Abstract

Inflammatory cell recruitment is an important step in the pathogenesis of autoimmune demyelinating diseases of the PNS. Chemokines might play a critical role in promoting leucocyte entry into the nervous system during immune‐mediated inflammation. Here, we report the expression pattern of the chemokine receptors CCR‐1, CCR‐2, CCR‐4, CCR‐5 and CXCR‐3 in sural nerve biopsies obtained from patients with classical Guillain–Barré syndrome (acute inflammatory demyelinating polyradiculoneuropathy), chronic inflammatory demyelinating polyradiculoneuropathy and various non‐inflammatory neuropathies. A consistent chemokine receptor expression pattern was immunohistochemically detected in inflammatory demyelinating neuropathies and quantitation of labelled mononuclear cells revealed significantly elevated cell counts compared with controls. CCR‐1 and CCR‐5 were primarily expressed by endoneurial macrophages, whereas CCR‐2, CCR‐4 and CXCR‐3 could be localized to invading T lymphocytes. Quantitative analysis revealed that CXCR‐3 was expressed at highest numbers by infiltrating T cells compared with the other receptors. Thus, expression and distribution of CXCR‐3 suggest a specific role of this receptor in chemokine‐mediated lymphocyte traffic into the inflamed PNS tissue. Therefore, we further analysed the expression of its ligands interferon‐γ‐inducible protein of 10 kDa (IP‐10) and monokine induced by interferon‐γ (Mig). Significantly increased levels of IP‐10 could be measured in the CSF of patients with inflammatory neuropathies, whereas no differences were observable for Mig. In situ hybridization for IP‐10 mRNA mirrored the distribution of the cognate receptor within the inflamed PNS, and delineated endothelial cells as the primary cellular source of IP‐10. Our results imply a pathogenic role for specific chemokine receptors and IP‐10 in the genesis of inflammatory demyelinating neuropathies.

Introduction

Guillain–Barré syndrome (GBS) and chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) are prototypic immune‐mediated neuropathies. Even with improved therapeutic options available, both diseases carry a grave prognosis as they are associated with sustained disability and even significant mortality (Winer et al., 1988; Hartung et al., 1998; Toyka, 1999).

Understanding of the underlying pathomechanisms is still incomplete. There is consensus that both GBS and CIDP result from aberrant humoral and cellular immune responses directed to peripheral nerve antigens. Collective evidence from animal studies points to a central role of T lymphocytes in the cellular arm of the immune response (Hughes et al., 1999; Hartung et al., 1995). The histopathology of classical GBS and CIDP is characterized by multifocal demyelination and mononuclear cellular infiltration (Cornblath et al., 1990; Arnason and Soliven, 1993; Schmidt et al., 1996). Since the PNS is largely separated from the systemic immune compartment by an anatomically rather tight interface between blood vessel wall and the neural parenchyma (the blood–nerve barrier), recent studies have focused on mechanisms involved in blood–nerve barrier disruption and extravasation of T lymphocytes into peripheral nerve. Two families of molecules are known to be involved in directing leucocytes into inflammatory sites: adhesion molecules and chemoattractants (Springer, 1994; Butcher and Picker, 1996; Archelos et al., 1999). Among the latter, chemokines, or chemoattractant cytokines, have gained particular interest, since these molecules selectively attract leucocyte subsets and activate leucocyte integrins to bind to cell adhesion molecule counter‐receptors on endothelial cells (Baggiolini, 1998; Campbell et al., 1998).

Chemokines have been implicated as important mediators in the pathogenesis of acute and chronic inflammatory disorders in man (Luster, 1998; Ward et al., 1998). Recently, multiple sclerosis, an immune‐mediated demyelinating disease of the CNS, has been added to this list (Hvas et al., 1997; McManus et al., 1998; Balashov et al., 1999; Ransohoff, 1999; Sørensen et al., 1999; van der Voorn et al., 1999). The expression of chemokines in the inflamed PNS has been studied in experimental autoimmune neuritis, an animal model for GBS (Fujioka et al., 1999; Zou et al., 1999; Kieseier et al., 2000). These investigations suggested that chemokines were upregulated in inflammatory demyelination of the PNS. Data on the expression of chemokines and their receptors in the human PNS have not been reported so far.

The present study focuses on the expression pattern of chemokine receptors in sural nerve biopsies from patients with classical GBS [acute inflammatory demyelinating polyradiculoneuropathy (AIDP)], CIDP and, for comparison, non‐inflammatory neuropathies (NIN). It is hoped that these new observations might help the design of novel therapeutic approaches specifically targeting the molecular mechanisms involved in the extravasation of blood‐borne inflammatory leucocytes into the neural parenchyma.

Material and methods

Human nerve biopsies

Sural nerve biopsies (n = 28) were obtained with informed consent from patients admitted to the Departments of Neurology at the Universities of Graz, Austria, Würzburg, Germany and Kyoto, Japan.

Three groups of patients were studied. The first group (n = 10) involved patients classified as GBS (AIDP) according to the criteria of Asbury and Cornblath (1990). The second group (n = 10) were diagnosed as CIDP according to accepted research criteria (Ad Hoc Subcommittee of the American Academy of Neurology AIDS Task Force, 1991). The third group served as control and included various NIN (n = 8) (Table 1). None of the patients with GBS and CIDP studied had received any immunomodulatory or immunosuppressive treatment within 3 months before biopsy.

We also analysed nerves and roots from two GBS autopsy cases from the Department of Neurology, Johns Hopkins University, School of Medicine, Baltimore, Md., USA.

Antibodies

The following primary antibodies were used for immunohistochemistry: goat polyclonal anti‐CCR‐1, anti‐CCR‐2b, anti‐CCR‐4 (Santa Cruz Biotechnology Inc., Santa Cruz, Calif., USA), mouse monoclonal anti‐CCR‐5 (R&D Systems Inc., Minneapolis, Minn., USA) and goat polyclonal anti‐CXCR‐3 (Leuko Site Inc., Cambridge, Mass., USA). Tissues were also analysed with murine monoclonal anti‐CD3 (Serotec, Oxford, UK), mouse monoclonal anti‐CD68 (DAKO, Hamburg, Germany) and corresponding secondary antibodies (Vector Laboratories, Burlingame, Calif., USA) to define inflammatory infiltrates.

