- Email: [email protected]
Effects of pH, CO2 and organic phosphates on oxygen affinity of sea turtle hemoglobins
75
Respiration Physiology (1982) 48, 75-87 Elsevier Biomedical Press
EFFECTS
OF
pH, C 0 2
AFFINITY
PETER
AND ORGANIC OF SEA TURTLE
PHOSPHATES
ON OXYGEN
HEMOGLOBINS*
L. L U T Z a n d G E O R G E
N. LAPENNAS
l
Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149 and Duke University Marine Laboratory, Beaufort, NC 28516, U.S.A.
Abstract. The affinities of adult-type hemoglobins of green (Chelonia mydas) and loggerhead (Caretta caretta) sea turtles were determined at 25.7°C as a function of pH, Pc02 and each of the organic phosphates found in erythrocytes of these species (ATP, IHP (substitute for IPP) and DPG). Each species showed a single hemoglobin band in cellulose acetate electrophoresis which isoelectric focusing resolved into one major and three minor components having isoelectric points near pH 7.2. Oxygen binding curves were recorded using a modified 'Hemoscan' (Instrument Division, Travenol Laboratories, Savage, MD). Hemolysates were kept saturated with CO during preparation to prevent methemoglobin formation. CO was driven off immediately before performing each oxygen binding determination, using the photodissociative effect of the Hemoscan light beam. In both species, phosphate-free hemoglobin has higher affinity than whole blood, and had lower pH sensitivity. CO 2 (37 Torr) reduces affinity, particularly in the green sea turtle, and eliminates pH sensitivity. Each organic phosphate tested reduces affinity, with IHP being most effective at a given concentration and DPG least effective. Organic phosphates are most effective at low pH, and restore both pH sensitivity and oxygen affinity to wholeblood levels taking into account the difference between plasma and intracellular pH. Reasons are suggested why the organic phosphate effect was not detected in a previous study. Hemolysates of both species are distinctive in giving upward-curving Hill plots, with n = 1 asymptotes at low saturation that are insensitive to pH, CO2 and organic phosphates. The oxygen affinities corresponding to these asymptotes are about 100 Torr in both species. Results are compared to oxygen dissociation curves obtained from whole blood. Diving reptile Hemoglobin Hill plot
Organic phosphates Oxygen affinity Ps0
Accepted for publication 31 December 1981 * Supported by National Science Foundation Grant PCM78-09281 and PCM79-0642 and NIH Grant HL07057. 1 Present address: Department of Physiology, Schools of Medicine and Dentistry, State University of New York, Buffalo, NY 14214, U.S.A. 0034-5687/82/0000-0000/$02.75 © Elsevier Biomedical Press
76
P.L. LUTZ AND G.N. LAPENNAS
In vertebrates, blood oxygen affinity is determined by the intrinsic hemoglobin oxygen affinity as modified by pH, Pco2 and organic phosphate compounds within the erythrocyte. The particular organic phosphates present vary in a broad taxonomic pattern (see review by Bartlett, 1980), and a few groups including cats, many ruminants (Bunn, 1980) and crocodilians (Bartlett, 1980), have no significant organic phosphate concentrations. Multiple organic phosphates are found in red cells of adult green (Chelonia mydas) and loggerhead (Caretta caretta) sea turtles. Adenosine triphosphate (ATP) predominates and is accompanied by smaller amounts of inositol pentaphosphate (IPP), and 2,3-diphosphoglycerate (DPG) (Isaacks, Harkness and Witham, 1978). Bartlett (1976) also found ATP and IPP in the green turtle and Pacific Ridley Lepidochelys olivacea). However, he noted small quantities of guanosine triphosphate (GTP) and found no DPG. As a rule, vertebrates that have red cell organic phosphates have hemoglobins whose oxygen affinities are affected by these compounds (Bunn, 1980). Conversely it appears that species with low or negligible concentrations of organic phosphate also have hemoglobins that are insensitive. It was therefore surprising when Isaacks et al. (1978) observed no effect of organic phosphates (ATP, IPP, DPG) on the hemoglobins of adult green and loggerhead sea turtles. One objective of the present study was to examine the organic phosphate sensitivity of sea turtle hemoglobins in more detail. We have shown earlier (Lapennas and Lutz, 1982) that sea turtle, whole blood has some very distinctive properties. In green and loggerhead blood, the Hill coefficient 'n' declines markedly at low oxygen saturation levels. In both species, the H ÷ Bohr effect and the CO2 Bohr effect are also strongly saturation dependent, falling to negligible values at low saturation. Although green and loggerhead sea turtles are of similar size and have somewhat similar mode of life, loggerhead blood has a much lower oxygen affinity (Isaacks et al., 1978; Lapennas and Lutz, 1982) while the green has a higher oxylabile CO2 sensitivity (Lapennas and Lutz, 1982). In view of these results, we considered it worthwhile to examine the origin of these whole blood properties in terms of the oxygen combining properties of sea turtle hemoglobins, including the effects of CO2 and organic phosphates over a range of pH. A preliminary report of our results has been presented elsewhere (Lapennas and Lutz, 1979). Materials and methods
Green and loggerhead sea turtles were obtained as hatchlings and raised to an age of 2 years (body mass about 3-4 kg). Blood was also obtained from an adult loggerhead for the purpose of establishing that the hemoglobin component composition in the younger turtles used for oxygen binding measurements was the same as in an adult. Turtles were maintained in outdoor tanks with running seawater, and were fed on a diet of fish and Purina Chow.
MARINE TURTLE HEMOGLOBINS
77
Blood samples were obtained by syringe from the jugular vein, anticoagulated by rinsing the syringe with sodium heparin solution. Blood was pooled from three green turtles, while loggerhead blood was obtained from a single individual. Hemolysate was prepared as described by Antonini and Brunori (1971). After removing cellular debris by centrifugation, the samples were saturated with CO to inhibit methemoglobin formation (Wallace et al., 1978). We have observed that autoxidation of the hemoglobins of these species is rapid in the absence of this precaution. Hemolysate was deionized and stripped of organic phosphates by 24-h dialysis against CO-containing deionized water followed by passage over mixed bed ion exchange resin (Fisher 'Rexyn') after Dintzis (1952). The hemolysates were free of methemoglobin as indicated by absence of spectral change on addition of potassium cyanide. (Hemolysate prepared similarly but without CO saturation contained 7-10~ methemoglobin). A small amount of unstripped hemolysate was prepared for the determination of hemoglobin absorption spectra. Fifty microliters of packed ceils (4 min at 13,500 x g) were lysed in 5.45 ml of deionized water, followed by addition of 0.5 ml of 1 M Tris-HC1 buffer, pH 8, and centrifugation to clear the solution of cell debris. One subsample of this solution was saturated with 02 to obtain the spectra of oxyhemoglobin. A second subsample was saturated with CO to obtain the carbonmonoxy spectrum, while a third subsample was used to determine the hemoglobin concentration according to the pyridine hemochromogen method (Porra and Jones, 1963). Spectra were recorded on an Aminco DW2A spectrophotometer with the bandpass set to 1 nm. Hemoglobin solutions were prepared for oxygen binding measurements as follows. Hemoglobin concentration was between 0.4 and 0.48 mM (tetramer), as determined using the extinction coefficients determined as described above (differences from human hemoglobin did not exceed a few percent). All solutions contained 0.05 M Hepes buffer (Sigma). Organic phosphates (adenosine triphosphate, ATP; inositol hexaphosphate, IHP; 2,3-diphosphoglycerate, DPG) were added in the form of stock solutions prepared as described by Bonaventura et al. (1972). pH of solutions was measured using a Radiometer Model 27 pH meter and capillary glass electrode calibrated at the experimental temperature, pH in the presence of carbon dioxide was measured after equilibration in a Radiometer ATM-1 Microtonometer. Oxygen binding measurements were performed using a modified 'HEM-O-SCAN' Oxygen Dissociation Analyzer (Instrument Division, Travenol Laboratories, Savage, MD.). The modifications to the instrument and its mode of use were as described by Lapennas and Lutz (1982). The accuracy of this method on hemoglobin solutions has been verified by comparison to a tonometric method. Using human hemoglobin A under conditions of pH and organic phosphate yielding a wide range of affinity (Ps0 from 0.5 to 55 Torr) there was excellent agreement between the two methods in affinity, pH dependence and shape of the binding curve (G. N, Lapennas and J. Bonaventura, unpublished results). The low oxygen affinity of sea turtle hemoglobins under certain conditions (see
