pschreiner_DissertationProposal

A Genomic Exploration of piRNA-mediated Deadenylation of
Protein-coding Genes
Patrick Schreiner
Genetics, Genomics, and Bioinformatics
Dissertation Research Proposal
8/25/14

Abstract
The Piwi-interacting sRNA (piRNA) subclass is the least characterized of the small
non-coding RNAs. piRNAshave been shown to function in suppressing transposable
elements (TEs) via Argonaute-mediated transcript slicing andepigenetic modification.
Further, the pi-RNA induced silencing complex (RISC) has been shown to have the ability
to regulate protein-coding genes as well. A complex involving PIWI proteins – Aub and
Ago3 – along with Smaug, and the CCR4deadenylase has been shown to associate with
target transcripts in a sequence-specific manner within the 3’ UTRto promote poly(A)
shortening, resulting in downstream transcript degradation. Although piRNA-mediated
deadenylation has been demonstrated in a protein-coding gene in Drosophila melanogaster,
Nanos,this mechanism of regulation has yet to be thoroughly explored on a genome-wide
scale. Our goal is to further characterize protein-coding gene targets of piRNA-
mediatedregulation via deadenylation by identifying several independent, genomic factors
indicative of this interaction.
Introduction
Transposable elements (TEs) are mobile genetic elements that encode the ability to
duplicate themselves within a genome (McClintock 1953). Although there are two classes of
TEs, both mechanisms of transposition produce RNA, albeit performing unique functions, to
facilitate their movement(Craig 2002). Unregulated transposition of TEs can cause integration
of the TE sequence in disadvantageous regions, which could yield phenotypic aberrations within
the organism (Levin et al., 2011). Therefore, a regulatory mechanism – via sRNAs – is crucial in
many species for proper development (Malone et al., 2009;Guoet al.,2013).

Small RNAs (sRNAs) are important non-protein coding RNAs that can regulate activity
within the cell. sRNAs can be classified into one of three major categories in species within
theAnimalia kingdom: small interfering-RNA (siRNA), microRNA (miRNA), or PIWI-
interacting RNA (piRNA). The Plantae kingdom sRNA system appears to have evolved
independently of Animalia (Millar 2005). The three subclasses of sRNA in the Animalia
kingdomdiffer in terms oftheir respective size, target specificity, and mechanisms of biogenesis
(Chung et al.,2008, Siomiet al.,2010).We are particularly interested in studying the least
thoroughly characterized among the subclasses of sRNA, the piRNA system.
The functionality of piRNAs has been shown to play an essential role in fertility and in
the maintenance of genome integrity. The foundation for this observation that was set decades
ago with the attempted cross of a female fly from a laboratory inbred line with a male fly taken
from the wild. The offspring of this crossyielded hybrid dysgenesis: sterility due to high
mutation rate, chromosomal rearrangement, or recombination (Bingham et al.,1982). On the
other hand, when a wild female fly was crossed with a male fly from a laboratory inbred line, a
normal, fertile phenotype was observed (Bregliano et al., 1980). More recent studies have
concluded that this phenomenon is likely due to the naivety of the laboratory female to the wild
male’s active TEs (Brennecke et al., 2008). Without maternal deposition of piRNAs, the
progeny was unable to inhibit the movement of active, genomic TEs. Therefore, it was shown
that the piRNA mechanism of genome defense is inherited maternally in offspring, rather than
paternally in Drosophila melanogaster, and that piRNA-mediated silencing of the active TEs
present in the progeny is necessary for proper development (Bingham et al., 1982).
piRNAs are sRNAs that function via association with members of the PIWI protein
family. piRNAs represent a subset of small RNAs that generally range between 24-32

nucleotides (nts) in length. Their mechanism of biogenesis is unique from that of other sRNAs
in that it is Dicer-independent. Rather, piRNAs can be classified according to their primary or
secondary mechanism of biogenesis (Guzzardoet al.,2013;Guoet al.,2013;Kharunaet al.,2010).
Those piRNAs generated from the primary mechanism of biogenesis are referred to as primary
piRNAs. Due to the observation that primary piRNAs predominately originate from the
antisense strand, it has been hypothesized that primary piRNAs originate from one long
transcript. Thislong transcript is thought to originate from a particular region, or several
particular heterochromatic genomic loci referred to as piRNA clusters. piRNA clusters can span
over 100,000 bps, depending on the organism, consisting mainly of fragments of inactivated
transposable elements (Ranganet al.,2011;Thomson et al.,2009;Guoet al.,2013;Kawaokaet
al.,2013;Kharunaet al.,2010). The transcript then undergoes downstream processing by other
members within the piRNA system in order to generate mature piRNAs, which are necessary for
the functionality of the system (Ishizuet al.,2012;Guoet al.,2013;Guzzardoet al.,2013).
Although many of the contributors to piRNA biogenesis are still under investigation,
several important factors have been well-described. An endonuclease, Zucchini, slices thelong,
intialtranscriptthat has originated from a piRNA cluster, into many fragments (Ishizu et al.,
2012;Guoet al., 2013;Guzzardoet al., 2013). The 3’ end of these fragments then undergoes a 2’-
O-methyl modification. This modification, modulated by HEN1,protects the 3’ end of the
piRNA from degradation or modification (Montgomery et al., 2012). Those piRNAs that have
been sliced to their proper length and whose 3’ end has been modified are referred to as ―mature
piRNAs‖(Kharunaet al.,2010; Thomson et al.,2009).
Mature piRNAs then associate with PIWI proteins, a subfamily of the Argonaute family
of proteins, to form what is referred to as a RNA-induced silencing complex (RISC). Among the

