Use of a B Cell Marker (B220) to Discriminate between Allergens and Irritants in the Local Lymph Node Assay

G. Frank Gerberick*,1, Lynn W. Cruse*, Cindy A. Ryan*, Ben C. Hulette*, Joel G. Chaney*, Robert A. Skinner{dagger}, Rebecca J. Dearman{dagger} and Ian Kimber{dagger}

* The Procter & Gamble Company, Miami Valley Laboratories, P.O. Box 538707, Cincinnati, Ohio 4523–8707; and {dagger} Syngenta Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, United Kingdom

Received January 9, 2002; accepted March 29, 2002


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
It has been shown that exposure of mice to contact allergens induces B cell activation in the draining lymph nodes (DLN), as seen by an increase in the percentage of B220+ or IgG/IgM+ cells. We have now examined whether the measurement of the percentage of B220+ cells could be used as an alternative or supplementary endpoint for the local lymph node assay (LLNA) to differentiate between allergenic responses and those few irritants that induce low-level proliferation in the DLN. Mice were treated on the ears, daily for 3 consecutive days, with various allergens (1-chloro-2,4-dinitrobenzene, {alpha}-hexylcinnamaldehyde, trinitrochlorobenzene, isoeugenol, and eugenol) or irritants (benzalkonium chloride, methyl salicylate, salicylic acid, and sodium lauryl sulfate). The DLN were excised 72 h following the final topical treatment, and the cells were prepared for B220 analysis using flow cytometry. The percentage of B220+ cells in lymph nodes derived from test and vehicle-treated animals was determined for 5 allergens and 4 irritants tested in multiple experiments (n = 3 to 17). As expected, the percentage of B220+ B cells was increased with each of the allergens tested, whereas irritant treatment did not cause similar increases. Moreover, the method was reproducible. For example, the strong allergen, 1-chloro-2,4-dinitrobenzene and the weak allergen, {alpha}-hexylcinnamaldehyde were identified as allergens in 17 of 17 and in 12 of 13 experiments, respectively. The percentage of B220 values for each chemical treatment (41 observations for allergens; 28 observations for irritants) versus the percentage of B220 values for the concurrent vehicle controls were plotted, and a classification tree model was developed that defined a B220 test:vehicle ratio cutoff of 1.25 for discriminating between allergens (>1.25) and irritants (<1.25). Using this B220 test:vehicle ratio of 1.25 in 93% of the 69 independent observations made, the allergens and irritants tested were identified correctly. Finally, to evaluate the performance of this model in a second independent laboratory, 3 allergens and 2 irritants were tested. Each of the allergens and irritants were classified correctly using the B220 test:vehicle ratio cutoff of 1.25. These data demonstrate that analysis of B220 expression in DLN may be useful in differentiating between allergen and irritant responses induced in chemically treated mice.

Key Words: allergens; irritants; local lymph node assay, B cells; flow cytometry.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
For the prediction of skin sensitization potential, the murine local lymph node assay (LLNA) is an alternative to the widely used guinea pig tests. For more than 10 years, this method has undergone extensive development, evaluation, and validation (reviewed in Gerberick et al., 2000Go). This predictive test measures proliferative responses in draining auricular lymph nodes during the induction phase of a contact sensitization response after epicutaneous exposure to materials (Kimber et al., 1986Go; Kimber and Basketter, 1992Go; Kimber and Weisenberger, 1989Go). This assay has been evaluated extensively for its reliability and sensitivity in predicting the contact sensitization potential of materials in both intra- and interlaboratory validation studies (Basketter et al., 1991Go; Kimber and Basketter, 1992Go; Kimber et al., 1991Go, 1995Go, 1998Go; Loveless et al., 1996Go; Scholes et al., 1992Go) and has been found to correlate well with guinea pig methods (Basketter et al., 1991Go; 1993Go; Basketter and Scholes, 1992Go; Gerberick et al., 2000Go; Kimber et al., 1990Go, 1991Go) and analyses in humans (Basketter et al., 1994Go; 2000Go; Gerberick et al., 2001Go; Ryan et al., 2000Go).

As with any toxicological test method, there are opportunities to improve assay performance. For example, the LLNA uses radiolabeled materials for assessing proliferation in the draining lymph node (DLN) that can make conduct of the assay difficult with regard to the handling of radioactive waste. This is especially relevant for certain countries where use of radioisotopes is regulated strictly. In addition, it has been noted that a limited number of irritants, shown to induce proliferation in the DLN, in theory may be regarded as false positives (Basketter et al., 1998Go; Gerberick et al., 1992Go; Montelius et al., 1994Go). It is our belief that an increased understanding of events in the DLN during the induction of contact sensitization could lead to the development of additional or supplementary endpoints for the LLNA that could improve the predictability, selectivity, and sensitivity of the assay as well as possibly providing opportunities to eliminate the requirement for radiolabeled materials. One approach taken has been to enumerate changes in the lymphocyte populations in the DLN following exposure to various allergens or to irritants not considered to display sensitizing activity.