78
P.L. LUTZ A N D G. N. LAPENNAS
Results) mean that quite high Po2 must be used in order to obtain full oxygen saturation. We raised the Po: to 650-700 Torr, but even so there was evidence of incomplete saturation under conditions giving the lowest affinity (i.e., saturation still increasing slowly with Po~ at 650 Torr; Hill plots tending to curve upward at the highest saturation levels). These unavoidable errors under lowest affinity conditions do not affect the general conclusions of this paper, i.e., (a) the presence of a substantial phosphate effect in sea turtle hemoglobins, (b) curvature of Hill plots, and (c) insensitivity of the lower n --- I asymptote of Hill plots to pH and organic phosphates. The consequence of incomplete saturation, which would naturally increase as affinity decreases at low pH or with organic phosphates, would be to cause an apparent decrease in the Bohr and phosphate effects at high saturation relative to lower saturation, i.e., the opposite of what we observed. Dissociation of CO from the hemoglobin solutions was performed immediately before running the oxygen binding curve on a given sample, to minimize the opportunity for methemoglobin formation. The sample assembly was placed in the Hemoscan sample chamber at a Po2 of 30-50 Torr. The light beam of the Hemoscan is sufficiently intense to photodissociate the CO from hemoglobin in a short time (5-10 min under the conditions of the present study). This process was monitored by recording the optical signal of the Hemoscan vs. time, and waiting until the signal stabilized before running the curve. It is important to remove the CO completely since any remaining CO would alter the oxygen affinity of the remaining binding sites (e.g., Collier, 1976). We have tested the effectiveness of this method for CO removal by running oxygen binding curves on a given hemoglobin solution before and after CO saturation. Identical binding curves were obtained, indicating that CO dissociation was effectively complete. Oxygen binding was measured in the absence of CO2 and in 5% CO2. The latter gives a Pco2 of 37 mm Hg at the experimental temperature of 25.7 °C + 0.2 °C. Methemoglobin formation during a determination was insignificant as evidenced by the excellent reproducibility of curves on the same sample. Electrophoresis was performed on cellulose acetate plates (Helena Laboratories) in 0.025 M Tris-EDTA-boric acid buffer, pH 8.4 (Schmidt and Brosious, 1978). Isoelectric focusing was performed in acrylamide gel according to Drysdale, Righetti and Bunn (1971), using ampholite range 6-8.
Results
Electrophoresis produced a single band in each species, having mobilities relative to human Hb A of 1.32 and 0.92 in green and loggerhead turtles, respectively. Isoelectric focusing resolved each hemolysate into one major and three minor components, with the major component making up about 70~o of the total and having an isoelectric point of 7.2. Our samples are comparable to the hemolysates of adults studied by Isaacks et al. since (a) in the green transition to the adult
79
MARINE TURTLE HEMOGLOBINS
2"0
100
1'5
-stripped ab:" . . . . . .
~
% f r ~ ' ~ : I [..I~"m- . . . .
-=- - - * " "=
O1
0
GREEN
O
1.0
, 7.0
67
, 7.5
10 7.8
pH Fig. 1. Comparison of oxygen affinity o f stripped hemoglobin and whole blood at Pco2 37 Ton"; 25.7 °C.
hemoglobin is completed within the first year (Isaacks et al., 1978) and (b) our isoelectric focusing experiments demonstrated the identity of hemolysates from 2-year old and adult loggerheads. In both species the oxygen affinity of phosphate free ('stripped') hemoglobin is markedly higher than that of whole blood, as measured at approximately physiological Pco2, 37 Torr (fig. 1). Stripped green turtle hemoglobin has higher intrinsic oxygen affinity than that of the loggerhead. Interestingly, for both species stripped hemoglobin is insensitive to pH at Pco~ = 37 Torr, whereas whole blood has a significant Bohr effect. Because of the differing pH sensitivities, the affinities of stripped hemoglobin and whole blood nearly converge at high pH, but are very different at lower pH. The effects of CO2 and organic phosphates on oxygen affinity are shown in figs. 2 and 3. In both species the oxygen affinity of stripped hemoglobin decreases 2.0
I
IHP~
~
'
•
'
'
1.8 LATP
1.4
""
""
• ........ 1'2
"'+
- - " - "'~o
strnpped
1.0 I
6.7
6-9
i 7.1
r
7.3
L
7-5
7-7
pH
Fig. 2. Effect of CO 2 and organic phosphates on green sea turtle hemoglobin. 25.7 °C, hemoglobin concentration 0.4-0.45 m M (tetramer); organic phosphate concentration ! raM; 0.05 M HEPES. ( ) = 0 Ton- CO 2 ; ( - - - - ) = 37 Ton" C O 2 .
80
P.L. LUTZ A N D G . N . L A P E N N A S
2.0 "
IHP=^
1.8 -
~"'-
o ~ 1-6 _o stripped
1.4
~
6.7
I
I
I
I
6.9
7.1
7.3
7-5
7.7
pH Fig. 3. Effect o f CO 2 and organic phosphates on loggerhead sea turtles hemoglobin. (Conditions same as in fig. 2.)
in the presence of CO2, the effect being greatest in the green sea turtle. All three organic phosphates reduce oxygen affinity in both species, both in the presence and absence of CO 2 (figs. 2 and 3), with the green being more strongly affected. In both species IHP is more effective than ATP, which in turn is more effective than DPG. Sensitivity to pH increases in the presence of organic phosphates. This increase in sensitivity is particularly evident when stripped and organic phosphate Bohr effects are compared in the presence of 37 Torr Pco~. The responses to the allosteric factors are highly pH dependent. In both species, ATP, IHP and DPG reduce affinity at low pH but have little effect at high pH, while CO2 reduces affinity most at high pH. In the presence of organic phosphates, addition of CO2 has no effect at low pH, but reduces affinity at high pH, except for the green turtle with IHP where C O 2 increases affinity. The effect of varying organic phosphate concentrations at pH 7.2 are shown in fig. 4. Although it was only in the case of ATP that the concentrations tested approached full saturation of the organic phosphate effect, it appears that saturating concentrations of the organic phosphates would affect oxygen affinity in the same sequence as in 1 mM conoentrations. Under most conditions, Hill plots were markedly curved. Except under the highest affinity conditions, Hill plots show a region of n approaching or reaching 1 at low oxygen saturation, and then curve upward with increasing saturation (e.g., figs. 5 and 6). The positions of the n = 1 asymptote are insensitive to changes in pH, CO: and organic phosphates, and correspond to a Ps0 of about 100 Torr in both species. When any of these factors act to reduce affinity, they do so by delaying departure from the lower asymptote. Figures 5 and 6 illustrate this point with respect to the effect of 1 mM organic phosphates at constant pH. As relates to
81
MARINE TURTLE HEMOGLOBINS
changing pH, the Bohr effect is negligible at low oxygen saturation and increases
progressively with increasing saturation. I El
b
IHP
i
i
1.8 IHP
ATP
1-6 8
1.4
1.2
1'0
'
0 "3
'
1.0
'
'
~
J O
2"0 3"0 -3 1" ORGANIC PHOSPHATE CONCENTRATION (raM)
2.~0
3"0
Fig. 4. (a) Effect o f organic phosphate concentration on green hemoglobin in absence of C O 2 • (Conditions as in fig. 2, except for organic phosphate concentration). (b) Effect of organic phosphate concentration on loggerhead hemoglobin in absence o f CO 2 .
1
,~ o~//~, ///
-2 ~-'J/i 0
, , 1 log PO2
, 2
,
l 3
torr
Fig. 5. Hill plots of green hemoglobin in the presence and absence o f organic phosphates (25.7 °C, Pco2 37 Torr).
