galnt1-mediatedglycosylationandactivationof sonic hedgehog ... · shh (santa cruz biotechnology,...

Tumor and Stem Cell Biology

GALNT1-MediatedGlycosylationandActivationofSonic Hedgehog Signaling Maintains the Self-Renewal and Tumor-Initiating Capacity of BladderCancer Stem CellsChong Li1, Ying Du1, Zhao Yang1, Luyun He1, Yanying Wang1, Lu Hao1, Mingxia Ding2,Ruping Yan2, Jiansong Wang2, and Zusen Fan1

Abstract

The existence of bladder cancer stem cells (BCSC) has beensuggested to underlie bladder tumor initiation and recurrence.Sonic Hedgehog (SHH) signaling has been implicated in pro-moting cancer stem cell (CSC) self-renewal and is activated inbladder cancer, but its impact on BCSC maintenance is unclear.In this study, we generated a mAb (BCMab1) against CD44þ

human bladder cancer cells that recognizes aberrantly glycosy-lated integrin a3b1. The combination of BCMab1 with an anti-CD44 antibody identified a BCMab1þCD44þ cell subpopula-tion as BCSCs with stem cell–like properties. Gene expressionanalysis revealed that the hedgehog pathway was activated inthe BCMab1þCD44þ subpopulation and was required forBCSC self-renewal. Furthermore, the glycotransferase GALNT1

was highly expressed in BCMab1þCD44þ cells and correlatedwith clinicopathologic features of bladder cancers. Mechanis-tically, GALNT1 mediated O-linked glycosylation of SHHto promote its activation, which was essential for the self-renewal maintenance of BCSCs and bladder tumorigenesis.Finally, intravesical instillation of GALNT1 siRNA and the SHHinhibitor cyclopamine exerted potent antitumor activity againstbladder tumor growth. Taken together, our findings identify aBCSC subpopulation in human bladder tumors that appearsto be responsive to the inhibition of GALNT1 and SHH sig-naling, and thus highlight a potential strategy for preventingthe rapid recurrence typical in patients with bladder cancer.Cancer Res; 76(5); 1273–83. �2015 AACR.

IntroductionBladder cancer is the second most common urological malig-

nancy with over 69,000 new cases and an estimated 150,000deaths worldwide each year (1–3). The big problem of bladdercancer is the extremely easy recurrence after treatment and norecognized second-line therapy (4). Thus, for new therapeuticstrategies, it is of great importance to delineate the mechanism ofbladder cancer tumorigenesis.

Accumulating evidence from leukemias, germ cell tumors, andmany solid tumors support a cancer stem cell (CSC) model inwhich cancers are initiated by a rare subpopulation of self-renew-ing and differentiating tumor-initiating cells (TIC; refs. 5–7). Thecancer stem cell (CSC) population can self-renew limitlessly and

differentiate into heterogeneous cell populations, resulting inphenotypic and functional heterogeneity of tumors (8, 9). Qui-escent CSCs may resist therapeutic intervention and confer torecurrence after initial therapy (7). Chan and colleagues identifieda subpopulation (Lineage�CD44þCK5þCK20�) from primaryhuman bladder cancer as bladder cancer stem cells (BCSC)(10), which shared similar expression markers to normal bladderbasal cells, suggesting the basal-like property of BCSCs. A sub-population of primary urothelial cancers, coexpressing the 67-kDa laminin receptor (67LR) and the basal cell–specific cytoker-atinCK17, possessesmore carcinogenic ability (11). These subsetsharbor high tumorigenicity, multipotency, and involvement inbladder tumor progression, suggesting that the CSCs exist inbladder tumors.

Bladder cancer arises from the urothelium, a transitional epi-thelium lining the bladder lumen. Thismultilayered epithelium iscomposed of a luminal layer of fully differentiated umbrella cellsthat overlie intermediate cells, and a basal layer of Sonic Hedge-hog (SHH)-expressing cells (12). The SHH-expressing basal cells,capable of regenerating all cell types of the urothelium, have beendefined as urothelial stem cells. The same group reported thatmuscle-invasive bladder cancers originate exclusively from theSHH-expressing basal stem cells of the basal urothelium inprocarcinogen N-butyl-N-4-hydroxybutyle nitrosamine (BBN)-induced mouse bladder cancer (13). These SHH-expressing basalstem cells act as BCSCs to initiate bladder oncogenesis. Aberrantactivation of the Hedgehog signaling is implicated in manymalignancies (14, 15). Emerging data support that Hh signalingplays a pivotal role in the self-renewal and differentiation of CSCs

1CAS Key Laboratory of Infection and Immunity, Institute of Biophys-ics, Chinese Academy of Sciences, Beijing, China. 2Department ofUrology, The Second Affiliated Hospital of Kunming Medical Univer-sity, Kunming, China.

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

C. Li, Y. Du, Z. Yang contributed equally to this article.

Corresponding Authors: Zusen Fan, Institute of Biophysics, Chinese Academyof Sciences, 15 Datun Road, Room 1611, Beijing 100101, China. Phone: 8610-6488-8457; Fax: 8610-6487-1293; E-mail: [email protected]; and Jiansong Wang,E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-15-2309

�2015 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 1273

on February 15, 2021. © 2016 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst December 16, 2015; DOI: 10.1158/0008-5472.CAN-15-2309

http://cancerres.aacrjournals.org/

(16, 17). Activation of Hh pathway causes CSC self-renewal andexpansion, while the SMO antagonist cyclopamine or the ligand-neutralizing antibody induces terminal differentiation and loss ofclonogenic growth capacity (16, 18). In this study, we defined thatHh signaling is essential for the self-renewal of BCSCs. GALNT1-mediated glycosylation is required for SHH activation in BCSCs.

Materials and MethodsReagents and mice

Mouse mAb BCMab1 was purified through protein A-Sephar-ose from ascites. Corresponding species-specific FITC-conjugatedsecondary antibodies were purchased from Sigma. The commer-cial primary antibodies were phycoerythrin (PE)-conjugated anti-CD44 (BDPharmingen; 550989), goat anti-humanC-terminus ofSHH (Santa Cruz Biotechnology, sc-33942), goat anti-human N-terminus of SHH (Santa Cruz Biotechnology; sc-1194), rabbitanti-Gli1 (Santa Cruz Biotechnology, sc-20687), goat anti-GALNT1 (Santa Cruz Biotechnology, sc-68491), and mouseanti-b-actin antibody (Sigma, A1978). NOD/SCID mice wereobtained from Vital River Laboratory Animal Technology Co.Ltd.. GALNT1�/� mice were obtained from Jackson Laboratories.BBN, benzyl-N-acetyl-D-galactosaminide (BG), and cyclopaminewere from Sigma. GALNT1 siRNA (human) and GALNT1 siRNA(mouse) were from Santa Cruz Biotechnology.

