Role of NK-Like CD8+ T Cells during Asymptomatic Borrelia burgdorferi Infection
ABSTRACT
Lyme disease (LD) due to Borrelia burgdorferi is the most prevalent vector-borne disease in the United States. There is a poor understanding of how immunity contributes to bacterial control, pathology, or both during LD. Dogs in an area of endemicity were screened for B. burgdorferi and Anaplasma exposure and stratified according to seropositivity, presence of LD symptoms, and doxycycline treatment. Significantly elevated serum interleukin-21 (IL-21) and increased circulating CD3+ CD94+ lymphocytes with an NK-like CD8+ T cell phenotype were predominant in asymptomatic dogs exposed to B. burgdorferi. Both CD94+ T cells and CD3− CD94+ lymphocytes, corresponding to NK cells, from symptomatic dogs expressed gamma interferon (IFN-γ) at a 3-fold-higher frequency upon stimulation with B. burgdorferi than the same subset among endemic controls. Surface expression of activating receptor NKp46 was reduced on CD94+ T cells from LD, compared to cells after doxycycline treatment. A higher frequency of NKp46-expressing CD94+ T cells correlated with significantly increased peripheral blood mononuclear cell (PBMC) cytotoxic activity via calcein release assay. PBMCs from dogs with symptomatic LD showed significantly reduced killing ability compared with endemic control PBMCs. An elevated NK-like CD8+ T cell response was associated with protection against development of clinical LD, while excess IFN-γ was associated with clinical disease.
INTRODUCTION
Lyme disease (LD) is the most common vector-borne disease in the United States, with an estimated ∼476,000 annual cases (1, 2). The range and prevalence of the causative spirochete, Borrelia burgdorferi, and its vector, Ixodes sp. ticks, have been steadily increasing since LD became nationally notifiable in 1991 (3). Following infection, Lyme arthritis occurs in ∼60% of untreated patients (4). Most cases are treatable with antibiotics; however, a subset of patients experience antibiotic-refractory joint inflammation, with a suspected immune-based etiology (4). Similar to humans, dogs are incidental hosts of B. burgdorferi infection. Most B. burgdorferi-seropositive dogs are asymptomatic (AS); Lyme arthritis is the most common clinical manifestation (5). As people do not seek clinical treatment unless they experience symptoms, dogs provide a unique natural model to identify immune processes associated with subclinical B. burgdorferi infection.
Invariant natural killer T (iNKT) cells were identified as critical for controlling experimental LD. Both mouse and human iNKT cells recognize B. burgdorferi diacylglycerol glycolipid antigen, leading to proliferation, cytokine production, and granzyme-mediated pathogen killing (6, 7). Mice lacking iNKT cells experience prolonged and more severe arthritis, increased carditis, and reduced spirochete clearance compared to wild-type controls (8, 9). Further, CD1d−/− mice, unable to present antigen to iNKT cells, have worsened arthritis and increased B. burgdorferi burden (10). iNKT cells were found elevated in LD patient blood, but only in synovial fluid of antibiotic-responsive patients (11). These studies suggest iNKT cells are protective against human LD, but roles for this cell subset in asymptomatic, controlled, natural infection have not been evaluated.
iNKT cells belong to a burgeoning family of unconventional T cells that bridge innate and adaptive immunity (12). While mice express only CD1d, humans and dogs express additional CD1 molecules—CD1a, CD1b, and CD1c (13–16). In humans, CD1 antigen presentation facilitates expansion of a much more diverse and abundant unconventional T cell compartment compared to that observed in mice (14). CD1d presentation of B. burgdorferi glycolipid antigen to iNKT cells is crucial for protection in experimental murine models; however, the potential contribution of other NKT-like cells to protection from clinical Lyme disease has yet to be explored.
Whether peripheral natural killer (NK) cells increase during LD is not clear (17, 18). Antibiotic-refractory human Lyme arthritis was associated with persistence of NK cells expressing CD16, a marker of antibody-dependent cell cytotoxicity (11). CD56bright NK cells in synovial fluid from antibiotic-refractory Lyme arthritis patients had a non-statistically significant, but larger, proportion of gamma interferon (IFN-γ)-producing cells compared to cells from treatment-responsive patients (11). Although T helper 1 (Th1)-type immunity may contribute to bacterial control, increased serum IFN-γ in erythema migrans-positive patients correlated with increased symptomatology (19). In humans, a higher ratio of Th1 to Th2 cells in synovial fluid directly correlates with arthritis severity (20). Furthermore, excessive type 1 inflammation is associated with increased risk of developing antibiotic-refractory Lyme arthritis (19, 21). These studies may indicate NK cell subsets promote inflammation and immunopathology.
Herein, a cohort of naturally infected dogs was screened for B. burgdorferi exposure and clinical signs. Due to previous studies indicating a potential protective role for iNKT cells in experimental murine models, we were particularly interested in investigating differences in NKT-like cell types and their critical outputs, inflammatory cytokine production, and cytotoxicity in dogs exposed to B. burgdorferi with and without clinical signs of LD. Our goal was to evaluate cellular responses associated with maintenance or loss of an asymptomatic state and presence of clinical LD. We evaluated inflammatory and cytotoxic functions of circulating lymphocytes expressing CD94, a transmembrane glycoprotein expressed by natural killer family cells (22–24).
RESULTS
Peripheral NK cell frequency not modulated during Lyme disease.
We wanted to determine if NK cells were differentially induced in dogs with asymptomatic (AS group) or symptomatic (SY group) Lyme disease. CD94 is a C-type lectin receptor specific for nonclassical major histocompatibility complex (MHC) class I molecules and preferentially expressed by NK cells, NKT cells, and related unconventional T cell subsets (25, 26). CD94 is expressed early in murine NK cell differentiation, and the anti-human CD94 clone HP-3D9 cross-reacts with canine CD94 (27–29). Graves et al. (29) found CD94 surface expression identified cells with an NK/NKT surface phenotype among canine peripheral blood mononuclear cells (PBMCs), and expansion of CD94+ cells by interleukin-2 (IL-2)/IL-15 resulted in a highly cytotoxic, electron-dense granular cell population (29). In our cohort, sorted CD94+ cells also displayed a granular phenotype by light microscopy (see Fig. S1 in the supplemental material). The majority of CD94+-expressing endemic control (EC group) lymphocytes coexpressed NKp46 (data not shown), and similar to humans and mice (22, 30), CD94 expression was observed at both intermediate and high levels on lymphocytes (Fig. S1C).
To determine if CD94+ cells were differentially regulated during Lyme disease, we used flow cytometry to analyze PBMCs from subsets of subjects within a cohort of B. burgdorferi-exposed dogs (see Table S1 in the supplemental material). Lymphocytes were subdivided into CD3-expressing or nonexpressing cells and then positively gated by CD94 surface expression (Fig. S1). Circulating CD3− CD94+ lymphocytes, thought to be predominately composed of NK cells, did not differ in frequency between endemic controls and B. burgdorferi-seropositive dogs (Fig. 1A and C). In endemic controls, CD94high-expressing NK cells were significantly enriched over the CD94dim subset (Fig. 1B). However, in B. burgdorferi-seropositive dogs, the CD3− CD94high subset was no longer dominant (Fig. 1B). This trend is also seen when comparing asymptomatic versus symptomatic Lyme disease. Endemic controls and doxycycline-treated symptomatic LD dogs had a higher ratio of CD3− CD94high cells over CD3− CD94dim cells (Fig. 1D).
FIG 1
Although peripheral NK cell frequency does not appear to be modulated during human LD, the phenotype of these cells differs among patients with antibiotic-responsive versus -refractory Lyme arthritis, suggesting NK cells play an active role in pathology (11). Granzyme B is a serine protease stored in intracellular granules by cytotoxic cells, including NK, NKT, and CD8+ T cells. Upon activation, granules are released onto target cells inducing cell death. NKp46, a conserved activating receptor expressed by mature NK cells (28, 31), clusters at the NK target cell immune synapse, contributing to cytotoxicity (32). Interestingly, the frequencies and geometric mean fluorescence intensities (gMFIs) of granzyme B and NKp46 were similar among CD3− CD94+ lymphocytes independent of B. burgdorferi serostatus (see Fig. S2A to D in the supplemental material).
