The influence of degree of labelling upon cellular internalisation of antibody-cell penetrating peptide conjugates

Antibody-based agents are increasingly used as therapeutics and imaging agents, yet are generally restricted to cell surface targets due to inefficient cellular internalisation and endosomal entrapment. Enhanced cell membrane translocation of antibodies can be achieved by the covalent attachment of cell-penetrating peptides, including the HIV-1-derived transactivator of transcription (TAT) peptide. This study evaluated the cellular internalisation properties of five anti-HER2 Herceptin–TAT conjugates with degrees of TAT labelling (DOLTAT) ranging from one to five. Herceptin–TAT conjugates were synthesised via a strain-promoted alkyne–azide cycloaddition reaction, characterised by UV-vis spectroscopy, MALDI-TOF, and gel electrophoresis, then radiolabelled with zirconium-89 to permit measurement of cellular internalisation by gamma counting. [89Zr]Zr–DFO–Her–TAT(0–5) conjugates were isolated in high radiochemical purity (>99%) and exhibited high stability in murine and human serum over 7 days at 37 °C. Significant increases in cellular internalisation were observed for [89Zr]Zr–DFO–Her–TAT conjugates with DOLTAT values of 2 and above in SKBR3 (high HER2) cells over 48 h, in contrast to low-level non-specific uptake in MDA-MB-468 (low HER2) cells that did not increase over time. Notably, [89Zr]Zr–DFO–Her–TAT conjugates with DOLTAT of 3, 4, and 5 reached uptake values in SKBR3 cells of 5, 6, and 8% of the applied dose at 48 h respectively, representing 9, 10, 14-fold increases relative to the TAT-free control conjugate, [89Zr]Zr–DFO–Her–TAT(0).


Introduction
Monoclonal antibodies (mAbs) are valuable tools in the era of precision medicine due to their ability to bind an array of clinically relevant cell surface proteins with high binding specicity, selectivity, and affinity. 1 Most mAb therapies are based on direct disruption of cancer cell activity (e.g. inhibition of growth factor receptor signalling) or immune activation (e.g. antibody-dependent cell-mediated cytotoxicity, ADCC), however they are also commonly modied with cytotoxic cargoes to yield antibody-drug conjugates (ADCs) with high therapeutic indices. A critical determinant of ADC efficacy is sufficient internalisation in target cell populations to ensure rapid intracellular delivery of the cytotoxin and minimise toxicity arising from extracellular delivery. It is also worth noting the promising utility of mAb-based companion diagnostics in positron emission tomography (immunoPET) investigations for aiding patient selection and enabling rapid assessment of treatment response. However, this otherwise attractive imaging modality is similarly limited by inefficient internalisation and endosomal entrapment of mAb-derived probes which restricts the scope of this application to a small subset of clinically relevant cell surface biomarkers, in contrast to the rich pool of biomarkers located inside cells. [2][3][4] Cell-penetrating peptides (CPPs, also known as protein transduction domains) are short peptides (typically 10 to 30 amino acids) capable of enhancing the cellular internalisation of either covalently conjugated or co-delivered cargos, including mAb-based imaging and therapeutic agents, 5-7 mostly by energy-dependant endocytosis mechanisms. 8,9 Notably, the TAT peptide (GRKKRRQRRRPPQGYG) derived from the transactivator of transcription protein of the HIV-1 virus has been widely applied as a CPP, including in the development of antibody-CPP conjugates which have emerged in recent years. 3 For example, an immunoPET agent based on a TAT-functionalised anti-gH2AX antibody has been developed to monitor DNA damage response signalling during tumorigenesis and cancer therapy. 10,11 Furthermore, two recent independent studies by Tietz et al. and Sauter et al. reported signicantly enhanced cytosolic delivery of mAb-derived agents via the integration of multimeric CPPs (trimeric and tetrameric, respectively), highlighting the importance of several CPP-related parameters, including degree of labelling, concentration, spatial proximity, and local charge density, upon cellular internalisation. 12,13 Each study also reported that non-specic cell uptake increased at higher CPP : mAb ratios which indicates the need to carefully optimise this key parameter to maximise specic uptake in target cell populations. 14 Here, we describe the synthesis, characterisation, and in vitro cellular internalisation properties of ve anti-HER2 Herceptin-TAT conjugates with degrees of TAT labelling (DOL TAT ) ranging from 1-5 with the aim to facilitate the future optimisation of CPP-modied antibody-based pharmaceutics. To this end, a bioconjugation strategy based on strain-promoted alkyneazide cycloaddition (SPAAC) 2 that enables UV/vis-based DOL TAT determination was utilised that distinguishes the Herceptin-TAT conjugates from those previously reported and synthesised using an amine-to-sulydryl (sulfo-SMCC) crosslinker. 15

