Engineering Conferences InternationalECI Digital Archives

Metabolic Engineering IX Proceedings

Summer 6-6-2012

Sustainable Production of Industrial ChemicalsUsing Microbial Biocatalysts: 1,4-Butanediolmark BurkGenomatica

Follow this and additional works at: http://dc.engconfintl.org/metabolic_ix

Part of the Biomedical Engineering and Bioengineering Commons

This Conference Proceeding is brought to you for free and open access by the Proceedings at ECI Digital Archives. It has been accepted for inclusion inMetabolic Engineering IX by an authorized administrator of ECI Digital Archives. For more information, please contact franco@bepress.com.

Recommended Citationmark Burk, "Sustainable Production of Industrial Chemicals Using Microbial Biocatalysts: 1,4-Butanediol" in "Metabolic EngineeringIX", E. Heinzle, Saarland Univ.; P. Soucaille, INSA; G. Whited, Danisco Eds, ECI Symposium Series, (2013).http://dc.engconfintl.org/metabolic_ix/16

GenomaticaSustainable Chemicals

GenomaticaSustainable Chemicals

Genomatica

• Sustainable chemicals

• Better economics

• Smaller footprint

Mark J. Burk, CTO

Metabolic Engineering IX

June 2012

Sustainable Production of Industrial Chemicals Using Microbial Biocatalysts: 1,4-Butanediol

Genomatica BDO Team - Past and Present

Molecular Biology

Harry Yim

Steve Van Dien

Bob Haselbeck

John Trawick

Wei Niu

Jeff Boldt

Laura Peiffer

Eric Van Name

Chris Wilson

Stephanie Culler

Microbiology

Catherine Pujol-

Baxley

Jazell Estadilla

Jesse Wooton

Jabus Tyerman

Jonathan Moore

Lars Knutstad

Paul Handke

Jonathan Joaquin

Enzymology

Brian Steer

Stefan Andrae

Cara Tracewell

Mike Kuchinskas

Wayne Liu

Brian Kinley

Amit Shah

Jacqueline Fritz

Process Engineering

Computational

Tony Burgard

Priti Pharkya

Robin Osterhout

Jun Sun

Tae Hoon Yang

Wyming "Lee" Pang

Fermentation

Dan Beacom

Sy Teisan

SABBernhard Palsson

Sang Yup Lee

Jens NielsenStephanie Culler

Brandon Chen

Kevin Hoff

Ewa Lis

Fannie Chau

Hongmei He

Shawn Bachan

Jingyi Li

Luis Reyes

Analytical Sciences

Julia Khandurina

Rosary Stephen

Lucy Zhao

Ahmed Alanjary

Blanca Ruvalcaba

Rainer Wagester

Korki Miller

Process Engineering

Joe Kuterbach

Michael Japs

Janardhan Garikipati

Fasil Tadesse

Ben Adelstein

Rachel Pacheco

Daric Simonis

Arvind Kaul

Ishmael Sonico

Christophe Schilling, CEO Bill Baum, CBO

Nelson Barton, VP R&D Jeff Lievense, EVP, Process Development

Sy Teisan

Brett Schreyer

Laurie Romag

Joseph Woodcock

Don Miller

Gian Oddone

Amruta Bedekar

Rebecca Bratcher

Jason Crater

Akhila Raya

Alex Navarro

Jens Nielsen

George Church

Lee Hood

Harvey Blanch

Bernhard Hauer

Sugars 1,4-Butanediol(BDO)