CD94+ T lymphocytes expanded during asymptomatic Lyme disease.
Contrary to the CD3− CD94+ population, the frequencies of CD3+ CD94+ T cells were significantly increased among circulating lymphocytes in B. burgdorferi-seropositive dogs compared to seronegative dogs (Fig. 1E). These cells were predominantly CD94high (Fig. 1F to H). When seropositive dogs were divided into asymptomatic or symptomatic LD, the frequency of CD94+ T cells in the peripheral blood of asymptomatic dogs was double the frequency observed among endemic control lymphocytes (P = 0.0041) (Fig. 1G). This subpopulation remained significantly increased in asymptomatic dogs over endemic controls when dogs were stratified by breed (data not shown).
Based on this, we expected circulating CD94+ T cells from asymptomatic dogs would have a more activated phenotype compared to those from symptomatic dogs. Intracellular granzyme B was found at similar levels in CD94+ T cells independent of B. burgdorferi serostatus (Fig. S2E and F). We expected more mature NKp46-expressing CD94+ T cells would be found in B. burgdorferi-seropositive compared to endemic controls. Interestingly, among the CD3+ CD94+ subset, there was a nonsignificant decline in the frequency of NKp46+ cells from both asymptomatic and symptomatic B. burgdorferi-seropositive dogs (Fig. S2G). When examining NKp46 geometric mean fluorescence intensity (gMFI), asymptomatic and symptomatic B. burgdorferi-seropositive dog CD94+ T cells had significantly less NKp46 surface expression compared to doxycycline-treated dogs (Fig. S2H).
Comparison of CD94dim- versus CD94high-expressing lymphocyte populations.
CD94 is differentially expressed during NK cell maturation (22, 30). The CD94+ T cell population from dogs seropositive for B. burgdorferi was predominately CD94high (Fig. 1F). Most evidence suggests CD94high NK cells represent a more mature version of CD94dim NK cells (30). To analyze potential differences in these subsets during asymptomatic and symptomatic LD, we compared levels of granzyme B and NKp46 expression between the two CD94+ lymphocyte subpopulations (Fig. 2).
FIG 2
CD94high T cells had a higher proportion of granzyme B-positive cells (Fig. 2A). However, CD94dim lymphocytes had more intracellular granzyme B on a per cell basis than their CD94high counterparts (Fig. 2B). Among CD94+ T cells, NKp46 was found to be expressed significantly more often and at a higher level on CD94dim T cells than CD94high T cells in endemic control, asymptomatic, and symptomatic B. burgdorferi-seropositive dogs (Fig. 2C and D).
CD94+ T cell subset has an NK-like CD8+ T cell phenotype.
Within CD3-expressing lymphocyte populations, CD94 has been shown to be expressed by NKT cells, some γδ T cells, and a subset of CD8+ T cells in both mouse and human systems. CD94+ T cells from endemic control dogs, as well as B. burgdorferi-seropositive dogs, were predominantly CD8+ CD4− (mean frequencies: EC, 88.2%; AS, 86.6%; SY, 75.2%) (Fig. 3A). CD8− CD4− double-negative cells made up roughly 10 to 20% of this population, depending on Lyme disease clinical state (mean frequencies: EC, 9.5%; AS, 7%; SY, 21.2%). CD4+ CD8−-expressing cells were less than 5% of the CD94+ T cells (mean frequencies: EC, 0.6%; AS, 4.5%; SY, 0.9%).
FIG 3
To establish whether these CD94+ T cells had an antigen binding pattern consistent with iNKT cells, we combined PBMCs with human CD1d tetramers loaded with PBS-57 ligand, an analog of α-galactosylceramide, which binds type 1 CD1d-restricted NKT cells (see Fig. S3 in the supplemental material). CD94+ T cells at 4.7%, 8.1%, and 13.3% were labeled by CD1d/PBS-57 tetramers from endemic control dogs, asymptomatic B. burgdorferi-seropositive dogs, and symptomatic B. burgdorferi-seropositive dogs, respectively (Fig. 3A). This indicates type 1 CD1d-restricted NKT cells comprise a minor portion of CD94+ T cells in this cohort. The frequency of CD1d/PBS-57 binding cells was significantly higher (1.7-fold, P = 0.025) in asymptomatic B. burgdorferi-exposed dogs compared to endemic control dogs and highest among cells from symptomatic exposed dogs (2.8-fold, P = 0.007) (Fig. 3A). Expansion of type 1 CD1d-restricted NKT cells occurred in dogs in response to B. burgdorferi infection, similar to observations from experimental LD (8–11).
Based on these findings, the CD1d restriction of CD94-expressing T cells was of interest. We compared CD1d/PBS-57 binding among CD3+ CD94− and CD3+ CD94+ lymphocytes from matched subjects (Fig. 3B). We found CD1d/PBS-57 binding cells among both CD94− and CD94+ T cell subsets. However, the proportion among the CD94+ subset was significantly larger, at double the proportion among CD94− T cells (8.4% and 3.7%, respectively; P < 0.0001). In agreement, sorted CD94+ T cells contained a significantly higher level of iTCRα mRNA than CD94− T cells compared to CD3− CD94− lymphocytes from matched subjects (Fig. 3C) (33). Therefore, type 1 invariant NKT cells are enriched among, but not restricted to, CD94+ T cells.
The majority of CD94+ T cells were CD8+ (Fig. 3A). NK-like CD8+ T cells (sometimes called innate memory CD8+ T cells) are a subset of conventional CD8+ T cells that arise during infection and tumor settings described in mice and humans (34, 35). These cells are most commonly defined as CD3+ CD8+ lymphocytes expressing killer immunoglobulin receptors (CD94/NKG2A, -C, -D, and -E), and distinct from type 1 NKTs by not being CD1d restricted and having a more diverse T cell receptor (TCR) repertoire (34, 36–38). These cells have functional features like those of NK cells, such as cytotoxic activity and rapid IFN-γ response to IL-12/IL-18 stimulation, but display an activated memory-like phenotype (34).
To explore whether the CD94+ CD8+ T cells in our study displayed an NK-like phenotype, we compared CD94− CD8+ CD3+ lymphocytes (conventional CD8+ T cells) to CD94+ CD8+ CD3+ lymphocytes in matched subjects (see Fig. S4 in the supplemental material). The CD94+ CD8+ T cells had a significantly higher frequency of intracellular granzyme B+ cells compared to conventional CD8+ T cells (65.8% versus 51.3%;, respectively, P < 0.0001) (Fig. 3D). Similarly, CD94+ CD8+ T cells had a significantly higher frequency of NKp46+ cells compared to conventional CD8+ T cells (15.3% versus 4.9%, respectively; P = 0.0017) (Fig. 3E). NKp46 expression was inducible on CD94+ CD8+ T cells, where the portion of NKp46+ cells increased significantly upon exposure to B. burgdorferi spirochetes (medium versus B. burgdorferi, P = 0.017), and recombinant canine IL-12 (rcIL-12) stimulation also trended to induce NKp46 expression (medium versus rcIL-12, P = 0.057) (Fig. 3F). In contrast, CD94− conventional CD8+ T cells from matched subjects did not upregulate surface NKp46 in response to either stimulus. Therefore, CD94+ CD8+ T cells respond to environmental stimuli by increasing expression of an NK cell cytotoxicity receptor and display an NK-like phenotype.
IL-21 is increased in asymptomatic, Borrelia-seropositive dog serum.
