Limits...
Comprehensive functional characterization of cancer-testis antigens defines obligate participation in multiple hallmarks of cancer.

Maxfield KE, Taus PJ, Corcoran K, Wooten J, Macion J, Zhou Y, Borromeo M, Kollipara RK, Yan J, Xie Y, Xie XJ, Whitehurst AW - Nat Commun (2015)

Bottom Line: FATE1 prevents the accumulation of the stress-sensing BH3-only protein, BCL-2-Interacting Killer (BIK), thereby permitting viability in the presence of toxic stimuli.Furthermore, ZNF165 promotes TGFβ signalling by directly suppressing the expression of negative feedback regulatory pathways.This action is essential for the survival of triple negative breast cancer cells in vitro and in vivo.

View Article: PubMed Central - PubMed

Affiliation: Simmons Comprehensive Cancer Center, UT-Southwestern Medical Center, Dallas, Texas 75390, USA.

ABSTRACT
Tumours frequently activate genes whose expression is otherwise biased to the testis, collectively known as cancer-testis antigens (CTAs). The extent to which CTA expression represents epiphenomena or confers tumorigenic traits is unknown. In this study, to address this, we implemented a multidimensional functional genomics approach that incorporates 7 different phenotypic assays in 11 distinct disease settings. We identify 26 CTAs that are essential for tumor cell viability and/or are pathological drivers of HIF, WNT or TGFβ signalling. In particular, we discover that Foetal and Adult Testis Expressed 1 (FATE1) is a key survival factor in multiple oncogenic backgrounds. FATE1 prevents the accumulation of the stress-sensing BH3-only protein, BCL-2-Interacting Killer (BIK), thereby permitting viability in the presence of toxic stimuli. Furthermore, ZNF165 promotes TGFβ signalling by directly suppressing the expression of negative feedback regulatory pathways. This action is essential for the survival of triple negative breast cancer cells in vitro and in vivo. Thus, CTAs make significant direct contributions to tumour biology.

Show MeSH

Related in: MedlinePlus

CTA expression is sufficient to activate oncogenic signallingcascades.(a) Indicated cDNAs were co-transfected with the HRE luciferasereporter into HEK293T cells. Following 10 h of stimulation withDMOG, luciferase activity was measured. Bars represent mean(n≥16)±s.e.m. P values calculated byunpaired Student's t-test.****P≤0.0001,***P≤0.0005,**P≤0.01, *P<0.05.(b) H1299 cells were transfected with indicated siRNAs for48 h prior to stimulation for 16 h with1 mM DMOG. Quantitative PCR (qPCR) was used to quantitaterelative mRNA expression of indicated genes normalized to DMOG-stimulatedsiCTRL. Bars represent mean (n=2)±range.(c) KM survival curves of patients from GSE42127 as a function ofIGF2BP3 high- and low-expressing tumours. HR and P value calculatedby Cox Regression Analysis. (d) Indicated cDNAs were co-transfectedwith the Wnt luciferase reporter into HEK293T cells. Following10 h of500 ng ml−1 Wnt-3astimulation, luciferase activity was measured. Bars represent mean(n≥8)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (e) Saos-2 cellswere transfected with indicated siRNAs for 72 h and exposed to500 ng ml−1 Wnt-3a for3 h. QPCR was then performed to quantitate AXIN2 and SPANX mRNAexpression. Bars represent mean (n=2)±range.(f) As in c except the KM plots are a function of SPANXA2expression levels. (g) Indicated cDNAs were co-transfected with theSBE luciferase reporter in HEK293T cells. Following 24 h of100 ng ml−1 TGFβstimulation, luciferase activity was measured. Bars represent mean(n≥4)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (h) WHIM12 cellswere transfected with indicated siRNAs for 48 h and then exposedto 100 ng ml−1 TGFβfor 3 h. QPCR was then performed to quantitate SNAI1 and ZNF165mRNA expression. Bars represent mean(n=2)±range. (i) As in c, exceptKM plots are using patients from the METABRIC data set and are a function ofZNF165 expression.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4660212&req=5

f4: CTA expression is sufficient to activate oncogenic signallingcascades.(a) Indicated cDNAs were co-transfected with the HRE luciferasereporter into HEK293T cells. Following 10 h of stimulation withDMOG, luciferase activity was measured. Bars represent mean(n≥16)±s.e.m. P values calculated byunpaired Student's t-test.****P≤0.0001,***P≤0.0005,**P≤0.01, *P<0.05.(b) H1299 cells were transfected with indicated siRNAs for48 h prior to stimulation for 16 h with1 mM DMOG. Quantitative PCR (qPCR) was used to quantitaterelative mRNA expression of indicated genes normalized to DMOG-stimulatedsiCTRL. Bars represent mean (n=2)±range.(c) KM survival curves of patients from GSE42127 as a function ofIGF2BP3 high- and low-expressing tumours. HR and P value calculatedby Cox Regression Analysis. (d) Indicated cDNAs were co-transfectedwith the Wnt luciferase reporter into HEK293T cells. Following10 h of500 ng ml−1 Wnt-3astimulation, luciferase activity was measured. Bars represent mean(n≥8)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (e) Saos-2 cellswere transfected with indicated siRNAs for 72 h and exposed to500 ng ml−1 Wnt-3a for3 h. QPCR was then performed to quantitate AXIN2 and SPANX mRNAexpression. Bars represent mean (n=2)±range.(f) As in c except the KM plots are a function of SPANXA2expression levels. (g) Indicated cDNAs were co-transfected with theSBE luciferase reporter in HEK293T cells. Following 24 h of100 ng ml−1 TGFβstimulation, luciferase activity was measured. Bars represent mean(n≥4)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (h) WHIM12 cellswere transfected with indicated siRNAs for 48 h and then exposedto 100 ng ml−1 TGFβfor 3 h. QPCR was then performed to quantitate SNAI1 and ZNF165mRNA expression. Bars represent mean(n=2)±range. (i) As in c, exceptKM plots are using patients from the METABRIC data set and are a function ofZNF165 expression.

