4846–4855 Nucleic Acids Research, 2000, Vol. 28, No. 24
© 2000 Oxford University Press
Characterization of the B lymphocyte-induced
maturation protein-1 (Blimp-1) gene, mRNA isoforms
and basal promoter
Chainarong Tunyaplin1, Miriam A. Shapiro2 and Kathryn L. Calame1,2,*
1Department
of Microbiology and 2Integrated Program in Biophysical, Cellular and Molecular Studies, Columbia
University College of Physicians and Surgeons, New York, NY 10032, USA
Received September 18, 2000; Revised and Accepted November 1, 2000
ABSTRACT
Blimp-1 is a transcriptional repressor that is both
required and sufficient to trigger terminal differentiation
of B lymphocytes and monocyte/macrophages. Here
we report the organization of the mouse Blimp-1
gene, an analysis of Blimp-1 homologs in different
species, the characterization of Blimp-1 mRNA
isoforms and initial studies on the transcription of
Blimp-1. The murine Blimp-1 gene covers ~23 kb and
contains eight exons. There are Blimp-1 homologs in
species evolutionarily distant from mouse (Caenorhabditis elegans and Drosophila melanogaster) but no
homolog was found in the unicellular yeast Saccharomyces cerevisiae. The three major Blimp-1 mRNA
isoforms result from the use of different polyadenylation
sites and do not encode different proteins. Run-on
transcription analyses were used to show that the
developmentally regulated expression of Blimp-1
mRNA in B cells is determined by transcription initiation.
Multiple Blimp-1 transcription initiates sites were
mapped near an initiator element and a region conferring
basal promoter activity has been identified.
DDBJ/EMBL/GenBank accession nos AF305534–AF305539
its role as a master regulator of terminal B cell differentiation.
Recently Blimp-1 has also been shown to be required for
terminal differentiation of monocyte/macrophages (11),
suggesting the possibility that it may play a role in terminal
differentiation of many cell lineages. Consistent with this
hypothesis, the Blimp-1 homologs in Xenopus (XBlimp1) and sea
urchin (SpKrox1) are required during early embryo development
of these two organisms (12,13).
Although the function of Blimp-1 has been the subject of
many studies, virtually nothing is known about the structure of the
Blimp-1 gene or regulation of Blimp-1 expression. Here we report
the organization of mouse Blimp-1 gene, the characterization of
the three most abundant Blimp-1 mRNA isoforms as well as a
minor splice variant. We also show that Blimp-1 expression in
terminally differentiated B cells is controlled by transcription
initiation, and identify the transcription initiation sites as well
as the basal promoter of the Blimp-1 gene.
MATERIALS AND METHODS
GenBank accession numbers
The accession numbers for mouse Blimp-1 gene are:
AF305534 for the promoter and exon 1; AF305535 for exon 2
and 3; AF305536 for exon 4; AF305537 for exon 5; AF305538
for exon 6 and 7; and AF305539 for exon 8.
INTRODUCTION
Genomic library screening
Terminal differentiation, giving rise to post-mitotic cells
performing dedicated functions, is a critical developmental
process in multicellular organisms that begins during embryogenesis and continues throughout life. One well-studied
example of terminal differentiation is the maturation of B
lymphocytes into end-stage, antibody-secreting plasma cells.
Although terminal differentiation of B cells involves complex
changes in the expression of many genes (1–4), a single transcription factor, B lymphocyte-induced maturation protein-1
(Blimp-1), has been identified as a ‘master regulator’ of this
process (5). Blimp-1 is a 98 kDa zinc finger-containing protein
that functions as a transcriptional repressor (6–9). Expression of
Blimp-1 is sufficient to trigger terminal differentiation of B-cell
lymphoma line in vitro (5), and the developmental stage-specific
expression of Blimp-1 in B cells in vivo (10) is consistent with
A genomic library from mouse (129 strain) in λfixII phage
(Stratagene, La Jolla, CA) was screened with Blimp-1 cDNA
using standard procedures (14,15). Three overlapping phage
clones, φ21, φ25 and φ17 were obtained. The restriction fragments
containing the Blimp-1 coding region were identified by
hybridization and subjected to DNA sequencing.
Plasmids
For transient transfection experiments, Blimp-pGL3 was generated by cloning the 1.1 kb SacI–ScaI fragment from φ21,
containing an ~900 bp region upstream of transcription initiation
sites, into firefly luciferase reporter pGL3 (Promega, Madison,
WI). The Renilla luciferase reporter pRL-tk was obtained from
Promega. For nuclear run-on analysis, the following plasmids were
used: pIg containing Cµ (16), pSK-Blimp-1 (5), pAlt-GAPDH (a
*To whom correspondence should be addressed at: 701 West 168th Street, Columbia University, HHSC 1202, New York, NY 10032, USA. Tel: +1 212 305 3504;
Fax: +1 212 305 1468; Email: [email protected]
Nucleic Acids Research, 2000, Vol. 28, No. 24 4847
gift from Dr M. Coutts) and pSV2-myc (17). For RNase
protection assay, a 300 bp SacII fragment from the genomic
phage, φ21, encompassing the putative transcriptional initiation sites was cloned into pBluescriptKS (Stratagene).
