0013.7227/94/1341-0177$03.00/0 Endocrinology Copyright 0 1994 by The Endocrine Society

Vol. 134, No. 1 Printed in U.S.A.

Calmodulin-Binding Peptides Interfere with Melanocyte- Stimulating Hormone Receptor Activity and Stimulate Adenosine 3’,5’-Monophosphate Production in M2R Mouse Melanoma Cells*

YOAV ESHEL AND YORAM SALOMONt

Department of Hormone Research, Weizmann Institute of Science, Rehovot 76100, Israel

ABSTRACT The MSH receptor belongs to a unique class of G-protein-coupled

receptors, in which calcium ions control the binding affinity of MSH by a yet unknown mechanism. Possible involvement of a calcium- binding protein [e.g. calmodulin (CaM)] in the regulation of MSH receptor activity has been studied in the M2R mouse melanoma cell line. In this study, we tested the inhibitory effects of a group of calmodulin-binding peptides (CBPs) on MSH receptor activities in intact M2R cells and membrane preparations derived from them. We also report here on stimulatory effects of CBPs on CAMP production in M2R cells that could not be produced in other cell lines lacking MSH receptors. This group of CBPs includes synthetic peptides corn- prising the CaM-binding domains of Ca’+/CaM-dependent enzymes, cytotoxic venom peptides, and peptide hormones that have been re- ported to directly interact with CaM. The results show that CBPs, at micromolar concentrations, inhibit MSH binding and consequent aden- ylate cyclase stimulation in a specific and concentration-dependent manner, but have no effect on adenylate cyclase stimulation by pros-

taglandin E,. On the other hand, when MSH was omitted and forskolin (0.5-l uM) was added instead, CBPs had the onposite effect on CAMP production, stimulating it in’ M2R cells, but -not in other cell types tested. Thus, these peptides can be considered as antagonists of MSH receptor and partial agonists of M2R adenylate cyclase. In contrast to MSH, the stimulatory effects of CBPs were unaffected by EGTA, suggesting a Ca’+-independent action of these peptides. Using phos- nholiaid vesicles and M2R cells. we recentlv showed that CBP activitv ‘In M2R cells may include direct partition Into the lipid bilayer of the cell membrane, permitting interaction with hydrophobic lipid-inserted domains of components of the signal transducing machinery. Based on these findings, we suggest that the mechanism of action of CBPs in the M2R cells includes two major components: 1) interaction with the cell surface membrane and penetration into the lipid milieu, and 2) interaction with exposed or lipid-embedded protein epitopes intrinsi- cally associated with the MSH-receptor system, thereby affecting the MSH receptor cascade. (Endocrinology 134: 177-l&5,1994)

H ORMONES, neurotransmitters, and many drugs initiate their biological actions by interacting with receptor

molecules on respective target cells. The ligand-receptor com- plex triggers a cascade of enzymatic reactions that involve intermediary guanine nucleotide-binding proteins (G-pro- teins). G-Proteins couple their receptors to specific effector enzymes to increase/decrease the level of second messengers (i.e. CAMP, inositol trisphosphate, Ca’+, etc. ) (1).

The MSH receptor was reported by us as being unique among G-protein-regulating receptors in its requirement for extracellular calcium ions that control the binding affinity for MSH and, consequently, the activity of adenylate cyclase (AC) (2-4). We demonstrated that at physiological hormone concentrations, binding of PMSH and subsequent activation of AC in intact M2R cells and plasma membrane preparations were strictly dependent on extracellular calcium ions and involved a freely reversible mechanism for the binding and/ or dissociation of the hormone (2). This effect of calcium was

Received June 17, 1993. * Y.E.‘s work was performed in partial fulfillment of his Ph.D. thesis

work at the Feinberg Graduate School of the Weizmann Institute of Science. This work was supported in part by research grants (to Y.S.) from the Joseph Cohn Center of Biomembrane Research at the Weiz- mann Institute of Science (Rehovot, Israel) and the Minerva Foundation (Munich, Germany).

t Tillie and Charles Lubin Professor of Biochemical Endocrinology. To whom all correspondence and requests for reprints should be ad- dressed.

not only confined to the association of the receptor with PMSH, but was also seen with the highly potent aMSH analog, [Nle4,0Phe7]aMSH (5), in studies of melanocortin receptors in lacrimal tissue (6).

It was then hypothesized that the putative calcium regu- latory site(s) involved in this phenomenon may reside on the receptor itself or be related to a putative Ca’+-binding protein that interacts with the MSH receptor. A suitable candidate for this purpose is calmodulin (CaM), a ubiquitous calcium receptor that regulates a host of calcium-mediated cellular and physiological phenomena in eukaryotic cells (7). To assess this theory, the effects of several CaM-binding pep- tides (CBPs), such as melittin (8), a bee venom-derived cytotoxic peptide (9), and the synthetic 17-amino acid pep- tide, M5 (lo), comprising the C-terminal CaM-binding do- main of myosin light chain kinase (MLCK) (1 l), on MSH receptor activity were tested in M2R cells (10) and lacrimal cell membranes (6). These CBPs were found to selectively inhibit MSH binding (ICso, 0.7 and 1.7 PM, respectively) to M2R mouse melanoma cells and consequent AC stimulation, presumably by interacting with a putative CaM-like site, thus interfering with MSH receptor functions (10).

The established common denominator of these peptides is their ability to be recognized by and interact with CaM (12). This is certainly the case for those peptides that function as CaM-binding domains in Ca*+/CaM-dependent enzymes, e.g. MLCK (1 l), human erythrocyte Ca*+-ATPase (13), and

177

on July 10, 2005 endo.endojournals.orgDownloaded from

178 CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS Endo. Vol134.No 1

Bordetella pertussis AC (14). These peptides retain their CaM- binding capacity when free in solution. Other members of this group include naturally occurring peptides, such as hormones, or cytotoxic peptides empirically found and re- ported to interact with CaM (15). Some of these peptides have been tested in this study (Table 1). Interactions of these peptides (e.g. M5) with CaM were established either by direct binding measurements (16, 17) or indirectly by their ability to interfere with the stimulation of CaM-dependent activa- tion of other Ca’+/CaM-regulated enzymes (18). Assuming that the enzyme-derived CBPs are the most specific amino acid sequences selected by evolution for recognition by CaM, we thought that they may serve as potential probes of putative CaM-dependent steps in pathways of higher com- plexity (e.g. the MSH receptor system). Being aware of the limitations of this approach, we do not advocate unequivo- cally earmarking the activities of CBPs described in this or previous studies to the presence or participation of CaM in the test system described. However, we found it intriguing that this group of peptides, in modest concentrations, af- forded rather selective effects on the MSH-responsive sig- naling pathway in intact M2R mouse melanoma cells, and we wished to explore its basis.