IL-21 is a pleotropic cytokine that augments NK and T cell type 1 polarization and promotes NKT cell expansion (39–41). We found significantly elevated serum IL-21 in asymptomatic seropositive dogs over control dogs (Fig. 4A). However, serum IL-21 was not statistically increased in dogs with symptomatic LD. As IL-21 can work with IL-23 to regulate type 17 immune responses, we also measured serum IL-23 levels in dogs at different stages of LD (42). However, IL-23 was not significantly modulated in either seropositive group compared to endemic controls (Fig. 4B).
FIG 4
B. burgdorferi induces a higher proportion of IFN-γ-producing CD94+ lymphocytes from symptomatic Lyme disease.
B. burgdorferi infection elicits a mixed adaptive immune in humans, including IFN-γ, a highly inflammatory cytokine (19, 20, 43). NK, NKT, and CD8+ T cells are strong producers of IFN-γ. After incubating PBMCs for 24 h with B. burgdorferi spirochetes, we measured CD3− CD94+ and CD3+ CD94+ lymphocyte intracellular IFN-γ by flow cytometry (Fig. 5).
FIG 5
Compared to no stimulation, B. burgdorferi induced a significant proportion of CD94+ T cells to become IFN-γ+, with moderate induction within asymptomatic dog PBMCs (1.8-fold over EC, P = 0.289) and a significantly greater induction in symptomatic LD PBMCs (2.9-fold over EC, P = 0.0037) (Fig. 5B). CD3− CD94+ lymphocytes also became IFN-γ+ after B. burgdorferi stimulation, independent of B. burgdorferi serostatus (Fig. 5C). Spirochete-stimulated PBMCs from symptomatic dogs resulted in a 2.7-fold-higher induction of IFN-γ+ CD3− CD94+ lymphocytes over similarly stimulated endemic control PBMCs (SY versus EC, P < 0.0001) or asymptomatic dog PBMCs (SY versus AS, P = 0.0003), while spirochete-stimulated asymptomatic dog PBMCs resulted in a nonsignificant induction of IFN-γ+ CD3− CD94+ lymphocyte cells over control PBMCs (AS versus EC, P = 0.815). This enhanced induction of IFN-γ-secreting cells might indicate CD94+ cells in symptomatic dogs require a lower threshold of activation.
We evaluated the correlation between surface CD94 expression and intracellular IFN-γ on cells exposed to spirochetes. There was a significant positive correlation between CD94 gMFI and IFN-γ gMFI on CD94+ T cells (Fig. 5D). However, for CD3− CD94+ lymphocytes, CD94 gMFI did not correlate with intracellular IFN-γ (Fig. 5E). NKT cells have been described to recognize B. burgdorferi antigens directly or to be indirectly activated by myeloid cell-derived cytokines within the PBMC fraction to induce IFN-γ production.
Peripheral lymphocytes during symptomatic Lyme disease have decreased cytotoxic activity.
After observing enhanced IFN-γ induction by CD94+ cells from symptomatic LD dogs, we were interested in evaluating whether other critical NK-like functions, particularly cytotoxicity, were enhanced during clinical disease. iNKT cells from mice infected with B. burgdorferi were able to directly kill extracellular spirochetes in a granzyme-dependent manner (6). In addition, B. burgdorferi is capable of infecting multiple cell types, including synovial cells, endothelial cells, fibroblasts, neurons, and glial cells (44–48). Finally, granzyme A/B-deficient mice have significantly higher joint bacteria burdens, implicating cytotoxic cells contribute to bacterial control (6).
Thus, we used the allogeneic canine MDCK cell line as target cells in an ex vivo cytotoxicity assay (Fig. 6). In this assay, target cells released calcein when killed by effector cells. PBMCs from both B. burgdorferi-seronegative and -seropositive dogs elicited target cell calcein release. We observed a significantly decreased released calcein ratio from symptomatic B. burgdorferi-exposed dog PBMCs compared to endemic control PBMCs (Fig. 6A). Asymptomatic B. burgdorferi-exposed dog PBMCs triggered intermediate calcein release that was not statistically different from that of endemic controls. Thus, contrary to IFN-γ production, peripheral lymphocytes from dogs with symptomatic LD had defective cytotoxicity.
FIG 6
We wanted to know which CD94+ cell parameters correlated with target cell calcein release (killing). Endemic control PBMCs had the greatest capacity to elicit calcein release. A higher frequency of CD94+ T cells expressing NKp46 significantly correlated with calcein release in endemic control cells, reinforcing this receptor’s role in cytotoxicity (r = 0.7, P = 0.04). We found no significant association between CD3− CD94+ cell marker expression and released calcein from killed target cells (Fig. 6B and C).
B. burgdorferi and Anaplasma coinfection.
Complicating LD, Ixodes ticks transmit additional medically important pathogens, including Anaplasma phagocytophilum, the causative agent of anaplasmosis. Anaplasmosis elicits immunosuppression with leukopenia, leaving subjects susceptible to secondary opportunistic infections (49). A substantial proportion of B. burgdorferi-seropositive dogs in our cohort were coexposed to Anaplasma (see Table S2 in the supplemental material). We hypothesized Anaplasma coexposure would exacerbate LD severity. We evaluated clinical outcome and NKT cell responses in dogs coexposed to B. burgdorferi and Anaplasma spp.
Upon physical exam, dogs coexposed to both B. burgdorferi and Anaplasma were no more likely to be LD symptomatic than dogs exposed to only B. burgdorferi (P = 0.606, Fisher’s exact test) (Table S2). Dogs coexposed to B. burgdorferi and Anaplasma did not have increased circulating CD3− CD94+ or CD3+ CD94+ lymphocyte frequencies compared to those observed in dogs exposed to B. burgdorferi alone (Fig. 7A and B).
FIG 7
A high proportion of both CD3− and CD3+ CD94+ lymphocytes from dogs with symptomatic LD produced IFN-γ in response to B. burgdorferi. Neither of these cell subsets from Anaplasma-coexposed dogs had enhanced IFN-γ production compared to B. burgdorferi singly exposed dogs (Fig. 7C and D). The proportion of IFN-γ+ CD3− CD94+ lymphocytes induced by spirochete stimulation was significantly lower in coexposed dogs compared to that in B. burgdorferi singly exposed dogs (P = 0.011) (Fig. 7D). Anaplasma coexposure does not exacerbate CD94+ lymphocyte responses during canine LD and may dampen proinflammatory responses to B. burgdorferi.
DISCUSSION
iNKT cells have previously been shown to be protective against experimental B. burgdorferi infection and disease severity, but roles for this cell subset in asymptomatic, controlled, natural infection were not evaluated. Within a cohort of naturally infected hunting dogs, asymptomatic B. burgdorferi exposure was associated with increased frequency of CD94+ T cells compared to seronegative endemic controls (Fig. 1G). We used the human CD1d tetramer and iTCRα gene expression to determine if CD94+ T cells found in dogs corresponded to iNKT cells.
We demonstrate for the first time that there was significantly increased binding of the PBS-57 ligand-loaded CD1d tetramer in both B. burgdorferi-exposed groups, indicating NKT-like cells are present in this population and importantly increased during clinical Lyme disease (Fig. S3). Previously, Yasuda et al. (33) used a murine α-GalCer/CD1d dimer and found a present, but minimal, binding population among canine PBMCs from one healthy lab-bred dog (33). The percentage of CD1d tetramer binding cells among CD3+ cells was higher in our study than that found by Yasuda et al. (33). This could be attributed to our use of the human CD1d tetramers, which have higher homology compared to the murine tetramers. The canine cohort used here is also more antigen experienced, which may increase the circulating iNKT cell compartment, as Yasuda et al. (33) only measured CD1d dimer binding in a dog raised under specific-pathogen-free conditions.
We observed a similar pattern when amplifying iTCRα mRNA corresponding to a putative canine invariant T cell receptor gene. iTCRα mRNA was enriched among CD94+ T cells (Fig. 3C). The canine iVa-iJa gene region used to design iTCRα primers has 72% amino acid homology with the human Vα24-JαQ region (33). The iTCRα mRNA fold change observed in CD94+ T cells in our study (∼117-fold) was lower than that reported by Yasuda et al. (33) (∼271-fold), which we attribute to their amplifying from a purer, flow-sorted α-GalCer/CD1d-binding PBMC population.
