No Correlation of hsa-miR-148a with Expression of PXR or CYP3A4 in Human Livers from Chinese Han Population

There is a huge variability of hepatic CYP3A4 level in human populations, which was believed to contribute to different responses to drugs among individuals. Transcription of CYP3A4 was regulated by transcription factors such as pregnane X receptor (PXR). MiRNA hsa-miR-148a was previously reported to influence PXR expression in HepG2 cells and in Japanese populations. In this study, we conducted a similar correlation study in Chinese Han population (N = 24). No significant correlation of hsa-miR-148a was found with PXR expression or CYP3A4 expression. Our results suggest that hsa-miR-148a does not play a major role in the regulation of PXR or CYP3A4 expression in human livers from Chinese Han population.


Introduction
Cytochrome P450 3A4 (CYP3A4), the most abundant hepatic and intestinal member of cytochrome P450 superfamily [1], contributes to the metabolism of over half of the current prescription drugs [2,3]. However, the CYP3A4 level has been reported to vary a lot (as great as 50-fold) in the general population, which cannot be explained by genetic polymorphisms alone [4]. Such huge inter-individual differences of CYP3A4 levels could explain the striking individual differences in drug responses, in terms of both therapeutic effects and adverse side effects [5]. Therefore, the regulation mechanism of CYP3A4 expression became a research hotspot in pharmacokinetics [6,7].
Previously, Takagi et al. reported that the protein level of pregnane X receptor (PXR), which is a major transcription factor of CYP3A4 [6,8,9], correlated significantly with the CYP3A4 mRNA and protein levels in Japanese population [10]. After in vitro investigation with reporter system, they further found that the translational efficiency of PXR (PXR protein/PXR mRNA ratio) was inversely correlated with the expression level of hsa-miR-148a in human livers [10]. Their results suggested the indirect regulatory function of hsa-miR-148a in CYP3A4 expression.
However, there are always controversial conclusions in clinical investigations, especially in different populations. Therefore, we asked whether their results could be replicated in Chinese Han population and whether the effect of hsa-miR-148a on PXR translation could indirectly influence CYP3A4 expression in the population. In the present study, we found that hsa-miR-148a might not play a major role in the regulation of PXR or CYP3A4 expression in human livers from Chinese Han population.

Liver Samples and Ethics Statement
Human liver samples from 24 Chinese Han donors were obtained in surgery from the First Affiliated Hospital of Zhengzhou University after informed consent was written. The donors included 16 males and 8 females with the age of 40.5564.63 (mean6SEM). Among them, six had mild cirrhosis and none had Hepatitis B or C. All donors included in this study had normal liver functions and didn't receive any preoperative medication of CYP3A4 activator or inhibitor (rifampicin, dexamethasone, propofol, etc) for 2 weeks, nor alcohol, grapefruit juice or caffeine-containing foods/beverages within 3 days before the surgery to avoid any inhibitory or inducing effects on CYP3A4 activity. The protocol complied with the Declaration of Helsinki and its subsequent revisions and was approved by the Ethics Committees of Zhengzhou University.

Immunoblot
Total proteins were extracted using RIPA lysis buffers containing protease inhibitor PMSF. Relative protein concentrations of PXR and CYP3A4 were accessed by immunoblot using rabbit anti-PXR antibody (AP8674c; Abgent, CA) and rabbit anti-CYP3A4 antibody (ab3572; Abcam, MA), respectively, normalized to GAPDH protein concentration.   The relative RNA levels of mRNAs and miRNA were measured using qPCR with the normalization to the GAPDH mRNA level and the U6 snRNA level, respectively. The relative protein levels were measured using immunoblot with the normalization to the GAPDH protein level. doi:10.1371/journal.pone.0059141.t001 . For miRNA quantification, reverse PCR and qPCR were performed using miRCURY LNA TM Universal RT microRNA PCR system (Exiqon, Denmark). U6 snRNA was measured as the endogenous control. All qPCRs were run on ViiA 7 instrument (Life technologies, CA) with three replicates.

Statistics
Linear regression, Student's t-test and Spearman's rank correlation were performed using SPSS for Windows, version 11.0 (IBM, NY). All tests were two-tailed and statistical significance was assumed at p,0.05.

