MYH9 binds to lncRNA gene PTCSC2 and regulatesFOXE1 in the 9q22 thyroid cancer risk locusYanqiang Wanga,b, Huiling Hea,b, Wei Lia,b, John Phayc, Rulong Shend, Lianbo Yue,f, Baris Hanciogluf,and Albert de la Chapellea,b,1

aHuman Cancer Genetics Program, The Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, OH 43210; bDepartmentof Cancer Biology and Genetics, The Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, OH 43210; cDepartment ofSurgery, The Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, OH 43210; dDepartment of Pathology, The OhioState University Comprehensive Cancer Center, The Ohio State University, Columbus, OH 43210; eCenter for Biostatistics, The Ohio State University,Columbus, OH 43210; and fDepartment of Biomedical Informatics, The Ohio State University, Columbus, OH 43210

Contributed by Albert de la Chapelle, December 7, 2016 (sent for review August 30, 2016; reviewed by Bryan Haugen and Shioko Kimura)

A locus on chromosome 9q22 harbors a SNP (rs965513) firmlyassociated with risk of papillary thyroid carcinoma (PTC). The locusalso comprises the forkhead box E1 (FOXE1) gene, which is impli-cated in thyroid development, and a long noncoding RNA (lncRNA)gene, papillary thyroid cancer susceptibility candidate 2 (PTCSC2).How these might interact is not known. Here we report that PTCSC2binds myosin-9 (MYH9). In a bidirectional promoter shared by FOXE1and PTCSC2, MYH9 inhibits the promoter activity in both directions.This inhibition can be reversed by PTCSC2, which acts as a suppres-sor. RNA knockdown of FOXE1 in primary thyroid cells profoundlyinterferes with the p53 pathway. We propose that the interactionbetween the lncRNA, its binding protein MYH9, and the coding geneFOXE1 underlies the predisposition to PTC triggered by rs965513.

lncRNA | MYH9 | transcriptional regulation | bidirectional promoter |thyroid cancer

Thyroid cancer is the most common endocrine malignancy.Based on the data from the National Cancer Institute for

2016, in total 64,300 new patients will be diagnosed with thyroidcancer, accounting for 3.8% of all cancer patients in the UnitedStates (https://seer.cancer.gov/statfacts/html/thyro.html). Thyroidcancer is classified into four main types: papillary, follicular,medullary, and anaplastic. Papillary thyroid carcinoma (PTC) ac-counts for 85–90% of all thyroid cancers (1). In efforts to explainthe genomic background to PTC, genome-wide association studies(GWASs) have disclosed an SNP marker, rs965513, in 9q22.33strongly associated with PTC in European populations [odds ratio(OR) = 1.75; P = 1.7 × 10−27] (2, 3). The association has beenobserved in other populations as well (4–8). Additionally, rs965513has been reported to be associated with levels of thyroid-relatedhormones, hypothyroidism, goiter, and other abnormal thyroidfunctions (2, 3, 9). SNP rs965513 resides in an intron of a recentlydetected long noncoding RNA (lncRNA), papillary thyroid cancersusceptibility candidate 2 (PTCSC2), which is located between twocoding genes, DNA damage recognition and repair factor (XPA)and forkhead box E1 (FOXE1) (Fig. S1).FOXE1, also known as thyroid transcription factor 2, is a single

exon coding gene belonging to the forkhead/winged helix-domainprotein family (10). FOXE1 knockout mice were born alive butexhibited an ectopic or completely absent thyroid gland and severecleft palate, accompanied by up-regulated TSH and down-regulatedfree thyroxine hormone levels (11). FOXE1 is essential for thyroidgland development and the maintenance of thyroid differentiatedstatus (12, 13). It also plays a role in the predisposition and devel-opment of thyroid cancer (14, 15). Moreover, somatic mutationalinactivation of FOXE1 has been found in PTC, suggesting FOXE1may contribute to tumorigenesis in a subset of thyroid cancers (16).PTCSC2 was discovered in the region harboring rs965513 on

9q22 (15). Similar to some 25% of all known lncRNA genes (17),PTCSC2 has one unspliced isoform and several spliced isoforms,all of which display thyroid-specific expression. In PTC tumors,the risk allele [A] of rs965513 is significantly associated with low

expression of both PTCSC2 and FOXE1 (15). Three enhancerelements are located in a 33-kb linkage disequilibrium (LD)block within PTCSC2. Previous studies showed that PTCSC2 istranscribed in the opposite direction of FOXE1. Exon 1 and intron 1of PTCSC2 isoform C overlap with the FOXE1 promoter region(15). Although the promoter region shared by PTCSC2 and FOXE1was found to be regulated via long-range looping interactions (18),the detailed mechanisms and their bearing on thyroid cancer needto be explored.In the present study, we identified myosin-9 (MYH9) as an

lncRNA binding protein that targets the FOXE1 promoter re-gion through interactions with PTCSC2 and performs its regu-latory function in thyroid cancer via downstream pathways.These findings provide a better defined description of the com-plex mechanisms involved in the 9q22 thyroid cancer locus.

