STRATEGIES OF OSMOADAPTATION IN ARCHAEA

The ability to adapt to fluctuations in external osmotic pressure (osmoadaptation) and the development of specific mechanisms to achieve this (osmoregulation) are fundamental to the survival of cells (6, 16, 67, 73, 76). Most cells maintain an osmotic pressure in the cytoplasm that is higher than that of the surrounding environment, resulting in an outward-directed pressure, turgor, whose maintenance is essential for cell division and growth. Changes in environmental osmolarity can trigger the flux of water across the cytoplasmic membrane. Thus, to avoid lysis under low-osmolarity or dehydration under high-osmolarity growth conditions, cells must possess active mechanisms that permit timely and efficient adaptation to changes in environmental osmolarity.

Eubacterial organisms have evolved several strategies that enable them to survive and proliferate in environments of varied ionic composition and salinity ranging from freshwater to hypersaline habitats: (i) the intracellular accumulation of inorganic ions such as potassium; (ii) the evolution of salt-tolerant (and in some cases salt-dependent) enzymes; and (iii) the accumulation, either by transport or synthesis, of selected organic molecules which may be negatively charged (and act as counterions for intracellular K⁺) or neutral. Accumulation of "compatible" solutes is a particularly ubiquitous response. The current understanding is that these compatible solutes maintain an equilibrium between macromolecule surface areas and the water phase by resisting drastic changes in intracellular water density. This is based on the ability of compatible solutes to accumulate at these interface regions (75). For example, the preferential hydration of ribonuclease is 600 mol of water per mol of protein. However, in the presence of sarcosine, a common osmolyte in eukaryotic organisms, the preferential hydration is reduced to 70 mol of water per mol of protein (51). These same solutes also protect macromolecules from thermal denaturation (e.g., glycine-based osmolytes provide an extraordinary degree of protection for hen egg white lysozyme [64]). Protein unfolding results in an increase of total protein surface area. Osmolytes oppose the increase in surface area by a preferential hydration of proteins (1) and favorable interactions with side chains (53). These interactions stabilize proteins by raising the chemical potential of the denatured protein, which leads to contraction of the random coil to a folded structure (53). A recent study quantifying the stability afforded by compatible solutes showed that the osmolyte trimethylamine oxide can increase the population of folded structures compared to denatured protein by nearly 5 orders of magnitude (3).

Archaea, which are often found in high-salt as well as high-temperature environments, use the same general strategies for osmoadaptation as eubacterial and eukaryotic organisms. However, they are notable for the unusual organic osmolytes accumulated. Specific examples of these osmolytes and factors that affect their accumulation are provided in the following sections.

Accumulation of inorganic ions. The ubiquity of K⁺ uptake systems in cell membranes, their high rates for ion transport, and the ability to gate this ion flow might, at first glance, make K⁺ a good candidate for use as an osmolyte in cells. Balanced against this is the observation that high concentrations of inorganic cations often have deleterious effects on enzyme catalytic rates (76). In most eubacteria, the accumulation of K⁺ is an early response to an increase in external NaCl. Glutamate is usually selectively accumulated as the counterion (2, 8). However, the increased K⁺ is often transient and is superseded by the accumulation of zwitterionic organic solutes such as proline or glycine betaine (GB) (16). Thus, osmoadapted eubacteria tend to use zwitterionic or nonionic solutes to counteract external osmotic pressure.

