Mechanistic Studies of the Protonation-Deprotonation Reactions for Type 1 and Type 2 Isopentenyl Diphosphate:Dimethylallyl Diphosphate Isomerase

Article abstract of DOI:10.1021/jacs.8b07274

Type 1 and type 2 isopentenyl diphosphate:dimethylallyl diphosphate isomerase (IDI-1 and IDI-2) catalyze the interconversion of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), the fundamental building blocks for biosynthesis of isoprenoid compounds. Previous studies indicate that both isoforms of IDI catalyze isomerization by a protonation-deprotonation mechanism. IDI-1 and IDI-2 are sluggish enzymes with turnover times of ～10 s^-1 and ～1 s^-1, respectively. We measured incorporation of deuterium into IPP and DMAPP in D₂O buffer for IDI-1 and IDI-2 under conditions where newly synthesized DMAPP is immediately and irreversibly removed by coupling its release to condensation with l-tryptophan catalyzed by dimethylallyltrytophan synthase. During the course of the reactions, we detected formation of d₁, d₂, and d₃ isotopologues of IPP and DMAPP, which were formed during up to five isomerizations between IPP and DMAPP during each turnover. The patterns for deuterium incorporation into IPP show that d₂-IPP is formed in preference to d₁-IPP for both enzymes. Analysis of the patterns of deuterium incorporation are consistent with a mechanism involving addition and removal of protons by a concerted asynchronous process, where addition substantially precedes removal, or a stepwise process through a short-lived (<3 ps) tertiary carbocationic intermediate. Previous work with mechanism-based inhibitors and related model studies supports a concerted asynchronous mechanism for the enzyme-catalyzed isomerizations.

Full text of DOI:10.1021/jacs.8b07274

Subscriber access provided by Kaohsiung Medical University
Article
Mechanistic Studies of the Protonation-Deprotonation Reactions for Type 1
and Type 2 Isopentenyl Diphosphate:Dimethylallyl Diphosphate Isomerase
Syam Sundar Neti, Jian-Jung Pan, and C Dale Poulter
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b07274 • Publication Date (Web): 05 Sep 2018
Downloaded from http://pubs.acs.org on September 5, 2018
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 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Mechanistic Studies of the ProtonationꢀDeprotonation
Reactions for Type 1 and Type 2 Isopentenyl
10
11
12
13
14
15
16
17
18
19
20
21
Diphosphate:Dimethylallyl Diphosphate Isomerase
Syam Sundar Neti, JianꢀJung Pan, and C. Dale Poulter*
Department of Chemistry, University of Utah, 315 South 1400 East, Salt Lake City, UT 84112
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
RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required according
to the journal that you are submitting your paper to)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 29
ABSTRACT
1
2
3
4
5
6
7
8
9
Type 1 and type 2 isopentenyl diphosphate:dimethylallyl diphosphate isomerase (IDIꢀ1 and IDIꢀ2)
catalyze the interconversion of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP),
the fundamental building blocks for biosynthesis of isoprenoid compounds. Previous studies indicate
that both isoforms of IDI catalyze isomerization by a protonationꢀdeprotonation mechanism. IDIꢀ1 and
IDIꢀ2 are “sluggish” enzymes with turnover times of ~10 sꢀ1 and ~1 sꢀ1, respectively. We measured
incorporation of deuterium into IPP and DMAPP in D2O buffer for IDIꢀ1 and IDIꢀ2 under conditions
where newly synthesized DMAPP is immediately and irreversibly removed by coupling its release to
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
condensation with ꢀtryptophan catalyzed by dimethylallyltrytophan (DMAT) synthase. During the
L
course of the reactions, we detected formation of d1, d2, and d3 isotopologues of IPP and DMAPP,
which were formed during up to five isomerizations between IPP and DMAPP during each turnover.
The patterns for deuterium incorporation into IPP show that d2ꢀIPP is formed in preference to d1ꢀIPP for
both enzymes. Analysis of the patterns of deuterium incorporation are consistent with a mechanism
involving addition and removal of protons by a concerted asynchronous process, where addition
substantially precedes removal, or a stepꢀwise process through a shortꢀlived (< 3 ps) tertiary
carbocationic intermediate. Work with mechanismꢀbased inhibitors and related model studies by
Richard and coworkers support a concerted asynchronous mechanism for the enzymeꢀcatalyzed
isomerizations.
ACS Paragon Plus Environment

Page 3 of 29
Journal of the American Chemical Society
Introduction
1
2
3
4
5
6
7
8
9
Isopentenyl diphosphate:dimethylallyl diphosphate isomerase (IDI) catalyzes the interconversion
of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP),1 the five carbon units
required for biosynthesis of more than 70,000 isoprenoid natural products.2 IPP and DMAPP are
biosynthesized by the mevalonate (MVA) pathway in Eukaryotes, Archaea, and some Bacteria3,4 or the
methylerythritol phosphate (MEP) pathway in most Bacteria and plant chloroplasts.5 IDI activity is
required for synthesis of DMAPP from IPP in organisms that utilize the MVA pathway6 and loss of IDI
activity is lethal. In contrast, IPP and DMAPP are synthesized as a mixture from 4ꢀhydroxyꢀDMAPP in
organisms that utilize the MEP pathway. Although IDI activity is not essential for their survival,7 most
organisms that rely on the MEP pathway have IDI activity,8 perhaps to balance the pools of IPP and
DMAPP.
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
Two convergently evolved and structurally distinct forms of IDI are known.9,10 IDIꢀ1,
discovered in the late 1950’s, is present in some Bacteria and the cytoplasm of Eukaryotes.8,11 IDIꢀ2 is
present in Archaea, other Bacteria, and plant chloroplasts.12,13 Both isoforms catalyze the same reaction,
although the organization of their active sites and catalytic residues responsible for isomerization are
different. IDIꢀ1 contains two divalent metal ions, typically Zn2+ and Mg2+ or Mn2+;14,15 whereas, IDIꢀ2
contains a single divalent metal, typically Mg2+, and flavin mononucleotide (FMN), which is fully
reduced in the active enzyme.10
The isomerizations catalyzed by IDIꢀ1 and IDIꢀ2 are stereoselective. Both enzymes bind IPP in
an extended conformation where C4 in IPP becomes the Eꢀmethyl group in DMAPP. Both remove the
pro-R hydrogen at C2 in IPP.16,17 However, IDIꢀ1 catalyzes an antarafacial isomerization of IPP to
DMAPP by adding a hydrogen to the reꢀface of C4 in IPP, while IDIꢀ2 catalyzes a suprafacial
isomerization by adding a hydrogen to the siꢀface of C4 (Scheme 1).16ꢀ20
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 29
DMAPP
IPP
1
2
3
4
5
6
7
B
H
B:
IDI-1
H
PPO
PPO
e
c
H
a
f
-
e
Antarafacial
H
HR
B:
H
H
r
H
B
8
9
B
H
B:
IDI-2
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
H
H
PPO
PPO
H
HR
B:
Suprafacial
H
H
H
B
Scheme 1. Stereochemistry for isomerization of IPP to DMAPP by IDIꢀ1 and IDIꢀ2.
1
When incubations of IDIꢀ1 or IDIꢀ2 with IPP in D2O are followed by HꢀNMR, the reversible
isomerization between IPP and DMAPP is accompanied by rapid incorporation of deuterium at the Eꢀ
methyl group of DMAPP and at C4 and pro-R locus at C2 of IPP in accord with the stereochemistries
reported for the two enzymes.16ꢀ20 For IDIꢀ1, all of the hydrogens, except those at C1, are eventually
replaced by deuterium in both molecules upon prolonged incubation, reflecting the stereoselective, but
not stereospecific, nature of the isomerization.21 In contrast, only one of the hydrogens at C2 and the
hydrogens at C4 are replaced during prolonged incubations with IDIꢀ2, suggesting that it is substantially
more stereoselective than IDIꢀ1.22 Careful analysis of the exchange reactions of IDIꢀ1 in D2O shows that
two atoms of deuterium are incorporated into IPP during each turnover of the IDIꢀ1·d0ꢀIPP complex.23
Several lines of evidence, including mechanismꢀbased inhibition, siteꢀdirected mutagenesis, and
studies with substrate analogues and transition state inhibitors, indicate that IDIꢀ1 and IDIꢀ2 catalyze
isomerization via a [1.3] protonationꢀdeprotonation mechanism.24ꢀ35 Activeꢀsite acids and bases were
identified from Xꢀray structures of IDIꢀ1 complexed with an ammonium transition state/reactive
intermediate analogue25 and both enzymes with covalently attached mechanismꢀbased irreversible
inhibitors.25,27,35 In the active site of IDIꢀ1, a Zn2+ꢀGlu/Tyr motif protonates the C3ꢀC4 double bond in
IPP and a cysteine removes the proton at C2.25 In contrast, a zwitterionic tautomer of reduced FMN
(FMNH2) is thought to be the acidꢀbase catalyst for protonation of the double bond in IPP and
abstraction of a proton from C2.36
ACS Paragon Plus Environment

