1. No category

Report
Nucleoporin Nup98 Associates with Trx/MLL and NSL
Histone-Modifying Complexes and Regulates Hox Gene
Expression
Graphical Abstract
Authors
Pau Pascual-Garcia, Jieun Jeong, Maya
Capelson
Correspondence
capelson@mail.med.upenn.edu
In Brief
Nuclear pore proteins regulate nuclear
transport but have also been implicated
in the regulation of transcription and
chromatin. Pascual-Garcia et al. now
report a physical and functional interaction between Nup98 and histone-modifying complexes NSL and Trx. The authors identify a component of the NSL
complex, MBD-R2, as targeting Nup98
to active gene promoters, providing
mechanistic insight into the connection
between nuclear pores and chromatin
state.
Highlights
Nup98 associates and colocalizes with the MBD-R2/NSL and
Trx complexes
MBD-R2 is required for recruitment of Nup98 to chromatin at a
number of genes
Nup98 regulates transcription of Trx targets, the Hox genes, in
development
Nup98 overexpression exhibits a homeotic transformation
phenotype
Pascual-Garcia et al., 2014, Cell Reports 9, 1–10
October 23, 2014 ª2014 The Authors
http://dx.doi.org/10.1016/j.celrep.2014.09.002
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
Cell Reports
Report
Nucleoporin Nup98 Associates with Trx/MLL
and NSL Histone-Modifying Complexes
and Regulates Hox Gene Expression
Pau Pascual-Garcia,1 Jieun Jeong,1 and Maya Capelson1,*
1Department of Cell and Developmental Biology, Penn Epigenetics Program, University of Pennsylvania, 9-101 Smilow Center for
Translational Research, 3400 Civic Center Boulevard, Philadelphia, PA 19104, USA
*Correspondence: capelson@mail.med.upenn.edu
http://dx.doi.org/10.1016/j.celrep.2014.09.002
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
SUMMARY
The nuclear pore complex is a transport channel
embedded in the nuclear envelope and made up of
30 different components termed nucleoporins
(Nups). In addition to their classical role in transport,
a subset of Nups has a conserved role in the regulation of transcription via direct binding to chromatin.
The molecular details of this function remain obscure,
and it is unknown how metazoan Nups are recruited
to their chromatin locations or what transcription
steps they regulate. Here, we demonstrate genomewide and physical association between Nup98 and
histone-modifying complexes MBR-R2/NSL and
Trx/MLL. Importantly, we identify a requirement for
MBD-R2 in recruitment of Nup98 to many of its
genomic target sites. Consistent with its interaction
with the Trx/MLL complex, Nup98 is shown to be
necessary for Hox gene expression in developing fly
tissues. These findings introduce roles of Nup98 in
epigenetic regulation that may underlie the basis of
oncogenicity of Nup98 fusions in leukemia.
INTRODUCTION
The nuclear pore complex (NPC) is a massive macromolecular
protein complex embedded in the nuclear envelope (NE). Its classically characterized function is to mediate the selective transport
between the nucleus and the cytoplasm. The NPC consists of
multiple copies of 30 different proteins termed nucleoporins
(Nups), which include the scaffold Nups that form the core ringlike structure of the NPC and the peripheral Nups that regulate
its selectivity barrier (D’Angelo and Hetzer, 2008) and can move
dynamically on and off the pore (Rabut et al., 2004). Mutations
in Nups are responsible for several human disorders, most notably
cases of acute myelogenous leukemia (AML), caused by oncogenic fusions of Nup98 to a number of different partners (Gough
et al., 2011; Xu and Powers, 2009). Metazoan Nups have been
shown to affect tissue-specific development such as neural
and muscle differentiation (D’Angelo et al., 2012; Lupu et al.,
2008), yet the mechanisms by which NPC components can lead
to tissue-specific pathologies and phenotypes remain unknown.
In addition to their classical role as transport channels, NPCs
have been demonstrated to regulate transcriptional programs
via physical binding to specific genes. In yeast, peripheral Nups
has been shown to bind and promote expression of the nutritionally inducible genes INO1 and GAL1 (Cabal et al., 2006; Light
et al., 2010; Schmid et al., 2006; Taddei et al., 2006). Remarkably,
this Nup-gene interaction was also required for transcriptional
memory of the gene’s active state, assisting its subsequent reactivation after several cell divisions (Brickner, 2009; Tan-Wong
et al., 2009). A similar role for Nups in epigenetic transcriptional
memory has been recently described in human cells, where interferon g (IFN-g)-inducible genes are activated more robustly with
repeated exposure to IFN-g, but this remembered response is
lost upon depletion of Nup98 (Light et al., 2013).
In Drosophila, genome-wide methods of polytene chromosome
staining, chromatin immunoprecipitation (ChIP), and DamID
demonstrated that several of the peripheral Nups, including
Nup98, were recruited to genes undergoing developmentally
induced transcriptional activation (Capelson et al., 2010; Kalverda
et al., 2010; Vaquerizas et al., 2010). In line with the mobile
behavior of peripheral Nups, it was found that such Nup-gene
contacts commonly occur in the nucleoplasm, away from the
NE-embedded NPCs. Human Nup98 has been similarly detected
at genes undergoing robust activation during neural differentiation
of embryonic stem cells (Liang et al., 2013). Together, these findings implicate Nups in direct regulation of developmental transcription programs, yet the molecular mechanism by which
Nups affect transcription or its memory remains obscure.
