DIRECTED EVOLUTION BY in vitro

RECOMBINATION

INTRODUCTION:

Two distinct ways to engineer proteins have appeared during the last few decades. These

include a rational protein design and a directed evolution. Here emphasis will only be on directed

evolution by in vitro recombination. This technique has been used to successfully to improve or

alter properties of many proteins (biocatalysts).

Directed evolution appeared as a conceptually similar experimental enzyme evolution in 1970's.

During the last decade, directed evolution became a key technology of molecular enzyme

engineering. Directed evolution on its own, i.e., not combined with the structural data, is more or

less accidental process similarly to natural evolution. During this process, successful and well

adapted mutants are selected for further rounds of improvement (iterative cycle). It passes through

the way of natural evolution and speeds up some slow steps that would normally proceed for

hundreds or thousands of years.

Contribution of directed evolution is substantial especially in particular cases, when

neither the three-dimensional (3-D) structures nor the catalytic mechanisms of the enzymes are

known. In comparison with rational design approach that exploits computer modeling techniques

and site-directed mutagenesis, directed evolution offers certain advantages. While rational design

emphasizes the understanding of protein structure and amino acids interactions at the beginning of

the process, directed evolution do not rely on such input data.

Figure: Comparison between rational design and directed evolution

1

Genetic recombination constructs libraries of hybrids by recombining fragments from two

or more parents, with the goal of discovering hybrids with beneficial properties such as improved

Thermostability

Activity

Substrate specificity 

Drug resistance

Stability binding affinity

Improved folding and solubility

New catalytic activity

Prerequisites for Directed Evolution:

For directed evolution we need:

1. Gene encoding protein(s) of interest

2. Effective method to create mutant library

3. Suitable expression (usually microbial) system

4. Suitable screening and selection system

Gene Diversification by in vitro Recombination:

Directed evolution by in vitro recombination can be done by:

1. DNA shuffling

2. Random priming (RPR)

3. Staggered extension process (StEP)

4. Random chimeragenesis on transient templates (RACHITT)

5. Incremental truncation hybrid (ITCHY)

6. SCRATCHY

7. Sequence Homology-Independent Protein Recombination (SHIPREC)

8. Recombined Extension on Truncated Templates (RETT)

9. Degenerate Oligonucleotides Gene Shuffling (DOGS)

DNA shuffling:

The method was first invented by W.P.C Stemmer .There are a number of techniques of in

vitro recombination but the main theme or principle is same which is based on simple DNA

shuffling. In this technique two or more homologous genes are taken and they are fragmented by

DNase I then these fragments are used to carry out a PCR like process and no external primers are

2

added these fragments themselves serve as primers. Then finally full-length gene is amplified.(1).

Following figure shows the simple DNA shuffling.

Figure: DNA shuffling

Random Priming (RPR):

It is a Simple and an efficient method. Random primers such as random hexamers are

annealed to the template DNAs and then extended by a DNA polymerase at or below room

temperature. The resulting DNA fragments are subsequently assembled into full-length genes by

repeated thermocycling in the presence of a thermostable DNA polymerase. (2).

It involves

Priming template with random-sequence primers .

Extension to generate a pool of short DNA fragments.

The fragments are reassembled during cycles of denaturation, annealing and

further enzyme-catalyzed DNA polymerization to produce a library of full-length

sequences.

Screening or selecting the expressed gene products leads to new variants with improved

functions

3

Figure: Schematic of random priming in vitro recombination (RPR). For simplicity, only two DNA templates are shown. Random hexanucleotide primers are annealed to the templates and extended by Klenow fragment to yield a pool of different sized random extension products. After the removal of the oligonucleotides and the templates, the homologous fragments are reassembled into full-length chimerical genes in a PCR-like process. The full-length genes will be amplified by a standard PCR and subcloned into an appropriate vector.

Example

DNA shuffling of a family of over 20 human interferon-α (Hu-IFN- α genes was used to derive

variantswith increased antiviral and antiproliferation activities in murine cells. A clone with

135,000-fold improved specific activity over Hu-IFN-a2a was obtained in the first cycle of

shuffling. After a second cycle of selective shuffling, the most active clone was improved

285,000-fold relative to Hu-IFN-a2a and 185-fold relative to Hu-IFN-a1. Remarkably, the three

most active clones were more active than the native murine IFNαs.These chimeras are derived

from up to five parental genes but contained no random point mutations. These results

demonstrate that diverse cytokine gene families can be used as starting material to rapidly evolve

cytokines that are more active, or have superior selectivity profiles, than native cytokine genes.(3)4

Staggered Extension Process (StEP)

It was introduced by Arnold and Kuchner in 1997.Staggered extension process (StEP) is simpler

and less labor intensive than DNA shuffling and other PCR-based recombination techniques that

require fragmentation, isolation, and amplification steps. StEP recombination is based on cross

hybridization of growing gene fragments during polymerase-catalyzed primer extension.

