J Med Dent Sci 2015； 62： 43－56
Original Article
Molecular Characterization of UKp83/68, a Widespread Nuclear Proteins that
Bind Poly(A) and Colocalize with a Nuclear Speckleʼs Component
Satoru Miyakura and Masayuki Hara
Department of Cellular and Environmental Biology Graduate School, Tokyo Medical and Dental University
1-5-45 Yushima Bunkyo-ku, Tokyo 113-8510, Japan
We have cloned a gene from a rat liver cDNA library,
representing alternatively spliced cDNAs encoding
83-kDa and 68-kDa proteins, which we have
designated as UKp83 and UKp68, respectively.
Both proteins have a predicted nuclear localization
signal and five CCCH motifs (zinc-binding motifs),
and share a degree of sequence similarity with
Nab2, a yeast protein that contains nucleic acidbinding motifs and tandem CCCH zinc fingers. Nab2
binds homopolymeric RNA and single-stranded
DNA and regulates poly(A) tail length and the
export of mRNA to the cytosol. The CCCH motifs
of UKp83/68 bound poly(A) and ssDNA strongly
and other RNA homopolymers and dsDNA less
efficiently. The UKp83/68 protein localized within
the nucleus with a fibrous or punctate structure
that reflected the distribution of SC35, a known
marker of nuclear speckles which are nuclear
domains enriched in pre-mRNA splicing factors
and located in the interchromatin regions of the
nucleoplasm of mammalian cells. The distribution
of UKp83/68 changed during the different
stages of mitosis. During prometaphase, when
the nuclear envelope disintegrates, the protein
becomes partially localized on the chromosomes;
at other times, transiently dispersed over the
cytoplasm with the formation of fibrous structure.
The transient expression of UKp83 in HEK293T
cells had no apparent effect on cellular function,
whereas the expression of an antisense sequence
Corresponding Author: Masayuki Hara, Ph.D.
General Isotope Center, Tokyo Medical and Dental University, 1-545 Yushima Bunkyo-ku, Tokyo 113-8510, Japan.
Tel: +81-3-5803-5790　Fax: +81-3-5803-5789
E-mail: [email protected]
Received October 24, 2014；Accepted March 6, 2015
or C-terminal domain of UKp83 induced apoptosis.
These results suggest that UKp83/68 is probably
essential for cell viability and may play important
role in mRNA processing.
Key words: RNA-binding protein, chromatin, CCCH
zinc finger, nuclei, apoptosis
Introduction
Although there are no membranous structures inside
the nucleus, many intranuclear subcompartments exist
such as chromatin, nucleolus, cajal bodies and
spliceosomes. The components of these structures selfassemble and biochemical reactions such as DNA
replication, transcription, mRNA processing, and assembly
of ribosomes are facilitated by increased concentrations
of various biomolecules in the individual compartments
that serve as “chemical factories” (Gadal and Nehrbass,
2002). These remarkable structures are formed by
elaborate protein–protein, protein-DNA and protein-RNA
interactions that are expected to involve DNA- and
RNA-binding proteins that play crucial roles in assembly
processes by recognizing specific DNA/RNA sequences
and recruiting many other proteins.
DNA- and RNA-binding proteins are characterized by
DNA- or RNA-binding motifs in their sequences. The zinc
finger motif is one of the typical DNA/RNA-binding
motifs. It contains several Cys and His residues at
regular intervals and has a structure stabilized by zinc
coordination. In most zinc finger proteins, several fingers
are tandemly repeated in a single polypeptide. The most
common type of zinc finger is the Cys2His2 (CCHH) type,
which is found in, for example, transcription factor
TFIIIA (Hall, 2005; Brown, 2005) and GATA-4 (Rojas et
al., 2009). Another type is Cys3His (CCCH), which is
found in murine tristetraprolin (TTP) (Hall, 2005; Brown,
44
S. Miyakura et al.
2005) and yeast Nab2 (Batisse et al., 2009).
In the course of our studies aimed at the identification
of nuclear structural proteins of rat hepatocytes, we
isolated a cDNA clone (clone 45), which encodes a
novel protein. Clone 45 was found to be a fragment
representing alternatively spliced cDNAs encoding 83kDa and 68-kDa proteins. We designated these proteins
UKp83 and UKp68, respectively. These proteins
contain nuclear localization signals, CCCH type zinc
finger motifs and have a degree of sequence similarity
with yeast Nab2.
Nab2 is a poly(A)-binding protein in yeast. It contains
two potential nucleic acid-binding motifs, an arginineglycine-glycine (RGG) repeat domain and seven
tandem CCCH zinc fingers and is known to bind
to homopolymeric RNA and single-stranded DNA
(Anderson et al., 1993). Further characterization of
Nab2 revealed that it shuttles between the nucleus
and the cytoplasm and is required for both nuclear
export and proper polyadenylation of mRNA transcripts
(Batisse et al., 2009). More recently, it was revealed
that ZC3H14, a human homolog of UKp68, which
contains CCCH zinc fingers homologous to those found
in Nab2, also specifically binds poly(A) RNA (Kelly et al.,
2007; Leung et al., 2009).
In this report we describe the structure, intracellular
and intranuclear localization and tissue distribution of
UKp83/68 and demonstrate that these proteins may be
closely related to a role for cell viability.
Materials and Methods
Isolation of cDNA Clones
A λ ZAPII cDNA expression library from rat liver cells
(Stratagene) was screened with a polyclonal antiserum
raised against a rat liver chromatin fraction as
described previously (Hara et al., 1999). One of the
cDNA clones obtained was selected, plaque-purified,
and released from phages by in vivo excision according
to the manufacturerʼs protocol. The resulting cDNA of
~1.6 kilobases (kb) (clone 45) was inserted into
pBluescript and characterized by sequencing from both
directions.
Since clone 45 was not a complete cDNA, the
sequences of the missing both 5′- and 3′-ends were
determined by the rapid amplification of cDNA ends
using the RACE procedure (rapid amplification of
cDNA ends) as described by Frohman et al. (Frohman
et al., 1988) with poly(A) + RNA isolated from rat
liver cells and the Marathon cDNA amplification kit
(Clontech) according to the manufacture’s protocol.
J Med Dent Sci
The two gene-specific primers used were 5′CTTGGTGGCACAGTAATAGTGGGGTGA-3′ and 5′CCGGGTCCCATAATCCTCATCTTCTTC-3′ for 5′-RACE
and 5′- GCTTCCACCAGGCTAATGTCAACAGTG-3′ and
5′-CAAGCCGAGATGACTG ACCTGAGTGTG-3′ for 3′
-RACE. The resulting PCR products were subcloned
into the pCR 2.1 plasmid using the Original TA Cloning
kit (Invitrogen) and sequenced. The full-length clone
was obtained by PCR using primers upstream of the
initiation codon and downstream of the stop codon.
