HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Asahi, M. Articles by MacLennan, D. H. Search for Related Content PubMed PubMed Citation Articles by Asahi, M. Articles by MacLennan, D. H. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 46, 32855-32862, November 12, 1999

Transmembrane Helix M6 in Sarco(endo)plasmic Reticulum Ca²⁺-ATPase Forms a Functional Interaction Site with Phospholamban
EVIDENCE FOR PHYSICAL INTERACTIONS AT OTHER SITES^*

Michio Asahi^§, Yoshihiro Kimura^¶, Kazimierz Kurzydlowski, Michihiko Tada, and David H. MacLennan^**

From the  Banting and Best Department of Medical Research, University of Toronto, Toronto, Ontario M5G 1L6, Canada, the ^¶ Department of Pharmacology, Yamaguchi University School of Medicine, 1-1-1 Minami-Kogushi, Ube, Yamaguchi 755-8505, Japan, and the  Department of Medicine and Pathophysiology, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565-0891, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In an earlier study (Kimura, Y., Kurzydlowski, K., Tada, M., and MacLennan, D. H. (1997) J. Biol. Chem. 272, 15061-15064), mutation of amino acids on one face of the phospholamban (PLN) transmembrane helix led to loss of PLN inhibition of sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA) molecules. This helical face was proposed to form a site of PLN interaction with a transmembrane helix in SERCA molecules. To determine whether predicted transmembrane helices M4, M5, M6, or M8 in SERCA1a interact with PLN, SERCA1a mutants were co-expressed with wild-type PLN and effects on Ca²⁺ dependence of Ca²⁺ transport were measured. Wild-type inhibitory interactions shifted apparent Ca²⁺ affinity of SERCA1a by an average of 0.34 pCa units, but four of the seven mutations in M4 led to a more inhibitory shift in apparent Ca²⁺ affinity, averaging 0.53 pCa units. Seven mutations in M5 led to an average shift of 0.32 pCa units and seven mutations in M8 led to an average shift of 0.30 pCa units. Among 11 mutations in M6, 1, Q791A, increased the inhibitory shift (0.59 pCa units) and 5, V795A (0.11), L802A (0.07), L802V (0.04), T805A (0.11), and F809A (0.12), reduced the inhibitory shift, consistent with the view that Val⁷⁹⁵, Leu⁸⁰², Thr⁸⁰⁵, and Phe⁸⁰⁹, located on one face of a predicted M6 helix, form a site in SERCA1a for interaction with PLN. Those mutations in M4, M6, or M8 of SERCA1a that enhanced PLN inhibitory function did not enhance PLN physical association with SERCA1a, but mutants V795A and L802A in M6, which decreased PLN inhibitory function, decreased physical association, as measured by co-immunoprecipitation. In related studies, those PLN mutants that gained inhibitory function also increased levels of co-immunoprecipitation of wild-type SERCA1a and those that lost inhibitory function also reduced association, correlating functional interaction sites with physical interaction sites. Thus, both functional and physical data confirm that PLN interacts with M6 SERCA1a.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Sarco(endo)plasmic reticulum Ca²⁺-ATPases (SERCAs)¹ are 110-kDa transmembrane proteins that transport Ca²⁺ ions from the sarcoplasm to the lumen of the membrane system at the expense of ATP hydrolysis (1). Three isoforms, SERCA1, SERCA2, and SERCA3, share 75-85% identity in overall amino acid sequences, but have different tissue-specific expression patterns (2). They have virtually identical hydropathy profiles and nearly 100% identity in key sequences, including the sequences of several transmembrane helices, suggesting that they share similar structures. The structure of the rabbit fast-twitch skeletal muscle Ca²⁺-ATPase, SERCA1a, and structure/function relationships within SERCA1a have been studied most intensively (3). A structural model for SERCA1a at 8 Å illustrates that the transmembrane domain contains 10 transmembrane helices, M1-M10 (4). Four of these predicted helices, M4, M5, M6, and M8, were proposed to form two Ca²⁺ binding sites (5), but subsequent analysis has led to the proposal that the two Ca²⁺ binding sites lie side-by-side within a right-handed, coiled-coil structure formed by helices M4, M5, and M6 (1, 6). A three-membered, right-handed coiled-coil has been resolved at 8 Å among the helices that make up the transmembrane domain of SERCA1a in the study of Zhang et al. (4). In a cross-sectional representation of this group of predicted helices in the transmembrane domain of SERCA1a, M5 and M6 have one surface exposed to the lipid bilayer, but M4 and M8 appear to be more completely surrounded by other helices (4).

Phospholamban (PLN) is a 52-amino acid, integral membrane protein that interacts with and, at low Ca²⁺ concentrations, reversibly inhibits the activity of the cardiac Ca²⁺-ATPase isoform, SERCA2a, by lowering its apparent affinity for Ca²⁺ (7). In its role as a regulator of the activity of the Ca²⁺ pump, PLN is a major regulator of the kinetics of cardiac contractility (8, 9). PLN has been predicted to contain three domains. Domain IA, amino acids 1-20, is largely helical (10) and contains sites of regulatory phosphorylation by protein kinase A at Ser¹⁶ and by calmodulin kinase at Thr¹⁷ (11, 12). Domain IB, amino acids 21-30, is unstructured and contains a high proportion of amidated residues. Domain II, amino acids 31-52, forms a transmembrane helix.

