Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Morii, H. Articles by Koga, Y. Search for Related Content PubMed PubMed Citation Articles by Morii, H. Articles by Koga, Y.

 Previous Article  |  Next Article 

Journal of Bacteriology, June 2007, p. 4053-4061, Vol. 189, No. 11
0021-9193/07/$08.00+0     doi:10.1128/JB.01875-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

In Vitro Biosynthesis of Ether-Type Glycolipids in the Methanoarchaeon Methanothermobacter thermautotrophicus

Hiroyuki Morii,^1* Tadashi Eguchi,² and Yosuke Koga¹

Department of Chemistry, School of Medicine, University of Occupational and Environmental Health, Yahatanishi-ku, Kitakyushu 807-8555,¹ Department of Chemistry and Materials Science, Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8551, Japan²

Received 13 December 2006/ Accepted 25 March 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The biosynthesis of archaeal ether-type glycolipids was investigated in vitro using Methanothermobacter thermautotrophicus cell-free homogenates. The sole sugar moiety of glycolipids and phosphoglycolipids of the organism is the ß-D-glucosyl-(16)-D-glucosyl (gentiobiosyl) unit. The enzyme activities of archaeol:UDP-glucose ß-glucosyltransferase (monoglucosylarchaeol [MGA] synthase) and MGA:UDP-glucose ß-1,6-glucosyltransferase (diglucosylarchaeol [DGA] synthase) were found in the methanoarchaeon. The synthesis of DGA is probably a two-step glucosylation: (i) archaeol + UDP-glucose MGA + UDP, and (ii) MGA + UDP-glucose DGA + UDP. Both enzymes required the addition of K⁺ ions and archaetidylinositol for their activities. DGA synthase was stimulated by 10 mM MgCl₂, in contrast to MGA synthase, which did not require Mg²⁺. It was likely that the activities of MGA synthesis and DGA synthesis were carried out by different proteins because of the Mg²⁺ requirement and their cellular localization. MGA synthase and DGA synthase can be distinguished in cell extracts greatly enriched for each activity by demonstrating the differing Mg²⁺ requirements of each enzyme. MGA synthase preferred a lipid substrate with the sn-2,3 stereostructure of the glycerol backbone on which two saturated isoprenoid chains are bound at the sn-2 and sn-3 positions. A lipid substrate with unsaturated isoprenoid chains or sn-1,2-dialkylglycerol configuration exhibited low activity. Tetraether-type caldarchaetidylinositol was also actively glucosylated by the homogenates to form monoglucosyl caldarchaetidylinositol and a small amount of diglucosyl caldarchaetidylinositol. The addition of Mg²⁺ increased the formation of diglucosyl caldarchaetidylinositol. This suggested that the same enzyme set synthesized the sole sugar moiety of diether-type glycolipids and tetraether-type phosphoglycolipids.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
More than 100 novel archaeal membrane lipid structures have been elucidated in the past 40 years (6, 14, 17, 18). These are unusual in the stereoconfiguration of the glycerol backbone, ether bonds, isoprenoid chains, and bipolar tetraether structures, which are one of the most remarkable features of archaea (17).

In vitro studies of the major pathway of polar lipid biosynthesis in archaea have been published over these past 15 years and extensively reviewed (16). Nishihara and Koga discovered sn-glycero-1-phosphate formation activity from dihydroxyacetone phosphate in a cell-free homogenate of Methanothermobacter thermautotrophicus (28). Poulter et al. (38, 39) have reported in vitro enzyme activities of first and second ether bond formation from sn-glycero-1-phosphate and geranylgeranylpyrophosphate in Methanothermobacter marburgensis. Concerning the attachment of polar head groups to digeranylgeranylglycerolphosphate (DGGGP), we showed the activities of CDP-archaeol synthase (25), by which DGGGP is activated by CTP to form CDP-archaeol, and archaetidylserine synthase, by which archaetidylserine (a diether analog of phosphatidylserine) is formed from CDP-archaeol and serine (23). The pathway was analogous to the bacterial pathway of phospholipid biosynthesis except for the differences in archaeon-specific lipid structures.

Other polar lipids in M. thermautotrophicus include two glycolipids and three phosphoglycolipids. The sole sugar moiety of glycolipids and phosphoglycolipids in M. thermautotrophicus is a gentiobiosyl group [ß-D-glucosyl-(16)-D-glucose]. Although glycolipid biosynthesis has been studied in vivo in Halobacterium cutirubrum (21), it has never been studied in vitro in archaea. In general, sugar moieties of glycolipids are simple in methanoarchaea, while glycolipid sugars of other groups of archaea are multifarious. Therefore, M. thermautotrophicus is considered to be a good source of enzymes for studies on glycolipid biosynthesis. Diglycosyl diacylglycerol is the dominant glycolipid (31) found in gram-positive bacteria and animal cells. Diacylglycerol (1,2-diacyl-sn-glycerol) is the initial lipid substrate and is sequentially glycosylated by nucleotide-linked sugar in almost all bacteria (31). As a first trial of an in vitro biosynthetic study of ether-type glycolipids, we used archaeol (diether analog of diacylglycerol) and UDP-glucose as the analogous substrates. The present paper reports that the synthesis of the glycolipids occurs in two steps: glucose is transferred from UDP-glucose to archaeol to form monoglucosylarchaeol (MGA), followed by a second transfer to form diglucosylarchaeol (DGA) (Fig. 1). The outline of the mechanism of glycolipid synthesis is analogous to the bacterial one, with certain differences between the archaeal and bacterial mechanisms.

View larger version (15K): [in this window] [in a new window]

FIG. 1. Possible pathways of ß-D-glucosyl-(16)-ß-D-glucosyl archaeol (DGA) from archaeol via ß-D-glucosyl archaeol (MGA).

