Role of Heat Shock Protein 70 and Cortisol Levels in Resistance to Paclitaxel Treatment

The Role of Heat Shock Protein 70 and Cortisol Levels in Resistance to Paclitaxel Treatment in Breast Cancer Cells

Abstract

The aim of our study is to assess the effect of HSP70 on breast cancer cell resistance to chemotherapeutic agents by overexpressing HSP70 and analysing its effect on MCF-7 cell viability combined with Paclitaxel treatment. HSP70 is a molecular chaperone that provides the cell with chemoresistance to multiple apoptotic pathways such as the ERK and JNK pathways (1,2,3,4,5) and reduces the efficacy of many chemotherapeutic drugs (1,3). Cell viability assays showed no clear change in cell survival rates in cells treated with Cortisol. Additionally, there was no clear trend in the effect of Cortisol on HSP70 protein expression in MCF-7 cells treated with Paclitaxel. However, only treating cells with varying doses of Cortisol had an apparent dose dependent effect in gene and protein expression of HSP70 in a bell-shaped manner. While no clear conclusions can be reached from the experiments performed so far, some promising results were obtained and new emerging research seems to support the idea of using HSP70 as a molecular target in cancer therapy (3,4).

Introduction

Breast cancer is the second most prevalent cancer-related cause of death and the most commonly diagnosed cancer in women (6,7). HSPs are a family of proteins that act as molecular chaperones. Their main role in healthy cells is to ensure protein stability and to prevent cellular damage (8). HSPs and in particular HSP90 are being studied as possible molecular targets in cancer therapy. Many HSP90 inhibitors are being developed however there is little research focusing on HSP70 as a potential target for chemotherapeutic drugs (3). HSP70s main function is assisting in the correct folding of proteins, protecting cells from heat shock, blocking protein aggregation and transporting proteins and vesicles (8,9). HSP70 has also been shown to inhibit apoptosis through means of different points of action along different apoptotic pathways including suppression of the Jun-NH2-kinase pathway (3,2), preventing the formation of the apoptosome through interaction with Apaf1, interfering mitochondrial membrane permeability by blocking Bax translocation from the cytosol to mitochondria (1,3) and decreasing senescence by acting of p53 and p21and shortening of cell cycle phases (1,3,10) (Fig.1). HSP70 family members include a stress inducible HSP70 also known as HSP72 as well as a constituently expressed HSP70 (3). Inducible HSP70 is expressed in many cancer cells lines and correlates with poor prognosis (4,11) as it directly inhibits apoptotic signalling and promotes cell migration and proliferation (1,2,4). Beere et al. illustrated the change in HSP70 concentration after heat shock in Jurkat cells showing an increase from ~0.4μM before stress to 10μM after (2). In their study Beere et al. also concluded that further addition of HSP70 either exogenously or by heat shock inhibits the apoptotic machinery (2). Chronic stress accompanied with high Cortisol levels have also been shown to stimulate HSP70 and HSP90 production (12). Additionally, oxidative stress has been linked to the increase in metastasis in MCF-7 cells (13) We hypothesised that as the concentration of Cortisol increases, the efficacy of Paclitaxel would decrease measure by an increase in cell viability compared to cells treated only with Paclitaxel. As seen in Fig.1 the increase in Cortisol would result in the overexpression of HSP70 (10,12) as more HSP70 would be needed to form mature glucocorticoid receptors capable of binding to Cortisol and transporting it to the nucleus (5). HSP70 would, in turn, also increase cytoprotection and chemotherapeutic resistance through various mechanisms mentioned previously. Knockdown of HSP70 has also been shown to enhance the chemosensitivity to anticancer chemotherapeutic drug cisplatin (1). Therefore, we hypothesized that increasing doses of cortisol, a form of oxidative stress, will lead to a decrease in the success of Paclitaxel, an anticancer chemotherapeutic drug, treatment in MCF-7 cells due to over induced expression of HSP70 which will increase proliferation rates and resistance to chemotherapeutic agents.

