optimized synthesis routes and biological ...doktori.bibl.u-szeged.hu/10082/1/dissertation...

OPTIMIZED SYNTHESIS ROUTES AND BIOLOGICAL APPLICATION

OF N-PEPTIDE-6-AMINO-D-LUCIFERIN CONJUGATES

PhD Thesis

Anita Kármen Kovács

Supervisors

Prof. Dr. Gábor Tóth

Department of Medical Chemistry

Dr. László G. Puskás

Avidin Ltd

Department of Medical Chemistry

University of Szeged

2019

TABLE OF CONTENTS

ABBREVIATIONS AND SYMBOLS i

PUBLICATIONS iv

1. INTRODUCTION AND AIMS 1

2. LITERATURE BACKGROUND 3

2.1 6-amino-2-cyanobenzothiazole synthesis 4

2.1.1 6-amino-2-cyanobenzothiazole synthesis with

6-amino-2-chlorobenzothiazole as immediate precursor 5

2.1.2 6-amino-2-cyanobenzothiazole synthesis with

2-cyano-6-nitrobenzothiazole as immediate precursor 8

2.2 N-peptide-aLuc conjugate synthesis 16

2.2.1 N-peptide-aLuc conjugate synthesis methods with carboxyl protected

aLuc as starting material 16

2.2.1.1 Phospho-azo method with carboxyl protected aLuc as

starting material 16

2.2.1.2 Mixed anhydride method with carboxyl protected aLuc as

starting material 17

2.2.2 N-peptide-aLuc conjugate synthesis methods with

6-amino-2-cyanobenzothiazole as starting material 18

2.2.2.1 Phospho-azo method with 6-amino-2-cyanobenzothiazole

as starting material 18

2.2.2.2 Mixed anhydride methods with 6-amino-2-cyanobenzothiazole

as starting material 19

2.2.2.3 DCC coupling method with protected amino acid

6-amino-2-cyanobenzothiazole as starting material 22

3. RESULTS AND DISCUSSION 23

3.1 6-amino-2-cyanobenzothiazole synthesis 23

3.1.1 2-chloro-6-nitrobenzothiazole synthesis 23

3.1.2 6-amino-2-chlorobenzothiazole synthesis 24

3.1.3 6-amino-2-cyanobenzothiazole synthesis 25

3.2 N-peptide-aLuc conjugate synthesis routes and biological applications 26

3.2.1 N-peptide aLuc conjugate synthesis with liquid/solid phase Fmoc strategy 26

3.2.1.1 N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole synthesis 26

3.2.1.2 N-Fmoc-Asp(OtBu)-aLuc synthesis 28

3.2.1.3 Attachment of N-Fmoc-Asp(OtBu)-aLuc to solid support 29

3.2.1.4 Building the peptide chain 29

3.2.1.5 Cleavage of the peptide from the resin 29

3.2.1.6 Limitations of the liquid/solid phase method 30

3.2.2 N-peptide aLuc conjugate synthesis with fragment condensation strategy 32

3.2.2.1 N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole synthesis 33

3.2.2.2 N-Fmoc-Gly-Pro-aLuc synthesis 33

3.2.2.3 Limitations of the fragment condensation method 34

3.2.3 Building block production for the Boc-strategy solid phase synthesis of

N-peptide-aLuc conjugates 34

3.2.3.1 Unsuccessful attempts 34

3.2.3.1.1 Unsuccessful Boc-anhydride attempt to produce N-Boc-aLuc 34

3.2.3.1.2 Unsuccessful Boc-anhydride attempt to produce

N-Boc-6-amino-2-cyanobenzothiazole 35

3.2.3.1.3 Unsuccessful triphosgene attempt to produce

N-Boc-6-amino-2-cyanobenzothiazole 36

3.2.3.2 Successful synthesis route 37

3.2.3.2.1 N-Boc-6-amino-2-cyanobenzothiazole synthesis 37

3.2.3.2.2 N-Boc-aLuc synthesis 38

3.2.4 Biological testing 39

3.2.4.1 Biological testing of Z-Asp-Glu-Val-Asp-aLuc 39

3.2.4.1.1 Biochemical and cellular testing 39

3.2.4.1.2 In vivo testing 40

3.2.4.2 Biological testing of N-Fmoc-Gly-Pro-aLuc 41

4. SUMMARY 43

5. REFERENCES 46

6. ACKNOWLEDGEMENTS 52

APPENDIX

I. TABLES 53

I.1 Literary overview of 6-amino-2-cyanobenzothiazole synthesis routes 53

I.2. Methods for the synthesis of 6-amino-2-cyanobenzothiazole 54

I.3 Comparison of starting materials 58

I.4 Comparison of solvents during chlorine-cyanide exchange 59

I.5 Comparison of the optimal and the non-optimal order of reduction

and cyanidation 60

II. FIGURES 61

II.1. 1H-NMR spectrum of 2-chloro-6-nitrobenzothiazole 61

II.2. TSQ mass spectrum of 2-chloro-6-nitrobenzothiazole 62

II.3. RP-HPLC profile of the crude 2-chloro-6-nitrobenzothiazole 63

II.4 1H-NMR spectrum of 6-amino-2-chlorobenzothiazole 64

II.5 13C-NMR spectrum of 6-amino-2-chlorobenzothiazole 65

II.6 ESI mass spectrum of 6-amino-2-chlorobenzothiazole 66

II.7 RP-HPLC profile of 6-amino-2-chlorobenzothiazole 67

II.8 1H-NMR of 6-amino-2-cyanobenzothiazole 68

II.9 13C-NMR spectrum of 6-amino-2-cyanobenzothiazole 69

II.10 ESI mass spectrum of 6-amino-2-cyanobenzothiazole 70

II.11 RP-HPLC profile of 6-amino-2-cyanobenzothiazole 71

II.12 TOF mass spectrum of N-Fmoc-Asp(OtBu)-6-amio-2-cyanobenzothiazole 72

II.13 RP-HPLC of the purified N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole 73

II.14 TOF mass spectrum of N-Fmoc-Asp(OtBu)-aLuc 74

II.15 RP-HPLC profile of the N-Fmoc-Asp(OtBu)-aLuc 75

II.16. 1H-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc 76

II.17 13C-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc 79

II.18 ESI mass spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc 81

II.19 RP-HPLC profile of the purified N-Z-Asp-Glu-Val-Asp-aLuc 82

II.20 RP-HPLC profile of the resulting material at the determination of load 83

II.21 ESI mass spectrum of N-Fmoc-Pro-6-aminodehydroluciferin 84

II.22 RP-HPLC profile of the N-Z-Gly-Pro-aminodehydroluciferin 85

II.23 ESI mass spectrum of the N-Z-Gly-Pro-6-aminodehydroluciferin 86

II.24 1H-NMR spectrum of N-Z-Gly-Pro-6-aminodehydroluciferin 87

II.25 ESI mass spectrum of the N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole 88

II.26 RP-HPLC profile of the N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole 89

II.27 1H-NMR of the N-Fmoc-Gly-Pro-aLuc 90

II.28 13C-NMR of the N-Fmoc-Gly-Pro-aLuc 91

II.29 ESI mass spectrum of the N-Fmoc-Gly-Pro-aLuc 92

II.30 RP-HPLC profile of the N-Fmoc-Gly-Pro-aLuc 93

II.31 ESI mass spectrum of the N-Boc-6-amino-2-cyanobenzothiazole 94

II.32 RP-HPLC profile of the N-Boc-6-amino-2-cyanobenzothiazole 95

II.33 1H-NMR spectrum of the N-Boc-aLuc 96

II.34 13C-NMR spectrum of the N-Boc-aLuc 97

II.35 ESI mass spectrum of the N-Boc-aLuc 98

II.36 Chiral HPLC profile of enantiomeric mixture of the N-Boc-6-amino-luciferin 99

II.37 Chiral HPLC profile of the untreated N-Boc-aLuc 100

II.38 The standard error of the mean (SEM) values of Figure 7A 101

II.39 Ac-915 induces the activation of caspase-3 in U87 cells (A) 102

III. MATERIALS AND METHODS 103

III.1 Materials 103

III.2 Chemical Methods 103

III.2.1 2-chloro-6-nitrobenzothiazole synthesis 103

III.2.2 6-amino-2-chlorobenzothiazole synthesis 104

III.2.3 6-amino-2-cyanobenzothiazole synthesis 104

III.2.4 N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole synthesis 105

III.2.5 N-Fmoc-Asp(OtBu)-aLuc synthesis 106

III.2.6 Attachment of N-Fmoc-Asp(OtBu)-aLuc to solid support 106

III.2.7 N-Z-Asp-Glu-Val-Asp-aLuc synthesis 107

III.2.7.1 Fmoc deprotection 107

III.2.7.2 SPPS peptide coupling 107

III.2.7.3 Cleavage of peptide from the resin 108

III.2.7.4 Purification of crude peptide 108

III.2.8 N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole synthesis 108

III.2.9 N-Fmoc-Gly-Pro-aLuc synthesis 109

III.2.10 Purification of crude N-Fmoc-Gly-Pro-aLuc 110

III.2.11 N-Boc-6-amino-2-cyanobenzothiazole synthesis 110

III.2.12 N-Boc-aLuc synthesis 111

III.2.13 Purification of N-Boc-aLuc 111

III.3. Analytical methods 112

IV. BIOLOGICAL INVESTIGATION 114

IV.1 N-Z-Asp-Glu-Val-Asp-aLuc biochemical assay 114

IV.2 N-Z-Asp-Glu-Val-Asp-aLuc cellular assay 114

IV.3 N-Fmoc-Gly-Pro-aLuc assay 115

IV.4 N-Z-Asp-Glu-Val-Asp-aLuc in vivo assay 115

IV.5 Statistics 116

V. RELATED ARTICLES

ABBREVIATIONS AND SYMBOLS

AcN acetonitrile

AcOH acetic acid

Ala alanine

aLuc 6-amino-D-luciferin

anh anhydrous

aq aquous

Asp aspartic acid

APCI atmospheric pressure chemical ionization

API atmospheric pressure ionization

Boc tert-butyloxycarbonyl

(Boc)2O di-tert-butyl dicarbonate

BSA bovine serum albumin

cc concentrated

CHAPS 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate

COMU 1-cyano-2-ethoxy-2-oxoethylidenaminooxy)dimethylamino-

morpholino-carbenium hexafluorophosphate

cps counts per seconds

DABCO 1,4-diazabicyclo[2.2.2]octane

DCC dicyclohexylcarbodiimide

DCM dichloromethane

Deoxo-Fluor Reagent bis(2-methoxyethyl)aminosulfur trifluoride

DIPEA N,N-diisopropylethylamine

i

DMAA N,N-dimethylacetamide

DMAP 4-dimethylaminopyridine

DMEM Dulbecco's modified eagle medium

DMEM-F12 Dulbecco's modified eagle medium nutrient mixture F-12

DMF N,N-dimethylformamide

DMSO dimethyl sulfoxide

DTT 1,4-dithiothreitol

EDTA disodium ethylenediaminetetraacetate dehydrate

equiv equivalent

EtOAc ethyl acetate

EtOH ethanol

FACS fluorescence activated cell sorter

FAP fibroblast activation protein alpha

FCS fetal calf serum

fmk fluoromethylketone

Fmoc 9-fluorenylmethoxycarbonyl

Glu glutamic acid

Gly glycine

HATU 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-

b]pyridinium 3-oxid hexafluorophosphate

HEPES 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid

HMPA hexamethylphosphoric acid triamide

HOBt 1-hydroxybenzotriazole

IBCF chloroformic acid isobutylester

ii

i.p. intraperitoneally

Me2O acetone

MeOH methanol

mp melting point

PBS phosphate-buffered saline

POP/PREP prolyl oligopeptidase

Pro proline

RP-HPLC reversed-phase high-performance liquid chromatography

SEM standard error of the mean

sicc siccum

SPPS solid phase peptide synthesis

tBu tert butyl

TCFH chloro-N,N,N′,N′-tetramethylformamidinium hexafluorophosphate

TEA triethylamine

TFA trifluoroacetic acid

TFFH fluoro-N,N,N′,N′-tetramethylformamidinium hexafluorophosphate

THF tetrahydrofuran

TLC thin layer chromatography

Tris tris(hydroxymethyl)aminomethane

Val valine

Z benzyloxycarbonyl

iii

PUBLICATIONS

Papers related to the thesis

[1] Anita K. Kovács, Péter Hegyes, Gábor J. Szebeni, Lajos I. Nagy, László G. Puskás, Gábor K.

Tóth:

Synthesis of N-peptide-6-amino-D-luciferin conjugates

Front. Chem. 2018, 6 (120), 1-11., doi: 10.3389/fchem.2018.00120 IF: 4.155

[2] Anita K. Kovács, Péter Hegyes, Gábor J. Szebeni, Krisztián Bogár, László G. Puskás, Gábor

K. Tóth:

Synthesis of N-peptide-6-amino-D-luciferin conjugates with optimized fragment condensation

strategy

Int J Pept Res Ther 2018, (ahead of print), doi: 10.1007/s10989-018-9768-8 IF: 1.132

Other publications

[3] Anikó Angyal, András Demjén, Edit Wéber, Anita K. Kovács, János Wölfling, László G.

Puskás, Iván Kanizsai

Lewis acid-catalyzed diastereoselective synthesis of multisubstituted N-acylaziridine-2-

carboxamides from 2H-azirines via Joullié-Ugi three-component reaction

J. Org. Chem. 2018, 83 (7), 3570–3581., doi: 10.1021/acs.joc.7b03189 IF: 4.805

[4] Gyula Telegdy, Anita K. Kovács, Kinga Rákosi, Márta Zarándi, Gábor K. Tóth

Antiamnesic properties of analogs and mimetics of the tripeptide human urocortin 3

Amino Acids 2016,48 (9), 2261-2266., doi: 10.1007/s00726-016-2268-2 IF: 3.173

iv

http://www.frontiersin.org/people/u/493211




https://scifinder.cas.org/scifinder/references/answers/28D20A3FX86F350ACX11EA1F6F1DA43529A6:28D363A7X86F350ACX526EC89140EDF18FE4/9.html?nav=eNpb85aBtYSBMbGEQcXIwsXYzNjRPMLCzM3Y1MDROcLUyMzV2cLS0MTA1cXN0MLN1QSoNKm4iEEwK7EsUS8nMS9dzzOvJDU9tUjo0YIl3xvbLZgYGD0ZWMsSc0pTK4oYBBDq_Epzk1KL2tZMleWe8qCbiYGhooCBgYEFaGBGCYO0Y2iIh39QvKdfmKtfCJDh5x_vHuQfGuDp517CwJmZW5BfVAI0obiQoY6BGaiPASianVsQlFqIIgoAUEU7cA&key=caplus_2018:364294&title=TGV3aXMgQWNpZC1DYXRhbHl6ZWQgRGlhc3RlcmVvc2VsZWN0aXZlIFN5bnRoZXNpcyBvZiBNdWx0aXN1YnN0aXR1dGVkIE4tQWN5bGF6aXJpZGluZS0yLWNhcmJveGFtaWRlcyBmcm9tIDJILUF6aXJpbmVzIHZpYSBKb3VsbGnCtGUtVWdpIFRocmVlLUNvbXBvbmVudCBSZWFjdGlvbg&launchSrc=reflist&pageNum=1&sortKey=ACCESSION_NUMBER&sortOrder=DESCENDING

https://scifinder.cas.org/scifinder/references/answers/28D20A3FX86F350ACX11EA1F6F1DA43529A6:28D363A7X86F350ACX526EC89140EDF18FE4/9.html?nav=eNpb85aBtYSBMbGEQcXIwsXYzNjRPMLCzM3Y1MDROcLUyMzV2cLS0MTA1cXN0MLN1QSoNKm4iEEwK7EsUS8nMS9dzzOvJDU9tUjo0YIl3xvbLZgYGD0ZWMsSc0pTK4oYBBDq_Epzk1KL2tZMleWe8qCbiYGhooCBgYEFaGBGCYO0Y2iIh39QvKdfmKtfCJDh5x_vHuQfGuDp517CwJmZW5BfVAI0obiQoY6BGaiPASianVsQlFqIIgoAUEU7cA&key=caplus_2018:364294&title=TGV3aXMgQWNpZC1DYXRhbHl6ZWQgRGlhc3RlcmVvc2VsZWN0aXZlIFN5bnRoZXNpcyBvZiBNdWx0aXN1YnN0aXR1dGVkIE4tQWN5bGF6aXJpZGluZS0yLWNhcmJveGFtaWRlcyBmcm9tIDJILUF6aXJpbmVzIHZpYSBKb3VsbGnCtGUtVWdpIFRocmVlLUNvbXBvbmVudCBSZWFjdGlvbg&launchSrc=reflist&pageNum=1&sortKey=ACCESSION_NUMBER&sortOrder=DESCENDING

https://scifinder.cas.org/scifinder/references/answers/28D20A3FX86F350ACX11EA1F6F1DA43529A6:28D363A7X86F350ACX526EC89140EDF18FE4/16.html?nav=eNpb85aBtYSBMbGEQcXIwsXYzNjRPMLCzM3Y1MDROcLUyMzV2cLS0MTA1cXN0MLN1QSoNKm4iEEwK7EsUS8nMS9dzzOvJDU9tUjo0YIl3xvbLZgYGD0ZWMsSc0pTK4oYBBDq_Epzk1KL2tZMleWe8qCbiYGhooCBgYEFaGBGCYO0Y2iIh39QvKdfmKtfCJDh5x_vHuQfGuDp517CwJmZW5BfVAI0obiQoY6BGaiPASianVsQlFqIIgoAUEU7cA&key=caplus_2016:947677&title=QW50aWFtbmVzaWMgcHJvcGVydGllcyBvZiBhbmFsb2dzIGFuZCBtaW1ldGljcyBvZiB0aGUgdHJpcGVwdGlkZSBodW1hbiB1cm9jb3J0aW4gMw&launchSrc=reflist&pageNum=1&sortKey=ACCESSION_NUMBER&sortOrder=DESCENDING

Selected scientific lectures

Anita K. Kovács, Péter Hegyes, Gábor Szebeni, László G. Puskás, Gábor K. Tóth

Synthesis of peptide-6-amino-D-luciferin conjugates

18th International Symposium on Bioluminescence and Chemiluminescence

Uppsala, Sweden, June 23-28, 2014, oral presentation

Anita K. Kovács, Péter Hegyes, Gábor J. Szebeni, László G. Puskás, Gábor K. Tóth

Boc strategy for the synthesis of peptide-6-amino-D-luciferin conjugates

19th International Symposium on Bioluminescence and Chemiluminescence

Tsukuba, Japan, 29 May – 02 June, 2016, oral presentation


Synthesis methods of peptide-6-amino-D-luciferin conjugates for protease activity detection

8th Conference Chemistry towards Biology

Brno, Czech Republic, August 28 – September 01, 2016, oral presentation


Comparison of Fmoc-, Boc- and fragment condensation strategies in the synthesis of peptide-6-

amino-D-luciferin conjugates

34th European Peptide Symposium

Leipzig, Germany, September 04-09, 2016, poster presentation


Comparison of Fmoc-, Boc- and fragment condensation strategies in the synthesis of peptide-6-

amino-D-luciferin conjugates

12th Australian Peptide Conference

Noosa Heads, Victoria, Australia, October 15-20, 2016, poster presentation

v

- 1 -

1. Introduction and aims

1. INTRODUCTION AND AIMS

In the recent years, numerous in vivo and in vitro analytical methods, based on fluorescence and

bioluminescence, have been developed for various biological objectives, including

immunoassays, gene expression assays, bioimaging, investigation of infectious diseases etc.1-7

Plate based, high-throughput viability assays addressing the detection of protease activity is in the

focus of intensive research.8 The advantage of bioluminescent systems over fluorescent ones lies

in their superior sensitivity and easy handling.9-12

In the bioluminescent methods, diverse sets of luciferases and their substrates, luciferins have

been applied in different cellular and animal models.4-5 The most ubiquitous enzyme-substrate

system in bioimaging is the American firefly (Photinus pyralis) luciferin-luciferase system.13-14

Substituting the 6-position hydroxyl group of native luciferin with an amino group, the resulting

6-amino-D-luciferin (2-(6-aminobenzo[d]thiazol-2-yl)-4,5-dihydrothiazole-4-carboxylic acid,

hereinafter: aLuc, Figure 1) can form an amide bond with a peptide, while retaining the transport

and bioluminescent properties of the original substrate, resulting in a good substrate for different

important proteases, which can be used for the determination of the enzymatic activity.15

Figure 1 6-amino-D-luciferin

These substrates can be used for measuring protease enzyme activity in the following way: the

protease enzyme to be measured recognizes the peptide part of the conjugate with the suitable

peptide sequence, then cleaves the amide bond between the peptide and the aLuc, thus aLuc is

released, which, in the presence of luciferase enzyme, emits light (Figure 2).

- 2 -

1. Introduction and aims

Figure 2 The operation of the bioluminescent system

The activity of the given protease enzyme can be determined from the amount of emitted light, as

the emitted light is directly proportional to the activity of the enzyme.16

The N-linked peptide-aLuc conjugates discussed in the dissertation can be substrates for

proteases such as metalloproteases, chymotrypsin-like, trypsin-like and caspase-like proteases.17

Caspases are potential drug targets due to their role as key mediators of apoptotic cell death.

Among them, caspase-3 is a frequently activated death protease, catalyzing the specific cleavage

of many key cellular proteins, it is widely used as a sensitive monitor of apoptosis induction for

general cytotoxicity screening. Fibroblast activation protein and prolyl oligopeptidase are

enzymes that are proposed therapeutic targets for major human diseases, including

neurodegenerative diseases and psychiatric disorders, inflammation and cancer.

Unfortunately, the synthesis methods of these conjugates published so far are complicated; and

the very few commercially available conjugates are very expensive. The aim of this dissertation

is to introduce novel routes for the scalable and economical synthesis of N-peptide-aLuc

conjugates.

The preparative work focuses on preparing a precursor (6-amino-2-cyanobenzothiazole, 3), two

conjugates (Z-Asp-Glu-Val-Asp-aLuc, 6, Fmoc-Gly-Pro-aLuc, 8) and an intermediate (Boc-

aLuc, 10).

The publications on which this thesis is based are given in square brackets, while other literature

references are given as superscripts.

- 3 -

2. Literature background

2. LITERATURE BACKGROUND

Despite their growing significance, in the past decades only a handful of publications have dealt

with the synthesis of peptide-aLuc conjugates and their precursors.9, 15, 20-24, 29-31

The logical method for the synthesis of these conjugates would start with the synthesis of aLuc.

This target compound has an aminobenzothiazole skeleton (Figure 3).

Figure 3 6-amino-D-luciferin

To produce an aLuc, D-cysteine needs to be added to this aminobenzothiazole skeleton, for which

step there are two possible substituents at position 2: a COOH group or a CN group. This means

that the two possible immediate precursors are 6-amino-2-benzothiazole carboxylic acid (6-

aminobenzo[d]thiazole-2-carboxylic acid) and 6-amino-2-cyanobenzothiazole (6-

aminobenzo[d]thiazole-2-carbonitrile), (Figure 4).

Figure 4 Possible immediate precursors of aLuc

What makes 6-amino-2-benzothiazole carboxylic acid less appealing as precursor is the fact that

its use requires a very complicated and expensive method: the amino group has to be protected,

then thiol- and carboxyl-protected D-cysteine has to be added, which step is followed by the

removal of the D-cysteine-protecting groups and the ring formation.18,19

- 4 -


Hence, most of the corresponding literature uses 6-amino-2-cyanobenzothiazole as immediate

precursor for the synthesis of aLuc. However, most of these publications are patents, and

unfortunately, they sometimes lack significant scientific data (e.g. concerning yields, protecting

groups).

2.1 6-amino-2-cyanobenzothiazole synthesis

So far very few methods have been published for the synthesis of the key molecule, 6-amino-2-

cyanobenzothiazole.15, 20-24 The key steps of all these methods are the nitration, the cyanidation and

the reduction; the methods differ in the starting material, the reagents, the solvents or the order of

transformations. However, taking a closer look at the different methods reveals that basically two

routes can be distinguished: one that uses 6-amino-2-chlorobenzothiazole as immediate precursor,

and one that uses 2-cyano-6-nitrobenzothiazole for that purpose (Figure 5, I.1).

Figure 5 Precursors of 6-amino-2-cyanobenzothiazole

- 5 -


2.1.1 6-amino-2-cyanobenzothiazole synthesis with 6-amino-2-chlorobenzothiazole as

immediate precursor

One of the possible immediate precursors of 6-amino-2-cyanobenzothiazole, 6-amino-2-

chlorobenzothiazole (2-chlorobenzo[d]thiazol-6-amine), was first synthetized by Drozdov and

Stavrovskaya.25 They reproduced Hofmann’s synthesis route 26 to produce 2-chloro-6-

nitrobenzothiazole (2-chloro-6-nitrobenzo[d]thiazole): 2-chloro-benzothiazole (2-

chlorobenzo[d]thiazole) was dissolved in cc H2SO4, the mixture was cooled down, then HNO3 was

added 27 (Scheme 1).

Scheme 1 Synthesis of 2-chloro-6-nitrobenzothiazole (Hofmann)

The resulting 2-chloro-6-nitrobenzothiazole was dissolved in EtOH upon heating, and acetic acid

and iron filings were added to the solution. Then, in the course of the reaction, 50 mL of water was

added. The reaction mixture was heated with stirring in a boiling-water bath for 1 hour. At the end

of the reduction, to prevent premature crystallization, EtOH was added, and the liquid was filtered.

After the addition of water to the filtrate, 6-amino-2-chlorobenzothiazole was obtained (Scheme

2).

Scheme 2 Synthesis of 6-amino-2-chlorobenzothiazole (Drozdov and Stavrovskaya)

- 6 -


Later Katz 28 applied certain modifications. The starting material, 2-chlorobenzothiazole, was first

nitrated with a mixture of KNO3 and cc H2SO4, keeping the temperature at 10-17 ºC (yield 80%).

The next step was the reduction of the nitro group. The material was added to the mixture of 95%

EtOH/AcOH/H2O/Fe powder, it was stirred at 80-85 ºC for 2 hours 45 minutes, thus preparing 6-

amino-2-chlorobenzothiazole (Scheme 3).

Scheme 3 Synthesis of6-amino-2-chlorobenzothiazole (Katz)

White et al 15 went on with the cyanidation of the substance: KCN was added to DMSO at 130-140

ºC, under continuous stirring overnight. Having cooled this mixture to 120 ºC, the 6-amino-2-

chlorobenzothiazole was added portionwise in an hour, then the mixture was stirred for 4.5 hours

at 120 ºC, thus producing the desired 6-amino-2-cyanobenzothiazole (Scheme 4).

Scheme 4 Synthesis of 6-amino-2-cyanobenzothiazole (White)

- 7 -


Takakura et al 20 used the same method for the nitration step as Katz, although with lower yield

(49%), then modified Katz’s reduction step: he used the same solvents but changed the reagents.

2-chloro-6-nitrobenzothiazole was dissolved in EtOH, water and HCl, then SnCl2·2H2O was added,

and the solution was refluxed at 120 ºC. When TLC showed complete consumption of the starting

material, the mixture was basified with aqueous NaOH solution. At this step Takakura’s yield was

significantly lower than Katz’s (61%). The final cyanidation step was the same as White’s,

although with a slightly lower yield (49%). (Scheme 5)

Scheme 5 Synthesis of 6-amino-2-cyanobenzothiazole (Takakura)

- 8 -


Hsu et al 21 used the Katz method, however, with longer reaction time and an extra reagent (FeCl3)

during the reduction, thus achieving a higher yield (88%) at this phase. The cyanidation step was

the replication of White’s method (Scheme 6)

Scheme 6 Synthesis of 6-amino-2-cyanobenzothiazole (Hsu)

2.1.2 6-amino-2-cyanobenzothiazole synthesis with 2-cyano-6-nitrobenzothiazole as immediate

precursor

The other possible immediate precursor of the target material is 2-cyano-6-nitrobenzothiazole (6-

nitrobenzo[d]thiazole-2-carbonitrile). Using this material, the last step in the synthesis of 6-amino-

2-cyanobenzothiazole is the reduction of the nitro group.

- 9 -


Gryshuk et al. 22 worked out four different synthesis routes.

During their first attempt, 6-nitrobenzothiazole (6-nitrobenzo[d]thiazole) was used as starting

material. It was suspended in water, then cc H2SO4 was added dropwise. In a different flask H2O2

was added to a solution of ethylpyruvate. The resultant oxyhydroperoxide solution and a solution

of FeSO4 was mixed and then added to the 6-nitrobenzothiazole mixture at 0 ºC. The mixture was

stirred for 30 minutes, then basified with NaHCO3. The resulting ethyl 6-nitrobenzothiazole-2-

carboxylate (ethyl 6-nitrobenzo[d]thiazole-2-carboxylate) was amidated: it was dissolved in

MeOH, then purged with NH3. During cyanidation, the 6-nitrobenzothiazole-2-carboxamide (6-

nitrobenzo[d]thiazole-2-carboxamide) was dissolved in anh pyridine at room temperature. It was

cooled down to 0 ºC and POCl3 was added dropwise. The mixture was stirred at room temperature

for 2 hours. The resultant 2-cyano-6-nitrobenzothiazole was finally reduced. Having dissolved it

in EtOH, SnCl2 was added and the mixture was stirred at 60 ºC for two hours. It was cooled down

to room temperature and the pH of the mixture was adjusted to 7, using NaHCO3 (Scheme 7).

