A distinct seven-residue trigger sequence is indispensable for proper coiled-coil formation of the human macrophage scavenger receptor oligomerization domain.

We have recently identified a distinct 13-residue sequence pattern that occurs with limited sequence variations in many two-stranded coiled coils but not in trimers, tetramers, or pentamers. This coiled-coil trigger pattern was demonstrated to be indispensable for the assembly of the oligomerization domain of the actin-bundling protein cortexillin I from Dictyostelium discoideum and the leucine zipper domain of the yeast transcriptional activator GCN4. With the aim to extend our knowledge on trigger sequences we have investigated the human macrophage scavenger receptor type A oligomerization domain as a representative of three-stranded coiled coils. We prepared a variety of recombinant N- and C-terminal deletion mutants from the full-length oligomerization domain by heterologous gene expression in Escherichia coli and assessed their ability to form trimeric coiled-coil structures by circular dichroism spectroscopy and analytical ultracentrifugation. Deletion mapping identified a distinct seven-residue sequence that was absolutely required for proper coiled-coil formation, supporting our previous results that heptad repeats alone are not sufficient for oligomerization. The finding that all fragments containing this particular sequence exhibited similar thermal stabilities indicates primarily a stabilizing function of the coiled-coil trigger. Based on sequence similarity, we suggest that functionally related sites are present in other three-stranded coiled-coil proteins.

We have recently identified a distinct 13-residue sequence pattern that occurs with limited sequence variations in many two-stranded coiled coils but not in trimers, tetramers, or pentamers. This coiled-coil trigger pattern was demonstrated to be indispensable for the assembly of the oligomerization domain of the actinbundling protein cortexillin I from Dictyostelium discoideum and the leucine zipper domain of the yeast transcriptional activator GCN4. With the aim to extend our knowledge on trigger sequences we have investigated the human macrophage scavenger receptor type A oligomerization domain as a representative of threestranded coiled coils. We prepared a variety of recombinant N-and C-terminal deletion mutants from the fulllength oligomerization domain by heterologous gene expression in Escherichia coli and assessed their ability to form trimeric coiled-coil structures by circular dichroism spectroscopy and analytical ultracentrifugation. Deletion mapping identified a distinct seven-residue sequence that was absolutely required for proper coiled-coil formation, supporting our previous results that heptad repeats alone are not sufficient for oligomerization. The finding that all fragments containing this particular sequence exhibited similar thermal stabilities indicates primarily a stabilizing function of the coiled-coil trigger. Based on sequence similarity, we suggest that functionally related sites are present in other three-stranded coiled-coil proteins.
The type I and II class A macrophage scavenger receptors (SR-AI and SR-AII) 1 are homotrimeric membrane glycoproteins that have been implicated in many macrophage-associated physiological and pathological processes including cell adhesion, binding, and endocytosis of multiple ligands, phagocytosis of apoptotic cells, host defense, and diseases like atherosclerosis, xanthoma, asbestosis, Alzheimer's disease, and other disorders of the central nervous system (for review, see Ref. 1 and references therein). SR-AI and SR-AII are the alter-natively spliced products of a single gene (2). Based on their amino acid sequences, SR-AI and SR-AII are predicted to contain six distinct structural domains (Refs. 3 and 4; Fig. 1): an N-terminal cytoplasmic domain, a single transmembrane domain, a spacer domain, an ␣-helical coiled-coil domain, a collagen-like domain, and a type-specific C-terminal domain of variable length. The C-terminal 22 amino acids of the collagenous domain, which contain five basic residues that are highly conserved among known species, are essential for both low density lipoprotein and acetyl low density lipoprotein binding (5). The ␣-helical coiled-coil domain is thought to be important for trimer formation and acid-dependent ligand dissociation (5,6).
