Outline - Screening•

(1)

Outline - Screening

• Performance in application conditions (multiple criteria)

(actual reaction using actual conditions)

• One critical feature, e.g., rate or selectivity

(actual reaction using more convenient conditions)

• One critical feature using a substrate analog

(more convenient reaction using more convenient conditions)

• Selection (growth depends on critical feature)

(rarely applies to biocatalysis-relevant reactions)

You only get what you screen for.

- Frances Arnold

faster

(2)

Measure performance using application conditions

• Usually hard: e.g., wash clothes, treat an oil well,

reaction under process conditions, works after six months storage

• Advantage - engineered protein will solve your problem; maintains other properties needed for good performance

• ^Examples

- carotenoid synthesis (unusually easy)

- esterase (substrate analog limits success)

- transaminase

(3)

Improving one property must not damage other properties

Efficiency Specificity

Stability Turnover frequency (k_cat)

Temperature profile

Specific activity (kat/kg, U/mg)

pH profile

Space-time yield

Product inhibition

Byproduct/ingredient inhibition

Producibility/expression yield

Temperature stability pH stability

Ingredient/byproduct stability

Solvent stability

Substrate range

Substrate specificity (K_m, k_cat/K_m)

Substrate regioselectivity and enantioselectivity

Substrate conversion (%), yield

6 5 4 3 2 1 1 2 3 4 5 6

laundry detergents⁸. Their versatility allows their use in many applications, including processes to degrade natural polymers such as starch, cellulose and proteins, as well as for the REGIOSELECTIVE or ENANTIOSELECTIVE

synthesis of asymmetric chemicals.

Global industrial enzyme sales were esti- mated at $2.3 billion in 2003 ^{REF. 15.} The main profits were divided among detergents ($789 million), food applications ($634 million), agriculture/feed ($376 million), textile processing ($237 million), and pulp/paper, leather and other applications including enzymes for the production of fine and bulk chemicals ($222 million). In the face of soar- ing energy costs, dwindling fossil resources, environmental pollution and a globalized economy, the large-scale use of biotechnology instead of, or to complement, traditional industrial production processes, particularly in the chemical sector, is viewed as both an opportunity and a necessity. In the future, novel biotechnological applications should boost the market for industrial enzymes.

White biotechnology

The notion that environmentally sound, commercially viable biotechnological processes can take their respected place in a global industrial environment has been acknowledged for some time¹⁶. Companies from Europe, Canada, Japan, South Africa and the USA reported on their experiences in

processes as diverse as the production of acryl- amide (Mitsubishi Rayon, Japan) and the use of enzymes in oil-well completion (British Petroleum Exploration, UK). Currently, the movement towards implementing sustainable technologies and processes is gaining momentum, particularly in Europe.

‘Industrial’ or WHITE BIOTECHNOLOGY is currently a buzzword in the European bio- business community. The term was coined in 2003 by the European Association for Bioindustries (EuropaBio), based on a case study report, and it denotes all industrially harnessed bio-based processes that are not covered by the RED BIOTECHNOLOGY (medical) or GREEN BIOTECHNOLOGY (plant) labels¹⁷. White biotechnology has its roots in ancient human history and its products are increasingly part of everyday life, from vitamins, medicines, biofuel and bioplastics to enzymes in detergents or dairy and bakery products. It is the belief of industrial promoters, analysts and policy makers that white biotechnology has the potential to impact industrial production processes on a global scale. The main long-term applications of white biotechnology will be focused on replacing fossil fuels with renewable resources (biomass conversion), replacing conventional processes with bioprocesses (including metabolic engineering) and creating new high-value bioprod- ucts, including nutraceuticals, performance chemicals and bioactives.

Besides the involvement of the food, feed, detergent and politically heavily- promoted biofuel industries, it is the globally operating chemical and pharmaceutical industries that are riding this ‘third wave of biotechnology’¹⁸ (a term illustrating the chronology of developments in which red and green biotechnology come first and sec- ond, respectively). As industries face increas- ing low-cost competition, particularly from East Asia, and political pressure to reduce their environmental impact and resource consumption to improve sustainability, it is felt that there is a strong need for smart and innovative technologies, processes and products to remain competitive.