Our investigations of structure/function relationships in PLN, in which we have identified clusters of amino acids in which mutation causes loss of the ability of PLN to diminish the apparent Ca²⁺ affinity of SERCA2a molecules, have allowed us to conclude that there are two, and possibly three, sites through which SERCA2a and PLN interact (13-17). Reconstitution of PLN inhibition with soluble PLN domain I peptides has failed to show that domain I can affect the affinity of SERCA2a for Ca²⁺ (18-21). However, mutations in domain I can cause both loss and gain of PLN inhibitory function (13, 17). Since domain IB-domain II constructs, from which residues 1-27 were removed or replaced, retained the ability to lower the affinity of SERCA2a for Ca²⁺ (15), it seems clear that inhibitory interactions that reduce SERCA2a Ca²⁺ affinity do occur in transmembrane sequences of PLN and SERCA2a and/or in a few residues at the cytoplasmic surface of the transmembrane domains (16, 17). On the basis of these results, we have proposed that PLN and SERCA2a interact via a four-base circuit through which long range inhibitory interactions might be propagated among a series of cytoplasmic and transmembrane interaction sites (16).

Since PLN exists in both pentameric and monomeric states (22), it is an important question whether inhibitory interactions between PLN and SERCA2a occur through PLN monomers or pentamers. Alanine-scanning mutagenesis of PLN domain II (16, 23-25) has demonstrated that mutations to one face of the transmembrane helix enhance monomer formation. The large concentration of Leu and Ile residues on this "PLN interaction face" has led to the postulate that the PLN pentamer is held together by a series of "leucine zippers" (25). In earlier analyses (16, 26) inhibitory function was greatly enhanced by those mutations in the transmembrane domain of PLN that led to monomerization. We proposed that the increase in monomer concentration led to increased inhibition through mass action, which would alter the equilibrium between inhibited and non-inhibited forms of SERCA2a, and, on this basis, we proposed that the monomer is the active form. Measurement of the effect of phosphorylation on the concentration of PLN monomers also supports the view that monomers are the active inhibitory species (27).

Mutation of amino acids on the opposite face of the PLN domain II helix led to loss of the ability of PLN to inhibit SERCA2a, leading us to postulate that it formed a "SERCA interaction face" (16). In this study, we have used a similar approach to identify a PLN interaction face among the predicted transmembrane helices in SERCA1a. We have used SERCA1a instead of SERCA2a because SERCA1a and SERCA2a are equally inhibited by wild-type and mutant PLN (28), because the sequences of transmembrane helices M4, M5, M6, and M8 are very highly conserved among all three SERCA isoforms (29), because structure/function analysis of SERCA1a is highly advanced, and because a large number of useful mutants derived from earlier alanine-scanning mutagenesis of SERCA1a transmembrane helices M4, M5, M6, and M8 (30) were available for this study. The effects of specific mutations on functional interactions, together with studies of co-immunoprecipitation of wild-type and mutant PLN and SERCA1a, implicate SERCA1a transmembrane helix M6 and Val⁷⁹⁵ and Leu⁸⁰², in particular, in the functional interaction between SERCA molecules and PLN.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Enzymes for DNA manipulation were obtained from New England Biolabs and Amersham Pharmacia Biotech. ³⁵S-Labeled dATP and [⁴⁵Ca]CaCl₂ were purchased from Amersham Pharmacia Biotech. G-Sepharose and a chemiluminescence kit for measurement of co-immunoprecipitation were purchased from Pierce.

Oligonucleotide-directed Mutagenesis and Expression-- Site directed mutagenesis was carried out as described previously (31). Wild-type and mutant SERCA1a cDNAs were ligated into the EcoRI site of the pMT2 expression vector (a generous gift from Dr. R. J. Kaufman). Wild-type and mutated PLN cDNAs were ligated into the XbaI and SalI sites of the pMT2 expression vector. Plasmid DNA was purified by absorption to and elution from Qiagen tip-500 columns. SERCA1a and/or PLN cDNAs were transfected into HEK-293 cells using the calcium phosphate precipitation method (32) at a concentration of 8 µg of each cDNA/10-cm plate in standard experiments. Cells were harvested 48 h after transfection, and microsomes were prepared as described previously (33).

Ca²⁺ Transport Activity-- Microsomes were assayed for Ca²⁺ transport activity as described previously (34). Total protein was measured by the method of Bradford (35) with bovine serum albumin as the standard. Data were analyzed for statistical significance using Student's unpaired t test.

