Introduction

Muscle contraction is triggered by an increase of the cytosolic-free Ca2+ concentration above the resting level (~ 10−7 M). For relaxation, new excitability Ca2+ must be removed from the actomyosin-associated regulatory proteins and transported out of the cytoplasm by membrane-bound Ca2+ ATPases (Rüegg 2012; Stammers et al. 2015).

In vertebrates, the high affinity Ca2+-binding protein parvalbumin has been assumed to act as a soluble relaxing factor by supporting the Ca2+ depletion of troponin and the Ca2+ uptake by the sarcoplasmic reticulum (SR) (Gerday and Gillis 1976; Pechere et al. 1977; Haiech et al. 1979; Gillis 1985; Arif 2009).

In contrast to vertebrate muscle, parvalbumin could not be detected in invertebrates. Instead, soluble or sarcoplasmic Ca2+-binding proteins (SCBP) were discovered and described in various invertebrate muscles. SCBPs are likewise to parvalbumin proposed to accelerate muscle relaxation by transporting Ca2+ from the myofibrils to the SR (Gerday 1988).

The amount of SCBP correlates with the contraction speed of the muscle type (Cox et al. 1976; Wnuk et al.1982; Cox 1990; Gao et al. 2006; White et al. 2011).

Protein sequencing data (Takagi et al. 1984, 1986) and X-ray crystallographic analysis (Cook et al. 1991) demonstrated that SCBPs belong to the EF-hand superfamily of calcium-binding proteins all sharing the common helix-loop-helix EF-hand domain for coordinative binding of Ca2+ ions (Kretsinger and Nockolds 1973; Mazumder et al. 2014).

SCBPs and parvalbumin are characterized by Ca2+-Mg2+ mixed sites with a high affinity for Ca2+ and Mg2+. Under physiological conditions where the Mg2+ concentration exceeds that of Ca2+, these mixed sites are mainly occupied by Mg2+. The Ca2+–Mg2+ exchange rates of both proteins are crucial for the acceleration of Ca2+ removal during a single contraction–relaxation cycle. According to various authors the exchange is not fast enough and therefore both proteins are considered as slow-onset buffers (Wnuk et al. 1982; Engelborghs et al. 1990; Schwaller 2010).

In previous studies, we have shown the presence of SCBPs in crude extracts of various muscles (leg muscles and the flight control muscles of the thorax) in the flies Drosophila melanogaster and Calliphora erythrocephala (Kiehl and D’Haese 1992). The results show that the specialized asynchronous muscles for flight, which contract more than once per nerve impulse, do not exhibit SCBP. However, in the synchronous steering muscles SCBP is present in high amounts. From the body wall muscle of the earthworm Lumbricus terrestris, three SCBP isoforms have been isolated and physicochemically characterized. In L. terrestris, more than a 1000-fold amount of SCBP has been calculated by sandwich ELISA in the body wall compared to the slow gizzard muscle (Huch and D’Haese 1992). Preliminary isometric measurements after single pulse and low-frequency stimulations of muscle stripes of the earthworm longitudinal and gizzard muscle indicate that in the longitudinal muscle the speed of relaxation is much faster than in the gizzard (Sturm et al. 1993). We could confirm a correlation between the amount of SCBP and speed of muscle contraction in the body wall and the gizzard muscle in L. terrestris by fluorescent in situ hybridization analysis (Thiruketheeswaran et al. 2016).

In the present study, we used SCBP, actomyosin, and fragmented SR all isolated from the earthworms’ obliquely striated muscle to analyze these components for their Ca2+ competition. In addition, we investigated the effect of SCBP on the Ca2+ diffusion to get more insight into the physiological significance of high amounts of SCBP in this invertebrate muscle.

Materials and methods

Inhibition of actomyosin ATPase and competition between actomyosin and SCBP for Ca2+

The actomyosin prepared according to D’Haese and Carlhoff (1987) and SCBP isolated as described elsewhere (Huch et al. 1988) were obtained from the body wall muscle of the earthworm Lumbricus terrestris. The ATPase activity of the actomyosin was determined as described by Carlhoff et al. (1987) in assay buffer containing 30 mM Tris-maleate pH 7.0, 20 mM KCl, 1 mM MgCl2, and the reaction was started by the addition of ATP (1 mM final concentration). Varying amounts of CaCl2 and EGTA were used to obtain free Ca2+ concentrations between 10−9 and 10−5 M (pCa2+: 9 to 5). Calculations of free Ca2+ were done according to Jewell and Rüegg (1966) using an apparent KD of 2 × 10−7 M for the Ca2+-EGTA complex. Liberated phosphate was determined photometrically according to Fiske and Subbarow (1925).

