An Efficient Chemoenzymatic Synthesis of Coenzyme A and Its Disulfide

Article abstract of DOI:10.1021/acs.oprd.6b00059

We have developed a chemoenzymatic route to coenzyme A (CoASH) and its disulfide that is amenable to gram-scale synthesis using standard laboratory equipment. By synthesizing the symmetrical disulfide of pantetheine (pantethine), we avoided the need to mask the reactive sulfhydryl and also prevented sulfur oxidation byproducts. No chromatography is required in our synthetic route to pantethine, which facilitates scale-up. Furthermore, we discovered that all three enzymes of the CoASH salvage pathway (pantetheine kinase, phosphopantetheine adenyltransferase, and dephospho-coenzyme A kinase) accept the disulfide of the natural substrates and functionalize both ends of the molecules. This yields CoA disulfide as the product of the enzymatic cascade, a much more stable form of the cofactor. Free CoASH can be prepared by in situ S-S reduction.

Full text of DOI:10.1021/acs.oprd.6b00059

Subscriber access provided by GAZI UNIV
Full Paper
An Efficient Chemoenzymatic Synthesis of Coenzyme A and Its Disulfide
Louis M.M. Mouterde, and Jon Dale Stewart
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00059 • Publication Date (Web): 08 Apr 2016
Downloaded from http://pubs.acs.org on April 11, 2016
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 free 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 accessible to all
readers and 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.
Organic Process Research & Development 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
Organic Process Research & Development
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
An Efficient Chemoenzymatic Synthesis of Coenzyme A and Its Disulfide
Louis M. M. Mouterde and Jon D. Stewart*
Department of Chemistry, 126 Sisler Hall, University of Florida, Gainesville, FL 32611 USA
Phone & Fax 352.846.0743, Eꢀmail jds2@chem.ufl.edu
*Author to whom correspondence should be addressed
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 22
1
2
3
4
Table of Contents Graphic
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
2
ACS Paragon Plus Environment

Page 3 of 22
Organic Process Research & Development
1
2
3
Abstract
4
5
6
7
8
We have developed a chemoenzymatic route to coenzyme A (CoASH) and its disulfide that is
amenable to gramꢀscale synthesis using standard laboratory equipment. By synthesizing the
symmetrical disulfide of pantetheine (pantethine), we avoided the need to mask the reactive sulfhydryl
and also prevented sulfur oxidation byꢀproducts. No chromatography is required in our synthetic route
to pantethine, which facilitates scaleꢀup. Furthermore, we discovered that all three enzymes of the
CoASH salvage pathway (pantetheine kinase, phosphopantetheine adenyltransferase and dephosphoꢀ
coenzyme A kinase) accept the disulfide of the natural substrates and functionalize both ends of the
molecules. This yields CoA disulfide as the product of the enzymatic cascade, a much more stable form
of the cofactor. Free CoASH can be prepared by in situ SꢀS reduction.
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
Keywords: Coenzyme A, chemoꢀenzymatic synthesis, disulfide, pantetheine, pantethine
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 22
1
2
3
4
Introduction
5
6
7
8
Coenzyme A (CoASH) is an essential acyl carrier and coꢀsubstrate for many biochemical processes.
In addition to recognizing the acyl moiety of CoA derivatives, cognate enzymes also commonly interact
with the cofactor portion and show little tolerance for structural perturbations.1 One must therefore
supply native CoA intermediates in stoichiometric quantities for such enzymes. This is particularly
relevant for enzymeꢀmediated aldol reactions and Claisen condensations, which are commonly used to
form C-C bonds in many metabolic pathways. Such reactions might also be highly useful in chemical
synthesis; unfortunately, enthusiasm for this approach must be tempered by the high cost of commercial
CoASH ($2600 / g with >85% purity). To remove this obstacle, we have developed a simple chemoꢀ
enzymatic route to CoASH that can be carried out on gram scales and requires no chromatographic
purification until the final target has been reached. Synthesis of the enzyme substrate requires only five
steps and has a 76% overall yield from commercially available βꢀalanine.
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
Preparative quantities of CoASH have been isolated from microbial sources such as bakers’ yeast 2,3
or Brevibacterium ammoniagenes IFO 12071.4,5 Dried cells of wildꢀtype or mutant B. ammoniagenes
were subsequently used in chemoꢀenzymatic routes since they contained all five of the necessary
CoASH biosynthetic enzymes.6,7 More recently, enzymeꢀassisted strategies have been developed by
several groups for CoASH and its analogs 8ꢀ10 based on fundamental biochemical studies by
Drueckhammer,11,12 Strauss and Begley,13ꢀ15 and Jackowski.16ꢀ20 Our route to CoASH was inspired by
the work of Strauss and Begley and employs the threeꢀenzyme cascade of the CoASH salvage pathway
(Scheme 1). All three steps are ATPꢀdependent and the proteins from Escherichia coli have been cloned
and overexpressed.12,19,21 Initial phosphorylation of 1 by pantetheine kinase (PanK) yields 2, whose
adenylation by phosphopantetheine adenyltransferase (PPAT) provides dephosphoꢀCoA 3. The final
step installs the 2'ꢀphosphate using dephosphoꢀcoenzyme A kinase (DCPK) to yield CoASH. In
principle, all three enzymes could be present simultaneously; however, most PPAT enzymes (including
4
ACS Paragon Plus Environment

Page 5 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the E. coli form) are subject to feedback inhibition by CoASH.18 This makes successive enzymeꢀ
catalyzed reactions more convenient for preparative purposes, an approach pioneered by Wright.9 The
three enzymeꢀcatalyzed steps in CoASH production need little or no optimization; what discourages
their largeꢀscale use is the lack of an economical source for the substrate. One key observation is that
the three enzymes tolerate significant structural alterations at the sulfhydryl moiety and this has allowed
CoASH analogs to be synthesized by the same route.8ꢀ10 Based on this, we hypothesized that the three
enzymes might therefore accept a disulfide dimer of pantetheine (known as pantethine) and
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
simultaneously functionalize both ends, ultimately producing the disulfide of CoASH.
