Brought to you by Portland Press Ltd.
Published on behalf of the International Federation for Cell Biology
Cancer Cell death Cell cycle Cytoskeleton Exo/endocytosis Differentiation Division Organelles Signalling Stem cells Trafficking
Cell Biology International (2006) 30, 533–540 (Printed in Great Britain)
Adenylyl cyclase signaling mechanisms of relaxin and insulin action: Similarities and differences
Marianna Pertsevaa*, Alexander Shpakova, Ludmila Kuznetsovaa, Svetlana Plesnevaa and Evgeniya Omeljaniukb
aSechenov Institute of Evolutionary Physiology and Biochemistry, Russian Academy of Sciences, Thorez pr. 44, 194223, St. Petersburg, Russia
bOtt Institute of Obstetrics and Gynecology, Russian Academy of Medical Sciences, Mendeleev line 3, 199034, St. Petersburg, Russia


Abstract

The adenylyl cyclase signaling mechanism (ACSM) of relaxin H2 action was discovered and deciphered in mammalian muscles. A study of signaling blocks involved in ACSM of relaxin in comparison with that of insulin previously detected showed a close similarity throughout the post-receptor signaling chain of both hormones. The inhibitory action of tyrosine kinase blockers on the hormone AC activating effect indicates that the relaxin receptor involved in ACSM is likely to be of the tyrosine kinase type. However, a recent discovery of a relaxin receptor with serpentine architecture leaves open the question concerning the existence of receptor of the tyrosine kinase type. The structural-functional organization of the ACSM due to the action of relaxin—shown here for the first time—can be presented as the following signaling sequence: relaxin receptorGi protein (βγ-dimer)phosphatidylinositol 3-kinaseprotein kinase CζGs proteinadenylyl cyclase. According to our hypothesis, the regulatory action of the insulin superfamily peptides on cell processes (proliferation, apoptosis, and metabolism) is mediated via ACSM.


Keywords: Adenylyl cyclase signaling, Relaxin, Insulin, Human, Rat.

*Corresponding author. Fax: +7 (812) 552 3012.


1 Introduction

In the last decade, interest in relaxin has continued to grow, probably because of the following three findings. First, in addition to the regulatory function in the reproductive system, this hormone has a variety of roles in the cardiovascular, renal, central nervous and other systems (Ivell and Einspanier, 2002). Second, new hormones of the relaxin family (relaxin-3, relaxin-like peptides and others) were discovered. And finally, some relaxin receptors of the serpentine type have been cloned (Hsu et al., 2000, 2002; Liu et al., 2003a,b; Scott et al., 2004).

Relaxin is one of the regulatory peptides of the insulin superfamily, which includes insulin, other insulin-like hormones, and growth factors of vertebrates and invertebrates. These peptides have a structural-functional similarity consistent with a common evolutionary origin (Chan et al., 1992; Murray-Rust et al., 1992). Relaxin is a pleiotropic endocrine and paracrine factor capable of regulating many cell processes, such as growth, differentiation, apoptosis and metabolism (Ivell and Einspanier, 2002; Lee et al., 2004; Schwabe and Bullesbach, 1994). However, the molecular mechanisms of the relaxin effects remain obscure. Second messengers, such as cyclic nucleotides, nitric oxide, phosphoinositides and calcium, are assumed to be involved in relaxin action (Bani, 1997; Nguyen et al., 2003; Nistri and Bani, 2003).

The study of molecular mechanisms involved in the regulatory effects of peptides of the insulin superfamily carried out in our laboratory for many years led to the discovery of the adenylyl cyclase signaling mechanism (ACSM) of insulin action in muscle of vertebrates and invertebrates that hitherto has been unknown (Pertseva et al., 1995, 1996, 2003; Plesneva et al., 2001). This novel mechanism of insulin action involves the following signaling chain: receptor-tyrosine kinaseGi protein (βγ-dimer)phosphatidylinositol 3-kinase (PI3K)protein kinase Cζ (PKCζ)Gs proteinadenylyl cyclase (AC)protein kinase A. The evolutionary relationship of the insulin superfamily peptides suggests the existence a similar mechanism to the action of relaxin. The involvement of the following signaling proteins: receptor-tyrosine kinase (on the basis of experiments with tyrosine kinase blocker, tyrphostin 47), Gs protein and AC in the AC-activating effect of relaxin have been shown earlier (Kuznetsova et al., 1999). The aim of the present work has been to identify: (i) the receptor of the tyrosine kinase type, using genistein, a tyrosine kinase blocker; (ii) G protein of stimulatory (Gs) and inhibitory (Gi) types, using the method of ADP-ribosylation by bacterial toxins; (iii) PI3K, using its specific inhibitor, wortmannin; (iv) PKCζ, by applying an antibody technique in the human myometrium and the rat skeletal muscles (see the preliminary report of Shpakov et al., 2004); and finally, on the basis of the data obtained, (v) to reveal similarity and difference in organization of relaxin and insulin ACSM.

2 Materials and methods

2.1 Experimental subjects

For the experiments, male rats Rattus norvegicus (weight 150–200g) were used. Pieces of human myometrium isolated at Cesarean section from the lower uterine segment of pregnant and laboring women were obtained from the clinic of Institute of Obstetrics and Gynecology, Russian Academy of Medical Sciences, and were frozen.