Immunohistochemistry

Formalin‐fixed, paraffin‐embedded tissue sections were placed on Superfrost Plus slides (Fisher Scientific Co., Pittsburgh, Penn., USA), deparaffinized in xylenes and rehydrated through graded ethanol into PBS (phosphate‐buffered saline). Antigen retrieval was performed, if necessary, by incubating the slides with protease XXIV (Sigma Chemical Co., St Louis, Mo., USA) for 30 min at 37°C, or microwaving in 1 mM EDTA, pH 8.0.

Sections were blocked with 10% human serum and incubated with primary antibodies at 4°C overnight. Thereafter, a biotinylated secondary antibody and avidin–biotin–horseradish peroxidase complex (Vectastain Elite, Vector Laboratories) were applied with 3,3′‐diaminobenzidine as peroxidase substrate, according to the manufacturer’s instructions. After development with DAB substrate, sections were counterstained with haematoxylin (Sigma), and mounted in 80% glycerol in PBS. For each antibody, optimal concentrations of primary and secondary antibody were determined. Endogenous peroxidase activity was suppressed by incubating the sections with 3% H2O2 in methanol for 10 min prior to adding the secondary antibody.

Quantitative assessment

Analysis of positive immunoreactivity was performed by an observer blinded to the patient group studied. Endoneurial and epineurial immunoreactivity was evaluated across an entire transverse section of each biopsy, and numbers of positive cells were related to the total area of each section as determined on digitized images using an Axiophot 2 microscope (Carl Zeiss, Jena, Germany). Only immunoreactivity associated with a cell nucleus was accepted.

In situ hybridization

Full‐length interferon‐γ‐inducible protein of 10 kDa (IP‐10) cDNA (kindly provided by Dr Jeffrey Ravetch, Rockefeller University) served as the transcription template. Radiolabelled hybridization probes of both polarities incorporating [3H]UTP and [3H]CTP (Amersham, Arlington Heights, Ill., USA) were generated by in vitro transcription and in situ hybridization (ISH), performed as described previously (Ransohoff et al., 1997). Emulsion autoradiography with NTB‐2 (Kodak, New Haven, Conn., USA) was performed for 4 weeks. Following development, sections were lightly counterstained with haematoxylin. Cells considered ISH‐positive contained more than eight grains. Sense‐strand probes were hybridized with each tissue section to establish baseline hybridization to genomic DNA and non‐specific background. Antisense hybridization analysis of control (NIN) nerves confirmed the relationship between ISH‐positive cells and sites of inflammation. Initial ISH using a β‐actin probe confirmed the presence of detectable mRNA in all cell‐types and verified that the quality of tissues and technical aspects of ISH were uniform.

ELISA (enzyme‐linked immunosorbent assay)

To measure expression levels of IP‐10 in human CSF, samples were obtained with informed consent from patients diagnosed as GBS (n = 6) or CIDP (n = 6) at the Graz Department of Neurology. CSF samples from patients with NIN diseases (n = 4) served as controls. The CSF white blood cell count was determined as [medians (ranges)] 4 (1–10) cells/µl in the GBS group, 5 (2–12) cells/µl in the CIDP group and 2 (0–4) cells/µl in the control group. The total protein content in these CSF samples was measured as [medians (ranges)] 0.67 (0.56–0.81) g/l in the GBS group, 0.62 (0.54–0.78) g/l in the CIDP group and 0.32 (0.23–0.41) g/l in the control samples. ELISA to quantitate IP‐10 and monokine induced by interferon‐γ (Mig) expression in the CSF was performed. Briefly, all frozen samples were thawed and IP‐10 and Mig levels quantified using paired antibody sandwich ELISA kits. Detection and capture antibody concentrations were optimized using recombinant human chemokine standards (R&D Systems, Abingdon, UK). The interassay variabilities were <10% and the lower detection limits were 10 pg/ml each.

Statistical analysis

Statistical analysis was carried out using the Kruskal–Wallis test and the Mann–Whitney U‐test for post hoc analysis with Bonferroni correction.

Results

CD3 and CD68

All tissue sections from patients with GBS and CIDP showed active disease with foci of inflammatory cells. We first analysed the presence of T lymphocytes and macrophages in each individual nerve using immunohistochemistry. Perivascular infiltrates could be detected in the epineurium and perineurium, predominantly consisting of CD3+ T cells, as well as in the endoneurium. Within the latter, we detected large numbers of CD68+ immunoreactive cells, indicative of macrophages. Analysis of nerves and roots from two GBS autopsy cases were in line with our findings in sural nerve biopsies (data not shown), indicating that inflammatory mechanisms defined in this study appear to be reflective of those operating in the entire PNS.

In the NIN group, inflammatory infiltrates were not observed, and immunoreactivity against CD3 and CD68 was sparse.

Control immunohistochemical analyses after omission of the primary antibody only showed background staining for each individual antibody used in the present study.

α‐Chemokine receptors

CXCR‐3

When sections from inflammatory demyelinating neuropathies were analysed with anti‐CXCR‐3 antibodies, small, round immunoreactive cells were noted in perivascular cuffs (Fig. 1). Such infiltrates were primarily located in the epineurium and perineurium, and were often clustered around arterioles or capillaries (Fig. 2). On serial sections these cells, morphologically consistent with lymphocytes, were similar in appearance and distribution to cells detected with the pan‐T‐cell marker anti‐CD3.

No immunoreactivity for CXCR‐3 could be discerned in the NIN group.

β‐Chemokine receptors

CCR‐1

In GBS and CIDP cases, CCR‐1 immunoreactivity was localized to mononuclear cells, and was highly expressed within the endoneurium (Fig. 3A, C and D). No immunoreactive cells were detectable in the epineurium and perineurium (Fig. 2). The distribution of CCR‐1+ cells was similar to the pattern found with the macrophage marker anti‐CD68.

