Ca2+ attenuates nucleation activity of leiomodin.

A transient increase in Ca2+ concentration in sarcomeres is essential for their proper function. Ca2+ drives striated muscle contraction via binding to the troponin complex of the thin filament to activate its interaction with the myosin thick filament. In addition to the troponin complex, the myosin essential light chain and myosin-binding protein C were also found to be Ca2+ sensitive. However, the effects of Ca2+ on the function of the tropomodulin family proteins involved in regulating thin filament formation have not yet been studied. Leiomodin, a member of the tropomodulin family, is an actin nucleator and thin filament elongator. Using pyrene-actin polymerization assay and transmission electron microscopy, we show that the actin nucleation activity of leiomodin is attenuated by Ca2+ . Using circular dichroism and nuclear magnetic resonance spectroscopy, we demonstrate that the mostly disordered, negatively charged region of leiomodin located between its first two actin-binding sites binds Ca2+ . We propose that Ca2+ binding to leiomodin results in the attenuation of its nucleation activity. Our data provide further evidence regarding the role of Ca2+ as an ultimate regulator of the ensemble of sarcomeric proteins essential for muscle function. SUMMARY STATEMENT: Ca2+ fluctuations in striated muscle sarcomeres modulate contractile activity via binding to several distinct families of sarcomeric proteins. The effects of Ca2+ on the activity of leiomodin-an actin nucleator and thin filament length regulator-have remained unknown. In this study, we demonstrate that Ca2+ binds directly to leiomodin and attenuates its actin nucleating activity. Our data emphasizes the ultimate role of Ca2+ in the regulation of the sarcomeric protein interactions.

INTRODUCTION

Striated muscle contraction is responsible for the beating of the heart and for muscle‐based motility. The sarcomere is the most basic contractile unit of striated muscle. Two of the sarcomere's fundamental components are the overlapping thin and thick filaments, made primarily of actin and myosin, respectively. Periodic transient increases in Ca2+ levels in striated muscle cells regulate the function of integral muscle proteins, such as troponin, myosin, myosin‐binding protein C, and titin. , , , , Ca2+ drives striated muscle contraction via binding to the troponin complex of the thin filament to activate its interaction with the myosin thick filament. Interaction between the myosin essential light chain and Ca2+ activates the myosin holoenzyme. Ca2+ restores the inhibitory capacity of phosphorylated myosin‐binding protein C. Ca2+ binding to the immunoglobulin domain of titin results in increased stability and stiffness. Another essential sarcomeric protein with a potential to bind cations is leiomodin (Lmod), a protein of the actin‐binding tropomodulin (Tmod) family. In particular, Lmod contains a long negatively charged region of unknown function (Figure 1) ; this region could potentially bind Ca2+.

FIGURE 1

Domain structure of Lmod2. Lmod2 is comprised of several functionally distinguished regions: TpmBS – Tropomyosin‐binding site, ABS – Actin‐binding site, EDRR – Glu/Asp rich region

The process of building, maintaining, and repairing of the thin filament is complex and involves Lmod, and Tmod, along with ZASP, nebulin, formins, profilin, and CapZ, at various stages of development (for review, see ). Tmod family proteins, with affinity for both actin and tropomyosin (Tpm), have a critical role in controlling thin filament lengths. , , , , Regulation of thin filament length is crucial to ensure an optimal overlap between thin and thick filaments and an optimal force generation during muscle contraction. Additionally, the ability of Lmod to nucleate actin was proposed to play a role in de novo formation of thin filaments. , Loss of Lmod causes aberrant sarcomere assembly and severe muscle dysfunction. , , , Yet, nothing is known about upstream regulation of Lmod function.

There are three Lmod isoforms, Lmod1, Lmod2, and Lmod3, each expressed primarily in smooth, cardiac, or skeletal muscle, respectively. , In this study, we sought to examine whether Ca2+ levels affect the nucleating activity of the cardiac isoform of Lmod, Lmod2. Lmod2 is a modular protein with a specific organization of Tpm‐ and actin‐binding sites (Figure 1; for review see ). In order of their appearance in the sequence from the N‐terminus, there is a Tpm‐binding site (TpmBS1), an actin‐binding site (ABS1), a leucine‐rich repeat (LRR) domain serving as the second actin‐binding site (ABS2), and a Wiskott‐Aldrich Homology 2 (WH2) domain, serving as the third actin‐binding site (ABS3). There are two disordered flexible linkers, the first one connecting ABS1 and ABS2 and the second one connecting ABS2 and ABS3 (Figure 1).

