Bioinfoqsar

Results
Romana

Project budget

Total cost of the project (all financial resources are included): 1.173.523 lei
The contract financial cost (national budget):1.108.887 lei
Co-financial resources: 64.636 lei

PHASE I – Year 2007

Specific objectives of the project first phase

1. Defining the research methodology, documentation, virtual screening of some large database in order to select the proper molecules and structure characterization of the target molecular classes; initiation of a personal database with antitumoural agents used nowadays in therapy.
2. Preliminary in vitro toxicity studies of nucleosides analogues.
3. Study and analyses of the software applications request; designing a QSAR application.

PHASE II –Year 2008

Specific objectives of the project second phase

1. Designing a specialized QSAR software
2. Developing QSAR models for the antitumoural activity of some conventional antitumour drugs
3. In vitro testing of the cellular model activity as well as the influence of the enzymatic substrates. .

Abstract of The Phase II

During this phase, the project aimed the developing of QSAR software using a series of standard agents with anticancer activity especially folic acid analogues (e.g. Metotrexat) and also some pyrimidine derivatives of 5-Fluorouracil (5-FU) class. It is important to mention that the in silico pharmacodinamic profile of a new obtained 5-FU derivate, Ent- Fluorouracil, demonstrated a potential anticancer activity similar to 5-FU. The QSAR models were proved to be predictive in case of some in vitro toxicological properties evaluated using cell cultures.

In this phase of the project, it was realised an in vitro study regarding the effects of some inhibitors of nucleotide biosynthesis on normal and tumour cells. For this purpose, Metotrexat (an inhibitor of dihydrofolate reductase and of tymidilate synthase), the active metabolite A771726 of leflunomide (an inhibitor of dihydroorate dehydrogenase) and also 5-fluorouracil were investigated in order to establish the effects of these antimetabolites on cellular viability/multiplication as well as on the cellular cycle.

Modelarea proceselor biologice induse de 5-FU asupra celulelor leucemice Jurkat si U937 reprezinta baza de studiu a unui nou analog nucleozidic bazat pe 5-FU, ent-27-5FU, sintetizat de INCD Chimico-Farmaceutice (colectivul Dr. Tanase Constantin), cu potentiale efecte anti-canceroase.

The computer modelling of the biological processes induced by 5-FU on Jurkat and U937 leukaemic cells was the basic study implemented also for a new nucleoside analogue derived from 5-FU (ent-27-%FU). This product has potential anticancer properties and it was synthesised at National Institute for Chemical and Pharmaceutical Research and Development – ICCF, Bucharest by PhD Tanase Constantin et colab. In order to demonstrate the interferences between some anticancer agents and the activity of some specific enzymes, it was realised an in vitro study aiming the effects of some biosynthesis nucleotide inhibitors on the responses of the tumour cells. The effect of metotrexate on glutathione-peroxidase which is a component of antioxidant protection system of U937 cells used for this study.

PHASE III - Year 2009 (15.09.2009)

Specific objects of the phase III

1. 1.Obtaining a specialized QSAR software
2. 1.Designing and testing of some robust and predictive QSAR/QPSR models for the antirheumatic effects of some antirheumatic agents with antitumoural component activity (Cytostatic Drug Modified Antirheumatic Drugs – DMARDs).
3. 1.In vitro evaluation based on cellular model of the activity of the standard therapeutical agents with antirheumatic effect.
4. 1.In vitro testing of the influence on enzymatic substrates.

Abstract of The Phase III

A. During the second phase of the project, it was structured a QSAR software for some standard agents with cytostatic DMARDs activity. The QSAR analyses were developed in order to find the molecular-structural profile throughout quantical descriptors and the QSAR/QSPR connection of the pharmacological activity prediction and the discrimination of the antirheumatic activity correlated with antitumour effect (cytostatic DMARDs).

