Peptides capable of interfering with the mdm2 / mdm4 heterodimer association and their use in cancer treatment

Abstract

The invention relates to short peptides, of amino acids in length, capable of binding with high affinity to the MDM2 protein and interfering with the MDM2 / MDM4 heterodimer activity, thereby hindering the inhibitory function of the heterodimer against the tumor suppressor p53. The invention also relates to the use of the aforementioned peptides, or nucleic acid molecules encoding them, in the treatment of a tumor, preferably a solid tumor expressing wild-type p53.

Description

[0001] Peptides capable of interfering with the MDM2 / MDM4 heterodimer association and their use in cancer treatment

[0002] The present invention relates to short peptides capable of interfering with the MDM2 / MDM4 heterodimer association, their formulation into nanoparticles, and their use in the therapeutic treatment of cancer. More specifically, the invention relates to short peptides capable of binding with high affinity to the carboxy-terminal end of the MDM2 protein, thus interfering with the MDM2 / MDM4 heterodimer association and hindering or preventing its inhibitory activity against the tumor suppressor p53. p53 protein is a transcription factor that regulates several cellular responses and serves as a tumor suppressor, or oncosuppressor.

[0003] The tumor suppressor p53 has long been considered an extremely interesting target in cancer studies. Even in tumors expressing wild-type p53, the hypothesis that its function is impaired has led to the development of a large number of reactivators. Since the discovery of MDM2 mouse double minute 2 homolog"), which represents the best characterized p53 negative regulator, researchers have developed numerous molecules with therapeutic potential. Unfortunately, none of them have reached clinical use at this time. The discovery of a second inhibitor of the p53 tumor suppressor, namely MDM4 ("double minute 4 human homolog"), partially explained this failure and prompted researchers to partially redesign strategies for reactivating the tumor suppressor function of p53 in human tumors. MDM2 inhibitors were then retested on MDM4, or new dual inhibitors were developed. At the same time, the heterodimer composed of MDM2 and MDM4 (MDM2 / MDM4) was found to be a more efficient p53 inhibitor than individual molecules, which led to the formulation of the proposal to target the activity of this heterodimer. However, the MDM2 / MDM4 interaction region is difficult to target with small molecules, as it includes a P-sheet structured protein portion with a relatively flat surface.

[0004] European patent application EP 2 639 240 A2 describes peptides of the MDM4 protein which are capable of hindering the inhibitory activity of the MDM2 / MDM4 heterodimer against p53. The shortest of the peptides exemplified in EP 2 639 240 A2 consists of the amino acid sequence corresponding to the carboxy-terminal portion of MDM4, from amino acid position 479 to amino acid position 490 of MDM4 (12 amino acids in length). In this description, this prior art peptide is identified as “Pep3”.

[0005] Patent application EP 2 639 240 A2 expressly mentions problems of intrinsic instability of the peptides described therein, which limit their therapeutic applicability. It is therefore clear that there is a need for new therapeutic agents for cancer treatment which overcome the disadvantages of the prior art.

[0006] In order to overcome these and other disadvantages, the present invention now provides isolated peptides derived from the MDM4 protein which are much shorter than the Pep3 peptide described in EP 2639240 A2 and which, despite their short length, are unexpectedly capable of interfering more effectively with the MDM2 / MDM4 heterodimer association than Pep3, as demonstrated by the significantly lower dissociation constant (KD) with regard to the MDM2 recombinant target (lower effective concentration compared to Pep3).

[0007] In addition, the peptides of the present invention unexpectedly exhibit some molecular and functional properties that differentiate them from Pep3, as will be described in more detail below.

[0008] A first object of the present invention is therefore a peptide of 5 amino acids in length selected from the group consisting of a peptide of sequence KVFIA (SEQ ID NO: 1), a peptide of sequence PWFRW (SEQ ID NO:2), and variants of the peptide of sequence SEQ ID NO:2 in which the proline residue (P) at position 1 is replaced with a 5-oxoproline residue (5oP), and optionally the tryptophan residue (W) at position 2 is replaced with a phenylalanine residue (F) or the phenylalanine residue (F) at position 3 is replaced with a homophenylalanine residue (homF).

