Hexachlorobenzene-induced Immunopathology in Brown Norway Rats is Partly Mediated by T Cells

Janine Ezendam*,{dagger},{ddagger},1, Ine Hassing*, Rob Bleumink*, Joseph G. Vos{dagger},{ddagger} and Raymond Pieters*

* Institute for Risk Assessment Sciences (IRAS), Immunotoxicology, Utrecht University, P.O. Box 80.176, 3508 TD Utrecht, Netherlands; {dagger} National Institute for Public Health and the Environment, Laboratory for Toxicology, Pathology and Genetics, Bilthoven, P.O. Box 1 3720 BA, Netherlands; and {ddagger} Faculty of Veterinary Medicine, Department of Pathobiology, Utrecht University, P.O. Box 80.150, 3508 TD Utrecht, Netherlands

1 To whom correspondence should be addressed at IRAS, Immunotoxicology, Utrecht University, P.O. Box 80.176, 3508 TD Utrecht, Netherlands. Fax: +31302535077. E-mail: J.Ezendam{at}iras.uu.nl.

Received July 16, 2003; accepted November 4, 2003

ABSTRACT

Hexachlorobenzene (HCB) is a persistent environmental pollutant with (auto)immune effects in humans and rats. The Brown Norway (BN) rat is very susceptible to HCB-induced immunopathology, and oral exposure causes inflammatory skin and lung lesions, splenomegaly, lymph node (LN) enlargement, and increased serum levels of IgE and anti-ssDNA IgM. The role of T cells in HCB-induced immunopathology is unclear and to elucidate this Cyclosporin A (CsA) was used. BN rats were exposed to either a control diet or a diet supplemented with 450 mg/kg HCB for 21 days. CsA treatment started 2 days prior to HCB exposure and rats were injected daily with 20 mg/kg body weight CsA. Treatment with CsA prevented the HCB-induced immunopathology significantly. The onset of skin lesions was delayed and the severity was also strongly decreased. Furthermore, CsA prevented the HCB-induced increase in spleen weight partly and the increase in auricular LN weight completely. The increase in serum IgE and IgM against ssDNA levels was prevented completely. Macrophage infiltrations into the spleen and lung still occurred but infiltrations of eosinophilic granulocytes into the lung were prevented. Restimulation of spleen cells with the T-cell mitogen ConA and the macrophage activator LPS clearly showed that CsA inhibited T-cell activation, but not macrophage activation. Together, our results show that both T cells and macrophages play a prominent role in HCB-induced immunopathology.

Key Words: hexachlorobenzene; T cells; macrophages; immunostimulation; Cyclosporin A; Brown Norway rat.

Hexachlorobenzene (HCB) is an environmental pollutant with pronounced adverse effects on the immune system of humans and rats (Michielsen et al., 1999bGo). An accidental poisoning in Turkey in the 1950s revealed several immunopathological effects of HCB in humans, such as splenomegaly, enlarged lymph nodes (LN), and painless arthritis (Cam, 1958Go; Peters, 1976Go). Breast-fed infants born to mothers exposed to HCB developed inflammatory lung and skin lesions, the latter characterized by infiltrations of lymphocytes and macrophages (Cam, 1960Go). Data from humans exposed occupationally to HCB indicated that HCB induced significantly elevated levels of IgM and IgG, as well as reduced neutrophil functions (Queiroz et al., 1998aGo,bGo).

HCB induced comparable immunostimulatory effects in rats (Schielen et al., 1993Go; Vos, 1986Go). The BN rat is a very susceptible strain and oral exposure to HCB caused splenomegaly, enlarged LN, inflammatory skin lesions characterized by an infiltration of eosinophilic granulocytes, inflammatory lung lesions characterized by macrophage infiltration, granuloma formation and perivascular eosinophilic infiltrates, and increased levels of total serum IgM, IgG, IgE, and IgM against ssDNA (Michielsen et al., 1997Go).

The mechanism of HCB-induced immunopathology in humans and rats is poorly understood. The involvement of T cells is still controversial and if T cells are involved it is not known how they become activated. Studies in Wistar rats have shown that HCB treatment enhanced thymus-dependent parameters such as primary and secondary IgM and IgG against tetanus toxoid, and mitogenic responses to T-cell mitogens PHA and ConA (Vos et al., 1979Go). To investigate the role of thymus-dependent T cells Michielsen et al. (1999a)Go exposed BN rats depleted of T cells by adult thymectomy, lethal irradiation, and bone marrow reconstitution to HCB and found that skin lesions appeared later in HCB-exposed T-cell-depleted rats than in T-cell-competent BN rats. The development of spleen and lung effects by HCB was not influenced by this T-cell depletion, so it seemed that HCB-induced inflammatory effects were not thymus-dependent, except for the skin lesions. A drawback of this study was that there were still functional T cells present in the spleen as demonstrated by flow cytometry. In thymectomized rats 4% T cells were present in the spleen as compared to 15% in control rats. Since Sado et al. have shown that T cells present in the spleen of thymectomized mice were able to proliferate after stimulation with ConA (Sado et al., 1980Go), these residual T cells may be responsible for the observed immune effects elicited by HCB in these thymectomized rats.

