Journal of Genetics and Genomics 44 (2017) 95e106

Contents lists available at ScienceDirect

Journal of Genetics and Genomics

Journal homepage: www.journals.elsevier.com/journal-of-genetics-
and-genomics/

Original research

Regulators of alternative polyadenylation operate at the transition

from mitosis to meiosis
Lingjuan Shan a, b, 1, Chan Wu a, b, 1, Di Chen b, Lei Hou c, Xin Li a, Lixia Wang a, Xiao Chu b, e,
Yifeng Hou d, Zhaohui Wang a, *
a
State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101,
China
b
The University of Chinese Academy of Sciences, Beijing 100049, China
c
Key Laboratory of Computational Biology, Chinese Academy of Sciences-Max Planck Partner Institute for Computational Biology, Shanghai Institutes for
Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China
d
State Key Laboratory of Plant Genomics and National Center for Plant Gene Research, Institute of Genetics and Developmental Biology, Chinese Academy of
Sciences, Beijing 100101, China
e
Key Laboratory of Genetic Network Biology, Collaborative Innovation Center of Genetics and Development, Institute of Genetics and Developmental
Biology, Chinese Academy of Sciences, Beijing 100101, China

a r t i c l e i n f o a b s t r a c t

Article history: In the sexually reproductive organisms, gametes are produced by meiosis following a limited mitotic
Received 8 November 2016 amplification. However, the intrinsic program switching cells from mitotic to meiotic cycle is unclear.
Received in revised form Alternative polyadenylation (APA) is a highly conserved means of gene regulation and is achieved by the
14 December 2016
RNA 30 -processing machinery to generate diverse 30 UTR profiles. In Drosophila spermatogenesis, we
Accepted 28 December 2016
observed distinct profiles of transcriptome-wide 30 UTR between mitotic and meiotic cells. In mutant
Available online 27 January 2017
germ cells stuck in mitosis, 30 UTRs of hundreds of genes were consistently shifted. Remarkably, altering
the levels of multiple 30 -processing factors disrupted germline's progression to meiosis, indicative of
Keywords:
Germ cell
APA's active role in this transition. An RNA-binding protein (RBP) Tut could directly bind 30 UTRs of 30 -
Mitosis to meiosis processing factors whose expressions were repressed in the presence of Tut-containing complex. Further,
Alternative polyadenylation we demonstrated that this RBP complex could execute the repression post-transcriptionally by recruiting
RNA-binding protein CCR4/Twin of deadenylation complex. Thus, we propose that an RBP complex regulates the dynamic APA
30 UTR profile to promote the mitosis-to-meiosis transition.
Copyright © 2017, The Authors. Institute of Genetics and Developmental Biology, Chinese Academy of
Sciences, and Genetics Society of China. Published by Elsevier Limited and Science Press. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction initiation remains the major obstacle for in vitro reconstitution of

Meiosis, which reduces a diploid germ cell to a haploid gamete,
is the fundamental event for sexual reproduction. Germ cells nor-
mally go through a limited mitotic amplification before entering
meiosis. Although the external cues, such as nutrient conditions for
yeast or retinoids for mammal, that trigger meiosis have been
identified (Honigberg and Purnapatre, 2003; Bowles et al., 2006;
Koubova et al., 2006), the internal program that switches germ
cells from mitotic to meiotic track is largely unknown. Meiosis
* Corresponding author.
E-mail address: zhwang@genetics.ac.cn (Z. Wang).
1
These authors contributed equally to this study.
gametogenesis (Sun et al., 2014).
Alternative polyadenylation (APA) of gene transcripts represents
an important layer of regulation during normal and pathological
development. APA events generate different 30 UTRs (untranslated
regions) of mRNA which may change the localization, stability,
translation efficiency of the transcripts (An et al., 2008; Sandberg et
al., 2008; Andreassi and Riccio, 2009; Mayr and Bartel, 2009; Pinto
et al., 2011), or even mediate protein-protein interactions without
changing the protein coding regions (Berkovits and Mayr, 2015;
Mayr, 2016). From yeast to human, 50%e70% of the genome
exhibit the usage of APA (Mangone et al., 2010; Ozsolak et al., 2010;
Wu et al., 2011; Derti et al., 2012; Smibert et al., 2012; Ulitsky et al.,
2012; Hoque et al., 2013). Characteristic 30 UTR profiles are