OH
H
H
N
N
HS
OR'
O
O
1
2
R' = H
R' = PO3
O
PanK, ATP
PPAT, ATP
2-
O
O
P O P O
3 R' =
A
O
O
O
H
OH
DCPK, ATP
(A = adenine)
OH
H
H
O
O
N
N
O
HS
O P O P O
A
O
O
O
O
O
O P O
O
OH
Coenzyme A
Scheme 1
We chose the disulfide derivative of 1 as the biocatalytic precursor for several reasons. First, a
disulfide eliminates the two steps needed to install and later remove the sulfhydryl protecting group. In
a sense, our “protecting group” is a second molecule of the target compound. This is an atom
economical strategy that eliminates carrying extraneous mass through the sequence. A disulfide also
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 22
1
2
3
4
diminishes the chances of forming sulfur oxidation byꢀproducts during the synthetic operations. The
5
6
7
8
dimeric intermediates were organicꢀsoluble, which allowed the desired products to be isolated by solvent
extraction with no need for chromatographic purification. This is a significant advantage when scaling
up the chemical synthesis steps. Finally, the commercial price of oxidized CoASH – the product of our
synthetic route – is almost 10ꢀfold greater than that of CoASH itself.22
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
In principle, the biocatalytic precursors (pantetheine 1 or pantethine 8) could also be synthesized
enzymatically or obtained commercially; however, we found it more convenient to prepare the building
block by traditional organic methodologies (Scheme 2). While routes to 1, 8 and analogs are known,8,23
our goals were to optimize each reaction and eliminate chromatographic separations that would hinder
scaleꢀup. We also sought a simple method to isolate the disulfide of CoASH on a preparative scale that
avoided HPLC.
6
ACS Paragon Plus Environment

Page 7 of 22
Organic Process Research & Development
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
Results and Discussion
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To synthesize panthethine 8, βꢀalanine 4 was reacted with a slight excess of Boc2O at 0ºC to afford
Nꢀprotected 5 in 90% yield.24 25 The crude product, isolated by simple solvent extraction, was
,
sufficiently pure to be used directly in the key coupling step with 0.50 equivalents of cysteamine
disulfide (cystamine). Initial attempts using PPh3, I2 and imidazole in CH2Cl2 gave 6 in only 15 – 20%
yield; in addition, the phosphine oxide byꢀproduct had nearly the same mobility on silica gel
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 22
1
2
3
4
5
chromatography as 6. One solution was to synthesize a linear polystyrene with pendant phosphines,26
which was used in place of free PPh3. Unfortunately, because both 6 and the polymeric reagent were
6
7
8
9
soluble in CH2Cl2, postꢀreaction separation remained an issue.27 We achieved smooth coupling of 5 and
cystamine by employing EDCI and HOBt.28 Several aspects of this process were optimized to facilitate
scaleꢀup. In addition to the originallyꢀreported solvent (CH2Cl2), we also screened several other
conditions. DMF at room temperature proved best along with dropwise addition of cystamine. Most
importantly, we eliminated column chromatography by judicious extractions that removed both HOBt
and the carbodiimide byꢀproduct and provided 6 in 95% yield with no need for further purification.
Formation of a symmetrical dimer with no contamination by monoacylation product was confirmed by
the observation of only 8 signals in the 13C NMR spectrum of 6.
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 Boc group was removed quantitatively using TFA to yield ammonium salt 7.29 This was
coupled with Dꢀpantolactone by heating an ethanolic solution at reflux in the presence of excess Et3N as
described by King et al.23 Pantethine 8 was isolated after solvent extraction. The four nonꢀenzymatic
steps proceeded in 76% overall yield on gram scales and no column chromatography was required. We
believe that this route is thus amenable to scaleꢀup. Moreover, this process give an easy access to
pantetheine 1 by TCEP reduction (or other reducing agent as 2ꢀmercaptoethanol).