Following denaturation, primers anneal and extend in a step whose brief duration and suboptimal

extension temperature limit primer extension. The partially extended primers randomly reanneal

to different parent sequences throughout the multiple cycles, thus creating novel recombinants.

(4). StEP recombination is based on template switching during polymerase catalyzed primer

extension.

This method uses full-length genes as templates for the synthesis of chimeric progeny

genes and does not involve fragmentation.

It consists of

priming denatured templates

followed by repeated cycles of denaturation and

extremely short annealing/extension steps

Recombinogenic events occur when the partially extended primers anneal randomly to different

templates based on sequence complementarily and extend further.Template removal is done by

passing the mixture through Microcon 100 filter. Due to template switching, most of the

polynucleotides contain sequence information from different parental sequences.

5

Figure: StEP recombination, illustrated for two gene templates. Only one primer and single strands from the two genes (open and solid blocks) are shown for simplicity. During priming, oligonucleotide primers anneal to the denatured templates. Short fragments are produced by brief polymerase-catalyzed primer extension that is interrupted by denaturation. During subsequent random annealing-abbreviated extension cycles, fragments randomly prime the templates (template switching) and extend further, eventually producing full-length chimeric genes. The recombinant full-length gene products can be amplified in a standard PCR

Example

StEP was used to convert Bacillus subtilis subtilisin E (serine protease) into an enzyme

functionally equivalent to its thermophilic homolog thermitase from Thermoactinomyces vulgaris.

Five generations of random mutagenesis, recombination and screening created subtilisin E 5-3H5,

whose half-life at 83°C is 3.5 min and Topt is 76°C,identical with those of thermitase. The Topt

of the evolved enzyme is 17°C higher and its halflife at is 200 times greater than that of wild-type

subtilisin E(5).

6

Random Chimeragenesis on Transient Templates (RACHITT):

Chimeragenesis on Transient Templates (RACHITT) has been used to create libraries

averaging 12 or even 19 crossovers per gene in a single round of gene family shuffling. The

heteroduplexed top strand fragments are stabilized on the template by a single, long annealing

step, taking advantage of full length binding by each fragment, rather than the binding of smaller

overlaps, and by carrying out reactions at relatively high ionic strength. Fragments containing

unannealed 5' or 3'-termini are incorporated after flap trimming using the endo and exo

nucleolytic activities of Taq DNA polymerase and Pfu polymerase, respectively. After gap filling

and ligation, the template, which was synthesized with uracils in place of thymidine, is rendered

non-amplifiable by uracil-DNA glycosylase (UDG) treatment. Other methods of DNA shuffling

by gene fragmentation and reassembly can result in reconstitution of one or all of the parental

genes at unacceptably high frequencies in the final shuffled library(6–8).

Figure: RACHITT begins with production of a single stranded bottom strand “Transient Template” containing uracil and production of single stranded top strand “Donor Fragments.” The fragments are annealed to the template and joined to form a continuous chimeric top strand. Anchor oligonucleotide (Anc) protects the template 5'-terminus from the nucleases used to trim unannealed fragment flaps. The template is then degraded and the chimeric top strand amplified and cloned to result in a gene family shuffled library.

7

Example

A library of variant enzymes was created by combined shuffling of the DNA encoding the

human Mu class glutathione transferases GST M1-1and GST M2-2. The parental GSTs are 84%

sequence identical at the protein level, but their specific activities with the substrates

aminochromeand 2-cyano-1, 3-dimethyl-1-nitrosoguanidine (cyanoDMNG) differ by more than

100-fold. Aminochrome is of particular interest as an oxidationproduct of dopamine and of

possible significance in the etiology of Parkinson’s disease, and cyanoDMNG is a model for

genotoxic and potentially carcinogenic nitroso compounds. GST M2-2 has at least two ordersof

magnitude higher catalytic activity with both of the substrates thanany of the other known GSTs,

including GST M1-1.Variant GST sequences were expressed in E. coli, and their enzymatic

activities with aminochrome, cyanoDMNG, and 1-chloro-2, 4-dinitrobenzene (CDNB) were

determined in bacterial lysates. Such screening of more than 70 clones demonstrated a continuous

range of activities covering at least two orders of magnitude for each of the substrates. For a given

clone, the activities with aminochrome and cyanoDMNG, in spite of their different chemistries,

were clearly correlated, whereas no strong correlation was found with CDNB. This functional

correlation suggests a common structural basis for the enzymatic mechanisms for conjugation of

aminochrome and denitrosation of cyanoDMNG. (9).

