Structural studies of a phosphocholine substituted beta-(1,3);(1,6) macrocyclic glucan from Bradyrhizobium japonicum USDA 110.

In our previous in vivo 31P study of intact nitrogen-fixing nodules (Rolin, D.B., Boswell, R.T., Sloger, C., Tu, S.I. and Pfeffer, P.E., 1989 Plant Physiol. 89, 1238-1246), we observed an unknown phosphodiester. The compound was also observed in the spectra of isolated bacteroids as well as extracts of the colonizing Bradyrhizobium japonicum USDA 110. In order to characterize the phosphodiester in the present study, we took advantage of the relatively hydrophobic nature of the material and purified it by elution from a C-18 silica reverse-phase chromatography column followed by final separation on an aminopropyl silica HPLC column. Structural characterization of this compound with a molecular weight of 2271 (FAB mass spectrometry), using 13C-1H and 31P-1H heteronuclear 2D COSY and double quantum 2D phase sensitive homonuclear 1H COSY NMR spectra, demonstrated that the molecule contained beta-(1,3); beta-(1,6); beta-(1,3,6) and beta-linked non-reducing terminal glucose units in the ratio of 5:6:1:1, respectively, as well as one C-6 substituted phosphocholine (PC) moiety associated with one group of (1,3) beta-glucose residues. Carbohydrate degradation analysis indicated that this material was a macrocyclic glucan, (absence of a reducing end group) with two separated units containing three consecutively linked beta-(1,3) glucose residues and 6 beta-(1,6) glucose residues. The sequences of beta-(1,3)-linked glucose units contained a single non-reducing, terminal, unsubstituted glucose linked at the C-6 position and a PC group attached primarily to an unsubstituted C-6 position of a beta-(1,3)-linked glucose.

Multiple-rotor-cycle 2D PASS experiments with applications to (207)Pb NMR spectroscopy.

Thetwo-dimensional phase-adjusted spinning sidebands (2D PASS) experiment is a useful technique for simplifying magic-angle spinning (MAS) NMR spectra that contain overlapping or complicated spinning sideband manifolds. The pulse sequence separates spinning sidebands by their order in a two-dimensional experiment. The result is an isotropic/anisotropic correlation experiment, in which a sheared projection of the 2D spectrum effectively yields an isotropic spectrum with no sidebands. The original 2D PASS experiment works best at lower MAS speeds (1-5 kHz). At higher spinning speeds (8-12 kHz) the experiment requires higher RF power levels so that the pulses do not overlap. In the case of nuclei such as (207)Pb, a large chemical shift anisotropy often yields too many spinning sidebands to be handled by a reasonable 2D PASS experiment unless higher spinning speeds are used. Performing the experiment at these speeds requires fewer 2D rows and a correspondingly shorter experimental time. Therefore, we have implemented PASS pulse sequences that occupy multiple MAS rotor cycles, thereby avoiding pulse overlap. These multiple-rotor-cycle 2D PASS sequences are intended for use in high-speed MAS situations such as those required by (207)Pb. A version of the multiple-rotor-cycle 2D PASS sequence that uses composite pulses to suppress spectral artifacts is also presented. These sequences are demonstrated on (207)Pb test samples, including lead zirconate, a perovskite-phase compound that is representative of a large class of interesting materials.

Phosphoglycerol substituents present on the cyclic beta-1,2-glucans of Rhizobium meliloti 1021 are derived from phosphatidylglycerol.

The synthesis of periplasmic cyclic beta-1,2-glucans is a property unique to species of the family Rhizobiaceae. For this reason, it is generally believed that these molecules may play an important role in the plant infection process. In the present study, we determined that the cyclic beta-1,2-glucans produced by Rhizobium meliloti 1021 were predominantly anionic in character and contained both phosphoglycerol and succinic acid substituents. In addition, we demonstrated that phosphatidylglycerol was the source of the phosphoglycerol substituents present on these oligosaccharides and that greater than 60% of the total phospholipid turnover in this organism involved this substitution reaction.

