Irreversible sortase a-mediated ligation driven by diketopiperazine formation

Article abstract of DOI:10.1021/jo4024914

Sortase A (SrtA)-mediated ligation has emerged as an attractive tool in bioorganic chemistry attributing to the remarkable specificity of the ligation reaction and the physiological reaction conditions. However, the reversible nature of this reaction limits the efficiency of the ligation reaction and has become a significant constraint to its more widespread use. We report herein a novel set of SrtA substrates (LPETGG-isoacyl-Ser and LPETGG-isoacyl-Hse) that can be irreversibly ligated to N-terminal Gly-containing moieties via the deactivation of the SrtA-excised peptide fragment through diketopiperazine (DKP) formation. The convenience of the synthetic procedure and the stability of the substrates in the ligation buffer suggest that both LPETGGisoacyl- Ser and LPETGG-isoacyl-Hse are valuable alternatives to existing irreversible SrtA substrate sequences.

Full text of DOI:10.1021/jo4024914

Featured Article
pubs.acs.org/joc
Irreversible Sortase A‑Mediated Ligation Driven by Diketopiperazine
Formation
Fa Liu,* Ethan Y. Luo, David B. Flora, and Adam R. Mezo
Lilly Research Laboratories, Indianapolis, Indiana 46285, United States
S
* Supporting Information
ABSTRACT: Sortase A (SrtA)-mediated ligation has
emerged as an attractive tool in bioorganic chemistry
attributing to the remarkable specificity of the ligation reaction
and the physiological reaction conditions. However, the
reversible nature of this reaction limits the efficiency of the
ligation reaction and has become a significant constraint to its
more widespread use. We report herein a novel set of SrtA
substrates (LPETGG-isoacyl-Ser and LPETGG-isoacyl-Hse)
that can be irreversibly ligated to N-terminal Gly-containing
moieties via the deactivation of the SrtA-excised peptide
fragment through diketopiperazine (DKP) formation. The
convenience of the synthetic procedure and the stability of the substrates in the ligation buffer suggest that both LPETGG-
isoacyl-Ser and LPETGG-isoacyl-Hse are valuable alternatives to existing irreversible SrtA substrate sequences.
reaction, thereby driving the forward ligation reaction toward
completion.12 As another example, both the Ploegh and
Turnbull groups replaced the Thr-Gly amide bond in the
LPETG recognition sequence of the substrate with an ester
bond. Upon ligation, the excised fragment contained a terminal
hydroxyl group, instead of a terminal amine, which dramatically
reduced its reactivity and prevented the reverse ligation.13,14
Among these methods, the depsipeptide approach from
Turnbull and co-workers14 appears particularly attractive;
however, the preparation of this depsipeptide requires a
nonstandard dipeptide which is not convenient to obtain
using standard solid-phase peptide synthesis. In addition, the
stability of this depsipeptide under standard SrtA ligation
conditions is limited because it is susceptible to hydrolysis via
its ester bond. This may become problematic should the
ligation proceed slowly. We report herein novel alternative
modifications of the SrtA recognition sequence designed to
generate an irreversible ligation reaction by deactivating the N-
terminal glycine of the excised peptide through diketopiper-
azine (DKP) formation (Scheme 1B).
