2 Immunotoxins by Protein Ligation

Subscriber access provided by Kaohsiung Medical University

Article

Generation of potent anti-HER1/2 immunotoxins by protein ligation using split inteins Thomas Pirzer, Kira-Sophie Becher, Marcel Rieker, Tobias Meckel, Henning D. Mootz, and Harald Kolmar ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.8b00222 • Publication Date (Web): 19 Jun 2018 Downloaded from http://pubs.acs.org on June 20, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

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

ACS Chemical Biology

79x39mm (300 x 300 DPI)

ACS Paragon Plus Environment

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

Figure 1: Scheme of the protein trans-splicing reactions and final immunotoxins. a) Antibodies were coupled to protein A agarose first and incubated with MBP-toxins in the presence of the reducing agent TCEP at 25 °C for 24 h. Subsequently, uncoupled toxins and TCEP were washed off, antibodies were re-oxidized and eluted with citrate buffer pH 3 from the solid support. b) Final immunotoxins: Trast-Gelonin, Trast-PE24, 7D9G-Gelonin and 7D9G-PE24. 7D9G stands for the two camelid nanobodies 7D12 and 9G8. Molecular masses of all four ITs were calculated from primary sequences. 139x86mm (300 x 300 DPI)


Page 2 of 23

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


Figure 2: Protein trans-splicing reactions and subsequent purification. a) PTS reaction on protein A beads of 7D9G-Fc-IntN with MBP-Gelonin (top) and MBP-PE24 (bottom). The reaction was carried out in the presence of 2 mM TCEP for 24 h at 25 °C. Molecular masses of Ab-toxins are close to the MBP-Toxin educts and are marked accordingly. Lane 1 = Ab load; 2 = Ab flow-through; 3 = Toxin load; 4 = Toxin flow-through; 5 = Toxin wash; 6 = after oxidation; 7–8 = elutions 1–2. b) Exemplary IMAC purification of generated TrastGelonin immunotoxin. Unconjugated HC can be seen in the flow-through. Enriched HC-Toxin is eluted with an approximately 2:1 ratio of conjugated to unconjugated HC. Lane 1 = IT input; 2 = flow-through; 3 = wash; 4–6 = elution fractions 1–3. c) Final ITs after PTS and IMAC purification with a TAR of approximately 1.3. Lane 1 = 7D9G-Gelonin; 2 = Trast-Gelonin; 3 = 7D9G-PE24; 4 = Trast-PE24. d) Table showing all protein fragments depicted in a)–c) with respective abbreviations and molecular masses. 138x96mm (300 x 300 DPI)



Figure 3: Determination of binding affinities of ITs on cells. Antigen over-expressing cells were incubated with varying concentrations of parental antibodies or immunotoxins and analyzed by flow cytometry. The relative mean fluorescence was plotted against antibody concentrations and KD values were determined using sigmoidal fitting with GraphPad Prism software (GraphPad Software, Inc.). 41x66mm (300 x 300 DPI)


Page 4 of 23

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


Figure 4: Cytotoxicity assay shows high toxicity for PE24 ITs and lower toxicity for gelonin derivatives. Different antigen presenting cells were treated for 72 h with antibodies, immunotoxins and toxins alone. The upper row shows data of cell lines treated with HER1 targeting 7D9G conjugates and the lower row shows data of HER2 targeting trastuzumab conjugates. CHO cells were used as negative control because of their lack of HER1 or HER2 expression. The assay was performed in triplicate. The relative survival to untreated cells was plotted against antibody concentrations and IC50 values were determined using sigmoidal fitting with GraphPad Prism (GraphPad Software, Inc.). 139x69mm (300 x 300 DPI)



Figure 5: Immunotoxins are endocytosed efficiently as shown by confocal microscopy. MDA-MB-468 cells (HER1 overexpressing) cells were treated for 1h with 7D9G ITs, washed with low pH glycine buffer to remove surface bound conjugates and stained directly or incubated for another 3h before staining. CHO cells were used as negative cell line. Anti-hIgG Fab-Alexa 488 (1:500) and anti-His6 Alexa 647 antibody (1:1000) were used to specifically visualize the antibody fraction and the toxin fraction of the ITs, respectively. Scale bar = 10 µm. 139x73mm (300 x 300 DPI)


Page 6 of 23

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


1

Generation of potent anti-HER1/2 immunotoxins by protein ligation using

2

split inteins

3 4 5

Thomas Pirzer1, Kira-Sophie Becher2, Marcel Rieker3, Tobias Meckel4, Henning D. Mootz2,

6

Harald Kolmar1*

7 8 9

1

Institute for Organic Chemistry and Biochemistry, Technische Universität Darmstadt, Alarich-

10

Weiss-Strasse 4, D-64287 Darmstadt, Germany

11

2

12

Germany

13

3

14

250, D-64293 Darmstadt, Germany

15

3

16

64293 Darmstadt, Germany

17

4

18

Universität Darmstadt, Alarich-Weiss-Str. 8, D-64287 Darmstadt, Germany

19

*

20

Harald Kolmar: Institute for Organic Chemistry and Biochemistry, Technische Universität

21

Darmstadt, Alarich-Weiss-Strasse 4, D-64287 Darmstadt, Germany

22

E-Mail: [email protected]

Institute of Biochemistry, University of Münster, Wilhelm-Klemm-Str. 2, D-48149 Münster, Antibody Drug Conjugates and Targeted NBE Therapeutics, Merck KGaA, Frankfurter Straße Protein Engineering and Antibody Technologies, Merck KGaA, Frankfurter Straße 250, DMacromolecular Chemistry & Paper Chemistry, Department of Chemistry, Technische

To whom correspondence should be addressed:

23 24 25

Keywords: Immunotoxin, split intein, protein ligation, antibody conjugation, protein

26

engineering,

27

-1ACS Paragon Plus Environment


28

Abstract

29

Cell targeting protein toxins have gained increasing interest for cancer therapy, aimed at

30

increasing the therapeutic window and to reduce systemic toxicity. Since recombinant

31

expression of immunotoxins consisting of a receptor-binding and a cell-killing moiety is

32

hampered by their high toxicity in an eukaryotic production host, most applications rely on

33

recombinant production of fusion proteins consisting of an antibody fragment and a protein

34

toxin in bacterial hosts such as Escherichia coli (E. coli). These fusions often lack beneficial

35

properties of whole antibodies like extended serum half-life or efficient endocytic uptake via

36

receptor clustering. Here we describe the production of full-length antibody immunotoxins

37

using self-splicing split inteins. To this end, the short (11 amino acids) N-terminal intein part

38

of the artificially designed split intein M86, a derivative of the Ssp DnaB intein, was

39

recombinantly fused to the heavy chain of trastuzumab, a human epidermal growth factor

40

receptor 2 (HER2) receptor targeting antibody and to a nanobody-Fc fusion targeting the HER1

41

receptor, respectively. Both antibodies were produced in Expi293F cells. The longer C-terminal

42

counterpart of the intein was genetically fused to the protein toxins gelonin or Pseudomonas

43

Exotoxin A, respectively and expressed in E. coli via fusion to maltose binding protein. Using

44

optimized in vitro splicing conditions we were able to generate a set of specific and potent

45

immunotoxins with IC50 values in the mid- to subpicomolar range.