The McKinsey consultancy predicts that, by 2010, biotechnology could be applied in the production of between 10% and 20% of all chemicals sold (amounting to a value of $160 billion) and that up to 60% of all fine chemicals (medium-volume products used as intermediates in the manufacturing of products such as pharmaceuticals, flavours, fragrances, agro-chemicals and detergents) might be produced using biotechnology¹⁷. Even for the traditional mainstay of the chemical industries, the polymer market (typical bulk-volume products), McKinsey predicts that biotechnology might account for up to 10% of the output.

The European Chemical Industry Council and EuropaBio have responded to a call from the European Commission, which asked for suggestions for ten central technological plat- forms, by proposing a ‘European Technology Platform for Sustainable Chemistry’. White biotechnology will be one of three pillars of the platform, together with materials science, and reaction and process design. As a corollary, this triggered the nomination of sustainable chemistry as one of the thematic priorities in the European Union’s seventh framework research programme, with the aim of strengthening the scientific and technological bases of European industry and encouraging its international competitive- ness. Clearly, this joint European initiative does not come without ‘stimulating’ prec- edent from forward-looking international peers in the United States or Japan, and leading European stakeholders (among them DSM¹⁹, Degussa, BASF, Henkel, Novozymes and Genencor) have urged policy makers to act to ensure a favourable competitive posi- tion. The promises of industrial biotechnology cannot be realized without coordinated and well-funded research and development (R&D) efforts, a supportive and guid- ing political framework and early public involvement to pre-empt concerns over new

Figure 1 | Multi-parameter footprint analysis. This figure illustrates the ideal biocatalyst concept. Each enzyme candidate from the metagenome is ranked, from low (rating of 1) to high (rating of 6) using a

specific set of criteria, to produce a multi-parameter fingerprint (shown in yellow). Criteria include in vitro enzyme activity, efficiency, specificity and stability. This decision matrix reveals the strengths and weaknesses of every candidate enzyme, so that the most promising candidate enzymes from diverse enzyme libraries can be selected for further process development by re-screening, protein engineering or directed

evolution methods. kat, catalytic reaction rate; k_cat, catalytic constant; K_m, Michaelis constant; U, unit.

NATURE REVIEWS | MICROBIOLOGY VOLUME 3 | JUNE 2005 | 511

F O C U S O N M E T A G E N O M I C S

Lorenz & Eck (2005) Metagenomics and industrial applications. Nature Rev. Microbiol. 3, 510.

(4)

Rare: improved variants easy to find

NATURE BIOTECHNOLOGY VOL 18 JULY 2000 http://biotech.nature.com 751

R E S E A R C H A R TI C L E S

neurosporene, as in Rhodobacter, or four desaturations to produce lycopene, as in Erwinia and other photosynthetic bacteria²⁰. The desaturase from Neurospora crassa introduces five double bonds into phytoene to synthesize 3,4-didehydrolycopene²¹. A desaturase capable of introducing six double bonds into phytoene would lead to the production of the fully conjugated carotenoid 3,4,3’,4’-tetradehydrolycopene.

The phytoene desaturase from Erwinia uredovora has been shown to synthesize small amounts of 3,4,3’,4’-tetradehydrolycopene under cer- tain conditions^22,23. Our first goal was to extend the desaturation pathway by evolving an efficient six-step desaturase in order to generate 3,4,3’,4’-tetradehydrolycopene as the major carotenoid in E. coli (Fig. 1).

To create variant enzyme libraries in the context of a biosynthetic pathway in E. coli requires co-transformation with two plasmids that together are stably propagated. Genes that produce the carotenoid precursors that serve as substrates for the target enzyme were cloned on a pACYC184-derived plasmid. Genes for the enzymes that were subjected to evolution in vitro were cloned on a pUC19-derived plasmid. All enzymes were individually expressed under the control of a lac promoter followed by an optimized Shine–Dalgarno sequence.

