Thermodynamics of oxygenation-linked proton and lactate binding govern the temperature sensitivity of O₂ binding in crustacean(Carcinus maenas) hemocyanin

The oxygenation of gas-transporting proteins is a function of their intrinsic O₂ affinities, as well as their interactions with allosteric effectors. The latter interactions govern O₂ affinity,and may modulate O₂ loading at the respiratory surfaces and/or O₂ unloading and in the metabolizing tissues in response to specific environmental conditions and metabolic requirements. The interactions of hemoglobins (Hbs) and hemocyanins (Hcs) with protons and CO₂(the Bohr effect, which decreases O₂ affinity and facilitates O₂ in the respiring tissues) are a well-known example.

Unlike the intensively investigated vertebrate Hbs, where protons, chloride and organic phosphate anions are major effectors that decrease Hb–O₂ affinity, crustacean Hcs commonly exhibit marked sensitivities to l-lactate and urate anions and to divalent cations that increase Hc–O₂ affinity(Mangum and Towle, 1977; Truchot, 1980; Mason et al., 1983). Lactate is an end product of anaerobic metabolism that enhances O₂ loading,so it may be a link in the negative feedback that favours aerobiosis in the face of decreasing O₂ availability(Truchot, 1980). In contrast to the tetrameric, intraerythrocytic vertebrate Hbs that consist of twoα and two β chains and have a sedimentation constants of 4.4 S,crustacean Hcs commonly consist of 24 S dodecamers and/or 16 S hexamers which,respectively, comprise twelve and six 5 S polypeptide chains(Markl et al., 1979; Dainese et al., 1998; Molon et al., 2000; Podda et al., 2007). However, Carcinus maenas Hc consists of 24 S dodecamers only(Eriksson-Quensel and Svedberg,1936; Markl et al.,1979).

A universal, although often neglected, factor governing O₂affinity in ectothermic invertebrates is temperature. Based on the exothermic nature of their oxygenation reactions(Klotz and Klotz, 1955), the O₂ affinities of metal-containing, gas-binding proteins decrease with increasing temperature. Apart from the intrinsic heat of oxygenation(ΔH^O₂), the overall heat of oxygenation(ΔH′) includes the heat of solution of oxygen(ΔH^sol, –13 kJ mol^–1) and heats of processes linked to oxygenation, such as the release or binding of protons[including the heat of ionization of buffers (ΔH^H+)] and of other allosteric effectors (ΔH^X,Y,etc).

The ΔH′ values of crustacean Hc at physiological pH(7.6–7.8) vary tremendously (from –67 to +134 kJ mol^–1) (cf. Jokumsen and Weber, 1982; Burnett et al.,1988; Brix et al.,1989; Adamczewska and Morris,1998; Chausson et al.,2004). The available literature has variously attributed adaptive advantages to both low and high temperature sensitivities, arguing that the former may stabilize tissue O₂ supply in the face of environmental thermal variations, and the latter may increase O₂ unloading in the respiring tissues, in parallel with temperature-induced increases in metabolic O₂ requirement. Extensive studies of vertebrate Hbs attribute adaptive reductions in temperature sensitivity to endothermic processes coupled to O₂ binding, such as changes in protein conformation(Wyman et al., 1977) or, more commonly, dissociation of proton, phosphate and chloride ions(Wyman, 1964; Weber et al., 1985; Fago et al., 1997b; Weber et al., 2003). However,no systematic studies appear to have been carried out on the interactive effects of ligand binding on the enthalpies of arthropod Hcs, notwithstanding the enormous variation in their temperature sensitivities and the radically different allosteric control mechanisms encountered compared to Hbs. In contrast to vertebrate Hbs, where protons and organic phosphates preferentially decrease O₂ affinity of the Hb molecules in the low-affinity Tense state (K_T) without significantly affecting that of the Relaxed state (K_R), in crab Callinectes sapidus Hc, l-lactate ions increase both K_T and K_R, whereas protons decrease both K_T and K_R, and Ca²⁺ions increase K_R(Johnson et al., 1988). Thus“...the allosteric interactions of l-lactate and crustacean hemocyanins... provide an interesting contrast to the extremely well-studied allosteric interactions of the vertebrate hemoglobins”(Graham, 1985).

Aiming to probe the contributions of the major endogenous allosteric effectors (protons and lactate) to the overall enthalpies of oxygenation, we measured the interactive effects of lactate, temperature and pH on the O₂ affinity of Hc in the blood of the shore crab Carcinus maenas, which is exposed to large variations in ambient temperature on a daily and seasonal basis.

