Sorcin Induces a Drug-Resistant Phenotype in Human Colorectal Cancer by Modulating Ca2+ Homeostasis

Therapeutics, Targets, and Chemical Biology

Sorcin Induces a Drug-Resistant Phenotype in HumanColorectal Cancer by Modulating Ca2þ Homeostasis

Francesca Maddalena1, Gabriella Laudiero3, Annamaria Piscazzi1, Agnese Secondo4, Antonella Scorziello4,Valentina Lombardi1, Danilo Swann Matassa3, Alberto Fersini2, Vincenzo Neri2, Franca Esposito3, andMatteo Landriscina1,5

AbstractThe Ca2þ-binding protein sorcin regulates intracellular calcium homeostasis and plays a role in the induction

of drug resistance in human cancers. Recently, an 18 kDa mitochondrial isoform of sorcin was reported toparticipate in antiapoptosis in human colorectal cancer (CRC), but information remains lacking about thefunctional role of the more abundant 22 kDa isoform of sorcin expressed in CRC.We found the 22 kDa isoform tobe widely expressed in human CRC cells, whether or not they were drug resistant. Its upregulation in drug-sensitive cells induced resistance to 5-fluorouracil, oxaliplatin, and irinotecan, whereas its downregulationsensitized CRC cells to these chemotherapeutic agents. Sorcin enhances the accumulation of Ca2þ in theendoplasmic reticulum (ER), preventing ER stress, and, in support of this function, we found that the 22 kDaisoform of sorcin was upregulated under conditions of ER stress. In contrast, RNAi-mediated silencing of sorcinactivated caspase-3, caspase-12, and GRP78/BiP, triggering apoptosis through the mitochondrial pathway. Ourfindings establish that CRC cells overexpress sorcin as an adaptive mechanism to prevent ER stress and escapeapoptosis triggered by chemotherapeutic agents, prompting its further investigation as a novel molecular targetto overcome MDR. Cancer Res; 71(24); 7659–69. �2011 AACR.

Introduction

Several mechanisms are responsible for inducing both drugresistance in human cancer cells and, among other functions,the upregulation of antiapoptotic genes (1). The investigationof these survival mechanisms responsible for drug resistance isa critical issue, because the likelihood of designing novelmolecular targeted strategies, which would rescue the sensi-tivity of tumor cells to anticancer agents, relies strongly onsuch knowledge. This issue is extremely relevant in the clinicalmanagement of human colorectal cancer (CRC), the secondmost common cause of cancer mortality (2, 3). Of note,although systemic therapy for metastatic CRC has been sig-nificantly improved, the outcome for this malignancy remainspoor, with a median overall survival rate ranging from 18 to 24

months (4). Thus, novel molecular targeted strategies areurgently needed to revert drug resistance and improve theefficacy of systemic therapy in human CRC.

We have previously reported that TRAP1, a mitochondrialchaperone with antioxidant and antiapoptotic functions (5–7),is involved in MDR in human CRC cells (8). Indeed, TRAP1 isupregulated in the majority of human CRCs (8), as well as inother cancer types (9, 10), and its upregulation results in aphenotype resistant tomultiple chemotherapeutics in CRC cells(8). Accordingly, the downregulation of TRAP1 by siRNA resen-sitizes tumor cells to apoptotic stimuli (8, 11). Starting from aproteomic analysis of TRAP1 coimmunoprecipitation com-plexes, we observed that TRAP1 specifically interacts with the18-kDa mitochondrial isoform of sorcin in a Ca2þ-dependentmanner and that this interaction is required for the antiapop-totic function of TRAP1. Interestingly, despite the significanthomology between the 18 and the 22 kDa isoforms of sorcin, the22 kDa isoform is not a TRAP1-interacting protein (11).

Sorcin is a Ca2þ-binding protein and is a member of thepenta-EF-hand protein family (12). Sorcin is a Ca2þ sensorwhich regulates the activity of the ryanodine receptor RyRs, theNaþ/Ca2þ exchanger NCX, and the voltage-dependent L-typeCa2þ channel (12–14) and that, by the interaction with thesetarget proteins, is involved both in regulating Ca2þ homeosta-sis and in modulating excitation–contraction coupling in theheart (15, 16). Some evidence suggests involvement of sorcin inthe drug resistance shown by human malignancies. Indeed,sorcin was purified from a vincristine-resistant lymphoma cellline (17), whereas its overexpression has also been associatedwith resistance to vincristine in gastric cancer cells (18),

Authors' Affiliations: Departments of 1Medical Sciences and 2SurgicalSciences, University of Foggia, Foggia; Departments of 3Biochemistry andMedical Biotechnology and 4Neuroscience, University of Naples FedericoII, Naples; and 5IRCCS CROB, Rionero in Vulture, Italy

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Authors: Matteo Landriscina, Dipartimento di ScienzeMediche e del Lavoro, Universit�a degli Studi di Foggia, Viale Pinto,1–71100 Foggia, Italy. Phone: 39-0881-736241; Fax: 39-0881-733614E-mail: [email protected] and Franca Esposito, Dipartimento di Bio-chimica e Biotecnologie Mediche, Universit�a degli Studi di Napoli FedericoII, Via S. Pansini, 5–80131 Naples, Italy. Phone: 39-081-7463145; Fax:39-081-7464359; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-11-2172

�2011 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 7659

American Association for Cancer Research Copyright © 2011 on January 16, 2012cancerres.aacrjournals.orgDownloaded from

Published OnlineFirst November 3, 2011; DOI:10.1158/0008-5472.CAN-11-2172

http://cancerres.aacrjournals.org/

http://www.aacr.org/

gemcitabine in non–small cell lung cancer (NSCLC; ref. 19) andCHOP regimen in diffuse large B-cell lymphoma (20). Otherreports suggest a correlation between the expression of sorcinand the MDR1/P-glycoprotein: it has been shown that sorcinknockdown induces the upregulation of MDR1 in HeLa cells(21); by contrast, a direct correlation between sorcin andMDR1expression and a role of sorcin in inducing a MDR phenotypehas been observed in human leukemia and gastric carcinomacells (22, 23). Finally, recent studies in oral squamous cellcarcinoma, NSCLC, and acute myeloid leukemia suggest thatsorcin may be responsible for drug resistance and poor prog-nosis (19, 24, 25).

Because the above studies suggest an involvement of the 22kDa sorcin isoform in the cytoprotective pathways of humancancer cells (17–23), without providing a more in-depth expla-nation of its function, and considering that we previouslyshowed that only themitochondrial isoform of sorcin interactswith TRAP1 and is crucial for its antiapoptotic function (11),we became interested in studying the expression of 22 kDasorcin in human CRC and in characterizing its role in the drug-resistant phenotype of this malignancy.

Materials and Methods

Chemicals, cell cultures, constructs, and siRNAsReagents were purchased from Sigma-Aldrich unless oth-

erwise specified. Human HT-29 and HCT-116 colon carcinomacells were purchased from American Type Culture Collection(ATCC) and cultured as previously reported (11). Cell lineswere routinely monitored in our laboratory by microscopicmorphology. Cell line authentication was done before startingthis study by evaluating, respectively, the presence of a muta-tion in codon 13 of the ras proto-oncogene in HCT-116 cells,according to the ATCC product description, and the presenceof the BRAF V600E mutation in HT-29 cells (26). HT-29 CRCcells resistant to single agents were selected as previouslyreported (27).

