- Email: [email protected]
Comparative morphofunctional study of dispersed mature canine cutaneous mast cells and BR cells, a poorly differentiated mast cell line from a dog subcutaneous mastocytoma
Veterinary Immunology and Immunopathology 62 Ž1998. 323–337
Comparative morphofunctional study of dispersed mature canine cutaneous mast cells and BR cells, a poorly differentiated mast cell line from a dog subcutaneous mastocytoma Gloria Garcıa ´ a, Pilar Brazıs ´ a, Natalia Majo´ b, Luis Ferrer b, Fernando de Mora a,1, Anna Puigdemont a,),1 b
a Departament de Farmacologia i Terapeutica UniÕersitat Autonoma de Barcelona, Barcelona, Spain ` ` Departament de Patologia i Produccio´ Animals, UniÕersitat Autonoma de Barcelona, Barcelona, Spain `
Accepted 18 November 1997
Abstract The dog mastocytoma BR cell line provides us with a permanent source of canine mast cells, allowing a characterization of secretory mediators that exert important effects in canine allergic and nonallergic diseases and in physiological processes. We studied the ultrastructural characteristics and histamine releasing activity after immunological and non-immunological stimuli of the dog mastocytoma BR cell line, and compared the cell line to normal skin mast cells enzymatically isolated from healthy dogs. The histamine content of BR cells was 0.04 " 0.002 pgrcell, approximately 100-fold less than that found in canine skin mast cells. Non-immunologic stimuli induced similar concentration-dependent histamine release from skin mast cells and BR cells: 29.3 " 0.9% vs. 12.7 " 0.7% Žcalcium ionophore A23187., 23.3 " 0.7% vs. 18.8 " 0.7% Žsubstance P. and 12.5 " 0.3% vs. 12.1 " 0.9% Žcompound 48r80., respectively. Immunologic stimulation, however, was only effective on canine skin mast cells, causing 30.9 " 1.7%, 27.7 " 0.6% and 12.2 " 0.9% histamine release in response to anti-canine IgE, concanavalin A, and antigen Asc S 1, respectively. The absence of functional IgE receptors in BR cells was
AbbreÕiations: MC, mast cells; Con A, concanavalin A; OPT, o-phthaldialdehyde; TPA, 12-O-tetradecanoylphorbol-13-acetate; MEM, Minimum essential medium; DMEM, Dulbecco’s modified eagle medium; X PBS, Phosphate buffered saline; HEPES, N-2-hydroxyethyl-piperazine-N -2-ethanesulphonic acid; FCS, foetal calf serum; HBSS, Hanks’ balanced salt solution; SP, substance P; TCA, trichloroacetic acid; IgE, immunoglobulin E; PKC, protein kinase C; Fc ´ RI, high-affinity IgE receptor ) Corresponding author. Facultat de Veterinaria, Universitat Autonoma de Barcelona, 08193 ` BellaterrarBarcelona, Spain. Fax: q34 3 581 2006; e-mail: [email protected] 1 This author has equally contributed to the elaboration of this project. 0165-2427r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 2 4 2 7 Ž 9 7 . 0 0 1 6 6 - 9
324
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
confirmed by the lack of response to anti-IgE and antigen Asc S 1 following passive sensitization with dog atopic serum and dog antigen sensitized serum. We conclude that BR cells are able to release histamine after non-immunologic stimulation in a similar manner to canine skin mast cells, but that there are morphological and functional differences possibly due to different states of maturity or differentiation. For this reason the study of the highly homogeneous BR cells could offer insights into dog mast cell biology in contexts where freshly isolated cells cannot be used because of low purity and recovery. q 1998 Elsevier Science B.V. Keywords: Skin mast cells; BR cells; Dog; Histamine
1. Introduction Canine cutaneous mast cells ŽMC. are considered key immune effector cells in the pathogenesis of canine atopic dermatitis, as suggested by increased skin MC number in biopsy specimens of lesions from dogs with atopic dermatitis, increased cutaneous histamine levels in lesional skin, and enhanced releasability both in vivo and in vitro of cutaneous MC from atopic dogs ŽTurner et al., 1989; DeMora et al., 1996.. We have studied skin MC isolated from normal and atopic dogs ŽDeMora et al., 1996. and evaluated the ability of several compounds to modulate MC histamine release ŽGarcıa ´ et al., 1997a,b., finding that while the dispersion of canine skin MC is a valuable tool, it nevertheless has limitations due to low recovery and low purity ŽDeMora et al., 1993.. The analysis of non-specific MC mediators that participate in the pathophysiology of dermal allergies and that might be released by other cells ŽGalli and Costa, 1995. requires 