Introduction

Parvalbumins (Parv) are small Ca2+-binding proteins (9–13 kDa, 105–115 amino acids) found in many different tissues of lower and higher vertebrates, including blood and brain (Baron et al. 1975; Blum et al. 1977; Kretsinger 1980; Kretsinger et al. 1988; Henzl and Tanner 2007, 2008). The tissue in which Parv has been found in the highest concentrations is skeletal muscle. Because Parv is found only in fast muscles, it is generally associated with speed. There have been a number of reports in which twitch speed is positively correlated with Parv concentration (Heizmann et al. 1982; Klug et al. 1988), and there are several direct demonstrations of Parv’s ability to speed relaxation (Muntener et al. 1995; Szatkowski et al. 2001). For instance, following injection of parvalbumin c-DNA into mammalian slow twitch fibers, which normally do not express very much Parv, these fibers began to express considerable Parv and their relaxation rate increased dramatically, comparable to that found in fast twitch fibers (Muntener et al. 1995). Similar results were obtained in heart muscle, where Parv is not normally expressed (Wahr et al. 1999). Finally, Schwaller and colleagues were able to produce a Parv knockout in mice. Fast twitch fibers from the knockouts had slower Ca2+ transients and slower relaxation rate than occurred in the wild type (Schwaller et al. 1999; Raymackers et al. 2000). Thus, one interpretation is that Parv is a major determinant of the speed of the Ca2+ transient, and its high concentration is a requisite of fast relaxation.

In no other muscle does the correlation with speed seem more evident than in the superfast toadfish swimbladder (SWB) muscle, the fastest vertebrate muscle known (Rome et al. 1996). It has been found that it also has the highest concentration of Parv measured (Appelt et al. 1991). Edds-Walton et al. (2002) found that toadfish give low duty cycle call (2.5–8%), that is, fish call for about 400 ms and then there is a long 5–15 s intercall interval. Based on a series of biophysical measurements reviewed by Rome (2006), it has been proposed that Ca2+ pumps cannot keep up with the released Ca2+ and that Parv rapidly binds most of the Ca2+ released during the call, and this Ca2+ subsequently unbinds and is pumped back into the SR at a relatively slow rate during the long intercall interval (i.e., only after the muscle has relaxed). In this role, Parv permits the muscle to have a superfast Ca2+ transient with only a relatively modest SR-Ca2+ pumping rate (Rome 2006).

Midshipman (Porichthys notatus) is another fish which calls for its mate using its SWB muscle. The communication system as well as the ultra structure of its muscle has been extensively studied by Bass and colleagues (Bass and Marchaterre 1989; Brantley and Bass 1994; Bass et al. 2001; Lewis et al. 2003). Males call at about 80–100 Hz at 12–15°C. However, unlike toadfish who call with a 2.5–8% duty cycle, midshipman call with a 100% duty cycle (i.e., they call continuously for hours at a time). Without a long intercall interval to unload Ca2+, Parv could not help take up Ca2+ as it would become quickly saturated. Hence, we might expect little Parv in midshipman leaving the mechanism of fast relaxation unknown. However, it is possible that what seems to be 100% duty cycle call could actually have short intercall-intervals embedded between very long calls. If Parv were to be helpful in this circumstance, however, it would need to be in much higher concentrations than found in toadfish SWB to bind more Ca2+ over a longer duration.

In this study we quantitatively measured the Parv content of the SWBs from toadfish and midshipman. We found that midshipman have about 1/8th the Parv as toadfish, demonstrating that Parv would have little effect on Ca2+ movements during calling in midshipman. Further, although the toadfish SWB has almost four times more Parv than white muscle in toadfish, the Parv concentration in midshipman SWB is nearly equal to that of the locomotory muscle in midshipman.

Materials and methods

Animals

Oyster toadfish, Opsanus tau, were obtained from the Marine Resources Center of the Marine Biological Laboratory (MBL), Woods Hole, Massachusetts, MA. Type I male and female midshipman fish, Porichthys notatus, were caught during breeding seasons along the coast of Bodega Bay, California and then shipped to us by Professor A. H. Bass (Cornell University, Ithaca, NY). All specimens were maintained in an aquarium filled with artificial seawater. They were held at 15°C, fed 2–3 times a week and generally used within several days after arrival. The sex of the fish was determined based on examination of the internal sex organs and also by some phenotypic and morphological sex-specific differences. On the day of use, the fish were first anaesthetized with MS-222, cooled down with the ice and then killed by cervical transection. The SWB was immediately removed, weighed and used for corresponding analyses.

