Data pertaining to aberrant intracellular calcium handling during androgen deprivation therapy in prostate cancer

The data generated here in relates to the research article “CaV1.3 enhanced store operated calcium promotes resistance to androgen deprivation in prostate cancer”. A model of prostate cancer (PCa) progression to castration resistance was employed, with untreated androgen sensitive LNCaP cell line alongside two androgen deprived (bicalutamide) sublines, either 10 days (LNCaP-ADT) or 2 years (LNCaP-ABL) treatment, in addition to androgen insensitive PC3. With this PCa model, qPCR was used to examined fold change in markers linked to androgen resistance, androgen receptor (AR) and neuron specific enolase (NSE), observing an increase under androgen deprivation. In addition, the gene expression of a range of calcium channels was measured, with only the L-type Voltage gated calcium channel, CACNA1D, demonstrating an increase during androgen deprivation. With CACNA1D knockdown the channel was found not to influence the gene expression of calcium channels, ORAI1 and STIM1. The calcium channel blocker (CCB), nifedipine, was employed to determine the impact of CaV1.3 on the observed store release and calcium entry measured via Fura-2AM ratiometric dye in our outlined PCa model. In both the presence and absence of androgen deprivation, nifedipine was found to have no impact on store release induced by thapsigargin (Tg) in 0mM Ca2+ nor store operated calcium entry (SOCE) following the addition of 2mM Ca2+. However, CACNA1D siRNA knockdown was able to reduce SOCE in PC3 cells. The effect of nifedipine on CaV1.3 in PCa biology was measured through cell proliferation assay, with no observed change in the presence of CCB. While siCACNA1D reduced PC3 cell proliferation. This data can be reused to inform new studies investigating altered calcium handling in androgen resistant prostate cancer. It provides insight into the mechanism of CaV1.3 and its functional properties in altered calcium in cancer, which can be of use to researchers investigating this channel in disease. Furthermore, it could be helpful in interpreting studies investigating CCB's as a therapeutic and in the development of future drugs targeting CaV1.3.


a b s t r a c t
The data generated here in relates to the research article "CaV1.3 enhanced store operated calcium promotes resistance to androgen deprivation in prostate cancer". A model of prostate cancer (PCa) progression to castration resistance was employed, with untreated androgen sensitive LNCaP cell line alongside two androgen deprived (bicalutamide) sublines, either 10 days (LNCaP-ADT) or 2 years (LNCaP-ABL) treatment, in addition to androgen insensitive PC3. With this PCa model, qPCR was used to examined fold change in markers linked to androgen resistance, androgen receptor ( AR ) and neuron Keywords: Prostate cancer Androgen deprivation Store operated calcium entry Voltage gated calcium channels CaV1. 3 Calcium channel blocker Store operated calcium channels ORAI STIM specific enolase ( NSE ), observing an increase under androgen deprivation. In addition, the gene expression of a range of calcium channels was measured, with only the L-type Voltage gated calcium channel, CACNA1D , demonstrating an increase during androgen deprivation. With CACNA1D knockdown the channel was found not to influence the gene expression of calcium channels, ORAI1 and STIM1 . The calcium channel blocker (CCB), nifedipine, was employed to determine the impact of CaV1.3 on the observed store release and calcium entry measured via Fura-2AM ratiometric dye in our outlined PCa model. In both the presence and absence of androgen deprivation, nifedipine was found to have no impact on store release induced by thapsigargin (Tg) in 0mM Ca 2 + nor store operated calcium entry (SOCE) following the addition of 2mM Ca 2 + . However, CACNA1D siRNA knockdown was able to reduce SOCE in PC3 cells. The effect of nifedipine on CaV1.3 in PCa biology was measured through cell proliferation assay, with no observed change in the presence of CCB. While siCACNA1D reduced PC3 cell proliferation. This data can be reused to inform new studies investigating altered calcium handling in androgen resistant prostate cancer. It provides insight into the mechanism of CaV1.3 and its functional properties in altered calcium in cancer, which can be of use to researchers investigating this channel in disease. Furthermore, it could be helpful in interpreting studies investigating CCB's as a therapeutic and in the development of future drugs targeting CaV1. 3

Value of the Data
• This data expands our understanding into the role of altered store operated calcium in androgen resistant prostate cancer, as well as improving our knowledge on calcium channel blocker (CCB) sensitivity of voltage gated calcium channels such as CaV1.3 in disease states such as cancer. • The outlined data can be used to support further investigations into altered store operated calcium in androgen resistant prostate cancer. In addition, it will aid in the development of future experiments investigating the mechanisms of CaV1.3 in disease. • In addition, it also provides further insight into previous or future studies using calcium channel blockers as a target for CaV1.3 in disease.