In contrast to eubacteria, most of the archaea examined have high intracellular concentrations of inorganic cations, primarily K⁺, under optimal growth conditions. Table 1 provides values for intracellular K⁺ concentrations in a range of archaea including organisms grown in low-ionic-strength media (e.g., Methanobacterium thermoautotrophicum) and extreme halophiles (e.g., Halobacterium halobium and Natronococcus occultus). Even in the nonhalophiles the intracellular K⁺ concentration is relatively high (>0.5 M), suggesting that under normal conditions these cells exist with high turgor pressure if the intracellular K⁺ is free and not tightly complexed to macromolecules. While there are no published data on whether there are immediate changes in intracellular K⁺ levels with increased external NaCl, steady-state intracellular K⁺ levels have been measured in cells grown at different NaCl concentrations (for examples, see Fig. 1). In many of the halotolerant methanogens (e.g., Methanococcus thermolithotrophicus [grown in medium with less than 1 M external NaCl], Methanogenium cariaci, and Methanosarcina thermophila) and the halophilic Methanohalophilus portucalensis, intracellular K⁺ varies with external NaCl concentration, and hence it can be considered an osmolyte. In other methanogens (e.g., Methanobacterium thermoautotrophicum), it is maintained at a constant level over a moderate range of external NaCl and contributes to high turgor pressure (15). However, in both M. thermolithotrophicus and Methanobacterium thermoautotrophicum there appear to be thresholds of external NaCl above which K⁺ accumulation (and organic counterion accumulation as well) is compromised and is seen to decrease (13, 60). At this point, if the cell is to survive (i.e., resist water efflux and dehydration) it must alter its osmoadaptation strategy. Methanobacterium thermoautotrophicum is unable to adapt effectively and does not grow at high concentrations of NaCl, while M. thermolithotrophicus alters its osmolyte pool and thus can survive and grow in medium containing higher concentrations of NaCl.

View larger version (12K): [in this window] [in a new window]   FIG. 1.   Intracellular K+ as a function of external NaCl in M. thermolithotrophicus () (data are from reference 60 [micromoles/milligram of protein] and were converted to moles/liter by using the conversion that 0.13 µmol of solute/mg of protein corresponds to 0.1 M as determined in reference 15), Methanobacterium thermoautotrophicum () (data from reference 13), and Methanohalophilus portucalensis () (data are from reference 36).

The highest accumulation of K⁺ among the archaea occurs within the halophilic family Halobacteriaceae (30). It has been reported that some Halobacterium spp. possess up to 5 M cell-associated K⁺, which is well beyond the solubility of KCl. This suggests that K⁺ is somehow bound to macromolecules in the cells, but analysis by ³⁹K-nuclear magnetic resonance (NMR) spectroscopy has indicated that the bulk of the K⁺ exists in a free state within these cells (66). For other Halobacterium spp., intracellular Na⁺ can be exchanged for K⁺, creating an intracellular environment that has reduced K⁺ and Na⁺ gradients (34) but which still maintains a very high monovalent cation concentration inside. The concentration of intracellular K⁺ need not be the same as that of extracellular Na⁺ in extreme halophiles. For example, N. occultus grows optimally in medium containing 3.4 M NaCl; the concentration of intracellular K⁺ in this organism is 1.1 M (18). Likewise, in Methanohalophilus portucalensis grown in 2.0 M NaCl, the intracellular concentration of K⁺ is 0.76 M (36). Presumably the cytosolic environment of these cells has other components that are effective at balancing external osmotic pressure.

Salt-tolerant enzymes. To cope with the high intracellular concentrations of K⁺, many archaea have evolved proteins that are exceedingly rich in acidic amino acids (glutamate and aspartate) compared to basic amino acid (arginine and lysine) residues (29). For example, the relative acidities of ribosomal proteins have been compared for a variety of archaea (primarily methanogens) and some eubacteria (29). Even archaea grown in low salt medium have acidic protein fractions (Table 2). The bias for acidic residues results in a net negative charge on the protein surface that presumably prevents folding into the native (active) structure unless a cation counterion like K⁺ is present. For example, malate dehydrogenase from H. halobium (67) requires high salt concentrations in order to maximize activity. In addition to this large excess of negatively charged amino acid residues, the hydrophobicity of the proteins of halophilic archaea is reduced. This in turn reduces the salting-out effects of K⁺ and allows the protein to retain its flexibility under conditions of extreme salinity. However, the requirement for K⁺ for proper protein folding and stability imposes a limit on the effectiveness of K⁺ as a compatible solute (76).

