Molecular Human Reproduction, Vol. 9, No. 6, 351-358,
June 2003
© 2003 European Society of Human Reproduction and Embryology
Article |
Gestational profile of matrix metalloproteinases in rat uterine artery
Submitted on December 5, 2002; accepted on February 25, 2003
1 Maternal and Fetal Research Unit, Department of Womens Health, Guy’s, King’s and St Thomas’ School of Medicine, 10th Floor St Thomas’ Hospital, Lambeth Palace Road, London SE1 7EH and 2 Department of Statistical Science, Glaxo SmithKline Pharmaceuticals, Harlow, Essex CM19 5AW, UK
3 To whom correspondence should be addressed. e-mail: brenda.a.kelly{at}kcl.ac.uk
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Mechanisms underlying structural reorganization of the uterine artery in pregnancy remain largely unknown. Matrix metalloproteinases (MMPs) which are involved in degradation of vascular wall matrix are likely to play a key role. In this investigation of rat uterine artery, key MMPs and the specific tissue inhibitors of MMPs (TIMPs) together with three housekeeping genes were studied before, during and after pregnancy, using real time PCR. Data were analysed by partial least squares analysis as well as by conventional univariate methods. Each gene studied [MMP-2, MMP-3, MMP-7, MMP-9, MMP-12, MMP-13, membrane-type 1 (MT1)-MMP, TIMP-1, TIMP-2, GAPDH, cyclophilin and ß-actin] increased in late pregnancy (day 21). MMP-2, MT1MMP, MMP-3 and TIMP-1 transcripts were also elevated at day 7. TIMP-1 and MMP-3 mRNA expression returned to virgin control values in the post-partum, whereas others remained elevated or increased further (MMP-9, MMP-13). Gelatin zymography showed maximum elevation of MMP-2 at day 21. A novel 43–45 kDa gelatinolytic doublet was observed which increased in density with gestation and may represent an active MMP-2 fragment. Together, these data strongly suggest that MMPs and TIMPs are likely to play an important role in remodelling uterine arteries in rat pregnancy and may represent means by which vasodilatation is maintained in later pregnancy. Continued elevated levels of some MMPs post-partum may contribute to vessel regression and return to a non-pregnant physiological state.
Key words: housekeeping genes/matrix metalloproteinase/pregnancy/remodelling/uterine artery
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
In normal pregnancy the maternal cardiovascular system undergoes functional and structural changes to accommodate the nutritional demands of the growing conceptus (Kelly et al., 1999). Nowhere is this more profound than in the uteroplacental vasculature. The marked increase in uterine blood flow during normal pregnancy is achieved by dramatic reduction in vascular resistance, facilitated by enhanced synthesis of endothelium-dependent vasodilators (Nelson et al., 2000; Magness et al., 2001) and reduced constrictor sensitivity (Weiner et al., 1992; Jovanovic et al., 2000). The fall in resistance is also aided by structural reorganization extending from the uterine and radial arteries (Palmer et al., 1992) to the spiral arteries of the myometrium and decidua (Pijnenborg et al., 1983).
The uterine and radial arteries in animal and human pregnancy become larger in caliber, longer and more distensible in both axial and radial directions (Moll et al., 1983; Moll and Gotz, 1985; Guenther et al., 1988; Nienartowicz et al., 1989; Osol and Cipolla, 1993; Cipolla and Osol, 1994). However, the cellular and molecular mechanisms underlying gestational growth and remodelling of these arteries remain largely unresolved. Structural modification is likely to occur through processes of expansive remodelling and non-branching angiogenesis as indicated by the increase in uterine artery diameter, which occurs in the absence of significant alteration in media thickness, and the increase in arterial length in association with endothelial and vascular smooth muscle cell proliferation (Nienartowicz et al., 1989; Cipolla and Osol, 1994). Degradation of extracellular matrix scaffold enables cell movement and tissue reorganization and is regulated by a family of enzymes, the matrix metalloproteinases (MMPs). These enzymes are therefore prime candidates for involvement in vascular remodelling. MMPs are regulated at several levels. While the expression of most of these enzymes is transcriptionally regulated, additional modulation exists at the level of activation of the secreted latent proenzyme. The activity of the resultant protease can be inhibited by several endogenous inhibitors, of which tissue inhibitors of metalloproteinases (TIMPs) appear most important. Many studies have addressed and endorsed a role for MMPs in pathological vascular remodelling. Both MMP-2 and -9 have been implicated in early compensatory outward remodelling of atherosclerotic human coronary arteries (Pasterkamp et al., 2000) and in elastolysis observed in aneurysmal expansion (Davis et al., 1998; Crowther et al., 2000). Altered expression of MMP-2 and -9, together with MMP-1, -3, -12, and -13 has been proposed to facilitate vascular smooth muscle cell (VSMC) migration and proliferation in atherosclerotic vessels (Galis and Khatri, 2002).
Although infrequently investigated, MMPs may also contribute to physiological vascular remodelling. MMP-2 and MMP-9 protein expression is increased in small endometrial arterioles during the proliferative phase of the menstrual cycle (Freitas et al., 1999). In addition, MMP-1, MMP-3 and MMP-9 protein expression is up-regulated in trophoblast invading distal spiral arteries in myometrium in early pregnancy (Blankenship and Enders, 1997).
