- Email: [email protected]
Tadpole Xenopus laevis hemoglobin correlation between structure and functional properties
J. Mol. Bid.
(1985) 181, 327-329
Tadpole Xenopus Zaevis Hemoglobin Correlation Between Structure and Functional Properties Perutz $ Brunori (1982) proposed that the COOH-terminal His and Ser F9 of the P-chains of fish and amphibian hemoglobins are responsible for their Root effect and part of their alkaline Bohr effect. Analysis of the kinetics of carbon monoxide binding by hemoglobin from the tadpole of Xenopus laevis supports that model and suggestsan explanation for t’hr absence of an alkaline Bohr effect in many aquatic Anura and Urodela.
Tn most vertebrates the globins synthesized during embryonic life differ from those of the adult. The switch of the globin genes has been thoroughly investigated in Anurans, such as Xenopus laevis, and has provided a model of the regulatory mechanisms of cell- and stage-relat’ed gene expression. The primary structures of the major component of both the adult and larval hemoglobins of X. laevis have been determined from their complementary DNA sequences (Kay et al., 1980; Williams et al., 1980: Banville et al., 1983; Andres et ccl., 1984). Knowledge of the sequence offers an opportunity for studying the relationships between the structure of the hemoglobins and their physiological function. The adult X. Zaevis hemoglobin exhibits the socalled Root effect (Perutz & Brunori, 1982), that is, a great reduct,ion of ligand athnity and cooperativity in acid solutions due t,o hyperstabilization of the low affinity (T) allosteric deoxy conformation. This phenomenon has been described for carp and trout hemoglobins (Noble et al., 1970; Brunori, 1975) and related to the physiology of the animals. Perutz & Brunori (1982) attributed the Root effect in hemoglobins from amphibians and fishes to two hydrogen bonds between the C-terminal His of the /I-chains (HisHC3/? (146)) one of the key residues of the Bohr effect, and SerFSb (93). Tn hemoglobins displaying the Root effect. this serine replaces the cysteine residue invariably found in mammals (Antonini & Brunori, 1971: Perutz & Brunori, 1982). When discussing the primary structure of the hemoglobin from a larval stage (stage 55) of X. laeois. Banville et al. (1983) pointed out that Ser F9/3 and His HC3/? (146), present in the adult,, are replaced. respectively, by Ala and Phe in the larva. Since both substitutions are thought to be crucial t,o function we decided to test that supposition by comparing the ligand binding properties of the adult and tadpole hemoglobins from this species. The relative instability of the larval hemoglobin, and the small size of the animals (less than 1 g at stage 55), forced us to employ the kinetics of
carbon monoxide recombination after flash photolysis, which has proved to be a powerful tool for comparing the functional characteristics of very small quantities of hemoglobin (Condo et al., 1983). Tadpoles were grown ab ovo at 26”C, and fed wit’h powdered nettles until stage 53 to 55. Blood obtained by heart puncture under microscope observation was collected in 3.2”/b (w/v) sodium citrat’e isotonic solution to prevent clot formation. Hemolysis was induced by adding two volumes of cold distilled water, and subsequently 0.5 iw-NaCl was added to precipitate ghosts. The hemoglobin in the supernatant was dialyzed overnight against the appropriate buffer under an atmosphere of pure CO. Flash photolysis experiments were performed at 20°C’ in 0.1 M-T& or bis-Tris buffer containing 0.1 ;n-NaCl with and without P,-inosit,ol, with an apparatus similar to that, described by Brunori & Giacometti (1981). Adult X. Zaevis hemoglobin, prepared by the procedure described by Condo et al. (1983), was used without further purification. Figure 1 shows the differences in the kinetics of CO recombination between larval and adult X. lacvis hemoglobins, both for the stripped and P,-inositol-saturated pigment. The hemoglobin from adult X. Zaevis clearly shows the Root effect’ described by Perutz & Brunori (1982). Thus, at pH 7.5 the time-course of CO combination is autocatalytic, indicating co-operative binding mediated by a fast transition from the T to the R state. and partial photolysis shows a, quickly reacting speciesindicative of a high-affinity R state (Antonini & Brunori, 1971). At pH 6.0, on the other hand. t,he main part of the time-course is slower (suggesting a reduction in ligand affinity), and pa.rt,ial photolysis has hardly any effect. indicating that at low pH values the molecule is stabilized in t’he low affinity (T) state both in the liganded and unliganded derivatives. The tadpole from the same species possessesa hemoglobin with different kinetic properties. First of all. the overall reaction is, if anything, faster at low than at’ high pH, contrary to the behavior of the adult’s pigment; secondly, the time-course is aut,ocatalytic at. both pH values: fina1l.v partial
JZ. Hrunori
Adult
et al.