The ␣-helical coiled coil is probably the most widespread subunit oligomerization motif found in proteins (7)(8)(9)(10). It is a type of protein structure consisting of two to five amphipathic ␣-helices that "coil" around each other in a left-handed supertwist (7)(8)(9)(10). The sequences of coiled coils are characterized by a heptad repeat of seven residues denoted a to g (see Fig. 2A) with a 3,4-hydrophobic repeat of mostly apolar amino acids at positions a and d (11,12). Interactions between the core residues a and d and its two flanking positions e and g determine the number of strands, the parallel or antiparallel orientation of ␣-helices, and the homo-or heterotypic association of subunits into a coiled coil (for review, see Refs. 7 and 8). Although the basic features contributing to the thermodynamic stability of coiled coils are well established, the mechanistic details of their assembly are not yet understood. A rather puzzling and frequently made observation is that relatively long heptadrepeat-containing polypeptide chain fragments derived from stable coiled-coil domains fail to associate into coiled-coil structures (13). This failure cannot be simply explained by instability due to the type of residues occupying the a and d positions of the heptad repeats or by electrostatic repulsion of the two chains. This raises the question of whether there exist distinct sites within heptad-repeat-containing amino acid sequences that are necessary to mediate coiled-coil formation. We have recently addressed this question in detail and reported a 13residue sequence pattern that occurs with limited sequence variations in many two-stranded coiled coils and that is absolutely required for the assembly of the Dictyostelium discoideum actin-bundling protein cortexillin I (14) and the yeast transcriptional activator GCN4 (15). According to their functional role in coiled-coil formation, we have termed these particular sequences "trigger" sites. The functional relationship between coiled-coil trigger sequences was manifested by replacing the intrinsic trigger motif of GCN4 with the related sequence from cortexillin I (15). We demonstrated that these trigger sequences represent autonomous helical folding units that, in contrast to arbitrarily chosen heptad repeats, can mediate coiled-coil formation (15). Our findings were supported by the results of Myers and Oas (16), who recently reinterpreted the folding pathway of GCN4-p1.
The aim of the present study was to extend our knowledge on trigger sequences. Therefore we have used the oligomerization domain of SR-A as a model system for three-stranded coiled coils. We prepared several recombinant polypeptide chain fragments of the three-stranded coiled-coil domain and investigated their structures and oligomeric states by circular dichroism (CD) spectroscopy and analytical ultracentrifugation. We found that one single heptad repeat is indispensable for proper assembly of the SR-A coiled-coil domain. Expression and Purification of Recombinant SR-A Polypeptide Chain Fragments-Escherichia coli JM109(DE3) host strain (Novagen) was used for all expression experiments. Purification of the 6ϫHis-tagged fusion proteins by immobilized metal affinity chromatography on Ni 2ϩ -Sepharose (Novagen) was performed under denaturing conditions as described in the manufacturer's instructions. Separation of the recombinant polypeptide chain fragments from the 6ϫHis-tagged carrier protein by thrombin cleavage was carried out as described previously (19). All fragments contain two additional residues, Gly and Ser, at their N termini that originate from the expression plasmid and are not part of the SR-A coding sequence. Recombinant proteins were analyzed in 5 mM sodium phosphate buffer (pH 7.4) supplemented with 150 mM sodium chloride. Concentrations of SR108 -272, SR205-272, SR187-272, SR180 -272, and SR173-272 were determined by tryptophan and/or tyrosine absorption in 6 M guanidine hydrochloride (20). The peptide concentrations of the other SR-A fragments were quantitated by the BCA assay (Pierce).

Construction of Expression
Gel Electrophoresis-SDS-PAGE and Tricine/SDS-PAGE (21) were performed on 12 ϫ 13-cm slab gels. Proteins were visualized by staining with Coomassie Brilliant Blue R-250. Apparent molecular masses were obtained by comparison with low molecular mass markers (Amersham Pharmacia Biotech and Sigma).
CD Spectroscopy-CD spectra were acquired on a Jasco J720 spectropolarimeter (Japan Spectroscopic Co.). Far ultraviolet spectra (200 -250 nm) were measured in a 1-mm path length quartz cell and represent averages of 10 accumulations. Spectra were normalized for concentration and path length to obtain the mean molar residue ellipticity after subtraction of the buffer contribution. The helix content was calculated according to Chen et al. (22). Temperature scans were recorded on a Cary 61 spectropolarimeter (Varian) equipped with a thermostatted 1-mm path length quartz cell. Thermal stability was determined by monitoring the change in the mean molar residue ellipticity at a fixed wavelength of 221 nm, [] 221 , as a function of temperature. A scan rate of 1°C/min was used. Data analysis was performed with the Jasco (Japan Spectroscopic Co.), LABView (National Instruments), and Sigma Plot (Jandel Scientific) software packages.