PIWIs, there are three proteins of particular importance to the piRNA system: Argonaute 3
(Ago3), Aubergine (Aub), and Piwi. These proteins are differentiable by their subcellular
localization, mechanism of repression, as well as their piRNA association (Table 1).
Table 1| Piwi Protein Comparison. Argonaute 3 (AGO3), and Aubergine (AUB) Comparison
in Drosophila melanogaster.
Similarly, those piRNAs that are generated via the secondary mechanism of biogenesis
are referred to as secondary piRNAs. Secondary piRNAs are generated as a product of RISC
slicing of target mRNAs via what has been termed the amplification loop. The amplification
loop, also referred to as the ping-pong cycle, generates piRNAs via PIWI protein-mediated
cleavage of target transcripts(Ishizuet al.,2012;Guzzaroet al.,2013). The PIWI domain of the
associated PIWI family proteinfacilitatesthe slicing of its target mRNA (Ishizuet
al.,2012).Primary piRNAs, generally from the antisense strand, associate with Aub which guides
the complex to its target mRNA molecule and results in Argonaute-mediated slicing of the
target. The fragments of sense transcript can then associate with Ago3, which then continues
slicing via the same mechanism on other targets complementary to the associated secondary
piRNA. In this manner, the secondary mechanism of piRNA biogenesis provides a positive
feedback loop of piRNA generation(Robineet al., 2009).
Although the mechanism of piRNA-mediated silencing was initially thought to occur
strictly via target slicing, it was then shown that the nuclear PIWI protein, Piwi, can induce
Piwi Aub Ago3
Localization Nuclear Cytoplasmic Cytoplasmic
Regulation Epigenetic Post- Transcriptional Post- Transcriptional
Strand of Associated
piRNA Origin
Antisense Antisense Sense
Associated piRNA
Bias
U at position 1 U at position 1 A at position 10

epigenetic modifications. In fact, the nuclear Piwi protein has been shown to be capable of
performing its function independent of its slicing, PIWIdomain (Darricarrereet al., 2013). Piwi
induces epigenetic modifications via association with other proteins. Piwi has been shown to
recruit Heterochromatin Protein 1a (HP1a)and histone methyltransferase, Su(var)3-9, and to
induce H3K9me2/3 enrichment(Brower-Toland et al., 2007). HP1a, guided by the Piwi-RISC,
can bind to histones, where Su(var)3-9 can induce methylation at the H3K9. Histone
methylationat a particular site results in an even more probable binding site for further HP1a
interaction (Huang et al., 2013). The association of HP1a, as the name implies, is known to
induce heterochromatin formation(James et al., 1986). Therefore, increased heterochromatin at
Piwi-RISC guided loci leads to reduced association of RNA polymerase II. It has been well
established that reduced RNA polymerase II association is correlated with lower levels of
transcription (Sims et al., 2004).
Outside of their role in TE repression, piRNAs have recently been implicated in a role
within protein-coding gene regulation. Protein-coding gene regulation mediated by piRNAs is
generally thought to occur, much like the epigenetic mechanism of repression, via PIWI family
proteins’association with other functional proteins.For example, piRNA-mediated deadenylation,
as shown in the model of the Nanos 3’UTR,requires the association of Aub and Ago3in a
common complex with Smaug: an RNA-binding protein (RBP) that facilitates the recruitment of
other machinery necessary for deadenylation.The mechanism of piRNA-mediated deadenylation
occurs by piRNA association with a PIWI protein, either Aub or Ago3, leading to the
recruitment of the Smaug RBP and CCR4 3’-5’ exonuclease (Rouget et. al., 2010). The CCR4
exonucleaseis then in a position to associate with the poly(A) tail of the transcript, leading to
poly(A) tail shortening.Transcripts with a shortenedpoly(A) tails are recognized and lead to

degradation (Salleset. al., 1999). It is important to note that although both miRNAs and piRNAs
have been shown to affect deadenylation in Nanos in Drosophila melanogaster, it has also been
demonstrated that the regulatory complex can bind to its target, to a lesser extent, in the absence
of RNA (Pinderet. al., 2012).This observation is indicative of the sRNA’s complementary, but
not absolute, role in promoting transcript deadenylation.
Objectives
Aim 1: Perform a genomic investigation of genes potentially regulated via piRNA-mediated
deadenylation
It has been shown that piRNA association within a protein complex is necessary to
promote deadenylation in Nanos in Drosophila melanogaster, but this analysis did not return a
comprehensive set of genes under this regulatory constraint (Rouget et al., 2010). We set to
explore the possibility of other protein-coding genes that may also be affected by a similar
mechanism of regulation by utilizing several independent factors related to the structure of the
Nanos 3’ UTR. The factors that we assess include transcript depletion in early embryos, an
accumulation of TE-derived piRNAs that exhibit a region of complementarity to a specific 3’
UTR, mapping of a potentially functional TE-derived piRNA to a defined piRNA cluster, as well
as the 3’ UTR of the gene of interest’s structural consistency with the Nanos 3’ UTR.
Aim 2: Assess Smaug as an important indicator of deadenylation in early Drosophila
melanogaster embryos
Smaug, in cooperation with Aub and Ago3, has been shown to be essential in recruiting
the CCR4 deadenylase to transcripts to promote poly(A) tail shortening (Dahanukar et al., 1999).
Therefore, factors of Smaug association and regulation of transcripts may also be useful in the