Baseꢀcatalyzed racemization of Dꢀpantolactone, e.g., when heated in DMF, has been reported.30
Because our conversion of 7 to 8 involved heating an ethanolic solution of Dꢀpantolactone in the
presence of Et3N, we carried out a control reaction to determine whether racemization might have
occurred prior to coupling. Optically pure Dꢀ(ꢀ)ꢀpantolactone was heated at reflux in EtOH overnight in
presence of 5 equivalents of Et3N, which mimicked the reaction conditions employed in the coupling
step. Chiralꢀphase GC analysis of the reaction mixture revealed only a single peak for pantolactone
within the detection limits of the instrument.31 Based on these results, and the 13C NMR spectrum of 8,
8
ACS Paragon Plus Environment

Page 9 of 22
Organic Process Research & Development
1
2
3
4
5
we believe that our reaction product is diastereomerically pure and that no significant racemization took
place.
6
7
8
9
Before attempting the enzymatic cascade with the disulfide, we carried out pilot experiments with
the known substrate 1. The disulfide bond of 8 was reduced quantitatively by a slight molar excess of
tris(2ꢀcarboxyethyl)phosphine (TCEP) in the buffer used for the enzymeꢀcatalyzed reactions (75 mM
TrisꢀCl, 25 mM KCl, 250 mM MgCl2, pH 7.5). The crude product was used directly as a substrate for
the three successive enzymeꢀcatalyzed reactions using PanK, PPAT and DPCK in the presence of excess
ATP to ensure that all conversions proceeded to completion (Scheme 1). This sequence yielded a
sample of CoASH whose HPLC and MS data were identical to those of commerciallyꢀsupplied material.
These reactions were performed on a small scale and quantitative conversion was observed at each step
according to HPLC analysis (Figure 1). As expected, the intermediates and final product were
somewhat unstable in the presence of oxygen, likely because of thiol oxidation.
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
To avoid sideꢀreactions involving the free thiol of 1 and its derivatives, we employed disulfide 8 as
an alternate substrate for the three CoASH salvage enzymes. The limited solubility of pyrophosphate in
the presence of Mg2+ caused precipitation during reactions with PPAT at high substrate concentrations,
which interfered with the enzymatic reaction. Filtration to remove the salt followed by adding
additional PPAT solved this issue. Quantitative conversions were observed for each step (as monitored
by HPLC; Figure 2).
To ensure that all the reactions had proceeded to completion at both ends of the symmetrical
substrates, samples were reduced after each step with TCEP and analyzed by HPLC. In each case, only
a single product was observed, corresponding to the intermediates in the sequence that commenced with
1.
We explored several strategies for purifying the disulfide of CoASH. Reversedꢀphase HPLC, while
effective, had very low capacity that greatly limited its practical scalability. We initially used anionꢀ
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 22
1
2
3
4
5
6
7
8
9
exchange resins developed for protein purification, e.g., MonoꢀQ. These provided high product purities,
but also suffered from limited capacities. Dowex anion exchange resin combined low material cost,
high loading capacity and simple operation. After binding, increasing concentrations of LiCl were used
to separate CoA disulfide 11 from byꢀproduct.32 A linear LiCl gradient yielded individual fractions
containing AMP, ADP, ATP and 11, which allowed recovery of the adenosine byꢀproducts and
regeneration into ATP if desired. For preparative purposes, however, we normally employed a
simplified step gradient in which all the nucleotide byꢀproducts eluted as an initial group and the desired
product 11 was eluted by 600 mM LiCl. CoA disulfide was desalted by absorption on activated
charcoal. Excess LiCl was removed by rinsing the activated charcoal with deionized water, then CoA
disulfide was eluted with a 40% acetone solution containing 0.028% ammonia. The solution was
concentrated under reduced pressure and the residual water was removed by lyophilization to yield the
desired product.32
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 structure of synthetic CoA disulfide was verified by 1H, 13C and 31P NMR along with HRMS.
All data supported the assignment and no extraneous peaks were observed. As expected, the 31P
spectrum showed only three peaks (a singlet at 0.60 ppm and doublets at ꢀ10.27 and ꢀ10.81 ppm),
corresponding to the three phosphate moieties of 11. No other significant phosphorusꢀcontaining
species were observed at the limits of 31P NMR detection, further confirming that the reactions by all
three CoASH salvage enzymes proceeded to completion at both ends of the molecule. The 31P NMR
data showed that anion exchange chromatography had successfully removed all nucleotide monoꢀ, diꢀ or
triphosphosphate byꢀproducts from the enzymeꢀcatalyzed steps. The chemical purity of CoA disulfide
purity was assessed by UV absorbance (ε260 = 14,320) and found to be ca. 82%. For biochemical
purposes, the cofactor can be used directly in this form.33 As HPLC and NMR analysis showed only
CoA disulfide, we attributed the mass impurity to inorganic salts (most likely LiCl carried over from ion
exchange purification). Consistent with this notion, the purity of CoA disulfide could be raised to 90%
10
ACS Paragon Plus Environment

Page 11 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
by a second adsorption / desorption treatment with activated charcoal.34 Similar values for purity were
found by both UV absorbance and elemental analysis. The latter suggested that 11 had been isolated as
a hexakisꢀammonium salt with two waters of hydration.