Xenograft establishmentHuman bladder cancer specimens were obtained with patient's

consent, as approved by the Research Ethics Board at The SecondAffiliated Hospital of Kunming Medical University (Kunming,China). For the generation of xenografts, cells were injected withMatrigel subcutaneously into the backs of NOD/SCID mice.Animal work was carried out in compliance with the ethicalregulations approved by the Animal Care Committee, Instituteof Biophysics, Chinese Academy of Sciences (Beijing, China).

Flow cytometry sortingPrimary tumor tissues were digested with 200U type I colla-

genase, 20U type IV collagenase (Sigma, C-0130, C-5138), andDNase at 37�C for 2 to 6 hours. Then, the cancer cells were stainedwith FITC-conjugated BCMab1 and PE-conjugated anti-CD44antibody or the same isotype FITC/PE–conjugated antibody for30minutes on ice. Labeled cells were analyzed and sorted by flowcytometry (BD FACSAria III).

DNA methylation analysisMethylation was assayed by using the procedure as described

previously (19). GenomicDNAwas treatedwith sodiumbisulfateusing the EZ DNA methylation direct kit (Zymo Research). PCRproducts were analyzed with the MassArray EpiTYPER system(Invitrogen).

Quantitative real-time PCRTotal RNA was isolated from primary sorted cells, human

bladder cancer cell lines, and mouse tissues using the RNAIsolation Kit (Tiangen Biotech). RNAwas then subjected to cDNAsynthesis using M-MLV Reverse Transcriptase (Promega). ThecDNAwas then processed for an amplification step with the SYBRGreen reaction system (Tiangen Biotech) running on an ABI 7300(AppliedBiosystems). Primer oligonucleotides for qPCRare listedin the Supplementary Table S1.

In vitro glycosylation assayRecombinant human wild-type (WT) SHH and T282A-SHH

mutant proteins were expressed in E. coli. WT SHHor T282A-SHHmutant was incubated with GALNT1 (R&D Systems) plusBCMab1�CD44� cell lysates at 37�C for 4 hours. Reaction pro-ducts were analyzed by immunoblotting.

Immunohistochemical stainingTumor specimens were fixed in 10% buffered formalin and

embedded in paraffin as described (20). The immunosignals weredetected using the ABC kit.

Intravesical administration of siRNA/liposomes andcyclopamine for treatment

Female 6-week-old NOD/SCID mice with a body weight ofapproximately 15 g were used and kept under specific pathogen-free conditions. BCMab1þCD44þ cells were labeled with lucifer-ase and transplanted into the murine bladder cavity via 24-gaugeangiocatheters (21). The mice were grouped (24 mice per group)and intravesically applied with 6 mmol/L siCtrl/liposomes orsiGALNT1/liposomes with or without 100 mmol/L cyclopaminenext day.

Statistical analysisThe relationship between the expression levels of GALNT1 and

clinicopathologic features was analyzed using the c2 or theKruskal–Wallis test. Kaplan–Meier analysis was used to estimatethe cumulative cause-specific survival rates, and the log-rank testwas used to correlate differences in patient survival with stainingintensity of GALNT1. The influence of GALNT1 on the growth ofbladder cancers was analyzed by the Student t test.

ResultsA BCMab1þCD44þ subpopulation of cancerous cellsacts as BCSCs

To identify unique biomarkers of BCSCs, we isolated CD44þ

BCSCs from human bladder cancer resection specimens (Fig. 1A),which were used as immunogen to immunize mice to generatemAbs. About 458 hybridomas were screened to bind BCSCs.Among these hybridomas, we identified several antibodies,including BCMab1 and BCMab37, which were determined torecognize the BCSC surface antigens, aberrantly glycosylatedintegrin a3b1 and CD47, respectively. We previously showedthat the BCMab1 antibody was also generated by T24 cell immu-nization, recognizing the aberrantly glycosylated integrin a3b1on bladder cancer cells (20). The aberrantly glycosylated integrina3b1 is highly expressed in bladder cancer tissues. CD47 was alsoconfirmed to be a surface marker of BCSCs (11).

By using BCMab1 and anti-CD44 antibodies, we distinguisheda BCMab1þCD44þ subset from the tumor cells of human bladdertumor specimens (Fig. 1A), comprising 3% to 5% of the total cellpopulation. BCMab1þCD44þ cells were further confirmed byimmunofluorescence staining (Fig. 1C). We next determinedwhether BCMab1þCD44þ cells weremore tumorigenic than theirBCMab1�CD44� counterparts. BCMab1þCD44þ cells freshlyisolated from bladder cancer specimens were inoculated subcu-taneously into NOD/SCID mice. As few as 10 BCMab1þCD44þ

cells could form xenograft tumors in vivo (Fig. 1D and E), whereasat least 1� 105 BCMab1�CD44� cells were necessary to establishxenografts. The xenograft tumors formed with BCMab1þCD44þ

Li et al.

Cancer Res; 76(5) March 1, 2016 Cancer Research1274




Figure 1.A BCMab1þCD44þ subpopulation of cancerous cells is defined as BCSCs. A, representative cystoscope of a typical bladder cancer. � , bladder tumor. B, a subset ofBCMab1þCD44þ cells was fractionated in one human bladder cancer specimen. Single tumor cells were stained with BCMab1 (B) and anti-CD44 (C) antibodiesfollowed by flow cytometry. C, immunofluorescence staining of BCMab1þCD44þ (left) and BCMab1�CD44� (right) cells. D, tumor formation frequencyof BCSCs from unsorted, BCMab1þCD44þ, and BCMab1�CD44� subpopulations derived from bladder cancer patient specimens. CI, confidence interval.Six bladder cancer samples got similar results. E, representative photographs demonstrating xenografted tumors (arrowheads). Bladder tumor cell subpopulations(BCMab1þCD44þ or BCMab1�CD44�) with limiting dilutions were injected subcutaneously into NOD/SCID mice with the indicated numbers of cells.F, FACS analysis of BCMab1 and CD44 expression in xenograft tumors generated from 1� 105 BCMab1þCD44þ or BCMab1�CD44� cells. G, injection of 10, 100, 1,000,10,000, and 100,000 freshly isolated BCMab1þCD44þ (left) or BCMab1�CD44� cells (right) derived from the primary graft tumors. Data represent means � SD;n ¼ 6. H, serial tumor formation assay. Tumor volumes were measured at the indicated time points and calculated as means � SD in BCMab1þCD44þ (left) orBCMab1�CD44� (right) cell transplantations; n ¼ 6. I, oncospheres were counted in five separate fields after two weeks and are shown as means � SD.

