Synthesis and evaluation of a library of fluorescent dipeptidomimetic analogues as substrates for modified bacterial ribosomes

Article abstract of DOI:10.1021/acs.biochem.6b00102

Described herein are the synthesis and photophysical characterization of a library of aryl-substituted oxazole- and thiazole-based dipeptidomimetic analogues, and their incorporation into position 66 of green fluorescent protein (GFP) in lieu of the natural fluorophore. These fluorescent analogues resemble the fluorophore formed naturally by GFP. As anticipated, the photophysical properties of the analogues varied as a function of the substituents at the para position of the phenyl ring. The fluorescence emission wavelength maxima of compounds in the library varied from ～365 nm (near-UV region) to ～490 nm (visible region). The compounds also exhibited a large range of quantum yields (0.01-0.92). The analogues were used to activate a suppressor tRNACUA and were incorporated into position 66 of GFP using an in vitro protein biosynthesizing system that employed engineered ribosomes selected for their ability to incorporate dipeptides. Four analogues with interesting photophysical properties and reasonable suppression yields were chosen, and the fluorescent proteins (FPs) containing these fluorophores were prepared on a larger scale for more detailed study. When the FPs were compared with the respective aminoacyl-tRNAs and the actual dipeptide analogues, the FPs exhibited significantly enhanced fluorescence intensities at the same concentrations. Part of this was shown to be due to the presence of the fluorophores as an intrinsic element of the protein backbone. There were also characteristic shifts in the emission maxima, indicating the environmental sensitivity of these probes. Acridon-2-ylalanine and oxazole 1a were incorporated into positions 39 and 66 of GFP, respectively, and were shown to form an efficient F?rster resonance energy transfer (FRET) pair, demonstrating that the analogues can be used as FRET probes.

Full text of DOI:10.1021/acs.biochem.6b00102

Subscriber access provided by Illinois Institute of Technology
Article
Synthesis and Evaluation of a Library of Fluorescent Dipeptidomimetic
Analogues as Substrates for Modified Bacterial Ribosomes
Sandipan Roy Chowdhury, Pradeep S Chauhan, Larisa M Dedkova,
Xiaoguang Bai, Shengxi Chen, POULAMI TALUKDER, and Sidney M. Hecht
Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b00102 • Publication Date (Web): 06 Apr 2016
Downloaded from http://pubs.acs.org on April 6, 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.
Biochemistry 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 48
Biochemistry
1
2
3
4
5
6
7
Synthesis and Evaluation of a Library of Fluorescent
Dipeptidomimetic Analogues as Substrates for Modified Bacterial
Ribosomes
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
Sandipan Roy Chowdhury,† Pradeep S. Chauhan,† Larisa M. Dedkova,† Xiaoguang
Bai,† Shengxi Chen,† Poulami Talukder and Sidney M. Hecht*,†
†Center for BioEnergetics, Biodesign Institute, and School of Molecular Sciences, Arizona State
University, Tempe, AZ 85287, United States
1
ACS Paragon Plus Environment

Biochemistry
Page 2 of 48
1
2
3
4
5
6
7
8
9
ABSTRACT: Described herein is the synthesis and photophysical characterization of a library
of aryl substituted oxazole and thiazole based dipeptidomimetic analogues, and their
incorporation into position 66 of green fluorescent protein (GFP) in lieu of the natural
fluorophore. These fluorescent analogues resemble the fluorophore formed naturally by GFP. As
anticipated, the photophysical properties of the analogues varied as a function of the substituents
at the para-position of the phenyl ring. The maximum fluorescence emission wavelength of
compounds in the library varied from ~365 nm (near UV region) to ~490 nm (visible region).
The compounds also exhibited a large range of quantum yields (0.01 to 0.92). The analogues
were used to activate a suppressor tRNACUA and were incorporated into position 66 of GFP using
an in vitro protein biosynthesizing system which employed engineered ribosomes selected for
their ability to incorporate dipeptides. Four analogues with interesting photophysical properties
and reasonable suppression yields were chosen and the fluorescent proteins (FPs) containing
these fluorophores were prepared on a larger scale for more detailed study. When the FPs were
compared with the respective aminoacyl-tRNAs and the actual dipeptide analogues, the FPs
exhibited significantly enhanced fluorescence intensities at the same concentrations. Part of this
was shown to be due to the presence of the fluorophores as an intrinsic element of the protein
backbone. There were also characteristic shifts in the emission maxima indicating the
environmental sensitivity of these probes. Acridon-2-ylalanine (Acd) and oxazole 1a were
incorporated into positions 39 and 66 of GFP, respectively, and were shown to form an efficient
Fӧrster resonance energy transfer (FRET) pair, demonstrating that the analogues can be used as
FRET probes.
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
2
ACS Paragon Plus Environment

Page 3 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
Fluorescent labeling of proteins and peptides can provide a powerful tool to monitor their
behavior in vitro as well as in vivo.1 Generally, fluorescent labeling is achieved by employing a
derivative of the fluorophore that reacts specifically with certain functional groups present in the
biomolecule.2 Numerous fluorescent probes are currently available which possess a wide range
of properties and functionalities to meet the requirements of the labeling technique and the
technical goals of specific studies. There are, however, certain caveats associated with the use of
synthetic fluorophores.2 For example, they should be small in size and chemically stable under
physiological conditions, they should not perturb the structure or function of the
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
peptides/proteins being labeled, and when used in intact cells or organisms, they should not be
cytotoxic and should be permeable to cell membranes. In addition, when used to measure protein
conformational changes, the covalent linkage between the synthetic probe and the specific
residue in the target molecule should provide the probe with a range of motion that is small
relative to the change in protein conformation being monitored.3
Following the successful cloning of green fluorescent protein4 (GFP) and its utilization as a
marker for gene expression,5 this application for GFP and its variants has been used widely.6
There are certain advantages of genetic tagging over chemical labeling of expressed proteins.7
For example, since the fluorescent protein is fused to the target protein, they are colocalized with
absolute specificity, which obviates the problem of non-specific post-synthetic labeling. The
fluorophore is embedded in the protein backbone and sheltered by the -barrel structure,
imparting a controlled environment for the probe. Nonetheless, there are significant limitations
of using fluorescent proteins as labels, including their need for posttranslational activation, their
limited stability and their large size.
Over a period of several years, we have reported the reengineering of E. coli ribosomes,
3
ACS Paragon Plus Environment

Biochemistry
Page 4 of 48
1
2
3
4
enabling them to incorporate -D-amino acids8,9 and -amino acids,10,11 while still incorporating
5
6
7
8
9
-L-amino acids with good fidelity. More recently, we have extended this strategy to the
selection of ribosomes capable of incorporating dipeptides and dipeptide analogues from an
activated suppressor tRNA into a specific site in a protein in a single ribosomal bond-forming
event by suppression of a nonsense codon.12,13 This has added importantly to the repertoire of
amino acid analogues potentially amenable to incorporation, since the modified ribosomes
apparently fail to recognize the amide bond connecting the two incorporated amino acids.
Accordingly, the modified ribosomes were capable of incorporating a thiodipeptide, as well as
fluorescent oxazole (1a) and thiazole (8a) heterocycles (Figure 1) substituted with NH2 and
COOH groups at distances which approximated those between such functional groups in a
dipeptide.12,13
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
The structures of the dipeptidomimetic analogues incorporated were particularly interesting
since they resemble the GFP fluorophore and are also fluorescent. When dipeptidomimetic
analogues 1a and 8a were incorporated into position 66 of GFP by suppression of a UAG codon
introduced into that position in lieu of the normal tyrosine codon, the resulting FPs had
fluorescence intensities significantly stronger than wild-type GFP.12,13 These dipeptidomimetic
analogues do not require activation by posttranslational modification; further, they are
completely stable chemically both as the free “dipeptides” and also as constituents of proteins.
The success of these studies12,13 suggested that it should be possible to identify additional
fluorescent dipeptidomimetic analogues having diverse photophysical properties. Described in
this report are the synthesis and photophysical properties of a library of dipeptidomimetic
analogues and their ribosomal incorporation into position 66 of Aequorea victoria GFP. Selected
fluorophores with interesting properties were made in greater amounts and characterized with
4
ACS Paragon Plus Environment