The following studies were performed:

a). The molecular topology was used to correlate the antirheumatic activity with the antitumoural activity in case of some therapeutically agents:
- 1. Metotrexat, classic antifolate analogues and derivatives of dehydrofolate dehydrogenase (pharmacological mechanism of action)
- 2. Leflunomid (its active metabolite – Teriflunomid) and derivatives of dihydro–orat dehydrogenase which are specific immunosuppressive in the arthritis rheumatoid treatment.
b). The space of the molecular descriptors were analysed in order to find the proper descriptors for the DMARDs agents with heterogenic molecular structures.

The theoretical data obtained by computer modelling were explored in vitro by determining the modification of some cellular parameters after the treatment with different antitumoural/antirheumatic agents (purchase from P3 and CO).

B. The effect of some standard agents with antirheumatic activity was studied in vitro using cellular model. The specific effects of Metotrexat and active metabolite of Leflunomide were studied using normal mononuclear cell and mononuclear cells isolated from patients suffering of arthritis rheumatoid.

The following aspects were of the main interest:

- The effect of the rheumatic agents on the profile of the cytokine secreted ex vivo by the pacients monocytes activated ex vivo via CD14. The monocytes were prelevated from patients with a major disequilibrium of the lymphocytes/monocytes ratio, with an increased number of monocytes and almost normal value lymphocytes.
- The effect of some rheumatic agents on the capacity of lymphocytes to incorporate thymidine in case of ex vivo polyclonal activation with phytohemaglutinine.
- The effect of some antirheumatic agents on HUVEC endothelial cells.
- The mononuclear cells were isolated from 20 patients suffering of arthritis rheumatoid. The cells were cultivated and were treated ex vivo with PHA and in vitro with antirheumatic agents. The cell lysates were tested by partner from University of Bucharest.

C. In order to demonstrate the operationally and the application, the software realised by Partner 3 (P3) was developed as a calculation instrument, friendly to use and easy to visualise, for modelling and prediction of the tested drugs therapeutic activity.

The best models of theoretical prediction were developed in order to use the software for each of the classes of tested drugs. In this purpose, mathematical equations were created in order to correlate the therapeutic properties with structural molecular descriptors of each class of drugs. Following these analyses, the discriminatory functions for the therapeutic effects were obtained. The models were tested for robustness and their predictive capacity for initiating a digital QSAR models database with application in individualised therapies.

PHASE IV – Year 2010 (15.10.2010)

Specific objectives of the Phase IV

1. Testing the operationally and utility of the ß version BIOINFOQSAR software
2. Application in individualised therapy using the QSAR/QSPR models based on the software
3. Ex vivo evaluation of the individual variation in case of the antirheumatic activity using standard therapeutic agents
4. Ex vivo evaluation of the individualised effect obtained with antirheumatic therapeutic standard agents on some relevant enzymatic substrates
5. Organizing an workshop about “The Application of QSAR/QSPR models in individualised therapy using antitumoural/antirheumatic agents”

Abstract of Phase IV

A. During this phase of the project, the QSAR computer programme was further applied and developed using congener series of standard agents with antitumour/antirheumatic activity (immunosuppressive drugs like derivatives of Metotrexate and Leflunomid).

The following activities were performed:

a). a)QSAR software application provided by the research team of Partner 3 (ASRC, Bucharest) was used for the QSAR computer modelling of the antirheumatic and immunosuppressive activity of some therapeutic agents:
- 1. The classic antifolate analogues of Metotrexate, the dihydrofolate dehydrogenase inhibitors (the pharmacological and molecular/cellular mechanism of action)
- 2. The derivates of Leflunomid with specific immunosuppressor effect used for the treatment of arthritis rheumatoid (dihydro–orat dehydrogenase inhibitors)
b). Analyses of the molecular descriptors family space in order to find both the proper discriminatory function for the antirheumatic activity and the QSAR models most selective and robust useful in individualised therapy applications.

The relationship between chemical structure- biological activities was analysed for some antifolate drugs with antitumour and immunosuppressive effects. Metotrexate, one of these tested compounds is known as “the golden standard” used for the treatment of individualised therapies. Moreover, 74 inhibitors of dihydrofolate reductase were tested and compared to Metotrexate used as standard substance.

