Phosphorylation of the RNase III enzyme Drosha
at Serine300 or Serine302 is required for its
Xiaoli Tang, Yingjie Zhang, Lynne Tucker and Bharat Ramratnam*
Laboratory of Retrovirology, Division of Infectious Diseases, Department of Medicine, Warren Alpert Medical
School of Brown University, Providence, RI 02903, USA
Received January 29, 2010; Revised May 27, 2010; Accepted May 28, 2010
The RNaseIII enzyme Drosha plays a pivotal role in
microRNA (miRNA) biogenesis by cleaving primary
miRNA transcripts to generate precursor miRNA in
the nucleus. The RNA binding and enzymatic
domains of Drosha have been characterized and
are on its C-terminus. Its N-terminus harbors a
truncated Drosha constructs, we narrowed down
the segment responsible for nuclear translocation
to a domain between aa 270 and aa 390. We
further identified two phosphorylation sites at
Serine300 (S300) and Serine302 (S302) by mass
spectrometric analysis. Double mutations of S!A
at S300 and S302 completely disrupted nuclear
localization. Single mutation of S!A at S300 or
S302, however, had no effect on nuclear localization
indicating that phosphorylation at either site is suf-
ficient to locate Drosha to the nucleus. Furthermore,
mimicking phosphorylation status by mutating S!E
at S300 and/or S!D at S302 restored nuclear local-
ization. Our findings add a further layer of complex-
ity to the molecular anatomy of Drosha as it relates
to miRNA biogenesis.
MicroRNAs (miRNAs) are a class of endogenous
non-protein-coding small RNAs of ?22nt in length that
impact gene expression by sequence-specific interaction
with homologous mRNA (1). Presently, it is thought
that miRNAs most commonly repress gene expression
by base-pairing with the 30-untranslated region (UTR) of
their target mRNAs. However, recent work has also
identified miRNA that target coding sequences of genes
(2). Additionally, miRNA may also target promoter
regions and thereby act as pro-transcriptional elements,
as in the case of miRNA-373 and the gene encoding
E-cadherin (3). One specific miRNA may inhibit many
target genes and one specific gene may be regulated by
miRNA-125a and b, the genes of which are located on
different chromosomes, target the p53 protein as both
miRNAs harbor similar seed sequences that share similar-
ity to the p53 30UTR (4,5). MiRNAs play increasingly
recognized roles in several basic processes including cell
signal transduction, tumorigenesis, tumor invasion and
metastasis, stem cell renewal, immune function, apoptosis
and reaction to stress (6–11). The vast majority of miRNA
genes are thought to be under the control of RNAP II
with others being recently identified as substrates of
RNAP III (12,13). Irrespectively, miRNA genes are ini-
tially transcribed to yield a primary, long transcript that
undergoes successive processing in both the nucleus and
cytoplasm. Nuclear processing is mediated by the RNase
Critical Region Gene 8) microprocessor complex to
generate precursor miRNAs (pre-miRNAs) of ?70nt in
length. Pre-miRNAs are subsequently transported to the
cytoplasm by export 5-Ran-GTP where they are cleaved
by the RNase III enzyme Dicer to generate mature
Investigation into the molecular mechanisms of miRNA
biogenesis at the transcriptional and translational levels
has been intensively pursued (19–22). Drosha plays a
central role in miRNA biogenesis and recent work
suggests that its expression level directly influences
clinical outcomes in malignant disease thus underlying
the importance of better understanding mechanisms that
impact Drosha expression and function (23). Along these
lines, Drosha has been found to interact with other host
proteins including DGCR8. Drosha and DGCR8 regulate
each other post-transcriptionally. The Drosha–DGCR8
complex destabilizes DGCR8 mRNA by cleaving its
hairpin structure. DGCR8, in turn, stabilizes Drosha via
*To whom correspondence should be addressed. Tel: +1 401 444 5219; Fax: +1 401 444 2939; Email: firstname.lastname@example.org
Nucleic Acids Research, 2010, Vol. 38, No. 19Published online 16 June 2010
? The Author(s) 2010. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/
by-nc/2.5), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
protein–protein interaction (22). In an elegant experiment,
deletion of Drosha N-terminal 390 amino acids had no
effect on its binding with DGCR8 and its ability to
process pri-miRNA in vitro (21). While these results sug-
gested that the N-terminal region is dispensable vis-a-vis
the protein’s catalytic activity, they also raised an interest-
ing question: what role, if any, does N-terminal sequence
play in Drosha function in vivo?
Here, we examine the subcellular localization of
wild-type Drosha and N-terminally deleted Drosha. We
find that wild-type Drosha locates to the nucleus, as
expected, but Drosha with N-terminal 390 amino acids
deleted is located exclusively in the cytoplasm. These
results were in agreement with previous work which
found that the N-terminus of Drosha is not dispensable
but rather essential for its nuclear localization and
ultimate miRNA processing function (24). We further
narrowed down the domain for nuclear localization to
the N-terminal region between aa 270 and aa 390. We
show by microscopy and functional assays that phosphor-
ylation at Serine300 or Serine302 is absolutely required for
the nuclear localization and subsequent miRNA process-
ing activities of Drosha.
