Thioether macrocycles of the microbisporicins via reductive desulfurization

Article abstract of DOI:10.1016/j.tet.2018.06.050

The microbisporicins are the most potent lantibiotics isolated to-date. Cyclic tetra-, hexa- and octapeptides, inspired by this family of antimicrobial agents, have been synthesized from linear peptides. Generalized reaction conditions are reported for th

Full text of DOI:10.1016/j.tet.2018.06.050

Accepted Manuscript
Thioether macrocycles of the microbisporicins via reductive desulfurization
Samuel K. Kutty, Joshua A. Lutz, Simon Felder, Philip Hahn, Carol M. Taylor
PII:
S0040-4020(18)30750-6
10.1016/j.tet.2018.06.050
TET 29646
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 26 May 2018
Revised Date: 17 June 2018
Accepted Date: 19 June 2018
Please cite this article as: Kutty SK, Lutz JA, Felder S, Hahn P, Taylor CM, Thioether macrocycles of the
microbisporicins via reductive desulfurization, Tetrahedron (2018), doi: 10.1016/j.tet.2018.06.050.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
Thioether Macrocycles of the Microbisporicins via Reductive Desulfurization
Samuel K. Kutty, Joshua A. Lutz , Simon Felder, Philip Hahn and Carol M. Taylor*
Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803
cmtaylor@lsu.edu
Abstract. The microbisporicins are the most potent lantibiotics isolated to-date. Cyclic
tetra-, hexa- and octapeptides, inspired by this family of antimicrobial agents, have been
synthesized from linear peptides. Generalized reaction conditions are reported for the
two-step conversion of linear peptides to thioether macrocycles: formation of a disulfide
followed by reductive desulfurization. 1H NMR analysis of the reduction reaction
mixture indicated the intermediacy of a dehydroalanine intermediate when excess
hexamethylphosphorus triamide (HMPT) was employed for the reduction. Maintaining a
stoichiometric amount of HMPT, in dilute methanolic solution, gave the corresponding
thioethers, retaining stereochemical integrity.
1. Introduction
Lantibiotics are lanthionine-containing peptides with antibiotic properties (e.g., 1,
Figure 1A).1 These compounds are synthesized ribosomally and undergo substantial post-
translational modifications, characterized by the formation of the bridging bis-amino
acids, meso-lanthionine (Lan, Ala-S-Ala, Figure 1B) and meso-β-methyl-lanthionine
(MeLan, Abu-S-Ala).2,3 Microbisporicins A1 (with a dihydroxylated proline at position
14) and A2 constitute antibiotic NAI-107, demonstrated to have submicromolar activity
against MRSA.4,5

ACCEPTED MANUSCRIPT
2
Figure 1. (A) Microbisporicin A2 (1). Dhb - dehydrobutyrine, ClTrp – 5-
chlorotryptophan, Dha = dehydroalanine, Abu = α-aminobutyric acid, Hyp = 4-
hydroxyproline. (B) Retrosynthetic analysis of the thioether macrocycle.
A major challenge in the chemical synthesis of lantibiotics is the formation of the
array of thioether bridges with the correct topology.6 While a number of approaches to
such macrocycles have been explored,7-10 the incorporation of a preformed lanthionine
unit,11 in combination with solid-phase peptide synthesis (SPPS), has gained momentum,
with cyclizations involving peptide bond formation rather than manipulation of the
thioether linkage. In 2010-11, there were some pioneering reports of the total synthesis
of lantibiotics via SPPS protocols. Notwithstanding these significant accomplishments,
this linear approach results in lengthy syntheses and limited quantities of peptides that
can be difficult to purify. Vederas reported the first SPPS of lactocin S, in 10% overall
yield for 71 steps.12 Tabor’s group reported the first synthesis of a lantibiotic fragment
with overlapping lanthionine bridges via SPPS,13 reporting that “the peptide proved
difficult to purify by HPLC due to tailing of impurities” and “a yield of less than 0.1 mg
of the peptide was obtained, which was not sufficient for detailed NMR analysis.”
Vederas reported a similar approach to both components of lacticin 3147,14 largely via

ACCEPTED MANUSCRIPT
3
SPPS although “the N-terminal peptide containing ring A was synthesized in solution
because of its extensive and densely packed modifications.” The van der Donk group has
pursued both chemical15 and bioengineering16 avenues to produce lantibiotics (e.g., their
work on epilancin 15X) signifying that neither approach is superior and complementary
approaches are valuable, even within the same research group. Since 2012, the chemical
synthesis of lantibiotics has slowed and literature searches indicate production of these
peptides mostly via microbial systems.
With a view to a more convergent, scalable approach to medium-sized peptide
fragments bearing thioethers, we sought to explore the reductive desulfurization of
disulfide macrocycles (Figure 1B). Harpp and Gleason originally described the
desulfurization of cystine to lanthionine using aminophosphines.17 Reductive
desulfurization has also been accomplished under alkaline conditions, with the caveat that
a dehydroalanine intermediate surrenders control over stereochemistry.18 The Harpp-
Gleason approach was applied to the first total synthesis of the prototypical lantibiotic
nisin by Shiba and coworkers in the early 1990s.19-23 Considerable optimization of
reaction conditions was required for each peptide. Little attention has been paid to this
approach in the intervening years. We sought to explore the reductive desulfurization
approach with a view to identifying generalized reaction conditions that ought to be
applicable to a range of peptides and their synthesis on a useful scale.
2.
Results and Discussion
Our investigation began with the synthesis of the tetrapeptides depicted in
Scheme 1. This tetrapeptide occurs in several lantibiotics, including nisin and subtilin,

ACCEPTED MANUSCRIPT
4
where it constitutes the B-ring. Short peptides containing this B-ring have been prepared
previously via diastereoselective conjugate addition of a liberated Cys thiol to an
appropriately positioned dehydroalanine (Dha). Toogood (H-Ala-Dha-Pro-Gly-L-Cys-
Ala-OBn),24 Bradley (Boc-Ile-Dha-Pro-Gly-L-Cys-Val-Gly-O-resin)25 and van der Donk
(H-Leu-Dha-Pro-Gly-L-Cys-Val-Gly-OH)26 all reported the formation of a single
compound, corresponding to meso-lanthionine.
Our synthesis of the B-ring tetrapeptides is summarized in Scheme 1. Dipeptide 4
was prepared by condensation of Fmoc-Gly-OH (2) and H-L-Cys(Trt)-NH2 (3). Peptide
4 was deprotected at the N-terminus to give amine 5, a common intermediate in all the B-
ring tetrapeptides. We synthesized dipeptide acids 8a and 8b, Fmoc- and Boc-protected
respectively, at the N-terminus. Our usual approach to dipeptide acids,27 using
DCC/NHS, was less effective than EDC/HOBT, presumably due to the hindered nature of
the prolyl amino component. Coupling of each dipeptide acid independently with amine
5 gave tetrapeptides 9a and 9b. The N-acetyl derivative 9c was prepared from 9a via
standard methods. Treatment of each peptide with iodine led to concomitant removal of
the trityl ethers and disulfide formation, viz. disulfide macrocycles 10a-c. The excess
iodine was typically quenched using sodium thiosulfate, followed by aqueous work-up.
However, for peptides with considerable water-solubility (e.g., 10c), work-up was better
achieved using Amberlite® IRA-400A resin in its chloride form28 to avoid water. A
single crystal of 10b was suitable for X-ray diffraction analysis, which revealed the prolyl
amide bond to be in the cis-conformation.