CCR‐2

The number of cells immunoreactive to CCR‐2 was low, and was detected primarily in GBS cases, only rarely in CIDP sections and never in the NIN controls. Immunoreactivity was localized to small, mononuclear cells in the blood vessels and in perivascular infiltrates found in the epineurium and perineurium (Fig. 2). On serial sections, appearance and distribution of the staining pattern was similar to the immunoreactivity observed with the anti‐CD3 antibody (Fig. 3B, E and F).

CCR‐4

The smallest number of immunoreactive cells was detected with the anti‐CCR‐4 antibody. CCR‐4+ cells were seen only in GBS, exclusively in the endoneurium (Fig. 2). On serial sections, the staining pattern resembled that obtained on immunohistochemistry with the anti‐CD3 antibody (Fig. 3G, I and K).

CCR‐5

In active demyelinating neuropathies, CCR‐5 immunoreactive mononuclear cells were detectable in large numbers in the endoneurium. CCR‐5+ cells were rarely observed in perivascular cuffs in the epineurium and perineurium (Fig. 2). On serial sections, the appearance and distribution of the immunoreactivity against CCR‐5 in the endoneurium resembled the pattern obtained with the anti‐CD68 antibody (Fig. 3H and L). In contrast, CCR‐5+ cells observed in perivascular inflammatory infiltrates in the epineurium and perineurium expressed a similar staining as detectable with the pan‐T‐cell marker CD3 (Fig. 1).

Quantitative analysis

To obtain insight into the frequency of chemokine receptor expression in the various disease groups, labelled mononuclear cells were quantified in each nerve and the results expressed as cells/mm2. For each individual chemokine receptor studied increased numbers were noted in inflammatory demyelinating disease. For all chemokine receptors these numbers were statistically significantly different when comparing the GBS with the CIDP and NIN groups. In contrast, the difference in numbers of CCR‐2‐ and CCR‐4‐expressing cells did not reach statistical significance when comparing CIDP with NIN cases. However, the expression of CXCR‐3, CCR‐1 and CCR‐5 was significantly higher in the CIDP group compared with the NIN cases (Fig. 4).

Since the number of CD3+ and CD68+ cells varies within each individual group, as has been reported previously (Kiefer et al., 1998; Bosboom et al., 1999), we further investigated the cellular expression of chemokine receptors as a fraction of the total number of the respective cell type. This analysis revealed that the percentage of chemokine receptor‐expressing cells is significantly higher in GBS compared with the CIDP and NIN groups. When comparing CIDP with NIN cases, percentages of chemokine receptor‐expressing leucocytes were statistically significant, except for CCR‐2 and CCR‐4 (Fig. 5). CXCR‐3 was the receptor expressed at highest rate by invading T cells, whereas CCR‐1 was the receptor with the highest expression rate found on CD68+ macrophages.

IP‐10

IP‐10 in the CSF

A statistically significant increase of IP‐10 protein in the CSF from patients with GBS and CIDP was observed in comparison with the NIN controls when analysed by Kruskal–Wallis test and Mann–Whitney U‐test (P < 0.05). In the GBS samples, mean IP‐10 protein content (± standard error) was 1343.36 (± 454.37) pg/ml, whereas in the CSF from CIDP patients 1064.76 (± 169.26) and in the NIN controls 483.85 (± 104.8) pg/ml was measured. This increase in IP‐10 protein levels did not correlate with CSF cell counts or the total protein levels.

IP‐10 in the inflamed peripheral nerve

IP‐10 immunoreactivity could not be demonstrated in fixed sections available for these studies, as was previously found in analysis of multiple sclerosis brain sections (Sørensen et al., 1999). Therefore, IP‐10 expression was analysed by ISH. We found abundant expression of IP‐10 mRNA associated with inflamed perineurial vessels in GBS sections (Fig. 6). IP‐10 mRNA expression was associated with intrinsic vascular cells, probably endothelial. Control nerves did not demonstrate ISH‐positive cells. However, β‐actin mRNA expression was found in large numbers in all tissue samples, confirming the presence of detectable mRNA (data not shown).

Mig in the CSF

Analysis of Mig protein expression in the CSF from patients with GBS, CIDP and NIN by ELISA did not reveal any statistically significant differences (data not shown).

Discussion

Selective accumulation of leucocyte subpopulations in an inflamed tissue is mediated by differential expression of leucocyte and endothelial adhesion receptors, chemokines and chemokine receptors. Working in concert, these components permit cells to localize selectively to parenchymal inflammatory sites. In the present study the specific expression pattern of α‐ and β‐chemokine receptors on mononuclear phagocytes and T cells in inflammatory demyelination of the PNS was investigated. We found consistent alterations in sural nerve biopsies in the acutely (GBS) as well as in the chronically inflamed (CIDP) PNS in comparison with non‐inflammatory (NIN) controls.

Studying the expression pattern of α‐chemokine receptors we detected CXCR‐3 immunoreactivity on perivascular mononuclear cells within the epineurium and perineurium. On serial sections, these cells were identified as CD3+ T lymphocytes. Quantitative analysis revealed that CXCR‐3 was expressed by a higher percentage of T cells than any other receptor studied. CXCR‐3 is preferentially expressed by T helper type 1 (Th1) lymphocytes and is recognized by the CXC chemokines IP‐10, Mig and I‐TAC (Loetscher et al., 1996a; Bonecchi et al., 1998; Lu et al., 1999; Moser and Loetscher, 2001). T lymphocytes play a critical role in the pathogenesis of inflammatory demyelination of the PNS. Especially Th1 cells, through the release of pro‐inflammatory cytokines, such as interferon‐γ and tumour necrosis factor‐α, are thought to render immigrating or resident macrophages activated to augment the generation and discharge of toxic molecules or to engage increased phagocytic activity (Hartung et al., 1998). It is conceivable that the local production of interferon‐γ by Th1 cells induces the production of IP‐10, Mig and I‐TAC, while downregulating the expression of other chemokines (Baggiolini, 1998), providing for selectivity in the recruitment of specific effector lymphocytes. In the present study, significantly elevated levels of IP‐10 were measured in the CSF of patients with GBS and CIDP, underlining an important role for this chemokine in the pathogenesis of inflammatory demyelination in the PNS. However, protein levels did not correlate with the number of cells within the CSF. Consistent with expectations based on classical studies, most of the samples analysed did not show pleocytosis. Increased chemokine expression in the CSF has been described in the context of other inflammatory disorders of the CNS, such as multiple sclerosis or HIV infections, but in these studies CSF samples were characterized by increased cell counts (Kolb et al., 1999; Sørensen et al., 1999). However, our finding suggests that chemokine levels will not necessarily be reflected in the cellular composition of the body fluid investigated. Rather, these proteins are secreted in the inflamed tissue, which apparently equilibrates with the body fluid.