Preparation of muscle homogenates

In preliminary series of experiments, three different ways of intracellular protein extraction were tested on toadfish SWB and locomotory muscles of both fish species. Freshly excised, ice-cold pieces of muscle (50–100 mg wet weight) were transferred into plastic micro tubes and immediately homogenized in 5 volumes (vol/wet Wt) of the following precooled solutions: (a) high-salt extracting solution (0.6 M KCl, 0.04 M NaHCO3, 0.01 M Na2CO3, 4 mM 2-mercaptoethanol, pH 9.1); (b) high-salt solution plus 2% of Triton X-100; (c) standard Laemmli sample buffer. All solutions contained protease inhibitors cocktail (Sigma, P-9599). Homogenization was performed with a custom electric homogenizer equipped with a Teflon pestle fitted to the size of micro tubes. Speed and homogenization time were optimized separately and kept unchanged (4,000 RPM, 3 × 10–12 s) for all samples. Next they were diluted in 1:1 ratio with corresponding extracting solutions, re-homogenized and diluted again with 4 volumes of buffer for native gel-electrophoresis (62.5 mM Tris–HCl, pH 6.8, 40% glycerol, 0.01% Bromophenol Blue), or with standard Laemmli sample buffer—for SDS-gel electrophoresis. SDS-samples were boiled for 10 min and analyzed. These experiments showed that the most complete extraction of Parv took place when homogenization of muscles was performed in a high-salt extraction buffer. Triton X-100 did not affect significantly Parv extraction level. Finally it was found that homogenization of muscles in Laemmli buffer alone resulted in incomplete protein extraction.

Gel-electrophoresis and Western blotting

Analytical electrophoretic procedures were performed using Bio-Rad Mini-Protean II cell. In some experiments larger vertical gel system SE-600 (16 × 16 cm, Hoefer Sci.) with cooling coil was required. Running times and other parameters were adjusted experimentally. The gels were then stained with Coomassie R-250 or G-250, or silver-stained with a Bio-Rad Silver Stain Plus kit. Immunoblotting was performed in Bio-Rad Trans-Blot Cell with plate electrodes, fitted with a cooling coil. Proteins were transferred on PVDF membranes overnight at 0.1 A in Towbin buffer without SDS (25 mM Tris, pH 8.3, 192 mM glycine, 20% methanol). Detection system included Bio-Rad Western Blot amplification module and Opti-4CN™ Substrate Kit.

Isolation and separation of Parv isoforms

Most methods for isolation of Parv include the steps of heating of muscle extract up to 60–80°C, precipitation of denatured proteins and additional purification of Parv fractions. However, we found this approach unacceptable in the case of fish SWB muscle as it reduced dramatically the yield of Parv and impaired its Ca2+-sensitivity. We hypothesize that this is a manifestation of specific thermal stability of Parv from SWB muscle.

The main separation step of our method is size exclusion by treatment of crude muscle extract with an organic solvent (acetone). Isolated muscle was minced with scissors on ice, slowly mixed with 9 volumes of the high-salt extracting solution (described above) and left overnight at 4°C. The following day, filtered extract was re-suspended with 10 volumes of de-ionized H2O and water-insoluble proteins were removed by centrifugation at 5,000g × 15 min (pellet from this step contained most of intracellular proteins, such as actin, myosin, etc., see lanes 4–7 on the Fig. 1a). The supernatant was then treated with the slow addition of 50–55% (vol/vol) of cold acetone under continuous stirring. This caused formation of insoluble precipitate mainly composed of proteins larger than 15 kDa. The precipitate was removed by centrifugation at 5,000g × 15 min. The next steps were performed on the supernatant which contained a mixture of different Parv isoforms (total Parv fraction—see lanes 8, 9 on the Fig. 1a). It was either additionally purified and concentrated by two step sedimentation in presence of 70 and 100% ammonium sulphate, as described by (Pechere et al. 1971), or was carefully titrated with HCl (Fig. 1b). The pH-titration was alternated with centrifugation steps which allowed us to precipitate different Parv isoforms. According to our electrophoretic data, different Parv isoform fractions were contaminated with 5–10% of the adjacent (on the pH-titration curve) isoforms. Before being used, ammonium sulphate-precipitated Parv fractions were extensively dialyzed against large volume of 0.1 M Tris–HCl buffer (pH 7.0) buffer contained 100 mM KCl and 4 mM 2-mercaptoethanol, and then were additionally clarified by a high-speed centrifugation at 100,000 × g for 1 h.