To begin unraveling this mechanism, we identified interacting
partners of Drosophila Nup98, which include proteins implicated
in gene activation and epigenetic memory. Importantly, we
pinpoint the chromatin-binding complex that is able to recruit
Nup98 to active genes and uncover a role for Nup98 in maintaining developmental expression of Hox genes. These findings
expose the link between NPC components and epigenetic regulators of tissue-specific gene expression, which may also be
central to the oncogenic roles of Nups in leukemia.
RESULTS AND DISCUSSION
Identification of Chromatin-Associated Interacting
Partners of Nup98
To address the molecular mechanism of the role of Nups in transcription and development, we aimed to identify proteins that
Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors 1
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
Figure 1. Nup98 Co-occupies Genomic Sites and Physically Associates with the MBD-R2/NSL and Trx Complexes
(A) Heatmap visualization of the enrichment matrix of ChIP-chip profiles of chromatin factors derived from modENCODE database with the Nup98 ChIP-chip data
set (green arrow). Redder colors denote higher enrichment, and bluer colors denote low enrichment. Enrichment of sets of regions A and B that have N(A) and N(B)
bps, respectively, and C(A,B) bps in common, is E(A,B) = G 3 C(A,B) / N(A)N(B), where G is the number of bps in the genome. The scale of colors is logarithmic.
(B) Heatmap of enrichment of the Nup98 ChIP-chip data set compared with tested data sets, ranked in the order of decreasing enrichment score.
(legend continued on next page)
2 Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
cooperate with Nup98 in gene regulation. To this end, we
compared the Nup98 ChIP-chip data set from embryonic S2
culture cells (Capelson et al., 2010) to the ChIP-chip profiles of
multiple factors from the same cell type, available through the
modENCODE project database. To uncover factors that correlate the most with genomic binding of Nup98, we carried out a
hierarchical clustering analysis on the selected data sets and
generated an enrichment matrix (Figure 1A). Interestingly,
Nup98 correlated the most with a group of chromatin regulatory
factors, including methyl binding domain-related 2 (MBD-R2),
components of the NURF chromatin remodeling complex
(NURF301 and ISWI), a well-known mediator of epigenetic memory (Trithorax [Trx]), and WD40 domain protein Wds (Figures 1A
and S1A). This is consistent with the known roles of Nup98 in
transcriptional activation and transcriptional memory and suggested that Nup98 coregulates the same gene targets as these
epigenetic complexes.
From our analysis, Nup98 shows the highest amount of overlap with MBD-R2 (Figure 1B). MBD-R2 is a component of the
males absent on the first (MOF)-containing nonspecific lethal
(NSL) complex, involved in genome-wide H4 K16 acetylation
and transcriptional activation (Prestel et al., 2010; Raja et al.,
2010). In mammals, the MOF complex has been reported to
cooperate with the mammalian equivalent of Trx, the histone
methyltransferase mixed lineage leukemia (MLL) complex (Wysocka et al., 2005; Zhao et al., 2013), with the homolog of
Wds, Wdr5, believed to be the main shared component. Like
the mammalian MOF and MLL, the fly NSL/MBD-R2 complex
has been reported to co-occupy its target promoters with Trx
(Feller et al., 2012). Overall, 77% of genomic binding sites of
Nup98 in S2 cells contain Trx and at least 44% of Nup98 sites
contain MBD-R2, while 43% of Nup98 sites are co-occupied
by both Trx and MBD-R2 (Figure S1B). Furthermore, three of
the four strongest conserved DNA binding motifs that we have
identified for Nup98 include DRE, E-box, and DMv2-like, which
are among the top consensus motifs previously identified for
the NSL complex in Drosophila (Figure S1E).
To test for physical association between Nup98 and its
genomic binding partners, we carried out a coimmunoprecipitation (co-IP) analysis using S2 cells extract enriched for the
nuclear soluble fraction, and we were able to confirm the interaction of Nup98 with MBD-R2 as well as with additional putative
NSL complex components, Z4 and Chromator (Chriz) (Mendjan
et al., 2006; Prestel et al., 2010) (Figure 1C). Incidentally, previously published analysis of the NSL protein complex yielded
other peripheral Nups such as Nup153 and Megator as interacting partners of MOF (Mendjan et al., 2006). We did not detect
ISWI in the Nup98 immunoprecipitated fraction (Figure 1C), supporting the specificity of the co-IP. On the other hand, both Trx
and Wds coimmunoprecipitated Nup98 from S2 cell extracts
(Figures 1D and S1D). As suggested by the identified role of
Nups in transcriptional initiation, Nup98 is found highly enriched
at promoter regions and correlates well with the transcription
start site (TSS) binding of Trx and Wds (Figure S1C).
