ABSTRACT The presence of trimethylamine N-oxide (TMAO) was recently discovered in the tissues of some Antarctic notothenioid fishes. This discovery raises the questions of whether the effects of TMAÖ on protein function vary with temperature and whether TMAO plays the same role in notothenioids as it does in elasmobranchs. Two temperature-sensitive kinetic properties - apparent Kn of pyruvate and kea- of muscle A4- LDH in Gillichthys mirabilis, a warm-temperature fish with negligible amounts of TMAC in its cells, and Parachaenichthys charcoti, an Antarctic fish, are not significantly affected by the presence of 150 mM TMAO at most temperatures from 0° to 30°C. The presence of TMAO did not appear to significantly affect the loss of activity in the LDH of any of the five teleost species studied over a period of 24 days. The Vmax of the LDH reaction in these species was slightly inhibited by 150 mM TMAO at both 0° and 20°C. TMAO may function by increasing the rigidity of the enzyme at hinge regions which play a role in establishing the rates of conformational change. TMAO is known to counteract urea’s perturbing effects on protein function in elasmobranchs at an approximately 2:1 sureal:[TMAO] ratio. However, the effects of temperature on this counteraction had not been studied. Below 20°C TMAO is able to fully counteract the denaturing effect of urea, while TMAO’s ability to counteract urea at those same concentrations appears diminished at 30°C or above. This varying counteracting effect, as observed on the Michaelis-Menten constant, appears to be consistent among species, including the subtropical cow-nosed ray (Rhinoptera bonasus) the temperate catshark (Parmaturus xaniurus) and the Antarctic skate (Bathyraja easoni). suggesting a lack of species-specific adaptation to solute effects. INTRODUCTION Trimethylamine N-oxide (TMAO), a small organic solute, was recently discovered in the tissues of some members of the notothenioid suborder of teleost fish (Raymond and DeVries, 1997). Notothenioids are primarily Antarctic species, with some members living in warmer waters such as those off the tip of South America. The function TMAO serves in these organisms is a matter of some speculation, as the molecule may not function in the same context in which it has previously been studied in different organisms. TMAO has been known to function as an osmolyte, but it also affects protein function and stability. Its presence in notothenioids may indicate a common function in different marine organisms, or it may reveal divergent roles for a single molecule. The ability of TMAO to counteract urea’s denaturing effects on proteins has been well studied in elasmobranchs, which use both molecules for osmotic regulation. While urea is known to denature proteins, TMAÖ is thought to have a counteracting, stabilizing effect. Past studies have shown that TMAO can counteract urea’s effects on protein function when both solutes occur at an approximately 1:2 concentration ratio of TMAO to urea (Yancey and Somero, 1980). Further studies, involving rabbit lactate dehydrogenase, showed that the increased rate of loss of protein activity over time caused by urea can be partially counteracted by the presence of TMAÖ, although TMAO alone causes a slightly increased rate of loss of activity (Baskakov and Bolen, 1998). Based on TMAO's effect on elasmobranch and rabbit proteins, TMAO may be hypothesized to serve any of several functions in notothenioids, including: 1) Although elasmobranchs are osmoconformers, notothenioids are hypo-osmotic, so they face the problem of freezing in Antarctic waters. In order to balance the high salt concentrations in their blood, notothenioids may use TMAÖ as an osmolyte which could function as a freezing point depressant. 2) Although urea is not present in significant amounts in notothenioids, TMAO may serve as a stabilizer at low temperatures, reducing the natural rate of protein degradation. If this were true, we would expect that the rate of loss of protein activity over extended periods of time would be reduced in the presence of TMAO at physiological concentrations. 