Engineering Saccharomyces cerevisiaefor pentose fermentation

Ton van MarisDelft University of TechnologyDepartment of BiotechnologyDelft, the Netherlands

Detmold, April 25, 2007

sunlight

mining

burning

geologicalprocesses

biomass(plants)

CO2

photosynthesis

fuel(gasoline)

naturalreservoirs

oiloil refinery

Mill

ion

s of

yea

rs

250

270

290

310

330

350

370

390

1850

atm

osph

eric

CO

2 (p

pm)

1900 1950 2000 2050

Year

The petrochemical carbon cycle

sunlight

mining

burning

geologicalprocesses

biomass(plants)

CO2

photosynthesis

fuel(gasoline)

naturalreservoirs

oiloil refinery

(a few) years

(ethanol) €/$

naturalreservoirs

oil

mining

geologicalprocesses

oil refinery

Ethanol Production: ~ 40 Mt.y-1 (2005)

starch, sucrose

glucose, fructose

hydrolysis

ethanol

S. cerevisiae(bakers’ yeast)

fermentation

Agricultural residues and waste streams:Abundantly available

CheapNo competition with food production

Current Technology: No Utilization of Crop Residues

‘corn stover’cornstover bagasse wheatstraw

Fuel Ethanol from Crop Residues: Strategy

Crop Residues

CelluloseHydrolysisPretreatment Fermentation

SugarMixture

Lignin

Ethanol

Combustion

Heat, electricity

Non-fermentable by S. cerevisiae

Sugars in Crop Residues: the Pentose Challenge

Corn stover Wheat straw Bagasse

Sugars (%)glucose 34.6 32.6 39.0

mannose 0.4 0.3 0.4

galactose 1.0 0.8 0.5

xylose 19.3 19.2 22.1

arabinose 2.5 2.4 2.1

uronic acids 3.2 2.2 2.2

Other (%)lignin 17.7 16.9 23.1

Grohmann & Bothast 1994Lee et al. 1997www.ecn.nl

D-xylose

Biochemistry of xylose metabolism

ethanolD-xylulose

PPP

pyruvate

D-xylulose-5-P

glucose

ATP

NADH

CO2

ATP

glucose-6-P

fructose-6-P

fructose-1,6-biP

DHAPG-3-P

PEP

NADH

ATP

ATP

ATP

glycerol

NADH

CO2

2 NADPH

xylulokinase

?

PPP

pyruvate

D-xylulose-5-P

D-xylose

xylitol

D-xylulose ethanol

glucose

NADPH

NADH

ATP

NADH

CO2

ATP

glucose-6-P

fructose-6-P

DHAP

fructose-1,6-biP

G-3-P

PEP

NADH

ATP

ATP

ATP

glycerol

NADH

CO2

2 NADPH

I

II

Xylose metabolism in non-Saccharomyces yeasts

• xylose-fermenting yeasts(e.g. Pichia stipitis)

• two-step conversion of xyloseinto xylulose:

I Xylose Reductase (XR)II Xylitol Dehydrogenase (XDH)

• Cloning of Pichia stipitisstructural genes for XRand XDH in S. cerevisiae

• Slow (aerobic) growth on xylose

• Ethanol production accompaniedby extensive production ofxylitol

Metabolic Engineering of S. cerevisiae for xylose fermentation: the beginning

6 xylose 10 ethanol + 10 CO2 + 10 ATP12 xylose 9 ethanol + 12 CO2 + 9 ATP + 6 xylitol6 xylose + 3 O2 9 ethanol + 12 CO2 + 9 ATP + 6 H2O

6 xylose6 xylitol

The XR-XDH Approach: Redox Constraints

3 O26 H2O

XR

6 ATP

6 xylose 6 xylitol 6 xylulose

6 C5P

6 NADH6 NAD

6 NADP6 NADPH

1 C6P

3 C6P

3 CO2

3 C5P

2 C6P

1 C3P 2 C3P4 CO2

8 ATP

1 ATP2 ATP

1 ethanol

4 C3P

4 ethanol

2 C3P

2 ethanol 2 ethanol

2 CO2

4 ATP1 CO2

2 ATP2 CO2

4 ATP

XDH

Research on XR-XDH strategy(1990 – present):

