Rapid Clearance of Lyme Disease Spirochetes Lacking OspC from Skin

Abstract

We previously demonstrated that Borrelia burgdorferi requires OspC to colonize a mammalian host. To delineate this requirement, we analyzed the clearance of ospC mutant spirochetes and found that they were eliminated within 48 h. We conclude that B. burgdorferi uses OspC to resist innate host defenses immediately after transmission.

Many vector-borne pathogens, like the spirochete Borrelia burgdorferi, alternate between arthropod and mammalian hosts. Each host environment requires a unique gene expression pattern that allows the bacterium to synthesize the products necessary for survival and multiplication in that host. Among the genes shown to be important for mammalian infection by B. burgdorferi is ospC, encoding a major outer surface protein (1, 3, 11, 21, 22). OspC synthesis increases while infected ticks feed and transmit spirochetes (13, 15, 17, 18) and is reduced after a mammalian infection is established (4, 8, 10, 12). OspC is analogous to Vtp of the related spirochete Borrelia hermsii (3), which is produced when this spirochete resides in its tick host and is replaced by a series of antigenically varying proteins after transmission to a mammal (16). In B. burgdorferi, an analogous antigenically varying family of proteins called VlsE follows OspC as the major spirochetal surface component in the mammalian host (4, 24).

Previous studies showed that animals inoculated with ospC mutant spirochetes by needle or tick bite did not seroconvert and that spirochetes were never recovered from any tissues, indicating that OspC is essential for infection of the mammalian host (7, 19, 20). Once infection is established, however, OspC is dispensable for persistence within a mammal (20), consistent with the pattern of ospC gene expression. This was demonstrated by establishing infection with ospC mutant spirochetes complemented with an unstable wild-type copy of the gene, which was subsequently lost in immunocompetent mice. Infections with ospC mutant spirochetes established by this method persisted for months, and B. burgdorferi reisolated from the mice did not contain the complementing plasmid or wild-type ospC gene. Feeding larval ticks can acquire ospC mutant spirochetes, which successfully carry out all subsequent stages of tick infection required for transmission to a naïve mouse (7, 20).

The early requirement for OspC suggests that it is involved in evasion of the host innate immune response. We have found, however, that ospC mutant spirochetes cannot infect mice lacking the adapter molecule MyD88, which is required for most Toll-like receptor-mediated responses (19). This implies that OspC provides protection from an aspect of host innate immunity that is MyD88 independent.

ospC mutant clearance from inoculation site.

Since the essential role of OspC appears to be in establishing mammalian infection, we wanted to further delineate the critical period in which the protein must be present. To do this, we inoculated wild-type and ospC mutant spirochetes into mouse skin and assessed whether spirochetes could be reisolated at various times soon after inoculation. We also assayed when spirochetes disseminated by determining when bacteria were first isolated from distal sites.

We conducted two independent experiments with wild- type B. burgdorferi (clone A3, derived from type strain B31 [6]), the deletion mutant ospCK1, and ospCK1 complemented with a shuttle vector carrying the ospC gene (ospCK1/pBSV2G-ospC [20]). C3H-HeN mice (Harlan Sprague-Dawley, Indianapolis, IN) were intradermally inoculated in the dorsal lumbar area with 5 × 10³ bacteria. At various times after inoculation, three mice per B. burgdorferi strain were euthanized. Samples from ears, bladders, ankle joints, and inoculation sites were cultured. In the first experiment, tissues were harvested at 3 h and 1, 2, 4, 8, and 21 days after inoculation with ospCK1 and A3. In the second experiment, tissues were harvested at 3, 7, and 12 h and 1, 2, 4, 8, and 21 days after inoculation with strains ospCK1 and ospCK1/pBSV2G-ospC. Wild-type and complemented mutant spirochetes were isolated from inoculation sites at all times (Table 1 shows pooled results) and were first isolated from distant sites at 8 days postinoculation. In contrast, ospC mutant spirochetes were isolated from inoculation sites through 24 h postinoculation (at which time five of six mice tested were negative) but were never isolated after this (Table 1). No serological response to B. burgdorferi was detected in mice inoculated with A3 or ospCK1 bacteria at 8 days postinoculation (data not shown). Mice inoculated with the ospC mutant remained seronegative at 21 days after inoculation, whereas mice inoculated with A3 or ospCK1/pBSV2G-ospC had seroconverted and recognized OspC, among other B. burgdorferi proteins (Fig. 1). We conclude that OspC plays a crucial role in B. burgdorferi infection within the first 48 h after introduction of spirochetes into a mammal, prior to dissemination or seroconversion of the host.