In this study we have investigated the longitudinal profiles of a number of MMPs and TIMPs previously implicated in angiogenesis and vascular remodelling in the uterine artery in normal pregnant rats. Using a combination of real time PCR and gelatin zymography, we have demonstrated that pregnancy is associated with altered expression of specific MMPs and TIMPs in this artery.
|
Materials and methods |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Animals
All animal procedures were carried out in accordance with the UK Animals (Scientific Procedures) Act 1986 and with institutional ethical approval. Female Sprague–Dawley rats (225–250 g; B&K Universal Ltd, UK) were housed under controlled conditions of temperature and humidity with a 12 h light/dark cycle and free access to tap water and rat chow (Breeding Diet no. 3; Special Diet Services, UK). For mating, an adult male and a virgin female rat were housed in a single cage. Successful mating was confirmed by the presence of a vaginal plug and designated day 0. Delivery usually occurred on day 22. On days 7, 14 and 21 of gestation and at day 7 post-partum, rats were culled by cervical dislocation. Female rats (225–250 g) were killed in the same manner on the estrus day of estrus cycle (as determined by daily vaginal smears) and served as virgin controls. Uterine arteries were rapidly dissected under a light microscope, snap-frozen in liquid nitrogen and stored at –80°C.
RNA extraction and RT
All reagents for RNA extraction, DNAase treatment and RT were purchased from Invitrogen, UK. Uterine arteries from each animal (n = 8 animals per group) were homogenized in 800 µl Trizol with 200 µg glycogen added and total RNA extracted according to the manufacturer’s instructions. The resultant RNA pellet was resuspended in PCR grade water and the concentration calculated in triplicate by A260 measurement. DNAase-treated RNA samples (1 µg) were then reverse-transcribed. Briefly, first strand cDNA was synthesised from 1 µg of each RNA sample using 0.01 mol/l dithiothreitol, 0.5 mmol/l each dNTP, 0.5 µg oligo(dT) primer, 40 IU RNaseOUT ribonuclease inhibitor and 200 IU Superscript II reverse transcriptase (final volume 20 µl). Triplicate RT were performed together with an additional reaction in which the reverse transcriptase enzyme was omitted to allow for assessment of genomic DNA contamination in each sample. cDNA synthesized was diluted 5-fold (equivalent to 10 ng starting RNA/µl), aliquoted and stored at –20°C.
Relative quantification of MMP and TIMP gene expression using real time PCR
TaqMan real time PCR assays for each gene target were performed on triplicate cDNA samples in 96-well optical plates using an ABI Prism 7700 Sequence Detection system (Applied Biosystems, USA). For real time PCR assays, all reagents were obtained from Applied Biosystems except primers (Sigma– Genosys, UK). For each 25 µl TaqMan reaction, 5 µl of cDNA was mixed with 12.5 µl of 2xTaqMan Universal Master Mix, 0.75 µl of each forward and reverse primer (10 µmol/l; final concentration 300 nmol/l), 1 µl TaqMan fluorogenic probe (5 mmol/l; final concentration 250 nmol/l) and 5 µl of molecular biology grade water. Forty PCR cycles were performed under standard thermal cycle conditions with an annealing temperature of 60°C. TaqMan primer and probes (listed in Table I) were designed from sequences in the GenBank database using Primer Express software (Applied Biosystems). All primer sets were tested under TaqMan PCR conditions using rat genomic DNA as a template. In all cases a single product of the appropriate size was detected by gel electrophoresis (data not shown). To enable quantification, additional reactions were performed on each 96-well plate using dilutions of known amounts of highly sheared rat genomic DNA to allow the construction of a standard curve relating threshold cycle to template copy number as described elsewhere (Harrison et al., 2000; Macdonald et al., 2001). Test gene values were then extrapolated from the standard curve and expressed in arbitrary units. Preliminary analyses showed that genomic DNA used for the standards amplified with similar efficiency to that of positive control cDNA template (data not shown).
|
Protein extraction
All chemicals were supplied by Sigma (UK) unless otherwise stated. Uterine arteries from each animal were pulverized under liquid nitrogen. Samples were vortexed briefly in 10 µl/mg tissue of phosphate-buffered saline (pH 7.4), 0.1% sodium dodecyl sulphate (SDS), 0.5% Triton X-100, 0.5% sodium deoxycholate, 0.02% sodium azide and EDTA-free Protease Inhibitor Cocktail V (Calbiochem, UK) and incubated on ice (10 min), then centrifuged (4°C, 20 min, 12 000 g). The supernatant was decanted and designated the ‘detergent fraction’. The remaining pellet was further extracted with a urea-based buffer (10 µl/mg starting tissue) consisting of TNC buffer [50 mmol/l Tris–HCl (pH 7.5), 150 mmol/l NaCl, 10 mmol/l CaCl2, 0.05% (w/v) Brij35, 0.02% sodium azide] containing 8 mol/l urea and EDTA-free Protease Inhibitor Cocktail V. Samples were agitated for 15 min at 4°C then centrifuged (12°C, 20 min, 12 000 g). The supernatant was decanted and designated the ‘urea fraction’. Preliminary analyses showed that further extraction of the remaining pellet did not yield any additional gelatinolytic activity and indicated that purified proMMP-2 and proMMP-9 standards processed by this protocol did not undergo significant activation (data not shown). Total protein concentration for each fraction was determined in triplicate using the DC protein assay (BioRad, UK) performed according to the manufacturer’s instructions.