• Important insights in kinetics of xylose metabolism

- role of xylulokinase- xylose transport- role of pentose-phosphate-pathway enzymes- role of aldose reductase (Gre3p)- inhibitor sensitivity- aeration studies

Bärbel Hahn-Hägerdal Nancy HoThomas Jeffries

et al., and coworkers

• Redox constraints remain a challenge

PPP

pyruvate

D-xylulose-5-P

D-xylose

D-xylulose ethanol

glucose

ATP

NADH

CO2

xylitolNADPH

NADH

ATP

glucose-6-P

fructose-6-P

DHAP

fructose-1,6-biP

G-3-P

PEP

NADH

ATP

ATP

ATP

glycerol

NADH

CO2

2 NADPH

The ‘bacterial’ pathway for xylose fermentation

XI

• xylose isomerase (XI) catalysesxylose/xylulose isomerisation

• common in Bacteria, Archaea

6 xylosebacterial & ArchaealXI genes known,but no efficientexpression in yeast

6 xylose 10 ethanol + 10 CO2 + 10 ATP

xylose isomerase

6 xylulose

6 C5P

4 C6P

8 C3P

8 ethanol

2 C3P

2 ethanol

6 ATP

4 ATP

8 CO2

16 ATP2 CO2

4 ATP

pentose phosphate pathway

glycolysis

Xylose Isomerase: no redox constraints

A Novel, Fungal Xylose Isomerase Gene

Research at Nijmegen University, The Netherlands:

• Anaerobic Piromyces fungus from elephant dung• Isolation of first fungal xylose isomerase gene (XylA)

Harhangi et al. 2003Arch. Microbiol. 180:134

Kuyper et al. 2003FEMS Yeast Research 4:69

Follow-up in Delft:

• High expression levels in S. cerevisiae• XI expression not sufficient for fast xylose fermentation

+ Xyl A

wt

Metabolic Engineering of Xylose Conversion Overexpression of 5 S. cerevisiae genes (pentose-phosphate pathway)

Deletion of GRE3 (aspecific aldose reductase)

6 xylulose

6 xylulose-5P

2 C3P + 4 C6P

6 xylose

glycolysis

xylitol

ethanol

Xylulokinase (XKS1)

Transketolase (TKL1)

Ribose-5P-isomerase (RKI1)

Transaldolase (TAL1)

Ribulose-5P-epimerase (RPE1)

Aldose reductase (GRE3)

XylA

Kuyper et al. 2005FEMS Yeast Research 5:399-409

< 0.01Xylitol yield (g.g-1)

0.42Ethanol yield (g.g-1)

0.10μ xylose anaerobic (h-1)

Xylose Fermentation by Engineered S. cerevisiae

Kuyper et al. 2005FEMS Yeast Research 5:399-409

Anaerobic batch cultivation of the XI-expressing, metabolicallyengineered S. cerevisiae strain RWB 217 on xylose

xylose

glycerol

(xylitol)

CO2ethanol

XylA + XKS1↑ TAL1↑ TKL1↑ RPE1↑ RKI1↑ gre3Δ

Further Improvements: ‘Evolution in the lab’

Generationtime~ 20 years

Generationtime~ 2 hour

Selection for Improved Xylose Affinity

Kuyper et al. 2005FEMS Yeast Research 5: 925-934

Long-term cultivation of RWB217 in anaerobic, xylose-limited chemostat(D = 0.06 h-1): decrease of residual xylose concentration

CO2

ethanol

xylose

glucose

Anaerobic Fermentation of a glucose-xylose Mixture

Kuyper et al. 2005FEMS Yeast Research 5: 925-934

glucose

At industrially relevant rates and yields!

