Nanofitin as a New Molecular-Imaging Agent for the Diagnosis of

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Article

Bioconjugate Chemistry 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.


Nanofitin as a new molecular imaging agent for diagnosis of EGFR overexpressing tumors Bioconjugate Chemistry 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.


Marine Goux, Guillaume Becker, Harmony Gorré, Sylvestre Dammicco, Ariane Desselle, Dominique Egrise, Natacha Leroi, François Lallemand, Mohamed Ali Bahri, Gilles Doumont, Alain Plenevaux, Mathieu Cinier, and André Luxen



Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.7b00374 • Publication Date (Web): 21 Aug 2017



Downloaded from http://pubs.acs.org on August 22, 2017



Just Accepted



“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical



Society provides “Just Accepted” as a free service to the research community to expedite the



dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts



appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been



fully peer reviewed, but should not be considered the official version of record. They are accessible to all



readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered



to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published



in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just



Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor



changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers



and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.


Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Bioconjugate Chemistry

ACS Paragon Plus Environment


% Cell viability

150 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Page 2 of 37

aEGFR_NF1-Pe38-KDEL aEGFR_NF2-Pe38-KDEL aEGFR_B10-Pe38-KDEL IrrNF-Pe38-KDEL

100

50

0 -13

-11

-9

-7

ACS Paragon Plus Environment log[Nanofitin-Pe38-KDEL], M

Page A.3 of 37

B.

IrrNF-FITC

B10-FITC + Cetuximab

B10-FITC 150

150

A431 cells

150

Events

0

100

101

102

0

103

100

101

102

103

90

90

H520 cells

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

IrrNF labeled anti-EGFR B10 Nanofitin Bioconjugate Chemistry with Alexa Fluor labeled with Alexa Fluor

0

100

101

0 Plus0Environment ACS 102 Paragon 103 10 101 102 103

FL-1

0

100

101

102

103

A.

B. Bioconjugate Chemistry

0.10

Page 4 of 37

0.10

nm

nm

1 0.05 2 0.05 3 4 0.00 0.00 5 150 300 450 600 150 300 450 600 6 Time (sec) Time (sec) 7 -0.05 -0.05 8 ACS Paragon Plus Environment 9 KD (M) kon (M-1.s-1) kdis (s-1) R2 KD (M) kon (M-1.s-1) kdis (s-1) R2 8.30x10-8 1.41x106 1.17x10-1 0.9801 102.76x10-8 1.90x106 5.24x10-2 0.9867

Page 5 of 37

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



A. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 C. 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

B.Chemistry Bioconjugate

Page 6 of 37

40% ID/g Liver

Intestine

Kidney

Bladder


Dorsal

Median

Ventral

0% ID/g

Page 7 of 37

Bioconjugate Chemistry DAPI

H520 tumor

A431 tumor

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Anti-EGFR


Merge


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19


Page 8 of 37

*

Page 9 of 37 Bioconjugate Chemistry

ns

3

2

1

10 B n N an of iti

C et ux i

m ab

0

C on tr ol

%ID/g (g-1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19



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

Table of contents only 82x20mm (300 x 300 DPI)


Page 10 of 37

Page 11 of 37

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


Nanofitin as a new molecular imaging agent for diagnosis of EGFR overexpressing tumors a

a,b

e

a

e

Marine Goux , Guillaume Becker , Harmony Gorré , Sylvestre Dammicco , Ariane Desselle , Dominique Egrisec,d, Natacha Leroif, François Lallemanda,b, Mohamed Ali Bahria,b, Gilles Doumontc, a,b ,

Alain Plénevaux

e

a†

Mathieu Cinier and André Luxen

a

Cyclotron Research Centre, University of Liège, Allée du 6 Août, 4000 Liège, Belgium

b

GIGA-CRC in vivo imaging, University of Liège, Allée du 6 Août, 4000 Liège, Belgium

c

Centre for Microscopy and Molecular Imaging, Université Libre de Bruxelles and Université de Mons,

8 Rue Adrienne Bolland, 6041 Gosselies, Belgium d

Service de Médecine Nucléaire, Hôpital Erasme, Université Libre de Bruxelles, Brussels, Belgium.

e

Affilogic SAS, 21 rue La Noue Bras de Fer, 44200 Nantes, France

f

GIGA-Cancer, Laboratory of Tumor and Development Biology, University of Liège, Avenue de

l’Hopital, 4000 Liège, Belgium

†

corresponding author’s information: André Luxen, [email protected], Cyclotron Research Centre,

University of Liège, Allée du 6 Août, 4000 Liège, Belgium.

