Trans-Dichlorotetracyanoplatinate(IV) as a Reagent for the Rapid and Quantitative Formation of Intramolecular Disulfide Bonds in Peptides

Article abstract of DOI:10.1021/jo981748r

Oxidation of cysteine thiol groups by trans-dichlorotetracyanoplatinate(IV) to form intramolecular peptide disulfide bonds has been studied for a series of dithiol peptides ranging from 4 to 15 amino acid residues in length. The dithiol peptides are rapidly and quantitatively transformed to their intramolecular disulfide forms by a slight excess of [Pt(CN)₄Cl₂]^2-, as shown by HPLC. Quantitative analyses by HPLC and by spectrophotometric titration confirm a [Pt(IV)]:[dithiol peptide] stoichiometry of 1:1. Under the low pH conditions used, oxidation to form a 38-membered ring in the case of reduced somatostatin is as rapid as that to form much smaller rings, suggesting that ring closure is not the rate-determining step. The oxidation rates increase as the pH is increased. Time-resolved spectra show two isosbestic points, indicating that no peptide-platinum intermediates accumulate to a significant amount. A reaction mechanism similar to that for reduction of [Pt(CN)₄Cl₂]^2- by monothiols is proposed. [Pt(CN)₄Cl₂]^2- is a mild oxidant and essentially substitution inert; its reduction product, [Pt(CN)₄Cl₂]^2-, is stable, has no redox chemistry with peptides, and does not form complexes with peptides. Moreover, [Pt(CN)₄Cl₂]^2- and [Pt(CN)₄Cl₂]^2- are nontoxic and readily separable from peptides by HPLC, and the cost of the Pt(IV) complex is negligible compared with that of peptides. The only unwanted side reaction observed with [Pt(CN)₄Cl₂]^2- is oxidation of the sulfur of methionine to the sulfoxide form. These characteristics and the results of this study suggest that [Pt(CN)₄Cl₂]^2- is an excellent reagent for the formation of intramolecular peptide disulfide bonds.

Full text of DOI:10.1021/jo981748r

4590
J . Org. Chem. 1999, 64, 4590-4595
tr a n s-Dich lor otetr a cya n op la tin a te(IV) a s a Rea gen t for th e Ra p id
a n d Qu a n tita tive F or m a tion of In tr a m olecu la r Disu lfid e Bon d s in
P ep tid es
Tiesheng Shi and Dallas L. Rabenstein*
Department of Chemistry, University of California, Riverside, California 92521
Received August 26, 1998
Oxidation of cysteine thiol groups by trans-dichlorotetracyanoplatinate(IV) to form intramolecular
peptide disulfide bonds has been studied for a series of dithiol peptides ranging from 4 to 15 amino
acid residues in length. The dithiol peptides are rapidly and quantitatively transformed to their
intramolecular disulfide forms by a slight excess of [Pt(CN)4Cl2]2-, as shown by HPLC. Quantitative
analyses by HPLC and by spectrophotometric titration confirm a [Pt(IV)]:[dithiol peptide]
stoichiometry of 1:1. Under the low pH conditions used, oxidation to form a 38-membered ring in
the case of reduced somatostatin is as rapid as that to form much smaller rings, suggesting that
ring closure is not the rate-determining step. The oxidation rates increase as the pH is increased.
Time-resolved spectra show two isosbestic points, indicating that no peptide-platinum intermediates
accumulate to a significant amount. A reaction mechanism similar to that for reduction of
[Pt(CN)4Cl2]2- by monothiols is proposed. [Pt(CN)4Cl2]2- is a mild oxidant and essentially
substitution inert; its reduction product, [Pt(CN)4]2-, is stable, has no redox chemistry with peptides,
and does not form complexes with peptides. Moreover, [Pt(CN)4Cl2]2- and [Pt(CN)4]2- are nontoxic
and readily separable from peptides by HPLC, and the cost of the Pt(IV) complex is negligible
compared with that of peptides. The only unwanted side reaction observed with [Pt(CN)4Cl2]2- is
oxidation of the sulfur of methionine to the sulfoxide form. These characteristics and the results of
this study suggest that [Pt(CN)4Cl2]2- is an excellent reagent for the formation of intramolecular
peptide disulfide bonds.
