- Email: [email protected]
Immunocompetent cells for in vitro screening of skin irritation
Toxic. in Vitro Vol. 7, No. 4, pp. 389-392, 1993 Printed in Great Britain.All rights reserved
0887-2333/93$6.00+ 0.00 Copyright © 1993PergamonPress Lid
IMMUNOCOMPETENT CELLS FOR IN VITRO SCREENING OF SKIN IRRITATION W. J. W. PAPE*, J. DEGWERT,F. STECKELand U. HOPPE c/o Beiersdorf AG, 4232--Biocompatibility, Cosmed Division--PGU Research Center, Unnastrasse 48, D-2000 Hamburg 20, Germany Abstract--The present studies were aimed at evaluating procedures for assessing the immunmodulatory effects of chemicals and preparations on macrophage differentiation and lymphocyte proliferation in cell cultures. The effects of l0 drugs and anti-inflammatory agents were monitored by determining thymidine incorporation into phytohaemagglutinin (PHA)-stimulated T cells in the lymphocyte transformation test (LTT) and the expression of two surface antigens on macrophages in the macrophage differentiation assay (MDA). One antigen was found on macrophages in acute inflamedtissue. The other was detected on those found in recoveringtissue. These parameters were compared with mean skin irritation scores for 12 known cosmetic products from epicutaneous patch testing. Finally, these parameters were also used to study six cosmetic test formulae with unknown irritation potentials subjected to blind testing during phase 2 of the "CTFA Evaluation of Alternatives Program". Immunosuppressiveagents were detected in both systems. Agents, thought to be pro-inflammatory, were monitored in the MDA by the acute inflammation marker. Skin irritation scores of known preparations correlated well with those of expressed acute inflammation markers in the MDA (rs = 0.714), but no clear relationship was detectable in the LTT. In contrast one of the CTFA samples tested blind revealed a strong response in both tests. The roll-on antiperspirant stimulated T-cell proliferation and induce a strong expression of the acute inflammation marker on macrophages. Based on these findings further studies are in progress to evaluate the usefulness of these in vitro tests for predicting dermal irritation.
Introduction
applied preparations can be derived from experiments with immunocompetent cells screening cellular responses that might be related to typical signs of adverse skin reactions. It is known that peripheral blood mononuclcar cells are centrally involved in acute inflammatory
In recent years, immunologists have focused their interest on the skin as the largest interfacial organ of the human body showing complex immune responses after exposure to different environmental toxins, irritants or allergens. It has become increasingly evident that the epidermis does not only serve as a barrier against environmental insults, but also as an initiation site of immune defence mechanisms, The skin can be the target of specific immunological as well as of non-specific inflammatory responses, The ceils involved have been summarized as the "skin immune system" (Bos and Kapsenberg, 1986), which is closely related to the "skin associated lymphoid tissue" proposed by Streilein (1983). With regard to the complex network of skin responses dermatologists have to discriminate between irritant and allergic dermatitis. During the last decade the incidence of adverse skin reactions seems to have increased. In order to identify irritant materials that might induce acute toxic reactions, we have recently proposed an in vitro test system (Pape and Hoppe, 1991) that is in use as a first approach to avoid animal testing. We feel that additional information about topically CsA=eyclosporin A; DMSO= dimethyl sulphoxide; LPS = lipopolysaecharide; LTT = lymphocyte transformation test; MDA = maerophage differentiation assay; PBMNC = peripheral blood mononuelear cells; PBS = phosphate buffered saline; PHA = phytohaemagglutinin;PWM = pokeweek mitogen,
Abbreviations:
processes. Specific immunomodulatory effects of different materials can be studied in the lymphocyte transformation test using, for instance, peripheral blood mononuclear cells (PBMNC) or cell lines as recently proposed by Steer et al. (1990). Another less specific approach could be to look at the differentiation of macrophages characterized by the sequential expression of cell surface molecules as defined by monoclonal antibodies. One of these antibodies (27E10) with specificity mainly directed against the early stage inflammation macrophage is found preferentially in acute inflammatory lesions (Zwadlo et al., 1986). The other (RM3/1) with specificity mainly associated with macrophages is found in the downregulatory phase of inflammatory processes (Zwadlo et al., 1987). In skin reactions other cells such as keratinocytes and endothelial cells are also essentially involved. If it can be shown that lymphocyte proliferation and macrophage differentiation are affected by test chemicals and preparations, such cellular tests could be incorporated in a battery as a further step to detect skin irritants by in vitro methods. In the present study the lymphocyte transformation test (LTT) and macrophage differentiation assay (MDA) with PBMNC were used to evaluate the
389
390
W.J.W. PAPEet al.