1 ACS Paragon Plus Environment


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

Page 12 of 37

Abstract Epidermal Growth Factor Receptor (EGFR) is involved in cell growth and proliferation, and is overexpressed in malignant tissues. Although anti-EGFR based immunotherapy became a standard of care for patients with EGFR-positive tumors, this strategy of addressing cancer tumors by targeting EGFR with monoclonal antibodies is less developed for patient diagnostic and monitoring. Indeed, antibodies exhibit a slow blood clearance which is detrimental for PET imaging. New molecular probes are proposed to overcome such limitations for patient monitoring, making use of low molecular weight protein scaffolds as alternatives to antibodies, such as Nanofitins with better pharmacokinetic profiles. Anti-EGFR Nanofitin B10 was reformatted by genetic engineering to exhibit a unique cysteine moiety at its C-terminus which allows the development of a fast and site-specific radiolabeling procedure with 18

F-4-fluorobenzamido-N-ethylamino-maleimide (18F-FBEM). The in vivo tumor targeting and imaging

profile of the anti-EGFR Cys-B10 Nanofitin was investigated in a double-tumor xenograft model by static small-animal PET at 2 h after tail vein injection of the radiolabeled Nanofitin 18F-FBEM-Cys-B10. The image showed that the EGFR-positive tumor (A431) is clearly delineated in comparison to the EGFR-negative tumor (H520) with a significant tumor-to-background contrast.

18

F-FBEM-Cys-B10,

demonstrated a significantly higher retention in A431 tumors than in H520 tumors at 2.5 h p.i. with an uptake ratio of A431-to-H520 of 2.53 ± 0.18 and a tumor-to-blood ratio of 4.55 ± 0.63. This study provides the first report of Nanofitin scaffold used as a targeted PET radiotracer for in vivo imaging of EGFR-positive tumor, with the anti-EGFR B10 Nanofitin as proof-of-concept. The fast generation of specific Nanofitins via a fully in vitro selection process, together with the excellent imaging features of the Nanofitin scaffold, could facilitate the development of valuable PET-based companion diagnostics. Keywords Nanofitin, scaffold, PET, molecular imaging, EGFR, companion diagnostic, FBEM, protein engineering, radiolabeling


Page 13 of 37

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


Introduction Treating cancer diseases has been found to be very challenging, merely due to their complex, heterogeneous and sometimes dispersed nature. Over the last decade, progress in molecular biology and genomics has led to the discovery of novel biomarkers that provide components to better understand the molecular and cellular mechanisms which drive tumor initiation, maintenance and progression

1,2

. These discoveries have fueled the development of more effective targeted

therapeutics with monoclonal antibodies; shifting away the standard cancer patient's care from an empirical treatment strategy based on the clinical-pathological profile. Thus creating a personalized medicine that is adapted to the biomarker profile of each tumor 3. In the era of personalized medicine, the ability to detect and visualize patient biomarkers with companion diagnostic tools has become a critical asset in drug development programs to facilitate 3

faster, safer, and more efficient clinical trials . Companion diagnostic assays can be used to monitor and improve treatment responses and patient outcomes by identifying and predicting patient subpopulations that are most likely to respond to a given treatment. In this regard, non-invasive imaging strategies have been developed to visualize tumor biomarkers by positron emission tomography (PET) 4

using targeted radioactively-labeled probes . Radionuclide-based imaging probes have the advantage of enabling the real time whole body assessment, hence providing information on tumor antigen 5

expression not only in primary tumors but also in distant metastases not amenable to biopsy . Radioimaging thus overcomes the inherent limitations of single-site and single time-point sampling of companion diagnoses that rely on histological analysis of biopsies or surgically resected tissues. These antiquated methods fail to provide information about heterogeneity of expression or changes in 6

expression that occur over time . A straightforward strategy for generating targeted PET-radiotracers has relied on the radiolabeling of 7

therapeutic monoclonal antibodies . Examples described in the literature include 89

Zr-Bevacizumab 9,

89

Zr-Panitunumab

10

,

89

Zr-Cetuximab as well as

64

89

8

Zr-Trastuzumab ,

Cu-Cetuximab

11,12

. Immuno-

PET radiotracers have shown some promises, but their sensitivity remains limited due to their full IgG format (~150 kDa) and related biophysical characteristics (extended residence in the bloodstream, slow tumor penetration and slow blood clearance) to low tissue contrast

15

13–15

. This results in high tumor accumulation, yields

and requires late imaging time point (at least 2 days after administration)

16

.