Since our previous work did not elucidate the precise role of T cells in HCB-induced immunopathology, we decided to investigate this more thoroughly by using Cyclosporin A (CsA) as CsA not only decreases peripheral T-cell number, but also inhibits antigen-induced T-cell activation (Hollander et al., 1994Go; Kosugi et al., 1989aGo).

MATERIALS AND METHODS

Rats and maintenance.
Three-week old specific pathogen-free female inbred Brown Norway (BN/SsNOlaHsD, termed BN) rats were purchased from Harlan (Blackthorn, UK). Rats were allowed to acclimatize 1 week before starting the experiment. All rats were housed at the animal facilities of the Utrecht University. They were kept two by two in filter-topped macrolon cages on bedding of chips and wood, under standard conditions (50-60% relative humidity, 12 h dark/12 h light cycle) with food and acidified drinking water ad libitum. The diet consisted of a semisynthetic diet (SSP/TOX, Hope Farms, Woerden, The Netherlands) either or not supplemented with crystalline HCB (99% purity; Aldrich Chemie, Bornem, Belgium) by mixing of homogeneity. The experiments were conducted according to the guidelines of the animal experiments committee of the Faculty of Veterinary Medicine of the Utrecht University.

Experimental protocol.
Rats were randomly assigned to the different experimental groups, either receiving control diet or the diet supplemented with 450 mg HCB per kg. One of the control groups and one of the HCB-fed groups received a daily injection of CsA (a gift from Novartis Pharma Inc., Basel, Switzerland). CsA was dissolved in olive oil at a concentration of 20 mg/ml. Rats treated with CsA received a daily injection of 20 mg/kg/day (= 1 ml/kg) via a sc injection in the flank; untreated rats received a daily injection of 1 ml olive oil per kg also in the flank. CsA treatment started 2 days prior to the beginning of HCB exposure. Rats were weighed three times per week and the development of skin lesions was evaluated daily. The severity of the lesional sites was rated as described before (Michielsen et al., 1997Go), irrespective of size, as 1 = minimal (some redness), 2 = moderate (redness), 3 = marked (dry desquamation and crusts), and 4 = severe (exudative lesions). A semiquantitative group score of the macroscopic skin lesions on day 21 was calculated by multiplying the fraction of rats showing skin lesions, the average severity of the skin lesions and the relative lesional area.

After 21 days of exposure to HCB rats were killed by a lethal dose of sodium pentobarbital (Euthesate, 0.3 g/kg body weight, ip). Blood was drawn from the vena cava and serum was collected after clotting and centrifugation. Liver, spleen, thymus (freed from adjacent lymph nodes), mesenteric LN (MLN), and auricular LN (ALN) were collected and weighed. LN and parts of the liver and spleen were fixed in phosphate-buffered 4% paraformaldehyde and lung was inflated and fixed with phosphate-buffered 4% paraformaldehyde to optimize morphology. Formaldehyde-fixed tissues were embedded in Paraplast and sections (5 µm) were stained with hematoxylin and eosin (HE). Part of the spleen was snap-frozen in liquid nitrogen and used for cryostat sections. To avoid bias at examination, sections were scored under code.

Single cell suspensions were prepared from part of spleen, part of thymus, ALN, and MLN in phosphate buffered saline (PBS) containing 1% bovine serum albumin (BSA). LN and thymus were minced between two slides and spleens were cut and gently pressed through a 70-µm pore sieve to prepare single cell suspensions. Suspensions were washed, resuspended in PBS/1% BSA, counted using a Coulter Counter (Coulter Electronics, Luton, UK) and then adjusted to 1 or 2.5 x 106 cells per ml.