ResultsIdentification of MYH9 as an LncRNA Binding Protein. To begin tounravel the underlying mechanisms by identifying proteins po-tentially interacting with PTCSC2, we performed biotin-labeledRNA pull-down assays with protein lysate from normal thyroidtissue. Two PTCSC2 isoforms—isoform C (1,947 nt) and isoformD (1,804 nt)—were used as baits by generating both sense andantisense (negative control) RNA probes for each of them (Fig. S1).As a result, a strand-specific binding protein was discovered between

Significance

Papillary thyroid carcinoma (PTC) is the most common endocrinecancer and displays strong heritability. So far, the most significantknown predisposing variant is rs965513 in 9q22. Although a longnoncoding RNA, papillary thyroid cancer susceptibility candidate 2(PTCSC2), has been characterized in this locus, its mode of action inthe carcinogenetic process is unknown. Here, we identify myosin-9 (MYH9) as a binding protein of PTCSC2 that regulates the bi-directional promoter shared by PTCSC2 and forkhead box E1(FOXE1). PTCSC2 can rescue the promoter inhibition caused byMYH9. The p53 pathway is profoundly affected by the inhibitionof FOXE1. Our study discovers fundamental roles for PTCSC2,MYH9, and FOXE1 in thyroid cancer and provides a description ofthe regulatory mechanism.

Author contributions: Y.W., H.H., and A.d.l.C. designed research; Y.W. performed research;W.L., J.P., and R.S. contributed new reagents/analytic tools; Y.W., L.Y., and B.H. analyzeddata; and Y.W., H.H., B.H., and A.d.l.C. wrote the paper.

Reviewers: B.H., University of Colorado School of Medicine; and S.K., National CancerInstitute/National Institutes of Health.

The authors declare no conflict of interest.

Data deposition: The data reported in this paper have been deposited in the Gene Ex-pression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no. GSE83919).1To whom correspondence should be addressed. Email: Albert.delaChapelle@osumc.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1619917114/-/DCSupplemental.

474–479 | PNAS | January 17, 2017 | vol. 114 | no. 3 www.pnas.org/cgi/doi/10.1073/pnas.1619917114

Dow

nloa

ded

by g

uest

on

June

3, 2

020

150 KD and 250 KD in size (Fig. 1A and Fig. S2). This proteinwas identified as MYH9, being 226 KD in size by mass spec-trometry (MS) (Fig. 1B). Another protein band that was noticedbetween 37 KD and 50 KD in size was identified as beta-actin(ACTB) (Fig. 1A and Figs. S2 and S3). As MYH9 is an ACTBbinding partner participating in important cellular processes(19), ACTB was not studied further here.The interaction between PTCSC2-isoform C and MYH9 was

confirmed by RNA immunoprecipitation (RIP) assay in BCPAPcells with PTCSC2-isoform C stable overexpression (Fig. 1C).PTCSC2 was readily enriched by MYH9 antibody compared withIgG-negative controls (>sixfold), and this enrichment was sig-nificantly higher than in the 18S RNA controls (P < 0.01),demonstrating the specificity of the assay.

MYH9 Binds to the FOXE1 Promoter Region. To determine whetherthe PTCSC2 binding protein MYH9 is involved in the transcrip-tional expression of the FOXE1 gene, four overlapping regions(R1, R2, R3, and R4) that cover the transcription factor enrichedregion (from ENCODE ChIP sequencing data) in the 5′UTR ofFOXE1 were chosen for ChIP analysis using MYH9 antibody (Fig.2A). Of note, the previously reported thyroid cancer risk-associ-ated SNP rs1867277 (14) is located in the R3 genomic region. Inthe KTC1 cell line, significant enrichment was found in the R1 andR3 regions (Fig. 2B). The enrichment was confirmed in non-tumorous thyroid tissue (Fig. 2C), suggesting the DNA bindingactivity of MYH9 is in the FOXE1 promoter, perhaps more spe-cifically in regions R1 and R3.

MYH9 Functions as a Suppressor in the Promoter Region of PTCSC2and FOXE1 with Bidirectional Activity. To investigate the role ofMYH9 binding to the FOXE1 promoter region, luciferase assayswere performed in cells transiently transfected with FOXE1promoter or empty constructs (without promoter), together withMYH9 expression or empty expression vectors.

As PTCSC2 and FOXE1 are transcribed in opposite directions(15) and there is only 174 bp between their transcription start sites(TSSs), we postulated that they are transcribed by the same pro-moter with bidirectional activity. To test this hypothesis, a shorter(1,607 bp) and a longer (2,520 bp) fragment cloned from theFOXE1 promoter region were used for luciferase reporter assay inthe BCPAP cell line with both forward (Promoter-S and Pro-moter-L) and inverted (Promoter-Inv-S and Promoter-Inv-L)orientation, as indicated (Fig. 3A). Although the inverted promotersshowed lower activity compared with their corresponding forwardones, the fragment with both orientations exhibited a significantlyhigher level of promoter activity in comparison with the no pro-moter control. Moreover, cotransfection with the MYH9 expressionvector showed that MYH9 can significantly inhibit the luciferaseactivity in both the forward and inverted promoter fragments. ThatMYH9 functions as a suppressor was disclosed in both the shorterand longer promoter fragments (Fig. 3 B and C).Thus, MYH9 had an impact on the FOXE1 gene, but its role in