ACCEPTED MANUSCRIPT
5
Scheme 1. Synthesis of B-ring tetrapeptides.
Desulfurization studies began with disulfide 10a. Important reaction parameters
such as desulfurization agent, solvent and concentration were considered. No reaction
was observed with PPh3 or P(OEt)329 over several days, whereas P(nBu)330 led to the
dithiol reduction product. Shiba and coworkers employed hexaethylphosphorus triamide
(HEPT),19-23 which we found to be incompatible with the Fmoc group. Cleavage could
be attributable to diethylamine present as an impurity, although the phosphorus triamide
itself is likely basic enough31 to promote the β-elimination of dibenzofulvene. Indeed,
dipeptide 4 could be converted to amine 5 in 48% yield on exposure to 6.1 mM HEPT (5
equivalents) in methanol. With a shift in focus to disulfide 10c, a range of solvents were
explored (PhH, PhMe, CH2Cl2, CHCl3, THF, MeOH, MeCN and DMF), using 20
equivalents of HEPT (30 mM). For tetrapeptides 10b and 10c, the best results for
contraction to the thioether, vis-à-vis minimizing byproducts and reaction time, were
obtained with methanol as solvent.

ACCEPTED MANUSCRIPT
6
Although less reagent was required for the reduction of 10c to 11c with Verkade’s
base,32 and the reaction was faster, epimerization was exacerbated (vide infra). A change
from HEPT to HMPT was advantageous for two reasons. Firstly, dimethylamine is more
volatile than diethylamine and thus the amine impurity is more readily removed by
evaporation. Separation of the peptide products from the more polar byproducts derived
from HMPT was typically also more facile. The high stoichiometric ratio of phosphorus
triamide to peptide, employed by Shiba et al.,19-23 compensated for the high dilution (0.3-
3.0 mM) invoked to encourage intramolecular thioether formation. In our hands, while
the desired thioether formation was observed in the presence of 5-40 equivalents of
HMPT, the reaction was cleaner, albeit slower, at lower concentrations. The impact of
HMPT concentration was studied by 1H NMR (Figure 2) leading to insights into the
mechanism of the reaction (Scheme 2). At ~50 mM HMPT, signals appeared at δ5.6 and
6.1 ppm, indicative of a dehydroalanine intermediate. When a stoichiometric amount of
HMPT was employed, this intermediate was not observed.
Harpp and Gleason proposed that the reaction proceeds via SN2 attack of a
thiolate anion at the β-position of a Cys-derived phosphonium salt (Scheme 2, path a).
Oku et al. reported the stabilization of such a phosphonium salt in a rotaxane.33 The
stereochemistry at Cα is unperturbed and the configuration at Cβ is inverted in the case
of β-methylcysteine. In the presence of base, including an excess of HMPT, elimination
of thiophosphoramide (path b) generates dehydroalanine (Dha, IV) which renders the Cα
stereogenic center susceptible to scrambling. Others have reported the intermediacy of
dehydroalanine intermediates.34,35

ACCEPTED MANUSCRIPT
7
Figure 2. Reductive desulfurization of 10c with HMPT in CD3OD monitored by 1H
NMR at 500 MHz. (A) 3 equiv. HMPT (51 mM) added in one portion; (B) 1 equivalent
of HMPT (13 mM) added slowly over 30 min (t = 5 min) followed by 0.3 equiv. on third
day for complete conversion to 11c.
Scheme 2. Pathways for reductive desulfurization.
Shiba and coworkers probed the stereochemical integrity of the lanthionine
residues in the macrocyclic peptide fragments they synthesized via their case-by-case
optimized reaction conditions.19-23 Acidic degradation, followed by amino acid analysis,

ACCEPTED MANUSCRIPT
8
showed that “racemization” increased with the concentration of acid, leading them to
conclude that the minor diastereomer/s arose from partial “racemization” on heating in
acid. There was little NMR data in their papers of 20-30 years ago. We recommend
against high concentrations of HMPT, maintaining only a stoichiometric amount of
reducing agent, with additional aliquots being added over time.
While disulfides 10b and 10c were each a single compounds, they did exist as
conformations about the prolyl amide bond at room temperatures. Based on our findings
above, disulfides 10b and 10c were each treated with 1.3 equivalents of HMPT, with the
same amount added again after 16 hours. After 30 hours, thioethers 11b and 11c were
isolated. NMR indicated that each thioether contained low levels of a species that did not
change in relative intensity as a function of temperature. While the studies illustrated in
Figure 2 did not demonstrate the presence of dehydroalanine under these conditions,
perhaps Dha was present at a low steady state concentration that was not detected. That
we observed low levels of epimerization, where others did not, could be explained by the
length of our peptide and, in particular, the size of the N-terminal group. We see more
of the minor isomer in 11c than in 11b (Ac smaller than Boc). In the precedents where
analogous B-ring macrocycles were generated via thiol addition to an unequivocal
dehydroalanine,24-26 the peptides were longer than four residues, with one more residue at
the N-terminus and one or two more in the C-terminal direction. This additional steric
bulk likely influences the overall conformation of the molecule and thereby the facial
selectivity of the conjugate addition.
We chose to study three other ring systems inspired by the microbisporicins
(Schemes 3-5).4,5,36,37 This suite of peptides enabled us to probe the reactions in different

ACCEPTED MANUSCRIPT
9
ring sizes and introduce the concept of a third, orthogonally protected, non-participating
cysteine residue.
Octapeptide 14 (Scheme 3) mimics the C-ring of microbisporicin A2. The linear
dithiol precursor 12 was assembled by SPPS, using THP protection38,39 for the thiol
functionality of the Cys residues. Disulfide formation from 12 did not require
concomitant detritylation as in Scheme 1. Therefore, we initially explored dilute
hydrogen peroxide as the oxidant, since work-up was inherently simpler than with iodine.
While this was successful on occasion, the reaction was very sensitive to peroxide
concentration and was not very reproducible. Iodine (3 equiv.) remained the oxidant of
choice, with the reaction conducted in 1:1 chloroform-methanol at a peptide
concentration of 0.5 mM. The reductive desulfurization of 13 was achieved over several
days with 1.3 equivalents of HMPT being added at 4-5 hour intervals over the course of a
week.
Scheme 3. Synthesis of a C-ring octapeptide thioether.
Hexapeptide 29, corresponding to the D-ring of the microbisporicins, was
prepared in solution according to Scheme 4. Tetrapeptide 23 was formed by
condensation of dipeptide acid 18 and dipeptide amine 22. Such condensations,
involving an activated acid of a peptide fragment are potentially susceptible to
epimerization of the C-terminal residue. We employed Carpino’s conditions,40 well

ACCEPTED MANUSCRIPT
10
established to minimize this side reaction, and saw no evidence of epimerization at any
stage of the synthesis. Hydrogenolysis of the benzyl ester in 23 afforded acid 24 that was
condensed with L-Cys building block 3 to give pentapeptide 25. Homologation with D-
Cys building block 6a, in the N-terminal direction, gave 26. The Fmoc group was
replaced by an acetamide, delivering 27.
Scheme 4. Synthesis of a D-ring hexapeptide thioether.
Disulfide 28 was formed and converted, under the conditions optimized above, to
give thioether 29. While we could see evidence of low levels of epimerization in the 1H
NMR spectra for the B-ring tetrapeptides (vide supra), the NMR and chromatographic
behavior of these proline-containing peptides were complicated by conformational issues.
Thus, it was not possible to accurately quantify the minor epimer by HPLC for the
peptides in Schemes 1 and 3. On the other hand, the two diastereomers of 29 could be
separated and quantified by HPLC, indicating that epimerization was less than 4%
(Figures S1 and S2).