Using both pyrene‐actin polymerization assay and transmission electron microscopy (TEM), we have found that Lmod2 nucleation activity is attenuated by Ca2+. Using nuclear magnetic resonance (NMR), we have shown that Ca2+ binds to the linker between the first two actin‐binding sites in Lmod2. We propose that Ca2+ binding to Lmod2 causes the observed decrease in Lmod2 nucleation activity. Lmod2 is the first pointed‐end binding protein, whose function has been shown to be Ca2+‐sensitive.

RESULTS

Effect of Ca2+ on Lmod2‐induced actin nucleation

To test whether Ca2+ has an effect on Lmod2‐induced actin nucleation, we measured polymerization of pyrene‐labeled G‐actin in the presence and absence of Lmod2 at high (pCa = 3.7) or low (pCa = 9) levels of Ca2+ (Figure 2). Since the pyrene‐labeled actin fluorescence is directly proportional to the extent of actin polymerization, the lag time and maximum polymerization rate can be deduced from the changes in fluorescence.

FIGURE 2

Effect of Ca2+ on Lmod2 nucleation observed by pyrene‐actin polymerization assay. (a) Representative time courses of each condition with their sigmoidal fits shown by solid lines. (b) Lmod2/control maximum polymerization rate ratios. (c) Lmod2/control lag time ratios. Bars and error bars are the average and standard deviation, respectively. Values of each repeat are represented by circles, and the P‐values were calculated by a paired t‐test, n = 4

It was shown that Ca2+ negatively affects the rate of actin polymerization. Therefore, to properly evaluate the effect of Lmod2, first, we established at which concentrations of G‐actin the polymerization rates at pCa = 9 and pCa = 3.7 were similar. We compared the polymerization rate of 3 μM G‐actin at pCa = 3.7 with polymerization rates of G‐actin in a range of concentrations from 1 to 2 μM at pCa = 9 (Figure S1). We found that 1.15 μM G‐actin at pCa = 9 displays a similar rate of polymerization as 3.0 μM G‐actin at pCa = 3.7. Therefore, these conditions were used to elucidate the effect of Ca2+ on Lmod2‐induced actin nucleation (Figure 2a). Analysis of the polymerization curves revealed that at either pCa = 9 or pCa = 3.7, Lmod2 increased the maximum polymerization rate to the same extent (Figure 2b). At the same time, the lag time was differentially affected by adding Lmod2. It was reduced at pCa = 3.7 and pCa = 9 by a factor of ~2.5 and ~5, respectively (Figure 2c). This suggests that the presence of Ca2+ reduces the nucleation activity of Lmod2, since the lag phase of polymerization reflects the time required for the formation of actin nuclei.

To confirm that the increase in pyrene fluorescence was due to the formation of actin filaments nucleated by Lmod2 we used TEM (Figure 3). The G‐actin concentration was optimized for the TEM to keep the distribution of filaments on the carbon coated grids suitable for visualization. First, we compared polymerization of 2 μM G‐actin at Ca2+‐free (Figure 3a‐c) and high Ca2+ (Figure 3d‐f) conditions. Consistent with the pyrene‐actin polymerization assay, at pCa = 9, G‐actin polymerized faster than at pCa = 3.7, producing noticeably more filaments after 2 min of spontaneous polymerization (compare Figure 3b and e). However, very few actin filaments were found after 30 s of spontaneous polymerization at either Ca2+ level (Figure 3a and d). In contrast, addition of Lmod2 yielded short but well‐organized actin filaments within the first 30 s (Figure 3g and j). The existence of those short filaments after such short times of polymerization is a hallmark of nucleation activity of Lmod2. Consistent with the pyrene‐actin polymerization assay, the Ca2+‐free condition was more favorable for Lmod2‐promoted actin polymerization, because, at pCa = 9 (Figure 3g), a larger number of short filaments were observed than at pCa = 3.7 (Figure 3j). The difference was even more noticeable after 2 min of incubation, since the Ca2+‐free condition yielded significantly denser population of actin filaments (compare Figure 3h and k). Therefore, our TEM data qualitatively confirmed the results obtained via the pyrene‐actin polymerization assay: high Ca2+ levels inhibit Lmod2's potency to nucleate G‐actin polymerization.