82
P.L. LUTZ AND G.N. LAPENNAS ,
,ff
,
S
//
/1/
~.',o° _o
-1
/
-2
/
/ i
0
i
I
1
2
log P02
i
torr
Fig. 6. Hill plots for loggerhead hemoglobin in the presence and absence of organic phosphate (25.7 °C, Pco~ 37 Torr).
Discussion
In contrast to an earlier report (Isaacks et al., 1978) we find that the hemoglobins of adult green and loggerhead sea turtles are affected by each of the organic phosphates present in their erythrocytes. The effect is not so large as in human hemoglobin under similar conditions (e.g., Bonaventura et al., 1975), but this is partly due to the low intrinsic affinity of the stripped hemoglobins in sea turtles. The qualitative effects of organic phosphates (reduced affinity and more negative Bohr effect) are appropriate to explain the affinity difference between stripped hemoglobin and whole blood. Affinity in the presence of 1 mM ATP, the predominant intracellular organic phosphate, is similar to that of whole blood, especially if the difference between intra- and extracellular pH is taken into account. (In preliminary experiments, measuring the pH of freeze lysed packed cells at 37 Torr CO 2 showed that in both species at 50~ saturation the intracellular pH was 7.2 when plasma was 7.4). The effectiveness of ATP would presumably be reduced somewhat in vivo due to complexing with intracellular magnesium ion (Bunn, Ransil and Chao, 1971 ; Weber and Lykkeboe, 1978). The most likely reasons that phosphate effects were undetected by Isaacks et al. (1978) are that their experiments were performed under conditions where the phosphate effect is rather small (pH 7.4, Pco2 ---40 Torr) and only low concentrations of organic phosphates were tested (maximum concentrations of 0.2, 0.1 and 0.4 mM for ATP, IPP and DPG, respectively, with hemoglobin concentration of 0.05 mM
83
MARINE TURTLE HEMOGLOBINS
(tetramer). (The reason why these phosphate concentrations might not be adequate to elicit the phosphate effect, despite being several times higher than the hemoglobin concentration, is discussed below.) Errors due to incomplete saturation of these low affinity hemoglobins may also have tended to obscure the phosphate effect. The method employed (Lian, Roth and Harkness, 1971) involved measurement of Po2 as oxygen is removed from the sample at a constant rate, and assumes that the initial rate of Po2 decline is entirely due to removal from physical solution. Errors that would tend to reduce the apparent phosphate effect would occur if the sample were not fully saturated at the initial Po2 (Lapennas et al., 1981). Isaacks et al. do not specify the initial Po: employed. As in whole blood, the intrinsic affinity of green turtle hemoglobin is higher than that of loggerhead. However, green turtle hemoglobin is more sensitive to organic phosphates. Interestingly, for human hemoglobin DPG is a more effective modifier of 02 affinity than ATP at equal concentrations (Bunn et al., 1971), while for sea turtle hemoglobin the sequence is reversed. IHP also shows the highest binding affinity with human hemoglobin (Arnone and Perutz, 1974). It is of interest to consider whether the minor organic phosphate components would have a significant effect on affinity in vivo. It has been noted, e.g. by Bartlett (1980), that intracellular organic phosphate must be present at a significant fraction of the hemoglobin concentration to exert an effect. IPP and DPG in sea turtle red cells are at concentrations less than 1/10 that of the hemoglobin (table 1). Duhm (1971) found that affinity of human red cells decreases rapidly with increasing ratios of DPG to hemoglobin, up to about 0.8, while further increase has little additional effect (except for that due to its influence o n intracellular pH). By analogy, it is reasonable to expect that IPP and DPG in sea turtle erythrocytes would have incremental effects over that of the ATP, since the total organic phosphate concentration is less than 0.8 that of the hemoglobin. In the course of our literature review on the effects of organic phosphates, we have frequently encountered the explicit or implicit assumption that a realistic measure of the in vivo effect of organic phosphates is provided directly by in vitro measurements of affinity on dilute hemoglobin solutions using the naturally
TABLE I Molar ratios of organic phosphates t o hemoglobin tetramer in sea turtle erythrocytesa
Green Loggerhead
[ATP]
[ATP] [Hb4]
[IPP]
[IPP] [Hb4]
[DPG]
[DPG] [Hb4]
[OP] [Hb4]
2.40 3.32
0.48 0.66
0.41 0.43
0.08 0.09
0.26 0.37
0.05 0.07
0.61 0.82
a Organic phosphate concentrations in mM from Isaaeks et al. (1978). [Hb4] assumed 5 mM (Lapennas and Lutz, 1982).
84
P.L. LUTZ A N D G.N. LAPENNAS
occurring molar ratios of phosphate to hemoglobin. An alternate form of this assumption is that if an organic phosphate has an effect, it should be detectable in vitro using dilute hemoglobin concentration if the molar ratio is several fold greater than the intracellular ratio. We suggest that this assumption is unjustified. It has long been appreciated that dilution of whole human hemolysate causes an increase in oxygen affinity (Forster, 1972), and this is thought to be due, in part, to the falling DPG concentration (e.g., Wood and Lenfant, 1976). Since the DPG/hemoglobin molar ratio is unaltered by dilution, these results demonstrate that a given molar ratio has less effect in dilute than in concentrated solution. Similarly, Torelli et al. (1977) found that higher molar ratios are required to obtain maximal effect in more dilute hemoglobin solutions. It would thus seem appropriate to use substantial organic phosphate concentrations (one to several mM) when testing for the presence of an effect on affinity, even though this results in unnaturally high molar ratios. In vitro studies would ideally be performed with hemoglobin, organic phosphate and inorganic constituents at their in vivo levels, but such investigations pose difficulties since hemoglobin is packed into the erythrocyte nearly to the limit of its solubility (Riggs, 1979). Except in the presence of IHP, CO 2 reduces 02 affinity by a specific effect (i.e., independent of its pH effect) in the hemoglobin of both species. As in whole blood, this effect is greater in the green in the presence of IHP, however, the effect is suppressed in loggerhead hemoglobin. A similar suppression of oxylabile carbamate formation has been found in some bird species at physiological levels of IPP (Baumann and Baumann, 1977), and in carp hemoglobin CO2 has no effect on oxygen affinity in the presence of organic phosphates (Weber and Lykkeboe, 1978). In the green turtle, on the other hand, CO2 increases the oxygen affinity in the presence of IHP. This effect is markedly pH dependent, declining as pH falls and it may even reverse at low pH. Interestingly, for human whole blood at low levels of DPG, increasing Pco2 at low pH may reduce oxygen affinity (fig. 1 ; Arturson et al., 1974) and for adult human hemoglobin at pH 7.2, CO2 actually increases affinity at high levels of DPG (Petschow et al., 1977). The reason for this varied response appears to lie in the fact that for some hemoglobins there is competition between the effect of CO2 and organic phosphates on blood oxygen affinity, both acting on a common site of the hemoglobin molecule (Kilmartin and RossiBernardi, 1973). The response to CO2, therefore, depends on the nature of the hemoglobin(s) and the conditions in the cell. For those hemoglobins where both CO2 and phosphate binding is oxylabile, the increase in oxygen affinity resulting from a reduction of Pco2 will be moderated by an increase in the organic phosphate binding and, if sufficient, this may even result in a net decrease in the oxygen affinity (Lutz, 1980). The organic phosphate effectiveness would be enhanced at low pH (Hlastala and Woodson, 1975). There are important similarities between the shapes of the oxygen dissociation curves of sea turtle whole blood (Lapennas and Lutz, 1982) and those of their hemoglobin in solution. In both species the Hill coefficient, n, declines, at low
MARINE TURTLE HEMOGLOBINS
85
saturation levels, to an asymptote of n = 1. In the region o f this asymptote, the curves show insensitivity to pH, and the hemoglobin solution curves are also insensitive to organic phosphates. Further study will be required to determine whether these properties reside in the individual hemoglobin components of each species, or whether they are due to the combination of hemoglobins having, individually, properties quite different from those of the mixture. The insensitivity of the bottom of the binding curves to allosteric factors indicates the presence either of one or more components that are totally insensitive or of ones whose 'T states' (Monod, Wyman and Changeux, 1965) are insensitive. The latter possibility would represent behavior similar to that of carp hemoglobin under certain conditions (Tan and Noble, 1973) and would conform to the assumption of the model of Monod et al. It would be contrary to the behavior of human Hb A in which the T state is strongly affected by pH and organic phosphates (Imai, 1973; Imai and Yonetani, 1975). We conclude that the particularly interesting features of the oxygen binding curves of green and loggerhead blood can be accounted for by the intrinsic properties of their hemoglobins. As with whole blood, phosphate free green turtle hemoglobin has a higher affinity than that of the loggerhead. Hemolysates of both species give upward curving Hill plots, with n -- 1 asymptotes at low saturation that are insensitive to pH, CO2 and organic phosphates. CO2 reduces affinity in both species, but as in whole blood the effect is greater in the green turtle. The hemoglobin of both species are sensitive to organic phosphates and organic phosphates are likely to play a role in affinity modulation. However, there is a complex interaction between these allosteric factors (pH, Pco2, organic phosphates) and any change in the level of one can markedly affect the influence of the other.