In tr od u ction
all, as will be exemplified by peptide 8 in Table 1 of this
work (vide infra). Potassium ferricyanide has been found
to be useful in the formation of single intramolecular
disulfide bonds in small peptides, in particular peptides
in the oxytocin and somatostatin families.1,2 However,
methionine and tryptophan residues in the peptides can
be oxidized, giving rise to side products.2 Further, even
in the oxidation of relatively small peptides in the
oxytocin family, significant amounts of side products,
including dimers and polymers, were found.9 These side
reactions can be interpreted in terms of an oxidation
mechanism that involves free radicals.10 Thus, it is not
surprising that there are no general rules for selecting
an oxidant for the formation of intramolecular peptide
disulfide bonds.
Recently, trans-dichlorotetracyanoplatinate(IV) ([Pt-
(CN)4Cl2]2-) has been used as a model compound for
anticancer-active platinum(IV) drugs, and the kinetics
and reaction mechanisms for reduction of the model
compound by thiols and methionine have been investi-
gated.11,12 Mechanistically, reduction of chloro-platinum-
(IV) complexes by thiols shows some features that
are similar to thiol-disulfide exchange reactions.11,13
Formation of intramolecular disulfide bonds by oxida-
tion of the corresponding free thiol precursors in solution
is usually the last step in the synthesis of disulfide-
containing peptides and still remains as a significant
challenge.1-3 Many oxidants, including oxygen, oxidized
glutathione (GSSG), dimethyl sulfoxide (DMSO), iodine,
ethoxycarbonylsulfenyl chloride (SceCl), and potassium
ferricyanide (K3Fe(CN)6), have been used for this pur-
pose.1-3 Recently, the use of Ellman’s reagent (5,5′-
dithiobis(2-nitrobenzoic acid)) bound through two sites
to a solid support has been used to form disulfide bonds.4
Although each oxidant has its advantages, most also have
drawbacks as well.1 For instance, advantages of DMSO
oxidation include applicability over the wide pH range
of 3-8, relatively fast reaction rates, and improved
solubility for the materials being oxidized.1,5,6 However,
removing DMSO from the final products can be dif-
ficult.7,8 Moreover, some peptides cannot be oxidized at
* Fax: 1-909-787-4713. Email: DLRAB@ucrac1.ucr.edu.
(1) Andreu, D.; Albericio, F.; Sole´, N. A.; Munson, M. C.; Ferrer,
M.: Barany, G. In Peptide Synthesis Protocols; Pennington, M. W.,
Dunn, B. M., Ed.; Humana Press: New J ersey, 1994; pp 91-169.
(2) Annis, I.; Hargittai, B.; Barany, G. Methods Enzymol. 1997, 289,
198-221.
(3) Moroder, L.; Besse, D.; Musiol, H.-J .; Rudolph-Bo¨hner, S.;
Siedler, F. Biopolymers 1996, 40, 207-234.
(4) Annis, I.; Chen, L.; Barany, G. J . Am. Chem. Soc. 1998, 120,
7226-7238.
(5) Tam, J . P.; Wu, C.-R.; Liu, W.; Zhang, J .-W. J . Am. Chem. Soc.
1991, 113, 6657-6662.
(6) Tam, J . P.; Shen, Z.-Y. Int. J . Pept. Protein Res. 1992, 39, 464-
(8) Ferrer, M.; Woodward, C.; Barany, G. Int. J . Pept. Protein Res.
1992, 40, 194-207.
(9) Moore, G. Biochem. J . 1978, 173, 403-409.
(10) Stochel, G.; Martinez, P.; van Eldik, R. J . Inorg. Biochem. 1994,
54, 131-140.
(11) Shi, T.; Berglund, J .; Elding, L. I. Inorg. Chem. 1996, 35, 3498-
3503.
(12) Shi, T.; Berglund, J .; Elding, L. I. J . Chem. Soc., Dalton Trans.
1997, 2073-2077.
(13) Shi, T.; Elding, L. I. J . Biol. Chem., to be submitted for
publication.
471.
(7) Munson, M. C.; Barany, G. J . Am. Chem. Soc. 1993, 115, 10203-
10216.