Data Description
This data provides supporting information to the outlined article "CaV1.3 enhanced store operated calcium promotes resistance to androgen deprivation in prostate cancer" [1] . Raw and analysed data in excel format relating to this article is provided at Mendeley data in the same order as outlined in this article [2] . This includes gene expression CT data for all genes outlined as well as housekeeping. CaV1.3 protein expression data values as quantified by ImageJ. 340/380 florescence measurements overtime for all calcium traces. In addition to absorbance values of WST-1 at 440nM relating to cell proliferation and colony counts for cell survival assays.
A model of disease progression during androgen deprivation therapy (ADT) was employed, representing untreated androgen sensitivity (LNCaP), early androgen resistance (LNCaP-ADT) and castrate resistant PCa (CRPC) (LNCaP-ABL). In addition to an androgen insensitive PCa cell line, PC3. With which the gene expression of androgen resistance markers were measured ( Fig. 1 ) [3][4][5] . Androgen receptor ( AR ) expression was significantly increased during ADT treatment in LNCaP-ADT (p < 0.01) and LNCaP-ABL (p < 0.05) compared to untreated LNCaP ( Fig. 1 A). In addition, neuron specific enolase ( NSE ) was also significantly increased due to ADT in LNCaP-ADT (p < 0.05), alongside a non-significant increase in LNCaP-Abl ( Fig. 1 B). The stages of PCa disease that the cellular model represents are summarised in Fig. 1 C.
The outlined primary article observed that high external potassium failed to mediate calcium influx following membrane depolarisation [1] . The addition of 56mM NaCl ( Fig. 3 A) and 75mM  NaCl ( Fig. 3 B) were tested as osmolarity controls, finding no change in basal Ca 2 + in any of the outlined PCa cell lines. Average basal cytosol calcium (Ca c 2 + ) in 0mM PSS was also measured ( Fig. 3 C), observing that ADT treatment induced a significant Ca c 2 + increase in both short term LNCaP-ADT cells (p < 0.05) and long-term ADT, LNCaP-ABL cells (p < 0.01).
The gene expression of a range of store operated calcium channel families was measured in LNCaP-ADT compared to untreated LNCaP ( Fig. 2 A). The expression of all calcium channels was increased due to ADT, but CACNA1D , was found to be predominately expressed with a
2-fold increase on average compared to other calcium channels ( Fig. 4 A). ORAI1 and STIM1 gene expression was measured individually due to their role in store operated calcium (SOC) under ADT in PCa [6] . No change was found in ORAI1 under ADT, nor did siCACNA1D have any effect ( Fig. 4 B). STIM1 was observed to have a significant increase in early-stage ADT treatment (LNCaP-ADT) versus untreated LNCaP (p < 0.001, Fig. 4 C), which was lost in androgen insensitive LNCaP-ABL. siCACNA1D failed to have any effect on STIM1 gene expression across all conditions. The linked publication demonstrated using siCACNA1D that increased CaV1.3 under ADT drove store operated calcium entry (SOCE) [1] . As a known CaV1.3 inhibitor the calcium channel blocker (CCB), nifedipine, was used to test its ability block CaV1.3 mediated SOCE ( Fig. 5 ). Calcium traces displayed an increase in store release following thapsigargin (Tg) treatment across