The splicing landscape is globally reprogrammed during male meiosis

The splicing landscape is globally reprogrammedduring male meiosisRalf Schmid1 Sushma Nagaraja Grellscheid23 Ingrid Ehrmann2 Caroline Dalgliesh2

Marina Danilenko2 Maria Paola Paronetto45 Simona Pedrotti56 David Grellscheid7

Richard J Dixon8 Claudio Sette56 Ian C Eperon1 and David J Elliott2

1Department of Biochemistry University of Leicester Leicester LE1 9HN UK 2Institute of Genetic MedicineNewcastle University Newcastle upon Tyne NE1 3BZ UK 3School of Biological and Biomedical SciencesDurham University Durham DH1 3LE UK 4Department of Health Sciences University of 00135 Rome lsquoForoItalicorsquo Rome Italy 5Laboratories of Neuroembryology and of Cellular and Molecular NeurobiologyFondazione Santa Lucia IRCCS 00143 Rome Italy 6Department of Public Health and Cell Biology Universityof Rome Tor Vergata 00133 Rome Italy 7Institute of Particle Physics Phenomenology Durham UniversityDurham DH1 3LE UK and 8Life Technologies Ltd Paisley PA4 9RF UK

Received January 23 2013 Revised August 15 2013 Accepted August 16 2013

ABSTRACT

Meiosis requires conserved transcriptional changesbut it is not known whether there is a correspondingset of RNA splicing switches Here we used RNAseqof mouse testis to identify changes associated withthe progression from mitotic spermatogonia tomeiotic spermatocytes We identified 150 splicingswitches most of which affect conserved protein-coding exons The expression of many key splicingregulators changed in the course of meiosis includingdownregulation of polypyrimidine tract bindingprotein (PTBP1) and heterogeneous nuclear RNP A1and upregulation of nPTB Tra2b muscleblind CELFproteins Sam68 and T-STAR The sequences near theregulated exons were significantly enriched in targetsites for PTB Tra2b and STAR proteins Reporterminigene experiments investigating representativeexons in transfected cells showed that PTB bindingsites were critical for splicing of a cassette exon in theRalgps2 mRNA and a shift in alternative 50 splice siteusage in the Bptf mRNA We speculate that nPTBmight functionally replace PTBP1 during meiosis forsome target exons with changes in the expression ofother splicing factors helping to establish meioticsplicing patterns Our data suggest that there are sub-stantial changes in the determinants and patterns ofalternative splicing in the mitotic-to-meiotic transitionof the germ cell cycle

INTRODUCTION

Most mammalian protein-coding genes are comprised ofmultiple exons and introns Introns are generally muchlonger than exons and are removed from the initial tran-script by pre-mRNA splicing In 90 of genes multiplemRNA isoforms are produced as a result of either theexistence of multiple splicing pathways (alternativesplicing) or the use of different promoters or terminationsites (1ndash3) Many alternative exons have been found (4)through sequencing of full-length mRNAs and expressedsequence tags Each human protein coding gene producesan average of 11 mRNA isoforms through alternativesplicing and recent estimates suggest there are gt82 000transcriptional initiation sites and 128 000 alternativepolyadenylation sites for 21 000 human protein codinggenes (5) As the use of many splice sites and alternativepromoters or polyadenylation sites is regulated inresponse to extracellular cues or during development al-ternative mRNA isoforms can determine the functions ofa gene in different circumstancesBecause splicing amplifies the functional content of the

genome there is currently great interest in how both RNAsplicing regulators and mRNA isoforms are modulated indevelopment (6) Extensive splicing switches have beenfound in the heart the immune system and brain (7ndash11)and some human diseases such as myotonic dystrophyare caused by defects in developmental splicing (12)Spermatogenesis is one of the most radical pathwaysof development still maintained in adult animalsSpermatogenesis involves alterations in both chromosome

To whom correspondence should be addressed Tel +44 191 241 8694 Fax +44 191 241 8666 Email DavidElliottnewcastleacukCorrespondence may also be addressed to Ian C Eperon Tel +44 116 229 7012 Fax +44 116 229 7018 Email ecileicesteracukCorrespondence may also be addressed to Ralf Schmid Tel +44 116 229 7023 Fax +44 116 229 7018 Email RSchmidleacukCorrespondence may also be addressed to Sushma N Grellscheid Tel +44 791 362 2690 Fax +44 191 334 1201 Email sushmacantabnet

The authors wish it to be known that in their opinion the first two authors should be regarded as joint First Authors

Nucleic Acids Research 2013 1ndash15doi101093nargkt811

The Author(s) 2013 Published by Oxford University PressThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby30) whichpermits unrestricted reuse distribution and reproduction in any medium provided the original work is properly cited

Nucleic Acids Research Advance Access published September 12 2013 at H

AM

-TM

C L

ibrary on September 26 2013

httpnaroxfordjournalsorgD

ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

at H

AM

-TM

C L



ownloaded from

number and cell morphology to convert a diploid stem cellin which chromatin is packaged with histones into a motilehaploid cell with a compact nucleus containing chromatinpackaged with protamines Exon-specific microarrayshave detected more alternative splicing in the wholeadult testis than in any other tissue except the brain(13) although at what stage in spermatogenesis thissplicing regulation originates is not known Perhaps themost important question concerning changes in alterna-tive splicing patterns during male germ cell development iswhether it is connected to meiosis Unlike cells in mitosisin which transcription is turned off meiotic cells arehighly transcriptionally active (14) In the single-celledyeast Saccharomyces cerevisiae meiosis is the only stagein the lifecycle to include alternative splicing In fact thetiming of the gene expression changes that drive meioticprogression in yeast is determined by a hierarchy ofmeiotic splicing events (1215)Adult male germ cell development takes around 30 days

in the testis (Figure 1A) Male mice are not born with afully developed male germ cell development pathway andinstead the first wave of spermatogenesis is initiated syn-chronously after birth (17) The testes of newborn micecontain germ cells arrested at the G0 and G1 stages ofthe cell cycle By 6 days post partum (dpp) these germcells have differentiated into spermatogonia a cell popu-lation that includes stem cells At 10 dpp somespermatogonia differentiate into meiotic spermatocytesMeiosis commences 12 dpp The first meiotic division(meiosis I) is complete by 21 dpp after which there is arapid second cell division (meiosis II) followed by progres-sive differentiation of the haploid cells into round sperm-atids elongating spermatids and finally mature sperm (thishaploid differentiation process is called spermiogenesis)Distinct programmes of transcriptional changes takeplace over animal meiosis which are critical for drivingdifferent molecular events such as the expression of genesencoding synaptonemal proteins like Sycp1 and the re-combinase Spo11 (18ndash20) It might be anticipated thatimportant splicing patterns represented only in testiswould be linked likewise to meiosis In the very laststages of male germ cell development nuclear DNA con-denses during the differentiation into elongating sperm-atids limiting transcriptional activity At this timetranslational control of pre-existing mRNAs transcribedearlier in meiosis plays a key role in gene expression (21)Much of the alternative splicing that has been detected

in whole human testis by exon-specific microarrays is notconserved in mice leads to the introduction of prematurestop codons and it occurs at individually low frequencies(13) These characteristics suggest that some alternativesplicing may represent lsquonoisersquo (22) arising from a lowerstringency of splicing control in the testis However thereare changes in the levels of expression of a number ofimportant RNA splicing regulators during spermatogen-esis including heterogeneous nuclear RNP (hnRNP)proteins (hnRNPA1 hnRNPG and PTB) (23ndash26) SR-like proteins (Tra2b) (27) and STAR proteins (Sam68and T-STAR) (2829) In this study we have set out toidentify whether there are high frequency changes inglobal splicing patterns that could affect protein isoform

production over meiosis and to comprehensively monitorthe expression of RNA splicing regulator proteins overthis timeframe Interestingly PTB is essential for malegerm cell development in fruitflies (3031) We speculatethat although polypyrimidine tract binding protein(PTBP1) is downregulated in meiosis nPTB might func-tionally replace it and thus ensure correct regulation ofPTB-dependent splicing events during germ cell differen-tiation Moreover other RNA splicing regulators alsochange in abundance during germ cell differentiation sug-gesting that they contribute to the meiotic patterns ofsplicing that we observe

MATERIALS AND METHODS

RNA sequencing

Sequencing reactions were done on pooled RNA samplesfrom 6 and 21 dpp mouse testis by Source BioscienceNottingham UK Four lanes of the flowcell were usedfor the sequencing of the 6 and 21 dpp samples on theGenome Analyzer II The Genome Analyzer (GA) wasrun for 38 cycles The images from the GA wereanalyzed with the GA pipeline software (v13 Illuminasoftware) on cycles 1ndash38 to undertake image analysisbase calling and sequence alignment to the referencegenome In all 5 345 040 and 5 561 352 quality filteredreads were obtained for the 6 dpp sample and 7 629 529and 7 610 503 were obtained for the 21 dpp sample Themouse NCBI Build 371 (mm9) was used as a referencegenome used for the read alignment Sequences werealigned with the ELAND software (lsquoELAND_rnarsquooption) which resulted in 3 922 430 4 085 374 5 515 372and 5 504 414 aligned reads for the 6 and 21 dppsamples respectively The aligned reads were used asinput for the Illumina CASAVA program (v10) tocount the sequence reads that align to genes exons andsplice junctions of the reference genome The mouse ref-erence feature files were supplied by Illumina and werederived from the mouse NCBI Build 371 The rawcounts of sequences aligning to features (gene exons andsplice junctions) were normalized by CASAVA bydividing the raw count by the length of the relevantfeature

The read counts per gene were used as input forDEGseq (32) and DESeq (33) to identify differentiallyexpressed genes Both tools are available via the statisticspackage R and Bioconductor DEGseq and DESeq usedifferent statistical approaches (Poisson distributionnegative binomial distribution) to estimate probabilitiesfor differential gene expression A P 0001 and a 2-foldchange (normalized) in expression levels were used as cut-off criteria Using these cut-offs DESeq identified 5835genes as differentially expressed whereas DEGseq found6362 differentially expressed genes The common set of5296 genes was taken as comprising the differentially ex-pressed genes for further analysis The resulting list wasread into the GOseq (34) BioconductorR-package toidentify GO terms that are over- or under-representedGOseq corrects for length bias in the detection of differ-ential expression in RNAseq The relationship between