Accumulation of organic solutes. Eukaryotic and eubacterial organisms generally adapt to increased osmotic stress by accumulating highly soluble organic compounds such as polyols (glycerol, arabitol, mannitol, and glucosylglycerol [6, 24]), low-molecular-weight nonionic carbohydrates (sucrose, trehalose, and glucose [28, 74]), free amino acids and their derivatives (proline, glutamate, glycine, -aminobutyrate, taurine, and -alanine [8, 16]), unique organic zwitterions (tetrahydropyrimidines such as ectoine [25]), methylamines (GB and trimethylamine-N-oxide [67]), and -dimethylsulfoniopropionate (71). In general, intracellular accumulation of organic solutes by transport from the culture medium is often preferred over their biosynthesis, and indeed, for certain solutes such as GB, high-affinity uptake systems are available to scavenge any of the material present in complex media.

The distribution of commonly occurring organic osmolytes found in archaea falls into the same major classes as for eubacteria and eukaryotes: (i) sugars and polyhydric alcohols (Table 3) and (ii) amino acids and derivatives including methylamines (Table 4). Archaea also accumulate some very unusual solutes that have no obvious eubacterial or eukaryote counterpart (Table 5). There are striking similarities among the carbohydrates and polyols that are used as osmolytes by eubacteria and archaea (Table 3) with the exception that the majority of the solutes in archaea are modified so that they are anionic. The addition of a negative charge is accomplished with carboxylate, phosphate, and sulfate groups. In the amino acid class, there are two osmolytes common to eubacteria and archaea: anionic L--glutamate and zwitterionic GB. However, methanogenic archaea have developed a novel strategy to produce amino acid-like molecules that are unlikely to interact with any metabolic or biosynthetic machinery in the cells. These anaerobic organisms synthesize and accumulate several -amino acids to balance external osmotic stress (Table 4). -Glutamate is the anionic solute in this group with -glutamine and N-acetyl--lysine as zwitterionic solutes.

(i) Sugars and polyhydric alcohols. Accumulation of polyol phosphodiesters represents a relatively unique osmolyte strategy in archaea. Glycerol is accumulated in response to external NaCl concentrations by several species of yeast and algae (6, 67, 73). The hyperthermophile Archaeoglobus fulgidus accumulates a novel charged version of this compound, diglycerol phosphate (DGP), as its major intracellular organic solute (44). In osmoadapted cells, DGP increases with both increased NaCl and growth temperature, indicating that it behaves as an osmolyte in this organism. Another common polyol used by plants and mammals for osmotic balance is myo-inositol. Archaea use a related phosphorylated inositol, di-myo-inositol-1,1'-phosphate (DIP), for osmotic balance. This rather unusual phosphodiester compound has been identified in hyperthermophilic organisms, including Methanococcus igneus (12), Pyrococcus woesei (65), P. furiosus (42), Thermotoga maritima (43), and Pyrodictium occultum (44). T. maritima is not an archaeon but shares many characteristics with archaeal hyperthermophiles. Both inositol rings of the DIP from M. igneus and Pyrococcus sp. have L-stereochemistry; DIP isolated from T. maritima is reported to occur in chiral and meso-forms. While intracellular DIP concentrations increase in M. igneus with external NaCl, other solutes, notably -glutamate, show a stronger increase with increasing NaCl (12, 43). Perhaps more strikingly, DIP levels also correlate with hyperthermophily. The optimal growth temperature of M. igneus is 85°C. In this organism, DIP is produced only at growth temperatures of 80°C or higher (12). This trend was also seen with P. furiosus, where the concentration of DIP was found to be higher at supraoptimal growth temperatures (98 to 101°C) and with Thermotoga neapolitana (43). These results suggest the possibility of DIP acting as a preferred thermoprotectant in addition to its role as an osmolyte.

(ii) -Amino acids and derivatives. Zwitterionic amino acids, such as proline, are important osmoprotectants in a variety of bacteria. These are usually internalized from the medium by osmotically induced proline transport systems, ProP and ProU (67), that have been shown to be sodium dependent (27). To date there are no reports of proline being accumulated in response to osmotic stress in archaea. Another relatively common eubacterial zwitterionic osmolyte is the cyclic amino acid derivative ectoine. Ectoine is synthesized de novo by the majority of heterotrophic halotolerant eubacteria under conditions of high salt (17, 24, 25) and is assumed to have a protective function similar to those of proline and GB (31). However, ectoine has not been detected in archaea.