Co-immunoprecipitation-- To quantitate association between SERCA1a and PLN, microsomes at 1 mg/ml protein in 0.25 M sucrose, 10 mM Tris-HCl, pH 7.5, 20 µM CaCl₂, 3 mM 2-mercaptoethanol, 150 mM KCl were mixed with an equal volume of 40 mM HEPES-NaOH, pH 7.5, 300 mM NaCl, 2 mM EDTA, 4 mM phenylmethylsulfonyl fluoride, 1% Tween 20, vortexed for 30 s, and centrifuged in an Eppendorf 5415C centrifuge for 30 min at 16,000 × g. The supernatants were rotated with bovine serum albumin-treated G-Sepharose for 30 min and centrifuged to remove proteins bound nonspecifically to G-Sepharose. The supernatants were then mixed with a G-Sepharose/PLN monoclonal antibody 1D11 complex, specific for an epitope between PLN residues 7 and 17. The samples were rotated for 30 min and centrifuged, and pellets were washed three times with a buffer composed of 20 mM HEPES, pH 7.5, 150 mM NaCl, 1 mM EDTA, and 0.5% Tween 20. The samples were loaded on 8% polyacrylamide gels, and proteins were separated using standard SDS-PAGE protocols (36) and transferred to nitrocellulose membranes. After blocking with a skim milk suspension, the membranes were treated with anti-SERCA1a monoclonal antibody A52, specific for an epitope between residues 657 and 672 in SERCA1a (37), washed in Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween 20 (Tris-buffered saline/0.1% Tween), and treated with horseradish peroxidase-conjugated anti-mouse secondary antibody (Promega). Membranes were washed in Tris-buffered saline/0.1% Tween, and the signals were detected with an enhanced chemiluminescence kit (Pierce Super Signal) The SERCA1a-PLN complex was quantified by scanning densitometry of each lane in the exposed films using a Kodak X-Omat Processor. The experiments could not be carried out in reverse order, since A52 did not precipitate SERCA1a under the conditions outlined above. Protein expression levels in all experiments were estimated by immunoblotting using antibodies A52 and 1D11.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Functional Interactions between PLN and SERCA1a Molecules Carrying Mutations in Transmembrane Helices M4, M5, M6, and M8-- In earlier studies, we showed that mutations on one face of the PLN transmembrane helix caused loss of functional interaction with SERCA2a (16). We deduced that mutations in the face of the transmembrane helix in SERCA2a that interacted with PLN would show a complementary loss of functional interaction. To test this hypothesis, we coexpressed wild-type PLN with a series of SERCA1a molecules carrying individual mutations on the hydrophobic face of helices M4, M5, M6, or M8. In each experiment, expression was monitored by immunoblotting with monoclonal antibody A52 against SERCA1a and with monoclonal antibody 1D11 against PLN. The results indicated that all of the mutants were expressed to about the same extent as wild-type SERCA1a. However, the activities of some potentially interesting mutants such as V798A or V798S were reduced relative to wild-type, so that functional analysis could not be carried out (30).

As an assay for functional interaction, we measured the effect of each SERCA1a mutation on the Ca²⁺ dependence of Ca²⁺ transport of SERCA1a co-expressed with wild-type PLN. Many of the SERCA1a mutations were characterized in an earlier study (30). Some of these mutations in SERCA1a increased or decreased apparent Ca²⁺ affinity (Table I), but, without exception, the apparent Ca²⁺ affinity of each mutant was diminished by co-expression with wild-type PLN (Table I and Fig. 1). For mutants V300A and V790A, a higher intrinsic Ca²⁺ affinity was associated with a higher inhibition by PLN, but for mutants V769A and G808A, a higher intrinsic Ca²⁺ affinity was associated with a slightly lower inhibition by PLN. For mutants V795A, T805A, and F809A, a lower intrinsic Ca²⁺ affinity was associated with a lower inhibition by PLN. However, mutants L802A and L802V, which were least inhibited by PLN, had intrinsic Ca²⁺ affinities values not significantly different from wild-type. Thus, there was no correlation between alterations of intrinsic Ca²⁺ affinity by SERCA1a mutations and alterations in the ability of the mutant SERCA1a to interact functionally with PLN.

View this table: [in this window] [in a new window]   Table I KCa of SERCA1a expressed in HEK-293 cells in the presence and absence of PLN *, p < 0.05; **, p < 0.01 (against WT SERCA1a for PLN column; against WT SERCA1a plus WT PLN for +PLN and KCa columns).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Effects of mutations in transmembrane helices M4, M5, M6, and M8 of SERCA1a on the affinity of SERCA1a for Ca2+. For each mutant, the normal amino acid residue is defined on the left in a single-letter code, its position in the sequence is identified by a number, and the newly introduced amino acid residue is defined on the right in a single-letter code. KCa is the Ca2+ concentration at which half-maximal Ca2+ uptake rates were observed. The vertical dashed line in the middle of each graph represents the difference between KCa in the presence and absence of wild-type PLN (KCa = 0.34 pCa units). The KCa values on the abscissa are negative relative to the KCa for SERCA1a alone, since the apparent affinity of SERCA1a for Ca2+ is decreased in the presence of PLN. Data are mean ± S.E. *, p < 0.05 versus +PLN, as judged by analysis of Student's unpaired t test. A, mutations in transmembrane helix M4; B, mutations in transmembrane helix M5; C, mutations in transmembrane helix M6; D, mutations in transmembrane helix M8.

The interaction between wild-type PLN and SERCA1a reduced the apparent Ca²⁺ affinity of SERCA1a by 0.34 pCa units. Among seven mutations on the most hydrophobic face of M4, V300A, V304A, and C318A reduced apparent Ca²⁺ affinity by more than 0.34 pCa units. We refer to these mutants as enhancers of PLN inhibitory function. The mutation L321A reduced apparent Ca²⁺ affinity by less than 0.34 pCa units. We refer to this mutant as a reducer of PLN inhibitory function. Mutations V307A, L311A, and V314A reduced apparent Ca²⁺ affinity by about 0.34 pCa unit. We refer to these mutants as non-effectors of PLN inhibitory function (Fig. 1A).