ACCEPTED MANUSCRIPT
11
Tetrapeptide 39 (Scheme 5) is the thioether analog of microbisporicin’s
aminovinylcysteine (AviCys)41,42 E-ring. A cyclic tetrapeptide, analogous to 39,
containing histidine in place of phenylalanine occurs in the D-ring of planosporicin.4 In
addition to the PheꢀHis difference, the E-ring of nisin bears a LanꢀMeLan
substitution.43 This specific tetrapeptide thioether has been predicted previously, based
on gene sequence, in subtilomycin44 and a bacteriocin produced by Listeria
monocytogenes.45 The E-ring thioether analog is thus of interest for structure-activity
studies and presents the opportunity to relay lessons learned with the B-ring tetrapeptide
to a tetrapeptide in which three of four residues are Cys.
Fmoc-Phe-OH (33) was activated and coupled with L-cysteine building block 34,
side-chain protected with the acetamidomethyl (Acm) group (Scheme 5). Dipeptide acid
35 was coupled with amine 32 to give 36. Homologation in the N-terminal direction,
with D-cysteine building block 6b gave tetrapeptide 37. The oxidation of tetrapeptide 37
was expected to proceed chemoselectively to give disulfide 38, with the trityl ethers
undergoing co-oxidation in the presence of the Acm-protected Cys residue. A modified
solvent system was required to afford 38 in reasonable yield. Trifluoroethanol (TFE) was
originally recommended by Kamber et al.46 and used to advantage by Shiba and
coworkers.22 We propose that the TFE masks the –Acm group through hydrogen
bonding interactions, allowing for selective activation of the STrt via the sulfenyl iodide.
Reductive desulfurization was accomplished over a 5 day period with the concomitant
transesterification to methyl ester 39. Subjection of Fmoc-D-Cys(Trt)-OAll (31) to
methanol for 5 days also led to transesterification, indicating that the HMPT was not

ACCEPTED MANUSCRIPT
12
necessarily involved and that the conversion of the allyl ester to the corresponding methyl
ester is merely a consequence of the high concentration of methanol.
Scheme 5. Synthesis of an E-ring tetrapeptide thioether analog.
STrt
allyl bromide
Cs2CO3, DMF
STrt
R-NH
COOAll
Fmoc-NH
Fmoc-NH
COOH
31 R = Fmoc
32 R = H
30
piperidine
CH2Cl2
Ph
H
N
Ph
1. NHS, DCC
COOH
SAcm
Fmoc-NH
COOH
2.
H2N
COOH
33
O
35
34
SAcm
O
32, HATU
CH2Cl2
collidine
Ph
STrt
COOAll
1. piperidine
CH2Cl2
H
N
Fmoc-NH
N
H
(98%)
2. 6b, HATU
CH2Cl2
O
SAcm
36
NHBoc
STrt
NHBoc
S
O
O
I2
CH2Cl2
CF3CH2OH
NH
NH
STrt
COOAll
NH
NH
S
Ph
Ph
H
N
H
N
O
O
COOAll
(58%)
AcmS
O
AcmS
O
37
38
HN
NHBoc
1
O
O
NH
NH
S
2
NH
S
Ph
Ph
P(NMe2)3
4
H
N
H
N
O
O
NH
3
COOMe
MeOH
(60%)
S
O
AcmS
O
E-ring of
microbisporicin A2
39
Analysis of thioether 39 by HPLC indicated a major diastereomer (>95%) and
minor diastereomers (<5%) (Figure S3). Partial purification by RP-HPLC gave the major
diastereomer in ~99% purity. LC-MS analysis showed two residual minor products, that
eluted slightly earlier than the major product, accounting for 0.38% and 0.85% of the
mixture (Figure S5). Compound 39 (~99% pure) and the two minor products each gave
rise to a species at m/z 636 with exact mass measurements consistent with their being
diastereomers arising from epimerization at either Cys1 or Cys4 via a dehydroalanine
intermediate. Interestingly, two species were observed in the 1H NMR of compound 39
(~99% pure) in a 2.6:1.0 ratio. The related ring systems of nisin47 and planosporicin48

ACCEPTED MANUSCRIPT
13
have been studied by NMR but only in the context of the parent peptides, with additional
conformational constraints. We note that the precursor disulfide 38 shows a single
species by NMR and much better solubility properties than the derived thioether 39,
suggesting that hydrogen bonding and a highly ordered structure are likely. Based on our
HPLC and LC-MS analysis, we were confident of the stereochemical integrity of the
product and so conducted assorted NMR experiments to investigate the nature of the two
species. No cross-peaks between signals of the two species were discernible in NOESY
or ROESY spectra. Variable temperature NMR experiments did not lead to a
coalescence of signals but two amide proton resonances demonstrated little temperature
dependence, indicative of intramolecular hydrogen-bonding (Figure S6). Notably, the
ring amide of L-Cys4 was more strongly hydrogen bonded in the minor species than in
the major; modeling suggests that the hydrogen bond acceptor is the amide carbonyl of
Phe2.
3.
Conclusions
In summary, we have developed a generalized reductive desulfurization method
for conversion of cyclic disulfides to thioether macrocycles with good stereochemical
control. A suite of microbisporicin-derived peptides enabled us to probe the reaction in
different ring sizes, while the E-ring analog introduced the concept of a third,
orthogonally protected cysteine residue. The method capitalizes on the facile synthesis of
disulfide macrocycles and has been applied to cyclic peptides containing four, six and
eight amino acid residues to generate 13-, 19- and 25-membered rings respectively.
Incorporation of trityl and acetamidomethyl protecting groups into a peptide provides

ACCEPTED MANUSCRIPT
14
control over the regioselectivity of disulfide formation. Future work will include
application of the reaction to longer sequences containing multiple thioether rings.
4.
Experimental Section
4.1
General Procedures. RP-HPLC was performed on a Waters 600 system
interfaced with Empower™ software and equipped with a Waters 2487 UV detector,
monitoring absorption at 218 and 254 nm. Columns were 250 mm long. Analytical work
was conducted on a 4.6 mm diameter Altima® C18 column with a flow rate of 1 mL min-
1. Semi-preparative work was conducted on a 10 mm diameter Altima® C18 column
with a flow rate of 3 mL min-1. Preparative work was conducted on a 22 mm diameter
Econosil® 10 m C18 column with a flow rate of 12 mL min-1. Solvent A is 0.1% w/v
TFA in water and Solvent B is 0.1% w/v TFA in MeCN. The configuration of Cys
residues is specified in each case: the C-terminal residue is L and the N-terminal residue
is D, as occurs in lantibiotics. Unless otherwise noted, amino acids are of the L-
configuration.
4.2
Synthesis of B-Ring Tetrapeptides
4.2.1 N-Fmoc-Gly-L-Cys(Trt)-NH2 (4). 1-Ethyl-3-(3-dimethylamino-propyl)
carbodiimide (232 mg, 1.2 mmol, 1.2 equiv.) was added to a solution of Fmoc-Gly-OH
(2) (300 mg, 1.0 mmol, 1 equiv.) in CH2Cl2 (30 mL) under N2 at rt and stirred for 15 min.
H.L-Cys(Trt)-NH2 (3) (329 mg, 0.91 mmol, 0.9 equiv.) was added and stirring continued
for 16 h. The reaction mixture was diluted with CH2Cl2 (100 mL) and washed with water
(50 mL). The aqueous layer was back-extracted with EtOAc (50 mL) and the organic

