Structure-Activity Relationships Mark Jason 

The Role of Structure-Activity Relationships and Expert Systems

In vitro human skin permeability coefficient data collected by Flynn (1990) have been analysed using multiple regression analysis.

An improved model for the prediction of permeability coefficients has been derived by the inclusion of the melting point as an independent variable in addition to the octanol-water partition coefficient (as logP) and molecular volume.

(The solubility of chemicals in water is related to logP and melting point through an algorithm; Suzuki, Journal of Computer-Aided Molecular Design 1991, 5, 149-166).

Examination of the dataset using principal components analysis confirmed that it could be divided into three distinct groups of chemicals–steroids, other pharmacologically active molecules.

A sub-set of hydrocortisone derivatives within the steroid group had measured permeability coefficients which were around 1.5 orders of magnitude greater than values predicted from the small molecular model.

By grouping together the small molecules and the steroids (excluding the hydrocortisone set)–60 molecules–a quantitative structure-activity relationship (QSAR) has been derived for their permeability coefficients which is dependent on logP, molecular volume and melting point, and which explains more than 90% of the variability in the data.

Although yet to be tested experimentally, this QSAR is expected to give accurate predictions of in vitro human skin permeability coefficients within the constraints of the dataset.

A historical database containing the results of 294 defined single substances tested in the guinea-pig maximisation test, carried out according to a single protocol, was used to derive a set of structural alerts for skin sensitisation, which have been incorporated into the expert system, DEREK.

Together with an assessment of percutaneous absorption, this system forms an integral part of a strategic approach to the identification of contact allergens.

Quantitative structure-activity relationships (QSARs) were derived for the skin corrosivity of organic acids and bases, and for the eye irritation potential of neutral organic chemicals.

The independent variables used for these analyses were selected on the basis of the putative mechanisms for skin irritation or corrosivity and for eye irritation, respectively.

Data sets were analysed using principal components analysis; plots of the first two principal components for different classifications of toxicological activity. The derived QSARs are expected to give useful predictions of skin corrosivity and eye irritancy for new or untested chemicals in these classes.

Although the development of these techniques is still at a very early stage, they are already able to play an important part in proposed strategies for the reduction of experimental animal usage.

In the long term, it should be possible to conduct safety evaluations using fewer experimental animals or no animals at all.

However, acceptance by regulatory authorities will be a key factor in realising the full benefits of the approach.

Quantitative structure-activity relationships (QSARs) have been derived by relating skin irritation and corrosivity data of neutral and electrophilic organic chemicals to their log(octanol/water partition coefficient) (logP), molecular volume, dipole moment and 1/molecular weight. Datasets were analysed using stepwise regression, discriminant and principal components analysis.

Discriminant analysis between irritant and non-irritant neutral and electrophilic organic chemicals using the above parameters, which broadly model skin permeability (logP and molecular volume), ‘reactivity’ (dipole moment) and 1/molecular weight to compensate for the fact that skin irritation/corrosivity testing is carried out using a fixed mass or volume of chemical, was found to discriminate well for only 73.1% of the dataset (67.3% cross-validated).

The poor discrimination at the irritant/non-irritant classification boundary is attributed largely to biological variability. Stepwise regression analysis of the Primary Irritation Index (PII) for the same dataset showed a poor correlation (r2=0.422; cross-validated r2=0.201) with a positive dependence on logP and dipole moment and a negative dependence on molecular volume, 1/molecular weight was not a significant variable.

While this QSAR for PII has little value as a predictive model, mainly because of the large biological variability evident in PII values, it is useful in confirming the putative model for skin irritation.

Discriminant analysis using logP, molecular volume and dipole moment, was able to dicriminate reasonably well (92.9% well classified; 92.9% cross-validated) between corrosive and non-corrosive electrophiles.

A plot of the first two principal components of the same parameters showed a clear demarcation between corrosive and non-corrosive electrophiles.

In contrast to the QSARs for skin irritation, increasing skin corrosivity was found to correlate with decreasing molecular olume, with incresing dipole moment, and with decreasing logP.