Cell lines and transient transfection
A plasmacytoma, P3X, and a pre-B cell line, 18-81, were
maintained in RPMI-1640 supplemented with 10% fetal calf
serum (FCS), 50 µM of β-ME and 10 µg/ml gentamicin. NIH
3T3 cells were maintained in DMEM supplemented with 10%
FCS and 10 µg/ml gentamicin. A promonocytic cell line,
U937, was maintained in RPMI-1640 supplemented with 10%
FCS and 10 µg/ml gentamicin. P3X, 18-81 and U937 cells
were transfected by electroporation using a Gene Pulser (Bio-Rad,
Richmond, CA). Briefly, for P3X and 18-81 cells, 3 × 106 cells
were pulsed in 300 µl of complete media at 960 µF and 240 V.
The cells were allowed to recover for 16–18 h in complete media
before harvesting. For U937 cells, 1 × 107 cells were pulsed in
300 µl of complete media at 960 µF and 300 V. The cells were
recovered in complete media for 8 h before harvesting. NIH
3T3 cells were transfected by calcium phosphate–DNA complex
(17). Briefly, 3 × 105 cells were incubated with DNA–calcium
phosphate complex for 12–14 h. The media were changed and
cells were harvested 24 h later. In all transfections, pRL-tk
(Promega) encoding Renilla luciferase under control of HSV-tk
promoter was co-transfected to correct for the transfection
efficiency.
Luciferase assay and data analysis
After transient transfection, cells were harvested and lysed in
passive lysis buffer (Promega). The levels of firefly and
Renilla luciferase in each sample were assayed as described
(17,18) and the light units were collected for 10 s with
luminometer (Berthold Lumat LB9501). The relative light
units from firefly luciferase were corrected for the transfection
efficiency by the relative light units obtained from Renilla
luciferase. All transfections were performed at least twice in
triplicate with two to three preparations of DNA.
Northern analysis of P3X RNA
Total RNA was prepared from P3X by Trizol (Gibco, Rockville,
MD). Total RNA (10–20 µg) was electrophoresed in a formaldehyde gel with 0.9% agarose (19) and transferred to a
Hybond-N membrane (Amersham Pharmacia Biotech, Piscataway, NJ). The hybridization was performed in 50% formamide,
5× SSC, 5× Denhardt’s, 1% SDS and 100 µg/ml yeast total
RNA at 45°C for DNA probes or 68°C for RNA probes or in
7% SDS, 10% PEG-8000, 1.5× SSPE and 100 µg/ml salmon
sperm DNA at 68°C. Exon specific probes used were as follows.
The exon 1 probe was a PCR product of φ21 DNA using primers
derived from the 5′ untranslated (UT) region (5′-AGAGTAGTCAGTCGGTCGCTCA-3′ and 5′-GGAGAGAGTACAGGTGGGTCAG-3′). Exons 4 and 8 were generated as
restriction fragments from phage φ25, which contains the
respective exons. Exon 6 and 3′ UT probes were generated
from cloned cDNA. Exon 6 was generated from the cDNA
clone as a 300 bp AvrII–AflIII fragment. The probe for the 5′
most portion of the 3′ UT was a 0.8 kb BglII fragment, the
probe for the middle portion of 3′ UT was a 1 kb BglII–XbaI
fragment and the probe for the 3′ end of the 3′ UT was a 0.4 kb
HincII–XhoI fragment.
Nuclear run-on
Nuclear run-on assays were performed according to standard
protocols (20) with the following modification for the cell lysis
step. One hundred million P3X and 18-81 cells were harvested
and resuspended in TKM (50 mM Tris pH 7.5, 50 mM KCl,
15 mM MgCl2). Then an equal volume of TKM with 60%
sucrose was added and cells were lysed by the addition of 10%
NP-40 to a final concentration of 0.05%. The nuclei were
collected and transcribed in the presence of 100 µCi UTP
(800 Ci/mmol) for 20 min as described (20). The newly transcribed RNA was purified and hybridized to 5 µg of denatured
cDNAs of Blimp-1, Cµ, GAPDH, c-myc and pUC19 immobilized on a nitrocellulose membrane. After the hybridization,
the membrane was washed and treated with RNase as
described (20). To determine the amount of newly transcribed
RNA for each cDNA, the radioactivity was quantified by a
PhosphoImager (Molecular Dynamics, Sunnyvale CA) and
corrected for the uracil content in the cDNAs.
PCR for Blimp-1 isoform
Single-strand cDNA was synthesized from 20 µg of P3X total
cellular RNA as described (21). The PCR reactions were
performed with 1 µg of cDNA in 30 cycles at 94°C for 1 min,
55°C for 1 min and 72°C for 1 min. The PCR primers used
(upstream
were:
5′-GAAGAAACAGAATGGCAAGA-3′
primer), 5′-AGTTGCCCTTCAGGT-3′ ( downstream primer a in
Fig. 3b), 5′-AAGACACTTTCAGACTGGT-3′ (downstream
primer b in Fig. 3b), 5′-CCTAAGAAGCAACACGAA-3′
(downstream primer c in Fig. 3b).
Primer extension
Primer extension was performed with 50 µg of total cellular
RNA as described (22) with the following modification. After
the overnight hybridization of RNA with the primer, the primer
extension reaction was performed using 4 U of Tth DNA
polymerase (Boehringer Mannheim, Indianapolis, IN) in the
Mn2+ buffer supplied by the manufacturer. The extension was
carried out at 50°C for 10 min and then at 65°C for 10 min. The
primer extension product was then electrophoresed on standard
8% PAGE with 7 M urea adjacent to a sequence ladder of the
genomic Blimp-1 gene. The sequencing reaction of genomic
Blimp-1 gene was generated with fmol sequencing kit from
Promega. The sequence of primer used in both primer
extension and sequencing reactions is 5′-CAGTGTCTGTGCGAGCGAGCGAGTGAGCGA-3′ and is specific for
nucleotides 21–50 of cloned Blimp-1 cDNA.
