Regulation of gene expression in a mixed-genus community: stabilized arginine biosynthesis in Streptococcus gordonii by coaggregation with Actinomyces naeslundii

Abstract

Interactions involving genetically distinct bacteria, for example, between oral streptococci and actinomyces, are central to dental plaque development. A DNA microarray identified Streptococcus gordonii genes regulated in response to coaggregation with Actinomyces naeslundii. The expression of 23 genes changed >3-fold in coaggregates, including that of 9 genes involved in arginine biosynthesis and transport. The capacity of S. gordonii to synthesize arginine was assessed using a chemically defined growth medium. In monoculture, streptococcal arginine biosynthesis was inefficient and streptococci could not grow aerobically at low arginine concentrations. In dual-species cultures containing coaggregates, however, S. gordonii grew to high cell density at low arginine concentrations. Equivalent cocultures without coaggregates showed no growth until coaggregation was evident (9 h). An argH mutant was unable to grow at low arginine concentrations with or without A. naeslundii, indicating that arginine biosynthesis was essential for coaggregation-induced streptococcal growth. Using quantitative reverse transcriptase PCR, the expression of argC, argG, and pyrA(b) was strongly (10- to 100-fold) up-regulated in S. gordonii monocultures after 3 h of growth when exogenous arginine was depleted. Cocultures without induced coaggregation showed similar regulation. However, within 1 h after coaggregation with A. naeslundii, the expression of argC, argG, and pyrA(b) in S. gordonii was partially up-regulated although arginine was plentiful, and mRNA levels did not increase further when arginine was diminished. Thus, A. naeslundii stabilizes S. gordonii expression of arginine biosynthesis genes in coaggregates but not cocultures and enables aerobic growth when exogenous arginine is limited.

FIG. 1.

Growth of S. gordonii and A. naeslundii monocultures and coaggregates in CDM. (A and B) Fluorescence micrographs of mixed-species cultures after 3 h of growth. Cells were stained with antibodies against S. gordonii (green) or A. naeslundii (orange). The structure of an intact coaggregate (A) and the culture after gentle sonication to facilitate the enumeration of cells (B) are shown. Bars, 10 μm. (C) Growth of S. gordonii in monoculture (•) or coaggregates (○) and A. naeslundii in monoculture (▪) or coaggregates (□) was quantified using selective agar for A. naeslundii or nonselective agar and a limited incubation time (24 h) for S. gordonii. Samples for microarray analysis were taken at 3 h.

FIG. 2.

In silico analysis of arginine biosynthesis gene loci and arginine metabolism in S. gordonii. (A) Predicted operon structure of arginine biosynthesis genes. Promoters (forward arrows) were identified with the BProm module of MolQuest bioinformatics software, and terminators (double verticle line and loop) were found using the FindTerm module. Promoters and terminators identified with a high degree of confidence are indicated by solid lines, and weak elements are shown with a dashed line. ARG box consensus elements for recognition by ArgR-type transcriptional regulators were identified using Prodoric software and are marked by a striped rectangle. Genes and levels of down-regulation in coaggregated S. gordonii cells compared with those in monocultures are indicated (Table 2). The pyrR gene, encoding a putative regulator of pyrimidine metabolism, was <3-fold decreased in coaggregates (indicated by parentheses). (B) Arginine biosynthesis and catabolism pathways in S. gordonii and major intermediary compounds. Biosynthetic steps are indicated by solid lines, and dashed lines represent catabolic reactions. Arginine is produced from l-glutamate using carbamoyl phosphate. This in turn is synthesized by the glutamine-dependent carbamoyl phosphate synthase complex encoded by pyrA_a and pyrA_b. Note that the transfer of an N-acetyl group from N-acetyl-l-ornithine to l-glutamate, catalyzed by the argJ gene product, represents both the first and fifth steps in arginine biosynthesis. Arginine is degraded by the arginine dihydrolase system that yields two molecules of ammonia per arginine molecule. Genes that were strongly down-regulated in coaggregates are indicated with a box, and the functions encoded by these genes are listed in Table 2. The arc genes encode components of arginine catabolism as follows: arcA, arginine deiminase; arcB, ornithine carbamoyltransferase; and arcC, carbamate kinase. It is possible that the arcB gene product also catalyzes the anabolic synthesis of l-citrulline from l-ornithine and carbamoyl phosphate (see the text).

FIG. 3.

Effect of arginine availability on growth of S. gordonii DL1 or the PK3337 argH::aphA3 isogenic mutant. Symbols represent DL1 (•) or PK3337 (○) in 0 mM arginine, DL1 (▪) or PK3337 (□) in 0.1 mM arginine, DL1 (♦) or PK3337 (⋄) in 0.5 mM arginine, and DL1 (▴) or PK3337 (▵) in 1 mM arginine.

FIG. 4.

Effect of coaggregation or coculture on S. gordonii growth at low (0.025 mM) arginine concentrations. S. gordonii cells were incubated in monoculture or were mixed with A. naeslundii cells with or without vigorous vortexing to induce the formation of coaggregates. Growth of S. gordonii DL1 (A) in monoculture (•), coculture without vortexing (▪), or coaggregates (▴) or growth of S. gordonii PK3337 (B) in monoculture (•) or coaggregates (▴). (C to H) Phase-contrast micrographs of S. gordonii DL1 coaggregates (C, E, and G) and cocultures (D, F, and H) with A. naeslundii after 0 h (C and D), 10 h (E and F), or 24 h (G and H). Large coaggregates are indicated by arrows. Bar, 20 μM.

FIG. 5.

Expression of arginine-related genes during growth of S. gordonii in CDM (0.5 mM arginine). The expression of arginine biosynthesis genes (argC, argG, and pyrA_b) or a control gene (amyB) was monitored during growth in monoculture (•) or in coculture (▪) or coaggregation (▴) with A. naeslundii. The abundance of transcripts was quantified using Q-RT-PCR, and expression levels relative to those in the monoculture at 1 h are shown.

FIG. 6.

Identification of ARG box motifs in the promoter regions of arginine metabolism-related operons. (A) Sequence logo representing the B. subtilis ARG box consensus matrix that was used to search promoter regions of S. gordonii genes. (B) High-scoring matches upstream of S. gordonii operons involved in arginine biosynthesis and transport, identified using Virtual Footprint software. The similarity score is the sum of the weighted probabilities of finding the given base at each position. The maximum possible score for the ARG box element is 18.9, and the scores of known B. subtilis ARG box elements range from 15.4 to 18.4. The ARG box motifs upstream of pyrRA_aAb, argGH, and argCJBD are on the negative strand relative to transcription, as indicated. The distances between the proximal base of the ARG box and the predicted transcription start sites, based on computational identification of σ⁷⁰ promoters, are shown.