http://dx.doi.org/10.1016/j.jgg.2016.12.007
1673-8527/Copyright © 2017, The Authors. Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and Genetics Society of China. Published by
Elsevier Limited and Science Press. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
96 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106
associated with specific tissues or developmental stages. For
example, testis-specific mRNAs tend to have much shorter 30 UTR
than the brain-specific ones in general (Smibert et al., 2012; Ulitsky
et al., 2012); genome-wide progressive 30 UTR lengthening was
observed during cell differentiation or embryo development (Ji
et al., 2009), whereas 30 UTR shortening in somatic reprograming
(Sandberg et al., 2008; Ji and Tian, 2009; Shepard et al., 2011). APA
has also been shown in association with diseases (Mayr and Bartel,
2009; de Klerk et al., 2012; Jenal et al., 2012; Batra et al., 2014;
Masamha et al., 2014). Nevertheless, the causal relationship and
mechanistic connection between APA and developmental regula-
tion remain to be clarified.
30 UTR of different length is generated by around 20 factors of
RNA 30 -processing machinery coordinating in the sequential steps
from recognizing the polyadenylation sites, executing 30 -end
cleavage after recognition, to adding polyA tails at the cleaved end
(Di Giammartino et al., 2011; Zheng and Tian, 2014). Manipulating
the protein levels of the core factors in 30 -processing machinery led
to alternative polyA site usage (Takagaki et al., 1996; Takagaki and
Manley, 1998; Kubo et al., 2006; Kim et al., 2010; Jenal et al.,
2012; Martin et al., 2012; Lackford et al., 2014; Masamha et al.,
2014; Zheng and Tian, 2014). Conceivably, the 30 -processing fac-
tors are the major targets to accomplish APA regulation under
various conditions. If so, how are they regulated?
RNA-binding proteins (RBPs) can interact with both RNAs and
various other proteins, which can connect the upstream signals and
their target RNAs. In a human cell line, protein occupancy profiling
showed that 30 UTRs contained widespread contacts with RBPs
(Baltz et al., 2012), which regulate APA by competing with 30 -pro-
cessing factors for binding sites on RNA (Shi, 2012; Zheng and Tian,
2014). Further, it is also possible that RBPs post-transcriptionally
modulate the expression of 30 -processing factors to influence APA
globally.
In Drosophila spermatogenesis, Bam (Bag of marbles) dosage
determines how many mitotic cycles germ cells undergo before
meiosis (Insco et al., 2009). However, the mechanism downstream
of Bam that switches germ cells from mitosis to meiosis is un-
known. We previously demonstrated that an RBP complex con-
taining Tut (Tumorous testis), Bam, and Bgcn (Benign gonial cell
neoplasm) is required for limiting the mitotic cycles (Chen et al.,
2014). One of the molecular targets of this RBP complex is mei-
P26 whose expression is repressed in the presence of Tut-Bam-
Bgcn. Interestingly, mei-P26 30 UTR varies in length in the testes
between wild type and any of the mutants (tut, bam, or bgcn).
However, restoring Mei-P26 expression is not sufficient to rescue
the over-amplification defect of tut, bam, or bgcn, indicating the
involvement of other genes and a more complicated scenario (Insco
et al., 2012; Chen et al., 2014). This prompted us to examine the
30 UTR profile genome-wide and its relation to the germline
development in wild type and in these RBP mutants.
In this study, we found by transcriptome RNA-seq that the 30 UTR
patterns are distinct between mitotic and meiotic germ cells in
Drosophila testis. Altering the levels of multiple 30 -processing fac-
tors disrupted the germ cell progression from mitosis to meiosis,
implying APA's active role in this transition. We then obtained
in vitro and in vivo evidence to show that Tut could directly bind the
30 UTRs of 30 -processing factors and suppress their expression.
Further, we provided genetic and biochemical data to propose that
the RBP complex of Tut-Bam-Bgcn regulates the components of 30 -
processing machinery at the post-transcriptional level by recruiting
CCR4/Twin of deadenylation complex. Thus, we discovered the
upstream regulators of APA in Drosophila germline, and proposed
the causal relationship between APA and meiosis entry.
2. Results
2.1. Distinct 30 UTR profiles of the germ cells from mitotic to meiotic
stage
To explore the global 30 UTR pattern in relation to the transition
from mitosis to meiosis, we performed transcriptome RNA-seq
using the accurately staged germ cells, the mitotic spermatogonia
(SG; small 2-to-8-cell cysts) and the meiotic spermatocytes (SC; big
16-cell cysts), by laser excision from wild-type Drosophila testes
(Figs. S1A and B; Materials and methods). Additionally, we used
bam, bgcn, and tut mutants to obtain germ cells retained at mitotic
stage for comparison (Fig. S1CeE). The enrichment of the cell types
in wild-type SG or SC samples was verified by the relative mRNA
levels of a SG marker (bam) or two SC markers (bol and sa) obtained
from the RNA-seq data (Figs. S1F and G).
Global 30 UTR analysis was based on RNA-seq data at over 100
coverage of the transcriptome. The 30 UTR profiles were compared
between wild-type SG and SC, or between mutant SG (SG-bam, SG-
bgcn, or SG-tut) and wild-type SC (Fig. 1A and B). In general, more
than 600 genes showed 30 UTR shifts in their transcripts between SG
(wild-type or mutant) and SC. Among these genes, the majority had
more long 30 UTR forms in SG (wild-type or mutant) than in SC
(Fig. 1A, blue dots). This polarization of 30 UTR profile was more
prominent in the wild-type SG than in any of the mutant SG
samples (Fig. 1A, compare the gene numbers on top of the blue and
orange bars).
Since Tut, Bam, and Bgcn act in the same protein complex in SG
differentiation (Insco et al., 2009; Chen et al., 2014) and if mean-
while they are involved in APA regulation, we expect that the 30 UTR
patterns of these three mutant SG would consistently display the
distinction from that of the meiotic cells (Fig. 1A). Indeed, among
the genes showing distinct 30 UTR profiles between SG and SC, 274
genes were common in all three mutant SG samples (Fig. 1B).
Pearson correlation analysis, which was based on the 30 UTR pat-
terns of 1457 genes that exhibited the differences between wild-
type SG, SG-bam, SG-tut, or SG-bgcn and wild-type SC, showed
the close relationship of all mutant SG (Fig. 1C, clustering of SG-
bam, SG-tut, and SG-bgcn). The distinction between mitotic SG and
meiotic SC was also indicated by the relatively similar patterns of
wild-type SG, SG-bam, SG-tut, and SG-bgcn in contrast to that of
wild-type SC (Fig. 1C). Hence, the 30 UTR profiles are highly distinct
between mitotic and meiotic germ cells.
2.2. Altering the levels of 30 -processing factors disrupts the mitosis-
to-meiosis switch
Since we observed a global change of 30 UTR length in mitotic SG
vs meiotic SC, we asked if RNA 30 -processing factors (components
listed in Table S1), the major executors of APA, are involved in the
transitional control from mitosis to meiosis. We tried to alter the
levels of 30 -processing factors by RNAi or overexpression using SG-
specific GAL4 driver (bam-GAL4), and scored the phenotype of SG
overproliferation (Fig. 2, compare the accumulation of small germ
cells brightly stained by DAPI in GFP control and others; only those
exhibiting SG overgrowth are shown). Under normal develop-
mental conditions in Drosophila melanogaster, SG mitotically
amplify to 16-cell per cyst before entering meiosis, whereas many
cysts containing much more than 16 cells were observed when we
downregulated the levels of some 30 -processing factors (Fig. 2A,
arrows). In the wild-type background, knockdown of Pabp2, Hrg,
and Sym could induce significant SG overproliferation (Fig. 2A and
C; see Table S1 for all 20 factors tested in the RNAi screen).
 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106 97

Fig. 1. 30 UTRs shift in the germ cells from mitotic to meiotic stage. A: Distinct 30 UTR patterns between mitotic and meiotic cells. SG: mitotic spermatogonia; SC: meiotic sper-
matocytes; SG-tut, SG-bam, and SG-bgcn are mutant SG stuck in mitosis. Only the genes showing differential 30 UTRs between two cell types are depicted in each scattered plot. Both
axes represent the proportion of the long isoform over the total 30 reads mapped on the common region in the corresponding cells. Each dot represents one gene, and the gene
numbers of each comparison pair are on top of the colored bars. B: Venn diagram indicates the portion of 30 UTR-shifted genes common among SG-tut, SG-bam, and SG-bgcn in
comparison to the wild-type SC. C: Pearson correlation analysis was based on the 30 UTR patterns of 1457 genes that exhibited the differences between wild-type SC and wild-type
SG, SG-bam, SG-tut, or SG-bgcn. D: RNA-seq read distribution of two representative genes exhibiting 30 UTR shift between SG and SC. Only the reads mapped to the junction of the
last exon (thicker blue bar) and 30 UTR (thinner blue bar) are shown. The vertical scale is optimized to show the full range of each sample in the 30 UTR region, thus only the relative
distribution over the 30 UTR length is comparable between samples.