Fig. 1
figure 1

Isolation and purification of different Parv fractions from toadfish SWB and locomotory muscles. a Coomassie-stained 18% polyacrylamide SDS-gel. 1—myoglobin (Mb); 2 and 3—myosin light chains and troponin standards from rabbit skeletal muscle (Sigma); 4 and 5—SWB homogenates from large and medium size toadfish; 6 and 7—homogenates from white and pink skeletal muscles; 8 and 9—total Parv fractions from white muscle and SWB; 10—proteins remaining in SWB after Parv extraction. b The pH-titration of total Parv fraction isolated from toadfish SWB (see lane 9, panel A). Optical density (OD) was continuously monitored during titration with a spectrophotometer. Different Parv fractions were collected by centrifugation (marked by arrows)

Full size image

Other biochemical measurements

The UV-absorption spectra of Parv preparations were recorded in 1 cm quartz cell on Ultrospec UV/Vis spectrophotometer (Pharmacia Biotech). Protein concentrations were determined by Bradford reaction, using Bio-Rad Protein assay reagent.

Amino acid composition of Parv isoforms

Having isolated several different isoforms with different spectroscopic properties we decided to evaluate the amino acid composition of the different isoforms. Analyses of amino acid composition of different Parv isoforms were performed by Lowell and Nancy Ericsson (AAA Laboratory, Mercer Island, WA) on Beckman 7300 amino acid analyzer coupled with the System Gold software. Analyses were performed by post-column derivatization of probes with ninhydrin, using the ion-exchange chromatographic methods which included 20 h hydrolysis of protein in 6 N-HCl/04% 2-mercaptoethanol/0.02% phenol at 115°C.

Results

Total Parv content

The low-molecular weight section of the gradient SDS-gel (Fig. 2, upper panel), shows large differences in total Parv content between the toadfish and midshipman SWBs. The quantitative differences are shown in the Figure 3. The highest content of Parv was found in SWB from toadfish (lanes 1 and 2 on the upper panel of Fig. 2 and bars 1 and 2 on Fig. 3). In both males and females concentration of Parv was very high and reached up to 13–15 g/kg of wet weight (i.e., 1.20–1.35 mM). That was 3.6–4.0 times higher than Parv concentration in white muscle, and 27–30 times higher than in pink skeletal muscle of toadfish (lanes 3 and 4 on the upper panel of Fig. 2 and bars 3 and 4 on Fig. 3). In contrast, midshipman SWB contained 6–8 times lower amounts of Parv than toadfish (lanes 1′ and 2′ on the upper panel of Fig. 2 and bars 1′ and 2′ on Fig. 3). Parv concentration was almost equal in SWB from males and females and was similar to that found in white skeletal muscle of this fish (i.e., ~2.0 g/kg wet weight, which is equal to ~0.18 mM—lanes 3′ and 4′ on the upper panel of Fig. 2 and corresponding bars on Fig. 3). This observation is surprising, because type 1 male midshipman possess the ability to generate high-frequency acoustic signals whose duration can be more than 50 times longer than in toadfish. Therefore one expectation could be that midshipman male might need more Parv than toadfish to produce these calls. However, our results do not support this hypothesis and clearly suggest that high intracellular content of Parv is not a requirement for the ability of muscle to contract and relax with superfast speed and at high frequencies.

Fig. 2
figure 2

Electrophoretic isolation of the total Parv fraction, separation and immuno-chemical identification of different Parv isoforms from SWB and locomotory muscles of toadfish and midshipman. 1 and 1′—males, 2 and 2′—females; 3 and 3′—white, 4 and 4′—pink skeletal muscles. Upper panel low molecular weight portion of 4–20% gradient polyacrylamide SDS-gel (Coomassie G-250 stain). Middle panel separation of Parv isoforms in muscle homogenates under non-denaturing conditions (18% linear polyacrylamide native gel, Coomassie R-250 stain); Parv isoforms (PA-Ia, PA-I, PA-II & PA-III/IIIa) were labeled according to their electrophoretic mobilities. Lower panel western blot of Parv isoforms shown on middle panel (obtained by using a monoclonal anti-Parv primary antibody (clone PARV-19, mouse IGG1, Sigma) and stained with Opti-4CN™ Detection Kit, Bio Rad)