Gelatin zymography
For each fraction, total protein (15 µg) from control, days 7, 14 and 21 pregnancy and day 7 post-partum (n = 4 or 5 per group) were mixed with sample loading buffer (62.5 mmol/l Tris–HCl pH 6.8, 2.5% SDS, 1% sucrose, 0.001% Bromophenol Blue) then resolved by electrophoresis in 10% SDS–polyacrylamide electrophoretic gels containing 1 mg/ml gelatin alongside purified MMP-2 (Calbiochem, UK) and MMP-2/-9 (kindly donated by Dylan Edwards, University of East Anglia, UK) recombinant protein standards. After electrophoresis, proteins in the gel were renatured by incubation in 2.5% Triton X-100 (2x30 min). Gels were subsequently incubated overnight at 37°C in TNC buffer and bands of lytic activity visualized as zones of clearing in Coomassie Blue-stained gels. To verify MMP activity, control gels were incubated under the same conditions in the presence of 25 mmol/l EDTA (an inhibitor of MMP), 2 mmol/l PMSF (a serine protease inhibitor) or 20 µmol/l E64 (a cysteine protease inhibitor). To further characterize lytic bands observed on zymography, purified recombinant human proMMP-2 standards were incubated in the presence of aminophenylmercuric acetate (2 h, 37°C; final concentration 1.5 mmol/l). Activated standards were then mixed with sample loading buffer and subjected to gelatin zymography as above. Gel images were captured by FluorS MultiImager (BioRad), inverted and the mean optical density product of individual lytic bands calculated (MultiAnalyst, BioRad).
Data analysis
For real time PCR analysis of gene expression, mean values were calculated for each set of reverse-transcribed RNA triplicates from each animal studied (n = 8 per time-point). Data were log-transformed (base 10) to meet the requirements of normality and homogeneity of variation and analyses of variance (ANOVA) performed using Genstat software V5.4.1 (Lawes Agricultural Trust, UK). Post-hoc pairwise comparisons between virgin control and remaining groups were performed using Dunnett’s t-test with a value of P < 0.05 considered significant. A further study of the relative changes in gene expression was performed by partial least squares (PLS), a multivariate analysis technique (Wold et al., 1984) that has been recommended in mRNA high throughput studies (Hilsenbeck et al., 1999; Hole et al., 2000; Bond et al., 2002). This method combines principal components analysis (PCA) with regression. By analysis of the expression of all the genes simultaneously, this method allows identification of relationships between sets of dependent variables (genes) with the set of independent variables (time, i.e. stage of gestation). Analysis was performed using SIMCA software V7 (Umetri AB, Sweden). Data were summarized by constructing a scatterplot of principal components loadings for the genes studied.
Densitometric data from gelatin zymography were generated through analysis of four or five uterine arteries per time-point and each assay was carried out in duplicate. Data were log transformed (base 10) to meet the requirements of normality and homogeneity of variation and ANOVA performed. Post-hoc pairwise comparisons between virgin control and remaining groups were performed using Dunnett’s t-test with a value of P < 0.05 considered significant.
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
MMP and TIMP gene expression
Figure 1A shows gestational profiles of gene expression. Data are presented as unadjusted geometric means of expression with 95% confidence intervals (CI). Temporal trends were evident, the most striking being elevation of expression of all MMP [MMP-2, -3, -7, -9, 12, -13, membrane-type 1 (MT1)-MMP] and TIMP (TIMP-1, -2) genes just prior to delivery, at day 21 of gestation (P = 0.01). In addition, early pregnancy (day 7) was associated with elevated expression of MMP-2, MT1MMP, MMP-3 and TIMP1 (P = 0.01). Day 14 was associated with elevation of TIMP1, MMP-2 and MT1MMP transcripts (P = 0.01) as well as increased TIMP-2 expression (P = 0.05). Seven days after delivery, expression of some genes reverted to pre-pregnancy levels of expression, e.g. TIMP-1, -2, MMP-3, while expression of the other MMP remained elevated or, in the case of MMP-9, further increased.
|
Also shown in Figure 1A are the results of parallel real time PCR assays run for three commonly used internal reference control genes, GAPDH, cyclophilin and ß-actin. All three were elevated in late gestation. While the magnitude of the gestational effect on expression of these control genes, e.g. GAPDH, was substantially less (1.8-fold increase over virgin control 95% CI: 1.2–2.8) than that of some MMP, e.g. MMP-12 (51.8-fold 95% CI: 10.3–260.5), adjusting test gene expression for that of the internal reference gene in this study might not only have removed any differences in mRNA quality but also pregnancy-associated effect giving rise to false negative or positive analyses.
