Journal of Neuroimmunology 97 1999.

5159

Chemokine mRNA expression in the cauda equina of Lewis rats with experimental allergic neuritis
Toshiki Fujioka 1, Enkhmaa Purev, Abdolmohamad Rostami
Received 18 November 1998; received in revised form 23 February 1999; accepted 8 March 1999 ) Department of Neurology, Uniersity of Pennsylania School of Medicine, 3400 Spruce Street, Philadelphia, PA 19104-4283, USA

Abstract Chemokines play an important role in the migration of leukocytes to inflammatory sites. In this study, using the quantitative competitive reverse transcriptase PCR method, we analyzed sequential expression of certain chemokine mRNAs in the cauda equina CE. of rats with experimental allergic neuritis EAN.. Interferon-g-inducible protein IP.-10, monocyte chemotactic protein MCP.-1, macrophage inflammatory protein MIP.-1a , the regulated upon activation normal T cell expressed and secreted chemokine RANTES., and lymphotactin were analyzed on days 0 pre-immunization., 7 preclinical stage., 10 disease onset., 13 clinical progression., 17 disease peak., as well as on days 20, 24, and 34 post-immunization p.i.. recovery.. MCP-1 message increased at the preclinical stage and peaked at day 17 p.i. The increase in the early stage was not detected in other tissues, indicating peripheral nerve-specific upregulation. MIP-1a and IP-10 messages surged at day 13, then returned to low in the recovery stage. RANTES message increased at day 13 and peaked at day 17 p.i.; however, unlike other chemokines, it showed a second peak of expression on day 24. Lymphotactin message was undetectable at any time point. MCP-1 protein was detected immunohistologically in endothelial cells at day 7 p.i. The sequential expression of these chemokines in relation to the inflammatory process in the nerve leading to demyelination is discussed. q 1999 Elsevier Science B.V. All rights reserved.
Keywords: Experimental allergic neuritis EAN.; Lewis rat; a-Chemokine; b-Chemokine; Messenger RNA; Polymerase chain reaction PCR.

1. Introduction Chemokines, a family of small proteins sharing aminoacid sequence homology, are implicated in the development of leukocyte migration from the blood stream to the site of inflammation Schall, 1994; Proost et al., 1996; Baggiolini, 1998; Luster, 1998. and activation of leukocytes Murphy et al., 1996; Taub et al., 1996.. Chemokines are classified into four subfamilies, CXC a ., CCb ., and C g ., and CX 3 C chemokines, according to the presence and positioning of several cysteine residues, and are one of the major products of activated lymphocytes Hedrick and Zlotnik, 1996; Baggiolini, 1998; Luster, 1998.. However, many other cell lineages besides immune system cells are

Corresponding author. Tel.: q1-215-8983253; Fax: q1-215-5732107 Present address: The Fourth Department of Internal Medicine, Toho University School of Medicine, 2-17-6 Ohashi, Meguro-ku, Tokyo 1538515, Japan.
1

known to be capable of producing chemokines Baggiolini et al., 1994; Vaddi et al., 1997.. Chemokines require specific receptors on their target cells Baggiolini et al., 1994; Murphy, 1994.. Because a particular chemokine receptor is expressed on a particular cell type, and in particular conditions Premack and Schall, 1996; Baggiolini, 1998; Baggiolini et al., 1998; Luster, 1998; Ward and Westeick, 1998., specific chemokines can selectively attract special cell types from the blood stream to the inflammation site. Understanding how the chemokine network functions will help in elucidating the mechanism of inflammatory disorders and in developing specific treatment for them. During autoimmune inflammation of the human peripheral nervous system PNS. as with GuillainBarre syn drome GBS., the most common acquired demyelinating neuropathy, or even in non-inflammatory PNS diseases like hereditary polyneuropathy, ischemic neuropathy or axonal degeneration after sciatic nerve transection, intense infiltration of mononuclear cells, i.e., lymphocytes, mono-

0165-5728r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 5 7 2 8 9 9 . 0 0 0 4 8 - X