To examine the influence of SCBP on the ATPase activity, SCBP was decalcified by the following procedure: SCBP-containing solutions were brought to 5 mM EGTA,  dialyzed first against assay buffer, then against assay buffer containing approx. 10 g of the ion exchange resin Chelex-100 (BioRad, Richmond, USA).

Prior to the actomyosin competition experiment, SCBP and actomyosin were decalcified as described before except that competition buffer (10 mM Hepes, 10 mM histidine pH 7.0, 0.1 M KCl, 5 mM NaN3) was used. After addition of 1 mM MgCl2, samples of both protein solutions were mixed and preincubated (room temperature, 10 min) in equimolar concentration of 5 µM as was estimated using an apparent molecular mass of 20 kDa for SCBP (Huch et al. 1988) and 1 MDa for the actomyosin complex, the latter being calculated on the basis of the molecular ratios and the molecular masses of its components (myosin, actin, paramyosin, tropomyosin, and troponin) determined in SDS–PAGE (Carlhoff 1988). The reaction was initiated by pipetting 80 µl of starting solution [0.1 mM 45CaCl2 (3700 Bq ml−1), 10 mM MgATP] into the test tubes (total assay volume of each tube: 0.8 ml) followed by vigorous shaking, and stopped immediately (0 min) and after 1, 5 and 20 min by centrifugation (11000g, 5 min). The resulting actomyosin pellet was resuspended overnight in 0.4 ml of incubation medium according to Laemmli (1970) and the suspension was used for liquid scintillation counting (LSC, see below). For ultrafiltration of the supernatant to measure the Ca2+ bound to SCBP a Centricon 10 (Merck Millipore, Darmstadt, Germany) with a cut off of 10 kDa was used. Control experiments contained SCBP or actomyosin alone.

Competition between SCBP and SR for Ca2+

Essentially, this experiment is a modified Ca2+ uptake assay using fragmented sarcoplasmic reticulum (FSR) vesicles prepared from earthworm body wall muscle according to Heilmann et al. (1977). SCBP was equilibrated in SR competition medium (10 mM Hepes, 10 mM histidine pH 7.0, 0.1 M KCl, 5 mM MgCl2, 5 mM NaN3, 10 µM CaCl2, and 5 mM K2C2O4) and  dialyzed against the same buffer containing 45Ca2+ (final activity: 7400 Bq ml−1) overnight. The uptake of Ca2+ by FSR vesicles was measured by a filtration method as described by Martonosi and Feretos (1964) with adaptations according to Semich and Volmer (1985). FSR vesicles suspended in SR competition medium (protein concentration: 20 µg ml− 1) were preincubated under gentle stirring for 5 min at 25 °C. Thereafter, the radioactive competition medium with and without SCBP was added. After another 5 min, the Ca2+ uptake reaction was started by addition of ATP (final concentration: 5 mM). Aliquots of 0.3 ml were taken after various periods of incubation and transferred into a disposable 1 ml syringe fitted with a 0.45 μm pore size filter (type SJHV; Merck Millipore). After filtration, the syringe filter with the retained FSR vesicles was washed with 1 ml of 0.1 M NaCl and taken for LSC. To determine the Ca2+ bound to SCBP the filtrate was further processed by ultrafiltration using Centricon 10 (Merck Millipore).

Measurement of facilitated Ca2+ diffusion

Ca2+ diffusion experiments were carried out essentially according to Feher (1983) in a three-chamber system. The effect of SCBP, present in the middle chamber, on the diffusion of Ca2+ was analyzed as the rate or flux at which 45Ca2+ diffuses from the left to the right chamber.