Materials and methods
All reagents were purchased from Fisher Scientic unless otherwise stated and used without further purication. Water was deionized using a Select Fusion ultrapure water deionisation unit (Suez) and had a resistance of >18.2 MU cm −1 at 25 C. Protein concentration measurements were obtained using a NanoDrop One C Microvolume UV-vis Spectrophotometer (NanoDrop Technologies, Inc.). MALDI-TOF mass spectrometry measurements were taken on a Bruker Microex LRF. Radioactivity measurements were obtained using a CRC-25 Dose Calibrator (Capintec, Inc.) or a Wizard 2480 Gamma Counter (PerkinElmer). Radioimmunoconjugate synthesis and serum stability studies were monitored by instant thin-layer chromatography using glass microber chromatography paper (iTLC-SA, Agilent). Radio-iTLC strips were measured by autoradiography (Amersham Typhoon Bioimager, GE) and analysed using ImageQuant soware (GE Healthcare). pH measurements were determined using pH indicator paper (Merck Millipore) or a pH Spear electrode (Eutech Instruments).

SPAAC conjugation of TAT-N 3 to mAb-DBCO
Azide-modied TAT (1 mM in PBS, TAT-N 3 ¼ GRKKRRQRRRPPQGYG-hA(N 3 ), Cambridge Peptides) was added in 10-fold molar excess to a 5 mg mL −1 solution of Her-DBCO in PBS (pH 7.2) then incubated at 25 C for 2 h with gentle shaking (450 rpm). The product, Her-TAT, was puried by passing the reaction mixture through a pre-rinsed 30 kDa molecular weight cut-off 0.5 mL centrifugal lter at 12 000 Â g for 10 minutes and washing three times (3 Â 500 mL) with PBS (pH 7.2). The absorbances of Her-TAT (1)(2)(3)(4)(5) conjugates at 280 and 309 nm were measured by UV-vis spectroscopy to determine the degree of TAT labelling and antibody concentration (see Section 2.4).

Degree of labelling determination
The average number of DBCO moieties attached to each antibody (DOL DBCO ) was calculated using the following equation: CF ¼ correction factor (A 280 /A 309 ), 3 309 ¼ molar attenuation coefficient at 309 nm, 3 1 % ¼ percent molar attenuation coefficient for a 10 mg mL −1 IgG solution.
Following the SPAAC reaction between TAT-N 3 and Her-DBCO and purication by size exclusion chromatography (Section 2.5), the average number of residual unreacted DBCO moieties attached to each antibody (DOL DBCO* ) was determined by the same approach. The average number of TAT peptides attached to each DBCO-modied antibody (DOL TAT ) was determined by subtraction of DOL DBCO* from the initial DOL DBCO (DOL TAT ¼ DOL DBCO − DOL DBCO* ).

SDS-PAGE
Her-TAT (0-5) conjugates were analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions. Antibody samples (1-2 mL, 1 mg mL −1 ) were prepared by the addition of NuPAGE 4Â LDS sample buffer (2.5 mL), NuPAGE 10Â Sample Reducing Agent (1 mL), and deionised water (4.5-5.5 mL) to a total volume of 10 mL. The resulting solutions were incubated at 70 C for 10 min at 450 rpm. Protein samples and molecular weight standards (Thermo Scientic PageRuler Unstained Broad Range Protein Ladder, 5 mL) were loaded on a 10-well protein gel (4-12% Bis-Tris) and run for 45 minutes at 200 V in NuPAGE MOPS SDS running buffer. The gel was subsequently washed three times in deionised water (200 mL, 5 min) then stained with SimplyBlue SafeStain (50 mL, Invitrogen) for 1 h. Destaining was performed by washing three times in deionised water (200 mL, 1 h).

