A polyphasic approach for identification of bacteria: the Burkholderia cepacia example

1
Microbial genomics
Tom Coenye
2

• Overview
• The sequencing of prokaryotic genomes
• Analysis of prokaryotic genome sequences
• Diversity of prokaryotic genome sequences
• Use of genome sequences in biodiversity studies
• Case studies

3

• The sequencing of prokaryotic genomes
• Growing the organism DNA isolation
• Approach 1 map-based cloning sequencing
• Approach 2 shotgun sequencing
• Annotation

4

• The sequencing of prokaryotic genomes
• Growing the organism DNA isolation
• Isolating DNA is normally not a problem
• Biggest challenge isolating DNA from
  unculturable organisms
• eg. Blochmannia floridanus abdomens of 100
  Camponotus floridanus ants were crushed, cell
  debris was removed by filtration, bacterial
  cells were collected and treated with Dnase I to
  remove ant DNA, then normal DNA preparation

5
Approach 1 map-based cloning sequencing
Order of fragments is known before start of
sequencing
6
Approach 2 shotgun sequencing Order of
fragments is not known before start of sequencing
7
Approach 2 shotgun sequencing
8
Approach 2 shotgun sequencing
9
Approach 2 shotgun sequencing
10
Approach 2 shotgun sequencing
11

• Annotation the search for genes
• Identification of genes is of utmost importance
  but not straight forward
• rRNA genes trough similarity with known rRNA
  genes
• tRNA genes trough similarity with known tRNA
  genes and/or use of statistical models (eg HMMs)
  
• Protein-coding genes computer models

12
(No Transcript)
13

• Annotation the search for genes
• Protein-coding genes computer models

SD start
stop ...AGGA........ATG(XXXXXX)nTAA
P
14

• Annotation the search for genes
• Biological confirmation remains necessary!
• E.g. over-annotation is a common problem

Skovgaard et al., 2001
15

• Sequencing centers
• Sanger Centre (UK) has sequenced more than 2
  billion bases between 4 October 1993 and now
• Equipment includes 250 ABI 3700 and MEGABACE
  capillairy sequencers

16
Evolution of sequencing
100000000
90000000
80000000
70000000
60000000
50000000
40000000
30000000
20000000
10000000
0
77
95
03
97
17
Evolution of sequencing
45,861,651,747 bases in 31,904,910 records
18

• Analysis of prokaryotic genomes
• In silico analysis
• Microarrays
• Subtractive hybridisation
• Proteomics
• Making knock-out mutants

19

• In silico analysis of prokaryotic genomes
• Comparison of genome sequences on the computer,
  using specialised bioinformatics tools
• Comparison of gene content
• Alignment of genes for phylogenetic purposes
• Looking for gene duplications
• ...

20
(No Transcript)
21

• Microarrays
• DNA Microarray miniaturised version of a
  dot-blot hybridisation
• Immobilised DNA is being hybridised with labelled
  probe DNA or RNA
• High throughput (up to 300000 spots/array)

22

• Microarrays
• Different technologies
• PCR product arrays (amplification of ORFs
  spotting or in situ PCR)
• Oligonucleotide arrays (immobilisation of
  pre-synthetised oligos or in situ synthesis)

23
E.g. XeoChips
24

• Microarrays
• Study of gene expression (are these ORFs
  expressed under certain conditions?)
• Comparative genomics
• Comparing gene content
• All genes or subset (eg. virulence genes)
• Allows to include organisms in comparative
  genomics research without the availability of a
  complete genome sequence

25

• Microarrays
• Eg identification of new genes in Klebsiella
  pneumoniae by means of an E. coli microarray

26
Microarrays
Green E. coli specific Red K. pneumoniae
specific Yellow common genes
27
Subtractive hybridisation PCR-based way to
isolate sequences present in genome 1 but not in
genome 2
28
Subtractive hybridisation
29

• Proteomics
• Study of the expressed protein-complement of the
  genome, expressed by specific cells at a
  specific moment under specific conditions
• Typical approach separation of proteins by
  2D-gel- electrophoresis (IEF SDS-PAGE) and
  identification of differentially expressed
  proteins by means of mass- spectrometry

30
Proteomics
Vb. induction of heat-shock proteins by
elevated temperature in Bradyrhizobium japonicum
(Munchbach et al., 1999)
31

• Knock-out mutants
• Final proof for biological function of gene
  product
• Time consuming
• Analysis not straight forward, for example
• Duplicated genes
• No phenotype
• ...