Escherichia coli cells co-transformed with pAC-crtE_EU-crtB_EU, expressing the GGDP synthase (crtB_EU) and the phytoene synthase (crtE_EU) from E. uredovora (EU), and with pUC-crtI_EU or pUC-crtI_EH expressing the phytoene desaturases (crtI) from E. uredovora and E.

herbicola (EH), respectively, produced lycopene as the exclusive carotenoid (Fig. 2A). These cells appeared orange to orange-red on plates and in liquid culture. A library of desaturases generated by in vitro homologous recombination (DNA shuffling⁵) of the genes from E. herbicola and E. uredovora was transformed into phytoene- synthesizing E. coli JM101 harboring pAC-crtE_EU-crtB_EU.

Colonies were transferred to nitrocellulose membranes, which pro- vide a white background for visual screening of the clones based on color. Approximately 10,000 colonies were screened; 30% appeared white as a result of inactivation of the desaturase. Twenty colonies were yellow, indicating the presence of carotenoids with fewer conjugated double bonds than lycopene. In addition, we identified one pink clone (I14), suggesting the introduction of additional double bonds into lycopene by this mutant. Analysis of cell extracts by high-pressure liquid chromatography (HPLC) showed that the desaturase of I14 introduces two double bonds in lycopene, which leads to the accumulation of the terminal desaturation product 3,4,3’,4’-tetradehydrolycopene in addition to lycopene (Fig. 2B). Yellow mutant I25, in contrast, introduces two double bonds in phytoene (Fig. 2C). Reflecting the stepwise

nature of desaturation, I25 synthesizes neurosporene and lycopene in addition to the main product, ζ- carotene (whereas wild type produces only lycopene—Fig. 2A). Although it is reported that wild-type Erwinia desaturases can produce dehy- drolycopenes^22,23, we do not observe even trace amounts under these cultivation conditions.

Sequence analysis of the I25 desaturase showed two amino acid changes, R332H and G470S, in the sequence of the crtI_EU, and no recombination.

G470S is located in a hydrophobic C-terminal domain that is thought to be involved in substrate binding and the dehydrogenation reaction, and is conserved among carotenoid desaturases²⁴. In mutant I14, the N terminus (residues 1–39) of the desaturase from E. uredovora is replaced with that of E. herbicola, which differs in only four residues (P3K, T5V, V27T, L28V). The I14 desaturase also contains two amino acid substitutions, F231L and A269V. We constructed two chimeras to determine whether the N-terminal recombination or the point mutations (or both) were responsible for the altered catalytic activity of mutant I14. Chimera I contained only the recombined N terminus, and chimera II contained only the two amino acid changes. Surprisingly, only chimera I exhibited the altered catalytic activity of mutant I14. The N terminus comprises a typical dinucleotide-binding domain (GXG(X)₂A/G(X)₃A(X)₆G) (ref.

25) not previously associated with substrate specificity. Thus, cofactor binding (FAD in Erwinia desaturases²²) appears to play an important role in controlling substrate specificity.

Molecular breeding of cyclic carotenoid biosynthesis. Starting from neurosporene in Rhodobacter or lycopene in other photosynthetic bacteria, diverse acyclic carotenoids are synthesized by further desaturation, hydroxylation, and methylation. Yet other bacteria, e.g. Erwinia, synthesize cyclic carotenoids from lycopene. These modifying enzymes show a high degree of promiscuity that allows them to act equally well on neurosporene and lycopene in engineered pathways^26–29. We reasoned that carotenoids with a further extended chromophore would also be modified by these enzymes or their evolved variants, leading to novel carotenoids in E. coli. The next step, therefore, was to generate new pathways for the biosynthesis of cyclic carotenoids by in vitro evolution of the cyclase.

