The genus Borrelia (B.) comprises the causative agents of Lyme disease (LD) and relapsing fever (1–3). Concerning LD spirochetes, there are distinct species belonging to the Borrelia burgdorferi sensu lato complex of which six species including B. burgdorferi sensu stricto (s.s.), Borrelia afzelii, Borrelia garinii, Borrelia spielmanii, Borrelia bavariensis (formerly referred to as B. garinii OspA serotype 4), as well as candidatus Borrelia mayonii, are associated with human LD (4, 5). While Borrelia valaisiana, Borrelia lusitaniae, and Borrelia bissettii have been detected in human biopsies, the pathogenicity of these and the remaining borrelial species remains largely unclear (6–8).

To survive and establish a persistent infection in the human host, pathogens must evade the first line of host defense by counteracting complement as an essential part of innate immunity. This powerful surveillance system comprises a network of precursors, regulatory and inhibitory proteins that can be immediately activated upon recognition of invading microorganisms (9, 10). Despite the effectiveness and abundance of complement, LD spirochetes are able to overcome its destructive defense mechanisms (11–13). While attempting to decipher the molecular mechanisms of complement evasion, distinct complement-interfering and -inhibiting molecules of B. burgdorferi s.s., B. afzelii, B. spielmanii, and B. bavariensis have been identified (14, 15). This review focuses on the current knowledge of the molecular principles utilized by LD spirochetes to counteract complement at certain activation levels and on the borrelial proteins known to take part in complement inactivation.

Complement operates as a cooperative network of inactive precursor molecules, fluid-phase and membrane-bound regulators, and inhibitors (16–20). The initiation of complement takes place in a cascade-like manner through three activation routes: the classical (CP), the lectin (LP), and the alternative pathway (AP), all of which converge in the generation of the highly reactive molecule, C3b (17, 19, 21, 22). The CP can be activated after binding of C1q to immune complexes (IgM and IgG) or charged molecules on the bacterial surface (23). In complex with C1q and C1r, C1s mediates cleavage of C4 and C2 leading to the formation of the C3 convertase, C4b2b. Activation of the LP is initiated by binding of mannan-binding lectin (MBL), ficolins (H-ficolin, L-ficolin, and M-ficolin), or collectins associated with MBL-associated serine proteases (MASP), to carbohydrates of microbial origin. After activation of MASP-2 by MASP-1, both proteases cleave C2 while MASP-2 is able to also cleave C4 to generate the identical C3 convertase. Finally, the AP is initiated by spontaneous hydrolysis of C3 followed by binding of C3b (opsonization) to different targets on the bacterial surface. Recruitment of Factor B (FB) followed by Factor D (FD)-mediated cleavage results in the formation of the membrane-bound C3 convertase C3bBb. To extend the half-life and to trigger the amplification of C3b (feedback loop), the C3 convertases of the AP are stabilized by properdin. Of note, deposition of large quantities of C3b on microbial surfaces is a prerequisite for opsonization and phagocytosis of invading pathogens. Upon binding of newly generated C3b molecules, the C4b2b and C3Bb complexes serve as precursors for the C5 convertases C4b2b3b and C3bBb3b. Cleavage of C5 into C5a and C5b by the C5 convertases initiates the unidirectional, sequential binding of the late components C6, C7, and C8 to C5b. Once the C5b–8 complex is formed, polymerization of multiple C9 molecules ensues, finally generating the pore-forming terminal complement complex [C5b–9, TCC, also referred to as membrane attack complex (MAC)], which promotes lysis of susceptible cells (19, 20). To avoid the detrimental effects of excessive complement activation, this surveillance system is tightly controlled by soluble and membrane-anchored regulators (21). The soluble regulators of the CP and LP include C1 esterase inhibitor (C1-INH) and C4b-binding protein (C4Bp), while the AP is primarily regulated by Factor H (FH) and Factor H-like protein 1 (FHL-1). Vitronectin, clusterin, and, in part, FH-related protein 1 (FHR-1) comprise the regulatory proteins of the terminal activation sequence (21, 24).

