Application of a systems biology approach for skin allergy risk assessment
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
AATEX 14, Special Issue, 381-388
Proc. 6th World Congress on Alternatives & Animal Use in the Life Sciences
August 21-25, 2007, Tokyo, Japan
Application of a systems biology approach for skin allergy risk assessment
Gavin Maxwell and Cameron MacKay
Unilever – Safety & Environmental Assurance Centre (SEAC)
Corresponding Author: Gavin Maxwell
Unilever – Safety & Environmental Assurance Centre (SEAC)
Colworth Science Park, Sharnbrook, Bedfordshire MK44 1LQ, UK
Phone: +(44)-1234-264888, Fax: +(44)-1234-264744, gavin.maxwell@unilever.com
Abstract
We have developed an in silico model of skin sensitization induction to characterise and quantify the
contribution of each pathway to the overall biological process; through this analysis we have developed a
rationale for the interpretation and potential integration of in vitro test data for risk assessment purposes.
The mouse local lymph node assay (LLNA) is now in widespread use for the evaluation of skin
sensitization potential. Recent changes in EU legislation [i.e. 7th Amendment to the EU Cosmetics Directive]
have made developing non-animal approaches to provide the data for skin sensitization risk assessment a key
business need. Several in vitro assays have already been developed for the prediction of skin sensitization;
however these are based on measuring a few pathways within the overall biological process and our
understanding of the relative contribution of these individual pathways to skin sensitization induction is
limited.
To address this knowledge gap and thereby guide further in vitro assay development, a "systems biology"
approach has been used to construct a computer-based mathematical model of the induction of skin
sensitization, in collaboration with Entelos Inc. Biological mechanisms underlying the induction phase of
skin sensitization are represented by non-linear ordinary differential equations, defined using qualitative and
quantitative information from a total of 496 published papers. From the model, we have identified knowledge
gaps for future investigative research, and key factors that have a major influence on the induction of skin
sensitization (e.g. TNF-α production in the epidermis); we have quantified their relative contribution to
the overall process. Through providing a biologically-relevant rationale for the interpretation and potential
integration of diverse types of in vitro data, the in silico model has helped us to define and evaluate a new
conceptual framework for skin sensitization risk assessment.
Keywords: in silico, in vitro, risk assessment, skin sensitization, systems biology
Introduction animal data. However, developing ways to integrate
Decisions about the consumer safety of our and determine the relative importance of these data
products are made on the basis of a risk assessment, to enable risk-based safety decisions to be made
in which data on the potential hazards of the represents a major challenge (Fentem et al., 2004;US
ingredients are interpreted in the context of the likely National Research Council, 2007).
exposure to the product, i.e. the concentration of the To ensure that products do not induce skin
ingredients in the product and how the product is (contact) allergy in consumers, information on the
used by consumers. Traditionally, much of the hazard concentrations of the ingredients in the product and
data on chemicals have been generated by applying how the product is used by consumers, together
technologies developed for histological and clinical with data generated in the mouse local lymph node
chemical analyses to animal models. Both the 6th and assay (OECD, 2002) is used to assess whether a
7th Amendments to the EU Cosmetics Directive have chemical ingredient has the potential to cause skin
stimulated considerable research into developing sensitisation, and thus skin allergy, in humans. The
alternative approaches to assess consumer safety (EU, current generation of in vitro predictive assays for
2003) with improved non-animal in vitro and in silico the skin sensitisation endpoint are primarily focussed
models and new technologies such as proteomics and in two key areas, both believed to be key events
bioinformatics approaches becoming increasingly in skin sensitisation induction; measurement of
available to generate and interpret new types of non- sensitizer haptenation events by detection of peptide