View larger version (12K): [in this window] [in a new window]   FIG. 2.   L--Glutamate levels in M. thermolithotrophicus () (data are from reference 60 [micromoles/milligrams of protein] and were converted to moles/liter by using the conversion that 0.13 µmol of solute/mg of protein corresponds to 0.1 M as determined in reference 15), Methanobacterium thermoautotrophicum () (data from reference 13), and Methanohalophilus portucalensis () (data from reference 36).

(iii) -Amino acids and derivatives. Several -amino acids, notably -glutamate, -glutamine, and N-acetyl--lysine (Table 4), function as compatible solutes in methanogenic archaea (36, 56-60, 62, 68, 69). -Glutamate (-aminoglutaric acid), an organic anion, accumulates in response to increasing external NaCl in several halotolerant thermophilic Methanococcus species (12, 60) as well as in mesophiles (e.g., Methanogenium cariaci, a marine methanogen [60]). A zwitterionic derivative, -glutamine, functions as the osmolyte of choice at the highest external NaCl concentrations in several Methanohalophilus species (36, 63). A rather different zwitterionic -amino acid derivative, N-acetyl--lysine, occurs as the predominant compatible solute in response to elevated external NaCl levels in various methanogenic archaea isolated from both marine and nonmarine origins (e.g., Methanosarcina thermophila, Methanogenium cariaci, Methanohalophilus sp., and Methanococcus sp. [36, 58, 60, 68, 69]). Like their L- isomers, the -amino acids are extremely soluble, but they are not metabolized significantly by cells. The stability of -amino acid pools to turnover has been shown by NMR analyses of ¹³C-pulse/¹²C-chase (or ¹⁵N-pulse/¹⁴N-dilution) labeling experiments (56, 60). In these experiments, isotope in L--glutamate pools is rapidly lost as this solute is used to synthesize other amino acids and proteins, etc., while label in the -amino acids turns over very slowly in growing cultures. This property makes the -amino acids true compatible solutes.

(iv) Miscellaneous osmolytes: cDPG and TCH. Several other unusual compounds have been detected at moderate to high concentrations in archaea. These organic polyanions are counterions to the high intracellular K⁺ concentration and contribute to the high turgor pressure of the organisms in which they occur. As noted before, the thermophile Methanobacterium thermoautotrophicum has a high intracellular K⁺ concentration that is relatively independent of external NaCl up to 0.4 M NaCl (see Fig. 1). The cell wall of this organism is extremely robust and is critical for maintaining cell volume under conditions where the organism is clearly not osmotically balanced. Both the Marburg and H strains of this species have similar growth requirements (5). However, they balance the high K⁺ by using different ratios of organic polyanions (13). In Methanobacterium thermoautotrophicum H, the major intracellular solutes are cyclic-2,3-diphosphoglycerate (cDPG) and L--glutamate; 1,3,4,6-tetracarboxyhexane (TCH) is a minor component (for structures see Table 5). In the Marburg strain TCH levels are considerably higher, and the level of this solute increases in response to decreased levels of L--glutamate and other solutes with increased external NaCl. Since TCH has a 4 charge and cDPG has a 3 charge at physiological pH, the intracellular concentrations of these anions are far lower than that of intracellular K⁺. None of these anionic organic solutes exhibits a large osmotic response over a wide range of external NaCl concentrations (0.01 to 0.4 M); the total concentration of negative charges on the solutes correlates directly with the K⁺ concentration (13). Both K⁺ and organic solute concentrations begin to decrease between 0.4 and 0.65 M NaCl. The observation of these novel solutes in Methanobacterium thermoautotrophicum connects other cellular processes to maintenance of turgor pressure (and K⁺ charge balance). cDPG has been shown to be involved in carbon fixation in this organism (26). High levels of cDPG have also been detected in other more thermophilic methanogenic archaea, including Methanopyrus kandleri and Methanothermus fervidus (44), where it is likely to have a role as an osmolyte and a counterion to intracellular K⁺. TCH, as a structural unit of methanofuran, connects osmotic effects and/or turgor pressure with methanogenesis components in these organisms (26).