Incremental Truncation Hybrid (ITCHY)

This method was introduced by Ostermeirer in 1999.The template here is a double stranded linear DNA fragment containing 2 or more linked genes.This method results in the production of hybrids or chimeric genes by use of exo-nuclease III and random incorporation of αS-dNTPs. Truncation of the targeted DNA can be achieved by two ways:

Exo III can be used to digest the DNA fragment first to produce 5` overhangs and then extending in presence of αS-dNTPs, also called as thio-itchy primer extension.

Alternatively, αS-dNTPs can be incorporated during PCR amplification of the entire plasmid, also called as thio-itchy PCR amplification.

o The whole plasmid is amplified in the presence of αS-dNTPs by Klenow

fragment.

o Then allowed to undergo Exo III cleavage. The phosphothioate

internucleotide linkages are resistant to 3′→5′ exonuclease hydrolysis, rendering the target DNA resistant to degradation in an exonuclease III treatment.

ITCHY does not rely on the parental genes containing regions of DNA sequence homology to create crossovers. Fusion of the truncated gene fragments by blunt end ligation then generates the ITCHY library.

8

Figure: Schematic overview of ITCHY using both thio itchy and primer extension.

SCRATCHY

SCRATCHY is a combination of the incremental truncation for the creation of hybrid

enzymes (ITCHY) technology and DNA shuffling. It generates combinatorial libraries of hybrid

proteins consisting of multiple fragments from two or more parental DNA sequences with no

restriction to DNA sequence identity between the original sequences. The experimental

implementation of SCRATCHY consists of two successive steps, an initial creation of an ITCHY

library, followed by a homologous recombination procedure such as DNA shuffling.

9

(a)

(b2)(b1)

(b3)(c)

(c)

(d)

(e)

(d)

Figure:The method requires two complementary vectors, carrying the genes A and B in alternating order. Following the generation of the ITCHY library, the linearized hybrid genes are selected for parental-size hybrid DNA constructs. After subcloning these DNA fragments into the NdeI and SpeI sites of pSALect, sequences with the correct reading frame result in the expression of a trifunctional fusion protein which renders the host cells resistant to ampicillin. The plasmid DNA from colonies grown under these conditions is recovered and can be used as starting material for DNA shuffling.

Sequence Homology-Independent Protein Recombination (SHIPREC)

SHIPREC results in a library of chimeras in which the hybrids generatedretain proper

sequence alignment with the parents. When parental genes arechosen that encode homologous

proteins, this type of recombination can producea library of chimeric proteins in which the

crossovers occur at structurallyrelated sites. The startingmaterial is a fusion of the two genes of

interest, with the C-terminus of thefirst gene and the N-terminus of the second gene joined

through a small linkercontaining a unique restriction site (e.g., PstI). The gene fusion is then

randomly truncated using DNase I and S1 nuclease, creating a library of fusions

10

ofvarying sizefrom this library, DNA corresponding to the length of the

parentalgenes is isolated and subsequently circularized. The size selection ensures that the

circularization produces chimeras which retain proper sequence alignment with the parental

genes. The resulting chimeras are linearized by cleaving the unique restriction site in the linker

(i.e., PstI), and the library is cloned into an appropriate vector for screening or selection. (13).

Figure:SHIPREC overview. A gene fusion comprised of the two parental genes connected by a unique restriction site is constructed. (1) This fusion is randomly fragmented using DNase I and S1 nuclease to produce a library of gene fusions exhibiting varying length and containing blunt ends. (2) Gene fusions corresponding to the length of the parental genes are isolated using gel electrophoresis and separated from the random digest pool. (3) Single-gene length fragments are circularized by intramolecular blunt-end ligation. (4) Circular DNA is linearized by treatment with a restriction endonuclease that cuts in the linker that separates 3' and 5'ends of the original parental genes. This yields a library of chimeric genes that encode for proteins with a N-terminal region originating from Parent B and a Cterminal region originating from Parent A. (5) The chimeras are amplified and cloned into an appropriate vector for screening/selection.

