| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, August 10, 1994; and in revised form, October 4,
1994) From the
Human C-reactive protein (hCRP) is a major acute-phase reactant
in man. The regulation of the hCRP gene in transgenic mice is similar
to that in humans. To map DNA regions required for the correct
regulation of the hCRP gene, several constructs have been generated,
and their expression in transgenic mice has been analyzed. Constructs
lacking DNA regions surrounding the poly(A) site of the gene are not
expressed either before or after induction in transgenic mice. Minimal
regions 540 base pairs upstream and 1.2 kilobases downstream of the
hCRP gene are sufficient for liver-specific expression. Extended 5`-
and 3`-flanking regions are required to silence the expression prior to
induction. Our findings demonstrate that regulatory sequences shown to
confer inducible expression of the hCRP gene in hepatoma cell lines are
insufficient in transgenic mice. The acute phase response is characterized by dramatic
alterations in the expression pattern of several liver proteins, hCRP ( In the present study, we have
attempted to delineate the DNA flanking regions of the hCRP gene that
are responsible for its in vivo regulation.
Figure 1:
Organization
of the human CRP gene. Restriction enzyme sites are positioned relative
to the cap site of the hCRP gene. The coding part of the two exons (I,
II), the 3`-untranslated region (3` UTR), and the CRP
pseudogene (
Construct 101 was generated by isolation of an AccI fragment containing 541 bp 5` of the cap and 1.2 kb 3` of
the poly(A) corresponding to position -0.541 to 3.5 kb in Fig. 1. This construct and the following construct 4 contained
the minimal elements for the in vitro regulation of the hCRP
gene (Arcone et al., 1988). Construct 4 contained 1.7 kb 5`
of the cap site and 1.6 kb 3` of the poly(A) site, corresponding to
-1.745 to +3.91 kb (Fig. 1). This construct was
microinjected alone and also co-microinjected with the matrix
attachment region (McKnight et al., 1992). Construct 97 was
generated by subcloning the BamHI-EcoRI fragment
(position -1.7 to +1.8 kb) with the polyadenylation signal
of SV40. This construct lacked the polyadenylation site and the 3`
surrounding sequences of construct 4. Construct 61 contained 12.7 kb
5` of the cap and 1.6 kb 3` of the poly(A) site, from position
-12.7 to +3.91 kb, and it represented construct 4 with
extended 5`-flanking sequences. Construct 47 contained 1.7 kb 5` of
the cap and 8.2 kb 3` of the poly(A) site, from position -1.745
to +10.5 kb (Fig. 1). In this construct and the following
ones various lengths of 3`-flanking sequences were added to construct
4. Construct 57 contained 1.7 kb 5` of the cap and 5.2 kb 3` of the
poly(A) site, corresponding to -1.745 to +7.51 kb. Construct 79 contained 1.7 kb of 5` and 3.8 kb of 3` sequences. The
construct was prepared by combining the BamHI-HindIII
fragment at -1.7 kb to +3.14 kb and KpnI-BamHI fragment from +7.5 kb to +10.5
kb. All constructs are illustrated in Fig. 2.
Figure 2:
Schematic presentation of the hCRP
constructs and their expression in transgenic mice. The expression data
are summarized adjacent to each construct. #, construct number; a, mean value of basal level of hCRP protein expression in
µg/ml of serum and standard deviation; b, mean value of
factor of induction upon LPS injection; c, number of
independent integration sites; d, expression not
detectable.
Total RNA was isolated from mouse organs as described previously
(Chomczynski and Sacchi, 1987). 20 µg of total RNA were separated
on a formaldehyde gel, blotted, and probed with DNA fragments specific
for the hCRP gene, mouse serum amyloid A2, and
glyceraldehyde-3-phosphate dehydrogenase. Data from mice expressing
mouse SAA in liver in the preinduction state were not considered for
interpretation as expression of this endogenous murine gene indicates
that the animal is inflamed, and therefore the preinduction level of
expression of the transgene could be unreliably high.
Glyceraldehyde-3-phosphate dehydrogenase was used to control the amount
of RNA loaded. To define the location of regulatory regions required for the
expression of the hCRP gene in transgenic mice, we have undertaken a
detailed analysis of the 31-kb genomic hCRP fragment (Ciliberto et
al., 1987a). For most constructs, at least 10 independent
integration sites were generated to analyze the inherent variation in
such an in vivo approach.