52

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159

cytesrmacrophages, has been observed in peripheral nerves Asbury et al., 1969; Cornblath et al., 1990; Venezie et al., 1995; Schmidt et al., 1996.. Since the ability of Schwann cells SC. to produce one of the b-chemokines, monocyte chemotactic protein MCP.-1, in vitro has recently been shown Polydefkis et al., 1998., and cell lineages other than SC found in the PNS such as fibroblasts and endothelia are a well-known cell source for chemokines, there is no doubt that these infiltrating cells are attracted and activated by some members of the chemokine family produced in the PNS. In the CNS of patients with multiple sclerosis MS. or in the CNS of experimental animals with experimental allergic encephalomyelitis EAE., the production of the a-chemokine interferon-g inducible protein IP.-10 and the b-chemokines MCP-1, macrophage inflammatory protein MIP.-1a , and the regulated upon activation normal T cell expressed and secreted chemokine RANTES. by infiltrating mononuclear cells and residential CNS cells like astrocytes, microglia, and endothelia has been demonstrated Godiska et al., 1995; Eng et al., 1996; Ransohoff et al., 1996; Glabinski et al., 1997; Simpson et al., 1998.. In addition, neutralization of MIP-1a prevents EAE, indicating the involvement of chemokines in the development of CNS demyelination Karpus et al., 1995.. The PNS is comprised of relatively fewer cell elements compared to the CNS Thomas et al., 1993. and the chemokine profile of inflammatory nerve lesions may not be similar to that of CNS inflammatory lesions. To date, chemokine expression in the PNS during inflammatory demyelination has not been well-characterized. Experimental allergic neuritis EAN. is an animal model of GBS and is considered to be a PNS version of EAE Rostami et al., 1985; Rostami, 1997.. Since EAN is a CD4q T cell Th1 cell.-mediated disease and the majority of infiltrating cells in the PNS during EAN are macrophages and lymphocytes Rosen et al., 1992; Zettl et al., 1996., the kinetics of chemokines which can attract mainly mononuclear cells are of particular interest. In the present study, we sequentially examined mRNA expression of mononuclear cell-attracting chemokines, achemokine IP-10, b-chemokines MCP-1, MIP-1a , and RANTES, and g-chemokine lymphotactin Ltn. Kelner et al., 1994; Hedrick et al., 1997. in the cauda equina of Lewis rats with EAN by quantitative competitive reverse transcriptase PCR Q-PCR. and demonstrated in-situ localization of MCP-1 protein using immunohistochemistry.

tide SP26 corresponding to the 5378 amino-acid sequence of the bovine myelin P2 protein supplied by Dr. J.D. Lambris; Protein Chemistry Lab., Department of Pathology and Laboratory Medicine, University of Pennsylvania, PA. was injected into the hind foot pads subcutaneously with 100 ml of phosphate buffered saline and 100 ml of CFA per rat Rostami et al., 1990.. Rats were examined daily for clinical signs of EAN and body weight. Clinical signs of EAN were scored as: 0 s normal, 1 s flaccid tail, 2 s previous sign plus inability to spread toes, 3 s paraplegia, 4 s quadriplegia Rostami et al., 1990.. 2.2. Tissue preparation Rats were sacrificed at days 0, 7, 10, 13, 17, 20, 24, and 34 four rats per group. post-immunization p.i... Rats were anesthetized with ether and perfused with ice-cooled 0.1 M phosphate buffer saline containing 0.2 unitsrml heparin via left ventricle. Quickly harvested cauda equina, from which the fragments of spinal cords had been carefully excluded, were rapidly frozen in isopentane precooled by liquid nitrogen, stored in y708C. 2.3. Quantitation of mRNA Total RNA was extracted by single step preparation using TRIzol reagent Gibco-BRL, Gaithersburg, MD. following the manufacturers instructions. A total of 2 mg of total RNA from each animal was reverse transcribed in 40 ml of first strand Buffer Gibco-BRL. containing a final concentration of 1.0 mM each of dATP, dCTP, dGTP, and dTTP Pharmacia, Piscataway, NJ., 5 mM of oligo dT. primers Promega, Madison, WI., 10 mM of MgCl 2 , 0.8 unitsrml of RNase inhibitor RNasin; Promega., 1 mM of DTT and 3 unitsrml of Molony Murine Leukemia Virus reverse transcriptase Gibco-BRL.. Replicated samples first strand cDNA equivalent of 10 ng of total RNA. from each animal were amplified individually with titrated amounts of the competitor as described previously Fujioka et al., 1998.. A synthetic competitor, gene plasmid containing oligonucleotide priming site for rat cytokines or housekeeping genes, for measuring b-actin message was kindly supplied by Dr. H.-D. Volk, Berlin, Germany Siegling et al., 1994.. The competitors for chemokines were constructed using PCR MIMICe construction kit Clontech, Palo Alto, CA.. Amplification of the synthetic competitor with oligonucleotide primers from the cytokine genes or housekeeping genes results in a product of a slightly different length than native cDNA, which can easily be resolved by agarose gel electrophoresis. Each gene message was amplified from sample cDNA in 50 ml of PCR Buffer Perkin-Elmer, Foster City, CA. containing 0.5 unit of Taq DNA polymerase Ampli-Taq Gold; Perkin-Elmer., 2.0 mM of MgCl 2 , 0.2 mM of each d-NTPs, and 0.4 mM of each primer sets using Perkin-

2. Materials and methods 2.1. Induction of EAN in Lewis rats Seven-week-old female Lewis rats weighing 110130 g were purchased from Charles River Laboratories, Raleigh, NC.. A total of 150 mg of HPLC-purified synthetic pep-