The two outer chambers are 0.8 cm high and 0.8 cm in diameter, both containing a magnetic stir bar (6 mm × 3 mm) driven by micromotors placed below the chambers. The middle chamber is represented by a ring of varying thickness (0.2, 0.6, or 1.0 cm), which can be fixed between the outer chambers, thus providing variable diffusion distances. All diffusion experiments were performed at a diffusion distance of 0.2 cm unless otherwise indicated. Dialysis membranes (Spectra/Por®, molecular mass cutoff: 6–8 kDa, Spectrum Laboratories, Los Angeles, USA) fixed on both sides of the middle chamber separated the compartments from each other. Protein samples were introduced into the middle chamber as liquid 0.5% agar gels (temperature approx. 50 °C) in diffusion buffer (10 mM Hepes pH 7.0, 140 mM KCl, 0.5 mM MgCl2, 10 µM CaCl2), where they were allowed to gel for at least 30 min. During the experiment, the right chamber was perfused with diffusion buffer and the left chamber with the same buffer complemented with 45Ca2+ (final radioactivity: 3700 Bq ml−1). Carefully adjusted high precision pumps [type 112 (Beckman Coulter, Fullerton, USA) or P-500 (Pharmacia LKB, Uppsala, Sweden)] were used to ensure identical flow rates of 12 ml h−1 through both outer compartments. Fractions of 0.4 h (= 4.8 ml) were collected from the perfundate of the right chamber and aliquots of 0.6 ml from each second fraction were taken for determination of radioactivity by LSC. The surrounding temperature was monitored continuously with a calibrated temperature sensor. The diffusion experiments were performed at a temperature in the range of 23 ± 3 °C. Calculations of the Ca2+ flux (ILR) data were performed according to Feher (1984). Relative Ca2+ flux denotes the relation of ILR in presence of SCBP divided by ILR in presence of 0.5% agar: ILR (SCBP)/ILR (agar). Control diffusion experiments were performed with the non-calcium-binding protein bovine serum albumin (BSA, Cohn fraction V; Sigma).

Liquid scintillation counting (LSC)

The 45Ca2+ radioactivity of syringe filters and liquid samples were measured in 4 ml of Ready Protein Plus scintillation mixture using a LS 5000 CE counter (Beckman Coulter).

Protein estimation

Protein concentrations were determined according to Bradford (1976) using a dye reagent concentrate (BioRad) and BSA as a standard.

Statistical analysis

Data represent the mean (± standard deviation, SD) of three independent experiments, each performed in triplicate. The statistical significance of differences between two means was determined by Student’s paired t test. Data were considered significantly different if the two-tailed p value was < 0.05.

Results

Ca2+ bound to actomyosin is depleted by SCBP

The actomyosin used for these experiments showed a sigmoidal Ca2+ dependence of the ATPase activity with a Ca2+ sensitivity of 80–90% as calculated with the following equation according to Carlhoff and D’Haese (1987): 1 – [ATPase activity at 2 mM EGTA/ATPase activity at 0.1 mM Ca2+] × 100 (%) (Fig. 1).

Fig. 1
figure 1

Inhibition of actomyosin ATPase by SCBP. The ATPase activity of the actomyosin was determined as described in methods. A sigmoidal curve was obtained demonstrating a drastic decrease of enzymatic activity below a free Ca2+ concentration of 2 × 10−7 M. With the addition of increasing amounts of Ca2+-free SCBP (indicated by arrows), the ATPase activity of the actomyosin was decreased. In the presence of 20 µM SCBP the ATPase activity was reduced to 319 ± 15 nmol Pi min−1 mg−1, indicating a free Ca2+ concentration of about 9 × 10−8 M. With 100 µM SCBP the actomyosin ATPase activity was further reduced to 170 ± 10 nmol Pi min−1 mg–1 equivalent to a decrease down to about 6 × 10–8 M of free Ca2+. In all preparations the actomyosin showed a Ca2+ sensitivity of about 80–90% as calculated with the following equation: 1 – [ATPase activity at 2 mM EGTA/ATPase activity at 0.1 mM Ca2+] × 100 (%). Bars indicate mean ± SD of three independent experiments. The sample size for each determination of ATPase activity in the presence of SCBP was N = 3

Full size image

Actomyosin with full enzymatic activity (560 ± 20 nmol Pi min−1 mg−1) was incubated with Ca2+ deprived SCBP in the presence of 10 µM Ca2+ (pCa2+: 5). With increasing amounts of SCBP the ATPase activity of the actomyosin was progressively decreased to about 170 ± 10 nmol Pi min−1 mg−1 in presence of 100 µM SCBP (Fig. 1).

