HCV is an enveloped, positive-stranded RNA virus belonging to the Hepacivirus genus of the Flaviviridae family, causing in the majority of cases (about 80%) a chronic infection [18]. On the basis of some conserved regions it can be divided in seven major genotypes and numerous subtypes, differently distributed in the world. In single infected patients, it circulates as a group of highly diversified viral variants, called quasispecies [19].

HCV genome is approximately 9,600 base pairs long and encodes a polyprotein precursor of about 3,000 amino acids. It is cleaved by viral and host proteases, resulting in a series of structural (core, E1 and E2) and nonstructural proteins (p7, NS2, NS3, NS4A, NS4B, NS5A and NS5B) [20]. Virions enter into the host cells, in particular hepatocytes, through a complex and finely regulated multistep process. In brief, the viral envelope type I membrane glycoproteins, E1 and E2 (HCV/E1-E2), allow clathrin-mediated virus endocytosis interacting consecutively with several entry cellular cofactors such as glycosaminoglycans [21,22,23], low-density lipoprotein receptor [24,25], scavenger receptor class B type I [26], the tetraspanin CD81 [27], the tight-junction proteins claudin-1 and occludin, and the recently described Niemann-Pick C1-like 1 cholesterol absorption receptor [28,29,30,31,32]. As expected, the envelope glycoproteins, in particular HCV/E2, are the major targets of the humoral anti-HCV response and, therefore, the most hypervariable HCV proteins [33,34,35]. Recently, increasing data have been evidencing a very complex interplay among different regions of this protein and antibodies (Abs) endowed with highly diverging biological activity, suggesting “novel” mechanisms of HCV escape [36,37,38,39].

Every HCV genotype have been found in infection-related MCII, even if different reports describe its higher prevalence among patients infected with HCV of genotype 1 and 2a/c [40,41,42,43,44,45,46]. The reported differences in the prevalence of HCV genotypes in different regions of the world could bias this observation, which should be therefore interpreted with caution.

The mechanisms by which HCV infection leads to RF production, MCII and B-NHL, as well as whether these conditions are related to the lack of some branches of the antiviral immune response are still unknown. The duration of HCV infection required for the development of cryoglobulinemic vasculitis is not well defined but appears to be at least a decade [47]. However, MCII does not display the molecular features of an in situ or occult B-cell lymphoma, as evidences show that the B-cell clonal expansion is not a consequence of a true neoplastic process but is probably the result of a pathogenic dysregulation of the host’s immune system. Cryoglobulins are thus the product of virus–host interactions, whose potential pathogenicity derives from several cofactors [48].

As anticipated, in HCV-induced MCII, cryoprecipitates are usually formed by polyclonal IgGs, frequently directed against the HCV core protein and the nucleic acid of HCV, as well as mono/oligoclonal IgM with RF activity [49,50]. Other constituents include also C1q, C-reactive protein (CRP), other HCV antigens (Ags), and molecules of the lectin complement pathway (MBL and MBL-associated serine protease-1), with the latter mostly associated with membranoproliferative glomerulonephritis [51].

Importantly, cryoprecipitation was directly correlated with anticore IgG concentration in the cryoprecipitate, thus inferring that its production is dependent on their selective binding to the Ag in the presence of IgM molecules with RF activity. Indeed, the concentration of HCV RNA in the cryoprecipitate was found to be 10 to 1,000-fold greater than in the supernatant [52,53]. This evidence has suggested a direct role of the HCV core protein in the cryoprecipitation phenomenon [49]. In fact, IgM RF acts as an incomplete cryoglobulin, precipitating at low temperature, probably following a conformational change induced by their binding to IgG with anticore reactivity. In particular, the core is supposed to be the most involved viral protein in cryocrit formation, as demonstrated in the skin and renal tissues of HCV-infected patients with MCII-associated active vasculitis and nephropathy, respectively [54]. In fact, nonenveloped core protein is overproduced during the viral life cycle, and in MCII patients, its plasmatic levels have been associated to cryoglobulinemia-associated symptoms [54]. Moreover, both IgG and IgM may be recognized by the globular heads of C1q interacting with their CH2 and CH3 or CH4 domains, respectively, and for this reason identified as a constituent of cryoprecipitates in some studies. In particular, IgM molecules are good acceptors of C1q and indeed can favor indirect binding of HCV core protein to endothelial cell surface [55,56].