Page 5 of 29
Journal of the American Chemical Society
The above work is consistent with either asynchronous concerted or stepwise mechanisms
1
2
(Scheme 2), with a few noteworthy exceptions. Incubation of IDIꢀ2 in D2O with a vinyl thiomethyl
analogue of IPP, where the methyl group was replaced with thiomethyl, resulted in rapid exchange of
the C4 protons without isomerization of the double bond.37 An IDIꢀ1 mutant, where the nucleophilic
cysteine that assists with proton removal was replaced by alanine, did not catalyze isomerization but
was covalently inactivated by the epoxy analogue of IPP. 25 An Xꢀray structure of the adduct showed
that the inhibitor alkylated the glutamate in the Zn2+ꢀGlu/Tyr protonation motif, suggesting protonation
of the analogue by the glutamate was followed by alkylation of that residue by the protonated epoxide.
Both observations are inconsistent with an asynchronous concerted protonationꢀdeprotonation process.
However, the carbocations generated by protonation of the thiomethyl and epoxy analogues of IPP are
substantially more stable than the putative tertiary carbocation derived from IPP or DMAPP, and it is
possible that they can be generated by the catalytic machinery in IDI, while the tertiary carbocation
from IPP or DMAPP is not. This suggestion is consistent with the observation that the C67A mutant of
IDIꢀ1, which was inactivated by the epoxy analog of IPP, did not catalyze isomerization. It is also
noteworthy that prolonged incubations of IDIꢀ1 and IDIꢀ2 with IPP/DMAPP do not result in alkylation
of the active site.21,22
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
PPO
H
H
Transition state
+H+/-H+
PPO
PPO
H
H
IDI-1 and IDI-2
DMAPP
IPP
-H+
+H+
PPO
H
H
Carbocationic
intermediate
Scheme 2. Stepꢀwise vs asynchronous concerted mechanisms for isomerization.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 29
We now report experiments that examine the incorporation of deuterium into IPP and DMAPP
for both IDIꢀ1 and IDIꢀ2 during incubations in D2O buffer by mass spectrometry, which reveal
formation of d1, d2, and d3 isotopologues of IPP and DMAPP during each turnover. Analysis of the
partitioning patterns of the isotopologues suggests that the protonationꢀdeprotonation sequence is
stepwise, involving a shortꢀlived carbocationic intermediate (τ1/2 < 3 ps), or a concerted asynchronous
reaction, where protonation significantly precedes deprotonation.
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
Results
Trapping experiments. In order to detect multiple isomerizations between IPP and DMAPP
during each turnover of IDI·d0ꢀIPP, we designed timeꢀcourse experiments where DMAPP produced by
isomerization of IPP by IDI is immediately and irreversibly consumed upon release by the
dimethylallyltryptophan (4ꢀDMAT) synthase catalyzed alkylation of tryptophan (Scheme 3). In a typical
experiment, IPP was incubated with E. coli IDIꢀ1 or S. pneumoniae IDIꢀ2 and Claviceps purpurea 4ꢀ
DMAT synthase in D2O buffer where the [Trp] was 5ꢀfold greater that the [IPP] and the specific activity
of DMAT synthase was 40ꢀfold higher than IDIꢀ1 and 1000ꢀfold higher than IDIꢀ2. All of the buffer
reagents, substrates, and proteins were thoroughly exchanged with deuterium by repeated lyophilization,
and the reactions were initiated by the addition of IDI. As the reactions proceeded, samples were
removed, and the enzymes were immediately separated from small molecules by ultrafiltration.
O
OH
NH2
N
PPO
H
IPP
4-DMAT
IDI
4-DMATS
-Trp
L
.
.
IDI
PPO
IDI
PPO
PPO
IDI
DMAPP
Scheme 3. Coupled reactions for removing DMAPP.
ACS Paragon Plus Environment

Page 7 of 29
Journal of the American Chemical Society
IPP and DMAT were separated on a reversed phase (RP) phenyl column (Figure 1). Fractions
1
2
containing DMAT were lyophilized and analyzed by positiveꢀion mass spectrometry. Under these
conditions, IPP eluted in the void volume. That fraction was lyophilized; the residue was dissolved in
water; and IPP was purified by ionꢀexchange chromatography before analysis by negative ion
electrospray mass spectrometry.38 In control experiments run in H2O, the intensities of 13C and 18O
isotope peaks in unlabeled DMAT and IPP agreed with theoretical values, which are ~ 18% and 0.4%
for DMAT and ~6% and 1.4% for IPP, respectively. There was no background at higher m/z values. In
control experiments where IDI and DMAT synthase were incubated with DMAPP instead of IPP, no
deuterium was incorporated into DMAT, demonstrating that DMAPP was efficiently trapped by DMAT
synthase and did not isomerize to IPP, even under conditions where the initial [DMAPP] was
substantially higher than the [DMAPP] achieved during isomerization of IPP (Figures S1 and S2).
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
Figure 1. HPLC chromatogram of the assay mixture showing the retention times for IPP (void volume,
Vo),
LꢀTrp (~11 min) and DMAT (~14.5 min) upon elution with a 0ꢀ100% gradient of H2O/acetonitrile
and detection at λ210.
Time courses for incorporation of IPP into DMAT, with immediate irreversible trapping of
released DMAPP during isomerizations catalyzed by IDIꢀ1 and IDIꢀ2, are shown in Figure 2. Peaks for
DMAT at m/z 296 (d1ꢀDMAT+Na+), 297 (d2ꢀDMAT+Na+), and 298 (d3ꢀDMAT+Na+) revealed that up
to three atoms of deuterium were incorporated into DMAPP during each turnover of IDIꢀ1 (Figures S3A
and S5ꢀS12) and IDIꢀ2 (Figures S3B and S12ꢀS19). The mole fractions of d1ꢀ, d2ꢀ, and d3ꢀDMAT at
each time point are given in Tables 1 and 2, respectively. These results show that newly released
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 29
DMAPP contained from one to three deuteriums (Tables 1 and 2), even at the earliest time points. It is
noteworthy that the relative ratios for d1ꢀ, d2ꢀ, and d3ꢀDMAT observed early in the time courses, when
most of the IPP in the sample did not contain deuterium, remained relatively constant up to ~25ꢀ40%
conversion of IPP to DMAPP. At higher levels of consumption of IPP, the relative amounts of the d2
and d3 isotopologues increased modestly relative to d1ꢀDMAT as deuterium was incorporated into IPP
(see below). Thus, the ratio of DMAPP isotopologues measured early in the reaction (up to ~25%
conversion) reflects the incorporation of deuterium during each turnover.
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
A
B
Figure 2. Percent conversion of IPP to DMAPP versus time for IDIꢀ1 (part A, three independent assays)
and IDIꢀ2 (Part B, three independent assays). Average of three independent runs.
ACS Paragon Plus Environment