The entire ~2.2kb cDNA sequence was sequenced
from both strands. Database searches were performed
by using the BLAST network service at the National
Center for Biotechnology Information (Altschul et al.,
1990).
Polyclonal antiserum
Clone 45, which contained amino acids(AA)115–605
of UKp68, was subcloned into pGEX-5X (Amersham
biosciences). A fusion protein with glutathione-S transferase (GST) was produced in Escherichia coli
BL21 cells and purified using glutathione-sepharose. A
rabbit was injected with 100 µg of the fusion protein
three times at intervals of a week for generation of
antibodies. The resulting antiserum harvested from the
rabbit was purified by Protein G Sepharose (GE
Healthcare). The fusion protein was also prepared as
GST-UKp68(AA115-605) to examine the binding of
CCCH motif and polynucleotide.
In order to obtain the anti-UKp83/68 antibodies, it
was subjected to the following. Cell nuclei from rat
liver cells were isolated, homogenized, and prepared
to preparative electrophoresis using Rotofor and Prep
Cell (Bio-Rad). Then, native UKp83 was purified on the
basis of the immunoblot with anti-GST-Clone45 and
was immunized to a rabbit. The resulting antiserum
harvested from the rabbit was purified by Protein
G Sepharose (GE Healthcare). And anti-GFP was
purchased from commercially available (Clontech).
Binding of GST-UKp68(AA115-605) to polynucleotide
Resin beads (25 µl each) which covalently bound
either of poly(G), poly(A), poly(C), poly(U), single
stranded DNA, or double stranded DNA (Sigma;
0.7–1.5 mg of polynucleotide per milliliter of resin)
were centrifuged and the precipitated resins were
resuspended in 250 µl of 10 mM Tris-HCl buffer (pH 7.4)
containing GST-UKp68(AA115–605) fusion protein (2
µg), 2.5 mM MgCl2, 1 mM ZnCl2, 0.5% Triton X-100, and
various concentrations of NaCl (0.05, 0.1, 0.3, or 1 M).
The suspensions were incubated at 4ºC for 10 min on
Molecular Characterization of UKp83/68
a rocking platform. The beads were spun down briefly
in a microcentrifuge, resuspended in the same solution
(but without GST-UKp68), and incubated for another
10 min at 4ºC on a rocking platform. The beads were
then washed four times in the same buffer. The washed
beads were suspended in 25 µl of SDS-polyacrylamide
gel electrophoresis (SDS-PAGE) loading buffer, boiled
for 3 min, and centrifuged. The supernatants were
subjected to SDS-PAGE.
Gel electrophoresis and immunoblotting
SDS-PAGE was performed according to Laemmli
(Laemmli, 1970) using 10% acrylamide and proteins
were either stained by Coomassie Brilliant blue R or
blotted to polyvinylidene difluoride (PVDF) membranes
(Bio-Rad) with a semidry electroblotter (Bio-Rad) for
l h at 1.8 mA/cm 2 . The amount of the sample was
constant on all the lanes and was 20-50 µg/lane.
For immunostaining of UKp83/68, the membrane
was blocked for 60 min in Tris-buffered saline (TBS)
containing 0.05% Tween 20 (TBST) and 5% nonfat dry
milk. The anti-UKp83/68 antibodies (diluted 1:10,000)
were incubated with the membranes for 3–5 h in
TBST/5% nonfat dry milk. After incubation with alkaline
phosphatase conjugated secondary antibodies diluted
1:5,000 in 100 mM Tris-HCl (pH 9.5) containing 100 mM
NaCl and 50 mM MgCl2 for 1 h, bound antibodies were
visualized with a color detection system using nitro
blue tetrazolium (NBT) and 5-bromo-4-chloro-3-indolylphosphate (BCIP).
RNA isolation and northern blot hybridization
Total RNA from rat tissues was prepared using Isogen
(Wako) according to the manufacturer’s protocol.
Poly(A) + RNAs (5 µg) prepared using Oligotex-dT30
(Takara) were fractionated on a 1% agarose gel
containing 6% formaldehyde and transferred to a
Hybond-N filter (Amersham). Hybridization and detection
of UKp83/68 RNA was performed using the DIG
Northern Starter kit (Boehringer Mannheim) essentially
according to the manufacturer’s protocol. The filters
were hybridized with the antisense single stranded DNA
probes made by asymmetric PCR using the clone 45 as
a template and the “PCR DIG probe synthesis kit
(Boehringer Mannheim) for 16 h at 50ºC, followed by
two 10-min washes with 2× SSC/0.1% SDS, two 20 min
washes with 1× SSC/0.1% SDS, and one 20-min wash
with 0.1× SSC/0.1% SDS. Oligonucleotides used for
asymmetric PCR labeling were as follows: 5′-CA
CTTACATACGGTTCTTCTCG-3′ (P1) and 5′-CATTT
CTGCGAGCACTTCTTC-3′ (P2) for Clone 45, 5′-AC
45
ACCAAACCAGGATTCGGGG-3′ (P3) and 5′-GCT
ACCATCTGGGTCCTCAAGC-3′ (P4) for Exon11+12
(see below), 5′-GTTGAACATCATTTTTGGTAAT-3′ (P5)
and 5′-TCTGTTTTCTGATACTTCCTGTA-3′ (P6) for 3′
UTR-2.
Expression of UKp83 and other cDNAs in HEK293T
cells
HEK293T cells were maintained in Dulbecco’s modified
Eagle’s medium (Invitrogen) with 10% fetal bovine serum
(Invitrogen). A cell line with younger passage number
was used and it was verified as free of any contaminants
including other cell types. For transient expression in
HEK293T cells, UKp83 and other cDNAs were subcloned
into a pDNR2 vector (Clontech). The constructs were
subcloned into the eukaryotic expression vector pLPEGFP-C1 (Clontech), which expresses fusion proteins
with EGFP. Transfections were performed using
Lipofectamine PLUS Reagent (Invitrogen) according to
the manufacturer’s protocol.
Immunofluorescence Microscopy
For immunofluorescence microscopy studies, cells
grown on coverslips were either fixed in methanol (7
min, -20ºC) or in phosphate-buffered saline (PBS)
containing 4% formaldehyde (10–20 min, room
temperature). For permeabilization, formaldehyde-fixed
cells were incubated for 10 min with PBS containing
0.5% Triton X-100.
