To examine the expression profiles of a large number of genes within typical vasculitic nerve lesions in patients with mixed cryoglobulinemia (MC) vasculitis in order to better characterize the molecules involved in cellular tissue activation and trafficking.
The quantitative expression of 19 genes coding for cytokines, chemokines, and their receptors in the nerve lesions of 9 patients with hepatitis C virus (HCV)–associated MC vasculitis, 7 with idiopathic polyarteritis nodosa (PAN) (rheumatic disease controls), and 8 patients with noninflammatory idiopathic neuropathy (noninflammatory neuropathy controls) was assessed using a real-time reverse transcriptase–polymerase chain reaction procedure.
Compared with the noninflammatory controls, HCV-MC vasculitis patients had a significantly higher expression of Th1 cytokines in vasculitic nerve lesions (mean ± SEM fold increase 33.7 ± 11.6 for interferon-γ and 7.2 ± 1.9 for tumor necrosis factor α), whereas Th2 cytokines were absent (interleukin-4 [IL-4], IL-5, and IL-13) or were not significantly different (IL-10). Chemokines involved in T cell and monocyte trafficking were also significantly up-regulated in the HCV-MC vasculitis patients (mean ± SEM fold increase 27.4 ± 8.3 for macrophage inflammatory protein 1α [MIP-1α], 19.9 ± 5.7 for MIP-1β, and 7.2 ± 1.5 for CXCR3). Compared with patients with idiopathic PAN, there was a trend toward higher expression of MIP-1α and CXCR3 in HCV-MC vasculitis patients (mean ± SEM fold increase 27.4 ± 8.3 versus 5.3 ± 3.4 for MIP-1α and 7.2 ± 1.5 versus 2.5 ± 0.9 for CXCR3).
This study is the first to demonstrate a role of cellular immunity and Th1 lymphocytes in the pathogenesis of HCV-MC vasculitic nerve lesions.
Type II mixed cryoglobulinemia (MC) is a systemic vasculitis, usually associated with hepatitis C virus (HCV), that mainly involves the skin, kidneys, and nervous system (1). MC is characterized by the proliferation of B cell clones producing pathogenic IgM with rheumatoid factor activity. The neurologic complications in HCV-infected patients occur predominantly in the peripheral nervous system and are mainly associated with MC (2). The reported prevalence of peripheral nervous system involvement in patients with HCV infection varies and can be as high as 50% of cases (3, 4). The most frequently described form is a distal sensory or sensorimotor polyneuropathy (3, 5).
The pathophysiology of HCV-MC vasculitis neuropathy is still a subject of controversy. Neutrophil infiltration with leukocytoclastic changes typical of immune complex–mediated vasculitis has seldom been found, while the presence of lymphohistiocytic infiltrates suggests a T cell–mediated pathogenesis and only a minor role for humoral mechanisms in the formation of the vasculitic lesions (5–7). We recently obtained evidence of an important role of vascular cell adhesion molecule 1, a molecule that is exclusively involved in mononuclear cell recruitment, in the pathogenesis of severe forms of HCV-MC vasculitis (8). Reported data indicate that peripheral blood monocytes and liver T cells from patients with HCV-related MC vasculitis are characterized by a strong Th1 response (9, 10). These results support the hypothesis that an HCV-driven cellular immune response plays an important role in the pathogenesis of HCV-MC vasculitis. However, there is as yet no available information regarding the cytokine profile in target organs of patients with MC vasculitis.
Cytokines play a major role in the immune response and inflammatory dysfunction. Classically, CD4+ T lymphocytes may differentiate into 2 functional subsets: Th1 cells and Th2 cells. Th1 cells secrete cytokines, such as interferon-γ (IFNγ) and interleukin-2 (IL-2), which promote cell-mediated immunity. In contrast, Th2 cells produce cytokines such as IL-4, IL-5, IL-10, and IL-13, which are involved in antibody-mediated immunity. Recent findings suggest that chemokines and their receptors play an important role in the pathophysiology of autoimmune diseases (11). Inflammatory chemokines are expressed in inflamed tissues by resident and infiltrating cells following stimulation by proinflammatory cytokines. These inflammatory chemokines play a pivotal role in lymphocyte trafficking through the expression and modulation of T cell chemokine receptors (12). Chemokines have also been shown to be involved in the effector and amplification mechanisms of polarized Th1-mediated and Th2-mediated immune responses (13).
