Efficient Isotope Editing of Proteins for Site-Directed Vibrational Spectroscopy

Article abstract of DOI:10.1021/jacs.5b12680

Vibrational spectra contain unique information on protein structure and dynamics. However, this information is often obscured by spectral congestion, and site-selective information is not available. In principle, sites of interest can be spectrally identi

Full text of DOI:10.1021/jacs.5b12680

Subscriber access provided by Hong Kong University of Science and Technology Library
Article
Efficient isotope editing of proteins for site directed vibrational spectroscopy
Sebastian Peuker, Hanna Andersson, Emil Gustavsson, Kiran Sankar Maiti, Rafal Kania,
Alavi Karim, Stephan Niebling, Anders Pedersen, Mate Erdelyi, and Sebastian Westenhoff
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b12680 • Publication Date (Web): 21 Jan 2016
Downloaded from http://pubs.acs.org on January 25, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Efficient isotope editing of proteins for site
directed vibrational spectroscopy
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
*
1
1
1
1
Sebastian Peuker , Hanna Andersson , Emil Gustavsson , Kiran Sankar Maiti , Rafal Kania ,
1
1
2
1
1*
Alavi Karim , Stephan Niebling , Anders Pedersen , Mate Erdelyi , Sebastian Westenhoff
*
Corresponding authors: peukers@gmail.com, westenho@chem.gu.se
Affiliations
1. Department of Chemistry and Molecular Biology, University of Gothenburg, 40530
Gothenburg, Sweden
2
. Swedish NMR Centre at the University of Gothenburg, PO Box 465, SE-405 30
Gothenburg
Keywords: cellꢀfree protein synthesis, infrared spectroscopy
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
Vibrational spectra contain unique information on protein structure and dynamics. However, this
information is often obscured by spectral congestions and siteꢀselective information is not
available. In principle, sites of interest can be spectrally identified by isotope shifts, but siteꢀ
specific isotope labeling of proteins is today only possible for favorable amino acids or with
prohibitively low yields. Here, we present an efficient cellꢀfree expression system for the siteꢀ
specific incorporation of any isotope labelled amino acid into proteins. We synthesized 1.6 mg of
green fluorescent protein with an isotopeꢀlabelled tyrosine from 100 ml cellꢀfree reaction extract.
We unambiguously identify spectral features of the tyrosine in the fingerprint region of the timeꢀ
resolved infrared absorption spectra. Kinetic analysis confirms the existence of an intermediate
state between photoexcitation and proton transfer, which lives for 3 ps. Our method lifts
vibrational spectroscopy of proteins to a higher level of structural specificity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ACS Paragon Plus Environment

Page 3 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Introduction
Proteins are in constant motion and their diverse functions are ultimately governed by this
1
,2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
dynamic character . Methods to study protein structural dynamics are in demand. Vibrational
3
spectroscopy is a very powerful mean to probe the conformational change of chemical entities .
A number of timeꢀresolved and multidimensional infrared spectroscopic techniques have been
developed and they can be used to directly monitor vibrational coupling of chemical groups in a
4
–8
timeꢀresolved manner . Unfortunately, vibrational spectra of proteins are crowded. The
assignment of spectral features to specific sites is at best tedious, but most often impossible. This
seriously restricts the application of vibrational spectroscopy to study proteins dynamics.
A strategy to overcome the crowdedness of the spectra is to label specific sites of interest in the
8
,9
protein. It is often possible to label cofactors or ligands . Moreover infraredꢀactive reporter
10
groups can be attached to the proteins after protein expression , by introducing unnatural amino
11–16
acids with infraredꢀactive functional groups during proteins synthesis
or by introducing
17–22
isotopeꢀedited amino acids
. Using infraredꢀactive reporter groups has the advantage that they
are typically easily identified in the spectra. These approaches have yielded remarkable insight
13
10
into the rearrangements of secondary structure and the solubilization of proteins , for example.
The latter method, labelling with isotopes, ensures that the chemical properties of the proteins
remain unaltered. This gives the method a distinct advantage over labelling with reporter groups
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
as it allows monitoring active sites that directly participate in the enzymatic activity of the
protein.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The increased mass of the isotopes causes chemical bonds to oscillate at lower frequencies, which
2
3
is observed as redꢀshifted peaks in the vibrational spectra . Infrared spectroscopy of isotope
1
9,20
labelled proteins has yielded unique insights into the folding dynamics of model peptides
and
1
7
amyloid fibres , it was used to solve the threeꢀdimensional dimer arrangement of a membraneꢀ
2
4
bound alpha helical protein , and it has helped to elucidate the photocycle of membrane proton
25
pumps .
Today, the widerꢀspread application of siteꢀspecific isotopeꢀlabelling is restricted by protein
synthesis. Most of the studies mentioned were proteins, which were produced by chemical
1
7,19,20,24
peptide synthesis
. This is a versatile method for siteꢀspecific insertion of isotope labelled
amino acids into proteins; however, it is also restricted to short peptide chains. Nevertheless,
some proteins can be labeled siteꢀselectively by connecting a chemically produced peptide chain
21
with a truncated version of the protein . In in vivo expression systems, siteꢀspecific isotope
labelling is possible when an amino acid occurs only a single time in a polypeptide chain and an
25,26
expression host auxotrophic for that amino acid is available
.
2
6
With in vitro expression systems , tyrosines can be labelled due to the existence of an orthogonal
2
5,27
tRNA/aminoꢀacyl transferase pair in E. coli
. The most general in vitro method is to introduce
4
ACS Paragon Plus Environment