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159

53

Elmer thermocycler 2400 following the manufacturers protocol. The cDNA was amplified through cycles of denaturing; 968C; 30 s, annealing; 638C b-actin., 658C MCP-1., 658C RANTES., 648C MIP1-a ., 678C IP-10., and 608C Ltn.; 35 s, and extension; 728C; 45 s. Rat b-actin-specific primer set yielded 762-bp product from the sample cDNA, and 601-bp product from the competitor, respectively Siegling et al., 1994.. Rat MCP-1-specific sense 5X-ATGCAGGTCTCTGTCACGCTTCTGG-3X . and antisense 5X-CAGAAGTGCTTGAGGTGGTTGTGGA-3X . primers yielded 418-bp product from the sample cDNA, and 350-bp product from the competitor, respectively Yoshimura et al., 1991.. Rat RANTES-specific sense 5X-ATGAAGATCTCTGCAGCTGCATCCC-3X . and antisense 5X-CCACTTCTTCTCTGGGTTGGCACAC-3X . primers yielded 243-bp product from the sample cDNA, and 376-bp product from the competitor, respectively GeneBank accession numbers U06436.. Rat MIP1aspecific sense 5X-CTGCCCTTGCTGTTCTTCTCTGCAC3X . and antisense 5X-GGCATTCAGTTCCAGCTCAGTGATG-3X . primers yielded 260-bp product from the sample cDNA, and 426-bp product from the competitor, respectively Shanley et al., 1995.. Rat IP-10-specific sense 5X-AGTGCTGCTGTCGTTCTCTGCCTCGTGCTG-3X . and antisense 5X-GGGCATGGCACATGCTGAAGAGATTAGTAC-3X . primers yielded 600-bp product from the sample cDNA, and 446-bp product from the competitor, respectively Wang et al., 1996.. Rat Ltn-specific sense 5X-GTCTGCTGCTTCGCCGCATGGGTTG-3X . and antisense 5X-TTACCCAGTCAGGGTTACTGCTGTG-3X . primers yielded 315-bp product from the sample cDNA, and 250-bp product from the competitor, respectively GeneBank accession numbers U23377.. The number of PCR cycles was changed depending on the samples: bactin s 32, MCP-1s 38, RANTESs 40, MIP-1a s 45, IP-10 s 40, and Ltn s 45. The fluorescent intensity of each amplicon on agarose gel containing ethidium bromide was measured using an image analysis system Gel Doc 1000; BIO-RAD, Hercules, CA. conducted by Molecular Analyst Ver. 1.4 BIO-RAD., and the input concentration of sample cDNA was determined using the ratio of fluorescent intensity between the competitor and the cDNA products. Chemokine message amounts were expressed as a molar ratio of b-actin message in the same samples in order to compensate for the varying efficiency of PCR amplification of each target message. All experiments were performed twice and the results were identical. 2.4. Immunohistochemistry for MCP-1 We selected MCP-1 to demonstrate intraneural production of chemokine protein since anti-rat MCP-1 antibody was the only available antibody against rat chemokine. CE of EAN rats removed at days 7 and 13 were frozen in isopentane pre-cooled by liquid nitrogen. Cryosections 56 mm thick. were fixed in ice-cooled 2% formaldehyde

in PBS for 20 min at room temperature. After blocking endogenous peroxidase activity by 0.5% hydrogen peroxide, sections were permeabilized by Earles balanced salt solutions EBSS; Gibco-BRL. containing 0.1% wrv. saponin Sigma, St. Louis, MO.. Sections were then incubated in EBSS containing 2% normal goat serumr0.1% saponin, then rabbit anti-rat MCP-1 antibody Serotec, Oxford, England. was added. First antibody was detected by biotinylated goat anti-rabbit antibody Pharmingen, San Diego, CA. followed by standard avidinbiotin complex peroxidase staining kit Pierce, Rockford, IL.. Diaminobenzidine Pierce. was used as chromogen with hematoxylin counterstain. For negative control, first Ab was excluded. Serial sections were also stained by standard immunostaining protocol with mAbs against rat mononuclear phagocyte mouse IgG1 , k: 1C7; Pharmingen. which recognizes ED1-like antigen, rabbit polyclonal Ab against S-100 protein Serotec., and rabbit polyclonal Ab against human von Willebrand factor DAKO, Carpinteria, CA.. 2.5. Statistical analysis Differences among the amounts of the chemokine messages at the different time points were evaluated by MannWhitneys U-test. Values of p - 0.05 were considered as significant.

3. Results 3.1. Induction of EAN The rats synchronously developed ascending paralysis that started from the tail at day 11 p.i. The maximal

Fig. 1. Clinical course of experimental allergic neuritis in Lewis rats. Bars indicate the standard deviations. Score 0 s normal. Score 1s flaccid tail. Score 2 s previous sign plus inability to spread toes. Score 3s paraplegia. Score 4 s quadriplegia. ns 28 from day 1 to 7 p.i.., 24 from day 8 to 10 p.i.., 20 from day 11 to 13 p.i.., 16 from day 14 to 17 p.i.., 12 from day 18 to 20 p.i.., 8 from day 21 to 24 p.i.., 4 from day 24 to 34 p.i.., respectively.

