Incorporation of chlorinated analogues of aliphatic amino acids during cell-free protein synthesis

Article abstract of DOI:10.1039/c0cc02879g

3-Chloro-Abu and 4-chloro-Nva are biosynthetically incorporated into E. coli peptidyl-Pro cis-trans isomerase B, as substitutes for Val and Leu, respectively. The extent of incorporation is up to ～90%, and substituted protein is catalytically active. By contrast, 4-chloro-Val is not an effective replacement for Ile.

Full text of DOI:10.1039/c0cc02879g

View Article Online / Journal Homepage / Table of Contents for this issue
This article is part of the
Enzymes & Proteins web
themed issue
This issue showcases high quality research in the field of enzymes and
proteins.
Please visit the website to access the other articles in this issue:-
http://www.rsc.org/chemcomm/enzymesandproteins

COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Incorporation of chlorinated analogues of aliphatic amino acids during
cell-free protein synthesiswz
Dannon J. Stigers,a Zachary I. Watts,a James E. Hennessy,a Hye-Kyung Kim,a
Romeo Martini,a Matthew C. Taylor,b Kiyoshi Ozawa,c Jeffrey W. Keillor,ya
Nicholas E. Dixonc and Christopher J. Easton*a
Received 29th July 2010, Accepted 30th October 2010
DOI: 10.1039/c0cc02879g
3-Chloro-Abu and 4-chloro-Nva are biosynthetically incorporated
into E. coli peptidyl-Pro cis-trans isomerase B, as substitutes for
Val and Leu, respectively. The extent of incorporation is up
to B90%, and substituted protein is catalytically active. By
contrast, 4-chloro-Val is not an effective replacement for Ile.
acid side chains with functionalised derivatives. It was known
already that the chloride 1a is incorporated into proteins14 and
acts as an antagonist of (S)-Val (4) to interfere with haemo-
globin biosynthesis,15 but the efficiency and extent of its
incorporation had not been investigated. The
Vmax/Km
value for its adenylation by Val-tRNA synthetase (ValRS)
of E. coli ML30 is 30% of that of (S)-Val (4),16 but this is
only one step towards protein synthesis. (2S,3R)-Thr17 and
(S)-Abu are also adenylated by the ValRS but they do not
even survive transfer to the related Val tRNA in the following
step.18 Even less information was available about the
chloride 1b. Its Vmax/Km value for adenylation by the ValRS
is 1.5% of that of (S)-Val (4)16 and it is 20% as active as the
diastereomer 1a as a Val antagonist against haemoglobin
biosynthesis.15 None of the chlorides 2a,b–3a,b had
been studied in this context, although the leucine analogue
2b had been found to have antibacterial activity that is
reversed by (S)-Leu (5).19
While protein biosynthesis typically employs only 20 amino
acids, many others are also incorporated by native bio-
synthetic machinery.1–3 Generally, each of the others has a
structure closely related to one of the 20 and therefore it
competes with the normal substrate to be loaded onto the
corresponding tRNA by the cognate aminoacyl-tRNA
synthetase. Cell-free protein production systems provide the
opportunity to bias this competition.1,4 These systems employ
cell extracts as the source of the ribosomes, aminoacyl-tRNA
synthetases and translation factors, combined with a medium
containing the other ingredients required for protein synthesis.
Excluding one of the normal amino acid substrates from the
medium, and adding a structurally analogous species instead,
increases the likelihood of incorporation of the analogue.
Auxotrophic cell strains which require exogenous supply of
one or more of the normal amino acids can also enable
incorporation of analogues.2 These approaches have been used
for the synthesis of proteins containing high levels of fluori-
nated,5 dehydro6 and isotopically-labelled7,8 amino acids, as
well as 3,4-dihydroxy-Phe9 and seleno-Met10 in place of Tyr
and Met, respectively.
Our interest in chlorinated aliphatic amino acids11 led us to
consider whether they would be incorporated. We chose to
investigate the chlorides 1a,b–3a,b as potential substitutes for
(S)-Val (4), (S)-Leu (5) and (2S,3S)-Ile (6), respectively, based
on the similar van der Waals radii of chloro (1.75 A)12 and
methyl (2.0 A)13 groups. Chlorine substitution is conceptually
attractive because it involves replacing non-reactive amino
We now report that while cell-free protein synthesis of
E. coli peptidyl-Pro cis-trans isomerase B (PpiB) shows no
indication of incorporation of either of the Ile analogues 3a,b,
the other chlorides 1a,b–2a,b are processed. When (S)-Leu (5)
is replaced with the chlorides 2a,b in the protein production
system, PpiB is still produced as a full length, soluble protein.