• Direct production• Meets application

Genomatica’s BDO Process in Engineered E.coli

1.4 M ton/year

BDO-producing E. coli

• Meets application specs

Strain, fermentation, process engineering → deliver BEP

BDO Strain Engineering Progress

2012108 g/L

1.2 M

2011

Genomatica’s Systems-Based Strain Engineering

Enzyme Evolution

ComputationalTechnologies

Pathway &

Strain Design

ProductFeedstockOrganism & Tools Select

Parent strain

Synthetic Biology Tools

Omics dataSystems analysis

Analyze &Interpret

Genomics

Metabolomics

Proteomics

Transcriptomics

>100 g/L BDOIn 3 years

Whole CellMutagenesis

Engineered strains

HT Screening: In vivo assays

Da

taLIMSFermentation development/scale-up

Transcriptomics

13C-Fluxomics

IterativeStrain

Engineering

Journey to a BDO Production Strain

Pathway Identification

glucoseex

acetyl-CoA citrate

phosphoenolpyruvate

glyoxylate

malate

pyruvate

ADP, NAD+

ATP, NADH

NAD(P)HCO2

NADH

CO2

NAD+

ATP

CoA

ADP

Pi

succinate

oxaloacetate

succinyl-CoA

fumarate

NAD+

NADH,CO2

CO2

hexose-P

-ketoglutarate

NAD(P)+

CoA

succinatesemialdehyde

CO2 NAD(P)+NAD(P)H

isocitrate

ATP

ADP

4-hy

drox

ybutyrate

4-hy

drox

ybutyrylCoA

acetate

AMP

quinol

quinone

4-hydroxybutyryl-aldehyde

1,4-butanediol

1,4-butanediolex

NAD(P)+

NAD(P)H

NAD(P)+

, CoA

ATP, CoA

NAD(P)H

sucA

sucD

4hbd

ald

adh

PTS system

ppc

ATPADP

sucroseex

glk

ATP

ADP

frk

NAD(P)+

NAD(P)H

acs

sucAB, lpdA

lpdA

acetyl-CoA

mqo

icdA

fumABC

sdhABCD

acn

aceA

aceB

gltA

ubiquinone

ubiquinol

pckA

ADP

CO2

ATP

cat2

renewable feedstock

sustainable chemicals

mdh

arcA

ldh

adhE

pflB

K.p.lpdD354KlpdA

gltAR163L

rrnC::cscAKB

Strain Design and Commercial Strain forPathway Identification

and Engineering

Strain Design and

Metabolic Engineering

Commercial Strain for

BDO Production

• Titer (g/L) - Impacts equipment sizing and energy needs• Rate (g/L/h) - Impacts # of fermentors, plant capacity• Yield (g/g) - Impacts feedstock cost contribution

Fermentation Metrics → Higher TRY = Lower COGS

TRY all inter-dependent → reduce by-products, increase rate and yield

BDO Pathway and Process

E. coli

Glycolysis

TCACycle

BDOPathway

BDO

E. coli

Sugars Glycolysis

TCACycle

BDOPathway

C H O + 0.5 O � C H O + 2 CO + H O

>100 g/L

• BDO pathway involves 4 reduction steps – redox intensive

• BDO pathway generates 1 extra NAD(P)H and no excess ATP

• Balance energy, redox and maintain high NAD(P)H/NAD(P)+ ratio

• Microaerobic production (DO ≈ 0) required for optimal performance

ATP = 0

NAD(P)H = +1

Oxidative TCA cycle flux required for redox needs of BDO pathway

ATP via oxidative

phosphorylation

C6H12O6 + 0.5 O2 � C4H10O2 + 2 CO2 + H2OMax yield = 1 mol/mol (0.50 g/g, 67 C-mol %)