MATERIALS AND METHODS
Plasmids and primers
Thomas Tuschl and deposited at Addgene (Addgene
plasmid 10828; 25). Wild-type and mutated Drosha were
subcloned into pEGFP-C1 vector (GenBank Accession
#:U55763) using restriction endonucleases HindIII and
cat.#R0142L, respectively). Point mutants were generated
with site-directed mutagenesis techniques as previously
described (26). The primers used for PCR were ordered
from Integrated DNA Technologies. The primer se-
psiCHECK2-miRNA-143 plasmid was created by insert-
ing antisense target sequences of miRNA-143 ds-oligos
AGCGGCCGCAAAAGGAAAA and antisense: TTTTC
CCTCGAGCGG) into psiCHECK2 vector (Promega)
using NotI and XhoI (New England Biolabs) restriction
endonuclease sites. All the cDNA constructs were verified
Biotechnology Resource Laboratory.
was originally created by
Yale KECK Foundation
Antibodies and reagents
antibody (sc-33778), Anti-IkB-a (C-21) rabbit polyclonal
antibody (sc-371), anti-Lamin A (H-102) rabbit poly-
clonal antibody (sc-20680), anti-GFP (B-2) mouse mono-
clonal antibody (sc-9996) and Protein G Plus-Agarose
Biotechnology. Pfu DNA Polymerase (cat.#600136) and
Stratagene. Lipofectamine Reagent (cat.#18324-020) and
III Drosha(H-300) rabbitpolyclonal
purchased from Invitrogen. Alexa Fluor 680 goat anti-
rabbit IgG (H&L) (cat.#A21076) and Hoechst 33342
(cat#H3570) were purchased from Invitrogen Molecular
Probes. IRDye800-conjugated Affinity Purified Anti-
Mouse IgG (H&L) (cat.#610-132-121) was purchased
from Rockland Immunochemicals.
(cat.#BP-190) were purchased from Boston Bioproducts.
LB Agar Kanamycin-50 Plates (cat.#L0643) and EDTA
(cat.#E7889) were purchased from Sigma. The 10?Tris-
Buffered Saline (cat.#170-6435)
solution (cat.#161-0781) were purchased from Bio-Rad
Inhibitor Cocktail Tablets (cat.#11 836 170 001) was
produced by Roche Diagnostics GmbH. QIAprep Spin
Miniprep Kit (cat.#27106), QIAquick PCR Purification
Kit (cat.#28106) and MinElute Gel Extraction Kit
(cat.#28604) were purchased from QIAGEN. Coomassie
Brilliant Blue R-250 Dye (cat.#20278) was purchased from
and 10% Tween20
Cell culture and transfection
HEK293T and Huh-7 cells were cultured in Dulbecco’s
modified Eagle’s medium (Invitrogen, Carlsbad, CA,
USA) with 10% fetal bovine serum and 2mM L-glutam-
ine. HeLa cells were grown in Eagle’s Minimum Essential
Medium supplemented with 10% fetal bovine serum,
2mM L-glutamine and non-essential amino acids. Cells
were trypsinized and reseeded in culture plates 1day
before transfection. HEK293T transfection was per-
formed with Lipofectamine when cell confluency was
?60%. Huh-7 and HeLa cells were transfected with
Lipofectamine 2000 when cell confluency was ?80%.
Cytoplasmic and nuclear fractionation
Cytoplasmic and nuclear fractionation was performed
using EZ Nuclei Isolation Kit (Sigma, cat.#NUC-101) ac-
cording to the manufacturer’s instructions. Briefly, cells
were harvested and washed once with cold PBS. The
cells were then suspended in EZ Nuclei Isolation buffer
and rotated at 4?C for 5min. After centrifugation at 4?C
for 5min, supernatant was collected containing the cyto-
plasmic fraction. Cell lysis and centrifugation were
repeated three times. The final pellets were collected as
the nuclear fraction and lysed in modified RIPA buffer
(50mM Tris–HCl pH 7.4, 150mM NaCl, 2mM EDTA,
1% NP-40, 1% Triton X-100).