[0009] Although the peptides of the present invention are characterized by an amino acid sequence consisting of only 5 amino acids, in molecular modelling (docking) studies they have been shown to be capable of interacting with MDM2 with high affinity and interfering very effectively with the MDM2 / MDM4 heterodimer activity. The KVFIA peptide (SEQ ID NO: 1), referred to herein as “Pep3S”, was the first to be identified by the present inventors, using molecular dynamics simulation studies performed on the Pep3-MDM2 complex, through which it was found that the amino-terminal portion of Pep3 does not contribute to the binding to the MDM2 target. This result was unexpected because, based on the structure, it was not possible to predict a priori whether the aminoterminal tail of Pep3, being a part of the MDM4 protein, would contribute to other properties, functional to the activity of the peptide.

[0010] The other peptides defined above were designed from Pep3S by means of a computational docking analysis, which allowed the identification of specific modifications of the amino acid sequence which increase the predicted binding affinity and / or stability of the peptide without however affecting its ability to interfere with the MDM2 / MDM4 heterodimer association. It is also noted that the two modifications, i.e., 5-oxoproline and homophenylalanine, which may be present in the peptide sequence, play a protective role against degradation. Homophenylalanine also confers an advantage to the interaction, as predicted by the docking analysis, by forming a more favourable peculiar hydrophobic - stacking interaction with the phenylalanine residue in position 490 of MDM2.

[0011] A further object of the present invention is an isolated nucleotide sequence encoding for a peptide as defined above.

[0012] Another object of the present invention is an expression vector comprising the nucleotide sequence as defined above.

[0013] A pharmaceutical composition comprising at least one peptide as defined above, or at least one nucleotide sequence as defined above, or at least one expression vector as defined above, in combination with one or more pharmaceutically acceptable excipients, carriers and / or adjuvants is also within the scope of the invention.

[0014] Given the antitumor activity of the peptides described herein, the scope of the invention also includes the peptides, nucleic acid sequences, expression vectors and pharmaceutical compositions as defined above for use in the therapeutic treatment of a tumor, preferably a solid tumor, even more preferably a solid tumor expressing wild-type p53, or of a benign hyperproliferative condition preferably expressing wild-type p53. Preferably, the subject being treated is a human patient. In this context, the term “tumor” refers to a primary tumor or a tumor metastasis. In vivo studies carried out by the inventors have in fact shown that treating a primary tumor with a peptide of the present invention results in a reduced formation of metastases, which is indicative of anti-metastatic efficacy.

[0015] The peptides, nucleic acid sequences, expression vectors and pharmaceutical compositions of the present invention may be administered through any suitable route of administration, for example selected based on the type of tumor to be treated, such as the intravenous or infusion route, or using liposomal vectors directly applicable to the tumor, or by gene therapy.

[0016] In a preferred embodiment, a peptide of the present invention is encapsulated in a polymeric nanoparticle, such as a poly(lactic-co-glycolic acid) nanoparticle. This formulation of the peptides of the invention is particularly preferred because, as will be described in detail below, encapsulation in poly(lactic-co-glycolic acid) nanoparticles (“PLGA-NPs”) improves the biodistribution and stability of the peptides.

[0017] Alternatively, in order to further improve the biodistribution and / or stability, a peptide of the present invention may be modified through further modifications capable of promoting cellular uptake, conferring protease resistance, and increasing overall stability. These techniques for modifying peptides are well known to those skilled in the art and can be applied to any peptide according to the present invention.

[0018] The examples that follow are provided for illustration purposes and do not limit the scope of the invention as defined in the appended claims.

[0019] Example 1 shows that the dissociation constant of the Pep3S peptide of the invention is significantly lower than that of the prior art Pep3 peptide (see Table 1 A in Example 1). In addition, Table IB in Example 1 shows that the predicted affinities of the peptides of the invention alternative to Pep3S (the values of which were obtained by in silico screening using the Autodock- Vina software) are even significantly lower than the affinity of Pep3S, therefore these alternative peptides represent a further improvement.

[0020] Example 2 shows that Pep3S is more effective than Pep3 at reducing cell viability and increasing cell death in 2D and 3D cultures of cancer cells in a p53-dependent manner. Specifically, Pep3S has been shown to reduce the growth of tumor xenografts in vivo (Example 2).

