The high oxygen affinity of foetal versus adult blood is an almost universal characteristic of viviparous vertebrates; a notable exception is the cat (Novy & Parer, 1969). There is also likely to be a difference in the oxygen affinities of foetal and adult blood in the teleostean fish, the seaperch Embiotoca lateralis Agassiz. Our studies of E. lateralis foetal and adult haemoglobins and erythrocytes have shown several mechanisms which are likely to facilitate oxygen transfer to the foetal fish by increasing foetal blood oxygen affinity with respect to that of adult blood. Proposed mechanisms include structurally distinct foetal and adult haemoglobins with foetal haemoglobin having the higher oxygen affinity, lower foetal erythrocyte nucleoside triphosphate (NTP, primarily ATP) concentrations, and possibly, lower mean corpuscular haemoglobin concentrations (MCHC) in foetal cells (Ingermann & Terwilliger, 1981a, b, 1982a). It is necessary, however, to see if the oxygen affinity of foetal whole blood is greater than that of adult blood as predicted from purified haemoglobin oxygen binding and erythrocyte characteristics. Consequently, we have measured the oxygen affinities of foetal and adult whole bloods under comparable conditions.

Adult E. lateralis, standard length greater than or equal to 22cm, were obtained near Cape Arago on the Oregon coast. Adult blood samples were collected into icecold heparinized test tubes. Oxygen dissociation curves were generated on these samples within 4 h. Mid-gestation foetuses obtained in early April, and late-gestation foetuses obtained in late June, were collected into ice-cold 40% sea water (approximately the salinity of ovarian fluid). Each foetus was blotted dry, its caudal peduncle severed, and approximately 3 μl blood per fish were collected onto a heparinized glass coverslip. Oxygen equilibria of foetal and adult whole blood samples were determined with a Hem-O-Scan oxygen dissociation analyser (SLM-AMINCO, Urbana, IL) essentially as described by Powers et al. (1979). Temperature was maintained at 20 °C in the Hem-O-Scan with an external circulator and temperature Control. Oxygen dissociation curves were generated in the absence and presence of 5·6 % CO2. For analysis without CO2, the sample was deoxygenated with medical grade N2 (verified to contain less than 0·1 % O2 with a Perkin Elmer mass spectrometer, MGA-1100) and reoxygenated with air (21 % O2). For each analysis with CO2, the 100 % oxyhaemoglobin absorbance was first established with air before the sample was exposed to CO2. The sample was then deoxygenated with 5·6% CO2 and 94·4% N2 and reoxygenated with 5·6% CO2, 25 % O2 and 69·4% N2. Thus, the P50 (partial pressure of oxygen at which 50% of the haemoglobin is saturated) obtained in the presence of CO2 represents the half saturation of the blood with oxygen in the presence of CO2 relative to full saturation with oxygen in the absence of CO2.

In the absence of CO2, mid-gestation foetal blood showed the lowest P50, or highest affinity for oxygen, with a P50 of 8·4 ± 2·4Torr (mean ± S.D.,7V = 5), and adult blood the lowest affinity, with a P50 of 21·2 ± 5·2Torr (N = 3) (Fig. 1). Late-gestation foetal blood had an intermediate value, with a P50 of 12·5 ± 2·5 Torr (N = 8). (Each N for foetal blood represents a single determination per individual; each N for adult blood represents a mean of quadruplicate determinations per individual.) The same basic pattern was seen in the presence of 5·6% CO2 ; however, all curves were shifted towards the right (Fig. 2). With CO2, mid-gestation foetal blood showed a P50 of 67 ± 13 Torr (N = 3 ) and late-gestation foetal blood had a P50 of 86 ± 12 Torr (N = 3 ). In the presence of CO2, one adult fish showed a blood P50 of about 106 Torr; however, samples from two other adults consistently did not exceed 25 % oxygen saturation at a of 150 Torr.

Fig. 1.

Oxygen equilibrium curves of mid-gestation foetal (A), late-gestation foetal (B) and adult (C) bloods in the absence of exogenous CO2.

Fig. 1.

Oxygen equilibrium curves of mid-gestation foetal (A), late-gestation foetal (B) and adult (C) bloods in the absence of exogenous CO2.

Fig. 2.

Oxygen equilibrium curves as in Fig. 1 in the presence of 5·6% CO2.

Fig. 2.

Oxygen equilibrium curves as in Fig. 1 in the presence of 5·6% CO2.

