| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,1
* Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh School of Medicine and Magee-Women’s Research Institute, Pittsburgh, Pennsylvania, USA;
Department of Zoology, University of Melbourne, Parkville, Victoria, Australia;
Department of Bioengineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA; and
Department of Physiology and Cell Biology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
1Correspondence: Department of Physiology and Functional Genomics, University of Florida College of Medicine, 1600 SW Archer Rd., M552, P.O. Box 100274, Gainesville, FL 32610-0274, USA. E-mail: kpconrad{at}ufl.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: vascular biology • relaxin knockout mouse • vasodilation • myogenic reactivity • passive compliance
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). However, recent evidence indicates that administration of either porcine or recombinant human (rh) relaxin to nonpregnant female and male rats exerts potent vascular effects that mimic the pregnant state. These actions include renal and systemic vasodilatation, antagonism of angiotensin (ANG) II-mediated vasoconstriction, increased global arterial compliance, as well as inhibited myogenic reactivity and increased passive compliance of isolated small arteries (reviewed in ref. 2
). Administration of relaxin to humans also promotes renal vasodilatation (3)
. By neutralizing endogenous, circulating relaxin during midterm pregnancy with antibodies or removing circulating hormone by ovariectomy, the renal and systemic circulatory changes of pregnancy are abolished in rats (4
, 5)
. Moreover, women who lack circulating relaxin due to ovarian failure and become pregnant through egg donation show a markedly subdued gestational increase in glomerular filtration rate (6)
. The renal vasodilatory action of relaxin in rats is mediated by up-regulation of vascular gelatinase activity, which processes big endothelin-1 at a gly-leu bond to ET1–32, thereby stimulating the endothelial ETB-NO vasodilatory pathway (7
, 8)
.
Humans have three relaxin genes designated H1, H2, and H3. Rats and mice each have two relaxin genes, designated relaxin-1 and relaxin-3 (9)
. Human H2 relaxin, as well as rat and mouse relaxin-1 gene products, are analogous insofar as they are secreted by the corpus luteum during pregnancy and circulate. To date, relaxin-3 expression has only been confirmed in the brain, specifically in the pars ventromedialis of the dorsal tegmental nucleus within the pons medulla (10)
. A third relaxin-like molecule, Leydig cell insulin-like peptide (INSL3), is expressed in the gonads and has an important role in testicular descent (11)
.
The receptors for relaxin are two leucine-rich repeat containing, guanine nucleotide binding (G-protein)- coupled receptors (GPCRs) known as LGR7 and LGR8 (12)
. Several pharmacological studies have reported that relaxin binds to both LGR7 and LGR8 and stimulates a Gs-cAMP-protein kinase A (PKA)-dependent signaling pathway, whereas INSL3 only activates LGR8 (13
, 14)
. Relaxin-1 promotes cell proliferation in the rat gubernaculum in vitro despite the lack of LGR7 receptors in this tissue (15)
, implying that relaxin-1 can produce an equivalent biological response to INSL3 by activating LGR8. However, overexpression of relaxin-1 in transgenic mice had no effect on testicular descent in male mice deficient in INSL3 (16)
. So, although there is some overlap of relaxin ligand binding in vitro, this does not seem to be the case in vivo at least in the gubernaculum; thus, LGR7 is most likely the predominant receptor for relaxin-1. Relaxin-3 also binds to LGR7 but with relatively low affinity compared with relaxin-1, and does not stimulate cAMP with the same potency (17)
. Two specific receptors for relaxin-3 have recently been identified as GPCR135 and GPCR142 (18
, 19)
. These receptors are highly expressed in the brain and testis of rodents, although GPCR142 is a pseudogene in rats.