In the course of study of archaeal glycolipid biosynthesis, one may encounter several questions concerning the biosynthesis of the specific archaeal lipid structures. One of the questions is which is the real substrate, unsaturated archaeol or saturated archaeol. The biosynthetic intermediates of archaeal phospholipids are ether compounds with unsaturated geranylgeranyl chains. The second question is whether this enzyme is specific to archaeal lipids with sn-2,3 stereoconfiguration of archaeol. The other problem is how to synthesize the tetraether-type glycolipid. The present work compares archaeal glycolipid biosynthesis to bacterial or eucaryal glycolipid synthesis and reports some properties of MGA synthase (archaeol:UDP-glucose ß-glucosyltransferase) and DGA synthase (MGA:UDP-glucose ß-1,6-glucosyltransferase) and the substrate specificity of MGA synthase in M. thermautotrophicus. The nomenclature of archaeal lipids proposed by Nishihara et al. (29) is used throughout this paper.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Growth of microorganisms. Cells of M. thermautotrophicus (DSM 1053) and Halobacterium salinarum (JCM 8981) were grown and harvested at log phase as previously described (25). Cells of M. thermautotrophicus were washed with buffer A or buffer B. Buffer A contained 50 mM Tris-HCl (pH 7.5), 1 mM Na₂-EDTA, and 10 mM 2-mercaptoethanol. Buffer B was 50 mM phosphate buffer (pH 6.5) containing 1 mM Na₂-EDTA and 10 mM 2-mercaptoethanol. The pelleted cell paste was stored at –20°C until use.

Preparation of lipid materials. Archaeol (2,3-di-O-phytanyl-sn-glycerol) and caldarchaeol were prepared from H. salinarum total lipid and M. thermautotrophicus total lipid, respectively, by HCl hydrolysis (concentrated aqueous HCl/methanol [1:5] incubated at 100°C for 3 h) followed by purification using preparative thin-layer chromatography (TLC) with a solvent system of light petroleum-diethyl ether-acetic acid (50:50:1). 1,2-Di-O-phytanyl-sn-glycerol was chemically synthesized from phytanol obtained from archaeol of H. salinarum (15) and 3-O-benzyl-sn-glycerol (Funakoshi, Tokyo, Japan) as described previously (25) except for the method for removal of the benzyl group in 1,2-di-O-phytanyl-3-O-benzyl-sn-glycerol at the last process. The benzyl group was removed by catalytic hydrogenolysis with 10% Pd-C (Nacalai, Kyoto, Japan) and a trace amount of HCl under an H₂ atmosphere at room temperature for 19 h. The identity of 1,2-di-O-phytanyl-sn-glycerol was confirmed by the molecular ion m/z 652 (M+H)⁺ revealed by positive fast atom bombardment-mass spectrometry (FAB-MS) and []_D = –7.1°, which confirmed the stereostructure (14). MGA was chemically synthesized from archaeol and glucose according to the method described in reference 20. Caldarchaetidylinositol was prepared by digestion of gentiobiosyl caldarchaetidylinositol from M. thermautotrophicus using a partially purified glycosidase mixture (22). Two milligrams of gentiobiosyl caldarchaetidylinositol was dispersed by sonication in acetate buffer (pH 4.5) containing 5 mM MgCl₂. The dispersion was incubated at 40°C for 2 days in the presence of 10 mg of the partially purified glycosidase mixture (Seikagaku Corp., Tokyo, Japan). During the reaction, another 10 mg of the enzyme was added 24 h after the incubation started. Archaetidylinositol, DGA, and diglucosyl caldarchaetidylinositol (DGCI) were purified from the total lipid of M. thermautotrophicus by TLC using solvent B (see below). Monoglucosylcaldarchaeol and 1,1'-diglucosylcaldarchaeol (caldarchaeol with a glucose unit on each hydroxyl group) were chemically synthesized from caldarchaeol in place of archaeol in the chemical synthesis reaction of MGA (see above). Negative FAB-MS of the products displayed m/z 1462 (M-H)^– and m/z 1624 (M-H)^–, respectively, which were consistent with the molecular weights of monoglucosylcaldarchaeol and 1,1'-diglucosylcaldarchaeol.

Enzyme preparation. Frozen cells (about 9 g [wet weight]) of M. thermautotrophicus were suspended in 10 ml of buffer A or buffer B (see above) containing 1 mg of DNase I (Sigma) and were passed through a French pressure cell operated at 1,400 kg/cm². This process was repeated three times. Cell debris and unbroken cells were removed by 10 min of centrifugation (10,000 x g). Each homogenate was centrifuged at 100,000 x g for 2 h to separate supernatant and membrane fractions. The membrane fraction was washed once and resuspended with the same buffer. The supernatant, membrane, and washing fractions prepared in buffer A were designated fraction AS, fraction AM, and fraction AW, respectively. The supernatant, membrane, and washing fractions prepared in buffer B were designated fraction BS, fraction BM, and fraction BW, respectively.

Enzyme assay. (i) MGA synthase. The complete assay mixture (final volume, 0.1 ml) contained the supernatant fraction of M. thermautotrophicus cell homogenate (fraction AS, 440 µg of protein) unless otherwise indicated, 0.1 M Bicine buffer (pH 8.0; Dojin Laboratories, Kumamoto, Japan), 0.5 M KCl, 20 nmol archaetidylinositol, 20 nmol archaeol, and 100 nmol UDP-[U-¹⁴C]glucose (18.5 Bq/nmol; Perkin-Elmer Life Sciences). Archaeol and archaetidylinositol were dispersed with the aqueous components of the reaction mixture, except for UDP-glucose and the enzyme preparation, in a 1.5-ml microtube by continuous sonication at 40°C for 30 min in a BRANSONIC 1210 (BRANSON) bath. After the addition of UDP-glucose and enzyme preparation, the reaction mixture was incubated in the 1.5-ml microtube at 60°C for 60 min. The reaction was stopped by adding 1 ml of 0.1 M HCl in methanol, and the reaction mixture was transferred to a 10-ml screw-cap glass tube with 1.5 ml of 0.1 M HCl in methanol and 2.5 ml CHCl₃. Finally, 2.15 ml of 1 M MgCl₂ (pH 2) was added to the mixture to partition the reactants and the products into aqueous and organic layers. After being washed twice with 0.1 M HCl/methanol-1 M MgCl₂ (pH 2) (1:0.8 [vol/vol]), the organic layer was evaporated to dryness and examined for radioactivity.