Scheme 7 Synthesis of 6-amino-2-cyanobenzothiazole (Gryshuk 1)

- 10 -


During their second attempt, ethyl benzothiazole-2-carboxylate (ethyl benzo[d]thiazole-2-

carboxylate) was used as starting material. First it was nitrated: it was suspended in cc H2SO4 at 0

ºC. KNO3 was added portionwise at 10 ºC in 30 minutes. After another 30 minutes stirring the

mixture was poured over ice/water and filtered. The resulting ethyl-6-nitrobenzothiazole-2-

carboxylate was amidated with the strategy described above: it was dissolved in MeOH, then

purged with NH3. The following steps also agreed with those of the previous method: cyanidation

- solution in anh pyridine at room temperature, cooling down to 0 ºC and POCl3 addition; reduction

- solution in EtOH, SnCl2 addition and pH adjustment with NaHCO3 (Scheme 8).


- 11 -


At the third attempt, ethyl benzothiazole-2-carboxylate was used as starting material again and the

transformations were the same but in a different order: in this case the first step was amidation

(solution in MeOH and purgation with NH3), which was followed by nitration (suspension in cc

H2SO4 and addition of KNO3), cyanidation (solution in anh pyridine at room temperature, cooling

down to 0 ºC and POCl3 addition) and reduction (solution in EtOH, SnCl2 addition and pH

adjustment with NaHCO3) (Scheme 9).


- 12 -


At the fourth attempt, the previous starting material, ethyl benzothiazole-2-carboxylate was used,

with another different order of transformations: amidation (solution in MeOH and purgation with

NH3), cyanidation (solution in anh pyridine at room temperature, cooling down to 0 ºC and POCl3

addition), nitration (suspension in cc H2SO4 and addition of KNO3) and reduction (solution in

EtOH, SnCl2 addition and pH adjustment with NaHCO3) (Scheme 10).


- 13 -


McCutcheon et al. 23 developed two synthesis routes. The disadvantage of these methods is that

exotic and expensive starting materials and reagents are needed. In the first route, 4-nitroaniline

was treated with 4,5-dichloro-1,2,3-dithiazol-1-ium chloride in a mixture of AcN and THF (2:1).

Pyridine was added dropwise, then Na2S2O3, dissolved in water, was added. The mixture was

stirred for 3 hours at room temperature. The resulting (4-nitrophenyl)-carbonocyanidothioic amide

was suspended in anh DMF/DMSO (1:1) in the presence of CuI, TBAB, PdCl2, and stirred for 3

hours at 130 ºC. During the next step, reduction, the 2-cyano-6-nitrobenzothiazole was dissolved

in EtOH and NH4Cl was added. After 5 minutes stirring at room temperature, Zn powder was

added, and the reaction mixture was vigorously stirred for 30 minutes, obtaining the target material,

6-amino-2-cyanobenzothiazole (Scheme 11).

Scheme 11 Synthesis of 6-amino-2-cyanobenzothiazole (McCutcheon 1)

- 14 -


During the second route, McCutcheon et al.23 used 6-nitrobenzothiazole as starting material,

similarly to Gryshuk et al in their first attempt.23 However, instead of the acylation of the 6-

nitrobenzothiazole, the thiazole ring was decomposed by suspending in N2H4·H2O and EtOH. The

mixture was stirred for 12 hours at room temperature, then cooled down. The resulting 2-amino-5-

nitrobenzenethiol was suspended in anh DCM. 4,5-dichloro-1,2,3-dithiazol-1-ium chloride added

to the suspension. The solution was refluxed in 12 hours. During the final step, reduction, the earlier

method was applied: the 2-cyano-6-nitrobenzothiazole was dissolved in MeOH and NH4Cl was

added. After 5 minutes stirring at room temperature, Zn powder was added, and the reaction

mixture was vigorously stirred for 30 minutes, obtaining the target material, 6-amino-2-

cyanobenzothiazole (Scheme 12).

Scheme 12 Scheme 11 Synthesis of 6-amino-2-cyanobenzothiazole (McCutcheon 2)

- 15 -


During their first step, nitration, Hauser et al.24 used the Katz method 28 with significantly longer

reaction time, 18 hours. Instead of the reduction, however, their next step was the cyanidation of

2-chloro-6-nitrobenzothiazole. Aqueous solution of NaCN, then DABCO catalyst was added to 2-

chloro-6-nitrobenzothiazole dissolved in AcN; the mixture was stirred at room temperature for 24

hours. Aqueous FeCl3 solution was added in order to quench the excess cyanide. The nitro-group

of the resulting 2-cyano-6-nitrobenzothiazole was finally reduced with iron powder and acetic acid

at room temperature in 24 hours (Scheme 13).

Scheme 13 Scheme 11 Synthesis of 6-amino-2-cyanobenzothiazole (Hauser)

- 16 -


2.2 N-peptide-aLuc conjugate synthesis

The synthesis routes were first published by Geiger and Miska 29, who developed four methods

(phospho-azo and mixed anhydride methods, both with both 6-amino-2-cyanobenzothiazole and

carboxyl protected aLuc as starting material) and produced 13 different conjugates (Unfortunately,

from the patent it is not clear which molecule was synthesized which way). Having examined these

and the methods published later, two main pathways can be observed, based on the starting

materials.

2.2.1 N-peptide-aLuc conjugate synthesis methods with carboxyl protected aLuc as starting

material

2.2.1.1 Phospho-azo method with carboxyl protected aLuc as starting material

A carboxyl-protected aLuc (protecting group not specified) was dissolved in anh pyridine at -10

ºC. PCl3, then protected amino acid or peptide was added to the solution. After 20-hour stirring,

the protecting groups of both the peptide moiety and the carboxyl group were removed, thus

obtaining the desired peptide-aLuc conjugate 29 (yield not given, Scheme 14).

Scheme 14 Phospho-azo method with carboxyl protected aLuc as starting material to produce

N-peptide-aLuc conjugates (Geiger and Miska)

- 17 -


2.2.1.2 Mixed anhydride method with carboxyl protected aLuc as starting material

A protected peptide was dissolved in anh THF at -15 ºC. The addition of TEA and IBCF was

followed by the addition of the carboxyl-protected aLuc (protecting group not specified). After

overnight stirring, the protecting groups were removed, thus obtaining the desired peptide-aLuc

conjugate 29 (yield not given, Scheme 15).

Scheme 15 Mixed anhydride method with carboxyl protected aLuc as starting material to

produce N-peptide-aLuc conjugates (Geiger and Miska)

- 18 -


2.2.2 N-peptide-aLuc conjugate synthesis methods with 6-amino-2-cyanobenzothiazole as

starting material

2.2.2.1 Phospho-azo method with 6-amino-2-cyanobenzothiazole as starting material

6-amino-2-cyanobenzothiazole was dissolved in anh pyridine at -10 ºC. PCl3, then protected

peptide was added to the solution. After 20-hour stirring, D-cysteine was added at pH 7.5, the

protecting groups were removed, thus obtaining the desired peptide-aLuc conjugate 29 (yield not

given, Scheme 16).

Scheme 16 Phospho-azo method with 6-amino-2-cyanobenzothiazole as starting material to


- 19 -


2.2.2.2 Mixed anhydride methods with 6-amino-2-cyanobenzothiazole as starting material

A protected peptide was dissolved in anh THF at -15 ºC. The addition of TEA and IBCF was

followed by the addition of 6-amino-2-cyanobenzothiazole. After overnight stirring, D-cysteine

was added at pH 7.5, the protecting groups were removed, thus obtaining the desired peptide-aLuc

conjugate 29 (yield not given, Scheme 17).

Scheme 17 Mixed anhydride method with 6-amino-2-cyanobenzothiazole as starting material to


- 20 -


The method was modified in a 2003 Promega patent.30 A protected peptide was dissolved in anh

THF at -10 ºC. N-methylmorpholine and IBCF were added, which was followed by the addition of

6-amino-2-cyanobenzothiazole. After 2 days stirring, D-cysteine was added at pH lower than 7.

After overnight stirring, the protecting groups were removed (yield not given, Scheme 18).


produce N-peptide-aLuc conjugates (2003 Promega patent)

- 21 -


In a modified route by Gryshuk et al. in 2011 31, a protected amino acid was dissolved in anh THF

at 0 ºC. The addition of N-methylmorpholine and IBCF was followed by the addition of 6-amino-

2-cyanobenzothiazole. After overnight stirring, the protecting group was removed, and D-cysteine

was added at pH 8. Then a protected peptide was dissolved in anh THF. The addition of N-

methylmorpholine and IBCF was followed by the addition of the amino acid-6-amino-2-

cyanobenzothiazole conjugate. After 72 hours stirring, the protecting group was removed (yield

not given, Scheme 19).


produce N-peptide-aLuc conjugates (2011 Gryshuk patent)

- 22 -


2.2.2.3 DCC coupling method with protected amino acid 6-amino-2-cyanobenzothiazole as

starting material

In a different route 9, a protected amino acid-6-amino-2-cyanobenzothiazole conjugate (6-

Asp(OtBu)amino-2-cyanobenzothiazole) and a protected peptide (Z-Asp(OtBu)-Glu(OtBu)-Val-

COOH) was joined using DCC and HOBt. The resulting conjugate (Z-Asp(OtBu)-Glu(OtBu)-Val-

Asp(OtBu)amino-2-cyanobenzothiazole) was reacted with D-cysteine and the protecting groups

were removed, thus obtaining the desired peptide-aLuc conjugate (Z-Asp-Glu-Val-Asp-aLuc).

Although in the article it is not defined explicitly, it seems feasible that the synthesis route started

with 6-amino-2-cyanobenzothiazole (yield not given, Scheme 20).

Scheme 20 DCC coupling method with protected amino acid 6-amino-2-cyanobenzothiazole as

starting material (O’Brien)

- 23 -

3.Results and discussion

3. RESULTS AND DISCUSSION

3.1 6-amino-2-cyanobenzothiazole (3) synthesis [1]

Having examined the published methods 15, 20-24, it can be seen that they have disadvantages:

a, certain synthesis routes require too many reagents, some of which are expensive (I.2)

b, choosing a less optimal starting material may require extra transformations (I.3)

c, the use of an ill-chosen solvent leads to low yield during the chlorine-cyanide exchange (I.4)

d, a less optimal order of the transformations results in low yield (I.5)

e, if the cyano group is present during the nitro group reduction, there is a high risk of its

reduction into an aminomethyl group (I.5)

3.1.1 2-chloro-6-nitrobenzothiazole (1) synthesis

As starting material, cheap, commercially available 2-chlorobenzothiazole was used, which was

nitrated with a mixture of KNO3 and cc H2SO4, keeping the temperature first under 15ºC than at

room temperature 28 (Scheme 21).

Scheme 21 Synthesis of 2-chloro-6-nitrobenzothiazole (1)

The structure of the resulting 2-chloro-6-nitrobenzothiazole (1) was attested by 1H-NMR (II.1),

TSQ-MS (II.2) and RP-HPLC (II.3). Recipe: III.2.1

- 24 -


3.1.2 6-amino-2-chlorobenzothiazole synthesis (2)

The next step was the reduction of the nitro group. First the mixture of EtOH/AcOH/Fe powder

was used 28. Although the reaction was successful, the resulting by-product (Fe(III) acetate) was

difficult to dispose of, so this method was dismissed. When trying the reduction with

SnCl2/AcOH/cc HCl, large amounts of by-product formed, in which the chlorine was split off or

substituted with a hydroxyl group, so this method also had to be dismissed. The third possibility

was the application of Na2S2O5, which also turned out to be unsatisfactory due to the reduction

giving a very low yield (10%). Using Zn/HCl led to the same results as with SnCl2.

Finally, applying EtOAc/H2O/NH4Cl/Fe powder system solved the problem and the reduction

was successful with a good yield (90%) (Scheme 22).

Scheme 22 Synthesis of 6-amino-2-chlorobenzothiazole (2)

Using a Soxhlet extractor turned out to be a solvent-sparing, thus environmentally-friendly

method and processing the obtained product was also simple: the solution had to be decanted in

order to get rid of the Fe powder and then extracted. The structure of the resulting 6-amino-2-

chlorobenzothiazole (2) was attested by 1H-NMR (II.4), 13C-NMR (II.5), ESI-MS (II.6), RP-

HPLC (II.7). Recipe: III.2.2

- 25 -


3.1.3 6-amino-2-cyanobenzothiazole synthesis (3)

The next step - the chlorine/nitrile exchange in the 6-amino-2-chlorobenzothiazole (2) - is the key

in the production of the desired conjugates. Six different methods had to be tried, the common

features of which were the polar aprotic non-aqueous solvent, the high temperature and the long

reaction time: a) anh DMSO/KCN, 160°C, 10h b) anh DMSO/KCN/KI, 160°C, 10h c) anh

DMSO/18-crown-6/KCN, 120°C, 8h d) anh DMF/KCN, 140°C, 12h e) anh HMPA/KCN, 140°C,

10h f) anh DMAA/KCN/KI, 120°C, 8h.

The first five methods had to be dismissed due to the low yield (15-20%). When trying DMAA,

however, it turned out that KCN is dissolved best in this solvent, resulting in relatively high yield

(80%) (Scheme 23).

Scheme 23 Synthesis of 6-amino-2-cyanobenzothiazole (3)

This means that the success of the chlorine/nitrile exchange depends on the rate of KCN

dissolution. The structure of the resulting 6-amino-2-cyanobenzothiazole (3) was attested by 1H-

NMR (II.8), 13C-NMR (II.9), ESI-MS (II.10), RP-HPLC (II.11). Recipe: III.2.3

- 26 -


3.2 N-peptide-aLuc conjugate synthesis [1]

Ideally, in the next step the amino group of the 6-amino-2-cyanobenzothiazole (3) is blocked with

a protecting group. However, the low nucleophilicity of the amino group makes its protection

problematic, resulting in very low yield (3%). Our goal was to find a more reliable method.

3.2.1 N-peptide aLuc conjugate synthesis with liquid/solid phase Fmoc strategy

The desired aLuc conjugate (N-Z-Asp-Glu-Val-Asp-aLuc, 6) was reached in a 5-step route (I.6):

a) attachment of the C-terminal amino acid of the target sequence b) cysteine addition c)

attachment to resin d) solid-phase peptide synthesis e) cleavage from resin

3.2.1.1 N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole synthesis (4)

An Fmoc-protected amino acid (Fmoc-Asp(OtBu)-OH) was coupled to the 6-amino-2-

cyanobenzothiazole (3). As, due to the deactivated amino group, the amide bond could not be

formed with the usual coupling reagent, DCC, a much more activating coupling agent was

necessary. Different agents were tested in different quantities: COMU,20, 32-33 HATU,34 Deoxo-

Fluor Reagent,34 TFFH 35 and TCFH,36 all with a ratio of 1:1.5 and 1:3. The best yield (97%) was

obtained with 1.5 equivalents of TCFH (Scheme 24, Table 1).

Scheme 24 Synthesis of N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole (4)

- 27 -


Coupling agent Quantity Yield

COMU 1.5 equiv 0 %

COMU 3.0 equiv 0 %

HATU 1.5 equiv 7 %

HATU 3.0 equiv 8 %

Deoxo-fluor reagent 1.5 equiv 48 %

Deoxo-fluor reagent 3.0 equiv 38 %

TFFH 1.5 equiv 59 %

TFFH 3.0 equiv 51 %

TCFH 1.5 equiv 97 %

TCFH 3.0 equiv 72%

Table 1 Coupling agents and corresponding yields

Other than the quantity of the different coupling reagents, all other conditions (solvent, reaction

time, temperature etc.) were kept the same. Although it was not checked with chiral

chromatography at this point, it became obvious following the achiral chromatography after the

cysteine addition that there was no racemization, because if there had been, during either this or

the previous step, we would have seen diastereomers. As the achiral chromatography after the

cysteine addition was indispensable anyway, we could save the rather complicated chiral

chromatography one step earlier. The structure of the resulting Fmoc-Asp(OtBu)-6-amino-2-

cyanobenzothiazole (4) was attested by TOF-MS (II.12) and RP-HPLC (II.13). Recipe: III.2.4

- 28 -


3.2.1.2 N-Fmoc-Asp(OtBu)-aLuc synthesis (5)

During the addition of D-cysteine 37 to the Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole (4),

the amino acid-heterocycle conjugate was dissolved in THF and MeOH, then D-cysteine

hydrochloride monohydrate was added. The resulting compound was dissolved in water, and then

the cysteine was released from its salt with NaHCO3. During the reaction (about 25 minutes) the

pH of the solution was kept between 7.3-7.4 by the addition of NaHCO3 aqueous solution,

monitoring the process with a pH-meter and the addition was carried out under argon atmosphere.

By this way, the desired amino acid-aLuc conjugate, Fmoc-Asp(OtBu)-aLuc (5) was obtained

(Scheme 25).

Scheme 25 Synthesis of N-Fmoc-Asp(OtBu)-aLuc (5)

The structure of the resulting conjugate (5) was attested by TOF-MS (II.14) and RP-HPLC

(II.14). Recipe: III.2.5

- 29 -


3.2.1.3 Attachment of N-Fmoc-Asp(OtBu)-aLuc (5) to solid support

During the next step this conjugate was attached to resin. Two types of resins (Figure 6) were

tested: 2-chlorotrityl chloride and p-alkoxybenzyl alcohol (Wang resin). Loading was checked in

both cases: with 2-chlorotrityl chloride resin it was 30%, while with Wang resin it was 50%, so

we decided to use the latter.

Figure 6 The two tested resins

3.2.1.4 Building the peptide chain

Classical solid phase peptide synthesis was carried out: the peptide chain was built with Fmoc

strategy. However, the N-terminal amino acid was always Z-protected, as this protecting group

gives higher biological stability to the peptide (Scheme 26). Recipe: III.2.6

3.2.1.5 Cleavage of the peptide from the resin

The obtained peptide-aLuc conjugate was removed from the resin with the mixture of TFA/water

(95:5 v/v) (Scheme 26). Recipe: III.2.7

- 30 -


Scheme 26 Solid phase synthesis of N-Z-Asp-Glu-Val-Asp-aLuc (8)

Finally, the resulting material, Z-Asp-Glu-Val-Asp-aLuc (6), was purified by preparative RP-

HPLC (Recipe: III.2.8) and then its structure was attested by 1H-NMR (II.16 part 1-3), 13C-NMR

(II.17 part 1-2) spectra, ESI-MS (II.18) and RP-HPLC (II.19).

3.2.1.6 Limitations of the liquid/solid phase method [2]

Following the same route, we attempted to produce another conjugate, N-Z-Gly-Pro-aLuc, which

could be used for the measurement of the activity of such protease enzymes like FAP and POP.

Attaching the protected amino acid-aLuc conjugate (in this case N-Fmoc-Pro-aLuc) to solid

support, the loading was 50%. In order to reach higher load, instead of 3 hours, the coupling

reaction mixture was shaken for 6 hours. The determination of load showed that no improvement

- 31 -


was achieved in loading and, according to the resulting material’s RP-HPLC profile (II.20) and

the mass spectrum (II.21), 20% of the coupled material was dehydrogenated among the

conditions mentioned above. As we were planning to purify the final, completed product, the

synthetizing process was not abandoned. The product on the resin was Fmoc-deprotected, which

was followed by coupling Z-protected glycine to the proline, and the resulting conjugate was

cleaved from the resin.

The RP-HPLC profile of the product (II.22) showed a homogeneous material, but, according to

the mass spectrum (II.23), it was not the expected N-Z-Gly-Pro-aLuc, but a fully dehydrogenated

conjugate, N-Z-Gly-Pro-6-aminodehydroluciferin (N-Z-Gly-Pro-2-(6-aminobenzo[d]thiazol-2-

yl)thiazole-4-carboxylic acid). This was also attested by its 1H-NMR-spectrum (II.24), on which

the -proton of the 2-thiazoline ring was not detectable; however, an olefin proton was, as proof

of dehydrogenation. The driving force behind this dehydrogenation was the aromatization of the

2-thiazoline ring, which led to its transformation into a thiazole ring (Scheme 27).

Scheme 27 The failed synthesis route to N-Z-Gly-Pro-aLuc

- 32 -


Since this reaction did not occur earlier either, it might have been the result of the longer

exposure to the basic conditions.

As dehydroluciferin is a very efficient inhibitor of luciferase 38,39 it is not suitable for the above-

mentioned measurement of enzymatic activity.

3.2.2 N-peptide aLuc conjugate synthesis with fragment condensation strategy [2]

Our goal was to find a method that is more reliable than the hybrid liquid/solid phase synthesis

method, a synthesis route that prevents the side reactions mentioned above. The problem was

solved with returning to the fragment-condensation method, which is used to avoid problems

occurring during stepwise solid phase synthesis.40 Having made modifications to achieve better

results than the standard method, the desired peptide-luciferin conjugate (N-Fmoc-Gly-Pro-aLuc)

was reached in a 2-step route:

a) attachment of the target peptide sequence (N-Fmoc-Gly-Pro-OH) to 6-amino-2-

cyanobenzothiazole (3) b) cysteine addition (Scheme 28, I.6):

Scheme 28 Synthesis of N-Fmoc-Gly-Pro-aLuc (8)

The optimized synthesis route of the key molecule (3), [1] and the modifications of the two steps

[2] make a significant improvement over the standard methods.9, 29-31

- 33 -


3.2.2.1 N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole synthesis (7)

A suitably protected, commercially purchased peptide, N-Fmoc-Gly-Pro-OH, was coupled with

the key molecule, 6-amino-2-cyanobenzothiazole (3). (As during the synthesis only a dipeptide

was coupled to the 6-amino-2-cyanobenzothiazole (3), it was reasonable to purchase a ready

material, rather than synthesize and purify one. In case of longer peptides, solid phase peptide

synthesis can be used.) Due to the deactivated amino group of the 6-amino-2-cyanobenzothiazole

(3), the amide bond could not be formed with the usual coupling reagents; therefore, a more

powerful coupling agent 36 was necessary. Excellent conversion (97%) of the 6-amino-2-

cyanobenzothiazole (3) was obtained with 1.5 equivalents of TCFH. Obtained yield,

corresponding to the crude product: 68%. (Scheme 28, II.25, II.26). Recipe: III.2.9

With this process, we could avoid the extremely long coupling time of the standard mixed

anhydride method 30,31 and reached adequate yield (68%).

3.2.2.2 N-Fmoc-Gly-Pro-aLuc (8) synthesis

The peptide-heterocycle conjugate (N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole, 7) was

dissolved in MeOH, then D-cysteine hydrochloride monohydrate was added. The resulting

substance was dissolved in water and the cysteine was released from its salt with NaHCO3.

During the reaction (about 25 minutes) the pH of the solution was kept between 7.3-7.4 with the

addition of NaHCO3 aqueous solution, the process was continuously monitored with a pH-meter,

under argon atmosphere. The Fmoc-protection of the the N-terminal amino-group of the peptide

was kept up because it gave higher biological stability to the conjugate (Scheme 28). The

structure of the resulting conjugate was attested by 1H-NMR (II.27), 13C-NMR (II.28), ESI-MS

(II.29) spectra and RP-HPLC (II.30). Recipe: III.2.10

This method is also an improvement over the standard practice 9,29-31 as the window between pH

7.3-7.4 a) rules out the racemization of D-cysteine b) ensures the release of the cysteine from its

salt.

- 34 -


3.2.2.3 Limitations of the fragment condensation method

Although our optimized fragment condensation method was successful [2], it had certain

limitations: a) when attaching longer peptides (ones that contain more than ten amino acids),

solubility problems may occur, which may make coupling difficult 40, b) when not glycine or

proline is chosen as the C-terminal amino acid, racemization might occur. 40

3.2.3 Building block production for the Boc-strategy solid phase synthesis of N-peptide-aLuc

conjugates

To avoid the aforementioned difficulties, an old-fashioned method, the Boc-strategy solid phase

peptide synthesis had to be tried to produce N-peptide-aLuc conjugates. The cornerstone of Boc-

strategy N-peptide-aLuc conjugate synthesis is the availability of Boc protected aLuc in large

quantities. Coupling a protected amino acid to the 6-amino-2-cyanobenzothiazole (3) would have

limited our possibilities. Our aim was to synthetize a general, all-purpose Boc protected aLuc to

which any peptide sequence can be attached. The desired substance was obtained in a 2-step

route.

3.2.3.1 Unsuccessful attempts

3.2.3.1.1 Unsuccessful Boc-anhydride attempt to produce N-Boc-aLuc

At our first attempt we tried the reaction of aLuc and Boc anhydride (solvent: acetone/water 1:1,

base: Na2CO3, room temperature, overnight). Due to the low reactivity of the amino function,

however, the method proved to be unsuccessful (Scheme 29).

Scheme 29 Unsuccessful Boc-anhydride attempt to produce N-Boc-aLuc

- 35 -


3.2.3.1.2 Unsuccessful Boc-anhydride attempt to produce N-Boc-6-amino-2-cyanobenzothiazole

The second attempt, reacting the precursor of aLuc, 6-amino-2-cyanobenzothiazole (3) with Boc

anhydride (solvent: acetonitrile/acetone/water 2:4:1, base: Na2CO3, room temperature,

overnight), also failed, the amino group was still not reactive enough. (Scheme 30).

Scheme 30 Unsuccessful Boc-anhydride attempt to produce N-Boc-6-amino-2-

cyanobenzothiazole

We did not want to use the method we used earlier, that is, coupling a protected amino acid to the

6-amino-2-cyanobenzothiazole (3), [1], because it would have limited our possibilities: our goal

was to synthetize a general, all-purpose Boc protected aLuc to which any peptide sequence can

be attached.

After a thorough examination of the methods in the literature 41-57, they had to be dismissed. For

our purposes, the disadvantage of these methods is that a) none of them is applied on substances

of such extremely low nucleophilicity as the one in our research – 6-amino-2-cyanobenzothiazole

(3); b) as our target substance serves as a building block, needed in great quantities for the

scalable solid phase Boc strategy of N-peptide-aLuc conjugates, these methods are highly

uneconomical.

- 36 -


3.2.3.1.3 Unsuccessful triphosgene attempt to produce N-Boc-6-amino-2-cyanobenzothiazole

When trying to incorporate a Boc protecting group into amines of low nucleophilicity, using

phosgene or a phosgene equivalent proves to be the solution.58 In our first triphosgene attempt

tBuOH and triphosgene were reacted in EtOAc at -15 C° (producing chloroformic acid tert-butyl

ester as intermediate) under continuous stirring for 1 hour, then the mixture of 6-amino-2-

cyanobenzothiazole (3) and DIPEA, both dissolved in EtOAc, was added dropwise in 1.5 hours at

-15°C. The new mixture was stirred for another 60 minutes at -15°C, then it was neutralized with

28% aquous NaOH solution at 0°C. It was extracted both with EtOAc and water 3 times, then

dried over sicc Na2SO4. The resulting material did contain N-Boc-6-amino-2-cyanobenzothiazole

(9), but due to the low yield (10% corresponding to the crude product, Scheme 31), the method

had to be dismissed and another method from the same patent 58 had to be applied.

Scheme 31 Unsuccessful triphosgene attempt to produce N-Boc-6-amino-2-cyanobenzothiazole

- 37 -


3.2.3.2 Successful synthesis route

3.2.3.2.1 N-Boc-6-amino-2-cyanobenzothiazole synthesis (9)

According to another method from the aforementioned patent,58 6-amino-2-cyanobenzothiazole

(3) was solved in EtOAc, then DIPEA was added. Triphosgene was added to this mixture

portionwise. The reaction system was stirred for 1 hour at 60°C (producing (2-

cyanobenzo[d]thiazol-6-yl)carbamic chloride as intermediate), then tBuOH was added. The new

mixture was stirred for another 1 hour at 60°C. Then the acidic solution was neutralized with

28% aquous NaOH solution at 0°C and extracted with EtOAc, water and saturated NaCl solution,

then dried over sicc Na2SO4 and evaporated. This time the yield of N-Boc-6-amino-2-

cyanobenzothiazole (9) was significantly better (74% corresponding to the crude product) (II.31,

II.32, Scheme 32), which might be the result of the fact that (2-cyanobenzo[d]thiazol-6-

yl)carbamic chloride is a much more stable intermediate than the intermediate chloroformic acid

tert-butyl ester. Recipe: III.11

Scheme 32 Synthesis of N-Boc-6-amino-2-cyanobenzothiazole (9)

- 38 -


3.2.3.2.2 N-Boc-aLuc synthesis (10)

The resulting N-Boc-6-amino-2-cyanobenzothiazole (9) was dissolved in a mixture of MeOH and

THF, then aqueous solution of D-cysteine hydrochloride monohydrate was added, and the

mixture was stirred for 20 minutes at room temperature. The cysteine was released from its salt

with 5 m/m% NaHCO3 solution. During the reaction (about 75 minutes) the pH of the solution

was kept between 7.3-7.4 with the addition of 5 m/m% NaHCO3 solution, the process was

continuously monitored with a pH meter, under argon atmosphere (Scheme 33). The solution was

evaporated then taken up in water and extracted with methyl tert-butyl ether. Dropping the

aqueous phase on a mixture of ice and glacial acetic acid, a fine yellow precipitate, N-Boc-aLuc

free carboxylic acid (10) formed, which was then filtered, yield: 85%.