54

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159

clinical score was 2.84 " 0.96 Mean " S.D., n s 16. at day 16 p.i. Recovery started at day 17 p.i. and continued through day 28 p.i. 0.63 " 0.63, n s 4.. Thereafter, the clinical signs became stable Fig. 1.. 3.2. Expression of chemokine mRNA in cauda equina 3.2.1. Expression of MCP-1 (Fig. 2) The increase in MCP-1 message started at day 7 p.i. It increased steadily during days 10 to 17 p.i., peaked at day 17 p.i., then abruptly returned to baseline level and stayed at a very low level through day 34 p.i. Fig. 2A.. The increase in MCP-1 message during days 7 to 17 p.i. was statistically significant p - 0.05 by MannWhitneys Utest, compared to day 0 p.i... Moreover, the increase in MCP-1 message at day 17 p.i. was significant compared with day 7 p.i. p - 0.05 by MannWhitneys U-test.. To exclude the possibility that the increase in MCP-1 message in the cauda equina during the preclinical stage days 7 and 10 p.i.. was not a tissue-specific increase, we quantitated the MCP-1 message in the various tissues

Fig. 3. Sequential expression of the regulated upon activation normal T cell expressed and secreted chemokine RANTES. mRNA in the cauda equina of Lewis rats during EAN. Bars indicate the standard errors. RANTES message increased significantly on days 13, 17, and 24 p.i. U 0.01- p- 0.05; compared to normal control..

kidneys, lungs, and livers, of the same EAN rats Fig. 2B.. No significant change was found in MCP-1 message in

Fig. 2. mRNA expression of MCP-1 during EAN. A. Sequential expression of MCP-1 mRNA in the cauda equina of Lewis rats during EAN. Bars indicate the standard errors. MCP-1 message increased on days 7, 10, 13, and 17 p.i. significantly U 0.01- p- 0.05; compared to normal control.. B. The comparison of MCP-1 messages between cauda equina and other tissues during the preclinical stage of EAN. The message of MCP-1 in liver, kidney, and lung maintained a low level which was almost identical to the normal level.

Fig. 4. Sequential expression of interferon-g IP-10 A. and MIP-1a mRNA B. in the cauda equina of Lewis rats during EAN. Bars indicate the standard errors. Both IP-10 message and MIP-1a message increased significantly on days 13, 17, and 20 p.i. U 0.01- p- 0.05; compared to normal control..

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159

55

Fig. 5 continued..

these tissues at days 7 and 10 p.i. compared to the normal rats, indicating that the increase in MCP-1 message in the cauda equina shown on days 7 and 10 p.i. was specific to peripheral nerve tissues. 3.2.2. Expression of RANTES The significant increase in RANTES mRNA expression was observed at days 13, 17, and 24 p.i. compared to day 0 p.i. p - 0.05 by MannWhitneys U-test.. The RANTES messages at days 20 and 34 p.i. returned to the baseline Fig. 3.. 3.2.3. Expression of MIP-1a and IP-10 These messages remained at a low level during days 0 to 10 p.i., surged at day 13 p.i., then abruptly returned to a low level at days 17 and 20 p.i., although still higher than the normal level p - 0.05 by MannWhitneys U-test.. After day 24 p.i. they returned to normal level Fig. 4.. 3.2.4. Expression of Ltn Ltn message could not be detected in the cauda equina at any time points, whereas it was detected in EAN spleen data not shown.. 3.3. Immunohistochemistry (Fig. 5) At day 7 p.i., MCP-1 was detected in the endothelial cells although mononuclear phagocytes were hardly seen

Fig. 5. Immunohistochemical detection of MCP-1 in cauda equina of EAN rats at day 13 p.i. A,B,C. and at day 7 p.i. D,E.. Serial cryosections were stained for MCP-1 A,D., von Willebrand factor endothelial cells.; E., mononuclear phagocytes ED1-like antigen.; B., and Schwann cells S-100 protein.; C.. Visualized by diaminobenzidine followed by hematoxtyline counterstain. Barss 20 mm. Two foci of perivascular cell infiltration are noted. MCP-1 positive cells in A. were mostly endothelial cells, which are negative for both ED1-like antigen in B. and S-100 in C. indicated by curved arrows. or macrophages, which are positive for ED1-like antigen in B. and negative for S-100 in C. indicated by large arrow heads.. Many macrophages found in the same perivascular lesions are negative for MCP-1 indicated by small arrow heads.. Some Schwann cells which are positive for S-100 in C. and negative for ED1-like antigen shown in B. are found as MCP-1 positive indicated by straight arrows.. On the other hand, at day 7 p.i., MCP-1 positive cells in D. are mostly endothelial cells which are positive for von Willebrand factor in E. indicated by arrows..

56

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159

Fig. 5D,E.. At day 13 p.i., MCP-1 was detected in perivascular lesions. Serial sections stained for mononuclear phagocytes demonstrated that these lesions were occupied principally by macrophages; however, not all of the macrophages expressed MCP-1. Endothelial cells in the perivascular lesions also exhibited MCP-1 immunoreactivity. Some SCs surrounding mononuclear cell infiltrations were also found positive for MCP-1 immunoreactivity. No signal for MCP-1 was found in the lesion without cell infiltration. Immunohistochemistry for other chemokines was not performed because antibodies against rat chemokines other than MCP-1 were not available.