GALNT1-Mediated SHH Activation in BCSCs

www.aacrjournals.org Cancer Res; 76(5) March 1, 2016 1275




cells and BCMab1�CD44� cells in Fig. 1E were used for furtheranalysis. Intriguingly, the tumors induced by BCMab1þCD44þ

cells were able to reconstitute a mixture of BCMab1þCD44þ cellsand differentiated BCMab1�CD44� tumor cells (Fig. 1F), but thetumors grafted by BCMab1�CD44� cells could not reconstitutethe BCMab1þCD44þ cell subpopulation.

In addition, the self-renewal capacity of BCMab1þCD44þ cellswas evaluatedbyan in vivo serialpassage assay.BCMab1þCD44þorBCMab1�CD44� cells were sorted from the primary mousexenografts induced by the BCMab1þCD44þ cells and implantedinto secondary recipient NOD/SCID mice accordingly. The sameprocedurewasused for the third and fourthpassages.All the seriallyxenografted mice injected with BCMab1þCD44þ cells were ableto form tumors (Fig. 1GandH),which grewmuch fasterwith serialpassages. However, xenografted tumors injected with BCMab1�

CD44� cells only formedwith high cell number inoculation,whilethe formed xenografts were smaller even with multiple passages.We next examined the self-renewal property of BCMab1þCD44þ

cells by an in vitro sphere formation assay. BCMab1þCD44þ

cells freshly isolated from human bladder cancer specimensexhibited significantly higher oncosphere-forming efficiencycompared with their respective BCMab1�CD44� counterparts(Fig. 1I). Taken together, the BCMab1þCD44þ subpopulation ishighly tumorigenic, with self-renewal and differentiation proper-ties. Thus, we defined this BCMab1þCD44þ subset as the BCSC.

Hedgehog signaling is essential for the self-renewal of BCSCsSHH-expressing basal stem cells, acting as BCSCs, initiate

tumorigenesis in BBN-induced mouse bladder cancer (15).These results suggest that Hh signaling may be implicatedin the stemness of BCSCs. To determine the Hh pathway ofBCMab1þCD44þ cells, we performed transcriptome microar-ray analysis of BCMab1þCD44þ and BCMab1�CD44� cellsisolated from human bladder tumor specimens. Microarrayanalysis revealed that several Hh signaling molecules, includ-ing SHH, Smo, Gli1, and Gli2, were highly expressed inBCMab1þCD44þ cells compared with their BCMab1�CD44�

counterparts (Fig. 2A). High Gli1 mRNA levels in BCMab1þ

Figure 2.Hh signaling is essential for the self-renewal of BCSCs. A, scatter plot comparing global gene expression profiles between BCMab1þCD44þ andBCMab1�CD44� cells.Black lines, 2-fold differences in either direction in gene expression levels. Arrowheads, Hh signaling–related genes. B, real-time PCR analysis ofGli1 mRNA levels in BCMab1þCD44þ cells derived from 72 bladder cancer specimens. �� , P < 0.01. C, immunohistochemical staining of human bladder cancertissues and pericarcinoma tissues with ant-SHH and anti-Gli1 antibodies. D, immunofluorescence staining in BCMab1þCD44þ or BCMab1�CD44� cells byBCMab1-FITC, anti-CD44-PE and anti-Gli1-PE–labeled antibodies. White arrowheads, nuclear localization of Gli1. E, Western blot analysis of SHH and Gli1 proteinlevels in BCMab1þCD44þ cells pooled from bladder cancer samples. F and G, cyclopamine significantly inhibits both oncosphere formation and anchorage-independent tumor growth of BCMab1þCD44þ cells. BCMab1þCD44þ cells were treated with cyclopamine for 24 hours prior to oncosphere formation (F) orsoft agar colony (G) assays. All data are representative of four independent experiments and are shown as means � SD. �� , P < 0.01. CPM, cyclopamine.

Li et al.





CD44þ cells were verified in BCMab1þCD44þ cells derivedfrom each specimen of 72 human bladder cancer samples(Fig. 2B). Bladder cancer tissues displayed high expression ofSHH and Gli1 by immunohistochemical staining (Fig. 2C). Inaddition, Gli1 was highly expressed in the nuclei of BCMab1þ

CD44þ cells, but undetectable in BCMab1�CD44� cells (Fig.2D). More importantly, SHH was highly expressed inBCMab1þCD44þ cells and apparently hydrolyzed to form itscatalyzed band (Fig. 2E), but undetectable in BCMab1�CD44�

cells. These data suggest that the Hh signaling pathway isactivated in BCSCs.

To further confirm the involvement of Hh signaling in BCSCs,we used the SMO antagonist cyclopamine to treat BCMab1þ

CD44þ cells for an in vitro sphere formation assay. We observedthat cyclopamine treatment dramatically impaired oncosphereformation (Fig. 2F). In addition, using a soft agar colony forma-tion assay, cyclopamine treatment also significantly repressedanchorage-independent tumor growth (Fig. 2G). Overall, the Hhsignaling pathway is essential for themaintenance of self-renewalof BCSCs.

GALNT1 is highly expressed in BCSCsWe performed transcriptome microarray of BCMab1þCD44þ

cells versus BCMab1�CD44� cells derived from bladder cancersamples. Microarray analysis displayed that the glycotransferaseGALNT1 mRNA level in BCMab1þCD44þ was over 7-fold higherthan that in BCMab1�CD44� cells (Fig. 3A). However, otherglycotransferases (GALNT4, 7, 8, 9, and 12) that the microarraydetected had no significant changes in BCMab1þCD44þ cellscompared with BCMab1�CD44� cells. As the GALNT family has12members, we then examinedother undetected glycotransferasemembers in the microarray, including GALNT2, 3, 5, 6, 10, and11, in the BCMab1þCD44þ cells. We noticed that these unde-tected glycotransferases were not expressed in the BCMab1þ

CD44þ cells (data not shown). These results were verified inBCMab1þCD44þ cells isolated from72bladder cancer specimensby using qPCR (Fig. 3B and C). Moreover, high expression ofGALNT1 in pooled BCMab1þCD44þ cells from ten bladdercancer specimens was further validated by immunoblotting (Fig.3D). High expression of GALNT1 in bladder cancer tissues wasconfirmed by immunohistochemical staining (Fig. 3E). In addi-tion, GALNT1-positive staining in bladder cancer tissues wasrelated to BCMab1 and CD44 staining.

