- Email: [email protected]
Crystals containing a single subunit type of Panulirus interruptus haemocyanin
J. Mol. Biol. (1983) 168, 197-201
Crystals Containing a Single Subunit Type of Panulirus interruptus Haemocyanin Crystals have been grown of the 94K subunit of Panulirus interruptus haemocyanin. The best crystals were obtained by the "two-layer liquid diffusion technique" with ammonium sulphate solutions as precipitating agent. The crystals have space group P6322 with cell dimensions a = b = 195"6A and c = 158"0A. The diffraction pattern extends to a resolution of approximately 3"0 A resolution. The unit cell may contain two hexameric molecules with 32 symmetry, i.e. one subunit per asymmetric unit. Haemocyanins are the copper-containing oxygen transp()rt proteins occurring in a large number of invertebrates. Detailed three-dimensional structural knowledge of these molecules is of interest for a number of quite different reasons. First, a comparison of the oxygen binding centres in haemoglobin, haemerythrin and haemocyanin reveals the three different modes by which nature has solved the problem of oxygen transport by carrier molecules. Second, the haemocyanins belong, together with copper containing enzymes like laccase, ceruloplasmin, ascorbate oxidase and tyrosinase, to the group of 'type 3' copper proteins (Urbaeh, 1981; Reinhammar, 1983). The electron paramagnetic resonance-silent binuclear copper site of this group of proteins has attracted great attention from spectroscopists (Urbaeh, 1981; Lontie & Witters, 1981; Reinhammar, 1983; Brown et al., 1980; Co et al., 1981; Co & Hodgson, 1981). In particular the spectrum of oxytyrosinase is remarkably similar to that of oxyhaemocyanin (Schoot-Uiterkamp & Mason, 1973; Himmelwright et al., 1980). Third, bioinorganic chemists are attempting to synthesize low molecular weight molecules that mimic the reversible oxygen binding properties of the haemocyanins (cf. Urbach, 1981; Hendriks et al., 1982; Birker & Reedijk, 1983). The haemocyanin of Panulirus interruptus is one of the smallest haemocyanins known. It is a hexamer with a molecular weight of about 500,000 (Kuiper et al., 1975). This hexamer appears to be a common building block of arthropodan haemocyanins, which occur as a variety of "multi-hexameric" aggregates (Van Bruggen et al., 1981). The subunits of P. interruptus haemocyanin show a marked heterogeneity (Van den Berg et al., 1977; Van Eerd & Folkerts, 1981; Folkerts & Van Eerd, 1981). Three major types of subunits have been described (Van Eerd & Folkerts, 1981) with molecular weights of 80,000, 90,000 and 94,000, hereafter called the 80K, 90K and 94K subunits. The 90K and 94K subunits are very similar in sequence, , but the 80K subunit deviates considerably from the two other types of subunits (Van Eerd & Folkerts, 1981). Other evidence exists that the molecular weight of the hexamerie molecule is 450,000, i.e. a molecular weight of 75,000 per subunit 197 0022-2836/83/210197-05 $03.00/0 9 1983 AcademicPress Inc. (London)Ltd.
198
W.P.J.
GAYKEMA ET. AL.