Page 5 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
unnatural amino acids in response to an amber stop codon with chemically aminoacylated
2
8,29
suppressor tRNA
. This strategy makes possible the incorporation of any amino acid, but it has
3
0
not yet been used for infrared spectroscopic studies . This is presumably because of the
moderate protein production yields. Fundamentally limiting the yield is the release factor 1 (RF1)
which competes with the suppressor tRNA for the amber stop codon. This problem has recently
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
1
been addressed in a modified E. coli strain, developed by the Otting group . The reaction
mixture from this strain can be reliably depleted of RF1 with affinity chromatography.
Here, we combine the two approaches for in vitro expression. We explore two strategies for the
synthesis of chemically aminoacylated suppressor tRNA and outline routes for the incorporation
1
8
13
of all native amino acids. Finally, we develop a simplified procedure for O Cꢀlabelled amino
1
3
acids from C precursors, which is of particular practical importance for infrared spectroscopy.
Combined, we were able to produce milligram quantities of siteꢀspecifically isotope labelled
proteins. Using this strategy, we assigned vibrations of a key tyrosine in the active site of green
fluorescent protein (GFP) to peaks in the infrared fingerprint spectral region.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Results
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1: Site-directed introduction of isotope labelled amino acids into proteins.
Scheme of the experimental principle applied to synthesize isotope labelled proteins. Details are given in the main text.
(
AA amino acid; RF1 release factor 1; PG protection group, pdCpA 5'-O-phosphoryl-2'-deoxycytidylyl-(3'--> 5')adenosine).
In the following sections the three major components of the method are described: The cellꢀfree
expression system, the synthesis of the chemically aminoacylated suppressor tRNAC AU AA , and the
synthesis of isotope labelled and appropriately protected amino acids. The final section describes
how isotope labelled GFP was used to assign chromophore vibrations in UV pump ꢀ infrared
probe spectra. Figure 1 is an overview of the synthesis.
Cell free expression system
32,33
We used a cellꢀfree protein expression system in batch mode
. In order to maximize the
incorporation efficiency of the modified amino acids, the S12 extract was produced from a
6
ACS Paragon Plus Environment

Page 7 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
genomically modified E. coli strain BL21 Star (DE3):RF1ꢀCBD3 developed by Otting and
3
1
coworkers . This strain allows the removal of release factor 1 protein (RF1) by means of affinity
chromatography. With this, the termination of the translation is avoided when the stop codon is
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4
reached . The complete removal of RF1 from the S12 was confirmed by Western blotting (Fig.
2
a), and the expression yield of the modified extract was 75 µg protein per mL reaction mixture.
Figure 2: Characterization of the site-directed labelling procedure.
a) Removal of RF1 verified by Western blotting. RF1 was detected by an anti-chitin binding domain antibody (α CBD).
The apparent molecular weight of RF1-(CBD) is 69.9 kDa. (b) Products of the run-off transcription and ligation verified
by a 10 % Novex® TBE-Urea Gel stained with SybrSafe. The length of tRNA(-CA) and tRNA*AA is 74 and 76 bases,
(
3
12 18
13
Phe , tRNATyr C O
Tyr C6
respectively. (c) Test of suppression efficacy of tRNA
, and tRNA
by cell-free expression
CUA
CUA
CUA
Phe8TAG
Tyr66TAG
and GFP
of GFP
respectively. Expression of GFP was quantified by means of fluorescence. (d) SDS-PAGE of
1
3
Tyr C
6
the purification of GFP using tRNA
29.8 kDa.
by Ni-affinity chromatography. The apparent molecular weight of GFP is
CUA
The synthesis of the chemically aminoacylated suppressor tRNACAUAA
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Chemically aminoacylated tRNA(ꢀCA)C AU AA was synthesized by enzymatically ligating 3’
truncated tRNA(ꢀCA)CUA with aminoacylꢀpdCpA (synthesis see below) in vitro using T4 RNA
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
5,36
Ligase 1 (Fig. 2b)
. The precursor, 3’ truncated tRNA(ꢀCA)CUA, was prepared by runꢀoff
transcription from FokI digested template DNA (pYRNA8) (Fig. 2b, left lane) and complete
digestion was observed using 0.25 U FokI per µg pYRNA8 in 15 min. As judged by a 10% TBE
Urea gel, the ligation of aminoacylꢀpdCpA to tRNA(ꢀCA)CUA was quantitative (Fig. 2b, right
lane).
We tested the efficiency of this system by the expression of green fluorescent protein (GFP). The
level of fluorescence is directly proportional to the level of functional full length protein. We
incorporated a phenylalanine at position 8 and a tyrosine at position 66 by expression of GFP
Phe8TAG
Tyr66TAG
from GFPcyc3
and GFPcyc3
templates, respectively. Fig. 2c shows that the level of
expressed GFP scales with the concentration of the respective tRNAC AU AA. The level of expressed
1
2 18
GFP scales equally for the F8 and Y66 labelled with Tyr C O, but the incorporation was lower
1
3
for Y66 labelled with Tyr C6. We attribute this difference to nonꢀoptimal deprotection of the
1
3
particular batch of tRNA CUA 6. Up to 30% of the tRNA AA was translated into proteins.
Tyr C
CUA
8
ACS Paragon Plus Environment