Figure 3:
hCRP mRNA expression pattern. Northern
blot analyses for four different constructs are presented. The
construct number is indicated above the corresponding panel. The numbers 1-5 indicate animals of
independent integration sites. The analysis of liver RNA isolated prior
to(-) and after induction (+) is shown. The filters were
sequentially hybridized with probes for hCRP, SAA2, and
glyceraldehyde-3-phosphate dehydrogenase. Animal 5 of construct no. 101
is the only CRP nonexpressing line as documented by the signal of
glyceraldehyde-3-phosphate dehydrogenase
probe.
We next analyzed
construct 101, which is only 4 kb in size (Fig. 2). Though this
construct still contained all the known in vitro defined acute
phase response elements, its regulation differed significantly from
that of construct 42. The transgenic mice generated showed a high basal
level of hCRP expression, which was induced 5-fold by LPS (Fig. 2). This finding was also reflected on the RNA level, five
examples of which are shown in Fig. 3. The high basal level of
CRP in serum of transgenic mice generated with construct 101 was not a
result of inflammation because expression of SAA2, a gene known to be
induced in mice in the inflammatory state, was absent prior to LPS
induction (Fig. 3). The result suggests that construct 101
lacked regulatory regions required to silence the transgene expression
prior to induction.
Figure 4:
Tissue-specific mRNA expression pattern of
hCRP. A Northern blot analysis of a mouse harboring construct 4 is
shown. The expression of hCRP in several organs is analyzed upon
induction with LPS. Expression in liver is also shown before (L) and after induction (L*). P, pancreas; L, liver; H, heart; Lg, lungs; O,
ovary, uterus; K, kidney; M, skeletal muscle. The upper panel (CRP) comes from a short exposure while
the lower panel (CRP*) represents a longer exposure
to ensure the absence of expression in other
organs.
To exclude the
possibility that the high basal level of expression of constructs 101
and 4 was due to the interference of surrounding sequences, construct 4
was co-injected with the matrix attachment region sequences from the
chicken lysozyme gene. The matrix attachment region sequences have the
property of allowing integration site-independent expression of
transgenes (McKnight et al., 1992). Three transgenic mice were
analyzed. all of which showed an expression pattern prior to and after
induction similar to that of construct 4 alone (data not shown).
The influence of 3`-flanking regions was surprising and therefore
further investigated by analyzing construct 57, which encompassed a
shorter length of 3`-flanking sequences compared with construct 47 (Fig. 2). Transgenic mice harboring this construct showed high
levels of CRP expression prior to induction, resembling the regulation
of constructs 101 and 4 ( Fig. 2and Fig. 3). This result
indicated that regions 5 kb downstream of the hCRP coding sequence (3`
of the +7.51 KpnI site (Fig. 1) are important for
inducible expression. These 3`-regions include the hCRP pseudogene
(Ciliberto et al., 1987b). We have therefore generated
construct 79, which contained the sequences deleted in construct 57
attached to the sequences of construct 4 (Fig. 2). This
construct exhibited a low basal expression level followed by strong
induction after LPS stimulation suggesting that the +7.5 to
+10.5 flanking region of the hCRP gene is important for negative
control of expression. In this study we have investigated the regions of the hCRP
gene that are necessary for liver-specific and inducible expression.
Constructs harboring various deletions 5` and 3` of the hCRP gene have
been introduced into mice and their expression analyzed on both the
protein and RNA level. The results indicate that 1) constructs lacking
sequences surrounding the poly(A) site of the gene were not expressed
either before or after induction in transgenic mice; 2) sequences 540
bp upstream and 1.2 kb downstream of the hCRP gene are sufficient for
liver-specific but constitutive expression; 3) DNA flanking regions
conferring tight control of the hCRP gene prior to induction are
present both in the 5`- and 3`-flanking regions; and 4) regulatory
sequences shown to confer inducible expression of the hCRP gene in
hepatoma cell lines are insufficient in transgenic mice. Regulation
of gene expression by sequences located 3` of the open reading frame
has been well documented, and examples include the human Further work is required to define the exact sequences
surrounding the poly(A) site of the hCRP gene that are required for the
transgene expression. A potential candidate is the consensus site for
LR1, a lipopolysaccharide-responsive factor with binding sites in the
immunoglobulin switch regions and also in the heavy-chain enhancer
(Williams and Maizels, 1991). The consensus site is GNCNAGGCTGA(A/C),
and that found in the human CRP gene is GACAAGGCTGAT. In the rabbit
where CRP is also an acute phase gene the site is also conserved
(TCCAAGGCTGAC). It is not known whether LR1 is required for the
expression of acute phase response genes; so far it has only been found
in cultured primary B and murine liver cells (Williams and Maizels,
1991). The presence of a potential binding site in the hCRP gene, in a
DNA fragment shown here to be essential for liver-specific transgene
expression, merits further investigation. The low level of
expression of the hCRP gene before induction is most likely due to
negative transcriptional regulation as has been shown for some
inducible genes, e.g. interferon. The promoter became
constitutively active when certain elements were deleted (for review,
see Clark and Docherty(1993)). Similar results have been found for the
hCRP promoter in human hepatoma cells (Li et al., 1990). Two
negative control elements were detected in the promoter of the gene.