Finally, HCV core protein has also been shown to promote immortalization in different cell lines, as well as being capable of blocking c-myc induced apoptosis and indeed could have a direct role in the pathogenesis of HCV-related lymphomas [57]. At this regard, focusing on animal models, core transgenic mice developed lymphoma with a high frequency (80%) at ages over 20 months [58].

The expression of proteins structurally similar to host defense proteins and immunomodulators is an important immune evasion strategy leading to persistence, a mechanism already described for several viruses [59,60]. Moreover, in the case of HCV, this evasion mechanism has been considered also as a possible factor involved in the development of MCII. In particular, some motifs on HCV/E2 glycoprotein have been suggested to be involved in a molecular mimicry of specific Ig portions. More in details, the N-terminal hypervariable region (HVR1) of HCV/E2 shares some conserved motifs with selected human Ig variable (IgV) domains, as well as with the T-cell receptor (TCR) α- and β-chains [61]. This observation, together with the fact that HVR1 acts as an immunodominant “decoy” region diverting the humoral response, would make the frequently elicited anti-HVR1 Abs potentially capable of cross-reacting with other Abs or with TCR [62]. In fact, a lower rate of HVR1 mutations in HCV-positive patients with MCII and presenting a monoclonal IgM expansion with RF activity has been observed [61]. This lower variability in the main target of anti-HCV humoral response was interpreted as a clear sign of impaired immune response, as already observed in agammaglobulinemic or in otherwise immunosuppressed patients [63,64]. Moreover, this low variability at the level of classically hypervariable regions on envelope glycoproteins has been observed also in the case of other viruses, like influenza viruses, suggesting a complex and well-regulated equilibrium between host immune response and viral evasion through variability [65]. Interestingly, a direct correlation has been observed between the degree of similarity of HVR1 to Ig and TCR molecules and the degree of immune escape and persistence in humans and experimentally infected chimpanzees [61]. This indicates that variation in HVR1 sequence is not only correlated with escape from neutralizing Abs, but also to an increased similarity to Ig, suggesting an additional immune evasion strategy through mimicry [66]. As a consequence, this mimicry could determine a chronic stimulation induced also by self-Ags, thus leading to the HCV-related lymphoproliferative disorders.

In addition, a similar mimicry of Ig motifs, not based on identical linear sequences, has been observed between the 1238-1334 amino acid region of HCV/NS3 (in particular in the 1238-1279 and 1251-1270 residues), the viral protease, and the CH3 domain on the Fc portion of human IgG (amino acid residues 345-355) [67]. This region is conserved among all IgG classes except for a single mutation in IgG4. Interestingly, the crystal-structure analysis of a complex between IgM RF and its IgG auto-Ag revealed relatively few Ab contact residues between the potential combining sites of IgM and IgG [68]. In particular, these residues are located on only one side of the combining site surface, indicating that IgM has another, entirely different, specificity than that featured by its RF activity. Autoreactive IgM may thus have originated in response to another Ag, such as HCV/NS3, and the reactivity with IgG Fc may be an unfortunate coincident cross-reactivity phenomenon.

At this regard, certain VH subfamily genes, like the VH4-34, have been described as endowed of binding activity outside the CDRs as clearly evidenced in the case of cold agglutinin-related disease [69,70]. This feature is characteristic of B-cell super-Ags that are supposed to directly activate B cells. As explained more in details below, this could be the case also for the HCV/E2 glycoprotein, due to its ability to stimulate the expansion of a restricted set of VH and VL subfamily expressing B cells [17,71,72]. A similar behavior has also been described for staphylococcal enterotoxins A and D, that function as human B-cell super-Ags rescuing B cell-expressing VH3 and VH4 (including VH4-34) genes inducing cell survival in in vitro experiments, and has also been suggested for HIV gp120 [73]. Moreover, certain portions of the FRs seem to be important for super-Ag binding, and thus determinant in the case of a super-Ag-driven selective pressure [70].