3) TMAO may have adaptive value to notothenioids by altering the binding affinity and/or efficiency of enzymes at low temperatures. If binding affinity were affected, we would expect that in the absence of TMAO, K,, values (apparent Michaelis-Menten constants) would be higher than the expected range for teleost fish, and that the presence of TMAO would lower those values into the expected range. If per molecule efficiency of enzymes were affected, we would expect that kea values (catalytic rate constants) would rise in the presence of TMAO at low temperatures in order to counteract the reduced protein flexibility caused by reduced thermal energy. 4) Finally, TMAO may serve no significant function at all in the context of protein structure and function; outside of its role as an osmolyte, it may be a remnant of evolutionary divergence, or a “Spandrel of San Marcos" (Gould and Lewontin, 1979). If this were the case, we would expect TMAÖ to have no effect on kinetic measurements such as Kn or keat, or on rates of loss of residual activity over time. I have addressed these questions using muscle lactate dehydrogenase (Az-LDH) from various notothenioid and elasmobranch species. LDH is a useful enzyme to study because it is a fairly universal element in the metabolic pathways of complex organisms. Furthermore, the enzyme’s sequence is extremely well conserved; orthologous homologues tend to differ at only a few positions. (Holland et al., 1997) LDH is easy to work with in the laboratory because of its stability across a wide range of conditions. The extensive research already completed on LDH provides a foundation for the characterization of temperature and solute effects. Specifically, the effects of TMAO and urea on protein stability and function have been studied using elasmobranch LDH, making this study relevant to past data (Yancey et al., 1982). The kinetic effects of TMAÖ were studied in a range of teleosts and elasmobranchs in order to compare the notothenioids, which use TMAÖ alone, with elasmobranchs which use the TMAO-urea counteraction system. Both warm and cold-water organisms were studied in order to elucidate possible temperature adaptations. Two Antarctic stenothermal notothenioids were studied: Parachaenichthys charcoti and Notothenia coriiceps. These fish live from -1.86°C to -1°C year-round. Two South American notothenioids also were studied: Patagonothen tessellata and Eleginops maclovinus, both of which live at approximately 2-10°C. P. tessellata is probably phylogenetically closer to the Antarctic species than E. maclovinus is, potentially revealing different solute adaptations (Balushkin, 1992). A fifth teleost, Gillichthys mirabilis, was studied in order to compare the four relative stenotherms to a temperate eurytherm. G. mirabilis, which lives at approximately 5-30°C, might be expected to show less adaptation to the presence of TMAO, because its tissue likely contains negligible amounts of the solute. Three elasmobranchs were studied: Parmaturus xaniurus, a cold temperate catshark species from Monterey Canyon which lives at approximately 2-10°C; Bathyraja easoni, an Antarctic stenothermal skate which lives at -1.86° to -1°C; and Rhinoptera bonasus, a cow-nosed ray species from Tampa Bay, Florida which lives at approximately 20-32°C. The catshark is known to use the TMAO-urea counteraction system, but the solute systems in the rajai remain to be characterized. 1 expected cold-water teleosts and elasmobranchs to have higher apparent Ky, curves than warmer acclimated species, since closely related species should have similar apparent Kn values at their respective physiological temperatures. Stenotherms would be expected to have steeper sloped apparent K,, curves than eurytherms due to their adaptations to narrow temperature ranges. Apparent Kn curves were generated for LDH in the presence of TMAO for the teleosts and in the presence of TMAÖ and urea for the elasmobranchs to determine whether shifting or flattening of the curves occurs, possibly indicating temperature-related adaptations of protein function in the presence of these solutes. Apparent Kn of NADH and Kn of pyruvate curves were generated in order to examine possible differences. In order to examine solute effects on the catalytic rate constants of LDII, I examined the relative maximal velocities of LDH in the absence and presence of TMAO. If TMAO slows enzyme activity by preventing substrate binding in a competitive manner, then relative maximal velocities should be unaffected. However, if TMAÖ works by changing the conformation or overall rigidity of LDH, then