After fixation, the cells were washed twice in PBS,
blocked for 60 min in TBS containing 5% nonfat dry
milk, and incubated with the anti-UKp83/68 antibodies
(diluted 1:5,000 in TBS/5% nonfat dry milk) for 3–5 h at
room temperature. After washing several times with
PBS, cells were incubated for 60 min with FITC-labeled
goat anti-rabbit IgG (Coppel, 1:1,000 in TBS/5% nonfat
dry milk), washed in PBS, counterstained with DAPI,
and mounted in Aqua-Poly/Mount (Polysciences).
For immunohistochemical studies of rat tissues, the
formalin-fixed and paraffin-embedded sections (2 µm)
were stained as previously described (Hara et al.,
1999). After deparaffinization, tissue sections were
incubated with TBS/5% nonfat dry milk for 3 h at room
temperature, and then treated with the anti-UKp83/68
antibodies (overnight) and the second antibody as
described above.
For double-labeling experiments, anti-SC35
monoclonal antibodies (Sigma, 1:2,000 in TBS/5%
nonfat dry milk) were mixed with the anti-UKp83/68
antibodies (diluted 1:2,000 in TBS/5% nonfat dry milk).
The antibodies were detected with Alexa488-labeled
46
J Med Dent Sci
S. Miyakura et al.
goat anti-mouse IgG (Molecular Probes, 1:1,000 in
TBS/5% nonfat dry milk) and TRITC-labeled swine antirabbit IgG (DAKO, 1:50 in TBS/5% nonfat dry milk),
respectively.
Cell number counting studies for transfected cells
were performed with NIH Image software using digital
fluorescent image obtained from GFP, as transient
expressed cells, and DAPI, as total cells. Measured
wells of incubation chambers were eight for each day
of each gene. Fields for analysis were three randomly
selected areas (each of 872.5 x 691.3 µm2) from one
well.
A Zeiss Axioskop microscope equipped with a 100
× fluorescence/differential interference contrast
objective was used for all cellular immunofluorescence
studies.
Confocal laser-scanning immunofluorescence
microscopy was performed on a Zeiss LSM 510 (Zeiss).
For simultaneous double-label fluorescence, an argon
ion laser operating at 488 nm and a helium-neon laser
operating at 543 nm were used together with a bandpass filter combination of 510–525 nm and 590–610 nm
for visualization of Alexa488 and TRITC fluorescence,
respectively.
Terminal deoxynucleotidyl transferase nick-end
labeling (TUNEL) assay
To detect DNA fragmentation in situ , a TUNEL assay
was performed in HEK293T cells expressing the GFPUKp83 fusion protein. Cell samples were harvested
after washing with PBS at 1, 2, and 3 days of posttransfection, fixed with 4% paraformaldehyde, and
stained with phycoerythrin (PE) using the MEBSTAIN
Apoptosis Kit II (MBL) according to the manufacturerʼs
guidelines. Cell samples irradiated with ultraviolet light
at 254 nm for 1 min with a UV-illuminator were used
as positive controls for TUNEL staining of apoptotic
nuclei. TUNEL positive and negative cells were counted
as expressed (GFP-positive) and non-expressed (GFPnegative) cells by flow cytometry (EPICS XL; Beckman
Coulter).
Statistical analysis
Data examined the percentage of cells transfected
with various genes were statistically analyzed for
relevance using StatPlus:mac LE Version 2009 software
(AnalystSoft Inc., Alexandria, VA). Comparisons between
continuous variables from more than two groups
were performed using one way analysis of variance
(ANOVA). If the result shows significant difference,
non-parametric Studentʼs t test was used to assess
difference between the two genes in all combinations.
The values are expressed as mean ± SEM. In all cases,
P value < 0.05 was considered statistically significant.
Nucleotide sequence accession numbers and Gene
ID
The accession numbers for the nucleotide sequences
of UKp83 and UKp68 mRNAs in the EMBL/GenBank/
DDBJ nucleotide sequence data bases are AB032932
and AB097075, respectively. Ukp83/68 gene is known
as Zc3h14 (Gene ID: 192359) and also known as
Npuk68.
Results
Structure of the UKp83/68 gene in rat genome
and Expression of its mRNA in various rat tissues
By searching the rat genome, the UKp83/68 gene
is found in chromosome 6 at 6q32 named Zc3h14
(zinc finger CCCH type containing 14) and composed
of 16 exons. The initiation codon is in exon 1 and
the stop codon is in exon 16 (Fig. 1A). We examined
mRNA containing the clone 45 sequence. Poly(A)+ RNA
prepared from rat liver was analyzed by northern blot
using an antisense single-stranded DNA probe (561bp long) prepared as a template from clone 45. As
shown in Figure 1B, a transcript of about 2.6 kb was
detected, along with three minor bands of about 2.2,
3.0, and 3.4 kb. To determine the structures of these
four transcripts we performed the RACE procedure
using adaptor primers and various internal primers.
Amplification and sequencing of the cDNA fragments
by PCR with various sets of primers revealed the
presence of two inserted sequences, as illustrated in
Figure 1A. The 3.4 and 2.6 kb transcripts contain all of
the exons, while the 3.0 and 2.2 kb transcripts lack the
11th and 12th exons. In the 2.6 and 2.2 kb transcripts,
polyadenylation is provided by a polyadenylation signal
ATTAAA at a position 240 nucleotides (indicated by 3′
UTR-1 in Fig. 1A) downstream of the stop codon in exon
16, while in the longer transcripts (3.0 kb and 3.4 kb),
another polyadenylation signal AATAAA is located 840
nucleotides (indicated by 3′UTR-2 in Fig. 1A) further
downstream in the same exon.
The 2.2 kb transcript, which lacked the inserted
sequence, encoded a 605-amino acid protein with a
molecular weight of 68,024 (pI = 7.54). We designated
this protein UKp68 (eukaryotic protein with a molecular
weight of 68k). The major 2.6 kb transcript encoded a
736-amino acid protein with a molecular weight of
82,628 (pI = 6.58), designated UKp83. The other two
Molecular Characterization of UKp83/68
transcripts, 3.0 and 3.4 kb, had another additional
sequence of 853 bp (3′UTR-2) in the 2.2 and 2.6 kb
transcripts in their 3′-untranslated region (Fig. 1A).
T hese t rans cri pts e n c o d e d UK p 6 8 a n d U kp8 3 ,
respectively. When an oligonucleotide with a sequence
complementary to the 11th and 12th exons was used as
a probe to detect the mRNA, only 2.6 and 3.4 kb
transcripts were detected (Fig. 1B). Similarly, an
oligonucleotide complementary to 3′UTR-2 detected
only the 3.0 and 3.4 kb transcripts (Fig. 1B), confirming
the structural relationships among these transcripts
47
indicated in Figure 1A.