10.1021/jo981748r CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/27/1999

Oxidation of Dithiol Peptides by [Pt(CN)4Cl2]2-
J . Org. Chem., Vol. 64, No. 13, 1999 4591
Ta ble 1. Dith iol P ep tid es Used in th e Stu d y of In tr a m olecu la r P ep tid e Disu lfid e Bon d F or m a tion by Rea ction w ith
[P t(CN)4Cl2]2-
no.
sequence of dithiol peptides
ring sizea
% acetonitrileb
1
2
3
4
5
6
7
8
9
N-Ac-Cys-Pro-Phe-Cys-NH2
N-Ac-Thr-Cys-Pro-Phe-Cys-Arg-NH2
N-Ac-Pro-Thr-Cys-Pro-Phe-Cys-Arg-Lys-NH2
N-Ac-Lys-Pro-Thr-Cys-Pro-Phe-Cys-Arg-Lys-Thr-NH2
N-Ac-Cys-Gly-Tyr-Cys-NH2
14
14
14
14
14
14
14
14
20
20
20
38
17
16
14.7
12.7
8
N-Ac-Thr-Cys-Gly-Tyr-Cys-His-NH2
N-Ac-Ile-Thr-Cys-Gly-Tyr-Cys-His-Lys-NH2
8
12
16
12
16
12
26
N-Ac-Asp-Ile-Thr-Cys-Gly-Tyr-Cys-His-Lys-Leu-His-Gly-Gln-Met-Lys-NH2
Cys-Tyr-Phe-Gln-Asn-Cys (reduced pressinoic acid)
Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Leu-Gly-NH2 (reduced oxytocin)
Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Arg-Gly-NH2 (reduced arginine-vasopressin)
Ala-Gly-Cys-Lys-Asn-Phe-Phe-Try-Lys-Thr-Phe-Thr-Ser-Cys (reduced somatostatin)
10
11
12
a
b
Ring size for the oxidized forms. Percentage of acetonitrile in the mobile phases used for HPLC separations.
[Pt(CN)4Cl2]2-can be readily prepared,14 is a stable
complex, is substitution inert, and does not hydrolyze
over a wide pH range if kept with excess chloride in
solution. It is a mild oxidant, reacting with strong
reductants such as sulfite, iodide, thioethers, and thiol
groups15 but not with most of the functional groups in
peptides, including phenols and disulfide bonds. More-
over, its reduction product, [Pt(CN)4]2-, is substitution
inert. Except for extremely strong oxidants such as
chlorine gas, most of the common oxidants do not react
with [Pt(CN)4]2- 15
. These characteristics suggested to us
that [Pt(CN)4Cl2]2- might be useful as an oxidant for the
formation of intramolecular disulfide bonds in peptides.
In this paper, we report the results of a study of the
oxidation of a series of dithiol peptides by [Pt(CN)4Cl2]2-
.
The results demonstrate [Pt(CN)4Cl2]2- to be an efficient
reagent for the rapid and quantitative formation of
intramolecular disulfide bonds in peptides. This is the
first report on the use of a platinum(IV) complex as an
oxidant in peptide synthesis.
Resu lts a n d Discu ssion
Table 1 lists the dithiol peptides used to study the
formation of intramolecular disulfide bonds in peptides
by [Pt(CN)4Cl2]2-. Peptides 1-4 are model peptides for
the active site of glutathione thioltransferase,16 peptides
5-8 are model peptides for the active site of the disulfide
bond-forming protein DsbC,17 and peptides 9-12 are the
reduced dithiol forms of four disulfide-containing peptide
hormones.
P ep tid es 1-8. The oxidation was monitored by re-
versed-phase HPLC. To illustrate, a typical oxidation
experiment for decapeptide 4 (Table 1) is shown in Figure
1. In this experiment, peptide 4 (final concentration of
19.4 µM) was dissolved in a mixture of 0.02 M HCl
and 20.0 µM N-acetyl-L-phenylalanine (which served as
an HPLC reference), sparged with helium, and then
injected into the HPLC. The chromatogram (Figure 1,
top) gives rise to three peaks: the internal standard at
8.3 min, 4 at 12.6 min, and the oxidized form of 4 at 23.5
min. The bottom chromatogram in Figure 1 was obtained
for a reaction mixture containing 19.4 µM 4, 40 µM
[Pt(CN)4Cl2]2-, and 20 µM internal standard in 0.02 M
HCl after reaction for 30 min. The peaks for [Pt(CN)4]2-
and [Pt(CN)4Cl2]2-, peaks B and C respectively, were
F igu r e 1. Chromatograms of 19.4 µM decapeptide 4 in 0.02
M HCl (top) and a reaction mixture of 19.4 µM 4 and 40 µM
[Pt(CN)4Cl2]2- in 0.02 M HCl (bottom). The mobile phase
contained 14.7% acetonitrile, 0.10 M NaH2PO4, and sufficient
H3PO4 to adjust the pH to 2.5. Peak assignments: A, solvent;
B, [Pt(CN)4]2- + solvent; C, [Pt(CN)4Cl2]2-; D, internal stan-
dard acetyl-L-phenylalanine; E, decapeptide 4; F, oxidized form
of 4.