RNase protection assay
RNase protection assay was carried out as described (23).
Briefly, the antisense riboprobe was generated from genomic
template encompassing the putative transcription initiation
sites and hybridized to 20 µg of P3X total cellular RNA. After
overnight hybridization, the reaction was treated with
RNaseA/RNase T1 mixture and the RNase-resistant riboprobe
was electrophoresed on standard 8% PAGE with 7 M urea. The
RNA ladder was synthesized by T3 RNA polymerase
transcription of PvuII-digested pKS and T7 RNA polymerase
transcription of PvuII- or XhoI-digested pKS.
4848 Nucleic Acids Research, 2000, Vol. 28, No. 24
Figure 1. (Above) (a) Organization of the murine Blimp-1 gene. The top panel shows a diagram of Blimp-1 cDNA. The coding region is highlighted in light green,
exon splice junctions are indicated by dotted red lines and the protein domains by colored blocks. The predicted poly(A) addition sites in the 3′ UT are indicated
with asterisks. The bottom panel shows the location of each exon in the three overlapping phage clones, φ21, φ25 and φ17. (Opposite) (b) Comparison of amino
acid sequences of Blimp-1 homologs from mouse (accession no. NP031574), human (accession no. NP001189), frog (accession no. AAF08791), fly (accession no.
AAF50759), worm (accession no. T21336) and sea urchin (accession no. AAF30407). From the lineup, consensus sequence was derived (not shown). Amino acids
identical to the derived consensus sequence are shown in green, amino acids with conservative change are shown in blue and semi-conservative changes in orange.
The functional domains in Blimp-1 protein are indicated. Exon junctions in mouse Blimp-1 protein are indicated by red triangles.
Gene analysis
A BLAST search (24,25) was performed on the whole
GenBank database using the murine Blimp-1 amino acid
sequence. The program used for the search was tblastn. Several
of the species that matched Blimp-1 were translated from
cDNA, however the Drosophila melanogaster and Caenorhabditis elegans amino acid sequences were derived from the
genomic DNA. In an attempt to verify the accuracy of the
predicted proteins, the genomic sequences from D.melanogaster and C.elegans were searched for splice sites using the
Neural Network splice site predictor on the Berkeley
Drosophila Genome Project page (http://www.fruitfly.org/
seq_tools/splice.html) and the NetGene2 World Wide Web
server which produces neural network predictions of splice
sites in C.elegans (http://130.225.67.199/services/NetGene2/).
Sites were used to construct the cDNA if they had a confidence
interval score above 0.90 and resulted in cDNA sequence that
was in-frame with the mouse Blimp-1 sequence. The resultant
amino acid sequence was compared with the amino acid
sequence predicted by the genome projects using the Clustal W
program (26).
RESULTS
Organization of the murine Blimp-1 gene
Three λ phage clones from a 129 mouse genomic phage library
were isolated by virtue of their hybridization with a probe
containing 5.3 kb Blimp-1 cDNA (Fig. 1a). Restriction
mapping showed that these overlapping clones together span
~33 kb of genomic DNA. Restriction fragments encoding
Blimp-1 mRNA were identified by hybridization to cDNA and
sequenced to identify all exon–intron junctions (Fig. 1a). The
Blimp-1 gene contains eight exons (Fig. 1a) spanning ~23 kb and
all exon–intron junctions conform to the gt/ag rule. In general, all
the identified functional domains (N-terminal acidic domain,
PEST domain with proline-rich domain and C-terminal acidic
domain) are encoded in separate exons with the exception of
the PR domain (8) and the five C2H2-type zinc finger domains
(Fig. 1a). The PR domain is encoded in exons 4 and 5 while the
zinc finger domains are encoded in exons 6, 7 and 8. Of particular
interest, zinc finger 1, one of two critical zinc fingers for
Blimp-1’s ability to bind DNA (27), is encoded in exons 6 and 7.
Comparison between the murine and human Blimp-1 genes
(GenBank accession no. AL358952 for human Blimp-1
genomic gene) shows that they have a very similar exon–intron
organization (data not shown). The only exception is that the 5′
UT in human is encoded in only one exon while in the mouse
it is encoded in two exons. Also, we could not find human
sequence encoding the conserved histidines of zinc finger 1 in
the human gene, because the sequence is still incomplete in
this region.
When a BLAST search of the GenBank database was
performed to identify Blimp-1 homologs in other species, nine
sequences received a score that was significant. These
sequences, from human, mouse, frog, worm, fly and sea
urchin, were more related to the mouse Blimp-1 sequence than
the mouse Blimp-1 sequence was to any other zinc finger or
PR family protein. The sequences are 69% identical in the area
of zinc fingers 1–4 and there is considerable homology in the
PR domain and 5′ acidic region as shown in Figure 1b.
Although there is little sequence homology in the proline-rich
domain, the sequence from each species contains many
prolines that are simply not aligned. The human (PRDI-BF1),
Xenopus (Xblimp1) and sea urchin (SpKrox-1) cDNAs have
been reported (7,12,13). However the worm and fly amino acid
and cDNA sequences are predicted from genomic DNA determined
by the C.elegans and Drosophila genome projects (28,29). We
verified these predictions to the best of our ability using neural
networks to predict splice sites. There was one major difference
Nucleic Acids Research, 2000, Vol. 28, No. 24 4849
4850 Nucleic Acids Research, 2000, Vol. 28, No. 24
between the predicted proteins found with the BLAST search
and the proteins that we constructed. In the C.elegans sequence
there is an acceptor site used in the protein from the genome
project that is not recognized by the neural network. The donor
site is predicted and the splicing is necessary for the change of
reading frame that allows the full protein to be produced. Additionally, there are several donor and acceptor sites that may or
may not be used that would add or delete very small sequences
that do not alter the match with mouse Blimp-1 significantly.