Removal of these elements resulted in an increase in the basal level of
gene expression. The negative control regions described (Li et
al., 1990) are included in all the constructs used in this report
and may indicate general differences between hepatoma cell lines and
the transgenic mouse system. Our results suggest that 5`- and
3`-regions of the hCRP gene can independently contribute to the low
level of expression prior to induction (constructs 61 and 47,
respectively). As we have not yet mapped the DNA elements responsible,
we cannot exclude that identical sequences may be present 5` and 3` of
the gene. Interestingly, the 3`-region involved in the negative
regulation of the hCRP gene includes the hCRP pseudogene. It is
possible that the pseudogene is a product of duplication, and certain
regulatory sequences have been retained. To characterize the elements
required for silencing the hCRP gene prior to induction, we will screen
for sequences that when deleted increase the preinduction level of
expression. This approach will finally lead to a fine mapping of the
elements involved in the regulation of the hCRP gene. Identification of
such regulatory elements will advance the understanding of the network
of cytokine action in the course of inflammation for which the hCRP is
a major marker protein.
Volume 270,
Number 2,
Issue of January 13, 1995 pp. 704-708
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)being the most abundant in humans (see Pepys and
Baltz(1983)). The level of expression of hCRP increases from a basal
serum level of less than 1 µg/ml to more than 300 µg/ml during
the acute phase response. The proteins induced during the acute phase
response are for the most part species-specific. In mouse, for
instance, the expression of serum amyloid proteins (SAA) is greatly
enhanced (Pepys and Baltz, 1983) while the expression of CRP is hardly
induced (Whitehead et al., 1990). Nevertheless, the expression
of human CRP was found to be regulated in transgenic mice as well as in
man (Ciliberto et al., 1987a). A very low basal level of hCRP
is strongly increased after induction. Therefore, all the sequences
required for the correct regulation of the hCRP gene were contained
within the 31-kb genomic fragment injected. The search for a suitable
experimental system to further investigate the hCRP gene regulation led
to the use of human hepatoma cell lines. However, even though these
cells are known to lose liver-specific properties (discussed in
Baumann(1989)), they were successfully used to demonstrate that both
interleukin 1 and 6 acute phase response elements (Arcone et
al., 1988; Ganter et al., 1989; Majello et al.,
1990) and binding sites for hepatocyte nuclear factor 1 (Toniatti et al., 1990a) were present in the hCRP promoter. Although
these in vitro studies were a first step toward the
characterization of the hCRP promoter, it was not known whether the
identified elements were sufficient for the correct in vivo regulation of the hCRP gene.
Generation of Constructs
Construct 42 contained
17 kb 5` of the cap site and 11.3 kb 3` of the poly(A) sequence
(corresponding to -17 kb to +13.6 kb in Fig. 1). The
31-kb ClaI fragment was isolated for microinjection from a
previously described human cosmid, pCOS-CRP1 (Ciliberto et
al., 1987a).
) are indicated.
Generation and Identification of Transgenic
Mice
Transgenic mice were generated according to standard
procedures (Hogan et al., 1986) by microinjection into the
fertilized eggs of a cross between F1 (C57BL/6 
SJL) animals.
Tail DNA was isolated at 14 days after birth and analyzed by Southern
blotting.LPS Induction of the Transgenic Mice and Estimation of
hCRP Concentration in Serum
The level of hCRP protein expression
in the serum was analyzed in all mice before and 15 h after induction
with an intraperitoneal injection of 100 µg of LPS (Sigma, L-3254).
The animals were analyzed at 4 weeks after birth to avoid fluctuations
in the gene expression that may occur at different ages. The 15-h time
point was chosen as it has been previously described (Ciliberto et
al., 1987a) that the peak hCRP expression occurs after 15 h in
transgenic mice. The serum samples were stored at -20 °C
until analyzed. hCRP concentration was estimated in the serum by an
enzyme-linked immunosorbent assay method using rabbit anti-hCRP
immunoglobulin fraction (Dako Corp., AO73) and conjugated anti-rabbit
immunoglobulin fraction peroxidase (Dako, P227). (
)hCRP
standard X923 was obtained from Dako Corp. The lower limit of detection
of hCRP in mouse serum ranged from 500 pg to 1 ng/ml.Isolation and Northern Blotting of Total RNA
At 8
weeks postpartum the mice were subjected to partial hepatectomies
before and 9 h after induction with an intraperitoneal injection of
LPS. Liver samples were stored at -80 °C until processed. The
9-h time point was chosen as it has been previously shown that the peak
of hCRP mRNA expression in transgenic mice occurs at this time
(Ciliberto et al., 1987a). To determine whether the transgenes
were liver-specific in their expression pattern, total RNA was prepared
from spleen, liver, heart, thymus, muscle, bone, kidney, pancreas,
gonads, and brain of two F
mice harboring each construct.