Based on their secondary structural properties, M5 (16), the synthetic 28-amino acid peptide C28W (representing the CaM-binding domain of the human erythrocyte plasma membrane calcium ATPase) (19), and other CBPs (12) were proposed to assume a predominantly a-helical conformation (~50%) upon contact with the hydrophobic core of CaM. Similar considerations may apply with respect to the inter- actions of mastoparan (20), C2OW (a truncated C28W) (21), C28W, and M5 (22) with phospholipid bilayers. To further examine the mode of action of CBPs in a cellular system, we employed biophysical methods and showed that M5 and C28W can interact with artificial phospholipid bilayers and with the lipid component of native M2R melanoma cells and cell membranes (22). Furthermore, we showed that CBPs actually partition into lipid bilayers (partition coefficients, 103-lo4 M-‘) (22). It was further proposed that this property allows for association of CBPs with lipid-inserted domains

TABLE 1. Primary sequences of CBPs used in this study

of membrane proteins, e.g. components of the MSH receptor system. Such an interaction may influence the activity of these proteins and be crucial for the observed effects of CBPs on the cellular responses reported here and previously (6, 8, 10, 23).

In this study we tested the inhibitory effects of a group of CBPs, including a number of structural analogs of M5 and C28W (Table l), on MSH binding and activity in M2R cell monolayers and membrane preparations derived from them to gain information concerning the structural requirements of their relevant site(s) of interaction. In addition, we describe here a novel independent effect of CBPs that led to enhanced CAMP production in M2R cells, but not in other cells that do not contain MSH receptors. The results suggest that, on the one hand, characteristic inhibition by CBPs of MSH binding in M2R cells is not confined to a specific primary sequence of CBPs, as reported for CBP-CaM interaction (12). On the other hand, analogs of M5 and C28W exhibit different inhibition efficacy, which emphasizes the relevance of spe- cific residues (e.g. Trp) in the peptide primary sequence. These phenomena will be interpreted in relation to the mechanism by which CBPs appear to affect MSH receptor activity. A model integrating the information gained in this and previous studies concerning the MSH receptor system is presented.

Materials and Methods

Materials

Chemicals. [‘*51]Iodine (carrier free) was supplied by Amersham (Ayles- bury, United Kingdom). [2-3H]Adenine (9.77 Ci/mmol) was obtained from Kamag (Dimona, Israel). Forskolin was purchased from Calbiochem Corp. (La Jolla, CA). BSA, soybean trypsin inhibitor, leupeptin, dithio- threitol, para-aminobenzamidine, prostaglandin Ei (PGEi), EGTA, iso- butylmethylxanthine, and isoproterenol were purchased from Sigma Chemical Co. (St. Louis, MO). All other reagents were of analytical grade. Buffers were prepared in double glass-distilled water.

Peptides. Melittin, mastoparan, porcine BMSH, and the synthetic analog of (YMSH, [Nle4,0Phe7]cuMSH, were obtained from Sigma. A synthetic analog of ACTH-(1-17) [Synchrodyn-(l-17)] was kindly provided by Dr. W. von Rechenberg (Hoechst AG, Frankfurt, Germany). A synthetic 20-amino acid peptide representing part of the calmodulin-binding

Designation Peptide sequence Group

1) Melittin G-I-G-A-V-L-K-V-L-T-T-G-L-P-A-L-I-S-W-I-K-R-K-R-Q-Q-~* Cytotoxic venom-de- 2) Mastoparan I-N-L-K-A-L-A-A-L-A-K-K-I-L-NH2 rived peptides CBPs derived from MLCK

3) M5 K-R-R-W-K-K-N-F-I-A-V-S-A-A-N-R-F-G-NH* Synthetic peptides rep- 4) Tyr-M5 Y-K-R-R-W-K-K-N-F-I-A-V-S-A-A-N-R-F-G-NH2 resenting part of the 5) AC-Met-M5 AC-M-K-R-R-W-K-K-N-F-I-A-V-S-A-A-N-R-F-G-NH* CaM-binding do- 6) M5AA K-R-R-W-K-K-A-F-I-A-V-S-A-A-A-R-F-G-NH2 main of Ca*+/CaM-

regulated enzymes CBPs derived from Ca’+-ATPase

7 c2ow L-R-R-G-Q-I-L-W-F-R-G-L-N-R-I-Q-T-Q-I-K 8) C20A L-R-R-G-Q-I-L-A-F-R-G-L-N-R-I-Q-T-Q-I-K 9) C28W L-R-R-G-Q-I-L-W-F-R-G-L-N-R-I-Q-T-Q-I-K-V-V-N-A-F-S-S-S 10) C28Y L-R-R-G-Q-I-L-_Y-F-R-G-L-N-R-I-Q-T-Q-I-K-V-V-N-A-F-S-S-S

11) BP AC R-E-R-I-D-L-L-W-K-I-A-R-A-G-A-R-S-A-V-G 12) Synchrodyn (l-17)” ~,A-Y-S-M-E-H-F-R-W-G-K-P-V-G-K-K-K-NH-(CH~)~-NH~ Synthetic and natural 13) VIP H-S-D-A-V-F-T-D-N-Y-T-R-L-R-K-Q-M-A-V-K-K-Y-L-N-S-I-L-N-~* peptide hormones 14) Glucagon H-S-E-G-T-F-T-S-D-Y-S-K-Y-L-D-S-R-R-A-E-D-F-V-E-W-L-M-D-T 15) GnRH pG-H-W-S-Y-G-L-R-P-G-NH2

a An ACTH-(l-17) analog.

on July 10, 2005 endo.endojournals.orgDownloaded from

CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS 179

domain of the B. pertussis AC (BPAC) was a present from Dr. A. Ullmann (Pasteur Institute, Paris, France). Glucagon- was a present from Dr. M. C. Lin (NIH, Bethesda, MD). AC-Met-M5 was a gift from Dr. A. I. Wand (Fox Chase Cancer Center; Philadelphia, PA). M5AA was a kind gift from Dr. D. K. Blumenthal (University of Utah, Salt Lake City, UT). CZOW, C20A, and C28Y were provided by Drs. T. Vorherr (Hoffman LaRoche, Basel, Switzerland) and E. Carafoli (ETH, Zurich, Switzerland). GnRH and vasoactive intestinal peptide (VIP) were supplied by Drs. Doron Leibovitz and Mati Fridkin of the Weizmann Institute of Science. C28W, M5, and Tyr-M5 were synthesized and purified as previously described for M5 (22). The pure peptides were shown to be homogene- ous (95-99%) bv analvtical HPLC and amino acid analvsis. Peutide

I , I , A identity was further confirmed by amino acid sequence analysis per- formed at the core facility of the Biological Services of the Weizmann Institute.

Methods

Cells and cell membranes. Culture of M2R mouse melanoma cells: M2R mouse melanoma cells were cultured in Ham’s F-12 and Dulbecco’s Modified Eagle’s Medium (DMEM; l:l), supplemented with 10% heat- inactivated horse serum, as previously described (24).

Culture of other cell lines: Baby hamster kidney (BHK), rat osteosar- coma (ROS), Ratl, HeLa, NIH-3T3, and neuroblastoma X glioma (NG108-15) cells were cultured in DMEM (ROS were cultured in 1:l Ham’s F-12-DMEM) supplemented with 10% heat-inactivated fetal calf serum. Neuroblastoma X glioma hybrid cells were additionally supple- mented with HAT medium [final concentrations, 10m4 M hypoxanthine, 4 X low7 M aminopterin, and 1.6 X 10m5 M thymidine). BHK were additionally supplemented with penicillin (200 IU/ml) and streptomycin (100 pg/ml). Cells were maintained in a humidified 8% COZ atmosphere at 37 C and subcultured every 5-7 days.

