bire-87-06-01 1..14

BIOLOGY OF REPRODUCTION (2012) 87(6):131, 1–14

Published online before print 3 October 2012.

DOI 10.1095/biolreprod.112.099663

SRY Induced TCF21 Genome-Wide Targets and Cascade of bHLH Factors During

Sertoli Cell Differentiation and Male Sex Determination in Rats

Ramji K. Bhandari, Ellyn N. Schinke, Md. M. Haque, Ingrid Sadler-Riggleman, and Michael K. Skinner

Center for Reproductive Biology, School of Biological Sciences, Washington State University, Pullman, Washington

ABSTRACT

Male sex determination is initiated through the testis-

determining factor SRY that promotes Sertoli cell differentiation

and subsequent gonadal development. The basic helix-loop-helix

(bHLH) gene Tcf21 was identified as one of the direct

downstream targets of SRY. The current study was designed to

identify the downstream targets of TCF21 and the potential

cascade of bHLH genes that promote Sertoli cell differentiation.

A modified ChIP-Chip comparative hybridization analysis

identified 121 direct downstream binding targets for TCF21.

The gene networks and cellular pathways potentially regulated

by these TCF21 targets were identified. One of the main bHLH

targets for TCF21 was the bHLH gene scleraxis (Scx). An

embryonic ovarian gonadal cell culture was used to examine the

functional role of Sry, Tcf21, and Scx to promote an in vitro sex

reversal and induction of Sertoli cell differentiation. SRY and

TCF21 were found to induce the initial stages of Sertoli cell

differentiation, whereas SCX was found to induce the later stages

of Sertoli cell differentiation associated with pubertal develop-

ment using transferrin gene expression as a marker. Therefore, a

cascade of SRY followed by TCF21 followed by SCX appears to

promote, in part, Sertoli cell fate determination and subsequent

differentiation. The current observations help elucidate the

initial molecular events involved in the induction of Sertoli cell

differentiation and testis development.

basic helix-loop-helix, cell differentiation, ChIP-Chip, SRY,

scleraxis, Scx, Sertoli cells, sex determination, testis development

INTRODUCTION

Male gonadal sex determination is induced by the

expression of the testis-determining factor SRY (sex-determin-

ing region on the Y chromosome) in a subset of somatic cells

that are induced to differentiate into Sertoli cells. Sertoli cells

are somatic cells that orchestrate germ cell differentiation and

nurture their development in the fetal, postnatal, pubertal, and

adult spermatogenic stages. Sertoli cells are the first cell type

known to differentiate within the gonad from the bipotential

precursor cell lineage to initiate testis differentiation. In

contrast, supporting cell precursors in XX female gonads

differentiate into granulosa cells and surround the meiotic germ

cells to form primordial follicles [1]. The initiation of testis

cord formation occurs at the onset of gonadal sex determina-

tion, whereas ovarian follicle assembly occurs after birth in the

rodent [2]. The current study was designed to investigate the

molecular events involved in promoting Sertoli cell fate

determination and subsequent differentiation.

During the differentiation of testicular Sertoli cells, SRY

induces the expression of its autosomal counterpart Sox9 by

acting synergistically with steroidogenic factor 1 (SF1) on its

testis-specific enhancer region on the promoter [3, 4]. After

reaching a threshold expression level, SOX9 acts on Sry to

repress expression [5, 6]. Concurrently, Sox9 expression is

maintained via a positive-feedback mechanism involving

fibroblast growth factor 9 (FGF9) and prostaglandins [7, 8].

Loss of function mutations of Sry and Sox9 produce a male-to-

female sex reversal phenotype in XY males, whereas the gain

of function causes induction of testis development in XX

females, suggesting SRY initiates and then SOX9 maintains

testis development. Despite extensive research regarding the

functions of SRY and SOX9 in mammals, downstream targets

and genome-wide actions remain poorly understood. Recently,

we used a genome-wide chromatin immunoprecipitation

(ChIP) followed by a promoter tiling array chip (Chip) in a

ChIP-Chip comparative hybridization approach to identify the

in vivo downstream targets of SRY and SOX9 in the rat gonad

[9]. This analysis identified 71 direct downstream binding

targets for SRY and 109 binding targets for SOX9, with only

five that overlap between the two. Recently, we also

demonstrated that the growth factor neurotropin 3 (Ntf3) and

the basic helix-loop-helix (bHLH) transcription factor Tcf21

(bHLHa23) are direct downstream targets of SRY [10, 11].

TCF21 was found to induce differentiation of Sertoli cells in

vitro in rat Embryonic Day (E) 13 ovary primary cell cultures.

Many cell differentiation events during early development

involve a cascade of bHLH gene expression, such as muscle

cell differentiation [12], neuronal differentiation [13–18], lung

cell morphogenesis [19], and cardiac cell differentiation, [20,

21]. Therefore, the current study investigated the potential that

a cascade of bHLH factors may be involved in Sertoli cell

differentiation and testis development.

The bHLH proteins are characterized by the helix-loop-helix

(HLH) domain, which mediates the interactions that form

homo- and heterodimers between these proteins [22, 23]. They

also contain a highly charged basic region upstream of the

HLH domain, which functions as a specific DNA-binding

domain that recognizes the bHLH consensus sequence known

as an E-box (CANNTG). The bHLH protein heterodimers,

which consist of a ubiquitous class of bHLH protein and more

tissue-specific class of bHLH proteins, bind at this conserved

E-box domain. The formation of these heterodimers can

promote cell-specific gene expression that influences cellular

differentiation and proliferation [23]. The bHLH proteins are

negatively regulated by another class of HLH proteins termed

the inhibitors of differentiation (Id). These Id proteins lack a

basic region, which allows them to inhibit transcriptional

activation with the bHLH proteins they bind [22]. The

phylogenetic and unified nomenclature of the bHLH family

of genes have been previously described [23].

Supported by an NICHD National Institutes of Health grant to M.K.S.

Correspondence: E-mail: [email protected]

Received: 2 February 2012.

First decision: 12 March 2012.

Accepted: 2 October 2012.

Ó 2012 by the Society for the Study of Reproduction, Inc.

eISSN: 1529-7268 http://www.biolreprod.org

ISSN: 0006-3363

1 Article 131

Downloaded from www.biolreprod.org.

A previous rat and mouse developmental microarray

database that covers several stages of testis development was

used to determine a potential cascade of bHLH expression [24,

25]. The corresponding early gonadal period of mouse E11.5 is

E13 in the rat, with no testis cords detected. Tcf21 was found to

be highly expressed in Sertoli cells during this rat develop-

mental period, corresponding to the onset of testis differenti-

ation [11]. Tcf21 has been shown to be a downstream-regulated

target of SRY, and TCF21 promotes Sertoli cell differentiation

[9, 11]. To investigate the cascade of bHLH genes associated

with Sertoli cell differentiation, a genome-wide ChIP-Chip

approach was used to identify downstream TCF21 binding

targets. A major bHLH target previously shown to be

associated with Sertoli cell differentiation was scleraxis (Scx)

(bHLHa41) [26]. The sequential actions of SRY, TCF21, and

SCX were found to induce, in part, fetal ovarian cell cultures to

differentiate in vitro to express an adult Sertoli cell gene (i.e.,

transferrin). Observations suggest SRY induces a cascade of

bHLH factors that promote Sertoli cell differentiation and testis

development.

MATERIALS AND METHODS

Tissue Preparation

Harlan Sprague-Dawley rats (Harlan, Inc.) were used for the experiment. All

rats were kept in a temperature-controlled environment (228C) and given food

and water ad libitum. Estrous cycles of female rats were monitored by cellular

morphology from vaginal smears. Rats in early estrus were paired with males

overnight, and mating was confirmed by sperm-positive smears, denoted as Day

0 of pregnancy. Pregnant rats were euthanized at E13 of pregnancy, and

embryonic gonads were collected for ChIP. Sex was determined by PCR using

primers specific for Sry on genomic DNA isolated from embryo tails as

previously described [27]. All procedures were approved by the Washington

State University Animal Care and Use Committee (IACUC approval 02568–26).

Immunohistochemistry

Immunolocalization of scleraxis was performed according to methods

described previously [11]. Briefly, embryonic testis sections (E16) were

deparaffinized, rehydrated through a graded ethanol series, boiled for 10 min in

10 mM sodium citrate buffer to expose the antigens, washed with 0.1% Triton-

X solution, and then blocked with 10% serum of the species the secondary

antibody was raised in for 60 min before incubation with 0.5 lg/ml of primary

rabbit anti-SCX antibody (Life Technologies) and 5.0 lg/ml of anti-Mu¨llerian

hormone (AMH) antibody (R&D Systems) for 18 h. The sections were then

washed with PBS and stained with diaminobenzidine according to the

manufacturer’s instructions (Vector Laboratories). Negative-control experi-

ments were performed using a nonimmune immunoglobulin (Ig) G at 0.5 and

5.0 lg/ml concentrations (Sigma).