INTRODUCTION
■
Staphylococcal sortase A (SrtA) is a transpeptidase that exists in
almost all Gram-positive bacteria and attaches surface proteins
to the bacterial cell wall through newly formed peptide amide
bonds.1 The SrtA mechanism is as follows: SrtA first cleaves the
Thr-Gly amide bond within the LPXTG (X = D, E, A, N, Q, or
K) recognition motif (substrate A) to form a Thr thioester
intermediate that can then react with a N-terminal Gly-
containing peptide (nucleophile B) to reform a Thr-Gly
peptide bond linking the two substrates (A−B, Scheme 1A).2,3
The remarkable specificity of the SrtA-mediated ligation and
the physiological reaction conditions have resulted in numerous
successful applications of this reaction, including (1) the
preparation of fusion constructs between small molecules,
peptides, proteins and antibodies,4−8 (2) the generation of
circular proteins,9,10 and (3) its application in protein
purification.11 When using synthetically generated peptides or
small molecules, this method can effectively circumvent the
challenge of introducing non-native amino acids or other
nonstandard features such as C-terminal amides into
recombinantly derived proteins. It can also efficiently combine
two recombinant proteins which otherwise could be difficult to
produce as a single fusion protein due to low expression yields
or improper folding.6,8
RESULTS AND DISCUSSION
■
Recombinant SrtA, containing a N-terminal deletion of 59
amino acids and a C-terminal hexa-His tag, was produced as
described15,16 and used throughout the study. The peptide
GGGGAEW-NH2 (1, Table 1) was used as the nucleophile in
all ligation experiments described below. In the search for
suitable R3 and R4 groups to enable the formation of DKP
(Scheme 1B), the first sequence evaluated was FLQLYGL-
However, despite the attractive features of SrtA-mediated
ligation, its broader application has been largely hampered by
the reversibility of the process which leads to less than
quantitative conversion yields. A few approaches have been
reported to improve conversion yields. For example, Nagamune
and co-workers incorporated a rigid β-hairpin secondary
structure flanking the LPETG motif in the ligated product.
The ligated product is a poor substrate for the reverse SrtA
Received: November 13, 2013
Published: December 30, 2013
© 2013 American Chemical Society
487
dx.doi.org/10.1021/jo4024914 | J. Org. Chem. 2014, 79, 487−492

The Journal of Organic Chemistry
Featured Article
Scheme 1. (A) Reversible SrtA-Mediated Ligation. (B) Irreversible SrtA Ligation Mediated by DKP Formation (R1, R2: Any
Moiety of Interest; R3: Any Suitable Group; R4: Any Leaving Group except Glycine)
Table 1. Sequence and Characterization of the Synthetic Model Peptides
MW (M + H)+
a
,
b
c
d
peptide no.
sequence
GGGGAEW-NH2
calcd
obsd
amt (mg)
yield (%)
purity (%)
e
1
632.3
1501.8
1458.8
1436.7
1350.7
1476.8
1464.8
1422.7
1237.6
1452.7
1604.8
632.0
1502.0
1458.1
1436.0
1350.9
1476.9
1464.9
1422.9
1238.0
1452.9
1604.9
50
79
73
72
66
85
7
96
80
90
98
98
98
98
98
95
98
93
e
2
FLQLYGLPETGSarHG-NH2
110
e
3
FLQLYGLPETGabaHG-NH2
105
4
FLQLYGLPETGG-isoacyl-Hse(Ac)-NH2
FLQLYGLPETGGG-NH2
95
115
10
9
10
11
12
13
14
15
FLQLYGLPETGP-isoacyl-Hse(Ac)-NH2
FLQLYGLPETGAib-isoacyl-Hse(Ac)-NH2
FLQLYGLPETGG-isoacyl-S(Ac)-NH2
FLQLYGLPET-OCH2CO-NH2
10
7
50
35
89
14
8
e
110
FLQLYGLPETGS-isoacyl-S(Ac)-NH2
20
12
FLQLYGLPETGY-isoacyl-Hse(Bz)-NH2
a
b
The nonunderlined residues were selected randomly. Unusual amino acid abbreviations: Sar, N-methylglycine; Gaba, γ-aminobutyric acid; Hse,
c
d
homoserine; Aib: 2-aminoisobutyric acid. The amount of product obtained from 0.10 mmol scale synthesis. Based on the RP-HPLC at 210 nm.
e
The weight of the crude peptide, which was used without further purification.
Scheme 2. Proposed Pathways for Deactivating the N-Terminus of Excised Peptide during SrtA-Mediated Ligation
PET0G1Sar2H3G-NH2 (2, the positions of selected residues are
labeled as the superscripted numbers, Table 1). To achieve
accelerated DKP formation, two hypotheses were pursued: (1)
Sar2 was incorporated into substrate peptide 2 to potentially
increase the cis-amide bond population, and (2) His3 was
incorporated to potentially activate the Sar-His amide bond
(Scheme 2A).17−20 However, it was found that the ligation did
not proceed, as no conversion was detected by HPLC (high-
performance liquid chromatography) and MS (mass spectrom-
etry) after 24 h incubation at 37 °C with 5% (mol/mol) SrtA. It
is likely that SrtA did not recognize substrate peptide 2,
possibly a result of the N-methylamino acid at the 2 position.