Scheme 33 Synthesis of N-Boc-aLuc (10)

The structure of the product was attested by 1H-NMR (II.33) and 13C-NMR (II.34) spectra and

mass spectrometry (II.35), its optical purity was attested with chiral chromatography (II.36,

II.37). Optical rotation was also measured. Recipe: III.2.12

- 39 -


3.2.4 Biological testing

3.2.4.1 Biological testing of N-Z-Asp-Glu-Val-Asp-aLuc (6)

3.2.4.1.1 Biochemical and cellular testing

The biological relevance of our N-Z-Asp-Glu-Val-Asp-aLuc (6) substrate was confirmed in a

biochemical reaction using a serial dilution of recombinant caspase-3 from 227 mU/reaction to

22.7µU/reaction. It has been published that both caspase-3 and caspase-7 digest Asp-Glu-Val-

Asp sequence, but caspase-3 has six-time higher Asp-Glu-Val-Asp digestion activity 61, 62 (Figure

7A and 7B). In order to verify that our N-Z-Asp-Glu-Val-Asp-aLuc (6) is a real substrate for

caspase-3, not only the enzyme but also the N-Z-Asp-Glu-Val-Asp-aLuc (6) substrate was titrated

in a concentration range from 100 µM to 1 µM. The recorded luminescence was linearly

proportional with the increasing enzyme activity and increased amount of the N-Z-Asp-Glu-Val-

Asp-aLuc (6) substrate, showing maximum cps at 227 mU caspase-3 and 100 µM N-Z-Asp-Glu-

Val-Asp-aLuc (6) (Figure 7A, II.38, IV.1) Moreover, the luminescence signal was completely

abolished by the application of equimolar pan-caspase inhibitor Z-Val-Ala-Asp-fmk in the

reaction of 2.27 mU caspase-3 and 10 µM N-Z-Asp-Glu-Val-Asp-aLuc (6),63 (Figure 7B). We

also verified the applicability of our N-Z-Asp-Glu-Val-Asp-aLuc (6) substrate to detect cellular

apoptotic cell death caused by a drug candidate molecule. The curcumin analogue C150 induces

caspase-3 activation (II.39, IV.2) and apoptosis of A549 human non-small cell lung carcinoma

cells 64, 65 and we could detect the activation of caspase-3 by N-Z-Asp-Glu-Val-Asp-aLuc (6) via

bioluminescence (Figure 7C, IV.3, IV.4).

- 40 -


Figure 7 Testing the N-Z-Asp-Glu-Val-Asp-aLuc (6) substrate in biochemical (A and B) and

cellular (C) systems. Control corresponds to untreated sample. The results are shown as

arithmetic mean values of three samples ± SEM

3.2.4.1.2 In vivo testing

To measure apoptosis directly in animals, an optical imaging experiment was performed in vivo,

administrating N-Z-Asp-Glu-Val-Asp-aLuc (6) to SCID mice (previously inoculated with the

stably expressing luciferase cell line U87-Luc) that had been treated with chemotherapeutics

previously. We used Ac-915, a lipid droplet binding thalidomide analogue inducing caspase-3

activation (II.40), and oxidative stress and apoptosis in different cancer cells. 66 Ac-915 enhanced

the bioluminescent signal already at 6 hours. Significantly fewer signals were detected from

control mouse having no Ac-915 treatment, but injected with only N-Z-Asp-Glu-Val-Asp-aLuc

(8) substrate, which represents the basal level of apoptosis. However, the limitation of the

widespread applicability of the luciferin conjugated peptides in vivo is that luciferase enzyme

activity is indispensable, therefore luciferase transgenic mouse or cells should be used in these

studies (IV.5)

- 41 -


Figure 8 Demonstration of applicability of the N-Z-Asp-Glu-Val-Asp-aLuc (6) substrate in

mouse bearing cancer, previous treated with an apoptosis inducer Ac-915. Aminoluciferin was

used as positive control (right)

3.2.4.2 Biological testing of N-Fmoc-Gly-Pro-aLuc (8)

N-Fmoc-Gly-Pro-aLuc (8) was tested in bioluminescence-based enzyme activity assays. The

substrate specificity of N-Fmoc-Gly-Pro-aLuc (8) was measured with two human proteases that

are involved in cancer, POP/PREP (Figure 9A) and FAP (Figure 9B), and with a bacterial non-

specific endoproteinase Pro-C (Figure 9C). All three enzymes accepted the substrate and

liberated aminoluciferin as a product, resulting in increased luminescence signal. Luminescence

was proportional to protease activity and N-Fmoc-Gly-Pro-aLuc (8) concentration. Enzymatic

degradation was confirmed with protease inhibitor, which completely abolished bioluminescent

signal increase (Figure 9D-F). Our novel substrate, therefore, could be used in different

biochemical assays with different proteases (IV.6).

- 42 -


Figure 9 (A) POP/PREP, (B) FAP and (C) Endoproteinase Pro-C protease activity on the

substrate N-Fmoc-Gly-Pro-aLuc (8). (D-F) The effect of protease inhibition. Each point

represents the average of 3 wells ± SD. Values are blank-subtracted (blank = no protease).

*p<0.05; ** p<0.01; ***p<0.001

- 43 -

5. Summary

5. SUMMARY

Efficient routes have been developed for the synthesis of N-peptide-aLuc conjugates. Two of the

routes are optimized versions of already published methods, while the third one is a completely

new method.

Precursor

As a starting step, all the three methods use a novel route for the synthesis of the precursor of the

desired final products, 6-amino-2-cyanobenzothiazole (3), which was obtained in a 3-step route:

a) nitration b) reduction c) chlorine-cyanide exchange

As starting material, cheap, commercially available 2-chlorobenzothiazole was used, which was

first nitrated. The resulting 2-chloro-6-nitrobenzothiazole (1) was reduced applying ethyl

acetate/water/ammonium chloride/iron powder system with a good yield. Using a Soxhlet extractor

turned out to be a solvent-sparing, thus environmentally-friendly method. Finally, the chlorine-

cyanide exchange was carried out in a polar aprotic non-aqueous solvent (DMAA), at high

temperature and with long reaction time.

Although the used transformations (nitration, reduction and cyanidation) are well-known in the

literature, our combination of these transformations, the equipment and the solvents make it

possible to prepare this material in larger quantities than the published strategies.

Mixed liquid/solid phase method

From the precursor, the desired N-peptide-aLuc conjugate (N-Z-Asp-Glu-Val-Asp-aLuc, 6) was

reached in a 5-step route:

a) attachment of the C-terminal amino acid of the target sequence b) cysteine addition

c) attachment to resin d) solid phase peptide synthesis e) cleavage from resin

An N-Fmoc-protected amino acid (N-Fmoc-Asp(OtBu)-OH) was coupled to the 6-amino-2-

cyanobenzothiazole (3). As, due to the deactivated amino group, the amide bond could not be

formed with the usual coupling reagents like DCC, a much more activating coupling agent was

necessary. Excellent yield (97%) was obtained with 1.5 equivalents of TCFH. After the addition

of D-cysteine, the desired amino acid-aLuc conjugate, N-Fmoc-Asp(OtBu)-aLuc (5) was obtained.

- 44 -

5. Summary

During the next step this conjugate was attached to Wang resin, then classical solid phase peptide

synthesis was carried out: the peptide chain was built with Fmoc strategy. The obtained N-peptide-

aLuc conjugate was removed from the resin with the mixture of TFA/water; and finally, the

resulting material was purified by preparative HPLC and successfully tested in vivo and

biochemical and cellular probes.

The advantage of the method over standard liquid phase methods lies in the fact that it requires

only one chromatographic purification step. However, the high risk of dehydrogenation poses

limitations to the method.

Fragment condensation method

The desired N-peptide-aLuc conjugate (N-Fmoc-Gly-Pro-aLuc, 8) was reached in a 2-step route:

a) attachment of the target peptide sequence to 6-amino-2-cyanobenzothiazole (3)

(b) cysteine addition

A suitably protected, commercially purchased peptide, N-Fmoc-Gly-Pro-OH, was coupled with the

key molecule, 6-amino-2-cyanobenzothiazole (3). Due to the deactivated amino group of the 6-

amino-2-cyanobenzothiazole (3), the amide bond could not be formed with the usual coupling

reagents; therefore, a more powerful coupling agent, TCFH, was necessary.

D-cysteine hydrochloride monohydrate was added to the peptide-heterocycle conjugate (N-Fmoc-

Gly-Pro-6-amino-2-cyanobenzothiazole, 7). During the reaction, the pH of the solution was kept

between 7.3-7.4. The resulting material was purified by preparative HPLC and successfully tested

in bioluminescence-based enzyme activity assays.

The novelty of the method lies in the fact that the modifications of the two steps make a significant

improvement over the standard method: the extremely long coupling time of the standard mixed

anhydride method can be avoided, furthermore, the window between pH 7.3-7.4 rules out the

racemization of the D-cysteine and ensures the release of the cysteine from its salt. The

disadvantage of the method is that when attaching longer peptides (ones that contain more than ten

amino acids), solubility problems may occur, which may make coupling difficult. Also, ill-chosen

C-terminal amino acid might lead to racemization. These are typical problems with the fragment

condensation methods in general.

- 45 -

5. Summary

Immediate precursor for Boc strategy solid phase method

We have successfully produced the building block of the Boc-strategy solid phase method, N-Boc-

aLuc.

N-Boc-aLuc was obtained in a 2-step route:

a) introduction of Boc protecting group to 6-amino-2-cyanobenzothiazole (3)

b) cysteine addition

Triphosgene was added to 6-amino-2-cyanobenzothiazole (3), producing (2-cyanobenzo[d]thiazol-

6-yl)carbamic chloride as intermediate, then tert-butanol was added. D-cysteine hydrochloride

monohydrate was added to the resulting N-Boc-6-amino-2-cyanobenzothiazole (9). The pH of the

solution was kept between 7.3-7.4. The resulting material, N-Boc-aLuc was purified by preparative

HPLC.

This building block can be attached to Boc-compatible resin, then the desired peptide sequence can

be built and cleaved from the resin.

- 46 -

V. References

V. REFERENCES

1, Sato A, Klaunberg B, Tolwani R: In vivo bioluminescence imaging. 2004, Comparative Med

54, 631-634.

2, Sadikot RTB, Timothy S: Bioluminescence imaging. 2005, Proc Am Thorac Soc 2, 537-540.

3, Chollet R, Ribault S: Use of ATP Bioluminescence for Rapid Detection and Enumeration of

Contaminants: The Milliflex Rapid Microbiology Detection and Enumeration System. 2012, In

Recent Advances in Oceanic Measurements and Laboratory Applications, Lapota D (ed) InTech,

China, pp. 99-118, ISBN: 978-953-307-940-0

4, Ioka S, Saitoh T, Iwano S, Suzuki K, Maki SA, Miyawaki A, Imoto M, Nishiyama S:

Synthesis of Firefly Luciferin Analogues and Evaluation of the Luminescent Properties. 2016,

Chem Eur J 22, 9330-9337.

5, Kaskova ZM, Tsarkova AS, Yampolsky IV: 1001 lights: luciferins, luciferases, their

mechanisms of action and applications in chemical analysis, biology and medicine. 2016, Chem

Soc Rev 45, 6048-6077.

6, Xu TT, Close D, Handagama W, Marr E, Sayler G, Ripp S: The expanding Toolbox of In Vivo

Bioluminescent imaging. 2016, Frontiers in Oncology 6:150.

7, Jiang YL, Zhu Y, Moore AB, Miller K, Broome AM: Biotinylated Bioluminescent Probe for

Long Lasting Targeted In vivo Imaging of Xenografted Brain Tumors in Mice. 2018, ACS

Chemical Neuroscience 9, 100-106.

8, Kepp O, Galluzzi L, Lipinski M, Yuan J, Kroemer G: Cell death assays for drug discovery.

2011, Nat Rev Drug Discov 10 (3), 221-237.

9, O'Brien MA, Daily WJ, Hesselberth PE, Moravec RA, Scurria MA, Klaubert DH, Bulleit RF,

Wood KV: Homogenous, Bioluminescent Protease Assays: Caspase-3 as a Model. 2005, J of

Biomol Scr 10 (2), 137-148.

10, Hickson J, Ackler S, Klaubert D, Bouska J, Ellis P, Foster K, Oleksijew A, Rodriguez L,

Schlessinger S, Wang B: Noninvasive molecular imaging of apoptosis in vivo using a modified

firefly luciferase substrate, Z-DEVD-aminoluciferin. 2010, Cell Death Differ 17 (6), 1003-522

1010.

11, De Saint-Hubert M, Devos E, Ibrahimi A, Debyser Z, Mortelmans L, Mottaghy FM:

Bioluminescence imaging of therapy response does not correlate with FDG-PET response in a

mouse model of Burkitt lymphoma. 2012, Am J Nucl Med Mol Imaging 2, 353-361.

- 47 -

V. References

12, Paley MA, Prescher JA: Bioluminescence: a versatile technique for imaging cellular and

molecular features. 2014, Medchemcomm 5 (3), 255-267.

13, Presiado I, Erez Y, Simkovitch R, Shomer S, Gepshtein R, da Silva LP, da Silva J, Huppert

D: Excited-State Proton Transfer of Firefly Dehydroluciferin. 2012, Journal of Physical

Chemistry A 116, 10770-10779.

14, Zhang YM, Pullambhatla M, Laterra J, Pomper MG: Influence of Bioluminescence Imaging

Dynamics by D-Luciferin Uptake and Efflux Mechanisms. 2012, Molecular Imaging 11, 499-

506.

15, White EH, Wörther H, Seliger HH, McElroy WD: Amino Analogs of Firefly Luciferin and

Biological Activity Thereof 1. 1966, J Am Chem Soc 88, 2015-2019.

16, Leippe, DM, Nguyen D, Zhou M, Good T, Kirkland TA, Scurria M, Bernad L, Ugo T,

Vidugiriene J, Cali JJ, Klaubert DH, O'Brien MA: A bioluminescent assay for the sensitive

detection of proteases. 2011, BioTechniques 51 (2), 105-110.

17, O'Brien MA, Moravec RA, Riss TL, Bulleit RF: Homogeneous, bioluminescent proteasome

assays. 2008, Methods Mol Biol. 414, 163-81.

18, Maki S, Kojima S, Niwa H: Luminescent substrate for luciferase. 2011, Pub.No.:

US2011/0033878 A1

19, Iwano S, Obata R, Miura C, Kiyama M, Hama K, Nakamura M, Amano Y, Kojima S, Hirano

T, Maki S, Niwa H: Development of simple firefly luciferin analogs emitting blue, green, red,

and near-infrared biological window light. 2013, Tetrahedron 69, 3847-3856.

20, Takakura H, Kojima R, Urano Y, Terai T, Hanaoka K, Nagano T: Aminoluciferins as

Functional Bioluminogenic Substrates of Firefly Luciferase. 2011, Chemistry - An Asian Journal

6 (7), 1800-1810.

21, Hsu HT, Trantow BM, Waymouth RM, Wender PA: Bioorthogonal Catalysis: A General

Method to Evaluate Metal-catalyzed Reactions in Real Time in Living Systems Using a Cellular

Luciferase Report System. 2016, Bioconjug Chem 138, 376-382.

22, Gryshuk AL, Perkins J, LaTour JV: Methods and systems for synthesis of a D-aminoluciferin

precursor and related compounds. 2013, Patent number: 8586759

23, McCutcheon DC, Porterfield WB, Prescher JA: Rapid and scalable assembly of firefly

luciferase substrates. 2015, Org Biomol Chem 13, 2117-2121.

24, Hauser JR, Beard HA, Bayana MF, Jolley KF, Warriner SI, Bon RS: Economical and scalable

synthesis of 6-amino-2-cyanobenzothiazole. 2016, Beilstein J Org Chem 12, 2019-2025.

- 48 -

V. References

25, Drozdov NS, Stavrovskaya VI: Derivatives of benzothiazole III. On some derivatives of 2-

aminobenzothiazole (in Russian). 1939, J Gen Chem USSR 9, (5), 409-414.

26, Hofmann AW: Ueber eine Reihe aromatischer, den Senfölen und Sulfocyanaten isomerer

Basen. 1880, Berichte 13, 8-22.

27, Drozdov NS, Stavrovskaya VI: Derivatives of benzothiazole II. On making 2-

chlorobenzothiazole and some of its derivatives (in Russian). 1937, J Gen Chem USSR 7, (23),

2813-2818.

28, Katz L: Antituberculous Compunds II. 2-Benzalhydrazinobenzothiazoles. 1951, J Am Chem

Soc 73, 4007-10.

29, Geiger R, Miska W: Aminoluciferin derivatives, processes for the production thereof and

their application in the determination of enzyme activities. 1991, US Patent US5035999A

30, O'Brian MA, Wood KV, Klaubert D, Daily B: Bioluminescent protease assay. 2003, US

Patent US20030211560A1

31, Gryshuk AL, Perkins J., LaTour JV: Methods and systems for synthesis of a D-aminoluciferin

precursor and related compounds. 2011, US Patent US20110224442A1

32, El-Faham A, Albericio F: COMU: A third generation of uronium-type coupling reagents.

2010, J of Pep Sci 16 (1), 6-9.

33, Chantell CA, Onaiyekan MA, Menakuru M: Fast conventional Fmoc solid-phase peptide

synthesis: a comparative study of different activators. 2012, J of Pep Sci 18 (2), 88-91.

34, El-Faham A, Subiros-Funosas R, Albericio F: A Novel Family of Onium Salts Based Upon

Isonitroso Meldrum's Acid Proves Useful as Peptide Coupling Reagents. 2010, Eur J of Org

Chem 19, 3641-3649.

35, Kangani CO, Kelley DE, Day BW: One-pot synthesis of aldehydes or ketones from

carboxylic acids via in situ generation of Weinreb amides using the Deoxo-Fluor reagent. 2006,

Tetrah Lett 47 (35), 6289-6292.

36, Carpino LA, Triolo SA, Griffin GW, Herman LW, Tarr G, Sole NA, Diekmann E, El-Faham,

A, Ionescu D, Albericio F, Edited by Epton R: Innovation and Perspectives in Solid Phase

Synthesis & Combinatorial Libraries: Peptides, Proteins and Nucleic Acids- Small Molecule

Organic Chemical Diversity, 1996, Collected Papers of 4th International Symposium, Edinburgh,

Sept. 12-16., 1995, 41-50.

37, Tulla-Puche J, Torres A, Calvo P, Royo M, Albericio F: N,N,N',N'-

Tetramethylchloroformamidinium Hexafluorophosphate (TCFH), a Powerful Coupling Reagent

for Bioconjugation. 2008, Bioconj Chem 19 (10), 1968-1971.

- 49 -

V. References

38, Ciuffreda P, Casati S, Meroni G, Santaniello E: A new synthesis of dehydroluciferin [2-(6 '-

hydroxy-2 '-benzothiazolyl)-thiazole-4-carboxylic acid] from 1,4-benzoquinone. 2013,

Tetrahedron 69, 5893-5897.

39, Fontes R, Dukhovich A, Sillero A, Sillero MAG: Synthesis of dehydroluciferin by firefly

luciferase: Effect of dehydroluciferin, coenzyme A and nucleoside triphosphates on the

luminescent reaction. 1997, Biochem Bioph Res Co 237, 445-450.

40, Nyfeler R: Peptide synthesis via fragment condensation. 1994, Methods Mol Biol 35, 303-

316.

41, Heydari A, Shiroodi RK, Hamadi H, Esfandyari M, Pourayoubi M: N-tert-

Butoxycarbonylation of amines using H3PW12O40 as an efficient heterogeneous and recyclable

catalyst. 2007, Tetrah Lett, 48, 5865-5868.

42, Sharma GVM, Reddy JJ, Lakshmi PS, Radha Krishna PR: Rapid and facile Lewis acid

catalysed Boc protection of amines. 2004, Tetrah Lett 45, 6963-6965.

43, Chakraborti AK, Chankeshwara SV: HClO4–SiO2 as a new, highly efficient, inexpensive and

reusable catalyst for N-tert-butoxycarbonylation of amines. 2006, Org Biomol Chem 4, 2769-

2771.

44, Chankeshwara SV, Chakraborti AK: Copper(II) tetrafluoroborate as a novel and highly

efficient catalyst for N-tert-butoxycarbonylation of amines under solvent-free conditions at room

temperature. 2006, Tetrahedron Lett 47, 1087-1091.

45, Bartoli G, Bosco M, Locatelli M, Marcantoni E, Massaccesi M, Melchiorre P, Sambri LA:

Lewis Acid-Mediated Protocol for the Protection of Aryl Amines as their Boc-Derivatives. 2004,

Synlett 10,1794-1798.

46, Chankeshwara SV, Chakraborti AK: Montmorillonite K-10 and montmorillonite KSF as new

and reusable catalysts for conversion of amines to N-tert-butylcarbamates. 2006, J Mol Catal A

253, 198–202.

47, Jahani F, Tajbakhsh M, Khaksar S, Reza Azizi M: An efficient and highly chemoselective N-

Boc protection of amines, amino acids, and peptides under heterogeneous conditions. 2011, Mon

Chem 142, 318 1035-1043.

48, Heydari A, Hosseini SE: Lithium Perchlorate‐Catalyzed Boc Protection of Amines and

Amine Derivatives. 2005, Adv Synth Cat 347, 1929-1932.

49, Upadhyaya DJ, Barge A, Stefania R, Cravotto G: Efficient, solventless N-Boc protection of

amines carried out at room temperature using sulfamic acid as recyclable catalyst. 2007, Tetrah

Lett 48, 8318-8322.

- 50 -

V. References

50, Pandey RK, Dagade SP, Upadhyay RK, Dongare MK, Kumara P: A facile procedure for tert-

butoxycarbonylation of amines promoted by yttria-zirconia based strong Lewis acid catalyst.

2002, ARKIVOC 7, 28-33.

51, Suryakiran N, Prabhakar P, Reddy TS, Rajesh K, Venkateswarlu Y: Facile N-tert-

butoxycarbonylation of amines using La(NO3)3·6H2O as a mild and efficient catalyst under

solvent-free conditions. 2006, Tetrah Lett 47, 8039-8042.

52, Sunitha S, Kanjilal S, Reddy PS, Prasad RBN: An efficient and chemoselective Brønsted

acidic ionic liquid-catalyzed N-Boc protection of amines. 2008, Tetrah Lett 49, 2527-2532.

53, Zolfigol MA, Khakyzadeh V, Moosavi-Zare AR, Chehardoli G, Derakhshan-Panah F, Zare

A, Khaledian O: Novel ionic liquid 1,3-disulfonic acid imidazolium hydrogen sulfate

{[Dsim]HSO4} efficiently catalyzed N-Boc protection of amines. 2012, Sci Iran Trans C 19,

1584-1590.

54, Akbari J, Heydari A, Ma’mani L, Hosseini SH: Protic ionic liquid [TMG][Ac] as an efficient,

homogeneous and recyclable catalyst for Boc protection of amines. 2010, C. R. Chimie 13, 544–

547.

55, Chankeshwara SV, Chakraborti AK: Catalyst-Free Chemoselective N-tert-

Butyloxycarbonylation of Amines in Water. 2006, Org Lett 8, 3259-3262.

56, Deb B, Debnath S, Deb A, Maiti DK, Majumdar S: Copper nanoparticles catalyzed N-H

functionalization: an efficient solvent-free N-tert-butyloxycarbonylation strategy. 2017, Tetrah

Lett 58, 346 629-633.

57, Amira A, K’tir H, Berredjem M, Aouf N-E: A simple, rapid, and efficient N-Boc protection

of amines under ultrasound irradiation and catalyst-free conditions. 2014, Monatsh Chem 145,

509-515.

58, Inoue H, Noda K: Process for preparation of t-butoxycarbonylamine compounds. United

States Patent, 2014, Patent No.: US 8,653,269 B2

59, Shinde R, Perkins J., Contag CH: Luciferin Derivatives for Enhanced in vitro and in vivo

Bioluminescence Assays. 2006, Biochemistry 45 (37), 11103-11112.

60, Reddy GR, Thompson WC, Miller SC: Robust light emission from cyclic alkylaminoluciferin

substrates for firefly luciferase. 2010, J Am Chem Soc 132, 13586-13587.

61, McStay GP, Salvesen GS, Green DR: Overlapping cleavage motif selectivity of caspases:

implications for analysis of apoptotic pathways. 2008, Cell Death and Diff. 15, (2), 322-331.

- 51 -

V. References

62, Talanian RV, Quinlan C, Trautz S, Hackett MC, Mankovich JA, Banach D, Ghayur T, Brady

KD, Wong WW: Substrate Specificities of Caspase Family Proteases. 1997, J Biol Chem 272,

(15), 9677–9682.

63, Ion G, Fajka-Boja R, Kovács F, Szebeni G, Gombos I, Czibula Á, Matkó J, Monostori É:

Acid sphingomyelinase mediated release of ceramide is essential to trigger the mitochondrial

pathway of apoptosis by galectin-1. 2006, Cell Sign 18, 1887–1896.

64, Nagy LI, Fehér LZ, Szebeni GJ, Gyuris M, Sipos P, Alföldi R, Ózsvári B., Hackler Jr L,

Balázs A, Batár P, Kanizsai I, Puskás LG: Curcumin and Its Analogue Induce Apoptosis in

Leukemia Cells and Have Additive Effects with Bortezomib in Cellular and Xenograft Models.

2015, BioMed Res Int 1-12.

65, Hackler Jr L, Ózsvári B, Gyuris M, Sipos P, Fábián G, Molnár E, Marton A, Faragó N,

Mihály J, Nagy LI, Szénási T, Diron A, Párducz A, Kanizsai I, Puskás LG: The Curcumin

Analog C-150, Influencing NF-κB, UPR and Akt/Notch Pathways Has Potent Anticancer

Activity In Vitro and In Vivo. 2016, PLoS One 11 (3)

66, Nagy LI, Molnár E, Kanizsai I, Madácsi R, Ózsvári B, Fehér LZ, Fábián G, Marton A, Vizler

C, Ayaydin F, Kitajka K, Hackler Jr L, Mátés L, Deák F, Kiss I, Puskás LG: Lipid droplet

binding thalidomide analogs activate endoplasmic reticulum stress and suppress hepatocellular

carcinoma in a chemically induced transgenic mouse model. 2013, Lipids Health Dis 12, 175.

67, Szebeni GJ, Balázs Á, Madarász I, Pócz G, Ayaydin F, Kanizsai I, Fajka-Boja R, Alföldi R,

Hackler Jr L, Puskás LG: Achiral Mannich-Base Curcumin Analogs Induce Unfolded Protein

Response and Mitochondrial Membrane Depolarization in PANC-1 Cells. 2017, Int J Mol Sci 18,

(10), 2105.

- 52 -

5. Acknowledgements

5. ACKNOWLEDGEMENTS

I am grateful to my supervisors, Professor Gábor K. Tóth, Head of the Department of Medical

Chemistry for providing me with the opportunity to perform my work at the department and Dr

László G. Puskás for his scientific guidance.

I would like to give my special thanks to Dr Péter Hegyes for his love and support during my

PhD years.

I am greatly indebted to Dr Gábor J. Szebeni and Dr Lajos I. Nagy for their contribution in

biological research, and also Dr Anasztázia Hetényi and Dr Krisztián Bogár for their

invaluable work in NMR spectroscopy.

The completion of my dissertation would not have been possible without the help of Professor

Alexey Trofimov and Mr Tibor Borbás.

Finally, I would like to thank my colleagues at the department for their help, encouragement and

friendship throughout the years.

APPENDIX

- 53 -

I.Tables

I.TABLES

Table I.1 Literary overview of 6-amino-2-cyanobenzothiazole synthesis routes

Hauser Takakura Hsu

4-nitroaniline 6-nitrobenzothiazole ethyl benzothiazole 2-carboxylate 2-chlorobenzothiazole

McCutcheon / 1 McCutcheon / 2 Gryshuk / 1 Gryshuk / 2 Gryshuk / 3 Gryshuk / 4

2-chloro-6-nitrobenzothiazole

2-cyano-6-nitrobenzothiazole 6-amino-2-chlorobenzothiazole

6-amino-2-cyanobenzothiazole

(4-nitrophenyl)-

carbonocyanido-

thioic amide

2-amion-5-

nitrobenzenethiol

6-nitrobenzothiazole-

2-carboxylate

ethyl-6-

nitrobenzothiazole-2-

carboxylate

benzothiazole-2-carboxamide

2-cyanobenzothiazole6-nitrobenzothiazole-2-carboxamide

- 54 -

I.Tables

Table I.2.1 Methods for the synthesis of 6-amino-2-cyanobenzothiazole

Author Starting material Transformation Yield (%)

Overall yield

Reagent Solvent Reaction time Temperature (°C)

Disadvantages Comments

Takakura et al. 2011

2-chlorobenzothiazole nitration 49

16

H2SO4, KNO3 H2SO4 1h 30m 0-15 -


reduction 61 SnCl2, HCl,

NaOH EtOH, H2O ? 120

A significant amount of by-product (2-hydroxy-6-aminobenzothiazole and 6-aminobenzothiazole) formed. The high temperature cannot be understood

6-amino-2-chlorobenzothiazole

cyanidation 54 KCN DMSO overnight 135

DMSO is not the best solvent for this step, as with distillation it cannot be regained from the aqueous solution.