assessment of genes that undergo piRNA-mediated deadenylation. Using additional factors to
those described in Aim I, we further assess potential targets of piRNA-mediated deadenylation
by the presence of a Smaug recognition element (SRE), direct transcript interaction with Smaug,
and transcript derepression in the absence of a functional Smaug protein.
Research Progress
Aim 1: Perform a genomic investigation of potential piRNA-mediated deadenylation in
Drosophila melanogaster
The only previousstudy regarding piRNA-mediated deadenylation was investigated
strictly in one gene, Nanos, in Drosophila melanogaster. We are interested in the assessment of
other potential genes that may be under a similar mechanism of piRNA-mediated deadenylation.
Nearly 30 million piRNAs, from seven publicly available sRNA libraries from the
previous investigation of Nanos piRNA-mediated deadenylation in earlyDrosophila
melanogaster embryos, 0-2 hours old,were used to assess the piRNA landscape (Leinonenet
al.,2010; Rouget et al., 2010). piRNA identification within these sRNA libraries is not trivial,
though.
miRNAsgenerally have anindicative 21-23 nucleotide length profile, depending on the
species of interest. piRNAs, however,typically have a more broad length spectrum. Primary
piRNAs tend to exhibit anuracil (U) at position 1, and secondary piRNAs tend to exhibit an
adenine (A) at position 10. These biases exist due to the mechanism of piRNA biogenesis via
the primary pathway and amplification loop, respectively. These biases are not indicative
enough of a piRNA to appropriately filter to all piRNAs, though. That is, many piRNAs exist
that do not show these biases. Therefore, the only sRNA filter that can be used to

predictpiRNAs is by the use of a sRNA length cutoff value. piRNAstend to be identified as
longer, generally at 24-32nts, in comparison to the other types of sRNA found in Drosophila
melanogaster. In our research, any sRNA that has a length greater than 24 nts is considered a
putative piRNA.
As a proof of concept that we can detect functional piRNAs within genomic 3’ UTRs, we
first aimed to confirm the presence of those previously identified functional piRNA with the 3’
UTR of Nanos within our genomic screen. We used the BLAST algorithm to assess piRNAs
that have a region of complementarity to genomic 3’ UTRs. We ran a NCBI BLAST with a
word size of 14 and an E-value of 100, as described in the discovery of the functional piRNAs
complementary to the Nanos 3’ UTR (Rouget et al., 2010). The piRNA motif mediating target
specificity has been shown to exist anywhere within the piRNA sequence, rather than strictly the
2-8 nucleotide seed region that is generally observed in miRNAs. Therefore, any piRNA
motifthat is complementary to a 3’ UTR, as returned in the BLAST search, was considered as a
potential guide of Aub or Ago3-RISC transcriptassociation.
Further, the functional piRNAs observed in the Nanos 3’ UTR were associated with the
TEs from which they had derived (Rouget et al., 2010). We acknowledge that all piRNAs do not
appear to be derived from TEs, but for the purposes of this analysis, we only considered those
piRNAs that can be mapped to a known TE in Drosophila melanogaster. TE sequences were
extracted from RepBase(Jurka 1998). We then performed a Bowtie2 alignment, with default
parameters, of all piRNA sequences to the sequences of all TEs in Drosophila melanogaster
(Langmeadet al., 2014). We defined those piRNAs that uniquely mapped to aDrosophila
melanogasterTE sequence as a TE-derived piRNA.

Those piRNAs that were both TE-derived and contain a significant region of
complementarity to a 3’ UTR wereused in downstream analysis. This analysis returned a gene
list of about 4,000 genes with over 500,000 TE-derivedpiRNAs with regions of complementarity
to a 3’ UTR [data not shown].
We thensubset theTE-derived piRNAs to specifically analyze the Nanos 3’ UTR for the
presence of previously established, functional, TE-derived piRNAs. For the functional piRNAs
deriving from the 412 LTR retrotransposon of the Gypsy family, we returned 50piRNAs that
contain the functional motif(Fig. 1A). Our analysis of 412-derived piRNAsto the Nanos 3’ UTR
disagreed with the previously published values of 75 functional piRNAs (Rouget et al.,2010).We
also returned about 600 other piRNAs that are TE-derived, have a region of complementarity to
the Nanos 3’ UTR, and map to a defined piRNA cluster in Drosophila melanogaster (Fig. 1B).
However, many of these TE-derived piRNAs are not downstream of the Smaug recognition
element (SRE), and therefore are unlikely to function in promoting deadenylation. The function
of these additional piRNAs is unknown. Since we made parameter calls as described in the
previous research, we attributed this difference to variation deriving from software
improvements within the local alignments and read mapping tools utilized and/or the improved
annotation of Drosophila melanogaster TEs. With that said, since these parameters were able to
generate a significant hit of the known functional piRNA motif, we deemed these parameters
sufficient to detect potential piRNA complementarity and can be utilized within the workflow.
Once the previous established functional piRNAs were detected, we were then confident
in an extrapolation of the workflow to detect genes that may be acted upon by a similar
regulatory mechanism (Fig. 2).

