Abstract P1-09-19: Tumor RNA disruption index as a tool to predict response to neoadjuvant chemotherapy in breast cancer: Optimizing timing of biopsy

BACKGROUND. Early detection of tumor response to neoadjuvant therapy (NAT) could be used to tailor therapy and lower toxicity from ineffective treatments. The CCTG MA.22 trial has shown that RNA disruption is associated with breast cancer response to NAT when measured by image guided core biopsy mid-treatment. The objectives of this study were: 1) to determine the optimal time to measure the tumor RNA Disruption Index (RDI) after initiation of NAT when assessed by fine needle aspiration (FNA) in an office setting and 2) to determine if RDI could predict response to a second chemotherapy agent in patients that had a suboptimal response.

METHODOLOGY. We performed a prospective pilot study including patients with palpable biopsy-proven breast cancer eligible for NAT. Chemotherapy and surgery were at the discretion of the treating physician. Two FNAs after cycles 1, 2, 3 and after initiation of a new chemotherapy agent were collected in RNA Protect Cell Reagent and sent to Rna Diagnostics Inc. to assess RDI. Prospectively recorded clinical tumor measurements and surgical pathology reports were obtained. Tumor pathological response (pR) after NAT was measured by pathological complete response (pCR: no invasive disease in breast) and residual cancer burden index (RCBI).

RESULTS. 30 patients were accrued to the study. One patient withdrew consent, one patient was found to have metastatic disease and did not undergo surgery, and one patient had bilateral breast cancer (n= 29 evaluable tumors). ER+HER2-: 38% (11/29), ER-Her2-: 28% (8/29) and HER2+: 34% (10/29). 89% (25/28) of patients received taxane and anthracycline containing regimens. All HER2+ received trastuzumab. Our pCR and RCBI 0-1 rates were 24% (7/29) & 38% (11/29) respectively. At cycles 1, 2 and 3, RDI could be evaluated in 72% (21/29), 73% (16/22) and 44% (7/16) of palpable tumors. After the switch to a new agent, RDI could only be evaluated in 30% (3/10) of patients with palpable tumors. Using a tumor RDA cutoff at 5 (non-responder RDI < 5 (NR) and responder RDI ≥ 5 (R)), responder status between cycle 1 and 2 was concordant in 73% (11/15). After 1 cycle, NR vs. R was associated with numerically lower pCR (13% (2/15) vs. 33% (2/6), p=0.54) and RCBI 0-1 at surgery (20% (3/15) vs. 33% (2/6), p=0.60). These findings were similar after cycle 2. Non-analyzable samples (NAS) because of absence of RNA were associated with high pR (pCR: 38% (3/8) and RBCI 0-1: 75% (6/8)). The 3 NR at cycle 1 that achieved a significant pR had either non-palpable tumors or NAS after switching to a new chemotherapy agent.

CONCLUSION. RDI can be measured by FNA in an office setting and could be helpful to identify early non-responders to NAT. The optimal time to perform RDI is after 1 or 2 cycles of treatment, which should be considered in ongoing and future trials. This study was underpowered to detect a statistically significant correlation between RDI and pR. NAS is associated with high pR and could represent responders to treatment. This early data suggests that RDI is unlikely to be helpful in assessing response to a second chemotherapy agent after receiving 4 cycles of standard chemotherapy. The use of RDI to tailor NAT needs to be evaluated in larger prospective trials.

Citation Format: Samkari A, Chung W, Parissenti A, Pritzker L, Trabulsi N, Basik M, Boileau J-F. Tumor RNA disruption index as a tool to predict response to neoadjuvant chemotherapy in breast cancer: Optimizing timing of biopsy [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-09-19.

Abstract P3-14-11: Microarray data analysis and long term outcomes of NCIC-CTG MA.22: Neoadjuvant epirubicin and docetaxel with pegfilgrastim support for locally advanced breast cancer

Background: 93 patients were enrolled in sequential phase 1 and 2 trials of epirubicin (E) and docetaxel (D) given at 3 (cohort A) or 2 weekly (cohort B) intervals. We previously reported clinical (93%) and pathologic response rates (pCR 7%) as well as the association of fall in tumour RNA integrity (RIN) with response (ASCO 2010). Here we report the results of microarray analysis of tumor specimens pre-and mid-treatment to determine genes which are differentially expressed in different groups. Methods: 6 core biopsies were collected for all patients pre-, mid- and post-treatment with ED. 3 cores were used for standard pathologic assessment while 3 were used for gene expression assessment using Agilent full genome microarrays. Pre- and mid-treatment cores were used with Agilent Feature Extraction Software to assess microarrays; and baseline continuous (% positive) immunohistochemical ER, PR, HER2, and Topo2 were investigated by schedule and dose. RNAs with RIN > = 5.0 were subjected to microarray analysis. NIH BRB array tools were used for investigations of differential gene expressions. K-M curves were generated for the phase 2 cohorts for disease free (DFS) and event free survival (EFS), and for ER− PR− and ER or PR+ subgroups. Results: Of the 93 patients, 47 were in cohort A and 46 in cohort B. Median overall survival was 6.34 years on study. DFS at 50 months was 55% for A (phase 2) and 67% for B (phase 2), while EFS was 55% for A and 63% for B. For ER or PR+ DFS and EFS were 60% and for ER− PR− DFS and EFS were 63%. 134 arrays were available in total: 57 from A, 68 from B with 11 reference breast tumour RNAs for standardization. Patients with and without microarrays were not significantly different. Pre-treatment, we found 3 differentially expressed genes in A and 6 in B between patients who did and did not have RIN > = 5.0 at mid-treatment. Comparing CR to non-CR (PR, SD, PD), 40 genes were found for A and 2 genes for B. Many genes were also differentially expressed in A and B when analyzed by pathologic factors ER, PR, HER2, Topo2, schedule and dose. Mid-treatment, 4,365 genes in A and 18,770 genes in B were significantly different from pre-treatment. Conclusion: At 50 months, DFS was 55% for 3 weekly and 67% for 2 weekly schedules of ED. The genes identified in each cohort pretreatment will be investigated further for relevance to predicting sensitivity or resistance to E or D.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-14-11.