Initial investigations showed that LD spirochetes differ substantially in their susceptibility to human serum and finally led to the classification of spirochetes into three main categories, serum-resistant, intermediately serum-resistant/sensitive or partially resistant, and serum-sensitive (25–27). It is worth mentioning that categorizing of spirochetes in these particular groups largely depends on technical parameters, e.g., serum collection and storage, serum and cell concentrations, incubation period, and the method of choice used to determine borrelial survival, making it difficult to compare the data published. Changing the experimental conditions can lead to differences in the phenotypic classification, in particular of intermediately serum-resistant/sensitive strains. Among LD spirochetes, B. burgdorferi s.s., B. afzelii, B. spielmanii, B. bavariensis, and Borrelia japonica are resistant to complement-mediated killing, B. bissettii was classified as intermediately serum-resistant and B. garinii, B. valaisiana, and B. lusitaniae comprise the group of highly susceptible spirochetes (13, 25–35). Furthermore, differences in serum susceptibility have been reported among certain B. valaisiana and B. garinii isolates (29, 36). Strikingly, the serum susceptibility pattern of LD spirochetes almost matches pathogenicity in humans with the exception of B. garinii known to frequently cause LD. The underlying molecular principles of how B. garinii circumvent complement-mediated killing are largely unknown and are still a matter of controversy. It is likely that pathogen-associated factors produced solely in the infected host or that interaction with host or tick-derived proteins upon the transmission process, e.g., plasminogen (37), Tick Salivary Lectin Pathway Inhibitor (TSLPI) (38), or Salp20 (39), may protect B. garinii from complement attack.

More recently, a novel immune evasion mechanism has been described by which B. burgdorferi s.s. specifically blocks CP activation (79). This study depicts BBK32 as the first protein that binds to C1r in a non-covalent manner and thereby preventing autocatalysis of this proenzyme and subsequently the cleavage of C1s, leaving the C1q complex in an inactive enzymatic state (Figure 1B; Table 1). By interfering with C1r, BBK32 acts as a potent inhibitor of the CP without affecting the LP and AP.

Unlike other serum-resistant LD spirochetes, B. bavariensis binds neither FH/FHL-1 nor other complement regulators such as C4Bp or C1-Inhibitor (45). Detailed analysis revealed two proteins, BGA66 and BGA71 as novel complement inhibitors. They belong to the PFam54 protein family and share 51 and 41% sequence identities, respectively, to CspA. Both proteins interact with components of the terminal pathway, in particular C7, C8, and C9, and also with the assembled TCC (45) (Figure 1B; Table 1). Binding of the borrelial proteins to various components of the terminal pathway affects TCC formation by (i) inhibiting C9 auto-polymerization, (ii) terminating TCC assembly, and (iii) preventing integration of the functional pore-forming complex. Moreover, BGA66 and BGA71 are simultaneously produced in all B. bavariensis strains investigated, but each protein by itself displays anti-complement activity and renders transformed spirochetes resistant to complement-mediated killing. Despite the structural similarities to CspA, BGA66 and BGA71 do not bind the potent complement regulators FH and FHL-1 (45). One might speculate that termination of the final activation steps may result in a somewhat weaker complement inactivation capacity. However, this does not appear to be the case as CspA and CspZ-producing and FH/FHL-1-binding spirochetes did not show a different resistance phenotype compared to BGA66- or BGA71-producing cells, indicating that inhibition of the terminal pathway is as efficient as blocking complement activation at the level of C3.

Borrelia burgdorferi s.s. also produces a CD59-like protein, which preferentially bind to C9 and to some extent to the β-subunit of C8 (80) (Figure 1B; Table 1). By using anti-CD59 antibodies, the functional activity of this protein could be blocked, rendering the spirochetes susceptible to complement-mediated killing. Although this surface-exposed protein has never been identified, the binding properties of the CD59-like protein suggest a role in inactivating the terminal pathway of complement. Despite the overlapping complement-inhibitory activities, the CD59-like protein is not identical with BGA66 and BGA71.

Besides the already mentioned molecules, which interacting with complement in multiple ways, additional proteins have been described that also bind human plasmin(ogen), e.g., OspA, OspC, and BBA70 (81–83) (Figure 1C). For the latter, degradation of C3 and, in part, C5 has been demonstrated. Whether the interactions of OspC and BBA70, known to be expressed in the mammalian host, might support complement inactivation in vivo is not known.

Initial studies of the molecular principles of complement resistance, focusing on the amounts of deposited complement components, revealed an amorphous structure of high density that surrounds the entire cell envelope and, apparently, acts as a physical barrier, preventing the insertion of the formed TCC into the bacterial membrane of serum-resistant cells (Figure 1C) (32). Currently, no data are available on the composition and content of this so-called “slime layer” and whether this structure is present in LD spirochetes other than B. afzelii (32). Thus, further studies are required to verify the precise nature of this morphological substance.

Over the last decades, numerous molecules have been identified that interact with the innate immune system in multiple ways to influence or terminate complement at distinct activation levels, e.g., initiation of the CP, C3 activation by the AP, and formation of the TCC. Knowledge of the proteins involved in the interaction with complement has allowed for a better understanding of molecular principles of complement evasion developed by LD spirochetes. Future investigations will undoubtedly identify additional complement-interacting molecules required for evading the innate immune response of different animals, including reservoir hosts, and provide insight into whether these proteins might function in a host-specific manner during the infection process.

PK prepared the figure and table and wrote the manuscript.

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.