immunomodulatory effects of chemical agents and cosmetic preparations, Material and Methods
Chemicals and test preparations. All drugs and agents were purchased from Sigma Chemical Co. (Munich). Cyclosporin A was used as the pharmaceutical preparation (Sandimmun from Sandoz, Nurnberg). Dithranol and Therafectin were supplied by the pharmacological department. All drugs and test samples were dissolved, if necessary, in dimethyl sulphoxide (DMSO) (cell culture grade, Sigma Chemical Co.) and diluted with medium to a maximum final test concentration of 0.05% DMSO. Lipopolysaccharide (LPS) from S. minnesota was purchased from Sigma. Cosmetic preparations were emulsions of the O/W- and W/O-type with different skin irritation potentials. The non-irritant materials were skin care and moisturizing products as well as sunscreens with different sunscreen protection factors. Different types of cleansing products were used as slightly to moderately irritant preparations. Six CTFAtest samples were chosen and used as supplied for phase 2 of the evaluation programme, Isolation o f peripheral blood mononuclear cells, Heparinized blood (50 U/ml) from healthy volunteers was diluted 1: 2 with balanced salt solution, layered onto Ficoll-Paque gradient and centrifuged at 200 g for 20 min at room temperature. The mononuclear cell fraction was collected, washed three times with phosphate buffered saline (PBS)and resuspended in RPMI-1640 medium containing 10% foetal calf serum. This fraction was used for the LTT and isolation of the monocyte fraction. Lymphocyte transformation test. Total PBMNC were plated into 96-well fiat-bottomed microtitre plates (3 x 105 cells/well) with 200#1 RPMI-1640. Cells were cultured with or without stimulating concentrations of the T-cell specific mitogen phytohaemagglutinin (PHA) (2.5pg PHA/ml) or the B lymphocyte specific pokeweek mitogen (PWM) (2.5#g PWM/ml) as described by Hume and Weidemann (1980). PHA-triggered lymphocyte reactivity in the presence of CTFA-samples was evaluated with a mitogen concentration of 1.25/tg PHA/ml. Serial dilutions of the test substances (100-0.01 pmol for chemicals and 100-0.1 pg/ml for other test materials) were added and incubated in triplicate. Test solutions exceeding a final test concentration of 0.05% were prepared adding DMSO as needed, Incubation was in microtitre plates at 37°C with 7% CO2 in a humidified atmosphere for 72 hr. During the last 8hr the cells were pulsed with 0.4pCi [3H]thymidine and harvested using a Titertek Cell Harvester (Flow Laboratories, Meckenheim) for the determination of [3H]thymidine incorporation, Thymidine incorporation rates were calculated as the percentage of PHA-stimulated minus unstimulated control ( = 100%).
Macrophage differentiation assay. Ficoll-Paque isolated mononuclear cells as described above were layered on a percoll-gradient according to the method of Zwadlo et al. (1986). The monocyte fraction was isolated, washed three times with McCoy's 5A medium and plated into fiat-bottomed microtitre plates to a final concentration of 5 x 105 cells/well in 200/~1 McCoy's 5A medium with 20% inactivated human serum. The cells were incubated for 96 hr at 37°C and 7% CO2 with or without varying concentration ranges of test substances. Test solutions were prepared as described above. Afterwards cells were washed with PBS and fixed with ice-cold fixing solution (1% acetic acid in aqueous ethanol [40%]) for 10 min. The fixed cells were stained by indirect immunoperoxidase staining with monoclonal antibodies 27E10 and RM 3/1 (Dianova, Hamburg) as described by Zwadlo et aL (1986 and 1987). Cells were stained with 3-amino-9-ethylcarbazol and analysed for antigen expression under the microscope. The visual scoring was as follows: 0 = no additional staining, + = defined and slightly intensifted staining, ÷ + = moderately intensified staining, + + + =strongly, and + + + + =very highly intensified staining compared with the untreated control. LPS-activated macrophages were used as a positive control. Human patch test. The backs of 20 healthy volunteers were exposed to 100 p l undiluted test samples for 24 hr under occlusion using patch test plasters (Leukotest, BDF Hamburg). 48 hr after application the test areas were inspected and skin reactions estimated according to the scoring system of Patrick and Maibach (1991): 0=negative, normal skin; + = questionable erythema not covering entire area; 1 = definite erythema; 2 = erythema and induration; 3 = vesiculation; 4 = bullous reaction. Statistical analysis. Spearman's rank correlation was performed according to the method described by Sachs (1984). The levels of significance (ct) were related to the correlation coefficient as follows (n = 12): r~> 0.4965 (ct = 0.050), rs > 0.6713 (~t =0.010), and r~>0.8182 (~=0.001). Results and Discussion
The modulating effects of 10 drugs and anti-inflammatory agents on PBMNCs were studied at concentrations between 0.1 and 100 nmol/ml. Typical results at a concentration of 1.0 nmol/ml are summarized in Table 1. At non-cytotoxic levels substances known to be immunosuppressive in vivo like cyclosporin A (CsA), dexamethasone, hydrocortisone, and chloroquine showed clearly detectable reductions of the maximum response of PBMNCs induced by 2.5/~g PHA/ml. Acetylsalicylic acid reduces T-cell proliferation to a level of about 80% over the whole range of concentrations tested, whereas CsA, like the corticosteroids, suppressed the T-cell responses effectively even at concentrations below 1 nmol/ml.