(ii) DGA synthase. The assay for DGA synthase was similar to that for MGA synthase except that the membrane fraction of the M. thermautotrophicus cell homogenate (fraction AM or fraction BM, 160 µg of protein) was usually used as an enzyme source unless specified otherwise, the lipid substrate was replaced by 40 nmol MGA, and 10 mM MgCl₂ was added.

TLC. TLC was carried out on a Silica Gel 60 plate (Merck) with the following solvents: solvent A, which consisted of chloroform, methanol, and 7 M ammonia (60:35:8); and solvent B, which consisted of chloroform, methanol, acetic acid, and water (80:30:15:5). Lipid spots were detected by spraying acid molybdate reagent (7) for phospholipids and subsequent charring for all lipids. Radioactive spots on a TLC plate were recorded with a Fujifilm FLA-2000 fluor image analyzer with an imaging plate (Fujifilm type BAS-MS).

Identification of reaction products. To confirm the structural identity of the enzyme reaction products from archaeol, MGA, or caldarchaeol, the reaction mixture was scaled up 20 to 400 times using nonradioactive UDP-glucose instead of UDP-[U-¹⁴C]glucose, and the incubation time was prolonged (3 h). Either the products (MGA and DGA) were purified by TLC with solvent A or monoglucosyl caldarchaetidylinositol (MGCI) and DGCI were purified with solvent B, and the products were analyzed by FAB-MS. Sugar moieties in MGA and DGA were analyzed by NMR spectrometry.

Analytical methods and physical measurements. Phosphate (2) and protein (34) were determined as described previously. A FAB-MS spectrum was recorded with a mass spectrometer (JEOL JMS DX-303) with a matrix of glycerol plus 15-crown-5 in a negative mode or with a matrix of glycerol in a positive mode. ¹H-NMR spectra (500 MHz) were recorded on a Bruker DRX-500 spectrometer in CDCl₃ for MGA and CDCl₃-CD₃OD (7:3) for DGA. The optical rotation was measured at room temperature at 589 nm with an automatic digital polarimeter (PM-201; Photal, Japan). Radioactivity was counted using a liquid scintillation spectrometer (LSC-3500E; Aloka, Japan) with Aquasol-2 (Packard) as the scintillator.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Activity of glucosylarchaeol synthesis. Because bacterial glyceroglycolipids are synthesized from diacylglycerol by sequential glycosylation with nucleotide-linked sugars, analogous substrates were tested for detection of in vitro glycolipid-synthesizing activity in M. thermautotrophicus cell homogenates. When archaeol was incubated with radioactive UDP-[U-¹⁴C]glucose in the presence of cell-free homogenates of M. thermautotrophicus under the conditions described in Materials and Methods, ¹⁴C was incorporated into chloroform-soluble materials. The incorporation was dependent on the incubation time and protein concentration of the added cell homogenates (0 to 660 µg) (data not shown) and continued for at least 60 min at 60°C (Fig. 2A). The addition of archaetidylinositol was essential for the activities (see below for details). The time courses of MGA and DGA synthesis from archaeol and of DGA synthesis from MGA are shown in Fig. 2A and C, respectively. Twenty to 45% of each lipid substrate was converted to the products during the reaction. To confirm that the chloroform-soluble products were really archaeol-derived glycolipids, they were analyzed by TLC with solvent A and FAB-MS. Two spots were detected by TLC after the reaction from archaeol (MGA synthase reaction) (Fig. 2B). One radioactive spot comigrating with standard MGA (R_f = 0.80) was the main product. The other minor spot (R_f = 0.50) coincided chromatographically with standard DGA. Only one spot was detected in the reaction from MGA (DGA synthase reaction) (Fig. 2D) and showed the same mobility in TLC as standard DGA. The lipid products enzymatically prepared from archaeol or MGA were purified by TLC with solvent A and confirmed by identification of their molecular ions. The FAB-MS of the major lipid product and the other minor lipid product from archaeol gave signals of m/z 814 (M-H)^– (Fig. 3A) and m/z 976 (M-H)^–, respectively, which are consistent with the molecular weights of MGA and DGA, respectively. FAB-MS of the product from MGA gave signals of m/z 976 (M-H)^– and m/z 814 (M-glucose)^– (Fig. 3B). The ¹H-NMR spectra of MGA and DGA clearly showed that all glucoside linkages had the ß configuration (MGA: 4.36, d, J = 8.0 Hz; DGA: 4.33, d, J = 7.8 Hz, 4.28, d, J = 7.8 Hz) (Fig. 4). Further, the ¹H-NMR spectrum of the enzymatically prepared DGA was identical to that of the synthetic ß-D-glucosyl-(16)-ß-D-glucosyl archaeol (1). These results show that M. thermautotrophicus cell homogenates contain activities that synthesize DGA via MGA from archaeol. These enzymes are MGA synthase (archaeol:UDP-glucose ß-glucosyltransferase) and DGA synthase (MGA:UDP-glucose ß-1,6-glucosyltransferase).

View larger version (22K): [in this window] [in a new window]

FIG. 2. Time courses of MGA and DGA synthesis from archaeol (A and B) and DGA synthesis from MGA (C and D). (A and B) Synthesis of glycolipids from 100 nmol UDP-[U-14C]glucose and 20 nmol archaeol catalyzed by the supernatant fraction of M. thermautotrophicus homogenate fraction AS incubated in the presence of 0.5 M KCl and 20 nmol archaetidylinositol in a final volume of 0.1 ml at 60°C, pH 8.0. (C and D) The enzyme assay condition was similar to that of the above assay except that the membrane fraction of M. thermautotrophicus homogenate fraction BM was used as an enzyme source, the lipid substrate was replaced by 40 nmol MGA, and 10 mM MgCl2 was added. (B and D) Autoradiograms of one-dimensional TLC of 14C products taken at indicated times (shown in minutes).