Page 5 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
regard to their environmental sensitivity and ability to act as a FRET donor following
incorporation into the -barrel structural domain of GFP, using acridon-2-ylalanine (Acd)14 as
the acceptor.
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
MATERIALS AND METHODS
All experiments requiring anhydrous conditions were conducted in flame-dried glassware fitted
with a rubber septum under a positive pressure of dry nitrogen or dry argon. Reactions were
performed at room temperature unless otherwise indicated. Analytical thin layer chromatography
was performed using glass plates pre-coated with silica gel (0.25 mm, 60 Å pore size, 230–400
mesh, Silicycle) impregnated with a fluorescent indicator (254 nm). TLC plates were visualized
by exposure to ultraviolet (UV) light. Flash column chromatography was performed employing
silica gel (60 Å pore size, 40–63 μm, standard grade, Silicycle). An acetone cooling bath was
used for the appropriate low temperature reactions by the addition of portions of dry ice.
1H NMR and 13C NMR spectra were recorded on Varian INOVA 400 (400 MHz) and Varian
INOVA 500 (500 MHz) spectrometers at 25 °C. Proton chemical shifts are expressed in parts per
million (ppm, δ scale) and are referenced to residual protium in the NMR solvent (CDCl3,
DMSO-d6 or CD3OD). Splitting patterns are designated as follows: s, singlet; br s, broad singlet;
d, doublet; dd, doublet of doublets; t, triplet; q, quartet; m, multiplet. High resolution mass
spectra were obtained at the Arizona State University CLAS High Resolution Mass Spectrometry
Facility or the Michigan State University Mass Spectrometry Facility. HPLC purification was
performed with a Waters 600 pump coupled with a Varian ProStar 340 detector and a Grace
Econosil C18 column (250  10 mm, 5 μm). The tetra-n-butylammonium (TBA) salt of pdCpA
was prepared using Dowex 50W X8, 200–400 mesh, activated in its TBA form.
5
ACS Paragon Plus Environment

Biochemistry
Page 6 of 48
1
2
3
4
5
6
7
8
9
The chemicals used for synthesis were purchased from Aldrich Chemical Co., Sigma Chemical
Co., Fisher Scientific, Oakwood Chemicals or Combi-Blocks. THF was distilled under argon
using sodium-benzophenone ketyl; CH2Cl2 was distilled under argon using calcium hydride.
Synthesis of pdCpA Derivatives of Dipeptidomimetic Analogues. The route employed
for the synthesis of the pdCpA derivative of dipeptidomimetic analogue 5a and its attachment to
suppressor tRNACUA-COH is outlined in Scheme 1. The syntheses of the pdCpA derivatives of
dipeptidomimetics 2a-4a, 6a, 7a and 9a-11a are outlined in Schemes S1 to S8. The syntheses of
the pdCpA derivatives of the dipeptidomimetic analogues 1a and 8a, and acceptor Acd have
been reported previously.12-14 The experimental procedures and compound characterizations are
provided in the Supporting Information.
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
Methyl 3-(4-(1H-Pyrrol-1-yl)phenyl)-2-(2-((tert-butoxycarbonyl)amino)acetamido)-3-
oxopropanoate (17). To a solution of 0.83 g (4.41 mmol) of acid 13 in 25 mL of THF were
added 1.59 mL (1.16 g, 18.6 mmol) of Et3N and 0.74 mL (0.78 g, 9.31 mmol) of isobutyl
chloroformate. The reaction mixture was stirred at room temperature for 2 h to obtain the acid
anhydride 14, which was used as a crude material in the next step without further purification.
To a solution of 1.17 g (4.63 mmol) of imine ester 15 in 100 mL of THF was added 4.63 mL
(4.63 mmol) of 1 M NaHMDS solution in THF at –78 °C. After 30 min, the crude acid anhydride
14 was added to the reaction mixture, which was stirred at –78 °C for 2 h. The reaction mixture
was quenched by the addition of conc HCl solution until pH ~ 2 was reached. The reaction
mixture was concentrated under diminished pressure to obtain the amine salt 16 as a colorless
solid which was used in the next reaction without further purification.
To a solution of 0.85 g (4.85 mmol) of Boc-Gly-OH in 30 mL of DMF were added 2.14 g (4.85
mmol) of BOP reagent and 0.67 mL (0.48 g, 4.85 mmol) of Et3N. After 2 min, the amine salt 16
6
ACS Paragon Plus Environment

Page 7 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
dissolved in 20 mL of DMF was added to the reaction mixture, which was stirred overnight at
room temperature. The reaction mixture was diluted with 300 mL of water and extracted with
two 50-mL portions of EtOAc. The organic phase was dried (MgSO4) and concentrated under
diminished pressure. The residue was purified by chromatography on a silica gel column (10 × 4
cm). Elution with 1:1 hexanes–ethyl acetate gave the desired product 17 as a colorless foam:
yield 1.06 g (58% overall yield from acid 13); silica gel TLC Rf 0.50 (1:1 hexanes–ethyl acetate);
1H NMR (CDCl3) 1.44 (s, 9H), 3.71 (s, 3H), 3.93 (d, 2H, J = 4.0 Hz), 5.39 (br s, 1H), 6.21 (d,
1H, J = 8.0 Hz), 6.38-6.39 (m, 2H), 7.16-7.17 (m, 2H), 7.47 (d, 2H, J = 12.0 Hz) 7.60 (d, 1H, J =
8.0 Hz) and 8.17 (d, 2H, J = 8.0 Hz); 13C NMR (CDCl3) 28.5, 44.3, 53.5, 57.9, 80.5, 111.3,
119.1, 119.2, 119.4, 123.3, 130.8, 131.7, 131.8, 145.2, 150.7, 156.2, 167.2, 169.9 and 189.9;
mass spectrum (APCI), m/z 416.1824 (M+H)+ (C21H26N3O6 requires m/z 416.1822).
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
Methyl 5-(4-(1H-Pyrrol-1-yl)phenyl)-2-(((tert-butoxycarbonyl)amino)methyl)oxazole-4-
carboxylate (18). To a stirred solution of 0.23 g (0.88 mmol) of triphenylphosphine and 0.22 g
(0.88 mmol) of iodine in 50 mL of CH2Cl2 was added 0.24 mL (0.17 g, 1.75 mmol) of Et3N. The
dark yellow solution was stirred for 5 min and 0.18 g (0.44 mmol) of ketoamide 17 was added to
the reaction mixture, which was stirred at room temperature for 2 h. The reaction mixture was
concentrated under diminished pressure and the residue was purified by chromatography on a
silica gel column (10 × 1 cm). Elution with 1:1 hexanes–ethyl acetate gave the desired product
18 as a colorless solid: yield 0.11 g (63%); silica gel TLC Rf 0.50 (1:1 hexanes–ethyl acetate); 1H
NMR (CDCl3) 1.45 (s, 9H), 3.91 (s, 3H), 4.51 (d, 2H, J = 8.0 Hz), 5.34 (br s, 1H), 6.34-6.35
(m, 2H), 7.11-7.13 (m, 2H), 7.42-7.45 (m, 2H) and 8.10-8.13 (m, 2H); 13C NMR (CDCl3) 28.5,
38.1, 53.6, 80.5, 111.4, 119.1, 119.3, 119.8, 120.2, 123.7, 123.7, 126.6, 128.1, 129.9, 142.0,
7
ACS Paragon Plus Environment

