Use of recombinant antigens of Borrelia burgdorferi and Anaplasma phagocytophilum in enzyme-linked immunosorbent assays to detect antibodies in white-tailed deer

Serum samples obtained from white-tailed deer (Odocoileus virginianus) in Connecticut (n=218) and South Carolina (n=20) (USA) during the period 1992-2002 were analyzed for antibodies to whole-cell or recombinant antigens (i.e., fusion proteins) of Borrelia burgdorferi sensu stricto and Anaplasma phagocytophilum, etiologic agents of Lyme borreliosis and granulocytic ehrlichiosis, respectively. In enzyme-linked immunosorbent assays (ELISAs) with whole-cell B. burgdorferi, the overall seropositivity rate for Connecticut (53%) exceeded that for South Carolina (30%). In separate tests of seven recombinant antigens of B. burgdorferi by an ELISA, seroprevalence for the VlsE antigen was highest (48%) in Connecticut followed by outer surface protein (OspF) (21%), whereas serum reactivities to the protein (p) 41-G antigen (55%) and VlsE (25%) were most frequent for South Carolina sera. In analyses for antibodies to the recombinant protein (p) 44 antigen of A. phagocytophilum, seroprevalences of 52% and 25% were recorded for Connecticut and South Carolina samples, respectively. These findings paralleled those determined by indirect fluorescent antibody staining methods with whole cells (43% and 30%). Moreover, there was good agreement (74%) in results of Western blot analyses and an ELISA when a subset of 39 sera was screened with whole-cell or recombinant p44 antigens of A. phagocytophilum. An ELISA with highly specific recombinant VlsE or p44 antigens can be used in conjunction with other antibody tests to determine whether deer living in different regions of eastern United States were exposed to B. burgdorferi or A. phagocytophilum.     

Antibodies to whole-cell or recombinant antigens of Borrelia burgdorferi, Anaplasma phagocytophilum, and Babesia microti in white-footed mice

Serum samples were obtained from white-footed mice (Peromyscus leucopus) in tick-infested areas of Connecticut during the period 2001 through 2003 and analyzed for antibodies to Borrelia burgdorferi, Anaplasma phagocytophilum, and Babesia microti. Emphasis was placed on the evaluations of highly specific recombinant VlsE or protein (p) 44 antigens of B. burgdorferi and A. phagocytophilum, respectively, in a newly developed enzyme-linked immunosorbent assay (ELISA) as well as testing sera with whole-cell antigens by conventional ELISA or indirect fluorescent antibody staining methods. Of the 414 mouse sera analyzed, 310 (75%) had antibodies to whole-cell B. burgdorferi, whereas 157 (38%) were positive to the VlsE antigen. The latter nearly equaled the overall antibody prevalence rate (37%) computed when sera were tested separately with the p44 antigen. Mice were exposed to these pathogens and B. microti (antibody prevalence = 25%) in extreme northern Connecticut as well as the southern coastal areas of the state, thus indicating further geographic expansion of these infections. Fifty-three (13%) sera from widely separated sites had antibodies to all three pathogens. With expression and immunological recognition of VlsE and p44 antigens in P. leucopus, separate incorporation of these fusion proteins in an ELISA was very helpful in confirming past or current infections and in identifying specific foci for B. burgdorferi and A. phagocytophilum.     

Comparative evaluation of two enzyme linked immunosorbent assay methods and three Western Blot methods for the diagnosis of culture-confirmed early Lyme borreliosis in Italy

This study investigated the onset and development of the immune response to Borrelia burgdorferi infection in 30 Italian patients with culture-confirmed Lyme Borreliosis in the stage of erythema migrans (EM). All patients received antimicrobial treatment when entering the study and were prospectively evaluated monthly for up to 30 days after enrolment. A total of 60 serially collected serum samples were tested by using two different commercial enzyme-linked immunosorbent assays (ELISAs): Anti-Borrelia plus VlsE ELISA, Euroimmun, and the synthetic peptide-based ELISA, Quick ELISA C6, Immunetics. Sixty-five potentially cross-reacting sera were also tested. Anti-Borrelia plus VlsE ELISA IgG was far more sensitive than Quick ELISA C6 (56.6% and 33.3%, respectively). Moreover, considering that 17 additional sera from the first bleeding group of Lyme disease patients were IgM positive when tested by Anti-Borrelia plus VlsE IgM, the sensitivity of Anti-Borrelia plus VlsE as a whole system rose to 85.0%. Nevertheless, due to the specificity values of Anti-Borrelia plus VlsE ELISA identified in this study (98.5% for IgG and 78.5% for IgM), the need of a confirmatory test for the diagnosis of Lyme disease remains. All the sera were also tested by two different commercial Western Blot (WB) assays: Euroline-WB against Borrelia, Euroimmun, and Qualicode B. burgdorferi WB, Immunetics, in comparison with a multispecies "home made" WB. Performances of the three WB methods for the detection of IgM were very similar. On the contrary, these WBs performed with different values of sensitivity and specificity when IgGs were evaluated. The most sensitive method was the "home-made" WB IgG (71.7%), followed by the Euroline-WB IgG against Borrelia (68.3%). Qualicode B. burgdorferi WB IgG demonstrated to be only 26.6% sensitive. Both "home-made" WB IgG and Qualicode B. burgdorferi WB IgG were 100% specific, whereas Euroline-WB IgG against Borrelia scored 12 cross-reacting samples as borderline, showing a specificity value of 80.0%.     

