Directed Evolution of a Genetic Circuit

1
Directed Evolution of a Genetic Circuit

• Balaji Srinivasan

2
Problem

• Want to design genetic circuits
• Understand naturally occuring networks
• Bioengineering applications
• But...
• circuits that work on paper never work in the
  lab...
• at least, not without optimization.

3
Making it work

• Several options
• Rational debugging (Weiss)
• Combinatorial synthesis (Guet)
• Directed evolution (this paper)

4
Goal Build Inverter

• Simplest Logic Gate

B
5
Inverter Implementation

• Two plasmid system
• 2 regulatory proteins
• IPTG input
• EYFP reporter

6
Relevant Genetics
7
Nonfunctional at first

• Leaky Plac ? product even w/o input
• low cI still represses EYFP expression
• Gate mismatch
• Want IPTG 0 ? EYFP 1
• Instead IPTG 0 interpreted as IPTG 1 from EYFP
  persp.
• Very familiar to EEs (impedance matching)

Low IPTG input
But still no fluorescence
8
Solution 1 Rational Design

• Reason that cI protein is too prevalent
• Damage RBS ? reduce cI protein levels
• Mutagenize Plac ? reduce leakiness
• Mutagenize PR ? reduce cI effectiveness
• This can be done (see Weiss)but its laborious

9
Solution 2 Evolve cI protein

• Cut plasmid
• PCR both pieces
• Mutagenic (Taq) PCR for cI gene and RBS
• Proofreading (Pfu) PCR for rest of plasmid
• Ligate (BsaI sites)

Pfu PCR
Taq PCR
Ligase
10
Screen Mutants

• Want
• IPTG 0, EYFP 1
• IPTG 1, EYFP 0

Jackpot!
IPTG low
IPTG high
11
Systems Analysis of Mutants

• Can now plot inverter device curves
• Compare to idealized inverter

12
Characterize Mutants

• Sequence successful cI mutant genes
• How did they solve the problem?
• What does this mean structurally?

13
Structural Mutants

• Many solutions
• Stop codons in cI ? remove C-term dimerization
  domain entirely!
• Disrupt start, RBS (reduce translation
  efficiency)
• Synonymous mutations (change transc/transl?)
• Protein-DNA interaction altered
• N and C term dimerization altered
• Model dimerization was another thing to tweak

14
Conclusion

• Directed evolution potentially potent tool for
  tuning circuits that should work
• Obvious research topics
• Selections rather than screens (must be careful)
• Multiplexing eliminate REAL cloning

15
References

• Guet comb. synthesis
• Elowitz repressilator
• Collins bistable inverter
• Hasty theory
• Leibler everything
• Barkai everything

16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)