To analyze how SCBP and actomyosin compete for Ca2+, a competition assay was conducted (see methods). Analysis of the 45Ca2+ radioactivity in the actomyosin pellets obtained from the actomyosin–SCBP competition experiment revealed a reduction of the Ca2+ initially bound to actomyosin with increasing incubation time. The most drastic decrease was observed after 1 min of incubation: the mean counting rate dropped from slightly more than 26 × 103 cpm down to approx. 18 × 103 cpm. After 20 min the counting rate in the pellet was further reduced to about 15 × 103 cpm (Fig. 2a). By mass determination of the actomyosin pellet, it was ensured that the reduction of the counting rate was not due to a loss of sedimentation efficiency in the course of the experiment. Protein estimation of the supernatant revealed that actomyosin was sedimented to 95% under the assay conditions used.

Fig. 2
figure 2

Actomyosin–SCBP competition assay. Analysis of the 45Ca2+ radioactivity in the actomyosin pellet (a) and SCBP supernatant (b) obtained from the actomyosin–SCBP competition experiment revealed a reduction of the 45Ca2+ initially bound to actomyosin while at the same time the 45Ca2+ uptake by SCBP increased with incubation time. The most drastic decrease was detected after one minute of incubation: the counting rate dropped from 26 × 103 cpm (0 min) down to approx. 18 × 103 cpm. After 20 min the counting rate in the pellet was further reduced to about 15 × 103 cpm. In contrast, within the SCBP-containing supernatant an increase in SCBP bound 45Ca2+ was measured. 45Ca2+ radioactivity was lowest, when actomyosin and SCBP were separated immediately after starting the competition reaction (8 × 103 cpm, 0 min) and increases with increasing incubation time (11 × 103 cpm, 20 min). Both competing proteins (actomyosin and SCBP) were present in equimolar concentration of approx. 5 µM. Bars indicate mean ± SD of three independent experiments

Full size image

In contrast to actomyosin, SCBP is not sedimented and the alteration of Ca2+ bound to SCBP was measured by a subsequent ultrafiltration of the supernatant. Within the SCBP-containing retentate after ultrafiltration, 45Ca2+ was accumulated. 45Ca2+ concentration was lowest, when actomyosin and SCBP were separated immediately after starting the competition reaction (0 min). With longer incubation times an increase in radioactivity was measured (Fig. 2b). This result suggests, that Ca2+ first binds to actomyosin and subsequently is taken over by SCBP. In control experiments using actomyosin or SCBP alone both proteins showed the highest Ca2+ binding already at the start point (0 min) and also at the following time intervals.

SCBP facilitates Ca2+ diffusion

The Ca2+ diffusion experiments were first performed with 0.5% agar and bovine serum albumin (50 and 250 µM) which does not belong to the family of high affinity Ca2+-binding proteins. A typical curve of an experiment with 0.5% agar shows a steep incremental phase turning into equilibrium or steady-state 45Ca2+ flux after 8 to 10 h (Fig. 3a). This curve represents the control situation with a mean steady-state 45Ca2+ flux (ILR) = 1 (0.041 ± 0.003 pmol Ca2+ cm−2 s−1) which was taken as a reference for all proteins tested. All measured 45Ca2+ flux values are relative to the steady-state flux in agar. Bovine serum albumin revealed no difference to 0.5% agar alone in the middle chamber as far as the incremental phase of 45Ca2+ radioactivity in the outflow from the right chamber and the steady-state Ca2+ flux are concerned.