While attempting to identify the expression of other ligands to CXCR‐3 we failed to detect significant differences in the expression levels of Mig in the CSF samples studied. It is unclear at present whether technical issues or differential regulatory events account for this observation.

In order to determine the cellular source, we detected increased expression of IP‐10 mRNA in the inflamed PNS and localized it to endothelial cells along the blood vessel walls, using ISH. This finding is consistent with a recent report demonstrating that vascular endothelium in the CNS is capable of producing IP‐10 in response to challenge with mouse adenovirus type‐1 (Charles et al., 1999). However, in the majority of autoimmune or virally induced inflammatory diseases of the CNS, expression of IP‐10 (either mRNA or protein) was primarily associated with astrocytes rather than endothelium (Ransohoff et al., 1993; Tani et al., 1996; Asensio and Campbell, 1997; Sørensen et al., 1999; Simpson et al., 2000). Since migration of autoreactive T cells from blood into the PNS is important in the genesis of autoimmune inflammatory demyelination, the localization in the endothelium assigns IP‐10 a strategic role in this process. In selected animal models, the restricted expression and selectivity for a single receptor on T cells suggest that IP‐10 is critically involved in lymphocyte recruitment to the nervous system. Our findings support this notion and emphasize the functional significance of IP‐10–CXCR‐3 interactions in recruiting T lymphocytes from the circulation into the neural parenchyma.

Coexpression of CXCR‐3 and CCR‐5 on T lymphocytes has been shown to predominate in some autoimmune diseases, such as rheumatoid arthritis, but not in multiple sclerosis (Qin et al., 1998; Sørensen et al., 1999). We addressed this question in the current study and found that only a minority of the infiltrating T cell population appeared to display both receptors on their surface. The number of CXCR‐3+ T lymphocytes was much higher in comparison with CD3+ cells expressing the β‐chemokine receptor CCR‐5. This observation may be explained by the fact that CCR‐5 expression on T lymphocytes is induced transiently by interleukin (IL)‐2 (Loetscher et al., 1996b), in contrast to CXCR‐3, which exhibits sustained expression on activated cells (Sallusto et al., 1998).

Although CCR‐5 immunoreactivity was associated with a minority of T cells in perivascular infiltrates in the epineurium and perineurium, this receptor was predominately expressed on mononuclear cells within the endoneurium, in all likelihood phagocytic macrophages in active demyelinating disease. This result is consistent with previous investigations in the CNS where macrophages, microglia cells and T lymphocytes were found to express CCR‐5 in lesional areas of multiple sclerosis brain tissue (Sørensen et al., 1999).

CCR‐1 was expressed by ∼50% of endoneurial macrophages in acute and chronic inflammatory demyelination of the PNS. We did not detect CCR‐1+ immunoreactivity on T cells. This finding indicates striking differences in the functional role of chemokine receptors in the pathogenesis of GBS/CIDP as compared with multiple sclerosis, since CCR‐1 was detected in multiple sclerosis lesions only on newly infiltrating monocytes in perivascular spaces and lesion borders (Sørensen et al., 1999; Trebst et al., 2001). Further studies are needed to clarify the specific role of this β‐chemokine receptor in inflammatory demyelination of the PNS.

Among the β‐chemokine receptors studied, CCR‐2 and CCR‐4 appeared to be expressed by T cells almost exclusively. CCR‐2 expression is primarily associated with circulating monocytes, although it is present on dendritic cells, chronically activated T lymphocytes, natural killer cells and basophils (Luster, 1998). In the present study, CCR‐2 was found on mononuclear cells located in the blood vessel lumen and in perivascular cuffs. On lymphocytes, CCR‐2 is expressed only after prolonged, high‐level stimulation by IL‐2 (Loetscher et al., 1996b). Since in the inflamed PNS this receptor appeared to be expressed by T cells in vessels and perivascular cuffs, it is likely that this staining depicted chronically activated T lymphocytes entering the PNS. Thus, our finding would support the present hypothesis that T cells migrate in response to chemokines after IL‐2‐mediated proliferation (Luster, 1998). Increased blood levels of soluble IL‐2, evidence of systemic immune activation, were reported in patients with GBS and CIDP (Hartung et al., 1991). In the present study, CCR‐2 was only detected in inflammatory demyelinating disease, with higher numbers in the acute GBS cases but never in the NIN controls. Therefore, this finding emphasizes the relevance of systemic immune activation in the pathogenesis of inflammatory demyelination (Kieseier et al., 1999).

CCR‐4 was detected occasionally on T lymphocytes in acute active demyelinating disease. Given that Th2 cells exhibit CCR‐4 (Luster, 1998), the present observation suggests that we stained a subpopulation of T lymphocytes in the inflamed PNS, which could stimulate immunoglobulin G (IgG) 1a production by B cells and, potentially, inhibit the production of proinflammatory cytokines in the ongoing disease process. However, given the fact that inflammatory demyelination of the peripheral nerve is a prominent Th1‐driven process, CCR‐4 could also function as a molecule involved in cell migration to the PNS, as suggested for the skin (Campbell et al., 1999). Further studies are clearly warranted to shed further light on this controversial topic.

Quantitative analysis of the chemokine receptor expression pattern in the present study revealed significantly increased numbers for all chemokine receptors investigated in the acute inflammatory demyelinating disease in comparison with the CIDP and NIN groups. This finding probably reflects ongoing invasion of the neural parenchyma by immunocompetent cells during the acute immune‐mediated inflammatory process, since chemokine receptors are downregulated after ligand engagement. In its chronic variant, CIDP, the number of chemokine receptors found to be expressed was lower, but in most cases still significantly higher than in NIN cases, pointing to the ongoing active immune response in a chronic progressive inflammatory disorder.