TABLE 1.

Isolation of B. burgdorferi strains at various times after intradermal inoculation of mice

Timea	Strain	Isolationb
		Distant sites	Inoculation site
3 h	A3	0/9	3/3
	ospCK1	0/18	6/6
	ospCK1/pBSV2G-ospC	0/9	3/3
7 h	A3	NDc	ND
	ospCK1	0/9	2/3
	ospCK1/pBSV2G-ospC	0/9	1/3
12 h	A3	ND	ND
	ospCK1	0/9	3/3
	ospCK1/pBSV2G-ospC	0/9	1/3
1 day	A3	0/9	3/3
	ospCK1	0/18	1/6
	ospCK1/pBSV2G-ospC	0/9	2/3
2 days	A3	0/9	3/3
	ospCK1	0/18	0/6
	ospCK1/pBSV2G-ospC	0/9	3/3
4 days	A3	0/9	3/3
	ospCK1	0/18	0/6
	ospCK1/pBSV2G-ospC	0/9	3/3
8 days	A3	7/9d	3/3
	ospCK1	0/18	0/6
	ospCK1/pBSV2G-ospC	2/9e	3/3
21 days	A3	9/9	3/3
	ospCK1	0/18	0/6
	ospCK1/pBSV2G-ospC	9/9	3/3

^aTime after intradermal inoculation of 5 × 10³ spirochetes at which mice were euthanized and tissues were harvested for culture of B. burgdorferi.

^bNumber of mouse tissues from which B. burgdorferi cells were isolated/number of mouse tissues cultured. Positive isolations are boldface. Three or six mice were inoculated per strain. Samples from the inoculation site, as well as from the ear skin, bladder, and joint (distant sites), were cultured from each mouse. Unless indicated otherwise, samples from all three distant sites from an individual mouse were either positive or negative. One-way analysis of variance, followed by Tukey's posttest, showed that the overall pattern of isolation for the ospCK1-inoculated mice differed significantly from the patterns for mice inoculated with A3 or ospCK1/pBSV2G-ospC (P < 0.05) and that the data for isolation from samples from 3 h to 1 day for the ospCK1-inoculated mice differed significantly from the data for isolation from samples from 2 days to 21 days (P < 0.01), whereas the same was not true for either the A3- or ospCK1/pBSV2G-ospC-inoculated mice (P > 0.05).

^cND, not done.

^dTwo mice were isolation positive for ear, bladder, and joint cultures, and one mouse was positive only for the joint culture.

^eTwo mice were isolation negative for cultures of all tissues, and one mouse was positive for bladder and joint cultures.

FIG. 1.

Immunoblot showing the serological response of mice 21 days after inoculation with B. burgdorferi. The lanes contained protein lysates of 10⁷ in vitro-grown high-passage B. burgdorferi bacteria not producing OspC (lanes 1) or B. burgdorferi bacteria producing OspC (lanes 2). Lysates were run on 12.5% acrylamide gels and blotted onto nitrocellulose membranes (7). Pairs of membrane strips were incubated with polyclonal rabbit anti-OspC sera (αOspC) (prepared by D. Bueschel and P. Rosa; 1:1,000 dilution in Tris-buffered saline-0.1% Tween 20) followed by peroxidase-conjugated goat anti-rabbit antibody or with the sera of mice inoculated with the B. burgdorferi strains indicated (wt, A3; mut, ospCK1; mut+ospC, ospCK1/pBSV2G-ospC; 1:200 dilution) followed by peroxidase-conjugated goat anti-mouse antibody (Sigma). Peroxidase activity was detected with SuperSignal reagents (Pierce). Although limited reactivity was detected with the serum from a mouse inoculated with the ospC mutant, we observed no difference between the reactivities of preinoculation serum and serum from 21 days postinoculation (data not shown). The migration positions of molecular mass standards (MW) are indicated on the left.

pBSV2G-ospC stability in infected mice.