In most cases increasing 30 -processing factors promotes prox-
imal polyA site usage, but CFIm factors tend to act in the opposite
direction (Shi, 2012; Zheng and Tian, 2014), which means that
different factors regulate 30 UTR in different ways. Thus, we
wondered if increasing these components such as CFIm factors
could disrupt the meiotic transition. In our germline over-
expression screen, we also found that the upregulation (instead of
RNAi) of CFIm25 (CG3689 in Drosophila; a cleavage factor) caused
SG overamplification, and so did the fly homolog of WDR33 (Fig. 2B
and D; see Table S2 for all factors tested in the overexpression
screen). In sum, we found that increase or decrease of multiple 30 -
processing factors (CFmI25, Hrg, Pabp2, Sym, and WDR33 homo-
log) led to the SG overproliferation, a sign of APA's role in the
transition to meiosis.
2.3. The RBP complex of Tut-Bam-Bgcn promotes mitosis-to-meiosis
transition by a 30 UTR-mediated repression of 30 -processing factors
Apparently changing the levels of 30 -processing factors in germ
cells had the same consequence as tut, bam, or bgcn mutant. Then
98 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106

Fig. 2. Altering the levels of 30 -processing factors retains spermatogonia in mitosis. A: The whole testes stained for DNA are shown in black and white images. EGFP RNAi served as a
normal control which does not contain any SG tumor. The apical part is shown in the high magnification image. Knockdown of Pabp2, hrg, or Sym driven by bam-GAL4 (active in SG)
led to SG tumors (DNA brightly stained by DAPI, and pointed by yellow arrows in the high magnifications) in fly testes. VASA is a specific marker for germ cells. B: The testes
overexpressing CFIm25 or WDR33 by bam-GAL4. C: The bar graph summary of the phenotype shown in A. Y axis represents the rate of testes containing SG tumors. ‘n’ is the total
number of testes scored. Public stock numbers of the fly lines used for each gene are provided. D: The bar graph of scoring the phenotype shown in B. ‘n’ is the total number of testes
scored for each sample. Color coding in panel A is valid for all panels. Scale bars in black and white panels: 100 mm. Scale bars in color panels: 50 mm. See Tables S1 and S2 for all 30 -
processing factors tested.
 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106
what is the relationship between Tut-Bam-Bgcn complex and the
30 -processing factors? The phenotypic similarity and the fact that
Tut is an RNA-binding protein (Chen et al., 2014) prompted us to
test if Tut directly bind the 30 UTR and affect the production of 30 -
processing factors. We thus examined the physical interaction
between Tut and the 30 UTRs of 30 -processing factors by yeast
three-hybrid screen, an efficient assay to detect a direct associa-
tion between protein and RNA (Bernstein et al., 2002). Interest-
ingly, Tut protein could bind the 30 UTRs of multiple 30 -processing
factors (Figs. 3A and S2; Table S3 for all 31 segments tested).
Strong assay signals were detected in different isoforms of Pabp2
or Pcf11 mRNA (Fig. 3A, Pabp2-RA3U, -RB3U; Pcf11-RC3U, -RH3U).
For CFIm25 transcripts, different signals were observed using
different segments of 30 UTR from the same isoform (Fig. 3A, *).
Apparently, Tut is able to bind the mRNA 30 UTR of at least ten 30 -
processing factors.
To test whether these Tut-associated 30 UTRs have any influence
on the expression of their corresponding transcripts, we cloned
each of these 30 UTRs downstream of Renilla luciferase coding
sequence and transfected them into the S2 cells in culture with or
without Tut-Bam-Bgcn protein constructs (Fig. 3B). Firefly lucif-
erase was co-transfected as a transfection control over which the
reporter expression associated with different 30 UTRs was normal-
ized. Although both 30 UTR isoforms of Pabp2, Pabp2-RA3U and
-RB3U, could bind Tut, only the RB-tagged reporter was suppressed
by the Tut-Bam-Bgcn complex (Fig. 3B). Interestingly, the RB iso-
form is relatively abundant in SG samples based on our RNA-seq
data sets (data not shown). Except CG4612 (a putative polyA
binding protein), all other 30 -processing factors, which were tested
positive in the binding assay between their 30 UTRs and Tut, showed
significantly reduced reporter expression in the presence of Tut-
Bam-Bgcn.
Next we tested the in vivo relationship between Tut-Bam-Bgcn
complex and 30 -processing factors by a GFP reporter fused to
CFIm25 30 UTR in fly testis (Fig. 3CeG). This reporter expression,
driven by bam-GAL4, was repressed in the Bam-positive SG cells in
comparison to the control reporter containing only SV40 30 UTR in
wild-type background (Fig. 3, compare C and D). In contrast, this
CFIm25 30 UTR-containing reporter was de-repressed in all SG cells
of tut, bam, or bgcn mutant testes even when Bam was present
(Fig. 3EeG). This clearly indicated that any member of Tut-Bam-
Bgcn complex was required for the repression of CFIm25 via its
30 UTR.
If Tut-Bam-Bgcn complex represses the expression of 30 -pro-
cessing factors in vivo, it would be hard to overexpress them in SG
under the wild-type condition, so that we examined the effect of
overexpressing 30 -processing factors in tut-compromised back-
ground. If Tut-containing complex promotes mitosis-to-meiosis
through 30 -processing factors, we expect to see that increasing
the levels of these factors would enhance the defects of tut mutant
testis. We used a hypomorphic tut allele which generally exhibits a
mild phenotype as the baseline control (Fig. 4A, compare TAP-
GFP;tut4/þ and TAP-GFP;tut4). Indeed, when we overexpressed 30 -
processing factors in SG by bam-GAL4, we observed much more
severe SG ‘tumor’ accumulation with eight members of the 30 -
processing machinery than that with the control (Fig. 4). For Pcf11,
overexpression of three different protein isoforms showed obvious
enhancement on tut mutant phenotype (Fig. 4, Pcf11-RB, -RC, or
-RD). In contrast, removing a copy of Pcf11 did not display any effect
in tut mutant background (data not shown). In sum, the in vitro and
in vivo experiments indicated that Tut, Bam, and Bgcn promote
mitosis-to-meiosis transition by a 30 UTR-mediated repression of 30 -
processing factors.
99
2.4. Tut can recruit RNA deadenylation factors to promote mitosis-
to-meiosis transition
As described above, Tut-Bam-Bgcn complex and their targets'
30 UTRs were required for the translational repression of 30 -pro-
cessing factors. Then what connects the two events of 30 UTR-
binding and translational repression in the process of mitosis-to-
meiosis switch? Deadenylation or polyA shortening commonly
precedes mRNA decay and compromised translation. CCR4-NOT
and PAN2-PAN3 are the two major deadenylation machineries
(Parker and Song, 2004; Wahle and Winkler, 2013; Wolf and
Passmore, 2014), and thus we explored their roles in the Tut-
mediated post-transcriptional regulation of 30 -processing factors
in germline by an RNAi screen of the major components of CCR4-
NOT and PAN2-PAN3 (Table S4). Among the 11 candidates, only
those showing SG tumors (yellow arrows) in RNAi screen were
illustrated in Fig. 5. Knockdown of CCR4-NOT components consis-
tently displayed more severe effect than that of PAN2-PAN3
(Fig. 5A, compare the DAPI staining of PAN3 with others; see the SG
tumor rates in Fig. 5B). Not surprisingly, reducing pAbp (Drosophila
homolog of PABPC1 which is involved in diverse aspects of cyto-
plasmic RNA, such as recruiting CCR4-NOT or PAN2-PAN3 complex
(Norbury, 2013)) also showed very prominent effect (Fig. 5B).
Nevertheless, the consequence of reducing deadenylation-related
proteins was similar to that of disturbing the levels of 30 -process-
ing factors, which kept the germ cells in mitosis.
To examine if the in vivo function of deadenylation complexes is
in the same pathway as tut in germ cell development, we did the
genetic interaction assay by crossing the available mutants of CCR4-
NOT complex with the weak allele of tut. The genetic association
with tut was revealed by the marked increase of SG tumor rate in
various mutants of Not1 or pAbp but not in those of PAN2 (Figs. 6A, B
and S3), suggesting that CCR4-NOT complex plays a more signifi-
cant role in the tut-dependent transition from mitosis to meiosis
than PAN2-PAN3.
Because CCR4-NOT complex does not bind to its specific RNA
targets on its own, we speculated if the RNA-binding Tut or Bam
could be the RBP recruiting the complex to its targets. In other
words, does Tut or Bam physically interact with any subunit of the
CCR4-NOT complex? Bam has been found to be physically associ-
ated with Twin (the Drosophila homolog of CCR4, a polyA-specific
ribonuclease) in Drosophila ovary (Fu et al., 2015), as well as with
Tut (Chen et al., 2014). It is possible that they are all in the same
complex. We therefore examined the physical association between
Tut and Twin in the cultured S2 cells. Not only we confirmed the
binding between Bam and Twin, we also detected the physical
association between Tut and Twin, and furthermore the presence of
Tut, Bam, Bgcn, and Twin in the same co-immunoprecipitation
complex (Fig. 6C). However, stable association of Tut and Pop2
(Drosophila homolog of CAF1, another polyA-specific ribonuclease)
was hard to detect (data not shown). It appeared that Tut and Bam
could recruit the CCR4-NOT complex to the specific target mRNAs
in germ cell progression to meiosis.
3. Discussion
Based on our genetic and biochemical evidence, a model of how
an RBP complex (Tut-Bam-Bgcn) regulates the levels of 30 -pro-
cessing machinery to switch germ cells from mitosis to meiosis is
illustrated in Fig. S4. In the mitotically amplifying germ cells, Tut,
Bam, and Bgcn form a complex, bind to the 30 UTR of selected 30 -
processing factors, and recruit the RNA deadenylation machinery to
downregulate the target genes. Consequently, 30 -processing
100 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106
 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106 101