Bacterial lycopene cyclases usually introduce β-ionone rings at both ends of lycopene to produce β,β-carotene³⁰ (Fig. 1). However, when neurosporene is produced by a three-step desaturase from Rhodobacter, or ζ-carotene is produced by a two-step desaturase from Synechococcus sp. in an engineered pathway, the cyclase is capable of cyclizing not only the ψ-end group (as in lycopene and at one end of neurosporene) to the β-end group, but also the 7,8-dihydro–end group (as at one end of neurosporene and in ζ-carotene) to the 7,8-dihydro–end group²⁸(see Fig. 1 for carotenoid structures). Synthesis of the respective monocyclic intermediates demonstrated that the enzyme acts on the two ends separate- ly. The proposed reaction mechanism for cyclization involves only the double bonds C1-C2 (C1’-C2’) and C5-C6 (C5$-C6’), which agrees with the observed broad substrate specificity³¹. Hence, we reasoned that breeding could generate a lycopene cyclase that efficiently cyclizes 3,4- didehydrolycopene, the precursor of 3,4,3’,4’-tetradehydrolycopene in the evolved extended desaturation pathway of I14.

The biosynthetic pathway consisting of GGDP synthase (crtB_EU), phytoene synthase (crtE_EU), and either wild-type phytoene desaturase (crtI_EU) or mutant I14 was extended with the genes for the lycopene cyclase (crtY) from E. uredovora orE. herbicola by cloning the desaturase genes into pAC-crtE_EU-crtB_EU to yield pAC-crtE_EU-crtB_EU-crtI_EU /I14 and complementation of E. coli pAC-crtE_EU-crtB_EU-crtI_EU/I14 with pUC- crtY_EU or pUC-crtY_EH. Escherichia coli cells expressing wild-type desat-

Figure 1. C₄₀ carotenoid biosynthesis branches into a variety of pathways to acyclic and cyclic carotenoids for which biosynthetic genes from bacteria have been cloned (for a review see refs 15, 17). Red arrows indicate how the central desaturation pathway has been extended to obtain the fully conjugated 3,4,3’,4’-tetradehydrolycopene and subsequent branching of this pathway for the synthesis of torulene.

pathways. Nat. Biotechnol. 18, 750-3.

752 NATURE BIOTECHNOLOGY VOL 18 JULY 2000 http://biotech.nature.com

R E S E A R C H A R TI C L E S

urase crtI

_EU

on pAC-c r tE

_EU

-c r tB

_EU

-c r t I

_EU

together with the lycopene cyclases crtY

_EU

or crtY

_EH

on pUC-c r tY

_EU

or pUC-c r tY

_EH

, respectively, synthesized predominantly β,β-carotene from lycopene and turned bright yellow-orange (Fig. 3A). A less-polar carotenoid with a spectrum typical for β-zeacarotene, the monocyclic product derived from neu- rosporene, was also produced (data not shown). In contrast, E . col i pAC-c r tE

_EU

-c r tB

_EU

-I14 expressing I14 desaturase together with the lycopene cyclases only synthesized β,β-carotene (not shown) and devel- oped a bright orange color (Fig. 3A). Neither 3,4,3’,4’-tetradehydroly- copene nor cyclization products of its precursor 3,4-didehydrolycopene are synthesized, suggesting that lycopene (the precursor to 3,4-didehy- drolycopene and 3,4,3’,4’-tetradehydrolycopene) is a good substrate for the wild-type cyclases. Desaturase variant I14 appears to have higher desaturation activity than the wild-type enzyme, since no neurosporene accumulates that can be cyclized to β-zeacarotene.

To access the cyclization products of the extended desaturase pathway, then, a library of lycopene cyclases was created by shuffling the genes c r tY

_EU

and c r tY

_EH

. This library was used to transform E . col i cells harboring pAC-c r tE

_EU

-c r tB

_EU

-I14. Among ∼4,500 clones screened, 20% were pink as a result of inactivation of the cyclase.

Twenty-five colonies with colors different from wild type were selected (Fig. 3B), including some that were orange-red to purple- red, indicating the possible cyclization of 3,4-didehydrolycopene.

The selected clones accumulated different ratios of lycopene,

3,4,3’4’-tetradehydrolycopene and β,β-carotene, whereas clones expressing wild-type enzymes exclusively formed β,β-carotene.