TMAO should lower the maximal velocity. MATERIALS & METHODS Purification of LDII White muscle was added to a 50 mM potassium phosphate (pHI 6.8) buffer in an approximately 1:1 volume ratio, and homogenized. The mixture was centrifuged at 11,951 g for 40 min at 4°C. The supernatant was filtered through glass wool, and 0.5 M potassium chloride, 0.2 mM dithiothreitol (DTT), and 200 uM NADH were added. A gravity column with oxamate agarose was rinsed with 100 ml 50 mM KzPO4 buffer (pH 6.8) containing 0.5 M KCl, and then with 100 ml of the same buffer containing 200 uM NADH. 100 ml of the homogenate supernatant solution were passed through the column, followed by 100 ml of the buffer with NADH. The LDH was eluted with 100 ml KzPO4 buffer containing 0.5 M KCl and 10 mM pyruvate. Fractions of elutant were collected and assayed for LDH activity spectrophotometrically at 340 nm. The active fractions were combined and placed in a dialysis bag. The dialysis bag was placed in 2 L of dialysis buffer (potassium phosphate, pH 6.80020°C) for 4 hours, and another 2 L for approximately 14 hours. Kn of NADII Kn of NADH values were measured for Az-lactate dehydrogenase at 0, 10, 20, and 30°C with and without 150 mM trimethylamine-N-oxide (TMAÖ Dihydrate, Sigma Chemical Company) present. Each K„, was calculated using a buffer containing 80mM imidazole-Cl (pH 7.0 ( 20°C), 2 mM pyruvate (Pyruvic acid, Sigma Chemical Company), and eight different concentrations of NADH (Boehringer Mannheim). Reaction velocities were measured by tracking NADH oxidation spectrophotometrically at 340 nm on a Perkin-Elmer Lambda 3B UV/ViS Spectrophotometer. The pyruvate reductase reaction was started by adding 25 ul of LDH at various dilutions depending on the reaction temperature. Apparent Kn, of pyruvate values were measured in the manner described above, except that NADH concentration was maintained at 150 uM and pyruvate was varied among eight concentrations. Kn of pyruvate values were measured for Az-LDH with no solute present, or in the presence of 150 mM TMAO for teleost LDHI reactions and 200 mM TMAO for elasmobranch LDH reactions. Elasmobranch enzymes were also measured in the presence of 400 mM urea (Bio-Rad Laboratories) and in the presence of both 400 mM urea and 200 mM TMAO. Teleost LDHs were incubated with 150 mM TMAO for the appropriate reaction velocity measurements. In order to prevent rapid enzyme degradation by urea, elasmobranch LDHs were not incubated with solutes present. Cold Denaturation of teleost Aj-LDIIs The cold denaturation experiments were designed to determine whether the rate of loss of LDH activity at 0°C over time changes in the presence of TMAO. Enzyme activity was measured at 20°C using reaction velocities measured spectrophotometrically at 340 nm with constant 150 uM NADH and 2 mM pyruvate concentrations. Three mixtures were tested for denaturation over time: LDH with no solute (control); LDH with 150 mM TMAO; and LDH with 600 mM TMAO. Bovine Serum Albumin (Sigma Chemical Company) at a concentration of I mg/ml was added to the enzyme to reduce LDH absorbance onto the walls of the container and to control for the effects of protein concentration on loss of enzyme activity. DTT at a concentration of 0.2 mM was added to the enzyme to prevent oxidation of cysteine side chains. An 80 mM imidazole buffer (pH 7.0 (20°C) was used, and the reaction mixtures were incubated on ice until the time point at which they were assayed. Activities were determined spectrophotometrically as described above at 20°C For each time point, duplicates were measured for each of the three reaction mixtures. Time points were measured at 0, 6, 24, 48, 96, 192, 408, and 576 hours. Relative Maximum Velocities The activity of the Aj-LDH of each of the five teleost species was measured in a buffer containing 80mM imidazole-Cl (pH 7.0 ( 20°C), 150 pM NADH, and 2mM pyruvate. Five replicates were run for each species with and without 200 mM TMAO. Velocities were measured spectrophotometrically at 340 nm at 20°C, and each reaction was started by adding 25 ul of LDH at various dilutions depending on the species. RESULTS The apparent K» of NADH values for A4 LDH of G. mirabilis were not significantly affected by 150 mM TMAO from 0°C to 30°C (Fig. 1). The apparent Kn of pyruvate values were not significantly affected by 150 mM TMAO from 0°C to 30°C in