Page 9 of 29
Journal of the American Chemical Society
Table 1. Timeꢀcourse for incorporation of deuterium into DMAT by IDIꢀ1.a,b
1
2
3
4
5
6
DMAT+Na+ a,b
(Mole fraction)
Time Conversion
(h)
(%)
7
8
9
m/z 296
m/z 297
m/z 298
d1
d2
d3
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
0.5
1
6 ± 3
0.75 ± <0.01 0.21 ± <0.01 0.04 ± <0.01
0.77 ± <0.01 0.20 ± <0.01 0.03 ± <0.01
15 ± 4
1.5
2.5
4
24 ± 8
40 ± 8
69 ± 10
75 ± 6
96 ± 1
0.78 ± 0.02 0.19 ± <0.01 0.03 ± <0.01
0.77 ± 0.01 0.20 ± < 0.01 0.03 ± <0.01
0.74 ± 0.01 0.22 ± <0.01 0.04 ± <0.01
6
0.71 ± 0.03
0.25 ± 0.02 0.04 ± <0.01
9
0.66 ± <0.01 0.27 ± <0.01 0.07 ± <0.01
a2 mM IPP, 12 mM
LꢀTrp, 0.1 mg/mL BSA, 10 mM MgCl2, 1 µM IDIꢀ1, 40 µM 4ꢀDMATS pD 7.8, 30
oC, reaction volume 2.5 mL. bAverage of three independent assays.
Table 2. Timeꢀcourse for incorporation of deuterium into DMAT by IDIꢀ2.a,b
DMAT+Na+ a,b
(Mole fraction)
Time Conversion
(h)
(%)
m/z 296
m/z 297
m/z 298
d1
d2
d3
1
4
5 ± 2
0.68 ± 0.02
0.26 ± 0.02
0.06 ± 0.02
22 ± 6
0.67 ± <0.01 0.26 ± <0.01 0.07 ± <0.01
6
8
35 ±8
47 ± 10
69 ± 11
100
0.65 ± <0.01 0.27 ± <0.01 0.08 ± <0.01
0.63 ± <0.01 0.28 ± <0.01 0.09 ± <0.01
0.59 ± <0.01 0.29 ± <0.01 0.12 ± <0.01
12
24
0.50 ± 0.03 0.31 ± <0.01
0.19 ± 0.03
a2 mM IPP, 10 mM
LꢀTrp, 0.5 mM FMN, 5 mM Na2S2O4, 0.1 mg/mL BSA, 10 mM MgCl2, 10 nM
o
b
spIDIꢀ2, 10 µM 4ꢀDMATS pD 7.8, 30 C, reaction volume 5.2 mL. Average of three independent
assays.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 29
The LCꢀMS peak intensities for unlabeled IPP and its deuterium isotopologues were measured
for the above time points for IDIꢀ1 (Figures S4A and S20ꢀ27) and IDIꢀ2 (Figures S4B and S28ꢀS34).
The mole fractions of d1ꢀ, d2ꢀ, d3ꢀIPP during the course of the reactions are given in Tables 3 and 4.
Incorporation of deuterium into IPP at low conversions of IPP to DMAPP under conditions where free
DMAPP is irreversibly trapped by DMAT synthase further supports the conclusion that interconversion
of IPP and DMAPP within the enzymeꢀsubstrate/product complex is fast relative to their release. As
seen for DMAPP, the distribution of IPP isotopologues during the first ~25% of the reaction reflects the
pattern for incorporation of deuterium during each turnover of the enzymes.
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
Table 3. Timeꢀcourse for incorporation of deuterium into IPP by IDIꢀ1.a,b
Time
(h)
Conversion
(%)
IPPꢀH+
(Mole fraction)
m/z 245
m/z 246
m/z 247
m/z 248
d0
d1
d2
d3
0.5
1
6 ± 2
15 ± 4
24 ± 8
40 ± 8
69 ± 10
75 ± 6
96
0.93 ± 0.03
0.92 ± 0.03
0.88 ± 0.04
0.80 ± 0.01
0.66 ± 0.01
0.41 ± <0.01
0.23
0.03 ± 0.02
(0.42)c
0.03 ± 0.01 0.01 ± <0.01
(0.42)c
(0.16)c
0.03 ± <0.01
(0.40)c
0.04 ± 0.03
(0.45)c
0.01 ± 0.01
(0.15)c
1.5
2.5
4
0.04 ± 0.01
(0.33)c
0.07 ± 0.01
(0.55)c
0.01 ± 0.01
(0.12)c
0.07 ± <0.01 0.11 ± <0.01 0.02 ± 0.01
(0.36)c
(0.54)c
(0.10)c
0.11 ± 0.01
(0.28)c
0.20 ± <0.01 0.05 ± <0.01
(0.57)c (0.15)c
6
0.14 ± 0.01
(0.24)c
0.31 ± <0.01 0.14 ± <0.01
(0.54)c
(0.23)c
9
0.12
0.4
0.25
(0.16)c
(0.52)c
(0.32)c
a2 mM IPP, 12 mM
LꢀTrp, 0.1 mg/mL BSA, 10 mM MgCl2, 1 µM IDIꢀ1, 40 µM 4ꢀDMATS pD 7.8, 30
b
c
oC, reaction volume 2.5 mL. 9h is not average. Average of two independent assays. Normalized mole
fractions for d1ꢀ, d2ꢀ, and d3ꢀIPP.
ACS Paragon Plus Environment