Fig. 4. Tut genetically antagonizes 30 -processing factors in mitosis-to-meiosis transition. A: Genetic interactions between hypomorphic tut4 and 30 -processing factors revealed by
bam-GAL4 driven overexpression. The flies were dissected within 12 h after eclosure. All images were testes stained with DNA dye DAPI, and only positive interactions are shown
other than the negative control. Green dotted lines divide the testis in half longitudinally, and the yellow dotted lines indicate the apical 1/6 section of the testis. Scale bar: 100 mm.
B: The scoring of genetic interaction screen of all tested factors. n ¼ total testes scored. The black bars indicate the proportion of testes containing SG tumors that occupy more than
1/6 of testis length.

Fig. 3. Tut binds the 30 UTRs of 30 -processing factors and suppresses their expression. A: Yeast three-hybrid assay to show the direct association between Tut protein and the 30 UTRs
of 30 -processing factors (see also Fig. S2 and Table S3 for all 31 segments tested). Tut was fused to AD (activation domain) and the tested RNAs fused to MS2 which can be bound by a
protein with a DNA-binding domain. The pairing of AD-IRP and IRE-MS2 is the positive control, and that of AD-Tut and MS2-GFP or AD-Tut and MS2-aTub84B-RA3U is the negative
control. CG4612-RC3U-2*, 1213e1584 nt of the isoform-RC's 30 UTR; CFIm25-RB3U-1*, 1e683 nt of the isoform-RB's 30 UTR; CFIm25-RB3U-2*, 664e1375 nt of the isoform-RB's 30 UTR;
all others contain the full length of 30 UTR of the indicated mRNA isoform. This assay was repeated at least twice for each 30 UTR construct. B: Luciferase assay to show the effect of
Tut-Bam-Bgcn complex on the expression of the reporter transcripts containing the 30 UTRs of 30 -processing factors. The Renilla reporter's production was normalized against firefly
reporter which served as a transfection control. This assay was repeated at least twice for each 30 UTR construct. Error bar indicates SD. ns, *, **, ***, and **** stand for ‘no significant
difference’, P < 0.05, P < 0.01, P < 0.001, and P < 0.0001 in Student's t-test, respectively. CeG: Tut-Bam-Bgcn complex is required in vivo for CFIm25 repression via its 30 UTR. This
experiment was repeated three times. Scale bars: 50 mm. CeC′′: Genotype: bam-GAL4/Y;UAS-EGFP-SV40/þ. Yellow dots outline the region of Bam-positive spermatogonia. Only the
apical part of the testis is shown. This control EGFP reporter was expressed in Bam-positive cells. DeD′′: Genotype: bam-GAL4/Y;UAS-EGFP-CFIm25RB-30 UTR-SV40/þ; þ/tut1 or þ/tut3.
EGFP expression was repressed in most Bam-positive spermatogonia. EeE′′: Genotype: bam-GAL4/Y;UAS-EGFP-CFIm25RB-30 UTR-SV40/þ;tut1/tut3. EGFP was de-repressed in tut
mutant. FeF′′: Genotype: bam-GAL4/Y;UAS-EGFP-CFIm25RB-30 UTR-SV40/þ;bamBG/bamD86. Yellow dots outline the GFP-positive spermatogonia. EGFP was de-repressed in bam
mutant. GeG′′: Genotype: bam-GAL4/Y;UAS-EGFP-CFIm25RB-30 UTR-SV40/þ;bgcnQS2/bgcn20093. EGFP was de-repressed in bgcn mutant.
102 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106

Fig. 5. Reduction of deadenylation-related genes retains the germ cells in mitotic growth. A: Knockdown of CCR4-NOT and PAN2-PAN3 components by bam-GAL4 led to SG
tumors in testes. All images of DAPI staining alone are of the same magnification (scale bar: 100 mm). The images of DAPI and VASA overlay show only part of the testis (scale bar:
50 mm). Yellow arrows point to SG tumors. B: The scoring of knockdown phenotypes shown in A. n, total testes scored for each RNAi line. Public stock numbers are given for the
genes tested.

machinery is maintained at the level to generate a characteristic
APA profile (Fig. 1A, SG) favoring the transition to meiosis (Fig. S4,
left panel). This transition is blocked when the APA profile is shifted
by disrupting any of the upstream steps, such as decreasing the
levels of the RBP complex or RNA-deadenylation components, or
increasing the levels of certain subunits of 30 -processing machinery
(Fig. S4, right panel). It is a coordinated teamwork of APA regulators
that drives germ cell from mitosis to meiosis.
3.1. The relationship between proliferation/differentiation and
30 UTR length
It has been documented that 30 UTR shortening is associated
with cell proliferation or naïve cell state, and lengthening with a
more differentiated state (Ji et al., 2009; Ji and Tian, 2009). We
observed in Drosophila germline that the proliferative cells
contained larger proportion of long 30 UTR isoforms than the
‘differentiated’ meiotic cells (Fig. 1, SG vs SC). Similar phenomenon
was observed in mouse testes (Li et al., 2016). Do germ cells behave
differently? Not entirely, at least in terms of the relation between
30 UTR length and differentiation. The development of a germ cell
towards a mature gamete may be considered a process of de-
differentiating to a ‘ground’ state in preparation for the totipotent
zygote. Additionally, it is beneficial to have shorter 30 UTRs for the
synthetically active, volumn-growing SC because shorter 30 UTRs
are less likely to get targeted by microRNAs (Sandberg et al., 2008).
3.2. Disturbance of 30 -processing factors and the mitosis-to-meiosis
transition
There are at least twenty 30 -processing factors whose homologs
can be found in Drosophila (Table S1). In our RNAi and
 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106 103

