ARCHIVES
OF BIOCHEMISTRY
AND BIOPHYSICS
Vol. 297, No. 2, September, pp. 383-387, 1992
COMMUNICATION Novel Protein
Inhibits in Vitro Precipitation
of Calcium
Carbonate’
Sybil K. Burgesq2 Dana M. Carey, and Sharon L. Oxendine Department
of Chemistry,
Received January
University
of North
Carolina at Wilmington,
Wilmington,
North Carolina 28403
7,1992, and in revised form June 3,1992
Organic molecules both coexist and interact with inorganic crystal lattices in biomineralizing tissues. Mineral precipitation and crystal morphology are tightly regulated by the actions of these molecules. Polyacrylamide gel electrophoresis studies on water soluble extracts from the cuticle of Callinectes sapidus (Atlantic blue crab) reveal the presence, in unmineralized nascent premolt cuticle, of proteins which are absent in the mineralized postmolt cuticle. In the present studies, homogenates from both premolt and postmolt C. sapidus cuticles have been tested for their effect on the in vitro precipitation of calcium carbonate. The role of protein in this process was determined by heat pretreatment and trypsin pretreatment of the cuticle homogenates prior to the precipitation assay. The results from these experiments indicate that proteins, with molecular weights of approximately 75,000 and between 10,000 and 20,000, concentrated in the C. sapidus premolt cuticle, inhibit calcium carbonate precipitation in vitro. The inhibitory activity of these proteins appears to be a result of specific interactions since trypsin, myoglobin, and ovalbumin are not inhibitory. The presence of lower amounts of these inhibitory proteins in C. sapidus postmolt cuticle may be responsible for the subsequent mineralization of this tissue. 0 1992 Academic Press, Inc.
Calcium carbonate is the biomineral found in invertebrate skeletal structures as well as the octoconia (balance organ) of humans. Regulation of calcium carbonate precipitation has been the subject of investigation by a number of laboratories (l-13). The organisms used in these studies have included sea urchins (l-4), crayfish (5), humans (6), garden snails (7), coral (8), oysters (9), nautiluses (lo), clams (ll), rats (12), and crabs (13). 1 This work was supported by a Greenwall Foundation, Inc., Grant of Research Corporation and by the University of North Carolina at Wilmington’s Center for Marine Science Research. ‘To whom correspondence should be addressed at Department of Chemistry, UNCW, 601 S. College Road, Wilmington, NC 28403-3297. 0003-9861/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
In vitro experiments with Ucapugilator (the sand fiddler crab) suggest that control of cuticle mineralization resides within the tissue’s organic matrix components (13). Furthermore, it has been shown by polyacrylamide gel electrophoresis studies with Callinectes sapidus (Atlantic blue crab) cuticle that certain proteins which are present in the crab’s premolt cuticle are present to a much lesser extent in the animal’s postmolt cuticle (13). To determine whether the appearance of specific proteins, in C. sapidus cuticle, at particular stages in the crab’s molt cycle was related to cuticle biomineralization, we compared the effect of premolt and postmolt cuticle homogenates on the in vitro precipitation of calcium carbonate and partially characterized the inhibitory substance found in the premolt cuticle. EXPERIMENTAL
PROCEDURES
Materials C. sqidus crabs were obtained from a local commercial seafood market. EDTA: NaOH, NaCl, KCl, CaCl,, NaHCO,, HCl, NaHPO,, NaH2P04, and calibration buffers were purchased from Fisher. Centricon-3, -10, and -30 ultracentrifugation devices were purchased from Amicon. BSA and Pierce Micro BCA Protein Assay Reagent were purchased from Pierce. Trypsin and myoglobin were purchased from Sigma.
Preparation
of Cuticle
Tissue Homogenates
Pieces of cuticle were removed from the dorsobranchial area of the crab during various molt stages. These pieces were stripped of underlying hypodermal tissue, cut into small pieces, and placed in 0.1 M EDTA (pH 8.0). The tissue pieces were then homogenized for 5 min in an ice bath. The resulting homogenate was then agitated at 4°C for 40 h. The tissue slurry was then rehomogenized and centrifuged at 20,OOOgat 4°C for 6 h. Supernatants were frozen until needed. Upon thawing, the soluble cuticle homogenates were subjected to ultrafiltration with Centricon 3 microconcentrators (molecular weight exclusion of 3000) at 3000g for 3 h.
Protein
Concentration
Determination
Protein concentrations were determined assay with BSA used as standard.
3 Abbreviations used: EDTA, vine serum albumin.
by the Pierce Micro
ethylenediaminetetraacetate;
BCA
BSA, bo-
383
384
BURGESS,
In Vitro Calcium Carbonate Precipitation
CAREY,
Assay
The precipitation assay involved the use of a Radiometer VIT 90 Titralab autotitration system. The instrument was operated in its pH stat mode with the pH set at 8.3 and 0.10 N NaOH as the titrant. The rate of calcium carbonate precipitation was equal to the rate of addition of titrant. The solutions used in this assay were A 1.0 M NaCl, 20 mM KCl, 20 mM CaClz and B: 20 mM NaHCOs (pH 8.5) and were stored at room temperature. This procedure was modified from that of Wheeler and Sikes (14). At the beginning of each set of assays, the burette of the pH stat was flushed with titrant and the instrument was calibrated with pH 7.0 and pH 10.0 buffer. After calibration, as well as between assays, the sample chamber and electrode were cleaned with water purified by a Milli-Q water purification system until the pH read approximately 5.5. Cleaning was facilitated by washing with 0.10 M HCl after the precipitation assays. For each assay, 25 ml of solution A was mixed with microliter volumes of cuticle homogenates followed by 25 ml of solution B, with continuous stirring. pH measurements were recorded after each addition. Assays were started immediately after the addition of solution B and continued for 60 to 1000 min. The assay temperature was maintained at 26.5”C in a water-jacketed sample container connected to a Lauda RM 20 refrigerated, circulating constant temperature bath (Brinkmann).