2.2 Chemicals and radiochemicals

Creatine phosphate, creatine phosphokinase, GTP, GTPγS, Gpp[NH]p, Tris–HCl, alumina for column chromatography, ATP, cAMP, NAD, NADP, thymidine, DTT, EDTA, imidazole were obtained from Sigma (USA). [α-32P]ATP (30Ci/mmol) was from Amersham (England).

2.3 Hormones

Human relaxin 2 was kindly provided by Dr J Wade (Howard Florey Institute, University of Melbourne, Australia). Porcine insulin was obtained from Sigma (USA).

2.4 Inhibitors

Genistein is a specific inhibitor of tyrosine kinase activity. Wortmannin is a potent, selective and irreversible inhibitor of PI3K (Yano et al., 1993). These agents were obtained from Sigma (USA).

The effects of hormones and inhibitors were studied in vitro. The agents were added to the sample for determination of AC activity. Relaxin and insulin were dissolved in 0.01N HCl and diluted with Tris–HCl (pH 7.5) immediately before use. Inhibitors were preincubated with the sample for 10min, followed by an addition of relaxin or insulin at the concentration inducing a maximal AC stimulating effect (Pertseva et al., 1995, 1996). The duration of action of relaxin or insulin was limited to 2.5min, time enough for AC effect to reach its peak (Kuznetsova et al., 1999; Pertseva et al., 1995). The samples were treated in the same manner with the medium used to dissolve the hormones, activators and blockers served as control.

2.5 Antibodies

For identification of PKCζ in relaxin signaling, specific antibodies against the mammalian (rabbit) enzyme (synthetic peptides corresponding to C-terminal region 577–592 of PKCζ) from Sigma (USA) were used.

2.6 Bacterial toxins

Pertussis (PT) and cholera toxins (ChT) were obtained from Sigma (USA). PT, an exotoxin isolated from the culture of Bordetella pertussis, has been an invaluable tool in the characterization of Gi/o proteins associated with different signaling systems. The toxin catalyzes the ADP-ribosylation of Giα-subunit (cysteine residue) leading to its inactivation and blockade of hormonal inhibition of AC activity (Milligan, 1988; Reisine, 1990). ChT A subunit isolated from Vibrio cholerae induces the ADP-ribosylation of arginine residue in Gsα-subunit. Such a modification inactivates the GTPase activity and maintains Gsα-subunit in permanently active state. As a result, ChT treatment enhances the catalytic activity of AC and decreases its response to the regulatory effects of hormones acting via Gs (Milligan, 1988; Reisine, 1990).

2.7 ADP ribosylation procedure

Samples of the membrane fraction (concentration of membrane protein was 0.95–1.0mg/mL) were incubated at 37°C for 45min with and without 10μg/mL PT or 100μg/mL ChT in 400μL of 50mM Tris–HCl (pH 7.8), 2mM MgCl2, 1mM EDTA, 10mM DTT, 0.1mM NAD, 1mM NADP, 0.1mM GTP (for PT) or Gpp[NH]p (for ChT), 1mM ATP and 10mM thymidine. The toxins had been previously activated in the presence of DTT and ATP for 15min at 37°C. After ADP-ribosylation, the suspension was diluted to a final volume of 5mL with an ice-cold 50mM Tris–HCl (pH 7.5) buffer and centrifuged at 100 000×g for 30min. Pellets were resuspended in 50mM Tris–HCl (pH 7.5) buffer and immediately used for determination of AC activity.

2.8 Membrane preparation

Sarcolemma membrane fractions were isolated from the leg skeletal muscle of the rat Rattus norvegicus (for each fraction, 4–6 rats were used) according to a modified method of Kidwai et al. (1973). To obtain a crude membrane fraction of human myometrium, tissue pieces were homogenized and centrifuged at 1000×g for 10min, and the resulting supernatant was centrifuged at 20 000×g for 20min. AC activity was measured in the pellet.

2.9 Adenylyl cyclase assay

AC (EC 4.6.1.1) activity was determined by a modified method of Salomon et al. (1974). The reaction mixture (final volume 50μL) contained 50mM Tris–HCl (pH 7.5), 5mM MgCl2, 0.1mM ATP, 1μCi [α-32P]ATP, 1mM cAMP, 20mM creatine phosphate, 0.2mg/mL creatine phosphokinase and 15–20mg of membrane protein. Incubation was carried out at 37°C for 2.5min. cAMP was determined according to White (1974) using alumina for column chromatography. The experiments were performed in triplicate at least three times.

2.10 Protein assay

The protein content was determined according to the method of Lowry et al. (1951), using bovine serum albumin as a standard.

2.11 Data analysis

The data are presented as the mean±SEM of three individual experiments. Each point represents the mean of triplicate values. Differences between control and hormone/inhibitor-treated groups were statistically assessed using ANOVA and considered significant at P<0.05.

3 Results

The first stage of the work consisted in demonstration of the ability of human relaxin 2, compared to mammalian insulin, to exercise the AC activating effect in the muscle tissues of the objects under study. Relaxin (10−8M), the same as insulin (10−8M), has an AC activating effect in the human myometrium and rat skeletal muscles (Table 1). The relaxin effect in human myometrium was stronger (+240%) than in the rat muscles (+103%) while the effect of insulin in human myometrium was lower (+72%) than in the rat muscles (+84%). The AC activating effect of relaxin and insulin in both tissues was potentiated in the presence of GTPγS (Table 1). Potentiation of relaxin effect by GTPγS (10−5M) was maximal in myometrium (394%). The data obtained are in agreement with species specificity of relaxin action (Ivell and Einspanier, 2002; Lee et al., 2004; Schwabe and Bullesbach, 1994), and with our earlier results (Kuznetsova et al., 1999).


