Lyme disease (LD) is the most prevalent tick-borne disease in North America and Europe with an estimate of ~476,000 annual cases in the United States (U.S.) alone (Mead, 2015; Kugeler et al., 2021). LD is caused by some members of Borreliella burgdorferi (Bb) sensu lato complex, which is divided into more than 20 genospecies. At least three genospecies (Borreliella afzelii, Bb sensu stricto, Borreliella garinii) have the capacity to cause LD in humans (Baranton et al., 1992; Canica et al., 1993; Kurtenbach et al., 2006). The horizontal transmission of LD is mediated by hard ticks of Ixodes spp., which acquire spirochetes from Bb-infected vertebrate carriers such as rodents and birds (Piesman and Gern, 2004; Barbour and Gupta, 2021). In the U.S., the white-footed mouse, Peromyscus leucopus, is considered the main mammalian reservoir of Bb spirochetes (Levine et al., 1985; Donahue et al., 1987). Clinical signs of LD patients are often flu-like symptoms that can be followed by arthritis and neurological or cardiac abnormalities (Smith et al., 2011; Radolf et al., 2021).

When Bb infects mammals, innate immune effectors such as toll-like receptor (TLR) 2 and myeloid differentiation antigen 88 (MyD88) become fully engaged (Hirschfeld et al., 1999; Wooten et al., 2002a; Wooten et al., 2002b; Bolz et al., 2004). MyD88 subsequently activates a toll/IL-1 receptor-domain-containing adapter-inducing interferon-β (TRIF) signaling (Petnicki-Ocwieja et al., 2013). Moreover, Bb outer surface lipoproteins lead to activation of nuclear factor kappa-light-chain-enhancer (NF-κB) (Wooten et al., 1996; Ebnet et al., 1997). This cascade leads to secretion of pro-inflammatory cytokines and chemokines for neutrophils, monocytes, and lymphocytes during LD infection in mice (Wooten and Weis, 2001; Verhaegh et al., 2017; Bockenstedt et al., 2021). The mounting immune response is also accompanied by production of interferons (IFN), IFNα and IFNβ (type I INF), and IFNγ in response to Bb antigens (Ganapamo et al., 2000; Crandall et al., 2006; Miller et al., 2008). As a result of the activated innate immune system, the acquired humoral immune response is developed, and this leads to lowering the spirochetal burden in various tissues (Barthold et al., 1997; McKisic and Barthold, 2000). In LD patients, Th1 lymphocytes and mononuclear cells mediate B lymphocyte production of neutralizing anti-Bb antibodies such as IgG1 and IgG3, which induce opsonization and activate complement defense mechanisms (Hechemy et al., 1988; Bockenstedt et al., 2021).

Similar to LD human patients, some laboratory mouse strains develop arthritis and carditis upon LD infection (Barthold et al., 1990; Armstrong et al., 1992; Brown and Reiner, 1999). The degree of susceptibility to Bb-induced arthritis in mice is predetermined by a number of factors including inter-strain genetic variability and the fitness variation of Bb strains (Barthold et al., 1992; Ma et al., 1998; Hanincová et al., 2008; Lin et al., 2014). The different capacity of mouse strains to regulate the localized inflammatory response is a major factor for the arthritis development (Ma et al., 1998). By using C3H/He mice, which develop severe Lyme arthritis (Barthold et al., 1990; Barthold et al., 1992), it was demonstrated that IFNα and IFNβ have an important role during an early LD infection. The upregulation of the interferon responsive genes contributes to the pathogenesis of LD arthritis in C3H/He mice (Crandall et al., 2006; Miller et al., 2008). Type I INF response can also be associated with differential regulation of genes that are involved in tissue repair processes (Crandall et al., 2006; Lochhead et al., 2021). In contrast to C3H strains, C57BL/6 mice, which are deficient in an intense interferon response to Bb infection (Crandall et al., 2006), develop only mild Lyme arthritis (Barthold et al., 1990; Crandall et al., 2006). In addition to type I IFN, interleukin 10 (IL-10), an anti-inflammatory cytokine, has a crucial role in the regulation of arthritis severity (Sonderegger et al., 2012). C57BL/6 mice lacking IL-10 exhibit more severe arthritis upon LD infection as opposed to their Bb-infected wild-type C57BL/6 controls (Brown et al., 1999).