46 47

Introduction

48

Antibodies (Abs) are currently widely used for therapy of several indications like inflammatory

49

diseases, haemophilia, autoimmune diseases and cancer. Besides classical antibodies that rely

50

on their agonistic/antagonistic properties or intrinsic effector functions, i.e. antibody-dependent

51

cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC), antibodies can

52

be modified resulting in antibody-drug conjugates (ADCs).1 These combine the specific

53

targeting properties of antibodies with highly toxic small molecules. Upon binding to a

54

malignant cell, the ADC is internalized by receptor-mediated endocytosis and guided to the

55

lysosome, where the toxic payload is released, either by proteolysis of the antibody or the

56

cleavage of a designed linker. Different classes of toxins are currently used, among them

57

microtubule inhibitors (e.g. auristatins) or DNA alkylating agents (e.g. duocarmycins).2,3

58

Beside small molecules, bacterial and plant protein toxins can be conjugated to antibody

59

fragments and the resulting immunotoxins (ITs) have been investigated for their potential

60

application in therapy.4 Compared to their small molecule counterparts, protein toxins can cause

61

over-stoichiometric cell damage due to their inherent enzymatic activities, thereby reducing the -2ACS Paragon Plus Environment

Page 8 of 23

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


62

necessary amount of molecules to reach the cytosol.5 The translation inhibiting Pseudomonas

63

Exotoxin A (PE) is one of the most prominent proteins in this class, mainly because of its good

64

producibility in Escherichia coli and its effective translocation mechanisms into the cytosol of

65

the target cell.6 In recent years five PE constructs entered clinical trials and currently at least

66

two next generation ITs are still under clinical investigation.4,7,8 One of the main problems is

67

still the high immunogenicity, although de-immunized variants have been developed to

68

circumvent this issue. The most recent de-immunized variant is a 24 kD large fragment of PE

69

(PE24) that encompasses an improved furin cleavage site and the cytotoxic domain III.9 After

70

tumor-specific uptake that is mediated by the fused antibody or antibody fragment the toxin is

71

released in the endosome by endosomal proteases and translocated into the cytoplasm, where it

72

induces cell killing via inhibition of ribosomal protein synthesis. Due to their high toxicity and

73

blockage of the translational machinery in an eukaryotic expression host, most immunotoxins

74

(ITs) are produced in bacterial hosts as fusions to antibody fragments like antigen binding

75

fragments (Fabs) or single chain variable fragments (scFvs), which limits their use.10–12

76

Besides bacterial toxins such as PE, some ribosome inactivating plant toxins were intensively

77

studied to be used as immunotoxins. They all belong to the family of ribosome inactivating

78

proteins (RIPs) that are divided in class 1 and 2, depending whether or not they contain a cell

79

targeting domain in addition to the ribosome inactivating domain. Class 1 RIPs like gelonin

80

have the advantage of a very low off-target toxicity since they are naturally lacking a cell

81

recognition domain.13 Gelonin inhibits cellular protein translation in eukaryotes by hydrolyzing

82

the 28S rRNA of the 60S ribosomal subunit.13 As for PE this is a catalytic process so that only

83

a few molecules are needed to kill a cell. Nevertheless it has been observed that a certain

84

threshold concentration has to be reached intracellularly for effective cell killing. 14 Transfer

85

from the endosomal lumen into the cytosol is still a major obstacle since gelonin has no active

86

translocation mechanism. Nevertheless, alternative ways for efficient endosomal release have

87

been investigated in the last years.15–17

88

For the generation of immunotoxins fusion protein expression can be considered or,

89

alternatively, chemical conjugation strategies as well as site-specific enzymatic conjugation

90

methods can be applied aimed at linking an antibody and a protein toxin. An elegant way of

91

generating protein-protein conjugates uses inteins, which are self-excising proteins that are

92

found in all domains of life (archaea, bacteria and eukaryotes).18 They are naturally flanked by

93

exteins, which are assembled through the splicing event. Because of their unique mechanism,

94

inteins can be regarded as single turnover enzymes that do not rely on an energy source or

95

cofactors.19 Besides classical inteins, buried in a pre-protein, naturally occurring split inteins -3ACS Paragon Plus Environment


96

have been reported that reassemble after mixing both N- and C-terminal intein parts (IntN and

97

IntC) and therefore perform the splicing reaction in trans.20–23 Designed artificial split inteins

98

have opened new possibilities for protein ligation.24–26 We have chosen an artificially evolved

99

split intein with an unusual short N-terminal part that showed very good splicing yields in a

100

previous study.27 This intein was chosen for this work since we expected not only good yields

101

but also minor effects of the shortened IntN on antibody production and stability.

102

In this study a truncated, optimized variant of Pseudomonas Exotoxin A and gelonin obtained

103

by expression in E. coli were used for the generation of immunotoxins derived from the full-

104

length antibody trastuzumab as well as a single-chain antibody targeting human epidermal

105

growth factor receptors HER2 and HER1/EGFR, respectively (Figure 1B). Upon optimization

106

of splicing conditions immunotoxins with high potency were obtained that also showed

107

selectivity.