The alteration of reference gene expression precluded normalization, a conventional practice in semi-quantitative gene expression studies. We elected to further test the hypothesis that MMP and TIMP gene expression was specifically altered in pregnancy by using the PLS method. This multivariate analysis has the added advantage of being independent of the need for a ‘control gene’. In considering the entire dataset it is possible to obtain insight into similarities in gene behaviour over the five time-points. Underlying latent variables (‘principal components’, PC) were calculated from each of the dependent (genes) and independent (gestational, virgin and post-partum time-points) variables. Components were made up of a linear combination of variables, whether as genes or gestational time-points, and were ordered by the proportion of variance they explain, the first PC (PC1) explaining the greatest proportion of the total variance. The contribution of a given variable to a component was reflected in its weighting or ‘loading’. A loading of zero indicated that that particular variable made no contribution to that component. These weights or loadings in themselves can be informative and, plotted against each other, can offer further insight into the relationship between genes. Figure 1B is a scatterplot of the principal component loadings for the 12 genes and five time-points studied. In this study, the first principal component (PC1) represented the difference between virgin animals and pregnant rats at day 21 of gestation and explained 64% of the variability in the genes. The second principal component (PC2) accounted for the next largest amount of variability in the dataset and reflected the difference between day 7/day 14 pregnancy and post-partum. This trend was not seen in all genes and explained only 16% of the variability in gene expression. Genes behaving in a similar way over the gestational time-points studied had similar loadings and appeared close to each other (see Figure 1B). The emergence of such clustering is one of the main advantages of this form of analysis reflecting qualitative similarities between profiles of different genes, which is not possible with conventional univariate analyses. In reviewing PC1, all genes including reference genes GAPDH, ß-actin and cyclophilin lie closer to the day 21 gestation than virgin time-point, suggesting that expression is higher in late gestation than before pregnancy. This broad clustering separates into more discrete groups when loadings in the second dimension/PC2 are considered with genes such as MMP-9 and MMP-12 lying closer to the post-partum time-point and furthest from day 7/day 14. In contrast, the loading for TIMP1 in PC2 is more closely related to day 7/day 14 than to the post-partum. MMP-3 lies close to the housekeeping genes GAPDH and cyclophilin while the loadings plot for MMP-2 fell close to that of TIMP-2 and MT1MMP.
Metalloproteinase activity in rat uterine artery
The expression of gelatinases was further investigated using gelatin zymography. Uterine arteries were pulverized and tissue serially extracted with a detergent-based buffer and a urea-based buffer to ensure removal of both soluble and matrix-bound MMP. Three gelatinolytic bands at 
78, 72 and 65 kDa were observed with both extraction methods and were common to all five groups (virgin, day 7, 14, 21 and day 7 post-partum) studied (Figure 2A, B). Based on size and inhibition with EDTA (data not shown), the 72 and 65 kDa bands were likely to represent latent and active MMP-2 respectively. The 78 kDa band may represent a latent form of MMP-2 or an active fragment of MMP-2 complexed with a TIMP. An additional gelatinolytic doublet, unique to the urea-extraction and inhibited by EDTA was observed at 43–45 kDa (Figure 2B). AMPA-activation of recombinant proMMP-2 yielded a cleavage product of similar molecular weight (Figure 2C). The gestational profile of this doublet was similar to that observed for the 65 kDa band suggesting that the doublet may represent active fragments of MMP-2 (Figure 3A, B). With the exception of this doublet, similar gestational profiles of latent and active enzyme in the detergent and urea fractions suggested that there was no difference in extractability of either latent or active MMP-2 between gestational stage nor between non-pregnant/pregnant vessels. Taking both fractions together, total gelatinolytic activity increased significantly with gestation (Figure 3C) with an early 2-fold elevation reaching maximum levels in late pregnancy (day 21; 6-fold 95% CI: 4.4–8.8). Notably, total activity remained elevated post-partum (4.2-fold CI: 3.0–6.0).
|
|
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
This study has demonstrated gestational modulation of MMPs and TIMPs in rat uterine artery. Maximal gene expression generally occurred in late pregnancy but specific mRNA transcripts including TIMP1 and MMP-2 were also elevated at day 7. Following delivery, both TIMP1 and MMP-3 mRNA returned to pre-pregnancy levels whereas the expression of other MMP remained elevated or even higher than in late gestation. MMP-2 gelatinolytic activity increased in pregnancy and remained elevated in the postnatal period. To our knowledge this is the first description of the longitudinal profile of MMPs and TIMPs in the uterine vasculature in pregnancy.
Temporal differences in expression of MMPs and TIMPs may indicate specific functional roles in gestation-related changes in the uterine artery. In the rat uteroplacental circulation, vascular adaptation appears to proceed in a specific temporal manner with maximum remodelling of the smaller more distal radial arteries reported to occur in mid-gestation (Cipolla and Osol, 1994) while growth and remodelling of the more proximal main uterine artery, as studied here, is greatest in late pregnancy (Nienartowicz et al., 1989; Cipolla and Osol, 1994). Post-partum, these structural modifications are rapidly reversed. Altered expression of MMPs and TIMPs in later gestation might therefore reflect a role in vascular proliferation while expression in the post-partum might signify a role in vascular resorption.
The profile of MMP-2 was most clearly characterized in this study, with enhanced expression and activity of MMP-2 occurring both during pregnancy and post-partum. This gelatinase has broad substrate specificity with the capacity to degrade elastin as well as key basement membrane proteins including collagen IV, laminin and entactin (Woessner and Nagase, 2000). Elevated vascular wall MMP-2 has previously been associated with VSMC migration (Aoyagi et al., 1998; Lijnen et al., 1999), elastin degradation (Li et al., 1999), aneurysmal formation (Crowther et al., 2000) and expansive remodelling (Pasterkamp et al., 2000). Moreover, MMP-2 may contribute to early events of angiogenesis (Fang et al., 2000).