Fig. 3
figure 3

Facilitated Ca2+ diffusion by SCBP and parvalbumin. a 45Ca2+ flux (ILR) values on the left ordinate are given in % which corresponds to ([45Ca2+]R/[45Ca2+]L) × 100. Relative 45Ca2+ flux denotes the relation of ILR in presence of SCBP, parvalbumin or bovine serum albumin (BSA) divided by ILR in presence of 0.5% agar alone: ILR(protein)/ILR(agar). BSA (circles) revealed no difference to 0.5% agar alone, i.e., a steep incremental phase turning into a steady-state 45Ca2+ flux (0.041 pmol cm−2 s−1) was observed after 10 h. The resulting flux data for SCBP (squares) and parvalbumin (triangles) were 1.9 (0.077 pmol cm−2 s−1) and 1.7 (0.072 pmol cm−2 s−1), respectively. There was no significant difference between the fluxes with SCBP and parvalbumin (two sample t test p > 0.05). All protein solutions (25 µM) have been dialyzed against 45Ca2+-containing diffusion buffer prior to the experiment (see “Materials and methods”). b With increasing concentration of SCBP the incremental phase of the 45Ca2+ flux is prolonged, from about 10 h at 5 µM to about 100 h at 100 µM SCBP. As compared to the 45Ca2+ flux in agar (ILR [agar]) 5, 50 and 100 µM SCBP yielded in an increase of about 30% (relative flux 1.3; 0.053 pmol cm−2 s−1), 60% (relative flux 1.6; 0.067 pmol cm−2 s−1), and 70% (relative flux 1.7; 0.070 pmol cm−2 s−1), respectively. Bars indicate mean ± SD of three independent experiments

Full size image

Earthworm SCBP was analyzed in various concentrations (5, 20, 25, 50, 100 and 200 µM) in the Ca2+ diffusion experiment. The results demonstrate an increase in steady-state Ca2+-flux as a function of SCBP concentration. Already 5 µM SCBP yielded an increase of the Ca2+ flux of about 30% (mean relative flux 1.3, 0.053 pmol cm−2 s−1) as compared to the control (0.5% agar). The amount of flux increase or the amount of diffusion facilitation brought about by SCBP tends to reach a maximum value of about 70% (relative flux 1.7, 0.070 pmol cm−2 s−1) at 100 µM SCBP (Fig. 3a, b). This can be concluded from the concentration differences and the corresponding flux increase. In addition, the steady-state Ca2+ flux obtained from an experiment with 200 µM SCBP was not significantly higher (two sample t test p > 0.05, data not shown). With increasing concentration of SCBP the incremental phase of Ca2+ flux is prolonged, from approx. 10 h at 5 µM to about 100 h at 100 µM SCBP (Fig. 3b).

When tested in equimolar concentration (25 µM) the resulting equilibrium Ca2+ flux for parvalbumin from white muscle of the chub Leuciscus cephalus was not significantly different to SCBP (mean relative flux 1.7 corresponding to 0.072 pmol cm−2 s−1, Fig. 3a).

It has been reported previously (Huch et al. 1988) that the SCBP isoforms differ in their Ca2+-binding capacity. Using the method of equilibrium dialysis, the isoform SCBP2 was found to bind 2 mol Ca2+ mol−1, whereas SCBP3 bound 3 mol Ca2+ mol−1. The resulting flux data from experiments with purified isoforms (25 µM each) did not significantly differ (two sample t test p > 0.05). The Ca2+ fluxes were 0.072 ± 0.005 pmol cm−2 s−1 for SCBP2 and 0.077 ± 0.004 pmol cm−2 9 s−1 for SCBP3 corresponding to relative fluxes of 1.74 and 1.87, respectively (N = 3, data not shown).

Ca2+ diffusion experiments with constant concentration of SCBP (100 µM) were performed to analyze the influence of the diffusion distance. At all three different diffusion distances tested (0.2, 0.6, and 1.0 cm) the steady-state Ca2+ flux increase was about 70% (Fig. 4). Applying the flux equations of Feher (1984) to our data, we calculated that SCBP increases the apparent Ca2+ diffusion coefficient by about 50%.

