Suppression of growth by all-trans retinoic acid requires prolonged induction of interferon regulatory factor 1 in cervical squamous carcinoma (SiHa) cells

Abstract

All-trans retinoic acid (ATRA) suppresses growth of cervical dysplasias in vivo, although the sensitivity to retinoids is frequently lost during cervical carcinogenesis. It has been suggested that prolonged treatment or use of higher doses of retinoids might offer favorable response rates. We found SiHa cervical squamous carcinoma cells that were virtually resistant to ATRA-induced growth-inhibitory effects at physiological doses (10(-7 to) 10(-6) M) to be more responsive at pharmacological doses (10(-5 to) 10(-4) M). The growth inhibition by high-dose (10(-4) M) ATRA was associated with a sustained activation of interferon regulatory factor 1 (IRF-1), while a low dose (10(-6) M) of ATRA activated IRF-1 only transiently. Antisense IRF-1 inhibited the high-dose (10(-4) M), ATRA-mediated growth arrest; forced expression of IRF-1 caused a significant reduction in cell growth. High-dose (10(-4) M) ATRA increased binding of NF-kappaB and STAT1 proteins to sequences that originated from the IRF-1 promoter region, while low-dose (10(-6) M) ATRA induced only NF-kappaB binding. A delayed tyrosine phosphorylation of the signal transducer and activator of transcription-1 (STAT1) was observed after high-dose (10(-4) M) but not low-dose (10(-6) M) ATRA treatment. In agreement with this, induction of IRF-1 mRNA by ATRA was only modest and transient in a STAT1 knockout cell line, suggesting the importance of STAT1 in sustained IRF-1 expression. Our data showed that ATRA is capable of inducing dose-dependent cellular changes, which might be appropriate to overcome resistance to retinoids that frequently develops during cervical carcinogenesis.

FIG. 1.

Effects of ATRA on cell growth. SiHa cells were seeded in a multiwell plate and treated with either low-dose (10⁻⁷ and 10⁻⁶ M) or high-dose (10⁻⁵ and 10⁻⁴ M) ATRA for the times indicated. An MTT assay was employed to determine the amounts of viable cells. Results are shown as means ± standard deviations (n = 3).

FIG. 2.

Effects of ATRA on IRF-1 protein levels in SiHa cells. Confluent cultures of SiHa cells were treated with either low-dose (10⁻⁶ M) or high-dose (10⁻⁴ M) ATRA for the times indicated. Top, Western immunoblotting was performed on the total cell lysate as described in Materials and Methods. Bottom, Densitometric results are expressed as fold change to the untreated controls. Values are shown as means ± standard deviations (n = 3).

FIG. 3.

Dose-dependent induction of IRF-1 by ATRA. Confluent cultures of SiHa cells were treated for 24 h with various doses of ATRA as indicated. Western immunoblotting was performed on the total cell lysate as described in Materials and Methods. Densitometric results are expressed as fold change to the untreated controls. Values are shown as means ± standard deviations (n = 3).

FIG. 4.

Prolonged induction of IRF-1 requires STAT1. STAT1 knockout (U3A) or STAT1 wild-type (2fTGH) cells were treated with ATRA (10⁻⁶ M) for different time points. IRF-1 mRNA levels were determined by reverse transcription-PCR together with the constitutively expressed G3PDH. IRF-1 levels are given as ratios of IRF-1/G3PDH. Values are shown as means ± standard deviations (n = 3).

FIG. 5.

Effects of different doses of ATRA on tyrosine phosphorylation of STAT1 in SiHa cells. Confluent cultures of SiHa cells were treated with low-dose (10⁻⁶ M) or high-dose (10⁻⁴ M) ATRA for the times indicated. Western immunoblotting was performed using anti-phosphoSTAT1 (STAT1p) as well as anti-STAT1 antibodies. Experimental data shown are representative of three independent experiments.

FIG. 6.

Effects of different doses of ATRA treatment on binding of cellular proteins to oligonucleotides representing NF-κB or GAS sites of the IRF-1 promoter. Nuclear extracts from SiHa cells treated with low-dose (A and C) or high-dose (B and D) ATRA were incubated with labeled oligonucleotides representing the NF-κB (A and B) or GAS (C and D) binding sites of the IRF-1 promoter, and an EMSA was performed as described in Materials and Methods. In panels A and B we showed only the 30-min time point that represented the first appearance of NF-κB binding. STAT1 binding to the GAS oligomer (C and D) appeared after 6 h of treatment. Next, 100× cold ligand was used to demonstrate specificity of both probes (data not shown). Tetradecanoyl phorbol acetate (TPA)- or gamma interferon (IFNγ)- treated cell extracts were used as positive controls. Preincubations with specific anti-NF-κB p65 and p50 or STAT1 antibodies were performed to demonstrate the presence of these proteins in the observed complexes (data not shown).