DNAzymes at Work: A DFT Computational Investigation on the

Subscriber access provided by UNIV OF TASMANIA

Computational Biochemistry

DNAzymes at Work: A DFT Computational Investigation on the Mechanism of 9DB1 Edoardo Jun Mattioli, Andrea Bottoni, and Matteo Calvaresi J. Chem. Inf. Model., Just Accepted Manuscript • DOI: 10.1021/acs.jcim.8b00815 • Publication Date (Web): 31 Jan 2019 Downloaded from http://pubs.acs.org on February 2, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

1

DNAzymes at Work: A DFT Computational

2

Investigation on the Mechanism of 9DB1

3

Edoardo Jun Mattioli, Andrea Bottoni* and Matteo Calvaresi*

4

Dipartimento di Chimica “G. Ciamician”, Alma Mater Studiorum - Università di Bologna, V. F. Selmi 2, 40126 Bologna, Italy.

5

6

ABSTRACT

7

The 9DB1 DNAzyme follows an addition–elimination (AN+DN) two-step mechanism, involving a

8

phosphorane intermediate, where the 3’-hydroxyl group (nucleophile) of one RNA fragment attacks the

9

5’-triphosphate of another RNA fragment. This mechanism does not involve a divalent metal cation in

10

agreement with the experimental evidence. The process is assisted by two proton transfers that activate

11

the nucleophile (first step) and the leaving group (second step). The dA13 nucleotide is not directly

12

involved in the reaction. However, it plays an important role in determining the regioselectivity of the

13

process: since the dA13 phosphate forms a strong hydrogen bond with the 2’-hydroxyl, only the 3’-

14

hydroxyl can behave as a nucleophile and form the new 3’-5’ bond. In silico mutagenesis, where the

15

dA13 phosphate oxygen involved in the hydrogen contact was replaced by a sulfur atom, causes a

16

significant rearrangement of the A50 ribose position with an increase of the activation barrier and a

17

consequent lower enzymatic activity in agreement with the experimental evidence. A similar effect is

18

determined by the replacement of the 2’-hydroxyl with different groups such as F, H, OMe.

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

1 2

INTRODUCTION.

3

After the discovery of catalytic RNA (ribozyme) in the first eighties, it made some sense to assume that

4

also single-stranded DNA could have a catalytic activity. This assumption was based on the fact that

5

DNA contains almost the same functional groups as RNA. The first catalytic DNA molecule, with an

6

RNAase Pb2+-dependent activity, was isolated in 1994 by Breaker and Joyce.1 They used the in vitro

7

selection technique2,3 and demonstrated for the first time the catalytic possibility of DNA. Since then,

8

using in vitro selection from synthetic random DNA libraries, various catalytic DNA molecules

9

(deoxyribozymes) were identified. Nowadays, a large number of DNA-catalyzed reactions are known,4-

10

19

with promising potential applications in medicinal chemistry,20,21 nanotechnology,22,23 analytical

11

chemistry,24,25 organic synthesis26 and informatics.27,28 Recently, the DNA catalyst 9DB1, characterized

12

by RNA ligase activity, was crystallized in the post-catalytic state. The crystallized deoxyribozyme is a

13

strand of 44 nucleotides that catalyzes the regioselective formation of a native phosphodiester bond

14

between the 3’-hydroxyl and the 5’-triphosphate group of two RNA fragments. The structural

15

characterization was achieved through single crystal X-ray crystallography and the structural

16

arrangement of a deoxyribozyme in its complex tertiary structure was revealed for the first time (Figure

17

S1 in the SI file).29 The crystallographic structure corresponds to a post-catalytic conformation where

18

9DB1 forms a complex with the ligated RNA product. These data show how 9DB1 uses the two

19

binding arm motifs on the RNA substrates leaving only the two nucleotides in the ligation junction

20

unpaired. Ribozymes with RNA ligase activity were already found in living organisms and from in

21

vitro selection experiments.30,31 Interestingly, the active site of RNA enzymes catalyzing the same

22

reaction as 9DB1, is characterized by the presence of two phosphate groups that coordinate a divalent

23

metal ion. This catalytic ion is supposed to provide electrostatic stabilization of the transition state

24

along with other specific groups. However, no electron density suggesting the presence of a catalytic ACS Paragon Plus Environment

Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

metal ion was observed in the core of 9DB1 and the possible involvement of a divalent ion is still under

2

debate.29 Another important structural feature evidenced by the crystallographic study, is the position

3

of nucleotide dA13. Because of its proximity to the ligation junction (about 3.1 Å) dA13 was supposed

4

to play an important catalytic role, as suggested by mutagenesis and kinetic experiments.29

5

The commonly proposed catalytic mechanism is a general SN2-like reaction where the 3’-hydroxyl

6

group (nucleophile) attacks the 5’-triphosphate with the assistance of the dA13 phosphate group near the

7

junction site. However, up-to-date the catalytic mechanism has not been understood in detail yet. In