Fig. 4
figure 4

Dependence of 45Ca2+ flux on the diffusion distance. The inverse of the 45Ca2+ flux (1/ILR) is plotted against the thickness of the middle compartment (ΔX) for flux data obtained in the absence (blank triangles) and presence (black squares) of SCBP (100 µM) in diffusion buffer (see Methods). In the absence of SCBP the equilibrium or steady-state 45Ca2+ flux was reached for 0.2 cm after 6–8 h (0.039 pmol cm−2 s−1), for 0.6 cm after 16–18 h (0.030 pmol cm−2 s−1), and for 1 cm after 30 h (0.022 pmol cm−2 s−1). In the presence of SCBP, the steady-state 45Ca2+ flux was reached for 0.2 cm after 72 h (0.070 pmol cm− 2 s− 1), for 0.6 cm after 192 h (0.053 pmol cm−2 s−1) and for 1 cm after 300 h (0.036 pmol cm−2 s−1). The slope of 1/ILR in this plot corresponds to DaC from the equation ILR = DaC/ΔX, where Da is the apparent diffusion coefficient of 45Ca2+ and C is the total 45Ca2+ concentration (Feher 1984). Assuming a C of 1 µM the Da is 3.8 × 10−5 and 5.8 × 10−5 cm−2 s−1 in 0.5% agar and 100 µM SCBP, respectively. The plot shows that SCBP enhances the apparent diffusion coefficient for 45Ca2+. At all three different diffusion distances tested the obtained steady-state 45Ca2+ flux increase in presence of SCBP was about 70%. Bars indicate mean ± SD of three independent experiments

Full size image

Ca2+ bound to SCBP is deprived by SR vesicles

The Ca2+ competition experiment carried out with SCBP and fragmented SR (FSR) vesicles as competing agents for Ca2+ was designed to analyze whether earthworm SR can take up Ca2+ previously bound to SCBP. As described in methods, the uptake of Ca2+ by FSR vesicles was measured by filtration according to Semich and Volmer (1985) in 1 ml syringe fitted with a 0.45 µm pore size filter. The 45Ca2+ radioactivity remaining on the filters was analyzed after incubation of FSR vesicles with or without SCBP, both dependent on incubation time. In presence of 1 mg ml−1 SCBP (50 µM), that has been equilibrated against 45Ca2+-containing buffer prior to the competition test, the mean counting rates after 20 min were about 19 × 103 cpm compared to approx. 4 × 103 cpm, when FSR (20 µg ml−1) was incubated without SCBP (Fig. 5a). Protein determinations yielded no significant difference between the protein content of the SCBP solution and the SCBP-containing filtrates obtained after the competition assay implicating, that the additional amount of Ca2+ taken up by the FSR is not due to a direct interaction between SCBP and FSR. The control situation of FSR alone illustrates the background signal in the range of 4 × 103 cpm caused by unspecific binding to the filters. The SCBP-containing filtrate obtained after filtration to separate FSR from SCBP was subsequently ultrafiltrated leaving SCBP enriched within the retentate. The counting rate in the control experiment (SCBP without FSR) was constant with incubation time, with a mean of 63 × 103 cpm (Fig. 5a). In contrast, when SCBP was incubated with FSR, the amount of 45Ca2+ radioactivity in the retentate was reduced to a mean of 32 × 103 cpm after 20 min (Fig. 5b). Therefore, it can be concluded that under the experimental conditions the Ca2+-binding state of SCBP is reduced by about 50% as a consequence of the ATP-driven Ca2+-pumping action of the SR vesicles. Thus, the source for the additional amount of 45Ca2+ taken up by the FSR vesicles is SCBP.

Fig. 5
figure 5

SR-SCBP 45Ca2+ competition experiment. a Time course of 45Ca2+ uptake by fragmented SR (FSR) in presence of SCBP (black dots) or in incubation medium without SCBP (triangles). In presence of SCBP that has been equilibrated against 45Ca2+-containing buffer prior to the competition test (see methods), the counting rates after 20 min were about 19 × 103 cpm compared to about 4 × 103 cpm, when FSR was incubated without SCBP. Protein concentration: 1 mg ml−1 (= 50 µM SCBP) and 20 µg ml−1 (FSR). Assay volume: 1 ml. b Time course of 45Ca2+ deprivation from SCBP by FSR. SCBP was incubated with FSR (black dots) or alone (triangles). The SCBP-containing filtrate obtained after filtration to separate FSR from SCBP was subsequently ultrafiltrated leaving SCBP enriched within the retentate. The counting rate in the control experiment (SCBP without FSR) was constant with incubation time, about 63 × 103 cpm. In contrast, when SCBP was incubated with FSR, the amount of 45Ca2+ radioactivity in the retentate was reduced to about 32 × 103 cpm after 20 min. Bars indicate mean ± SD of three independent experiments