The pH of foetal and maternal blood was measured under conditions comparable to those of the Hem-O-Scan. Adult and late-gestation foetal blood samples were first equilibrated with either N2, air, N2 plus CO2 or air plus CO2 at 20 °C. Blood pH was then measured with a Bio-Rad combination pH microelectrode, calibrated versus an Orion glass electrode. Six measurements were made per equilibration gas. Under N2, both foetal and adult bloods were at pH 7·7 ±0·1; when these samples were reoxygenated with air, their pH values were 7·7 ±0·1 and 7·9 ±0·1, respectively. In the presence of CO2, deoxygenated foetal and adult samples showed pH values of 7·3 ±0·1 and 7·2 ±0·1, respectively; reoxygenated values were 7·2 ± 0·2 and 7·1, respectively. One can thus conclude that since the pH values of the bloods were nearly the same, E. lateralis foetal blood had a higher oxygen affinity than adult blood when the CO2 concentration was either zero or high, two very different conditions.

The blood-oxygen binding data above do not necessarily represent the in vivo state, which is dependent upon variable physiological parameters such as lactate concentration and CO2 tension. However, the data do provide direct evidence that the intrinsic oxygen affinity of adult blood is significantly less than that of either mid- or late-gestation foetal blood when measured under comparable conditions. This is consistent with earlier studies which reported indirect evidence that in this teleost the blood oxygen affinity of the foetus is higher than that of the adult (Ingermann & Terwilliger, 1981a, b, 1982a).

Previous reports have shown that E. lateralis mid-gestation and late-gestation foetal bloods do not differ in NTP levels (mol NTP/mol haemoglobin tetramer) and only slightly in MCHC (Ingermann & Terwilliger, 19816, 1982a). The haemogl bins of these two foetal stages, however, are structurally and functionally different Mid-gestation foetal haemoglobin, stripped of organic phosphates, has a higher oxygen affinity than late-gestation stripped foetal haemoglobin; the latter haemoglobin appears to be a mixture of foetal and adult haemoglobin structures (Ingermann & Terwilliger, 1981a). Thus, the apparent decrease in foetal blood oxygen affinity with increasing development in 71. lateralis during the last half of gestation is probably due to the replacement of foetal haemoglobin by adult haemoglobin.

Szabo & Karplus (1976) have theorized that at low haemoglobin to organic phosphate concentrations, the ratio of stripped haemoglobin P50 in the presence of organic phosphate to stripped haemoglobin P50 in the absence of organic phosphate is a function of the relative binding constants of organic phosphate to deoxy- and oxyhaemoglobin. Ratios of P50 of mid-gestation foetal, late-gestation foetal and adult E. lateralis stripped haemoglobins in the presence of 1 mm-ATP to P50 in its absence are very similar ; e.g. atpH7-4, ratios were found to be 1·3, 1·3 and 1·4, respectively (Ingermann & Terwilliger, 1981a). This strongly suggests that at a given concentration of ATP and haemoglobin, ATP binds and affects E. lateralis haemoglobins similarly. The ratio of P50 of blood to P50 of stripped haemoglobin at pH 7·4 (approximate intraerythrocytic pH) for mid- and late-gestation foetuses was about 1·5–1·6. The low oxygen affinity of blood relative to the affinity of stripped haemoglobin was probably caused by the presence of appreciable intraerythrocytic ATP (Ingermann & Terwilliger, 1981b). Since ATP exerts a similar effect on these haemoglobins, an adult blood P50 of about 16 Torr would be predicted if the structural and functional differences in haemoglobins were the only difference between foetal and adult bloods. However, the actual measured adult blood P50 value was 21 Torr. This implies that the differences in haemoglobin oxygen affinities alone could not account for the high adult blood P50 and that the ATP concentration, or some other factor in the blood or within erythrocytes, decreased the oxygen affinity of adult blood more than that of foetal blood. A likely explanation is the differences in mean corpuscular NTP concentrations; these concentrations are about 4·1, 4·4 and 11 mm for mid-gestation foetal, late-gestation foetal and adult erythrocytes, respectively (calculated as NTP/Hb4 × MCHC, data from Ingermann & Terwilliger, 198lb, 1982a). Additionally, the differences between adult and foetal MCHCs may contribute to the maternal-foetal blood oxygen affinity difference by the mechanism previously discussed (Ingermann & Terwilliger, 1982a). In most viviparous animals which have been studied, facilitation of foetaboxygen uptake is based either upon structurally different adult and foetal haemoglobins or upon different concentrations of intraerythrocytic organic phosphates. E. lateralis appears to incorporate both strategies, and possibly a third - different MCHCs - to facilitate foetal oxygen uptake.