However it is impossible to tell if these are included in the endogenous protein by computer predictions alone. Despite our attempts
at predicting the amino acid sequence of the fly and worm Blimp-1,
there are probably errors since there are stretches where these two
sequences match neither each other nor the sequences from any
other species. These non-homologous areas are unlikely to be due
to evolutionary distance since the sea urchin Blimp-1 (SpKrox1)
does not contain these unmatched gaps. Interestingly, the very 3′
end of sea urchin Blimp-1 contains a leucine-zipper that was not
evolutionary conserved.
Blimp-1 has multiple transcription initiation sites and
lacks a TATA element
In order to map the transcription initiation sites of murine
Blimp-1 gene, a primer extension assay was performed. A
primer was chosen from the 5′ UT sequence of the cloned
Blimp-1 cDNA (GenBank accession no. U08185). The results
showed multiple extension products ranging from 40 to 80 bases
(Fig. 2b), suggestive of multiple transcription initiation sites
within a region of ~40 bp. However, some of the complexity
could have been caused by the inability of the reverse transcriptase to transcribe completely in this GC-rich region.
Furthermore, this assay indicates the length of 5′ UT mRNA
from the primer but cannot show whether it is encoded by one
or more exons. Therefore, a riboprobe protection assay was
also performed to confirm the primer extension results and to
map the location of transcription initiation sites in Blimp-1. A
riboprobe from the genomic DNA encompassing the known 5′
UT and the putative initiation sites determined by primer
extension was used (Fig. 2c). Multiple protected fragments
were observed (Fig. 2c), in agreement with the primer extension
results. The size of the two largest protected fragments, 167 and
160 bp ± 7 bp, is consistent with the size of the largest primer
extension products (Fig. 2b). Taken together, these data show that
Blimp-1 is transcribed from multiple transcription initiation sites
located between 244 and 251 bp from the ATG start codon.
The occurrence of multiple transcription initiation sites is the
hallmark of transcription from TATA-less promoters that
usually contain initiator elements (30,31). We examined the
DNA sequence in the vicinity of Blimp-1 transcriptional start
sites and found no TATA sequence. However, there is an
initiator element located 13 bp upstream to the most 5′ transcription initiation site. It is likely that this initiator is responsible for the transcription of Blimp-1 gene (Fig. 2d).
Three major Blimp-1 mRNA isoforms are generated by
differential polyadenylation but a minor splice variant
lacking exon 7 is also expressed
Knowing the exon–intron structure and the location of transcription initiation sites in the gene allowed us to characterize
Blimp-1 mRNA isoforms. Blimp-1 mRNA is found in three
major isoforms of ~5.7, 4.3 and 3.6 kb, determined by northern
blot analysis (Fig. 3a) (11). The original cDNA clone (5) is
5.3 kb exclusive of poly(A), corresponding to the 5.7 kb
mRNA isoform. Probes corresponding to all the proteincoding exons and to the 5′ UT hybridized to all three mRNA
isoforms in northern blots of total RNA from P3X cells (Fig. 3a
and data not shown) suggesting that the three isoforms encode
the same protein. Analysis of the 2.5 kb 3′ UT portion of the
cDNA sequence revealed several potential polyadenylation
sites [poly(A) sites]. We therefore performed northern
analyses using probes from different regions of the 3′ UT to
determine whether the smaller isoforms were generated by the
use of different poly(A) sites. A probe derived from the first
400 bp of the 3′ UT and a probe derived from the middle
portion of the 3′ UT region detected the 5.7 and 4.3 kb
isoforms but did not hybridize to the 3.6 kb isoform. A probe
from the very 3′ end of the cDNA detected only the 5.7 kb
isoform and did not hybridize to the 3.6 or 4.3 kb isoforms.
Thus, the 3.6 kb isoform appears to result from polyadenylation at the poly(A) site located ~3.3 kb of the cDNA. The
4.3 kb isoform could result from usage of intermediate poly(A)
site located at ~4.5 kb and the 5.7 kb isoform results from
usage of the most 3′ poly(A) site. Thus, the three major Blimp-1
mRNA isoforms are generated by differential use of multiple
poly(A) sites and all encode the same protein.
mRNAs encoding some zinc finger-containing proteins are
expressed as alternatively spliced isoforms, differing in the
presence of exons encoding specific zinc fingers (32–34).
Since all of zinc finger 2 and parts of zinc fingers 1 and 3 are
encoded by a small exon, exon 7, we examined whether an
alternatively spliced isoform of Blimp-1 lacking exon 7 might
be expressed. (Due to the small differences in size, such an
isoform would not be revealed by northern analysis.) PCR
primers for cDNA sequence 5′ of the region encoding zinc
finger 1 and 3′ of the region encoding zinc finger 5 were used
(Fig. 3b) in an RT–PCR reaction using RNA from plasmacytoma P3X. Each primer pair gave a product predicted for
full-length Blimp-1 mRNA as well as another product ~120 bp
smaller (Fig. 3b). The size of the smaller RT–PCR product is
consistent with an alternatively spliced isoform lacking exon 7.