FIGURE 3

Spontaneous and Lmod2‐nucleated polymerization of G‐actin observed by electron microscopy of negatively stained samples at low and high Ca2+ levels. G‐actin was polymerized in the absence (a‐f) or presence (g‐l) of 50 nM Lmod. Each trial was done at either low (a‐c and g‐i) or high (d‐f and j‐l) levels of Ca2+, at 30 sec, 2 min and 10 min time points

Interaction of Lmod2 with cations

Attenuation of Lmod2's nucleation activity by Ca2+ implies that Ca2+ binds directly to Lmod2. There is an acidic region located within the 107‐residue linker between the first two actin‐binding sites of Lmod2 (Figure 1). Almost half of amino acid residues in this region are Glu and Asp residues. To study the putative Ca2+‐binding properties of Lmod2, we chose a fragment (Gly90‐Tyr157) covering the acidic region. Hereafter we will refer to this fragment as EDRR (Glu(E)/Asp(D)‐Rich Region).

We cloned and purified the fragment of Lmod2 representing EDRR. The fragment is a charged low complexity polypeptide, which makes it likely to be unstructured or poorly structured. With the exception of the well‐structured ABS2 represented by the LRR domain, Lmod2 is mostly disordered (Figure S2). Across all of the disorder predictors in the D2P2 database, there is strong agreement that EDRR (Gly90‐Tyr157) is a part of the disordered regions. To prove the disordered state of EDRR experimentally, we measured its circular dichroism (CD) spectrum (Figure S3). The shape of the spectrum was consistent with EDRR being a mostly disordered polypeptide, since the major minimum was at 197.5 nm. The helical content of EDRR calculated from the CD spectrum was only ~10%. Upon the addition of either Ca2+ or Mg2+, at different cation to peptide ratios, we did not observe obvious changes in the shape of the EDRR CD spectrum (Figure S3), which suggests that neither cation causes a detectable change in the helical content. This is not unusual as cation binding is not always accompanied by a change in such secondary structure elements as α‐helices and β‐strands. , ,

To further evaluate the cation‐binding ability of the EDRR fragment, we used NMR. Since the chemical shifts of nuclei are sensitive to their environment, the interaction of EDRR with cations could be detected as a change in the 1D proton NMR spectrum. Titrating the peptide with increasing concentrations of cations would not only allow us to conclude whether the observed changes in chemical shifts are distinct with respect to the chemical nature of the cation, but also whether they are dose‐dependent. Therefore, we compared EDRR 1D proton NMR spectra using a range of either Ca2+ or Mg2+ concentrations (Figure 4).

FIGURE 4

NMR titrations of the Lmod2 EDRR with Ca2+ and Mg2+. Displayed are NMR spectral changes in three chemical shift regions upon addition of (a) Ca2+ and (b) Mg2+. Molar excesses (0, 0.5, 1.0, 2.0, 3.0, and 5.0) of [cation] over [peptide] are shown to the left of the relevant spectra. Left and middle panels in both titrations (chemical shift regions 8.25–8.65 ppm and 7.95–8.35 ppm, respectively) correspond to resonance peaks of amide (backbone) protons. Right panels (chemical shift region 2.5–3.2 ppm) correspond to resonance peaks of side‐chain protons. Solid vertical lines show initial (prior to cation addition) positions of the most affected peaks. Dashed vertical lines at ~2.6 ppm in (b) correspond to ~2.6 ppm solid lines in (a). The dashed lines demonstrate that side‐chain‐resonance peaks affected by Ca2+ are shifted markedly less when the peptide is titrated with Mg2+

The left (8.25–8.65 ppm) and the middle (7.95–8.35 ppm) panels in the titrations represent chemical shift ranges that primarily contain resonances belonging to backbone amide (NH) protons (Figure 4). In both Ca2+‐ and Mg2+‐titrations, we observed similar changes in the amide proton spectra upon addition of either Ca2+ or Mg2+. Therefore, 1D spectra of the region corresponding to the NH protons demonstrated that EDRR binds Ca2+ and Mg2+, but, judging by the changes in this spectral region alone, there was no indication that the binding is specific to a particular cation. Nevertheless, contrary to the similarity of the effect of Ca2+ and Mg2+ on the NH resonances, chemical shift changes of resonance peaks located in the side‐chain region (represented by the right panels in the Figure 2, 4, 2.5–3.2 ppm) were more pronounced upon the addition of Ca2+ rather than when Mg2+ was used. This observation suggested that the binding was specific with respect to the type of the cation used in titration. The response of the side‐chain resonances to Ca2+ is indicative of a distinct secondary structure forming upon Ca2+ binding to EDRR.