To further verify the high expression of GALNT1 in BCMab1þ

CD44þ cells, we examined the 50 CpG sites at the GALNT1promoter of genomic DNAs by bisulfite genomic sequencinganalysis. Each group displayed ten independent sequencingclones. The GALNT1 promoter region was over 90% unmethy-lated in BCMab1þCD44þ cells, whereas the same cytosine resi-dues were about 60% methylated in BCMab1�CD44� cells (Fig.3F). Similar results were obtained with other three bladder cancerspecimens. In sum, these results suggest that the GALNT1 isindeed actively transcribed in BCMab1þCD44þ cells.

GALNT1 is required for themaintenance of stemness for BCSCsWe next wanted to determine whether GALNT1 takes part in

the regulation of BCSCs. We silenced GALNT1 in BCMab1þ

CD44þ cells derived from one human bladder cancer specimenand established GALNT1-depleted cell lines (Fig. 4A). Wefound that GALNT1 knockdown significantly declined onco-sphere formation and anchorage-independent colony forma-tion (Fig. 4B and C). BG is a competitive substrate inhibitor forall glycotransferases (22). BG treatment obtained similar resultsto GALNT1 depletion (Fig. 4B and C). We detected mRNA levelsof Gli1 and GALNT1 in BCMab1þCD44þ cells derived from 72bladder cancer samples by qPCR. Interestingly, Gli1 displayed agood correlation with GALNT1 expression (Pearson correlationcoefficient, r ¼ 0.945; Fig. 4D).

Figure 3.GALNT1 is highly expressed in BCSCs.A, heatmap represents the relativeexpression of GALNT1, integrin a3b1,CD44, and Gli1 in BCMab1þCD44þ andBCMab1�CD44� cells. B, real-timePCR analysis of mRNA levels ofGALNTs in BCMab1þCD44þ andBCMab1�CD44� cells derived frompatient samples. �� , P < 0.01; n ¼ 6.C and D, real-time PCR (C) andWestern blot analysis (D) of GALNT1expression levels in BCMab1þCD44þ

and BCMab1�CD44� cells derivedfrom 72 bladder cancer specimens.E, immunohistochemical staining ofhuman bladder cancer tissues withBCMab1, anti-CD44, and anti-GALNT1antibodies. H&E, hematoxylin andeosin. F, bisulfite genomic sequencinganalysis of the 50 CpG sites at theGALNT1 promoter. Each groupincludes ten independent clones.Black dots, methylated CpG; graydots, unmethylated CpG.






Importantly, we observed that GALNT1 knockdown or BGtreatment dramatically reduced Gli1 expression in BCMab1þ

CD44þ cells (Fig. 4E). These results were verified by immunoflu-orescence staining (Fig. 4F). We next overexpressed Gli1 and Gli2in BCMab1�CD44� cells followed by oncosphere formationassays. We observed that Gli1 but not Gli2 overexpression couldpromote oncosphere formation (Supplementary Fig. S1A), sug-gesting that Gli1 was activated in BCSCs. In addition, GALNT1knockdown or BG treatment significantly declined active SHH-Nprotein levels in BCMab1þCD44þ cells (Fig. 4G), suggesting that

GALNT1 depletion impaired SHH processing. Consequently,GALNT1-silenced BCMab1þCD44þ cells formed smaller tumorburden and displayed much lower xenograft tumor formationrates compared with shCtrl-treated cells (Fig. 4H and I).

We next rescued SHH and/or GALNT1 in GALNT1-silencedBCMab1�CD44� cells. We found that rescue of both SHH andGALNT1 was able to process SHH and induce Gli1 expression(Supplementary Fig. S1B). In contrast, rescue of SHH in GALNT1-silenced BCMab1�CD44� cells neither hydrolyzed SHH nortriggered Gli1 expression (Supplementary Fig. S1B). These data

Figure 4.GALNT1 is required for maintenance of stemness of BCSCs. A, GALNT1 was silenced in BCMab1þCD44þ cells derived from one human bladder cancer specimen. Ascrambled sequence (shCtrl) was used as a control. B and C, GALNT1 silencing or BG treatment declines oncosphere formation (B) and anchorage-independenttumor growth (C) of BCMab1þCD44þ cells. Oncospheres or colonies were counted in five separate fields after two weeks. �� , P < 0.01. D, mRNA levels ofGALNT1 and Gli1 were detected by RT-PCR from 72 patient samples. P ¼ 0.001; r ¼ 0.945, Pearson correlation. E, real-time PCR analysis of Gli1 mRNA levels inGALNT1-silenced or shCtrl BCMab1þCD44þ cells. BG-treated BCMab1þCD44þ cells were also analyzed. Data were normalized to b-actin and are shown as means�SD. n ¼ 12; �� , P < 0.01. F, immunofluorescence staining of GALNT1 and Gli1 in shCtrl or GALNT1-silenced BCMab1þCD44þ cells. GALNT1 was labeled with FITC,and Gli1 was labeled with PE. G, ELISA analysis of SHH-N levels shCtrl or GALNT1-depleted BCMab1þCD44þ cells. BG-treated cells were also analyzed;�� , P < 0.01. H, tumor volumes were measured at the indicated time points and calculated as means� SD (left). Representative tumors at week 8 are shown on theright; �� , P < 0.01. n ¼ 12. I, xenograft BCSCs sorted by FACS and infected with retrovirus expressing either scramble vector (shCtrl) or shGALNT1, followedby transplantation into mice 24 hours after infection (n ¼ 10 for each group).

Li et al.





validate that GALNT1-mediated SHH processing can induce Gli1expression. Finally, we examined expression levels of SHH and itsreceptor PTCH1 in BCSCs from bladder cancer samples. Wenoticed that PTCH1 was expressed in the BCMab1þCD44þ cells(Supplementary Fig. S1C), and SHH was processed in theseBCSCs. These results suggest that SHH processing in the BCSCsis directly related to its own Hh signaling in the same type of cell.Thus, we conclude that SHH is an autocrine factor in theBCMab1þCD44þ cells. Altogether, these data suggest thatGALNT1 contributes to the activation of Hh signaling and tumor-igenesis of bladder cancer.