lmmunocompetent cells in skin irritation
391
Table I. Immunmodulationinduced by anti-inflammatoryand pro-inflammatoryagentsand drugs in the LTT and the MDA MDA Test material (1.0 nmol/ml) Cyclosporin A (0.6 nmol/ml)* Dexamethasone* Hydrocortisone* Chloroquine* Cyclophosphamide* Therafectin (92 nmol/ml) Therafectin (46 nmol/ml) lndomethacin Acetylsalicylicacid Dithranol Retinoic acid, trans
PHA-LTT (%)t 15.9 23.9 33.7 51.9 82.4 4.6 87.5 92.4 78.5 82.5 70.1
RM 3/1:~ + + + + + + + + 0 0 nd + + + + + 0 0
27EI0~ 0 0 0 + 0 nd + 0 + + + + + + + +
nd= not done *Substances known to be immunosuppressivein vivo. tPercentage of PHA-induced lymphocyteproliferation (2.5 #g PHA/ml). ~:RM 3/1 antibody binding to surface marker seen on macrophagesduring recovery. §27EI0 antibody binding to early-stageinflammationmarker on macrophages. All values are means of triplicates.
Therafectin, an isopropylidene-dimethylaminopropylglucofuranose, described as a T-cell suppressing agent required a higher concentration to down-regulate T-cell proliferation, Cyclophosphamide, although known to have an immunosuppressive effect after metabolic transformation in vivo neither suppressed the mitogen-induced T-cell response nor stimulated macrophages in vitro to express the .corresponding surface antigen detectable by antibody RM3/1. The same 10 agents were examined in the M D A for the expression o f two differentiation markers as described and characterized by Zwadlo et al. (1986 and 1987). M a c r o p h a g e subsets found in acute inflamed dermal tissues o f h u m a n patients (Czarnetzki et al., 1990) were detected by means o f 27E10 m o n o clonal antibody, directed against a protein expressed on their surface. It has been reported that macrophages can be detected in recovering or healing tissue using monoclonal antibody R M 3 / I . Immunosuppressive drugs like CsA, dexamethasone, hydrocortisone, and Therafectin induced the expression o f the down-regulatory inflammation marker detected by the antibody R M 3 / I . Although
there was a reduced T-cell blastogenic reaction to chloroquine in the LTT, macrophages did not respond. Drugs thought to have a pro-inflammatory effect, such as dithranol and retinoic acid, activated macrophages to give a strong expression o f the early-stage inflammatory marker. However, there was no expression o f the second marker and the in vitro lymphocyte responses were not affected. Both substances are k n o w n to evoke primary skin reactions. Based on these observations we c o m p a r e d visual estimations o f dermal erythema induced during epicutaneous patch testing o f 12 selected cosmetic emulsions to evaluate how well the expression o f the macrophage markers correlates with dermal responsiveness. The m e a n values o f the dermal erythema scores ranking from non-irritant to moderately irritant were c o m p a r e d with estimated intensities o f staining o f the expressed early-stage inflammation marker. Results shown in Table 2 reveal an acceptable correlation o f the intensity o f erythema with the expression o f the acute inflammation marker as analysed by rank correlation according to Spearman (Sachs, 1984). The
Table 2. Comparisonof macrophagedifferentiationassay in vitro with epicutaneouspatch test results MDAt Test product Epicutaneous type reactions* RM3/I 27E10 Emollient cream 0.0 0/+ 0/+ Body care lotion 0.06 + 0/+ Moisturizing cream 0.06 + 0 After sun body lotion 0.13 + + Hand and body lotion 0.19 + +/+ + Cleansing face milk 0.50 0 0 Facial make-up remover 0.50 + 0/+ Moisturizing lotion 0.75 + + + Cleansing lotion 1.50 + + + + Cleansing cream 1.56 + + +/+ + + Skin clearing milk 1.75 + +/+ + Cleansing paste 2.00 0/+ + + *Original concentration 100 #l/test area; mean values of 48-hr scores. tTest concentration0.01 (g/litre)dissolvedin DMSO (0.05% final concn); 27EI0 antibody binding to early-stage inflammation marker; RM3/1 antibody binding to surface marker seen during recovery; all values are means from triplicates. TIV 7/4---G
W . J . W . PAPE et al.
