Fluorescence Studies: A9 Peptide, Functionalized with a Fluorogenic Probe, Interacts with Its Receptor Model HER2-DIVMP

Abstract

A recently developed synthetic protocol allowed for the functionalization of the active peptide A9 with a fluorogenic probe, which is useful for studying biomolecular interactions. Essentially, a nucleophilic attack on a halo-substituted benzofurazan is selectively performed by a cysteine sulfhydryl group. The process is assisted by the basic catalysis of activated zeolites (4 Å molecular sieves) and promoted by microwave irradiation. Fluorescence studies revealed that a donor–acceptor pair within the peptide sequence was introduced, thus allowing a deeper investigation on the interaction process between the peptide ligand and its receptor fragment. The obtained results allowed us to come full circle for all the currently understood structural determinants that were found to be involved in the binding process.

Over the past 20 years, fluorescence spectroscopy has been widely employed in the field of biological chemistry, since it provides a quite rapid information about the interaction among biomolecules. FRET (fluorescence resonance energy transfer) has also been implemented, allowing further characterizations that concern the supersolved optical measurements of intra- and intermolecular distances within the molecular target.¹

Fluorescent tags based on substituted 2,1,3-benzoxadiazole (benzofurazan) have been long used to label proteins and peptides for biological assays.²⁻⁷ Recently, we have introduced a new method for the bioconjugation of cysteine containing peptides with benzofurazan halogenides as fluorescent tags.⁸ Such a procedure relies on the selective S-alkylation of the cystein thiol group by benzofurazan halogenides, promoted by the mild basic catalysis of activated zeolites (4 Å molecular sieves). We have shown that this reaction is very chemoselective, and it can be performed even in the presence of other unprotected nucleophilic groups (Met, Trp, Thr, His, Lys) without compromising the yield. The high chemoselectivity for the cysteine sulfur atom over other potentially competing nucleophiles was confirmed in a number of model peptides by fluorescence and NMR spectroscopy.⁸

In this report, we describe the application of our synthetic method to functionalize the bioactive peptide called A9 with the ionic 7-sulfobenzofurazan as the fluorescent tag.⁹⁻¹¹ Peptide A9 is a nine-amino acid peptide that binds specifically to the extracellular domain IV of the HER2 receptor with a nanomolar dissociation constant. The design, synthesis, and validation of this peptide was previously described, as well as its application as a radiolabeled tracer for molecular imaging of HER2.¹²⁻¹⁵ The A9 sequence was rationally designed by analyzing the X-ray structure of the bimolecular complex between the extracellular region of the HER2 receptor and the antigen-binding fragment (Fab) of Herceptin, an antibody specific for HER2. The A9 sequence represents the minimal sequence derived from Herceptin Fab making closer contacts with the HER2 extracellular domain. In the same study, the minimal binding region of the receptor was identified as well. Such a region, called HER2-DIVMP (Domain IV mimicking peptide), was chemically synthesized and validated as a synthetic model of the Herceptin binding domain. It consists in a modified fragment of the receptor HER2 that was previously proved as fully representative of the receptor domain IV binding properties, as well as easy to obtain and implement in reproducible screening of ligands.¹⁶

As the plain A9 peptide does not contain cysteine, we have elongated the N-terminus of A9 with a N-acetyl-cysteine residue in order to perform an S-conjugation with 4-chloro-7-sulfo-benzofurazan (Cl-Sbf). The key feature of the synthetic protocol consists in using activated molecular sieves (MS) as the basic catalyst to activate the thiol function for nucleophilic substitution.^17,18 Peptide A10 (i.e., the N-terminus elongated A9 peptide, Scheme 1) was dissolved in DMF, under an argon atmosphere, and added with an excess of 1.2 equiv of 4-chloro-7-sulfobenzofurazan. Molecular sieves (4 Å MS activated at 280 °C for 4 h under vacuum, then allowed to cool to room temperature) were finally added to the reaction mixture. This was irradiated with microwaves and stirred for 5 min at 40 °C in a microwave source apparatus. (Scheme 1).

Scheme 1. Incorporation of Sulfo-Benzofurazan in Peptide A10.

