Rhizobium meliloti 1021 - Journal of Bacteriology - American Society ...

13 downloads 0 Views 609KB Size Report
Feb 14, 1991 - produced were then linked by reductive amination (5) to bovine serum albumin and used byAntibodies, Inc., to immunize rabbits. Test of the ...
JOURNAL OF BACTERIOLOGY, May 1991, p. 3021-3024 0021-9193/91/093021-04$02.00/0 Copyright ©) 1991, American Society for Microbiology

Vol. 173, No. 9

Biosynthesis and Excretion of Cyclic Glucans by Rhizobium meliloti 1021 OTTO

GEIGER, AUDREY C. WEISSBORN, AND EUGENE P. KENNEDY*

Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 02115 Received 26 December 1990/Accepted 14 February 1991

Cyclic beta-1,2-glucans produced by Agrobacterium and Rhizobium species play an important role in the interaction of these bacteria with plant hosts. In this study, we show that (i) the neutral cyclic glucans are the biosynthetic precursors of anionic cyclic glucans; (ii) the conversion of neutral to anionic glucans is much more rapid and more extensive in exponentially growing cultures than in cultures in the stationary phase, although the latter synthesize large amounts of glucan; and (iii) the excretion of glucan, as well as the total amount synthesized, is strongly influenced by the medium.

Members of the family Rhizobiaceae produce not only neutral, unsubstituted cyclic glucans but also anionic species substituted with sn-1-phosphoglycerol (3, 12) or O-succinyl ester (10) residues. The functional and metabolic relationship of neutral and anionic glucans is obscure. To test the possible role of neutral glucans as precursors of the anionic glucans in a pulse-chase experiment, cultures (5 ml) of Rhizobium meliloti 1021, generously provided by F. M. Ausubel, growing in the log phase in TY medium (4) were labeled by the addition of [1-3H]glucose (0.2 mM; 4.5 x 106 cpm/,Imol). After growth for 30 min, cells were harvested by centrifugation at 25°C, washed twice with unlabeled TY medium, and resuspended in fresh, unlabeled TY medium at 30°C for continuation of growth. During this period of chase, samples (usually 5 ml) were withdrawn, and the cells were harvested by centrifugation, resuspended in 1 ml of a solution of bovine serum albumin (10 mg/ml), and extracted with 1 ml of 4% (wt/vol) trichloroacetic acid. The neutralized extract was chromatographed on Sephadex G-50 to obtain the total cyclic glucan and further separated into neutral and anionic fractions on DEAE-cellulose essentially as described by Miller et al. (12). At the beginning of the period of chase, the neutral glucan fraction (peak I) was the largest single fraction recovered from the DEAE-cellulose column, comprising 38% of the total count (Fig. 1A and Table 1). This value fell steadily during the chase, with a clear shift of the label to more anionic products (Fig. 1B and Table 1). At the end of 6 h of chase, the neutral fraction constituted only 4% of the total count recovered from the DEAE column. The experiment shows that the neutral fraction is the precursor of anionic fractions and that there is a continued conversion of less anionic to more anionic peaks (compare Fig. 1A and B). Treatment of anionic glucans with 0.1 N NaOH at 37°C for 30 min, which removes O-succinyl residues, made only slight changes in their chromatographic behavior, suggesting that the anionic substituents are principally phosphoglycerol residues derived from phosphatidylglycerol (10). These findings differ in one important aspect from those that might been expected from prior work on periplasmic glucans from Escherichia coli. In the latter organism, neutral glucans are *

not released into the periplasm in significant amounts (8). Substitution with phosphoglycerol or other residues appears to be the signal for their cleavage from a membrane carrier, presumably a polyprenolphosphate derivative. In contrast, in Rhizobium spp., unsubstituted glucans under certain conditions of culture are the principal products. For further identification of the labeled materials described in Table 1, a sample isolated by chromatography on Sephadex G-50, representing total glucan, was treated with 0.5 N NaOH at 100°C for 80 min to cleave phosphoglycerol residues from glucans (9). The labeled glucans were converted to neutral products that were not bound to DEAEcellulose. After complete hydrolysis in 1 N H2SO4 for 3 h at 100°C, the neutral product yielded 29 nmol of glucose, as determined by hexokinase-catalyzed phosphorylation with [_y-32P]ATP of known specific activity, in good agreement with the value of 30 nmol of glucose equivalent of total carbohydrate found with the phenol-sulfuric acid method (6), indicating that glucose is the sole sugar. Partial acid hydrolysis (0.8 N H2SO4 at 100°C for 1 h) yielded mainly glucose (47%) and a single disaccharide (26%) which was identified as sophorose (D-glucosyl-P-1,2-glucose) by thin-layer chromatography and, after reduction to sophoritol with sodium borohydride, by its highly characteristic electrophoretic behavior (15). To prepare an antiserum against P-1,2-glucan, authentic cyclic glucan obtained essentially by the procedures of Miller et al. (11) was subjected to partial acid hydrolysis to open the ring structure. The set of linear P-1,2-glucans so produced were then linked by reductive amination (5) to bovine serum albumin and used by Antibodies, Inc., to immunize rabbits. Test of the antiserum so prepared revealed its high affinity for open-chain ,-1,2-glucans containing six to eight glucose residues; shorter P-1,2-glucans, including sophorose, were bound with much lower affinity. No binding of a-linked disaccharides could be detected (data not shown). No significant binding of labeled ,-glucan to preimmune serum was found. Labeled, neutral glucan was subjected to partial acid hydrolysis, neutralized, and incubated with specific immunoglobulin G. Up to 60% of radioactivity was bound to the antibody (Fig. 2). The hydrolysate consisted of a mixture of glucans, some of which, like glucose (12%), sophorose (15%), and uncleaved cyclic glucan, react not at all or very

Corresponding author. 3021

J. BACTERIOL.

NOTES

3022

A. A

v

A

1600

200

A

b0

A

S

A A-

1200

8c

A A A

120

A

800

A A

0.0

400

40

AAA

40

0

20

30

40

50

TUBE NUMBER

B. A

A

200

A A 0

120

~~~~~~~~

0