Full-length 22 kDa sorcin cDNA was cloned in pRC-CMVvector (Invitrogen) and stably transfected in HCT-116 cells.siRNA SRI3, which downregulates both sorcin isoforms, waspurchased from Qiagen (catalogue no. S10048118). siRNAspecific for the 22-kDa sorcin isoform was custom designedas reported in Supplementary Methods. For knockdownexperiments, siRNAs were transfected by using HiPerFectTransfection Reagent according to the manufacturer's proto-col (Qiagen).

Cytotoxicity assaysApoptosis was evaluated by cytofluorimetric analysis of

Annexin V and 7-amino-actinomycin D–positive cells as pre-viously reported (11).

Membrane fractionation, proteinase K digestion, andimmunoblot analysis

Mitochondria and endoplasmic reticulum (ER) were puri-fied using the QproteomeMitochondria Isolation Kit (Qiagen),according to the manufacturer's protocol. In specific experi-ments, the ER fraction was treated with 1 and 4 mg/mL

proteinase K in the presence and absence of NP40 (Igepal)according to Hassink and colleagues (28).

Immunoblot analysis was done as previously reported (11).Specific proteins were detected by using rabbit polyclonal anti-sorcin (Ab57991; Abcam), rabbit polyclonal anti-sorcin (a kindgift fromProf. E. Chiancone, University of Rome "La Sapienza"),mouse monoclonal anti-GAPDH (glyceraldehyde-3-phosphatedehydrogenase; sc-47724; Santa Cruz Biotechnology), mousemonoclonal anti-tubulin (sc-8035; Santa Cruz Biotechnology),mouse monoclonal anti-TRAP1 (sc-13557; Santa Cruz Biotech-nology), rabbit polyclonal anti-Calnexin (M-108, sc-5627; SantaCruz Biotechnology), rabbit polyclonal anti–Caspase-12 fulllength (SPA-827; StressGen), rabbit polyclonal anti–Caspase-12 cleaved form (Ab62463; Abcam), mouse monoclonal anti–Caspase-3 (sc-56051; Santa Cruz Biotechnology), mousemono-clonal anti-COX IV antibodies (MS407; Mitosciences).

RNA extraction and real-time RT-PCR analysisTotal RNA from cell pellets and tumor specimens was

extracted using the TRIzol Reagent (Invitrogen). Primers andexperimental conditions are reported as SupplementaryMethods.

[Ca2þ]i and [Ca2þ]m measurement and mitochondrialmembrane potential

[Ca2þ]i wasmeasured by single-cell computer-assisted videoimaging (29). The equation of Grynkiewicz and colleagues wasused for calibration (30). Ca2þ content into ERwas evaluated ascytosolic Ca2þ release by using the irreversible and selectiveinhibitor of the sarco(endo)plasmic reticulum Ca2þ ATPase(SERCA) thapsigargin (Tg; 1 mmol/L; ref. 31). [Ca2þ]m wasassessed using the fluorescent dye X-Rhod1 (32). Mitochon-drial membrane potential was assessed using the fluorescentdye tetramethyl rhodamine ethyl ester in the "redistributionmode" (32).

Statistical analysisThe paired Student's t and the 1-way ANOVA, followed by

Newman–Keuls tests were used to establish the statisticalsignificance between, respectively, different levels of apoptosis,mitochondrial membrane potential, and [Ca2þ] in transfectedcells and the respective scramble controls. Spearman rank andKendall tau tests were used to establish the statistical corre-lation between sorcin protein levels and, respectively, histo-pathologic parameters and TRAP1 protein levels.

Results

The 22 kDa sorcin isoform is upregulated in humanCRCsand in drug-resistant human CRC cells

The protein levels of sorcin were evaluated in a series of 59human CRCs by immunoblot analysis, and mRNA expressionof 22 kDa sorcin was assessed in a subgroup of 25 CRCs. Figure1A shows an immunoblot analysis of sorcin in 4 CRCs, chosenas representative analytical results of all samples. Character-istics of patient are reported in Supplementary Table S1,whereas protein and mRNA levels in the whole series arereported in Supplementary Table S2. Immunoblot analysis

Maddalena et al.

Cancer Res; 71(24) December 15, 2011 Cancer Research7660





showed that the 22 kDa isoform of sorcinwas upregulated in 28of 59 (47.5%) carcinomas, whereas real-time reverse transcrip-tase PCR (RT-PCR) analysis revealed that only 5 of 14 colorectaltumors with increased 22 kDa sorcin protein levels werecharacterized by a concomitant upregulation of its transcript(Supplementary Table S2), suggesting that both transcriptionaland posttranscriptional mechanisms are involved in the reg-ulation of 22 kDa sorcin expression. By contrast, the 18 kDamitochondrial isoform of sorcin was upregulated in a minorityof human CRCs (11/59 tumors, 18.6%), and its upregulationwas also dependent on either transcriptional or posttran-scriptional mechanisms (data not shown). Interestingly,besides this different expression profile, a significant correla-tion was observed between the protein levels of 18 and22 kDa sorcin isoforms (Spearman rank, P ¼ 0.001; Kendaltau, P ¼ 0.003).The protein expression of sorcin isoforms was analyzed for

the major histopathologic parameters of human CRC. A sig-nificant correlation was observed between the expression of 22kDa sorcin and either tumor grading (Kendal tau, P ¼ 0.01),depth of intestine wall invasion (Spearman rank, P ¼ 0.04;Kendal tau, P ¼ 0.005), presence of lymph node (Spearmanrank, P ¼ 0.05; Kendal tau, P ¼ 0.004), or distant (Kendal tau,P ¼ 0.01) metastases, whereas a trend toward a positivecorrelation was observed between 22 kDa sorcin levels andtumor stage (Kendal tau, P ¼ 0.08). No statistical correlationwas observed between 18 kDa sorcin levels and the samehistopathologic parameters. The expression of 22 kDa sorcinwas also analyzed for TRAP1 protein levels (SupplementaryTable S2) because our previous studies showed that theexpression of this mitochondrial TRAP1/HSP75 chaperone isincreased in 65% of human CRCs (8), and a significant coex-

pression of the two proteins was observed in tumor specimens(Spearman rank, P ¼ 0.002; Kendall tau, P ¼ 0.00001).

The major interest of our group is to identify newmolecularmechanisms/targets for colon cancer chemoresistance. To thisaim, sorcin expression was evaluated in HT-29 CRC cellsresistant to fluorouracil (FU), oxaliplatin (l-OHP), and irino-tecan (IRI), and a significant upregulation of the 22 kDaisoform was observed (Fig. 1B). Real-time RT-PCR showed amodest upregulation of 22 kDa sorcin transcript only in IRI-resistant HT-29 cells (data not shown), consistently with therelevance of posttranscriptional mechanisms in the regulationof sorcin protein expression.

The fractionation of subcellular compartments revealedthat the 22 kDa sorcin band is present in the cytoplasm andin the ER (Fig. 1C and Supplementary Fig. S1), as previouslyreported in ventricular cardiac myocytes (15), whereas the 18kDa isoform, known to be a TRAP1 interactor (11), is restrictedto themitochondria both in HCT-116 CRC cells (Fig. 1C) and inCRCHT-29 FU-resistant cells (Supplementary Fig. S1A), as wellas in a representative case of human CRC (Supplementary Fig.S1B). Considering the functional role of sorcin in ER calciumhomeostasis, the "topology" of sorcin ER localization wasfurther evaluated through biochemical assays on the basis ofprotease digestion. Interestingly, the exposure to proteinase Kof the ER subfraction revealed that 22 kDa sorcin is locatedwithin the ER, similarly to the molecular chaperone calnexin, awell-known ER-resident protein (ref. 33; Fig. 1D).

