1
Synthetic protein biology II.
Directed evolution, combined approaches & screening
and selection technologies
Modifying protein structure and function for biotechnology,
biomedicine and basic research
Dr. Martin Marek
Loschmidt Laboratories
Faculty of Science, MUNI
Kamenice 5, bld. A13, room 332
martin.marek@recetox.muni.cz
2
Short recapitulation of previous lesson
• What is protein engineering? Concept, strategies etc…
• Why protein engineering is important for synthetic biology?
• How rational protein design works? Workflow, pros and cons
• Computational de novo protein design…?
• Computational design of new enzyme…?
• What to do after the design is finished…?
3
What will we talk about
• Introduction to protein engineering and design
• definition, goals and applications
• Rational protein design (knowledge-based strategies)
• concepts, methodology, limitations, success stories
• Directed evolution (lab-based brute force engineering)
• strategies, methodology, disadvantages, success stories
• Integrative (combined) approaches
• the best of both approaches, beneficial synergy, examples
• Selection and screening technologies
• classical versus emerging technologies, unmet challenges
4
The two major strategies of protein engineering
5
Directed evolution
Concepts
Methods
Applications
6
Expanding the synthetic protein universe by guided
evolutionary concepts
Poluri and Gulati: Protein Engineering Techniques,
Springer, pp 27-54 (2017)
• The genetic information of a cell is maintained by the sequence composition of
the DNA
• The changes in the nucleotide content may potentially alter its transcriptional and
translational events thus influencing the properties of the newly synthesized proteins
• The nature’s alterations can be helpful in the evolution of proteins with novel or improved
functionalities
• Unravelling the principles of such molecular evolutionary process is resourceful to
implement it for the benefit of the mankind through
laboratory techniques
• The laboratory process of synthesizing new proteins in
a constructive way through evolutionary guided
principles is called “directed evolution“
7
What it is directed evolution and
why did it win the Nobel prize
Prof. Frances Arnold herself summed it up nicely:
In directed evolution we provide a new niche in the laboratory, so to speak, and
encourage evolution of enzymes to catalyse commercially useful reactions.
Enzymes can speed up reactions dramatically while carrying out their work in
water at room temperature. They are also really good at making one specific bond
– without messing around with other functional groups – and are often
enantioselective too.
So it’s easy to understand why chemists like using enzymes to catalyse reactions.
However, many bonds chemists are interested in aren’t made by any natural
enzyme. This is simply because organisms have never needed to evolve the
ability to make, for example, carbon-silicon bonds.
https://www.chemistryworld.com/news/what-is-directed-evolution-and-why-did-it-win-the-chemistry-nobel-prize/3009584.article
8
Directed evolution workflow
https://www.kapabiosystems.com/technology/overview
• Directed evolution (DE) is a method of protein engineering that simulates
natural selection in the lab
• The process starts with a gene coding
for a “wild-type”, or unmodified, protein
of interest
• Random variation is introduced into
the gene through a process of
mutagenesis, generating a library
of millions of genes each coding for
a unique protein variant
• A functional selection pressure is then
applied to the library, and only the genes
that coded for the highest performing
proteins “survive”
• This process of random mutation
and selection is repeated until the
desired enzyme function evolves
9
Directed evolution is a mimic of the natural evolution
cycle in a laboratory setting
Directed evolution is analogous to climbing a hill on a „fitness landscape“ where
elevation represents the desired property. Each round of selection samples
mutants on all sides of the starting template (1) and selects the mutant with the
highest elevation, thereby climbing the hill. This is repeated until a local peak is
reached (2).
https://en.wikipedia.org/wiki/Directed_evolution
10
Advantages of directed evolution
• Rational design of a protein relies on an in-depth knowledge of the protein
structure, as well as its mode of action (e.g. catalytic mechanism)
• Specific changes are then made by site-directed mutagenesis in an attempt to
change the function of the protein
• A drawback of this is that even when the structure and mechanism of action of
the protein are well known, the change due to mutation is still difficult to predict
• Therefore, an advantage of directed evolution is that there is no need to
understand the mechanism of desired activity or how mutations would affect it
11
Methods for gene diversification and library generation
12
Error-prone PCR (epPCR)
• The error rate of Taq DNA polymerase is 0.001-0.002 % per nucleotide per
replication cycle under standard conditions which is sufficient to create mutant
libraries of large genes but not for small genes
• Error-prone PCR (epPCR) takes advantage of the inherently low fidelity of Taq
DNA Polymerase, which may be further decreased by the addition of Mn2+,
increasing the Mg2+ concentration, and using unequal dNTP concentrations.