ACCEPTED MANUSCRIPT
15
layers were separately washed with 5% aq. NaHCO3 (50 mL), 1 M aq. KHSO4 (50 mL)
and brine (50 mL), filtered through MgSO4, and concentrated. The residue was subjected
to flash chromatography, eluting with 50-70% EtOAc in hexanes to yield the dipeptide N-
Fmoc-Gly-L-Cys(Trt)-NH2 (4) as a colorless solid (510 mg, 79%). Rf 0.15 (1:1 EtOAc-
hex); 0.35 (9:1 CH2Cl2-MeOH). [α]D25 +6.3 (c 0.84, CHCl3). 1H NMR (CDCl3, 500
MHz) δ 2.55 (dd, J = 13.1, 5.5 Hz, 1H), 2.78 (dd, J = 12.8, 6.9 Hz, 1H), 3.73 (d, J = 5.4
Hz, 2H), 4.11-4.13 (m, 1H), 4.15 (t, J = 6.7 Hz, 1H), 4.34 (d, J = 6.7 Hz, 2H), 5.54 (br s,
2H), 6.09 (br s, 1H), 6.48 (br d, J = 7.3 Hz, 1H), 7.18 (t, J = 7.0 Hz, 2H), 7.23-7.30 (m,
9H), 7.35-7.40 (m, 8H), 7.54 (t, J = 6.2 Hz, 2H), 7.73 (d, J = 7.5 Hz, 2H); 13C NMR
(125 MHz, CDCl3) δ 33.2, 44.5, 47.1, 51.9, 67.3, 120.0, 125.0, 127.0, 127.1, 127.8,
128.1, 129.5, 141.3, 143.7, 144.3, 156.7, 169.0, 172.0. HRMS (ESI-TOF) m/z: [M + Na]+
Calcd for C39H35N3NaO4S 664.2240; Found 664.2223.
4.2.2 H-Gly-L-Cys(Trt)-NH2 (5). Piperidine (3 mL) was added to a solution of dipeptide
4 (1.2 g, 1.8 mmol) in CH2Cl2 (12 mL) under N2 at rt, the mixture stirred for 30 min .
The mixture was concentrated and the residue was subjected to flash chromatography,
eluting with 1-10% MeOH in CH2Cl2 to yield H-Gly-L-Cys(Trt)-NH2 (5) as a colorless
solid (700 mg, 92%) which was used directly in the next step. Rf 0.25 (9:1 CH2Cl2-
MeOH).
4.2.3 Fmoc-D-Cys(Trt)-Pro-OH (8a). 1-Ethyl-3-(3-dimethylamino-propyl)
carbodiimide (117 mg, 0.61 mmol, 1.2 equiv.) and HOBt (94 mg, 0.61 mmol, 1.2 equiv.)
were added sequentially to a solution of Fmoc-D-Cys(Trt)-OH (6a) (300 mg, 0.51 mmol,

ACCEPTED MANUSCRIPT
16
1.0 equiv.) in CH2Cl2 (15 mL) at rt under N2. The mixture was stirred for 30 min and then
H-L-Pro-OH (7) (58 mg, 0.51 mmol, 1.0 equiv) was added. After 3 h, the mixture was
diluted with CH2Cl2 (60 mL) and washed successively with 1 M aq. KHSO4 (60 mL),
water (60 mL) and brine (60 mL), dried over MgSO4, filtered and concentrated. Flash
chromatography was performed, eluting with 1-10% MeOH in CH2Cl2 to afford Fmoc-D-
Cys(Trt)-Pro-OH (8a) as a colorless solid (298 mg, 85%) that was used directly in the
next step. Rf 0.35 (95:5 CH2Cl2-MeOH).
4.2.4 Boc-D-Cys(Trt)-Pro-OH (8b). By analogy to the preparation of 8a on a scale of
0.65 mmol of Boc-D-Cys(Trt)-OH (6b). The mixture was stirred for 23 h, diluted with
EtOAc (60 mL) and washed with 10% aq. citric acid (60 mL), water (60 mL) and brine
(60 mL), dried over MgSO4, filtered and concentrated. Flash chromatography was
performed eluting with of 1-7% MeOH in CH2Cl2 to give dipeptide acid Boc-D-Cys(Trt)-
Pro-OH (8b) as a colorless foam (333 mg, 92%) that was used directly in the next step. Rf
0.15 (93:7 CH2Cl2: MeOH).
4.2.5 Nα-Fmoc Linear Tetrapeptide 9a. 1-Ethyl-3-(3-dimethylamino-propyl)
carbodiimide (129 mg, 0.68 mmol, 1.6 equiv.) and HOBt (103 mg, 0.68 mmol, 1.6
equiv.) were added sequentially to a solution of dipeptide acid Fmoc-D-Cys(Trt)-Pro-OH
(8a) (289 mg, 0.42 mmol, 1.0 equiv.) in CHCl3 (30 mL) under N2 at rt and stirred for 15
min, followed by the addition of dipeptide amine H-Gly-L-Cys(Trt)-NH2 (5) (172 mg,
0.42 mmol, 1.0 equiv.) as a solid. Stirring was continued for 16 h. The mixture was
diluted with CHCl3 (100 mL) and washed with water (50 mL). The aqueous layer was

ACCEPTED MANUSCRIPT
17
back-extracted with EtOAc (50 mL) and each organic layer was washed separately with
5% aq. NaHCO3 (50 mL), 1 M aq. KHSO4 (50 mL) and brine (50 mL). The combined
organic layers were filtered through MgSO4 and concentrated. The residue was subjected
to flash chromatography eluting with 1-4% MeOH in CH2Cl2 to yield tetrapeptide 9a as a
colorless foam (387 mg, 84%). Rf 0.51 (95:5 CH2Cl2-MeOH). [α]D25 -11.7 (c 0.91,
CHCl3). 1H NMR (500 MHz, CDCl3) δ 1.82-1.89 (m, 1H), 1.91-1.96 (m, 1H), 1.98-2.04
(m, 1H), 2.10-2.18 (m, 1H), 2.48 (dd, J = 13.0, 7.3 Hz, 1H), 2.66 (dd, J = 13.0, 7.3 Hz,
1H), 2.74 (dd, J = 13.1, 5.2 Hz, 1H), 2.79 (dd, J = 13.1, 9.5 Hz, 1H), 3.19 (dd, J = 16.8,
7.3 Hz, 1H), 3.52 (dd, J = 17.0, 5.9 Hz , 1H), 3.56-3.60 (m, 1H), 3.66 (dd, J = 12.7, 6.7
Hz, 1H), 3.88 (dd, J = 17.0, 6.3 Hz, 1H), 3.97 (t, J = 7.0 Hz, 1H), 4.05 (dd, J = 10.2, 6.3
Hz, 1H), 4.16 (td, J = 8.5, 5.2 Hz, 1H), 4.28 (dd, J = 10.3, 7.9 Hz, 1H), 4.32 (dd, J = 8.5,
4.7 Hz, 1H), 5.09 (d, J = 5.2 Hz, 1H), 5.13 (br s, 1H), 6.51 (br s, 1H), 7.09 (t, J = 7.2 Hz,
2H), 7.15 (t, J = 7.5 Hz, 6H), 7.23-7.26 (m, 5H), 7.29-7.35 (m, 15H), 7.40-7.45 (m, 9H),
7.72 (dd, J = 9.7, 7.8 Hz, 2H); 13C NMR (125 MHz, CDCl3) δ 24.7, 28.9, 32.6, 33.1,
43.6, 46.7, 47.7, 52.5, 53.0, 61.5, 66.9, 67.4, 67.7, 120.0, 124.9, 125.0, 126.7, 127.1,
127.2, 127.3, 127.8, 127.9, 128.0, 128.2, 129.5, 129.6, 141.2, 141.3, 143.1, 143.8, 144.1,
144.5, 146.9, 156.9, 169.1, 171.0, 172.4, 172.8. HRMS (ESI-TOF) m/z: [M + Na]+ Calcd
for C66H61N5NaO6S2 1106.3955; Found 1106.3984.
4.2.6 Nα-Boc Linear Tetrapeptide 9b. 1-Ethyl-3-(3-dimethylaminopropyl)
carbodiimide (149 mg, 0.77 mmol, 1.4 equiv.) and HOBt (102 mg, 0.77 mmol, 1.4
equiv.) were added sequentially to a solution of Boc-D-Cys(Trt)-Pro-OH (8b) (307 mg,
0.55 mmol, 1.0 equiv.) in CH2Cl2 (20 mL) at rt under N2. After 15 min, solid H-Gly-L-