pH 6.0
Adult
pH 7.5
Time
(msl
t
Tadpole
Time
pH 6.0
(ms)
Figure 1. Effect of pH on the time-course of CO recombination by adult and tadpole X. laevis hemoglobin. Each panel shows the time-course after complete photolysis (0) and partial (N 16%) photolysis (0). Conditions: 0.1 M-bisTris buffer containing 3 mM-P,-inositol and 0.1 iv-NaCl. Concentration of CO = 25 PM; concentration of hrmoglobin = 4 PM; observation wavelength 1 = 436 nm. Temperature -20°C. The hemoglobin from the adult at pH 6.0 shows approx. 30% of a quickly reacting component both in the full and the partial photolysis; this has been attributed t.o the presence of free a/l dimers in the CO-derivative (similar to what is observed for human hemoglobin). photolysis always leads to the appearance of the quickly reacting form, indicating that ligand-bound tadpole hemoglobin is in the R quaternary state also at low pH in the presence of P,-inositol. Thus, no Root effect is seen in tadpole hemoglobin. Perutz (for a review, see Perutz, 1979) attributed a major part of the alkaline Bohr effect to a salt bridge between the imidazole group of His HC3/3 (146) and the carboxyl group of FGl (94) Asp (for hemoglobin) or Glu (for fish and X. Zaewisadult hemoglobins), which stabilizes the deoxy or T structure. Accordingly, any hemoglobin with a substitution inhibiting that salt bridge should display a greatly reduced or no alkaline Bohr effect. This prediction has been confirmed by the data on trout I hemoglobin (Brunori et al.? 1973; Brunori, 1975). Perutz & Brunori (1982) suggested that the Root effect is related to the nature of amino acid residue F9; a serine (found in those hemoglobins which display a Root effect) is likely to donate a hydrogen bond to the free carbonyl oxygen of the C-terminal
His HC3, and accept a hydrogen bond from the NH of this histidine. On the other hand, the more hydrophobic cysteine residue invariably found in mammals cannot form these hydrogen bonds and stabilizes the T-structure by van der Waals’ interactions, which are much weaker. The behavior of the hemoglobin from the tadpole of X. laevis is consistent with the proposed model: the replacement of His HC3P (146) by Phe abolishes or possibly reverses the Bohr effect and, together with the replacement of Serb93 by Ala (which cannot form any hydrogen bond), also inhibits the Root effect. The hydrophobic Ala at F9fl (93) and the replacement of the C-terminal His by Phe have also been found in component I (hemoglobin I) from trout hemolysate (Barra et al., 1983), whose oxygen equilibrium curve is independent of the concentration of heterotropic ligands, in contrast to component IV, whose oxygen equilibrium curve varies strongly with pH and organic phosphate concentration. The presence of these two
Letters to the Editor components has been correlated with the swimming habits of trout (Brunori, 1975). The present results, together with information from other species, suggest why the absence of a Bohr effect, or its reversal, may be physiologically advantageous to many aquatic Urodela (Triturus &status: Condo et al., 1981; Amphiuma means: Bonaventura et al., 1977; Ambystoma tigrinum: Amiconi et al., 1970) and bo the aquatic stage “tadpole” of many Anura (Rana clamitans: Manwell, 1966; Rana catesbeiana: Watt & Riggs, 1975; X. laevis: this paper). The lack of a Bohr effect or its reversal may be correlated (1) with the large production of lactate originating from anaerobic glycolysis and accumulated in the blood of lower vertebrates (Bennet, 1978), and (2) with the lower buffering capacity of the blood of aquatic amphibians compared with the terrestrial species (Hutchison & Miller, 1979). In addition many aquatic animals do not need an alkaline Bohr effect for the transport’ of CO2 to the gills or lungs because the extensive vascular network in their skin allows it to diffuse into the surrounding water. These biochemical peculiarities not only permit, but perhaps even demand the kind of pa-insensitive oxygen carrier found in aquatic amphibians. Investigations along this line on several species of amphibians are being continued in our laboratories. The authors are very grateful to Professor J. G. Williams (Imp. Cancer Res. Fund, London) for the generous gift of X. lar,vis toad hemoglobins and to Professor M. F. Perutz (M.R.C.. Cambridge) for stimulating suggestions and construct,ive discussion of the experiments reported above. A grant from t)he M.P.I. of It)aly is gratefully acknowledged. M. S. A. B.