In summary, we demonstrated consistent alterations of specific chemokine receptor expression in inflammatory demyelinating diseases of the PNS. Similar to recent observations in inflammatory demyelination of the CNS (Tani and Ransohoff, 1994; Balashov et al., 1999; Mennicken et al., 1999; Sørensen et al., 1999; Misu et al., 2001), increased expression of CXCR‐3 and CCR‐5 could be observed in the inflamed PNS. The cellular source of the CXCR‐3 ligand IP‐10, however, differed in comparison with recent studies in the CNS. Morover, we found striking differences in the expression pattern of the chemokine receptors CCR‐1, CCR‐2 and CCR‐4 in comparison with previous investigations in multiple sclerosis. It will be of interest to clarify the functional implications of this differential display in the pathogenesis of immune‐mediated demyelination in the CNS and PNS.

Because mononuclear cellular infiltration is one of the pathological hallmarks of classical GBS and some of its variants (Asbury et al., 1969; Honavar et al., 1991; Griffin et al., 1996; Hafer‐Macko et al., 1996) and is considered to be of paramount importance in the pathogenesis of immunoinflammatory demyelination (Griffin et al., 1993; Hartung, 1995), chemokines and their receptors appear to assume a strategic role in the process of leucocyte recruitment. It is likely that chemokines might play a deleterious role in GBS and CIDP, but further studies are required to comprehend fully the respective role of each chemokine in the various phases of the immune response and to establish their action on Schwann cells and axons. Demonstrating a selective expression pattern of specific chemokine receptors, and the α‐chemokine IP‐10 in the inflamed human PNS and CSF of affected patients, our data support the hypothesis that chemokines and their specific receptors participate in the pathogenesis of autoimmune demyelinating diseases in the human and imply that these molecules should be considered as potential targets for therapeutic intervention.

Acknowledgements

We thank T. Phan for expert assistance with figure preparation and W. N. Löscher for statistical advice. The technical assistance of B. Kirby and H. Pischel is gratefully acknowledged. This study was supported by the Gemeinnützige Hertie Stiftung and research grants from the Austrian Federal Ministry of Education, Science, and Culture (GZ 70.057/2‐Pv/4/99) to H.‐P.H., the National Institutes of Health, Maryland, USA (NS 32151, NS 38667) to R.M.R., and a fellowship support from the National Multiple Sclerosis Society, New York, USA to M.T.

Fig. 1 Expression of CXCR‐3 (type 3 CXC chemokine receptor) in a sural nerve biopsy from a patient with GBS. Perivascular infiltrates, primarily consisting of CD3+ T cells, exhibit strong immunoreactivity to CXCR‐3. In contrast, no immunostaining against CCR‐2b or CCR‐4 could be observed, whereas some invading mononuclear cells expressed CCR‐5 as well (arrows). Original magnification ×600.

Fig. 1 Expression of CXCR‐3 (type 3 CXC chemokine receptor) in a sural nerve biopsy from a patient with GBS. Perivascular infiltrates, primarily consisting of CD3+ T cells, exhibit strong immunoreactivity to CXCR‐3. In contrast, no immunostaining against CCR‐2b or CCR‐4 could be observed, whereas some invading mononuclear cells expressed CCR‐5 as well (arrows). Original magnification ×600.

Fig. 2 Representative immunocytochemical stains mirroring the distribution pattern of various chemokine receptors (arrows) in a sural nerve biopsy from a patient with Guillain‐Barré syndrome: CXCR‐3 and CCR‐2b could primarily be localized to perivascular infiltrates in the epineurium and perineurium; CCR‐1 was found in large numbers within the endoneurium only; CCR‐4 was detectable in low numbers within the endoneurium; and CCR‐5 could be observed within the endoneurium and within perivascular infiltrates in the epineurium and perineurium. Original magnification ×200.

Fig. 2 Representative immunocytochemical stains mirroring the distribution pattern of various chemokine receptors (arrows) in a sural nerve biopsy from a patient with Guillain‐Barré syndrome: CXCR‐3 and CCR‐2b could primarily be localized to perivascular infiltrates in the epineurium and perineurium; CCR‐1 was found in large numbers within the endoneurium only; CCR‐4 was detectable in low numbers within the endoneurium; and CCR‐5 could be observed within the endoneurium and within perivascular infiltrates in the epineurium and perineurium. Original magnification ×200.

Fig. 3 Immunohistochemical analysis of the expression of various chemokine receptors in a sural nerve biopsy from a patient with GBS. (A) Within the endoneurium immunoreactive cells, staining for CCR‐1 could be detected (arrows) in a pattern and distribution similar to CD68+ macrophages, as shown on serial sections (C). (D) In a case of hereditary motor and sensory neuropathy type I, no positive staining against CCR‐1 could be found. (A′) High‐power view of a CCR‐1+ cell. (B) CCR‐2 immunoreactivity (arrows) could be localized to small, mononuclear cells in blood vessels (L = lumen) and (E) in perivascular infiltrates. (F) On serial sections these cells were identified as CD3+ T cells. (G) CCR‐5 immunoreactive mononuclear cells (arrows) were detectable in large numbers within the endoneurium, but rarely in perivascular cuffs in the perineurium and epineurium (I). (K) CCR‐5+ cells were identified to be primarily CD68+ macrophages. (H) Mononuclear cells positive for CCR‐4 were detectable in the endoneurium, resembling the staining pattern in single cells obtained with an anti‐CD3 antibody (L). Original magnification ×348 (AD, G, H, L) and ×522 (A′, E, F, I, K).

