The genome of erythromycinproducing Saccharopolyspora erythraea Markiyan Samborskyy and Peter F' Lea

1
The genome of erythromycin-producing
Saccharopolyspora erythraeaMarkiyan Samborskyy
and Peter F. LeadlayDepartment of Biochemistry,
University of Cambridge, 80 Tennis Court Road,
CB2 1GA, Cambridge, UK.

• Sequencing of the S. erythraea BIO1714
  overproducer strain has revealed
• No obvious major alterations in the genome
  sequence (large deletions, rearrangements,
  duplications - though this needs checking by, e.g
  Pulsed Field Gel Electrophoresis analysis).
• 2. Of a total of 7198 CDSs, only 296 (4) have
  altered nucleotide sequence
• 62 are silent (no changes at protein level)
• 160 are minor (one or a few amino acid changes)
• 71 are major (very likely to compromise
  function).
• 3. When putative regulatory genes are considered,
  out of a total of 366 CDSs, only 19 are altered
• 4 are silent (no changes at the protein level)
• 8 are minor (one or a few amino acid changes)
• 7 are major (very likely to compromise function)
• 4. The mutations are mainly found in the non-core
  region of the chromosome, although no part of the
  chromosome is spared (Fig. 3).

Introduction Saccharopolyspora erythraea is a
Gram-positive filamentous bacterium from
Actinomycetes class, originally identified as
Streptomyces erythraeus but later assigned to the
genus Saccharopolyspora. S. erythraea is an
industrial producer of erythromycin A, an
important broad-spectrum antibiotic against
pathogenic Gram-positive bacteria. Semisynthetic
derivatives of erythromycin have a total annual
market value of 5 billion.
Results
Fig. 1. a. S. erythraea NRRL23338 growing on
solid agar media b. electron micrograph of S.
erythraea NRRL23338 c. Erythromycin A. The
commercial importance of erythromycin has
fostered intensive research into its
biosynthesis, and genetic engineering of the
pathways involved promises to enhance production
of potentially valuable analogues of polyketide
secondary metabolites. This has revived efforts
to increase strain productivity. Historically,
wild-type actinomycete strains have been
subjected to multiple rounds of random
mutagenesis and selection to obtain overproducing
mutants for industrial production of a desired
secondary metabolite. However, genome-scale
information might allow such actinomycete strains
to be more quickly optimized for production. We
present here the complete sequence of the S.
erythraea genome and initial comparative
statistics of its differences with an
industrially-derived overproducer strain, S.
erythraea BIO1417. Although a catalogue of such
differences obviously does not provide instant
answers to the molecular basis of overproduction,
it offers a starting point for hypothesis and
experimentation.

• Fig. 3. S. erythraea genome distribution of
  altered ORFs between wild type (NRRL23338) and
  Bulgarian (BIO1714) strains.
• Perspective
• Only a subset of the noted genome differences are
  likely to contribute to the higher fermentation
  yield of erythromycin in the over-producing
  strain. The sequencing results provide a
  framework for further work to help identify these
  key differences. Approaches include
• differential transcriptome analysis
• differential proteome analysis
• focus on regulatory genes
• focus on specific feeder pathways
• Candidate genes highlighted by any of these
  approaches can then be readily manipulated in S.
  erythraea and the effects on erythromycin
  production determined.
• Reference
• M. Oliynyk, M. Samborskyy, J. B. Lester, T.
  Mironenko, N. Scott, S. Dickens, S. F. Haydock
  P. F. Leadlay. Complete genome sequence of the
  erythromycin-producing bacterium
  Saccharopolyspora erythraea NRRL23338 Nature
  Biotech. 25, 447 - 453 (2007).
• Ikeda, H. et al. Complete genome analysis and
  comparative analysis of the industrial
  microorganism Streptomyces avermitilis. Nat.
  Biotechnol. 21, 526531 (2003).
• Bentley, S.D. et al. Complete genome sequence of
  the model actinomycete Streptomyces coelicolor
  A3(2). Nature 417, 141147 (2002).
• Acknowledgement
• I would like to thank John Lester, the DNA
  Sequencing Facility, Department of Biochemistry,
  University of Cambridge and the Cambridge
  Commonwealth Trust for their support Markiyan
  Oliynyk for help with annotation and comparative
  analysis Tatiana Mironenko and Nataliya Scott
  for help with sequencing.

Fig. 2. Schematic representation of the S.
erythraea chromosome. The main features of the
chromosome sequence are shown in Figure 2.
Starting from the outside, the rings show CDS
distribution on the forward and reverse strands
genes essential for core metabolism secondary
metabolite biosynthetic genes (including
erythromycin repetitive elements and rRNA
operons GC content and GC skew. The blue arc
defines the core region containing most of the
essential genes. The genome is circular. The
genome is comparable in size to the linear
genomes of S. coelicolor M145 (8.7 Mbp)3 and S.
avermitilis MA-4680 (9.0 Mbp)2. However, the S.
erythraea genome is (unexpectedly) circular, a
topology it shares with other actinobacteria such
as Mycobacterium tuberculosis and Clostridium
diphtheriae. The S. erythraea chromosome contains
7,198 predicted protein-coding sequences
(CDSs)1. The genome reveals the presence of at
least 18 different gene clusters for polyketide
and non-ribosomal peptide secondary metabolites
(Fig. 3)
Methods Sequencing and assembly. Whole-genome
shotgun sequencing of the S. erythraea NRRL2338
genome was done by conventional Sanger
sequencing. The final assembly contained 7.1-fold
coverage with an estimated error rate of 1 per
100,000 bases of the consensus sequence. The
overproducer strain (BIO1714) was sequenced by
combining in-house whole-genome shotgun
sequencing with 454 genome sequencing data
obtained from 454 Life Sciences, Branford, CT,
USA. The current overproducer assembly contains
202 contigs covering 98 of the
genome. Comparative genome analysis and
annotation. CDSs were predicted and annotated
using the program fgenesB45 (http//www.softberry.
com/), trained ab initio, and manually curated
using Artemis (version 8) and a set of in-house
PERL scripts (http//131.111.43.95). The
comparison of genomes was done using BLAST and a
set of PERL scripts. Any differences with an
error probability lt0.001 per base were considered
real. For an initial survey, the differences in
CDSs were (fairly arbitrarily) categorized
according to the identity level of the predicted
protein sequences 100 - silent 85-99.9 -
minor 0-85 major.
Fig. 3. S. erythraea gene clusters for polyketide
synthases (yellow) and non-ribosomal peptide
synthetases (blue) .