Cyclic peptides and small proteins in molecular recognition

Abstract

The present thesis is based on the work developed about the use of peptides and small proteins in protein surface molecular recognition.

Peptides present several advantages when compared with small molecules and biologics: (i) flexibility, which is translated into adaptability to large surfaces; (ii) easy modularity, which increases structural diversity and consequently allows higher selectivity and potency; (iii) size, which limits accumulation in tissue; and (iv) complete biocompatibility, which means low toxicity in humans. The last two features are highly desirable given the growing interest in PPIs as therapeutic targets. Moreover, non- natural building blocks and various chemical scaffolds can be incorporated into a peptide sequence to create a palette of modified peptides with a wide range of functionalities and chemical diversity. However, progress towards the development of therapeutic peptide PPI modulators is hindered by the following drawbacks of these molecules: low stability against degradation by proteolytic enzymes of the digestive system and blood plasma; rapid removal from the circulation; poor ability to cross physiological barriers; and potential immunogenicity. In spite of these limitations, the large number of successful peptide PPI modulators reported so far and the great effort to tackle the bottlenecks that impair their use as pharmaceutics are impressive. In addition to features that allow cell and tissue permeability, many chemical modifications and smart linker conjugations have been introduced into PPI modulators in order to reduce proteolytic degradation and improve bioavailability.

In the first chapter of this theses we described the use of cyclic peptides able to bind VEGF. For this purpose we screened a library of cyclic hexapeptides against VEGF by NMR. We found a cyclic hexapeptide (c(EpWEpW)) that bound VEGF in the receptor binding epitope with a low affinity. In order to gain more insight into how VEGF binding was affected by different parameters such as: C2 symmetry, amino acid replacement and stereochemistry, we evaluated a set of derivatives of peptide c(EpWEpW). Unfortunately none of the evaluated peptides displayed a better affinity than the parent compound. Finally we observed that ring size expansion was key to improve VEGF-binding and selectivity. We found a dodecapeptide that binds VEGF in the receptor binding domain with µM affinity. In this chapter we also described the conformational study of a set of cyclic hexapeptides.

In the second chapter we described the conformational analysis done with different apamin analogues in order to determine their conformational behavior to explain the differences observed in BBB- permeability when these peptides were tested in vitro.

Finally, the last chapter describes the work that I did during my short stay in Prof. Imperiali’s lab at MIT, where I was working on the development of fluorescence-based biosensors for protein detection. In this sense I performed the development and optimization of a protocol for the expression, purification and labeling with 4-DMN derivatives of Sso7d-based cysteine mutants. Yeast surface

display was also used in order to obtain Sso7d-based binders of a relevant target (hEGF). We obtained the first Sso7d-based hEGF binders that were also characterized in order to obtain their sequences. Nowadays, the conversion of these Sso7d-based hEGF binders into future biosensors is being done in order to detect hEGF in vivo.

Los péptidos presentan varias ventajas sobre las pequeñas moléculas y las proteínas: (i) flexibilidad,

(ii) fácil modularidad, (iii) tamaño reducido, y (iv) completa biocompatibilidad. Sin embargo, el progreso hacia el desarrollo de péptidos terapéuticos para modular interacciones proteicas se ve obstaculizado por los siguientes inconvenientes de estas moléculas: baja estabilidad frente a la degradación por encimas proteolíticas del sistema digestivo y del plasma sanguíneo; rápida eliminación del sistema circulatorio; dificultad para atravesar barreras fisiológicas; y potencial inmunogenicidad. A pesar de estas limitaciones, un gran número de péptidos han llegado a ser exitosos moduladores de interacciones proteína-proteína y el gran esfuerzo realizado para abordar este cuello de botella que perjudica su uso como fármacos es impresionante.

En el primer capítulo se describe el uso de péptidos cíclicos capaces de unir a VEGF. Para ello primero se realizó un cribado de la librería EXORIS frente a VEGF que dio lugar a la obtención de un hexapéptido cíclico (c(EpWEpW)) capaz de unir a VEGF en su región de unión a receptores con una afinidad baja. Mediante la evaluación de varios parámetros tales como: simetría C2, sustitución de aminoácidos, o la estereoquímica de este tipo de péptidos cíclicos, se examinaron varios derivados del péptido c(EpWEpW) pero ninguno de ellos dio mejores resultados en cuanto a afinidad. Finalmente se observó que la ampliación del tamaño del péptido pasando de hexapéptido a dodecapéptidos daba lugar a una mejora significativa de la afinidad así como la selectividad de unión a VEGF. Se obtuvo así un péptido cíclico capaz de unir a VEGF. En este capítulo se realizó también un estudio conformacional de un set representativo de hexapéptido cíclicos.

En el segundo capítulo se describe el estudio realizado para determinar la conformación de unos análogos de apamina que presentaron distintas permeabilidades frente a la barrera hematoencefálica.

Finalmente en el último capítulo de la tesis se describe el trabajo realizado durante mi estancia en el MIT, donde llevé a cabo el desarrollo y optimización de un protocolo para la expresión, purificación y marcaje de derivados de la proteína Ss07d para ser usados como biosensores capaces de detectar proteínas in vivo. En este capítulo se llevó a cabo también el desarrollo de nuevos “scaffold” basados en Sso7d capaces de unir a hEGF mediante la técnica de YSD (“yeast surface display”).