Effect of pre‐incubation of Lmod2 with Ca2+ on Mg‐actin nucleation

Despite the evidence that Ca2+ binds to the EDRR region of Lmod2, we could not completely eliminate the possibility that the observed effect of Ca2+ on the actin nucleation by Lmod2 arose from the effect of Ca2+ on the structure of the actin nucleus that might change the affinity of Lmod2 for G‐actin. In the polymerization experiments presented above, we used G‐actin with Ca2+ in its nucleotide‐binding cleft. Upon adding Mg2+‐containing polymerization buffer, Ca2+ is expected to partially exchange with Mg2+. The rate of exchange depends on the [Ca2+]/[Mg2+] ratio, and therefore, pCa = 9 and pCa = 3.7 conditions yield different populations of Ca2+‐ and Mg2+‐bound actin monomers. In such case, the indeterminate Ca/Mg‐actin populations will introduce an uncertainty in our nucleation experiments.

To overcome this uncertainty, we exchanged the cation bound to actin from Ca2+ to Mg2+ prior to polymerization. We pre‐incubated Lmod2 in a buffer containing Ca2+ to allow Ca2+ to bind to the protein. To choose the preferable Ca2+ concentration for the pre‐incubation, we tested different Ca2+ concentrations in pyrene‐actin polymerization assay in the absence of Lmod2 (Figure S4). Based on these experiments, we chose the Ca2+ concentration of 0.01 mM, which did not significantly affect the polymerization kinetics of Mg2+‐bound G‐actin. Comparison of the actin nucleation and polymerization after addition of either Lmod2 or Lmod2 pre‐incubated with Ca2+ revealed that the maximal rate of polymerization was not affected by pre‐incubation with Ca2+, while the lag time was noticeably increased (Figure 5). Therefore, our data suggested that the effect of Ca2+ on Lmod‐induced nucleation was primarily due to Ca2+ binding to Lmod2.

FIGURE 5

Effect of pre‐incubation of Lmod2 with Ca2+ on nucleation of Mg‐actin observed by pyrene‐actin polymerization assay. (a) Representative time courses of each condition with sigmoidal fits represented by solid lines. (b) Lmod2/control maximum polymerization rate ratios. (c) Lmod2/control lag time ratios. Values of each repeat are represented by circles, and the P‐values were calculated by a paired t‐test, n = 4

DISCUSSION

Striated muscles rely on a Ca2+ transient for the contraction mechanism to proceed, and the Ca2+ transient affects the properties/functions of many essential muscle proteins, for example, troponin, myosin, titin, and myosin‐binding protein C. , , , However, the effect of Ca2+ on the activity of the members of the Tmod family proteins has never been examined.

In this study, we demonstrated that Ca2+ affects Lmod2 nucleation ability most likely by binding to the acidic region. The acidic region is present in the ABS1‐ABS2 linker of the striated muscle isoforms, Lmod2 and Lmod3, but it is absent in Lmod1, the smooth muscle isoform. In Lmod2, this region displays a large degree of homology across species (Figure S5), which suggests that it performs an important cellular function.

With respect to Lmod2 molecular interactions, most previous research efforts have been focused on the structure of interfaces between Lmod consensus binding sites (Figure 1) and Tpm or actin. , , , , , Much less attention has been paid to the linkers connecting the actin‐binding domains; thus, their functional significance has been unknown. Based on its sequence, it has been suggested that the ABS1‐ABS2 linker should be disordered, and the disorder prediction confirms this (Figure S2). The ABS1‐ABS2 linker does not bind Tpm, and it has been shown that a region between Glu124 and Gly201 within the linker might interact with actin.