Table 1.

The influence of relaxin (10−8 M), insulin (10−8 M) and the hormone plus GTPγS (10−5 M) in vitro on the adenylyl cyclase activity in the human myometrium and the rat skeletal muscles


Adenylyl cyclase activity (pmol cAMP/min per mg protein)
Human myometriumRat muscles
Control17.4 ± 1.3 (100)19.7 ± 1.7 (100)
GTPγS32.4 ± 2.7 (186)43.0 ± 1.9 (218)
Relaxin59.2 ± 2.8 (340)39.9 ± 2.0 (203)
Relaxin +GTPγS142.7 ± 8.9 (820) [394]98.8 ± 4.0 (502) [181]
Insulin30.0 ± 2.8 (172)36.2 ± 2.6 (184)
Insulin +GTPγS52.2 ± 4.1 (300) [44]75.1 ± 3.1 (381) [79]


The second stage was aimed at further deciphering of the AC activating mechanism of relaxin on the basis of functional organization of ACSM of insulin action discovered in previous studies (Pertseva et al., 2003; Plesneva et al., 2001).

3.1 Receptor tyrosine kinase: action of genistein

For the elucidation of the type of relaxin receptor involved in the putative ACSM of hormone action, experiments were carried out using genistein, a specific tyrosine kinase blocker. Genistein in a dose-dependent manner (1.25–10.0mM) inhibited the AC activating effect of the insulin superfamily peptides, relaxin and insulin, in the myometrium of pregnant women (Fig. 1).


Fig. 1

The influence of genistein on the AC activating action of relaxin (10−8M) and insulin (10−8M) in the presence and absence of Gpp[NH]p in the human myometrium. Vertical axis, AC stimulating effect of the hormones in percentage (AC maximal effect taken as 100%). Horizontal axis, genistein concentration (mM). AC activity in control is 17.4±1.3pmol cAMP/min per mg protein.


In the case of genistein, the AC activating effect of relaxin was decreased more than with insulin. In the presence of Gpp[NH]p, the blocking action of genistein was also evident. As a result, the potentiating effect of guanine nucleotide on the AC activating action of both peptides vanished. A higher sensitivity of relaxin AC effect to genistein was denoted by an IC50 at1.25mM (the genistein concentration for the peptide AC effect to be inhibited by 50%), which was lower than for insulin (IC50 8.4mM).

The data is in accord with our earlier findings of the inhibitory action of tyrphostin 47, another tyrosine kinase blocker, on the AC activating effect of relaxin in human myometrium (Kuznetsova et al., 1999), and of insulin and IGF-1 in rat skeletal muscles (Pertseva et al., 1996; Plesneva et al., 2001).

3.2 G proteins: action of bacterial toxins

3.2.1 Pertussis toxin (PT)

PT treatment of human myometrium and rat muscle membranes (Table 2) led to an increase of AC activity in the controls, and to a significant decrease in AC stimulating effect of relaxin and insulin, as well as with the hormones combined with GTPγS. According to the mechanism of PT action (Milligan, 1988; Reisine, 1990), the toxic effect is due to ADP-ribosylation of Gi protein, which leads to a loss of functional activity of the latter. As a result, the Gi protein inhibitory influence on the basal AC activity is abolished, and Gi protein function in hormonal signal transduction is impaired. The effect of PT is generally associated with transduction of inhibitory signals. However, as was shown in earlier studies devoted to the novel ACSM of insulin action (Pertseva et al., 2003; Plesneva et al., 2001), PT treatment led unexpectedly to a blockade of stimulatory signal transduction, probably connected with a loss of the ability of Gi protein subunits to dissociate and release βγ-dimer. The latter is capable of activating of PI3K, a downstream block of insulin-competent ACSM (Plesneva et al., 2001). An analogous effect might have occurred with the ACSM of relaxin's action. Data is available showing that the G protein βγ-dimer activates PI3K (Brock et al., 2003; Shpakov, 2002).


Table 2.

The influence of pertussis toxin on the relaxin- and insulin-stimulated AC activity in the human myometrium and the rat skeletal muscles


Adenylyl cyclase activity (pmol cAMP/min/mg protein)
Human myometrium
Rat muscles
−PT+PT−PT+PT
Control16.7 ± 1.0 (100)22.5 ± 2.1 [100]25.7 ± 2.0 (100)65.3 ± 5.2 [100]
Relaxin, 10−9 M58.6 ± 8.0 (351)29.3 ± 4.3 [130]55.3 ± 4.7 (215)74.8 ± 5.7 [115]
Relaxin, 10−9 M, plus GTPγS115.1 ± 7.9 (689)29.0 ± 2.0 [129]105.9 ± 10.1 (412)76.7 ± 7.9 [118]
Insulin, 10−8 M29.3 ± 2.2 (175)24.9 ± 1.9 [111]46.7 ± 3.9 (182)78.2 ± 6.9 [120]
Insulin, 10−8 M, plus GTPγS52.4 ± 3.5 (314)32.6 ± 2.2 [145]90.5 ± 7.9 (352)90.4 ± 8.5 [139]


3.2.2 Cholera toxin (ChT)

ChT treatment of human myometrium and rat muscle membranes led to an increase of GTPγS-stimulated AC-activity compared without ChT (Fig. 2), which has been ascribed to stabilization of the GTP-binding form of Gsα subunit, and to inhibition of its GTPase activity, both leading to a permanently activated state of Gs protein (Milligan, 1988; Reisine, 1990). Being highly activated, AC decreases its reactivity to the hormone. Indeed, in our experiments with ChT treatment, the relaxin and insulin effects on GTPγS-stimulated AC activity were practically negligible. This data is evidence for the involvement of Gs protein in relaxin and insulin AC stimulating action, confirmed by the ability of guanine nucleotide (GTPγS) to potentiate hormonal effects, known to be implemented via Gs protein (Table 1).