AND
OXENDINE
with 3000, 10,000, and 30,000 molecular weight exclusion limits, respectively. Concentrates were assayed for protein and were tested for their effects on the in vitro precipitation of calcium carbonate, as described above. Sephadex G-100 column (0.45 X (b) Gel filtration chromatography. 15 cm) was run at 4.0°C at 2.0 ml/h with 50 mM sodium phosphate buffer, pH 8.0, as the eluent. The column was calibrated with albumin (MW 67,000), ovalbumin (MW 43,000), chymotrypsinogen A (MW 25,000), and ribonuclease A (MW 13,700). Protein concentrations were routinely monitored by an in-line absorbance detector (LKB 2138 Uvicord S) set at 280 nm. Premolt cuticle homogenate which had been concentrated by ultrafiltration through a Centricon(10,000 molecular weight exclusion) was applied to the column (100 ~1,0.738 mg of protein). Fractions were collected (0.5 ml), assayed for protein by the Pierce Micro BCA protein assay, and pooled to obtain enough protein for calcium carbonate precipitation inhibition assays. Pools were centrifuged in Centricondevices. Concentrates were resuspended in Milli-Q water, and recentrifuged in Centricondevices a second time. This solvent exchange step was necessary since the in vitro calcium carbonate precipitation assay cannot be conducted in the presence of buffer. Concentrates were then tested for their effect on the inhibition of calcium carbonate precipitation in uitro.
RESULTS
AND
DISCUSSION
Dose-Response
In Vitro Calcium Carbonate Precipitation
The effect of various amounts of cuticle homogenates on in uitro calcium carbonate precipitation was determined by the assay described above. Premolt cuticle protein amounts were 0, 26.1, 51.4, and 103 ng. Postmolt cuticle protein amounts were 0, 26.1, 52.2, and 94 pg.
Figure 1 shows a typical curve from a control in vitro calcium carbonate precipitation assay. The assay was based on the fact that as calcium carbonate precipitates from a solution, the solution concentration of carbonate ion decreases (Eq. [l]).
Assay
Heat Treatment The effect of heat treatment of premolt cuticle homogenates on their ability to influence in vitro calcium carbonate precipitation was determined. Premolt cuticle homogenate (26.1 pg of protein in 3.2 ~1 water) was incubated at 57°C for 940 min, cooled to room temperature, and quantitatively added to the precipitation assay described above. A second premolt cuticle homogenate sample, which also contained 26.1 ag of protein, was heated under more rigorous conditions (at 118°C for 1433 min) prior to the precipitation assay. A Thermolyne Dry-Bath heating block was used for the heating steps.
Trypsin
Treatment
The effect of trypsin digestion of premolt cuticle homogenates on their ability to influence in vitro calcium carbonate precipitation was determined. Premolt cuticle homogenate (26.1 pg of protein in 3.2 ~1) was incubated with trypsin (1.13 pg in 3.2 pl of 20 mM NaHCOa buffer, pH 8.5) for 1700 min at 37°C in the Thermolyne Dry-Bath heating block.
Protein Controls
In solution, this carbonate ion is in equilibrium carbonate ion and hydrogen ion (Eq. [2]).
with both bi-
Therefore, as the solution concentration of carbonate decreased, the equilibrium is reestablished by the formation of more carbonate ion and hydrogen ion. The increased concentration of hydrogen ion was detected by the pH electrode of the pH stat.
3 2.5 E - 2 B i I.5
The effects of trypsin (100 pg in 284 pl of 20 mM sodium bicarbonate buffer, pH 8.5), myoglobin (100 pg in 255 pl in 50 mM sodium phosphate buffer, pH 8.0), and ovalbumin (100 pg in Milli-Q water) on the in vitro precipitation of calcium carbonate were determined in the assay described above.
Z J B
' 0.5 0
Determination
of Molecular
Weight
The molecular weight of the premolt inhibitory substance was determined by ultracentrifugation and gel filtration chromatography. Premolt cuticle homogenate samples were (a) Ultracentrifugation. centrifuged at 3000g in Centricon(3 h), -10 (2 h), and -30 (2 h) devices
0
50
loo
150
200 200
250
300
350
400
lime (mm)
FIG. 1. Effect of premolt and postmolt cuticle homogenates on the in vitro precipitation of calcium carbonate at pH 8.3; 26.1 ng of protein was used.
NOVEL
0
lo
20
30
40
50
PROTEIN
60
70
INHIBITS
80
93
lal
CALCIUM
110
amount of protein (micrograms)
CARBONATE
385
PRECIPITATION
alytic interaction between the proteins and the calcium ions or crystal nuclei. Postmolt cuticle homogenates also inhibited in vitro calcium carbonate precipitation in a dose-dependent manner. However, this inhibition was not as extensive as that with the premolt samples. For example, a postmolt sample which contained 94 ag of protein increased the nucleation time to only 107 min (Fig. 2). This inhibition was less than that caused by a premolt sample which contained only 26.1 pg of protein (nucleation time’ of 250 min). Trace amounts of the premolt inhibitory substance retained in the postmolt tissue may be responsible for the slight inhibition associated with the latter
FIG. 2. Premolt and postmolt cuticle protein dose-response curve for at pH 8.3. the inhibition of in uitro calcium carbonate precipitation
instrument automatically added sufficient 0.10 N sodium hydroxide to the solution to react with the additional hydrogen ion produced, and keep the pH of the solution at the set point of 8.3 (Eq. [3]).