100% pure populations only obtainable through cell lines. In vitro dog MC lines maintained in continuous culture provide a large number of pure cells and allow the characterization of a broader range of MC secretory mediators. In recent years, Gold and colleagues established 4 cell lines derived from dog subcutaneous mastocytomas ŽC 2 , C 1 , G and BR. propagated subcutaneously in athymic mice ŽLazarus et al., 1986; DeVinney and Gold, 1990.. Because cell lines are functionally heterogeneous, specific lines, such as BR, can be studied to further our understanding of canine MC biology ŽThomas et al., 1991.. To date BR cell line has mainly been studied to characterize protease content from freshly isolated mastocytoma ŽCaughey et al., 1987. but as yet no histamine release or ultrastructural studies have been performed with cells passaged in long-term in vitro culture for comparison with normal canine skin MC. In the current work, we compare the ultrastructural characteristics and the histamine release activity of BR cells and cutaneous MC enzymatically isolated from healthy dogs after activation by immunological and non-immunological stimuli. 2. Materials and methods 2.1. Materials The following chemicals and enzymes were purchased from Sigma Chemical ŽSt. Louis, MO, USA.: calcium ionophore A23187, concanavalin A Žtype IV. ŽCon A.,
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
325
substance P ŽSP., compound 48r80, 12-O-tetradecanoylphorbol-13-acetate ŽTPA., ophthaldialdehyde ŽOPT., histamine diphosphate salt, histaminase, collagenase Žtype I., hyaluronidase Žtype I-S., protease Žpronase E, type XIV. and bovine albumin Žfraction V.. Penicillin-streptomycin, Eagle’s minimum essential medium ŽMEM. containing 10% foetal calf serum ŽFCS., Dulbecco’s modified eagle medium ŽDMEM., Hanks’ balanced salt solution ŽHBSS. containing N-2-hydroxyethyl-piperazine-N X-2-ethanesulphonic acid ŽHEPES. and 2% FCS, Phosphate buffered saline ŽPBS., glutamine, histidine, fungizone and gentamicin were obtained from Gibco ŽIzasa, Barcelona, Spain.. Spurr resin was purchased from Tash Laboratories Equipment ŽUK.. Anti-canine IgE was kindly supplied by Dr. D.J. DeBoer ŽUniversity of Wisconsin, Madison, USA. ŽDeBoer et al., 1993.. Antigen Asc S 1 from Ascaris suum ŽGreer Lab., Lenoir, NC, USA.. The BR canine mastocytoma cell line was kindly provided by Dr. W.M. Gold ŽCardiovascular Research Institute, University of California, San Francisco, CA, USA.. 2.2. Isolation of canine cutaneous MC Skin biopsies from healthy dogs were obtained at the Barcelona animal pound ŽCentre de Zoonosi., where dogs are sacrificed twice a week, for reasons related to city sanitation policy. The dogs were kept under observation by a veterinarian and any dog showing signs of disease was withdrawn from the study. Skin samples were obtained from the abdominal craniolateral region of healthy dogs, and the MC were isolated by an enzymatic procedure developed in our department and described elsewhere ŽDeMora et al., 1993.. Briefly, after subcutaneous fat tissue was removed, the skin was chopped into approximately 0.5–1 mm3 fragments, washed twice with MEM containing 10% FCS and incubated Ž180 min, 378C. in an enzyme mixture Ž34.5 mg collagenase, 18 mg hyaluronidase and 12 mg protease. supplemented with bovine albumin and antibiotics. Cells dispersed by this procedure were separated from undissociated tissue by filtration, washed once with MEM and incubated for 16 h at 378C with enriched DMEM in a humidified atmosphere with 5% CO 2 . Later, cells were washed twice with Ca2q- and Mg 2q-free HBSS ŽCMF-HBSS.. MC numbers was assessed in a Neubaeur hemocytometer after a Kimura metachromatic stain ŽKimura et al., 1973.. Cell viability was checked with trypan blue dye and was always greater than 95%. Before activation for histamine release, cells were resuspended in an appropriate volume of prewarmed HBSS with Ca2q and Mg 2q. 2.3. BR cell culture BR mastocytoma cells were seeded in 75-cm2 tissue culture flasks at a density of 10 6 cellsrml of enriched DMEM supplemented with 2 mM glutamine, 1.6 mM histidine, 25 mM HEPES ŽpH 7.4., 100 Urml penicillin, 100 m grml streptomycin, 2.5 m grml fungizone and 5% FCS. Cells were incubated at 378C in a humidified atmosphere with 5% CO 2 and fresh medium was added every 48 h. Cells maintained in this medium were split 1:4 and the cultured medium was changed at weekly intervals. BR cells were serially passaged to obtain the required number of cells.