Increasing Rate and Lowering By-products

Key Advances

• Backflux

• Enzymes

• Redox

2 PEP

PYR

AceACCOAOA

G6P

CIT

+ATP

+NADH

+2 NADH

1 GLUCOSE

CO2 loss

Acetate

Ethanol-2 NADH

More metabolic steps in a pathway increases avenues for by-products

• Redox

• ATP supply

• Regulation

• Balanced expression

• Fermentation PD

CIT

ICIT

AKG

SUCSAL

4HB

4HBALD

BDO

-ATP

+NADPH

- NADH

- NAD(P)H

- NAD(P)H

Glutamate

CO2 loss via

oxidative TCA

4-HBSUCCOA

TCA

Cycle

GBL

+ATP

BDO Biosynthetic Pathways

Sugar

• Prioritized pathways proceed through 4-hydroxybutyrate

• Downstream enzymes function on non-natural substrates

ALDCat2 ADH

Alternative routes

1. Developed enzyme assays and analytical methods for all metabolites

2. Screened libraries of gene candidates for each step – >100 in some cases

3. Demonstrated seven different functional BDO pathways in E. coli

Yim et al., Nature Chem. Biol., 2011

Omics Analysis of Reverse C-flux

Sugar

Endogenous genes responsible

for reverse C-flux

ALDCat2 ADH

•13C-Flux analysis identified significant drain due to competing pathways

• Microarray analysis identified candidate genes involved in backflux

• Intracellular metabolite measurements indicated downstream bottleneck

Deletion of sad and gabD

∆∆∆∆sad

Deleted endogenous succinate

semialdehyde dehydrogenases

Sugar

mm

olC

O2

per

L f

erm

enta

tion5000

4000

3000

2000

1000

0

500

400

300

200

100

0

mM

in F

erm

enta

tion B

roth

Eckh-432

Eckh-432 ∆∆∆∆sad ∆∆∆∆gabD

∆∆∆∆sad

∆∆∆∆gabD

ALDCat2 ADH

• 30% increase in BDO, 3 fold 4-hydroxybutyrate

• Significant decrease in pyruvate, acetate, CO2

Deletion of Endogenous Alcohol Dehydrogenases

Sugar

KO all 4 ADHs

No Backflux

• Fractionation/proteomics

• Deleted 4 ADH’s

• All 4 backflux ADH’s

NADP+-dependent

E. coli ADH backflux eliminated

Ba

ckfl

ux

Act

ivit

y

ALDCat2 ADH

Backflux from

endogenous

ADH’s

Improving the Downstream Pathway: Cat2

4-HB 4-HB-CoA 4-HBal BDO

ALDHO

O

OHHO

O

SCoAHO

O

HHO

OHor

Cat2

BK/PTB

ADH

Cat2 inhibited >90%

by high [BDO]

Enzyme Discovery - Bioinformatics Directed Enzyme Evolution

New discovered Cat2 has >6 X rate in 1M BDO

0

2

4

6

8

10

1209 1210034001 033

Cat2 Activity in 1M BDO

Spec

ific

Act

ivit

y

(µm

ol∙m

in-1

∙mg-1

)

Cat2 Enzyme

0

2

4

6

8

10

1209 1210034001 033

Cat2 Activity in 1M BDO

Spec

ific

Act

ivit

y

(µm

ol∙m

in-1

∙mg-1

)

Cat2 Enzyme

New Cat2

Original Cat2’s

Enzyme Discovery - Bioinformatics

6x

Evolution increased rate of 033 2.5 X in 1M BDO

2.5x

Improving the Downstream Pathway: ADH

4-HB 4-HB-CoA 4-HBal BDO

ALDHO

O

OHHO

O

SCoAHO

O

HHO

OHor

Cat2

BK/PTB

ADH

Screened over 200 ADH enzymes

14 fold

increase

Evolved best ADH parent - 956

increase

in kcat/KM

Improving the Downstream Pathway: ALD

4-HB 4-HB-CoA 4-HBal BDO

ALDHO

O

OHHO

O

SCoAHO

O

HHO

OHor

Cat2

BK/PTB

ADH

Evolution

Enzyme Discovery

Evolution

• Combined discovery and evolution

• BDO productivity improved 8 fold

• Evolved ALD uses NADH and NADPH

Constitutive Promoter Libraries (σ70)

RFP

pConstitutive

TTGACAGCTAGCTCAGTCCTAGGTATAATGCTAGC-35* -10*

Consensus σσσσ70 PromoterTransform promoter

variants into the BDO

production strain

RPU = RFP fluorescence of promoter RFP fluorescence of reference

promoter p100

1.40

1.60

1.80

Re

lati

ve

Pro

mo

ter

Un

its

(R

PU

)

Control promoters (p100,p104,p111,p119)1.40

1.60

1.80

Re

lati

ve

Pro

mo

ter

Un

its

(R

PU

)

Control promoters (p100,p104,p111,p119)

*Sites randomized via

degenerate primers

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

P2

00

P1

11

P1

19

P2

01

P1

04

P2

02

P2

03

P2

04

P2

05

P2

06

P2

07

P2

08

P2

09

P2

10

P2

11

P2

12

P2

13

P2

14

P2

15

P2

16

P2

17

P1

00

P2

18

P2

19

P2

20

P2

21

P2

22

P2

23

P2

24

P2

25

P2

26

P2

27

P2

28

P2

29

P2

30

P2

31

P2

32

P2

33

P2

34

P2

35

P2

36

P2

37

P2

38

P2

39

P2

40

P2

41

P2

42

P2

43

P2

44

P2

45

P2

46

P2

47

P2

48

P2

49

P2

50

Re

lati

ve

Pro

mo

ter

Un

its

(R

PU

)

Promoter Variants(M9 media, OD600= 1.0)

Control promoters (p100,p104,p111,p119)