In Vivo ChIP Assay

Carrier ChIP assay was adopted from O’Neill et al. [28] and performed

according to Bhandari et al. [11]. The testis was removed from the

mesonephros. However, residual mesonephros cannot be eliminated, so some

contamination may persist. Male gonads from 30 E13 (12- to 18-tail somite

stage) rat embryos were used per array. Drosophila SL2 cells were used as a

carrier. Densely grown cells (;5 3 10

cells) were pelleted and washed three

times in ice-cold PBS, then resuspended in NB buffer (15 mM Tris-HCL [pH

7.4], 60 mM KCl, 15 mM NaCl, 5 mM MgCl

, 0.1 mM ethyleneglycoltetra-

acetic acid, 0.5 mM 2-mercaptoethanol, and 0.1 mM PMSF). Testis samples

were mixed with SL2 cells and homogenized to isolate nuclei. After a minute of

homogenization in a glass homogenizer (15 dounces), the lysate was put on ice

for 3 min before repeating the homogenization and icing for a total of 20

repeats. Nuclei were pelleted, resuspended in NB buffer and 5% sucrose, then

pelleted and resuspended again in digestion buffer (50 mM Tris-HCl [pH 7.4],

0.32 M sucrose, 4 mM MgCl

, 1 mM CaCl

, and 0.1 mM PMSF). Following

micrococcal nuclease digestion, the digested samples were gently centrifuged

(,1000 3 g) and the supernatant set aside on ice. A fraction of the supernatant

was used as input. The remaining supernatant was incubated with either

nonimmune IgG or anti-TCF21 antibody at 48C overnight. After incubation

with preswollen protein A-Sepharose beads, the bead-bound immunoprecipi-

tates were centrifuged gently (,1000 3 g) and washed five times with TE

buffer (50 mM Tris-HCl [pH 7.5], 10 mM ethylenediaminetetra-acetic acid, 5

mM sodium butyrate, and 50–150 mM NaCl). Immunoprecipitated DNA was

eluted with elution buffer (TE plus 1% SDS) and extracted with a phenol/

chloroform extraction. Final concentration of immunoprecipitated DNA varied

from 200 to 500 ng/assay. Three different experiments with three different

biological sample ChIPs were performed. Exactly 30 ng of immunoprecipitated

DNA from each assay were amplified with a whole-genome amplification kit

developed by Sigma (WGA2-50 RXN). At least five separate whole-genome

amplifications were performed for each sample, and DNA was pooled. Pooled

whole-genome amplified DNA was purified by using the Wizard SV40 PCR

Clean-up System (A9281; Promega). Purified DNA was checked on the gel and

sent to Roche NimbleGen for ChIP-chip hybridization, and a three-plex

promoter array was used for competitive hybridizations. Confirmation of the

selected candidate binding targets used a semiquantitative PCR method [9] as

described below.

Analysis of ChIP-Chip Data

For ChIP-chip hybridization, Roche NimbleGen’s Rat ChIP-Chip 3x720K

RefSeq Promoter Array was used. The enrichment for each probe on the array

was calculated as the log ratio of the intensities of hybridization for TCF21

ChIP DNA (Cy5) to control DNA from IgG ChIP control (Cy3). Arrays

contained on average 4000 bp of promoter for each of 15 287 promoters in the

rat genome, corresponding to 15 600 RefSeq transcripts (;3880 bp upstream

and 970 bp downstream from transcription start site).

The bioinformatics analysis of ChIP-Chip data was performed as previously

described [29]. For each hybridization experiment, raw data from both the Cy3

and Cy5 channels were imported into R (http://www.R-project.org), checked

for quality, and converted to MA values (M ¼Cy5 Cy3; A ¼[Cy5

Cy3]/2).

The R codes that were used are available at http://skinner.wsu.edu/microarray.

html.

Within each array, probes were separated into groups by guanine-cytosine

(GC) content, and each group was separately normalized using the Loess

normalization procedure [30]. After each array was normalized within array, the

arrays were then normalized across arrays using the A-quantile normalization

procedure [31]. Both Loess and A-quantile normalization were performed using

the limma library [32]. Following normalization, each probe’s normalized M

(and then A) values were replaced with the median value of all probe normalized

M (and then A) values across all arrays within a 600-bp window [33–35]. The

average size of the DNA fragment was 600 bp, so the window correlated.

Following normalization, the median intensity difference between Cy5 and Cy3

of a 600-bp window was determined. Significance was assigned to probe

differences between experimental (TCF21) and IgG control by calculating the

median value of the intensity differences as compared to a normal distribution

scaled to the experimental mean and SD of M. Regions of interest were then

determined by combining consecutive probes with P-values less than 13 10

3

Z-score and P-value were computed from that distribution with the use of R code

analysis [29]. The statistically significant peaks of hybridization were identified

and the P-value associated with each peak presented. Each peak of interest was

then annotated for the gene. Every promoter exceeding the intensity threshold

was considered to be positive for TCF21 binding. The final list of TCF21 targets

includes the promoter-proximal regions that made the threshold in an average for

all three experiments. Hybridization signals for all the candidate promoters that

were within the cutoff line (P  1 3 10

7

) were plotted (average of the three

experiments). The genes that were not in the list but seemed to be masked by

IgG-negative signals were designated as questionable positives. Selected

questionable positive promoters were manually chosen and confirmed by PCR.

Gene Network Analysis

To move beyond examining single-gene effects on a developmental

process, gene network analysis can be employed to identify groups (e.g.,

modules) of genes for which expression is regulated in a coordinated and

functionally interconnected manner (gene network). One important end product

of gene coexpression network analysis is the construction of gene modules

composed of highly interconnected genes. In the current study, a network for

TCF21 downstream binding target genes was constructed separately using the

Pathway Studio software (Ariadne Genomics, Inc) and previously published

criteria for developmental network analysis [36].

Pathway Analysis

The pathway analysis of direct downstream target genes was performed

according to the protocol previously described by Nilsson et al. [36]. The

downstream binding targets of TCF21 and their interconnections were analyzed

BHANDARI ET AL.

2 Article 131

Downloaded from www.biolreprod.org.

for KEGG (Kyoto Encyclopedia for Genes and Genome, Kyoto University,

Japan) pathway enrichment using Pathway Express, a Web-based tool freely

available as part of the Onto-Tools (http://vortex.cs.wayne.edu). The Partek

software was used to identify gene functional categories. A program based on

literature analysis (Pathway Studio) was used to evaluate a gene network of

TCF21 targets.

Fetal Gonadal Cell Preparation and Culture

Cell preparation and culture were performed are previously described [11].

E13 embryos (12- to 18-tail somite stage) were collected from timed-pregnant

females as described above. Gonads from E13 animals were dissected, and each

pair of gonads from individual animals was placed into one well of a 24-well

plate with 500 ll of Ham F-12 medium until embryos could be sexed as

described above. The female gonads were then pooled and digested with

trypsin (2.5%) and collagenase (1 mg/ml, type I) plus DNase (3 mg/ml) to

disassociate the cells. All cells from the digested ovaries were then cultured on

100-mm plates in Ham F-12 with 10% bovine calf serum (Sigma). Cells

initially multiplied well in culture and were split two times (1:2) as they reached

confluence, at which point cell division slowed considerably. Cells were

maintained in culture, with the medium changed every 3 days, until growth

plaques were observed at approximately 1 mo. These growth plaques were then

collected for further propagation, and frozen stocks were prepared for

subsequent cell culture such that cells could be maintained at fewer than 12

subcultures.

Transfection

The E13 ovarian cell cultures between subcultures 8 and 12 were

transfected using Lipofectamine 2000 (Invitrogen). For each well of a six-

well plate, 2 lg of expression plasmid in 250 ll of Opti-Mem medium (Sry-

pCMV-HA, Tcf21-pCMV-MYC, Scx-pCMV-HA; Invitrogen) were mixed

with 10 ll of Lipofectamine 2000 in 250 ll of Opti-MEM medium and

incubated for 20 min at room temperature. This 500-ll mix was added to each

well containing approximately 90% confluent, cultured E13 ovarian cells in 2

ml of Ham F-12 without antibiotics and incubated at 328C for 24 h. After 24 h,

medium was aspirated from cells and replaced with 2 ml of Ham F-12 with

10% serum. Cells were cultured for 2–10 days and collected in TRIzol reagent

(Sigma) for RNA extraction.

PCR and ChIP Confirmation

Total RNA was extracted from frozen cells using the TRIzol extraction

method. A total of 2 lg of total RNA was used to generate cDNA using

Moloney Murine Leukemia Virus Reverse Transcriptase (Invitrogen) according

to the manufacturer’s instruction. At the end, cDNA was diluted to 20 ng/ll,

and exactly 50 ng of cDNA were used for PCR. PCR was performed using the

following conditions: 958C for 5 min; followed by 25–35 cycles of 958C for 30

sec to 1 min, 588C for 30 sec, and 728C for 40 sec; followed by 728C for 7 min.

PCR product was visualized in 2% agarose gel with ethidium bromide. Primers

were designed for ChIP confirmation for selected statistically significant targets

from the region of probe hybridization peaks and tested using WGA kit-

amplified ChIP DNA in a PCR. Primers are listed in Supplemental Table S1 (all

Supplemental Data are available online at www.biolreprod.org). The ChIP-

PCR was performed with 25–30 cycles, so it was in the linear curve of the PCR

and used as a semiquantitative procedure, as previously described [9].