Preparation of a plasma membrane-enriched fraction from M2R mouse melanoma cells: Plasma membrane fractions were prepared from cultured M2R cells by differential centrifugation, as previously described (24). Cell membrane fractions, essentially melanosome free, in 1 rn~ bicar- bonate buffer (pH 7.0) were divided into aliquots (0.5 ml/tube) and kept frozen in liquid nitrogen until used. - _

‘25Z iodination of MSH: i2’I iodination of BMSH was uerformed as previously described (24). I251 iodination of INle4,0Phe’lorMSH was performed by the chloramine-T method, as

Binding of pz511-NIe’.oPhe7101MSH and/or “ IL >

brane preparations: MSH binding was deter&red at 30 C in a final volume of 0.1 ml PBS containing 1 mg/ml BSA, 10 rg/ml leupeptin, 10 pg/ml soybean trypsin inhibitor, 1 rnr.4 para-aminobenzamidine, and 50-100 kg membrane protein. In binding inhibition experiments, CBPs (O-100 PM) were added to the incubation medium and kept on ice for 15 min before the binding reaction started. Binding was initiated by the addition of 1’251-Nle4,0Phe71aMSH and/or Iu51]PMSH (lo5 cpm/tube; 1 nM). The incubation was terminated after 20 n&r, when’bindmg reached eauilibrium. bv the addition of 2 ml ice-cold 1 mer/ml BSA in PBS, and tubes were placed on ice. The membrane suspen&ns were then filtered on nitrocellulose filters, presoaked in 1 mg/ml BSA-PBS, and washed twice with 2 ml of the same buffer, as described previously (24). Filters were counted in an autogamma spectrometer. Specific [‘r51]MSH binding was calculated by subtracting the values of nonspecific binding obtained in the presence of excess (1 MM) unlabeled [Nle4,0Phe7]cuMSH and/or PMSH from the total radioactivity bound. These controls comprised less than 2% of the total radioactivity added and less than 10% of the bound label. Values of specific binding obtained in the absence of the inhibitors were taken as 100% for normalizing purposes. Data points represent the mean + SD of triplicate determinations. Experiments were repeated three to five times independently.

rz5Z](3MSH binding to M2R cell monolayers: Binding of [?]iodo-PMSH to M2R cell monolayers, grown in 24-well culture plates, was determined as previously described (24). In brief, [iZ51]flMSH (l-3 X lo5 cpm; l- 1000 nM), in the absence or presence of CBPs (O-100 PM), in F12:DMEM solution containing 0.1 mg/ml BSA was added (0.2 ml/well) to M2R cell monolayers (2-3 x 10s cells/well) and incubated at 37 C. Binding was terminated after 30 min at equilibrium by aspiration of the medium, followed by two washes (1 ml each) in F-12-DMEM containing 1 mg/ ml BSA. Monolayers were then taken up in 0.5 ml 1 N NaOH and counted in an autogamma spectrometer. Specific MSH binding was calculated by subtracting nonspecific binding, determined in the pres-

ence of excess (1 PM) unlabeled PMSH, from the total radioactivity bound. These controls comprised less than 2% of the total radioactivity added and less than 10% of the bound label. Values of specific binding obtained in the absence of the inhibitors were taken as 100% for normalizing purposes. Data points represent the mean + SD of triplicate determinations. Experiments were repeated three times independently.

Measurement of CAMP production in M2R cell monolayers: Intracellular [3H]cAMP production in M2R cell monolayers (24) was measured using the [2-3H]adenine prelabeling method, described previously (25). In brief, M2R cell monolayers (2-3 X lo5 cells/well) were preincubated in 24.well plates for 2 h with 2-5 PCi [3H]adenine (1 ml culture medium/ well). The medium was changed, and the cells were incubated further in the absence or presence of agonist(s) and/or inhibitor(s) for 20 min at 37 C. The medium was then removed, and intracellular [3H]cAMP was isolated and determined as described previously (25). Intracellular production of [2-3H]cAMP was calculated as a percentage of the total [2-3H]adenine uptake per well. Data points represent the mean + SD of triplicate determinations. Experiments were repeated three to five times independently.

Results

Effect of CBPs on r251]/3MSH binding

We have previously shown that synthetic M5 inhibits [lz51] PMSH binding to M2R cell monolayers and membranes derived from them in a dose-dependent manner (10). To test the generality of these observations and the possible rela- tionship of these findings to the CaM-binding properties of these peptides, we first examined the ability of other CBPs (Table 1) to interfere with [‘251]PMSH binding to M2R cell membranes (Fig. 1A). As can be seen, the inhibitory effect observed for all CBPs tested was dose dependent and re- stricted for each tested CBP to a narrow concentration range. Specifically, C2OW inhibited PMSH binding [I&, 2.9 f 0.45 (Z&EM) PM; n = 41 in a fashion similar to M5 (ICsO, 3.8 f 0.85 PM; n = 5). Melittin was slightly less effective (IC50, 6.4 + 1 .l PM; n = 4), and mastoparan (26) was much less effective in this regard (I&, 67 + 12 PM; n = 3). In contrast, we found that GnRH, which is not known as a CBP and was used here as a nonrelevant control, was ineffective even at concentra- tions as high as 100 PM (IC5,,, >>lOO PM).

In experiments with plasma membrane preparations, the tested peptides could, in principle, interact with both cyto- plasmic and extracellular faces of the membrane. We, there- fore, examined the effect of the CBPs on [‘251]PMSH binding to M2R cell monolayers, where their access to the membrane is restricted to the extracellular face only. Nevertheless, an almost identical pattern of inhibition of [‘251]/?MSH binding was observed with respect to C2OW and M5 compared with the results obtained with cell membranes (Fig. 1B). M5 was slightly less effective (IC50, 6.2 + 0.93 PM; n = 3) than reported above. VIP, a known CBP (27), had a strong inhib- itory effect on MSH binding (IC5,,, 7.5 + 1.3 PM; n = 3) at a concentration range about lOOO-fold higher than that de- scribed for its actions as a hormone in respective target organs (28). Another synthetic CBP tested represents part of a CaM- binding domain of the BPAC (14). The inhibitory effect of this peptide (I&, 59 & 9.5 PM; n = 3) assigns it, together with mastoparan, to a second, less effective group of CBPs that inhibit the binding of MSH in the submillimolar range. GnRH, as shown above (Fig. lA), had no significant effect on MSH binding (Fig. 1B).

on July 10, 2005 endo.endojournals.orgDownloaded from

180 CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS Endo. 1994 Voll34. No 1

” 0 L E

8

0’ “““” “““’ “““’ . . ..-I 10 .7 10 -5 10 -5 10 -4 10 4

Peptide Concentration (M) FIG. 1. Dose-dependent inhibition of [‘251]PMSH binding to M2R cells and membrane preparations derived from them by CBPs. A, M2R cell membranes (50 pg/assay) were incubated on ice for 10 min with the indicated concentrations of CBPs. The reaction was initiated by the addition of [‘251]PMSH (1 nM). Binding was determined under standard binding assay conditions, as described in Materials and Methods. B, M2R cell monolayers were incubated at 37 C with [?]@MSH (1 nM) in the presence of the indicated concentrations of CBPs. Binding was determined as described in Materials and Methods (n = 3-5). 100% specific binding represents values obtained in the absence of inhibitors. Values are the mean of triplicate determinations + SD.