[0021] At the molecular level, Pep3S has been shown to be capable of inducing stronger oxidative stress than Pep3 and inducing a specific subset of targets downstream of p53, some of which are unique for Pep3S, more efficiently than Pep3 (Example 3). Moreover, Pep3S exhibited activation of a specific pathway mediated by alteration of cell cycle-dependent cyclins, enhancing cell death. A comparative analysis of p53 transcriptional pathways showed divergent activation of p53 targets by Pep3S compared to Nutlin-3 (the best-known p53 activating molecule), confirming that the MDM2 / MDM4 interaction region constitutes a different pathway for wild-type p53 reactivation in human tumors and envisaging a promising use of Pep3S for an alternative reactivation of wt-p53 in human tumors (Example 3).

[0022] Example 4 also demonstrates that the Pep3S peptide encapsulated in PLGA nanoparticles (PLGAPep3S NPs) reduces cell growth and causes cell death in 3D cell cultures more efficiently than naked Pep3S. PLGAPep3S are also shown to reduce tumor volume in xenografts. It is also noted that PLGA nanoparticles have a selective tropism towards the tumor microenvironment, which advantageously reduces toxicity to non-tumor cells.

[0023] Example 5 refers to in silico studies carried out by the present inventors, examining the ability of a series of small peptides, generated from KVFIA (SEQ ID NO: 1), to affect the MDM2 / MDM4 interaction. The series was generated by gradually modifying the original KVFIA peptide (SEQ ID NO: 1) based on the predicted energy obtained after the molecular docking procedure, replacing one to four amino acid residues. A list of 72 peptides was selected from all possible permutations (corresponding to 3,200,000 elements). Among them, the peptide of sequence PWFRW (SEQ ID NO:2) was shown to bind to MDM2 in a more stable manner than the original KVFIA peptide (SEQ ID NO: 1).

[0024] Example 1: Physicochemical properties of the peptides of the invention

[0025] The average effective dose of the Pep3 peptide of the prior art was high, implying high production costs and difficult administration. Therefore, the present inventors carried out dynamic simulation studies, which allowed them to find that the interaction of Pep3 with MDM2 is mainly mediated by the last 5 amino acids of the C-terminal tail, whereas the first 7 do not participate in the interaction. However, it could not be predicted whether the first 7 amino acids of Pep3 contributed in any way to the efficacy properties of the peptide, i.e., intracellular distribution and availability.

[0026] The inventors thus synthesized a short peptide composed only of the last 5 amino acids of Pep3 (“Pep3S”, SEQ ID NO: 1) (Figure 1) and assessed its binding affinity with recombinant MDM2. For this purpose, a recombinant MDM2 with a purity level higher than 90% was used to achieve a reliable affinity value. The pGEX4T-l-Flag-HsMdm2 construct, encoding for a recombinant form of the MDM2 protein fused to a GST tag, was used to transfect prokaryotic cells (E.coli BL21 strain), cultured in appropriate medium for the growth of bacteria. When appropriate, protein synthesis was induced with IPTG and supported by the addition of ZnCh. The protein was subsequently extracted by sonicating in appropriate lysis buffer. A fractional precipitation procedure with ammonium sulfate and subsequent purifications with dedicated chromatography columns resulted in the pure protein free of the GST tag which was removed by enzymatic digestion with thrombin. The purity of the protein was assessed by SDS-PAGE electrophoresis; the protein was then concentrated and quantified using a Bradford assay.

[0027] The binding of the Pep3S, Pep3 and scramble peptides, collectively referred to as “ligands”, to the recombinant form of MDM2 thus obtained was assessed through the extinction of the intrinsic fluorescence signal of the protein collected at 340 nm after excitation at 280 nm. The spectra were recorded at 25°C on a Jasco FP-8200 spectrofluorimeter equipped with a thermostat. Upon addition of each ligand concentration, five traces were collected to obtain a signal relating to the average emission. Ligand concentrations ranged from low picomolar up to 1 pM. After the addition of each ligand concentration, the system was allowed to equilibrate for a few minutes under stirring. The dissociation constant, indicated as KD, was obtained by nonlinear regression following a Langmuir isotherm model, for the formation of a 1 : 1 protein: ligand complex, as follows:

[0028] (Fmax-Fi) / (Fmax-Fmin )=[L]i / ([L]i +KD) where Fi is the fluorescence measured after the addition of the 1thconcentration of the ligand, indicated with L, Fmax is the intrinsic fluorescence of MDM2 in the absence of L, Fmin is the fluorescence of the protein considered fully complexed following the addition of 1 pM L, KD is the dissociation constant for the test ligand, and [L]i is the concentration of the ligand added in the 1thstep.