The Minimal Elements Required for Inducible Expression of the
hCRP Gene in Hepatoma Cells Are Insufficient in Transgenic
Mice
The expression profile of construct 42, encompassing the
entire 31-kb human genomic CRP fragment (Fig. 1), showed a low
constitutive level of CRP expression that was induced 30-fold after LPS
induction (Fig. 2). The protein values in serum were reflected
in the RNA expression patterns. Five random examples are shown in Fig. 3. Furthermore, the hCRP mRNA was restricted to liver, and
no correlation was found between the expression levels and the
transgene copy number (data not shown). All these findings were in
agreement with previous results (Ciliberto et al., 1987a) and
represented the starting point for our comparisons.
Effect of DNase-hypersensitive Sites on the Inducible
Expression of the hCRP Gene
Three LPS-inducible liver-specific
DNase I-hypersensitive sites were found in transgenic mice harboring
construct 42 (Toniatti et al., 1990b). Two mapped to the
5`-region adjacent to the cap site and at approximately -250 bp.
The third site mapped to approximately -600 bp. Interestingly,
this third DNase I-hypersensitive site mapped to a region that is
conserved in the human haptoglobin and 
-antitrypsin
genes (Bensi et al., 1985; De Simone et al., 1987).
The sequence, TGGACACAGG, is 100% identical at position -671 of
the hCRP gene. The regulation in mice of construct 4, which, compared
with construct 101, contained extra 5`- and 3`-flanking DNA sequences
including the latter motif, did not differ substantially from that of
construct 101 except that it exhibited an even higher basal level of
expression prior to induction (Fig. 2). Therefore, this DNase
I-hypersensitive site is not involved in the silencing of the gene
prior to induction. Since expression of constructs 4 (Fig. 4)
and 101 (data not shown) was exclusively found in the liver, tissue
specificity and silencing of hCRP expression prior to induction seem to
be controlled by different regulatory regions.
Sequences Surrounding the Poly(A) Site of the hCRP Gene
Are Necessary for Its Expression
In the uninduced liver of
transgenic mice harboring construct 42, a strong DNase I-hypersensitive
site was identified in the vicinity of the poly(A) site of the hCRP
gene (Toniatti et al., 1990b). To investigate the requirement
of this region for the proper regulation of hCRP, we generated
construct 97, which contained the poly(A) signal of SV40 (Fig. 2). Five transgenic mice were analyzed, none of which
expressed the transgene in any organ either before or after induction.
Thus, the sequences in the vicinity of the poly(A) site are necessary
for hCRP promoter activity.Investigation of Extended 5`- and 3`-Flanking
Regions
Since none of the short constructs (e.g. no. 4)
analyzed so far revealed a low basal level of expression that was
comparable with that of construct 42, we determined the effects of
increased 5`- or 3`-flanking regions. To this end constructs 61 and 47
with either additional 5`- or 3`-flanking regions, respectively, were
generated (Fig. 2). Both constructs exhibited reasonably low
basal serum level, which was strongly induced after LPS stimulation, a
result that was also confirmed by RNA expression analysis (Fig. 3). Thus, increasing either the 5`- or 3`-flanking DNA
clearly decreased the basal level of expression prior to induction
indicating the presence of regulatory regions in these sequences.
-globin,
CD2, and keratin 18 genes (Grosveld et al., 1987; Greaves et al., 1990; Neznanov et al., 1993). These genes are
constitutively expressed in a copy number-dependent and integration
site-independent manner in transgenic mice only when 3` elements are
included. Similarly, in the case of the hCRP gene we have not observed
any expression of the transgene in the absence of the sequences
surrounding the poly(A) site. This was also the case when we attempted
to express the SV40 T antigen under the control of the hCRP promoter.
It was necessary to include the region surrounding the poly(A) signal
for transgene expression (Rüther et al.,
1993). In contrast to the copy number-dependent expression observed
with the -globin, CD2, and keratin 18 genes, we have not found any
indication for copy number-dependent expression in the case of the hCRP
gene.