GALNT1-mediated glycosylation is required for SHH activationin BCSCs

To further test whether GALNT1 influences SHH activation, wedetected the SHH processing status in GALNT1-silenced or BG-treated BCMab1þCD44þ cells. Interestingly, we found thatGALNT1 knockdown significantly impaired SHH hydrolysis inBCMab1þCD44þ cells derived from bladder cancer specimens(Fig. 5A), whereas shCtrl-treated BCMab1þCD44þ cells exhibitedapparent SHH processing. Moreover, the full-length SHH dis-played a smaller size on the gel blot in GALNT1-depleted

BCMab1þCD44þ cells, suggesting that SHH might not be glyco-sylated after depletion of GALNT1. BG treatment showed similarresults (Fig. 5A). Peanut agglutinin (PNA) lectin can specificallybind to the terminal galactose residues of O-glycans (23). Toexamine whether SHH undergoes O-linked glycosylation, weused PNA to probe SHH immunoprecipitates derived fromGALNT1-silenced or shCtrl-treated BCMab1þCD44þ cells. Inter-estingly,we found that PNAcouldbind to SHHand its hydrolyzedSHH-C fragment (Fig. 5A), suggesting the SHH underwent O-linked glycosylation in BCMab1þCD44þ cells. To further confirmO-linked glycans, we used peptide N-glycosidase F and O-glyco-sidase to treat BCMab1þCD44þ cell lysates, respectively. PNA stillbound to the SHH from N-glycosidase–treated cell extracts, butnot the SHH from O-glycosidase–treated cell extracts (data notshown). These results verify that the O-linked glycosylation takesplace in SHH of BCSCs. We observed that immunofluorescencestaining appeared in WT or shCtrl-treated BCMab1þCD44þ cells,but not in GALNT1-silenced or BG-treated BCMab1þCD44þ cells(Fig. 5B). These data indicate that GALNT1 participates in theprocessing activation of SHH.

We next predictedO-linked glycosylation sites of SHHby usingthe NetOGlyc 3.1 Server tool. Three potential candidate O-linked

Figure 5.GALNT1-mediated glycosylation ofSHH is needed for SHH activation inBCSCs. A, GALNT1-silenced or BG-treated BCMab1þCD44þ cell lysateswere probed with anti-C-terminalSHH, anti-GALNT1, and anti-Gli1antibodies. IP, immunoprecipitation;IB, immunoblotting. B,immunofluorescence staining withC- or N-terminal SHH antibody inGALNT1-silenced or BG-treatedBCMab1þCD44þ cells. The C-terminalof SHH antibody was labeled withFITC, and the N-terminal of SHHantibody was labeled with PE. C, SHHand its three mutations (T282A,S288A, S289A) were constructed intovectors and cotransfected withGALNT1 into BCMab1�CD44� cells for48 hours, respectively. D, WT SHH orT282A-SHH mutant was incubatedwithGALNT1 plus BCMab1�CD44� celllysates. Reaction products wereanalyzed by immunoblotting. E,biotinylated peptides were incubatedwith GALNT1 and UDP-GalNAcfollowed by detection of absorbance(450 nmol/L). F and G, cotransfectionof SHH and GALNT1 can rescueformation of BCMab1þCD44þ cellswith self-renewing ability. BCMab1�

CD44� cells were cotransfected withthe indicated vectors for 48 hoursfollowed by FACS analysis (F) and asphere formation assay (G).Oncospheres were counted in fiveseparate fields after two weeks andare shown asmeans� SD; �� , P < 0.01.H, tumors with the indicatedtreatment were subcutaneouslyinjected into the backs of NOD/SCIDmice (n ¼ 12).






glycosylation sites of SHHprotein, including Thr282, Ser288, andSer289, were selected (Supplementary Fig. S2). To identify theexact glycosylation sites, we generated three mutation constructvectors of SHH (T282A-SHH, S288A-SHH, and S289A-SHH). Wecotransfected GALNT1 and SHH (WT) or its three respectivemutants into BCMab1�CD44� cells. We noticed that onlyT282A-SHH mutation abolished SHH hydrolysis, but WT SHHand other two mutants (S288A-SHH and S289A-SHH) were stillprocessing to generate its hydrolyzed forms (Fig. 5C). Further-more, T282A-SHH mutant showed a smaller size of SHH on theblot, suggesting T282A-SHH mutation could not undergo glyco-sylation. In addition, T282A-SHHmutant could not bind to PNA(Fig. 5C), whereas WT SHH and other two mutants (S288A-SHHand S289A-SHH)were able to bind to PNA, indicating that theO-linked glycosylation did not take place in T282A-SHH mutation.Overall, Thr282 was defined as the glycosylation site for the O-linked glycosylation of SHH.

To further verify that the SHH hydrolysis is mediated byGALNT1 glycosylation, we generated recombinant GALNT1, WTSHH, and T282A-SHH mutant proteins for an in vitro glycosyla-tion assay. As expected, the T282A-SHH mutant incubated withGALNT1 was not processed, and displayed a smaller size (Fig.5D). In contrast, WT SHH incubated with GALNT1 showedapparent hydrolysis and displayed a bigger size. Moreover, WTSHH incubated with GALNT1 could bind to PNA, but not for theT282A-SHH mutant incubated with GALNT1 (data not shown).SHH processing in BCMab1þCD44þ cells served as a positivecontrol. To further validate the glycosylation of SHH and its threemutants, we constructed biotinylated peptides for SHH and itsmutants to analyze GALNT1-mediated SHH glycosylation (Sup-plementary Fig. S3). We found that only the T282A-SHH–mutantpeptide could not be glycosylated by GALNT1 (Fig. 5E), whileWTSHH and the other two mutants (S288A-SHH and S289A-SHH)indeed exhibited glycosylation. We conclude that the GALNT1-mediated O-linked glycosylation of SHH is necessary for SHHprocessing.

We next examined rescue function of GALNT1 and SHH incotransfected BCMab1�CD44� cells from bladder cancer speci-

mens. Notably, GALNT1 and SHH cotransfection could induceformation of a subpopulation of BCMab1þCD44þ cells (Fig. 5F).However, GALNT1, SHH or T282A-SHH alone, or GALNT1 plusT282A-SHH–mutant transfection failed to generate this subset. Inaddition, GALNT1 and SHH cotransfection were able to generateoncospheres efficiently (Fig. 5G). GALNT1 and SHH cotransfec-tion formed xenograft tumors rapidly after inoculation (Fig. 5H).In contrast, GALNT1, SHHor T282A-SHH alone, or GALNT1 plusT282A-SHH–mutant transfection had no such capacity. Takentogether, GALNT1-mediated activation of SHH is indispensablefor the maintenance of self-renewal of BCSCs.