	BIOSYNTHESIS OF OSMOLYTES

A detailed understanding of the osmotic behavior of archaea (or any organism for that matter) requires knowledge of the biosynthetic pathways or uptake systems for osmolytes and how they are linked to external NaCl. A number of the osmolytes used by archaea are novel compounds and as such present new wrinkles in standard biochemistry and in some cases novel enzymatic transformations. Several of these are discussed below with comments on which steps in a given pathway are the most likely to be regulated or affected by external NaCl.

Conversion of glucose-6-phosphate to DIP: novel enzymes, novel compounds, and high temperatures. DIP biosynthesis in M. igneus (and presumably the other hyperthermophilic archaea where it occurs) consists of the following four steps (9) starting with D-glucose 6-phosphate (Fig. 3): (i) conversion of glucose-6-phosphate to L-myo-inositol-1-phosphate (I-1-P) by I-1-P synthase; (ii) conversion of I-1-P to myo-inositol by a specific phosphatase, I-1-P phosphatase; (iii) conversion of I-1-P to cytidine diphosphoinositol (CDP-I) with CTP as the carrier molecule in the phosphoryl transfer of the CTP:I-1-P cytidylyltransferase reaction, and (iv) nucleophilic attack of the free hydroxyl group of C-1 of myo-inositol on CDP-I to produce DIP (DIP synthase). Steps i and ii have been elucidated in detail for plants and other organisms. Analogs of step iii occur in many cells as part of phospholipid biosynthesis, although this is not the way phosphatidylinositol is synthesized in cells (CDP-diacylglycerol is the activated species). Step iv, the novel DIP synthesis reaction, is based on phosphatidylinositol synthesis in both eukaryotic and bacterial cells with myo-inositol attacking CDP-I instead of the CDP-diacylglycerol. The enzymatic steps unique to DIP and not to inositol lipids are the CTP:I-1-P cytidylyltransferase and DIP synthase reactions. Thus, one might suggest that these would be steps sensitive to external osmotic pressure.

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Biosynthetic pathway proposed for DIP (adapted from reference 9).

Archaea include a large number of hyperthermophiles, and these are often isolated from environments with moderate to high concentrations of NaCl. Whereas growth temperature appears to have little effect on the accumulation of amino acid type osmolytes, it can have a large impact on synthesis and accumulation of sugar and polyol solutes. The effects of growth temperature on DIP biosynthesis in M. igneus have been explored in detail. In the case of M. igneus DIP does not contribute to the osmotic balance until the growth temperature is higher than 80°C (12).

Why is DIP accumulated only at high growth temperatures? In vitro studies of the temperature dependence of enzyme activities in M. igneus involved in biosynthesis of DIP provide some insight into this question. The condensation of CDP-I and myo-inositol by DIP synthase in crude protein fractions has a high activation energy, ~120 kJ · mol¹. For comparison, the first two steps in the DIP biosynthetic pathway, I-1-P synthase and I-1-P phosphatase, exhibit much lower activation energies, 60 to 70 and ~50 kJ · mol¹, respectively (9). At lower growth temperatures, any I-1-P generated may be hydrolyzed to myo-inositol, possibly for incorporation into lipids. I-1-P cannot build up to levels needed for conversion to CDP-I and eventual DIP synthesis. As the growth temperature is increased, however, not only is more I-1-P generated but DIP synthase activity is enhanced. DIP synthesis in other thermophiles should be examined to see if this hierarchy of activation energies is part of a universal trend or is unique to this methanogen.

