Global analysis of Borrelia burgdorferi genes regulated by mammalian host-specific signals

Abstract

Lyme disease is a tick-borne infection that can lead to chronic, debilitating problems if not recognized or treated appropriately. Borrelia burgdorferi, the causative agent of Lyme disease, is maintained in nature by a complex enzootic cycle involving Ixodes ticks and mammalian hosts. Many previous studies support the notion that B. burgdorferi differentially expresses numerous genes and proteins to help it adapt to growth in the mammalian host. In this regard, several studies have utilized a dialysis membrane chamber (DMC) cultivation system to generate "mammalian host-adapted" spirochetes for the identification of genes selectively expressed during mammalian infection. Here, we have exploited the DMC cultivation system in conjunction with microarray technology to examine the global changes in gene expression that occur in the mammalian host. To identify genes regulated by only mammal-specific signals and not by temperature, borrelial microarrays were hybridized with cDNA generated either from organisms temperature shifted in vitro from 23 degrees C to 37 degrees C or from organisms cultivated by using the DMC model system. Statistical analyses of the combined data sets revealed that 125 genes were expressed at significantly different levels in the mammalian host, with almost equivalent numbers of genes being up- or down-regulated by B. burgdorferi within DMCs compared to those undergoing temperature shift. Interestingly, during DMC cultivation, the vast majority of genes identified on the plasmids were down-regulated (79%), while the differentially expressed chromosomal genes were almost entirely up-regulated (93%). Global analysis of the upstream promoter regions of differentially expressed genes revealed that several share a common motif that may be important in transcriptional regulation during mammalian infection. Among genes with known or putative functions, the cell envelope category, which includes outer membrane proteins, was found to contain the most differentially expressed genes. The combined findings have generated a subset of genes that can now be further characterized to help define their role or roles with regard to B. burgdorferi virulence and Lyme disease pathogenesis.

FIG. 1.

SDS-polyacrylamide gel electrophoresis and silver stain profiles of clonal isolate B31c8 under different cultivation conditions. Whole-cell-lysates from 10⁷ B31c8 organisms either cultured at 23°C (lane 23°), temperature shifted from 23°C to 37°C (lane TS), or cultivated within rat DMCs (lane DMC) were separated on a 12.5% polyacrylamide gel and stained with silver. The OspA and OspC proteins are indicated. Molecular mass standards (in kilodaltons) are indicated to the left.

FIG. 2.

Comparative analysis of QRT-PCR and microarray data. Twenty ORFs were selected at random for QRT-PCR for comparison with microarray data. Additionally, ospA, ospB, ospC, ospD, and flaB were included in the analysis. All QRT-PCRs were performed in triplicate to determine the average ΔCt for subsequent statistical analysis, which resulted in a correlation coefficient of r = 0.92.

FIG. 3.

Dichotomy between chromosomally and plasmid-encoded ORFs in their direction of differential expression. Of the 125 ORFs found to be differentially expressed within rat DMCs, 58 were up-regulated, and 67 were down-regulated. A total of 44 are encoded on the chromosome, 41 of which are up-regulated (93%). The plasmids contained 81 differentially expressed ORFs, 61 of which are down-regulated (79%).

FIG. 4.

Genomic distribution of ORFs differentially expressed during DMC cultivation. The total percentage of ORFs encoded by each genetic element found to be up- or down-regulated is shown. The percent up-regulated is indicated above the line (open bars), and the percent down-regulated is displayed below the line (solid bars). The total number of ORFs differentially expressed on each genetic element is noted above each bar.

FIG. 5.

Differentially expressed ORFs encoded by lp54. All 76 ORFs encoded by lp54 are displayed as boxes with bars extending up or down from specific ORFs indicating the fold up- or down-regulated during DMC cultivation. Solid and open boxes indicate gene orientations, and asterisks denote ORFs with leader peptides. Five genes discussed in the text are also labeled.

FIG. 6.

ORFs differentially expressed during DMC cultivation separated by functional category. The total percentage of ORFs in each of the 23 different functional categories found to be up- or down-regulated is shown. The percent up-regulated is indicated above the line (open bars), and the percent down-regulated is displayed below the line (solid bars).

FIG. 7.

Phenogram analysis of upstream regions from ORFs differentially expressed during DMC cultivation and multiple sequence alignment of selected promoters. (A and B) Phenogram analysis of the 100-bp regions just upstream of ORFs found to be up-regulated (A) or down-regulated (B). Four unrelated ORFs that were down-regulated and contained promoters that clustered together in panel B are indicated in boldface. Phenogram branch lengths are inversely proportional to the overall sequence identity between promoters. (C) Upstream sequences shown in boldface from panel B were subjected to a ClustalW multiple sequence alignment. The dashed rectangle encompasses the most highly conserved sequences that were shared by all four promoters. Solid lines below the lineup indicate the tandem repeats identified upstream of ospD. The Shine-Dalgarno ribosomal binding site (SD) and the −10 and −35 hexamers for RNA polymerase binding are indicated above the lineup.