IgE ELISA.
Total serum IgE levels were determined by an isotype-specific sandwich ELISA as described before (Michielsen et al., 1997Go). Individual test serum samples were measured in duplicate. The concentration of IgE in test serum samples was calculated using a standard curve with known amounts of monoclonal rat IgE (Serotec, Oxford, UK) and results were expressed in total serum IgE concentrations (ng/ml).

ssDNA Elisa.
ELISA procedures to determine IgM levels against ssDNA were performed essentially as described previously with some modifications (Schielen et al., 1993Go). In short, serial dilutions of the individual test sera were added in duplicate to wells coated with ssDNA (50 µg/ml; Sigma, St. Louis, MO). Each plate contained serial dilutions of NRS to obtain a reference curve. After incubation alkaline phosphatase labeled goat-anti-rat IgM (µ) (Brunswich, Amsterdam, The Netherlands) was added to the wells. For the color reaction wells were incubated with a solution of 0.01% 4-nitrophenylphospate in diethanolaminobuffer (pH 9.8). The reaction was stopped by adding 50 µl 10% (w/v) EDTA in distilled water. Absorbance was measured at 405 nm. To calculate IgM serum levels against ssDNA calibration curves of the reference serum (NRS) were used. Therefore, absorbance values of serial dilutions of the reference serum were plotted against the 2 log of the dilution. The dilution of the test serum that would result in the same absorbance as NRS was calculated by linear regression and was used to obtain the ELISA index, which is the ratio of the test dilution used and the calculated dilution of NRS at the same absorbance. Mean ELISA index of the control group was transformed to 1 and data of the treatment groups was expressed relative to the index of the control group.

Immunohistology.
Cryostat sections (6 µm) of spleen were stained with ED3 (monoclonal antibody that recognizes a subpopulation of macrophages). ED3 was a kind gift from Dr. C. Dijkstra (Free University, Amsterdam, The Netherlands). Peroxidase-conjugated rabbit anti-mouse Ig (Dako AS, Glostrup, Denmark) was used as a second step agent. As a substrate 3-amino-9-ethylcarbazole (AEC; Janssen, Beerse, Belgium) was used. Slides were counterstained with hematoxylin.

Flow cytometry.
All antibodies for flow cytometry were obtained from Pharmingen (San Diego, CA). The following conjugated mAbs were used in duplo- or triplostaining to phenotype lymphocytes: fluorescein-isothiocyanate (FITC)-conjugated OX-33 (anti-rat CD45RA), phycoerythrin (PE)-conjugated R73 (anti-rat TCR{alpha}ß), OX39 (anti-rat CD25), OX-8 (anti-rat CD8{alpha}), and biotinylated OX-35 (anti-rat CD4). Single cell suspensions (2 x 105 cells/well) in PBS with 1% BSA, 0.1% NaN3, 4% normal rat serum were centrifuged in a 96-well microtiter plate, resuspended and incubated with combinations of optimal dilutions of FITC, -PE- and biotin-conjugated mAbs (30 min, in the dark, 4°C). Cells were washed twice and the biotin-conjugated antibodies were stained in a second step with streptavidin-CyChrome (30 min, in the dark, 4°C). Samples were analyzed on a FACScan® flow cytometer with standard FACSflow using CellQuest software (BD Biosciences, Franklin Lakes, NJ).

Ex vivo restimulation of spleen cells.
For cytokine measurements spleen cells were restimulated ex vivo with the mitogens LPS and ConA or with medium alone. LPS (2 µg/ml), ConA (5 µg/ml), and complete medium (RPMI 1640 with glutamax; Invitrogen Life Technologies, Paisley, Scotland) supplemented with 10% fetal bovine serum (FBS; ICN Biomedicals, Costa Mesa, CA), 2% penicillin-streptomycin (Invitrogen Life Technologies, Paisley, UK) were plated in a 96-well microtitre plate. Spleen cells in complete medium were added to the different stimuli (3.75 x 105 cells per well). Cells were incubated for 48 h at 37°C, 5% CO2. After incubation supernatants were collected and stored at -70°C until cytokine measurement.

IL-2 measurement.
The IL-2-dependent cell line CTLL-2 was kindly provided by Dr Janssen (Immunotherapy, Utrecht Medical Center, Utrecht, The Netherlands). CTLL-2 cells were grown in Iscove's medium supplemented with 10% FBS, 2% penicillin-streptomycin, L-glutamin, 1% mercapto-ethanol and 20 units/ml of IL-2. For the IL-2 bioassay cells were washed in medium without IL-2 and plated in 96-well plates at a concentration of 5000 cells per well. Diluted samples and a standard curve were added to the wells and incubated for 24 h at 37°C in 5% CO2. Thereafter, cells were labeled for 18 h with 0.5 µCi/well 3H-thymidine. Plates were harvested with a cell harvester onto filter paper and dried. The filter paper was placed into a bag containing scintillation fluid and 3H-Thymidine incorporation was measured with a beta-plate liquid scintillation counter. The standard curve was used to calculate IL-2 (in units/ml) in the individual samples.

