A monovalent ion in the DNA binding interface of the eukaryotic junction-resolving enzyme GEN1

Abstract GEN1 is a member of the FEN/EXO family of structure-selective nucleases that cleave 1 nt 3′ to a variety of branchpoints. For each, the H2TH motif binds a monovalent ion and plays an important role in binding one helical arm of the substrates. We investigate here the importance of this metal ion on substrate specificity and GEN1 structure. In the presence of K+ ions the substrate specificity is wider than in Na+, yet four-way junctions remain the preferred substrate. In a combination of K+ and Mg2+ second strand cleavage is accelerated 17-fold, ensuring bilateral cleavage of the junction. We have solved crystal structures of Chaetomium thermophilum GEN1 with Cs+, K+ and Na+ bound. With bound Cs+ the loop of the H2TH motif extends toward the active site so that D199 coordinates a Mg2+, buttressed by an interaction of the adjacent Y200. With the lighter ions bound the H2TH loop changes conformation and retracts away from the active site. We hypothesize this conformational change might play a role in second strand cleavage acceleration.

. The sequences, secondary structures and cleavage sites of the different branched substrates used to examine cleavage activity by CtGEN1 in the presence of Na + and K + ions. The common element is the B arm, and the [5'-32 P]-labeled x strand (red). In these depictions 10 nt have been removed from each end of the 50 nt long strands, and from the 5' and 3' ends of the 5'h and 3'r strands respectively. The full sequences are shown in Table S1. Cleavage sites (1 nt 3' to the junction) are shown by the red arrows. The second site in the four-way junction is shown black, because this is not detected using the x-strand labelling scheme. Figure S3. The sequence of the pHRX plasmid cruciform shown as an inverted repeat (left) and as an extruded cruciform structure (right). In its cruciform conformation the four-way junction comprises the H, R and X arms of the standard junction 3. The central 24 bp of the inverted repeat comprises a (AT)12 repeat that confers facile extrusion of the cruciform structure in the negatively supercoiled plasmid. This cruciform has been used for the kinetic measurements reported in this work. Figure S4. Scheme showing the principle of the cruciform cleavage experiment. The cruciform structure is not stable in non-supercoiled DNA but is stabilized in negatively supercoiled DNA due to its local twist change. Once a cleavage is introduced the cruciform by the action of a nuclease the supercoiling is released and the cruciform is no longer stable and is absorbed into the body of the plasmid. Thus if the resolving enzyme dissociates after a unilateral cleavage, no further cleavage is possible because the substrate no longer exists, and the product is nicked circular DNA. However bilateral cleavage leads to the formation of a linear DNA product. The first cleavage event occurs with rate k1, and the second with k2. Supercoiled, nicked and linear DNA can be readily separated by agarose gel electrophoresis.  Binding isotherms for binding in K + ions (closed circles) and Na + ions (open circles). The data have been fitted to the Hill equation (lines) from which dissociation constants for binding were calculated. Note that the lowest concentration of CtGEN1 is really zero, i.e. corresponds to protein-free DNA junction. Figure S7. Affinity of CtGEN1 binding to a three-way junction in K + and Na + ions.
A. 0.5 nM junction was incubated with an increasing concentration of CtGEN1 and electrophoresed in a 6 % polyacrylamide gel in the presence of either 50 mM Na + or K + ions. The concentrations were (tracks 1 -10) : 0, 0.59, 0.989, 1. 63, 2.72, 4.54, 7.56, 12.6, 21.0 and 35.0 nM. B. Binding isotherms for binding in K + ions (closed circles) and Na + ions (open circles). The data have been fitted to the Hill equation (lines) from with dissociation constants for binding were calculated. Note that the lowest concentration of CtGEN1 is really zero, i.e. corresponds to protein-free DNA junction. Figure S8. The structure of the H2TH loop of CtGEN1 with a K + ion (K) bound. The protein is colored yellow except for the H2TH region that is colored pink. The view shown is similar to that presented in Figure 4A for the structure with Cs + bound. The active site and the M2 Mg 2+ ion (Mg, green) lies on the left in this view. Note that D199, Y200 do not approach the active center in this structure.

SUPPLEMENTARY TABLES
All DNA sequences are written 5' to 3'

Four-way junction 3
All strands 50 nt in length b-strand CCTCGAGGGATCCGTCCTAGCAAGGGGCTGCTACCGGAAGCTTACAGATG h-strand CATCTGTAAGCTTCCGGTAGCAGCCTGAGCGGTGGTTGAATTCACAGATG r-strand CATCTGTGAATTCAACCACCGCTCAACTCAACTGCAGTCTAGAACACATG x-strand CATGTGTTCTAGACTGCAGTTGAGTCCTTGCTAGGACGGATCCCTCGAGG

Three-way junction (junction 3 B, R and X arms)
All strands 50 nt in length br-strand CCTCGAGGGATCCGTCCTAGCAAGGACTCAACTGCAGTCTAGAACACATG r-strand CATCTGTGAATTCAACCACCGCTCAACTCAACTGCAGTCTAGAACACATG x-strand CATGTGTTCTAGACTGCAGTTGAGTCCTTGCTAGGACGGATCCCTCGAGG

Splayed arm junction
Strands 50 nt in length b-strand CCTCGAGGGATCCGTCCTAGCAAGGGGCTGCTACCGGAAGCTTACAGATG x-strand CATGTGTTCTAGACTGCAGTTGAGTCCTTGCTAGGACGGATCCCTCGAGG Table S1. Sequences of oligonucleotides used to construct the various DNA junctions used to analyze CtGEN1 substrate specificity in Na + and K + ions.

4H
D199A 3 x 10 -4 1.1 x 10 -3 4H Y200F 1.7 x 10 -4 1 x 10 -3 4H D199A Y200F 1.5 x 10 -5 2 x 10 -5 ___________________________________________________________________________ Table S2. Rates and binding affinities for CtGEN1. Rates of cleavage and dissociation constants have been measured for the four-way DNA junction 3. Rates were determined under single-turnover conditions in the presence of Na + or K + ions for the branched species shown in Figure S1. Rates of cleavage were measured in the presence of Rb + or Cs + ions for the four-way junction only. Cleavage rates were also measured in the presence of Na + or K + ions for the four-way junction using single and double mutants of CtGEN1. Dissociation constants for wild-type CtGEN1 were measured by electrophoretic retardation analysis in the presence of Na + or K + ions for the four branched species. Data were fitted to the Hill equation, and the Hill coefficients are reported in parenthesis.  6GRD   Table S3. Details of data collection and refinement statistics for the data as deposited in the PDB. Statistics for the highest resolution shell are in parenthesis.