Although contact sensitization is generally considered to be a T cell-mediated immune response, mice exposed to contact allergens show increases in B lymphocytes in the DLN, measured as a function of the percentage of B220+ or IgG/IgM+ cells (Gerberick et al., 1997Go, 1999Go; Manetz and Meade, 1999Go; Sikorski et al., 1996Go). In addition, a selective modulation of certain B cell activation markers on murine DLN cells has been demonstrated following allergen, but not irritant, treatment (Gerberick et al., 1999Go). Specifically, in mice treated with allergens, an increase in the median intensity of I-AK (MHC class II) and CD86 on B220+ or IgG/IgM+ B cells was observed when compared with mice treated with irritants or vehicles. Kraal and Twisk (1984) reported that B cells were retained specifically in the allergen-stimulated lymph nodes when compared with unstimulated peripheral DLN cells. In another study, Kuhn et al. (1995) demonstrated that the phenotype of cells in lymph nodes following exposure to the potent contact allergen oxazolone (1%) in the murine LLNA (4-day exposure) consisted of 37.8% CD4+ cells, 16.6% CD8+ cells, and 44.7% B220+ cells. More recently, Manetz and Meade (1999) reported that the B220+ DLN population becomes significantly elevated following exposure to allergens, but not to irritants. Collectively, these data suggest that an increase in the percentage of B220+ B cells in DLN is associated selectively with allergen exposure.

The purpose of the present study was to determine if B cell B220 analyses could be used to discriminate between allergens and irritants in the LLNA. We show that a selective increase in the percentage of B220+ cells occurs following treatment of mice with allergens, but not with irritants. Using data collected from multiple experiments, we developed a predictive model based on the B220 test:vehicle ratio to classify chemicals as allergens or irritants, with the tested chemicals being classified correctly in 93% of cases. This approach was found to be very robust and appeared to be transferred readily to a second independent laboratory where all of the materials tested were classified correctly. The results support the notion that measurement of the percentage of B cells in DLN may provide an alternative or supplementary endpoint for the LLNA that does not require the use of radioisotopes.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Participating laboratories.
Participating laboratories were at the Procter & Gamble Company, Miami Valley Laboratories, Cincinnati, OH (Laboratory 1) and Syngenta Central Toxicology Laboratory, Macclesfield, Cheshire, U.K. (Laboratory 2).

Animals.
Animals used by Laboratory 1 were female CBA/J mice, 6 to 8 weeks of age, purchased from Jackson Labs (Bar Harbor, ME). Animals were housed, fed, and handled in compliance with standards set forth by the U.S. Animal Welfare Act or recommendations in National Institutes of Health "Guide for the Care and Use of Laboratory Animals." All procedures performed on animals were reviewed and approved by a veterinarian and the Institutional Animal Care and Use Committee. Animals used by Laboratory 2 were female CBA/Ca mice, 6 to 8 weeks of age, purchased from Harlan SeraLab (Bicester, Oxfordshire, U.K.). The mice were housed under standard conditions and all procedures were approved by the UK Home Office and carried out in compliance with the Animals (Scientific Procedures) Act 1986 under a Home Office-granted Project License.

Test chemicals.
Test chemicals used in studies conducted by Laboratory 1 were as follows: benzalkonium chloride (BZC), 1-chloro-2,4-dinitrobenzene (DNCB), methyl salicylate (MS) and sodium lauryl sulfate (SLS) were purchased from Sigma Chemical Co. (St. Louis, MO). a-Hexylcinnamaldehyde (HCA), salicylic acid (SA), eugenol (EUG), and isoeugenol (ISOE) were purchased from Aldrich Chemical Co. (Milwaukee, WI). Trinitrochlorobenzene (TNCB) was obtained from PolySciences, Inc. (Warrington, PA). Acetone (ACE), 20% aqueous ethanol (EtOH), or acetone:olive oil (4:1, AOO) were used as vehicle controls. For materials formulated in acetone, the concentrations were as follows: DNCB, 0.25%; HCA, 50%; EUG, 25%; TNCB, 0.5%; ISOE, 5%; BZC, 2% and SA, 40%. SLS was used at 20% in 20% aqueous ethanol and MS was formulated at 10% in AOO. Laboratory 2 purchased DNCB and SLS from the Sigma Chemical Company and BZC and HCA were obtained from Aldrich Chemical Co. (Gillingham, Kent, U.K.). EUG used by Laboratory 2 was purchased from Quest Laboratories (Ashford, Kent, U.K.). Formulation of test materials was consistent between laboratories in terms of both concentration and vehicle used, and, in all instances, dosing solutions were prepared volumetrically and immediately prior to each treatment.

In vivo treatment.
Mice were treated using the exposure regimen of the standard LLNA as described previously (Gerberick et al., 2000Go; Kimber et al., 1995Go; Loveless et al., 1996Go). Briefly, groups of mice (n = 3–5) were treated by topical application on the dorsum of both ears with 25 µl of test material or with vehicle alone. Treatments were performed daily for 3 consecutive days. Approximately 72 h after the final treatment, the animals were euthanized, and the bilateral draining auricular lymph nodes were excised and pooled for each treatment group.

Lymph node single-cell suspensions and staining.
Single-cell suspensions were prepared from the pooled lymph nodes for each treatment group by gentle disruption through a 100-micron nylon mesh filter (The Spectra Co., Los Angeles, CA) or stainless steel mesh. The cells were collected and washed once in phosphate-buffered saline without calcium chloride or magnesium chloride (PBS) and resuspended at a concentration of 107 cells/ml in FACS buffer (PBS supplemented with 10 mM HEPES [Sigma], 0.01% sodium azide [Sigma], and 1% heat-inactivated fetal bovine serum [Gibco BRL], Laboratory 1; or PBS containing 2% fetal calf serum, Laboratory 2). Cell numbers were determined using either an automated cell counter (Laboratory 1; Coulter ZM, Coulter Electronics, Inc., Miami Lakes, FL) or manual counting of viable cells by trypan blue exclusion (Laboratory 2). Cells were stained in a 96-well round-bottom plate (106 cells/well) with optimal (typically 1 µg per 106 cells) concentrations of fluorescein isothiocyanate (FITC) conjugated rat monoclonal antimouse CD45R/B220 (clone RA3-6B2, IgG2a{kappa}, Pharmingen, San Diego, CA) and phycoerythrin (PE) conjugated hamster monoclonal antimouse CD3e (clone 145–2c11, IgG, Pharmingen). Cells were incubated for 30 min on ice, washed, and resuspended in FACS buffer to a concentration of 106 cells/ml for analysis by flow cytometry. FITC-labeled rat IgG2a{kappa} and PE-labeled hamster IgG (clone R35-38 and clone UC8-4B3, Pharmingen) controls were used.