Abstract P1-08-24: High tumor RNA disruption is associated with increased survival in patients with locally advanced breast cancer

Background: We recently reported that epirubicin/docetaxel treatment of locally advanced breast cancer patients in the MA.22 clinical trial reduced tumor RNA integrity and that low mid-treatment tumor RNA integrity was associated with a pathologic complete response (pCR) post-treatment [Breast Cancer Res. Treat. 119:347-356]. A novel method for detection of chemotherapy-dependent loss of RNA integrity, the RNA disruption assay (RDA), has since been developed, which stratifies RNA disruption into 3 zones. Zone 1 represents drug effect inadequate to obtain a pCR [RNA disruption index (RDI) ≤ 10, NPV>0.99]. Zone 2, an intermediate zone (RDI >10 and ≤35) had 1 pCR in our study, while Zone 3 (RDI >35) had the most pCRs (n = 7).

Methods: We have now evaluated the relationship between tumor RNA disruption and both treatment response and patient survival in all evaluable MA.22 patients (n = 85). RNA was isolated from 2-3 tumor biopsies/patient prior to and after 3 (Schedule A) or 4 (Schedule B) cycles of chemotherapy at 3- or 2-weekly intervals, respectively. RNA quality was assessed on an Agilent 2100 Bioanalyzer and RNA disruption quantified using RDA.

Results: Chemotherapy exposure reduced maximum tumor RNA content from 197 ± 24 ng/μl (mean ± standard error) pre-treatment to 94.0 ± 12.8 ng/μl mid-treatment, with maximum tumor RDI values increasing from 26.0 ± 5.8 to 88.1 ± 19.1 (p<0.01 for both observations by Sign test). At the time of assessment, there were similar numbers of deceased patients across tumor RDA zones 1, 2, and 3 mid-treatment (7, 8, and 9 patients, respectively). In contrast, there were 2.9-fold more living patients with tumors in RDA zone 3 (29) than in zone 1 (10). Disease-free survival (DFS) was significantly greater (by 23.6 months) for zone 3 patients (56.9 ± 5.6 months) compared to zone 1 patients (33.3 ± 6.8 months) (p = 0.0096 by Mann Whitney Wilcoxon Test for this and all subsequent statistical tests). Living patients with tumor RDA values in zone 3 exhibited 24.5 months greater DFS than living patients with tumors in zone 1 (68.2 ± 5.5 months versus 43.7 ± 9.5 months, respectively; p = 0.015). Given their high frequency within the patient population, we were also able to conclude that greater DFS (an additional 28.6 months) could be seen for patients with estrogen receptor positive (ER+) tumors in zone 3 (59.2 ± 7.2 months) compared to patients with similar tumors in zone 1 (30.6 ± 6.3 months; p = 0.0088). Living patients with ER+ tumors in zone 3 had even greater DFS (+42.5 months) compared to living patients with ER+ tumors in zone 1 (71.8 ± 5.8 months versus 29.3 ± 8.7 months, respectively; p = 0.0012). There were also 2.5-fold more living patients with ER+ tumor RDA values in zone 3 (15) than in zone 1 (6). None of the above significant differences were observed when pre-treatment tumor RDI values or pre-treatment patient characteristics were assessed. Irrespective of breast tumor receptor subtype, chemotherapy-dependent RNA disruption mid-treatment was typically higher in patients that achieved a pCR post-treatment than for patients that did not.

Conclusions: Our findings suggest that tumor RNA disruption measurements during treatment will be highly useful to predict response and survival for locally advanced breast cancer patients. Our data demonstrate that non-responding patients (zone 1 by RDA) have considerably reduced DFS with very little chance of achieving a pCR. Such patients can be spared the toxicities associated with continuing an ineffective regimen and can be considered quickly for alternate treatments, including surgery, radiation therapy, or other anti-cancer drugs. The impact of treatment schedule and dose on RDA's ability to predict patient response and survival is currently being assessed.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-24.

