| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Zoology and Genetics (J.T.F., R.T.H.) and Biochemistry, Biophysics and Molecular Biology (M.N.-H.), Iowa State University, Ames, Iowa 50011-3260
Address all correspondence and requests for reprints to: M. Nilsen-Hamilton, Iowa State University, Department of Biochemistry, Biophysics and Molecular Biology, 1210 Molecular Biology Building, Ames, Iowa 50011-3260. E-mail: marit{at}iastate.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, Cv, E through L, and d/t-PRP) (7, 8, 9, 10, 11, 12), and the
mitogen-regulated proteins (proliferins; MRP/PLF) (13). MRP/PLFs are
glycoproteins consisting of a polypeptide core of 21 kDa and appear by
gel electrophoresis to be of average sizes varying from 27K to 38K.
Like some other members of the PRL/GH family, the MRP/PLFs are
synthesized in the placenta, and their synthesis is temporally
regulated during gestation (6, 14). Expression of the
mrp/plf genes in the placenta begins as early as E7, peaks
at midgestation (E9–11), then declines through the remainder of
gestation. In situ hybridization and immunohistochemistry
have identified the trophoblastic giant cells as the site of MRP/PLF
synthesis in the placenta, from which it is released into both the
maternal bloodstream and the amniotic fluid (15). MRP/PLFs are believed
to function during pregnancy as placental angiogenesis factors through
their interaction with the mannose-6-phosphate receptor (16) and as
uterine growth factors through another receptor that is specific for
MRP/PLFs and does not recognize mannose-6-phosphate (17). PLF1 also has
been described as having an intracellular role, in preventing muscle
differentiation (18, 19). The mouse genome contains between four and six mrp/plf genes (20) and previous to this report, three complementary DNAs (cDNAs) had been cloned: plf1 from serum-stimulated 3T3 cells (21), plf2 from a Balb/c mouse placental cDNA library (22), and mrp3 from a Swiss mouse embryo fibroblast library (23). The three MRP/PLFs are approximately 98% identical in amino acid sequence and the variations in sequence do not disrupt any predicted structural motifs. Thus it is likely that all three previously identified MRP/PLFs have a similar structure. Each of the identified MRP/PLFs contain three N-glycosylation motifs within the sequence of the mature secreted protein, and all placental and 3T3 forms are glycosylated (13, 22, 24). Because MRP/PLF binds at least two different receptors during gestation, only one of which recognizes mannose-6-phosphate, the level of glycosylation could be important in determining the functional interactions of the MRP/PLFs with their receptors in vivo (16, 17, 25).
Here we describe a fourth mrp/plf messenger RNA (mRNA) (mrp4) that we have cloned, sequenced and characterized. The mature protein produced from this newly discovered member of the mrp/plf gene family is 91% identical in amino acid sequence to the other members of the family. MRP4 lacks two of the three glycosylation sites found on the other MRP/PLFs, and when expressed, it appears to contain less carbohydrate than the other MRP/PLFs. Expression of mrp4 mRNA is consistent with the presence of MRP/PLF protein found in the placenta in the latter half of gestation and that has a lower apparent molecular weight than the plasma form of MRP/PLFs. MRP4 is also the only mrp/plf mRNA species detected in hair follicles of the tail and the ears of adult mice. These results suggest that, as well as functioning to stimulate uterine growth and placental vascularization during pregnancy, the MRP/PLFs also may have a specific function in adult hair follicles. Because mrp4 is the only member of the mrp/plf gene family expressed in the tail and the ear, this gene may play a singular role in these tissues.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Polyclonal anti-MRP rabbit sera and preimmune sera were prepared as described (26). pRSV-plf1 and pRSV-plf2 plasmids and a set of cosmids containing mrp/plf-hybridizing sequences were gifts from Jiandie Lin and Daniel Linzer (Northwestern University). The pBluescript II KS(+) plasmid (pBKSII+) was obtained from Stratagene (La Jolla, CA), and the pcDNA3 expression vector was obtained from Invitrogen (Carlsbad, CA). The full mrp3 gene extending from -1450 to 600 bp past the poly A site (including introns) was obtained from David Denhardt (Rutgers University) in four separate pieces. The full-length genomic clone from -1450 was reconstructed almost completely except for an approximately 230 bp AccI-PvuII fragment of intron I and a 300 bp EcoRI fragment of intron 3, which were removed during cloning. The reconstructed gene was placed downstream of the CMV (cytomegalovirus) enhancer (obtained from pcDNA3) to produce pCMV-MRP3. Tunicamycin was from Sigma (St. Louis, MO). Dulbecco-Vogt’s modified Eagles medium with high glucose (DMEG) was from Life Technologies, Inc. (Rockville, MD).