None of the seven mutations carried out on the most hydrophobic surface of M5 affected PLN inhibitory function (Fig. 1B). A single mutation, C910A, in the most hydrophobic face of M8 led to enhancement of PLN inhibitory function, but six other mutations did not affect PLN inhibitory function (Fig. 1D). By contrast, 5 of 11 mutations on the most hydrophobic face of M6, V795A, L802A, L802V, T805A, and F809A, reduced PLN inhibitory function and 1, Q791A, enhanced PLN inhibitory function (Fig. 1C). These observations are consistent with the hypothesis that the hydrophobic face of M6 is the site in SERCA1a that interacts with the hydrophobic face of PLN.

Physical Evidence for Association between SERCA1a and PLN-- In order to obtain physical evidence for specific interaction sites between SERCA1a transmembrane helix M6 and the transmembrane helix of PLN, we attempted to form disulfide cross-links (6) between amino acids in M6 and PLN predicted to be the most likely sites of PLN/SERCA1a transmembrane interactions. We co-expressed mutant SERCA1a (for example, mutant L802C) with mutant PLN (for example, L31C), isolated microsomal fractions and exposed the microsomes to conditions that would promote disulfide bonds to form between SERCA1a and PLN (6). In these experiments, we explored whether cross-links might form between the following mutant proteins: L802C/L31C, L802C/N34C, L802C/F35C, L802C/I38C, L802C/L42C, L802C/I48C, L802C/V49C, L802C/L52C, T805C/N34C, T805C/F35C, T805C/I38C, F809C/L31C, F809C/N34C, and F809C/F35C. Unfortunately, we could not detect the formation of any cross-links in these hydrophobic regions of the two molecules.

We then attempted to detect physical interaction by co-immunoprecipitation. Data presented in Fig. 2 (A-C) illustrate that association of SERCA1a with PLN could be detected by co-immunoprecipitation using the PLN-specific antibody, 1D11, but association of PLN with SERCA1a could not be detected by co-immunoprecipitation using the SERCA1a-specific antibody because A52 did not immunoprecipitate SERCA1a. Data calculated from at least four replicates of experiments like those shown in Fig. 2A are presented in Table II.

View larger version (64K): [in this window] [in a new window]   Fig. 2.   Co-immunoprecipitation of wild-type and mutant SERCA1a with wild-type and mutant PLN. Microsomal fractions from HEK-293 cells expressing various combinations of SERCA1a and PLN at a final concentration of 0.5 mg/ml protein were solubilized with 0.5% Tween 20 and incubated with monoclonal anti-PLN antibody 1D11, as described under "Experimental Procedures." Washed immunoprecipitates were subjected to 8% SDS-PAGE and transferred to nitrocellulose membranes. The membranes were then treated with monoclonal anti-SERCA1a antibody A52 and horseradish peroxidase-conjugated anti-mouse secondary antibody (Promega). The chemiluminescent signals (Pierce Super Signal) were detected with a Kodak M35A X-Omat processor, and mutant signal densities were compared against appropriate wild-type signal densities. Synthesis of SERCA1a and PLN was monitored in each experiment by Western blotting of microsomal fractions using anti-SERCA1a antibody A52 and anti-PLN antibody 1D11. A, co-immunoprecipitation of different SERCA1a mutants with wild-type PLN. B, co-immunoprecipitation of wild-type and L802A mutant SERCA1a with different PLN mutants. C, co-immunoprecipitation of wild-type and L802A mutant SERCA1a with PLN single and double mutants transfected at a ratio of 8 µg of SERCA1a cDNA to 1 µg of PLN cDNA/10-cm plate. The mutations in the SERCA1a/PLN combinations shown in lanes 1-26 in B and C are identified in the lines following C. The co-transfections in A and B were carried out at a SERCA1a/PLN ratio of 1:1 (8 µg of each cDNA/plate). p, pentamer; m, monomer.

Associations between PLN and SERCA1a M6 mutants V795A, L802A, T805A, and F809A were all diminished. In particular, physical associations between PLN and SERCA1a mutants V795A and L802A, which were least able to interact functionally with PLN (Fig. 1), were reduced to 53% and 44%, respectively. The mutant Q791A, which enhanced PLN inhibitory function (Fig. 1), also had diminished ability to associate with PLN. By contrast, M4 mutants V300A, V304A, L311A, and C318A, which enhanced PLN inhibitory function, and M4 mutant L321A, which reduced PLN inhibitory function, all associated with PLN to about the same extent as wild-type SERCA1a. This was also true for M5 mutants V769A and A780V and M8 mutants M899A and V903A, which did not alter PLN inhibitory function. These data highlight M6 as the SERCA1a helix most likely to interact physically and functionally with the PLN transmembrane helix, resulting in SERCA1a inhibition.

Co-immunoprecipitation of SERCA1a with PLN Mutants-- Our earlier studies illustrated that mutations in PLN domains IA, IB, and II can lead to either gain or loss of inhibitory function (13, 16, 17). Thus, it was of interest to use co-immunoprecipitation to test for physical association between SERCA1a and a variety of PLN mutants with differing effects on inhibitory function. The results of co-immunoprecipitation are presented in Fig. 2, B and C in Table III.

View this table: [in this window] [in a new window]   Table III Effects of PLN and SERCA1a mutations on co-immunoprecipitation of SERCA1a with PLN *, p < 0.05; ND, not determined; KCa is defined in the legend to Fig. 1.