392
Table 3. Reactivity of macrophages and PHA-triggered lymphocytes to test products from the CTFA study (phase 2) MDA* PHA-LTTt CTFA test sample (No.) 27EI0 RM3/I (%) Emollient cream 5 0 0/+ 91.7 Cuticle remover 6 + + + 0 86.4 Ointment 0 % SDS) 7 + + 0/+ 78.3 Cleansing cream II 10 + + 0/+ 105.0 Ointment (2% SDS) 14 + + 0 86.8 Roll-on antiperspirant 18 + + + + 0 145.1 *Test concentration 0.01 (g/litre) dissolved in DMSO (0.05% final conch); 27E10 antibody binding to early-stage inflammation marker. $Percentages of maximum stimulation induced by 2.5 #g PHA/ml; test concentration as above. All values are means of triplicates. correlation coefficient r s was 0.7138 (~t < 0.010, n = 12). There was n o correlation with the m a r k e r expressed during d o w n regulation of the inflammatory process ( R M 3 / I ) . T-lymphocyte proliferation was n o t affected by any o f these test materials (data n o t shown). Finally, we have studied six cosmetic f o r m u l a t i o n s f r o m the " C T F A E v a l u a t i o n o f Alternatives Prog r a m " (phase 2) t h a t have given inconsistent results in different in vitro irritation tests (Gettings et al., 1991; Pape a n d Hoppe, 1991). Table 3 shows t h a t the L T T results were hardly affected by samples 5, 6, 7, 10 a n d 14. W i t h the exception o f sample 5 in M D A these p r o d u c t s showed a n up-regulated expression o f the early-stage of inflammation marker, which indicates a n irritation potential. In the LTT, sample 18 represented a strong dose-dependent increase o f 4 5 % relative to the P H A - i n d u c e d effect alone. This stimulatory response c o r r e s p o n d s to the surprisingly high expression o f the early-stage i n f l a m m a t i o n m a r k e r o n macrophages, suggesting t h a t the roll-on antiperspir a n t m i g h t contain a n i m m u n o s t i m u l a t i n g agent. Moreover, these findings were s u p p o r t e d by the dermal reactions observed after accidental skin contact during experimentation. P W M - i n d u c e d Blymphocyte proliferation was not influenced. In addition, there was n o induction of the staining o f m a c r o p h a g e s with a n t i b o d y R M 3 / 1 . F u r t h e r examination o f test m e t h o d s using i m m u n o c o m p e t e n t cells are presently in progress to evaluate the usefulness o f such in vitro a p p r o a c h e s for predicting
dermal irritation. REFERENCES Bos J. D. and Kapsenberg M. L. (1986) The skin immune
system. Its cellular constituents and their interaction. ImmunologyToday 7, 235-239. Czarnetzki B. M., Zwadlo-Klarwasser G., Br6cker E.-B. and Sorg C. (1990) Macrophage subsets in different types of urticaria. Archives of Dermatological Research 282, 93-97. Gettings S. D., Teal J. J., Bagley D. M., Demetrulias J. L., DiPasquale L. C., Hintze K. L., Rozen M. G., Weise S. L , Chudkowsky M., Marenus K. D., Pape W. J. W., Roddy M., Schnetzinger R., Silber P. M., Glaza S. M. and Kurtz P. J. (1991) The CTFA Evaluation of Alternatives Program: an evaluation of in vitro alternatives to the Draize primary eye irritation test. (Phase 1) Hydroalcoholic formulations; (part 2) data analysis and biological significance. In Vitro Toxicology 4, 247-288. Hume D. A. and Weidemann M. J. (1980) Mitogenic lymphocyte transformation. In Methods in Enzymology. Vol. 150. Elsevier, New York. Pape W. J. W. and Hoppe U. (1991) In vitro methods for the assessment of primary local effects of topically applied preparations. Skin Pharmacology 4, 205-212. Patrick E. and Maibach H. I. (1991) Predictive skin irritation tests in animals and human. In Dermatotoxicology. By G. Di Sabato. Vol. 150.4th Ed. Edited by F. N. Marzulli and H. I. Maibach. pp. 201-222, Hemisphere Publ. Corp. NY. Sachs L. (1984) Angewandte Statistik--Statistiche Methoden und l h r e Anwendung. 6th Ed. pp. 309-312. Springer-Verlag, Berlin. Steer S., Lasek W., Clothier R. H. and Balls M. (1990) An in vitro test for immunomodulators? Toxicology in Vitro 4, 360-362. Streilein J. W. (1983) Skin associated lymphoid tissue (SALT): origins and function. Journal of Investigative Dermatology80 (suppl.) 125-165. Zwadlo G., Schlegel R. and Sorg C. (1986) A monoclonal antibody to a subset of human monocytes found only in the peripheral blood and inflammatory tissues. Journal of Immunology137, 512-518. Zwadlo G., Voegeli R., Schulze-Osthoff K. and Sorg C. (1987) A monoclonal antibody to a novel differentiation antigen on human macrophages associated with downregulatory phase of the inflammatory process. ExperimentalCell Biology 55, 295-304.