assigned by use of authentic samples of the two com-
plexes, whereas the peak at 23.5 min was assigned to
the oxidized form of 4 by mass and NMR spectroscopies.18
Chromatograms were also run for the samples in Figure
1 using a higher percentage of acetonitrile in the mobile
phase to decrease the retention times; the chromato-
grams were run for the same length of time as in Figure
1 to determine if there were any late-eluting peaks. No
additional peaks were observed, indicating a rapid and
clean formation of the intramolecular disulfide bond.
Similar results were obtained for the oxidation of 1-3
and 5-7. The chromatographic peaks for peptides 1-4
and their oxidized forms are broad as a result of the slow
cis-trans isomerization of the cysteine-proline peptide
bonds,19-23 whereas those for peptides 5-7 are sharp.
(14) Shi, T.; Elding, L. I. Inorg. Chim. Acta 1998, 282, 55-60.
(15) Shi, T. Ph.D. Thesis, Lund University, Sweden, 1997.
(16) Gan, Z.-R.; Wells, W. W. J . Biol. Chem. 1986, 261, 996-1001.
(17) Zapun, A.; Missiakas, D.; Raina, S.; Creighton, T. E. Biochem-
istry 1995, 34, 5075-5089.
(18) Rabenstein, D. L.; Gu, J .; Fernandez, M.; Huang, H., unpub-
lished results.

4592 J . Org. Chem., Vol. 64, No. 13, 1999
Shi and Rabenstein
F igu r e 3. Chromatograms of ca. 35 µM peptide 8 in 0.02 M
HCl (top) and a reaction mixture of ca. 35 µM peptide 8 and
50 µM [Pt(CN)4Cl2]2- in 0.02 M HCl after reaction for 1 h
(bottom). The mobile phase contained 16% acetonitrile, 0.10
M NaH2PO4, and sufficient H3PO4 to adjust the pH to 2.5.
Peak assignments: A, solvent; B, [Pt(CN)4]2- + solvent; C,
[Pt(CN)4Cl2]2-; D, oxidized form of 8 (methionine side chain
oxidized to methionine sulfoxide); E, 8.
F igu r e 2. Chromatograms of 80.5 µM tetrapeptide 1 in 0.02
M HCl (top) and a reaction mixture of 402.5 µM 1 and 500
µM [Pt(CN)4Cl2]2- in 0.02 M HCl, which was diluted five
times (bottom). The mobile phase contained 17% acetonitrile,
0.10 M NaH2PO4, and sufficient H3PO4 to adjust the pH to
2.5. Peak assignments: A, solvent; B, [Pt(CN)4]2- + solvent;
C, [Pt(CN)4Cl2]2-; D, tetrapeptide 1; E, oxidized form of 1.
Previous experience with peptide synthesis1,2 suggests
that oxidation reactions to form intramolecular peptide
disulfide bonds should be carried out at high dilution,
typically at 10-100 µM, to avoid aggregation and inter-
molecular disulfide bond formation. With [Pt(CN)4Cl2]2-
as an oxidant, however, oxidation to form intramolecular
disulfide bonds was found to be quantitative even at
millimolar concentrations, as evidenced by the reaction
between the Pt(IV) complex and 1 (Figure 2).
It was observed that peptide 8 was not oxidized to
its disulfide form by some commonly used oxidants,
including oxygen, GSSG, DMSO, and iodine. When [Pt-
(CN)4Cl2]2- was used, the oxidation was clean and com-
plete within 1 h in a 0.02 M HCl medium. Chromato-
grams for this oxidation are shown in Figure 3. From the
mass spectrum of the oxidized form of peptide 8, [peptide
+ 3H]3+ ) 621.3 (base peak) or molecular weight )
1860.9, indicating that not only was the 14-membered
disulfide bond-containing ring formed but the methionine
residue was oxidized to methionine sulfoxide as well. The
extent of methionine oxidation depends on the [dithiol
peptide]:[Pt(IV)] ratio; at a ratio g 1:2, the oxidation is
quantitative, as indicated by a single peak in the chro-
matogram. The oxidation of the methionine side chain
by the Pt(IV) complex is an expected result.12 The
methionine sulfoxide group can be conveniently reduced
with ammonium iodide while preserving the disulfide
bond.24,25
Oxid a tion of Red u ced Hor m on es. The reduced
dithiol forms of the peptide hormones pressinoic acid,
oxytocin, arginine-vasopressin, and somatostatin (9-12
in Table 1) are also rapidly and cleanly oxidized by
[Pt(CN)4Cl2]2-, as illustrated for reduced somatostatin 12
by the chromatograms in Figure 4. The peak assignments
and reaction conditions are given in the figure legend.