326
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
Before activation, cells were washed twice in Ca2q- and Mg 2q-free PBS ŽCMF-PBS. at 238C. Cell number and viability were assessed by applying trypan blue stain. Viability was always greater than 95%. Finally, BR cells were resuspended in an appropriate volume of prewarmed PBS with Ca2q and Mg 2q before experimental use. 2.4. Canine skin MC and BR cell actiÕation To assess anti-canine IgE and antigen Asc S 1 induced histamine release, BR cells and dispersed skin MC were passively sensitized by incubation for 2 h at 378C in MEM containing 15% serum from either atopic dogs or Ascaris suum spontaneously sensitized dogs to achieve 100% occupancy of IgE receptors. Before stimulation, sensitized cells were washed twice in an appropriate buffer. Isolated canine skin MC and BR cells were distributed at a rate of 15 000 and 2 = 10 6 cells per 180 m l and stimulated by adding 20 m l of each stimulus Žanti-canine IgE, antigen Asc S 1, Con A, SP, calcium ionophore A23187, compound 48r80 or TPA. at increasing concentrations. Activation was halted at 08C and after centrifugation at 2630 rcf for 20 min at a 48C, the supernatant was collected. Twenty microliters of 55% trichloroacetic acid ŽTCA. was added to 180 m l of the supernatant to precipitate proteins before the histamine assay was performed. All conditions were performed in duplicate. Spontaneous histamine release was measured in samples incubated in the absence of any stimulus. Total histamine per MC was calculated by adding the histamine content remaining in the pellet after centrifuging to the histamine present in the supernatant and dividing by the number of cells. The cytotoxicity of each stimulus at different concentrations was assessed by trypan blue exclusion. 2.5. Histamine assay The classic spectrofluorimetric method of Shore et al. Ž1959., with the modifications of Bergendorf and Uvnas ¨ Ž1972., was used. One hundred and seventy microliters of the supernatant was taken and centrifuged Ž2000 rcf, 2 min. in the presence of 10% TCA. The medium was made alkaline Žto pH 12.5. with 0.3 M NaOH and the histamine fluorophore derivative was formed by incubating at room temperature for 4 min with OPT Ždiluted to 10 mgrml in methanol.. The histamine-OPT condensation reaction was halted at pH 3.5 by the addition of 6 N HCl. The spectrofluorometric reading conditions were excitation, 350 nm, and emission, 437 nm. At the concentrations used, none of the stimuli interfered with the fluorometric histamine assay. The presence of histamine in the samples was confirmed by degradation with histaminase Ž0.30 mgrml. for 3 h at 378C. Additionally, the histamine assay proved to have an excellent recovery rate Ž98%. and on 2- and 10-fold dilutions gave readings for histamine that were inversely proportional to the dilution rate. 2.6. Morphological study Stimulated and non-stimulated canine skin MC and BR cultured cells were treated identically in order to study their ultrastructure by transmission electron microscopy.
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
327
Canine skin MC were previously purified using a discontinuous Percoll gradient Ž40, 60 and 80%.. Freshly disgregated skin MC and cultured BR cells suspended in PBS were fixed in an equal volume of the fixative for 4 h at 48C. The fixative contained 5% glutaraldehyde in 0.1 M Sorensen’s phosphate buffer. Samples were then postfixed in 1% osmium tetroxide in 0.1 M Sorensen’s phosphate buffer for 1 h at 48C. After repeated washes with the same buffer, samples were dehydrated using a graded series of ethanol and embedded in Spurr resin following conventional procedures. Before every change of liquid or resin, suspensions were centrifuged for 10 min and the supernatant was removed. Sections 1 m m thick were stained with 1% toluidine blue for examination under a light microscope. Areas containing a higher MC density were selected for ultrathin sectioning. Sections 40 to 60 nm thick were cut using glass knives on a NOVA ultramicrotome and collected on 300 mesh copper grids. Thereafter, the sections were stained 10 min with lead citrate and examined using a Hitachi 7000 transmission electron microscope. 2.7. Data analysis Results are expressed as the arithmetical means " standard error Žmean " SEM. of 8 or 10 experiments. Differences between means were tested for significance using a Student’s t test for paired data with a level of significance of 0.05. Net histamine release after the addition of each stimulus was expressed as a percentage of total histamine corrected for spontaneous release.