Mentions: We next turned our investigation to the CTAs returned from screens of regulatorysignalling modules. We prioritized siRNAs from these screens based on the extentand genetic penetrance of their activity, which returned a large set(>10) of CTAs as putative regulators of each pathway (See Methods andSupplementary Data 4). Wereasoned that this large set could be due to multiple indirect mechanismsimpinging on these signalling modules. Therefore, we asked whether CTAs returnedfrom these screens were sufficient to activate the pathways for which they wererequired. Of the six CTAs tested from the HIF screen, LDHC, XAGE1B, PAGE4,CTAGE1 and IGF2BP3 were all sufficient to enhancedimethyloxalylglycine-N-(methoxyoxoacetyl)-glycine methyl ester(DMOG)-mediated activation of the HIF response element (HRE) reporter (Fig. 4a). We further examined IGF2BP3, which exhibited thestrongest effect. IGF2BP3 is expressed in trophoblasts during the firsttrimester of pregnancy, and low expression of IGF2BP3 is associated withattenuated trophoblast invasion and preeclampsia24. Measurementof canonical HIF targets showed that IGF2BP3 depletion attenuated the endogenousexpression of HIF target genes in H1299 lung adenocarcinoma cells (Fig. 4b). Our findings here suggest that reactivation ofIGF2BP3 could inappropriately activate the HIF pathway, promoting invasive andmetastatic behaviour and reducing overall survival. Indeed, in NSCLC, elevatedIGF2BP3 portends poor overall survival (HR=1.84;P=0.0063; Cox Regression) (Fig. 4c).


Comprehensive functional characterization of cancer-testis antigens defines obligate participation in multiple hallmarks of cancer.

Maxfield KE, Taus PJ, Corcoran K, Wooten J, Macion J, Zhou Y, Borromeo M, Kollipara RK, Yan J, Xie Y, Xie XJ, Whitehurst AW - Nat Commun (2015)

CTA expression is sufficient to activate oncogenic signallingcascades.(a) Indicated cDNAs were co-transfected with the HRE luciferasereporter into HEK293T cells. Following 10 h of stimulation withDMOG, luciferase activity was measured. Bars represent mean(n≥16)±s.e.m. P values calculated byunpaired Student's t-test.****P≤0.0001,***P≤0.0005,**P≤0.01, *P<0.05.(b) H1299 cells were transfected with indicated siRNAs for48 h prior to stimulation for 16 h with1 mM DMOG. Quantitative PCR (qPCR) was used to quantitaterelative mRNA expression of indicated genes normalized to DMOG-stimulatedsiCTRL. Bars represent mean (n=2)±range.(c) KM survival curves of patients from GSE42127 as a function ofIGF2BP3 high- and low-expressing tumours. HR and P value calculatedby Cox Regression Analysis. (d) Indicated cDNAs were co-transfectedwith the Wnt luciferase reporter into HEK293T cells. Following10 h of500 ng ml−1 Wnt-3astimulation, luciferase activity was measured. Bars represent mean(n≥8)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (e) Saos-2 cellswere transfected with indicated siRNAs for 72 h and exposed to500 ng ml−1 Wnt-3a for3 h. QPCR was then performed to quantitate AXIN2 and SPANX mRNAexpression. Bars represent mean (n=2)±range.(f) As in c except the KM plots are a function of SPANXA2expression levels. (g) Indicated cDNAs were co-transfected with theSBE luciferase reporter in HEK293T cells. Following 24 h of100 ng ml−1 TGFβstimulation, luciferase activity was measured. Bars represent mean(n≥4)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (h) WHIM12 cellswere transfected with indicated siRNAs for 48 h and then exposedto 100 ng ml−1 TGFβfor 3 h. QPCR was then performed to quantitate SNAI1 and ZNF165mRNA expression. Bars represent mean(n=2)±range. (i) As in c, exceptKM plots are using patients from the METABRIC data set and are a function ofZNF165 expression.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4660212&req=5

