Search

Background: CDK4/6 inhibitors—palbociclib, ribociclib, and abemaciclib—result in prolonged progression free survival for advanced estrogen receptor positive breast cancer patients. However, the optimum CDK4/6 inhibitor for each patient remains unclear until the emergence of side effects, which are the key differentiators between the two treatments. Furthermore, an optimum treatment after acquisition of resistance is poorly understood. Therefore, we elucidated the difference in CDK4/6 inhibitor resistance mechanisms using ribociclib- and abemaciclib-resistant cell lines in vitro.Methods: We established hormone-resistant cell lines. Estrogen deprivation-resistant (EDR) cell lines (EDR1:ER-positive, EDR2:ER-negative) and fulvestrant-resistant (MFR) cell lines were established from the MCF-7 cell line. We established ribociclib-resistant cell lines (RIBR) and abemaciclib-resistant cell lines (ABER) from EDR1 by long-term culture (7 months for RIBR and 12 months for ABER) with ribociclib and abemaciclib, respectively. We also established CDK6 overexpressed cell lines (MCF7-C6) from MCF-7 cell line using the stably transfected CDK6 expression vector.Results: First, we identified the efficacies of ribociclib and abemaciclib in several cell lines. Luminal cell lines (MCF-7 and T-47D) exhibited high sensitivity to ribociclib and abemaciclib compared to non-luminal cell lines (MDA-MB-231, BT-20, and SKBR3). The hormone-resistant cell lines (EDR1, EDR2, and MFR) exhibited similar sensitivity to ribociclib and abemaciclib as that of MCF-7; this was independent of hormone resistant mechanism.Subsequently, we compared RIBR and ABER, which are the models of acquired ribociclib- and abemaciclib-resistant cell lines, respectively. We confirmed that RIBR and ABER decreased ribociclib and abemaciclib sensitivity, respectively. RIBR showed low sensitivity to abemaciclib. Similarly, ABER exhibited low sensitivity to ribociclib. Immunoblot analysis showed that compared to EDR1, RIBR and ABER had extremely low levels of total Retinoblastoma (RB). Interestingly, compared to EDR1, CDK6 levels were upregulated in ABER and remained unaltered in RIBR. On the other hand MCF7-C6 reduced sensitivity not only abemaciclib but also ribociclib compared to MCF-7 cell line. Furthermore, compared to EDR1, RIBR exhibited lower levels of p21 and p27 compared to EDR1 (P21 levels were extremely reduced, but p27 levels were slightly reduced). Additionally, ABER showed markedly lower levels of p27; however, the p21 level was slightly increased compared to EDR1. PI3K inhibitor and mTOR inhibitor suppressed cell growth in RIBR and ABER to the same extent as EDR1, indicating that resistance to CDK4/6 inhibitors was still dependent on the PI3K/Akt/mTOR pathway. Chemotherapeutic drugs were effective both in RIBR and ABER.Discussion & Conclusion: There are various mechanisms of CDK4/6 inhibitor resistance, such as loss of RB, CDK6 amplification, and high CDK2 activity. However, the mechanism of resistance between ribociclib and abemaciclib might be different because the CDK6 level was different, although RB level was decreased in both RIBR and ABER. Furthermore, both p21 and p27 are considered as the internal inhibitors of CDK2, although more factors in RIBR and ABER were different. CDK4/6 inhibitor resistant cell lines acquired cross resistance to another CDK4/6 inhibitor. Although the mechanism of resistance between ribociclib and abemaciclib might be different, use of drugs targeting the PI3K/Akt/mTOR pathway and chemotherapeutic drugs would be an effective strategy following the development of resistance. Our findings suggest that the mechanism of resistance differs depending on the type of CDK4/6 inhibitors used, and provide some insights into the treatment strategies after acquisition of CDK4/6 inhibitor resistance. Citation Format: Masafumi Iida, Daichi Toyosawa, Misato Nakamura, Emi Tokuda, Toshifumi Niwa, Ryuichi Yoshida, Takanori Ishida, Shin-ichi Hayashi. Different mechanism of CDK4/6 inhibitor resistance between ribociclib and abemaciclib [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr PD2-04.

Presently, hormonal therapy targeting estrogen receptors is the most effective treatment available for luminal breast cancer. However, many patients relapse after the therapy. It has been suggested that cancer stem-like cells are involved with hormonal therapy resistance; in the present study, we evaluated this hypothesis. In the present study, we used our previously established hormonal therapy-resistant cell lines, including aromatase inhibitor (AI)-resistant cells (Type 1 and Type 2) and fulvestrant-resistant cells (MFR). AI-resistant cell lines expressing ER (Type 1 V1 and V2) showed high cancer stemness in terms of their CD44/CD24 expression and side populations, which were stimulated by the addition of estrogen and inhibited by fulvestrant. However, ALDH activity was lower than in the ER-negative resistant cells, suggesting that the stemness of luminal cells is distinct from that of basal-like breast cancer cells. The migration and invasion activity of the ER-positive Type 1 V1 and V2 cells were higher than in the ER-negative cell lines, Type 2 and MFR. Fractionation of parental cells based on CD44/CD24 expression and colony formation assay indicated that CD44+/CD24+ cells might be the origin of hormonal therapy-resistant cells. This population reconstituted various other subpopulations under estrogen deprivation. These results indicate that hormonal therapy resistance is closely related to the cancer stem cell-like properties of luminal breast cancer.

Hormone therapy is the most effective treatment for patients with estrogen receptor α-positive breast cancers. However, although resistance occurs during treatment in some cases and often reflects changed estrogen receptor α status, the relationship between changes in estrogen receptor α expression and resistance to therapy are poorly understood. In this study, we identified a mechanism for altered estrogen receptor α expression during disease progression and acquired hormone therapy resistance in aromatase inhibitor-resistant breast cancer cell lines. Subsequently, we investigated promoter switching and DNA methylation status of the estrogen receptor α promoter, and found marked changes of methylation at a single CpG site (CpG4) in resistant cells. In addition, luciferase reporter assays showed reduced transcriptional activity from this methylated CpG site. This CpG region was also completely conserved among species, suggesting that it acts as a methylation-sensitive Ets-2 transcription factor binding site, as confirmed using chromatin immunoprecipitation assays. In estrogen receptor α-positive tumors, CpG4 methylation levels were inversely correlated with estrogen receptor α expression status, suggesting that single CpG site plays an important role in the regulation of estrogen receptor α transcription.

Cyclin-dependent kinase (CDK) 4/6 inhibitors represent a significant advancement in the treatment of estrogen receptor (ER)-positive human epidermal growth factor receptor 2-negative advanced breast cancer. However, mechanisms of alterations after acquired resistance to CDK4/6 inhibitors and the optimal treatment options are still not established. Abemaciclib-resistant cell lines were established from the models of estrogen deprivation-resistant cell lines which retained ER expression and activated ER function derived from MCF-7 breast cancer cell lines. Ribocilib-resistant cell lines were established in the same method as previously reported. Both abemaciclib- and ribociclib-resistant cell lines showed decreased ER expression. ER transcriptional activity was maintained in these cell lines; however, the sensitivity to 4-hydroxytamoxifen and fulvestrant was almost completely lost. These cell lines did not exhibit any ERα gene mutation. Abemaciclib-resistant cell lines demonstrated low sensitivity to other CDK4/6 inhibitors; sensitivities to PI3K inhibitor, mTOR inhibitor, and chemotherapeutic drugs were maintained. Dependence on ER signaling appears to decrease after the development of acquired resistance to CDK4/6 inhibitors. Further, CDK4/6 inhibitor-resistant cells acquired cross-resistance to other CDK4/6 inhibitors, PI3K/Akt/mTOR inhibitor therapy and chemotherapeutic drugs might serve as optimal treatment options for such breast cancers.