• The rate of mutagenesis achieved by error-prone PCR is in the range
of 0.6-2.0 %
https://lifescience.canvaxbiotech.com/product/pickmutant-error-prone-pcr-kit/
13
Mutator strains
• Mutator strains of E. coli are deficient in one or more of DNA repair genes,
leading to single base substitutions at a rate of approximately 1 mutation per
1000 base pairs
• Generation of mutant libraries
• Process is simple
14
Insertion and deletion (InDel) mutagenesis
• Gain or lost of one or more nucleotides produces frameshift mutations
(triplet reading frame)
• Triplet InDel mutagenesis may trigger
protein backbone changes essential
for evolvability
• Insertion and deletion mutations can
enhance proteins through structural
rearrangements not possible by
substitution mutations alone
Arpino et al., Structure 22: 889-898 (2014)
Using directed evolution, green fluorescent protein
(GFP) was observed to tolerate residue deletions,
particularly within short and long loops, helical
elements, and at the termini of strands. A variant
with G4 removed from a helix (EGFPG4Δ) conferred
significantly higher cellular fluorescence.
15
Site saturation mutagenesis (SSM)
• Site saturation mutagenesis is used to substitute targeted residues to any other naturally
occurring amino acid
• The core of a SSM experiment lies in the codon degeneracy or randomness. A completely
randomized codon (NNN, where N=A, C, G or T) results in a library size of 64 different
sequences encoding all 20 amino acids and 3 stop codons
• When an experiment targets multiple codons, the library size can be considerably higher,
making it difficult to perform a complete screening (e.g. targeting three NNN codons has
262,144 unique codon configurations
16
Combinatorial active-site saturation test (CAST)
• The Combinatorial Active-site Saturation Test (CAST) was developed to
increase the enantioselectivity and/or the substrate specificity of
enzymes
• The basis of the method is the generation
of small libraries of mutant enzymes
that are easy to screen for activity
• The mutants are produced by
simultaneous randomization of sets
of two or three spatially close amino
acids, whose side chains form part
of the substrate-binding pocket
17
DNA shuffling
• DNA shuffling is a method for in vitro
recombination of homologous genes
• The genes to be recombined are randomly
fragmented by DNaseI, and fragments of the
desired size are purified from an agarose gel
• These fragments are then reassembled using
cycles of denaturation, annealing, and
extension by a polymerase
• Recombination occurs when fragments from
different parental templates anneal at a region
of high sequence identity
• Following this reassembly reaction, PCR
amplification with primers is used to generate
full-length chimeras suitable for cloning into
an expression vector
• Moving from DNA shuffling to whole genome
shuffling is known as GENOME SHUFFLING
18
Random chimeragenesis on transient templates
(RACHITT)
• RACHITT creates chimeric genes by
aligning parental gene “donor” fragments on
a full-length DNA template
• The heteroduplexed top strand fragments
are stabilized on the template by a single,
long annealing step
• Fragments containing unannealed 5′ or 3′termini
are incorporated after flap trimming
using the endo and exonucleolytic 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
19
Incremental truncation for the creation of hybrid (ITCHY)
• ITCHY is a directed evolution technique for
randomly recombining two genes
• The chief advantage of ITCHY is that there
is no requirement for the two genes to
share any sequence similarity
• This distinguishes ITCHY from directed
evolution methods that are based on
homologous recombination, such as DNA
shuffling
• In ITCHY, Escherichia coli exonuclease III
is used to incrementally truncate one of the
parental genes from its 3' end and the other
from its 5' end
• Ligation of the randomly truncated gene
fragments yields a combinatorial library of
chimeras
20
Combining ITCHY and DNA shuffling
1. ITCHY
2. DNA shuffling
21
Continuous directed evolution strategies
• Continuous evolution systems couple mutagenesis and selection
• The optimal continuous evolution system is performed in a turbidostat which
consists of a medium supply container, fermentation tank, and a waste
collection container
• Selection pressures can be imposed in the fermentation tank to enable
growth-based enrichment of improved mutants
Zhou and Alper, Journal of Chemical Technology & Biotechnology, 94:366-376 (2019)
22
Phage-assisted continuous evolution (PACE)
Packer et al., Nature Communications, 8: 956 (2017)
• The PACE is a form of directed evolution, a laboratory process that uses the phage virus
bacterial infection cycle to generate multiple rounds of DNA sequence changes and
selection for DNA changes in a target gene that result in a desired structure or activity in
the encoded protein
23
Evolution of protein functions with PACE: examples
• Bt toxin evolution to targeting new receptor
(a).
• TEV protease evolution to target new
cleavage site (b).
• aaRS evolution to get specificity with ncAA
and suppressor tRNA (c).
• Eukaryotic protein evolution to improve
solubility in E. coli (d).