32

• Diversity of prokaryotic genomes
• Which organisms have been sequenced ?
• Genome size and organisation
• Number of genes, functional groups
• Visualisation of properties of prokaryotic genomes

33
Which organisms have been sequenced
? http//igweb.integratedgenomics.com/GOLD/
34
Which organisms have been sequenced
? http//www.ncbi.nlm.nih.gov/genomes/MICROBES/Com
plete.html http//www.tigr.org http//www.jgi.d
oe.gov/JGI_microbial/html/index.html http//www.s
anger.ac.uk/
35
Which organisms have been sequenced ?
Finished In progress Total Bacteria 128
391 519 Archaea 17 22
39 Total 145 413 558
36

• Which organisms have been sequenced ?
• Available genomes are NOT representative of total
  biodiversity
• Emphasis on
• Medically important organisms (humans and
  animals)
• Organisms with biotechnological applications

37

• Which organisms have been sequenced ?
• Example 1 Escherichia coli
• 4 completed genomes available
• 7 additional genomes in progress
• Strains K12 and O157 have been sequenced twice

• Example 2 Agrobacterium tumefaciens C58 has
  been sequenced twice

38

• Genome size and organisation
• Genome size of 580074 bp (Mycoplasma genitalium)
  to 9105828 bp (Bradyrhizobium japonicum)

39

• Genome size and organisation
• Genome organisation
• 1 circular chromosome (eg. Escherichia coli 4.6
  5.4 Mbp)
• Multiple circular chromosomes (eg. Ralstonia
  solanacearum 3.7 Mbp en 2.1 Mbp Burkholderia
  cenocepacia 3.8 Mbp, 3.2 Mbp en 0.9 Mbp)
• 1 linear chromosome (eg. Borrelia burgdorferi 0.9
  Mbp)
• 1 linear and 1 circular chromosome (eg.
  Agrobacterium tumefaciens 2.8 en 2.1 Mbp)

40

• Genome size and organisation
• Plasmids can also be present
• Borrelia burgdorferi 1 linear chromosome (0.91
  Mbp)
• Also contains 21 circular plasmids (total 0.61
  Mbp)

41

• Genome size and organisation
• GC 22.5 (Wigglesworthia brevipalpis) to
  72.1 (Streptomyces coelicolor)

42

• Genome size and organisation
• Number of genes 480 (Mycoplasma genitalium) to
  8317 (Bradyrhizobium japonicum)

43

• Genome size and organisation
• Number of genes 480 (Mycoplasma genitalium) to
  8317 (Bradyrhizobium japonicum)
• Coding density 75 (Rickettsia prowazekii) to
  94 (Campylobacter jejuni)
• Very compact when compared to eukaryote genomes
  (eg Fugu rubripes only 25 of genome is
  coding)

44

• Relationship genome size/number of genes
• (Konstatinidis Tiedje, 2003)
• Linear relationship between genome size and
  number of genes (r20.97)
• Linear relationship between genome size and
  coding density (r20.72)

Larger genomes have more genes but not more junk
DNA !
45

• Relationship genome size/number of genes
• (Konstatinidis Tiedje, 2003)
• Regulatory genes (signal transduction and
  transcriptional control) overrepresented in
  larger genomes
• Genes involved in metabolism and transport also
  overrepresented in larger genomes
• Gene involved in DNA metabolism are
  underrepresented in larger genomes