[0029] The dissociation constant of Pep3S was found to be approximately half that of Pep3, indicating that Pep3S binds with a higher affinity to MDM2 than Pep3 (see Table 1A). In addition, the dose of Pep3S necessary to inhibit 50% of cell viability (IC50) was found to be 5.9 pM (Figure 2), compared to 10 pM IC50 for Pep3 (Pellegrino M, et al. Targeting the MDM2 / MDM4 interaction interface as a promising approach for p53 reactivation therapy. Cancer Res. 2015; 75:4560-72). Table IB below also shows the predicted affinity values for the alternative peptides to Pep3S falling within the scope of the invention (affinity values expressed in Kcal / mol), which were even better compared to Pep3S.

[0030] Table 1A. KD values for the interaction of Pep3, Pep3S, or control peptides with MDM2

[0031] Pep3S* Pep3* scramble

[0032] KD 0.152±0.007 0.245±0.029 0.447±0.028

[0033] *Mean of two independent experiments ± SD (Pep3S vs Pep3, p=0.047)

[0034] Table IB. Predicted affinities

[0035] Predicted affinity

[0036] Peptide Note Improvement over Pep3 S

[0037] (Kcal / mole)

[0038] KVFIA (Pep3S) -5.3 original

[0039] PWFRW -7.4 >30 5oPWhomFRW -7.7 >60

[0040] 5oPWFRW -7.9 >80

[0041] 5oPFFRW -7.8 >70

[0042] Given the high hydrophobicity of Pep3S, its ability to penetrate 3D tumor spheroids was tested and Pep3S was found to be capable of efficiently penetrating spheroids, persisting in this model for up to 4 days (Figure 3).

[0043] Example 2. In vitro and in vivo tumor suppressive activity

[0044] Based on the binding and distribution data, the inventors compared the tumor suppressive activity of Pep3S with that of Pep3 through in vitro and in vivo studies.

[0045] For 2D in vitro studies, equal numbers of human cells derived from colorectal cancer cell lines, HCT116 and LoVo, were plated and then the individual plates were treated every other day with Pep3 or Pep3 S peptide and compared with cells treated with the respective scramble peptides, i.e., Scramble3S (VAIKF, SEQ ID NO:3) or Scramble3 (VIFVIKAKEIQL, SEQ ID NO:4), and with untreated cells (Untreated) or cells treated with the same concentration of the solvent alone in which the various peptides are dissolved (DMSO).

[0046] For 3D in vitro studies, 2xl03cells were resuspended in culture medium in the presence of methyl cellulose (1 mg / ml) and inserted into “ultra low attachment” (ULA) plates. After a brief centrifugation of the plate, the cells were allowed to form spheroids over the next 48 hours and were then treated as described for the 2D cultures.

[0047] Both Pep3S and Pep3 reduced HCT116 and LoVo cell growth compared to controls, in both 2D and 3D cell models, but Pep3S did so significantly more effectively than Pep3 (Figures 4 A and 4B).

[0048] An additional assay based on caspase-3 / 7 activity was used to assess the extent of cell death in 2D and 3D models. In this case, a fluorescent caspase substrate (Apo-ONE Homogeneous caspase 3 / 7 assay) was added to the cells treated as described above for 4 days, and a fluorescent reading was performed 1 hour later. The results showed that, following treatment of cells with Pep3S, caspase activity was strongly increased, compared to cells treated with Pep3 and with the control scramble peptide (Figures 4C, D). Antagonism by the caspase inhibitor ZVAD (Figures 4C, D) confirmed that this cell death is due to increased cell apoptosis.

[0049] In vivo, Pep3S suppresses tumor growth, as indicated by the results obtained with subcutaneous HCT116 xenografts treated with Pep3S or Pep3 and compared with the results observed on the contralateral flank treated with DMSO (Figures 5A, 5B).

[0050] In this experiment, both flanks of NU / NU strain mice, with reduced immune response, were inoculated with 3X106HCT116 cells subcutaneously. The cells had previously been engineered to stably express the luciferase gene and resuspended in matrigel. After 10 days or in any case after the tumor reached a defined level of luminescence (measured as ROI), the mice were treated every other day on one flank with Pep3 or Pep3S (injected on the tumor surface), whereas the contralateral flank was treated with the solvent alone (DMSO) (Figure 5A).

[0051] Tumor growth was monitored through the luminescence signal emitted by the cells.

