New Serologic Approaches to the Diagnosis of LD

The Lyme Disease Network
Conference Abstract

Title:	New Serologic Approaches to the Diagnosis of LD
Authors:	Dattwyler RJ
Conference:	10th Annual International Scientific Conference on Lyme Disease & Other Tick-Borne Disorders, National Institutes of Health, Bethesda, MD April 28-30, 1997
Presenter:	Raymond J. Dattwyler, M.D. Professor of Medicine, Chief of Allergy & Immunology Director of LD Center, SUNY at Stony Brook School of Medicine

Abstract:
B. burgdorferi expresses a large number of proteins. Outer surface proteins (Osp) A through F have been described and others may be defined in the future. Three proteins, OspA, OspB and OspC are best characterized and appear to be the most important immunologically. OspA and OspB are encoded on a 49 kilobase (kB) linear plasmid and OspC is encoded on a 28kB circular plasmid. OspA is the major surface protein expressed by this spirochete in the tick and when cultured. However, once tick feeding begins, OspA expression stops and OspC becomes the dominant outer surface protein. Thus, in the initial phase of mammalian infection, OspC is the major surface protein.

Definitive diagnosis of Bb infection is complicated by the following four key factors:
a. The varied nature of the presentation of the clinical symptoms
b. The overlap of these symptoms with numerous other infections and non-infectious diseases.
c. The difficulty of isolating and culturing the infectious agent from clinical samples.
d. The organism's extensive cross-reactivity with other infectious agents and immunopathologic factors.

Testing for Bb infection is frequently an early step in the differential diagnosis of patients with rheumatologic or neurologic symptoms. Definitive diagnosis or exclusion requires one or more serologic assays. With more than 5 million LD tests performed annually in the US, accurate, reliable assays are essential both to ensure early treatment of infected individuals and to exclude the large majority of uninfected patients with "Lyme-like" symptoms.

Except in patients with EM, Bb is rarely observed in clinical samples. Direct diagnosis via microbiological techniques is not practical at present. Instead, Bb infection is defined indirectly by detection of a humoral immune response to the organism. Antibody responses in the infected host follow the usual pattern. IgM appears first, followed by increasing levels of IgG and IgA during the second and third months of infection. Once the latter responses are established, they may remain detectable for years. In most patients, the earliest immune response to Bb is directed against the proteins flagellin (p41 or Fla) and OspC (25kd): the 35-37 kd and 39 kd proteins also elicit an early response in many individuals. The repertoire of antibody responses continues to expand as the disease progresses.

One serious issue in detecting antigens is that many Bb antigens comprise epitopes that cross-react with antibodies directed against other common infectious agents. For example, Bruckbauer et al. demonstrated extensive cross-reactivity between Bb and the bacterial pathogens B. hermsii, T. pallidum, L. interrogans, N. meningitidis, H. influenzae, Y. enterocolitica,, C. jejuni, L. monocytogens, P. aeruginosa, E. coli, S. typhimurium, S. flexeri, and L. micdadei. The Borrelia proteins in the 60-75 kd range, p41, p33, and two proteins of about 20kd are among the most highly cross-reactive. The immunodominant 41 kd flagellin antigen is not cross-reactive with non-flagellated bacteria, but it is highly cross reactive with similar protein from other spirochetes: greater than 50% of normal healthy adults with no history of Bb infection have circulating anti-41 kd IgG. Many individuals also have antibodies directed against other spirochetal antigens, most commonly the 60 kd common bacterial and 66 kd antigens. In addition, the accuracy of indirect assays for Bb infection can be compromised by immunopathogical conditions such as rheumatoid arthritis and systemic lupus erythematosus. Hence the positive predictive value of commercially available serological assays is poor because of the relatively high incidence of false positives.

Currently, the serological tests commonly used to detect antibodies against Bb are indirect immunofluorescence assays (IFA) or enzyme-linked immunosorbent assays (ELISA). These tests use whole cell Bb preparations. Bb are used as the antigen substrate for IFA and crude fractions of sonicated organisms are used for most ELISA tests. The use of whole Borrelia as the source of antigen introduces a high degree of variability and a large number of proteins that contain cross-reactive epitopes. In addition, since the assays are not standardized, there are significant differences in how the tests are performed and reported. For example, ELISA tests of Bb performed at five academic research centers and the CDC was recently evaluated. It was noted that there was only limited concordance among the results from laboratories that did not use immunoblotting to supplement the ELISA results. Moreover, the group in the study that used a commercial test had the poorest results by virtually all criteria. This study, together with other reports of poor results in terms of the accuracy, precision, and concordance among the various test kits on the market, has led to recommendations that all positive whole cell Borrelia ELISA and IFA test results be confirmed by immunoblotting assays. While immunoblotting has been shown to improve the accuracy of LD testing, the procedure is cumbersome and adds significantly to the time and cost of definitive diagnosis of LD.

We have identified a series of highly specific epitopes on several Bb antigens. Recombinant DNA techniques have then been used to isolate the genomic sequences coding for these epitopes and to incorporate them into vectors that can be expressed in E. coli, Gerber et al. demonstrated that an ELISA using recombinant OspC is more sensitive than a whole cell ELISA in detecting Borrelia antibodies in patients with erythema migrans. The use of single recombinant protein, however, may limit the utility of the assay. By combining specific portions of key proteins, into unique chimeric recombinant proteins, specific proteins have been developed that contain a higher ration specific to non-specific epitopes than the native proteins.

Unique ID: 97LDF028

The Lyme Disease Network of NJ, Inc.
43 Winton Road
East Brunswick, NJ 08816
http://www.lymenet.org/