Page 11 of 29
Journal of the American Chemical Society
1
2
3
4
Table 4. Timeꢀcourse for incorporation of deuterium into IPP by IDIꢀ2.a,b
5
6
7
8
9
Time Conversion
IPPꢀH+
(Mole fraction)
(h)
(%)
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
m/z 245
m/z 246
m/z 247
m/z 248
d0
d1
d2
d3
1
4
5 ± 2
22 ± 6
35 ± 8
47 ± 10
69 ± 11
0.81 ± 0.09
0.74 ± 0.03
0.67 ± 0.04
0.53 ± 0.03
0.15 ± 0.04
0.02 ± 0.01
(0.11)c
0.17 ± 0.06 0.01 ± <0.01
(0.85)c
(0.04)c
0.04 ± 0.01
(0.16)c
0. 20 ± 0.02
(0.80)c
0.01 ± 0.01
(0.04)c
6
0.06 ± <0.01
(0.18)c
0.24 ± 0.02
(0.70)c
0.04 ± 0.01
(0.12)c
8
0.07 ± 0.02
(0.15)c
0.34 ± 0.01
(0.72)c
0.06 ± 0.01
(0.13)c
12
0.04 ± 0.01
(0.05)c
0.68 ± 0.01
(0.80)c
0.13 ± 0.03
(0.15)c
a2 mM IPP, 10 mM
LꢀTrp, 0.5 mM FMN, 5 mM Na2S2O4, 0.1 mg/mL BSA, 10 mM MgCl2, 10 nM
spIDIꢀ2, 10 µM 4ꢀDMATS pD 7.8, 30 C, reaction volume 5.2 mL. Average of two independent
assays. cNormalized mole fractions for d1ꢀ, d2ꢀ, and d3ꢀIPP.
o
b
The results for incorporation of deuterium into IPP during isomerizations catalyzed by IDIꢀ1 and
IDIꢀ2 show that IDIꢀ1 and IDIꢀ2 catalyze up to five interconversions of IPP and DMAPP before
releasing the substrate/product from the active site. While there is some scatter in the data for the
isotopologues at early reaction times where the mole fraction of unlabeled IPP is high, the trend in the
patterns for incorporation of deuterium is evident. Deuterium was incorporated into IPP early during the
incubations with IDIꢀ1 or IDIꢀ2 and the amount of labeled IPP slowly increased as the reactions
proceeded. The earliest samples from incubations of IDIꢀ1 and IDIꢀ2 with IPP contained a mixture of
d1ꢀIPP, d2ꢀIPP, and d3ꢀIPP isotopologues where d2ꢀIPP was the predominant isotopologue for both
enzymes. As the reactions progressed, the mole fraction of d1ꢀIPP decreased relative to d2ꢀIPP and d3ꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 29
IPP (see amounts in parenthesis, Tables 3 and 4), the mole fraction of d2ꢀIPP did not change
appreciably, and the mole fraction of d3ꢀIPP increased. The distribution of isotopologues clearly
demonstrates that multiple protonationꢀdeprotonation events occur between binding and release of IPP
by the enzyme. The relative values of the mole fractions for the d1, d2, and d3 isotopologues of IPP and
DMAPP from incubations with IDIꢀ1 and IDIꢀ2 did not change substantially until more than ~25% of
IPP was consumed. Afterward the mole fractions of d2ꢀIPP, and d3ꢀIPP isotopologues increased.
Analysis of deuterium incorporation. The deuterium labeling patterns for each turnover of IDIꢀ1
and IDIꢀ2 can be determined by extrapolation of the mole fractions of deuteriumꢀcontaining
isotopologues measured during the time course of the reaction back to t = 0. Our data show that the
ratio of isotopologues did not change appreciably during the first ~25% of the reaction, where the
amount of labeled IPP was low to relative to unlabeled IPP. We selected the mole fractions of IPP and
DMAPP isotopologues determined at ~23% consumption of IPP for IDIꢀ1 and IDIꢀ2, respectively, as
representative of the extrapolated values for our calculations. At this point in the time courses for IDIꢀ1
and IDIꢀ2, the reactions had proceeded to the point where the mole fractions of the isotopologues could
be measured reliably, but the IPP pools still mostly consisted of d0ꢀIPP – 88% for IDIꢀ1 and 74% for
IDIꢀ2.
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
Consider the distributions of deuterium isotopologues of IPP and DMAPP resulting from the
first turnover of IDIꢀ1, assuming that each isomerization reaction involves addition of a deuterium to the
carbonꢀcarbon double bond in IPP or DMAPP. At ~23% conversion of IPP to DMAPP, as measured by
formation of DMAT, 12% of the remaining IPP contained deuterium, as measured by mass
spectrometry. This corresponds to incorporation of label into 9% of the IPP [12% x (77/100)] present at
the beginning of the reaction. Thus, at 23% conversion to DMAPP, 32% (9% + 23%) of the IPP present
at t = 0 had been bound to IDIꢀ1, deuterated, and released as IPP (28%) or DMAPP (72%). These
percentages should closely approximate the values for each turnover. In a similar manner for IDIꢀ2,
26% of the remaining IPP contained deuterium, which corresponds to incorporation of label into 20% of
the IPP [26% x (77/100)] present at the beginning of the reaction. Thus, at 23% conversion to DMAPP,
ACS Paragon Plus Environment

Page 13 of 29
Journal of the American Chemical Society
43% (20% + 23%) of the IPP present at t = 0 had been bound to IDIꢀ2, deuterated, and released as IPP
(47%) or DMAPP (53%). The mole fractional composition (χ) of the deuterated isotopologues produced
during each turnover of IDIꢀ1 and IDIꢀ2, calculated from values for the distribution of d1, d2, and d3
isotopologues of IPP and DMAPP in Tables 1ꢀ4. The respective percentages of deuterated IPP and
DMAPP formed at ~23% conversion of IPP to DMAPP, are shown in Table 5.
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
Table 5. Mole fractions (χ) of IPP and DMAPP isotopologues formed during each turnover of d0ꢀIDIꢀ
1·IPP and d0ꢀIDIꢀ2·IPP.
χIDIꢀ1
χIDIꢀ2
d1ꢀIPP
d2ꢀIPP
0.09
0.16
0.08
0.37
d3ꢀIPP
0.03
0.56
0.14
0.02
0.02
0.35
0.14
0.04
d1ꢀDMAPP
d2ꢀDMAPP
d3ꢀDMAPP
The simplest scenario for formation of the isotopologues of IPP and DMAPP during a turnover
of IDIꢀ1·d0ꢀIPP and IDIꢀ2·d0ꢀIPP is the concerted mechanism presented in Scheme 4. Other possibilities
are considered in the Discussion. While our experiments do not allow us to determine the relative rates
for proton exchange with bulk solvent versus isomerization, the extensive incorporation of multiple
deuteriums into IPP and DMAPP during each turnover demonstrates that a large fraction of the protons
released during isomerization are not available for isomerization beyond the initial conversion of IPP to
DMAPP. Thus, isomerization reactions where a deuteron is introduced and a proton is lost (+D+/ꢀH+)
are essentially irreversible. These steps include removal of a proton from C2 during the initial
isomerization of d0ꢀIPP to d1ꢀDMAPP and removal of protons from the Eꢀmethyl group in DMAPP
during subsequent DMAPP to IPP isomerizations. At 23% conversion of d0ꢀIPP, steps involving
dissociation of the enzymeꢀligand complexes are also essentially irreversible. The mole fractions of the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 29
isotopologues in Table 5 correspond to the amount of each that is released from IDI during repeated
isomerizations. The mole fraction of the original substrate (d0ꢀIPP) that passes through each irreversible
step in the sequence of isomerizations can be calculated by subtracting the mole fractions of products
released from IDI·IPP/DMAPP after the previous irreversible step from the mole fraction of
IDI·IPP/DMAPP that passed through the previous irreversible step. Thus, for IDIꢀ1·d0ꢀIPP ꢀ IDIꢀ1·d1ꢀ
DMAPP, χ = 1.00 – 0 = 1.00 for IDIꢀ1·d1ꢀDMAPP; for IDIꢀ1·d1ꢀDMAPP ꢀ IDIꢀ1·d2ꢀIPP, χ = 1.00 –
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
(0.56 + 0.09) = 0.35 for IDIꢀ1·d2ꢀIPP; and for IDIꢀ1·d2ꢀDMAPP ꢀ IDIꢀ1·d3ꢀIPP, χ = 0.35 – (0.16 +
0.14) = 0.05 for IDIꢀ1·d3ꢀIPP (Figure 4A). A similar analysis for IDIꢀ2 gives the values shown in Figure
4B.
Scheme 4. Formation of IPP and DMAPP isotopologues during each turnover of d0ꢀIDIꢀ1·IPP (part A)
and IDIꢀ2 (part B). Irreversible and reversible steps within the enzyme ligand complex are indicated by
+D+/ꢀH+ and +D+/ꢀD+, respectively. Ligand release steps are indicated by blue arrows. Mole fractions
(χ) are given for products formed irreversibly.
ACS Paragon Plus Environment

