Optimization of estrogen receptor analysis by immunocytochemistry in random periareolar fine-needle aspiration samples of breast tissue processed as thin-layer preparations

Epigenetic regulation of transcription factor promoter regions by low-dose genistein through mitogen-activated protein kinase and mitogen-and-stress activated kinase 1 nongenomic signaling

Background

The phytoestrogen, genistein at low doses nongenomically activates mitogen-activated protein kinase p44/42 (MAPK_p44/42) via estrogen receptor alpha (ERα) leading to proliferation of human uterine leiomyoma cells. In this study, we evaluated if MAPK_p44/42 could activate downstream effectors such as mitogen- and stress-activated protein kinase 1 (MSK1), which could then epigenetically modify histone H3 by phosphorylation following a low dose (1 μg/ml) of genistein.

Results

Using hormone-responsive immortalized human uterine leiomyoma (ht-UtLM) cells, we found that genistein activated MAPK_p44/42 and MSK1, and also increased phosphorylation of histone H3 at serine10 (H3S10ph) in ht-UtLM cells. Colocalization of phosphorylated MSK1 and H3S10ph was evident by confocal microscopy in ht-UtLM cells (r = 0.8533). Phosphorylation of both MSK1and H3S10ph was abrogated by PD98059 (PD), a MEK1 kinase inhibitor, thereby supporting genistein’s activation of MSK1 and Histone H3 was downstream of MAPK_p44/42. In proliferative (estrogenic) phase human uterine fibroid tissues, phosphorylated MSK1 and H3S10ph showed increased immunoexpression compared to normal myometrial tissues, similar to results observed in in vitro studies following low-dose genistein administration. Real-time RT-PCR arrays showed induction of growth-related transcription factor genes, EGR1, Elk1, ID1, and MYB (cMyb) with confirmation by western blot, downstream of MAPK in response to low-dose genistein in ht-UtLM cells. Additionally, genistein induced associations of promoter regions of the above transcription factors with H3S10ph as evidenced by Chromatin Immunoprecipitation (ChIP) assays, which were inhibited by PD. Therefore, genistein epigenetically modified histone H3 by phosphorylation of serine 10, which was regulated by MSK1 and MAPK activation.

Conclusion

Histone H3 phosphorylation possibly represents a mechanism whereby increased transcriptional activation occurs following low-dose genistein exposure.

Electronic supplementary material

The online version of this article (doi:10.1186/s12964-016-0141-2) contains supplementary material, which is available to authorized users.

Incorporating Biomarkers in Studies of Chemoprevention

Despite Food and Drug Administration approval of tamoxifen and raloxifene for breast cancer risk reduction and endorsement by multiple agencies, uptake of these drugs for primary prevention in the United States is only 4% for risk eligible women likely to benefit from their use. Side effects coupled with incomplete efficacy and lack of a survival advantage are the likely reasons. This disappointing uptake, after the considerable effort and expense of large Phase III cancer incidence trials required for approval, suggests that a new paradigm is required. Current prevention research is focused on (1) refining risk prediction, (2) exploring behavioral and natural product interventions, and (3) utilizing novel translational trial designs for efficacy. Risk biomarkers will play a central role in refining risk estimates from traditional models and selecting cohorts for prevention trials. Modifiable risk markers called surrogate endpoint or response biomarkers will continue to be used in Phase I and II prevention trials to determine optimal dose or exposure and likely effectiveness from an intervention. The majority of Phase II trials will continue to assess benign breast tissue for response and mechanism of action biomarkers. Co-trials are those in which human and animal cohorts receive the same effective dose and the same tissue biomarkers are assessed for modulation due to the intervention, but then additional animals are allowed to progress to cancer development. These collaborations linking biomarker modulation and cancer prevention may obviate the need for cancer incidence trials for non-prescription interventions.