46
Visualisation
AGACCGAAATTTACGCACCTGTGGACAATCTGGGGAGAATTTTGAACAGT
TCCGTCTTATTCCAGTAAT TCACAGGCGTCTCGAAGACGAGAGACGCCA
CTTGCGGATTGTGGAAAAACACCACCTTATTCACCCCGCG
GCTCGGCCCGTCGGACAATTCAGAGATTTGTCCCGGTTTATCAACAGGGG
GAGAAAAACAGCGTGGAGAA CAAAAAAAGCTTCTTCCATCTGCACCTGA
TTTCGGACTCGACGGGAGAGACTCTGATGTCGGCCGGCCGC
GCCGTCTCGGCGCAGTTTCATACATCCATGCCGGTGGAACATGTCTATCC
GATGATCCGCAACCAGAAGC AGCTCGCGCAGGTCATCGATCTCATCGAC
AAGGAGCCCGGCATTGTTCTTTATACAATCGTTGATCAGCA
GCTGGCGGAATTCCTGGATCTGCGCTGCCATGCGATTGGCGTGCCCTGCG
TCAACGTTCTCGAACCGATC ATCGGCATTTTCCAGACCTATCTCGGCGC
GCCGTCCAGGCGGCGGGTGGGTGCGCAACACGCGCTGAATG
CCGATTATTTCGCGCGGATCGAAGCACTCAATTTCGCCATGGACCATGAT
GACGGGCAGATGCCGGAGAC CTATGACGATGCGGATGTCGTCATCATCG
GCATCAGCCGCACGTCGAAAACACCAACCAGCATCTATCTT
GCTAACAGGGGCATAAAGACTGCCAATATCCCGGTCGTTCCCAATGTGCC
TTTGCCCGAAAGCCTATATG CCGCGACCCGGCCGTTGATCGTCGGTCTC
GTCGCGACATCGGATCGCATATCGCAGGTTCGTGAGAACAG
GGATCTGGGTACAACCGGCGGATTTGACGGTGGCCGTTACACGGATCGCG
CCACCATCATGGAAGAGCTG AAATATGCGCGTGCGCTCTGCGCCCGCAA
CAATTGGCCGCTGATCGACGTCACACGCCGTTCCATCGAGG
AAACGGCCGCGGCGATCCTTGCCCTGCGCCCGAGGACGCGATAATCCGAA
TCGCATCATCAGGAGCAGAC AGTCGATGAAACAAGAGTTGATCCTCGCC
TCATCCAGCGCATCCCGGCAGATGCTGATGCGCAATGCGGG
GCTGACATTTTCGGCAATACCCGCGGATATTGATGAGCGTGCGCTTGATG
AGCAACTGGAACGGGACGGC GCCAGCCCCGAAGAGGTTGCGCTGGAACT
TGCGCGGGCGAAGGCTCTTGCAGTCAGTGCGCTCCATCCAG
AAGCACTGGTTCTTGGCTGCGACCAGACCATGGCGCTCGGCACACGCGTT
TATCACAAGCCAAAAAACAT GGCGGAAGCCGCGACGCATCTGCTGTCGT
TGTCCGGCAAGGTCCACCGCCTGAACAGCGCGGCTGTTCTC
GTTCACAACGGAAAGGTGGTGTGGCAGACCGTTTCCAGTGCAGAGCTTGC
CGTTCGAACCTTGAGCGCTG AGTTTGTGTCCCGCCACCTGCAGCGGGTG
GGAGAAAAGGCGCTCAGCAGCGTCGGCGCTTACCAGCTTGA
GAGGGAAGGAATCCAGCTATTCACCTCCATAGAGGGGGATTATTTCACGA
TCCTCGGTTTGCCGCTTCTG CCTCTTTTATCGAAACTACGCGACATGGA
TGTCATCGATGGCTGATTCACGTGAAACATTAACTATAAAT
GCCTTCGTTGTCGGTTACCCGATCAAACATTCCCGGTCGCCGATCATCCA
TTCCTATTGGCTGAAAAAAT TCGGTATCGCCGGTTCCTATACGGCAGTT
GAGGTCTCCCCAGACGATTTCCCGAAGTTCATTGCAACGCT
GAAGGAAGGCAAGCCGGGTGCAGCGGTGGGCGGTAACGCCACCATTCCGC
ACAAGGAAGCGGCTTACCGG TTGGCCGATCATCCCGATGCCTTGGCGGA
AGAACTCGGCGCCGCCAACACCATCTGGATGGAGGAGGGTA
AACTCCACGCGACCAACACGGATGGTTACGGTTTCGTCTCGAACCTGGAC
GAGCGGCATCCGGGCTGGGA TAAGACCCAGCGCGCGGTGGTGTTCGGCG
CCGGCGGTGCAAGCCGGGCCGTCATTCAGTCGCTGCGTGAT
CGGGATGTTGCGGAAATTCACGTCGTGAACCGTACGGTCGAGCGCGCTCG
CGAACTGGCCGACCGCTTTG GCCCACGGGTCTTTTCCCATCCCCAGGCA
GCGCTTCAGGAGGTCATGCACGGCGCGGGGTTGTTCGTGAA
47

• Visualisation
• Genome Atlas (http//www.cbs.dtu.dk/services/Genom
  eAtlas/)
• TIGR CMR (http//www.tigr.org/tigr-scripts/CMR2/ch
  oose_genome.spl)
• Artemis (http//www.sanger.ac.uk/Software/Artemis/
  )
• ACT (http//www.sanger.ac.uk/Software/ACT/)
• Apollo
• SeqVISTA
• EMBOSS, JEMBOSS, EMBASSY, ...

48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
Novel approaches in taxonomy based on
whole-genome sequences

• Supertree approach
• Presence/abscence of characteristics
  (genes/gene families/protein folds/conserved
  indels)
• Differences in gene content
• Gene order / synteny
• Differences in nucleotide composition

53
Supertrees

• Availability of genome sequences allows to
  determine which genes are shared between a
  number of organisms
• These genes can then be used for phylogenetic
  analysis
• Analysis of protein-coding genes can deliver
  important additional information compared to 16S
  rRNA gene sequences
• Combined alignments of conserved proteins

54
Presence/abscence analysis
55
Presence/abscence analysis

• Genes

• Gene families

56
Presence/abscence analysis

• Genes

• Gene families

• Protein folds (b-propeller, TIM-barrel,
  Zn-b-lactamases, ...)