RESULTS

Downstream Binding Targets of TCF21 During Male

Gonadal Sex Determination

A bHLH protein, TCF21 was previously shown to be a

direct downstream target of SRY to induce Sertoli cell

differentiation [11]. The initial objective of the current study

was to identify the genome-wide binding targets for TCF21,

with a focus given to bHLH targets. Observations would

extend previous observations involving the identification of the

downstream targets of SRY and SOX9 [9] and help elucidate

the molecular events involved in the induction of Sertoli cell

differentiation and testis development. A modified ChIP

followed by a Chip involving a comparative hybridization of

a TCF21 antibody ChIP versus a control nonimmune IgG ChIP

was used. This novel modified ChIP-Chip corrects for false-

positive targets due to nonimmune IgG nonspecific binding, as

previously described [9]. The rat Chip used contains

approximately 4 kb of the majority of promoters in the

genome. An R code was used to identify the positive-binding

target genes, remove false positives, and determine statistical

significance for the binding. The E13 rat male gonad was

microdissected and identified with an embryonic tail clip and

Sry PCR to confirm XY male genotype. The chromatin was

isolated from this E13 male gonad and used in the ChIP. An

antibody for TCF21 and corresponding nonimmune IgG were

also used in the ChIP. The two different ChIP DNA samples

were labeled with unique fluorescent colors and used in a

competitive hybridization rat Chip analysis. A total of 1383

promoters were detected with P , 0.001, and a statistical

cutoff of P , 1 3 10

7

was used to select potential targets to

examine hybridization profiles and determine the downstream

binding targets. The final analysis identified 121 downstream

direct gene binding targets (Table 1). The functional catego-

rization of these genes, chromosome location, and statistical

significance are presented. The hybridization profiles for these

gene targets are provided in Supplemental Figure S1. A

minimum of three adjacent oligonucleotides on the tiling array

had to have a statistically significant hybridization of P ,

0.001 to be considered. Representative examples of direct

binding targets are presented in Figure 1. A confirmation of the

ChIP analysis of selected downstream targets was performed

with a TCF21 antibody ChIP followed by PCR. The anti-

TCF21 ChIP-PCR confirmation data are presented next to the

hybridization profiles in Figure 1. All 121 targets had bHLH

response elements (i.e., E-box) present, as indicated in Table 1.

The ChIP-Chip comparative hybridization data profile

identifies positive peaks representing the TCF21 ChIP pull-

down, whereas negative peaks represent the nonimmune IgG

ChIP (Fig. 1 and Supplemental Fig. S1). In the event a strong

negative IgG peak is adjacent to a positive TCF21 peak, the

positive target can be masked and considered to be question-

able. A list of questionable binding targets is presented in Table

S2. Representative profiles for these questionable targets are

presented in Figure 2. A ChIP-PCR was performed to confirm

the target and is presented. Clearly, the comparative hybrid-

ization allows the detection of nonspecific IgG binding

(negative bars presented) and eliminates false positives in the

ChIP-Chip assay. However, strong negative hybridization of

the IgG can mask some positive targets (Fig. 2).

The current study was designed to identify the binding

targets for TCF21 and not to indicate direct transcriptional

regulation during the E13 period. However, a previous

experiment using the E13 ovarian cell culture overexpressed

the TCF21 with a transient transfection of an expression

construct [11]. A microarray of the cells identified genes that

had significantly altered expression due to the overexpression

of TCF21 [11]. A comparison of this regulated gene set with

the TCF21 binding targets identified in the current study

identified a number of binding targets that are transcriptionally

regulated by TCF21. These statistically significant (P , 0.05)

regulated TCF21 binding targets include Arl1, Rpl24, Stk3,

Tipin, Pfkp, RGD1306839, Mcdcom1, Sl00a3, Phox2a, F7,

Akap4, and Itih4. Therefore, more than 10% of the binding

targets identified were previously shown to be regulated by

TCF21 [11]. Because TCF21 may also have a role in gene

repression and may regulate genes at later stages of

development, not all genes are anticipated to be regulated at

E13. Observations confirm a transcriptional role for TCF21 at

some of the binding targets identified.

Analysis of the TCF21 DNA-binding site for the 121

downstream targets identified a consensus binding motif (Fig.

BASIC HLH CASCADE DURING MALE SEX DETERMINATION

3 Article 131

Downloaded from www.biolreprod.org.

TABLE 1. TCF21 downstream binding promoter target regions.

Gene symbol

GenBank/reference

sequence no. Chromosomal location P-value

E-box Gene name

Apoptosis

Faim3 NM_001014843 chr13:43816092–43817078 2.46E-09 Yes Fas apoptotic inhibitory molecule 3

Thap11 NM_001107422 chr19:35690656–35691256 6.56E-08 Yes THAP domain containing 11

Cell cycle

Cenpt NM_001024257 chr19:35690656–35691256 6.56E-08 Yes Centromere protein T

Cytoskeleton-extracellular matrix

Cldn22 NM_001110143 chr16:47625353–47626043 2.90E-09 Yes Claudin 22

Krtap13-2 NM_001109325 chr11:28609435–28610035 2.69E-09 Yes Keratin associated protein 13–2

Madcam1 NM_019317 chr7:11555057–11555657 1.02E-08 Yes Mucosal vascular addressin cell adhesion

molecule 1

RGD1562043 NM_001110144 chr16:47625353–47626043 2.90E-09 Yes Similar to claudin-22

Tspan33 NM_001109227 chr4:56583349–56584085 2.75E-14 Yes Tetraspanin 33

Development

Brwd1 NM_001107106 chr11:36388682–36389686 4.04E-09 Yes Bromodomain and WD repeat domain

containing 1

Crygn NM_001106573 chr4:5769730–5770516 3.62E-10 Yes Crystallin, gamma N

LOC298109 BC086943 chr5:78494238–78494913 7.45E-08 Yes Alpha-2u globulin PGCL2

LOC691093 NM_001109621 chr7:11555057–11555657 1.02E-08 Yes Similar to gene trap ROSA b-geo 22

Lrrc56 NM_001024902 chr1:201386300–201386900 2.66E-08 Yes Leucine-rich repeat containing 56

Meox1 NM_001108837 chr10:90943030–90943630 6.82E-08 Yes Mesenchyme homeobox 1

RGD1565657 NM_001106728 chrX:82516111–82516711 1.91E-13 Yes Similar to germ cell-less

DNA repair

Lig1 NM_001024268 chr1:73729084–73730186 3.61E-11 Yes Ligase I, DNA, ATP-dependent

Tipin NM_001025287 chr9:112832611–112833286 8.54E-15 Yes Timeless interacting protein

Electron transport

Cyp3a23/3a1 NM_013105 chr12:9598130–9598730 1.28E-15 Yes Cytochrome P450, family 3, subfamily a,

polypeptide 23/polypeptide 1

Tmx4 NM_001100529 chr3:122638091–122639096 3.33E-13 Yes Thioredoxin-related transmembrane protein 4

Epigenetics

Aicda NM_001100779 chr4:159003247–159003847 5.17E-11 Yes Activation-induced cytidine deaminase

Cbx3 NM_001008313 chr4:79703674–79704274 3.55E-10 Yes Chromobox homolog 3 (HP1 gamma homolog,

Drosophila)

Cdx2 NM_023963 chr12:8294748–8295348 5.56E-09 Yes Caudal type homeo box 2

Ehmt2 NM_212463 chr20:4033685–4034285 2.94E-08 Yes Euchromatic histone lysine N-methyltransferase

Ruvbl2 NM_001025405 chr1:95910251–95911046 2.37E-08 Yes RuvB-like 2 (E. coli)

Wbscr22 NM_001135743 chr12:22726326–22727079 4.49E-16 Yes Williams-Beuren syndrome chromosome region

Growth factors

Ins2 NM_019130 chr1:202935144–202935854 9.27E-08 Yes Insulin 2

Tgfa NM_012671 chr4:120352748–120353348 3.52E-08 Yes Transforming growth factor alpha

Immune response

C4-2 NM_001002805 chr20:4103885–4104595 4.08E-09 Yes Complement component 4, gene 2

C4b AY149995 chr20:4103885–4104595 4.08E-09 Yes Complement component 4B (Chido blood

group) /// complement component 4, gene 2

F7 NM_152846 chr16:81361265–81361865 2.02E-08 Yes Coagulation factor VII (serum prothrombin

conversion accelerator)

LOC259244 AB039824 chr5:78166654–78167254 8.03E-13 Yes Alpha-2u globulin PGCL3

Rt1.aa NW_001088093 chr20:3408490–3409090 2.97E-10 Yes MHC class I RT1.Aa alpha-chain

RT1-CE7 NM_001008845 chr20:3408490–3409090 2.97E-10 Yes RT1 class I, locus CE7

Metabolism and transport

Afg3l2 NM_001134864 chr18:63975093–63975767 1.94E-08 Yes AFG3 (ATPase family gene 3)-like 2 (yeast)

Akr7a3 NM_013215 chr5:158139006–158139707 4.40E-09 Yes Aldo-keto reductase family 7, member A3

(aflatoxin aldehyde reductase)

Aldoa NM_001177305 chr1:185970974–185971574 5.45E-11 Yes Aldolase A, fructose-bisphosphate

Alg2 NM_001100710 chr5:64097238–64098265 8.53E-08 Yes Asparagine-linked glycosylation 2, alpha-1,3-

mannosyltransferase homolog (S. cerevisiae

)

x3l1 NM_001008522 chr9:56944915–56945515 4.62E-11 Yes Aldehyde oxidase 3-like 1

Car2 NM_019291 chr2:88092498–88093184 1.61E-11 Yes Carbonic anhydrase II

Ces2l NM_133586 chr1:267886199–267886799 4.24E-08 Yes Carboxylesterase 2-like

Cml5 NM_080884 chr4:120001822–120002422 1.47E-08 Yes Camello-like 5

Cnga1 NM_053497 chr14:38015769–38016369 1.52E-08 Yes Cyclic nucleotide gated channel alpha 1

Ctu2 NM_001037094 chr19:52766059–52766659 7.15E-09 Yes Cytosolic thiouridylase subunit 2 homolog (S.

pombe)

Dhodh NM_001008553 chr19:39476737–39477337 3.21E-15 Yes Dihydroorotate dehydrogenase

Dhrs7b NM_001008507 chr10:46969055–46970158 4.67E-10 Yes Dehydrogenase/reductase (SDR family) member

Fetub NM_053348 chr11:80284316–80285096 4.50E-08 Yes Fetuin B

Gys1 NM_001109615 chr1:95910251–95911046 2.37E-08 Yes Glycogen synthase 1, muscle

Hibch NM_001013112 chr9:45648153–45648944 8.01E-08 Yes 3-Hydroxyisobutyryl-coenzyme A hydrolase

Nat8 NM_022635 chr4:120001822–120002422 1.47E-08 Yes N-acetyltransferase 8

BHANDARI ET AL.