2 Nucleic Acids Research 2013

Exonskipping

Alternative 3rsquo splice sites


Mutually exclusiveexons

50

100

110

Nu

mb

ers

of a

lter

nat

ive

even

ts

Unannotated exonsUTR only exonsCoding exons

Type of alternative event

Conserved events in human

Mouse meioticallyregulated events

Coding exon regulated in mouse meiosis

A

540nt141nt

Day

6 R

NA

Adul

t RN

A

Bptf gene post-meiosis Ymax=26

Bptf gene pre-meiosis Ymax=18

1 2 3

RNAseqreads

mRNAisoforms

Alternative splice site regulated in mouse meiosis

Direction of transcription

126nt248nt

Day

6 R

NA

Day

21

RNA

Vapa gene postmeiotic testis Ymax=108

Vapa gene premeiotic testis Ymax=86


RNAseqreads

mRNAisoforms

1 2

Unannotated meiotically regulated exonE

Day

6 R

NA

Day

21

RNA

Adul

t RN

A

187nt109nt

Ralgps2 gene post-meiotic testis day21 Ymax=34

Alternative exonConstitutive exon

Constitutive exon

Ralgps2 gene pre-meiotic testis day6 Ymax=7


1 2 3

RNAseqreads

mRNAisoforms

B

C

Elongatingspermatids(Spd)

Spermatogonia (Spg)

Meiotic cells (Spc)

Post-meiotichaploid cells (Rtd)

Days post partum

Cell types in testis

Sertoli cells (SC)

0

10

21

31

D

RNAseq reads

RNAseq reads

RNAseq reads

RT-PCR validation

RT-PCR validation

RT-PCR validation

Figure 1 High frequency switches in mRNA isoforms take place between the mouse pre-meiotic and meiotic testis transcriptomes (A) Cartoonshowing major cell types in the mouse testis and the timing of their appearance in the postnatal mouse (B) Summary of the patterns of alternativesplicing found by comparative RNAseq analysis to change between 6 and 21 dpp in the mouse testis transcriptome (this table summarizes infor-mation in Supplementary File S1) (C) Example of a meiotically regulated alternative cassette exon in the Ralgps2 gene (D) Example of a meioticallyregulated alternative splice site in the Bptf gene (E) Example of a previously unannotated exon discovered in the mouse Vapa4 gene which isregulated during meiosis In parts (C and D) the patterns of exon inclusion were monitored during meiosis by direct visualization of RNAseq readson the Savant genome browser (16) indicating the maximum peak (Ymax) of reads at 6 and 21 dpp (left hand panels for each gene) and by RT-PCRusing primers in flanking exons (Supplementary Table S1) followed by agarose gel electrophoresis (right hand panels for each gene)

Nucleic Acids Research 2013 3

gene expression for every gene in our data set before andafter meiosis (6 and 21 dpp respectively) was representedusing scatter plots prepared using an in-house Pythonscript Read counts per gene were used as an input andwere derived from CASAVAThe MISO pipeline (35) was used to identify differential

alternative splicing across the 6 and 21 dpp samplesBriefly MISO requires a library file of annotated alterna-tive events and alignment files for the two stages as inputThe mm9 alternative event annotation file (36) as providedwith the MISO software was used as a library file For theevents defined in the library file MISO measures for dif-ferential expression using Bayesian inference To generateMISO-compatible alignment files the quality filteredreads for the two stages were re-aligned against the mm9mouse reference genome with Tophat (37) using theIllumina mm9 genome feature file to improve the detec-tion of splicing junctions The Fastmiso version of theMISO package was run with default settings A combin-ation of different cut-offs and filters was tested in theanalysis of the MISO output culminating in the use of aBayes factor of 10 as cut-off value to detect differentialalternative splice events RNAseq reads were visualized onthe mouse genome using the Savant genome browser (16)

Analysis of enriched sequences associated with meioticsplicing regulation

K-mer analysis was carried out using custom scripts asdescribed previously and the total set of cassette exonspredicted as meiosis-regulated by MISO (3536) Wechose a bayes factor value of gt10 from the MISOresults as a cut-off to define exons that were alternativelyspliced in meiosis and exons over 500 bases (9 exons) wereremoved to yield a total of 251 exons that are alterna-tively spliced in meiosis The background data set wasdefined as the set of exons smaller than 500 bp and witha bayes factor value below 01 indicating that althoughthey were expressed in our data set they were not alter-natively spliced (276 exons) We analysed alternativeexons and 250 bases of flanking introns with correspond-ing background data sets Activated (159 exons) andrepressed (92 exons) data sets predicted by MISO wereanalysed separately with the same background data setto identify enriched 5-mers that were over-represented inmeiotic regulated exons The 5-mer counts werenormalized to the corresponding data set size (frequency)as well against the background K-mers were ranked inorder of the highest difference to the background andsignificance was measured using a t-test The completek-mer list with counts for all possible 5-mers is presentedin the Supplementary Data Potential binding sites forPTB were analysed as described (38) with the spacingrelaxed to YCUN(1ndash6)CUN(1ndash8)YCU where N is any nu-cleotide For analysis at lower stringency a match wasonly required at 7 of the 8 nt specified inYCUN(1ndash6)CUN(3ndash8)YCU

Amplification of different mRNA isoforms

Candidate meiotically regulated splice isoforms werecharacterized by RT-PCR using the primer sequences

given in Supplementary File S1 followed by eitheragarose gel electrophoresis or capillary gel electrophor-esis for quantitation Percentage Splicing Inclusion(PSI) values were calculated as the concentrationof isoform including alternative event(concentration ofisoform including alternative event+concentration ofisoform excluding alternative event) 100 Heat mapswere drawn using httpwwwhivlanlgovcontentsequenceHEATMAPheatmaphtml

Cell isolation

Spermatogonia were obtained from 7dpp CD1 mice(Charles River Italy) as previously described (39)Sertoli cells were prepared from 7 and 17 dpp CD1 miceas previously described (40) Testes from 28ndash30 dpp CD1mice were used to obtain pachytene spermatocytes andround spermatids by elutriation (41) Purified germ cellswere collected washed with phosphate-buffered saline(PBS) and used for RNA and protein extraction Toanalyse the timing of splicing events in meiosis RNAsamples were analysed from 13 dpp testis (latest stageearly meiosis) 16 dpp testis (latest stage early pachytene)18 dpp (latest stage late pachytene and meiotic divisions)and day 21 (meiosis complete)

RNA and protein extraction

Purification of RNA from tissuesTotal RNA from whole postpartum testes or adult

mouse tissues was isolated using TRIZOL (Invitrogen)Poly A+ RNA was purified using a Dynabeads mRNApurification kit (Invitrogen) Parallel samples were fixedusing Bouinrsquos and mounted in paraffin wax followed byHampE staining using standard procedures as previouslydescribed (42) Total RNA from isolated germ cells orSertoli cells was prepared using TRIZOL (Invitrogen) ac-cording to the manufacturerrsquos instructions DNase diges-tion was performed using RQ1 RNase free DNase(Promega) at 37C for 20min One microgram of RNAwas used for RT-PCR with the Superscript III reversetranscriptase (Invitrogen) according to manufacturerrsquos in-structions A total of 5 of the RT reaction was used astemplate for the PCR reaction Oligonucleotides used asPCR primers are listed in the Supplementary File S1

For protein extraction cells were washed in ice-coldPBS homogenized and lysed in lysis buffer (50mMHepes (pH 74) 150mM NaCl 15mM MgCl2 15mMEGTA 1 Triton X-100 10 glycerol 20mMb-glycerophosphate 1mM DTT 05 mM Na3VO4) andprotease inhibitors (Sigma Aldrich) After 10min on icecell lysates were centrifuged at 10 000 g for 10min at 4CCell extracts were diluted in Laemmli sample buffer andboiled for 5min

Western blot analysis

Proteins were separated on 10 SDSndashpolyacrylamidegels and transferred to polyvinylidene fluorideImmobilon-P membranes (GE-Healthcare) using a wetblotting apparatus (Bio-Rad) Membranes were saturatedwith 5 BSA at room temperature and incubated with thefollowing primary antibodies (11000 dilution) at 4C


overnight a-nPTB mouse a-hnRNP A1 a-hnRNPA2B1 a- hnRNP C1C2 a-SC35 (Sigma Aldrich)mouse a- hnRNP FH (Abcam) rabbit a-SRp55 a-SRp20 a-SRp40 a-ERK2 and goat a-hnRNP I (SantaCruz Biotechnology) mouse a-ASFSF2 (USBiological) Secondary anti-mouse anti-goat or anti-rabbit IgGs conjugated to horseradish peroxidase(Amersham) were incubated with the membranes for 1 hat room temperature at a 110000 dilutionImmunostained bands were detected by a chemilumines-cent method (Santa Cruz Biotechnology)

Minigene analysis

Minigenes were cloned into pXJ41 using the primers inSupplementary File S1 and mutagenesis was carried outby overlap PCR as previously described (27)

RESULTS

High frequency switches in mRNA isoforms take placebetween the mouse pre-meiotic and meiotic testistranscriptomes

Previous transcriptome-wide analyses of gene expressionchanges in meiosis have detected only a single expressionsignal per gene and so have been unable to detect changesin mRNA isoforms (18ndash20) To comprehensively profilegene expression changes taking place during meiosis weinitially took advantage of the synchronous onset ofmeiosis in the testes of new-born mouse to separate geneexpression changes in meiosis from those associated withthe later processes of morphological differentiation (17)Testes were dissected from mice before (6 dpp) and at theend of meiosis (21 dpp) (Figure 1A) PolyA+ RNA wasisolated from testes at both ages and then analysed bydeep sequencing (RNAseq)