LGR7 receptors have been localized in epithelial and stromal compartments of the uterine endometrium as well as showing strong expression in the myometrium, cervix, nipple, fetal membranes, and placental villi in humans and/or rodents (20
21
22)
. None of these studies assessed LGR7 receptors in vascular elements. Relaxin binding sites were identified earlier in blood vessels of reproductive organs using biotinylated porcine relaxin in the pig and human uterus (23
, 24)
. Based on the potent vascular effects of circulating relaxin, we postulated that blood vessels express the major receptor for relaxin-1, LGR7, thereby mediating local vasodilatation and increases in passive compliance. It is also possible that a relaxin peptide is produced by arteries and acts locally to regulate arterial function. Therefore, we had two primary objectives in this study: first, to investigate the expression of gene transcripts and, where possible, proteins for relaxin-1 and its putative receptors (LGR7 and LGR8) in the thoracic aortas, small renal and mesenteric arteries from mice, rats, and a marsupial, the tammar wallaby. Three different species were investigated in order to explore how widespread arterial relaxin-1 and LGR7 expression is across the animal kingdom, with the view that conservation among species may indicate important functional roles. This was expanded to include a quantitative analysis of gene expression in the small renal arteries of mice to establish which relaxin receptors are predominantly expressed in these blood vessels and whether there are differences between males and females. Second, functional studies were conducted on isolated small renal arteries of mice to elucidate the effects of relaxin-1 deficiency on myogenic reactivity and passive compliance in nonpregnant females and males.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals
Long-Evans male and female rats 10–14 wk of age were purchased from Harlan Sprague-Dawley (Frederick, MD, USA). C57BLK/6J mice 2–3 or 4–6 months of age were purchased from Hilltop Lab Animals, Inc (Scottsdale, PA, USA). The functional studies used relaxin-1 gene-deficient (Rlx–/–) mice backcrossed on a C57BLK/6J background to the F14 generation. In these mice, a 750 bp relaxin-1 genomic DNA fragment encoding 90 C-terminal amino acids and 17 N-terminal amino acids of the A chain was replaced with a neomycin transferase gene (25)
. WT (Rlx+/+), heterozygotes (Rlx+/–), and Rlx–/– littermates were available at 4–6 months of age. Due to the limited availability of these mice, investigation of arterial myogenic reactivity was confined to females, and the complete study of arterial passive mechanics was conducted in males. Female tammar wallaby (Macropus eugenii) vascular tissues were collected from colony-derived animals at the CSIRO Sustainable Ecosystems (Gungahlin, Canberra, ACT, Australia) under permit AEEC 03/04–13. All investigations conformed to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85–23, revised 1996). All studies in rodents and wallabies were approved by the relevant Institutional Animal Care and Use Committees.
Dissection of arteries for molecular analyses
Rats and wallabies were euthanized with sodium pentobarbital, and mice were anesthetized with isoflurane prior to cervical dislocation. Arteries were isolated for molecular studies as previously reported (7
, 8)
.
Reverse transcription-PCR (RT-PCR)
Isolation of total RNA from pools of arteries was conducted using RNAwiz (Ambion, Inc., Austin, TX, USA). cDNA was synthesized from 1–5 µg of total RNA using SuperScript II RNase H-Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA). Polymerase chain reaction (PCR) was conducted using TaqDNA Polymerase (Promega, Madison, WI, USA). The species-specific PCR primers and PCR conditions used in this study are outlined in supplemental Table 1 and are based on earlier studies (26
27
28)
. PCR products were previously sequenced (29
30
31)
.
Quantitative reverse transcription-polymerase chain reaction (qRT-PCR)
The relative expression of relaxin-1, LGR7, and LGR8 was assessed in the small renal arteries of male and female mice aged 2–3 months as described in earlier studies (26
, 27)
. To obtain sufficient RNA, pools of arteries were obtained from three mice (n=4–5 pools/artery/sex). Forward/reverse primers and FAM (6-carboxyl fluorescein)-labeled TaqMan® probes (outlined in supplemental Table 2) specific for mouse genes were designed to span introns to eliminate false positives that might arise from genomic DNA contamination. The endogenous reference gene was ribosomal 18S. Quantitative PCR was carried out in the ABI 7700 PCR machine (Applied Biosystems, Foster City, CA, USA). In these Q-PCR experiments, the comparative CT method was used according to the ABI Users’ Bulletin #2. The gene of interest and the endogenous reference gene (18S) were assessed in separate PCR reactions.