Under a similar hypothesis, the N-terminal Gaba (γ-amino-
butyric acid) moiety of the excised fragment from peptide
FLQLYGLPET0Gaba1,2H3G-NH2 (3, Table 1) could undergo
five-membered ring γ-lactam formation to form 2-pyrroli-
done,19 thereby exposing the SrtA-unreactive histidine at the N-
terminus (Scheme 2B). However, similar to peptide 2, SrtA-
mediated ligation of peptides 3 and 1 did not provide any
ligated product after 24 h at 37 °C with 5% SrtA, reinforcing
the need for natural amino acids at the 1 and 2 position in the
consensus recognition sequence.
488
dx.doi.org/10.1021/jo4024914 | J. Org. Chem. 2014, 79, 487−492

The Journal of Organic Chemistry
Featured Article
Scheme 3. Synthesis of Isoacyl Peptides 4 and 12
Figure 1. (A) Time-course of SrtA-mediated ligation of peptides 4, 9, 12, or 13 with 1 (1.20 equiv) and 2% SrtA (mol/mol). (B) Efficiency of
peptides 4, 12 (1.25−3.0 equiv), and 9 (1.5 equiv) for converting nucleophile 1 to ligation products. (C) Stability of substrate peptides 4, 12, and 13.
All experiments were conducted in 50 mM HEPES, 150 mM NaCl and 10 mM CaCl2 at 37 °C; pH 8.0 for panels A and B, pH 7.5 and 8.0 for panel
C.
Learning from these two unsuccessful attempts with peptides
2 and 3, subsequent model substrate peptides utilized glycine at
both positions 1 and 2 to ensure efficient recognition by SrtA.
In order to promote DKP formation, an ester bond between
Gly2 and isoacyl-Hse3 (FLQLYGLPETG1G2-isoacyl-Hse3(Ac)-
NH2, peptide 4) was employed in place of the regular amide
bond (Table 1 and Scheme 2C). The synthesis of peptide 4
began with loading Fmoc-Hse(Trt)-OH onto Rink amide resin,
followed by treatment with piperidine and acetic anhydride to
provide resin 6 (Scheme 3). The trityl group of resin 6 was
removed with 1% TFA (trifluoroacetic acid) and 5% TIS
(triisopropylsilane) in DCM (dichloromethane), and the
Fmoc-Gly-OH was coupled to resin 7 with HBTU (O-
(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluoro-
phosphate)/HOBt (hydroxybenzotriazole)/DIEA (N,N-diiso-
propylethylamine) and a catalytic amount of DMAP (4-
(dimethylamino)pyridine). The remaining residues were then
assembled on resin 8 by means of an automated peptide
synthesizer using standard Fmoc/t-Bu protocols. In order to
effectively compare the ligation efficiency of peptide 4, a
comparator peptide was also synthesized which contained all
natural amino acids (peptide 9, Table 1). The ligation of
peptides 4 or 9 was performed with 1.2 equiv of nucleophile 1
and 2% SrtA (mol/mol, based on substrate peptides 4 or 9) at
37 °C, and the ligation reaction was monitored by analytical
HPLC. It was found that peptide 4 was progressively consumed
and the conversion of 4 to its ligated product was complete by
10 h. In contrast, the ligation of peptides 9 and 1 reached
equilibrium at 2 h with only 58% conversion of peptide 9 to its
ligated product (Figure 1A; Figures S1A and S1B and Table S1,
Supporting Information).