We analysed this RNAseq data (35) to identify a pool ofalternative splicing changes that occur between the 6 and21dpp testis transcriptomes From the total alternativeevents predicted by the MISO programme we selected104 exon skipping events 11 alternative 50 splice sites 28alternative 30 splice sites and 5 mutually exclusive exons byvisual inspection (Figure 1B and Supplementary File S2)We experimentally confirmed 15 of 20 tested events fromthese regulated events using RT-PCR analysis a validationrate of 75 (eg Figure 1CndashE right panels andSupplementary File S2) Although we detected alternativesplicing of some 50 UTR and poison exons most detectedalternative splicing events regulated in meiosis introducedexon sequences that comprised integer multiples of threenucleotides and were protein coding (Figure 1B) Suchevents included meiotic inclusion of a cassette exon withinthe Ralgps2 mRNA which encodes a ras-specific guaninenucleotide-releasing factor and an alternative 50 splice sitein the Bptf mRNA which encodes a bromodomain PHDtranscription factor (Figure 1C and D respectively)

Several of the exons regulated during postnatal mousetestis development were also annotated as alternativeevents in the human genome including Ralgps2 and Bptf(43) (Figure 1B) RNAseq analysis also predicted meioticsplicing regulation of a number of exons currently

unannotated on the mouse genome browser includingone in the mouse Vapa4 mRNA which we confirmed ex-perimentally (Figure 3D) Some of these currentlyunannotated exons (including that in Vapa4) mapped toregions of chromosome conservation between species andwere already annotated as either alternative or constitutiveexons in the human genome (Supplementary File S2)

Regulated splicing events take place betweenspermatogonia and spermatocytes

We confirmed the cell type-specificity of the observedsplicing changes using RT-PCR analysis of RNApurified from cell types in the adult testis (Figure 2AndashCleft hand panels) In 13 of 14 alternative splices tested inthis way splicing changed between spermatogonia andspermatocytes confirming their splicing was regulatedduring meiosis (Figure 2 and Supplementary File S2)Confirmed meiotic splicing changes included activationof the Ralgps2 cassette exon and the downstream splicesite in Bptf (Figure 2B and C left hand panels) We alsoobserved a switch to complete repression of the cassetteexons in the Odf2 and Ezh2 mRNA (Figure 2A left handpanel)Analysis of purified cell types indicated that for some

exons splicing regulation also occurs in Sertoli cellsGenerally developmental splicing switches in Sertolicells occurred at a lower frequency than those observedin meiotic cells An exception was for alternative splicingregulation of the Lrrc16a mRNA which encodes aleucine-rich protein (Figure 2C left hand panel)Lrrc16a showed a similar switch in mRNA spliceisoforms between spermatogonia and spermatocytes andbetween 7 and 17 dpp Sertoli cells Although most splicingisoform switches established in meiosis were maintained inround spermatids Lrrc16a again was an exceptionLrrc16a mainly produced the exon-skipped mRNAsplice isoform in spermatogonia and spermatids and theexon-included isoform in meiosisAlthough the aforementioned experiments analysed the

profile of mRNA splice isoform switches which take placebetween pre-meiotic and meiotic cells meiosis itself takesplace over 12 days in the mouse To monitor more pre-cisely the timing of splicing regulation during mousemeiosis we analysed splicing patterns of this samepanel of exons during the first wave of spermatogenesis(Figure 2AndashC right hand panels) Meiotic switches inmany mRNA isoforms (including Odf2 Ezh2 Add3)started early in meiosis (by day 13 which is 1 day aftermeiosis initiates in male mice) Later events includedRapgef1 (13 dpp) and Vapa4 (16 dpp) (Figure 2C righthand panel) Consistent with the results from purifiedcell types the splicing pattern of Lrrca16a switched backto mainly the exon-skipped form in adult testis

Most meiotically enriched splice isoforms aretestis-enriched rather than meiosis-specific

The aforementioned analyses indicate the existence of apool of meiotic splicing switches These events mightoccur only in the testis during and after meiosis or theymight occur elsewhere in the body in response to different


regulatory signals To test this we purified RNA fromother mouse tissues and analysed splicing patterns usingRT-PCRWhen splicing inclusion levels were analysed in different

tissues of the adult mouse (horizontal clustering in

Figure 3) the testis formed an outlier group for bothmeiosis-activated and meiosis-repressed exons indicatingthat meiosis-regulated splicing events are differentiallyregulated in the mouse testis compared with othertissues Complete exclusion of both the Odf2 and Ezh2

Ezh2

Odf2

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

Stage in meiosis

1 2 3 4

Odf2

Ezh2SC

7dp

pSC

17d

ppSp

gSp

cRt

d

1 2 3 4 5

A

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

1 2 3 4 5

C

Total testis RNA

Nasp

Picalm

Bptf

Add3

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

Nasp

Bptf

Picalm

Add3

1 2 3 4 5

P

B

Total testis RNA

Total testis RNA

RT-PCR analysis in purified cell types

RT-PCR analysis in whole testis

Stage in meiosis1 Earlymeiosis

2 Earlypachytene

3 Late pachyteneand meiotic divisions

4 Meiosis complete

Figure 2 Splicing events that change between spermatogonia and spermatocytes (A) Cassette exons in the Ezh2 and Odf2 genes are repressed duringmeiosis (B) An downstream 50 site in the Bptf gene and cassette exons in the Picalm Add3 and Nasp genes are activated during meiosis (C) Cassetteexons in the Ralgps2 Rapgef1 Vapa4 Lrrc16a and Nxt1 genes are activated during meiosis Left hand panels Levels of the different mRNAisoforms were detected by RT-PCR in RNA from purified cell types using primers in flanking exons (Supplementary Table S1) followed by agarosegel electrophoresis The different kinds of splicing event are annotated as in Figure 1 with protein coding events in red UTR exons in blue andpreviously unannotated events in grey Right hand panels levels of PSI in the testis at different days after birth (the first wave of meiosis ishighlighted in red) SC sertoli cells (isolated at 7 and 17 dpp) Spg spermatogonia Spc primary spermatocytes Rtd round spermatids


meiosis-repressed exons was only found in the testis andsplicing inclusion of the cassette exon in Vapa4 was onlyobserved in the testis However most meioticallyregulated exons in mouse testis were included to someextent in other mouse tissues as well For example theNasp-T exon is spliced into mRNAs in the mouse heartand the Add3 cassette exon is included at high levels in themouse gut and kidney

We also used the RNAseq data to compare overall geneexpression patterns of genes with activated and repressedcassette exons between the 6 and 21 dpp testis Many geneswith meiotically regulated cassette exons also increased inoverall gene expression between the 6 and 21 dpp testistranscriptomes (Supplementary Figure S1A and B andSupplementary File S3) For the Nasp and Odf2 genes(which have known important roles in germ cell develop-ment see lsquoDiscussionrsquo section) we also found that thatdistinct transcriptional initiation sites were used inmeiosis (indicated by red arrows in SupplementaryFigure S1C and D) To validate these gene expressionpatterns inferred from the RNAseq data set weanalysed the patterns of expression of genes alreadyknown to be regulated over meiosis (SupplementaryFigure S2 and Supplementary File S4) Genes known tobe involved in the mouse meiotic gene expression pro-grammes (18) were more highly expressed in the 21 dpptestis including Ccna1 Aurkc Spdy1 Acrbp Adam2Adam18 Pla2g6 Ribc2 Tcfl5 Ppp3r2 Smcp and Spag6In contrast known members of the core mitotic

programme (Gata4 Dmrt1 Osr2 Pcdh18 and Abca1)were more highly expressed in the 6 dpp testis than the21 dpp testis (18)

Comprehensive analysis of splicing factor geneexpression show global changes in the meioticsplicing regulator landscape

RNA splicing regulation is under combinatorial controlwith an important role for RNA-binding protein expres-sion (4445) To comprehensively analyse changes in thesplicing landscape in meiosis we monitored the expressionof all known RNA splicing regulators between the 6 and21 dpp testis (Figure 4 Supplementary Files S5 and S6)Identified changes in expression included the 2-folddownregulation of Ptbp1 (encoding PTBP1 protein)whereas Ptbp2 (encoding nPTB protein) was upregulated5-fold with a similar isoform switch at the protein level(Figure 4A and B) Interestingly transcription of Raver2which encodes a protein that interacts with PTB (46) wasalso significantly downregulated in the 21 dpp testis tran-scriptome consistent with a coordinate modulation ofPTB activity in meiotic cellsAmongst the other genes encoding hnRNP proteins we

observed an isoform switch between expression of the Xchromosome-encoded Rbmx gene before meiosis to theautosomal retrogene Rbmxl2 during and after meiosis(Figure 4A) (2647) RNAseq analysis also detected adecrease in expression of Hnrnpa1 mRNA between 6and 21 dpp HnRNP A1 protein is already known to beexpressed only in spermatogonia and Sertoli cells (23)Western blotting showed an even more dramaticdecrease in protein expression levels in purified celltypes with the corresponding hnRNP A1 protein virtuallydisappearing in purified meiotic cells (Figure 4B) Otherdetected meiotic changes in the expression of RNAsplicing regulators included activation of each of thegenes encoding CUG-binding proteins Celf4-6 mRNAswere upregulated over 2-fold during meiosis (Figure 4Aand Supplementary Files S5 and S6) and there was alsoan almost 2-fold upregulation of the Cugbp1 (Celf1) andCugbp2 (Celf2) genes (Supplementary Files S5 and S6)The expression levels of both Mbnl1 and Mbnl2encoding muscleblind proteins [Mbnl1 interacts withPTB (48)] were downregulated over meiosis(Supplementary Files S5 and S6) Not all changes inmRNA levels resulted in changes in protein expressionAlthough RNAseq indicated increased or decreased ex-pression of the various Hnrnph genes at the transcriptlevel (Supplementary Files S5 and S6) no overall changein expression of the family was seen at the protein level(Figure 4B)The expression of Tra2b mRNA (which encodes the