Our earlier alanine scanning mutagenesis in domain IA showed that both charged (Glu², Arg⁹, and Arg¹⁴) and hydrophobic (Val⁴, Leu⁷, and Ile¹²) amino acids are essential for functional interactions between PLN and SERCA2a (13, 17). We tested the effects of mutations E2A and V4R on the ability of PLN to immunoprecipitate SERCA1a, but other mutations were not tested, since the 1D11 epitope is composed of amino acids 7-17 in PLN domain IA (38). Association between PLN and SERCA1a was reduced to 55% of wild-type by the E2A mutation and to 34% by the V4R mutation, consistent with the view that these residues are involved in physical interactions between PLN and SERCA1a.

Three mutations in PLN domain IB result in large gains of PLN inhibitory function without an effect on PLN pentameric structure (17). In these cases, gain of inhibitory function is likely to be due to enhanced affinity between PLN and SERCA1a (16, 17). In line with this postulate, the pentameric superinhibitory mutants N27A and N30A both enhanced the association between PLN and SERCA1a over 2-fold.

PLN domain II mutations in one face of the PLN transmembrane helix result in gain of function, presumably due to an increased concentration of PLN monomer which increases SERCA1a inhibition, probably by mass action (16). The association of mutants L37A (about 90% monomeric) and I40A (100% monomeric) with SERCA1a was enhanced by 1.6- and 2.5-fold, respectively. Mutations in the other face of the PLN transmembrane helix result in loss of inhibitory function, presumably because they represent the site in PLN that interacts with SERCA1a helix M6, as described above. The association of pentameric nonfunctional mutants L31A and N34A with SERCA1a was reduced to 27% and 35% of wild-type, respectively. Mutants C36A and I45A, which had relatively little effect on PLN inhibitory function, also had relatively little effect on the association between SERCA1a and PLN.

We tested the ability of PLN double mutants to associate with SERCA1a. The double mutant N27A/N34A combined a pentameric gain of function mutation in domain IB with a pentameric loss of function mutation in domain II. The resulting protein had reduced ability to alter SERCA1a Ca²⁺ affinity (K_Ca = 0.15) and reduced association (61%) with SERCA1a. The double mutants N27A/I40A and N30A/I40A combined pentameric gain of function mutations in domain IB with a monomeric gain of function mutation in domain II. The N27A/I40A and N30A/I40A mutant proteins had increased association (240% and 280%, respectively) with SERCA1a and increased ability to diminish SERCA1a Ca²⁺ affinity (K_Ca = 1.07 and 0.74, respectively). The double mutant N34A/I40A combined a pentameric loss of function mutation in domain II with a monomeric loss of function mutation in domain II. The resulting protein had very little ability to alter SERCA1a Ca²⁺ affinity (K_Ca = 0.03), but, surprisingly, had increased physical association (229%) with SERCA1a. Thus, the I40A mutation is dominant over the N34A mutation in terms of physical interaction (Table III, part B), even though N34A is dominant over I40A in terms of functional interaction (16).

The SERCA1a mutant, L802A, had reduced association with wild-type PLN (44%). It had very low association with the PLN mutant N34A, a pentameric loss of function mutation in domain II (25%), but its association with two monomeric gain of function mutations in domain II, PLN L37A (111%) and I40A (151%), was enhanced.

If we assume that the monomer is the active species that binds to SERCA1a and that the monomer content in some PLN domain II mutations is up to 10-fold higher than the monomer content in some pentameric proteins, then it may not be appropriate to make direct comparisons of SERCA1a binding to monomeric and pentameric PLN mutants. Accordingly, we altered the protocol for some of the experiments reported in Table III (part A) that involve the highly monomeric I40A mutation. In these experiments, we used a PLN/SERCA1a cDNA ratio of 1:8 instead of the ratio of 1:1 used in normal transfections. By setting the binding of SERCA1a to PLN I40A as 1.0 under these conditions, we observed that the binding of N27A/I40A was increased over I40A alone (143%), although the binding of the comparable mutant N30A/I40A, was not enhanced over wild-type (Table III, part B). The binding of SERCA1a to N34A/I40A was not reduced when compared with I40A alone in this assay, as in the standard assay, confirming the observation that there is a synergism between these two PLN mutations that enhances physical association with SERCA1a, even though functional interaction is diminished (16).

We also tested the effect of the SERCA1a L802A mutant on association with the PLN I40A mutant. We confirmed that the SERCA1a L802A mutant binds less effectively (41%) than wild-type SERCA1a to the PLN I40A mutant. We further confirmed the synergism between the N34A and I40A PLN mutants, since the PLN N34A/I40A mutant bound more SERCA1a L802A mutant (80%) than the PLN I40A mutant alone (41%).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Functional Interactions between PLN and SERCA1a-- Numerous studies support the view that PLN forms a single transmembrane helix (16, 23-25). In earlier studies we demonstrated that one face of the PLN transmembrane helix is likely to form inhibitory interactions with a transmembrane helix in SERCA molecules. Our primary objective in this study was to obtain proof of this hypothesis by identifying the transmembrane helix in SERCA molecules with which PLN interacts. We deduced that the PLN transmembrane helix would be most likely to interact with one of the key Ca²⁺ binding helices, M4, M5, M6, or M8, which have been identified in earlier studies as forming the Ca²⁺ binding and translocation domain (5, 6). Since we knew that PLN interacts as effectively with SERCA1a as it does with SERCA2a (28), and since we had carried out alanine-scanning mutagenesis of these four helices in SERCA1a, we co-expressed wild-type PLN with mutant SERCA1a molecules and looked for reduction of PLN inhibitory function. Because the PLN transmembrane helix is hydrophobic throughout and because the four SERCA1a helices each have one face that is more hydrophobic, we concentrated our search in the hydrophobic face of each SERCA1a helix.