4. Discussion In this study, sequential expression of messages of one a-chemokine, IP-10, of three major b-chemokines, MCP-1, MIP-1a , and RANTES, and of a g-chemokine, lymphotactin, in the cauda equina of EAN rats was examined. MCP-1, MIP-1a , RANTES and IP-10 were upregulated during the active stage of disease. Several unique findings could be emphasized from our results: 1. an early upregulation of MCP-1 message in endothelial cells; 2. biphasic upregulation of RANTES message in both the active and recovery stages of disease; and 3. SC, endothelial cells and macrophages as the cell source for MCP-1 in the PNS. MCP-1 message started increasing before the clinical onset of EAN, whereas the other chemokine messages increased after clinical onset. The MCP-1 expression in kidney, liver, or lung at the same period remained at normal level, suggesting the tissue-specific upregulation of MCP-1 in the cauda equina. MCP-1 protein was detected immunohistologically only in endothelial cells at day 7 p.i. The peak expression of MCP-1 and RANTES was observed at day 17 p.i., coinciding with the clinical disease peak, while the expression of IP-10 and MIP-1a surged at day 13 p.i., followed by rapid decrease. During the clinical recovery stage of EAN, only MCP-1 message stayed at normal level, whereas the other chemokines remained at a relatively higher than normal level. In particular, a unique pattern in RANTES expression was notable, showing a distinct biphasic profile during recovery. It is still unclear which cells initiate the pathological process of EAN. In EAE, a CNS counterpart for EAN, encephalitogenic T cells produce IL-1b, IL-2, TNF-a, IFN-g upon stimulation with a specific antigen Sun et al., 1995.. Since MCP-1 can be produced by monocytesrmacrophages, endothelial cells, and glial cells after stimulation with IL-1b, TNF-a, LPS, etc. Baggiolini et al., 1994; Berman et al., 1996; Douglas et al., 1997. and the ability of SCs to produce MCP-1 in response to TNF-a and IFN-g has been demonstrated Polydefkis et al., 1998., similar mechanisms can be speculated in EAN. Moreover,

the increase in the number of IL-1b-producing cells has been observed in the peripheral nerves in the early stage of EAN Zhu et al., 1998.. Therefore, it is speculated that the early upregulation of the intraneural MCP-1 expression found in this study was produced by endothelial cells stimulated by infiltrating neuritogenic T cells within the cauda equina via proinflammatory cytokines. Since MCP-1 attracts monocytes and activated T cells via b-chemokine receptors CCR.-2 and 4 Baggiolini et al., 1994; Premack and Schall, 1996., this early upregulation of MCP-1 in endothelial cells appears very important for establishing an intense cell infiltration and subsequent paralysis. In this study, an upregulation of MIP-1a and IP-10 at day 13 p.i. was observed. In our previous report, it was shown that IFN-g message is upregulated during the disease-accelerating stage, whereas IL-10 message peaked at disease peak stage in EAN Fujioka et al., 1998.. These chemokines, especially IP-10, are known to be induced by IFN-g Luster and Ravetch, 1987.. IP-10 requires the a-chemokine receptor CXCR.-3 while MIP-1a requires CCR-1, 4, or 5 Premack and Schall, 1996; Baggiolini et al., 1998.. In humans, CXCR-3 and CCR-5 are known to be expressed on activated T cells, particularly Th1 cells Qin et al., 1998; Sallusto et al., 1998.. In addition to its chemoattracting role, IP-10 can upregulate MHC-II expression on astrocytes in an autocrine fashion Vanguri et al., 1996.. Therefore, together with the upregulation of IFN-g, the upregulation of these chemokines IP-10 and MIP-1a . at day 13 p.i. may indicate an accumulation of Th1 cells and an enhancement of Th1-mediated immune response in the peripheral nerves of the rat EAN model, although a difference in species may cause a difference in immune response. After their peak expression, both IP-10 and MIP-1a messages declined quickly, and remained at a relatively low level at days 17 and 20 p.i., although still significantly high compared to the normal level. This profile is similar to the IFN-g profile Fujioka et al., 1998., suggesting the influence of this cytokine on the expression of IP-10 and MIP-1a. The significance of this low level upregulation for these chemokines is still unknown; however, it can be postulated that these chemokines play some role in assisting remyelination since MIP-1a can stimulate SC proliferation in vitro Khan and Wigley, 1994.. In contrast to the chemokines mentioned above, RANTES exhibited two distinct peaks of expression: at days 1317 p.i. and 24 p.i. Between these two peaks, the earlier one was simultaneous with the peak expression of IL-10 Fujioka et al., 1998. and MCP-1 in the cauda equina. Again, the chemotactic roles for MCP-1 and RANTES are mediated by their specific receptors: CCR1, 3, 4 and 5 for RANTES and CCR2 and 4 for MCP-1. CCR3 and CCR4 are expressed on activated Th2 cells, whereas CCR5 is expressed on Th1 cells Sallusto et al., 1998.. Moreover, MCP-1 is involved in both Th1 response in an established Th1 clone Murphy et al., 1996; Taub et

