Structure activity relationship sar studies

structure activity relationship sar studies

Structure-Activity Relationships (SAR) studies of benzoxazinones, their degradation products and analogues. phytotoxicity on standard target species ( STS). Structure-activity relationships (SAR) explore the relationship between The results of structure-activity relationship (SAR) studies can provide. Structure-activity relationships (SAR) are the traditional practices of medicinal chemistry which try to modify the effect or the potency (i.e. activity) of bioactive.

The key is to identify aspects of structure pertaining to the rate-determining, molecular-triggering event in the mechanism of action for the chemical and biological actions of interest.

Hence, the mechanism of action is a guiding concept in determining both the groupings of chemicals suitable for study and the molecular descriptors potentially most relevant to activity. Ultimately, it is the linkage of SAR to mechanism that enables a scientific rationale to be constructed to account for activity variations in existing chemicals.

structure activity relationship sar studies

This, in turn, provides the most sound scientific basis for predicting the activity of new and untested chemicals. Having stated the ideal case, we are faced with the reality that many toxicity endpoints are complex, often poorly understood and characterized, and not resolvable to the level of a common mechanism of action.

To the extent that we can resolve the toxicity problem, SARs may be capable of global discrimination among different mechanisms, e. A good illustration of the former is provided by a recent report of enhanced MultiCASE models, developed in collaboration with the FDA, for predicting rodent carcinogenicity potential for pharmaceutical databases Matthews and Contrera, The classifications are based primarily on MultiCASE-identified structural alerting features of molecules, and are potentially applicable to rough screening of a wide diversity of chemical structures and mechanisms of carcinogenicity.

In contrast, a prominent example of a mechanism-based SAR application that has impacted risk assessment is the modeling of Ah receptor-binding capability of dioxin-like compounds, including the structurally related polychlorinated dioxins, dibenzofurans, and biphenyls, in particular.

Extensive studies by Safe and coworkersand others, demonstrating rank order correlations between Ah-receptor binding and various measures of response, both in vivo and in vitro, were effective in establishing a common mechanism of action for these toxic responses.

structure activity relationship sar studies

This, in turn, led to the development and use of toxic equivalency factors TEFs to arrive at concentrations of dioxin equivalents TEQs in human and ecological risk assessments involving exposures to complex mixtures of these compounds as they occur in real-world environments Van den Berg et al. In deriving TEF values, a variety of available data, including from in vivo, in vitro, and QSAR studies are usually weighed in using a tiered approach.

This overall approach is basically a relative potency-ranking scheme in which the relative potency of each chemical is expressed as some fraction of the potency of 2,3,7,8-tetrachlorodibenzo-p-dioxin TCDD. Other ways of representing molecules may extend beyond those based on 2D structure, atoms and bonds, to those based on 3D structure, steric and electrostatic fields.

The latter are most appropriate if a receptor-mediated mechanism is known or suspected. Finally, appropriate methods of analysis are needed for relating the activities and chemical structures of interest, which will depend on the nature of the activity measure e. The goal is to strive at every step in the process to consider what is chemically and biologically plausible, to reasonably constrain the problem in these terms, and to derive models that have a strong scientific rationale and basis for interpretation.

SAR Models An SAR model is defined and limited by the nature and quality of the data used in model development and is strictly applicable only in relation to the data set that was used to generate it, but that possibly has predictive capability within some reasonable boundary outside that data set.

In evaluating an SAR model, it is important to define boundaries of application, by considering what sorts of molecules, and range of descriptor values, have activities that can be confidently predicted, and statistical measures of fit, significance, and robustness. Models can also lead to mechanistic hypotheses that guide future testing and validation. A process for model validation should test predictive capability, as well as explore the boundaries for model application and challenge the mechanistic hypotheses suggested by a well-constructed model.

SAR models are useful in research for purposes beyond prediction. They can offer rationalization of activity variations in existing data, argue for a common mechanism of activity and additivity of effect for a series of chemicals Richard and Hunter III,identify outliers due to either experimental error or alternative mechanisms Lipnick,narrow a dose range-finding experiment by using a predicted dose as a first estimateserve as a metric for comparison of different biological endpoints Hansch et al.

The ideal SAR model should consider sufficient numbers of molecules for adequate statistical representation, have a broad range of quantitative activities orders of magnitude or adequate distribution of molecules in each activity class active and inactiveand yield to mechanistic interpretation Hermens, In toxicology modeling problems, this ideal is rarely encountered.

For many toxicity endpoints of interest, diverse chemical structures, lack of knowledge of mechanisms, and large data gaps are more frequently the norm.