Fig. 3 Immunohistochemical analysis of the expression of various chemokine receptors in a sural nerve biopsy from a patient with GBS. (A) Within the endoneurium immunoreactive cells, staining for CCR‐1 could be detected (arrows) in a pattern and distribution similar to CD68+ macrophages, as shown on serial sections (C). (D) In a case of hereditary motor and sensory neuropathy type I, no positive staining against CCR‐1 could be found. (A′) High‐power view of a CCR‐1+ cell. (B) CCR‐2 immunoreactivity (arrows) could be localized to small, mononuclear cells in blood vessels (L = lumen) and (E) in perivascular infiltrates. (F) On serial sections these cells were identified as CD3+ T cells. (G) CCR‐5 immunoreactive mononuclear cells (arrows) were detectable in large numbers within the endoneurium, but rarely in perivascular cuffs in the perineurium and epineurium (I). (K) CCR‐5+ cells were identified to be primarily CD68+ macrophages. (H) Mononuclear cells positive for CCR‐4 were detectable in the endoneurium, resembling the staining pattern in single cells obtained with an anti‐CD3 antibody (L). Original magnification ×348 (AD, G, H, L) and ×522 (A′, E, F, I, K).

Fig. 4 Total number of immunoreactive cells within entire transverse sections of each biopsy expressed as immunoreactive cells/mm2. Values within each group of patients studied are given as mean ± standard deviation. Differences in the expression of chemokine receptors are statistically significant (P < 0.05) when analysed with the Kruskal–Wallis test and the Mann–Whitney U‐test. Biopsies are from patients with GBS (n = 10), CIDP (n = 10) and NIN (n = 8). An asterisk above the error bar indicates statistical significance compared with CIDP and with NIN, and a circle indicates statistical significance compared with NIN. n.s. = not significant. Note different ordinate scale.

Fig. 4 Total number of immunoreactive cells within entire transverse sections of each biopsy expressed as immunoreactive cells/mm2. Values within each group of patients studied are given as mean ± standard deviation. Differences in the expression of chemokine receptors are statistically significant (P < 0.05) when analysed with the Kruskal–Wallis test and the Mann–Whitney U‐test. Biopsies are from patients with GBS (n = 10), CIDP (n = 10) and NIN (n = 8). An asterisk above the error bar indicates statistical significance compared with CIDP and with NIN, and a circle indicates statistical significance compared with NIN. n.s. = not significant. Note different ordinate scale.

Fig. 5 Density of CD68+ macrophages carrying the chemokine receptors CCR‐1 and CCR‐5 and density of CD3+ T lymphocytes expressing the chemokine receptors CCR‐2, CCR‐4 and CXCR‐3 in biopsies from patients with GBS (n = 10), CIDP (n = 10) and NIN (n = 8) (box and whisker plots). The central horizontal bar denotes the median value, the top and bottom lines of the box denote the 75th and 25th percentiles, respectively, and the top and bottom ends of the error bar denote the 95th and 5th percentiles. ***P < 0.001, **P < 0.01, *P < 0.05; n.s. = not significant, when the respective disease groups were analysed with the Kruskal–Wallis test and the Mann–Whitney U‐test.

Fig. 5 Density of CD68+ macrophages carrying the chemokine receptors CCR‐1 and CCR‐5 and density of CD3+ T lymphocytes expressing the chemokine receptors CCR‐2, CCR‐4 and CXCR‐3 in biopsies from patients with GBS (n = 10), CIDP (n = 10) and NIN (n = 8) (box and whisker plots). The central horizontal bar denotes the median value, the top and bottom lines of the box denote the 75th and 25th percentiles, respectively, and the top and bottom ends of the error bar denote the 95th and 5th percentiles. ***P < 0.001, **P < 0.01, *P < 0.05; n.s. = not significant, when the respective disease groups were analysed with the Kruskal–Wallis test and the Mann–Whitney U‐test.

Fig. 6 Expression of the chemokine IP‐10 in GBS sections. Representative (A) brightfield and (B) corresponding darkfield autoradiographs after hybridizing for IP‐10 mRNA, which was found to be associated with inflamed perineurial vessels (arrows). Original magnification ×400.

Fig. 6 Expression of the chemokine IP‐10 in GBS sections. Representative (A) brightfield and (B) corresponding darkfield autoradiographs after hybridizing for IP‐10 mRNA, which was found to be associated with inflamed perineurial vessels (arrows). Original magnification ×400.

Table 1

Summary of patient characteristics

Diagnosis No. of cases Age (years) (range) Sex ratio (M: F) Duration of disease* (range) 
GBS (AIDP) 10 48 (19–69) 7: 3 8 days (3–14) 
CIDP 10 62 (43–74) 8: 2 3 years (0.3–15) 
NIN 52 (27–78) 6: 2 n.a. 
 HNPP I  3    
 HMSN I  1    
 HMSN II  2    
 Normal nerves  2    
Diagnosis No. of cases Age (years) (range) Sex ratio (M: F) Duration of disease* (range) 
GBS (AIDP) 10 48 (19–69) 7: 3 8 days (3–14) 
CIDP 10 62 (43–74) 8: 2 3 years (0.3–15) 
NIN 52 (27–78) 6: 2 n.a. 
 HNPP I  3    
 HMSN I  1    
 HMSN II  2    
 Normal nerves  2    

M = male; F = female; HNPP I = hereditary neuropathy with liability to pressure palsy; HMSN I, II = hereditary motor and sensory neuropathies type I, type II; n.a. = not applicable. *Before biopsy. Figures denote median values and ranges.