References Antonini, E. and M. Brunori (1971). Hemoglobin and Myoglobin in Their Reactions with Ligands. New York, American Elsevier. Arnone, A. and M.F. Perutz (1974). Structure of inositol hexaphosphate-human deoxyhaemoglobin complex. Nature 249: 34-36. Arturson, G., L. Garby, B. Wranne and B. Zaar (1974). Effect of 2,3-diphosphoglycerate on the oxygen affinity and on the proton- and carbamino-linked oxygen affinity of hemoglobin in human whole blood. Acta Physiol. Scand. 92: 332-340. Bartlett, G. R. (1976). Phosphate compounds in red cells of reptiles, amphibians and fish. Comp. Biochem. Physiol. 55A: 211-214. Bartlett, G.R. (1980). Phosphate compounds in vertebrate red blood cells. Am. Zool. 20: 103-114. Baumann, F.H. and R. Baumann (1977). A comparative study of oxygen transport in bird blood. Respir. Physiol. 31 : 333-343. Bonaventura, J., C. Bonaventura, B. Giardina, E. Antonini, M. Brunori and J. Wyman (1972). Partial restoration of normal fttlactional properties in carboxypeptidase A-digested hemoglobin. Proc. Natl. Acad. Sci. (USA), 69: 2174-2178. Bonaventura, J., C. Bonaventura, B. Sullivan and G. Godette (1975). Hemoglobin Deer Lodge (B2-His-Arg). J. Biol. Chem. 250: 9250-9255.
86
P.L. LUTZ AND G.N. LAPENNAS
Bunn, H.F. (1971). Differences in the interaction of 2,3-diphosphoglycerate with certain mammalian hemoglobins. Science 172: 1049-1050. Bunn, H. F., B.J. Ransil and A. Chao (1971). The interaction between erythrocyte organic phosphates, magnesium ion and hemoglobin. J. Biol. Chem. 246: 5273-5279. Bunn, H.F. (1980). Regulation of hemoglobin function in mammals. Am. Zool. 20:199-211. Collier, C. R. (1976). Oxygen affinity of human blood in presence of carbon monoxide. J. Appl. Physiol. 40: 487490. Dintzis, H. (1952). Studies on the dilectric properties of human serum mercaptalbumin solutions. Ph.D. Thesis, Harvard University, Cambridge, MA. Drysdale, J.W., P. Righetti and H.F. Bunn (1971). The separation of human and animal hemoglobins by isoelectric focusing in polyacrylamide gel. Bioc'him. Biophys. Acta 229: 42-50. Duhm, J. (1971). Effects of 2,3-diphosphoglycerate and other organic compounds on oxygen affinity and intracellular pH of human erythrocytes. Pfliigers Arch. 326: 334-356. Forster, R.E. (1972). The effect of dilution in saline on the oxygen affinity of human hemoglobin. In: Oxygen Affinity of Hemoglobin and Red Cell Acid Base Status, edited by M. R~rth and P. Astrup. Copenhagen, Munksgaard. Hlastala, M.P. and R.D. Woodson (1975). Saturation dependency of the Bohr effect: interactions among H ÷, CO2, and DPG. J. Appl. Physiol. 38: 1126-1131. Imai, K. (1973). Analysis of oxygen equilibria of native and chemically modified human adult hemoglobins on the basis of Adair's stepwise oxygenation theory and the allosteric model of Monod, Wyman and Changeux. Biochemistry 12: 798-808. Imai, K. and T. Yonetani (1975). pH dependent of the Adair constants of human hemoglobin. J. Biol. Chem. 250: 2227-2231. Isaacks, R. E., D. R. Harkness and P. R. Witham (1978). Relationship between the major phosphorylated metabolic intermediates and oxygen affinity of whole blood in the loggerhead (Caretta caretta) and the green sea turtle (Chelonia mydas) during development. Dev. Biol. 62: 344-353. Kilmartin, J.V. and L. Rossi-Bernardi (1973). Interaction of hemoglobin with hydrogen ions, carbon dioxide and organic phosphates. Physiol. Rev. 53: 836-890. Lapennas, G.N. and P.L. Lutz (1979). Oxygen affinity of green and loggerhead blood. The effect of organic phosphates and carbon dioxide. Am. Zool. 19: 982. Lapennas, G. N., J.M. Colacino and J. Bonaventura (1981). Thin-layer methods for determination of oxygen binding curves of hemoglobin solutions and blood. In: Hemoglobins, Methods in Ensymology, Vol. 76, edited by E. Antonini, L. Rossi-Bernardi and E. Chiancone. New York, Academic Press, pp. 449~70. Lapennas, G.N. and P. L. Lutz (1982). Oxygen affinity of sea turtle blood. Respir. Physiol. 48: 59-74. Lian, C.Y., S. Roth and D.R. Harkness (1971). The effect of alteration of intracellular 2,3-DPG concentration upon oxygen binding of intact erythrocytes containing normal and mutant hemoglobins. Biochem. Biophys. Res. Commun. 45: 151-158. Lutz, P.L. (1980). On the oxygen affinity of bird blood. Am. Zool. 20: 187-198. Monod, J., J. Wyman and J.P. Changeux (1965). On the nature of allosteric transitions: a plausible model. J. Mol. Biol. 12: 88-118. Petschow, D., I. Wurdinger, R. Baumann, J. Duhm, G. Braunitzer and C. Bauer (1977). Causes of high blood 02 affinity of animals living at high altitude. J. Appl. Physiol. 42: 139-143. Porra, R.J. and O. T. G. Jones (1963). Studies on ferrochelatase: assay and properties of ferrochelatase from a pig liver and mitochondrial extract. Biochem. J. 87: 181-185. Riggs, A. (1979). Studies on the hemoglobins of Amazonian fishes: an overview. Comp. Biochem. Physiol. 62A : 257-272. Schmidt, R. M. and E. M. Brosious (1978). Basic Laboratory Methods of Hemoglobinopathy Detection. DHEW Publication # (CDC) 78-8266. Tan, A.L. and R.W. Noble (1973). Conditions restricting allosteric transitions in carp hemoglobin. J. Biol. Chem. 248: 2880-2888. Torelli, G., F. Celentano, G. Corteli, E. D'Angelo, A. Cazziniga and E. P. Radford (1977). Hemoglobin-
MARINE TURTLE HEMOGLOBINS
87
oxygen equilibrium at different hemoglobin and 2,3-diphosphoglycerate concentrations. Physiol. Chem. Physics 9: 21-38. Wallace, W.J., R.A. Houtchens, J.M. Holt and W.S. Caughey (1978). Mechanisms for hemoglobin oxidation : the response of abnormal human and other hemoglobins to different oxidative pathways. In: Biochemical and Clinical Aspects of Hemoglobin Abnormalities, edited by W.S. Caughey. New York, Academic Press, pp. 475-490. Weber, R.E. and G. Lykkeboe (1978). Respiratory adaptations in carp blood, influences of hypoxia, red cell organic phosphates, divalent cations and CO 2 on hemoglobin-oxygen affinity. J. Comp. Physiol. 128: 127-137. Wood, S.C. and C. Lenfant (1976). Respiration: Mechanics, control and gas exchange. In: Biology of the Reptilia, Vol. 5, edited by C. Gans and W.R. Dawson. New York, Academic Press, pp. 225-274.