Full size image
Fig. 3
figure 3

Total Parv content in SWB (1′ and 1′—male, 2 and 2′—female), white (3 and 3′) and pink (4 and 4′) muscles of toadfish and midshipman

Full size image

Distribution of Parv isoforms

High resolution native gel electrophoresis revealed presence of at least five bands in toadfish SWB homogenates which corresponded to different Parv isoforms (lanes 1 and 2 on the middle panel of Fig. 2). All these bands were immunochemically identified as different Parv isoforms (corresponding lanes on the lower panel). Three Parv isoforms were found in white skeletal muscle and two in pink skeletal muscle of toadfish (lanes 3 and 4 on the middle and lower panels). Two major Parv isoforms were present in about equal amounts in midshipman muscles; very low amounts of PA-III could be seen in SWB and in white muscle and trace amounts of this isoform were present in pink muscle of this fish (lanes 1′–4′ on middle and lower panels).

The Parv isoforms were labeled as PA-I/Ia, PA-II and PA-III/IIIa from top to the bottom on the gel according to the accepted classification based on their increasing electrophoretic mobility (middle panel on Fig. 2; Note that PA-I in toadfish SWB is not necessarily the same protein as “PA-I” in other muscles or species). The major isoform in toadfish SWB appeared to be PA-I, which together with a minor PA-Ia isoform comprised about 63–66% of total Parv in both, males and females (Fig. 4). PA-II isoform comprised 19–22% and PA-III + PA-IIIa comprised only 15% of total Parv in both, male and female toadfish. Thus, the ratio of these three major Parv isoforms in toadfish SWB was 3.0:1.0:0.7. This distribution of Parv isoforms was closer to the white, than to the pink skeletal muscle.

Fig. 4
figure 4

Distribution of different Parv isoforms in SWB and locomotory muscles from toadfish (TF) and midshipman (Msh). Note that the values inside of the bars are given as relative % of values for total Parv content, normalized to 100%. Based on the gel electrophoresis in non-denaturing conditions

Full size image

The SWB from midshipman also appeared to have three Parv isoforms. Both genders contained almost equal amounts of PA-I and PA-II isoforms which were complemented with a minor amount (~10%) of PA-III isoform. In general, Parv isoforms distribution in midshipman SWB was more similar to the pink than white skeletal muscle (Fig. 4).

Amino acid composition of Parv isoforms from toadfish SWB

High Parv content in relatively large swimbladder muscles makes toadfish a very good source for isolation of Parv and separation of its isoforms. Therefore, in this study we also performed the amino acid and spectral analyses of the major Parv isoforms from this muscle. In general, amino acid composition of all studied isoforms were similar to each other (78–91% similarity). Their amino acid compositions were close to the compositions described in earlier studies on Parv from various fish muscles [see, e.g., (Closset 1976; Gerday et al. 1979)]. Figure 5 illustrates amino acid profiles for various Parv isoforms from toadfish SWB. They appear similar; however, even small differences in their amino acid composition can produce large differences in their functional activity. Hence the amino acid composition of different Parv isoforms are shown in more detail in Table 1. First, this table shows that the ratio of “acidic/basic” amino acids [i.e., (Glx + Asx)/(Lys + His + Arg)] decreased in order PA-Ia:PA-I:PA-II:PA-III/IIIa as 1.86:1.71:1.31:1.17. This can be an indication of decreasing acidity of these Parv isoforms going from PA-Ia to PA-III/IIIa. Conversely, the large increase in the number of His residues (2:1:4:5) can significantly reduce or, even, reverse this tendency. It is also interesting that differences in amino acid compositions of various Parv isoforms were much larger between PA-Ia and PA-I, than between PA-II and PA-III/IIIa. In fact, PA-Ia and PA-I differed by 24 residues, whereas PA-II and PA-III/IIIa differed by only 10 residues; and PA-I differed from PA-II and PA-III/IIIa by 14 residues. These two groups of Parv isoforms also differed by His content: two and one residue in PA-Ia and PA-I, respectively, and four and five residues in PA-II and III/IIIa, respectively.