PXR mRNA Level is Correlated with PXR Protein Level in Human Livers
We first examined the PXR mRNA level and protein level in our human liver samples (N = 24) by qPCR and immunoblot, respectively, and investigated the relationship between them. All qPCR and immunoblot data were shown in Table 1. As shown in Figure 1, significant linear correlation was observed between the PXR mRNA and protein levels (p = 0.004, R 2 = 0.324).  The CYP3A4 mRNA level in human liver samples was measured by qPCR, and was investigated about its relationship with the PXR protein level. As shown in Figure 2, significant linear correlation was observed between the PXR protein level and CYP3A4 mRNA level (p = 0.005, R 2 = 0.301).

Hsa-miR-148a did not Affect Expression of PXR in Chinese Han Population
We calculated the PXR protein/PXR mRNA ratio as an index of the translational efficiency of PXR, according to Takagi et al. [10]. There was no significant correlation between the translational efficiency of PXR and the hsa-miR-148a level in liver samples using Spearman's rank method (p = 0.850) or linear regression (p = 0.217), as shown in Figure 3A. The hsa-miR-148a level was not significantly correlated with PXR protein level either (Spearman's rank p = 0.923; Figure 3B).

Hsa-miR-148a did not Show Influence on the Expression of CYP3A4 in Chinese Han Population
To investigate whether hsa-miR-148a could indirectly regulate CYP3A4 transcription, we examined the relationship between hsa-miR-148a level and CYP3A4 mRNA level. As shown in Figure 4, hsa-miR-148a level in human livers was not significantly in Spearman's rank correlation with CYP3A4 mRNA level (p = 0.848) or protein level (p = 0.248).

Hsa-miR-148a Grouping or Cirrhosis Grouping didn't Associate with the Expression of PXR or CYP3A4
The relative expression of hsa-miR-148a seems to aggregate into two groups with a split point as 1.0 as shown in Figure 3. We compared the relative expression of PXR and CYP3A4 between high and low hsa-miR-148a expression groups. As shown in Table 2, there was no significant difference between these two groups. We also analyzed and found that mild cirrhosis status had no significant influence on PXR or CYP3A4 levels ( Table 2). To be noticed, mild cirrhosis status didn't show significant correlation with hsa-miR-148a expression (p = 0.472).

Discussion
PXR is a major transcription factor of CYP3A4 gene, and regulates CYP3A4 expression. In 2008, Takagi et al. found that hsa-miR-148a could bind to the 39-UTR region of PXR and influence its expression both in vitro and in vivo. However, our results suggest that hsa-miR-148a may not be involved in the regulation of PXR expression or further CYP3A4 expression.
The significant correlation between PXR protein level and CYP3A4 mRNA level is in line with the consensus that PXR is a transcription factor of CYP3A4, also consistent with Takagi et al.'s results [10]. Nevertheless, the linear correlation between PXR mRNA and protein levels is inconsistent with Takagi et al.'s results. They explained the non-significant correlation as the involvement of post-transcriptional regulation. Our significant correlation result, however, still couldn't rule out the involvement of miRNA's regulation, considering the R 2 value is small (R 2 = 0.324).
As a replication study, we tried to correlate the hsa-miR-148a level with the translational efficiency of PXR in our liver samples. The correlation was non-significant no matter using traditional linear regression or the Spearman's rank method as Takagi et al. used. To confirm our non-significant result, we also correlated the miRNA level and the PXR protein level. The correlation was still statistically non-significant. Furthermore, hsa-miR-148a did not show significant effect on CYP3A4 expression either.
It is a general problem that the clinical correlation study in one population cannot be replicated in another population. Several differences may contribute to the inconsistency of the influence of hsa-miR-148a between Takagi 3) The techniques to quantify the concentration of miRNA were different between the two studies. In this study, we used miRCURY LNA TM technology to make sure better sensitivity and more importantly higher specificity. (4) The regulation of PXR expression is complex and involves multiple signaling mechanisms, while the effect of hsa-miR-148a on the regulation of PXR expression might be weak. Besides, both these two studies had only small sample sizes, which impaired the statistical power of the conclusion. To offset the probability of false-negatives caused by limited sample size, we re-evaluated the alpha threshold using G*Power version 3.0 [11]. With a medium size effect (f 2 = 0.2), alpha threshold should be set as 0.191 to achieve power = 0.8. The non-significant correlation of hsa-miR-148a level with PXR and CYP3A4 expression still exceeded this more stringent threshold.
In summary, we failed to replicate the significant correlation between liver hsa-miR-148a level and the translational efficiency of PXR in Chinese Han population. Thus, our results didn't support the significant effect of hsa-miR-148a in PXR expression in human livers. Further replications with larger sample sizes and different ethnic populations were warranted to fully elucidate the exact role of hsa-miR-148a in the regulation of PXR expression and then CYP3A4 transcription.