References

1.
Ad Hoc Subcommittee of the American Academy of Neurology AIDS Task Force. Research criteria for diagnosis of chronic inflammatory demyelinating polyneuropathy (CIDP). [Review].
Neurology
 
1991
;
41
:
617
–18.
Archelos JJ, Previtali SC, Hartung H‐P. The role of integrins in immune‐mediated diseases of the nervous system. [Review].
Trends Neurosci
 
1999
;
22
:
30
–8.
Arnason BGW, Soliven B. Acute inflammatory demyelinating polyradiculoneuropathy. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo J, editors. Peripheral neuropathy. 3rd ed. Philadelphia (PA): W. B. Saunders;
1993
. p.
1437
–97.
Asbury AK, Cornblath DR. Assessment of current diagnostic criteria for Guillain‐Barré syndrome. [Review].
Ann Neurol
 
1990
;
27
Suppl:
S21
–4.
Asbury AK, Arnason BG, Adams RD. The inflammatory lesion in idiopathic polyneuritis. Its role in pathogenesis.
Medicine (Baltimore)
 
1969
;
48
:
173
–215.
Asensio V, Campbell IL. Chemokine gene expression in the brains of mice with lymphocytic choriomeningitis.
J Virol
 
1997
;
71
:
7832
–40.
Baggiolini M. Chemokines and leukocyte traffic. [Review].
Nature
 
1998
;
392
:
565
–8.
Baggiolini M, Dewald B, Moser B. Human chemokines: an update. [Review].
Annu Rev Immunol
 
1997
;
15
:
675
–705.
Balashov KE, Rottman JB, Weiner HL, Hancock WW. CCR5+ and CXCR3+ T cells are increased in multiple sclerosis and their ligands MIP‐1alpha and IP‐10 are expressed in demyelinating brain lesions.
Proc Natl Acad Sci USA
 
1999
;
96
:
6873
–8.
Bonecchi R, Bianchi G, Bordignon PP, D’Ambrosio D, Lang R, Borsatti A, et al. Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s.
J Exp Med
 
1998
;
187
:
129
–34.
Bosboom WM, Van den Berg LH, De Boer L, Van Son MJ, Veldman H, Franssen H, et al. The diagnostic value of sural nerve T cells in chronic inflammatory demyelinating neuropathy.
Neurology
 
1999
;
53
:
837
–45.
Butcher EC, Picker LJ. Lymphocyte homing and homeostasis. [Review].
Science
 
1996
;
272
:
60
–6.
Campbell JJ, Hedrick J, Zlotnik A, Siani MA, Thompson DA, Butcher EC. Chemokines and the arrest of lymphocytes rolling under flow conditions.
Science
 
1998
;
279
:
381
–4.
Campbell JJ, Haraldsen G, Pan J, Rottman J, Qin S, Ponath P, et al. The chemokine receptor CCR4 in vascular recognition by cutaneous but not intestinal memory T cells.
Nature
 
1999
;
400
:
776
–80.
Charles PC, Chen X, Horwitz MS, Brosnan CF. Differential chemokine induction by the mouse adenovirus type‐1 in the central nervous system of susceptible and resistant strains of mice.
J Neurovirol
 
1999
;
5
:
55
–64.
Cornblath DR, Griffin DE, Welch D, Griffin JW, McArthur JC. Quantitative analysis of endoneurial T‐cells in human sural nerve biopsies.
J Neuroimmunol
 
1990
;
26
:
113
–8.
Fujioka T, Purev E, Rostami A. Chemokine mRNA expression in the cauda equina of Lewis rats with experimental allergic neuritis.
J Neuroimmunol
 
1999
;
97
:
51
–9.
Griffin JW, George R, Ho T. Macrophage system in peripheral nerves. A review. [Review].
J Neuropathol Exp Neurolol
 
1993
;
52
:
553
–60.
Griffin JW, Li CY, Ho TW, Tian M, Gao CY, Xue P, et al. Pathology of the motor‐sensory axonal Guillain‐Barré syndrome.
Ann Neurol
 
1996
;
39
:
17
–28.
Hafer‐Macko CE, Sheikh KA, Li CY, Ho TW, Cornblath DR, McKhann GM, et al. Immune attack on the Schwann cell surface in acute inflammatory demyelinating polyneuropathy.
Ann Neurol
 
1996
;
39
:
625
–35.
Hartung H‐P. Pathogenesis of inflammatory demyelination: implications for therapy. [Review].
Curr Opin Neurol
 
1995
;
8
:
191
–9.
Hartung H‐P, Reiners K, Schmidt B, Stoll G, Toyka KV. Serum interleukin‐2 concentrations in the Guillain‐Barré syndrome and chronic idiopathic demyelinating polyradiculoneuropathy: comparison with other neurological diseases of presumed immunopathogenesis.
Ann Neurol
 
1991
;
30
:
48
–53.
Hartung H‐P, Pollard JD, Harvey GK, Toyka KV. Immunopathogenesis and treatment of the Guillain‐Barré syndrome, Part I and II. [Review].
Muscle Nerve
 
1995
;
18
:
137
–64.
Hartung H‐P, Toyka KV, Griffin JW. Guillain‐Barré syndrome and chronic inflammatory demyelinating polyradiculoneuropathy. In: Antel J, Birnbaum G, Hartung H‐P, editors. Clinical neuroimmunology. Malden (MA): Blackwell Science;
1998
. p.
294
–306.
Honavar M, Tharakan JK, Hughes RA, Leibowitz S, Winer JB. A clinicopathological study of the Guillain‐Barré syndrome. Nine cases and literature review.
Brain
 
1991
;
114
:
1245
–69.
Hughes RAC, Hadden RDM, Gregson NA, Smith KJ. Pathogenesis of Guillain‐Barré syndrome. [Review].
J Neuroimmunol
 
1999
;
100
:
74
–97.
Hvas J, McLean C, Justesen J, Kannourakis G, Steinman L, Oksenberg JR, et al. Perivascular T cells express the pro‐inflammatory chemokine RANTES mRNA in multiple sclerosis lesions.
Scand J Immunol
 
1997
;
46
:
195
–203.
Kiefer R, Kieseier BC, Brück W, Hartung H‐P, Toyka KV. Macrophage differentiation antigens in acute and chronic autoimmune polyneuropathies.
Brain
 
1998
;
121
:
469
–79.
Kieseier BC, Storch MK, Archelos JJ, Martino G, Hartung H‐P. Effector pathways in immune mediated central nervous system demyelination. [Review].
Curr Opin Neurol
 
1999
;
12
:
323
–36.
Kieseier BC, Krivacic K, Jung S, Pischel H, Toyka KV, Ransohoff RM, et al. Sequential expression of chemokines in experimental autoimmune neuritis.
J Neuroimmunol
 
2000
;
110
:
121
–9.
Kolb SA, Sporer B, Lahrtz F, Koedel U, Pfister HW, Fontana A. Identification of a T cell chemotactic factor in the cerebrospinal fluid of HIV‐1‐infected individuals as interferon‐gamma inducible protein 10.
J Neuroimmunol
 