We next investigated the expression of UKp83/68
mRNA in various rat tissues. As shown in Figure 1C, all
4 transcripts were detected in the tissues tested.
Among these tissues, the testis has the highest
expression, followed by the prostate. Lung tissue had
the lowest expression. In most tissues, the 2.6 kb
transcript was the major component, whereas the 2.2
and 3.0 kb transcripts were predominant in the
cerebrum and cerebellum (Fig. 1C).
Figure 1. Polymorphism of UKp83/68 transcripts and their distributions in rat tissues
(A) The diagram depicts exons predicted for the UKp83/68 splice variants encoding four UKp83/68 gene transcripts
(3.4k, 3.0k, 2.6k, and 2.2k) and Clone45. The sequences and approximate positions of predicted classical NLS motifs
(cNLSs) are indicated at the top of the diagram. The location of the translation start (ATG) and stop (TGA) sites are also
indicated as are the approximate locations of the proline-tryptophan-isoleucine domain (PWI) and the tandem Cys3His zinc
finger domain (CCCH). Positions of primer pairs (P1 and P2, P3 and P4, and P5 and P6) used to detect the splice variants
are indicated at the bottom of the diagram. (B) Northern blot analysis of UKp83/68 in the rat liver were performed.
Polyadenylated RNA isolated from rat liver was fractionated on a 1% formaldehyde-agarose gel and probed with an
asymmetric PCR fragment of UKp83/68 using each primer pair indicated. (C) Detection of UKp83/68 gene transcripts in
various rat tissues were performed. Polyadenylated RNA was isolated from various rat tissues and analyzed as described
for Figure 1B using primer pair P1 and P2. A GAPDH gene transcript (1.2 kb) was also detected in each tissue as a
control. The lower panel shows the densitometric values of the four transcripts in each tissue divided by those of GAPDH
in the same tissue.
48
S. Miyakura et al.
UKp83/68 protein structure
Both proteins contain at least two interesting
structural motifs. The first is a 17-amino-acid motif
(292–308) containing a basic amino acid cluster
predicted to be a classical nuclear localization signal
(cNLS) (Dingwall et al., 1988) found in a wide array of
nuclear proteins (Fig. 1A). The second and the most
important feature of UKp83/68 is the CCCH putative
zinc-binding motif. The CCCH motif, or so-called “zincfinger motif,” is known to bind zinc ion and forms DNAor RNA-binding domains. In UKp83/68, it is repeated 5
times in the C-terminal region (indicated by 5 x CCCH in
Fig.1A). The interval between Cys-Cys or Cys-His is
CX5CX5CX3H in the first two fingers, and CX5CX4CX3H
in the next three fingers. Moreover, the N-terminal
domain of UKp83/68 conserved the similarity of
proline-tryptophan-isoleucine (PWI) domain, which is
found either at the N terminus or at the C terminus of
eukaryotic proteins involved in pre-mRNA processing.
Detection of UKp83/68 protein
We next attempted to detect localization of UKp83/68
in various cells. To do this, antiserum against the
purified UKp83 was prepared and purified as described
in Materials and Methods. Then, the resulting antiUKp83/68 antibodies were tested for its specificity.
The anti-UKp83/68 antibodies were reacted with
UKp68, UKp83, and Luciferase (Luc) that were
expressed in HEK293T cells, respectively, as a fusion
protein with green fluorescent protein (GFP) and
prepared for immunoblot analysis after homogenization
and SDS-PAGE preparation. As a result, only GFPUKp68 (95 kDa) and GFP-UKp83 (110 kDa) were
detected by the anti-UKp83/68 antibodies, however
GFP-Luc (89 kDa) was not detected (Fig. 2A left panel).
On the other hand, when it was with the anti-GFP
antibodies, all of three kinds of fusion proteins were
detected (Fig. 2A right panel). By the way, in the result
using the anti-UKp83/68 antibodies, two more bands
were detected at the position of about 68k and 83k in
all the lanes, including the non-treated cell fraction (NT)
(indicated by arrows in Fig. 2A left panel). This result
was considered that the anti-UKp83/68 antibodies
have detected native UKp68 and UKp83 in HEK293T
cells.
Histochemical studies also indicated that the antiUKp83/68 antibodies detect UKp68 and UKp83
specifically. The antibodies were investigated on the
same conditions as described above except for
histological preparations. The GFP signals of the GFP
fusion protein of UKp83 and UKp68 were detected at
J Med Dent Sci
only nuclei of the cells transfected, respectively, and
the signals of Luc fusion protein were detected in the
cytoplasm of the transfected cells (Fig. 2B). On the
other hand, the anti-UKp83/68 antibodies detected
strongly UKp83 and UKp68 that are expressed as the
GFP fusion proteins, respectively, however, did not
react with GFP-Luc fusion protein (Fig. 2B). And antiUKp83/68 antibodies expressed signals in the nuclei of
all the cells, with or without any fusion proteins as well
as the experiments of an immunoblotting. It is thought
that anti-UKp83/68 antibodies have detected the
nuclear localization of native UKp83 and UKp68 of
HEK293T (Fig. 2B).
Comparison of the nuclear localization of UKp83/
68 in various cell types
Figure 3 shows the results of immunostaining of
UKp83/68 in various cells. UKp83/68 was expressed
in all cells examined, although its localization in the
nucleus was variable.
These staining patterns can be classified into three
types. The first is a thick fibrous type shown in Figures
3a-e and is found in pyramidal cells of the cerebrum,
Purkinje’s cells of the cerebellum, skeletal muscle cells
of the tongue, heart muscle cells, and epidermal cells of
the skin. The second type is characterized by fine mesh
or small spots as shown in Figures 3i-n. This type is
found in epithelial cells of the stomach, pancreatic
acinar cells, epithelial cells of the lung, proximal tubular
cells of the kidney, spermatocytes of the testis, and
lymphocytes of the spleen. The last type is intermediate
in form between the first and the second as shown in
Figures 3f-h and is found in hepatocytes, the duodenum,
and the colon. The staining patterns of UKp83/68 in
some cells are clearly different relative to the
neighboring cells of the same tissue (Figs. 3n and 3o).
In this classification, all the ectodermal tissues
investigated in this experiment were classified into the
first type, however, the endodermal tissues were not
into the type at all. On the other hand, the mesodermal
tissues were classified into the first type and the
second type, but there was no intermediate type.
Although the significance of differences in the staining
patterns in various cells is unclear, these results may
suggest that the distribution of UKp83/68 in cell nuclei
reflects a state of cellular activity.