f4: CTA expression is sufficient to activate oncogenic signallingcascades.(a) Indicated cDNAs were co-transfected with the HRE luciferasereporter into HEK293T cells. Following 10 h of stimulation withDMOG, luciferase activity was measured. Bars represent mean(n≥16)±s.e.m. P values calculated byunpaired Student's t-test.****P≤0.0001,***P≤0.0005,**P≤0.01, *P<0.05.(b) H1299 cells were transfected with indicated siRNAs for48 h prior to stimulation for 16 h with1 mM DMOG. Quantitative PCR (qPCR) was used to quantitaterelative mRNA expression of indicated genes normalized to DMOG-stimulatedsiCTRL. Bars represent mean (n=2)±range.(c) KM survival curves of patients from GSE42127 as a function ofIGF2BP3 high- and low-expressing tumours. HR and P value calculatedby Cox Regression Analysis. (d) Indicated cDNAs were co-transfectedwith the Wnt luciferase reporter into HEK293T cells. Following10 h of500 ng ml−1 Wnt-3astimulation, luciferase activity was measured. Bars represent mean(n≥8)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (e) Saos-2 cellswere transfected with indicated siRNAs for 72 h and exposed to500 ng ml−1 Wnt-3a for3 h. QPCR was then performed to quantitate AXIN2 and SPANX mRNAexpression. Bars represent mean (n=2)±range.(f) As in c except the KM plots are a function of SPANXA2expression levels. (g) Indicated cDNAs were co-transfected with theSBE luciferase reporter in HEK293T cells. Following 24 h of100 ng ml−1 TGFβstimulation, luciferase activity was measured. Bars represent mean(n≥4)±s.e.m. P values calculated by unpairedStudent's t-test as in a. (h) WHIM12 cellswere transfected with indicated siRNAs for 48 h and then exposedto 100 ng ml−1 TGFβfor 3 h. QPCR was then performed to quantitate SNAI1 and ZNF165mRNA expression. Bars represent mean(n=2)±range. (i) As in c, exceptKM plots are using patients from the METABRIC data set and are a function ofZNF165 expression.
Mentions: We next turned our investigation to the CTAs returned from screens of regulatorysignalling modules. We prioritized siRNAs from these screens based on the extentand genetic penetrance of their activity, which returned a large set(>10) of CTAs as putative regulators of each pathway (See Methods andSupplementary Data 4). Wereasoned that this large set could be due to multiple indirect mechanismsimpinging on these signalling modules. Therefore, we asked whether CTAs returnedfrom these screens were sufficient to activate the pathways for which they wererequired. Of the six CTAs tested from the HIF screen, LDHC, XAGE1B, PAGE4,CTAGE1 and IGF2BP3 were all sufficient to enhancedimethyloxalylglycine-N-(methoxyoxoacetyl)-glycine methyl ester(DMOG)-mediated activation of the HIF response element (HRE) reporter (Fig. 4a). We further examined IGF2BP3, which exhibited thestrongest effect. IGF2BP3 is expressed in trophoblasts during the firsttrimester of pregnancy, and low expression of IGF2BP3 is associated withattenuated trophoblast invasion and preeclampsia24. Measurementof canonical HIF targets showed that IGF2BP3 depletion attenuated the endogenousexpression of HIF target genes in H1299 lung adenocarcinoma cells (Fig. 4b). Our findings here suggest that reactivation ofIGF2BP3 could inappropriately activate the HIF pathway, promoting invasive andmetastatic behaviour and reducing overall survival. Indeed, in NSCLC, elevatedIGF2BP3 portends poor overall survival (HR=1.84;P=0.0063; Cox Regression) (Fig. 4c).

Bottom Line: FATE1 prevents the accumulation of the stress-sensing BH3-only protein, BCL-2-Interacting Killer (BIK), thereby permitting viability in the presence of toxic stimuli.Furthermore, ZNF165 promotes TGFβ signalling by directly suppressing the expression of negative feedback regulatory pathways.This action is essential for the survival of triple negative breast cancer cells in vitro and in vivo.

View Article: PubMed Central - PubMed

Affiliation: Simmons Comprehensive Cancer Center, UT-Southwestern Medical Center, Dallas, Texas 75390, USA.

ABSTRACT
Tumours frequently activate genes whose expression is otherwise biased to the testis, collectively known as cancer-testis antigens (CTAs). The extent to which CTA expression represents epiphenomena or confers tumorigenic traits is unknown. In this study, to address this, we implemented a multidimensional functional genomics approach that incorporates 7 different phenotypic assays in 11 distinct disease settings. We identify 26 CTAs that are essential for tumor cell viability and/or are pathological drivers of HIF, WNT or TGFβ signalling. In particular, we discover that Foetal and Adult Testis Expressed 1 (FATE1) is a key survival factor in multiple oncogenic backgrounds. FATE1 prevents the accumulation of the stress-sensing BH3-only protein, BCL-2-Interacting Killer (BIK), thereby permitting viability in the presence of toxic stimuli. Furthermore, ZNF165 promotes TGFβ signalling by directly suppressing the expression of negative feedback regulatory pathways. This action is essential for the survival of triple negative breast cancer cells in vitro and in vivo. Thus, CTAs make significant direct contributions to tumour biology.

Show MeSH
Related in: MedlinePlus