Lyme Disease caused by infection with Borrelia burgdorferi is an emerging infectious disease and already by far the most common vector-borne disease in the U.S. Similar to many other infections, infection with B. burgdorferi results in strong antibody response induction, which can be used clinically as a diagnostic measure of prior exposure. However, clinical studies have shown a sometimes-precipitous decline of such antibodies shortly following antibiotic treatment, revealing a potential deficit in the host’s ability to induce and/or maintain long-term protective antibodies. This is further supported by reports of frequent repeat infections with B. burgdorferi in endemic areas. The mechanisms underlying such a lack of long-term humoral immunity, however, remain unknown. We show here that B. burgdorferi infected mice show a similar rapid disappearance of Borrelia-specific antibodies after infection and subsequent antibiotic treatment. This failure was associated with development of only short-lived germinal centers, micro-anatomical locations from which long-lived immunity originates. These showed structural abnormalities and failed to induce memory B cells and long-lived plasma cells for months after the infection, rendering the mice susceptible to reinfection with the same strain of B. burgdorferi. The inability to induce long-lived immune responses was not due to the particular nature of the immunogenic antigens of B. burgdorferi, as antibodies to both T-dependent and T-independent Borrelia antigens lacked longevity and B cell memory induction. Furthermore, influenza immunization administered at the time of Borrelia infection also failed to induce robust antibody responses, dramatically reducing the protective antiviral capacity of the humoral response. Collectively, these studies show that B. burgdorferi-infection results in targeted and temporary immunosuppression of the host and bring new insight into the mechanisms underlying the failure to develop long-term immunity to this emerging disease threat.

Long-lived T-dependent B cell responses fail to develop during persistent infection of mice with Borrelia burgdorferi, the causative agent of Lyme disease, raising questions about the induction and/or functionality of anti–B. burgdorferi adaptive immune responses. Yet, a lack of reagents has limited investigations into B. burgdorferi–specific T and B cells. We attempted two approaches to track B. burgdorferi–induced CD4 T cells. First, a B. burgdorferi mutant was generated with an influenza hemagglutinin (HA) peptide, HA111–119, inserted into the B. burgdorferi arthritis-related protein (Arp) locus. Although this B. burgdorferi arp::HA strain remained infectious, peptide-specific TCR transgenic CD4 T cells in vitro, or adoptively transferred into B. burgdorferi arp::HA–infected BALB/c mice, did not clonally expand above those of recipients infected with the parental B. burgdorferi strain or a B. burgdorferi mutant containing an irrelevant peptide. Some expansion, however, occurred in B. burgdorferi arp::HA–infected BALB/c SCID mice. Second, a (to our knowledge) newly identified I-Ab–restricted CD4 T cell epitope, Arp152–166, was used to generate Arp MHC class II tetramers. Flow cytometry showed small numbers of Arp-specific CD4 T cells emerging in mice infected with B. burgdorferi but not with Arp-deficient Borrelia afzelii. Although up to 30% of Arp-specific CD4 T cells were ICOS+PD-1+CXCR5+BCL6+ T follicular helper cells, their numbers declined after day 12, before germinal centers (GCs) are prominent. Although some Arp-specific B cells, identified using fluorochrome-labeled rArp proteins, had the phenotype of GC B cells, their frequencies did not correlate with anti-Arp serum IgG. The data suggest a failure not in the induction, but in the maintenance of GC T follicular helper and/or B cells to B. burgdorferi.