Enzyme-linked immunosorbent assays (ELISAs) are essential for monitoring various biomarkers. Competitive and noncompetitive (sandwich) assay formats are used to determine hapten and macromolecule levels, respectively. Both formats require more sensitive detection of reporter enzymes for greater assay sensitivities. We previously reported the utility of wild-type Gaussia luciferase (wtGLuc) as a fusion partner with antibody single-chain Fv fragments (scFvs) for developing sensitive luminescent ELISAs. Here, we evaluated utility of NanoLuc luciferase (NLuc), a recently developed luciferase, as fusion partner with scFvs from the view of comparison with wtGLuc and a mutant of alkaline phosphatase (ALP’). Thyroxine (T4) and T4-labeled albumin were chosen as model haptenic and macromolecular antigens, respectively. An in-house-prepared anti-T4 scFv was fused with NLuc, wtGLuc, or ALP’. The scFv−NLuc fusion protein showed 47-fold and 29-fold lower limit of detection [LOD; 59 zmol (per assay)] than the wtGLuc- and ALP’-fusions, respectively. In a competitive T4 ELISA, the NLuc-fusion showed 9.3- and 6.3-fold lower LOD, (0.67 pg) than the wtGLuc- and ALP’-fusions, respectively, with a higher specificity in clinical applications. A typical colorimetric ELISA using a peroxidase-labeled second antibody showed 70-fold higher LOD than NLuc-based ELISA. Another advantage of the NLuc-fusion was shown in the sandwich assays; the LOD of T4-labeled albumin (5.0 fmol) was >6-fold lower than that of the other luminescent ELISAs. In an additional sandwich assay developed to count bacteriophage particles, NLuc enabled more sensitive determination than wtGLuc, whereas ALP’ showed nearly equivalent performance. Its slowest alteration rate for light intensity after starting the enzyme reaction should enable robust batch-by-batch assay operations. Thus, we concluded that scFv−NLuc fusions serve as suitable probes in various types of immunoassays and may facilitate higher sensitivities with practical specificities.

Cortisol levels in bodily fluids represent a useful index for pituitary-adrenal function, and thus practical anti-cortisol antibodies are required. We have studied "antibody-breeding" approaches, which involve in vitro evolution of antibodies to improve their antigen-binding performances. Here, we produced an antibody fragment to measure serum cortisol levels with over 30-fold enhanced affinity after single mutagenesis and selection steps. A mouse anti-cortisol antibody, Ab-CS#3, with insufficient affinity for practical use, was chosen as the prototype antibody. A "wild-type" single-chain Fv fragment (wt-scFv; Ka, 3.4×108 M-1) was prepared by bacterial expression of a fusion gene combining the VH and VL genes for this antibody. Then, random point mutations were generated separately in VH or VL by error-prone PCR, and the resulting products were used to assemble scFv genes, which were displayed on filamentous phages. Repeated panning of the phage library identified a mutant scFv (scFv#m1-L10) with an over 30-fold enhanced affinity (Ka 1.2×1010 M-1). Three amino acid substitutions (Cys49Ser, Leu54Pro, and Ser63Gly) were observed in its VL sequence. In a competitive enzyme-linked immunosorbent assay (ELISA), the mutant scFv generated dose-response curves with measuring range ca. 0.03-0.6 ng/assay cortisol, midpoint of which (0.15 ng/assay) was 7.3-fold lower than that of wt-scFv. Although cortisone, 11-deoxycortisol, and prednisolone showed considerable cross-reactivity, the mutant scFv should enable sensitive routine cortisol assays, except for measurement after metyrapone or high-dose of prednisolone administrations. Actually, cortisol levels of control sera obtained with the scFv-based ELISA were in the reference range.

Hormone therapy targeting the estrogen receptor (ER) pathway is the most common treatment used for ER-positive breast cancer. However, some patients experience de novo or acquired resistance, which becomes a critical problem. Activation of the insulin-like growth factor (IGF) pathway allows breast cancer cells to proliferate and is associated with the ER pathway. Little is known about the role of the IGF pathway in hormone therapy and resistance; therefore, we investigated whether the inhibition of this pathway may represent a novel therapeutic target for overcoming hormone therapy resistance in ER-positive breast cancers. Crosstalk between the ER and IGF pathways was analyzed in breast cancer cell lines by inhibiting or stimulating either one or both pathways. We studied the effect of insulin-like growth factor one receptor (IGF1R) inhibition in aromatase inhibitor-resistant breast cancer cell lines and fulvestrant-resistant cell lines which were uniquely established in our laboratory. Under normal conditions, IGF signaling is controlled by ER signaling to promote cell growth. Temporary disruption of the estrogen supply results in attenuated ER signaling, and IGF-1 dramatically increased relative growth compared with normal conditions. In addition, IGF1R inhibitor strongly suppressd cell growth in hormone-resistant breast cancer cells where ER remains than cells where ER decreased or was almost lost. Our study suggests that inhibition of the IGF pathway may be an effective strategy for ER-positive breast cancer therapy, even in hormone therapy-resistant cases.

In this study, we show the utility of Gaussia luciferase (GLuc), which is much smaller than previously found luciferases, as the fusion partner with artificial antibody species for developing sensitive immunoassay systems. As an example, we constructed a bioluminescent enzyme-linked immunosorbent assay (BL-ELISA) system determining the major glucocorticoid cortisol. A monoclonal antibody was newly elicited against a cortisol-albumin conjugate, and the genes encoding its variable domains (VH and VL) were cloned and combined to encode a single-chain Fv fragment (scFv). scFv was then linked to the wild-type GLuc gene or that encoding GLuc mutants reported to show improved emission kinetics and expressed in the periplasmic space of several Escherichia coli strains. Notably, the wild-type GLuc fusion protein (scFv-wtGLuc) showed the most suitable luminescent properties for BL-ELISAs. In our system, scFv-wtGLuc was reacted competitively with the analyte and immobilized cortisol moieties, and the bound GLuc activity was monitored with coelenterazine as the substrate. Successful batch-type luminescence detection was achieved using a plate reader without built-in injectors. The midpoint and limit of detection in a typical dose-response curve were 4.1 and 0.26 pg/assay, respectively, thus exhibiting much more sensitivity than conventional cortisol immunoassays. Serum cortisol levels (as the sum with cortisone) for healthy subjects, determined without any pretreatment, were compatible with reported reference ranges. The scFv-wtGLuc probe was stable over a year under storage as periplasmic extracts at -30 °C or with repeated freeze-thawing. These results suggest that GLuc fusions with antibody fragments might serve as useful and highly sensitive immunoassay probes in various clinical settings.

Aromatase inhibitors (AI) are commonly used to treat postmenopausal estrogen-receptor (ER)-positive breast carcinoma. However, resistance to AI is sometimes acquired, and the molecular mechanisms underlying such resistance are largely unclear. Recent studies suggest that AI treatment increases androgen activity during estrogen deprivation in breast carcinoma, but the role of the androgen receptor (AR) in breast carcinoma is still a matter of controversy. The purpose of this study is to examine the potential correlation between the AR- and AI-resistant breast carcinoma. To this end, we performed immunohistochemical analysis of 21 pairs of primary breast carcinoma and corresponding AI-resistant recurrent tissue samples and established two stable variant cell lines from ER-positive T-47D breast carcinoma cell line as AI-resistance models and used them in in vitro experiments. Immunohistochemical analysis demonstrated that the expression of prostate-specific antigen (PSA) and Ki-67 were significantly higher and ER and progesterone receptor (PR) were lower in recurrent lesions compared to the corresponding primary lesions. Variant cell lines overexpressed AR and PSA and exhibited neither growth response to estrogen nor expression of ER. Androgen markedly induced the proliferation of these cell lines. In addition, the expression profile of androgen-induced genes was markedly different between variant and parental cell lines as determined by microarray analysis. These results suggest that in some cases of ER-positive breast carcinoma, tumor cells possibly change from ER-dependent to AR-dependent, rendering them resistant to AI. AR inhibitors may thus be effective in a selected group of patients.