© 2008, Japanese Society for Alternatives to Animal Experiments 381Gavin Maxwell and Cameron MacKay Fig. 1. Skin Sensitization PhysioLab® Platform – CD4+ T cell Life Cycle example CD4+ T cell Life Cycle is shown as an example of the underlying functionality of the Skin Sensitization Physiolab® Platform. In the upper panel, the node and arc architecture can be seen, while in the lower panel a reference note (top right) and comparison chart (lower left) demonstrate how relevant literature is stored within the platform and how in silico experimental results are visualised, respectively. modifications in vitro (Gerberick et al., 2007) and initiated with Entelos® Inc. to construct a computer- measurement of Dendritic cell (DC) activation based mathematical model of the induction of skin (assessed by detecting changes in co-stimulatory sensitisation (termed Skin Sensitisation PhysioLab® receptor expression) following sensitizer treatment (SSP) platform) (MacKay et al., 2007). (Ade et al., 2005;Sakaguchi et al., 2006;Ashikaga et The aim of this study was to build a transparent, al., 2006;Sakaguchi et al., 2007;Python et al., 2007). robust, mechanistic and quantitative in silico The identified strategy moving forward is that data model that captured our current understanding of from these in vitro assays will be integrated with the biological pathways, processes and mediators other hazard information (e.g. skin bioavailability involved in skin sensitisation in vivo. The SSP predictions) to generate a new measure of sensitizer platform has been constructed such that the biology potential/potency that can be used to inform risk can be interrogated computationally in an iterative, assessment decisions, in the absence of new animal hypothesis-driven manner (Fig. 1). The publications data. Some relatively simple scoring approaches to used to build and define the biological relationships integrating these data have been considered (Jowsey are both accessible and can be removed and replaced et al., 2006) and these together with more complex with other information, if/when it becomes available. statistical approaches are currently being assessed. The model is not intended to be an exhaustive record However we were also interested to explore whether of all biological pathways that could potentially be mathematical modelling approaches (also termed implicated in skin sensitisation induction but rather systems biology) could be used to (a) generate new it represents the pathways known to be required (i.e. biological understanding; (b) guide our experimental quantitative data exists in the literature to support research programme; (c) focus our development their role). Therefore, the value of the model is in of new predictive in vitro assays; and (d) inform quantifying the accepted theory of skin sensitisation, our risk assessments. To do so, a collaboration was in order to determine the influence of each pathway 382
on the predictive endpoint (e.g. antigen-specific T cell modelled properties such as the release of epidermal
proliferation in the draining lymph node). cytokines in response to sensitizer stimulus or
the migration rate of LC in response to epidermal
Materials and Methods cytokines. The majority of the biological data used to
In Silico Modelling develop the model was derived from murine models
The software package PhysioLab Modeller (Entelos due to the relative absence of skin sensitization-
Inc.) was used to develop the SSP platform (Fig. 1). relevant datasets derived using human tissues. Data
The software enables the user to graphically construct on rate constants, dose response curves and other
models of large-scale biological systems in a manner temporal data were directly parameterised within
akin to pathway diagrams. Cellular and molecular the model. Where the required parameters were
interactions are represented quantitatively using not directly available, the values in the model were
ordinary differential equations (ODEs) the numerical manually adjusted within a physically feasible range
solution of which simulates the dynamic response of in order to quantitatively and qualitatively reproduce
the biological system. The form and parameterization the published results. For example, in Cumberbatch
of the ODEs (e.g. Michaelis-Menten, first-order, 2005, (Cumberbatch et al., 2005), the LC cell density
second-order or more complicated kinetics) are in a mouse ear was measured at 4 and 17 hours
chosen to ensure consistency of the simulation with following exposure to 1% DNCB. Additionally, the
the published theory and experimental data. In cases number of DC appearing in the LN was measured at
where multiple accepted mechanisms or hypotheses 24, 48 and 72 hours following an identical exposure.
exist, an alternate parameterisation of the model In order to ensure that the model reproduced this data,
can be defined for each mechanism, subject to the the parameters governing LC migration rates were
constraints of experimental data. The model can then adjusted until the simulation results for an exposure
be interrogated for each parameterisation in order to of 1% DNCB were in qualitative and quantitative
obtain conclusions that are robust to the uncertainty agreement with the observed experimental results. In
in the mechanism. In cases where the modelled order to simulate this experiment, it was necessary
mechanism cannot explain the experimental data, to first use other experimental data to determine
the biology must be revisited to determine what the the parameters governing the release of epidermal
likely alternate mechanism is. The modelling is thus cytokines that drive LC migration in response to a 1%
an iterative process in which the accepted theory is DNCB exposure. If when adjusting the parameters
modelled, compared to experimental data and either the chosen model could not reproduce the desired
accepted or refined. experimental results, the assumptions made about the
biology were re-visited and a new model proposed.