The predominant parameter in determining the skin corrosivity of electrophilic organic chemicals appears to be the molar dose at which they are tested; this arises because skin corrosivity testing is conducted using a fixed mass or volume of chemical.

A stepwise approach to the skin corrosivity/irritation classification of neutral and electrophilic organic chemicals is outlined. The derived QSARs should be useful for the prediction of the skin corrosivity potential of new or untested electrophiles. (Non-electrophilic neutral organic chemicals, as a category, do not generally appear to be corrosive).

Discrimination between some non-irritant and irritant neutral and electrophilic organic chemicals using these techniques is also possible.

The Role of Structure-Activity Relationships and Expert SystemsFor a large number of chemicals whose irritation potentials lie in a fairly broad and around the irritant/non-irritant, classifiction boundary, no firm prediction of classification is possible.

Keratinocyte Mark Jason 

Human Keratinocyte Cultures as Models of Cutaneous Esterase Activity

As the quest to develop useful in vitro test methodologies to replace rabbit eye and skin irritation tests continues, validation projects still depend on historical rabbit eye and skin irritation data as the benchmark against which to measure the performance of the in vitro assays. 

During our daily contact with sponsors, we are asked more and more about the relationship between in vitro test methods and the results of human trials, especially in the area of skin irritancy. 

To date, there is a limited amount of data available in the literature which compare the results of in vitro methodologies with those of human volunteer trials.  In this study, data obtained in a simple cytotoxicity assay, the 3T3 neutral red uptake (3T3 NRU) assay, will be compared with data from a standard three-application patch test.

A selection of 20 personal care and cosmetic products will be tested in a single-blind, within-subject comparison three-application patch tests on a panel of 30 healthy human volunteers. 

Skin irritation potential will be determined by assessing and scoring the volunteers’ reactions to the products. 

Products will be ranked according to their irritancy potential in vivo and in vitro, and the relationship between each of the EC values of the NRU assay and patch test results will be determined.

A reproducible, quantifiable assay has been developed for the measurement of esterase activity in human keratinocyte culture, using the model substrate 4-methyl umbelliferyl heptanoate (MUH) which is hydrolysed to the fluorescent metabolite 4-methyl umbelliferone (MU).

Activity was assessed in two human keratinocyte cell lines, NCTC 2544 and SVK-14, and in freshly isolated human breast keratinocytes from primary culture to passage 3. Vmaxvalue for MUH hydrolysing activity in the two cell lines showed that the less differentiated cell line NCTC 2544 (Vmax = 23.00 ± 2.84) expresses a much higher activity than SVK-14s (Vmax = 13.28 ± 1.42) which are more differentiated and able to form a cornified envelope.

Activity in the freshly isolated human breast keratinocytes decreased with time in culture in all three donors tested, which is also likely to relate to the extent of cell differentiation.

In human skin, xenobiotic esters penetrating the stratum corneum may be exposed to changing levels of hydrolysing esterases as they are absorbed across the epidermal cell layers.

The assay for MUH hydrolysis will be a useful tool for the study of esterase activity in populations of human keratinocytes in vitro.

“…Use of collagen, both native and modified, as a substrata for growth of cells from different tissues has been reported extensively [4,5,6,7,9,17,23].

In the present investigation, we have employed sources of collagen that are available in relatively pure form and have also not been extensively investigated earlier.

The serosal layer of bovine intestines and bovine Achilles tendon are very pure sources of bovine collagen and are available as byproducts of the slaughtered animals.

This study is aimed at the evaluation of wound dressings prepared from these two sources and cross-linked with agents 1,6-diisocyanatohexane (HDI) and basic chromium sulfate (BCS), for clinical use…”

Cytotoxic Mark Jason 

Cytotoxic and Morphological Effects of Phenylpropanolamine

The neutral red cytotoxicity assay was used in vitro to evaluate the potencies of phenylpropanolamine (PPA), nicotine, caffeine, some of their metabolites, and related chemicals.

The human cell types used as targets included fibroblast (HFF), melanoma (SK-Mel/27), and hepatoma (HepG2) cell lines and early passage endothelial (ENDO) cells and keratinocytes (NHEK).