RNA isolation and RT-PCR
Immediately after their removal, tissues were frozen under
liquid nitrogen and stored at -70 C until use. Frozen tissue was
pulverized in liquid nitrogen using a mortar and pestle, and RNA was
isolated using Tri-Reagent (Life Technologies, Inc.)
according to the manufacturer’s instructions. For RT, 250 ng of total
RNA was incubated with or without 40 U of Superscript II reverse
transcriptase (Life Technologies, Inc.) in 5-µl reaction
mixtures containing 75 mM KCl, 3 mM
MgCl2, 10 mM DTT, 112 ng poly dT
primers, 400 µM each of the four dNTPs (dATP, dGTP, dCTP,
dTTP), and 50 mM Tris·HCl (pH 8.3 at RT). The reaction
mixtures were incubated at 42 C for 50 min, 50 C for 10 min, then 75 C
for 15 min.
For PCR, the following primers were used; mrp/plf downstream exon I (DE1), 5'taagcctgggtaggactctgc3' (+45 to +65), or (DE1B) 5'ctctgcagagatgctcccttc3' (+59 to +79) and mrp/plf upstream exon V (NUE5), 5'catgatatttcagaagcagagcac3' (+778 to +756). Briefly, 0.5 µl of the RT reaction mix was added to a 25-µl volume containing 1 U of Taq Polymerase, 20 pmoles of each primer, 200 µM of each of the four dNTPs, 1.5 mM MgCl2, 50 mM KCl, and 10 mM Tris-HCl (pH 8.3). The thermal cycling conditions were as follows: 95 C for 2 min to denature, then 40 cycles at 95 C (30 sec), 66 C (1 min), and 72 C (1.5 min) with a final 3 min at 72 C. The PCR products were resolved by electrophoresis through a 1.5% agarose gel and examined by ethidium bromide staining.
Diagnostic restriction digestion to distinguish between the mrp/plf
mRNAs
The diagnostic RT-PCR assay for the mrp/plfs is based
on minor differences in the cDNA sequences of these closely related
gene products that result in the presence or absence of restriction
sites within the cDNA that can be used to distinguish between the
mrp/plf mRNAs (Fig. 1
). For
restriction digestion, one-fifth of the reaction mixture from the
RT-PCR reaction was subjected to one additional round of PCR using DE1B
and NUE5 primers and 62.5 µM of each of the
four dNTPs, 0.9 mM MgCl2
and 1 µCi 32P-dCTP in a 25-µl volume. The
sample was heated to 98 C for 4 min, 64 C for 1 min, then 72 C for 20
min. The radiolabeled cDNA was precipitated with 70% ethanol in the
presence of 0.3 M sodium acetate with 10 µg
yeast transfer RNA as a carrier. Digestion was carried out using 1 U
each of BsoFI (New England Biolabs, Inc., Beverly, MA) and
BstXI at 55 C overnight, or 5 U of NdeI
(Life Technologies, Inc.) at 37 C including 100 µg/ml
BSA (New England Biolabs, Inc.). The radiolabeled and
digested products were resolved by electrophoresis through a 2%
agarose gel or an 8% or 10% polyacrylamide gel. The gel was dried,
and the amount of radioactivity associated with each band was
determined using a phosphorimager and ImageQuant software
(Molecular Dynamics, Inc., Sunnyvale, CA). Positive
controls for identification of specific mrp/plf cDNAs were
amplified from pRSV-plf1 or pRSV-plf2 cDNAs, or were produced using
mRNA isolated from COS cells that had been transiently transfected with
pCMV-MRP3. The diagnostic RT-PCR was validated using known mixes of
mrp/plf cDNAs followed by amplification and diagnostic
RT-PCR. The amount of each cDNA was then determined quantitatively
after restriction digestion and found to reflect the expected ratio of
each cDNA.