Similarly, each of the chlorides 1a,b is an effective substitute
for (S)-Val (4), producing soluble PpiB, at concentrations as
high as those formed from (S)-Val (4). Depending on the
experimental conditions, the average levels of incorporation
are well above 90% and around 50% for the chlorides 1a,b,
respectively, and PpiB is formed that has up to 30% of the
protein in which every one of the 16 residues of (S)-Val (4) is
replaced by the chloride 1a. Furthermore, the catalytic activity
of that protein sample is very similar to that of native PpiB. In
a competitive experiment the chloride 1a is incorporated
a ARC Centre of Excellence for Free Radical Chemistry and
Biotechnology, Research School of Chemistry,
Australian National University, Canberra ACT 0200, Australia.
E-mail: easton@rsc.anu.edu.au
b CSIRO Entomology, Canberra ACT 2601, Australia
c School of Chemistry, University of Wollongong,
Wollongong NSW 2522, Australia
w This article is part of the ‘Enzymes and Proteins’ web-theme issue for
ChemComm.
z Electronic supplementary information (ESI) available: Synthetic
procedures; cell-free protein synthesis; hydrolysis of chlorides 1a,b,
2a,b and 3a,b; ESI mass spectrometry of purified proteins; activity
assay for PpiB; IleRS activity assay. See DOI: 10.1039/c0cc02879g
y Current address: Departement de Chimie, Universite de Montreal,
´ ´ ´
Montreal Quebec H3C 3J7, Canada.
´
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 1839–1841 1839

of PpiB, since they were not removed by ion affinity chroma-
tography. As discussed in more detail below, high levels of
substitution are observed by mass spectrometry with the Val
analogues 1a,b. These data are also consistent with the gel,
which shows that addition of the chlorides 1a,b–2a,b in the
absence of (S)-Val (4) and (S)-Leu (5), respectively, increases
the level of synthesis of soluble PpiB.
The chlorides 2a,b hydrolyse to the corresponding alcohols/
lactones, with half-lives of around 2.0 and 2.6 h at 37 1C under
the conditions required for protein synthesis, whereas the
chlorides 1a,b and 3a,b are stable, showing less than 5%
decomposition after 6 h. It seems therefore that the misincor-
poration of the Val and Leu substitutes 1a,b and 2a,b but not
the Ile analogues 3a,b is not related to their stability, but is a
true reflection of the relative extents to which they are
activated and incorporated. The chlorides 1a,b and 2a,b are
incorporated into PpiB so they must be suitable substitutes for
(S)-Val (4) and (S)-Leu (5) at each and every step of protein
synthesis. Incubation of the chlorides 3a,b (2 mM) with IleRS
(1.5 mM) resulted in the production of around 100%
more AMP (210 mM) than when the same concentration of
(2S,3S)-Ile (6) was used, indicating that at least one of the
chlorides 3a,b is efficiently adenylated by the synthetase. The
lack of incorporation of either of the chlorides 3a,b into PpiB
must therefore be due to selectivity post adenylation.21
Apparently, the editing process distinguishes between the
ClCH2 substituent of the primary chlorides 3a,b and the
MeCH2 group of (2S,3S)-Ile (6), but not between the ClCHMe
group of the secondary chlorides 1a,b and 2a,b and the
MeCHMe group of (S)-Val (4) and (S)-Leu (5).
Fig. 1 SDS-PAGE gel of synthesised His6-PpiB with: (a) no DNA;
(b) DNA; (c) no (S)-Val (4); (d) no (S)-Val (4), but with the chlorides
1a,b; (e) no (S)-Leu (5); (f) no (S)-Leu (5), but with the chlorides 2a,b;
(g) no (2S,3S)-Ile (6); and (h) no (2S,3S)-Ile (6), but with the chlorides
3a,b.
into PpiB with an overall efficiency of B8% of that of
(S)-Val (4).
Cell-free synthesis of PpiB with an N-terminal His6 tag was
carried out with an E. coli [BL21(DE3)star] S30 extract
essentially as described previously.7,9,20 The native protein
was prepared using each of the 20 normal amino acids.