The
The instrument plotted and recorded the volume of sodium hydroxide titrant consumed by the reaction over a period of time. The rate of this titration was directly proportional to the rate of calcium carbonate precipitation from the solution. Since 0.10 N sodium hydroxide was used as a titrant, 1.0 ml of this titrant was added for every 100 pmol of calcium carbonate which precipitated. To initiate the calcium carbonate precipitation process, calcium and carbonate ions interact with one another to form minute crystal nucleation centers. This nucleation involves diffusion and loss of ionic hydration layers. In control assays, the amount of time required for crystal nucleation was 8.7 + 1.3 min. This value was the average of three measurements + the standard deviation. After the nucleation phase was over, the rate of calcium carbonate precipitation increased dramatically. This precipitation phase was monitored by measuring the rate of titrant added, as well as by visually observing the accumulation of calcium carbonate in the sample. In all samples, the precipitation rates gradually increased and then decreased as the equilibrium between dissolved solute and precipitate was reached. Since plots of volume of titrant vs. time for the precipitation phases were sigmoidal in shape, they were difficult to quantitatively evaluate. Nucleation times were much more reliable and were used as in the following experiments as indices of inhibitory ability.
Determination of Protein Character of Inhibitor of in Vitro Calcium Carbonate Precipitation by Heat Treatment and Trypsin Treatment of Premolt Cuticle Homogenates Heat treatment of premolt cuticle homogenates mitigated but did not completely eliminate their inhibition of in vitro calcium carbonate precipitation (Fig. 3). Heat-treated (57”C, 940 min) premolt cuticle homogenate gave a nucleation time (157 min) less than that of the premolt control (250 min). Longer incubations at higher temperatures (118’C, 1433 min) were associated with even shorter nucleation times (55 min). Trypsin treatment was much more effective than heat treatment at diminishing inhibition of in vitro calcium carbonate precipitation (Fig. 4). However, even in this case, some inhibitory activity remained (nucleation time of 18.4 min). Both trypsin and high heat destroy active conformations of proteins. Heat treatment causes destruction of secondary and tertiary protein structure and trypsin treatment breaks proteins into peptide fragments at lysyl and arginyl residues. Since premolt cuticle homogenates, subjected to either treatment, lost some inhibitory activity, this activity must be associated with proteinaceous substances present in the premolt cuticle. However, neither trypsin nor heat treatment of premolt cuticle homogenates completely destroyed their ability to inhibit in vitro calcium carbonate precipitation. Therefore, inhibitor proteins, present in these homogenates, retained par-
Inhibition of in Vitro Calcium Carbonate Precipitation by Premolt and Postmolt Cuticle Homogenates Premolt cuticle homogenates inhibited in vitro calcium carbonate precipitation by increasing the nucleation time in a dose-dependent manner. The maximum amount of protein used, 103 Kg, resulted in a nucleation time of 673 min (Fig. 2). The magnitude of this effect with such a relatively small amount of protein and a large amount of calcium and carbonate ions suggests that the mechanism may involve a cat-
0
1 0
50
100
I50
200
250
MO
350
al
A50
500
time (mln)
FIG. 3. Effect of heat-treated premolt cuticle homogenates on the in uitro precipitation of calcium carbonate at pH 8.3.
BURGESS, CAREY, AND OXENDINE
386 3 2.5 g $ 2 i 1.5 B 3e ' D 0.5 0 0
50
loo
150
200
250
300
150
400
450
lime (mio)
FIG. 4. Effect of trypsin-treated premolt cuticle homogenates on the in vitro precipitation of calcium carbonate at pH 8.3.
tial activity after both heat denaturation into a random coil conformation and tryptic digestion into peptide fragments. It must be recognized that some proteins, which have been unfolded by heating, may refold upon cooling and, second, that some proteins are resistant to trypsin digestion. However, it can be speculated based on this retention of activity after these treatments that these inhibitor proteins may contain various surface acidic calcium-binding groups, as does the bone formation inhibitor protein, osteocalcin (15). The calcium binding abilities of these surface groups would be only partially dependent on cooperative interactions created by polypeptide chain folding. Alternatively, regions rich in acidic groups may be present within the inhibitor proteins, as has been postulated for oyster shell matrix proteins (16-20). After tryptic digestion, some peptide fragments would contain clusters of acidic groups and consequently have calcium binding ability. Experiments to determine if these inhibitor proteins are acidic and/or bind calcium are planned.