View larger version (5K): [in this window] [in a new window]

FIG. 3. Negative ion FAB mass spectra of MGA (A) and DGA (B) enzymatically prepared from archaeol and MGA, respectively.

View larger version (10K): [in this window] [in a new window]

FIG. 4. 1H-NMR spectra of MGA (A) and DGA (B) (500 MHz in CDCl3 for MGA and in CDCl3-CD3OD (7:3) for DGA).

Cellular localization of the activities. Because many polar lipid-synthesizing enzymes are associated with membranes, the cellular localization of the glycolipid-synthesizing enzymes was examined. Table 1 shows the distribution of the MGA synthase and DGA synthase activities between the particulate (membrane) and soluble fractions of the cell homogenates prepared at pH 7.5 and 6.5. The synthesis activities of the two glycolipids were found in both the soluble and membrane fractions. Almost all the MGA synthase activity was recovered in the soluble fraction when the homogenates were prepared at pH 7.5 (Table 1, fraction AS), while a greater proportion of DGA synthase was found to be associated with the membrane fraction (Table 1, fraction AM). The specific activities showed that MGA synthase was concentrated in the soluble fraction (Table 1, fraction AS), with DGA synthase in the membrane fraction (Table 1, fraction AM). The distribution of the activities between the two fractions (soluble and membrane) varied depending on the pH under which the cell homogenates had been prepared. When prepared at pH 6.5, more activities were recovered in the membrane fraction. The membrane fraction prepared at pH 6.5 (Table 1, fraction BM) contained the largest amount and highest specific activity of DGA synthase. Because fraction AS contained almost all of the MGA synthase activity, it was usually used as the source of MGA synthase (unless otherwise indicated). Fraction BM was usually used as the source of DGA synthase. Although these fractions contained other enzyme activity along with the major activity, the two enzyme activities could be distinguished by their different requirements for Mg²⁺ (see below).

View this table: [in this window] [in a new window]

TABLE 1. Enzyme activity of each cell fractiona

Properties of MGA synthase and DGA synthase. Both MGA synthase and DGA synthase were stimulated by the addition of K⁺ ions, displaying maximum activities at a concentration of 0.5 M (Fig. 5). The optimal pH for both reactions was 8.0 (Fig. 6). The addition of archaetidylinositol is essential for the activities of both MGA and DGA synthase (Table 2).

View larger version (12K): [in this window] [in a new window]

FIG. 5. Effects of K+ concentration on MGA synthase activity and DGA synthase activity.

View larger version (10K): [in this window] [in a new window]

FIG. 6. Effects of pH on MGA synthase activity (A) and DGA synthase activity (B).

View this table: [in this window] [in a new window]

TABLE 2. Effects of archaetidylinositol on MGA synthase activity and DGA synthase activity

The two activities were different in terms of the effect of Mg²⁺. DGA synthase required 10 mM Mg²⁺ for maximum activity (Fig. 7A), while MGA synthase did not require the addition of Mg²⁺ ions. Figure 7B and C showed the effect of Mg²⁺ concentration on the proportion of the products (MGA and DGA) of the MGA synthase reaction using archaeol as a substrate. When no Mg²⁺ was added to the reaction mixture containing either fraction AS or fraction BM as an enzyme source, almost all products were MGA, and only a trace amount of DGA was formed. This indicates that the effect of Mg on the MGA synthase activity was not due to the state of the enzymes (whether these were membrane-associated or not). The pattern of total glucose incorporation into lipid varied depending on the enzyme source. In the case of the MGA synthase assay of fraction AS from archaeaol as a substrate, less glucose was incorporated into lipid as the Mg²⁺ ion concentration increased (Fig. 7B), while more glucose was incorporated when fraction BM was used as an enzyme source for the same assay (Fig. 7C). This difference is mainly derived from the difference in the proportion of activity of MGA synthase and DGA synthase in the fractions. In fraction AS, DGA synthase has one-third of the activity of MGA synthase (Table 1). On the other hand, DGA synthase has an activity two times higher than that of MGA synthase in fraction BM (Table 1). The difference in the Mg²⁺ ion requirement for the activities discriminates MGA synthase and DGA synthase. That is, it is concluded that these two activities are effected by two different enzymes. MGA synthase activity in the homogenate containing some DGA synthase can be expected to be assayed without interference with DGA synthase activity in the absence of the Mg²⁺ ion. In fact, when archaeol and UDP-[U-¹⁴C]glucose were reacted in the absence of Mg²⁺ with fraction AS, which is the condition of the MGA synthase assay, 94% of the incorporated [¹⁴C]glucose appeared in MGA and 6% appeared in DGA. Considering that DGA is synthesized via MGA from archaeol and that DGA has two glucose units, this corresponds to a product proportion of 97 mol% MGA and 3 mol% DGA. At 10 mM Mg²⁺, the proportion of DGA formed from archaeol increased to approximately 25 mol% of the total products (Fig. 7B).

View larger version (11K): [in this window] [in a new window]

FIG. 7. Effects of Mg2+ concentration on DGA synthase activity (A) and (B and C) MGA synthase assay by use of the supernatant fraction AS (B) and the membrane fraction BM (C) of M. thermautotrophicus homogenate as an enzyme source. Glc, glucose.

A fairly small amount of DGA was synthesized without the addition of Mg²⁺ (Fig. 2A and B). This is determined according to the presence of a small amount of Mg²⁺ in the cell homogenates. Ca²⁺ was also effective, but less so than Mg²⁺, on the activity of DGA synthase. The addition of 5 or 10 mM Ca²⁺ showed 20% of the Mg²⁺ effect on DGA synthase activity. Addition of 1 mM EDTA to the assay mixture completely abolished DGA synthesis from MGA.

Mode of reaction of DGA synthase. Chloroplasts of plant cells have galactolipid:galactolipid galactosyltransferase (EC 2.4.1.184), which catalyzes an interlipid exchange of galactose between two molecules of monogalactosyldiacylglycerol and leads to the formation of digalactosyldiacylglycerol and diacylglycerol (8). This kind of activity was examined as follows. [¹⁴C]MGA enzymatically prepared by the MGA synthase reaction was incubated with cell homogenates in the presence or absence of nonradioactive UDP-glucose. [¹⁴C]DGA was synthesized only in the presence of UDP-glucose (Fig. 8). This result indicates that DGA synthase is MGA:UDP-glucose glucosyltransferase and not glucosylarchaeol:glucosylarchaeol glucosyltransferase.