Flow cytometric analysis.
The fluorescence intensities of the various surface markers were measured using a Coulter Elite ESPTM flow cytometer (Beckman Coulter Inc., Fullerton, CA). Listmode data were analyzed using EXPO2TM software (Beckman Coulter). Instrument alignment was validated using Flow-CheckTM Fluorospheres (Beckman Coulter). Fluorescence intensities were calibrated by adjusting gains and photomultiplier voltages to meet target values, using ImmunoBriteTM Fluorospheres (Beckman Coulter) for the FITC and PE channels. Fluorescence compensation was adjusted using samples stained with a single fluorescent-labeled antibody. Listmode data were first gated to exclude dead cells by forward-angle light scatter and 90° light scatter based on propidium iodide dye exclusion (data not shown). Dual parameter dot plots (CD3e-PE versus CD45R/B220-FITC) were produced for at least 10,000 live cells and percentage of B220+ cells was determined relative to matched isotype controls using quad stats.

Statistical analyses.
For each allergen and irritant, descriptive statistics (mean and standard deviation) were calculated for total cells per node and percentage of B220+ cells by individual treatment. These descriptive statistics were also obtained for the corresponding vehicle control cells per node and percentage of B220+ cells (by treatment) and for the test:vehicle ratio (by treatment). Statistical comparisons were made between total cells/node for chemical-treated groups and total cells/node for vehicle-treated groups, using 2-sample t-tests. When an F-test for equality of variances (Steel and Torrie, 1980Go) indicated that the chemical group and vehicle group variances significantly differ (p < 0.01), an approximate t-test was conducted (Satterthwaite, 1946Go). Similar tests were conducted in order to compare the percentage of B220+ cells in DLN derived from chemical with vehicle-treated groups. Differences between chemical- and vehicle-treated groups at the level of p < 0.01 were declared statistically significant.

In order to distinguish allergens from irritants, classification tree models (Breiman et al, 1983Go) were developed and assessed using the recursive partitioning routines implemented in S-Plus 2000 software (Insightful Corp., Seattle, WA).

To compare the accuracy of the classifications made using the prediction model, based on B220 test:vehicle ratio to the predicted human classification, a 2 x 2 contingency table was used. This procedure is a recommended means for assessing data from validation studies (Holzhütter et al., 1996Go).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cellular analysis of lymphocytes in the DLN.
In a series of multiple experiments conducted over a 5-year period in Laboratory 1, mice were treated with allergens (DNCB, TNCB, ISOE, EUG, HCA), nonsensitizing irritants (SLS, BZC, MS, SA), or their corresponding vehicles. The concentrations selected for the allergens used were based on levels that have been shown to give a positive response in a standard LLNA (Gerberick et al., 1992Go; Loveless et al., 1996Go). The irritant concentrations used were based on concentrations tested previously in Laboratory 1 and which, in some instances, have been reported to be associated with a positive LLNA response (Gerberick et al., 1992Go; Loveless et al., 1996Go). Although isotope incorporation has been shown to be a more sensitive indicator of lymph node cell proliferation compared with assessment of total lymph node cellularity (Gerberick et al., 1992Go), in these experiments, DLN cellularity was measured to avoid the use of radioactive materials. In each separate experiment, the cell number and % B220+ cells per node were assessed for the treated and vehicle control groups. The mean values (± SD) listed for each test material, as shown in Table 1Go, represent the results of 3 or more independent experiments. The mean test:vehicle ratio for cells per node for each of the allergens and irritants tested is shown in Table 2Go. Historically, in Laboratory 1, the number of cells isolated per node for naïve mice has been 1.6 x 106 ± 0.2, which is similar to levels observed following vehicle treatment (data not shown). The number of cells isolated per node following treatment with allergen (TNCB, DNCB, EUG, and HCA) was significantly increased (p < 0.01) over values obtained with the relevant concurrent vehicle-treated controls. While some increases in DLN cellularity were observed following exposure to ISOE compared with the concurrent vehicle-treated control, these changes did not reach statistical significance. The observed mean test:vehicle cell number ratios for all of the allergens were greater than 2-fold with values ranging from 2.7 for ISOE to 7.7 and 11.0 for the strong allergens DNCB and TNCB, respectively. While treatment with BZC resulted in a mean test:vehicle ratio of 3.8, treatment with the other irritants, SA, SLS, or MS stimulated more modest increases in the number of cells/node with mean test:vehicle ratios of 1.9, 1.4, and 1.5, respectively. Indeed, exposure only to BZC and MS induced statistically significant increases in DLN cell numbers.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Total DLN Cells Per Node and % B220+ Cell Values following Allergen and Irritant Treatment (Laboratory 1)
 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Classification of Allergens and Irritants Using the Test:Vehicle B220 Ratio (Laboratory 1)
 