Inhibition of [Nle4,0Phe7]aMSH binding to M2R cell membranes by C20 W, C28 W, and M5, and their respective analogs C2OA, C28Y, Ac-Met-M5, Tyr-M5, and M5AA

Most of the CBPs contain two or three aromatic residues in their primary sequence (one of which is tryptophan; see Table 1). O’Neil et al. (29) pointed out that the tryptophanyl residues may play an important role in the mechanism of interaction between CaM and its target enzymes.

Based on these considerations, we tested whether the tryptophanyl residue in the tested CBPs also plays a signifi- cant role in the inhibition of MSH binding. We compared the abilities of the Trp-containing CBPs, representing the CaM-binding domain of the Ca2+ pump, C2OW and C28W with those of their respective analogs C20A (TrptAla) and C28Y (Trp+Tyr) to inhibit [‘251-Nle4,DPhe7]&MSH binding to M2R cell membranes. Although we found only a slight decrease in the ICso of C20A [7.5 f 1.25 (~SEM) PM; n = 31 compared to that of C2OW (3.2 f 0.56 PM; n = 3), the Trp+ Tyr substitution in C28W reduced the efficacy of this peptide more profoundly by about 8-fold (from 5.5 + 0.78 SEM PM

(n = 3) in C28W to 40 + 4.6 PM (n = 3) in C28Y, respectively; Fig. 2A]. This finding is of particular interest, because the substitution in this case is between similar aromatic residues.

Another clue concerning the structural requirements for the inhibitory effect of CBP on MSH receptor binding came from experiments with M5 analogs (Fig. 2B). The addition of

60

40 - AC-Met-MS +

MSAA

Tyr-MS

oJ 7 ““” c “““’ r “” 10 -’ 10 -” 10 .2 10 -4

Peptide Concentration (M) FIG. 2. Dose-dependent inhibition of [‘251-Nle4,DPhe7]aMSH binding to M2R cell membranes by C2OW, C28W, and M5, and their respective analogs. M2R cell membranes (50 pg/assay) were incubated on ice for 10 min with the indicated concentrations of the tested peptides (A, C2OW, C28W, C20A, and C28Y; B, M5, Tyr-MS, Ac-Met-M5, and M5AA). The reaction was initiated by the addition of [1251-Nle’,DPhe’] (uMSH (-1 nM). Binding was determined under standard binding assay conditions, as described in Materials and Methods (n = 3). 100% specific binding represents values obtained in the absence of inhibitors. Values are the mean of triplicate determinations + SD.

acetylated methionine to the N-terminus in AC-Met-M5 did not affect its inhibitory ability and may even have enhanced it slightly compared to that of M5 (ICso, 1.85 + 0.37 and 4.1 + 0.72 PM, respectively; n = 3). On the other hand, the addition of Tyr to the N-terminus of M5 (Tyr-M5) decreased the inhibitory potential of this analog by about 5-fold (IC5o, 19.6 + 2.7 PM; n = 3). Most profound were the double Asn+ Ala substitutions in positions 7 and 15, M5AA (30), in which the inhibitory effect of this analog on [‘251-Nle4,0Phe7]aMSH binding was essentially lost (Fig. 2B).

Inhibition of MSH-stimulated CAMP production in M2R mouse melanoma cell monolayers by CBPs

Gerst and Salomon (10) showed that M5 inhibits /3MSH stimulation of CAMP production in intact M2R cells. The effect of M5 on MSH stimulation appears to concur well with the ability of this CBP to inhibit MSH binding, as described above (Figs. 1 and 2) and by Gerst and Salomon (10). On the basis of those observations, we examined the modulatory activity of several CBPs (Table 1) on CAMP production in M2R mouse melanoma cells. We chose C28W and M5 as representative peptides, because other CBPs (C2OW, Ac-Met-M5, Tyr-M5, and VIP) gave similar results (data not shown). We found that C28W and M5 (10 PM each) profoundly inhibited ,f3MSH stimulation of AC in M2R cell monolayers (89% and 82% inhibition, respectively; Fig. 3A).

on July 10, 2005 endo.endojournals.orgDownloaded from

CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS 181

16.0

c2aw M5 GnRH

FIG. 3. Inhibition of MSH-stimulated CAMP production in cultured M2R cell monolayer by C28W and M5. M2R cell monolayers, prela- beled with [2-3H]adenine for 2 h, were incubated for 20 min at 37 C without (basal) or with the indicated additions. /3MSH concentration was 10 nM; all other test substances were used at 10 pM. A, Effects of C28W and M5 on OMSH stimulation. B, Effect of C28W on PGE, stimulation. [3H]cAMP production was determined as described in Materials and Methods (n = 3-6). Values are the mean of triplicate determinations + SD.

In contrast, stimulation of CAMP production in these cells by PGEl was not affected by C28W (Fig. 3B), which is in full agreement with a previous observation made with M5 (lo), suggesting that these inhibitory effects of CBPs are confined to the MSH receptor cascade. GnRH, which served as a control, did not inhibit CAMP production stimulated by PMSH in these cells (Fig. 3A). CBPs, on their own, did not stimulate [2-3H]cAMP production above basal levels to any significant extent (Fig. 3).

Stimulation of CAMP production in M2R cell monolayers by CBPs

Preliminary experiments indicated that M5 in the presence of low concentrations of forskolin (1 PM) can act as a partial agonist and stimulate CAMP production in M2R cells (Gerst, J. E., and Y. Salomon, unpublished data). This observation provided us with an additional and independent way to assess the interactions of CBPs with M2R mouse melanoma cells. As shown in Fig. 4, C28W and M5, in combination with forskolin, induced a synergistic and significant stimu- lation of CAMP levels of about 6-fold over forskolin-stimu- lated control values. Glucagon, another CBP (27), although a much weaker CBP compared to M5 and C28W, served as a negative control in this assay. It should be noted that the low forskolin concentrations used in these experiments,

BASAL Forskolin p-MSH C26W M5 GLUCAGON

Forskolin: - + + + +

FIG. 4. CAMP production induced by C28W and M5 in cultured M2R cell monolayers. M2R cell monolayers, prelabeled with [2-3H]adenine for 2 h, were incubated for 20 min at 37 C without (basal) or with the indicated additions, as described in Fig. 3, in the absence or presence of forskolin (0.5 PM). [3H]cAMP production was determined as de- scribed in Materials and Methods (n = 4). The value 39 f 2 represents the extent of CAMP production in the presence of DMSH plus forskolin. Values are the mean of triplicate determinations + SD.

