Rational design of a DNA sequence-specific modular protein tag by tuning the alkylation kinetics

A design principle for sequence-specific DNA modifiers driven by the specific DNA recognition was proposed based on the kinetic parameters for DNA binding and modification reactions.


Table S1
Nucleotide sequences of Alexa Fluor 488 (A488) modified ODN used to determine equilibrium disassociation constants (K D ), association (k on ) and dissociation (k off ) rate constants for the complexes of modular adaptors with ODN by fluorescence polarization measurements.

Oligo name
Sequence (from 5' to 3') T A = A488 modified T A488-ODN-ZF CGCGACGCCCACGCGCGTT A TTCGCGCGTGGGCGTCGC A488-ODN-AZ CGCGATGCCACGTAGCGTT A TTCGCTACGTGGCATCGC A488-ODN-AP CGTTCATGACTCATGAGTT A TTCTCATGAGTCATGAAC  Table S3 Kinetic parameters for the cross-linking reaction between 5'-32 P-end-labeled BC-modified ODN derivatives and modular adaptors containing CLIP-tag. Rate constants for the matched pairs (k app ) are in bold, and for the unmatched pairs (k' app )) in plane.       The koff value for the complex of ZF-CLIP and ODN-AZ was not determined because of the fast kinetics under our experimental condition (koff < 1.0 s -1 ). Thus, koff for the complex between ZF-CLIP and ODN-AZ was calculated from its KD (1000 nM) deduced from the titration of fluorescence polarization (Table S2) and kon (1.3 × 10 6 M -1 s -1 ) because kon of the unmatched complex is similar to that of the matched complex S4,6-9 .

Rate constant (M -
*2 KD for the complex between ZF-CLIP and ODN-ZF was consistent with the value obtained from the titration of fluorescence polarization (Table S2).

Table S13
Nucleotide sequences of staple strands including zif268, AZP4 and GCN4 binding sites with BC modified T BC . The GCN4, zif268 and AZP4 binding sites on the staple strands were colored in red, blue and green, respectively.

Oligo name Sequence (from 5' to 3')
8g-AP-2BC     Table S15 Amino acid sequences and molecular weights of modular adaptor derivatives. at the apparent reaction rates (10 2 -10 6 M -1 s -1 ) were simulated by using Tenua S1-S3 as the kinetic simulator. The concentrations of substrate tethered ODN and modular adaptor were set to 10 nM and 100 nM, respectively.

Derivatives
By comparing the yields of cross-linking reaction for k app and k' app , the one order of magnitude difference of reaction rate is not sufficient to warrant the orthogonality for these cross-linking reactions. For example, when k app was set to 1 × 10 5 M -1 s -1 , the yield of cross-linking reaction reached 100% within 10 min. When k' app was 1 × 10 4 M -1 s -1 , the undesired cross-linking reaction proceeded to 50% at 10 min. When k' app was 1 × 10 3 M -1 s -1 , the yield of undesired cross-linking reaction was less than 10% in the reaction time. These simulation results strongly indicated that the difference of more than two orders of magnitude between k app and k' app was required to warrant the orthogonality for these cross-linking reactions.

Figure S2
Simulation of the apparent reaction rates (k app ) (M -1 s -1 ) with various rate constants for the covalent bond formation (k cov ) (10 -6 -10 s -1 ). As the simulation condition, the association rate constant (k on ) was set to 1 × 10 6 M -1 s -1 for both matched and unmatched pair and the dissociation rate constant (k off ) was set to 0.01 s -1 for matched pair and 1 s -1 to unmatched pair, respectively. These parameters are comparable to the reported k on and k off values S4 and our determined value from the fluorescence polarization analysis ( Figure S11 and Table S12).
The apparent reaction rates of matched (k app ) and unmatched (k' app ) pairs changed with varying the k cov value.
In the case of k cov = 1 s -1 (Line 1), k app is 1 × 10 6 M -1 s -1 and k' app is around 1 × 10 6 M -1 s -1 , thus the cross-linking reactions of matched and unmatched pairs proceed within the same time scale. In the case of k cov = 1 × 10 -5 s -1 (Line 3), k app is 1 × 10 3 M -1 s -1 and k' app is around 10 M -1 s -1 , giving two orders of magnitude difference, but required long incubation time to complete the reaction ( Figure S1). In the case of k cov = 4 × 10 -3 s -1 (Line 2), k app is 3 × 10 5 M -1 s -1 and k' app is around 4 × 10 3 M -1 s -1 , showing almost two orders of magnitude difference with maintaining enough reactivity for the matched pair. These simulation results strongly indicate that the tuning the rate constant of covalent bond formation is an effective way to realize the sequence selectivity between matched and unmatched pairs.  (Table S1) with increasing concentrations of modular adaptor (0.5 nM to 1024 nM) at ambient temperature. Estimated equilibrium dissociation constants are listed in Table S2.
s12 with ZF-CLIP to obtain the rate constants (k app ) (right). The determined rate constants are listed in Table S3.  Table S4.
s14  Table S8.  Table S10 and S11.  (Table S1), and red arrow indicates the addition of 2 µM unlabelled ODN-ZF as the competitor for alexa488 modified ODN. The complex formation process was analysed with addition of various concentrations of modular adaptor (20 nM ~ 100 nM) to 1 nM alexa488 modified ODN. The kinetic parameter k obs (s -1 ) was obtained by fitting to a reaction model assuming the first-order kinetics (eq S1), and then k on (M -1 s -1 ) and k off (s -1 ) was determined by following eq S2. The dissociation process was also monitored by addition of an excess amount of unlabeled ODN (2 µM) to the reaction mixture. The estimated values are shown in Table S12.   Table S15.