Western blot analysis
The preparation of tissues and Western blot procedures were conducted as described previously with minor modifications (8)
. Immunoblots were incubated overnight at 4°C with antirat relaxin monoclonal antibodies (MCA1) or with rabbit anti-LGR7 antibodies directed against amino acid residues 107–119 or residues 696–710 of the mouse and rat LGR7 protein (all antibodies provided by Dr. O.D. Sherwood). LGR7 antibodies were first purified from immunized rabbit serum using protein A/G PLUS-agarose (Santa Cruz Biologicals, Santa Cruz, CA, USA).
Immunohistochemistry
Standard techniques were utilized as published (8)
. Immunohistochemistry was performed using a monoclonal antibody (mAb) directed against rat relaxin, designated MCA1. After testing a dose response (1–10 µg/ml), a final concentration of 3 µg/ml was chosen as optimal. For a negative control, the same concentration of mouse IgG1k was substituted for the primary antibody (Ab).
Renal artery myogenic reactivity
Myogenic reactivity is the vasoconstrictory response of a small artery to a step increase in intraluminal pressure. A branch of the main renal artery was dissected free of surrounding tissue (unpressurized inner diameter, 
100 µm). An arterial segment was then transferred to the isobaric arteriograph (Living Systems, Burlington, VT, USA) and mounted on two microcannulae suspended in the chamber (7)
. After the diameter was recorded at 60 mmHg, the intraluminal pressure was rapidly increased in a stepwise manner to 80 mmHg. The data were expressed as percent change in diameter at 80 mmHg compared with the diameter at 60 mmHg. The three responses from each vessel were averaged. After the initial set of responses, the bath was rinsed with warm buffer and L-arginine or D-arginine were added to achieve a final concentration of 100 µM. The arteries were incubated for 30 min, then the series of pressure step was repeated. Finally, arterial passive mechanics were assessed from 60 to 80 mmHg (see below). The data were analyzed by single factor randomized block design ANOVA. If a significant main effect was observed, then group means were contrasted by the Fisher least significant difference (LSD) test.
Arterial passive mechanics
The isobaric arteriograph was used to measure the distensibility of the renal arteries and to detect changes in passive mechanics as described earlier (32)
. Briefly, the arteries were incubated in calcium-free, EGTA-containing buffer with papavarine to inhibit vascular smooth muscle function. Pressure-diameter relationships were generated, and this information was used for subsequent calculations. The relationship between circumferential wall stress (
) and midwall radius was determined as reported by Cholley and colleagues (33)
. This relationship describes the mechanical properties of the vessel. These parameters were normalized for wall thickness and therefore characterize the stiffness of the components that comprise the vascular wall. Least squares regression analysis was performed on the –Rm relationship. Analysis of excess variance (or extra sum of squares) was used to compare the relationships among mice genotypes (34)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). A 290 bp amplicon consistent with relaxin-1 in the positive control tissue (ovary) was expressed in thoracic aortas, small renal, and mesenteric arteries (Fig. 1A
). These data were obtained from vessels pooled from three mice of each sex. Two LGR7 gene transcripts (450 and 320 bp) were present in the positive control tissue (myometrium) and also in all arteries examined (Fig. 1B
). These correspond to a full-length form of LGR7 and a splice variant previously reported in the mouse (26)
. The full-length variant was the predominantly expressed transcript. There were no obvious differences in gene expression between males and females as revealed by nonquantitative RT-PCR. The consistent expression of ß-actin in all tissues indicated that the quality of cDNA synthesis was comparable for all samples. For logistical reasons, the function of small renal arteries was investigated in older mice of 4–6 months of age (see below), and we further established that both relaxin-1 and LGR7 were expressed in the arteries of mice from this older age group (Fig. 2
).