Biochemistry
Page 8 of 48
1
2
3
4
5
6
155.4, 155.7, 159.7 and 162.6; mass spectrum (APCI), m/z 398.1714 (M+H)+ (C21H24N3O5
requires m/z 398.1716).
7
8
9
Methyl 5-(4-(1H-Pyrrol-1-yl)phenyl)-2-(pent-4-enamidomethyl)oxazole-4-carboxylate
(5b). To a solution of 0.20 g (0.55 mmol) of Boc-protected amine 18 in 10 mL of CH2Cl2 was
added 10 mL of CF3COOH. The reaction mixture was stirred overnight and concentrated under
diminished pressure to afford the deprotected amine which was used in the next step without
purification.
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
To the solution of deprotected amine in 15 mL of THF was added 0.11 g (0.55 mmol) of 4-
pentenoyloxysuccinimide followed by 0.14 mL (0.10 g, 1.00 mmol) of Et3N. The reaction
mixture was stirred overnight at room temperature and then concentrated under diminished
pressure. The residue was purified by chromatography on a silica gel column (10 × 1 cm).
Elution with 1:10 MeOH–ethyl acetate gave the desired product 5b as a colorless solid: yield
0.14 g (72%); silica gel TLC Rf 0.75 (1:10 MeOH–ethyl acetate); 1H NMR (CDCl3) 2.31-2.38
(m, 4H), 3.86 (s, 3H), 4.59 (d, 2H, J = 4.0 Hz), 4.92-5.03 (m, 2H), 5.76-5.78 (m, 1H),
6.31-6.32 (m, 2H), 6.82-6.85 (m, 1H), 7.08-7.09 (m, 2H), 7.38-7.41 (m, 2H) and
8.04-8.06 (m, 2H); 13C NMR (CDCl3) δ 29.5, 35.5, 36.6, 52.4, 111.4, 115.8, 119.0,
119.2, 119.3, 119.6, 122.9, 123.5, 126.4, 129.9, 131.6, 136.9, 142.0, 155.3, 159.4,
162.4 and 172.8; mass spectrum (APCI), m/z 380.1607 (M+H)+ (C21H22N3O4
requires m/z 380.1610).
Cyanomethyl 5-(4-(1H-Pyrrol-1-yl)phenyl)-2-(pent-4-enamidomethyl)oxazole-4-
carboxylate (19). To a solution of 0.12 g (0.31 mmol) of ester 5b in 6 mL of 1:1 MeOH and
THF was added 0.46 mL (0.46 mmol) of 1 M aq LiOH. The reaction mixture was stirred
8
ACS Paragon Plus Environment

Page 9 of 48
Biochemistry
1
2
3
4
5
6
overnight at room temperature and then concentrated under diminished pressure to obtain the
free acid.
7
8
9
To a solution of the acid in 15 mL of DMF was added 0.06 mL (0.07 g, 0.93 mmol) of
chloroacetonitrile followed by 0.22 mL (0.16 g, 1.54 mmol) of Et3N. The reaction mixture was
stirred overnight at room temperature and then concentrated under diminished pressure. The
residue was purified by chromatography on a silica gel column (10 × 1 cm). Elution with 1:10
MeOH–ethyl acetate gave the desired product 19 as a yellow solid: yield 70.0 mg (56%); silica
gel TLC Rf 0.75 (1:10 MeOH–ethyl acetate); 1H NMR (CDCl3) 2.37-2.44 (m, 4H), 4.66 (d, 2H,
J = 8.0 Hz), 4.95 (s, 2H), 4.97-5.10 (m, 2H), 5.80-5.84 (m, 1H), 6.27 (br s, 1H), 6.38-6.39 (m,
2H), 7.15-7.16 (m, 2H), 7.48-7.51 (m, 2H) and 8.11-8.14 (m, 2H); 13C NMR (CDCl3) δ 29.6,
35.7, 36.8, 49.1, 111.8, 114.3, 116.1, 119.2, 119.9, 122.9, 124.7, 130.0, 130.3, 137.0, 142.7,
157.4, 159.7, 160.6, 170.1 and 172.7; mass spectrum (APCI), m/z 405.1563 (M+H)+
(C22H21N4O4 requires m/z 405.1563).
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
5-(4-(1H-Pyrrol-1-yl)phenyl)-2-(pent-4-enamidomethyl)oxazole-4-carboxylic acid pdCpA
Ester (20). To a stirred solution containing 8.50 mg (6.37 μmol) of pdCpA tetrabutylammonium
salt15 in 100 L of 9:1 anhydrous DMF–Et3N was added 6.40 mg (15.9 μmol) of cyanomethyl
ester 19. The reaction mixture was sonicated for 6 h. The reaction mixture was purified by C18
reversed phase HPLC (250  10 mm) using a gradient of 1% to 65% acetonitrile in 50 mM
ammonium acetate, pH 4.5, over a period of 1 h. The retention time of the desired product was
24.6 min. The fractions containing the product were lyophilized to afford 20 as a colorless solid:
yield 2.0 mg (32%); mass spectrum (ESI), m/z 982.2283 (M_H)- (C39H42N11O16P2 requires m/z
982.2286).
9
ACS Paragon Plus Environment

Biochemistry
Page 10 of 48
1
2
3
4
5
6
Measurement of the Photophysical Properties of Fluorescent Dipeptidomimetics.
The UV/vis absorption spectra (220-400 nm) were recorded using a Cary 60 UV/vis
7
8
9
spectrophotometer. Fluorescence quantum yields (Table 1) were determined using the gradient
method.16 Dipeptidomimetic analogues were dissolved in methanol. Solutions of each compound
were made such that the UV absorptions at the maximum wavelength were within the range of
0.01 and 0.1. Anthracene (ΦF 0.27, λex 340 nm) was used as a reference standard to calculate the
fluorescence quantum yields of the dipeptidomimetic analogues according to the formula Φx =
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
2
Φs × (Gradx × nx )/(Grads × ns2 ), where Grad is gradient of the plot of integrated intensity
versus absorbance, n is the refractive index of the solvent, s is the standard of known ΦF, and x is
the tested sample.16
Ligation of Suppressor tRNACUA-COH with Dipeptidomimetics and Deprotection of
N-pentenoyl Group. The activation of suppressor tRNACUA and tRNACCCG was carried out as
described previously.17,18 Briefly, a 100-µL reaction mixture of 100 mM Na Hepes, pH 7.5,
contained 1.0 mM ATP, 15 mM MgCl2, 100 µg of suppressor tRNACUA-COH or tRNACCCG-COH,
0.5 A260 unit of N-pentenoyl-protected aminoacyl-pdCpA, 15% DMSO, and 100 units of T4
RNA ligase. The reaction mixture was incubated at 37 °C for 1.5 h and quenched by the addition
of 0.1 vol of 3 M NaOAc, pH 5.2. The N-protected aminoacylated tRNA was precipitated with 3
vol of cold EtOH. The efficiency of ligation was estimated by 8% polyacrylamide–7 M urea gel
electrophoresis (pH 5.0).19 The N-pentenoyl-protected aminoacyl-tRNACUAs were deprotected by
treatment with 5 mM aqueous I2 at 25 °C for 15 min. The solution was centrifuged, and the
supernatant was adjusted to 0.3 M NaOAc and then treated with 3 vol of cold EtOH to
precipitate the aminoacylated tRNA. The tRNA pellet was collected by centrifugation, washed
with 70% aq EtOH, air dried and dissolved in 10 µL of RNase-free H2O.
10
ACS Paragon Plus Environment

