From OpenWetWare

Jump to: navigation, search


Zinc finger protein binding staples

Basic guideline of our approach was to avoid unnecessary use of covalent modifications of staple strands intended for DNA origami functionalization, which is the main advantage of our protein DNA origami add-ons approach. We achieved positional control of zinc finger binding proteins through replacement of appropriate standard staples with such that contained additional nucleotide sequences, which did not anneal to single stranded M13 mp18 DNA but formed double stranded DNA hairpins. These hairpins contained zinc finger protein variants target sequences. We prepared three types of hairpins: type A, B and C. Type A and type B hairpins are essentially the same but they form on opposite sides of DNA rectangle. Such design was intended for the formation of multiple layered stacks using protein tethers. Additionally we designed special dumbbell staples with which we wanted to make the binding of the protein on the surface of DNA origami more rigid.

Zinc finger protein Target DNA sequence Target sequence length [bp]
Zif268 variants* 5' gcgtgggcg 3' 9
6F6 5' gatcgggcggtaatgagat 3' 19
2C7 5' gcgtgggcggcgtgggcg 3' 18
AZPA4 5' gcgtgggcggcgtgggcg 3' 18
* Soluble Zif268 variant is an "in-house" designed protein with amino acid that are not essential for binding substituted with charged amino acids to improve solubility.

Type A and type B hairpins

Both A and B type hairpins are inverted repeats of ZFP target DNA sequences that fold onto themselves. The difference between types A and B is in the position of binding sequence within the staple strand. Type A hairpins are positioned at 3'-terminus of staple strands, whereas type B hairpins are integrated in the middle of staple strands. We included thymines at positions where steric hinderance might pose a problem.

Figure 45: Type A and B hairpins. Target DNA sequences depicted are binding sites for 2C7 zinc finger protein.

Type C hairpin

Type C hairpins were inspired by Rothemund's original dumbbell staple strand design (Rothemund, Nature 2006). We speculated we might have to bind chimeric zinc fingers more rigidly to the surface of DNA origami to achieve higher contrasting with the AFM. The main difference between type A/B and type C design is the position of the binding DNA helix relative to DNA origami rectangle. The DNA helix is perpendicular in the former design and parallel to the rectangle in the latter.

DNA tethers for DNA origami stacking

Figure 46: Type C hairpin
Using DNASequenceGenerator_v1.01b we designed the first set of ten unique binding sequences that were 20 bp long. The pool of sequences, which served as a reference for unique sequence calculations contained 216 standard staple strands used for construction of DNA origami rectangle. The uniqueness was set to be 7, which means that any heptad or higher (and its complementary sequence) should not be contained either in the pool of standard staple (and their complementary strands) either in the binding strands already generated and added to the pool. These unique staple strands were added to the 3' termini of sequences of selected staple strands, except when stated otherwise. The first set of binding sequences protrudes from the rectangle in the upward direction.

Sequence position in the DNA origami modelNucleotide sequence of generated binding sequences
Aa3 5' ctatcttcgactctcgattc 3'
AS1 5' cgagcgtccattgagttata 3'
Af3 5' ggtgttgcacgatacatacc 3'
Cb3 5' cctgcatcacatctagcttc 3'
CS1 5' gggaggccaaattaggatat 3'
CS4*5' gtagcactaagaaggcttgt 3'
Ce3 5' catacaccacaagaccactc 3'
Ea1 5' gcgtatagctgtataatggc 3'
ES4*5' cgatgctctcaggctattag 3'
Ef1 5' ggaattgctcaactattcgc 3'
* Soluble Zif268 variant is an "in-house" designed protein with amino acid that are not essential for binding substituted with charged amino acids to improve solubility.

After generating the first set of binding sequences we determined the set of complementary binding staples. Positions within the model were chosen such to allow for rectangles in the same spatial orientation to stack on top of each other. For this purpose we also had to be sure for the complementary sets of binding sequences to project from the rectangle on the opposite site as the first set. Staple strands at selected positions were divided in halves. The first halves were elongated with appropriate binding staples complementary to the first set at 3'-terminus, except when stated otherwise. The second halves were added to fill up the gaps that remained.

Sequence position in the DNA origami model Nucleotide sequence of generated binding sequences
Ba1 5' gaatcgagagtcgaagatag 3'
AS5 5' tataactcaatggacgctcg 3'
Bf1 5' ggtatgtatcgtgcaacacc 3'
Db1 5' gaagctagatgtgatgcagg 3'
CS5 5' atatcctaatttggcctccc 3'
CS8* 5' acaagccttcttagtgctac 3'
De1 5' gagtggtcttgtggtgtatg 3'
Ea3 5' gagtggtcttgtggtgtatg 3'
ES8* 5' ctaatagcctgagagcatcg 3'
Ef3 5' gcgaatagttgagcaattcc 3'

Figure 47: Arrangement of 10 binding staples on the opposing faces of DNA origami rectangle. Asymmetric arrangement decreases the likelihood of staggered overlay.


BioNanoWizards - BioMod 2011 team Slovenia. Design by Free CSS Templates.

Personal tools