Cell lysates (100mg protein each) were separated by 10%
SDS–PAGE electrophoresis and electroblotted to nitrocel-
lulose membrane (Bio-Rad, cat.#162-0115). Blotted mem-
antibodies rotating at 4?C overnight. The membranes
were washed three times in TBST buffer and probed
with secondary antibody (Alexa Fluor 680 goat anti-
rabbit IgG or IRDye800-conjugated Affinity Purified
Anti-Mouse IgG, respectively) at room temperature
for 1h. Membranes were then washed three times in
TBST buffer and direct infrared fluorescence detection
Nucleic Acids Research, 2010,Vol.38, No. 196611
was performed with a Licor Odyssey?Infrared Imaging
LC-MS/MS mass spectrometry analysis of protein with
HEK293T cells were transfected with 20mg cDNA of
GFP–Drosha. Two milligrams of protein lysate was
incubated with 5mg anti-GFP monoclonal antibody at
4?C for 1h. Protein G Plus-Agarose beads (80ml) were
added into the lysate. The lysate-antibody-bead mixture
was rotated at 4?C overnight. The immunoprecipitated
beads were subsequently washed three times with RIPA
buffer, resolved by 10% SDS–PAGE and visualized using
Coomassie blue staining. The GFP–Drosha protein band
was excised and subjected to mass spectrometric analysis
performed by the Taplin Biological Mass Spectrometry
Facility (Boston, MA, USA; 26).
Confocal fluorescent imaging
HEK293T, Huh-7 or HeLa cells were transfected with
constructs expressing GFP or GFP-tagged wild-type or
mutated Drosha, respectively. Twenty-four hours post-
transfection, the cells were trypsinized and reseeded at
1:10 dilution. The cells were incubated for another 24h.
Hoechst 33342 (final concentration 1mg/ml) was added to
cell culture 30min before confocal fluorescent imaging
which was performed with a LeicaTCS SP2 AOBS
confocal laser microscope.
Cell sorting and real-time PCR
HEK293T cells were transfected with constructs express-
ing GFP alone (as EV control) or GFP-tagged wild-type
or mutated Drosha, respectively. Twenty-four hours
post-transfection, the cells were trypsinized and resus-
pended in a sort buffer (1?Ca/Mg+ +-free phosphate
buffered saline, 1mM EDTA, 25mM HEPES pH 7.0,
DNase I). GFP-positive cells were sorted out with
FACSVantage SE (Becton Dickinson) to respective collec-
tion tubes. The collected cells were used for RNA extrac-
tion. Total RNA was extracted by TRIZOL (Invitrogen)
and 1mg of total RNA was used for cDNA synthesis using
MMLV reverse transcriptase (New England Biolabs).
TaqMan?microRNA assay (Applied Biosystems) that
include RT primers and TaqMan probes was used to
quantify the levels of mature miRNA and 18S RNA was
used for normalization. All PCR reactions were run in
We used the psi-CHECK2 system (Promega) to create
sensorassays for quantifying
function byplacing the
miRNA-143 in the 30UTR of the gene encoding Renilla
luciferase. In the presence of mature miRNA-143, the
luciferase activity of Renilla decreases through the classic-
al RNAi pathway. HEK 293T cells were co-transfected
promoter driven miRNA-143 expression vector) and
Drosha expression constructs or empty vector. Firefly
and Renilla luciferase activities were quantified using the
and Renilla luciferase activity was normalized to firefly
luciferase activity. For each experiment, a control employ-
ing an empty vector was used and corrected luciferase
values were averaged, arbitrarily set to a value of ‘1’ and
served as a reference for comparison of fold-differences in
Assay System (Promega)
A miRNA-143 expression plasmid and Drosha wt or
mutant constructs were transfected into HEK293T cells
using lipofectamine reagent. Forty-eight hour post-
transfection, total RNA was prepared with Trizol
reagent. Twenty micrograms of total RNA was used for
northern blotting analysis following conventional proto-
cols but using a biotin-labeled probe and BrightStar
BioDectectTMNonisotopic Detection Kit (Ambion).
The Drosha protein sequence was mined for potential
phosphorylation sites using the PredictProtein program
(proc mixed, SAS version 9.2, SAS Institute, Cary, NC,
USA) were used for statistical analysis as described previ-
GFP-tagged Drosha locates in the nucleus
We first quantified Drosha expression in a number of
human cell lines including HEK293T, Huh-7 and HeLa
cells by cytoplasmic and nuclear fractionation followed by
western blot analysis. As expected, our data (Figure 1A,
top panel) revealed that Drosha exclusively localized to
the nuclei of the tested cells. Cytoplasmic marker IkBa
and nuclear marker Lamin A were used as controls to
confirm the stringency of our nuclear isolation methods
(Figure 1A, middle panel and bottom panel, respectively).
We then subcloned Drosha into pEGFP-C1 vector to
Importantly, sequences encoding GFP do not harbor a
nuclear localization signal (NLS) or a nuclear export
signal (NES). As expected, introduction of the parental
GFP construct into HEK293T cells led to diffuse GFP
expression (Figure 1B, top panel). In contrast, introduc-
tion of GFP–Drosha was associated with exclusive GFP
expression in the nuclei of HEK293T cells (Figure 1B,
bottom panel). We repeated these experiments in other
humancell lines such
Figure S1) and Huh-7 and obtained exactly similar
results. These experiments suggested that sequences
encoding Drosha harbored a putative NLS that was not
affected by the addition of sequence encoding GFP.
The N-terminus of Drosha harbors a NLS
As GFP–Drosha was exclusively expressed in the nucleus,
we reasoned that there must be a strong NLS in Drosha.