Fig. 6. Tut-Bam-Bgcn recruits CCR4-NOT to control mitosis-to-meiosis transition. A: Genetic interaction test of the hypomorphic tut4 and RNA decay related factors. The flies were
dissected within 2 days after eclosure. SG tumor cells were brightly stained by DAPI. VASA is a germ cell specific marker. Scale bar, 100 mm. B: The bar graph summary of the genetic
interaction shown in A. The dark gray bars indicate the percentage of testes containing only SG tumor and no SC at all. C: The co-immunoprecipitation assay to show the physical
interactions between Tut and CCR4/Twin, Bam and CCR4/Twin. The blots on the right reveal that Bam, Bgcn, and CCR4/Twin were present in the complex immunoprecipitated by
TAP-Bam. *, a non-specific band. This experiment was repeated three times.

overexpression screens of 30 -processing factors, either decrease
(Hrg, Pabp2, or Sym) or increase (CFmI25 or WDR33 homolog) of
certain components retained the germ cells in mitotic cycle (Fig. 2;
Tables S1 and S2). If Tut-Bam-Bgcn complex acts as a repressor on
the production of 30 -processing machinery, how do we explain the
fact that knocking down some of the 30 -processing components
also retained germ cells in mitosis? First, this may simply reflect the
effect of positive or negative regulators that different 30 -processing
components are in contact with, so that different effects are
exhibited when the levels of different factors are changed. Sec-
ondly, it implies a close connection between APA events and germ
cell development. That is, whenever APA is misregulated by simply
altering the level of only one component of 30 -processing ma-
chinery, the germ cells would be stuck at mitotic stage and unable
104 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106
to progress to meiosis.
3.3. APA-regulated network and meiosis initiation
When any member of the Tut-Bam-Bgcn complex was
compromised, the 30 UTR profile of the mutant mitotic germ cells
exhibited a shift in a similar set of genes (Fig. 1B and C). We spec-
ulate that a set of genes critical for meiosis initiation is under APA
regulation, which constitutes the barrier between mitosis and
meiosis. Further exploration combining transcriptome analyses of
finely manipulated cells and candidate genetic screens is required
to unveil this network promoting meiosis.
Although the scenario of meiosis is highly conserved from yeast
to human, is there a common molecular program governing its
initiation? Retinoic acid (RA)-STRA8 signaling pathway acts before
pre-meiotic DNA replication to switch germ cells from mitosis to
meiosis in mouse (Baltus et al., 2006; Koubova et al., 2006;
Anderson et al., 2008). STRA8 induction by RA depends on the
presence of DAZL (Baltus et al., 2006; Lin et al., 2008). Very recently,
MYC family partner MAX has been found to negatively regulate
meiosis entry (Suzuki et al., 2016). Drosophila homologs of DAZL
and MAX are Boule and Max, respectively; however, Boule is
required for the completion of meiosis instead of initiation
(Eberhart et al., 1996; Xu et al., 2003). Max function in fly gonad is to
be examined. The molecular program of meiosis initiation is
probably conserved because the human or mouse homolog can
successfully restore the fertility of fly Doa mutants, which also
shows mitotic over-amplification as tut, bam, or bgcn mutants
(Zhao et al., 2013). Not surprisingly, Doa encodes a highly conserved
protein kinase. Though Tut, Bam, and Bgcn are not conserved
proteins, the global APA shift that they regulate in fly spermato-
genesis has been observed in mouse testes (Li et al., 2016), and
should be explored for APA's role in the transition to meiosis in
mammals.
4. Materials and methods
4.1. RNA-seq library construction and sequencing
The SG and SC samples were isolated by laser microdissection
(Leica LMD7000, Germany) using His-GFP (BL24163, Bloomington
Drosophila Stock Center) as a marker (Fig. S1B; ~800e1000 SG/
sample; ~300 SC/sample; triplicate repeats). The cDNA libraries
were constructed following SMARTer Ultra Low Input RNA Kit for
Sequencing-v3 (634848, Clonetech, USA) with 12 PCR cycles, and
purified using the Agencourt Ampure XP Kit (A63880, Beckman
Coulter, USA). The full-length cDNA outputs were processed with
the Nextera XT DNA Sample Preparation Kit (FC-131-1024, Illumina,
USA).
Total RNAs from tut (tut3), bam (bamBG/bamD86) or bgcn
(bgcn20093/bgcnQS2) mutant testes (~100 pairs/sample) were
extracted using Trizol (15596-026, Invitrogen, USA). mRNAs were
enriched from total RNAs using Dynabeads mRNA DIRECT Kit
(61012, Invitrogen). RNA-seq libraries were generated using the
mRNA-Seq Sample Preparation kit (RS-930-1001, Illumina)
following manufacturer instructions. Libraries were sequenced
using 125 paired-end on Illumina HiSeq 2500 sequencer. Triplicate
sampling for wild-type SG and SC, and duplicate sampling for all
mutants were prepared. At least 70 M reads of each sample were
obtained. The sequencing data and the data analysis related to Fig. 1
were submitted to NCBI (GSE86974).
4.2. RNA-seq data analysis
RNA-seq reads were mapped to Drosophila melanogaster
genome (dm6) using TopHat (version 2.1.0) (Kim et al., 2013) with
parameters only allowing uniquely mapped reads. The mapped
reads were assembled using Cufflinks (version 2.2.1) with dm6
annotation as the reference (Trapnell et al., 2013). The assembled
transcripts were merged and annotated using Cuffcompare
(version 2.2.1) with dm6 annotation as the reference. The expres-
sion levels of each gene were then calculated by the fragments per
kilobase of exons per million fragments mapped (FPKM) using
Cuffdiff (version 2.2.1) with default parameters.
4.3. Identification of 30 UTR extensions
Only the extension of 30 UTR but not new isoforms due to
alternative splicing was the focus of this study. To identify 30 UTR
extensions, RNA-seq reads that mapped to the downstream of the
shortest annotated 30 UTR were collected. The initial 30 UTR exten-
sions were contiguously assembled from the 30 end of the shortest