When the smaller RT–PCR product was cloned and sequenced,
it was indeed shown to be an alternatively spliced form lacking
exon 7 (Fig. 3c).
RT–PCR (Fig. 3b) indicated that the isoform lacking exon 7
(∆exon7) constitutes <10% of total Blimp-1 mRNA in P3X
cells. Riboprobe protection assays confirmed this proportion
and showed that it did not vary when RNAs from various
mouse tissues were examined (data not shown). The protein
encoded by the ∆exon7 isoform lacks the functional zinc finger
1 and 2 and is not predicted to bind DNA efficiently since
Keller and Maniatis (27) showed that these two zinc fingers of
Blimp-1 are critical for DNA binding activity. Indeed, the zinc
finger domain synthesized in 293T cells from the ∆exon7
cDNA did not bind to the Blimp-1 sequence from the c-myc
promoter while binding of the zinc finger domain synthesized
from the full-length cDNA was clearly evident (data not
shown).
Developmental stage-specific expression of Blimp-1 in B
cells is determined by transcription initiation
Blimp-1 protein and steady-state mRNA are present in plasmacytoma lines but are absent from cell lines representing earlier
Nucleic Acids Research, 2000, Vol. 28, No. 24 4851
Figure 2. Mapping transcription initiation site(s) on the Blimp-1 gene. (a) Rationale for the primer extension and RNase protection assays. The primer for primer
extension is specific to nucleotides 21–50 of cloned cDNA. The riboprobe template was Blimp-1 genomic DNA encompassing the putative transcription initiation
sites mapped by primer extension. (b) Primer extension assay. The first four lanes are a sequence ladder and the next lane size markers. The last two lanes show
extension products derived from total RNA from plasmacytoma P3X or from control tRNA. The arrows indicate the putative transcription initiation sites on the
Blimp-1 sequence. (c) RNase protection assay. The two major protected fragments are indicated by arrows. The sizes of RNA markers are indicated. (d) The
sequence of transcription initiation sites. The sites mapped by primer extension assay are indicated by dots. The sites mapped by RNase protection assay are
indicated by capital letters of the nucleotide sequence. The 5′-most transcription initiation site is assigned the +1 position. The initiator element likely to be responsible
for Blimp-1 transcription initiation is underlined.
stages of B cell development (5,9). To determine if this B cell
stage-specific expression is the result of differential transcriptional
initiation, transcription run-on assays were performed. Nuclei
were isolated from two B-lineage cell lines: P3X, a plasmacytoma
that expresses Blimp-1, and 18-81, a pre-B line that does not
express Blimp-1. The results showed no polymerase loading
on the Blimp-1 gene in 18-81 nuclei, but significant
polymerase loading on the Blimp-1 gene in P3X (Fig. 4). Transcription of Cµ, GAPDH and c-myc genes was also measured
to provide positive controls for the assay. The amount of
4852 Nucleic Acids Research, 2000, Vol. 28, No. 24
Figure 3. Alternative polyadenylation and alternative splicing of Blimp-1 mRNA. (a) Northern blot analysis of Blimp-1 mRNAs. The diagram indicates portions
of cDNA used as probes. Probes specific for the 5′ UT and the coding region were derived from either cDNA or genomic phage clones and are exon-specific. Probes
derived from 3′ UT were derived from the cDNA. The asterisks in the diagram denote the putative poly(A) sites conserved between mouse and human Blimp-1
gene. The size of mRNA [without poly(A) tail] expected from the usage of each poly(A) site is indicated. (b) RT–PCR analysis of Blimp-1 mRNA. The diagram
shows the possible alternative splicing involving exon 7 and the rationale for the PCR primers design. The expected sizes of the PCR products from the cloned
cDNA and the potential ∆exon7 isoform are indicated. The PCR products from each pair of primers were separated on 5% PAGE. +/– RT indicate the PCR
products of cDNA synthesized in the presence or absence of AMV–RT respectively. (c) cDNA sequence and deduced amino acid sequence of the zinc finger
domain of ∆exon7 Blimp-1 isoform. The arrow indicates the exon splice junction.
Nucleic Acids Research, 2000, Vol. 28, No. 24 4853
Figure 4. Polymerase loading on the Blimp-1 gene in B cell lines. Nuclear run-on
assay showing the transcription of Cµ, Blimp-1, GAPDH and c-myc genes in
nuclei from plasmacytoma (P3X) and pre-B (18-81) cells. pUC19 DNA was
used as negative control. The amount of each transcript was quantified using a
PhosphoImager and corrected for uridine content. The relative amount of each
transcript, compared to c-myc, is shown.
Figure 5. Identification of the Blimp-1 basal promoter. Transient transfection
of a luciferase reporter dependent upon the Blimp-1 gene from –918 to +207 bp.
(GenBank accession no. AF305534) in cell lines: P3X, 18-81, NIH 3T3 and
U937. P denotes the reporter containing the promoter region; V denotes the
parental reporter. The data were corrected for the transfection efficiency using
Renilla luciferase and the mean ± standard deviation of triplicate data points is
shown.
polymerase loading on Blimp-1 was 5-fold higher than that on
c-myc, and ~7.5-fold lower than Cµ in P3X cells. These data
show that developmental stage-specific expression of Blimp-1
in B cells is regulated by transcription initiation.
expression of Blimp-1 is controlled by transcription initiation.
Functional analyses showed that a region 900 bp 5′ of the
Blimp-1 transcription initiation sites is sufficient for basal, but
not B-cell stage-specific, promoter activity.