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159

57

al., 1996. and Th2 response in nave T cells in vitro Chensue et al., 1996; Karpus et al., 1997.. Thus, RANTES and MCP-1 can promote both Th1 and Th2 response. Furthermore, the inhibitory activity of MCP-1 against Th1 cytokine production has been found in intestinal mucosa, Payers patch, and mesenteric lymphnodes Karpus et al., 1998.. It is possible that MCP-1 can inhibit Th1 response at the inflammatory site. It is probable that MCP-1 first promotes a greater Th1 response, then afterwards, a greater Th2 response, which is reflected by the increase in IL-10 expression in the cauda equina Fujioka et al., 1998.. This upregulated IL-10 also suppresses Th1 response and finally suppresses MCP-1 production Kucharzik et al., 1998. in the next step, resulting in recovery from EAN. The same sequential modulation system can be postulated for RANTES. As mentioned above, two of four CCRs for RANTES are expressed on Th2 cells. Since EAN is a Th1-mediated disease, the main subpopulation of activated T cells in the early stage are Th1 cells. These cells can be attracted by not only RANTES but also MCP-1, MIP-1a , and IP-10; however, as the disease progresses, Th2 cells become activated and sensitive to chemoattractants. Probably, the increase in RANTES and MCP-1 at the disease peak stage is beneficial for the recruitment of Th2 cells at this time point. The significance of the transient increase in RANTES expression at day 24 p.i. is unclear. At this time point, inflammatory cell recruitment is complete and clinical recovery is obvious. Recently, it has been shown that a migration and a differentiation of dorsal root ganglia cells in vitro were stimulated by RANTES as well as by nerve growth factor Bolin et al., 1998.. This fact suggest that the transient upregulation of RANTES during the recovery stage of EAN may reflect active axonal regeneration after axonal degeneration caused by severe demyelination. The cell source for this RANTES message upregulation should be elucidated for a better understanding of peripheral nerve regeneration. Finally, we examined the intraneural expression of lymphotactin, a member of the recently discovered new chemokine family, g-chemokine. Since lymphotactin expression is rapidly upregulated upon stimulation and it attracts exclusively lymphocytes Hedrick and Zlotnik, 1997., the role for lymphotactin in the early stage of EAN was expected. However, despite the detection of other chemokines and the detection of lymphotactin in the spleen of the same animals, this chemokine was not detectable in the cauda equina at any time points. Lymphotactin is produced by activated CD8q T cells, CD4 CD8 double negative ab T cells Kelner et al., 1994., CD4q NKq T cells, NK cells Hedrick et al., 1997., and gd T cells Boismenu et al., 1996., which are not thought to be a majority of peripheral nerve-infiltrating cells in EAN. Moreover, the predominant infiltrating cells during EAN are monocytesrmacrophages, not lymphocytes Rosen et al., 1992; Zettl et al., 1996.. Thus, the role for lympho-

tactin within the peripheral nerves during EAN may be negligible. In conclusion, intraneural mRNA expression of several chemokines which attract monocytes and lymphocytes was studied during EAN by Q-PCR. MCP-1 message preceded clinical disease and other chemokine messages, suggesting a disease-initiating role for this chemokine. Immunohistochemistry revealed that SCs and endothelial cells are the main source of MCP-1 in the preclinical stage. The expression of MIP-1a and IP-10 message correlated with the temporal profile of IFN-g. MCP-1 and RANTES messages reached their peak expression in concert with the disease peak, later than IP-10 and MIP-1a. A close relationship of both IP-10 and MIP-1a with Th1 response was suggested. MCP-1 and RANTES seemed to have a broader connection to both Th1 and Th2 response; moreover, an active role for RANTES in recovery from EAN was suggested. For a better understanding of the chemokine milieu, we need further elucidation of the cellular sources for chemokines other than MCP-1, as well as the kinetics of chemokine receptor expression in the peripheral nerves.

Acknowledgements We thank Katherine Regan for her assistance in preparing the manuscript. This work was supported in part by NIH Grant NS 11037.

References
Asbury, A.K., Arnason, B.G., Adams, R.D., 1969. The inflammatory lesion in idiopathic polyneuritis. Its role in pathogenesis. Medicine 48, 173215. Baggiolini, M., 1998. Chemokines and leukocyte traffic. Nature 392, 565568. Baggiolini, M., Dewald, B., Moser, B., 1994. Interleukin-8 and related chemotactic cytokinesCXC and CC chemokines. Adv. Immunol. 55, 97179. Baggiolini, M., Dewald, B., Moser, B., 1998. Human chemokines: an update. Annu. Rev. Immunol. 15, 675705. Berman, J.W., Guida, M.P., Warren, J., Armat, J., Brosnan, C.F., 1996. Localization of monocyte chemoattractant peptide-1 expression in the central nervous system in experimental autoimmune encephalomyelitis and trauma in the rat. J. Immunol. 156, 30173023. Boismenu, R., Feng, L., Xia, Y.Y., Chang, J.C.C., Havran, W.L., 1996. Chemokine expression by intraepithelial gr d T cells. Implication for the recruitment of inflammatory cells to damaged epithelia. J. Immunol. 157, 985992. Bolin, L.M., Murray, R., Lukacs, N.W., Strieter, R.M., Kunkel, S.L., Schall, T.J., Bacon, K.B., 1998. Primary sensory neurons migrate in response to the chemokines RANTES. J. Neuroimmunol. 81, 4957. Chensue, S.W., Warmington, K.S., Ruth, J.H., Sanghi, P.S., Lincoln, P., Kunkel, S.L., 1996. Role of monocyte chemoattractant protein-1 MCP-1. in Th1 Mycobacterial. and Th2 Schistosomal. antigen-induced granuloma formation. Relationship to local inflammation, Th cell expression, and IL-12 production. J. Immunol. 157, 46024608.