View larger version (54K): [in this window] [in a new window]

FIG. 8. Requirement of UDP-glucose for DGA synthase activity. DGA synthase reactions were carried out with [14C]MGA in the presence (left lane) or absence (right lane) of nonradioactive UDP-glucose. The products were analyzed by one-dimensional TLC. An autoradiogram of the TLC was shown.

Substrate specificity of MGA synthase. Archaeal glycolipids and phospholipids are primarily characterized by the stereoconfiguration of the glycerol backbone on which two hydrocarbon chains are bound at the sn-2 and -3 positions (sn-2,3 stereostructure). Intermediates of the phospholipid biosynthesis pathway in archaea have unsaturated hydrocarbon chains, which are then saturated to the final form in the membrane. The structural requirements for the substrate of the enzymes involved in glycolipid biosynthesis are, therefore, of special interest. We examined the activities of MGA synthase towards a variety of chemically synthesized substrate analogs, including archaeol analogs with sn-1,2 stereostructure or unsaturated hydrocarbons, and tetraether-type lipids (Table 3). Ordinary saturated archaeol was the most active substrate for MGA synthase. The stereostructure of the glycerol backbone was critical to MGA synthase activity; the synthetic enantiomeric saturated archaeol with an sn-1,2 stereostructure showed quite low activity compared with saturated archaeol with an sn-2,3 stereostructure. Unsaturated archaeol with a natural sn-2,3 stereostructure exhibited low activity (29%) compared with saturated natural archaeol. It is concluded that MGA synthase recognizes the stereostructure of the glycerol backbone and saturation of the hydrocarbon chains.

View this table: [in this window] [in a new window]

TABLE 3. Substrate specificity of MGA synthasea

Tetraether lipids (caldarchaetidylinositol and caldarchaeol) exhibited activities similar to that of the saturated natural diether lipid when incubated with fraction AS in the absence of Mg²⁺ (Table 3). Even when archaetidylinositol was omitted from the reaction mixture, caldarchaetidylinositol exhibited a similar activity as in the presence of archaetidylinositol because caldarchaetidylinositol has the same polar group as archaetidylinositol (phosphoinositol). The chloroform-soluble ¹⁴C-labeled products from caldarcharchaetidylinositol displayed two spots on TLC, with solvent B comigrating with authentic MGCI and DGCI (R_f values of 0.27 and 0.09, respectively). Table 4 shows the proportion of the products from caldarchaetidylinositol as a substrate with the soluble or membrane fraction as an enzyme in the presence or absence of archaetidylinositol and Mg²⁺. The addition of Mg²⁺ stimulated the synthesis of diglucosyl caldarchaetidylinositol, as seen in the case of DGA synthesis from archaeol (Fig. 7B and C). Since caldarchaeol has two hydroxyl groups per molecule, similar activity was exhibited even if the concentration of the lipid substrate was lowered to half of the standard assay condition (0.1 mM) (Table 3). The chloroform-soluble ¹⁴C-labeled products from caldarchaeol displayed four spots on TLC with solvent A. The two main spots comigrated with chemically synthesized monoglucosyl caldarchaeol (R_f = 0.87) and 1,1'-diglucosyl caldarchaeol (caldarchaeol with a glucose unit on each hydroxyl group) (R_f = 0.68). The other two products may be triglucosylcaldarchaeol with two glucose moieties on one side and one glucose moiety on the other side and tetraglucosylcaldarchaeol with two glucose moieties on both sides, although these were not experimentally verified. The ratio of monoglucosyl caldarchaeol, 1,1'-diglucosyl caldarchaeol, and the other products was 55:39:6.

View this table: [in this window] [in a new window]

TABLE 4. Product composition for reactions of caldarchaetidylinositol and UDP-glucose with fraction AS or BM under conditions similar to those of MGA synthase reaction with or without Mg2+

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The activities of MGA synthase and DGA synthase were found in M. thermautotrophicus homogenates. The two enzymes catalyzed transfer of a glucose unit from UDP-glucose to archaeol and MGA to form MGA and DGA, respectively. This is the first report on the in vitro biosynthesis of glycolipids in archaea. It is likely that the two activities are carried out by two separate enzymes based on the following observations. (i) Although the two enzymes were loosely associated with the membranes, they are easily released from the membranes during cell disruption in buffer solutions at different rates. DGA synthase was less easily released from the membrane than MGA synthase. (ii) DGA synthase required Mg²⁺ for the activity but MGA synthase did not. In Bacillus subtilis, one enzyme catalyzes these two steps of glycosylation in contrast to the archaeal enzymes (11). We used fraction AS as a source of MGA synthase, which contained 93% of the total activity of MGA synthase with the highest specific activity among the homogenate fractions. For DGA synthase we used fraction BM, which contained 52% of the total activity of DGA synthase with the highest specific activity among the homogenate fractions. Although these fractions represent each of the particular enzyme activities, each fraction still contained the other enzyme. However, the activities of MGA synthase and DGA synthase can almost be determined separately with or without the addition of Mg²⁺, even though the enzymes were not completely separated from each other.