Because SAR ultimately draws its validity from linkage to mechanism, however, any success achieved with these methods rests on the degree to which the global models are able to discern and adequately represent the mechanism-based SAR components of the larger data set Lewis, ; Richard, ; Wagner et al. Prediction Systems Two main types of commercial toxicity prediction systems are currently available: The biggest drawback of such systems is the ease with which a prediction is generated versus the need for careful scrutiny of the results.

The rule-based systems typically are more limited in their application than the more correlative type approaches, but they may offer greater chemical and biological interpretableness for the chemicals they do predict. These methods attempt to identify spatially-localized features across a series of molecules that correlate with activity, and represent requirements for ligand binding and complementarity to a postulated receptor binding site Green and Marshall, ; Marshall and Cramer, III, The modeled compound, and not its metabolites or other transformation products, is responsible for the biological effect.

The proposed or modeled conformation is the bioactive one. All compounds are binding in the same way to the same site. The biological activity is largely explained by enthalpic processes steric, electrostatic, hydrogen bonding, etc. Entropic terms are similar for all compounds. The system is at equilibrium. Common solvent effects—diffusion, transport, etc.

Although enjoying much more extensive use in the area of drug design, the process of 3D-QSAR specifically as applied in comparative molecular field analysis CoMFA will be described here in the context of its limited applications in the area of toxicology prediction. In CoMFA, non-covalent ligand-receptor interactions are represented by steric Lennard-Jones and electrostatic Coulombic interactions with the ligand. The steric and electrostatic interactions of probe atoms with the ligand are calculated at uniform grid points, then tabulated for each molecule row in the series.

The resulting matrix is analyzed with multi variate statistics partial least squares or PLSyielding an equation that relates the CoMFA field value to the activity. This process also highlights those features of the putative receptor that are being probed by the structure-activity data set. In general, the objective of this and other related 3D-QSAR procedures is to place molecules with common alignments in a 3D grid or regioncalculate interaction values for each grid point, and place the values for each point in a QSAR table.

Then create an equation, based on PLS regression, to describe the relationship between the values and the reported activities, verify the predictive ability of the QSAR by cross-validation and determine the optimal number of componentsvisualize the final QSAR model by plotting coefficients in the corresponding regions of space, and use the final QSAR equation to estimate the biological activity for other new compounds not included in the model.

Requirements for successful development of a 3D-QSAR model include selecting appropriate compounds and biological data to serve as the training set and identifying a useful and meaningful alignment of the molecules for study. A general guideline is that at least 20 compounds are required to derive a QSAR, although useful QSARs have been obtained with as few as 7 compounds in the model. The quality and choice of biological data to be modeled is critical to successful development of a model.

The range and distribution of biological data are also important, with a normal distribution of data across as wide a range of activities as possible minimum of 3 log units.

The initial challenge is to choose structural conformers as close to the actual bioactive conformers as possible. It is important to realize that the choice of modeling technique can influence the to what extent and in how much detail an SAR can be explored. For example, statistical QSAR approaches bsed on 2D descriptors that ignore stereochemistry can be miss key elements of an SAR that depend on chirality [ 8 ].

However, it generally remains true that 3D approaches are preferable when a crystal structure is available and when a few chemical series are being explored. Much of traditional QSAR is based on statistical models that link chemical structure characterized by numerical descriptors to biological activities. In either case, one develops a model based on a training set of molecules. The model can then be used to predict the activity for new molecules. While much of QSAR has focused on various forms of linear regression ranging from ordinary least squares to more robust methods such as PLS or ridge regressionthere is no reason to assume, a priori, the structure-activity relationship is linear.

Indeed, for nost biological systems it is unreasonable to expect linear relationships simply because of the multiple, complex processes occurring in vivo. Thus, modern non-linear methods such as neural networks and support vector machines have seen extensive use and tend to exhibit high accuracy.

However, building a predictive model is just the first step. For certain scenarios, such as virtual screening, one can apply the model and simply obtain numerical predictions of activity.

However, the focus of this article is to consider the use of such predictive models for exploring SAR. Key to such exploration is the ability to interpret the model and understand how exactly it correlates activity to specific structural features [ 1011 ].

For interpretive purposes, a model should be understandable - both in terms of the descriptors used and the underlying model itself. The predictive ability of the model is not primary though of course, for statistical models, they should be statistically significant.