(Kuiper et al., 1975). Therefore, an intermediate value of 500,000 will be used in this letter. Kuiper et al. (1975) described P. interruptu8 haemocyanin crystals obtained from solutions containing all three types of subunits. These monoclinic crystals have recently resulted in an electron density distribution at 5 A resolution (Van Schaick et al., 1982), which revealed the symmetry of the haemocyanin hexamer and the shape of the subunits. At 4"0 A resolution the positions of both copper ions per subunit could be determined (Gaykema et al., 1983). These monoclinic crystals appeared, upon dissolving, to contain a negligible amount of the 80K subunit and the ratio of the content of 90K and 94K subunits was approximately 1:1 (Van Eerd, unpublished results). This means that, in certain regions of the electron density at higher resolution, some uncertainties may arise due to differences in length and in the nature of the amino acid sidechains of the 90K and 94K subunits. Therefore, it was considered desirable to obtain crystals containing only one type of subunit. As considerable progress has been made in elucidating the sequence of the 94K subunit (Vereyken et al., 1982 ; Vereyken & Soeter, unpublished results), it was most logical to attempt to obtain good quality crystals of this subunit. Initially, we tried to crystallize the 94K component using conditions similar to those used for the successful crystallization of "mixed" haemocyanin (Kuiper et al., 1975). This at first appeared to be quite a successful approach: large, wellshaped crystals of the 94K component were obtained under a wide variety of conditions. A typical crystallization experiment consisted of the following steps. (1) Freeze-dried 94K haemocyanin, purified as described by Van Eerd & Folkerts (1981), was dialyzed against 50 mM-Tris" HCI buffer (pH 7"2), l0 mM-CaC12, in order to remove excess sucrose, which had been added as stabilizing agent. (2) Dialysis of the haemoeyanin solution against 10 mM-sodium acetate/acetate buffer (pH 5"5). During this step a slight precipitate often developed, which was removed by centrifugation. The haemocyanin concentration was approximately 7-5 mg/ml. (3) Dialysis at 4~ of the haemocyanin solution, in samples of about 0"5 ml, against 25 ml of 0-1 M-sodium acetate/acetate buffer (pH 3"8 to 5). (4) After two weeks the dialysis was stopped and the vials were kept sealed at 4~ Crystals developed within a few weeks or months. Despite the beautiful shape and size of the crystals, they appeared to be useless for high-resolution studies because of considerable disorder. The crystals were probably orthorhombic, with cell dimensions of 123 A x 196 A x 209 A. The disorder prevented a determination of the space group. Numerous attempts were made to eliminate the disorder of the crystals, mostly by adding various components to the dialysis solution in step (3) above, without success, however. Next, a search for suitable crystallization conditions at higher pH values and higher concentrations of precipitating agents was conducted. This yielded two quite different sets of circumstances under which large crystals developed. The crystals were grown by the "two-layer liquid diffusion technique", which has given useful results for a number of proteins in our laboratory: subtilisin Novo (Drenth & Hol, 1967), bovine liver rhodanese (Drenth & Smit, 1971), p-hydroxy-
LETTERS TO THE EDITOR
199
benzoate hydroxylase (Drenth et al., 1975), several phospholipases A 2 (Drenth et al., 1976; Dijkstra et al., 1978,1982) and lipoamide dehydrogenase (Schierbeek et al., 1983). After two initial dialysis steps as described above, 15 t~l protein solution was introduced into a capillary tube with a diameter of about 1"4 mm. Subsequently, 15 ~l of a second solution was very carefully layered on top of the protein solution. Beautiful crystals were obtained using the following compositions for this second layer: (i) 12"5 to 15~/o polyethylene glycol (pH 6"0), 5 mM-CaC12; or (ii) 50 to 65~/o saturated ammonium sulphate (pH 5"2 to 5"5), 5 mM-CaC12. At room temperature slightly better crystals were obtained than at 4~ Dimensions of the hexagonal crystals were up to 0"3 mm • 0"3 mm x l'0 mm. The molecular weight of polyethylene glycol was unimportant. The "polyethylene glycol crystals" appeared to be very fragile. Nevertheless, from a number of precession photographs the space group could be determined as being P6322, with cell dimensions: a - - b - - 196"3 A and c - 159"0 A. The reflections do not go beyond 4 A resolution. The "ammonium sulphate crystals" were much more stable although the space group was the same, P6322, and the cell dimensions were nearly identical: a -- b --- 195"6 A and c -- 158"0 A. The reflections go out to approximately 3"0 A resolution. They can be transferred to ammonium sulphate solutions up to pH 8"0 without losing their diffracting power. Soaking in 0"75 mM-o-mercury phenol, one of the successful heavy-atom derivatives for the monoclinic "mixture" crystals (Van Schaick et al., 1982), produced promising changes in intensities. Also, soaking crystals in solutions of 5 to 10 mM-ErC13 gave small, but significant, intensity changes. This may be helpful in determining the position of crucial calcium ions, as it is known that erbium ions can replace a subset of the calcium ions bound by this haemocyanin (Kuiper et al., 1981). The packing of the haemocyanin hexamers, which have point group 32 (Van Schaick et al., 1982), in the hexagonal crystals is probably quite "loose". With two hexamers per unit cell and one per asymmetric unit, each hexamer could be nicely placed at the two positions in the unit cell having 32 point symmetry. The molecular centres are then situated at (1/3, 2/3, 1/4) and (2/3, 1/3, 3/4). If a similar shape is assumed for the hexamer as that in the monoclinic crystals (Van Schaick et al., 1982), the molecules would just touch each other in such an arrangement. Therefore, this seems a likely mode of packing. Assuming a molecular weight of 500,000, the value for Vm is then approximately 4"9 A 3 per dalton. This is considerably higher than the average value for protein crystals of about 2"6 A 3 per dalton (Matthews, 1968), so other packing modes or hexamers with other subunit arrangements may need to be considered. From the above it can be concluded that the P6322 haemocyanin crystals, grown from ammonium sulphate solutions and containing a single subunit type only, would be very interesting material for further structural studies at reasonably high resolution. We are indebted to Drs J. Drenth, N. M. Soeter and J. J, Beintema for stimulating and critical discussions, and to Mr E. H. Schwander for technical assistance. This research was
200
W.P.J.