Non-fermentable by S. cerevisiae

The pentose challenge continued: L-arabinose

Corn stover Wheat straw Bagasse

Sugars (%)glucose 34.6 32.6 39.0

mannose 0.4 0.3 0.4

galactose 1.0 0.8 0.5

xylose 19.3 19.2 22.1

arabinose 2.5 2.4 2.1

uronic acids 3.2 2.2 2.2

Other (%)lignin 17.7 16.9 23.1

Grohmann & Bothast 1994Lee et al. 1997www.ecn.nl

PPP

pyruvate

(XKS1 / XYL3)

D-xylulose-5-P

D-xylose

D-xylulose

L-arabinose

(xylA)ethanol

glucose

ATP

NADH

CO2

L-ribulose L-ribulose-5-PATP

(araB)

(araD)

(araA)

ATP

glucose-6-P

fructose-6-P

fructose-1,6-biP

DHAPG-3-P

PEP

NADH

ATP

ATP

ATP

glycerol

NADH

CO2

2 NADPH

(GRE3 /XYL1)L-arabinitol

L-xylulose

xylitol

(XYL2)

NADH

NADPH

NADH

(lxr1)

(lad1)

NADPH

Pathways for L-arabinose fermentation

Arabinose fermentation by engineered S. cerevisiae

M.E. strategy Ethanol productivity

Fungal pathway

P. stipidis XYL1 + XYL2, T. reesei lad1 + lxr1, S. cerevisiae XKS1

0.35 mg g-1h-1 Richard et al. (2003)

Bacterial pathway

E. coli AraABD - Sedlak & Ho (2001)

B. subtilis AraA, E. coli AraBD+ evolution

60-80 mg g-1h-1 Becker & Boles (2003)

• Bacterial pathway most promising (Becker & Boles 2003)• Challenges: rate, anaerobicity, arabinitol formation

Strain construction

pAKXAD11440 bps

T_cyc1pMB1 ori

AmpR

URA3

P_TDH3AraA

T_ADH1P_HXT7

AraD

T_PGI1

2mu

P_TPI XylA

pRS305XKS1LpAraB11286 bps

T_ADH1

AraBP_PGI1

P_ADH1

XKS1

T_CYC

LEU2

Amp

• Host: XylA-expressing S. cerevisiae strain optimized for xylose fermentation

• Expression of AraA, AraB and AraD from Lactobacillus plantarum

Wouter Wisselink et al.unpublished

0 20 40 60 80 100 120 140 160 180 2000.00

0.05

0.10

0.15

time (days)

μ ( h

−1 )

Serial transfer in shake-flask cultures

serial transfers in synthetic medium with 2% (w/v) arabinose

Wouter Wisselink et al.unpublished

Evolutionary engineering for growth on L-arabinose

0 20 40 60 80 100 120 140 160 180 2000.00

0.05

0.10

0.15

time (days)

μ ( h

−1 )

Serial transfer in shake-flask cultures

Anaerobic SBR

Strain IMS0001

Single cell isolateIMS0002

Evolutionary engineering for growth on L-arabinose

serial transfers in synthetic medium with 2% (w/v) arabinose

Wouter Wisselink et al.unpublished

Characterisation of single-cell line:anaerobic batch cultures on L-arabinose and

glucose/L-arabinose mixture

glucose arabinose ethanol CO2 glycerol

0 10 20 30 40 50 60 70 800

20

40

60

80

100

120

140

160

0

100

200

300

400

500

time (h)

arab

inos

e, g

lyce

rol (

mM

)

etha

nol,

CO

2 (m

M)

0 10 20 30 40 50 60 70 800

20

40

60

80

100

120

140

160

0

100

200

300

400

500

time (h)

gluc

ose,

ara

bino

se, g

lyce

rol (

mM

)

etha

nol,

CO

2 (m

M)

qara= 0.66 ± 0.01 g g-1 h-1

qeth= 0.27 ± 0.01 g g-1 h-1qara= 0.46 ± 0.03 g g-1 h-1

qeth= 0.21 ± 0.02 g g-1 h-1

Wouter Wisselink et al.unpublished

Non-fermentable by S. cerevisiae

The challenge continued: uronic acids?