G. mirabilis, while the same values were not significantly affected in P. charcoti except at 30°C, where TMAÖ significantly decreased the apparent Kn of pyruvate value (Fig. 2). The apparent K,, values rose with temperature for both species, with a steeper curve for LDH of P. charcoti. Whereas the apparent K„ of pyruvate for LDH of P. charcoti increased by 12% at 10°C in the presence of 150 mM TMAO, the apparent K» dropped by 19% in the presence of 200 mM TMAO (from 0.2776 + 0.0108 without TMAO to 0.2248 + 0.0054 with TMAO) The relative maximum velocities of the five teleost LDHs in the presence of 150 mM TMAO were approximately 88-94% of the velocities without solute present at 0°C, and approximately 93-99% of the velocities without solute present at 20°C. (Fig. 3). A chi¬ squared test showed that the (Vmag, 150 mM TMAO)/(Vmax, no TMAO) ratio was higher at 20°C than at 0°C (p=1.54x10**), but an analysis of variance showed that the ratios are not significantly different when variances are considered. During cold denaturation, the relative rates of loss of activity of the teleost LDHs were not significantly affected by the presence of 150 mM or 600 mM TMAO over 576 hours (data not shown). In all three elasmobranch species, the apparent K», values rose significantly in the presence of 400 mM urea at 5°C or above and dropped significantly in the presence of 200 mM TMAO at 20°C or below, relative to the values in the absence of any solute. (Fig 4-9) For orthologs of the catshark and the cow-nosed ray, the values in the presence of the TMAO-urea combination did not significantly differ from the values with only urea present at 30°C or above. At 20°C or below in all three species, the TMAO-urea values did not differ significantly from the values in the absence of any solute. DISCUSSION The lack of a significant effect of TMAO at a concentration of 150 mM on the binding affinities of either G. mirabilis or P. charcoti LDH across a wide range of temperatures suggests that the solute does not affect teleost LDH by competitive binding. Since the binding affinity is apparently unaffected but maximum velocities are slightly lowered by TMAO, the solute may affect LDH by altering only the aspects of the enzyme which are operational once the substrate is bound. The catalytic domain of LDH is comprised primarily of alpha helical regions connected by regions known as hinges (Abad-Zapatero et al., 1987); these hinges are where much of the inter-species sequence variation occurs (Holland et al., 1997). Because the enzyme has a catalytic loop which closes around the substrate and cofactor during the pyruvate reduction, the flexibility of the hinge regions has been hypothesized to be a key determinant of the overall reaction speed. TMAO may alter the rigidity of a specific hinge in the enzyme, slowing down the closing of the catalytic loop prior to pyruvate reduction or the opening of the catalytic loop after the conversion. The similarity of TMAO's effect on apparent Ky values and maximum velocities between the Antarctic and South American stenotherms and the temperate eurytherm suggests that no species-specific adaptation to the presence of TMAO has taken place, despite the fact that TMAÖ is present in much larger quantities in the notothenioids than in the goby The larger effect on apparent Kn, of P. charcoti LDH seen in the presence of 200 mM TMAO compared to 150 mM TMAO may indicate that teleosts contain the solute in levels low enough to avoid any significant effect of enzyme function. If these fish were to accumulate greater than 200 mM TMAO, their enzymes might be affected the way elasmobranch enzymes are in those conditions. TMAO’s effect on the maximum velocity of LDH’s pyruvate reduction is only slightly affected by temperature, with a reduced inhibition at 20°C. The reduction of maximum velocity seen in the presence of 150 mM TMAÖ is consistent with TMAO's hypothesized protein-stabilizing role, since decreased flexibility would reduce the per-molecule turnover rate of the enzyme. According to the mechanism proposed by Arakawa and Timasheff (1985), TMAO’s unfavorable interaction with water would result in the exclusion of the solute from the enzyme's contact area with the water around it, making a small surface area more favorable for the protein and thereby causing stabilization. If this mechanism is true, the change