Gryshuk et al. 1 2013

6-nitrobenzothiazole acylation 61

8

H2SO4, H2O2, FeSO4,

NaHCO3 H2O more than 30m rt, 0

The step itself is good, but the starting material is ill-chosen.

The starting material is not optimal, consequently the method is too complicated, the desired end-product is reached in four steps, instead of the others’ three.

6-nitrobenzothiazole-2-carboxylate

amidation 80 NH3 MeOH 20m rt -

6-nitrobenzothiazole-2-carboxamide

cyanidation 24 POCl3, H2O,

pyridine EtOAc 2h 20m 0, rt -

2-cyano-6-nitrobenzothiazole

reduction 65 SnCl2,

NaHCO3 EtOH, H2O 2h rt, 60

If the cyano group is present during the NO2 reduction, there is a high risk of its reduction into an aminomethyl group.

- 55 -

I.Tables



Overall yield




ethyl benzothiazole-2-carboxylate

nitration 10

5

H2SO4, HNO3

H2O more than 1h 0-15

The ill-chosen starting material can be nitrated with very low yield, also, there is a high risk of hydrolysis under the aqueous-acidic conditions.

The desired end-product is reached in four steps, instead of the others’ three.

ethyl-6-nitrobenzothiazole-2-carboxylate

amidation 80 NH3 MeOH 20m rt -





reduction 65 SnCl2,


If the cyano group is present during the NO2 reduction, there is a high risk of its reduction into an aminomethyl group



amidation ?

(<5)

NH3 MeOH 15-30m rt -

In step 1 there is 100% conversion rate, but due to the lack of purification, no known yield.


nitration 30 H2SO4, HNO3

H2O overnight 0-15 There is a risk of hydrolysis of the amide under the aqueous-acidic conditions.





reduction 65 SnCl2,



- 56 -

I.Tables



Overall yield





amidation ?

(<62)

NH3 MeOH 15-30m rt - In step 1, 100% conversion, but due to the lack of purification, no known yield, In step 3, yield was not determined, therefore the overall yield cannot be determined.




2-cyanobenzothiazole

nitration ? H2SO4, HNO3 H2O 6h 0 -


reduction 65 SnCl2, NaHCO3 EtOH, H2O 2h rt, 60

If the cyano group is present during the NO2 reduction, there is a high risk of its reduction into an aminomethyl group

McCutcheon et al. 1 2015

4-nitroaniline thiocyanidation 80

56

pyridine, sodium thiosulphate, 4,5-

dichloro-1,2,3-dithiazol-1-ium

chloride

H2O, acetonitrile,

THF 1h 30m rt

Complicated method, fewer reagents would be more reasonable. 4,5-dichloro-1,2,3-dithiazol-1-ium chloride is very expensive.

Extremely complicated route.

(4-nitrophenyl)-carbonocyanidothioic amide

cyanidation 74 CuI, Bu4N+ •Br-, Catalyst: PdCl2

DMF, DMSO

4h 130 Extremely complicated method, special catalysts are required.

6-nitro-2-cyanobenzothiazole

reduction 95 Zn, NH4Cl MeOH 35m rt



6-nitrobenzothiazole

decomposition of the thiazole

ring

62

37

N2H4

EtOH, DCM

12h

rt, reflux

-

2-amino-5-nitrobenzenethiol

cyanidation

62

4,5-dichloro-1,2,3-dithiazol-1-

ium chloride

DCM

12 h

reflux

4,5-dichloro-1,2,3-dithiazol-1-ium chloride is very expensive.


reduction

95

NH4Cl, Zn

MeOH

35m

rt


- 57 -

I.Tables



Overall yield

Reagent Solvent Reaction time

Temperature (°C)


Hsu et al. 2016


27

H2SO4, KNO3 H2SO4 4h 0-15 -

Too much by-product and waste.


reduction 88 Fe, FeCl3 EtOH,

acetic acid 24h 80

The resulting by-product (Fe(III)-acetate) is difficult to separate from the desired product.


cyanidation 45 KCN DMSO 24h 120

DMSO is not the best solvent for this step, as with distillation it cannot be regained from the aqueous solution.

Hauser et al. 2016


44

H2SO4, KNO3 H2SO4 18h 0-15 -

Too much by-product.


cyanidation 75 NaCN, Dabco catalyst, FeCl3

H2O, acetonitrile

24h rt

Two byproducts (2-carbethoxy-6-nitrobenzothiazole and 2-ethoxy-nitrobenzothiazole) formed.


reduction 71 Fe acetic acid 24h rt

If the cyano group is present during the NO2 reduction, also, there is a high risk of its reduction into an aminomethyl group. The resulting by-product (Fe(III)-acetate) is difficult to separate from the desired product.

Kovács et al 2018.

2-chlorobenzothiazole nitration 83 57 H2SO4, KNO3 H2SO4 5h 0-15 - -


reduction

88

NH4Cl, H2O, Fe powder

EtOAc

8h

reflux

-

Using Soxhlet extractor, the EtOAc can be fully regained.


cyanidation

78

KCN

DMAA

12h

110

The reagent is toxic, but was reacted with KH2PO4 in order to get non-toxic KOCN.

The use of DMAA instead of other solvents results in significantly better yield during cyanidation.

- 58 -

I.Tables

Table I.3 Comparison of starting materials


Reagent Solvent Reaction time Temperature

(°C) Disadvantages Comments

Takakura et al. 2011 2-chlorobenzothiazole

nitration 49 H2SO4, KNO3 H2SO4 1h 30m 0-15 _


6- nitrobenzothiazole

acylation 61 H2SO4, H2O2,

FeSO4, NaHCO3 H2O more than 30m rt, 0

The step itself is good, but the starting material is ill-chosen, which makes the following step uneconomical.

.



nitration 10 H2SO4, HNO3 H2O more than 1h 0-15

The ill-chosen starting material can be nitrated with very low yield, also, there is a high risk of hydrolysis under the aqueous-acidic conditions.



amidation ? NH3 MeOH 15-30m rt _

100% conversion, but due to the lack of purification, yield was not determined



amidation ? NH3 MeOH 15-30m rt _

100% conversion, but due to the lack of purification, yield was not determined


4-Nitroaniline thiocyanidation 80

pyridine, sodium thiosulphate, 4,5-dichloro-

1,2,3-dithiazol-1-ium chloride

H2O, acetonitrile,

THF 1h 30m rt

Complicated method, fewer reagents would be more reasonable. 4,5-dichloro-1,2,3-dithiazol-1-ium chloride is very expensive.


6- nitrobenzothiazole

decomposition of the thiazole

ring 62 N2H4 EtOH, DCM 12h rt, reflux _

Hsu et al. 2016

2-chlorobenzothiazole

nitration 67 H2SO4, KNO3 H2SO4 4h 0-15 _

Hauser et al. 2016 2-chlorobenzothiazole

nitration 82 H2SO4, KNO3 H2SO4 18h 0-15 _

- 59 -

I.Tables

Table I.4 Comparison of solvents during chlorine-cyanide exchange

Author Starting material Transformation Yield (%) Reagent Solvent Reaction time Temperature (°C) Disadvantages



cyanidation 54 KCN DMSO overnight 135 DMSO is not the best solvent for this step, as with distillation it cannot be regained from the aqueous solution.

Hsu et al. 2016


cyanidation 45 KCN DMSO 24h 120 DMSO is not the best solvent for this step, as with distillation it cannot be regained from the aqueous solution.

Hauser et al. 2016


cyanidation 75 NaCN, Dabco catalyst, FeCl3

H2O, acetonitrile 24h rt

Two byproducts (2-carbethoxy-6-nitrobenzothiazole and 2-ethoxy-nitrobenzothiazole) formed.

- 60 -

I.Tables

Table I.5 Comparison of the optimal and the non-optimal order of reduction and cyanidation


Reagent Solvent Reaction time Temperature (°C) Disadvantages



reduction 61 SnCl2, HCl, NaOH EtOH, H2O ? 120

- 6-amino-2-chlorobenzothiazole

cyanidation 54 KCN DMSO overnight 135

Gryshuk et al. 1, 2 and 3

2013


cyanidation 24 POCl3, H2O, pyridine EtOAc 2h 20m 0, rt If the cyano group is present during the NO2 reduction, there is a high risk of its reduction into an aminomethyl group.





cyanidation 95 POCl3, H2O, pyridine EtOAc 2h 20m 0, rt If the cyano group is present during the NO2 reduction, there is a high risk of its reduction into an aminomethyl group.

2-cyanobenzothiazole nitration ? H2SO4, HNO3 H2O 6h 0




(4-nitrophenyl)-carbonocyanidothioic amide

cyanidation 74 CuI, Bu4N+ •Br-, Catalyst:

PdCl2 DMF, DMSO 4 h 130



reduction 95 Zn, NH4Cl MeOH 35m rt

McCutcheon et al 2. 2015

2-amino-5-nitrobenzenethiol

cyanidation 62 4,5-dichloro-1,2,3-

dithiazol-1-ium chloride DCM 12 h reflux



reduction 95 NH4Cl, Zn MeOH 35m rt

Hauser et al. 2016


cyanidation 75 NaCN, Dabco catalyst,

FeCl3 H2O, acetonitrile 24h rt

If the cyano group is present during the NO2 reduction, also, there is a high risk of its reduction into an aminomethyl group. The resulting by-product (Fe(III)-acetate) is difficult to separate from the desired product..


reduction 71 Fe acetic acid 24h rt

- 61 -

II. Figures

II. FIGURES

II.1. 1H-NMR spectrum of 2-chloro-6-nitrobenzothiazole (1)

(CDCl3, 500 MHz) 8.77 (s, 1H), 8.41 (d, J= 9.0 Hz, 1H), 8.10 (d, J= 9.0 Hz, 1H)

- 62 -

II. Figures

II.2. TSQ mass spectrum of 2-chloro-6-nitrobenzothiazole (1)

213.93= [M-H]-

- 63 -

II. Figures

II.3. RP-HPLC profile of the crude 2-chloro-6-nitrobenzothiazole (1)

70-100% B in 15 min + 1005 B in 5 min, tR1= 3.878 min: 2-chloro-5-nitrobenzothiazole,

tR2= 7.899 min: 2-chloro-6-nitrobenzothiazole

- 64 -

II. Figures

II.4 1H-NMR spectrum of 6-amino-2-chlorobenzothiazole (2)

([D6]DMSO, 600 MHz ) (d, J = 8.4 Hz, 1H), 7.04 (d, J = 2.4 Hz, 1H), 6.78 (dd, J1 = 1.8

Hz, J2 = 8.4 Hz, 1H), 5.53 (bs, 2H)

- 65 -

II. Figures

II.5 13C-NMR spectrum of 6-amino-2-chlorobenzothiazole (2)

([D6]DMSO, 150 MHz) 148.24, 145.51, 142.00, 137.86, 123.13, 115.51, 104.18

- 66 -

II. Figures

II.6 ESI mass spectrum of 6-amino-2-chlorobenzothiazole (2)

185.0= [M+H]+

- 67 -

II. Figures

II.7 RP-HPLC profile of 6-amino-2-chlorobenzothiazole (2)

5-80% B in 25 min + 3 min up to 100% B + 100% B in 5 min

tR= 11.527 min

- 68 -

II. Figures

II.8 1H-NMR of 6-amino-2-cyanobenzothiazole (3)

([D6]DMSO, 600 MHz ) (d, J = 9.0 Hz, 1H), 7.22 (d, J = 1.8 Hz, 1H), 7.01 (dd, J1 = 1.8 Hz,

J2 = 9.0 Hz, 1H), 4.40 (bs, 2H)

- 69 -

II. Figures

II.9 13C-NMR spectrum of 6-amino-2-cyanobenzothiazole (3)

([D6]DMSO, 150 MHz) 150.06, 144.10, 138.75, 128.86, 125.62, 118.19, 114.71, 103.54

- 70 -

II. Figures

II.10 ESI mass spectrum of 6-amino-2-cyanobenzothiazole (3)

176.0= [M+H]+

- 71 -

II. Figures

II.11 RP-HPLC profile of 6-amino-2-cyanobenzothiazole (3)

5-80% B in 25 min + 3 min up to 100% B + 5 min 100% B

tR= 16.692 min

- 72 -

II. Figures

II.12 TOF mass spectrum of N-Fmoc-Asp(OtBu)-6-amio-2-cyanobenzothiazole (4)

569.2070= [M+H]+

- 73 -

II. Figures

II.13 RP-HPLC of the purified N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole (4)

50-100% B in 25 min

tR= 10.591 min

- 74 -

II. Figures

II.14 TOF mass spectrum of N-Fmoc-Asp(OtBu)-aLuc (5)

673.1882= [M+H]+

- 75 -

II. Figures

II.15 RP-HPLC profile of the N-Fmoc-Asp(OtBu)-aLuc (5)

50-100% B in 25 min

tR= 21.046 min

- 76 -

II. Figures

II.16. part 1 1H-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc (6)

([D6]DMSO, 600 MHz) 10.26 (bs, 1H), 8.61 (s, 1H), 8.42 (d, J = 7.2 Hz, 1H), 8.08 (d, J =

9.0 Hz, 2H), 7.81 (bs, 1H), 7.64 (dd, J1 = 9.0 Hz, J2 = 29.4 Hz, 2H), 7.34 (s, 5H), 5.42 (t, J =

8.4 Hz, 1H), 5.02 (s, 2H), 4.69 (d, J = 7.2 Hz, 1H), 4.34 (dd, J1 = 5.4 Hz, J2 = 29.4 Hz, 2H),

4.12 (bs, 1H), 3.77 (t, J = 10.8 Hz, 1H), 3.68 (dd, J1 = 8.4 Hz, J2 = 11.4 Hz, 1H), 2.77 (bs,

1H), 2.61-2.68 (m, 2H), 2.25-2.46 (m, 3H), 1.76-1.99 (m, 2H), 0.83-0.86 (m, 7H)

- 77 -

II. Figures

II.16 part 2 1H-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc (6)





1H), 2.61-2.68 (m, 2H), 2.25-2.46 (m, 3H), 1.76-1.99 (m, 2H), 0.83-0.86 (m, 7H)

- 78 -

II. Figures

II.16 part 3 1H-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc (6)





1H), 2.61-2.68 (m, 2H), 2.25-2.46 (m, 3H), 1.76-1.99 (m, 2H), 0.83-0.86 (m, 7H)

- 79 -

II. Figures

II.17 part 1 13C-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc (6)

([D6]DMSO, 150 MHz) 174.54, 172.00, 171.61, 171.38, 170.24, 159.62, 156.32, 149.14,

138.66, 137.31, 136.71, 128.83, 128.28, 128.19, 124.65, 120.28, 112.04, 78.64, 66.01, 58.21,

52.53, 51.86, 51.24, 36.77, 36.31, 35.23, 30.98, 30.52, 27.59, 19.53

- 80 -

II. Figures

II.17 part 2 13C-NMR spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc (6)

([D6]DMSO, 150 MHz) 174.54, 172.00, 171.61, 171.38, 170.24, 159.62, 156.32, 149.14,

138.66, 137.31, 136.71, 128.83, 128.28, 128.19, 124.65, 120.28, 112.04, 78.64, 66.01, 58.21,

52.53, 51.86, 51.24, 36.77, 36.31, 35.23, 30.98, 30.52, 27.59, 19.53

- 81 -

II. Figures

II.18 ESI mass spectrum of the N-Z-Asp-Glu-Val-Asp-aLuc (6)

872.3= [M+H]+

- 82 -

II. Figures

II.19 RP-HPLC profile of the purified N-Z-Asp-Glu-Val-Asp-aLuc (6)

5-80% B in 25 min + 3 min up to 100% B + 5min 100% B

tR= 18.555 min

- 83 -

II. Figures

II.20 RP-HPLC profile of the resulting material at the determination of load

50-100% B in 25 min + 100% B in 5 min

tR1= 12.234 min: Fmoc-Pro-OH, tR2= 14.779 min: N-Fmoc-Pro-aLuc

tR3=15.299 min: N-Fmoc-Pro-6-aminodehydroluciferin

- 84 -

II. Figures

II.21 ESI mass spectrum of N-Fmoc-Pro-6-aminodehydroluciferin

597.1= [M+H]+

- 85 -

II. Figures

II.22 RP-HPLC profile of the N-Z-Gly-Pro-aminodehydroluciferin

5-80% B in 25 min + 3 min up to 100% B + 100% B in 3 min

tR= 21.240 min

- 86 -

II. Figures

II.23 ESI mass spectrum of the N-Z-Gly-Pro-6-aminodehydroluciferin

566.1= [M+H]+, 1131.2= [2M+H]+, 1413.6= [5M+2H]+, 1697.6= [3M+H]+

- 87 -

II. Figures

II.24 1H-NMR spectrum of N-Z-Gly-Pro-6-aminodehydroluciferin

([D6]DMSO, 500 MHz) 13.39 (s, 1H), 8.70 (s, 1H), 8.61 (s, 1H), 8.06 (d, J = 8.90 Hz, 1H),

7.64 (d, J = 8.80 Hz, 1H), 7.38 -7.29 (m, 4H), 5.03 (s, 2H), 4.48 (dd, J1 = 2.80 Hz, J2 = 7.92

Hz, 2H), 3.88 (ddd, J1 = 6.33 Hz, J2 = 17.20 Hz, J3 = 47.52 Hz, 4H), 2.22-2.12 (m, 2H), 2.05-

1.99 (m, 2H), 1.98-1.89 (m, 2H)

- 88 -

II. Figures

II.25 ESI mass spectrum of the N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole (7)

552.0= [M+H]+

- 89 -

II. Figures

II.26 RP-HPLC profile of the N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole (7)

50-100% B in 25 min + 3 min up to 100% B +100% B in 5 min

tR1= 8.973 min: Fmoc-Gly-Pro-OH

tR2= 17.868 min: Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole

- 90 -

II. Figures

II.27 1H-NMR of the N-Fmoc-Gly-Pro-aLuc (8)

([D6]DMSO, 500 MHz) 10.39 (s, 1H), 8.59 (t, J = 15.85 Hz, 1H), 8.09 (d, J = 8.98 Hz, 1H),

7.88 (d, J = 7.43 Hz, 2H), 7.71 (d, J = 7.48 Hz, 2H), 7.66-7.60 (m, 1H), 7.48 (t, J = 5.65 Hz,

1H), 7.39 (q, J1 = 7.60 Hz, J2 = 15.29 Hz, 2H), 7.30 (q, J1 = 6.78 Hz, J2 = 13.76 Hz, 2H), 5.43

(t, J = 8.98 Hz, 1H), 4.47 (dd, J1 = 2.92 Hz, J2 = 5.21 Hz, 1H), 4.29-4.25 (m, 1H), 4.21 (q, J1

= 6.68 Hz, J2 = 14.87 Hz, 1H), 3.95-3.67 (m, 4H), 3.65-3.48 (m, 4H), 2.20-2.11 (m, 1H),

2.06-1.99 (m, 1H), 1.97-1.88 (m, 2H)

- 91 -

II. Figures

II.28 13C-NMR of the N-Fmoc-Gly-Pro-aLuc (8)

([D6]DMSO, 125 MHz) 171.15, 171.05, 167.43, 164.43, 159.04, 156.55, 148.58, 143.86,

140.70, 138.38, 136.28, 127.61, 127.08, 125.27, 124.20, 120.10, 119.64, 111.52, 78.11, 65.70,

60.47, 46.62, 45.92, 42.72, 34.78, 29.28, 24.52

- 92 -

II. Figures

II.29 ESI mass spectrum of the N-Fmoc-Gly-Pro-aLuc (8)

656.0= [M+H]+, 1311.5= [2M+H]+

- 93 -

II. Figures

II.30 RP-HPLC profile of the N-Fmoc-Gly-Pro-aLuc (8)

70-100% B in 15 min

tR= 12.608 min

- 94 -

II. Figures

II.31 ESI mass spectrum of the N-Boc-6-amino-2-cyanobenzothiazole (9)

276.1= [M+H]+

- 95 -

II. Figures

II.32 RP-HPLC profile of the N-Boc-6-amino-2-cyanobenzothiazole (9)

0-100% B in 30 min + 100% B in 5 min

tR= 29.708 min

- 96 -

II. Figures

II.33 1H-NMR spectrum of the N-Boc-aLuc (10)

([D6]DMSO, 500 MHz) d 9.83 (s, 1H), 8.41 (s, 1H), 8.03 (d, J =8.93 Hz, 1H), 7.53 (d, J =

8.99 Hz, 1H), 5.42 (t, J = 9.01 Hz, 1H), 3.73 (dt, J1 =10.52 Hz, J2 = 45.0 Hz, 2H), 1.50 (s, 9H)

- 97 -

II. Figures

II.34 13C-NMR spectrum of the N-Boc-aLuc (10)

([D6]DMSO, 125 MHz) d 171.19, 164.41, 158.32, 152.75, 147.89, 139.17, 136.53, 124.17,

118.87, 109.72, 79.73, 78.14, 34.73, 28.08

- 98 -

II. Figures

II.35 ESI mass spectrum of the N-Boc-aLuc (10)

380.0= [M+H]+

- 99 -

II. Figures

II.36 Chiral HPLC profile of enantiomeric mixture of the N-Boc-6-amino-luciferin

MeOH/AcOH/TEA (100/0.1/0.1 v/v/v), isocratic elution

tR1= 5.56 min: Boc-6-amino-L-luciferin

tR2= 8.52 min: Boc-6-amino-D-luciferin

- 100 -

II. Figures

II.37 Chiral HPLC profile of the untreated N-Boc-aLuc (10)

MeOH/AcOH/TEA (100/0.1/0.1 v/v/v), isocratic elution

tR= 8.01 min: N-Boc-aLuc

- 101 -

II. Figures

II.38 The standard error of the mean (SEM) values of Figure 7A

SEM was calculated by Microsoft Excel from triplicate values obtained as described in IV.

Biological Investigation.

(N-Z-DEVD-aLuc= N-Z-Asp-Glu-Val-Asp-aLuc, 8)

- 102 -

II. Figures

II.39 Ac-915 induces the activation of caspase-3 in U87 cells (A).

Representative FL1-FSC dot plots (B) arithmetic means of percentages ± SEM of cells with

active caspase-3 show data of cells treated with Ac-915 with the indicated concentrations

(μM) ont he graph for 72 hours. Active caspase-3 was analyzed by flow cytometry as

described in IV. Biological Investigation.

- 103 -

III. Materials and methods

III. MATERIALS AND METHODS

III.1 Materials

2-chlorobenzothiazole, HATU, TCFH, DCC, D-Cys∙HCl∙H2O and triphosgene were obtained from

AK Scientific Inc. (Union City, CA, USA). Z-Asp(OtBu)-OH and COMU were sourced from

Bachem (Bubendorf, Switzerland). N-Fmoc-amino acids were purchased from Orpegen

(Heidelberg, Germany) and Bachem (Bubendorf, Switzerland). N-Fmoc-Gly-Pro-OH was

purchased from Iris Biotech GmbH (Marktredwitz, Germany), Wang resin from Rapp Polymere

GmbH (Tuebingen, Germany), TFFH from Fluorochem Ltd (Hadfield, UK). The HOBt was

sourced from Carbosynth Ltd (Compton, UK), TFA gradient grade from VWR International

(Radnor, PA, USA). The following reagents were purchased from Sigma-Aldrich (St. Louis, MO,

USA): Deoxo-Fluor Reagent, PBS, trypsin, HEPES, CHAPS, DTT, EDTA, DMEM-F12,

penicillin, streptomycin, 0.1% saponin, Endoproteinase Pro-C, DTT and Bovine Serum Albumin.

Alexa Fluor® 488 was bought from Thermo Fisher Scientific (Waltham, MA, USA). C150 and

Ac-915 were synthesized by Avidin Ltd., (Szeged, Hungary).

III.2 Chemical Methods

III.2.1 2-chloro-6-nitrobenzothiazole synthesis (1)

438 mL cc H2SO4 was cooled to 10C in a 2-litre triple-neck round-bottomed flask. 100 g (0.59

mol) 2-chlorobenzothiazole was dripped to the H2SO4 over a period of 2 hours, meanwhile the

reaction mixture was stirred vigorously and the temperature was held under 15C. 66 g (0.66 mol)

powdered KNO3 was added to the reaction mixture in small quantities in 45 minutes, the

temperature was still kept under 15C. Then the reaction mixture was allowed to warm up to room

temperature, and stirring was continued at room temperature for 2 hours. It was poured into 4 litres

of ice and water. Yellow precipitation formed, which was filtered and washed until the pH of the

filtrate became neutral. The crystalline compound was dried at room temperature, followed by its

recrystallization from EtOAc in order to get rid of 2-chloro-5-nitrobenzothiazole as the single side

product. The resulting material was a pale yellow crystal, its weight was 104.90 g (0.49 mol), yield

83%, mp 191-192C (EtOAc), (lit. mp 190-191C 28). 1H-NMR (CDCl3, 500 MHz) 8.77 (s, 1H),

8.41 (d, J= 9.0 Hz, 1H), 8.10 (d, J= 9.0 Hz, 1H) (II.1). The spectral data matched that in the

- 104 -


literature 59, m/z (TSQ): 213.93 [M-H]- (II.2), RP-HPLC: 70-100% B in 15 min + 100% B in 5 min,

tR= 7.899 min (II.3), TLC: n-hexane/dioxane = 2:1; Rf: 0.42.

III.2.2 6-amino-2-chlorobenzothiazole synthesis (2)

25 g (0.12 mol) 2-chloro-6-nitrobenzothiazole (1) packed in a paper cup and 500 mL EtOAc, 30 g

NH4Cl (0.56 mol), 200 mL water and 20 g reduced Fe powder in a 1-litre round-bottomed flask

was put in a Soxhlet apparatus and heated under reflux for eight hours while continuously stirring

the mixture. This way, the continuous dissolution of the starting material, which has low solubility

in EtOAc, was ensured, allowing for unmonitored and unmanaged operation while we could

efficiently recycle a small amount of EtOAc to dissolve a larger amount of 2-chloro-6nitro-

benzothiazole (1), thus making the procedure more economical. In order to get rid of the remaining

water/NH4Cl/Fe-powder as lower part, the upper part EtOAc layer was decanted, and this process

was repeated twice with 100 mL EtOAc, respectively. Decantation was employed instead of using

a separatory funnel because the lower aqueous phase too viscous. We had no iron waste as in our

method no chemical transformation of the iron occurred, the iron was 100% recyclable: the iron

powder was filtered off, and then washed on a Büchner funnel with distilled water. The combined

organic phase was dried over anh Na2SO4, filtered and evaporated on rotary evaporator. The

resulting material was a yellow crystal, its weight was 18.90 g (0.10 mol), yield 88%, mp 154-156

C (EtOAc) (lit.155-157C 28). 1H-NMR ([D6]DMSO, 600 MHz ) (d, J = 8.4 Hz, 1H), 7.04 (d, J

= 2.4 Hz, 1H), 6.78 (dd, J1 = 1.8 Hz, J2 = 8.4 Hz, 1H), 5.53 (bs, 2H) (II.4), 13C-NMR ([D6]DMSO,

150 MHz) 148.24, 145.51, 142.00, 137.86, 123.13, 115.51, 104.18 (II.5) The spectral data

matched that in the literature.60 m/z (ESI): 185.0 [M+H]+ (II.6), RP-HPLC: 5-80% B in 25 min + 3

min up to 100% B + 100% B in 5 min, tR= 11.527 min (II.7), TLC: n-hexane/dioxane = 2:1; Rf:

0.76.

III.2.3 6-amino-2-cyanobenzothiazole synthesis (3)

In order to get a suspension, 6.1 g (93.0 mMol) KCN was sonicated in 400 mL DMAA for 3 x 15

minutes. The suspension was heated in an oil-bath at 98-100C under argon atmosphere and then

6.86 g (37.15 mMol) 6-amino-2-chlorobenzothiazole (2), dissolved in 20 mL DMAA, was dripped

to this reaction mixture over a period of 50 minutes. This resulting mixture was heated in an oil-

bath at 110C and stirred continuously under argon atmosphere for 12 hours. After 12 hours stirring

- 105 -


there was still starting material in the mixture. An increased conversion from the starting ratio of

2.5:1 for KCN/6-amino-2-chlorobenzothiazole (2) to the ratio of 3.4:1 was achieved by adding 2.20

g (33.80 mMol) of KCN. This was followed by 5 hours stirring under the conditions described

above, and after that procedure the remaining amount of 6-amino-2-chlorobenzothiazole (2) was

insignificant. The reaction mixture was poured on a mixture of 200 g ice, 400 mL 1M KH2PO4 and

300 mL diethyl ether. The organic phase was separated from the aqueous phase. The latter was

extracted with 2 x 250 mL diethyl ether, then with 2 x 200 mL EtOAc. The combined organic

phases were washed with 2 x 300 mL water and 1 x 300 mL brine, dried over anh Na2SO4 and

concentrated on rotary evaporator. The resulting material was a pale brown solid, its weight was

6.4g (crude). The material was recrystallized from acetone, and the impurities were removed by

adding activated charcoal to the solution. The weight of the desired purified material was 5.09 g

(29.10 mMol), yield 78%, mp 218-219°C (EtOAc) (lit. mp 216-218C 15). 1H -NMR ([D6]DMSO,

600 MHz ) (d, J = 9.0 Hz, 1H), 7.22 (d, J = 1.8 Hz, 1H), 7.01 (dd, J1 = 1.8 Hz, J2 = 9.0 Hz, 1H),

4.40 (bs, 2H) (II.8), 13C-NMR (150 MHz, ([D6]DMSO) 150.06, 144.10, 138.75, 128.86, 125.62,

118.19, 114.71, 103.54 (II.9) The spectral data matched that in the literature. 23 m/z (ESI) 176.0

[M+H]+ (II.10), RP-HPLC: 5-80% B in 25 min + 3 min up to 100% B + 5 in 100% B, tR= 16.692

min (II.11), TLC: n-hexane/dioxane = 2:1; Rf: 0.48. The remaining KCN was reacted with KH2PO4

in order to get non-toxic KOCN.