Interestingly, ScFv antibody fragments (~25 kDa) show faster clearance but only a moderate tumor



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

Page 14 of 37

14

accumulation

. Studies focused on analyzing the relationship between the size of molecules and

their tumor uptake

13,17

, showing a U-shaped relationship between these two parameters with a

minimal accumulation of molecules of the size of a ScFv. Smaller molecules such as nonimmunoglobulin alternative scaffolds (4 to 20 kDa) might provide the optimal balance between a rapid clearance from blood and non-target tissue, vascular extravasation and tissue penetration, while being amenable to high enough affinity to provide tumor targeting for imaging applications

18

. The growing

number of imaging ligands recently developed, including Affibodies 19–21, Fibronectins 22,23 and Darpins 24,25

, have illustrated the suitability of non-IgG scaffolds for in vivo tumor visualization. With their very

short biological half-life (in minutes)

18

, alternative scaffolds are perfectly well suited for fast imaging

protocols in combination with rapid decaying radioisotopes such as

68

Ga (68 min) and

18

F (110 min).

Additionally, fast imaging protocols with short physical half-life radioisotopes facilitate handling of radioactive waste that can be contained within the hospital facility with the current common protocols already in place for 18F-fluorodeoxyglucose (18F-FDG). Nanofitins are cysteine-free protein scaffolds derived from the hyperstable DNA-binding protein Sac7d (7 kDa, 66 amino acids) of Sulfolobus acidocaldarius

26

. High-affinity Nanofitins have been easily

engineered by ribosome-display over a wide range of targets (cell surface proteins GFP

29

, IgG

30

, cytokines

31

26

, enzymes

27,28

,

, etc) by the full randomization of 10 to 14 amino acid residues localized in

the DNA-binding site of Sac7d (Fig. 1). This process allows to fully redirect the initial DNA specificity of Nanofitins to the binding of a target of interest

26–32

. Nanofitins have kept their extreme stability from

Sac7d origin and have been proven robust enough to remain active after one cycle of steam sterilization 29. Taking advantage of their robustness and small size, Nanofitins are currently developed for challenging non-systemic administration; orally deliverable anti-TNF alpha Nanofitins have entered a pre-clinical development program in inflammatory bowel disease

31

. Here, we introduce the use of

radiolabeled Nanofitins as a new class of molecular imaging probes with a suitable pharmacokinetic profile for the non-invasive diagnosis and monitoring of high-EGFR-expressing solid tumors. The epidermal growth factor receptor 1 (EGFR) is a transmembrane tyrosine-kinase receptor frequently found overexpressed or mutated in a variety of human tumors Nanofitin Cys-B10 was site-specifically labeled with group

18

33,34

. In this study, the anti-EGFR

F by site-specific conjugation with the prosthetic

18

F-4-fluorobenzamido-N-ethylamino-maleimide (18F-FBEM), using a unique cysteine residue

specifically introduced in C-terminus (Fig. 1). The resulting probe,

18

F-FBEM-Cys-B10, was then


Page 15 of 37

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


injected in a double-bearing tumor model to evaluate the biodistribution and the ability of the radiolabeled protein to specifically target in vivo the over-EGFR-expressing A431 tumor.

Figure 1: Scheme of

18

F-FBEM-Cys-B10 radiosynthesis. Positions randomized in the Nanofitin

libraries are labelled in red



Results Anti-EGFR B10 Nanofitin. Anti-EGFR Nanofitins were identified upon several rounds of ribosome display

29

using the recombinant extracellular domain of human EGFR fused to a Fc fragment

(Creative Biomart) as a bait. Fusion of anti-EGFR Nanofitins to Pe38-KDEL toxins allowed the evaluation of their internalization potential by monitoring IC50 of the constructs on A431 cells, in comparison to a negative control involving the anti-egg white lysozyme H4 Nanofitin

27,28,35–37

and

referred hereafter as IrrNF (Fig. 2).