We propose that MMP-2 plays an important role in physiological vascular growth and remodelling in pregnancy. The functional significance of increased MMP-2 activity at day 7 of gestation is difficult to assess as structural changes of this artery at this gestation have not been studied. At day 14, the reduction in active MMP-2 is consistent with the previously reported low proliferative index at this gestation (Cipolla and Osol, 1994). Our observation of elevated gelatinolytic activity in late pregnancy (day 21) favours a role for MMP-2 in basement membrane turnover and matrix scaffolding degradation, thus facilitating cellular proliferation, known to be maximal at this time (Cipolla and Osol, 1994). MMP-2 might also contribute to vascular proliferation indirectly through altered mitogenic factor bioavailability. It is known, for example, that proteolytic degradation of decorin by MMP-2 mediates release of decorin-associated transforming growth factor-ß in connective tissue (Imai et al., 1997). A balance between proliferative and antiproliferative mechanisms is likely to be important in maintaining vascular growth within physiological limits and it is pertinent that cell-surface activation of MMP-2 can also generate potent antiproliferative autolytic fragments (Brooks et al., 1998). Persistent elevation of MMP-2 post-partum suggests a role in postnatal resorption of gestation-related uterine artery growth and is consistent with a previous study implicating MMP-2 in vessel regression observed in a rat model of angiogenesis (Zhu et al., 2000).
In this study, extraction of arterial wall MMP was maximized by two phases of protein extraction (Woessner, 1995). Zymography of urea-extracted protein revealed a gelatinolytic 43–45 kDa doublet in addition to bands at 65 and 72 kDa. These later bands are likely to be active and latent MMP-2 respectively. As far as we are aware this is the first description of gelatinolytic activity at 43–45 kDa in intact vasculature and is likely to represent an active MMP-2 species. In other tissues, cell surface-activated MMP-2 can be further processed to generate active species of 42–45 kDa with high specific activity (Howard and Banda, 1991; Itoh et al., 1998). Because of the lack of a C-terminal domain, these fragments are less susceptible to TIMP-2 inhibition (Itoh et al., 1998). Since in the present study, isolation of this doublet occurred only under harsher extraction conditions, the 43–45 kDa species may be matrix-bound in vivo. The binding of MMP-2 to gelatin or collagen I, which occurs in vitro through the fibronectin type II-like domain (Allan et al., 1995), can reduce the rate of inactivation of MMP-2 by TIMP-2 (Itoh et al., 1998). Sequestration of 43–45 kDa gelatinolytic species in uterine artery ECM may thus provide a mechanism that maintains and even enhances MMP-2 activity.
MMP-2 is often constitutively expressed and activity of this protease is considered to be predominantly regulated through activation of the latent proenzyme (Sternlicht and Werb, 2001). Our results, however, suggest that MMP-2 is also transcriptionally regulated in pregnancy, as a close relationship was observed in the temporal pattern of protein and gene expression. Several factors may contribute to pregnancy-associated MMP-2 up-regulation including estradiol (Bian and Sun, 1997; Wingrove et al., 1998). The triphasic pattern of plasma estradiol in rat pregnancy, which shows elevation between days 6 and 12, then a fall followed by a maximal increase just prior to delivery (Gibori and Sridaran, 1981), parallels the observed MMP-2 profile. Intracellular reactive oxygen species may also contribute. Flow, which is maximal in late gestation in rat uterine artery (Dowell and Kauer, 1997), is a potent stimulus to both NAD(P)H oxidase activity and therefore of superoxide generation (Laurindo et al., 1994; De Keulenaer et al., 1998).
The temporally related up-regulation of MT1MMP and TIMP-2 transcripts with MMP-2 observed is consistent with the recognized co-operation between these proteases during cell-surface activation of MMP-2 (Butler et al., 1997). The biological significance of up-regulation in late pregnancy of several apparently unrelated genes such as MMP-3, -9 and -12 is less clear. These enzymes may act synergistically to promote matrix turnover, rapid cell proliferation and subsequent vascular remodelling. MMP-3, -9 and -12 are also capable of cleaving plasminogen to yield angiostatin, a potent antiproliferative peptide (Cornelius et al., 1997). Although significantly up-regulated, the expression of these transcripts is several orders of magnitude lower than other MMPs and TIMPs studied. This may reflect cell-specific expression. For example, MMP-12, a potent elastase with striking elevation at day 21 gestation, is predominantly synthesized by macrophages (Belaaouaj et al., 2001). There may also be selective spatial expression; for instance two studies have suggested that remodelling in the uterine artery is largely confined to the cervical end of the vessel (Nienartowicz et al., 1989; Dowell and Kauer, 1997). The simultaneous up-regulation of TIMP (TIMP-1, -2) with the MMP may protect against excessive proteolysis.
The evolution of mRNA quantitative technologies such as real time PCR yields greater insight into the complex nature of gene expression. Such complexities may not always be amenable to conventional statistical analyses. In semi-quantitative studies of mRNA expression, the accepted method for correcting for sample-to-sample variation is to normalize test gene expression to an internal reference. An essential requirement is that the expression of this gene is constant and remains unaffected by the experimental treatment. In this study, we demonstrated that pregnancy influenced uterine artery expression of GAPDH, ß-actin and cyclophilin, three ‘housekeeping genes’ commonly adopted as internal reference genes. Altered expression of the reference genes studied here has also been described under a variety of experimental conditions in vitro (Bustin, 2000). We therefore considered it erroneous to ‘normalize’ test gene expression to these reference genes and elected to present the univariate analysis of gene expression in an unadjusted manner. It could be argued that part of the ‘gestational’ effect on both target and reference gene expression reflected common non-specific variation in, for example, ease of RNA extraction or sample quality. In addition, given that GAPDH, ß-actin and cyclophilin participate in processes of cellular metabolism and cytoskeleton assembly, it is possible that their up-regulation in late gestation expression may reflect a ‘global’ effect of proliferation on gene expression.