8

principle, different reaction channels are possible. In particular: (i) The nucleophile (3’-hydroxyl) is

9

first activated by a proton transfer that could be accomplished by the dA13 phosphate. (ii) The

10

nucleophilic attack occurs without a preliminary activation and is followed by a deprotonation by dA13

11

of the resulting acidic trivalent oxygen. (iii) The process is concerted and the nucleophilic attack and

12

deprotonation occurring in the same kinetic step. Since a deeper understanding of the intimate

13

mechanism is essential to define effective tools that can help to improve catalytic efficiency,32 we

14

carried out a quantum-mechanical (QM) computational study at the DFT level to elucidate in detail the

15

actual mechanism of 9DB1. Several examples are available in literature showing that this

16

computational methodology is successful to study enzymatic mechanisms.33-45

17

COMPUTATIONAL DETAILS

18

MD computations. To build a suitable model-system for quantum-mechanical (QM) computations, we

19

started from the crystallographic structure in the post-catalytic state (see Figure S1) and we generated

20

in silico the pre-catalytic state by breaking the bond between the two nucleotides G51 and A50. Then, we

21

added the triphosphate group, thus obtaining the reactant nucleotide GTP51. We carried out a 100ns

22

molecular dynamics (MD) on this pre-catalytic state (broken G51-A50 bond). This MD simulation was

23

carried out in the subspace of the full conformational space i.e. the subspace corresponding to the ACS Paragon Plus Environment


Page 4 of 22

1

motion of the reactant nucleotides involved in the reaction (A50 and GTP51). Thus, only these

2

nucleotides were free to move while the other nucleotides were “frozen” at their crystallographic

3

coordinates. We used a Quenched Molecular Dynamics (QMD) protocol to examine the 100ns MD

4

trajectory. 1000 snapshots (one snapshot every 0.1 ns) were considered along the trajectory and the

5

corresponding structures were optimized at the MM level. The lowest energy structure (corresponding

6

to optimized positions of the two nucleotides A50 and GTP51) was chosen as representative of the pre-

7

catalytic state and used to generate the model-system for the QM computations (starting point for the

8

subsequent QM optimization). The MD calculations were carried out with the AMBER1245 package in

9

implicit water using the generalized Born (GB) solvation model.46

10

QM computations. To reduce the size of the QM model-system we included all the residues within a

11

radius of 5 Å from the center of the reacting nucleotides GTP51 and A50. Also, for the residues

12

interacting via Watson-Crick base-pairing we removed the sugar ring and the phosphate moiety.

13

However, the dA13 phosphate group was included in the model since mutagenesis studies

14

unequivocally demonstrate its importance in the catalysis. The resulting model-system included (a) the

15

reacting nucleotides GTP51 and A50; (b) the nucleobase portion of dA15, dG27, dT29, dT30 (for each

16

nucleobase the removed sugar moiety was replaced by a methyl group); (c) the dA13 phosphate group

17

interacting with A50. A schematic representation of the entire QM model-system is given in Figure S2.

18

We carried out all computations with the Gaussian09 series of programs47 using the M06-2X

19

functional.48 Two basis sets of different accuracy were first used to describe the system: a 6-31G*

20

basis47 for the atoms involved directly in the reacting process and a 3-21G* basis47 for all remaining

21

atoms (the atoms described at different levels of accuracy are indicated in Figure S3 of the SI file). We

22

denoted this double-layer approach as M06-2X//6-31G*/3-21G* computational level. To check the

23

accuracy of our computational approach we re-optimized all critical points involved in the process

24

(reactants Rx, transition states TS1 and TS2, intermediate Int and products Pd) using the 6-311++G** ACS Paragon Plus Environment

Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

basis on the atoms participating actively in the reaction and the 6-31+G* basis for all remaining atoms

2

(M06-2X//6-311++G**/6-31+G* computational level).

3

To preserve the geometry of the active site cavity and emulate the constraining effect of the RNA-

4

ligated DNA, during the geometry optimization we “froze” the hydrogens that we added to replace the

5

removed atoms along the cut bonds and the atoms at the edges of the model-system at their original

6

crystallographic positions (the “frozen” atoms are evidenced in Figure S4). Also, we carried out IRC

7

computations on the transitions state of the rate-determining step (TS1) in both reactant and

8

intermediate directions. Frequency computations were carried to check the nature of the various critical

9

points. Since the reactive system is completely exposed to water, all geometry optimizations were

10

carried out using the IEF-PCM solvation model49 with the water dielectric constant  = 78.36.

11

RESULTS AND DISCUSSION.

12

The computed reaction surface is depicted in Figure 1. A conventional schematic representation of the

13

reaction is given at the bottom of the figure.



E (kcal mol-1)

Page 6 of 22

TS1 TS2

Int 22.8 [25.2]

18.7 [20.7]

15.6 [18.0]

Rx

r.c.