Page 15 of 29
Journal of the American Chemical Society
The formation of d1, d2, and d3 isotopologues of IPP and DMAPP clearly demonstrates that rates
1
2
3
4
5
6
7
8
for isomerization within and dissociation of the enzyme·IPP/DMAPP complexes are competitive. The
mole fractions for IPP and DMAPP isotopologues shown in Scheme 4 reflect how newly bound IPP
partitions during each turnover. As shown in Scheme 5, IDI·d1ꢀDMAPP partitions in three different
9
directions, two of which – to IDI + d1ꢀDMAPP and to IDI·d2ꢀIPP – are irreversible. Thus, χd ꢀDMAPP and
1
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
χIDI·d ꢀIPP are directly related to k1 and k2, and χd ꢀDMAPP/χIDI·d ꢀIPP = k1/k2. Since the E-methyl group of
2
1
2
d1-DMAPP only has two hydrogen atoms, the intrinsic value of k2 for unlabeled DMAPP is roughly
50% greater than the corresponding rate constant for deprotonation of d1-DMAPP. Furthermore,
assuming that the deuteration of DMAPP does not significantly alter the rate of product
dissociation (i.e., k1) and the secondary isotope effect on k2 is near unity, it follows that k1 ~ k2 for
IDI-1·DMAPP and k1 ~ 0.4 k2 for IDI-2·DMAPP. Thus, the rates for dissociation (k1) and isomerization
(k2) are similar, and the lifetimes of the IDI·IPP/DMAPP complexes can be estimated from turnover
times for E. coli IDIꢀ1 (τ = 12 s based on kcat = 0.08 sꢀ1) and S. pneumonia IDIꢀ2 (τ = 1 s based on kcat
=
1 sꢀ1).
Scheme 5. Partitioning of IDI·d1ꢀDMAPP during each turnover.
IDI·d1ꢀDMAPP isomerizes to IDI·d2ꢀIPP (k2) by loss of a proton and to IDI·d1ꢀIPP (k3) by loss
of a deuteron. The mole fraction ratios χIDI·d ꢀIPP/χd ꢀIPP = 3.9 for IDIꢀ1 and χIDI·d ꢀIPP/χd ꢀIPP = 7.1 for
2
1
2
1
IDIꢀ2 strongly suggest a primary kinetic isotope effect (KIE), kH/kD, represented by relative rates for
conversion of IDI·d1ꢀDMAPP to IDI·d2ꢀIPP (k2) and IDI·d1ꢀIPP (k3). Since removal of hydrogen is
favored statistically over deuterium by a factor of two, kH/kD = k2/2k3. Unfortunately, the equilibrium
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 29
between IDI·d1ꢀDMAPP and IDI·d1ꢀIPP prevents us from determining unique relative values for k1, k2
and k3. Relative values for rate constants shown in Scheme 5 were estimated from simulations using the
mole fractions of d1ꢀDMAPP, d1ꢀIPP, and IDI·d2ꢀIPP in Schemes 4A and 4B along with different values
for k2/k3 (Figures S62ꢀ63). None of the values for k1–k5 were unreasonably large or small within the
range 2 < k2/k3 < 3.5 for IDIꢀ1 and 2 < k2/k3 < 4 for IDIꢀ2, but outside of these ranges one or more of
the rate constants rapidly increased or decreased.
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
Primary deuterium kinetic isotope effect studies. We measured primary deuterium KIEs (kH/kD)
for removal of a proton/deuteron from a ꢀCDH2 group in the elimination products produced during
solvolysis of two extensivelyꢀstudied systems, cumyl trifluoroacetate (1ꢀOTFA) and 2ꢀmethylꢀ2ꢀchloroꢀ
4ꢀ(4ꢀmethyoxyphenyl)butane (3ꢀCl) under solvolysis conditions reported by Richard and coworkers
(Scheme 6).39,40 The elimination products were separated from unreacted starting material and
substitution products by chromatography. Fractions containing elimination products were combined,
and the relative amounts of d1 and d2 isotopologues were determined by GCꢀMS. During preliminary
experiments, we discovered that the relative intensities of mass spectral peaks characteristic of the
isotopologues changed when the chromatographic peak was sampled at different stages of elution,
suggesting the isotopologues partially resolved during chromatography. To address this issue, we
collected ion currents in 0.1 m/z increments at 500 ms at intervals during the entire elution of the
chromatographic peak and summed the ion currents for each m/z increment.
ACS Paragon Plus Environment