Figure 1. Graphical abstract of our hypothesis showing Cortisol entering a cell binding to a glucocorticoid receptor which releases HSP70 which in turs blocks various apoptotic pathways of the cell. (Adapted from Dick DM, Richard IM (5)).

Materials and methods

Cell culture

MCF-7 cells used in this study were sourced from the American Type Culture Collection cell bank and were stored in a liquid nitrogen freezer in the vapour phase. The cells were plated in DMEM with 10% foetal bovine serum (FBS) and grown at 37^oC under 5% CO2.

Cortisol and Paclitaxel treatment.

Paclitaxel was purchased from Sigma (T7191) and diluted in DMSO to a 100nM working solution. Cortisol was purchased from Sigma (H0888) and diluted in EtOH (VWR Chemicals, 20821.365) to a 2μM working solution. To record the effects of the combined effect of half maximal inhibitory concentration Paclitaxel and varying doses of Cortisol the IC50 dose of Paclitaxel needed to be determined. After performing three independent CVA analysis the IC50 dose of Paclitaxel in MCF-7 cells was wrongly determined to be 19nM (data not shown). Based on the findings of our fellow students all cells used in this study requiring combined treatment with IC50 Paclitaxel and Cortisol used a concentration of 7nM Paclitaxel (true IC50). To analyse the effect of Cortisol on HSP70 expression and of overexpressed HSP70 on MCF-7 cells, cells were treated with a range of Cortisol concentrations from 15,625nM to 1000nM.

Cell Viability Assay.

MCF-7 cells were seeded in 200μl of media at a 40-50% confluency in a 96-well culture plate by Timothy Oates. Firstly, to assess the effect of Cortisol in cell viability of MCF-7 cells and therefore HSP70, MCF-7 cells were treated with a serial dilution of Cortisol from 31,25nM to 1000nM by a constant factor of 1:2 with triplicates of each treatment. They were then incubated at 37 37^oC under 5% CO2 for 72 hours. The cells were seeded by Timothy Oates.

Secondly, to determine whether an increase in Cortisol concentration reduces the efficacy of Paclitaxel, per our hypothesis, cell viability of cells treated with a combination of 7nm Paclitaxel and a serial dilution of Cortisol from 31,25nM to 1000nM by a constant factor of 1:2 was performed. Control cells were left untreated, treated with 0.004% EtOH as a Cortisol control vehicle, 0.00014% DMSO as a Paclitaxel control vehicle or with a combination of both control vehicles. In both treatments changes in cell viability were quantified by staining attached cells with crystal violet dye and measuring intensity of staining with a spectrophotometer at 595nm. MCF-7 treated cells were incubated with 0.5% Crystal violet in 30% (v/v) ethanol sourced from Sigma. The average absorbance reading was calculated from the triplicates of each sample and normalised to the ethanol vehicle or the combination vehicle absorbance values (depending on treatment). Cell viability was expressed as a percentage.

Western Blot Analysis

Looking for expression of HSP70 in dependence of Cortisol concentration cells were treated with 7nM Paclitaxel and six Cortisol concentrations ranging from 31,25nM to 1000nM. As a control cells were left untreated, treated with Cortisol vehicle, Paclitaxel vehicle and a combination of both. Cells were also treated with 7nM Cortisol and with 500nM Cortisol to assess effect of Cortisol on untreated MCF-7 cells. Samples containing the same amount of protein were loaded to hand cast 10% polyacrylamide gels and separated by SDS-PAGE based on their molecular weight. They were then transferred to Polyvinylidene difluoride (PVDF) membranes from Millipore (IPFL00010). The antibodies used were HSP70 primary antibody raised in rat (Cell signalling technology, #4873), GAPDH primary antibody raised in rabbit (Cell signalling technology #5174), HSP70 secondary anti-rat (Cell signalling technology #7077S) and GAPDH secondary anti-rabbit (Cell signalling technology #70745) all in a 1:1000 dilution in 5% BSA. The Horseradish peroxidase (HRP) labelled secondary antibodies were detected using ECL, the chemiluminescence produced was analysed by iBright System following manufacturers instructions. Band size from the captured images was then quantified using ImageJ.