Finally, considering the HCV core protein participation in the formation of immune complexes and suppression of T-cell response by interacting with the globular domain of C1q complement receptor (gC1qR), it has been suggested that this interaction may play a key role in determining complement activation, a critical interdependent regulator of the size and solubility of immune aggregates [55,56,74]. In particular, the structural similarity between HCV core Ag and C1q may explain the presence of cross-reactive anti-C1q in HCV-associated MCII [56].

Early after its discovery, it was shown that HCV is also a lymphotropic virus, thus suggesting its direct role in lymphoproliferative disorders [75,76]. Interestingly, a lymphotropic isolate of HCV (SB strain) has been obtained from an infected B-cell lymphoma that produces IgM displaying reactivity against the HCV/NS3 protein [77]. Moreover, several studies have described the presence of negative-stranded HCV RNA in B lymphocytes, but these observations were not confirmed in other works [78,79].

HCV/E2 Ag binding to the ubiquitously expressed CD81 receptor on B-cells, as part of the CD81/CD19/CD21 complex, and concomitantly to close anti-HCV/E2 B cell receptors (BCR), could induce a strong B-cell proliferation signal. [80,81]. The binding of HCV/E2 to CD81 induces double-strand DNA breaks and hypermutation of VH (that has been observed also using anti-CD81 Abs) that lower Ab affinity and Ab-mediated complement-dependent cytotoxicity and consequently Ab-mediated neutralization, suggesting a novel escape mechanism of HCV [82]. This process was demonstrated to be dependent on activation-induced cytidine deaminase and related to an increase in the production of TNF-α [83,84]. Moreover, an Ig cloned from B-NHL biopsy specimens have been found to bind HCV/E2 [85].

Indeed, it has been postulated that also anti-HCV IgG complexes with the virus and HCV lipoproteins (VLDL) complexes may act as another B-cell super-Ags inducing the synthesis of non-HCV reactive IgM with RF activity [86]. Then, these auto-Abs, in turn, form immune complexes, which circulate throughout the body and are deposited in small to medium blood vessels, resulting in complement activation and extrahepatic injury.

Several studies demonstrated the presence, in patients with HCV-related MCII, of IgV gene mutations compatible with a germinal center (GC) or post-GC derivation, a replacement/silent (R/S) mutation ratio consistent with the maintenance of a functional structure of the BCR, and the presence of intraclonal heterogeneity. Conceivably, the similarity in the structure of the variable BCR region and the restricted recruitment of certain IgV gene subfamilies, both for heavy and light chains, may account for selection of B cells expressing specific and similar reactivity. This suggests the possible role of a common Ag, possibly endowed of super-Ag-like features [87,88,89,90,91].

In particular, the WA cross-idiotype Abs, frequently encountered among IgMκ type II mixed cryoglobulins, possess heavy chains belonging to discrete VH subfamily genes, such as VH1-69 and VH3-7 paired with specific Vκ products, such as Vκ3-20 and Vκ3-15, respectively [92]. In fact, it has been observed a major involvement of Vκ expressing B cells, as demonstrated by highly skewed κ/λ ratios and as corroborated by the Vκ usage belonging to these restricted subfamily gene segments. In particular, Vκ3 genes utilize a Jκ1 joining segment for expression, while VH1 genes preferentially use a D3-22 diversity gene segments, with VH1-69 rearranging to JH4 and D region consensus 1, and VH3-7 rearranging to JH3 or JH4 gene segments and D region consensus 2 [93]. Other restricted Ig VH subfamilies have been described in HCV infection and HCV-related lymphoproliferative disorders. Among them the mostly described are the previously mentioned VH4-34 as well as the VH4-59, VH3-7, VH3-21, VH3-23, VH3-30 and VH3-48 [94,95].