Abstract P1-14-13: Increased Pathologic Complete Response Rate and Reduced Tumour RNA Levels Upon Treatment of Locally Advanced Breast Cancer with Neoadjuvant Concurrent Chemotherapy and Radiation

Introduction: Patients with locally advanced breast cancer (LABC) have a 5-year survival rate of 50% using standard treatment that includes neo-adjuvant chemotherapy, surgery and adjuvant radiation. Given this poor prognosis, we proposed use of an adjuvant regimen in the neoadjuvant setting, involving radiation with docetaxel for radiosensitization. Pathological complete response (pCR) at surgery (the best surrogate for overall survival) was the primary end-point of this single-arm prospective Phase II trial. Given that dose-dependent reductions in tumour RNA integrity mid-treatment were correlated with post-treatment pCRs in similar patients in the NCIC-CTG-MA.22 clinical trial (Breast Cancer Research and Treatment 119:347–56), we also assessed whether changes in tumour RNA integrity and concentration could be observed following this regimen.

Methodology: Thirty-two patients with stage III non-metastatic LABC were enrolled at a single institution from 2009–2011. They were treated with neoadjuvant 5-Fluoro-uracil, Epirubicin, and Cyclophosphamide (FEC) q3 weekly for 4 cycles followed by weekly Docetaxel (35 mg/m2) concurrently with regional radiation (45 Gy with 16 Gy boost in 25 & 5 fractions) for the first 6 of 9 weeks (D/R). Serial 14 gauge tumour core biopsies were taken from the patients pre-, mid- and post-treatment and 1 mm3 samples used to assess RNA quantity and quality using an Agilent 2100 Bioanalyzer. This was followed by a modified radical mastectomy.

Results: Treatment-related toxicities included grade 3 resolving pneumonitis (6 patients), grade 3 dermatitis (6 patients) and one treatment-related death. The pCR rate was 23%–about twice the provincial rate for LABC. At a mean 21 months follow-up, the relapse-free survival was 100% in the pCR cohort and 65% among partial responders (PRs). While biopsies were not normalized for mass, little difference in RNA concentration was observed between pre-treatment and post-FEC tumour biopsies (means of 50 ± 15 nanograms/microliter and 50 ± 12 nanograms/microliter, respectively). In contrast, the mean tumour RNA concentration was reduced to 10 ± 2.1 nanograms/microliter post-D/R. While the post-treatment biopsies often yielded insufficient RNA for assignment of tumour RNA integrity (RIN) values, treatment response appeared to be associated with low tumour RNA integrity, since patients with disease progression or stable disease had higher RIN values than patients exhibiting PRs or pCRs.

Conclusions: This is the first study to use a full chemotherapy regimen with radiation in the neo-adjuvant setting for LABC. Although not without toxicity, the regimen appeared to significantly improve the pCR rate for this high-risk group, resulting in a much-improved outcome even at short-term follow-up. In addition, the FEC→D/R regimen produced reductions in tumour RNA concentration, and treatment response appeared to be associated with reduced RNA integrity.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P1-14-13.

miR-135a contributes to paclitaxel resistance in tumor cells both in vitro and in vivo

Cancer cell resistance to paclitaxel continues to be a major clinical problem. In this study, we utilized microRNA (miRNA) arrays to screen for differentially expressed miRNAs in paclitaxel-resistant cell lines established in vitro. We observed concordant upregulation of miR-135a in paclitaxel-resistant cell lines representing three human malignancies. Subsequently, the role of miRNA-135a was evaluated in an in vivo model of paclitaxel resistance. In this model, mice were inoculated subcutaneously with a non-small cell lung carcinoma cell line and treated with paclitaxel for a prolonged period. In paclitaxel-resistant cell lines, established either in vitro or in vivo, blockage of miR-135a sensitized resistant cell lines to paclitaxel-induced cell death. We further demonstrated a correlation between paclitaxel response and miR-135a expression in paclitaxel-resistant subclones that were established in vivo. The paclitaxel-resistant phenotype of these subclones was maintained upon retransplantation in new mice, as shown by decreased tumor response upon paclitaxel treatment compared with controls. Upregulation of miR-135a was associated with reduced expression of the adenomatous polyposis coli gene (APC). APC knockdown increased paclitaxel resistance in parental cell lines. Our results indicate that paclitaxel resistance is associated with upregulation of miR-135a, both in vitro and in vivo, and is in part determined by miR-135a-mediated downregulation of APC.

Reductions in tumor RNA integrity associated with clinical response to epirubicin/docetaxel chemotherapy in breast cancer patients

Abstract #6068

Background: Optimal doses and dosing intervals for neoadjuvant anthracycline/taxane chemotherapy are poorly studied. Moreover, biomarkers for measuring response to such combination regimens are unknown. This study investigated these issues in sequential phase I/II cohorts of women with locally advanced or inflammatory breast cancer treated with epirubicin and docetaxel with pegfilgrastim support at 3- or 2-weekly intervals in association with the NCIC-CTG MA.22 clinical trial.
 Methods: Accrual has been completed for the 3-weekly regimen (maximum tolerated dose: epirubicin 105 mg/m2, taxotere 75 mg/m2), and continues for the phase II portion of the 2-weekly regimen. Six core biopsies were obtained from 50 patients pre-, mid-, and post-treatment. Immunohistochemical staining was performed to determine baseline levels of ER, PR, HER2 and Topo II expressed as % positive stain. Tumour RNA integrity (RIN) and tumor extent were measured pre-, mid- and post-treatment by capillary electrophoresis and light microscopy after haematoxylin/eosin staining, respectively. Associations between maximum and average RIN at the three time points and tumour extent, clinical response, pathologic complete response, or baseline levels of ER, PR, HER2 and Topo II were assessed using Spearman correlation coefficients after data transformation to improve symmetry and stabilize variances. The association between both RIN and tumour extent, and baseline drug dose was assessed using a 1-way ANOVA.
 Results: Low mid-treatment maximum RIN was associated with high drug dose level (p=0.05) and eventual pathologic complete response (p=0.01). Post-treatment, low maximum and average RIN were found to be associated with low tumor extent (p=0.004 and p=0.01, respectively). As well, low average RIN was significantly associated with clinical complete response post-treatment (p=0.01). As expected, post-treatment low tumor extent was significantly associated with pathologic complete response (p=0.01). High pre-treatment Topo II levels were also significantly associated with high RIN (p = 0.03). No association was observed between RIN and HER2, ER or PR.
 Discussion: The association of RIN with tumour extent, pathologic complete response, clinical response, a known risk factor (Topo II), and drug dose suggests that the RIN may represent an important new biomarker for measuring response to anthracycline/taxane combinations (and possible other chemotherapy regimens) in breast cancer patients.

Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6068.

Evaluation of sICAM-1, sVCAM-1, and sE-Selectin levels in patients with metastatic breast cancer receiving high-dose chemotherapy

Soluble forms of some cell adhesion molecules (CAM), sICAM-1, sVCAM-1, and sE-selectin, are elevated in the sera and plasma of patients with inflammation, arthritis, diabetes, and cancer. Increased levels of these soluble molecules in patients with cancer have been shown to correlate with disease progression and survival. This suggests that increased expression of the soluble forms of CAMs may play an important role in cancer cell growth and metastasis and may be prognostic and/or predictive of malignant disease. In this retrospective study, we assessed the clinical significance of sICAM-1, sVCAM-1, and sE-selectin in 95 patients with metastatic breast cancer enrolled in clinical trials of high-dose chemotherapy (HDC) and autologous stem cell transplantation (ASCT). The significance of soluble HER-2 (sHER-2) and sFAS status, determined in previous studies for this group of patients, was also included in this analysis. Univariate analysis showed that sICAM-1, sVCAM-1, sFas, sHER-2 positive status, and the presence of liver metastases were significant prognostic factors for both progression-free survival (PFS) and overall survival (OS) in the total patient group. In multivariable analysis, HER-2 and sFAS were shown to be independent prognostic factors for PFS and OS. Within the various treatment groups examined, sICAM-1 was a prognostic factor for clinical outcome for patients with metastatic breast cancer enrolled in trials with cyclophosphamide- and carboplatin-based or vinblastine-based HDC, but not in trials with paclitaxeland cyclophosphamide-based HDC.

Soluble Fas (CD95) is a prognostic factor in patients with metastatic breast cancer undergoing high-dose chemotherapy and autologous stem cell transplantation

The Fas/Fas ligand (FasL) system plays an important role in cellular apoptosis and is involved in cancer cell death induced by the immune system and anticancer drugs. Increased serum levels of soluble Fas (sFas) are associated with a number of different disease states and with tumor progression and metastasis in patients. In this study, we examined the plasma levels of sFas in 94 women with metastatic breast cancer undergoing high-dose chemotherapy (HDCT) treatment with autologous stem cell transplantation (ASCT) using a quantitative enzyme-linked immunosorbent assay (ELISA) method. Thirty-one patients (31/94, 33%) had plasma sFas levels greater than the optimum cut point of 1.90 ng/ml (median 2.47, range 1.98-13.54 ng/ml) and were designated as sFas positive. Sixty-three patients (63/94, 67%) had sFas levels below 1.90 ng/ml (median 1.14, range 0.47-1.89 ng/ml). In univariate analysis, patients with sFas-positive status, HER-2 overexpression, and the presence of liver metastases had a significantly shorter time to disease progression (PFS) and significantly decreased overall survival (OS). Multivariable analysis (Cox proportional hazards model) for PFS determined that sFas status significantly predicted disease progression (p = 0.004) with an adjusted hazard ratio (HR) of 2.0 (95% CI, 1.3-3.3). HER-2 status and liver metastases were also significant independent predictors of disease progression (p < 0.001) for both. sFas level was also an independent prognostic factor for OS with an adjusted HR of 2.0 (p = 0.006; 95% CI, 1.2-3.4). HER-2 status and liver metastases also remained highly significant independent prognostic factors for OS (HER-2: p < 0.001, HR 2.3, and liver metastases: p = 0.001, HR 2.7). In conclusion, these results suggest that plasma levels of sFas may be a valuable clinical prognostic factor in predicting outcome (PFS and OS) for patients with metastatic breast cancer undergoing HDCT with ASCT.