Incorporation of other amino acids was investigated through
their substitution for one of the 20. As reported previously,11
B1 : 1 mixtures of the diastereomeric chlorides 1a,b–3a,b were
prepared by chlorination of (S)-Abu, (S)-Nva and (S)-Val (4),
respectively. Initial examination for incorporation into PpiB
involved the use of the crude diastereomeric mixtures, which
also contained the corresponding starting materials and other
regioisomeric chlorination products. In this form the chlorides
1a,b–3a,b were examined at a concentration of 2 mM for each
diastereomer, as substitutes for 1 mM (S)-Val (4), (S)-Leu (5)
and (2S,3S)-Ile (6), respectively.
The chlorides 1a,b–2a,b are two diastereomeric pairs. It was
impractical to separate the Leu analogues 2a,b, due to their
instability, but the diastereomeric Val analogues 1a,b were
prepared and studied separately. Some key mass spectral data
for native PpiB and the protein that formed using each of the
chlorides 1a,b at concentrations of 2, 0.2 and 0.02 mM, instead
of adding 1 mM (S)-Val (4), are shown in Fig. 2, S2 and S3.
The corresponding mass spectrum of PpiB prepared in a
competitive experiment using the chloride 1a (2.0 mM) and
(S)-Val (4) (0.5 mM) is also illustrated in Fig. 2. The results
show that both of the chlorides 1a,b are effective substitutes
The proteins produced in these experiments were purified by
metal-ion affinity chromatography and then analysed by 20%
SDS-PAGE, as shown in Fig. 1. Mass spectral analysis of the
PpiB showed that only native protein is obtained using the 20
normal amino acids, as well as from the control experiments
when either (S)-Val (4), (S)-Leu (5) or (2S,3S)-Ile (6) is not
added, and when the mixture of the chlorides 3a,b is added
instead of (2S,3S)-Ile (6). The production of native protein
without added (S)-Val (4), (S)-Leu (5) or (2S,3S)-Ile (6) is
attributable to the background concentrations of these amino
acids, that were shown through separate HPLC analyses to
increase from negligible to around 0.01 mM, in each case,
during the course of the experiments. This could be the result
of amino acid synthesis or release through peptide degrada-
tion. The mass spectral evidence of lack of misincorporation of
either of the chlorides 3a,b as a (2S,3S)-Ile (6) substitute is
consistent with analysis of the gel, which shows that their
addition to the medium has negligible effect on the amount of
PpiB produced. On the other hand, mass spectra of PpiB
produced in the presence of the chlorides 1a,b–2a,b showed
their incorporation. For the Leu analogues 2a,b, the average
level of substitution is around 90%, with around 60% of the
protein having all 5 of the (S)-Leu (5) residues replaced (see
ESI). The gel also shows evidence of some smaller proteins
being formed in this case, that are likely to be truncated forms
Fig. 2 Mass spectra of: (a) PpiB produced using the chloride 1a
(2 mM) instead of (S)-Val (4); (b) native PpiB; (c) PpiB produced using
the chloride 1a (2 mM) and (S)-Val (4) (0.5 mM); and (d) PpiB
produced using the chloride 1b (2 mM) instead of (S)-Val (4).
c
1840 Chem. Commun., 2011, 47, 1839–1841
This journal is The Royal Society of Chemistry 2011

for (S)-Val (4), producing soluble PpiB at concentrations
similar to those formed when (S)-Val (4) is added. Native
PpiB contains 16 (S)-Val (4) residues, and the dominant peaks
in the spectra of protein obtained using 2, 0.2 and 0.02 mM of
the chloride 1a, correlate with the replacement of all 16, 13–14,
and 9 of these, respectively. The corresponding values for the
chloride 1b are 7–9, 5–7, and 3–5. At 2 mM concentration, the
chlorides 1a,b show average levels of incorporation of well
above 90% and around 50%, respectively. The PpiB produced
in the former case is around 30% comprised of protein in
which every one of its 16 residues of (S)-Val (4) is replaced by
the chloride 1a.
Notes and references
1 I. Hirao, T. Kanamori and T. Ueda, in Protein Engineering,
ed. C. Kohrer and U. L. RajBhandary, Nucleic Acids and Mole-
¨
cular Biology, Springer–Verlag, Berlin, 2009, vol. 22, pp. 271–290.