• Cas9 evolution to recognize new PAM (e)
24
The CRISPR/Cas9 system
• The Cas9 (CRISPR associated protein 9) is a protein which plays a vital role in the
immunological defense of bacteria against DNA viruses, and which is used in genetic
engineering. Its main function is to cut DNA and therefore it can alter a cell's genome
• Structurally, Cas9 is an RNA-guided DNA endonuclease enzyme associated with
CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) adaptive immunity
system in Streptococcus pyogenes
• Cas9 performs this by unwinding foreign
DNA and checking for sites complementary
to the 20 bp spacer region of the guide RNA
• If the DNA substrate is complementary
to the guide RNA, Cas9 cleaves
the invading DNA
Crystal structure of S. pyogenes Cas9 in complex
with sgRNA and its target DNA at 2.5 Å resolution,
Nishimazu et al., Cell 156: 935–49 (2014)
25
The CRISPR/Cas9 system: key elements
• Cas9 nuclease specifically cleaves double-stranded
DNA activating double-strand break repair
machinery
• In the absence of a homologous repair template
non-homologous end joining can result in indels
disrupting the target sequence
• Alternatively, precise mutations and knock-ins can
be made by providing a homologous repair template
and exploiting the homology directed repair pathway
26
CRISPR/Cas9-directed evolution (CDE) in plants
27
Advantages of CRISPR/Cas9-mediated mutagenesis
• The CRISPR/Cas9 system requires only the redesign of the crRNA to change
target specificity
• This contrasts with other genome editing tools, including zinc finger and
TALENs, where redesign of the protein-DNA interface is required
• Furthermore, CRISPR/Cas9 enables rapid genome-wide interrogation of
gene function by generating large gRNA libraries for genomic screening
Zinc-finger nucleases TALENs Cas9
28
Multiplex Automated Genome Engineering (MAGE)
https://wyss.harvard.edu/technology/multiplex-automated-genomic-engineering-mage/
• MAGE is a technique capable of editing the genome by making small changes in
existing genomic sequences
• These properties make MAGE a highly useful tool for synthetic biology, allowing
researchers to easily modify the bacterial genome and generate diversity within a
population
29
A
How MAGE works
http://2011.igem.org/Team:Harvard/Technology/MAGE
• The MAGE is inserting single-stranded oligos that contain
the desired mutations into the cell, and more than one gene
can be targeted at a time simply by using multiple oligos
• Mediated by λ-Red ssDNA-binding protein β, the oligos
are incorporated into the lagging strand of the replication fork
during DNA replication, creating a new allele
• The efficiency of oligo incorporation depends on several
factors (mutS), but the frequency of the allele can be
increased by performing multiple MEGA rounds on the same cell culture
• The process can be automated allowing to
run many rounds of MAGE in succession
B
30
Basic design and cycling of MAGE
• The MAGE starts with growing the
cells at 30 °C until cell density reaches
the mid-log phase and lambda (λ) red
proteins are expressed under the
control of pL promoter which is
regulated by temperature sensitive CI
• Then cells are moved to 42 °C for
15 min for the heat shock induction of
λ-red proteins.
• Cells are moved to 4 °C to repress the
λ-red and prevent degradation. Cells
were subsequently washed and
resuspended in chilled distilled water
• Single stranded oligos were introduced
into cells via electroporation and
incorporated into the lagging strand of
the replication fork during DNA
replication
• The cells were kept at 30 °C for 2–3 h
for recovery of generated sequences
diversity before proceeding into the
next MAGE cycle
31
• Enzymes that catalyze carbon–silicon bond formation are unknown in nature, despite the
natural abundance of both elements. Such enzymes would expand the catalytic repertoire
of biology, enabling living systems to access chemical space previously only open to
synthetic chemistry
• Discovery that heme proteins catalyze the formation of organosilicon compounds under
physiological conditions via carbene insertion into silicon–hydrogen bonds. The reaction
proceeds both in vitro and in vivo, accommodating a broad range of substrates with high
chemo- and enantioselectivity
• Using directed evolution the catalytic function of cytochrome c from Rhodothermus marinus
achieved more than 15-fold higher turnover than state-of-the-art synthetic catalysts. This
carbon–silicon bond-forming biocatalyst offers an environmentally friendly and highly
efficient route to producing enantiopure organosilicon molecules
DE of cytochrome c for carbon–silicon bond formation
An example of directed evolution I.
Kan et al., Science, 354: 1048-1051 (2016)
32
Screening and selection technologies
Concepts
Methods
Applications
33
High-throughput screening is important part of DE cycle
1. Random mutations are introduced into a target gene (1), e.g. error-prone PCR
2. The mutated genes are transferred into a suitable host organism and expressed,
thereby resulting in a large library of protein or enzyme variants
3. Improved variants are then identified either by high-throughput screening or by
selection for a desired property
4. The gene(s) encoding such improved
variant(s) are isolated and used for
another cycle of directed evolution
34
The size of sequence space
• Important concept when considering a protein's amino acid sequence is that of
(its) sequence space, i.e. the number of variations of that sequence that can
possibly exist
• Straightforwardly, for a protein that contains just the 20 main natural amino
acids, a sequence length of N residues has a total number of possible
sequences of 20N. For N = 100 (a rather small protein) the number 20100 is
already far greater than the number of atoms in the known universe
35
Principles of selection in directed evolution
• The goal of all selection (and
screening) platforms is to partition a
potentially large population (shown
in grey as the bulk diversity) by
function (phenotype) ensuring the
recovery of the genetic information
that accounts for that phenotype.
• Strong phenotype–genotype
linkages allow efficient isolation of
mutants with the desired function
(green).
• Breakdown of that linkage results in
false negatives (variants that have
the desired function but that are not
efficiently recovered–yellow) and
false positives (variants that are
recovered independently of the
desired function–blue), which are
integral aspects of all selection
strategies.Tizei et al., Biochem. Soc. Trans. 44: 1165-1175 (2016)
36
In vivo and ex vivo directed evolution
• Both strategies use the cell (or virus
particle) as the physical linkage between
genotype and phenotype through the
directed evolution process.