HER-2 expression is a prognostic factor in patients with metastatic breast cancer treated with a combination of high-dose cyclophosphamide, mitoxantrone, paclitaxel and autologous blood stem cell support

The expression levels of a circulating extracellular domain of HER-2 can be detected in the plasma and serum of patients with metastatic breast cancer using an enzyme immunoassay (ELISA) method. In this study, we evaluated the clinical significance of high and low levels of HER-2 in the plasma of 46 patients with metastatic breast cancer enrolled in a clinical trial of high-dose chemotherapy (HDCT) using cyclophosphamide, mitoxantrone, and paclitaxel with autologous stem cell transplantation (ASCT). Using 2500 U/ml as the cut-point, 20 patients (46%) had elevated HER-2 levels (HER-2 positive). Our results suggest that patients with metastatic breast cancer and high soluble plasma HER-2 have a significantly poorer overall (OS) and progression-free survival (PFS) following high-dose chemotherapy with paclitaxel and ASCT. The median OS of patients with low levels of HER-2 was significantly longer (P < 0.01) than the median OS of patients with high levels of HER-2 (29.8 months vs 15.9 months). PFS was also significantly longer (P < 0.01) for patients who were HER-2-negative, than for patients who were HER-2-positive (13.0 vs 8.6 months). Univariate analysis showed that patients with liver or lung metastases had significantly reduced OS and PFS. Patients with metastases to two or more sites also had a significantly reduced time to disease progression, but not OS. In multivariable analysis, lung metastases contributed along with HER-2-positive status to determine a group of patients with significantly poorer OS. However, HER-2-positive status remained the only independent predictor of PFS.

Regulation of protein kinase C by the cytoskeletal protein calponin

Previous studies from this laboratory have shown that, upon agonist activation, calponin co-immunoprecipitates and co-localizes with protein kinase Cepsilon (PKCepsilon) in vascular smooth muscle cells. In the present study we demonstrate that calponin binds directly to the regulatory domain of PKC both in overlay assays and, under native conditions, by sedimentation with lipid vesicles. Calponin was found to bind to the C2 region of both PKCepsilon and PKCalpha with possible involvement of C1B. The C2 region of PKCepsilon binds to the calponin repeats with a requirement for the region between amino acids 160 and 182. We have also found that calponin can directly activate PKC autophosphorylation. By using anti-phosphoantibodies to residue Ser-660 of PKCbetaII, we found that calponin, in a lipid-independent manner, increased auto-phosphorylation of PKCalpha, -epsilon, and -betaII severalfold compared with control conditions. Similarly, calponin was found to increase the amount of (32)P-labeled phosphate incorporated into PKC from [gamma-(32)P]ATP. We also observed that calponin addition strongly increased the incorporation of radiolabeled phosphate into an exogenous PKC peptide substrate, suggesting an activation of enzyme activity. Thus, these results raise the possibility that calponin may function in smooth muscle to regulate PKC activity by facilitating the phosphorylation of PKC.

Role of specific apoptotic pathways in the restoration of paclitaxel-induced apoptosis by valspodar in doxorubicin-resistant MCF-7 breast cancer cells

Paclitaxel (Taxol) kills tumor cells by inducing both cellular necrosis and apoptosis. A major impediment to paclitaxel cytotoxicity is the establishment of multidrug resistance whereby exposure to one chemotherapeutic agent results in cross-resistance to a wide variety of other drugs. For example, selection of MCF-7 breast cancer cells for resistance to doxorubicin (MCF-7ADR cells) results in cross-resistance to paclitaxel. This appears to involve the overexpression of the drug transporter P-glycoprotein which can efflux both drugs from tumor cells. However, MCF-7ADR cells possess a deletion mutation in p53 and have considerably reduced levels of the Fas receptor, Fas ligand, caspase-2, caspase-6, and caspase-8, suggesting that paclitaxel resistance may also stem from a bona fide block in paclitaxel-induced apoptosis in these cells. To address this issue, we examined the ability of the P-glycoprotein inhibitor valspodar to restore paclitaxel accumulation, paclitaxel cytotoxicity, and paclitaxel-induced apoptosis. Compared to drug sensitive MCF-7 cells, MCF-7ADR cells accumulated >6-fold less paclitaxel, were approximately 100-fold more resistant to killing by the drug, and were highly resistant to paclitaxel-induced apoptosis. In contrast, MCF-7ADR cells pretreated with valspodar were indistinguishable from drug-sensitive cells in their ability to accumulate paclitaxel, in their chemosensitivity to the drug, and in their ability to undergo paclitaxel-induced apoptosis. Valspodar, by itself, did not affect these parameters. This suggests that the enhancement of paclitaxel toxicity in MCF-7ADR cells involves a restoration of apoptosis and not solely through enhanced drug-induced necrosis. Morever, it appears that changes in the levels/activity of p53, the Fas receptor, Fas ligand, caspase-2, caspase-6, or caspase-8 activity have little effect on paclitaxel-induced cytotoxicity and apoptosis in human breast cancer cells.

Lack of modulation of MDR1 gene expression by dominant inhibition of cAMP-dependent protein kinase in doxorubicin-resistant MCF-7 breast cancer cells