the presence of lncRNA PTCSC2 was not known. We transientlycotransfected the PTCSC2 expression vector in the BCPAP cell linein the presence or absence of the MYH9 expression construct.Because there was no significant difference in promoter activitybetween the two fragments of different sizes, only the longer pro-moter sets were used. Interestingly, an increase in promoter lucif-erase activity was observed with cotransfection of the PTCSC2expression vector, irrespective of MYH9 overexpression. The ef-fect was significant on the forward Promoter-L, whereas it was notsignificant on the inverted Promoter-Inv-L, suggesting a rescuingeffect on the inhibition caused by MYH9 (Fig. 4 A and B).To further validate the roles of MYH9 and PTCSC2 in regu-

lating the endogenous FOXE1 promoter, a transient transfectionassay with MYH9 and PTCSC2 expression vectors was per-formed in BCPAP and TPC1 cells. The expression of the en-dogenous FOXE1 gene showed a significant decrease in thepresence of MYH9 overexpression, which is consistent with therole of MYH9 as an inhibitor in our luciferase assay. This in-hibitory effect becomes nonsignificant in the presence ofPTCSC2, indicating that FOXE1 is regulated by PTCSC2 andMYH9 differentially (Fig. 4 C and D). When we examined thePTCSC2 transcript levels and protein levels of MYH9 andFOXE1 in 6 thyroid cancer cell lines and in nontumorous thyroidtissue, we did not notice a correlation among the expressionlevels of these three genes (Fig. S4).SNP rs1867277 is in partial LD with rs965513 and is reported as

a functional variant in the regulation of FOXE1 expression andconfers thyroid cancer susceptibility (14). As rs1867277 is locatedin the MYH9 interacting region R3 (Fig. 2A), we tested the reg-ulatory effects of MYH9 and PTCSC2 on the FOXE1 promoterwith either the wild-type allele (G) or the thyroid cancer risk allele(A). BCPAP cells were transiently transfected with either MYH9or both MYH9 and PTCSC2 constructs using promoters withopposite orientations. The promoter containing the A alleleexhibited significantly lower activity than the promoter with theG allele, and a similar difference was found in both forward andinverted promoters (Fig. S5).

FOXE1 Is Involved in the P53 Pathway in Human Primary Thyroid Cells.FOXE1 is an important transcriptional regulator in the physio-logical functions of the thyroid as well as in thyroid cancer.Lower expression level of FOXE1 is associated with PTC risk inthe presence of the risk allele of SNP rs965513 (15). To identifygenes involved in FOXE1-mediated signaling pathways andtranscriptional regulation in thyroid, knockdown of FOXE1 wasperformed with siRNAs in nontumorous human primary thyroidcells (20). Nontumorous human primary thyroid cells from thesame patient sample transfected with scrambled siRNAs wereused as the control. Knockdown of FOXE1 resulted in numerouschanges in gene expression determined by RNA deep sequencing

Fig. 1. Identification of MYH9 as a PTCSC2 binding protein. (A) RNA pull-down experiment with nontumorous thyroid tissue extract. Biotin pull-downassays followed by SDS/PAGE separation were used to isolate the bindingprotein of PTCSC2 isoform C. Antisense RNA of PTCSC2 isoform C was used asthe negative control. The arrows indicate the binding protein bands ∼226KD and 42 KD in size, respectively. (B) Information for the identification ofMYH9 by MS. The full length of MYH9 protein is 1,960 amino acids, as shownin the lower diagram. Blue boxes indicate the peptides identified by MSanalysis. (C) qRT-PCR detection of the indicated RNAs retrieved by MYH9antibody (RIP assay) in the BCPAP cell line with stable PTCSC2 isoform Coverexpression. IgG was used as a negative control. The relative fold changewas normalized to the IgG control. **P < 0.01. Student’s t test.

Wang et al. PNAS | January 17, 2017 | vol. 114 | no. 3 | 475

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

June

3, 2

020

(fold change > 1.5, P < 0.001). Of the dysregulated genes (totalN = 107; up-regulated n = 80), 89.7% were coding genes (TableS1). Biological functional analysis using Ingenuity Pathway Anal-ysis (IPA) software showed that cancer was in the top categoryof “Diseases and Disorders” with the second lowest P value (Fig.S6). The top five categories of “Molecular and Cellular Func-tions”—namely, cellular movement, cell death and survival, cel-lular development, cellular growth and proliferation, and cellsignaling—suggested its possible relevance to pathways related tocell apoptosis, migration, or invasion (Table 1). These effects havebeen widely reported as the major effects of the well-known p53signaling pathway (21, 22). Of the top 25 dysregulated genes, thethrombospondin 1 (THBS1, also known as TSP-1) and insulin-likegrowth factor binding protein 3 (IGFBP3) genes were found to beup-regulated in FOXE1 knockdown cells (Fig. 5A and Table S1).This inverse relationship was subsequently confirmed by quanti-tative RT-PCR (qRT-PCR) in primary thyroid cells (Fig. 5B) andalso in the BCPAP and TPC1 cell lines (Fig. S7). We also analyzedgene expression data from 59 pairs of tumor/nontumor PTCsamples available in the TCGA data portal (https://tcga-data.nci.nih.gov/docs/publications/tcga). In this analysis, the expressionof FOXE1 was significantly (P = 2.02 × 10−05) higher in non-tumorous tissue, whereas the expression of both THBS1 andIGFBP3 are significantly higher in tumorous tissue (P = 1.73 × 10−02 and P = 4.19 × 10−05, respectively) (Table S2). This is consistentwith our data and further supports the putative role of FOXE1 as asuppressor of THBS1 and IGFBP3 in thyroid. Both THBS1 andIGFBP3 are important members of the p53 pathway involved inapoptosis, inhibition of the IGF1/mTOR pathway, and inhibitionof angiogenesis and metastasis, suggesting interactions betweenFOXE1 and the p53 pathway in thyroid cells (Fig. 5C).To further assess the function of FOXE1 in thyroid cells, we

