frozen stocks. As used throughout these examples, the term "low passage" refers to up to six passages in culture for strain B31, and a single passage in culture for strain JD-1. The term "high passage" for both strains refers to greater than 30 passages in culture for either strain.

These were expanded in BSK II medium supplemented with 6% rabbit serum (Pel-Freez Biologicals,

Rogers, AK). One liter expansion cultures were harvested in late log phase (day 5), pelleted by centrifugation at 10.240 g for 20 minutes at20cC. Spirochetes were washed with PBS, 5 mM MgCl2 and centrifuged again. Pellets were resuspended in 10 mM Tris, 1 mM EDTA (TE) at 30 ml/g wet weight of cells. The spirochetes were lysed in a Dounce homogenizer and protein concentration was adjusted to 2 mg/ml. These preparations were then mixed with equal volumes of 2X SDS-PAGE sample buffer, aliquoted, and stored at -700C.

Example 10 - Infection of Mice bv Tick Bite

Adult mice of ages 4-6 weeks were obtained from

Jackson Laboratories, Bar Harbor, MA (BALB/c, B10,B1O.BR, and B1O.D2) or from the specific pathogen free (SPF) colony of ICR outbred mice maintained at DVBID,

NCID, CDC at Fort Collins, CO. These four recombinant inbred strains and one outbred strain of mice were used as infection targets for transmission of B. burgdorferi infection by Ixodes ticks.

The mice used in this study were BALB/c (H-2d),

C57BL/1OJ (B10) (H-2b), and the recombinant inbred strains on the B10 background, B1O.BR (H-2k) and B1O.D2 (H-2d).

The BALB/c and B10 mice differ in their immunoglobulin (Ig) gene allotypes (BALB/c has Ig allotype "a"; B10 have

Ig allotype "b") and the outbred mice have mixed MHC and

Ig haplotypes.

Groups of four mice from each strain as well as

ICR outbred mice were numbered by ear punch and preimmune serum samples were drawn. Each mouse was exposed to ten to twelve B. burgdorferi, strain B31 infected ticks as described above in Example 9. Table IV shows the number of ticks that successfully attached and fed on each animal as well as the results of culturing ear punch biopsies for the Borrelia spirochete 30 days postinfection.

Table IV

Infection of inbred and outbred strains of mice with the

B31 isolate of B. burgdorferi by infected Ixodes dammini.

Strain ID Mouse No. not No. Culture

ID# Attached Engorged Positive2

ICR outbred 1 1 7 +

2 0 12 +

3 2 9 +

4 1 7 +

BALB/c 1 0 13 +

2 0 7 +

3 09. +

4 0 5 +

C57B1/10 5 3 8 + (B10) 6 0 4 +

7 1 4 +

8 3 6 +

B1O.BR 5 4 4 +

6 0 4 +

7 0 6. +

8 1 3 +

B1O.D2 3 ' 0 9 +

4 4 6 +

7 3 9 +

8 2 4 + 1 Ticks that were recovered without feeding.

2 Infection was assessed by culturing ear punch

biopsies at day 30 of infection.

Example 11 - Response of Inbred Strains of Mice to

Infection with B.burudorfer'i Resulting from Tick Bite using the Homoloaous Bacterial Strain as Antigen

Serum samples were drawn from each mouse on days 5, 12, 19, 26, and 60 post infection and these serums were used in western blot analysis using whole lysates of Borrelia burgdorferi, strain B31, low passage (80 g), as antigen.

Samples were run in a discontinuous SDS-PAGE system using a Hoefer Scientific Instruments model SE600 vertical slab gel electrophoresis system. The stacking and resolving gels were 4% and 10% acrylamide respectively. Bacterial lysates were run at 80pg/gel in a single preparative well. Molecular weight standards (14 to 106 kD) were run in flanking lanes. This was run on a 10% SDS-PAGE and electrophoretically transferred to a 0.2ssm nitrocellulose filter as follows.