• Ex vivo platforms tend to focus diversity (a)
on to a single target gene, whereas in vivo
platforms can extend that to metabolic
pathways or even whole genomes.
• Once generated, the diverse repertoires are
partitioned (b) with active (blue) variants
preferentially recovered over inactive
variants (orange).
• Partition by phenotype is linked to genotype
recovery and amplification (c) which can
take place in a single step if cells are still
viable (as is the norm for in vivo
methodologies). Alternatively, as shown for
the ex vivo selection (light green boxes),
genotype recovery and amplification can be
separated, introducing different limitations
to the process. The amplified recovered
genotypes are the starting point of a
subsequent round of selection.
Tizei et al., Biochem. Soc. Trans. 44: 1165-1175 (2016)
37
In vitro selection
• Platforms for in vitro selection can be
broadly divided by the available
redundancy of phenotype and
genotype linkages.
• In a number of selection strategies,
the link is unique – a lone genotype
molecule is linked to a lone molecule
that may have the phenotype being
selected (a).
• Compartmentalization strategies
enable redundancy in the system with
one-to-many [redundant genotype to
lone phenotype (b) or lone genotype
to pooled phenotype (not shown)] and
many-to-many [redundant genotype to
pooled phenotype (c)] mappings
between phenotype and genotype
available.
Tizei et al., Biochem. Soc. Trans. 44: 1165-1175 (2016)
38
Overview of screening and selection strategies in DE cycle
• Clonally isolated variants can be screened as colonies
on solid media or as wells in liquid culture. Fluorescent
or colorimetric reporters are measured by automated
microtitre plate readers, chromatography, MS or NMR
• Fluorescence-activated cell sorting (FACS) enables the
fluorescence measurement of individual cells and the
separation of distinct subpopulations by electrostatic
deflection
• Yeast display techniques enable FACS screens of
protein–protein interactions, bond formation and
peptide bond cleavage
• In vitro compartments entrap DNA, translated proteins
and fluorogenic substrates, allowing the fluorescenceactivated
sorting of functional variants
39
Screening and selection methods
• Virus display methods
phage display, baculovirus display
• Cell surface display systems
bacteria, insect and mammalian
cells
• Cell-free display systems
mRNA display, ribosome display,
covalent and non-covalent display,
in vitro compartmentalization
40
Phage display as the tool for directed evolution
• Since its introduction in 1985, phage display
has had a tremendous impact on the
discovery of peptides that bind to a variety of
receptors, the generation of binding sites
within predefined scaffolds, and the creation
of high-affinity antibodies without
immunization
• Its application to enzymology has required
the development of techniques that couple
enzymatic activity to selection protocols
based on affinity chromatography
• The protein of interest is fused to the Nterminal
of g3p. The foreign gene is inserted
between the sequences encoding the signal
peptide and the mature coat protein g3p
• On phage morphogenesis, the fusion protein
is displayed on the phage surface
41
Baculovirus display system
• Baculovirus is a large DNA insect-infecting
virus
• Baculovirus surface glycoprotein Gp64 is
expressed early and late in the infection of
an insect cell
• It is a 64 kDa protein which forms trimers
and locates in the BV envelope with a
polarized distribution
• As Gp64 is a transmembrane protein that
exposes an outer domain, it can be used to
display a selected protein on the BV surface
• A chimeric Gp64 can be constructed to
contain the protein of interest allowing it to
be incorporated in the BV structure upon
infection of insect cells
42
Cell surface display systems
• Cell surface display (bacteria, yeasts,
insect or mammalian cells) is a strategy
used for in vitro protein evolution
• Libraries of proteins displayed on the
surface of cells can be screened using
flow cytometry
• This technique allows us to link the
function of a protein with the gene that
encodes it
• Cell surface display can be used to find
target proteins with desired properties,
e.g. to make high-affinity ligands,
creation of novel vaccines, identification
of enzyme substrates
Target protein
Linker
Cell membrane
protein
43
mRNA display
Puromycin
• mRNA display is a display technique used for in
vitro protein evolution
• The process results in translated proteins that are
linked with their mRNA via a puromycin linkage
• Puromycin is an analogue of the 3’ end of a
tyrosyl-tRNA with a part of its structure mimics a
molecule of and the other part mimics a molecule of
tyrosine
• Compared to the cleavable ester bond in a tyrosyltRNA,
puromycin has a non-hydrolysable amide
bond. As a result, puromycin interferes with
translation, and causes premature release of
translation products
• The complex then binds to an immobilized target in
a selection step, and mRNA-protein fusions that
bind well are then reverse transcribed to cDNA and
their sequence amplified via a PCR
• The result is a nucleotide sequence that encodes a
peptide with high affinity for the molecule of interest
44
mRNA display workflow
45
Ribosome display
• Ribosome display is an in vitro evolution technology for proteins
• It is based on in vitro translation, but prevents the newly synthesized protein and the mRNA
encoding it from leaving the ribosome
• It thereby couples phenotype and genotype
• Since no cells need to be transformed,
very large libraries can be used directly
in selections, and the in vitro amplification
provides a very convenient integration of
random mutagenesis that can be
incorporated into the procedure.