% Cell viability

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

Page 16 of 37

Figure 2: Measurement of A431 cells viability after incubation with Nanofitins fused to the Pe38 toxin. A431 cells were incubated in presence of a concentration range of Nanofitins-Pe38 fusion (10-8 to 1013

M) and viability was measured after 7 days using a XTT assay. Data points represent the mean and

standard deviation of triplicate results. αEGFR_B10_Pe38-KDEL: Anti-EGFR B10 Nanofitin; αEGFR_NF1- and NF2-Pe38-KDEL: Other anti-EGFR Nanofitins; IrrNF-Pe38_KDEL: irrelevant Nanofitin.

Pe38-KDEL toxin is a derivative of pseudomonas exotoxin A which is lacking the cell entry domain required for its intracellular mechanism of action. Only Nanofitins triggering the internalization of the toxin, i.e. with the Nanofitin moiety replacing the toxin cell entry domain, allow to restore the internalization of the chimeric construct and its subsequent cytotoxic activity. As with the negative


Page 17 of 37

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


control, no effect on the cell viability could be measured over concentrations of the B10-PE38-KDEL construct ranging from 10-8 to 10-13 M. This non-internalizing profile appeared not to be shared with the two other anti-EGFR Nanofitins NF1 and NF2. Toxicity of the latter two Nanofitin-Pe38-KDEL constructs is observed to a different extent with an IC50 of 1.16x10-9 and 1.94x10-11 M, respectively, which we attributed to a different ability to trigger internalization.

Beside its non-internalizing behavior, the specificity and efficiency of B10 at localizing at the membrane of EGFR overexpressing A431 cell lines was evaluated by cell imaging and flow cytometry (Fig. 3).

Figure 3: In vitro specificity experiments of the Nanofitin B10 (A) Cell surface labeling of A431 cells. Labeling of the A431 cells with a 2 µM solution of an irrelevant Nanofitin conjugated to Alexa Fluor 488 and a 1 µM solution of the anti-EGFR Nanofitin B10 conjugated to Alexa Fluor 488. Images were captured by time-lapse microscopy (7 images/sec, magnification x20) for 18 sec and the images given



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

Page 18 of 37

were obtained after 6 sec of exposure. (B) Binding and specificity analysis of the Nanofitins B10-FITC and IrrNF-FITC (10 µM) binding to H520 (bottom) or A431 cells (top), and cross-blocking experiment on A431 cells with B10-FITC pre-incubated with Cetuximab (1.1 µM). The EGFR binding of B10-FITC is represented by solid line, of IrrNF-FITC by dotted line and control cells without staining by grayshaded profiles. For the mean fluorescent intensity, please see Supplementary information (Fig. S1).

Time-lapse microscopy on A431 cells incubated with either Alexa Fluor 488 labeled anti-EGFR B10 (1 µM) or anti-egg white lysozyme H4 Nanofitin (2 µM, negative control) revealed a fast accumulation (visible after few seconds) of fluorescence on cells membrane for B10, while no targeting was observed with the irrelevant Nanofitin (Fig. 3). The effective localization of B10 at the membrane of the cells was further confirmed by fluorescence recovery after photobleaching A431 cell membranes (Fig. S2), which resulted in instantaneous recovery of fluorescence, accounting for localization of the labeling on the membrane. The analysis of the specificity of B10 Nanofitin was complemented by a set of flow cytometry experiments using A431 and H520 cells as EGFR-positive and EGFR-negative cell lines, respectively. While no labeling could be observed with the irrelevant Nanofitin neither on A431 or H520 cell lines, B10 was able to label only the A431 EGFR-positive cell line in an EGFR-dependent manner as demonstrated by the full inhibition of its labeling efficiency on cells pre-incubated with the anti-EGFR therapeutic antibody Cetuximab. Additionally, B10 was demonstrated to bind human and mouse forms of EGFR with an affinity of 27 and 83 nM, respectively, as measured by biolayer interferometry. While the kinetic profiles appear similar between the two forms of EGFR, displaying a fast on-rate, the binding of B10 to the human EGFR is marked with a slightly slower off-rate; highlightening the difference in overall equilibrium dissociation constants (Kd) (Fig. 4).