Several observations argued for additional gene-specific changes having occurred amongst the MMPs and TIMPs. In the univariate analysis, the magnitude of fold-change differed substantially between these transcripts and the reference genes. Further support came from the multivariate analysis by PLS. This offered two advantages: first, it was independent of the need for a reference gene; second, in considering the entire dataset and in plotting principal component loadings, insight into similarities in gene behaviour was facilitated. Thus genes gestationally regulated in a similar manner had loading scores which fell close to each other in space even if magnitude of fold-change may have differed. If observed changes in MMP and TIMP gene expression were truly non-specific then one would expect superimposition of PC loadings for these and reference genes in the scatterplot in Figure 1B. While some similarity does exist between the reference genes and some of the MMP genes, e.g. MMP-3, clearly not all genes studied conform to this pattern, e.g. MMP-2, -9, -12 and TIMP-1.
In summary, we propose an important role for MMPs and TIMP1/2 in growth and remodelling of rat uterine artery in pregnancy. The temporal profile of expression of the proteases may reflect specific roles in regulating these physiological processes. We have demonstrated that the expression and activity of MMP-2 is particularly enhanced in late pregnancy when cellular proliferation and arterial growth is maximal. The up-regulation of MT1MMP and TIMP-2 expression may be important in regulating cell-surface activation of MMP-2. Enhanced MMP-2 activity together with elevated expression of less abundant transcripts such as MMP-3, MMP-9 and MMP-12 in late gestation may act synergistically to promote, then arrest, vascular proliferation. Following delivery, persistent elevation of specific proteases including MMP-2 and MMP-12 together with down-regulation of TIMP1 expression may contribute to resorption and regression of neovasculature.
|
Acknowledgements |
|---|
We would like to thank Paul Murdock (GlaxoSmithKline Beecham, Harlow, UK) for donating the ß-actin and cyclophilin primers and probes and for advice on real time PCR assay development, and Mike Mitchell (Haemophilia Unit, St Thomas’ Hospital, London, UK) for use of the TaqMan. We are grateful to Yoshifumo Itoh (Kennedy Institute of Rheumatology, Imperial College, London, UK) for his advice on protein extraction and for critical review of this manuscript. This research was funded by the British Heart Foundation (BAK; Clinical PhD studentship FS99044) and by Tommy’s the Baby Charity (registered charity no. 106058).
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Allan, J.A., Docherty, A.J., Barker, P.J., Huskisson N.S., Reynolds J.J. and Murphy G. (1995) Binding of gelatinases A and B to type-I collagen and other matrix components. Biochem. J., 309, 299–306.[ISI][Medline]
Aoyagi, M., Yamamoto, M., Azuma, H., Nagashima, G., Niimi, Y., Tamaki, M., Hirakawa, K. and Yamamoto, K. (1998) Immunolocalization of matrix metalloproteinases in rabbit carotid arteries after balloon denudation. Histochem. Cell Biol., 109, 97–102.[CrossRef][ISI][Medline]
Belaaouaj, A., Shipley, J.M., Kobayashi, D.K., Zimonjic, D.B., Popescu, N., Silverman, G.A. and Shapiro, S.D. (1995) Human macrophage metalloelastase–genomic organization, chromosomal location, gene linkage, and tissue-specific expression. J. Biol. Chem., 270, 14568–14575.
Bian, J. and Sun, Y. (1997) Transcriptional activation by p53 of the human type IV collagenase (gelatinase A or matrix metalloproteinase 2) promoter. Mol. Cell. Biol., 17, 6330–6338.[Abstract]
Blankenship, T.N. and Enders, A.C. (1997) Trophoblast cell-mediated modifications to uterine spiral arteries during early gestation in the macaque. Acta Anat., 158, 227–236.[ISI][Medline]
Bond, B.C., Virley, D.J., Cairns, N.J., Hunter, A.J., Moore, G.B., Moss, S.J., Mudge, A.W., Walsh, F.S., Jazin, E. and Preece, P. (2002) The quantification of gene expression in an animal model of brain ischaemia using TaqMan(TM) real-time RT–PCR. Mol. Brain Res., 106, 101–116.[Medline]
Brooks, P.C., Silletti, S., von Schalscha, T.L. Friedlander, M. and Cheresh, D.A. (1998) Disruption of angiogenesis by PEX, a noncatalytic metalloproteinase fragment with integrin binding activity. Cell, 92, 391–400.[CrossRef][ISI][Medline]
Bustin, S.A. (2000) Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J. Mol. Endocrinol., 25, 169–193.[Abstract]
Butler, G.S., Will, H., Atkinson, S.J. and Murphy, G. (1997) Membrane-type-2 matrix metalloproteinase can initiate the processing of progelatinase A and is regulated by the tissue inhibitors of metalloproteinases. Eur. J. Biochem., 244, 653–657.[ISI][Medline]
Cipolla, M. and Osol, G. (1994) Hypertrophic and hyperplastic effects of pregnancy on the rat uterine arterial wall. Am. J. Obstet. Gynecol., 171, 805–811.[ISI][Medline]
Cornelius, L.A., Nehring, L.C., Harding, E., Bolanowski, M., Welgus, H.G., Kobayashi, D.K., Pierce, R.A. and Shapiro, S.D. (1998) Matrix metalloproteinases generate angiostatin: effects on neovascularization. J. Immunol., 161, 6845–6852.