In support of our hypothesis, we found that several mutations in M6 led to reduction in the ability of PLN to inhibit SERCA1a function (Table I; Fig. 1). The most effective mutations involved amino acids Val⁷⁹⁵, Leu⁸⁰², Thr⁸⁰⁵, and Phe⁸⁰⁹, which, together with Val⁷⁹⁸, would form a hydrophobic face on M6 if M6 were helical throughout. The diminished Ca²⁺ transport activity with mutants V798A and V798S obviated the possibility of testing the role of Val⁷⁹⁸ in PLN interaction. We propose that this hydrophobic face of M6 forms a site of interaction with the hydrophobic face of the PLN transmembrane helix that contains the key interacting residues Leu³¹, Asn³⁴, Ile³⁸, and Leu⁴², as indicated in Fig. 3. A model of the arrangement of helices MI-MIO in the transmembrane domain of SERCA1a at a resolution of 8 Å predicts that M6 is located in a shallow groove with key residues Val⁷⁹⁵, Leu⁸⁰², Thr⁸⁰⁵, and Phe⁸⁰⁹, facing the lipid bilayer (4). This groove could provide the PLN helix with access to the interaction site with M6.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   The proposed relationship of the transmembrane helix of PLN to the transmembrane sequence of M6, in transverse section. In Ref. 16, PLN residues Leu31, Asn34, Phe35, Ile38, Leu42, Ile48, Val49, and Leu52 were shown to form a helical face on PLN in which mutation caused loss of functional interaction with SERCA2a. In this study, residues Val795, Leu802, Thr805, and Phe809 form one face of a proposed M6 helix in which mutagenesis causes loss of ability to interact with PLN. These residues are located on the opposite face of the proposed helix to those which interact with M6 (6) and bind Ca2+ (5). A break in the helical structure of M6 between Thr799 and Gly801 (39) might allow realignment of two separate N- and C-terminal helices in M6 so that the overall alignment of the hydrophobic face is similar to that predicted for a single transmembrane -helix.

Recent analysis of the structure of the M6 sequence using NMR (39) suggests that M6 might assume two different conformations in a hydrophobic environment. In dodecylphosphocholine, the N-terminal sequence between Ile⁷⁸⁸ and Thr⁷⁹⁹ was helical, but the C terminus, between Gly⁸⁰¹ and Asn⁸¹⁰ was disordered. The addition of 20% trifluoroethanol led to the formation of an additional helix between Gly⁸⁰¹ and Leu⁸⁰⁷. Thus, in both conformations, the key Ca²⁺ binding residues Thr⁷⁹⁹ and Asp⁸⁰⁰ (5) appear to lie in an unstructured region critical to the formation of a Ca²⁺-binding cavity.

Our earlier analysis of cross-linking between M4 and M6 was consistent with interaction throughout the length of two helices oriented as a right-handed, coiled-coil (6). The cross-linking was strongest in the N and C termini, but was weak in the Ca²⁺ binding strata near Asp⁸⁰⁰. The conditions under which cross-linking was carried out (presence of vanadate, absence of Ca²⁺) would drive the enzyme into the E2 conformation. Our current data, summarized in Fig. 3, are also consistent with interaction between helical sequences in M6 and PLN. Studies of the interaction of PLN with SERCA2a suggest that PLN stabilizes SERCA2a in the E2 conformation (40). Thus, both our current data and the data of Rice et al. (6) suggest that M6 is helical at both N and C termini in the E2 conformation and define both the hydrophobic helical face that interacts with PLN (this study) and the more hydrophilic helical face that interacts with M4 (6). Our strongest PLN interactions were with Val⁷⁹⁵, which would lie in the N-terminal helix and Leu⁸⁰², Thr⁸⁰⁵, and Phe⁸⁰⁹, which would lie in the C-terminal helix in the new model of M6 (39). The lack of helical structure between residues Thr⁷⁹⁹ and Gly⁸⁰¹, proposed in the new model of M6, might, fortuitously, allow realignment of these faces in N- and C-terminal helices so that M6 appears as a single helical structure in our PLN interaction and cross-linking studies.

Physical Interactions between PLN and SERCA1a-- While functional studies provided strong evidence for interactions between specific amino acids in M6 of SERCA1a and PLN, they did not prove physical interaction at this site. Evidence for a physical interaction was provided by analysis of co-immunoprecipitation. The pattern of co-immunoprecipitation was well correlated with the pattern of loss of function that we had observed in our functional studies of SERCA1a mutations. Thus, SERCA1a mutants V795A, L802A, T805A and F809A, with reduced ability to interact functionally with PLN, were also reduced in their ability to interact physically with PLN. There was no correlation, however, between gain of inhibitory function and increased co-immunoprecipitation for mutants in M4, M6, and M8, suggesting that the gain of inhibitory function for M4 and M6 mutants does not involve direct physical interaction with PLN.