For all of these cells, nicotine was more cytotoxic than cotinine, its major metabolite.

In turn, cotinine was more cytotoxic than chemically related compounds such as nicotinic acid and nicotinamide.

Nicotine, but neither cotinine, nicotinic acid, nor nicotinamide, induced cytoplasmic vacuolization in all the cell types tested. Except for the ENDO cells, caffeine and its metabolite, theophylline, showed approximately equivalent cytotoxic potencies.

However, for the ENDO cells, caffeine was more cytotoxic than theophylline.

Furthermore, the ENDO cells were 2-3 times more sensitive to caffeine and theophylline than were the other cell types.

The phenylpropanolamine induced cytoplasmic vacuolization only in the ENDO cells. Combinations of caffeine + PPA interacted synergistically in their cytotoxicity towards the HepG2 cells; a similar synergistic interaction was not noted with the ENDO cells. 

Fourteen cosmetic products reflecting a range of irritancy levels were evaluated for cutaneous irritation potential using dermal equivalent (DE) and skin equivalent (SE) kits. Our two in vitro models are presented in 12 culture inserts (Transwell, Costar) allowing air-liquid interphase culture and topical application.

Dermal equivalent includes a collagen-glycosaminoglycans-chitosan porous matrix populated by human normal fibroblasts and skin equivalent is realized by seeding keratinocytes onto the DE.

The pure, cosmetic products were applied in triplicate by topical application (10 ul) onto a surface delimited by a silicone assay ring placed onto DE and SE.

After 24 hr contact with the cosmetic products, the residual cellular viability was measured using a MTT test on treated and untreated tissues.

The purpose of this preliminary validation study was to evaluate to what extent the in vitro results can predict in vivo skin irritation.

Consequently, 11 cosmetic products with known Draize irritation classes were tested on DE.

Ten of 11 were correlated if we considered only the irritancy potential prediction (irritant or nonirritant) but considering the binary correlation with three classes of irritation (irritant, slightly irritant, or nonirritant) 8 of 11 were correlated.

Moreover, the parameters of validation were calculated. Second, three cosmetic products with known Draize primary irritation index (PDII) were tested both on DE and SE. The correlation of the in vitro MTT values to the in vivo data using a regression line was found to be r = 0.99 for DE and r = 0.99 for SE.

These preliminary results are encouraging and suggest that the two models, DE kit and SE kit, could be used as in vivo alternative methods after a complete validation study involving the testing of the different chemical classes and various cosmetic forms.

It is known that alkylating processes occur after exposure to sulfur mustard (HD). Although direct alkylation of DNA and RNA has been widely described, we hypothesize additional alkylation events of potential importance in skin injury.

We have determined by immunocytochemistry levels of interleukin-1 b (IL-1b ) in cultures of normal human epidermal keratinocytes (NHEK) after exposure to HD.

The expression of IL-1b in NHEK was found to be related to cell culture donor age. In neonatal NHEK exposed to HD (2 mM), IL-1b response is increased.

Electron paramagnetic resonance (EPR) spectroscopy was used to show the formation of EPR detectable, g=2.04, feature characteristic of iron-nitrosyl complex formation, and the generation of this induced complex by NHEK exposed to HD (1 mM for 18 hours) was blocked by Nw -Nitro-L-arginine (L-NOARG), a competitive inhibitor of nitric oxide synthase (NOS). Nitric oxide (·NO) has been implicated as the effector molecule that mediates IL-1b (Corbett et al, 1993).

Our results show the release of nitric oxide during cytokine expression, IL-1b , when keratinocytes ae exposed to HD.

The combination of the nitric oxide with the chloride (Cl- ) released from sulfur mustard ([ClCH2CH2]2S) on cyclization to the sulfonium ion may lead to the formation of nitrosyl chloride (NOCl), a known potent alkylating agent.

If NOCl is formed as a result of HD exposure, then it may play a role in the skin injury.