Corn stover Wheat straw Bagasse

Sugars (%)glucose 34.6 32.6 39.0

mannose 0.4 0.3 0.4

galactose 1.0 0.8 0.5

xylose 19.3 19.2 22.1

arabinose 2.5 2.4 2.1

uronic acids 3.2 2.2 2.2

Other (%)lignin 17.7 16.9 23.1

Grohmann & Bothast 1994Lee et al. 1997www.ecn.nl

A next challenge for metabolic engineering: pectine

+ Beetpulp

up to 30% pectine (galacturonic acid)

Syntheticmedium

From the Lab… to the Real World

cornstover bagasse wheatstraw

From the Lab to the Real World…

Syntheticmedium

Planthydrolysate

From the Lab to the Real World and back again…

Fed-batch Batch

0

1

2

3

4

5 20

10

10

ethanol

glucose

xylose

20 30 40 50

Suga

r (g

/L)

Etha

nol (

g/L)

Ethanol yield:0.47 g ethanol/g sugar238 L ethanol/ton dry

biomass

Ethanol Production from Wheat-Straw Hydrolysate(‘academic’ xylose-fermenting strain)

De Laat et al. 2006unpublished

0 6 12 18 24 30 360

5

10

15

20

25

0

2

4

6

8

10

Glucose Xylose Glycerol Acetic AcidEthanol BIomass

Time (hr)

Met

abol

ites

(g/L

)B

iomass (O

D 660 nm

)

Fermentation of glucose-xylose mixture by S. cerevisiae RWB218pH 3.5

Anaerobic batch culture, synthetic medium, 20 g.l-1 glucose and 20 g.l-1 xylose

Bellissimi et al. 2006unpublished

Low pH as such is not a problem forxylose-fermenting S. cerevisiae….

0 6 12 18 24 30 36 42 480

5

10

15

20

25

0

2

4

6

8

10

Glucose Xylose Glycerol Acetic AcidEthanol Biomass

Time (hr)

Met

abol

ites

(g/L

) Biom

ass (660 nm)

Fermentation of glucose-xylose mixture by S. cerevisiae RWB218pH 3.5, 3 g.l-1 acetic acid

Anaerobic batch culture, synthetic medium, 20 g.l-1 glucose and 20 g.l-1 xylose

Bellissimi et al. 2006unpublished

….but acetic acid specifically inhibitsxylose fermentation

• ethanol yieldon xylosenot affected

0 6 12 18 24 30 36 42 480

4

8

12

16

20

24

0

10

20

30

40

50

60

Glucose Xylose Acetic AcidEthanol BiomassGlycerol

Time (hr)

Met

abol

ites

(g/L

) Ethanol (g/L)

Continuous glucosefeed (2.4 g.l-1.h-1)

Bellissimi et al. 2006unpublished

Glucose co-feeding alleviates acetic acid inhibition ofxylose fermentation (pH 3.5 + 3 g.l-1 acetic acid)

Similar to simultaneous saccharification and fermentation (SSF)

• metabolic engineering and evolutionary engineering are powerful tools in industrial microbiology

• Isomerase-based pathways are the key for robust pentose fermentation by S. cerevisiae

• Inhibitor tolerance: experiments with real-life plant biomass hydrolysates (wheat straw, corn stover) are promising

• Future: Towards industrial scale fermentations?

Conclusions

AcknowledgementsDelft Industrial Microbiology

Derek AbbottMarinka AlmeringEleonora BellissimiJoost van den Brink Jean-Marc DaranPascale Daran-LapujadeLeonie van DijkHans van DijkenAndreas GombertDiana HarrisLucie HazelwoodEline HuisjesErik de HulsterZita van der KrogtMarko KuyperMarijke LuttikTon van Maris Jack PronkIshtar SnoekMaurice ToirkensHan de WindeWouter WisselinkRintze Zelle

A.J.A.vanMaris@TUDelft.NL

sunlight

mining

burning

geologicalprocesses

biomass(plants)

CO2

photosynthesis

fuel(gasoline)

naturalreservoirs

oiloil refinery

Mill

ion

s of

yea

rs

250

270

290

310

330

350

370

390

1850 1900 1950 2000 2050

Year

atm

osph

eric

CO

2 (p

pm)

(a few) years

(ethanol) €/$

Research on XR-XDH strategy(1990 – present):