))
We thank David Stoddart and Barbara Stace for
excellent animal husbandry and Rocio Garcia de Veas Lovillo and Birgit
Bosse for technical assistance. We thank Lothar Hennighausen for the
matrix attachment region construct, and Alfred Nordheim and Michael
Cahill for critical reading of the manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Xing, F. G. Hage, Y.-F. Chen, M. A. McCrory, W. Feng, G. A. Skibinski, E. Majid-Hassan, S. Oparil, and A. J. Szalai Exaggerated Neointima Formation in Human C-Reactive Protein Transgenic Mice Is IgG Fc Receptor Type I (Fc{gamma}RI)-Dependent Am. J. Pathol., January 1, 2008; 172(1): 22 - 30. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kovacs, P. Tornvall, R. Nilsson, J. Tegner, A. Hamsten, and J. Bjorkegren Human C-reactive protein slows atherosclerosis development in a mouse model with human-like hypercholesterolemia PNAS, August 21, 2007; 104(34): 13768 - 13773. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wang, S. Oparil, Y.-F. Chen, M. A. McCrory, G. A. Skibinski, W. Feng, and A. J. Szalai Estrogen Treatment Abrogates Neointima Formation in Human C-Reactive Protein Transgenic Mice Arterioscler. Thromb. Vasc. Biol., October 1, 2005; 25(10): 2094 - 2099. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. P. Grocott, W. D. White, R. W. Morris, M. V. Podgoreanu, J. P. Mathew, D. M. Nielsen, D. A. Schwinn, M. F. Newman, and for the Perioperative Genetics and Safety Outcomes Genetic Polymorphisms and the Risk of Stroke After Cardiac Surgery Stroke, September 1, 2005; 36(9): 1854 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Burke, A. Pridmore, N. Harraghy, A. Collick, J. Brown, and T. Mitchell Transgenic Mice Showing Inflammation-Inducible Overexpression of Granulocyte Macrophage Colony-Stimulating Factor Clin. Vaccine Immunol., May 1, 2004; 11(3): 588 - 598. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Russell, D. S. Cunninghame Graham, C. Shepherd, C. A. Roberton, J. Whittaker, J. Meeks, R. J. Powell, D. A. Isenberg, M. J. Walport, and T. J. Vyse Polymorphism at the C-reactive protein locus influences gene expression and predisposes to systemic lupus erythematosus Hum. Mol. Genet., January 1, 2004; 13(1): 137 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Wiekowski, M. W. Leach, E. W. Evans, L. Sullivan, S.-C. Chen, G. Vassileva, J. F. Bazan, D. M. Gorman, R. A. Kastelein, S. Narula, et al. Ubiquitous Transgenic Expression of the IL-23 Subunit p19 Induces Multiorgan Inflammation, Runting, Infertility, and Premature Death J. Immunol., June 15, 2001; 166(12): 7563 - 7570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kanzler, A. W. Lohse, A. Keil, J. Henninger, H. P. Dienes, P. Schirmacher, S. Rose-John, K. H. Meyer Zum Buschenfelde, and M. Blessing TGF-beta 1 in liver fibrosis: an inducible transgenic mouse model to study liver fibrogenesis Am J Physiol Gastrointest Liver Physiol, April 1, 1999; 276(4): G1059 - G1068. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Klein, T. Klein, U. Ruther, and B. Kyewski CD4 T Cell Tolerance to Human C-reactive Protein, an Inducible Serum Protein, Is Mediated by Medullary Thymic Epithelium J. Exp. Med., July 1, 1998; 188(1): 5 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Szalai, F. W. van Ginkel, S. A. Dalrymple, R. Murray, J. R. McGhee, and J. E. Volanakis Testosterone and IL-6 Requirements for Human C-Reactive Protein Gene Expression in Transgenic Mice J. Immunol., June 1, 1998; 160(11): 5294 - 5299. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Basile, A. Sica, E. d'Aniello, F. Breviario, G. Garrido, M. Castellano, A. Mantovani, and M. Introna Characterization of the Promoter for the Human Long Pentraxin PTX3. ROLE OF NF-kappa B IN TUMOR NECROSIS FACTOR-alpha AND INTERLEUKIN-1beta REGULATION J. Biol. Chem., March 28, 1997; 272(13): 8172 - 8178. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Zhang, M. Sun, D. Samols, and I. Kushner STAT3 Participates in Transcriptional Activation of the C-reactive Protein Gene by Interleukin-6 J. Biol. Chem., April 19, 1996; 271(16): 9503 - 9509. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