in the water’s surface free energy caused by the presence of TMAO does not appear to vary significantly with higher thermal energy. A slightly lessened inhibition of Vmax at 20° relative to 0° complements the data showing a decreased counteraction effect of TMAÖ at higher temperatures; at higher thermal energies, TMAO's effect on enzyme function may be decreased. The cold denaturation portion of the study revealed long-term (at least 24 days) resistance to denaturation of LDH of all five teleost species. The data (not shown) illustrate the lack of any significant difference in the rate of loss of activity in the presence of different TMAO concentrations. The percent of initial activity data generated by Baskakov and Bolen (1998) for rabbit Az-LDH showed a large drop (approximately 25%) in relative activity in the presence of 600 mM TMAO relative to no TMAO by about one week. By comparison, none of the five teleost species I studied showed less than 90% relative activity after more than three weeks. This discrepancy may be due to the different buffer used by Baskakov and Bolen (0.2 M tris-Cl, versus 0.16 M imidazole-Cl in my experiment) or inherent differences between mammalian and teleost orthologs of the enzyme. The effects of TMAO and urea on the apparent K» values of Aj-LDHs of the elasmobranchs are consistent with TMAO’s role as a protein stabilizer and urea’s role as a denaturant. TMAO increases LDH's binding affinity for pyruvate, while urea lowers this affinity. In the catshark and the Antarctic skate, the 1:2 ratio of TMAO to urea appears to give almost complete counteraction of these kinetic effects, as is consistent with past knowledge of solute counteraction. This does not necessarily mean that these two species have adapted to the presence of TMAO and urea; the ability of the two solutes to counteract each other may depend on conditions other than LDH sequence variation, such as other small molecules present in the organism, or the counteraction may be inherent to aqueous protein systems (Lin and Timasheff, 1994). The lack of complete counteraction at high temperatures in the warm water ray further supports the idea that the elasmobranch LDH has not adapted to the presence of these two solutes, since we would expect that TMAO and urea’s combined effect on LDH function would not differ significantly over temperatures within the animal’s physiological range. This change from almost complete counteraction at 20°C and below to only slight counteraction at 30°C and above is consistent between the Florida ray and the catshark, even though 30° is well above the physiological range of the shark. This similarity suggests that TMAO’s ability to counteract urea’s effects on protein function diminishes as temperature increases, and furthermore that this diminishing effect is not the result of species-specific adaptation. The implications of the solute effects studied in this paper depend on the assumption that the elasmobranchs studied share common solute concentrations. Measurements of the urea and TMAO levels in the tissues of these species could reveal different TMAO levels among species which live in very different environments. If this is the case, I expect that elasmobranchs which live in warm, shallow waters would contain more TMAO than those which live in cold, deep water. Increased TMAÖ levels may compensate for TMAO’s possibly diminished counteraction of urea at high temperatures. LITERATURE CITED Abad-Zapatero, C., et al. 1987. Resined crystal structure of dogsish M, apo-lactate dehydrogenase. Journal of Molecular Biology. 198:445-467. Arakawa, T., and S. N. Timasheff. 1985. The stabilization of proteins by osmolytes. Biophysical Journal. 47:411-415. Balushkin, A. V. 1992. Classification, phylogenetic relationships, and origins of the families of the suborder Notothenioidei (Perciformes). Voprosy ikhtiologii. 32(3):3-19. Baskakov, I., and D. W. Bolen. 1998. Time-dependent effects of TMAO/urea on LDH activity: An unexplored dimension of the adaptation paradigm. Biophysical Journal. In press. Baskakov, I., et al.. 1998. TMAO counteracts urea essects on rabbit muscle LDH function: A test of the counteraction hypothesis. Biophysical Journal. In press. Gould, S. J., and R. C. Lewontin. 