Page 9 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: Chemical intermediates in the preparation of aminoacyl-tRNAs
a) Aminoacylated dinucleotide derivatives 4, 9 and 15 were synthesized (Supplementary Scheme 1–3) and used in the
(
preparation of tRNAs for site-specific incorporation into GFP. The iodine labile 4-PO protecting group in red was used
for phenylalanine derivative 4, while the photolabile NVOC and NB protecting groups in blue and green were used for
18
tyrosine derivatives 9 and 15, respectively. (b) A simplified procedure for O labelling of carboxylic acids, such as that of
1
8
amino acid derivative 6 above, has been developed utilizing O-enriched water, 1,4-dioxane, and commercially available
M HCl in 1,4-dioxane as acid source (Supplementary Scheme 2).
4
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Overview of most common amino acid protecting group strategies applicable for the
preparation of chemically aminoacylated tRNAs.
Protecting group strategy
Examples of natural amino acids
αꢀAmine
Sideꢀchain
Nonꢀpolar; Gly, Ala, Val, Leu, Ile, Met, Pro, Phe
Polar basic; Arg
4
ꢀPO
ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Polar; Ser, Thr, Asn, Gln
Polar acidic; Asp, Glu
Polar basic; Lys
Nonꢀpolar; Gly, Ala, Val, Leu, Ile, Met, Pro, Phe, Trp
Polar basic; Arg
4
4
ꢀPO
ꢀPO
NV/NB
NVOC
ꢀ
NVOC
Polar; Ser, Thr, Cys, Asn, Gln, Tyr
Polar acidic; Asp, Glu
Polar basic; His
NVOC
NVOC
NV/NB
NVOC
Polar basic; Lys
10
ACS Paragon Plus Environment