Page 19 of 37

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: Determination of the binding characteristics of the anti-EGFR Nanofitin by interferometry on human EGFR (A) and on mouse EGFR (B). The experiment was run on Octet RED96 using the antiEGFR Nanofitin at a range of concentrations of 1000, 500, 250, 125, 62.5, 31.25, 15.625 nM. Fittings (1:1 model) are represented as solid grey line.

Chemistry and radiochemistry. To ensure a controlled and site-specific labeling of the B10 Nanofitin with FBEM derivatives, the probe was genetically engineered to display a unique cysteine residue at its C-terminus extremity and named Cys-B10 (Fig. S3). The coupling of FBEM to the Cys-B10 Nanofitin was first investigated with non-radioactive cysteine in presence of TCEP/HCl,

19

F-FBEM compound. After reduction of the

19

F-FBEM was added in excess and the reaction was allowed to

proceed for up to 90 min. Reaction media after 45 and 90 min of incubation were analyzed on a C18 reverse-phase column (Fig. S4). In both cases, only a single peak was obtained with a retention time of 23 min, as observed for the Nanofitin injected alone. Further investigations of the peak at 23 min by mass spectrometry confirmed the completion of the reaction at 45 min. A major mass peak (around 90 %) with a calculated molecular weight of 9429 Da corresponding to the expected product was observed for both reaction times (Fig. S5). The Cys-B10 Nanofitin was then conjugated to

18

F-FBEM

(Fig. 1) with a decay-corrected radiolabeling yield of 45 ± 14 % (n = 6). The automatic radiosynthesis of the

18

F-FBEM prosthetic group, purification and coupling to the Nanofitin were completed in

approximately 3 h. Stability of the radiolabeled Nanofitin. Before evaluating Nanofitin biodistribution, stability of the radioconjugate at pH 7.4 and in contact with plasmatic proteins was investigated by incubating the 18FFBEM radiolabeled Nanofitin respectively in phosphate-buffered saline (PBS) and fetal bovine serum (FBS) at 37 °C. After 2 h of incubation in FBS and 2.5 h in PBS, the mixtures were injected on radioHPLC by C18 reverse-phase column. While

18

F-FBEM is expected to elute at 20 min (Fig. 5A),

chromatograms of the incubation mixtures (Fig. 5B and C) revealed only a single peak with a retention time of 23 min corresponding to the Nanofitin radiolabeled with

18

F-FBEM. A slight shoulder was

observed at 18 min, which was not attributed to either impurity in the Nanofitin

18

F-FBEM sample (Fig.

S6) nor degradation of the Nanofitin conjugate in FBS. These results confirmed that the prosthetic group is covalently conjugated to the Nanofitin, and suggested that

18

F-FBEM-Cys-B10 is stable in



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

Page 20 of 37

complex media over a time compatible with both the serum half-life of Nanofitins and the decay halflife of 18F.

Figure 5: In vitro stability assay of the radiolabeled Nanofitin at 37 °C. (A) Co-injection in radio-HPLC of

18

F-FBEM (retention time of 20 min) and

18

F-FBEM-Cys-B10 (retention time of 23 min) in PBS.

Injection in radio-HPLC of the radiolabeled Nanofitin after 2 h of incubation in FBS (B) and 2.5 h in PBS (C). Spectrum of 18F-FBEM-Cys-B10 in PBS is provided in Supplementary information (Fig. S6).

Biodistribution study and PET imaging in BALB/c mice. The in vivo biodistribution of Cys-B10 was investigated in BALB/c mice at 2.5 h after injection.

18

F-FBEM-

18

F-FBEM-Cys-B10 displayed an

uptake of radioactivity in kidney (6.56 ± 2.40 %ID/g), liver (4.92 ± 2.06 %ID/g) and intestine (24.39 ± 5.85 %ID/g) (Fig. 6A). Dynamic scanning of

18

F-FBEM-Cys-B10 has shown an uptake in liver and

kidneys with peaks at about 13 min and 32 min, respectively, which rapidly decreases after these times (Fig. 6B). The decrease of radioactivity in the kidney was followed by a progressive accumulation in the bladder, only visible at the later time points (Fig. 6C). Taken together, these data suggest a renal excretion of

18

F-FBEM-Cys-B10, which is consistent for a radioconjugate having a

molecular weight below the cut-off for glomerular filtration. Moreover, based on imaging and biodistribution studies, the 18F-FBEM-Cys-B10 exhibits low accumulations in blood (0.99 ± 0.20 %ID/g) and most normal organs such as heart (0.48 ± 0.14 %ID/g), lung (0.84 ± 0.59 %ID/g) and spleen (0.77 ± 0.22 %ID/g).