RNA Analysis and qPCR

Quantitative PCR was performed to assess whether Cortisol had an effect on HSP70 gene expression Cells were treated with a Cortisol serial dilution ranging from 15,625nM to 250nM, untreated cells were used as a control. Cells were frozen by Timothy Oates. 7 days after treatment cells were lysed and RNA was purified using the Qiagen RNeasy® Mini Kit and was then quantified using a photodetector, the Nanodrop Spectrophotometer (Thermo Scientific). 2μl of cDNA synthesised from the RNA samples were combined with 8μl master mix containing HSP70 forward primer (5’-TGTGGCTTCCTTCGTTATTGG-3’) and reverse primer (5’-GCCAGCATCATTCACCACCAT-3’) or GAPDH forward primer (5’-AGCCACATCGCTCAGACAC-3’) and reverse primer (5’-GCCCAATACGACCAAATCC-3’), all sourced from Sigma, in the wells of a qPCR plate. Both master mixes used Power SYBR Green purchased from AB/LT (A25741) and RNase free water. Quantification of the cDNA sequence of our test gene, HSP70, and our internal control gene, GAPDH, was achieved by measuring the increase of fluorescence, resulting from SYBR green dye binding to the amplified double stranded product, with a qPCR machine. Each cDNA sample was run in triplicate with triplicates of ‘no enzyme’ controls, qPCR water controls and cDNA water controls. To calculate gene expression, the double delta Ct method was used, the relative changes in HSP70 gene expression compared to the untreated sample were then analyzed.

Results

Cell Viability Assay

Visual assessment of Figure 2 shows no correlation between Cortisol treatment and variation in MCF-7 cell viability. Figure 3 shows that 7nM Paclitaxel resulted in half maximal cell death and that Cortisol treatment did not increase cell viability when combined with Paclitaxel either.

Western Blot

Figure 4, 5 and 6 show no clear trend in the combined effect of Cortisol and Paclitaxel on HSP70 protein expression. The large difference between the untreated sample and EtOH vehicle in figure 5 indicates unreliability of the results.

Unfortunately, figure 4, 5 and 6 are technical replicates that used the same protein lysates and therefore no statistical analysis can be performed. Had they been independent biological replicates a t-test were our null hypothesis would be that Cortisol treatment has no significant effect on HSP70 protein expression in MCF-7 cells treated with Paclitaxel could be performed comparing the average HSP70 expression in IC50 Paclitaxel treated cells and in cells treated with 500nM Cortisol and the IC50 dose of Paclitaxel (our highest dose in two of the three trials).

Alternatively, an ANOVA test could also be performed to determine what concentration of Cortisol induces the largest difference in HSP70 expression compared to cells treated with 7nM paclitaxel. From this we could observe the effect said concentration has on cell viability and, per our hypothesis, prove (or disprove) a relationship between Cortisol induced HSP70 expression and cell viability of MCF-7 cells.

Contrastingly, figure 7 appears to show an increase in HSP70 expression in cells treated with Cortisol.

qPCR 

Figure 8 seems to indicate that there was an increase of HSP70 expression in MCF-7 cells with treatment of Cortisol in a bell-shaped dose dependant manner.

Discussion and conclusion

The obtained Western Blot results seem to indicate that Cortisol had no effect on HSP70 protein expression in MCF-7 cells treated with Paclitaxel. This would explain the cell viability results whereby treatment with Cortisol showed no significant inhibition of normal apoptotic pathways or any decreased chemosensitivity to Paclitaxel. We suspect this was due to the already high concentration of HSP70 in MCF-7 cells or to the inability of Cortisol to induce HSP70 protein expression in MCF-7 cells. However, the first option seems more likely as it would relate with previous literature (Yaglom JA. et al that showed that MCF-7 cells had the largest relative levels of inducible HSP70 compared to MCF10F, HeLa and HCT116 (4)) and our own qPCR (Fig.7) and Western Blot results (Fig.6) where a similar trend is seen in both HSP70 gene and protein expression of MCF-7 cells treated exclusively with Cortisol. It must be noted that no biological or technical repeats where performed for either of these experiments. Therefore, the results are still to be confirmed as being an accurate representation of the effect of Cortisol on MCF-7 cells.