To further validate interactions that we have identified, we performed the co-IP analysis in the presence of DNase I and RNase
A. As depicted in Figures 1E and 1F, the same associations could
be detected in both Nup98 and MBD-R2 immunoprecipitated
fractions from nuclear extracts treated with DNase and DNase +
RNase. The DNase and RNase treatments were validated by
assessing the integrity of nucleic acids, isolated from the same
extracts (Figure S1F). These results argue that the association
of Nup98 with MBD-R2 and Trx is not a consequence of pulling
down genomic fragments with closely bound complexes.
The genomic overlap and physical association between
Nup98 and Trx is particularly intriguing in light of the shared
leukemic phenotypes of human Nup98 and MLL, both of which
can be part of oncogenic protein fusions leading to AML (Tenney
and Shilatifard, 2005). Our findings in the fly system suggest that
Nup98 and Trx/MLL may in fact be components of the same
chromatin-bound complex. In support of our hypothesis, a
recent sequence analysis of protein domain architecture of
Nups across a large range of eukaryotes reported a repeated
evolutionary occurrence of Nup98 homologs containing a SET
domain, a conserved histone methyltransferase domain that is
present in Trx/MLL (Katsani et al., 2014).
Nup98 Can Be Recruited to Chromatin via MBD-R2
It has not been determined how (i.e., through which molecular
factors) components of the nuclear pore interact with chromatin
in animal cells. Thus, we next sought to determine whether any of
the identified interacting partners of Nup98 can mediate the
recruitment of Nup98 to its target genomic sites.
To validate genomic overlap observed in S2 cells, we
compared genome-wide binding of Nup98 on polytene chromosomes of salivary glands from third-instar larvae to those of its
newly identified interacting partners, by immunofluorescence.
We observed that the staining patterns of MBD-R2 and Z4 (Figure S2A) closely overlap on polytene chromosomes, supporting
the previously suggested association of Z4 with the NSL complex. Similarly, comparison of the reported ChIP-sequencing
profiles of Chriz and Z4 in Kc cells (Van Bortle et al., 2014) with
MBD-R2 showed that 71% of the reported MBD-R2 peaks
contained Chriz and 73% contained Z4 (Figure S2D). Coimmunofluorescence analysis of Nup98 with Z4 as a marker of the
MBD-R2/NSL complex revealed their frequent colocalization
genome-wide (Figure S2C), supporting the identified interaction
between Nup98 and the MBD-R2/NSL complex. On the other
hand, little colocalization was observed between Trx and Z4 (Figure S2B), suggesting that the functions of these two chromatin
regulatory complexes are distinct in this tissue.
Next, we utilized the salivary gland-specific driver Sgs3-GAL4 to
induce upstream activating sequence (UAS)-driven RNAi against
MBD-R2, which resulted in a visible reduction of MBD-R2 on
(C) Coimmunoprecipitation (co-IP) analysis with Nup98 and control immunoglobulin G (IgG) in S2 cell extracts enriched for nuclear soluble fraction, western
blotted against labeled factors.
(D) Co-IP analysis with Trx, Nup98, and control IgG on S2 cell extracts, western blotted for Nup98.
(E and F) Co-IP analysis with Nup98, MBD-R2, and control IgG on S2 cell extracts, prepared as in (C), and either untreated or treated with DNase or DNase/RNase,
as indicated, and western blotted against Trx (E) and Nup98 and MBD-R2 (F).
Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors 3
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
Figure 2. Nup98 Requires MBD-R2 for Binding to Chromatin
(A) Immunofluorescence analysis of polytene chromosomes from third-instar larvae of wild-type (WT) versus Sgs3-driven MBD-R2 RNAi genotypes, stained
against Nup98 and Mod2.2 as control. DNA is labeled in blue by Hoechst.
(B) Expression or mRNA levels of Nup98 target genes Hph and CG10851, assessed by quantitative PCR in S2 cells, treated with control dsRNA against White or
against Nup98.
(legend continued on next page)
4 Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
chromatin (Figure S2E). Strikingly, the depletion of MBD-R2
resulted in partial loss of Nup98 from the chromosomes (Figure 2A), indicating that MBD-R2 is necessary for recruitment of
Nup98 to chromatin, at least at a certain fraction of Nup98 binding
sites. On the other hand, RNAi against Trx, Wds, or ISWI did not
affect chromatin binding of Nup98 (data not shown).
To further address the question of Nup98 recruitment,
we tested the binding of Nup98 at specific target promoters in
S2 cells by ChIP analysis. Target genes such as Hph and
CG10851, where Nup98 co-occupies the promoter/TSS region
with MBD-R2, Trx, and Wds, depend on Nup98 for their expression (Figure 2B). As expected, ChIP analysis with anti-Nup98
antibodies at these promoters, as compared to a control region
not bound by Nup98, revealed an enrichment of Nup98 signal,
which was strongly reduced upon RNAi treatment against
Nup98 (Figure 2D). In agreement with results obtained with polytene chromosomes (Figure 2A), RNAi against MBD-R2 in S2 cells
similarly resulted in 40%–50% loss of Nup98 occupancy at the
tested promoters, while RNAi against Trx or Wds did not significantly affect the levels of Nup98 at the target genes (Figures 2D,
S2F, and S2G). The protein levels of Nup98 were not observed to
decrease upon depletion of any of the tested factors (Figure 2C).