Crowther, M., Goodall, S., Jones, J.L., Bell, P.R.F. and Thompson, M.M. (2000) Localization of matrix metalloproteinase 2 within the aneurysmal and normal aortic wall. Br. J. Surg., 87, 1391–1400.[CrossRef][ISI][Medline]
Davis, V., Persidskaia, R., Baca-Regen, L., Itoh, Y., Nagase, H., Persidsky, Y., Ghorpade, A. and Baxter, B.T. (1998) Matrix metalloproteinase-2 production and its binding to the matrix are increased in abdominal aortic aneurysms. Arterioscler. Thromb. Vasc. Biol., 18, 1625–1633.
De Keulenaer, G.W., Chappell, D.C., Ishizaka, N. Nerem, R.M., Alexander, R.W. and Griendling, K.K. (1998) Oscillatory and steady laminar shear stress differentially affect human endothelial redox state: role of a superoxide-producing NADH oxidase. Circ. Res., 82, 1094–1101.
Dowell, R.T. and Kauer, C.D. (1997) Maternal hemodynamics and uteroplacental blood flow throughout gestation in conscious rats. Meth. Find. Exp. Clin. Pharmacol., 19, 613–625.[ISI][Medline]
Fang, J., Shing, Y., Wiederschain, D., Yan, L., Butterfield, C., Jackson, G., Harper, J., Tamvakopoulos, G. and Moses, M.A. (2000) Matrix metalloproteinase-2 is required for the switch to the angiogenic phenotype in a tumor model. Proc. Natl Acad. Sci. USA, 97, 3884–3889.
Freitas, S., Meduri, G., Le Nestour E., Bausero, P. and Perrot-Applanat, M. (1999) Expression of metalloproteinases and their inhibitors in blood vessels in human endometrium. Biol. Reprod., 61, 1070–1082.
Galis, Z.S. and Khatri, J.J. (2002) Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ. Res., 90, 251–262.
Gibori, G. and Sridaran, R. (1981) Sites of androgen and estradiol production in the second half of pregnancy in the rat. Biol. Reprod., 24, 249–256.[Abstract]
Guenther, A.E., Conley, A.J., van Orden, D.E., Farley, D.B. and Ford, S.P. (1988) Structural and mechanical changes of uterine arteries during pregnancy in the pig. J. Anim. Sci., 66, 3144–3152.
Harrison, D.C., Medhurst, A.D., Bond, B.C. Campbell, C.A., Davis, R.P. and Philpott, K.L. (2000) The use of quantitative RT–PCR to measure mRNA expression in a rat model of focal ischemia—caspase-3 as a case study. Mol. Brain Res., 75, 143–149.[Medline]
Hilsenbeck, S.G., Friedrichs, W.E., Schiff, R., O’Connell, P., Hansen, R.K., Osborne, C.K. and Fuqua, S.A. (1999) Statistical Analysis of Array Expression Data as Applied to the Problem of Tamoxifen Resistance. J. Natl Cancer Inst., 91, 453–459.
Hole, S.J.W., Howe, P.W.A., Stanley, P.D. and Hadfield S.T. (2000) Pattern recognition analysis of endogenous cell metabolites for high throughput mode of action identification: removing the postscreening dilemma associated with whole organism high throughput screening. J. Biomol. Screen., 5, 335–342.[Abstract]
Howard, E.W. and Banda, M.J. (1991) Binding of tissue inhibitor of metalloproteinases 2 to two distinct sites on human 72-kDa gelatinase. Identification of a stabilization site. J. Biol. Chem., 266, 17972–17977.
Imai, K., Hiramatsu, A., Fukushima, D., Pierschbacher, M.D. and Okada, Y. (1997) Degradation of decorin by matrix metalloproteinases: identification of the cleavage sites, kinetic analyses and transforming growth factor-beta1 release. Biochem. J., 322, 809–814.[ISI][Medline]
Itoh, Y., Ito, A., Iwata, K.,Tanzawa, K., Mori, Y. and Nagase, H. (1998) Plasma membrane-bound tissue inhibitor of metalloproteinases (TIMP)-2 specifically inhibits matrix metalloproteinase 2 (Gelatinase A) activated on the cell surface. J. Biol. Chem., 273, 24360–24367.
Jovanovic, S., Grbovic, L. and Jovanovic, A. (2000) Pregnancy is associated with altered response to Neuropeptide Y in uterine artery. Mol. Hum. Reprod. 6 352–360.
Kelly, B., Stone, S. and Poston, L. (1999) Cardiovascular adaptation to pregnancy: the role of altered vascular structure. Fetal Matern. Med. Rev., 11 105–116.
Laurindo, F.R., Pedro, M.A., Barbeiro, H.V., Pileggi, F., Carvalho, M.H., Augusto, O. and da Luz, P.L. (1994) Vascular free radical release. Ex vivo and in vivo evidence for a flow-dependent endothelial mechanism. Circ. Res. 74, 700–709.
Li, Z., Froehlich, J., Galis, Z.S. and Lakatta, E.G. (1999) Increased expression of matrix metalloproteinase-2 in the thickened intima of aged rats. Hypertension, 33, 116–123.
Lijnen, H.R., Lupu, F., Moons, L., Carmeliet, P., Goulding, D. and Collen, D. (1999) Temporal and topographic matrix metalloproteinase expression after vascular injury in mice. Thromb. Haemost., 81, 799–807.[ISI][Medline]
Macdonald, R., Bingham, S., Bond, B.C., Parsons, A.A. and Philpott, K.L. (2001) Determination of changes in mRNA expression in a rat model of neuropathic pain by Taqman (TM) quantitative RT–PCR. Mol. Brain Res., 90, 48–56.[Medline]
Magness, R.R., Sullivan, J.A, Li, Y., Phernetton, T.M. and Bird, I.M. (2001) Endothelial vasodilator production by uterine and systemic arteries. VI. Ovarian and pregnancy effects on eNOS and NOx. Am. J. Physiol., 280, H1692–H1698.