6612 Nucleic Acids Research, 2010,Vol.38, No. 19
To pinpoint its location, we constructed a series of GFP–
Drosha deletion constructs as illustrated in Figure 2A.
A construct in which the C-terminal 524 amino acids
was deleted (Drosha?C524) localized exclusively to the
nucleus (Figure 2B, top panel). In contrast, when we
deleted the entire N-terminus (aa 1–390), Drosha localized
exclusively to the cytoplasm (Figure 2B, upper middle
panel) providing a general location for its NLS. We then
focused on better defining the molecular anatomy of the
N-terminus which localizes to the nucleus exclusively
(Figure 2B, down middle panel). We deleted N-terminal
217 amino acids and a corresponding reporter construct
GFP–Drosha?N217 still localized to the nucleus exclu-
sively (Figure 2B, bottom panel). All these results were
confirmed in other cell lines including Huh-7 and HeLa
cells (Supplementary Figure S1).
A domain between aa 270 and aa 390 of Drosha
harbors a NLS
In silico screening of the Drosha protein sequence predicted
two potential NLSs: NLS1 (243-RHRSLDRRER-252) and
constructed GFP–Drosha reporter constructs with either
NLS1 or NLS2 or both being deleted (Figure 3A). GFP–
Drosha270–1374 with NLS1 deleted localized to the
nucleus (Figure 3B, top panel) indicating that this
not essentialfor correct
cellular localization. We next focused on the NLS2 site.
Surprisingly, we encountered similar results in that the
NLS2 deleted variant also localized to the nucleus
(Figure 3B, middle panel). Lastly, a reporter construct
and NLS2 were deleted also localized to the nucleus
(Figure 3B, bottom panel) clearly suggesting that a
nuclear localization mechanism distinct from the canonic-
al NLS in the domain between aa 270 and aa 390 of
Drosha is operational.
Identification of phosphorylation sites by mass
Protein phosphorylation plays an important role in
nuclear localization (28,29). For example, phosphoryl-
ation of extracellular signal-regulated kinase (ERK)-2 at
Ser244 and Ser246 induced its nuclear translocation.
Additionally, phosphorylation of b-catenin at Ser191
and Ser605 controls its nuclear localization. We next
asked if Drosha is constitutively phosphorylated and
whether phosphorylation status impacts nuclear localiza-
tion. The cell lysate from HEK293T cells transfected with
GFP–Drosha for48h without
immunoprecipitated with anti-GFP mouse monoclonal
antibody and Protein G Plus-Agarose beads. The precipi-
tate was resolved by SDS–PAGE and stained with
Coomassie Brilliant Blue R-250 Dye. The gel showed a
clear band corresponding to the calculated size of GFP–
Drosha (Figure 4A). This band was excised and subjected
to LC-MS/MS mass spectrometry. The results confirmed
that the excised band was indeed purified GFP–Drosha
Serine302 being identified (Figure 4B).
sites atSerine300 and
Phosphorylation at Serine300 or Serine302 locates
Drosha to the nucleus
Based upon the MS results, we made a series of GFP–
involving the putative phosphorylation sites as indicated.
Single S!A mutation at S300 or S302 did not change the
nuclear localization of GFP–Drosha (Figure 5A and B,
respectively). Double mutations at S300 and S302,
however, totally prevented nuclear localization producing
a cellular staining pattern similar to GFP without NLS
(Figure 5C). These results indicated that S300 and S302
were constitutively phosphorylated and that phosphoryl-
ation at either site was sufficient to locate Drosha to the
nucleus. Thiswas further
Serine300 to glutamic acid (E), and Serine302 to aspartic
acid (D). Indeed, all of these phosphorylation mimics
DroshaS300E/S302D) located Drosha in the nuclei of
HEK293T cells (Figure 5D–F). Again, these results were
reproducible in other cell lines including HeLa cells
(Supplementary Figures S1 and S2).
Nuclear localization of Drosha is critical for its
functionality in miRNA processing
The fact that Drosha is indispensible for cellular health
posed a challenge for assigning a critical in vivo function to
Figure 1. Drosha localizes to the nucleus. (A) Endogenous Drosha
protein was detected in the nuclei of HEK293T, Huh-7 and HeLa
cells by western blot analysis. Cells were harvested when 100% conflu-
ent. Cytoplasmic and nuclear fractionation was performed. IkBa and
Lamin A served as cytoplasmic and nuclear markers, respectively
(C, cytoplasmic; N, nuclear). (B) Overexpression of a construct
encoding Drosha tagged with GFP on the N-terminus localizes to the
nucleus. HEK293T cells were transfected with GFP backbone vector
alone or GFP-tagged Drosha plasmid, respectively. Hoechst 33342 was
used to stain the nuclei 30min before fluorescent imaging was per-
formed. Top panel: GFP backbone vector expression showing diffuse
localization. Bottom panel: GFP–Drosha expression showing nuclear
Nucleic Acids Research, 2010,Vol.38, No. 196613
the N-terminus using our various constructs. The purest
experiment, after all, would be to use a Drosha null cell
and introduce our various constructs as well as miRNA
expression cassettes to examine their downstream effect on
miRNA processing. To our knowledge, no Drosha null
human cell line exists. We therefore relied upon ectopic
expression of both our various Drosha constructs as well
as miRNA expression cassettes. We first confirmed that all
our Drosha constructs produced similar levels of protein
upon cellular transfection, thereby removing the possibil-
ity that differential protein expression of our constructs
was the ultimate effector of miRNA processing activity
(Figure 6A). As previously reported, HEK293T cells
harbor relatively low levels of mature miRNA-143 (27).