In contrast to C3H mice, previous studies demonstrated that experimentally Bb-infected P. leucopus mice do not develop disease (Schwanz et al., 2011). Both Bb-infected and uninfected control P. leucopus mice were shown to have similar white blood cell counts and wheel-running activity (Schwanz et al., 2011). The resistance of persistently Bb-infected P. leucopus mice to the disease was also shown to be age-dependent since only infant mice could develop carditis and multifocal arthritis upon Bb infection (Moody et al., 1994). Of note, however, the tolerance of white-footed mice to Bb-induced disease is not absolute. A previous study demonstrated that a fraction (e.g., ~7-10%) of wild-caught P. leucopus population had clinical signs of LD (Burgess et al., 1990). The lack of detectable LD in the majority of Bb-infected P. leucopus population can potentially be explained by a long-term co-existence between the host and the pathogen. The outcome of this host-pathogen adaptation is likely to have created a balance between the strong but non-sterilizing immune response of reservoir hosts (Schwan et al., 1989; Baum et al., 2012; Cook and Barbour, 2015; Rogovskyy et al., 2015) and the persistent presence of Bb spirochetes. Unfortunately, to date, the exact factors and mechanism of tolerance or susceptibility of P. leucopus mice to Bb-induced disease are unknown. In the present study, we have explored infection-induced changes in gene expression to understand how P. leucopus mice are able to tolerate Bb infection. For that, we compared responses to Bb infection between P. leucopus and Mus musculus (C3H/HeJ, referred to here as C3H) mice at the transcriptomic level. The overall data showed that the spleen transcriptome of P. leucopus mice was much more quiescent compared to that of the C3H mice.

In the present investigation, comparative transcriptomic analyses of spleens harvested from Bb-infected and uninfected P. leucopus mice have been performed. The spleen was selected because it is one of the major lymphoid organs that together with the bone marrow contains most immune cells (Zhao et al., 2012; Bronte and Pittet, 2013). The experimental design also included the laboratory “control” mouse strain – Bb-infected and uninfected C3H mice. To understand how P. leucopus mice are able to tolerate Bb infection, the transcriptome data generated from spleens of infected and uninfected mice were analyzed within each animal species by three complementing bio-molecular tools. The overall data demonstrated that Bb-infected P. leucopus mice had five times lower number of DEGs (n=81) than infected C3H mice (n=421). Out of the 81 DEGs, approximately 57% (n=46) and 43% (n=35) of genes were found to be down- and up-regulated, respectively. As opposed to the 46 down-regulated genes, many of whom were associated with immune response, the 35 up-regulated genes were not related to any immune response pathways. The overall data demonstrated that the Bb-induced response of the spleen transcriptome was more quiescent in P. leucopus mice compared to that of C3H mice.

The present results are similar to the findings of a previous study, where the spleen transcriptome analyses of B. afzelii-infected and uninfected bank voles (Myodes glareolus) and yellow-necked mice (Apodemus flavicollis) were performed (Zhong et al., 2020). Although both animal species are considered natural reservoir hosts of B. afzelii in Europe (Zhong et al., 2019), Borreliella-infected bank voles have on average a magnitude higher spirochetal loads in their tissues (Raberg, 2012; Zhong et al., 2019). The comparative analyses between infected and uninfected animals within each species identified only 8 and 5 DEGs in B. afzelii-infected bank voles and yellow-necked mice, respectively (Zhong et al., 2020). Furthermore, similar to the current data, other study had also shown no significant involvement of inflammation or innate immunity genes, when transcriptomes of skin and blood samples taken at 5 weeks postinfection were compared between Bb strain Sh-2-82-infected and uninfected P. leucopus mice (Long et al., 2019). Overall, the current and previous results could collectively suggest that reservoir hosts of LD pathogens moderate the immune system to avoid development of exaggerated (immune) responses.