SR-like protein Tra2b) was upregulated almost 2-foldduring meiosis (Supplementary Files S5 and S6) Incontrast the expression levels of the classical SRproteins ASFSF2 (SRSF1) SC35 (SRSF2) SRp40(SRSF5) and SRp20 (SRSF3) remained similar at bothRNA and protein levels between the pre-meioticand meiotic testis (Figure 4C and Supplementary FilesS5 and S6) but subsequently there was a dramatic

Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp

0 100percent spliced in (psiΨ)

Exons skipped in meiosis

Exons activatedin meiosis

Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts

Splicing pattern clustered by tissue

Figure 3 Most meiotically enriched splice isoforms are testis-enrichedrather than meiosis-specific Heat map showing PSI levels of each of themeiotically regulated exons in different mouse tissues PSI levels areclustered according to tissue (horizontal axis) and splicing pattern(vertical axis) Patterns of expression are organized so that the exonsshowing the highest levels of inclusion in the testis are seen at the topof the vertical axis PSI levels were measured using RT-PCR analysisusing RNA purified from different mouse tissues using the primers inSupplementary File S1


loss of expression of ASFSF2 SRp20 and SRp40during the haploid stages of differentiation Strong in-creases in expression during meiosis (11- and 25-foldrespectively) were observed for the Sfrs14 mRNA(also known as Sugp2) which encodes a relativelyuncharacterized SR protein and for Sfrs15 whichencodes an SR-like protein (Sca4) that couples tran-scription and RNA splicing Expression levels from theSrpk1 and Srpk2 genes which encode serine kinasesthat phosphorylate SR proteins (and also protamines)(49) also increased between the 6 and 21 dpp testistranscriptomes

Specific RNA sequences are associated with meioticallyregulated exons in the mouse

To unravel the potential roles of changes in RNA proteingene expression in coordinating changes in meioticsplicing profiles we identified 5mer motifs that were sig-nificantly enriched in and around the meiotically regulatedcassette exons (Figure 5 and Supplementary Table S1 Thestatistical significance of enriched 5mers is included inSupplementary Table S1) Identified motifs includedknown binding sites for PTB (5051) PTB binding siteswere enriched downstream both of activated and repressed

SRp55SRp40

ASFSF2SC35

SRp20

ERK2

purified germcell extracts

SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB

hnRNP FhnRNP HhnRNP C1hnRNP C2

hnRNP A1

hnRNP B1hnRNP A2hnRNP B0

ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3

Genes for RNA splicing regulators activatedin meiosis

Jmjd6cwc15Hnrnph3

Gemin5

Genes for RNA splicing regulators repressedin meiosis

A

B C

Figure 4 Comprehensive analysis of splicing factor gene expression showing changes in the meiotic splicing regulator landscape (A) Scatterplotshowing expression levels of genes encoding known RNA splicing regulators (shown as green dots) that change expression gt2-fold (broken diagonalline) between the 6 and 21 dpp testis transcriptomes A full alphabetical list showing changes in RNA splicing factor expression between the 6 and21 dpp testis of all known RNA splicing regulators is given in Supplementary File S3 (B) Western blot analysis of hnRNP proteins in extracts madefrom cell types purified from the adult mouse testis Spg spermatogonia I Spc primary spermatocytes II SpcSpd secondary spermatocytes andelongated spermatids Rtd round spermatids (C) Western blot analysis of SR proteins in extracts made from cell types purified from the adult testisThe asterisk indicates a non-specific band detected by the a-SC35 antibody


exons similar to the pattern observed downstream ofexons positively and negatively regulated in muscle cells(652) Binding motifs for PTB upstream of or within anexon are associated with repression by PTB whereasdownstream motifs or motifs close to the splice sites ofthe adjacent constitutive exon are associated with activa-tion (5354) Intriguingly an analysis of the potentialbinding sites for PTB (38) around the regulated exon inRalgps2 suggested that the highest affinity binding siteswere downstream of the exon (Figure 6AndashC) eventhough it was activated during meiosis when PTBP1levels fell (Figure 2)

As germ cells are difficult to transfect in vitro we testedwhether the expression of this Ralgps2 exon might beregulated by PTB using a cell line model We cloned theregulated exon and its flanking intron sequences into anexon trap vector Co-transfection of this Ralgps2 minigeneinto cells with GFP resulted in production of mainlythe exon skipped isoform (Figure 6D lane 1) Howeverco-transfection of either PTBP1 or nPTB with theminigene dramatically increased splicing inclusion of the

meiosis-regulated Ralgps2 exon (Figure 6D compare lane1 with lanes 4 and 5) as would be expected if PTB boundto the downstream sitesOur analysis of PTB-binding possibilities which is

based on the sequence preferences of the RNA-bindingdomains inter-domain spacing and the number ofpossible arrangements of binding (38) identified tworegions downstream of the Ralgps2 exon to which PTBmight bind of these the one to the 30 side appeared tobe much more favourable (Figure 6AndashC) To test theindividual functions of these sites they were mutatedby converting cytosines in the core CT-rich motif intoadenosines (the sequences mutated are underlined inFigure 6C) Mutation of the lower affinity site did notblock splicing activation by PTBP1 (lanes 1ndash3 in Figure6E) but interestingly it did prevent splicing activation bythe nPTB protein suggesting a slightly different require-ment for splicing regulation of this exon by these twohighly homologous RBPs On the other hand mutationof the higher affinity site prevented splicing activation onco-expression with either PTBP1 or nPTB (compare

Upstream intron

Downstream intronRegulated

cassette exonUpstream exon downstream exon

Upstream intronCAUUU (PTB)CCCCC AAUAU (STAR family) ACAAU AAUAC UCAUU ACAGU CCAUA GAAUA (TRA2B) AUCCC UUUAC AAACC CCCUC (PTB)UGCUC AUAUA UUUUU AAUCG

ExonGAAGU (TRA2B) GGGAA (hnRNPH)AUAUG UAAAU (STAR family) UACAU GUAAA (STAR family) UUAAA (STAR family) GAGAC CAGGG (hnRNPH)ACUAA UAGAU AAGUA AACCA

Downstream IntronUUUUU (Sam68 PTB)UAUUA (STAR family)AAAGU AUUUU (STAR family) CUAUU AUAUU (STAR family) UAUUG AGUCA AUGAU UCUAU (PTB)GAAGU GCACU UAAAA UGAAA UUUGU UUUAA ACAUC UUCUU (PTB)

Downstream intron CCUCC (PTB)CUCCC (PTB)CCCUC (PTB)CUCCU (PTB)UCCUC (PTB)UCCCU (PTB)UCUCC (PTB)AGCAG CCUCU (PTB)CUCUA (PTB)UUUUU UUUCU (PTB)UGGAA (hnRNPA1)

ExonCGCGC rich (MBNL1 RBM4) UUAGG (hnRNPA1)UUUAG (hnRNPA1)UUCUU (PTB)

Upstream intronCCCCC UAUUC UUUAU AUUCA CAUCU CCAUC AGUCA AUAAA CCCUC AAUCG

Sequences associatedwith meiotically activatedexons

Sequences associatedwith meiotically repressedexons

Figure 5 Specific RNA sequences are associated with meiotically regulated exons in the mouse Frequently occurring 5mers found in and aroundmeiotically regulated exons are shown In some cases the RNA binding proteins that might interact with these motifs are indicated Full details ofidentified 5mers and their statistical significance are given in Supplementary Table S1


lanes 4ndash6 in Figure 6E) We conclude that the exon inRalgps2 that is activated in meiosis can be regulated byboth PTBP1 and nPTB both of which act via down-stream binding sites to cause inclusion Although theaforementioned data comes from a reconstituted cellline system it is suggestive for a potential role for PTBin regulating this Ralgps2 exon in mouse germ cells Wespeculate further that the general enrichment of pyrimi-dine-rich sequences around the regulated exons is con-sistent with roles for PTBP1 and nPTB in the regulationof splicing in meiosis

Other statistically significant motifs shown in Figure 5associated with inclusion are (G+A)-rich sequenceswithin the exon and UAAAA and similar motifs to thedownstream side These motifs are likely to includebinding sites for Tra2b (GAA core site) (5556) and forSam68 (5758) and T-STAR (5960) which are each highlyexpressed in testis and upregulated in meiosis (Figure 4and Supplementary Files S5 and S6) As both Tra2b andKhdrbs1 gene expression changes just lt2-fold overmeiosis they are not annotated on Figure 4 although apredicted binding site for Tra2b was the most significant

0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5

wild type Ralgps2 minigene

Ralgps2splice pattern

p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB

Mutation 1(low affinity site)

Mutation 2(high affinity site)


ggaatccaacagGAAGAACAGATTATACCATTCTCTCGGCCCGGTGACAAGAGTGCCGCGAAGAAATGGCTATCGAAGCCACACGAAGAAGGCCAGCAGgtacaatcccctgcatcaggggccatagaactcccttctggtgttggtggctggctcatatgggtgtggtctgacatttttttctcttcttcctcagctaattggtttta

D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40

Figure 6 Dissection of the meiotically regulated Ralgps2 cassette exon The Ralgps2 exon and its flanking intron sequences were screened for (A)high affinity and (B) lower affinity PTB binding sites (the position of the regulated exon is shown on the x-axis as a red rectangle) (C) Sequence ofthe meiotically regulated Ralgps2 exon (upper case) and its flanking intron sequence (lower case) Candidate binding sites for Tra2b in the regulatedexons are shown in bold The intronic PTB binding sites are shown in green (low affinity site with core motif underlined) and red (high affinity sitewith core motif underlined) (D) Splicing pattern of transcripts made from a Ralgps2 minigene in HEK293 cells after co-transfection of expressionvectors for different proteins (E) Affect on splicing pattern of transcripts made from the Ralgps2 minigene after mutation of the low or high affinityPTB binding sites In parts (D) and (E) the top panel shows a capillary gel electrophoresis analysis from a single experiment and the bottom panel isa bar chart representing data from three biological replicates


of the recovered 5mers shown in Supplementary Table S1We tested whether the GAA motif might indicate regula-tion by Tra2 b using the meiotically regulated exon ofRalgps2 which has been shown by CLIP to bind Tra2 b[(27) and data not shown] and contains three GAA motifs(Figure 6C) Co-transfection of a Ralgps2 minigene withTra2b caused a small but statistically significant increasein inclusion of the Ralgps2 cassette exon whereasT-STAR had no effect (Figure 6D lanes 1ndash3)