Page 11 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Synthesis of aminoacylꢀpdCpAs
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Aminoacylated dinucleotide derivatives (aminoacylꢀpdCpAs) were synthesized by
35
aminoacylation of pdCpA with the desired amino acid (Fig. 1) . To prevent sideꢀreactions and to
increase the stability of the amino acidꢀnucleotide ester linkage, the αꢀamino and sideꢀchain
3
7
functional groups were protected. We explored two protecting group strategies . The iodine
labile 4ꢀpentenoyl (4ꢀPO) group was used as an αꢀamino protecting group in the preparation of
phenylalanylꢀpdCpA 4 (Fig. 3 and Supplementary Scheme 1). It is ideal for amino acids without
3
8
sideꢀchain functionality (Table 1) , but can also be applied in orthogonal protecting strategies,
e.g. together with photolabile sideꢀchain protecting groups. However, it is unsuitable for amino
acids having oxidizable sideꢀchains. Tyrosine falls into this category and therefore the photolabile
6ꢀnitroveratryloxycarbonyl (NVOC) and 2ꢀnitrobenzyl (NB) groups were used as αꢀamino and
sideꢀchain protecting groups in the synthesis of tyrosinylꢀpdCpA 9 and 15, respectively (Fig. 3,
Supplementary Schemes 2 and 3). 6ꢀNitroveratryl (NV) is a possible substitute for NB and Table
1
summarizes suitable protection strategies for all natural amino acids. Compound 4 was
produced with an overall yield of 21% starting from ꢀphenylalanine, and 9 and 15 were
L
produced with overall yields of 4% and 6% starting from the commercially available Lꢀtyrosine
13
and
L
ꢀ[phenylꢀ C ]tyrosine, respectively.
6
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
18
A simple method for O labelling of amino acids
1
3
18
Labelling the backbone carbonyl function as C= O is particularly important for probing the
3
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
amide I region of infrared spectra . The amide I vibrations gives rise to strong absorption peaks
ꢀ1
13
18
ꢀ1
at approximately 1640 cm and the introduction of C= O labels results in a ~60 cm red shift
39
13
18
which enables spectral isolation . Unfortunately, C= Oꢀlabelled amino acids are generally not
1
3
commercially available. Synthesis may be achieved from Cꢀlabelled precursors by refluxing the
18
40
compound in a waterꢀfree dioxane/H O mixture at acidic pH . We applied a simplified
2
1
8
procedure for O labelling of the backbone carbonyl in the synthesis of compound 9. The
exchange was performed under acidic conditions, adding 4 M HCl/1,4ꢀdioxane to the amino acid
18
derivative that was premixed with Oꢀenriched water in 1,4ꢀdioxane under nitrogen atmosphere
(
Fig. 3 and Supplementary Scheme 2). The reaction mixture was heated to 100°C using an oil
1
8
bath in a sealed microwave vial overnight. LCꢀMS confirmed quantitative O labelling within 20
h (See the Supplementary Information for analytical data). In contrast to earlier reports, using this
1
8,40
method neither in situ generation of HClꢀgas
, nor repetition of the reaction is necessary for
full conversion.
12
ACS Paragon Plus Environment

Page 13 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Tyr66-13C6
Figure 4: UV pump infrared probe spectra of GFP
.
(a) DFT calculations of infrared absorption spectra of model compounds representing the chromophore of GFP as
illustrated. The isotope labelled part of the molecule is highlighted. The single modes underlying the spectra shown in
grey are plotted as Gaussian functions with a full width at half maximum of 10 cm . (b) UV pump - infrared probe
(red). Three spectral windows were recorded with center photon energies of
370 cm , 1259 cm , and 1165 cm . The averaged and aligned spectra drawn in bold lines and the underlying data sets are
drawn in grey. (c) Difference spectra at two delay times after photoexcitation. The spectrum denoted as ‘650 fs’ is the
average of all spectra from 500 fs to 800 fs and the spectrum denoted as ‘6.5 ps’ is integrated from 3 ps to 10 ps. (d)
Kinetics at the wavenumbers as indicated. The kinetics are integrated over +/- 3 cm .
-
1
Tyr66-13C6
spectra of wild type (black) and GFP
-
1
-1
-1
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
14
ACS Paragon Plus Environment

Page 15 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Isotope edited infrared spectroscopy
We used our isotope labelling method to unambiguously assign the fingerprint vibrations in
difference infrared spectra of tyrosine 66 in GFP. Tyr66 is part of the chromophore and
deprotonates upon photoexcitation within a few picoseconds. The proton is transferred past a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
,41,42
ꢀ1
water molecule and a serine sideꢀchain to glutamate 222
. We target the 1100 cm to
ꢀ1
1400 cm fingerprint region of the infrared spectrum. While this spectral range contains a wealth
of information on chemical structure and dynamics, the features are also notoriously difficult to
3
assign .
To guide the protein synthesis, we calculated the spectral shifts of different isotope labelling
patterns on tyrosine using density functional theory (DFT) calculations. While DFT calculations
are usually not very reliable in predicting the peak position of fingerprint vibrations, the
computation of isotope shifts should be much more robust. Replacing all carbons in the phenyl
13
13
ring by C (Tyr C ) produced the largest predicted shift (Fig. 4a). We therefore synthesized
6
Tyr66Tyr13C6
GFPcyc3
in a preparative scale and purified it by affinity chromatography (Fig. 2d).
Tyr66Tyr13C6
From 100 ml cellꢀfree reaction, 1.6 mg purified and concentrated GFPcyc3
was
Tyr13C6
CUA
obtained, which relates to about 13% of the tRNA
incorporated into the purified protein.
This sample was subjected to UV pump ꢀ infrared probe difference spectrometry using our home
built setup (see methods for details). GFP was photoꢀexcited using laser pulses centered at 400
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ꢀ1
ꢀ1
nm. Infrared difference spectra from 1426 cm to 1117 cm are shown in Fig. 4b for the labelled
GFP (red line) and for a wt GFP sample (black line). Inspection of these spectra reveals that the
ꢀ1
ꢀ1
ꢀ1
position of the positive peaks at 1375 cm and 1188 cm , and the negative peaks at 1292 cm ,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ1
ꢀ1
1
174 cm , and 1152 cm are redꢀshifted in the isotope labelled sample. We conclude that these
peaks must be due to sideꢀchain vibrations of Tyr66. The assignment also allows us to conclude
ꢀ1
that the negative peak at 1240 cm does not originate from the Tyr66 residue. This peak is likely
caused by a different amino acid participating in the photoreactions of GFP.
We also performed DFT computations of the truncated chromophore. DFT calculations are often
used to aid assignment of infrared spectra, however, the computed absolute spectral positions
usually carry a significant uncertainty. This was also the case for our calculations and we
adjusted the wavenumber scales by a constant factor to match the wildꢀtype spectrum. When it
comes to predicting isotopeꢀinduced spectral shifts, DFT should be much more robust, because
the influence of the increased mass of the isotopes on the reduced mass of the vibrations can be
estimated with precision. Indeed, the predicted isotope shifts agree with the observed shifts (Fig
4a and b).
With the tyrosine peaks assigned based on isotopeꢀinduced spectral shifts, we now inspect the
ꢀ1
peaks at 1375 cm closer. The peak converts from a double peak at early times after
photoexcitation (650 fs, Fig. 4c) into a single peak (6.5 ps, Fig 4c). The conversion has a
16
ACS Paragon Plus Environment