Analytical Ultracentrifugation-Sedimentation equilibrium and sedimentation velocity experiments were performed at 20°C on a Beckman Optima XL-A analytical Ultracentrifuge (Beckman Instruments) equipped with 12-mm Epon double-sector cells in an An-60 Ti rotor. The recombinant peptides were analyzed in 5 mM sodium phosphate buffer (pH 7.4) supplemented with 150 mM sodium chloride, and protein concentrations were adjusted to 0.1-1 mg/ml. Sedimentation velocity runs were performed at a rotor speed of 56,000 rpm, and sedimenting material was assayed by its absorbance at 234 or 278 nm. Sedimentation coefficients were corrected to standard conditions (water, 20°C (23)). Sedimentation equilibrium scans were carried out at 10,000 to 42,000 rpm depending on molecular mass. Molecular masses were evaluated from lnA versus r 2 plots, where A is the absorbance, and r is the distance from the rotor center (23). A partial specific volume of 0.73 ml/g was used for all calculations. Heptad repeats were assigned according to the COILS2 algorithm (39) and are shown as blocks of seven amino acid residues denoted a to g. The sequence crucial for proper coiled-coil formation is underlined. B, schematic representation of the full-length SR-A coiled coil and polypeptide chain fragments derived thereof. The coiled-coil trigger sequence is marked as a black box, and the stutter discontinuity is shown as a gray box.

The Three-stranded ␣-Helical Coiled-coil Domain of Human SR-A Contains Two Regions of Different Stabilities-Recently
Suzuki et al. (24) reported that the three-stranded coiled-coil domain of the bovine SR-A contains two parts of different stabilities: the N-terminal half retains its ability to form a stable three-stranded coiled coil, whereas the C-terminal half shows a monomeric, random structure at physiological buffer conditions. Based on the findings by Suzuki et al. (24), we selected the human SR-A as a model system for the identification of a trigger sequence in a three-stranded coiled coil. Like the bovine oligomerization domain, the human SR-A coiled coil comprises a series of 22 complete heptad repeats with a threeresidue discontinuity, termed stutter (25), approximately in the middle ( Fig. 2A). To corroborate the results of Suzuki et al. (24) for the human SR-A, we produced three recombinant polypeptide chain fragments; the full-length coiled-coil domain (SR-A108 -272) and the N-and C-terminal halves divided by the stutter heptad repeat discontinuity (SR-A108 -204 and SR-A205-272) (Fig. 2B). The homogeneity of the affinity-purified peptides was assessed by Tricine/SDS-PAGE (21). Single bands migrating with mobilities corresponding to their expected molecular masses were detected (data not shown).
Far-ultraviolet CD spectroscopy was used to test for the secondary structure of the peptides. At 5°C and a total chain concentration of 15 M, the CD spectra recorded from the full-length coiled-coil domain SR108 -272, and the N-terminal fragment SR108 -204 were characteristic for ␣-helical structures with minima near 208 and 222 nm (Fig. 3, A and C). Based on [] 222 values of Ϫ37,000 and Ϫ29,600 deg cm 2 dmol Ϫ1 , helical contents of Ͼ90 and 80 -90% were calculated for SR-A108 -272 and SR-A108 -204, respectively. In contrast, the spectrum recorded from SR-A205-272 was more difficult to interpret (Fig. 3C). Although similar to the spectra of randomly coiled proteins, the less pronounced minimum near 200 nm indicates some limited secondary structure.
The stabilities of the recombinant SR-A polypeptide chain fragments were assessed by thermal unfolding profiles recorded by CD at 221 nm (Fig. 3, B and D). For SR-A108 -272 and SR-A108 -204, the thermal unfolding profiles exhibited the sigmoid shape typical for coiled coils. At total chain concentrations of 15 M, the profiles recorded from SR-A108 -272 and  Table I). As expected from its random structure, the C-terminal fragment SR205-272 yielded no significant thermal unfolding profile (data not shown).
The oligomeric states of the three fragments were investigated by analytical ultracentrifugation. As shown in Table I, sedimentation equilibrium of SR108 -272 and SR108 -204 revealed average molecular masses that are consistent with trimeric structures. However, both peptides tended to aggregate into higher order oligomers. In contrast, SR205-272 was predominantly monomeric (Table I).
Taken together, these findings demonstrate that the fulllength (SR-A108 -272) and the N-terminal half (SR-A108 -272) of the SR-A oligomerization domain fold into stable threestranded coiled-coil structures. Because the C-terminal half (SR-A205-272) of the SR-A coiled coil is unstructured, the N-terminal part of the SR-A oligomerization domain contains a region that is indispensable for proper coiled-coil formation of SR-A.
