Investigation of excess toxicity across different species using in silico methods
Investigation of excess toxicity across different species using in silico methods
INTRODUCTION:
In silico techniques is an alternative to the experimental evaluation in predicting (estimating) excess toxicity as well as quantitative structure activity relationships (QSARs) without the need for animal testing. Using In silico approaches has many advantages, for example it could assist scientists in estimating the potential chemical properties of the pharmaceuticals studied, which can aid in the production of safer compounds that can then easily be degraded in in surface water.
There are also disadvantages in In silico approaches such as there being a great need for predictions of acute oral toxicity, this is not only to meet development and regulatory needs but also to aid in the rationalization of data previously obtain. A relationship between chemical structure and biological activity is defined by a useful tool known as QSARs. Pharmaceutical companies have been using QSARs for over 50 years that has lead to the production of non-toxic, safe and efficient medications that have been thoroughly checked before released in the markets. Baseline toxicity is also represented by QSAR models that are based on narcosis mechanisms. In silico approaches aim to complement in vivo and in vitro toxicity tests potentially minimize the time and cost of toxicity tests and improve safety assessments as well as toxicity prediction. In addition, in silico methods provide a faster approach in predicting toxicity compared to traditional testing. Using chemicals, which use relevant molecular physico-chemical properties to predict important treatment responses, is considered as an alternative to the experimental evaluation whereas in human toxicology great amount of effort has been placed into estimating end-points, for example skin sensitization, carcinogenicity and mutagenicity.
In order for In silico prediction of acute toxicity in different species to be successful, certain factors have to be taken into consideration such as, the route of administration and route of application of the toxicant species. Precision and accuracy of data is low despite there being many data available publically. Acute toxicity of species is very complicated in many cases and not correctly identified. Administration of a single dose to species can cause limitations relating to bio-availability. Another complication includes metabolism as this can lead to chemicals changing or remaining unchanged in the body. Acute toxicity has been estimated considerably from numerous cytotoxicity data, however successful results obtained from the data has been limited.
Organic chemicals have the potential to cause narcosis, their ability to do this sits mainly governed by their concentration, and their ability to cause more serious toxic effects. This would then mask the effect of narcosis caused on any chemical change. The categories for the prediction of toxicity of chemicals depends on the rational grouping of chemicals. With respect to severe toxicity in mammalian and other species, it is identified that a baseline effect can be recognised, this recognized baseline effect can be related with chemicals by narcosis. This is the disruption of physiological function that is brought about by reversible and non-specific hydrophobic binding of chemicals to membranes and proteins up to the stage lethality is tackled. The most fundamental molecular mechanism of acute toxicity is represented by narcosis for the largest function of non-electrolytic organic compounds. Such chemicals include, aliphatic alcohols, ketones, alkanes, halogenated benzenes and alkyl alcohols. This then aids the formation of categories based on the concept of baseline toxicity. To demonstrate this, QSARs can be resultant with hydrophobicity for toxicity data for saturated monohydric alcohols and other chemical classes that are well known to be related with a narcotic mechanism. These relationships found can then be used as an analysis to study the toxicological behaviour of other non-electrolytic organic compounds, and they also form mechanistic categories, for example if they are well modelled by QSAR they can be presumed to belong to this mechanism of action.
There has been discussions as to whether the association of narcotic acute potency and hydrophobicity is either a linear or non-linear relationship. Linear QSAR relationships will fail to show from chemicals acted by the narcosis mechanism if pseudo steady-state equilibrium partitioning is not achieved between the toxicant donor phase and the bio phase site of action. When a chemical compound is taken orally, there is a slight delay between oral administration and exertion as it needs to go through first pass metabolism. In this short amount of time the drug will partition in the lipophilic site. Therefore there is kinetic relationship between the starting dose and response rather than thermodynamic. As the value of log P increases above 2, toxicity will then exceed and this indicates a positive correlation between the two variables. The more soluble a compound is, suggests the less likely it will be toxic.
Log C-1 = a1 logP + a2 (logP)2 + b
C = concentration (molar units)
Log P = the logarithm of the octanol-water partition coefficient
a1 and a2 = regression coefficients
b = regression intercept
Evidence from previous data confirmed that non-equilibrium data seemed to have fit better to a non-linear QSAR model. In chemicals that have more specific mechanisms will have an increased toxic potency above this baseline. Excess toxicity is defined as the chemicals that possess these specific mechanisms that are associated with this narcosis level.