´
2 (a) T. L. Hendrickson, V. de Crecy-Lagard and P. Schimmel, Annu.
Rev. Biochem., 2004, 73, 147–176; (b) S. Lepthien, L. Merkel and
N. Budisa, Angew. Chem., Int. Ed., 2010, 49, 5446–5450.
3 (a) P. M. England, Biochemistry, 2004, 43, 11623–11629; (b) J. Xie
and P. G. Schultz, Nat. Rev. Mol. Cell Biol., 2006, 7, 775–782;
(c) K. J. Rodgers and N. Shiozawa, Int. J. Biochem. Cell Biol.,
2008, 40, 1452–1466.
4 (a) A. S. Spirin, V. I. Baranov, L. A. Ryabova, S. Y. Ovodov and
Y. B. Alakhov, Science, 1988, 242, 1162–1164; (b) T. Kigawa,
T. Yabuki, Y. Yoshida, M. Tsutsui, Y. Ito, T. Shibata and
S. Yokoyama, FEBS Lett., 1999, 442, 15–19; (c) A. S. Spirin,
Trends Biotechnol., 2004, 22, 538–545; (d) Y. Shimizu, Y. Kuruma,
B.-W. Ying, S. Umekage and T. Ueda, FEBS J., 2006, 273,
4133–4140.
The chloride 1b is incorporated less efficiently than the
diastereomer 1a. In this context it is noteworthy that
(2S,3R)-Thr is adenylated by the ValRS of E. coli ML30,
but (2S,3S)-allo-Thr is not.16 In this first step of incorporation,
an OH group is accommodated in place of one Me substituent
of (S)-Val (4) but not the other. The pattern of incorporation
of the chlorides 1a,b parallels that for adenylation of
(2S,3R)-Thr and (2S,3S)-allo-Thr, with the Cl being preferen-
tially accommodated in place of the Me that may be
substituted for OH. The competitive experiment using the
chloride 1a (2.0 mM) and (S)-Val (4) (0.5 mM) shows an
average degree of substitution of around 30% (5 of a possible
16, Fig. 2c), indicating that the overall efficiency of incorpora-
tion of the chloride 1a is B8% (2.0/0.5 mM ꢀ 30%) that of
(S)-Val (4). In one regard this value is not surprising as it
is not too dissimilar to the relative efficiency of adenylation
by ValRS (30%),16 but it is nevertheless remarkable because
in direct contrast to the situation with (2S,3R)-Thr and
(S)-Abu for example, the chloride 1a survives editing post-
adenylation. Apparently the proof-reading sieve of ValRS18
accommodates the HOCHMe and CH2Me side chains
of (2S,3R)-Thr and (S)-Abu, respectively, but excludes the
MeCHMe and ClCHMe groups of (S)-Val (4) and the
chloride 1a.
5 (a) B. Holzberger, M. Rubini, H. M. Moller and A. Marx, Angew.
¨
Chem., Int. Ed., 2010, 49, 1324–1327; (b) J. K. Montclare, S. Son,
G. A. Clark, K. Kumar and D. A. Tirrell, ChemBioChem, 2009, 10,
84–86.
6 M. L. Mock, T. Michon, J. C. M. van Hest and D. A. Tirrell,
ChemBioChem, 2006, 7, 83–87.
7 K. Ozawa, M. J. Headlam, P. M. Schaeffer, B. R. Henderson,
N. E. Dixon and G. Otting, Eur. J. Biochem., 2004, 271,
4084–4093.
8 (a) K. Ozawa, P. S. C. Wu, N. E. Dixon and G. Otting, FEBS J.,
2006, 273, 4154–4159; (b) Y. Xun, P. Tremouilhac, C. Carraher,
C. Gelhaus, K. Ozawa, G. Otting, N. E. Dixon, M. Leippe,
J. Grotzinger, A. J. Dingley and A. V. Kralicek, Protein Expression
¨
Purif., 2009, 68, 22–27.
9 K. Ozawa, M. J. Headlam, D. Mouradov, S. J. Watt, J. L. Beck,
K. J. Rodgers, R. T. Dean, T. Huber, G. Otting and N. E. Dixon,
FEBS J., 2005, 272, 3162–3171.