performed cell viability and apoptosis assays in FOXE1 knock-down BCPAP cells. The down-regulation of FOXE1 led to re-duced cell viability in CellTiter-Glo assay (Fig. 5D). Consistently,apoptosis assay with flow cytometric analysis indicated that in-creases of apoptotic cells at both early and late apoptosis stageswere found in the FOXE1 knockdown cells (Fig. 5E).

DiscussionLncRNAs are important regulators of tissue physiology and diseaseprocesses including cancer. The regulatory effectiveness of lncRNAs

is dependent on their expression. Many lncRNAs show tissue-spe-cific expression patterns. Several thyroid tissue- and cancer-associ-ated lncRNAs were recently discovered (23). Only a few lncRNAsrelated to thyroid and thyroid cancer have been well characterized,including PTCSC2, PTCSC3, and NAMA (15, 24, 25), but even inthese cases, the relevant molecular mechanisms are not known. It isof special interest that these lncRNAs are involved in the geneticpredisposition to thyroid cancer (germ-line involvement). Therefore,their impact on nontumorous thyroid tissue needs to be explored,which we here have endeavored to accomplish. We identifiedMYH9 as a noncoding RNA binding partner, which had not beenreported before.MYH9 protein is a member of the nonmuscle myosin II

(NMII) group, which belongs to the myosin II subfamily (26). Asa subunit of myosin II heavy chains, MYH9 takes part in thegeneration of cell polarity, cell migration, cell–cell adhesionprocesses, and maintaining cytoskeleton structure by binding toactin filaments (19). Recently, additional functions of MYH9have been reported. For example, MYH9 affects the expressionof PAX5 by interacting with Thy28 (27), and it can activateAKT through RAC1 and PAK1 (28), suggesting a role in generegulation. In 2014, a role for MYH9 as a tumor suppressorgene was discovered in squamous cell carcinomas (SCCs). Thisreport implicated MYH9 in tumor development by regulatingposttranscriptional p53 stabilization, but the underlying mech-anism is still unknown (29). Our results are consistent with thisfinding and provide further evidence for a regulatory model bywhich a specific noncoding RNA, PTCSC2, and MYH9 worktogether by regulating FOXE1 promoter activity. Knockdown ofFOXE1 in primary human thyroid cells revealed that FOXE1 canregulate events in the p53 pathways.As a common occurrence, bidirectional transcription of two

protein coding genes has been discovered and studied duringthe past decades (30, 31). More recently, many noncodingRNAs that are transcribed in the proximity of coding genes anddriven by the same bidirectional promoter have been reported(32, 33). Some noncoding transcripts exert repression of theirnearby coding genes by competing for the same polymerasesand accessory factors or by transcriptional interference or his-tone modification (34–36). A number of highly expressed an-tisense transcripts derived from bidirectional TSSs are able toenhance the expression of the corresponding protein coding gene

Fig. 2. Validation of the MYH9 binding site in theFOXE1 promoter region by ChIP assay. (A) Schematicview of a 900-bp transcription factor enrichmentregion in the FOXE1 promoter region. Genomic co-ordinate, chr9:100,615,431–100,616,330 (GRC37/hg19).DNase I hypersensitive sites, CpG islands, and ENCODEdata were obtained from the UCSC Genome Browser.Red bars indicate the tested regions in ChIP assays.(B) ChIP assay using KTC1 cell line. The results representfour independent experiments, each in duplicates.Results are shown as means ± SD. *P < 0.05. Student’st test. (C) ChIP assay using nontumorous thyroid tissuesample. The relative abundance was normalized tothe input control.

476 | www.pnas.org/cgi/doi/10.1073/pnas.1619917114 Wang et al.