13.5 [11.2]

Pd dC12

O O H O O O O P O O O P O P O O O O H

O

NH

H N O N

1

dC12 O O P O O dA13 H O

N N dG27

A50

TS1

G51 OH

A50

O

O

O O P O

dA13

O H O

O

O H O P O G51 O O P O OH O O O P O O H NH H N N O N

O

O

O

dC12 O O O P dA13 O

A50

O H O O P O O O

TS2

G51

O OH H NH

H N O N

N

H O

O

P O

O O

P

O

O

N dG27

N dG27

2

Figure 1. The energy profile computed at the M06-2X//6-31G*/3-21G* computational level for the

3

9DB1catalyzed reaction. In square brackets the values obtained at the M06-2X//6-311++G**/6-31+G*

4

level. A conventional representation of the reaction mechanism is given on the bottom side of the

5

figure.

6

Two-dimensional and three-dimensional representations of the various critical points are given in

7

Figure 2 (Rx and TS1) and Figure 3 (Int and TS2). The nucleobases dA15, dT30 and dT29 (all included

8

in the QM computations) are deleted in the pictures since the interactions of these fragments with the

9

reacting core are approximately constant in the course of the reaction. In Rx the dA13 phosphate moiety

10

forms two strong hydrogen bonds with the 2’- and 3’-hydroxyl groups: the corresponding O1…HO2 and

11

O1…HO3 distances are 1.78 and 1.83 Å, respectively. The distance between the 3’-hydroxyl and the -

12

phosphorous atom of GTP51 is still rather large, the O3…P distance being 5.82 Å. Two important


Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

hydrogen bonds involve the dG27 nucleobase (nitrogen-bonded H atoms) and the second phosphate unit

2

of the GTP51 tri-phosphate moiety (O6 and O7 oxygens): the corresponding NH…O6 and NH…O7

3

distances are 1.72 and 1.90 Å, respectively.

4 5

Figure 2. A schematic two-dimensional (top) and three-dimensional (bottom) representation of the

6

reactant complex Rx and transition state TS1 (bond lengths are in ångstroms).

7

A transition state TS1 (Figure 2) was located for the nucleophilic attack of the 3’-hydroxyl on the -

8

phosphorous atom of the 5’-triphosphate group. We found that this attack is “assisted” by a transfer of

9

the hydroxyl proton. The 3’-hydroxyl proton is “captured” by one of the negative oxygen (O4) bonded

10

to the -phosphorous atom of the GTP51 triphosphate moiety. In TS1 the proton transfer and the

11

nucleophilic attack occur simultaneously and are an example of “substrate-assisted mechanism”.50 The



Page 8 of 22

1

O3-P distance (new forming bond) is 2.11Å and the proton is approximately half-way between O3 and

2

O4, the H…O3 and H…O4 distances being 1.20 and 1.23Å, respectively.

3

The two hydrogen contacts involving the dG27 nucleobase and the second phosphate unit of GTP51

4

are replaced by similar rather strong interactions with the terminal phosphate: the NH…O8 and NH…O9

5

distances are 1.75 and 1.67 Å, respectively. These interactions are important in maintaining the tri-

6

phosphate moiety in the position suitable for the nucleophilic attack. Interestingly, the dA13 phosphate

7

does not “assist” directly the nucleophilic attack. However, the strong hydrogen contact involving the

8

2’-hydroxyl and the dA13 phosphate group is conserved during the transformation (O1…HO2 distance =

9

1.78 and 1.90 Å in Rx and TS1, respectively). This contact can be considered as a key-factor

10

responsible for the regioselectivity of the reaction: being the 2’-hydroxyl group engaged in a strong H-

11

bond, only the 3’-hydroxyl is available as a nucleophile to form the new bond between the two RNA

12

nucleotides. The computed activation energy for TS1 (rate-determining step of the process) is 22.8 kcal

13

mol-1, a value in good agreement with the activation free energy (22.0 kcal mol-1) that was obtained

14

from the experimental kinetic constant using the Eyring’s equation.29 After inclusion of the zero-point

15

energy corrections the activation barrier does not change significantly, being 21.2 kcal mol-1 (a

16

complete energy profile including zero-point energy corrections is given in Figure S5 of the SI file). A

17

re-optimization of Rx and TS1 at the M06-2X//6-311++G**/6-31+G* computational level provided a

18

barrier of 25.2 kcal mol-1 (values in square brackets in Figure 1) in good agreement with the previous

19

M06-2X//6-31G*/3-21G*value. This barrier becomes 23.6 kcal mol-1 when we consider zero-point

20

energy corrections (Figure S5). This suggests that double-layer M06-2X//6-31G*/3-21G* computations

21

provide a reliable description of the system (cartesian coordinates corresponding to the re-optimized 6-

22

311++G**/6-31+G* structures are reported in Table S3 of the SI file).


Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Alternative reaction channels where the nucleophile (3’-hydroxyl) is activated by the dA13 phosphate

2

or no preliminary activation occurs, were examined. To this purpose a detailed scan of the

3

corresponding regions of the potential surface was carried out. These reaction paths were discarded on

4

energy ground since in all cases they lead to regions of the potential surface which are more than 50

5

kcal mol-1 above reactants.

