Streptomyces genome mining reveals enzymatic expansions underpinning its metabolic

robustness and plasticityPablo Cruz-Morales and Francisco Barona-Gómez. (pcruz@ira.cinvestav.mx)

Metabolic Diversity Evolution Laboratory, National Laboratory of Genomics for Biodiversity (LANGEBIO)

Summary

Functional redundancy has a primordial role in the arising of new enzymatic functions (i.e. gene duplication and lateral gene transfer). However, the evolutionary implications of the presence of more than one gene encoding for enzymes capable of

performing the same chemical conversions has been largely neglected in Streptomyces species. Here, we define this apparent functional redundancy as Enzymatic Expansions (EE), which were analyzed trough comparative genomics. We assume that

the Streptomyces genome should encode for traits involved on the production of their overwhelming metabolic diversity. With the release of several genome sequences from Streptomyces species it is now possible to perform a comprehensive survey

of the distinctive metabolic features of this genus. A particular emphasis was put on enzymes taking part on precursor supply central pathways (PSCP), which are hypothesized to be recruited by natural product biosynthetic pathways to perform new

functions. A novel bioinformatic pipeline was developed after such premises, and it was used to identify EEs. Our analysis implicates a particular group of central metabolic enzymes with radiation of the Streptomyces genus (type I), while another

group of EEs seems to be related to the diversification of natural product biosynthesis (type II).

A B C D E F G H I J K

Statistical analysis

Strhyg

Corglu5

ArtchlA6

TrowhiTW

Micaraf

Strros15

Bsub

Artaur

Mycvan

TrostrTw

Cordip

Brelin

Bacamy

Kocrhi

StrACTE

Clamic

Rhojos

Mycsme

CorjeiK

Corgluc

Strgri

FranCcI

BiflonLeistr

Strvir

Strcoe

Acicel

Coracc

StrACT1

Strsca

RhoerySK

StrSPB74

Strliv

Mycavi1

Strpri

Janibac

Strave

Coreff

Nocfar

Biflac

Corstr

Strfla

Nocardi

Mycmar

Mycbov

Sacery

Rhoopa

Corjei7

Myclep

RhoeryPR

Strgha

Baccer

StrspC

Sacvir

Strros11

StrSPB78

Myculc

Coramy

Kinrad

Fraaln

Rensal

StrMg

FranEA

CorgluR

Myctub

StrgriTu

Salare

Strcla

Strsvi

Saltro

Corure

Miclut

StralbJ

Propacn

Mycavia

Bifado

Corkro

Precursor supply central pathways (PSCP)

Ph

ylo

gen

y

Sequence search

PSCPHomologs

Precursor supply central pathways

(PSCP)

Ph

ylo

gen

y

EE (% conservation)

Enzyme recruitmentsEnzymatic expansions

P. acnes

S. arenicola

M. smegmatis

C. glutamicum

S. hygroscopicus

S. albus

S. sp SPB74

S. sp SPB78

S. avermitilisS. sviceus

S. scabiei

S. ghanaensis

S. viridochromogenes

S. sp MG1

S. sp CS. clavuligerus

S. pristinaespiralis

S. flavogriseus

S. griseoflavus

S. lividans

S. coelicolor

S. sp ACTE

S. roseosporus 15998

S. roseosporus 11379

S. sp ACT1S. griseusParsing

against EEs

Sequence search

Natural products enzyme families

Curated natural product

biosynthetic gene clusters

Box 1

Box 3

Box 2

Box 4

Box 5

Box 2. PSCP Sequence search.

Sequence searches were performed with Blast (4) using PSCP queries on anactynomecetes amino acid sequence database from 74 actinobacterialgenomes*. This database included twenty two streptomycetes. A cutoff score =>150 was used.

Box 3. Statistical analysis.

The collection of PSCP homologs was filtered by statistical significance. Thenumber of enzyme homologs (HN) related to a function in a particular organismwas compared against the values for the same function in the remaining species.If the value was equal or bigger than the media plus the standard deviation, thenan enzymatic expansion event (EE) was recorded.

EEHN ≥ µ + σ

Box 4. Curated natural product biosynthetic gene cluster database (cNPGC).