Dow

nloa

ded

by g

uest

on

June

3, 2

020

in a tissue-specific manner (37). The PTCSC2–FOXE1 pair in ourstudy further supports this regulatory model, which can be directedby the same inhibitor. Considering the complexity of the 9q22 re-gion including nearby CpG islands (38), it is reasonable to assumethat other transcription factors and some epigenetic modificationsmight be involved in the regulation of this bidirectional promoter, inaddition to MYH9 and PTCSC2. It will be of interest to learnwhether other regulatory factors have any impact on MYH9, orvice versa.Cell line models were used in most previous reports on

FOXE1 in the thyroid (39, 40). Because several thyroid markersincluding lncRNAs lose their expression in most cell lines (41),

we consider thyroid primary cell culture to be a superior modelfor pathway and network analysis. In this study, we use humanthyroid primary cell cultures that were recently established (20).Two important members of the p53 pathway, THBS1 andIGFBP3, have been reported to be involved in various cancertypes. The regulation of tumor growth and metastasis by THBS1has been widely described (42). THBS1 is regulated by BRAFV600E

in thyroid cancer cells. The silencing of THBS1 results in decreasedcell proliferation, adhesion, migration, and invasion (43). Moreover,THBS1 silencing can also cause changes in the levels of integrinreceptors and anaplastic thyroid cancer cell morphology (44).THBS1 promotes human follicular thyroid carcinoma cell invasionmainly through up-regulation of urokinase plasminogen activator(PLAU) (45). This is consistent with our RNA-seq results, as PLAUis also one of the top 25 dysregulated genes in FOXE1 knockdownprimary thyroid cells (Fig. 5A and Table S1). IGFBP3 inhibits tumorgrowth by promoting apoptosis and inhibiting cell proliferation insome cancer types, but in other circumstances, it increases cellsurvival and stimulates proliferation (46). Our data further em-phasize the roles played by these two genes (THBS1 and IGFBP3)in thyroid cancer.In conclusion, our study addresses the mechanisms and pro-

vides a biological characterization of the widely reported GWASlocus 9q22 in thyroid cancer. We propose MYH9 as a bindingprotein of thyroid-specific noncoding RNA that may underlie thepredisposition to thyroid cancer.

Fig. 4. Effect of PTCSC2 on FOXE1 bidirectional promoter. Dual reporterluciferase assay using the long promoter constructs with forward (A) orinverted (B) orientations cotransfected with PTCSC2, MYH9, or empty ex-pression vectors. The values at the bottom of each column represent theportion of the total expression vectors. Results are shown as means ± SD offour independent experiments, each in four replicates. *P < 0.05. Student’s ttest. (C) qPCR detection of endogenous FOXE1 in the BCPAP cell line and (D)the TPC1 cell line transiently transfected with PTCSC2, MYH9, or empty ex-pression vectors as indicated (n = 3). All values were normalized with thevalues of the corresponding empty vector-transfected groups. Results areshown as means ± SD. *P < 0.05; NS, not significant. Student’s t test.

Fig. 3. Effect of MYH9 on FOXE1 promoter with bidirectional activity. (A) Sche-matic view of the vectors used in luciferase assays. Red boxes indicate the UTRregions of FOXE1; the yellow box indicates the FOXE1 coding sequences; blueboxes indicate the exons of PTCSC2; the thick gray line indicates the intron ofPTCSC2; the thin black line indicates the intergenic genomic region. All of theposition numbers are labeled relative to the FOXE1 TSS as +1. Arrows representthe fragment direction in the constructs according to the position in the ge-nome. Dual reporter luciferase assays in BCPAP cells cotransfected with con-structs containing a short promoter fragment (B) or long promoter fragment (C)having either forward or inverted orientation, with or without MYH9 over-expression vector. Values are presented as fold of the no promoter controltreated with empty expression vector. Results are shown as means ± SD of fourindependent experiments, each in four replicates. *P < 0.05; **P < 0.01; ***P <0.001. Student’s t test.

Wang et al. PNAS | January 17, 2017 | vol. 114 | no. 3 | 477

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

June

3, 2

020

Materials and MethodsThe study was approved by the Institutional Review Board at The Ohio StateUniversity (OSU), and all subjects gave written informed consent beforeparticipation.

Cell Lines and Thyroid Tissue Samples. The human thyroid carcinoma cell linesused in this study were incubated in antibiotic-free RPMImedium 1640 (KTC1,BCPAP, C643, SW1736) orDMEM(TPC1, FTC133) supplementedwith10%(vol/vol)FBS (Gibco) at 37 °C in humidified air with 5% (vol/vol) CO2.Thyroid non-tumorous samples were snap-frozen in liquid nitrogen and kept at –80 °Cafter being obtained from patients with PTC during surgery. All cases werehistologically diagnosed as PTC; clinical information on these samples will bemade available on request.

In Vitro RNA Probe Synthesis and RNA Pull-Down. In vitro RNA synthesisprocedures using MEGAscript T7 Transcription Kit for target probe andMEGAscript SP6 Transcription Kit for antisense control probe (Life Technol-ogies) were performed according to the manufacturers’ instructions. Thewhole-protein lysate from the human nontumorous thyroid sample wasextracted using T-PER Tissue Protein Extraction Reagent (Thermo FisherScientific). Protein concentration was determined by using the Bio-RadProtein Assay Dye Reagent Concentrate (Bio-Rad) kit. RNA pull-down assaywas performed using Pierce Magnetic RNA-Protein Pull-Down Kit (ThermoFisher Scientific) according to the standard instructions. A detailed de-scription can be found in SI Materials and Methods.