Page 17 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
characteristic time constant of 3 +/ꢀ 1 ps (Fig 4d) and the final spectral shape is stable for > 1 ns
(data not shown). We interpret this conversion to be a sign of deprotonation of the
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
3
chromophore . Directly after photoexcitation, the chromophore relaxes in its electronically
excited, protonated state and after 3 ps the proton departs from the phenolꢀOH group, leaving the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
tyrosinate behind .
Discussion
We have demonstrated that the active site of a protein can be siteꢀspecifically isotopeꢀlabelled in
sufficient yields to perform timeꢀresolved infrared spectroscopy. The production yield in our cellꢀ
free system (75 µg protein per mL reaction mixture) was limited by the concentration of
aminoacylated tRNA (Fig S1 and 2c). This compound has to be chemically synthesized for each
amino acid and is also the most expensive reactant. The overall material cost of a labelled protein
was approximately 6000 Dollar/mg protein. The T4 RNA Ligase 1 (New England Biolabs) and
the pdCpa (ATDBio), which are needed in the synthesis of the aminoacylated tRNA, are the main
costꢀdrivers. It would therefore be interesting to explore alternative methods for more efficient
production of the aminoacylated tRNA.
With the label in place, we were readily able to assign peaks in the infrared difference absorption
spectra to specific amino acid vibrations, while conserving the functionality of the wildꢀtype
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
protein. Previously these assignments were not possible, despite the abundance of knowledge
8
,41,42
3
about the GFP photocycle
, spectroscopic experience , and highly developed computational
methods. Specific isotope labelling, as demonstrated here, enabled the spectral interpretation. If
the threeꢀdimensional structure of a protein is known, such definite assignments add a spatial
dimension to infrared spectroscopy.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The fingerprint region of infrared spectra yields highly specific information. This is illustrated by
ꢀ1
the spectral evolution of the peak at 1375 cm (Fig 4c and 4d). From this peak we find that the
deprotonation of Tyr66 occurs after a few picoseconds of dwell time, which confirms earlier
8
,22
proposals . The deprotonation could be limited by additional structural rearrangements of the
protein or the chromophore. However, the dwell time could also reflect the tunneling barrier of
the protons. The latter suggestion is supported by the lengthening of the proton relay time when
8
,41
the protons are exchanged for deuterium . It is conceivable that our method can be used to
deduct specific information about transient states in other photoactive proteins, such as
photoreceptor proteins or photosynthetic proteins.
Here, we used isotopeꢀinduced shifts to assign absorption peaks in difference infrared spectra.
However, isotope shifts can also be used to isolate peaks in absolute spectra. This has, for
17
example, been exploited to study the aggregation behavior of amyloids and unfolding of
1
9
peptides with multidimensional infrared spectroscopy. With the method presented in this paper,
18
ACS Paragon Plus Environment