Figure 1 | Nanos 3’ UTR piRNA Landscape.(A)Validation of the known, functional 412
transposable element-derived piRNAs to the Nanos 3’ UTR. The gray regions are the defined
SREs within the TCE of the Nanos 3’ UTR. Sense (blue) and antisense (red) 412-derived
piRNAs are represented as having a region of complementarity to the Nanos 3’ UTR. (B)
DMRT1B transposable element-derived piRNAs that map to a defined piRNA cluster, but are
not downstream of the SREs.

Since not every TE-derived piRNA with complementarity to a 3’ UTR has an easily
ascertained function, we propose gene target filtering based on other independent factors
indicative piRNA-mediated deadenylation. Further, as there is no published data regarding the
sequences to which the Aub or Ago3 RISC complexes associate, it is most advantageous to
initially filter out those genes whose 3’ UTRs appear to be incapable of regulation via piRNA-
mediated deadenylation. Factors indicative of anincapability to undergo piRNA-mediated
regulation via deadenylation includeno apparent degradation of transcript in early embryos, a
lack of TE-derived piRNA accumulation in a defined 3’ UTR,and a lack of complementary TE-
derived piRNAs that can be mapped to a defined piRNA cluster in Drosophila melanogaster.
Previous investigation of Nanos has shown that Smaug is present and able to degrade
transcript between 0-2 hr and 2-4 hrembyos in Drosophila melanogaster(Rouget et al., 2010).
We assess the degree of degradation between 0-2hr embryos and 2-4 hr embryos based upon
RNA-seqexpression profile data from the modENCODE group, as seen on FlyBase(St. Pierre et
al., 2014). As the Nanos expression profile has an RPKM-normalized expression profile of 243
transcripts in 0-2 hr embryo, and 22 in 2-4 hr embryos, we explore other genes withsimilar,
significant transcript during this timeframe(Fig. 4)(Celniker et al., 2009). With that said, we
acknowledge that a larger hairpin within the secondary structure of the 3’ UTR, as well as other,
potentially unknown factors, may influence the rate of degradation as well. Therefore, we do
consider the possibility of genes being degraded by this mechanism to various extents (Laver et
al., 2013).
In the established model of piRNA-mediated deadenylation, TE-derived
piRNAscomplementarity to the Nanos 3’ UTR are shown to guide the piRNA-mediated
deadenylation complex to target transcripts. Therefore, we explore other genes of interest by the

observation ofa substantial number of TE-derived piRNAs that have a region of complementary
to theirspecific 3’ UTRs. In this analysis, we only considered genes that had more than 300 TE-
derived piRNAs complementary to their 3’ UTR.
Finally, we utilize the default settings of Bowtie2 to map the TE-derived piRNAs, with
regions of complementarity to a specific 3’ UTR, to sequences that have been defined as piRNA
clusters in Drosophila melanogaster. We extracted genomic loci from the piRNA Bank that
define regions of the genome that have been considered piRNA clusters (Lakshmi et al., 2008).
The mapping ofTE-derived piRNAs to a specific, defined piRNA cluster provides an indication
for the origin of the piRNAs.
Taken together, these independent factors returned a subset of 54 genes that appear to be
regulated via piRNA-mediated deadenylation. Probabilities have been calculated to represent the
chance that a gene satisfies each independent factor (Table 2). Further investigation of the genes
that satisfied these filters is required to prioritize the functional assay of potential piRNA-
mediated deadenylation. Additional factors such as expression profiles, piRNA sequence, and
putative function can be useful in prioritizing the assay of these genes (Fig. 2). We also plan to
pursue a Gene Ontology term enrichment analysis to assess the possibility of functional links
between the genes returned in this analysis. That is, the investigation regarding whether or not
specific groups of genes, in terms of function, are more susceptible to piRNA-mediated
deadenylation relative to other groups.

Independent Factors
Number of
hits
Total Hits
Possible
Probability of
Occurrence
Transcript Depletion in Early
Embryos
286 17,262 1.66%
Presence of >300 TE-derived
piRNAs complementarity to a
specific 3' UTR
15,888 30,277 52.48%
TE-derived piRNAs
complementary to genic 3' UTR
also maps to defined piRNA
cluster in Drosophila
melanogaster
1,855 5,548 33.44%
Probability of all 3 independent
factors
2.91E-3
Table 2 | Smaug-Independent Filters.Probability that a gene or transcript satisfies each of the
Smaug-Independentfilter criteria considered individually, and combined.

Figure 2 | Smaug-Independent Workflow.The series of filtersregarding the prediction of
piRNA functionality in facilitating deadenylation of protein-coding transcripts.
Aim 2: Assess Smaug as an important regulator of the piRNA system in early Drosophila
melanogaster embryos
The RNA-binding protein, Smaug, has been shown to immunoprecipitate (IP) with other
proteins that are supposed to be in the same complex: Aub, Ago3, CAF1, and CCR4 (Rouget et
al., 2010). Smaug, like most RBPs, has many targets to which it associates and represses.
Smaug’s ability to bind transcripts and guide protein complexes to target transcripts potentially
renders it an interesting asset to this analysis. We acknowledge, though, that the guiding of this
deadenylase complex to target transcripts may be a result of strictly Smaug binding, strictly
piRNA binding, or a cooperative effort amongst the two. Previous data suggests that piRNAs
are likely playing a supplementary role in the recruitment of the deadenylase complex (Rouget et