|
Cloning and sequencing of the mrp4 cDNA
The mrp4 complementary DNA (cDNA) was cloned from day
13 CF-1 mouse placenta. Total placental RNA was isolated and reverse
transcribed as described under RNA isolation and RT-PCR. The
fragment was amplified by PCR using the following primers: DE1;
5'taagcctgggtaggactctgc3' (+45 to +65), ZMRPUE5;
5'gaaaagcttgtcaacaacaaattcaaaag3' (+820 to +801). Twenty-seven cycles
of 95 C (30 sec), 64 C (1 min) and 72 C (1.5 min) were performed using
a low error rate amplification kit (Expand High Fidelity PCR System,
Roche Molecular Biochemicals, Indianapolis, IN). The PCR
product was precipitated with ethanol, then digested 5' and 3' of the
coding regions using PstI and HincII restriction
endonucleases respectively. The resulting cDNA fragment was ligated
into pBKSII+ that had been cut with these same restriction
endonucleases. The plasmid pBKS-mrp4 was sequenced by the DNA Synthesis
and Sequencing Facility at Iowa State University to verify the sequence
of the mrp4 cDNA insert. Two of the clones were sequenced.
Because the two sequences differed in sequence at five base positions,
total mRNA isolated from two different tail samples were amplified by
RT-PCR and the amplicons sequenced. The resulting sequences matched
each other and clone 2 with the exception of a single base at position
221 in clone 2 which was A. The base in this position in the two
amplified total tail RNA samples was G, which matched that of the other
three cloned mrp/plfs. The base at position 221 was
therefore taken to be G because the amplified tail mRNA sequences were
the sum of sequences of many mRNAs in each sample whereas the cDNA
clone 2 was the result of a single copy of one mRNA.
Expression of MRP4 in COS cells
The mrp4 cDNA was subcloned into pcDNA3 by digestion
of pBKS-mrp4 with SmaI and XhoI and insertion of
the mrp4 cDNA into the blunt-ended HindIII and
XhoI sites of pcDNA3 to make pcDNA3-mrp4. pcDNA3-Plf1 was
made by cloning the HindIII/XbaI fragment of
pRSV-Plf 1 into the same sites in pcDNA3. COS cells were transiently
transfected with pcDNA3-MRP4 or pcDNA3-PLF1 using DEAE-Dextran as
coprecipitant (27). After transfection, the cells were grown in DMEG
10% calf serum for 48 h, then switched to 2% calf serum with or
without 5 µg/ml tunicamycin for 24 h. The culture medium was
collected and spun in a microcentrifuge for 1 min and then stored at 4
C. The cells were scraped into lysis buffer (0.5% NP40, 100 U/ml
Trasylol, 10-5
M PMSF, 20 mM HEPES, pH
7.25) and stored frozen at -70 C.
Western blot analysis
Tissue homogenates for Western blot analysis were prepared
as follows: halves of two or more placentae from pregnant mice were
collected on either day 11 or day 14 of gestation, pooled according to
day of gestation and homogenized in the same lysis buffer used for
cultured cells above. Large tissue debris were spun out, and the
supernatants were saved. Transfected COS cell lysates or homogenates
from day 11 or day 14 placentae were separated on a 15%
polyacrylamide-SDS gel as described previously (28). The proteins were
electroblotted to nitrocellulose membranes (Nitrobind; Micron
Separations, Inc., Westborough, MA), and the MRP/PLFs were detected
using rabbit anti-MRP/PLF antiserum at a dilution of 1/200, followed by
treatment with horseradish peroxidase-conjugated protein A
(Sigma) and Western blotting detection reagents
(Amersham International, Buckinghamshire, UK).
