Campylobacter jejuni and Campylobacter coli are foodborne enteric pathogens of human beings (Allos 2001), with poultry meat being a major source of Campylobacter infection (Park 2002). Recently, increasing levels of resistance in C jejuni and C coli to macrolide and fluoroquinolone antibiotics have become a worldwide problem as these are the therapeutic agents used for Campylobacter infections in humans (Aarestrup and Engberg 2001). The aims of this study were to investigate the antimicrobial resistance of Campylobacter isolates from human beings and poultry meat in Kanagawa, Japan, and to detect point mutations related to macrolide and quinolone resistance by using a line probe assay (MQ-LiPA), designed for the rapid and easy simultaneous detection of macrolide resistance mutations in 23S rDNA and quinolone resistance mutations in gyrA (Niwa and others 2003). In total, 249 isolates were used in this study: 193 human Campylobacter strains and 56 poultry meat strains were isolated in Kanagawa prefecture from 1979 to 2001 and from 1997 to 2000, respectively, and identified by biological and biochemical tests in the Kanagawa Prefectural Public Health Laboratory. All strains were cultured on Mueller-Hinton (MH) agar plates (Oxoid) at 42°C under microaerobic conditions (85 per cent nitrogen, 5 per cent oxygen, 10 per cent carbon dioxide) for 24 hours. The following antimicrobial agents were tested: nalidixic acid (Sigma), ofloxacin (Sigma), erythromycin (Sigma), tetracycline (Wako), ampicillin (Sigma), gentamicin (Sigma) and phosphomycin (Wako). Minimum inhibitory concentrations (MICs) were determined by an agar dilution method using MH agar plates. All MIC plates were inoculated with approximately 104 colony-forming units of each strain per spot using a multipoint inoculator. A control plate without antibiotics was also incubated at the end of each procedure. The plates were incubated for 24 hours at 42°C under microaerobic conditions. The MIC was defined as the lowest concentration producing no visible growth. The breakpoints were used according to the recommendations of the National Committee for Clinical Laboratory Standards (NCCLS 2002) except for ofloxacin and phosphomycin. No criteria were shown for ofloxacin or phosphomycin in the NCCLS document, so a breakpoint of 8 mg/litre was used for ofloxacin (Sanchez and others 1994) and 128 mg/litre for phosphomycin (Gomez-Garces and others 1995). The line probe assay (MQ-LiPA) was performed by the method described by Niwa and others (2003). This method is based on PCR and reverse hybridisation to a number of oligonucleotide probes immobilised as parallel lines on a nylon membrane strip. Oligonucleotide probes were tailed with dTTP (Gibco BRL). Poly (dT)-tailed probes and biotinylated DNA used as a colour development control were immobilised as parallel lines on the nylon membrane strip by ultraviolet exposure. The PCR was performed using primers designed according to the sequences of 23S rDNA and gyrA. The results of the antimicrobial susceptibility tests for the C jejuni and C coli isolates are shown in Table 1. In human isolates, the frequency of strains resistant to the quinolones, nalidixic acid and ofloxacin, increased remarkably, while tetracycline-resistant strains decreased during the study periods 1979 to 1989 and 1990 to 2001. Quinolone-resistant Campylobacter strains were observed from 1977 in both human and poultry meat isolates in this study. In Japan, many kinds of fluoroquinolones were introduced from the second half of the 1980s and the first half of the 1990s in the medical and veterinary fields, respectively. Poultry meat is a main source of Campylobacter infection in Japan (Ono and Yamamoto 1999) and various other countries (Padungton and Kaneene 2003), and the increase in the incidence of fluoroquinolone resistance in human isolates appears to be related to that in poultry. There was a marked decrease in erythromycin-resistant strains in human C jejuni isolates between the two study periods (1979 to 1989 and 1990 to 2000). The resistance level of all erythromycin-resistant C jejuni strains was low (MIC 8 to 16 mg/litre), while high-level (MIC ≥128 mg/litre) erythromycin-resistant strains were observed only in C coli. Low-level erythromycin-resistant strains might not be truly resistant, because the amount of carbon dioxide in the atmosphere increases the MICs of erythromycin (Rautelin and others 1991). Range of MIC (mg/litre) Profile by MQ-LiPA* Nalidixic acid Ofloxacin Erythromycin