Page 17 of 29
Journal of the American Chemical Society
OTFA
CH2D
OTFE
CH2D
DH2C
CH2
DH2C
CHD
DH2C
DH2C
1
2
3
3:1 v/v
AcN:TFE
[2H2]-1-OTFA
[2H2]-2
[2H]-2
[2H2]-1-OTFE
4
5
6
7
8
Cl
HO
TFEO
CH2D
CH2D
CH2D
CH2D
CH2D
CH2D
H3CO
[2H2]-3-Cl
1:1 v/v
TFE:aq 1M NaClO4
H3CO
H3CO
[2H2]-3-OH
[2H2]-3-OTFE
9
CH2
CH2D
CHD
CH2D
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
H3CO
H3CO
[2H2]-4
[2H]-4
CH2D
CH2D
H3CO
[2H2]-5
Scheme 6. Solvolysis of 1-OTFA and 3-Cl.
KIEs for the elimination reactions leading to 2 and 4 were determined from the relative
intensities of peaks characteristic of their isotopologues. The molecular ion (M) for unlabeled 2 (m/z =
118) was accompanied by peaks at Mꢀ1 and Mꢀ3, which complicated analysis of peak intensities for the
M and M+1 (13C) peaks of the d1 and d2 isotopologues. However, the Mꢀ15 region (loss of a methyl
group) was simpler with a strong peak at Mꢀ15 and a much smaller peak at Mꢀ1. This region in the
spectrum for d0ꢀ and d1ꢀαꢀmethylstyrene (loss of CH2D) had overlapping peaks at m/z 103, 104, and
105 for the C8H7 and C8H6D fragments and their 13C isotopologues. Their relative intensities were
determined from total ion current versus m/z plots by curve fitting as described in SI (Figure S35). A
similar analysis was performed for the d1 and d2 isotopologues of olefin 4 based on their overlapping M
and M+1peaks at m/z 177, 178, and 179 (Figure S36). Primary KIEs calculated from the relative
H
D
intensities of peaks in the Mꢀ15 cluster for 2 and the M cluster for 4 were k2 /k2 = 2.96 ± 0.05 and
k4 /k4D = 2.93 ± 0.18, respectively.
H
Discussion
IDIꢀ1 and IDIꢀ2 are convergently evolved enzymes that catalyze isomerization of IPP to
DMAPP by a protonationꢀdeprotonation reaction. A large body of work suggests that catalysis by both
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 29
enzymes proceeds by a stepꢀwise mechanism through a tertiary carbocationic intermediate or by a
concerted asynchronous mechanism through a late transition state with substantial carbocationic
character.18,19, 24, 27, 30,34,36,37
1
2
3
4
5
6
7
8
9
The Xꢀray structures of IDIꢀ1 complexed with a molecule of an amino analogue of IPP (NIPP), a
transition state analogue for the carbocation (Figure 3),25 and IDIꢀ2 complexed with IPP (Figure 4)30 are
compatible with a protonationꢀdeprotonation mechanism. In both enzymes, the ligands are bound in a
pocket formed by highly conserved aromatic residues oriented to stabilize the formation of positive
charge at C3 in the isopentenyl moiety through πꢀcation interactions. In IDIꢀ1, an acidic glutamate
(E116) and a basic cysteine (C67), are located on opposite faces of the isopentenyl group for
protonation of reꢀface of the double bond and removal of the proꢀR hydrogen at C2, consistent with the
antarafacial stereochemistry for isomerization (see Scheme 1). In IDIꢀ2, the zwitterionic tautomer of
fully reduced FMN is both the acid and the base catalyst for the protonation/deprotonation steps.30,36
The isopentenyl moiety in IPP is located in a plane ~3.5 Å above FMNH2. The proꢀR hydrogen at N5 in
FMNH2 forms a hydrogen bond with the backbone carbonyl group of M67 and the proꢀS hydrogen is
positioned to protonate the siꢀface of the C3ꢀC4 double bond. The proꢀR hydrogen at C2 in IPP is
positioned to transfer to the flavin cofactor, consistent with the suprafacial stereochemistry seen for IDIꢀ
2. Q160, located on the opposite face of IPP, provides dipolar stabilization of developing positive
charge in IPP.
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
ACS Paragon Plus Environment

Page 19 of 29
Journal of the American Chemical Society
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
W161
C67
SH
Mg2+
Oδ
-
-
N
H
δ O
δ-
P
NH
-O
O
O
+
-
-
δ O
Oδ
Y104
HO
E116
OH
H69
HN
O
Zn+
N
Figure 3. Active site of Escherichia coli IDIꢀ1·NIPP.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 29
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
H11
HN
Q160
O
N
W225
H2N
+
K193
NH3
N
H
R
Mg2+
O
N
O
N
O
P
O
P
N
H
N
O
-O
O
H H
O
O
O
NH
N
H
~
N
N
~
H155
H
M67
Figure 4. Active site of Sulfolobus shibatae IDIꢀ2·IPP/DMAPP.
When the isomerizations catalyzed by IDIꢀ1 and IDIꢀ2 were run in D2O buffer, one of the
hydrogens at C2 and those at C4 in IPP and the Eꢀmethyl hydrogens in DMAPP were rapidly replaced
by deuterium.21,22 This pattern is consistent with the regioꢀ and stereochemical preferences for the
protonationꢀdeprotonation reactions catalyzed by the enzymes. In related experiments, Richard and
coworkers discovered that IDIꢀ1 catalyzes at least two isomerizations ꢀ IPP to DMAPP and DMAPP
back to IPP – during each turnover.23 In the current study, we found that hydrogen atoms in IPP and
ACS Paragon Plus Environment

Page 21 of 29
Journal of the American Chemical Society
DMAPP were replaced sequentially by deuterium in up to five isomerization reactions during each
turnover, which gave d1, d2, and d3 isotopologues of both IPP and DMAPP. The lifetimes of IDIꢀ
1·IPP/DMAPP (τ = ~12 s) and IDIꢀ2·IPP/DMAPP (τ = ~1 s) are relatively long, and the rates of product
release and additional isomerization are similar for each isomerization within the IDI·IPP/DMAPP
complexes.
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
The reactions in Schemes 4A and 4B show the sequence of reactions for formation of deuterium
isotopologues by a concerted asynchronous mechanism, where each time a deuterium is added to IPP or
DMAPP, the compound is converted to its isomer. In this scenario d1ꢀDMAPP is formed during the first
isomerization and is the immediate precursor for d1ꢀIPP and d2ꢀIPP. A more complex stepꢀwise
mechanism is shown in Scheme 7, where the d1ꢀcarbocation generated from unlabeled IPP can give
unlabeled IPP by removal of the deuterium at C4 or d1ꢀIPP or d1ꢀDMAPP by removal of a hydrogen
from C2 or C4, respectively. Addition of another deuterium, to d1ꢀIPP or d1ꢀDMAPP is required to give
d2ꢀIPP.
.
.
E
E
PPO
-D+
d1-DMAPP
d1-IPP
+D+
-H+
-H+
.
PPO
E
.
PPO
PPO
E
D
D
D
+D+
+D+
-D+
-D+
-H+
-H+
D
D
D
.
.
.
.
PPO
-D+
PPO
E
PPO
E
E
PPO
E
D
D
D
D
D
+D+
-H+
d2-DMAPP
d2-IPP
.
E
PPO
D
.
PPO
E
D
D
d1-IPP
d2-IPP
Scheme 7. Formation of d1ꢀ and d2ꢀisotopologues by a stepꢀwise mechanism.
In a concerted asynchronous mechanism, the only direct route to d1ꢀIPP and d2ꢀIPP is from d1ꢀ
DMAPP, where a primary KIE for removing a deuterium/hydrogen from C4 should favor formation of
the d2ꢀisotopologue. In contrast, both d1ꢀIPP and d1ꢀDMAPP can be formed directly from the initial d1ꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 29
carbocationic intermediate in a stepꢀwise mechanism. d2ꢀIPP can only be formed by further
isomerization of d1ꢀIPP or d1ꢀDMAPP. In this mechanism, one would anticipate that each turnover of
unlabeled IPP would result in more d1ꢀIPP than d2ꢀIPP, in contrast to our observations. However, this
result assumes that rotation of the deuteromethyl group in the d1ꢀcarbocation is sufficiently fast relative
to deprotonation to interconvert the d1ꢀC+D and d1ꢀC+H rotomers (Scheme 8). If the lifetime of d1ꢀC+D is
substantially less than that for deuteromethyl rotation, formation of d1ꢀIPP does not compete with d1ꢀ
DMAPP and the predicted labeling patterns for formation of the IPP and DMAPP isotopologues in
Schemes 4A and 7 become equivalent.
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
D+
H+
CH3
H
CH3
H
H
H
H
H
H
H
PPO
PPO
d0-IPP
d1-IPP
+
+
H
D
CH3
H
CH3
H
D
H
H
H
PPO
PPO
H
H
d1-C+
d1-C+
D
H
D
H
H
CH3
H
PPO
d1-DMAPP
H+
Scheme 8. Deuterium addition and hydrogen removal from conformational isomers of the d1ꢀ
carbocation.
Based on these considerations, our results are consistent with an asynchronous concerted
mechanism or a stepꢀwise mechanism, where the lifetime of the putative tertiary carbocation is less than
that required for the Eꢀmethyl group to rotate by 120o. Although the barrier for rotation of a methyl
group attached to the trigonal center in a tertiary carbocation is not known, one would expect it to be ~2ꢀ
ACS Paragon Plus Environment