Immunohistocemistry
Tissues were fixed in 4% paraformaldehyde in PBS (0.14
M NaCl, 2.7 mM KCl, 4 mM
Na2HPO4, 14.7
mM KH2PO4, pH
7.4) for 1–2 h at 4 C immediately after collection. Samples were then
rinsed in PBS and stored in 70% ethanol until sectioned. For
immunodetection of MRP, 6 µm sections were rehydrated and stained as
previously described (29) using a polyclonal rabbit anti-MRP/PLF serum
(89rb13a) or a preimmune serum from the same animal, each at a dilution
of 1/500. Primary antibody was detected using biotinylated goat
antirabbit and horse radish peroxidase conjugated to avidin, and
visualized by a peroxidase substrate, diamino benzidine
tetrahydrochloride (ABC kit; Vectastain;Burlingame, CA).
Samples were counter-stained with hematoxylin and eosin, dehydrated,
and mounted with permount (Fisher Scientific, Pittsburgh,
PA) with a coverslip on top.
Identification of the promoters for the different mrp/plf genes
Cosmid clones containing mrp/plf-hybridizing
sequences and a reconstructed mrp3 gene from -1450 to +600
bp past the poly A site and genomic DNA were amplified using primers
that recognize the sequence from +310 to +335 (exon 3;
5'acaaaagccccatgagatgcaatac3') and from +455 to +435 (exon 4;
5'actcactagatcgtccagagg3'). PCR-amplified fragments were digested with
EcoRI for 4 h at 37 C and then digested with either
BsoFI or BstXI as described for diagnostic RT-PCR. The
fragments were resolved through a 2% agarose gel. Individual clones
that produced diagnostic PCR digestion patterns consistent with a
particular mrp/plf cDNA were sequenced across the promoter
region using a primer +71 to +43 (UE1;
5'catctctgcagagtcctacccaggcttag3') that hybridizes with the 5' end of
mrp4 and the three mrp/plfs identified previous
to this report (21, 22, 23).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
to determine the relative
proportions of mrp/plf mRNAs expressed in the placenta. By
incorporating 32PCTP into the final round of
amplification, we were able to quantitatively determine, using a
phosphorimager, the amount of each restriction fragment. The major form
of mrp/plf expressed during gestation was mrp3.
This can be seen visually in Fig. 1 in which the majority of the
radioactivity is associated with the 289- and 246-bp fragments that are
diagnostic for mrp3. Less radioactivity was associated with
the 318-bp fragment that is diagnostic for plf1, and no
radioactivity was associated with the 275-bp fragment diagnostic for
plf2. However, we also found a form of mrp/plf
mRNA that was not digested at the internal BsoF1 and BstXI
sites present in the three cloned mrp/plfs. The undigested
form of mrp/plf was found in the placenta at lower levels
before day 11 and peaked between days 12 and 14 (Fig. 1) (14). These
results suggested the presence of a fourth, previously unidentified,
mrp/plf mRNA that is expressed in the placenta during late
gestation that does not contain the internal BsoFI or BstXI
sites present in the three previously cloned mrp/plfs.
Cloning of mrp4
The fourth mrp/plf cDNA that was identified as uncut by
the diagnostic RT-PCR procedure was cloned from day 13 placental RNA
and sequenced as described in Materials and Methods to
reveal a previously unknown mrp/plf cDNA (mrp4)
with 91% amino acid identity to plf1 (Fig. 2). Two noteworthy differences between
mrp4 and the other three mrp/plfs are changes in
amino acids 77 (Ser to Pro) and 88 (Asp to His) that would disrupt two
of the three putative N-glycosylation sites found on the
other three MRP/PLF mature protein sequences.
|
carbons between 185 and
188.
Mrp4 expression in the placenta
Among the nucleotide sequence differences between the
mrp/plfs is the presence of a NdeI site at +649
that is unique to mrp4. To obtain further evidence that the
mrp4 gene accounts for the uncut portion of cDNA observed
during later gestation, rather than a different mrp/plf that
also lacks the BsofI/BstxI sites, diagnostic RT-PCR was carried out on
placental RNAs using NdeI in place of BsofI and BstxI.