Fig. 5
figure 5

Amino acid composition profiles of Parv fractions from toadfish SWB muscle. Profiles were obtained on different Parv isoforms isolated from SWB of large size toadfish. The composition of one isoform sequenced by Gerday et al. (1989) is shown for comparison

Full size image
Table 1 Amino acid composition of Parv isoforms from toadfish SWB
Full size table

Importantly, these isoforms also contained different amount of Tyr residues: one in PA-I and PA-Ia, and two in PA-II and PA-III/IIIa. Because other two aromatic amino acids (i.e., Trp and Phe) were present in similar amounts in all isoforms, these differences could be responsible for the shapes of their UV-absorption spectra (Fig. 6). In fact, PA-I and PA-Ia showed significantly lower Tyr/Trp(around 265 and 275 nm) and higher Phe (259 nm) absorption than PA-II and PA-III. This led to higher A275/259 nm ratios for PA-II and PA-III/IIIa (1.13 and 1.15) than for PA-I and PA-Ia (1.0 and 0.96, respectively).

Fig. 6
figure 6

UV-absorption spectra for different Parv isoforms from toadfish SWB. Conditions: 50 mM Tris–HCl (pH = 7.0 at 20°C), 100 mM KCl, 0.2 mM CaCl2, 1.0 mM MgCl2, 4.0 mM 2-mercaptoethanol and 0.15 mg/ml PA

Full size image

Discussion

Total Parv content

An important finding of this study is that although toadfish appear to need a very high Parv concentration, midshipman SWB has a relatively low concentration similar to that found in its locomotory muscles. While it has been proposed that toadfish need Parv to serve as an auxiliary Ca2+ sequestor to provide a very fast Ca2+ transient whilst saddled with a modest Ca2+ pumping rate, midshipman apparently do not need this help. It is unclear at this time how rapid relaxation is accomplished in midshipman. Preliminary evidence shows that they have a rapid twitch, however the Ca2+ transient has not been measured nor have the Ca2+ exchange per twitch (i.e., Ca2+ release and reuptake) been determined. The fact that Parv concentration in midshipman SWB is the same as in the much slower contracting locomotory muscles is further evidence that Parv does not play a crucial role in providing rapid twitches in this species. This is further supported by the fact that Parv concentration is the same in the swimbladder muscles of calling (Type 1) males as it is in the non-calling females.

Rome and colleagues (Rome et al. 1996, 1999; Rome and Lindstedt 1998; Rome and Klimov 2000; Young and Rome 2001; Young et al. 2003; Rome 2006) have argued that superfast muscles are endowed with general physiological traits which are almost certainly common to all fibers of that type. These include having a fast Ca2+ transient, a fast off-rate of Ca2+ from troponin and a fast crossbridge detachment rate constant. It has further been shown that because the detachment rate constant must be so fast, that at a steady state few crossbridges are attached and hence these muscles generate low forces (Rome et al. 1999). Finally, it has been shown that in terms of mechanical function, superfast fibers and locomotory fibers are mutually exclusive. The locomotory fibers are too slow to power sound production at high frequencies, and at the low frequencies used for locomotion, superfast fibers cannot generate sufficient mechanical power to power locomotory movement due to their low force.

Aside from these general attributes, superfast muscles appear to develop a whole complex of biochemical, histochemical, structural and morphological adaptations for speed. Not enough information is available to know if some of these are general for all superfast muscles or if some are specific to a given species. For instance in toadfish, total myosin content in SWB muscle appeared to be almost 3 times lower than in locomotory muscles (Rome et al. 1999) which given the requirement for high SR volume is likely common to other species. Further, in fish SWB, radial morphology and small diameter of fibers (Fawcett and Revel 1961; Ono and Poss 1982; Fine et al. 1993; Loesser et al. 1997; Hirsch et al. 1998; Parmentier et al. 2003) appear common, but from a metabolic viewpoint, some species show abundant mitochondria (midshipman) but other do not (toadfish) (Ono and Poss 1982; Fine et al. 1993; Connaughton et al. 1997). Others have also found modifications of myosin light chains, troponin, and SR ATPase of toadfish SWB (Hamoir et al. 1980; Appelt et al. 1991; Feher et al. 1998). Others found important structural changes in this muscle include specific packing and high density of T-tubules of sarcoplasmic reticulum (Franzini-Armstrong and Nunzi 1983; Appelt et al. 1991), high Ca2+ capacity of reticulum (Feher et al. 1998). In midshipman SWB, species-specific adaptations are the highly reinforced elastic desmin cytoskeleton and extremely enlarged 1.0 μm wide Z-bands which does not occur in other species (Bass and Marchaterre 1989; Lewis et al. 2003).