Proteins were transferred to 15xl5cm nitrocellulose filters, 0.2im pore size (BioRad,

Richmond, CA) using a Hoefer Semiphor TE 70 transblot system at 100 mA for 50 minutes. Nitrocellulose filters were placed in preblocking solution containing 3%

BSA-Fraction V, 0.9% NaCl,lOmM Tris-HCl, pH 7.4, and 0.1% sodium azide for 16 hours at 40C. Filters were washed in0.05 Tween 20, 0.9% NaCl, andl0mM Tris HCl pH 7.4 (wash buffer), for 20 minutes. Filters were mounted in an Immunetics Miniblotter 25 and washed again. Wells were suctioned dry and loaded with the appropriate dilution of antisera or monoclonal antibody (MAb) in wash buffer.Primary antibody was incubated for 3 hours at room temperature, filters were washed 3X each for 5 minutes, and alkaline phosphatase conjugated Goat antimouse Ig G + M (H and L) (Jackson Immunoresearch,

Vineland, PA) at 1:1000 dilution was added to all wells.

After 1 hour, filters were washed 2X and removed from the apparatus and washed once more. Filters were quickly washed 2X with deionized water and BCIP/NBT substrate (Kirkegaard and Perry Laboratories, Inc. Gaithersburg,

MD) was added for exactly 5 minutes to detect bound alkaline phosphatase.

Preimmune sera as well as day 5 sera from all animals were negative in western blot using B.

burgdorferi strain B31. On day 12, all mice began to generate antibodies that can be found in serum to a variety of Borrelia proteins. Further, the inbred strains of mice react differentially to these antigens.

For example, the B10 animals all show a strong response to a protein migrating at a molecular weight of just above 80kD, most likely the P83 reported by Dorward et al., cited above. The balance of the animals in this study have not responded to this protein at this stage of infection.

There is a variable response to P17 antigen [described by Wilske petal., Third Intern'lSymposium of

Lvme Borreliosis, Sweden (1990)] in that only the B10 congenics respond to this protein at day 12. The P21 antigen, also described by these authors, induces a weak response in all strains with the exception of BALB/c.

Further, there appears to be a variable response to a group of proteins in the 20 to 22 kD molecular weight range as well as the 55 to 60 kD range. The antibody response is still very weak to these antigens at this early stage of the infection.

Example 12 -Earlv response induced bv proteins in the 39kD molecular weight range

There are at least two bands in the 39 kD region but this is not believed to be reactivity with the41kD flagellin protein since the MAb to flagellin detects a band that migrates just above the P39 doublet. Two other proteins migrating at 14 and 18 kD molecular weight also induce a variable response this early in infection.

Finally, there is a significant response by all mice to a protein migrating close to the 45kD marker. In Example 5 above, this protein was described in natural infection in hamsters and designated the antigen P43.

At this stage of the infection, no response to

OspA, OspB, or flagellin was detected, all proteins which induce an early response from these as well as other strains of mice when challenged by needle inoculation of cultured Borrelia burgdorferi [U. E. Schaible, et al.,

Eur. J. Immunol., 21:2397-2405 (1991), E. Fikrig, et al.,

Infection and Immunitv, 53:713-714 (1986)]. Though there is a marginal reactivity with bands at 30kD (close to

OspA) and 32kD (OspB) this banding is not stoichiometric with the Coomassie blue staining of the acrylamide gel or the Ponceau S stain of the nitrocellulose filter where

OspA and OspB are very prominent bands. This was interpreted as an antibody response to different proteins migrating coincident with OspA and OspB. Further, these sera have been analyzed by western blot and ELISA using purified, whole lipidated OspA from a cloned gene product expressed in E. coli and detect no anti-OspA activity in the sera of any mice used in this study.

By day 19 post infection, the antibody response to a number ofbacteria antigens intensified as detected on western blots of B31 low passage. The set of five bands migrating at 14, 17, 18, 21, and 22 kD were detected by sera from all five mouse strains with the exception of Bl0 mice not responding to the18kD band.

Sera from the BlO.BR mice were not detecting the 3OkD band migrating just below OspA, and there was a variable response to a group of proteins migrating coincident to

OspB. All mice responded to the P39 band(s) and one mouse, B1O.BR #5, had antibodies specific for flagellin (41kD). All mice also maintained the response to P43 and a band(s) migrated at about 52 to 55kD. B10.BR, and to a lesser degree B10.D2, detected a band slightly higher, at 58kD, and as on day 12, only the B10 animals responded to the 83kD band.