Measured in this way, mrp4 is found at higher relative
levels on days 13–15 than days 9–11 of gestation as was shown
previously with the BsofI/BstxI cleavage (Fig. 1) (14). Thus, both
diagnostic RT-PCR methods identify the same gene product,
mrp4, as having a different expression pattern relative to
plf1 and mrp3.
Measurements of the relative amounts of plf1,
mrp3, and mrp4 in the placenta as a function of
gestation showed that the proportion of mrp3 varied between
about 50% and 75% of the total mrp/plf mRNA throughout
gestation. The relative amounts of plf1 and mrp4
varied reciprocally (Fig. 3). The maximum
proportion of mrp4 was reached on day 12 when it constituted
about 40% of the mrp/plf mRNA in the placenta. At this
time, the proportion of plf1 is at its lowest of about 13%
from a high of about 35% on day 7 of gestation.
|
). As expected, PLF1 migrated with an
apparent molecular weight ranging between 32K and 36K (13). By
comparison, placental minces secrete MRP/PLFs of 27 kDa and 38 kDa. The
38-kDa MRP/PLF band was previously identified as containing MRP3 and
PLF1 (14). MRP4 migrated at the same rate as the 27-kDa placental
MRP/PLF. The 27-kDa MRP/PLF has a time course of appearance in
gestation similar to that of the mrp4 mRNA.
|
). The nonglycosylated forms of the
MRP/PLFs, produced by transfected cells that were incubated with
tunicamycin, migrated as expected for the 21K molecular weight proteins
predicted from their cDNA sequences. These results demonstrate that the
difference in electrophoretic migration rates between the two MRP/PLFs
is due to their glycosylation. Thus, the loss of two glycosylation
sites in the MRP4 sequence results in the addition of less carbohydrate
to this protein compared with PLF1. The placental MRP with the same
apparent size as the recombinant MRP4 produced by COS cells is more
predominant in day 14 placental lysates than day 11 lysates. This is
consistent with the delayed expression of MRP4 mRNA relative to PLF1
and MRP3 in the placenta (Fig. 3) (14).
Differential cell and tissue-specific expression of mrp/plf
genes
Although mrp/plfs were initially demonstrated by
Northern blot analysis to be expressed only in the placenta (6), we
show here that the mrp/plfs are differentially expressed in
a variety of mouse tissues (Fig. 5).
Mrp/plf mRNAs were detected in the small intestine and tail
but not in the liver, lungs, backskin, adult ribs, or large intestine.
The same expression pattern was found in males and females.
Mrp/plfs were detected in some, but not all, stomach samples
and at low levels in some samples of backskin (data not shown).
Mrp/plfs were also detected in 3T3 cells as expected from
previous studies (13). The major mrp/plf expressed by 3T3
cells is plf1 consistent with the cloning of this
mrp/plf from these cells (21). Samples incubated without
reverse transcriptase showed no visible bands (Fig. 5A). Restriction
analysis of radiolabeled cDNAs showed that the mouse tail produces only
mrp4, whereas the small intestine and 3T3 cells produce
mostly plf1 with lower levels of mrp3 (Fig. 5B).
Mrp4 seems to be expressed at lower relative levels in the
small intestine but is absent from 3T3 cells. The PCR product from cDNA
reverse transcribed from whole tail RNA was sequenced, and the results
confirmed that the mrp/plf sequence of the mRNA expressed in
the tail is the same as that of the mrp4 cDNA cloned from
day 13.5 placentae.
|
). Although
mrp/plf cDNA was not detected in most normal backskin
samples (Fig. 5A), mrp/plf cDNAs were regularly identified
in both the ears and tailskin where mrp4 was the only
identified mrp/plf (Fig. 5B). Mrp4 also was the
major mrp/plf species detected in the remainder of the tail
after the skin was removed (data not shown).
Identification of the mrp4 promoter
To identify the mrp4 promoter, a series of cosmid
clones containing mrp/plf sequences were screened for the
presence of the mrp4 gene by several PCR-based assays.