Respiration Physiology (1982) 48, 75-87 Elsevier Biomedical Press
EFFECTS
OF
pH, C 0 2
AFFINITY
PETER
AND ORGANIC OF SEA TURTLE
PHOSPHATES
ON OXYGEN
HEMOGLOBINS*
L. L U T Z a n d G E O R G E
N. LAPENNAS
l
Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149 and Duke University Marine Laboratory, Beaufort, NC 28516, U.S.A.
Abstract. The affinities of adult-type hemoglobins of green (Chelonia mydas) and loggerhead (Caretta caretta) sea turtles were determined at 25.7°C as a function of pH, Pc02 and each of the organic phosphates found in erythrocytes of these species (ATP, IHP (substitute for IPP) and DPG). Each species showed a single hemoglobin band in cellulose acetate electrophoresis which isoelectric focusing resolved into one major and three minor components having isoelectric points near pH 7.2. Oxygen binding curves were recorded using a modified 'Hemoscan' (Instrument Division, Travenol Laboratories, Savage, MD). Hemolysates were kept saturated with CO during preparation to prevent methemoglobin formation. CO was driven off immediately before performing each oxygen binding determination, using the photodissociative effect of the Hemoscan light beam. In both species, phosphate-free hemoglobin has higher affinity than whole blood, and had lower pH sensitivity. CO 2 (37 Torr) reduces affinity, particularly in the green sea turtle, and eliminates pH sensitivity. Each organic phosphate tested reduces affinity, with IHP being most effective at a given concentration and DPG least effective. Organic phosphates are most effective at low pH, and restore both pH sensitivity and oxygen affinity to wholeblood levels taking into account the difference between plasma and intracellular pH. Reasons are suggested why the organic phosphate effect was not detected in a previous study. Hemolysates of both species are distinctive in giving upward-curving Hill plots, with n = 1 asymptotes at low saturation that are insensitive to pH, CO2 and organic phosphates. The oxygen affinities corresponding to these asymptotes are about 100 Torr in both species. Results are compared to oxygen dissociation curves obtained from whole blood. Diving reptile Hemoglobin Hill plot
Organic phosphates Oxygen affinity Ps0
Accepted for publication 31 December 1981 * Supported by National Science Foundation Grant PCM78-09281 and PCM79-0642 and NIH Grant HL07057. 1 Present address: Department of Physiology, Schools of Medicine and Dentistry, State University of New York, Buffalo, NY 14214, U.S.A. 0034-5687/82/0000-0000/$02.75 © Elsevier Biomedical Press
76
P.L. LUTZ AND G.N. LAPENNAS
In vertebrates, blood oxygen affinity is determined by the intrinsic hemoglobin oxygen affinity as modified by pH, Pco2 and organic phosphate compounds within the erythrocyte. The particular organic phosphates present vary in a broad taxonomic pattern (see review by Bartlett, 1980), and a few groups including cats, many ruminants (Bunn, 1980) and crocodilians (Bartlett, 1980), have no significant organic phosphate concentrations. Multiple organic phosphates are found in red cells of adult green (Chelonia mydas) and loggerhead (Caretta caretta) sea turtles. Adenosine triphosphate (ATP) predominates and is accompanied by smaller amounts of inositol pentaphosphate (IPP), and 2,3-diphosphoglycerate (DPG) (Isaacks, Harkness and Witham, 1978). Bartlett (1976) also found ATP and IPP in the green turtle and Pacific Ridley Lepidochelys olivacea). However, he noted small quantities of guanosine triphosphate (GTP) and found no DPG. As a rule, vertebrates that have red cell organic phosphates have hemoglobins whose oxygen affinities are affected by these compounds (Bunn, 1980). Conversely it appears that species with low or negligible concentrations of organic phosphate also have hemoglobins that are insensitive. It was therefore surprising when Isaacks et al. (1978) observed no effect of organic phosphates (ATP, IPP, DPG) on the hemoglobins of adult green and loggerhead sea turtles. One objective of the present study was to examine the organic phosphate sensitivity of sea turtle hemoglobins in more detail. We have shown earlier (Lapennas and Lutz, 1982) that sea turtle, whole blood has some very distinctive properties. In green and loggerhead blood, the Hill coefficient 'n' declines markedly at low oxygen saturation levels. In both species, the H ÷ Bohr effect and the CO2 Bohr effect are also strongly saturation dependent, falling to negligible values at low saturation. Although green and loggerhead sea turtles are of similar size and have somewhat similar mode of life, loggerhead blood has a much lower oxygen affinity (Isaacks et al., 1978; Lapennas and Lutz, 1982) while the green has a higher oxylabile CO2 sensitivity (Lapennas and Lutz, 1982). In view of these results, we considered it worthwhile to examine the origin of these whole blood properties in terms of the oxygen combining properties of sea turtle hemoglobins, including the effects of CO2 and organic phosphates over a range of pH. A preliminary report of our results has been presented elsewhere (Lapennas and Lutz, 1979). Materials and methods
Green and loggerhead sea turtles were obtained as hatchlings and raised to an age of 2 years (body mass about 3-4 kg). Blood was also obtained from an adult loggerhead for the purpose of establishing that the hemoglobin component composition in the younger turtles used for oxygen binding measurements was the same as in an adult. Turtles were maintained in outdoor tanks with running seawater, and were fed on a diet of fish and Purina Chow.
MARINE TURTLE HEMOGLOBINS
77
Blood samples were obtained by syringe from the jugular vein, anticoagulated by rinsing the syringe with sodium heparin solution. Blood was pooled from three green turtles, while loggerhead blood was obtained from a single individual. Hemolysate was prepared as described by Antonini and Brunori (1971). After removing cellular debris by centrifugation, the samples were saturated with CO to inhibit methemoglobin formation (Wallace et al., 1978). We have observed that autoxidation of the hemoglobins of these species is rapid in the absence of this precaution. Hemolysate was deionized and stripped of organic phosphates by 24-h dialysis against CO-containing deionized water followed by passage over mixed bed ion exchange resin (Fisher 'Rexyn') after Dintzis (1952). The hemolysates were free of methemoglobin as indicated by absence of spectral change on addition of potassium cyanide. (Hemolysate prepared similarly but without CO saturation contained 7-10~ methemoglobin). A small amount of unstripped hemolysate was prepared for the determination of hemoglobin absorption spectra. Fifty microliters of packed ceils (4 min at 13,500 x g) were lysed in 5.45 ml of deionized water, followed by addition of 0.5 ml of 1 M Tris-HC1 buffer, pH 8, and centrifugation to clear the solution of cell debris. One subsample of this solution was saturated with 02 to obtain the spectra of oxyhemoglobin. A second subsample was saturated with CO to obtain the carbonmonoxy spectrum, while a third subsample was used to determine the hemoglobin concentration according to the pyridine hemochromogen method (Porra and Jones, 1963). Spectra were recorded on an Aminco DW2A spectrophotometer with the bandpass set to 1 nm. Hemoglobin solutions were prepared for oxygen binding measurements as follows. Hemoglobin concentration was between 0.4 and 0.48 mM (tetramer), as determined using the extinction coefficients determined as described above (differences from human hemoglobin did not exceed a few percent). All solutions contained 0.05 M Hepes buffer (Sigma). Organic phosphates (adenosine triphosphate, ATP; inositol hexaphosphate, IHP; 2,3-diphosphoglycerate, DPG) were added in the form of stock solutions prepared as described by Bonaventura et al. (1972). pH of solutions was measured using a Radiometer Model 27 pH meter and capillary glass electrode calibrated at the experimental temperature, pH in the presence of carbon dioxide was measured after equilibration in a Radiometer ATM-1 Microtonometer. Oxygen binding measurements were performed using a modified 'HEM-O-SCAN' Oxygen Dissociation Analyzer (Instrument Division, Travenol Laboratories, Savage, MD.). The modifications to the instrument and its mode of use were as described by Lapennas and Lutz (1982). The accuracy of this method on hemoglobin solutions has been verified by comparison to a tonometric method. Using human hemoglobin A under conditions of pH and organic phosphate yielding a wide range of affinity (Ps0 from 0.5 to 55 Torr) there was excellent agreement between the two methods in affinity, pH dependence and shape of the binding curve (G. N, Lapennas and J. Bonaventura, unpublished results). The low oxygen affinity of sea turtle hemoglobins under certain conditions (see