Model Scope In this way the various pathways of the model were
Creation of the Skin Sensitization PhysioLab® platform calibrated until the dynamics of the system-level
required a detailed literature review of approximately response to which they contribute, i.e. Ag-specific
500 publicly available publications. This process proliferation, could be simulated.
helped to determine the key elements that should be The model was evaluated by linking together each
represented in the platform and identify the appropriate of the individually calibrated pathways in order
data that described those elements. The model describes that the full system-level response be simulated.
the induction of skin sensitization over a period of 10 Among the chemical-specific information utilized for
days following chemical exposure and is composed evaluation were data for 2,4-dinitrochlorobenzene
of two main tissue compartments, the epidermal layer (DNCB), oxazolone (OX), isoeugenol, cinnamic
of skin and the local draining lymph node (dLN) (Fig. aldehyde, hexyl cinnamic aldehyde, and hydroquinone
1). The epidermal compartment of the model contains (chemical sensitizers), as well as p-aminobenzoic
keratinocytes, Langerhans' cells (LC), mast cells and T acid and glycerol (non-sensitizers). Greatest
cells while the lymph node compartment is composed of emphasis was placed on DNCB and OX data, as the
DC, T cells (CD4+ and CD8+) and B cells. In addition majority of chemical information available for both
to modelling the biological events occurring during the calibration and evaluation was restricted to various
induction of skin sensitization, a number of modules have concentrations of these two chemicals. Only unknown
been included to perform miscellaneous calculations. chemical-specific parameters were adjusted during
Generally, the function of these calculations is to enable the evaluation process, while underlying biological
experimentally observed properties to be compared with parameters defined during the calibration process
those modelled. remained unchanged.
Use of experimental data Sensitivity Analysis
To ensure model consistency with published data, Sensitivity analysis was used to determine the
a calibration process was performed at the individual key pathways driving the sensitization response.
pathway level. Published data was gathered on The analysis was performed by modulating various
383Gavin Maxwell and Cameron MacKay
pathways in the model and comparing simulation
results for selected outputs. In these simulations a
standard protocol was implemented for sensitizer
exposure, parameters associated with the selected
pathways were modulated around their baseline
values and the relative impact on the selected outputs
was assessed. The exposure protocol employed
was that of 3 consecutive exposures at 0, 24 and 48
hours [i.e. that of the murine local lymph node assay
(LLNA)] to a sensitizer of a prescribed potency class
(OECD, 2002).
A key aspect of sensitivity analysis was the
selection of outputs to be analyzed. The maximum
rate of Ag-specific CD4+ and CD8+ T cell expansion
was chosen as the primary output for analysis as it
was decided that this was the model output giving
the best measure of the extent of skin sensitization
induction. Maximum Ag-specific T cell expansion
was recorded by measuring the peak level of antigen-
specific T cell proliferation observed over a ten-day
course following exposure. Additionally, the LLNA
SI (OECD, 2002) was also used as secondary output.
Using chemical specific LLNA data model settings
were defined for a prototypical weak, moderate and
strong sensitizer. A reference dose was then selected Fig. 2. Implementation of 'recruitment hypothesis' within Skin
for the sensitivity analysis performed on each Sensitization PhysioLab® Platform
prototypical chemical. For the prototypical weak, A) Despite adjusting the parameters for maximum T-cell
moderate and strong sensitizers, reference doses expansion, the SSP platform cannot reproduce the observed
of 1%, 1% and 0.1% were chosen respectively and fold-increase in cellularity for 0.25% DNCB. While the absolute
corresponded to a response at the low, moderate and number of proliferating CD4+ and CD8+ T cells is in agreement
with observed data, the percentage of proliferating cells exceeds
high end of the dose response. Thus for the weak
that seen in the same sources (inset), suggesting the need for
sensitizer the reference dose was such that pathways factors that increase the non-dividing cell population to be
amplifying the response would be detected, for the included. [* % proliferating cells reflects the CD62LloCD44hi
moderate sensitizer pathways either amplifying percent +/- SEM cited in (Gerberick et al., 1999b), ** Fold
or dampening the response would be detected and increase compared to naïve control, ***Cellularity data from
for the strong sensitizer pathways dampening the (Sikorski et al., 1996;Gerberick et al., 1999b;Van Och et al.,
response would be detected. The pathway sensitivity 2000;Suda et al., 2002;Soderberg et al., 2005;Goutet et al.,
index was defined as the range (in logs) between 2005)
the maximal and minimal normalized outcome B) With enhanced recruitment implemented in the SSP platform,
model predictions (solid bars) are consistent with in vivo data
value at that reference dose. Thus, for example, a for fold increase in cellularity, Ag-specific proliferation (inset),
sensitivity score of 1 indicates a 10-fold difference and total cellularity in the dLN after stimulation with 0.25%
between the outcome value observed when the DNCB and other chemical sensitizers reported in the literature.