4 Article 131

Downloaded from www.biolreprod.org.

TABLE 1. Continued.

Gene symbol

GenBank/reference

sequence no. Chromosomal location P-value

E-box Gene name

Ostalpha NM_001107087 chr11:70112264–70112954 4.44E-09 Yes Organic solute transporter alpha

Pla2g6 NM_001005560 chr7:117306739–117307639 4.25E-08 Yes Phospholipase A2, group VI (cytosolic, calcium-

independent)

RGD1309808 NM_001134801 chr7:115651493–115652093 4.27E-14 Yes Similar to apolipoprotein L2; apolipoprotein L-II

Sec61b NM_001106654 chr5:64097238–64098265 8.53E-08 Yes Sec61 beta subunit

Slc22a7 NM_053537 chr9:9932589–9933384 3.61E-10 Yes Solute carrier family 22 (organic anion

transporter), member 7

Tgm2 NM_019386 chr3:148862141–148862954 3.54E-08 Yes Transglutaminase 2, C polypeptide

Trappc6a NM_001109410 chr1:78859327–78860050 1.14E-09 Yes Trafficking protein particle complex 6A

Proteolysis

Itih4 NM_019369 chr16:6317228–6318354 4.94E-25 Yes Inter alpha-trypsin inhibitor, heavy chain 4

LOC299282 NM_182474 chr6:128433633–128434350 6.96E-12 Yes Serine protease inhibitor

Rnf166 NM_001002279 chr19:52766059–52766659 7.15E-09 Yes Ring finger protein 166

Receptors and binding proteins

Il1rap NM_012968 chr11:76225521–76226121 6.45E-09 Yes Interleukin 1 receptor accessory protein

Il1rapl1 NM_177935 chrX:74472070–74472670 1.34E-12 Yes Interleukin 1 receptor accessory protein-like 1

Olr1320 NM_001000471 chr8:42711659–42712259 7.91E-09 Yes Olfactory receptor 1320

Olr1365 NM_214824 chr10:12301149–12301839 4.62E-08 Yes Olfactory receptor 1365

Olr1631 NM_001000837 chr15:26595553–26596254 8.92E-08 Yes Olfactory receptor 1631

Olr1641 NM_001000100 chr15:27870103–27870792 7.52E-14 Yes Olfactory receptor 1641

Olr1688 NM_001000275 chr20:538973–539573 3.01E-08 Yes Olfactory receptor 1688

Olr1718 NM_214460 chr20:1128784–1129384 7.70E-09 Yes Olfactory receptor 1718

Olr542 NM_001000566 chr3:70577439–70578118 7.54E-08 Yes Olfactory receptor 542

Olr675 NM_001000632 chr3:73353375–73354052 6.07E-10 Yes Olfactory receptor 675

Olr803 NM_001000853 chr4:70273009–70273609 6.27E-08 Yes Olfactory receptor 803

Olr853 NM_001000398 chr5:70398861–70399461 6.88E-09 Yes Olfactory receptor 853

Pear1 NM_001134959 chr2:179828477–179829187 6.00E-09 Yes Platelet endothelial aggregation receptor 1

S100a3 NM_053681 chr2:182884071–182884859 4.63E-09 Yes S100 calcium-binding protein A3

S100a4 NM_012618 chr2:182884071–182884859 4.63E-09 Yes S100 calcium-binding protein A4

Tshr NM_012888 chr6:115021810–115022410 2.31E-08 Yes Thyroid-stimulating hormone receptor

Vom1r29 AY510346 chr1:63195048–63195648 2.89E-09 Yes Vomeronasal 1 receptor, 29

Vom1r41 AY510299 chr1:71260215–71260815 1.57E-08 Yes Vomeronasal 1 receptor, 41

Vom2r69 NM_001099468 chr14:741194–741794 1.28E-09 Yes Vomeronasal 2 receptor, 69

Signaling

Adcy7 NM_053396 chr19:20074415–20075095 1.03E-08 Yes Adenylate cyclase 7

Akap4 NM_024402 chrX:27527026–27527727 6.40E-10 Yes A kinase (PRKA) anchor protein 4

Camkk1 NM_031662 chr10:59909090–59909990 1.15E-10 Yes Calcium/calmodulin-dependent protein kinase

kinase 1, alpha

Cmtm1 NM_001029914 chr19:621615–622312 5.26E-09 Yes CKLF-like MARVEL transmembrane domain

containing 1

Fermt3 NM_001127543 chr1:209693441–209694041 2.75E-11 Yes Fermitin family homolog 3 (Drosophila)

Grinl1a NM_183402 chr8:75957303–75957903 3.47E-12 Yes Glutamate receptor, ionotropic, N-methyl-

D-

aspartate-like 1A

Hras NM_001130441 chr1:201386300–201386900 2.66E-08 Yes Harvey rat sarcoma virus oncogene

Itpripl1 NM_001025043 chr3:114686510–114687200 4.95E-09 Yes Inositol 1,4,5-triphosphate receptor interacting

protein-like 1

Pbx3 NM_001107834 chr3:13406614–13407214 6.72E-12 Yes Pre-B-cell leukemia homeobox 3

Pde2a NM_031079 chr1:158951501–158952284 6.05E-33 Yes Phosphodiesterase 2A, cGMP-stimulated

Pfkp NM_206847 chr17:74761586–74762186 4.51E-08 Yes Phosphofructokinase, platelet

Ppp1r7 NM_001009825 chr9:92620909–92621609 1.60E-09 Yes Protein phosphatase 1, regulatory (inhibitor)

subunit 7

Ppp4c NM_134359 chr1:185970974–185971574 5.45E-11 Yes Protein phosphatase 4, catalytic subunit

Rab3c NM_133536 chr2:41673441–41674135 1.22E-25 Yes RAB3C, member RAS oncogene family

Rab43 NM_001024331 chr4:121999604–122000287 3.53E-10 Yes RAB43, member RAS oncogene family

RGD1562638 NM_001100944 chr16:68750676–68751465 2.09E-37 Yes Similar to MAP/microtubule affinity-regulating

kinase 3

Sgk493 NM_001108007 chr6:6551181–6551966 4.63E-08 Yes Protein kinase-like protein SgK493

Sgsm1 ENSRNOT00000000898 chr12:44329558–44330263 4.76E-11 Yes Small G protein-signaling modulator 1

Skp1 NM_001007608 chr10:37662699–37663391 3.34E-09 Yes S-phase kinase-associated protein 1

Transcription

Dnajc30 NM_001109024 chr12:22726326–22727079 4.49E-16 Yes DnaJ (Hsp40) homolog, subfamily C, member

Mtf1 NM_001108677 chr5:144134250–144134850 8.11E-08 Yes Metal-regulatory transcription factor 1

Pask NM_001009362 chr9:92620909–92621609 1.60E-09 Yes PAS domain containing serine/threonine kinase

Phox2a NM_053869 chr1:159272192–159273082 5.16E-09 Yes Paired-like homeobox 2a

Reck NM_001107954 chr5:60338635–60339805 5.17E-08 Yes Reversion-inducing-cysteine-rich protein with

kazal motifs

Translation and protein modification

A1cf NM_133400 chr1:235862714–235863720 7.62E-11 Yes APOBEC1 complementation factor

Eif4a2 NM_001008335 chr11:79958862–79959547 1.17E-08 Yes Eukaryotic translation initiation factor 4A2

Arl1 NM_022385 chr7:25400533–25401133 8.75E-10 Yes ADP-ribosylation factor-like 1

BASIC HLH CASCADE DURING MALE SEX DETERMINATION

5 Article 131

Downloaded from www.biolreprod.org.

3). The TCF21 DNA-binding motif identified contained an E-

box consensus sequence and flanking sequence on both sides.

The largest nucleotide shown in Figure 3 represents the most

prominent nucleotide present, but others are possible, provid-

ing flexibility in TCF21 binding. Although a more specific

TCF21 binding site was identified, the core E-box sequence

was present. A predominant E-box sequence present was

CANNTC, which previouslywas shown to be used by other

bHLH factors [37, 38].

TCF21 Downstream Gene Target Pathway and Gene

Network Analysis

The TCF21 downstream binding target gene functional

categories are presented in Table 1 and Figure 4. Predominant

gene categories include metabolism/transport, signaling, recep-

tive binding proteins, and translation/promoter modification and

development (Fig. 4). Further analysis of the TCF21 gene target

list for specific pathways and cellular processes is presented in

Table 2. The pathways containing more than three TCF21 target

genes were considered. The total number of genes in the

pathways and impact of the correlated genes is also presented in

Table 2. The pathway with the highest number of genes present

was the olfactory transduction pathway (Supplemental Fig, S2).

Other predominant developmentally relevant pathways include

the cell adhesion molecule, chemokine signaling pathway, and

mitogen-activated protein kinase pathway (Table 2).

A gene network analysis was performed by considering all

the TCF21 binding target genes and the use of a literature-

TABLE 1. Continued.

