TEXT

Top Abstract Text References

The presence of hepatitis C virus (HCV) in blood compartments is well known (1), and detection of HCV RNA in human secretions like saliva, seminal fluid, urine, and bile has been described (4, 5, 7, 9). In particular, the detection of HCV RNA in bile was a reason for us to examine fecal specimens from patients chronically infected with HCV since fecal specimens can be more easily obtained. Hence, the rationale for the experiments described here was to investigate the possibility that HCV might be excreted into fecal specimens of patients chronically infected with HCV. Thus far, fecal specimens have not been studied extensively, most likely because of impaired recovery and inhibition of amplification of DNA (3) and RNA, and data on both the frequency and the load of HCV in feces are not available (6). Here we report on the presence and load of HCV RNA in fecal specimens from chronically infected patients.

Six patients chronically infected with HCV from whom one or more fecal specimens were available were studied. In addition, fecal specimens from six subjects with an HCV-negative serostatus (EIA 3.0; Abbott Laboratories, Chicago, Ill.) were used as a control group. Approximately 30% (vol/vol) suspensions were made from fecal specimens by mixing the specimens with broth (nutrient broth no. 2 [Oxoid, Hampshire, England], 500 IU of penicillin G [Sigma, St. Louis, Mo.] per ml, 500 µg of streptomycin [Fisiopharma, Milan, Italy] per ml, and 3 µg of amphotericin B [Fungizone; Bristol-Myers Squibb, New Brunswick, N.J.] per ml). The fecal specimens were stored at 20°C; EDTA-anticoagulated plasma specimens were stored at 70°C.

HCV RNA was detected and quantified in plasma with the COBAS AMPLICOR system according to the manufacturer's manual (Roche Diagnostics Systems, Inc., Branchburg, N.J.).

For fecal specimens the procedure was adapted as described below. HCV RNA was extracted from 50 µl of fecal suspension by a modification of the procedure of Van der Hoek et al. (11). In short, 50 µl of fecal suspension was added to 900 µl of lysis buffer L6 (2); next, 84 molecules of internal control RNA (Roche) for qualitative detection or approximately 2,000 molecules (lot number specific) of quantification standard RNA (Roche) for quantitative detection were added. After 10 min at ambient temperature, the tubes were centrifuged (for 2 min at 12,000 × g). Nine-hundred microliters of supernatant was transferred to a new tube containing 50 µl of a modified silica suspension which was prepared as described by Boom et al. (2) but which contained 2,400 µl of 32% (vol/vol) HCl per 60 ml of silica suspension rather than 600 µl. After a 10-min binding step, the silica pellets were washed and dried as described previously (2). RNA was eluted in 100 µl of TE buffer (10 mM Tris, 1 mM EDTA; pH 8.0). For qualitative detection of HCV RNA from feces, 10 µl of RNA was added to 40 µl of HCV diluent (Roche). For quantitative detection, 5 µl of RNA was added to 45 µl of HCV diluent (Roche). The resulting mixtures were added to 50 µl of mastermix (Roche) and further processed with the COBAS AMPLICOR, version 2.0, system according to the manufacturer's instructions for qualitative or quantitative detection. Viral loads (expressed as the number of copies of HCV RNA per milliliter of feces) were calculated by the COBAS AMPLICOR software and were corrected for input volumes. By following this procedure, detection and quantitation of HCV RNA in feces are controlled for both HCV RNA recovery and the presence of inhibitors.

Internal control RNA was detected for all fecal specimens tested (n = 19). All six fecal specimens from HCV-seronegative controls were HCV RNA negative. To study the reliability of quantitation of HCV RNA in feces by the COBAS AMPLICOR system, an HCV-negative fecal specimen (Table 1, patient E) was supplemented with known amounts of HCV RNA by the addition of serial dilutions of a plasma specimen for which the HCV RNA load had been quantified. The expected and calculated values were in excellent accordance for viral loads of 3 × 10³ to 3 × 10⁶ copies of HCV RNA/ml (Fig. 1). HCV RNA was detected in feces from four of six patients (67%) with HCV levels up to 2.8 × 10⁵ copies/ml of feces. No clear relation was found between HCV RNA levels in plasma and feces. For two patients (patients E and F), HCV RNA levels in plasma exceeded 10⁶ copies/ml, whereas the corresponding fecal specimens were HCV RNA negative (Table 1).

View larger version (15K): [in this window] [in a new window]   FIG. 1.   An HCV-negative fecal specimen was supplemented with known amounts of HCV by addition of serial dilutions of a plasma specimen for which the HCV RNA load had been quantified (3 × 106 to 3 × 103 copies of HCV RNA/ml). The correlation coefficient (r = 0.999) and slope (1.008) were obtained by linear regression analysis after logarithmic transformation.

To confirm the specificity of HCV detection in feces, amplicons obtained by the COBAS AMPLICOR, version 2.0, assay were used for direct sequencing by using the TruGene HCV Genotyping Assay and the OpenGene automated DNA sequencing system (Visible Genetics Inc., Toronto, Ontario, Canada). The same HCV genotypes were found in the plasma and fecal specimens of three patients. In one patient (patient D), the genotype of the HCV in feces could not be determined, presumably due to the small amount of HCV RNA (Table 1).

In the COBAS AMPLICOR system only plasma and serum samples are suitable as input materials. The method described here might be a useful tool for the detection and quantitation of HCV RNA in clinical specimens other than plasma or serum samples. In the present study, HCV RNA was frequently found in the feces of chronically infected patients in relatively large amounts. Fecal specimens have not been studied extensively, most likely because of impaired recovery and inhibition of amplification of DNA (3) and RNA. Internal control RNA was detected for all fecal specimens (n = 19), suggesting that the rate of recovery of RNA was high and that inhibition of reverse transcription and amplification was not present. The significance of HCV RNA in feces is unknown, and several mechanisms could attribute to the detection of HCV RNA in feces. The presence of blood in feces may be obvious and can be tested with freshly obtained fecal specimens by tests for occult blood. However, in our study only frozen fecal suspensions were available, and these are unsuitable for use in tests for occult blood. Alternatively, extrahepatocellular HCV replication in lymphocytes was recently described (8) and may contribute to the presence of HCV RNA in feces. Leakage of HCV virions in body fluids by liver damage may also occur, and detection of HCV RNA in biliary epithelial cells (10) and the presence of HCV RNA in bile from 8 of 10 seropositive patients (7) have recently been described. Therefore, bile and feces are human secretions in which the virus is possibly excreted. Whether HCV was excreted in an infectious form remains to be addressed. We conclude that HCV RNA was frequently found in large amounts in feces from patients chronically infected with HCV.