pathway modulation was set to the lowest value [****OX data from (Dearman et al., 1999;Gerberick et al.,
and the outcome value observed when the pathway 1999b), DNCB data from (Gerberick et al., 1999b;Van Och et
modulation was set to the highest value. al., 2000;Suda et al., 2002;Goutet et al., 2005), HCA data from
(Dearman et al., 1999;Gerberick et al., 1999b;Gerberick et al.,
Results 2002), and ISO data from (Dearman et al., 1999;Gerberick et
al., 1999b).]
Model development required extensive quantitative
data analysis at the level of individual pathways as that the calibrated and evaluated Skin Sensitization
well as whole-scale system dynamics. Calibration PhysioLab® platform reproduced many of the key
experiments were performed to ensure that the key experimental benchmarks observed in several studies.
biological mechanisms and dynamics of individual Data on dLN cellularity and composition as well as
pathways were consistent with published data. Several Ag-specific cell proliferation were used to create
evaluation experiments were performed to ensure reasonable ranges that defined an acceptable response.
that the individually calibrated pathways, when For example, given the standard application of 0.25%
acting together in a system, reproduced the dynamics of the sensitizing chemical DNCB in the LLNA,
observed experimentally without any changes to the predicted dLN cell response and the dLN proliferative
biological representation (see Methods). Results show
384responses were in line with those reported in several based solely on the number of immigrant LCs in
studies. The platform was similarly able to capture the dLN. Therefore, we sought evidence for other
the cellular and proliferative responses to other doses immune processes that might result in significant
of DNCB as well as other chemicals. Furthermore, increases in cellularity. While not usually considered a
the platform ultimately reproduced the key dynamics primary driver of the skin sensitization response, data
observable with the LLNA and also allowed in other immunological contexts (e.g. inflammation
exploration of underlying mechanisms (often difficult and virus infection) supported the phenomenon of
to measure in vivo) giving rise to the observed increased non-specific recruitment of cells to the
phenomena. dLN following infection (Martin-Fontecha et al.,
2003;Soderberg et al., 2005). These findings resulted
Novel insights from model building in an exploration of alternate mechanisms to explain
Due to the quantitative nature of modelling the increase in dLN cellularity during sensitization.
experimental data, it is often possible to recognize
inconsistencies that would otherwise have gone Significance of T cell recruitment
unnoticed. During model building and calibration, The platform captures dynamics of increased
it was found that the expansion of Ag-specific T recruitment through enhanced influx of naïve
cells alone was not sufficient to reproduce the dLN lymphocytes in response to maturing LCs in the dLN.
cellularity reported in the published literature. By including the dynamics of enhanced recruitment in
Experiments investigating application of a 0.25% the platform, we were able to reproduce key aspects
dose of DNCB in the LLNA revealed that the of LLNA data (Fig. 2). To test the hypothesis that
overall cellularity of the dLN could increase by as enhanced recruitment is vital to reproduce this data,
much as 14-fold over control conditions (Dearman we created a parameterization of the model without
et al., 1999;Gerberick et al., 1999a;Van Och et al., the hypothesized enhanced recruitment mechanisms.
2000;Suda et al., 2002). Despite accurate calibration In this model version, simulation of response to a
of the dynamics associated with T-cell proliferation, standard LLNA protocol with a strong sensitizer
however, the model could not achieve such high resulted in a mismatch with data across several of
levels of dLN cellularity (Fig. 2). In order to the criteria defined for validation, particularly in
investigate this apparent discrepancy, we formulated extremely low total T cell counts in spite of high
a simpler model of proliferation in the dLN in which percentages of CD4+ and CD8+ T cell proliferation.
we considered the number of Ag-specific T cells No adjustments in other parameters were able
able to proliferate. The number of naïve Ag-specific to increase total T cells numbers or decrease the
clones (i.e., the number of different T cell clones number of proliferating cells to acceptable (in range)
able to recognize Ag generated by exposure to the values. Therefore, our conclusion was that increased
chemical), the clonal frequency (i.e., the frequency recruitment was necessary to explain existing data on
with which each T cell clone is likely to occur in the LN cellularity and cellular proliferation.