32.6k3.3 +

L- -

- 7

7 Basal 8-MSH PMSH C28W C28W PGE, PGE,

Forskolin: - - + + + + + + + + EGTA : - + - + - + - + - +

FIG. 5. The effect of EGTA on CAMP production induced by C28W and /YMSH in cultured M2R cell monolayers. M2R cell monolayers, prelabeled with [2-3H]adenine for 2 h, were incubated for 20 min at 37 C without (basal) or with the indicated additions, as described in Fig. 3, in the absence or presence of forskolin (0.5 PM) and without or with EGTA (5 mM). [3H]cAMP production was determined as described in Materials and Methods (n = 3). The value 32.6 + 3.3 represents the extent of CAMP production in the presence of @MSH plus forskolin. Values are the mean of triplicate determinations f SD.

when added alone (Figs. 4 and 5) caused a very small increase in CAMP levels compared to PMSH (Fig. 3).

Gerst et al. (2) showed that the activation of the AC in M2R cells by MSH, but not by PGE1, is calcium dependent. It was of special interest to test whether the stimulatory effect of the CBPs on CAMP production is also calcium dependent. The activity of most of these CBPs in their native biological role as CaM targets is strictly dependent on the interaction of calcium ions with CaM (31). To our surprise, we found that the stimulation of CAMP production in M2R cells by C28W in the presence of forskolin was not affected by the presence of 5 mM EGTA in the incubation medium (Fig. 5).

on July 10, 2005 endo.endojournals.orgDownloaded from

182 CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS Endo. 1994 Voll34. No 1

It can clearly be seen that under the same conditions, stim- ulation of CAMP by PMSH, but not by PGEi, was almost completely abolished by the addition of EGTA. The same calcium-independent stimulatory effect observed with C28W was also demonstrated for other tested CBPs (M5, C2OW, C28Y, C28A, and VIP; data not shown). None of the tested CBPs stimulated CAMP production in M2R cells by itself, which is in agreement with the results using C28W and M5 (Fig. 3). In comparison with the stimulatory effects of M5 and C28W in the presence of forskolin in M2R cells (5- to 6- fold), we found that other CBPs stimulated CAMP production to various degrees (C28Y, l-fold; C2OW, 2- to 3-fold; VIP, 12- to 13-fold above the respective forskolin-stimulated CAMP levels; data not shown). In contrast, M5AA was totally inactive in this regard, in agreement with its low inhibitory effects on MSH binding (Fig. 2B). GnRH, as mentioned above, was ineffective in stimulating CAMP levels in M2R cells in the presence of forskolin (data not shown).

To determine whether the effects of the CBPs on CAMP production are cell specific, i.e. confined to melanoma cells that bear MSH receptors, we tested their ability to raise CAMP levels in other cell types that are not known to have MSH receptors. Results of experiments performed using six arbitrarily selected cell lines are summarized in Table 2. It can be seen that except for two cases, one in HeLa cells, in which VIP plus forskolin stimulated CAMP production, and the second in ROS cells, in which [Nle4,nPhe7]crMSH plus forskolin stimulated CAMP production, no other effect was elicited by CBPs alone or in combination with forskolin in any of the cell lines tested (Table 2). As will be discussed later, we attribute these stimulatory effects by VIP and MSH to the presence of intrinsic receptors for these hormones in the respective cell types.

Discussion

Using independent experimental approaches, we demon- strated in this study that several CBPs from natural and synthetic sources affect the activity of the MSH receptor and

TABLE 2. The CAMP response of several cells to stimulation with MSH or CBPs

Cell typea PGE, Isoproterenol MSH CBPs

(10 PM) (10 PM) (0.1 PM)* (10 PM)’

M2R + NGlO&15 + HeLa

ROS BHK - NIH-3T3 + RAT1

+ +c3,7,3,9.10.12,14,

- ~(3.9,10,13)

+ A%9,10.14)

+(W + +d -(3.6,7.9,12,13)

+ -W,9,13,14)

- - -63) + -(3)

Cultured cell monolayers were prelabeled with ]2-3H]adenine for 2 h and incubated for 20 min at 37 d with the indicated additions in the nresence of forskolin (0.5 uM). as described in Fies. 3 and 4. 13HlcAMP accumulation was determined as described in I\;iateriaLs anh k&hods (n = 3). PGE, and isoproterenol served as positive controls for [3H] CAMP accumulation.

a NG, Neuroblastoma x glioma; ROS, rat osteosarcoma; BHK, baby hamster kidney. Experiments were performed three times or more.

* pMSH or [Nle4,DPhe7]cyMSH. ’ CBP numbers are given in Table 1. ’ [Nle4,nPhe7]aMSH.

the level of CAMP in M2R mouse melanoma cells, Two major effects of CBPs were described: 1) interference with MSH actions at the binding and AC stimulation levels, and 2) intrinsic stimulatory activity of CAMP levels. All of the CAMP production studies were performed in the presence of satu- rating isobutylmethylxanthine concentrations, conditions in which the activity of CAMP phosphodiesterase, a CaM- dependent enzyme (32, 33) is largely inhibited. We, there- fore, excluded the possible involvement of CAMP phospho- diesterase in the mechanism of action of CBPs.

In spite of considerable variability in primary sequence and chain length, the structural requirements for CBPs in this system resemble certain aspects in the structure-function relationship described for their interaction with CaM (34). For instance, the higher efficacy of C2OW and C28W on M2R cells compared to that of their respective Tip-deficient analogs C20A and C28Y demonstrates similarity between the structural requirements for interaction of these peptides with CaM, on the one hand, and with the site mediating their inhibition of MSH binding, on the other. It was reported that the tryptophanyl residues might be important for the interaction between CaM and synthetic enzyme-derived CBPs (35-37), hormonal (27) and cytotoxic peptides such as mastoparan-X (38), and melittin (9). The results reported herein concur well with the above reports and emphasize the importance of the Trp residue in the interaction of CBPs (excluding VIP and mastoparan) with M2R cells. The results shown in Fig. 2B suggest that the positively charged amino acids in M5 are more important for the inhibitory effect than the hydrophobic residues. The loss of activity of M5AA emphasizes the importance of the two asparagine groups in positions 7 and 15, probably as points of attachments be- tween the peptide and its binding site(s) on M2R cells. In addition, the decrease in the activity of Tyr-M5 of about 5- fold relative to that of M5 suggests the importance of the positively charged cluster at the N-terminus modified by the Tyr addition.

The rather steep concentration-dependent inhibition curves obtained with the various CBPs, affording the full range of activity within a IO-fold concentration range, sug- gest a relatively high degree of specificity. Together, these features suggest that the putative CBP-binding site(s) that mediates the effects in the M2R cells is not likely to result from nonspecific interactions of CBPs with membrane com- ponents. Rather, we can cautiously consider its behavior to resemble that of a protein-associated site, even though we have no direct evidence of its exact molecular nature.

Although we do not presently have direct evidence to support this notion, the putative site of interaction may be the MSH receptor molecule itself. Three lines of evidence support this idea. Firstly, CPBs inhibit MSH binding to M2R cells, as shown here and previously (8, lo), and to lacrimal cell membranes that contain high levels of MSH receptor (6). Secondly, CBPs inhibit CAMP stimulation by MSH, but not by PGE1, in the same cells (10). Thirdly, CBPs demonstrate intrinsic stimulatory activity on CAMP stimulation in M2R cells, but not in a number of MSH receptor-deficient cell types.