The cellular composition of DLN in naïve mice has been reported previously to be approximately 79.0% CD3+ T cells and 10.1% B220+ B cells (Sikorski et al., 1996Go). Studies from Laboratory 1 (Gerberick et al., 1997Go; 1999Go; Sikorski et al., 1996Go) have demonstrated that allergen treatment causes an increase in the percentage of B220+ B cells. Consistent with those previous reports, the present investigations also demonstrated an increase in the % B220+ cells found in DLN of allergen-treated mice (Tables 1 and 2GoGo). The increase was significant (p < 0.01) for each of the allergens tested with the exception of ISOE, where the increase was approximately 2-fold greater than its vehicle control, but was not statistically significant. Approximately 30% of DLN cells were B220+ following exposure to the potent contact allergens TNCB and DNCB, and more than 17% B220+ cells were recorded for the other allergens tested (ISOE, EUG and HCA), compared with 9.4 to 12.9% B220+ cells in DLN isolated from vehicle-treated mice. In contrast to allergen treatment, irritant treatment (BZC, SLS, MS, SA) did not cause significant changes in the percentage of B220+ B cells when compared with their matched vehicle controls (ranging from 1.0 to 1.4 fold increase; Table 2Go). Thus, the increase in percentage of B cells (B220+) was apparently selective for allergen treatment.

Classification of Allergens and Irritants Using B220 Marker Analysis.
Classification tree models were developed for a number of B and T cell markers and cells/node in order to ascertain which parameter(s) would serve as the best discriminator(s) for classifying allergens and irritants. In addition to % B220+ and cell number per node, several other B and T cell markers were examined in developing the model including CD3, CD4, CD8 and CD86 as well as two subclasses of CD4+ and CD8+ cells, CD44l°CD62Lhi and CD44hiCD62Ll° (Gerberick et al., 1997Go; 1999Go). All parameters were examined together and, based on model fitting, B220 proved to be the best predictor (data not shown). A B220 test:vehicle ratio of 1.25 was determined as the optimal cutoff value for discriminating between allergens and irritants. On this basis, if a calculated test:vehicle B220 ratio was greater than 1.25 the chemical was classified as an allergen whereas if the ratio was less than 1.25 the material was classified as an irritant. The % B220 values for each chemical treatment were plotted against the % B220 values for the matched vehicle controls in order to illustrate classification ability; a total of 41 observations (data points) for the 5 allergens tested and 28 observations for the 4 irritants tested (Fig. 1Go).



View larger version (15K):
[in this window]
[in a new window]
 
FIG. 1. Scatter plot illustrating differences between the percentage of B220 test:vehicle ratios from allergen-treated (squares) and irritant-treated (triangles) mice.

 
The mean test:vehicle ratios for percentage B220+ cells for each of the allergens and irritants tested in Laboratory 1 are presented in Table 2Go. The mean test:vehicle B220 ratios for the allergens were each above the 1.25 classification cutoff. On an individual experiment basis, the majority of allergens were correctly classified using the 1.25 classification cutoff. In fact, the only allergen to be misclassified in a single experiment was HCA. Moreover, the reproducibility of the B220 test:vehicle ratio was excellent, as evidenced by the fact that correct classifications were obtained for the strong allergen DNCB in 17 of 17 experiments and for the weak allergen, HCA, in 12 of 13 experiments. For 3 of 4 irritants (BZC, SLS, MS), the mean test:vehicle B220 ratio was below 1.25. The mean test:vehicle B220 ratio for SA (1.4) was above the classification cutoff. On an individual experiment level, the proportion of irritants correctly classified was high, with only 4 misclassifications from the 28 independent experiments conducted with the irritants. As with the allergens, the reproducibility of the B220 test:vehicle ratio for the nonsensitizing irritants tested was excellent.

A 2 x 2 contingency table was used to compare the predicted human sensitization classification with the predictions obtained using the B220 classification model for the 9 chemicals tested in Laboratory 1 in a total of 69 individual experiments (Table 3Go). The accuracy or closeness of agreement between the B220 test result and the predicted human classification was found to be 93%. Sensitivity and specificity values of 98 and 86% were calculated and reflect the proportion of all correctly identified positive and negative materials, respectively. The percentage of chemicals predicted to be allergens using the B220 ratio (ratio > 1.25) that are positive in humans (positive predictivity) was 91%, and the percentage of B220 negatives (classified as nonallergens; ratio < 1.25) that are human negatives (negative predictivity) was 96%.


View this table:
[in this window]
[in a new window]
 
TABLE 3 Comparison of Predicted Human Classification Response with B220 Test:Vehicle Ratio Response
 
Finally, to determine the ability of the model to transfer successfully to another laboratory, a select number of 3 allergens (DNCB, EUG, HCA) and 2 irritants (BZC, SLS) were tested in Laboratory 2, using the same treatment protocol and concentration of test materials. Each chemical was tested in a single experiment with the exception of DNCB, which was tested in 5 separate experiments. The mean cells/node and % B220 values (±SD) for DNCB are shown in Table 4Go along with the results of single experiments for the other chemicals. The total number of cells isolated per node from DNCB-treated mice was significantly increased compared with concurrent vehicle controls. For the other allergens and irritants, there was also an increase in the number of cells per node. Interestingly, the total cells per node values observed in Laboratory 2 (Table 4Go) were consistently higher for both the chemical- and vehicle-treated mice than the values observed in our initial studies in Laboratory 1 (Table 1Go). However, when the cell number ratios were determined, the values were very similar in both Laboratories (see Tables 2 and 5GoGo). Consistent with previous results from Laboratory 1, allergen treatment caused an increase in the percentage of B220+ B cells in Laboratory 2 (Table 4Go). The increase was significant for the DNCB-treated animals and greater than a 2-fold increase was observed with the other allergens evaluated (EUG,HCA). In contrast to allergen treatment, little or no increase was observed with irritant treatment (BZC, SLS). Most importantly, the test:vehicle B220 ratio for each of the allergens tested was above the 1.25 classification cutoff, while the ratios for both of the irritants were below the cutoff (Table 5Go). Thus, in Laboratory 2, all of the allergens and irritants tested were classified correctly using the B220 test:vehicle ratio cutoff of 1.25.