• Important insights in kinetics of xylose metabolism

- role of xylulokinase- xylose transport- role of pentose-phosphate-pathway enzymes- role of aldose reductase (Gre3p)- inhibitor sensitivity- aeration studies

Bärbel Hahn-Hägerdal Nancy HoThomas Jeffries

et al., and coworkers

• Redox constraints remain a challenge

• Cloning of Pichia stipitisstructural genes for XRand XDH in S. cerevisiae

• Slow (aerobic) growth on xylose

• Ethanol production accompaniedby extensive production ofxylitol

Metabolic Engineering of S. cerevisiae for xylose fermentation: the beginning

Anaerobic Mixed-Substrate Utilisation

xylose

CO2

glucose

ethanol

Suboptimal kinetics of xyloseutilisation: an affinity problem?

Kuyper et al. 2005FEMS Yeast Research 5:399-409

Anaerobic growth of S. cerevisiae RWB217 on 20 g/L glucose and 20 g/L xylose

CEN.PK113-7D RWB217 RWB218Reference Genetically Evolved

engineered

Glucose Glucose Glucose

Xylose Xylose

Transcriptome analysis

• Affymetrix GeneChips • triplicate chemostat cultures(anaerobic, C-limited, D = 0.05 h-1)for each strain/condition

Transcriptome Analysis (1)Genes involved in hexose transport

Bellissimi et al. unpublished

Transcript LevelsGenes

Parental strainRWB 217

Evolved strainRWB 218

HXT 1 14.2 ± 3.87 91.7 ± 26.3 6.44 0.034

HXT 2 394.9 ± 131 2645.4 ± 756

HXT 3 2194 ± 816 39.3 ± 12.8

9.5 0.025

-39 0.18

HXT 4 325 ± 258 1872 ± 430

HXT 5 394.9 ± 46.9 109.9 ± 32.3

8.4 0.008

-3.59 0.002

HXT 6 4144.9 ± 359 3696 ± 459 -1.12 0.26

HXT 7 3573 ± 201 3824.8 ± 441 1.07 0.44

HXT 8 12.03 ± 0.058 13.6 ± 2.14 1.13 0.34

HXT 9 12 12 1.0 -

HXT 10 12.23 ± 0.4 14.13 ± 3.7 1.16 0.47

HXT 12 56.4 ± 6.5 43.2 ± 11.1 -1.3 0.17

HXT 14 12.97 ± 1.67 12.83 ± 1.27 -1.01 0.92

HXT 16 834.4 ± 231.5 26.6 ± 2.51 -31 0.026

GAL 2 12.03 ± 0.058 12.1 ± 0.17 1.01 0.58

RGT 2 36.6 ± 7.80 55.1 ± 8.51 1.39 0.08

SNF 3 29.6 ± 3.66 25.2 ± 3.46 -1.17 0.21

Fold Change

p-value

Altered kinetics of U-14C xylose transport

CEN.PK113-7D RWB217 RWB218Reference Genetically Evolved

engineered

Glucose Glucose Glucose

Xylose Xylose

Transcriptome analysis

• Affymetrix GeneChips • triplicate chemostat cultures(anaerobic, C-limited, D = 0.05 h-1)for each strain/condition

RWB 217(genetic engineering only)

70 transcripts higher on xylose70 transcripts higher on xylose25 transcripts higher on glucose25 transcripts higher on glucose

100/6000 = 1.7 % of transcriptome100/6000 = 1.7 % of transcriptome

Transcriptome analysis (2)Transcriptional responses to carbon source (glucose, xylose)

2-fold threshold, statistical analysis with SAM algorithm

Bellissimi, Kuyper et al. 2006unpublished

0 transcripts higher on xylose0 transcripts higher on xylose6 transcripts higher on glucose6 transcripts higher on glucose

6/6000 = 0.1 % of transcriptome6/6000 = 0.1 % of transcriptome

RWB 218(genetic engineering &evolutionary engineering)