Based on our phenotyping, CD1d tetramer binding, and iTCRα gene expression results, it is likely that the CD94+ T cell population described in this work contains a mixture of NK-like CD8+ T cells and a small portion of CD1d tetramer binding iNKT cells. This tetramer-ligand combination does not identify group 2 type 2 CD1d-restricted NKT cells or group 1 NKT-like cells, which may also be present within this population (Fig. 8). As observed in experimental and human Lyme disease, we were very interested to see significantly more CD1d/PBS-57 binding cells among PBMCs from dogs with clinical Lyme disease (Fig. 3A). These findings demonstrate the human CD1d/PBS-57 tetramer has utility for further studies of NKT cell responses in dogs during Lyme disease and other relevant pathological settings.
FIG 8
The majority of circulating canine CD94+ T cells were CD8+ (Fig. 3A). In addition to CD94, these CD8+ T cells expressed granzyme B and NKp46 more often than CD94− CD8+ T cells. This phenotype is consistent with NK-like CD8+ T cells that have been recently described (35, 37). NK-like CD8+ T cells are described to possess an effector-memory-reexpressing CD45RA (TEMRA) phenotype (37). Similar to conventional NKT cells, NK-like CD8+ T cells are thought to bridge innate immunity and adaptive immunity, allowing them to respond quickly to both IL-12/IL-18 stimulation and TCR-mediated signals with IFN-γ production and cytotoxic activity (34). Due to the high degree of functional overlap between these cell types, we hypothesize NK-like CD8+ T cells are likely to also be protective during Lyme disease. This would be consistent with their expansion only in asymptomatic B. burgdorferi-exposed dogs.
A caveat of using a naturally infected cohort is that the exact time point postexposure of subjects is not known. This could increase the variability among cellular response observations within each clinical group, if sampling occurred at different phases of the immune response. Further, categorization of dogs as asymptomatic in this study was based on serology and lack of clinical signs. However, antibodies against B. burgdorferi can persist for many months postexposure, unless antibiotic treatment occurs (50–52). The medical histories of the dogs indicated dogs categorized as asymptomatic had not received treatment for Lyme disease. Since B. burgdorferi PCR is unreliable in peripheral blood, we cannot differentiate asymptomatic dogs harboring active spirochetes from those that have cleared a recent infection (53–55).
In mice, IFN-γ produced during B. burgdorferi infection does not contribute to disease resistance (56). In human Lyme disease, the serum IFN-γ concentration is significantly elevated in active erythema migrans patients, correlates with an increased number of symptoms, and may promote chronic inflammation (19, 56–58). Although neither circulating CD3− CD94+ nor CD3+ CD94+ lymphocyte frequencies were modulated among symptomatic dog PBMCs in this report, a significantly augmented fraction of these cells produced IFN-γ when exposed to B. burgdorferi, indicating differential activation of these cells (Fig. 5B and C). NK cell priming by various cytokines, including IFN-α/β, IL-2, IL-12, and IL-18, has been described, where NK cells cultured with these cytokines secreted significantly increased inflammatory cytokines, including IFN-γ (59, 60). IL-2 and IL-18 have been shown to be elevated in acute Lyme disease patient sera: thus, increased sensitivity of canine CD94+ lymphocytes to B. burgdorferi during symptomatic Lyme disease may be due to cytokine priming (61).
Interestingly, we found IL-21 was significantly increased in asymptomatic, but not symptomatic, B. burgdorferi-seropositive dog serum. Activated CD4+ T cells, T follicular helper cells, and NKT cells are most commonly observed sources of IL-21 (62, 63). IL-21 can synergize with IL-2, IL-7, and IL-15 to decrease apoptosis and enhance proliferation of NKT cells and CD8+ T cells (39–41, 62, 64–66). Increased serum IL-21 may contribute to the expanded CD94+ T cell population in dogs with asymptomatic Lyme disease; however, this association needs to be strengthened with correlation analyses between serum IL-21 concentration and CD94+ cell frequencies among a larger sample of subjects.
How IL-21 exerts regulation is complex, often dependent on dose, differentiation state of targets, and cytokine milieu (64, 65). In peripheral human NK cells and CD8+ T cells, IL-21 reduced expression and activity of the activating receptor NKG2D, while upregulating other activating receptors, such as 2B4 and NKp30 on NK cells and CD28 on CD8+ T cells (67). Thus, IL-21 may alter the activation profile of NK-like CD8+ T cells in Borrelia-exposed dogs to tune their response.
Despite exacerbated IFN-γ induction by symptomatic dog CD94+ lymphocytes in response to B. burgdorferi, we observed decreased cytotoxicity by PBMCs from symptomatic dogs compared to seronegative dog PBMCs. One previous study found PBMCs from active Lyme disease patients also showed a defect in cytotoxicity (68). This study found that addition of B. burgdorferi spirochetes, or B. burgdorferi culture supernatant, within cytotoxicity assay cultures significantly reduced the ability of healthy control PBMCs to exert cytotoxic activity. Thus, the potential presence of B. burgdorferi spirochetes or secreted factors in actively infected dogs may affect their cytotoxic ability.
NK cell exhaustion has also been described in settings of chronic antigen exposure and inflammation, with both decreased NKp46 expression and impairment of NK cell cytolytic function characteristic of exhausted cells (69). In our assay, there was a significant positive association between calcein release and NKp46 expression on endemic control dog CD94+ T cells (70–72). However, this correlation was lost among B. burgdorferi-seropositive dog cells. In Fig. 1D, CD94+ T cells from both asymptomatic and symptomatic dogs expressed less NKp46 compared to posttreatment. However, cytotoxicity was only significantly reduced in the symptomatic group. IL-21 can augment IFN-γ production in some settings and is a potent enhancer of cytotoxic activity (40, 41, 66, 73). IL-21-expanded murine NKTs display increased intracellular granzyme B, cytotoxicity in vitro, and tumor control in vivo (39, 41, 66). As IL-21 is known to enhance cytotoxicity, we hypothesize increased serum IL-21 in asymptomatic dogs may buffer the cytotoxic capability of NK, NKT, and NK-like CD8+ T cells in these dogs. Further experiments using recombinant IL-21 during in vitro cytotoxicity assays would be needed to confirm this interaction.
In regions where B. burgdorferi is endemic, Ixodes scapularis also transmits Anaplasma phagocytophilum and coexposure is common (74–76). Although a subset of dogs was seropositive to both pathogens, our analyses did not reveal a negative synergistic effect of coexposure on CD94+ lymphocyte subsets (Fig. 7). This agrees with physical exam findings; dogs coexposed to B. burgdorferi and Anaplasma were no more likely to be symptomatic than dogs exposed to only B. burgdorferi (Table S2). Although the IDEXX 4Dx SNAP Plus test detects both A. phagocytophilum and Anaplasma platys, based on geography and B. burgdorferi coexposure, A. phagocytophilum is the most likely coinfecting species. Therefore, our data do not support an exacerbatory effect of A. phagocytophilum exposure on Lyme disease in this hunting dog cohort.
Together, we found expansion of an NK-like CD8+ T subset with reduced propensity to express IFN-γ and elevated serum IL-21 characterized dogs with asymptomatic Lyme disease. These cells were cytotoxic without excessive inflammation, leading to maintenance of a subclinical infection. CD94+ lymphocytes prone toward IFN-γ production with reduced cytotoxic activity characterized dogs with symptomatic disease. These biomarkers may aide diagnosis of Lyme disease in environments with transmission of multiple tick-borne pathogens. Regulatory therapies directed at limiting NK and NKT-like cell IFN-γ expression may improve the clinical course.
MATERIALS AND METHODS
Lyme disease cohort.