A few examples of immune related genes that, during the persistent Bb infection, were down-regulated in P. leucopus mice and remained unresponsive in C3H mice include TLR7-encoding gene, type I IFN-associated genes, and members of the lymphocyte antigen-6 (Ly6) gene family. TLR7 recognizes bacterial RNA and peptidoglycan, and becomes activated in response to bacterial exposure by dendritic cells (DCs) (Mancuso et al., 2009). It was shown that TLR7 mRNA became up-regulated in Bb-exposed human oligodendrocyte cell line (Parthasarathy and Philipp, 2018). Overall, activation of TLRs by Bb lipoproteins leads to the induction of type I IFN (Petzke et al., 2009; Love et al., 2014a; Lochhead et al., 2021), which in turn primes macrophages for producing proinflammatory cytokines (Bockenstedt et al., 2021).

Through the use of laboratory mouse strains of Mus musculus, it has been consistently demonstrated that the severity of LD arthritis is driven by type I IFN (Crandall et al., 2006; Miller et al., 2008). Early type I IFN induction also positively correlates with the Bb capacity to disseminate in C3H/HeJ mice (Petzke et al., 2016). Type I IFN signaling may also lead to the accumulation of naïve B lymphocytes in Bb-infected mice (Hastey et al., 2014). Surprisingly, the present study did not identify any DEGs associated with the interferon signaling pathways in Bb-infected C3H mice. The latter cannot be solely explained by the lack of a functional TLR4 pathway in this particular mouse strain (Poltorak et al., 1998), partly because the classic enterobacterium-type lipopolysaccharide (LPS) (Brubaker et al., 2015), one of TLR4 ligands, is not produced by Bb spirochetes (Takayama et al., 1987). In contrast, a number of types I IFN-associated genes (n=6) were down-regulated in Bb-infected P. leucopus mice. These findings are consistent with the previous study, which demonstrated that a set of genes involved in IFNα signaling was down-regulated in B. afzelii-infected bank voles and yellow-necked mice compared to their uninfected controls (Zhong et al., 2020). In addition to type I IFN-associated genes, two members of Ly6 gene family were also found to be down-regulated in Bb-infected P. leucopus mice. The 6E and 6A-2/6E-1-like proteins are known to be expressed by different types of mouse and human immune cells (e.g., B and T lymphocytes, natural killer (NK) cells, monocytes, and DCs) and involved in regulation of activation, proliferation, and differentiation of T lymphocytes (Ortega et al., 1986; Toulon et al., 1988; Lee et al., 2013).

Notably, of the 421 DEGs identified in Bb-infected C3H mice, a number of genes (n=38) were associated with different aspects of hemostasis. In the blood coagulation pathway alone, there were a total of 19 down-regulated genes with some involved in the regulation of blood coagulation (n=10) and others associated with platelet activation (n=9). Similar results were obtained by an earlier study, where blood, spleen and liver transcriptomes of LPS-treated-treated Mus musculus (BALB/cAnNCrl, referred here as BALB/c) and P. leucopus mice were compared to their respective non-treated controls (Balderrama-Gutierrez et al., 2021). The previous data showed that it was only in the blood of LPS-treated BALB/c mice where DEGs associated with negative regulation of blood coagulation pathways were detected (Balderrama-Gutierrez et al., 2021). The present study also demonstrated that the Cela2a and Cd40lg genes were up-regulated only in Bb-infected C3H mice. Cela2a hydrolyzes elastin and reduces platelet hyperactivation (Esteghamat et al., 2019), and Cd40lg is involved in immunoglobulin class switching, activation of B cell and T cell proliferation, enhancement of IL10 production, and induction of NF-kappa-B production (Armitage et al., 1992; Borrow et al., 1996; Van Kooten and Banchereau, 1996). As opposed to P. leucopus, Bb-infected C3H mice had also an overabundance of up-regulated cell cycle genes (n=50), whose products are known to play an important role in the host immune response (e.g., macrophage expansion and antibody class switching) (Laphanuwat and Jirawatnotai, 2019).