Interestingly, these cells also express relatively low levels
of endogenous Drosha when compared to other cell lines
such as HCT116, NIH3T3, etc. Accordingly, we trans-
fected cells with the GFP vector alone (empty vector),
GFP–Drosha WT or the various GFP–Drosha mutated
constructs along with an expression plasmid encoding
miRNA-143. FACS was used to obtain GFP expressing
cells.Real-timePCRwas performedto measure
Figure 2. The N-terminus of Drosha is critical for nuclear localization. (A) Schematic illustration of domain deletion constructs of Drosha tagged
with GFP at the N-terminus. (B) Cellular localization of different Drosha deletion mutants. Top panel: nuclear localization of GFP–Drosha with
C-terminal 524 amino acid deletion. Upper middle panel: cytoplasmic localization of GFP–Drosha C-terminus. Lower middle panel: nuclear local-
ization of GFP–Drosha N-terminus. Bottom panel: nuclear localization of GFP–Drosha with N-terminal 217 amino acid deletion.
6614 Nucleic Acids Research, 2010,Vol.38, No. 19
miRNA-143 and northern blot was performed to identify
pre- and mature miRNA-143 species. Our data revealed
that overexpression of GFP–Drosha WT, S300A, S302A,
S300E, S302D or S300E/S302D led to the production of
pre- and mature miRNA-143 (Figure 6B and D).
However, cells expressing GFP–DroshaS300/302A or
GFP–Drosha C-terminus (aa 391–1374) did not retain
the ability to produce mature or pre-miRNA-143. These
results were further validated by miRNA sensor assays.
For example, co-transfection of the sensor along with
GFP–Drosha led to a statistically significant reduction
in Renilla activity compared to empty vector. GFP–
DroshaS300A, S302A, S300E, S302D or S300E/S302D
behaved similarly to GFP–Drosha. In contrast, cellular
introduction of GFP–DroshaS300/302A had no signifi-
cant effect on sensor activity compared to empty vector
(Figure 6C). We then repeated this experiment but
used sensors for miRNA-26b and 125a. Again, similar
results were encountered with the stereotypical loss
of miRNA activity in transfections involving only
(aa 391–1374; Supplementary Figure S3). Lastly, we
further reasoned thatthese
mutants should impede the processing of pri-miRNA
and thereby lead to an accumulation of this species.
Indeed, real-time PCR quantification of pri-miRNA-143
revealed a 2-fold increase in experiments involving
391–1374)] compared to
variants (Supplementary Figure S4).
A myriad of proteins exist in a single cell and their correct
subcellular localization is critical for protein function and
cellular health. The RNaseIII enzyme Drosha is a nuclear
protein and correct localization is essential to its function
in processing pri-miRNA species in the canonical path-
way of miRNA biogenesis. The nuclear transporting
Figure 3. The domain for nuclear localization. (A) Schematic illustration of domain deletion constructs of Drosha tagged with GFP at the
N-terminus. (B) Cellular localization of different Drosha deletion mutants. Top panel: nuclear localization of GFP–Drosha with N-terminal 269
amino acid deletion. Middle panel: nuclear localization of GFP–Drosha with NLS2 deletion. Bottom panel: nuclear localization of GFP–Drosha
with N-terminal 269 amino acid and NLS2 deletion.
Nucleic Acids Research, 2010,Vol.38, No. 196615
mechanisms involving Drosha are unknown. Given the
central role of miRNA in many basic cellular processes,
any factor that adversely affects the nuclear location of
Drosha is likely to disturb miRNA biogenesis and even-
tually influence the normal function of the cell.
Most nuclear proteins use a NLS to target themselves to
the nucleus. Typical NLS consist of an amino acid
sequence with positively charged lysines or arginines
(30). In silico screening predicted two potential NLSs
(NLS1 and NLS2) on Drosha. To determine if NLS1
and NLS2 were functional for Drosha, we generated a
series ofwild-type and mutant
with GFP tagged to the N-termini to monitor the
protein localization. GFP is a widely used reporter
protein that does not change the function and subcellular
locationwhen added as
Furthermore, GFP does not harbor a NLS or a nuclear
export signal (NES), thereby accounting for its diffuse
expression upon cellular introduction. GFP–Drosha and
GFP–Drosha1-390 localized to the nucleus. However,
GFP–Drosha391–1374 exclusively localized to the cyto-
plasm. These results raised the possibility that the
C-terminus of Drosha might contain a NES which
perhaps was inhibited by a strong NLS in the full-length
context. Deletion of the N-terminus may have exposed
this NES or abolished its inhibition.
