Differential expression of microRNAs in early-stage neoplastic transformation in the lungs of F344 rats chronically treated with the tobacco carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone

Abstract

While numerous microRNAs (miRNAs) have been reported to alter their expression levels in human lung cancer tissues compared with normal tissues, the function of these miRNAs and their contribution to the long process of lung cancer development remains largely unknown. We applied a tobacco-specific carcinogen-induced cancer model to investigate the involvement of miRNAs in early lung cancer development, which could also provide information on potential, early biomarkers of lung cancers. Male F344 rats were first chronically treated with 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), a carcinogen present in tobacco products, for up to 20 weeks. The expression profiles of miRNAs in rat lungs were then determined. As measured by miRNA microarrays and confirmed by Northern blot and real-time polymerase chain reaction analyses, NNK treatment reduced the expression of a number of miRNAs, such as miR-101, miR-126*, miR-199 and miR-34. Significantly, these miRNAs overlap with previously published reports on altered miRNA expression in human lung cancer samples. These miRNAs might, therefore, represent early-response miRNAs that signify the molecular changes associated with pulmonary tumorigenesis. Moreover, we identified cytochrome P450 (CYP) 2A3, a critical enzyme in rat lungs that activates NNK to render it carcinogenic, as a potential target of miR-126*. NNK treatment in rats repressed miR-126* but induced CYP2A3 expression, a mechanism that may potentiate the oncogenic effects of NNK.

Fig. 1.

Relative miRNA expression determined by microarray studies. Expression levels of the indicated miRNAs in the lungs of 1-, 5-, 16- and 20-week, control and NNK-treated rats were compared with those in the control, 1-week rats, which were arbitrarily set at 100 (y-axis). Shown are the averages, standard deviations (error bars) and P-values.

Fig. 2.

Northern blot validation of microarray results. Total RNAs pooled from two rats at 1 or 20 weeks of the same treatments were separated on a denaturing gel and transferred to a membrane. The membrane was then probed for miRNAs or the control U6 snRNA. Results were analyzed using a PhosphorImager. Levels of the indicated miRNAs, after normalized to the U6 signals and compared with those of the control, 1-week RNA arbitrarily set at 100, were shown below the phosphorimages.

Fig. 3.

Real-time PCR analyses of miRNA expression. RNA samples were reverse transcribed and then amplified using miRNA-specific primers or a primer for the normalization control, U6 snRNA. ΔC_T values (y-axis) for the miRNAs were calculated as C_T-miRNA−C_T-U6. Averages, standard deviations (error bars) and P-values from two to three biological replicates are shown.

Fig. 4.

CYP2A3 as a potential target of miR-126*. (A) Sequence alignment between miR-126* and its partially complementary 3′ UTRs of CYP2A family members from several species. Underlined nucleotides were deleted in the mutant CYP2A3 reporter construct; a relatively large deletion was necessary to disrupt the interaction between miR-126* and the 3′ UTR. (B) Inhibition of reporter expression by miR-126*. 293T cells were cotransfected with a reporter construct expressing the firefly luciferase linked to the wild-type or mutant CYP2A3 3′UTR, pRL-CMV and short hairpin RNA that expressed a control miRNA or miR-126*. Relative reporter expression is represented by firefly luciferase activities divided by the Renilla luciferase activities expressed from pRL-CMV, set at 100% for the control RNA transfections. Shown are the averages and standard deviations (error bars) of three to four experiments. The P-value was calculated using two-tailed, paired Student’s t-test. (C) CYP2A3 mRNA expression in the lungs of 1-, 5-, 16- and 20-week, control and NNK-treated rats, as measured by real-time PCR. The β-actin mRNA was the normalization control, and ΔC_T values (y-axis) were calculated as C_T-CYP2A3−C_T-actin. Averages and standard deviations (error bars) are shown. To assess the differences in CYP2A3 mRNA expression based on the ΔC_T values in control and NNK-treated, 16- and 20-week rats, two-tailed, heteroscedastic Student’s t-test was performed to yield a P-value of 0.003. (D) Protein expression in the lungs of individual, control and NNK-treated rats, as detected by western blotting. The ratios between CYP2A3 and actin band intensities are listed on top. Shown is a representative of three experiments that gave similar results.

Fig. 5.

A model to explain some of the consequences of NNK exposure. When administered to rats, NNK is activated by CYP2A3 in the lungs, which then reduces the expression of certain miRNAs such as miR-126* and miR-34. CYP2A3 is itself a target of miR-126*, so lowered miR-126* contributes to the increased production of CYP2A3, further activating NNK in a positive feedback loop. As miR-34 represses the expression of many genes involved in cell cycle control, apoptosis, etc., loss of miR-34 deregulates those important genes. These proposed molecular changes, together with others, eventually lead to the development of lung cancer.