Page 11 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
Preparation of S-30 Extracts from Cells Having Modified Ribosomes. Aliquots (5-10
μL) from liquid stocks of E. coli BL-21(DE-3) cells, harboring plasmids with a wild-type or
modified rrnB gene, were placed on LB agar supplemented with 100 g/mL of ampicillin and
grown at 37 ºC for 16-18 h. One colony was picked from each agar plate and transferred into 3
mL of LB medium supplemented with 100 g/mL of ampicillin and 0.5 mM IPTG.
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
The cultures were grown at 37 ºC for 3-6 h in a thermostated shaker until OD600 ~0.15-0.3 was
reached, then diluted with LB medium supplemented with 100 g/mL ampicillin, 1 mM IPTG
and 3 µg/mL of erythromycin (for selectively enhancing the modified ribosome fraction) until
OD600 0.01 was reached, and then grown at 37 ºC for 12-18 h. The optimal concentration of the
final cultures was OD600 0.5-1.0. Cells were harvested by centrifugation (5000 × g, 4 ºC, 10
min), washed three times with S-30 buffer (1 mM Tris-OAc, pH 8.2, containing 1.4 mM
Mg(OAc)2, 6 mM KOAc and 0.1 mM DTT) supplemented with β-mercaptoethanol (0.5 mL/L)
and once with S-30 buffer having 0.05 mL/L β-mercaptoethanol.
The weight of the wet pellet was estimated and 1.27 mL of S-30 buffer was added to suspend
each 1 g of cells. The volume of the suspension was measured and used for estimating the
amount of other components. Pre-incubation mixture (0.3 mL) (0.29 M Tris, pH 8.2, containing
9 mM Mg(OAc)2, 13 mM ATP, 84 mM phosphoenol pyruvate, 4.4 mM DTT and 5 µM amino
acids mixture), 15 units of pyruvate kinase and 10 μg of lysozyme were added per 1 mL of cell
suspension and the resulting mixture was incubated at 37 ºC for 30 min. The incubation mixture
was then frozen at – 80 ºC (~30 min), melted (37 ºC, 30 min), and again frozen and melted at
room temperature (~30 min). Ethylene glycol tetraacetic acid (EGTA) was then added to 2.5 mM
final concentration and the cells were incubated at 37 ºC for 30 min. The same molar
11
ACS Paragon Plus Environment

Biochemistry
Page 12 of 48
1
2
3
4
5
6
7
8
9
concentration of CaCl2 was added, mixed well and frozen (–80 ºC, 30 min). The frozen mixture
was centrifuged (15,000 × g, 4 ºC, 1 h) and the supernatant was stored in aliquots at –80 ºC.
In vitro Protein Translation. Protein translation reactions were carried out in 15-1700 L of
incubation mixture containing 0.3 L/L of S-30 system, 170 ng/L of plasmid, 35 mM Tris
acetate, pH 7.4, 190 mM potassium glutamate, 30 mM ammonium acetate, 2 mM DTT, 0.2
mg/mL total E. coli tRNA, 3.5% PEG 6000, 20 g/mL folinic acid, 20 mM ATP and GTP, 5
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
mM CTP and UTP, 100 M amino acids mixture, 0.5 Ci/L of 35S-methionine (for 15-L
reactions) and 1 μg/mL rifampicin. In the case of plasmids having a gene with a TAG codon, an
activated suppressor tRNA was added to a concentration of 1.5 g/L. In the case of the plasmid
having a gene with TAG and GGGC codons to enable the incorporation of two unnatural amino
acids, two aminoacyl-tRNAs (aminoacyl-tRNACUA and aminoacyl-tRNACCCG) were added.
Reactions were carried out at 37 C for 1 h (for 15 µL reaction mixtures) to 1.5 h (for 1700 uL
reaction mixtures) and terminated by chilling on ice. Aliquots from in vitro translation mixtures
were analyzed by SDS-PAGE followed by quantification of the radioactive bands by
phosphorimager analysis (for 15 µL reaction mixtures).
Purification of GFP. Samples of GFP were diluted with 50 mM Tris-HCl, pH 8.2, and applied
to a 100 µL Ni-NTA agarose column that had been equilibrated with the same buffer. The
column was washed with 1 mL of the same buffer and GFP was eluted with 500 µL of the 50
mM Tris-HCl, pH 8.2, containing 250 mM imidazole. A 50-µL column of DEAE-Sepharose was
equilibrated with two 500-µL portions of 25 mM Tris-HCl, pH 7.4. Samples of GFP purified by
Ni-NTA chromatography were diluted 3-fold in the same buffer and applied to the resin. The
column was washed with four 500-µL portions of the same buffer. GFP was eluted from the
resin with 25 mM Tris-HCl, pH 7.4, containing 0.25 M NaCl. The fractions were analyzed on a
12
ACS Paragon Plus Environment

Page 13 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
15% polyacrylamide gel, stained using Coomassie R-250, and protein-containing fractions were
combined and concentrated by the use of an Amicon Ultra 10K filter device. The concentrations
of all modified proteins were estimated by comparison with the corresponding wild-type
proteins, which had been prepared in vivo and purified using the same protocols. The
concentrations of the reference proteins were determined based on OD280 determination and the
extinction coefficient for the protein. The wild-type protein of known concentration was then
used as a reference standard at several concentrations in an SDS-PAGE experiment visualized by
Coomassie staining in comparison with the modified proteins.
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
Protein Fluorescence Measurements. The fluorescence spectra of modified GFPs
containing compounds 1a, 3a, 4a and 8a at positions 66, as well as compound 1a at position 66
and Acd at position 39, were measured using a Varian Cary Eclipse Fluorescence
Spectrophotometer with the excitation slit set at 10 nm and emission slit at 10 nm in 25 mM Tris-
HCl, pH 7.4, containing 0.5 M NaCl.
The emission spectra of modified GFPs containing compounds 1a, 3a, 4a and 8a in position 66
were measured after excitation at appropriate wavelengths (300 nm for 1a and 8a; 320 nm for
3a; 360 nm for 4a). The emission spectra of the GFP analogue having Acd at position 39 and
compound 1a at position 66 were measured after excitation at two wavelengths (296 and 370
nm), corresponding to excitation maxima for compounds 1a and Acd, respectively.
RESULTS
Synthesis of the Fluorescent Dipeptidomimetic Analogues and their pdCpA
Esters. The synthesis of the aminoacylated pdCpA derivative of dipeptidomimetic analogue 5a
(Figure 1) began with the conversion of commercially available 4-(1H-pyrrol-1-yl)benzoic acid
13
ACS Paragon Plus Environment