The Blimp-1 basal promoter is contained within 900 bp 5′
of the transcription initiation sites
Evolution of the Blimp-1 gene
A transfection assay was used to search for the Blimp-1
promoter. A region from –918 to +204 bp relative to the most
5′ transcription initiation site of the Blimp-1 gene was engineered into a plasmid containing the firefly luciferase cDNA
but lacking a TATA or initiator element. Upon transfection
into P3X plasmacytoma cells, the plasmid containing Blimp-1
sequence, showed 3–4-fold more luciferase activity than the
parent plasmid after correction for transfection efficiency
(Fig. 5). Thus, the region containing ~900 bp 5′ of the initiation
sites is sufficient to confer basal promoter activity. To determine if
this region also contained elements capable of conferring B-cell
developmental stage-specificity or tissue-specificity, these
plasmids were also transfected into 18-81 preB cells, NIH 3T3
cells and human U937 promonocytic cells, which do not
express endogenous Blimp-1. The Blimp-1 plasmid had
similar promoter activity in all cell lines tested. Thus we
conclude that additional regulatory elements, outside the
1100 bp present on our reporter, are likely to be required for
tissue- and developmental stage-specific expression of Blimp-1.
DISCUSSION
In this report, we analyzed the murine Blimp-1 gene and found
that it contains eight exons, a transcriptional initiator and
multiple transcription initiation sites. The organization of the
human Blimp-1 gene is similar to that of the mouse. The three
major Blimp-1 mRNA isoforms are generated by differential
poly(A) site usage and a minor isoform, encoding a protein
lacking zinc fingers 1 and 2, is generated by differential
splicing. We showed that B-cell developmental stage-specific
Blimp-1 was originally identified in human (7) and murine (5)
cells; subsequently Blimp-1 homologs were identified in
Xenopus (12) and sea urchin (13). In this work we compared
these Blimp-1 homologs as well as those in two more primitive
eukaryotes, D.melanogaster and C.elegans (Fig. 1b) that have
been identified by genome sequencing. However, Blimp-1
homologs were not found in the genomes of lower, unicellular
eukaryotes, such as Saccharomyces cerevisiae, for which
genome sequence is available. Thus, Blimp-1 appears to have
arisen during early evolution of multicellular organisms,
consistent with the suggestion of its role as an inducer for
terminal differentiation.
In all of the Blimp-1 homologs, conservation of each known
protein domain is evident, with the zinc finger domains having
the highest degree of conservation. This domain conservation
suggests that some/many biological functions of Blimp-1 are
likely to be conserved between lower forms and mammals.
Consistent with this idea is the finding of de Souza et al. (12)
that mouse Blimp-1 can substitute XBlimp1 during frog
embryogenesis. In mouse, Blimp-1 is important for terminal
differentiation of B lymphoid and myeloid cells (9,11,35), is
expressed and may play a role in other end-stage cells (11) and
is required for development of the embryo (M.Davis, personal
communication). Blimp-1 is a DNA-binding transcriptional
repressor. One target of Blimp-1-dependent repression in
hematopoietic cells is c-myc (9,11), a gene that plays key roles
in cell growth and apoptosis (36,37) and which is repressed in
terminally differentiated cells. However, we (J.Piskurich and
K.Lin, unpublished data) and others (7) have also identified
other genes that are targets of Blimp-1 repression in mice and
human cells. It is interesting to note that although Xenopus (38)
4854 Nucleic Acids Research, 2000, Vol. 28, No. 24
and Drosophila (39) have c-myc homologs, C.elegans does
not. It will be particularly interesting to learn if Blimp-1 regulates
cell proliferation or terminal differentiation in C.elegans, where,
presumably, there are Blimp-1 targets not related to c-myc.
Blimp-1 mRNA isoforms
Although three major and one minor isoform of Blimp-1
mRNA have been identified, their significance is not readily
apparent. The three major isoforms differ with respect to 3′ UT
sequence but encode the same protein. The three major
isoforms are found in human as well (11) where they are also
likely to be the result of alternative polyadenylation. Although
the 3′ UT region of human Blimp-1 cDNA has not been
cloned, we could identify in the human Blimp-1 gene the same
putative poly(A) addition sequences identified in the mouse 3′
UT region. Different 3′ UT sequence and/or poly(A) could affect
mRNA stability or translatability. However, the proportion of
these isoforms to one another is similar in most tissues of the
adult mouse we have investigated (D.Chang, unpublished
results). The data suggest, therefore, that these isoforms have
neither functional nor regulatory significance.
The minor isoform lacking exon 7 encodes a protein that
does not bind DNA. Theoretically this protein could interfere
with the function of the full-length Blimp-1 either by forming
a heterodimer with altered function or by sequestering Blimp-1associating proteins required for function. A recent report
shows that an aberrant splice product of the zinc finger protein
Ikaros fails to bind DNA and blocks normal Ikaros function by
dimerization (40). At present there is no evidence that Blimp-1
heterodimerizes, making this mechanism unlikely. However,
Blimp-1 does require association with hGroucho (41) and
HDAC (6) in order to function as a transcriptional repressor. If
the small, non-DNA binding form is present at sufficient
levels, it might sequester one or both of these proteins and
block Blimp-1-dependent repression. Our analyses show that
steady-state mRNA encoding the small form is usually 10-fold
lower than that encoding full-length protein, making this
mechanism unlikely unless the small protein is preferentially
translated or has increased stability. When antisera become
available to distinguish the two forms, it will be important to
determine their relative abundance in different cells.