58

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159 Proost, P., Wuyts, A., van Damme, J., 1996. The role of chemokines in inflammation. Int. J. Clin. Lab. Res. 26, 211223. Qin, S., Rottman, J.B., Myers, P., Kassam, N., Weinblatt, M., Loetscher, M., Koch, A.E., Moser, B., Mackay, C.R., 1998. The chemokine receptors CXCR3 and CCR5 mark subset of T cells associated with certain inflammatory reactions. J. Clin. Invest. 101, 746754. Ransohoff, R.M., Glabinski, A., Tani, M., 1996. Chemokines in immune-mediated inflammation of the central nervous system. Cytokine Growth Factor Rev. 7, 3546. Rosen, J.L., Brown, M.J., Rostami, A., 1992. Evolution of the cellular response in P2-induced experimental allergic neuritis. Pathobiology 60, 108112. Rostami, A., 1997. P2 -reactive T cells in inflammatory demyelination of the peripheral nerve. J. Inf. Dis. 176, S160S163. Rostami, A., Burns, J.B., Brown, M.J., Rosen, J., Zweiman, B., Lisak, R.P., Pleasure, D.E., 1985. Transfer of experimental allergic neuritis with P2-reactive T-cell lines. Cell. Immunol. 91, 354361. Rostami, A., Gregorian, S.K., Brown, M.J., Pleasure, D.E., 1990. Induction of severe experimental autoimmune neuritis with a synthetic peptide corresponding to the 5378 amino acid sequence of the myelin P2 protein. J. Neuroimmunol. 30, 145151. Sallusto, F., Lenig, D., Mackay, C.R., Lanzavecchia, A., 1998. Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes. J. Exp. Med. 187, 875883. Schall, T.J., 1994. In: Thomson, A.W. Ed.., The Cytokine Handbook. Academic Press, London, pp. 419460. Schmidt, B., Toyka, K.V., Kiefer, R., Full, J., Hartung, H.P., Pollard, J., 1996. Inflammatory infiltrates in sural nerve biopsies in Guillain Barre syndrome and chronic inflammatory demyelinating neuropathy. Muscle and Nerve 19, 474487. Shanley, T.P., Schmal, H., Friedl, H.P., Jones, M.L., Ward, P.A., 1995. Role of macrophage inflammatory protein-1 alpha MIP-1 alpha. in acute lung injury in rats. J. Immunol. 154, 47934802. Siegling, A., Lehmann, M., Platzer, C., Emmrich, F., Volk, H.-D., 1994. A novel multispecific competitor fragment for quantitative PCR analysis of cytokine gene expression in rats. J. Immunol. Meth. 177, 2328. Simpson, J.E., Newcombe, J., Cuzner, M.L., Woodroofe, M.N., 1998. Expression of monocyte chemoattractant protein-1 and other b-chemokines by resident glia and inflammatory cells in multiple sclerosis lesions. J. Neuroimmunol. 84, 238249. Sun, D., Hu, X.Z., Shah, R., Coleclough, C., 1995. The pattern of cytokine gene expression induced in rat T cells specific for myelin basic protein depends on the type and quality of antigenic stimulus. Cell. Immunol. 166, 18. Taub, D.D., Turcovski-Corrales, S.M., Key, M.L., Longo, D.L., Murphy, W.J., 1996. Chemokines and T lymphocyte activation: I. Beta chemokines costimulate human T lymphocyte activation in vitro. J. Immunol. 156, 20952103. Thomas, P.K., Berthold, C.-H., Ochoa, J., 1993. In: Dyck, P.J., Thomas, P.K., Griffin, J.W., Low, P.A., Poduslo, J.F. Eds.., Peripheral Neuropathy, Vol. 1. Saunders, Philadelphia, pp. 2873. Vaddi, K., Keller, M., Newton, R.C., 1997. The Chemokine Factsbook. Academic Press, San Diego. Vanguri, P., Cho, S.Y., Chi, C.M., 1996. Role of muIP-10 in interferongamma induction of Ia in rat astrocytes. Mol. Immunol. 33, 1079 1087. Venezie, R.D., Toews, A.D., Morell, P., 1995. Macrophage recruitment in different models of nerve injury: lysozyme as a marker for active phagocytosis. J. Neurosci. Res. 40, 99107. Wang, X., Yue, T.-L., Ohlstein, E.H., Sung, C., Feuerstein, G.Z., 1996. Interferon-inducible protein-10 involves vascular smooth muscle cell migration, proliferation, and inflammatory response. J. Biol. Chem. 271, 2428624293. Ward, S.G., Westeick, J., 1998. Chemokines; understanding their role in T-lymphocyte biology. Biochem. J. 333, 457470. Yoshimura, T., Takeya, M., Takahashi, K., 1991. Molecular cloning of