The 22 kDa isoform of sorcin is involved in MDR inhuman CRC cells

Some evidence suggests that sorcin may be responsible forinducingMDR in human tumors (17–23), but no information is

Cases

Sorcin

Sorcin

Tubulin

HCT-116 cells HCT-116 cells

Cytosol ER Mito PK 4 μg

/mL

PK 1 μg

/mL

PK 4 μg

/mL

PK 1 μg

/mL

Lysa

te ER

NP40 1%

Sorcin

Sorcin

Calnexin

C FU l-OHP IRI

1 2.1 2.2 2.1

HT-29 cells

GAPDHGAPDH

25M TM TM T M T

28

A

C

B

D

29 32

22 kDa18 kDa

34 kDa

50 kDa

22 kDa

90 kDa

22 kDa18 kDa

22 kDa18 kDa

34 kDa

Figure 1. Sorcin expression and distribution in human colorectal cancer cells. A and B, total cell lysates from 4 human CRCs and the respective noninfiltratedperitumoral mucosas (A) and from wild type, FU-, IRI-, and l-OHP-resistant HT-29 CRC cells (B) were separated by SDS–PAGE and immunoblotted with theindicated antibodies. B, densitometric band intensities for 22 kDa sorcin are indicatedby numbers by assuming protein levels of the control (C) equal 1. C, totalcell lysates from HCT-116 CRC cells were separated into cytosolic (Cytosol), mitochondrial (Mito), and ER fractions, separated by SDS–PAGE, andimmunoblotted with the indicated antibodies. D, ER fraction obtained from HCT-116 cells was treated with 1 and 4 mg/mL proteinase K (PK) in the presenceand absence of NP40, separated by SDS–PAGE, and immunoblotted with the indicated antibodies.

Antiapoptotic Role of Sorcin in Human Colorectal Cancer

www.aacrjournals.org Cancer Res; 71(24) December 15, 2011 7661





available on the specific function of the 22 kDa isoform in CRCdrug resistance. Thus, pRC- or 22 kDa sorcin-HCT-116–stabletransfectants (Supplementary Fig. S2A) were cultured for 48hours in the presence of FU, IRI, or l-OHP to evaluate the rate ofapoptotic cell death. Cell sensitivity to each drug is expressedas the ratio between drug-induced and vehicle-induced apo-ptosis (Table 1). Interestingly, protection by 22 kDa sorcin

upregulation against the programmed cell death induced byFU, l-OHP, and IRI was observed. Consistently, the selectivedownregulation of 22 kDa sorcin by siRNA (SupplementaryFig. S2B and C) enhanced the proapoptotic activity of the samethree antiblastic agents by a magnitude similar to thatinduced by the simultaneous knockdown of both sorcin iso-forms (Table 2).

Table 1. Rates of apoptotic cell death in colorectal carcinoma cells transfected with sorcin

Apoptosis (%�SD) Ratio (�SD) Apoptosis (%�SD) Ratio (�SD) P

pRC 22 kDa sorcin

HCT-116 cells scrambleVehicle 5.1 � 0.2 6.5 � 0.110 mmol/L FU 8.8 � 0.5 1.7 � 0.2 4.5 � 0.4 0.7 � 0.1 0.00120 mmol/L FU 21.4 � 0.9 4.2 � 0.3 7.6 � 0.5 1.2 � 0.1 <0.000110 mmol/L l-OHP 28.7 � 2.4 5.6 � 0.7 9.1 � 1.5 1.4 � 0.3 <0.000110 mmol/L IRI 27.0 � 1.7 5.3 � 0.5 14.3 � 1.2 2.2 � 0.2 <0.0001

HCT-116 siRNA TRAP1Vehicle 4.5 � 0.3 5.1 � 0.320 mmol/L FU 34.8 � 2.7 7.7 � 1.2 31.9 � 2.9 6.2 � 1.0 <0.000110 mmol/L l-OHP 51.7 � 3.1 11.5 � 1.5 42.3 � 3.4 8.3 � 1.2 <0.000110 mmol/L IRI 44.4 � 2.2 9.9 � 1.2 38.6 � 3.7 7.5 � 1.3 0.002

NOTE: HCT-116 cells were stably transfectedwith pRC vector control or 22 kDa sorcin constructs and treatedwith FU, l-OHP, or IRI atthe indicated concentrations for 48 hours or with the 3 antiblastic agents for 48 hours upon transient (siRNA) downregulation of TRAP1gene expression. Ratios are calculated between rates of apoptosis in drug- and vehicle-treated cells. P values indicate the statisticalsignificance between the ratios of apoptosis in sorcin- and pRC-transfected cells and in sorcin-overexpressing HCT-116 cellstransfected with TRAP1 or Scramble siRNA.

Table 2. Rates of apoptotic cell death in colorectal carcinomacells upon transient downregulation of sorcin

Apoptosis (%�SD) Ratio (�SD) P Apoptosis (%�SD) Ratio (�SD) P

Vehicle Cyclosporine

ScrambleControl 4.1 � 0.2 5.4 � 0.2FU 6.7 � 0.3 1.6 � 0.2 6.4 � 0.3 1.2 � 0.3l-OHP 10.1 � 0.2 2.5 � 0.1 4.9 � 0.3 0.9 � 0.1IRI 20.3 � 0.4 4.9 � 0.3 12.0 � 0.5 2.2 � 0.2

siRNA Sor22Control 4.1 � 0.1 7.4 � 0.4FU 14.7 � 0.2 3.6 � 0.1 <0.001 13.0 � 0.5 1.7 � 0.2 n.s.l-OHP 22.6 � 0.3 5.5 � 0.2 <0.001 6.7 � 0.2 0.9 � 0.1 n.s.IRI 31.5 � 0.4 7.7 � 0.3 <0.001 17.4 � 0.5 2.3 � 0.2 n.s.

siRNA Sor18/22Control 5.4 � 0.2 7.5 � 0.5FU 16.1 � 0.3 3.0 � 0.1 <0.001 13.8 � 0.6 1.8 � 0.3 n.s.l-OHP 25.2 � 0.3 4.7 � 0.2 <0.001 6.2 � 0.3 0.8 � 0.1 n.s.IRI 37.1 � 0.5 6.9 � 0.3 0.001 16.6 � 0.5 2.2 � 0.1 n.s.

NOTE: HCT-116 cells were treated with 10 mmol/L FU, l-OHP, or IRI for 48 hours in the presence and the absence of 1 mmol/Lcyclosporine A upon transient (siRNA) downregulation of 22 kDa sorcin or both isoforms of sorcin. Ratios are calculated between ratesof apoptosis in drug- and vehicle-treated cells.P values indicate the statistical significance between the ratios of apoptosis induced byantiblastic agents in transfected cells and in the siRNA negative control.Abbreviation: n.s., not significant.

Maddalena et al.