An indirect hemagglutination antibody test to detect antibodies to Borrelia burgdorferi in patients with Lyme disease

An indirect hemagglutination antibody (IHA) test was evaluated for its ability to detect borrelial antibodies in serum samples from patients with Lyme disease. The key test reagent developed for this antibody detection system was tannic acid-treated and glutaraldehyde-fixed sheep red blood cells (SRBC) containing Borrelia burgdorferi (Bb) antigens attached to the outer surface of the SRBC. In order to establish suitable cut-off titers, initial specificity and sensitivity measurements were made using sera from 100 anonymous healthy volunteers and 30 additional pre-determined samples known to be non-reactive or reactive for Lyme disease or syphilis. These results were compared with those obtained using a commercially available ELISA. At titers >/=64, the IHA test had a combined 98% specificity and 100% sensitivity for these 130 serum samples, 30 of which were known positives or negatives, whereas the ELISA was less specific (93%) and much less sensitive (80%). Subsequent testing was performed on sera from 65 patients with the erythema migrans (EM) rash and 20 patients with early disseminated (cardiac/neurologic) symptoms or with Lyme arthritis. At initial presentation, 46-48% of the EM patients had IHA reactivity, with titers >/=128, while 42% were positive in the ELISA. Follow-up testing performed on these EM patients, 8-12 days after receiving antibiotic treatment, revealed that Bb antibodies were detected best by the IHA test (83-86% reactive) relative to the ELISA (81% reactive). Bb antibodies were readily detectable on all of the serum samples from the early disseminated and late stage Lyme disease cases in both assay systems. Based on these results and because of its technical and interpretive simplicity, the IHA test should be considered as a useful and convenient alternative for the serological analysis of Bb infections.     

Serologic survey of selected zoonotic disease agents in black-tailed jack rabbits from western Texas

A serologic survey for the agents of Rocky Mountain spotted fever (RMSF) (Rickettsia rickettsii), Borrelia spp. including the causative agent for Lyme disease (Borrelia burgdorferi), and plague (Yersinia pestis) was conducted on blood samples collected from 30 and 46 black-tailed jack rabbits (Lepus californicus) from an urban environment in Lubbock, Texas (USA) during winter 1987 and the following spring 1988, respectively. Antibody titers to the agents of RMSF and borreliosis were detected in sera of 28 and 1% of the jack rabbits, respectively. Neither organisms (rickettsiae and/or spirochetes) nor their associated antigens were detected in any of the tissue or whole blood samples; plague antibodies were not detected in the 76 jack rabbits sampled. Four of 18 ticks (Dermacentor parumapertus) removed from 12 jack rabbits were positive for RMSF using the fluorescent antibody test. The black-tailed jack rabbit is a common wildlife species living in close proximity to higher density human populations in many areas of the southwestern United States. Our results indicate the potential importance of urban populations of this mammal as reservoirs for at least one important zoonotic disease, RMSF, in western Texas.     

Dynamics of humoral immune response to Treponema pallidum proteins p17 and p41 at early stages of syphilis

In 60 blood sera from syphilis patients the titers of IgG to T. pallidum antigens p17 and p41 were detected with the use of the test system based on the recombinant analogues of T. pallidum proteins. The study revealed that primary syphilis was characterized by considerable prevalence of IgG to protein p41 with the total antibody level being low, while early latent syphilis was characterized mainly by considerable prevalence of IgG to protein p17 in the presence of high titers of antibodies. In secondary syphilis the sera contained a high total antibody level and a wide range of IgG ratios to individual antigens. On the basis of the data obtained the dynamics of immune response to antigens p17 and p41 at the early stages of the disease was hypothetically plotted. The curves of antibody levels had a wave-like character with the phase shifts of peaks for individual proteins and very low antibody titers (less than 1:100) in the negative peak areas. Conclusions were made that it was necessary to use the mixture of antigens in the production of the test systems and, when designing reference panels of sera, to include sera with extremely low titers of antibodies to individual proteins.     