al., 2010). Therefore, building upon our previous analysis (Aim I), we further assess genes of
interest based on factors indicative of Smaug interaction and repression.
In this analysis, we further assess potential targets of piRNA-mediated deadenylation
bySmaug-dependent factors such as the presence of a SRE, overrepresentation of Smaug binding
to target mRNA in immunoprecipitations (IPs), and transcript recovery in the absence of
functional Smaug.
The first step in the workflow is to identify 3’ UTRs that contain SREs. A SREs is a 5-
mer region within the 3’ UTR, generally towards the 5’ end of the 3’ UTR, which is indicative of
Smaug association. In the Nanos 3’UTR, there are two SREs, each of the sequence, CTGGC.
Both of the SREs are contained in the Nanos translational control elements, which is found in the
5’ region of the 3’ UTR. Only one of these SREs appears to be necessary for the recruitment of
the piRNA-mediated deadenylase complex (Crucs et al., 2000). The TE-derived piRNAs, with a
region of complementarity to the 3’ UTR, exist downstream of the SREs in the Nanos model.
Given that RNA has the ability to fold into proper conformation, we further consider any targets
that contain an accumulation of TE-derived piRNAs at any region downstream of the SRE.
It is important to note that through a comparative analysis of SRE conservation within the
other drosopholids’ Nanos 3’ UTRs, as well as evidence from recent literature, suggests that
SREs may have a more general motif, CNGGN [data not shown] (Clark et al., 2007; Chen et al.,
2014). We plan to reassess genomic 3’ UTRs in a similar manner, incorporating the detection of
this more general SRE motif.
Next, we investigate previous research that has shown that Smaug binding to 3’ UTRs of
transcripts facilitates target degradation (Dahanukar et al., 1999). There exists data regarding the

transcript targets of Smaug binding and their expression levels in early embryos in the absence of
a functional Smaug protein (Chen et al., 2014). Smaug IPs, followed by RNA elutions, were
performed to assess a transcriptome-wide scale overrepresentation of Smaug association relative
to control IPs. Also, independently, a transcriptome-wide assay in which the RNA-binding
domain of Smaug was made non-functional, and transcript level within the cells was assessed.
With these data, we can further assess our genes of interest on whether or not the transcripts of
interest were shown to significantly associate with Smaug, as well as whether or not the gene of
interest’s transcripts recovered in the absence of functional Smaug. Presumably, in the absence
of functional Smaug, there is an absence of this mechanism of deadenylation.
We have calculated the probability that a particular gene would satisfy each particular,
Smaug-dependentcriterion. Assuming that each of these factors occurs independently of one
another, we also return the probability that a gene would satisfy all of the six criteria (Aims I &
II) strictly by chance (Table 3).
When applying the six independent filters to all transcripts in Drosophila melanogaster,
only two genes were returned: Nanos and CG5010. As piRNA-mediated deadenylation has
already been described in Nanos, we focused on further exploration of CG5010. The function of
CG5010 has not been previously identified.

Independent
Factors
Number of
hits
Total Hits
Possible
Probability of
Occurrence
Presence of SRE in 3'
UTR
2,976
(CNGGN: 10,943)
30,277
9.83%
(36.14% with more
general motif: CNGGN)
Overrepresented Smaug
Association
312 17,262 1.81%
Transcript Recovery in
the Absence of Functional
Smaug
1,814 17,262 10.51%
Probability of all 6
independent factors
5.45E-7
Table 3 | Smaug-dependent Filters. Probability that a gene or transcript satisfies each of the
Smaug-dependentfilter criteria considered individually. The final probability calculation takes
into account both Smaug-independent and Smaug-dependent filters.
Figure 3|Smaug-dependent Workflow.The series of filters regarding the Smaug-dependent
prediction of piRNA functionality in facilitating deadenylation of protein-coding transcripts.

We performed further investigation regarding whether or notthe CG5010 expression
profile is consistent with the Nanos profile. Although, Nanos transcript is more abundant in
Drosophila melanogaster embryos 0-2 hrs relative to CG5010, both Nanos and CG5010
transcript abundance falls nearly 90% in 2-4 hr embryos (Fig. 4). Since initial transcript
abundance presumably has little to do with the recruitment of the deadenylase complex, yet
transcript abundance falls on a similar scale, we concluded that the CG5010 expression profile is
consistent with the Nanos profile in terms of having the capabilityof being regulated via a similar
mechanism.
Figure 4 |
Expression Profile Comparison.Bar plot comparing the RPKM values of transcript expression
of Nanos and CG5010 in early embryos.

CG5010 TE-derived piRNAs complemented the 3’ UTR downstream of the SRE and
mapped to a non-LTR, Jockey family retrotransposon named FW (Fig. 5). It is worth noting that
the previously established functional piRNAs in the Nanos 3’ UTR derived from a Gypsy family
transposon, 412. The motif regarding the TE-derived piRNA region of complementarity to the
CG5010 3’ UTRof those that mapped to a defined piRNA clusteris CAAAACGAAAACGTA.
Although the motif doesn’t occur in the same position in all of the piRNAs, most of the piRNAs
exhibit this motif from the 11th to the 25th nucleotides.With that said, about 30% of the piRNAs
exhibited the motif start in the first 4 nucleotides (Table 4).