One clone Y2 appeared bright red compared to the yellow-orange color of the wild type (Fig. 3A); its extract showed a marked absorp- tion maximum of 480 nm. Analysis by HPLC revealed small amounts of the acyclic carotenoids lycopene and 3,4,3’,4’-tetradehydrolycopene and the lycopene cyclization products β,β-carotene and β, ψ- carotene, as well as a new, major carotenoid (Fig. 4A). The absorption maxima

²³

, mass, and polarity of this new product show it to be toru- lene, the cyclization product of 3,4-didehydrolycopene. Expression of the Y2 cyclase together with the wild-type desaturase, in contrast, resulted in the synthesis of monocyclic β, ψ-carotene and dicyclic β,β- carotene, and no torulene (Fig. 4B). Sequence analysis of mutant Y2 revealed two amino acid changes, R330H and F367S, in the sequence of the E . u r e d ovo r a cyclase and no recombination. Neither mutation is located in motifs conserved among various cyclases

³⁰

.

Figure 2. HPLC analysis of carotenoid extracts of E. coli transformants carrying plasmids pAC-crtE_EU-crtB_EU and (A) pUC- crtI_EU expressing the wild-type phytoene desaturase; (B) pUC-I14 expressing desaturase mutant I14; (C) pUC-I25 expressing desaturase mutant I25. The following carotenoids were identified:

peak 1, 3, 4, 3’, 4’-tetradehydrolycopene (λmax = 480 510 540 nm); peak 2, lycopene (λmax = 444 470 502 nm); peak 3, neurosporene (λmax = 415 440 468 nm); peak 4, ζ-carotene (λmax = 378 400 425 nm). Double peaks indicate different geometrical isomers. Insets: recorded absorption spectra of individual HPLC peaks. Corresponding carotenoid extracts are shown. Results for pUC-crtI_EH were similar to pUC-crtI_EU.

Figure 3. Cell pellets of E. coli transformants expressing wild-type and mutant cyclases. (A) JM109 carrying plasmid pUC-crtY_EU or pUC- Y2, together with pAC-crtE_EU-crtB_EU-crtI_EU or pAC-crtE_EU-crtB_EU-I14.

(B) JM109 transformants carrying pAC-crtE_EU-crtB_EU-I14 and various cyclase mutants.

Figure 4. HPLC analysis of carotenoid extracts of E. coli transformants carrying plasmids (A) pAC-crtE_EU-crtB_EU-I14 and pUC- Y2 expressing desaturase mutant I14 together with cyclase mutant Y2 and (B) pAC-crtE_EU-crtB_EU-crtI_EU and pUC-Y2 expressing wild-type desaturase together with cyclase mutant Y2. The following carotenoids were identified: peak 1, 3, 4, 3’, 4’-tetradehydrolycopene (λmax = 480 510 540 nm; M+ at m/e = 532.4); peak 2, lycopene (λmax = 444 470 502 nm, M+ at m/e = 536.4); peak 3, torulene (λmax = 454 480 514 nm, M+ at m/e = 534.5); peak 4, β,ψ-carotene (λmax = 435 450 478 nm, M+ at m/e = 536.4); peak 5, β,β-carotene (λmax = 425 450 478 nm, M+ at m/e = 536.4). Double peaks represent different geometrical isomers. Insets: recorded absorption spectra of individual peaks.

A

B

A

B A

B

C

(5)

N H₂N

O Cl

O O

target

NO₂ N H₂N

O Cl

O O

NO₂ chromogenic

analog

relative rate with chromogenic analog

relative rate with target

J. C. Moore, F. H. Arnold (1996) Directed evolution of a p-nitrobenzyl esterase for aqueous-organic solvents, Nature Biotechnol. 14, 458-467.

Best variant was 24-times more active than wt to chromogenic analog, but only 4-times more active toward target.

(6)

Savile et al. (2010) Biocatalytic asymmetric synthesis of chiral amines from ketones applied to sitagliptin manufacture, Science, 329, 305.

