A 1-step RT-PCR assay, targeting a 730 bp fragment of the nucleocapsid (N) gene of bovine coronavirus (BCV), and a nested PCR assay, targeting a 407 bp fragment of the N gene, were developed to detect BCV in nasal swab and fecal samples of calves experimentally exposed to BCV. Both 1-step RT-PCR and nested PCR recognized cell culture passaged isolates of 10 bovine respiratory coronavirus (BRCV), 5 calf diarrhea (CD) and 8 winter dysentery (WD) strains of BCV, but not transmissible gastroenteritis coronavirus or bovine rotavirus. The sensitivity of the 1-step RT-PCR and nested PCR was compared to that of an antigen-capture ELISA. The lowest detection limit of the 1-step RT-PCR and nested PCR as determined by using tenfold serial dilutions of the BRCV 255 and 440 strains in BCV negative nasal swab suspensions from preexposure gnotobiotic calves was 2 x 10(4) and 2 x 10(2) TCID50/0.1 ml for each strain, respectively. The lowest detection limit of the antigen-capture ELISA as determined by using the same serially diluted samples was 1 x 10(6) TCID50/0.1 ml for each strain. Therefore, the 1-step RT-PCR and nested PCR assays were 50 and 5000 times, respectively more sensitive than the antigen-capture ELISA to detect BRCV in nasal swab suspensions. To investigate in vivo cross-protection between the BRCV and CD or WD strains of BCV and to detect nasal and fecal shedding of BCV using the 1-step RT-PCR, nested PCR and antigen-capture ELISA, 6 colostrum-deprived and two gnotobiotic calves were inoculated with a BRCV, a CD or a WD strain of BCV and then challenged 3-4 weeks later with either BRCV, CD or WD strains of BCV. All calves developed diarrhea after inoculation and BCV antigen (ELISA) or RNA (RT-PCR) was detected in the diarrheic fecal samples or the corresponding nasal swab samples. In addition, low amounts of BCV were also detected only by nested PCR in the fecal and nasal swab samples before and after diarrhea. No respiratory clinical signs were observed during the entire experimental period, but elevated rectal temperatures were detected during diarrhea in the BCV-inoculated calves. All calves recovered from infection with the BRCV, CID, or WD strains of BCV were protected from BCV-associated diarrhea after challenge exposure with either a heterologous or homologous strain of BCV. However, all calves challenged with heterologous BCV strains showed subclinical BCV infection evident by detection of nasal and fecal shedding of BCV RNA detected only by nested PCR. Such results confirm field and experimental data documenting reinfection of the respiratory and enteric tracts of cattle, suggesting that, in closed herds, respiratory or enteric tract reinfections may constitute a source of BCV transmissible to cows (WD) or neonatal or feedlot calves. In addition, the present 1-step RT-PCR and nested PCR assays were highly sensitive to detect BCV in nasal swab and fecal specimens. Therefore. these assays should be useful to diagnose BCV infections in calves and adult cows.