The downregulation of 22 kDa sorcin in drug-resistantCRC cells restores the sensitivity to antiblastic agentsin vitroWe further evaluated whether the selective silencing of 22

kDa sorcin by siRNAs is able to restore sensitivity tochemotherapeutics in drug-resistant cells. To address thisissue, we used wild-type HT-29 cells and HT-29 cells resis-tant to FU, l-OHP, and IRI, which had been previouslycharacterized by increased expression of 22 kDa sorcin

(Fig. 1B). Cells were exposed to FU, l-OHP, or IRI uponsilencing of 22 kDa sorcin or both sorcin isoforms bysiRNAs (Supplementary Fig. S2D). As reported in Table 3,wild-type HT-29 cells treated with FU, l-OHP, and IRIexhibited a significant increase in apoptotic rates, whereasHT-29 cells adapted to growth in the presence of the sameconcentrations of these agents did not exhibit any increasein apoptotic cell death. Interestingly, drug-resistant HT-29cells, depleted of 22 kDa sorcin by siRNAs, revealed

Table 3. Rates of apoptosis inwild-type colorectal carcinomaHT-29 cells and in drug-resistant HT-29 cellsupon transient downregulation of sorcin

Apoptosis (%�SD) Ratio (�SD) P

Wild-type HT-29 cellsScrambleVehicle 4.1 � 0.2FU 11.3 � 1.3 2.7 � 0.5 <0.0001l-OHP 10.4 � 0.9 2.5 � 0.4 <0.0001IRI 12.1 � 1.5 3.0 � 0.5 <0.0001

siRNA Sor22Vehicle 4.2 � 0.4FU 33.7 � 2.3 8.0 � 1.5 <0.0001l-OHP 27.8 � 2.0 6.6 � 1.2 <0.0001IRI 20.7 � 1.8 4.9 � 1.0 <0.0001

siRNA Sor18/22Vehicle 4.6 � 0.3FU 25.4 � 1.9 5.5 � 0.6 <0.0001l-OHP 20.6 � 2.2 4.5 � 08 <0.0001IRI 19.9 � 1.5 4.3 � 0.7 <0.0001

HT-29 FU-R cellsScrambleFU 1.7 � 0.2

siRNA Sor22FU 19.2 � 1.8 11.3 � 2.7 <0.0001

siRNA Sor18/22FU 16.1 � 1.5 9.5 � 2.2 <0.0001

HT-29 l-OHP-R cellsScramblel-OHP 1.7 � 0.1

siRNA Sor22l-OHP 4.4 � 0.4 2.6 � 0.4 <0.0001

siRNA Sor18/22l-OHP 7.1 � 0.9 4.2 � 0.8 <0.0001

HT-29 IRI-R cellsScrambleIRI 2.1 � 0.3

siRNA Sor22IRI 9.0 � 1.1 4.3 � 1.3 <0.0001

siRNA Sor18/22IRI 7.8 � 0.8 3.7 � 1.1 <0.0001

NOTE: Wild-type HT-29 cells and HT-29 cells resistant to FU (FU-R), l-OHP (l-OHP-R), or IRI (IRI-R) were treated with 20 mmol/L FU, 3mmol/L l-OHP, or 1 mmol/L IRI upon transient downregulation of 22 kDa sorcin or both isoforms of sorcin by siRNAs. Ratios arecalculated between drug- or vehicle-induced rates of apoptosis in wild-typeHT-29 cells and between rates of apoptosis in transfectedcells and siRNA negative control in drug-resistant cells. P values indicate the statistical significance between the ratios of apoptosisinduced by antiblastic agents in siRNA transfected and in scramble cells.







increased rates of apoptosis upon treatment with therespective resistance antiblastic agent (Table 3).

The antiapoptotic activity of 22 kDa sorcin involvesthe regulation of Ca2þ homeostasis in the ER

Because sorcin is involved in the regulation of Ca2þ homeo-stasis in cardiomyocytes, mainly by regulating the activity ofthe ryanodine receptor RyRs (12, 15), we questioned whetherthe cytoprotective activity of 22 kDa sorcinmight depend on itsability tomodulate intracellular Ca2þ levels. To this aim, Tg, aninhibitor of the SERCA pump, was used to induce the release ofCa2þ from the ER, thus providing an indirect measure of Ca2þ

levels into the ER (31). In fact, by blocking Ca2þ uptake fromcytosol, Tg was able to induce a progressive and complete iondepletion fromER (Fig. 2A andB). Interestingly, the exposure ofHCT-116 cells to FU determined changes in the ER Ca2þ

content that were dependent on drug concentration. Indeed,after 48 hours of exposure to 10 mmol/L FU, a greater accu-mulation of Ca2þ was detected in all treatment conditions, asan initial response to this toxic stimulus (Fig. 2C). Interestingly,the Tg-induced Ca2þ release reached significantly higher levelsin HCT-116 sorcin transfectants than in all other experimentalconditions, suggesting that the upregulation of 22 kDa sorcinfavors accumulation of Ca2þ into ER. This effect was preventedin 22 kDa sorcin-depleted cells. By contrast, the exposure of thesame CRC cell lines to 40 mmol/L FU for 48 hours, a treatmentthat induces higher levels of apoptosis (data not shown),resulted in significant depletion of ER Ca2þ levels in pRC- andsiRNA-transfected cells and a further increase in ER Ca2þ

accumulation among 22 kDa sorcin transfectants (Fig. 2C). Inthe same experimental conditions, a significant increase incytosolic Ca2þ levels was observed only in pRC- and siRNA-transfected cells exposed to 40 mmol/L FU for 48 hours (seelegend to Fig. 2). Similarly, 22 kDa sorcin HCT-116 transfec-tants treated with 30 mmol/L l-OHP for 48 hours exhibited anincreased release of Ca2þ from the ER compared with therespective untreated conditions (Fig. 2D). Such evidence sug-gests that the overexpression of 22 kDa sorcin in CRC cellsinduces the accumulation of Ca2þ into the ER and that thiscorrelates with resistance to apoptosis.

The 22 kDa sorcin is involved in protecting from ERstress and preventing the opening of the mitochondrialtransition pore

Because the perturbation of Ca2þ homeostasis in the ER is amajor event which triggers ER stress (34), we evaluated wheth-er a correlation between sorcin expression and ER stress couldbe observed in our experimental systems. A significant upre-gulation of 22 kDa sorcin was observed upon treatment ofHCT-116 cells with 2 mmol/L Tg for 7 hours, a condition thatinduced ER stress (ref. 35; Fig. 3A). Accordingly, the selectivedownregulation of 22 kDa sorcin in HCT-116 CRC cells treatedwith 10 mmol/L FU resulted in reduced levels of caspase-3 andthe proteolytic cleavage of the caspase-12 precursor, a molec-ular event which triggers apoptotic signaling in response to ERstress (ref. 35; Fig. 3B). Furthermore, the downregulation of 22kDa or both sorcin isoforms by siRNAs resulted in increasedupregulation of GRP78/BiP, a well-known molecular ER chap-

erone (36), in response to Tg (Fig. 3C). By contrast, the 22 kDasorcin exhibited a protective activity toward ER stress, asrevealed by the reduced levels of GRP78/BiP mRNA upon Tgstimulation compared with pRC-transfected cells (Fig. 3C).Finally, because exposure to 40 mmol/L FU resulted in acondition of ER stress with depletion of Ca2þ levels in the ERand a parallel increase in cytosolic Ca2þ (Fig. 2C), we ques-tioned whether the localization of sorcin was modified by thistreatment. Interestingly, the exposure of HCT-116 cells to 40mmol/L FU for 48 hours resulted in a translocation of 22 kDasorcin from the ER to the cytosolic compartment (Fig. 3D).These results suggest that the upregulation of the 22 kDa sorcinisoform prevents the ER stress response and that its cytopro-tective function might depend on its role in Ca2þ homeostasisin the ER.