Chromatogram a is for a solution containing only 12,
chromatogram b is for a solution containing 12 plus less
than a stoichiometric amount of [Pt(CN)4Cl2]2-, chro-
matogram c is for 12 plus an excess of [Pt(CN)4Cl2]2-
,
and chromatogram d is for the same solution as in c
but after a longer reaction time. The chromatograms
indicate that, when less than a stoichiometric amount of
[Pt(CN)4Cl2]2- is used, only a fraction of the reduced
somatostatin is converted to the oxidized form. In the
presence of excess [Pt(CN)4Cl2]2-, all of the reduced
somatostatin is converted to the disulfide form, and the
disulfide bond is stable toward the excess [Pt(CN)4Cl2]2-
.
Formation of the disulfide bond of somatostatin was
found to be as rapid as formation of the disulfide bonds
of the other peptides in Table 1, which suggests that,
under the conditions used, ring closure to form the
disulfide bond is probably not the rate-controlling step
in the oxidation reaction for this series of peptides. It also
is interesting to note that the disulfide forms of the
peptide hormones elute from the column before the
(19) Larive, C. K.; Guerra, L.; Rabenstein, D. L. J . Am. Chem. Soc.
1992, 114, 7331-7337.
(20) Larive, C. K.; Rabenstein, D. L. J . Am. Chem. Soc. 1993, 115,
2833-2836.
(21) Rabenstein, D. L.; Yeo, P. L. J . Org. Chem. 1994, 59, 4223-
4229.
(24) Yajima, H.; Fujii, N.; Funakoshi, S.; Watanabe, T.; Murayama,
E.; Otaka, A. Tetrahedron 1988, 44, 805-819.
(25) Nicolas, E.; Vilaseca, M.; Giralt, E. Tetrahedron 1995, 51, 5701-
5710.
(22) Rabenstein, D. L.; Yeo, P. L. Bioorg. Chem. 1995, 23, 109-118.
(23) Rabenstein, D. L.; Weaver, K. H. J . Org. Chem. 1996, 61, 7391-
7397.

Oxidation of Dithiol Peptides by [Pt(CN)4Cl2]2-
J . Org. Chem., Vol. 64, No. 13, 1999 4593
F igu r e 5. Absorbance at 256 nm for solutions with constant
[Pt(IV)] and increasing [1]. Conditions: [Pt(CN)4Cl22-] ) 0.10
mM, [HCl] ) 0.02 M, and room temperature.
F igu r e 4. Chromatograms for the oxidation of reduced
somatostatin 12: (a) 27.04 µM 12; (b) a reaction mixture of
27.04 µM 12 and 20 µM [Pt(CN)4Cl2]2- after reaction for 1 h
at pH ≈ 2.3; (c) a reaction mixture of 27.04 µM 12 and 50 µM
[Pt(CN)4Cl2]2- after reaction for 1 h at pH ≈ 2.3; and (d) the
same reaction mixture as in (c) after reaction for 12 h.
Peak assignments: A, solvent; B, [Pt(CN)4]2- + solvent; C,
[Pt(CN)4Cl2]2-; D, somatostatin; E, reduced somatostatin 12.
reduced forms, in contrast to the chromatographic be-
havior of peptides 1-7 (Figures 1 and 2).
Qu a n tita tion of th e Oxid a tion Rea ction s. The
reaction of [Pt(CN)4Cl2]2- with dithiol peptides was
quantitatively characterized by two experiments. In the
first experiment, a somatostatin sample obtained from
Bachem (peptide content reported to be (89.2 ( 3)%,
which was confirmed previously by NMR23) was used as
a standard to prepare an HPLC peak area vs somatosta-
tin concentration calibration curve. Reduced somatostatin
12 was then prepared by reduction of somatostatin with
a 10-fold excess of dithiothreitol in pH 7.0 buffer, the pH
was reduced to 2.5, and the reduced somatostatin was
isolated from the reaction mixture by semipreparative
scale HPLC. The chromatogram of the sample isolated
(Figure 4a) indicates the absence of native somatostatin.