Unlike other mutants in M6, the V790A and Q791A mutants enhanced PLN inhibitory function, but co-immunoprecipitation did not support enhancement of physical interactions. Unlike other mutants in M4, the L321A mutant reduced PLN inhibitory function, but had only a borderline effect on physical interaction. These mutants all lie near the proposed boundaries of predicted transmembrane helices M6 and M4 and may not exert their effects through direct interactions among transmembrane helices.

We did not uncover any evidence that M5 or M8 might interact with PLN. However, several mutations in M4 enhanced PLN inhibitory function, while one, L321A, reduced PLN inhibitory function. None of those mutations in M4 that enhanced PLN inhibitory function increased SERCA1a/PLN physical interactions significantly, although association appeared to be reduced for the L321A mutation that reduced PLN inhibitory function (Fig. 2A and Table II).

M4 and M6 are intimately associated throughout their lengths in some conformations of the enzyme (6). Therefore, it is not surprising that mutations in M4 could affect functional interactions between M6 and PLN. It is clear that interactions among many residues in both PLN and SERCA1a contribute to the stabilization of specific conformations of SERCA1a, leading to the inhibited state (39). Long range conformational changes induced by the binding of PLN to M6 might alter interhelix interactions, changing the microenvironment of both Ca²⁺ binding sites and altering Ca²⁺ affinity. Specific conformations, induced following PLN/SERCA1a interactions, might be blocked or stabilized by mutations in other helices that interact with M6. Thus, it is conceivable that mutations in helices other than M6 could increase or decrease K_Ca, even though they are not involved directly in the interaction between PLN and SERCA1a. Our data suggest that mutations that affect K_Ca through long range interactions might be identified as those that affect functional interactions without affecting physical interactions.

Mutations at Other Sites in PLN-- In previous studies (13), we showed that residues between Glu² and Ile¹⁸ in cytoplasmic domain IA play an important role in the inhibitory function of PLN. We identified a series of charged and hydrophobic residues in both PLN and SERCA2a that, when mutated, abrogate the inhibitory interaction. In this study, we examined the effect of two of these mutations, E2A and V4R, on co-immunoprecipitation. Data presented in Table III show that both of these mutations have profound effects on co-immunoprecipitation, reducing the amount of SERCA1a bound by more than 50%.

In other earlier studies, we showed that two mutations in cytoplasmic domain IB, N27A and N30A, result in gain of inhibitory PLN function (17). Both of these mutations caused more than 2-fold increases in the amount of SERCA1a co-immunoprecipitated by mutant PLN. Since these PLN mutants were pentameric, we proposed earlier that their ability to inhibit SERCA2a function was due to an increase in the affinity of the PLN/SERCA2a complex (16, 17). The co-immunoprecipitation data support this postulate.

Two other classes of PLN amino acids that enhance or diminish PLN inhibitory function are found on two separate faces of the PLN transmembrane helix. One is involved in interactions between PLN monomers leading to pentamer formation and the other is involved in interaction with SERCA molecules, leading to the inhibited SERCA/PLN complex. The correlation between monomerization and gain of inhibitory function led us to propose that the PLN monomer is the active inhibitory species and that enhancement of monomer formation leads to gain of inhibitory function through mass action. The retention of pentamer structure and loss of inhibitory function led us to propose that the other class represents the interaction site with SERCA2a.

Mutants L31A and N34A, which lost the ability to interact functionally with PLN, also showed diminished ability to interact physically with SERCA1a. Mutants L37A and I40A, which gained inhibitory function, also showed an enhanced ability to interact physically with SERCA1a. It is of interest that I40A, which is more completely monomeric than L37A, also had a greater capacity to interact physically with SERCA1a. Mutants C36A and I45A, which were unaltered in their ability to interact functionally with SERCA2a, were also unchanged in their ability to interact physically with SERCA1a. All of these studies provide an excellent correlation between functional and physical interactions, supporting the view that co-immunoprecipitation is a valid method for analysis of PLN/SERCA interactions. Validation of this assay also supports our premise that PLN/SERCA interactions occur through M6.

Double mutations provided further evidence for our identification of M6 as the site of PLN/SERCA interaction. The combination of SERCA1a mutant L802A with PLN mutant N34A led to a further reduction in the amount of SERCA1a co-immunoprecipitated with PLN. The co-immunoprecipitation of L802A with either L37A or I40A mutants was diminished in comparison with wild-type SERCA1a.

The results of co-immunoprecipitation of both single and double PLN mutants with wild-type SERCA1a led to results that were in line with some of our earlier postulates and other results that will require refinement of those postulates. Our prediction that highly inhibitory domain IB mutations, N27A or N30A, gain inhibitory function because they gain a higher affinity for SERCA is supported by the enhancement of SERCA1a precipitation by these mutants. Similarly, our postulate that loss of function domain II mutants, L31A and N34A, lose affinity for SERCA is supported by a decrease in SERCA1a precipitation by these mutants. Our prediction that highly inhibitory domain II mutants such as I40A gain inhibitory function because of monomerization is not fully supported by co-immunoprecipitation of SERCA1a. If the only effect of the I40A mutation were to increase the supply of inhibitory monomers, then its physical association with SERCA1a should not have been enhanced. In experiments where PLN/SERCA1a transfection ratios were 1:1, association was enhanced nearly 2.5-fold. This might have been due to mass action, but, when PLN/SERCA1a ratios were reduced to 1:8 to account for higher monomer concentrations, association was still high. The PLN I40A mutant had reduced affinity for the SERCA1a L802A mutant (41%), and the PLN N34A mutant had reduced affinity for the SERCA1a L802A mutant (25%). However, the combination of the PLN N34A mutation with I40A increased the association of the mutant PLN with either wild-type or L802A mutant SERCA1a. These results suggest that I40A is not only monomeric, but also has enhanced affinity for SERCA1a. This affinity appears to be increased synergistically by the presence of the N34A mutation in the same PLN molecule. The structure of the PLN I40A and N34A/I40A mutant proteins, the interaction between PLN I40A and SERCA1a and the role of N34A in facilitating this interaction will be the subject of future investigation.