[0052] The diagram in Figure 5B shows luciferase activity levels measured twice a week. The data were referred to the ROI level at the starting point of the treatment set to 1. N=5.

[0053] Pep3S was found to maintain its specificity towards p53, as demonstrated by the lack of activity in HCT116 p53' / _(Figure 6), which confirms its indication against tumors characterized by the presence of wild type p53.

[0054] Example 3. Molecular mechanism

[0055] Pep3 had previously demonstrated the ability to induce p53 transcriptional function.

[0056] To test whether Pep3S maintains the same specificity, the inventors compared the transcriptional targets in Pep3 S-treated cells and Pep3-treated cells, using a p53-transcription array (Life Technologies). Cells treated for 48 hours as indicated were lysed and their RNA was extracted. After quantification, the RNAs were transcribed into cDNAs and inserted into the different wells of the array. The expression of the genes of interest was normalized using B2M and actin housekeepers, for HCT116 cells and LoVo cells, respectively.

[0057] Pep3S induced a reduced number of p53 targets compared to Pep3, which indicates that Pep3S activity is at least partially different and suggests greater specificity (Figure 7A). The two peptides share the regulation of 14 targets (Figure 7A). Most of these targets (12 / 14) were down-regulated by both peptides relative to their levels in scramble-treated cells. For 4 of these genes, the down-regulation activity of Pep3S was stronger compared to Pep3 (Figure 7B). Of note, only Pep3S and not Pep3 was able to significantly reduce p21 levels in cell culture (Figure 7C) and in tumor samples (Figure 7D). Given the anti-apoptotic activity of p21, its reduction enhances the cell death activity of Pep3S.

[0058] As for cell response mechanisms, Pep3 increased the oxidative stress as the main mechanism for inducing cell death. Pep3S induced a significantly greater stress response than Pep3 in the two different colorectal cancer cell lines (HCT116 and LoVo), which confirms its greater effectiveness (Figure 8).

[0059] Example 4. Encapsulation in PLGA nanoparticles (PLGA NPs)

[0060] To improve the efficiency of in vivo administration of Pep3S, nanoparticle encapsulation studies were carried out.

[0061] Two different vectors were selected as possible candidates: hyaluronic acid-coated chitosan nanoparticles (CS / HA NPs) or poly(lactic-co-glycolic acid) nanoparticles (PLGA NPs).

[0062] The CS / HANPs + Pep3S formulation is performed in an aqueous buffer (10 mM phosphate at pH 6.5). Under these conditions, the amount of recovered peptide (measured using the fluorescein assay) was less than 5%, indicating that in this aqueous environment the peptide forms insoluble clusters. Encapsulation in poly(lactic-co-glycolic acid) nanoparticles (PLGA-NPs) was therefore chosen.

[0063] Due to the low solubility of Pep3 in aqueous solution, the PLGA nanoparticles were formulated by dissolving the polymer in DMSO.

[0064] Low molecular weight (MW = 7-17 kDa) or high molecular weight (MW = 54-69 kDa) PLGA was loaded with different amounts of Pep3, mixed with the surfactant Pluronic Fl 27, dialysed in PBS, and subsequently filtered to remove non-loaded and clustered peptides.

[0065] Dynamic light scattering (DLS) analysis of the 1% and 10% peptide / PLGA mass ratios showed that nanoparticles loaded with 1% peptide maintained dimensions (Z average) and polydispersity (PD) very similar to empty particles (Table 2).

[0066] In contrast, a 10% amount of peptide led to high polydispersity, indicating that the peptide interferes with nanoprecipitation and suggesting the existence of a mixture of encapsulated and clustered peptide. Therefore, the 1% mass peptide / PLGA ratio was chosen.

[0067] Table 2. Formulation and loading of PLGA NPs

[0068] Peptide

[0069] PLGA Pep3 S / PLGA Z average size

[0070] PD recovery

[0071] (MW) mass ratio (%) (diameter, nm)

[0072] High 0 53.4±1.5 6.1±5.63

[0073] High 1 49.6±1.7 7.7±4.9 74±1

[0074] High MW PLGA supported high peptide loading (89% of the original amount) with an estimated 2.21 pg / ml (3,57pM) of encapsulated peptide. Based on these data, the high MW PLGA formulation loaded with a peptide / PLGA mass ratio of 1% w / w was considered the most efficient of the tested conditions and was chosen for the in vitro and in vivo experiments.