First, diagnostic PCR was performed between exons 3 and 4 on isolated
cosmid clones containing mrp/plf hybridizing genes to
identify specific mrp/plf-containing cosmids. Individual
clones were identified that produced the predicted restriction
digestion pattern for each of the
mrp/plfs. Clear restriction
patterns were obtained for five cosmid clones; two cosmids gave
patterns specific for plf1 (cos4 and cos5), two gave
mrp3-specific patterns (cos2 and cos3) and one gave the
mrp4 restriction pattern (cos6). Clones that produced PCR
products with mrp/plf-specific digestion patterns were also
sequenced across the promoter region after PCR amplification starting
with a primer in the common 5'UTR. By this method, the promoter known
as plf42 was shown to be contiguous with the gene encoding
plf1, the promoter plf149 corresponded to mrp4
and the mrp3 promoter corresponded to mrp3.
To confirm the promoter cDNA contiguities deduced from the cosmid clones, RT-PCR was performed using 5' primers that would amplify transcripts initiating at upstream promoters located between 140 or 230 bp of the normal transcriptional start site (data not shown). When performed on RNA from the tail (using primers DAP1 and NUE5), the promoter for mrp4 was confirmed to be plf149. Similarly for 3T3 cells, which express predominantly plf1, RT-PCR using M3V55 and UE5 produced upstream transcripts that corresponded to the plf42 promoter. As expected, the placental mrp3 cDNA was found to be linked to the mrp3 promoter when RT-PCR was performed using DAP1 and NUE5.
Localization of MRP4 in the hair follicles of the mouse tail
To determine the location of the MRP4 protein in the adult tail,
immunohistochemistry was performed using a rabbit polyclonal
anti-MRP/PLF serum. Binding of anti-MRP/PLF antibody was detected in
the hair follicles of the tail skin (Fig. 6). The location of MRP/PLF
in the hair follicles appeared to be limited to the keratinocytes that
make up the outer root sheath. Preimmune sera did not stain these
structures.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MRP4 is more different in sequence from the other three MRP/PLFs than are the other three MRP/PLFs from each other. Whereas MRP4 is 91% identical with PLF1, the other three MRP/PLFs (PLF1, PLF2 and MRP3) are 99% identical with each other. The MRP4 protein also is differently glycosylated because two differences in the amino acid sequence result in the loss of two potential N-glycosylation sites from MRP4 compared with the other MRP/PLFs. This difference in glycosylation might alter the stability of the protein, its interaction with receptors or its ability to move from one tissue to another. For example, we have demonstrated that the 27-kDa form of MRP/PLF in the placenta that we identified as MRP4 is not found in the bloodstream or the amniotic fluid of the pregnant mouse as are other MRP/PLFs (14).
Mrp4 expression peaks on day 12 of gestation. This is at a time during which expression of the other two mrp/plfs is decreasing (6, 14, 31). Interestingly, it is the same time at which proliferin-related protein (PRP) peaks (32). PRP is closely related in sequence to the mrp/plfs with 37% identity to PLF1 in amino acid sequence and 95% identity to PLF1 in the first 97 nucleotides (6). PRP opposes the effect of bFGF on angiogenesis, and the delayed increase in PRP expression in the placenta is proposed to effect a rapid inhibition of the angiogenic effects of the MRP/PLFs (16). Similar to MRP4, PRP is not found in the amniotic fluid, but unlike MRP4, PRP is found in maternal serum (32).
Individual members of the mrp/plf family are differentially regulated in vivo in both tissue-specific and temporal patterns. As well as being expressed in the placenta, the mrp/plfs are expressed in the small intestine, tail, and ear and sometimes were observed in the stomach. Although not detected in most samples of backskin, occasionally samples showed low levels of mrp/plf expression. This expression is mainly of mrp3 (Fassett, J. T., and M. Nilsen-Hamilton, manuscript in preparation). The mrp/plfs were not detected in the liver, lungs, ribs, and large intestines. Different mrp/plf genes are expressed in different tissues. Plf1 is the major form of mrp/plf mRNA expressed in the stomach, small intestine and 3T3 cells, while mrp3 is the major form expressed by the midgestational placenta. Mrp4 is the only form of mrp/plf expressed in the ear and tail skin. As we have shown a broader tissue expression pattern of the mrp/plfs beyond the placenta, which is their site of highest expression, so are other genes of the PRL/GH family found to be expressed in high levels in one primary tissue and secondarily in other tissues. Beyond its primary site of expression in the pituitary, PRL, the evolutionary precursor to the rodent utero/placental gene family, also is expressed in other tissues, including the dermis (33). The placental lactogens I and II are expressed in the testes in addition to their high levels of expression in the placenta (34).