In conclusion, previous research has already established the increase in senescence (4) and sensitivity to chemotherapeutic agents, such as Cisplatin, (1) due to HSP70 knockdown and while this paper aimed at consolidating said effect in MCF-7 cells we have been unable to overexpress HSP70 in a consistent manner. Nevertheless, it is not uncommon for HSP70 to already be constituently overexpressed in tumorigenic cells (4,2,3) these results just reiterate the importance of studying the different senescent and apoptotic pathways HSP70 inhibits in cancer cells. Looking to future experiments we suggest attempting to overexpress HSP70 through other sources of stress (we have already performed cell viability assays of MCF-7 cells treated with epinephrine (data not shown)) and to quantify their effect on various chemotherapeutic drugs. It might also be interesting to observe the effect of knocking down HSP70 on MCF-7 cells treated with Paclitaxel. The more we comprehend the mechanisms through which HSP70 can be overly expressed and how in turn this increase in expression relates to poor prognosis (1,3,4) the more accurately we can design inhibitors to sensitise cancer cells to chemotherapy

References

1. Yoshidomi K, Murakami A, Yakabe K, Sueoka K, Nawata S, Sugino N. Heat shock protein 70 is involved in malignant behaviors and chemosensitivities to cisplatin in cervical squamous cell carcinoma cells. J Obstet Gynaecol Res 2014 May 01;40(5):1188-1196.
2. Beere HM, Wolf BB, Cain K, Mosser DD, Mahboubi A, Kuwana T, et al. Heat-shock protein 70 inhibits apoptosis by preventing recruitment of procaspase-9 to the Apaf-1 apoptosome. Nat Cell Biol 2000;2(8):469.
3. Kim LS, Kim JH. Heat Shock Protein as Molecular Targets for Breast Cancer Therapeutics. Journal of Breast Cancer 2011;14(3):167-174.
4. Yaglom JA, Gabai VL, Sherman MY. High levels of heat shock protein Hsp72 in cancer cells suppress default senescence pathways. Cancer Res 2007;67(5):2373.
5. Dick DM, Richard IM. Molecular chaperones and the stress of oncogenesis. Oncogene 2004;23(16):2907.
6. Sharma G, Dave R, Sanadya J, Sharma P, Sharma K. Various types and management of breast cancer: An overview. Journal of Advanced Pharmaceutical Technology & Research 2010;1(2):109-126.
7. Rostas JW, Dyess DL. Current Operative Management of Breast Cancer: An Age of Smaller Resections and Bigger Cures. International Journal of Breast Cancer 2012;2012(2012).
8. Saibil H. Chaperone machines for protein folding, unfolding and disaggregation. Nature Reviews Molecular Cell Biology 2013;14(10):630.
9. María RF, José María Valpuesta. Hsp70 chaperone: a master player in protein homeostasis version 1; referees: 3 approved]. F1000Research 2018;7.
10. Barnes JA, Dix DJ, Collins BW, Luft C, Allen JW. Expression of inducible Hsp70 enhances the proliferation of MCF-7 breast cancer cells and protects against the cytotoxic effects of hyperthermia. Cell Stress Chaperones 2001;6(4):316.
11. Seo J, Park YM, Kim JI, et al. T cell lymphoma in transgenic mice expressing the human Hsp70 gene. Biochem Biophys Res Commun 1996;218:582–7.
12. Celi M, Vazzana M, Sanfratello MA, Parrinello N. Elevated cortisol modulates Hsp70 and Hsp90 gene expression and protein in sea bass head kidney and isolated leukocytes. Gen Comp Endocrinol 2012 February 01;175(3):424-431.
13. Prathap Kumar SM, Kamaleshwar PS. Chronic oxidative stress increases growth and tumorigenic potential of MCF-7 breast cancer cells. PLoS ONE 2014;9(1):e87371.