78
P.L. LUTZ A N D G. N. LAPENNAS
Results) mean that quite high Po2 must be used in order to obtain full oxygen saturation. We raised the Po: to 650-700 Torr, but even so there was evidence of incomplete saturation under conditions giving the lowest affinity (i.e., saturation still increasing slowly with Po~ at 650 Torr; Hill plots tending to curve upward at the highest saturation levels). These unavoidable errors under lowest affinity conditions do not affect the general conclusions of this paper, i.e., (a) the presence of a substantial phosphate effect in sea turtle hemoglobins, (b) curvature of Hill plots, and (c) insensitivity of the lower n --- I asymptote of Hill plots to pH and organic phosphates. The consequence of incomplete saturation, which would naturally increase as affinity decreases at low pH or with organic phosphates, would be to cause an apparent decrease in the Bohr and phosphate effects at high saturation relative to lower saturation, i.e., the opposite of what we observed. Dissociation of CO from the hemoglobin solutions was performed immediately before running the oxygen binding curve on a given sample, to minimize the opportunity for methemoglobin formation. The sample assembly was placed in the Hemoscan sample chamber at a Po2 of 30-50 Torr. The light beam of the Hemoscan is sufficiently intense to photodissociate the CO from hemoglobin in a short time (5-10 min under the conditions of the present study). This process was monitored by recording the optical signal of the Hemoscan vs. time, and waiting until the signal stabilized before running the curve. It is important to remove the CO completely since any remaining CO would alter the oxygen affinity of the remaining binding sites (e.g., Collier, 1976). We have tested the effectiveness of this method for CO removal by running oxygen binding curves on a given hemoglobin solution before and after CO saturation. Identical binding curves were obtained, indicating that CO dissociation was effectively complete. Oxygen binding was measured in the absence of CO2 and in 5% CO2. The latter gives a Pco2 of 37 mm Hg at the experimental temperature of 25.7 °C + 0.2 °C. Methemoglobin formation during a determination was insignificant as evidenced by the excellent reproducibility of curves on the same sample. Electrophoresis was performed on cellulose acetate plates (Helena Laboratories) in 0.025 M Tris-EDTA-boric acid buffer, pH 8.4 (Schmidt and Brosious, 1978). Isoelectric focusing was performed in acrylamide gel according to Drysdale, Righetti and Bunn (1971), using ampholite range 6-8.
Results
Electrophoresis produced a single band in each species, having mobilities relative to human Hb A of 1.32 and 0.92 in green and loggerhead turtles, respectively. Isoelectric focusing resolved each hemolysate into one major and three minor components, with the major component making up about 70~o of the total and having an isoelectric point of 7.2. Our samples are comparable to the hemolysates of adults studied by Isaacks et al. since (a) in the green transition to the adult
79
MARINE TURTLE HEMOGLOBINS
2"0
100
1'5
-stripped ab:" . . . . . .
~
% f r ~ ' ~ : I [..I~"m- . . . .
-=- - - * " "=
O1
0
GREEN
O
1.0
, 7.0
67
, 7.5
10 7.8
pH Fig. 1. Comparison of oxygen affinity o f stripped hemoglobin and whole blood at Pco2 37 Ton"; 25.7 °C.
hemoglobin is completed within the first year (Isaacks et al., 1978) and (b) our isoelectric focusing experiments demonstrated the identity of hemolysates from 2-year old and adult loggerheads. In both species the oxygen affinity of phosphate free ('stripped') hemoglobin is markedly higher than that of whole blood, as measured at approximately physiological Pco2, 37 Torr (fig. 1). Stripped green turtle hemoglobin has higher intrinsic oxygen affinity than that of the loggerhead. Interestingly, for both species stripped hemoglobin is insensitive to pH at Pco~ = 37 Torr, whereas whole blood has a significant Bohr effect. Because of the differing pH sensitivities, the affinities of stripped hemoglobin and whole blood nearly converge at high pH, but are very different at lower pH. The effects of CO2 and organic phosphates on oxygen affinity are shown in figs. 2 and 3. In both species the oxygen affinity of stripped hemoglobin decreases 2.0
I
IHP~
~
'
•
'
'
1.8 LATP
1.4
""
""
• ........ 1'2
"'+
- - " - "'~o
strnpped
1.0 I
6.7
6-9
i 7.1
r
7.3
L
7-5
7-7
pH
Fig. 2. Effect of CO 2 and organic phosphates on green sea turtle hemoglobin. 25.7 °C, hemoglobin concentration 0.4-0.45 m M (tetramer); organic phosphate concentration ! raM; 0.05 M HEPES. ( ) = 0 Ton- CO 2 ; ( - - - - ) = 37 Ton" C O 2 .
80
P.L. LUTZ A N D G . N . L A P E N N A S
2.0 "
IHP=^
1.8 -
~"'-
o ~ 1-6 _o stripped
1.4
~
6.7
I
I
I
I
6.9
7.1
7.3
7-5
7.7
pH Fig. 3. Effect o f CO 2 and organic phosphates on loggerhead sea turtles hemoglobin. (Conditions same as in fig. 2.)
in the presence of CO2, the effect being greatest in the green sea turtle. All three organic phosphates reduce oxygen affinity in both species, both in the presence and absence of CO 2 (figs. 2 and 3), with the green being more strongly affected. In both species IHP is more effective than ATP, which in turn is more effective than DPG. Sensitivity to pH increases in the presence of organic phosphates. This increase in sensitivity is particularly evident when stripped and organic phosphate Bohr effects are compared in the presence of 37 Torr Pco~. The responses to the allosteric factors are highly pH dependent. In both species, ATP, IHP and DPG reduce affinity at low pH but have little effect at high pH, while CO2 reduces affinity most at high pH. In the presence of organic phosphates, addition of CO2 has no effect at low pH, but reduces affinity at high pH, except for the green turtle with IHP where C O 2 increases affinity. The effect of varying organic phosphate concentrations at pH 7.2 are shown in fig. 4. Although it was only in the case of ATP that the concentrations tested approached full saturation of the organic phosphate effect, it appears that saturating concentrations of the organic phosphates would affect oxygen affinity in the same sequence as in 1 mM conoentrations. Under most conditions, Hill plots were markedly curved. Except under the highest affinity conditions, Hill plots show a region of n approaching or reaching 1 at low oxygen saturation, and then curve upward with increasing saturation (e.g., figs. 5 and 6). The positions of the n = 1 asymptote are insensitive to changes in pH, CO: and organic phosphates, and correspond to a Ps0 of about 100 Torr in both species. When any of these factors act to reduce affinity, they do so by delaying departure from the lower asymptote. Figures 5 and 6 illustrate this point with respect to the effect of 1 mM organic phosphates at constant pH. As relates to
81
MARINE TURTLE HEMOGLOBINS
changing pH, the Bohr effect is negligible at low oxygen saturation and increases
progressively with increasing saturation. I El
b
IHP
i
i
1.8 IHP
ATP
1-6 8
1.4
1.2
1'0
'
0 "3
'
1.0
'
'
~
J O
2"0 3"0 -3 1" ORGANIC PHOSPHATE CONCENTRATION (raM)
2.~0
3"0
Fig. 4. (a) Effect o f organic phosphate concentration on green hemoglobin in absence of C O 2 • (Conditions as in fig. 2, except for organic phosphate concentration). (b) Effect of organic phosphate concentration on loggerhead hemoglobin in absence o f CO 2 .
1
,~ o~//~, ///
-2 ~-'J/i 0
, , 1 log PO2
, 2
,
l 3
torr
Fig. 5. Hill plots of green hemoglobin in the presence and absence o f organic phosphates (25.7 °C, Pco2 37 Torr).
82
P.L. LUTZ AND G.N. LAPENNAS ,
,ff
,
S
//
/1/
~.',o° _o
-1
/
-2
/
/ i
0
i
I
1
2
log P02
i
torr
Fig. 6. Hill plots for loggerhead hemoglobin in the presence and absence of organic phosphate (25.7 °C, Pco~ 37 Torr).