ACCEPTED MANUSCRIPT
18
Cys(Trt)-NH2 (5) (253 mg, 0.66 mmol, 1.1 equiv.) was added with stirring. After 19.5 h,
the mixture was diluted with CH2Cl2 (80 mL) and washed with water (50 mL), 10% aq.
Na2CO3 (50 mL), water (50 mL), 10% aq. citric acid (50 mL), water (50 mL) and brine
(50 mL). The organic layer was filtered through MgSO4 and concentrated. The residue
was subjected to flash chromatography eluting with 1-4% MeOH in CH2Cl2 to afford the
25
tetrapeptide 9b as a colorless foam (326 mg, 62%). Rf 0.52 (9:1 CH2Cl2-MeOH). [α]D
-
13.6 (c 0.95, CHCl3). 1H NMR (CDCl3, 500 MHz) δ 1.17 (s, 9H), 1.74-1.80 (m, 1H),
1.83-1.92 (m, 2H), 2.04-2.12 (m, 1H), 2.37 (dd, J = 13.0, 7.7 Hz, 1H), 2.48 (dd, J = 13.0,
7.2 Hz, 1H), 2.62 (dd, J = 13.2, 9.8 Hz, 1H), 2.70 (dd, J = 13.2, 4.8 Hz, 1H), 3.10 (dt, J =
9.8, 7.1 Hz, 1H), 3.38 (dd, J = 17.0, 5.6 Hz, 1H), 3.46 (app q, J = 7.0 Hz, 1H), 3.55-3.60
(m, 1H), 3.88 (dd, J = 17.0, 6.6 Hz, 1H), 4.09 (td, J = 9.0, 4.7 Hz, 1H), 4.21 (dd, J = 8.7,
4.5 Hz, 1H), 4.68 (d, J = 4.0 Hz, 1H), 5.07 (br s, 1H), 6.49 (s, 1H), 7.06-7.09 (m, 3H),
7.13-7.17 (m, 10H), 7.20-7.23 (m, 6H) 7.31-7.33 (m, 12H), 7.36 (t, J = 6.1, 1H). 13C
NMR (CDCl3, 125 MHz) δ 24.7, 28.2, 29.1, 32.5, 33.0, 43.8, 47.7, 52.2, 53.1, 61.6, 67.0,
67.6, 81.0, 126.7, 127.2, 128.0, 128.2, 129.6, 129.7, 144.1, 144.7, 156.4, 169.2, 171.3,
172.6, 173.1. HRMS (ESI-TOF) m/z: [M + Na]+ calcd for C56H59N5NaO6S2 984.3799;
Found 984.3803.
4.2.7 Nα-Acetyl Linear Tetrapeptide 9c. Piperidine (2 mL) was added to a solution of
tetrapeptide 9a (387 mg, 0.35 mmol) in CH2Cl2 (11 mL) under N2 at rt. The mixture was
stirred for 30 min and concentrated. The residue was subjected to flash chromatography,
eluting with 1-10% MeOH in CH2Cl2 to yield the intermediate amine as a colorless solid
(211 mg, 69%). Rf 0.25 (95:5 CH2Cl2-MeOH).

ACCEPTED MANUSCRIPT
19
Acetic anhydride (51 µL, 55 mg, 0.53 mmol, 2.2 equiv.) was added to a solution
of the amine (211 mg, 0.24 mmol, 1.0 equiv.) in EtOAc (15 mL) under N2 at 0 °C,
followed by K2CO3 (85 mg, 0.61 mmol, 2.5 equiv). The mixture was stirred for 5 min at
0 °C, warmed to rt and, after 20 min, diluted with EtOAc (100 mL) and washed with
sat’d aq. K2CO3 (50 mL). The basic aqueous layer was further extracted with EtOAc (2 ×
40 mL) and the organic layers were combined, washed with water (50 mL) and brine (50
mL), filtered through MgSO4 and concentrated. The residue was subjected to flash
chromatography eluting with 1-5% MeOH in CH2Cl2 to yield tetrapeptide 9c as a
colorless solid (212 mg, 96%). Rf 0.25 (95:5 CH2Cl2-MeOH). [α]D25 -21.4 (c 0.95,
CHCl3). 1H NMR (500 MHz, CDCl3) δ 1.65 (s, 3H), 1.85-1.91 (m, 1H), 1.94-2.03 (m,
2H), 2.13-2.21 (m, 1H), 2.54 (dd, J = 12.8, 7.6 Hz, 1H), 2.68 (dd, J = 12.8, 7.5 Hz, 1H),
2.81 (dd, J = 13.1, 5.2 Hz, 1H), 2.83 (dd, J = 13.1, 9.4 Hz, 1H), 3.25 (dt, J = 9.9, 7.1 Hz,
1H), 3.52 (dd, J = 17.0, 6.1 Hz, 1H), 3.73-3.77 (m, 1H), 3.84 (td, J = 7.4, 5.1 Hz, 1H),
3.91 (dd, J = 17.0, 6.1 Hz, 1H), 4.15 (td, J = 8.9, 5.3 Hz, 1H), 4.31 (dd, J = 8.6, 4.2 Hz,
1H), 5.20 (br s, 1H), 5.79 (d, J = 4.9 Hz, 1H), 6.63 (br s, 1H), 7.18-7.21 (m, 3H), 7.23-
7.26 (m, 10H), 7.29-7.35 (m, 12H), 7.40-7.42 (m, 6H), 7.46 (t, J = 6.0 Hz, 1H); 13C NMR
(125 MHz, CDCl3) δ 22.7, 24.7, 29.0, 32.2, 33.3, 43.6, 47.7, 51.8, 53.1, 61.7, 66.9, 67.7,
126.9, 127.2, 128.0, 128.2, 129.5, 129.6, 144.1, 144.5, 169.2, 170.9, 172.3, 172.5, 173.0.
HRMS (ESI-TOF) m/z: Calcd for [M + Na]+ C53H53N5NaO5S2 926.3380; Found
926.3395.
4.2.8 Nα-Fmoc Tetrapeptide Disulfide 10a. A solution of linear tetrapeptide 9a (36
mg, 0.03 mmol, 1.0 equiv.) in CHCl3 (15 mL) and MeOH (15 mL) was added dropwise

ACCEPTED MANUSCRIPT
20
to a solution of iodine (33 mg, 0.13 mmol, 4.0 equiv.) in MeOH (30 mL) over a period of
30 min. The mixture was stirred for an additional 20 min, followed by the dropwise
addition of 0.2 M aq. Na2S2O3 until the reaction mixture became colorless. The mixture
was concentrated and the residue dissolved in CHCl3 (50 mL), washed with 0.1 M aq.
Na2S2O3 (20 mL), water (20 mL) and brine (20 mL). The organic layer was filtered
through MgSO4 and concentrated. The residue was subjected to flash chromatography
eluting with 1-6% MeOH in CH2Cl2 to yield tetrapeptide disulfide 10a as a colorless
solid (20 mg, 91%). Rf 0.16 (95:5 CH2Cl2-MeOH). [α]D25 +94.2 (c 0.53, MeOH). 1H
NMR (500 MHz, DMSO-d6) δ 1.63-1.72 (m, 1H), 1.76-1.83 (m, 1H), 1.99-2.03 (m, 1H),
2.20-2.27 (m, 1H), 2.77 (dd, J = 14.1, 3.7 Hz, 1H), 2.83 (dd, J = 14.1, 3.7 Hz, 1H), 2.98-
3.11 (m, 2H), 3.36-3.39 (m, 1H), 3.49 (ddd, J = 11.7, 8.4, 3.2 Hz, 1H), 3.67 (br d, J =
11.1Hz, 1H), 3.77 (dd, J = 14.0, 3.8 Hz, 1H ), 4.20 (dd, J = 9.7, 7.7 Hz, 1H), 4.25 (app. t,
J = 7.0 Hz, 1H), 4.32 (ddd, J = 12.0, 8.8, 3.8 Hz, 1H), 4.37-4.40 (m, 1H), 4.43 (dd, J =
9.7, 6.4 Hz, 1H ), 4.92 (d, J = 7.4 Hz, 1H), 7.23 (br s, 1H), 7.32 (tdd, J = 7.5, 3.9, 0.6 Hz,
2H), 7.42 (t, J = 7.5 Hz, 2H), 7.48 (br s, 1H), 7.61 (d, J = 8.5 Hz, 1H), 7.70 (t, J = 8.5 Hz,
2H), 7.88 (d, J = 7.5 Hz, 2H), 8.00 (d, J = 8.5 Hz, 1H), 8.75 (t, J = 5.9 Hz, 1H); 13C NMR
(125 MHz, DMSO-d6) δ 22.3, 32.1, 38.7, 41.2, 43.8, 47.0, 47.4, 50.1, 50.9, 60.4, 66.5,
120.5, 125.5, 125.6, 127.6, 127.7, 128.3, 141.1, 143.7, 143.9, 157.1, 170.4, 170.5 172.3,
173.4. HRMS (ESI-TOF) m/z: [M + H]+ Calcd for C28H32N5O6S2 598.1789; Found
598.1793.
4.2.9 Nα-Boc Tetrapeptide Disulfide 10b. A solution of linear tetrapeptide 9b (350
mg, 0.37 mmol, 1.0 equiv.) in CHCl3 (100 mL) and MeOH (100 mL) was added