When the same sera, drawn on day 19 post infection, was used to blot B31 high passage, a very different pattern was seen. In the low molecular weight range, only outbred mice responded to P22 band and all mice except the BALB/c's responded to P17. These sera did not detect the P14, P18, and P21 antigens observed in the blots of B31 low passage. A similar reactivity to the band at 30 kD was seen with all mice responding exceptB1O.BR. No reactivity at all was detected between 31 and 39kD and all mice responded to the P39 band(s).

There was no reactivity with P41 (flagellin), P43, or any other bands in this molecular weight range seen in the western blots of B31 low passage. The response to the band at 52 to 55 kD was detectable but the B10 response to p83 was not seen in this blot.

Immunoblots performed with sera drawn on day 26 post-infection continued to show an increase in the intensity of the response to many Borrelia antigens. The strong response to the five bands in the 14 to 24kD molecular weight range was observed in all mice except for the weak response to P18 by the B10 animals. There was a reactivity to a protein at approximately 28kD that was only detected in serum from the B10.BR animals and there was still a variable and minimal response to antigens in the same molecular weight region as OspA(30kD).

The variability in the response to a group of proteins in the 32 to 34kD range, including OspB, was maintained at day 26. For example, outbred and B1O.BR mice had an antibody response to multiple bands in this region as did the B10 mice with the exception of mouse #6. The B1O.D2 and BALB/c mice (both H-2d) did not respond to these proteins with the exception of BALB/c mouse #3 which responded to a single band. Apparently these proteins were not efficiently presented by H-2d MHC proteins.

All animals made antibody specific for the P39 bands and by day 26 all the B10.BR animals were responding to P41 flagellin. As seen at earlier time points, all animals responded to P43. A pair of bands at 50 and 52 kD gave a minimal signal using these sera and all animals had antibodies to a series of proteins in the 55 to 65 kD molecular weight region that included the heat shock proteins. At this time in the course of the infection only the B10 animals were responding to P83.

The final sera for this study were drawn on day 60 post infection and immunoblots using these sera and

B31 low passage as antigen were performed by conventional techniques. At this time post infection, sera of all these animals had significant antibodies specific for all of the 5 proteins in the lower molecular weight range as discussed above. At this later stage of infection, all animals were responding to the protein migrating at approximately 28kD. The response at 30kD continued to be variable between the strains tested in this study.

The response to proteins migrating in the region of OspB had become more pronounced and consistent between strains with the exception of B10 mouse #6, as noted at earlier time points, and B10.D2 mice # 7 and 8.

Again, all animals responded to the P39 bands and animals appeared to be responding to P41 flagellin. As had been seen throughout the infection, all animals made a strong response to P43. Also at this later stage of infection all animals appeared to respond to the pair of bands at approximately 50 and 52 kD, although the B10.BR and

B10.D2 response was very weak.

There was a variability in the intensity of the signal in the response of these mice to the triplet of bands in the 55 to 65 kD molecular weight region. By day 60 post infection, all strains were generating antibody to p83 with the exception of B1O.D2. and BALB/c animals #2 and #4, all H-2d haplotypes.

When these sera were used to blot strain B31 high passage as antigen, the blotting pattern was very different. The only antigen detected in the low molecular weight region of the gel was P17. A minimal response to P22 and P24 could be seen in the outbred,

BALB/c and B1O.D2 animals and the B10 and B1O.BR were essentially negative. Because the antibody binding to

P17 in the B31 low passage and B31 high passage gel was comparable, the difference in detection of the other proteins using these sera indicated a qualitative difference in the bacterial proteins expressed by the two derivatives of B31. Presumably, after many passages in culture, these antigens were no longer expressed by the spirochete, expressed at greatly lower levels, or their antigenic structures have been modified such that these serum antibodies could no longer bind.

These sera also failed to detect the P28 antigen and showed variable binding in the OspA and OspB region. All sera again detected the P39 bands but not

P41 flagellin. There was no reactivity in the region of

P43 or the 52-54 kD range yet a similar binding pattern is seen in the region of 55 to 65 kD. The ability of these sera to bind the P83 antigen was exactly the same as in the low passage blot. In general, when comparing the immunoblots onB31 low passage and high passage using the same amount of protein antigen, sera from mice infected with B. burgdorferi by exposure to tick bite taken at both day 19 and day 60 detected many more antigens in the low passage material.In the case of the antigens detected in both lysate preparations, the signal in the low passage blots was more intense than the same reactivity detected against high passage derivative with the exception of P17 in the blots with day 60 sera as discussed above.