GAYKEMA ET. AL.
supported by the Netherlands Foundation for Chemical Research (SON} with financial aid from the Netherlands Organisation for. the Advancement of Pure Research (ZWO). Biomoleeular Study Center (BIOS) Department of Chemistry University of Groningen Nijenborgh 16 9747 AG Groningen The Netherlands Received 15 March 1983
W. P. J. GAYKEMA H. GROENDIJK G. DOORTEN J. M. VEI~.EYKEN W. G. J. HOL
REFERENCES Birker, P. J. M. W. L. & Reedijk, J. (1983). In Biochemical and Inorganic Perspectives in Copper Coordination Chemistry (Karlin, K. D. & Zubieta, J., eds), Adenine Press, Albany, New York, in the press. Brown, J. M., Powers, L., Kineaid, B., Larrabee, J. A. & Spiro, T. G. (1980). J. Amer. Chem. Soc. 102, 4210-4216. 9Co, M. S. & Hodgson, K. O. (1981). J. Amer. Chem. Soc. 103, 3200-3201. Co, M. S., Hodgson, K. 0., Eeeles, K. & Lontie, R. (1981). J. Amer. Chem. Soc. 103, 984-986. Dijkstra, B. W., Drenth, J., Kalk, K. H. & Vandermaelen, Ph.J. (1978). J. Mol. Biol. 124, 53-60. Dijkstra, B. W., Van Nes, G. J. H., Kalk, K. H., Brandenburg, N. P., Hol, W. G. J. & Drenth, J. (1982). Acta Crystallogr. sect. B, 38, 793-799. Drenth, J. & Hol, W. G. J. (1967). J. Mol. Biol. 28, 543-544. Drenth, J. & Smit, J. D. G. (1971). Biochem. Biophys. Res. Commun. 45, 1320-1321. Drenth, J., Hol, W. G. J. & Wierenga, R. K. (1975). J. Biol. Chem. 250, 5268-5269. Drenth, J., Enzing, C. M., Kalk, K. H. & Vessies, J. C. A. (1976). Nature (London), 264, 373-377. Folkerts, A. & Van Eerd, J. P. (1981). In Invertebrate Oxygen Binding Proteins (Lamy, J. & Lamy, J., eds), pp. 215-225, Marcel Dekker, New York. Gaykema, W. P. J., Van Schaick, E. J. M., Schutter, W. G. & Hol, W. G. J. {1983). Chemica Scripta, 21, 19-23. Hendriks, H. M. J., Birker, P. J. M. W. L., Van Rijn, J., Versehoor, G. C. & Reedijk, J. {1982). J. Amer. Chem. Soc. 104, 3607-3617. Himmelwright, R. S., Eiekman, N. C., LuBien, C. D., Lerch, K. & Solomon, E. I. (1980). J. Amer. Chem. Soc. 102, 7339-7344. Kuiper, H. A., Gaastra, W., Beintema, J. J., Van Bruggen, E. F. J., Sehepman, A. M. H. & Drenth, J. (1975). J. Mol. Biol. 99, 619-629. Kuiper, H. A., Zolla, L., Finazzi-Agro, A. & Brunori, M. (1981). J. Mol. Biol. 149, 805-812. Lontie, R. & Witters, R. {1981). In Metal Ions in Biological Systems, vol. 13, Copper Proteins (Siegel, H., ed.), pp. 229-258, Marcel Dekker, New York. Matthews, B. W. (1968). J. Mol. Biol. 33, 491-497. Reinhammar, B. (1983). In Coordination Chemistry of Metalloenzymes in Hydrolytic and Oxidative Processes (Reidel, D., ed.), Dordrecht, Holland, in the press. Schierbeek, A. J., Van der Laan, J. M., Groendijk, H., Wierenga, R. K. & Drenth, J. (1983). J. Mol. Biol.,165, 563-564. Sehoot-Uiterkamp, A. J. M. & Mason, H. S. (1973). Proc. Nat. Acad. Sci., U.S.A. 70, 993-996. Urbach, F. L. (1981). In Metal Ions in Biological Systems, vol. 13, Copper Proteins (Siegel, H., ed.), pp. 73-115, Marcel Dekker, New York. Van Bruggen, E. F. J., Schutter, W. G., Van Breemen, J. F. L., Bijholt, M. M. C. & Wichertjes, T. (1981). In Electron Microscopy of Proteins (Harris, M., ed.), vol. 1, pp. 1-38, Academic Press, New York, London.
LETTERS TO THE EDITOR
201
