Inhibitory effect of ferulic acid and isoferulic acid on the production of macrophage inflammatory protein-2 in response to respiratory syncytial virus infection in RAW264.7 cells.

We investigated the effect of ferulic acid (FA) and isoferulic acid (IFA), which are the main active components of the rhizoma of Cimicifuga heracleifolia (CH), an anti-inflammatory drug used frequently in Japanese traditional medicine, on the production of macrophage inflammatory protein-2 (MIR-2) in a murine macrophage cell line, RAW264.7, in response to respiratory syncytial virus (RSV) infection. Following the exposure of cells to RSV for 20h, the MIP-2 level in condition medium was increased to about 20 ng/ml, although this level in mock-infected cells was negligible. In the presence of either FA or IFA, RSV-infected cells reduced MIP-2 production in a dose-dependent manner. These data suggest that FA and IFA might be responsible, at least in part, for the anti-inflammatory drug effect of CH extract through the inhibition of MIP-2 production.


Introduction
Respiratory syncytial virus (RSV) is the major c ause of acute low e r respiratory tract illne sses such as bronchiolitis and pneumonia in infants and young childre n, and can c ause se ve re, e ve n fatal, infec tions in the e lderly. 1 Seve ral studies have indicate d that RSV is release d initially from upper resp iratory tract infe ctions and then re aches the bronchoalveolar region, w here viruses induce inflammation of the airw ay epithe lium ac companied by pe ribronchiolar infiltrations of neutrophils. 1,2 It is w ell know n that neutrophil ac cumulation is an important charac te ristic of inflammation and modulate s various inflammatory reactions . 3 Sinc e the initial disc overy of interle ukin-8 as a nove l neutrophil chemotactic c ytokine , 4 -6 it has sinc e bee n named a chemokine and bee n found in the conditioned medium (CM) of various c ells in response to stimulation w ith lipopolysac charide (LPS) as w ell as se ve ral virus infe ctions. We previously reporte d that RSV infe ction strongly induc es the production of macrophage inflammatory protein-2 (MIP-2), a murine counterpart of che mokine family, 7 in RAW 264.7 cells. 8 ,9 On the othe r hand, the rhizoma of Cim icifug a sp., such as Cim icifu ga h e ra c le ifo lia (CH) Komarov and Cim icifu ga da h u rica Max im are used frequently as antip yre tic, analge sic and anti-inflammatory drugs in Japanese traditional me dic ine. Fe rulic acid (FA), 3-(4-hydrox y-3-me thox yphenyl)-2-propenic ac id and its is omer isoferulic acid (IFA), 3-(3-hydrox y-4-me thox yphenyl)-2-propenic, have bee n re ce ntly rec ognized as the main active components of CH ex tract in the inflammation model in rats. 10 In the light of these findings, w e ex amine d w he the r FA and IFA show inhibitory effects on MIP-2 production in response to RSV infe ction.

Preparation of drugs
FA and IFA, purchased from Carl Roth GmbH (Karlsruhe, Ge rmany), w ere fre shly pre pare d in serum-free Dulbec co's modified Eagle's minimal e ssential me dium (DMEM) at a c once ntration of 5 mM. The dissolve d drugs w e re strilize d through a Millipore filter before use.

Cell
A murine mac rophage ce ll line , RAW264.7, w as obtained from Americ an Type Culture Colle ction and maintained in culture w ith DMEM supple mented w ith 10% fetal bovine serum (FBS), in a humidifie d atmosphe re containing 5% CO 2 at 37°C. He p-2 cells w ere c ulture d in MEM suppleme nted w ith 10% of FBS in the same manne r as above .
Virus RSV A2 strain, w hich w as kindly supplie d by Dr Watanabe (Rational Drug De sign Laboratories, Fukushima, Japan), w as propagate d in a conflue nt monolaye r of Hep-2 ce lls in a mainte nance me dium (MEM suppleme nte d w ith 2% FBS). 1 When the cytopathic effe ct reache d 80%, the cultures w ere proce sse d w ith three c ycles of fre ezing and thaw ing. After c entrifugation at 20003 g for 10 min, the culture supe rnatant w as c ollec te d and store d in a small portion at -84°C as a virus stock solution. Virus titers w ere de te rmined as plaque -forming units (PFU) on He p-2 cells as de scribe d previously. 1 RSV infection RAW 264.7 ce lls w ere inoculated into a 96-w e ll mic roplate at a density of 13 10 5 cells/w e ll and cultured ove rnight at 37°C. Thereafter, the ce lls w ere w ashe d once w ith serum-free media and ex pose d for 3 h at 37°C to 20 m l of virus solution at a multiplicity of infec tion (MOI) of 3 PFU/c ell. After virus adsorption, 200 m l of se rum-fre e me dium w ith or w ithout drugs w e re added (0 h post-infec tion) and further cultured for additional time periods as indic ated.