VH1-69 is certainly the most recurrent and described Ig VH subfamily induced during anti-HCV infection [96,97,98,99], as well as against other viral pathogens determining acute infections, like influenza viruses [100,101,102,103,104,105,106], and chronic infections, like HIV [107,108]. This VH subfamily gene commonly encodes polyreactive natural Abs and, in HCV-associated cryoglobulinemic patients, it frequently undergoes somatic mutation, probably during affinity maturation. Again, this observation further suggests that HCV-associated MCII and lymphomas may originate in B cells responding to a common Ag. In particular, the VH1-69 gene is highly represented in the anti-HCV/E2 humoral response [71,109]. Preferential use of this gene has also been seen in 10%–20% of patients with CD5+ B cell chronic lymphocytic leukaemia [45,110]. In addition, biased use of VH1-69 has been demonstrated for salivary gland mucosa-associated lymphoid tissue (MALT) lymphomas in which 61% of the patients express this gene [111].

Moreover, a significant homology among CDR3 belonging to VH1-69 expanded memory B cells of a patient with HCV-associated MCII and the correspondent region amplified from a monoclonal nonneoplastic B-cell expansion of a Sjögren’s syndrome patient has been described [111]. This suggests that the stimulatory agents underlying these disorders may share common antigenic determinants. Moreover, in the same report, in all the analyzed patients with HCV-associated MCII, 18%–98% of circulating B cells express the VH1-69 gene and, in one-third of these patients, these cells coexpress the Vκ3-20 gene, constituting the previously mentioned WA cross-idiotype [47,71,110,112]. In fact, it has been described that in HCV-infected patients, although the absolute number of circulating B cells was within the normal limits, in some cases, they were almost completely represented by VH1-69 monoclonal B cells and, in patients that had in addition to HCV-associated MCII a splenic lymphoma, a leukemia-like monoclonal expansion of VH1-69 B cells was present [111,113]. This evidence indicates that a clonal population of VH1-69-expressing B cells progressively invades the circulating B cell repertoire of patients with HCV-associated MCII. Moreover, some of these clones have CDR3 sequences identical to RF IgMs isolated from patients with MALT neoplasms [114]. Thus, these nonneoplastic B cells appear to evade the homeostatic mechanisms that regulate the Ag-driven clonal expansion and genetic events may cause further escape from control leading to an absolute lymphocytosis.

In this regard, our group previously reported that the restricted VH1-69 gene usage could be responsible for the B-cell expansion of restricted clones that share the capability of reacting against Igs belonging to this VH subfamily [94]. In particular, it has been observed that the immune repertoire of a patient with HCV-associated MCII contains IgM clones able to react specifically against anti-HCV/E2 Abs belonging to VH1-69 subfamily and derived from the same patient [115]. Indeed, we found that 61% of IgM reactive to anti-HCV/E2 VH1-69 Fab fragments belonged only to two VH subfamilies, VH3-23 (39%) and VH3-21 (22%), that are frequently described in autoimmune disorders [116]. Furthermore, the mutational pattern of selected anti-HCV/E2 IgM showed that almost all clones had a natural origin, evidenced by the high homology to the germline counterpart. Finally, more in details, we found that the VH3-23 subfamily showed a preferential binding of the VH1-69-derived IgG1 Fabs. These data suggest that VH3-23 IgM may be naturally prone to recognize some conserved regions of specific VH subfamilies, as VH1-69, the VH gene described to be elicited in the humoral response against HCV/E2 [35,96,98,99,117]. Considering these data, the HCV/E2-driven stimulation of the immune system may cause the expansion of specific B cells expressing Abs encoded by the VH1-69 subfamily gene and recognized by some natural IgM-encoded subfamily Abs. This could lead to the formation of circulating immune complexes and the cross-linking of BCR by auto-Abs that may allow a chronic activation and a clonal expansion of anti-HCV/E2 B cells. Indeed, a widely held hypothesis is that with increased duration of HCV infection, monoclonal IgM RFs arise from the population of polyclonal IgM RFs that are present before the development of type III cryoglobulins, consisting of polyclonal IgGs and polyclonal IgM RFs, a condition that is believed to precede MCII, characterized by a monoclonal expansion of IgM RF and hypothesized to derive from the population of polyclonal IgM RFs in type III cryoglobulins. However, in this regard, there are conflicting studies [47].