[0075] This formulation of PLGA NPs showed no change in molecular weight over 48 hours of dialysis in PBS, neither for scramble-loaded PLGA NPs nor for Pep3S-loaded PLGA NPs, indicating their stability in solution.

[0076] The ability of the NPs to penetrate and release Pep3S into 3D tumor spheroids was tested by using the chosen formulation (high molecular weight PLGA loaded with 1% w / w Pep3 mass ratio, corresponding to 3,57pM Pep3) and compared with the same peptide without the carrier. The PLGA-NPs were diluted 1 :5 to obtain a Pep3S concentration of 0.714 pM.

[0077] Analysis of the distribution of Pep3 conjugated with the fluorescent FITC molecule carried by the PLGA NPs showed that Pep3S reached the inner spheroid region as early as after 2 days, persisting abundantly for up to 6 days after the initial treatment. Therefore, PLGA NPs promote widespread distribution and long-term availability of Pep3S, more effectively than naked Pep3S (see above description of Figure 3), especially when considering the concentration used (0.714 pM versus 10 pM of naked Pep3S).

[0078] To assess the efficacy of PLGA-Pep3S NPs on tumor cells, spheroids with different human tumor cell lines were treated with increasing doses of PLGA-Pep3S NPS and compared with PLGA-Scramble NPs or PLGA-DMSO NPs. The resulting dose-dependent response indicated that the IC50 is 100 nM, i.e., approximately 1 / 50 and 1 / 100 below the values of naked Pep3S (see above description of Figure 2) and naked Pep3, respectively (Pellegrino M, et al. Targeting the MDM2 / MDM4 interaction interface as a promising approach for p53 reactivation therapy. Cancer Res. 2015; 75:4560-72).

[0079] Comparison of the viability of spheroids confirmed the higher efficacy of PLGA-Pep3 NPs compared to naked Pep3.

[0080] The in vivo therapeutic potential of PLGA-Pep3S NPs was then assessed through a mouse model of subcutaneous xenograft of the human tumor cell line HCT116 (Figure 9). PLGA- Pep3S NPs were injected subcutaneously near the tumor region at a dose of 8.2 pg / kg body weight (compared to the dose of lOmg / kg body weight for the naked peptide). The formulation of Pep3S in PLGA nanoparticles was shown to significantly reduce tumor growth compared to empty PLGA-NPs (PLGA-DMSO) (Figure 9). No disease, poor health or death has been observed in either group at the end of the treatment, suggesting that the PLGA nanoparticles do not induce adverse effects.

[0081] Example 5. In silico studies on PWFRW (SEQ ID NO:2)

[0082] To further exploit the possibility of targeting the MDM2 / MDM4 interaction by small peptides, the inventors examined several small peptides, spanning from peptides with a single mutation with respect to the original one (KVFIA, SEQ ID NO: 1) up to those with four out of five residues being changed. The series has been generated starting from KVFIA and then gradually modifying the original peptide based on the predicted energy obtained after the molecular docking procedure. Among all the possible permutations (which would equal 3,200,000 entries), the inventors selected a group of 72 peptides.

[0083] Interestingly, this procedure led, among others, to the recognition of some general features responsible for enhancing the predicted binding energy: a Trp residue at the C-terminal position gave lower energies (virtually, higher affinity), and most strikingly, Phe resulted as a mandatory feature to achieve good binding energy as it is essential for the specific binding of MDM4 to MDM2.

[0084] The peptide with sequence PWFRW (SEQ ID NO: 2) showed a -7.4 kcal / mol score (in the Autodock- Vina run), indicating a neat increase over the original KVFIA (-5.7 kcal / mol for the best binding pose).

[0085] The inventors further investigated the PWFRW peptide by molecular dynamics (MD), followed by a procedure known as “umbrella sampling”. The inventors started from the binding pose of PWFRW to get an initial estimation of its binding position and further investigated the dynamics under fine-tuned conditions. The initial MD simulation reached a stationary phase within a few nanoseconds, which is necessary to get a starting structure for subsequent umbrella sampling, and, more importantly, showed that PWFRW did not detach from RING / MDM2, unlike KVFIA (Figure 10 A).

[0086] The inventors then used Root Mean Square Fluctuation (RMSF) to measure the average deviation of atomic positions, thus obtaining a measure of the stability of the complex peptide / MDM2. By comparing RMSF profiles obtained for alpha carbons of PWFRW and KVFIA, the inventors observed that the former induced the RING / MDM2 to acquire a less flexible and more stable structure (Figure 10B).