Various studies were conducted presented by a lot of literature using in silico approaches to predict excess toxicity of pharmaceutical compounds on mammalian and fish species. [mammalian paper] For instance in the study carried out for mammalian experiment, baseline toxicity was calculated using 80 different chemicals on the basis of experimental values for oral acute toxicity (LD50) to the rat and mouse, QSAR models were obtained to derive the baseline toxicity. Data were collected from KOWWIN 1.66 programme.
The background has not been very clearly presented. Think carefully about what you want to say at each point and complete one aspect before starting another.
It would be useful to give example QSARs for baseline toxicity (e.g. Konnemann, Austin baseline toxicity models for fish), then refer to examples of excess toxicity.
You need to demonstrate that you have read and understood a wide range of relevant references – all material must be referenced throughout.
Aim
The purpose of this investigation is to try and see for which compounds toxicity can be predicted using log-p and which compounds toxicity
calculate 1-octanol/water partition coefficient-dependent (LogP) baseline toxicity and excess toxicity of various chemical compounds to different species.
This is achieved by using in silico techniques such as Quantitative structure-activity relationships (QSARs) which is a powerful tool for the prediction of toxicity of untested chemicals.
Objectives-
- Collate data on toxicity and physical chemical properties) using literature and online resources such as web of science and PubMed.
- calculate log P using appropriate software such as EPISuite
- Investigate structural features that maybe responsible for toxicity using appropriate software such as Ochem
- Investigatterelationship between properties and activities using statistical methods such as minitab.
Method
Dataset from literature:
Dataset containing 60 organic compounds collected from different literature sources, including alcohols, esters, ketones, as well as compounds containing more than one functional group that differs from the rest. The different chemicals are known to act by narcotic mechanism of action. Furthermore, a number of journals were found using different websites such as web of science and science Direct to obtain information on pharmaceutical toxicity on different species.
Physico-chemical properties:
Different number of software such as EPISuite and minitab were used to obtain the physico-chemical properties. However, KOWINN software was used specifically to obtain data for the logarithm of the 1-octanol-water partition (logP), where possible measured logP values were verified and used in preference to calculated values.
Statistical analysis:
QSAR model was performed from the relationship between logP and inverse of logLD50 values of the saturated alcohols, ketones and esters. Linear regression within the Minitab software was used to develop the QSAR model. The reason for choosing those chemical compounds is because there are well studied classic non-polar narcotics. Measurements of logP values are easy to be found for these compounds. Solubility is the key during toxicity testing, however these compounds are polar chemicals, therefore issues around solubility are minimized. Measurement of logP values were predicted for the remaining chemicals using the QSAR model. The statistical parameters for the QSAR calculations were revised from literature and calculated from equations provided in those literatures.
Calculation of baseline toxicity and excess toxicity:
Species sensitivity:
Different organisms possess different levels of sensitivity to chemicals, for example, changes in the transport, site of action and metabolic transformation. Hence why linear regression analysis of toxicity values and different species were performed. Organisms will show their effect to
Sara – this draft version is clearly quite incomplete. Try to organise your thoughts carefully and ask if you need help.
The introduction should cover what is baseline toxicity, what is excess toxicity (any previous QSARs for these – with references). Include specific referenced information.
How may excess toxicity be observed across different species.
Methods should state what you will do.
Ensure you complete all sections e.g. Gantt chart, resources, references etc
REFRENCES:
Eriksson L, et al. Methods for Reliability and Uncertainty Assessment and for Applicability Evaluations of Classification- and Regression-Based QSARs. Environmental health prospectives. 2003;1361-1375. Available from: http://researchonline.ljmu.ac.uk/2306/ (accessed 18th November 2016)
Zhiming L, et al. Quantitative Structure Activity Relationship Models for the Antioxidant Activity of Polysaccharides. PLoS ONE. 2016; 1-22. Available from:
http://eds.b.ebscohost.com/eds/pdfviewer/pdfviewer?sid=9b928011-c028-49f7-9c24-dae3279a6531%40sessionmgr101&vid=4&hid=127 (accessed 18th November 2016)