N O NH₂

F F

F

N N N

CF₃ N

F F O

F

N N N

CF₃ O

NH₂ O

detect by HPLC

!j#

Solvent % Conversion

J7!_# Q.S#

7,)%/8*1# Q.Z#

;)%/8*1# Q.U#

P(*+&*+/8*1# Q.V#

#

Isopropylamine:#P8')'/1 (5&,,8'8N#'>,8)'?',>#"#7#i-H&IM:#/(#*+)'</1#5*85,8)&/)'*8#?*&#,8F9<,#

&/),2#6=)#=8?/0*&/61,#?*&#,8F9<,#()/6'1')9.#@%,#i-H&IM:#5*85,8)&/)'*8#R/(#</'8)/'8,>#/)#[QQ#<7#'8#

&*=8>(#"#/8>#:#)*#,8(=&,#)%/)#/5)'0,#`6=)#+*(('619#=8()/61,c#0/&'/8)(#R,&,#'>,8)'?',>.#@%'(#R/(#'85&,/(,>#)*#

"#7#*85,#(=??'5',8)#,8F9<,#/5)'0')9p()/6'1')9#R/(#,()/61'(%,>.#;<+1*9'8N#/#1/&N,#,O5,((#*?#)%,#/<'8,#

>*8*&#idH&IM:#'(#/1(*#'<+*&)/8)#?*&#(%'?)'8N#,i='1'6&'=<#)*R/&>#+&*>=5)#(')/N1'+)'82#(*#%'N%,&#

5*85,8)&/)'*8(#R,&,#>,('&/61,.#

Table S8. !5&,,8'8N#5*8>')'*8(#?*&#,/5%#&*=8>#*?#,0*1=)'*8.

Round # 1 and 2 3 4 5 6 7-9 10-11

Substrate, g/L :# [# "Q# TQ# "QQ# "QQ# [Q#

[i-PrNH2], M Q.[# Q.[# ".Q# ".Q# ".Q# ".Q# ".Q#

Cosolvent [k#######

J7!_#

[k#

7,_M#

[k##

7,_M#

"Qk#

7,_M#

:Qk#

7,_M#

:[k#

J7!_#

[Qk#

J7!_#

pH Z.[# Z.[# j.[# j.[# j.[# j.[# j.[#

Temp, °C ::# VQ# VQ# T[# T[# T[# T[#

Table S9. X'6&/&9#)9+,(#/8>#8=<6,&#*?#0/&'/8)(#(5&,,8,>#`B*=8>#Vd""c.^##@%,#)*)/1#8=<6,&#*?#0/&'/8)(#

(5&,,8,>#'8#,/5%#/((/9#*0,&#)%,#5*=&(,#*?#)%,#,0*1=)'*8#'(#N'0,8#?*&#)%,#>'??,&,8)#1'6&/&9#)9+,(.#@%,#

%*<*1*N92#&/8>*<#<=)/N,8,('(2#&/)'*8/1#>,('N8#/8>#('),#(/)=&/)'*8#<=)/N,8,('(#1'6&/&',(#R,&,#=(,>#)*#

N,8,&/),#>'0,&(')9.##@%,#H&*!DB#1'6&/&',(#R,&,#=(,>#)*#('?)#)%&*=N%#,O'()'8N#>'0,&(')9#/(#R,11#/(#N,8,&/),#

>'0,&(')9#'8#)%,#?*&<#*?#&/8>*<#<=)/)'*8(.#

(7)

Protease nomenclature

Naming of the binding site of proteases according to Schechter and Berger (1967). The acyl part of the amide link to be cleaved lies in the S1, S2, S3, etc. binding sites, while the amino part of the amide link to be cleaved lies in the S1', S2', etc binding sites. The substrate residues are called P1, P2, P3, etc, and P1', P2', etc. according to their location relative to the amide link being cleaved.

H₂N CHC CH₂

OEt

P1 O P1’

(8)

Trypsin to chymotrypsin

• Trypsin and chymotrypsin have the same

protein fold and same catalytic machinery, but different specificity

• Trypsin favors Lys, Arg while chymotrypsin favors Phe, Tyr & Trp. This specificity can be

rationalized by the different residues in the S1 pocket.

• Can you switch the specificity? Not easily.

Replace fifteen residues in the S1 site and two surface loops: specificity switched, but rate

<1% of the rate for chymotrypsin.

J. J. Perona et al. (1995) Structural origins of substrate discrimination in trypsin and chymotrypsin, Biochemistry, 34, 1489.

H₂N CHC CH₂

OEt O

(9)

Outline - Screening•