Naturally B. burgdorferi-exposed, owned hunting dogs were enrolled after caretakers provided informed consent (76). Subjects included foxhound and English springer spaniel breeds at high risk of exposure due to their hunting activities. All animal work was reviewed and approved by the University of Iowa Institutional Animal Care and Use Committee. A total of 717 dogs from Midwestern and Eastern United States were screened for seropositivity to Borrelia burgdorferi, Anaplasma phagocytophilum, Anaplasma platys, Ehrlichia ewingii, and Ehrlichia canis (SNAP 4Dx Plus test; IDEXX Laboratories, Inc.) (Table S1). All B. burgdorferi-seropositive samples were also tested via C6 quantitative enzyme-linked immunosorbent assay (ELISA), and those with a C6 value of >10 U/mL were enrolled (IDEXX Reference Laboratories).
Caretakers provided relevant medical history, and veterinarians performed a physical exam for signs of tick-borne illness, including lymphadenopathy, rash or petechiation, erythema, scleral injection, injected or pale mucous membranes, loss of body condition, and more specific signs of Lyme disease, including shifting lameness or other arthropathy without history of musculoskeletal injury or neurological signs. Doxycycline was administered to symptomatic dogs, and a subset of these cases were redrawn approximately 4 weeks posttreatment. Dogs were allocated into groups: endemic control (EC) (negative SNAP 4Dx Plus test), asymptomatic (AS) (B. burgdorferi seropositive with no clinical signs [33.5% of dogs B. burgdorferi seropositive]), symptomatic (SY) (B. burgdorferi seropositive and 2+ clinical signs [18.7% of dogs B. burgdorferi seropositive]), and treated (doxycycline-treated symptomatic dogs) (Table S1). There was no significant association between age or sex and Lyme disease clinical outcome in B. burgdorferi-seropositive dogs. Additionally we included evaluation of responses in Anaplasma-coexposed (B. burgdorferi and Anaplasma seropositive [47.8% of dogs Borrelia seropositive]) (Table S2).
Serum cytokine bioplex.
IL-21 and IL-23 serum protein concentrations were measured using a custom Milliplex MAP magnetic bead panel assay according to the manufacturer’s instructions (Millipore Sigma). Detection was performed on a Luminex 200 (Bio-Rad) instrument in the University of Iowa Flow Cytometry Core.
Peripheral blood mononuclear cell isolation.
Diluted blood was underlaid with Ficoll-Paque PLUS (GE Healthcare) and counted as previously described (77). Peripheral blood mononuclear cells (PBMCs) were resuspended at 5 × 106 cells/mL for further analyses in complete tissue culture medium (RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum [FBS], 2 mM l-glutamine, 1× penicillin-streptomycin, and 1× nonessential amino acids). In some experiments, recombinant canine IL-12 (rcIL-12) (R&D Systems) was used to stimulate PBMC cultures at 10 ng/mL.
Immunolabeling for flow cytometry and cell sorting.
PBMCs were blocked with 2 mg/mL cells dog gamma globulin (Jackson ImmunoResearch) in 50% FBS/PBS. Mouse anti-canine NKp46 clone 48A generated by J. A. Foltz and D. Lee from Nationwide Children’s Hospital (now commercially available as MABF2109 from Millipore Sigma) was applied at 4 μg/mL, followed by 1:50 anti-mouse IgG conjugated with allophycocyanin (APC) (Jackson ImmunoResearch) (31). A 1:25 dilution of anti-canine CD3 conjugated with fluorescein isothiocyanate (FITC) (clone CA17.2A12), 1:100 anti-canine CD8 conjugated with A700 (clone YCATE55.9; Bio-Rad Antibodies), and 1:25 anti-human CD94 conjugated with BV421 (clone HP-3D9; BD Biosciences) were used (29). Intracellular staining used 1:50 anti-human granzyme B conjugated with phycoerythrin (PE) and Texas Red clone GB11 (78–80) or 5 μg/mL anti-canine IFN-γ (MAB781; R&D) conjugated with Zenon R-PE (Thermo Fisher).
Unloaded or PBS-57-loaded human CD1d tetramers fluorescently tagged with APC were obtained from the NIH Tetramer Core Facility and stored at 4°C. For labeling of PBMCs, tetramers were included with other surface staining antibodies at 1:50 dilution for 60 min on ice, protected from light.
“Fluorescence minus one” controls were performed. Cell events were acquired at University of Iowa Flow Cytometry Core on an LSRII flow cytometer (BD Biosciences) and analyzed via FlowJo v10 software.
For flow-assisted cell sorting, PBMCs were stimulated with 10 ng/mL recombinant canine IL-2 (R&D Systems). Anti-human CD94-BV421 alone or in combination with anti-canine CD3-FITC was applied as described above. Cells were sorted using the FACS Aria II or FACS Aria Fusion fluorescence-activated cell sorter (Becton Dickenson) at the University of Iowa Flow Cytometry Core. Sorted PBMCs were cytospun onto SuperFrost slides, fixed briefly in methanol, and stained with Hema-3 solutions I and II. Coverslips were mounted with Permount, and bright-field images were obtained using cellSans Standard software on an Olympus BX51 microscope with a 100× objective and 1.30 aperture under oil immersion.
iTCRα gene expression.
The invariant variable-junction segment of the iTCRα gene utilized the iV4 forward primer (5′-CAGAGCTCCTTGCACATCACA-3′) and iJ reverse primer (5′-CAGGCCATACAGTTAGCTGAG-3′). The constant region segment was amplified using C2 forward primer (5′-GGCCCCCTCCGGTTTT-3′) and C3 reverse primer (5′-TGCCTGTTCACCGATTTTGA-3′). Real-time quantitative PCR (qPCR) was performed using Power SYBR master mix on a QuantStudio 3 thermocycler. Cycling conditions were based on those used by Yasuda at al. (33): 55 cycles of 30 s at 94°C, 35 s at 55°C, and 1 min at 72°C in triplicate. Fold change of the iVα-iJα mRNA segment was calculated using the threshold cycle (ΔΔCT) method with Cα as the internal reference gene and sorted CD3− CD94− lymphocytes used as the control group.
Borrelia burgdorferi culture.
Borrelia burgdorferi strain B31 (clone 5A1), NR-13251 (BEI Resources, NIAID, NIH), was grown in 10 mL of BSK-H medium with 6% rabbit serum (Sigma) at 34°C. Late-exponential-phase bacteria were enumerated with a hemocytometer by phase-contrast microscopy. B. burgdorferi spirochetes were applied to cells at a 1:10 ratio of PBMCs to spirochetes.
IFN-γ production assay.
PBMCs were cultured with medium, B. burgdorferi (1:10 B. burgdorferi), or phorbol myristate acetate (PMA) plus ionomycin (10 ng/mL and 0.5 μg/mL, respectively). PBMCs were incubated for 24 h. Brefeldin A (10 μg/mL) (Sigma) was added 8 h before harvest. IFN-γ-producing cells were analyzed by intracellular cytokine staining and flow cytometry, with gating as shown in Fig. 5A.
Cytotoxicity assay.
The cytotoxicity assay was based on that of Somanchi et al. (81), with modifications. PBMCs were stimulated with 10 ng/mL recombinant canine IL-2 overnight (R&D). Nonadherent PBMCs added to 100:1 effector-to-target (E:T) wells diluted to 50:1, 25:1, and 12.5:1 E:T. Calcein AM (Life Technologies)-loaded Madin-Darby Canine Kidney (MDCK) target cells were added, and plates were centrifuged at 100 × g to initiate cell contact. After 4 h of incubation, released calcein was read in supernatant (excitation, 485 nm; emission, 530 nm) (Omega software). The ratio of released calcein signal was calculated for each animal.
Statistical analyses.
Prior to analysis, outliers were removed by ROUT test (Q = 1%), and normality was assessed using D’Agostino-Pearson test. Nonparametric or parametric tests of variance were performed as appropriate. Replicate-matched two-way analysis of variance (ANOVA) with Sidak’s posttest was performed for comparisons of cell populations within the same animal. For all analyses, significance is indicated as follows: *, P ≤ 0.05; **, P < 0.01; ***, P < 0.001; and ****, P < 0.0001. The mean and standard deviation (SD) are indicated by error bars.