The concentration of reduced somatostatin was deter-
mined spectrophotometrically by reaction with Ellman’s
reagent to be 33.8 µM. Aliquots of the reduced soma-
tostatin solution were then reacted with [Pt(CN)4Cl2]2-
at pH ≈ 2.3 using four reaction conditions: 27.04 µM 12
+ 50 µM Pt(IV) reacted for 1 h; 27.04 µM 12 + 200 µM
Pt(IV) reacted for 1 h; 27.04 µM 12 + 100 µM Pt(IV)
reacted for 1.5 h; and 27.04 µM 12 + 50 µM Pt(IV) aged
for 12 h. The reaction solutions were analyzed by HPLC.
Using the peak area/concentration calibration curve,
reduced somatostatin was found to be converted to native
somatostatin with a yield of (93 ( 8)%. The large error
is estimated from four sources: the somatostatin content
of the sample used to prepare the calibration curve
Figu r e 6. Time-resolved spectra for reduction of [Pt(CN)4Cl2]2-
by tetrapeptide 1. Conditions: [Pt(CN)4Cl22-] ) 0.10 mM, [1]
) 0.0858 mM, [HCl] ) 0.02 M, and room temperature. A
solution of 0.02 M HCl was used as reference. The first
spectrum was obtained ca. 10 s after mixing [Pt(CN)4Cl2]2-
and 1. The time interval between scans was 2 min. The last
scan was obtained 1 h after mixing; no spectral changes were
observed thereafter.
((3%); errors in the determination of the concentration
of reduced somatostatin using Ellman’s reagent ((3%);
weighing errors ((2%); and sample injection errors
((2%).
In the second quantitation experiment, a spectropho-
tometric titration was performed with peptide 1. The
absorption peak for [Pt(CN)4]2- at 256 nm (cf. Figure 6)
was used. Absorption spectra were measured for a series
of solutions in which the Pt(IV) concentration was held
constant at 0.100 mM and the concentration of 1 was
increased from 0 to 0.20 mM. After oxidation was
complete (checked by scanning the UV-vis spectrum),
the absorbance was measured at 256 nm for each
solution. The titration curve shown in Figure 5 gives rise
to a stoichiometry of [Pt(IV)]:[1] ) 1:(0.96 ( 0.06). If
errors from the concentration assay of 1 ((3%) and from
locating the end point in the titration curve ((3%) are
taken into account, this stoichiometry is in good agree-
ment with the HPLC results. Thus, these two experi-

4594 J . Org. Chem., Vol. 64, No. 13, 1999
Shi and Rabenstein
Sch em e 1
ments indicate quantitative formation of the disulfide
bond, with no significant formation of intermolecular
disulfides to give higher polymers.
The stoichiometry for oxidation of dithiol peptides
(without a methionine residue) by [Pt(CN)4Cl2]2- can be
described by eq 1:
(CN)4Cl2]2-, as in the present work, the [Pt(IV)]:[RSH]
stoichiometry deviates significantly from 1:2.26 Under
these conditions, RSOH can accumulate to a significant
amount and be further oxidized by the excess Pt(IV) to
RSO2H and RSO3H.26
A reaction mechanism similar to that proposed previ-
ously for reduction of [Pt(CN)4Cl2]2- by monothiols11,15 is
presented in Scheme 1.