Page 23 of 29
Journal of the American Chemical Society
3 kcal/mol based on the ~2ꢀ3 kcal/mol barrier for methyl rotation in 2ꢀmethylpropene,42ꢀ45
corresponding to a lifetime (τ) of ~10ꢀ11 s for individual rotomers. These values are similar to the
lifetimes for a variety of tertiary carbocations in aqueous solvents reported by Richard and coworkers,
including the dimethylphenyl (cumyl) carbocation,39 where τ ~ 10ꢀ10 s, represents carbocationꢀsolvent
1
2
3
4
5
6
7
8
9
combination, and a dimethylalkyl tertiary alkyl substituted carbocation,40 where τ ~ 10ꢀ11 s represents
rotation of a solvent molecule into a reactive conformation for combination. In both systems, the
Richard group40,45 found variations in the relative amounts of elimination and substitution products,
which led them to conclude that substitution proceeded through a carbocationic intermediate, while
elimination proceeded through a concerted process where the leaving group facilitated removal of the
proton.
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
The primary hydrogen/deuterium KIEs we measured for removal of a methyl proton in the
elimination reactions of 1ꢀOTFA and 3ꢀCl, kH/kD = 2.9, were identical within experimental error. While
we were not able to extract unique values for the k2/k3 ratio during isomerization of IDI·d1ꢀDMAPP to
IDI·d1ꢀIPP and IDI·d2ꢀIPP (Scheme 5), the KIEs measured for solvolytic elimination fall in the middle
of the range of reasonable simulated values and are consistent with similar concerted mechanisms for
hydrogen removal in the solvolytic elimination and enzymeꢀcatalyzed isomerization reactions.
A concerted asynchronous mechanism for the enzyme catalyzed isomerization of IPP to DMAPP
that bypasses a highly electrophilic carbocationic intermediate has obvious advantages for preventing
alkylation of nucleophilic residues in the active site. Both IDI isoforms catalyze highly regioꢀ and
stereoselective isomerizations. For IDIꢀ2, only one of the hydrogens at C2 and both C4 hydrogens in
IPP and all of the Eꢀmethyl hydrogens in DMAPP are exchanged, even upon prolonged incubation in
D2O. In contrast, IDIꢀ1 is more promiscuous and all of the hydrogens in IPP and DMAPP, except those
at C1 are exchanged. Thus, IDIꢀ1 catalyzes proton exchange when the substrate is bound in different
conformations that allow access of the active site glutamate and cysteine resides to hydrogens at C2, C4,
and C5 in IPP and DMAPP without concomitant alkylation of active site nucleophiles or addition of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 29
water. In contrast, epoxide and diene mechanismꢀbased inhibitors, which are more readily protonated
than carbonꢀcarbon double bonds and are likely to generate protonated intermediates, readily alkylate
C67 and E116. These observations are consistent with an asynchronous concerted mechanism for
isomerization of the less reactive double bond in IPP and a stepwise protonation/alkylation mechanism
for the epoxide and diene inhibitors. A concerted mechanism for IDIꢀ2 requires that the zwitterionic
form of FMNH2 serves as both the acid and the base for protonation/deprotonation. The Xꢀray structure
of IDIꢀ2·IPP/DMAPP (Figure 4), shows protonated N5 and the carbonyl oxygen at C4 are positioned to
facilitate a concerted [1.3] hydrogen shift through a concerted asynchronous sixꢀmembered transition
state. Having both the acidic and basic functional groups located in each molecule may contribute to the
enhanced stereoselectivity of IDIꢀ2.
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
Conclusions
IDIꢀ1 and IDIꢀ2 are convergently evolved enzymes with different active site architectures. While
turnover is rather sluggish, both enzymes catalyze up to five isomerization reactions each time IPP
binds. Analysis of the patterns for incorporation of multiple deuterium atoms during each turnover
reveals a marked preference formation of d2ꢀIPP over d1ꢀIPP, which is consistent with the kinetic
isotope effects for removal of a hydrogen in related elimination reactions of tertiary alkyl derivatives.
This behavior is also consistent with the extensive studies by Richard and coworkers indicating that
solvolytic elimination reactions of tertiary alkyl derivatives follow concerted asynchronous pathways
that are parallel to competing stepꢀwise solvolytic substitution reactions. We conclude that IDIꢀ1 and
IDIꢀ2 catalyze concerted asynchronous double bond isomerizations, which reduce the possibility of
substrate inactivation by alkylation of their active site residues.
ACS Paragon Plus Environment