GALNT1 deficiency inhibits bladder tumor growthWe induced bladder cancer in GALNT1-deficient mice or lit-

termate control mice (WT) by feeding BBN. At 4 months of BBNexposure, 23 WT out of 24 mice displayed robust tumors inbladders, with bigger tumor sizes (Fig. 6A and B). In contrast,only 5 GALNT1-deficient mice of 24 GALNT1�/� mice showedbladder tumors with smaller sizes. Tumors in WT or GALNT1-deficient mice were confirmed by histology staining (Fig. 6C). Weprepared single tumor cells fromWT orGALNT1�/�mice for an invitro sphere formation assay.We observed thatGALNT1�/� tumorcells formed much fewer oncospheres than WT tumor cells (Fig.6D). As expected, mRNA and protein levels of Gli1 were unde-tectable in GALNT1�/� tumors (Fig. 6E and F), whereas Gli1 washighly expressed inWT tumors. In sum, GALNT1 deletion inmicesuppresses BBN-induced bladder tumors.

GALNT1 silencing or blockade of Hh signaling displaysantitumor efficacy of bladder cancer in vivo

The implanted mice were grouped (24 mice/group) and intra-vesically administrated GALNT1 siRNA/liposomes, or GALNT1siRNA/liposomes plus cyclopamine the next day. Intriguingly,both GALNT1 siRNA and cyclopamine could significantly sup-press tumor growth (Fig. 7A and B). Importantly, GALNT1 siRNAplus cyclopamine were able to abrogate tumor formation. Todetect the delivery efficiency of siRNAwith liposome, we analyzed

Figure 6.GALNT1 deficiency inhibits BBN-induced bladder tumor growth. A and B, GALNT1�/� mice show a dramatic decrease in tumor formation rates (A) and sizes (B) ofBBN-induced bladder cancer. Tumor formation rates of BBN-induced bladder cancer in mice were calculated after four months with BBN exposure. Then,the mice were sacrificed for measurement of tumor volumes. (n ¼ 24); �� , P < 0.01. C, hematoxylin and eosin counterstaining for GALNT1þ/þ and GALNT1�/� micetreated with BBN. � , bladder tumor. D, tumors induced from GALNT1�/� mice suppress oncosphere formation. Sphere formation assays wereperformed for tumor cells induced from GALNT1þ/þ and GALNT1�/� mice. E and F, GALNT1 and Gli1 expression levels in bladder tumors from GALNT1þ/þ orGALNT1�/� mice analyzed by PCR (E) and immunoblotting (F).

Li et al.





GALNT1 expression in bladder cancer tissues after 5 weeks ofintravesical instillation of GALNT1 siRNA. GALNT1 siRNA wasmixedwith liposome for intravesical administration as previouslydescribed (24). GALNT1 was almost undetectable in GALNT1siRNA/liposome-treated bladder cancer tissues (Fig. 7C), whereasGALNT1was abundantly expressed in siCtrl-treated control mice.These data suggest that intravesical instillation ofGALNT1 siRNA/liposome could efficiently deliver into bladder tissues and suc-cessfully deplete GALNT1 expression. Notably, both GALNT1siRNA and cyclopamine treatment were able to prolong survivalrates of treatedmice compared with siCtrl-treatedmice. However,over 60% mice treated with GALNT1 siRNA plus cyclopaminelived as long as WT mice (Fig. 7D). Therefore, the therapeuticefficacy is due to the elimination of BCSCs.

We next examined whether GALNT1 affects metastasis of blad-der cancer. BCMab1þCD44þ cells injected via tail veins inducedsevere lung metastasis after 5 weeks, whereas BCMab1�CD44�

cellswere unable to formapparent tumors in lungs or other tissues(Fig. 7E). We established GALNT1-silenced BCMab1þCD44þ celllines sorted from human bladder cancer samples and injectedthem through tail veins. We observed that GALNT1-silencedBCMab1þCD44þ cells significantly inhibited tumor metastasis(Fig. 7E), while shCtrl-treated cells still showed robust lungmetastasis.

To further assess GALNT1 function in BBN-induced tumori-genesis of bladder cancer, we induced mouse bladder cancer byfeeding BBN and intravesically applied GALNT1 siRNA everyweek. Then, we used ultrasound to observe tumor volumes everytwo weeks. At 12 weeks, siCtrl treated mice displayed robustbladder tumors in most mice (Fig. 7F). In contrast, GALNT1siRNA administration effectively suppressed bladder tumorgrowth. In addition, GALNT1 siRNA treatment significantly

declined tumor formation rates compared with siCtrl treatment(Fig. 7G). Altogether, GALNT1 is implicated in the tumorigenesisof bladder cancer.

DiscussionBCSCs were considered as TICs (25), which could be

responsible for bladder cancer initiation and recurrence. Inthis study, we identified a BCMab1þCD44þ cell population asBCSCs. The BCMab1 antibody, generating against CD44þ cellsfrom human bladder cancer specimens, was defined to rec-ognize aberrantly glycosylated integrin a3b1 in our previousstudy (20), which is a novel biomarker for BCSCs. We dem-onstrate that BCMab1þCD44þ cells have increased tumorige-nicity and self-renewal property in vivo and in vitro. We definedthat the GALNT1-mediated Hh signaling activation is essentialfor the maintenance of self-renewal of BCSCs and bladdertumorigenesis.

The mammalian secreted Hh ligands include SHH, IndianHedgehog, and Desert Hedgehog, which initiate a signalingcascade via engaging with their common receptor Patched1(PTCH1; ref. 17). Upon Hh engagement, PTCH1 releases itsinhibitory role in the positive regulatory transmembrane proteinSMO, resulting in activation of Gli family transcription factors(Gli1, Gli2, and Gli3; refs. 26, 27). The Hh pathway is largelyinactive in the adult, except for its action in tissue repair andmaintenance (14, 28). The role of Hh signaling in the tumori-genesis of bladder cancer remains controversial. Inconsistently,loss-of-function mutations of PCTH1 were rarely identified inbladder cancer (29, 30). Recently, Robbins and his colleaguesshowed that the Hh signaling is highly expressed in bladdercancer tissues and implicated in the tumorigenesis of bladder