The acquisition of estrogen-deprivation resistance and estrogen receptor (ER) signal-independence in ER-positive breast cancer is one of the crucial steps in advancing the aggressiveness of breast cancer; however, this has not yet been elucidated in detail. To address this issue, we established several estrogen-deprivation-resistant (EDR) breast cancer cell lines from our unique MCF-7 cells, which had been stably transfected with an ERE-GFP reporter plasmid. Three cell lines with high ER activity and another 3 cell lines with no ER activity were established from cell cloning by monitoring GFP expression in living cells. The former three ERE-GFP-positive EDR cell lines showed the overexpression of ER and high expression of several ER-target genes. Further analysis of intracellular signaling factors revealed a marked change in the phosphorylation status of ERα on Ser167 and Akt on Thr308 by similar mechanisms reported previously; however, we could not find any changes in MAP-kinase factors. Comprehensive phospho-proteomic analysis also indicated the possible contribution of the Akt pathway to the phosphorylation of ERα. On the other hand, constitutive activation of c-Jun N-terminal kinase (JNK) was observed in ERE-GFP-negative EDR cells, and the growth of these cells was inhibited by a JNK inhibitor. An IGF1R-specific inhibitor diminished the phosphorylation of JNK, which suggested that a novel signaling pathway, IGF1R-JNK, may be important for the proliferation of ER-independent MCF-7 cells. These results indicate that ER-positive breast cancer cells can acquire resistance by more than two mechanisms at a time, which suggests that multiple mechanisms may occur simultaneously. This finding also implies that breast cancers with different resistance mechanisms can concomitantly occur and mingle in an individual patient, and may be a cause of the recurrence of cancer.

[Background] Estrogen receptor (ER) positive breast cancer was composed of about 70% of clinical breast cancer and therapy for the ER positive breast cancer have progressed, but some patients recur after endocrine therapy including aromatase inhibitors (AIs). The switch of proliferation drive to human epidermal growth factor receptor 2 was reported as one of AI resistance mechanisms, but ER status of many recurrent patients after AI treatment was maintained or slightly decreased. Moreover, plasma concentration of estrone-sulfate (E1S) is mostly unchanged after menopause and treatment of AI. Thus, we established AI resistant model with conserved ER expression and its activity from MCF-7 cell, and found that E1S contributed to proliferation of the AI-resistant cells. [Material and Method] We introduced plasmids carrying ERE-green fluorescent protein (GFP) gene to MCF-7 (E10), and then aromatase gene was additionally introduced to E10 (E10arom). E10arom was cultured in estrogen-depleted medium including testosterone and letrozole, monitoring GFP as ER activity for about three months. We finally established ER positive Letrozole-Resistant cell lines (LR cells). In addition, to confirm whether E1S-dependent growth in LR cells could be reproduced in ER positive breast cancer cell line, we also established human steroid sulfatase (STS) gene-transfected MCF-7 (MCF-STS) and organic anion transporter peptide (OATP) 4A1 gene-transfected MCF-7 (MCF-4A1). ER activity of these cell lines was examined by ERE-luciferase reporter assay, and response to each steroid and drug was investigated by cell proliferation assay. Quantitative PCR was used for comparing expression of mRNA in these cell lines. [Results] mRNA expression of STS and four OATPs including OATP4A1 increased in LR cells comparing to E10arom. Moreover, there was no change in mRNA expressions of efflux drug transporters, aromatase-metaboic enzyme and enzymes related with production for estrogenic androgen. ERE-luciferase assay with E1S showed higher ER activity of LR cells than E10arom, and LR cells were more proliferated by E1S in dose-dependent manner than the parental cell. STS inhibitor alone and its combination with letrozole were inhibitory to proliferation of LR cells. LR cells were also inhibited by selective estrogen receptor modulator (SERM) and downregulator (SERD). In proliferation assay of MCF-STS and MCF-4A1, we observed that these cell lines also proliferated by E1S in dose-dependent manner as LR cells did. [Discussion] Since ER status of recurrent patients after AI treatment was mostly unchanged, our model could be more practical for analysis of AI resistance in ER positive breast cancer. These results showed that combined induction of STS and OATP contributed to estradiol (E2) production from E1S in LR cells, suggesting that analysis of E1S as alternative E2 productive pathway needed investigation of both STS and transporter of E1S. The combination of STS inhibitor with letrozole was effective to LR cells, and SERM and SERD were also effective to these cells, suggesting that combined therapy of STS inhibitor with AI or switching AI to SERM or SERD was beneficial for AI-resistant breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-09-16.

Aromatase inhibitors (AI) are the common treatment for postmenopausal estrogen receptor (ER) positive breast cancer. However, not a few of these patients acquire resistance to AI.To investigate this molecular mechanism of resistance, we have established several breast cancer cell lines as AI resistant models. Recently, we reported on AD-EDR cells (androgen metabolite-dependent and estrogen depletion resistant cells) derived from MCF-7, in which ER activity depends on one of androgen metabolite, 5a-androstane-3β, 17β-diol (3βdiol). It is supposed that tumor heterogeneity contributes to the acquired resistance to drugs.To examine this diversity, we established another AI resistant model derived from T-47D, to compare characters of these AI resistant models from different parents. Then, we found that the androgen receptor (AR) signal acquired oncogenic role in this model in contrast to AD-EDR cells established from MCF-7. We cultured ER positive breast cancer cells, T47D-TE8, stably transfected with ERE-GFP reporter into T47D, under estrogen-depleted and androgen-supplemented conditions for more than three months, simulating the AI treatment, and established estrogen deprivation-resistant cells (TE8-EDR cells). TE8-EDR cells had little ERE-GFP activity and proliferation response to estrogen though estrogen stimulated proliferation of parental T47D-TE8. The mRNA for ER and progesterone receptor was hardly detected in TE8-EDR cells. These results suggested that they had lost the ER-mediated pathway of growth. On the contrary, Dihydrotestosterone (DHT) increased growth of T47D-TE8 cells, while it was more remarkable in TE8-EDR, and this proliferative effect was reduced by AR antagonist, bicalutamide. The mRNA expressions of AR and its target gene, PSA, were extremely increased in TE8-EDR cells. These results indicated that TE8-EDR enhanced AR-regulated pathway for growth. Increased AR expression was presumed to be one of the factors elevating AR activity. To identify other elements enhancing AR-mediated signal, we compared the gene expression of TE8-EDR cells with that of T47D-TE8 by microarray analysis. The outcome showed that the expression of some well-known androgen-related genes were increased in TE8-EDR cells. Among these genes, we focused on DDC (L-dopa decarboxylase) which has been reported as one of the AR coactivator, and its overexpression increase proliferative effect in prostate cancer cell lines.By real-time RT-PCR, we confirmed that the mRNA expression of DDC in TE8-EDR was higher over one-hundred fold than T47D-TE8. In addition, DHT-induced proliferation was cancelled by DDC inhibitor, NSD-1015. This study suggests that TE8-EDR cells shift their growth pathway from ER to AR signal, and the AR activity is enhanced by high-expression of DDC. Whereas, MCF7-derived AD-EDR cells still depends on ER signal for their proliferation. So far, few studies have shown the proliferative effects of androgen as AI resistant mechanism. Our results provide new mechanism of acquired AI-resistance in breast cancer. Moreover, the fact that the character of AI resistant models varied with their parent cells supports the hypothesis that tumor heterogeneity contributes to diversify the mechanism of acquired resistance to hormonal therapy. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P5-09-10.