all three cell lines that was similar in the presence (p < 0.0 0 01) or absence (p < 0.0 0 01) of nifedipine ( Fig. 5 A + B). Further analysis found that ADT significantly increased the Tg peak ( Fig. 5 C) in long term CRPC (LNCaP-Abl, p < 0.005), which was not altered by nifedipine treatment. This was also mimicked in the analysis of the Tg-induced slope, with an increase in LNCaP-Abl which is no altered by nifedipine (p < 0.05, Fig. 5 D). Following the addition of 2mM Ca 2 + a significant increase in SOCE was witnessed again across all three cell lines in both the DMSO control (P < 0.0 0 01) and nifedipine treated (p < 0.0 0 01) ( Fig. 5 A + B). Comparing the Ca 2 + peak a nonsignificant stepwise increase was observed in short (LNCaP-ADT) to long term ADT (LNCaP-Abl) compared to untreated LNCaP ( Fig. 5 E). No significant differences were observed in cells between DMSO control and nifedipine treatment. Analysis of Ca 2 + entry slope found no significant difference between any cell type either with ADT or nifedipine ( Fig. 5 F). Calculation of the total change in Ca 2 + through SOC was measured by area under the curve analysis, here a step wise increase was observed from short (LNCaP-ADT) to long term (LNCaP-Abl) ADT treatment, but this failed to reach significance ( Fig. 5 G). Furthermore, nifedipine was found to have no significant effect.
The effect of nifedipine compared to DMSO control on cell proliferation was tested ( Fig. 6 ). The outlined cell line model displayed a significant reduction in LNCaP-ADT proliferation verses untreated LNCaP in the DMSO control (p < 0.001), with nifedipine displaying no effect. Long term ADT CRPC cell line LNCaP-Abl displayed a significant increase in cell proliferation (p < 0.01) but nifedipine did not have any impact.
CaV channels are known to produce a C-terminus that can influence gene transcription [7] , using protein fractionation the presence of the CaV1.3 C-terminus in the nucleus was measured ( Fig. 7 ). Its presence was found in both LNCaP and early-stage ADT resistant LNCaP-ADT, the latter of which displayed a 3-fold increase. In castrate resistant LNCaP-ABL, expression was completely lost. Note, this is preliminary data without a nucleus control so further work is require to confirm this.
Androgen insensitive PC3 cell line was used to measure the role of CaV1.3 on SOC and proliferation using siCACNA1D . PC3 cell lines displayed a significant increase in store release under Tg treatment (P < 0.0 0 01) followed by a significant increase in SOCE (p < 0.0 0 01) ( Fig. 8 A). Following CACNA1D siRNA knockdown store release was decreased but did not reach significance ( Fig. 8 B), however SOCE did (p < 0.05, Fig. 8 C). PC3 proliferation was significantly increased compared to androgen sensitive LNCaP (p < 0.01, Fig. 8 D), which was reduced by siCACNA1D (p < 0.01, Fig. 8 D) to levels observed with LNCaP.
Bioinformatic analysis was conducted in the original paper using the samples from the Taylor et al data set [8] . These samples had been spilt into adjacent, primary, and metastatic for which the numbers are outlined ( Table 1 ). The numbers used in each cohort for Gleason score, hormone therapy and ETS-related gene ( ERG) status are all outlined in the table.