By immunohistochemistry, the MRP4 protein is localized in the keratinocytes of the outer root sheath of the hair follicle in the tail. These cells may not be the site of the protein’s initial synthesis, but they are apparently the site of highest accumulation of MRP4. Thus, it is likely that the keratinocytes are either the source of MRP4 or the site of its action or perhaps both.
The observation that the mrp/plfs are expressed in a tissue-specific manner suggests that these genes are regulated by distinct factors. As well, the site of expression of MRP4 is specific to the location in the body, e.g. tail vs. back. This observation parallels those of the keratin gene family where the keratin pair of 48 kDa and 65 kDa is expressed specifically in suprabasal cells of the tail epidermis, in the filiform papillae of the tongue, and at low levels in suprabulbar cells of newborn hair follicles (35). In mice, a 70-kDa keratin (mk2e) is expressed specifically in the ear, tail, and footsole epidermis (36). The ear and tail are also the specific sites of mrp4 gene expression.
Here we have shown that the promoter for mrp4 is the previously cloned plf149 promoter and the plf42 promoter is the sequence upstream of the plf1 coding sequence (37). Although 98% homologous to the promoter for plf1 (plf42), the mrp4 (plf149) promoter is not stimulated by the addition of serum (37) or bFGF (38). Interestingly, there are a number of differences between the plf149 and mrp3 and plf42 promoters that affect recognizable regulatory elements. One of these changes affects the FGF response element (FRE) present in mrp3, which is disrupted in the other two promoters (38). This and other alterations in regulatory sequences are likely to be responsible for the differential expression of the mrp/plfs in mouse tissues.
In summary, we have identified a fourth mrp/plf gene with a distinct tissue expression pattern from the other mrp/plf genes. Despite the high identity in nucleotide sequence, it is striking that a large number of the differences in nucleotide sequence of the first 550 bp of these promoters result in alterations of recognizable regulatory element sequences. These observations suggest that the mrp/plf family may have evolved to take advantage of specific combinations of transcriptional regulatory elements for the regulation of mrp/plf gene expression in particular cell types or under special physiological conditions. The resulting MRP/PLF proteins, which are also 91% to 98% identical in amino acid sequence, may not differ in their function. However, there may be other differences. For example, the difference in glycosylation of MRP4 compared with MRP3 and PLF1 could result in altered stability or tissue access for MRP4, or MRP4 may interact with a different set of receptors. Thus, we suggest that MRP4 may function as a growth factor and/or angiogenesis factor in the vicinity of the hair follicle in the tail and ear as MRP3 and PLF1 are proposed to do in the placenta.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received September 23, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
: structure and
expression. Endocrinology 139:5157–5163This article has been cited by other articles:
 |
|
 |
|
  D. O. Wiemers, L.-j. Shao, R. Ain, G. Dai, and M. J. Soares The Mouse Prolactin Gene Family Locus Endocrinology, January 1, 2003; 144(1): 313 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Toft, S. B. Rosenberg, G. Bergers, O. Volpert, and D. I. H. Linzer Reactivation of proliferin gene expression is associated with increased angiogenesis in a cell culture model of fibrosarcoma tumor progression PNAS, October 16, 2001; (2001) 231364798. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Fassett and M. Nilsen-Hamilton Mrp3, a Mitogen-Regulated Protein/Proliferin Gene Expressed in Wound Healing and in Hair Follicles Endocrinology, May 1, 2001; 142(5): 2129 - 2137. [Abstract] [Full Text]
|
|
|
|
|
|
D. J. Toft, S. B. Rosenberg, G. Bergers, O. Volpert, and D. I. H. Linzer Reactivation of proliferin gene expression is associated with increased angiogenesis in a cell culture model of fibrosarcoma tumor progression PNAS, November 6, 2001; 98(23): 13055 - 13059. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
