Structure based design beats natural diversity

1
Structure based design beats natural diversity?
2
GFP structure

• ?-barrel structure
• Chromophore maturation
• After Heim et al., PNAS 91, 12501 (1994) by S.
  Jonda
• Spectral properties depend on chromophore
  environment

3
Spectral Properties

• wtGFP - has absorption maxima at 400nm and 470nm,
  due to different protonation status of
  chromophore
• S65 mutations eliminate absorption at 400nm

RS-GFP F64M, S65G, Q69L EYFP S65G, V68L, S72A,
T203Y Creemers et al., PNAS 97 p2974 (2000)
4

• Which positions should we target for mutagenesis?
• What aminoacids can we mutate to?

• Use ORBIT to identify targets

• Use ORBIT to identify targets
• Examine results of natural selection - look at
  a published alignment of GFP variants (Shagin et
  al., Mol.Bio. And Evol. V21 p.841 (2004) Supp.
  Mat.)

5
MSA-based library design

• Create Frequency Matrix from alignment
• At each position, find frequency of each
  aminoacid in alignment
• Set a threshold identify all aminoacids with
  frequencies above threshold
• Of these, eliminate the combinations that do not
  satisfy genetic code constraint
• Bias towards WT by enforcing each solution set to
  contain the WT aminoacid
• Lower threshold until library exceeds desired
  library size

6

• Idea use degenerate primers
• mixture of nucleotides at certain positions
• allow a subset of aminoacids to be encoded by
  that particular codon
• Advantages
• degenerate codons are equally represented in the
  library of mutant sequences
• Constraint
• the mutant sequences must be related by
  degenerate codons

7
http//www.tamu.edu/classes/plan/magill/gene310/we
btests/source.html
8
The Nucleotide Alphabet
Standard Bases A, C, G, U
Mixed Bases (IDT) R A, G Y C, U M A, C K
G, U S C, G W A, U H A, C, U B C, G, U V
A, C, G D A, G, U N/X A, C, G, U
Ser/Trp USG
Ser/Pro YCN
Trp/Pro NOT possible!
9
(No Transcript)
10
BsaXI

• Type IIB restriction enzyme
• cleaves both DNA strands upstream and downstream
  of a specific recognition site
• Recognition sequence

• Generates a 27bp fragment
• leaves 3overhangs (3 nt long), sequence
  independent

Tengs et al., Nucleic Acid Res. V32 (15) - 2004
11
The pBAD-GFP

• Arabinose-inducible pBAD promoter1
• In the absence of inducer, AraC binds promoter
• In the presence of inducer, AraC activates
  promoter
• GFP(S65T) gene inserted downstream of pBAD (at
  SfiI site)

• Sequence encoding aa. 57-72 replaced by BsaXI
  restriction site

1Guzman et al., J.Bacteriol. V177 p4121 (1995)
12
BsaXI Implementation

• 5 NNNNNNNNNACNNNNNCTCCNNNNNNNNNN 3
• 3 NNNNNNNNNNNNTGNNNNNGAGGNNNNNNN 5

5 NNNNNNNNNACNNNNNCTCCGCTNAGCNNN 3 3
NNNNNNNNNNNNTGNNNNNGAGGCGANTCG 5

• Can insert ANY library to replace residues 57-72

• Engineer-in BlpI restriction site inside BsaXI
  fragment
• BlpI digestion kills any vector incompletely
  digested with BsaXI that religated to itself

13
Library Evaluation protocol

• MiniPrep vector from DH5?
• BsaXI digest CIP treat vector
• Anneal phosphorylate primers
• Vector Library ligation
• BlpI digestion
• Transform into NM554 cells (RecA-) plate to
  obtain 3X oversampling
• Inoculate all colonies in 96-well plates
• Wash Take pictures fluorescence spectra
• Normalize for cell density in each well
• Count fraction functional estimate diversity
• Sequence interesting clones

14
Libraries designed so far
15
(No Transcript)
16
(No Transcript)
17
Conclusions

• Method works, is promising
• We will take spectra to assess functional
  diversity
• In this case one method will probably perform
  better than the other
• In general, more than one method of library
  design needs to be considered before judging
• Useful for applications in industry