A Distinct Seven-residue Sequence Is Necessary for Proper Coiled-coil Formation of SR-A-To further characterize this trigger site, our strategy was to successively remove heptad repeats from the N and C termini of the SR-A oligomerization domain and to assess the ability of these mutants to form coiled-coil structures by analytical ultracentrifugation and CD spectroscopy. Toward this goal, seven recombinant polypeptide chain fragments were produced by heterologous gene expression in E. coli. SR-A108 -151, SR-A108 -165, SR-A108 -172, and SR-A108 -179 were designed to consist of the first 6, 8, 9, and 10 N-terminal heptad repeats, respectively, and SR-A187-272, SR-A180 -272, and SR-A173-272 comprise the last 11, 12, and 13 C-terminal heptad repeats, respectively, of the SR-A coiled-coil domain (Fig. 2B). The homogeneity of the affinitypurified recombinant polypeptide fragments was verified by Tricine/SDS-PAGE. All peptides showed single bands with mobilities corresponding to their calculated monomer molecular masses (not shown). As the minimum length required for formation of stable coiled coils has been reported to be in the range of 21-23 residues (26 -28), all these recombinant fragments that contain a minimum of six heptad repeats were potentially long enough to fold into stable coiled-coil structures.
Deletion of more than 10 heptad repeats from the N terminus of SR-A (SR-A180 -272 and SR-A187-272) completely abol-ished coiled-coil formation. The CD spectra recorded from these polypeptide chain fragments at 5°C and total chain concentrations of 15 M were similar to randomly coiled proteins with minima near 200 nm (Fig. 3E). However, the less pronounced minimum in the spectrum of SR-A187-272 suggests some limited secondary structure. Analytical ultracentrifugation sedimentation equilibrium of SR-A187-272 yielded a molecular mass indicative of monomers, whereas SR-A180 -272 apparently formed a mixture of monomers and higher aggregates (Table I).
In contrast, deletion of nine heptad repeats from the Nterminal end of SR-A did not affect coiled-coil formation. At 5°C and a total chain concentration of 15 M, SR-A173-272 yielded a CD spectrum typical for coiled coils with an ␣-helix content of 70 -80% (Fig. 3F). The corresponding thermal unfolding profile at a chain concentration of 15 M exhibited a sigmoid shape with a T m of 47°C (Fig. 3G). Sedimentation equilibrium revealed that the SR-A173-272 fragment forms a trimeric structure (Table I).
Taken together, these findings indicate that the ninth Nterminal heptad repeat of SR-A comprises the N-terminal boundary of the coiled-coil trigger sequence. Based on this finding and to delimit the C terminus of the trigger sequence, we next focused on C-terminal heptad repeats of the SR-A oligomerization domain.
Deletion of 13 and more heptad repeats from the C terminus of SR-A (SR-A108 -151, SR-A108 -165, and SR-A108 -172) resulted in a loss of the ability of the fragments to form proper coiled-coil structures. Although all fragments formed ␣-helical aggregates, as evidenced by CD (Fig. 3H) and analytical ultracentrifugation analysis (Table I), they were not able to form specific three-stranded coiled-coil structures. In particular SR-A108 -151 and SR-A108 -165 showed ␣-helical contents of Ͻ40% (Fig. 3H), and all three recombinant polypeptide chain fragments exhibited very broad noncooperative thermal unfolding transitions (Fig. 3I).