Moll, W. and Gotz, R. (1985) Pressure diameter curves of mesometrial arteries of guinea pigs demonstrate a non-muscular, estrogen-inducible mechanism of lumen regulation. Pflügers Archiv-Eur. J. Physiol., 404, 332–336.[CrossRef][ISI][Medline]
Moll, W., Espach, A. and Wrobel, K.H. (1983) Growth of mesometrial arteries in guinea pigs during pregnancy. Placenta, 4, 111–123.[CrossRef][ISI][Medline]
Nelson, S.H., Steinsland, O.S., Wang,Y., Yallampalli, C., Dong, Y.L. and Sanchez, J.M. (2000) Increased nitric oxide synthase activity and expression in the human uterine artery during pregnancy. Circ. Res., 87, 406–411.
Nienartowicz, A., Link, S. and Moll, W. (1989) Adaptation of the uterine arcade in rats to pregnancy. J. Dev. Physiol., 12, 101–108.[ISI][Medline]
Osol, G. and Cipolla, M. (1993) Pregnancy-induced changes in the 3-dimensional mechanical properties of pressurised rat uteroplacental (radial) arteries. Am. J. Obstet. Gynecol., 168, 268–274.[ISI][Medline]
Palmer, S.K., Zamudio, S., Coffin, C., Parker, S., Stamm, E. and Moore, L.G. (1992) Quantitative estimation of human uterine artery blood flow and pelvic blood flow redistribution in pregnancy. Obstet. Gynecol., 80, 1000–1006.
Pasterkamp, G., Schoneveld, A.H., Hijnen, D.J., de Kleijn, D.P.V., Teepen, H., van der Wal, A.C. and Borst, C. (2000) Atherosclerotic arterial remodeling and the localization of macrophages and matrix metalloproteases 1, 2 and 9 in the human coronary artery. Atherosclerosis, 150, 245–253.[CrossRef][ISI][Medline]
Pijnenborg, R., Robertson, W.B. and Brosens, I. (1983) Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. Placenta, 4, 397–414.[ISI][Medline]
Sternlicht, M.D. and Werb, Z. (2001) How matrix metalloproteinases regulate cell behaviour. Annu. Rev. Cell. Dev. Biol., 17, 463–516.[CrossRef][ISI][Medline]
Weiner, C.P., Thompson, L.P., Liu, K.Z. and Herrig, J.E. (1992) Endothelium-derived relaxing factor and indomethacin-sensitive contracting factor alter arterial contractile responses to thromboxane during pregnancy. Am. J. Obstet. Gynecol., 166, 1171–1178.[ISI][Medline]
Wingrove, C.S., Garr, E., Godsland, I.F. and Stevenson, J.C. (1998) 17beta-oestradiol enhances release of matrix metalloproteinase-2 from human vascular smooth muscle cells. Biochim. Biophys. Acta, 1406, 169–174.[Medline]
Woessner, J.F. (1995) Quantification of matrix metalloproteinases in tissue samples. Methods Enzymol., 248, 510–528.[ISI][Medline]
Woessner, J.F. and Nagase, H. (2000) Protein substrates of the MMPs. In Matrix Metalloproteinases and TIMPs, 1st edn. Oxford University Press, Oxford, pp. 87–97.
Wold, S., Albano, C., Dunn, W.J.I., Edlund, U., Esbensen, K., Geladi, P., Hellberg, S., Johansson, E., Lindberg, W. and Sjostrom, M. (1984) Multivariate data analysis in chemistry. In Kowalski, B.R. (ed.), Chemometrics: Mathematics and Statistics in Chemistry. Reidel, Dordrecht, pp. 71–96.
Zhu, W.H., Guo, X., Villaschi, S. and Nicosia, R.F. (2000) Regulation of vascular growth and regression by matrix metalloproteinases in the rat aorta model of angiogenesis. Lab. Invest., 80, 545–555.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. O'Brien, D. O'Shaughnessy, E. Ahamide, J. J. Morrison, and T. J. Smith Differential expression of the metalloproteinase MMP3 and the {alpha}5 integrin subunit in human myometrium at labour Mol. Hum. Reprod., September 1, 2007; 13(9): 655 - 661. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.A. Kelly, B.C. Bond, and L. Poston Aortic adaptation to pregnancy: elevated expression of matrix metalloproteinases-2 and -3 in rat gestation Mol. Hum. Reprod., May 1, 2004; 10(5): 331 - 337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Dalrymple, D. M. Slater, L. Poston, and R. M. Tribe Physiological Induction of Transient Receptor Potential Canonical Proteins, Calcium Entry Channels, in Human Myometrium: Influence of Pregnancy, Labor, and Interleukin-1{beta} J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1291 - 1300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Jeyabalan, J. Novak, L. A. Danielson, L. J. Kerchner, S. L. Opett, and K. P. Conrad Essential Role for Vascular Gelatinase Activity in Relaxin-Induced Renal Vasodilation, Hyperfiltration, and Reduced Myogenic Reactivity of Small Arteries Circ. Res., December 12, 2003; 93(12): 1249 - 1257. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||