ACCEPTED MANUSCRIPT
21
dropwise to a solution of iodine (325 mg, 1.27 mmol, 3.5 equiv.) in MeOH (200 mL)
over a period of 1 h under N2. The mixture was stirred for an additional 20 min, followed
by the dropwise addition of 0.2 M aq. Na2S2O3 (approx. 3 mL) until the mixture became
colorless. The mixture was concentrated, and the residue dissolved in CHCl3 (150 mL),
washed with 0.1 M aq. Na2S2O3 (100 mL), water (150 mL) and brine (150 mL). The
organic layer was filtered through MgSO4 and concentrated. The residue was subjected to
flash chromatography eluting with 1-8% MeOH in CH2Cl2 to yield tetrapeptide disulfide
10b as a colorless solid (109 mg, 63%). Rf 0.4 (9:1 CH2Cl2-MeOH). [α]D25 +110.5 (c
0.53, MeOH). 1H NMR* (500 MHz, CD3OD) δ 1.46 (s, 9H), 1.78-1.88 (m, 1H), 1.90-
1.96 (m, 1H), 2.13-2.18 (m, 1H), 2.40-2.47 (m, 1H), 2.81 (dd, J = 14.3, 11.7 Hz, 1H),
2.93 (dd, J = 14.6, 3.4 Hz, 1H), 3.08 (app. t, J= 10.8, 1H), 3.27-3.29 (m, 1H, obscured by
CH3OH signal), 3.47-3.53 (m, 1H), 3.58-3.63 (m, 1H), 3.78 (d, J = 16.5 Hz, 1H), 4.06 (d,
J = 16.4 Hz, 1H), 4.40 (dd, J = 11.4, 3.4 Hz, 1H), 4.45-4.52 (m, 1H), 4.93 (d, J = 7.3 Hz,
1H), 13C NMR* (125 MHz, CD3OD) δ 22.1, 27.4, 31.8, 39.1, 42.0, 43.1, 47.1, 49.3, 51.4,
60.1, 79.8, 157.0, 170.8, 171.1, 172.8, 173.6. HRMS (ESI-TOF) m/z: [M + Na]+ Calcd
for C18H29N5NaO6S2 498.1451; Found 498.1466.
*peaks reported for major conformer only
4.2.10 Nα-Acetyl Tetrapeptide Disulfide 10c. A solution of tetrapeptide 9c (212 mg,
0.23 mmol, 1.0 equiv.) in CHCl3 (75 mL) and MeOH (75 mL) was added dropwise to a
solution of iodine (208 mg, 0.82 mmol, 3.5 equiv.) in MeOH (100 mL) over a period of 1
h 30 min. The mixture was stirred for an additional 20 min, followed by the addition of
pre-washed Amberlite® IRA-400 Type II basic resin (10 g) and stirred gently until the

ACCEPTED MANUSCRIPT
22
reaction mixture became colorless. The mixture was filtered to remove the resin, which
was washed with MeOH. The filtrate was concentrated and the residue subjected to flash
chromatography eluting with 5-15% MeOH in CHCl3 to yield tetrapeptide disulfide 10c
as colorless solid (97 mg, 99%). Rf 0.12 (9:1 CHCl3-MeOH). [α]D25 +145.2 (c 0.52,
MeOH). 1H NMR* (500 MHz, CD3OD) δ 1.80-1.86 (m, 1H), 1.89-1.94 (m, 1H), 1.99 (s,
3H), 2.13-2.18 (m, 1H), 2.36-2.44 (m, 1H), 2.82 (dd, J = 14.5, 11.7 Hz, 1H), 2.96 (dd, J =
14.5, 3.5 Hz, 1H), 3.05-3.14 (m, 1H), 3.16-3.21 (m, 1H), 3.45-3.51 (m, 1H), 3.61 (ddd, J
= 11.7, 8.2, 3.6 Hz, 1H), 3.86 (app. d, J = 15.9 Hz, 1H), 3.98 (app. d, J = 15.9, 1H), 4.49-
4.58 (m, 1H), 4.71 (dd, J = 11.5, 3.5 Hz, 1H), 5.00 (d, J = 6.7 Hz, 1H), 8.02 (d, J = 7.4
Hz, 1H), 8.72 (t, J = 5.8 Hz, 1H); 13C NMR* (125 MHz, CD3OD) δ 20.8, 22.0, 31.8,
38.2, 41.5, 43.4, 47.0, 48.4, 51.3, 60.4, 170.4, 170.7, 171.9, 172.9, 173.7. HRMS (ESI-
TOF) m/z: [M + Na]+ Calcd for C15H23N5NaO5S2 440.1033; Found 440.1036.
* peaks reported for major conformer only
4.2.11 Nα-Boc Tetrapeptide Thioether 11b. A solution of HMPT (20 µL, 17.8 mg, 0.11
mmol, 1.3 equiv.) in dry methanol (~200 µL) was added dropwise over 20 min to a
solution of tetrapeptide disulfide 10b (40 mg, 0.08 mmol, 1.0 equiv.) in dry, degassed
methanol (40 mL) under N2 and the mixture was stirred at rt for 16 h. Another aliquot of
HMPT (20 µL in ~200 µL dry methanol) was added after 16 h and stirring continued for
a total of 40 h. After completion of reaction, as evidenced by NMR of aliquots, the
mixture was concentrated, the solid was triturated with ether (3 × 10 mL) and the
resulting solid further purified by flash chromatography eluting with 5-15% CH3OH in
CH2Cl2 to afford the tetrapeptide thioether 11b as a colorless solid (23 mg, 61%). Rf 0.37