On the other hand, RNAi-mediated depletion Nup98 did not have
a strong affect on recruitment of MBD-R2 at the same targets,
suggesting that Nup98 does not participate in binding of the
MBD-R2/NSL complex to chromatin (Figure S2H).
Together, these results reveal MBD-R2 as a factor that can
mediate recruitment of a nuclear pore component to chromatin.
Although the MBD-R2/NSL complex appears to be one such targeting factor, it is likely that Nup98 utilizes other, yet-unidentified
genome-targeting partners to be recruited to sites that lack the
NSL complex.
Nup98 Is Required for Hox Gene Expression in
Developing Drosophila Tissues
We were next interested in the functional significance of the
interactions we identified for Nup98. As mentioned above, the
MBD-R2/NSL complex is thought to carry out H4 K16 acetylation
on autosomes. The best-characterized functions of Trx are maintaining tissue-specific Hox gene expression (Breen and Harte,
1993; Petruk et al., 2001) and the activity of its complex as an
H3K4 methyltransferase (Mohan et al., 2011).
The gene coding for Nup98 also codes for another NPC
component, Nup96, which is a stable Nup required for NPC assembly. Nup98 and Nup96 are transcribed and translated
together, after which the Nup98-Nup96 peptide is autoproteolytically cleaved into individual Nup98 and Nup96 components
(Figure S3A). To assess the developmental function of Nup98
exclusively, we generated transgenic flies that express UAScontrolled Nup96, which localized properly to nuclear pores (Figure S4A). We utilized the UAS-driven RNAi line, which depletes
the transcript that carries both Nup98 and Nup96 (Figure S3A),
and combined it with the UAS-Nup96 construct (UAS-Nup98/
96 RNAi; UAS-Nup96), leading to the knockdown of the endogenous Nup98-Nup96 transcript but with compensatory expression of Nup96 (Figure S3B).
During the third-instar stage of fly development, one of the
best-known gene targets of Trx-mediated active gene memory,
Ultrabithorax (Ubx), is highly expressed in the haltere imaginal
disc, particularly in the haltere blade region (Figure 3A, top panel).
Using protein extracts from third-instar larvae, we observed a
robust co-IP of Nup98 with Trx in these tissues (Figure S3C).
To determine whether Nup98 plays a role in Ubx expression,
we drove UAS-Nup98/96 RNAi; UAS-Nup96 with a tissuespecific GAL4 under the control of the Nubbin promoter (NubGal4), the expression pattern of which is shown in Figure S3D.
We observed a significant decrease in Ubx expression in the haltere discs of Nup98-depleted animals (Figure 3A). Quantification
of fluorescence intensity of anti-Ubx staining in the blade area,
where Nub-GAL4 is expressed, relative to the total fluorescence
intensity of the haltere disc, showed a reproducible decrease in
the levels of Ubx in Nup98-depleted animals versus wild-type,
while levels of the control staining with anti-Lamin DmO were
not affected (Figure 3B). Furthermore, this decrease was not
observed in Nub-GAL4-driven RNAi of Nup107 (Figures 3B and
S3B), suggesting that these affects are not the result of general
nuclear pore disruption. We could confirm these effects by western blot analysis of imaginal disc tissues using the engrailedGAL4 (en-Gal4) (which is expressed in every imaginal disc),
which demonstrated that the levels of Hox genes Ubx and Antennapedia (Antp) are decreased by Nup98 RNAi, but not by Nup107
RNAi (Figure 3C).
If the effect of Nup98 on Ubx expression is carried out by a
mechanism linked to Trx, then the disruption of Ubx should occur
at the level of transcription as opposed to a potential mRNA
export defect due to Nup98 depletion. To address this question,
we analyzed the levels of Ubx mRNA and nascent (unspliced)
transcript in the haltere discs of control, Nup98 RNAi, and Trx
RNAi animals (Figure 3D). As expected, we observed a significant
reduction of Ubx mRNA in the haltere discs upon RNAi-mediated
depletion of Trx (Figure 3D, left). A similar level of downregulation
of Ubx mRNA was observed in the haltere discs of Nub-GAL4
driven UAS-Nup98/96 RNAi; UAS-Nup96 animals (Figure 3D,
left). Furthermore, reduction of the nascent transcript of Ubx
was clearly detected in Nup98-depleted haltere discs, using
four different primer combinations specific to the nascent unspliced Ubx RNA (Figure 3D, right). The genomic controls for
these reactions, without addition of the reverse transcriptase
(RT), produced no signal (data not shown). The downregulation
of Ubx at the level of nascent transcript supports our hypothesis
that this role of Nup98 is executed through its transcriptional
function, not its mRNA export role.
Using the Nub-GAL4 driver approach, we detected no
changes in the levels of H3K4 trimethylation upon Nup98 knockdown in larval wing imaginal discs, while RNAi against Wds resulted in a strong depletion of the H3K4 trimethyl mark in the
(C) Western blot analysis with indicated antibodies of S2 cell extracts, treated with indicated dsRNAs.