From "X" to -glutamate to -glutamine ... . -Glutamate is the most ubiquitous of the -amino acids found in archaea and a few eubacteria as well. However, the biosynthetic pathway for this species is not yet known, although information is available on what potential pathways do not generate -glutamate in methanogens. Studies of transaminase activities in M. thermolithotrophicus clearly show that -ketoglutarate cannot be converted to -glutamate by using alanine or aspartate as nitrogen donors in the reaction (41a). In the same fashion both ¹³C- and ¹⁵N-labeling experiments and incubation of crude protein extracts with -glutamate do not yield the -isomer (which would require a glutamate aminomutase activity) (41a). Two other routes to -glutamate that involve soluble precursors can be proposed. (i) -Ketoglutarate is reduced to -hydroxyglutarate, which is then dehydrated to glutaconic acid. Addition of ammonia to the glutaconic acid would then generate -glutamate. (ii) A three-carbon propionate unit with an aldehyde or amino group at C-3 is complexed with pyridoxal-5-phosphate and then condensed with an activated two-carbon unit (likely to be acetyl-coenzyme A). Since -hydroxyglutarate has been detected in methanogens, it is possible that the first scheme is operational although there is at present no firm evidence. Clearly this is an area for further investigation.

Biosynthesis of -glutamine in Methanohalophilus portucalensis occurs by the activity of a glutamine synthetase on -glutamate. The accumulation of -glutamine as an osmolyte at high concentrations of external NaCl is interesting in view of what is known about glutamine synthetases (GS) from a large number of organisms. In general this multimeric, highly (and diversely) regulated enzyme converts -glutamate to -glutamine poorly (i.e., high K_m and low V_max compared to those of -glutamate as a substrate) (33). The one archaeal GS that has been purified (from Methanobacterium ivanovi) and characterized was not examined for its ease in converting -glutamate to -glutamine. In vitro, Methanohaliphilus portucalensis uses -glutamate as a direct precursor of -glutamine (63a), suggesting that the GS from this methanogen has different kinetic and regulatory properties from the GS of eukaryotes and eubacteria. The V_max for this conversion is ~0.14 that for L--glutamate conversion to -glutamine. However, while cells supplemented with exogenous -glutamate (under conditions where the -glutamate was internalized) showed enhanced -glutamine, the neutral -amino acid was not the major solute (63). This might suggest that GS conversion of -glutamate to -glutamine and accumulation of this zwitterion (as one of the trio of GB, N-acetyl--lysine, and -glutamine) are regulated indirectly by external NaCl.

N-Acetyl--lysine: control of lysine aminomutase? At low osmolarity (<1 M NaCl) -glutamate and -glutamate are the primary organic solutes in M. thermolithotrophicus. When adapted to growth at higher salinity (Fig. 4), the organism accumulates N-acetyl--lysine, and it rapidly becomes the major solute as external NaCl is increased still further (59, 62). Nothing is known about what regulates the transition from accumulation of the anionic glutamate isomers to accumulation of the zwitterion at high external NaCl. Perhaps K⁺ levels reach a saturation point where further cation accumulation would be detrimental to cell function. In order to avoid this the cell adapts by accumulation of something zwitterionic. Another possibility is that N-acetyl--lysine biosynthesis is controlled by Na⁺ and/or K⁺ and this solute is synthesized only when intracellular K⁺ levels are high. The synthesis of N-acetyl--lysine involves lysine 2,3-aminomutase conversion of -lysine to -lysine followed by acetylation of the side chain amino group (56, 60). Since -lysine is necessary for all protein biosynthesis, it is unlikely that any of its biosynthetic enzymes are osmotically regulated. A more likely way of controlling N-acetyl--lysine synthesis and accumulation would be to regulate lysine 2,3-aminomutase. Lysine aminomutase activities from bacteria that metabolize -lysine as a carbon and nitrogen source have been purified and characterized (10, 22), and a very oxygen labile lysine aminomutase activity has been detected in anaerobic extracts of M. thermolithotrophicus (14).

View larger version (17K): [in this window] [in a new window]   FIG. 4.   (A) Intracellular organic osmolytes in M. thermolithotrophicus as a function of adapted growth in different concentrations of NaCl. , L--glutamate; , -glutamate; , N-acetyl--lysine; ×, K+. (B) Scheme indicating how M. thermolithotrophicus responds to an increase in external NaCl. (1) Initially the cell is balanced at a particular concentration of external NaCl; S represents intracellular solutes. (2) Immediately after the cells are subjected to increased external NaCl, water will tend to rush out of the cells. (3) The most immediate osmotic response of these cells is to internalize K+ (within minutes) at a concentration that is in excess of those of the organic solute counterions available in the cell; over a time scale of 10 to 15 min a new intracellular K+ concentration is reached. (4) Over a longer period of time (25 to 60 min), L--glutamate is accumulated as a counterion to the K+ (cells are still not growing and are in a lag phase). (5) After cells begin to grow, there is synthesis and accumulation of the other organic osmolytes in the ratios seen in adapted cultures.