Fig. 2

The influence of relaxin and insulin on GTPγS-stimulated AC activity in cholera toxin-treated membranes of the human myometrium (A, B) and the rat skeletal muscles (C, D). (A, C) Without toxin; (B, D) ChT-treated membranes in the presence of GTPγS. 1, without hormone; 2, relaxin, 10−8M; 3, insulin, 10−8M. Vertical axis, AC activity in percentage as compared to the control taken as 100%. The basal AC activity: 16.7±1.0 and 25.7±2.0pmol cAMP/min per mg protein for myometrium and skeletal muscles, respectively.


Thus, the ADP-ribosylation technique (i) shows the involvement of G proteins of the inhibitory and stimulatory types in the AC stimulating action of relaxin, and (ii) confirms the involvement of both G proteins in the ACSM of insulin action discovered earlier (Pertseva et al., 2003; Plesneva et al., 2001).

3.3 Phosphatidylinositol 3-kinase: effect of wortmannin

The influence of wortmannin (10−10–10−7 M) on AC-activating effects of the peptides in human myometrium and rat muscles was a strong inhibition of the AC stimulating effect of relaxin and insulin in a dose-dependent manner (Fig. 3A,B). The specificity of wortmannin blocking action on AC effect of peptides of the insulin superfamily was tested in the experiments using β-adrenergic agonist isoproterenol. The action of adrenergic agonists is usually due to another signaling pathway involving three molecular blocks: receptors of the serpentine type, Gs protein, and AC. In our experiments, wortmannin did not inhibit the stimulatory action of isoproterenol on AC activity, therefore the data is not shown. This confirms the specificity of the blocking effect of wortmannin on ACSM of relaxin and insulin action, and provides evidence for the involvement of PI3K in the mechanism of their action on AC activity in human myometrium and rat muscles. The involvement of PI3K in ACSM of insulin action had been shown earlier (Pertseva et al., 2003; Plesneva et al., 2001).


Fig. 3

The inhibitory effect of wortmannin on relaxin and insulin AC stimulating effects in the human myometrium (A) and in the rat skeletal muscles (B). Vertical axis, percentage of the AC stimulating effect of hormones. 1. without wortmannin; 2–5, 10−10, 10−9, 10−8 and 10−7M wortmannin; respectively. Asterisk, a statistical significance of differences from the control (p<0.05). AC activity without GTPγS: 17.4±1.3 and 25.7±2.0pmol cAMP/min per mg protein for myometrium and rat skeletal muscles, respectively.


3.4 Protein kinase Cζ: influence of specific antibodies against PKCζ

In the human myometrium (Fig. 4A) and the rat skeletal muscles (Fig. 4B), antibodies against mammalian PKCζ diluted 1:10 to 1:1000 showed a dose-dependent inhibition of the AC activating effects of relaxin and insulin. In the myometrium, a combined effect of hormones plus Gpp[NH]p was also blocked with PKCζ antibodies, leading to the disappearance of the guanine nucleotide potentiating action (data not shown). The same was shown in our earlier work devoted to the insulin AC activating action in rat skeletal muscles (Pertseva et al., 2003; Plesneva et al., 2001).


Fig. 4

Relaxin and insulin AC activating effect in PKCζ antibody-treated membranes of human myometrium (A) and rat skeletal muscles (B). Vertical axis, AC activity in percentage as compared to the control taken as 100%. (A) 1, without antibodies; 2 and 3, with antibodies diluted 1:500 and 1:100, respectively. (B) 1, without antibodies; 2, 3 and 4, with antibodies diluted 1:1000, 1:100 and 1:10, respectively. Asterisk, a statistical significance of differences from the control (p<0.05). AC activity (control) in myometrium and rat skeletal muscles: 17.4±1.3 and 19.7±1.7pmol cAMP/min per mg protein, respectively.


4 Discussion

The putative AC activating mechanism of relaxin action was revealed and compared with the ACSM of insulin action described in our previous investigations (Pertseva et al., 1995, 1996, 2003; Plesneva et al., 2001). Prior to our earlier findings, involvement of the AC signaling system in regulatory effects of peptides of the insulin superfamily had been ruled out. Only scant data was available to show that relaxin and insect bombyxin are capable of inducing cAMP accumulation in some tissues.

Studying the AC activating action of human relaxin 2 in the muscle tissues of mammals and mollusks (Kuznetsova et al., 1999), we had found that the hormone acts via the following signaling chain: receptor-tyrosine kinase, Gs protein, AC. Proceeding from the evolutionary relationship of insulin and relaxin, and a novel ACSM of insulin action (reviewed by Pertseva et al., 2003), we suggested that there would be similarity in the structural-functional organization of ACSM of the two hormones.

In the present study the following data on the architecture of relaxin ACSM have been obtained.

4.1 Characteristics of the receptor type

The experiments with tyrosine kinase blocker genistein confirmed our earlier finding made with tyrphostin 47 (Kuznetsova et al., 1999; Pertseva et al., 1996), which showed inhibition of the AC activating effects of relaxin and insulin.