the lower molecular weight molecules. Experiments are planned to investigate this possibility. It is also possible that some of the smaller inhibitory proteins may have been blocked by other molecules from passing through the 30,000-Da membrane in the Centricondevice, resulting in some inhibitory activity in this concentrate. These molecular weight values have been supported by gel filtration chromatography experiments. Inhibitory activity peaks were found in two pooled column fractions, one which corresponded to a molecular weight range of lO,OOO-20,000 and another which corresponded to an approximate molecular weight of 75,000. Since these ranges have now been established, further gel filtration chromatography experiments with larger initial amounts of protein can be conducted. A more precise molecular weight can be determined in these experiments since pools of smaller volume will be used. These pools should contain enough protein to determine protein heterogeneity by electrophoretic analysis. In summary, our results suggest that water soluble proteins with molecular weights in at least two ranges (lO,OOO-20,000 and approximately 75,000), which are partially sensitive to heat and trypsin denaturation and are concentrated in the premolt cuticle of Callinectes sapidus, inhibit in vitro calcium carbonate precipitation at pH 8.3. This in vitro inhibition may be related to the ability of the crab to retain the demineralized nature of its premolt cuticle until postmolt. Definitive experiments which relate these in vitro and in viuo effects await purification of these inhibitory proteins. ACKNOWLEDGMENTS The authors greatly appreciate the assistance of Dr. Robert D. Roer of the Center for Marine Science Research and Department of Biology, UNCW, Wilmington, North Carolina, and Mr. Charles G. Miller, of Burroughs Wellcome, Research Triangle Park, North Carolina, in collecting and staging the crabs as well as initial preparation of cuticle tissue homogenates.
Protein Controls
REFERENCES
Neither myoglobin, trypsin, nor ovalbumin had an inhibitory effect on the in vitro inhibition of calcium carbonate precipitation. The nucleation time for calcium carbonate precipitation was 7.5 min in the presence of myoglobin, 7.4 min in the presence of trypsin, and 4.3 min in the presence of ovalbumin. These results suggest that the inhibitory effect of the premolt cuticle protein on in vitro calcium carbonate precipitation is a characteristic property of these inhibitor proteins and is not a nonspecific effect. The lack of inhibition by ovalbumin is particularly interesting since it is an acidic protein (PI 4.6).
1. Berman, A., Addadi, L., and Weiner, S. (1988) Nature (London) 331,546-548. 2. Carson, D. D., Farach, M. C., Earles, D. S., Decker, G. L., and Lennarx, W. J. (1985) Cell 41, 639-643. 3. Mintz, G. R., DeFrancesco, S., and Lennarz, W. J. (1981) J. Biol. Chem. 256, 13105-13111. 4. Swift, D. M., Sikes, C. S., and Wheeler, A. P. (1986) J. Erp. Zool. 240,65-73. 5. Durliat, M., and Vranckx, R. (1986) Comp. Biochem. Physiol. B 85, 264-274. Fischer, L. W., Hawkins, G. R., Tuross, N., and Termine, J. D. 6. (1987a) J. Biol. Chem. 262,9702-9708. 7. Howard, B., Mitchell, P. C. H., Ritchie, A., Simkiss, K., and Taylor, M. (1981) Biochem. J. 194, 507-511. 8. Kingsley, R. J., and Watabe, N. (1983) Comp. Biochem. Physiol. 76B,443-447. 9. Krampitz, G., Drolshagen, H., and Hotta, S. (1983b) Erperientia 39, 1104-1105. 10. Lowenstam, H. A., Traub, W., and Weiner, S. (1984) Paleobiology 10,268-279. 11. Marsh, M. E., and Sass, R. L. (1983) J. Exp. Zool. 226, 193-203.
Molecular
Weight
Inhibitory activity was observed in the concentrates from Centricons-3, -10, and -30 with 3000, 10,000, and 30,000 molecular weight exclusions, respectively. However, the specific inhibitory values, defined as minutes of nucleation per microgram of protein, were 0.635,1.34, and 0.445 min/pg, respectively. This suggests that some of these inhibitory proteins have molecular weights between 10,000 and 30,000 and that possibly there are other inhibitory proteins present with molecular weights above 30,000. These larger inhibitory proteins may be aggregates of
NOVEL
PROTEIN
INHIBITS
CALCIUM
12. Prince, C. W., Oosawa, T., Butler, WW. T., Tomana, M., Bhown, A. S., Bhown, M., and Schrohenloher, R. E. (1987) J. Biol. Chem. 262,2900-2907. 13. Roer, R. D., Burgess, S. K., Miller, C. G., and Dail, M. B. (1988) in Chemical Aspects of Regulation of Mineralization (Sikes, C. S., and Wheeler, A. P., Eds.), Univ. of South Alabama Publication Service, Mobile. 14. Wheeler, A. P., and Sikes, C. S. (1989) in Biomineralization: Chemical and Biochemical Perspectives (Mann, S., Webb, J., and Williams, R. J. P., Eds.) pp. 95-131, VCH Publishers, New York.
CARBONATE
PRECIPITATION
387
15. Williams, R. J. P., and Frausto da Silva, J. J. R. (1991) The Biological Chemistry of the Elements: The Inorganic Chemistry of Life, pp. 292-293, Clarendon Press, Oxford. 16. Weiner, S., and Traub, W. (1980) FEBS Lett. 111, 311-316. 17. Weiner, S., Talmon, Y., and Trauc, W. (1983) Znt. J. Biol. Mucromolecules 5, 325-328. 18. Weiner, S. (1984) Philos. Trans. R. Sot. B 304, 425-434. 19. Weiner, S., and Hood, L. (1975) Science 190, 987-989. 20. Wheeler, A. P., Rusenko, K. W., and Sikes, C. S. (1988) in Chemical Aspects of Regulation of Mineralization (Sikes, C. S., and Wheeler, A. P., Eds.), pp. 9-13, Univ. of South Alabama Publication Service, Mobile.