Discussion
In contrast to an earlier report (Isaacks et al., 1978) we find that the hemoglobins of adult green and loggerhead sea turtles are affected by each of the organic phosphates present in their erythrocytes. The effect is not so large as in human hemoglobin under similar conditions (e.g., Bonaventura et al., 1975), but this is partly due to the low intrinsic affinity of the stripped hemoglobins in sea turtles. The qualitative effects of organic phosphates (reduced affinity and more negative Bohr effect) are appropriate to explain the affinity difference between stripped hemoglobin and whole blood. Affinity in the presence of 1 mM ATP, the predominant intracellular organic phosphate, is similar to that of whole blood, especially if the difference between intra- and extracellular pH is taken into account. (In preliminary experiments, measuring the pH of freeze lysed packed cells at 37 Torr CO 2 showed that in both species at 50~ saturation the intracellular pH was 7.2 when plasma was 7.4). The effectiveness of ATP would presumably be reduced somewhat in vivo due to complexing with intracellular magnesium ion (Bunn, Ransil and Chao, 1971 ; Weber and Lykkeboe, 1978). The most likely reasons that phosphate effects were undetected by Isaacks et al. (1978) are that their experiments were performed under conditions where the phosphate effect is rather small (pH 7.4, Pco2 ---40 Torr) and only low concentrations of organic phosphates were tested (maximum concentrations of 0.2, 0.1 and 0.4 mM for ATP, IPP and DPG, respectively, with hemoglobin concentration of 0.05 mM
83
MARINE TURTLE HEMOGLOBINS
(tetramer). (The reason why these phosphate concentrations might not be adequate to elicit the phosphate effect, despite being several times higher than the hemoglobin concentration, is discussed below.) Errors due to incomplete saturation of these low affinity hemoglobins may also have tended to obscure the phosphate effect. The method employed (Lian, Roth and Harkness, 1971) involved measurement of Po2 as oxygen is removed from the sample at a constant rate, and assumes that the initial rate of Po2 decline is entirely due to removal from physical solution. Errors that would tend to reduce the apparent phosphate effect would occur if the sample were not fully saturated at the initial Po2 (Lapennas et al., 1981). Isaacks et al. do not specify the initial Po: employed. As in whole blood, the intrinsic affinity of green turtle hemoglobin is higher than that of loggerhead. However, green turtle hemoglobin is more sensitive to organic phosphates. Interestingly, for human hemoglobin DPG is a more effective modifier of 02 affinity than ATP at equal concentrations (Bunn et al., 1971), while for sea turtle hemoglobin the sequence is reversed. IHP also shows the highest binding affinity with human hemoglobin (Arnone and Perutz, 1974). It is of interest to consider whether the minor organic phosphate components would have a significant effect on affinity in vivo. It has been noted, e.g. by Bartlett (1980), that intracellular organic phosphate must be present at a significant fraction of the hemoglobin concentration to exert an effect. IPP and DPG in sea turtle red cells are at concentrations less than 1/10 that of the hemoglobin (table 1). Duhm (1971) found that affinity of human red cells decreases rapidly with increasing ratios of DPG to hemoglobin, up to about 0.8, while further increase has little additional effect (except for that due to its influence o n intracellular pH). By analogy, it is reasonable to expect that IPP and DPG in sea turtle erythrocytes would have incremental effects over that of the ATP, since the total organic phosphate concentration is less than 0.8 that of the hemoglobin. In the course of our literature review on the effects of organic phosphates, we have frequently encountered the explicit or implicit assumption that a realistic measure of the in vivo effect of organic phosphates is provided directly by in vitro measurements of affinity on dilute hemoglobin solutions using the naturally
TABLE I Molar ratios of organic phosphates t o hemoglobin tetramer in sea turtle erythrocytesa
Green Loggerhead
[ATP]
[ATP] [Hb4]
[IPP]
[IPP] [Hb4]
[DPG]
[DPG] [Hb4]
[OP] [Hb4]
2.40 3.32
0.48 0.66
0.41 0.43
0.08 0.09
0.26 0.37
0.05 0.07
0.61 0.82
a Organic phosphate concentrations in mM from Isaaeks et al. (1978). [Hb4] assumed 5 mM (Lapennas and Lutz, 1982).
84
P.L. LUTZ A N D G.N. LAPENNAS
occurring molar ratios of phosphate to hemoglobin. An alternate form of this assumption is that if an organic phosphate has an effect, it should be detectable in vitro using dilute hemoglobin concentration if the molar ratio is several fold greater than the intracellular ratio. We suggest that this assumption is unjustified. It has long been appreciated that dilution of whole human hemolysate causes an increase in oxygen affinity (Forster, 1972), and this is thought to be due, in part, to the falling DPG concentration (e.g., Wood and Lenfant, 1976). Since the DPG/hemoglobin molar ratio is unaltered by dilution, these results demonstrate that a given molar ratio has less effect in dilute than in concentrated solution. Similarly, Torelli et al. (1977) found that higher molar ratios are required to obtain maximal effect in more dilute hemoglobin solutions. It would thus seem appropriate to use substantial organic phosphate concentrations (one to several mM) when testing for the presence of an effect on affinity, even though this results in unnaturally high molar ratios. In vitro studies would ideally be performed with hemoglobin, organic phosphate and inorganic constituents at their in vivo levels, but such investigations pose difficulties since hemoglobin is packed into the erythrocyte nearly to the limit of its solubility (Riggs, 1979). Except in the presence of IHP, CO 2 reduces 02 affinity by a specific effect (i.e., independent of its pH effect) in the hemoglobin of both species. As in whole blood, this effect is greater in the green in the presence of IHP, however, the effect is suppressed in loggerhead hemoglobin. A similar suppression of oxylabile carbamate formation has been found in some bird species at physiological levels of IPP (Baumann and Baumann, 1977), and in carp hemoglobin CO2 has no effect on oxygen affinity in the presence of organic phosphates (Weber and Lykkeboe, 1978). In the green turtle, on the other hand, CO2 increases the oxygen affinity in the presence of IHP. This effect is markedly pH dependent, declining as pH falls and it may even reverse at low pH. Interestingly, for human whole blood at low levels of DPG, increasing Pco2 at low pH may reduce oxygen affinity (fig. 1 ; Arturson et al., 1974) and for adult human hemoglobin at pH 7.2, CO2 actually increases affinity at high levels of DPG (Petschow et al., 1977). The reason for this varied response appears to lie in the fact that for some hemoglobins there is competition between the effect of CO2 and organic phosphates on blood oxygen affinity, both acting on a common site of the hemoglobin molecule (Kilmartin and RossiBernardi, 1973). The response to CO2, therefore, depends on the nature of the hemoglobin(s) and the conditions in the cell. For those hemoglobins where both CO2 and phosphate binding is oxylabile, the increase in oxygen affinity resulting from a reduction of Pco2 will be moderated by an increase in the organic phosphate binding and, if sufficient, this may even result in a net decrease in the oxygen affinity (Lutz, 1980). The organic phosphate effectiveness would be enhanced at low pH (Hlastala and Woodson, 1975). There are important similarities between the shapes of the oxygen dissociation curves of sea turtle whole blood (Lapennas and Lutz, 1982) and those of their hemoglobin in solution. In both species the Hill coefficient, n, declines, at low
MARINE TURTLE HEMOGLOBINS
85
saturation levels, to an asymptote of n = 1. In the region o f this asymptote, the curves show insensitivity to pH, and the hemoglobin solution curves are also insensitive to organic phosphates. Further study will be required to determine whether these properties reside in the individual hemoglobin components of each species, or whether they are due to the combination of hemoglobins having, individually, properties quite different from those of the mixture. The insensitivity of the bottom of the binding curves to allosteric factors indicates the presence either of one or more components that are totally insensitive or of ones whose 'T states' (Monod, Wyman and Changeux, 1965) are insensitive. The latter possibility would represent behavior similar to that of carp hemoglobin under certain conditions (Tan and Noble, 1973) and would conform to the assumption of the model of Monod et al. It would be contrary to the behavior of human Hb A in which the T state is strongly affected by pH and organic phosphates (Imai, 1973; Imai and Yonetani, 1975). We conclude that the particularly interesting features of the oxygen binding curves of green and loggerhead blood can be accounted for by the intrinsic properties of their hemoglobins. As with whole blood, phosphate free green turtle hemoglobin has a higher affinity than that of the loggerhead. Hemolysates of both species give upward curving Hill plots, with n -- 1 asymptotes at low saturation that are insensitive to pH, CO2 and organic phosphates. CO2 reduces affinity in both species, but as in whole blood the effect is greater in the green turtle. The hemoglobin of both species are sensitive to organic phosphates and organic phosphates are likely to play a role in affinity modulation. However, there is a complex interaction between these allosteric factors (pH, Pco2, organic phosphates) and any change in the level of one can markedly affect the influence of the other.