[0087] More in particular, Figure 10 A shows the RMSD profile for the PWFRW peptide calculated over the full simulation time (200ns). A long stationary phase is recognizable just after the first few nanoseconds of equilibrium phase. The inset reports the superposition of the same RMSD of PWFRW (blue trace) alongside the RMSD of KVFIA (red trace). PWFRW shows a stable behaviour, whereas KVFIA detaches from MDM2 after 160ns, resulting is a spike region in the latest portion of the plot.

[0088] Figure 10 B shows the superposition of the RMSF plots, residue-wise, of PWFRW (blue) and KVFIA (red). The plot shows the lower degree of fluctuations in case of PWFRW with respect to KVFIA. The inset is relative to the difference RMSF / KVFIA - RMSF / PWFRW: the more a value deviates from zero (reported as a magenta line), the more a portion of the structure subdues to a larger fluctuation.

[0089] Further analysis of the most representative structure obtained from molecular dynamics showed that the two peptides KVFIA and PWFRW bind MDM2 differently (Figure 11) and explore different regions of the MDM2 RING domain and contact some different amino acids (Figure 12, Table 3).

[0090] Particularly, the picture in Figure 11 shows the position of KVFIA (in red) and PWFRV (in cyan) peptides within the MDM2 RING domain from different angles (A, B, C) obtained by subsequent 90° rotation, assuming an arbitrary reference system.

[0091] The picture in Figure 12 shows the different binding regions of KVFIA and PWFRV peptides within the MDM2 RING domain from different angles obtained by subsequent 90° rotation (A, B, C), assuming an arbitrary reference system. The amino acids in (A) represent the major polar contact points of the two peptides as reported in Table 3. Table 3 shows that KVFIA interacts strongly with G456 and L458, whereas PWFRW interacts strongly and uniquely with P491 and L487. Since the amino acids contacted uniquely by the two peptides are quite distant from each other, the data further prove that the two peptides explore two different binding pockets (Figures 11, 12). It is interesting to note that PWFRW polar interactions are contributed by all five residues, whereas KVFIA polar interactions are exerted by only K, F, and A residues. This aspect may contribute to a tighter stability of the interaction observed from RMSF analysis.

[0092] Table 3. Polar contacts by the two peptides to the MDM2 RING domain.

[0093] The cell colors are related to the specificity of peptide interaction with MDM2: the light grey ones indicate the MDM2 amino acids exclusively contacted by KVFIA; the dark grey indicates the one exclusively contacted by PWFRV; the white indicates the amino acids contacted by both peptides.

Claims

CLAIMS1. A peptide of 5 amino acids in length selected from the group consisting of:(i) a peptide of sequence PWFRW (SEQ ID NO:2) and variants of the peptide of sequence SEQ ID NO:2 in which the proline residue (P) at position 1 is replaced with a 5-oxoproline residue (5oP), and optionally the tryptophan residue (W) at position 2 is replaced with a phenylalanine residue (F) or the phenylalanine residue (F) at position 3 is replaced with a homophenylalanine residue (homF);(ii) a peptide of sequence KVFIA (SEQ ID NO: 1).

2. The peptide according to claim 1, which is encapsulated in a poly(lactic-co-glycolic acid) (PLGA) nanoparticle.

3. The peptide according to claim 2, wherein PLGA is high molecular weight PLGA.

4. The peptide according to claim 3, wherein the peptide / PLGA mass ratio is 1% w / w.

5. The peptide according to any one of the preceding claims, wherein the amino acid sequence of the peptide is KVFIA (SEQ ID NO: 1).

6. An isolated nucleic acid encoding for a peptide according to any one of the preceding claims.

7. An expression vector comprising a nucleic acid encoding for a peptide according to any one of claims 1 to 5.

8. A pharmaceutical composition including at least one peptide, or nucleic acid, or expression vector according to any one of claims 1 to 7.

9. The pharmaceutical composition, peptide, nucleic acid or expression vector according to any one of claims 1 to 8, for use in the therapeutic treatment of a tumor.

10. The pharmaceutical composition, peptide, nucleic acid or expression vector for use according to claim 9, wherein the tumor is a solid tumor.

11. The pharmaceutical composition, peptide, nucleic acid or expression vector for use according to claim 9 or 10, wherein the tumor expresses wild type p53.

Images