To gain further insight into the pathogenesis of cryoglobulinemia vasculitis neuropathy, we used a quantitative real-time reverse transcriptase–polymerase chain reaction (RT-PCR) procedure to study the expression profiles of a large number of genes within typical nerve lesions from patients with MC vasculitis and to compare them with the expression profiles in nerves from patients with idiopathic polyarteritis nodosa (PAN) and noninflammatory idiopathic neuropathy. We also assessed the level of expression of 19 cytokine and chemokine genes that have been shown to be involved in cellular activation or trafficking associated with autoimmune diseases (14).
PATIENTS AND METHODS
The characteristics of the patients with HCV-MC vasculitis and idiopathic PAN are summarized in Table 1. The study population consisted of 9 patients with HCV-MC vasculitis (mean age 58.1 years, age range 33–73 years), 7 patients with idiopathic PAN (mean age 61 years, age range 45–76 years) (rheumatic disease controls), and 8 patients with noninflammatory idiopathic axonopathy (mean age 58.8 years, range 52–68 years) (noninflammatory neuropathy controls). Informed consent was obtained from each patient, and the study conformed to the ethical guidelines of the Declaration of Helsinki.
Table 1. Main features of patients with hepatitis C virus (HCV)–associated mixed cryoglobulinemia (MC) vasculitis and polyarteritis nodosa (PAN)
The patients with HCV-MC vasculitis had a serum level of mixed cryoglobulins >0.05 gm/liter on at least 2 occasions and were positive for serum HCV RNA. Control patients with idiopathic PAN or noninflammatory neuropathy were negative for HCV infection. All patients and controls were negative for the human immunodeficiency and hepatitis B viruses. Patients with HCV-MC vasculitis and idiopathic PAN were untreated and had clinically active disease. The type of vasculitis was defined according to the Chapel Hill Consensus Conference criteria (15).
Biopsy specimens from all patients showed severe axonal degeneration and an inflammatory process involving the nerves. Necrotizing vasculitis was present in the 7 patients with idiopathic PAN and in 5 of the 9 patients with HCV-MC vasculitis (Table 1). Lymphocytes clearly predominated within the cellular infiltrates in the nerves from HCV-MC vasculitis patients, whereas the cellular infiltrate in nerve lesions from the idiopathic PAN patients (except for patient 7) was equally distributed between lymphocytes and polymorphonuclear leukocytes (mainly neutrophils and eosinophils).
All patients and controls underwent full-thickness open biopsy of the superficial peroneal nerve and peroneus brevis muscle in the most affected limb. Specimens were immediately snap-frozen in liquid nitrogen and stored at −80°C. Part of the specimen was paraffin-embedded, and both transverse and longitudinal sections were stained with either hematoxylin and eosin, periodic acid–Schiff, or Congo red. In each case, the presence of inflammatory vascular lesions was evaluated in both muscle and nerve samples by examination of serial sections of paraffin-embedded specimens.
For immunohistochemistry studies, consecutive serial sections of inflammatory vascular lesions were analyzed by the same pathologist (TM). The degree of staining was graded as follows: mild = labeling of <10 cells per infiltrate, moderate = labeling of 10–30 cells per infiltrate, and marked = labeling of >30 cells per infiltrate. Paraffin-embedded serial sections of nerve were labeled using anti-CD3 (F7.2.38), anti-CD20 (L26), anti-CD4 (MT310), and anti-CD8 (08/144B) monoclonal antibodies (all from DakoCytomation, Trappes, France) or an anti-IgG antibody as a negative control. Labeling was revealed by the immunoperoxidase method.
Quantification of cytokine messenger RNA (mRNA).
The theoretical and practical aspects of real-time quantitative RT-PCR using the ABI Prism 7700 Sequence Detection system (PE Applied Biosystems, St. Quentin-en-Yvelines, France) have been described in detail elsewhere (16). Briefly, total RNA was reverse-transcribed before real-time PCR amplification was performed. Quantitative values are obtained from the threshold cycle (Ct) number, the point at which the increase in the signal associated with exponential growth of PCR products begins to be detected by the PE Applied Biosystems analysis software, which was used according to the manufacturer's instructions. The precise amount of total RNA added to each reaction mixture (based on optical density) and its quality (i.e., lack of extensive degradation) are both difficult to assess. We therefore also quantified transcripts of the TBP gene, which encodes the TATA box binding protein (a component of the DNA binding protein complex transcription factor IID), as the endogenous RNA control, and each sample was normalized on the basis of its TBP content.
