Application of quantitative real-time reverse transcription-PCR in assessing drug efficacy against the intracellular pathogen Cryptosporidium parvum in vitro

Abstract

We report here on a quantitative real-time reverse transcription-PCR (qRT-PCR) assay for assessing drug efficacy against the intracellular pathogen Cryptosporidium parvum. The qRT-PCR assay detects 18S rRNA transcripts from both parasites, that is, the cycle threshold for 18S rRNA from parasites (C(T)([P18S])) and host cells (C(T)([H18S])), and evaluates the relative expression between parasite and host rRNA levels (i.e., deltaC(T) = C(T)([P18S]) - C(T)([H18S])) to minimize experimental and operational errors. The choice of qRT-PCR over quantitative PCR (qPCR) in this study is based on the observations that (i) the relationship between the logarithm of infected parasites (log[P]) and the normalized relative level of rRNA (deltadeltaC(T)) is linear, with a fourfold dynamic range, by qRT-PCR but sigmoidal (nonlinear) by qPCR; and (ii) the level of RNA represents that of live parasites better than that of DNA, because the decay of RNA (99% in approximately 3 h) in dead parasites is faster than that of DNA (99% in approximately 24 to 48 h) under in vitro conditions. The reliability of the qRT-PCR method was validated by testing the efficacies of nitazoxanide and paromomycin on the development of two strains of C. parvum (IOWA and KSU-1) in HCT-8 cells in vitro. Both compounds displayed dose-dependent inhibitions. The observed MIC50 values for nitazoxanide and paromomycin were 0.30 to 0.45 micro/ml and 89.7 to 119.0 microg/ml, respectively, comparable to the values reported previously. Using the qRT-PCR assay, we have also observed that pyrazole could inhibit C. parvum development in vitro (MIC50 = 15.8 mM), suggesting that the recently discovered Cryptosporidium alcohol dehydrogenases may be explored as new drug targets.

FIG. 1.

Standard curves for IOWA strain (A) and KSU-1 strain (B) of C. parvum infection in HCT-8 cells as determined by qRT-PCR (solid line) and qPCR (dashed line) by detecting the parasite and host cell 18S rRNA and 18S rRNA gene. At each oocyst inoculum level, the mean ΔΔC_T values were derived from two sample replicates, and each sample was subjected to at least two qRT-PCRs or qPCRs. The qRT-PCR curve was derived by a linear regression between the ΔΔC_T values and the logarithm of oocyst inoculum (presented here as the number of oocysts per 10⁶ host cells), whereas the qPCR curve was derived by a nonlinear regression by using a four-parameter logistic equation (see equation 9 in Materials and Methods). Bars represent standard errors of the means.

FIG. 2.

Decay of RNA and DNA in dead C. parvum sporozoites spiked into the HCT-8 cells in 24-well plates. After incubation at 37°C for the specified periods of time, total RNA and DNA were isolated for the detection of parasite and host cell 18S rRNA and the 18S rRNA gene by qRT-PCR and qPCR, respectively. The mean of the ΔΔC_T values at each time point of incubation was derived from at least two sample replicates, and each sample was subjected to at least two qRT-PCRs or qPCRs. (A) The ΔΔC_T values required for detection of parasite 18S rRNA (solid line) or the 18S rRNA gene (dashed line); (B) the percent decay of parasite RNA and DNA derived from the ΔΔC_T values by using the standard curves shown in Fig. 1. Bars represent standard errors of the means.

FIG. 3.

Efficacies of nitazoxanide (A) and paromomycin (B) on the growth of C. parvum (IOWA and KSU-1 strains) in vitro, as determined by the qRT-PCR assay. The percent inhibition curves were derived by a nonlinear regression by using a hyperbolic equation (A) or sigmoidal model (B). Bars represent standard errors of the means derived from at least four replicates.

FIG. 4.

Efficacy of pyrazole on the growth of C. parvum (IOWA strain) in vitro, as determined by the qRT-PCR assay. (A) Percent inhibition curve derived by a nonlinear regression by using a fourth-order polynomial equation. A similar curve was also obtained with a hyperbolic equation, albeit with a slightly higher variation (curve not shown). The inset shows the reversed sigmoidal curve representing the relative parasite growth against the logarithm of drug. (B) Cytotoxicity of pyrazole on HCT-8 cells, as determined by MTT assay. ***, value significantly different from that for the control (by t test, P < 0.001); OD, optical density. Bars represent standard errors of the means derived from at least four replicates.