View this table:
[in this window]
[in a new window]
 
TABLE 4 DLN Cells Per Node and % B220 Values following Allergen and Irritant Treatment (Laboratory 2)
 

View this table:
[in this window]
[in a new window]
 
TABLE 5 Classification of Allergens and Irritants Using B220 Test:Vehicle Ratio (Laboratory 2)
 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
It is known that contact allergens induce a number of characteristic changes within draining lymph nodes, including a marked increase in the absolute number of CD4+ and CD8+ T cells (Sikorski et al., 1996Go) and a selective increase in the percentage of T cells expressing the T cell activation/memory phenotype CD62Ll°CD44hi (Gerberick et al., 1997Go). Although allergic contact dermatitis is generally thought of as a T cell-mediated immune response, we have shown that mice exposed to allergens, but not to irritants, show a selective and reproducible increase in the percentage of B220+ B cells within DLN (Gerberick et al., 1997Go, 1999Go; Sikorski et al., 1996Go). The increase in percentage of B220+ B cells was seen 24 h after the final application of material, plateaued by 48 h, and was still elevated by 96 h (Sikorski et al., 1996Go). Moreover, in allergen-treated mice, percentages of B220+ cells increased in a dose-dependent manner. Other investigators have reported a preferential increase in the percentage or proliferation of B220+ cells in murine DLN after allergen treatment (Homey et al., 1998Go; Kuhn et al., 1995Go; Manetz and Meade, 1999Go). The investigations described here were designed to explore further the utility of the measurement of % B220+ cells as an alternative or supplementary endpoint for the LLNA in differentiating between allergen and irritant responses. To that end, we have measured the B220 response induced by a total of 9 chemicals tested multiple times and in two independent laboratories.

There have been few studies examining the role of B cells in contact sensitization. Kraal and Twisk (1984) reported B cells were retained specifically in allergen-stimulated nodes when they compared unstimulated peripheral DLN cells to cells that had been primed with the contact allergen oxazolone. Kuhn et al. (1995) demonstrated an increase in B220+ cells (44.7%) in DLN cells isolated from oxazolone-treated mice. These two studies using oxazolone support our findings that B cells accumulate in DLN following topical application of contact allergens. Speculation regarding the cellular mechanisms underlying the preferential accumulation of B lymphocytes in the DLN following allergen treatment has led to 2 possible explanations. First, this observed phenomenon might be attributed to development of an antibody response to the hapten. Several investigators have shown that epicutaneous application of dinitrofluorobenzene, TNCB, and oxazolone can elicit antihapten antibodies (Askenase and Hayden, 1974Go; Dearman and Kimber, 1991Go, 1992Go; Takahashi et al., 1978Go; Taylor and Iverson, 1971Go; Thomas et al., 1976Go). Still other investigators suggest that anti-idiotypic antibodies may arise in animals as a consequence of contact sensitization and serve to regulate this immune response (Sy et al., 1979Go). The second possible explanation is that the B cell response may be the outcome of T cell activation. It is known that cytokines from both Th1 and Th2 T cells can stimulate B lymphocytes. Data from Neumann et al. (1992) indicate such a B cell response may result from T cell activation. They demonstrated a dose-dependent increase in the percentage of B220+ cells in the popliteal lymph nodes after in vitro T cell activation using anti-CD3 antibody treatment as well as a decrease in the percentage of Thy 1.2, CD4 and CD8 positive T cells. This increase in B220+ cells seemed to reflect an influx of B lymphocytes rather than an increase in proliferation because few of the B220+ cells were cycling in the S/G2M phase of the cell cycle. Interestingly, preferential accumulation of B lymphocytes in draining lymph nodes has been seen in other T cell-mediated immune (CMI) responses. Fojtasek et al. (1993) studied the CMI response (which is the principal host defense) to Histoplasma infection in the DLN of B6C3F1 mice after respiratory exposure. Seven days after infection, the percentage of B cells in the DLN rose to 43% compared with 26% in uninfected control mice and this elevated response remained throughout the course of infection. Constant and Wilson (1992) also saw a proportionally greater increase in B cells compared with T cells in draining lymph nodes following immunization with the attenuated larvae of Schistosoma mansoni for which the principal host defense is T cell-mediated. Thus, the increase in percentage of B cells in DLN of mice seems to be a consistent finding for several T cell-mediated immune responses including contact sensitization.

One potential application for examining B220+ B cells may be in its use as an alternative or supplementary endpoint for the murine LLNA. This assay measures the proliferative response (3H-thymidine incorporation) in the draining auricular lymph nodes during the induction phase of a contact sensitization response after epicutaneous exposure to materials (Basketter et al., 1996Go; Gerberick et al., 2000Go; Kimber et al., 1986Go; Kimber and Weisberger, 1989Go). It is based on the observation that the induction phase of a contact sensitization response to allergen is characterized by lymphocyte proliferation and hyperplasia in the lymph nodes draining the site of topical exposure. Measuring the percentage of B220+ B cells in DLN might provide an alternative, nonradioisotope approach for assessing skin sensitization potential. Consistent with previous findings, we showed a significant increase in the percentage of B220+ cells in DLN isolated from mice treated with allergens, but not with irritants (Table 1Go). The strong allergens (DNCB and TNCB) gave a more robust B220 response than did weaker allergens (EUG and HCA). However, it is important to point out that in these studies the allergens and irritants were tested at only one concentration; thus, it is not possible to get a good indication of allergenic potency as is the case in the standard LLNA (Basketter et al., 2000Go).