Page 19 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
larger proteins could in principle be studied in a similar way. Whether the desired peaks can be
isolated requires further investigation, because spectral crowding becomes increasingly
restricting with protein size. Also, it is often not easy to predict how the excitonic coupling of
amide groups would changed by isotope labelling of one specific site. Nevertheless, experiments
that require siteꢀspecific isotope labelling of larger proteins have been suggested. One intriguing
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
4
idea is to determine the ion occupancy in the selection filter of the ion channel KcsA with twoꢀ
3
5
dimensional spectroscopy .. This proposed experiment has not yet been performed because siteꢀ
selective isotope labelling of the protein was not possible. The method presented here can open
this bottleneck.
Methods
Wild-type GFP
45
33
Wild type GFP was produced by cloning GFPcyc3 into the vector pIVEX2.4d (Roche Applied
Science) and heterologous expression in E. coli BL21 Star (DE3) (Invitrogen). Purification was
2+
by Ni ꢀaffinity chromatography (NiꢀNTA Agarose, Quiagen).
Extract preparation, and RF1 depletion
3
1
33
S12 extract was prepared from BL21 (DE3):RFIꢀCBD3 as described by Pedersen . To remove
RF1, the S12 extract was passed over chitin resin (New England Biolabs) directly after the
incubation and removal was confirmed by Western blotting (Fig. 2a). After addition of 1 mM
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DTT the S12 extract was dialyzed twice against extract buffer supplemented with βꢀ
mercaptoethanol (1 ml/l), flash frozen, and stored at ꢀ80 °C.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cell-free protein expression
3
3
Cellꢀfree protein expression was performed in batch mode as described . For every batch of S12
2
+
extract the optimal Mg was determined by maximizing GFP fluorescence at 520 nm. To
AA*NVOC/NB
preprare tRNA(ꢀCA)
, aminoacylated pdCpA (1–15) was synthesized as described in
CUA
the extended methods second (SI) and coupled to tRNA(ꢀCA)CUA, prepared by runꢀoff
46,47
48
transcription
, by using T4 RNA Ligase 1 (New England Biolabs). Deprotection of ~25 µM
AA*NVOC/NB
tRNA(ꢀCA)
was done at acidic pH, 2 ºC at a volume of 400 µl in a 1.5 ml in a
CUA
Protein LoBind tube (Eppendorf, serves as an additional HP filter >320 nm). The sample was
illuminated for 10 min using a 500 W Xe lamp equipped with a water filter and a 320 nm HP
filter (Schott). After exposure, samples were immediately flash frozen, stored at ꢀ80 ºC, and used
AA*pentenyl
without further purification. Deprotection of ~25 µM tRNA(ꢀCA)
was achieved by
CUA
treating the suppressor tRNA with 5 mM iodine (10% (v/v) tetrahydrofuran) for 45 min at 25 ºC.
AA
CUA
After deprotection 0.3 M NaOAc (pH 5.2) was added, tRNA(ꢀCA)
isopropanol, resuspended in 1 mM NaOAc (pH 4.5) and stored at ꢀ80 ºC.
precipitated by
Isotope-labelling
20
ACS Paragon Plus Environment

Page 21 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Details of the synthesis are provided in the SI
DFT calculations
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Density functional theory (DFT) calculations were performed with a system comprising the GFP
chromophore and a coordinating water molecule in the gas phase with the software package
TURBOMOLE 6.6. More details are found in the SI.
Pump-probe spectroscopy
For timeꢀresolved pump probe infrared spectroscopy the sample was excited by the second
harmonic (400 nm, diameter of ~120 µm, 250 nJ/pulse) of a modeꢀlocked femtosecond
Ti:sapphire amplifier at 5 kHz repetition rate (Spitfire Ace, SpectraꢀPhysics) and probed by
infrared pulse generated by a Topas Twin (Light conversion) and nonꢀcollinear difference
frequency generation (approximately 3 nJ/pulse and a diameter of ~180 µm at the sample
position). In a first set of experiments (Fig 4b), the excitation beam was passed through a 10 cm
long glass rod (15 mm aperture, 100 mm L High NA, Hexagonal Light Pipe) to stretch the pulse
to a duration of approximately 2 ps. A second set of experiments (Fig 4c and 4d) was performed
with femtosecond pulses with a duration of approximately 100fs. Reference probe pulses, which
were note guided through the sample, were used to correct for shotꢀtoꢀshot laser fluctuations. A
iHR 320 spectrometer (Horiba) equipped with a grating (5 grooves/mm, 8000 nm blaze
wavelength, Horiba) and two 64 element mercury cadmium telluride detector arrays (MCTꢀ8ꢀ
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
x64, Infrared Associates) with associated digitizing software (Infrared System Development
Corporation, FPAS 0144) were used for detection. To limit photo degradation the sample was
2
moved perpendicular to the probe beam direction over a total area of ~1.1 cm using two linear
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
stages in xyꢀconfiguration (4218Sꢀ04ꢀ01R0, Newmark).
22
ACS Paragon Plus Environment