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
We found that solid CoA disulfide could be stored at room temperature in the presence of oxygen
for at least one month with no evidence of byꢀproduct formation. Coinjecting purified CoA disulfide
with a commercial sample of CoASH gave two peaks using reversedꢀphase HPLC, as expected. When a
slight molar excess of TCEP was added to the sample of CoA disulfide prior to HPLC analysis,35 only
one peak was observed at the tr corresponding to free CoASH (11.47 min). This allows the cofactor to
be stored as the stable disulfide, then reduced in situ as needed for biochemical applications.
Because of its high value, we also prepared the disulfide of 3'ꢀdephosphoꢀCoA using an analogous
sequence of two enzymeꢀcatalyzed steps (but leaving out the third, DCPK). While the synthetic effort
itself was successful, the desired product coꢀeluted with ADP when the standard ionꢀexchange
purification method was employed. Future efforts toward 3'ꢀdephosphoꢀCoA will require a revised
purification method.
Using the optimized methods and standard laboratory equipment, we prepared 5 g batches of 8 in a
single campaign. This was converted to CoA disulfide in lots of 1.89 g per reaction (each divided into
two anion exchange runs) with a purity ≥90%. This has allowed us to generate multiꢀgram amounts of
the cofactor in approximately two weeks. The yield of purified CoA disulfide from precursor 8 was
95%; the overall yield of the sequence starting from βꢀalanine was 72%.
Conclusion
A simple chemoꢀenzymatic route to CoASH has been developed that emphasizes high yield and
scalability. No chromatographic purification steps are required until the final target has been reached.
Because the CoASH salvage pathway enzymes (PanK, PPAT and DPCK) can all accommodate disulfide
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 22
1
2
3
4
5
6
7
8
9
dimer substrates, we could prepare an even higherꢀvalue CoASH derivative, eliminate thiol protection /
deprotection steps and avoid the possibility of sulfur oxidation byꢀproducts. We hope that this chemoꢀ
enzymatic methodology can enable further progress in using CoAꢀdependent enzymes in chemical
synthesis by making the substrates more easily obtainable. While we currently have no plans to increase
the reaction scales further, none of the procedures would appear to be problematic should future needs
dictate further scaleꢀup.
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
Experimental Section
General. Moistureꢀ and O2ꢀsensitive reactions were carried out in flameꢀdried glassware under Ar.
Evaporations were conducted under reduced pressure at temperatures below 35°C unless otherwise
1
noted. H NMR spectra were recorded at 300 MHz at 25°C in the indicated solvent and referenced to
residual protons (CDCl3, 7.26 ppm; CD3OD, 4.87 ppm; D2O, 4.75 ppm). 13C NMR spectra were
recorded at 75 MHz at 25°C in the indicated solvent and referenced to solvent (CDCl3, 77.2 ppm;
CD3OD, 49.2 ppm). 31P NMR spectra were recorded at 121 MHz at 25°C in D2O and referenced to the
solvent deuterium lock frequency.
HPLC methods. Enzymatic reactions were monitored by reversedꢀphase HPLC using a 150 × 4.6
mm Synergi™ HydroꢀRP 80Å column using 50 mM NaPi pH 5.0 and 50 mM NaPi, pH 5.0 / 20%
acetonitrile as Solvents A and B, respectively at a flow rate of 1 mL / min. Initial conditions (3% B)
were maintained for 2.5 min, then a linear increase to 18% B over 5 min was immediately followed by a
linear increase to 28% B over 3.5 min, then a linear increase to 90% B over 10 min. After a 5 min hold
at 90% B, a linear decrease to 3% B over 3 min was followed by a 5 min hold at the initial conditions
(3% B). The eluant was monitored by UV absorbance simultaneously at both 220 and 260 nm (to detect
peptide bonds and adenyl moieties, respectively). These conditions allowed baseline separation of all
relevant analytes in this study (pantetheine 1, tr = 14.64 min; 4’ꢀphosphoꢀpantetheine 2, tr = 7.39 min;
12
ACS Paragon Plus Environment

Page 13 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
dephospho CoASH 3, tr = 13.36 min; CoASH tr = 11.49 min; pantethine 8, tr = 20.86 min; 4’ꢀphosphoꢀ
pantethine 9, tr = 14.67 min; dephosphoꢀCoASH disulfide 10, tr = 16.27 min; CoA disulfide 11, tr =
14.18 min).