Full size image

Discussion

Two physiological functions have been proposed and are under continuous discussion for soluble Ca2+-binding proteins (SCBP) and parvalbumin (PV) during muscle relaxation namely as Ca2+ buffer and Ca2+ shuttle transporting Ca2+ between myofibrils and sarcoplasmic reticulum (SR). Both roles result from kinetic considerations of their ability to bind and exchange Ca2+ and Mg2+, respectively (Wnuk et al. 1982). Evidence supporting a role as buffer for SCBPs comes from structural studies which showed no significant Ca2+-dependent conformational changes (Engelborghs et al. 1990; Cook et al. 1993; Sillen et al. 2003) in contrast to “Ca2+ sensors” (e.g., calmodulin; Chin and Means 2000; Johnson 2006), where binding of Ca2+ induces a conformational change which enables them to interact with specific targets. In equilibrium dialysis and Scatchard plot analysis of the earthworm SCBP isoforms, SCBP2 bound two Ca2+ with a KD of 1.5 × 10−7 M and one Ca2+ with 1 × 10−5 M. For SCBP3 the binding of three Ca2+ with a KD of 1.3 × 10−7 M has been shown (Huch et al. 1988). Considering the different selectivity and affinity constants for Ca2+ (Zot and Potter 1984; Celio et al. 1996; Schwaller 2010), three and two mixed Ca2+–Mg2+ sites can be derived for SCBP2 and SCBP3, respectively. Under resting conditions where the free Mg2+ is several magnitudes higher than Ca2+ ([Ca2+]: about 0.01 µM, [Mg2+]: 0.5–1 mM; Berridge et al. 2000; Romani and Scarpa 1992) it is likely that the mixed Ca2+–Mg2+ binding sites are occupied by Mg2+. As shown for PV from frog muscle fibers and the SCBP of the marine annelid Nereis the dissociation rate of Mg2+ at the mixed Ca2+-Mg2+ binding sites are too slow so that Ca2+ is taken up by SCBP with a delay (Engelborghs et al. 1990; Hou et al. 1992). Therefore the rate of Ca2+ uptake by SCBP is determined by the rate of dissociation of Mg2+ from SCBP. Measuring the time dependent binding of Ca2+ in a mixture of SCBP and actomyosin both pretreated to be in a Ca2+-free and Mg2+-loaded form the added Ca2+ is first bound to a higher extent by the actomyosin and is thereafter progressively bound by SCBP. Apparently, the slow off-dissociation rate of Mg2+ from SCBP allows Ca2+ first to bind to Ca2+ specific sites of the regulatory proteins of the dually regulated earthworm actomyosin (troponin C and myosin light chains; D’Haese and Ditgens 1980) to trigger muscle contraction. Thereafter, SCBP is able to inhibit the enzymatic activity of body wall actomyosin in a chelator-like manner. As shown in our assay in Fig. 1, the addition of Ca2+-deprived and Mg2+-loaded SCBP led to a dose-dependent decrease of the free Ca2+ exhibiting a stronger binding to SCBP than to the Ca2+-binding proteins associated with the actomyosin. This function is related to their Ca2+-buffering feature (Schwaller 2010) and would be of importance in situations of “Ca2+ stress” in the cytosol, such as prolonged cell activation with high Ca2+ levels. Our experimental setup did not allow for short time measurements. The spatial arrangement of the actomyosin and SR in the earthworm muscle probably has great influence on the Ca2+ movements. In our assays, the concentrations of actomyosin and SCBP were chosen to correspond closely to the proportions of both components in muscle but for experimental reason the concentrations were about 10 times lower. This may be the reason why the kinetic of actomyosin decalcification by SCBP appears very slow. But it enabled us to detect a shift of Ca2+ from actomyosin to SCBP which is nearly complete within the first minute (Fig. 2). Use of higher concentrations is difficult as the actomyosin in rigor state was used with a three-dimensional network of actin and reassembled myosin. The Ca2+ depletion of actomyosin has been so far only shown for PV with isolated skeletal myofibrils (Gerday and Gillis 1976). Furthermore, it was demonstrated that after prolonged contraction the removal of Ca2+ in PV-deficient mice was much slower than in wild type muscle (Raymackers et al. 2000).