Tumor necrosis factor-{alpha} and nitric oxide measurement.
The tumor necrosis factor-{alpha} (TNF-{alpha})-sensitive murine fibroblast cell line L929 was kindly provided by Dr. Wulferink (PharmaAware, Utrecht University, Utrecht, The Netherlands). L929 cells were seeded in 96-well flat-bottom microtiter plates (4 x 104 cells per well) in complete RPMI and allowed to adhere o.n. at 37°C in a 5% CO2 atmosphere. After removal of supernatant, cells were pretreated with 2 µg/ml Actinomycin-D (Acros Organics, Geel, Belgium). Thereafter, diluted samples were added to the wells. A standard curve of recombinant rat TNF-{alpha} (Biosource, Camarillo, CA) was included on each plate. Plates were incubated for 18 h at 37°C in a 5% CO2 atmosphere. Cell viability was determined by measuring the decrease in mitochondrial activity using the MTT (3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide; Sigma Chemical Co., St. Louis, MO) reduction assay. The standard curve was used to calculate TNF-{alpha} production (ng/ml). Nitric oxide (NO) production after LPS restimulation of splenocytes was measured by the Griess reaction that measures the accumulation of nitrite in the culture medium as described before (Green et al., 1982Go). A standard curve of sodium nitrite was used to calculate nitrite production (µM).

Statistical analysis.
Significant differences of treatment groups with the control group or between the HCB treated groups with or without CsA treatment were determined by ANOVA with the Bonferroni post hoc test for contrasts by using SPSS software. Values deviating more than two standard deviations from group means were considered as outliers and not included in statistical analyses. Preceding statistical analysis, all data were transformed to log 10 values to homogenize variance.

RESULTS

Macroscopic Skin Lesions
As described previously (Michielsen et al., 1997Go), oral treatment with HCB caused skin lesions in the head and neck region of BN rats. Table 1 shows that rats treated with HCB alone developed skin lesions after 9 or 10 days of treatment, which is according to previous reports. Skin lesions increased in size and severity resulting in hemorrhagic lesions often with exudative crusts. CsA caused a delay of the development of skin lesions with 9-11 days as the first lesions were visible on days 19, 20, or 21. As can be seen in Table 1 all rats displayed skin lesions on day 21, but rats treated with CsA had less severe lesions and the affected area was much smaller.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Effect of CsA on the Onset, Incidence, Severity, and Lesional Area of Skin Lesions Induced by HCB

 
Body Weight Gain and Relative Organ Weights
Neither HCB nor CsA treatment influenced the gain in body weight during the course of the experiment (data not shown). As shown in Figure 1, relative spleen and ALN weights were significantly increased in both groups treated with HCB as compared to control rats. CsA treatment inhibited this increase in relative spleen weight moderately but spleen weights were still increased when compared to the control group. In contrast, CsA treatment prevented the HCB-induced increase in ALN weight completely. ALN weight of HCB-fed rats treated with CsA was not significantly different from control rats, but was significantly higher than control rats treated with CsA.



View larger version (16K):
[in this window]
[in a new window]
 
FIG. 1. Relative spleen (A) and ALN (B) weights after 21 days of exposure of female BN rats to control diet or HCB diet with or without CsA treatment. Data represent mean ± SE of four animals per group. The asterisks above the bars indicate significance from control group (*p < 0.05; **p < 0.01). Significance from HCB treated group is depicted with ##p < 0.01. §§Indicates significance from the control group treated with CsA (§§p < 0.01).

 
Total Serum IgE Levels and IgM Levels against ssDNA
As has been described previously, HCB caused a significant increase in total serum IgE levels in BN rats (Michielsen et al., 1997Go). Table 2 shows that treatment with CsA prevented this increase in IgE levels completely. The BN rat is a strain with high basal IgE levels and in control rats the total serum IgE levels were 195 ng/ml. After CsA treatment total serum IgE levels in both control rats and HCB-fed rats were low and in both groups only one out of four rats had detectable IgE levels (detection limit 40 ng/ml). Only the values of the positive animals are given in Table 2. Serum levels of IgM against ssDNA are also shown in Table 2. Oral exposure to HCB increased anti-ssDNA IgM levels with a factor of 1.7 and treatment with CsA suppressed this completely to control levels.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Total Serum IgE and Serum IgM Levels against ssDNA

 
Histopathology
Lung.
HE stained sections of formalin-fixed lung show that the morphology of rats fed the control diet either treated with CsA or not appeared normal (Fig. 2) except for some scattered microgranulomas consisting of eosinophilic granulocytes. The presence of these microgranulomas in the lung of untreated BN rats have been described before (Ohtsuka et al., 1997Go). Treatment with HCB caused an infiltration of eosinophilic granulocytes around small blood vessels, granuloma formation and also intra-alveolar accumulation of macrophages. These HCB-induced effects are according to those described before (Michielsen et al., 1997Go). Rats fed with HCB and treated with CsA still had focal accumulations of macrophages comparable with HCB treated rats, but the perivascular infiltrate of eosinophilic granulocytes did not occur.