UKp83/68 is associated with chromosomal DNA
throughout the cell cycle
We next examined the intracellular localization of
UKp83/68 during mitosis using an immunohistochemical
Molecular Characterization of UKp83/68
49
Figure 2. Identification of UKp83/68 in HEK293T cells expressed with GFP fusion protein of UKp83, UKp68, or
Luciferase, respectively
(A) Proteins in HEK293T cells which were transfected with expression vector for GFP fusion protein of UKp83 (GFPUKp83), UKp68 (GFP-UKp68), and Luciferase (GFP-Luc), respectively, were fractionated by SDS-PAGE and visualized by
immunostaining with two antibodies, anti-UKp83/68 (left panel) and anti-GFP (right panel). Proteins in HEK293T cells with
no treatment were also prepared with the same manner above, as a negative control (NT). The asterisks indicate fusion
proteins visualized and the arrows indicate native proteins. (B) Immunohistochemical analysis of UKp83/68 in HEK293T
cells transfected as described in (A) was performed. HEK293T cells were stained with anti-UKp83/68 antibodies (antiUKp83/68) (red). The localization of GFP fusion proteins was also examined by direct GFP fluorescence (GFP) (green) and
DAPI (blue) was used to stain DNA and indicate the position of the nucleus. A merged image is also shown.
method. The localization of UKp83/68 displayed
characteristic changes during mitosis in rat liver cells
(Ac2F). During interphase, the protein was stained in
the nucleoplasm with a finely punctate pattern (Fig. 4,
interphase). In the prophase, the protein was partially
diffused and formed fibrous structures (Fig. 4,
prophase). When the nuclear envelope disintegrated
during prometaphase, most of the UKp83/68 appeared
to be bound to the chromosomes, although partial
dispersal of the protein was visible throughout the
cytoplasm, accompanied by formation of fibrous
structures (Fig. 4, prometaphase). This profile was
retained until telophase, and during cytokinesis, the
protein re-concentrated around the chromosomal
masses with a finely mesh pattern in the daughter
nuclei (Fig. 4, cytokinesis).
Colocalization of UKp83/68 with splicing factor
SC35
When nuclei were observed at high magnification in
histochemical studies indicated the nuclear localization
of UKp83/68, the staining pattern obtained with the
anti-UKp83/68 antibodies was found to be finely
punctate in certain regions in rat liver cells, Ac2F (Fig.
5A). The observation that UKp83/68 was distributed
with mottled pattern in the nucleoplasm prompted us to
compare the localization of UKp83/68 with the
distribution of SC35, a known marker of nuclear
speckles, which represent subnuclear compartments
enriched in snRNPs and other splicing factors
50
S. Miyakura et al.
J Med Dent Sci
Figure 3. Intranuclear localization of UKp83/68 in various rat tissues
Various rat tissue sections were stained with anti-UKp83/68 antibodies (green) and counterstained with DAPI (red), and a
merged image is also shown; a: pyramidal cell of cerebrum, b: Purkinje’s cell of cerebellum, c: skeletal muscle cell of
tongue, d: heart muscle cell, e: epidermal cell of skin, f: hepatocyte, g: epithelial cell of duodenum h: epithelial cell of colon,
i: epithelial cell of stomach, j: pancreatic acinar cell, k: epithelial cell of lung, l: proximal tubular cell of kidney, m:
spermatocyte of testis, n and o: lymphocytes of spleen. Down the figure, the images of three staining types classified in
the text were shown as follows, fibrous; thick fibrous type, intermediate; intermediate type, and mesh or spot; fine mesh or
small spots type.
(spliceosomes).
By immunohistochemical examination of the rat liver
using confocal laser scanning microscopy, we directly
compared the localization of UKp83/68 with that of
SC35. As illustrated in Figure 5A, it could be said that
the intranuclear distribution of the two proteins is
almost identical.
UKp83/68 binds poly(A) + and ssDNA in vitro
To test whether the CCCH motif in UKp83/68 has
functional RNA/ssDNA-binding activity, we performed
an in vitro RNA/DNA-binding assay. The GSTUKp68(AA115-605) fusion protein was incubated with
synthetic polynucleotides bound to resin beads. After
the beads were washed with buffers containing various
concentrations of NaCl, the fusion protein bound to
the resins was analyzed by SDS-PAGE as described
in Materials and Methods. As shown in Figure 5B, in
50 mM NaCl, the fusion protein bound to the poly(A)
resins most efficiently (70% of the added protein) and
more or less to all of the polynucleotides tested. In
0.1 M NaCl, there was essentially no protein bound to
poly(G) and dsDNA, whereas the poly(A) and ssDNA
resins retain the protein even in 0.3 M NaCl. On the
other hand, rGST without the fusion of UKp68 did not
bind to any of the polynucleotides (Fig. 5B). Thus, GSTUKp68(AA115-605), which contains the zinc finger
domain, but not GST alone, bound most efficiently to
poly(A) and ssDNA, suggesting that the CCCH motif may
be responsible for the binding to these polynucleotides.
Expression of UKp83-antisense causes cell death
To estimate the function of UKp83, we prepared four
constructs as shown in Figure 6A and transiently
expressed them in HEK293T cells. GFP-UKp83, full
length UKp83 fused with GFP, and GFP-UKp83c,
lacking N-terminal domain (but retaining the nuclear
localiz at ion s ignal) appear ed t o localiz e i n t h e
nucleoplasm with a fibrous structure visualized by
fluorescence microscopy (Fig. 6B). When UKp83 cDNA
Molecular Characterization of UKp83/68
51
Figure 4. Intracellular localization of UKp83/68 during mitosis
Cells of the rat hepatocyte cell line Ac2F were fixed, stained with anti-UKp83/68 antibodies (green), and counterstained
with DAPI (red) and a merged image is also shown. Cell images at each mitotic phase were selected.
was expressed in HEK293T cells with ligation in the
reverse direction (i.e., sense sequence of GFP ligated
with antisense sequence of UKp83, GFP-AS), where a
stop codon appears just downstream of the GFP
sequence, warped cellular and nuclear shapes were
observed as shown in Figure 6B. In these cells, GFP
was localized in the cytoplasm, apparently as a result
of the absence of the nuclear localization signal.
To examine the viability of the transfected cells, we
counted cells after transfection. One day after
transfections, there were no significant differences in
the fractions of GFP-positive cells among the cells
expressing GFP-UKp83, GFP-UKp83c, GFP-Luc, and
GFP-AS. However, three days after transfection, the
fractions of GFP-positive cells sharply declined with
respect to GFP-UKp83c and GFP-AS expression
whereas the decrease in the fractions of GFP-positive
cells was only slightly with respect to GFP-UKp83 and
GFP-Luc expression (Fig. 6C).