Predictive in silico models were obtained for Leflunomid and its derivatives with immunosuppressive effect useful for the treatment of arthritis rheumatoid.

The study aimed also checking of the BIOINFOQSAR software operation. This software is original and allows the calculation of the quantical-molecular pharmacological properties. The study results based on the interpretation of their chemical structure demonstrated that the application is useful to obtain conclusive information about the biological properties of some drugs.

The accuracy and robustness (not affected by noise factors) of the QSAR models were checked both statistically and by correlation with the data obtained from the ex vivo studies and enzymatic substrates.

B. The individual variability of some standard agents with antirheumatic activity was evaluated ex vivo by Partner 1 (INCD Victor Babes). The peripheral immunological profile of some patients was monitored during the treatment of arthritis rheumatoid.

The following parameters were noted:

- The distribution of the peripheral lymphocytes populations and sub-populations
- The oxidative activity of peripheral phagocytes (monocytes, granulocytes)
- Cytokine profiles secreted ex vivo by the peripheral monocytes after the experimental activation

C. The individualised effect of the antirheumatic standard agents obtained on enzymatic substrates was evaluated ex vivo by the Partner 2 (University of Bucharest). The following aspects were noted:

- The G6PDH activity varies during the treatment with Metotrexate and anti-TNFa. The catalytic activity is modified in time comparing to the initial moment.
- When the activity of G6PDH has a normal initial level, its evolution is the same with the immunological parameters.
- If the G6PDH initial activity value is increased, the immunological parameters will increase even in case of an enzymatic activity decrease.
- No correlation between the variation of both G6PDH and LDH enzymatic activity and the DAS28 clinical score could be observed.
- A large inter-individual variability of the G6PDH and LDH enzymatic activity was noted for the lymphocytes collected from the patients suffering of arthritis rheumatoid. The main causes of this variability are individual genetic particularities and the complexity of the cellular response activated by therapeutic agents and the disease evolution.

D. The operation and application of the new software created by research team of Partner 3 (ASCR Bucharest) were tested in order to be a friendly to use calculation instrument for modeling and predicting the therapeutic activity of the tested drugs.

Testing the software created by Partner 3 on each class of the studied drugs, it was possible to create “the best models” of theoretical prediction as mathematical equations of the quantical therapeutic/pharmacologic properties correlated with structural-molecular descriptors. After repeating the algorithm, new discriminatory functions were obtained and it was check the robustness and predictive capacity of the models. Moreover, it was initiated a Digital Database of QSAR models very useful in individualized therapy.

In the final phase of the project, it was organized the workshop entitled “Application of the QSAR/QSPR models in individualized therapies with antitumour/antirheumatic drugs”. The results of the project were presented during the workshop.

The main objective of the project was completely reached. It was created a new specialized bioinformatics system useful for in silico modelling and drugs development. QSAR models permitted the prediction of the pharmacological activity by correlating different predictive technologies with in vitro tests using tumour cell lines, normal human cell cultures and different enzymatic substrates. Based on QSAR/QSPR principles, it was developed an optimal, efficient and robust database for predicting the antirheumatic/antitumour activity in case of some conventional classes of drugs. The database was focused on the application within a new priority domain – the individualised therapies and personalised medicine.

The result indicator was entirely fulfilled. A QSAR specialized software was created. It was successfully applied for the quantical-molecular description of some antitumour/antirheumatic agents like cytostatic DMARDs. The software can be applied in almost any therapeutic domain.

The specialized software was entirely designed and developed by partner 3 – ASCR, Bucharest within this project carrying same name: BIOINFOQSAR. This software is a useful instrument for both calculation the quantical-molecular descriptors and correlation of the pharmacological/biological activity. This software was permanently improved and adapted by processing the feed-back signals from all users (the end-users are all the project partners). The software operation and application were tested on some standard agent derivatives of metotrexate and leflunomid – not all drugs were proved to be active. A ß-version of the software was designed and successfully tested on other classes of DMARDs active functional analogues. The technological platform of the bioinformatics system could be take over and applied for innovative medical field. QSAR models Digital Library was based within this project by in silico modelling methods.