Gene symbol

GenBank/reference

sequence no. Chromosomal location P-value

E-box Gene name

Arl9 NM_212459 chr14:33459931–33461186 4.49E-18 Yes ADP-ribosylation factor-like 9

Hnrnpa2b1 NM_001104613 chr4:79703674–79704274 3.55E-10 Yes Heterogeneous nuclear ribonucleoprotein A2/

Hnrnpa3 NM_001111294 chr3:58349049–58349742 1.34E-08 Yes Heterogeneous nuclear ribonucleoprotein A3

Mrpl17 NM_133539 chr1:163555460–163556060 1.05E-09 Yes Mitochondrial ribosomal protein L17

Rtf1 NM_001108958 chr3:106189897–106190497 4.65E-12 Yes Rtf1, Paf1/RNA polymerase II complex

component, homolog (S. cerevisiae)

Rbm16 NM_139094 chr1:38097295–38098088 1.92E-24 Yes RNA binding motif protein 16

RGD1561102 NM_001106777 chr7:25400533–25401133 8.75E-10 Yes Similar to ribosomal protein S12

Miscellaneous and unknown

Fam3b NM_001107102 chr11:37508106–37508821 9.74E-09 Yes Family with sequence similarity 3, member B

Otud5 NM_001037496 chrX:26696636–26697315 4.58E-09 Yes OTU domain containing 5

RGD1306595 NM_001025626 chr10:68345133–68345813 6.04E-08 Yes Similar to hypothetical protein

Tm9sf2 NM_001005554 chr15:107224987–107225587 6.16E-10 Yes Transmembrane 9 superfamily member 2

Tmem119 NM_001107155 chr12:43859955–43860555 1.66E-09 Yes Transmembrane protein 119

All targets have statistical significance at P , 1 3 10

7

FIG. 1. A and B) Hybridization signals of selected direct downstream binding target genes (Brwd1 and Camkk1) in the ChIP-Chip analysis and the PCR

confirmation of target genes. PCR was performed with primers designed from the region of significant binding, indicated by arrows. The size marker

(Marker), genomic DNA (Input), IgG ChIP (IgG), and TCF21 ChIP (aTCF21) are presented. The relative hybridization signal (black bar) for each

oligonucleotide probe on the tiling array in the region of interest is presented. The positive bars represent TCF21 ChIP signal, and the negative bars

represent the IgG ChIP signal. Data are representative of three different experiments.

BHANDARI ET AL.

6 Article 131

Downloaded from www.biolreprod.org.

FIG. 2. A and B) Hybridization signals of selected questionable positive TCF21 binding target genes in the ChIP-Chip analysis and the PCR confirmation

of target genes. PCR was performed with primers designed from the region of significant binding, indicated by arrows. The DNA ladder (Marker), genomic

DNA (Input), IgG ChIP (IgG), and TCF21 ChIP (aTCF21) are presented. The relative hybridization signal for each probe on the tiling array in the region of

interest is presented. The positive bars represent TCF21 ChIP signal, and the negative bars represent the IgG ChIP signal. Data are representative of three

different experiments.

FIG. 3. Binding motif identified to be highly conserved in DNA sequences identified in TCF21 binding target gene promoters. Analysis was done using

MEME software with TOMTOM follow up. Approximately 500–800 bp of all direct downstream target gene promoters were aligned and used for analysis.

The core sequence for TCF21 binding was found to be CANNTC, which is common for many bHLH factors. A) The 22-bp core sequence for TCF21

binding. B) The reverse complementary sequence for TCF21 binding motif.

BASIC HLH CASCADE DURING MALE SEX DETERMINATION

7 Article 131

Downloaded from www.biolreprod.org.

based bioinformatics tool to assess gene connectivity (Pathway

Studio). A direct connection gene network for the TCF21 target

genes is presented in Figure 5. Central to this gene network is

the insulin signaling system along with a number of signaling

and transcriptional genes. This potential regulatory network of

TCF21 downstream binding target genes was identified and

speculated to be involved in the downstream regulation of

Sertoli cell differentiation and testis development.

TCF21 Downstream bHLH Target Genes

Tcf21 (bHLHa23) is a member of the bHLH family of

transcription factors [23]. It has been previously shown that the

bHLH proteins heterodimerize to functionally induce tran-

scription of target genes, and as stated, some bHLH genes

should be associated with TCF21. The current study concen-

trated on TCF21 gene targets belonging to the bHLH

transcription factor family. Manual screening of genes was

performed to reveal bHLH gene promoters beyond a statistical

cutoff of P , 1 3 10

7

. A total of six bHLH genes were found

within the cutoff of P , 1 3 10

4

. During the comparative

hybridization, the signals generated by negative-control IgG

often masked the signal generated by anti-TCF21 antibody for

several bHLH genes. Such genes were tested by PCR for

presence in the enriched ChIP-DNA and termed questionable

positive downstream targets (Fig. 2 and Supplemental Table

S2). Four additional bHLH genes were also questionable

targets. A total of 10 bHLH genes were found to be direct

binding targets of TCF21 (Table 3).

The expression profiles of the bHLH candidates indicated

scleraxis had the highest level of expression following the

expression of Tcf21 [23, 24]. The bHLH family transcription

factor scleraxis (Scx)(bHLHa41) was one of the downstream

targets in the questionable positive group. SCX has previously

been shown to be localized to Sertoli cells and to promote

Sertoli cell differentiation in neonatal and prepubertal rats [26].

No transferrin expression occurs during fetal development, and

the only cell type that expresses transferrin is the Sertoli cell.

Therefore, transferrin gene expression is a marker of pubertal

and adult Sertoli differentiation [26]. The presence of Scx in

TCF21 ChIP-enriched DNA was confirmed by ChIP-PCR (Fig.

2). Localization of SCX to Sertoli cells in E13 rat testis was

confirmed by immunohistochemistry (Fig. 6). No SCX was

detected in the interstitial or precursor Leydig cells at E13, only

in Sertoli cells, but SCX has been shown to localize to Leydig

cells in the E16 testis [26]. A control of AMH is presented to

confirm Sertoli localization. Previously, SCX was localized to

postpubertal Sertoli cells [26]. Therefore, scleraxis was selected

as the primary TCF21 bHLH target for further analysis.

Functional Analysis of the Role of TCF21 and SCX in the

Induction of Sertoli Cell Differentiation

The functional role of TCF21 and SCX was investigated

with an E13 fetal gonad stem cell-like culture system as

FIG. 4. TCF21 binding target gene functional categories. Total numbers of target genes associated with a specific category are presented on the y-axis

and gene functional categories on the x-axis.

TABLE 2. Number of regulated TCF21 downstream target genes in the

pathways as analyzed by KEGG, with total number of genes in pathway

indicated.

Pathway name

No. of

genes

Total

pathway genes

Impact

factor

Olfactory transduction 8 732 1.4

HTLV-I infection 6 297 NA

Cell adhesion molecules (CAMs) 6 151 7.6

Endocytosis 5 231 NA

Phagosome 5 191 NA

Autoimmune thyroid disease 5 69 9.8

Type I diabetes mellitus 5 68 9.7

Antigen processing and presentation 4 100 6.0

Allograft rejection 4 61 7.5

Bile secretion 4 74 NA

Graft-versus-host disease 4 63 NA

Viral myocarditis 4 109 NA

Protein processing in endoplasmic 3 165 NA

Chemokine signaling pathway 3 178 NA

Pancreatic secretion 3 106 NA

Insulin signaling pathway 3 130 2.8

MAPK signaling pathway 3 253 1.3

Prostate cancer 3 91 3.6

Fc epsilon RI signaling pathway 3 70 4.2

GnRH signaling pathway 3 91 3.6

Oocyte meiosis 3 115 NA

Regulation of actin cytoskeleton 3 204 1.7

NA, not available.

BHANDARI ET AL.

8 Article 131

Downloaded from www.biolreprod.org.

previously described [11]. The objective was to transiently

overexpress expression constructs for SRY, TCF21, or SCX in

the fetal E13 ovarian cells to induce an in vitro sex reversal and

promote Sertoli cell differentiation. Previously, this was

accomplished with the overexpression of SRY or TCF21 to

induce expression of the Sertoli cell-specific marker gene Amh

[11]. AMH is a marker of perinatal and prepubertal Sertoli cell

differentiation that is lost at the onset of puberty. A pubertal

and adult marker of Sertoli cell differentiation that was selected

was transferrin gene expression [39]. Transferrin is specifically

expressed in Sertoli cells in the pubertal and adult testis. This

fetal ovarian cell culture was used to investigate the ability of

TCF21 and SCX to induce the cascade of events leading to

Sertoli cell differentiated gene expression. This experiment was

not designed to suggest a normal developmental process but,

rather, simply the ability of Tcf21 and Scx to induce the cascade

of events leading to the induction of a more adult state of

Sertoli cell differentiation, similar to other stem cell-type

experiments.

The ability of TCF21 to induce Scx expression was

investigated with the in vitro E13 ovary culture system. The

cell culture of E13 rat ovaries was transfected with a Tcf21

expression construct, and then the expression of Scx was tested

by PCR. Preliminary studies indicated that E13 testis cell

cultures express Scx but that ovarian cell cultures do not. After

48 h of culture, TCF21 induced the expression of Scx in

ovarian cell cultures in vitro (Fig. 7A). In addition to TCF21

inducing Scx expression, SRY also induced Scx expression as

an upstream gene to TCF21. The Postnatal Day (P) 20 testis

was used as a positive control and had Scx expression [26]

(Fig. 7A). Observations suggest that TCF21 induces the

expression of Scx that correlates with the promotion of Sertoli

FIG. 5. Gene network of TCF21 downstream binding target genes. Cell membrane and organelles have been graphically presented. Direct

interconnected genes are shown as solid lines with positive or negative relationships and indirectly connected genes as dotted lines.