To assess the potential use of analysis of the B cell marker B220 as an alternative endpoint for the LLNA, we developed a model based on the % B220+ cells to classify chemicals as allergens or irritants. A scatterplot of % B220 values for each chemical treatment (41 and 28 independent observations with allergens and irritants, respectively) versus the concurrent vehicle controls was used to illustrate differences between allergens and irritants (Fig. 1Go). A classification tree model approach was used to calculate a B220 test:vehicle ratio cutoff of 1.25 for discriminating between allergens and irritants. A compound with a % B220/vehicle ratio above 1.25 is classified as an allergen and a ratio below 1.25 would be classified as an irritant. Once a cutoff value was established with the dataset we evaluated each chemical individually and found that the model had an accuracy of 93%. Moreover, the model was very reproducible, with the strong allergen DNCB being classified correctly in 17 of 17 experiments and the weak allergen HCA classified correctly in 12 of 13 experiments (Table 2Go). More importantly, the performance of the model was demonstrated successfully in a second independent laboratory (Tables 4 and 5GoGo) illustrating the potential ease of transfer of the model into other laboratories.

Although the great majority of nonsensitizing irritants are negative in the LLNA, cellular proliferation has been observed with a few irritants (Basketter et al., 1998Go; Gerberick et al., 1992Go; ICCVAM, 1999Go; Ikarashi et al., 1993Go; Montelius et al., 1994Go; Sikorski et al., 1996Go). This proliferation has been thought to be nonspecific and nonantigen directed. Analysis of the % B220+ cells in the LLNA may provide a potential means for improving the capability of the assay to differentiate irritant from allergic responses. Some of the irritants used in these studies were chosen because they were previously shown to induce proliferative activity in the DLN, as measured by 3H-thymidine incorporation (Gerberick et al., 1992Go; Kimber et al., 1998Go; Loveless et al., 1996Go). However, in the majority of studies conducted with irritants in the experiments reported herein, despite causing some increase in DLN cellularity in some cases, the chemicals were classified correctly as irritants using the 1.25 % B220:vehicle ratio cutoff. For example, SLS, which has been reported to give a false positive response in the LLNA (Basketter et al., 1998Go; Loveless et al., 1996Go), was classified correctly in 6 of 6 experiments in Laboratory 1 as well as being classified as an irritant in a single experiment in Laboratory 2 with respect to the percentage of B220+ cells (Tables 2 and 5GoGo). Another chemical irritant that has been reported to induce proliferation in a slightly modified version of LLNA is BZC (Gerberick et al., 1992Go). Indeed, in the current experiments BZC stimulated a significant increase in DLN cellularity to levels similar to those seen for 3 out of 5 of the allergens. Using the 1.25 % B220:vehicle ratio cutoff, however, BZC was classified correctly as an irritant in 11 of 13 experiments in Laboratory 1 and in the single experiment performed in Laboratory 2 (Tables 2 and 5GoGo). While SA had a mean B220:vehicle ratio of 1.44, it was classified as an irritant in 2 of the 4 experiments (Table 2Go). Perhaps further testing of SA may have produced a greater number of correct classifications and thus reduced the mean B220:vehicle ratio. It is very important to note that other criteria can be used to determine whether or not a chemical may be producing a false positive LLNA response (Basketter et al., 1998Go). For example, one needs to consider whether or not the material is a known strong irritant, or if its chemical structure contains known skin sensitization structural alerts or that it induces a proliferative response only at high test concentrations and the nature of the dose response. Regardless, further evaluation of the % B220 test:vehicle ratio cutoff value of 1.25 with a wider spectrum of irritants and weak allergens will be needed to determine the usefulness of this approach for differentiating allergen from irritant responses.

In summary, the present investigations reported herein demonstrate that measuring the increase in the percentage of B220+ B cells in DLN after chemical treatment is useful in discriminating between allergen and irritant responses. Similar to the standard LLNA, a range of concentrations should be tested when assessing new chemicals by this method in order to obtain dose response data. The method is reproducible and appears to be sufficiently robust for effective transfer to other laboratories. Further evaluation and validation of the % B220 test:vehicle ratio using a wider range of test materials and expanded interlaboratory trials, is needed to assess fully its potential as an alternative endpoint for the LLNA that eliminates the need for radioisotopes and increases discrimination between allergen and irritant responses. Moreover, in addition to being used for hazard identification, the B220 method may also prove to be useful in providing information on allergenic potency.


    ACKNOWLEDGMENTS
 
We gratefully acknowledge detailed review of this manuscript by Drs. J. Nash and Michael K. Robinson and assistance with flow cytometry analysis by Dr. Gregg Ridder and Ms. Susan M. Dawes.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (513) 627-0400. E-mail: gerberick.gf{at}pg.com. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Askenase, P. W., and Hayden, B. J. (1974). Cytophillic antibodies in mice contact-sensitized with oxazolone. Immunochemical characterization and preferential binding to a typsin-sensitive macrophage receptor. Immunology 27, 563–576.[ISI][Medline]

Basketter, D. A., Balikie, D. A., Dearman, R. J., Kimber, I., Ryan, C. A., Gerberick, G. F., Harvey, P., Evans, P., White, I. R., and Rycroft, R. J. G. (2000). Use of the local lymph node assay for the estimation of relative contact allergenic potency. Contact Dermatitis 42, 344–348.[ISI][Medline]