The perturbation of Ca2þ homeostasis in the ER and theconsequent ER stress represent conditions that favor apoptosisthrough the opening of the mitochondrial transition pore(MTP; ref. 37); we thus questioned whether sorcin-dependentprotection of apoptosis may involve the mitochondrial anti-apoptotic pathway and the modulation of mitochondrial func-tion. To this aim, the rate of apoptosis induced by antiblasticagents was analyzed (i) in 22 kDa sorcin HCT-116 transfectantsafter selective downregulation of TRAP1, a mitochondrialchaperone known to antagonize the activity of cyclophilinD, an immunophilin that induces mitochondrial cell death(5), and (ii) in HCT-116 cells depleted of 22 kDa sorcin andtreated with cyclosporine A, an inhibitor of MTP opening (38).Interestingly, TRAP1 interference by siRNA restored the sen-sitivity to FU-, l-OHP-, and IRI-induced cell death in 22 kDasorcin HCT-116 transfectants (Table 1), whereas the inhibitionof MTP opening by cyclosporine A prevented apoptosisinduced by antiblastic agents in 22 kDa sorcin-depletedHCT-116 cells (Table 2). These results suggest that the cyto-protective activity of 22 kDa sorcin involves the modulation ofmitochondrial function.

The role of 22 kDa sorcin in themodulation ofmitochondrialactivity was further supported by experiments aimed at mea-suring the mitochondrial membrane potential and calciumconcentration ([Ca2þ]m). Indeed, the silencing of 22 kDa sorcininduced a significant mitochondrial membrane hyperpolari-zation and a reduction in [Ca2þ]m (Fig. 4A), whereas 22 kDasorcin overexpression induced a significant mitochondrialdepolarization and a slight mitochondrial calcium increase(Fig. 4C). These results suggest that, by modulating mitochon-drial membrane potential, 22 kDa sorcin plays a crucial role inthe regulation of mitochondrial calcium efflux. Interestingly,treatment with cytotoxic concentrations of l-OHP (40 mmol/L)in 22 kDa sorcin-silenced HCT-116 cells, although reducingmitochondrial calcium levels to the same extent as untreatedsilenced cells, restored the mitochondrial membrane potential(Fig. 4B). These effects might be responsible for greater sus-ceptibility of 22 kDa sorcin-depleted HCT-116 cells to apopto-sis induced by antiblastic agents. Consistently, the restoring ofmitochondrial membrane potential in 22 kDa sorcin-silencedHCT-116 cells treated with l-OHP, together with lowering of[Ca2þ]m, might preserve metabolic cell rate and ATP produc-tion, thus contributing to apoptotic cell death.

Maddalena et al.






Discussion

Several observations suggest that sorcin is involved in acytoprotective pathway in human malignancies, beingresponsible for drug resistance (17–23). However, to ourknowledge, no evidence has been reported for a role of the 22

kDa sorcin isoform in the MDR phenotype of human CRCs,whereas the functional mechanism proposed for its anti-apoptotic activity is still quite elusive. Sorcin has beenextensively studied in cardiomyocytes due to its role as aCa2þ-sensitive protein, localized in the cytosolic and mem-branous compartments (15) and involved in controlling

130

120

110

100

90

80

70

60

130

120

110

100

90

80

70

60

130

120

110

100

90

80

70

60

0 20 40 60 80

pRC (-FU)pRC (+10 μmol/L FU)

pRC (+40 μmol/L FU)

Sorc (+40 μmol/L FU)

Sorc (+10 μmol/L FU)

5-FU

(10 μmol/L/48 h)

5-FU

(40 μmol/L/48 h)l-OHP

(5 μmol/L/48 h)

l-OHP

(30 μmol/L/48 h)

l-OHP

(40 μmol/L/48 h)

Sorc (-FU)

Tg

A

C D

B

Tg Tg Tg

Tg

Time (sec.)

Time (sec.)

Tg

100 120 140160

0 20 40 60 80 100 120 140160 0 20 40 60 80 100 120 140160 0 20 40 60 80 100 120 140 160

0 20 40 60 80 100 120 140160 0 20 40 60 80 100 120 140 160

[Ca

2+] i

nm

ol/L

120

pRC

Sorcin

si S

or22 Scr

pRC

Sorcin

si S

or22 Scr

pRC

Sorcin

si S

or22 Scr

pRC

Sorcin

si S

or22 Scr

pRC

Sorcin

si S

or22 Scr

pRC

Sorcin

si S

or22 Scr

pRC

Sorcin

si S

or22 Scr

100

80

60

40

Tg-induced [C

a2+] i

rele

ase

(Δ%

incre

ase a

bove

basal va

lues)

Tg-induced [C

a2+] i

rele

ase

(Δ%

incre

ase a

bove

basal va

lues)

20

0

120

100

80

60

40

20

0

130

120

110

100

90

80

70

60

50

130

120

110

100

90

80

70

60

50

130

120

110

100

90

80

70

60

50

[Ca

2+] i

nm

ol/L

**

*

**

* **

**

***

Figure 2. ER Ca2þ levels in HCT-116 wild-type CRC cells and HCT-116 CRC cells overexpressing sorcin. A, superimposed single-cell traces of [Ca2þ]irepresentative of Tg-inducedCa2þ release fromER inHCT-116 cells transfectedwith pRCvector in control conditions andafter treatmentwith 10or 40mmol/LFU for 48 hours. B, superimposed single-cell traces of [Ca2þ]i representative of Tg-induced Ca2þ release from ER in HCT-116 cells transfected with22 kDa sorcin cDNA in control conditions and after treatment with 10 or 40 mmol/L FU for 48 hours. C and D, bar graph for the quantification of Tg effects on[Ca2þ]i increase, calculated as D%of peak/basal value, in HCT-116 cells transfected with pRC vector, 22 kDa sorcin cDNA, siRNA against 22 kDa sorcin (siSor22) or its scramble sequence (Scr) in control conditions (white bars) and after 48 hours of treatment with 10 or 40 mmol/L FU (black bars; C) or after 48 hoursof treatment with 5, 30, or 40 mmol/L l-OHP (black bars; D). For each experiment, 40 to 65 individual cells were monitored. Each bar represents the mean(� SEM) of the aforementioned experimental values studied in 3 independent experimental sessions. C, &, P < 0.05 versus all treatments without drug;�, P < 0.05 versus their respective controls; ��, P < 0.01 versus its control without drug; £, P < 0.05 versus their respective controls and versus 10 mmol/L FU;��,P < 0.05 versus all. In control conditions, the basal values of [Ca2þ]i were 66� 1.7 for pRC, 64� 3.6 for sorcin, 70� 2.7 for siRNA against sorcin, and 74� 4for scramble sequence. No significant change in [Ca2þ]i was detected after 48 hours of treatment with 10 mmol/L FU, whereas with 40 mmol/L FU thebasal values of [Ca2þ]i were 87.5� 4 for pRC (P < 0.05 vs. its respective control condition), 54.3� 2 for sorcin, 88� 3 for siRNA against sorcin (P < 0.05 vs. itsrespective control condition), and 103 � 6 for scramble sequence (P < 0.05 vs. its respective control condition). D, &, P < 0.05 versus all treatmentswithout drug; �, P < 0.05 versus their respective controls without drug; £, P < 0.05 versus their respective controls and versus 30 mmol/L l-OHP. ��, P < 0.05versus all. No significant change in [Ca2þ]i was detected after 48 hours exposure to 5, 30, and 40 mmol/L l-OHP.







intracellular Ca2þ homeostasis, regulating the activity ofseveral Ca2þ transporters (12, 13, 15) and modulating exci-tation–contraction coupling (15, 16).