View larger version (109K):
[in this window]
[in a new window]
 
FIG. 2. Representative HE stained lung sections of female BN rats exposed to control diet (A), 450 mg HCB per kg diet (B), or 450 mg HCB per kg diet plus CsA treatment (C and D). Feeding rats with HCB induced perivascular eosinophilic infiltrates in the lung (arrows in B). CsA treatment prevented this infiltration of eosinophils (C) but HCB-induced accumulations of focal accumulations of macrophages (asterisks) did still occur (D).

 
Spleen.
The most prominent alterations induced by HCB in the spleen were comparable with effects observed previously in rats (Michielsen et al., 1999aGo; Vos et al., 1979Go). HCB induced hyperplasia of the B-cell compartments (marginal zone [MZ] and follicles) and increased extramedullary hemopoiesis. Morphology of spleens of HCB-fed rats treated with CsA showed depleted T-cell areas (PALS), while some but not all of the B-cell compartments were hyperplastic. This hyperplasia was less when compared with the hyperplasia observed in HCB-fed rats not treated with CsA. HCB-induced hemopoiesis was inhibited by CsA (data not shown). Staining of cryostat sections with the MAb ED3, which stains a subpopulation of macrophages, revealed an increase of the amount of ED3+ macrophages (Fig. 3) leading to broadening of the rim these macrophages in the MZ and scattered infiltrates of ED3+ in the white pulp, as described previously in Wistar rats (Schielen et al., 1993Go). This infiltration of ED3+ macrophages still occured after CsA treatment.



View larger version (77K):
[in this window]
[in a new window]
 
FIG. 3. ED3 staining of representative spleen sections from control rat (A), HCB treated rat (B), and HCB and CsA treated rat (C). HCB induced an increase of ED3+ macrophages leading to broadening of the rim of these macrophages in the marginal zone (MZ) and scattered infiltrates of ED3+ in the white pulp (arrows in B and C). This increase was not influenced by CsA treatment.

 
CD25 Expression in the MLN
Figure 4 shows the expression of CD25 in the MLN. HCB treatment induced a strong increase in CD25 expression from 5.6% in control rats to 13%, which was prevented completely by CsA treatment. In control rats CsA treatment decreased CD25 expression slightly to 3.3% (not significant).



View larger version (7K):
[in this window]
[in a new window]
 
FIG. 4. CD25 expression in MLN expressed as percentage cells staining positive. The asterisks above the bars indicate significance from control group (**p < 0.01) and significance from the HCB group is indicated with ##p < 0.01. Data represent mean ± SE from four animals per group.

 
Effects of CsA on Thymic and Peripheral T Cells
The effects of CsA on thymocyte development have been studied intensively (Kosugi et al., 1989aGo,bGo). In the present study CsA treatment inhibited the development of mature single-positive thymocytes from their double-positive precursors, resulting in a reduction of mature TCR{alpha}ß single-positive CD4+ and to a lesser extent a reduction of mature TCR{alpha}ß single-positive CD8+ thymocytes (data not shown). Table 3 shows the relative and absolute numbers of T and B cells and the CD4/CD8 ratio in the spleen and ALN. CsA treatment caused a decrease of relative and absolute T-cell number in spleen and ALN in both the control and the HCB-treated group. The decreased CD4/CD8 ratio shows that predominantly CD4+ T cells were affected by CsA treatment, which is in line with previous work (Fukuzawa et al., 1989Go). In both spleen and ALN absolute T-cell and B-cell numbers were increased in the HCB treated group, whereas the relative T- and B-cell numbers in this group were not changed indicating that both cell types contributed equally to the observed enlargement of spleen and ALN.