• Conserved insertions and deletions (indels)

57
Presence/abscence analysis
glmU (UDP-N-acetylglucosamine pyrophosphorylase)
shows a 17 aa insertion in Archae and Chlamydiae
58
Presence/abscence analysis
59
Differences in gene content and gene order

• Simple approach consider 2 genomes as bags
  filled with genes and compare content of both
• Extension if the same genes are present, are
  they present in the same order?

60
Differences in gene content and gene order

• Synteny less conserved than gene content
• Syntenie lt shared genes lt identity between shared
  genes

61
Differences in nucleotide composition

• GC (A, C, G, T)
• Relative abundance of di/tri/tetra nucleotides
  (AA, AC, AG, AT, CA, ...)
• Mathematical
• rXY fXY/fXfY (X, Y A, C, G, T XY AA,
  AC, AG, AT, ..., TT)
• d(f,g) 1/16 S r(f) r(g) (within species
  lt 20)
• Similar statistics apply to higher-order
  nucleotides
• Codon statistics (codon usage, GC content of
  synonymous 3rd position)

62
Examples

• Genome-scale metabolic model of Helicobacter
  pylori 26695. Schilling et al., J. Bact.
  1844582-4593 (2002)
• Phylogenetic position of the Aquificales
• Comparative analysis of the genome sequences of
  Bordetella pertussis, Bordetella parapertussis
  and Bordetella bronchiseptica. Parkhill et al.
  Nature Genetics (2003) (can be downloaded at
  http//www.sanger.ac.uk/Projects/B_pertussis/Borde
  tella_genomes.pdf)

63

• Goal
• Thorough study of the metabolic potential of
  Helicobacter pylori based on genome sequence
• Calculate the growth demands

64
(No Transcript)
65

• Several key enzymes from glycolysis (PFK, PK) are
  absent deficient in hexose metabolism
• All pentose phosphate reactions are present
  except G6P-DH oxidative branch of pentose
  phosphate pathway incomplete
• Entner-Doudoroff pathway is present C can be
  transferred from pentose phosphate pathway to
  G3P en pyruvate
• No PEP carboxylase PEP can not be transformed
  to oxaloacetate
• PEP can not be transformed to TCA intermediate
• TCA cycle is complete but in a slightly modified
  form

66
(No Transcript)
67

• Required components Ala, Arg, His, Ile, Leu,
  Met, Phe, Val, thiamine, phosphate, oxygen and
  sulphate/cysteine
• No requirement for purines (de novo synthesis is
  possible)
• Ala and Arg are important C sources - use of AA
  as C sources is coupled to production of ammonia
• 14 alternative C sources were identified
• Oxygen necessary for production of NAD, NADP,
  CTP, UTP, dCTP, dUTP necessary because there
  is need for an e- acceptor to allow oxidation of
  FADH to FAD

68

• 9 essential AA for humans, 8 for H. pylori 6 in
  common
• Strategy of metabolic design seems ideal and
  could be coupled to co-evolution of this
  organism to its host (ie humans)
• Elimination of genes required for energetically
  not favoured AA biosynthesis (proteolysis in
  stomach releases unlimited supply of AA)
• Use of AA as C source results in production of AA
  what helps to neutralise the acid pH in the
  stomach

69
A genomic perspective on the taxonomic
position of Aquifex aeolicus

• Genus Aquifex marine, hyperthermophilic,
  microaerophilic, hydrogen- oxidising, recovered
  from marine sediments near Iceland
• Based on 16S rRNA sequence data one of the
  deepest branching lineages
• But, considerable debate regarding its true
  taxonomic postion
• Deep (type of ribosomes, EF-G and EF-Tu
  sequences, ribosomal proteins, rpoB and rpoC
  sequences)
• Not deep but close to e-Proteobacteria
  (Helicobacter, Campylobacter, Wolinella)(cytochro
  me bc, four amino-acid insert in
  ala-tRNA- synthetase, ultrastructure)

70
16S rRNA gene
71
Differences in gene content
72
Dinucleotide relative abundance
73
The supertree
74
Conclusions (?)

• Coming to a consistent picture is not
  straightforward
• No particular relationship between Aquifex and
  e-Proteobacteria
• Aquifex can be considered as a primitive species
  with primitive genes