46
Ribosome display: step by step
• Ribosome display starts with a DNA library
• Each sequence is transcribed, and then translated in vitro into polypeptide
• However, the DNA library coding for a particular library of binding proteins is genetically
fused to a spacer sequence lacking a stop codon before its end
• The lack of a stop codon prevents release factors from binding and triggering the
disassembly of the translational complex
• It results in a complex
of mRNA, ribosome,
and protein which can
bind to surface-bound
ligand or substrate
47
Ribosome display vs. mRNA display
48
Semi-rational protein design
Concepts
Methods
Applications
49
Combining rational design and directed evolution
• Combined, 'semi-rational' approaches are being investigated to address the
limitations of both rational design and directed evolution
• Beneficial mutations are rare, so large numbers of random mutants have to be
screened to find improved variants
• 'Focused libraries' concentrate on randomising regions thought to be richer in
beneficial mutations for the mutagenesis step of DE. A focused library contains
fewer variants than a traditional random mutagenesis library and so does not
require such high-throughput screening
• Creating a focused library requires some knowledge of which residues in the
structure to mutate. For example, knowledge of the active site of an enzyme may
allow just the residues known to interact with the substrate to be randomised.
• Alternatively, knowledge of which protein regions are variable in nature can guide
mutagenesis in just those regions.
50
Combining protein engineering approaches
51
Discovery of PET-breaking enzyme (PETase)
• PET plastic, short for polyethylene terephthalate, is
the fourth most-produced plastic, used to make
things such as beverage bottles and carpets, most
of which are not being recycled
• PETase is an esterase class of enzymes that
catalyze the hydrolysis of polyethylene terephtalate
(PET) plastic to monomeric mono-2-hydroxyethyl
terephtalate (MHET)
Han et al., Nature Communications, 8: 2106 (2017)
52
Engineering of PET-breaking enzyme (PETase)
The S238F mutation provides new π-stacking and hydrophobic interactions to adjacent terephthalate moieties, while
the conversion to His159 from the bulkier Trp allows the PET polymer to sit deeper within the active-site channel.
53
CARBIOS Chief Scientific Officer, Prof. Alain Marty will explain the development of novel
enzymes allowing PET plastics such as bottles, packaging and textiles to be recycled in an
eco-responsible way. Drawing on the company’s proprietary enzyme engineering
technologies and the extensive experience of its team and partners, CARBIOS has
developed a breakthrough solution to infinitely recycle all kind of PET plastics and polyester
fibers.
54
An example of coupling design and experiment
• a) The Kemp elimination proceeds by means of a single transition state, which can be
stabilized by a base deprotonating the carbon and the dispersion of the resulting negative
charge; a hydrogen bond donor can also be used to stabilize the partial negative charge on
the phenolic oxygen.
• b) Examples of active site motifs highlighting the two choices for the catalytic base (a
carboxylate (left) or a His–Asp dyad (right)) used for deprotonation, and a π-stacking
aromatic residue for transition state stabilization. For each catalytic base, all combinations
of hydrogen bond donor groups (Lys, Arg, Ser, Tyr, His, water or none) and π-stacking
interactions (Phe, Tyr, Trp) were input as active site motifs into RosettaMatch.
Röthlisberger et al., Nature 453: 190-195 (2008)
The Kemp elimination: a model reaction for proton transfer from carbon atom
55
An example coupling design and experiment
1. Theozyme containing a carboxylate base, an
aromatic side-chain for substrate binding,
and a hydrogen bond donor to stabilize
developing negative charge in the transition
state was used as the template for design
2. The crystal structure of the first-generation
enzyme (blue) complexed with the transition
state analog shows good qualitative
agreement with the computational design
model (green). However, the ligand in the
crystal structure (orange) is flipped relative to
the designed orientation (pink)
3. The catalytic efficiency of the starting enzyme
(R1) relative to the nonenzymic acetatepromoted
reaction was increased by more
than two orders of magnitude over thirteen
rounds of mutagenesis and screening (R2–
R13)
4. The structure of the best evolved variant R13
(blue) cocrystallized with benzotriazole
(orange)
Khersonsky et al., PNAS 109:10358-10363 (2012)
Kries et al., Curr. Opin. in Chem. Biol. 17:221-8 (2013)
Generation of an improved Kemp eliminase with a kcat/KM of 570.000 M−1 s−1
56
Rational protein design vs. directed evolution
57
Questions
58
Dr. Martin Marek
Loschmidt Laboratories
Faculty of Science, MUNI
Kamenice 5, bld. A13, room 332
martin.marek@recetox.muni.cz
59
Supplementary materials
60
Directed evolution as a tool for synthetic biology
• Despite the diversity and versatility of selection platforms available, novel ones are regularly
being developed–delivering custom solutions to ever growing challenges. As molecular
biology methods and technologies develop, novel strategies to diversify and partition a
biopolymer population become available, increasing experimental control, throughput and
pace.