III.2.4 N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole synthesis (4)

6.30 g (15.30 mMol, 1.5 equiv) N-Fmoc-Asp(OtBu)-OH, which was previously dried in a vacuum

desiccator, and 4.30 g (15.30 mMol, 1.5 equiv) TCFH were solved in 35 mL anh DCM. The mixture

was stirred for 60 minutes at room temperature. First 3.06 mL (18.36 mMol, 1.8 equiv) DIPEA,

then 1.79 g (10.20 mMol, 1 equiv) 6-amino-2-cyanobenzothiazole (3), which was previously dried

in a vacuum desiccator, were added. Further 200 mL dry DCM was added to get complete

dissolution of the materials. After stirring the reaction mixture overnight at room temperature, it

was transferred into a separatory funnel and washed with water (2 x 30 mL), with saturated

NaHCO3-solution (2 x 30 mL), then with water again (2 x 30 mL), and finally with brine (2 x 30

mL). It was dried over anh Na2SO4, finally concentrated on rotary evaporator. The resulting crude

material was a yellowish-brown powder, its weight was 6.17 g. RP-HPLC analysis showed a yield

- 106 -


of 73%. m/z (TOF) 569.2070 [M+H]+ (II.12), RP-HPLC (for the purified compound): 50-100% B

in 25 min, tR= 10.591 min (II.13), TLC: EtOH/toluene 50:7.5; Rf: 0.58.

III.2.5 N-Fmoc-Asp(OtBu)-aLuc synthesis (5)

2.96 g (5.20 mMol) N-Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole (4) was dissolved in the

mixture of 35 mL MeOH and 20 mL THF. 1.37 g (7.80 mMol) D-cysteine∙HCl∙H2O, dissolved in

10 mL distilled water, was added to the mixture at room temperature under argon atmosphere,

while stirring continuously under pH control (starting pH: 1.67). After 20 minutes stirring at room

temperature 16 mL 5% (m/m) NaHCO3 was added dropwise over a period of one hour to the

mixture in order to release cysteine from its salt while continuously checking pH. Reaching pH

2.5, a fine, yellow solid material, N-Fmoc-Asp(OtBu)-aLuc free carboxylic acid, started to

precipitate. At pH 6.1, this material started to dissolve, and at pH 7.36, it dissolved completely.

Here then N-Fmoc-Asp(OtBu)-aLuc formed Na-salt, which dissolved under the basic conditions.

After an additional 20 minutes stirring at room temperature the organic solvent was removed under

reduced pressure. Water and MeOH forms an azeotrope, and the two solvents were therefore

removed together through the distillation. Due to the decrease in the concentration of the water,

from the remaining aqueous solution a yellow solid material, N-Fmoc-Asp(OtBu)-aLuc Na-salt,

precipitated. This, however, was just an irrelevant event during the process, as our goal was the

removal of the MeOH. The aforementioned precipitate is water soluble, so then it was dissolved

again in 20 mL water and extracted with 1 x 15 mL EtOAc in order to get rid of possible impurities.

Having dropped this solution on a mixture of ice and AcOH (adjusted to pH 3), a fine yellow

precipitate formed, N-Fmoc-Asp(OtBu)-aLuc free carboxylic acid. It was allowed to settle for 10

minutes, filtered and washed with 3 x 10 mL water, then air-dried to constant weight, which was

2.83 g (4.20 mMol), yield 81%. m/z (TOF) 673.1882 [M+H]+ (II.14), RP-HPLC: 50-100% B in 25

min, tR= 21.046 min (II.15), TLC: toluene/EtOH 50:30 saturated with water, Rf: 0.58.

III.2.6 Attachment of N-Fmoc-Asp(OtBu)-aLuc to solid support

Solid phase peptide synthesis was performed manually by using a solid phase vessel attached to a

rotating apparatus. 0.127 g (0.10 mMol, 1 equiv) p-alkoxybenzyl alcohol resin was allowed to swell

in anh DCM for 20 minutes. After the removal of the DCM, 0.202 g (0.30 mMol, 3 equiv) N-Fmoc-

Asp(OtBu)-aLuc, 0.062 g (0.30 mMol, 3 equiv) DCC, 0.041 g (0.30 mMol, 3 equiv) HOBt and

- 107 -


0.037 g (0.10 mMol, 1 equiv) DMAP, dissolved in 10 mL anh DCM, was added to the resin. The

coupling reaction was shaken for 3 hours at room temperature. After the removal of the coupling

mixture, the resin was rinsed with DCM (3 x 10 mL), MeOH (1 x 10 mL) and then with DCM (3

x 10 mL) again. The coupling reaction was repeated with the soln of 0.067 g (0.10 mMol, 1 equiv)

N-Fmoc-Asp(OtBu)-aLuc, 0.062 g (0.30 mMol, 3 equiv) DCC, 0.041 g (0.30 mMol, 3 equiv) HOBt

and 0.037 g (0.10 mMol, 1 equiv) DMAP in 5 mL anh DCM at room temperature for 2 hours. The

resin was drained and rinsed with DCM (3 x 10 mL), MeOH (1 x 10 mL), DCM (3 x 10 mL), then

dried to constant weight.

Determination of load: 5 mg of dried loaded resin was treated with a mixture of TFA/water (500

L, with the ratio of 95:5) for 1 hour at room temperature. This was followed by the addition of

500 L water to the cocktail, which was then filtered off. 10 L from the filtrate was injected to

analytical RP-HPLC and the area of the N-Fmoc-Asp-aLuc on the resulted chromatogram was

compared with the area of 10 uL N-Fmoc-Asp-aLuc stock solution with the concentration of 1

mg/mL. The resulting load was 47.8%.

III.2.7 N-Z-Asp-Glu-Val-Asp-aLuc (6) synthesis

III.2.7.1 Fmoc deprotection

Fmoc deprotection was carried out by suspending the resin in 20% (v/v) piperidine/DMF (5 mL)

and agitating the vessel at room temperature for 2 x 10 minutes. The suspension was then filtered

and the resin was washed with DMF (3 x 5 mL), MeOH (3 x 5 mL), DMF (3 x 5 mL).

III.2.7.2 SPPS peptide coupling

N-Fmoc-Val-OH (3 equiv), DCC (3 equiv) and HOBt (3 equiv) dissolved in DMF were added to

the previously swollen and Fmoc-deprotected loaded resin (1 equiv). The resulting suspension was

agitated at room temperature for 2 hours and the resin was then rinsed with DMF (3 x 5 mL),

MeOH (3 x 5 mL), DMF (3 x 5 mL).

The same procedure was carried out with N-Fmoc-Glu(OtBu)-OH (3 equiv) and N-Z-Asp(OtBu)-

OH, (3 equiv). The presence or absence of the N-free amino group was monitored using the Kaiser

test.

- 108 -


III.2.7.3 Cleavage of peptide from the resin

The peptide-resin was treated with a solution of TFA/water (95:5 v/v) for 2 hours at room

temperature. After the removal of the cleaving mixture, the resin was rinsed with AcN (3 x 10 mL),

MeOH (1 x 10 mL) and with AcN (3 x 10 mL) again. The resulting material is a yellow liquid,

which was lyophilized afterwards. RP-HPLC for the crude compound: 5-80% B in 25 min + 3 min

up to 100% B + 5 min in 100% B, tR= 18.483 min.

III.2.7.4 Purification of crude peptide

26 mg crude peptide was dissolved in AcOH/water (1.5 mL, with the ratio of 1:1), then filtered,

using a 0.45m nylon filter. Gradient elution was used, 0-60% eluent B in 60 minutes at a 3 mL

min-1 flow rate with detection at 220 nm. Pure fractions were collected and lyophilized to give a

pale yellow material, the weight of which was 11.4 mg (0.013 mMol).1H-NMR (600 MHz,

[D6]DMSO) 10.26 (bs, 1H), 8.61 (s, 1H), 8.42 (d, J = 7.2 Hz, 1H), 8.08 (d, J = 9.0 Hz, 2H), 7.81

(bs, 1H), 7.64 (dd, J1 = 9.0 Hz, J2 = 29.4 Hz, 2H), 7.34 (s, 5H), 5.42 (t, J = 8.4 Hz, 1H), 5.02 (s,

2H), 4.69 (d, J = 7.2 Hz, 1H), 4.34 (dd, J1 = 5.4 Hz, J2 = 29.4 Hz, 2H), 4.12 (bs, 1H), 3.77 (t, J =

10.8 Hz, 1H), 3.68 (dd, J1 = 8.4 Hz, J2 = 11.4 Hz, 1H), 2.77 (bs, 1H), 2.61-2.68 (m, 2H), 2.25-2.46

(m, 3H), 1.76-1.99 (m, 2H), 0.83-0.86 (m, 7H) (II.16 part 1, 2, 3), 13C-NMR (150 MHz,

[D6]DMSO) 174.54, 172.00, 171.61, 171.38, 170.24, 159.62, 156.32, 149.14, 138.66, 137.31,

136.71, 128.83, 128.28, 128.19, 124.65, 120.28, 112.04, 78.64, 66.01, 58.21, 52.53, 51.86, 51.24,

36.77, 36.31, 35.23, 30.98, 30.52, 27.59, 19.53 (II.17 part 1, 2), m/z (ESI) 872.3 [M+H]+ (II.18),

RP-HPLC: 5-80% B in 25 min + 3 min up to 100% B + 5 min in 100% B, tR= 18.555 min (II.19).

III.2.8 N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole synthesis (7)

2.03 g (5.145 mMol, 1.5 equiv) anh N-Fmoc-Gly-Pro-OH and 1.44 g (5.145 mMol, 1.5 equiv) anh

TCFH were solved in 7 mL anh DCM. The mixture was stirred for 60 minutes at room temperature.

First 1 mL (6.174 mmol, 1.8 equiv) DIPEA, then 0.600 g (3.43 mMol, 1 equiv) anh 6-amino-2-

cyanobenzothiazole (3) was added. After stirring the reaction mixture overnight at room

temperature (according to HPLC analysis the conversion was 97%), it was washed with water (2 x

7 mL), with saturated NaHCO3-solution (2 x 7 mL), then with water again (2 x 7 mL), and finally

with saturated NaCl-solution (2 x 7 mL). It was dried over sicc Na2SO4, finally concentrated in

vacuo. The resulting crude material was a pale yellow powder, its weight was 1.28 g, yield

- 109 -


corresponding to the crude product: 68%. m/z (ESI) 552.0 [M+H]+ (II.25). RP-HPLC (for the

purified compound): 50-100% B in 25 min + 3 min up to 100% B +100% B in 5 min, tR1= 8.973

min: N-Fmoc-Gly-Pro-OH, tR2= 17.868 min: N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole (7),

(II.26).

III.2.9 N-Fmoc-Gly-Pro-aLuc synthesis (8)

5.512 g (0.010 mol, 1 equiv.) N-Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole (7) was dissolved

in 25 mL MeOH. 2.634 g (0.015 mol, 1.5 equiv.) D-cysteine∙HCl∙H2O, solved in 19 mL distilled

water, was added to the solution at room temperature, under argon atmosphere, then the mixture

was stirred continuously under pH control (starting pH: 2.27).

After 20 minutes’ stirring at room temperature, 30 mL, 5% (m/m) NaHCO3 was added dropwise

over a period of one hour to the mixture in order to release cysteine from its salt, while checking

pH continuously. Reaching pH 2.6, a fine, yellow solid material, N-Fmoc-Gly-Pro-aLuc free

carboxylic acid, started to precipitate. At pH 6.3, this material started to dissolve, and at pH 7.40,

it dissolved completely. Here the N-Fmoc-Gly-Pro-aLuc formed Na-salt, which dissolved under

these conditions.

After another 20 minutes’ stirring at room temperature, the organic solvent was evaporated. From

the remaining aqueous solution, a pale yellow solid material, N-Fmoc-Gly-Pro-aLuc Na-salt

precipitated partly. This aqueous mixture was extracted with 3 x 15 mL EtOAc, in order to get rid

of possible impurities. The combined organic layers were extracted with saturated NaCl solution.

Having dropped the resulting solution on a mixture of ice and cc HCl, a fine yellow precipitate, N-

Fmoc-Gly-Pro-aLuc free carboxylic acid, formed. It was allowed to settle for 10 minutes, filtered

and washed with 2 x 5 mL water, then air-dried to constant weight, which was 5.115 g (7.80 mMol),

yield corresponding to the crude product: 78%. 1H NMR (500 MHz, [D6]DMSO) δ 10.39 (s, 1H),

8.59 (t, J = 15.85 Hz, 1H), 8.09 (d, J = 8.98 Hz, 1H), 7.88 (d, J = 7.43 Hz, 2H), 7.71 (d, J = 7.48

Hz, 2H), 7.66-7.60 (m, 1H), 7.48 (t, J = 5.65 Hz, 1H), 7.39 (q, J1 = 7.60 Hz, J2 = 15.29 Hz, 2H),

7.30 (q, J1 = 6.78 Hz, J2 = 13.76 Hz, 2H), 5.43 (t, J = 8.98 Hz, 1H), 4.47 (dd, J1 = 2.92 Hz, J2 =

5.21 Hz, 1H), 4.29-4.25 (m, 1H), 4.21 (q, J1 = 6.68 Hz, J2 = 14.87 Hz, 1H), 3.95-3.67 (m, 4H),

3.65-3.48 (m, 4H), 2.20-2.11 (m, 1H), 2.06-1.99 (m, 1H), 1.97-1.88 (m, 2H) (II.27). 13C NMR (125

MHz, [D6]DMSO) δ 171.15, 171.05, 167.43, 164.43, 159.04, 156.55, 148.58, 143.86, 140.70,

138.38, 136.28, 127.61, 127.08, 125.27, 124.20, 120.10, 119.64, 111.52, 78.11, 65.70, 60.47,

- 110 -


46.62, 45.92, 42.72, 34.78, 29.28, 24.52 (II.28). m/z (ESI) 656.0 [M+H]+ (II.29). RP-HPLC: 70-

100% B in 15 min, tR= 12.608 min (II.20). TLC: toluene/EtOH 50:30 saturated with water, Rf:

0.44.

III.2.10 Purification of crude N-Fmoc-Gly-Pro-aLuc (8)

160 mg crude peptide (45% desired material content, 72 mg) was dissolved in 1 mL DMF, then

filtered, using a 0.45m nylon filter. Gradient elution was used, 40-70% eluent B in 60 minutes at

a 4 mL min-1 flow rate with detection at 220 nm. Pure fractions were collected and lyophilized to

give a pale yellow material, the weight of which was 23 mg (0.035 mmol), yield corresponding to

the isolated pure product: 32%.

III.2.11 N-Boc-6-amino-2-cyanobenzothiazole synthesis (9)

0.81 g (4.62 mMol, 1 equiv) 6-amino-2-cyanobenzothiazole (3) was dissolved in 60 mL anh EtOAc

in a round-bottomed flask, then 2.45 mL (14.10 mMol, 3 equiv) DIPEA was added. 2.04 g (6.87

mMol, 1.5 equiv) triphosgene was added in small amounts, over a period of 30 minutes at room

temperature while vigorously stirring the solution. The mixture was stirred at 60 C for a period of

60 minutes. In the following step, the flask was put in cold water in order to cool down the solution

to room temperature. A pale-yellow intermediate, (2-cyanobenzo[d]thiazol-6-yl)carbamic chloride

precipitated. At room temperature 4.75 mL (50.10 mMol, 11 equiv) tBuOH was added to this

material, the mixture was heated again to 60 C in an oil bath, then stirred over a period of 1 hour.

During processing, the flask was first put in cold water in order to cool down the solution. In order

to neutralize the acidic solution, it was poured into a mixture of ice and 28 m/m% NaOH, then the

organic phase was separated from the aqueous phase. The latter was extracted with 3 x 30 mL

EtOAc. The combined organic phases were washed with 3 x 30 mL water and 1 x 30 mL saturated

NaCl, dried over sicc Na2SO4, filtered and concentrated on a rotary evaporator. The resulting

material was a pale brown solid, its weight was 0.995 g, yield corresponding to the crude product:

74%. m/z (ESI) 276.1 [M+H]+ (II.31). RP-HPLC: 0-100% B in 30 min + 100% B in 5 min, tR=

29.708 min (II.32). TLC: toluene/EtOH 50:7.5, Rf: 0.84.

- 111 -


III.2.12 N-Boc-aLuc synthesis (10)

0.995 g crude N-Boc-6-amino-2-cyanobenzothiazole (9) was dissolved in the mixture of 20.5 mL

MeOH and 2.5 mL THF. 1.58 g (9.05 mmol) D-cysteine∙HCl∙H2O, dissolved in 10 mL distilled

water, was added to the mixture at room temperature under argon atmosphere, while stirring

continuously under pH control (starting pH: 0.66). After 20 minutes stirring at room temperature,

21.6 mL 5 m/m% NaHCO3 was added to the mixture dropwise over a period of 75 minutes in order

to release the cysteine from its salt, while continuously monitoring pH. Reaching pH 2.99, a fine,

yellow solid material, N-Boc-aLuc free carboxylic acid, started to precipitate. At pH 7.13, this

material started to dissolve, and at pH 7.24, it dissolved completely. Here the N-Boc-6-aLuc formed

Na-salt, which dissolved under the basic conditions. After an additional 10 minutes stirring at room

temperature the organic solvents was removed under reduced pressure. Due to the decrease in the

concentration of the water, from the remaining aqueous solution a yellow solid material, N-Boc-

aLuc Na-salt, precipitated. This, however, was an irrelevant event during the process, as our goal

was the removal of the organic solvents. The aforementioned precipitate is water soluble, so then

it was dissolved again in 20 mL water and extracted with 1 x 15 mL methyl tert-butyl ether in order

to get rid of possible impurities. Having dropped the aqueous phase on a mixture of ice and AcOH

(adjusted to pH 3.43), a fine yellow precipitate formed, N-Boc-aLuc free carboxylic acid. It was

allowed to settle for 10 minutes, filtered and washed with 3 x 10 mL water, then air-dried to

constant weight, which was 0.846 g, yield corresponding to the crude product: 85%. TLC:

toluene/ethanol 50:15 saturated with water, Rf: 0.15.

III.2.13 Purification of N-Boc-aLuc (10)

40 mg crude material (85% desired material content, 34 mg) was dissolved in 1 mL DMF, then

filtered, using a 0.45μm nylon filter. Gradient elution was used, 20-70% eluent B in 50 minutes at

a 3 mL min-1 flow rate with detection at 220 nm. Pure fractions were collected and lyophilized to

give a pale yellow material, the weight of which was 17 mg (0.045 mMol), yield corresponding to

the isolated pure product: 44%. 1H-NMR (500 MHz, [D6]DMSO) d 9.83 (s, 1H), 8.41 (s, 1H), 8.03

(d, J =8.93 Hz, 1H), 7.53 (d, J = 8.99 Hz, 1H), 5.42 (t, J = 9.01 Hz, 1H), 3.73 (dt, J1 =10.52 Hz, J2

= 45.0 Hz, 2H), 1.50 (s, 9H) (II.33). 13C-NMR (125 MHz, [D6]DMSO) d 171.19, 164.41, 158.32,

- 112 -


152.75, 147.89, 139.17, 136.53, 124.17, 118.87, 109.72, 79.73, 78.14, 34.73, 28.08 (II.34). m/z

(ESI) 380.0= [M+H]+ (II.35)

The product’s optical purity was proved with chiral chromatography the following way:

5 mL 1M Na2CO3 solution was added to the previously purified product (5 mg, 0.013 mMol) and

the mixture was stirred for 10 minutes at room temperature at pH 9 in order to get an enantiomeric

mixture. The resulting material was examined with chiral chromatography, which showed that the

L-enantiomer was eluted at 5.5 min, while the D-enantiomer was eluted at 8.5 min (II.36). Then the

untreated purified product was also chromatographed, in order to prove its optical purity. Having

found it optically pure (II.37). Its optical rotation was also measured: [𝛼]𝐷20 = +5 (c 0.250, EtOH).

III.3. Analytical methods

TLC was performed on silica gel plates 60 F254 from Merck (Darmstadt, Germany). Melting points

were determined using Büchi (Flawil, Switzerland) melting point apparatus Model B-545. pH

values were measured with a Hanna HI 8424 pH meter. Cellulose extraction thimbles were

purchased from Whatman (Maidstone, UK). Analytical reversed-phase high-performance liquid

chromatography was performed on an Agilent 1200 series separations module with diode array and

multiple wavelength detector (Waldbronn, Germany), with a Luna C18(2) 100Å column (10 µm,

250 x 4.6 mm) Phenomenex, (Torrance, CA, USA). The experiments were carried out at room

temperature with a flow rate maintained at 1.2 mL min-1 at 220 nm wavelength (mobile phases

solvent A: 0.1% TFA in Milli-Q water and solvent B: 0.1% TFA in AcN) using gradient elution.

Separation was achieved on a Shimadzu (Kyoto, Japan) semi-preparative system with a Jupiter

C18 300Å column (10 µm, 250 x 21.20 mm), also from Phenomenex (mobile phases solvent A:

0.1% TFA in Milli-Q water and solvent B: 0.1% TFA in AcN) using gradient elution. Chiral

chromatography was carried out on a Waters HPLC system consisting of an M-600 low-pressure

gradient pump, an M-996 photodiode-array detector and an Empower 2 data manager software

(Waters Chromatography, Milford, MA, USA) with a Chirobiotic T 250 x 4.6 mm ID column

(Astec, Whippany, USA) The experiments were carried out at room temperature with a flow rate

maintained at 0.8 mL min-1 at 250 nm wavelength. MeOH/AcOH/TEA (100/0.1/0.1 v/v/v) was

used as the mobile phase solvent in isocratic elution mode. Optical rotation was measured with a

Perkin-Elmer 341 polarimeter. Mass spectrometry data for the compound 1 were collected on a

- 113 -


Finnigan MAT TSQ 7000 (Waltham, MA, USA) instrument, operating with APCI in negative ion

mode, data for materials 2, 3, 6, 7, 8, 9, 10 were collected on Waters (Milford, MA, USA) SQ

Detector with electrospray ionization (ESI) in positive ion mode; data for compounds 4 and 5 were

recorded with Waters Q-TOF Premier Mass Spectrometer. 1H NMR and 13C NMR spectra were

recorded using a Bruker DR X 500 spectrometer at 600 MHz and 150 MHz, respectively in

[D6]DMSO. Chemical shifts were reported on the scale and J values were given in Hz.

- 114 -

IV. Biological Investigation

IV. BIOLOGICAL INVESTIGATION

IV.1 N-Z-Asp-Glu-Val-Asp-aLuc (6) biochemical assay

Caspase-3 and the assay buffer were used from the caspase-3 inhibitor drug screening kit

(BioVision, Milpitas, CA, USA). Caspase-3 was used in ten-fold serial dilution starting from 227

mU to 22.7 μU/reaction. N-Z-Asp-Glu-Val-Asp-aLuc (6) substrate was applied in 100 μM to 1 μM

in 25 μL final reaction volume in a black plastic microtiter plate. The effect of the pan-caspase

inhibitor Z-Val-Ala-Asp-fmk was tested in equimolar ratio of N-Z-Asp-Glu-Val-Asp-aLuc (6) at

10 μM with 2.27 mU/reaction caspase-3. After 45 minutes incubation at 37°C we added 25 μL

luminescence detection reagent to each well. Luminescence was recorded as cps by a plate reader

within 5 minutes. Blank wells contained each component except caspase-3. Presented values were

blank-subtracted.

IV.2 N-Z-Asp-Glu-Val-Asp-aLuc (6) cellular assay

A549 non-small cell lung carcinoma cells were purchased from the ATCC (Manassas, VA, USA)

and U87-Luc glioblastoma cell line from Perkin Elmer (Waltham, MA, USA). Complete protease

inhibitor cocktail was from Roche Basel, Switzerland.

A549 non-small cell lung carcinoma cells (2 x 106) were plated in 60 mm dishes (Corning, NY,

USA) in DMEM-F12 media (Gibco BRL, Gaithersburg, MD, USA).. After cell attachment (24h),

the cells were treated with curcumin analogue C150 (5 μM to 1.25 μM) in order to induce apoptosis

in 5 μL final volume.64 After 24h incubation supernatant was harvested and kept on ice. Cells were

washed with PBS and trypsinized (5 minutes, 37°C). Supernatant, washing PBS and media blocked

trypsin were mixed and centrifuged down (5 minutes, 4°C, 1800 g). Lysis buffer was diluted with

distilled water (five times concentrated lysis buffer: 250 mM HEPES, pH 7.4, 25 mM CHAPS, 25

mM DTT, and 50 μL 1x concentration lysis buffer was added to the pellet, resuspended and kept

on ice for 15 minutes. Samples were centrifuged down for 10 minutes (4°C, 11000 g). Supernatant

of the lysate was harvested and kept on ice for analysis. N-Z-Asp-Glu-Val-Asp-aLuc (6) was

dissolved in DMSO at 10 mM and used in 10 μM in the assay. Assay buffer was diluted in distilled

water (ten times concentrated assay buffer: 200 mM HEPES, pH 7.4, 1% (v/v) CHAPS, 50 mM

DTT, 20 mM EDTA, 95 μL 1x concentration assay buffer containing 10 μM N-Z-Asp-Glu-Val-

Asp-aLuc (6) was measured in 96-well tissue culture plate (Corning) and 5 μL lysate was assayed

- 115 -


in triplicates to detect luminescence proportional to caspase-3 activity. Z-Val-Ala-Asp-fmk from

Calbiochem (San Diego, CA, USA) was used as internal control for caspase-3 inhibition. Blank

contained 5 μL assay buffer instead of the lysate. After 4 hours reaction 50 μL reaction mixture

was measured into black plastic microtiter plate and 50 μL luminescent detection reagent was

added. Luminescence was recorded as cps by a plate reader within 5 minutes. Presented values

were blank-subtracted.

IV.3 N-Fmoc-Gly-Pro-aLuc (8) assay

N-Fmoc-Gly-Pro-aLuc (8) was tested in bioluminescence-based enzyme activity assays with three

enzymes. POP/PREP and recombinant human FAP alpha were obtained from R&D Systems

(Minneapolis, MN, USA) and a bacterial non-specific endoproteinase Pro-C was from Sigma

(Budapest, Hungary). Enzymes were used at equivalent protease activity in ten-fold serial dilution

starting from 32 fmol/min/reaction to 320 pmol/min/reaction. Assay buffer for POP/PREP and

Endoproteinase Pro-C contained 25 mMol tris(hydroxymethyl)aminomethane HCl-salt pH 7.4, 250

mMol NaCl, 2.5 mMol 1,4-dithiothreitol and the assay buffer for FAP contained 50 mMol Tris∙HCl

pH 7.4, 1 M NaCl, 1 mg/mL BSA. The N-Fmoc-Gly-Pro-aLuc (8) substrate was applied in 1 μmol

to 100 μmol in 25 μL final reaction volume in a black plastic microtiter plate (Tomtec). The effect

of protease inhibition was prepared by dissolving one tablet in 2 mL POP/PREP buffer and used

in 2.5-fold dilution in each reaction with 10 μmol N-Fmoc-Gly-Pro-aLuc (8) and 32

pmol/min/reaction protease activity. After 2 hours incubation at 37°C 25 μL Luminescence

Detection Reagent was added to each well. Luminescence was recorded as cps by a plate reader

(Perkin Elmer Wallac VICTOR 1420 (Waltham)) within 5 minutes. Blank wells contained each

component except POP/PREP or recombinant human FAP enzymes. Presented values were blank-

subtracted.

IV.4 N-Z-Asp-Glu-Val-Asp-aLuc (6) in vivo assay

Male SCID mice (6 weeks old, 22–24 g body weight) were housed in sterile cages at Avidin Ltd.

The mice were fed autoclaved food and sterile water ad libitum. For inoculation, the U87-Luc cells

were trypsinized, washed and resuspended in sterile PBS. The mice were injected subcutaneously

with this suspension (3 x 106 cells in 0.2 mL), in the dorsal region, unilaterally. All operative

procedures and animal care conformed strictly to the Hungarian Council on Animal Care

- 116 -


guidelines. Approval provided by Head of Food Chain Safety and Animal Health of the Csongrad

County Government Office. Document number: CSI/01/126/2013; valid until 8th of January 2018.

18 days after inoculation the mice were treated with Ac-915, a lipid droplet binding thalidomide

analogue inducing oxidative stress and apoptosis in glioblastoma cells.66 Ac-915 was dissolved in

DMSO:solutol 3:1 mixture, then diluted in PBS four times and injected i.p. at a 20 mg/kg dose

except negative control mice which were injected by PBS only. Six hours after drug administration,

to monitor apoptosis induction, the mice were injected i.p., with 50 mg/kg N-Z-Asp-Glu-Val-Asp-

aLuc (6) in PBS, followed by anesthetization in 2–3% isoflurane atmosphere. 30 min after the

injection of the substrate, the mice were imaged using a charge coupled device camera in the IVIS

100 imaging instrument.