Most immunoassays use probes that convert concentrations of analytes into signal intensity. To prepare the probes, analytes are usually linked to a reporter protein (e.g., enzymes) with the aid of chemical reagents. However, these conventional methods yield a mixture of heterogeneous products and consequently reduce assay performance. "Clonable" homogeneous probes, i.e., recombinant molecules in which a target protein is genetically fused to a reporter with a defined coupling ratio, are now available for analyzing protein biomarker concentrations. Here, we have expanded this strategy to measure small biomarkers (haptens) by genetically fusing proteinaceous molecules that mimic target haptens with enzymes. 11-Deoxycortisol (11-DC) was chosen as the model hapten, and the β-type anti-idiotype antibodies (βId-Abs) that recognize the paratope of anti-hapten antibodies were used as the target hapten mimic. The V(H) and V(L) genes of a βId-Ab, targeting a mouse anti-11-DC antibody (CET-M8), were assembled to encode a single-chain Fv fragment (βId-scFv), which was then fused with a gene encoding a variant of alkaline phosphatase. The product, βId-scFv-ALP' protein, had satisfactory enzyme activity and bound to CET-M8 in a competitive manner with 11-DC. A colorimetric enzyme-linked immunosorbent assay (ELISA) for 11-DC, based on the competitive reaction between the analyte and βId-scFv-ALP' against immobilized CET-M8, was found to be sensitive (limit of detection = 22 pg/assay) and specific (cross-reactivity with cortisol, 0.24%) for clinical use and could be used to determine serum 11-DC levels after a simple solvent extraction. The anti-idiotype scFv-enzyme fusion proteins proposed here can be prepared reproducibly as homogeneous products with a 1:1 coupling ratio and would facilitate standardization of immunoassays for small biomarkers.

Aromatase inhibitors (AIs) have been reported to exert their antiproliferative effects in postmenopausal women with hormone receptor-positive breast cancer not only by reducing estrogen production but also by unmasking the inhibitory effects of androgens such as testosterone (TS) and dihydrotestosterone (DHT). However, the role of androgens in AI-resistance mechanisms is not sufficiently understood. 5α-Androstane-3β,17β-diol (3β-diol) generated from DHT by 3β-hydroxysteroid dehydrogenase type 1 (HSD3B1) shows androgenic and substantial estrogenic activities, representing a potential mechanism of AI resistance. Estrogen response element (ERE)-green fluorescent protein (GFP)-transfected MCF-7 breast cancer cells (E10 cells) were cultured for 3 months under steroid-depleted, TS-supplemented conditions. Among the surviving cells, two stable variants showing androgen metabolite-dependent ER activity were selected by monitoring GFP expression. We investigated the process of adaptation to androgen-abundant conditions and the role of androgens in AI-resistance mechanisms in these variant cell lines. The variant cell lines showed increased growth and induction of estrogen-responsive genes rather than androgen-responsive genes after stimulation with androgens or 3β-diol. Further analysis suggested that increased expression of HSD3B1 and reduced expression of androgen receptor (AR) promoted adaptation to androgen-abundant conditions, as indicated by the increased conversion of DHT into 3β-diol by HSD3B1 and AR signal reduction. Furthermore, in parental E10 cells, ectopic expression of HSD3B1 or inhibition of AR resulted in adaptation to androgen-abundant conditions. Coculture with stromal cells to mimic local estrogen production from androgens reduced cell sensitivity to AIs compared with parental E10 cells. These results suggest that increased expression of HSD3B1 and reduced expression of AR might reduce the sensitivity to AIs as demonstrated by enhanced androgen metabolite-induced ER activation and growth mechanisms. Androgen metabolite-dependent growth of breast cancer cells may therefore play a role in AI-resistance.

Approximately 70% of breast cancers express estrogen receptor α (ERα), which plays critical roles in breast cancer development. Fulvestrant has been effectively used to treat ERα-positive breast cancer, although resistance remains a critical problem. To elucidate the mechanism of resistance to fulvestrant, we established fulvestrant-resistant cell-lines named MFR (MCF-7 derived fulvestrant resistance) and TFR (T-47D derived fulvestrant resistance) from the ERα-positive luminal breast cancer cell lines MCF-7 and T-47D, respectively. Both fulvestrant-resistant cell lines lost sensitivity to estrogen and anti-estrogens. We observed diminished ERα expression at both the protein and mRNA levels. To address the mechanism of gene expression regulation, we examined epigenetic alteration, especially the DNA methylation level of ERα gene promoters. MFR cells displayed high methylation levels upstream of the ERα gene, whereas no change in DNA methylation was observed in TFR cells. Hence, we examined the gene expression plasticity of ERα, as there are differences in its reversibility following fulvestrant withdrawal. ERα gene expression was not restored in MFR cells, and alternative intracellular phosphorylation signals were activated. By contrast, TFR cells exhibited plasticity of ERα gene expression and ERα-dependent growth; moreover, these cells were resensitized to estrogen and anti-estrogens. The difference in epigenetic regulation among individual cells might explain the difference in the plasticity of ERα expression. We also identified an MFR cell-activating HER/Src-Akt/MAPK pathway; thus, the specific inhibitors effectively blocked MFR cell growth. This finding implies the presence of multiple fulvestrant resistance mechanisms and suggests that the optimal therapies differ among individual tumors as a result of differing epigenetic mechanisms regulating ERα gene expression.

Background: CDK4/6 inhibitors have received FDA breakthrough therapy designation as 1st line treatment for advanced estrogen receptor positive breast cancer patients. However, the benefit offered by CDK4/6 inhibitors is individually different and furthermore acquired resistance to the drugs is monumental challenges. It is urgent need to search for the biomarker and understand the drug sensitivity and its alteration after acquired resistance. Results: To identify the efficacy of CDK4/6 inhibitors, we assessed IC50 of ribociclib in several cell lines. Luminal cell lines (MCF-7, T-47D) exhibited lower ribociclib IC50 than HER-2 (SK-BR-3) and triple negative cell lines (MDA-MB-231, BT-20). Immunoblot analysis of Luminal cell lines showed extremely lower levels of CDK6 compared with others. CDK6-transfected MCF-7 by means of expression vector reduced the sensitivity equivalent to MDA-MB-231 not only to ribociclib but also to palbociclib and abemaciclib. Protein level of ERα in CDK6-transfected MCF-7 stayed unchanged and fulvestrant sensitivity was unaltered as well. Subsequently, we detect the efficacy of ribociclib in hormone resistant cell lines. Estrogen deprivation-resistant (EDR) cells (EDR1:ER-positive, EDR2:ER-negative) and fulvestrant resistance (MFR) cells (loss of ER expression) established from MCF-7 maintained ribocilib sensitivity to the same degree with MCF-7. No marked difference in IC50 was observed between EDR1/2 and MFR, and CDK6 expressions were comparable to MCF-7. These results suggest that high level of CDK6 expression weaken the sensitivity to CDK4/6 inhibitors. The inhibitors would provide more effective benefits to tumors expressing lower level of CDK6 than the higher, independent of hormone sensitivity. To understand the characteristics in acquired resistance, ribociclib resistant cell lines (RIBR1/2) were established from EDR1 by long-term culture with ribociclib. RIBR designed lower level of p21, p27 and ERα by immunoblot analysis. EDR1 were promoted cell growth by estrogen addition, while RIBR were not. Further, ER activity of RIBR was intensely decreased, and mRNA levels of the ER target genes, PgR and EGR3 were also decreased. Therefore, the responsiveness to tamoxifen and fulvestrant were lost. On the other hand, PI3K inhibitor and mTOR inhibitor suppressed cell growth to the same extent as EDR1, suggesting that RIBR were reduced ER dependence and remain reliant on PI3K/Akt/mTOR pathway. Conclusion: The possibility of CDK6 as a biomarker is corroborated by the finding that low level of CDK6 expression is positively correlated with the efficacy of CDK4/6 inhibitor. Further ER dependence had decreased after acquired CDK4/6 inhibitor resistance whereas the dependence on PI3K/Akt/mTOR pathway still remain, indicating the inhibition of PI3K/Akt/mTOR pathway would be amenable to therapeutic target. Citation Format: Iida M, Nakamura M, Tokuda E, Niwa T, Ishida T, Hayashi S-I. CDK6 might be a key factor for efficacy of CDK4/6 inhibitor and the hormone sensitivity following acquired resistance [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-04-02.

