Backbone N-modified peptides: beyond N-methylation

Abstract

Backbone N-methylation is becoming an increasingly important tool in peptide drug design, and has been widely used to optimize the activity and selectivity of peptide ligands as a result of conformational modulation. However, no systematic research has been conducted on modifying the peptide backbone with other N-alkyl substituents. The present doctoral thesis is aimed at introducing novel N-substituents into peptides, and comparing the conformational and biological properties of the resulting N-substituted peptides with those of their N-Me homologues.

In a first project, we studied the effect of replacing backbone N-Me groups by an N-triethylene glycol (N-TEG) chain on hydrophobicity and conformation. For that, we chose Sansalvamide A peptide as a model, and we incorporated N-Me and N-TEG amino acids at the different positions of its cyclopentapeptide structure. We found that Fmoc-protected amino acids bearing the N-TEG group [i.e. N-CH2CH2(OCH2CH2)2OCH3] can be easily prepared in solution, and they are straightforward to incorporate into a resin-bound peptide. The acylation of N-TEG amines can be achieved in solid-phase by activating the following amino acid with triphosgene. In this way, N-TEG peptides are accessible by the same synthetic repertoire as that already established for N-Me peptides. Comparison of NMR data of our N-TEG vs. N-Me analogs gives evidence of similar conformational preferences for those peptides with the same N-alkylation pattern. Furthermore, comparison of their chromatographic retention parameters indicates that the incorporation of an N-TEG chain into a peptide provides a higher hydrophobicity than an N-Me group.

In a second study, we chose Cilengitide as model peptide, and we replaced its backbone N-Me group by various N-oligoethylene glycol (N-OEG) chains of increasing size: namely N-OEG2, N-OEG11, and N-OEG23, which are respectively composed of 2, 11 and 23 ethylene oxide monomer units. The N-OEG2 cyclopeptide analog was straightforward to synthesize in solid-phase, using the same methodology as for the N-TEG analogs of Sansalvamide A peptide. The syntheses of the N-OEG11 and N-OEG23 cyclopeptides are hampered due to the increased steric hindrance exerted by the N-substituent, and could only be achieved by segment coupling, which takes place with epimerization and thus requires extensive product purification. The different N-OEG cyclopeptide analogs and the parent peptide were compared with respect to biological activity and lipophilicity. The N-OEG2 analog displayed the same capacity as Cilengitide to inhibit integrin-mediated adhesion of HUVEC and DAOY cells to their ligands vitronectin and fibrinogen. The N-OEG11 and N-OEG23 analogs also inhibited cell adhesion, though with less potency. Thus, replacement of the backbone N-Me group of Cilengitide by a short N-OEG chain provides a more lipophilic analog with a similar biological activity. Upon increasing the size of the N-OEG chain, lipophilicity is enhanced, but synthetic yields drop and the longer polymer chains may impede receptor binding.

On the basis of our finding that N-alkyl chains exert similar conformational constraints as a backbone N-Me group when incorporated into a cyclic peptide, we studied the N-(4-azidobutyl) group as a linker to permit conjugation in peptides that lack derivatizable groups (i.e. N-terminus, C-terminus, and side-chain functionalities). We developed a robust strategy for the introduction of this linker into a peptide using standard solid-phase peptide synthesis techniques. With this methodology, we synthesized an analog of Cilengitide in which its backbone N-Me group was replaced by the N-(4-azidobutyl) group. This N-(4-azidobutylated) analog was used to prepare several conjugates with a polydisperse PEG chain (2 KDa), showing that our linker allows conjugation either via click chemistry or -after azide reduction- via acylation or reductive alkylation. This linker is orthogonal to protecting groups and resins commonly used in peptide chemistry, and chemically inert to a wide range of functionalities. NMR data indicated that Cilengitide and its N-(4-azidobutylated) analog have the same backbone conformation. Therefore, substitution of a backbone N-Me group by the N-(4-azidobutyl) linker is a valuable strategy to provide a reactive site for the attachment of molecules whilst preserving the original peptide sequence and conformation.

In summary, we have found that peptides bearing larger N-substituents than an N-Me group can be easily synthesized, but difficulties arise upon increasing the size of N-alkyl group. For Sansalvamide A peptide and Cilengitide, replacement of a backbone N-Me group by a short N-OEG chain resulted in analogs with similar biological activity and conformational features. This concept was then employed for the design of the N-(4-azidobutyl) linker, which allows bioorthogonal conjugation of a desired molecule with minimal perturbation of a target peptide structure. Considering the high abundance of N-Me groups in biologically active peptides, we contend that modification at this position is a feasible alternative to introduce chemical diversity or alter pharmacologically important parameters when modification at any other position of the peptide is not wished or possible.

En química medicinal, la N-metilació de l’esquelet peptídic s’ha utilitzat àmpliament per a imposar restriccions conformacionals en pèptids i així optimitzar la seva activitat i selectivitat. D’altra banda, la introducció de grups N-Me en pèptids d’interès terapèutic també és una estratègia per a millorar la seva biodisponibilitat, ja que els pèptids N-metilats són més hidrofòbics, més resistents al trencament proteolític, i -en general- més permeables a través de les membranes biològiques. No obstant, s’han descrit molts pocs exemples en els quals s’hagi modificat l’esquelet peptídic amb d’altres grups N-alquil. Això es pot atribuir a la dificultat d’acilar residus N-alquilats amb grups més grans que N-Me, degut al major impediment estèric.

L’objectiu principal de la present tesi ha estat explorar la viabilitat sintètica d’introduir nous N-substituents en pèptids, i comparar les propietats d’aquests nous pèptids N-substituïts amb les dels seus homòlegs N-metilats.

En aquesta tesi demostrem que els pèptids modificats amb una cadena de N-trietilenglicol (N-TEG) es poden preparar amb mètodes ja establerts per a la síntesi de pèptids N-metilats. En incrementar la llargada de la cadena de N-oligoetilenglicol (N-OEG), l’acoblament sobre el residu N-alquilat no és viable en fase sòlida, però es pot aconseguir solució utilitzant un clorur d’àcid. Per a dos ciclopèptids model, vam sintetitzar diversos N-OEG anàlegs, i vam trobar que la introducció del grup N-OEG augmenta la hidrofobicitat de forma proporcional a la llargada de la cadena. També vam trobar que el reemplaçament del grup N-Me present en aquests pèptids per una cadena curta de N-OEG provoca una mínima pertorbació de la seva conformació i activitat biològica.

En base aquesta observació, vam estudiar el grup N-(4-azidobutil) com a linker per a permetre la conjugació en pèptids que no poseeixen grups funcionals derivatitzables. Demostrem que el grup N-(4-azidobutil) es pot introduir en un pèptid utilitzant mètodes estàndard de síntesi en fase sòlida, i que la substitució d’un grup N-Me present en un pèptid pel nostre linker no altera la conformació del pèptid. També demostrem que grup N-(4-azidobutil) permet la conjugació mitjançant diverses transformacions químiques. Es tracta d’un linker ortogonal a la majoria de grups protectors emprats en síntesi de pèptids, i químicament inert a una gran varietat de grups funcionals.

En conclusió, la modificació de l’esquelet peptídic amb d’altres N-substituents més grans que N-Me és factible, però sorgeixen dificultats sintètiques en incrementar el tamany del grup N-alquil. Considerant que el grup N-Me es troba present en nombrosos pèptids biològicament actius, la seva substitució per d’altres entitats químiques és una alternativa viable per a introduir diversitat estructural o alterar propietats farmacològiques importants quan no és possible o no interessa modificar d’altres posicions d’un pèptid.