Page 23 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Acknowledgement
We thank JeanꢀMichel Betton and Cordula Jany for providing the GFP cycle 3 vector, Sydney
Hecht for the pYRNA8 vector, and Gabriela Kovácsová for providing the blot in Fig. 2d. Ashley
Hughes and Matthijs Panman are acknowledged for critically reading the manuscript. We thank
Linnea Isaksson for help with the additional assay (SI Fig 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
13
Supporting Information. Extended Methods, Reaction Schemes, H and C NMR Spectra, CD
spectra, and HPLC (ESI /ESI /UV). This material is available free of charge via the Internet at
http://pubs.acs.org.
+
ꢀ
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
1)
Henzler-Wildman, K.; Kern, D. Nature 2007, 450 (7172), 964.
(2)
Vinson, V. J. Science (80-. ). 2009, 324 (5924), 197.
(
3)
Barth, A. Biochim. Biophys. Acta 2007, 1767 (9), 1073.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(4)
Hamm, P.; Lim, M. H.; Hochstrasser, R. M. J. Phys. Chem. B 1998, 102 (31), 6123.
(
5)
Kolano, C.; Helbing, J.; Kozinski, M.; Sander, W.; Hamm, P. Nature 2006, 444 (7118), 469.
Zheng, J.; Kwak, K.; Asbury, J.; Chen, X.; Piletic, I. R.; Fayer, M. D. Science (80-. ). 2005, 309 (5739), 1338.
Zhao, W.; Wright, J. C. Phys. Rev. Lett. 1999, 83 (10), 1950.
(6)
(
(
(
(
7)
8)
9)
10)
Fang, C.; Frontiera, R. R.; Tran, R.; Mathies, R. A. Nature 2009, 462 (7270), 200.
Deng, H.; Callender, R. In Infrared and Raman Spectroscopy of Biological Materials, eds Gremlich and Yan; 2001; pp 477–515.
King, J. T.; Kubarych, K. J. J. Am. Chem. Soc. 2012, 134 (45), 18705.
(11)
Waegele, M. M.; Culik, R. M.; Gai, F. J Phys Chem Lett 2011, 2, 2598.
(
12)
Liu, C. C.; Schultz, P. G. Annu. Rev. Biochem. Vol 79 2010, 79, 413.
(13)
Ye, S. X.; Zaitseva, E.; Caltabiano, G.; Schertler, G. F. X.; Sakmar, T. P.; Deupi, X.; Vogel, R. Nature 2010, 464 (7293), 1386.
Bazewicz, C. G.; Lipkin, J. S.; Smith, E. E.; Liskov, M. T.; Brewer, S. H. J. Phys. Chem. B 2012, 116 (35), 10824.
Schultz, K. C.; Supekova, L.; Ryu, Y.; Xie, J.; Perera, R.; Schultz, P. G. J. Am. Chem. Soc. 2006, 128 (43), 13984.
Getahun, Z.; Huang, C.-Y.; Wang, T.; De León, B.; DeGrado, W. F.; Gai, F. J. Am. Chem. Soc. 2003, 125 (2), 405.
Middleton, C. T.; Marek, P.; Cao, P.; Chiu, C.; Singh, S.; Woys, A. M.; de Pablo, J. J.; Raleigh, D. P.; Zanni, M. T. Nat Chem 2012,
(
(
(
(
14)
15)
16)
17)
4
(5), 355.
(18)
Middleton, C. T.; Woys, A. M.; Mukherjee, S. S.; Zanni, M. T. Methods 2010, 52 (1), 12.
(
(
19)
Ihalainen, J. A.; Paoli, B.; Muff, S.; Backus, E. H. G.; Bredenbeck, J.; Woolley, G. A.; Caflisch, A.; Hamm, P. Proc. Natl. Acad. Sci.
U. S. A. 2008, 105 (28), 9588.
20)
Smith, A. W.; Lessing, J.; Ganim, Z.; Peng, C. S.; Tokmakoff, A.; Roy, S.; Jansen, T. L. C.; Knoester, J. J. Phys. Chem. B 2010, 114
(
34), 10913.
(21)
Schwans, J. P.; Sunden, F.; Gonzalez, A.; Tsai, Y.; Herschlag, D. Biochemistry 2013, 52 (44), 7840.
24
ACS Paragon Plus Environment