The interest of our group in studying sorcin originated froma recent observation showing that a new mitochondrial sorcin

isoform (18 kDa), identified by our group, is a TRAP1 inter-acting protein, playing a critical role in the mitochondrialantiapoptotic pathway (11). Indeed, the Ca2þ-dependent inter-action between TRAP1 and 18 kDa sorcin is required for sorcinmitochondrial localization/stability and seems crucial for

Thapsigargin 2 μmol/L

A B

C D

Tubulin

Tubulin

Calnexin

Tg

Sorcin

45’CTR

L

2 h

FU

Scramble

Full length

Cleaved form

Caspase 3

Caspase 1

2

Sorcin

Sorcin

GAPDH

si Sor22

55 kDa

39 kDa

34 kDa

22 kDa

55 kDa

90 kDa

22 kDa

34 kDa

– –

FU

Cytosol ER

––

–––––

4 h

22 kDa

<0.0001

Scramble si Sor22 si Sor18/22 pRc Sor22

<0.0001

<0.0001

100

80

60

40

20

0GR

P78/B

iP r

ela

tive

units

50 kDa

7 h

Figure 3. Sorcin expression and localization in conditions of ER stress and role of sorcin in protecting from ER stress. A, total cell lysates from HCT-116 CRCcells treated with 2 mmol/L Tg for 45 minutes, 2, 4, and 7 hours, were separated by SDS–PAGE and immunoblotted with the indicated antibodies. B, total celllysates from HCT-116 cells, transfected with siRNA against 22 kDa sorcin (siSorc) or its scramble sequence and exposed to 10 mmol/L FU for 48 hours, wereseparated by SDS–PAGE and immunoblotted with the indicated antibodies. Arrow indicates caspase 12 band. C, real-time RT-PCR analysis of GRP78/BiPexpression in HCT-116 cells transfected with pRC vector, 22 kDa sorcin cDNA (Sor22), siRNA against 22 kDa sorcin (si Sor22), siRNA against both sorcinisoforms (si Sor18/22), or their scramble sequence in basal conditions and after 12 hours of treatmentwith 1 nmol/L Tg. D, total cell lysates fromHCT-116 cellswere separated into cytosolic (Cytosol) and ER fractions, separated by SDS–PAGE, and immunoblotted with the indicated antibodies.

180

160

140

120

100

80

60

40

20

0

180

160

140

120

100

80

60

40

20

0Scr

1-OHP

40 μmol/L

1-OHP

40 μmol/L

pRC Sorcin

22

pRC Sorcin

22

si Sor22

A C

B

Scr si Sor22

Scr si Sor22 Scr si Sor22

TM

RE

Arb

itra

ry u

nits o

f F

luore

scence

TM

RE

Arb

itra

ry u

nits o

f fluore

scence

120

100

80

60

40

20

0

120

100

80

60

40

20

0

120

100

80

60

40

20

0

TM

RE

Arb

itra

ry u

nits o

f F

luore

scence

120

100

80

60

40

20

0

X-R

hod 1

Arb

itrary

units

of flu

ore

scence

X-R

hod 1

Arb

itrary

units

of flu

ore

scence

X-R

hod

1

Arb

itrary

un

its o

f fluore

scence

*

*

*

*

*

Figure 4. Sorcin-dependentmodulation of mitochondrialmembrane potential andmitochondrial calciumconcentration. A and B,mitochondrial membrane potential(white bars) and [Ca2þ]m (black bars)in HCT-116 cells transfected withsiRNA against 22 kDa sorcin (siSor22) or its scramble sequence(Scr) in basal condition (A) or aftertreatment with 40 mmol/L l-OHP for48 hours (B). C, mitochondrialmembrane potential (white bars)and [Ca2þ]m (black bars) inHCT-116cells transfected with pRC vector or22 kDa sorcin cDNA in basalconditions. Data are expressed aspercentage of its respective control� SE; �, P < 0.05 versus itsrespective control condition.

Maddalena et al.






TRAP1 regulation of cell survival (11). Although this observa-tion provides mechanistic insights into the cytoprotective roleof the mitochondrial sorcin isoform, suggesting that its anti-apoptotic function depends on its participation in the TRAP1pathway (11), it does not provide any information on thecytoprotective function of the 22 kDa isoform of sorcin, whichis the most abundant cellular isoform and is not a TRAP1interacting protein (11). Here, we report that 22 kDa sorcin isupregulated in about 50% of humanCRCs, and its upregulationinduces protection against drug-induced apoptosis. Because asignificant amount of the protein is localized within the ER andprevious studies have shown that sorcin is involved in mod-ulating the ryanodine receptor RyRs (12, 15), an ER-residentprotein responsible for Ca2þ efflux from the ER (39), wequestioned whether the cytoprotective activity may be medi-ated by its ability to regulate Ca2þ homeostasis in the ER. Thishypothesis is in agreement with several lines of evidencearguing in favor of a role of ER Ca2þ content in the resistanceto stress and apoptotic stimuli (34, 40). Our results suggest thatthe treatment of sorcin overexpressing HCT-116CRC cells withFU and l-OHP induced a dramatic response to Tg, an inhibitorof SERCA, responsible for the depletion of ER Ca2þ contentand, thus, awidely used tool formeasuringCa2þ levels in theER(31). Therefore, it is likely that the upregulation of 22 kDa sorcinfavors the accumulation of Ca2þ into the ER, possibly byinhibiting RyRs receptors (15), and this may be responsiblefor the cytoprotective function exerted by the calcium-bindingprotein. Indeed, sorcin-dependent accumulation of Ca2þ intothe ER seems to be critical in preventing ER stress and inprotecting tumor cells fromapoptosis, because 22 kDa sorcin isupregulated in response to conditions of ER stress. Further-more, ER-associated sorcin translocates to cytosol under con-ditions of high apoptotic cell death, thus determining greaterERCa2þ depletion. By contrast, the selective downregulation of22 kDa sorcin results in the activation of caspase-3 andcaspase-12, in the upregulation of GRP78/BiP, and in thesensitization of tumor cells to drug-induced apoptosis bylowering the threshold of MTP opening. Indeed, downregula-tion of 22 kDa sorcin is associated with hyperpolarization ofthe mitochondrial membrane and reduced calcium content inmitochondria which, in turn, might promote drug-inducedapoptosis.Such evidence is consistent with previous studies suggesting

that modulation of Ca2þ homeostasis is a crucial step in theregulation of cell response to stress conditions and in favoringdrug resistance in tumors (40, 41). Indeed, neuronal cellsexposed to oxygen and glucose deprivation activate theNaþ/Ca2þ exchanger NCX1, which is known to be a targetprotein of sorcin (14), and this is correlated with accumulationof Ca2þ into the ER, and prevention of ER stress and apoptosis(42). Furthermore, increased levels of Ca2þ in the ER have alsobeen correlated with MDR phenotypes in cancer cells (41, 43).Thus, it is intriguing to speculate that sorcin isoforms are bothinvolved in regulating Ca2þ homeostasis in separate cell com-partments and that this function is relevant for their anti-apoptotic activities in tumor cells. Indeed, the 2 proteins,though translated from 2 independent transcripts, are almostidentical in their structure because they share 5 EF-hand