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
3ꢀ((tertꢀbutoxycarbonyl)amino)propanoic acid 5. To a stirred solution of 3ꢀaminoꢀpropanoic acid
(5.00 g, 55.7 mmol) in 10% NaOH (50 mL) at 0°C was slowly added 1.2 equiv. of Boc anhydride (15
mL, 66.7 mmol). The resulting mixture was allowed to warm to room temperature. The reaction was
monitored by TLC (10% MeOH/CHCl3) and when complete (3 hr), the mixture was diluted with EtOAc
(50 mL), brought to pH 2 with 1 M HCl, and the layers separated. The organic layer was washed with
water (15 mL) and brine (15 mL), then dried over MgSO4. The drying agent was rinsed with 20 mL of
iceꢀcold hexanes and concentrated in vacuo. The resulting oil was dissolved in a minimum volume of
CH2Cl2 and added slowly to a stirred solution of pentane (100 mL) cooled to 0ºC. The resulting white
solid was collected by filtration (9.99 g, 90%). m.p. 76ꢀ77ºC, lit. m.p. 75ꢀ77ºC.36 1H and 13C NMR data
matched those reported by McCubbin et al.37
diꢀtertꢀButyl(((disulfanediylbis(ethaneꢀ2,1ꢀdiyl))bis(azanediyl))bis(3ꢀoxopropaneꢀ3,1ꢀ
diyl))dicarbamate 6. To a stirred solution of 5 (4.00 g, 21.1 mmol) in DMF (40 mL) at rt was added
HOBt (2.45 g, 21.1 mmol) followed by EDCI (3.71 mL, 21.1 mmol). The reaction was stirred for 30
min, then a solution containing 0.5 equiv. of cystamine (2.38 g, 10.6 mmol) and 2.2 equiv. of Et3N (6.30
mL, 46.4 mmol) in DMF (10 mL) was added dropwise. The mixture was stirred overnight, then diluted
5ꢀfold with water and extracted with Et2O (3 × 200 mL). The combined organic layers were washed
successively with 30 mL 1 M HCl, 10% NaHCO3 and water, then dried over MgSO4. The solution was
concentrated in vacuo and the desired product 6 was obtained as a white solid (4.95 g, 95%). m.p. 102ꢀ
1
103ºC. HRMS: m/z [M + H]+ calcd for C20H38O6N4S2: 495.2306; found: 495.2326. H NMR (300
MHz, CDCl3): δ 6.72 (s, 2H), 5.28 (s, 2H), 3.57 (q, J = 6 Hz, 4H), 3.41 (q, J = 6 Hz, 4H), 2.83 (t, J = 6
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 22
1
2
3
4
5
Hz, 4H), 2.45 (t, J = 6 Hz, 4H), 1.44 (s, 18H) ppm. 13C NMR (75 MHz, CDCl3) : δ 171.9, 155.9, 78.9,
38.2, 37.4, 36.6, 35.9, 28.1 ppm.
6
7
8
9
N,N'ꢀ(Disulfanediylbis(ethaneꢀ2,1ꢀdiyl))bis(3ꢀaminopropanamide) · (CF3CO2H)2 7. Excess
CF3CO2H (37.3 mL, 487.19 mmol) was added to a solution of 6 (4.82 g, 9.74 mmol) in CH2Cl2 (100
mL) at rt. The reaction mixture was allowed to stir for 6 hr, then the solvent was removed in vacuo to
afford the crude product as the bisꢀtrifluoroacetate salt (5.08 g, 99%). HRMS: m/z [M + H]+ calcd for
C10H22O2N4S2: 295.4320; found: 295.1248. 1H NMR (300 MHz, CD3OD): δ 8.31 (s, 6H), 8.01 (s, 2H),
3.46 (t, J = 6 Hz, 4H), 3.18 (t, J = 6 Hz, 4H), 2.79 (t, J = 6 Hz, 4H), 2.61 (t, J = 6 Hz, 4H) ppm. 13C
NMR (75 MHz, CD3OD): δ 172.4, 159.0 (d, 2J = 42 Hz), 116.0 (q, 1J = 283 Hz), 39.6, 38.3, 37.2, 33.0
ppm.
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
(S)ꢀNꢀ((R)ꢀ17,19ꢀDihydroxyꢀ18,18ꢀdimethylꢀ3,12,16ꢀtrioxoꢀ7,8ꢀdithiaꢀ4,11,15ꢀtriazanonadecyl)ꢀ
2,4ꢀdihydroxyꢀ3,3ꢀdimethylbutanamide 8. Three equivalents of Dꢀ(ꢀ)ꢀpantolactone (3.73 g, 28.7
mmol) and 2.5 equiv. of Et3N (3.23 mL, 23.92 mmol) were added to a solution of 7 (5 g, 9.57 mmol) in
EtOH (40 mL). The solution was allowed to stir overnight at reflux. The solvent was evaporated under
reduced pressure and excess Dꢀ(ꢀ)ꢀpantolactone was removed by triturating the residue with diethyl ether
(2 × 50 mL). Solvent was removed in vacuo to afford 8 as an orange solid (5.04 g, 90%). mp = 63ꢀ
1
65°C. HRMS: m/z [M + Na]+ calcd for C10H23O2N4S2: 577.1300; found: 577.2309. H NMR (300
MHz, CD3OD) : δ 3.95 (s, 2H), 3.49 (m, 12H), 2.82 (t, J = 6 Hz, 4H), 2.48 (t, J = 6 Hz, 4H), 0.89 (2s,
12H). 13C NMR (75 MHz, D2O): 174.9, 173.8, 75.6, 68.3, 38.5, 38.0, 36.4, 35.3, 35.2, 20.4, 19.1 ppm.