|
|
To begin exploring how widespread arterial relaxin-1 and LGR7 gene expression is across the animal kingdom, we also examined vessels in the rat and tammar wallaby. Both relaxin-1 and LGR7 gene transcripts were detected in thoracic aortas and in mesenteric and small renal arteries of male and female rats (Fig. 1)
. The initial rat RT-PCR reaction for relaxin-1 yielded the predicted amplicons of 519 and 214 bp, respectively, but the signals were inconsistent or weak because of lower than expected RNA concentrations (data not shown). Therefore, the two primer sets were combined for nested primer RT-PCR, which resulted in a 214 bp amplicon consistent with relaxin-1 gene expression in the ovary (Fig. 1C
). A nested primer RT-PCR reaction also yielded a 286 bp amplicon consistent with LGR7 gene expression in the myometrium (Fig. 1D
). The upper band in each sample of the LGR7 reaction did not correspond to a known LGR7 splice variant, but there was insufficient DNA extracted from the gel to conduct sequence analysis. One mesenteric artery sample lacked reverse transcriptase in the cDNA synthesis and was an additional negative control for genomic DNA contamination in relaxin-1 and ß-actin PCR reactions (asterisk in Fig. 1C
).
In the tammar wallaby, the small renal arteries of one late pregnant and one nonpregnant female were assessed. Both relaxin-1 and LGR7 gene transcripts were detected in this artery as well as in the renal papilla (Fig. 1E
). The 350 bp relaxin-1 amplicon was consistent with relaxin-1 gene expression in the ovary and yolk sac membrane (data not shown). There was only one LGR7 transcript (253 bp) in the positive control tissue, the nipple. To date, no splice variant for LGR7 has been demonstrated in any tissue in the wallaby (30)
. There was strong expression of GAPDH in all tissues, indicating equivalent quality of cDNA synthesis in all samples.
Quantitative analysis of gene expression
Q-PCR confirmed the presence of relaxin-1, LGR7, and LGR8 receptor expression in small renal arteries from female and male mice (Fig. 3
A–C). Gel electrophoresis of the Q-PCR products obtained after 40 cycles demonstrated a single amplicon for each gene, equivalent to the product detected in the positive control samples (Fig. 3)
. Comparative analyses between receptor genes and sexes were also conducted by relating all data to LGR7 gene expression in the small renal arteries of female mice. On average, there was higher expression of LGR7 and LGR8 in the female small renal arteries compared with males (Fig. 3D
), although the ranges of gene expression were considerable.
|
Relaxin and LGR7 protein expression in arteries
Few specific antibodies are available for detection of relaxin and LGR7 proteins; therefore, we could only investigate rat relaxin and rat/mouse LGR7 protein expression. Upon Western blot analysis (nonreducing conditions), we detected both purified rat relaxin and mature relaxin in ovaries from gravid rats (positive control, 6 kDa; Fig. 4
A). In addition, higher molecular mass bands consistent with prorelaxin (15, 16.5, and 18 kDa) and preprorelaxin (21 kDa) were observed in the ovaries. As expected, these proteins were expressed more abundantly in ovaries from late than midterm pregnant rats. All of these immunoreactive relaxin protein bands were observed in the small renal arteries from a male rat, although the mature form was less consistently detected and strongly expressed relative to the positive control (Fig. 4A
). Decreasing amounts of protein loaded on the gel resulted in immunoreactive bands of less intensity. In addition, there was a possible dimer of preprorelaxin at 42 kDa in the arteries (Fig. 4A
) that completely disappeared under reducing conditions (data not shown). When mouse IgG1
was substituted for the MCA1 Ab as a control, immunoreactive bands were absent or virtually so (Fig. 4B
).