Further work must be done to better understand what controls the osmoadapted response. However, what is known is that K⁺ plays an important role in the immediate (and transient) response of this organism to hyperosmotic (increased NaCl) or hypoosmotic (decreased NaCl) shock (14). Within the first few minutes after transfer to the higher-NaCl-containing medium, M. thermolithotrophicus internalizes K⁺. This is followed by a decrease to steady-state levels over a time course of 15 to 20 min. Once the K⁺ reaches a new steady-state concentration, synthesis and accumulation of L--glutamate occur. The K⁺--glutamate pair functions as a temporary osmolyte. Once growth of the M. thermolithotrophicus culture begins, typically 60 to 90 min after the increase in NaCl concentration, the nonmetabolizable zwitterion (N-acetyl--lysine) is synthesized and accumulated. A schematic diagram suggesting the steps in the osmotic response of this organism is shown in Fig. 4B. That N-acetyl--lysine accumulation occurs in osmoadapted cells (i.e., after the lag phase) is consistent with the hypothesis that one of its biosynthetic enzymes is not available in cells grown and adapted to lower NaCl.

GB: methylation of glycine depends on intracellular K⁺. Methanohalophilus portucalensis is relatively unique in the archaeal domain in that it can synthesize GB de novo (36). The biosynthesis of GB in this organism occurs by methylation of glycine (56) with S-adenosylmethionine as the methyl donor (37). More importantly, in vitro studies show that the methylation reactions are K⁺ dependent, with little betaine accumulating below 0.4 M K⁺; at this potassium concentration sarcosine was the major product. K⁺ concentrations above 0.4 M enhanced GB synthesis from sarcosine and N,N-dimethylglycine (37). This observation suggests that control of betaine synthesis is modulated by intracellular K⁺; this cation may also function as an intracellular signal for osmoregulation in M. portucalensis.

	FUTURE DIRECTIONS

Studies of osmoadaptation in many organisms have shown that this is a complex process that involves osmolyte accumulation, solute discrimination, control of ion flux, signaling, and change in protein expression. The last of these, change in protein expression, has been explored in yeast and mammals (7) but has not been examined in archaea, with the exception of the halophilic archaeon Haloferax volcanii. Protein expression, as determined at the gene and protein levels, in this archaeon is affected by changes in external NaCl (21, 47). Differential expression occurs under both low and high salt conditions. Several of these proteins may be heat shock proteins, but it is possible that others may have more specific roles in osmoregulation (synthesis of osmolytes or transport of ions or solutes, etc.). The end result of osmoadaptation, specific solute accumulation, has been characterized for many archaea. However, the physiological and biochemical details of how these solutes are accumulated are not well understood. A number of unanswered questions remain regarding these cells. For example, the transient response of M. thermolithotrophicus to increased external NaCl is similar to that of the eubacterium Escherichia coli. In E. coli, there is a rapid accumulation of excess K⁺, which then decreases to a steady state, followed by accumulation of L--glutamate. Further changes in the L--glutamate pool over a longer time course lead to a redistribution of solutes that reflects the composition of osmoadapted cells at the higher NaCl growth conditions. Is this a general response in other archaea as well? Additionally, why do many of these cells synthesize several different solutes (i.e., Methanohalophilus portucalensis uses three zwitterions and M. igneus uses three anions) rather than produce just one solute for osmotic balance? What is the biochemical basis for why some cells change osmolyte strategies as a function of external NaCl or growth temperature (i.e., why is N-acetyl--lysine produced in M. thermolithotrophicus only at high salt concentrations). Given the unusual collection of archaeal osmolytes, a deeper understanding of the protective effect of these solutes on macromolecular structure will also be needed to shed light on mechanisms of osmoregulation and osmoadaptation.