PosterPDF Available

Generating Binding Proteins - via in vivo directed evolution in E. coli

Authors:

Abstract and Figures

Evobodies - molecular speed dating We developed a novel and easy to use system for the generation of binding proteins in E. coli via in vivo directed evolution. Resulting proteins called Evobodies have the potential to bind specific to target proteins enabling various medical and analytical applications. Big advantages of our low-cost system are the short hands on time and the short generation time. As the starting point of our system, we designed and synthesized genetic libraries encoding binding proteins based on Nanobodies as well as Monobodies and submitted them as a new entity of BioBricks. The diversity of the respective library in E. coli is continuously increased by co-expressing a special DNA-Polymerase conferring plasmid restricted error-prone replication of the binding protein expressing plasmids. Finally, binding proteins with high affinity to the target protein are selected using a bacterial two-hybrid system providing growth advantage to antibiotics in relation to protein-protein interaction strength. Ultimately, desired clones are enriched during fermentation in batch or in continuous culture. Binding capacities of our libraries, efficiency of our selection system and potential of our mutagenesis system were demonstrated. Moreover, library diversity and mutation system characteristics were analyzed in detail by high-throughput sequencing.
Content may be subject to copyright.
Abstract
We developed a novel and easy to use system for the generation of binding proteins in E. coli via in vivo directed
evolution. Resulting proteins called Evobodies have the potential to bind specically to target proteins enabling
various medical and analytical applications. Great advantages of our low-cost system are the short hands on time and
the short generation time.
As the starting point of our system, we designed and synthesized a library of a new entity of BioBricks encoding
binding proteins in E. coli based on Nanobodies as well as Monobodies. The diversity of the library in E. coli is
continuously increased by co-expressing a special DNA-Polymerase inducing plasmid restricted error-prone
replication of the binding protein encoding plasmids. Finally, binding proteins with high anity to the target protein
are selected using a bacterial two-hybrid (B2H) system providing growth advantage to antibiotics in relation to
protein-protein interaction strength. Ultimately, desired clones are enriched during fermentation in batch or in
continuous culture.
Binding capability of our libraries, eciency of our selection system and potential of our mutagenesis system were
demonstrated. Moreover, library diversity and mutation system characteristics were analyzed in detail by
high-throughput sequencing (HTS).
Bianca Frommer, Carsten Hain, Niklas Homann, Judith Kampa, Cassandra Königs, Marten Linder,
Sebastian Perez Knoche, Fabian Roelos, Mikail Sahin, Pascal Schmidt and Marius Schöller
Center for Biotechnology (CeBiTec), Bielefeld University, Universitätsstraße 27, D-33615 Bielefeld, Germany
E
V
O
B
O
D
I
E
S
m
o
l
e
c
u
l
a
r
s
p
e
e
d
d
a
t
i
n
g
GEM
BIELEFELD 2016
GENERATING BINDING PROTEINS
via 
 E.
Because scientic language seems to
non-scientic people like a foreign language, we
created a dictionary on synthetic biology to
translate scientic language and especially
everything about Evobodies into regular English phrasing. We collected about 500
terms associated with our project or synthetic biology in general. This tool should
support non scientic people to understand science related media and give
scientists the chance to nd simple words when explaining their research to a
public audience.
In vivo mutagenesis system. By using either an error-prone polymerase I or a combination of
global mutator genes a single Evobody coding sequence can be evolved to various dierent vari-
ants, each with a unique binding site.
We transported the idea of
directed evolution to human
practices and started a series of
presentations at high schools.
Gradually we were able to im-
prove our presentation enhancing
the general understanding.
0,25
0,5
0,75
1
1234
Score
Evolutional Cycles
General Synthetic Biology
With our selection system, we want to nd the best adapted E. coli strain, and we
want to nd this strain as quickly as possible. Therefore, the prediction should
provide optimal parameters for laboratory use. The analysis of the program results
showed that a genome integration of the reporter gene with a strong promoter and
RBS is best for the B2H system. Additionally, we tted the growth curves of our
program to cultivation curves measured under dierent ampicillin concentrations.
Cultivation of E. coli JS200 with ampicillin
resistance. Each curve represents a culture.
Dierent concentrations of ampicillin, ranging
from 0.1 μg/mL to 10 μg/mL, were used in the
dierent cultures and t our model to this.
0.02
Amp concentration in periplasm [M]
0
0.2
0.4
0.6
1.4
1.2
1.6
1.8
1
0.8
x10-5
00.01 0.03 0.04 0.05 0.06
Expression rate of selectable marker [M/s]
0.02
50 Plasmids with strong promoter and RBS
Genome with strong promoter and RBS
Genome with strong promoter and RBS
50 plasmids with strong promoter and RBS
Change of expression rate in relation to ampicilin in the
periplasm. It is shown, how the number of genes change the
expression rate. The result of this modeling is that the
integration of the reporter gene in the genome is best suited for
our purpose .
Tecan-results of in vivo reporter activity. Two cul-
tures of E. coli were measured on their ability to pro-
duce RFP through the reporter construct. A culture
with one fusion protein SH2-cI expressing is com-
pared with a culture carrying both fusion proteins
SH2-cI (target protein fused to cI) and HA4-RpoZ
(binding protein fused to RpoZ). The RFP intensity
of the culture with both fusion proteins is signi-
cantly higher than in the culture with only one
fusion protein.
The system applied for the selection of functional Evobodies
was a B2H system [5]. With this method, bacterial cultures
can be selected based on the binding strength of two
fusion proteins of which one is the Evobody and the
other is the desired target. Successful binding
subsequently leads to expression of a selectable
marker. Important parameters of the system
were checked by showing successful
expression of the fusion proteins,
interaction of the positive controls
and the binding of the cI DNA-
binding domain to its target site.
However, the most important
achievement was the
demonstration of a whole
working B2H system.
In the Evobody generating process, we want to obtain a high anity binding protein by further evolving binding
proteins with an innate ability to interact with our target. By continuously mutagenizing binding proteins, we enable
an ongoing adaptation to the target.
We successfully utilized two dierent approaches for in vivo mutagenesis: a highly error prone polymerase I (EP) [3]
(BBa_K2082106) and a genome wide mutator system [4] (BBa_K2082117).
We investigated both mutators by reversion assays and HTS.
Dictionary of synthetic biology. Containing
500 words in a synthetic biology context.
Students talks. Presenting at schools and handing out surveys
helped us improve our public presence. The gained understanding is
documented in the bar graph on the right.
(a)
(b)
We implement a basic diversity of initial binding proteins by creating a minimalistic
high-quality library consisting of over one hundred thousand Monobodies and over
one hundred sixty thousand Nanobodies, respectively. We achieved the theoretical
diversity of 1,073,741,824 protein sequences for each library by randomizing
variable regions of the binding proteins preferring amino acids that favor
higher anities.
We veried our binding protein libraries with Phagemid display and HTS.
Thereby, we showed the occurrence of initial binders against diverse
targets and the correct distribution of the bases on randomized
regions [1; 2]. Construction of our part collection (BBa_K2082000,
BBa_K2082004, BBa_K2082005 and BBa_K2082009) was a
foundation for our project and a cornerstone for utilizing
libraries in iGEM.
Scheme of the binding proteins. Plasmids and 3D structures
with variable regions highlighted in light blue.
Nanobody randomized CDR3 of 24 colonies. Top to bottom: Ordered sequence, chromatogram and
sequencing result of 24 plasmids each representing a unique binding protein sequence.
Quantity of actually feasible diversity compared to theoretical diversity.
3,5 x 1016-fold more of the theoretical library can be covered by using our mini-
malistic scheme than with the conventional scheme including codons for all
amino acids but stops (NNK).
RNAP
promoter
blapromoter
RNAP
RNAP
promoter
promoter
RNAP
ampicilin
amp
icilin
bla
bla
bla
constant region
1,073,741,824
variants
variable region
perfect binder
target
binding proteins
N
a
n
o
b
o
d
y
C
D
S
s
e
l
e
c
t
i
o
n
p
l
a
s
m
i
d
M
o
n
o
b
o
d
y
C
D
S
s
e
l
e
c
t
i
o
n
p
l
a
s
m
i
d
constant region
variable region