A high concentration (0.5 M) of K⁺ ions was also necessary for both enzyme activities. This corresponds with the intracellular concentration of the K⁺ ion (0.62 to 0.78 M) in M. thermautotrophicus (35). The K⁺ ion dependence of the polar lipid biosynthetic enzyme in this methanoarchaeon has been also observed in CDP-archaeol synthase (25). Although the intracellular pH (6.7) (10) is not optimal for activity, both enzymes are still active at pH 6.7. An anionic phospholipid (archaetidylinositol) was essential for both enzyme activities. In the case of bacterial glycolipid synthases, the monoglucosyl diacylglycerol (MGDG) synthase of Acholeplasma laidlawii is absolutely dependent upon the presence of anionic phospholipids such as phosphatidylglycerol (5, 13). This seems to constitute a rate-keeping connection between the archaetidylinositol and glucolipid pathways in vivo, which yields the constant surface charge density of the lipid bilayer membrane as discussed by Christiansson et al. (4, 37). Although the outline of the mechanism of archaeal glycolipid synthesis is the same as the bacterial one, there are some differences in enzyme properties among species. Both enzymes of Acholeplasma require detergent and Mg²⁺ ions, while both enzymes of Methanothermobacter did not require any detergent and the first enzyme (MGA synthase) did not require Mg²⁺. MGA synthase was inhibited by 0.2 or 1% Triton X-100. Because, according to the data of Sprott and Jarrell (35), the intracellular Mg²⁺ concentration in M. thermautotrophicus increases from 1.4 mM (mid-log phase) to 12 mM (stationary phase) as the cultures aged, it is presumed that DGA synthase activity may vary several times during its growth. In fact, gentiobiose as a polar head group of polar lipid of this methanoarchaeon increased from 43% to 50% between the log and stationary phases of growth, while other polar groups of its phospholipids decreased during the same period (24).

2,3-Di-O-geranylgeranyl-sn-glycerol-1-phosphate (unsaturated archaetidic acid, DGGGP) is an important intermediate in archaeal phospholipid biosynthesis (25). Although it is not yet known how saturated archaeol, which is the most active substrate of MGA synthase, is formed in vivo, it is possibly made via this unsaturated intermediate (DGGGP). In order to determine the real substrate for glycolipid biosynthesis, we investigated the substrate specificity. MGA synthase preferred fully saturated archaeol (2,3-di-O-phytanyl-sn-glycerol) to fully unsaturated archaeol (2,3-di-O-geranylgeranyl-sn-glycerol). On the other hand, archaetidylserine synthase of M. thermautotrophicus exhibited similar activities when CDP-unsaturated archaeol and CDP-saturated archaeol were used as substrates (23). Nishimura and Eguchi have reported that 2,3-di-O-geranylgeranyl-sn-glycerophospholipid reductase is detected in cell extracts of Thermoplasma acidophilum (30). Similar geranylgeranyl reductase was also reported in Archaeoglobus fulgidus by Murakami et al. (26). These enzymes exhibit broad substrate specificity; hydrogenation of unsaturated phospholipids can occur before or after the attachment of polar head groups. In the psychrotrophic archaeon Halorubrum lacusprofudi, unsaturated lipids were detected in trace amounts in cells grown at 20°C but at 12°C accounted for 70% of the phospholipids and 10% of the glycolipid (9), possibly as a result of incomplete saturation of unsaturated precursors. These in vivo results, in combination with our in vitro results, suggest that while phospholipids are hydrogenated after attachment of polar head groups, the reduction step (hydrogenation) of the unsaturated intermediate may occur before the attachment of the glucose moiety in glycolipid (MGA) biosynthesis. It is worth noting that the hydrogenation step in glycolipid synthesis is different from that in phospholipid synthesis. The real substrate of MGA synthase in this methanoarchaeon is probably saturated archaeol or archaeol with few double bonds in hydrocarbon chains.

Tetraether lipids (caldarchaetidylinositol and caldarchaeol) were also actively glucosylated by the homogenate of M. thermautotrophicus. The products of glucosylation of caldarchaetidylinositol were MGCI and diglucosyl caldarchaetidylinositol. The addition of Mg²⁺ ions to the reaction mixture increased the proportion of diglucosyl caldarchaetidylinositol, which was similar to the effect of the Mg²⁺ ions in the glucosylation of archaeol. The tetraether-type phosphoglycolipid was synthesized as shown in Fig. 9. This finding is consistent with the conclusion from in vivo experiments in Thermoplasma acidophilum using terbinafine, an inhibitor of tetraether lipid biosynthesis (27). It is suggested that sugar residues are attached to the tetraether lipid core after the condensation of two molecules of diether-type phospholipid and removal of one phosphate-containing polar head group. The fact that diether-type phospholipid but not diether-type glycolipid accumulated during terbinafine inhibition (27) can be explained by the difference in the synthetic mechanisms of tetraether-type phospholipids and glycolipids.

View larger version (18K): [in this window] [in a new window]

FIG. 9. Possible pathways of DGCI synthesis from caldarchaetidylinositol. G, glucose.

The mode of glucose transfer reaction of MGA synthase is similar to that of 1,2-diacylglycerol 3-glucosyltransferase (MGDG synthase) from A. laidlawii. That is, glucose is transferred from UDP-glucose to diradylglycerol core lipid. The requirement of anionic phospholipid is also similar. A different enzyme catalyzes transfer of the second glucose moiety. In spite of these similar biochemical properties, the genes encoding glycolipid synthesizing enzymes have not been identified in the M. thermautotrophicus genome (33). The sequence properties of MGDG synthase from A. laidlawii are well characterized. Similar amino acid sequences are encoded in Pyrococcus horikoshii (3). Although many glycosyltransferases were detected by a BLAST search on the amino acid sequence of MGDG synthase of A. laidlawii, no enzyme with sequence similarity in M. thermautotrophicus was detected. One of the explanations of this situation may be the difference in substrate specificities of the enzymes (see below).

MGA synthase showed a relatively high activity with the sn-2,3 stereoconfiguration of the glycerol backbone of the lipid substrate. In the case of bacterial glycolipid synthases, mannosyltransferase in Micrococcus lysodeikticus (19) and glucosyltransferase in Streptococcus faecalis (32) display a relatively high specificity for the sn-1,2 stereoconfiguration of diacylglycerol substrate. In A. laidlawii MGDG synthase activity with a racemic mixture of diacylglycerol was approximately half of the activity gained with sn-1,2 diacylglycerol (12). Assuming that glycolipids and glycolipid-synthesizing enzymes were present in a precell stage with racemic diradylglycerol core lipids (36), the ancestors of MGA synthase and MGDG synthase developed in this stage, i.e., when the common mode of sugar transfer reaction was established. The substrate specificities of both enzymes, particularly the stereospecificities, were developed after the differentiation of Archaea and Bacteria, which might cause significant sequence diversity. In contrast to glycolipid-synthesizing enzymes, the archaetidylserine synthase of M. thermautotrophicus exhibits a loose substrate specificity (it does not recognize ether/ester linkages, saturated/unsaturated hydrocarbons, isoprenoid/straight chain hydrocarbons, or streoconfiguration of the glycerol backbone). We have speculated that the gene encoding phosphatidylserine synthase was transferred from a gram-positive bacterium to a methanoarchaeon after differentiation from other Archaea groups (23).