T cell population) and the number of cell divisions
were set to established values (Gudmundsdottir et Sensitivity analysis identified key pathways driving
al., 1999;Kehren et al., 1999;Jelley-Gibbs et al., sensitization
2000;Kaech and Ahmed, 2001;Vocanson et al., As a first step in determining the relative
2005). However, even in the scenario of no cellular contribution of individual pathways on skin
efflux from the dLN, our simple model showed that sensitization induction we conducted a sensitivity
the experimentally observed cellularity could not be analysis on the model. Sensitivity analysis is a
achieved within the 4 to 10 cell divisions reported method of assessing key drivers of an overall
in the literature (Gudmundsdottir et al., 1999;Jelley- response such as sensitization. Parameters associated
Gibbs et al., 2000;Kaech et al., 2001). Achievement with these pathways are selected for analysis and are
of the observed cellularity requires biologically modulated around their baseline values to assess the
infeasible values of naïve Ag-specific clones or clonal resulting impact on selected outputs. As described
frequency. Interestingly, experimental evidence in the Methods, performing a sensitivity analysis
suggested that increased cellularity in an excised dLN enabled determination of the relative contribution of
could also be observed upon treatment with sodium specific pathways to the overall response. The output
lauryl sulfate (SLS) (a non-sensitizer) (Gerberick of primary interest for the assessment of sensitizer
et al., 1999a;Lee et al., 2002). Although SLS is potency was the ability to elicit CD4+ and CD8+ Ag-
recognised not to be a skin sensitizer, increased specific T cell proliferation. Therefore we measured
migration of LCs to the dLN has been observed in the impact of several pathways on this output variable
mice following SLS treatment (Cumberbatch et al., and explored sensitivities for a prototypical weak,
1993). However, the increase in dLN cellularity moderate (data not shown) and strong sensitizer (Fig.
following SLS treatment cannot be accounted for 3).
385Gavin Maxwell and Cameron MacKay Fig. 3. Sensitivity of Maximum Ag-specific proliferation and Local Lymph Node Assay Stimulation Index to key pathways Sensitivity is defined as the range (in logs) between maximal and minimal value of output attained through modulation of each pathway. Maximum Ag-specific proliferation (panels a & b) and LLNA SI (panels c & d) were selected as the outputs for the sensitivity analysis. Results for two prototypical chemicals are shown: a weak sensitizer (panels a & c) and strong sensitizer (panels b & d) are shown. Pathway category sensitivities are computed by taking an average of the associated individual pathway sensitivity. Antigen-specific proliferation. An analysis was to the same biological pathways were also assessed. performed to assess the relative contribution of Using LLNA SI as the output variable, we found a functional pathways in various biological categories significant difference in the relative contribution of on maximum Ag-specific proliferation. Contribution the pathway categories between weak and strong was measured by identifying the most sensitive prototypic chemicals (Fig. 3c and 3d). With prototypic pathways, i.e. those that, when modulated, resulted strong sensitizers all pathway categories appear in the greatest change in increasing or decreasing to contribute in a comparative fashion to LLNA maximal Ag-specific proliferation. Pathways were SI (Fig. 3, panel d). However, this profile changes grouped into generalised categories and the range of substantially when prototypic weak sensitizers are modulation for each pathway was chosen to represent analysed, with 'Epidermal activation and irritation' a reasonable operating range for that pathway. and 'Non-specific response' pathways constituting the Surprisingly, pathways other than those associated majority of influence upon LLNA SI (Fig. 3, panel c). with 'Antigen presentation' were among the top Consequently it can be summarised that the relative drivers of Ag-specific proliferation (Fig. 3a and 3b). contribution of the pathway categories and therefore Indeed all pathway categories were found to be strong the individual pathways to the LLNA SI is likely to drivers of Ag-specific T-cell proliferation. Within vary with sensitizer potency. the 'Epidermal activation and irritation pathway' category, TNF-α proved to have the most significant Discussion contribution (data not shown). These results suggest The SSP platform, was developed to (a) generate epidermal cytokine production and its effects on new biological understanding; (b) guide our LC maturation and migration are key steps in the experimental research programme; (c) focus our induction of sensitization. Analyzing the Ag-specific development of new predictive in vitro assays; and (d) output for both weak and strong sensitizers did not inform our risk assessments. From our experiences to alter the conclusions. date we believe this new approach has applications Local Lymph node assay Stimulation Index in each of these areas and that it will continue to (LLNA SI). The sensitivities of the LLNA SI output generate new biological insights through combining 386
previously disparate data sets and/or drawing parallels tool that can be iteratively improved as new data
from other disease processes. The implementation of become available.