RIP Assay. In total, 2 × 106 BCPAP cells with PTCSC2 isoform C overexpressionwere used for RIP lysis preparation using Harsh Lysis Buffer of Imprint RIP Kit

(Sigma-Aldrich) following the instructions. Briefly, 5 μg antibody of rabbitanti-MYH9 (Santa Cruz, sc-98978) or rabbit IgG (Sigma, I5006-1MG) wasused to incubate with the cell lysate in 1 mL RIP wash buffer with ProteaseInhibitor Mixture (Roche) and also RNase Inhibitor overnight at 4 °C afterligation to Maganetic Beads. The RIP products were harvested by washingwith 1 mL washing buffer up to five times. Finally, the washed productresuspended in 200 μL buffer was subjected to the purification step usingTRIzol reagent (Life Technologies). The quantification of target RNA inthe final RIP products was determined by real-time qRT-PCR using theTaqman method.

ChIP Assay.Adetailed description of the chromatin immunoprecipitation (ChIP)assays to determine the degree of MYH9 enrichment on the KTC1 cell line andfrozen thyroid tissue samples can be found in SI Materials and Methods. Theprimer sequences are provided in Table S3.

Fig. 5. Gene expression profile of FOXE1 knock-down in thyroid primary cells indicating its in-volvement in the p53 pathway. (A) Gene expressiondifferences of the top 25 genes caused by FOXE1knockdown in primary thyroid cells on day 5 of cul-ture. The expression is plotted with heat-map colorscale using relative expression fold change (FOXE1knockdown culture vs. control culture) (fold change >1.5, P < 0.001). Exp. 1, Exp. 2, and Exp. 3 representthree independent primary culture experiments de-rived from different patient samples. (B) Confirma-tion by qRT-PCR of two up-regulated genes, IGFBP3(Left) and THBS1 (Right), in FOXE1 knockdown pri-mary cells. si-Control represents the primary cellsample treated with scrambled siRNA control. Resultsare shown as means ± SD of three independent ex-periments, each in three replicates. All values werenormalized with the values of the correspondingnegative control siRNA-treated groups. *P < 0.05;***P < 0.001. Student’s t test. (C) Part of the p53signaling pathway including typical gene membersand their functions from Kyoto Encyclopedia ofGenes and Genomes (KEGG). THBS1 is labeled in apurple ellipse, and IGFBP3 is labeled in two red el-lipses. (D) Cell viability changes of BCPAP cell line atthree time points (24, 48, and 72 h) with FOXE1knockdown. For each time point, experiments wereperformed in four replicates of three biological rep-licates. All values were normalized with the values ofthe corresponding negative control siRNA-treatedgroups. (E) Example plots of cell apoptosis assays inBCPAP cell line treated with FOXE1 siRNA or negativecontrol siRNA. Cells stained with only annexin V wereevaluated as being in early apoptosis stage (lowerright quadrant); cells stained with both annexin Vand PI (propidium iodide) were evaluated as being inlate apoptosis stage (upper right quadrant).

Table 1. Molecular and cellular functions of the dysregulatedgenes in FOXE1 knockdown cells by IPA analysis

Function P value Molecules

Cellular movement 1.30 × 10−03–7.65 × 10−10 42Cell death and survival 1.27 × 10−03–8.74 × 10−10 46Cellular development 1.22 × 10−03–1.20 × 10−08 53Cellular growth and

proliferation1.13 × 10−03–1.20 × 10−08 47

Cell signaling 9.58 × 10−04–6.75 × 10−08 19

478 | www.pnas.org/cgi/doi/10.1073/pnas.1619917114 Wang et al.

Dow

nloa

ded

by g

uest

on

June

3, 2

020

Transfection and Dual Luciferase Reporter Assay. For the luciferase reporterassay, BCPAP cells were transiently transfected with reporter plasmid usingLipofectamine 3000 Reagent (Thermo Fisher Scientific) according to themanufacturer’s instructions. Briefly, cells were seeded in a 24-well plate at 0.7 ×105 cells per well and cultured overnight. The following day, 250 ng lucif-erase reporter plasmids and 250 ng effector plasmids were cotransfectedalong with 1.25 ng Renilla plasmid pRL-TK (Promega) as a control for eachwell. Cells were lysed with 100 μL passive lysis buffer (Promega) for luciferaseactivity analysis using the Dual-Luciferase Reporter Assay System (Promega)24 h after transfection. A 20-μL aliquot of cell lysate was then assayed forluciferase activity using GloMax 96 Microplate Luminometer (Promega).

Cell Viability Assay and Cell Apoptosis Assay. Cell viability in the BCPAP cell linetransfected with FOXE1 siRNA or negative control siRNA at different timepoints (24, 48, and 72 h) was analyzed by using the CellTiter-Glo assay

(Promega). The numbers of viable cells were represented by the luminescentsignals, which were obtained by using GloMax 96 Microplate Luminometer(Promega). Experiments were performed according to the manufacturer’s in-structions. Apoptotic cell death in the BCPAP cell line transfected with FOXE1siRNA or negative control siRNA after 72 h was measured using the FITCAnnexin V Apoptosis Detection Kit (BD Biosciences) by flow cytometryaccording to the manufacturer’s instructions. Three biological replicateswere performed.