The drug transporter P-glycoprotein (P-gp) appears to play an important role in the ability of tumor cells to evade killing by chemotherapeutic agents. Using pharmacological inhibitors of cAMP-dependent protein kinase (PKA), it has been suggested that, similar to rodent model systems, the human P-gp gene (MDR1) is also under PKA-dependent control and that PKA inhibition may prove useful in reducing drug resistance in human cancer cells. To test this hypothesis, we stably transformed doxorubicin (Adriamycin)-resistant human MCF-7 breast cancer cells (MCF-7(ADR)) with a vector that inhibits PKA activity by inducing over-expression of mutant type Ialpha PKA regulatory (RIalpha) subunits. Two transformants (MCF-7(ADR-A) and MCF-7(ADR-B)) were found to express mutant RIalpha subunits and to possess markedly reduced PKA activity; another transformant (MCF-7(ADR-9)) lacked mutant RIalpha subunit expression and exhibited no inhibition of PKA activity. In contrast with findings in Chinese hamster ovary and Y1 adrenal cells, P-gp levels and cellular sensitivity to drugs which are P-gp substrates were unchanged in the PKA-inhibited transformants, suggesting that P-gp expression and function are not under PKA-dependent control in MCF-7(ADR) cells. Growth and saturation densities of the cell lines were highly correlated with level of PKA catalytic activity, suggesting that PKA inhibition may prove useful in inhibiting growth of breast tumor cells, even upon establishment of resistance to doxorubicin. However, our results challenge current proposals that drug sensitivity in P-gp-expressing human tumor cells may be restored by blocking MDR1 gene expression through inhibition of PKA activity.

Carbon source-dependent regulation of cell growth by murine protein kinase C epsilon expression in Saccharomyces cerevisiae

Protein kinase C is known to play a role in cell cycle regulation in both lower and higher eucaryotic cells. Since mutations in yeast proteins involved in cell cycle regulation can often be rescued by the mammalian homolog and since significant conservation exists between PKC-signalling pathways in yeast and mammalian cells, cell cycle regulation by mammalian PKC isoforms may be effectively studied in a simpler genetically-accessible model system such as Saccharomyces cerevisiae. With this objective in mind, we transfected S. cerevisiae cells with a plasmid (pYECepsilon) coding for the expression of murine protein kinase C epsilon (PKCepsilon) under the control of a galactose-inducible promoter. Unlike mock-transfected cells, yeast cells transformed with pYECepsilon expressed, in a galactose-dependent manner, an 89 kDa protein that was recognized by a human PKCepsilon antibody. Extracts from these pYECepsilon-transfected cells could phosphorylate a PKCepsilon substrate peptide in a phospholipid/phorbol ester-dependent manner. Moreover, this catalytic activity could be inhibited by a fusion protein in which the regulatory domain of murine PKCepsilon was fused in frame with GST (GST-Repsilon), further confirming the successful expression of murine PKCepsilon. Induction of PKCepsilon expression by galactose in cells transformed with pYECepsilon increased Ca++ uptake by the cells approximately 5-fold and resulted in a dramatic inhibition of cell growth in glycerol. However, when glucose was used as the carbon source, PKCepsilon expression had no effect on cell growth. This was in contrast to what was observed upon bovine PKCalpha or PKCbeta-I expression in yeast, where expression of these PKC isoforms strongly and moderately inhibited growth in glucose, respectively. Visualization of the cells by phase contrast microscopy indicated that murine PKCepsilon expression in the presence of glycerol resulted in a significant increase in the number of yeast cells exhibiting very small buds. Since overall growth of the cells was dramatically decreased, the data suggests that PKCepsilon expression potently inhibits the progression of yeast cells through the cell cycle after the initiation of budding. In addition, a small amount of the PKCepsilon-expressing yeast cells (1-2%) exhibited gross alterations in cell morphology and defects in both chromosome segregation and septum formation. This suggests that for those cells which do complete DNA synthesis, murine PKCepsilon expression may nevertheless inhibit yeast cell growth by retarding and/or imparing cell division. Taken together, the data suggests murine PKCepsilon expression potently reduces the growth of yeast cells in a carbon source-dependent fashion by affecting progression through multiple points within the cell cycle. This murine PKCepsilon-expressing yeast strain may serve as a very useful tool in the elucidation of mechanism(s) by which external environmental signals (possibly through specific PKC isoforms) regulate cell cycle progression in both yeast and mammalian cells.

Inhibitory properties of the regulatory domains of human protein kinase Calpha and mouse protein kinase Cepsilon

Two fusion proteins in which the regulatory domains of human protein kinase Calpha (Ralpha; amino acids 1-270) or mouse protein kinase Cepsilon (Repsilon; amino acids 1-385) were linked in frame with glutathione S-transferase (GST) were examined for their abilities to inhibit the catalytic activities of protein kinase Calpha (PKCalpha) and other protein kinases in vitro. Both GST-Ralpha and GST-Repsilon but not GST itself potently inhibited the activities of lipid-activated rat brain PKCalpha. In contrast, the fusion proteins had little or no inhibitory effect on the activities of the Ser/Thr protein kinases cAMP-dependent protein kinase, cGMP-dependent protein kinase, casein kinase II, myosin light chain kinase, and mitogen activated protein kinase or on the src Tyr kinase. GST-Ralpha and GST-Repsilon, on a molar basis, were 100-200-fold more potent inhibitors of PKCalpha activity than was the pseudosubstrate peptide PKC19-36. In addition, a GST-Ralpha fusion protein in which the first 32 amino acids of Ralpha were deleted (including the pseudosubstrate sequence from amino acids 19-31) was an effective competitive inhibitor of PKCalpha activity. The three GST-R fusion proteins also inhibited protamine-activated PKCalpha and proteolytically activated PKCalpha (PKM), two lipid-independent forms of PKCalpha; however, the IC50 values for inhibition were 1 order of magnitude greater than the IC50 values obtained in the presence of lipid. These results suggest that part of the inhibitory effect of the GST-R fusion proteins on lipid-activated PKCalpha may have resulted from sequestration of lipid activators. Nonetheless, as evidenced by their abilities to inhibit the lipid-independent forms of the enzyme, the GST-R fusion proteins also inhibited PKCalpha catalytic activity through direct interactions. These data indicate that the R domains of PKCalpha and PKCepsilon are specific inhibitors of protein kinase Calpha activity and suggest that regions of the R domain outside the pseudosubstrate sequence contribute to autoinhibition of the enzyme.