|
The relaxin Ab used in the Western blot analysis was also used in immunohistochemistry. Immunoreactive relaxin was detected in the corpora lutea of late pregnant but not virgin rats (Fig. 5
A, B). These tissues represent positive and negative tissue controls, respectively (1)
. There was strong positive immunostaining for relaxin and precursor forms in both the endothelium and vascular smooth muscle of a small renal artery from a virgin female rat (Fig. 5D
). Incubation of either the late pregnant ovary or small renal artery with mouse IgG1 instead of the MCA1 primary Ab negated the relaxin immunoreactivity, further demonstrating specificity of the MCA1 Ab for relaxin in both of these tissues (Fig. 5C, E
).
|
It was initially necessary to validate the two different polyclonal LGR7 antibodies for use in Western blot analysis in mouse and rat tissues (Fig. 6
A, B, D). Cardiac atrium and gravid myometrium served as positive control tissues based on strong mRNA expression in earlier studies. Immunoreactive proteins of 100 kDa were detected in mouse atrium and gravid myometrium by both LGR7 antibodies. A second 87 kDa immunoreactive protein was detected in the myometrium. Two immunoreactive LGR7 proteins of >100 kDa were detected in rat atrium compared with the one 100 kDa protein in the mouse atrium (Fig. 6D
). These immunoreactive proteins were not observed or were markedly diminished when either rabbit IgG or preimmune serum was substituted for the primary Ab or when the primary Ab was preabsorbed with the immunizing LGR7 peptide (Fig. 6C, E, F
). Therefore, the two antibodies were specific for LGR7 proteins in the atrium and myometrium in both mice and rats.
|
Figure 7A demonstrates an immunoreactive LGR7 protein of 100 kDa in small renal and mesenteric arteries as well as thoracic aortas from male and female mice. This protein aligns with the 100 kDa protein in the positive control tissues. The 87 kDa protein observed in the mouse myometrium was less intense in the arteries. Similar data were obtained in the rat (Fig. 7B
). Two immunoreactive LGR7 proteins of >100 kDa were present in the small renal arteries of males and females and aligned with the two bands in the rat atrium.
|
Despite their successful application in Western blot analysis, neither LGR7 Ab worked satisfactorily in immunohistochemistry. Therefore, we were not able to immunolocalize LGR7 expression in arteries.
Effects of relaxin-1 deficiency on myogenic reactivity and passive compliance in small renal arteries
To test the potential impact of a local, vascular-derived relaxin-1/LGR7 system on the function of arteries, we first investigated myogenic reactivity of small renal arteries from relaxin-1-deficient (Rlx–/–) mice and WT (Rlx+/+) female littermates aged 4–6 months. In response to a 20 mmHg increase in intraluminal pressure from 60 mmHg to 80 mmHg, myogenic reactivity was comparably robust under control conditions between the two genotypes (Fig. 8
A, B). However, when physiological concentrations of L-arginine, but not D-arginine, were added to the media to provide substrate for endothelial NOS, the myogenic reactivity of small renal arteries from Rlx+/+ mice was significantly inhibited relative to the control or D-arginine conditions (P<0.004 by ANOVA; Fig. 8
). Similar results were observed for heterozygous (Rlx+/–) female mice (% change from baseline: control 0.9±1.1; L-Arg 5.4±1.6; D-Arg 0.8±2.7), but these changes did not reach statistical significance (P=0.08 by ANOVA). In contrast, there was no significant inhibition of myogenic reactivity in the small renal arteries of Rlx–/– mice after addition of L-arginine (P>0.05; Fig. 8
).
|
After measurements of myogenic reactivity, calcium-free buffer containing EGTA and papavarine was substituted and passive mechanical properties were assessed during a 60 to 80 mmHg increase in intraluminal pressure. The reduction in myogenic reactivity of small renal arteries isolated from the WT and heterozygous mice and incubated with L-arginine as noted above was not significantly different from the passive mechanical changes observed with a 60 to 80 mmHg increase in intraluminal pressure (% change from baseline: Rlx+/+ 7.8±1.0; Rlx+/– 7.4±0.9; and Rlx–/– 6.6±1.3).