MALDI-TOF
A saturated solution of a-cyano-4-hydroxycinnamic acid (CHCA, 20 mg) in acetone (500 mL) was prepared as Mix 1. A precursor saturated solution of CHCA (20 mg) in 70% acetonitrile with 5% formic acid (500 mL) was prepared alongside a saturated solution of 2,5-dihydroxybenzoic acid (DHB, 20 mg, Sigma-Aldrich) in 70% acetonitrile with 0.1% triuoroacetic acid (500 mL). Solutions were prepared at room temperature and vortexed thoroughly for 60 seconds before use. DHB (100 mL) and CHCA (100 mL) solutions were then combined to prepare Mix 2. Mix 1 (0.5 mL) was spotted onto a polished steel target plate (Bruker) and evaporated quickly to leave a thin layer of CHCA. A 0.5 mL aliquot of protein sample (typically 10 mM in PBS) was spotted directly onto the layer. Then, 0.5 mL of Mix 2 was added to the liquid droplet and allowed to dry. Where required, antibody samples were reduced by incubation with dithiothreitol (DTT, 10 mM) at 60 C for 30 min.

Radiolabelling
Zirconium-89 in 1 M oxalic acid ($5-10 MBq, PerkinElmer) was adjusted to pH 7 by the addition of 1 M Na 2 CO 3 . The solution was briey shaken and le at room temperature for 2-3 min until the evolution of CO 2(g) had stopped. The neutralised 89 Zr solution was added to DFO-Her-TAT (0-5) ($100 mg, 2 mg mL −1 ) and then incubated at 25 C for 1 h with gentle shaking (450 rpm). The radiolabelling efficiency and radiochemical purity was assessed by radio-instant thin layer chromatography (radio-iTLC) using 50 mM EDTA (pH 5.5) as the mobile phase, where 89 Zr-labelled antibody conjugates remained at the origin (R f ¼ 0) and free 89 Zr migrated to the top of the iTLC strip (R f ¼ 0.8-1). 16 Radioimmunoconjugates were used in in vitro assays when RCP was >99%.

Cell culture
The human breast cancer cell lines SKBR3 (high HER2 expression) and MDA-MB-468 (low HER2 expression) were obtained from American Type Culture Collection (ATCC). Cell lines were maintained in RPMI (Sigma-Aldrich, R8758) and DMEM (Gibco, 41965039) medium, respectively. Media were supplemented with 10% foetal bovine serum (FBS, Sigma-Aldrich, F9665), 2 mM L-glutamine, penicillin (100 units per mL), and streptomycin (0.1 mg mL −1 , Sigma-Aldrich, G6784), and maintained in a 5% CO 2(g) humidied atmosphere at 37 C. Cells were harvested and passaged when they reached 85-95% conuency using trypsin-EDTA solution (Sigma-Aldrich, T4049). The cumulative length of culture was less than 6 months following retrieval from liquid nitrogen storage. Cells were tested for the absence of mycoplasma at regular intervals.

Cell internalisation assay
SKBR3 and MDA-MB-468 cells were seeded in 0.5 mL media in 24-well plates at the necessary density to achieve 50 000 cells per well at each time point of the experiment. To each well, 1.47 mL was added from respective [ 89 Zr]Zr-DFO-Her-TAT (0-5) stock solutions (0.21 AE 0.02 MBq, 2.42 mg) diluted in 0.5 mL of media. At 0.75, 1.5, 4, 24 and 48 h incubation, the supernatant media (free fraction) was transferred to allocated counting tubes and combined with 0.5 mL of PBS (pH 7.2) used to wash the cells. Cells were then incubated in cold 0.1 M glycine HCl (0.5 mL, pH 2.5, Sigma-Aldrich) for 10 min and this solution was then transferred to a separate series of counting tubes (membranebound fraction) followed by the addition of a 0.5 mL PBS (pH 7.2) cell wash. Lastly, RIPA lysis and extraction buffer (0.1 mL, Thermo Scientic) was added to each well and plates were placed on ice for 5 min before transferring the lysed cell solutions (internalised fraction) to a third series of counting tubes, followed by the addition of a 0.5 mL PBS (pH 7.2) cell wash. The activity in each counting tube was measured as counts per minute (cpm) on a gamma counter, converted to activity units (MBq) using recent calibration data, and decay corrected to the start of the experiment.