annotated 30 UTR with a 10-nt resolution. The initial extension was
able to extend to next 10-nt bin only if the FPKM of the assembled
extension (FPKMext) 1 and the FPKM of the 10-nt bin (FPKMbin)
1. Next, the assembled transcript extension in one condition was
compared to that in the other condition, and only the overhangs
were defined as final 30 UTR extensions. The final extensions were
kept if their lengths were at least 200 nt and their FPKMs were at
least 2-fold greater in one condition than another condition.
P < 0.05 and adjusted P < 0.1 were used as cutoffs to define
differentially expressed 30 UTR extensions. The P value and adjusted
P value were calculated using DESeq2.
4.4. Drosophila strains and husbandry
The fly strains used: tut1 (Chen et al., 2014); tut3 (Chen et al.,
2014); tut4 (Chen et al., 2014); bamD86 (McKearin and Ohlstein,
1995); bamBG (Chen and McKearin, 2005); bgcnQS2 (Ohlstein et al.,
2000); bgcn20093 (Jin et al., 2008); UAS-dcr2 (a gift from T. Tabata
lab); UAS-GFP (a gift from T. Xie lab); bam-GAL4VP16 (Chen and DM,
2003); His-GFP, nos-GAL4VP16, Not1MI07631, pAbpk10109, pAbpEY11561,
and Not3KG10496 from Bloomington Drosophila Stock Center; UAS-
TAP-GFP was generated in our lab. Fly stocks were maintained un-
der standard culture conditions and all flies were dissected 0e2
days after eclosure unless otherwise indicated.
For RNAi experiments, the flies were ordered from Bloomington
Drosophila Stock Center, NIG-Fly, and Tsinghua Fly Center, and the
stocks' numbers are shown in Figs. 2 and 5. Flies were cultured at
25 C for 2e3 days and transferred to 30 C for another 10e12 days
before dissection. For overexpression of 30 -processing factors, the
flies were cultured at 25 C for 2 days and transferred to 30 C for
another 18 days before dissection. For tut related experiments, flies
were cultured at 24 C and the dissection time is described in
related figures.
4.5. Transgenic flies
The p51D flies were chosen as the hosts for attB-attP mediated
transgenesis (Bischof et al., 2007). UAS-Pcf11-RC, UAS-CG4612-RA,
UAS-Pabp2-RA, UAS-CstF77-RE, UAS-cbc-RA, UAS-CFIm25-RB, UAS-
IntS9-RA, UAS-CstF50-RA, UAS-hrg-RA, UAS-IntS11-RA, and UAS-
CG1109-RA were generated by injecting Drosophila Genomics
Resource Center (DGRC) clones UFO10877, UFO02873, UFO06451,
UFO08917, UFO01678, UFO03148, UFO03412, UFO01987,
UFO05604, UFO03259, and UFO07896 into the p51D flies, respec-
tively. UAS-CPSF73-RA, UAS-CstF64-RA, UAS-Pcf11-RB, UAS-Pcf11-RD,
UAS-Sym-RA, and UAS-Cpsf100-RA were generated by cloning the
coding sequences from DGRC clones FMO06340, FMO13854,
UFO10877, RE43027, LD11480, BS10621, and FMO13689 into the
 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106
pUAST-attB vector, respectively. UAS-EGFP-SV40 was generated by
cloning EGFP coding sequence in pUAST-attB vector, and UAS-EGFP-
CFIm25-RB30 UTR-SV40 by inserting CFIm25-RB30 UTR between EGFP
coding sequence and the SV40 element.
4.6. Immunofluorescence
Fly testes were prepared and immunostained as previously
described (Li et al., 2007). The primary antibodies used: rabbit anti-
VASA (1:8000; against (KLH)-MSDDWDDEPIVDTRGARC-OH),
guinea pig anti-VASA (1:4000; against 6  His-VASA produced in
E. coli), mouse anti-Bam (1:2000) (a gift from Dahua Chen). Fluo-
rescent images were collected by OLYMPUS FV1000 Confocal
microimaging system.
4.7. Yeast three-hybrid assay
30 UTRs of 30 -processing factors were cloned from w1118 cDNA
and inserted into pIIIA/MS2-1 and pIIIA/MS2-2 vectors (Bernstein
et al., 2002), respectively. Yeasts were cultured on SD/-Leu/-Ura
medium (DDO) to confirm the transformation of testing plasmid
DNA. Three independent clones were selected and cultured in SD/-
Leu/-Ura medium (DDO) overnight. The yeasts were collected and
subjected to qualitative (filter) assays as described (Bernstein et al.,
2002).
4.8. Cell culture and luciferase assay
S2 cells were cultured in SFM serum free medium (10902, Gibco,
USA). Transfection was performed following Cellfectin Reagent
(10362-100, Invitrogen). As shown in Fig. 3B, pAc5.1-Renilla
luciferase-X 30 UTR-SV40 30 UTR and pAc5.1-Firefly luciferase-SV40
30 UTR constructs were co-transfected with act-GAL4, pUAST
(-Flag-Tut, -HA-Bam), and pAc5.1-Flag-Myc-Bgcn expression vec-
tors. Cells were collected 60 h after transfection and manipulated
according to the manufacturer's instructions of Dual-Luciferase
Reporter Assay System (E1910, Promega,).
4.9. Immunoprecipitation and Western blot analysis
Extracts were obtained from the cultured S2 cells transfected
with relevant constructs. Immunoprecipitation and Western blot
analysis were described previously (Chen et al., 2014). Original
Twin construct was a gift from T. Xie and subcloned into pUAST-
Flag vector. IgG-beads (A2909, Sigma) were used in co-
immunoprecipitation.
Acknowledgments
We are deeply grateful to Drs. Xiaofeng Cao, Dahua Chen, Xun
Huang, Tetsuya Tabata, Marvin Wickens, Ting Xie, and Weicai Yang
for sharing research materials; to Bloomington Drosophila Stock
Center, NIG-Fly, Tsinghua Fly Center (THFC), and Drosophila Geno-
mics Resource Center for providing fly stocks and cDNA clones. This
work was supported by National Key Basic Research Program of
China (No. 2013CB945000) and National Science Foundation of
China (No. 31471345).
Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jgg.2016.12.007.
105
References
An, J.J., Gharami, K., Liao, G.-Y., Woo, N.H., Lau, A.G., Vanevski, F., Torre, E.R.,
Jones, K.R., Feng, Y., Lu, B., Xu, B., 2008. Distinct role of long 30 UTR BDNF mRNA
in spine morphology and synaptic plasticity in hippocampal neurons. Cell 134,
175e187.
Anderson, E.L., Baltus, A.E., Roepers-Gajadien, H.L., Hassold, T.J., de Rooij, D.G., van
Pelt, A.M., Page, D.C., 2008. Stra8 and its inducer, retinoic acid, regulate meiotic
initiation in both spermatogenesis and oogenesis in mice. Proc. Natl. Acad. Sci.
U. S. A. 105, 14976e14980.
Andreassi, C., Riccio, A., 2009. To localize or not to localize: mRNA fate is in 30 UTR