1999
;
93
:
172
–81.
Loetscher M, Gerber B, Loetscher P, Jones SA, Piali L, Clark‐Lewis I, et al. Chemokine receptor specific for IP10 and Mig: structure, function, and expression in activated T‐lymphocytes.
J Exp Med
 
1996
a;
184
:
963
–9.
Loetscher P, Seitz M, Baggiolini M, Moser M. Interleukin‐2 regulates CC chemokine receptor expression and chemotactic responsiveness in T lymphocytes. J Exp Med
1996
b;
184
:
569
–77.
Lu B, Humbles A, Bota D, Gerard C, Moser B, Soler D, et al. Structure and function of the murine chemokine receptor CXCR3.
Eur J Immunol
 
1999
;
29
:
3804
–12.
Luster AD. Chemokines – chemotactic cytokines that mediate inflammation. [Review].
New Engl J Med
 
1998
;
338
:
436
–45.
McManus C, Berman JW, Brett FM, Staunton H, Farrell M, Brosnan CF. MCP‐1, MCP‐2 and MCP‐3 expression in multiple sclerosis lesions: an immunohistochemical and in situ hybridization study.
J Neuroimmunol
 
1998
;
86
:
20
–9.
Mennicken F, Maki R, de Souza EB, Quirion R. Chemokines and chemokine receptors in the CNS: a possible role in neuroinflammation and patterning. [Review].
Trends Pharmacol Sci
 
1999
;
20
:
73
–8.
Misu T, Onodera H, Fujihara K, Matsushima K, Yoshie O, Okita N, et al. Chemokine receptor expression on T cells in blood and cerebrospinal fluid at relapse and remission of multiple sclerosis: imbalance of Th1/Th2‐associated chemokine signaling.
J Neuroimmunol
 
2001
;
114
:
207
–12.
Moser B, Loetscher P. Lymphocyte traffic control by chemokines. [Review].
Nat Immunol
 
2001
;
2
:
123
–8.
Qin S, Rottman JB, Myers P, Kassam N, Weinblatt M, Loetscher M, et al. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory reactions.
J Clin Invest
 
1998
;
101
:
746
–54.
Ransohoff RM. Mechanisms of inflammation in MS tissue: adhesion molecules and chemokines. [Review].
J Neuroimmunol
 
1999
;
98
:
57
–68.
Ransohoff RM, Hamilton TA, Tani M, Stoler MH, Shick HE, Major JA, et al. Astrocyte expression of mRNA encoding cytokines IP‐10 and JE/MCP‐1 in experimental autoimmune encephalomyelitis.
FASEB J
 
1993
;
7
:
592
–600.
Ransohoff RM, Tani A, Glabinski A, Chernosky A, Krivacic K, Peterson J, et al. Chemokines and chemokine receptors in model neurological pathologies: molecular and immunocytochemical approaches.
Methods Enzymol
 
1997
;
287
:
319
–48.
Sallusto F, Lenig D, Mackay CR, Lanzavecchia A. Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes.
J Exp Med
 
1998
;
187
:
875
–83.
Schmidt B, Toyka KV, Kiefer R, Full J, Hartung H‐P, Pollard J. Inflammatory infiltrates in sural nerve biopsies in Guillain‐Barré syndrome and chronic inflammatory demyelinating neuropathy.
Muscle Nerve
 
1996
;
19
:
474
–87.
Simpson JE, Newcombe J, Cuzner ML, Woodroofe MN. Expression of the interferon‐gamma‐inducible chemokines IP‐10 and Mig and their receptor, CXCR3, in multiple sclerosis lesions.
Neuropathol Appl Neurobiol
 
2000
;
26
:
133
–42.
Sørensen TL, Tani M, Jensen J, Pierce V, Lucchinetti C, Folcik VA, et al. Expression of specific chemokines and chemokine receptors in the central nervous system of multiple sclerosis patients.
J Clin Invest
 
1999
;
103
:
807
–15.
Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm.
Cell
 
1994
;
76
:
301
–14.
Tani M, Ransohoff RM. Do chemokines mediate inflammatory cell invasion of the central nervous system parenchyma? [Review].
Brain Pathol
 
1994
;
4
:
135
–43.
Tani M, Glabinski AR, Tuohy VK, Stoler MH, Estes ML, Ransohoff RM. In situ hybridization analysis of glial fibrillary acidic protein mRNA reveals evidence of biphasic astrocyte activation during acute experimental autoimmune encephalomyelitis.
Am J Pathol
 
1996
;
148
:
889
–96.
Trebst C, Sørensen TL, Kivisàkk P, Cathcart MK, Hesselgessen J, Horuk R, et al. CCR1+/CCR5+mononuclear phagocytes accumulate in the central nervous system of patients with multiple sclerosis. Am J Pathol
2001
;
159
:
1701
–10.
Toyka KV. Eighty three years of the Guillain‐Barré syndrome: clinical and immunopathologic aspects, current and future treatments. [Review].
Rev Neurol (Paris)
 
1999
;
155
:
849
–56.
vander Voorn P, Tekstra J, Beelen RH, Tensen CP, van der Valk P, de Groot CJ. Expression of MCP‐1 by reactive astrocytes in demyelinating multiple sclerosis lesions.
Am J Pathol
 
1999
;
154
:
45
–51.
Ward SG, Bacon K, Westwick J. Chemokines and T lymphocytes: more than an attraction. [Review].
Immunity
 
1998
;
9
:
1
–11.
Winer JB, Hughes RA, Osmond C. A prospective study of acute idiopathic neuropathy. I. Clinical features and their prognostic value.
J Neurol Neurosurg Psychiatry
 
1988
;
51
:
605
–12.
Zou LP, Pelidou SH, Abbas N, Deretzi G, Mix E, Schaltzbeerg M, et al. Dynamics of production of MIP‐1alpha, MCP‐1 and MIP‐2 and potential role of neutralization of these chemokines in the regulation of immune responses during experimental autoimmune neuritis in Lewis rats.
J Neuroimmunol
 
1999
;
98
:
168
–75.