Characterization of the antibodies to p62 nucleoporin in primary biliary cirrhosis using human recombinant antigen

Reactivity of sera from patients with primary biliary cirrhosis (PBC) with a 60 kDa component of nuclear pore complexes (NPCs), purified by affinity chromatography on wheat-germ agglutinin (WGA)-Sepharose, was previously detected. Recently, clinical significance of the anti-NPC antibodies in PBC became evident. In the light of recent reports, indicating the correlation of the anti-NPC antibodies with severity and progression of the disease, the characterization of the reactive antigens is becoming essential in the clinical management of patients with PBC. Since accurate autoantibody detection represents one of the fundamental requirements for a reliable testing, we have generated a human recombinant p62 protein and validated an immunoprecipitation assay for the detection of anti-p62. We also demonstrated that the generated human recombinant p62 nucleoporin was modified by N-acetylglucosamine residues. More than 50% of tested PBC sera precipitated (35)S-radioactively labeled p62 recombinant nucleoporin and 40% recognized this recombinant antigen by immunoblotting. We compared the reactivity of PBC sera with rat and human nucleoporin. The incidence of anti-p62 nucleoporin positive PBC sera increased by 15% when human recombinant antigen was used. The titer of autoantibodies in p62-positive PBC samples strongly varied. Preadsorption of the PBC sera with p62 recombinant protein completely abolished their reactivity with the antigen. In conclusion, this study unequivocally proves that autoantibodies reacting with the 60 kDa component of NPCs target p62 nucleoporin and, more importantly, provide a better antigen source for future evaluations of the clinical role of anti-p62 in PBC.     

Antibodies to meningococcal H.8 (Lip) antigen fail to show bactericidal activity

Purified H.8 (Lip) antigen was coupled to tresyl-activated Sepharose 4B and used in affinity columns to purify anti-Lip antibodies from convalescent patient sera and from immune rabbit sera. Affinity-purified anti-Lip antibodies isolated from two convalescent patient sera contained 1000 and 1280 ELISA units of antibody and included antibodies of IgG, IgA, and IgM isotypes. An anti-Lip mouse monoclonal ascites (2-1-CA2) had 28,400 ELISA units of antibody. Bactericidal assays were performed using three different case strains of Neisseria meningitidis group B, namely 44/76, 8532, and 8047. Neither preparation of purified human anti-Lip antibodies had detectable bactericidal activity against strains 44/76 and 8532, but one of the two had a titer of 1:4 against strain 8047. Anti-Lip antibodies that were purified from immune rabbit serum and contained 1600 ELISA units of anti-Lip antibodies also failed to show detectable bactericidal activity. The rabbits were immunized with purified Lip antigen and showed specific antibody levels of 2000-2200 units by ELISA, but even the unfractionated sera had little or no bactericidal activity against the test strains. The high titer mouse monoclonal ascites had no bactericidal activity against the test strains. The poor bactericidal activity associated with monoclonal and polyclonal antibodies to the Lip antigen suggest that in spite of other attractive properties it may not be useful as a meningococcal vaccine.     

Antibodies to Borrelia burgdorferi in deer and raccoons.

An enzyme-linked immunosorbent assay (ELISA) was developed to detect serum antibodies to Borrelia burgdorferi, the causative agent of Lyme borreliosis, in deer (Odocoileus virginianus) and raccoons (Procyon lotor). Blood samples were collected from these mammals in Connecticut, Maryland, North Carolina, Georgia and Florida. Seropositivity for deer was highest in Connecticut (56% of 353 sera) and Maryland (51% of 35 sera). Raccoons in Connecticut, Maryland, North Carolina, and Florida also had antibodies to B. burgdorferi, but prevalence of positive sera was highest in Maryland (79% of 14 samples). Based on adsorption tests, the immunoglobulins detected in these mammals were probably specific to B. burgdorferi. The ELISA was more sensitive than an indirect fluorescent antibody staining method and was more suitable for analyzing large numbers of serum samples.     

Seroprevalence of Lyme disease in gray wolves from Minnesota and Wisconsin.