ACKNOWLEDGMENTS. We thank Jerneja Tomsic and Sandya Liyanarachchifor helpful discussions about data analysis, Ann-Kathrin Eisfeld for help withknockdown experiments, Jan Lockman and Barbara Fersch for administrativehelp, and the OSU Comprehensive Cancer Center (OSUCCC) ProteomicsShared Resource for help with MS. This work was supported by NationalCancer Institute Grants P30CA16058 and P50CA168505.

1. Davies L, Welch HG (2006) Increasing incidence of thyroid cancer in the United States,1973-2002. JAMA 295(18):2164–2167.

2. Gudmundsson J, et al. (2009) Common variants on 9q22.33 and 14q13.3 predispose tothyroid cancer in European populations. Nat Genet 41(4):460–464.

3. Gudmundsson J, et al. (2012) Discovery of common variants associated with low TSHlevels and thyroid cancer risk. Nat Genet 44(3):319–322.

4. Ai L, et al. (2014) Associations between rs965513/rs944289 and papillary thyroid car-cinoma risk: A meta-analysis. Endocrine 47(2):428–434.

5. Takahashi M, et al. (2010) The FOXE1 locus is a major genetic determinant for radi-ation-related thyroid carcinoma in Chernobyl. Hum Mol Genet 19(12):2516–2523.

6. Matsuse M, et al. (2011) The FOXE1 and NKX2-1 loci are associated with susceptibility topapillary thyroid carcinoma in the Japanese population. J Med Genet 48(9):645–648.

7. Zhu H, Xi Q, Liu L, Wang J, Gu M (2014) Quantitative assessment of common geneticvariants on FOXE1 and differentiated thyroid cancer risk. PLoS One 9(1):e87332.

8. Wang YL, et al. (2013) Confirmation of papillary thyroid cancer susceptibility lociidentified by genome-wide association studies of chromosomes 14q13, 9q22, 2q35and 8p12 in a Chinese population. J Med Genet 50(10):689–695.

9. Denny JC, et al. (2011) Variants near FOXE1 are associated with hypothyroidism andother thyroid conditions: Using electronic medical records for genome- and phenome-wide studies. Am J Hum Genet 89(4):529–542.

10. Zannini M, et al. (1997) TTF-2, a new forkhead protein, shows a temporal expressionin the developing thyroid which is consistent with a role in controlling the onset ofdifferentiation. EMBO J 16(11):3185–3197.

11. De Felice M, et al. (1998) A mouse model for hereditary thyroid dysgenesis and cleftpalate. Nat Genet 19(4):395–398.

12. Fernández LP, López-Márquez A, Santisteban P (2015) Thyroid transcription factors indevelopment, differentiation and disease. Nat Rev Endocrinol 11(1):29–42.

13. De Felice M, Di Lauro R (2011) Minireview: Intrinsic and extrinsic factors in thyroidgland development: An update. Endocrinology 152(8):2948–2956.

14. Landa I, et al. (2009) The variant rs1867277 in FOXE1 gene confers thyroid cancersusceptibility through the recruitment of USF1/USF2 transcription factors. PLoS Genet5(9):e1000637.

15. He H, et al. (2015) Genetic predisposition to papillary thyroid carcinoma: Involvement ofFOXE1, TSHR, and a novel lincRNA gene, PTCSC2. J Clin Endocrinol Metab 100(1):E164–E172.

16. MondM, et al. (2015) Somatic mutations of FOXE1 in papillary thyroid cancer. Thyroid25(8):904–910.

17. Derrien T, et al. (2012) The GENCODE v7 catalog of human long noncoding RNAs: Analysisof their gene structure, evolution, and expression. Genome Res 22(9):1775–1789.

18. He H, et al. (2015) Multiple functional variants in long-range enhancer elementscontribute to the risk of SNP rs965513 in thyroid cancer. Proc Natl Acad Sci USA112(19):6128–6133.

19. Conti MA, Adelstein RS (2008) Nonmuscle myosin II moves in new directions. J Cell Sci121(Pt 1):11–18.

20. Wang Y, et al. (2016) Primary cell culture systems for human thyroid studies. Thyroid26(8):1131–1140.

21. Hwang CI, et al. (2011) Wild-type p53 controls cell motility and invasion by dualregulation of MET expression. Proc Natl Acad Sci USA 108(34):14240–14245.

22. Fridman JS, Lowe SW (2003) Control of apoptosis by p53. Oncogene 22(56):9030–9040.

23. Iyer MK, et al. (2015) The landscape of long noncoding RNAs in the human tran-scriptome. Nat Genet 47(3):199–208.

24. Jendrzejewski J, et al. (2012) The polymorphism rs944289 predisposes to papillarythyroid carcinoma through a large intergenic noncoding RNA gene of tumor sup-pressor type. Proc Natl Acad Sci USA 109(22):8646–8651.

25. Yoon H, et al. (2007) Identification of a novel noncoding RNA gene, NAMA, that isdownregulated in papillary thyroid carcinoma with BRAF mutation and associatedwith growth arrest. Int J Cancer 121(4):767–775.

