Background
Previous studies, including ours, have shown that loss of heterozygosity (LOH) on the long arm of chromosome 16 is one of the most frequent genetic events in breast, gastric and prostate cancers, implying the presence of one or more tumor suppressor genes (TSGs) at this location [
1‐
7]. In breast cancer, the gene encoding
E-cadherin at 16q22.1 was identified as a TSG, but only in the histological subgroup of lobular carcinoma [
8]. Recently, the
AT-motif binding factor 1 (
ATBF1)-
A gene (GenBank: L32832), which has been assigned to chromosome 16q22.3-23.1 [
9], was identified as a reasonable candidate for tumor suppressor activity in solid tumors, based on its functional inhibition of cell proliferation and high rate of mutations in prostate cancer [
10].
ATBF1-A was originally identified as a negative transcriptional factor for the
alpha-fetoprotein (AFP) gene through binding with the AT-rich motif in the AFP enhancer element I [
11,
12]. In gastric cancer, absence of
ATBF1-A is a distinct feature of AFP-producing cancer cells, which are characterized by a high malignant potential [
7,
13]. Moreover,
ATBF1-A negatively regulates the c-Myb oncoprotein [
14] and transactivates the cell cycle inhibitor
CDKN1A [
15]. Therefore, the
ATBF1-A gene is considered to be a good TSG candidate in solid tumors.
Previously, we reported that reduced
ATBF1-A mRNA levels in tumors correlated with axillary lymph node metastasis and estrogen receptor (ER)-α negative status in breast cancer, and with a worse prognosis [
16]. Sun et al. confirmed the presence of reduced
ATBF1-A mRNA levels in breast cancer cell lines [
17]. However, the reduced
ATBF1-A mRNA expression was attributed neither to promoter methylations nor to frequent somatic mutations [
17]. Therefore, the authors concluded that
ATBF1-A plays a role in breast cancer through transcriptional down-regulation rather than promoter methylation or mutations.
In addition to promoter methylations or mutations, LOH resulting from a deletion spanning one or more genes is one of the mechanisms by which the function of genes is lost. However, there are no papers in which has been reported the associations between LOH at the
ATBF1-A locus [
10] in the 16q22 minimal region and
AFBF1-A mRNA levels, or between LOH at this locus and the clinicopathological factors in breast cancer. We performed LOH analysis at the 16q22 minimal region and mutational analysis focusing on specific loci in the
ATBF1-A gene, which have been reported previously in prostate cancer[
10]. Our analysis shows that
ATBF1-A mRNA levels are not regulated by genetic machinery, LOH, or mutations. These findings could support the view that the
ATBF1-A gene plays a role in breast cancer through transcriptional down-regulation rather than through LOH and mutations.
Discussion
Using polymorphic microsatellite markers spanning chromosome 16q22, we performed fluorescent-labeled PCR amplification and capillary electrophoresis to investigate LOH status at the 16q22 minimal region in paired specimens of blood and primary breast tumors from 127 patients. We assessed the relationship between LOH status at the
ATBF1-A locus and mRNA levels of
ATBF1-A, as well as the clinicopathological factors in breast cancer. Quantitative assessment of
ATBF1-A mRNA expression according to LOH status at the
ATBF1-A locus demonstrated no relationship between these factors. This finding completely consists with the result repoted by Kim et al., studied in hepatocellular carcinoma (HCC)[
25]. We therefore concluded that a mechanism other than LOH was involved in regulating the transcriptional level of the
ATBF1-A gene in breast cancer as like in HCC.