In this reaction mechanism, dithiol peptide and its
deprotonated species reduce [Pt(CN)4Cl2]2- in parallel
reactions, giving two short-lived intermediates (I and II),
which then undergo intramolecular attack to form the
disulfide bond. The Pt(IV) complex is substitution inert
and no Pt(IV)-dithiol peptide reaction intermediates are
observed, as evidenced by the time-resolved spectra
(Figure 6). The rate of oxidation of dithiol peptides
increases when the pH of the reaction medium is in-
creased. A similar pH dependence was observed previ-
ously for reduction of [Pt(CN)4Cl2]2- by monothiols and
is due to the much faster reaction of the thiolate form
with the Pt(IV) complex.11 Although there is very little
of the thiolate species present at the low pH used in this
work, there is nevertheless some reaction by the lower
pathway in Scheme 1 because the rate of reaction of
[Pt(CN)4Cl2]2- with the thiolate group is orders of mag-
nitude faster than with the thiol group.11 A low pH was
used to minimize potential side reactions in which the
free thiol group in intermediates I and II would react
with the excess [Pt(CN)4Cl2]2- before it could undergo
intramolecular attack at the SCl or SHCl+ group to form
the disulfide bond. Working in acidic media is also
convenient because thiol-disulfide exchange and air
oxidation side reactions are essentially eliminated. Theo-
retically, an intermolecular attack on the intermediates
by another peptide could produce a dimer as in the case
of monothiol oxidations.11 However, no chromatographic
peaks that could be assigned to intermolecular disulfides
were observed, suggesting that the rate for intermolecu-
lar attack is much less than that for intramolecular
attack under the conditions used. In other words, rapid
conformational change of intermediates I and II to make
ring closure possible is a highly favorable step. In this
regard, the possibility for formation of the hydrolyzed
intermediate similar to RSOH as discussed above is also
less likely.
Rea ction Mech a n ism . Time-resolved spectra were
measured for the oxidation of tetrapeptide 1; the results
are shown in Figure 6. The growing peak at 256 nm is
ascribed to formation of [Pt(CN)4]2-. Two clear-cut isos-
bestic points are observed at 243.4 and 286.6 nm, indi-
cating that no platinum-containing intermediates ac-
cumulate to any significant concentration. Thus, [Pt(CN)4-
Cl2]2- is directly reduced to [Pt(CN)4]2-, further confirm-
ing the quantitative nature of the oxidation reaction.
Reduction of [Pt(CN)4Cl2]2- by monothiols (RSH) was
shown previously to proceed according to eq 2 in the
[Pt(CN)4Cl2]2- + 2RSH f [Pt(CN)4]2- + RSSR +
2Cl- + 2H+ (2)
presence of a large excess of RSH.11 Reduction of plati-
num(IV) halide complexes is rationalized in terms of
halide-bridged electron transfer,11-15 and reduction by
thiols is assumed to take place by attack of the sulfur on
a coordinated chloride. Accompanying the attack is a Cl+
transfer from the Pt(IV) center to the incoming sulfur
atom, resulting in RSCl and RSHCl+ as initial short-lived
products.11,15 RSCl and RSHCl+ can hydrolyze quickly,
giving rise to RSOH, which subsequently is trapped by
RSH and RS- according to reactions 3 and 4.27 When
RSOH + RSH f RSSH + H2O
RSOH + RS- f RSSH + OH-
(3)
(4)
monothiols are oxidized under conditions of excess [Pt-
(26) Shi, T., unpublished results.
(27) Allison, W. S. Acc. Chem. Res. 1976, 9, 293-299.

Oxidation of Dithiol Peptides by [Pt(CN)4Cl2]2-
J . Org. Chem., Vol. 64, No. 13, 1999 4595
Oxid a tion P r otocol. Oxidation reactions were carried out
in 20 mM HCl solution at room temperature. A stock solution
of 5.00 mM K2[Pt(CN)4Cl2] in 20 mM HCl was prepared;
[Pt(CN)4Cl2]2- is stable indefinitely under these conditions. A
100 µM to 1 mM solution of the dithiol peptide (or crude
product cleaved from the synthesis resin) was also prepared
in 20 mM HCl. The dithiol peptide and [Pt(CN)4Cl2]2- solutions
were mixed at [dithiol peptide]:[Pt(IV)] ratios ranging from
1:1.2 to 1:2 and then flushed for several minutes with nitrogen
or argon. The solutions were then allowed to react for ∼1 h,
after which conversion of the dithiol peptide to its disulfide
form was complete.
Mon itor in g th e Oxid a tion Rea ction s by HP LC. Oxida-
tion of dithiol peptides by [Pt(CN)4Cl2]2- was monitored by
isocratic HPLC using a 100 mm × 3.2 mm ODS (C18) column
(particle size 3 µm). The detector was set at 215 nm. Mobile
phases were prepared by addition of NaH2PO4 (0.10 M final
concentration) and acetonitrile to water, and then the pH was
adjusted to 2.5 with 1 M H3PO4. Chromatographic conditions
were optimized for separation of the reduced dithiol and
oxidized forms of the peptides by varying the percentage of
acetonitrile in the mobile phase;30 the percentage of acetonitrile
used for each peptide is given in Table 1.