Figure 5 |
CG5010 3’ UTR piRNA Landscape.The gray regions are the defined SREs within the CG5010
3’ UTR. Sense (blue) and antisense (red) 412-derived piRNAs are represented as having a
region of complementarity to the CG5010 3’ UTR. (A) The FW transposable element-derived
piRNAs that also mapped to a defined piRNA cluster.

piRNA Position of
Motif Start Site
Number of piRNAs
1-4 119
5-9 43
10-14 227
15-18 2
Table 4 | piRNA Motif Position.Table representing the number of piRNAs that had the
complementary motif to the 3’ UTR start at the beginning (nts 1-4), early middle(nts 5-9), late
middle (nts 10-14) and end (nts 15-18) of the piRNA.
Therefore, we present the possibility that piRNAs may promote the deadenylation of
CG5010, among other potential protein-coding candidates, in early Drosophila
melanogasterembryos.Neither the post-transcriptional regulation of CG5010 transcripts, nor the
effect of CG5010derepression in early embryos hasbeenpreviously explored, but may lead to a
better understanding as to the breadth of the piRNA system.
Future Directions
Upon establishment of genes of interest, like CG5010, potentially under piRNA-mediated
deadenylation, experimental validation is required to establishthe significance of the regulation.
First, it is necessary to demonstrate that deadenylation of CG5010 transcripts is taking
place in early, wild-type embryos. In order to assess whether or not deadenylation is occurring
in CG5010 transcripts, we will observe the length of the poly(A) tails of CG5010 transcripts in a
time series in early embryos via poly(A) tests (PAT). PATs are Polymerase Chain Reactions
(PCRs) that indicate the length of poly(A) regions in transcripts of interest. Initially, an adaptor

is ligated onto the 3’ end of RNAs within the sample. The adaptor sequence can then act as a
priming site for reverse transcriptase to convert the molecules from RNA to cDNA. Then, using
the cDNA, primers can be chosen to amplify the region between a unique sequence within the
transcript of interest to the adapter site beyond the poly(A) tails of the transcripts of interest.
Finally, size selection of the cDNA amplified product on an agarose gel electrophoresis would
provide the length of the poly(A) tail (Sallés et al., 1999). PAT assays in Nanos showed poly(A)
tails between 100-250 nucleotides in length. It is reasonable to predict that the length of the
poly(A) tails of CG5010 transcripts would be similar. If CG5010 is undergoing deadenylation,
then the poly(A) tails should shorten in the wild-type time series.
Upon the establishment of deadenylation in early, wild-type embryos, we can proceed to
investigate whether or not piRNAs are playing a role in facilitating the regulation. We would
explore the necessity of the piRNA complementarity to the target 3’ UTR by injecting anti-
piRNAs into a separate time series in earlyDrosophila melanogaster embryos. Anti-piRNAs are
single-stranded oligonucleotides that have a complementary sequence to the piRNAs supposedly
guiding the deadenylation complex. In theory, the anti-piRNA should bind to the
complementary, potentially functional, piRNA to render the guiding element non-functional
(Rouget et al., 2010). Therefore, if the piRNAs of interest are playing a role in guiding the
deadenylation complex to CG5010 transcripts, the efficiency of this mechanism should be
compromised. A change in phenotype could be assessed in this experiment by observing the
poly(A) tail length in the subsequent stages of embryos (2-6 hrs) relative to 0-2 hr embryos. If
piRNAs are promoting recruitment of the CCR4 deadenylase complex to shorten poly(A) tails of
target transcripts, then upon injection of anti-piRNAs, poly(A) tail shortening should become
less efficient or cease altogether. We would perform a control injection using only injection

buffer to demonstrate that any observed phenotypic aberration is not resulting from the innate
stress of an injection in the embryos. Further, we would also inject a piRNAs that do not contain
the potentially functional motif in order to demonstrate that the piRNA-mediated deadenylation
is occurring in a sequence-specific manner. The degree of phenotypic difference between the
wild-type embryos, the previous Nanos anti-piRNA assays, as well as our assays in CG5010 will
provide further information as to the extent of regulation resulting from piRNAs in this
mechanism.
Concluding Remarks
Taken together,the Smaug-independent criteria aim to predict other protein-coding
genesundergoing piRNA-mediated deadenylation. Although piRNA-mediated deadenylation has
already been established in a protein-coding gene, Nanos, the deadenylation mechanism of
piRNA-mediated regulation has yet to be explored on a transcriptome-wide scale.We aim to
explore other potential protein-coding transcripts that may be under regulation via piRNA-
mediated deadenylation using several independent factors indicative of this type of regulation in
the Nanos 3’ UTR.
Smaug has been shown to be involved in a common complex with PIWI proteins, Aub
and Ago3, which can then recruit the CCR4 exonuclease to perform deadenylation on its target
transcripts. Therefore, we propose the investigation of targets of Smaug-mediated deadenylation
to aid in the identification of transcripts potentially undergoing piRNA-mediated deadenylation.
We have identified and seek to further investigate the protein-coding gene, CG5010, as a
potential target of piRNA-mediated deadenylation. CG5010 is the one gene, with the exception