Basketter, D. A., Gerberick, G. F., and Kimber, I. (1998). Strategies for identifying false positive responses in predictive skin sensitization tests. Food Chem. Toxicol. 36, 327–333.[ISI][Medline]

Basketter, D. A., Gerberick, G. F., Kimber, I., and Loveless, S. E. (1996). The local lymph node assay: A viable alternative to currently accepted skin sensitization tests. Food Chem. Toxicol. 34, 985–987.[ISI][Medline]

Basketter, D. A., and Scholes, E. W. (1992). Comparison of the local lymph node assay with the guinea-pig maximization test for the detection of a range of contact allergens. Food Chem. Toxicol. 30, 65–69.[ISI][Medline]

Basketter, D. A., Scholes, E. W., and Kimber, I. (1994). The performance of the local lymph node assay with chemicals identified as contact allergens in the human maximization test. Food Chem. Toxicol. 32, 543–547.[ISI][Medline]

Basketter, D. A., Scholes, E. W., Kimber, I., Botham, P. A., Hilton, J., Miller, K., Robbins, M. C., Harrison, P. T. C., and Waite, S. J. (1991). Interlaboratory evaluation of the local lymph node assay with 25 chemicals and comparison with guinea pig-test data. Toxicol. Methods 1, 30–43.

Basketter, D. A., Selbie, E., Scholes, E. W., Lees, D., Kimber, I., and Botham, P. A. (1993). Results with OECD-recommended positive control sensitizers in the maximization: Buehler and local lymph node assays. Food Chem. Toxicol. 31, 63–67.[ISI][Medline]

Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J. (1983). Classification and Regression Trees. Wadsworth, Belmont, CA.

Constant, S. L., and Wilson, R. A. (1992). In vivo lymphocyte responses in the draining lymph nodes of mice exposed to Schistosoma mansoni: Preferential proliferation of T cells is central to the induction of protective immunity. Cell Immunol. 139, 145–161.[ISI][Medline]

Dearman, R. J., and Kimber, I. (1991). Differential stimulation of immune function by respiratory and contact chemical allergens. Immunology 72, 563–570.[ISI][Medline]

Dearman, R. J., and Kimber, I. (1992). Divergent immune responses to respiratory and contact chemical allergens: Antibody elicited by phthalic anhydride and oxazolone. Clin. Exp. Allergy 22, 241–250.[ISI][Medline]

Fojtasek, M. F., Sherman, M. R., Garringer, T., Blair, R., Wheat, L. J. and Schnizlein-Bick, C. T. (1993). Local immunity in lung-associated lymph nodes in a murine model of pulmonary histoplasmosis. Infect. Immun. 61, 4607–4614.[Abstract]

Gerberick, G. F., Cruse, L. W., Miller, C. M., and Ridder, G. M. (1999). Selective modulation of B cell activation markers CD86 and I-Ak on murine draining lymph node cells following allergen or irritant treatment. Toxicol. Appl. Pharmacol. 159, 142–151.[ISI][Medline]

Gerberick, G. F., Cruse, L. W., Miller, C. M., Sikorski, E. E., and Ridder, G. M. (1997). Selective modulation of T cell memory markers CD62L and CD44 on murine draining lymph node cells following allergen and irritant treatment. Toxicol. Appl. Pharmacol. 146, 1–10.[ISI][Medline]

Gerberick, G. F., House, R. V., Fletcher, E. R., and Ryan, C. A. (1992). Examination of the local lymph node assay for use in contact sensitization risk assessment. Fundam. Appl. Toxicol. 19, 438–445.[ISI][Medline]

Gerberick, G. F., Robinson, M. K., Ryan, C. A., Dearman, R. J., Kimber, I., Basketter, D. A., Wright, Z. and Marks, J. G. (2001). Contact allergenic potency: Correlation of human and local lymph node assay data. Am. J. Contact Dermatitis 12, 156–161.[Medline]

Gerberick, G. F., Ryan, C. A., Kimber, I., Dearman, R. J., Lea, L. J., and Basketter, D. A. (2000). Local lymph node assay: Validation assessment for regulatory purposes. Am. J. Contact Dermatol. 11, 3–18.

Holzhütter, H.-G., Archer, G., Dami, N., Lovell, D. P., Saltelli, A., and Sjöström, M. (1996). Recommendations for the application of biostatistical methods during the development and validation of alternative toxicological methods. ATLA 24, 511–530.[ISI]

Homey, B., von Schilling, C., Blumel, J., Schupper, H.-C., Ruzicka, T., Ahr, H. J., Lehmann, P., and Vohr, H.-W. (1998). An integrated model for the differentiation of chemical-induced allergic and irritant skin reactions. Toxicol. Appl. Pharmacol. 153, 83–94.[ISI][Medline]

Ikarashi, Y., Tsukamoto, Y., Tsuchiya, T., and Nakamura, A. (1993). Influence of irritants on lymph node cell proliferation and the detection of contact sensitivity to metal salts in the murine local lymph node assay. Contact Dermatitis 29, 128–132.[ISI][Medline]

ICCVAM (1999). The murine local lymph node assay: A test method for assessing the allergic contact dermatitis potential of chemicals/compounds. Interagency Coordinating Committee on the Validation of Alternative Methods Report. NIH Publication No.: 99–4494.