Page 25 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
Associated Content
Supporting Information
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
The Supporting Information is available free of charge on the ACS Publications website at DOI:
Synthesis and solvolysis experiments, including NMR spectra.. Measurement of kinetic isotope
effects. Enzyme preparation and deuterium incorporation assays. Mass spectra for isotopologues
and kinetic isotope effect determinations. Analysis of peak areas for kinetic isotope experiments.
Simulations of time courses for partitioning of IDIꢁd1-DMAPP.
Author Information
Corresponding Author
*poulter@chemistry.utah.edu
Orcid
Syam Sundar Neti: 0000-0003-3343-8642
Jian-Jung Pan: 0000-0001-6859-7470
C. Dale Poulter: 0000-0001-7682-3095
Acknowledgements: We thank Dr. James Cox, Dr. Krishna Parsawar and Dr. James Muller for the
assistance with mass spectrometry measurements; Dr. Jeffery Rudolf and Dr. Seoung Choi for 4-
DMATS and IDI-1 plasmids; and Ms. Iris Guo for performing simulations of the time courses for
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 29
partitioning of IDI·d1-DMAPP. This work was supported by NIH grant GM 25521 from the
1
2
3
institute for General Medical Sciences (CDP).
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
References
(1) Kuzuyama, T.; Seto, H. Nat. Prod. Rep. 2004, 20, 171-183.
(2) Dictionary of Natural Products (http://dnp.chemnetbase.com) Version 25.1, 2016, accessed
07/30/2016.
(3) Chaykin, S.; Law, J.; Phillips, A. H.; Tchen, T. T.; Bloch, K. Proc. Natl. Acad. Sci. 1958, 44, 998-
1004.
(4) Kuzuyama, T. Biosci. Biotechnol. Biochem. 2002, 66, 1619 – 1627.
(5) Rohmer, M. Pure Appl. Chem. 2003, 75, 375 – 387.
(6) Anderson, M. S.; Muehlbacher, M.; Street, I. P.; Proffitt, J.; Poulter, C. D. J. Biol. Chem. 1989, 264,
19169-19175.
(7) Hahn, F. M.; Hurlburt, A. P.; Poulter, C. D. J. Bacteriol. 1999, 181, 4499-4504.
(8) Rohmer, M. A mevalonate-independent route to isopentenyl diphosphate, in Comprehensive
Natural Products Chemistry (Cane, D., Ed.) 1999, 45-67, Pergamon Press, New York.
(9) Agranoff, B. W.; Eggerer, H.; Henning, U.; Lynen, F. J. Am. Chem. Soc. 1958, 81, 1254-1255.
(10) Kaneda, K.; Kuzuyama, T.; Takagi, M.; Hayakawa, Y.; Seto, H. Proc. Natl. Acad. Sci. USA 2001,
98, 932-937.
(11) Hahn, F. M.; Xuan, J. W.; Chambers, A. F.; Poulter, C. D. Arch. Biochem. Biophys. 1996, 332, 30-
34.
(12) Amashita, S.; Hemmi, H.; Ikeda, Y.; Nakayama, T.; Nishino, T. Eur. J. Biochem. 2004, 271, 1087-
1093.
(13) Laupitz, R.; Hecht, S.; Amslinger, S.; Zepeck, F.; Kaiser, J.; Richter, G.; Schramek, N.;
Steinbacher, S.; Huber, R.; Arigoni, D.; Bacher, A.; Eisenreich, W.; Rohdich, F. Eur. J.Biochem. 2004,
271, 2658-2669.
(14) Agranoff, B. W.; Eggerer, H.; Henning, U.; Lynen, F. J. Biol. Chem. 1960, 235, 326–332.
ACS Paragon Plus Environment

Page 27 of 29
1
Journal of the American Chemical Society
(15) Carrigan, C. N.; Poulter, C. D. J. Am. Chem. Soc. 2003, 125, 9008-9009.
2
3
4
5
(16) Popjak, G.; Conforth, J. W. Biochem. J. 1966, 101, 553-568.
6
7
8
9
(17) Kao, C.-l.; Kittleman, W.; Zhang, H.; Seto, H.; Liu, H.-w. Org. Lett. 2005, 7, 5677-5680.
(18) Reardon, J. E.; Abeles, R. H. Biochemistry 1986, 25, 5609-5616.
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
(19) Muehlbacher, M.; Poulter, C. D. Biochemistry 1988, 27, 7315-7328.
(20) Calveras, J.; Thibodeaux, C. J.; Mansoorabadi, S. O.; Liu, H. ChemBioChem 2012, 13, 42 46.
(21) Street, I. P.; Christensen, D. J.; Poulter, C. D. J. Am. Chem. Soc. 1990, 112, 8577-8578.
(22) Barkley, S. J.; Desai, S. B.; Poulter, C. D. Org. Lett. 2004, 6, 5019-5021.
(23) Jonnalagadda, V.; Toth, K.; Richard J. P. J. Am. Chem. Soc. 2012, 134, 6568−6570.
(24) Lee, S.; Poulter, C. D. J. Am. Chem. Soc. 2006, 128, 11545-11550. ꢀ
(25) Wouters, J.; Oudjama, Y.; Barkley, S. J.; Tricot, C.; Stalon, V.; Droogmans, L.; Poulter, C. D. J. Biol.
Chem. 2003, 278, 11903-11908.
(26) Thibodeaux, C. J.; Mansoorabadi, S. O.; Kittleman, W.; Chang, W.; Liu, H. Biochemistry 2008,
47, 2547-2558.
(27) Johnston, J. B.; Walker, J. R.; Rothman, S. C.; Poulter, C. D. J. Am. Chem. Soc. 2007, 129, 7740-
7741.
(28) Sharma, N. K.; Pan, J.-J.; Poulter, C. D. Biochemistry 2010, 49, 6228-6233.
(29) Thibodeaux, C. J.; Chang, W.; Liu, H. J. Am. Chem. Soc. 2010, 132, 9994-9996.
(30) Unno, H.; Yamashita, S.; Ikeda, Y.; Sekiguchi, S.-Y.; Yoshida, N.; Yoshimura, T.; Kusunoki, M.;
Nakayama, T.; Nishino, T.; Hemmi, H. J. Biol. Chem. 2009, 284, 9160-9167.
(31) de Ruyck, J.; Janczak, M. W.; Neti, S. S.; Rothman, S. C.; Schubert, H.; Cornish, R.; Matagne, A.;
Wouters, J.; Poulter, C. D. ChemBioChem 2014, 15, 1452−1458.
(32) Neti, S. S.; Eckert, D. M.; Poulter C. D. Biochemistry 2016, 55, 4229-4238.
(33) Walker, J. R.; Rothman, S. C.; Poulter, C. D. J. Org. Chem. 2008, 73, 726-729.
(34) Rothman, S. C., Helm, T. R., and Poulter, C. D. Biochemistry 2007, 46, 5437-5445.
(35) Wouters, J.; Oudjama, Y.; Stalon, V.; Droogmans, L.; Poulter, C. D. PROTEINS: Structure,
Function, and Bioinformatics 2004, 54, 216–221.
(36) Nagai, T.; Unno, H.; Janczak, M. W.; Yoshimura, T.; Poulter, C. D.; Hemmi, H. Proc. Natl. Acad.
Sci. U.S.A. 2011, 108, 20461-20466.
(37) Heaps, N. A.; Poulter, C. D. J. Am. Chem. Soc. 2011, 133, 19017–19019.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 29
(38) Zhang, D.; Poulter, C. D. Anal. Biochem. 1993, 213, 356-361.
1
2
3
4
5
6
7
8
(39) Richard, J. P.; Amyes, T. L.; Vontor, T. J. Am. Chem. Soc. 1991, 113, 5871-5873.
(40) Toteva, M. M.; Richard, J. P. J. Am. Chem. Soc. 1996, 118, 11434-11445.
(41) Lide, D. R.; Mann, D. E. J. Chem. Phys. 1957, 27, 868-873.
9
(42) Wiberg, K. B.; Martin, E. J. Am. Chem. Soc. 1985, 107, 5035-5041.
(43) Gutowsky, H. S.; Germann, T. C. J. Mol. Spect. 1991, 147, 91-99.
(44) Livingston, R. C.; Grant, D. M.; Pugmire, R. J. J. Chem. Phys. 1973, 48, 1438-1445.
(45) Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc. 1991, 113, 8960-8961.
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
ACS Paragon Plus Environment

Page 29 of 29
Journal of the American Chemical Society
1
2
3
TOC
4
5
6
7
8
9
Mechanistic Studies of the Protonation-Deprotonation
Reactions for Type 1 and Type 2 Isopentenyl
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
Diphosphate:Dimethylallyl Diphosphate Isomerase
Syam Sundar Neti, Jian-Jung Pan, and C. Dale Poulter*
ACS Paragon Plus Environment