0887-2333/93$6.00+ 0.00 Copyright © 1993PergamonPress Lid
IMMUNOCOMPETENT CELLS FOR IN VITRO SCREENING OF SKIN IRRITATION W. J. W. PAPE*, J. DEGWERT,F. STECKELand U. HOPPE c/o Beiersdorf AG, 4232--Biocompatibility, Cosmed Division--PGU Research Center, Unnastrasse 48, D-2000 Hamburg 20, Germany Abstract--The present studies were aimed at evaluating procedures for assessing the immunmodulatory effects of chemicals and preparations on macrophage differentiation and lymphocyte proliferation in cell cultures. The effects of l0 drugs and anti-inflammatory agents were monitored by determining thymidine incorporation into phytohaemagglutinin (PHA)-stimulated T cells in the lymphocyte transformation test (LTT) and the expression of two surface antigens on macrophages in the macrophage differentiation assay (MDA). One antigen was found on macrophages in acute inflamedtissue. The other was detected on those found in recoveringtissue. These parameters were compared with mean skin irritation scores for 12 known cosmetic products from epicutaneous patch testing. Finally, these parameters were also used to study six cosmetic test formulae with unknown irritation potentials subjected to blind testing during phase 2 of the "CTFA Evaluation of Alternatives Program". Immunosuppressiveagents were detected in both systems. Agents, thought to be pro-inflammatory, were monitored in the MDA by the acute inflammation marker. Skin irritation scores of known preparations correlated well with those of expressed acute inflammation markers in the MDA (rs = 0.714), but no clear relationship was detectable in the LTT. In contrast one of the CTFA samples tested blind revealed a strong response in both tests. The roll-on antiperspirant stimulated T-cell proliferation and induce a strong expression of the acute inflammation marker on macrophages. Based on these findings further studies are in progress to evaluate the usefulness of these in vitro tests for predicting dermal irritation.
Introduction
applied preparations can be derived from experiments with immunocompetent cells screening cellular responses that might be related to typical signs of adverse skin reactions. It is known that peripheral blood mononuclcar cells are centrally involved in acute inflammatory
In recent years, immunologists have focused their interest on the skin as the largest interfacial organ of the human body showing complex immune responses after exposure to different environmental toxins, irritants or allergens. It has become increasingly evident that the epidermis does not only serve as a barrier against environmental insults, but also as an initiation site of immune defence mechanisms, The skin can be the target of specific immunological as well as of non-specific inflammatory responses, The ceils involved have been summarized as the "skin immune system" (Bos and Kapsenberg, 1986), which is closely related to the "skin associated lymphoid tissue" proposed by Streilein (1983). With regard to the complex network of skin responses dermatologists have to discriminate between irritant and allergic dermatitis. During the last decade the incidence of adverse skin reactions seems to have increased. In order to identify irritant materials that might induce acute toxic reactions, we have recently proposed an in vitro test system (Pape and Hoppe, 1991) that is in use as a first approach to avoid animal testing. We feel that additional information about topically CsA=eyclosporin A; DMSO= dimethyl sulphoxide; LPS = lipopolysaecharide; LTT = lymphocyte transformation test; MDA = maerophage differentiation assay; PBMNC = peripheral blood mononuelear cells; PBS = phosphate buffered saline; PHA = phytohaemagglutinin;PWM = pokeweek mitogen,
Abbreviations:
processes. Specific immunomodulatory effects of different materials can be studied in the lymphocyte transformation test using, for instance, peripheral blood mononuclear cells (PBMNC) or cell lines as recently proposed by Steer et al. (1990). Another less specific approach could be to look at the differentiation of macrophages characterized by the sequential expression of cell surface molecules as defined by monoclonal antibodies. One of these antibodies (27E10) with specificity mainly directed against the early stage inflammation macrophage is found preferentially in acute inflammatory lesions (Zwadlo et al., 1986). The other (RM3/1) with specificity mainly associated with macrophages is found in the downregulatory phase of inflammatory processes (Zwadlo et al., 1987). In skin reactions other cells such as keratinocytes and endothelial cells are also essentially involved. If it can be shown that lymphocyte proliferation and macrophage differentiation are affected by test chemicals and preparations, such cellular tests could be incorporated in a battery as a further step to detect skin irritants by in vitro methods. In the present study the lymphocyte transformation test (LTT) and macrophage differentiation assay (MDA) with PBMNC were used to evaluate the
389
390
W.J.W. PAPEet al.