This all indicated that long term culture of

Borrelia burgdorferi leads to a significant change in the expression of a majority of the bacterial antigens that could be detected by murine antibodies.

Example 13 - Response to a heterologous strain of

Borreliaburodorferi

Western blot analysis of sera from five mouse strains binding B. burgdorferi, a low passage isolate of strain JD-1, was performed. The results of this analysis showed the ability of these sera to detect the same antigens from a different strain of B. burgdorferi. Sera drawn on day 12 showed no response to the five low molecular weight bands and a similar response to the band at 3OkD as with theB31 low passage material. All mice responded to the P39 bands and there was a very marginal detection of P43. There was also a good response to P58 band. Compared to the blots using these sera on the low passage isolate of strain B31, fewer bands were detected from JD-1 and the bands that were detected show less intensity.

Sera from day 19 post infection showed a little more intensity of signal to the bands detected by day 12 sera with the only addition that some variable reactivity to P17 and P18 was then detectable. Using day 26 sera, the reactivity to P17 by all animals exceptB10 was apparent. The response to P18 was more consistent across these mouse strains butB10 animals had a marginal response and serum from B10.BR #6 did not detect this antigen. There was a variable detection of an antigen migrating just below OspA and all sera detected the P39 bands. All of these sera gave a good signal to P43 and the reactivity to the 58 to 62 molecular weight antigens was more pronounced. As in the case ofB31 strain used as antigen, only sera fromB10 animals detected P83.

Sera from day60 post infection blotted against this low passage isolate of strain JD-1 showed a consistent crossreactivity with the P17 and P18 bands but no such binding to P14, P22, and P24 was detectable.

There was a tremendous variability to all of the bands migrating between 28 and 35 kD and as with earlier bleeds, all sera detected the P39 doublet as well as the

P43 antigen. The reactivity to the band at 58kD was consistent with all sera and the bands above that were variably detected. Specifically, P83 was detected by allB10 and B10.BR sera, variably by BALB/c and outbred sera, and not at all by B10.D2. As in the case of using high passage strainB31 as antigen, all of these blots showed less intensity of signal even though they were done with the same serum samples on the equivalent amount of protein antigen.

The only prominent reactivity detected on sera drawn on day 60 post infection on a high passage isolate of JD-1 was anti-P39 response. There was marginal detection of the band at 58 kD by all sera and only

BALB/c and outbred mice detected anyP18. There was a barely detectable binding of serum antibodies fromB10 animals for the P83 band. This striking result indicated how different a high passage isolate of a different strain, JD-1, was from the infective agent used in this study, strainB31 in infected ticks.

Example 14 - Antigen Capture ELISA

The antigen capture ELISA was evaluated by varying a number of different parameters in the test including plates, both polystyrene (Immulon II) and polyvinyl chloride plates (Dynatech, Costar, catalogue numbers 2597 and 2787). Blocking reagents tested included BSA (1%, 1.5%, 2.5%, 5%), Casein (0.5%, 1%), 0.5% Boiled Casein and 0.5% Boiled Casein with 0.05%

Tween 20; powdered skim milk (2.5%, 5%) with or without 0.05% Tween 20.

Monoclonal antibodies tested for antigen capture included antibodies against OspA (184.1 and

H5332), OspB [H6831], the 41 kd flagellin (H9724), the p60 molecule (149.1). These antibodies are available commercially from Symbicom (Umea, Sweden).

Polyclonal antisera against B. burgdorferi made in rabbits and hyperimmune ascites made in mice were biotin-labelled according to the manufacturer's recommendations (Amersham) and evaluated for signal recognition together with both streptavidin-horseradish peroxidase conjugates and streptavidin-alkaline phosphatase conjugates with ABTS and TMB as substrates for horseradish peroxidase and Sigma 104-105 as substrate for alkaline phosphatase.