1979. The spandrels of San Marco and the Panglossian paradigm: a critique of the adaptationist programme. Proceedings of the Royal Society of London. B 205:581-598. Holland, L. Z., et al. 1997. Evolution of lactate dehydrogenase-A homologs of barracuda fishes (Genus Sphyraena) from different thermal environments: Disserences in kinetic properties and thermal stability are due to amino acid substitutions outside the active side. Biochemistry. 36:3207-3215. Lin, T., and S. N. Timashesf. 1994. Why do some organisins use a urca-methylamine mixture as osmolyte? Thermodynamic compensation of urea and trimethylamine N-oxide interactions with protein. Biochemistry. 33:12695-12701. Raymond, J. A., and A. L. DeVries. 1997. Elevated concentrations and synthetic pathways of trimethylamine oxide and urea in some telcost sishes of McMurdo Sound, Antarctica. Fish Physiology and Biochemistry. In press. Yancey, P. H., et al. 1982. Living with water stress: Evolution of osmolyte systems. Science. 217:1214- 1222. Yancey, P. H., and G. N. Somero. 1980. Methylamine osmoregulatory solutes of clasmobranch fishes counteract urea inhibition of enzymes. Journal of Experimental Zoology. 212:205-213. FIGURE LEGENDS Fig. 1: Temperature vs. K» of NADII for A--LDII of G. mirabilis. LDH was incubated on ice and added in 25ul aliquots to reaction mixture containing 80 mM imidazole-Cl buffer (pH 7.00020°C), 2 mM pyruvate and varying concentrations of NADH. Reaction velocities were determined spectrophotometrically at 340 nm in triplicate. Error bars indicate one standard deviation as determined by Willman 4.0 software. The presence of TMAO had no significant effect on apparent K,, values from 0° to 30° Fig. 2: Temperature vs. Kn of pyruvate for A--LDII. Reaction mixtures contained 80 mM imidazole-Cl buffer (pH 7.0(020°C), 150 uM NADH, and varying concentrations of pyruvate. Reaction velocities and error bars were determined as described above Except for P. charcoti at 30°C, apparent Kn, values were not significantly affected by 150 mM TMAO. Fig. 3: Relative Vmax for Ag-LDII (150 mM TMAO over no solute). Reaction mixtures included 80 mM imidazole-Cl buffer (pH 7.00020°C), 150 uM NADH and 2 mM pyruvate. Error bars represent one standard deviation based on five velocity measurements made at each temperature for each species. Fig. 4: Temperature vs. Kn of pyruvate for A--LDII of Parmaturus xaniurus. See description of Fig. 2 for procedures. Fig. 5: Temperature vs. K» of pyruvate for A--LDII of Bathyraja easoni. See description of Fig. 2 for procedures. Fig. 6: Temperature vs. K., of pyruvate for AJ-LDII of Rhinoptera bonasus. See description of Fig. 2 for procedures. Fig. 7: Relative Kn of pyruvate for A--LDII of Parmaturus xaniurus. Values are relative to apparent Kn values with no TMAÖ or urea present. Fig. 8: Relative Kn of pyruvate for A--LDII of Bathyraja easoni. Values are relative to apparent Kn values with no TMAO or urea present. Fig. 7: Relative K,, of pyruvate for A--LDII of Rhinoptera bonasus. Values are relative to apparent K,, values with no TMAÖ or urea present. Figure 1 15 .............. 10 5 — No solute -E- 150 mM TMAO 5 10 15 20 Temperature (?C) Figure 2 0.8 —— G. mirabilis, no solute 0.7 E— G. mirabilis, 150 mM TMAO -Q— P. charcoti, no solute 0.6 —+ P. charcoti, 150 mM TMAO — 0.5 0.4- 0.3 0.2 0.1 0.0 5 10 20 Temperature (°C) 25 30 35 .................... 25 30 35 Figure 3 1.4 G. mirabilis P. charcoti 1.2 N. coniceps E. maclovinus 1.0 IP. tessellata 0.8 0.6 8 0.4 0.2 0.0 20 Temperature (?C) Figure 4 0.45 — No solute 0.40 —A- 200 mM TMAO 0.35 •- 400 mM Urea -E— TMAO + Urea 0.30 0.25 0.20 0.15 0.10 005 .... ...... 0.00 - -5 0 5 10 25 30 35 20 Temperature (°C) Figure 5 0.15 —— No solute E- 200 mM TMAO —A- 400 mM Urea •- TMAO + Urea 0.10 0.05 0.00 -5 15 20 Temperature (°C) Figure 6 0.50 — No solute ....................................... 0.45 -A- 200 mM TMAO 0.40 - 400 mM Urea -E— TMAO + Urea 0.35 **................. ......... 0.30 0.25 0.20 0.15 W 0.10 O5 0.00 30 35 40 10 15 20 25 Temperature (?C) 25 Figure 7 2.0 1.8 5 16 4 214 . 1.2 — . ....... 1.0 — 9 06 •- 200 mM TMAO 04 - 400 mM Urea 02 • TMAO + Urea 0.0 5 10 15 20 25 30 35 Temperature (?C) Figure 8 2.0 — 18 E 4— 1.2 1 0.8 - 200 mM TMAO .... 0.4 - 400 mM Urea —- TMAO + Urea E 0.2 0.0 10 20 Temperature (°C) 25 Figure 2.0 - 18- 1.6 ................................................. 1.4 1.2 1.0 S — 0.8 0.6 200 mM TMAO 0.4 400 mM Urea -0.2 • TMAO + Urea 0.0 25 10 20 30 35 Temperature (°C) 40