ACKNOWLEDGMENTS
We thank the animal caretakers for their time and participation in this study. We acknowledge Tracy Stokol and Martin Goodier for valuable discussion of markers to use to identify canine NK(T) cell subsets. Thank you to the NIH Tetramer Core Facility for providing CD1d tetramer materials. The data presented herein were obtained at the Flow Cytometry Facility, which is a Carver College of Medicine/Holden Comprehensive Cancer Center core research facility at the University of Iowa. The facility is funded through user fees and the generous financial support of the Carver College of Medicine, Holden Comprehensive Cancer Center, and Iowa City Veteran's Administration Medical Center.
This work was supported by the American Kennel Club Canine Health Foundation OAK 02383. This work was performed while Breanna Scorza was supported by University of Iowa Interdisciplinary Immunology Postdoctoral Training Grant NIH/NIAID T32AI007260.
We declare no conflict of interest.
Supplemental Material
File (iai.00555-21-s0001.pdf)
- Download
- 5.01 MB
REFERENCES
1.
Nelson CA, Saha S, Kugeler KJ, Delorey MJ, Shankar MB, Hinckley AF, Mead PS. 2015. Incidence of clinician-diagnosed Lyme disease, United States, 2005–2010. Emerg Infect Dis 21:1625–1631.
2.
Kugeler KJ, Schwartz AM, Delorey MJ, Mead PS, Hinckley AF. 2021. Estimating the frequency of Lyme disease diagnoses, United States, 2010–2018. Emerg Infect Dis 27:616–619.
3.
Eisen RJ, Eisen L. 2018. The blacklegged tick, Ixodes scapularis: an increasing public health concern. Trends Parasitol 34:295–309.
4.
Steere AC, Strle F, Wormser GP, Hu LT, Branda JA, Hovius JWR, Li X, Mead PS. 2016. Lyme borreliosis. Nat Rev Dis Primers 2:16090.
5.
Littman MP, Goldstein RE, Labato MA, Lappin MR, Moore GE. 2006. ACVIM small animal consensus statement on Lyme disease in dogs: diagnosis, treatment, and prevention. J Vet Intern Med 20:422–434.
6.
Lee W-Y, Sanz M-J, Wong CHY, Hardy P-O, Salman-Dilgimen A, Moriarty TJ, Chaconas G, Marques A, Krawetz R, Mody CH, Kubes P. 2014. Invariant natural killer T cells act as an extravascular cytotoxic barrier for joint-invading Lyme Borrelia. Proc Natl Acad Sci USA 111:13936–13941.
7.
Kinjo Y, Tupin E, Wu D, Fujio M, Garcia-Navarro R, Benhnia MR-E-I, Zajonc DM, Ben-Menachem G, Ainge GD, Painter GF, Khurana A, Hoebe K, Behar SM, Beutler B, Wilson IA, Tsuji M, Sellati TJ, Wong C-H, Kronenberg M. 2006. Natural killer T cells recognize diacylglycerol antigens from pathogenic bacteria. Nat Immunol 7:978–986.
8.
Tupin E, Benhnia MR-E-I, Kinjo Y, Patsey R, Lena CJ, Haller MC, Caimano MJ, Imamura M, Wong C-H, Crotty S, Radolf JD, Sellati TJ, Kronenberg M. 2008. NKT cells prevent chronic joint inflammation after infection with Borrelia burgdorferi. Proc Natl Acad Sci USA 105:19863–19868.
9.
Olson CM, Bates TC, Izadi H, Radolf JD, Huber SA, Boyson JE, Anguita J. 2009. Local production of IFN-gamma by invariant NKT cells modulates acute Lyme carditis. J Immunol 182:3728–3734.
10.
Kumar H, Belperron A, Barthold SW, Bockenstedt LK. 2000. Cutting edge: CD1d deficiency impairs murine host defense against the spirochete, Borrelia burgdorferi. J Immunol 165:4797–4801.
11.
Katchar K, Drouin EE, Steere AC. 2013. Natural killer cells and natural killer T cells in Lyme arthritis. Arthritis Res Ther 15:R183.
12.
Godfrey DI, Uldrich AP, McCluskey J, Rossjohn J, Moody DB. 2015. The burgeoning family of unconventional T cells. Nat Immunol 16:1114–1123.
13.
Looringh van Beeck FA, Zajonc DM, Moore PF, Schlotter YM, Broere F, Rutten VPMG, Willemse T, Van Rhijn I. 2008. Two canine CD1a proteins are differentially expressed in skin. Immunogenetics 60:315–324.
14.
Godfrey DI, Stankovic S, Baxter AG. 2010. Raising the NKT cell family. Nat Immunol 11:197–206.
15.
Schjaerff M, Keller SM, Affolter VK, Kristensen AT, Moore PF. 2016. Cellular endocytic compartment localization of expressed canine CD1 molecules. Vet Immunol Immunopathol 182:11–21.
16.
Schjaerff M, Keller SM, Fass J, Froenicke L, Grahn RA, Lyons L, Affolter VK, Kristensen AT, Moore PF. 2016. Refinement of the canine CD1 locus topology and investigation of antibody binding to recombinant canine CD1 isoforms. Immunogenetics 68:191–204.
17.
Stricker RB, Winger EE. 2001. Decreased CD57 lymphocyte subset in patients with chronic Lyme disease. Immunol Lett 76:43–48.
18.
Marques A, Brown MR, Fleisher TA. 2009. Natural killer cell counts are not different between patients with post-Lyme disease syndrome and controls. Clin Vaccine Immunol 16:1249–1250.
19.
Strle K, Sulka KB, Pianta A, Crowley JT, Arvikar SL, Anselmo A, Sadreyev R, Steere AC. 2017. T-helper 17 cell cytokine responses in Lyme disease correlate with Borrelia burgdorferi antibodies during early infection and with autoantibodies late in the illness in patients with antibiotic-refractory Lyme arthritis. Clin Infect Dis 64:930–938.
20.
Gross DM, Steere AC, Huber BT. 1998. T helper 1 response is dominant and localized to the synovial fluid in patients with Lyme arthritis. J Immunol 160:1022–1028.
21.
Strle K, Shin JJ, Glickstein LJ, Steere AC. 2012. Association of a Toll-like receptor 1 polymorphism with heightened Th1 inflammatory responses and antibiotic-refractory Lyme arthritis. Arthritis Rheum 64:1497–1507.
22.
Hasmim M, Khalife N, Zhang Y, Doldur M, Visentin G, Terry S, Giron-Michel J, Tang R, Delhommeau F, Dulphy N, Bourhis J-H, Louache F, Chouaib S. 2017. Expression of CD94 by ex vivo-differentiated NK cells correlates with the in vitro and in vivo acquisition of cytotoxic features. Oncoimmunology 6:e1346763.
23.
Van Beneden K, Stevenaert F, De Creus A, Debacker V, De Boever J, Plum J, Leclercq G. 2001. Expression of Ly49E and CD94/NKG2 on fetal and adult NK cells. J Immunol 166:4302–4311.
24.
Kawamura T, Takeda K, Kaneda H, Matsumoto H, Hayakawa Y, Raulet DH, Ikarashi Y, Kronenberg M, Yagita H, Kinoshita K, Abo T, Okumura K, Smyth MJ. 2009. NKG2A inhibits invariant NKT cell activation in hepatic injury. J Immunol 182:250–258.
25.
Backstrom E, Kristensson K, Ljunggren HG. 2004. Activation of natural killer cells: underlying molecular mechanisms revealed. Scand J Immunol 60:14–22.
26.
Iwaszko M, Bogunia-Kubik K. 2011. Clinical significance of the HLA-E and CD94/NKG2 interaction. Arch Immunol Ther Exp (Warsz) 59:353–367.
27.