Biochemistry
Page 14 of 48
1
2
3
4
5
6
7
8
9
(13) to activated intermediate 14 by treatment with isobutylchloroformate and triethylamine
(Scheme 1). This anhydride was condensed with the monoanion of the diphenylmethylene-
glycine methyl ester (15), followed by imine hydrolysis using ~6 N HCl to afford amine salt 16,
which was used in the next step without further purification.20 Amine salt 16 was coupled with
Boc-Gly-OH using BOP21 as a coupling agent to form α-amido-β-ketoester 17 in 58% overall
yield from acid 13. Cyclodehydration of 17 using triphenylphosphine in presence of iodine and
triethylamine in CH2Cl2 afforded oxazole 18 in 63% yield.20 Subsequent removal of the Boc
group in 18 using CF3COOH, followed by treatment with pentenoyloxysuccinimide gave the
amide 5b in 72% yield.18 The methyl ester moiety of 5b was saponified using ~1 N LiOH
solution and the corresponding acid was activated as cyanomethyl ester 19 in 56% yield for two
steps.18,22 Treatment of cyanomethyl ester 19 with the tris(tetrabutylammonium) salt of pdCpA
(TBA-pdCpA) in anhydrous DMF gave the corresponding aminoacylated pdCpA ester 20 in
32% yield.15 Finally, activated pdCpA derivative 20 was ligated to an abbreviated suppressor
tRNACUA-COH transcript using T4 RNA ligase, providing the respective misacylated suppressor
tRNA transcript.
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
The syntheses of the aminoacylated pdCpA derivative of dipeptidomimetics 1a and 8a have
been reported previously.12,13 The synthesis of the pdCpA derivative of compound 2a is
described in Scheme S1 of the Supporting Information. The synthesis began with the
condensation of commercially available 3,4,5-trimethoxybenzoyl chloride with the monoanion of
the diphenylmethylene-glycine methyl ester (15), followed by imine hydrolysis using conc HCl
to afford the hydrochloride salt of amine 21. The latter was coupled directly with CBz-Gly-OSu
in presence of triethylamine to form the α-amido-β-ketoester 22 in 79% overall yield from imine
14
ACS Paragon Plus Environment

Page 15 of 48
Biochemistry
1
2
3
4
5
6
15. The remaining steps for the synthesis of the pdCpA derivative of compound 2a were similar
to those described in Scheme 1.
7
8
9
The synthesis of the aminoacylated pdCpA derivative of oxazole 3a (Scheme S2 of the
Supporting Information) and the corresponding thiazole 9a (Scheme S6 of the Supporting
Information) began from known glycine derivative 26.23 Thus, fully protected glycine 26 was
deprotonated using 1 M NaHMDS and condensed with 4-thiomethylbenzoyl chloride to obtain
ketoester 27. The Boc groups were removed from 27 using CF3COOH followed by coupling
with Boc-Gly-OH using the BOP reagent21 to obtain α-amido-β-ketoester 28 in 72% overall yield
from 26. Cyclodehydration of 28 using triphenylphosphine in presence of iodine and
triethylamine in CH2Cl2 gave oxazole 29 in 62% yield (Scheme S2 of the Supporting
Information).24 Treatment of 28 with the Lawesson's reagent gave the corresponding thiazole 47
in 87% yield (Scheme S6 of the Supporting Information).20 The remaining steps for the synthesis
of pdCpA derivatives of the oxazole dipeptidomimetic 3a and the corresponding thiazole
dipeptidomimetic 9a were similar to those described in Scheme 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
Similar reaction schemes were employed for the syntheses of the aminoacylated pdCpA
derivatives of the oxazole dipeptidomimetics 4a, 6a and 7a, and these are described in Schemes
S3, S4 and S5, respectively, of the Supporting Information. Likewise, the reaction schemes for
the synthesis of the aminoacylated pdCpA derivatives of the thiazole dipeptidomimetics 10a and
11a are shown in Schemes S7 and S8, respectively, of the Supporting Information.
Photophysical Properties of the Fluorescent Dipeptidomimetic Analogues. We
characterized the photophysical properties of the library of oxazole and thiazole based
dipeptidomimetic analogues, including molar absorptivities, quantum yields, emission and
absorption spectra of the corresponding N-pentenoyl methyl ester derivatives in methanol (Table
15
ACS Paragon Plus Environment

Biochemistry
Page 16 of 48
1
2
3
4
5
6
7
8
9
1; Figure S1). Ideally, the photophysical data should have been measured in aqueous buffer to
allow direct correlations with the same compounds present within a protein on attached to a
tRNA. However, this was not possible due to the very limited solubility of some of the
fluorophores in aqueous media. The absorbance maxima of the analogues ranged from 296 to
351 nm, while the emission maxima ranged from 365 to 490 nm. The photophysical properties of
1b and 8b have been described previously using samples having purity considered routine for
synthetic products, although the careful purification of the compounds by reversed phase HPLC
have now provided samples with reevaluated quantum yields.12,13 The UV absorption and
fluorescence emissions of 4b and 10b were the most red-shifted of the analogues studied. The
emission peak wavelength of 10b (490 nm) was closest to the emission peak wavelength of wild-
type GFP (509 nm) with a large Stokes shift (139 nm). Nonetheless, 10b proved to be a poor
fluorescent probe since it has very low quantum yield (0.04). We verified (UV, HPLC analysis)
that compound 10b was unaltered chemically as a consequence of its excitation. Compound 4b
has quite favorable properties as a fluorescent probe. It has a high quantum yield (0.60) and high
molar absorptivity (20000 M-1 cm-1), emits in the blue region (458 nm) of the spectrum and has a
large Stokes shift as compared with BFP (107 vs 75 nm). Compounds 3b and 9b were the best
oxazole_thiazole pair in terms of quantum yield; both exhibited very high quantum yields, 0.92
and 0.91, respectively. Interestingly, the other oxazole-thiazole pairs differed significantly from
each other in quantum yields; the thiazoles had significantly lower quantum yields compared to
the corresponding oxazoles. This likely reflects the electron donating and withdrawing properties
of the substituents by which these analogues differ. Apart from this, 3b and 9b emit in the visible
region (419 and 447 nm, respectively), have large Stokes shifts (100 nm and 134 nm,
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
respectively) and large molar absorptivities (21000 and 13100 M-1 cm-1, respectively) which
16
ACS Paragon Plus Environment

Page 17 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
make them efficient fluorescent probes. Although the absorption maximum of 2b was similar to
1b, the emission maximum was red-shifted by ~35 nm. The molar absorptivity of 2b (14300 M-1
cm-1) was roughly 80% that of 1b (17600 M-1 cm-1), such that the brightness of 2b was less than
that of 1b (10600 vs 15800 M-1 cm-1). In addition to 3b and 9b, which were quite bright (19300
and 12000 M-1 cm-1, respectively), analogue 4b also exhibited favorable brightness (12000 M-1
cm-1). Compounds 5b, 7b, 10b and 11b exhibited one or more photophysical properties that
diminish their brightness, and thus their potential utility as fluorescent probes.
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
Activation of Suppressor tRNACUA and Synthesis of Modified GFPs. The individual
N-pentenoyl protected aminoacylated pdCpA derivatives were ligated to an abbreviated
suppressor tRNACUA-COH using T4 RNA ligase. The N-pentenoyl groups were removed using 5
mM aqueous I2 at room temperature for 15 min. The free aminoacyl-tRNAs so obtained were
then utilized in a cell free coupled transcription‒translation system containing an E. coli S-30
preparation and pETGFP66 plasmid. The plasmid pETGFP66 contained a modified GFP gene
having a TAG codon at position 66. As shown in Figure 2 and Table 2, the dipeptidomimetic
analogues were all incorporated into GFP, but showed varying suppression efficiencies. In
replicate experiments, the lowest suppression yield was ~3%, observed for compound 5a, and the
highest suppression yield was ~12%, observed for compound 4a. Among the four
oxazole/thiazole pairs (1a/8a; 3a/9a; 4a/10a; 6a/11a), the best incorporation efficiency was
shown by pair 4a /10a. Compound 4a was incorporated in ~12% yield and compound 10a in
~10% yield. For the pairs 1a/8a, and 3a/9a, the suppression yields realized with the oxazoles
were reproducibly higher than those observed for the corresponding thiazoles. Compounds 6a
and 11a displayed roughly the same incorporation efficiencies, ~5% yields. Compounds 2a, 5a
and 7a were incorporated in ~6, ~3 and ~11% yields, respectively. Five of the analogues had
17
ACS Paragon Plus Environment