the recruitment hypothesis within the SSP to 'balance' In addition to fulfilling its role as a research
dLN cell dynamics is a good example of where tool, the SSP can also be used in order to assist in
pathogen infection literature was used to identify determining and refining the key information required
and explain a previously underappreciated pathway in a weight-of-evidence approach to consumer
within skin sensitisation induction (Fig. 2). Clearly, safety risk assessment. Jowsey et al. (Jowsey et
this hypothesis will now need to be confirmed al., 2006) published an initial concept of a weight-
experimentally before any firm conclusions can of-evidence approach to integrate non-animal data
be drawn and other unidentified, uncharacterised inputs as surrogates for several key processes known
sensitizer-specific effects may also be involved; to be important mechanistically in the induction of
however it remains debatable that this question would skin sensitisation. Whilst recognising that the list
even have been posed without first modelling the of assays or data requirements is not definitive, this
available dLN cell data. Consequently, there exists a conceptual approach is a pragmatic starting point
belief that further biological insights will be yielded given the limited availability of new types of data. As
through iterative modelling of experimental data more data are generated, statistical and mathematical
moving forward, as has been shown elsewhere (Lewis approaches will be used to model and interpret
et al., 2001;Rullman et al., 2005). Therefore it is data in a more robust way and it is here where the
our intention to incorporate data generated from our understanding gained from in silico modelling can
future research and the scientific literature into the be applied. Using the SSP sensitivity analysis results
model, where possible. in order to propose the key data required for an
With respect to guiding future in vitro assay integrated weight-of-evidence approach will allow
development, the sensitivity analysis of the SSP the retention of a biological rationale in deriving a
platform increased our understanding and confirmed new measure of sensitizer potential/potency. Indeed,
previously discussed hypotheses in equal measure. it may be that once all the key pieces of information
For example, the identified role for both antigen- have been determined, modelling approaches such as
specific and antigen non-specific pathways in the one described here, may prove themselves to be a
driving sensitizer-induced proliferation in the dLN suitable method of dispassionately integrating diverse
was already appreciated within the field (Kimber types of increasingly complex, multi-parameter in
and Dearman, 2002). However, the dissection of vitro datasets.
the relative contribution of key pathways to LLNA
SI (or antigen-specific T cell proliferation) and Acknowledgements
the assessment of how these pathways vary with The research is part of Unilever's ongoing effort to
sensitizer potency (using prototypic chemicals) develop novel ways of delivering consumer safety.
represents a unique analysis that is potentially The SSP platform was developed in collaboration
invaluable for guiding in vitro model development. with Entelos Inc., US.
For example, the identified strong influence of
epidermal inflammatory signals upon sensitizer- References
induced DC activation highlights that there may 1. Ade N, Teissier S, Pallardy M, and Rousset F. Contact
be value in investigating the incorporation of Sensitizers Induce Apoptosis and CD86 Expression on
key inflammatory mediators within existing DC U937 cells in an independent manner. 2005. Ref Type:
Conference Proceeding.
activation models. While the apparent importance of 2. Ashikaga T, Yoshida Y, Hirota M, Yoneyama K, Itagaki H,
a recruitment effect for inducing a draining lymph Sakaguchi H, Miyazawa M, Ito Y, Suzuki H, and Toyoda
node proliferation may suggest that traditional DC: T H (2006) Development of an in vitro skin sensitization test
cell co-culture protocols e.g. (Krasteva et al., 1996) using human cell lines: the human Cell Line Activation
need to be viewed in context of the proportion of the Test (h-CLAT). I. Optimization of the h-CLAT protocol.
immune repertoire they can effectively scan. However Toxicol In Vitro, 20, 767-773.