Then, we investigated whether the decreases in the
fractions of GFP-positive cells with respect to
expression of GFP-AS and GFP-UKp83c were due to
apoptosis by examining nuclear DNA fragmentation in
situ . The transfected cells were subjected to TdTmediated dUTP nick end labeling (TUNEL) with
phycoerythrin (PE) and analyzed by flow cytometry
according to the fluorescence intensities of GFP and
PE (Fig.6D and 6E). The percentage of cells labeled
with PE was calculated for both the GFP-positive and
GFP-negative groups, respectively (Fig. 6F). When
either GFP-UKp83 or GFP-Luc was expressed, the
fraction of PE-positive labeling in GFP-positive cells
was similar to that observed in GFP-negative cells and
essentially remained the same for the 3 days after
transfection (~10%, Fig. 6F). On the other hand, the
fractions of PE-positive labeling in both GFP-UKp83cand GFP-AS-expressed cells drastically increased by
30–40% during the same period.
52
S. Miyakura et al.
J Med Dent Sci
Figure 5. Comparison of the localization between UKp83/68 and splicing factor SC35 and Binding of GSTUKp68(AA115-605) fusion protein to polynucleotide
(A) Cells of the rat hepatocyte cell line Ac2F were fixed and double-stained with anti-UKp83/68 antibodies and anti-SC35
monoclonal antibody. The images of two nuclei are shown as representative examples.
(B) (Left panel) GST-UKp68(AA115-605) fusion protein was incubated with resin beads, which bind to one of the
polynucleotides indicated. The fusion protein bound to the beads was analyzed by SDS-polyacrylamide gel electrophoresis
after elution with buffers containing various concentrations of NaCl (0.05, 0.1, 0.3, or 1.0 M) as described in Materials and
Methods. Recombinant GST (rGST) was treated in the same way as the fusion protein and used as a control. Total: the
fusion protein or rGST untreated with resins. (Right panel) The intensities of the fusion protein bands were evaluated by
densitometric analysis with “total” set at as 100%.
Discussion
We have identified and characterized a nuclear
protein, UKp83/68. This is the first study describing the
molecular characterization of rat UKp83/68. A notable
feature of this protein is that it contains a CCCH zinc
finger domain. This zinc finger motif interacts with RNA
in a number of regulatory proteins of eukaryotic cells
(Hall, 2005). The best characterized group of proteins
involved in these interactions is the tristetraprolin
(TTP) family of mRNA regulators, which destabilize
specific cytokine transcripts involved in immune
responses (Taylor et al., 1991). The destabilization
is triggered by binding of TTP to AU-rich elements
in the 3′-untranslated region of target mRNAs and
recruitment of mRNA degrading machinery (Liang et
al., 2009; Cao et al.,2007). In Caenorhabditis elegans ,
a TTP homolog, MEX-5, and other CCCH-type zinc
finger proteins regulate the expression of maternal
mRNA, which controls cellular differentiation in early
development (Pagano et al., 2007). In addition to the
TTP family of proteins, several other groups of CCCHcontaining molecules with various functions have
been characterized in eukaryotic cells. These include
proteins related to mRNA processing machinery such as
Nab2 (Kelly et al., 2007) (which controls the length of
poly(A) tails and export of mRNAs in yeast), Muscleblind
(Pascual et al., 2006) (which regulates alternative
splicing), and ZAP (Gao et al., 2002) (which provides
protection against viral infection by destabilizing
Molecular Characterization of UKp83/68
Figure 6. Expression of GFP-UKp83 fusion protein and UKp83 antisense RNA in HEK293T cells.
(A) Schematic representation of the plasmids used for transfection experiments. UKp83 and luciferase cDNAs were ligated
downstream of the GFP gene of the pLP-EGFP-C1 vector, respectibly. GFP-UKp83: GFP + full-length UKp83, GFPUKp83c: GFP + UKp83 lacking amino terminal domain (291 amino acids), GFP-AS: GFP + UKp83 in the reverse direction,
GFP-Luc: GFP + luciferase. NLS: nuclear localization signal; CCCH: Zinc-finger motif. (B) The cells transfected with
plasmids shown in (A) were fixed at day 1 and counterstained with DAPI. GFP and DAPI are shown as green and blue,
respectively, and a merged image is also shown. (C) Diurnal proportion alteration of GFP-positive cells in the total cell
number. Each plasmid shown in (A) was transfected to HEK293T cells (Day 0). From the next day, the diurnal cell
proportion of GFP-positive in the total cell number was analyzed by each plasmid. Data are expressed as the mean ± SEM
of three independent experiments. *P < 0.05 versus GFP-Luc and GFP-UKp83. (D) Condition setting of flow cytometric
analysis. The cells (HEK293T) transfected with plasmids shown in (A) were fixed, treated with TdT-mediated dUTP nick
end labeling (TUNEL) with phycoerythrin (PE), and then analyzed by flow cytometry. In the left panel, a histogram of SS/FS
is shown. By setting the gate to the A region, cell debris were excluded. For analysis of DNA fragmentation, 5000 cells of
A region were targeted. Relationships between signal intensity (horizontal axis) and cell number (ordinate) of cells gating
the A region are represented at GFP/Count and PE/Count. Black line shows the distribution of untreated cells. Green line
shows the distribution of transfected cells for GFP. B represents positive for GFP intensity. Red line shows the distribution
of cells in the TUNEL method with UV irradiation. C represents positive for PE intensity. In the right panel, a dot plot of
GFP/PE is shown. Two lines in the plot indicate the boundary B and C, respectively, and they divide the plot area into four
regions indicated by 1-4. (E) Dot plots of GFP/PE of the actual experimental samples. Plots were expressed with GFP
intensity (horizontal axis) and PE intensity (vertical axis). The type of the transfected plasmid is shown above the plots and
the number of days from the date of transfection is shown rightward. The number of cells was counted for each of the four
regions 1-4 for each plot. (F) DNA fragmentation is expressed as percentage of PE positive cells. Percentage of PE
positive cells in GFP-positive cells (closed circle) is calculated by 2/(2+4) from the cell number counted in the same
number's region of each plot and percentage of PE positive cells in GFP-negative cells (open circle) is calculated by 1/
(1+3) in the same manner. Data are expressed as the mean ± SEM of three independent experiments. *P < 0.05 versus
GFP-negative cells and GFP-positive cells transfected with GFP-Luc or GFP-UKp83.
53
54
S. Miyakura et al.
retroposon transcripts).