Biochemistry
Page 18 of 48
1
2
3
4
5
6
7
8
9
suppression yields ≥10%, which should enable quite reasonable quantities of proteins containing
these analogues to be prepared. It may be noted that oxazole 1a has previously been incorporated
into DHFR in suppression yields up to 15%.12
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
Fluorescent Properties of the Oxazoles and Thiazoles, as tRNA Esters and
Incorporated in Proteins and Peptides. GFP analogues having compounds 1a, 3a, 4a and
8a at position 66 were prepared on a larger scale and purified by Ni-NTA agarose
chromatography. The fluorescence emission spectra of these modified GFPs were compared after
excitation of each at the absorption maxima of the corresponding free fluorophores. The
fluorescence emission maxima of proteins containing 1a, 3a, 4a and 8a were observed at shorter
wavelengths (377, 410, 445 and 406 nm, respectively) than the emission maxima observed for
the individual dipeptide analogues (cf Tables 1 and 3). It must be noted, however, that the
monomeric fluorophores were investigated in MeOH, while the protein and activated tRNAs
containing the same fluorophores were studied as aqueous solutions.
These emission wavelengths were characterized in more detail, by comparing the fluorescence
emission spectra of analogues 1a, 3a, 4a or 8a present either at position 66 of GFP, or at the 3′-
end of a suppressor tRNA. As shown in Figure 3, in which the concentrations of the activated
tRNAs and GFPs had been adjusted to afford emission spectra of roughly comparable intensities,
the emission maxima for the species containing 1a, 3a and 8a differed by ~30-35 nm. In
comparison, the emission spectrum of the activated tRNA and GFP containing 4a differed by
only 15 nm, and the emission spectrum of the activated tRNA was clearly narrower than that of
the modified GFP.
The effect of environment on the fluorescence emission intensity of dipeptidomimetic analogue
1a was studied by comparing the emission spectra of the free oxazole with those of the
18
ACS Paragon Plus Environment

Page 19 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
corresponding aminoacyl-tRNA, and the modified GFP containing 1a at position 66 (Figure 4).
As shown, 100 nM free oxazole 1a (green trace) and the same concentration of the suppressor
tRNA activated with oxazole 1a (orange trace) exhibited emission spectra of comparable
intensity, with the emission maximum of the spectrum due to the activated tRNA being at longer
wavelength (cf Table 3). In comparison, the modified GFP having 1a at position 66 emitted at
much shorter wavelength (377 nm) with several-fold greater intensity (Figure 4). The effect of
solvent on the fluorescence emission wavelength and intensity of oxazole 1a was also studied
(Figure S2), and it was found that much of the observed differences in emission wavelength was
due to the solvent. The slightly longer wavelength emission of the fluorophores attached to the
3′-end of tRNAs (e.g. Figure S2) presumably reflects a more polar environment induced by the
tRNA, which contains multiple charged and polar functional groups.
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
The source of the enhanced fluorescence emission intensity observed for the modified GFP
having 1a at position 66 was explored by embedding oxazole 1a in the tetrapeptide Gly-Gly-Gly-
Gly to determine whether the simple presence of the fluorophore within a peptide backbone
would influence its photophysical properties. The route used for the synthesis of the requisite
peptide 12 is shown in Scheme 2, and that for the preparation of the requisite photophysical
control (N-Boc 1a methyl ester, 63) is provided in Scheme S9. As noted in Table S1, the
photophysical properties of Boc-deprotected peptide 12 differed in interesting ways from those
of the analogously deprotected oxazole 63 (1a methyl ester). While both had the same absorption
maxima when measured in MeOH (300 nm), the emission maximum of Boc-deprotected 12 was
at shorter wavelength (381 vs 391 nm), analogous to the shifts observed when the oxazoles and
thiazoles were present at position 66 of GFP (cf Table 3). Further, while their quantum yields
were similar (0.88 and 0.92 for deprotected 12 and 63, respectively), compound 12 may have
19
ACS Paragon Plus Environment

Biochemistry
Page 20 of 48
1
2
3
4
5
6
7
8
9
exhibited somewhat greater molar absorptivity, and consequently greater brightness than 63
(Table S1), although these differences were not statistically significant. The emission spectra of
12 and 63 were also recorded in MeOH, and reflected the same trends (Figure S3). When the
same two Boc-deprotected compounds were studied in 25 mM Tris-HCl, pH 7.4, containing 0.5
M NaCl as had been done for the modified GFPs (Table 3), the overall differences in their
photophysical properties were more pronounced (Table 4). The molar absorptivities and quantum
yields decreased for both compounds, but the decrease in quantum yield was much greater for
deprotected oxazole 63. As a consequence deprotected 12 was more than three-fold brighter than
deprotected 63 (Table 4). Once again, these trends were reflected in the fluorescence emission
spectra of 63 and 12 themselves taken in the same buffer (Figure 5). Thus the several-fold
differences in fluorescence intensity noted in Figure 4 must be due in part to the presence of the
fluorophore as an intrinsic structural constituent of the protein backbone (rather than as the usual
electronically isolated substituent attached to the peptide backbone), and in part to its additional
presence within the -barrel structure of GFP.
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
To explore the capability of the dipeptidomimetic analogues as FRET probes, we selected
compound 1a as the donor. The emission peak wavelength of 1a after incorporation into position
66 of GFP was at 377 nm (Table 3). As the fluorescence acceptor we chose acridon-2-ylalanine
(Acd) which has absorption peaks at 388 nm and 407 nm in water as a monomer, and emission
peaks at ~420 and 450 nm. As position 66 is within the β-barrel structure of GFP, it was chosen
for donor 1a whereas position 39, outside of the -barrel, was chosen for acceptor Acd. The
spectral overlap between the fluorescence emission of oxazole 1a and the absorption spectrum of
Acd is shown in Figure S4. The predicted R0 value for these two fluorophores was calculated25 as
32.8 Å, while the distance between the -carbon atoms of residues 39 and 66 is 17.0 Å based on
20
ACS Paragon Plus Environment

Page 21 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
X-ray crystallography (Figure 6B). By decoding the four base codon CGGG with the acridon-2-
ylalanyl-tRNACCCG and the nonsense codon UAG with the oxazolyl-tRNACUA, Acd and
compound 1a were incorporated into GFP at positions 39 and 66, respectively. The incorporation
yield was 5% compared to wild-type GFP. After purification, the doubly modified GFP was
excited at 296 nm. As demonstrated in Figure 6A, the modified GFP containing fluorescence
acceptor Acd at position 39 and compound 1a at position 66 displayed a FRET signal at ~450
nm. The theoretical maximum intensity of the FRET peak is defined by the orange curve in
Figure 6A, which shows the emission spectrum of Acd following its direct excitation by
irradiation at 370 nm. A control experiment involving a GFP having glycine at position 66 and
Acd at position 39 gave only weak emission at 450 nm when irradiated at 296 nm (Figure S5).
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
DISCUSSION
We have recently described the recognition by engineered bacterial ribosomes of suppressor
tRNAs activated with dipeptide analogues containing oxazole and thiazole cores, and the
incorporation of these heterocyclic species into dihydrofolate reductase and green fluorescent
protein.12,13 The incorporations have been verified by mass spectrometric analysis of tryptic
digests of the modified proteins. The strongly fluorescent nature of the derived proteins
prompted the preparation and study of a library of substituted oxazoles and thiazoles to enable a
better understanding of the nature and scope of our initial observations. The synthesis of 1,3-
oxazoles and 1,3-thiazoles has been the focus of many studies as a consequence of their
widespread occurrence in natural products. Nature biosynthesizes oxazoles through
cyclodehydration and oxidation of acylserine derivatives,26 and many synthetic routes for
oxazoles follow this general strategy.27,28 Another versatile method for the synthesis of 1,3-
21
ACS Paragon Plus Environment