IV.5 Statistics

Statistical significance was calculated with unpaired t-test (two-tailed, homoscedastic) between

untreated and one treated sample. Each point represents the average of 3 wells ± SEM. Values are

blank-subtracted (blank = no caspase). *p<0.05; ** p<0.01; ***p<0.001.

V. RELATED ARTICLES

ORIGINAL RESEARCHpublished: 19 April 2018

doi: 10.3389/fchem.2018.00120

Frontiers in Chemistry | www.frontiersin.org 1 April 2018 | Volume 6 | Article 120

Edited by:

Ramon Rios,

University of Southampton,

United Kingdom

Reviewed by:

Jun Wang,

University of Arizona, United States

Bruno Linclau,

University of Southampton,

United Kingdom

*Correspondence:

Anita K. Kovács

[email protected]

Gábor K. Tóth

[email protected]

Specialty section:

This article was submitted to

Organic Chemistry,

a section of the journal

Frontiers in Chemistry

Received: 22 November 2017

Accepted: 30 March 2018

Published: 19 April 2018

Citation:

Kovács AK, Hegyes P, Szebeni GJ,

Nagy LI, Puskás LG and Tóth GK

(2018) Synthesis of

N-peptide-6-amino-D-luciferin

Conjugates. Front. Chem. 6:120.

doi: 10.3389/fchem.2018.00120

Synthesis ofN-peptide-6-amino-D-luciferinConjugates

Anita K. Kovács 1,2*, Péter Hegyes 2, Gábor J. Szebeni 2,3, Lajos I. Nagy 2,

László G. Puskás 2,3 and Gábor K. Tóth 1*

1Department of Medical Chemistry, University of Szeged, Szeged, Hungary, 2 Avidin Ltd., Szeged, Hungary, 3Department of

Genetics, Biological Research Center, Hungarian Academy of Sciences, Szeged, Hungary

A general strategy for the synthesis of N-peptide-6-amino-D-luciferin conjugates has

been developed. The applicability of the strategy was demonstrated with the preparation

of a known substrate, N-Z-Asp-Glu-Val-Asp-6-amino-D-luciferin (N-Z-DEVD-aLuc).

N-Z-DEVD-aLuc was obtained via a hybrid liquid/solid phase synthesis method,

in which the appropriately protected C-terminal amino acid was coupled to

6-amino-2-cyanobenzothiazole and the resulting conjugate was reacted with D-cysteine

in order to get the protected amino acid-6-amino-D-luciferin conjugate, which was then

attached to resin. The resulting loaded resin was used for the solid-phase synthesis

of the desired N-peptide-6-amino-D-luciferin conjugate without difficulties, which was

then attested with NMR spectroscopy and LC-MS, and successfully tested in a

bioluminescent system.

Keywords: bioluminescence, aminoluciferin, conjugate, protease activity, solid-phase peptide synthesis

INTRODUCTION

In the recent years, numerous in vivo and in vitro analytical methods have been developed based onfluorescence and bioluminescence, including immunoassays, gene expression assays, bioimaging,investigation of infectious diseases etc., (Ioka et al., 2016; Kaskova et al., 2016); plate based,high-throughput viability assays addressing the detection of protease activity is in the focus ofintensive research (Kepp et al., 2011). Protease activity can be detected with both fluorescent andbioluminescent detection systems, but with the latter the detection threshold is orders of magnitudelower than that of the fluorescent technique (O’Brien et al., 2005; Hickson et al., 2010; Gilbert andBoutros, 2016).

In the bioluminescent methods, diverse sets of luciferases and their substrates, luciferins havebeen applied in different cellular and animal models (Ioka et al., 2016; Kaskova et al., 2016).Aminoluciferin (aLuc) is a luciferin with its 6-position hydroxyl group substituted with an aminogroup. This modification allows aLuc to form amide bond with a peptide, while retaining thetransport and bioluminescent properties of luciferin, resulting in a good substrate for differentimportant proteases, which can be used for the determination of the enzymatic activity mentionedabove (White et al., 1966).

N-linked peptide-6-amino-D-luciferins can be substrates for different proteases, includingmetalloproteases, chymotrypsin-like, trypsin-like, and caspase-like proteases (O’Brien et al., 2008)They can be used for measuring protease enzyme activity in the following way: the protease enzymeto be measured recognizes the peptide part of the conjugate with the suitable peptide sequence,then cleaves the amide bond between the peptide and the aLuc, thus aLuc is released, which, in the

https://www.frontiersin.org/journals/chemistry

https://www.frontiersin.org/journals/chemistry#editorial-board




https://doi.org/10.3389/fchem.2018.00120

http://crossmark.crossref.org/dialog/?doi=10.3389/fchem.2018.00120&domain=pdf&date_stamp=2018-04-19


https://www.frontiersin.org

https://www.frontiersin.org/journals/chemistry#articles

https://creativecommons.org/licenses/by/4.0/

mailto:[email protected]

mailto:[email protected]

https://doi.org/10.3389/fchem.2018.00120

https://www.frontiersin.org/articles/10.3389/fchem.2018.00120/full

http://loop.frontiersin.org/people/493211/overview



Kovács et al. Synthesis of N-peptide-6-amino-D-luciferin Conjugates

presence of luciferase enzyme, emits light (Figure 1). The activityof the given protease enzyme can be determined from the amountof emitted light, as the emitted light is directly proportional to theactivity of the enzyme (Leippe et al., 2011).

In the ongoing research the authors have developed andoptimized amore efficientmethod for the synthesis ofN-peptide-6-amino-D-luciferin conjugates in general, with a simpler set-up under milder conditions. N-Z-DEVD-aLuc was chosen todemonstrate the applicability of the strategy because it is acommercially available but very expensive compound, whichis used to measure the activity of caspase-3, and consequentlythe efficiency of apoptosis-inducing drugs (Talanian et al., 1997;McStay et al., 2008).

Literary OverviewThe logical method for the synthesis of N-peptide-6-amino-D-luciferin conjugates would start with the synthesis of thekey molecule, 6-amino-2-cyanobenzothiazole. So far very fewmethods have been published for this step (Takakura et al., 2011;Gryshuk et al., 2013; McCutcheon et al., 2015; Hauser et al.,2016; Hsu et al., 2016; see Supplementary Table 1). The keysteps of all these methods are the nitration, the cyanidation andthe NO2 reduction; the methods differ in the starting material,the reagents, the solvents and the order of the transformations.Having examined these methods, it can be seen that they havedisadvantages:

a. A less optimal starting material may require an extratransformation during the synthesis (see SupplementaryTable 2).

b. Certain synthesis routes require too much reagents, someof which are expensive (see Comments in SupplementaryTable 1).

c. The use of an ill-chosen solvent leads to low yield during thechlorine-cyanide exchange (see Supplementary Table 3).

Abbreviations: AcN, acetonitrile; anh, anhydrous; APCI, atmosphericpressure chemical ionization; API, atmospheric pressure ionization;CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate;COMU, 1-cyano-2-ethoxy-2-oxoethylidenaminooxy)dimethylamino-morpholino-carbenium hexafluorophosphate; cps, counts per seconds; DCC,dicyclohexylcarbodiimide; DCM, dichloromethane; Deoxo-Fluor Reagent,bis(2-methoxyethyl)aminosulfur trifluoride; DIPEA, N,N-diisopropylethylamine;DMAA, N,N-dimethylacetamide; DMAP, 4-dimethylaminopyridine; DMEM,Dulbecco’s modified eagle medium; DMEM-F12, Dulbecco’s modified eaglemedium nutrient mixture F-12; DMF, N,N-dimethylformamide; DMSO, dimethylsulfoxide; DTT, 1,4-dithiothreitol; EDTA, disodium ethylenediaminetetraacetatedehydrate; EtOAc, ethyl acetate; FACS, fluorescence activated cell sorter;FCS, fetal calf serum; Fmoc, 9-fluorenylmethoxycarbonyl Fmoc; HATU, 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxidhexafluorophosphate; HEPES, 4-(2-hydroxyethyl)piperazine-1-ethanesulfonicacid; HMPA, hexamethylphosphoric acid triamide; HOBt 1-hydroxybenzotriazole;i.p., intraperitoneally; MeOH, methanol; mp, melting point; PBS, phosphate-buffered saline; RP-HPLC, reversed-phase high-performance liquidchromatography; SEM, standard error of the mean; SPPS, solid phasepeptide synthesis; TCFH, chloro-N,N,N′ ,N′-tetramethylformamidiniumhexafluorophosphate; TFA, trifluoroacetic acid; TFFH, fluoro-N,N,N′,N′-tetramethylformamidinium hexafluorophosphate; THF, tetrahydrofuran;TLC, thin layer chromatography; Z, benzyloxycarbonyl; Z-VAD-fmk,benzyloxycarbonyl-valyl-alanyl-aspartyl-[O-methyl]- fluoromethylketone.

d. A less optimal order of the transformations results in lowyield (by transformation and, consequently, overall). With theoptimal order, however, a transformation on one functionalgroup does not result in a side reaction on the other functionalgroup (see Supplementary Table 4).

Ideally, in the next step the amino group of the 6-amino-2-cyanobenzothiazole is blocked with a protecting group. However,the low nucleophilicity of the amino group makes its protectionproblematic, resulting in very low yield. Therefore, a differentsynthesis route is needed. One method has been published(Gryshuk et al., 2011; see Supplementary Table 5). Insteadof a protecting group, the amino group of the 6-amino-2-cyanobenzothiazole is blocked with the protected C-terminalamino acid of the target sequence, and then the protecting groupis removed from the C-terminal amino acid. In the followingstep, the remaining part of the target sequence is added and theside chain protecting groups of the peptide portion are removed;finally cysteine is added. However, the route has disadvantages:

a. The mixed anhydride method for the acylation is not optimal,because it is not economical.

b. Due to the basic conditions (pH 8) during the cysteineaddition, there is a risk of racemization.

c. The resulting materials are purified twice during the route,which is unnecessary.

d. Yields were not determined, therefore it is difficult to evaluatethe synthesis route.

MATERIALS AND METHODS

Materials2-chlorobenzothiazole, HATU, TCFH, DCC, and D-Cys·HCl·H2O were obtained from AK Scientific Inc. (UnionCity, CA, USA). Z-Asp(OtBu)-OH and COMU were sourcedfrom Bachem (Bubendorf, Switzerland). Fmoc-amino acids werepurchased from Orpegen (Heidelberg, Germany) and Bachem(Bubendorf, Switzerland); Wang resin from Rapp PolymereGmbH (Tuebingen, Germany), TFFH from Fluorochem Ltd.,(Hadfield, UK). The HOBt was sourced from CarbosynthLtd (Compton, UK), trifluoroacetic acid gradient grade fromVWR International (Radnor, PA, USA). The following reagentswere purchased from Sigma-Aldrich (St. Louis, MO, USA):Deoxo-Fluor Reagent, PBS, trypsin, HEPES, CHAPS, DTT,EDTA, DMEM-F12, penicillin, streptomycin, 0.1% saponin.Alexa Fluor R© 488 was bought from Thermo Fisher Scientific(Waltham, MA, USA). N-Z-DEVD-aLuc, C150 (Nagy et al.,2015; Hackler et al., 2016), and Ac-915 (Nagy et al., 2013) weresynthesized by Avidin Ltd., (Szeged, Hungary).

Thin layer chromatography was performed on silica gel plates60 F254 from Merck (Darmstadt, Germany). Melting pointswere determined using Büchi (Flawil, Switzerland) melting pointapparatus Model B-545. pH values were measured with a HannaHI 8424 pH meter. Cellulose extraction thimbles were purchasedfrom Whatman (Maidstone, UK). Analytical reversed-phasehigh-performance liquid chromatography was performed on anAgilent 1,200 series separations module with diode array andmultiple wavelength detector (Waldbronn, Germany), with a






FIGURE 1 | The operation of the bioluminescent system.

Luna C18(2) 100 Å column (10µm, 250× 4.6mm) Phenomenex,(Torrance, CA, USA). The experiments were carried out at roomtemperature with a flow rate maintained at 1.2ml min−1 at220 nm wavelength (mobile phases solvent A: 0.1% TFA in Milli-Q water and solvent B: 0.1% TFA in AcN) using gradient elution.Separation was achieved on a Shimadzu (Kyoto, Japan) semi-preparative system with a Jupiter C18 300 Å column (10µm,250× 21.20mm), also from Phenomenex (mobile phases solventA: 0.1% TFA in Milli-Q water and solvent B: 0.1% TFA in AcN)using gradient elution. Mass spectrometry data for the 2-chloro-6-nitrobenzothiazole (1) were collected on a Finnigan MAT TSQ7,000 (Waltham, MA, USA) instrument, operating with APCIin negative ion mode, data for materials 2, 3, 8 were collectedon Waters (Milford, MA, USA) SQ Detector with API massspectrometer in positive ion mode; data for compounds 4 and5 were recorded with Waters Q-Tof Premier Mass Spectrometer.1H NMR and 13C NMR spectra were recorded using a BrukerDR × 500 spectrometer at 600 MHz and 150 MHz, respectivelyin [D6]DMSO. Chemical shifts were reported on the δ scale andJ values were given in Hz.

Caspase-3 and the assay buffer were used from the Caspase-3inhibitor drug screening kit from BioVision (Milpitas, CA, USA).Black plastic microtiter plates were purchased from Tomtec(Budapest, Hungary), Z-VAD-fmk from Calbiochem (SanDiego, CA, USA) and Merck Millipore, (Billerica, MA, USA),luminescence detection reagent from Promega (Madison, WI,USA). 9661S caspase-3 antibody from Cell Signaling Technology(Leiden, TheNetherlands). Luminescence was recorded as countsper seconds by a plate reader Perkin ElmerWallac VICTOR 1420(Waltham, MA, USA). A549 non-small cell lung carcinoma cellswere purchased from the ATCC (Manassas, VA, USA) and U87-Luc glioblastoma cell line from Perkin Elmer (Waltham, MA,USA). Tissue culture dishes (60mm dishes. 96-well plates) werepurchased from Corning (Corning, NY, USA). DMEM and FBSwere purchased fromGibco BRL (Gaithersburg, MD, USA). MaleSCID mice (6 weeks old, 22–24 g body weight) were supplied byInnovo Ltd. (Budapest, Hungary). We used IVIS 100 imaginginstrument from Xenogen, (Alameda, CA, USA). CellQuestTM

software was bought from Becton Dickinson (Franklin Lakes, NJ,USA), GraphPad Prism R© 5 from GraphPad Software (La Jolla,CA, USA).

The mouse studies were performed according to theInstitutional and National Animal Experimentation andEthics Guidelines in possession of an ethical clearance(XXIX./3610/2012), provided by the Head of Foodchain-safety and Animal Health of the Csongrad County GovernmentOffice. Document number: CSI/01/126/2013. Valid until 8th ofJanuary 2018.

MethodsPreparation of 2-chloro-6-nitrobenzothiazole (1)Four hundred and thirty eight milliliter cc H2SO4 was cooledto 10◦C in a 2-liter triple-neck round-bottomed flask. 100 g(0.59mol) 2-chlorobenzothiazole was dripped to the sulfuricacid over a period of 2 h, meanwhile the reaction mixturewas stirred vigorously and the temperature was held under15◦C. Sixty-Six grams (0.66mol) powdered KNO3 was addedto the reaction mixture in small quantities in 45min, thetemperature was still kept under 15◦C. Then the reactionmixturewas allowed to warm up to room temperature, and stirringwas continued at room temperature for 2 h. It was pouredinto 4 liters of ice and water. Yellow precipitation formed,which was filtered and washed until the pH of the filtratebecame neutral. The crystalline compound was dried at roomtemperature, followed by its recrystallization from ethyl acetatein order to get rid of 2-chloro-5-nitrobenzothiazole as the singleside product. The resulting material was a pale yellow crystal,its weight was 104.90 g (0.49mol), yield 83%, mp 191-192◦C(EtOAc), (lit. mp 190–191◦C, Katz, 1951). 1H-NMR (CDCl3,500 MHz) δ 8.77 (s, 1H), 8.41 (d, J = 9.0Hz, 1H), 8.10(d, J = 9.0Hz, 1H) (Supplementary Figure 1). The spectraldata matched that in the literature (Shinde et al., 2006) m/z(TSQ): 213.93 [M-H]− (Supplementary Figure 2), RP-HPLC:70–100% B in 15min + 100% B in 5min, tR = 7.899min(Supplementary Figure 3), TLC: n-hexane/dioxane = 2:1;Rf: 0.42.

Preparation of 6-amino-2-chlorobenzothiazole (2)Twenty-five gram (0.12mol) 2-chloro-6-nitrobenzothiazole (1)packed in a paper cup and 500ml ethyl acetate, 30 g NH4Cl(0.56mol), 200ml water, and 20 g reduced Fe powder in a1-liter round-bottomed flask was put in a Soxhlet apparatusand heated under reflux for 8 h while continuously stirring themixture. This way, the continuous dissolution of the startingmaterial, which has low solubility in ethyl acetate, was ensured,allowing for unmonitored and unmanaged operation while wecould efficiently recycle a small amount of ethyl acetate todissolve a larger amount of 2-chloro-6nitro-benzothiazole, thusmaking the procedure more economical. In order to get ridof the remaining water/NH4Cl/Fe-powder as lower part, theupper part ethyl acetate layer was decanted, and this processwas repeated twice with 100ml ethyl acetate, respectively.Decantation was employed instead of using a separatory funnelbecause the lower aqueous phase too viscous. We had noiron waste as in our method no chemical transformation ofthe iron occurred, the iron was 100% recyclable: the ironpowder was filtered off, and then washed on a Büchner






funnel with distilled water. The combined organic phase wasdried over anhydrous Na2SO4, filtered and evaporated onrotary evaporator. The resulting material was a yellow crystal,its weight was 18.90 g (0.10mol), yield 88%, mp 154–156◦C(EtOAc) (lit.155–157◦C, Katz, 1951). 1H NMR ([D6]DMSO,600 MHz) δ (d, J = 8.4Hz, 1H), 7.04 (d, J = 2.4Hz,1H), 6.78 (dd, J1 = 1.8Hz, J2 = 8.4Hz, 1H), 5.53 (bs,2H) (Supplementary Figure 4), 13C NMR ([D6]DMSO, 150MHz) δ 148.24, 145.51, 142.00, 137.86, 123.13, 115.51, 104.18(Supplementary Figure 5). The spectral data matched that inthe literature (Reddy et al., 2010). m/z (ESI): 185.0 [M + H]+

(Supplementary Figure 6), RP-HPLC: 5–80% B in 25min +

3min up to 100% B + 100% B in 5min, tR = 11.527min(Supplementary Figure 7), TLC: n-hexane/dioxane = 2:1;Rf: 0.76.

Preparation of 6-amino-2-cyanobenzothiazole (3)In order to get a suspension, 6.1 g (93.0 mmol) KCN wassonicated in 400ml DMAA for 3 × 15min. The suspension washeated in an oil-bath at 98–100◦C under argon atmosphere andthen 6.86 g (37.15 mmol) 6-amino-2-chlorobenzothiazole (2),dissolved in 20ml DMAA, was dripped to this reaction mixtureover a period of 50min. This resulting mixture was heatedin an oil-bath at 110◦C and stirred continuously under argonatmosphere for 12 h. After 12 h stirring there was still startingmaterial in the mixture. An increased conversion from thestarting ratio of 2.5:1 for KCN/6-amino-2-chlorobenzothiazole(2) to the ratio of 3.4:1 was achieved by adding 2.20 g (33.80mmol) of KCN. This was followed by 5 h stirring underthe conditions described above, and after that procedure theremaining amount of 6-amino-2-chlorobenzothiazole (2) wasinsignificant. The reaction mixture was poured on a mixtureof 200 g ice, 400ml 1M KH2PO4 and 300ml ether. Theorganic phase was separated from the aqueous phase. Thelatter was extracted with 2 × 250ml ether, then with 2× 200ml ethyl acetate. The combined organic phases werewashed with 2 × 300ml water and 1 × 300ml brine,dried over anhydrous Na2SO4 and concentrated on rotaryevaporator. The resulting material was a pale brown solid,its weight was 6.4 g (crude). The material was recrystallizedfrom acetone, and the impurities were removed by addingactivated charcoal to the solution. The weight of the desiredpurified material was 5.09 g (29.10 mmol), yield 78%, mp 218–219◦C (EtOAc) (lit. mp 216–218◦C, White et al., 1966). 1HNMR ([D6]DMSO, 600 MHz) δ (d, J = 9.0Hz, 1H), 7.22(d, J = 1.8Hz, 1H), 7.01 (dd, J1 = 1.8Hz, J2 = 9.0Hz,1H), 4.40 (bs, 2H) (Supplementary Figure 8), 13C NMR(150 MHz, ([D6]DMSO) δ 150.06, 144.10, 138.75, 128.86,125.62, 118.19, 114.71, 103.54 (Supplementary Figure 9). Thespectral data matched that in the literature (McCutcheonet al., 2015). m/z (ESI) 176.0 [M + H]+ (SupplementaryFigure 10), RP-HPLC: 5–80% B in 25 + 3min up to100% B + 5 in 100% B, tR = 16.692min (SupplementaryFigure 11), TLC: n-hexane/dioxane= 2:1; Rf: 0.48. The remainingKCN was reacted with KH2PO4 in order to get non-toxicKOCN.

Preparation of

Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole (4)6.30 g (15.30 mmol, 1.5 equiv) Fmoc-Asp(OtBu)-OH, which waspreviously dried in a vacuum desiccator, and 4.30 g (15.30 mmol,1.5 equiv) TCFH were solved in 35ml dry DCM. The mixturewas stirred for 60min at room temperature. First 3.06ml (18.36mmol, 1.8 equiv) DIPEA, then 1.79 g (10.20 mmol, 1 equiv) 6-amino-2-cyanobenzothiazole, which was previously dried in avacuum desiccator, were added. Further 200ml dry DCM wasadded to get complete dissolution of the materials. After stirringthe reaction mixture overnight at room temperature, it wastransferred into a separatory funnel and washed with water (2× 30ml), with saturated NaHCO3-solution (2 × 30ml), thenwith water again (2 × 30ml), and finally with brine (2 × 30ml).It was dried over anhydrous Na2SO4, finally concentrated onrotary evaporator. The resulting crude material was a yellowish-brown powder, its weight was 6.17 g. RP-HPLC analysis showeda yield of 73%. m/z (TOF) 569.2070 [M + H]+ (SupplementaryFigure 12), RP-HPLC (for the purified compound): 50–100% Bin 25min, tR = 10.591min (Supplementary Figure 13), TLC:ethanol (EtOH)/toluene 50:7.5; Rf: 0.58.

Synthesis of Fmoc-Asp(OtBu)-6-amino-D-luciferin (5)2.96 g (5.20 mmol) Fmoc-Asp(OtBu-)-6-amino-2-cyanobenzothiazole was dissolved in the mixture of 35mlMeOH and 20ml THF. 1.37 g (7.80 mmol) D-cysteine·HCl·H2O,dissolved in 10ml distilled water, was added to the mixtureat room temperature under argon atmosphere, while stirringcontinuously under pH control (starting pH: 1.67). After 20minstirring at room temperature 16ml 5% (m/m) NaHCO3 wasadded dropwise over a period of 1 h to the mixture in orderto release cysteine from its salt while continuously checkingpH. Reaching pH 2.5, a fine, yellow solid material, Fmoc-Asp(OtBu)-6-amino-D-luciferin free carboxylic acid, started toprecipitate. At pH 6.1, this material started to dissolve, and atpH 7.36, it dissolved completely. Here the Fmoc-Asp(OtBu)-6-amino-D-luciferin formed Na-salt, which dissolved under thebasic conditions. After an additional 20min stirring at roomtemperature the organic solvent was removed under reducedpressure. Water and methanol forms an azeotrope, and thetwo solvents were therefore removed together through thedistillation. Due to the decrease in the concentration of the water,from the remaining aqueous solution a yellow solid material,Fmoc-Asp(OtBu)-6-amino-D-luciferin Na-salt, precipitated.This, however, was just an irrelevant event during the process, asour goal was the removal of the methanol. The aforementionedprecipitate is water soluble, so then it was dissolved againin 20ml water and extracted with 1 × 15ml ethyl acetate inorder to get rid of possible impurities. Having dropped thissolution on a mixture of ice and glacial acetic acid (adjusted topH 3), a fine yellow precipitate formed, Fmoc-Asp(OtBu)-6-amino-D-luciferin free carboxylic acid. It was allowed to settlefor 10min, filtered and washed with 3 × 10ml water, thenair-dried to constant weight, which was 2.83 g (4.20 mmol),yield 81%. m/z (TOF) 673.1882 [M + H]+ (SupplementaryFigure 14), RP-HPLC: 50–100% B in 25min, tR = 21.046min






(Supplementary Figure 15), TLC: toluene/EtOH 50:30 saturatedwith water, Rf: 0.58.

Attachment of Fmoc-Asp(OtBu)-6-amino-D-luciferin

to Solid Support (6)Solid phase peptide synthesis was performed manually by usinga solid phase vessel attached to a rotating apparatus. 0.127 g(0.10 mmol, 1 equiv) p-alkoxybenzyl alcohol resin was allowedto swell in anh DCM for 20min. After the removal of theDCM, 0.202 g (0.30 mmol, 3 equiv) Fmoc-Asp(OtBu)-amino-D-luciferin, 0.062 g (0.30 mmol, 3 equiv) DCC, 0.041 g (0.30mmol, 3 equiv) HOBt, and 0.037 g (0.10 mmol, 1 equiv) DMAP,dissolved in 10ml anh DCM, was added to the resin. Thecoupling reaction was shaken for 3 h at room temperature. Afterthe removal of the coupling mixture, the resin was rinsed withDCM (3 × 10ml), MeOH (1 × 10ml) and then with DCM (3× 10ml) again. The coupling reaction was repeated with thesoln of 0.067 g (0.10 mmol, 1 equiv) Fmoc-Asp(OtBu)-amino-D-luciferin, 0.062 g (0.30 mmol, 3 equiv) DCC, 0.041 g (0.30 mmol,3 equiv) HOBt and 0.037 g (0.10 mmol, 1 equiv) DMAP in 5mlanh DCM at room temperature for 2 h. The resin was drainedand rinsed with DCM (3 × 10ml), MeOH (1 × 10ml), DCM (3× 10ml), then dried to constant weight.

Determination of load5mg of dried loaded resin was treated with a mixture ofTFA/water (500 µl, with the ratio of 95:5) for 1 h at roomtemperature. This was followed by the addition of 500 µl waterto the cocktail, which was then filtered off. 10 µl from thefiltrate was injected to analytical RP-HPLC and the area of theFmoc-Asp-6-amino-D-luciferin on the resulted chromatogramwas compared with the area of 10 ul Fmoc-Asp-6-amino-D-luciferin stock solution with the concentration of 1 mg/ml. Theresulting load was 47.8%.

Preparation of N-Z-DEVD-aLuc (8)

Fmoc deprotectionFmoc deprotection was carried out by suspending the resin in20% (v/v) piperidine/DMF (5ml) and agitating the vessel at roomtemperature for 2× 10min. The suspension was then filtered andthe resin was washed with DMF (3 × 5ml), MeOH (3 × 5ml),DMF (3× 5ml).

SPPS peptide coupling (7)Fmoc-Val-OH (3 equiv), DCC (3 equiv), and HOBt (3 equiv)dissolved in DMF were added to the previously swollenand Fmoc-deprotected loaded resin (1 equiv). The resultingsuspension was agitated at room temperature for 2 h and the resinwas then rinsed with DMF (3 × 5ml), MeOH (3 × 5ml), DMF(3× 5ml).

The same procedure was carried out with Fmoc-Glu(OtBu)-OH (3 equiv) and Z-Asp(OtBu)-OH, (3 equiv). The presence orabsence of the Nα-free amino group was monitored using theKaiser test.

Cleavage of peptide from the resinThe peptide-resin (7) was treated with a solution of TFA/water(95:5 v/v) for 2 h at room temperature. After the removal of the

cleaving mixture, the resin was rinsed with AcN (3 × 10ml),MeOH (1× 10ml) and with AcN (3× 10ml) again. The resultingmaterial is a yellow liquid, which was lyophilized afterwards. RP-HPLC for the crude compound: 5–80% B in 25min + 3min upto 100% B + 5min in 100% B, tR = 18.483min (SupplementaryFigure 16).