View this table:
[in this window]
[in a new window]
 
TABLE 3 T- and B-Cell Number and CD4/CD8 Ratio in Spleen and ALN

 
Cytokine and NO Production by Spleen cells after Mitogen Restimulation
To establish the effect of CsA on T-cell activation, spleen cells were restimulated with ConA and subsequently IL-2 levels in the supernatant were measured. Table 4 shows that HCB treatment alone caused a significant increase of IL-2 levels and that after CsA treatment IL-2 levels in both the control group and the HCB-fed group were lower than the untreated control group. To investigate whether CsA treatment affected macrophage function spleen cells were also restimulated with LPS and TNF-{alpha} and NO production were measured. TNF-{alpha} levels were twice as high after HCB treatment and this increase was not influenced by CsA treatment (Table 4). HCB also induced an increase in NO production and this increase was also not affected by CsA treatment.


View this table:
[in this window]
[in a new window]
 
TABLE 4 IL-2, TNF-{alpha}, and NO Production after Restimulation of Splenocytes with the Mitogens ConA and LPS

 
DISCUSSION

The present study showed that CsA treatment delayed the onset of HCB-induced skin lesions and decreased the severity of these skin effects, but did not affect the incidence. CsA prevented increase in spleen weight, but not to control values. In contrast, the increase in ALN weight was prevented completely and infiltration of eosinophils and increases in serum IgE and anti-ssDNA IgM were absent or at control level, respectively, in HCB-fed animals treated with CsA. The T-cell-specific nature of the CsA inhibitory effect was confirmed by the reduction of ConA-induced IL-2 production and the unaffected LPS-induced TNF-{alpha} and NO production by splenocytes. In line with this, macrophage infiltrations into lung and spleen were still present regardless of the CsA treatment. Together, these data show that the immunopathological effects of HCB are to a large extent mediated by T cells.

The observed upregulation of CD25, or IL-2 receptor, in the MLN supports the relevance of T cells in HCB-induced immunopathology. The finding that CD25 expression in control rats and HCB-fed rats treated with CsA was not altered is in accordance with knowledge that CsA does not act at the level of the IL-2 receptor, but downregulates IL-2 mRNA (Gelfand et al., 1987Go) and production of IL-2 (see above).

Present observations are partly in line with studies investigating the role of T cells in HCB-induced immunopathology by using thymectomized, lethal irradiated, and bone marrow reconstituted BN rats (Michielsen et al., 1999aGo). These results already suggested that thymus-dependent T cells were not required for induction of inflammatory skin effects by HCB, but that T cells accelerated the occurrence of these skin lesions. In contrast to our results splenomegaly and eosinophilia in the lung did still occur in these T-cell-depleted rats. This previous study, however, was not completely conclusive since after the T-cell-depleting procedure (thymectomy, lethal irradiation plus bone marrow reconstitution) these BN rats still had functional T cells left in the periphery, which could account for some T-cell-dependent processes. By using CsA we circumvented this, since CsA not only reduces the number of T cells, but also inhibits antigen-induced T-cell activation by interfering with the cytokine production (Gelfand et al., 1987Go). Furthermore, we show that HCB-induced lung eosinophilia is not dependent on mediators derived from other cell types than T cells as was suggested previously (Michielsen et al., 1999aGo).

The increase in IgM levels against ssDNA has previously been related to the ability of HCB to activate B1-cells, a subset of B cells present in the MZ of the spleen (Schielen et al., 1995Go) and known to produce low-affinity IgM antibodies against autoantigens. This increase was absent in HCB-fed rats treated with CsA. Remarkably, in these rats in particular the MZ was still hyperplastic, although less than in rats treated only with HCB. The HCB-induced influx of ED3+ macrophages, which constitute an important cell population in the MZ and thought to be capable of activating MZ B cells (Damoiseaux et al., 1991), was not affected by CsA treatment. Interestingly, our data suggest that ED3+ macrophages are responsible for the expansion the MZ, but that T cells are required for the augmentation of the production of ssDNA-IgM by B-1 cells.

Remarkably, CsA suppressed the severity and delayed the onset of skin lesions but did not influence the incidence, since all rats eventually developed skin lesions. This might be due to HCB-induced macrophage activation, leading to the production of inflammatory cytokines, NO and oxidative stress which was not inhibited by CsA. CD8{alpha} macrophages have been observed in the dermis of HCB treated rats before (Michielsen et al., 1999aGo) and these macrophages might become activated by uptake of the inert HCB and initiate inflammatory responses and lesions in the skin. Nevertheless T cells are indispensable in further acceleration of skin pathology.

Together our results indicate that T cells play a prominent role in HCB-induced eosinophilia and humoral responses but that macrophages do account for at least a part of the observed adverse immune effects. Importantly, we show that macrophage activation and T-cell activation by HCB are separate entities.