• Directed evolution performs the design, build and test cycle of synthetic biology on a scale
that is unnatural in engineering: it would be the equivalent of building millions (or even
trillions) of slightly different machines (e.g. watches) in search of a specific improvement
(e.g. more precise time keeping). On an engineering scale, such approach would be
prohibitive, if even possible. However, on a biological scale, millions are still small numbers,
barely able to cover the immediate sequence neighbourhood of even a small protein.
• Directed evolution has successfully been used to isolate novel and optimize existing
function on natural and synthetic biopolymers. But its key strength lies on how it deals with
uncertainty. Even in the absence of complete understanding of complex biological systems,
directed evolution is a powerful tool to re-engineer even the most central truths of life on our
planet–that life is based on DNA and RNA, and that life requires (or is optimal with only) 20
amino acids.
Tizei et al., Biochem. Soc. Trans. 44: 1165-1175 (2016)
61
The cycle of knowledge in directed evolution
• Both structure-based design and a more empirical data-driven approach can
contribute to the evolution of a protein with improved properties, in a series of
iterative cycles.
62
Basic elements of a mixed computational and
experimental programme in directed evolution
63
The CRISPR/Cas9 system on YouTube
https://www.youtube.com/watch?v=bXnWIk8FgKc
https://www.youtube.com/watch?v=OjNrbPMXyMA
https://www.youtube.com/watch?v=0dRT7slyGhs
https://www.youtube.com/watch?v=2pp17E4E-O8
64
SELEX
Systematic Evolution of Ligands by EXponential enrichment
• Systematic evolution of ligands by
exponential enrichment (SELEX) is well
established allowing not only directed
evolution of functional RNA molecules, but
also functional DNA and synthetic nucleic
acids (XNAs).
• In most cases, selection involves isolating
functional nucleic acids, converting them
to DNA (to allow efficient amplification by
PCR), and the regeneration of the
functional nucleic acid repertoire for
further rounds of selection.
(a) Diverse repertoires of XNA molecules may be synthesized (using engineered DNA-dependent XNA
polymerases. (b) Catalytic XNAs (XNAzymes) may be selected by reacting libraries of XNAs tagged with
substrates of interest (e.g., a disease-related RNA) and isolating on the basis of change in substrate (e.g.,
urea-PAGE gel shift upon RNA hydrolysis). (c,d) XNAzymes may be subsequently reverse-transcribed
using engineered XNA-dependent DNA polymerases (c), yielding cDNA that may be amplified to enable
either deep sequencing or generation of templates for XNA synthesis (d), and further rounds of X-SELEX.
65
66
Rational protein design workflow
67
68
Additivity: Additivity implies simple continuing fixing of improved mutations, and follows from
a model in which selection in natural evolution quite badly disfavours lower fitnesses, a
circumstance known from Gillespie as ‘strong selection, weak mutation’. For small changes
(close to neutral in a fitness or free energy sense), additivity may indeed be observed, and
has been exploited extensively in DE. If additivity alone were true, however (and thus there is
no epistasis for a given protein at all) then a rapid strategy for DE would be to synthesise all
20L amino acid variants at each position (of a starting protein of length L) and pick the best
amino acid at each position. However, the very existence of convergent and divergent
evolution implies that landscapes are rugged (and hence epistatic), so at the very least
additivity and epistasis must coexist.
Epistasis: The term ‘epistasis’ in DE covers a concept in which the ‘best’ amino acid at a
given position depends on the amino acid at one or more other positions. In fact, we believe
that one should start with an assumption of rather strong epistasis, as did Wright. Indeed the
rugged fitness landscape is itself a necessary reflection of epistasis and vice versa. Thus,
epistasis may be both cryptic and pervasive, the demonstrable coevolution goes hand in hand
with epistasis, and “to understand evolution and selection in proteins, knowledge of
coevolution and structural change must be integrated”.
Promiscuity. The concept of enzyme promiscuity mainly implies that some enzymes may bind,
or catalyse reactions with, more than one substrate, and this is inextricably linked to how one
can traverse evolutionary landscapes. It clearly bears strongly on how we might seek to effect
the directed evolution of biocatalysts.
69
SELEX
SELEX—A (r)evolutionary method to generate high-affinity nucleic acid
ligands
https://experiments.springernature.com/articles/10.1038/nprot.2015.104
https://www.nature.com/articles/nprot.2015.104/figures/1
70
Genome shuffling
71
Design principle of a CRISPR/Cas9 expression vector for construction of large-scale libraries. The vector
requires two minimal components, i.e., the single-guide RNA (sgRNA) sequence and the Cas9 gene that can
both be expressed from a single-vector system. The sgRNA is composed of a variable crRNA and a constant
tracrRNA. The gene sequence can be inserted by a simple adaptor cloning step in the 20 bp crRNA region.
