The developmental process is characterized by a temporal cascade of development-specific gene expression, which is absolutely dependent upon cell-cell signaling (9). Five classes (A to E) of intercellular signaling mutations have been identified, designated asg for A-signal, bsg for B-signal, etc., each of which arrests development at a characteristic stage (4, 7, 9, 12). Each class of signaling mutant is thought to be defective in the production of a distinct extracellular signal that is required for continued progress through the developmental program.

In a genetic approach to identifying additional components of the bsg-dependent regulatory pathway, bypass suppressors were previously identified which allow bsgA mutants to form mature fruiting bodies containing nearly normal numbers of viable, sonication-resistant myxospores (5; J. Cusick and R. E. Gill, submitted for publication). Among the suppressors that bypassed the bsgA requirement were those that inactivated spdR, a gene that encodes an NtrC-like component of a two-component regulatory system (8).

To further characterize the relationship between spdR and intercellular signaling during development, the spdR2134 mutation was introduced into a representative of each of the five classes of signaling mutant [DK480 (asgB480), M853 (bsgA330), LS523 (csgA205), DK429 (dsg429), and JD258 (esgA)] (2, 4, 13). The spdR2134 mutation was introduced into these strains using homologous recombination to integrate plasmid pREG2134, which contains the internal NcoI fragment from within the spdR coding sequence (8) cloned into the EcoRI site of pCR2.1 (Invitrogen). Plasmid integration was predicted to disrupt the spdR gene, resulting in strains that contained only two truncated, nonfunctional copies of spdR. This genotype was confirmed by Southern blot analysis (data not shown). The transcriptional organization of the spdR locus has not been determined, and so the possibility that the spdR2134 mutation may have a polar effect on as yet unidentified downstream genes cannot be ruled out.

If the spdR mutation in these strains bypassed the signal requirement, then one would expect an increased efficiency of development in the double mutant compared to its nondeveloping parent. Development was induced by concentrating exponentially growing cells to a density of 5 × 10⁹ cells per ml in TPM (10 mM Tris-hydrochloride [pH 7.6], 1 mM KH₂PO₄, 8 mM MgSO₄), and spotting 10-μl spots onto TPMP-agarose medium (TPM supplemented with 0.1% sodium pyruvate and solidified with 1% [wt/vol] electrophoresis-grade agarose [Bio-Rad]). Media solidified with agarose were used for assessing the phenotype of asg mutants and their derivatives because certain asg mutants have a particularly high background level of sporulation on medium containing the standard bacteriologic grade of agar.

Duplicate plates containing 10 spots each were prepared for each strain and incubated at 30^°C for 7 days. Spots were gently scraped from the agar surface, resuspended in 1 ml of TPM buffer, and sonicated. Refractile spores were counted microscopically using a Petroff-Hauser counting chamber. The numbers of viable spores were determined by plating on nutrient CTT medium after heating to 50°C for 2 h to kill vegetative cells and sonicating to disperse the surviving spores. Numbers of refractile and viable spores were within 20% agreement.

For the bsg mutant, the spdR bsgA double mutant produced mature fruiting bodies that were morphologically similar to those of wild-type cells and nearly wild-type numbers of viable spores. This result confirmed that the method used in these studies to disrupt spdR suppresses the B-signaling defect imposed by bsgA.

There was no observable effect on development when the spdR mutation was introduced into the C-, D-, or E-signaling mutants However, the spdR mutation resulted in a dramatic improvement in the development of the asgA480 mutant, defective in A-signaling (Fig. 1). Morphologically, the fruiting bodies formed by the spdR asgB480 double mutant were considerably smaller and more numerous than those formed by the wild type or the spdR mutant itself, and a prominent ridge of fruiting boding structures formed a ring at the outer margin of the spotted cell suspension. Fruiting bodies contained refractile spores; however, unusually large numbers of spores also formed in the cell mat outside of the fruiting bodies. The spdR mutation increased the numbers of spores produced by the asgB480 mutant by nearly 300-fold (average of two isolates) compared with the asgB480 mutant itself, and to greater than 50% of the number produced by wild-type or spdR mutant cells.

FIG. 1. Phenotype of an asg mutant containing the spdR suppressor mutation. Exponentially growing cells in CTT medium (1) were washed and resuspended in TPM at a concentration of 5 × 109 cells per ml, spotted onto TPM-agarose medium, and photographed (more ...)

The spdR mutation did not correct the pigmentation or motility defect associated with asgB (data not shown). The spdR asgB double mutant retained the tan pigmentation and reduced swarming motility characteristic of the asgB parent. It is likely that the reduced motility contributes to the aberrant morphology and distribution of fruiting bodies produced by the double mutant.