OF BIOCHEMISTRY
AND BIOPHYSICS
Vol. 297, No. 2, September, pp. 383-387, 1992
COMMUNICATION Novel Protein
Inhibits in Vitro Precipitation
of Calcium
Carbonate’
Sybil K. Burgesq2 Dana M. Carey, and Sharon L. Oxendine Department
of Chemistry,
Received January
University
of North
Carolina at Wilmington,
Wilmington,
North Carolina 28403
7,1992, and in revised form June 3,1992
Organic molecules both coexist and interact with inorganic crystal lattices in biomineralizing tissues. Mineral precipitation and crystal morphology are tightly regulated by the actions of these molecules. Polyacrylamide gel electrophoresis studies on water soluble extracts from the cuticle of Callinectes sapidus (Atlantic blue crab) reveal the presence, in unmineralized nascent premolt cuticle, of proteins which are absent in the mineralized postmolt cuticle. In the present studies, homogenates from both premolt and postmolt C. sapidus cuticles have been tested for their effect on the in vitro precipitation of calcium carbonate. The role of protein in this process was determined by heat pretreatment and trypsin pretreatment of the cuticle homogenates prior to the precipitation assay. The results from these experiments indicate that proteins, with molecular weights of approximately 75,000 and between 10,000 and 20,000, concentrated in the C. sapidus premolt cuticle, inhibit calcium carbonate precipitation in vitro. The inhibitory activity of these proteins appears to be a result of specific interactions since trypsin, myoglobin, and ovalbumin are not inhibitory. The presence of lower amounts of these inhibitory proteins in C. sapidus postmolt cuticle may be responsible for the subsequent mineralization of this tissue. 0 1992 Academic Press, Inc.
Calcium carbonate is the biomineral found in invertebrate skeletal structures as well as the octoconia (balance organ) of humans. Regulation of calcium carbonate precipitation has been the subject of investigation by a number of laboratories (l-13). The organisms used in these studies have included sea urchins (l-4), crayfish (5), humans (6), garden snails (7), coral (8), oysters (9), nautiluses (lo), clams (ll), rats (12), and crabs (13). 1 This work was supported by a Greenwall Foundation, Inc., Grant of Research Corporation and by the University of North Carolina at Wilmington’s Center for Marine Science Research. ‘To whom correspondence should be addressed at Department of Chemistry, UNCW, 601 S. College Road, Wilmington, NC 28403-3297. 0003-9861/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
In vitro experiments with Ucapugilator (the sand fiddler crab) suggest that control of cuticle mineralization resides within the tissue’s organic matrix components (13). Furthermore, it has been shown by polyacrylamide gel electrophoresis studies with Callinectes sapidus (Atlantic blue crab) cuticle that certain proteins which are present in the crab’s premolt cuticle are present to a much lesser extent in the animal’s postmolt cuticle (13). To determine whether the appearance of specific proteins, in C. sapidus cuticle, at particular stages in the crab’s molt cycle was related to cuticle biomineralization, we compared the effect of premolt and postmolt cuticle homogenates on the in vitro precipitation of calcium carbonate and partially characterized the inhibitory substance found in the premolt cuticle. EXPERIMENTAL
PROCEDURES
Materials C. sqidus crabs were obtained from a local commercial seafood market. EDTA: NaOH, NaCl, KCl, CaCl,, NaHCO,, HCl, NaHPO,, NaH2P04, and calibration buffers were purchased from Fisher. Centricon-3, -10, and -30 ultracentrifugation devices were purchased from Amicon. BSA and Pierce Micro BCA Protein Assay Reagent were purchased from Pierce. Trypsin and myoglobin were purchased from Sigma.
Preparation
of Cuticle
Tissue Homogenates
Pieces of cuticle were removed from the dorsobranchial area of the crab during various molt stages. These pieces were stripped of underlying hypodermal tissue, cut into small pieces, and placed in 0.1 M EDTA (pH 8.0). The tissue pieces were then homogenized for 5 min in an ice bath. The resulting homogenate was then agitated at 4°C for 40 h. The tissue slurry was then rehomogenized and centrifuged at 20,OOOgat 4°C for 6 h. Supernatants were frozen until needed. Upon thawing, the soluble cuticle homogenates were subjected to ultrafiltration with Centricon 3 microconcentrators (molecular weight exclusion of 3000) at 3000g for 3 h.
Protein
Concentration
Determination
Protein concentrations were determined assay with BSA used as standard.
3 Abbreviations used: EDTA, vine serum albumin.
by the Pierce Micro
ethylenediaminetetraacetate;
BCA
BSA, bo-
383
384
BURGESS,
In Vitro Calcium Carbonate Precipitation
CAREY,
Assay
The precipitation assay involved the use of a Radiometer VIT 90 Titralab autotitration system. The instrument was operated in its pH stat mode with the pH set at 8.3 and 0.10 N NaOH as the titrant. The rate of calcium carbonate precipitation was equal to the rate of addition of titrant. The solutions used in this assay were A 1.0 M NaCl, 20 mM KCl, 20 mM CaClz and B: 20 mM NaHCOs (pH 8.5) and were stored at room temperature. This procedure was modified from that of Wheeler and Sikes (14). At the beginning of each set of assays, the burette of the pH stat was flushed with titrant and the instrument was calibrated with pH 7.0 and pH 10.0 buffer. After calibration, as well as between assays, the sample chamber and electrode were cleaned with water purified by a Milli-Q water purification system until the pH read approximately 5.5. Cleaning was facilitated by washing with 0.10 M HCl after the precipitation assays. For each assay, 25 ml of solution A was mixed with microliter volumes of cuticle homogenates followed by 25 ml of solution B, with continuous stirring. pH measurements were recorded after each addition. Assays were started immediately after the addition of solution B and continued for 60 to 1000 min. The assay temperature was maintained at 26.5”C in a water-jacketed sample container connected to a Lauda RM 20 refrigerated, circulating constant temperature bath (Brinkmann).