Biochemistry
Page 22 of 48
1
2
3
4
5
6
7
8
9
oxazoles involves the cyclodehydration of α-acylaminoketones, esters or amides;29 the oxazole
synthesis described in this report falls in this category. The α-amido-β-ketoesters were cyclized
directly using a dehydrating mixture of two equivalent of triphenylphosphine, iodine and
triethylamine,24 and the reactions were typically complete within 15 minutes at room
temperature.
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
There are also number of methods for the synthesis of 1,3-thiazoles, the most common
involving the condensation of substituted -haloketones and thioamides. In this report, the
thiazoles were obtained by the cyclization of -amido--ketoesters using Lawesson’s reagent.30
This method was followed since the -amido--ketoester was a common intermediate for both
oxazole and thiazole syntheses, which simplified the overall process of library construction.
Aequorea victoria GFP chromophore is formed by a dehydrative cyclization, involving the Tyr
moiety at position 66 in the nascent polypeptide. The phenolic OH group of the Tyr moiety is
particularly important for the photophysical properties of GFP.31 There have been an extensive
structure‒activity studies of the GFP fluorophore. In addition, site-directed mutagenesis has
enabled the mutation of GFP chromophore by Phe analogues with different para-substituents in
place of Tyr66.32,33 These studies indicated that the wavelength maxima for both absorption and
emission peaks of GFP fluorophore increased with increasing electron-donating ability of these
substituents. In order to build a library of the dipeptidomimetic analogues with a wide variety of
photophysical properties, several oxazole and thiazole analogues containing either electron-
donating (NMe2, SMe, OMe) or electron-withdrawing (NO2, pyrrole, CN) substituents were
synthesized. The wavelength of emission peaks for both oxazoles and thiazoles were in the order
NMe2 > SMe > OMe ≈ NO2 ≈ pyrrole > CN, i.e. roughly in the order predicted by studies of the
GFP fluorophore. As the most electron-donating substituent the presence of NMe2 elicited the
22
ACS Paragon Plus Environment

Page 23 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
maximum red shift of the absorption and emission peaks. Compounds 4b and 10b both had
absorption maxima at 351 nm and had emission maxima at 458 and 490 nm in methanol,
respectively. Nitrile 6b exhibited the shortest wavelength absorption and emission peaks (285
and 365 nm, respectively). Compound 2b can be considered as a more electron dense congener
of compound 1b because it contains methoxyl groups at both meta-positions in addition to the
one at the para-position. It was anticipated correctly that compound 2b would exhibit a red shift
in absorption and emission peak wavelengths compared to compound 1b, although the shift was
not dramatic, presumably due to the placement of the additional OMe groups at the meta
positions. In all cases, the emission maxima of the thiazoles were ~20-30 nm red-shifted
compared to the corresponding oxazoles. The molar absorptivity values for the oxazoles were
also greater compared to the corresponding thiazoles. Similarly, the quantum yields for the
oxazoles were generally greater than for the corresponding thiazoles, with the exception of
compounds 3b and 9b, both of which exhibited very high quantum yields. In comparison, the
compounds having electron-withdrawing substituents (5b, 6b, 7b and 11b) displayed very low
quantum yields.
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
In order to study the basis for the observed photophysical properties of the oxazoles and
thiazoles, four compounds (1b, 3b, 4b and 8b) were used to prepare the respective activated
suppressor tRNAs and modified GFPs on a larger scale. As shown in Table 3 and Figure 4, when
these compounds were incorporated at the 3′-end of tRNAs as activated esters, the fluorescence
emission maxima were generally shifted to longer wavelength, albeit without any significant
increase in fluorescence intensity. In contrast, when the four compounds were incorporated into
position 66 of GFP the resulting GFP analogues had several-fold stronger fluorescent intensities
23
ACS Paragon Plus Environment

Biochemistry
Page 24 of 48
1
2
3
4
5
6
than the free compounds and the corresponding aminoacyl-tRNACUAs, and the fluorescence
emission occurred at shorter wavelengths in each case (Figures 3 and 4).
7
8
9
The oxazoles and thiazoles employed in the current study are unique as synthetic protein
fluorophores in that the conjugated systems which are responsible for their fluorescent properties
actually become part of the polypeptide/protein backbone. Accordingly, it seemed possible that
the effects noted in Table 3, and Figures 3 and 4, related to alterations of fluorescence intensity
and emission wavelength maxima, might be influenced by the simple incorporation of these
fluorophores into a polypeptide. As shown in Table 4 and Figure 5, this is clearly the case.
Embedding oxazole 1a within the tetrapeptide Gly-Gly-Gly-Gly caused its fluorescence emission
wavelength maximum to shift to shorter wavelength (from 407 to 399 nm), as also noted for the
oxazoles and thiazoles incorporated at position 66 of GFP (cf Table 3). The embedded oxazole
also exhibited about a three-fold greater brightness of its fluorescence signal as compared with
the free oxazole when measured in Tris-HCl buffer (Table 4). This increase in brightness was
associated with an increased quantum yield, plausibly due to suppression of non-radiative
deactivation pathways by the rigid peptide matrix. This observation parallels an earlier finding12
in which we noted that the incorporation of oxazole 1a at position 39 of GFP (outside the -
barrel structure) produced a stronger emission than that observed for free oxazole 1b. Inclusion
of the embedded oxazole at position 66 (which is within the -barrel structure, Figure 6B)
effected a further enhancement in fluorescence emission intensity. More specifically, we found
that the absorption extinction coefficient for GFP containing 1a at position 39 was 50,200 M-1
cm-1, while that for GFP itself was 25,000.12 The quantum yield for the GFP analogues
containing 1a at position 39 was 0.84 vs 0.79 for GFP itself. When 1a was present at position 66
of GFP, the absorption extinction coefficient was 90,300 M-1 cm-1 and the quantum yield was
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
24
ACS Paragon Plus Environment

Page 25 of 48
Biochemistry
1
2
3
4
5
6
7
8
9
0.91. Thus introducing 1a at position 39 of GFP produced a protein about two-fold brighter than
native GFP, while introducing this fluorophore at position 66 resulted in a protein about four-
fold brighter than wild-type GFP.12 This change was associated with a change in the absorption
extinction coefficient of 1b. Embedding oxazole 1a in GFP at position 66 enhanced its
brightness ~25 to 30-fold. It is worth noting that our novel GFP, having 1 at position 66, is also
two- to three-fold brighter than a “class 2” GFP34 having an absorption extinction coefficient
about twice that of GFP that we employed for modification. While oxazole 1a is roughly
isosteric with a dipeptide, and has been introduced into the -barrel structure in lieu of a single
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
amino acid (Tyr66), there appears to be ample room within the -barrel structure to
accommodate this larger molecular entity (Figure 6B). Finally, the much smaller Stokes shift
observed for oxazole 1a within the -barrel of GFP12 may reasonably be attributed to a
suppression of solvent relaxation. While the environmental sensitivity of these fluorophores
should provide the wherewithal for enhanced definition of the microenvironment of proteins and
proteins complexes, there will also be a corresponding increase in the complexity of
quantification as a consequence of the aforementioned environmental sensitivity.
We conclude that the significant observed changes in emission wavelength and intensity are
due both to incorporation of the fluorophore within the protein backbone, and also to the
environmental sensitivity of these fluorophores, and that this property may potentially be
exploited to utilize such compounds as reporters of their environment, and conformational
changes which alter that environment. It may be noted that the introduction of oxazole 1a into
position 39 of GFP, which is not within the -barrel structure, afforded a much more significant
increase in fluorescence intensity of the oxazole12 than did its present inclusion within a
tetraglycine peptide. Accordingly, it might be anticipated that alteration of peptide structure, e.g.
25
ACS Paragon Plus Environment