 
We are grateful to Mami Ohga for the culturing of microorganisms.

This work was partly supported by JSPS KAKENHI 16580067.

 
* Corresponding author. Mailing address: Department of Chemistry, School of Medicine, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807-8555, Japan. Phone: 81-93-603-1611. Fax: 81-93-693-9921. E-mail: h-morii{at}health.uoeh-u.ac.jp

Published ahead of print on 6 April 2007.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Arakawa, K., T. Eguchi, and K. Kakinuma. 2001. 36-Membered macrocyclic diether lipid is advantageous for archaea to thrive under the extreme thermal environments. Bull. Chem. Soc. Jpn. 74:347-356.[CrossRef]
2
2. Bartlett, G. R. 1959. Phosphorus assay in column chromatography. J. Biol. Chem. 234:466-468.[Free Full Text]
3
3. Berg, S., M. Edman, L. Li, M. Wikstrom, and A. Wieslander. 2001. Sequence properties of the 1,2-diacylglycerol 3-glucosyltransferase from Acholeplasma laidlawii membranes: recognition of a large group of lipid glycosyltransferases in eubacteria and archaea. J. Biol. Chem. 276:22056-22063.[Abstract/Free Full Text]
4
4. Christiansson, A., L. E. G. Eriksson, J. Westman, R. Demel, and A. Wieslander. 1985. Involvement of surface potential in regulation of polar membrane lipids in Acholeplasma laidlawii. J. Biol. Chem. 260:3984-3990.[Abstract/Free Full Text]
5
5. Dahlqvist, A., S. Nordstrom, O. P. Karlsson, D. A. Mannock, R. N. McElhaney, and A. Wieslander. 1995. Efficient modulation of glucolipid enzyme activities in membranes of Acholeplasma laidlawii by the type of lipids in the bilayer matrix. Biochemistry 34:13381-13389.[CrossRef][Medline]
6
6. De Rosa, M., A. Gambacorta, and A. Gliozzi. 1986. Structure, biosynthesis, and physicochemical properties of archaebacterial lipids. Microbiol. Rev. 50:70-80.[Free Full Text]
7
7. Dittmer, J. C., and R. L. Lester. 1964. A simple, specific spray for the detection of phospholipids on thin-layer chromatograms. J. Lipid Res. 5:126-127.
8
8. Dorne, A. J., M. A. Block, J. Joyard, and R. Douce. 1982. The galactolipid:galactolipid galactosyltransferase is located on the outer surface of the outer membrane of the chloroplast envelope. FEBS Lett. 145:30-34.[CrossRef]
9
9. Gibson, J. A. E., M. R. Miller, N. W. Davies, G. P. Neill, D. S. Nichols, and J. K. Volkman. 2005. Unsaturated diether lipids in the psychrotrophic archaeon Halorubrum lacusprofundi. Syst. Appl. Microbiol. 28:19-26.[CrossRef][Medline]
10
10. Jarrell, K. F., and G. D. Sprott. 1981. The transmembrane electrical potential and intracellular pH in methanogenic bacteria. Can. J. Microbiol. 27:720-728.[Medline]
11
11. Jorasch, P., F. P. Wolter, U. Zahringer, and E. Heinz. 1998. A UDP glucosyltransferase from Bacillus subtilis successively transfers up to four glucose residues to 1,2-diacylglycerol: expression of ypfP in Escherichia coli and structural analysis of its reaction products. Mol. Microbiol. 29:419-430.[CrossRef][Medline]
12
12. Karlsson, O. P., A. Dahlqvist, S. Vikstroem, and A. Wieslander. 1997. Lipid dependence and basic kinetics of the purified 1,2-diacylglycerol 3-glucosyltransferase from membranes of Acholeplasma laidlawii. J. Biol. Chem. 272:929-936.[Abstract/Free Full Text]
13
13. Karlsson, O. P., A. Dahlqvist, and A. Wieslander. 1994. Activation of the membrane glucolipid synthesis in Acholeplasma laidlawii by phosphatidylglycerol and other anionic lipids. J. Biol. Chem. 269:23484-23490.[Abstract/Free Full Text]
14
14. Kates, M. 1978. The phytanyl ether-linked polar lipids and isoprenoid neutral lipids of extremely halophilic bacteria. Prog. Chem. Fats Other Lipids 15:301-342.[CrossRef][Medline]
15
15. Kates, M., B. Palameta, and L. S. Yengoyan. 1965. Aliphatic diether analogs of glyceride-derived lipids. II. Synthesis of naturally occurring L-2,3,-di-O-3',7',11',15'-tetramethylhexahexadecyl glycerol and its D isomer. Biochemistry 4:1595-1598.[CrossRef][Medline]
16
16. Koga, Y., and H. Morii. 2007. Biosynthesis of ether-type polar lipids in Archaea and evolutionary considerations. Microbiol. Mol. Biol. Rev. 71:97-120.[Abstract/Free Full Text]
17
17. Koga, Y., and H. Morii. 2005. Recent advances in structural research on ether lipids from Archaea including comparative and physiological aspects. Biosci. Biotechnol. Biochem. 69:2019-2034.[CrossRef][Medline]
18
18. Koga, Y., M. Nishihara, H. Morii, and M. Akagawa-Matsushita. 1993. Ether polar lipids of methanogenic bacteria: structures, comparative aspects, and biosynthesis. Microbiol. Rev. 57:164-182.[Abstract/Free Full Text]
19
19. Lennarz, W. J., and B. Talamo. 1966. The chemical characterization and enzymatic synthesis of mannolipids in Micrococcus lysodeikticus. J. Biol. Chem. 241:2707-2719.[Abstract/Free Full Text]