Specimens of shore crabs Carcinus maenas L., with carapace width of 5–7 cm, were collected in Aarhus Bay, Denmark, and maintained in well-aerated seawater (30–32 p.p.t., 12°C, P_O2 >135 mmHg; 12 h:12 h photoperiod) for at least 4 days without feeding. Prebranchial hemolymph samples (1–2 ml)were withdrawn from the branchial sinuses by piercing the arthropodial membrane at the bases of the walking legs using a syringe fitted with a 22-G hypodermic needle. Samples from individual crabs were kept separately on ice. Those that did not coagulate were pooled into a Petri dish and whipped with a glass spatula to separate coagulating fibrinogen that was removed by filtration. Hemolymph was frozen at –80°C in 120 μl aliquots that were freshly thawed for oxygen binding studies.

Hc concentrations were estimated from peak absorbances near 335 nm, using an extinction coefficient of 17.5 mmol l^–1cm^–1, based on the A_1%,1cm value of 2.33 reported for C. maenas Hc (Nickerson and Van Holde, 1971) and a functional-unit mass of 75 kDa. Reagent grade l(+)-lactate, lithium salt(C₃H₅O₃Li) was purchased from Sigma-Aldrich Chemicals (St Louis, MO, USA). l-lactate concentration in freshly collected hemolymph samples of crabs kept in normoxic seawater) was measured according to Lowry and Passonneau's `Method I'(Lowry and Passonneau,2006).

Hemolymph samples were buffered at varying pH by adding 1 mol l^–1 Bis-Tris buffers to a final buffer concentration of 0.1 mol l^–1. O₂ equilibria of 4 μl hemolymph samples were recorded at 10 and 20°C using a modified gas diffusion chamber connected to Wösthoff gas mixing pumps (Bochum, Germany) that mix air and ultrapure (>99.998%) N₂ to increase O₂tensions stepwise (Weber,1981; Weber et al.,1987), while absorbance is continuously monitored at 365 nm. pH values were measured in triplicate in separate 110 μl sub-samples (to avoid KCl contamination from the calomel electrode) at the same temperatures as the O₂ equilibrium measurements, using a BMS 2 Mk 2 microelectrode coupled to a PHM 64 Research pH meter (Radiometer, Copenhagen, Denmark). For each O₂ binding curve at least four equilibrium steps between 20%and 80% saturation were recorded, and P₅₀ and n₅₀ values (O₂ tensions and Hill's cooperativity coefficients at 50% O₂ saturation, respectively) were interpolated from Hill plots, log[S/(1–S)] vslogP_O2, where S is the fractional O₂ saturation.

The effects and stoichiometry of l-lactate binding were investigated by recording Hc–O₂ equilibria at varying l-lactate concentrations, and at pH values close to 6.9 and 7.7,and thereafter interpolating the P₅₀ and n₅₀ values at these exact pH values (6.90 and 7.70) from regression analysis of logP₅₀vs pH.

For precise O₂ equilibrium measurements focusing on extreme, low and high O₂ saturations (extended Hill plots, see below), a subsample of the freshly prepared hemolymph was concentrated approximately twofold by centrifugation at 2500 g and 4°C for 1 h in Ultrafree-4 Millipore tubes with 10 000 Da molecular mass cut-off membranes.

Lactate concentration in hemolymph of crabs kept in aerated seawater was 0.034±0.023 mmol l^–1 (N=5), confirming the low level previously reported for normoxic, resting C. maenas [0.1 mmol l^–1 (Lallier and Truchot, 1989)].

Dose–response curves measured at two pH values(Fig. 1) show that lactate increases O₂ affinity, exerting the greatest effect in the 50–100 mmol l^–1 concentration range, without markedly affecting the Bohr factor (depicted by the vertical distance between the two curves). Interpolated from the slopes of logP₅₀vs log[lactate] plots in this range, C. maenas Hc binds maximally 0.30 and 0.35 lactate ions per O₂ molecule at pH 7.0 and 7.7, respectively. Plotting logP₅₀ values against free lactate concentrations (estimated assuming that each functional Hc subunit binds 0.3 lactate ions at half-saturation) does not tangibly alter these coefficients, which compare with values of 0.11–0.54 observed in decapod crustacean Hc under different experimental conditions [(cf. Zeis et al., 1992; Adamczewska and Morris, 1998)and studies cited therein].

Extended Hill plots of precise O₂ equilibria of the hemolymph at pH 7.84 (Fig. 2) show slopes of unity in the lower and upper extremities of the curves, reflecting non-cooperative binding of the first and last O₂ molecules to the functional units. Fitting the MWC model to the data in the absence and in the presence of 10 and 50 mmol l^–1 lactate, yields q-values (the number of interacting O₂-binding sites) of 4.5, 4.7 and 3.2, respectively (Table 1). Fitting the model with q fixed at 4 (the average integral value found when the model was fitted with q floating) shows P₅₀ to be 7.7 mmHg, O₂ binding to be strongly cooperative (n₅₀ and n_max=3.2), and O₂ association constants for the Hc in the deoxy (Tense) and the oxygenated (Relaxed) states (K_T and K_R) to be 0.01 and 1.61 mmHg^–1,respectively (Table 1). The identity between n₅₀ and n_max values and the close agreement between P₅₀ and P_m values (Table 1) indicate symmetry of the O₂ binding curves,justifying rigorous analysis of the allosteric effects in terms of P₅₀ values (cf. Wyman,1964).