Molecular strategies for the dominant inhibition of protein kinase C

The regulatory (R) domain of PKC alpha fused to glutathione-S-transferase (GST-R alpha) competitively inhibited PKC activity associated with extracts of Y1 mouse adrenocortical tumor cells and the activities of several specific PKC isozymes. GST-R alpha did not inhibit the activities of cAMP-dependent protein kinase, cGMP-dependent protein kinase or calmodulin-dependent myosin light chain kinase. GST-R alpha inhibited PKC activities 20 times more potently than did a synthetic peptide corresponding to the pseudosubstrate sequence of PKC alpha. In intact yeast cells, the R domain prevented PKC beta-1-induced inhibition of growth and cytokinesis. These results indicate that the R domain of PKC alpha acts as a specific, dominant inhibitor of PKC activity, and suggest that the PKC alpha R domain may provide a useful genetic tool to assess the roles of PKC in various signal transduction processes.

Regulatory domain of human protein kinase C alpha dominantly inhibits protein kinase C beta-I-regulated growth and morphology in Saccharomyces cerevisiae

This study demonstrates that the isolated regulatory (R) domain (amino acids 1-270) of human protein kinase C alpha (PKC alpha) is a potent inhibitor of PKC beta-I activity in a yeast expression system. The PKC alpha R domain fused to glutathione-S-transferase competitively inhibited the activity of yeast-expressed rat PKC beta-I in vitro (Ki = 0.2 microns) and was 400-fold more potent than a synthetic pseudosubstrate peptide corresponding to amino acids 19-36 from PKC alpha. In contrast, the fusion protein did not affect the activity of the purified catalytic subunit of cAMP-dependent protein kinase. The PKC alpha R domain (without glutathione-S-transferase [GST]) also was tested for its ability to inhibit PKC beta-I activity in vivo, in a yeast strain expressing rat PKC beta-I. Upon treatment with a PKC-activating phorbol ester, yeast cells expressing rat PKC beta-I were growth-inhibited and a fraction of the cells appeared as long chains. Coexpression of the R domain with rat PKC beta-I blocked the phorbol ester-induced inhibition of yeast cell growth and the phorbol ester-dependent alterations in yeast cell morphology. These results indicate that the R domain of PKC alpha acts as a dominant inhibitor of PKC activity in vivo and thus provides a useful genetic tool to assess the roles of PKC in various signal transduction processes.

Reconstitution of protein kinase C alpha function by the protein kinase C beta-I carboxy terminus

The Ca(2+)- and phospholipid-dependent Ser/Thr kinase protein kinase C (PKC) plays important roles in the transduction of cellular signals. Various PKC isoforms exist in mammalian cells which share conserved and variable regions as defined by cDNA sequence comparisons. To test whether carboxyl (C) terminal sequences of distinct isoforms can complement each other to yield functional chimeric molecules, we have constructed a PKC chimera in which amino acids 595-672 at the C-terminus of bovine PKC alpha (a) were replaced with the corresponding C-terminal amino acids (598-671) of rat PKC beta-I (b) to yield the chimera alpha/beta-I (ab). The chimera was then characterized biochemically and functionally, and compared with the parental isoforms. Since structure/function analysis of PKC in mammalian experimental systems is complicated by multiple PKC isoforms and by cellular complexity, we stably introduced the PKC constructs into the yeast Saccharomyces cerevisiae, a simple, lower eukaryote with a short doubling time and well established molecular genetics. In yeast, the faithfully expressed PKCab chimera and normal PKC isoforms bound radiolabelled phorbol ester and were recognized on immunoblots by PKC-specific antibodies. The chimera phosphorylated substrate peptides in a PMA- and Ca(2+)-dependent manner, and, upon activation, increased the cell doubling time and the rate of Ca2+ uptake into cells. In addition, PKCab displayed characteristics distinct from normal PKCb, but virtually indistinguishable from normal PKCa. Our findings indicate the reconstitution of PKCa function by the PKCb carboxyl terminus.

Phorbol ester activation of functional rat protein kinase C beta-1 causes phenotype in yeast

The phorbol ester receptor protein kinase C (PKC) gene family encodes essential mediators of various eukaryotic cellular signals. The molecular dissection of its mechanisms of action has been limited in part by the genetic inaccessibility and complexity of signaling in mammalian cells. Here we present a novel approach to study rat PKC beta-1 action in yeast, a simple lower eukaryotic genetic model. Expression of its cDNA in Saccharomyces cerevisiae introduces novel phorbol ester binding sites which stimulate a specific calcium- and phospholipid-dependent catalytic activity in vitro consistent with a fully functional protein which phosphorylates cellular yeast proteins in vivo. Phorbol ester activation of PKC beta-1 in vivo results in biological responses which include stimulation of extracellular calcium uptake, changes in cell morphology, and an increase in the cell doubling time. These PKC functions are not affected by truncation of 12 amino terminal amino acids; however, they are completely abolished by truncation of 15 or more carboxyl terminal amino acids which likely result in inactivation of the kinase. The increase in the yeast doubling time caused by PKC beta-1 activation provides a phenotype which can be exploited as a screen for the activity of random PKC cDNA mutations. Our findings indicate that rat PKC beta-1 is functional in yeast and leads to biological responses which suggest compatible aspects of higher and lower eukaryotic signaling pathways and the feasibility of dissecting parts of the action of common signaling mediators in a simple genetic model.