Clearly, Ca2+ affects the ability of Lmod2 to nucleate actin (Figures 2 and 5). The search for a consensus Ca‐binding motif (e.g., an EF hand motif found in TnC and myosin light chains ) did not give us any matches in Lmod2. Therefore, our data suggest that the ABS1‐ABS2 linker harbors a unique Ca2+‐binding site. It might be similar to the so‐called condensed‐charge motif that has been shown to bind Ca2+, among other divalent cations, in proteins, such as starmaker, α‐synuclein, and prothymosin α. In proteins, Ca2+ is typically coordinated by side‐chain oxygen atoms of Asp, Glu, and Asn. In addition to 29 Glu and 4 Asp, there are 2 Asn residues located within the acidic region (Figure S5). Since the coordination geometry of Ca2+ is usually octahedral, we can estimate that 3–6 residues are required to form the complete coordination sphere. Based on the total number of residues Asp, Glu, and Asn alone, the EDRR can potentially bind up to 6–12 Ca2+ cations; however, considering the coordination geometry requirements, this number is likely to be smaller. Moreover, only a few of these residues may be responsible for the specific Ca2+ binding that affects Lmod2 function. Further studies are necessary to determine residues that are responsible for the specific Ca2+ binding.

The geometric requirements should also reduce the number of conformations that the mostly disordered ABS1‐ABS2 linker containing the EDRR can adopt. As a result, an average end‐to‐end distance of the linker might be reduced upon binding Ca2+. This could make the formation of nuclei by ABS1 and ABS2 binding to actin monomers less favorable, manifesting itself as a reduction in the nucleation activity of Lmod2 when it is Ca2+‐bound. Indeed, the cooperation between ABS1 and ABS2 is important for Lmod2 nucleation.

The present study calls for further investigation into the connection between the apparent Ca2+ sensitivity of Lmods and the role of Lmods in thin filament maintenance. It is possible that for Lmod to promote the formation of new thin filaments, it should happen when the muscle is relaxed and the free Ca2+ is low. Further characterization of the Ca2+ sensitivity of Lmods could change our understanding of the thin filament length regulation mechanism in striated muscles.

MATERIALS AND METHODS

Cloning of the recombinant Lmod2 Glu/asp‐rich region EDRR

The amino acid sequence of mouse Lmod2 (NP_444328.1) was from the National Center for Biotechnology Information (Bethesda, MD). DNA primers for cloning were from Integrated DNA Technologies (Coralville, IA). Restriction enzymes XhoI and EcoRI, OneTaq DNA polymerase, T4 ligase, and relevant buffers and solutions were supplied by New England Biolabs (Ipswich, MA). Synthesis of template DNA optimized for Escherichia coli expression was performed by GenScript Biotech (Piscataway, NJ). The DNA sequence encoding residues Gly90‐Tyr157 of the mouse Lmod2 was cloned as an MFH‐fusion protein by following the protocol of cloning Tmod2‐TpmBS2 described previously. To facilitate the fragment production and spectral data collection, Cys118 was replaced with Tyr, which is a highly conservative residue in mammalian Lmod2. Generation of a mutagenized expression vector incorporating the C118Y substitution was performed as we did before using two partially overlapping primers 5′‐GGAAGTT‐TAC‐ACGGAAGATGAAGAGGAATCTCAGGAAGAAGA‐3′ and 5’‐CTTCCGT‐GTA‐AACTTCCTCCGATACCTCGGAATTAGATTCAG‐3′. In the primers, complementary codons encoding Tyr118 are shown in bold. DNA sequences of the recombinant constructs were confirmed by Sanger sequencing (performed at GeneWiz, South Plainfield, NJ).

EDRR expression and purification

Expression in LB medium and purification of the MFH‐fusion protein were performed following the same protocols that we used for other MFH‐fused constructs. The eluted from Ni‐NTA fusion protein was dialyzed against 5 mM HCl with 2 dialysis solution changes, brought to 0.2 M in HCl, and cleaved overnight with a 400‐fold molar excess of CNBr. The reaction mixture was applied onto a Sep‐Pak C18 cartridge from Waters (Milford, MA) and washed with six bed volumes of 0.1% trifluoroacetic acid (TFA). The MFH tag and the uncleaved MFH‐fusion protein were removed from the column with 60% acetonitrile in 0.1% TFA. The Lmod2 EDRR was eluted from the Sep‐Pak column with a mixture of acetonitrile and 1% ammonium bicarbonate (6:4 vol:vol), dialyzed against 5 mM HCl and lyophilized. Concentration of the peptide was determined by measuring its extinction at 280 nm and comparing it with the molar peptide extinction predicted by an online ProtParam Expasy tool or using the difference in spectra in 6 M GuHCl between pH 12.5 and pH 7.0 as described in.