Biochemistry
Page 26 of 48
1
2
3
4
5
6
by the inclusion of aromatic or lipophilic amino acids, would also affect the brightness of the
embedded oxazole.
7
8
9
Fluorescent proteins have been used as probes of molecular assembly in FRET experiments,35
although their large size does impose constraints on the utility of this approach. The
environmental sensitivity of the oxazoles and thiazoles studied here, as well as their small size,
chemical stability and stronger fluorescence as compared with natural GFPs,36 suggests that they
may have superior potential as reporter groups. Accordingly, we prepared a modified GFP
having a strong oxazole fluorophore at position 66 within the -barrel structure, and a
fluorescence acceptor (Acd) whose absorption spectrum overlaps the emission spectrum of the
oxazole outside the -barrel at position 39. The R0 value of this potential FRET pair was
calculated25 as 32.8 Å, while the distance calculated from an X-ray crystal structure37 was 17.0 Å
(Figure 6B). As predicted, when the doubly modified GFP was irradiated at the excitation
maximum of the oxazole at position 66 (296 nm), a FRET peak was evident (Figure 6A).
Where spectral overlap exists, the R0 value for a given FRET pair increases in direct proportion
to the quantum yield of the donor fluorophore and molar absorptivity of the acceptor. Given the
very favorable molar absorptivities and quantum yields of many of the novel fluorophores
studied here, as well as their environmental sensitivity and ability to be incorporated into
proteins with reasonable suppression efficiencies, it is anticipated that they may find utility as
reporter groups in experiments involving protein conformation and binding interactions with
other (macro)molecules. Further, the activated tRNAs whose fluorescence emission properties
are summarized in Table 3 may plausibly be of utility in reporting on the environment of the
aminoacyl moiety during the partial reactions of protein biosynthesis. Presumably, such
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
26
ACS Paragon Plus Environment

Page 27 of 48
Biochemistry
1
2
3
4
5
6
interactions may be unique to protein biosynthesizing systems in which the ribosomes have been
modified to accommodate dipeptidyl-tRNAs as substrates.
7
8
9
It should be noted that the present experiments were all carried out in cell free systems, and
application of the present artificial fluorescent proteins in a cellular context would require
significant extensions of the current technology.
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
CONCLUSIONS
A library of several fluorescent oxazole- and thiazole-based dipeptidomimetic analogues has
been synthesized. The analogues are structurally similar to the GFP chromophore, but are stable
chemically and do not require activation to exhibit fluorescence. In order to encompass a variety
of photophysical properties, the compounds were designed with different electron-donating and
electron-withdrawing substituents at the position analogous to the p-OH group in the GFP
fluorophore. The compounds displayed reasonable efficiency of incorporation into position 66 of
GFP using modified ribosomes and misacylated suppressor tRNACUAs. In addition, four mutant
GFPs each containing a dipeptide analogue with favorable fluorescent properties and formed in
good suppression yield were prepared in larger quantities and were characterized in more detail.
All of the modified GFPs exhibited several-fold higher fluorescence intensities compared to the
corresponding aminoacyl tRNACUAs, and were more strongly fluorescent than GFP itself. The
fluorescent properties (emission wavelength and intensity) were shown to be responsive to
environment, both within the protein backbone and especially within a secondary structure such
as a -barrel. Embedding the oxazole within a simple tetrapeptide also enhanced the intensity of
its fluorescence emission. A successful FRET experiment demonstrated the potential utility of
the fluorophores as reporter groups to monitor protein conformational changes and molecular
interactions.
27
ACS Paragon Plus Environment

Biochemistry
Page 28 of 48
1
2
3
4
ASSOCIATED CONTENT
5
6
7
8
9
Supporting Information
The Supporting Information is available free of charge on the ACS Publications website at DOI
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
AUTHOR INFORMATION
Corresponding Author
* Email: sidney.hecht@asu.edu. Phone: (480) 965-6625. Fax: (480) 965-0038.
Funding
This study was supported by Research Grant GM 103861 from the National Institute of General
Medical Sciences, National Institutes of Health.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS
APCI, atmospheric pressure chemical ionization; ESI, electrospray ionization; GFP, green
fluorescent protein; FP, fluorescent protein; HPLC, high-performance liquid chromatography;
TLC, thin layer chromatography; DMSO, dimethylsulfoxide; Rf, ratio of fronts; Hz, Hertz; THF,
tetrahydrofuran; DMF, dimethylformamide; TBA, tetrabutylammonium; BOP, N-
[oxytris(dimethylamino)phosphonium]benzotriazole hexafluorophosphate; EGTA, ethylene
glycol tetraacetic acid; PEG, polyethylene glycol; SDS-PAGE, sodium dodecyl sulfate‒
polyacrylamide gel electrophoresis; Ni-NTA, nickel nitrilotriacetic acid; DIPEA,
diisopropylethylamine; UV, ultraviolet.
28
ACS Paragon Plus Environment

Page 29 of 48
Biochemistry
1
2
3
4
REFERENCES
5
6
7
8
(1) Chen, I., and Ting, A. Y. (2005) Site-specific labeling of proteins with small molecules in
live cells. Curr. Opin. Biotechnol. 16, 35‒40.
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
(2) Sahoo, H. (2012) Fluorescent labeling techniques in biomolecules: a flashback. RSC Adv. 2,
7017‒7029.
(3) Chen, S., Fahmi, N. E., Wang, L., Bhattacharya, C., Benkovic, S. J., and Hecht, S. M. (2013)
Detection of dihydrofolate reductase conformational change by FRET using two fluorescent
amino acids. J. Am. Chem. Soc. 135, 12924‒12927.
(4) Heim, R., Prasher, D. C., and Tsien, R. Y. (1994) Wavelength mutations and
posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 91,
12501‒12504.
(5) Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994) Green
fluorescent protein as a marker for gene expression. Science 263, 802‒805.
(6) Walker, C. L., Lukyanov, K. A., Yampolsky, I. V., Mishin, A. S., Bommarius, A. S., Duraj-
Thatte, A. M., Azizi, B., Tolbert, L. M., and Solntsev, K. M. (2015) Fluorescence imaging using
synthetic GFP chromophores. Curr. Opin. Chem. Biol. 27, 64‒74.
(7) Marks, K. M., and Nolan, G. P. (2006) Chemical labeling strategies for cell biology. Nature
Methods 3, 591‒596.
(8) Dedkova, L. M., Fahmi, N. E., Golovine, S. Y., and Hecht, S. M. (2003) Enhanced D-amino
acid incorporation into protein by modified ribosomes. J. Am. Chem. Soc. 125, 6616‒6617.
(9) Dedkova, L. M., Fahmi, N. E., Golovine, S. Y., and Hecht, S. M. (2006) Construction of
modified ribosomes for incorporation of D-amino acids into proteins. Biochemistry 45, 15541‒
15551.
29
ACS Paragon Plus Environment