motifs, typical calcium-binding sites, and diverge at theN-terminus due to the absence of 15 amino acids in themitochondrial isoform. Although the role of 18 kDa sorcin inthe regulation of mitochondrial calcium homeostasis needs tobe confirmed by further studies, it is likely that the 22 kDasorcin isoform is one of several ER stress proteins involved inthe control of Ca2þ levels in the ER, preventing ER stress andthe subsequent apoptotic events, and may co-operate with 18kDa sorcin in controlling Ca2þ homeostasis in mitochondriacontributing to the regulation of the MTP opening. Thishypothesis, although still preliminary, sheds some light on therelevance of the TRAP1/18 kDa sorcin interaction and the roleof this sorcin isoform in the TRAP1 cytoprotective pathway.Indeed, the activity of TRAP1 chaperone may be crucial for 18kDa sorcin transport/stability in mitochondria because sorcinlacks a mitochondrial localization sequence (11), and thisinteraction contributes to sorcin-induced regulation of Ca2þ

homeostasis in mitochondria and Ca2þ-dependent MTP reg-ulation. In addition, recent evidence suggests that TRAP1 maybe also involved in the regulation of the unfolded proteinresponse induced by ER stress (44, 45), suggesting a potentialcross-talk between mitochondria and the ER stress responsepathways (46). This hypothesis is consistent with the obser-vation that, in our series of human CRCs, TRAP1 expressioncorrelates with 22 kDa sorcin levels. Thus, it is likely that 22 kDasorcin and TRAP1 are components of a coordinated adaptiveresponse of tumor cells to counteract ER stress conditions andapoptotic signaling.

To our knowledge, this study is the first evidence of a roleplayed by sorcin in resistance to FU, IRI, and l-OHP, 3chemotherapeutics that represent the backbone of humanCRC treatment (47). Apoptosis assays upon selective RNAiknockdown of 22 kDa sorcin in drug-resistant CRC cellssuggest that targeting sorcin may form the basis for a noveltherapeutic strategy for improving the efficacy of che-motherapeutics in CRCs. Indeed, previous studies in severalhuman tumor cell models evaluated verapamil, an agentblocking Ca2þ influx through L-type Ca2þ channels, andsuggested that such a strategy may antagonize P-glycopro-tein–mediated MDR in vitro, but it provided uncertainresults in clinical trials (41). However, because sorcin over-expression has been correlated with upregulation of MDR1/P-glycoprotein (22, 23) and because P-glycoprotein–depen-dent MDR phenotype seems to be related to intracellularCa2þ homeostasis (41), our results suggest that targeting 22kDa sorcin may represent an innovative strategy to preventCa2þ accumulation in the ER and likely revert the MDRphenotype. Thus, further studies are needed to confirm therelationship between sorcin, Ca2þ homeostasis, and theMDR phenotype and to design agents able to modulate ERCa2þ through the inhibition of 22 kDa sorcin activity.Interestingly, several signaling pathways induced by ERstress are currently regarded as novel molecular targets forcancer therapy (34).

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.







Grant Support

This work was supported by AIRC (IG8780 toM. Landriscina and F. Esposito),MIUR (PRIN 2008 to M. Landriscina and F. Esposito), and Fondazione Berlucchi(M. Landriscina and F. Esposito).

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received June 29, 2011; revised September 15, 2011; accepted October 6, 2011;published OnlineFirst November 3, 2011.

References1. Wilson TR, Johnston PG, Longley DB. Anti-apoptotic mechanisms of

drug resistance in cancer. Curr Cancer Drug Targets 2009;9:307–19.2. Ferlay J, Autier P, Boniol M, Heanue M, Colombet M, Boyle P.

Estimates of the cancer incidence and mortality in Europe in 2006.Ann Oncol 2007;18:581–92.

3. Herszenyi L, Tulassay Z. Epidemiology of gastrointestinal and livertumors. Eur Rev Med Pharmacol Sci 2010;14:249–58.

4. Cunningham D, Atkin W, Lenz HJ, Lynch HT, Minsky B, Nordlinger B,et al. Colorectal cancer. Lancet 2010;375:1030–47.

5. Kang BH, Plescia J, Dohi T, Rosa J, Doxsey SJ, Altieri DC. Regulationof tumor cell mitochondrial homeostasis by an organelle-specificHsp90 chaperone network. Cell 2007;131:257–70.

6. Landriscina M, Maddalena F, Laudiero G, Esposito F. Adaptation tooxidative stress, chemoresistance, and cell survival. Antioxid RedoxSignal 2009;11:2701–16.

7. MontesanoGesualdi N,ChiricoG, Pirozzi G, Costantino E, LandriscinaM, Esposito F. Tumor necrosis factor-associated protein 1 (TRAP-1)protects cells from oxidative stress and apoptosis. Stress 2007;10:342–50.

8. Costantino E, Maddalena F, Calise S, Piscazzi A, Tirino V, Fersini A,et al. TRAP1, a novel mitochondrial chaperone responsible for multi-drug resistance and protection from apoptotis in human colorectalcarcinoma cells. Cancer Lett 2009;279:39–46.

9. Leav I, Plescia J, Goel HL, Li J, Jiang Z, Cohen RJ, et al. Cytoprotectivemitochondrial chaperone TRAP-1 as a novel molecular target inlocalized and metastatic prostate cancer. Am J Pathol 2009;176:393–401.

10. Fang W, Li X, Jiang Q, Liu Z, Yang H, Wang S, et al. Transcriptionalpatterns, biomarkers and pathways characterizing nasopharyngealcarcinoma of Southern China. J Transl Med 2008;6:32.

11. Landriscina M, Laudiero G, Maddalena F, Amoroso MR, Piscazzi A,Cozzolino F, et al. Mitochondrial chaperone Trap1 and the calciumbinding protein Sorcin interact and protect cells against apoptosisinduced by antiblastic agents. Cancer Res 2010;70:6577–86.

12. Meyers MB, Pickel VM, Sheu SS, Sharma VK, Scotto KW, FishmanGI.Association of sorcin with the cardiac ryanodine receptor. J Biol Chem1995;270:26411–8.

13. MeyersMB, Puri TS, Chien AJ, Gao T, HsuPH, HoseyMM, et al. Sorcinassociateswith the pore-forming subunit of voltage-dependent L-typeCa2þ channels. J Biol Chem 1998;273:18930–5.

14. Zamparelli C, Macquaide N, Colotti G, Verzili D, Seidler T, Smith GL,et al. Activation of the cardiac Na(þ)-Ca(2þ) exchanger by sorcin viathe interaction of the respective Ca(2þ)-binding domains. J Mol CellCardiol 2010;49:132–41.

15. Farrell EF, Antaramian A, Rueda A, Gomez AM, Valdivia HH. Sorcininhibits calcium release and modulates excitation-contraction cou-pling in the heart. J Biol Chem 2003;278:34660–6.

16. Meyers MB, Fischer A, Sun YJ, Lopes CM, Rohacs T, Nakamura TY,et al. Sorcin regulates excitation-contraction coupling in the heart. JBiol Chem 2003;278:28865–71.