Brunorit G. Condb Bellelli Giardina
Institute of Chemistry, Faculty of Medicine and CNR Center of Molecular Biology University of Rome “La Sapienza” Department of Experimental Medicine and Biochemical Sciences, University of Rome “Tor Vergata”, 00185 Rome. Italy ~.--~__~ t Author to whom all correspondence should be addressed at: Istituto di Chimica, Facoltl di Medicina e Chirurgia. rnivrrsiti “La Sapienza”, 00185 Rome, Italy.
329
G. Micheli Centro Acidi Nucleici, CNR Institute of General Physiology Faculty of Science University of Rome “La Sapienza”. Received 3 August 23 October 1984
Rome. Italy
1984, and in revised form
References Amiconi, G.. Brunori, M., Antonini, E., Sorcini? M. & Tentori, L. (1970). Int. J. B&hem. 1, 582-588. Andres, A. C., Hosbach, H. A. & Weber. R. (1984). Biochim. Biophys. Actu, 781, 294-301. Ant,onini, E. & Brunori, M. (1971). Hemoglobin and Myoglobin in their Reactions with Ligands, North Holland, Amsterdam. Banville, D.. Kay, R. M., Harris, R. & Williams. J. G. (1983). J. Biol. Chem. 258, 7924-7927. Barra, D.. Petruzzelli, R., Bossa. F. 8: Brunori. M. (1983). Biochim. Biophys. Acta, 742, 72-77. Bennett, A. F. (1978). Annu. Rev. Physiol. 40. 447-469. Bonaventura, C.. Sullivan, B., Bonaventura. ,J. & Bourne, S. (1977). Nature (London), 265. 474-476. Brunori, M. (1975). Curr. Top. Cell. Regul. 9, l-39. Brunori, M. & Giacometti, G. M. (1981). Methods Enzymol. 76, 582~-585. Brunori, M. (1975). Curr. Top. Cell. Regul. 9, l-39. Brunori, M., Bonaventura, J., Bonaventura, c’.. Giardina, B., Bossa, F. & Antonini, E. (1973). I~oZ. CPU. Biochem. 1, 189-196. Condb, S. G., Giardina, B., Lunadei, M.. Ferracin, A. & Brunori. M. (1981). Eur. J. Biochem. 120. 323-327. Condb, S. G., Giardina, B., Bellelli, A.. Lunadei, M., Ferracin, A. & Brunori, M. (1983). Comp. Biochem. Physiol. 74A, 545-548. Hutchison. V. H. & Miller, K. (1979). Pomp. Biochem. Physiol. 63A. 213-216. Kay, R. M., Harris, R., Patient, R. K. & Williams, J. G. (1980). Nucl. Acids Res. 8, 2691-2707. Manwell, C. (1966). Comp. Biochem. Physiol. 17. 805-813. Noble, R. W., Parkhurst, L. J. & Gibson. Q. H. (1970). J. Biol. Chem. 245, 6628-6633. Perutz, M. F. (1979). Annu. Rev. Biochem. 44X 327-386. Perutz, M. F. & Brunori, M. (1982). AVdurr (Londo?~), 299. 421-426. Watt, K. W. K. 8 Riggs, A. (1975). J. Bid. (‘hem. 250, 5934-5944. Williams. ,J. G., Kay. R. M. & Patient, R. K. (1980). .Vucl. Acids Res. 8, 4247-4258.