Results, which were expressed as the N-fold difference in target gene expression relative to the TBP gene (termed Ntarget), were determined by the following formula: Ntarget = 2. The ΔCt value of the sample was determined by subtracting the Ct value of the target gene from the Ct value of the TBP gene. The Ntarget values of the samples were subsequently normalized as the median value in samples from the controls with noninflammatory idiopathic neuropathies; the Ntarget values were 1. Genes were considered to be markedly overexpressed when they were ≥5 times the value in the noninflammatory neuropathy controls.
Primers for TBP and the 19 target genes were chosen with the assistance of the Oligo 5.0 computer program (National Biosciences, Plymouth, MN). The nucleotide sequences of the primers we used are shown in Table 2. To avoid amplification of contaminating genomic DNA, 1 of the 2 primers was placed at the junction between 2 exons. The thermal cycling conditions consisted of an initial denaturation step at 95°C for 10 minutes and 50 cycles at 95°C for 15 seconds and 65°C for 1 minute.
Statistical analyses were performed using StatView software (SAS Institute, Cary, NC). Data are expressed as the mean ± SEM. Comparison of the gene expression level between different groups was performed with the nonparametric Mann-Whitney U test. Values were considered significant only after adjustment for multiple testing. We used Hochberg's stepwise adjusted P values for strong control of Type I error (17). The level of significance was set at 0.05 for all statistical tests.
Findings of immunohistochemical analyses.
Immunohistochemical analysis of vasculitic nerve tissue was used to investigate the pattern of expression of CD3 (T cells), CD20 (B cells), CD4+, and CD8+ T lymphocytes in inflammatory perivascular infiltrates from 5 patients with HCV-MC vasculitis (patients 1, 3, 5, 6, and 8) (Table 1). No signal was detected using the IgG control (results not shown). Marked expression of CD3 (Figure 1A), which was especially prominent in the inflammatory mononuclear cell infiltrate, was observed in all HCV-MC vasculitis patients, whereas CD20 (Figure 1B) showed moderate expression. As shown in Figures 1C and D, the T lymphocyte infiltrate was equally distributed between CD4+ and CD8+ cells.
Intralesional cytokine expression.
Local quantification of IFNγ, IL-2, IL-4, IL-5, IL-6, IL-10, IL-13, and tumor necrosis factor α (TNFα) mRNA was performed using quantitative real-time PCR of nerve biopsy samples from patients with HCV-MC vasculitis, idiopathic PAN (rheumatic disease controls), and noninflammatory idiopathic neuropathies (noninflammatory neuropathy controls) (Table 3 and Figure 2).
Table 3. Gene expression profile in the nerves of patients with HCV-MC vasculitis and idiopathic PAN*
HCV-MC vasculitis (n = 9)
Idiopathic PAN (n = 7)
Values are the mean ± SEM fold difference compared with nerves from controls with noninflammatory neuropathy, after normalization of each measurement. HCV-MC = hepatitis C virus–associated mixed cryoglobulinemia; PAN = polyarteritis nodosa; IFNγ = interferon-γ; TNFα = tumor necrosis factor α; IL-10 = interleukin-10; MCP-1 = macrophage chemoattractant protein 1; MIP-1α = macrophage inflammatory protein 1α.
P < 0.05 versus noninflammatory neuropathy controls.
Significant differential expression, as high as 5 times the value in noninflammatory neuropathy controls, was observed for 2 genes in the nerves of HCV-MC vasculitis patients, IFNγ and TNFα (mean ± SEM fold increase 33.7 ± 11.6 versus 1 ± 0.1 in controls [P = 0.01] and 7.2 ± 1.9 versus 1 ± 0.4 in controls [P = 0.03], respectively), whereas IL-2 was not detected. IL-6 and IL-10 were not significantly different in the HCV-MC vasculitis patients compared with the noninflammatory neuropathy controls (mean ± SEM fold increase 2.1 ± 0.5 versus 1 ± 0.3 [P = 0.6] and 2.7 ± 0.6 versus 1 ± 0.9 [P = 0.5], respectively). IL-4, IL-5, and IL-13 were undetectable in all of the nerve biopsy specimens from patients with HCV-MC vasculitis, idiopathic PAN, and noninflammatory neuropathy.
Compared with patients with idiopathic PAN, cytokine expression was not significantly different in patients with HCV-MC vasculitis (Table 3). However, there was a trend toward higher expression of IFNγ and TNFα in MC vasculitis (mean ± SEM fold increase 33.7 ± 11.6 versus 27.9 ± 11.4 and 7.2 ± 1.9 versus 2.6 ± 0.7, respectively).