AND
OXENDINE
with 3000, 10,000, and 30,000 molecular weight exclusion limits, respectively. Concentrates were assayed for protein and were tested for their effects on the in vitro precipitation of calcium carbonate, as described above. Sephadex G-100 column (0.45 X (b) Gel filtration chromatography. 15 cm) was run at 4.0°C at 2.0 ml/h with 50 mM sodium phosphate buffer, pH 8.0, as the eluent. The column was calibrated with albumin (MW 67,000), ovalbumin (MW 43,000), chymotrypsinogen A (MW 25,000), and ribonuclease A (MW 13,700). Protein concentrations were routinely monitored by an in-line absorbance detector (LKB 2138 Uvicord S) set at 280 nm. Premolt cuticle homogenate which had been concentrated by ultrafiltration through a Centricon(10,000 molecular weight exclusion) was applied to the column (100 ~1,0.738 mg of protein). Fractions were collected (0.5 ml), assayed for protein by the Pierce Micro BCA protein assay, and pooled to obtain enough protein for calcium carbonate precipitation inhibition assays. Pools were centrifuged in Centricondevices. Concentrates were resuspended in Milli-Q water, and recentrifuged in Centricondevices a second time. This solvent exchange step was necessary since the in vitro calcium carbonate precipitation assay cannot be conducted in the presence of buffer. Concentrates were then tested for their effect on the inhibition of calcium carbonate precipitation in uitro.
RESULTS
AND
DISCUSSION
Dose-Response
In Vitro Calcium Carbonate Precipitation
The effect of various amounts of cuticle homogenates on in uitro calcium carbonate precipitation was determined by the assay described above. Premolt cuticle protein amounts were 0, 26.1, 51.4, and 103 ng. Postmolt cuticle protein amounts were 0, 26.1, 52.2, and 94 pg.
Figure 1 shows a typical curve from a control in vitro calcium carbonate precipitation assay. The assay was based on the fact that as calcium carbonate precipitates from a solution, the solution concentration of carbonate ion decreases (Eq. [l]).
Assay
Heat Treatment The effect of heat treatment of premolt cuticle homogenates on their ability to influence in vitro calcium carbonate precipitation was determined. Premolt cuticle homogenate (26.1 pg of protein in 3.2 ~1 water) was incubated at 57°C for 940 min, cooled to room temperature, and quantitatively added to the precipitation assay described above. A second premolt cuticle homogenate sample, which also contained 26.1 ag of protein, was heated under more rigorous conditions (at 118°C for 1433 min) prior to the precipitation assay. A Thermolyne Dry-Bath heating block was used for the heating steps.
Trypsin
Treatment
The effect of trypsin digestion of premolt cuticle homogenates on their ability to influence in vitro calcium carbonate precipitation was determined. Premolt cuticle homogenate (26.1 pg of protein in 3.2 ~1) was incubated with trypsin (1.13 pg in 3.2 pl of 20 mM NaHCOa buffer, pH 8.5) for 1700 min at 37°C in the Thermolyne Dry-Bath heating block.
Protein Controls
In solution, this carbonate ion is in equilibrium carbonate ion and hydrogen ion (Eq. [2]).
with both bi-
Therefore, as the solution concentration of carbonate decreased, the equilibrium is reestablished by the formation of more carbonate ion and hydrogen ion. The increased concentration of hydrogen ion was detected by the pH electrode of the pH stat.
3 2.5 E - 2 B i I.5
The effects of trypsin (100 pg in 284 pl of 20 mM sodium bicarbonate buffer, pH 8.5), myoglobin (100 pg in 255 pl in 50 mM sodium phosphate buffer, pH 8.0), and ovalbumin (100 pg in Milli-Q water) on the in vitro precipitation of calcium carbonate were determined in the assay described above.
Z J B
' 0.5 0
Determination
of Molecular
Weight
The molecular weight of the premolt inhibitory substance was determined by ultracentrifugation and gel filtration chromatography. Premolt cuticle homogenate samples were (a) Ultracentrifugation. centrifuged at 3000g in Centricon(3 h), -10 (2 h), and -30 (2 h) devices
0
50
loo
150
200 200
250
300
350
400
lime (mm)
FIG. 1. Effect of premolt and postmolt cuticle homogenates on the in vitro precipitation of calcium carbonate at pH 8.3; 26.1 ng of protein was used.
NOVEL
0
lo
20
30
40
50
PROTEIN
60
70
INHIBITS
80
93
lal
CALCIUM
110
amount of protein (micrograms)
CARBONATE
385
PRECIPITATION
alytic interaction between the proteins and the calcium ions or crystal nuclei. Postmolt cuticle homogenates also inhibited in vitro calcium carbonate precipitation in a dose-dependent manner. However, this inhibition was not as extensive as that with the premolt samples. For example, a postmolt sample which contained 94 ag of protein increased the nucleation time to only 107 min (Fig. 2). This inhibition was less than that caused by a premolt sample which contained only 26.1 pg of protein (nucleation time’ of 250 min). Trace amounts of the premolt inhibitory substance retained in the postmolt tissue may be responsible for the slight inhibition associated with the latter
FIG. 2. Premolt and postmolt cuticle protein dose-response curve for at pH 8.3. the inhibition of in uitro calcium carbonate precipitation
instrument automatically added sufficient 0.10 N sodium hydroxide to the solution to react with the additional hydrogen ion produced, and keep the pH of the solution at the set point of 8.3 (Eq. [3]).