Background Mutations in phosphatidylinositol 3-kinase (PI3K), which encodes the catalytic subunit of PI3Kα, is one of the most frequent genomic alterations and is found in about 40% of estrogen receptor (ER)-positive, HER2 negative breast cancers. PIK3CA mutations promote the growth and proliferation of cancers via activation of the PI3K/mammalian target of rapamycin pathway and can mediate resistance to endocrine therapies in breast cancer. Although several clinical trials for PI3K inhibitors (PI3Kis) in ER-positive metastatic breast cancers are ongoing, the effects of PI3Kis on endocrine-resistant breast cancers with various characteristics and definitive biomarkers of PI3Kis are unclear. Using endocrine-resistant cells established in our laboratory, we evaluated the efficacy of PI3Ki in these cell lines and identified the characteristics associated with acquired resistance to PI3Kis in endocrine-resistant cells. Results Long-term estrogen deprivation-resistant (EDR) cell lines and fulvestrant-resistant cell lines (MFR and TFR) were established from MCF-7 and T-47D cells in our previous studies. These cell lines showed different ER expression levels, including high expression (EDR-1), low expression (EDR-2, -3), or no expression (MFR, TFR); all of these cell lines had the same PIK3CA mutations as the parental cell lines. The pan-class1 PI3Ki buparlisib (BKM120) and α-specific PI3K inhibitor alpelisib (BYL719) inhibited the proliferation of endocrine-resistant cell lines when compared with their parental cells. Among endocrine-resistant cells, MFR cells were dramatically inhibited by PI3Kis. Colony formation assays indicated that MFR cells were more sensitive to PI3Kis than other cells lines. Next, we established PI3Ki- and everolimus-resistant cell lines from EDR-1, EDR-2, and MFR cells using BKM120, BYL719, and everolimus. Analysis of the time required to generate resistant cells showed that MFR cells required twice as long to acquire resistance compared with EDR cells. Furthermore, the time required to acquire resistance to BYL719 was shorter than that for BKM120. BYL719-resistant (BYL-R) cells were effectively inhibited by BKM120 to a degree similar to that of parental cells; however, BYL-R cells lost sensitivity to BYL719 and everolimus. Evelolimus-resistant (EVE-R) cells were also the same. In contrast, BKM120-resistant (BKM-R) cells showed less sensitivity to BKM120, BYL719, and everolimus. In other words, the pan-PI3Ki BKM120 was able to inhibit the growth of BYL-R and EVE-R cells, whereas BYL719 and everolimus were not able to inhibit BKM-R cells sufficiently. In addition, there were no changes in ER expression in EDR-1, EDR-2, and MFR cells exposed to PI3Kis for 1 h. Interestingly, ER expression on EDR-2 cells with acquired resistance to PI3Kis was increased compared with that in parental cells. Conclusion Our findings showed that PI3Kis exhibited remarkable efficacy in all types of EDR cells, particularly in fulvestrant-resistant cell lines. In PI3Ki- and everolimus-resistant cell lines, BKM120 and BYL719 showed different effects, and BYL719 and everolimus may exhibit cross-resistance. Furthermore, PI3Kis were likely to change the expression of ER. Citation Format: Tokuda E, Komatsu T, Sasaki S, Tsuboi K, Nakamura M, Iida M, Niwa T, Saito M, Hayashi S-I. Effects of PI3K inhibitors on endocrine-resistant cell lines and differences in the characteristics of ER positive breast cancer cells after acquired resistance to the inhibitors [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P4-03-12.

Cortisol levels in body fluids are useful for monitoring the function of the pituitary-adrenal axis. Here, we established an "enzyme-linked immunometric assay" (a noncompetitive-type ELISA) for cortisol based on idiotype-anti-idiotype reactions. Six different anti-idiotype monoclonal antibodies that recognized the variable regions of a newly established anti-cortisol antibody were generated using hybridoma technology; these were two beta-type and four alpha-type anti-idiotype antibodies, recognizing the paratope and framework regions, respectively. An immunometric assay was established using a combination of a selected alpha-type and a selected beta-type antibody. The analyte (cortisol) was captured by an excess amount of anti-cortisol antibody immobilized on microplates, and the unoccupied paratope was saturated with the beta-type antibody. Hapten-occupied anti-cortisol antibody, with less steric hindrance, was then selectively bound by the alpha-type antibody, labeled with biotin. The amount of biotin residue on the microplates was colorimetrically monitored using a peroxidase-labeled streptavidin. This assay had an approximately threefold higher sensitivity (detection limit: 90 pg = 248 fmol cortisol) than a competitive ELISA using the same anti-cortisol antibody, as well as a practical specificity for providing reasonable determination of normal urinary cortisol levels.

A single-chain Fv fragment (scFv) against estradiol-17beta (E(2)) was generated to begin the construction of a library of various mutated anti-steroid antibodies with an improved affinity and/or specificity. A hybridoma clone secreting a specific anti-E(2) antibody (Ab#E4-4) was established by the cell fusion using splenocytes from a mouse immunized with an immunogenic E(2)-carrier conjugate. DNA fragments encoding the variable heavy and light domains (V(H) and V(L)) of the Ab#E4-4 were cloned and combined to give the scFv gene fragment encoding the sequence 5'-V(H)-(GGGGS)(3)-V(L)-3'. Compared to the Ab#E4-4 Fab fragment, soluble scFv (scFv#E4-4) protein showed a similar affinity to E(2) (K(a)=8.6x10(7)M(-1)) and a similar cross-reaction profile. To further study the fundamentals for creating a comprehensive library of mutated scFvs, the scFvV(H) and V(L) genes were amplified using error-prone PCR conditions and the frequency and pattern of incorporated mutations were investigated. For this, regular Taq polymerase was used in the presence of unequal concentrations of dNTPs. At 1.0mM MnCl(2), the error frequency reached to 8.5% and 11% for the V(H) and V(L) respectively, although a significant transition/transversion bias was observed. ScFv#E4-4 and the mutated polyclonal scFvs were then displayed on filamentous phage under various packaging conditions. Cultivation of the transformed bacteria was more suitable at 25 degrees C than at higher temperatures for the packaging of scFv-bearing phagemid particles. Based on these experimental conditions, an scFv-displaying phage library, each scFv member in which has mutated complementarity-determining region (CDR) H2, H3, L1, and L3, was constructed. A soluble scFv clone (scFv#m1-e7) with a mutated amino acid (I--V) in CDR L1, isolated from this library, showed threefold higher affinity (K(a)=2.6 x 10(8)M(-1)) than that of scFv#4-4.

To overcome the sensitivity limit in immunoassays for small molecules (haptens), we established a noncompetitive immunoenzymometric assay (IEMA) format that can detect attomole-range hapten molecules. We selected 11-deoxycortisol (11-DC; Mr 346.5), a corticosteroid serving a diagnostic index for pituitary-adrenal function, as a model target hapten. A fusion of a single-chain Fv fragment (scFv) specific for 11-DC and alkaline phosphatase (ALP) was generated for use as an enzyme-labeled antibody, instead of the conventional chemically linked enzyme-antibody conjugates. After binding reaction of 11-DC and fixed amounts of the fusion protein (scFv-ALP), the unbound fusion protein was removed by incubation with a mouse beta-type anti-idiotype antibody recognizing the scFv paratope. These complexes were captured by magnetic separation using anti-mouse IgG antibody-coated magnetic beads. Following magnetic sedimentation of the beads, immune complexes of scFv-ALP and 11-DC remained in the supernatant were further purified by capture on microtiter plates with immobilized alpha-type anti-idiotype antibody. As measured fluorometrically, ALP activity from bound immune complexes on the plates increased with increasing 11-DC, which is characteristic of a noncompetitive relationship. This IEMA afforded an extremely low detection limit (20 amol/assay), a very wide measurable range, and practical specificity. The plasma 11-DC levels determined for healthy subjects were validated as reliable.