In contrast, deletion of the 12 C-terminal heptad repeats from SR-A retained the ability of SR-A108 -179 for coiled-coil formation. The fragment had all features characteristic of a three-stranded coiled coil as revealed by its helical CD spectrum (␣-helix content of 80 -90% at a chain concentration of 15 M; Fig. 3J), its sigmoid shaped thermal CD unfolding profile (T m value of 48°C at a chain concentration of 15 M; Fig. 3K; Table I), and its analytical ultracentrifugation sedimentation velocity profile and sedimentation equilibrium boundary characteristic for an elongated trimeric molecule (Table I). Taken together, we have identified by deletion mapping a seven-residue sequence within the SR-A oligomerization domain that is indispensable for proper coiled-coil formation. Fig. 4. The characteristic feature of all the sequences is the strict conservation of the isoleucine residue at the heptad d position and the positively charged residue at position f as well as the possibility of at least one intrahelical i, iϩ3 or i, iϩ4 salt bridge between charged residues in positions b and f or c and f. In addition, the e and g positions of this particular heptad repeat are occupied by polar serine or threonine residues. Screening of known three-stranded coiled-coil proteins with the two derived sequence patterns (I/V/L)X(D/ E)IX(R/K)X and (I/V/L)(D/E)XIX(R/K)X yielded positive matches for several laminin chains, hemagglutinin, and mouse matrilin-3 (Fig. 4). However, the trigger sequence pattern could not be found in other three-stranded coiled-coil proteins. It should be noted that by screening protein sequences using our rather restrictive search patterns (zero mismatches were allowed) additional potential coiled-coil trigger sequences (even in the same heptad-repeat-containing segment) may go undetected. Furthermore, the polarity of the possible i, iϩ3 or i, iϩ4 intrahelical salt bridges was also not changed in our search patterns. Whether the related sequences in laminin chains, hemagglutinin, and matrilin-3 play a functionally equivalent role in triggering coiled-coil formation remains to be elucidated.

A Distinct Seven-residue Sequence Is Required for Proper
Assembly of the SR-A Coiled Coil-To generalize our concept on the existence of coiled-coil trigger sequences, the aim of this study was to characterize such a site within a three-stranded coiled-coil domain. We have chosen the SR-A coiled coil as a model polypeptide for the following reasons. (i) The SR-A coiled-coil domain was reported to form a parallel homotrimer that contains two regions of different stabilities; the N-terminal half forms a stable three-stranded coiled coil, whereas the C-terminal part is unstructured at physiological buffer conditions (24). This finding already suggested the existence of a trigger region within the SR-A oligomerization domain. (ii) The full-length SR-A coiled-coil domain is of intermediate length.
As the minimum length required for the formation of stable coiled coils has been reported to be in the range of 21-23 residues (26 -28), it was possible to generate polypeptide chain fragments from the full-length oligomerization domain that were potentially long enough to fold into stable coiled-coil structures. (iii) SR-A108 -272 exhibits a sigmoid unfolding profile, which suggests that the trimer represents a uniform and entire folding domain.
Using the same systematic N-and C-terminal deletion mapping strategy as described for cortexillin I (14), we have identified a seven-residue segment Ile 173 -Ser 179 that was indispensable for triggering proper coiled-coil formation of SR-A. Deletion of this specific coiled-coil trigger site abolished trimerization of up to 12-heptad-repeat-long SR-A fragments. This observation further supports our previous findings that the presence of heptad repeats is not sufficient for stable coiled-coil formation.
Characteristic Features of the SR-A Trigger Sequence-Our identified trigger sequence comprising amino acids 173 IDEISKS 179 of SR-A possesses some characteristic features that make it an ideal heptad repeat to stabilize a threestranded coiled-coil structure. (i) It contains isoleucine residues at the hydrophobic a and d (see Fig. 4) positions of the heptad repeat. Isoleucine residues are known to play a major role in the specification of the oligomerization state of coiled-coil structures. By using GCN4 leucine zipper mutants, in which all the a and d positions had been simultaneously mutated, Kim and co-workers (29) demonstrate that an even distribution of isoleucine residues at the a and d positions favors trimer formation (29). The preference for specific amino acids at the core positions of trimers can be explained in terms of the packing geometry of side chains at the a and d positions (30). Isoleucine side chains at both the a and d positions can be best accommodated in a trimer and, thus, have been used in the de novo design of trimeric coiled coils (31,32). (ii) The SR-A coiled-coil trigger sequence contains two potential favorable electrostatic intrahelical interactions between Asp 174 and Lys 178 (i, iϩ4) and Glu 175 and Lys 178 (i, iϩ3). Whereas the contribution of electrostatic interchain interactions is currently fiercely debated in the literature (33)(34)(35)(36), there is evidence accumulating that favorable intrachain i, iϩ3 and i, iϩ4 electrostatic interactions significantly contribute to coiled-coil stability (37). By introduction of interactive combinations of surface salt bridges referred to as complex salt bridges into the GCN4 leucine zipper, Spek FIG. 4. Alignment of known SR-A trigger sites and potential trigger sequences from three-stranded coiled coils. Potential trigger sequences were localized using the GCG Wisconsin sequence analysis software package. A sequence pattern search against SWISS-PROT with (I/V/L) X(D/E)IX(R/K) X and (I/V/L)(D/E)XIX(R/K)X has been applied, and zero mismatches were allowed. Mouse ␣1, ␤1, and ␥1 chains are shown as representatives for laminin subunits. In addition, the sequence patterns were found in hemagglutinin and matrilin-3. Heptad positions are indicated by lowercase letters. Conserved potential intrahelical attractive electrostatic interactions between ionizable side chains are possible between positions b and f (i, iϩ4) and/or positions c and f (i, iϩ3). Numbers refer to the amino acid (aa) positions within the native proteins. X is any residue.