The
The instrument plotted and recorded the volume of sodium hydroxide titrant consumed by the reaction over a period of time. The rate of this titration was directly proportional to the rate of calcium carbonate precipitation from the solution. Since 0.10 N sodium hydroxide was used as a titrant, 1.0 ml of this titrant was added for every 100 pmol of calcium carbonate which precipitated. To initiate the calcium carbonate precipitation process, calcium and carbonate ions interact with one another to form minute crystal nucleation centers. This nucleation involves diffusion and loss of ionic hydration layers. In control assays, the amount of time required for crystal nucleation was 8.7 + 1.3 min. This value was the average of three measurements + the standard deviation. After the nucleation phase was over, the rate of calcium carbonate precipitation increased dramatically. This precipitation phase was monitored by measuring the rate of titrant added, as well as by visually observing the accumulation of calcium carbonate in the sample. In all samples, the precipitation rates gradually increased and then decreased as the equilibrium between dissolved solute and precipitate was reached. Since plots of volume of titrant vs. time for the precipitation phases were sigmoidal in shape, they were difficult to quantitatively evaluate. Nucleation times were much more reliable and were used as in the following experiments as indices of inhibitory ability.
Determination of Protein Character of Inhibitor of in Vitro Calcium Carbonate Precipitation by Heat Treatment and Trypsin Treatment of Premolt Cuticle Homogenates Heat treatment of premolt cuticle homogenates mitigated but did not completely eliminate their inhibition of in vitro calcium carbonate precipitation (Fig. 3). Heat-treated (57”C, 940 min) premolt cuticle homogenate gave a nucleation time (157 min) less than that of the premolt control (250 min). Longer incubations at higher temperatures (118’C, 1433 min) were associated with even shorter nucleation times (55 min). Trypsin treatment was much more effective than heat treatment at diminishing inhibition of in vitro calcium carbonate precipitation (Fig. 4). However, even in this case, some inhibitory activity remained (nucleation time of 18.4 min). Both trypsin and high heat destroy active conformations of proteins. Heat treatment causes destruction of secondary and tertiary protein structure and trypsin treatment breaks proteins into peptide fragments at lysyl and arginyl residues. Since premolt cuticle homogenates, subjected to either treatment, lost some inhibitory activity, this activity must be associated with proteinaceous substances present in the premolt cuticle. However, neither trypsin nor heat treatment of premolt cuticle homogenates completely destroyed their ability to inhibit in vitro calcium carbonate precipitation. Therefore, inhibitor proteins, present in these homogenates, retained par-
Inhibition of in Vitro Calcium Carbonate Precipitation by Premolt and Postmolt Cuticle Homogenates Premolt cuticle homogenates inhibited in vitro calcium carbonate precipitation by increasing the nucleation time in a dose-dependent manner. The maximum amount of protein used, 103 Kg, resulted in a nucleation time of 673 min (Fig. 2). The magnitude of this effect with such a relatively small amount of protein and a large amount of calcium and carbonate ions suggests that the mechanism may involve a cat-
0
1 0
50
100
I50
200
250
MO
350
al
A50
500
time (mln)
FIG. 3. Effect of heat-treated premolt cuticle homogenates on the in uitro precipitation of calcium carbonate at pH 8.3.
BURGESS, CAREY, AND OXENDINE
386 3 2.5 g $ 2 i 1.5 B 3e ' D 0.5 0 0
50
loo
150
200
250
300
150
400
450
lime (mio)
FIG. 4. Effect of trypsin-treated premolt cuticle homogenates on the in vitro precipitation of calcium carbonate at pH 8.3.
tial activity after both heat denaturation into a random coil conformation and tryptic digestion into peptide fragments. It must be recognized that some proteins, which have been unfolded by heating, may refold upon cooling and, second, that some proteins are resistant to trypsin digestion. However, it can be speculated based on this retention of activity after these treatments that these inhibitor proteins may contain various surface acidic calcium-binding groups, as does the bone formation inhibitor protein, osteocalcin (15). The calcium binding abilities of these surface groups would be only partially dependent on cooperative interactions created by polypeptide chain folding. Alternatively, regions rich in acidic groups may be present within the inhibitor proteins, as has been postulated for oyster shell matrix proteins (16-20). After tryptic digestion, some peptide fragments would contain clusters of acidic groups and consequently have calcium binding ability. Experiments to determine if these inhibitor proteins are acidic and/or bind calcium are planned.