To determine the seroprevalence of Lyme disease in gray wolves (Canis lupus) from various counties of Minnesota and Wisconsin (USA), 589 serum samples were collected from 528 wolves from 1972 to 1989. An indirect fluorescent antibody (IFA) test was used to detect the presence of antibodies against Borrelia burgdorferi. Titers of greater than or equal to 1:100 were considered positive. Results were confirmed by testing a few selected sera by Western blotting. Of the 589 sera tested, 15 (3%) had IFA titers of greater than or equal to 1:100. Three of the positive samples were collected from Douglas County in Wisconsin and twelve were from Minnesota counties. This study indicates that wolves are exposed to B. burgdorferi and are susceptible to Lyme disease.     

AN INDIRECT ENZYME-LINKED IMMUNOSORBENT ASSAY FOR THE DETECTION OF LEPTOSPIRA INTERROGANS SEROVAR POMONA ANTIBODIES IN BOVINE SERA

An indirect ELISA was developed and initially evaluated for the detection of bovine antibodies to Leptospira interrogans serovar pomona. The antigen used in this ELISA was extracted from a serovar pomona culture supernatant by a combination of centrifugation, digestion with proteinase K and ultra-centrifugation. The antigen showed little cross-reaction with immune rabbit sera to L. interrogans serovars copenhageni, grippotyphosa, hardjo and sejroe and, Leptospira biflexa serovar patoc. Some cross-reaction was observed with immune rabbit serum to L. interrogans serovar canicola. The relative sensitivity of the ELISA was 94.76% confidence interval =± 3.32%) when estimated with bovine sera (n=172) with serovar pomona microscopic agglutination test (MAT) titers of 100. The relative specificity of the ELISA was 99.28% (95% confidence interval = 1.40%) when estimated with bovine sera (n=139) with MAT titers of <100 to L. interrogans serovars canicola, copenhageni, grippotyphosa, hardjo, pomona and sejroe. Thirty six of 258 field sera (13.95%) with serovar pomona MAT titers of <100, gave positive reactions in the ELISA.     

Cryptic and exposed invariable regions of VlsE, the variable surface antigen of Borrelia burgdorferi sl

Borrelia burgdorferi, the Lyme disease spirochete, possesses a surface protein, VlsE, which undergoes antigenic variation. VlsE contains two invariable domains and a variable one that includes six variable and six invariable regions (IRs). Five of the IRs are conserved among strains and genospecies of B. burgdorferi sensu lato. IR(6) is conserved, immunodominant, and exposed at the VlsE surface but not at the spirochete surface, as assessed in vitro. In the present study, the remaining conserved IRs (IR(2) to IR(5)) were investigated. Antisera to synthetic peptides based on each of the IR(2) to IR(5) sequences were produced in rabbits. Antipeptide antibody titers were similarly high in all antisera. Native VlsE was immunoprecipitable with antibodies to IR(2), IR(4), and IR(5) but not to IR(3), indicating that the first three sequences were exposed at the VlsE surface. However, negative surface immunofluorescence and in vitro antibody-mediated killing results indicated that none of the IRs were accessible to antibody at the spirochetal surface in vitro.     

Serologic evidence of leptospirosis in woodchucks (Marmota monax) in central New York State.

Serum samples from 153 woodchucks (Marmota monax) from Tompkins County, New York, obtained in 1976 and 1977, were examined by plate agglutination tests for antibodies against five Leptospira antigens. Fourteen sera showed significant titers against either L. hardjo, L. icterohemorrhagiae and/or L. pomona. Reactions against L. hardjo were the most frequent. Woodchucks collected from two dairy farms with histories of bovine leptospirosis did not have a greater prevalence of antibodies than woodchucks collected from other locations. Each of two woodchucks experimentally-inoculated with L. hardjo developed titers to L. hardjo. Maximum titers occurred approximately 30 days post-inoculation. L. hardjo was not observed in urine specimens of these animals.     

T cell and antibody reactivity with the Borrelia burgdorferi 60-kDa heat shock protein in Lyme arthritis.

The reactivity of cloned T cells and serum antibodies, obtained from patients with chronic Lyme arthritis, with expressed recombinant B. burgdorferi 60-kDa heat shock protein homologue (HSP60) was analyzed. The expressed recombinant Borrelia burgdorferi HSP60 was bound by antibodies in the sera of patients with Lyme arthritis, but not by control sera. A T cell clone (CR253), isolated from one of four patients examined, exhibited an HLA-DR2 restricted proliferative response to the expressed recombinant B. burgdorferi HSP60. This T cell clone specifically recognized the HSP60 of B. burgdorferi and did not proliferate in response to the human, mycobacterial, or Escherichia coli HSP60 homologues. The epitope recognized by this cloned T cell, located between amino acids 260 and 274, is in a region of the spirochetal HSP60 that is not conserved between bacteria and eukaryotes.     