We also investigated whether modulations in PTB con-centration might regulate other types of high amplitudesplicing events which change over male meiosisCandidate PTB binding sites (38) were also identifiedjust downstream of the upstream meiosis-regulated 50

splice site in the Bptf gene (Figure 7A) To enable us totest the function of these PTB binding sites on selection ofthe upstream and downstream Bptf 50 splice sites wecloned a minigene containing the meiosis-regulated Bptfexon with both available 50 splice sites between b globinexons When this Bptf minigene was co-transfected inHEK293 cells with GFP we observed mainly use ofthe upstream 50 splice site (Figure 7B lane 1 This is thesplicing pattern seen in the mitotically active cells of thetestis) In contrast co-transfection with PTBP1 (but notnPTB) strongly activated use of the downstream 50 splicesite (Figure 7B lanes 2 and 3 This is the splicing patternseen in post-meiotic cells in the testis) Splicing control ofBptf 50splice site selection was specific to PTBP1 in theseexperiments and no effect on Bptf splicing regulation wasseen following Sam68 co-transfection

DISCUSSION

Here we have used RNAseq to identify global changes inalternative exon splicing inclusion and parallel switches inthe RNA splicing environment during mouse malemeiosis Our data reveal that quantitatively significantprotein-coding splicing changes occur during mousemale meiosis The work described here builds onprevious work that detected extremely high levels ofoverall alternative splicing in the whole testis but whichconcluded that much of this is likely to be non-functionalbased on the low amplitude of the changes poor conser-vation and low protein-coding potential (13) In contrastthe meiotically regulated switches we describe here havehigh fold changes and are also regulated at some fre-quency in other tissues For example the meiosis-selected Bptf alternative 50 splice site is also selected inthe heart and muscle as well as the testis Exon skippingwas the most frequently identified form of alternativesplicing regulation between the 6 and 21 dpp testis tran-scriptomes (Figure 1B) and exon skipping is also thehighest frequency alternative splice event in the mousetranscriptome (61)

Most exons are under combinatorial control from dif-ferent splicing regulator proteins and also contributionsfrom transcription-related effects (4445) Although thecassette exon splice switches in the Odf2 and Nasp geneswere also associated with the concurrent use of alternativepromoters in meiosis our data suggest that global changes

in the concentration of RNA splicing regulators duringmeiosis make important contributions to the observedswitches in splicing One striking change is a switchbetween Ptbp1 and Ptbp2 gene expression in meiosis Asimilar switch is seen in neurogenesis (62ndash64) Both theencoded PTB proteins (PTBP1 and nPTB) are generallyseen as repressors of splicing (6566) although it is notclear whether nPTB is a weaker repressor than PTBP1as originally suggested (646567) Both PTB proteins arealso able to activate splicing although the dependence ofactivation versus inhibition of an exon on the location ofthe PTB binding sites is not clear (5354) In HeLa cells itappears that the two proteins affect the same targets(5368) whereas in neuroblastoma cells the proteins alsoaffect separate sets of exons (62) It is therefore difficultto predict whether the switch from PTBP1 to nPTB wouldcontribute to the observed splicing changes accompanyingmeiosis In the case of the meiotic exon of Ralgps2 bothPTBP1 and nPTB proteins increased inclusion in trans-fected cells using minigene constructs (Figure 6) Directinvestigation of the regulation of these exons in situ willrequire the utilization of appropriate mouse knockoutmodels (germ cells are not easily transfected in vitro)Interestingly though whereas both PTBP1 and nPTBdepended on the presence of a good candidate down-stream binding site for their splicing effect nPTB alsorequired a further weaker site that would not have beendetected by the common practice of searching for se-quences containing UCUU or (CU)n It would be inter-esting to know whether the presence of such additionalmotifs is a characteristic of exons regulated by nPTBOther regulatory proteins that might be important in

activating meiotic splicing of the Ralgps2 exon includeTra2b The Tra2b gene was also upregulated in meiosisand the Ralgps2 exon contained GAA target motifs andwas activated by Tra2b The Ralgps2 exon was alsoidentified as a Tra2b-CLIP tag in mouse testis (AJ Bestand DJ Elliott data not shown) Other exons identified byRNAseq here that are known from CLIP analysis in themouse testis to be bound strongly in vivo by Tra2b are thecassette exon of Nasp-T and poison exon of Tra2b (2769)Our transcriptome-wide analysis also identified changes

affecting the expression of other proteins that regulatesplicing These include the replacement of RBMX withRBMXL2 (26) and the meiotic upregulation of T-STARand Sam68 (282960) Predicted target sites for Sam68and T-STAR splicing regulators were enriched down-stream of activated exons and Sam68 protein is knownto regulate a cassette exon in the Sgce gene in meiosis thathas a downstream UAAA-rich site (70) Expression ofthese RNA-binding proteins is known to be importantfor male germ cell development Haploinsufficiency ofRbmxl2 causes infertility in mice (47) and Sam68 nullmice are infertile (7172) A number of unanticipatedchanges were also found in splicing regulator gene expres-sion Members of the CELF protein group includingCUG-BP2 were upregulated in meiosis This change islikely to be important as the Celf1 gene encoding CUG-BP1 is essential for spermatogenesis in mice (73) CELFproteins often work in antagonism to the muscleblindproteins (74) which were themselves transcriptionally


repressed during meiosis Target binding sites for CUG-BP2 and muscleblind proteins were also respectivelyenriched within activated and repressed exons (Figure 5)Previous data have shown that the transcription of a

core panel of genes changes during meiosis and providesmany of the structural components needed for this uniquedivision cycle (18ndash20) Many of the genes affected are ex-pressed only in the testis (eg the genes encoding synapto-nemal complex proteins) (18) In contrast many of theexons identified here as being under meiotic splicingcontrol are included to some extent in other mousetissues However two of the substantial switches insplicing patterns identified here by RNAseq have alreadybeen associated with important roles in animal germ celldevelopment Meiotic skipping of the Odf2 exon isassociated with a switch in protein function from asomatic intracellular role in organising microtubules

within the centriole to a post-meiotic role in organizingmicrotubules in the sperm tail (7576) Alternative splicingof the Nasp gene creates a protein isoform associated withmeiotic chromosomes that forms part of the machinerythat monitors DNA integrity during meiosis (77ndash79)Quantitative meiotic splicing regulation also takes placein other genes implicated in key roles in germ cell devel-opment The Ezh2 gene encodes an important chromatinmodifier that can affect development (80) and might playan important role in normal fertility (8182) A mutuallyexclusive exon is selected in the Ate1 gene and the meioticAte1 mRNA isoform is the major mRNA made from thisgene in the mouse testis (Supplementary File S1)Knockout of the Ate1 gene prevents germ cell develop-ment in the mouse (83) The major switches in alternativesplicing events discovered here might thus underlie essen-tial changes in the expression of meiotic protein isoforms

A

B

Figure 7 Dissection of the meiotically regulated Bptf cassette exon (A) The Bptf exon and its flanking intron sequences were screened for highaffinity and lower affinity PTB binding sites as in Figure 6 The sequences included as a result of the alternative 50 site selection are shaded grey andthe high affinity PTB sites are shown in red just downstream of the alternative 50 splice site Exon sequence is shown in upper case and intronsequence is in lower case The positions of the cloning oligonucleotides used to make the minigene are underlined (B) Splicing pattern of transcriptsmade from a Bptf minigene in HEK293 cells after co-transfection of expression vectors for different proteins


that play significant roles in preparing the cell for themorphological transformations that lie ahead

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online

ACKNOWLEDGEMENTS

The authors thank Dr Julian Venables for comments onthe manuscript and David Dolan for assistance withstatistics

FUNDING

Wellcome Trust [WT080368MA and WT089225Z09Zto DJE] BBSRC [BBD0139171 and BBI0069231 toDJE] Telethon Grant [GGPGGP09154] AssociazioneItaliana Ricerca sul Cancro (AIRC) 2010 (to CS)Addison Wheeler trust (to SG) Funding for openaccess Wellcome Trust

Conflict of interest statement None declared

REFERENCES

1 KwanT BenovoyD DiasC GurdS ProvencherCBeaulieuP HudsonTJ SladekR and MajewskiJ (2008)Genome-wide analysis of transcript isoform variation in humansNat Genet 40 225ndash231

2 MortazaviA WilliamsBA McCueK SchaefferL andWoldB (2008) Mapping and quantifying mammaliantranscriptomes by RNA-Seq Nat Methods 5 621ndash628

3 PanQ ShaiO LeeLJ FreyBJ and BlencoweBJ (2008)Deep surveying of alternative splicing complexity in the humantranscriptome by high-throughput sequencing Nat Genet 401413ndash1415

4 DreszerTR KarolchikD ZweigAS HinrichsASRaneyBJ KuhnRM MeyerLR WongM SloanCARosenbloomKR et al (2012) The UCSC Genome Browserdatabase extensions and updates 2011 Nucleic Acids Res 40D918ndashD923

5 DjebaliS DavisCA MerkelA DobinA LassmannTMortazaviA TanzerA LagardeJ LinW SchlesingerF et al(2012) Landscape of transcription in human cells Nature 489101ndash108

6 LlorianM and SmithCW (2011) Decoding muscle alternativesplicing Curr Opin Genet Dev 21 380ndash387

7 BlandCS WangET VuA DavidMP CastleJCJohnsonJM BurgeCB and CooperTA (2010) Globalregulation of alternative splicing during myogenic differentiationNucleic Acids Res 38 7651ndash7664

8 IpJY TongA PanQ ToppJD BlencoweBJ andLynchKW (2007) Global analysis of alternative splicing duringT-cell activation RNA 13 563ndash572