Furthermore, multiple target sequences of the same gene can be inserted to increase efficiency and reduce offtarget
effects. At the 3′-end of the crRNA sequence, a 2 bp (base pair) protospacer adjacent motif (PAM – green
box) of the sequence GG is essential, but AG may be used to a lesser extend as well. A 12 bp seed region (blue
box) at the 3′-end is required for sequence tar-geting with additional 8 bp at the 5′-end contributing to specificity
(red box). sgRNA expression can be driven by a U6 promoter or alternatively by a H1 promoter. The Cas9
component is currently available in three different forms: as a wild-type (hCas9) or a mutant (D10A) version for
gene editing purposes and a catalytically dead (dCas9) version for gene silencing approaches. Exp-ression of
Cas9 can be driven by any mammalian expression promoter (e.g., EF1A, CMV, etc.) or retroviral promoter (LTR).
For nuclear targeting, the Cas9 gene requires multiple nuclear localization signals (NLS) and general expression
of Cas9 can be enhanced by inclusion of a woodchuck post-transcriptional regulatory element (WPRE) at
the 3′-end. A polyA-tail is required for expression vectors, while it should be deleted from lenti-/retroviral
expression vectors that have a polyA recognition sequence in their 3′-LTR. Variations of this design are possible
with respect to the Cas gene used or further modifications such as tagging the Cas gene with GFP or NLSs in
order to optimize Cas9 nuclear targeting.
72
Continuous directed evolution techniques
73
Continuous directed evolution strategies
Zhou and Alper, Journal of Chemical Technology & Biotechnology, 94:366-376 (2019)
Strategies for in vivo continuous directed evolution approaches include:
• Phage-assisted continuous evolution– PACE (B)
• In vivo continuous evolution – ICE (C)
• CRISPR/Cas9-based hypermutation strategies – CRSPR-X (D) and CREATE (E)
• GREACE (F)
74
Overview of phage-assisted continuous evolution (PACE)
• A culture of host E. coli continuously dilutes a fixed-volume vessel containing an evolving
population of selection phage (SP) in which essential phage gIII has been replaced by a protease
gene.
• These host cells contain an arabinose-inducible mutagenesis plasmid (MP) and an accessory
plasmid (AP) that supplies gIII.
• The expression of gIII is made
protease-dependent through
the use of a protease-activated
RNA polymerase (PA-RNAP)
consisting of T7 RNA
polymerase fused through a
cleavable substrate linker to
T7 lysozyme, a natural inhibitor
of T7 RNAP transcription.
• If an SP encodes a protease
capable of cleaving the substrate
linker, then the resulting
liberation of T7 RNAP leads to
the production of pIII and
infectious progeny phage
encoding active proteases.
• Conversely, SP encoding
proteases that cannot cleave the PA-RNAP yield non-infectious progeny phage
Packer et al., Nature Communications, 8: 956 (2017)
75
MAGE: the principle
The technique of MAGE is practically begun by growing the cells at 30 °C until the cell density reaches mid-log
phase. In this process, the λ-red proteins are expressed under the control of pL promoter which is regulated
by temperature sensitive λCI. The λCI represses the expression of λ-red proteins. The cells are transferred
to 42 °C for 15 min for heat shock induction where λCI is inactivated due to high temperature and pL
promoter expresses the λ-red proteins. The Exo protein degrades dsDNA in the 5′–3′ direction. The Beta
protein binds to ssDNA and generates recombination while Gam plays a key role in binding to the RecBCD
protein complex and subsequently preventing this complex from binding to dsDNA ends. It also helps to
induce the high efficiency of ssDNA recombination. The cells are then moved to 4 °C to repress λ-red and
prevent degradation, and then washed with chilled distilled water thrice. A pool of single stranded oligos is
introduced into cells via eletroporation and these oligos become incorporated into the lagging strand of the
replication fork during DNA replication.
Growth medium is added to the culture, which is then transferred to 30 °C for 2–3 h for the recovery of cells
with different sequence diversity. MAGE cycling should be repeated many times as required by the
experimental design. Each cell of the generated heterogeneous population contains a different set of
mutations. There are a number of applications of MAGE which facilitates the rapid and continuous
generation of a diverse set of genetic changes including for example, mismatches, insertions, and deletions.
The oligo mediated allelic replacement is the capability of introducing a number of genetic changes at high
efficiency. With MAGE, up to 30 bp mismatch mutations and insertion could be possible while up to the 45
kbp may be chromosomally sequences deleted and two-state hybridization free energy delta G between
oligo and the targeted complement region in the genome was also predicted. It indicates that a pool of oligos
with generate sequences have lesser frequency of incorporation than highly homologous sequences. MAGE
also provides a highly efficient, inexpensive and automated solution to concurrently modify many genomic
locations across different length scales from the nucleotide to genome level.