Calculated asΔ G=RTln[<(L+1)(Lc^q+1)>/<(Lc+1)(L^q–1+1)>],where L is the allosteric constant and c=K_T/K_R(Imai, 1982), the free energy of cooperativity equals 11.84 kJ mol^–1 in the absence of lactate and q=4 (Table 1). As shown (Fig. 2, Table 1) lactate increases O₂ affinity (decreases P₅₀) by lowering L and raising K_T, without markedly markedly altering K_R. This indicates that the underlying molecular mechanism of the lactate effect is to shift the allosteric equilibrium towards the high affinity R state and to increase the O₂ affinity of the T state molecules, indicating that the T-state Hc becomes less stable in the presence of added lactate. The decreased K_R/K_T ratio decreases n₅₀ values (slopes of the plots at log[Y/(1–Y)]=0) from 3.2 in the absence of lactate to 2.4 and 2.2 in the presence of 10 and 50 mmol l^–1 lactate,respectively, and correspondingly lowers ΔG from 11.84 to 7.90 and 6.77, respectively.

Measurement of the O₂ affinity of C. maenas Hc at 10 and 20°C, in the absence and presence of lactate over a wide pH range where the Bohr effect (oxygenation-linked proton binding) changes from normal to absent and reverse (Fig. 3)provides an ideal opportunity to analyse the contributions of individual,oxygenation-linked processes to the overall enthalpy of oxygenation.

As shown (Fig. 3) the Hc exhibits a strong normal Bohr effect φ(=ΔlogP₅₀/ΔpH)=–0.80 at alkaline pH range, which encompasses physiological values in decapods [7.44–7.84(Mangum and Shick, 1972)], no Bohr effect (φ=0) at pH 6.9, and a distinct, reversed Bohr effect(φ=+0.46) at lower pH. The data show a drastic increase in the temperature effect with decreasing pH (Fig. 4), indicating that high temperatures will adversely affect O₂ loading under acidotic and hypoxic conditions.

Fig. 5 shows the relationship between ΔH′ and pH in the absence and presence(Fig. 5 top, curves A and B,respectively) of lactate. In the absence of lactate(Fig. 5 top, curve A) the numerical value of ΔH′ increases from ∼–10 kJ mol^–1 at pH 8 to plateaus of approx. –42 at pH 7.0 and∼60 at pH 6.5, revealing a clear inverse relationship with the Bohr factors (–0.8, 0 and +0.46) at these respective pH values. This indicates that (i) as with vertebrate Hb, oxygenation-linked proton dissociation is endothermic and thus counterbalances the heat released upon Hc oxygenation (ΔH^O₂) in the pH range where the Bohr effect is normal, (ii) the overall ΔH′ value at pH 7.0 is devoid of contributions from proton binding (Δ^H+) and thus reflects the intrinsic heat of O₂ binding by the Hc, (iii)oxygenation-linked proton association (reverse Bohr effect) is exothermic and thus increases the temperature effect contributing about +18 (=–60 to–42) kJ mol^–1 and (iv) in the absence of other effectors, proton concentrations play the dominating role in determining the temperature sensitivity of O₂ affinity in crustacean Hc.

This contention is supported by numerous data in the literature. Particular examples are Hcs of (i) crab Holthuisana transversa, where an exceptionally low Bohr factor (–0.13) correlates with highly exothermic oxygenation reactions (ΔH′=–48 to –62 kJ mol^–1) observed over a range of operating conditions(Morris et al., 1988), (ii)hermit crab Pagurus bernhardus, where a massive Bohr effect[φ=–1.55 (Jokumsen and Weber,1982)] tallies with a complete lack of temperature sensitivity at pH 7.2–7.8, whereas the temperature effect is large at low pH where the Bohr effect is zero (A. Jokumsen and R.E.W., unpublished), and (iii) krill Meganyctiphanes norvegica, and the hydrothermal vent crab Segonzacia mesatlantica, where even larger Bohr factors(φ=–1.9 and –2.9, respectively) are associated with overall endothermic heats of oxygenation [ΔH′=+134 and +17 kJ mol^–1 (Brix et al.,1989; Chausson et al.,2004)].