(D) Chromatin immunoprecipitation, assayed by quantitative PCR, using Nup98 or control IgG antibodies, in S2 cells, treated with indicated dsRNAs, at promoter
regions of CG10851 and Hph genes, labeled red (Nup98 ChIP) or gray (IgG ChIP), or at control regions (Table S1), labeled in black, plotted as % of Input Recovery.
Error bars are derived from triplicate experiments.
Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors 5
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
Figure 3. Nup98 Is Necessary for Maintaining Hox Gene Expression
(A) Immunofluorescence staining of haltere and third-leg imaginal discs of third-instar larvae of wild-type (w1118), Nub-GAL4-driven Nup98 RNAi, with Nup96
compensated (Nub-GAL4; Nup98RNAi, Nup96) or Nup107 RNAi genotypes, stained for Ubx (red), Lamin DmO (green), and DAPI (blue). The dashed circled area
represents the blade area of Nub-GAL4 expression, where Ubx expression is normally highest, and the solid circled area represents the entire haltere disc.
(B) Quantification of fluorescent pixel intensity in the blade region over the total remaining fluorescent intensity in the haltere disc, for Ubx and Lamin stainings, in
the indicated genotypes. Each dot on the top plot represents an individual measurement from a single haltere. The bottom plot shows average blade/total
measurement for Ubx or Lamin.
(legend continued on next page)
6 Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
blade region (Figure S3E). Furthermore, we found no pronounced or reproducible changes in H4K16 acetylation (Figure S3F). Together, these results identify a function for Nup98
in maintaining normal levels of Hox gene expression and further
support the suggested role of Nup98 in Trx-mediated epigenetic
memory of active gene expression.
Nup98 Modifies Mutant Phenotypes of Trx and MBD-R2
To determine the potential in vivo cooperation between Nup98
and chromatin complexes with which it interacts, we sought to
test for possible genetic interactions between Nup98 and Trx
and MBD-R2. Since Trx is required for maintaining Ubx expression in the developing haltere, depletion of Trx results in partial
homeotic transformation of haltere to wing (Breen and Harte,
1991), often manifested as ectopic bristles on the haltere (which
normally lacks bristles altogether). Driving RNAi against Trx with
Nub-GAL4 expectedly resulted in a mild transformation of haltere to wing, manifested as one to two bristles per haltere on
average in 73% of the progeny (Figure 4A). RNAi-mediated
depletion of Nup98 with Nub-GAL4 results in almost complete
lethality in the pupal stages, making it difficult to assess adult
phenotypes. Thus, we generated a Nup98-overexpressing
transgene, UAS-Nup98-myc (Figure S4A), to test for possible
genetic interactions with Trx and MBD-R2.
Surprisingly, Nub-GAL4-driven overexpression (OE) of Nup98
alone resulted in a mild haltere transformation in approximately
25% of the progeny, while OE of Nup96 did not show any haltere
transformation (Figure 4A). Combining Nup98 OE with Trx RNAi
strongly enhanced the haltere transformation phenotype of Trx,
resulting in 91% of the progeny exhibiting the homeotic transformation (Figure 4A) and an average of three to four ectopic bristles
per haltere (Figure 4A, right chart). Nup96 OE also had a mild effect on the Trx RNAi haltere phenotype, resulting in 79% of the
progeny with transformed halteres and producing two to three
ectopic bristles per haltere on average, but this effect was less
severe that that of the Nup98 OE (Figure 4A). Interestingly, these
results reveal that OE of Nup98 appears to act in the same direction as the Nup98 depletion, at least in the context of Trx-related
phenotypes. In support of this idea, we observed a significant
decrease in the levels of Ubx in imaginal discs from animals
with patched (ptc)-GAL4 driven Nup98 OE (Figure S4B). One
possibility that would explain how depletion and overexpression
of Nup98 result in a similar phenotype is the potential ability of
Nup98 to oligomerize. The FG repeat domains of Nups, present
in Nup98, are known to self-interact and/or polymerize, which is
thought to underlie their ability to set up a permeability barrier
through the NPC (Milles et al., 2013; Xu and Powers, 2013).
Too much Nup98 may interfere with Trx-directed effects on activation, perhaps via uncoupling interactions between Trx and
other downstream effectors.
We next analyzed the effects of Nup98 OE in the eye, where
ey-GAL4 driven RNAi against either Trx or MBD-R2 results in
malformation and reduced size of the eye (Figure 4B). As
observed in the haltere, OE of Nup98 results in the enhancement
of the mutant phenotype of Trx RNAi, such that the average estimated eye area is drastically reduced from 75% of the total eye
area in Trx RNAi alone to 17% in the Trx RNAi combined with
Nup98 OE (Figures 4B and 4C). OE of Nup96, combined with
Trx RNAi, did not alter the eye phenotype of Trx RNAi significantly, resulting in 72% average eye area (Figures 4B and
4C). On the other hand, OE of Nup98 had an opposing effect
on MBD-R2 RNAi, resulting in an increase in the average eye
area relative to MBD-R2 RNAi alone, while OE of Nup96 did
not (Figure 4B and 4C). These enhancing and suppressing trends
were also observed in the overall viability of the genotypes, such
that OE of Nup98 was found to decrease the viability of Trx RNAi
animals but to increase the viability of MBD-R2 RNAi flies (Figure 4B, bottom panel). One possibility that these results are
consistent with is that Nup98 functions effectively downstream
of MBD-R2 but upstream of or in parallel with Trx in the context
of transcriptional activation (Figure 4D).