ends. Trends Cell Biol. 19, 465e474.
Baltus, A.E., Menke, D.B., Hu, Y.C., Goodheart, M.L., Carpenter, A.E., de Rooij, D.G.,
Page, D.C., 2006. In germ cells of mouse embryonic ovaries, the decision to enter
meiosis precedes premeiotic DNA replication. Nat. Genet. 38, 1430e1434.
Baltz, A.G., Munschauer, M., Schwanhausser, B., Vasile, A., Murakawa, Y.,
Schueler, M., Youngs, N., Penfold-Brown, D., Drew, K., Milek, M., Wyler, E.,
Bonneau, R., Selbach, M., Dieterich, C., Landthaler, M., 2012. The mRNA-bound
proteome and its global occupancy profile on protein-coding transcripts. Mol.
Cell 46, 674e690.
Batra, R., Charizanis, K., Manchanda, M., Mohan, A., Li, M., Finn, D.J., Goodwin, M.,
Zhang, C., Sobczak, K., Thornton, C.A., Swanson, M.S., 2014. Loss of MBNL leads
to disruption of developmentally regulated alternative polyadenylation in RNA-
mediated disease. Mol. Cell 56, 311e322.
Berkovits, B.D., Mayr, C., 2015. Alternative 30 UTRs act as scaffolds to regulate
membrane protein localization. Nature 522, 363e367.
Bernstein, D.S., Buter, N., Stumpf, C., Wickens, M., 2002. Analyzing mRNA-protein
complexes using a yeast three-hybrid system. Methods 26, 123e141.
Bischof, J., Maeda, R.K., Hediger, M., Karch, F., Basler, K., 2007. An optimized trans-
genesis system for Drosophila using germ-line-specific phiC31 integrases. Proc.
Natl. Acad. Sci. U. S. A. 104, 3312e3317.
Bowles, J., Knight, D., Smith, C., Wilhelm, D., Richman, J., Mamiya, S., Yashiro, K.,
Chawengsaksophak, K., Wilson, M.J., Rossant, J., Hamada, H., Koopman, P., 2006.
Retinoid signaling determines germ cell fate in mice. Science 312, 596e600.
Chen, D., DM, M., 2003. A discrete transcriptional silencer in the bam gene de-
termines asymmetric division of the Drosophila germline stem cell. Develop-
ment 130, 1159e1170.
Chen, D., McKearin, D., 2005. Gene circuitry controlling a stem cell niche. Curr. Biol.
15, 179e184.
Chen, D., Wu, C., Zhao, S., Geng, Q., Gao, Y., Li, X., Zhang, Y., Wang, Z., 2014. Three
RNA binding proteins form a complex to promote differentiation of germline
stem cell lineage in Drosophila. PLoS Genet. 10, e1004797.
de Klerk, E., Venema, A., Anvar, S.Y., Goeman, J.J., Hu, O., Trollet, C., Dickson, G., den
Dunnen, J.T., van der Maarel, S.M., Raz, V., t Hoen, P.A., 2012. Poly(A) binding
protein nuclear 1 levels affect alternative polyadenylation. Nucleic Acids Res.
40, 9089e9101.
Derti, A., Garrett-Engele, P., Macisaac, K.D., Stevens, R.C., Sriram, S., Chen, R.,
Rohl, C.A., Johnson, J.M., Babak, T., 2012. A quantitative atlas of polyadenylation
in five mammals. Genome Res. 22, 1173e1183.
Di Giammartino, D.C., Nishida, K., Manley, J.L., 2011. Mechanisms and consequences
of alternative polyadenylation. Mol. Cell 43, 853e866.
Eberhart, C.G., Maines, J.Z., Wasserman, S.A., 1996. Meiotic cell cycle requirement for
a fly homologue of human Deleted in Azoospermia. Nature 381, 783e785.
Fu, Z., Geng, C., Wang, H., Yang, Z., Weng, C., Li, H., Deng, L., Liu, L., Liu, N., Ni, J.,
Xie, T., 2015. Twin promotes the maintenance and differentiation of germline
stem cell lineage through modulation of multiple pathways. Cell Rep. 13,
1366e1379.
Honigberg, S.M., Purnapatre, K., 2003. Signal pathway integration in the switch
from the mitotic cell cycle to meiosis in yeast. J. Cell Sci. 116, 2137e2147.
Hoque, M., Ji, Z., Zheng, D., Luo, W., Li, W., You, B., Park, J.Y., Yehia, G., Tian, B., 2013.
Analysis of alternative cleavage and polyadenylation by 30 region extraction and
deep sequencing. Nat. Methods 10, 133e139.
Insco, M.L., Bailey, A.S., Kim, J., Olivares, G.H., Wapinski, O.L., Tam, C.H., Fuller, M.T.,
2012. A self-limiting switch based on translational control regulates the tran-
sition from proliferation to differentiation in an adult stem cell lineage. Cell
Stem Cell 11, 689e700.
Insco, M.L., Leon, A., Tam, C.H., McKearin, D.M., Fuller, M.T., 2009. Accumulation of a
differentiation regulator specifies transit amplifying division number in an
adult stem cell lineage. Proc. Natl. Acad. Sci. U. S. A. 106, 22311e22316.
Jenal, M., Elkon, R., Loayza-Puch, F., van Haaften, G., Kuhn, U., Menzies, F.M., Oude
Vrielink, J.A., Bos, A.J., Drost, J., Rooijers, K., Rubinsztein, D.C., Agami, R., 2012.
The poly(A)-binding protein nuclear 1 suppresses alternative cleavage and
polyadenylation sites. Cell 149, 538e553.
Ji, Z., Lee, J.Y., Pan, Z., Jiang, B., Tian, B., 2009. Progressive lengthening of 30 un-
translated regions of mRNAs by alternative polyadenylation during mouse
embryonic development. Proc. Natl. Acad. Sci. U. S. A. 106, 7028e7033.
Ji, Z., Tian, B., 2009. Reprogramming of 30 untranslated regions of mRNAs by
alternative polyadenylation in generation of pluripotent stem cells from
different cell types. PLoS One 4, e8419.
Jin, Z., Kirilly, D., Weng, C., Kawase, E., Song, X., Smith, S., Schwartz, J., Xie, T., 2008.
Differentiation-defective stem cells outcompete normal stem cells for niche
occupancy in the Drosophila ovary. Cell Stem Cell 2, 39e49.
Kim, D., Pertea, G., Trapnell, C., Pimentel, H., Kelley, R., Salzberg, S.L., 2013. TopHat2:
accurate alignment of transcriptomes in the presence of insertions, deletions
106 L. Shan et al. / Journal of Genetics and Genomics 44 (2017) 95e106
and gene fusions. Genome Biol. 14, R36.
Kim, S., Yamamoto, J., Chen, Y., Aida, M., Wada, T., Handa, H., Yamaguchi, Y., 2010.
Evidence that cleavage factor Im is a heterotetrameric protein complex con-
trolling alternative polyadenylation. Genes Cells 15, 1003e1013.
Koubova, J., Menke, D.B., Zhou, Q., Capel, B., Griswold, M.D., Page, D.C., 2006. Reti-
noic acid regulates sex-specific timing of meiotic initiation in mice. Proc. Natl.
Acad. Sci. U. S. A. 103, 2474e2479.
Kubo, T., Wada, T., Yamaguchi, Y., Shimizu, A., Handa, H., 2006. Knock-down of 25
kDa subunit of cleavage factor Im in Hela cells alters alternative polyadenylation
within 30 -UTRs. Nucleic Acids Res. 34, 6264e6271.
Lackford, B., Yao, C., Charles, G.M., Weng, L., Zheng, X., Choi, E.A., Xie, X., Wan, J.,