Overexpression and purification of PanK, PPAT, and DPCK. Individual plasmids that
overexpressed E. coli PanK, PPAT, and DPCK (kindly provided by Prof. Steve Bruner, University of
Florida; also available from Addgene) were used to transform E. coli BL21(DE3) cells with selection for
kanamycin resistance (50 ꢁg / mL) on LB plates. Single colonies of the appropriate strains were used to
14
ACS Paragon Plus Environment

Page 15 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
inoculate 50 mL portions of liquid LB medium containing 50 µg / mL kanamycin and cultures were
grown overnight at 37°C and 250 rpm. Aliquots (40 mL) were diluted into 4 L of the same medium in a
New Brunswick M9 fermenter. Cultures were grown at 37ºC with stirring at 600 rpm and air flow at 6 L
/ min until reaching an optical density of 0.6 at 600 nm. Overexpression of PanK, PPAT, or DPCK was
induced by adding sterile isopropyl βꢀDꢀthiogalactopyranoside (IPTG) to a final concentration of 0.4
mM, along with a solution of 16 g of dextrose in 100 mL of water. The culture was incubated for an
additional 4 h at 30°C, then the cells were harvested by centrifugation (6000 × g at 4ºC for 15 min). The
cell paste was resuspended in 20 mL of cold lysis buffer (50 mM TrisꢀHCl, pH 8.0) and disrupted by
two passages through a French pressure cell at 18,000 psi. The lysate was clarified by centrifugation
(18,000 × g for 20 min at 4ºC) and applied to a 5 mL Ni–NTA agarose column (GE Healthcare)
previously equilibrated with 20 mL of binding buffer (20 mM NaPi, 500 mM NaCl, 20 mM imidazole,
pH 7.4). The column was washed with binding buffer until the A280 value returned to its baseline value,
then the desired protein was eluted by washing with 20 mM NaPi, 500 mM NaCl, 500 mM imidazole,
pH 7.4. The protein solution was concentrated at 4ºC by ultrafiltration and typical yields of purified
enzymes from 4 L cultures were 130 ꢀ 150 mg. Purified proteins were stored in 50% glycerol in 1 mL
aliquots at ꢀ80°C.
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
CoASH synthesis. The reaction mixture contained 20 mM KCl, 10 mM MgCl2, 5 mM ATP (5
ꢁmol), 50 mM Tris–Cl, pH 7.5, and 2.5 mM pantethine 8 (2.5 ꢁmol) in a final volume of 1 mL. Two
equivalents of solid TCEP (3 ꢁmol) were added to produce pantetheine 1 in situ.35 The initial
phosphorylation reaction was initiated by adding 0.17 mg of purified PanK and incubating at rt for 2 hr.
After this time, 0.16 mg of purified PPAT and an additional 5 ꢁmol of ATP was added and incubation
was continued (overnight at rt). Finally, 0.16 mg of purified DPCK along with an additional 5 ꢁmol of
ATP were added and incubation was continued overnight at rt. Samples were analyzed by reversedꢀ
phase HPLC to ensure that conversions had proceeded to completion before the next enzyme was added.
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 22
1
2
3
4
5
The purified product was characterized by MALDI MS: calcd. for C21H36N7O16P3S [M + H]+ 768.1152;
found 768.1242.
6
7
8
9
CoA disulfide synthesis. The reaction mixture contained 25 mM KCl, 250 mM MgCl2, 30 mM
ATP (3.0 mmol), 75 mM Tris–Cl, pH 7.5, and 13 mM pantethine 8 (1.3 mmol, 0.720 g) in a final
volume of 100 mL. The reaction was initiated by adding 3 mg of purified PanK and incubating at rt for
2 hr. After this time, 3 mg of purified PPAT along with an additional 3.0 mmol of ATP was added and
incubation was continued (overnight at rt). The Mg·PPi complex that precipitated during the reaction
was removed by filtration. Finally, 3 mg of purified DPCK and 3.0 mmol of ATP were added and
incubation was continued (overnight at rt). Samples were analyzed by reversedꢀphase HPLC to ensure
that conversions had proceeded to completion before the next enzyme was added.
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
CoA disulfide was purified by injecting half of the crude reaction mixture (50 mL, containing ca. 14
mM CoA disulfide) onto a 650 × 25 mm Dowex 1 × 2 anion exchange column that had been
equilibrated with solvent A (10 mM HCl, 300 mM LiCl). The column was washed with solvent A for
180 min (to elute ATP, ADP and AMP), then the buffer was changed 100% solvent B (10 mM HCl, 600
mM LiCl) for 120 min to elute CoA disulfide. The flow rate was 10 mL / min throughout. Fractions
containing CoA disulfide were then evaporated under vacuum. The material from both runs was
combined. The resulting solid (33 g) was dissolved in a minimum volume of water (200 mL) and
brought to pH ≈ 3, then activated charcoal (20 g) was added. The resulting suspension was stirred for 30
minutes at rt and then the solid was collected by vacuum filtration. The filter cake was washed with 1 L
of deionized water (adjusted to pH ≈ 3) to remove unwanted inorganic salts, then CoA disulfide 11 was
eluted with a 40% (v/v) aqueous acetone solution containing 0.028% ammonia. The combined solutions
containing 11 were concentrated in vacuo to remove acetone, then water was removed by lyophilization
to give the desired product 11 as a pale yellow solid (1.97 g). The adsorption onto activated charcoal
was repeated a second time to provide analyticallyꢀpure 11 (1.89 g, 95%). Purity was assessed by
16
ACS Paragon Plus Environment

Page 17 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
dissolving 5.0 mg of purified 11 in 100 mL of 1 M TrisꢀCl, pH 7.5; this solution yielded A260 = 0.8013.