3. Cumberbatch M, Clelland K, Dearman RJ, and Kimber I
these predictions should be interpreted in the context
(2005) Impact of cutaneous IL-10 on resident epidermal
of the underlying models and technologies that were Langerhans' cells and the development of polarized
used to derive the data. For example good quality, immune responses. Journal of Immunology, 175, 43-50.
quantitative, sensitizer-specific epidermal cytokine 4. Cumberbatch M, Scott RC, Basketter DA, Scholes EW,
release data was difficult to identify and the majority Hilton J, Dearman RJ, and Kimber I (1993) Influence
of the data used to develop the SSP platform was of Sodium Lauryl Sulfate on 2,4-Dinitrochlorobenzene-
Induced Lymph-Node Activation. Toxicology, 77, 181-191.
derived from murine models; consequently our future
5. Dearman RJ, Hilton J, Basketter DA, and Kimber I (1999)
efforts will focus on filling the identified data gaps Cytokine endpoints for the local lymph node assay:
with quantitative data generated using human models Consideration of interferon-gamma and interleukin 12.
and in this way the platform represents an evolving Journal of Applied Toxicology, 19, 149-155.
387Gavin Maxwell and Cameron MacKay
6. EU (2003) Directive 2003/15/EC of the European Parliament 21. Lewis AK, Paterson T, Leong CC, Defranoux N, Holgate
and of the Council of 27 February 2003 amending Council ST, and Stokes CL (2001) The Roles of Cells and
Directive 76/768/EEC on the approximation of the laws of Mediators in a Computer Model of Chronic Asthma.
the Member States relating to cosmetic products. Official International Archives of Allergy Immunology, 124,
Journal of the European Union, L66, 26-35. 282-286.
7. Fentem J, Chamberlain M, and Sangster B (2004) The 22. MacKay C, Bajaria S, Shaver G, Kudrycki K, Ramanujan
feasibility of replacing animal testing for assessing S, Paterson T, Friedrich C, Maxwell G, Jowsey I, Lockley
consumer safety: a suggested future direction. Altern Lab D, Reynolds F, and Fentem J (2007) In silico modelling of
Anim, 32, 617-623. skin sensitisation. Toxicology, 231, 103.
8. Gerberick GF, Cruse LW, Miller CM, and Ridder GM (1999a) 23. Martin-Fontecha A, Sebastiani S, Hopken UE, Uguccioni M,
Selective modulation of B-cell activation markers CD86 Lipp M, Lanzavecchia A, and Sallusto F (2003) Regulation
and I- A(k) on murine draining lymph node cells following of dendritic cell migration to the draining lymph node:
allergen or irritant treatment. Toxicology and Applied impact on T lymphocyte traffic and priming. J Exp Med,
Pharmacology, 159, 142-151. 198, 615-621.
9. Gerberick GF, Cruse LW, and Ryan CA (1999b) Local 24. OECD (2002) OECD Guideline for the Testing of
lymph node assay: Differentiating allergic and irritant Chemicals, No. 429. Skin Sensitisation: Local Lymph Node
responses using flow cytometry. Methods-A Companion to Assay, 7pp, OECD, Paris.
Methods in Enzymology, 19, 48-55. 25. Python F, Goebel C, and Aeby P (2007) Assessment of
10. Gerberick GF, Cruse LW, Ryan CA, Hulette BC, Chaney the U937 cell line for the detection of contact allergens.
JG, Skinner RA, Dearman RJ, and Kimber I (2002) Use of Toxicol Appl Pharmacol, 220, 113-124.
a B cell marker (B220) to discriminate between allergens 26. Rullman JAC, Struemper H, Defranoux NA, Ramanujan
and irritants in the local lymph node assay. Toxicological S, Meeuwisse CML, and van Elesen A (2005) Systems
Sciences, 68, 420-428. biology for battling rheumatoid arthritis: application of the
11. Gerberick GF, Vassallo JD, Foertsch LM, Price BB, Chaney Entelos Physiolab. IEE Proceeding Systems Biology, 152,
JG, and Lepoittevin JP (2007) Quantification of chemical 256-262.
peptide reactivity for screening contact allergens: A 27. Sakaguchi H, Ashikaga T, Miyazawa M, Yoshida Y, Ito Y,
classification tree lmodel approach. Toxicological Sciences, Yoneyama K, Hirota M, Itagaki H, Toyoda H, and Suzuki
97, 417-427. H (2006) Development of an in vitro skin sensitization test
12. Goutet M, Pepin E, Langonne I, Huguet N, and Ban using human cell lines; human Cell Line Activation Test
M (2005) Identification of contact and respiratory (h-CLAT). II. An inter-laboratory study of the h-CLAT.
sensitizers using flow cytometry. Toxicology and Applied Toxicol In Vitro, 20, 774-784.