An enzyme-linked immunosorbent assay for detection of chronic subclinical Pasteurella pneumotropica infection in mice.

Serum samples from seventy-five, 3- to 12-week-old and 16 retired breeder male Swiss mice from a conventional colony with enzootic chronic subclinical Pasteurella pneumotropica infection were tested by enzyme-linked immunosorbent assay (ELISA) and Western blots for IgG antibodies to whole cell (WC) and lipooligosaccharide (LOS) antigens of P. pneumotropica. In 3- to 12-week-old mice, serum antibody levels to LOS exceeded those to the WC preparation. Western blots of sera from mice in this age group substantiated that a major component of the early IgG antibody response was directed against LOS antigens. Higher antibody levels to both antigen preparations in 3-week-old mice compared to mice 4 and 6 weeks old were interpreted as reflecting a decline in antibodies acquired from the dam. Active immunity indicative of infection was first detected at 8 weeks of age. Serum samples from retired breeder mice (28 weeks of age) also had substantial antibody titers to LOS but, in contrast to sera from mice in the younger age groups, retired breeders had significantly greater IgG reactivity to WC preparations than to LOS antigens. The superior specificity of the LOS antigen compared to the WC preparation in the ELISA was demonstrated by testing serum samples from retired breeder mice against WC and LOS antigens from P. ureae, P. multocida, and P. hemolytica. The reactivity of IgG against LOS antigens from these organisms was negligible, whereas substantial titers were evident to WC antigens. This ELISA, using LOS preparations as antigen, is a useful serologic assay for the detection of subclinical P. pneumotropica infection in mice.     

Effect of thimerosal on the whole yeast phase antigen of Histoplasma capsulatum

Summary Thimerosal (merthiolate) and formalin treated whole-cell yeast phase antigens ofHistoplasma capsulatum were prepared and their reactivities with sera from cases of histoplasmosis, blastomycosis and coccidioidomycosis were compared. Thimerosal treated antigens often gave complement fixation titers with heterologous sera 2 to 8 fold lower than the titers obtained with formalin treated antigens. However, with certain anti-histoplasmosis sera, thimerosal killed antigens had less reactivity with homologous antisera also. In virtually all cases an equal or higher specificity ratio was obtained with thimerosal killed antigens. The effects of thimerosal and formalin were independent, indicating different sites of reactivity of these reagents. Uptake of thimerosal at several concentrations suggested two types of reactions with live yeast phase cells. Analyses of the cellular fractions for thimerosal showed it was present only in the soluble fractions from which it was readily removed by dialysis. Cellular fractions killed with thimerosal retained several of the same physical and antigenic characteristics of those fractions isolated from frozen and thawed cells.     

A new type of anti-ganglioside antibodies present in neurological patients

High titers of anti-GA1 antibodies have been associated with neurological syndromes. In most cases, these antibodies cross-react with the structurally related glycolipids GM1 and GD1b, although specific anti-GA1 antibodies have also been reported. The role of specific anti-GA1 antibodies is uncertain since the presence of GA1 in the human nervous system has not been clarified. A rabbit was immunized with GD1a and its sera were screened for antibody reactivity by standard immunoassay methods (HPTLC-immunostaining and ELISA). Anti-GD1a antibodies were not detected but, unexpectedly, anti-GA1 IgG-antibodies were found. Antibody binding to GA1 was inhibited by soluble GA1 but also by GD1a. These results indicate that the rabbit produced antibodies that recognize epitopes present on the glycolipids, that are absent or not exposed on solid phase adsorbed GD1a. We investigated the presence of these unusual anti-ganglioside antibodies in normal and neurological patient sera. Approximately, 10% of normal human sera contained low titer of specific anti-GA1 IgG-antibodies but none of them recognized soluble GD1a. High titers of IgG-antibodies reacting only with GA1 were detected in 12 patient sera out of 325 analyzed. Of these, 6 sera showed binding that was inhibited by soluble GD1a and four of them also by GM1. This new type of anti-ganglioside antibodies should be considered important elements for understanding of the pathogenesis of these diseases as well as their diagnosis.     