immunomodulatory effects of chemical agents and cosmetic preparations, Material and Methods
Chemicals and test preparations. All drugs and agents were purchased from Sigma Chemical Co. (Munich). Cyclosporin A was used as the pharmaceutical preparation (Sandimmun from Sandoz, Nurnberg). Dithranol and Therafectin were supplied by the pharmacological department. All drugs and test samples were dissolved, if necessary, in dimethyl sulphoxide (DMSO) (cell culture grade, Sigma Chemical Co.) and diluted with medium to a maximum final test concentration of 0.05% DMSO. Lipopolysaccharide (LPS) from S. minnesota was purchased from Sigma. Cosmetic preparations were emulsions of the O/W- and W/O-type with different skin irritation potentials. The non-irritant materials were skin care and moisturizing products as well as sunscreens with different sunscreen protection factors. Different types of cleansing products were used as slightly to moderately irritant preparations. Six CTFAtest samples were chosen and used as supplied for phase 2 of the evaluation programme, Isolation o f peripheral blood mononuclear cells, Heparinized blood (50 U/ml) from healthy volunteers was diluted 1: 2 with balanced salt solution, layered onto Ficoll-Paque gradient and centrifuged at 200 g for 20 min at room temperature. The mononuclear cell fraction was collected, washed three times with phosphate buffered saline (PBS)and resuspended in RPMI-1640 medium containing 10% foetal calf serum. This fraction was used for the LTT and isolation of the monocyte fraction. Lymphocyte transformation test. Total PBMNC were plated into 96-well fiat-bottomed microtitre plates (3 x 105 cells/well) with 200#1 RPMI-1640. Cells were cultured with or without stimulating concentrations of the T-cell specific mitogen phytohaemagglutinin (PHA) (2.5pg PHA/ml) or the B lymphocyte specific pokeweek mitogen (PWM) (2.5#g PWM/ml) as described by Hume and Weidemann (1980). PHA-triggered lymphocyte reactivity in the presence of CTFA-samples was evaluated with a mitogen concentration of 1.25/tg PHA/ml. Serial dilutions of the test substances (100-0.01 pmol for chemicals and 100-0.1 pg/ml for other test materials) were added and incubated in triplicate. Test solutions exceeding a final test concentration of 0.05% were prepared adding DMSO as needed, Incubation was in microtitre plates at 37°C with 7% CO2 in a humidified atmosphere for 72 hr. During the last 8hr the cells were pulsed with 0.4pCi [3H]thymidine and harvested using a Titertek Cell Harvester (Flow Laboratories, Meckenheim) for the determination of [3H]thymidine incorporation, Thymidine incorporation rates were calculated as the percentage of PHA-stimulated minus unstimulated control ( = 100%).
Macrophage differentiation assay. Ficoll-Paque isolated mononuclear cells as described above were layered on a percoll-gradient according to the method of Zwadlo et al. (1986). The monocyte fraction was isolated, washed three times with McCoy's 5A medium and plated into fiat-bottomed microtitre plates to a final concentration of 5 x 105 cells/well in 200/~1 McCoy's 5A medium with 20% inactivated human serum. The cells were incubated for 96 hr at 37°C and 7% CO2 with or without varying concentration ranges of test substances. Test solutions were prepared as described above. Afterwards cells were washed with PBS and fixed with ice-cold fixing solution (1% acetic acid in aqueous ethanol [40%]) for 10 min. The fixed cells were stained by indirect immunoperoxidase staining with monoclonal antibodies 27E10 and RM 3/1 (Dianova, Hamburg) as described by Zwadlo et aL (1986 and 1987). Cells were stained with 3-amino-9-ethylcarbazol and analysed for antigen expression under the microscope. The visual scoring was as follows: 0 = no additional staining, + = defined and slightly intensifted staining, ÷ + = moderately intensified staining, + + + =strongly, and + + + + =very highly intensified staining compared with the untreated control. LPS-activated macrophages were used as a positive control. Human patch test. The backs of 20 healthy volunteers were exposed to 100 p l undiluted test samples for 24 hr under occlusion using patch test plasters (Leukotest, BDF Hamburg). 48 hr after application the test areas were inspected and skin reactions estimated according to the scoring system of Patrick and Maibach (1991): 0=negative, normal skin; + = questionable erythema not covering entire area; 1 = definite erythema; 2 = erythema and induration; 3 = vesiculation; 4 = bullous reaction. Statistical analysis. Spearman's rank correlation was performed according to the method described by Sachs (1984). The levels of significance (ct) were related to the correlation coefficient as follows (n = 12): r~> 0.4965 (ct = 0.050), rs > 0.6713 (~t =0.010), and r~>0.8182 (~=0.001). Results and Discussion
The modulating effects of 10 drugs and anti-inflammatory agents on PBMNCs were studied at concentrations between 0.1 and 100 nmol/ml. Typical results at a concentration of 1.0 nmol/ml are summarized in Table 1. At non-cytotoxic levels substances known to be immunosuppressive in vivo like cyclosporin A (CsA), dexamethasone, hydrocortisone, and chloroquine showed clearly detectable reductions of the maximum response of PBMNCs induced by 2.5/~g PHA/ml. Acetylsalicylic acid reduces T-cell proliferation to a level of about 80% over the whole range of concentrations tested, whereas CsA, like the corticosteroids, suppressed the T-cell responses effectively even at concentrations below 1 nmol/ml.