Therefore, elucidation of possible mechanisms of contraction/relaxation of superfast muscles requires a complex approach in which both generalized and species-specific adaptations are considered. Although prior to this study, elevated Parv concentration found in toadfish SWB may have been assumed to be a generalized trait, our findings show that it appears to be a species-specific adaptation not found in some species. Our findings also suggest that some modification to mechanisms involving the Ca2+ release and reuptake by the sarcoplasmic reticulum is probably involved in the prolonged high frequency calling in midshipman.

The number and distribution of Parv isoforms

In this study we also examined the distribution of Parv isoforms in SWB of toadfish and midshipman of both genders. Despite toadfish SWB being the best studied superfast muscle, it was still not characterized in terms of the age and sex-specific distribution of Parv isoforms in its SWB. Previous authors have mentioned presence of multiple Parv isoforms in young and only two of them in adult toadfish SWB (Hamoir et al. 1980). In this study we have found three major (PA-I,II and III) and two minor (PA-Ia and IIIa, <5% of total Parv content) Parv isoforms in SWB of large adult toadfish. We also for the first time compared Parv isoforms composition in SWB and different skeletal muscles of midshipman fish.

Figure 4 illustrates the main trends of Parv distribution in studied muscles. In toadfish SWB this distribution was closer to that in white skeletal muscle, whereas in midshipman it was closer to that in pink muscle. It should be also noticed that in toadfish SWB major isoform PA-I (taken together with minor isoform Ia) made up 63–66% of total Parv and the ratio of different isoforms (i.e., PA-I + PA-Ia/PA-II/PA-III + PA-IIIa) was 3.0/1.0/0.7. These data are important because some authors have found good correlation between Parv isoform composition and relaxation rates in muscles from fish and some other vertebrates (Le Peuch et al. 1978; Celio and Heizmann 1982; Klug et al. 1988; Simonides and van Hardeveld 1989; Coughlin et al. 2007). According to these data, predomination of one or more different Parv isoform(s) can affect relaxation kinetics of muscle fibers presumably due to differences in Ca2+ and Mg2+ binding characteristics (Erickson et al. 2005). Therefore, physiologically, it is very important to know the type of Parv, or its isoform distribution in a given muscle. Finally, some previous authors (Hamoir et al. 1980) and our preliminary studies revealed different number and variable ratio of Parv isoforms in different size (i.e., age) fishes. For instance, we have found that small/young toadfish SWB expresses up to 5 different Parv isoforms in about equal amounts (unpublished data) and three (plus 2 minor) isoforms are expressed in large size adult fish (present study). No determination has been made as to whether the isoforms are the same in small and large fish and hence only differ in proportions. Therefore, when studying the SWB muscle it is very important to standardize the fish by their age (and size) as much as possible.

Amino acid composition of Parv isoforms

In general, all three major Parv isoforms (PA-I, PA-II, and PA-III/IIIa) showed very similar amino acid compositions, both among themselves and between themselves and previously identified parvalbumins. The similarities include: (a) relatively high amount of acidic (Glu and Asp) and basic (Lys) amino acids; (b) high amount of certain polar (Ser and Thr) and non-polar (Ala, Phe, Ile, and Gly) neutral amino acids; (c) only one or two residues of Cys/2; and (d) very low amount of Arg, Tyr, Pro, Trp, and Met. Nonetheless, differences among isoforms PA-I, PA-II, and PA-III/IIIa were apparent and support the view that the isoforms differ in sequence, rather than in simply post-translational modification. For example, the ratio of acidic to basic amino acids [i.e. (Glx + Asx)/(Lys + His + Arg)] decreases in the following order: 1.86 for PA-Ia; 1.71 for PA-I; 1.31 for PA-II; and 1.17 for PA-III/IIIa. It is this decreasing ratio that likely underlies the detected differences in isoelectric points, differences that would be sensitive to phosphorylation status of the proteins and protonation of His residues.

Another feature distinguishing the isoforms is the presence of one Tyr in PA-Ia and PA-I, yet two in PA-II and PA-III/IIIa. The lower Tyr content of PA-Ia and PA-I accounts for the weaker absorption of these isoforms, when compared with PA-II and PA-III/IIIa, at wavelengths longer than 265 nm. The differing content of Tyr, as well as Phe, consequently contributes to the distinct ratios of absorbance at 275 and 259 nm (A275/259 nm), which were 0.96 for PA-Ia, 1.0 for PA-I, 1.13 for PA-II, and 1.15 for PA-III/IIIa.

Finally, determining the Ca2+ binding kinetics and relating these to the sequence differences of the isoforms, should provide an useful approach to better understand the physiological significance of the different Parv isoforms and how they evolved.