This interaction of CBPs with the MSH receptor may take one of the following forms: 1) association of the CBP with

on July 10, 2005 endo.endojournals.orgDownloaded from

CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS 183

the MSH receptor as a ligand in the classical sense, i.e. with an external component of the receptor. In this case, our data would predict that CBPs and MSH do not compete for the same binding site due to the noncompetitive relationship of the two ligands reported previously (8, 10); and 2) interaction with lipid-embedded regions of the MSH receptor; this op- tion is based on the ability of CBPs to interact with and insert into lipid bilayers, including M2R cell membranes, as recently shown (22). This property, which appears essential for the effects of CBPs on M2R cells, may also include interaction of CBPs with portions of other membrane proteins, including components of the MSH receptor cascade within the lipid milieu, and affect their function. This novel aspect of peptide- cell membrane interactions is of general interest and will have to be explored further in the future. In addition, the possibility that CBPs affect M2R cells by changing ion perme- ability and membrane potential (22) should be considered. However, the selective effect of these peptides on the re- sponse of the cells to MSH us. PGE, makes it less likely. Considering the two possibilities mentioned above, we sug- gest that the MSH receptor molecule may contain an intrinsic CBP-binding site. Alternatively, CBPs may compete with the MSH receptor for a CaM-like protein or for other proteins, e.g. G-protein, that are essential for MSH receptor activity. The interaction of mastoparan with G-proteins and the effect on their function in a reconstituted system have been as- cribed to its amphiphilic nature and its ability to interact with a portion of the G,-subunit that is normally recognized by the receptor molecule (39-42). Consequently, in a situation of cells in culture, as described here, one can visualize how CBPs may demonstrate their inhibitory effects, at the level of MSH binding and stimulation of CAMP production, by interfering with receptor-G-protein interactions. However, the mild nature of the enzyme-derived CBPs induces the above-mentioned cellular effects without causing cell lysis, as is often the case for the cytotoxic venom peptides. The phenotypic effect of the CBP will also depend on the type of G-protein involved. For instance, CBPs may reduce the apparent affinity of the receptor for MSH by preventing the normal course of the GTPase cycle (43-45). This effect is expected to interfere with MSH binding and AC stimulation by MSH. The interaction of CBPs with G, may enhance CAMP production and be synergized by forskolin to appre- ciable levels. This MSH receptor-independent mechanism may explain the intrinsic stimulatory effect of CBPs in M2R cells and be compatible with the calcium-independent mode of this response to CBPs. At this point, it is worth mentioning that the question of whether calcium ions are involved in the inhibition of MSH binding and AC stimulation by CBPs could not be probed due to the essential role of calcium in the MSH-binding process in M2R cells (2).

Direct interaction of CBPs with the MSH receptor may still be viewed in the simple-minded, partial agonist manner, producing a low effector response. Finally, as a partial agonist acting via the MSH receptor, but at a site distinct from the MSH-binding domain, the action of the CBPs may be calcium independent. Under all circumstances, the interactions of MSH and CBP with the MSH receptor, according to our model, are mutually exclusive (Fig. 6). It should be noted that inhibition by melittin and M5 of MSH binding and

PARTIALLY ACTIVE

INACTIVE ACTIVE

A I

FIG. 6. Activation of the MSH receptor system; involvement of cal- cium ions, MSH, and CBPs: a model. The MSH receptor (R) contains three putative binding sites: 1) a calcium-binding site, 2) a MSH- binding site, and 3) a CBP-binding site. The inactive MSH receptor binds calcium and subsequently interacts with MSH to give an active comnlex (MSH recentor-MSH). Unon addition of EGTA (5 mM). the MSfi receptor-MSB complex ‘dissociates to give an inactive receptor state. Alternatively, the MSH receptor can bind a CBP molecule in a calcium-independent manner and produce a partially active state.

consequent adenylate cyclase activation was previously de- termined to be noncompetitive with respect to MSH, using plasma membrane preparations derived from M2R cells (8, 10). Determination of the exact type of inhibition afforded by the various CBPs in intact cells, as reported here, will await additional studies in which MSH concentrations will also be varied. Other circumstances, i.e. the absence of MSH receptor, different lipid compositions, and the identity of G- proteins or CaM-like proteins, may explain why CBPs had no effect on CAMP levels in any of the other cell types tested.

To the best of our knowledge, this is the first demonstra- tion of AC activation in any system by synthetic peptides of this kind. The ability to demonstrate this phenomenon in the presence of forskolin, specifically in M2R cells, may relate to another unique property of these cells. Treatment of M2R cells with MSH and forskolin, as described in this study, stimulates CAMP to phenomenally high levels (>30% con- version). As reported by us previously, nearly 50% of the total adenine nucleotide pool can be consumed (24), deplet- ing ATP levels by about the same extent (46, 47). Conse- quently, interaction with CBPs, even if they represent very poor partial agonists, eliciting no activity on their own, may be amplified beyond proportion, specifically in these cells. The positive response of HeLa cells to VIP may be ascribed to specific receptors for VIP, as reported recently (48). The CAMP response to MSH in ROS cells can be explained by the low MSH receptor content that has been found by us (data not shown). Yet, no other CBP was found to be active in these or any of the other cells tested in the presence of forskolin.

The relative effectiveness of C28W and C2OW, as reported in this study, was basically similar and within the low 2-6 PM concentration range. C2OW was about 2-fold more effi- cient than C28W when acting as an MSH inhibitor, but about 2-fold less efficient than C28W as a CAMP stimulant (data

on July 10, 2005 endo.endojournals.orgDownloaded from

184 CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS Endo. 1994 Vol134 -No 1

not shown). A similar ratio, in the same concentration range, was reported (13) for the inhibitory effects of these peptides on the CaM-independent activity of a proteolyzed Ca*+ pump preparation. Here the peptides competed for the en- dogenous CBP acceptor site of the pump where C28W was about 2-fold more effective than C2OW (I&, 3 and 7 PM,

respectively). However, when tested as CaM antagonists in the same system, the peptides acted in the nanomolar con- centration range, where C28W was 5-fold more efficient than C2OW (I&, 8 us. 40 nM, respectively). The dissociation constants for CaM-peptide complexes were also in the na- nomolar range, but varied largely between the peptides (Kd of C28W, 0.1 nM; Kd of C2OW, 50 nM) (13). Other CBPs have been described to interact with CaM in the nanomolar range (12). In avian MLCK, however, the affinity for CaM is a lot lower (49). The comparison, therefore, of the efficiency of these CBPs is complex and dependent on the type of activity studied and possibly the site and mechanism by which the peptide affords its effect, including the case of M2R cells described here. Furthermore, partition of CBPs into the lipid bilayer of the membrane, which may precede their interac- tion with membrane proteins, as proposed recently by us (22), will make it even more difficult to accurately estimate and compare such affinities.