The nucleophilic substitution of the cysteine sulfhydryl occurred selectively and with a good yield (85%), even in the presence of potentially competing nucleophilic groups, such as the carboxylic function of aspartic acid and the indolic nitrogen of tryptophan. The chemical structure of the final conjugate (compound A10-Sbf) was unambiguously confirmed by NMR spectroscopy. All ¹H NMR resonances were assigned according to the sequence specific method. The expected short distances between the cysteine H_α/H_β protons and the benzofurazan H5 ring proton yielded clear NOE peaks in 2D-NOESY spectra (Figure 1). No NOE distances connecting the benzofurazan ring protons with side chain protons of potentially competing alkylation sites (especially aspartic acid and tryptophan) were detected.

Figure 1.

Overlay between the 2D-TOCSY (black) and the 2D-NOESY (red) NMR spectra of A10-Sbf (600 MHz, H₂O/D₂O 600:50, 288 K) showing the expansion over the fingerprint region. 2D-TOCSY strips with amino acid assignment is shown. NOE cross-peaks connecting the benzofurazan ring H5 proton and cysteine H_α and H_β protons are indicated.

Next, the binding affinity of the fluorescent peptide ligand toward its receptor model HER2-DIVMP was studied by means of fluorescence spectroscopy titrations. As the method we have optimized uses excitation at 280 nm, we first investigated the emission spectrum of the ligand A10-Sbf upon excitation at such wavelength. As shown in Figure 2A, two emission peaks can be found: one at λ_max = 354 nm related to the tryptophan residue, and second one at λ_max = 533 related to the sulfobenzofurazan fluorescent tag. It is worth noting that the latter emission requires excitation at 380 nm, while it should be negligible upon excitation at 280 nm. In addition, it was demonstrated by a control peptide AcCGTVANH₂–Sbf, which contains a sulfobenzofurazan group conjugated to cysteine (similarly to A10-Sbf) but lacks of any tryptophan residue. Excitation of this control peptide at 280 nm yields negligible emission at 520 nm, while it could clearly be observed by direct excitation at 380 nm (Figure 2B), likewise to what was observed for A10-Sbf (Figure 2A). Thus, we were allowed to conclude that in A10-Sbf a transfer of resonance energy takes place from tryptophan to the benzofurazan group, such that both the tryptophan and the Sbf emission were observed.¹⁹⁻²¹

Figure 2.

Emission fluorescence spectra of A10-Sbf upon excitation at 280 nm (panel A) and of a generic peptide conjugated to a sulfo-benzofurazan moiety upon excitation at 280 and 380 nm (panel B).

This phenomenon was expected because the spectral overlap of donor emission (Tryptophan) and acceptor absorption (S-linked sulfo-benzofurazan) provides the requirement necessary for the energy transfer (Figure 3A). It is worth noting that the same efficient overlap was not observed for the peptide control that lacks a tryptophan residue, so that any efficient overlap could be generated between a potential donor and the benzofurazan acceptor (Figure 3B).²²

Figure 3.

Overlap of donor fluorescence emission and acceptor absorption (absorbance) spectra normalized for A10-Sbf (panel A) and for a control peptide AcCGTVANH2-Sbf (panel B).

Fluorescence binding titrations were performed according to the limit reagent methos, by exciting at 280 nm and measuring the change of fluorescence intensity at 354 nm (ΔF₃₅₄) between solutions containing an increasing amount of the ligand in the presence or absence of a fixed amount of the receptor. We have previously shown that, after the formation of the complex between A9 and the HER2-DIVMP receptor fragment, the emission at 354 nm increases (positive sign of ΔF₃₅₄). This effect was explained in terms of fluorescence resonance energy transfer from the donor tyrosine Y568 of HER2-DIVMP to the acceptor tryptophan W35 of the A9 ligand.⁹

When applied to the A10-Sbf/HER2-DIVMP ligand/receptor interaction, we observed ΔF₃₅₄ with negative sign, indicating that the complex formation decreased the fluorescence emission intensity at 354 nm. This can be explained in terms of a complex energy transfer among multiple fluorophores (W35 and the Sbf tag of A10, and Y568 of the receptor (Figure 4).

Figure 4.

Upon the binding process, the FRET phenomenon occurs from Y568 of the receptor target to the W39 of the ligand peptide. A second FRET phenomenon also occurs within the peptide, from W39 to the Sbf conjugated to the cysteine sulfur.

These fluorescence titrations were repeated at receptor concentrations in the range 0.9–2.5 μM, always giving a change of fluorescence emission at 354 nm with negative sign. The plot of ΔF₃₅₄ as a function of ligand concentration yielded consistently hyperbola-shaped curves (a representative curve is shown in Figure 5). Computer aided best-fitting of such hyperbola according to the model for saturable specific binding²³ (see SI) led to a dissociation constant K_d of 337 nM.