TABLE 3. Basic helix-loop-helix genes that are identified as downstream targets (both direct and questionable) of TCF21 in the rat testis during sex

determination.

Gene symbol

GenBank/reference

sequence no. Binding location in chromosome Gene name

Mxd3 NM_145773 chr17:15346632–15347232 Max dimerization protein 3

Hand1 NM_021592 chr10:43421274–43421984 Heart and neural crest derivatives expressed 1

Arnt NM_012780 chr2:190332146–190332746 Aryl hydrocarbon receptor nuclear translocator

Arntl NM_024362 chr1:171060977–171061577 Aryl hydrocarbon receptor nuclear translocator-like

Clock NM_021856 chr14:34214873–34215473 Clock homolog

Id2 NM_013060 chr6:42785587–42786187 Inhibitor of DNA binding 2

Scx NM_001130508 chr7:114503475–114504470 Scleraxis

Usf2 NM_031139 chr1:85993659–85994361 Upstream transcription factor 2, c-fos interacting

Nhlh1 NM_001105970 chr13:88053994–88054594 Nescient helix-loop-helix 1

Mlx NM_001034112 chr10:90101857–90102652 MAX-like protein X

BASIC HLH CASCADE DURING MALE SEX DETERMINATION

9 Article 131

Downloaded from www.biolreprod.org.

cell differentiation. This supports the observation that Scx is a

downstream target for TCF21.

Induction of Scx by TCF21 suggested a cascade of bHLH

genes may be involved in the induction of Sertoli cell

differentiation. Previous studies suggested that TCF21 can

induce Sertoli cell differentiation in vitro by causing sex

reversal of fetal ovarian cell cultures [11]. The final experiment

of the current study examined if SCX can induce the

expression of an adult Sertoli cell differentiated marker in

vitro. Scx was overexpressed in the fetal ovarian cell culture,

and the presence of transferrin was examined by PCR.

Transferrin is a cell-specific marker gene of mature Sertoli

FIG. 6. Immunohistochemical localization of scleraxis protein in the E13 testis. A) Negative-control staining with nonimmune IgG. B) Localization of

Sertoli cells identified by anti-AMH antibody. C and D) Localization of scleraxis identified by antibody against scleraxis (higher-magnification view in D).

Data are representative of three different experiments. Bar ¼ 50 lm.

FIG. 7. E13 gonadal cell culture gene expression analysis. A) In vitro induction of scleraxis (Scx) expression in primary E13 ovarian cell cultures. PCR

was performed with Scx primers, and DNA bands represent scleraxis. DNA ladder (Marker), overexpressed empty expression plasmid (Empty plasmid),

SRY expression plasmid (Sry), Tcf21 expression plasmid (Tcf21), P20 testis cDNA, and water are presented. L19 was used as an internal control. B) In vitro

induction of mature Sertoli cell marker transferrin (Tf) gene expression in E13 primary cell cultures with transfected Scx, Tcf21, and empty expression

constructs. PCR bands indicate Tf and L19 control for 4- and 10-day culture durations. Data are representative of three or more different experiments.

BHANDARI ET AL.

10 Article 131

Downloaded from www.biolreprod.org.

cell function and differentiation [40] and is not expressed

prepubertally. After 4 days of cell culture, transferrin

expression was seen in the cells overexpressing SCX in vitro

(8/10 experiments) (Fig. 7B). Interestingly, TCF21 did not

consistently promote transferrin expression within this 4-day

culture duration (5/10 experiments) but did consistently

promote transferrin expression after 10 days of culture (8/10

experiments) (Fig. 7B). Therefore, an intermediate step, such as

SCX expression, appears to be required. The P20 testis was

used as a positive control for transferrin expression. These

observations suggest that SCX can induce further maturation of

a more adult Sertoli cell differentiation state in vitro. Because

TCF21 was not able to consistently promote transferrin

expression within the 4-day culture duration used but was

able to promote SCX expression, a cascade of SRY-induced

TCF21 followed by TCF21-induced SCX appears to be

required to promote, in part, Sertoli cell differentiation (Fig.

8). However, a large cascade of multiple factors likely is

required in the in vivo differentiation of Sertoli cells.

DISCUSSION

Sertoli cell differentiation is one of the earliest events in sex

determination and testis morphogenesis, which is initiated by

expression of the male sex-determining gene Sry. During gonadal

sex determination and morphogenesis, precursor Sertoli cells

aggregate with primordial germ cells and then become

surrounded by peritubular myoid cells to form cord-like

structures. The cords in the adult testis develop into seminiferous

tubules and support the process of spermatogenesis. As Jost

proposed in his theory of sex determination in the 1940s,

chromosomal or genetic sex is followed by gonadal sex

determination, which is induced by a testis-determining factor

on the Y chromosome to promote phenotypic sex through

endocrine processes [41]. The sex-determining region on the Y

chromosome (SRY) was identified in 1990 as the testis-

determining factor [42, 43] that induces Sertoli cell differentiation

to then promote testis development and male sex determination.

Although SRY is known to act as Jost’s testis-determining factor

(TDF) and induce Sertoli differentiation, the downstream targets

have only recently been elucidated [9]. Because Sertoli cell

differentiation is essential for male gonadal sex determination

and adult male fertility, the downstream targets of SRY and the

cascade of molecular events involved are important to under-

stand. The current study was designed to further investigate the

SRY-induced molecular events that lead to Sertoli cell

differentiation and male gonadal sex determination.

The initial downstream target for SRY identified was Sox9

[5, 6]. SRY acts in concert with SF1 to regulate Sox9

expression [3]. Recently, we identified the neurotrophin 3

(Ntf3) growth factor as a downstream target of SRY [10]. A

genome-wide analysis of the SRY direct downstream binding

targets identified 71 different genes and a large number of

atypical binding targets not involving apparent direct DNA

binding [9]. The downstream binding targets of SOX9 were

found to be distinct from SRY, with negligible overlap [9].

Observations identified gene networks and cellular pathways

involved in the SRY-induced Sertoli cell differentiation and

testis determination.

One of the previously identified primary downstream SRY

targets was the bHLH protein TCF21 (bHLHa23) [11]. Tcf21

was also detected in the genome-wide analysis of SRY target

genes [9]. Tcf21 is a bHLH gene involved in the development

of a number of tissues, such as the kidney, lungs, and heart

[12–19, 21, 44]. The bHLH transcription factor family is

critical for cellular differentiation and tissue development for a

wide number of organ systems [12–18, 21, 44]. Although

previously shown to be primarily expressed at E13 in Sertoli

cells [11], potential Tcf21 expression in other cell types needs

to be considered as a limitation to any data interpretations.

Often, a cascade of bHLH genes orchestrate cellular differen-

tiation and development [45, 46]. An example involves the

myod and myogenin cascade for muscle development [12].

Tissue-specific bHLH factors exist for muscle (myod),

neurogenesis (Neurogenin) [47–49], and heart (Hand) [50].

In addition to the cell-specific focus of bHLH genes, more

widely expressed bHLH proteins integrate with other bHLH

proteins to promote tissue-specific development [12–19, 44,

45]. A phylogenetic analysis of the bHLH family of genes in

several different species identified groups of bHLH genes with

functional similarities [23]. The Tcf21 gene clusters with a

number of different genes (e.g., Musc, Tcf23, Hand, Atoh,

Twist, NeuroG, NeuroD, Lyl, Tal, etc.) and appears to be

highly conserved between species. Because the TCF21 bHLH

protein is involved in cellular differentiation and a direct

downstream target of SRY, the current study was designed to

investigate the downstream targets of TCF21, with a focus on

bHLH targets. An objective was to determine if a cascade of

bHLH genes has a role in the induction and promotion of

Sertoli cell differentiation.

A modified ChIP followed by a promoter tiling microarray

(Chip) was performed to identify the downstream targets of

TCF21. A ChIP-Chip involving a competitive hybridization of

the antibody to TCF21-ChIP compared to the nonimmune IgG-

ChIP was used to eliminate false-positive sites due to

nonspecific binding of IgG. The comparative hybridization in

the ChIP-Chip allows the direct comparison not possible in

next-generation sequencing, ChIP-Seq, that requires different

experiments to be compared. The hybridization profiles clearly

identified TCF21-specific binding sites (position peaks) (Fig. 2

and Supplemental Fig. S1). This novel competitive hybridiza-

tion clearly indicates nonspecific IgG binding is a critical issue

to address in ChIP-Chip analysis. A limitation to this analysis

is that nonspecific IgG-binding peaks can mask true-positive

TCF21 binding sites, but the elimination of the false-positive

binding sites is a significant advance to previous ChIP-Chip

analysis and is not possible in ChIP-Seq analysis. Therefore,

the number of TCF21 binding targets identified likely will be a

subset of a larger number of targets. In addition, a Chip was

used with approximately 4 kb of upstream promoter, such that

more distal binding sites would not be detected. Therefore, the

ChIP-Chip analysis used will identify targets within 4 kb of the

promoter without a strong nonspecific binding of IgG. This

FIG. 8. Schematic diagram of the hypothesized cascade of bHLH

transcription factors involved in Sertoli cell differentiation and gonadal sex

determination.

BASIC HLH CASCADE DURING MALE SEX DETERMINATION

11 Article 131

Downloaded from www.biolreprod.org.