Figure 7.Intravesical administration of GALNT1siRNA and cyclopamine eradicatesbladder tumor in vivo. A, intravesicalinstillation of GALNT1 siRNA orcyclopamine significantly suppressesbladder tumor growth.Bioluminescence images wereobtained by IVIS at 8 weeks afterBCMab1þCD44þ cells transplantationwith intravesical instillation of theindicated agents. B, photon counts ofeach mouse are indicated by pseudo-color scales. Growth curves oforthotopical tumorsweremeasured byIVIS (n ¼ 24 per group); �� , P < 0.01.C, GALNT1 expression levels wereanalyzed by immunoblotting. D,comparison of survival rates with theindicated treatment. n¼ 24 per group.E, representative bioluminescenceimages showing bladder cancermetastasis. Mice were implanted withthe indicated human bladder cancercells via tail vein injection for 5 weeks.F, imaging by stereoscope (middle)and hematoxylin and eosin (H&E)counterstaining (bottom) at 12 weeksafter intravesical administration ofGALNT1 siRNA with BBN exposure(top). G, tumor formation rates ofGALNT1 siRNA or siCtrl-treated micewith exposure of BBN.






cancer (31, 32). However, the role of Hh signaling pathway inBCSCs is largely unknown.

Newly synthesized Hh proteins require a series of posttransla-tional processing reactions to generate their respective activeforms for signal transduction. Many properties of Hh autoproces-singwere defined from studies of theDrosophila protein (33). It iswidely considered that the Hh mature processing procedure ishighly conserved in various species. It has been revealed that thesecreted Hh signaling domain is covalently modified by choles-terol and palmitate (26). The N-terminal cysteine is catalyzed bythe acyl-transferase Skinny hedgehog to form the amide linkage ofpalmitate (34, 35). Hydrophobic modifications confer mem-brane affinity such that the signaling domain is successfullyassociated with its membrane receptors, leading to a signalingcascade (36, 37). Here we defined a novel glycosylation modifi-cation for SHH in BCSCs, which is uniquely mediated byGALNT1.

Superficial tumors are usually treated by surgical resection andintravesical chemotherapy and/or immunotherapy. The difficultyinmanaging bladder cancer is unable to predictwhich tumorswillrecur or progress (38). A previous report showed that intravesicalinstillation of polo-like kinase (PLK-1) siRNA can effectivelyprevent bladder cancer growth (24). siRNA-mediated RNAi issequence-specific gene silencing and widely used to silence geneexpression (39, 40). Here we showed that intravesical adminis-tration of GALNT1 siRNA can significantly suppress bladdertumor growth in orthotopic bladder cancer mouse models andBBN-induced bladder tumors. The SMO antagonist cyclopamineand its derivatives havebeen reported tobe effective agents againstseveral tumors (18, 41). One of cyclopamine derivative GDC-449has been approved to treat advanced unresectable basal cellcarcinoma as a first-line therapy. Of note, intravesical adminis-tration of GALNT1 siRNA and cyclopamine can eradicateimplanted tumor formationwithout overt adverse effects. Finally,

therapeutic use of GALNT1 siRNA, the SMO inhibitor, and theBCMab1 antibody will provide a potent antitumor strategyagainst bladder cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: C. Li, J. Wang, Z. FanDevelopment of methodology: C. Li, Z. YangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): C. Li, Y. Du, Z. Yang, L. He, Y. Wang, R. Yan, J. Wang,Z. FanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): C. Li, Z. Yang, L. He, Y. Wang, R. Yan, Z. FanWriting, review, and/or revision of the manuscript: C. Li, Z. Yang, Z. FanAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): C. Li, Y. Du, L. Hao, Z. FanStudy supervision: C. Li, M. Ding, J. Wang, Z. Fan

AcknowledgmentsThe authors thank Drs. Junfeng Hao and Qiang Hou for their technical

support.

Grant SupportThis work was supported by the National Natural Science Foundation of

China (81330047, 81472413); the State Projects of Essential Drug Research andDevelopment (2012ZX09103301-041); 973 Program of the MOST of China(2010CB911902), and the Strategic Priority Research Programs of the ChineseAcademy of Sciences (XDA01010407).

The costs of publication of this articlewere defrayed inpart 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 August 24, 2015; revised November 11, 2015; accepted November30, 2015; published OnlineFirst December 16, 2015.

References1. Dinney CP, McConkey DJ, Millikan RE, Wu X, Bar-Eli M, Adam L, et al.

Focus on bladder cancer. Cancer Cell 2004;6:111–6.2. Wu XR. Urothelial tumorigenesis: a tale of divergent pathways. Nat Rev

Cancer 2005;5:713–25.3. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA: A Cancer J Clin

2010;60:277–300.4. von der Maase H, Sengelov L, Roberts JT, Ricci S, Dogliotti L, Oliver T, et al.

Long-term survival results of a randomized trial comparing gemcitabineplus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatinin patients with bladder cancer. J Clin Oncol 2005;23:4602–08.

5. Magee JA, Piskounova E, Morrison SJ. Cancer stem cells: impact, hetero-geneity, and uncertainty. Cancer Cell 2012;21:283–96.

6. Rosen JM, Jordan CT. The increasing complexity of the cancer stem cellparadigm. Science 2009;324:1670–1673.

7. Visvader JE, Lindeman GJ. Cancer stem cells: current status and evolvingcomplexities. Cell Stem Cell 2012;10:717–28.

8. Reya T,Morrison SJ, ClarkeMF,Weissman IL. Stem cells, cancer, and cancerstem cells. Nature 2001;414:105–111.

9. Polyak K, Hahn WC. Roots and stems: stem cells in cancer. Nat Med2006;12:296–300.

10. Chan KS, Espinosa I, Chao M, Wong D, Ailles L, Diehn M, et al. Identi-fication, molecular characterization, clinical prognosis, and therapeutictargeting of human bladder tumor-initiating cells. Proc Natl Acad Sci U S A2009;106:14016–021.

11. He X,Marchionni L, Hansel DE, YuW, Sood A, Yang J, et al. Differentiationof a highly tumorigenic basal cell compartment in urothelial carcinoma.Stem Cells 2009;27:1487–95.

12. Shin K, Lee J, Guo N, Kim J, Lim A, Qu L, et al. Hedgehog/Wnt feedbacksupports regenerative proliferation of epithelial stem cells in bladder.Nature 2011;472:110–14.

13. Shin K, Lim A, Odegaard JI, Honeycutt JD, Kawano S, Hsieh MH, et al.Cellular origin of bladder neoplasia and tissue dynamics of its progressionto invasive carcinoma. Nat Cell Biol 2014;16:469–78.