Statistical analysis
All statistical and regression analyses were performed using GraphPad Prism v9 (GraphPad Soware, San Diego, CA, USA). A condence interval of 95% (P < 0.05) was considered statistically signicant. Unpaired t-tests or two-way ANOVA followed by Sídák's multiple comparisons test was used to compare multiple groups. All data was obtained in at least triplicate, and results were reported and graphed as mean AE standard deviation, unless stated otherwise. Statistical signicance is shown by asterisks where ns ¼ not signicant, * ¼ p < 0.05, ** ¼ p < 0.01, *** ¼ p < 0.001, **** ¼ p < 0.0001.

Cellular internalisation
The baseline control conjugate without TAT, [ 89 Zr]Zr-DFO-Her (0) , had low uptake in SKBR3 (high HER2) cells that marginally increased over time to 0.58 AE 0.04% of the applied dose at 48 h and remained consistently higher than in MDA-MD-468 (low HER2) cells which showed no signicant uptake (Fig. 5). A similar uptake prole was observed for the radioimmunoconjugate modied with a single TAT, [ 89 Zr]Zr-DFO-Her-TAT (1) , which revealed no enhancement of internalization in either cell line relative to the baseline conjugate, indicating that a DOL TAT > 1 is needed to increase uptake (Fig. S5 †).
In contrast, [ 89 Zr]Zr-DFO-Her-TAT (2) was internalized in SKBR3 cells to a greater extent than the TAT-free control at each time interval and achieved an uptake value of 0.78 AE 0.03% at 0.75 h that steadily increased to 1.01 AE 0. 11 (Fig. S6 †). The average non-specic contribution was 0.67 AE 0.05, 0.93 AE 0.07 and 1.40 AE 0.17% of applied dose at 48 h for conjugates with DOL TAT values of 3, 4, and 5, respectively.
The steady time prole of internalization in this low HER2 cell line suggests that this uptake is being offset by externalization which may occur via a mechanism proposed by Rayne et al. that describes externalization of TAT by phosphatidylinositol-(4,5)-bisphosphate on the inner leaet of the cell membrane. 17

Conclusions
This study has examined the relationship between DOL TAT and cellular internalisation over time using a clinically relevant model based on Herceptin. The isolation of Herceptin-TAT conjugates with well-dened DOL TAT values between 1 and 5 was readily achieved in high yields using a synthetic route based on strain-promoted alkyne-azide cycloaddition. For Herceptin conjugates with DOL TAT values $ 2, signicant enhancements in internalisation relative to a TAT-free Herceptin control were observed in HER2 positive SKBR3 cells but not in HER2 negative MDA-MB-468 cells, indicating HER2 specic uptake in cells. The extent of this enhancement was most prominent aer 48 h incubation and increased towards the highest DOL TAT , although conjugates with DOL TAT values of 2 and 3 showed a lower proportion of uptake in HER2 negative cells which suggests that TAT modication in this range may represent an optimal balance between total uptake and specicity. The signicant increases in cellular internalisation observed in this study will be of interest to the drug development and molecular imaging communities as it may serve to increase the therapeutic indices of ADCs and expand the scope of available imaging biomarkers for immunoPET.
The systematic nature of this investigation has yielded data that contributes to our understanding of the inuence of CPPs upon the cellular internalisation characteristics of antibodies. A limitation of this study, however, is its examination of just one of a great multitude of CPP candidates, and while TAT was selected on the basis of its prominent application in biomedical research over recent decades it is likely that other CPPs (e.g. cyclic-and multimeric-TAT, 12,13,18 r9, 19 CPP12 (ref. 20)) may yield even greater enhancements in internalisation and pharmacological properties, 21 and the optimisation of this nextgeneration of antibody-CPP conjugates would benet from a similar investigation to that described herein. With the principal aim of developing a facile, modular strategy capable of enhancing the therapeutic index of antibody-based drugs, future work will examine the applicability of these ndings to other clinically relevant antibodies and ADCs, and systematically assess the impact of DOL TAT upon key factors, including cellular internalisation mechanisms, immunoreactivity, pharmacokinetics, and biodistribution in preclinical models.

Conflicts of interest
There are no conicts to declare.