Cornblath, D.R., Griffin, D.E., Welch, D., Griffin, J.W., McArthur, J.C., 1990. Quantitative analysis of endoneurial T-cells in human sural nerve biopsies. J. Neuroimmunol. 26, 113118. Douglas, M.S., Ali, S., Rix, D.A., Zhang, J.G., Kirby, J.A., 1997. Endothelial production of MCP-1: modulation by heparin and consequences for mononuclear cell activation. Immunology 92, 512518. Eng, L.F., Ghirnikar, R.S., Lee, Y.L., 1996. Inflammation in EAE: role of chemokinercytokine expression by resident and infiltrating cells. Neurochem. Res. 21, 511525. Fujioka, T., Jimi, T., Hilliard, B.A., Ventura, E.S., Rostami, A., 1998. The expression of cytokine mRNA in the cauda equina of Lewis rats with experimental allergic neuritis. J. Neuroimmunol. 84, 223229. Glabinski, A.R., Tani, M., Strieter, R.M., Tuohy, V.K., Ransohoff, R.M., 1997. Synchronous synthesis of alpha- and beta-chemokines by cells of diverse lineage in the central nervous system of mice with relapses of chronic experimental autoimmune encephalomyelitis. Am. J. Pathol. 150, 617630. Godiska, R., Chantry, D., Dietsch, G.N., Gray, P.W., 1995. Chemokine expression in murine experimental allergic encephalomyelitis. J. Neuroimmunol. 58, 167176. Hedrick, J.A., Zlotnik, A., 1996. Chemokines and lymphocyte biology. Curr. Opin. Immunol. 8, 343347. Hedrick, J.A., Zlotnik, A., 1997. In: Horuk, R. Ed.., Chemokines, Vol. 287. Academic Press, New York, pp. 206215. Hedrick, J.A., Saylor, V., Figueroa, D., Mizoue, L., Yuming, X., Menon, A., Abrams, J., Handel, T., Zlotnik, A., 1997. Lymphotactin is produced by NK cells and attracts both NK cells and T cells in vivo. J. Immunol. 158, 15331540. Karpus, W.J., Lukacs, N.W., McRae, B.L., Strieter, R.M., Kunkel, S.L., Miller, S.D., 1995. An important role for the chemokine macrophage inflammatory protein-1 alpha in the pathogenesis of the T cell-mediated autoimmune disease, experimental autoimmune encephalomyelitis. J. Immunol. 155, 50035010. Karpus, W.J., Lukacs, N.W., Kennedy, K.J., Smith, W.S., Hurst, S.D., Barrett, T.A., 1997. Differential CC chemokine-induced enhancement of T helper cell cytokine production. J. Immunol. 158, 41294136. Karpus, W.J., Kennedy, K.J., Kunkel, S.L., Lukacs, N.W., 1998. Monocyte chemotactic protein 1 regulates oral tolerance induction by inhibition of T helper cell 1-related cytokines. J. Exp. Med. 187, 733741. Kelner, G.S., Kennedy, J., Bacon, K.B., Kleyensteuber, S., Largaespada, D.A., Jenkins, N.A., Copeland, N.G., Bazan, J.F., Moore, K.W., Schall, T.J., Zlotnik, A., 1994. Lymphotactin; a cytokine that represents a new class of chemokine. Science 266, 1395. Khan, S., Wigley, C., 1994. Different effects of a macrophage cytokine on proliferation in astrocytes and Schwann cells. Neuroreport 5, 13811385. Kucharzik, T., Lugering, N., Pauels, H.G., Domschke, W., Stoll, R., 1998. IL-4, IL-10 and IL-13 down-regulate monocyte-chemoattracting protein-1 MCP-1. production in activated intestinal epithelial cells. Clin. Exp. Immunol. 111, 152157. Luster, A.D., 1998. Chemokineschemotactic cytokines that mediate inflammation. N. Engl. J. Med. 338, 436445. Luster, A.D., Ravetch, J.V., 1987. Biochemical characterization of a g interferon-inducuble cytokine IP-10.. J. Exp. Med. 166, 10841097. Murphy, P.M., 1994. The molecular biology of leukocytes chemoattractant receptors. Annu. Rev. Immunol. 12, 593633. Murphy, W.J., Tian, Z.G., Asai, O., Funakoshi, S., Rotter, P., Henry, M., Strieter, R.M., Kunkel, S.L., Longo, D.L., Taub, D.D., 1996. Chemokines and T lymphocyte activation: II. Facilitation of human T cell trafficking in severe combined immunodeficiency mice. J. Immunol. 156, 21042111. Polydefkis, M.J., Krivacic, K., Tani, M., Ransohoff, R.M., Griffin, J.W., Ho, T.W., 1998. Rat Schwann cells upregulate MCP-1 synthesis in response to TNF alpha and IFN gamma. Neurology 50, A184. Premack, B.A., Schall, T.J., 1996. Chemokine receptors: gateways to inflammation and infection. Nat. Med. 2, 11741178.

T. Fujioka et al.r Journal of Neuroimmunology 97 (1999) 5159 rat monocyte chemoattractant protein-1 MCP-1. and its expression in rat spleen cells and tumor cell lines. Biochem. Biophys. Res. Commun. 174, 504509. Zettl, U.K., Gold, R., Toyka, K.V., Hartung, H.P., 1996. In situ demonstration of T cell activation and elimination in peripheral nervous

59

system during experimental autoimmune neuritis in the Lewis rat. Acta Neuropathol. 91, 360367. Zhu, J., Mix, E., Link, H., 1998. Cytokine production and the pathogenesis of experimental autoimmune neuritis and GuillainBarre syn drome. J. Neuroimmunol. 84, 4052.