Edited by G. A. Gilbert
(1985) 181, 327-329
Tadpole Xenopus Zaevis Hemoglobin Correlation Between Structure and Functional Properties Perutz $ Brunori (1982) proposed that the COOH-terminal His and Ser F9 of the P-chains of fish and amphibian hemoglobins are responsible for their Root effect and part of their alkaline Bohr effect. Analysis of the kinetics of carbon monoxide binding by hemoglobin from the tadpole of Xenopus laevis supports that model and suggestsan explanation for t’hr absence of an alkaline Bohr effect in many aquatic Anura and Urodela.
Tn most vertebrates the globins synthesized during embryonic life differ from those of the adult. The switch of the globin genes has been thoroughly investigated in Anurans, such as Xenopus laevis, and has provided a model of the regulatory mechanisms of cell- and stage-relat’ed gene expression. The primary structures of the major component of both the adult and larval hemoglobins of X. laevis have been determined from their complementary DNA sequences (Kay et al., 1980; Williams et al., 1980: Banville et al., 1983; Andres et ccl., 1984). Knowledge of the sequence offers an opportunity for studying the relationships between the structure of the hemoglobins and their physiological function. The adult X. Zaevis hemoglobin exhibits the socalled Root effect (Perutz & Brunori, 1982), that is, a great reduct,ion of ligand athnity and cooperativity in acid solutions due t,o hyperstabilization of the low affinity (T) allosteric deoxy conformation. This phenomenon has been described for carp and trout hemoglobins (Noble et al., 1970; Brunori, 1975) and related to the physiology of the animals. Perutz & Brunori (1982) attributed the Root effect in hemoglobins from amphibians and fishes to two hydrogen bonds between the C-terminal His of the /I-chains (HisHC3/? (146)) one of the key residues of the Bohr effect, and SerFSb (93). Tn hemoglobins displaying the Root effect. this serine replaces the cysteine residue invariably found in mammals (Antonini & Brunori, 1971: Perutz & Brunori, 1982). When discussing the primary structure of the hemoglobin from a larval stage (stage 55) of X. laeois. Banville et al. (1983) pointed out that Ser F9/3 and His HC3/? (146), present in the adult,, are replaced. respectively, by Ala and Phe in the larva. Since both substitutions are thought to be crucial t,o function we decided to test that supposition by comparing the ligand binding properties of the adult and tadpole hemoglobins from this species. The relative instability of the larval hemoglobin, and the small size of the animals (less than 1 g at stage 55), forced us to employ the kinetics of
carbon monoxide recombination after flash photolysis, which has proved to be a powerful tool for comparing the functional characteristics of very small quantities of hemoglobin (Condo et al., 1983). Tadpoles were grown ab ovo at 26”C, and fed wit’h powdered nettles until stage 53 to 55. Blood obtained by heart puncture under microscope observation was collected in 3.2”/b (w/v) sodium citrat’e isotonic solution to prevent clot formation. Hemolysis was induced by adding two volumes of cold distilled water, and subsequently 0.5 iw-NaCl was added to precipitate ghosts. The hemoglobin in the supernatant was dialyzed overnight against the appropriate buffer under an atmosphere of pure CO. Flash photolysis experiments were performed at 20°C’ in 0.1 M-T& or bis-Tris buffer containing 0.1 ;n-NaCl with and without P,-inosit,ol, with an apparatus similar to that, described by Brunori & Giacometti (1981). Adult X. Zaevis hemoglobin, prepared by the procedure described by Condo et al. (1983), was used without further purification. Figure 1 shows the differences in the kinetics of CO recombination between larval and adult X. lacvis hemoglobins, both for the stripped and P,-inositol-saturated pigment. The hemoglobin from adult X. Zaevis clearly shows the Root effect’ described by Perutz & Brunori (1982). Thus, at pH 7.5 the time-course of CO combination is autocatalytic, indicating co-operative binding mediated by a fast transition from the T to the R state. and partial photolysis shows a, quickly reacting speciesindicative of a high-affinity R state (Antonini & Brunori, 1971). At pH 6.0, on the other hand. t,he main part of the time-course is slower (suggesting a reduction in ligand affinity), and pa.rt,ial photolysis has hardly any effect. indicating that at low pH values the molecule is stabilized in t’he low affinity (T) state both in the liganded and unliganded derivatives. The tadpole from the same species possessesa hemoglobin with different kinetic properties. First of all. the overall reaction is, if anything, faster at low than at’ high pH, contrary to the behavior of the adult’s pigment; secondly, the time-course is aut,ocatalytic at. both pH values: fina1l.v partial
JZ. Hrunori
Adult
et al.