References Antonini, E. and M. Brunori (1971). Hemoglobin and Myoglobin in Their Reactions with Ligands. New York, American Elsevier. Arnone, A. and M.F. Perutz (1974). Structure of inositol hexaphosphate-human deoxyhaemoglobin complex. Nature 249: 34-36. Arturson, G., L. Garby, B. Wranne and B. Zaar (1974). Effect of 2,3-diphosphoglycerate on the oxygen affinity and on the proton- and carbamino-linked oxygen affinity of hemoglobin in human whole blood. Acta Physiol. Scand. 92: 332-340. Bartlett, G. R. (1976). Phosphate compounds in red cells of reptiles, amphibians and fish. Comp. Biochem. Physiol. 55A: 211-214. Bartlett, G.R. (1980). Phosphate compounds in vertebrate red blood cells. Am. Zool. 20: 103-114. Baumann, F.H. and R. Baumann (1977). A comparative study of oxygen transport in bird blood. Respir. Physiol. 31 : 333-343. Bonaventura, J., C. Bonaventura, B. Giardina, E. Antonini, M. Brunori and J. Wyman (1972). Partial restoration of normal fttlactional properties in carboxypeptidase A-digested hemoglobin. Proc. Natl. Acad. Sci. (USA), 69: 2174-2178. Bonaventura, J., C. Bonaventura, B. Sullivan and G. Godette (1975). Hemoglobin Deer Lodge (B2-His-Arg). J. Biol. Chem. 250: 9250-9255.
86
P.L. LUTZ AND G.N. LAPENNAS
Bunn, H.F. (1971). Differences in the interaction of 2,3-diphosphoglycerate with certain mammalian hemoglobins. Science 172: 1049-1050. Bunn, H. F., B.J. Ransil and A. Chao (1971). The interaction between erythrocyte organic phosphates, magnesium ion and hemoglobin. J. Biol. Chem. 246: 5273-5279. Bunn, H.F. (1980). Regulation of hemoglobin function in mammals. Am. Zool. 20:199-211. Collier, C. R. (1976). Oxygen affinity of human blood in presence of carbon monoxide. J. Appl. Physiol. 40: 487490. Dintzis, H. (1952). Studies on the dilectric properties of human serum mercaptalbumin solutions. Ph.D. Thesis, Harvard University, Cambridge, MA. Drysdale, J.W., P. Righetti and H.F. Bunn (1971). The separation of human and animal hemoglobins by isoelectric focusing in polyacrylamide gel. Bioc'him. Biophys. Acta 229: 42-50. Duhm, J. (1971). Effects of 2,3-diphosphoglycerate and other organic compounds on oxygen affinity and intracellular pH of human erythrocytes. Pfliigers Arch. 326: 334-356. Forster, R.E. (1972). The effect of dilution in saline on the oxygen affinity of human hemoglobin. In: Oxygen Affinity of Hemoglobin and Red Cell Acid Base Status, edited by M. R~rth and P. Astrup. Copenhagen, Munksgaard. Hlastala, M.P. and R.D. Woodson (1975). Saturation dependency of the Bohr effect: interactions among H ÷, CO2, and DPG. J. Appl. Physiol. 38: 1126-1131. Imai, K. (1973). Analysis of oxygen equilibria of native and chemically modified human adult hemoglobins on the basis of Adair's stepwise oxygenation theory and the allosteric model of Monod, Wyman and Changeux. Biochemistry 12: 798-808. Imai, K. and T. Yonetani (1975). pH dependent of the Adair constants of human hemoglobin. J. Biol. Chem. 250: 2227-2231. Isaacks, R. E., D. R. Harkness and P. R. Witham (1978). Relationship between the major phosphorylated metabolic intermediates and oxygen affinity of whole blood in the loggerhead (Caretta caretta) and the green sea turtle (Chelonia mydas) during development. Dev. Biol. 62: 344-353. Kilmartin, J.V. and L. Rossi-Bernardi (1973). Interaction of hemoglobin with hydrogen ions, carbon dioxide and organic phosphates. Physiol. Rev. 53: 836-890. Lapennas, G.N. and P.L. Lutz (1979). Oxygen affinity of green and loggerhead blood. The effect of organic phosphates and carbon dioxide. Am. Zool. 19: 982. Lapennas, G. N., J.M. Colacino and J. Bonaventura (1981). Thin-layer methods for determination of oxygen binding curves of hemoglobin solutions and blood. In: Hemoglobins, Methods in Ensymology, Vol. 76, edited by E. Antonini, L. Rossi-Bernardi and E. Chiancone. New York, Academic Press, pp. 449~70. Lapennas, G.N. and P. L. Lutz (1982). Oxygen affinity of sea turtle blood. Respir. Physiol. 48: 59-74. Lian, C.Y., S. Roth and D.R. Harkness (1971). The effect of alteration of intracellular 2,3-DPG concentration upon oxygen binding of intact erythrocytes containing normal and mutant hemoglobins. Biochem. Biophys. Res. Commun. 45: 151-158. Lutz, P.L. (1980). On the oxygen affinity of bird blood. Am. Zool. 20: 187-198. Monod, J., J. Wyman and J.P. Changeux (1965). On the nature of allosteric transitions: a plausible model. J. Mol. Biol. 12: 88-118. Petschow, D., I. Wurdinger, R. Baumann, J. Duhm, G. Braunitzer and C. Bauer (1977). Causes of high blood 02 affinity of animals living at high altitude. J. Appl. Physiol. 42: 139-143. Porra, R.J. and O. T. G. Jones (1963). Studies on ferrochelatase: assay and properties of ferrochelatase from a pig liver and mitochondrial extract. Biochem. J. 87: 181-185. Riggs, A. (1979). Studies on the hemoglobins of Amazonian fishes: an overview. Comp. Biochem. Physiol. 62A : 257-272. Schmidt, R. M. and E. M. Brosious (1978). Basic Laboratory Methods of Hemoglobinopathy Detection. DHEW Publication # (CDC) 78-8266. Tan, A.L. and R.W. Noble (1973). Conditions restricting allosteric transitions in carp hemoglobin. J. Biol. Chem. 248: 2880-2888. Torelli, G., F. Celentano, G. Corteli, E. D'Angelo, A. Cazziniga and E. P. Radford (1977). Hemoglobin-
MARINE TURTLE HEMOGLOBINS
87
oxygen equilibrium at different hemoglobin and 2,3-diphosphoglycerate concentrations. Physiol. Chem. Physics 9: 21-38. Wallace, W.J., R.A. Houtchens, J.M. Holt and W.S. Caughey (1978). Mechanisms for hemoglobin oxidation : the response of abnormal human and other hemoglobins to different oxidative pathways. In: Biochemical and Clinical Aspects of Hemoglobin Abnormalities, edited by W.S. Caughey. New York, Academic Press, pp. 475-490. Weber, R.E. and G. Lykkeboe (1978). Respiratory adaptations in carp blood, influences of hypoxia, red cell organic phosphates, divalent cations and CO 2 on hemoglobin-oxygen affinity. J. Comp. Physiol. 128: 127-137. Wood, S.C. and C. Lenfant (1976). Respiration: Mechanics, control and gas exchange. In: Biology of the Reptilia, Vol. 5, edited by C. Gans and W.R. Dawson. New York, Academic Press, pp. 225-274.