There is mounting evidence that A-signaling is a far more complex regulatory step than originally envisioned (3, 6). To determine whether suppression of the asgB developmental defect by spdR is a common characteristic of other classes of A-signaling mutants, we determined the effect of spdR mutations on the behavior of mutants carrying two additional A-signaling mutations, asgA476 and asgC767 (Table 1). Introduction of the spdR2134 mutation into either the asgA476 or asgC767 mutant background restored the level of sporulation by more than 100-fold, to levels approaching that of the wild type (12.9 and 70.2% of wild-type level, respectively).

TABLE 1. Restoration of spore formation of representative asg mutants by the spdR2134 mutationa

Previous studies have shown that fruiting body formation, spore formation, and transcription of developmental genes by the asgB mutants are all restored by exogenous addition of conditioned starvation medium from wild-type cells undergoing development in suspension (11). The conditioned medium contains at least two active components, a heat-labile, trypsin-like protease activity, and a pool of peptides and amino acids (presumably the product of the protease activity), and are collectively known as A-factor. This activity can be mimicked by the addition of either a heterologous protease (e.g., trypsin) or an appropriate concentration of specific amino acids. A-factor activity is absent or drastically reduced in asgB mutants, including the asgB480 mutant used in these studies. However, the role of the asgB product in the synthesis or release of A-factor components has not been described.

It is possible that mutations in spdR bypass the requirement for asgB either by allowing release of authentic A-factor or by fortuitously allowing the release of an unrelated protease or pool of amino acids from the cell. In either case, a functional source of A-factor would be provided. To test this possibility, developmental supernatants of two spdR asgB double mutant isolates were assayed for A-factor activity, and the levels of activity were compared to those of the wild type and the asgB parent (Table 2). A-factor production was assayed as previously described (10, 11). Exponentially growing cells in CTT medium were washed and resuspended in MC7 (10 mM morpholinepropanesulfonic acid [MOPS; pH 7.0], 1 mM CaCl₂) at a density of 5 × 10⁹ cells per ml. The suspension was shaken vigorously at 32°C for 2 h. Cells were removed by centrifugation, and the supernatant was assayed for A-factor activity. A-factor activity was assayed by measuring the restored expression of the asg-dependent lacZ fusion Ω4521 by strain DK4324 (asgB480 Tn5 lacZΩ4521).

TABLE 2. A-factor activity in developmental supernatantsa

Transcription of this lacZ fusion occurs within the first hours of development in an A-signal-dependent fashion, and so occurs in asg⁺ cells or in asg mutants upon addition of an exogenous source of A-factor. Exponentially growing DK4324 test cells were washed and resuspended in MC7 at 5 × 10⁹ cells per ml, and 25 μl of cell suspension was mixed with dilutions of conditioned medium and adjusted to a final volume of 400 μl. Suspensions were incubated for 20 h at 32°C, followed by determination of β-galactosidase activity (as a measure of expression of the Ω4521 lacZ fusion). One unit of A-factor activity was defined as the amount required to produce 1 U of β-galactosidase (1 nmol of o-nitrophenyl [ONP] per min per 1.25 × 10⁸ cells).

A-factor activity was readily detectable in the wild-type developmental supernatant (12 U/ml of supernatant), but was present in supernatants of the parent asgB strain at only very reduced levels (0.8 U/ml, 7% of the wild-type activity). These values are comparable to those reported previously (11). When assayed under these conditions, the level of A-factor activity in developmental supernatants of the two spdR asgB double mutant isolates (0.95 and 0.6 U/ml, respectively) was not significantly different from that of the asgB parent.

These data indicate that bypass of the developmental defect in the asgB mutant by the spdR mutation does not occur by release of either authentic A-factor or a surrogate activity that functionally substitutes for A-factor. Rather, the data suggest that mutations in spdR alter the regulatory pathways in early development, rendering them independent of the A-signaling event. Furthermore, when combined with the results of earlier studies (8), these results show that mutations that disrupt spdR bypass the developmental requirement for two genetically and biochemically distinct signaling steps in M. xanthus development and suggest the possibility that the A- and B-signaling pathways share regulatory components.

Interestingly, a second bsgA suppressor has recently been identified that bypasses the requirement for B- and C-signaling but not A-signaling (J. Cusick, E. Hager, and R. E. Gill, submitted for publication). These observations suggest two parallel pathways of regulation early in development, one requiring B- and C-signaling, and the other requiring B- and A-signaling.

This work was funded by awards to R.E.G. from the National Institutes of Health (GM31900) and National Science Foundation (MCB-9631365).

We thank Heidi Kaplan for helpful discussions and Elizabeth Hager and Martin Pato for critical evaluation of these experiments and careful reading of the manuscript.