Examples include linear regression and random forests. Those studies were conducted by Broka inCorey in, andMcCombie in andBuszek inSchmalz inKociensk inand Harroweven in complete information on this topic can be found cited and widely commented in Ref. However, those efforts have provided information on improvement of their biological activity, pharmacophore and mechanism of action.

structure activity relationship sar studies

Moreover, it is worth noting that semi-synthetic alkoxy or phenoxy substitution such as ether and acetate derivatives of pseudopterosins are under patent protection [ 21 ]. At this point, we want to mention the recent studies reported in [ 40 ] where simplified synthetic analogs of pseudopterosins 78—87 were prepared by Fenical and colleagues using a new and efficient synthesis taking into account the following general structural modifications: Nine of the 10 compounds evaluated as racemic mixtures were active in the mouse-ear assay the most active one was twice more active than PsA and no statistical differences were identified among compounds.

Additionally, the synthetic route involving only six steps leads to derivatives without substitutents at C-1 and C-3 reducing the number of stereoisomers and allows for the preparation of multigram amounts of them.

Muricea austera Specimens of Muricea austera were collected in the Pacific coast of Panama during an expedition of the Smithsonian Tropical Research Institute [ 49 ]. The MeOH extract of M. Bioassay-guided fractionation using vacuum liquid chromatography followed by flash chromatography and normal-phase HPLC purification yielded six compounds: The structures of the compounds were determined based on their spectroscopic data.

Several synthetic analogs were obtained under basic hydrolysis and perbenzoylation reactions. All natural compounds and synthetic analogs were evaluated against a drug-resistant Plasmodium falciparum and intracellular form of Trypanosoma cruzi.

The antiplasmodial activity of analogs with stereochemistry as d -arabinopyranose 96 and 97 and d -galactosides 98 and 99 were also evaluated. The octocoral genus Paragorgia has been barely studied; however, some diterpenoids and steroids were reported in [ 50 ]. InSpanish researchers collected Paragorgia sp. The sample was extracted with isopropanol, and a bioguided isolation procedure allowed to isolate three novel cytotoxic steroids derivatives named parathiosteroids A-C — The structures incorporate an A-ring with different degrees of unsaturation, and a side chain containing both a thioester and an acetamide groups.

These structural novelties do not have precedents in marine natural products chemistry [ 51 ]. Related compounds were detected in an aerobic degradation study of bile acid cholate by a Pseudomonas sp. The synthesis includes oxidation of C hydroxymethylene to carboxylic acid followed by thioesterification of the carboxylic acid with N-acetylcysteamine.

The unsaturation pattern of A-ring at was obtained by Birch reduction of followed by bromination at C-2 and subsequent dehydrohalogenation. In this way, more than 20 steroids were prepared. Colonies of Lobophytum sp.

On Exploring Structure Activity Relationships

All compounds were evaluated for cell growth inhibitory activity against three different cell lines: H9c2 cardiacmyoblastsC6 glioma and HeLa epithelial carcinoma. This fact indicated that the growth inhibitory activity of should be attributed to the presence of the hydroperoxy group in this molecule.

Based on the availability of high amounts of decaryiolit was subjected to several reactions acetylations, oxidations and epoxidations to obtain six semi-synthetic derivatives with the purpose of extending the structure-activity relationship knowledge.

These results allowed to establish that minor structural changes on the cembranoid skeleton of decaryiol can radically affect the activity and selectivity as cell growth inhibitors.

Sarcophyton glaucum Sarcophine is a bioactive cembranoid diterpenes with anticancer activity isolated by Kashman group in [ 5455 ] from the Red Sea soft coral Sarcophyton glaucum. Continued studies of structure-activity relationship as mentioned in Hassan et al. In addition, later experiments confirmed the importance of macrocyclic double bonds to the mentioned activity, and showed that bromination of sarcophine improved the antiproliferative activity against malignant breast cancer cells.

Thus far, the reported studies clearly demonstrate the need to further optimize the epoxide functionality of sarcophine in relation to its anticancer activity. The analogs thus prepared were subjected to evaluation of their ability to inhibit the proliferation and migration of the human metastatic prostate cancer PC-3 and breast cancer MDA-MB cell lines using MTT and wound healing assays.

Most analogs exhibited enhance antimigration activity and lack of cytotoxicity toward the cancer cells.

Structure–activity relationship - Wikipedia

Sinularia lochmodes An interesting example related to the topic of this chapter is the one published by Tanaka et al. This study mentions lectin SLL-2 isolated from octocoral Sinularia lochmodes as an important mediator in the symbiotic relationship of this animal with its zooxanthellae the symbiotic microalgae Symbiodinium on which the coral depends for energy and nutrients.

V3 SAR of Penicillins Essential features

This lectin SLL-2 influences the transformation of Symbiodinium cells into a non-flagellated coccoid form from a flagellated-swimming form.