Purification of crude peptide26mg crude peptide was dissolved in acetic acid/water (1.5ml,with the ratio of 1:1), then filtered, using a 0.45µm nylonfilter. Gradient elution was used, 0–60% eluent B in 60min ata 3ml min−1 flow rate with detection at 220 nm. Pure fractionswere collected and lyophilized to give a pale yellow material,the weight of which was 11.4mg (0.013 mmol). 1H NMR (600MHz, [D6]DMSO) δ 10.26 (bs, 1H), 8.61 (s, 1H), 8.42 (d,J = 7.2Hz, 1H), 8.08 (d, J = 9.0Hz, 2H), 7.81 (bs, 1H), 7.64 (dd,J1 = 9.0Hz, J2 = 29.4Hz, 2H), 7.34 (s, 5H), 5.42 (t, J = 8.4Hz,1H), 5.02 (s, 2H), 4.69 (d, J = 7.2Hz, 1H), 4.34 (dd, J1 = 5.4Hz,J2 = 29.4Hz, 2H), 4.12 (bs, 1H), 3.77 (t, J = 10.8Hz, 1H), 3.68(dd, J1 = 8.4Hz, J2 = 11.4Hz, 1H), 2.77 (bs, 1H), 2.61–2.68(m, 2H), 2.25–2.46 (m, 3H), 1.76–1.99 (m, 2H), 0.83–0.86 (m,7H) (Supplementary Figure 17 part 1,2,3), 13C NMR (150 MHz,[D6]DMSO) δ 174.54, 172.00, 171.61, 171.38, 170.24, 159.62,156.32, 149.14, 138.66, 137.31, 136.71, 128.83, 128.28, 128.19,124.65, 120.28, 112.04, 78.64, 66.01, 58.21, 52.53, 51.86, 51.24,36.77, 36.31, 35.23, 30.98, 30.52, 27.59, 19.53 (SupplementaryFigure 18 part 1,2), m/z (ESI) 872.3 [M + H]+ (Figure 2), RP-HPLC: 5–80% B in 25min + 3min up to 100% B + 5min in100% B, tR = 18.555min (Figure 3).

N-Z-DEVD-aLuc Biochemical AssayCaspase-3 and the assay buffer were used from the caspase-3inhibitor drug screening kit. Caspase-3 was used in ten-fold serialdilution starting from 227 mU to 22.7 µU/reaction. N-Z-DEVD-aLuc substrate was applied in 100µM to 1µM in 25 µl finalreaction volume in a black plastic microtiter plate. The effect ofthe pan-caspase inhibitor Z-VAD-fmk was tested in equimolarratio of N-Z-DEVD-aLuc at 10µM with 2.27 mU/reactioncaspase-3. After 45min incubation at 37◦C we added 25 µlluminescence detection reagent to each well. Luminescence wasrecorded as cps by a plate reader within 5min. Blank wellscontained each component except caspase-3. Presented values areblank-subtracted.

Cell Lines and CultureA549 cells and U87-Luc glioblastoma cells were maintainedin DMEM-F12 and DMEM cell culture media, respectively.Both type of medium were supplemented with 10% (v/v)heat-inactivated FBS, 100 units/ml penicillin and 100 mg/mlstreptomycin at 37◦C in a humidified atmosphere containing5% CO2.

N-Z-DEVD-aLuc Cellular AssayA549 non-small cell lung carcinoma cells (2 × 106) were platedin 60mm dishes in DMEM-F12 media. After cell attachment(24 h), the cells were treated with curcumin analog C150 (5µM to1.25µM) in order to induce apoptosis in 5ml final volume (Nagyet al., 2015). After 24 h incubation supernatant was harvested and






FIGURE 2 | Mass spectrum (ESI) of the purified N-Z-DEVD-aLuc (8) 872.3 = [M+H]+.

FIGURE 3 | RP-HPLC profile of the purified N-Z-DEVD-aLuc (8), 5–80% B in 25 + 3min up to 100% B + 5min in 100% B, tR = 18.555min.

kept on ice. Cells were washed with PBS and trypsinized (5min,37◦C). Supernatant, washing PBS and media blocked trypsinwere mixed and centrifuged down (5min, 4◦C, 1,800 g). Lysis

buffer was diluted with distilled water (five times concentratedlysis buffer: 250mM HEPES, pH 7.4, 25mM CHAPS, 25mMDTT, and 50 µl 1x concentration lysis buffer was added to the






pellet, resuspended and kept on ice for 15min. Samples werecentrifuged down for 10min (4◦C, 11,000 g). Supernatant of thelysate was harvested and kept on ice for analysis. N-Z-DEVD-aLuc was dissolved in DMSO at 10mM and used in 10µM inthe assay. Assay buffer was diluted in distilled water (ten timesconcentrated assay buffer: 200mM HEPES, pH 7.4, 1% (V/V)CHAPS, 50mM DTT, 20mM EDTA, 95 µl 1x concentrationassay buffer containing 10µM N-Z-DEVD-aLuc was measuredin 96-well tissue culture plate and 5 µl lysate was assayedin triplicates to detect luminescence proportional to caspase-3activity. Blank contained 5 µl assay buffer instead of the lysate.After 4 h reaction 50µl reactionmixture wasmeasured into blackplastic microtiter plate and 50 µl luminescent detection reagentwas added. Luminescence was recorded as cps by a plate readerwithin 5min. Presented values are blank-subtracted.

In Vivo Animal ModelMale SCID mice (6 weeks old, 22–24 g body weight) were housedin sterile cages at Avidin Ltd. The mice were fed autoclaved foodand sterile water ad libitum. For inoculation, the U87-Luc cellswere trypsinized, washed and resuspended in sterile PBS. Themice were injected subcutaneously with this suspension (3× 106

cells in 0.2ml), in the dorsal region, unilaterally. All operativeprocedures and animal care conformed strictly to the HungarianCouncil on Animal Care guidelines. (Approval provided by Headof Foodchain-safety and Animal Health of the Csongrad CountyGovernment Office. Document number: CSI/01/126/2013; validuntil 8th of January 2018.) 18 days after inoculation the micewere treated with Ac-915, a lipid droplet binding thalidomideanalog inducing oxidative stress and apoptosis in glioblastomacells (Nagy et al., 2013). Ac-915 was dissolved in DMSO:solutol3:1 mixture, then diluted in PBS four times and injected i.p. at a20 mg/kg dose except negative control mice which were injected

by PBS only. Six hours after drug administration, to monitorapoptosis induction, the mice were injected i.p., with 50 mg/kgN-Z-DEVD-aLuc in PBS, followed by anesthetization in 2–3%isoflurane atmosphere. 30min after the injection of the substrate,the mice were imaged using a charge coupled device camera inthe IVIS 100 imaging instrument.

StatisticsStatistical significance was calculated with unpaired t-test (two-tailed, homoscedastic) between untreated and one treatedsample. Each point represents the average of 3 wells ± SEM.Values are blank-subtracted (blank = no caspase). ∗p < 0.05;∗∗p < 0.01; ∗∗∗p < 0.001.

RESULTS AND DISCUSSION

The desired peptide-luciferin conjugate (N-Z-DEVD-aLuc) wasreached in an 8-step route (Figure 4, Supplementary Table 6):

nitration → reduction → chlorine-cyanide exchange →

attachment of the C-terminal amino acid of the target sequence→ cysteine addition → attachment to resin → solid-phasepeptide synthesis→ cleavage from resin.

As starting material, cheap, commercially available 2-chlorobenzothiazole was used, which was first nitrated with amixture of potassium nitrate and concentrated sulphuric acid,keeping the temperature at 0◦C (Katz, 1951). The structureof the resulting 2-chloro-6-nitrobenzothiazole (1) was attestedby 1H-NMR spectrum. The next step was the reduction ofthe nitro-group. First the mixture of ethanol/glacial aceticacid/iron powder was used (Katz, 1951). Although the reactionwas successful, the resulting by-product (iron(III) acetate)was difficult to dispose of, so this method was dismissed.When trying the reduction with tin(II) chloride/glacial acetic

FIGURE 4 | The synthetic route to N-Z-DEVD-aLuc. Reagents and conditions: (a) ccH2SO4/KNO3, 0–15◦C, 5h (b) EtOAc, NH4Cl, H2O, Fe powder, reflux, 8h (c)

KCN, DMAA, 110◦C, 12h (d) Fmoc-Asp(OtBu)-OH, TCFH, dry DCM, DIPEA, overnight, rt (e) D-Cys Hcl H2O, THF, MeOH, H2O, 5 m/m% NaHCO3, 2h, rt (f)

p-alkoxybenzyl alcohol resin, dry DCM, DCC, HOBt, DMAP, 5 h 30min (g) Fmoc -/Z-protected amino acid. DCC. HOBt, DMF, 2 h, rt; 20% (v/v) piperidine for Fmoc

deprotection, 20min, rt (h) TFA/H2O 95:5 (v/v), 2 h, rt.






acid/concentrated hydrogen chloride, large amounts of by-product formed, in which the chlorine was split off or substitutedwith a hydroxyl group, so this method also had to be dismissed.The third possibility was the application of sodium pyrosulphite,which also turned out to be unsatisfactory due to the reductiongiving a very low yield. Using zinc/hydrogen chloride led to thesame results as with tin(II) chloride.

Finally, applying ethyl acetate, water, ammonium chloride,and iron powder system solved the problem and the reductionwas successful with a good yield. Using a Soxhlet extractorturned out to be a solvent-sparing, thus environmentally-friendlymethod and processing the obtained product was also simple: thesolution had to be decanted in order to get rid of the iron powderand then extracted.

The next step—the chlorine/nitrile exchange in the 6-amino-2-chlorobenzothiazole (2)—is the key in the production of thedesired conjugates. Six different methods had to be tried, thecommon features of which were the polar aprotic non-aqueoussolvent, the high temperature and the long reaction time (a)anh DMSO/KCN, 160◦C, 10 h; (b) anh DMSO/KCN/KI, 160◦C,10 h; (c) anh DMSO/18-crown-6/KCN, 120◦C, 8 h; (d) anhDMF/KCN, 140◦C, 12 h; (e) anh HMPA/KCN, 140◦C, 10 h; (f)f) anh DMAA/KCN/KI, 120◦C, 8 h).

The first five methods had to be dismissed due to the lowyield (15–20%). When trying DMAA, however, it turned out thatKCN is dissolved best in this solvent, resulting in relatively highyield (75–80%) This means that the success of the chlorine/nitrileexchange depends on the rate of KCN dissolution.

An Fmoc-protected amino acid (Fmoc-Asp(OtBu)-OH) wascoupled to the 6-amino-2-cyanobenzothiazole (3). As, due tothe deactivated amino group, the amide bond could not beformed with the usual coupling reagent, DCC, a much moreactivating coupling agent was necessary. Different agents weretested in different quantities: COMU (El-Faham and Albericio,2010; Takakura et al., 2011; Chantell et al., 2012), HATU (El-Faham et al., 2010), Deoxo-Fluor Reagent (El-Faham et al., 2010),TFFH (Kangani et al., 2006), and TCFH (Carpino et al., 1996), allwith a ratio of 1:1.5 and 1:3. The best yield (97%) was obtainedwith 1.5 equivalents of TCFH (Table 1). Other than the quantityof the different coupling reagents, all other conditions (solvent,reaction time, temperature etc.) were kept the same. Although

TABLE 1 | Coupling agents and yields.

Coupling agent Quantity Yield (%)

COMU 1.5 equiv 0

COMU 3.0 equiv 0

HATU 1.5 equiv 7

HATU 3.0 equiv 8

Deoxo-fluor reagent 1.5 equiv 48

Deoxo-fluor reagent 3.0 equiv 38

TFFH 1.5 equiv 59

TFFH 3.0 equiv 51

TCFH 1.5 equiv 97

TCFH 3.0 equiv 72

it was not checked with chiral chromatography at this point,it became obvious following the achiral chromatography afterthe cysteine addition that there was no racemization, because ifthere had been, during either this or the previous step, we wouldhave seen diastereomers. As the achiral chromatography after thecysteine addition was indispensable anyway, we could save therather complicated chiral chromatography one step earlier.

During the addition of D-cysteine (Tulla-Puche et al.,2008) to the resulting conjugate (Fmoc-Asp(OtBu)-6-amino-2-cyanobenzothiazole, (4), the amino acid-heterocycle conjugatewas dissolved in THF andMeOH, then D-cysteine hydrochloridemonohydrate was added. The resulting compound was dissolvedin water, and then the cysteine was released from its salt withNaHCO3. During the reaction (about 25min) the pH of thesolution was kept between 7.3-7.4 by the addition of NaHCO3

aqueous solution, monitoring the process with a pH-meterand the addition was carried out under argon atmosphere. Bythis way, the desired amino acid-6-amino-D-luciferin conjugate,Fmoc-Asp(OtBu)-6-amino-D-luciferin (5) was obtained.

During the next step this conjugate was attached to resin (6).Two types of resins were tested: 2-chlorotrityl chloride and p-alkoxybenzyl alcohol (Wang resin). Loading was checked in bothcases: with 2-chlorotrityl chloride resin it was 30%, while withWang resin it was 50%, so we decided to use the latter.

Classical solid phase peptide synthesis was carried out: thepeptide chain was built with Fmoc strategy. However, the N-terminal amino acid was always Z-protected, as this protectinggroup gives higher biological stability to the peptide. Theobtained peptide-aLuc conjugate was removed from the resinwith the mixture of TFA/water (95:5 v/v); finally, the resultingmaterial was purified by preparative HPLC.

The material—N-Z-DEVD-aLuc (7)—was successfully testedin a bioluminescent system. It has been published that bothcaspase-3 and caspase-7 digest DEVD sequence, but caspase-3has six time higher DEVD digestion activity (Talanian et al., 1997;McStay et al., 2008) so we tested N-Z-DEVD-aLuc by the activityof caspase-3 in vitro (Figure 5) and caspase-3 in vivo (Figure 6).The purity of the peptide is demonstrated in SupplementaryFigure 19. The biological relevance of our N-Z-DEVD-aLucsubstrate was confirmed in a biochemical reaction using a serialdilution of recombinant caspase-3 from 227 mU/reaction to22.7 µU/reaction. In order to verify that our N-Z-DEVD-aLucis a real substrate for caspase-3, not only the enzyme but alsothe N-Z-DEVD-aLuc substrate was titrated in a concentrationrange from 100 to 1µM. The recorded luminescence waslinearly proportional with the growing enzyme activity andincreased amount of the N-Z-DEVD-aLuc substrate showingmaximum cps at 227 mU caspase-3 and 100µM N-Z-DEVD-aLuc (Figure 5A, Supplementary Figure 20) Moreover, theluminescence signal was completely abolished by the applicationof equimolar pan-caspase inhibitor Z-VAD-fmk in the reactionof 2.27 mU caspase-3 and 10µM N-Z-DEVD-aLuc (Ion et al.,2006; Figure 5B). We further verified the applicability of our N-Z-DEVD-aLuc substrate to detect cellular apoptotic cell deathcaused by a drug candidate molecule. The curcumin analog C150induces caspase-3 activation (Szebeni et al., 2017; SupplementaryFigure 21) and apoptosis of A549 human non-small cell lung






FIGURE 5 | Luminescence is proportional to caspase activity and N-Z- DEVD-aLuc concentration. (A) Recombinant human caspase-3 (227 mU to 22.7 µU) and

N-Z-DEVD-aLuc (100 to 1µM) were titrated and assayed for 45min. (B) The effect of the pan-caspase inhibitor Z-VAD-fmk was tested in equimolar ratio of

N-Z-DEVD-aLuc at 10µM with 2.27 mU/reaction caspase-3. (C) Caspase-3 and -7 activation induced by the curcumin analog C150 (5–1.25µM) on A549 cells was

detected by N-Z-DEVD-aLuc, control corresponds to untreated sample. The results are shown as arithmetic mean values of three samples ± SEM. Anyhow, SEM

values are too small to be visible on the logarithmic scale of Figure 5A. ***p < 0.001.

carcinoma cells (Nagy et al., 2015; Hackler et al., 2016) and wecould detect the activation of caspase-3 by N-Z-DEVD-aLuc viabioluminescence (Figure 5C).

To measure apoptosis directly in animals, an opticalimaging experiment was performed in vivo, administrating N-Z-DEVD-aLuc to SCID mice (previously inoculated with thestably expressing luciferase cell line U87-Luc) that had beentreated with chemotherapeutics previously. We used Ac-915,a lipid droplet binding thalidomide analog inducing caspase-3activation (Supplementary Figure 22) and oxidative stress andapoptosis in different cancer cells (Nagy et al., 2013). Ac-915enhanced the bioluminescent signal already at 6 h. Significantlyfewer signals were detected from control mouse having noAc-915 treatment, but injected with only N-Z-DEVD-aLucsubstrate, which represents the basal level of apoptosis. However,the limitation of the widespread applicability of the luciferinconjugated peptides in vivo is that luciferase enzyme activityis indispensable, therefore luciferase transgenic mouse or cellsshould be used in these studies.

CONCLUSION

A general strategy for the synthesis of N-peptide-6-amino-D-luciferin conjugates has been developed. The strategy is based ona newly established sequence of different transformations:

nitration → reduction → chlorine-cyanide exchange →

attachment of the C-terminal amino acid of the target sequence(variable, depending on the protease to bemeasured)→ cysteine

FIGURE 6 | In vivo test of N-Z-DEVD-aLuc. N-Z-DEVD-aLuc (100 mg/kg, i.p.)

was administered to all SCID mice previously inoculated with U87-Luc

glioblastoma cells (middle). Apoptosis was induced by Ac-915 in all mice

except negative controls administered by PBS (left). Aminoluciferin was used

as positive control (right).

addition → attachment to resin → solid-phase peptidesynthesis (variable, depending on the protease to be measured)→ cleavage from resin.






The cornerstone of N-peptide-6-amino-D-luciferin conjugatesynthesis is the availability of 6-amino-2-cyanobenzothiazole inlarge quantities. Although the used transformations (nitration,NO2 reduction and cyanidation) are well-known in the literature,our combination of these transformations, the equipment andthe solvents make it possible to prepare this material in largerquantities than the published strategies; the key of the methodis the chlorine-cyano group exchange, the success of whichdepends on the solvent. Using this larger quantity of 6-amino-2-cyanobenzothiazole, we preparedN-Z-DEVD-aLuc with Fmocsolid phase peptide synthesis. The material has already beensuccessfully used in in vivo optical imaging experiments.

The fact that in step 4 and 7 any amino acid can be usedmeans that our method provides a practical and scalable way forpreparation of other N-peptide-6-amino-D-luciferin conjugatesas well, which compounds have a crucial role in the developmentof plate based, high-throughput viability assays.

AUTHOR CONTRIBUTIONS

The project was conceived by LP, GT designed and coordinatedthe research. AK and PH chemically synthesized and analyzed the

materials, GS performed in vitro assays. LN performed in vivoexperiments. AK, GS, LP, and GT analyzed and compiled the dataand co-wrote the manuscript. The final manuscript was read andapproved by all the authors.

FUNDING

This research was supported by the GINOP-2.3.2-15-2016-00030 grant. GS was supported by János Bolyai ResearchScholarship of the Hungarian Academy of Sciences. (no. BO/00139/17/8).

ACKNOWLEDGMENTS

We would like to thank Anasztázia Hetényi for the NMRspectra.

SUPPLEMENTARY MATERIAL

The Supplementary Material for this article can be foundonline at: https://www.frontiersin.org/articles/10.3389/fchem.2018.00120/full#supplementary-material

REFERENCES

Carpino, L. A., Triolo, S. A., Griffin, G. W., Herman, L. W., Tarr, G., Sole, N.A., et al. (1996). “Innovation and perspectives in solid phase synthesis &combinatorial libraries: peptides, proteins and nucleic acids–small moleculeorganic chemical diversity,” in Collected Papers, International Symposium, 4th

Sept. 12–16, 1995 (Edinburgh), 41–50.Chantell, C. A., Onaiyekan, M. A., and Menakuru, M. (2012). Fast conventional

Fmoc solid-phase peptide synthesis: a comparative study of different activators.J. Pep. Sci. 18, 88–91. doi: 10.1002/psc.1419

El-Faham, A., and Albericio, F. (2010). COMU: a third generation of uranium-typecoupling reagents. J. Pep. Sci. 16, 6–9. doi: 10.1002/psc.1204

El-Faham, A., Subiros-Funosas, R., and Albericio, F. (2010). A novelfamily of onium salts based upon isonitroso meldrum’s acid provesuseful as peptide coupling reagents. Eur. J. Org. Chem. 19, 3641–3649.doi: 10.1002/ejoc.201000314

Gilbert, D. F., and Boutros, M. (2016). A protocol for a high-throughputmultiplex cell viability assay. Methods Mol. Biol. 1470, 75–84.doi: 10.1007/978-1-4939-6337-9_6

Gryshuk, A. L., Perkins, J., and LaTour, J. V. (2011). Methods and Systems

for Synthesis of a D-Aminoluciferin Precursor and Related Compounds. US2011/0224442 A1.

Gryshuk, A. L., Perkins, J., and LaTour, J. V. (2013). Methods and Systems for

Synthesis of a D-Aminoluciferin Precursor and Related Compounds. Patentnumber: 8586759. Livermore, CA: Lawrence Livermore National Security, LLC.

Hackler, L. Jr., Ózsvári, B., Gyuris, M., Sipos, P., Fábián, G., Molnár, E., et al. (2016).The curcumin analog C-150, influencing NF-κB, UPR and Akt/Notch pathwayshas potent anticancer activity in vitro and in vivo. PLoS ONE 11:e0149832.doi: 10.1371/journal.pone.0149832

Hauser, J. R., Beard, H. A., Bayana, M. F., Jolley, K. F., Warriner, S. I., and Bon, R.S. (2016). Economical and scalable synthesis of 6-amino-2-cyanobenzothiazole.Beilstein J. Org. Chem. 12, 2019–2025. doi: 10.3762/bjoc.12.189

Hickson, J., Ackler, S., Klaubert, D., Bouska, J., Ellis, P., Foster, K., et al. (2010).Noninvasive molecular imaging of apoptosis in vivo using a modified fireflyluciferase substrate, Z-DEVD-aminoluciferin. Cell Death Differ. 17, 1003–1010.doi: 10.1038/cdd.2009.205

Hsu, H. T., Trantow, B. M., Waymouth, R. M., and Wender, P. A. (2016).Bioorthogonal catalysis: a general method to evaluate metal-catalyzed reactions

in real time in living systems using a cellular luciferase report system. Bioconjug.Chem. 138, 376–382. doi: 10.1021/acs.bioconjchem.5b00469

Ioka, S., Saitoh, T., Iwano, S., Suzuki, K., Maki, S. A., Miyawaki, A.,et al. (2016). Synthesis of firefly luciferin analogues and evaluation of theluminescent properties. Chem. Eur. J. 22, 9330–9337. doi: 10.1002/chem.201600278

Ion, G., Fajka-Boja, R., Kovács, F., Szebeni, G., Gombos, I., Czibula, Á., et al. (2006).Acid sphingomyelinase mediated release of ceramide is essential to trigger themitochondrial pathway of apoptosis by galectin-1. Cell. Sign. 18, 1887–1896.doi: 10.1016/j.cellsig.2006.02.007

Kangani, C. O., Kelley, D. E., and Day, B. W. (2006). One-pot synthesisof aldehydes or ketones from carboxylic acids via in situ generation ofWeinreb amides using the Deoxo-Fluor reagent. Tetrah. Lett. 47, 6289–6292.doi: 10.1016/j.tetlet.2006.06.121

Kaskova, Z. M., Tsarkova, A. S., and Yampolsky, I. V. (2016). 1001 lights:luciferins, luciferases, their mechanisms of action and applications inchemical analysis, biology and medicine. Chem. Soc. Rev. 45, 6048–6077.doi: 10.1039/C6CS00296J

Katz, L. (1951). Antituberculous compunds. II. 2-Benzalhydrazinobenzothiazoles.J. Am. Chem. Soc. 73, 4007–4010. doi: 10.1021/ja01152a132

Kepp, O., Galluzzi, L., Lipinski, M., Yuan, J., and Kroemer, G. (2011). Celldeath assays for drug discovery. Nat. Rev. Drug Discov. 10, 221–237.doi: 10.1038/nrd3373

Leippe, D. M., Nguyen, D., Zhou, M., Good, T., Kirkland, T. A., Scurria, M.,et al. (2011). A bioluminescent assay for the sensitive detection of proteases.Biotechniques 51, 105–110. doi: 10.2144/000113716

McCutcheon, D. C., Porterfield, W. B., and Prescher, J. A. (2015). Rapid andscalable assembly of firefly luciferase substrates. Org. Biomol. Chem. 13,2117–2121. doi: 10.1039/C4OB02529F

McStay, G. P., Salvesen, G. S., and Green, D. R. (2008). Overlapping cleavagemotif selectivity of caspases: implications for analysis of apoptotic pathways.Cell Death Diff. 15, 322–331. doi: 10.1038/sj.cdd.4402260

Nagy, L. I., Fehér, L. Z., Szebeni, G. J., Gyuris, M., Sipos, P., Alföldi, R., et al.(2015). Curcumin and its analogue induce apoptosis in leukemia cells and haveadditive effects with bortezomib in cellular and xenograft models. BioMed Res.

Int. 2015:968981. doi: 10.1155/2015/968981Nagy, L. I., Molnár, E., Kanizsai, I., Madácsi, R., Ózsvári, B., Fehér, L.

Z., et al. (2013). Lipid droplet binding thalidomide analogs activate


https://www.frontiersin.org/articles/10.3389/fchem.2018.00120/full#supplementary-material

https://doi.org/10.1002/psc.1419

https://doi.org/10.1002/psc.1204

https://doi.org/10.1002/ejoc.201000314

https://doi.org/10.1007/978-1-4939-6337-9_6

https://doi.org/10.1371/journal.pone.0149832

https://doi.org/10.3762/bjoc.12.189

https://doi.org/10.1038/cdd.2009.205

https://doi.org/10.1021/acs.bioconjchem.5b00469

https://doi.org/10.1002/chem.201600278

https://doi.org/10.1016/j.cellsig.2006.02.007

https://doi.org/10.1016/j.tetlet.2006.06.121

https://doi.org/10.1039/C6CS00296J

https://doi.org/10.1021/ja01152a132

https://doi.org/10.1038/nrd3373

https://doi.org/10.2144/000113716

https://doi.org/10.1039/C4OB02529F

https://doi.org/10.1038/sj.cdd.4402260

https://doi.org/10.1155/2015/968981





endoplasmic reticulum stress and suppress hepatocellular carcinoma in achemically induced transgenic mouse model. Lipids Health Dis. 12:175.doi: 10.1186/1476-511X-12-175

O’Brien, M. A., Daily, W. J., Hesselberth, P. E., Moravec, R. A., Scurria,M. A., Klaubert, D. H., et al. (2005). Homogenous, bioluminescentprotease assays: caspase-3 as a model. J. Biomol. Screen 10, 137–148.doi: 10.1177/1087057104271865

O’Brien, M. A., Moravec, R. A., Riss, T. L., and Bulleit, R. F. (2008).Homogeneous, bioluminescent proteasome assays. Methods Mol. Biol. 414,163–181. doi: 10.1007/978-1-59745-339-4_13

Reddy, G. R., Thompson, W. C., and Miller, S. C. (2010). Robust light emissionfrom cyclic alkylaminoluciferin substrates for firefly luciferase. J. Am. Chem.

Soc. 132, 13586–13587. doi: 10.1021/ja104525mShinde, R., Perkins, J., and Contag, C. H. (2006). Luciferin derivatives for enhanced

in vitro and in vivo bioluminescence assays. Biochemistry 45, 11103–11112.doi: 10.1021/bi060475o

Szebeni, G. J., Balázs, Á., Madarász, I., Pócz, G., Ayaydin, F., Kanizsai, I., et al.(2017). Achiral mannich-base curcumin analogs induce unfolded proteinresponse and mitochondrial membrane depolarization in PANC-1 cells. Int. J.Mol. Sci. 18:2105. doi: 10.3390/ijms18102105

Takakura, H., Kojima, R., Urano, Y., Terai, T., Hanaoka, K., and Nagano,T. (2011). Aminoluciferins as functional bioluminogenic substrates offirefly luciferase. Chem. Asian J. 6, 1800–1810. doi: 10.1002/asia.201000873

Talanian, R. V., Quinlan, C., Trautz, S., Hackett, M. C., Mankovich, J. A., Banach,D., et al. (1997). Substrate specificities of caspase family proteases. J. Biol. Chem.272, 9677–9682. doi: 10.1074/jbc.272.15.9677

Tulla-Puche, J., Torres, A., Calvo, P., Royo, M., and Albericio, F. (2008).N,N,N’,N’-tetramethylchloroformamidinium hexafluorophosphate (TCFH), apowerful coupling reagent for bioconjugation. Bioconjug. Chem. 19, 1968–1971.doi: 10.1021/bc8002327

White, E. H., Woerther, H., Seliger, H. H., and McElroy, W. D. (1966). Aminoanalogs of firefly luciferin and biological activity thereof. J. Am. Chem. Soc. 88,2015–2018. doi: 10.1021/ja00961a030

Conflict of Interest Statement: The authors declare that the research wasconducted in the absence of any commercial or financial relationships that couldbe construed as a potential conflict of interest.

The reviewer, BL, and handling Editor declared their shared affiliation.