Effect of Phosphate Limitation on Synthesis of Periplasmic Cyclic (beta)-(1,2)-Glucans.

Rhizobium meliloti and Agrobacterium tumefaciens synthesize periplasmic cyclic (beta)-(1,2)-glucans during adaptation to hypoosmotic environments. It also appears that these glucans provide important functions during the interactions of these bacteria with plant hosts. A large fraction of these glucans may become modified with anionic substituents such as phosphoglycerol or succinic acid; however, the role(s) of these substituents is unknown. In this study, we show that growth of these bacteria in phosphate-limited media leads to a dramatic reduction in the levels of phosphoglycerol substituents present on the periplasmic cyclic (beta)-(1,2)-glucans. Under these growth conditions, R. meliloti 1021 was found to synthesize anionic cyclic (beta)-(1,2)-glucans containing only succinic acid substituents. Similar results were obtained with R. meliloti 7154 (an exoH mutant which lacks the ability to succinylate its high-molecular-weight exopolysaccharide), revealing that succinylation of the cyclic (beta)-(1,2)-glucans is mediated by an enzyme system distinct from that involved in the succinylation of exopolysaccharide. In contrast, when A. tumefaciens C58 was grown in a phosphate-limited medium, it was found to synthesize only neutral cyclic (beta)-(1,2)-glucans.

Cell-associated oligosaccharides of Bradyrhizobium spp.

We report the initial characterization of the cell-associated oligosaccharides produced by four Bradyrhizobium strains: Bradyrhizobium japonicum USDA 110, USDA 94, and ATCC 10324 and Bradyrhizobium sp. strain 32H1. The cell-associated oligosaccharides of these strains were found to be composed solely of glucose and were predominantly smaller than the cyclic beta-1,2-glucans produced by Agrobacterium and Rhizobium species. Linkage studies and nuclear magnetic resonance analyses demonstrated that the bradyrhizobial glucans are linked primarily by beta-1,6 and beta-1,3 glycosidic bonds. Thus, the bradyrhizobia appear to synthesize cell-associated oligosaccharides of structural character substantially different from that of the cyclic beta-1,2-glucans produced by Agrobacterium and Rhizobium species.

Observation by 13C NMR of the EPSP synthase tetrahedral intermediate bound to the enzyme active site.

Direct observation of the tetrahedral intermediate in the EPSP synthase reaction pathway was provided by 13C NMR by examining the species bound to the enzyme active site under internal equilibrium conditions and using [2-13C]PEP as a spectroscopic probe. The tetrahedral center of the intermediate bound to the enzyme gave a unique signal appearing at 104 ppm. Separate signals were observed for free EPSP (152 ppm) and EPSP bound to the enzyme in a ternary complex with phosphate (161 ppm). These peak assignments account for our quantitation of the species bound to the enzyme and liberated upon quenching with either triethylamine or base. A comparison of quenching with acid, base, or triethylamine was conducted; the intermediate could be isolated by quenching with either triethylamine or 0.2 N KOH, allowing direct quantitation of the species bound to the enzyme. After long times of incubation during the NMR measurement, a signal at 107 ppm appeared. The compound giving rise to this resonance was isolated and identified as an EPSP ketal [Leo et al. (1990) J. Am. Chem. Soc. (in press)]. The rate of formation of the EPSP ketal was very slow, 3.3 X 10(-5) s-1, establishing that it is a side product of the normal enzymatic reaction, probably arising as a breakdown product of the tetrahedral intermediate. A slow formation of pyruvate was also observed and is attributable to the enzymatic hydrolysis of EPSP, with 5% of the enzyme sites occupied by EPSP and hydrolyzing EPSP at a rate of 4.7 X 10(-4) s-1. To look for additional signals that might arise from a covalent adduct which has been postulated to arise from reaction of enzyme with PEP, an NMR experiment was performed with an analogue of S3P lacking the 4- and 5-hydroxyl groups.(ABSTRACT TRUNCATED AT 250 WORDS)