17. Wang SL, Tam MF, Ho YS, Pai SH, Kao MC. Isolation and molecularcloning of human sorcin a calcium-binding protein in vincristine-resistant HOB1 lymphoma cells. Biochim Biophys Acta 1995;1260:285–93.

18. Yang YX, Chen ZC, Zhang GY, Yi H, Xiao ZQ. A subcelluar proteomicinvestigation into vincristine-resistant gastric cancer cell line. J CellBiochem 2008;104:1010–21.

19. QuY,YangY, LiuB, XiaoW.Comparative proteomicprofiling identifiedsorcin being associated with gemcitabine resistance in non-small celllung cancer. Med Oncol 2010;27:1303–8.

20. Maxwell SA, Cherry EM, Bayless KJ. Akt, 14-3-3zeta, and vimentinmediate a drug-resistant invasive phenotype in diffuse large B-celllymphoma. Leuk Lymphoma 2011;52:849–64.

21. Kawakami M, Nakamura T, Okamura N, Komoto C, Markova S,Kobayashi H, et al. Knock-down of sorcin induces up-regulation ofMDR1 in HeLa cells. Biol Pharm Bull 2007;30:1065–73.

22. He Q, Zhang G, Hou D, Leng A, Xu M, Peng J, et al. Overexpression ofsorcin results in multidrug resistance in gastric cancer cells with up-regulation of P-gp. Oncol Rep 2011;25:237–43.

23. Zhou Y, Xu Y, Tan Y, Qi J, Xiao Y, Yang C, et al. Sorcin, an importantgene associated with multidrug-resistance in human leukemia cells.Leuk Res 2006;30:469–76.

24. Nagpal JK, Das BR. Identification of differentially expressed genes intobacco chewing-mediated oral cancer by differential display-poly-merase chain reaction. Eur J Clin Invest 2007;37:658–64.

25. Tan Y, Li G, Zhao C, Wang J, Zhao H, Xue Y, et al. Expression of sorcinpredicts poor outcome in acute myeloid leukemia. Leuk Res 2003;27:125–31.

26. Fukuyo Y, Inoue M, Nakajima T, Higashikubo R, Horikoshi NT, Hunt C,et al. Oxidative stress plays a critical role in inactivating mutant BRAFby geldanamycin derivatives. Cancer Res 2008;68:6324–30.

27. Barone C, Landriscina M, Quirino M, Basso M, Pozzo C, Schinzari G,et al. Schedule-dependent activity of 5-fluorouracil and irinotecancombination in the treatment of human colorectal cancer: in vitroevidence and a phase I dose-escalating clinical trial. Br J Cancer2007;96:21–8.

28. Hassink GC, Zhao B, Sompallae R, Altun M, Gastaldello S, Zinin NV,et al. The ER-resident ubiquitin-specific protease 19 participates inthe UPR and rescues ERAD substrates. EMBO Rep 2009;10:755–61.

29. Secondo A, Molinaro P, Pannaccione A, Esposito A, Cantile M,Lippiello P, et al. Nitric oxide stimulates NCX1 and NCX2 but inhibitsNCX3 isoform by three distinct molecular determinants. Mol Pharma-col 2011;79:558–68.

30. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2þindicators with greatly improved fluorescence properties. J Biol Chem1985;260:3440–50.

31. Janczewski AM, Lakatta EG. Thapsigargin inhibits Ca2þ uptake, andCa2þ depletes sarcoplasmic reticulum in intact cardiacmyocytes. AmJ Physiol 1993;265:H517–22.

32. SecondoA,StaianoRI, ScorzielloA, SirabellaR,Boscia F,AdornettoA,et al. BHK cells transfected with NCX3 are more resistant to hypoxiafollowed by reoxygenation than those transfected with NCX1 andNCX2: Possible relationship with mitochondrial membrane potential.Cell Calcium 2007;42:521–35.

33. Cala SE, Ulbright C, Kelley JS, Jones LR. Purification of a 90-kDaprotein (Band VII) from cardiac sarcoplasmic reticulum. Identificationas calnexin and localization of casein kinase II phosphorylation sites.J Biol Chem 1993;268:2969–75.

34. WangG,Yang ZQ, ZhangK. Endoplasmic reticulum stress response incancer: molecular mechanism and therapeutic potential. Am J TranslRes 2010;2:65–74.

35. Martinez JA, Zhang Z, Svetlov SI, Hayes RL, Wang KK, Larner SF.Calpain and caspase processing of caspase-12 contribute to the ERstress-induced cell death pathway in differentiated PC12 cells. Apo-ptosis 2010;15:1480–93.

36. Ni M, Lee AS. ER chaperones in mammalian development andhuman diseases. FEBS Lett 2007;581:3641–51.

37. Rasola A, Sciacovelli M, Pantic B, Bernardi P. Signal transductionto the permeability transition pore. FEBS Lett 2010;584:1989–96.

Maddalena et al.






38. Boengler K, Hilfiker-Kleiner D, Heusch G, Schulz R. Inhibition ofpermeability transition pore opening by mitochondrial STAT3 and itsrole in myocardial ischemia/reperfusion. Basic Res Cardiol 2010;105:771–85.

39. Ong DS, Mu TW, Palmer AE, Kelly JW. Endoplasmic reticulum Ca2þincreases enhance mutant glucocerebrosidase proteostasis. NatChem Biol 2010;6:424–32.

40. Mattson MP. Neuronal life-and-death signaling, apoptosis, andneurodegenerative disorders. Antioxid Redox Signal 2006;8:1997–2006.

41. Sulova Z, Seres M, Barancik M, Gibalova L, Uhrik B, Polekova L, et al.Does any relationship exist between P-glycoprotein-mediated multi-drug resistance and intracellular calcium homeostasis. Gen PhysiolBiophys 2009;28:F89–95.

42. Sirabella R, Secondo A, Pannaccione A, Scorziello A, Valsecchi V,Adornetto A, et al. Anoxia-induced NF-kappaB-dependent upregula-tion of NCX1 contributes to Ca2þ refilling into endoplasmic reticulumin cortical neurons. Stroke 2009;40:922–9.

43. Padar S, van Breemen C, Thomas DW, Uchizono JA, Livesey JC,Rahimian R. Differential regulation of calcium homeostasis in adeno-carcinoma cell line A549 and its Taxol-resistant subclone. Br J Phar-macol 2004;142:305–16.

44. SiegelinMD,Dohi T,RaskettCM,OrlowskiGM,PowersCM,Gilbert CA,et al. Exploiting the mitochondrial unfolded protein response for cancertherapy in mice and human cells. J Clin Invest 2011;121:1349–60.

45. Takemoto K, Miyata S, Takamura H, Katayama T, Tohyama M. Mito-chondrial TRAP1 regulates the unfolded protein response in theendoplasmic reticulum. Neurochem Int 2011;58:880–7.

46. Amoroso MR, Matassa DS, Laudiero G, Egorova AV, Polishchuk RS,Maddalena F, et al. TRAP1 and the proteasome regulatory particleTBP7/Rpt3 interact in the endoplasmic reticulum and control cellularubiquitination of specific mitochondrial proteins. Cell Death Differ.Epub 2011 Oct 7.

47. O'Neil BH, Goldberg RM. Chemotherapy for advanced colorectalcancer: let's not forget how we got here (until we really can). SeminOncol 2005;32:35–42.







Sorcin Induces a Drug-Resistant Phenotype in Human Colorectal Cancer by Modulating Ca2+ Homeostasis

Documents