In silico analysis revealed
putative NLS sites in the N-terminus of Drosha.
thepresence of two
GFP–Drosha270–1374 with deletion of NLS1 located in
the nucleus. GFP–Drosha270–1374?NLS2 with deletion
of both NLS1 and NLS2 still located in the nucleus
indicating that both NLS1 and NLS2 were dispensable
for the nuclear localization of Drosha. This prompted us
to hypothesize that Drosha might use a nuclear localiza-
tion mechanism other than the canonical NLS to locate to
the nucleus. In reality, the canonical NLS motif is absent
in many proteins that localize to the nucleus. Proteins
without canonical NLS may contain Serine–Proline–
Serine (SPS) motif. Phosphorylation at either serine can
facilitate binding with nuclear transporter (28). This
prompted us to identify putative phosphorylation sites in
the N-terminus of Drosha. A series of reporter constructs
revealed that phosphorylation at Serine 300 or Serine 302
is criticalfor Drosha
phosphorylated SPS motif acting as a NLS was initially
found in signal proteins such as ERKs, Smad3 and MEK1
(28). Our results suggest that Drosha may use a similar
pathway for correct nuclear localization. We note that
phosphorylation at either or both serine sites could
change the conformation of Drosha and thereby alter its
affinity for putative binding partners. Further investiga-
tion is clearly warranted to identify proteins that interact
with Drosha on the SPS motif as this motif appears highly
conserved among mammals (Supplementary Figure S5).
Reversible phosphorylation of proteins is an important
regulatory mechanism in which kinases (phosphorylation)
Figure 4. Identification of phosphorylation sites by mass spectrometry. (A) Immunoprecipitation of GFP–Drosha with anti-GFP monoclonal
antibody. HEK293T cells were transfected with a GFP–Drosha construct (or empty vector as a control) using Lipofectamine. Whole-cell lysate
was prepared with modified RIPA buffer 48h post-transfection and resultant immunoprecipitants were submitted for MS analysis (lane 1, GFP–
Drosha; lane 2, a transfection with an irrelevant construct as a negative control). (B) Identification of phosphorylation sites by mass spectrometry.
The Coomassie stained GFP–Drosha protein band was excised and mass spectrometry analysis was performed. The asterisk shows the
6616 Nucleic Acids Research, 2010,Vol.38, No. 19
and phosphatases (dephosphorylation) are involved.
Phosphorylation plays critical roles in signal transduction,
protein localization and degradation (33–35). We have
S300 and S302 from immunoprecipitated endogenous
analysis. The potential kinases which phosphorylate
S300 and S302 could include cyclin dependent kinase
by mass spectrometry
activated protein kinase(MAPK)3, MAPK2, MAPK8,
protein kinase A(PKA), protein kinase Ca(PKCa),
protein kinase B(PKB), casein kinaseI (CKI), etc. All of
our data indicate that these serine sites might be constitu-
tively phosphorylated by multiple kinases. Identification
of the specific enzymes which affect Drosha phosphoryl-
ation status may reveal new insights into miRNA
Figure 5. Phosphorylation at S300 or S302 locates Drosha to the nucleus. Experimental procedures were the same as in Figure 1B but used different
constructs as indicated. (A) Expression of GFP–Drosha with Serine300 mutated to Alanine showing nuclear localization. (B) Expression of GFP–
Drosha with Serine302 mutated to Alanine also showed nuclear localization. (C) Expression of GFP–Drosha with both serines mutated to alanines
showing diffuse cellular localization similar to that of GFP without NLS. (D) Expression of GFP–Drosha with Serine300 mutated to glutamic acid
showed nuclear localization. (E) Expression of GFP–Drosha with Serine302 mutated to aspartic acid showing nuclear localization. (F) Expression of
GFP–Drosha with Serine300 mutated to glutamic acid and Serine302 mutated to aspartic acid also showed nuclear localization.
Nucleic Acids Research, 2010,Vol.38, No. 196617
deregulation that is seen in diverse disease states including
Supplementary Data are available at NAR Online.
We thank Ross Tomaino for help in mass spectrometry
analysis, Ginny Hovanesian for assistance in fluorescent
imaging, Sara Spangenberger for assistance in cell sorting
and Jin Song Gao for technical support in performing the
miRNA sensor assays.
(P30AI042853); National Institutes of Health (number
T32DA013911 to X.T.). Funding for open access charge:
National Institutes of Health.
Health (P20RR025179 and
Conflict of interest statement. None declared.