The inhibitory action of tyrosine kinase blockers on cAMP accumulation induced by relaxin has also been observed by other authors (Bartsch et al., 2001; Sanborn et al., 2001). This led us to suggest that the relaxin receptor, like that of insulin, is of the tyrosine kinase type. The same view has been expressed by other authors (Goldsmith et al., 2001; Hollingsworth et al., 2001). Due to discovery of the receptor of relaxin and relaxin-related peptides with serpentine organization (Hsu et al., 2000, 2002; Liu et al., 2003a,b; Scott et al., 2004), the problem now concerning the nature of relaxin receptor is apparently more complicated, and the question of the types of relaxin receptors remains open. The detection in organisms of a number of relaxin-like peptides and their receptors gives reason to suppose that among the latter there may be receptors of the tyrosine kinase and serpentine types scattered in different target-tissues. This view is in agreement with the idea that the different effects of relaxin in various tissues are mediated via several signaling pathways that vary from cell to cell depending on its type (Koos and Pillai, 2001). It may also be the case that relaxin's insulin-like effects—such as growth promoting, anti-apoptotic and metabolic changes—are elicited by a receptor of the tyrosine kinase type and via ACSM, in accord with our hypothesis, as is the case with insulin (Pertseva et al., 2003; Plesneva et al., 2003). At the same time, relaxin effects that are responsible for regulation of the reproductive system are probably mediated through G protein-coupled receptors of the serpentine type structurally related to the receptors of hormones (such as LH and FSH) belonging to the group of reproductive polypeptide hormones (Hsu, 2003; Ivell et al., 2003; Kumagai et al., 2002). To solve such an intriguing dilemma concerning the type(s) of relaxin receptor requires further analysis.

There is additionally a view that the non-receptor tyrosine kinase signaling cascade is involved in the relaxin regulatory effect on phosphodiesterase/AC activity (Bartsch et al., 2001). This offers an alternative explanation of the data on the interruption of relaxin signaling by inhibitors of tyrosine kinases.

4.2 G proteins involved in relaxin ACSM

Inhibition of relaxin signal transduction via the AC system as a result of the PT treatment of muscle membranes, seen in the present study, implies the involvement of Gi protein in this process. This is in agreement with the results of our previous study, devoted to the ACSM of insulin action (Plesneva et al., 2001). Recently direct data were obtained showing the insulin-induced insulin receptor-Gi protein complex formation (Kreuzer et al., 2004), which mediates Gi protein-positive regulation of the process of autophosphorylation of insulin receptor. Besides, data exists showing that receptors coupled with Gi protein exercise a stimulating action on effector proteins due to release of Gi protein βγ-dimer (Brock et al., 2003). This was the case with PI3K and other downstream components of mitogenic signaling (Brock et al., 2003). Moreover, activation of the receptor/Gi protein (α2-adrenoreceptor) can in some instances lead to potentiation of Gs-stimulated AC activity (Mhaouty-Kodja et al., 1997). This supports our suggestion of a downstream localization of Gs protein in the ACSM of relaxin action (Kuznetsova et al., 1999, 2001). The influence of ChT on the Gs protein-coupled AC system in our experiments (Fig. 2) favors such a supposition.

The unexpected finding that PT inhibits relaxin (and insulin, earlier data) stimulatory effect on AC observed in the present study and our interpretation of this fact is confirmed by the work of Marjamaki et al. (1997). Studying integration of different signals directed to several AC isoforms, they found that (i) ligand activation of α2-adrenoreceptor induces both positive (Gβγ) and negative (Giα) signals which act on Gsα-stimulated AC, and (ii) both signals are blocked by PT. Our present and earlier findings concerning the ACSM of insulin (Plesneva et al., 2001) and relaxin action (Kuznetsova et al., 1999) coincide with available data showing the involvement of Gi- and Gs-proteins in the regulatory action of both peptides of the insulin superfamily (Kompa et al., 2002; Kreuzer et al., 2004; Mhaouty-Kodja et al., 1997; O'Brien et al., 1987).

4.3 Phosphatidylinositol 3-kinase

The suppressing influence of wortmannin on the relaxin AC activating effect provides evidence for involvement of this enzyme in the putative ACSM of the hormone action. In the smooth muscles of the freshwater bivalve mollusk, Anodonta cygnea, the AC activating effect of relaxin is also blocked by wortmannin (Kuznetsova et al., 2001). Similar results were obtained in the case of the ACSM of insulin action (Plesneva et al., 2001). According to Nguyen et al. (2003), wortmannin exercises a dose-dependent inhibition of relaxin-induced increase in cAMP, which agrees with their data on stimulatory effect of relaxin on PI3K activity in the human monocyte cell line, TPH-1. The fact that PI3K is involved in relaxin and insulin AC signaling coincides with our finding of the involvement in it of Gi protein as a donor of βγ-subunits.

4.4 Protein kinase C zeta (PKCζ)

One of the targets of the regulatory action of insulin, and probably of relaxin mediated through PI3K, is atypical PKCζ. This was found by using specific antibodies against the mammalian (rabbit) PKCζ. The treatment of membrane fractions of the human myometrium and the rat muscles with antibodies led to interruption of relaxin signal transduction via the AC signaling system (Fig. 4). The same true of ACSM with regard to insulin action in rat skeletal muscles (Pertseva et al., 2003). The involvement of PKCζ in relaxin stimulating action on the cAMP accumulation was also demonstrated in THP-1 cells (Nguyen and Dessauer, 2005).