How macrophage and T-cell activation leads to pathology is not completely clear as yet. But one can envisage that the HCB is first taken up by macrophages and that responses similar to those observed after silica exposure take place. Silica is an inert chemical that induces autoimmune-like phenomena due to chronic activation of macrophages, leading to the production of pro-inflammatory cytokines like TNF-{alpha} and reactive oxygen species, which results in chronic inflammation and tissue damage (Driscoll et al., 1990Go; Kim et al., 1999Go). Subsequently, these adjuvant signals could polyclonally activate the available T cell-pool, including autoreactive T cells. In addition, inflammatory cells, in particular polymorphonuclear cells attracted by the presence of HCB, might induce myeloperoxidase-dependent formation of reactive metabolites of HCB. It is known that pentachlorophenol, an oxidative metabolite of HCB, can be oxidized into tetrachlorohydroquinone (TCHQ) and TCBQ by peroxidases (Samokyszyn et al., 1995Go). Recently, we have shown by using the mouse popliteal lymph node assay that TCBQ is capable of inducing neoantigen specific T-cell hapten (Ezendam et al., 2003Go). Once formed in vivo, these reactive intermediates would be able to haptenize to proteins to form neoantigens and then trigger hapten-specific T cells as well. Since cytochrome P450 catalyzed oxidation is not involved in HCB-induced immunopathology (Schielen et al., 1995Go), extrahepatic metabolism by phagocytes might be implicated, as has been shown before for other allergenic or autoimmunogenic chemicals (reviewed in Uetrecht, 1997Go).

Our future research will focus on the specificity of T cells and the role of macrophages in HCB-induced immunopathology. Interestingly, the present data stresses the suitability of the HCB-Brown Norway model to separately evaluate cellular and molecular interactions between innate and adaptive immune phenomena in relation to autoimmune-like syndromes. As such it adds to already existing Brown Norway models using metals or pharmaceuticals (Pelletier et al., 1988Go; Tournade et al., 1990Go).

REFERENCES

Cam, C. (1958). Cases of skin porphyria related to hexachlorobenzene intoxication. Saglik Dergisi 32, 215–216.

Cam, C. (1960). Une nouvelle dermatose épidémique des enfants. Anales de Dermatologie 87, 393–397.

Driscoll, K. E., Lindenschmidt, R. C., Maurer, J. K., Higgins, J. M., and Ridder, G. (1990). Pulmonary response to silica or titanium dioxide: Inflammatory cells, alveolar macrophage-derived cytokines, and histopathology. Am. J. Respir. Cell. Mol. Biol. 2, 381–390.[ISI][Medline]

Ezendam, J., Vissers, I., Bleumink, R., Vos, J. G., and Pieters, R. (2003). Immunomodulatory effects of tetrachlorobenzoquinone, a reactive metabolite of hexachlorobenzene. Chem. Res. Toxicol. 16, 688–694.[ISI][Medline]

Fukuzawa, M., Sharrow, S. O., and Shearer, G. M. (1989). Effect of cyclosporin A on T cell immunity. II. Defective thymic education of CD4 T helper cell function in cyclosporin A-treated mice. Eur. J. Immunol. 19, 1147–1152.[ISI][Medline]

Gelfand, E. W., Cheung, R. K., and Mills, G. B. (1987). The cyclosporins inhibit lymphocyte activation at more than one site. J. Immunol. 138, 1115–1120.[Abstract/Free Full Text]

Green, L. C., Wagner, D. A., Glogowski, J., Skipper, P. L., Wishnok, J. S., and Tannenbaum, S. R. (1982). Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids. Anal. Biochem. 126, 131–138.[ISI][Medline]

Hollander, G. A., Bierer, B. E., and Burakoff, S. J. (1994). Molecular and biological actions of cyclosporin A and FK506 on T cell development and function. Transfus. Sci. 15, 207–220.[ISI][Medline]

Kim, J. K., Lee, W. K., Lee, E. J., Cho, Y. J., Lee, K. H., Kim, H. S., Chung, Y., Kim, K. A., and Lim, Y. (1999). Mechanism of silica- and titanium dioxide-induced cytotoxicity in alveolar macrophages. J. Toxicol. Environ. Health A 58, 437–450.[CrossRef][ISI][Medline]

Kosugi, A., Sharrow, S. O., and Shearer, G. M. (1989a). Effect of cyclosporin A on lymphopoiesis. I. Absence of mature T cells in thymus and periphery of bone marrow transplanted mice treated with cyclosporin A. J. Immunol. 142, 3026–3032.[Abstract/Free Full Text]