Figure 6. Nuclear localization of Drosha is critical for its functionality in miRNA processing. (A) Protein expression levels of wt Drosha and
mutants. GFP-tagged Drosha wt or mutant constructs were transfected into HEK293T cells using lipofectamine reagent. Forty-eight hours
post-transfection, protein lysates were prepared from the transfected cells and protein levels were measured by western blot analysis. GAPDH
was used as a loading control. All constructs produced equivalent levels of Drosha protein. (B) Quantification of mature miRNA-143 levels/function
by real-time PCR (B), miRNA sensor assays (C) and northern blot (D). GFP vector alone (empty vector, EV), GFP–Drosha wide type (WT) or
different mutant constructs as indicated were transfected into HEK 293T cells along with a miRNA-143 expression vector. Twenty-four hours
post-transfection, GFP-positive cells were cell sorted for RNA extraction. Endogenous mature miRNA-143 level was quantified by using a specific
miRNA-143 Taqman real-time PCR kit. All experiments were performed in triplicate (compared to EV, *P<0.05). (C) miRNA sensor assays
revealed that compared to EV control conditions, cells that had been transfected with Drosha constructs that localized to the cytoplasm
(GFP–DroshaC’ and GFP–DroshaS300/302A) were associated with impaired miRNA function. All experiments were performed in triplicate
(compared to EV, *P<0.05). (D). Effects of overexpressed Drosha wt or mutants on miRNA-143 biogenesis by northern blotting analysis. Cells
transfected with cytoplasmic Drosha localized constructs did not retain the ability to process pri-miRNA-143 with the absence of pre- and mature
species in GFP–DroshaC’ and GFP–DroshaS300/302A treated cells.
6618 Nucleic Acids Research, 2010,Vol.38, No. 19
1. Lee,R.C., Feinbaum,R.L. and Ambros,V. (1993) The C. elegans
heterochronic gene lin-4 encodes small RNAs with antisense
complementarity to lin-14. Cell, 75, 843–854.
2. Tay,Y., Zhang,J., Thomson,A.M., Lim,B. and Rigoutsos,I. (2008)
MicroRNAs to Nanog, Oct4 and Sox2 coding regions modulate
embryonic stem cell differentiation. Nature, 455, 1124–1128.
3. Place,R.F., Li,L.C., Pookot,D., Noonan,E.J. and Dahiya,R.
(2008) MicroRNA-373 induces expression of genes with
complementary promoter sequences. Proc. Natl Acad. Sci. USA,
4. Zhang,Y., Gao,J.S., Tang,X., Tucker,L.D., Quesenberry,P.,
Rigoutsos,I. and Ramratnam,B. (2009) MicroRNA 125a and its
regulation of the p53 tumor suppressor gene. FEBS Lett., 583,
5. Le,M.T., Teh,C., Shyh-Chang,N., Xie,H., Zhou,B., Korzh,V.,
Lodish,H.F. and Lim,B. (2009) MicroRNA-125b is a novel
negative regulator of p53. Genes Dev., 23, 862–876.
6. John,B., Enright,A.J., Aravin,A., Tuschl,T., Sander,C. and
Marks,D.S. (2004) Human microRNA targets. PLoS Biol., 2,
7. Zhang,W., Dahlberg,J.E. and Tam,W. (2007) MicroRNAs in
tumorigenesis: a primer. Am. J. Pathol., 171, 728–738.
8. Gregory,R.I. and Shiekhattar,R. (2005) MicroRNA biogenesis
and cancer. Cancer Res., 65, 3509–3512.
9. Ma,L., Teruya-Feldstein,J. and Weinberg,R.A. (2007) Tumour
invasion and metastasis initiated by microRNA-10b in breast
cancer. Nature, 449, 682–688.
10. Garzon,R., Fabbri,M., Cimmino,A., Calin,G.A. and Croce,C.M.
(2006) MicroRNA expression and function in cancer. Trends Mol.
Med., 12, 580–587.
11. Leung,A.K. and Sharp,P.A. (2007) microRNAs: a safeguard
against turmoil? Cell, 130, 581–585.
12. Lee,Y., Kim,M., Han,J., Yeom,K.H., Lee,S., Baek,S.H. and
Kim,V.N. (2004) MicroRNA genes are transcribed by RNA
polymerase II. EMBO J., 23, 4051–4060.
13. Borchert,G.M., Lanier,W. and Davidson,B.L. (2006) RNA
polymerase III transcribes human microRNAs. Nat. Struct. Mol.
Biol., 13, 1097–1101.
14. Lee,Y., Ahn,C., Han,J., Choi,H., Kim,J., Yim,J., Lee,J.,
Provost,P., Ra ˚ dmark,O., Kim,S. et al. (2003) The nuclear RNase
III Drosha initiates microRNA processing. Nature, 425, 415–419.
15. Landthaler,M., Yalcin,A. and Tuschl,T. (2004) The human
DiGeorge syndrome critical region gene 8 and its D.
melanogaster homolog are required for miRNA biogenesis.
Curr. Biol., 14, 2162–2167.