108 109

RESULTS AND DISCUSSION

110

Antibody and toxin constructs. Two antibodies were used to generate immunotoxins and

111

evaluate their functionality. Trastuzumab is a well characterized HER2 binding antibody that

112

is approved by the FDA as single treatment as well as in the form of the ADC trastuzumab-

113

emtansin (T-DM1) for the treatment of aggressive breast tumors.28 A glycine serine linker

114

(GGGSGGG), followed by the 11 amino acids (aa) of the IntN, was attached to the C-terminus

115

of the heavy chain (Supplementary Figure S1). The second targeting antibody was constructed

116

from two camelid heavy-chain antibody domains (VHHs) targeting HER1 (EGFR) that were

117

already described and well characterized in literature. The 7D12 and 9G8 nanobodies bind to

118

HER1/EGFR with an affinity in the double digit nanomolar range on cells and were used in the

119

optimized 7D12-9G8 orientation (later called 7D9G) with a linker length of 10 aa, which leads

120

to an optimized KD in the single digit nanomolar range.29 The VHH domains were shown to

121

bind different epitopes in the domain III and in the junction of domain II and III of HER1 and

122

induce rapid internalization by receptor clustering.30,31 Both VHH domains were placed on an

123

IgG1 Fc domain and terminated with a slightly longer C-terminal glycine serine linker

124

((GGGGS)3) and the IntN sequence (Supplementary Figure S1).

125

Both antibodies were then expressed in Expi293F cells and purified from the supernatant by

126

protein A chromatography. Both antibodies showed a monomer content >95 % in size exclusion

127

chromatography (SEC) analysis (Supplementary Figure S2). The toxin constructs were

128

composed of a maltose binding protein for better solubility and producibility in E. coli followed

129

by the 143 aa long C-terminal intein part (IntC) and the toxin itself (Supplementary Figures S1 -4ACS Paragon Plus Environment

Page 10 of 23

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


130

and S5). PE24 already contained an optimized furin cleavable sequence, which was also

131

inserted into the gelonin construct to provide efficient proteolytic release in the

132

endosome/lysosome.9 Both toxins were produced in E. coli BL21 (DE3) and purified by

133

immobilized metal ion affinity chromatography (IMAC) and SEC (Supplementary Figure S3).

134

The monomeric fractions were then used for the trans-splicing reaction.

135

Protein trans-splicing (PTS) on protein A beads. Antibodies were coupled to protein A

136

agarose beads for 30 min at room temperature (RT). The toxins were added at an approximately

137

6-fold molar excess (3-fold relative to each heavy chain) together with 2 mM TCEP to achieve

138

complete reduction of the splicing-active N-terminal cysteine in the IntN. The reaction was

139

stopped after 24 h by column wash and immunotoxins were eluted after antibody re-oxidation

140

with dehydroascorbic acid. The whole procedure is depicted in Figure 1A. Since coupling

141

efficiencies were in the range of 50–70 %, as determined by densitometry of SDS-PAGE gels

142

(Figure 2A), unconjugated antibody was removed by a subsequent IMAC (Figure 2B). It was

143

not possible to distinguish molecules with a toxin/antibody ratio (TAR) of 2 from those with an

144

incomplete coupling. The resulting final constructs showed a TAR of approximately 1.3, as

145

determined by densitometry from a reducing SDS-PAGE gel and were used for further analysis

146

(Figure 2C).

147

Immunotoxins retain enzymatic function and inhibit protein translation in vitro. Since

148

elution from protein A includes a low pH step that might harm the structural and functional

149

integrity of the respective toxin protein, the activity of ITs was assessed in a protein translation

150

assay in vitro using rabbit reticulocyte lysate. Various concentrations were added to the lysate

151

and the translation of luciferase was analyzed by measuring the luminescence of a chromogenic

152

substrate. While gelonin showed inhibition in the picomolar range (IC50 = 251 pM) as reported

153

before (Supplementary Figure S4), PE24 did not show inhibition of rabbit ribosomal protein

154

synthesis in this assay (data not shown). This is probably due to the fact that NAD+ is needed

155

for activity of PE24 and not enough of this co-factor may be included in the rabbit lysate.32

156

Both trastuzumab and 7D9G conjugates to gelonin showed IC50 values in the same range

157

compared to the parental toxin with half maximal inhibitions at 183 pM and 84 pM, respectively

158

(Supplementary Figure S4). The IC50 values of ITs were expected to be slightly lower than for

159

the parental toxins because of a stoichiometry of 1–2 toxins per antibody.

160

Immunotoxins retain cell binding properties of parental antibodies. To study whether

161

attached toxins have an influence on the binding affinities of the antibody, both parental

162

antibodies and respective immunotoxins were titrated on target protein expressing cells to

163

determine binding constants. Both gelonin and PE24 ITs showed only slight deviations in KD -5ACS Paragon Plus Environment


164

values compared to the parental antibodies (Figure 3). 7D9G binding to MDA-MB-468 cells

165

showed an expected high affinity binding with a KD of 4.5 ± 1.0 nM. The gelonin and PE24

166

conjugates had slightly decreased binding affinities of 5.9 ± 0.6 nM and 6.1 ± 0.4 nM,

167

respectively. The same was observed for trastuzumab. While the parental antibody bound to

168

HER2 overexpressing SK-BR-3 cells with a KD of 8.2 ± 0.3 nM, the gelonin and PE24

169

conjugates showed dissociation constants of 9.1 ± 1.3 and 9.7 ± 0.4, respectively. Thus, the

170

attachment of such bulky additives at the C-terminus of both antibodies does not impair their

171

binding properties.

172

High toxicity for both PE24 and gelonin derived immunotoxins. As biochemical properties

173

of all ITs looked promising, they were tested in cellular assays to assess their cytotoxicity and

174

specificity. Four different cell lines with different receptor densities on their surfaces were used.

175

A549 have low to medium EGFR expression, MDA-MB-468 show high EGFR expression, SK-

176

BR-3 are HER2 high-expressing cells and CHO-KI were used as negative cell line. Antibodies,

177

toxins and ITs were incubated with the cells for 72h before measuring cell viability in a standard

178

MTS assay. Different concentration scales had to be used for gelonin and PE24 in the

179

nanomolar and picomolar range, respectively, because of different inherent toxicities.

180

On the negative cell line CHO, none of the investigated ITs and controls displayed growth

181

inhibition up to the maximal tested concentrations (Figure 4 left). The single-chain IT with

182

7D9G targeting HER1 showed potent cell killing of MDA-MB-468 cells compared to the

183

antibody and the toxin alone (Figure 4 upper middle). The effective ratio (ER) of IC50 value

184

(IC50 of IT divided by that of the toxin, Table 1) of 7D9G-Gelonin to MBP-Gelonin of about

185

771 showed the high selectivity and enhanced uptake of the antibody-toxin conjugate. With a

186

half-maximal inhibition at only 68.0 ± 4.2 pM, this conjugate is highly toxic to this target cell

187

line. Interestingly, the same IT did not show any toxicity in the other EGFR expressing cell line

188

A549, which has a receptor density about 50–fold lower than that of MDA-MB-468.33 The

189

HER2 targeting Trast-Gelonin conjugate showed an even lower inhibitory constant of 11.8 ±

190

1.6 pM (Table 1) on overexpressing SK-BR-3. Combined with an ER of over 2.5*107 this is

191

the highest gap between toxin and antibody-toxin conjugates in this whole data set.