Acknowledgement
References
Prof. Dr. Jörn Kalinowski, Prof. Dr. Kristian M. Müller, Julian Droste, Boas Pucker, Dr. Petra Lutter
[1] Fellouse, Frederic A. et al. (2004), In: Proceedings of the National Academy of Sciences of the United States of America 101 (34), S. 12467–12472.
[2] Koide, Shohei; Sidhu, Sachdev S. (2009), In: ACS chemical biology 4 (5), S. 325–334.
[3] Camps, Manel et al. (2003), In: Proceedings of the National Academy of Sciences of the United States of America 100 (17), S. 9727–9732.
[4] Badran, Ahmed H.; Liu, David R. (2015), In: Nature communications 6, S. 8425.
[5] Badran, Ahmed H. et al. (2016): In: Nature 533 (7601), S. 58–63.
B2H. Interaction between the binding protein (a) and the target protein (b) lead to recruitment of RNA
polymerase (RNAP) to the promoter upstream of the reporter cassette and subsequent expression of
the reporter gene. In this case the reporter is beta-lactamase which expression leads to degradation of
ampicillin and survival of the bacteria.
0
1000
2000
3000
4000
5000
6000
RFU
SH2
SH2+HA4
* *
Mutation rate determined by HTS. The mutation rate of the EP Pol I and
the genome wide mutator is increased in comparison to wild type Pol I
(WT) or a non-mutated control. Induction of the promoter in control of
the genome wide mutator yields additional mutation eect.

Actually feasible
(conventional
scheme)
Actually feasible
(our scheme)
0,00 0,02 0,04 0,06 0,08 0,10
Fraction of useable Evobodies [%]
3,5x1016-fold

LIBRARY
 
PuBLIC ENGAGEMENT

LIBRARY

SELECTION
Mutation
SELECTION


Mutation Modeling
ResearchGate has not been able to resolve any citations for this publication.
  • Ahmed H Badran
Badran, Ahmed H. et al. (2016): In: Nature 533 (7601), S. 58-63.
  • Frederic A Fellouse
Fellouse, Frederic A. et al. (2004), In: Proceedings of the National Academy of Sciences of the United States of America 101 (34), S. 12467–12472.
  • Manel Camps
Camps, Manel et al. (2003), In: Proceedings of the National Academy of Sciences of the United States of America 100 (17), S. 9727–9732.