With the encouraging data from peptide 4, we again
hypothesized that the rates of DKP formation could be
increased by inducing a greater proportion of the cis-
conformation in the C-terminal amide of Gly1 via the
replacement of Gly2 with proline or 2-aminoisobutyric acid
(Aib). However, the SrtA-mediated ligation of peptide 1 with
both substrate peptides 10 (FLQLYGLPET0G1P2-isoacyl-Hse-
(Ac)-NH2) and 11 (FLQLYGLPET0G1Aib2-isoacyl-Hse(Ac)-
NH2) proceeded much more slowly than the ligation of
peptides 4 and 1, possibly a result of poor recognition of
sequences LPETG1P2 and LPETG1Aib2 by SrtA as compared to
LPETG1G2. Our focus then shifted to the possibility of
increasing the leaving ability of the alcohol to also accelerate
DKP formation. Serine was tested instead of Hse at position 3
of peptide 4 since the hydroxyl of the serine side-chain
possesses a lower pKa than the hydroxyl of Hse, and therefore
serine was expected to be more reactive. The resulting peptide
12 (FLQLYGLPETG1G2-isoacyl-S3(Ac)-NH2) was evaluated
under the identical ligation conditions as for 4 and 9, and it was
found that full conversion of 12 was achieved at 4 h (vs 10 h for
4) (Figure 1A; Figure S1C and Table S1, Supporting
Information). To compare with the previously reported
depsipeptide approach,14 substrate peptide 13 (FLQLYGL-
PET-OCH2CO-NH2) was prepared and tested, and it was
found to be fully converted to ligation product after 2 h (Figure
489
dx.doi.org/10.1021/jo4024914 | J. Org. Chem. 2014, 79, 487−492

The Journal of Organic Chemistry
Featured Article
Scheme 4. Determination of the DKP Formation during the Ligation between Peptides 15 and 1
Figure 2. Analytical RP-HPLC traces of the ligation between peptides 15 and 1 (at 210 nm).
1A; Figure S1D and Table S1, Supporting Information). It
should be noted that although the ligations proceeded to
completion, hydrolysis of the substrate peptides 4, 12, and 13
was observed to a similar extent of approximately 15% for each
(Table S2, Supporting Information). To test if other residues at
position 2 could reduce the extent of hydrolysis, substrate
peptide 14 (FLQLYGLPETG1S2-isoacyl-S3(Ac)-NH2), with
serine at position 2, was synthesized and tested in the ligation
reaction. It was found that the extent of hydrolysis of 14 was
approximately 20% under the same ligation conditions as for
peptide 12 (Figure S3, Supporting Information). Considering
the low efficiency of coupling residues other than glycine to the
side-chain hydroxyl of serine, additional amino acids at position
2 were not pursued.
To examine the efficiency of the substrate peptides 4 and 12
on converting a fixed amount of nucleophile 1 into the SrtA
ligation product, a range of molar ratios of 4 and 12 to 1
(1.25−3.0) were evaluated. It was found that 1.5 equiv of
substrate peptide 4 converted 91% nucleophile 1 within 20 h,
while the same amount of peptide 12 achieved 96% conversion
of 1 within 6 h (Figure 1B; Figures S2A−F, Supporting
Information). As described earlier,14 SrtA (2 mol % of the
substrate peptides) maintains an equilibrium between the
product and the thioester intermediate making it difficult to
achieve 100% conversion of nucleophile 1. As a comparison, 1.5
equiv of the natural amino acid-containing substrate peptide 9
only converted 57% of peptide 1 under identical ligation
conditions (Figure 1B; Figure S2G, Supporting Information)
In addition to overcoming the reversibility of the reaction,
the stability of the substrate peptides in the ligation buffer is
also important for achieving high conversion yields. The t1/2
(half-life time) of peptides 4, 12, and 13 was measured in the
ligation buffer of 50 mM HEPES, 150 mM NaCl, and 10 mM
CaCl2 at pH 7.5 or 8.0 at 37 °C. It was determined that at pH
8.0 the t1/2 was ∼15 h for peptide 4, ∼5 h for peptide 12, and
∼3 h for peptide 13 (Figure 1C; Table S3, Supporting
Information). The trends at pH 7.5 were similar with t1/2’s
approximately 2-fold higher for each peptide. The longer t1/2’s
of peptides 4 and 12 could be beneficial in cases where the
ligation proceeds slowly (e.g., due to steric hindrance of the
substrates) and the reaction needs more time for completion.