FIG. 5. Fractional helicity in the SR-A coiled-coil monomer predicted by AGADIR. The predicted helix content of the SR-A coiled coil appears to be periodic. In contrast to the coiled-coil trigger sequences of cortexillin I and GCN4, only a low helix content is predicted for the SR-A trigger site (residues Ile173 to Ser179). et al. (37) observed an increase in thermal stability of 22°C of the mutant peptide relative to the wild type. Interestingly, in the x-ray structure of the full-length coiled-coil oligomerization domain of cortexillin I, the region of the trigger site can be distinguished from the other heptad repeats by an accumulation of intra-and intermolecular salt bridges. 2 Furthermore, intrahelical salt bridges appear also critical for the helicity of 16-residue peptides containing the coiled-coil trigger sequences of GCN4 and cortexillin I (see below and Refs. 14 and 15).
Thus, the specific distribution of hydrophobic and hydrophilic residues of this particular sequence appears critical for proper coiled-coil formation of the SR-A oilgomerization domain. The alignment of known SR-A sequences (Fig. 4) reveals that at least one potential intrahelical salt bridge and the isoleucine residue at the heptad d position are strictly conserved. Interestingly, all fragments containing this sequence exhibit similar stabilities as indicated by their T m values. Therefore we conclude that this heptad repeat has a major stabilizing function.
Proposed Model for Coiled-coil Formation-We recently demonstrated that the coiled-coil trigger sequences we identified within the two-stranded coiled-coil domains of cortexillin I and GCN4 represent autonomous helical folding units (14,15). Although monomeric, 16-residue peptides containing the coiled-coil trigger sequences revealed significant helicity in aqueous solution (15). Ionic strength-dependent stability of the GCN4 trigger peptide indicated that favorable intrahelical electrostatic interactions play an important role. On the basis of these observation and together with the data published in the literature we have proposed a mechanism for coiled-coil formation in GCN4. The presence of a helical segment within GCN4 monomers significantly limits the number of possible chain conformations and provides an ideal scaffold for the interaction of critical core residues. Stable dimer formation may involve the interaction of two helical trigger sites at some stage in the folding pathway. Interacting helices then "zip up" along the dimer to finally form the stable coiled-coil structure. It should be noted that such a mechanism would ideally arrange the two-stranded coiled coil in parallel register. Our suggested mechanism for the folding pathway of GCN4 is supported by recent studies on the folding kinetics of the GCN4-p1 leucine zipper (16). It has been observed that partial helix formation indeed precedes dimerization in GCN4 leucine zipper folding. Interestingly, the AGADIR program (39) predicts a helix content for the GCN4 trigger peptide GCN4p16 -31 of 31% that is in excellent agreement with the experimental value of 40% (15). Also the coiled-coil trigger sequence of cortexillin I is located in the region with the highest predicted helical propensity. 3 Only a low degree of helicity is predicted for the full-length human SR-A coiled-coil domain (Fig. 5). Interestingly, there appears to be a periodicity in the predicted helix content of the SR-A coiled coil. In contrast to the coiled-coil trigger sequences of cortexillin I and GCN4, however, only a low helix content is predicted for the SR-A trigger site (residues Ile 173 to Ser 179 ). Although mechanistic differences between two-and three-stranded trigger sites may exist, this observation demonstrates that coiled-coil trigger sequences cannot be unambiguously identified on the basis of their predicted ␣-helicity.
Our results further demonstrate that the helical content of heptad repeat-containing polypeptide chain fragments per se is not sufficient for specific coiled-coil formation. Although helical, neither of the N-terminal SR-A fragments lacking the trigger site was able to fold into trimeric coiled-coil structures. Kinetic studies of wild-type and mutant SR-A oligomerization domains and fragments thereof will now be necessary to clarify the exact role of the trimeric coiled-coil trigger sequence.