6

TS1 leads to intermediate Int (15.6 kcal mol-1 higher than the starting complex) where the -

7

phosphorous atom is penta-coordinated (Figure 3). The Minimun Energy Path (MEP) connecting the

8

reactant Rx and the intermediate Int is given in Figure S6 and in the video V1 in SI.

9



1

Figure 3. A schematic two-dimensional (top) and three-dimensional (bottom) representation of the

2

intermediate Int and transition state TS2 (bond lengths are in ångstroms).

Page 10 of 22

3

The two strong hydrogen contacts involving the dG27 nucleobase and the terminal GTP51 phosphate

4

group do not change significantly with respect to the initial complex and the previous transition state.

5

The hydrogen bond between the 2’-hydroxyl and the dA13 phosphate group also remains significant,

6

the O1…HO2 distance being 2.63 Å. In the final step (transition state TS2, 18.7 kcal mol-1 above Rx) the

7

P-O5 bond breaks with the consequent dissociation of the diphosphate group. The breaking of the

8

P-O5 bond is again “assisted” by a proton transfer from O4 to O6, i.e the oxygen belonging to the

9

detaching diphosphate group. This mechanism agrees with the evidence recently obtained for the

10

hydrolysis reactions of phosphate triesters that involve phosphorane intermediates.51 The reaction is

11

significantly exothermic, the energy of the products Pd being 13.5 kcal mol-1 lower than reactants.

12

The energy values of Int and TS2 become 18.0 and 20.7 kcal mol-1 after re-optimization at the more

13

accurate M06-2X//6-311++G**/6-31+G* level. The exothermicity of the reaction changes from 13.5 to

14

11.2 kcal mol-1. These results are a further validation of the reliability of the M06-2X//6-31G*/3-21G*

15

approach. Cartesian coordinates corresponding to the re-optimized structures of Int, TS2 and Pd are

16

reported in Table S3 of the SI file.

17

The involvement of a phosporane intermediate suggests an addition–elimination pathway

18

(AN+DN),52,53 in which the nucleophilic attack leads to a trigonal-bipyramidal pentacoordinate

19

phosphorus that breaks down into products in a subsequent step. In the present case the formation and

20

breakdown of phosphorane is triggered by two “substrate assisted” proton transfers: one involving the

21

3’-hydroxyl and the other the hydroxyl bonded to the -phosphorous atom.


Page 11 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

When we include zero-point energy corrections, Int and TS2 become almost degenerate, their energy

2

relative to reactants being 17.3 and 17.7 kcal mol-1, respectively (see Figure S5 of the SI file). The

3

same trend is observed at the M06-2X//6-311++G**/6-31+G* level: in that case Int and TS2 become

4

exactly degenerate i.e. 19.7 kcal mol-1. The negligible (or non-existing) barrier between Int and TS2,

5

suggests that the entire process could be a concerted SN2-like substitution with the initial attack of the

6

nucleophile and the expulsion of the diphosphate (leaving group) occurring very asynchronously in two

7

different phases of a unique kinetic step.

8

Our computations clearly indicate that the H contact between the 2’-hydroxyl and the dA13 phosphate

9

(O1…HO2 contact) engages the 2’-hydroxyl and favors the regioselective formation of the 3’-5’ bond.

10

At the same time, it contributes to stabilize transition state TS1. This finding confirms the observation

11

of Salvatierra and co-workers29 on the role of the 2’-hydroxyl. These authors carried out mutagenesis

12

experiments by substituting the 2’-OH with other groups. We carried out similar in silico mutagenesis

13

and we replaced the 2’-OH with F, H and OMe. We re-optimized Rx and TS1 for each mutated form

14

and we computed the corresponding activation barriers (Figure 4). In these new calculations we used

15

the M06-2X//6-31G*/3-21G* approach, since this computational level was demonstrated to provide a

16

reliable description of the entire mechanism at a lower computational cost.

17



Page 12 of 22

1 2

Figure 4. In silico mutagenesis operated on (a) the 2’-hydroxyl group and (b) dA13 phosphate. The

3

corresponding computed activation barriers (kcal mol-1) for TS1 are reported. At the bottom of the

4

figure the distances between 3’-OH and oxygen O1 of dA13 phosphate are collected.

5 6

We found that the activation barrier slightly increases for F (25.4 kcal mol-1) and, then, further

7

increases for H (31.2 kcal mol-1) and OMe (31.6 kcal mol-1). The kinetic trend suggested by these

8

values is: OH (non-mutated) → F → H → OMe, in agreement with the experimental evidence29

9

showing that the reaction becomes slower when OH is replaced by F and much slower for H and OMe.