Strain construction

pAKXAD11440 bps

T_cyc1pMB1 ori

AmpR

URA3

P_TDH3AraA

T_ADH1P_HXT7

AraD

T_PGI1

2mu

P_TPI XylA

pRS305XKS1LpAraB11286 bps

T_ADH1

AraBP_PGI1

P_ADH1

XKS1

T_CYC

LEU2

Amp

• Host: XylA-expressing S. cerevisiae strain optimized for xylose fermentation

• Expression of AraA, AraB and AraD from Lactobacillus plantarum

Wouter Wisselink et al.unpublished

0 6 12 18 24 30 360

5

10

15

20

25

0

2

4

6

8

10

Glucose Xylose Glycerol Acetic AcidEthanol BIomass

Time (hr)

Met

abol

ites

(g/L

)B

iomass (O

D 660 nm

)

Fermentation of glucose-xylose mixture by S. cerevisiae RWB218pH 3.5

Anaerobic batch culture, synthetic medium, 20 g.l-1 glucose and 20 g.l-1 xylose

Bellissimi et al. 2006unpublished

Low pH as such is not a problem forxylose-fermenting S. cerevisiae….

0 6 12 18 24 30 36 42 480

5

10

15

20

25

0

2

4

6

8

10

Glucose Xylose Glycerol Acetic AcidEthanol Biomass

Time (hr)

Met

abol

ites

(g/L

) Biom

ass (660 nm)

Fermentation of glucose-xylose mixture by S. cerevisiae RWB218pH 3.5, 3 g.l-1 acetic acid

Anaerobic batch culture, synthetic medium, 20 g.l-1 glucose and 20 g.l-1 xylose

Bellissimi et al. 2006unpublished

….but acetic acid specifically inhibitsxylose fermentation

• ethanol yieldon xylosenot affected

Why is xylose fermentation more sensitive to acetic acidthan glucose fermentation ?

CH3COO- + H+ ↔ CH3COOH

CH3COO- + H+ ↔ CH3COOH pH 3.5pH 7

H+

ATP

ADPADP

ATP

glucose/xylose

ethanol + CO2

Bellissimi et al. 2006unpublished

maisloof tarwestro

Oil, Ethanol & Sugar

Ethanolproductie USA

18 billion litrein 2005!

stop climate change

independence fromoil import

The petrochemical carbon cycle: Consequences

stop climate change

independence fromoil import

The Bio-ethanol Boom: a Weird & Wonderful Coalition

new marketsfor farmers

This corresponds to 30 million tonnesof fuel ethanol per year

Target: 5,75 % replacement of all petrol and diesel for transport purposes by December 31, 2010

From the Real World back to the Lab:Effects of low pH and acetic acid

Syntheticmedium

Planthydrolysate

0 6 12 18 24 30 360

5

10

15

20

25

0

2

4

6

8

10

Glucose Xylose Glycerol Acetic AcidEthanol BIomass

Time (hr)

Met

abol

ites

(g/L

)B

iomass (O

D 660 nm

)

Fermentation of glucose-xylose mixture by S. cerevisiae RWB218pH 3.5

Anaerobic batch culture, synthetic medium, 20 g.l-1 glucose and 20 g.l-1 xylose

Bellissimi et al. 2006unpublished

Low pH as such is not a problem forxylose-fermenting S. cerevisiae….

0 6 12 18 24 30 36 42 480

5

10

15

20

25

0

2

4

6

8

10

Glucose Xylose Glycerol Acetic AcidEthanol Biomass

Time (hr)

Met

abol

ites

(g/L

) Biom

ass (660 nm)

Fermentation of glucose-xylose mixture by S. cerevisiae RWB218pH 3.5, 3 g.l-1 acetic acid

Anaerobic batch culture, synthetic medium, 20 g.l-1 glucose and 20 g.l-1 xylose

Bellissimi et al. 2006unpublished

….but acetic acid specifically inhibitsxylose fermentation

• ethanol yieldon xylosenot affected

Why is xylose fermentation more sensitive to acetic acidthan glucose fermentation ?

CH3COO- + H+ ↔ CH3COOH

CH3COO- + H+ ↔ CH3COOH pH 3.5pH 7

H+

ATP

ADPADP

ATP

glucose/xylose

ethanol + CO2

Bellissimi et al. 2006unpublished