Our findings present interacting partners of Nup98, which are
components of chromatin regulatory complexes and mediators
of epigenetic memory. Interestingly, the role of Nup98 in IFNg-induced transcriptional memory has been shown to rely on
H3K4 dimethylation (Light et al., 2013), which is put on by the histone methyltransferase activities of the MLL-related complexes,
supporting the idea that the connection between Nup98 and
MLL is evolutionary conserved. These findings provide insights
into the causative role of Nups in leukemia progression, suggesting that Nup98 and MLL form a complex that plays a central role
in regulation of Hox genes and that leukemic fusions of either
Nup98 or MLL disrupt the normal activities of this complex.
Overall, our results introduce new mechanistic directions for
elucidating the role of nuclear pore components in tissue-specific development and in pathologies that have been linked to
Nups.
EXPERIMENTAL PROCEDURES
Analysis of the Genomic Distribution
Details of the bioinformatics analysis (Figures 1A, 1B, S1C, and S1E) are
detailed in Supplemental Experimental Procedures.
Fly Strains and Transgenics
All strains are described in Supplemental Experimental Procedures. For haltere and eye phenotypes, 15–50 individual progeny from each cross were
scored. For estimated eye area, each eye was assigned into an approximate
percentage category, as shown in Figure 4B.
Immunoprecipitation and Western Blotting
The sources and dilutions for all antibodies are listed in Supplemental Experimental Procedures. Immunoprecipitations (IPs) were performed from 2 3 107
S2 cells lysed hypotonically to enrich for nuclear soluble fraction; further details of which can be found in Supplemental Experimental Procedures. Protein
extracts from third-instar larvae were prepared by lysing the anterior tissues,
(C) Western blot analysis of Ubx and Antp protein levels in imaginal discs from w1118, Nup107 RNAi and Nup98 RNAi; Nup96 larvae, driven by en-GAL4, with
anti-Tubulin as loading control. Quantification of Ubx and Antp levels, normalized to Tubulin, are represented as plots.
(D) Diagrams of the Ubx exon-intron structure and relative positions of primers used in the quantitative RT-PCR analysis (plots below) of Ubx mRNA and nascent
transcript levels in the haltere discs of indicated genotypes, normalized to Rp49 (see Table S1 for primers). Error bars are derived from triplicate quantitative PCR
experiments performed on haltere discs of indicates genotypes.
Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors 7
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
Figure 4. Overexpression of Nup98 Modifies Phenotypes of Trx and MBD-R2
(A) The left chart represents percentage of progeny of a given genotype that exhibit transformed halteres, defined as at least one bristle per haltere, for Nub-GAL4
driven RNAi against Trx, with or without Nup98 or Nup96 overexpression (OE), and for Nup98 OE or Nup96 OE alone. The right chart and panel show quantification
and examples of homeotic transformation of halteres to wings, represented as the number of bristles scored per individual haltere, for adults of indicated genotypes.
(B) The top chart represents average eye size phenotypes of adults with ey-GAL4-driven RNAi against Trx or MBD-R2, with or without Nup98 OE or Nup96 OE,
and of Nup98 OE or Nup96 OE alone. The fly eye images below represent the % scale of eye size, used for scoring estimated eye area. The bottom chart
represents obtained viability from the same crosses.
(C) Examples of eye phenotypes for each indicated genotype.
(D) Possible model of the role of Nup98 in regulation of Hox genes and other targets, showing the ability of MBD-R2/NSL complex to recruit Nup98 (blue arrow)
and the ability of Nup98 to influence the functions of Trx or functions downstream of Trx, which lead to transcriptional activation (red arrows).
8 Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
containing salivary glands, brains, and imaginal discs, in RIPA buffer (10 mM
Tris [pH 7.4], 150 mM NaCl, 1% NP40, and 0.1% SDS) for 20 min. Proteins
were resolved by SDS-PAGE and subjected to western blot analysis with primary antibodies in blocking buffer (PBS + 5% milk), and horseradish peroxidase or fluorescent secondary antibodies.
Breen, T.R., and Harte, P.J. (1993). Trithorax regulates multiple homeotic
genes in the bithorax and Antennapedia complexes and exerts different tissue-specific, parasegment-specific and promoter-specific effects on each.
Development 117, 119–134.
Immunofluorescence
Immunostaining of polytene chromosomes and larval imaginal discs was
carried out as described previously (Capelson et al., 2010). For imaginal disc
imaging, z stack projections of 20–25 confocal sections of each disc were
quantified by ImageJ. Pixel intensity of Ubx and Lamin stainings was assessed
in the blade region of the haltere disc (delimited according to the morphology
of the disc) and normalized to the intensity measured from the whole disc.
Cabal, G.G., Genovesio, A., Rodriguez-Navarro, S., Zimmer, C., Gadal, O.,
Lesne, A., Buc, H., Feuerbach-Fournier, F., Olivo-Marin, J.C., Hurt, E.C., and
Nehrbass, U. (2006). SAGA interacting factors confine sub-diffusion of transcribed genes to the nuclear envelope. Nature 441, 770–773.