Example

Sequence homology–independent protein recombination (SHIPREC) is used to create

libraries of single-crossover hybrids of unrelated or distantly related proteins. The method

maintains the proper sequence alignment between the parents and introduces crossovers mainly at

11

structurally related sites distributed over the aligned sequences. SHIPREC is used to create a

library of interspecies hybrids of a membrane-associated human cytochrome P450 (1A2) and the

heme domain of a soluble bacterial P450 (BM3). By fusing the hybrid gene library to the gene for

chloramphenicol acetyl transferase (CAT), researchers were able to select for soluble and properly

folded protein variants. Screening for 1A2 activity (diethylation of 7-ethoxyresorufin) identified

two functional P450 hybrids that were more soluble in the bacterial cytoplasm than the wild-type

1A2 enzyme.(13)

Recombined Extension on Truncated Templates (RETT)

Its starting material is RNA. Fragments can be generated by using random primers for

reverse transcription or by unidirectional serial truncation of cDNA with exonuclease III. A

specific primer is afterwards annealed to complementaryssDNA fragmentation and extended by

PCR. Short fragments extended fromthis specific primer (like in StEP) are annealed to another

ssDNA fragmentand thus switch templates. This extension is repeated until full length genesare

generated, which then is used to generate dsDNA by PCR. (14).

In spite of the great importance of in vitro recombination techniques in directed gene

evolution, the current techniques have still some drawbacks to be overcome for more

efficient generation of gene library

DNA fragmentation process in DNA shuffling and RACHITT method may not be

random because DNase I hydrolyzes DNA preferentially at sites adjacent to

pyrimidine nucleotides, which introduces a sequence bias into recombination library

For the recombination of genes with low or no homology, ITCHY and SHIPREC were

developed.

These methods have limitations that only two parental genes can be recombined and

the created hybrids are limited to one crossover.

RETT does not use DNA endonucleases for generation of shuffling blocks

In RETT, unidirectional single-stranded DNA (ssDNA) fragments are created by either

DNA polymerase in the presence of random primers or serial deletion with

exonuclease

These unidirectional ssDNA fragments only act as templates in PCR, not as primers

RETT generates random recombinant gene library by template-switching of

unidirectionally growing polynucleotides from primers in the presence of

unidirectional ssDNA fragments pool used as templates

12

According to truncation pattern of unidirectional ssDNA fragments, two methods are

described separately.

(a) Two homologous genes are presented to simplify the model. Unidirectional ssDNA

fragments are prepared by reverse transcription using in vitro-transcribed target RNA as

template in the presence of random primers.

Recombinational synthesis reaction is conducted as follow:

(i) Specific primer is annealed to ssDNA fragment.

(ii) Specific primer is extended during one cycle of PCR.

13

(iii, iv) Short fragments extended from primer are annealed to other ssDNA fragment by

template-switching and extended during another cycle of PCR.

(v) Steps are repeated until full-length ssDNA genes are generated.

(b) Unidirectional ssDNA fragments are prepared by serial deletion with exonuclease III.

Recombinational synthesis reaction is conducted by steps (i–v).

Degenerate Oligonucleotides Gene Shuffling (DOGS)

A method for enhancing the frequency of recombination with family shuffling

It was designed to decrease the amount of parental DNA reassembled from shuffling

procedures. For DOGS complementary degenerate primers are designed for conserved motives

found in the candidate genes.Each of these segments is flanked by primers and individually

amplified.For the reassembly procedure the library of fragments can be put togetherat different

ratios generating many biased libraries containing no parentalgenes. (15).

14

Figure:complementary degenerate primers are designed for conserved motives found in the candidate genes. Each of these segments is flanked by primers and individually amplified. For the reassembly procedure the library of fragments can be put together at different ratios generating many biased libraries containing no parental genes.

Examples

Thermophilic β-xylanase was made more active. It is important in paper industry

for bleaching of paper pulp. It is obtained from Dictyoglomusthermophilum strain.

(15).