Protein purification

The G‐actin was purified according to protocols in, with some modifications. Reconstituted F‐actin pellets, after polymerization and removal of Tpm and TnT, were combined and sonicated in a bath sonicator on ice for 30 min. After dialysis to depolymerize actin, the solution was sonicated with a probe sonicator for 2 min then centrifuged at 80,000 rpm (rotor MLA‐80) and 4°C for 2 h and 10 min. G‐actin in the supernatant was purified on a Sephacryl S‐300 gel filtration column equilibrated with Buffer A. The P‐actin was prepared and purified according to protocols in. Lmod2 was expressed and purified as previously described. All protein concentrations were determined by measuring the difference in spectra in 6 M GuHCl between pH 12.5 and pH 7.0.

Nuclear magnetic resonance titration

The NMR samples were prepared in 20 mM HEPES‐d18 (pH 6.8), which was obtained from the Cambridge Isotope Laboratories (Tewksbury, MA). The samples contained 10% D2O and 0.2 mM DSS, both purchased from the Cambridge Isotope Laboratories. The concentration of the Lmod2 EDRR was 0.3 mM. The sample was titrated with small aliquots of 10 mM and 30 mM CaCl2 or MgCl2 stock solutions (adjusted to pH ~6.8). 1H NMR spectra were recorded at 25°C on a Varian Inova 500 spectrometer (500 MHz) equipped with a 5 mm triple‐resonance probe. The spectra were processed, visualized, and analyzed with Mnova (Mestrelab Research, Santiago de Compostela, Spain).

Circular dichroism

CD measurements were taken using an Aviv model 420 spectropolarimeter (Lakewood, NJ). The spectrum of 6 μM EDRR in 10 mM HEPES, pH 7.0, containing 100 mM NaCl was recorded in a 1 mm cuvette at 10°C. Helical content was calculated as in. Disordered regions in Lmod2 were predicted using the Database of Disordered Protein Predictions (D2P2) at https://d2p2.pro/.

Fluorescence measurement of Lmod2 nucleation at pCa = 3.7 and pCa = 9

Actin polymerization was measured as the change in fluorescence of pyrene‐actin on a PTI fluorometer (10 mm cuvette, 366 nm excitation, 387 nm emission, and 2 nm slit). Pyrene‐actin was added to 10% of the actin concentration. To measure the effect of Ca2+ on actin polymerization and Lmod2 nucleation of actin, aliquots of concentrated 20X polymerization buffer with or without 20X CaCl2 were added to initiate polymerization at room temperature. The final compositions of the polymerization buffers in the polymerization mixtures were: 100 mM KCl, 2 mM MgCl2, 2 mM EGTA, 1 mM DTT, 0.1 mM ATP, 25 mM imidazole pH 8.0, with (pCa 3.7) or without (pCa 9) 2 mM CaCl2. For our experiments, pCa (pCa = −log[Ca 2+], where [Ca 2+] is free Ca2+ concentration) was calculated by using a web‐based pCa calculator called CHELATOR. The calculation was based on the buffer contents and the ionic strength, and the ionic strength was calculated using Ionic Strength Calculator http://calistry.org/calculate/ionic-strength-calculator. We first polymerized 3.0 μM G‐actin at pCa = 3.7 and 1.0 μM, 1.15 μM, 1.5 μM, and 2.0 μM G‐actin at pCa = 9 (Figure S1). The maximum polymerization rates of 3.0 μM at pCa 3.7 and 1.15 μM actin at pCa = 9 were ~4.2 and ~4.5 respectively (Figure S1). So we chose to continue with 3.0 μM actin at pCa = 3.7 and 1.15 μM actin at pCa = 9 to elucidate the effects of Ca2+ on Lmod2‐indiuced nucleation; 5 nM Lmod2 was added right before adding polymerization buffer with or without CaCl2. Polymerization of actin in the absence of Lmod2 was measured as a control. The curves were fitted to a four‐parameter sigmoid function in SigmaPlot 12, the rate of polymerization was calculated as the first derivative of the function, and a tangent line to the point of the maximum rate was plotted. The lag time for actin polymerization was calculated by finding the intersection between the tangent line and the minimum fluorescence value (y‐value) of the curve as in.