10

The optimized geometries for the various mutated forms show that, in the absence of the hydrogen ACS Paragon Plus Environment

Page 13 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

contact, the dA50 ribose moiety moves far away from the dA13 phosphate. Since the new position

2

corresponds to a more stable structural arrangement of the reactant complex, this determine an increase

3

of the reaction barrier. To test the role of the dA13 phosphate we carried out a second in silico

4

mutagenesis by replacing either of the non-bridging oxygen atoms with a sulfur atom. The computed

5

activation barriers for the two resulting stereoisomers Rp-P(S) and Sp-P(S) are 21.6 and 25.5 kcal mol-1,

6

respectively. These barriers follow the kinetic trend evidenced by the experiment of Salvatierra.29 In the

7

former case we evidenced a slight increase of the enzymatic activity (decrease of the activation barrier

8

of the rate-determining step TS1), while in the latter case the process became slower (increase of the

9

activation barrier) as experimentally observed. Interestingly, in Rp-P(S), the strong hydrogen bond

10

involving the 2’-OH group and O1 is maintained (O1…HO2 distance is 1.90 Å) and the dA50 ribose does

11

not move significantly with respect to its original position in the non-mutated form. In Sp, where the

12

sulfur atom replaces oxygen O1, the nature of the hydrogen contact 2’-OH changes significantly. Since

13

the new hydrogen contact involving sulfur is much weaker with respect to that of the non-mutated form

14

(S…HO2 distance becomes 2.43 Å), the A50 ribose can move significantly from its original position

15

leading to a more stable arrangement of the reactant complex and a consequent increase of the

16

activation barrier.

17 18

CONCLUSIONS

19

In this paper we have shown that, in analogy to many ribozymes,54,55 a possible catalytic mechanism

20

of 9DB1 follows a AN+DN reaction pattern, involving a phosphorane intermediate, where the 3’-

21

hydroxyl group (nucleophile) of one RNA fragment attacks the 5’-triphosphate of another RNA

22

fragment. This mechanism does not require the presence of a divalent metal cation in agreement with

23

the experimental indications of Salvatierra29 who did not find any evidence for electron density of a

24

catalytic metal ion. The process is assisted by two proton transfers occurring in different steps of the



Page 14 of 22

1

reaction: one from the 3’-hydroxyl to the oxygen bonded to the -phosphorous atom of the

2

triphosphate moiety of GTP51 (nucleophile activation assisted by the substrate); the other from the

3

-phosphorous atom to the second phosphate group of GTP51 (leaving group activation).

4 5

An accurate benchmark of the computational level demonstrated that the M06-2X//6-31G*/3-21G* approach provide a reliable description of the reaction mechanism and its energetics.

6

We found that dA13 is not directly involved in the reaction. However, this nucleotide, because of its

7

proximity to the ligation junction, plays an important catalytic role, as suggested by mutagenesis and

8

kinetic experiments. The dA13 phosphate forms a strong hydrogen bond with the 2’-hydroxyl and

9

determines the regioselectivity of the process: since the 2’-hydroxyl is engaged in a strong hydrogen

10

contact, only the 3’-hydroxyl can behave as a nucleophile and form the new 3’-5’ bond. Also, the

11

hydrogen contact maintains the A50 ribose in the right position to carry out the ligation process easily.

12

In silico mutagenesis, where the dA13 phosphate oxygen involved in the hydrogen contact was replaced

13

by a sulfur atom, is accompanied by a significant rearrangement of the ribose position with an increase

14

of the activation barrier. This leads to a lower enzymatic activity in agreement with the experimental

15

evidence. Similarly, the replacement of the 2’-hydroxyl with different groups (such as F, H, OMe) and

16

the consequent disappearing of the strong hydrogen contact with the dA13 phosphate, causes a

17

displacement of the A50 ribose from its original position and a consequent increase of the activation

18

barrier with a corresponding decrease of the enzymatic activity.

19 20

ASSOCIATED CONTENT

21

The following file is available free of charge (schemes of the QM model, Cartesian coordinates of the

22

critical points)

23

AUTHOR INFORMATION


Page 15 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Corresponding Author

2

AB: [email protected]

3

MC: [email protected]

4 5 6 7 8 9

REFERENCES (1)

Breaker, R. R., and Joyce, G. F. (1994) A DNA Enzyme that Cleaves RNA. Chem. Biol. 4,

618-622. (2)

Mills, D. R., Peterson, R. L., and Spiegelman, S. (1967) An Extracellular Darwinian

Experiment with a Self Duplicating Nucleic Acid Molecule. Proc. Natl. Acad. Sci. USA 58, 217-224.

10

(3)

Joyce, G. F. (2007) Forty Years of In Vitro Evolution. Angew. Chem. Int. Ed. 46, 6420-6436.

11

(4)

Chandra, M., and Silverman, S. K., (2008) DNA and RNA Can Be Equally Efficient Catalysts

12 13 14 15 16 17

for Carbon-Carbon Bond Formation. J. Am. Chem. Soc. 130, 2936-2937. (5)

Mohan, U., Burai, R., and McNaughton, B.R. (2013) In Vitro Evolution of a Friedel-Crafts

Deoxyribozyme. Org. Biomol. Chem. 11, 2241-2244. (6)

Li, Y. F., and Sen, D., (1996) A Catalytic DNA for Porphyrin Metallation. Nat. Struct. Biol. 3,

743-747. (7)

Flynn-Charlebois, A., Wang, Y., Prior, T. K., Rashid, I., Hoadley, K. A., Coppins, R. L., Wolf,

18

A. C., and Silverman, S. K. (2003) Deoxyribozymes with 2’-5’ RNA Ligase Activity. J. Am. Chem.

19

Soc. 125, 2444-2454.