TCF21 ChIP-Chip analysis used a stringent statistical cut off of

P , 1 3 10

7

, which may not have considered less significant

binding sites. Therefore, the analysis used likely is a subset of

TCF21 targets but includes the most highly significant

detectable binding targets.

The TCF21 ChIP-Chip analysis identified 121 direct

binding target genes for the fetal male gonad at the rat E13

stage of development. This is a period when SRY expression is

high and testis development has been initiated (corresponding

to the mouse E10.5–11.5 stage of development). Our previous

studies have demonstrated the expression of Tcf21 and Scx at

this stage of gonadal development in rat and mouse microarray

studies [24, 25] and protein expression studies [11]. A recent

study using a transgenic mouse line and fluorescent automated

cell sorting to sort gonadal cell populations and examine gene

expression with microarrays [51] did not detect Tcf21

expression at mouse E11.5 or Tcf21 or Scx at E12.5. This

information, which is in conflict with the previous literature

and the current study, remains to be elucidated, but may be an

element of the mouse-versus-rat model and of the analysis used

[51]. In the current study, the TCF21 target gene set was

analyzed with bioinformatics protocols to identify potential

common cellular pathways and processes. A number of major

pathways were identified, with the most predominant being the

olfactory transduction pathway, a G protein-coupled receptor

family of genes (Supplemental Fig. S2). The functional role of

this pathway in testis development is unclear, but evidence

suggests a nonolfactory role. A gene network analysis of the

TCF21 targets also identified a network of interconnected

genes (Fig. 5). The insulin signaling was central to this

network, with signal transduction and transcriptional regulation

components. These cellular pathways and gene networks now

can be further investigated to help elucidate potential roles in

Sertoli cell differentiation and testis development. The

combination of the previous gene networks and binding targets

for SRY and SOX9 [9] with the current TCF21 targets provide

a genome-wide set of molecular events in male gonadal sex

determination. A comparison of the SRY [9] and TCF21 target

genes indicated that 12 genes overlapped (i.e., Tmen95,

Afg312, Arfgef2, Gata1, Il1rep11, Olr853, Pdcd6ip, Phox2a,

RGD1562638, Rpl24, Rtf1, Sec1, and Timm86).

The analysis of TCF21 downstream targets identified 10

different bHLH genes. All these bHLH targets now need to be

considered in regards to Sertoli cell and/or testis development.

One gene that had a high level of fetal testis gene expression

[24] and that had previously been shown by our laboratory to

be associated with Sertoli cell differentiation was scleraxis

[26]. Scleraxis (Scx)(bHLHa41) is involved in connective

tissue development and cartilage [52, 53]. Scx clusters with

Tcf15 (bHLHa40) in a phylogenic manner [23] and appears to

have functions in a variety of tissues [54–57]. In adult Sertoli

cells, Scx is expressed during pubertal and adult periods and

can influence differential functions, such as transferrin

expression [26]. Because Scx is a direct target for TCF21, is

primarily localized in Sertoli cells at the E13 stage of

development (Fig. 6), and has a role in pubertal and adult

Sertoli cell differentiation [26], the functional role of Scx in a

cascade of events associated with SRY and TCF21 was

investigated.

Previously, a fetal gonadal cell culture system was

established to examine the ability of SRY and TCF21 to

promote an in vitro sex reversal and induction of Sertoli cell

differentiation in vitro [11]. In the current study, TCF21 was

found to induce Scx expression in the fetal ovarian cells,

providing support for its role as a direct target for TCF21. The

overexpression of SCX was found to induce the expression of

transferrin gene expression in the fetal ovarian cell culture,

suggesting an in vitro sex reversal and induction of mouse

adult Sertoli cell differentiation. Transferrin gene expression is

not detected in ovaries at this stage of development; it is only

expressed in pubertal and adult Sertoli cells in the testis [24–

26]. Interestingly, TCF21 alone did not have the ability to

consistently promote transferrin gene expression (Fig. 7B)

within the 4-day culture duration, but an extended 10-day

duration of TCF21 overexpression was found to more

consistently promote transferrin gene expression. This suggests

an intermediate, such as SCX, in the TCF21 induction of

transferrin expression. Therefore, the hypothesis is that a

sequential cascade of SRY acting on Tcf21, which then acts on

Scx, is part of a cascade that can induce Sertoli cell

differentiation through fetal development to a more adult

differentiated state (Fig. 8). The assumption is that the ultimate

molecular in vivo cascade of events involved in Sertoli cell

differentiation and testis development will be more complex

and involve many of the other gene networks and bHLH

factors identified. The current gonadal cell culture experiment

was designed to determine the potential that TCF21 and SCX

can promote, in a stem cell-like culture system, the differen-

tiation of Sertoli cells, not to suggest the in vivo timing or

developmental programming. A similar approach was recently

made to differentiate Sertoli cells in vitro from a stem cell-like

culture [58]. Observations demonstrate that TCF21 and SCX

can promote fetal gonadal ovarian cells to sex reverse and

induce Sertoli cells to become differentiated. Clearly, in vivo, a

more complex cascade of molecular events will be involved

over the developmental period of normal male testis develop-

ment. Future studies need to examine more thoroughly the

expression profiles and cascade of molecular events and to

utilize conditional knockouts for Tcf21 and Scx in Sertoli cells

to investigate this phenomenon further. The current mechanism

proposed provides a framework for these further investigations.

Observations support a role for bHLH in Sertoli cell

differentiation and provide further insights regarding the

molecular processes in male sex determination and testis

development.

ACKNOWLEDGMENT

We thank Ms. Tiffany Hylkema for assistance in PCR confirmation of

downstream targets, Dr. Marina Savenkova for gene family and network

data analysis, Dr. Mohan Manikkam and Ms. Rebecca Tracey for time-

pregnant animals, Dr. Eric Nilsson for critical review of the manuscript,

and Ms. Heather Johnson for assistance in preparation of the manuscript.

REFERENCES

1. DeFalco T, Capel B. Gonad morphogenesis in vertebrates: divergent

means to a convergent end. Annu Rev Cell Dev Biol 2009; 25:457–482.

2. Skinner MK. Regulation of primordial follicle assembly and development.

Hum Reprod Update 2005; 11:461–471.

3. Sekido R, Lovell-Badge R. Sex determination and SRY: down to a wink

and a nudge? Trends Genet 2009; 25:19–29.

4. Sekido R, Lovell-Badge R. Sex determination involves synergistic action

of SRY and SF1 on a specific Sox9 enhancer. Nature 2008; 453:930–934.

5. Morais da Silva S, Hacker A, Harley V, Goodfellow P, Swain A, Lovell-

Badge R. Sox9 expression during gonadal development implies a

conserved role for the gene in testis differentiation in mammals and birds.

Nat Genet 1996; 14:62–68.

6. Sekido R, Bar I, Narvaez V, Penny G, Lovell-Badge R. SOX9 is up-

regulated by the transient expression of SRY specifically in Sertoli cell

precursors. Dev Biol 2004; 274:271–279.

7. Kim Y, Bingham N, Sekido R, Parker KL, Lovell-Badge R, Capel B.

Fibroblast growth factor receptor 2 regulates proliferation and Sertoli

differentiation during male sex determination. Proc Natl Acad Sci U S A

2007; 104:16558–16563.

8. Wilhelm D, Hiramatsu R, Mizusaki H, Widjaja L, Combes AN, Kanai Y,

BHANDARI ET AL.

12 Article 131

Downloaded from www.biolreprod.org.

Koopman P. SOX9 regulates prostaglandin D synthase gene transcription

in vivo to ensure testis development. J Biol Chem 2007; 282:

10553–10560.

9. Bhandari R, Haque Md. M, Skinner M. Global genome analysis of the

downstream binding targets of testis-determining factor SRY and SOX9.

PLoS ONE 2012; 7:e43380.

10. Clement TM, Bhandari RK, Sadler-Riggleman I, Skinner MK. Sry directly

regulates the neurotrophin-3 promoter during male sex determination and

testis development in rats. Biol Reprod 2011; 85:227–284.

11. Bhandari R, Sadler-Riggleman I, Clement TM, Skinner M. Basic helix-

loop-helix transcription factor TCF21 is a downstream target of the male

sex-determining gene SRY. PLoS ONE 2011; 6:e19935.

12. Baylies MK, Bate M, Ruiz Gomez M. Myogenesis: a view from

Drosophila. Cell 1998; 93:921–927.

13. Ghysen A, Dambly-Chaudiere C, Jan LY, Jan YN. Cell interactions and

gene interactions in peripheral neurogenesis. Genes Dev 1993; 7:723–733.

14. Modolell J, Campuzano S. The achaete-scute complex as an integrating

device. Int J Dev Biol 1998; 42:275–282.

15. Goulding SE, zur Lage P, Jarman AP. amos, a proneural gene for

Drosophila olfactory sense organs that is regulated by lozenge. Neuron

2000; 25:69–78.

16. Huang ML, Hsu CH, Chien CT. The proneural gene amos promotes

multiple dendritic neuron formation in the Drosophila peripheral nervous

system. Neuron 2000; 25:57–67.

17. Jarman AP, Grau Y, Jan LY, Jan YN. atonal is a proneural gene that

directs chordotonal organ formation in the Drosophila peripheral nervous

system. Cell 1993; 73:1307–1321.

18. Ma Q, Kintner C, Anderson DJ. Identification of neurogenin, a vertebrate

neuronal determination gene. Cell 1996; 87:43–52.