9 KalsotraA XiaoX WardAJ CastleJC JohnsonJMBurgeCB and CooperTA (2008) A postnatal switch of CELFand MBNL proteins reprograms alternative splicing in thedeveloping heart Proc Natl Acad Sci USA 105 20333ndash20338

10 McKeeAE NerettiN CarvalhoLE MeyerCA FoxEABrodskyAS and SilverPA (2007) Exon expression profilingreveals stimulus-mediated exon use in neural cells Genome Biol8 R159

11 FagnaniM BarashY IpJY MisquittaC PanQSaltzmanAL ShaiO LeeL RozenhekA MohammadNet al (2007) Functional coordination of alternative splicing in themammalian central nervous system Genome Biol 8 R108

12 KalsotraA and CooperTA (2011) Functional consequences ofdevelopmentally regulated alternative splicing Nat Rev Genet12 715ndash729

13 KanZ Garrett-EngelePW JohnsonJM and CastleJC (2005)Evolutionarily conserved and diverged alternative splicing eventsshow different expression and functional profiles Nucleic AcidsRes 33 5659ndash5666

14 MonesiV (1964) Ribonucleic acid synthesis during mitosis andmeiosis in the mouse testis J Cell Biol 22 521ndash532

15 MundingEM IgelAH ShiueL DorighiKM TrevinoLRand AresM Jr (2010) Integration of a splicing regulatorynetwork within the meiotic gene expression program ofSaccharomyces cerevisiae Genes Dev 24 2693ndash2704

16 FiumeM WilliamsV BrookA and BrudnoM (2010) Savantgenome browser for high-throughput sequencing dataBioinformatics 26 1938ndash1944

17 RussellLD EttlinR Sinha-HikimAP and CleggED (1990)Histological and Histopathological Evaluation of the Testis CacheRiver Press Clearwater FL

18 ChalmelF RollandAD Niederhauser-WiederkehrCChungSS DemouginP GattikerA MooreJ PatardJJWolgemuthDJ JegouB et al (2007) The conservedtranscriptome in human and rodent male gametogenesis ProcNatl Acad Sci USA 104 8346ndash8351

19 ClementeEJ FurlongRA LovelandKL and AffaraNA(2006) Gene expression study in the juvenile mouse testisidentification of stage-specific molecular pathways duringspermatogenesis Mamm Genome 17 956ndash975

20 EllisPJ FurlongRA WilsonA MorrisS CarterDOliverG PrintC BurgoynePS LovelandKL andAffaraNA (2004) Modulation of the mouse testis transcriptomeduring postnatal development and in selected models of maleinfertility Mol Hum Reprod 10 271ndash281

21 LeeK HaugenHS CleggCH and BraunRE (1995)Premature translation of protamine 1 mRNA causes precociousnuclear condensation and arrests spermatid differentiation inmice Proc Natl Acad Sci USA 92 12451ndash12455

22 MelamudE and MoultJ (2009) Stochastic noise in splicingmachinery Nucleic Acids Res 37 4873ndash4886

23 KammaH PortmanDS and DreyfussG (1995) Cell type-specific expression of hnRNP proteins Exp Cell Res 221187ndash196

24 XuM and HechtNB (2007) Polypyrimidine tract bindingprotein 2 stabilizes phosphoglycerate kinase 2 mRNA in murinemale germ cells by binding to its 3rsquoUTR Biol Reprod 761025ndash1033

25 ElliottDJ OgheneK MakarovG MakarovaOHargreaveTB ChandleyAC EperonIC and CookeHJ(1998) Dynamic changes in the subnuclear organisation of pre-mRNA splicing proteins and RBM during human germ celldevelopment J Cell Sci 111(Pt 9) 1255ndash1265

26 ElliottDJ VenablesJP NewtonCS LawsonD BoyleSEperonIC and CookeHJ (2000) An evolutionarily conservedgerm cell-specific hnRNP is encoded by a retrotransposed geneHum Mol Genet 9 2117ndash2124

27 GrellscheidS DalglieshC StorbeckM BestA LiuYJakubikM MendeY EhrmannI CurkT RossbachK et al(2011) Identification of evolutionarily conserved exons asregulated targets for the splicing activator tra2beta indevelopment PLoS Genet 7 e1002390

28 ParonettoMP ZalfaF BottiF GeremiaR BagniC andSetteC (2006) The nuclear RNA-binding protein Sam68translocates to the cytoplasm and associates with the polysomesin mouse spermatocytes Mol Biol Cell 17 14ndash24

29 VenablesJP DalglieshC ParonettoMP SkittLThorntonJK SaundersPT SetteC JonesKT andElliottDJ (2004) SIAH1 targets the alternative splicing factorT-STAR for degradation by the proteasome Hum Mol Genet13 1525ndash1534

30 RobidaM SridharanV MorganS RaoT and SinghR (2010)Drosophila polypyrimidine tract-binding protein is necessary forspermatid individualization Proc Natl Acad Sci USA 10712570ndash12575


31 RobidaMD and SinghR (2003) Drosophila polypyrimidine-tractbinding protein (PTB) functions specifically in the male germlineEMBO J 22 2924ndash2933

32 WangL FengZ WangX and ZhangX (2010) DEGseq an Rpackage for identifying differentially expressed genes from RNA-seq data Bioinformatics 26 136ndash138

33 AndersS and HuberW (2010) Differential expression analysisfor sequence count data Genome Biol 11 R106

34 YoungMD WakefieldMJ SmythGK and OshlackA (2010)Gene ontology analysis for RNA-seq accounting for selectionbias Genome Biol 11 R14

35 KatzY WangET AiroldiEM and BurgeCB (2010) Analysisand design of RNA sequencing experiments for identifyingisoform regulation Nat Methods 7 1009ndash1015

36 WangET SandbergR LuoS KhrebtukovaI ZhangLMayrC KingsmoreSF SchrothGP and BurgeCB (2008)Alternative isoform regulation in human tissue transcriptomesNature 456 470ndash476

37 TrapnellC PachterL and SalzbergSL (2009) TopHatdiscovering splice junctions with RNA-Seq Bioinformatics 251105ndash1111

38 ChernyD GoodingC EperonGE CoelhoMBBagshawCR SmithCW and EperonIC (2010) Stoichiometryof a regulatory splicing complex revealed by single-moleculeanalyses EMBO J 29 2161ndash2172

39 RossiP DolciS AlbanesiC GrimaldiP RiccaR andGeremiaR (1993) Follicle-stimulating hormone induction of steelfactor (SLF) mRNA in mouse Sertoli cells and stimulation ofDNA synthesis in spermatogonia by soluble SLF Dev Biol 15568ndash74

40 GrimaldiP PiscitelliD AlbanesiC BlasiF GeremiaR andRossiP (1993) Identification of 3rsquo5rsquo-cyclic adenosinemonophosphate-inducible nuclear factors binding to the humanurokinase promoter in mouse Sertoli cells Mol Endocrinol 71217ndash1225

41 SetteC BarchiM BianchiniA ContiM RossiP andGeremiaR (1999) Activation of the mitogen-activated proteinkinase ERK1 during meiotic progression of mouse pachytenespermatocytes J Biol Chem 274 33571ndash33579

42 ElliottDJ MillarMR OgheneK RossA KiesewetterFPryorJ McIntyreM HargreaveTB SaundersPT VogtPHet al (1997) Expression of RBM in the nuclei of human germcells is dependent on a critical region of the Y chromosome longarm Proc Natl Acad Sci USA 94 3848ndash3853

43 FujitaPA RheadB ZweigAS HinrichsAS KarolchikDClineMS GoldmanM BarberGP ClawsonH CoelhoAet al (2011) The UCSC Genome Browser database update 2011Nucleic Acids Res 39 D876ndashD882

44 HertelKJ (2008) Combinatorial control of exon recognitionJ Biol Chem 283 1211ndash1215

45 SmithCW and ValcarcelJ (2000) Alternative pre-mRNAsplicing the logic of combinatorial control Trends Biochem Sci25 381ndash388

46 GromakN RideauA SouthbyJ ScaddenAD GoodingCHuttelmaierS SingerRH and SmithCW (2003) The PTBinteracting protein raver1 regulates alpha-tropomyosin alternativesplicing EMBO J 22 6356ndash6364

47 EhrmannI DalglieshC TsaousiA ParonettoMPHeinrichB KistR CairnsP LiW MuellerC JacksonMet al (2008) Haploinsufficiency of the germ cell-specific nuclearRNA binding protein hnRNP G-T prevents functionalspermatogenesis in the mouse Hum Mol Genet 17 2803ndash2818

48 GoodingC EdgeC LorenzM CoelhoMB WintersMKaminskiCF ChernyD EperonIC and SmithCW (2013)MBNL1 and PTB cooperate to repress splicing of Tpm1 exon 3Nucleic Acids Res 41 4765ndash4782

49 PapoutsopoulouS NikolakakiE ChalepakisG KruftVChevaillierP and GiannakourosT (1999) SR protein-specifickinase 1 is highly expressed in testis and phosphorylatesprotamine 1 Nucleic Acids Res 27 2972ndash2980

50 KafaslaP MickleburghI LlorianM CoelhoM GoodingCChernyD JoshiA Kotik-KoganO CurryS EperonIC et al(2012) Defining the roles and interactions of PTB Biochem SocTrans 40 815ndash820

51 OberstrassFC AuweterSD EratM HargousY HenningAWenterP ReymondL Amir-AhmadyB PitschS BlackDLet al (2005) Structure of PTB bound to RNA specific bindingand implications for splicing regulation Science 309 2054ndash2057

52 BarashY CalarcoJA GaoW PanQ WangX ShaiOBlencoweBJ and FreyBJ (2010) Deciphering the splicing codeNature 465 53ndash59

53 LlorianM SchwartzS ClarkTA HollanderD TanLYSpellmanR GordonA SchweitzerAC de la GrangeP AstGet al (2010) Position-dependent alternative splicing activityrevealed by global profiling of alternative splicing events regulatedby PTB Nat Struct Mol Biol 17 1114ndash1123