Copyright © 2018 Kovács, Hegyes, Szebeni, Nagy, Puskás and Tóth. This is an open-

access article distributed under the terms of the Creative Commons Attribution

License (CC BY). The use, distribution or reproduction in other forums is permitted,

provided the original author(s) and the copyright owner are credited and that the

original publication in this journal is cited, in accordance with accepted academic

practice. No use, distribution or reproduction is permitted which does not comply

with these terms.


https://doi.org/10.1186/1476-511X-12-175

https://doi.org/10.1177/1087057104271865

https://doi.org/10.1007/978-1-59745-339-4_13

https://doi.org/10.1021/ja104525m

https://doi.org/10.1021/bi060475o

https://doi.org/10.3390/ijms18102105

https://doi.org/10.1002/asia.201000873

https://doi.org/10.1074/jbc.272.15.9677

https://doi.org/10.1021/bc8002327

https://doi.org/10.1021/ja00961a030

http://creativecommons.org/licenses/by/4.0/








Vol.:(0123456789)1 3

International Journal of Peptide Research and Therapeutics https://doi.org/10.1007/s10989-018-9768-8

Synthesis of N-Peptide-6-Amino-d-Luciferin Conjugates with Optimized Fragment Condensation Strategy

Anita K. Kovács1,2 · Péter Hegyes2 · Gábor J. Szebeni3 · Krisztián Bogár4 · László G. Puskás2,3 · Gábor K. Tóth1

Accepted: 19 September 2018 © Springer Nature B.V. 2018

AbstractThe synthesis of peptide-luciferin conjugates has a pivotal role in the development of bioluminescent detection systems that are based on the determination of protease enzyme activity. This work describes the optimized synthesis of an N-peptide-6-amino-d-luciferin conjugate (Fmoc-Gly-Pro-6-amino-d-luciferin) with a simple fragment condensation method in adequate yields. Fmoc-Gly-Pro-6-amino-d-luciferin was produced from a previously synthesized Fmoc-Gly-Pro-OH and also previ-ously synthesized 6-amino-2-cyanobenzothiazole with an optimized method, to which conjugate cysteine was added in an also improved way. The resulting conjugate was successfully used in a bioluminescent system, in vitro, demonstrating the applicability of the method.

Graphical Abstract

Keywords Bioluminescence · Protease activity · Aminoluciferin · Conjugate · Fragment condensation

Electronic supplementary material The online version of this article (https ://doi.org/10.1007/s1098 9-018-9768-8) contains supplementary material, which is available to authorized users.

* Anita K. Kovács [email protected]

* Gábor K. Tóth [email protected]

1 Department of Medical Chemistry, University of Szeged, Szeged 6720, Dóm tér 8., Hungary

2 Avidin Ltd, Szeged 6726, Alsó Kikötő sor 11/D., Hungary3 Biological Research Center, Hungarian Academy

of Sciences, Szeged 6726, Temesvári krt. 62., Hungary4 Institute of Pharmaceutical Analysis, University of Szeged,

Szeged 6720, Somogyi u. 4., Hungary

http://orcid.org/0000-0001-9805-1647

http://orcid.org/0000-0002-1829-4786

http://orcid.org/0000-0002-3604-4385

http://crossmark.crossref.org/dialog/?doi=10.1007/s10989-018-9768-8&domain=pdf

https://doi.org/10.1007/s10989-018-9768-8

International Journal of Peptide Research and Therapeutics

1 3

Introduction

Bioluminescent detection systems, used for in vivo and in vitro analytical methods, have been in the spotlight of research in the past decades (Jiang et al. 2018; Xu et al. 2016; Sadikot and Timothy 2005; Chollet and Ribault 2012; Sato et al. 2004). The advantage of these sys-tems over fluorescent systems lies in their superior sen-sitivity and easy handling (De Saint-Hubert et al. 2012; Hu et al. 2012). The most ubiquitous enzyme-substrate system in bioimaging is the American firefly (Photinus pyralis) luciferin-luciferase system (Presiado et al. 2012; Zhang et al. 2012). Substituting this luciferin’s 6-position hydroxyl group with an amino group, the resulting ami-noluciferin (aLuc) can form amide bond with a peptide, while retaining the transport and bioluminescent proper-ties of the original substrate, resulting in a good substrate for different important proteases, which can be used for the determination of the enzymatic activity (White et al. 1966).

Earlier this year a general synthesis method for N-pep-tide-6-amino-d-luciferin conjugates, via a hybrid liquid/solid phase method, was published (Kovács et al. 2018). The applicability of the strategy was demonstrated with the preparation of a known substrate (O’Brian et al. 2005; Hickson et al. 2010), N-Z-Asp-Glu-Val-Asp-6-amino-d-lu-ciferin (N-Z-DEVD-aLuc, Z: benzyloxycarbonyl) used to measure the activity of caspase-3, and consequently the efficiency of apoptosis-inducing drugs. Our goal was to use the same strategy to produce N-Z-Gly-Pro-6-amino-d-luciferin (N-Z-GP-aLuc), which can be used to measure the activity of other protease enzymes, namely Fibroblast activation protein alpha (FAP) and Prolyl Oligopeptidase (POP/PREP), two prolyl-specific serine proteases, the lev-els of which are elevated in many cancers and may have roles in promoting angiogenesis and in immunotolerant tumour microenvironment (Christiansen et al. 2013).

Literary Background

Synthesis routes of peptide-6-amino-d-luciferin conju-gates have been published five times. The backgrounds were worked out by Geiger and Miska (1991), who developed four methods (phospho-azo and mixed anhy-dride methods, both with both 6-amino-2-cyanoben-zothiazole and carboxyl protected 6-amino-d-luciferin as starting material) and produced 13 different conju-gates. The mixed anhydride/6-amino-2-cyanobenzo-thiazole method was modified in a 2003 Promega pat-ent (O’Brian et al. 2003), and by Gryshuk et al. (2011). O’Brian et al. (2005) followed a different path, when

6-amino-2-cyanobenzothiazole was coupled with a pro-tected peptide with the help of DCC and HOBt. Kovács et al. (2018) synthetized a conjugate with a different, liq-uid/solid phase synthesis method with better yields than the other methods (unfortunately, in some cases yields and conversion rates were not determined in the literature and the cited patents, also, HPLC, MS and NMR analysis is incomplete or missing (Tables S1–S4)).

Unsuccessful Attempt

Our original plan was to synthesize N-Z-GP-aLuc, apply-ing the above-mentioned 2018 method (Kovács et al. 2018). First the key molecule, 6-amino-2-cyanobenzothiazole (1) was produced (Kovács et al. 2018) which was then coupled to Fmoc-Pro-OH (Fmoc: 9-fluorenylmethoxycarbonyl). The resulting protected amino acid-heterocycle conjugate (Fmoc-Pro-6-amino-2-cyanobenzothiazole, 2, Figs. S1, S2) was reacted with d-cysteine. The structure of the resulting Fmoc-Pro-6-amino-d-luciferin was attested with 1H-NMR (Fig. S3) and 13C-NMR Fig. S4), LC-MS analysis was also carried out (Figs. S5, S6). Then the protected amino acid-aminoluciferin conjugate was attached to p-alkoxybenzyl alcohol resin. However, attaching the protected amino acid-aminoluciferin conjugate to solid support, the loading was only 50%. This meant that we had a significant loss of the conjugate, even though some of it can be regained with dif-ferent purification methods and reused during further resin attachment. In order to reach higher load, instead of 3 h, the coupling reaction mixture was shaken for 6 h. The determi-nation of load showed that no improvement was achieved in loading and, according to the resulting material’s RP-HPLC profile (Fig. S7) and the mass spectrum (MS) (Fig. S8), 20% of the coupled material was dehydrogenated among the con-ditions mentioned above. As we were planning to purify the final, completed product, the synthetizing process was not abandoned. The product on the resin was Fmoc-deprotected, which was followed by coupling Z-protected glycine to the proline, and the resulting conjugate was cleaved from the resin (Kovács et al. 2018).

The RP-HPLC profile of the product (Fig. S9) showed a homogeneous material, but, according to the mass spectrum (Fig. S10), it was not the expected N-Z-GP-aLuc, but a fully dehydrogenated conjugate, N-Z-Gly-Pro-6-aminodehydrolu-ciferin. This was also attested by its 1H-NMR-spectrum (Fig. S11), on which the α-proton of the 2-thiazoline ring was not detectable; however, an olefin proton was, as proof of dehy-drogenation. The driving force behind this dehydrogenation was the aromatization of the 2-thiazoline ring, which led to its transformation into a thiazole ring (Fig. 1). Since this reaction did not occur earlier either, it might have been the result of the longer exposure to the basic conditions.


1 3

As dehydroluciferin is a very efficient inhibitor of lucif-erase (Ciuffreda et al. 2013; Fontes et al. 1997), it is not suitable for the above-mentioned measurement of enzymatic activity.

Therefore, our goal was to find a method that is more reli-able than the hybrid liquid/solid phase synthesis method, a synthesis route that prevents the side reactions mentioned above. The problem was solved with returning to the frag-ment-condensation method, which is used to avoid problems occurring during stepwise solid phase synthesis (Nyfeler 1994). However, we had to make modifications to achieve better results than the standard method (O’Brian et al. 2005; Geiger and Miska 1991; O’Brian et al. 2003; Gryshuk et al. 2011).

Results and Discussion

The desired peptide-luciferin conjugate (N-Fmoc-GP-aLuc) was reached in a 2-step route:

(a) attachment of the target peptide sequence to 6-amino-2-cyanobenzothiazole → (b) cysteine addition (Fig. 2, Table S5):

The optimized synthesis route of the key molecule and the modifications of the two steps make a significant improvement over the standard method.

Fig. 1 The unsuccessful synthesis route to produce the desired N-Z-Gly-Pro-6-amino-d-luciferin

Fig. 2 The 2-step synthetic route to N-Fmoc-GP-aLuc (3). Reagents and conditions: a Fmoc-Gly-Pro-OH, TCFH, dry DCM, DIPEA, overnight, rt, yield 68% b d-Cys∙HCl∙H2O, MeOH, H2O, 5 m/m% NaHCO3, 2 h, rt, yield 78%


1 3

Synthesis of 6‑Amino‑2‑Cyanobenzothiazole (1)

We started with the synthesis of the key molecule, 6-amino-2-cyanobenzothiazole (1) with an improved method (Kovács et al. 2018): commercially available 2-chlorobenzothiazole was nitrated with a mixture of cc H2SO4 and KNO3, at 0 °C (Katz 1951). The structure of the resulting 2-chloro-6-ni-trobenzothiazole was attested by 1H-NMR spectrum (83% yield corresponding to the isolated pure product). In the next step the nitro group was reduced with ethyl acetate/water/ammonium chloride/iron powder in a Soxhlet extractor with a good yield (88%, corresponding to the crude product). The chlorine/nitrile exchange in the 6-amino-2-chlorobenzothi-azole was carried out in N,N-dimethylacetamide (DMAA), a polar aprotic non-aqueous solvent, with KCN, at high temperature (110 °C) in 12 h, resulting in relatively high yield (78%, corresponding to the crude product). During procession the remaining KCN was reacted with KH2PO4, keeping pH above 7 to avoid the production of HCN, then first FeSO4 was added, forming K4[Fe(CN)6], then Fe(III) salt was added, forming insoluble Berlin blue, which can be filtered from the solution. The product was purified with recrystallization (Kovács et al. 2018).

Synthesis of Fmoc‑Gly‑Pro‑6‑Amino‑2‑Cyanobenzothiazole (2)

A suitably protected, commercially purchased peptide, Fmoc-Gly-Pro-OH, was coupled with the key molecule, 6-amino-2-cyanobenzothiazole (1). (As during the syn-thesis only a dipeptide was coupled to the 6-amino-2-cy-anobenzothiazole (1), it was reasonable to purchase a ready material, rather than synthesize and purify one. In case of longer peptides, solid phase peptide synthesis can be used.) Due to the deactivated amino group of the 6-amino-2-cy-anobenzothiazole (1), the amide bond could not be formed with the usual coupling reagents; therefore, a more powerful coupling agent (Carpino et al. 1996) was necessary. Excel-lent conversion (97%) of the 6-amino-2-cyanobenzothiazole was obtained with 1.5 equivalents of chloro-N,N,N′,N′-tetramethylformamidinium hexafluorophosphate (TCFH) (Kovács et al. 2018). Obtained yield, corresponding to the crude product: 68%. (Figs. S12, S13).

With this process, we could avoid the extremely long cou-pling time of the standard mixed anhydride method (O’Brian et al. 2003; Gryshuk et al. 2011) and reached adequate conversion/yield.

Synthesis of Fmoc‑Gly‑Pro‑6‑Amino‑d‑Luciferin (3)

The peptide-heterocycle conjugate (Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole, 2) was dissolved in methanol (MeOH), then d-cysteine hydrochloride monohydrate was

added. The resulting substance was dissolved in water and the cysteine was released from its salt with NaHCO3. During the reaction (about 25 min) the pH of the solution was kept between 7.3 and 7.4 with the addition of NaHCO3 aque-ous solution, the process was continuously monitored with a pH-meter, under argon atmosphere. The Fmoc-protection of the the N-terminal amino-group of the peptide was kept up because it gave higher biological stability to the conju-gate. The structure of the resulting conjugate was attested with 1H-NMR (Fig. S14), 13C-NMR (Fig. S15) and LC-MS (Figs. S16, S17).

This method is also an improvement over the standard practice (O’Brian et al. 2005; Geiger and Miska 1991; O’Brian et al. 2003; Gryshuk et al. 2011) as the window between pH 7.3–7.4 a, rules out racemization b, ensures the release of the cysteine from its salt.

Analysis of Fmoc‑Gly‑Pro‑6‑Amino‑d‑Luciferin (3) in Enzyme Activity Assays

The product (3) was tested in bioluminescent-based enzyme activity assays. The substrate specificity of N-Fmoc-GP-aLuc was measured with two human proteases that are involved in cancer, POP/PREP and FAP, and with a bacte-rial non-specific endoproteinase Pro-C. All three enzymes accepted the substrate and liberated aminoluciferin as a product resulting in increased luminescence signal (Fig. 3). Enzymatic degradation was confirmed with protease inhib-itor, which completely abolished bioluminescent signal increase. Our novel substrate, therefore, could be used in different biochemical assays.

Materials and Methods

Materials

TCFH and d-Cys·HCl·H2O were obtained from AK Scien-tific Inc. (Union City, CA, USA). Fmoc-Gly-Pro-OH was purchased from Iris Biotech GmbH (Marktredwitz, Ger-many). Trifluoroacetic acid (TFA) gradient grade came from VWR International (Radnor, PA, USA). POP/PREP and recombinant human FAP alpha were obtained from R&D Systems (Minneapolis, MN, USA). Endoproteinase Pro-C, DTT and Bovine Serum Albumin (BSA) were purchased from Sigma–Aldrich (St. Louis, MO, USA), black plastic microtiter plates were from Tomtec (Budapest, Hungary). Complete protease inhibitor cocktail was from Roche Basel, Switzerland and Luminescence Detection Reagent was from Promega (Madison, WI, USA).

TLC was performed on silica gel plates 60 F254 from Merck (Darmstadt, Germany). pH values were measured with a Hanna HI 8424 pH meter. Analytical reversed-phase


1 3

high-performance liquid chromatography (RP-HPLC) was performed on an Agilent 1200 series separations module with diode array and multiple wavelength detector (Wald-bronn, Germany), with a Luna C18(2) 100 Å column (10 µm, 250 × 4.6 mm) Phenomenex, (Torrance, CA, USA). The experiments were carried out at room temperature (rt) with a flow rate maintained at 1.2 mL min−1 at 220 nm wave-length (mobile phases solvent A: 0.1% TFA in Milli-Q water and solvent B: 0.1% TFA in acetonitrile (AcN)) using gradi-ent elution. Separation was achieved on a Shimadzu (Kyoto, Japan) semi-preparative system with a Jupiter C18 300 Å column (10 µm, 250 × 21.20 mm), also from Phenomenex (mobile phases solvent A: 0.1% TFA in Milli-Q water and solvent B: 0.1% TFA in AcN) using gradient elution. Mass spectrometry data were collected on Waters (Milford, MA, USA) SQ Detector with atmospheric pressure ionization (API) mass spectrometer in positive ion mode; 1H NMR and 13C NMR spectra were recorded using a Bruker DR X 500 spectrometer at 600 MHz and 150 MHz, in deuterated dime-thyl sulfoxide ([D6]DMSO). Chemical shifts were reported on the δ scale and J values were given in Hz.

Methods

Synthesis of Fmoc-Gly-Pro-6-Amino-2-Cyanobenzothiazole (2)

2.03 g (5.145 mmol, 1.5 equiv) anhydrous (anh) Fmoc-Gly-Pro-OH and 1.44 g (5.145 mmol, 1.5 equiv) anh TCFH were solved in 7 mL anh dichloromethane (DCM). The

mixture was stirred for 60 min at room temperature. First 1 mL (6.174 mmol, 1.8 equiv) N,N-diisopropylethylamine (DIPEA), then 0.600 g (3.43 mmol, 1 equiv) anh 6-amino-2-cyanobenzothiazole (1) was added. After stirring the reaction mixture overnight at room temperature (according to HPLC analysis the conversion was 97%), it was washed with water (2 × 7 mL), with saturated NaHCO3-solution (2 × 7 mL), then with water again (2 × 7 mL), and finally with saturated NaCl-solution (2 × 7 mL). It was dried over sicc Na2SO4, finally concentrated in vacuo. The resulting crude material was a pale yellow powder, its weight was 1.28 g, yield corresponding to the crude product: 68%. m/z [M + H]+ calcd for C30H25N5O4S 551.62, found 552.0 (Fig. S12). RP-HPLC (for the purified compound): 50–100% B in 25 min + 3 min up to 100% B + 100% B in 5 min, tR1 = 8.973 min: Fmoc-Gly-Pro-OH, tR2 = 17.868 min: Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole (Fig. S13).

Synthesis of Fmoc-Gly-Pro-6-Amino-d-Luciferin (3)

5.512 g (0.010 mol, 1 equiv.) Fmoc-Gly-Pro-6-amino-2-cyanobenzothiazole (2) was dissolved in 25 mL MeOH. 2.634 g (0.015 mol, 1.5 equiv.) d-cysteine∙HCl∙H2O, solved in 19 mL distilled water, was added to the solution at room temperature, under argon atmosphere, then the mixture was stirred continuously under pH control (starting pH: 2.27).

After 20 min’ stirring at room temperature, 30 mL, 5% (m/m) NaHCO3 was added dropwise over a period of 1 h to the mixture in order to release cysteine from its salt, while checking pH continuously. Reaching pH 2.6, a fine, yellow

Fig. 3 Luminescence is proportional to protease activity and N-Fmoc-GP-aLuc concentration. Different amount of a POP/PREP, b FAP and c Endoproteinase Pro-C proteases were titrated and assayed with 1–100 µmol N-Fmoc-GP-aLuc for 2 h. (d–f) The effect of protease inhibition was tested using the Complete protease inhibi-

tor cocktail, in each reaction with 10 µmol N-Fmoc-GP-aLuc and 32 pmol/min/reaction protease activity. Each point represents the average of 3 wells ± SD. Values are blank-subtracted (blank = no pro-tease). *p < 0.05; **p < 0.01; ***p < 0.001


1 3

solid material, Fmoc-Gly-Pro-6-amino-d-luciferin free car-boxylic acid, started to precipitate. At pH 6.3, this material started to dissolve, and at pH 7.40, it dissolved completely. Here the Fmoc-Gly-Pro-6-amino-d-luciferin formed Na-salt, which dissolved under these conditions.

After another 20 min’ stirring at room temperature, the organic solvent was evaporated. From the remaining aque-ous solution, a pale yellow solid material, Fmoc-Gly-Pro-6-amino-d-luciferin Na-salt precipitated partly. This aque-ous mixture was extracted with 3 × 15 mL ethyl acetate, in order to get rid of possible impurities. The combined organic layers were extracted with saturated NaCl solution. Hav-ing dropped the resulting solution on a mixture of ice and cc HCl, a fine yellow precipitate, Fmoc-Gly-Pro-6-amino-d-luciferin free carboxylic acid, formed. It was allowed to settle for 10 min, filtered and washed with 2 × 5 mL water, then air-dried to constant weight, which was 5.115 g (7.80 mmol), yield corresponding to the crude product: 78%. 1H NMR (500 MHz, [D6]DMSO) δ 10.39 (s, 1H), 8.59 (t, J = 15.85 Hz, 1H), 8.09 (d, J = 8.98 Hz, 1H), 7.88 (d, J = 7.43 Hz, 2H), 7.71 (d, J = 7.48 Hz, 2H), 7.66–7.60 (m, 1H), 7.48 (t, J = 5.65 Hz, 1H), 7.39 (q, J1 = 7.60 Hz, J2 = 15.29 Hz, 2H), 7.30 (q, J1 = 6.78 Hz, J2 = 13.76 Hz, 2H), 5.43 (t, J = 8.98 Hz, 1H), 4.47 (dd, J1 = 2.92 Hz, J2 = 5.21 Hz, 1H), 4.29–4.25 (m, 1H), 4.21 (q, J1 = 6.68 Hz, J2 = 14.87 Hz, 1H), 3.95–3.67 (m, 4H), 3.65–3.48 (m, 4H), 2.20–2.11 (m, 1H), 2.06–1.99 (m, 1H), 1.97–1.88 (m, 2H) (Fig. S14). 13C NMR (125 MHz, [D6]DMSO) δ 171.15, 171.05, 167.43, 164.43, 159.04, 156.55, 148.58, 143.86, 140.70, 138.38, 136.28, 127.61, 127.08, 125.27, 124.20, 120.10, 119.64, 111.52, 78.11, 65.70, 60.47, 46.62, 45.92, 42.72, 34.78, 29.28, 24.52 (Figure S15). m/z [M + H]+ calcd for C33H29N5O6S2 655.74 found 656.0 (Figure S16). RP-HPLC: 70–100% B in 15 min, tR= 12.608 min. TLC: tolu-ene/EtOH 50:30 saturated with water, Rf: 0.44.

Purification of Crude Fmoc-Gly-Pro-6-Amino-d-Luciferin (3)

160 mg crude peptide (45% desired material content, 72 mg) was dissolved in 1 mL N,N-dimethylformamide (DMF), then filtered, using a 0.45∝ m nylon filter. Gradient elution was used, 40–70% eluent B in 60 min at a 4 mL min−1 flow rate with detection at 220 nm. Pure fractions were collected and lyophilized to give a pale yellow material, the weight of which was 23 mg (0.035 mmol), yield corresponding to the isolated pure product: 32%.

Fmoc-Gly-Pro-aLuc Assay

The following proteases were used in the assay: recom-binant human POP/PREP, recombinant human FAP and Endoproteinase Pro-C at equivalent protease activ-ity in tenfold serial dilution starting from 32 fmol/min/

reaction to 320 pmol/min/reaction. Assay buffer for POP/PREP and Endoproteinase Pro-C contained 25 mmol tris(hydroxymethyl)aminomethane HCl-salt (Tris∙HCl) pH 7.4, 250 mmol NaCl, 2.5 mmol 1,4-dithiothreitol (DTT) and the assay buffer for FAP contained 50 mmol Tris∙HCl pH 7.4, 1 M NaCl, 1 mg/mL BSA. The N-Fmoc-GP-aLuc substrate was applied in 1 µmol to 100 µmol in 25 µL final reaction volume in a black plastic microtiter plate. The effect of protease inhibition (Complete protease inhibitor cocktail) was prepared by dissolving one tablet in 2 mL POP/PREP buffer and used in 2.5-fold dilution in each reaction with 10 µmol N-Fmoc-GP-aLuc and 32 pmol/min/reaction protease activity. After 2 h incubation at 37 °C 25 µL Luminescence Detection Reagent was added to each well. Luminescence was recorded within 5 min. The blank wells contained each component except proteases. Pre-sented values are blank-subtracted.

Statistics

Statistical significance was calculated by unpaired t-test (two-tailed, homoscedastic) between untreated and inhibi-tor containing samples.

Conclusion

We have developed an improved route for the synthesis of N-peptid-6-amino-d-luciferin conjugates. The method is more reliable than the standard practice as the preparation of one of the building blocks and two operations have been improved, which has led to better yields and significantly faster production time. The produced N-Fmoc-GP-aLuc was successfully used to measure FAP and POP/PREP enzyme activity in vitro.

This optimized method provides a practical and scalable way for the preparation of other N-peptide-6-amino-d-lucif-erin conjugates as well.

Author Contributions Conceptualization: László G. Puskás, Investiga-tion: Anita K. Kovács, Péter Hegyes, Gábor J. Szebeni, NMR analysis: Krisztián Bogár, Writing - original draft, review & editing: Anita K. Kovács, Supervision: Gábor K. Tóth.

Funding This work was partly supported by the following Grants: GINOP-2.3.2-15-2016-00030 and GINOP-2.3.2-15-2016-00001 from the National Research, Development and Innovation Office (NKFI), Hungary. Gábor J. Szebeni was supported by János Bolyai Research Scholarship of the Hungarian Academy of Sciences (BO/00139/17/8).

Compliance with Ethical Standards

Conflict of interest The authors declare no conflict of interest.


1 3

References

Carpino LA, Triolo SA, Griffin GW, Herman LW, Tarr G, Sole NA, Diekmann E, El-Faham A, Ionescu D, Albericio F (1996) Inno-vation and Perspectives. In: Solid phase synthesis & combinato-rial libraries: peptides, proteins and nucleic acids-small molecule organic chemical diversity. Collected Papers, 4th International Symposium, pp. 41–50, 1996

Chollet R, Ribault S (2012) Use of ATP Bioluminescence for rapid detection and enumeration of contaminants: the milliflex rapid microbiology detection and enumeration system. In Lapota D (ed) Recent advances in oceanic measurements and laboratory applica-tions, InTech, China, pp 99–118, ISBN: 978-953-307-940-0

Christiansen VJ, Jackson KW, Lee KN, Downs TD, McKee PA (2013) Targeting inhibition of fibroblast activation protein-alpha and pro-lyl oligopeptidase activities on cells common to metastatic tumor microenvironments. Neoplasia 15:348–358

Ciuffreda P, Casati S, Meroni G, Santaniello E (2013) A new syn-thesis of dehydroluciferin [2-(6 ‘-hydroxy-2 ‘-benzothiazolyl)-thiazole-4-carboxylic acid] from 1,4-benzoquinone. Tetrahedron 69:5893–5897

De Saint-Hubert M, Devos E, Ibrahimi A, Debyser Z, Mortelmans L, Mottaghy FM (2012) Bioluminescence imaging of therapy response does not correlate with FDG-PET response in a mouse model of Burkitt lymphoma. Am J Nucl Med Mol Imaging 2:353–361

Fontes R, Dukhovich A, Sillero A, Sillero MAG (1997) Synthesis of dehydroluciferin by firefly luciferase: effect of dehydroluciferin, coenzyme A and nucleoside triphosphates on the luminescent reaction. Biochem Bioph Res Co 237:445–450

Geiger R, Miska W (1991) US Patent US5035999AGryshuk AL, Perkins J, LaTour JV (2011) US Patent

US20110224442A1Hickson J, Ackler S, Klaubert D, Bouska J, Ellis P, Foster K, Oleksijew

A, Rodriguez L, Schlessinger S, Wang B, Frost D (2010) Nonin-vasive molecular imaging of apoptosis in vivo using a modified firefly luciferase substrate, Z-DEVD-aminoluciferin. Cell Death Diff 17:1003–1010

Hu H, Liu JT, Yao LP, Yin JP, Su N, Liu XQ, Cao F, Liang JM, Nie YZ, Wu KC (2012) Real-time bioluminescence and tomographic imaging of gastric cancer in a novel orthotopic mouse model. Oncol Rep 27:1937–1943

Jiang YL, Zhu Y, Moore AB, Miller K, Broome AM (2018) Bioti-nylated bioluminescent probe for long lasting targeted in vivo imaging of xenografted brain tumors in mice. ACS Chem Neu-rosci 9:100–106

Katz L (1951) Antituberculous Compunds II. 2-Benzalhydrazinoben-zothiazoles. J Am Chem Soc 73:4007–4010

Kovács AK, Hegyes P, Szebeni GJ, Nagy LI, Puskás LG, Tóth GK (2018) Synthesis of N-peptide-6-amino-d-luciferin Conjugates. Front Chem 6:120

Nyfeler R (1994) Peptide synthesis via fragment condensation. Meth-ods Mol Biol 35:303–316

O’Brian MA, Wood KV, Klaubert D, Daily B (2003) US Patent US20030211560A1

O’Brian MA, Daily WJ, Hesselberth PE, Moravec RA, Scurria MA, Klaubert DH, Bulleit RF, Wood KV (2005) Homogeneous, bio-luminescent protease assays: caspase-3 as a model. J Biomol Scr 10:137–148

Presiado I, Erez Y, Simkovitch R, Shomer S, Gepshtein R, da Silva LP, da Silva J, Huppert D (2012) Excited-state proton transfer of firefly dehydroluciferin. J Phys Chem A 116:10770–10779

Sadikot RTB, Timothy S (2005) Bioluminescence imaging. Proc Am Thorac Soc 2:537–540

Sato A, Klaunberg B, Tolwani R (2004) In vivo bioluminescence imag-ing. Comp Med 54:631–634

White EH, Wörther H, Seliger HH, McElroy WD (1966) Amino ana-logs of firefly luciferin and biological activity thereof 1. J Am Chem Soc 88:2015–2019

Xu TT, Close D, Handagama W, Marr E, Sayler G, Ripp S (2016) The expanding toolbox of in vivo bioluminescent imaging. Front Oncol 6:150

Zhang YM, Pullambhatla M, Laterra J, Pomper MG (2012) Influence of bioluminescence imaging dynamics by d-luciferin uptake and efflux mechanisms. Mol Imaging 11:499–506

optimized synthesis routes and biological ...doktori.bibl.u-szeged.hu/10082/1/dissertation...

Documents