These studies provide the first clear evidence for the transmembrane site in SERCA molecules which binds PLN. It is not surprising that PLN should associate with M6, one of four helices involved in Ca²⁺ binding and one of the three helices that have been proposed to form a right-handed, coiled-coil, Ca²⁺-binding structure in the SERCA transmembrane domain (1). By binding to one key helix, PLN would be in a favorable position to influence the Ca²⁺ binding properties of the four-helix structure.

	ACKNOWLEDGEMENTS

We thank Dr. Robert G. Johnson, Merck Research Laboratories, for the gift of the Anti-PLN antibody 1D11 and for his continued interest in these studies; Dr. N. Michael Green for helpful discussion of this work; Dr. R. J. Kaufman, Genetics Institute, for the gift of the pMT2 vector; Dr. R. Kopito, Stanford University, for the gift of HEK-293 cells; and Stella de Leon for oligonucleotide synthesis.

	FOOTNOTES

^* This work was supported in part by a grant (to D. H. M.) from the Heart and Stroke Foundation of Ontario and by a grant (to Y. K.) from the Ministry of Education, Science and Culture, Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Postdoctoral fellow of the Heart and Stroke Foundation of Canada.

^** To whom correspondence should be addressed: Banting and Best Dept. of Medical Research, University of Toronto, Charles H. Best Inst., 112 College St., Toronto, Ontario M5G 1L6, Canada. Tel.: 416-978-5008; Fax: 416-978-8528; E-mail: david.maclennan@utoronto.ca.

	ABBREVIATIONS

The abbreviations used are: SERCA, sarco(endo)plasmic reticulum Ca²⁺-ATPase; PLN, phospholamban; PAGE, polyacrylamide gel electrophoresis; WT, wild-type.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. N. Ha, N. J. Traaseth, R. Verardi, J. Zamoon, A. Cembran, C. B. Karim, D. D. Thomas, and G. Veglia Controlling the Inhibition of the Sarcoplasmic Ca2+-ATPase by Tuning Phospholamban Structural Dynamics J. Biol. Chem., December 21, 2007; 282(51): 37205 - 37214. [Abstract] [Full Text] [PDF]

				A. A. Werdich, F. Baudenbacher, I. Dzhura, L. H. Jeyakumar, P. J. Kannankeril, S. Fleischer, A. LeGrone, D. Milatovic, M. Aschner, A. W. Strauss, et al. Polymorphic ventricular tachycardia and abnormal Ca2+ handling in very-long-chain acyl-CoA dehydrogenase null mice Am J Physiol Heart Circ Physiol, May 1, 2007; 292(5): H2202 - H2211. [Abstract] [Full Text] [PDF]

				A. O. Gramolini, M. G. Trivieri, G. Y. Oudit, T. Kislinger, W. Li, M. M. Patel, A. Emili, E. G. Kranias, P. H. Backx, and D. H. MacLennan Cardiac-specific overexpression of sarcolipin in phospholamban null mice impairs myocyte function that is restored by phosphorylation PNAS, February 14, 2006; 103(7): 2446 - 2451. [Abstract] [Full Text] [PDF]

				K. Haghighi, F. Kolokathis, A. O. Gramolini, J. R. Waggoner, L. Pater, R. A. Lynch, G.-C. Fan, D. Tsiapras, R. R. Parekh, G. W. Dorn II, et al. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy PNAS, January 31, 2006; 103(5): 1388 - 1393. [Abstract] [Full Text] [PDF]

				Z. Chen, D. L. Stokes, and L. R. Jones Role of Leucine 31 of Phospholamban in Structural and Functional Interactions with the Ca2+ Pump of Cardiac Sarcoplasmic Reticulum J. Biol. Chem., March 18, 2005; 280(11): 10530 - 10539. [Abstract] [Full Text] [PDF]

				A. J. Lokuta, N. A. Maertz, S. V. Meethal, K. T. Potter, T. J. Kamp, H. H. Valdivia, and R. A. Haworth Increased Nitration of Sarcoplasmic Reticulum Ca2+-ATPase in Human Heart Failure Circulation, March 1, 2005; 111(8): 988 - 995. [Abstract] [Full Text] [PDF]

				A. R. Tupling, A. O. Gramolini, T. A. Duhamel, H. Kondo, M. Asahi, S. C. Tsuchiya, M. J. Borrelli, J. R. Lepock, K. Otsu, M. Hori, et al. HSP70 Binds to the Fast-twitch Skeletal Muscle Sarco(endo)plasmic Reticulum Ca2+-ATPase (SERCA1a) and Prevents Thermal Inactivation J. Biol. Chem., December 10, 2004; 279(50): 52382 - 52389. [Abstract] [Full Text] [PDF]