-10 Library variants

-35 Library variants

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

P2

00

P1

11

P1

19

P2

01

P1

04

P2

02

P2

03

P2

04

P2

05

P2

06

P2

07

P2

08

P2

09

P2

10

P2

11

P2

12

P2

13

P2

14

P2

15

P2

16

P2

17

P1

00

P2

18

P2

19

P2

20

P2

21

P2

22

P2

23

P2

24

P2

25

P2

26

P2

27

P2

28

P2

29

P2

30

P2

31

P2

32

P2

33

P2

34

P2

35

P2

36

P2

37

P2

38

P2

39

P2

40

P2

41

P2

42

P2

43

P2

44

P2

45

P2

46

P2

47

P2

48

P2

49

P2

50

Re

lati

ve

Pro

mo

ter

Un

its

(R

PU

)

Promoter Variants(M9 media, OD600= 1.0)

Control promoters (p100,p104,p111,p119)

-10 Library variants

-35 Library variants

Optimization of BDO Pathway Gene ExpressionBDO production influenced by promoter strength and ALD protein levels

40

60

80

100

[BD

O]

g/L

2 L Fermentations

Promoter strength

0

20

ALD

ALD levels determined by Western blot

p1

11

p1

08

p1

04

p1

00

p1

05

p1

15

Higher soluble acCve protein = higher BDO → not always strongest promoter

Lowering By-Products: γ-Butyrolactone (GBL)

4-HB 4-HB-CoA 4-HBal BDO

ALDHO

O

OHHO

O

SCoAHO

O

HHO

OHor

Cat2

BK/PTB

ADH

OOhydrolase

• GBL formed by cyclization of 4-HB-CoA (C-yield loss)

• GBL formation enzyme induced and spontaneous

• Enzymes identified by microarray experiments

Two approaches to eliminate GBL by-product:

1. Delete genes responsible for GBL

2. ID and introduce new hydrolase

GBL

0.0

100.0

200.0

300.0

400.0

500.0

600.0

700.0

800.0

879 1889 1889 +hydrolase

mM BDO mM GBL

0.0

100.0

200.0

300.0

400.0

500.0

600.0

700.0

800.0

879 1889 1889 +hydrolase

mM BDO mM GBL

BDO

GBLNo GBL

KO hydrolase

Lowering By-Products: Excess CO2

C6H12O6 + 0.5 O2 � C4H10O2 + 2 CO2 + H2O

2 PEP

PYR

ACCOAOA

G6P

CIT

+ATP

+NADH

+2 NADH

1 GLUCOSE

X

• Excess CO2 from complete TCA

• Delete sucCD gene – lose 1 ATP

• Eliminated “futile” energy drains

CIT

ICIT

AKG

SUCSAL

4HB

4HBALD

BDO

-ATP

- NADH

- NAD(P)H

- NADHCO2 loss via

oxidative TCA

SUCCOA

TCA

Cycle

+ATP

sucCD

In vivo Flux Distribution during BDO Production

4

38 h timepoint

∆∆∆∆sucCD strain

• 13C flux analysis demonstrates little oxidative TCA flux in ∆∆∆∆sucCD strain

• Irreversibility introduced between BDO pathway and central metabolism

BDO

CO2

∆∆∆∆sucCD StrainCO2

BDO

Lowering Excess CO2 via sucCD Deletion

Parent Strain

∆sucCD Strain: higher BDO, much lower CO2

BDO Scale-up and CommercializationJoint Development Partnership with Tate & Lyle

• $4.3B per year• Operates four corn wet mills• JV with DuPont: PDO, 100-140M lb/yr

40M lb/yr

100M+ lb/yrper plant

Genomatica taking proven path:� Same base organism� Same scale-up factor� Similar chemical� Similar cost model

Demonstration

Commercial

2009 2010 2011 2012 2013 2014

Lab

Pilot

Non-integrated Integrated

30L3,000L

13,000L240,000L

600,000LDemonstration

13,000 L fermentorin Decatur demo plant

Bio-BDO® Becoming a Commercial Reality

2008 2013first production of 1,4-BDO from carbohydrates

commercial scale production (40M lbs/yr)

Development of a Robust BDO Production Strain

� Systems-based approach20122012

Pathway Identification

and Engineering

Strain Design and

Metabolic Engineering

Commercial Strain for

BDO Production

� Systems-based approach

� Over 35 genetic manipulations

� Eliminated reverse C-flux

� Improved pathway enzymes

� Promoter libraries/expression tuning

� Eliminated/reduced by-products

� Fully integrated, constitutive strain

� Achieved commercial TRY targets

� Process scale-up to 13,000 L (1 ton/wk)

2011

2012108 g/L

1.2 M

2011

2012108 g/L

1.2 M

Questions?

Thank you.

Mark BurkMark Burk

mburk@genomatica.com