19. Quaggin SE, Schwartz L, Cui S, Igarashi P, Deimling J, Post M, Rossant J.

The basic-helix-loop-helix protein pod1 is critically important for kidney

and lung organogenesis. Development 1999; 126:5771–5783.

20. Porcher C, Swat W, Rockwell K, Fujiwara Y, Alt FW, Orkin SH. The T

cell leukemia oncoprotein SCL/tal-1 is essential for development of all

hematopoietic lineages. Cell 1996; 86:47–57.

21. Srivastava D, Thomas T, Lin Q, Kirby ML, Brown D, Olson EN.

Regulation of cardiac mesodermal and neural crest development by the

bHLH transcription factor, dHAND. Nat Genet 1997; 16:154–160.

22. Norton JD, Deed RW, Craggs G, Sablitzky F. Id helix-loop-helix proteins

in cell growth and differentiation. Trends Cell Biol 1998; 8:58–65.

23. Skinner MK, Rawls A, Wilson-Rawls J, Roalson EH. Basic helix-loop-

helix transcription factor gene family phylogenetics and nomenclature.

Differentiation 2010; 80:1–8.

24. Clement TM, Anway MD, Uzumcu M, Skinner MK. Regulation of the

gonadal transcriptome during sex determination and testis morphogenesis:

comparative candidate genes. Reproduction 2007; 134:455–472.

25. Small CL, Shima JE, Uzumcu M, Skinner MK, Griswold MD. Profiling

gene expression during the differentiation and development of the murine

embryonic gonad. Biol Reprod 2005; 72:492–501.

26. Muir T, Sadler-Riggleman I, Skinner MK. Role of the basic helix-loop-

helix transcription factor, scleraxis, in the regulation of Sertoli cell

function and differentiation. Mol Endocrinol 2005; 19:2164–2174.

27. Levine E, Cupp AS, Miyashiro L, Skinner MK. Role of transforming

growth factor-alpha and the epidermal growth factor receptor in embryonic

rat testis development. Biol Reprod 2000; 62:477–490.

28. O’Neill LP, VerMilyea MD, Turner BM. Epigenetic characterization of

the early embryo with a chromatin immunoprecipitation protocol

applicable to small cell populations. Nat Genet 2006; 38:835–841.

29. Guerrero-Bosagna C, Settles M, Lucker B, Skinner M. Epigenetic

transgenerational actions of vinclozolin on promoter regions of the sperm

epigenome. PLoS ONE 2010; 5:e13100.

30. Smyth GK, Speed T. Normalization of cDNA microarray data. Methods

2003; 31:265–273.

31. Bolstad BM, Irizarry RA, Astrand M, Speed TP. A comparison of

normalization methods for high density oligonucleotide array data based

on variance and bias. Bioinformatics 2003; 19:185–193.

32. Smyth GK. Limma: linear models for microarray data. In: Gentleman R,

Carey VJ, Huber W, Irizarry RA, Dudoit S (eds.), Bioinformatics and

Computational Biology Solutions Using R and Bioconductor. New York:

Springer; 2005:397–420.

33. Tukey J. Exploratory Data Analysis. Reading, MA: Addison-Wesley;

1977.

34. Hardle W, Steiger W. Algorithm AS 296: optimal median smoothing. J R

Stat Soc Ser C Appl Stat 1995; 44(2):258–264.

35. Toedling J, Skylar O, Krueger T, Fischer JJ, Sperling S, Huber W.

Ringo—an R/Bioconductor package for analyzing ChIP-chip readouts.

BMC Bioinformatics 2007; 8:221.

36. Nilsson EE, Savenkova MI, Schindler R, Zhang B, Schadt EE, Skinner

MK. Gene bionetwork analysis of ovarian primordial follicle development.

PLoS ONE 2010; 5:e11637.

37. Liu Y, Ray SK, Yang XQ, Luntz-Leybman V, Chiu IM. A splice variant

of E2-2 basic helix-loop-helix protein represses the brain-specific

fibroblast growth factor 1 promoter through the binding to an imperfect

E-box. J Biol Chem 1998; 273:19269–19276.

38. Lei H, Fukushige T, Niu W, Sarov M, Reinke V, Krause M. A widespread

distribution of genomic CeMyoD binding sites revealed and cross

validated by ChIP-Chip and ChIP-Seq techniques. PLoS ONE 2010; 5:

e15898.

39. Skinner MK, Griswold MD. Sertoli cells synthesize and secrete

transferrin-like protein. J Biol Chem 1980; 255:9523–9525.

40. Chaudhary J, Skinner MK. E-box and cyclic adenosine monophosphate

response elements are both required for follicle-stimulating hormone-

induced transferrin promoter activation in Sertoli cells. Endocrinology

1999; 140:1262–1271.

41. Jost A, Price D, Edwards G. Hormonal factors in the sex differentiation of

the mammalian fetus. Phil Trans R Soc Lond B Biol Sci 1970; 259:

119–131.

42.

Gubbay J, Collignon J, Koopman P, Capel B, Economou A, Munsterberg

A, Vivian N, Goodfellow P, Lovell-Badge R. A gene mapping to the sex-

determining region of the mouse Y chromosome is a member of a novel

family of embryonically expressed genes. Nature 1990; 346:245–250.

43. Sinclair AH, Berta P, Palmer MS, Hawkins JR, Griffiths BL, Smith MJ,

Foster JW, Frischauf AM, Lovell-Badge R, Goodfellow PN. A gene from

the human sex-determining region encodes a protein with homology to a

conserved DNA-binding motif. Nature 1990; 346:240–244.

44. Murre C, McCaw PS, Vaessin H, Caudy M, Jan LY, Jan YN, Cabrera CV,

Buskin JN, Hauschka SD, Lassar AB, Weintraub H, Baltimore D.

Interactions between heterologous helix-loop-helix proteins generate

complexes that bind specifically to a common DNA sequence. Cell

1989; 58:537–544.

45. Henke RM, Savage TK, Meredith DM, Glasgow SM, Hori K, Dumas J,

MacDonald RJ, Johnson JE. Neurog2 is a direct downstream target of the

Ptf1a-Rbpj transcription complex in dorsal spinal cord. Development

2009; 136:2945–2954.

46. Jung P, Hermeking H. The c-MYC-AP4-p21 cascade. Cell Cycle 2009; 8:

982–989.

47. Brunet JF, Ghysen A. Deconstructing cell determination: proneural genes

and neuronal identity. Bioessays 1999; 21:313–318.

48. Verzi MP, Anderson JP, Dodou E, Kelly KK, Greene SB, North BJ,

Cripps RM, Black BL. N-twist, an evolutionarily conserved bHLH protein

expressed in the developing CNS, functions as a transcriptional inhibitor.

Dev Biol 2002; 249:174–190.

49. Powell LM, Jarman AP. Context dependence of proneural bHLH proteins.

Curr Opin Genet Dev 2008; 18:411–417.

50. Barnes RM, Firulli BA, VanDusen NJ, Morikawa Y, Conway SJ, Cserjesi

P, Vincentz JW, Firulli AB. Hand2 loss-of-function in Hand1-expressing

cells reveals distinct roles in epicardial and coronary vessel development.

Circ Res 2011; 108:940–949.

51. Jameson SA, Natarajan A, Cool J, DeFalco T, Maatouk DM, Mork L,

Munger SC, Capel B. Temporal transcriptional profiling of somatic and

germ cells reveals biased lineage priming of sexual fate in the fetal mouse

gonad. PLoS Genet 2012; 8:e1002575.

52. Cserjesi P, Brown D, Ligon KL, Lyons GE, Copeland NG, Gilbert DJ,

Jenkins NA, Olson EN. Scleraxis: a basic helix-loop-helix protein that

prefigures skeletal formation during mouse embryogenesis. Development

1995; 121:1099–1110.

53. Murchison ND, Price BA, Conner DA, Keene DR, Olson EN, Tabin CJ,

Schweitzer R. Regulation of tendon differentiation by scleraxis distin-

guishes force-transmitting tendons from muscle-anchoring tendons.

Development 2007; 134:2697–2708.

54. Alberton P, Popov C, Pragert M, Kohler J, Shukunami C, Schieker M,

Docheva D. Conversion of human bone marrow-derived mesenchymal

stem cells into tendon progenitor cells by ectopic expression of scleraxis.

Stem Cells Dev 2011; 21:846–858.

55. Wang L, Bresee CS, Jiang H, He W, Ren T, Schweitzer R, Brigande JV.

Scleraxis is required for differentiation of the stapedius and tensor tympani

tendons of the middle ear. J Assoc Res Otolaryngol 2011; 12:407–421.

56. Furumatsu T, Shukunami C, Amemiya-Kudo M, Shimano H, Ozaki T.

Scleraxis and E47 cooperatively regulate the Sox9-dependent transcrip-

tion. Int J Biochem Cell Biol 2010; 42:148–156.

BASIC HLH CASCADE DURING MALE SEX DETERMINATION

13 Article 131

Downloaded from www.biolreprod.org.

57. Levay AK, Peacock JD, Lu Y, Koch M, Hinton RB Jr, Kadler KE, Lincoln

J. Scleraxis is required for cell lineage differentiation and extracellular

matrix remodeling during murine heart valve formation in vivo. Circ Res

2008; 103:948–956.

58. Buganim Y, Itskovich E, Hu YC, Cheng AW, Ganz K, Sarkar S, Fu D,

Welstead GG, Page DC, Jaenisch R. Direct reprogramming of fibroblasts

into embryonic Sertoli-like cells by defined factors. Cell Stem Cell 2012;

11:373–386.

BHANDARI ET AL.

14 Article 131

Downloaded from www.biolreprod.org.