Schuberth HJ, Kucinskiene G, Chu RM, Faldyna M. 2007. Reactivity of cross-reacting monoclonal antibodies with canine leukocytes, platelets and erythrocytes. Vet Immunol Immunopathol 119:47–55.
28.
Narni-Mancinelli E, Chaix J, Fenis A, Kerdiles YM, Yessaad N, Reynders A, Gregoire C, Luche H, Ugolini S, Tomasello E, Walzer T, Vivier E. 2011. Fate mapping analysis of lymphoid cells expressing the NKp46 cell surface receptor. Proc Natl Acad Sci USA 108:18324–18329.
29.
Graves SS, Gyurkocza B, Stone DM, Parker MH, Abrams K, Jochum C, Gallo S, Saad M, Johnson MM, Rosinski SL, Storb R. 2019. Development and characterization of a canine-specific anti-CD94 (KLRD-1) monoclonal antibody. Vet Immunol Immunopathol 211:10–18.
30.
Yu J, Wei M, Mao H, Zhang J, Hughes T, Mitsui T, Park I-k, Hwang C, Liu S, Marcucci G, Trotta R, Benson DM, Caligiuri MA. 2009. CD94 defines phenotypically and functionally distinct mouse NK cell subsets. J Immunol 183:4968–4974.
31.
Foltz JA, Somanchi SS, Yang Y, Aquino-Lopez A, Bishop EE, Lee DA. 2016. NCR1 expression identifies canine natural killer cell subsets with phenotypic similarity to human natural killer cells. Front Immunol 7:521.
32.
Hadad U, Thauland TJ, Martinez OM, Butte MJ, Porgador A, Krams SM. 2015. NKp46 clusters at the immune synapse and regulates NK cell polarization. Front Immunol 6:495.
33.
Yasuda N, Masuda K, Tsukui T, Teng A, Ishii Y. 2009. Identification of canine natural CD3-positive T cells expressing an invariant T-cell receptor alpha chain. Vet Immunol Immunopathol 132:224–231.
34.
Jacomet F, Cayssials E, Basbous S, Levescot A, Piccirilli N, Desmier D, Robin A, Barra A, Giraud C, Guilhot F, Roy L, Herbelin A, Gombert J-M. 2015. Evidence for eomesodermin-expressing innate-like CD8(+) KIR/NKG2A(+) T cells in human adults and cord blood samples. Eur J Immunol 45:1926–1933.
35.
Pita-Lopez ML, Pera A, Solana R. 2016. Adaptive memory of human NK-like CD8(+) T-cells to aging, and viral and tumor antigens. Front Immunol 7:616.
36.
Wang C, Liu X, Li Z, Chai Y, Jiang Y, Wang Q, Ji Y, Zhu Z, Wan Y, Yuan Z, Chang Z, Zhang M. 2015. CD8(+)NKT-like cells regulate the immune response by killing antigen-bearing DCs. Sci Rep 5:14124.
37.
Barbarin A, Cayssials E, Jacomet F, Nunez NG, Basbous S, Lefèvre L, Abdallah M, Piccirilli N, Morin B, Lavoue V, Catros V, Piaggio E, Herbelin A, Gombert J-M. 2017. Phenotype of NK-like CD8(+) T cells with innate features in humans and their relevance in cancer diseases. Front Immunol 8:316.
38.
Correia MP, Stojanovic A, Bauer K, Juraeva D, Tykocinski L-O, Lorenz H-M, Brors B, Cerwenka A. 2018. Distinct human circulating NKp30(+)FcepsilonRIgamma(+)CD8(+) T cell population exhibiting high natural killer-like antitumor potential. Proc Natl Acad Sci USA 115:E5980–E5989.
39.
Smyth MJ, Wallace ME, Nutt SL, Yagita H, Godfrey DI, Hayakawa Y. 2005. Sequential activation of NKT cells and NK cells provides effective innate immunotherapy of cancer. J Exp Med 201:1973–1985.
40.
Coquet JM, Kyparissoudis K, Pellicci DG, Besra G, Berzins SP, Smyth MJ, Godfrey DI. 2007. IL-21 is produced by NKT cells and modulates NKT cell activation and cytokine production. J Immunol 178:2827–2834.
41.
Ngai H, Tian G, Courtney AN, Ravari SB, Guo L, Liu B, Jin J, Shen ET, Di Pierro EJ, Metelitsa LS. 2018. IL-21 selectively protects CD62L(+) NKT cells and enhances their effector functions for adoptive immunotherapy. J Immunol 201:2141–2153.
42.
Liu SM, King C. 2013. IL-21-producing Th cells in immunity and autoimmunity. J Immunol 191:3501–3506.
43.
Brown CR, Reiner SL. 1999. Experimental Lyme arthritis in the absence of interleukin-4 or gamma interferon. Infect Immun 67:3329–3333.
44.
Ma Y, Sturrock A, Weis JJ. 1991. Intracellular localization of Borrelia burgdorferi within human endothelial cells. Infect Immun 59:671–678.
45.
Girschick HJ, Huppertz HI, Russmann H, Krenn V, Karch H. 1996. Intracellular persistence of Borrelia burgdorferi in human synovial cells. Rheumatol Int 16:125–132.
46.
Livengood JA, Gilmore RD, Jr. 2006. Invasion of human neuronal and glial cells by an infectious strain of Borrelia burgdorferi. Microbes Infect 8:2832–2840.
47.
Chmielewski T, Tylewska-Wierzbanowska S. 2010. Interactions between Borrelia burgdorferi and mouse fibroblasts. Pol J Microbiol 59:157–160.
48.
Petnicki-Ocwieja T, Kern A. 2014. Mechanisms of Borrelia burgdorferi internalization and intracellular innate immune signaling. Front Cell Infect Microbiol 4:175.
49.
Brown WC. 2012. Adaptive immunity to Anaplasma pathogens and immune dysregulation: implications for bacterial persistence. Comp Immunol Microbiol Infect Dis 35:241–252.
50.
Appel MJ, Allan S, Jacobson RH, Lauderdale TL, Chang YF, Shin SJ, Thomford JW, Todhunter RJ, Summers BA. 1993. Experimental Lyme disease in dogs produces arthritis and persistent infection. J Infect Dis 167:651–664.
51.
Levy SA, O'Connor TP, Hanscom JL, Shields P, Lorentzen L, Dimarco AA. 2008. Quantitative measurement of C6 antibody following antibiotic treatment of Borrelia burgdorferi antibody-positive nonclinical dogs. Clin Vaccine Immunol 15:115–119.
52.
Chandrashekar R, Beall MJ, Thatcher B, Saucier JM, Tyrrell P, Lappin MR. 2017. Serologic responses to peptides of Anaplasma phagocytophilum and Borrelia burgdorferi in dogs infested with wild-caught Ixodes scapularis. Vet J 226:6–11.
53.
Goodman JL, Bradley JF, Ross AE, Goellner P, Lagus A, Vitale B, Berger BW, Luger S, Johnson RC. 1995. Bloodstream invasion in early Lyme disease: results from a prospective, controlled, blinded study using the polymerase chain reaction. Am J Med 99:6–12.
54.
Aguero-Rosenfeld ME, Wormser GP. 2015. Lyme disease: diagnostic issues and controversies. Expert Rev Mol Diagn 15:1–4.
55.
Liang L, Wang J, Schorter L, Nguyen Trong TP, Fell S, Ulrich S, Straubinger RK. 2020. Rapid clearance of Borrelia burgdorferi from the blood circulation. Parasit Vectors 13:191.
56.
Glickstein L, Edelstein M, Dong JZ. 2001. Gamma interferon is not required for arthritis resistance in the murine Lyme disease model. Infect Immun 69:3737–3743.
57.
Widhe M, Jarefors S, Ekerfelt C, Vrethem M, Bergström S, Forsberg P, Ernerudh J. 2004. Borrelia-specific interferon-gamma and interleukin-4 secretion in cerebrospinal fluid and blood during Lyme borreliosis in humans: association with clinical outcome. J Infect Dis 189:1881–1891.