16. Winter,J., Jung,S., Keller,S., Gregory,R.I. and Diederichs,S.
(2009) Many roads to maturity: microRNA biogenesis pathways
and their regulation. Nat. Cell Biol., 11, 228–234.
17. Han,J., Lee,Y., Yeom,K.H., Nam,J.W., Heo,I., Rhee,J.K.,
Sohn,S.Y., Cho,Y., Zhang,B.T. and Kim,V.N. (2006) Molecular
basis for the recognition of primary microRNAs by the
Drosha-DGCR8 complex. Cell, 125, 887–901.
18. Kim,V.N., Han,J. and Siomi,M.C. (2009) Biogenesis of small
RNAs in animals. Nat. Rev. Mol. Cell Biol., 10, 126–139.
19. Zeng,Y. and Cullen,B.R. (2005) Efficient processing of primary
microRNA hairpins by Drosha requires flanking nonstructured
RNA sequences. J. Biol. Chem., 280, 27595–27603.
20. Nilsen,T.W. (2007) Mechanisms of microRNA-mediated gene
regulation in animal cells. Trends Genet., 23, 243–249.
21. Han,J., Lee,Y., Yeom,K.H., Kim,Y.K., Jin,H. and Kim,V.N.
(2004) The Drosha-DGCR8 complex in primary microRNA
processing. Genes Dev., 18, 3016–3027.
22. Han,J., Pedersen,J.S., Kwon,S.C., Belair,C.D., Kim,Y.K.,
Yeom,K.H., Yang,W.Y., Harssler,D., Blelloch,R. and Kim,V.N.
(2009) Posttranscriptional crossregulation between Drosha and
DGCR8. Cell, 136, 75–84.
23. Merritt,W.M., Lin,Y.G., Han,L.Y., Kamat,A.A., Spannuth,W.A.,
Schmandt,R., Urbauer,D., Pennacchio,L.A., Cheng,J.F.,
Nick,A.M. et al. (2008) Dicer, Drosha, and outcomes in patients
with ovarian cancer. N. Engl. J. Med., 359, 2641–2650.
24. Lee,Y., Han,J., Yeom,K.H., Jin,H. and Kim,V.N. (2006) Drosha
in primary microRNA processing. Cold Spring Harb. Symp.
Quant. Biol., 71, 51–57.
25. Landthaler,M., Yalcin,A. and Tuschl,T. (2004) The human
DiGeorge syndrome critical region gene 8 and Its D.
melanogaster homolog are required for miRNA biogenesis.
Curr. Biol., 14, 2162–2167.
26. Tang,X., Gao,J.S., Guan,Y.J., McLane,K.E., Yuan,Z.L.,
Ramratnam,B. and Chin,Y.E. (2007) Acetylation-dependent signal
transduction for type I interferon receptor. Cell, 131, 93–105.
27. Gao,J.S., Zhang,Y., Li,M., Tucker,L.D., Machan,J.T.,
Quesenberry,P., Rigoutsos,I. and Ramratnam,B. (2010) Atypical
transcription of microRNA gene fragments. Nucleic Acids Res.,
28. Chuderland,D., Konson,A. and Seger,R. (2008) Identification and
characterization of a general nuclear translocation signal in
signaling proteins. Mol. Cell, 31, 850–861.
29. Wu,X., Tu,X., Joeng,K.S., Hilton,M.J., Williams,D.A. and
Long,F. (2008) Rac1 activation controls nuclear localization of
beta-catenin during canonical Wnt signaling. Cell, 133, 340–353.
30. Kalderon,D., Roberts,B.L., Richardson,W.D. and Smith,A.E.
(1984) A short amino acid sequence able to specify nuclear
location. Cell, 39, 499–509.
31. Tsien,R.Y. (1998) The green fluorescent protein. Annu. Rev.
Biochem., 67, 509–544.
32. Lin,M.Z., Glenn,J.S. and Tsien,R.Y. (2008) A drug-controllable
tag for visualizing newly synthesized proteins in cells and whole
animals. Proc. Natl Acad. Sci. USA, 105, 7744–7749.
33. Gong,X., Tang,X., Wiedmann,M., Wang,X., Peng,J., Zheng,D.,
Blair,L.A., Marshall,J. and Mao,Z. (2003) Cdk5-mediated
inhibition of the protective effects of transcription factor MEF2
in neurotoxicity-induced apoptosis. Neuron, 38, 33–46.
34. Tang,X., Wang,X., Gong,X., Tong,M., Park,D., Xia,Z. and
Mao,Z. (2005) Cyclin-dependent kinase 5 mediates
neurotoxin-induced degradation of the transcription factor
myocyte enhancer factor 2. J. Neurosci., 25, 4823–4834.
35. Vervoorts,J., Lu ¨ scher-Firzlaff,J. and Lu ¨ scher,B. (2006) The ins
and outs of MYC regulation by posttranslational mechanisms.
J. Biol. Chem., 281, 34725–34729.
Nucleic Acids Research, 2010,Vol.38, No. 196619