26. Sellers JR (2000) Myosins: A diverse superfamily. Biochim Biophys Acta 1496(1):3–22.

27. Fujita T, Kitaura F, Fujii H (2015) A critical role of the Thy28-MYH9 axis in B cell-specificexpression of the Pax5 gene in chicken B cells. PLoS One 10(1):e0116579.

28. Zhao B, et al. (2015) The non-muscle-myosin-II heavy chain Myh9 mediates colitis-induced epithelium injury by restricting Lgr5+ stem cells. Nat Commun 6:7166.

29. Schramek D, et al. (2014) Direct in vivo RNAi screen unveils myosin IIa as a tumorsuppressor of squamous cell carcinomas. Science 343(6168):309–313.

30. Beck CF, Warren RA (1988) Divergent promoters, a common form of gene organi-zation. Microbiol Rev 52(3):318–326.

31. Trinklein ND, et al. (2004) An abundance of bidirectional promoters in the humangenome. Genome Res 14(1):62–66.

32. Seila AC, et al. (2008) Divergent transcription from active promoters. Science322(5909):1849–1851.

33. Pickrell JK, et al. (2010) Understanding mechanisms underlying human gene expres-sion variation with RNA sequencing. Nature 464(7289):768–772.

34. Neil H, et al. (2009) Widespread bidirectional promoters are the major source ofcryptic transcripts in yeast. Nature 457(7232):1038–1042.

35. Berretta J, Pinskaya M, Morillon A (2008) A cryptic unstable transcript mediatestranscriptional trans-silencing of the Ty1 retrotransposon in S. cerevisiae. Genes Dev22(5):615–626.

36. Camblong J, Iglesias N, Fickentscher C, Dieppois G, Stutz F (2007) Antisense RNAstabilization induces transcriptional gene silencing via histone deacetylation inS. cerevisiae. Cell 131(4):706–717.

37. Uesaka M, et al. (2014) Bidirectional promoters are the major source of gene acti-vation-associated non-coding RNAs in mammals. BMC Genomics 15:35.

38. Abu-Khudir R, et al. (2014) Role for tissue-dependent methylation differences in theexpression of FOXE1 in nontumoral thyroid glands. J Clin Endocrinol Metab 99(6):E1120–E1129.

39. Santisteban P, Acebrón A, Polycarpou-Schwarz M, Di Lauro R (1992) Insulin and in-sulin-like growth factor I regulate a thyroid-specific nuclear protein that binds to thethyroglobulin promoter. Mol Endocrinol 6(8):1310–1317.

40. Venza I, et al. (2011) MSX1 and TGF-beta3 are novel target genes functionally reg-ulated by FOXE1. Hum Mol Genet 20(5):1016–1025.

41. Pilli T, Prasad KV, Jayarama S, Pacini F, Prabhakar BS (2009) Potential utility andlimitations of thyroid cancer cell lines as models for studying thyroid cancer. Thyroid19(12):1333–1342.

42. Sid B, et al. (2004) Thrombospondin 1: A multifunctional protein implicated in theregulation of tumor growth. Crit Rev Oncol Hematol 49(3):245–258.

43. Nucera C, et al. (2010) B-Raf(V600E) and thrombospondin-1 promote thyroid cancerprogression. Proc Natl Acad Sci USA 107(23):10649–10654.

44. Duquette M, Sadow PM, Lawler J, Nucera C (2013) Thrombospondin-1 silencing down-regulates integrin expression levels in human anaplastic thyroid cancer cells withBRAF(V600E): New insights in the host tissue adaptation and homeostasis of tumormicroenvironment. Front Endocrinol (Lausanne) 4:189.

45. Sid B, et al. (2008) Thrombospondin-1 enhances human thyroid carcinoma cell in-vasion through urokinase activity. Int J Biochem Cell Biol 40(9):1890–1900.

46. Baxter RC (2014) IGF binding proteins in cancer: Mechanistic and clinical insights. NatRev Cancer 14(5):329–341.

47. Dulyaninova NG, House RP, Betapudi V, Bresnick AR (2007) Myosin-IIA heavy-chainphosphorylation regulates the motility of MDA-MB-231 carcinoma cells. Mol Biol Cell18(8):3144–3155.

48. Kim D, et al. (2013) TopHat2: Accurate alignment of transcriptomes in the presence ofinsertions, deletions and gene fusions. Genome Biol 14(4):R36.

49. Liao Y, Smyth GK, Shi W (2014) featureCounts: An efficient general purpose programfor assigning sequence reads to genomic features. Bioinformatics 30(7):923–930.

50. Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dis-persion for RNA-seq data with DESeq2. Genome Biol 15(12):550.

51. Gordon A, Glazko G, Qiu X, Yakovlev A (2007) Control of the mean number of falsediscoveries, Bonferroni and stability of multiple testing. Ann Appl Stat 1(1):179–190.

Wang et al. PNAS | January 17, 2017 | vol. 114 | no. 3 | 479

BIOCH

EMISTR

Y

Dow

nloa

ded

by g

uest

on

June

3, 2

020