Skin Irritation Mark Jason 

Skin Irritation Research – Part 1

Skin barrier disruption caused by organic solvents to human cadaver dermatomed skin was evaluated using an in vitro model system.

Resultant changes in transepidermal water loss (TEWL), as measured with an evaporimeter, were recorded after topical application of either acetone, chloroform:methanol 2:1, hexane, hexane:methanol 2:3, or the control, water, for exposure times of 1, 3, 6 and 12 min.

The resultant lipid/solvent mixture was removed and analyzed for its lipid content.

The ability of the different solvents to induce changes in the skin’s barrier function was assessed by comparing pre- to post-solvent exposure TEWL.

When compared to the controls, water and unexposed skin, chloroform:methanol 2:1 caused the greatest significant increase in TEWL, followed by hexane:methanol 2:3.

Acetone and hexane showed no difference in TEWL from the controls.

Besides solvent, exposure time was a significant independent variable for predicting change in TEWL, and the interaction of the two (exposure time and solvent type together) was the strongest predictor.

Lipid analysis of the extracts revealed that all the solvents removed comparable quantities of the surface lipids (triglycerides, wax esters, squalene, cholesterol esters).

Stratum corneum lipids–ceramides, free fatty acids, and cholesterol–extracted by chloroform:methanol 2:1 induced a significantly greater change in TEWL than hexane:methanol 2:3.

Additionally, no individual lipid class extracted by either chloroform:methanol 2:1 or hexane:methanol 2:3 proved to be a significant or accurate variable for predicting change in TEWL.

This suggests that the mechanism by which topical chloroform:methanol 2:1 and hexane:methanol 2:3 exposure induce a change in TEWL involves more than pure lipid extraction.

 The principal objective of a skin irritation test is to determine the irritation potential of a substance so that a hazard assessment can be made and possible risk to humans evaluated.

The traditional test for evaluating skin irritation has been in vivo tests in the rabbit, often referred to as Draize tests.

hese have been employed for almost 50 years and are regarded as standard requirements for estimating the hazards associated with human skin exposure to test materials.

Since 1980, the validity and propriety of these tests have been increasingly questioned.

Much more attention has been given to the search for alternative test procedures in the hope that methods could be developed that would be both more humane and more predictive of human response.

RIFM wanted to investigate one of the new in vitro skin models for detecting skin irritation potential.

A key factor in effectively replacing animals with alternative methods is validation, which requires databases to correlate historical in vivo results with new in vitro results.

We had accumulated irritation data on approximately 100 fragrance materials using the procedure detailed in the EEC Directive and we felt that a study of representatives from this group of materials was a good place to start.

Skin2 was the model we investigated.

We chose MTT uptake, LDH release, and PGE2 release as endpoints to be measured.

Correlation of the results of these assays with those of the rabbit irritation test was not exact.

In vitro sensitisation assays have been proposed by various research groups and are actively being developed in our laboratories. The test described here is based on in vitro cultured dendritic-like cells (DCs) derived from human peripheral blood mononuclear cells.

These cells serve as replacement for Langerhans cells which are the most important antigen presenting cells in the skin.

After application of contact allergens to DCs, very rapid and selective increases of IL-1b mRNA expression followed by cell maturation have been reported (for example, Reutter et al. Toxicology in Vitro 11, 619-626, 1997 and Aiba et al. European Journal of Immunology 27, 3031-3038, 1997).

In the present work, the cell culture and the reverse transcriptase-polymerase chain reaction (RT-PCR) conditions have been optimized.

The expression of IL-1b mRNA has been measured 15 minutes after application of the test substance by a semi-quantitative RT-PCR procedure using some newly developed internal standards.

Maturation of the DCs observed as a time-dependent modulation of the CD86 positive cell population was then measured between 4 and 48 hours by flow cytometry.

Known irritants and sensitisers, as well as other potential sensitisers, were evaluated in this test format. Our results indicate that exposure to sensitisers but not to irritants induced reproducible increases in the CD86 positive cell population after 24 hours.

Specific and discriminating increases in IL-1b mRNA expression could also be observed in many, but not all, experiments.