10 (a) T. Kigawa, E. Yamaguchi-Nunokawa, K. Kodama,
T. Matsuda, T. Yabuki, N. Matsuda, R. Ishitani, O. Nureki and
S. Yokoyama, J. Struct. Funct. Genomics, 2002, 2, 29–35;
(b) V. Bergo, S. Mamaev, J. Olejnik and K. J. Rothschild, Biophys.
J., 2003, 84, 960–966.
11 Z. I. Watts and C. J. Easton, J. Am. Chem. Soc., 2009, 131,
11323–11325.
12 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
13 A. A. Bondi, Physical Properties of Molecular Crystals, Liquids and
Glasses, Wiley, New York, 1968.
In conclusion, the chlorides 1a,b–2a,b are efficiently
incorporated into PpiB as substitutes for (S)-Val (4) and
(S)-Leu (5), respectively, but the chlorides 3a,b do not
substitute for (2S,3S)-Ile (6). In preliminary studies we have
also observed a virtually identical pattern of incorporation
into ubiquitin. These results are of intrinsic interest, especially
since protein with high levels of incorporation of chlorinated
amino acids seems to be soluble and natively folded, and
they also provide a way to obtain novel proteins that
may have interesting properties. Conversely, it is remarkable
in this regard, that PpiB formed using the chloride 1a (2 mM),
and incorporating more than 90% of that amino acid
residue in place of (S)-Val (4), catalyses the interconversion
of the cis- and trans-amide isomers of N-succinyl-(S)-Ala-(S)-
Ala-(S)-Pro-(S)-Phe-4-nitroanilide with an activity similar to
that of native PpiB. The KM and kcat values of the native PpiB
14 (a) W. F. Prouty, J. Virol., 1975, 16, 1090–1093; (b) J. D. Etlinger
and A. L. Goldberg, J. Biol. Chem., 1980, 255, 4563–4568;
(c) K. W. Lanks, Biochim. Biophys. Acta, Gen. Subj., 1990, 1034,
62–66.
15 (a) M. Rabinovitz and H. McGrath, J. Biol. Chem., 1959, 234,
2091–2095; (b) M. Rabinovitz and J. M. Fisher, Biochim. Biophys.
Acta, 1962, 61, 644–646; (c) M. Rabinovitz and J. M. Fisher,
Biochim. Biophys. Acta, 1964, 91, 313–322.
16 F. H. Bergmann, P. Berg and M. Dieckmann, J. Biol. Chem., 1961,
236, 1735–1740.
17 The chloride 1a and (2S,3R)-Thr are structurally analogous, except
for the substitution of OH for Cl, which changes the order of
priority of the substituents at the 2-position when assigning
stereochemistry.
18 S. Fukai, O. Nureki, S.-i. Sekine, A. Shimada, J. Tao,
D. G. Vassylyev and S. Yokoyama, Cell, 2000, 103, 793–803.
19 S. Narayanan, M. R. S. Iyengar, P. L. Ganju, S. Rengaraju,
T. Shomura, T. Tsuruoka, S. Inouye and T. Niida, J. Antibiot.,
1980, 33, 1249–1255.
20 (a) L. Guignard, K. Ozawa, S. E. Pursglove, G. Otting and
159–162;
were determined to be 380 ꢁ 25 mM and 1940 ꢁ 60 sꢂ1
,
N.
E.
Dixon,
FEBS
Lett.,
2002,
524,
(b) M. A. Apponyi, K. Ozawa, N. E. Dixon and G. Otting, in
Structural Proteomics: High Throughput Methods, ed. B. Kobe,
M. Guss and T. Huber, Methods in Molecular Biology, Humana
Press, Totowa, NJ, 2008, vol. 426, pp. 257–268.
respectively, while the corresponding values for the
chlorinated protein are 590 ꢁ 50 mM and 1110 ꢁ 50 sꢂ1, a
less than two-fold difference in each (see ESI).
21 (a) O. Nureki, D. G. Vassylyev, M. Tateno, A. Shimada,
T. Nakama, S. Fukai, M. Konno, T. L. Hendrickson,
P. Schimmel and S. Yokoyama, Science, 1998, 280, 578–582;
(b) A. R. Fersht, Science, 1998, 280, 541.
We gratefully acknowledge financial support received from
the Australian Research Council, and assistance of Mai-Loan
Huynh with mass spectral analysis.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 1839–1841 1841