Anti-idiotype antibodies recognizing the variable regions of a particular anti-hapten antibody are valuable tools, which can be used in sensitive hapten immunoassays based on a noncompetitive format. Here, we describe the production and characterization of monoclonal anti-idiotype antibodies against idiotopes on the variable regions of an antibody showing high affinity and specificity to 11-deoxycortisol (11-DC). 11-DC is the biosynthetic precursor of cortisol and a diagnostic index for the assessment of pituitary-adrenal function. BALB/c or A/J mice were repeatedly immunized with the anti-11-DC antibody conjugated with keyhole limpet hemocyanin and their spleen cells were then fused with P3/NS1/1-Ag4-1 myeloma cells. Seven kinds of anti-idiotype antibodies were generated, one of which was a beta-type antibody recognizing the paratope and others which were alpha-type antibodies recognizing the framework region. A noncompetitive ELISA based on idiotype-anti-idiotype reactions was established using one of these alpha-type antibodies in combination with the beta-type antibody and with the anti-11-DC antibody. This noncompetitive assay system provided improved sensitivity (detection limit: 1.0 pg=2.9 fmol), which is approximately 10 times higher than the corresponding competitive enzyme immunoassay, and offered a practical specificity for clinical use. Appropriate serum 11-DC levels were obtained for normal subjects [0.16+/-0.09 (S.D.) microg/l (n=6), ranging from 0.086 to 0.316 microg/l] using the present assay system.

A combined method of immunoaffinity extraction with high-performance liquid chromatography has been developed for the enantioselective determination of bufuralol and its metabolites in human plasma. The antibodies having high affinity toward the asymmetric center at the C-1 position of bufuralol and its 1′-oxidized metabolites and low affinity to their antipodes were elicited by immunization of rabbits with immunogens, (1 R )- and (1 S )-1′-oxobufuralol O -carboxymethyl oxime-bovine serum albumin conjugates, respectively. 0.5 ml Of the immunoaffinity adsorbent (7.6 mg·ml −1 for anti-(1 S )-antibody and 28.8 mg·ml −1 for anti-(1 R )-antibody) prepared by immobilization of an antibody was capable of retaining up to 1 μg of ( R )- and ( S )-bufuralol and up to 500 ng of other metabolites. The adsorbates were recovered quantitatively by elution with methanol–10 mM ammonium acetate buffer (pH 5) (95:5, v/v) without any interfering peaks on the high-performance liquid chromatogram. The proposed method was evaluated to be useful for the simultaneous determination of optically active bufuralol and its metabolite in plasma with acceptable recovery and precision.

Aromatase inhibitors (AIs) exert antiproliferative effects by reducing local estrogen production from androgens in postmenopausal women with hormone-responsive breast cancer. Previous reports have shown that androgen metabolites generated by the aromatase-independent enzymes, 5α-androstane-3β, 17β-diol (3β-diol), androst-5-ene-3β, and 17β-diol (A-diol), also activate estrogen receptor (ER) α. Estradiol (E2) can also reportedly be generated from estrone sulfate (E1S) pooled in the plasma. Estrogenic steroid-producing aromatase-independent pathways have thus been proposed as a mechanism of AI resistance. However, it is unclear whether these pathways are functional in clinical breast cancer. To investigate this issue, we assessed the transcriptional activities of ER in 45 ER-positive human breast cancers using the adenovirus estrogen-response element-green fluorescent protein assay and mRNA expression levels of the ER target gene, progesterone receptor, as indicators of ex vivo and in vivo ER activity, respectively. We also determined mRNA expression levels of 5α-reductase type 1 (SRD5A1) and 3β-hydroxysteroid dehydrogenase type 1 (3β-HSD type 1; HSD3B1), which produce 3β-diol from androgens, and of steroid sulfatase (STS) and 17β-hydroxysteroid dehydrogenase type 1 (17β-HSD type 1; HSD17B1), which produce E2 or A-diol from E1S or dehydroepiandrosterone sulfate. SRD5A1 and HSD3B1 expression levels were positively correlated with ex vivo and in vivo ER activities. STS and HSD17B1 expression levels were positively correlated with in vivo ER activity alone. Elevated expression levels of these steroid-metabolizing enzymes in association with high in vivo ER activity were particularly notable in postmenopausal patients. Analysis of the expression levels of steroid-metabolizing enzymes revealed positive correlations between SRD5A1 and HSD3B1, and STS and HSD17B1. These findings suggest that the SRD5A1-HSD3B1 as well as the STS-HSD17B pathways, could contributes to ER activation, especially postmenopause. These pathways might function as an alternative estrogenic steroid-producing, aromatase-independent pathways.

Conventional immunization-based antibodies are not always applicable to monitor trace amounts of clinical biomarkers due to insufficient antigen-binding affinity. The development of in vitro affinity maturation for native antibodies has paved the way to solve this problem, but there has been minimal success in obtaining improved antibody mutants for small molecule biomarkers (haptens). We here describe a "molecular breeding" approach, in which a practical anti-hapten antibody mutant has successfully been generated from a prototype antibody with poor binding properties. Random point mutations were introduced by error-prone polymerase chain reaction in the V(H) and V(L) genes of a second-generation mutant of a single-chain Fv fragment (scFv) that binds estradiol-17β (E2), whose parent mouse Fab fragment showed poor affinity for E2 (Ka = 5.2 × 10(7) M(-1)). The resulting scFv gene library was expressed on filamentous phage particles. Phage clones with strong E2 binding were isolated with dissociation-independent methods using newly developed reagents. The binding characteristics and clinical applicability of the soluble scFvs prepared from the selected clones were examined in a competitive enzyme-linked immunosorbent assay (ELISA). We finally obtained a third-generation scFv mutant, scFv#m3-a18 (11 amino acid substitutions), which showed ~250-fold greater Ka (1.3 × 10(10) M(-1)) than the parent Fab, and yielded sensitive ELISA dose-response curves for E2 (limit of detection0.5 pg/assay) with practical specificity. The average human serum E2 levels, determined after acetonitrile extraction, were compatible with reported reference ranges. The present results will prompt a new era for preparing diagnostic reagents.

A specific enzyme immunoassay (EIA) for ursodeoxycholic acid 7-N acetylglucosaminides (UDCA-7-NAGs), novel metabolites of ursodeoxycholic acid, has been developed. The EIA was based upon antisera elicited against the glycine conjugated UDCA-7-NAG-bovine serum albumin conjugate and horseradish peroxidase-labeled-UDCA-7-NAG. The sensitivity was significantly improved by the utilization of the bridge heterologous combination system. An appropriate dose response curve of EIA for glycine conjugated UDCA-7-NAG was obtained in the range of 0.3-6ng/tube. The specificity of ETA proved to be satisfactory in terms of cross-reactivities to related bile acids. The proposed method is useful for the determination of UDCA-7-NAGs in human urine with acceptable reliability.

Aromatase inhibitors (AIs) effectively treat hormone receptor-positive postmenopausal breast cancer, but some patients do not respond to treatment or experience recurrence. Mechanisms of AI resistance include ligand-independent activation of the estrogen receptor (ER) and signaling via other growth factor receptors; however, these do not account for all forms of resistance. Here we present an alternative mechanism of AI resistance. We ectopically expressed aromatase in MCF-7 cells expressing green fluorescent protein as an index of ER activity. Aromatase-overexpressing MCF-7 cells were cultured in estrogen-depleted medium supplemented with testosterone and the AI, letrozole, to establish letrozole-resistant (LR) cell lines. Compared with parental cells, LR cells had higher mRNA levels of steroid sulfatase (STS), which converts estrone sulfate (E1S) to estrone, and the organic anion transporter peptides (OATPs), which mediate the uptake of E1S into cells. LR cells proliferated more in E1S-supplemented medium than did parental cells, and LR proliferation was effectively inhibited by an STS inhibitor in combination with letrozole and by ER-targeting drugs. Analysis of ER-positive primary breast cancer tissues showed a significant correlation between the increases in the mRNA levels of STS and the OATPs in the LR cell lines, which supports the validity of this AI-resistant model. This is the first study to demonstrate the contribution of STS and OATPs in E1S metabolism to the proliferation of AI-resistant breast cancer cells. We suggest that E1S metabolism represents a new target in AI-resistant breast cancer treatment.