Figure 5.

Typical titration curve for the binding of A10-Sbf to HER2-DIVMP (10 mM phosphate buffer, pH= 7.2). The receptor concentration was 2.4 μM. The ordinate represents the residual fluorescence signal at 354 nm after subtraction of the individual contributions of HER2-DIVMP and the peptide ligand.

However, the fluorescence signals collected at λ_em = 520 nm increased because of a net energy transfer to the Sbf tag (acceptor). However, the scattering points of the titration curve did not allow the fitting calculation (Figure 6)

Figure 6.

Typical titration curve with ordinates representing the residual fluorescence signal at 520 nm after subtraction of the individual contribution of the peptide ligand.

It can be concluded that a multiple FRET coupling is observed upon interaction between A10-Sbf and our HER2-DIVMP receptor fragment. Indeed, the introduction of the Sbf moiety allows for the formation of a second donor–acceptor pair, since appropriate spectral overlap and proximity requirements are satisfied. Namely, the excited tryptophan donor transfers energy to a suitable and nearby acceptor fluorophore such as the S-sulfo-benzofurazan (Figure 4)

Next, we studied the binding process of the complex HER2-DIVMP/A10-Sbf by investigating the fluorescent properties of the S-Sbf moiety. It is worth remembering that the fluorescent intensity of 7-sulfobenzofurazan is strongly affected by the solvent polarity. Namely, it increases sharply in nonpolar solvent or in hydrophobic environment, such as receptor binding sites. Thus, fluorescence titrations were also carried out by fixing the excitation wavelength at 380 nm, where it can be assumed that the emission originates only from the sulfobenzofurazan moiety of the peptide ligand. In fact, the fluorescence emission from tyrosine as well as tryptophan can be considered negligible at this excitation wavelength. After subtraction of the individual signals of the peptide ligand A10-Sbf, no hyperbolic modification of the residual fluorescence was detected (see SI). This confirmed that the binding process between A10-Sbf and HER2-DIVMP does not involve the N-terminal region of the peptide ligand to any relevant extend.¹⁰

In conclusion, we applied our mild, efficient, and chemoselective procedure to label a preformed bioactive peptide with a benzofurazan-based fluorophore. The A10-Sbf tracer was obtained with good yield and purity, demonstrating the great potential, reliability, and versatility of the synthetic protocol.

The chemical modification introduced in A9 peptide sequence allowed a further investigation of the binding process of the active peptide ligand with its synthetic model HER2-DIVMP, essentially confirming the structural determinants already studied. However, the binding affinity of A10-Sbf decreased by approximately 1 order of magnitude in comparison to parent A9, although molecular modeling studies about the A9/HER2-DIVMP complex showed that N-terminus modifications of A9 may be well tolerated.¹⁰ In the future, we planned the design of a flexible spacer to insert between the fluorophore and the targeting peptide, in order address further binding studies.

Glossary

ABBREVIATIONS

AcCGTVANH₂

Ac-Cys-Gly-Tyr,Val,Ala-NH₂

A9

Ac-Gln-Asp-Val-Asn-Thr-Ala-Val-Ala-Trp-NH₂

A10

Ac-Cys-Gln-Asp-Val-Asn-Thr-Ala-Val-Ala-Trp-NH₂

Cl-Sbf

4-chloro-7-sulfo-benzofurazan

HER2-DIVMP

domain IV mimicking peptide

¹H NMR

proton nuclear magnetic resonance

FRET

fluorescence resonance energy transfer

HER2

human epidermal growth factor receptor 2

MD

molecular dynamic

MS

molecular sieves

MW

microwave

NOESY

the nuclear overhauser effect

SI

Supporting Information

TOCSY

total correlation spectroscopy

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.2c00026.

• Experimental section, HPLC profiles, MS spectra, NMR characterization, absorption spectroscopy, fluorescence study and treatment of binding data of functionalization of peptide A9 with a fluorogenic probe (PDF)

Author Contributions

The manuscript was written through contributions of all authors. S.D.L. conceptualized manuscript and performed the writing. All authors have given approval to the final version of the manuscript.

Università del Piemonte Orientale, Fondi di Ateneo per la Ricerca 2017 (FAR2017) Consiglio Nazionale delle Ricerche, Progetto DCM.AD007.079-Targeting e delivery molecolare

The authors declare no competing financial interest.