192 193 194 195 196 197 -6ACS Paragon Plus Environment

Page 12 of 23

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

198 199 200


Table 1: Cytotoxicity of immunotoxins on target and control cell lines. Antibodies, toxins and immunotoxins were tested for cytotoxicity in a MTS cell viability assay. IC50 values were calculated by sigmoidal curve fitting using GraphPad Prism (GraphPad Software, Inc.). Values were calculated from triplicate samples. CHO Target

HER1

HER2

A549

MDA-MB-468

SK-BR-3

Protein

IC50 (nM)

ER

IC50 (pM)

ER

IC50 (pM)

ER

IC50 (pM)

ER

7D9G-Fc

>300

n.d.

>300000

n.d.

>300000

n.d.

n.d.

n.d.

7D9G-Gelonin

>300

n.d.

>300000

n.d.

68.0 ± 4.2

771

n.d.

n.d.

7D9G-PE24

>1

n.d.

3.9 ± 0.5

412

0.8 ± 0.1

290

n.d.

n.d.

Trastuzumab

>300

n.d.

n.d.

n.d.

>300000

n.d.

>300000

n.d.

Trast-Gelonin

>300

n.d.

n.d.

n.d.

n.d.

n.d.

11.8 ± 1.6

>2.5E+7

Trast-PE24

>10

n.d.

n.d.

n.d.

59.9 ± 3.2

4

2.4 ± 0.1

233

MBP-Gelonin

>300

1

>300000

1

5244 ± 1316

1

>300000

1

MBP-PE24

>10

1

1644 ± 247

1

239 ± 26

1

568 ± 58

1

201 202

Similar results were obtained for the PE24 ITs. In the analyzed concentration range up to

203

10 nM, no toxicity can be observed in CHO cells for all analytes (Figure 4 lower panel). The

204

7D9G-PE24 constructs show high potency in both EGFR positive cell lines A549 and MDA-

205

MB-468 compared to the toxin alone with IC50 values of 3.9 ± 0.5 and 0.8 ± 0.1 pM, respectively

206

(Table 1, Figure 4). Effective ratios of 412 and 290 for A549 and MDA-MB-468, respectively,

207

show a high contribution of receptor mediated internalization of the antibody fraction to the

208

ultimate toxicity. The sub-picomolar IC50 value of 7D9G-PE24 of 0.8 ± 0.1 pM depicts the

209

highest toxicity in this study. In this case, the lower number of surface exposed receptors on

210

A549 cells did not have such a high impact as for the gelonin construct which indicates that

211

also the efficiency of endosomal escape may play an important role for IT efficacy.

212

Additionally, the Trast-PE24 IT showed a 233-fold decreased IC50 of 2.4 ± 0.1 pM compared

213

to the toxin alone.

214

Time-resolved toxin internalization to target cells. To investigate why PE24 is several orders

215

of magnitude more efficient in cell killing, confocal microscopy was applied aimed at

216

monitoring a time-dependent release of the toxin into the cytosol. Cells were treated with ITs

217

for one or four hours and the respective route of the IgG and the toxin were followed by staining

218

them with specific antibodies. The antibody fraction was visualized using an Alexa488 labeled

219

-IgG Fab fragment and the toxin was stained with an Alexa647 labelled -His6 antibody.

220

HER1 positive MDA-MB-468 cells showed endocytic uptake of 7D9G-Gelonin and 7D9G-

221

PE24 after 1h of incubation, marked by a punctate pattern in the 488-channel inside the cell

222

(Figure 5). This was also confirmed for the trastuzumab ITs on SK-BR-3 cells (Supplementary

223

Figure S5). Additionally, after 1 h, signals for the antibody part and the toxin part colocalize

224

for both the gelonin and PE24 IT. After four hours, however, the toxin fluorescence of the PE24

225

IT is clearly reduced in the vesicles. Gelonin still seems to be attached to the antibody at this -7ACS Paragon Plus Environment


226

time point as a high degree of colocalization of antibody and gelonin is observed. This may be

227

explained by the retrograde transport mechanism of PE246 that may result in a more efficient

228

and faster transport to the cytosol than gelonin.

229

Discussion. In this study we showed the conjugation of two toxic proteins to two antibody

230

scaffolds by protein trans-splicing. With this site-specific ligation, four immunotoxins were

231

produced, which retained their target binding and toxic properties, hence being highly toxic on

232

tumor cells expressing the target receptors. Although the intein splicing reaction was not

233

quantitative, the purified ITs contained at least one toxin molecule per antibody.

234

Different protein conjugation strategies have been reported in previous years, ranging from

235

enzymatic reactions using sortase A34, microbial transglutaminase (mTG)35, or SpyLigase36.

236

Although sortase A has a high substrate specificity and has already been used to fuse the

237

trastuzumab Fab fragment to gelonin to increase its potency11, generally a high excess (10-20

238

fold) of the oligoglycine containing substrate is needed to drive the equilibrium reaction to the

239

side of the desired product.37 Microbial transglutaminase forms a stable isopeptide bond

240

between the -carboxamide group of glutamine and the primary -amino group of lysine.35,38,39

241

Although mTG has been used successfully in the generation of ADCs40 it shows to be highly

242

promiscuous towards the acyl acceptor41 and thus makes coupling of proteins with several

243

superficial lysines a trial and error approach. Furthermore, it showed to have a substrate

244

preference for glutamines in special sequence and structural context that is currently little

245

understood. While IgG1 antibodies have no acyl donor glutamine, they can be specifically

246

conjugated with mTG after deglycosylation or by introduction of a glutamine containing

247

tag.42,43 In human growth hormone two of several available glutamines are addressed resulting

248

in a heterogeneous reaction product.44 The SpyTag/SpyCatcher technology proved to be highly

249

specific with no or only a slight excess of one reaction partner over the other needed for a

250

quantitative turnover to the final product. This method nevertheless has the disadvantage of

251

significant peptide tags (23 aa) remaining in the product.36,45 Protein trans-splicing thus offers

252

an advantageous combination of specificity, efficiency and speed without inserting large

253

foreign sequences into the final construct.