20
20. Minamikawa, H., T. Murakami, and M. Hato. 1994. Synthesis of 1,3-di-O-alkyl-2-O-(ß-glycosyl)glycerols bearing oligosaccharides as hydrophilic groups. Chem. Phys. Lipids 72:111-118.[CrossRef][Medline]
21
21. Moldoveanu, N., and M. Kates. 1988. Biosynthetic studies of the polar lipids of Halobacterium cutirubrum. Formation of isoprenyl ether intermediate. Biochim. Biophys. Acta 960:164-182.
22
22. Morii, H., and Y. Koga. 1994. Asymmetric topology of diether and tetraether type polar lipids in the membrane of Methanobacterium thermoautotrophicum cells. J. Biol. Chem. 269:10492-10497.[Abstract/Free Full Text]
23
23. Morii, H., and Y. Koga. 2003. CDP-2,3-di-O-geranylgeranyl-sn-glycerol:L-serine O-archaetidyltransferase (archaetidylserine synthase) in the methanogenic archaeon Methanothermobacter thermautotrophicus. J. Bacteriol. 185:1181-1189.[Abstract/Free Full Text]
24
24. Morii, H., and Y. Koga. 1993. Tetraether type polar lipids increase after logarithmic growth phase of Methanobacterium thermoautotrophicum in compensation for the decrease of diether lipids. FEMS Microbiol. Lett. 109:283-288.[CrossRef]
25
25. Morii, H., M. Nishihara, and Y. Koga. 2000. CTP:2,3-di-O-geranylgeranyl-sn-glycero-1-phosphate cytidyltransferase in the methanogenic archaeon Methanothermobacter thermoautotrophicus. J. Biol. Chem. 275:36568-36574.[Abstract/Free Full Text]
26
26. Murakami, M., K. Shibuya, T. Nakayama, T. Nishino, T. Yoshimura, and H. Hemmi. 2007. Geranylgeranyl reductase involved in the biosynthesis of archaeal membrane lipids in the hyperthermophilic archaeol Archaeoglobus fulgidus. FEBS J. 274:805-814.[CrossRef][Medline]
27
27. Nemoto, N., Y. Shida, H. Shimada, T. Oshima, and A. Yamagishi. 2003. Characterization of the precursor of tetraether lipid biosynthesis in the thermoacidophilic archaeon Thermoplasma acidophilum. Extremophiles 7:235-243.[Medline]
28
28. Nishihara, M., and Y. Koga. 1995. sn-Glycerol-1-phosphate dehydrogenase in Methanobacterium thermoautotrophicum: key enzyme in biosynthesis of the enantiomeric glycerophosphate backbone of ether phospholipids of archaebacteria. J. Biochem. 117:933-935.[Abstract/Free Full Text]
29
29. Nishihara, M., H. Morii, and Y. Koga. 1987. Structure determination of a quartet of novel tetraether lipids from Methanobacterium thermoautotrophicum. J. Biochem. 101:1007-1015.[Abstract/Free Full Text]
30
30. Nishimura, Y., and T. Eguchi. 2006. Biosynthesis of archaeal membrane lipids: digeranylgeranylglycerophospholipid reductase of the thermoacidophilic archaeon Thermoplasma acidophilum. J. Biochem. 139:1073-1081.[Abstract/Free Full Text]
31
31. Pieringer, R. A. 1983. Formation of bacterial glycerolipids, p. 255-306. In P. D. Boyer (ed.) The enzymes, vol. 16. Academic Press, New York, NY.
32
32. Pieringer, R. A. 1968. The metabolism of glyceride glycolipids. J. Biol. Chem. 243:4894-4903.[Abstract/Free Full Text]
33
33. Reeve, J. N., J. Nolling, R. M. Morgan, and D. R. Smith. 1997. Methanogenesis: genes, genomes, and who's on first? J. Bacteriol. 179:5975-5986.[Free Full Text]
34
34. Smith, P. K., R. I. Krohn, G. T. Hermanson, A. K. Mallia, F. H. Gartner, M. D. Provenzaro, E. K. Fujimoto, N. M. Goeke, B. J. Olson, and D. C. Klenk. 1985. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150:76-85.[CrossRef][Medline]
35
35. Sprott, G. D., and K. F. Jarrell. 1981. K⁺, Na⁺, and Mg²⁺ content and permeability of Methanospirillum hungatei and Methanobacterium thermoautotrophicum. Can. J. Microbiol. 27:444-451.[Medline]
36
36. Wächtershäuser, G. 2003. From pre-cells to Eukarya—a tale of two lipids. Mol. Microbiol. 47:13-22.[CrossRef][Medline]
37
37. Wieslander, A., S. Nordstrom, A. Dahlqvist, L. Rilfors, and G. Lindblom. 1995. Membrane lipid composition and cell size of Acholeplasma laidlawii strain A are strongly influenced by lipid acyl chain length. Eur. J. Biochem. 227:734-744.[Medline]
38
38. Zhang, D.-L., L. Daniels, and C. D. Poulter. 1990. Biosynthesis of archaebacterial membranes. Formation of isoprenyl ethers by a prenyl transfer reaction. J. Am. Chem. Soc. 112:1264-1265.[CrossRef]
39
39. Zhang, D. L., and C. D. Poulter. 1993. Biosynthesis of archaebacterial ether lipids. Formation of ether linkages by prenyltransferases. J. Am. Chem. Soc. 115:1270-1277.[CrossRef]

---

Journal of Bacteriology, June 2007, p. 4053-4061, Vol. 189, No. 11
0021-9193/07/$08.00+0     doi:10.1128/JB.01875-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Morii, H. Articles by Koga, Y. Search for Related Content PubMed PubMed Citation Articles by Morii, H. Articles by Koga, Y.