Metabolic reconstructions of 67 biosynthetic gene clusters of selected naturalproducts were obtained from the literature*. The enzyme sequences, excludingthe assembly complexes (PKSs and NRPSs), regulatory and transport proteins,conformed the cNPGC database.

Box 5. Natural product enzyme families Parsing against EE.

The natural product enzyme families collection was parsed against the EEs foundin the previous analysis. The enzyme functions found in both databases wereconsidered enzyme recruitment events, that is, functions present among centraland natural products metabolic pathways.

Natural product precursors

Natural products

Biomass and energy

Natural product biosynthetic gene clusters

Central metabolism

Type II expansions

Type I expansions

% Enzyme name (E.C.) Natural product biosynthetic gene cluster* Precursor supply central pathways

100 Phosphoglycerate mutase (5.4.2.1) None Glycolysis

100 Pyruvate kinase (2.7.1.40) None Glycolysis

100 Aspartate-semialdehyde dehydrogenase (1.2.1.11) None Oxalacetate amino acids

100 Branched-chain-amino-acid transaminase (2.6.1.42) None Pyruvate and Thr amino acids

100 Threonine synthase (4.2.3.1) None Oxalacetate amino acids

82 Tyr/Phe transaminase (2.6.1.1) Dehydrophos E4P and PEP amino acids

64 Fructose-bisphosphate aldolase (4.1.2.13) Clorobiocin, Novobiocin Glycolysis

55 Glyceraldehyde-3-phosphate dehydrogenase (1.2.1.12) Pentalenolactone Glycolysis

45 3-phosphoshikimate 1-carboxyvinyltransferase (2.5.1.19) ECO-02301, Enterocin, Moenomycin, TriostinA, Simocyclone, Clorobiocin, Neocarzinostatin E4P and PEP amino acids

41 Succinyldiaminopimelate transaminase (2.6.1.17) Spectinomycin, A54145, clavamic acid Oxalacetate amino acids

41 Acetilornithine aminotransferase (2.6.1.11) Spectinomycin, A54145, lipoidomycin, clavams α-ketoglutarate amino acids

36 Enolase (4.2.1.11) Phosphinotricin Glucolysis

32 Dihydroxy-acid dehydratase (4.2.1.9) SalinosporamideA Pyruvate and Thr amino acids

32 Acetolactate synthase (2.2.1.6) Actagardine, nanchangmycin, clavamic acid Pyruvate and Thr amino acids

32 dUTP diphosphatase (3.6.1.23) Puromycin Pirymidines

27 Aspartate transaminase (2.6.1.1) Dehydrophos Oxalacetate amino acids

23 Aconitate hydratase (4.2.1.3) Phosphinotricin TCA cycle

23 Cysteine synthase (2.5.1.47) D-Cycloserine 2-Phosphoglycerate amino acids

23 Phosphoglycerate dehydrogenase (1.1.1.95) Phosphinotricin 2-Phosphoglycerate amino acids

18 Histidinol-phosphate transaminase (2.6.1.9) Thiostrepton R5P amino acids

18 Anthranilate synthase (4.1.3.27) Candicidin, CDA, Indanomycin, Meridamycin, Neoaureothin, Tetracyclin, Simocyclone

Chloramphenicol Salinosporamide A, Novobiocin

E4P and PEP amino acids

18 3-deoxy-7-phosphoheptulonate synthase (2.5.1.54) Tomamycin, CDA, Salinosporamide, Tetracycline, Chloramphenicol E4P andPEP amino acids

14 Malate dehydrogenase (1.1.1.37) Dehydrophos TCA cycle

14 Asparagine synthase (6.3.5.4) Bleomycin, Fredericamycin, Lysolipin, Moenomycin, Thiostrepton, Zorbamycin, Tetracyclin Oxalacetate amino acids

14 3-dehydroquinate synthase (4.2.3.4) Neomycin E4P and PEP amino acids

14 Prephenate dehydrogenase (1.3.1.12) Enduracidin, CDA, Clorobiocin, Novobiocin E4P and PEP amino acids

9.1 Aspartate kinase (2.7.2.4) Grixazone Oxalacetate amino acids

9.1 2,3,4,5-tetrahydropyridine-2,6-dicarboxylate N-

succinyltransferase (2.3.1.117)