The occurrence of CaM-like binding sites in several en- zymes has previously been suggested (50). The catalytic domain of MLCK, which is recognized and autoregulated by M5, its C-terminal peptide (1 l), can, therefore, be considered such an example. C28W, the CaM-binding domain of the CaM-regulated Ca2+ pump, has also been described to bind with a defined internal sequence of this enzyme (13, 51). Likewise, the 4-5 intradiscal loop of visual rhodopsin has been proposed to contain homology with CaM itself (52). Most recently, the involvement of intrinsic CaM-like domains in the autoregulation of plant protein kinases has been reported (53).

Based on the experimental observations described in this study and previous reports (8, 10) and considerations raised in the discussion, we propose a model for MSH receptor activation by calcium ions, MSH, and CBPs (Fig. 6). In this model, it is assumed that the MSH receptor contains separate binding sites for MSH, Ca’+, and CBP. Binding of MSH to the receptor-Ca2+ complex activates the receptor. Chelation of Ca2+ with EGTA will reverse this step (2). On the other hand, upon CBP binding in the absence of calcium, the MSH receptor becomes partially active, even in the absence of MSH. The bound CBPs, acting as weak partial agonists, may synergize with forskolin and stimulate CAMP production in M2R cells.

The recent cloning of the MSH receptor (54-56) will enable future experiments to identify a putative CBP-binding do- main that appears to mediate the antagonistic/partial agon- istic action of the CBPs in MSH receptor-containing cells.

Acknowledgments

We wish to thank Rachel Benjamin for excellent secretarial assistance; Anat Azrad, Ester Shai, and Josepha Schmidt-Sole for excellent technical help; and to M. Zohar and Prof. Arie Moran for helpful discussions.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

References

Birnbaumer L 1990 Transduction of receptor signal into modulation of effector activitv bv G oroteins: the first 20 vears or so. FASEB J4:3178-3188 ’ ’ - Gerst JE, Sole J, Salomon Y 1987 Dual regulation of P-melanotropin receptor function and adenylate cyclase by calcium and guanosine nucleotides in the M2R melanoma cell line. Mol Pharmacol 31:81- 88 Salomon Y 1990 Melanocortin receptors: targets for control by extracellular calcium. Mol Cell Endocrinol 70:139-145 Salomon Y, Zohar M, DeJordy JO, Eshel Y, Shafir I, Leiba H, Gartv NB. Schmidt-Sole 1. Azrad A, Shai E, Denani H 1993 Signaling mechanisms co&olled by melanocorbns x melanoma, lacrimal and brain astroglial cells. Ann NY Acad Sci 680:539-588 Sawyer TK, Sanfilippo PJ, Hruby VJ, Engel MH, Heward JB, Hadley ME 1980 4-Norleucine, 7-o-phenylalanine-cY-melanocyte- stimulating hormone: a highly potent a-melanotropin with ultralong biological activity. Proc Nat1 Acad Sci USA 77:5754-5758 Leiba H, Garty NB, Schmidt-Sole J, Piterman 0, Azrad A, Salo- mon Y 1990 The melanocortin receptor in the rat lacrimal gland: a model system for the study of MSH (melanocyte stimulating hor- mone) as a potential neurotransmitter. Eur J Pharmacol 181:71-82 Manalan AS, Klee CB 1984 Calmodulin. Adv Cyclic Nucleotide Protein Phospho Res 18:227-260 Gerst JE, Salomon Y 1987 Inhibition by melittin and fluphenazine of melanotropin receptor function and adenylate cyclase in M2R cell membranes. Endocrinology 12131766-1772 . Comte M, Maulet Y, Cox IA 1983 Ca’+-dependent high-affiiitv complex formation between calmodulin and melittin. Biochem J 209:269-272 Gerst JE, Salomon Y 1988 A synthetic analog of the calmodulin- binding domain of myosin light chain kinase inhibits melanotropin receptor function and activation of adenylate cyclase. J Biol Chem 263:7073-7078 Kennelly PJ, Edelman AM, Blumenthal DK, Krebs EG 1987 Rabbit skeletal muscle myosin light chain kinase: the calmodulin- binding domain as a potential active site- directed inhibitory domain. J Biol Chem 262:11958-11963 O’Neil KT, DeGrado WF 1990 How calmodulin binds its targets: sequence independent recognition of amphiphilic cu-helices. Trends Biochem Sci 15:59-64 Enyedi A, Vorherr T, James P, McCormick DJ, Filoteo AG, Carafoli E, Penniston JT 1989 The calmodulin binding domain of the plasma membrane Ca2+ pump interacts both with calmodulin and with another part of the pump. J Biol Chem 264:12313-12321 Ladant D, Michelson S, Sarfati R, Gilles A-M, Predeleanu R, Barzu 0 1989 Characterization of the calmodulin-binding and of the catalytic domains of Bordetella pertussis adenylate cyclase. J Biol Chem 264:4015-4020 Anderson SR, Malencik DA 1986 Peptides recognizing calmodulin. In: Cheung WY (ed) Calcium Cell Function, Academic Press, New York, vol 6~1-42 Garone L, Steiner RF 1990 The interaction of calmodulin with the C-terminal M5 peptide of myosin light chain kinase. Arch Biochem Biophys 276:12-18 Ikura M, Clore GM, Gronenborn AM, Zhu G, Klee CB, Bax A 1992 Solution structure of a calmodulin-target peptide complex by multidimensional NMR. Science 256:632-638 Blumenthal DK, Charbonneau H, Edelman AM, Hinds TR, Ro- senberg GB, Storm DR, Vincenzi FF, Beavo JA, Krebs EG 1988 Synthetic peptides based on the calmodulin-binding domain of myosin light chain kinase inhibit activation of other calmodulin- dependent enzymes. Biochem Biophys Res Comm 156:860-865 Vorherr T, James P, Krebs J, Enyedi A, McCormick DJ, Penniston JT, Carafoli E 1990 Interaction of calmodulin with the calmodulin binding domain of the plasma membrane Ca2+ pump. Biochemistry 29:355-365 Higashijima T, Wakamatsu K, Takemitsu M, Fujino M, Nakajima T, Miyazawa T 1983 Conformational change of mastoparan from wasp venom on binding with phospholipid membranes. FEBS Lett 152:227-230 Brodin P, Falchetto R, Vorherr T Carafoli E 1992 Identification of two domains which mediate the binding of activating phospholipids to the plasma-membrane Ca2+ pump. Eur J Biochem 204:939-946

on July 10, 2005 endo.endojournals.orgDownloaded from

CALMODULIN-BINDING PEPTIDES AFFECT MELANOMA CELLS

22.

23.

24.

Eshel Y, Shai Y, Vorherr T, Carafoli E, Salomon Y 1993 Synthetic peptides corresponding to the calmodulin-binding domains of skel- etal muscle myosin light chain kinase and human erythrocyte Ca’+ pump interact with and permeabilize liposomes and cell membranes. Biochemistry 32:6721-6728 Eshel Y, Shai Y, Salomon Y, Calmodulin-binding peptides desta- bilize a&ial liposomes and inhibit MSH bindi& tb melanoma cells. 15th International Congress of Biochemistry, Jerusalem, Israel, 1991 (Abstract 71) Gerst IE, Sole I, Mather TP, Salomon Y 1986 Rermlation of aden- ylate iyclase by p- melano&opin in the M2R mt?anoma cell line. Mol Cell Endocrinol46:137-147

25.