Combining the data obtained in the present work with that of our earlier studies (Kuznetsova et al., 1999, 2001), the AC signaling cascade involved in relaxin action in the mammalian tissues can be presented as follows: receptor (tyrosine kinase or serpentine type)Gi protein (βγ-dimer)PI3KPKCζGs proteinAC. Thus, it was shown that in the ACSM of relaxin action, at least six signal proteins are involved. In the case of insulin ACSM, there are also six functional blocks plus protein kinase A, whose involvement in ACSM of relaxin action is yet to be confirmed experimentally. The data provides evidence for the participation of protein kinase A in the production of a number of relaxin regulatory effects (Hollingsworth et al., 2001; Sanborn et al., 2001).

A comparison of the structural-functional organization of ACSM of relaxin and insulin action revealed both some similar and some different features. The difference concerns the type of receptor. As is well known, the insulin receptor belongs to receptor of tyrosine kinase types. Despite inhibitory action of tyrosine kinase blockers on relaxin AC stimulating effect, which we have already shown (Kuznetsova et al., 1999, and the present study) and other authors (Bartsch et al., 2001; Sanborn et al., 2001), the structural homology has also been revealed between some receptors of relaxin peptides and receptors of the serpentine type, which hinders identification of the nature of relaxin receptors. In the mechanism of relaxin action the possibility that the non-receptor tyrosine kinase cascade is involved cannot be excluded. This was suggested by Bartsch et al. (2001), concerning the mechanism of relaxin action on phosphodiesterase/AC activity. Thus the types of relaxin receptors remain questionable.

As far as the type of heterotrimeric G protein coupled with putative relaxin receptor is concerned, one idea is based on a structural study of insulin receptor-tyrosine kinase and receptors of the serpentine type. These were found to have regions with similar amino acid sequences responsible for interaction with Gi and Gs proteins (Okamoto et al., 1993; Shpakov, 1996). This gives grounds that the receptor of relaxin similar to that of insulin is probably coupled with Gi protein (Pertseva et al., 2003; Plesneva et al., 2001). Recently, direct data has appeared confirming the involvement of Gi2 protein in the process of autophosphorylation of the insulin receptor-tyrosine kinase (Kreuzer et al., 2004). This view is supported by the inhibitory influence of PT on both insulin and relaxin signal transduction, as shown in the present work. The participation of Gi protein in ACSM of insulin and relaxin action is in agreement with the involvement in subsequent downstream signaling of the blocks (PI3K, PKCζ, Gs protein and AC) identified by us in both hormonal signal cascades.

It follows from the above that post-receptor components of ACSM of the two related hormones are alike. According to our hypothesis that insulin-competent ACSM mediates the hormone regulatory action on the main processes in a cell (Pertseva, 2000; Pertseva et al., 2003; Plesneva et al., 2001), it can be assumed that relaxin-competent ACSM will in all probability also interfere in the regulation of cell growth, apoptosis and anabolism. This had been experimentally shown to be so for insulin (Plesneva et al., 2003). Data is available that relaxin also stimulates cell proliferation and retards apoptosis in different tissues (Lee et al., 2004).

In conclusion, we propose a novel adenylyl cyclase signaling mechanism of relaxin action, which finds confirmation in the data of Nguyen et al. (2003), Nguyen and Dessauer (2005).

Acknowledgements

We thank Dr John Wade (Howard Florey Institute, University of Melbourne, Australia) for a generous gift of human relaxin 2, and Inga Menina for linguistic assistance. This research was supported by Grant 03-04-49114 from the Russian Fund for Fundamental Studies and Grant from Russian Special Program “Fundamental Science for Medicine”.

References

Bani D. Relaxin: a pleiotropic hormone. Gen Pharmacol 1997:28:13-22
Medline   1st Citation  

Bartsch O, Bartlick, B, Ivell, R. Relaxin signalling links tyrosine phosphorylation to phosphodiesterase and adenylyl cyclase activity. Mol Hum Reprod 2001:7:799-809
Crossref   Medline   1st Citation   2nd   3rd   4th  

Brock C, Schaefer, M, Reush, HP, Czupalla, C, Michaelke, M, Spicher, K. Roles of Gβγ in membrane recruitment and activation of p110γ/p101 phosphoinositide 3-kinase γ. J Cell Biol 2003:160:89-99
Crossref   Medline   1st Citation   2nd   3rd  

Chan SJ, Nagamatsu, S, Cao, QP, Steiner, DF. Structure and evolution of insulin and insulin-like growth factors in chordates. Progress in Brain Research 1992:vol. 92:15-24
1st Citation  

Goldsmith LT, Weiss, G, Palejwala, S. Mechanism of action of relaxin human cervix. Relaxin-2000 2001:285-9
1st Citation  

Hollingsworth M, Rudkin, S, Douning, S. Myometrial relaxant action of relaxin. Relaxin-2000 2001:291-9
1st Citation   2nd  

Hsu SY. New insights into the evolution of the relaxin-LGR signaling system. Trends Endocrinol Metab 2003:14:303-9
Crossref   Medline   1st Citation  

Hsu SY, Kudo, M, Chen, T, Nakabayashi, K, Bhalla, A, van der Spek, PJ. The three subfamilies of leucine-rich repeat-containing G protein-coupled receptors (LGR): identification of LGR6 and LGR7 and the signaling mechanism for LGR7. Mol Endocrinol 2000:14:1257-71
Crossref   Medline   1st Citation   2nd  