20 21

(8)

Sheppard, T. L., Ordokhanian, P., and Joyce, G. F. (2000) A DNA Enzyme with N-Glycosylase

Activity. Natl. Acad. Sci. U.S.A. 97, 7802-7807. ACS Paragon Plus Environment


1 2 3 4 5 6 7 8

(9)

Page 16 of 22

Chiuman, W., and Li, Y. (2006) Evolution of High-Branching Deoxyribozymes from Catalytic

DNA with a Three-Way Junction Chem. Biol. 13, 1061-1069. (10) Gellert, M., Lipsett, M. N., and Davies, D. R. (1962) Helix Formation by Guanylic Acid. Proc. Natl. Acad. Sci. U.S.A. 48, 2013-2018. (11) Arnott, S., Chandrasekaran, R., and Martilla C. M (1974) Structures for Polyinosinic Acid and Polyguanylic Acid. Biochem. J. 141, 537-543. (12) McManus, S. A., and Li, Y. (2010) The Structural Diversity of Deoxyribozymes. Molecules 15(9), 6296-6284.

9

(13) Semlow, D. R., and Silverman, S. K. (2005) Parallel Selections in Vitro Reveal a Preference for

10

2’-5’ RNA Ligation By Deoxyribozyme-Mediated Opening of a 2’-3’-Cyclic Phosphate. J. Mol. Evol.

11

61, 205-217.

12

(14) Hoadley, K. A., Purtha, W. E., Wolf, A. C., Flynn-Charlebois, A., and Silverman, S. K. (2005)

13

Zn2+-Dependent Deoxyribozymes That Form Natural and Unnatural RNA Linkages. Biochemistry 44,

14

9217-9231.

15

(15) Kost, D. M., Gerdt, J. P., Pradeepkumar, P. I., and Silverman, S. K. (2008) Controlling the Direction

17

Deoxyribozymes That Use 2',3'-Cyclic Phosphate RNA Substrates. Org. Biomol. Chem. 6, 4391-4398.

18 19

of

Site-Selectivity

and

Regioselectivity

in

RNA

Ligation

by

Zn2+-Dependent

16

(16) Silverman, S. K. (2009) Deoxyribozymes: Selection Design and Serendipity in the Development of DNA Catalysts. Accounts of Chemical Research 42, 1521-1531.


Page 17 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1


(17) Purtha, W. E., Coppins, R. L., Smalley, M. K., and Silveman, S. K.

(2005) General

2

Deoxyribozyme-Catalyzed Synthesis of Native 3’-5’ RNA Linkages. J. Am. Chem. Soc. 126, 13124-

3

13125.

4

(18) Coppins, R. L, and Silverman, S. K. (2005) A Deoxyribozyme That Forms a Three-Helix-

5

Junction Complex With Its RNA Substrates and Has General RNA Branch-Forming Activity. J. Am.

6

Chem. Soc. 127, 2900-2907.

7 8

(19) Wang, Y., and Silverman, S. K. (2005) Directing the Outcome of Deoxyribozyme Selections to Favor Native 3'-5' RNA Ligation. Biochemistry, 44, 3017-3023.

9

(20) Bennett, C. F., and Swayze, E. E. (2010) RNA Targeting Therapeutics: Molecular Mechanisms

10

of Antisense Oligonucleotides as a Therapeutic Platform. Annu. Rev. Pharmacol. Toxicol. 50, 259-293.

11

(21) Goyenvalle, A., Griffith, G., Babbs, A., Andaloussi, S., Ezzat, K., Avril, A., Dugovic, B.,

12

Chaussenot, R., Ferry, A., Voit, T., Amthor, H., Bühr, C., Schürch. S., Wood, M. J., Davies, K.

13

E., Vaillend, C., Leumann, C., and Garcia, L. (2015) Functional Correction in Mouse Models of

14

Muscular Dystrophy Using Exon-Skipping Tricycle-DNA Oligomers. Nat. Med 21, 270-275.

15 16 17 18

(22) Aldaye, F. A., Palmer, A. L., and Sleiman, H. F. (2008) Assembling Materials with DNA as the Guide. Science 321, 1795-1799. (23) Jones, M. R., Seeman, N. C., and Mirkin, C. A. (2015) Nanomaterials. Programmable Materials and the Nature of the DNA Bond. Science 347, 840-852.

19

(24) Turner, A. P. (2013) Biosensors: Sense and Sensibility. Chem. Soc. Rev. 42, 3184-3196.