Expression Analysis, Cell Culture, and RNA Interference
Drosophila S2 cells were grown at 25 C in Schneider’s medium (Life Technologies) supplemented with 10% fetal bovine serum (Gibco). Double-stranded
RNAs (dsRNAs) were synthesized with a Megascript T7 kit (Ambion) using fly
genomic DNA or cDNAs as templates. Primers used are listed in Table S1.
Knockdown cells were subjected to two rounds of RNAi using 20 mg of dsRNA
per 1 3 106 cells and incubation for 3 days in each round. Total RNA was isolated from cells or wandering third-instar larvae using TRIzol (Ambion) and
reverse transcribed with one-step RT-PCR kit (QIAGEN). Each quantitative
RT-PCR was repeated three times from groups of five to ten larvae each
and the values were normalized to the rp49 or Act5C primers.
Chromatin Immunoprecipitation
ChIP experiments were carried out as described here with some minor modifications (http://www.epigenesys.eu/en/protocols/chromatin-immunoprecipi
tation-chip Prot 48), as detailed in Supplemental Experimental Procedures.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures,
four figures, and one table and can be found with this article online at http://
dx.doi.org/10.1016/j.celrep.2014.09.002.
AUTHOR CONTRIBUTIONS
Brickner, J.H. (2009). Transcriptional memory at the nuclear periphery. Curr.
Opin. Cell Biol. 21, 127–133.
Capelson, M., Liang, Y., Schulte, R., Mair, W., Wagner, U., and Hetzer, M.W.
(2010). Chromatin-bound nuclear pore components regulate gene expression
in higher eukaryotes. Cell 140, 372–383.
D’Angelo, M.A., and Hetzer, M.W. (2008). Structure, dynamics and function of
nuclear pore complexes. Trends Cell Biol. 18, 456–466.
D’Angelo, M.A., Gomez-Cavazos, J.S., Mei, A., Lackner, D.H., and Hetzer,
M.W. (2012). A change in nuclear pore complex composition regulates cell differentiation. Dev. Cell 22, 446–458.
Feller, C., Prestel, M., Hartmann, H., Straub, T., So¨ding, J., and Becker, P.B.
(2012). The MOF-containing NSL complex associates globally with housekeeping genes, but activates only a defined subset. Nucleic Acids Res. 40,
1509–1522.
Gough, S.M., Slape, C.I., and Aplan, P.D. (2011). NUP98 gene fusions and hematopoietic malignancies: common themes and new biologic insights. Blood
118, 6247–6257.
Kalverda, B., Pickersgill, H., Shloma, V.V., and Fornerod, M. (2010). Nucleoporins directly stimulate expression of developmental and cell-cycle genes inside the nucleoplasm. Cell 140, 360–371.
Katsani, K.R., Irimia, M., Karapiperis, C., Scouras, Z.G., Blencowe, B.J., Promponas, V.J., and Ouzounis, C.A. (2014). Functional genomics evidence unearths new moonlighting roles of outer ring coat nucleoporins. Sci Rep 4, 4655.
Liang, Y., Franks, T.M., Marchetto, M.C., Gage, F.H., and Hetzer, M.W. (2013).
Dynamic association of NUP98 with the human genome. PLoS Genet. 9,
e1003308.
Light, W.H., Brickner, D.G., Brand, V.R., and Brickner, J.H. (2010). Interaction
of a DNA zip code with the nuclear pore complex promotes H2A.Z incorporation and INO1 transcriptional memory. Mol. Cell 40, 112–125.
P. P.-G. and M.C. designed, performed, and analyzed the experiments and
wrote the paper, and J.J. carried out the bioinformatics analysis and provided
input on the writing.
Light, W.H., Freaney, J., Sood, V., Thompson, A., D’Urso, A., Horvath, C.M.,
and Brickner, J.H. (2013). A conserved role for human Nup98 in altering chromatin structure and promoting epigenetic transcriptional memory. PLoS Biol.
11, e1001524.
ACKNOWLEDGMENTS
Lupu, F., Alves, A., Anderson, K., Doye, V., and Lacy, E. (2008). Nuclear pore
composition regulates neural stem/progenitor cell differentiation in the mouse
embryo. Dev. Cell 14, 831–842.
We thank members of the Capelson lab, particularly Jessica Talamas, for valuable input and imaging help. We are very grateful to Martin Hetzer and Robbie
Schultz for sharing of reagents and to all members of the Hetzer lab for comments and discussions. We sincerely thank P. Becker for anti-MBD-R2 and
anti-Z4, K. Johansen for anti-Chriz, A. Mazo for anti-Trx, J.T. Kadonaga for
anti-ISWI, E.P. Lei for anti-Mod2.2, L. Jones for anti-tubulin, G. Karpen for
anti-Wds, and P. Fisher for anti-Lamin antibodies. We also thank the Bloomington, VDRC, and DGRC stock centers and the Developmental Studies Hybridoma Bank for reagents. M.C. is supported by the March of Dimes Foundation
and by the W. W. Smith Charitable Trust Research award.