Polymerization of Mg‐actin with Lmod2 pre‐incubated with Ca2+

At pCa = 9, in the presence of 1 nM Ca2+ and 2 mM Mg2+, Ca‐actin is converted to Mg‐actin with t1/2 ~ 40 sec, according to the competitive exchange model. At pCa = 3.7, in the presence of 0.2 mM Ca2+, 2 mM MgCl2, and 2 mM CaCl2, Ca‐actin is converted to Mg‐actin with t1/2 ~ 450 sec. In each condition, there is a different population of Ca2+ or Mg2+ bound actin over time. Exchanging the actin‐bound cation to Mg2+ before performing nucleation experiments largely averts this problem. To exchange the Ca‐actin to Mg‐actin, pyrene‐actin was first added to 10% of the actin concentration, then incubated in 0.05 mM EGTA and 0.05 mM MgCl2 at room temperature for 6 min as in. The actin was then applied to and eluted from a Cytiva PD‐10 MiniTrap desalting column packed with Sephadex G‐25 resin using an actin buffer containing 2 mM Tris–HCl pH 8.0, 0.05 mM MgCl2, 0.01% NaN3, 0.5 mM DTT, and 0.2 mM ATP. Concentrated 20X polymerization buffer was added to allow polymerization to begin at room temperature. The final composition of the polymerization buffer in the polymerization mixtures was: 150 mM KCl, 0.1 mM MgCl2, 1 mM DTT, 0.1 mM ATP, 25 mM imidazole, pH 8.0. We first polymerized 1.5 μM Mg‐actin in increasing concentrations of Ca2+, from 0 to 0.05 mM (Figure S2A). We chose 0.01 mM Ca2+ based on the observation that higher Ca2+ concentrations affect the polymerization of Mg‐actin (Figure S2B). To prepare working solutions of Lmod2, we diluted a 12.8 μM Lmod2 stock to 0.5 μM, with or without 0.5 mM Ca2+. After at least 10 min of incubation time, Lmod2 with or without Ca2+ was added right before adding polymerization buffer, to make the final concentrations 10 nM Lmod2 and 0.01 mM Ca2+. The curves were analyzed as before.

Transmission electron microscopy

Two μM G‐actin was polymerized at room temperature using the same conditions as described for fluorescent measurements. Aliquots of the polymerizing actin were used for negative staining at 30 s, 2 min, and 10 min time points. A small volume (7 μL) of each sample was incubated for 30 s on carbon‐coated, glow‐discharged grid (Electron Microscopy Sciences, FCF300‐Cu‐SB) and blotted with filter paper. After negative staining with 2% (w/v) uranyl acetate, grids were analyzed using JEM‐1200EX (JEOL) Transmission Electron Microscope at an accelerating voltage of 70 kV to record micrographs at a nominal magnification of 30,000×.

AUTHOR CONTRIBUTIONS

Garry E. Smith Jr.: Data curation (equal); formal analysis (equal); investigation (equal); resources (equal); validation (equal); visualization (equal); writing – original draft (lead); writing – review and editing (equal). Dmitri Tolkatchev: Data curation (equal); investigation (equal); resources (equal); visualization (equal); writing – original draft (lead); writing – review and editing (equal). Cristina Risi: Data curation (equal); investigation (equal); resources (equal); writing – review and editing (equal). Madison Little: Data curation (equal); writing – review and editing (equal). Carol C. Gregorio: Funding acquisition (equal); resources (equal); writing – review and editing (equal). Vitold E. Galkin: Conceptualization (equal); data curation (equal); funding acquisition (equal); investigation (equal); project administration (equal); resources (equal); supervision (equal); validation (equal); visualization (equal); writing – original draft (equal); writing – review and editing (lead). Alla S. Kostyukova: Conceptualization‐Lead, Funding acquisition‐Equal, Investigation‐Equal, Project administration‐Equal, Resources‐Equal, Supervision‐Lead, Validation‐Equal, Writing ‐ original draft‐Equal, Writing ‐ review & editing‐Lead.

Appendix S1 Figures

Click here for additional data file.