Stimulation of calcium uptake in Saccharomyces cerevisiae by bovine protein kinase C alpha

Ca2+ plays essential roles as a second messenger often in synergism with the calcium- and phospholipid-dependent phorbol ester receptor, protein kinase C (PKC), which stimulates Ca2+ influx in various cell types in a potential positive feedback mechanism. To address the compatibility of these mechanisms between lower eukaryotes and mammals, we have stably expressed bovine PKC alpha in the yeast Saccharomyces cerevisiae. We find that phorbol ester binding sites are created which stimulate a specific calcium- and phospholipid-dependent catalytic activity in vitro. Phorbol ester activation in vivo stimulates PKC down-regulation, uptake of extracellular Ca2+, Ca2+ dependence of cell viability, and changes in cell morphology. This may represent activation of a putative PKC-mediated signaling pathway utilized by functional yeast homologs of mammalian PKC isoforms. These are suggested by some protein data; however, their genes have not yet been characterized (Simon, A. J., Milner, Y., Saville, S. P., Dvir, A., Mochly-Rosen, D., and Orr, E. (1991) Proc. R. Soc. Lond. B 243, 165-171). Our findings indicate that bovine PKC alpha is functional in yeast and stimulates calcium uptake in a manner similar to some of its responses in mammalian cells, which suggests compatible aspects of higher and lower eukaryotic signaling pathways and the feasibility of dissecting parts of the action of common signaling mediators in a simple genetic model.

Identification of promoter elements in the mouse 21-hydroxylase (Cyp21) gene that require a functional cyclic adenosine 3',5'-monophosphate-dependent protein kinase

The constitutive and cAMP-induced expression of the mouse steroid 21-hydroxylase gene (Cyp21) are impaired in adrenal cell mutants harboring mutations in cAMP-dependent protein kinase (cAMPdPK). These requirements for a functional cAMPdPK have been mapped to the proximal 330 basepairs of the Cyp21 promoter. This study attempts to identify specific promoter elements of Cyp21 that require cAMPdPK for constitutive activity by comparing their abilities to enhance the expression of a reporter gene in Y1 adrenocortical tumor cells and Y1 Kin mutants defective in cAMPdPK activity. As determined in transient transfection assays, Cyp21 promoter elements at -65, -140, -170, -210, and -280 each enhanced the expression of a human GH reporter gene in parent Y1 cells. The relative order of effectiveness of each of these elements was: -170 >> -280 > -140 > -65 > or = -210. The -170 element was 25-fold more effective in enhancing gene expression from the reporter construct in Y1 cells than in Kin mutant cells; the elements at -65, -140, and -210 were 3-fold more effective in Y1 cells than in Kin mutant cells; the -280 element was equally effective in the parent and Kin mutant clones. These studies suggest that the promoter elements at -170, -65, -140, and -210 mediate the requirement for a functional cAMPdPK in the expression of Cyp21. As determined by gel mobility shift assays with these elements, the dependence of the Cyp21 promoter elements on a functional cAMP-dependent protein kinase did not result from decreased expression or binding affinities of their respective DNA-binding proteins.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of DNA and synthetic oligodeoxyribonucleotides on the binding properties of a cGMP-binding protein from Dictyostelium discoideum

Previously, we identified a cGMP-binding protein (cGBP) in Dictyostelium discoideum that can exist in two forms: a fast-dissociating (F-type) activity and a slow-dissociating (S-type) activity. Moreover, the F-type activity was converted effectively to S-type by the addition of nucleic acids, especially DNA (Parissenti, A.M. and Coukell, M. B. (1989) J. Cell Sci. 92, 291-301). In this study, we examined the effects of heterologous DNA and various synthetic homo-oligodeoxyribonucleotides on the cGMP-binding properties of partially purified F-type activity. Equilibrium and kinetic binding experiments revealed that DNA increased the affinity of the protein for cGMP without altering the number of binding sites. However, the presence of DNA decreased only slightly the apparent Kd of the protein for cGMP because the nucleic acid also reduced the rate of cGMP association. Addition of oligo(dGMP)8 or oligo(dCMP)8 to the protein increased both total cGMP binding and the conversion of F-type activity to S-type; in contrast, oligo(dAMP)8 or oligo(dTMP)8, at the same concentration, had no effect. Oligodeoxycytidylic acids with chain lengths less than about eight nucleotides were also ineffective or inhibitory. Analysis of cGMP binding to intact, filipin-permeabilized cells revealed a binding activity with association and dissociation rates comparable to isolated S-type activity. This observation suggests that in vivo the cGBP might exist in its S-form.