254

A PTS efficiency of up to 70 % was achieved, which is fairly high considering the molecular

255

sizes of both splicing partners of 150 and 90 kDa. In the original assay, the evolved M86 split

256

intein showed up to 90 % splicing efficiency, which was performed with a 30 kDa IntC and an

257

IntN peptide with less than 2 kDa.27 While trastuzumab contained a shorter linker of only 7 aa

258

and glycine at the –1 position of the IntN, which is favored for the M86 intein, the 7D9G

259

antibody contained a longer linker of 15 aa providing more flexibility but has a serine at the –1 -8ACS Paragon Plus Environment

Page 14 of 23

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


260

position, possibly decreasing PTS efficiency.27,46 Thus, optimization of linker length and amino

261

acids composition could further improve the splicing reaction. Besides that, increasing the

262

equivalents of the toxin portion to the antibody could also help increasing the coupling

263

efficiency, although protein solubility could be a limiting factor. Protein solubility of the IntC

264

containing splicing partner was also critical because inefficient splicing was observed for

265

aggregated portions after concentrating the protein. Choosing a different split intein could also

266

be an alternative, as shown for the generation of bispecific antibodies by Han and coworkers in

267

2017.26 They used a split Npu DnaE intein which was reported to have extremely fast splicing

268

kinetics and could achieve efficient conjugations of antibody scaffolds in the same size range

269

as they were used here.47 Another engineered Npu DnaE split intein was used for direct labelling

270

of antibodies in culture medium.48 However, the much larger IntN domain (123 amino acids

271

compared to 11) might impact antibody production yields and stability.

272

To date, mostly antibody fragments have been linked to toxic proteins to target them to receptor

273

expressing cells.11,49–51 This was mainly driven by the fact that the complete immunotoxin can

274

be produced in prokaryotes and can thus be generated in larger amounts. There are also reports

275

that made use of toxins containing Fc binding domains from protein A or G which proved to

276

induce cell-specific toxicity when bound to a full antibody.49,52 Nevertheless, this approach is

277

useful only in screening campaigns and not for therapeutic purposes. Full-length antibodies

278

have other advantages over antibody fragments such as an extended half-life in blood53, low

279

immunogenicity and increased stability mediated through the glycosylated Fc domain.54

280

Additionally, an antibody is naturally bivalent and therefore in the position to trigger efficient

281

cellular uptake through receptor clustering.31,55 The VHH antibody format that was applied in

282

this study may combine the benefits of both antibody fragments and conventional antibodies.

283

The glycosylated Fc domain adds physicochemical stability and increases the serum half-life.

284

The VHH construct has a more simple structure due to the lack of an additional light chain and

285

retains a bivalent, or in this case, even tetravalent binding to its antigen, leading to an increased

286

receptor-mediated uptake. This may also be the reason for an increased efficacy of the generated

287

ITs in this study, compared to previously reported conjugates. All produced Pseudomonas

288

Exotoxin A conjugates in this study were more potent than the most toxic PE IT reported to

289

date, with an IC50 of 7D9G-PE24 of 0.8 pM on MDA-MB-468 cells being more potent by a

290

factor of 24.56 Although the toxic moiety is comparable in this case, different receptors were

291

targeted and ITs were tested on different cell lines, which may account for some difference in

292

efficacy. More comparability is given for the gelonin conjugates to trastuzumab that were tested

293

on SK-BR-3 cells. The same combination was tested with a Fab-toxin conjugates in 2014 with -9ACS Paragon Plus Environment


294

an EC50 of 600 pM on the same cell line which is less potent by a factor of 50 compared to the

295

tested full-length Trast-Gelonin in this study.11 The endosomal escape mechanism may also

296

play a role in the differences of toxicities between gelonin and PE24. For the latter, a defined

297

escape mechanism via a retrograde ER transport has been described in detail,57 while for

298

gelonin no such mechanism is known. As a consequence, PE24 has the advantage over gelonin

299

of an active transport mechanism that likely contributes to its remarkable potency. To sum up,

300

we generated highly potent immunotoxins by applying protein trans-splicing with split inteins.

301

This method is fast and yields site-specific conjugates with minimal changes to the native

302

sequences.

303 304

METHODS

305

Detailed information on used cell lines and cultivation conditions, protein productions and

306

purifications, cloning of expression plasmids, immunostainings, confocal microscopy and in

307

vitro protein translation assay are available in the Supplementary.

308

Protein trans-splicing on protein A agarose beads. Protein splicing reactions were performed

309

on a solid support as described previously for coupling cytotoxic agents to reduced cysteines of

310

antibodies.58 Abs were first incubated with protein A agarose (protein A HP SpinTrap, GE

311

Healthcare) for 30 min under shaking in intein buffer (50 mM Tris/HCl, 300 mM NaCl, pH 7).

312

The supernatant was removed by centrifugation and the slurry washed 2x with intein buffer.

313

Toxin at 6-fold molar excess towards the antibody was then added with 2 mM TCEP as reducing

314

agent and incubated with the beads for 18–24 h at 25 °C under continuous shaking. Unreacted

315

toxin was removed by centrifugation and residual reducing agent was washed away 3x with

316

PBS. Immunotoxins were re-oxidized in PBS with 100 eq. of dehydroascorbic acid for 2 h at

317

37 °C. ITs were eluted after 2 washing steps with elution buffer, diluted in IMAC A buffer and

318

subjected to IMAC.

319

Cytotoxicity assay. Cell viability after addition of immunotoxins was assessed with the

320

CellTiter 96® AQueous One Solution Cell Proliferation Assay. 5,000 cells were seeded per

321

well in a 96-well plate in the corresponding medium with additional pen/strep solution and

322

incubated for 24 h. Cells were then treated in triplicate with varying concentrations of

323

antibodies, toxins or immunotoxins. After 72 h, 20 µl of the MTS solution were added per well

324

and the plates were incubated for 0.5 to 3 h under standard conditions. Last, the absorption was

325

measured at 485 nm in a Tecan reader. IC50 values were calculated by using sigmoidal fitting

326

with GraphPad Prism software (GraphPad Software, Inc.) and were not normalized to the TAR - 10 ACS Paragon Plus Environment

Page 16 of 23

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


327

since all ITs displayed the same TAR of 1.3. Cell viability of reference wells with untreated

328

cells was set to 100 %.

329

Determination of KD by flow cytometry. Receptor over-expressing cells were incubated with

330

the analytes at different concentrations and subsequently with specific fluorogenic secondary

331

antibodies. Cells were then conducted to flow cytometry. For more details, see Supplementary

332

section 2.