The differences between ΔH′ in the presence and absence of lactate (depicted by curve B–A, Fig. 5 bottom) reflects the pH-dependent heats of reaction with lactate ions and of coupled reactions. As evident, lactate drastically raises theΔH′ values at alkaline (physiological) pH. Given that lactate increases O₂ affinity (Fig. 1) and thus undergoes oxygenation-linked binding (and deoxygenation-linked dissociation), the data show that lactate binding is exothermic, strongly pH-dependent, and peaks at a pH of approximately 7.6,where it contributes approximately –30 kJ mol^–1 per O₂ molecule bound/released and decreases at higher and lower pH values. The exothermic, oxygenation-linked binding of lactate ions that increases the overall oxygenation enthalpy in crab Hc contrasts sharply with vertebrate Hbs, where endothermic, oxygenation-linked dissociation of organic effectors (2,3-diphosphoglycerate, DPG, and ATP) reduces the overall heat of oxygenation and thus the temperature sensitivity of O₂affinity.

The single previously reported value for the enthalpic contribution of lactate binding to Hc [–25 kJ mol^–1 at pH 7.5 for Hc of the crab Calappa granulata(Olianas et al., 2006)] is in good agreement with our data. Remarkably, when measured directly by calorimetric methods, the enthalpy change for urate binding to Homarus vulgaris Hc is much higher [135 kJ mol^–1 at pH 8.0(Menze et al., 2001)]. The enthalpies of lactate binding to crab Hc are also considerably lower than for DPG binding to human Hb, where similar values (∼46–55 kJ mol^–1) were obtained from van't Hoff plots and calorimetric methods (Benesch et al., 1969; Bunn et al., 1971; Nelson et al., 1974). Literature on mammalian Hbs illustrate how the tyranny of the van't Hoff equation may be bypassed by endothermic processes (e.g. conformational changes and proton and chloride binding) (Wyman et al., 1977; Ikeda-Saito et al.,1983; De Rosa et al.,2004) that reduce the temperature sensitivity of O₂affinity. Arguably, a low ΔH′ value is an advantageous trait for O₂ loading at high temperature when an effector (such as lactate)increases O₂ affinity and thereby raises the exothermic nature of oxygenation, just as low ΔH values resulting from ligand reactions coupled to oxygen binding are advantageous for maintaining a high O₂ affinity in vertebrate Hbs.

Although lactate that is produced under anaerobic conditions increases Hc–O₂ affinity and thus may favour O₂ loading and survival under hypoxic conditions(Truchot, 1980), our data show that a thermodynamic consequence of oxygenation-linked lactate binding is an increased temperature sensitivity in the physiological pH range (cf. Fig. 5), which may compromise O₂ loading at high temperature. While this specific `lactate effect' (ΔH^lact) is maximal near 7.5, the overall temperature sensitivity (ΔH′) remains high at lower pH values (imparted by lactate accumulation), where the Bohr effect successively disappears and reverses (resulting in loss of O₂-linked endothermic proton dissociation and initiation of ectothermic proton binding, respectively). However, apart from hampering O₂ binding at high temperature,increased temperature sensitivity will promote O₂ loading in the gills at low temperature. In this regard it may be significant that increases in lactate levels may induce behavioural hypothermia, since lactate injection reduces the preferred temperature in C. maenas(De Wachter et al., 1997).

In conclusion, the large variations in the effects of temperature on O₂ affinity of crustacean Hcs (observed between different species and within the same individuals at different times) appear to be directly related to variations in the heats of reaction of processes coupled to the oxygenation reaction, notably proton and lactate binding, indicating that hemolymph levels of these ions are major factors controlling the temperature sensitivity of oxygen binding to crustacean hemocyanin.

In this light it would be interesting to investigate the enthalpic components in Hcs that lack a lactate effect [as in the hydrothermal vent crab Syanagraea praedator (Chausson et al., 2001)] or have an opposite lactate effect [as in the land crab Gecarcoidea natalis, where lactate decreases O₂affinity (Adamczewska and Morris,1998)] as well as the thermodynamic consequences of binding other Hc effectors, such as divalent cations, whose concentrations may vary greatly(with water salinity, ionoregulatory capacities and the moult cycle) and urate, which may bind to Hc with a higher affinity than l-lactate[40-fold higher in Homarus vulgaris Hc(Nies et al., 1992)]. In Carcinus maenas, whose Hc has about twice as many urate (and caffeine) binding sites than that of H. vulgaris(Hellmann et al., 2004),hemolymph urate concentrations increase drastically in response to hypoxic exposure (Lallier et al.,1987). In line with the effect of lactate at pH >7.2 reported here (Fig. 5), urate increases the temperature sensitivity of slipper lobster Scyllarides latus Hc,which has four urate binding sites per hexamer, at pH 7.5(Sanna et al., 2004).

Supported by the Danish Natural Science Research Council (FNU) and the Carlsberg Foundation. We thank Anny Bang and Dorte Olsson for valuable technical assistance.