In a second functional study, we investigated passive compliance of small renal arteries obtained from male Rlx+/+ and Rlx–/– mice of 4–6 months of age. After pharmacological inhibition of vascular smooth muscle function, intraluminal pressure was increased in a stepwise fashion, and the vessel diameter was assessed after a brief stabilization period. The arteries were significantly stiffer (less compliant) in the Rlx–/– mice compared with the WT mice (P<0.001; Fig. 8C
).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Relaxin-1 and LGR7 gene expression were detected in isolated small renal and mesenteric arteries, and in thoracic aortas from male and female mice using standard RT-PCR. To begin to address how widespread the phenomenon of vascular expression of relaxin-1 and LGR7 might be among mammalian species, we investigated rats and tammar wallabies. In both male and female rats, relaxin-1 and LGR7 mRNA were also expressed by isolated small renal and mesenteric arteries, and thoracic aortas. In a nonpregnant and pregnant female tammar wallaby, both relaxin-1 and LGR7 were expressed by small renal arteries (other arteries were not investigated). These data on vascular expression of relaxin corroborate a report of H1 and H2 relaxin gene transcripts in human saphenous vein and mammary artery (35)
. The identification of relaxin and its major receptor, LGR7, in blood vessels of several mammalian species highlights the conserved and likely important functional role for relaxin in arteries. Based on these results, further investigation of human arteries is clearly indicated.
The results from real-time RT-PCR both corroborated and extended those obtained by nonquantitative RT-PCR. First, the quantitative analysis confirmed the expression of relaxin-1 and LGR7 mRNA in small renal arteries from male and female mice. Second, it demonstrated the presence of a second potential relaxin receptor, LGR8, in these arteries. Third, the results showed that small renal arteries from female mice expressed more LGR7 and LGR8 mRNA than those from males. A difference was also observed by standard RT-PCR for LGR7 between the genders in rats, with greater expression in arteries from females. Despite these differences in receptor expression between the sexes, there was no difference in relaxin-1 gene expression.
It was important to verify the findings of relaxin-1 and LGR7 mRNA in arteries with protein expression data. Because there is no specific Ab for mouse relaxin, we investigated relaxin protein expression in rat arteries with a well-characterized Ab to rat relaxin (MCA1; ref. 36
). Using nonreducing conditions in order to preserve the integrity of mature relaxin, which has two interchain disulfide bonds, we observed immunoreactive bands consistent with prepro-, pro-, and mature relaxin in the pregnant rat ovary (positive control; ref. 37
). As expected, immunoreactivity was greater in late than in midterm pregnant ovary, and the mature relaxin of 6 kDa aligned with purified rat relaxin. Immunoreactive bands consistent with prepro- and prorelaxin were consistently detected in arteries from nonpregnant female and male rats, but the mature form was not as consistently or strongly expressed. One explanation for the stronger expression of prepro- and prorelaxin relative to the mature form in arteries is that relaxin processing may be less complete than in the ovary, where it was more extensive and poised toward mature relaxin, presumably for secretion. However, prorelaxin has been found to be as active as mature relaxin in the rat atrial inotropy and chronotropy bioassay (38)
, and so has the potential to be a major bioactive relaxin molecule in blood vessels. The same Ab was used to localize relaxin in small renal arteries of rats. Intense immunostaining was observed in ovaries from late pregnant, but not virgin, rats and in positive and negative tissue controls, respectively (1)
. Immunoreactive relaxin was present in both vascular smooth muscle and the endothelium of these blood vessels. As the MCA1 Ab also cross-reacts with the preprorelaxin and prorelaxin molecules, it is not possible to clarify the exact molecular forms of the protein expressed in each vascular compartment. Moreover, it is not known whether the MCA1 Ab specifically detects the relaxin-1 peptide (vs. relaxin-3).