To confirm that the SrtA-mediated ligation of substrate
peptides 4 or 12 was indeed driven by DKP formation as
originally hypothesized, peptide 15 (FLQLYGLPETGY-isoacyl-
Hse(Bz)-NH2) was designed with tyrosine and benzamide
incorporated into the SrtA-excised fragment to facilitate
monitoring of the byproduct by HPLC/UV and MS. Upon
analysis of the ligation mixture of peptides 15 and 1 using LC−
MS, a side product with the molecular weight of 220 Da was
identified which unambiguously confirmed the formation of
Gly-Tyr DKP 18 (Scheme 4 and Figure 2; Figure S4,
Supporting Information).
The practical utility of substrate peptides 4 and 12, in
comparison to peptide 13, was evaluated by performing a
ligation with insulin lispro 20. Conveniently, insulin lispro
contains a glycine at the N-terminus of its A-chain. Each of the
490
dx.doi.org/10.1021/jo4024914 | J. Org. Chem. 2014, 79, 487−492

The Journal of Organic Chemistry
Featured Article
Scheme 5. SrtA-Mediated Ligation between Insulin Lispro 20 and Substrate Peptides 4, 12, and 13
Figure 3. Analytical RP-HPLC traces of the SrtA-mediated ligation between insulin lispro 20 and peptide 4 (at 210 nm). 93% of insulin lispro 20 was
converted to 21 at 18 h. Asterisk indicates the hydrolyzed product of 4 as confirmed by mass spectrometry.
purchased from Rapp Polymere GMBH. Fmoc-chemistry based
amino acid cartridges were purchased from several vendors and used
directly. LC-MS system: Agilent 1100 series liquid chromatography
mass spectrometer, model G1956A/B. General analytical RP-HPLC
condition: All samples were analyzed using a Waters SymmetryShield
RP18 Column (cat. no. 186000179, 3.5 μm, 4.6 × 100 mm.): a linear
gradient from 6% aqueous ACN (0.1% trifluoroacetic acid) to 60%
aqueous ACN (0.1% trifluoroacetic acid) over 14 min at 60 °C and a
flow rate of 1.0 mL/minute.
General Procedure for Solid-Phase Synthesis of Peptides 1−
3, 9, and 13. Peptide synthesis was conducted using 0.10 mmol of
Rink Amide resin (Rapp Polymere GMBH) and a ABI433 peptide
synthesizer, using a standard Fmoc_HOBT_DCC_0.10 mmol
method (Table S4, Supporting Information). Cleavage was conducted
by treatment of this resin with 10 mL of a TFA solution containing
2.5% TIS, 2.5% H2O at rt, with gentle shaking for 1.5 h. The resin was
filtered and washed with TFA (2 mL × 2). The combined filtrate was
precipitated with cold ether (80 mL). The precipitate was collected by
centrifugation, then washed with cold ether (80 mL × 2), and purified
by preparative RP-HPLC using a Waters SymmetryPrep C18 Column
(cat. no. WAT066245, 7 μm, 19 × 300 mm). Purification was achieved
using a linear gradient from 5% aqueous ACN (0.1% trifluoroacetic
acid) to 50% aqueous ACN (0.1% trifluoroacetic acid) over 80 min at
a flow rate of 15 mL/minute. Selected fractions were pooled and
lyophilized to provide the target peptides as lyophilized powders.
Identity was confirmed using mass spectrometry (Table 1; Figure S6,
Supporting Information).
General Procedure for Solid-Phase Synthesis of Peptides 4
and 10−12, 14, and 15. Rapp Polymere Rink amide resin (Rapp
Ploymere, cat no. H 400 23) (0.10 mmol) was treated with 20%
piperidine in DMF (10 min × 2) and washed with DMF (3 mL × 6).
In a separate vial, Fmoc-Hse(Trt)-OH (1.0 mmol) or Fmoc-Ser(Trt)-
OH (1.0 mmol), HBTU (1.0 mmol), HOBt (1.0 mmol), and DIEA
(1.0 mmol) were mixed in DMF (3 mL) for 5 min before being
transferred to the above resin. The resulting slurry was gently mixed
for 2 h at rt, washed with DMF (3 mL × 3), treated with 20%
piperidine in DMF (10 min × 2), and washed with DMF (3 mL × 3)
before being mixed with acetic anhydride (2 mmol) and DIEA (2
mmol) in DCM (5 mL) at rt for 1 h. The resulting resin was washed
substrate peptides (1.5 equiv each; 4, 12, or 13) was incubated
with lispro 20 in the presence of 2% SrtA at 37 °C (Scheme 5).