Expression of chemokines and their receptors in nerve biopsy specimens.
For a better analysis of cellular involvement in the pathogenesis of HCV-MC vasculitis neuropathy, we analyzed a set of chemokines and their receptors (Table 3). Although not statistically different, both HCV-MC vasculitis and idiopathic PAN patients had a marked overexpression of IL-8 compared with the noninflammatory neuropathy controls (P = 0.06 and P = 0.06, respectively) (Table 3). The expression of mRNA for CXCR1, CXCR2, and CXCL6, which are involved in the migration of neutrophils, was not significantly different in HCV-MC vasculitis patients as compared with noninflammatory neuropathy controls (mean ± SEM fold increase 1.4 ± 0.8 versus 1 ± 0.1 [P = 0.4], 1.4 ± 0.8 versus 1 ± 0.3 [P = 0.4], and 4.3 ± 1 versus 1 ± 0.9 [P = 0.1], respectively) (Table 3).
Compared with patients with idiopathic PAN, there was a trend toward lower expression of CXCL6 and IL-8 in the nerves of HCV-MC vasculitis patients (Figure 2).
In contrast, all of the following chemokines act as chemoattractants for T cell and monocytes: macrophage chemoattractant protein 1 (MCP-1), macrophage inflammatory protein 1α (MIP-1α), MIP-1β, RANTES (CCL5), CCR5 (chemokine receptor for MIP-1α, MIP-1β, and CCL5), and IFNγ-inducible protein 10 (CXCL10) and its receptor CXCR3. Except for CCL5 (mean ± SEM fold increase 4.9 ± 1.3), all of these chemokines were markedly overexpressed (≥5 times the control value) in HCV-MC vasculitis patients as compared with the noninflammatory neuropathy controls. However, significantly different expression was observed only for MIP-1α, MIP-1β, and CXCR3 (mean ± SEM fold increase 27.4 ± 8.3 versus 1 ± 0.3 [P = 0.01], 19.9 ± 5.7 versus 1 ± 0.4 [P = 0.01], and 7.2 ± 1.5 versus 1 ± 0.5 [P = 0.03], respectively).
Compared with patients with idiopathic PAN, there was a trend toward higher mRNA expression of MIP-1α and CXCR3 in patients with HCV-MC vasculitis (27.4 ± 8.3 versus 5.3 ± 3.4 [P = 0.08] and 7.2 ± 1.5 versus 2.5 ± 0.9 [P = 0.06], respectively) (Figure 2).
It is well known that there is a pathologic distinction in HCV-associated MC neuropathy, ranging from perivascular infiltrate to necrotizing arteritis. To address this issue, we analyzed the pattern of cytokine and chemokine expression according to the clinical features of the HCV-MC vasculitis patients. HCV-MC patients with necrotizing vasculitis had markedly overexpressed IL-8 as compared with those with only perivascular infiltration (mean ± SEM fold increase 13.2 ± 4.4 versus 3.1 ± 1). The pattern of cytokine and chemokine expression in the subgroups of HCV-MC vasculitis patients with predominantly lymphocytic (n = 5) or polymorphonuclear (PMN) (n = 4) infiltration is shown in Figure 3. TNFα and IL-8 were clearly overexpressed in the PMN group as compared with the lymphocytic group (mean ± SEM fold increase 12.8 ± 4.2 versus 2.8 ± 0.9 and 14 ± 4.6 versus 3.3 ± 1, respectively).
The pathogenesis of HCV-MC vasculitis is complex and is likely to involve many mechanisms. The Arthus phenomenon and its equivalents are considered good experimental models for immune complex vasculitis and MC. They are characterized histologically by the association of tissue edema, hemorrhage, and neutrophil infiltration (18). HCV-MC vasculitis appears to be, at least in part, histologically different from the Arthus model (19).
In order to better characterize the pathogenesis of HCV-MC vasculitis neuropathy, we studied the involvement of cytokines, chemokines, and their receptors in cryoglobulin vasculitic nerve lesions. Our findings provide strong evidence for a role of the cellular immune response in the pathogenesis of HCV-MC vasculitis neuropathy. T cells predominated within vasculitic infiltrates. Both CD4+ and CD8+ T cells accumulated in vasculitic tissues. The increased expression of IFNγ and TNFα in the vasculitic nerves, associated with the absence of typical Th2 cytokines (IL-4, IL-5, and IL-13), points to a strong polarized type 1 immune response. The up-regulation of MIP-1α, MIP-1β, and CXCL10 and its specific receptor CXCR3, all of which influence type 1 T cell differentiation, is also consistent with these findings.