The identification of LGR7 mRNA in arteries of all three mammalian species examined was a crucial finding. However, it was important to substantiate the gene expression data and demonstrate LGR7 protein in blood vessels. In fact, to our knowledge the present data represent the first Western blot analyses for LGR7. Initially, antibodies generated against identical peptide sequences in the mouse and rat receptor were characterized using positive control tissues such as the cardiac atria and myometrium. The predicted molecular mass of the LGR7 is 87 kDa (39)
. There is also a known splice variant in both the human and mouse LGR7 genes (39)
we detected by our primer set in mouse vascular tissues by RT-PCR. But because this splice variant lacks 102 base pairs or 34 amino acids (from the NH2-terminal region of its ectodomain) compared with the full-length form, the predicted difference in molecular mass is only 4 kDa less, and was not resolved by our Western blot analysis. The data showed bands at 87 and/or 100 kDa in the positive control tissues. The former is the predicted molecular mass of the full length-form of LGR7 and the latter is consistent with a glycosylated protein as members of this family of receptors are glycosylated (39)
. The same bands were detected using two different rabbit polyclonal antibodies, each generated against a different peptide sequence in the LGR7 receptor. In various arteries from male and female mice, a predominant band of 100 kDa was detected, with the 87 kDa form being expressed less. In rat tissues, two bands—one of 100 kDa and the other slightly greater than 100 kDa—were detected in the positive control tissues, consistent with glycosylated forms of the LGR7 receptor. In the small renal arteries from male and female rats, the 100 kDa glycosylated form predominated. Relaxin binding sites have been identified in blood vessels of the pig and human uterus using biotinylated porcine relaxin (23
, 24)
. One shortcoming of these data is that biotinylated relaxin would have bound to LGR8 as well as LGR7. Our data can now confirm that receptors for relaxin-1 are present in blood vessels, as the two LGR7 antibodies used in this study are specific for LGR7. This study was not designed to investigate potential differences in LGR7 protein expression in arteries between female and male rodents. Moreover, the anatomical localization of LGR7 within the vascular wall remains to be identified pending the availability of suitable antibodies.
Although we have demonstrated a local, arterial-derived relaxin hormone receptor system, it was important to identify its functional significance, if any, on arterial behavior. In this study we used a relaxin-1 gene knockout (Rlx–/–) mouse to investigate potential aberrant behavior in small renal arteries. The prediction was that these arteries would show increased myogenic reactivity (i.e., enhanced vasoconstriction in response to an increase in intraluminal pressure) because exposure of arteries from rats to relaxin results in blunted myogenic reactivity in a NO-dependent fashion (2
, 40)
. As well, the reduced myogenic reactivity of small renal arteries is abolished during pregnancy in rats by neutralizing endogenous circulating relaxin with antibodies (4)
. The small renal arteries from female Rlx-1+/+ and Rlx-1+/– mice showed a loss of myogenic reactivity but only when incubated with physiological concentrations of L-arginine, the substrate for the generation of NO. Thus, the inhibition of myogenic reactivity in small renal arteries from mice expressing the relaxin-1 gene was dependent on arginine transport and consistent with one view of the so-called arginine paradox (41)
. This inhibition of myogenic reactivity was profound because the increase in diameter virtually matched that observed when calcium-free buffer containing EGTA and papavarine were subsequently tested in the same arteries. In contrast, the small renal arteries from female Rlx-1–/– mice failed to demonstrate a significant reduction in myogenic reactivity despite incubation with L-arginine. This indicates an important role for vascular-derived relaxin-1 in mediating reduced myogenic reactivity in the WT and heterozygous mice via an NO-dependent vasodilatory pathway. Thus, arterial-derived relaxin-1 exerts a local vasodilatory action in nonpregnant female mice. Most likely, this effect is mediated by the major relaxin-1 receptor, LGR7, and up-regulation of vascular gelatinase and endothelial ETB receptor activity (2
, 7
, 8)
, although this remains to be proved. Because arterial resistance is inversely proportional to radius raised to the fourth power, we expect significantly lower vascular resistance and higher blood flow in those mice that express an arterial-derived relaxin system, a hypothesis to be tested in future studies.