Regulation of Blimp-1 gene expression
In frogs, sea urchins and mice, Blimp-1 mRNA is expressed
early in development (12,13). It is also expressed in various
terminally differentiated cells of mouse and human and is
induced upon terminal differentiation of both B lymphoid and
myeloid cells. Thus, expression of Blimp-1 mRNA is subject
to both lineage and developmental stage-specific regulation.
The data presented here show that the developmental stagespecific expression of Blimp-1 mRNA in B cells is regulated at
the level of transcription initiation. Transfection analyses showed
that a 1.1 kb region containing the Blimp-1 transcription initiation
sites, initiator element, and ~900 bp 5′ to the initiation sites has
basal promoter activity. However, our studies did not reveal
tissue or developmental stage specificity conferred by this
region. More detailed analyses, possibly utilizing transgenic mice,
may be necessary to unravel the transcriptional regulatory
elements of the Blimp-1 gene. It is possible that regulation is
mediated by transcriptional repressors that could not be identified by transient transfection. Furthermore, some or all of the
transcriptional regulation may be mediated by elements such
as enhancers or locus control regions, located distal to the
promoter. The gene organization presented in this study
provides the basis for a systematic search for such regulatory
elements.
ACKNOWLEDGEMENTS
We are grateful to Dr M. Davis for providing the Blimp-1
cDNA, Dr M. Coutts for providing the GAPDH cDNA, Dr G. Siu
for providing the genomic phage libraries and to members of
the Calame laboratory for helpful discussions. This work was
supported by RO1AI4576 to K.C. C.T. was supported by a
Fellowship from the Leukemia and Lymphoma Society (no.
5295-98) and M.S. is supported by NIH Immunology Training
Grant (5T32A107525).
REFERENCES
1. Lamson,G. and Koshland,M.E. (1984) Changes in J chain and mu chain
RNA expression as a function of B cell differentiation. J. Exp. Med., 160,
877–892.
2. Krenacs,L., Himmelmann,A.W., Quintanilla-Martinez,L., Fest,T.,
Riva,A., Wellmann,A., Bagdi,E., Kehrl,J.H., Jaffe,E.S. and Raffeld,M.
(1998) Transcription factor B-cell-specific activator protein (BSAP) is
differentially expressed in B cells and in subsets of B-cell lymphomas.
Blood, 92, 1308–1316.
3. Oliver,A.M., Martin,F. and Kearney,J.F. (1997) Mouse CD38 is
down-regulated on germinal center B cells and mature plasma cells.
J. Immunol., 158, 1108–1115.
4. Sanderson,R.D., Lalor,P. and Bernfield,M. (1989) B lymphocytes express
and lose syndecan at specific stages of differentiation. Cell Regul., 1, 27–35.
5. Turner,C.A.,Jr, Mack,D.H. and Davis,M.M. (1994) Blimp-1, a novel zinc
finger-containing protein that can drive the maturation of B lymphocytes
into immunoglobulin-secreting cells. Cell, 77, 297–306.
6. Yu,J., Angelin-Duclos,C., Greenwood,J., Liao,J. and Calame,K. (2000)
Transcriptional repression by blimp-1 (PRDI-BF1) involves recruitment
of histone deacetylase. Mol. Cell. Biol., 20, 2592–2603.
7. Keller,A.D. and Maniatis,T. (1991) Identification and characterization of
a novel repressor of beta-interferon gene expression. Genes Dev., 5, 868–879.
8. Huang,S. (1994) Blimp-1 is the murine homolog of the human
transcriptional repressor PRDI-BF1 [letter]. Cell, 78, 9.
9. Lin,Y., Wong,K. and Calame,K. (1997) Repression of c-myc transcription
by Blimp-1, an inducer of terminal B cell differentiation. Science, 276,
596–599.
10. Angelin-Duclos,C., Cattaretti,G., Lin,K.-I. and Calame,K.L. (2000)
Commitment of B lymphocytes to a plasma cell fate is associated with
Blimp-1 expression in vivo. J. Immunol., 165, 5462–5471.
11. Chang,D., Angelin-Duclos,C. and Calame,K. (2000) BLIMP-1: trigger for
differentiation of myeloid lineage. Nature Immunol., 1, 169.
12. de Souza,F.S., Gawantka,V., Gomez,A.P., Delius,H., Ang,S.L. and Niehrs,C.
(1999) The zinc finger gene Xblimp1 controls anterior endomesodermal cell
fate in Spemann’s organizer. EMBO J., 18, 6062–6072.
13. Wang,W., Wikramanayake,A.H., Gonzalez-Rimbau,M., Vlahou,A.,
Flytzanis,C.N. and Klein,W.H. (1996) Very early and transient vegetalplate expression of SpKrox1, a Kruppel/Krox gene from
Stronglyocentrotus purpuratus. Mech. Dev., 60, 185–195.
14. Quertermous,T. (1989) In Ausubel,F.M. (ed.), Current Protocols in
Molecular Biology. John Wiley & Son, Inc., New York, Vol. I, Unit 6.1.
15. Strauss,W.M. (1989) In Ausubel,F.M. (ed.), Current Protocols in
Molecular Biology. John Wiley & Son, New York, Vol. I, Unit 6.3.
16. Avitahl,N. and Calame,K. (1996) A 125 bp region of the Ig VH1 promoter
is sufficient to confer lymphocyte-specific expression in transgenic mice.
Int. Immunol., 8, 1359–1366.