76
Selection strategies
• Affinity selection identifies library members that bind to an immobilized target. Methods for covalently linking proteins with
their corresponding genes during selection include display on phage particles via protein fusion to the coat protein pIII (left),
covalent attachment to their encoding mRNA transcript via a puromycin linkage (middle) and the non-covalent attachment of
both mRNA and nascent polypeptide to stalled ribosomes (right). b | Compartmentalized self-replication (CSR) selects for
DNA and RNA polymerases that can amplify, by PCR, their own genes within water emulsion droplets (blue circle) isolated
from one another by an oil phase (brown rectangle). c | In compartmentalized partnered replication (CPR), the evolving
activity must trigger expression of Taq polymerase. For example, aminoacyl tRNA synthetase (aaRS) activity promotes
amber stop codon suppression, leading to the expression of full-length Taq polymerase. Individual Escherichia coli cells are
then isolated in water–oil emulsion droplets and lysed by heat. Higher Taq expression leads to better PCR amplification of
the active library members. d | During phage-assisted
continuous evolution (PACE), host E. coli cells
continuously dilute an evolving population of ~1010
filamentous bacteriophages in a fixed-volume vessel
(cell stat; blue rectangle). Phage encoding active
variants trigger host cell expression of the missing
phage protein (pIII) in proportion to the desired activity
and consequently produce infectious progeny, whereas
phage with inactive variants produce progeny that
are not infectious and are diluted out of the vessel.
77
Screening protocols
• In screening protocols, individual clones are examined in small liquid cultures or
on Petri dishes to determine the properties of an enzyme that they produce.
Their efficiency depends on the concentration and specific activity of the
catalyst, the sensitivity of the test, the rate of the corresponding uncatalysed
reaction, and possible interference with other enzymes. The term selection is
reserved for situations where only clones producing a certain level of activity
survive. In vivo selection protocols are the simplest but require the expression of
the gene to confer a significant biological advantage. In vitro selection
techniques are based on a physical link between a protein and its encoding
gene – the best known system is phage display. In principle, selection
processes are more efficient because they can handle larger libraries (frequently
≥107 clones), whereas screening protocols are better suited for libraries of a
few-thousand clones. In both strategies, however, success depends on the
quality of the designed protocols.
78
CRISPR/Cas-directed evolution (CDE) platform
• All possible sgRNAs targeting the whole coding sequence of a gene are designed
• The sgRNA library is constructed via oligo synthesis and annealing
• The annealed oligos are cloned with sgRNA scaffold in the binary vector
• The sequences are confirmed by Sanger sequencing.
• All the plasmids are pooled and transformed into Agrobacterium
• The Agrobacterium cells are washed from plates with transformation medium and used for callus
transformation
• After two consecutive selections on hygromycin, the callus is regenerated under selection pressure (e.g.,
inhibitor)
• The resistant seedlings are recovered and the resistant plants are further analyzed by exhaustive
phenotyping under
selection pressure
• The plants are
genotyped by amplicon
sequencing, and protein
variants are identified
79
Examples of computationally designed enzymes
80
Phage-assisted continuous evolution (PACE)
DNAs harboring beneficial mutations are propagated by compartmentalization technique and PACE. Genetic
cassettes convert diverse protein functions such as protein–protein binding, protein–DNA interaction, protein
specific activity, and protein solubility to changes in the expression levels of taq-polymerase or of phage
infection protein. In the taq-polymerase case, the amount of amplified target product depends on taqpolymerase
content which is expressed in a cell. Each cell containing a plasmid carrying target DNA and taqpolymerase
is encapsulated together with PCR mixture (PCR buffer, dNTPs, primers). During emulsion
PCR, cells are disrupted and expose plasmid as template and expressed taq-polymerase. In M13 protein III
(pIII) case, expression of pIII is regulated. When beneficial mutations are occurred and increasing pIII
expression, phage carrying these mutations can generate more progeny
81
Continous directed evolution techniques
82
Cell surface display systems
83
Protein C-terminal end labelling
Puromycin
84
Protein synthesis using unnatural elements
• The in vitro translation can also be
done in a PURE (protein synthesis
using recombinant elements) system.
PURE system is an E. Coli cell-free
translation system in which only
essential translation components are
present. Some components, such as
amino acids and aminoacyl-tRNA
synthases (AARSs) can be omitted
from the system. Instead, chemically
acylated tRNA can be added into the
PURE system. It has been shown that
some unnatural amino acids, such as
N-methyl-amino acid accylated tRNA
can be incorporated into peptides or
mRNA-polypeptide fusions in a PURE
syste
85
The choice of screening strategy
• The choice of a screening or selection method
can be depicted as a decision tree that operates
primarily on the properties of the protein and
phenotype to be evolved. Although many
techniques can be extended to alternative
phenotypes, this figure focuses on the most
popular methods for each set of conditions. b |
Diversification strategies must be chosen both at
the outset of an evolution project and between
rounds of screening or selection. Considerations
can and should change over the course of a
project due to the phenotypes and genotypes
within the evolving population. This decision tree
attempts to distil these considerations with an
emphasis on focused mutagenesis methods that
have the maximum potential to identify functional
variants. epPCR, error-prone PCR; CPR,
compartmentalized partnered replication; CSR,
compartmentalized self-replication; FACS,
fluorescence-activated cell sorting; gIII, gene III;
NMR, nuclear magnetic resonance; PACE, phageassisted
continuous evolution; REAP,
reconstructed evolutionary adaptive path; SeSaM,
sequence saturation mutagenesis.
86
Dr. Martin Marek
Loschmidt Laboratories
Faculty of Science, MUNI
Kamenice 5, bld. A13, room 332
martin.marek@recetox.muni.cz