Page 21 of 37

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 6: Biodistribution of the radiolabeled anti-EGFR Nanofitin in BALB/c mice. (A) Uptake and distribution of

18

F-FBEM-Cys-B10 at 2h30 p.i.. Data are expressed in percentage of injected dose per

gram of organ [%ID/g] ± SD after intravenous injection of the probe in balb/c (n = 4). (B) Data are expressed in percentage of injected dose per gram of organ [%ID/g] and obtained from co-registered coronal sections of MRI and two-hour duration PET after injection of

18

F-FBEM-Cys-B10 (2 - 6

MBq/100 µL) under isoflurane anesthesia (n = 4). Arrows show maximum of uptake in the liver and left kidney. (C) Co-registered coronal sections of MRI and two-hour duration PET after injection under isoflurane anesthesia of

18

F-FBEM-Cys-B10 (5.6 MBq/100 µL).

Biodistribution study in BALB/c nude xenograft model. For the in vivo biology assessment, we developed a double bearing-tumor animal model using two human carcinoma cell lines with opposite EGFR expression levels. We injected in the right flank of nude mice overexpressing EGFR-positive A431 cell line

38,39

and in the left flank EGFR-negative H520 cell line

40,41

. Immunohistochemical

staining on fixed tumor sections confirmed a high-expression levels of EGFR on cell surface of A431 tumor while no EGFR was detected on H520 tumor cell surface. Results indicated that the model was suitable for comparing uptake of the anti-EGFR Nanofitin in A431 and H520 tumors (Fig. 7).



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

Page 22 of 37

Figure 7: Immunohistochemistry of A431 and H520 tumors grown on xenograft model. EGFR expression level was evaluated by staining tumors slices with a rabbit anti-human EGFR antibody and a secondary goat anti-rabbit antibody conjugated to Alexa Fluor 488. Nuclei were counter stained with DAPI (x10).

18

F-FDG PET were firstly performed on xenograft model as control of metabolic activity of the tumors

prior the specific targeting assay with the anti-EGFR Nanofitin. The transverse image showed retention of

18

F-FDG in the two tumors indicating high levels of metabolism for both (Fig. 8A).

Figure 8: Targeting of the EGFR-positive tumor A431 by the radiolabeled anti-EGFR Nanofitin (A) Coregistered transversal sections of PET and CT 1 h after injection of

18

F-FDG (9 MBq) in xenograft

model under isoflurane anesthesia (blood glucose level: 73 mg/dL, weight: 29 g). He: heart. (B) Co-


Page 23 of 37

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


registered transversal sections of PET and MRI 2 h after injection in xenograft model under isoflurane anesthesia of

18

F-FBEM-Cys-B10 (19 MBq/100µL).

Biodistribution studies of

18

F-FBEM-Cys-B10 in xenograft model (Table 1A and Fig. S7) displayed an

uptake of radioactivity in the kidney and liver of 1.55 ± 0.57 and 1.13 ± 0.52 %ID/g 2.5 h p.i., respectively (Table 1A). Overall, liver-to-blood and kidney-to-blood uptake ratios at 2.5 h after injection appeared of similar order in the xenograft model and in BALB/c mice (3.90 ± 1.33 % and 4.82 ± 0.84 % versus 4.80 ± 1.11 % and 6.74 ± 2.43 %, respectively), suggesting a similar excretion profile (Table 1B). B.