17. Artandi,S.E., Cooper,C., Shrivastava,A. and Calame,K. (1994) The basic
helix-loop-helix-zipper domain of TFE3 mediates enhancer-promoter
interaction. Mol. Cell. Biol., 14, 7704–7716.
18. Matthews,J.C., Hori,K. and Cormier,M.J. (1977) Substrate and substrate
analogue binding properties of Renilla luciferase. Biochemistry, 16,
5217–5220.
Nucleic Acids Research, 2000, Vol. 28, No. 24 4855
19. Brown,T. (1989) In Ausubel,F.M. (ed.), Current Protocols in Molecular
Biology. John Wiley & Son, Inc., New York, Vol. I, Unit 4.9.
20. Greenberg,M.E. (1989) In Ausubel,F.M. (ed.), Current Protocols in
Molecular Biology. John Wiley & Son, Inc., New York, Vol. I, Unit 4.10.
21. Klickstein,L.B. (1989) In Ausubel,F.M. (ed.), Current Protocols in
Molecular Biology. John Wiley & Son, Inc., New York, Vol. I, Unit 5.5.
22. Maniatis,T., Fritsch,E.F. and Sambrook,J. (1989) Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory Press, New York,
NY.
23. Gilman,M. (1989) In Ausubel,F.M. (ed.), Current Protocols in Molecular
Biology. John Wiley & Son, Inc., New York, Vol. I, Unit 4.7.
24. Gish,W. and States,D.J. (1993) Identification of protein coding regions by
database similarity search. Nature Genet., 3, 266–272.
25. Altschul,S.F., Gish,W., Miller,W., Myers,E.W. and Lipman,D.J. (1990)
Basic local alignment search tool. J. Mol. Biol., 215, 403–410.
26. Thompson,J.D., Higgins,D.G. and Gibson,T.J. (1994) CLUSTAL W:
improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, position-specific gap penalties and weight
matrix choice. Nucleic Acids Res., 22, 4673–4680.
27. Keller,A.D. and Maniatis,T. (1992) Only two of the five zinc fingers of the
eukaryotic transcriptional repressor PRDI-BF1 are required for sequencespecific DNA binding. Mol. Cell. Biol., 12, 1940–1949.
28. The C.elegans sequencing consortium (1998) Genome sequence of the
nematode C.elegans: a platform for investigating biology. [published
errata appear in Science (1999) 283, 35 and (1999) 283, 2103 and (1999)
285, 1493]. Science, 282, 2012–2018.
29. Adams,M.D., Celniker,S.E., Holt,R.A., Evans,C.A., Gocayne,J.D.,
Amanatides,P.G., Scherer,S.E., Li,P.W., Hoskins,R.A., Galle,R.F. et al.
(2000) The genome sequence of Drosophila melanogaster. Science, 287,
2185–2195.
30. Lo,K. and Smale,S.T. (1996) Generality of a functional initiator
consensus sequence. Gene, 182, 13–22.
31. Javahery,R., Khachi,A., Lo,K., Zenzie-Gregory,B. and Smale,S.T. (1994)
DNA sequence requirements for transcriptional initiator activity in
mammalian cells. Mol. Cell. Biol., 14, 116–127.
32. van Eyndhoven,W.G., Gamper,C.J., Cho,E., Mackus,W.J. and
Lederman,S. (1999) TRAF-3 mRNA splice-deletion variants encode
isoforms that induce NF-kappaB activation. Mol. Immunol., 36, 647–658.
33. Hahm,K., Ernst,P., Lo,K., Kim,G.S., Turck,C. and Smale,S.T. (1994) The
lymphoid transcription factor LyF-1 is encoded by specific, alternatively
spliced mRNAs derived from the Ikaros gene. Mol. Cell. Biol., 14,
7111–7123.
34. Cabanillas,A.M. and Darling,D.S. (1996) Alternative splicing gives rise to
two isoforms of Zfhep, a zinc finger/homeodomain protein that binds
T3-response elements. DNA Cell Biol., 15, 643–651.
35. Schliephake,D.E. and Schimpl,A. (1996) Blimp-1 overcomes the block in
IgM secretion in lipopolysaccharide/anti-mu F(ab’)2-co-stimulated B
lymphocytes. Eur. J. Immunol., 26, 268–271.
36. Dang,C.V. (1999) c-myc target genes involved in cell growth, apoptosis
and metabolism. Mol. Cell. Biol., 19, 1–11.
37. Prendergast,G.C. (1999) Mechanisms of apoptosis by c-myc. Oncogene,
18, 2967–2987.
38. Principaud,E. and Spohr,G. (1991) Xenopus laevis c-myc I and II genes:
molecular structure and developmental expression. Nucleic Acids Res.,
19, 3081–3088.
39. Gallant,P., Shiio,Y., Cheng,P.F., Parkhurst,S.M. and Eisenman,R.N.
(1996) Myc and Max homologs in Drosophila. Science, 274, 1523–1527.
40. Sun,L., Goodman,P.A., Wood,C.M., Crotty,M.L., Sensel,M., Sather,H.,
Navara,C., Nachman,J., Steinherz,P.G., Gaynon,P.S. et al. (1999)
Expression of aberrantly spliced oncogenic ikaros isoforms in childhood
acute lymphoblastic leukemia [see comments]. J. Clin. Oncol., 17,
3753–3766.
41. Ren,B., Chee,K.J., Kim,T.H. and Maniatis,T. (1999) PRDI-BF1/Blimp-1
repression is mediated by corepressors of the Groucho family of proteins.
Genes Dev., 13, 125–137.