Kosugi, A., Zuniga-Pflucker, J. C., Sharrow, S. O., Kruisbeek, A. M., and Shearer, G. M. (1989b). Effect of cyclosporin A on lymphopoiesis. II. Developmental defects of immature and mature thymocytes in fetal thymus organ cultures treated with cyclosporin A. J. Immunol. 143, 3134–3140.[Abstract/Free Full Text]

Michielsen, C. C., Bloksma, N., Klatter, F. A., Rozing, J., Vos, J. G., and van Dijk, J. E. (1999a). The role of thymus-dependent T cells in hexachlorobenzene-induced inflammatory skin and lung lesions. Toxicol. Appl. Pharmacol. 161, 180–191.[CrossRef][ISI][Medline]

Michielsen, C. P., Bloksma, N., Ultee, A., van Mil, F., and Vos, J. G. (1997). Hexachlorobenzene-induced immunomodulation and skin and lung lesions: A comparison between Brown Norway, Lewis, and Wistar rats. Toxicol. Appl. Pharmacol. 144, 12–26.[CrossRef][ISI][Medline]

Michielsen, C. C., van Loveren, H., and Vos, J. G. (1999b). The role of the immune system in hexachlorobenzene-induced toxicity. Environ. Health Perspect. 107, 783–792.

Ohtsuka, R., Doi, K., and Itagaki, S. (1997). Histological characteristics of respiratory system in Brown Norway rat. Exp. Anim. 46, 127–133.[CrossRef][ISI][Medline]

Pelletier, L., Pasquier, R., Rossert, J., Vial, M. C., Mandet, C., and Druet, P. (1988). Autoreactive T cells in mercury-induced autoimmunity. Ability to induce the autoimmune disease. J. Immunol. 140, 750–754.[Abstract/Free Full Text]

Peters, H. A. (1976). Hexachlorobenzene poisoning in Turkey. Fed. Proc. 35, 2400–2403.[ISI][Medline]

Queiroz, M. L., Bincoletto, C., Perlingeiro, R. C., Quadros, M. R., and Souza, C. A. (1998a). Immunoglobulin levels in workers exposed to hexachlorobenzene. Hum. Exp. Toxicol. 17, 172–175.[CrossRef][ISI][Medline]

Queiroz, M. L., Quadros, M. R., Valadares, M. C., and Silveira, J. P. (1998b). Polymorphonuclear phagocytosis and killing in workers occupationally exposed to hexachlorobenzene. Immunopharmacol. Immunotoxicol. 20, 447–454.[ISI][Medline]

Sado, T., Kamisaku, H., and Aizawa, S. (1980). Nature of T-cells resident in spleens of thymectomized, lethally irradiated, bone marrow-reconstituted mice. Cell. Immunol. 49, 51–63.[ISI][Medline]

Samokyszyn, V. M., Freeman, J. P., Maddipati, K. R., and Lloyd, R. V. (1995). Peroxidase-catalyzed oxidation of pentachlorophenol. Chem. Res. Toxicol. 8, 349–355.[ISI][Medline]

Schielen, P., Den Besten, C., Vos, J. G., Van Bladeren, P. J., Seinen, W., and Bloksma, N. (1995). Immune effects of hexachlorobenzene in the rat: Role of metabolism in a 13-week feeding study. Toxicol. Appl. Pharmacol. 131, 37–43.[CrossRef][ISI][Medline]

Schielen, P., Schoo, W., Tekstra, J., Oostermeijer, H. H., Seinen, W., and Bloksma, N. (1993). Autoimmune effects of hexachlorobenzene in the rat. Toxicol. Appl. Pharmacol. 122, 233–243.[CrossRef][ISI][Medline]

Tournade, H., Pelletier, L., Pasquier, R., Vial, M. C., Mandet, C., and Druet, P. (1990). D-penicillamine-induced autoimmunity in Brown-Norway rats. Similarities with HgCl2-induced autoimmunity. J. Immunol. 144, 2985–2991.[Abstract/Free Full Text]

Uetrecht, J. P. (1997). Current trends in drug-induced autoimmunity. Toxicology 119, 37–43.[CrossRef][ISI][Medline]

Vos, J. G. (1986). Immunotoxicity of hexachlorobenzene. In Hexachlorobenzene: Proceedings of an International Symposium (C. R. Morris, and J. R. Cabral, Eds.), pp. 347–356. IARC Scientific Publications, Lyon.

Vos, J. G., Van Logten, M. J., Kreeftenberg, J. G., and Kruizinga, W. (1979). Hexachlorobenzene-induced stimulation of the humoral response in rats. Ann. N.Y. Acad. Sci. 320, 535–550.[Abstract]