Van den Berg, A. A., Gaastra, W. & Kuiper, H. A. (1977). In Structure and Function of Hemocyanin (Bannister, J. V., ed.), pp. 6-12, Springer-Verlag, Berlin. Van Eerd, J. P. & Folkerts, A. (1981). In Invertebrate Oxygen Binding Proteins (Lamy, J. & Lamy, J., eds), pp. 139-149, Marcel Dekker, New York. Van Schaick, E. J. M., Sehutter, W. G., Gaykema, W. P. J., Schepman, A. M. H. & Hol, W. G. J. (1982). J. Mol. Biol. 158,457-485. Vereyken, J. M., Schwander, E. H., Soeter, N. M. & Beintema, J. J. (1982). Eur. J. Biochem. 123, 283-289.
Edited by S. Brenner
Crystals Containing a Single Subunit Type of Panulirus interruptus Haemocyanin Crystals have been grown of the 94K subunit of Panulirus interruptus haemocyanin. The best crystals were obtained by the "two-layer liquid diffusion technique" with ammonium sulphate solutions as precipitating agent. The crystals have space group P6322 with cell dimensions a = b = 195"6A and c = 158"0A. The diffraction pattern extends to a resolution of approximately 3"0 A resolution. The unit cell may contain two hexameric molecules with 32 symmetry, i.e. one subunit per asymmetric unit. Haemocyanins are the copper-containing oxygen transp()rt proteins occurring in a large number of invertebrates. Detailed three-dimensional structural knowledge of these molecules is of interest for a number of quite different reasons. First, a comparison of the oxygen binding centres in haemoglobin, haemerythrin and haemocyanin reveals the three different modes by which nature has solved the problem of oxygen transport by carrier molecules. Second, the haemocyanins belong, together with copper containing enzymes like laccase, ceruloplasmin, ascorbate oxidase and tyrosinase, to the group of 'type 3' copper proteins (Urbaeh, 1981; Reinhammar, 1983). The electron paramagnetic resonance-silent binuclear copper site of this group of proteins has attracted great attention from spectroscopists (Urbaeh, 1981; Lontie & Witters, 1981; Reinhammar, 1983; Brown et al., 1980; Co et al., 1981; Co & Hodgson, 1981). In particular the spectrum of oxytyrosinase is remarkably similar to that of oxyhaemocyanin (Schoot-Uiterkamp & Mason, 1973; Himmelwright et al., 1980). Third, bioinorganic chemists are attempting to synthesize low molecular weight molecules that mimic the reversible oxygen binding properties of the haemocyanins (cf. Urbach, 1981; Hendriks et al., 1982; Birker & Reedijk, 1983). The haemocyanin of Panulirus interruptus is one of the smallest haemocyanins known. It is a hexamer with a molecular weight of about 500,000 (Kuiper et al., 1975). This hexamer appears to be a common building block of arthropodan haemocyanins, which occur as a variety of "multi-hexameric" aggregates (Van Bruggen et al., 1981). The subunits of P. interruptus haemocyanin show a marked heterogeneity (Van den Berg et al., 1977; Van Eerd & Folkerts, 1981; Folkerts & Van Eerd, 1981). Three major types of subunits have been described (Van Eerd & Folkerts, 1981) with molecular weights of 80,000, 90,000 and 94,000, hereafter called the 80K, 90K and 94K subunits. The 90K and 94K subunits are very similar in sequence, , but the 80K subunit deviates considerably from the two other types of subunits (Van Eerd & Folkerts, 1981). Other evidence exists that the molecular weight of the hexamerie molecule is 450,000, i.e. a molecular weight of 75,000 per subunit 197 0022-2836/83/210197-05 $03.00/0 9 1983 AcademicPress Inc. (London)Ltd.