Clavamic acid α-ketoglutarate amino acids

4.5 Histidinol-phosphatase (3.1.3.15) Puromycin R5P amino acids

4.5 3-isopropylmalate dehydrogenase (1.1.1.85) TriostinA Pyruvate and Thr amino acids

4.5 Indole-3-glycerol-phosphate synthase (4.1.1.48) CDA E4P and PEP amino acids

Conclusions.On one hand, type I expansions were found to be universally conserved throughout the streptomycetes there were analyzed. Interestingly, enzyme recruitment events, from PSCP into natural product biosynthesis, were not found amongst this type of EE. The functions encoded within the type I expansions seem to be responsible for the deviation of precursors from the PSCPs into natural product biosynthetic pathways. The complete set of type I EEs is expected to be present in any streptomycete characterized in the future. We hypothesize that this set of enzymes was already present in the last common ancestor of Streptomyces, and that it was essential for the radiation of the genus.

On other hand, type II EEs were not conserved, and their distribution seems to be related with the natural product production profiles in a species-specific manner. These enzyme families are promising for the development of novel natural product genome mining approaches to study and exploit the metabolic diversity of the Streptomyces genus.

References.1. Borodina I, Krabben P, Nielsen J. Genome-scale analysis of Streptomyces coelicolor A3(2) metabolism. Genome Res. 2005;15(6):820-92. Kjeldsen KR, Nielsen J. In silico genome-scale reconstruction and validation of the Corynebacterium glutamicum metabolic network. Biotechnol Bioeng. 2009 ;102(2):583-97.3. Jamshidi N, Palsson BØ. Investigating the metabolic capabilities of Mycobacterium tuberculosis H37Rv using the in silico strain iNJ661 and proposing alternative drug targets. BMC Syst Biol. 2007 ;1:264. Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. Basic local alignment search tool. J. Mol. Biol. 1990; 215:403-410* References for genome sequences and natural product metabolic pathways are available upon request.

Pathway Main products Steps QuerysA Glycolysis Pyruvate, Phosphoenolpyruvate and ATP 10 49B Pentose phosphate pathway Fructose, E4P, R5P 8 38

C Citric acid cycle Acetyl CoA, oxalacetate, α-Ketoglutarate 8 94D Amino acids from 3PGA Gly, Ser, Cys 6 25E Amino acids from E4P and PEP Tyr, Phe, Trp 17 56F Amino acids from R5P His 10 37

G Amino acids from Oxalacetate Asp, Asn, Thr, Met, Lys 18 64H Amino acids from THR and PYR Ala,Ile, Leu, Val 14 42I Amino acids from AKG Glu, Gln, Pro, Arg 13 50J Purines ADP,GTP, dADP, d GTP 21 88

K Pyrimidines UTP, CTP, dCTP, dUTTP 18 66TOTAL 143 609

NON REDUNDANT QUERYS 499

Box 1. Precursor supply central pathways (PSCP) database.

Metabolic pathways were delimited from the metabolic reconstructions of Streptomyces coelicolor (1), Corynebacterium glutamicum (2), and Mycobacterium tuberculosis (3).

Fig 1. (A) EE analysis of precursor supply central pathways on actinomycetes . The genome-level survey of the EE produced profiles for each organism; as can be seen EE isoverrepresented in streptomycetes (blue). (B) Conservation and recruitment of expanded enzyme families in streptomycetes. The conservation of these enzymatic expansionsignatures, across the metabolic pathways, suggested the existence of two groups (Table 1). Type I, which are Streptomyces universally conserved EEs, and type II, which are enzymesrecruited by natural product biosynthetic gene clusters.

Table 1. Type I and type II EEs found in streptomycetes sorted by conservation (%).

(A)

(B)

Figure 2. Unifying metabolic view of the EEs in Streptomyces.

This project is supported by the National Council of Science and Technology of México (Conacyt).(PhD Fellowship 28830 and grant number 82319)P. C thanks financial support from the organizing committee of the "Microbial Metabolites: Signals to Drugs“ summer school in applied molecular microbiology