Pharmacology, 205, 259-270. 28. Sakaguchi H, Miyazawa M, Yoshida Y, Ito Y, and Suzuki
13. Gudmundsdottir H, Wells AD, and Turka LA (1999) H (2007) Prediction of preservative sensitization potential
Dynamics and requirements of T cell clonal expansion in using surface marker CD86 and/or CD54 expression
vivo at the single-cell level: effector function is linked to on human cell line, THP-1. Archives of Dermatological
proliferative capacity. J Immunol, 162, 5212-5223. Research, 298, 427-437.
14. Jelley-Gibbs DM, Lepak NM, Yen M, and Swain SL (2000) 29. Sikorski EE, Gerberick GF, Ryan CA, Miller CM, and
Two distinct stages in the transition from naive CD4 T cells Ridder GM (1996) Phenotypic analysis of lymphocyte
to effectors, early antigen-dependent and late cytokine- subpopulations in lymph nodes draining the ear following
driven expansion and differentiation. J Immunol, 165, exposure to contact allergens and irritants. Fundamental
5017-5026. and Applied Toxicology, 34, 25-35.
15. Jowsey IR, Basketter DA, Westmoreland C, and Kimber 30. Soderberg KA, Payne GW, Sato A, Medzhitov R, Segal SS,
I (2006) A future approach to measuring relative skin and Iwasaki A (2005) Innate control of adaptive immunity
sensitising potency: a proposal. J Appl Toxicol, 26, via remodeling of lymph node feed arteriole. Proceedings
341-350. of the National Academy of Sciences of the United States of
16. Kaech SM and Ahmed R (2001) Memory CD8+ T America, 102, 16315-16320.
cell differentiation: initial antigen encounter triggers a 31. Suda A, Yamashita M, M, Taguchi K, Vohr HW, Tsutsui
developmental program in naive cells. Nat Immunol, 2, N, Suzuki R, Kikuchi K, Sakaguchi K, Mochizuki K,
415-422. and Nakamura K (2002) Local lymph node assay with
17. Kehren J, Desvignes C, Krasteva M, Ducluzeau MT, non-radioisotope alternative endpoints. J Toxicol Sci, 27,
Assossou O, Horand F, Hahne M, Kagi D, Kaiserlian 205-218.
D, and Nicolas JF (1999) Cytotoxicity is mandatory for 32. US National Research Council (2007) Toxicity Testing
CD8(+) T cell-mediated contact hypersensitivity. J Exp in the Twenty-first Century: A Vision and a Strategy.
Med, 189, 779-786. Committee on Toxicity and Assessment of Environmental
18. Kimber I and Dearman RJ (2002) Allergic contact Agents, National Research Council, Washington, DC.
dermatitis: The cellular effectors. Contact Dermatitis, 46, 33. Van Och FM, Slob W, De Jong WH, Vandebriel RJ,
1-5. and Van Loveren H (2000) A quantitative method for
19. Krasteva M, Moulon C, Peguet-Navarro J, Courtellemont assessing the sensitizing potency of low molecular weight
P, Redziniak G, and Schmitt D (1996) In vitro sensitization chemicals using a local lymph node assay: employment
of human T cells with hapten-treated Langerhans cells: a of a regression method that includes determination of the
screening test for the identification of contact allergens. uncertainty margins. Toxicology, 146, 49-59.
Curr Probl Dermatol, 25, 28-36. 34. Vocanson M, Saint-Mezard P, Tailhardat-Cluzel M,
20. Lee JK, Park JH, Park SH, Kim HS, and Oh HY (2002) A Benetiere J, Ducluzeau MT, Chavagnac C, Tedone R,
nonradioisotopic endpoint for measurement of lymph node Berard F, Kaiserlian D, and Nicolas JF (2005) CD8+T cells
cell proliferation in a murine allergic contact dermatitis are effector cells of contact dermatitis to common skin
model, using bromodeoxyuridine immunohistochemistry. allergens. Journal of Investigative Dermatology, 125, A39.
Journal of Pharmacological and Toxicological Methods,
48, 53-61.
388You can also read