Kimber, I., and Basketter, D. A. (1992). The murine local lymph node assay: A commentary on collaborative studies and new directions. Food Chem. Toxicol. 30, 165–169.[ISI][Medline]

Kimber, I., Hilton, J., and Botham, P. A. (1990). Identification of contact allergens using the murine local lymph node assay: Comparisons with the Buehler occluded patch test in guinea pigs. J. Appl. Toxicol. 10, 173–180.[ISI][Medline]

Kimber, I., Hilton, J., Botham, P. A., Basketter, D. A., Scholes, E. W., Miller, K., Robbins, M. C., Harrison, P. T. C., Gray, T. J. B., and Waite, S. J. (1991). The murine local lymph node assay: Results of an interlaboratory trial. Toxicol. Lett. 55, 203–213.[ISI][Medline]

Kimber, I., Hilton, J., Dearman, R. J., Gerberick, G. F., Ryan, C. A., Basketter, D. A., Lea, L., House, R. V., Ladics, G. S., Loveless, S. E., and Hastings, K. L. (1998). Assessment of the skin sensitization potential of topical medicaments using the local lymph node assay: An interlaboratory evaluation. J. Toxicol. Environ. Health A 53, 563–579.[ISI][Medline]

Kimber, I., Hilton, J., Dearman, R. J., Gerberick, G. F., Ryan, C. A., Basketter, D. A., Scholes, E. W., Ladics, G. S., Loveless, S. E., House, R. V., et al. (1995). An international evaluation of the murine local lymph node assay and comparison of modified procedures. Toxicology 103, 63–73.[ISI][Medline]

Kimber, I., Mitchell, J. A., and Griffin, A. C. (1986). Development of a murine local lymph node assay for the determination of sensitizing potential. Food. Chem. Toxicol. 24, 585–586.[ISI]

Kimber, I., and Weisenberger, C. (1989). A murine local lymph node assay for the identification of contact allergens. Assay development and results of an initial validation study. Arch. Toxicol. 63, 274–282.[ISI][Medline]

Kraal, G., and Twisk, A. (1984). The interaction of high endothelial venules with T and B cells in peripheral lymph nodes after antigenic stimulation. Eur. J. Immunol. 14, 586–588.[ISI][Medline]

Kuhn, U., Lempertz, U., Knop, J., and Becker, D. (1995). A new method for phenotyping proliferating cell nuclear antigen-positive cells using flow cytometry: Implications for analysis of the immune response in vivo. J. Immunol. Methods 179, 215–222.[ISI][Medline]

Loveless, S. E., Ladics, G. S., Gerberick, G. F., Ryan, C. A., Basketter, D. A., Scholes, E. W., House, R. V., Hilton, J., Dearman, R. J., and Kimber, I. (1996). Further evaluation of the local lymph node assay in the final phase of an international collaborative trial. Toxicology 108, 141–152.[ISI][Medline]

Manetz, T. S., and Meade, B. J. (1999). Development of a flow cytometry assay for the identification and differentiation of chemicals with the potential to elicit irritation, IgE-mediated, or T cell-mediated hypersensitivity responses. Toxicol. Sci. 48, 206–217.[Abstract]

Montelius, J., Wahlkvist, H., Boman, A., Fernstrom, P., Grabergs, L., and Wahlberg, J. E. (1994). Experience with the murine local lymph node assay: Inability to discriminate between allergens and irritants. Acta Derm. Venereol.(Stockh.) 74, 22–27.

Neumann, C.M., Oughton, J.A., and Kerkvliet, N.I. (1992). Anti-CD3-induced T-cell activation in vivo—I. Flow cytometric analysis of dose-responsive, time-dependent, and cyclosporin A-sensitive parameters of CD4+ and CD8+ cells from the draining lymph nodes of C57B1/6 mice. Int. J. Immunopharmacol. 14, 1295–1304.

Ryan, C. A., Gerberick, G. F., Cruse, L. W., Basketter, D. A., Lea, L., Blaikie, L., Dearman, R. J., Warbrick, E. V., and Kimber, I. (2000). Activity of human contact allergens in the murine local lymph node assay. Contact Dermatitis 43, 95–102.

Satterthwaite, F. W. (1946). An approximate distribution of estimates of variance components. Biom. Bull. 2, 110–114.

Scholes, E. W., Basketter, D. A., Sarll, A. E., Kimber, I., Evans, C. D., Miller, K., Robbins, M. C., Harrison, P. T. C., and Waite, S. J. (1992). The local lymph node assay: Results of a final interlaboratory validation under field conditions. J. Appl. Toxicol. 12, 217–222.

Sikorski, E. E., Gerberick, G. F., Ryan, C. A., Miller, C. M., and Ridder, G. M. (1996). Phenotypic analysis of lymphocyte subpopulations in lymph nodes draining the ear following exposure to contact allergens and irritants. Fundam. Appl. Toxicol. 34, 25–35.

Steel, R. G. D., and Torrie, J. H. (1980), Principles and Procedures of Statistics, 2nd ed. McGraw-Hill, New York.

Sy, M. S., Moorhead, J. W., and Claman, H. N. (1979). Regulation of cell-mediated immunity by antibodies: Possible role of anti-receptor antibodies in the regulation of contact sensitivity to DNFB in mice. J. Immunol. 123, 2593–2598.

Takahashi, I., Nishikawa, S., Katsura, Y., and Izumi, T. (1978). Anti-DNP antibody response after the topical application of DNFB in mice. Immunology 22, 589–596.

Taylor, R. B., and Iverson, G. M. (1971). Hapten competition and the nature of cell cooperation in the antibody response. Proc. R. Soc. Lond. B. Biol. Sci. 176, 393–418.

Thomas, W. R., Asherson, G. L., and Watkins, M. C. (1976). Reaginic antibody produced in mice with contact sensitivity. J. Exp. Med. 144, 1386–1390.