Received: March 7, 2014
Revised: July 19, 2014
Accepted: August 28, 2014
Published: October 9, 2014
Mendjan, S., Taipale, M., Kind, J., Holz, H., Gebhardt, P., Schelder, M., Vermeulen, M., Buscaino, A., Duncan, K., Mueller, J., et al. (2006). Nuclear pore
components are involved in the transcriptional regulation of dosage compensation in Drosophila. Mol. Cell 21, 811–823.
Milles, S., Huy Bui, K., Koehler, C., Eltsov, M., Beck, M., and Lemke, E.A.
(2013). Facilitated aggregation of FG nucleoporins under molecular crowding
conditions. EMBO Rep. 14, 178–183.
Mohan, M., Herz, H.M., Smith, E.R., Zhang, Y., Jackson, J., Washburn, M.P.,
Florens, L., Eissenberg, J.C., and Shilatifard, A. (2011). The COMPASS family
of H3K4 methylases in Drosophila. Mol. Cell. Biol. 31, 4310–4318.
Petruk, S., Sedkov, Y., Smith, S., Tillib, S., Kraevski, V., Nakamura, T., Canaani, E., Croce, C.M., and Mazo, A. (2001). Trithorax and dCBP acting in a
complex to maintain expression of a homeotic gene. Science 294, 1331–1334.
REFERENCES
Prestel, M., Feller, C., Straub, T., Mitlo¨hner, H., and Becker, P.B. (2010). The
activation potential of MOF is constrained for dosage compensation. Mol.
Cell 38, 815–826.
Breen, T.R., and Harte, P.J. (1991). Molecular characterization of the trithorax
gene, a positive regulator of homeotic gene expression in Drosophila. Mech.
Dev. 35, 113–127.
Rabut, G., Doye, V., and Ellenberg, J. (2004). Mapping the dynamic organization of the nuclear pore complex inside single living cells. Nat. Cell Biol. 6,
1114–1121.
Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors 9
Please cite this article in press as: Pascual-Garcia et al., Nucleoporin Nup98 Associates with Trx/MLL and NSL Histone-Modifying Complexes and
Regulates Hox Gene Expression, Cell Reports (2014), http://dx.doi.org/10.1016/j.celrep.2014.09.002
Raja, S.J., Charapitsa, I., Conrad, T., Vaquerizas, J.M., Gebhardt, P., Holz, H.,
Kadlec, J., Fraterman, S., Luscombe, N.M., and Akhtar, A. (2010). The nonspecific lethal complex is a transcriptional regulator in Drosophila. Mol. Cell 38,
827–841.
Schmid, M., Arib, G., Laemmli, C., Nishikawa, J., Durussel, T., and Laemmli,
U.K. (2006). Nup-PI: the nucleopore-promoter interaction of genes in yeast.
Mol. Cell 21, 379–391.
Taddei, A., Van Houwe, G., Hediger, F., Kalck, V., Cubizolles, F., Schober, H.,
and Gasser, S.M. (2006). Nuclear pore association confers optimal expression
levels for an inducible yeast gene. Nature 441, 774–778.
Tan-Wong, S.M., Wijayatilake, H.D., and Proudfoot, N.J. (2009). Gene loops
function to maintain transcriptional memory through interaction with the nuclear pore complex. Genes Dev. 23, 2610–2624.
Tenney, K., and Shilatifard, A. (2005). A COMPASS in the voyage of defining
the role of trithorax/MLL-containing complexes: linking leukemogensis to covalent modifications of chromatin. J. Cell. Biochem. 95, 429–436.
Van Bortle, K., Nichols, M.H., Li, L., Ong, C.T., Takenaka, N., Qin, Z.S., and
Corces, V.G. (2014). Insulator function and topological domain border strength
scale with architectural protein occupancy. Genome Biol. 15, R82.
10 Cell Reports 9, 1–10, October 23, 2014 ª2014 The Authors
Vaquerizas, J.M., Suyama, R., Kind, J., Miura, K., Luscombe, N.M., and
Akhtar, A. (2010). Nuclear pore proteins nup153 and megator define transcriptionally active regions in the Drosophila genome. PLoS Genet. 6,
e1000846.
Wysocka, J., Swigut, T., Milne, T.A., Dou, Y., Zhang, X., Burlingame, A.L.,
Roeder, R.G., Brivanlou, A.H., and Allis, C.D. (2005). WDR5 associates with
histone H3 methylated at K4 and is essential for H3 K4 methylation and vertebrate development. Cell 121, 859–872.
Xu, S., and Powers, M.A. (2009). Nuclear pore proteins and cancer. Semin. Cell
Dev. Biol. 20, 620–630.
Xu, S., and Powers, M.A. (2013). In vivo analysis of human nucleoporin repeat
domain interactions. Mol. Biol. Cell 24, 1222–1231.
Zhao, X., Su, J., Wang, F., Liu, D., Ding, J., Yang, Y., Conaway, J.W., Conaway,
R.C., Cao, L., Wu, D., et al. (2013). Crosstalk between NSL histone acetyltransferase and MLL/SET complexes: NSL complex functions in promoting histone
H3K4 di-methylation activity by MLL/SET complexes. PLoS Genet. 9,
e1003940.