54 XueY ZhouY WuT ZhuT JiX KwonYS ZhangCYeoG BlackDL SunH et al (2009) Genome-wide analysis ofPTB-RNA interactions reveals a strategy used by the generalsplicing repressor to modulate exon inclusion or skipping MolCell 36 996ndash1006

55 CleryA JayneS BenderskaN DominguezC StammS andAllainFH (2011) Molecular basis of purine-rich RNArecognition by the human SR-like protein Tra2-beta1 NatStruct Mol Biol 18 443ndash450

56 TsudaK SomeyaT KuwasakoK TakahashiM HeFUnzaiS InoueM HaradaT WatanabeS TeradaT et al(2011) Structural basis for the dual RNA-recognition modes ofhuman Tra2-beta RRM Nucleic Acids Res 39 1538ndash1553

57 GalarneauA and RichardS (2009) The STAR RNA bindingproteins GLD-1 QKI SAM68 and SLM-2 bind bipartite RNAmotifs BMC Mol Biol 10 47

58 LinQ TaylorSJ and ShallowayD (1997) Specificity anddeterminants of Sam68 RNA binding Implications for thebiological function of K homology domains J Biol Chem 27227274ndash27280

59 StossO OlbrichM HartmannAM KonigH MemmottJAndreadisA and StammS (2001) The STARGSG familyprotein rSLM-2 regulates the selection of alternative splice sitesJ Biol Chem 276 8665ndash8673

60 VenablesJP VernetC ChewSL ElliottDJCowmeadowRB WuJ CookeHJ ArtztK and EperonIC(1999) T-STARETOILE a novel relative of SAM68 thatinteracts with an RNA-binding protein implicated inspermatogenesis Hum Mol Genet 8 959ndash969

61 KerenH Lev-MaorG and AstG (2010) Alternative splicingand evolution diversification exon definition and function NatRev Genet 11 345ndash355

62 BoutzPL StoilovP LiQ LinCH ChawlaG OstrowKShiueL AresM Jr and BlackDL (2007) A post-transcriptionalregulatory switch in polypyrimidine tract-binding proteinsreprograms alternative splicing in developing neurons Genes Dev21 1636ndash1652

63 LillevaliK KullaA and OrdT (2001) Comparative expressionanalysis of the genes encoding polypyrimidine tract bindingprotein (PTB) and its neural homologue (brPTB) in prenatal andpostnatal mouse brain Mech Dev 101 217ndash220

64 LicatalosiDD YanoM FakJJ MeleA GrabinskiSEZhangC and DarnellRB (2012) Ptbp2 represses adult-specificsplicing to regulate the generation of neuronal precursors in theembryonic brain Genes Dev 26 1626ndash1642

65 MarkovtsovV NikolicJM GoldmanJA TurckCWChouMY and BlackDL (2000) Cooperative assembly of anhnRNP complex induced by a tissue-specific homolog ofpolypyrimidine tract binding protein Mol Cell Biol 207463ndash7479

66 PolydoridesAD OkanoHJ YangYY StefaniG andDarnellRB (2000) A brain-enriched polypyrimidine tract-bindingprotein antagonizes the ability of Nova to regulate neuron-specificalternative splicing Proc Natl Acad Sci USA 97 6350ndash6355

67 TangZZ SharmaS ZhengS ChawlaG NikolicJ andBlackDL (2011) Regulation of the mutually exclusive exons 8aand 8 in the CaV12 calcium channel transcript by polypyrimidinetract-binding protein J Biol Chem 286 10007ndash10016

68 SpellmanR LlorianM and SmithCW (2007) Crossregulationand functional redundancy between the splicing regulator PTBand its paralogs nPTB and ROD1 Mol Cell 27 420ndash434


69 StoilovP DaoudR NaylerO and StammS (2004) Humantra2-beta1 autoregulates its protein concentration by influencingalternative splicing of its pre-mRNA Hum Mol Genet 13509ndash524

70 ParonettoMP MessinaV BarchiM GeremiaR RichardSand SetteC (2011) Sam68 marks the transcriptionally activestages of spermatogenesis and modulates alternative splicing inmale germ cells Nucleic Acids Res 39 4961ndash4974

71 ParonettoMP MessinaV BianchiE BarchiM VogelGMorettiC PalombiF StefaniniM GeremiaR RichardSet al (2009) Sam68 regulates translation of target mRNAs inmale germ cells necessary for mouse spermatogenesis J CellBiol 185 235ndash249

72 RichardS TorabiN FrancoGV TremblayGA ChenTVogelG MorelM ClerouxP Forget-RichardA KomarovaSet al (2005) Ablation of the Sam68 RNA binding proteinprotects mice from age-related bone loss PLoS Genet 1 e74

73 KressC Gautier-CourteilleC OsborneHB BabinetC andPaillardL (2007) Inactivation of CUG-BP1CELF1 causesgrowth viability and spermatogenesis defects in mice Mol CellBiol 27 1146ndash1157

74 WangGS and CooperTA (2007) Splicing in disease disruptionof the splicing code and the decoding machinery Nat RevGenet 8 749ndash761

75 HuberD GeislerS MoneckeS and Hoyer-FenderS (2008)Molecular dissection of ODF2Cenexin revealed a short stretch ofamino acids necessary for targeting to the centrosome and theprimary cilium Eur J Cell Biol 87 137ndash146

76 RivkinE TresLL and KierszenbaumAL (2008) Genomicorigin processing and developmental expression of testicularouter dense fiber 2 (ODF2) transcripts and a novel nucleolarlocalization of ODF2 protein Mol Reprod Dev 75 1591ndash1606

77 AlekseevOM RichardsonRT and OrsquoRandMG (2009) Linkerhistones stimulate HSPA2 ATPase activity through NASP bindingand inhibit CDC2Cyclin B1 complex formation during meiosis inthe mouse Biol Reprod 81 739ndash748

78 RichardsonRT AlekseevOM GrossmanG WidgrenEEThresherR WagnerEJ SullivanKD MarzluffWF andOrsquoRandMG (2006) Nuclear autoantigenic sperm protein(NASP) a linker histone chaperone that is required for cellproliferation J Biol Chem 281 21526ndash21534

79 RichardsonRT BatovaIN WidgrenEE ZhengLXWhitfieldM MarzluffWF and OrsquoRandMG (2000)Characterization of the histone H1-binding protein NASP as acell cycle-regulated somatic protein J Biol Chem 27530378ndash30386

80 Di MeglioT KratochwilCF VilainN LocheA VitobelloAYoneharaK HrycajSM RoskaB PetersAH EichmannAet al (2013) Ezh2 orchestrates topographic migration andconnectivity of mouse precerebellar neurons Science 339204ndash207

81 HinzS MagheliA WeikertS SchulzeW KrauseHSchraderM MillerK and KempkensteffenC (2010)Deregulation of EZH2 expression in human spermatogenicdisorders and testicular germ cell tumors World J Urol 28631ndash635

82 LambrotR JonesS Saint-PharS and KimminsS (2012)Specialized distribution of the histone methyltransferase Ezh2 inthe nuclear apical region of round spermatids and its interactionwith the histone variant H1t2 J Androl 33 1058ndash1066

83 LeuNA KurosakaS and KashinaA (2009) Conditional Tekpromoter-driven deletion of arginyltransferase in the germ linecauses defects in gametogenesis and early embryonic lethality inmice PLoS One 4 e7734


number and cell morphology to convert a diploid stem cellin which chromatin is packaged with histones into a motilehaploid cell with a compact nucleus containing chromatinpackaged with protamines Exon-specific microarrayshave detected more alternative splicing in the wholeadult testis than in any other tissue except the brain(13) although at what stage in spermatogenesis thissplicing regulation originates is not known Perhaps themost important question concerning changes in alterna-tive splicing patterns during male germ cell development iswhether it is connected to meiosis Unlike cells in mitosisin which transcription is turned off meiotic cells arehighly transcriptionally active (14) In the single-celledyeast Saccharomyces cerevisiae meiosis is the only stagein the lifecycle to include alternative splicing In fact thetiming of the gene expression changes that drive meioticprogression in yeast is determined by a hierarchy ofmeiotic splicing events (1215)Adult male germ cell development takes around 30 days

in the testis (Figure 1A) Male mice are not born with afully developed male germ cell development pathway andinstead the first wave of spermatogenesis is initiated syn-chronously after birth (17) The testes of newborn micecontain germ cells arrested at the G0 and G1 stages ofthe cell cycle By 6 days post partum (dpp) these germcells have differentiated into spermatogonia a cell popu-lation that includes stem cells At 10 dpp somespermatogonia differentiate into meiotic spermatocytesMeiosis commences 12 dpp The first meiotic division(meiosis I) is complete by 21 dpp after which there is arapid second cell division (meiosis II) followed by progres-sive differentiation of the haploid cells into round sperm-atids elongating spermatids and finally mature sperm (thishaploid differentiation process is called spermiogenesis)Distinct programmes of transcriptional changes takeplace over animal meiosis which are critical for drivingdifferent molecular events such as the expression of genesencoding synaptonemal proteins like Sycp1 and the re-combinase Spo11 (18ndash20) It might be anticipated thatimportant splicing patterns represented only in testiswould be linked likewise to meiosis In the very laststages of male germ cell development nuclear DNA con-denses during the differentiation into elongating sperm-atids limiting transcriptional activity At this timetranslational control of pre-existing mRNAs transcribedearlier in meiosis plays a key role in gene expression (21)Much of the alternative splicing that has been detected

in whole human testis by exon-specific microarrays is notconserved in mice leads to the introduction of prematurestop codons and it occurs at individually low frequencies(13) These characteristics suggest that some alternativesplicing may represent lsquonoisersquo (22) arising from a lowerstringency of splicing control in the testis However thereare changes in the levels of expression of a number ofimportant RNA splicing regulators during spermatogen-esis including heterogeneous nuclear RNP (hnRNP)proteins (hnRNPA1 hnRNPG and PTB) (23ndash26) SR-like proteins (Tra2b) (27) and STAR proteins (Sam68and T-STAR) (2829) In this study we have set out toidentify whether there are high frequency changes inglobal splicing patterns that could affect protein isoform

production over meiosis and to comprehensively monitorthe expression of RNA splicing regulator proteins overthis timeframe Interestingly PTB is essential for malegerm cell development in fruitflies (3031) We speculatethat although polypyrimidine tract binding protein(PTBP1) is downregulated in meiosis nPTB might func-tionally replace it and thus ensure correct regulation ofPTB-dependent splicing events during germ cell differen-tiation Moreover other RNA splicing regulators alsochange in abundance during germ cell differentiation sug-gesting that they contribute to the meiotic patterns ofsplicing that we observe