lmmunocompetent cells in skin irritation
391
Table I. Immunmodulationinduced by anti-inflammatoryand pro-inflammatoryagentsand drugs in the LTT and the MDA MDA Test material (1.0 nmol/ml) Cyclosporin A (0.6 nmol/ml)* Dexamethasone* Hydrocortisone* Chloroquine* Cyclophosphamide* Therafectin (92 nmol/ml) Therafectin (46 nmol/ml) lndomethacin Acetylsalicylicacid Dithranol Retinoic acid, trans
PHA-LTT (%)t 15.9 23.9 33.7 51.9 82.4 4.6 87.5 92.4 78.5 82.5 70.1
RM 3/1:~ + + + + + + + + 0 0 nd + + + + + 0 0
27EI0~ 0 0 0 + 0 nd + 0 + + + + + + + +
nd= not done *Substances known to be immunosuppressivein vivo. tPercentage of PHA-induced lymphocyteproliferation (2.5 #g PHA/ml). ~:RM 3/1 antibody binding to surface marker seen on macrophagesduring recovery. §27EI0 antibody binding to early-stageinflammationmarker on macrophages. All values are means of triplicates.
Therafectin, an isopropylidene-dimethylaminopropylglucofuranose, described as a T-cell suppressing agent required a higher concentration to down-regulate T-cell proliferation, Cyclophosphamide, although known to have an immunosuppressive effect after metabolic transformation in vivo neither suppressed the mitogen-induced T-cell response nor stimulated macrophages in vitro to express the .corresponding surface antigen detectable by antibody RM3/1. The same 10 agents were examined in the M D A for the expression o f two differentiation markers as described and characterized by Zwadlo et al. (1986 and 1987). M a c r o p h a g e subsets found in acute inflamed dermal tissues o f h u m a n patients (Czarnetzki et al., 1990) were detected by means o f 27E10 m o n o clonal antibody, directed against a protein expressed on their surface. It has been reported that macrophages can be detected in recovering or healing tissue using monoclonal antibody R M 3 / I . Immunosuppressive drugs like CsA, dexamethasone, hydrocortisone, and Therafectin induced the expression o f the down-regulatory inflammation marker detected by the antibody R M 3 / I . Although
there was a reduced T-cell blastogenic reaction to chloroquine in the LTT, macrophages did not respond. Drugs thought to have a pro-inflammatory effect, such as dithranol and retinoic acid, activated macrophages to give a strong expression o f the early-stage inflammatory marker. However, there was no expression o f the second marker and the in vitro lymphocyte responses were not affected. Both substances are k n o w n to evoke primary skin reactions. Based on these observations we c o m p a r e d visual estimations o f dermal erythema induced during epicutaneous patch testing o f 12 selected cosmetic emulsions to evaluate how well the expression o f the macrophage markers correlates with dermal responsiveness. The m e a n values o f the dermal erythema scores ranking from non-irritant to moderately irritant were c o m p a r e d with estimated intensities o f staining o f the expressed early-stage inflammation marker. Results shown in Table 2 reveal an acceptable correlation o f the intensity o f erythema with the expression o f the acute inflammation marker as analysed by rank correlation according to Spearman (Sachs, 1984). The
Table 2. Comparisonof macrophagedifferentiationassay in vitro with epicutaneouspatch test results MDAt Test product Epicutaneous type reactions* RM3/I 27E10 Emollient cream 0.0 0/+ 0/+ Body care lotion 0.06 + 0/+ Moisturizing cream 0.06 + 0 After sun body lotion 0.13 + + Hand and body lotion 0.19 + +/+ + Cleansing face milk 0.50 0 0 Facial make-up remover 0.50 + 0/+ Moisturizing lotion 0.75 + + + Cleansing lotion 1.50 + + + + Cleansing cream 1.56 + + +/+ + + Skin clearing milk 1.75 + +/+ + Cleansing paste 2.00 0/+ + + *Original concentration 100 #l/test area; mean values of 48-hr scores. tTest concentration0.01 (g/litre)dissolvedin DMSO (0.05% final concn); 27EI0 antibody binding to early-stage inflammation marker; RM3/1 antibody binding to surface marker seen during recovery; all values are means from triplicates. TIV 7/4---G
W . J . W . PAPE et al.