Based on the extinction coefficient (ε260 = 14,320), the solution contained 4.5 mg of 11 in 100 mL.
Elemental analysis calcd. for C42H70N14O32P6S2 · 6 NH4 ·2 H2O: %C 30.15, %H 5.66, %N 16.74; found
%C 27.42, %H 5.40, %N 15.31. MALDI MS calcd. for C42H70N14O32P6S2 [M + H]+ 1533.2159; found
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
1533.2234. H NMR (300 MHz, D2O) : δ 8.47 (s, 2H), 8.14 (s, 2H), 6.11 (d, J = 6 Hz, 2H), 4.67 (s, 4H),
4.55 (s, 2H), 4.21 (t, J = 3 Hz, 4H), 3.96 (s, 2H), 3.78 (m, 2H), 3.53 (m, 2H), 3.38 (m, 7H), 2.69 (q, J = 6
Hz, 4H), 2.41 (t, J = 6 Hz, 4H), 1.34,1.24,0.84,0.71 (4s, 12H). 31P (121 MHz, D2O): 0.60 (s), ꢀ10.27 (d,
J = 22 Hz), ꢀ10.81 (d, J = 22 Hz). 13C NMR (75 MHz, D2O): 174.6, 173.7, 155.0, 152.4, 148.9, 139.6,
118.2, 86.5, 83.5, 74.1, 73.9, 71.8, 65.5, 58.0, 38.3, 36.5, 35.3, 35.2, 20.8, 18.2 ppm.
Acknowledgements
This work was partially supported by the National Science Foundation (CHEꢀ1111791).
Supporting Information
NMR, HRMS and HPLC data for all intermediates and CoA disulfide (17 pages).
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 22
1
2
3
4
5
6
References and Notes
(1) Polyketide synthases and nonꢀribosomal peptide synthases are exceptions to this rule and often
7
8
9
accept acylated Nꢀacetylcysteamines. Such synthases first catalyze transacylation to a covalentlyꢀ
linked thioester intermediate, which is the actual substrate for subsequent conversions.
(2) Beinert, H.; Von Korff, R. W.; Green, D. E.; Buyske, D. A.; Handschumacher, R. E.; Higgins, H.;
Strong, F. M. J. Biol. Chem. 1953, 200, 385ꢀ400.
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
(3) Reece, M. C.; Donald, M. B.; Crook, E. M. J. Biochem. Microbiol. Technol. Eng. 1959, 1, 217ꢀ
228.
(4) Shimizu, S.; Tani, Y.; Ogata, K. Meth. in Enzymol. 1979, 62, 236ꢀ245.
(5) Shiozaki, S.; Shimizu, S.; Yamada, H. Agric. Biol. Chem. 1984, 48, 2293ꢀ2300.
(6) Shimizu, S.; Esumi, A.; Komaki, R.; Yamada, H. Appl. Environ. Microbiol. 1984, 48, 1118ꢀ1122.
(7) Shimizu, S.; Komaki, R.; Tani, Y.; Yamada, H. FEBS Lett. 1983, 151, 303ꢀ306.
(8) Mandel, A. L.; La Clair, J. J.; Burkart, M. D. Org. Lett. 2004, 6, 4801ꢀ4803.
(9) Nazi, I.; Koteva, K. P.; Wright, G. D. Anal. Biochem. 2004, 324, 100ꢀ105.
(10) Rootman, I.; de Villiers, M.; Brand, L. A.; Strauss, E. ChemCatChem 2010, 2, 1239ꢀ1251.
(11) Martin, D. P.; Drueckhammer, D. G. Biochem. Biophys. Res. Commun. 1993, 192, 1155ꢀ1161.
(12) Mishra, P. K.; Park, P. K.; Drueckhammer, D. G. J. Bacteriol. 2001, 183, 2774ꢀ2778.
(13) Strauss, E.; Begley, T. P. J. Biol. Chem. 2002, 277, 48205ꢀ48209.
(14) Strauss, E.; Kinsland, C.; Ge, Y.; McLafferty, F. W.; Begley, T. P. J. Biol. Chem. 2001, 276,
13513ꢀ13516.
(15) Brand, L. A.; Strauss, E. J. Biol. Chem. 2005, 280, 20185ꢀ20188.
(16) Leonardi, R.; Chohnan, S.; Zhang, Y.ꢀM.; Virga, K. G.; Lee, R. E.; Rock, C. O.; Jackowski, S. J.
Biol. Chem. 2005, 280, 3314ꢀ3322.
(17) Leonardi, R.; Zhang, Y.ꢀM.; Rock, C. O.; Jackowski, S. Prog. Lipid Res. 2005, 44, 125ꢀ153.