After 18 h, the conversion of lispro 20 to the ligation product
21 was nearly identical for all three substrate peptides: 92% for
peptide 4, 93% for peptide 12, and 92% for peptide 13 (Figures
3; Figures S5A−C, Supporting Information).
CONCLUSION
■
In conclusion, we have developed a novel set of SrtA substrates
that, through DKP formation of the N-terminal glycine of the
SrtA-excised peptide fragment, can be irreversibly ligated to the
N-terminus of Gly-containing nucleophiles. Peptides compris-
ing both LPETGG-isoacyl-Ser (4) and LPETGG-isoacyl-Hse
(12) were demonstrated to be effective, irreversible SrtA
substrates, and each peptide can be readily prepared through
standard solid-phase procedures using commercially available
amino acid building blocks. Additionally, the convenience of
the synthetic procedures and the higher stability (vs substrate
peptide 13) in the ligation buffer suggest that both LPETGG-
isoacyl-Ser and LPETGG-isoacyl-Hse are valuable alternatives
to existing irreversible SrtA substrate sequences.
EXPERIMENTAL SECTION
■
General Information. All solvents [N,N′-dimethylforamide
(DMF), methanol (MeOH), dichloromethane (DCM), acetonitrile
(ACN), diethyl ether (Et2O)] and reagents [hydroxybenzotriazole
(HOBT), N,N′-diisopropylcarbodiimide (DIC), O-(benzotriazol-1-
yl)-N,N,N′,N′-tetramethyluronium hexafluorophosphate (HBTU),
piperidine, trifluoroacetic acid (TFA), triisopropylsilane (TIS),
HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), tris-
(hydroxymethyl)aminomethane (Tris), collidine, acetic anhydride,
triphosgene] were purchased and used directly. The dipeptide Fmoc-
Thr(t-Bu)-OCH2CO2H was purchased from the Chinese Peptide
Company. Insulin lispro was obtained from Eli Lilly (Indianapolis,
IN). Water (H2O) was obtained from a Milli-Q water purification
system (Millipore). Polystyrene Rink amide (RAM) resin was
491
dx.doi.org/10.1021/jo4024914 | J. Org. Chem. 2014, 79, 487−492

The Journal of Organic Chemistry
Featured Article
with DCM (3 mL × 6), treated with 5 mL of 1% TFA and 5% TIS in
DCM for 2 min, drained, and retreated 5 more times. Formation of the
ester bond in peptides 4, 10, 12, 14, and 15 was achieved by mixing
Fmoc-Gly-OH (1.0 mmol), HBTU (1.0 mmol), HOBt (1.0 mmol),
DIEA (1.0 mmol), and DMAP (0.6 mg, 5% of the resin substitution)
in 3 mL of DMF and then adding this mixture to the resin for 2 h at rt.
This coupling was repeated once. Formation of the ester bond in
peptide 11 was achieved by in situ acyl chloride formation using
triphosgene:21 the resin was pretreated with DIEA/THF (0.60 mL, 1:1
by volume) for 10 min. In a separate vial, Fmoc-Aib−OH (1.0 mmol)
and BTC (triphosgene, 100 mg, 0.33 mmol) were dissolved in
anhydrous THF (5 mL). To this mixture was added collidine (0.40
mL) and the resulting slurry was mixed for 1 min, and transferred to
the above DIEA/THF pretreated resin followed by the addition of
DMAP (1.2 mg). The resin was gently agitated at rt for 4 h. For both
coupling methods, the resulting resin-bound isoacyl peptides were
transferred to the ABI433 peptide synthesizer to assemble the
remainder of the peptide sequence using the standard Fmo-
c_HOBT_DCC_0.10 mmol method (Table S4, Supporting Informa-
tion). Resin cleavage and peptide purification was conducted as
described above for the synthesis of peptides 1−3, 9, and 13.