198
W.P.J.
GAYKEMA ET. AL.
(Kuiper et al., 1975). Therefore, an intermediate value of 500,000 will be used in this letter. Kuiper et al. (1975) described P. interruptu8 haemocyanin crystals obtained from solutions containing all three types of subunits. These monoclinic crystals have recently resulted in an electron density distribution at 5 A resolution (Van Schaick et al., 1982), which revealed the symmetry of the haemocyanin hexamer and the shape of the subunits. At 4"0 A resolution the positions of both copper ions per subunit could be determined (Gaykema et al., 1983). These monoclinic crystals appeared, upon dissolving, to contain a negligible amount of the 80K subunit and the ratio of the content of 90K and 94K subunits was approximately 1:1 (Van Eerd, unpublished results). This means that, in certain regions of the electron density at higher resolution, some uncertainties may arise due to differences in length and in the nature of the amino acid sidechains of the 90K and 94K subunits. Therefore, it was considered desirable to obtain crystals containing only one type of subunit. As considerable progress has been made in elucidating the sequence of the 94K subunit (Vereyken et al., 1982 ; Vereyken & Soeter, unpublished results), it was most logical to attempt to obtain good quality crystals of this subunit. Initially, we tried to crystallize the 94K component using conditions similar to those used for the successful crystallization of "mixed" haemocyanin (Kuiper et al., 1975). This at first appeared to be quite a successful approach: large, wellshaped crystals of the 94K component were obtained under a wide variety of conditions. A typical crystallization experiment consisted of the following steps. (1) Freeze-dried 94K haemocyanin, purified as described by Van Eerd & Folkerts (1981), was dialyzed against 50 mM-Tris" HCI buffer (pH 7"2), l0 mM-CaC12, in order to remove excess sucrose, which had been added as stabilizing agent. (2) Dialysis of the haemoeyanin solution against 10 mM-sodium acetate/acetate buffer (pH 5"5). During this step a slight precipitate often developed, which was removed by centrifugation. The haemocyanin concentration was approximately 7-5 mg/ml. (3) Dialysis at 4~ of the haemocyanin solution, in samples of about 0"5 ml, against 25 ml of 0-1 M-sodium acetate/acetate buffer (pH 3"8 to 5). (4) After two weeks the dialysis was stopped and the vials were kept sealed at 4~ Crystals developed within a few weeks or months. Despite the beautiful shape and size of the crystals, they appeared to be useless for high-resolution studies because of considerable disorder. The crystals were probably orthorhombic, with cell dimensions of 123 A x 196 A x 209 A. The disorder prevented a determination of the space group. Numerous attempts were made to eliminate the disorder of the crystals, mostly by adding various components to the dialysis solution in step (3) above, without success, however. Next, a search for suitable crystallization conditions at higher pH values and higher concentrations of precipitating agents was conducted. This yielded two quite different sets of circumstances under which large crystals developed. The crystals were grown by the "two-layer liquid diffusion technique", which has given useful results for a number of proteins in our laboratory: subtilisin Novo (Drenth & Hol, 1967), bovine liver rhodanese (Drenth & Smit, 1971), p-hydroxy-