The protocol for the optimal antigen capture

ELISA was as follows. Wells of polyvinylchloride plates (Costar) were coated with 50pl of protein-A purified Mab

H5332 at a concentration of 10pg/ml in phosphate buffered saline, pH 7.4 (PBS). Following an overnight incubation at 40C, wells were washed 3 times with PBS with 0.05% Tween 20 (PBST) and 200pl of 2.5% skim milk in PBS with 0.05% Tween 20 were added to each well for one hour at room temperature.

Laboratory strains of Borrelia to be tested in the assay were grown in BSKII and washed 3 times with PBS prior to counting. Borrelia to be tested were diluted in 0.25% nonidet P-40 in PBS (NP40-PBS) for 10 minutes prior to a 5 fold addition of 1% bovine serum albumen in PBS (BSA-PBS). Ixodes dammini nymphs to be assayed were homogenized in 20Ml of NP40-PBS in microtubes (Biorad) using a pestle shaped to fit the tube and attached to a power drill. Homogenization was facilitated by the addition of several grains of sterile sand to the microtube. 1301 of BSA-PBS was then added to nymph extracts. Adult I. dammini were homogenized in 50ijl of

NP40-PBS as described and 250 1 of BSA-PBS were added.

Following the blocking step, wells were emptied by aspiration and 50i1 of Borrelia spp or tick extracts to be tested were added to individual wells for 2 hours at room temperature (RT). Wells were then washed 3 times with PBST and 50pl of biotin-labelled polyclonal rabbit serum at a.concentration of 2.5Mg/mL was added to each well for one hour at RT. Following an hour incubation at

RT, streptavidin-horseradish peroxidase (Amersham) at a 1:3000 dilution in BSA-PBS was added to each well.

Following an hour incubation at RT, wells were washed 3 times as described and 100l of ABTS (Kirkegaard and

Perry Labs) was added. Plates were read at 405 nm after an hour incubation.

Specificity of the ELISA was evaluated against 20 strains of Borrelia burgdorferi. These strains are

B31,WC-H1, NY14, NY-944, JDl, HB19, GA-747, PA606,

PA622, PA624, WI-206, WI-210, CA4, CA5, CA-808, CA1238,

UK814, IP90, CH2223, CH2246. B. hermsii, B. turicatae,

B. coriaceae, B. parkeri were also evaluated. In addition, uninfected I. dammini, I. scapularis, I.

pacificus, I. cookei, I. kingi or Ambylomma americanum were tested in the assay.

Sensitivity of the assay was evaluated against standard curves of Borrelia spp starting with a maximum concentration of 5,000 per well.

I. dammini from laboratory colony 2324, originally isolated from Massachusetts [CDC, Fort

Collins, CO] were tested as nymphs and adults, either uninfected or infected with the B31, JD1 or HFAI strains of B. burgdorferi. Replete and flat nymphs were homogenized after storage at -200C, in 70% ethanol, dried or alive.

The antigen capture ELISA was found to recognize all North American stains of B.bussgdorferi tested. The only exception was a Californian strain(CA1238) which was in its 52nd passage in the laboratory.

Coomassie blue stains revealed that this isolate no longer produces OspA. In addition one strain, WI-206, gave a diminished signal. However, the ELISA was still sensitive enough to detect infected I. dammini. The assay was unable to detect B. burgdorferi isolates from the United Kingdom, Russia and one of two Chinese isolates tested. The assay did not recognize other

Borrelia species tested. The sensitivity of the assay is approximately 50-200 spirochetes.

The ELISA was able to distinguish B.

burgdorferi-infected from uninfected I. dammini (larvae, nymphs or adults) using a third of a tick extract, thereby leaving adequate material for confirmational testing. The assay did not recognize tick antigens of I.

dammini, I. scapularis, I. pacificus, I. kingi or

Ambylomma americanum. Infected I. dammini nymphs were estimated to harbor between 1000 and 10000 B. burgdorferi spirochetes. In addition to live ticks, the assay was able to detect B. burgdorferi in ticks that had been stored frozen, dried at room temperature, or in70t ethanol.

Example 15 - Cure andRechallense Experiment

As a preliminary to vaccine trials using any B.

burgdorferi antigen, a study was performed to investigate whether previous infection with a given strain of

Borrelia burgdorferi would protect against a second exposure to the same strain of Borrelia. This "cure and rechallenge" experiment was carried out using the same batch of B31 infected Ixodes dammini nymphal ticks to transmit infection.