(13) Antos, J. M.; Chew, G. L.; Guimaraes, C. P.; Yoder, N. C.;
Grotenbreg, G. M.; Popp, M. W.; Ploegh, H. L. J. Am. Chem. Soc. 2009,
131, 10800.
(14) Williamson, D. J.; Fascione, M. A.; Webb, M. E.; Turnbull, W. B.
Angew. Chem., Int. Ed. 2012, 51, 9377.
(15) Ilangovan, U.; Ton-That, H.; Iwahara, J.; Schneewind, O.;
Clubb, R. T. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 6056.
(16) Ton-That, H.; Liu, G.; Mazmanian, S. K.; Faull, K. F.;
Schneewind, O. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 12424.
(17) Suaifan, G. A. R. Y.; Mahon, M. F.; Arafat, T.; Threadgill, M. D.
Tetrahedron 2006, 62, 11245.
(18) Santos, C.; Mateus, M. L.; dos Santos, A. P.; Moreira, R.; de
Oliveira, E.; Gomes, P. Bioorg. Med. Chem. Lett. 2005, 15, 1595.
(19) Gomes, P.; Vale, N.; Moreira, R. Molecules 2007, 12, 2484.
(20) Goolcharran, C.; Borchardt, R. T. J. Pharm. Sci. 1998, 87, 283.
(21) Thern, B.; Rudolph, J.; Jung, G. Tetrahedron Lett. 2002, 43,
5013.
General Procedure for SrtA Ligation. The substrate peptides
(2−4 and 9−15) and the nucleophile peptides (1 or insulin lispro 20)
were mixed in a HEPES ligation buffer (50 mM HEPES, 5 mM CaCl2,
150 mM NaCl, pH 8.0) at the desired concentration and molar ratio
with the presence of SrtA (2% mol/mol based on the substrate
peptides) at 37 °C or room temperature (21 °C). The reaction was
monitored by analytical HPLC and/or LC−MS.
ASSOCIATED CONTENT
■
S
* Supporting Information
RP-HPLC traces and mass spectrometry data for the SrtA
ligations and the synthesized peptides. This material is available
free of charge via the Internet at http://pubs.acs.org
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: liufx@lilly.com.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Yuewei Qian and Ms. Xiaohua He (Eli Lilly) for
the expression and purification of SrtA.
REFERENCES
■
(1) Hendrickx, A. P.; Budzik, J. M.; Oh, S. Y.; Schneewind, O. Nat.
Rev. Microbiol. 2011, 9, 166.
(2) Ton-That, H.; Mazmanian, S. K.; Faull, K. F.; Schneewind, O. J.
Biol. Chem. 2000, 275, 9876.
(3) Aulabaugh, A.; Ding, W.; Kapoor, B.; Tabei, K.; Alksne, L.;
Dushin, R.; Zatz, T.; Ellestad, G.; Huang, X. Anal. Biochem. 2007, 360,
14.
(4) Mao, H.; Hart, S. A.; Schink, A.; Pollok, B. A. J. Am. Chem. Soc.
2004, 126, 2670.
(5) Proft, T. Biotechnol. Lett. 2010, 32, 1.
(6) Tsukiji, S.; Nagamune, T. ChemBioChem 2009, 10, 787.
(7) Wu, Z.; Guo, Z. J. Carbohydr. Chem. 2012, 31, 48.
(8) Popp, M. W.; Ploegh, H. L. Angew. Chem., Int. Ed. 2011, 50, 5024.
(9) Antos, J. M.; Popp, M. W.; Ernst, R.; Chew, G. L.; Spooner, E.;
Ploegh, H. L. J. Biol. Chem. 2009, 284, 16028.
(10) Parthasarathy, R.; Subramanian, S.; Boder, E. T. Bioconjugate
Chem. 2007, 18, 469.
(11) Mao, H. Protein Expr. Purif. 2004, 37, 253.
(12) Yamamura, Y.; Hirakawa, H.; Yamaguchi, S.; Nagamune, T.
Chem. Commun. (Cambridge) 2011, 47, 4742.
492
dx.doi.org/10.1021/jo4024914 | J. Org. Chem. 2014, 79, 487−492