Borrelia burgdorferi (Bb), the bacterial agent of Lyme disease, has evolved numerous immune evasion mechanisms that result in Bb persistence in mice, its natural host, following infection via tick-bite. IgG antibodies control Bb-tissue loads but cannot clear the infection. We previously observed abnormalities in T-dependent B cell responses in mice following infection. Specifically, we demonstrated premature germinal center (GC) collapse, a lack of affinity maturation, memory B cell and long-lived plasma cell responses, and Ig responses that were skewed toward IgM. Due to the central role of CD4+ follicular T helper (TFH) cells in initiating and maintaining GCs, we hypothesized that Bb infection causes TFH dysfunction. Since Bb-specific CD4 T cell responses have not previously been characterized, we first identified an I-Ab restricted epitope on Arthritis related protein (Arp) of Bb, and used tetramers to characterize these Bb-specific CD4 T cells by FACS. We observed a surprisingly robust induction of Arp-specific TFH by day 14 of infection, which was followed by a precipitous decline in total and Arp-specific TFH by day 21 that preceded the observed collapse of GC. In order to explain this sharp TFH decline we studied the expression of inhibitory receptors on TFH. We also followed the induction of FOXP3+ regulatory T follicular (TFR) cells, which are known to limit GC responses. Not only were we able to demonstrate induction of a significant population of TFR, we also found that expression of the inhibitory Programmed death receptor 1 (PD-1) was significantly higher on Arp-specific TFH compared to overall TFH. Our data suggest that overshooting TFH regulatory mechanisms may inhibit T-dependent B cell responses during Bb infection.

Tick-borne infection with Borrelia burgdorferi (Bb) causes Lyme disease in humans and persistent infection of mice, its natural reservoir host. IgG responses critically control Bb tissue burden but do not clear the infection. We previously showed induction of germinal centers (GC) in mice after Bb infection followed by their rapid collapse within 30 days, despite ongoing infection. This was consistent with the lack of long-lived and high-affinity antibody responses in these mice. Our recent experiments showed that serum IgG from Bb-infected mice bound more strongly to the inhibitory FcγRIIb on B cells, as well as recombinant FcγRIIb than those from uninfected mice, as assessed by flow cytometry and ELISA, respectively. The data prompted us to investigate whether the interaction between Bb infection-induced IgG and FcγRIIb results in the suppression of GC. Indeed, Bb-infected mice lacking FcγRIIb signaling (FcγRIIb−/−) maintained GC B cells and CD4 TFH for at least 90 days after infection, and GC B cells showed reduced apoptosis, as assessed by staining for Annexin V. Furthermore, transfer of serum from Bb-infected wild type but not AID−/−sIgM−/− mice lacking IgG, IgM and from non-infected wild type mice, into 14-day Bb-infected wild type mice resulted in reduced GC at day 21. However, despite GC maintenance and the resulting increases in Bb-specific, T cell-dependent IgG responses, enhanced antibody affinity maturation was not observed and Bb tissue burdens of FcγRIIb−/− and control mice were comparable. The data show the importance for FcγRIIb in the regulation of GC during Bb infection, while additional mechanisms currently under investigation seem to underlie their functional deficits, enabling the establishment of Bb persistence.

Appropriate polarization of effector CD4 T cells is essential for eliminating invading pathogens. Borrelia burgdorferi (Bb), the spirochetal agent of Lyme disease, evades immune clearance in infected mice and humans, establishing bacterial persistence. Here we aimed to explore the functionality of CD4 T cell as a possible target of Borrelia-induced immune evasion. In response to Bb infection, CD4 T cell polarization is considered to be Th1 dominant. However, when we stimulated CD4 T cells from Bb infected mice ex vivo with Bb surface Ag Arthritis Related Protein (Arp), we were unable to detect IFNγ, prompting us to re-examine the polarization state of lymph node and blood effector CD4 T cells throughout Bb infection by quantitative RT-PCR and flow cytometry. In contrast to prevailing views, we found only modest numbers of Th1-polarized activated CD4 T cells, as measured by expression of Tbet (Tbx21), and little to no induction of Ifng, including in Bb-specific CD4 T cells, identified by a novel I-Ab tetramer to Arp. This could not be explained by enhanced polarization to Th2 or Th17, as GATA-3 and RORγt expression were also largely absent. Nor could it be explained by polarization at a T cell effector site, the skin, as skin-resident T cells also showed no polarization even 4 months after infection. Furthermore, Th1 polarization was not necessary for controlling bacterial dissemination, as Bb infection of Tbet−/− mice showed significant reductions in Borrelia numbers in the tibiotarsal joints, which correlated with enhanced Th17 polarization as measured by RORγt expression. Together, our data demonstrate a failure of CD4 T cell polarization after Bb infection, indicating effector CD4 T cell polarization as a novel target of Bb immune evasion.