LETTERS TO THE EDITOR
199
benzoate hydroxylase (Drenth et al., 1975), several phospholipases A 2 (Drenth et al., 1976; Dijkstra et al., 1978,1982) and lipoamide dehydrogenase (Schierbeek et al., 1983). After two initial dialysis steps as described above, 15 t~l protein solution was introduced into a capillary tube with a diameter of about 1"4 mm. Subsequently, 15 ~l of a second solution was very carefully layered on top of the protein solution. Beautiful crystals were obtained using the following compositions for this second layer: (i) 12"5 to 15~/o polyethylene glycol (pH 6"0), 5 mM-CaC12; or (ii) 50 to 65~/o saturated ammonium sulphate (pH 5"2 to 5"5), 5 mM-CaC12. At room temperature slightly better crystals were obtained than at 4~ Dimensions of the hexagonal crystals were up to 0"3 mm • 0"3 mm x l'0 mm. The molecular weight of polyethylene glycol was unimportant. The "polyethylene glycol crystals" appeared to be very fragile. Nevertheless, from a number of precession photographs the space group could be determined as being P6322, with cell dimensions: a - - b - - 196"3 A and c - 159"0 A. The reflections do not go beyond 4 A resolution. The "ammonium sulphate crystals" were much more stable although the space group was the same, P6322, and the cell dimensions were nearly identical: a -- b --- 195"6 A and c -- 158"0 A. The reflections go out to approximately 3"0 A resolution. They can be transferred to ammonium sulphate solutions up to pH 8"0 without losing their diffracting power. Soaking in 0"75 mM-o-mercury phenol, one of the successful heavy-atom derivatives for the monoclinic "mixture" crystals (Van Schaick et al., 1982), produced promising changes in intensities. Also, soaking crystals in solutions of 5 to 10 mM-ErC13 gave small, but significant, intensity changes. This may be helpful in determining the position of crucial calcium ions, as it is known that erbium ions can replace a subset of the calcium ions bound by this haemocyanin (Kuiper et al., 1981). The packing of the haemocyanin hexamers, which have point group 32 (Van Schaick et al., 1982), in the hexagonal crystals is probably quite "loose". With two hexamers per unit cell and one per asymmetric unit, each hexamer could be nicely placed at the two positions in the unit cell having 32 point symmetry. The molecular centres are then situated at (1/3, 2/3, 1/4) and (2/3, 1/3, 3/4). If a similar shape is assumed for the hexamer as that in the monoclinic crystals (Van Schaick et al., 1982), the molecules would just touch each other in such an arrangement. Therefore, this seems a likely mode of packing. Assuming a molecular weight of 500,000, the value for Vm is then approximately 4"9 A 3 per dalton. This is considerably higher than the average value for protein crystals of about 2"6 A 3 per dalton (Matthews, 1968), so other packing modes or hexamers with other subunit arrangements may need to be considered. From the above it can be concluded that the P6322 haemocyanin crystals, grown from ammonium sulphate solutions and containing a single subunit type only, would be very interesting material for further structural studies at reasonably high resolution. We are indebted to Drs J. Drenth, N. M. Soeter and J. J, Beintema for stimulating and critical discussions, and to Mr E. H. Schwander for technical assistance. This research was
200
W.P.J.
GAYKEMA ET. AL.