Xing, Y., Freudenberg, J.M., Yang, P., Jothi, R., Hu, G., Shi, Y., 2014. Fip1 regulates
mRNA alternative polyadenylation to promote stem cell self-renewal. EMBO J.
33, 878e889.
Li, C.Y., Guo, Z., Wang, Z., 2007. TGFbeta receptor saxophone non-autonomously
regulates germline proliferation in a Smox/dSmad2-dependent manner in
Drosophila testis. Dev. Biol. 309, 70e77.
Li, W., Park, J.Y., Zheng, D., Hoque, M., Yehia, G., Tian, B., 2016. Alternative cleavage
and polyadenylation in spermatogenesis connects chromatin regulation with
post-transcriptional control. BMC Biol. 14, 6.
Lin, Y., Gill, M.E., Koubova, J., Page, D.C., 2008. Germ cell-intrinsic and -extrinsic
factors govern meiotic initiation in mouse embryos. Science 322, 1685e1687.
Mangone, M., Manoharan, A.P., Thierry-Mieg, D., Thierry-Mieg, J., Han, T.,
Mackowiak, S.D., Mis, E., Zegar, C., Gutwein, M.R., Khivansara, V., Attie, O.,
Chen, K., Salehi-Ashtiani, K., Vidal, M., Harkins, T.T., Bouffard, P., Suzuki, Y.,
Sugano, S., Kohara, Y., Rajewsky, N., Piano, F., Gunsalus, K.C., Kim, J.K., 2010. The
landscape of C. elegans 30 UTRs. Science 329, 432e435.
Martin, G., Gruber, A.R., Keller, W., Zavolan, M., 2012. Genome-wide analysis of pre-
mRNA 30 end processing reveals a decisive role of human cleavage factor I in the
regulation of 30 UTR length. Cell Rep. 1, 753e763.
Masamha, C.P., Xia, Z., Yang, J., Albrecht, T.R., Li, M., Shyu, A.B., Li, W., Wagner, E.J.,
2014. CFIm25 links alternative polyadenylation to glioblastoma tumour sup-
pression. Nature 510, 412e416.
Mayr, C., 2016. Evolution and biological roles of alternative 30 UTRs. Trends Cell Biol.
26, 227e237.
Mayr, C., Bartel, D.P., 2009. Widespread shortening of 30 UTRs by alternative cleavage
and polyadenylation activates oncogenes in cancer cells. Cell 138, 673e684.
McKearin, D., Ohlstein, B., 1995. A role for the Drosophila Bag-of-marbles protein in
the differentiation of cystoblasts from germline stem cells. Development 121,
2937e2947.
Norbury, C.J., 2013. Cytoplasmic RNA: a case of the tail wagging the dog. Nat. Rev.
Mol. Cell Biol. 14, 643e653.
Ohlstein, B., Lavoie, C.A., Vef, O., Gateff, E., McKearin, D.M., 2000. The Drosophila
cystoblast differentiation factor, benign gonial cell neoplasm, is related to
DExH-box proteins and interacts genetically with bag-of-marbles. Genetics 155,
1809e1819.
Ozsolak, F., Kapranov, P., Foissac, S., Kim, S.W., Fishilevich, E., Monaghan, A.P.,
John, B., Milos, P.M., 2010. Comprehensive polyadenylation site maps in yeast
and human reveal pervasive alternative polyadenylation. Cell 143, 1018e1029.
Parker, R., Song, H., 2004. The enzymes and control of eukaryotic mRNA turnover.
Nat. Struct. Mol. Biol. 11, 121e127.
Pinto, P.A.B., Henriques, T., Freitas, M.O., Martins, T., Domingues, R.G.,
Wyrzykowska, P.S., Coelho, P.A., Carmo, A.M., Sunkel, C.E., Proudfoot, N.J.,
Moreira, A., 2011. RNA polymerase II kinetics in polo polyadenylation signal
selection. EMBO J. 30, 2431e2444.
Sandberg, R., Neilson, J.R., Sarma, A., Sharp, P.A., Burge, C.B., 2008. Proliferating cells
express mRNAs with shortened 30 untranslated regions and fewer microRNA
target sites. Science 320, 1643e1647.
Shepard, P.J., Choi, E.A., Lu, J., Flanagan, L.A., Hertel, K.J., Shi, Y., 2011. Complex and
dynamic landscape of RNA polyadenylation revealed by PAS-Seq. RNA 17,
761e772.
Shi, Y., 2012. Alternative polyadenylation: new insights from global analyses. RNA
18, 2105e2117.
Smibert, P., Miura, P., Westholm, J.O., Shenker, S., May, G., Duff, M.O., Zhang, D.,
Eads, B.D., Carlson, J., Brown, J.B., Eisman, R.C., Andrews, J., Kaufman, T.,
Cherbas, P., Celniker, S.E., Graveley, B.R., Lai, E.C., 2012. Global patterns of tissue-
specific alternative polyadenylation in Drosophila. Cell Rep. 1, 277e289.
Sun, Y.C., Cheng, S.F., Sun, R., Zhao, Y., Shen, W., 2014. Reconstitution of gameto-
genesis in vitro: meiosis is the biggest obstacle. J. Genet. Genomics 41, 87e95.
Suzuki, A., Hirasaki, M., Hishida, T., Wu, J., Okamura, D., Ueda, A., Nishimoto, M.,
Nakachi, Y., Mizuno, Y., Okazaki, Y., Matsui, Y., Izpisua Belmonte, J.C., Okuda, A.,
2016. Loss of MAX results in meiotic entry in mouse embryonic and germline
stem cells. Nat. Commun. 7, 11056.
Takagaki, Y., Manley, J.L., 1998. Levels of polyadenylation factor CstF-64 control IgM
heavy chain mRNA accumulation and other events associated with B cell dif-
ferentiation. Mol. Cell 2, 761e771.
Takagaki, Y., Seipelt, R.L., Peterson, M.L., Manley, J.L., 1996. The polyadenylation
factor CstF-64 regulates alternative processing of IgM heavy chain pre-mRNA
during B cell differentiation. Cell 87, 941e952.
Trapnell, C., Hendrickson, D.G., Sauvageau, M., Goff, L., Rinn, J.L., Pachter, L., 2013.
Differential analysis of gene regulation at transcript resolution with RNA-seq.
Nat. Biotechnol. 31, 46e53.
Ulitsky, I., Shkumatava, A., Jan, C.H., Subtelny, A.O., Koppstein, D., Bell, G.W., Sive, H.,
Bartel, D.P., 2012. Extensive alternative polyadenylation during zebrafish
development. Genome Res. 22, 2054e2066.
Wahle, E., Winkler, G.S., 2013. RNA decay machines: deadenylation by the Ccr4-not
and Pan2-Pan3 complexes. Biochim. Biophys. Acta 1829, 561e570.
Wolf, J., Passmore, L.A., 2014. mRNA deadenylation by Pan2-Pan3. Biochem. Soc.
Trans. 42, 184e187.
Wu, X., Liu, M., Downie, B., Liang, C., Ji, G., Li, Q.Q., Hunt, A.G., 2011. Genome-wide
landscape of polyadenylation in Arabidopsis provides evidence for extensive
alternative polyadenylation. Proc. Natl. Acad. Sci. U. S. A. 108, 12533e12538.
Xu, E.Y., Lee, D.F., Klebes, A., Turek, P.J., Kornberg, T.B., Reijo Pera, R.A., 2003. Human
BOULE gene rescues meiotic defects in infertile flies. Hum. Mol. Genet. 12,
169e175.
Zhao, S., Chen, D., Geng, Q., Wang, Z., 2013. The highly conserved LAMMER/CLK2
protein kinases prevent germ cell overproliferation in Drosophila. Dev. Biol. 376,
163e170.
Zheng, D., Tian, B., 2014. RNA-binding proteins in regulation of alternative cleavage
and polyadenylation. Adv. Exp. Med. Biol. 825, 97e127.