20

(25) Kim, H. N., Ren, W. X., Kim, J. S., Yoon, J. (2012) Fluorescent and Colorimetric Sensors for

21

Detection of Lead, Cadmium, and Mercury Ions. Chem. Soc. Rev. 41, 3210-3244.



Page 18 of 22

1

(26) Boersma, A. J., Coquière, D., Geerdink, D., Rosati, F., Feringa, B. L., and Roelfes, G. (2010)

2

Catalytic Enantioselective Syn Hydration of Enones in Water Using a DNA-Based Catalyst. Nat. Chem.

3

2, 991-995.

4 5

(27) Guo, Y., Zhou, L., Xu, L., Zhou, X., Hu, J., and Pei, R. (2014) Multiple Types of Logic Gates Based on a Single G-Quadruplex DNA Strand. Sci. Rep. 4, 7315.

6

(28) Wu, C., Wan, S., Hou, W., Zhang, L., Xu, J., Cui, C., Wang, Y., Hu, J., and Tan, W. (2015) A

7

Survey of Advancements in Nucleic-Acid Based Logic Gates and Computing Applications in

8

Biotechnology and Biomedicine. Chem. Commun. 51, 3723-3734.

9 10 11 12 13 14 15 16 17 18 19 20

(29) Ponce-Salvatierra, A., Wawrzyniak-Turek, K., Steuerwald, U., Hobartner, C., and Pena, V. (2016) Crystal Structure of a DNA Catalyst. Nature 529, 231-234. (30) Bartel, D. P., and Szostak, J. W. (1993) Isolation of New Ribozymes from a Large Pool of Random Sequence. Science 261, 1411-1418. (31) Joyce, G. F. (2004) Directed Evolution of Nucleic Acid Enzymes. Annu. Rev. Biochem. 73, 791-836. (32) Wirmer-Bartoschek, J., and Schwalbe, H. (2016) Understanding How DNA Enzymes Work. Angew. Chem. Int. Ed. 55, 5376-5377. (33) Calvaresi, M.; Garavelli, M.; Bottoni, A. (2008) Computational Evidence for the Catalytic Mechanism of Glutaminyl Cyclase. A DFT Investigation. Proteins 73, 527-538. (34) Siegbahn, Per E. M., and Borowski, T. (2011) Comparison of QM-only and QM/MM Models for the Mechanism of Tyrosinase. Faraday Discuss. 148, 109-117.


Page 19 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

(35) Bottoni, A., Miscione G. P., Calvaresi, M. (2011) Computational Evidence for the Substrate-

2

Assisted Catalytic Mechanism of O-GlcNAcase. A DFT Investigation. Phys. Chem. Chem. Phys. 13,

3

9568-9577

4

(36) Williams, R. T., and Wang, Y. (2012) A Density Functional Theory Study on the Kinetics and

5

Thermodynamics of N-Glycosidic Bond Cleavage in 5-Substituted 2′-Deoxycytidines. Biochemistry

6

51, 6458-6462.

7

(37) Wójcik, A., Broclawik, E., Siegbahn, Per, E. M., and Borowski, T. (2012) Mechanism of

8

Benzylic Hydroxylation by 4-Hydroxymandelate Synthase. A Computational Study. Biochemistry 51,

9

9570-9580.

10

(38) Sheng, X., and Liu, Y. (2013) Theoretical Study of the Catalytic Mechanism of E1 Subunit of

11

Pyruvate Dehydrogenase Multienzyme Complex from Bacillus Stearothermophilus. Biochemistry 52,

12

8079−8093.

13

(39) Aranda, J., Cerqueira, N. M. F. S. A., Fernandes, P. A., Roca, M., Tuñon, I., and Ramos, M. J.

14

(2014) The Catalytic Mechanism of Carboxylesterases: A Computational Study. Biochemistry 53,

15

5820−5829.

16

(40) Marforio, T. D.; Giacinto, P.; Bottoni, A.; Calvaresi, M. (2015) Computational Evidence for the

17

Catalytic Mechanism of Tyrosylprotein Sulfotransferases: A Density Functional Theory Investigation.

18

Biochemistry 54, 4404–4410.

19

(41) Fick, R. G., Clay, C. C., Lee, L. V., Scheiner, S., Al-Hashimi, H., and Trievel, R. C. (2018)

20

Water-Mediated Carbon−Oxygen Hydrogen Bonding Facilitates S‑Adenosylmethionine Recognition in

21

the Reactivation Domain of Cobalamin-Dependent Methionine Synthase, Biochemistry 57, 3733-3740.



Page 20 of 22

1

(42) Rankin, J. A., Mauban, R. C, Fellener, M., Desguin, B., McCracken, J. Hu, J., Varganov, S. A.,

2

and Hausinger, R. P. (2018) Lactate Racemase Nickel-Pincer Cofactor Operates by a Proton Coupled

3

Hydride Transfer Mechanism. Biochemistry 57, 3244-3251.

4 5

(43) Lintuluoto, M. and Lintuluoto, J.M. (2016) DFT Study on Enzyme Turnover Including Proton and Electron Transfers of Copper-Containing Nitrite Reductase Biochemistry, 55, 4697-4707.