Page 25 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(22)
Stoner-Ma, D.; Jaye, A. A.; Matousek, P.; Towrie, M.; Meech, S. R.; Tonge, P. J. J. Am. Chem. Soc. 2005, 127 (9), 2864.
Decatur, S. M. Acc. Chem. Res. 2006, 39 (3), 169.
(
23)
(24)
Remorino, A.; Korendovych, I. V; Wu, Y. B.; DeGrado, W. F.; Hochstrasser, R. M. Science (80-. ). 2011, 332 (6034), 1206.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
(
25)
26)
Sonar, S.; Lee, C. P.; Coleman, M.; Patel, N.; Liu, X. M.; Marti, T.; Khorana, H. G.; Rajbhandary, U. L.; Rothschild, K. J. Nat.
Struct. Biol. 1994, 1 (8), 512.
Bernhard, F.; Tozawa, Y. Curr Opin Struct Biol 2013, 23 (3), 374.
(27)
Yabuki, T.; Kigawa, T.; Dohmae, N.; Takio, K.; Terada, T.; Ito, Y.; Laue, E.; Cooper, J.; Kainosho, M.; Yokoyama, S. J. Biomol.
NMR 1998, 11 (3), 295.
(
(
(
28)
29)
30)
Noren, C. J.; Anthony-Cahill, S. J.; Griffith, M. C.; Schultz, P. G. Science (80-. ). 1989, 244 (4901), 182.
Ellman, J. A.; Volkman, B. F.; Mendel, D.; Schultz, P. G.; Wemmer, D. E. J Am Chem Soc 1992, 114 (20), 7959.
Sobhanifar, S.; Reckel, S.; Junge, F.; Schwarz, D.; Kai, L.; Karbyshev, M.; Lohr, F.; Bernhard, F.; Dotsch, V. J. Biomol. NMR 2010,
46 (1), 33.
(
31)
Loscha, K. V; Herlt, A. J.; Qi, R. H.; Huber, T.; Ozawa, K.; Otting, G. Angew. Chemie-International Ed. 2012, 51 (9), 2243.
Isaksson, L.; Enberg, J.; Neutze, R.; Göran Karlsson, B.; Pedersen, A. Protein Expr. Purif. 2012, 82 (1), 218.
Pedersen, A.; Hellberg, K.; Enberg, J.; Karlsson, B. G. N Biotechnol 2011, 28 (3), 218.
(32)
(
33)
(34)
Park, H. S.; Hohn, M. J.; Umehara, T.; Guo, L. T.; Osborne, E. M.; Benner, J.; Noren, C. J.; Rinehart, J.; Soll, D. Science (80-. ).
2011, 333 (6046), 1151.
(
(
35)
36)
Robertson, S. A.; Ellman, J. A.; Schultz, P. G. J Am Chem Soc 1991, 113 (7), 2722.
England, T. E.; Uhlenbeck, O. C. Nature 1978, 275 (5680), 560.
(37)
Hecht, S. M. Protein Engineering: Nucleic Acids and Molecular Biology; Springer, 2009; Vol. 22.
Lodder, M.; Golovine, S.; Laikhter, A. L.; Karginov, V. A.; Hecht, S. M. J Org Chem 1998, 63 (3), 794.
(
38)
(39)
Ihalainen, J. A.; Bredenbeck, J.; Pfister, R.; Helbing, J.; Chi, L.; van Stokkum, I. H. M.; Woolley, G. A.; Hamm, P. Proc. Natl.
Acad. Sci. U. S. A. 2007, 104 (13), 5383.
(
(
40)
41)
Marecek, J.; Song, B.; Brewer, S.; Belyea, J.; Dyer, R. B.; Raleigh, D. P. Org Lett 2007, 9 (24), 4935.
van Thor, J. J. Chem. Soc. Rev. 2009, 38 (10), 2935.
(42)
Di Donato, M.; van Wilderen, L. J. G. W.; Van Stokkum, I. H. M.; Stuart, T. C.; Kennis, J. T. M.; Hellingwerf, K. J.; van
Grondelle, R.; Groot, M. L. Phys. Chem. Chem. Phys. 2011, 13 (36), 16295.
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(43)
Chattoraj, M.; King, B. A.; Bublitz, G. U.; Boxer, S. G. Proc. Natl. Acad. Sci. 1996, 93 (16), 8362.
(
44)
Köpfer, D. A.; Song, C.; Gruene, T.; Sheldrick, G. M.; Zachariae, U.; de Groot, B. L.; Kopfer, D. A. Science (80-. ). 2014, 346
(
6207), 352.
(
(
(
45)
46)
47)
Crameri, A.; Whitehorn, E. A.; Tate, E.; Stemmer, W. P. Nat Biotechnol 1996, 14 (3), 315.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Gao, R.; Zhang, Y.; Choudhury, A. K.; Dedkova, L. M.; Hecht, S. M. J Am Chem Soc 2005, 127 (10), 3321.
Robertson, S. A.; Noren, C. J.; Anthony-Cahill, S. J.; Griffith, M. C.; Schultz, P. G. Nucleic Acids Res. 1989, 17 (23), 9649.
England, T. E.; Bruce, A. G.; Uhlenbeck, O. C. Methods Enzym. 1980, 65 (1), 65.
(48)
26
ACS Paragon Plus Environment

Page 27 of 27
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Insert Table of Contents artwork here
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27
ACS Paragon Plus Environment