A. Tissues

Xenograft model

Uptake ratio

Xenograft model

Balb/c

Blood

0.32 ± 0.07

liver-to-blood

3.90 ± 1.33

4.80 ± 1.11

Brain

0.02 ± 0.01

kidney-to-blood

4.82 ± 0.84

6.74 ± 2.43

Bone

0.20 ± 0.01

A431-to-kidney

0.98 ± 0.29

Liver

1.13 ± 0.52

A431-to-liver

1.43 ± 0.52

Kidney

1.55 ± 0.57

A431-to-H520

2.53 ± 0.18

Heart

0.17 ± 0.03

A431-to-lung

2.53 ± 0.89

Spleen

0.27 ± 0.08

A431-to-blood

4.55 ± 0.63

Skin

0.28 ± 0.12

A431-to-heart

8.56 ± 1.34

Muscle

0.12 ± 0.03

Lung

0.63 ± 0.31

Tumor A431

1.42 ± 0.18

Tumor H520

0.56 ± 0.10

Table 1: Biodistribution of

18

F-FBEM-Cys-B10 at 2.5 h p.i. (A) Biodistribution in xenograft model. Data

are expressed in percentage of injected dose per gram of organ [%ID/g] ± SD after intravenous injection of the probe. (B) Tumor-to-non-target tissue count density ratios were calculated from the corresponding %ID/g values at 2.5 h after injection of the probe in BALB/c and xenograft model. Specificity of tumor targeting. The in vivo tumor targeting and imaging profile of the anti-EGFR CysB10 Nanofitin were investigated studies by static small-animal PET at 2 h after tail vein injection of 18FFBEM-Cys-B10 into two separate studies, using i) the double-tumor xenograft model (Fig. 8B) and ii) blocking amounts of unlabeled tracer to saturate receptors (Fig. 9)



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

Page 24 of 37

In the double-tumor xenograft model, the image showed that the EGFR-positive tumor A431 is clearly delineated in comparison to the EGFR-negative tumor H520 with a significant tumor-to-background contrast (Fig. 8B). This specific tumor targeting was confirmed by biodistribution studies;

18

F-FBEM-

Cys-B10 Nanofitin (13 - 25 MBq) demonstrated a significantly higher retention in A431 tumors (1.42 ± 0.18 %ID/g) than in H520 tumors (0.56 ± 0.10 %ID/g) at 2.5 h p.i. (p = 0.028, Mann-Whitney test) with an uptake ratio of A431-to-H520 of 2.53 ± 0.18 (Table 1B). To access the specificity of

18

F-FBEM-Cys-B10 tumor uptake, two blocking studies were performed

with either the anti-EGFR Cetuximab antibody or the unlabeled B10 Nanofitin. Cetuximab and unlabeled Nanofitin are intended to saturate overexpressed EGFR sites at the tumor. Cetuximab was injected 48 h before

18

F-FBEM-Cys-B10 and

18

F-FBEM-Cys-B10 was co-injected with a large excess

of the unlabeled Nanofitin B10. Quantification analysis of PET images showed lower tumor uptake for mice injected with cold B10 or Cetuximab compared with the control group (20 µg spiking dose). The co-injection of Cys-B10 specifically reduced the tumor uptake of

18

F-FBEM-Cys-B10 from 1.99 ± 0.82

to 0.93 ± 0.50% ID/g at 2 h after injection (p = 0.026, Mann-Whitney test) and the blocking study with Cetuximab showed a reduction of the tumor uptake from 1.99 ± 0.82 to 1.25 ± 0.39% ID/g (p = 0.171, Mann-Whitney test).

Figure 9: A431 tumor targeting and specificity of

18

F-labeled B10 Nanofitin. Uptake in tumors of

18

F-

FBEM-Cys-B10 (2.1 – 9.1 MBq) in mice carrying EGFR-expressing A431 tumors, without blocking (n = 6) or with blocking amounts of non-labeled B10 (500 µg, n = 6) or Cetuximab (45 µg, n = 4) injected 48h p.i.. Data are expressed in percentage of injected dose per gram of organ [%ID/g] ± SD after


Page 25 of 37

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


intravenous injection of the probe. For Tumor PET images, please see Supplementary information (Fig. S9).



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

Page 26 of 37

Discussion In accordance with patient comfort, the ideal PET radiotracer for cancer imaging must combine fast clearance, efficient uptake, and specific tumor targeting with a significant tumor-to-background ratio shortly after tracer administration. Here, we introduced a first preclinical proof-of-concept of radiolabeled Nanofitins as a new class of molecular imaging probes showing a suitable pharmacokinetic profile for non-invasive monitoring of tumor response. Nanofitins are hyperstable, small single chain, alternative scaffold proteins amenable to high affinity against a defined target over several rounds of ribosome display