ACCEPTED MANUSCRIPT
23
(9:1 CH2Cl2-MeOH). [α]D25 +54.4 (c 0.41, MeOH). 1H NMR* (500 MHz, CD3OD) δ
1.46 (s, 9H), 1.89-1.98 (m, 2H), 2.05-2.15 (m, 1H), 2.35-2.43 (m, 1H), 2.55 (dd, J = 15.0,
3.3 Hz, 1H), 2.60 (dd, J = 14.0, 3.6 Hz, 1H), 2.94 (dd, J = 14.0, 11.3 Hz, 1H), 3.43 (dd, J
= 15.0, 4.4 Hz, 1H), 3.48-3.663 (m, 2H), 3.87 (d, J = 16.3 Hz, 1H), 3.95 (d, J = 16.4 Hz,
1H), 4.21 (dd, J = 11.2, 3.5 Hz, 1H), 4.76 (dd, J = 8.9, 3.2 Hz, 1H), 4.87 (app. t, J = 3.7
Hz, 1H); 13C NMR* (125 MHz, CD3OD) δ 23.7, 28.8, 33.1, 37.9, 38.1, 45.3, 48.5, 53.8,
55.3, 61.4, 81.1, 158.7, 172.0, 172.1, 174.5, 175.6. HRMS (ESI-TOF) m/z: [M + H]+
Calcd for C18H30N5O6S 444.1911; Found 444.1912.
*peaks reported for major conformer only
4.2.12 Nα-Acetyl Tetrapeptide Thioether 11c. A solution of HMPT (17 µL, 15.2 mg,
0.09 mmol, 1.3 equiv.) in dry methanol (~150 µL) was added dropwise over 30 min to a
solution of tetrapeptide disulfide 10c (30 mg, 0.07 mmol, 1.0 equiv.) in dry, degassed
MeOH (30 mL) under N2 and the mixture stirred at rt for 14 h. Another equivalent of
HMPT (17 µL in ~150 µL dry methanol) was added and stirring continued for a total of
30 h. After completion of reaction, as evidenced by NMR of aliquots, the mixture was
concentrated and the solid was triturated with ether (3 × 10 mL). The residual solid was
further purified by HPLC with a gradient of 100% A for 3 min; 0-3% B over 5 min;
3ꢀ45% B over 27 min; 45ꢀ95% over 3 min; hold at 95% B for 5 min; 95ꢀ0% B
over 2 min and hold at 100% A for 5 min. Upon freezedrying of appropriate fractions,
thioether 11c was obtained as a colorless solid (19 mg, 67%). [α]D25 +51.5 (c 0.26,
1
CH3OH). H NMR* (500 MHz, CD3OD) δ 1.87-1.93 (m, 2H), 1.98 (s, 3H), 2.04-2.10
(m, 1H), 2.34-2.43 (m, 1H), 2.61 (dd, J = 15.1, 3.3 Hz, 1H), 2.67 (dd, J = 14.2, 3.8 Hz,

ACCEPTED MANUSCRIPT
24
1H), 2.98 (dd, J = 14.2, 11.2 Hz, 1H), 3.39 (dd, J = 15.1, 4.7 Hz, 1H), 3.47-3.52 (m, 1H),
3.59 (ddd, J = 12.1, 7.3, 4.8 Hz, 1H), 3.85 (app. d, J = 16.4 Hz, 1H), 3.94 (app. d, J =
16.4, 1H), 4.47 (dd, J = 11.2, 3.7 Hz, 1H), 4.79 (dd, J = 8.8, 3.5 Hz, 1H), 4.85 (1H,
obscured by H2O signal but evident in COSY); 13C NMR* (125 MHz, CD3OD) δ 22.2,
23.8, 33.1, 37.6, 38.1, 45.3, 48.5, 54.0, 54.4, 61.5, 171.4, 172.0 173.8, 174.4, 175.6.
HRMS (ESI-TOF) m/z: [M + Na]+ Calcd for C15H23N5NaO5S 408.1312; Found
408.1312.
*peaks reported for major conformer only
4.3 Synthesis of C-Ring Octapeptides
4.3.1 Ac-D-Cys-Hyp-Gly-Gly-Gly-D-Ala-Asn-L-Cys-NH2 (12). The octapeptide was
synthesized by SPPS on a 100 µmol scale. Rink Amide SS resin (189 mg, 0.53 mmol g-1,
1 equiv.) was placed in a vessel and end capped. DMF (3 mL) was added and the resin
left to swell for 30 min. The solvent was drained from the resin. Fmoc cleavage was
carried out by adding 20% piperidine in NMP (3 mL) to the resin in the stoppered vessel
and was shaken for 5 min. The solution was drained and the process repeated twice. The
resin was washed with DMF (3 mL) for 30 s. A solution of Fmoc-L-Cys(Thp)-OH (213.5
mg, 0.5 mmol, 5 equiv.), HCTU (206.8 mg, 0.5 mmol, 5 equiv.) and 0.4 M 4-
methylmorpholine (NMM) in NMP (3 mL) was prepared in a vial and the mixture stirred
for 2 min to activate. The activated amino acid solution was added to the resin and the
vessel shaken for 20 min. The resin washed with DMF for 30 seconds (5 x 3 mL).
Fmoc cleavage was performed as described above and, after washing, a solution
of Fmoc-Asn(Trt)-OH (298.4 mg, 0.5 mmol, 5 equiv.) and HCTU (206.8 mg, 0.5 mmol,

ACCEPTED MANUSCRIPT
25
5 equiv.) and 0.4 M 4-methylmorpholine (NMM) in NMP (3 mL) were added and
procedure repeated as described for the coupling of Fmoc-L-Cys(Thp)-OH. This protocol
for elongation was repeated for remaining amino acids to give the octapeptide, Fmoc-D-
Cys-Hyp-Gly-Gly-Gly-D-Ala-Asn-L-Cys-NH2.
Fmoc-cleavage was performed as described above. A solution of NMP/acetic
anhydride/NMM (3:1:1; 3 mL) was added and the mixture shaken for 20 min. The
solution was drained and resin washed with DMF for 30 seconds (5 x 3 mL). The resin
was washed with DCM (3 mL) for 30 seconds and dried completely.
A cleavage cocktail of TFA/EDT/water/TIPS (94:2.5:2.5:1; 3 mL) was added and
the suspension left to stand for 2 h. The solution was drained into a 50 mL centrifuge
tube. Cold ether (30 mL) was added to precipitate the peptide and the supernatant was
removed after centrifugation for 10 min (× 5). The precipitate was dissolved in 0.1%
aqueous TFA (4 mL) and lyophilized to give linear octapeptide 12 as a colorless powder
that was used directly in the next step. [α]D25 -1.2 (c 0.34, H2O). 1H NMR* (500 MHz,
DMSO-d6) δ 1.19 (d, J = 7.0 Hz, 3H), 1.86 (s, 3H), 1.90-1.96 (m, 1H), 1.99-2.05 (m, 1H),
2.31 (dd, J = 9.4, 7.7 Hz, 1H), 2.38 (t, J = 8.5 Hz, 1H), 2.45 (dd, J = 15.6, 6.9 Hz, 1H),
2.52-2.57 (m, 1H), 2.60 (dd, J = 15.6, 6.5 Hz, 1H), 2.72-2.79 (m, 1H), 2.85 (ddd, J =
13.7, 9.6, 4.2 Hz, 1H), 3.38 (d, J = 4.1 Hz, 1H, partially obscured by H2O), 3.48 (app. d, J
= 10.7 Hz, 1H, completely obscured by H2O but visible in COSY), 3.64 (dd, J = 10.8, 4.4
Hz, 1H), 3.70-3.78 (m, 6H), 4.19-4.23 (m, 2H), 4.34-4.36 (m, 2H), 4.48 (app. q, J = 6.8
Hz, 1H), 4.58 (app. q, J = 7.0 Hz, 1H), 6.97 (s, 1H), 7.20 (s, 1H), 7.30 (s, 1H), 7.41 (s,
1H), 7.92 (d, J = 8.3 Hz, 1H), 7.99 (t, J = 5.7 Hz, 1H), 8.05 (t, J = 5.8 Hz, 1H), 8.11 (d, J
= 6.7 Hz, 1H), 8.25 (m, 1H), 8.30 (d, J = 7.6 Hz, 1H); 13C NMR* (125 MHz, DMSO-d6)