MATERIALS AND METHODS

RNA sequencing

Sequencing reactions were done on pooled RNA samplesfrom 6 and 21 dpp mouse testis by Source BioscienceNottingham UK Four lanes of the flowcell were usedfor the sequencing of the 6 and 21 dpp samples on theGenome Analyzer II The Genome Analyzer (GA) wasrun for 38 cycles The images from the GA wereanalyzed with the GA pipeline software (v13 Illuminasoftware) on cycles 1ndash38 to undertake image analysisbase calling and sequence alignment to the referencegenome In all 5 345 040 and 5 561 352 quality filteredreads were obtained for the 6 dpp sample and 7 629 529and 7 610 503 were obtained for the 21 dpp sample Themouse NCBI Build 371 (mm9) was used as a referencegenome used for the read alignment Sequences werealigned with the ELAND software (lsquoELAND_rnarsquooption) which resulted in 3 922 430 4 085 374 5 515 372and 5 504 414 aligned reads for the 6 and 21 dppsamples respectively The aligned reads were used asinput for the Illumina CASAVA program (v10) tocount the sequence reads that align to genes exons andsplice junctions of the reference genome The mouse ref-erence feature files were supplied by Illumina and werederived from the mouse NCBI Build 371 The rawcounts of sequences aligning to features (gene exons andsplice junctions) were normalized by CASAVA bydividing the raw count by the length of the relevantfeature

The read counts per gene were used as input forDEGseq (32) and DESeq (33) to identify differentiallyexpressed genes Both tools are available via the statisticspackage R and Bioconductor DEGseq and DESeq usedifferent statistical approaches (Poisson distributionnegative binomial distribution) to estimate probabilitiesfor differential gene expression A P 0001 and a 2-foldchange (normalized) in expression levels were used as cut-off criteria Using these cut-offs DESeq identified 5835genes as differentially expressed whereas DEGseq found6362 differentially expressed genes The common set of5296 genes was taken as comprising the differentially ex-pressed genes for further analysis The resulting list wasread into the GOseq (34) BioconductorR-package toidentify GO terms that are over- or under-representedGOseq corrects for length bias in the detection of differ-ential expression in RNAseq The relationship between


Exonskipping




50

100

110

Nu

mb

ers

of a

lter

nat

ive

even

ts






A

540nt141nt

Day

6 R

NA

Adul

t RN

A



1 2 3

RNAseqreads

mRNAisoforms



126nt248nt

Day

6 R

NA

Day

21

RNA




RNAseqreads

mRNAisoforms

1 2


Day

6 R

NA

Day

21

RNA

Adul

t RN

A

187nt109nt



Constitutive exon



1 2 3

RNAseqreads

mRNAisoforms

B

C


Spermatogonia (Spg)

Meiotic cells (Spc)


Days post partum


Sertoli cells (SC)

0

10

21

31

D

RNAseq reads

RNAseq reads

RNAseq reads

RT-PCR validation

RT-PCR validation

RT-PCR validation










Cell isolation










Minigene analysis


RESULTS
















Ezh2

Odf2

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

Stage in meiosis

1 2 3 4

Odf2

Ezh2SC

7dp

pSC

17d

ppSp

gSp

cRt

d

1 2 3 4 5

A

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

1 2 3 4 5

C

Total testis RNA

Nasp

Picalm

Bptf

Add3

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

Nasp

Bptf

Picalm

Add3

1 2 3 4 5

P

B

Total testis RNA

Total testis RNA




2 Earlypachytene


4 Meiosis complete










Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp




Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts







SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES























































































Exonskipping




50

100

110

Nu

mb

ers

of a

lter

nat

ive

even

ts






A

540nt141nt

Day

6 R

NA

Adul

t RN

A



1 2 3

RNAseqreads

mRNAisoforms



126nt248nt

Day

6 R

NA

Day

21

RNA




RNAseqreads

mRNAisoforms

1 2


Day

6 R

NA

Day

21

RNA

Adul

t RN

A

187nt109nt



Constitutive exon



1 2 3

RNAseqreads

mRNAisoforms

B

C


Spermatogonia (Spg)

Meiotic cells (Spc)


Days post partum


Sertoli cells (SC)

0

10

21

31

D

RNAseq reads

RNAseq reads

RNAseq reads

RT-PCR validation

RT-PCR validation

RT-PCR validation










Cell isolation










Minigene analysis


RESULTS
















Ezh2

Odf2

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

Stage in meiosis

1 2 3 4

Odf2

Ezh2SC

7dp

pSC

17d

ppSp

gSp

cRt

d

1 2 3 4 5

A

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

1 2 3 4 5

C

Total testis RNA

Nasp

Picalm

Bptf

Add3

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

Nasp

Bptf

Picalm

Add3

1 2 3 4 5

P

B

Total testis RNA

Total testis RNA




2 Earlypachytene


4 Meiosis complete










Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp




Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts







SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES






























































































Cell isolation










Minigene analysis


RESULTS
















Ezh2

Odf2

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

Stage in meiosis

1 2 3 4

Odf2

Ezh2SC

7dp

pSC

17d

ppSp

gSp

cRt

d

1 2 3 4 5

A

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

1 2 3 4 5

C

Total testis RNA

Nasp

Picalm

Bptf

Add3

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

Nasp

Bptf

Picalm

Add3

1 2 3 4 5

P

B

Total testis RNA

Total testis RNA




2 Earlypachytene


4 Meiosis complete










Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp




Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts







SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES
























































































Minigene analysis


RESULTS
















Ezh2

Odf2

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

Stage in meiosis

1 2 3 4

Odf2

Ezh2SC

7dp

pSC

17d

ppSp

gSp

cRt

d

1 2 3 4 5

A

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

1 2 3 4 5

C

Total testis RNA

Nasp

Picalm

Bptf

Add3

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

Nasp

Bptf

Picalm

Add3

1 2 3 4 5

P

B

Total testis RNA

Total testis RNA




2 Earlypachytene


4 Meiosis complete










Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp




Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts







SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES


























































































Ezh2

Odf2

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

Stage in meiosis

1 2 3 4

Odf2

Ezh2SC

7dp

pSC

17d

ppSp

gSp

cRt

d

1 2 3 4 5

A

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

Rapgef1

Ralgps2

Vapa

Lrrc16a

Nxt1

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

1 2 3 4 5

C

Total testis RNA

Nasp

Picalm

Bptf

Add3

20

40

60

80

100

0Perc

enta

ge

Splic

ing

Incl

usi

on

6 d

ay

8 d

ay

13 d

ay

18 d

ay

16 d

ay

21 d

ay

Ad

ult

1 2 3 4

Stage in meiosis

SC 7

dpp

SC 1

7dpp

Spg

Spc

Rtd

Nasp

Bptf

Picalm

Add3

1 2 3 4 5

P

B

Total testis RNA

Total testis RNA




2 Earlypachytene


4 Meiosis complete










Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp




Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts







SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES






























































































Testis

Ovary

Heart

Gut

Kidney

Uterus

Muscle

Thymus

Spleen

Brain

Lung

Liv er

ODF2

EZH2

Lrrca1

BPTF

VAPA4

Add

Picalm

NXT1

Ppap2a

Nasp




Ppap2a

Nxt1

Picalm

Add3

Vapa4

Bptf

Lrrca1

Ezh2

Odf2

Nasp-T

Mouse tissues

Pan

el o

f mei

oti

c sp

lice

even

ts







SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES


























































































SRp55SRp40

ASFSF2SC35

SRp20

ERK2


SR p

rote

ins

SpgI S

pcII S

pcSpd

Rtd


PTBnPTB


hnRNP A1


ERK2

hn

RN

PS

SpgI S

pcRtd

II Spc S

pd

Rbfox2

T-STAR

Raver2

Ptbp2

Rbmxl2

Esrp2

Sfrs14Srpk1

Hnrnpa1Rbmx

Celf 4

Celf5Celf6

Ybx1

Ddx39Rsrc1

Nono

Ddx20Lsm2

Strap

Txnl4b

Srpk2Ddx46

Tsen2

Tsen34

Fox1

zcrb1

Prpf38a Prpf3


Jmjd6cwc15Hnrnph3

Gemin5


A

B C







Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES



























































































Upstream intron















0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES

























































































0

10

20

30

40

50

NS

p=00185

p=00002p=00009

Perc

enta

ge

Splic

ing

Incl

usi

on

GFP

GFP

T-ST

AR

Tra2

β

PTB

nPTB

1 2 3 4 5

1 2 3 4 5



p=00059

NS

NS NS

Perc

enta

ge

Splic

ing

Incl

usi

on

0

10

20

30

40

50

1 2 3 4 5 6

1 2 3 4 5 6

GFP

GFP

PTB

nPTB

GFP

GFP

PTB

nPTB





D

C

E

A

B

Co

mb

inat

ion

nu

mb

er

1

2

3

100

200

300

400

500

600

700

800

Base number

0

High stringency

Co

mb

inat

ion

nu

mb

er

10

Base number

Low stringency

100

200

300

400

500

600

700

8000

20

30

40






DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES


























































































DISCUSSION










A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES


























































































A

B




SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES
























































































SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES






























































































































































The splicing landscape is globally reprogrammed during male meiosis

Documents