6

(44) Messiha, H. L., Ahmed, S. T., Karuppiah, V., Suardiaz, R., Ascue Avalos, G. A., Fey, N.,

7

Yeates, S. Toogood, H. S., Mulholland, A. J., and Scrutton, N. S (2018) Biocatalytic Routes to Lactone

8

Monomers for Polymer Production. Biochemistry 57, 1997-2008.

9

(45) Fisher, G., Thomson, C. M., Stroek, R, Czekster, C. M., Hirschi, J. S., and da Silva, R. G.

10

(2018)

Allosteric

Activation

Shifts

the

Rate-Limiting

11

Phosphoribosyltransferase. Biochemistry 57, 4357–4367

Step

in

a

Short-Form

ATP

12

(46) Case, D. A., Darden, T. A., Cheatham, T. E., Simmerling, C. L., Wang, J., Duke, R. E., Luo,

13

R., Walker, R. C., Zhang, W., Merz, K. M., Roberts, B., Hayik, S., Roitberg, A., Seabra, G., Swails, J.,

14

Götz, A. W., Kolossváry, I., Wong, K. F., Paesani, F., Vanicek, J., Wolf, R. M., Liu, J., Wu, X.,

15

Brozell, S. R., Steinbrecher, T., Gohlke, H., Cai, Q., Ye, X., Wang, J., Hsieh, M.-J., Cui, G., Roe, D.

16

R., Mathews, D. H., Seetin, M. G., Salomon-Ferrer, R., Sagui, C., Babin, V., Luchko, T., Gusarov, S.,

17

Kovalenko, A., and Kollman, P. A. (2012), AMBER 12, University of California, San Francisco

18

(47) Frisch, M., Trucks, G., Schlegel, H., Scuseria, G., Robb, M., Cheeseman, J., Scalmani, G.,

19

Barone, V., Mennucci, B., Petersson, G., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H., Izmaylov,

20

A., Bloino, J., Zheng, G., Sonnenberg, J., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J.,

21

Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, J., Jr., Peralta, J.,

22

Ogliaro, F., Bearpark, M., Heyd, J., Brothers, E., Kudin, K., Staroverov, V., Kobayashi, R., Normand,


Page 21 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

J., Raghavachari, K., Rendell, A., Burant, J., Iyengar, S., Tomasi, J., Cossi, M., Rega, N., Millam, J.

2

M., Klene, M., Knox, J., Cross, J., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.,

3

Yazyev, O., Austin, A. J., Cammi, R., Pomelli, C., Ochterski, J., Martin, R., Morokuma, K.,

4

Zakrzewski, V., Voth, G., Salvador, P., Dannenberg, J., Dapprich, S., Daniels, A., Farkas, Ö.,

5

Foresman, J., Ortiz, J., Cioslowski, J., and Fox, D. (2009) Gaussian09, Revision D.1; Gaussian, Inc.:

6

Wallingford, CT.

7

(48) Zhao, Y., and Truhlar, D. G. (2008) The M06 Suite of Density Functional for Main Group

8

Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States, and Transition

9

Elements: Two New Functional and Systematic Testing of Four M06-Class Functionals and 12 Other

10 11 12 13 14 15 16 17 18

Functional. Theor. Chem. Acc. 120, 215−241. (49) Tomasi, J., Menucci, B., and Cammi, R. (2005) Quantum Mechanical Continuum Solvation Models. Chem. Rev. 105, 2999-3093 (50) Klahn, M., Rosta, E., and Warshel, A. (2006) On the Mechanism of Hydrolysis of Phosphate Monoesters Dianions in Solutions and Proteins. J. Am. Chem. Soc. 128, 47, 15310-15323 (51) Prasad, B. R., Plotnikov, N. V., and Warshel, A. (2013) Addressing Open Questions about Phosphate Hydrolysis Pathways by Careful Free Energy Mapping. J. Phys. Chem. B 117, 153-163. (52) Kirby, A. J., and Nome F., (2015) Fundamentals of Phosphate Transfer Acc. Chem. Res. 48, 1806-1814

19

(53) Pereira, E. S., Da Silva, J. C. S., Brandao, T. A. S., and Rocha, W. R. (2016) Phosphorane

20

Lifetime and Stereo-Electronic Effects Along the Alkaline Hydrolysis of Phosphate Esters. Phys.

21

Chem. Chem. Phys. 18, 18255-18267.



1 2 3 4

(54) Lilley, D. M. J., (2017) How RNA Acts as a Nuclease: Some Mechanistic Comparisons in the Nucleolytic Ribozymes. Biochem. Soc. Transact 45, 683-691. (55) Lilley, D. M. J. (2011) The Chemical Origins of Life and its Early Evolution: An Introduction. Phil. Trans. R. Soc. B 366, 2910-2917.

5 6

Page 22 of 22

TOC

7


DNAzymes at Work: A DFT Computational Investigation on the

Recommend Documents