Assessment of RNA in human breast tissue sampled by random periareolar fine needle aspiration and ductal lavage and processed as fixed or frozen specimens

Ductal lavage (DL) and random periareolar fine needle aspiration (RPFNA) have both been proposed as minimally invasive techniques to sample breast tissue during breast cancer prevention trials. Laser capture microdissection (LCM), linear RNA amplification and quantitative real-time polymerase chain reaction (qPCR) theoretically overcome the limitations of small specimen size obtained with DL and RPFNA. In order to test the yield, relative stability and amplifiability of RNA from fixed and archived RPFNA and DL specimens, breast tissue was sampled from individual high risk women (n=9) by both DL and RPFNA. RPFNA samples showed good RNA/cDNA yield and amplification while only 2 of 9 of the paired DL specimens had cDNA of adequate quality for subsequent PCR. One and two rounds of linear amplification provided approximately a 200- and 20,000-fold enrichment of RNA, respectively. PCR analysis consistently detected ER and COX-1 mRNA in the majority of RPFNA samples examined while pS2, PCNA, VEGF and survivin expression varied with subject. RNA yield and/or stability was greater for fixed and archived RPFNA than DL specimens of breast tissue. In a subsequent study examining an expanded biomarker gene panel in fixed vs. frozen RPFNA samples, mRNA profiles and ranked relative mRNA abundance were similar (r=0.89) for frozen and fixed RPFNA specimens. In summary, frozen RPFNA samples may be optimal for RNA endpoints in human breast cancer prevention trials but fixed RPFNA specimens allow similar analyses with greater convenience.

Reduction in proliferation with six months of letrozole in women on hormone replacement therapy

The objective of this study was to determine if 6 months of the aromatase inhibitor letrozole, administered to postmenopausal women taking a stable dose of hormone replacement remedy, would be safe and would modulate biomarkers of breast cancer risk. The intent was to reduce the proliferation marker Ki-67 while maintaining adequate systemic levels of estradiol so as to avoid perimenopausal symptoms. Postmenopausal women at high risk for development of breast cancer and taking a stable dose of estrogen or estrogen plus progestin were screened by random periareolar fine needle aspiration (RPFNA). To be eligible, the acquired breast epithelial cells had to be characterized as cytologic atypia or borderline atypia with > or =1,000 epithelial cells on the cytomorphology slide; plus > or =500 epithelial cells on a slide processed for Ki-67 immunocytochemistry. Forty-two women were enrolled in the one arm study and received 2.5 mg letrozole per day for 6 months, followed by repeat assessment of biomarkers. Ki-67 was reduced by a median relative value of 66%. There was no significant change in breast cell cytomorphology; ER weighted index score; serum estradiol, testosterone, or IGF-1:IGFBP-3 ratio; mammographic breast density, or frequency or severity of perimenopausal symptoms. Given the dramatic reduction in proliferation, the effect of letrozole on risk and response biomarkers should be explored further in a randomized, placebo-controlled Phase IIB breast cancer chemoprevention trial.

Estrogen Receptor Expression in Benign Breast Ductal Cells Obtained from Random Periareolar Fine Needle Aspiration Correlates with Menopausal Status and Cytomorphology Index Score

INTRODUCTION

Estrogen receptor (ER) expression in breast epithelial cells has potential as a risk marker for development of breast cancer and as a response marker for preventive interventions.

AIM

The purpose of this study was to determine if ER expression in benign cytologic specimens acquired by random periareolar fine needle aspiration (RPFNA) increases with morphologic abnormality as has been reported for histologic preparations.

METHODS

ER expression was assessed in 122 women at high risk for development of breast cancer who had RPFNA hyperplasia +/- atypia and were being screened for entry into one of two chemoprevention trials. ER was assessed using antigen retrieval at 90 degrees C for 2 min and the DAKO ER monoclonal antibody (Clone number 1D5). The proportion of cells with definitive staining at each intensity level (0-4) was recorded as a percentage of the total cells counted, to give a weighted intensity score (IS).

RESULTS

Of 122 women, 65% exhibited hyperplasia and 35% exhibited hyperplasia with atypia in their RPFNA specimens. A majority (66%) of subjects had at least 10% of ductal cells exhibiting nuclear staining for ER. Median percent of cells with > or =1+ staining was 20% and the median ER IS was 0.23. There was a strong correlation between ER IS and percentage of ER positive cells (R(2) = 0.88). By univariate analysis ER IS was statistically significantly higher in women older than median age of 48 years (P = 0.025), in postmenopausal women on HRT (P < 0.017), and in women with a Masood cytomorphology index score of > or =14 (P = 0.005). On multivariable analysis, ER IS was significantly associated with postmenopausal status (P = 0.038) and cytomorphology as measured by Masood score (P = 0.043).

CONCLUSION

ER can be readily measured in cytologic specimens obtained by RPFNA with the use of antigen retrieval method. Further, ER expression in cytologic specimens is influenced by postmenopausal status and morphologic abnormality.