the lower molecular weight molecules. Experiments are planned to investigate this possibility. It is also possible that some of the smaller inhibitory proteins may have been blocked by other molecules from passing through the 30,000-Da membrane in the Centricondevice, resulting in some inhibitory activity in this concentrate. These molecular weight values have been supported by gel filtration chromatography experiments. Inhibitory activity peaks were found in two pooled column fractions, one which corresponded to a molecular weight range of lO,OOO-20,000 and another which corresponded to an approximate molecular weight of 75,000. Since these ranges have now been established, further gel filtration chromatography experiments with larger initial amounts of protein can be conducted. A more precise molecular weight can be determined in these experiments since pools of smaller volume will be used. These pools should contain enough protein to determine protein heterogeneity by electrophoretic analysis. In summary, our results suggest that water soluble proteins with molecular weights in at least two ranges (lO,OOO-20,000 and approximately 75,000), which are partially sensitive to heat and trypsin denaturation and are concentrated in the premolt cuticle of Callinectes sapidus, inhibit in vitro calcium carbonate precipitation at pH 8.3. This in vitro inhibition may be related to the ability of the crab to retain the demineralized nature of its premolt cuticle until postmolt. Definitive experiments which relate these in vitro and in viuo effects await purification of these inhibitory proteins. ACKNOWLEDGMENTS The authors greatly appreciate the assistance of Dr. Robert D. Roer of the Center for Marine Science Research and Department of Biology, UNCW, Wilmington, North Carolina, and Mr. Charles G. Miller, of Burroughs Wellcome, Research Triangle Park, North Carolina, in collecting and staging the crabs as well as initial preparation of cuticle tissue homogenates.
Protein Controls
REFERENCES
Neither myoglobin, trypsin, nor ovalbumin had an inhibitory effect on the in vitro inhibition of calcium carbonate precipitation. The nucleation time for calcium carbonate precipitation was 7.5 min in the presence of myoglobin, 7.4 min in the presence of trypsin, and 4.3 min in the presence of ovalbumin. These results suggest that the inhibitory effect of the premolt cuticle protein on in vitro calcium carbonate precipitation is a characteristic property of these inhibitor proteins and is not a nonspecific effect. The lack of inhibition by ovalbumin is particularly interesting since it is an acidic protein (PI 4.6).
1. Berman, A., Addadi, L., and Weiner, S. (1988) Nature (London) 331,546-548. 2. Carson, D. D., Farach, M. C., Earles, D. S., Decker, G. L., and Lennarx, W. J. (1985) Cell 41, 639-643. 3. Mintz, G. R., DeFrancesco, S., and Lennarz, W. J. (1981) J. Biol. Chem. 256, 13105-13111. 4. Swift, D. M., Sikes, C. S., and Wheeler, A. P. (1986) J. Erp. Zool. 240,65-73. 5. Durliat, M., and Vranckx, R. (1986) Comp. Biochem. Physiol. B 85, 264-274. Fischer, L. W., Hawkins, G. R., Tuross, N., and Termine, J. D. 6. (1987a) J. Biol. Chem. 262,9702-9708. 7. Howard, B., Mitchell, P. C. H., Ritchie, A., Simkiss, K., and Taylor, M. (1981) Biochem. J. 194, 507-511. 8. Kingsley, R. J., and Watabe, N. (1983) Comp. Biochem. Physiol. 76B,443-447. 9. Krampitz, G., Drolshagen, H., and Hotta, S. (1983b) Erperientia 39, 1104-1105. 10. Lowenstam, H. A., Traub, W., and Weiner, S. (1984) Paleobiology 10,268-279. 11. Marsh, M. E., and Sass, R. L. (1983) J. Exp. Zool. 226, 193-203.
Molecular
Weight
Inhibitory activity was observed in the concentrates from Centricons-3, -10, and -30 with 3000, 10,000, and 30,000 molecular weight exclusions, respectively. However, the specific inhibitory values, defined as minutes of nucleation per microgram of protein, were 0.635,1.34, and 0.445 min/pg, respectively. This suggests that some of these inhibitory proteins have molecular weights between 10,000 and 30,000 and that possibly there are other inhibitory proteins present with molecular weights above 30,000. These larger inhibitory proteins may be aggregates of
NOVEL
PROTEIN
INHIBITS
CALCIUM
12. Prince, C. W., Oosawa, T., Butler, WW. T., Tomana, M., Bhown, A. S., Bhown, M., and Schrohenloher, R. E. (1987) J. Biol. Chem. 262,2900-2907. 13. Roer, R. D., Burgess, S. K., Miller, C. G., and Dail, M. B. (1988) in Chemical Aspects of Regulation of Mineralization (Sikes, C. S., and Wheeler, A. P., Eds.), Univ. of South Alabama Publication Service, Mobile. 14. Wheeler, A. P., and Sikes, C. S. (1989) in Biomineralization: Chemical and Biochemical Perspectives (Mann, S., Webb, J., and Williams, R. J. P., Eds.) pp. 95-131, VCH Publishers, New York.
CARBONATE
PRECIPITATION
387
15. Williams, R. J. P., and Frausto da Silva, J. J. R. (1991) The Biological Chemistry of the Elements: The Inorganic Chemistry of Life, pp. 292-293, Clarendon Press, Oxford. 16. Weiner, S., and Traub, W. (1980) FEBS Lett. 111, 311-316. 17. Weiner, S., Talmon, Y., and Trauc, W. (1983) Znt. J. Biol. Mucromolecules 5, 325-328. 18. Weiner, S. (1984) Philos. Trans. R. Sot. B 304, 425-434. 19. Weiner, S., and Hood, L. (1975) Science 190, 987-989. 20. Wheeler, A. P., Rusenko, K. W., and Sikes, C. S. (1988) in Chemical Aspects of Regulation of Mineralization (Sikes, C. S., and Wheeler, A. P., Eds.), pp. 9-13, Univ. of South Alabama Publication Service, Mobile.
