MasABK proteins interact with proteins of the type IV pilin system to affect social motility of Myxococcus xanthus

Abstract

Gliding motility is critical for normal development of spore-filled fruiting bodies in the soil bacterium Myxococcus xanthus. Mutations in mgl block motility and development but one mgl allele can be suppressed by a mutation in masK, the last gene in an operon adjacent to the mgl operon. Deletion of the entire 5.5 kb masABK operon crippled gliding and fruiting body development and decreased sporulation. Expression of pilAGHI, which encodes type IV pili (TFP) components essential for social (S) gliding, several cryptic pil genes, and a LuxR family protein were reduced significantly in the Δmas mutant while expression of the myxalamide operon was increased significantly. Localization and two-hybrid analysis suggest that the three Mas proteins form a membrane complex. MasA-PhoA fusions confirmed that MasA is an integral cytoplasmic membrane protein with a ≈100 amino acid periplasmic domain. Results from yeast two-hybrid assays showed that MasA interacts with the lipoprotein MasB and MasK, a protein kinase and that MasB and MasK interact with one another. Additionally, yeast two-hybrid analysis revealed a physical interaction between two gene products of the mas operon, MasA and MasB, and PilA. Deletion of mas may be accompanied by compensatory mutations since complementation of the Δmas social gliding and developmental defects required addition of both pilA and masABK.

Figure 1. The mas operon is predicted to encode three proteins, MXAN_1927 (MasA), MXAN_1928 (MasB) and MXAN_1929 (MasK).

A: MasA possesses two transmembrane helices, and a periplasmic domain of 100 amino acid residues (green). MasB is a membrane bound lipoprotein (purple), anchored to the inner membrane at residue 25 (black dot). MasK is a membrane-associated protein kinase (red) . B: Genomic context of the masABK operon. The mas operon is 327 bp upstream of the mglBA operon (blue arrows) and is transcribed from the opposite strand. The mas operon coordinates are indicated below the ORFs for reference. The spacing of the three predicted open reading frames (ORF) suggests that the genes are cotranscribed. C: Primer extension analysis of cDNA transcript. cDNA transcript was generated using a primer specific for MasK, 810, and the template was amplified using primers 801 and 810 (illustrated using inward arrows). The resulting amplicon was 5.4 kb and confirms expression of the masABK operon involves the formation of a polycistronic mRNA transcript.

Figure 2. MasA and MasK are membrane proteins.

A. MasA has two predicted membrane spanning regions (TM  =  transmembrane). B. PhoA fusions demonstrate that MasA has a periplasmic domain. MasA was cloned into pCR Blunt II TOPO under a lac promoter to form pSF8. pSF20 is a derivative of pSF8 with an in-frame insertion of the E. coli phoA gene at the PstI site (*), while pACW1 has an in-frame fusion at the MluI site (†) shown in panel A. Overnight cultures were adjusted to an OD₆₀₀ of 1.0 and serially diluted. 5 μl of each dilution was incubated on LB agar with 35 µg/ml X-P at 37°C for 16 hours. C. Fluorescence from a MasK-mcherry fusion expressed in an otherwise WT strain was concentrated at one pole of the cell. Yellow arrows indicate the direction of movement of cells on 1.5% agar surface.

Figure 3. Deletion of the masABK operon affects swarming.

A: WT and the Δmas strain were grown in CTPM to a concentration of 5×10⁸ cells/ml. After 10-fold concentration, 3 µl were plated on 1.5% and 0.3% agar plates as previously described . Swarm diameters was measured after 96 hours. Data represent an average of nine samples plated in three separate assays. B: Photomicrographs of the colony edge after 4 days on CTPM 1.5% agar reveal isolated cells indicative of A-motility. White arrows designate the leading edge of the colony.

Figure 4. qRT-PCR confirms pilA expression decline is specific.

A: Relative quantity of S-motility gene expression in M xanthus WT and Δmas strains. Gene expression in the Δmas strain was normalized to the WT reference (set to 1) and relative quantitation are represented as a ratio of strain/WT, such as Δmas/WT. Values represent the average of three PCR reactions. pilA is shown in red (panels A and B), pilS is shown in orange and pilT is in lighter blue. B: Genomic context of the pilA gene within the 20 kb pil gene cluster. Arrows denote known promoter regions for pilSR, pilAGHI, gspO and pilS2/R2 transcription units , .

Figure 5. Deletion of pilA does not affect masB transcription.

RNA was harvested from 1×10⁸ cells and mRNA was used to generate cDNA using a random hexanucleotide primer. Gene specific primers for masB and the 16S gene (endogenous control) were used in the amplification reaction with cDNA. masB expression was amplified from cDNA using 50, 5 and 0.5 ng template, while 16 S expression was amplified using 200, 20 and 2 pg cDNA. Both the WT and pilA strain express masB (black band in lanes 1 and 4 respectively). Hence, deletion of pilA does not adversely affect the expression of masB in M. xanthus. Deletion of masABK abolished masB expression (lanes 7–9).

Figure 6. Addition of pilA and masABK restores swarming to the Δmas mutant.

Strains were grown to a density of 5×10⁸ cells per ml and concentrated 10-fold. Aliquots (3 µl) were spotted on 0.3% agar and incubated at 32°C for 96 hr. Bars represent the average of 3 assays with 3 replicates each. Error bars represent the standard deviation. The dashed line (− −) is included for reference and represents the WT average.

Figure 7. Expression of masB is observed from the masABK pSF26 construct in vivo.

See Figure (sqRT-PCR) for RT-PCR conditions. 5 μl of reaction was electrophoresed through TAE agarose. The masB transcript is detectible in strains that carry the masABK complementing construct (lanes 4 and 7), but absent in lane 1, a strain that lacks masABK at either the chromosomal or ectopic site.

Figure 8. Immunoblot analysis shows that expression of PilA is abolished in a mas deletion strain and is only rescued by addition of a pilA construct.

Lane 1 represents the WT level of PilA expression, whereas deletion of the pilA gene is sufficient to abolish all detectible expression (lane 2). Addition of pilA at the chromosomal site rescues the defect as shown in lane 3. Lane 4 represents a strain deficient in pilA but merodiploid for the masABK locus, and does not rescue the pilA defect. ΔmasABK mutants lack pilA expression as seen in lane 5. Addition of masABK at the att site does not rescue the phenotype, as shown in lane 7, but addition of a WT copy of the pilA gene and necessary promoter region rescues expression of PilA in M. xanthus (lanes 6 and 8). Immunoblot was performed as described in materials and methods, with a polyclonal anti-pilin antibody diluted to 1∶1000.

Figure 9. masABK and pilA are both required for fruiting body formation.

M. xanthus strains were concentrated to a density of 5×10⁹ cells/ml and 20 µl aliquots were plated on starvation (TPM) agar as described in Methods. After 96 hours incubation, spots were photographed under using the 10× objective as seen above. The black bar in panel A denotes 100 µm. A: DK1622 (WT), B: DK10407, C: MxH2604, D: MxH2622, E: MxH2609, F: MxH2624.

Figure 10. MasAB-PilA interaction drives transcription from the GAL promoter in a Y2H assay.

Diploid yeast strains containing both a pGAD-C1 and pGBD-C1 derived vectors. Dilutions of overnight culture were plated on a low stringency medium (SC-LTH) Panel A, and medium stringency medium (SC-LTA), Panel B as described in Materials and Methods. Row 1: FOS/JUN (+) control, Row 2: Empty vector (pGAD, pGBD negative control), Row 3: pGADmasA, pGBDpilA, Row 4, pGADmasB, pGBDpilA.