14. Amakye D, Jagani Z, DorschM. Unraveling the therapeutic potential of theHedgehog pathway in cancer. Nat Med 2013;19:1410–22.

15. Ng JM, Curran T. The Hedgehog's tale: developing strategies for targetingcancer. Nat Rev Cancer 2011;11:493–501.

16. Merchant AA, Matsui W. Targeting Hedgehog–a cancer stem cell pathway.Clin Cancer Res 2010;16:3130–40.

17. Teglund S, Toftgard R. Hedgehog beyond medulloblastoma and basal cellcarcinoma. Biochim Biophys Acta 2010;1805:181–208.

18. Kim J, Aftab BT, Tang JY, Kim D, Lee AH, Rezaee M, et al. Itraconazole andarsenic trioxide inhibit Hedgehog pathway activation and tumor growthassociated with acquired resistance to smoothened antagonists. CancerCell 2013;23:23–34.

19. Siegfried Z, Eden S, Mendelsohn M, Feng X, Tsuberi BZ, Cedar H.DNA methylation represses transcription in vivo. Nat Genet 1999;22:203–206.

20. Li C, Yang Z, Du Y, Tang H, Chen J, Hu D, et al. BCMab1, a monoclonalantibody against aberrantly glycosylated integrin alpha3beta1, has potentantitumor activity of bladder cancer in vivo. Clin Cancer Res2014;20:4001–13.

21. Chong L, Ruping Y, Jiancheng B, Guohong Y, Yougang F, JiansongW, et al.Characterization of a novel transplantable orthotopic murine xenograft


Li et al.




model of a human bladder transitional cell tumor (BIU-87). Cancer Biol.Ther 2006;5:394–98.

22. Ulloa F, Real FX. Benzyl-N-acetyl-alpha-D-galactosaminide induces astorage disease-like phenotype by perturbing the endocytic pathway.J Biol Chem 2003;278:12374–83.

23. Foulquier F, Vasile E, Schollen E, Callewaert N, Raemaekers T,Quelhas D, et al. Conserved oligomeric Golgi complex subunit 1deficiency reveals a previously uncharacterized congenital disorderof glycosylation type II. Proc Natl Acad Sci U S A 2006;103:3764–69.

24. Nogawa M, Yuasa T, Kimura S, Tanaka M, Kuroda J, Sato K, et al.Intravesical administration of small interfering RNA targeting PLK-1successfully prevents the growth of bladder cancer. J Clin Invest2005;115:978–85.

25. Ho PL, Kurtova A, Chan KS. Normal and neoplastic urothelialstem cells: getting to the root of the problem. Nat Rev Urol2012;9:583–94.

26. RobbinsDJ, FeiDL, RioboNA. TheHedgehog signal transduction network.Sci Signal 2012;5:re6.

27. Varjosalo M, Taipale J. Hedgehog: functions and mechanisms. Genes Dev2008;22:2454–72.

28. van denBrinkGR.Hedgehog signaling in development and homeostasis ofthe gastrointestinal tract. Physiol Rev 2007;87:1343–75.

29. Aboulkassim TO, LaRue H, Lemieux P, Rousseau F, Fradet Y. Alteration ofthe PATCHED locus in superficial bladder cancer. Oncogene 2003;22:2967–71.

30. Xie J, Johnson RL, Zhang X, Bare JW, Waldman FM, Cogen PH, et al.Mutations of the PATCHED gene in several types of sporadic extracuta-neous tumors. Cancer Res 1997;57:2369–72.

31. Fei DL, Li H, Kozul CD, Black KE, Singh S, Gosse JA, et al. Activation ofHedgehog signaling by the environmental toxicant arsenic may con-tribute to the etiology of arsenic-induced tumors. Cancer Res 2010;70:1981–88.

32. Fei DL, Sanchez-Mejias A, Wang Z, Flaveny C, Long J, Singh S, et al.Hedgehog signaling regulates bladder cancer growth and tumorigenicity.Cancer Res 2012;72:4449–58.

33. Mann RK, Beachy PA. Novel lipidmodifications of secreted protein signals.Annu Rev Biochem 2004;73:891–923.

34. Chamoun Z, Mann RK, Nellen D, von Kessler DP, Bellotto M, Beachy PA,et al. Skinny hedgehog, an acyltransferase required for palmitoylation andactivity of the hedgehog signal. Science 2001;293:2080–84.

35. Dawson PE, Muir TW, Clark-Lewis I, Kent SB. Synthesis of proteins bynative chemical ligation. Science 1994;266:776–79.

36. Feng J, White B, Tyurina OV, Guner B, Larson T, Lee HY, et al. Synergisticand antagonistic roles of the Sonic hedgehog N- and C-terminal lipids.Development 2004;131:4357–70.

37. Grover VK, Valadez JG, Bowman AB, Cooper MK. Lipid modifications ofSonic hedgehog ligand dictate cellular reception and signal response. PLoSOne 2011;6:e21353.

38. Kaufman DS, Shipley WU, Feldman AS. Bladder cancer. Lancet 2009;374:239–49.

39. Dorsett Y, Tuschl T. siRNAs: applications in functional genomics andpotential as therapeutics. Nat Rev Drug Discov 2004;3:318–29.

40. Fellmann C, Lowe SW. Stable RNA interference rules for silencing. Nat CellBiol 2014;16:10–18.

41. Tremblay MR, Lescarbeau A, Grogan MJ, Tan E, Lin G, Austad BC, et al.Discovery of a potent and orally active hedgehog pathway antagonist(IPI-926). J Med Chem 2009;52:4400–18.






2016;76:1273-1283. Published OnlineFirst December 16, 2015.Cancer Res Chong Li, Ying Du, Zhao Yang, et al. of Bladder Cancer Stem Cells

CapacitySignaling Maintains the Self-Renewal and Tumor-Initiating GALNT1-Mediated Glycosylation and Activation of Sonic Hedgehog

Updated version

10.1158/0008-5472.CAN-15-2309doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2015/12/15/0008-5472.CAN-15-2309.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/76/5/1273.full#ref-list-1

This article cites 41 articles, 14 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/76/5/1273.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/76/5/1273To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-15-2309

http://cancerres.aacrjournals.org/content/suppl/2015/12/15/0008-5472.CAN-15-2309.DC1

http://cancerres.aacrjournals.org/content/76/5/1273.full#ref-list-1

http://cancerres.aacrjournals.org/content/76/5/1273.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/76/5/1273


galnt1-mediatedglycosylationandactivationof sonic hedgehog ... · shh (santa cruz biotechnology,...

Documents