During the relaxation phase, a Ca2+ movement from the contractile proteins to the SR takes place. Our results obtained from the Ca2+ diffusion experiments demonstrate that the unidirectional Ca2+ flux is increased by Ca2+-loaded SCBP. The maximum amount of facilitation of Ca2+ diffusion (approx. 70%) implicates a saturation effect, which can be explained by a mutual hindrance of the SCBP molecules with increasing concentrations. In vivo SCBP may start to facilitate Ca2+ diffusion to the SR, provided that SCBP is Ca2+ saturated. As the exchange rates at the mixed Mg2+–Ca2+ sites are not fast enough during a single twitch, repeated stimulations are needed to provide a Ca2+ saturated SCBP to facilitate Ca2+ diffusion. For longitudinal muscle a tetanic contraction and a catch-like activity of the earthworm muscle has been shown (Tashiro and Yamamoto 1971). In these cases, SCBP could accelerate the relaxation rate after a prolonged intracellular Ca2+ increase by facilitation of Ca2+ transport from myofilaments to the SR. Considering facilitated Ca2+ diffusion two properties of the transporting protein are of major interest, which are its Ca2+-binding capacity and its affinity for Ca2+. Indications for the significance of Ca2+-binding affinity were gained by Feher et al. (1989) from comparative diffusion experiments performed with calmodulin, a ubiquitous Ca2+ sensor protein (Haeseleer and Palczewski 2002) and the intestinal calbindin-D9K which facilitates the absorption of Ca2+ (Balmain 1991). The effect of calmodulin was considerably less than that of calbindin-D9K. The authors attributed this difference to the lower association constant for calmodulin. Assuming an apparent KD of 2.5 × 10−6 M, it is conceivable that calmodulin was not fully saturated with Ca2+ (Zot and Potter 1984; Feher et al. 1989). Thus, as the incremental phase of 45Ca2+ flux is dependent on SCBP concentration and prolonged with increasing SCBP concentration (Fig. 3b) this phase may represent the process of equilibration of SCBP with 45Ca2+. This can also be concluded from the finding, that dialysis of the protein against the diffusion buffer prior to the experiment resulted in a shortening of the incremental phase by about 50%.

When regarding SCBP as a Ca2+-transporting molecule it is of interest to consider the interaction at the interface between cytosol and the SR. This issue was addressed by the Ca2+ competition experiment using SCBP and fragmented SR (FSR) vesicles. It has been shown by Stössel and Zebe (1968), that FSR vesicles from earthworm body wall muscle are able to inhibit the actomyosin ATPase by lowering the Ca2+ concentration. The vesicular fraction we used for Ca2+ competition showed Ca2+ uptake properties, which were dependent on ATP and free Ca2+ concentration very similar as FSR from vertebrate skeletal muscle (Rüegg 2012). The Ca2+ competition experiment showed that SCBP enhances Ca2+ uptake by FSR, which can be explained by dissociation of Ca2+ ions from SCBP that compensate for the reduction of Ca2+ in the medium due to the activity of the SR Ca2+ ATPase. In a previous study it was suggested that PV may activate Ca2+ uptake by directly binding to SR (Ushio and Watabe 1994). However, our filtrate experiments clearly showed that a direct interaction between SCBP and fragmented SR (FSR) can be excluded. The SCBP-containing filtrate obtained after filtration to separate FSR from SCBP was subsequently ultrafiltrated leaving SCBP enriched within the retentate. Protein determinations yielded no significant difference between the protein content of the applied SCBP solution and the SCBP-containing ultrafiltrates obtained after the competition assay implicating, that the additional amount of Ca2+ taken up by the FSR is not due to a direct interaction between SCBP and FSR.

Though in lower amounts PV has been shown to be present also in various other non-muscle tissues like brain and kidney (Berchtold et al. 1984; Heizmann 1988; Bastianelli 2003; Belge et al. 2007; Olinger et al. 2012). Likewise the presence of SCBP has also been described in neuronal tissue (Hermann and Cox 1995; Kelly et al. 1997; Thiruketheeswaran et al. 2016). In these organs the soluble Ca2+ binding proteins may function differently and play a role as a cytosolic Ca2+ buffer, as protectors against high Ca2+ levels, Ca2+ shuttle facilitating Ca2+ diffusion or even as Ca2+ sensors (Schwaller 2010).

In three different approaches using competition assays and Ca2+ diffusion experiments, our results consistently demonstrated a possible involvement of SCBPs in the relaxation mechanism of the fast obliquely striated muscle fibers of Lumbricus terrestris.