333 334

ASSOCIATED CONTENT

335

Supporting Information

336

The Supporting Information is available free of charge on the ACS Publications website.

337

Supporting figures (S1–S5), protein sequences and supplemental methods (PDF).

338 339

ACKNOWLEDGEMENTS

340

This work was supported by Deutsche Forschungsgemeinschaft in the course of priority program

341

SPP 1623.

342 343

AUTHOR CONTRIBUTIONS

344

T.P. wrote the manuscript, conducted the cloning of antibody constructs, produced all proteins

345

and conducted all biochemical and cell biological characterizations. H.K. and H.M. directed

346

and supervised the project. K.-S.B. provided the intein sequences and performed cloning of

347

MBP-Gelonin vector. M.R. provided pTT5 expression plasmids for trastuzumab and gave

348

advices for the cytotoxicity assays, antibody analytics and cell culture. T.M. performed confocal

349

microscopy. All authors reviewed the manuscript.

350 351

Corresponding autor:

352

*E-Mail: [email protected]

353 354

- 11 ACS Paragon Plus Environment


355

REFERENCES

356

(1) Thomas, A., Teicher, B. A., and Hassan, R. (2016) Antibody–drug conjugates for cancer

357

therapy. Lancet Oncol. 17, e254–e262.

358

(2) Chen, H., Lin, Z., Arnst, K. E., Miller, D. D., and Li, W. (2017) Tubulin inhibitor-based

359

antibody-drug conjugates for cancer therapy. Molecules 22.

360

(3) Elgersma, R. C., Coumans, R. G. E., Huijbregts, T., Menge, W. M. P. B., Joosten, J. A. F.,

361

Spijker, H. J., De Groot, F. M. H., Van Der Lee, M. M. C., Ubink, R., Van Den Dobbelsteen,

362

D. J., Egging, D. F., Dokter, W. H. A., Verheijden, G. F. M., Lemmens, J. M., Timmers, C. M.,

363

and Beusker, P. H. (2015) Design, synthesis, and evaluation of linker-duocarmycin payloads:

364

Toward selection of HER2-targeting antibody-drug conjugate SYD985. Mol. Pharm. 12, 1813–

365

1835.

366

(4) Alewine, C., Hassan, R., and Pastan, I. (2015) Advances in Anticancer Immunotoxin

367

Therapy. Oncologist 20, 176–185.

368

(5) Kreitman, R. J., and Pastan, I. (1998) Accumulation of a recombinant immunotoxin in a

369

tumor in vivo: Fewer than 1000 molecules per cell are sufficient for complete responses. Cancer

370

Res. 58, 968–975.

371

(6) Smith, D. C., Spooner, R. A., Watson, P. D., Murray, J. L., Hodge, T. W., Amessou, M.,

372

Johannes, L., Lord, J. M., and Roberts, L. M. (2006) Internalized pseudomonas exotoxin A can

373

exploit multiple pathways to reach the endoplasmic reticulum. Traffic 7, 379–393.

374

(7) ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000 Feb

375

29 - . Identifier NCT01362790, SS1P and Pentostatin Plus Cyclophosphamide for

376

Mesothelioma; 2011 May 30, [cited 2018 Apr 30]; [about 10 screens]. Available from:

377

https://clinicaltrials.gov/ct2/show/NCT01362790.

378

(8) ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000 Feb

379

29 - . Identifier NCT02798536, Mesothelin-Targeted Immunotoxin LMB-100 in People With

380

Malignant Mesothelioma; 2016 June 14 [cited 2018 Apr 30]; [about 9 screens]. Available from:

381

https://clinicaltrials.gov/ct2/show/NCT02798536.

382

(9) Bauss, F., Lechmann, M., Krippendorff, B. F., Staack, R., Herting, F., Festag, M., Imhof-

383

Jung, S., Hesse, F., Pompiati, M., Kollmorgen, G., da Silva Mateus Seidl, R., Bossenmaier, B.,

384

Lau, W., Schantz, C., Stracke, J. O., Brinkmann, U., Onda, M., Pastan, I., Bosslet, K., and

385

Niederfellner, G. (2016) Characterization of a re-engineered, mesothelin-targeted Pseudomonas

386

exotoxin fusion protein for lung cancer therapy. Mol. Oncol. 10, 1317–1329.

387

(10) Shin, M. C., Min, K. A., Cheong, H., Moon, C., Huang, Y., He, H., and Yang, V. C. (2016)

388

Preparation and Characterization of Gelonin-Melittin Fusion Biotoxin for Synergistically - 12 ACS Paragon Plus Environment

Page 18 of 23

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


389

Enhanced Anti-Tumor Activity. Pharm. Res. 33, 2218–2228.

390

(11) Kornberger, P., and Skerra, A. (2014) Sortase-catalyzed in vitro functionalization of a

391

HER2-specific recombinant Fab for tumor targeting of the plant cytotoxin gelonin. MAbs 6,

392

354–366.

393

(12) Cao, Y., Marks, J. D., Marks, J. W., Cheung, L. H., Kim, S., and Rosenblum, M. G. (2009)

394

Construction and characterization of novel, recombinant immunotoxins targeting the Her2/neu

395

oncogene product: In vitro and in vivo studies. Cancer Res. 69, 8987–8995.

396

(13) Stirpe, F., Olsnes, S., and Pihl, A. (1980) Gelonin , a New Inhibitor of Protein Synthesis ,

397

Nontoxic to Intact Cells. J. Biol. Chem. 255, 6947–6953.

398

(14) Pirie, C. M., Hackel, B. J., Rosenblum, M. G., and Wittrup, K. D. (2011) Convergent

399

potency of internalized gelonin immunotoxins across varied cell lines, antigens, and targeting

400

moieties. J. Biol. Chem. 286, 4165–4172.

401

(15) Yang, N. J., Liu, D. V., Sklaviadis, D., Gui, D. Y., Vander Heiden, M. G., and Wittrup, K.

402

D. (2015) Antibody-mediated neutralization of perfringolysin o for intracellular protein

403

delivery. Mol. Pharm. 12, 1992–2000.

404

(16) Selbo, P. K., Sandvig, K., Kirveliene, V., and Berg, K. (2000) Release of gelonin from

405

endosomes and lysosomes to cytosol by photochemical internalization. Biochim. Biophys. Acta

406

- Gen. Subj. 1475, 307–313.

407

(17) Pirie, C. M., Liu, D. V, and Wittrup, K. D. (2013) Targeted cytolysins synergistically

408

potentiate cytoplasmic delivery of gelonin immunotoxin. Mol. Cancer Ther. 12, 1774–82.