MIP-2 assay
MIP-2 le ve ls in the CM w e re ex amined by antibodysandw ich enzyme-linke d immunosorbent assay (ELISA) in w hich rabbit unlabe led and biotinylate d anti-murine MlP-2 antibodie s w e re use d as the capture and sec ond-laye r antibodie s, re spec tively, as de scribed previously. 11 Purifie d MIP-2 w as used for standardization of MIP-2 le vels, and MIP-2 leve ls in ex perime ntal groups w ere ex pre sse d as perce nt of c ontrol culture, in w hich infe cte d ce lls w e re c ulture d in the abse nce of drugs. For statistical analysis, three w ells w ere use d for e ach ex p erimental point.

Statistical analysis
The results w e re ex pressed as mean± standard deviation (SD). The diffe renc es w ere ex amined by Bonferroni's least-significant differe nce test.

Results and discussion
In the control c ulture , MIP-2 le ve ls in the CM inc re ased to the le ve l of 20.4± 0.8 ng/ml until 20 h after infe ction in a time -de pendent manner. In the uninfec te d c ells, no significant produc tion (less than 0.5 ng/ml) w as observe d throughout the ex perime nts (data not show n). When the infected c ells w ere c ulture d in the presenc e of FA and IFA, the se drugs show ed inhibitory effe cts on MIP-2 production in a dose-dep endent manne r. At 20 h, FA reduced MIP-2 levels to 74.4 and 42.8% of c ontrol at doses of 50 and 500 m M, respe ctive ly (Fig. 1). In addition, IFA also re duce d MIP-2 levels to 61.8, 58.0 and 35.6% of c ontrol at doses of 5, 50 and 500 m M, respective ly (Fig. 2). These value s w ere significantly S. Sa k a i e t al.  In Japane se oriental medicine s, the rhizoma of CH are use d fre que ntly as antip yre tic, analgesic and antiinflammatory drugs, and FA and IFA have be en rec ently rec ognized as the main active components of CH ex trac t in the inflammation model in rats. 1 0 How e ver, the me chanisms of anti-inflammatory effect of FA and IFA are not unde rstood c omple te ly. In addition to a previous report indicating that FA inhibits MIP-2 production in resp onse to LPS stimulation in RAW264.7 ce lls, 1 2 this study confirmed that both FA and IFA inhibit MIP-2 produc tion even in response to RSV infec tion in RAW264.7 c ells in a dose-depende nt manner. Since peribronchiolar infiltrations of ne utrophils are important pathoge nesis in RSV infec tions, 1 3,1 4 MIP-2-inducing activity of RSV should play a ke y role. Taking into acc ount the fac t that neutrophils c an cause tissue injurie s, such as lung damage in adult resp iratory distre ss syndrome and other inflammatory disease s, through the p roduc tion of supe rox ide or some enzymes, in addition to killing inge ste d or invading mic robe s, 15 -1 7 these data sugge st that FA and IFA might contribute at least in part to the anti-inflammatory drug e ffec t of CH ex tract through the reduction of MIP-2 levels, and the reby the reduction of neutrophil infiltratio n into the inflammatory site s. In addition, the use of FA and IFA or CH ex tract seems to be a novel and attractive strate gy for control of pneumonia induce d by ce rtain virus infe ctions, including RSV infe c tion.