To predict the efficacy of hormonal therapy at the individual-level, immunohistochemical methods are used to analyze expression of classical molecular biomarkers such as estrogen receptor (ER), progesterone receptor (PgR), and HER2. However, the current diagnostic standard is not perfect for the individualization of diverse cases. Therefore, establishment of more accurate diagnostics is required. Previously, we established a novel method that enables analysis of ER transcriptional activation potential in clinical specimens using an adenovirus estrogen response element–green fluorescence protein (ERE-GFP) assay system. Using this assay, we assessed the ERE transcriptional activity of 62 primary breast cancer samples. In 40% of samples, we observed that ER protein expression was not consistent with ERE activity. Comparison of ERE activity with clinicopathological information revealed that ERE activity was significantly correlated with the ER target gene, PgR, rather than ER in terms of both protein and mRNA expression. Moreover, subgrouping of Luminal A-type breast cancer samples according to ERE activity revealed that ERα mRNA expression correlated with ER target gene mRNA expression in the high-, but not the low-, ERE-activity group. On the other hand, the low-ERE-activity group showed significantly higher mRNA expression of the malignancy biomarker Ki67 in association with disease recurrence in 5% of patients. Thus, these data suggest that ER expression does not always correlate with ER transcriptional activity. Therefore, in addition to ER protein expression, determination of ERE activity as an ER functional marker will be helpful for analysis of a variety of diverse breast cancer cases and the subsequent course of treatment.

Measurement of cortisol levels in body fluids is important for monitoring pituitary gland and adrenal functions. To develop a specific and standardized enzyme-linked immunosorbent assay (ELISA), a novel monoclonal anti-cortisol antibody has been generated using a reasonably designed haptenic derivative. Spleen cells were prepared from the BALB/c or A/J mouse, which had repeatedly been immunized with a conjugate of 4-(2-carboxyethylthio)cortisol (CET) and bovine serum albumin, to be fused with P3/NS1/1-Ag4-1 myeloma cells. After four fusion experiments, one hybridoma clone secreting a practical antibody has been established. The resulting monoclonal antibody CS#38 (isotype gamma1, kappa) showed an affinity constant (K(a)) for cortisol of 1 x 10(9) M(-1) and provided a practical calibration curve (detection limit, 0.26 ng per assay) in a homologous ELISA system employing horseradish peroxidase-labeled CET as a labeled antigen. Cross-reactivities with related C-21 steroids were acceptably low: 11-deoxycortisol (4.3%), cortisone (4.0%), corticosterone (1.9%), progesterone (1.6%), 17alpha-hydroxyprogesterone (12%), 6beta-hydroxycortisol (8.4%), and tetrahydrocortisol (< 0.1%). Urinary and serum cortisol levels of healthy volunteers were determined by this method after methylene chloride extraction to be 39.0 +/- 17.0 microg/day (n = 7) and 80.8 +/- 38.9 ng/mL (n = 10), respectively, both of which are in the reference range.

[Introduction] Aromatase inhibitors (AI) are the common treatment for postmenopausal estrogen receptor (ER)-positive breast cancer. However, some patients acquire resistance to AI. To investigate the mechanism of resistance, we have established several breast cancer cell lines as AI-resistant models. Recently, we have reported on AMD-EDR cells (androgen metabolite-dependent and estrogen-depletion-resistant cells) derived from MCF-7, in which ER activity depends on one of the androgen metabolites. In this study, another AI-resistant mechanism was shown by T-47D-derived AI-resistant cells, indicating the character of the AI-resistant models differ with their parental cells. [Methods] T-47D was cultured under estrogen-depleted and androgen-supplemented conditions for more than 3 months, simulating the AI treatment, which resulted in two stable variant cell lines. To examine the response of variant cell lines to estrogen and androgen, cell proliferation assay and luciferase reporter assay were performed. The mRNA and protein levels were measured by real-time PCR and western blotting. To identify the elements promoting the AI-resistance mechanism in the variant cell lines, the gene expression of variant cell line was compared with that of T-47D by microarray analysis. Immunohistchemical staining for AR and PSA was performed on 20 pairs of primary and recurrent tissue samples from AI-resistant breast cancer. [Results] The variant cell lines had neither growth response to estrogen nor expression of ER. On the other hand, androgen markedly induced the proliferation of these cell lines, and this effect was inhibited by the androgen receptor (AR) antagonist, bicalutamide. Increased expression of AR and androgen-responsive genes in the variant cell lines indicated that the AR-dependent growth pathway was enhanced. Moreover, L-dopa decarboxylase (DDC), an AR coactivator, was extremely induced, suggesting it is an enhancer of AR activity in the variant cell lines. Immunohistochemistry analysis revealed that all primary tumors did not express PSA, while two recurrent tumors did express PSA, suggesting that AR transcriptional activity is enhanced in some AI-resistant breast cancers. [Conclusion] During the AI treatment simulation these variant cell lines transformed from ER-dependent to AR-dependent, and overexpression of AR and DDC possibly enhanced the AR-mediated proliferative pathway. PSA-expression in the AI-resistance recurrent tumors suggests that AR-mediated pathway possibly plays a role in aromatase inhibitor-resistance mechanism. The AI-resistance mechanism in T-47D-derived variant cell lines was different from that in MCF-7-derived AMR-EDR cells. The fact that the character of the AI-resistant models differed with their parental cells supports the hypothesis that the intristic character of individual primary tumors contributes to diversification of AI-resistance mechanisms. Citation Format: Rika Fujii, Toru Hanamura, Toshifumi Niwa, Yuri Yamaguchi, Takanori Ishida, Hironobu Sasano, Noriaki Ohuchi, Shin-ichi Hayashi. Androgen receptor acquires an oncogenic role in the AI resistance mechanism in breast cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4754. doi:10.1158/1538-7445.AM2014-4754

Introduction: Aromatase inhibitors (AIs) have been reported to exert their anti-proliferative effects not only by reducing estrogen production but also by unmasking the inhibitory effect of androgens such as testosterone (TS) or dihydrotestosterone (DHT) in postmenopausal women with hormone receptor-positive breast cancer. The behavior of androgens in AI-resistance mechanisms is not sufficiently understood. 5α-Androstane-3β,17β-diol (3β-diol) generated from DHT by 3β-hydroxysteroid dehydrogenase type 1 (3β-HSD type 1: HSD3B1) has androgenic activity and substantial estrogenic activity, representing a potential mechanism of AI resistance. Methods: To investigate these issues, ERE-GFP-transfected MCF-7 cells (E10 cells) were cultured for 3 months under steroid-depleted, TS supplemented conditions. Among the surviving cells, two stable variants (V1 and V2 cells) showing ER activity depending on androgen metabolites were selected by monitoring GFP expression. Using these variant cell lines, we investigated the process of adaptation to androgen-abundant conditions and the role of androgens in AI-resistance mechanisms. Results: The variant cell lines showed increased growth and induction of estrogen-responsive genes rather than androgen-responsive genes by androgens or 3β-diol. Further analysis of these cell lines suggested that increased expression of HSD3B1 and reduced expression of androgen receptor (AR) provide adaptability to androgen-abundant conditions, as proved by increased conversion of DHT into 3β-diol by 3β-HSD type 1 and AR signal reduction. Corresponding to these hypotheses, in parental E10 cells, ectopic expression of HSD3B1 or inhibition of AR resulted in adaptation to androgen-abundant conditions. In coculture with stromal cells mimicking the local estrogen production from androgen, these cell lines showed lower sensitivity to an AI compared with parental E10 cells. Immunohistochemistry on 9 pairs of primary and recurrent tissue samples from AI-resistant breast cancer revealed the decrease of AR protein expression in all cases. Conclusion:In the present study, we successfully cloned two stable variants that show ER activity depending on androgen metabolites. Investigation of these cell lines suggested that the increased expression of HSD3B1 and reduced expression of AR might decrease the sensitivity to aromatase inhibitors as proved by enhancement of androgen metabolite-induced ER activation and growth mechanism. The androgen metabolite-dependent growth of hormone receptor positive breast cancer possibly plays a certain role as an aromatase inhibitor-resistance mechanism. Citation Format: Toru Hanamura, Toshifumi Niwa, Sayo Nishikawa, Hiromi Konno, Tatsuyuki Gohno, Chika Chika Tazawa, Yasuhito Kobayashi, Masafumi Kurosumi, Hiroyuki Takei, Yuri Yamaguchi, Ken-ichi Ken-ichi Ito, Shin-ichi Hayashi. The androgen metabolite-dependent growth of hormone receptor-positive breast cancer as a possible aromatase inhibitor resistance mechanism. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1313. doi:10.1158/1538-7445.AM2013-1313