409

(18) Perler, F. B. (2002) InBase: the Intein Database. Nucleic Acids Res. 30, 383–384.

410

(19) Paulus, H. (2001) Inteins as enzymes. Bioorg. Chem. 29, 119–129.

411

(20) Martin, D. D., Xu, M. Q., and Evans, T. C. (2001) Characterization of a naturally occurring

412

trans-splicing intein from Synechocystis sp. PCC6803. Biochemistry 40, 1393–1402.

413

(21) Iwai, H., Züger, S., Jin, J., and Tam, P. H. (2006) Highly efficient protein trans-splicing

414

by a naturally split DnaE intein from Nostoc punctiforme. FEBS Lett. 580, 1853–1858.

415

(22) Wu, H., Hu, Z., and Liu, X.-Q. (1998) Protein trans-splicing by a split intein encoded in a

416

split DnaE gene of Synechocystis sp. PCC6803. Proc. Natl. Acad. Sci. 95, 9226–9231.

417

(23) Thiel, I. V., Volkmann, G., Pietrokovski, S., and Mootz, H. D. (2014) An atypical naturally

418

split intein engineered for highly efficient protein labeling. Angew. Chemie - Int. Ed. 53, 1306–

419

1310.

420

(24) Shibuya, Y., Haga, N., Asano, R., Nakazawa, H., Hattori, T., Takeda, D., Sugiyama, A.,

421

Kurotani, R., Kumagai, I., Umetsu, M., and Makabe, K. (2016) Generation of camelid VHH

422

bispecific constructs via in-cell intein-mediated protein trans-splicing. Protein Eng. Des. Sel. - 13 ACS Paragon Plus Environment


423

1–7.

424

(25) Braner, M., Kollmannsperger, A., Wieneke, R., and Tampé, R. (2016) “Traceless” tracing

425

of proteins – high-affinity trans-splicing directed by a minimal interaction pair. Chem. Sci.

426

2646–2652.

427

(26) Han, L., Chen, J., Ding, K., Zong, H., Xie, Y., Jiang, H., Zhang, B., Lu, H., Yin, W., Gilly,

428

J., and Zhu, J. (2017) Efficient generation of bispecific IgG antibodies by split intein mediated

429

protein trans-splicing system. Sci. Rep. 7, 8360.

430

(27) Appleby-Tagoe, J. H., Thiel, I. V., Wang, Y., Wang, Y., Mootz, H. D., and Liu, X. Q.

431

(2011) Highly efficient and more general cis- and trans-splicing inteins through sequential

432

directed evolution. J. Biol. Chem. 286, 34440–34447.

433

(28) Lewis Phillips, G. D., Li, G., Dugger, D. L., Crocker, L. M., Parsons, K. L., Mai, E.,

434

Blättler, W. A., Lambert, J. M., Chari, R. V. J., Lutz, R. J., Wong, W. L. T., Jacobson, F. S.,

435

Koeppen, H., Schwall, R. H., Kenkare-Mitra, S. R., Spencer, S. D., and Sliwkowski, M. X.

436

(2008) Targeting HER2-positive breast cancer with trastuzumab-DM1, an antibody-cytotoxic

437

drug conjugate. Cancer Res. 68, 9280–9290.

438

(29) Roovers, R. C., Vosjan, M. J. W. D., Laeremans, T., El Khoulati, R., De Bruin, R. C. G.,

439

Ferguson, K. M., Verkleij, A. J., Van Dongen, G. a M. S., and Van Bergen En Henegouwen, P.

440

M. P. (2011) A biparatopic anti-EGFR nanobody efficiently inhibits solid tumour growth. Int.

441

J. Cancer 129, 2013–2024.

442

(30) Schmitz, K. R., Bagchi, A., Roovers, R. C., Van Bergen En Henegouwen, P. M. P., and

443

Ferguson, K. M. (2013) Structural evaluation of EGFR inhibition mechanisms for

444

nanobodies/VHH domains. Structure 21, 1214–1224.

445

(31) Heukers, R., Vermeulen, J. F., Fereidouni, F., Bader, A. N., Voortman, J., Roovers, R. C.,

446

Gerritsen, H. C., and van Bergen En Henegouwen, P. M. P. (2013) Endocytosis of EGFR

447

requires its kinase activity and N-terminal transmembrane dimerization motif. J. Cell Sci. 126,

448

4900–12.

449

(32) Pastrana, D. V., and FitzGerald, D. J. (2006) A nonradioactive, cell-free method for

450

measuring protein synthesis inhibition by Pseudomonas exotoxin. Anal. Biochem. 353, 266–

451

271.

452

(33) Sellmann, C., Doerner, A., Knuehl, C., Rasche, N., Sood, V., Krah, S., Rhiel, L.,

453

Messemer, A., Wesolowski, J., Schuette, M., Becker, S., Toleikis, L., Kolmar, H., and Hock,

454

B. (2016) Balancing Selectivity and Efficacy of Bispecific EGFR x c-MET Antibodies and

455

Antibody-Drug Conjugates. J. Biol. Chem. jbc.M116.753491.

456

(34) Schmohl, L., and Schwarzer, D. (2014) Sortase-mediated ligations for the site-specific - 14 ACS Paragon Plus Environment

Page 20 of 23

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


457

modification of proteins. Curr. Opin. Chem. Biol. 22, 122–128.

458

(35) Strop, P. (2014) Versatility of microbial transglutaminase. Bioconjug. Chem. 25, 855–862.

459

(36) Veggiani, G., Zakeri, B., and Howarth, M. (2014) Superglue from bacteria: Unbreakable

460

bridges for protein nanotechnology. Trends Biotechnol. 32, 506–512.

461

(37) Guimaraes, C. P., Witte, M. D., Theile, C. S., Bozkurt, G., Kundrat, L., Blom, A. E. M.,

462

and Ploegh, H. L. (2013) Site-specific C-terminal and internal loop labeling of proteins using

463

sortase-mediated reactions. Nat. Protoc. 8, 1787–1799.

464

(38) Lorand, L., and Graham, R. M. (2003) Transglutaminases: Crosslinking enzymes with

465

pleiotropic functions. Nat. Rev. Mol. Cell Biol.

466

(39) Pasternack, R., Dorsch, S., Otterbach, J. T., Robenek, I. R., Wolf, S., and Fuchsbauer, H.