26.

Salomon Y 1990 Determination of cellular responsiveness to hor- mones and neurotransmitters by the conversion of [3H]adenine to 13Hlcvclic AMP in cell monolavers, cell suspensions and tissue slices. beigods Enzymol 195:22-28. Malencik DA, Anderson SR 1983a Hiah affinitv binding of the mastoparans by calmodulin. Biochem” Biophys Res &mmun 114:50-56

27.

28.

29.

30.

Malencik DA, Anderson SR 1983b Binding of hormones and neuropeptides by calmodulin. Biochemistry 22:1995-2001 Laburthe M, Couvineau A 1988 Molecular analysis of vasoactive intestinal peptide receptors. A comparison with receptors for VIP- related peptides. Proc NY Acad Sci 527:296-313 O’Neil KT, Wolfe HRJ, Erickson-Viitanen S, DeGrado WF 1987 Fluorescence properties of calmodulin-binding peptides reflect al- pha-helical periodicity. Science 236:1454-1456 Blumenthal DK, Charbonneau H, Starovasnik MA, Klevit RE 1988 Solution mapping of calmod&n-myosin light chain kinase interactions usine svnthetic MLCK ueotides. In: Yad K, Mivazaki T ” ,

(eds) Calcium and Signal Response’. fapan Scientifyc Socieiy Press, Tokyo, pp 303-305

31.

32.

Blumenthal DK, Krebs EG 1988 Calmodulin-binding domains on target proteins. In: Cohen P, Klee CB (eds) Calmodulin. Elsevier, Amsterdam, pp 341-356 Cheung WY 1970 Cyclic 3’,5’-nucleotide phosphodiesterase: dem- onstration of an activator. Biochem Biophys Res Commun 38:533- 538

33.

34.

35

Minocherhomjee M, Selfe S, Flowers NJ, Storm DR 1987 Direct interaction between the catalytic subunit of the calmodulin-sensitive adenylate cyclase from bovine brain with ‘251-labeled wheat germ agglutinin and ?-labeled calmodulin. Biochemistry 26:4444-4448 O’Neil KT, DeGrado WF 1989 The interaction of calmodulin with fluorescent and photoreactive model peptides: evidence for a short interdomain separation. Proteins 6:284-293 Blumenthal DK, Takio K, Edelman AM, Charbonneau H, Titani K, Walsh KA, Krebs EG 1985 Identification of the calmodulin- binding domain of skeletal muscle myosin light chain kinase. Proc Nat1 Acad Sci USA 82:3187-3191

36 Buschmeier 8, Meyer HE, Mayr GW 1987 Characterization of the calmodulin-binding sites of muscle phosphofructokinase and com- parison with known calmodulin-binding domains. J Biol Chem 262:9454-9462

‘27 I , . Lukas TJ, Burgess WH, Prendergast FG, Lau W, Watterson MD

1986 Calmodulin binding domains: characterization of a phosphor- ylation and calmodulin binding site from myosin light chain kinase. Biochemistry 25:1458-1464

38. McDowell L, Sanyal G, Prendergast FG 1985 Probable role of

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49

50.

51.

52.

53.

54.

55.

56.

amphiphilicity in the binding of mastoparan to calmodulin. Bio- chemistry 24:2979-2984 Weingarten R, Ransnas L, Mueller H, Sklar LA, Bokoch GM 1990 Mastoparan interacts with the carboxyl terminus of the (Y subunit of Gi. J Biol Chem 265:11044-11049 Higashijima T, Burnier J, Ross EM 1990 Regulation of Gi and G, by mastoparan, related amphiphilic peptides, and hydrophobic amines. J Biol Chem 265:14176-14186 Higashijima T, Ross EM 1991 Mapping of the mastoparan-binding site on G proteins: cross-linking of [’ I-Ty?,Cys”]mastoparan to G,. J Biol Chem 266:12655-12661 Tomita U, Inanobe A, Kobavashi I, Takahashi K, Ui M, Katada T 1991 Direct interactions of mastoparan and compound 48/80 with GTP-binding proteins. J Biochem 109:184-189 Case1 D, Levkovitz H, Selinger Z 1977 The regulatory GTPase cvcle of turkev ervthrocvte adenvlate cvclase. 1 Cvclic Nucleotide lies 3:393-408 ’ ’ ’ ’ ’ Rodbell M 1980 The role of hormone receutors and GTP-regulatorv proteins in membrane transduction. Nat&e 284:17-22 - I Bourne HR. Sanders DA, McCormick F 1990 The GTPase super- family: a conserved s&h for diverse cell functions. Na’ture 348:125-132 Degani H, DeJordy JO, Salomon Y 1991 Stimulation of CAMP and phosphomonoester production by melanotropin in melanoma cells: ‘I’ NMR studies. Proc Nat1 Acad Sci USA 88:1506-1510

Salomon Y, DeJordy JO, Degani H 1992 Signal transduction as studied in living cells by 31P-NMR: adenylate cyclase activation and a novel phosphoethanolamine synthesizing pathway are stimulated by MSH in melanoma cells. In: Papa S, Azzi A, Tager JM (eds) Adenine Nucleotides in Cellular Energy Transfer and Signal Trans- duction. Birkhauser Verlag, Basel, pp 345-355 Raymond JR, Middleton JP, Dennis VW 1990 HeLa cells express CAMP-inhibitable sodium-dependent phosphate uptake. Am J I’hys- iol 258:F433-F437 Nunnallv MH, Blumenthal DK, Krebs EG, Stull JT 1987 Proper- ties of a*mon&lonal antibody directed to the calmodulin-binding domain of rabbit skeletal muscle myosin light chain kinase. Bio- chemistry 26:5885-5890 Jarrett HW, Madhavan R 1991 Calmodulin-binding proteins also have a calmodulin-like binding site within their structure. J Biol Chem 266:362-371 Falchetto R, Vorherr T, Brunner J, Carafoli E 1991 The plasma membrane Ca2+ pump contains a site that interacts with its calmod- ulin-binding domain. J Biol Chem 266:2930-2936 O’Tousa JE, Baehr W, Martin RL, Hirsh J, Pak WL, Applebury ML 1985 The drosophila ninaE gene encodes an opsin. Cell 40:839- 850 Roberts DM 1993 Protein kinases with calmodulin-like domains: novel targets of calcium signals in plants. Curr Opin Cell Biol5:242- 246 Chhajlani V, Wikberg JE 1992 Molecular cloning and expression of the human melanocyte stimulating hormone receptor cDNA. FEBS Lett 309:417-420 Mountjoy KG, Robbins LS, Mortrud MT, Cone RD 1992 The cloning of a family of genes that encode the melanocortin receptors. Science 257:1248-1251 Gantz I, Konda Y, Tashiro T, Shimoto Y, Miwa H, Munzert G, Watson SJ, DelValle J, Yamada T 1993 Molecular cloning of a novel melanocortin receptor. J Biol Chem 268:8246-8250

on July 10, 2005 endo.endojournals.orgDownloaded from