Research ArticleMICROBIOLOGY

Antibiotic production in Streptomyces is organized by a division of labor through terminal genomic differentiation

See allHide authors and affiliations

Science Advances  15 Jan 2020:
Vol. 6, no. 3, eaay5781
DOI: 10.1126/sciadv.aay5781
  • Fig. 1 Emergence of phenotypic heterogeneity in colonies of S. coelicolor.

    (A) WT (top) and mutant (bottom) colonies and the frequency that mutants emerge from WT colonies on SFM agar (right). (B) Phenotypically diverse progeny (top) emerges after restreaking mutant colonies that vary in size, shape, and pigmentation. Representative colonies are shown. The bottom graph depicts the range of distinct morphologies that emerge after restreaking 15 random colonies. Each color represents a distinct colony phenotype.

  • Fig. 2 Genome diversity of mutant colonies determined from whole-genome sequencing and PFGE.

    Values in (A) correspond to the size (in kilobases) of genome deletions, while the hexagons represent an ~297-kb genome amplification. Each line in (B) depicts the range of deletion sizes (gray) in each mutant class, together with their respective frequencies from 30 sampled mutant strains.

  • Fig. 3 Secondary metabolite production in mutant strains determined by 1H NMR, quantitative proteomics, or zones of inhibition on B. subtilis or 40 different natural streptomycete isolates.

    (A) Principal components (PC) analysis plot of 1H NMR data. Each cluster enclosed in a colored ellipse (with 95% confidence interval) corresponds to a mutant class with a different phenotype and degree of genomic instability: WT-like strains (gray), CamSArg+ strains (blue), and CamSArg strains (red). (B and C) Volcano plots of MS-based quantitative proteomics of two representative strains 9H1A (CamSArg) (B) and 9H1B (CamSArg+) (C). Protein level is indicated by the size of the dot, and genes with ≤2-fold change and/or P ≥ 0.05 are grayed out. (D) Zones of inhibition of each strain when grown with a B. subtilis soft agar overlay. Colors represent the same mutant classes as in (A). The large dot represents the mean of four replicates, while error bars represent the SE. (E) Partial least-squares (PLS) plot of 1H NMR data partitioned by the same clusters as in (A). The heat map indicates the size of the zone of inhibition on B. subtilis. (F) Zones of inhibition of four representative mutant strains with an overlay of 40 different natural streptomycetes, each represented by a different line. Statistics are given in the main text.

  • Fig. 4 Fitness of mutant strains grown alone or during coculture with the WT strain and the effects of genome deletions on fitness and antibiotic production.

    (A) Fitness [colony-forming unit (CFU)] of mutant strains. (B) Decreases in genome size negatively correlate with fitness (top) and positively correlate with antibiotic production (bottom). (C) Division of labor during coculture of the WT and strain 9H1A at different starting frequencies. Increasing frequencies of 9H1A cause increased antibiotic production (F2,7 = 107.7, r2 = 0.969, P < 0.001) (red) but only negatively affect colony fitness at frequencies >~50% (F2,7 = 37.95, r2 = 0.916, P < 0.001) (black). Quadratic regression lines include the 95% confidence interval.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/3/eaay5781/DC1

    Fig. S1. Mutant frequencies in different media.

    Fig. S2. PacBio sequencing results of nine selected strains.

    Fig. S3. PFGE results of all sampled strains.

    Fig. S4. Volcano plots of proteomics from four mutant strain.

    Fig. S5. Trade-off between fitness and antibiotic production.

    Fig. S6. Extended evidence for division of labor during coculture of the WT and three mutant strains at different starting frequencies.

    Fig. S7. Competition assays between WT and mutant strain 9H1A at different starting frequencies.

    Table S1. Filtered proteomics data.

    Table S2. 1H NMR signals ranked by X and Y weights (w* and c) for PLS component 1.

    Table S3. Mutant strains used in this study.

  • Supplementary Materials

    The PDFset includes:

    • Fig. S1. Mutant frequencies in different media.
    • Fig. S2. PacBio sequencing results of nine selected strains.
    • Fig. S3. PFGE results of all sampled strains.
    • Fig. S4. Volcano plots of proteomics from four mutant strain.
    • Fig. S5. Trade-off between fitness and antibiotic production.
    • Fig. S6. Extended evidence for division of labor during coculture of the WT and three mutant strains at different starting frequencies.
    • Fig. S7. Competition assays between WT and mutant strain 9H1A at different starting frequencies.

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Table S1 (.pdf format). Filtered proteomics data.
    • Table S2 (.pdf format). 1H NMR signals ranked by X and Y weights (w* and c) for PLS component 1.
    • Table S3 (.pdf format). Mutant strains used in this study.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article