Finally, to reconcile the fact that expansion of B-cell clones carrying the VH1-69 subfamily BCR characterizes the HCV humoral response, as well as that against other viral pathogens, and patient-derived monoclonal IgGs belonging to this subfamily are frequently endowed of a neutralizing activity against their respective pathogens, while IgMs belonging to this subfamily frequently harbor RF characteristics, recent evidences suggest that somatic hypermutation, as well as class switching, abrogate germline BCR reactivity, revealing additional or altered antigenic specificities [118,119,120,121].

In this regard, a pauciclonality of the peripheral memory B-cell population has been described as a unique feature of spontaneous resolving acute HCV-infected patients compared to chronically evolving patients. This finding, considered characteristic only of patients with HCV-associated lymphoproliferative disorders, suggests that the B-cell clones potentially involved in clearance of the virus may also be subjected to abnormal proliferation [122]. However, it is widely accepted that the intrinsic genetic instability of B cells during somatic hypermutation and class-switching processes, may favor genetic aberrations responsible for prolonged B-cell survival and proliferation, thus allowing them to escape from the homeostatic balance controlling clonal expansion. In particular, among chromosomal aberrations, loss of chromosome 2q, gain of the long arm of chromosome 3, 7q deletion and gains of 1q and 8q, have been described [123,124,125]. All these events could lead the lymphoproliferation to become independent of antigenic stimulation, exposing the patients to the risk of developing a frank B-cell malignancy.

Analogously to VH1-69, the VH4-34 subfamily usage has been observed in several autoimmune and lymphorpoliferative disorders, such as diffuse large-cell lymphoma, primary central nervous system lymphoma, B-chronic lymphocytic leukemia, and autoimmune disorders [115]. Moreover, it has also been implicated in HCV response and HCV-associated MCII and lymphomas. Additionally, it is well known that, independently from the DH and JH gene segments, as well as light chains from different subfamilies and isotype associated, the VH4-34 is a naturally autoreactive subfamily [70]. In fact, the VH4-34 gene is found in virtually all cases of cold agglutinin disease in which the red blood cell I/I Ags bind to the FR1 domain of Ig, with a minor involvement of the CDR3 region [69]. Therefore, the high frequency of the VH4-34 gene usage and the intrinsic molecular features of its FR and CDR domains, suggest a possible role of yet unknown B-cell super-Ag in driving HCV-related lymphoproliferative disorders [79].

Low levels of complement components, such as C1, C4 and C2, have been reported in cryoglobulinemic patients, suggesting ongoing complement activation and consumption both via the classical pathway as well as via the MBL pathway. In particular, the C4 level provides a “signature” which may be used to anticipate the presence of significant (> 1 mg/mL) amounts of type II cryoglobulins in the blood [126].

Moreover, in MCII the IgM and IgG deposits are frequently accompanied by the glomerular deposition of MBL, C4, C3 and C1q as a consequence of complement activation at the level of cryoprecipitates, consisting in immune complexes of Ig, Ags of HCV particles and other serum components, such as CRP and complement components [127]. Moreover, possibly present IgMs with RF activity could cover the Fc fragments of IgGs, thus altering their interference with Fc receptors as well as complement activation that is necessary for immune precipitates solubilization with a consequent augmented risk of inflammatory tissue damage. Furthermore, this phenomena could be accentuated with temperature drops or saturation and accumulation of IgM RF-IgG complexes [51].