Hsu SY, Nakabayashi, K, Nishi, S, Kumagai, J, Kudo, M, Sherwood, OD. Activation of orphan receptors by the hormone relaxin. Science 2002:295:671-4
Crossref   Medline   1st Citation   2nd  

Ivell R, Balvers, M, Pohnke, Y, Telgmann, R, Bartsch, O, Milde-Langosch, K. Immunoexpression of the relaxin receptor LGR7 in breast and uterine tissues of humans and primates. Reprod Biol Endocrinol 2003:1:114
Crossref   Medline   1st Citation  

Ivell R, Einspanier, A. Relaxin peptides are new global players. Trends Endocrinol Metab 2002:13:343-8
Crossref   Medline   1st Citation   2nd   3rd  

Kidwai AM, Radcliffe, MA, Lee, EY, Daniel, EE. Isolation and properties of skeletal muscle membranes. Biochim Biophys Acta 1973:298:593-607
Crossref   Medline   1st Citation  

Kompa AR, Samuel, CS, Summers, RJ. Inotropic responses to human gene 2 (B29) relaxin in a rat model of myocardial infarction (MI): effect of pertussis toxin. Br J Pharmacol 2002:137:710-8
Crossref   Medline   1st Citation  

Koos RD, Pillai, SB. Intersection of the relaxin and estrogen signaling pathways in the uterus. Relaxin-2000 2001:101-8
1st Citation  

Kreuzer J, Nurnberg, B, Krieger-Brauer, HI. Ligand-dependent autophosphorylation of the insulin receptor is positively regulated by Gi-proteins. Biochem J 2004:380:831-6
Crossref   Medline   1st Citation   2nd   3rd  

Kumagai J, Hsu, SY, Mtsumi, H, Roh, JS, Wade, JD, Bathagte, RA. INSL3/Leiydig insulin-like peptide activates the LGR8 receptor important in testis descent. J Biol Chem 2002:277:31283-6
Crossref   Medline   1st Citation  

Kuznetsova L, Plesneva, S, Derjabina, N, Omeljaniuk, E, Pertseva, M. On the mechanism of relaxin action: the involvements of adenylyl cyclase signalling system. Regul Pept 1999:80:33-9
Crossref   Medline   1st Citation   2nd   3rd   4th   5th   6th   7th   8th   9th   10th  

Kuznetsova L, Plesneva, S, Omeljaniuk, E, Shpakov, A, Pertseva, M. Relaxin activating effect on the adenylyl cyclase and glycogen synthase in muscles of mammals and molluscs. Relaxin-2000 2001:317-9
1st Citation   2nd   3rd  

Lee HY, Zhao, S, Fields, PA, Sherwood, OD. The extent to which relaxin promotes proliferation and inhibits apoptosis of cervical epithelial and stromal cells is greatest during late pregnancy in rats. Endocrinology 2004:
Medline   1st Citation   2nd   3rd  

Liu C, Chen, J, Sutton, S, Roland, B, Kuei, C, Farmer, N. Identification of relaxin-3/INSL7 as a ligand for GPCR142. J Biol Chem 2003:278:50765-70
Crossref   Medline   1st Citation   2nd  

Liu C, Eriste, E, Sutton, S, Chen, J, Roland, B, Kuei, C. Identification of relaxin-3/INSL7 as an endogenous ligand for the orphan G-protein coupled receptor GPCR135. J Biol Chem 2003:278:50754-64
Crossref   Medline   1st Citation   2nd  

Lowry OH, Rosenbrough, NJ, Farr, AL, Randall, RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951:195:265-75
1st Citation  

Mhaouty-Kodja S, Bouet-Alard, R, Limon-Boulez, G, Maltier, JP, Legrand, C. Molecular diversity of adenylyl cyclases in human and rat myometrium. Correlation with global adenylyl cyclase activity during mid- and term pregnancy. J Biol Chem 1997:272:31100-6
Crossref   Medline   1st Citation   2nd  

Milligan G. Techniques used in the identification and analysis of function of pertussis toxin-sensitive guanine nucleotide binding proteins. Biochem J 1988:255:1-13
Medline   1st Citation   2nd   3rd   4th  

Marjamaki A, Sato, M, Bouet-Alard, R, Yang, Q, Limon-Boulez, G, Maltier, JP. Factors determining the specificity of signal transduction by guanine nucleotide-binding protein-coupled receptors. Integration of stimulatory and inhibitory input to the effector adenylyl cyclase. J Biol Chem 1997:272:16466-73
Crossref   Medline   1st Citation  

Murray-Rust J, Macleod, AN, Blundell, TL, Wood, SP. Structure and evolution of insulins: implications for receptor binding. BioEssays 1992:14:325-31
Crossref   Medline   1st Citation  

Nistri S, Bani, D. Relaxin receptors and nitric oxide synthases: search for the missing link. Reprod Biol Endocrinol 2003:1:5
Crossref   Medline   1st Citation  

Nguyen BT, Yang, L, Sanborn, BM, Dessauer, CW. Phosphatidylinositol 3-kinase activity is required for biphasic stimulation of cyclic adenosine 3′,5′-monophosphate by relaxin. Mol Endocrinol 2003:17:1075-84
Crossref   Medline   1st Citation   2nd   3rd  