ACCEPTED MANUSCRIPT
26
δ 17.8, 22.3, 25.3, 25.8, 36.7, 37.4, 41.9, 42.1, 42.2, 48.6, 50.0, 53.0, 55.1, 55.3, 59.0,
68.6, 168.8, 169.2, 169.3 (2C), 169.4, 170.9, 171.7 (2C), 171.9, 172.5. HRMS (ESI-TOF)
m/z: [M + Na]+ Calcd for C26H42N10NaO11S2 757.2368; Found 757.2364.
*peaks reported for major conformer only
4.3.2 Octapeptide disulfide 13. A solution of linear octapeptide 12 (80 mg, 0.11 mmol,
1.0 equiv.) in MeOH (60 mL) and H2O (15 mL) was added dropwise to a solution of
iodine (55 mg, 0.22 mmol, 3.5 equiv.) in MeOH (80 mL) over a period of 1 h. The
mixture was stirred for an additional 20 min, followed by the addition of pre-washed
Amberlite® IRA-400 Type II basic resin (3 g). The mixture was stirred gently until it
became colorless, filtered to remove the resin, washing well with MeOH, and filtrate
concentrated and lyophilized. The residue was subjected to reverse-phase HPLC to yield
octapeptide disulfide 13 as colorless solid (RT 6 min, 69 mg, 86%). [α]D25 +23.5 (c 0.41,
1
H2O). H NMR* (500 MHz, DMSO-d6) δ 1.17 (d, J = 7.1 Hz, 3H), 1.86 (s, 3H), 1.88-
1.93 (m, 1H), 1.99-2.07 (m, 1H), 2.36 (dd, J = 15.7, 7.2 Hz, 1H), 2.56-2.63 (m, 1H), 2.87
(dd, J = 13.6, 7.0 Hz, 1H), 2.98-3.04 (m, 1H), 3.13-3.17 (m, 1H), 3.46 (app. d, J = 11.2
Hz, 1H, partially obscured by H2O), 3.59-3.85 (m, 1H), 3.69-3.85 (m, 6H), 4.25 (app. t, J
= 7.0 Hz, 1H), 4.30-4.35 (m, 2H), 4.39-4.47 (m, 1H), 4.50 (app. q, J = 7.2 Hz, 1H), 4.67
(app. q, J = 6.9 Hz, 1H), 6.97 (s, 1H), 7.22 (s, 1H), 7.30 (s, 1H), 7.41 (s, 1H), 7.85 (t, J =
6.0 Hz, 1H), 7.97 (t, J = 6.1 Hz, 1H), 8.14 (d, J = 6.9 Hz, 1H), 8.16 (d, J = 8.2 Hz, 1H),
8.25 (d, J = 8.4 Hz, 1H), 8.30 (d, J = 7.8 Hz, 1H), 8.42-8.48 (m, 1H) 13C NMR* (125
MHz, DMSO-d6) δ 18.3, 22.8, 37.4, 37.5, 41.4, 42.2, 42.6, 43.2, 48.8, 49.7, 50.3, 50.6,
54.2, 56.1, 60.4, 67.3, 69.3, 169.3, 169.8, 169.9, 170.2, 170.3, 171.4, 172.0, 172.4, 172.6,

ACCEPTED MANUSCRIPT
27
172.9, 173.7. HRMS (ESI-TOF) m/z: [M + Na]+ Calcd for C26H40N10NaO11S2 755.2212;
Found 755.2207.
*peaks reported for major conformer only
4.3.3 Octapeptide thioether 14. Octapeptide disulfide 13 (40 mg, 0.05 mmol, 1.0
equiv.) was suspended in dry, degassed methanol (100 mL) under N2 and sonicated to
enhance solubility. To this suspension was added a solution of HMPT (13 ꢀL, 11.6 mg,
0.07 mmol, 1.3 equiv.) in dry methanol (~100 ꢀL) dropwise over 20 min. Additional
aliquots of HMPT (13 ꢀL in ~100 ꢀL dry methanol) were added 4-5 times per day, with
at least 4 h intervals between additions. After 3 d, the reaction mixture was monitored
periodically by MS to check for residual disulfide. After 7.5 d, the total amount of HMPT
added was 416 ꢀL (376 mg, 41.6 equiv.) and disulfide could no longer be detected by
MS. The mixture was concentrated and the residue triturated with Et2O (3 × 10 mL), the
Et2O decanted carefully to leave the crude product that was further purified by HPLC
using a gradient of 0% B for 3 min; 0ꢀ5% B over 7 min; 5ꢀ45% B over 25 min;
45ꢀ95% B over 3 min and 95% B for 5 min to give 14 as colorless solid (RT 5.7-6.7
min, 25.5 mg, 67 %). [α]D25 +16.8 (c 0.91, H2O). 1H NMR* (500 MHz, DMSO-d6) δ 1.19
(d, J = 6.9 Hz, 3H), 1.87 (s, 3H), 2.03-2.08 (m, 2H), 2.36-2.43 (m, 1H), 2.58-2.62 (m,
1H), 2.78-2.89 (m, 3H), 3.43-3.51 (m, 1H, obscured by H2O), 3.55-3.60 (m, 1H), 3.62-
3.85 (m, 6H), 4.06-4.18 (m, 1H), 4.24 (q, J = 7.1 Hz, 1H), 4.32-4.37 (m, 2H), 4.48 (app.
q, J = 7.3 Hz, 1H), 4.65 (app. q, J = 7.1 Hz, 1H), 6.96 (s, 1H), 7.11-7.21 (m, 1H), 7.38 (s,
1H), 7.86 (d, J = 8.5 Hz, 1H), 7.93-8.02 (m, 1H), 8.16 (d, J = 8.7 Hz, 1H), 8.24 (d, J =
6.3 Hz, 1H), 8.28-8.36 (m, 1H), 8.47 (d, J = 7.3 Hz, 1H), 8.58-8.69 (m, 1H); 13C NMR*

ACCEPTED MANUSCRIPT
28
(125 MHz, DMSO-d6) δ 17.7, 22.3, 36.4, 37.1, 41.8, 42.2, 42.8, 48.4, 48.6, 49.3, 50.3,
50.6, 52.6, 55.7, 59.8, 68.9, 69.8, 169.0, 169.3, 169.4, 170.0, 170.9, 171.5, 172.0, 172.1,
172.2, 172.8. HRMS (ESI-TOF) m/z: [M + H]+ Calcd for C26H41N10O11S 701.2671;
Found 701.2667.
*peaks reported for major conformer only
4.4
Synthesis of D-Ring Hexapeptides
4.4.1 Fmoc-Asn(Trt)-Ala-OBn (17). EDC (312 mg, 2.7 mmol, 1.6 equiv.), HOBt (364
mg, 2.7 mmol, 1.6 equiv.) and iPr2NEt (379 µL, 1.9 mmol, 1.2 equiv.) were added
sequentially to a solution of Fmoc-Asn(Trt)-OH (15) (953 mg, 1.6 mmol, 1 equiv.) in
CH2Cl2 (50 mL) at rt under N2. After 5 min, TsOH.H-Ala-OBn (16) (615 mg, 1.7 mmol,
1.1 equiv.) was added and stirring continued for 14 h. The mixture was diluted with
CH2Cl2 (200 mL), washed with water (100 mL), 5% aq. NaHCO3 (100 mL), 1 M aq.
KHSO4 (100 mL) and brine (100 mL), filtered through MgSO4, and concentrated. The
residue was subjected to flash chromatography eluting with 15-30% EtOAc in hexanes to
1
yield the dipeptide 17 as a colorless solid (483 mg, 40%). Rf 0.41 (1:1 EtOAc-hex). H
NMR (500 MHz, CDCl3) δ 1.30 (d, J = 7.2 Hz, 3H), 2.60 (dd, J = 15.6, 5.9 Hz, 1H), 3.08
(d, J = 15.6 Hz, 1H), 4.19 (t, J = 6.8, 1H), 4.33-4.41 (m, 2H), 4.50 (t, J = 7.2, 1H), 4.53-
4.60 (m, 1H), 5.08 (d, J = 12.3 Hz, 1H), 5.18 (d, J = 12.3 Hz, 1H), 6.38 (d, J = 6.4 Hz,
1H), 6.80 (br s, 1H), 7.16-7.19 (m, 6H), 7.23-7.29 (m, 11H), 7.30-7.33 (m, 5H), 7.39 (t, J
= 7.2, 2H), 7.56 (d, J = 7.0 Hz, 2H), 7.75 (d, J = 7.4 Hz, 2H); 13C NMR (125 MHz,
CDCl3) δ 17.6, 38.5, 47.1, 48.6, 51.0, 67.1, 67.3, 70.9, 120.0, 125.2, 127.1, 127.2, 127.7,
128.1, 128.2, 128.4, 128.6, 128.7, 135.4, 141.3 (2C), 143.7, 143.8, 144.2, 156.2, 170.6