pH 6.0
Adult
pH 7.5
Time
(msl
t
Tadpole
Time
pH 6.0
(ms)
Figure 1. Effect of pH on the time-course of CO recombination by adult and tadpole X. laevis hemoglobin. Each panel shows the time-course after complete photolysis (0) and partial (N 16%) photolysis (0). Conditions: 0.1 M-bisTris buffer containing 3 mM-P,-inositol and 0.1 iv-NaCl. Concentration of CO = 25 PM; concentration of hrmoglobin = 4 PM; observation wavelength 1 = 436 nm. Temperature -20°C. The hemoglobin from the adult at pH 6.0 shows approx. 30% of a quickly reacting component both in the full and the partial photolysis; this has been attributed t.o the presence of free a/l dimers in the CO-derivative (similar to what is observed for human hemoglobin). photolysis always leads to the appearance of the quickly reacting form, indicating that ligand-bound tadpole hemoglobin is in the R quaternary state also at low pH in the presence of P,-inositol. Thus, no Root effect is seen in tadpole hemoglobin. Perutz (for a review, see Perutz, 1979) attributed a major part of the alkaline Bohr effect to a salt bridge between the imidazole group of His HC3/3 (146) and the carboxyl group of FGl (94) Asp (for hemoglobin) or Glu (for fish and X. Zaewisadult hemoglobins), which stabilizes the deoxy or T structure. Accordingly, any hemoglobin with a substitution inhibiting that salt bridge should display a greatly reduced or no alkaline Bohr effect. This prediction has been confirmed by the data on trout I hemoglobin (Brunori et al.? 1973; Brunori, 1975). Perutz & Brunori (1982) suggested that the Root effect is related to the nature of amino acid residue F9; a serine (found in those hemoglobins which display a Root effect) is likely to donate a hydrogen bond to the free carbonyl oxygen of the C-terminal
His HC3, and accept a hydrogen bond from the NH of this histidine. On the other hand, the more hydrophobic cysteine residue invariably found in mammals cannot form these hydrogen bonds and stabilizes the T-structure by van der Waals’ interactions, which are much weaker. The behavior of the hemoglobin from the tadpole of X. laevis is consistent with the proposed model: the replacement of His HC3P (146) by Phe abolishes or possibly reverses the Bohr effect and, together with the replacement of Serb93 by Ala (which cannot form any hydrogen bond), also inhibits the Root effect. The hydrophobic Ala at F9fl (93) and the replacement of the C-terminal His by Phe have also been found in component I (hemoglobin I) from trout hemolysate (Barra et al., 1983), whose oxygen equilibrium curve is independent of the concentration of heterotropic ligands, in contrast to component IV, whose oxygen equilibrium curve varies strongly with pH and organic phosphate concentration. The presence of these two
Letters to the Editor components has been correlated with the swimming habits of trout (Brunori, 1975). The present results, together with information from other species, suggest why the absence of a Bohr effect, or its reversal, may be physiologically advantageous to many aquatic Urodela (Triturus &status: Condo et al., 1981; Amphiuma means: Bonaventura et al., 1977; Ambystoma tigrinum: Amiconi et al., 1970) and bo the aquatic stage “tadpole” of many Anura (Rana clamitans: Manwell, 1966; Rana catesbeiana: Watt & Riggs, 1975; X. laevis: this paper). The lack of a Bohr effect or its reversal may be correlated (1) with the large production of lactate originating from anaerobic glycolysis and accumulated in the blood of lower vertebrates (Bennet, 1978), and (2) with the lower buffering capacity of the blood of aquatic amphibians compared with the terrestrial species (Hutchison & Miller, 1979). In addition many aquatic animals do not need an alkaline Bohr effect for the transport’ of CO2 to the gills or lungs because the extensive vascular network in their skin allows it to diffuse into the surrounding water. These biochemical peculiarities not only permit, but perhaps even demand the kind of pa-insensitive oxygen carrier found in aquatic amphibians. Investigations along this line on several species of amphibians are being continued in our laboratories. The authors are very grateful to Professor J. G. Williams (Imp. Cancer Res. Fund, London) for the generous gift of X. lar,vis toad hemoglobins and to Professor M. F. Perutz (M.R.C.. Cambridge) for stimulating suggestions and construct,ive discussion of the experiments reported above. A grant from t)he M.P.I. of It)aly is gratefully acknowledged. M. S. A. B.