Nguyen BT, Dessauer, CW. Relaxin stimulates protein kinase Cζ translocation:requirement for cyclic adenosine 3′,5′-monophosphate production. Mol Endocrinol 2005:19:1012-23
Crossref   Medline   1st Citation   2nd  

O'Brien RM, Houslay, MD, Milligan, G, Siddle, K. The insulin receptor tyrosyl kinase phosphorylates holomeric forms of the guanine nucleotide regulatory proteins Gi and Go. FEBS Lett 1987:212:281-8
Crossref   Medline   1st Citation  

Okamoto J, Okamoto, T, Murayama, Y, Hayashi, Y, Ogata, E, Nishimoto, I. GTP-binding protein-activator sequences in the insulin receptor. FEBS Lett 1993:334:143-8
Crossref   Medline   1st Citation  

Pertseva MN. Hypothesis of the key coordinating role of adenylate cyclase signal mechanism and cAMP in regulatory effect of insulin superfamily peptides on fundamental cellular processes: cell growth, apoptosis, metabolism. J Evol Biochem Physiol 2000:36:641-53
Crossref   1st Citation  

Pertseva MN, Plesneva, SA, Kuznetsova, LA, Shpakov, AO, Derkach, KV. On the tyrosine kinase mechanism of the novel effect of insulin and insulin-like growth factor-I: stimulation of adenylyl cyclase system in muscle tissues. Biochem Pharmacol 1996:52:1867-74
Crossref   Medline   1st Citation   2nd   3rd   4th   5th  

Pertseva MN, Plesneva, SA, Shpakov, AO, Rusakov, YuI, Kuznetsova, LA. Involvement of adenylyl cyclase signalling system in the action of insulin and mollusc insulin-like peptide. Comp Biochem Physiol 1995:112:689-95
Crossref   1st Citation   2nd   3rd   4th  

Pertseva MN, Shpakov, AO, Plesneva, SA, Kuznetsova, LA. A novel view on the mechanisms of action of insulin and other insulin superfamily peptides: involvement of adenylyl cyclase signaling system. Comp Biochem Physiol 2003:134:11-36
1st Citation   2nd   3rd   4th   5th   6th   7th   8th   9th   10th   11th   12th  

Plesneva SA, Pospelova, TV, Kuznetsova, LA, Bykova, TV, Shpakov, AO, Pertseva, MN. New insulin-competent adenylate cyclase signaling system as putative mechanism of antiapoptotic effect of insulin and insulin-like growth factor I. Dokl Biochem Biophys 2003:393:334-6
Crossref   Medline   1st Citation   2nd  

Plesneva SA, Shpakov, AO, Kuznetsova, LA, Pertseva, MN. A dual role of protein kinase “C” in insulin signal transduction via adenylyl cyclase signaling system in muscle tissues of vertebrates and invertebrates. Biochem Pharmacol 2001:61:1277-91
Crossref   Medline   1st Citation   2nd   3rd   4th   5th   6th   7th   8th   9th   10th   11th   12th   13th   14th  

Reisine T. Pertussis toxin in the analysis of receptor mechanisms. Biochem Pharmacol 1990:39:1499-504
Crossref   Medline   1st Citation   2nd   3rd   4th  

Salomon Y, Londos, C, Rodbell, M. A highly sensitive adenylate cyclase assay. Anal Biochem 1974:58:541-8
Crossref   Medline   1st Citation  

Sanborn BM, Dodge, KL, Yue, C, Ku, CY. Relaxin and scaffolding proteins in signaling crasstalk. Relaxin-2000 2001:279-83
1st Citation   2nd   3rd  

Schwabe C, Bullesbach, EE. Relaxin: structures, functions, promises, and nonevolution. FASEB J 1994:8:1152-60
Medline   1st Citation   2nd  

Scott DJ, Layfield, S, Riesewijk, A, Morita, H, Tregear, GW, Bathgate, RA. Identification and characterization of the mouse and rat relaxin receptors as the novel orthologues of human leucine-rich repeat-containing G-protein-coupled receptor-7. Clin Exp Pharmacol Physiol 2004:31:828-32
Crossref   Medline   1st Citation   2nd  

Shpakov AO. The homology of the primary structure of the third cytoplasmic domains in rhodopsin type receptors and of the cytoplasmic tail in the beta-subunit of the insulin receptor. Tsitologia 1996:38:1179-90
1st Citation  

Shpakov AO. The role of βγ-dimers of GTP-binding proteins in the process of hormone signal transduction. J Evol Biochem Physiol 2002:38:512-29
1st Citation  

Shpakov AO. . Shpakov AO, Pertseva MN, Kuznetsova LA, Plesneva SA. A novel, adenylyl cyclase, signaling mechanism (ACSM) of relaxin H2 action. Relaxin-2004. Wyoming, USA; 2004. p. 62.
1st Citation  

Yano H, Nakanishi, S, Kimura, K, Hanai, N, Saitoh, Y, Fukui, Y. Inhibition of histamine secretion by wortmannin through the blockage of phosphoinositol 3-kinase in RBL-2H3 cells. J Biol Chem 1993:268:25846-56
Medline   1st Citation  

White AA. Separation and purification of cyclic nucleotides by alumina column chromatography. Methods Enzymol 1974:380:41-6
1st Citation  


Received 20 March 2005/18 November 2005; accepted 30 December 2005

doi:10.1016/j.cellbi.2005.12.015


ISSN Print: 1065-6995
ISSN Electronic: 1095-8355
Published by Portland Press Limited on behalf of the International Federation for Cell Biology (IFCB)