A polymorphic multigene family encoding an immunodominant protein from Babesia microti

Abstract

Human babesiosis in the United States is caused predominantly by Babesia microti, a tick-transmitted blood parasite. Improved testing methods for the detection of infection with this parasite are needed, since asymptomatic B. microti infection represents a potential threat to the blood supply in areas where B. microti is endemic. We performed immunoscreening of an expression library of genomic DNA from a human isolate of B. microti (strain MN1). Among 17 unique immunoreactive clones, we identified 9 which represent a related family of genes with little sequence homology to other known sequences but with an architecture resembling that of several surface proteins of Plasmodium. Within this family, a tandem array of a degenerate six-amino-acid repeat (SEAGGP, SEAGWP, SGTGWP, SGTVGP) was found in various lengths between relatively well conserved segments at the N and C termini. In order to examine within-clone variation, we developed a PCR protocol for direct recovery of a specific bmn1-6 homologue directly from 30 human blood isolates, 4 corresponding hamster isolates, and 5 geographically corresponding Peromyscus leucopus (white-footed mouse) isolates. Isolates from the hamsters had the same sequences as those found in the corresponding human blood, suggesting that genetic variation of bmn1-6 does not occur during passage. However, clones from different patients were often substantially different from each other with regard to the number and location of the degenerate repeats within the bmn1-6 homologue. Moreover, we found that strains that were closely related geographically were also closely related at the sequence level; nine patients, all from Nantucket Island, Mass., harbored clones that were indistinguishable from each other but that were distinct from those found in other northeastern or upper midwestern strains. We conclude that considerable genetic and antigenic diversity exists among isolates of B. microti from the United States and that geographic clustering of subtypes may exist. The nature of the bmn1-6 gene family suggests a mechanism of antigenic variation in B. microti that may occur by recombination, differential expression, or a combination of both mechanisms.

FIG. 1

Amino acid sequence alignment of the BMN1-6-related family of genes recovered from the B. microti MN1 genomic expression library screened using B. microti-infected patient sera. The relatively well conserved amino and carboxy termini flank a variable length of pattern-encoded degenerate repeats composed of six amino acids. Oligonucleotide primers BMN16-5′, BMN16-3′, BMN13-5′, BMN13-3′, and BMN112-3′ (listed in Table 1) correspond to the solid underlined amino acids. Oligonucleotide primers BMN16A-5′ and BMN16B-3′ in Table 1 relate to the dashed overlined amino acids. Boxed amino acids indicate bmn1 downstream primers RCBB6B and RCBB6D in Table 1. Boldface letters indicate amino acids which differ from the consensus sequence. In the downstream sequence alignment, dots designate termination of the clone sequence and asterisks designate consensus with the bmn1-6 sequence. The key to the shaded sequences appears at the bottom.

FIG. 2

Ethidium bromide-stained 2% agarose gel visualization of RT-PCR products. Lane M, 100- to 800-bp marker; lanes 1 and 2, primers BMN13-5′ and BMN13-3′ and MN1 RNA template with and without the addition of reverse transcriptase, respectively; lanes 3 and 4, primers BMN16-5′ and BMN16-3′ and the MN1 RNA template with and without the addition of reverse transcriptase, respectively; lanes 5 and 6, primers BMN13-5′ and BMN112-3′ and the MN1 RNA template with and without the addition of reverse transcriptase, respectively.

FIG. 3

(A) Immunoblot showing mini-induction of clone bmn1-7 in the pET17b vector screened with the positive patient serum pool. T0 the culture lysate with an A₅₆₀ of 0.5 at time zero; T3, the culture lysate 3 h after the addition of IPTG. (B) Immunoblot showing reactivity of infected patient serum pool or hamster serum with increasing amounts (5, 10, and 20 μl) of crude B. microti lysate. λ, microliters. Numbers on the left are in base pairs.

FIG. 4

(A) Representative amino acid sequence alignment of B. microti from patients or animals amplified and sequenced with the BMN16-5′ and BMN16-3′ primers or, in some cases, the BMN16A-5′ and BMN16B-3′ primer set. Solid underlined regions indicate the former primer set. Boldface letters indicate amino acids that differ from the bmn1-6 consensus sequence. Numbers to the right designate the number of patients with the same motif. The key to the shaded sequences appears at the bottom. (B) Dendrogram generated from the Genetics Computer Group Pileup program demonstrating the clustering of all patient and animal isolate sequences generated from the bmn1-6 primer sets (BMN16-5′ and BMN16-3′ or BMN16A-5′ and BMN16B-3′). The left column is the patient or animal code. The center column is the known or probable location of infection (BI, Block Island; LI, Long Island; the other abbreviations represent states). The right column indicates the source of DNA used in the analysis.