Brunorit G. Condb Bellelli Giardina
Institute of Chemistry, Faculty of Medicine and CNR Center of Molecular Biology University of Rome “La Sapienza” Department of Experimental Medicine and Biochemical Sciences, University of Rome “Tor Vergata”, 00185 Rome. Italy ~.--~__~ t Author to whom all correspondence should be addressed at: Istituto di Chimica, Facoltl di Medicina e Chirurgia. rnivrrsiti “La Sapienza”, 00185 Rome, Italy.
329
G. Micheli Centro Acidi Nucleici, CNR Institute of General Physiology Faculty of Science University of Rome “La Sapienza”. Received 3 August 23 October 1984
Rome. Italy
1984, and in revised form
References Amiconi, G.. Brunori, M., Antonini, E., Sorcini? M. & Tentori, L. (1970). Int. J. B&hem. 1, 582-588. Andres, A. C., Hosbach, H. A. & Weber. R. (1984). Biochim. Biophys. Actu, 781, 294-301. Ant,onini, E. & Brunori, M. (1971). Hemoglobin and Myoglobin in their Reactions with Ligands, North Holland, Amsterdam. Banville, D.. Kay, R. M., Harris, R. & Williams. J. G. (1983). J. Biol. Chem. 258, 7924-7927. Barra, D.. Petruzzelli, R., Bossa. F. 8: Brunori. M. (1983). Biochim. Biophys. Acta, 742, 72-77. Bennett, A. F. (1978). Annu. Rev. Physiol. 40. 447-469. Bonaventura, C.. Sullivan, B., Bonaventura. ,J. & Bourne, S. (1977). Nature (London), 265. 474-476. Brunori, M. (1975). Curr. Top. Cell. Regul. 9, l-39. Brunori, M. & Giacometti, G. M. (1981). Methods Enzymol. 76, 582~-585. Brunori, M. (1975). Curr. Top. Cell. Regul. 9, l-39. Brunori, M., Bonaventura, J., Bonaventura, c’.. Giardina, B., Bossa, F. & Antonini, E. (1973). I~oZ. CPU. Biochem. 1, 189-196. Condb, S. G., Giardina, B., Lunadei, M.. Ferracin, A. & Brunori. M. (1981). Eur. J. Biochem. 120. 323-327. Condb, S. G., Giardina, B., Bellelli, A.. Lunadei, M., Ferracin, A. & Brunori, M. (1983). Comp. Biochem. Physiol. 74A, 545-548. Hutchison. V. H. & Miller, K. (1979). Pomp. Biochem. Physiol. 63A. 213-216. Kay, R. M., Harris, R., Patient, R. K. & Williams, J. G. (1980). Nucl. Acids Res. 8, 2691-2707. Manwell, C. (1966). Comp. Biochem. Physiol. 17. 805-813. Noble, R. W., Parkhurst, L. J. & Gibson. Q. H. (1970). J. Biol. Chem. 245, 6628-6633. Perutz, M. F. (1979). Annu. Rev. Biochem. 44X 327-386. Perutz, M. F. & Brunori, M. (1982). AVdurr (Londo?~), 299. 421-426. Watt, K. W. K. 8 Riggs, A. (1975). J. Bid. (‘hem. 250, 5934-5944. Williams. ,J. G., Kay. R. M. & Patient, R. K. (1980). .Vucl. Acids Res. 8, 4247-4258.
Edited by G. A. Gilbert