Accordingly, we screened mutations of the specific loci in the
ATBF1-A gene that have previously been reported in prostate cancer [
10]. The 12 tumors investigated for mutations had been predicted to have a higher frequency of gene alterations in the
ATBF1-A gene, because they demonstrated lower
ATBF1-A mRNA levels but showed ROH at the
ATBF1-A locus (data not shown). In the mutational analyses, although none of the 12 samples studied showed somatic mutations with substitution of amino acids or frameshift, two germ line gene alterations were recorded. One of these alterations predicted a substitution of methionine for valine at codon 2572. This alteration also reported in gastric cancer as a somatic mutation at one allele, while the other allele had lost[
7]. Although the gene alteration was not reported in the NCBI SNP and JSNP databases, pathogenic significance of the gene alteration was also not verified by functional assay. The remaining gene alteration produced a 6-nucleotide deletion without frameshift in glutamic acid rich domain in exon10. Although the 3 – 24 nucleotides deletions are prevalent in various kind of tumors[
6,
7,
10,
26], its pathogeny is controversial among those tumors. In prostate cancer, Xu et. reported the deletions in germline and its relevance with the susceptibility to prostate cancer[
26]. Contrast to that, in breast cancer, Cleton-Jansen et al. reported no association between these germline deletions and suscetability to breast cancer[
6]. This indicate that these germline deletions has different impacts in the suscetabilities to these tumors. As in the present study, the deletion was also germline, they might be benign polymorphisms regarding with the susceptibility to breast cancer.
A larger case-control study compared between cancer patients and healthy individuals and functional analysis in each gene alteration should be performed to reveal their pathogenic significances in tumorigenesis. As, consequently, somatic mutations with substitution of amino acids or frameshift were not seen in these samples, we concluded that
ATBF1-A mRNA levels may be regulated at the transcriptional stage, but are not regulated by genetic mechanisms, deletions (LOH), or mutations in breast cancer. Infrequent somatic mutations in
ATBF1-A gene have previously been reported, except for prostate cancer. The frequencies were 8.6% in gastric cancer[
7], 0% in HCC[
25], 0% in breast cancer[
6]. In addition, Sun et al. and Kim et al. reported infrequent methylation at
ATBF1 gene promoter in breast cancer and HCC, respectively[
17,
25]. Based on these reports, we speculate that the methylations at ATBF1-A gene are also not attributed to the downregulation of ATBF1-A transcripts, though we did not performed the methylation analysis.
Recently, posttranscriptional mRNA repression associated with microRNA (miRNA) have been discussed as an alternative mechanism of mRNA modulation at the posttranscriptional level in mammals as well as metazoa [
27]. According to the prevailing model, posttranscriptional repression by miRNA is determined by the complementarity to the miRNA of the target mRNA[
27]. Therefore, we may speculate that posttranscriptional cleavage of mRNA by miRNA-associated machinery is the molecular mechanism of
ATBF1-A mRNA regulation. This may explain why there is a discrepancy between LOH status at
ATBF1-A locus and
ATBF1-A mRNA levels.
LOH status at the
ATBF1-A locus significantly correlated with positivity of PR (
P = 0.013) and negativity of HER2 (
P = 0.024) status. Similar results were previously reported by Wang [
28], who found that 16q23-24 genetic loss significantly correlated with ER-α positivity and HER2 negativity. These biological features, hormonal receptor positivity, and HER2 negativity, are reminiscent of the "luminal-type" tumors described by Perou et al. [
29]. Based on the study by Wang [
28] and the present study, a target gene at the narrowed locus spanning from 16q22 to 16q24 may determine the biological features of the luminal-type. Recent cytogenetic approaches, such as comparative genomic hybridization[
30], may help reveal the new target gene at this locus in such cohorts.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
K.K. performed the molecular genetic studies, participated in the LOH analysis and sequence alignment, and drafted the manuscript. Z.Z. performed the sequence alignment and scored the status of immunohistochemical staining. H.Y. was involved in the study design and scored the status of immunohistochemical staining. Y.M. was involved in the study design and assisted in drafting of the manuscript. Y.Y. and H.I. conceived the study, participated in its design and coordination, and assisted in drafting of the manuscript. All authors have read and approved the final manuscript.