Aharoni et al. recently reported the functional expression of a mammalian

paraoxonase (PON) enzyme in Escherichia coli. PONs have gained interest due to

their role in prevention of human disease. Through DNA family shuffling of four

wild type PON1 genes derived from human, mouse, rat, and rabbit sources, a library

of PON1 mutants was created and screened for esterase activity. Further analysis of

several PON1 mutants showed that improvement was largely due to increased

solubility rather than changes in kinetic parameters. (18)

15

REFERENCES

1. Stemmer, W. P. C. (1994). DNA shuffling by random fragmentation and reassembly: In vitro

recombination for molecular evolution. Proc. Natl. Acad. Sci. USA91: 10,747–10,751.

2. Shao, Z., Zhao, H., Giver, L., and Arnold, F. H. (1998). Random-priming in vitro recombination:

an effective tool for directed evolution. Nucleic Acids Res,26: 681–683.

3. Chia-Chun J. Chang, Teddy T. Chen, Brett W. Cox, Glenn N. Dawes, Willem P.C. Stemmer,

JuhaPunnonen, and Phillip A. Patten.(1999).Evolution of a cytokine using DNA familyshuffling.

Nature Biotechnology,17: 1-5

4. Zhao, H., Giver, L., Shao, Z., Affholter, J. A., and Arnold, F. H. (1998). Molecular evolution by

staggered extension process (StEP) in vitro recombination. Nat.Biotechnol, 16: 258–261.

5. Huimin Zhao and Frances H.Arnold. (1999). Directed evolution converts subtilisin E into a

functional equivalent of thermitase.Protein Engineering,12:47–53.

6. Kikuchi, M., Ohnishi, K., and Harayama, S. (1999). Novel family shuffling methods for the in

vitro evolution of enzymes. Gene,236: 159–167.

7. Moore, G. L., Maranas, C. D., Lutz, S., and Benkovic, S. J. (2001). Predicting cross

overgeneration in DNA shuffling. Proc. Natl. Acad. Sci. USA, 98: 3226–3231.

8. Joern, J. M., Meinhold, P., and Arnold, F. H. (2002). Analysis of shuffled gene libraries . J. Mol.

Biol,316: 643–656.

9. Lars O. Hansson, Robyn Bolton-Grob, TaherehMassoudandBengtMannervik. (1999). Evolution

of differential substrate specificities in Muclass glutathione transferases probed by DNA

shuffling.J. Mol. Biol,287: 265-276.

10. Ostermeier, M., Nixon, A. E., Shim, J. H., and Benkovic, S. J. (1999).Combinatorial protein

engineering by incremental truncation. Proc. Natl. Acad. Sci. USA,96: 3562–3567.

11. Lutz, S., Ostermeier, M., and Benkovic, S. J. (2001). Rapid generation of incremental truncation

libraries for protein engineering using α-phosphothioate nucleotides. Nucleic Acids Res,29: e16.

16

12. Putney, S. D., Benkovic, S. J., and Schimmel, P. R. (1981). A DNA fragment withan alpha-

phosphorothioate nucleotide at one end is asymmetrically blocked fromdigestion by exonuclease

III and can be replicated in vivo. Proc. Natl. Acad. Aci.USA,78: 7350–7354.

13. Sieber, V., Martinez, C.A. & Arnold, F.H. (2001). Libraries of hybrid proteins from distantly

related sequences. Nat. Biotechnol. 19: 456–460.

14. Lee SH, Ryu EJ, Kang MJ, Wang ES, Piao Z, Choi YJ, Jung KH, Jeon JYJ, Shin YC. (2003). A

new approach to directed gene evolution by recombined extension on truncated templates

(RETT). J Mol Catalysis B-Enzymatic, 26:119–129.

15. Gibbs MD, Nevalainen KM, Bergquist PL. (2001). Degenerate oligonucleotide gene shuffling

(DOGS): a method for enhancing the frequency of recombination with family shuffling.

Gene,271:13–20.

16. Gavin J. Williams, SilvieDomann, Adam Nelson, and Alan Berry.(2003).Modifying the

stereochemistry of an enzyme catalyzed reaction by directed evolution.PNAS, 100: 3143–3148.

17. Wen-Chen Suen, Ningyan Zhang, Li Xiao,Vincent Madison and Aleksey Zaks.(2004). Improved

activity and thermostability ofCandida antarctica lipase B by DNA shuffling.Protein

Engineering, Design & Selection,17:133-140.

18. Aharoni, A. Gaidukov, L. Yagur, S.; Toker, L. Silman, I. Tawfik, D.S. (2004).Directed evolution

of mammalian paraoxonasesPON1 and PON3 for bacterial expression andcatalytic specialization.

Proc. Natl. Acad. Sci. USA, 101: 482-487.

17