Tissue culture
The human androgen receptor (AR) positive PCa cell line LNCaP (ATCC # CRL-1740) was used as an androgen sensitive untreated control. Two androgen resistant sublines were developed using anti-androgen, 10μM bicalutamide (Sigma). LNCaP-ADT was treated with ADT for 10 days, the timepoint associated with significant AR and neuroendocrine gene expression linked to treatment resistance ( Fig. 1 ) [3][4][5] . Alongside a previously published 2 year ADT insensitive CRPC, LNCaP-ABL [9] . In addition to androgen independent cell line PC3. All lines were maintained in RPMI (Gibco) with 10% fetal bovine serum (Gibco) in a humidified incubator at 37 °C and 5% CO 2 . In addition, they were mycoplasma negative and had their origin confirmed via short tandem array (STR) profiling.

Calcium channel blocker Nifedipine
Nifedipine stock concentration of 100 mM was made up by adding 34.63 mg Nifedipine (Acros Organics) to 1 ml DMSO. Media containing nifedipine was produced by adding 0.1 μl/ml of 100mM stock Nifedipine to cell culture media above resulting in a final concentration of 10 μM Nifedipine.

siRNA transfection
siRNA was received as lyophilised powder and reconstituted with PCR grade H 2 O to a concentration of 5μM, aliquoted into 20 μl and stored at -20 °C. Cells were seeded into the relevant plates and allowed to grow to 80% confluency in relevant media, without antibiotic.  efficiency was assessed by qPCR and western blotting with a minimum of 70% reduced genetic expression and 87% protein expression observed.

RNA extraction and cDNA synthesis
Total ribonucleic acid (RNA) was extracted using High Pure RNA Isolation Kit (Roche) as per the supplied instructions. Total RNA was quantified using NanoDrop 10 0 0 TM (Thermo Scientific). 1μl RNA extraction was used to give quantity in ng/μl. Purity of RNA was assessed using absorbance 260nm and 280nm (A260/A280) with a ratio of 1.8-2.2 accepted as pure RNA. Complimentary deoxyribonucleic acid (cDNA) was synthesised from RNA extractions using Transcriptor first strand cDNA synthesis kit (Roche) using 2.5μM Anchored-oligo primer and 60μM random hexamer primer. Producing 1ug cDNA in 20μl which was stored at -20 °C.

Primer design
All primers were designed using primer 3 ( https://primer3.ut.ee/ ), using Exon spanning sequences where possible. Oligo Calc ( http://biotools.nubic.northwestern.edu/OligoCalc.html ) was Table 1 Sample details used for bioinformatic analysis. PCa primary and metastatic tumour and match normal samples were acquired from radical prostatectomy. (A) Tumour type vs origin. This sample set contained expression information from a total of N = 176 samples of which n = 130 where primary cancers and n = 18 metastatic (The 19 metastatic samples were from lung, neck, spine, bladder, testes, bone, brain and colon tissue). A subset of primary cancers also had n = 28 paired adjacent normal samples. (B) Sample size and tumour type for each hormone therapy group used. (C) Sample source and Combined Gleason score showed that most samples had scores of 6 or 7 than 8 or 9. (D) Sample source and ERG status showed that most were ERG-negative (73%) and a minority were ERG-positive (27%) and that these proportions were similar in both metastatic (78% and 22%) and primary tumours (72% and 28%   used to detect any potential hairpin formations or self-annealing sites. All primers were acquired from Sigma Aldrich as lyophilised powder which were reconstituted with PCR grade H 2 O to stock concentration 100μM, from which working stock primers (5μM) were made.

Real time polymerase chain reaction (qPCR)
Master mix was prepared for multiple reactions by adding 5μl FastStart Essential DNA Green Master (Roche), 1μl forward primer (0.5μM) and 1 μl reverse primer (0.5μM) per reaction. 2μl cDNA (50ng) was added to each reaction alongside the 7μl of the master mix and the volume made up to 10μl with PCR grade H2O. The PCR strip was placed in the LightCycler Nano TM (Roche) which was programmed to run as outlined in the FastStart Essential DNA Green Master protocol, with the annealing temperature adjusted according to the primer melting temperature. The housekeeping gene, Hypoxanthine Phosphoribosyltransferase 1 ( HPRT1), was used to determine the relative expression of each gene. The relative gene expression data was analysed using the 2-CT method. The sequences the PCR primers used through are included ( Table 2 ).

Cellular fractionation and protein quantification
To determine the cellular location of expressed proteins a cell fractionation kit was used (Abcam). The kit extracted three fractions, the cytosolic fraction, the membrane fraction, and the nuclear fraction. Cells were grown on 100mm cell culture dishes to approximately 6 * 10 6 cells and processed as stated in the provided manual. Protein was quantified using the Pierce BCA Protein Assay Kit (Thermo Scientific). Protein samples were diluted 1:10 in PBS and 25μl were plated in triplicate. Working reagent (WR) was made up by mixing 50 parts of BCA Reagent A with 1 part BCA Reagent B. 200μl of WR was added to each well and the plate was incubated for 30mins at 37 °C. The absorbance was detected at 562nm. The standards were used to create a standard curve from which the concentration of each protein sample was calculated