One possible explanation for the absence of IL-2 is that it is secreted in the early stage of the immune response. Indeed, IL-2 is crucial in initiating the adaptive immune response and determines whether a naive T cell will proliferate. This cytokine is also very short-lived because of its unstable sequence, which prevents sustained cytokine production. It is likely that infiltrating T cells within the vasculitic lesions of HCV-MC patients are armed effector cells that do not need IL-2 costimulation.
IL-10 expression within the vasculitic nerve was quite similar to that observed in nerves from patients with idiopathic PAN and noninflammatory idiopathic neuropathy. The predominant type 1 polarization and the absence of Th2 cytokines suggest that the origin of IL-10 is likely not Th2. IL-10 is probably produced by macrophages, epithelial cells, or regulatory T cells and may play a role in preventing a more severe inflammatory response.
The role of chemokines in cell recruitment is 2-fold. First, they act on the leukocyte as it rolls along the endothelial cells at sites of inflammation. Second, the chemokines direct the migration of the leukocyte along a gradient of the chemokine that increases in concentration toward the site of inflammation (12). The high levels of MCP-1, MIP-1α, MIP-1β, CCR5, CXCL10, and CXCR3 mRNA, which are the chemokines mainly responsible for T cell and monocyte trafficking, may thus reflect the important in situ recruitment of both T cells and monocytes in the nerves of HCV-MC vasculitis patients. Consistent with the different subsets of cell infiltrates in the vasculitic nerves of patients with idiopathic PAN, there was a trend toward higher MIP-1α and CXCR3 mRNA expression in patients with HCV-MC vasculitis. There was also a higher expression of IL-8 in the nerves of idiopathic PAN patients as compared with HCV-MC vasculitis patients. However, both groups of patients had a marked IL-8 overexpression as compared with the noninflammatory neuropathic controls. This inflammatory chemokine, which is mainly involved in neutrophil recruitment, was clearly overexpressed in the subgroup of HCV-MC patients with necrotizing vasculitis and with the presence of polymorphonuclear cells in vasculitic infiltrates. The clinical distinction of HCV-MC patients with and without necrotizing arteritis may also be reflected at the cytokine level.
T cell involvement in the pathogenesis of HCV-MC vasculitis has been observed in previous studies (20–23). Patients with HCV-MC vasculitis often have a disturbed peripheral blood T cell repertoire, with a high frequency of T cell expansions (21). CD4+,CD25+ regulatory T cells, which have been shown to control autoimmunity, are significantly reduced in HCV-MC vasculitis patients (22). Further evidence that T cells aid in the production of cryoglobulins is the influence of HLA type II polymorphism on the production of HCV-related cryoglobulins (23). The present study is the first to report the observation of an increased production of Th1 cytokines in the vasculitic nerve lesions of HCV-MC patients. This is consistent with previous data obtained from peripheral blood mononuclear cells (9) and liver (10), ruling out the possibility of a discrepancy between the response of peripheral, liver, and nerve T cells that could occur because of compartmentalization of T cells. These data are also concordant with the observation that the major IgG isotypes in HCV-associated type II and type III MC are mainly IgG1 and IgG3 subclasses (24, 25). TNFα has been shown to drive the differentiation of dendritic cell subsets that will present the Fc fragment of immunoglobulins, eventually leading to the generation of rheumatoid factor (anti-Ig Fc response) (26), which is typical of MC vasculitis.
In conclusion, our study is the first to demonstrate that cryoglobulin nerve lesions may be related to cell-mediated immunity, since a high increase in the expression of Th1 cytokines (IFNγ and TNFα) and chemokines (MIP-1α, MIP-1β, CXCL10, and CXCR3) were found in the vasculitic nerves of HCV-MC patients. The up-regulation of these inflammatory cytokines and chemokines may lead to the recruitment, activation, and differentiation of effector monocytes and T cells at the site of inflammation. These findings are consistent with morphologic studies of HCV-MC–related peripheral neuropathy showing inflammatory vascular lesions with preferential lymphocytic infiltrates and a predominance of T cells. Taken together, these data support the hypothesis that an HCV-driven cellular immune response plays an important role in the pathogenesis of HCV-MC vasculitis neuropathy.
We would like to thank Mathieu Resche Rigon and the staff of the Centre d'Investigation Clinique, Hôpital Saint-Louis (Paris, France), for their assistance in the statistical analysis.