26,29,32

. The ribosome display, as a fully in vitro selection process,

offers the possibility of a fine tuning of the selection parameters ranging from the frame of the targeted epitope by the use of competitors to a forced cross-binding over multi-species analogues; the latter being of particular interest in the development of biopharmaceutics to avoid the tedious involvement of surrogates in the pre-clinical phases or misestimation of the biodistribution behaviors. Here, we engaged in the evaluation of the potential of Nanofitins as targeted PET-radiotracers to develop suitable companion diagnostics which ideally should not alter the pattern of cell surface antigens. We postulated that a non-internalizing Nanofitin would have a lesser propensity to affect the representation of EGFR at the surface of the tumor cells than an internalizing Nanofitin, and thus selected the anti-EGFR B10 Nanofitin which demonstrated high efficiency at labeling EGFR overexpressing A431 cells (Fig. 3) without triggering EGFR-mediated internalization in a Pe38 toxin fusion assay (Fig. 2). Moreover, binding on both human and murine EGFR is highly relevant for preclinical evaluation in mice with xenograft tumor expressing the human EGFR, in regard to off-target uptake of the anti-EGFR pharmaceutical. Despite a high identity of ~ 80 % between the extracellular domain of human and murine EGFR, therapeutic monoclonal anti-EGFR antibodies such as Cetuximab do not bind murine EGFR. In our case, the B10 Nanofitin binds to the extracellular domain of both human and murine EGFR, with affinities measured by biolayer interferometry of 27 and 83 nM, respectively (Fig. 4). While high affinity ligands have been found limited to the external layer of tumors due to the -8

-9

binding site barrier phenomenon, it has been reported that affinities in the range of 10 /10 M would provide an optimal balance between specificity of the targeting and capability to localize deeply into the tumor

42

.

Genetic engineering of the B10 Nanofitin allowed the insertion of a unique cysteine moiety at its Cterminus, which was then involved for its site-specific radiolabeling with the

18

F-FBEM prosthetic


Page 27 of 37

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


group. The resulting probe,

18

F-FBEM-Cys-B10, displayed a low accumulation in most normal organs

including heart, lung, brain, skin, testes, muscles, and bones. The very low uptake in bone (0.32 ± 0.09 %ID/g in BALB/c and 0.20 ± 0.01 %ID/g in the xenograft model at 2.5 h p.i.) is supporting the absence of defluorination of the

18

F-FBEM-Cys-B10 tracer, which is consistent with the in vitro stability

of the probe demonstrated over a 2 h incubation in FBS at 37 °C (Fig. 5). The highest uptake at 2.5 h after injection was found in the kidney (6.56 ± 2.40 %ID/g), liver (4.92 ± 2.06 %ID/g) and intestine (24.39 ± 5.85 %ID/g) (Fig. 6A). Uptake of the radioactivity in the kidney was shown to decrease over time to the profit of an increase in the bladder suggesting a renal clearance of the Nanofitin-based tracer, which is consistent for probes with a molecular weight below the cut-off of glomerular filtration (Fig. 6B and C). The absence of specific renal uptake is a major advantage for

18

F-FBEM-Cys-B10 as

molecular imaging probe, as it allows reduced dosage to the organ. The radioactivity uptake in the liver and the intestine might reflect the hepatic metabolism of the radioconjugate. However, considering the ability of B10 Nanofitin to bind to both human and murine EGFR with a similar affinity, the high uptake in liver could be explained by the elevated EGFR expression in this organ also supported by the lower liver uptake of

18

43

, which is

F-FBEM-Cys-B10 when co-injected with high excess of

unlabeled B10 Nanofitin during the blocking experiments (Fig. S10). While this off-target signal is not clinically limiting from an imaging perspective for non-hepatic tumors, further investigations are required to evaluate the potential contribution of additional parameters on the liver uptake of

18

F-

FBEM-Cys-B10 such as net charge, presence of a purification tag (hexa-histidine tag in this study) and choice of the radioelement; all of which having been already described as potential drivers of hepatic uptake 44–46. The

18

F-FBEM-Cys-B10 displayed an excellent tumor imaging quality of the over-EGFR-expressing

A431 tumor which is clearly delineated at 2 h after injection (Fig. 8B). The radiolabeled Nanofitin demonstrated a significantly higher retention in A431 tumor (1.42 ± 0.18 %ID/g at 2.5 h p.i.) than in H520 tumor (0.56 ± 0.10 %ID/g at 2.5 h p.i.) showing high specificity of the Nanofitin for EGFR-highexpressing tumors (p

Nanofitin as a New Molecular-Imaging Agent for the Diagnosis of

Recommend Documents