Tim e-Resolved Sp ectr a a n d Sp ectr op h otom etr ic Ti-
tr ation . Time-resolved spectra for the reaction of [Pt(CN)4Cl2]2-
with peptide 1 in 20 mM HCl were recorded with a diode array
spectrophotometer. The reaction solution was contained in 1.00
cm quartz cells. Solutions of 2.0 × 10-4 M [Pt(CN)4Cl2]2- and
of 1.716 × 10-4 M 1 were prepared fresh in 20 mM HCl and
were sparged with helium for ca. 10 min before mixing and
recording the spectra. For the spectrophotometric titration, a
series of solutions with [Pt(IV)] ) 0.100 mM, [HCl] ) 20 mM,
and 0 e [1] e 0.200 mM were freshly prepared and flushed
with helium, and their absorbances were measured at 256 nm
(the absorption peak of [Pt(CN)4]2-) after the solutions had
reacted for ca. 3 h.
Con clu sion s. This work has demonstrated that
[Pt(CN)4Cl2]2- is an efficient oxidant for the rapid and
quantitative formation of intramolecular peptide disulfide
bonds. The Pt(IV) complex has several advantages: (1)
it is a mild oxidant and essentially substitution inert; (2)
its reduced product [Pt(CN)4]2- is also substitution inert;
(3) both [Pt(CN)4Cl2]2- and [Pt(CN)4]2- are nontoxic and
readily separable from peptides by HPLC; and (4) the
cost of the Pt(IV) complex is negligible compared to
that of peptides. The only drawback observed is that
[Pt(CN)4Cl2]2- can oxidize the methionine side chain to
methionine sulfoxide.
Exp er im en ta l Section
Ch em ica ls. K2[Pt(CN)4]‚3H2O (J ohnson Matthey), dithio-
threitol (DTT, Aldrich), N-acetyl-L-phenylalanine (Sigma), and
5,5′-dithiobis(2-nitrobenzoic acid) (Ellman’s reagent, Sigma)
were used as received. K2[Pt(CN)4Cl2] was synthesized by the
procedure described previously.14 Stock solutions of 1.00 mM
and 5.00 mM K2[Pt(CN)4Cl2] in 0.02 M HCl were prepared.
Sodium dihydrogen phosphate, phosphoric acid, hydrochloric
acid, trifluroacetic acid (TFA), and acetonitrile (Optima) were
obtained from Fisher Scientific Co. The peptide hormones
arginine-vasopressin, oxytocin, pressinoic acid, and somatosta-
tin were obtained from Bachem Inc. (Torrance, CA). Peptides
1-8 were synthesized by solid-phase peptide synthesis using
FMOC methodology and characterized by LC-MS and NMR
spectroscopy in this laboratory.18 Water was purified with a
Millipore water purification system. Mobile phases were
filtered through a 0.45 µm cellulose nitrate filter, sparged with
helium for ca. 15 min just before the experiment and used for
less than a week.
P u r ifica tion of th e P ep tid es. The reduced dithiol forms
of the peptide hormones were prepared by reduction of the
oxidized forms with a 10-fold excess of DTT at pH 7.0 and were
isolated immediately by reversed-phase HPLC on a 100 mm
× 250 mm C18 column. The reduced peptides were eluted with
an acetonitrile-water gradient containing 0.1% TFA. The pH
of the mobile phases was low (ca. 2.5) to minimize oxidation
of the reduced peptides by air. The other peptides used in this
work were also purified by HPLC. The reduced peptide content
was determined spectrophotometrically by use of Ellman’s
reagent at pH 7.4 in a phosphate buffer containing 1 mM
EDTA.28,29
Ack n ow led gm en t. This research was supported by
National Institutes of Health grant GM 37000. We
thank Drs. Haihong Huang and J uan Gu and Mr.
Maximillian Fernandez in this laboratory for the prepa-
ration and characterization of peptides 2-8.
J O981748R
(29) Riddles, P. W.; Blakeley, R. C.; Zerner, B. Anal. Biochem. 1979,
94, 75-81.
(30) Yeo, P. L.; Rabenstein, D. L. Anal. Chem. 1993, 65, 3061-3066.
(28) Ellman, G. L. Arch. Biochem. Biophys. 1959, 82, 70-77.