of Nanos, that has satisfied all six of the filters specified in Aims I and II in Drosophila
melanogaster.
Further description of the targets of the piRNA system will contribute to a better
understanding of this relatively poorly understood class of sRNA. A better understanding of the
piRNA class of sRNA can provide further explanation in piRNA function, transposable element
regulation, and even the regulation of protein-coding genes. Also, as the functional mechanisms
of the sRNA system are relatively well conserved among the Animalia kingdom, these findings
within Drosophila melanogaster embryos can likely be extrapolated to other species of interest.
Bibliography
1. Craig N. Mobile DNA: An introduction. In: Craig N et al, eds. Mobile DNA II. American
Society for Microbiology Press. 2002:3–11.
2. McClintock, B. (1953). ―Induction of Instability at Selected Loci in Maize‖. Genetics.
38:579–599.
3. Levin, H. L. & Moran, J. V. (2011). ―Dynamic interactions between transposable elements
and their hosts‖. Nature Reviews Genetics, 12(9), 615–627. Nature Publishing Group.
4. Malone, C. D. & Hannon, G. J. (2009). ―Small RNAs as guardians of the genome‖. Cell,
136(4), 656–668. Elsevier.
5. Chung, W.-J., Okamura, K., Martin, R. & Lai, E. C. (2008). ―Endogenous RNA Interference
Provides a Somatic Defense against Drosophila Transposons‖. Current Biology, 18(11), 795–
802. Elsevier.
6. Bregliano, J. C., et al. "Hybrid dysgenesis in Drosophila melanogaster‖. Science 207.4431
(1980): 606-611.
7. Brennecke, J. et al. (2008). "An epigenetic role for maternally inherited piRNAs in
transposon silencing". Science322: 1387–1392.
8. Bingham, Paul M., Margaret G. Kidwell, and Gerald M. Rubin. "The Molecular Basis of P-
M Hybrid Dysgenesis: The Role of the P Element, a P-strain-specific Transposon
Family‖.Cell 29 (July 1982): 995-1004.
9. Khurana, J. S. &Theurkauf, W. (2010). ―piRNAs, transposon silencing, and Drosophila
germline development‖. The Journal of cell biology, 191(5), 905–913. Rockefeller Univ
Press.
10. Siomi, Mikiko C. Tomohiro Miyoshi, and Haruhiko Siomi. (2010). ―piRNA-mediated
silencing in Drosophila germlines.‖ Seminars in Cell &Devlopmental Biology‖ Vol. 21. No.7
Academic Press.

11. Guzzardo, P. M., Muerdter, F. & Hannon, G. J. (2013). ―The piRNA pathway in flies:
highlights and future directions‖. Current opinion in genetics & development, 23(1), 44–52.
Elsevier.
12. Rangan, P., Malone, C. D., Navarro, C., Newbold, S. P., Hayes, P. S., Sachidanandam, R.,
Hannon, G. J., et al. (2011). ―piRNA Production Requires Heterochromatin Formation in
Drosophila‖. Current Biology, 21(16), 1373–1379. Elsevier.
13. Thomson, Travis, and Haifan Lin. "The biogenesis and function PIWI proteins and piRNAs:
progress and prospect". Annual review of cell and developmental biology 25 (2009): 355.
14. Guo, M. & Wu, Y. (2013). ―Fighting an old war with a new weapon—silencing transposons
by Piwi-interacting RNA‖. IUBMB life. Wiley Online Library.
15. Millar AA, Waterhouse PM. (2005). ―Plant and animal microRNAs: similarities and
differences‖. FunctIntegr Genomics. 5:129–135.
16. Montgomery, T. A., Rim, Y.-S., Zhang, C., Dowen, R. H., Phillips, C. M., Fischer, S. E.
&Ruvkun, G. (2012). ―PIWI associated siRNAs and piRNAs specifically require the
Caenorhabditiselegans HEN1 ortholog henn-1‖. PLoS genetics, 8(4), e1002616. Public
Library of Science.
17. Kawaoka, S., Hara, K., Shoji, K., Kobayashi, M., Shimada, T., Sugano, S., Tomari, Y., et al.
(2013). ―The comprehensive epigenome map of piRNA clusters‖. Nucleic acids research,
41(3), 1581–1590. Oxford Univ Press.
18. Ishizu, H., Siomi, H., and Mikiko C. Siomi. ―Biology of PIWI-interacting RNAs: new
insights into biogenesis and function inside and outside of germlines.‖ Genes and
Development 26.21 (2012): 2361-2373.
19. Robine, N., Lau, N. C., Balla, S., Jin, Z., Okamura, K., Kuramochi-Miyagawa, S., Blower,
M. D., et al. (2009). ―A broadly conserved pathway generates 3′ UTR-directed primary
piRNAs‖. Current Biology, 19(24), 2066–2076. Elsevier.
20. Sallés, F. J., Richards, W. G. & Strickland, S. (1999). ―Assaying the polyadenylation state of
mRNAs‖. Methods, 17(1), 38–45. Elsevier.
21. Brower-Toland, Brent, et al. ―Drosophila PIWI associates with chromatin and interacts
directly with HP1a.‖ Genes & Development 21.18 (2007): 2300-2311.
22. Rouget, Christelet al.―Maternal MRNA Deadenylation and Decay by the PiRNA Pathway in
the Early Drosophila Embryo‖. Nature 467 (October 2010): 1128-132.
23. Pinder, Benjamin D, Smibert, Craig A. ―microRNA-independent recruitment of Argonaute 1
to nanos mRNA through the Smaug RNA-binding protein‖. European Molecular Biology
Organization 14 (November 2012): 80-86.
24. Darricarrere, Nicole, et al. ―Function of Piwi, a nuclear Piwi/Argonaute protein, is
independent of its slicer activity.‖ Proceedings of the National Academy of Sciences 110.4
(2013): 1297-1302.
25. Huang, Xiao A., et al. "A Major Epigenetic Programming Mechanism Guided by
piRNAs‖.Developmental cell (2013).
26. James, T.C., Elgin, S.C (1986). "Identification of a nonhistone chromosomal protein
associated with heterochromatin in Drosophila melanogaster and its gene". Mol. Cell. Biol.6
(11): 3862–3872.