Introduction: Aromatase inhibitors (AIs) have been reported to exert their anti-proliferative effects not only by reducing the production of estrogens but also by unmasking the inhibitory effect of androgens such as testosterone (TS) or dihydrotestosterone (DHT) in postmenopausal women with hormone receptor-positive breast cancer. The behavior of androgens in AI-resistance mechanisms is not sufficiently understood. 5α-androstane-3β,17β-diol (3β-diol) generated from DHT by 3β-hydroxysteroid dehydrogenase type 1 (3β-HSD type 1: HSD3B1) has androgenic activity and substantial estrogenic activity, representing a potential mechanism of AI resistance. Methods: To investigate these issues, ERE-GFP-transfected MCF-7-E10 cells were cultured for 3 months under steroid-depleted, TS-supplemented conditions which is the similar as the AI treatment. Among the surviving cells, two stable variants that show ER activity depending on androgen metabolites were selected as AD-EDR (androgen metabolite-dependent and estrogen depletion-resistant) by monitoring GFP expression. Using these cell lines, we investigated the process of adaptation to androgen-abundant conditions and the role of androgens in AI-resistance mechanisms. Results: AD-EDR cell lines showed increased growth and induction of estrogen-responsive genes rather than androgen-responsive genes by androgens or 3β-diol. Further analysis revealed increased expressions of HSD3B1 and reduced expression of androgen receptor (AR) in these cell lines. In parental MCF-7-E10 cells, ectopic expression of HSD3B1 or inhibition of AR resulted in adaptation to estrogen-deprived and androgen-abundant conditions. In coculture with stromal cells replicating the local estrogen production from androgen, AD-EDR cell lines showed AI resistance compared with parental MCF-7-E10 cells. Immunohistochemistry and real-time PCR analyses on 9 pairs of primary and recurrent tissue samples from AI-resistant breast cancer revealed the decrease of AR protein expression in all cases and increase of HSD3B1 mRNA expression in 5. Conclusion: In the present study, we successfully cloned two stable variants that show ER activity depending on androgen metabolites. Investigation of these cell lines suggested that the increased function of 3β-HSD type 1 and reduced function of AR contribute to AI resistance by enhancing the androgen metabolite-dependent growth and reducing the inhibitory effect of androgens. Our data of clinical samples suggest that this mechanism also acts as an AI-resistance in clinical breast cancer in some cases. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P6-04-17.

The effects of estradiol, progesterone, and tamoxifen on the activity of estradiol 2- and 16 alpha-hydroxylases were studied in human breast cancer cell cultures using a radiometric assay. After 5 days' exposure to these compounds, incubations in the presence of either [2-3H]estradiol or [16 alpha-3H]estradiol as substrate were carried out. In MCF-7 cells, estradiol (10(-8) M), progesterone (10(-6) M) and tamoxifen (10(-6) M) significantly increased 16 alpha-hydroxylase activity (estradiol; 21% progesterone 10% to 32%; tamoxifen 21% to 31%; P less than 0.01). Synergistic effects were observed when the cells were successively exposed to tamoxifen and progesterone. Simultaneous treatment with tamoxifen plus estradiol or estradiol plus progesterone showed no change from estradiol alone. On the other hand, although estradiol had no direct effects on 2-hydroxylase activity, tamoxifen decreased this enzymatic activity significantly at 10(-6) M (23% to 37%). Progesterone acted synergistically to further decrease this reaction. Treatment with only progesterone caused an increase in 2-hydroxylation. In contrast, a subline of MCF-7 cells with low estrogen receptor levels showed only minimal enzyme-hormone responses. Likewise, treatment of the estrogen receptor-negative MDA-MB-231 human breast cancer cell line with these compounds showed no effects on either 2- or 16 alpha-hydroxylase activity. In the progesterone receptor-rich T47D cell line, estradiol decreased both activities while progesterone increased both.(ABSTRACT TRUNCATED AT 250 WORDS)

The preparation and antigenic properties of estriol 16-glucuronide-bovine serum albumin (BSA) and estradiol 17-glucuronide-BSA conjugates in which the hapten molecule is linked to the carrier protein through an (O-carboxymethyl)oxime bridge at the C-6 position on the steroid nucleus, have been described. Antibodies raised against the two immunogens in the rabbit possessed excellent specificity to estriol 16-glucuronide and estradiol 17-glucuronide, respectively, exhibiting no significant cross-reactions with estrogen glucuronides having a substituent in ring A and no cross-reactivities with free estrogens, their sulfates and related steroids.

Using a radiometric assay the effects of estradiol upon the activity of estradiol 2- and 16 alpha-hydroxylases in MCF-7 human breast cancer cells in culture were studied. After 5 days of treatment and 36 h of withdrawal, incubation in the presence of either 2- or 16 alpha-tritiated substrate was carried out. Estradiol (10 nM) significantly increased 16 alpha-hydroxylase activity (21%, P less than 0.01), while no effects on 2-hydroxylase activity was observed. Treatment for 6 weeks caused a major increase in 16 alpha-hydroxylation (65%, P less than 0.001) and a significant reduction in 2-hydroxylation (21%, P less than 0.05). These effects were not observed in estrogen receptor-negative MDA-MB-231 human breast cancer cells. Our observations suggest that these two metabolic pathways in MCF-7 cells are independently regulated and that this regulation is affected via the estrogen receptor.

Estrogen receptors were found in the nuclear fractions of granulosa cells of the largest (F 1 ) and the second largest (F 2 ) preovulatory follicle in the ovary of the hen. The maximum number of binding sites (NBSmax) of the nuclear estrogen receptors was greater in F 2 than in F 1 , while the equilibrium dissociation constant ( K d ) was not different. During the ovulatory cycle, the NBSmax of the estrogen receptors showed a change parallel to the change in the estradiol-17β content in the follicular membranes. The results suggest that estradiol-17β produced in the follicular membrane may exert a direct action on the granulosa cells in the hen ovary.

Although cyclin-dependent kinase (CDK) 4/6 inhibitors have exhibited remarkable results for patients with estrogen receptor (ER)-positive breast cancer in clinical trials, the mechanism of CDK4/6 inhibitor resistance remains unclear. Thus, this study aimed to investigate the mechanism of CDK4/6 inhibitor resistance using two CDK4/6 inhibitor resistant breast cancer cell lines. We established CDK6 overexpressed cell lines (MCF7-C6) from MCF-7 cells using the stably transfected CDK6 expression vector. Additionally, acquired ribociclib-resistant (RIBR) cell lines were created using ER-positive hormone-resistant cell lines by long-term exposure to ribociclib. CDK6 overexpression and the knockdown of CDK4 experiments highlight the significance of high levels of CDK4 and low levels of CDK6 in CDK4/6 inhibitor sensitivity. Moreover, RIBR cell lines did not exhibit incremental CDK6 compared with ER-positive hormone-resistant cell lines. In MCF7-C6 and RIBR cell lines, p21 levels decreased, and p21 levels were proportional to CDK4/6 inhibitor sensitivity. This study suggests that overexpression of CDK6 is one of the many possible mechanisms of resistance to CDK4/6 inhibitors. Furthermore, p21 levels have the potential to serve as a marker for CDK4/6 inhibitors independent of the resistance mechanism.