UKp83/68, which contains five CCCH zinc fingers,
was shown to bind to poly(A) in vitro . In UKp83/68, five
repeats of a CCCH motif are located at the C-terminal
domains with a consensus sequence of CX5CX4-5CX3H,
whereas in many RNA-binding proteins of the TTP
family, there are two repeats of a CCCH motif with a
CX 8CX 5CX 3H consensus (Lian et al., 2009; Pagano
et al., 2007; Blackshear et al., 2005; Stumpo et al.,
2004). In yeast RNA polymerases I, II, and III, a single
CCCH motif exists in each large subunit, with a CX2CX612 CX 2 H consensus (Yano and Nomura, 1991). The
consensus motif, CX5CX4-5CX3H, found in UKp83 is also
found in Nab2 (Anderson et al., 1993), and a degree
of sequence similarity was identified between the two
proteins in the zinc finger region. Nab2 contains seven
tandem CCCH zinc fingers, additional potential RNAbinding motifs, and an arginine-glycine-glycine (RGG)
repeat domain, but it lacks RRM domains (Anderson et
al., 1993; Marfatia et al., 2003). Nab2 was originally
identified as a component of a heterogeneous nuclear
ribonucleoprotein (hnRNP), which was copurified
with poly(A)+ RNA transcripts (Anderson et al., 1993).
Subsequent studies revealed that Nab2 shuttles
between the nucleus and the cytoplasm and is required
for both nuclear export and proper polyadenylation of
mRNA transcripts (Batisse et al., 2009).
UKp83/68 also has significant single-stranded DNAbinding activity. Although RNA is the target of most zinc
finger proteins of the CCCH type, certain proteins such
as Mcm10 are reported to be DNA-binding proteins,
which function in DNA replication (Warren et al., 2008).
Thus, we should keep in mind the ssDNA-binding
activity of UKp83/68.
Immunolocalization studies in interphase cells
revealed that UKp83/68 was present in the nucleus
and particularly enriched in distinct nuclear domains
but lacking in nucleoli and heterochromatin. This
intranuclear localization pattern is reminiscent of that
observed with antibodies directed against proteins
involved in the splicing of pre-mRNAs (Bogolyubova et
al., 2009). The localization profile of UKp83/68 in the
nucleus is similar to that of SC35, one of the typical
splicing factors.
Moreover, the results obtained from
immunolocalization studies of UKp83/68 during mitosis
showed that UKp83/68 was mostly distributed diffusely
throughout the cytoplasm. This behavior of UKp83/68
also appears to be similar to that of most splicing
factors (Bubulya et al., 2004). However, a significant
portion of UKp83/68 remains on the chromosomes
J Med Dent Sci
during mitosis and shows an extensive overlap with
DNA stained by DAPI.
These results suggest the possibility that UKp83/68
has an important role in pre-mRNA processing through
binding to polyA tails of mRNAs. And, previous study
revealed that the transcripts of human ZC3H14 which
is a human homolog of UKp68 also colocalize with
SC35 and play a role in mRNA processing (Leung et al.,
2009) strongly supports our findings. And, the retention
of UKp83/68 in chromatin during mitosis may suggest
another role of this protein in association with mitotic
chromatin.
Northern blot analysis of UKp83/68 indicated that
the transcription of the UKp83/68 gene is tissue
specific, the highest expression being observed in the
testis. The profiles of alternative splicing were also
tissue specific. In most tissues including the liver, heart,
kidney, testis and spleen, 2.6K and 3.4K transcripts
encoding UKp83 were predominant, while the 2.2K and
3.0K transcripts encoding UKp68 were the major
transcripts in the cerebrum and cerebellum. The tissue
specificity of distinct splicing variants of UKp83/68
expression might suggest that if UKp83/68 is involved
in mRNA processing, it is involved in the processing of
a specific group of mRNAs. In this respect, it is quite
important to identify the RNA and protein species that
UKp83 or UKp68 can bind to.
Another structural feature of UKp83/68 is a PWI
domain which is a RNA/DNA-binding domain that has an
equal preference for single- and double-stranded
nucleic acids and is likely to have multiple important
functions in pre-mRNA processing (Szymczyna et al.,
2014). Although we do not presently know whether
UKp83/68 can interact with DNA or RNA, this structure
might provide clues for elucidating the function of
UKp83/68.
In the transfection analysis indicated in figure 6, there
was no differences about percentage of PE-positive
cells between GFP-positive and negative cells, which
transfected with GFP-UKp83 or GFP-Luc expression
vector, for three days observed (Fig. 6F). The cells,
which expressed either of GFP-UKp83 or GFP-Luc,
showed that fragmentation of DNA occurred to the
same degree as those negative cells. On the other
hand, there were about 20-30% increase of PE-positive
cells in GFP-positive cells compared with PE-positive
cells in GFP-negative cells, which transfected with
GFP-UKp83c or GFP-AS expression vector, at the day
3 (Fig. 6F). These indicate that fragmentation of DNA
occurred to the high frequency in cells that expressed
either of GFP-UKp83c or GFP-AS rather than those
Molecular Characterization of UKp83/68
negative cells. It is expected that transfection of GFPUKp83c and GFP-AS expressed cells c-terminal side
of UKp83/68 conjugated with GFP and antisense RNA
of UKp83/68, respectively. Therefore, as a cause
that introduced those DNA fragmentations, it could be
considered that endogenous UKp83/68 was influenced
as an effect of dominant negative by GFP-UKp83c
and as a suppression effect of gene expression by
GFP-AS, respectively. Then we investigated the
effects on the endogenous protein by immunoblot and
immunohistochemistry, however, we were not able to
detect them. Because it can be considered as reasons
that the transfection efficiency was low, acceleration
of proteolysis was high by cell death, and the titer of
antibodies was low. NAB2 is considered to be putative
orthologue of UKp83/68 in a yeast (Kelly et al., 2014),
and to be essential for cell growth (Anderson et al.,
1993), so it is also expected that the native role of
UKp83/68 is closely related to cell viability.
UKp83/68 is ubiquitously expressed in all tissues
examined at both the mRNA and protein levels.
Although further investigations are required to identify
the specific function of UKp83/68, it is very clear that
the proteins play important roles in a fundamental
cellular process.
Acknowledgments
We thank Mr. Akio Noto, Mr. Hiratsugu Yokota, and
Ms. Sachiyo Ohtani for their assistance.
References
Altschul, SF, Gish, W, Miller, W, Myers, EW, Lipman, DJ. Basic
local alignment search tool. J Mol Biol 1990; 215(3): 40310.
Anderson, JT, Wilson, SM, Datar, KV, Swanson, MS (1993)
NAB2 - A yeast nuclear polyadenylated RNA-binding
protein essential for cell viability. Mol. Cell. Biol., 13,
2730-2741.
Batisse, J, Batisse, C, Budd, A, Boettcher, B, Hurt, E (2009)
Purification of nuclear poly(a)-binding protein nab2
reveals association with the yeast transcriptome and a
messenger ribonucleoprotein core structure. J. Biol.