In a recent study (20), we found that many B. burgdorferi isolates from mice initially infected with ospCK1/pBSV2G-ospC lacked the complementing plasmid. When bacteria were derived from mice that had been infected for 8 weeks or longer, none of the isolates retained pBSV2G-ospC. Our interpretation of these results was that OspC is required only during the initial phase of infection. The OspC protein can be a target for neutralizing antibodies (21-23), so spirochetes must reduce OspC production after the essential period in order to evade the acquired immune response and persist in the mammal. In our previous study, we detected some loss of pBSV2G-ospC after 21 days but did not look earlier, so we could not narrow the window of OspC requirement before that point. Since we obtained reisolates of ospCK1/pBSV2G-ospC from inoculation sites of mice harvested at every time in the current study, we were able to assess the loss of pBSV2G-ospC over the course of the experiment. The presence of pBSV2G-ospC was determined by plating the reisolated spirochetes in solid BSKII medium and screening individual colonies by PCR, using primers flg5′-AvrII (5′-CCTAGGTAATACCCGAGCTTCAAGGAG) and aacC13′-NheI (5). The PCR conditions were 5 min at 95°C, followed by 30 cycles of 1 min at 95°C, 30 s at 55°C, and 3 min at 68°C. We found that there was little plasmid loss through 8 days postinoculation (285 of 288 colonies screened were positive), but pBSV2G-ospC was not detected in any of the 72 colonies screened from inoculation site isolates derived from mice infected for 21 days. Two events are required for ospC-negative bacteria to arise: first, the period in which OspC is essential must pass, and second, the shuttle vector carrying the ospC gene must be lost. Since we have shown that pBSV2G-ospC is relatively stable during bacterial growth in SCID mice, which lack acquired immunity against OspC (20), we hypothesize that the development of antibodies against OspC between 8 and 21 days after inoculation selects against spirochetes containing the complementation plasmid. Because few if any cells of inoculation site isolates contained pBSV2G-ospC by 21 days after inoculation, the requirement for OspC must end before that point in mammalian infection, which is consistent with the timing of reduced OspC production in mice infected with the wild-type strain (4, 8, 10, 12).

Conclusions.

Our results demonstrate that OspC performs its crucial protective function within 48 h after transmission into the mammal. OspC protein is also required when spirochetes are transmitted by tick bite (7, 20). In this case the timing, but probably not the mechanism, of clearance may differ, since tick saliva contains many immunomodulatory products, including Salp15 (14). Less than 19 days after transmission, OspC-mediated protection is no longer necessary, and production of OspC can cease. Another spirochetal product, such as VlsE, may assume the role that OspC carries out initially, or the spirochete may otherwise manipulate its environment in order to avoid clearance. When the wild-type ospC gene is present at its normal location on the essential plasmid cp26 (2), OspC is typically eliminated by modulation of gene expression rather than by plasmid loss. Since continued production of OspC in the face of host humoral immunity is disadvantageous (23), an alternative escape mechanism for the complemented ospC mutants in our experiments is to lose the shuttle vector carrying ospC.

The early requirement for OspC indicates that the protein does not have a role in evading the acquired immune response, which is consistent with our previous finding that the defect in ospC mutants is not overridden in immunodeficient SCID mice (7, 20). We favor the possibility that the protein directly or indirectly protects spirochetes against an MyD88-independent aspect of innate immunity. Although natural killer T cells have recently been shown to recognize a glycolipid antigen of B. burgdorferi (9), these cells would be absent in SCID mice, in which the OspC protein is still essential. Phagocytic cells in the skin likely first detect B. burgdorferi deposited by tick bite or needle inoculation. These cells could engulf spirochetes directly or initiate inflammatory responses to signal effector cells to eliminate the bacteria. Our working model is that OspC inhibits phagocytosis of B. burgdorferi, perhaps by limiting opsonization by complement, allowing the bacteria to evade clearance immediately after transmission to the mammalian host.