467

L. (1998) Bacterial pro-transglutaminase from Streptoverticillium mobaraense - Purification,

468

characterisation and sequence of the zymogen. Eur. J. Biochem. 257, 570–576.

469

(40) Dennler, P., Chiotellis, A., Fischer, E., Brégeon, D., Belmant, C., Gauthier, L., Lhospice,

470

F., Romagne, F., and Schibli, R. (2014) Transglutaminase-Based Chemo-Enzymatic

471

Conjugation Approach Yields Homogeneous Antibody–Drug Conjugates. Bioconjug. Chem.

472

25, 569–578.

473

(41) Gundersen, M. T., Keillor, J. W., and Pelletier, J. N. (2014) Microbial transglutaminase

474

displays broad acyl-acceptor substrate specificity. Appl. Microbiol. Biotechnol. 98, 219–230.

475

(42) Siegmund, V., Schmelz, S., Dickgiesser, S., Beck, J., Ebenig, A., Fittler, H., Frauendorf,

476

H., Piater, B., Betz, U. A. K., Avrutina, O., Scrima, A., Fuchsbauer, H. L., and Kolmar, H.

477

(2015) Locked by Design: A Conformationally Constrained Transglutaminase Tag Enables

478

Efficient Site-Specific Conjugation. Angew. Chemie - Int. Ed. 54, 13420–13424.

479

(43) Jeger, S., Zimmermann, K., Blanc, A., Grünberg, J., Honer, M., Hunziker, P., Struthers,

480

H., and Schibli, R. (2010) Site-specific and stoichiometric modification of antibodies by

481

bacterial transglutaminase. Angew. Chemie - Int. Ed. 49, 9995–9997.

482

(44) Mero, A., Spolaore, B., Veronese, F. M., and Fontana, A. (2009) Transglutaminase-

483

mediated PEGylation of proteins: Direct identification of the sites of protein modification by

484

mass spectrometry using a novel monodisperse PEG. Bioconjug. Chem. 20, 384–389.

485

(45) Zakeri, B., and Howarth, M. (2010) Spontaneous intermolecular amide bond formation

486

between side chains for irreversible peptide targeting. J. Am. Chem. Soc. 132, 4526–4527.

487

(46) Ludwig, C., Schwarzer, D., and Mootz, H. D. (2008) Interaction studies and alanine

488

scanning analysis of a semi-synthetic split intein reveal thiazoline ring formation from an

489

intermediate of the protein splicing reaction. J. Biol. Chem. 283, 25264–25272.

490

(47) Zettler, J., Schütz, V., and Mootz, H. D. (2009) The naturally split Npu DnaE intein - 15 ACS Paragon Plus Environment


491

exhibits an extraordinarily high rate in the protein trans-splicing reaction. FEBS Lett. 583, 909–

492

914.

493

(48) Stevens, A. J., Brown, Z. Z., Shah, N. H., Sekar, G., Cowburn, D., and Muir, T. W. (2016)

494

Design of a Split Intein with Exceptional Protein Splicing Activity. J. Am. Chem. Soc. 138,

495

2162–2165.

496

(49) Fukuhara, T., Kim, J., Hokaiwado, S., Nawa, M., Okamoto, H., Kogiso, T., Watabe, T.,

497

and Hattori, N. (2017) A novel immunotoxin reveals a new role for CD321 in endothelial cells.

498

PLoS One 1–12.

499

(50) Kawa, S., Onda, M., Ho, M., Kreitman, R. J., Bera, T. K., and Pastan, I. (2011) The

500

improvement of an anti-CD22 immunotoxin: Conversion to single-chain and disulfide

501

stabilized form and affinity maturation by alanine scan. MAbs 3, 479–486.

502

(51) Weldon, J. E., Xiang, L., Zhang, J., Beers, R., Walker, D. A., Onda, M., Hassan, R., and

503

Pastan, I. (2013) A Recombinant Immunotoxin against the Tumor-Associated Antigen

504

Mesothelin Reengineered for High Activity, Low Off-Target Toxicity, and Reduced

505

Antigenicity. Mol. Cancer Ther. 12, 48–57.

506

(52) Klausz, K., Kellner, C., Derer, S., Valerius, T., Staudinger, M., Burger, R., Gramatzki, M.,

507

and Peipp, M. (2015) The novel multispecies Fc-specific Pseudomonas exotoxin A fusion

508

protein α-Fc-ETA’ enables screening of antibodies for immunotoxin development. J. Immunol.

509

Methods 418, 75–83.

510

(53) Roopenian, D. C., and Akilesh, S. (2007) FcRn: The neonatal Fc receptor comes of age.

511

Nat. Rev. Immunol. 7, 715–725.

512

(54) Zheng, K., Bantog, C., and Bayer, R. (2011) The impact of glycosylation on monoclonal

513

antibody conformation and stability. MAbs 3.

514

(55) Moody, P. R., Sayers, E. J., Magnusson, J. P., Alexander, C., Borri, P., Watson, P., and

515

Jones, A. T. (2015) Receptor Crosslinking: A General Method to Trigger Internalization and

516

Lysosomal Targeting of Therapeutic Receptor:Ligand Complexes. Mol. Ther. 23, 1888–1898.

517

(56) Hollevoet, K., Mason-Osann, E., Liu, X. -f., Imhof-Jung, S., Niederfellner, G., and Pastan,

518

I. (2014) In Vitro and In Vivo Activity of the Low-Immunogenic Antimesothelin Immunotoxin

519

RG7787 in Pancreatic Cancer. Mol. Cancer Ther. 13, 2040–2049.

520

(57) Jackson, M. E., Simpson, J. C., Girod, A., Pepperkok, R., Roberts, L. M., and Lord, J. M.

521

(1999) The KDEL retrieval system is exploited by Pseudomonas exotoxin A, but not by Shiga-

522

like toxin-1, during retrograde transport from the Golgi complex to the endoplasmic reticulum.

523

J. Cell Sci. 112 ( Pt 4, 467–75.

524

(58) Puthenveetil, S., Musto, S., Loganzo, F., Tumey, L. N., O’Donnell, C. J., and Graziani, E. - 16 ACS Paragon Plus Environment

Page 22 of 23

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


525

(2016) Development of Solid-Phase Site-Specific Conjugation and Its Application toward

526

Generation of Dual Labeled Antibody and Fab Drug Conjugates. Bioconjug. Chem. 27, 1030–

527

1039.

528 529 530

For Table of Contents Only

531 532

- 17 ACS Paragon Plus Environment

2 Immunotoxins by Protein Ligation

Recommend Documents