We also predicted that arteries harvested from Rlx-1–/– mice would manifest reduced passive compliance because administration of rhRLX to nonpregnant rats results in enhanced passive compliance of isolated arteries ex vivo, which is consistent with alterations in vascular structure (32)
. In this study, the small renal arteries harvested from male relaxin-1 knockout mice were less compliant or stiffer than the arteries from the WT animals. Although the nature of the changes in the vascular structure and underlying mechanisms in response to relaxin are presently unknown and beyond the scope of the current work, it is of interest that, in addition to initiating the vasodilatory pathway described above, gelatinase activity is also involved in matrix turnover in the vascular wall (42)
. Thus, overlapping hormonal and molecular mechanisms (e.g., relaxin and gelatinase activity) are likely to mediate both vasodilatory and compliance changes. This sharing of molecular mechanisms is one way to ensure a temporal and spatial coordination of both steady and pulsatile arterial loads that is critical to cardiovascular homeostasis.
In conclusion, we provide strong evidence for the existence of an arterial-derived relaxin hormone-relaxin receptor system that locally acts to reduce myogenic reactivity and increase arterial compliance. It is possible that aberrant expression of arterial relaxin or its receptor contributes to vascular pathologies (e.g., a deficiency might contribute to increased arterial constriction and stiffness associated with aging, both normal and accelerated), and an excess might contribute to aortic aneurysm formation and dissection.
|
ACKNOWLEDGMENTS |
|---|
Received for publication April 20, 2006. Accepted for publication July 19, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
expression, but not relaxin receptors (LGR7), in the myometrium of late pregnant relaxin gene knockout mice. Endocrinology 144,4272-4275This article has been cited by other articles:
|
|
 |
|
  L. C. Cardoso, A. R. Nascimento, C. Royer, C. S. Porto, and M. F. M. Lazari Locally produced relaxin may affect testis and vas deferens function in rats Reproduction, January 1, 2010; 139(1): 185 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Resch, R. Wiest, L. Moleda, S. Fredersdorf, B. Stoelcker, J. A. Schroeder, J. Scholmerich, and D. H. Endemann Alterations in mechanical properties of mesenteric resistance arteries in experimental portal hypertension Am J Physiol Gastrointest Liver Physiol, October 1, 2009; 297(4): G849 - G857. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kern, D. Hubbard, A. Amano, and G. D. Bryant-Greenwood Cloning, Expression, and Functional Characterization of Relaxin Receptor (Leucine-Rich Repeat-Containing G Protein-Coupled Receptor 7) Splice Variants from Human Fetal Membranes Endocrinology, March 1, 2008; 149(3): 1277 - 1294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Baccari, S. Nistri, M. G. Vannucchi, F. Calamai, and D. Bani Reversal by relaxin of altered ileal spontaneous contractions in dystrophic (mdx) mice through a nitric oxide-mediated mechanism Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2007; 293(2): R662 - R668. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CT ± SD. Primers and the FAM-labeled probe were specific to the full-length variant of the murine Lgr7. Confirmation that a single amplicon was obtained after 40 PCR cycles is shown in the right panels. +ve control tissues were the ovary for relaxin-1, cervix for Lgr7 and endometrium for Lgr8; NTC: dH2O replaced the cDNA template. D) Comparative analysis of Lgr7 and Lgr8 gene expression in the small renal arteries of female and male mice aged 2–3 months (n=4 pools of 3 mice) using Q-PCR. Values are 2–