27. Sims, R. J. 3., Mandal, S. S., Reinberg, D. (June 2004). "Recent highlights of RNA-
polymerase-II-mediated transcription". Current opinion in cell biology16 (3): 263–271.
28. Handler, D., Meixner, K., Pizka, M., Lauss, K., Schmied, C., Gruber, F. S. &Brennecke, J.
(2013). ―The Genetic Makeup of theDrosophila piRNA Pathway‖. Molecular cell, 50(5),
762–777. Elsevier.
29. Aravin, A. A. & Chan, D. C. (2011). ―piRNAs meet mitochondria‖. Developmental cell,
20(3), 287–288. Elsevier.
30. Crucs, S., Chatterjee, S. &Gavis, E. R. (2000). ―Overlapping but Distinct RNA Elements
Control Repression and Activation of nanosTranslation‖. Molecular cell, 5(3), 457–467.
Elsevier.
31. Clark, A. G., Eisen, M. B., Smith, D. R., Bergman, C. M., Oliver, B., Markow, T. A.,
Kaufman, T. C., et al. (2007). ―Evolution of genes and genomes on the Drosophila
phylogeny‖. Nature, 450(7167), 203–218. Nature Publishing Group.
32. Chen, L., Dumelie, J. G., Li, X., Cheng, M. H., Yang, Z., Laver, J. D., Siddiqui, N. U., et al.
(2014). ―Global regulation of mRNA translation and stability in the early Drosophila embryo
by the Smaug RNA-binding protein‖. Genome Biol, 15(1), R4.
33. Gӧtze, M. &Wahle, E. (2014). ―Smaug destroys a huge treasure‖. Genome biology, 15(1),
101. BioMed Central Ltd.
34. Benoit, B., He, C. H., Zhang, F., Votruba, S. M., Tadros, W., Westwood, J. T., Smibert, C.
A., et al. (2009). ―An essential role for the RNA-binding protein Smaug during the
Drosophila maternal-to-zygotic transition‖. Development, 136(6), 923–932. The Company of
Biologists Limited.
35. Leinonen, R., Sugawara, H. & Shumway, M. (2010). ―The sequence read archive‖. Nucleic
acids research, gkq1019. Oxford Univ Press.
36. Celniker, S. E., Dillon, L. A., Gerstein, M. B., Gunsalus, K. C., Henikoff, S., Karpen, G. H.,
Kellis, M., et al. (2009). ―Unlocking the secrets of the genome‖. Nature, 459(7249), 927–
930. Nature Publishing Group.
37. St. Pierre SE, Ponting L, Stefancsik R, McQuilton P, and the FlyBase Consortium (2014).
―FlyBase 102 - advanced approaches to interrogating FlyBase‖.Nucleic Acids Res.
42(D1):D780-D788.
38. Laver, J. D., Li, X., Ancevicius, K., Westwood, J. T., Smibert, C. A., Morris, Q. D.
&Lipshitz, H. D. (2013). ―Genome-wide analysis of Staufen-associated mRNAs identifies
secondary structures that confer target specificity‖. Nucleic acids research, gkt702. Oxford
Univ Press.
39. Sai Lakshmi S, Shipra Agrawal. (2008) "piRNABank: a web resource on classified and
clustered Piwi-interacting RNAs". Nucleic Acids Research(Database issue): D173–D177.
40. Dahanukar, A., Walker, J. A. & Wharton, R. P. (1999). ―Smaug, a Novel RNA-Binding
Protein that Operates a Translational Switch in Drosophila”. Molecular cell, 4(2), 209–218.
Elsevier.
41. Jurka, J. ―Repeats in genomic DNA: mining and meaning‖. Curr. Opin. Struct. Biol. 8:333-
337 (1998).
42. Langmead, B. &Salzberg, S. L. (2012). ―Fast gapped-read alignment with Bowtie 2‖. Nature
methods, 9(4), 357–359. Nature Publishing Group.
43. Stephen F. Altschul, Thomas L. Madden, Alejandro A. Schäffer, Jinghui Zhang, Zheng
Zhang, Webb Miller, and David J. Lipman (1997), "Gapped BLAST and PSI-BLAST: a new
generation of protein database search programs". Nucleic Acids Res. 25:3389-3402.

44. Gentleman, R. C., Carey, V. J., Bates, D. M., Bolstad, B., Dettling, M., Dudoit, S., Ellis, B.,
et al. (2004). ―Bioconductor: open software development for computational biology and
bioinformatics‖. Genome biology, 5(10), R80. BioMed Central Ltd.
45. Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, Marshall KA,
Phillippy KH, Sherman PM, Holko M, Yefanov A, Lee H, Zhang N, Robertson CL, Serova
46. Huang DW, Sherman BT, Lempicki RA. ―Bioinformatics enrichment tools: paths toward the
comprehensive functional analysis of large gene lists‖. Nucleic Acids Res. 2009;37(1):1-13.

pschreiner_DissertationProposal

More Related Content

What's hot (20)

Similar to pschreiner_DissertationProposal (20)

pschreiner_DissertationProposal