392
Table 3. Reactivity of macrophages and PHA-triggered lymphocytes to test products from the CTFA study (phase 2) MDA* PHA-LTTt CTFA test sample (No.) 27EI0 RM3/I (%) Emollient cream 5 0 0/+ 91.7 Cuticle remover 6 + + + 0 86.4 Ointment 0 % SDS) 7 + + 0/+ 78.3 Cleansing cream II 10 + + 0/+ 105.0 Ointment (2% SDS) 14 + + 0 86.8 Roll-on antiperspirant 18 + + + + 0 145.1 *Test concentration 0.01 (g/litre) dissolved in DMSO (0.05% final conch); 27E10 antibody binding to early-stage inflammation marker. $Percentages of maximum stimulation induced by 2.5 #g PHA/ml; test concentration as above. All values are means of triplicates. correlation coefficient r s was 0.7138 (~t < 0.010, n = 12). There was n o correlation with the m a r k e r expressed during d o w n regulation of the inflammatory process ( R M 3 / I ) . T-lymphocyte proliferation was n o t affected by any o f these test materials (data n o t shown). Finally, we have studied six cosmetic f o r m u l a t i o n s f r o m the " C T F A E v a l u a t i o n o f Alternatives Prog r a m " (phase 2) t h a t have given inconsistent results in different in vitro irritation tests (Gettings et al., 1991; Pape a n d Hoppe, 1991). Table 3 shows t h a t the L T T results were hardly affected by samples 5, 6, 7, 10 a n d 14. W i t h the exception o f sample 5 in M D A these p r o d u c t s showed a n up-regulated expression o f the early-stage of inflammation marker, which indicates a n irritation potential. In the LTT, sample 18 represented a strong dose-dependent increase o f 4 5 % relative to the P H A - i n d u c e d effect alone. This stimulatory response c o r r e s p o n d s to the surprisingly high expression o f the early-stage i n f l a m m a t i o n m a r k e r o n macrophages, suggesting t h a t the roll-on antiperspir a n t m i g h t contain a n i m m u n o s t i m u l a t i n g agent. Moreover, these findings were s u p p o r t e d by the dermal reactions observed after accidental skin contact during experimentation. P W M - i n d u c e d Blymphocyte proliferation was not influenced. In addition, there was n o induction of the staining o f m a c r o p h a g e s with a n t i b o d y R M 3 / 1 . F u r t h e r examination o f test m e t h o d s using i m m u n o c o m p e t e n t cells are presently in progress to evaluate the usefulness o f such in vitro a p p r o a c h e s for predicting
dermal irritation. REFERENCES Bos J. D. and Kapsenberg M. L. (1986) The skin immune
system. Its cellular constituents and their interaction. ImmunologyToday 7, 235-239. Czarnetzki B. M., Zwadlo-Klarwasser G., Br6cker E.-B. and Sorg C. (1990) Macrophage subsets in different types of urticaria. Archives of Dermatological Research 282, 93-97. Gettings S. D., Teal J. J., Bagley D. M., Demetrulias J. L., DiPasquale L. C., Hintze K. L., Rozen M. G., Weise S. L , Chudkowsky M., Marenus K. D., Pape W. J. W., Roddy M., Schnetzinger R., Silber P. M., Glaza S. M. and Kurtz P. J. (1991) The CTFA Evaluation of Alternatives Program: an evaluation of in vitro alternatives to the Draize primary eye irritation test. (Phase 1) Hydroalcoholic formulations; (part 2) data analysis and biological significance. In Vitro Toxicology 4, 247-288. Hume D. A. and Weidemann M. J. (1980) Mitogenic lymphocyte transformation. In Methods in Enzymology. Vol. 150. Elsevier, New York. Pape W. J. W. and Hoppe U. (1991) In vitro methods for the assessment of primary local effects of topically applied preparations. Skin Pharmacology 4, 205-212. Patrick E. and Maibach H. I. (1991) Predictive skin irritation tests in animals and human. In Dermatotoxicology. By G. Di Sabato. Vol. 150.4th Ed. Edited by F. N. Marzulli and H. I. Maibach. pp. 201-222, Hemisphere Publ. Corp. NY. Sachs L. (1984) Angewandte Statistik--Statistiche Methoden und l h r e Anwendung. 6th Ed. pp. 309-312. Springer-Verlag, Berlin. Steer S., Lasek W., Clothier R. H. and Balls M. (1990) An in vitro test for immunomodulators? Toxicology in Vitro 4, 360-362. Streilein J. W. (1983) Skin associated lymphoid tissue (SALT): origins and function. Journal of Investigative Dermatology80 (suppl.) 125-165. Zwadlo G., Schlegel R. and Sorg C. (1986) A monoclonal antibody to a subset of human monocytes found only in the peripheral blood and inflammatory tissues. Journal of Immunology137, 512-518. Zwadlo G., Voegeli R., Schulze-Osthoff K. and Sorg C. (1987) A monoclonal antibody to a novel differentiation antigen on human macrophages associated with downregulatory phase of the inflammatory process. ExperimentalCell Biology 55, 295-304.