supported by the Netherlands Foundation for Chemical Research (SON} with financial aid from the Netherlands Organisation for. the Advancement of Pure Research (ZWO). Biomoleeular Study Center (BIOS) Department of Chemistry University of Groningen Nijenborgh 16 9747 AG Groningen The Netherlands Received 15 March 1983
W. P. J. GAYKEMA H. GROENDIJK G. DOORTEN J. M. VEI~.EYKEN W. G. J. HOL
REFERENCES Birker, P. J. M. W. L. & Reedijk, J. (1983). In Biochemical and Inorganic Perspectives in Copper Coordination Chemistry (Karlin, K. D. & Zubieta, J., eds), Adenine Press, Albany, New York, in the press. Brown, J. M., Powers, L., Kineaid, B., Larrabee, J. A. & Spiro, T. G. (1980). J. Amer. Chem. Soc. 102, 4210-4216. 9Co, M. S. & Hodgson, K. O. (1981). J. Amer. Chem. Soc. 103, 3200-3201. Co, M. S., Hodgson, K. 0., Eeeles, K. & Lontie, R. (1981). J. Amer. Chem. Soc. 103, 984-986. Dijkstra, B. W., Drenth, J., Kalk, K. H. & Vandermaelen, Ph.J. (1978). J. Mol. Biol. 124, 53-60. Dijkstra, B. W., Van Nes, G. J. H., Kalk, K. H., Brandenburg, N. P., Hol, W. G. J. & Drenth, J. (1982). Acta Crystallogr. sect. B, 38, 793-799. Drenth, J. & Hol, W. G. J. (1967). J. Mol. Biol. 28, 543-544. Drenth, J. & Smit, J. D. G. (1971). Biochem. Biophys. Res. Commun. 45, 1320-1321. Drenth, J., Hol, W. G. J. & Wierenga, R. K. (1975). J. Biol. Chem. 250, 5268-5269. Drenth, J., Enzing, C. M., Kalk, K. H. & Vessies, J. C. A. (1976). Nature (London), 264, 373-377. Folkerts, A. & Van Eerd, J. P. (1981). In Invertebrate Oxygen Binding Proteins (Lamy, J. & Lamy, J., eds), pp. 215-225, Marcel Dekker, New York. Gaykema, W. P. J., Van Schaick, E. J. M., Schutter, W. G. & Hol, W. G. J. {1983). Chemica Scripta, 21, 19-23. Hendriks, H. M. J., Birker, P. J. M. W. L., Van Rijn, J., Versehoor, G. C. & Reedijk, J. {1982). J. Amer. Chem. Soc. 104, 3607-3617. Himmelwright, R. S., Eiekman, N. C., LuBien, C. D., Lerch, K. & Solomon, E. I. (1980). J. Amer. Chem. Soc. 102, 7339-7344. Kuiper, H. A., Gaastra, W., Beintema, J. J., Van Bruggen, E. F. J., Sehepman, A. M. H. & Drenth, J. (1975). J. Mol. Biol. 99, 619-629. Kuiper, H. A., Zolla, L., Finazzi-Agro, A. & Brunori, M. (1981). J. Mol. Biol. 149, 805-812. Lontie, R. & Witters, R. {1981). In Metal Ions in Biological Systems, vol. 13, Copper Proteins (Siegel, H., ed.), pp. 229-258, Marcel Dekker, New York. Matthews, B. W. (1968). J. Mol. Biol. 33, 491-497. Reinhammar, B. (1983). In Coordination Chemistry of Metalloenzymes in Hydrolytic and Oxidative Processes (Reidel, D., ed.), Dordrecht, Holland, in the press. Schierbeek, A. J., Van der Laan, J. M., Groendijk, H., Wierenga, R. K. & Drenth, J. (1983). J. Mol. Biol.,165, 563-564. Sehoot-Uiterkamp, A. J. M. & Mason, H. S. (1973). Proc. Nat. Acad. Sci., U.S.A. 70, 993-996. Urbach, F. L. (1981). In Metal Ions in Biological Systems, vol. 13, Copper Proteins (Siegel, H., ed.), pp. 73-115, Marcel Dekker, New York. Van Bruggen, E. F. J., Schutter, W. G., Van Breemen, J. F. L., Bijholt, M. M. C. & Wichertjes, T. (1981). In Electron Microscopy of Proteins (Harris, M., ed.), vol. 1, pp. 1-38, Academic Press, New York, London.
LETTERS TO THE EDITOR
201
Van den Berg, A. A., Gaastra, W. & Kuiper, H. A. (1977). In Structure and Function of Hemocyanin (Bannister, J. V., ed.), pp. 6-12, Springer-Verlag, Berlin. Van Eerd, J. P. & Folkerts, A. (1981). In Invertebrate Oxygen Binding Proteins (Lamy, J. & Lamy, J., eds), pp. 139-149, Marcel Dekker, New York. Van Schaick, E. J. M., Sehutter, W. G., Gaykema, W. P. J., Schepman, A. M. H. & Hol, W. G. J. (1982). J. Mol. Biol. 158,457-485. Vereyken, J. M., Schwander, E. H., Soeter, N. M. & Beintema, J. J. (1982). Eur. J. Biochem. 123, 283-289.
Edited by S. Brenner