For this preliminary experiment, four strains of outbred mice, BALB/c, B10, and C3H were used. These animals were infected by placing up to ten B31 infected ticks on each animal. Three weeks after tick infestation, the mice were assayed for infection by taking an ear punch biopsy and culturing the tissue in

BSKII media for spirochetes. After one week, all mice were positive for Borrelia by culture, indicating a systemic infection in the corresponding mouse. Infected mice were then treated with tetracycline at 1 mg/mL in the drinking water for two weeks. Tetracycline water was changed three times per week because of the light sensitivity of the antibiotic.

After the antibiotic treatment, a second ear punch biopsy was taken to demonstrate that mice cultured negative, indicating that they had been cured of the infection. These mice were then rechallenged by infesting them with ten B31 infected ticks. After the ticks had fed to repletion, they were collected, dissected, and assayed for the presence of spirochetes by immunofluorescence using polyclonal antisera.

The mice were assayed for infection three weeks after the second tick challenge. This was done as before by taking an ear punch biopsy and culturing the tissue in

BSKII media. The results of this experiment are shown in

Table V below. The results indicated that reinfection is likely even though the mice were infected with the same strain of Borrelia. Further, mice that did not get reinfected showed 40% or more of the feeding ticks were cleared of the infection, presumably as a result of feeding on an immune animal. This conclusion is based on the observation that no significant clearing was observed in the ticks that fed on age matched naive control mice and all of those animals became infected.

TABLE V

Rechallenge

Outbred Ticks Ticks Ear Biopsy

Mouse No.enor9ed infected (%) Culture

2 8 7 (87.5) +

3 7 4 (57.1)

4 9 5 (55.6)

5 7 6 (85.7) +

C3H Ticks Ticks Ear Biopsy

Mouse No.engorged infected (%) Culture

1 7 7 (100) +

3 9 3 (33.3)

4 5 5 (100) +

5 9 8(88.8). +

C3H Naive Ticks Ticks Ear Biopsy

Controlsengoraed infected(%) Culture

1 9 8 (88.8) +

2 9 7 (77.7) +

BALB/c Ticks Ticks Ear Biopsy

Mouse No. engorged infected (%) Culture

6 4 4 (100) +

7 6 6 (100) +

8 6 6 (100) +

9 5 4 (100) 1 dead +

10 4 4 (100) +

Naive Ticks Ticks Ear Biopsy

BALB/c engorged infected (%) Culture

1 2 1 (100) 1 dead dead

2 2 1 (100) 1 dead dead

Example 16 - Monoclonal Antibodies

The source of B cell blasts for B cell fusions for monoclonal antibodies to antigens of Borrelia burgdorferi were spleen cells from mice infected with B.

burgdorferi by exposure to infected ticks. These mice were shown to be infected, cured by administering tetracycline (see Example 15), and exposed a second time to infected ticks. The spleen cells were harvested five to seven days after the second exposure (post tick drop off).

Two B cell fusions have been completed and approximately 1000 independent B cell hybridomas have been screened. The first fusion used immune B cells from a BALB/c mouse fused to the B cell lymphoma, P3X63Ag8.

This fusion was named BB31 for BALB/c anti-B31. Six monoclonal antibodies have been isolated from this fusion. The second fusion was done by the same protocol using a C57B1/10 (B10) mouse as a source of immune B cells. This fusion was designated B10B31 for B10 anti

B31. From this fusion five more monoclonal antibodies have been isolated. A summary of these antibodies and their specificity is given in Table VI.

TABLE VI

Monoclonal Antibodv Molecular Weight of Detected Band

BB31-42.1 75 kD

BB31-109.1 30 kD

BB31-77.9 35 kD

BB31-290.1 22 kD

BB31-292.9 18 kD

BB31-426.11 52 and 55 kDB10B31-64.4 36 kDB10B31-103.1 55 kDB10B3 1-158.6 70 kDB10B31-221.1 17 kD

Numerous modifications and variations of the present invention are included in the above-identified specification and are expected to be obvious to one of skill in the art. Such modifications and alterations to the compositions and processes of the present invention are believed to be encompassed in the scope of the claims appended hereto.