58.
Dame TM, Orenzoff BL, Palmer LE, Furie MB. 2007. IFN-gamma alters the response of Borrelia burgdorferi-activated endothelium to favor chronic inflammation. J Immunol 178:1172–1179.
59.
Chaix J, Tessmer MS, Hoebe K, Fuséri N, Ryffel B, Dalod M, Alexopoulou L, Beutler B, Brossay L, Vivier E, Walzer T. 2008. Cutting edge: priming of NK cells by IL-18. J Immunol 181:1627–1631.
60.
Sabry M, Zubiak A, Hood SP, Simmonds P, Arellano-Ballestero H, Cournoyer E, Mashar M, Pockley AG, Lowdell MW. 2019. Tumor- and cytokine-primed human natural killer cells exhibit distinct phenotypic and transcriptional signatures. PLoS One 14:e0218674.
61.
Soloski MJ, Crowder LA, Lahey LJ, Wagner CA, Robinson WH, Aucott JN. 2014. Serum inflammatory mediators as markers of human Lyme disease activity. PLoS One 9:e93243.
62.
Leonard WJ, Wan CK. 2016. IL-21 signaling in immunity. F1000Res 5:224.
63.
Solaymani-Mohammadi S, Eckmann L, Singer SM. 2019. Interleukin (IL)-21 in inflammation and immunity during parasitic diseases. Front Cell Infect Microbiol 9:401.
64.
Parrish-Novak J, Dillon SR, Nelson A, Hammond A, Sprecher C, Gross JA, Johnston J, Madden K, Xu W, West J, Schrader S, Burkhead S, Heipel M, Brandt C, Kuijper JL, Kramer J, Conklin D, Presnell SR, Berry J, Shiota F, Bort S, Hambly K, Mudri S, Clegg C, Moore M, Grant FJ, Lofton-Day C, Gilbert T, Rayond F, Ching A, Yao L, Smith D, Webster P, Whitmore T, Maurer M, Kaushansky K, Holly RD, Foster D. 2000. Interleukin 21 and its receptor are involved in NK cell expansion and regulation of lymphocyte function. Nature 408:57–63.
65.
Parrish-Novak J, Foster DC, Holly RD, Clegg CH. 2002. Interleukin-21 and the IL-21 receptor: novel effectors of NK and T cell responses. J Leukoc Biol 72:856–863.
66.
Seo H, Jeon I, Kim B-S, Park M, Bae E-A, Song B, Koh C-H, Shin K-S, Kim I-K, Choi K, Oh T, Min J, Min BS, Han YD, Kang S-J, Shin SJ, Chung Y, Kang C-Y. 2017. IL-21-mediated reversal of NK cell exhaustion facilitates anti-tumour immunity in MHC class I-deficient tumours. Nat Commun 8:15776.
67.
Burgess SJ, Marusina AI, Pathmanathan I, Borrego F, Coligan JE. 2006. IL-21 down-regulates NKG2D/DAP10 expression on human NK and CD8+ T cells. J Immunol 176:1490–1497.
68.
Golightly M, Thomas J, Volkman D, Dattwyler R. 1988. Modulation of natural killer cell activity by Borrelia burgdorferi. Ann N Y Acad Sci 539:103–111.
69.
Bi J, Tian Z. 2017. NK cell exhaustion. Front Immunol 8:760.
70.
De Maria A, Fogli M, Costa P, Murdaca G, Puppo F, Mavilio D, Moretta A, Moretta L. 2003. The impaired NK cell cytolytic function in viremic HIV-1 infection is associated with a reduced surface expression of natural cytotoxicity receptors (NKp46, NKp30 and NKp44). Eur J Immunol 33:2410–2418.
71.
Sivori S, Pende D, Bottino C, Marcenaro E, Pessino A, Biassoni R, Moretta L, Moretta A. 1999. NKp46 is the major triggering receptor involved in the natural cytotoxicity of fresh or cultured human NK cells. Correlation between surface density of NKp46 and natural cytotoxicity against autologous, allogeneic or xenogeneic target cells. Eur J Immunol 29:1656–1666.
72.
Dons'koi BV, Osypchuk DV, Chernyshov VP. 2019. Enumeration of peripheral blood NKp46 positive NK lymphocytes reflects NK cytotoxic activity in vitro. J Immunol Methods 474:112639.
73.
Smyth MJ, Cretney E, Kelly JM, Westwood JA, Street SEA, Yagita H, Takeda K, van Dommelen SLH, Degli-Esposti MA, Hayakawa Y. 2005. Activation of NK cell cytotoxicity. Mol Immunol 42:501–510.
74.
Diuk-Wasser MA, Vannier E, Krause PJ. 2016. Coinfection by Ixodes tick-borne pathogens: ecological, epidemiological, and clinical consequences. Trends Parasitol 32:30–42.
75.
Nieto NC, Foley JE. 2009. Meta-analysis of coinfection and coexposure with Borrelia burgdorferi and Anaplasma phagocytophilum in humans, domestic animals, wildlife, and Ixodes ricinus-complex ticks. Vector Borne Zoonotic Dis 9:93–102.
76.
Mahachi K, Kontowicz E, Anderson B, Toepp AJ, Lima AL, Larson M, Wilson G, Grinnage-Pulley T, Bennett C, Ozanne M, Anderson M, Fowler H, Parrish M, Saucier J, Tyrrell P, Palmer Z, Buch J, Chandrashekar R, Scorza B, Brown G, Oleson JJ, Petersen CA. 2020. Predominant risk factors for tick-borne co-infections in hunting dogs from the USA. Parasit Vectors 13:247.
77.
Esch KJ, Juelsgaard R, Martinez PA, Jones DE, Petersen CA. 2013. Programmed death 1-mediated T cell exhaustion during visceral leishmaniasis impairs phagocyte function. J Immunol 191:5542–5550.
78.
Lee S-H, Shin D-J, Kim Y, Kim C-J, Lee J-J, Yoon MS, Uong TNT, Yu D, Jung J-Y, Cho D, Jung B-G, Kim S-K, Suh G-H. 2018. Comparison of phenotypic and functional characteristics between canine non-B, non-T natural killer lymphocytes and CD3(+)CD5(dim)CD21(−) cytotoxic large granular lymphocytes. Front Immunol 9:841.
79.
Rabiger FV, Bismarck D, Protschka M, Köhler G, Moore PF, Büttner M, von Buttlar H, Alber G, Eschke M. 2019. Canine tissue-associated CD4+CD8alpha+ double-positive T cells are an activated T cell subpopulation with heterogeneous functional potential. PLoS One 14:e0213597.
80.
Pantelyushin S, Ranninger E, Guerrera D, Hutter G, Maake C, Markkanen E, Bettschart-Wolfensberger R, Rohrer Bley C, Läubli H, Vom Berg J. 2021. Cross-reactivity and functionality of approved human immune checkpoint blockers in dogs. Cancers (Basel) 13:785.
81.
Somanchi SS, Senyukov VV, Denman CJ, Lee DA. 2011. Expansion, purification, and functional assessment of human peripheral blood NK cells. J Vis Exp.
Information & Contributors
Information
Published In
Infection and Immunity
Volume 90 • Number 5 • 19 May 2022
eLocator: e00555-21
Editor: De'Broski R. Herbert, University of Pennsylvania
PubMed: 35416707
Copyright
© 2022 American Society for Microbiology. All Rights Reserved.
History
Received: 11 October 2021
Returned for modification: 11 November 2021
Accepted: 25 March 2022
Published online: 13 April 2022
Keywords
Contributors
Editor
De'Broski R. Herbert
Editor
University of Pennsylvania
Metrics & Citations
Metrics
Note:
- For recently published articles, the TOTAL download count will appear as zero until a new month starts.
- There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
- Citation counts come from the Crossref Cited by service.
Citations
Citation
2022.
Role of NK-Like CD8+ T Cells during Asymptomatic Borrelia burgdorferi Infection. Infect Immun
90:e00555-21.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.