4738
P.-S. Pan et al. / Bioorg. Med. Chem. 14 (2006) 4731–4739
4.4. General acid deprotection
with ammonium chloride. After back extraction of
aqueous layers with methylene chloride, organic layers
were combined, dried over sodium sulfate, filtered, and
concentrated. All macrocycles were purified using re-
verse phase HPLC, and a gradient of acetonitrile and
DI water with 0.1% TFA.
Acids were deprotected using 4 equiv of lithium hydrox-
ide (or enough was added until pH ꢀ11) in methanol
(0.1 M). The peptide was placed in a flask, along with
lithium hydroxide and methanol, and stirred overnight.
Within 12 h the acid was usually deprotected. Work-up
of reactions involved the acidification of reaction solu-
tion using HCl to pH 1. The aqueous solution was
extracted three times with methylene chloride, and the
combined organic layer was dried, filtered, and concen-
trated in vacuo.
4.6. RuvC HJ binding and cleavage assays
RuvC protein was purified as described.21 A synthetic
50 bp HJ substrate containing an 11 bp mobile core
(J11) was labeled with 32P using T4 polynucleotide ki-
nase.22 One of the HJ constituent oligonucleotides was
used as a ssDNA substrate and annealed to its comple-
mentary sequence to give the dsDNA substrate. Binding
assays were performed in 50 mM Tris–HCl, pH 8.0,
5 mM EDTA, 1 mM dithiothreitol, 5% glycerol, and
100 lg/ml BSA. Samples were incubated on ice for
15 min before separation on 4% PAGE in 6.7 mM
Tris–HCl, pH 8.0, 3.3 mM sodium acetate, and 2 mM
EDTA. HJ cleavage was assayed at 37 °C for 30 min
in 50 mM Tris–HCl, pH 8.0, 1 mM dithiothreitol,
100 lg/ml BSA, and 10 mM MgCl2. Reactions (20 ll)
were terminated by the addition of 5 ll of 100 mM
Tris–HCl, pH 8.0, 2.5% SDS, 100 mM EDTA, and
10 mg/ml proteinase K, and incubated for a further
10 min at 37 °C. Following addition of 5 ll of loading
buffer (0.25% w/v bromophenol blue, 0.25% w/v xylene
cyanol, and 15% v/v Ficoll type 400), 15 ll was electro-
phoresed on 10% polyacrylamide gels in 90 mM Tris–
borate, 2 mM EDTA. Gels were dried onto filter paper
and analyzed by autoradiography.
4.5. Macrocyclization procedure (in situ)
All hexa- and octa-peptides were deprotected using HCl
in methanol and the presence of the free amine was ver-
ified using LCMS. The reaction was then neutralized
with lithium hydroxide. Upon neutralization, LiOH
was added (ꢀ4 equiv) to bring the pH up to ꢀ11. The
acid deprotection was verified via LCMS. Upon acid
deprotection the reaction was concentrated in vacuo
and the crude, dry, double deprotected peptide (free acid
and free amine) was dissolved in a minimum of dry ace-
tonitrile. Three coupling agents were initially used:
DEPBT, HATU, and TBTU (ꢀ0.5 to 0.75 equiv each).
These coupling agents were dissolved in a calculated vol-
ume of dry 50% acetonitrile and 50% methylene chloride
that would give a 0.01 M solution when including the
volume used for the deprotected peptide. The coupling
agents were then added to the deprotected peptide solu-
tion. Three to five equivalents of DIPEA were then add-
ed to the reaction to ensure the pH was kept at or
greater than 8. If the solution was not clear, DMF or
methylene chloride was added but not more than 20%
of the volume used for the overall reaction. Note: in
some cases methylene chloride addition improved the
solution clarity more than DMF, this depended on the
number of methyl groups on the compound (i.e., the
hydrophobicity). With at least one methyl group it was
found that methylene chloride was a better solvent than
DMF for clarity; with no methyl groups, DMF was the
better solvent. It is important to recognize that the cou-
pling agents are typically not very soluble in acetonitrile,
which is why an additional solvent is often used.
4.7. Bacterial growth inhibition assays
Escherichia coli K12 strains AB2463 (recA13) and
GS1481 (ruvC64::kan) are derivatives of the wild-type,
AB1157. S. epidermidis ATCC14990 was used as a
wild-type strain. Bacteria were cultured in LB broth at
37 °C and growth monitored at A650nm. A single 60 ml
culture was grown and divided into aliquots at an
A650nm of 0.18. Different compounds were added at a
concentration of 0.025 lM and growth monitored every
15 min. Appropriate dilutions of cultures at an A650nm of
0.6 were spotted on LB agar plates and colonies counted
to determine any loss in viability. Dilutions of peptides
were spotted onto plates carrying a lawn of bacteria in
qualitative growth inhibition experiments.
After 24 h, TLC and LCMS (where the LCMS sample
was worked up prior to injection) were taken, if no
clear distinct product spot was visible, then typically
PyAOP was added (0.5 equiv), and sometimes, depend-
ing on reaction clarity, 0.5 equiv of HATU were also
added. The comparison for Rf value in the product
spot on TLC was the protected linear hexa- or octa-
peptide. The reaction was allowed to run another
24 h, and checked again by TLC and LCMS. If the
reaction still failed to show a clear product spot, then
0.25 equiv of DEPBT were added and the reaction con-
tinued for 24–48 h. At this point we found the reaction
always demonstrated a product spot, although it was
sometimes difficult to determine if it was complete
(monitoring the starting material deprotected hexa- or
octapeptide via LCMS was the easiest method). Upon
completion, the reaction was worked up by washing
Acknowledgments
We thank John Rafferty (University of Sheffield, UK)
for modeling RuvC bound to the Holliday structure.
We are grateful to the following: Pfizer (La Jolla) for
equipment and financial donations as well as their fel-
lowships to IM (2003–2005) and CLC (SURF, Summer
2004); the SDSU McNair program for support to CLC
(Summer 2003 and 2004) and IM (Summer 2004); the
Howell Foundation for support for CLC (Spring
2004); and the MIRT program for travel support for
IM (2003–2004) and CLC (2004). We also thank San
Diego State University for additional funding. S.R.M.