Chem., 284, 34911-34917.
Blackshear, PJ, Phillips, RS, Ghosh, S, Ramos, SB, Ramos, SV,
Richfield, EK, Lai, WS. Zfp36l3, a rodent X chromosome
gene encoding a placenta-specific member of the
Tristetraprolin family of CCCH tandem zinc finger
proteins. Biol Reprod 2005; 73(2): 297-307.
Bogolyubova, I, Bogolyubov, D, Parfenov, V. Localization of
poly(A)(+) RNA and mRNA export factors in interchromatin
granule clusters of two-cell mouse embryos. Cell Tissue
55
Res 2009; 338: 271–281.
Brown, RS. Zinc finger proteins: getting a grip on RNA. Curr
Opin Struct Biol 2005; 15(1): 94-8.
Bubulya, PA, Prasanth, KV, Deerinck, TJ, Gerlich, D,
Beaudouin, J, Ellisman, MH, Ellenberg, J, Spector, DL.
Hypophosphorylated SR splicing factors transiently
localize around active nucleolar organizing regions in
telophase daughter nuclei. J Cell Biol 2004; 167(1): 5163.
Cao, H, Deterding, LJ, Blackshear, PJ. Phosphorylation site
analysis of the anti-inflammatory and mRNA-destabilizing
protein tristetraprolin. Expert Rev Proteomics 2007; 4(6):
711-26.
Dingwall, C, Robbins, J, Dilworth, SM, Roberts, B, Richardson,
WD. The nucleoplasmin nuclear location sequence is
larger and more complex than that of SV-40 large T
antigen. J Cell Biol 1988; 107(3): 841-9.
Frohman, MA, Dush, MK, Martin, GR. Rapid production of fulllength cDNAs from rare transcripts: amplification using a
single gene-specific oligonucleotide primer. Proc Natl
Acad Sci U S A 1988; 85(23): 8998-9002.
Gadal, O, Nehrbass, U. Nuclear structure and intranuclear
retention of premature RNAs. J Struct Biol 2002; 140(1):
140-6.
Gao, G, Guo, X, Goff, SP. Inhibition of retroviral RNA production
by ZAP, a CCCH-type zinc finger protein. Science 2002;
297(5587): 1703-6.
Hall, TM. Multiple modes of RNA recognition by zinc finger
proteins. Curr Opin Struct Biol 2005; 15(3): 367-73.
Hara, M, Igarashi, J, Yamashita, K, Iigo, M, Yokosuka, M, OhtaniKaneko, R, Hirata, K, Herbert, DC. Proteins recognized by
antibodies against isolated cytological heterochromatin
from rat liver cells change their localization between cell
species and between stages of mitosis (interphase vs
metaphase). Tissue Cell 1999; 31(5): 505-13.
Kelly, SM, Leung, SW, Pak, C, Banerjee, A, Moberg, KH, Corbett,
AH. A conserved role for the zinc finger polyadenosine
RNA binding protein, ZC3H14, in control of poly(A) tail
length. RNA 2014; 20(5): 681-8.
Kelly, SM, Pabit, SA, Kitchen, CM, Guo, P, Marfatia, KA,
Murphy, TJ, Corbett, AH, Berland, KM. Recognition of
polyadenosine RNA by zinc finger proteins. Proc Natl
Acad Sci U S A 2007; 104(30): 12306-11.
Laemmli, UK. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 1970;
227(5259): 680-5.
Liang, J, Lei, T, Song, Y, Yanes, N, Qi, Y, Fu, M. RNAdestabilizing factor tristetraprolin negatively regulates
NF-kappaB signaling. J Biol Chem 2009; 284(43): 2938390.
Marfatia, KA, Crafton, EB, Green, DM, Corbett, AH. Domain
analysis of the Saccharomyces cerevisiae heterogeneous
nuclear ribonucleoprotein, Nab2p. Dissecting the
requirements for Nab2p-facilitated poly(A) RNA export. J
Biol Chem 2003; 278(9): 6731-40.
Pagano, JM, Farley, BM, McCoig, LM, Ryder, SP. Molecular
basis of RNA recognition by the embryonic polarity
56
S. Miyakura et al.
determinant MEX-5. J Biol Chem 2007; 282(12): 888394.
Pascual, M, Vicente, M, Monferrer, L, Artero, R. The Muscleblind
family of proteins: an emerging class of regulators of
developmentally programmed alternative splicing.
Differentiation 2006; 74(2): 65-80.
Rojas, A, Schachterle, W, Xu, SM, Black, BL. An endodermspecific transcriptional enhancer from the mouse Gata4
gene requires GATA and homeodomain protein-binding
sites for function in vivo. Dev Dyn 2009; 238(10): 258898.
Leung, SW, Apponi, LH, Cornejo, OE, Kitchen, CM, Valentini, SR,
Pavlath, GK, Dunham, CM, Corbett, AH. Splice variants of
the human ZC3H14 gene generate multiple isoforms of a
zinc finger polyadenosine RNA binding protein. Gene
2009; 439(1): 71-8.
Stumpo, DJ, Byrd, NA, Phillips, RS, Ghosh, S, Maronpot, RR,
Castranio, T, Meyers, EN, Mishina, Y, Blackshear, PJ.
Chorioallantoic fusion defects and embryonic lethality
resulting from disruption of Zfp36L1, a gene encoding a
CCCH tandem zinc finger protein of the Tristetraprolin
J Med Dent Sci
family. Mol Cell Biol 2004; 24(14): 6445-55.
Szymczyna, Bowman, J, McCracken, S, Pineda-Lucena, A, Lu,
Y, Cox, B, Lambermon, M, Graveley, BR, Arrowsmith, CH,
Blencowe, BJ. Structure and function of the PWI motif: a
novel nucleic acid-binding domain that facilitates premRNA processing. Genes Dev 2003; 17(4): 461-75.
Taylor, Lai, WS, Oakey, RJ, Seldin, MF, Shows, TB, Eddy,
Blackshear, PJ. The human TTP protein: sequence,
alignment with related proteins, and chromosomal
localization of the mouse and human genes. Nucleic
Acids Res 1991; 19(12): 3454.
Warren, EM, Vaithiyalingam, S, Haworth, J, Greer, B, Bielinsky,
AK, Chazin, WJ, Eichman, BF. Structural basis for DNA
binding by replication initiator Mcm10. Structure 2008;
16(12): 1892-901.
Yano, R, Nomura, M. Suppressor analysis of temperaturesensitive mutations of the largest subunit of RNA
polymerase I in Saccharomyces cerevisiae: a suppressor
gene encodes the second-largest subunit of RNA
polymerase I. Mol Cell Biol 1991; 11(2): 754-64.