Western blot
Protein lysates were separated on a Bolt Bis-Tris Plus pre-cast Gels (4-12%, Invitrogen) using the Mini Gel Tank (Invitrogen). In a microcentrifuge tube 50μg total protein was added to 5μl Bolt LDS sample buffer (Invitrogen) and 2μl Bolt reducing agent (Invitrogen). The total volume was made up to 20μl with sterile dH2O. The protein sample was heated at 70 °C for 10mins to denature. The samples were loaded into the relevant wells and the tank filled with x1 Bolt MOPS SDS running buffer (Invitrogen). The tank was connected to the electrical supply and run at 200V for 30mins. The gel was then transferred to a PVDF membrane (methanol activated) 0.4μm in Transfer buffer (Bolt Transfer buffer (Invitrogen) with Bolt Antioxidant (Invitrogen) at 20V for 60mins. The membrane was incubated in blocking buffer (Tris buffered saline, 0.1% tween 20 and 5% milk) for 1 hour at room temperature, followed by three 5 minutes washes in TBST. The membrane was subsequently incubated with CaV1.3 C-terminus antibody (Abcam, 1:500 dilution, mouse monoclonal, AB-84811) in TBST containing 5% milk over night at 4 °C with gentle agitation. Followed with a TBST wash before 1hr incubation with secondary HRP-Anti Mouse (HRP-Anti-Mouse, BD Pharmigen, Goat Polycloncal,1:10 0 0). The membrane was then incubated in 1:4 dilution of Supersignal West Dura Chemiluminescent Substrate (Thermo Scientific) with gentle agitation for 1min at room temperature. Afterwards it was exposed to a piece of X-ray film in a cassette in the dark room before placed in developer solution, followed by fixer. Resultant blots where scanned and analysed using ImageJ.

Calcium measurements
Cells were grown in 6 well dishes with relevant test conditions for experimental parameters. Cells were then removed, counted, and resuspended in a tube in OptiMEM TM low serum media at a concentration of 5 * 10 5 cells/ml. Fura 2-AM was added to the tube to a final concentration of 2uM and incubated at RT in the dark for 45mins. The cells were centrifuged at 80 0 0g for 5mins and the supernatant was discarded. The pellet was then resuspended in 1ml OptiMEM TM and incubated in the dark for 30mins. This process was repeated, and the cells were incubated in the dark for 45mins to allow complete de-esterification of the dye. Tubes were then centrifuged and the pellet was resuspended in 500μl physiological saline solution PSS 0% Ca 2 + (NaCl 142mM, MgCl 2 1mM, KCl 4mM, D-glucose 11.1mM,1 mM EGTA and HEPES 10mM pH balanced to 7.4 with NaOH). From this100μl was added to each well of a black walled 96 well plate to give 1 * 10 5 cells/well. Measurements were made using the VICTOR multilabel plate reader with excitation wavelengths set at 340 and 380nm and the emission wavelength set at 510nm with measurements recorded every 5s. The plate reader was preloaded with the Tg and Ca 2 + solutions which were injected at a volume of 20μl/well. Baseline Ca 2 + levels were determined for 100s then Tg was injected at final concentration of 4μM and recorded for 400s after which CaCl 2 was injected at a final concentration of 2mM with measurements recorded for a further 200s. Tg results in the depletion of the ER Ca 2 + stores by inhibiting the action of SERCA. Introducing Ca 2 + to the extracellular fluid allows us to see the level of extracellular Ca 2 + that is taken into the cell as a result of ER store depletion. The addition of high external potassium of 56 mM KCl (K60) or 76 mM KCl (K80) was used to induce membrane depolarization. Associated osmolarity controls of 56 mM NaCl (Na196) or 76 mM NaCl (Na216) was used. These measurements where contacted in PSS (in mM): NaCl 140, MgCl2 1, KCl 4, CaCl2 2, D-glucose 11.1 and HEPES 10, adjusted to pH 7.4 with NaOH, as well as potassium channel blocker NS1619 (50μM) to induce membrane hyperpolarised.

Cell proliferation
Proliferation was assayed using a Wst-1 assay kit (Roche), measuring the formation of formazan through the enzymatic cleavage of tetrazolium salt by the mitochondrial dehydrogenases