18
ACS Paragon Plus Environment

Page 19 of 22
Organic Process Research & Development
1
2
3
4
5
(18) Rock, C. O.; Park, H.ꢀW.; Jackowski, S. J. Bacteriol. 2003, 185, 3410ꢀ3415.
(19) Song, W.ꢀJ.; Jackowski, S. J. Bacteriol. 1992, 174, 6411ꢀ6417.
6
7
8
9
(20) Virga, K. G.; Zhang, Y.ꢀM.; Leonardi, R.; Ivey, R. A.; Hevener, K.; Park, H.ꢀW.; Jackowski, S.;
Rock, C. O.; Lee, R. E. Bioorg. Med. Chem. 2006, 14, 1007ꢀ1020.
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
(21) Geerlof, A.; Lewendon, A.; Shaw, W. V. J. Biol. Chem. 1999, 274, 27105ꢀ27111.
(22) The current price is approximately $200 for 10 milligrams with >85% purity.
(23) King, T. E.; Stewart, C. J.; Cheldelin, V. H. J. Am. Chem. Soc. 1953, 75, 1290ꢀ1292.
(24) Bock, M. G.; Gaul, C.; Gummadi, V. R.; Moebitz, H.; Sengupta, S. WO2012035078, Novartis
A.G, 2012.
(25) Initial studies in which the Fmocꢀprotected βꢀalanine acid chloride was reacted with cystamine
gave poor results, prompting us to substitute Boc protection.
(26) Kwok, M.; Choi, W.; He, H. S.; Toy, P. H. J. Org. Chem. 2003, 68, 9831ꢀ9834.
(27) Adding crossꢀlinks to the polymeric reagent likely would have solved this issue; however, the need
to regenerate reduced phosphine from the polymeric phosphine oxide byꢀproduct discouraged our
further efforts in this area.
(28) Maehara, T.; Kanno, R.; Yokoshima, S.; Fukuyama, T. Org. Lett. 2012, 14, 1946ꢀ1948.
(29) Guaragna, A.; Amoresano, A.; Pinto, V.; Monti, G.; Mastrobuoni, G.; Marino, G.; Palumbo, G.
Bioconjugate Chem. 2008, 19, 1095ꢀ1104.
(30) Dueno, E. E.; Chu, F. X.; Kim, S. I.; Jung, K. W. Tetrahedron Lett. 1999, 40, 1843ꢀ1846.
(31) FID detection using a 0.25mm x 25 m, ChirasilꢀDEX CB column. Racemic pantolactone analyzed
under the same conditions gave baseline separation of the two enantiomers.
(32) Shimizu, S.; Miyata, K.; Tani, Y.; Ogata, K. Agric. Biol. Chem. 1973, 37, 615ꢀ619.
(33) In one example, we have prepared succinylꢀCoA from disulfide CoA. In brief, an excess of TCEP
was added to 100 ꢁL of a 10 mM solution of disulfide CoA dissolved in water. After incubating at
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 22
1
2
3
4
5
6
7
8
9
rt for 45 min, HPLC indicated complete disulfide reduction. The mixture was diluted with 800 ꢁL
of sterile water and cooled to 0ºC, then an excess of finelyꢀground succinic anhydride was added
and the mixture was kept on ice with periodic mixing for 30 min. The reaction was quenched by
adding 100 ꢁL of 1 M NaHCO3 and HPLC indicated compete conversion of CoASH to succinylꢀ
CoA
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
(34) Both UV absorbance and elemental analysis results were consistent with this purity assessment.
(35) Burns, J. A.; Butler, J. C.; Moran, J.; Whitesides, G. M. J. Org. Chem. 1990, 56, 2648ꢀ2650.
(36) Yasui, A.; Douglas, A. J.; Walker, B.; Magee, D. F.; Murphy, R. F. Int. J. Peptide Protein Res.
1990, 35, 301ꢀ305.
(37) McCubbin, J. A.; Maddess, M. L.; Lautens, M. Org. Lett. 2006, 8, 2993ꢀ2996.
20
ACS Paragon Plus Environment

Page 21 of 22
Organic Process Research & Development
1
2
3
Figure Legends
4
5
6
7
8
9
Figure 1. HPLC characterization of pantatheine conversions. A. Pantetheine 1. B.
Phosphopantetheine 2. C. Dephosphoꢀcoenzyme A 3. D. Coenzyme A.
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
0.39
A
0.26
A220
0.13
0.00
0.39
B
0.26
A220
0.13
0.00
0.39
C
A260
0.26
0.13
0.00
0.39
D
0.26
A260
0.13
0.00
5
10
15
20
25
Time (min)
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 22
1
2
3
4
Figure 2. HPLC characterization of pantathine conversions. A. Pantethine 8. B.
5
6
Phosphopantethine 9. C. Dephosphoꢀcoenzyme A disulfide 10. D. Disulfideꢀcoenzyme A 11.
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
0.42
A
0.28
A220
0.14
0.00
0.39
B
0.26
A220
0.13
0.00
0.39
C
A260
0.26
0.13
0.00
0.39
D
A260
0.26
0.13
0.00
10
15
20
25
Time (min)
22
ACS Paragon Plus Environment