Webb & Brett (1972) found that ovarian fluid decreases and foetal oxygen consumption per unit weight increases as foetal development approaches term in the seaperch, Rhacochilus vacca, a close relative of E. lateralis. Therefore, it might be expected that as the E. lateralis foetus approaches the end of gestation and encounters decreased ovarian oxygen tensions, blood oxygen affinity should increase to ensure oxygen loading. We found, however, that the late-gestation foetal blood had a significantly lower blood oxygen affinity than that of the mid-gestation foetus. The physiological significance of this finding is not clear, but it is not without precedent. Decreases in foetal blood oxygen affinity with increased development have been reported in man and the viviparous lizard, Sphenomorphus quoyii (Bard & Teasdale, 1979; Grigg & Harlow, 1981).

We found in this study that CO2 had a dramatic effect on the extent of oxygenation of whole blood. This is consistent with our earlier studies which showed that both E. lateralis mid-gestation foetal and adult stripped haemoglobins demonstrate a Root effect which is accentuated by ATP (Ingermann & Terwilliger, 1982b). The shift of the blood oxygen dissociation curve by CO2 (primarily a pH phenomenon, Root Irving, 1943) was probably related to the marked sensitivity of these haemoglobins to low pH. The pronounced shift of the adult curve was likely to be related as well to the high ATP concentrations within the adult erythrocytes. The significance of this finding, if any, to maternal-foetal oxygen transfer is unclear. However, it is possible that at physiological oxygen tensions, a Root effect may be operative in the foetus as well as in the adult.

We are grateful to Dr Robert Koler for use of the Hem-O-Scan analyser and to Mrs Marilyn Jones for her technical assistance. We also appreciate the help of Dr Michael Stock with mass spectrometric analysis of equilibration gases. RLI was supported by an Institutional National Research Service Award, No. HD 07084; this work was also supported in part by NSF Grant No. PCM 8207548 to RCT.

Bard
,
H.
&
Teasdale
,
F.
(
1979
).
Red cell oxygen affinity, hemoglobin type, 2,3-diphosphoglycerate, and pH as a function of fetal development
.
Pediatrics
64
,
483
487
.
Grigg
,
G. C.
&
Harlow
,
P.
(
1981
).
A fetal-maternal shift of blood oxygen affinity in an Australian viviparous lizard, Sphenomorphus quoyii (Reptilia, Scincidae)
.
J. comp. Physiol
.
142
,
495
499
.
Ingermann
,
R. L.
&
Terwilliger
,
R. C.
(
1981a
).
Oxygen affinities of maternal and fetal hemoglobins of the viviparous seaperch, Embiotoca lateralis
.
J. comp. Physiol
.
142
,
523
531
.
Ingermann
,
R. L.
&
Terwilliger
,
R. C.
(
1981b
).
Intraerythrocytic organic phosphates of fetal and adult seaperch (Embiotoca lateralis) : their role in maternal-fetal oxygen transport
.
J. comp. Physiol
.
144
,
253
259
.
Ingermann
,
R. L.
&
Terwilliger
,
R. C.
(
1982a
).
Blood parameters and facilitation of maternal-fetal oxygen transfer in a viviparous fish {Embiotoca lateralis)
.
Comp. Biochem. Physiol
.
73A
,
497
501
.
Ingermann
,
R. L.
&
Terwilliger
,
R. C.
(
1982b
).
Presence and possible function of Root effect hemoglobins in fishes lacking a functional swimbladder
.
J. exp. Zool
.
220
,
171
177
.
Novy
,
M. J.
&
Parer
,
J. T.
(
1969
).
Absence of high blood oxygen affinity in the fetal cat
.
Respir. Physiol
.
6
,
144
150
.
Powers
,
D. A.
,
Fyhn
,
H. J.
,
Fyhn
,
U. E. H.
,
Martin
,
J. P.
,
Garlick
,
R. L.
&
Wood
,
S. C.
(
1979
).
A comparative study of the oxygen equilibria of blood from 40 genera of Amazonian fishes
.
Comp. Biochem. Physiol
.
62A
,
67
85
.
Root
,
R. W.
&
Irving
,
L.
(
1943
).
The effect of carbon dioxide and lactic acid on the oxygen-combining power of whole and hemolyzed blood of the marine fish Tautoga onitis (Linn
.).
Biol. Bull. mar. biol. Lab., Woods Hole
84
,
207
212
.
Szabo
,
A.
&
Karplus
,
M.
(
1976
).
Analysis of the interaction of organic phosphates with hemoglobin
.
Biochemistry, N.Y
.
15
,
2869
2877
.
Webb
,
P. W.
&
Brett
,
J. R.
(
1972
).
Oxygen consumption of embryos and parents, and oxygen transfer characteristics within the ovary of two species of viviparous seaperch, Rhacochilus vacca and Embiotoca lateralis
.
J. Fish. Res. Bd Can
.
29
,
1543
1553
.