Overall, there were only 3 DEGs that were commonly identified in both animal species when infected with Bb ( Table 1 ). The HMGB3 gene was up- and down-regulated in Bb-infected C3H and P. leucopus mice, respectively. As multifunctional molecules, HMGB proteins are proposed to play a role in the nucleic-acid-mediated activation of innate immune responses (Yanai et al., 2009). Being part of the cytosolic receptor–IRF3/NF-kB signaling pathway (Yanai et al., 2009), HMGB proteins can also stimulate TLR2, which, through its engagement with Bb lipoproteins, initiates expression of NF-κB-dependent cytokines (Shin et al., 2008). Another example of “shared” DEGs includes the Slc16a2 gene, which encodes solute carrier family 16 member 2 (monocarboxylate transporter 8) in P. leucopus and C3H mice. This active and specific thyroid hormone transporter mediates cellular uptake and transmembrane transporter activity of thyroxine (T4), triiodothyronine (T3), reverse triiodothyronine (rT3) and diidothyronine (Friesema et al., 2003; Friesema et al., 2005). Thyroid hormones have an important role in activation of neutrophils, NK cells, DCs, and B and T lymphocytes during infection (Rubingh et al., 2020). The corresponding genes were up- and down-regulated in Bb-infected P. leucopus and C3H mice, respectively. Lastly, both mouse species had only one “shared” gene up-regulated in response to Bb infection. The Try5 gene encodes anionic trypsin-2-like protein and trypsin 5 in P. leucopus and C3H mice, respectively.

Some limitations of this study include small sample sizes and the use of male mice only. Another caveat is that spirochetal burdens were not determined in spleens of the infected mice to see if potentially higher bacterial loads in C3H mice would explain overall higher numbers of DEGs identified in the laboratory mice compared to P. leucopus mice. In addition to addressing the above limitations, future studies are also warranted to confirm the identified DEGs through other approaches (e.g., qRT-PCR, proteomics).

In summary, the present investigation is one of the few that have examined the transcriptome response of mammalian reservoir hosts to Borreliella infection. Although the experimental design of this study is different from the above-mentioned earlier investigations (e.g., reservoir host, Borreliella strain, time point of harvest, tissue type analyzed), the collective results of all studies have consistently demonstrated a very limited response of natural reservoir hosts to the persistent infection of LD pathogens. In order to decipher tolerance mechanism of P. leucopus and other reservoir hosts to Bb infection, future investigations that would pursue some of the data leads identified in the present and earlier studies are highly warranted.

The data presented in the study are deposited in the GenBank Sequence Read Archive (SRA) depository and the SRA accession numbers are SAMN32740077, SAMN32740078, SAMN32740079, SAMN32740080, SAMN32740081, SAMN32740082, SAMN32740083, SAMN32740084, SAMN32740085, SAMN32740086, SAMN32740087, SAMN32740088.

The animal study was reviewed and approved by the Institutional Animal Care and Use Committee of Texas A&M University.

AG analyzed the data and produced the manuscript draft. AH performed sequencing and raw data assembly. IM and AZ were involved in the data analysis. CN and HP analyzed the data, contributed to the data visualization and writing of the manuscript. DT provided resources and contributed to the data analysis. AR concepted and developed the study, provided resources, oversaw the project, and wrote the manuscript. All authors contributed to the article and approved the submitted version.