Mapping the immunodeterminants of the complete streptococcal M6 protein molecule. Identification of an immunodominant region

The immune response to the complete streptococcal M6 protein was examined by kinetic ELISA to determine the reactivity of rabbit and human sera to M6 peptides representing 82% of the native molecule. The results revealed that rabbits immunized with purified native M6 protein or whole streptococci responded by reacting early and predominantly to one of the three sequence repeat regions of the molecule, the B-repeat, antibodies which have been shown to be non-opsonic. Antibodies to peptides representing the hypervariable N-terminal and adjacent A-repeat regions appear when opsonic antibodies are detected in the serum. Antibodies to peptides located within the conserved C-terminal half of the molecule (proximal to the cell) were restricted even after several immunizations. An examination of human sera from individuals with no recent streptococcal infection (greater than 3 yr), revealed that those sera opsonic for M6 streptococci contained antibodies reactive predominantly to the N-terminal and A-repeat regions, supporting the view that opsonic antibodies are long lived. Nonopsonic human sera to M6 streptococci exhibited a low reactivity to all peptides. However, by Western blot analysis, all human sera tested contained antibodies to the conserved region of the molecule, whereas only sera opsonic for M6 streptococci reacted with the variable region. Evidence is presented supporting the view that antibodies to the conserved regions of the M molecule may be conformation dependent.     

Comparison of immunological responses to the various types circumsporozoite proteins of Plasmodium vivax in malaria patients of Korea

We compared the seroreactivities against four synthetic peptide antigens (VK210, VK247, Korean type 1, and type 2) and a full length recombinant circumsporozoite protein (CSP) antigen of Plasmodium vivax (P. vivax ) in samples of sixty-three tertian malaria patients in Korea. Among the various CSP antigens, the full-length recombinant CSP showed the highest reactivity in malaria-exposed groups (85.7%, 54/63). No significant difference was found in the percentage of malaria patients with antibodies among four peptides examined, except a full-length recombinant CSP. Absorbance values from the peptide-based ELISA showed high correlations (r > 0.9, P < 0.05) at significant values. Five sera without the immunoaffinity against peptides were reactive towards the full-length recombinant CSP in ELISA. Sera, which were not reactive to a full length recombinant CSP antigen, were not recognized by any of peptide based ELISA. These data suggested that peptide structures included in Korean isolates, GNGAGGQAA, and VK247 peptides had immune reactivity and recognition epitopes. Among the antigens, GNGAGGQAA was less recognized by patients exposed to Korean strains of P. vivax in comparison to the VK210 structures.     

Complement-mediated killing of Borrelia burgdorferi by nonimmune sera from sika deer

Various species of cervid deer are the preferred hosts for adult, black-legged ticks (Ixodes scapularis and Ixodes pacificus) in the United States. Although frequently exposed to the agent of Lyme disease (Borrelia burgdorferi), these animals, for the most part, are incompetent as transmission reservoirs. We examined the borreliacidal activity of normal and B. burgdorferi-immune sera from sika deer (Cervus nippon) maintained in a laboratory setting and compared it to that of similar sera from reservoir-competent mice and rabbits. All normal deer sera (NDS) tested killed > 90% of B. burgdorferi cells. In contrast, normal mouse and rabbit sera killed < or = 22% of the Borrelia. Anti-B. burgdorferi antibodies could not be detected in any normal sera by indirect fluorescent antibody assay (IFA). Sera collected from deer 6 wk after exposure to B. burgdorferi by tick feeding exhibited IFA titers of 1:256, whereas sera from mice and rabbits similarly exposed had titers of > 1:1,024. Heat treatment (56 C, 30 min) of NDS reduced borreliacidal activity, with < 20% of the B. burgdorferi cells killed, suggesting complement-mediated killing. The chelators EGTA and EDTA were used to block the classical or both the classical and alternative complement pathways, respectively. Addition of 10 mM EGTA to NDS had a negligible effect on borreliacidal activity, with > 90% of the cells killed. Addition of 10 mM EDTA reduced the killing to approximately 30%, whereas the addition of Mg2+ (10 mM) restored borreliacidal activity to NDS. The addition of zymosan A, an activator of the alternative pathway, increased the survival of B. burgdorferi cells to approximately 80% in NDS. These data suggest that the alternative complement activation pathway plays a major role in the borreliacidal activity of NDS. Additionally, 10 mM EGTA had almost no effect on the killing activity of B. burgdorferi-exposed deer sera, suggesting that the classical pathway is not involved in Borrelia killing, even in sera from B. burgdorferi-exposed deer.