Fig. 1. Non-stimulated canine dispersed skin MC. Numerous oval or round granules can be seen in the cytoplasm. The granules are filled with a highly electron-dense, homogeneous material with little or no lucent halo. Bar s1.5 m m.
328
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
3. Results The mean histamine content of BR cells was 0.04 " 0.002 pgrcell, approximately 100-fold less than that found in canine dispersed cutaneous MC Ž4.93 " 0.39 pgrcell.. Histamine content and histamine release from BR cells remained stable over time after 20 culture passages, allowing us to study their releasability adequately.
4. Electron microscopy Examination under electron microscope demonstrated significant differences between normal canine skin MC and the BR cells. Non-stimulated canine skin MC were generally oval or round and their diameters ranged from 7.5 to 12 m m. The nucleus was oval or irregular and usually located eccentrically. Nucleoli were observed in a few cells. The plasma membranes presented some folds or villous projections and some mitochondria, abundant microfilaments and numerous granules were observed in the cytoplasm. The granules of these cells were oval or round and highly uniform in size and appearance. They ranged from 255 to 410 nm in size and were filled with a highly electron-dense, homogeneous and amorphous material with little or no lucent halo ŽFig. 1.. Stimulated canine skin MC were rounder and their average diameter was similar to that of non-stimulated canine MC. The nuclei were generally more irregular and
Fig. 2. Stimulated canine dispersed skin MC. An irregular nucleus and a very prominent nucleolus are present. The plasma membrane has numerous folds and some degranulation channels are in the cytoplasm. No cytoplasmic granules are visible. Bar s 2 m m.
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
329
Fig. 3. Non-stimulated BR cells. Few moderately electron-dense granules with a visible lucent halo Žarrowheads. are present in the cytoplasm. Bar s 2.5 m m.
Fig. 4. Stimulated BR cells. Many degranulation channels can be seen in the cytoplasm. No cytoplasmic granules are visible. Bar s1.5 m m.
330
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
centrally situated. The plasma membranes had numerous folds, and many degranulation channels lined with common membranes were observed in the cytoplasm. Most MC did not show cytoplasmic granules. However, a few cells presented scarce cytoplasmic granules, similar to those described in the non-stimulated MC ŽFig. 2.. Non-stimulated BR cells were generally round and diameters ranged from 5.8 to 10 m m. The nucleusrcytoplasm ratio was much higher in these cells than in normal MC. The nucleus was usually oval and eccentric and a nucleolus was observed in many cells ŽFig. 3.. The plasma membrane had very few irregular folds and few granules were visible in the cytoplasm. These granules were homogeneous, moderately electron-dense and lucent halos were visible ŽFig. 3.. Stimulated BR cells were also round and had an average diameter similar to that of non-stimulated cells. The ultrastructural morphology of stimulated BR cells was very similar to that of non-stimulated ones, except for the presence of some degranulation channels in the cytoplasm and the absence of granules in stimulated cells ŽFig. 4..
5. Histamine release induced by Fc ´RI-independent stimuli 5.1. Calcium ionophore A23187 Dispersed skin MC and BR cells showed concentration-dependent histamine release in the presence of A23187 ŽFig. 5A.. At a concentration of 1 m M, A23187 induced a significantly lower percentage of net histamine release from BR cells than from dispersed MC Ž12.7 " 0.7% vs. 29.3 " 0.9%. ŽFig. 5A.. Histamine release from BR cells increased to 20.6 " 1.1% when TPA at 40 nM and ionophore were added simultaneously ŽFig. 6.. The 40 nM concentration of TPA had been determined previously to be the one that gave the highest synergistic effect with A23187. 5.2. TPA TPA at concentrations of 5 to 100 nM induced slight histamine release from both MC types ŽFig. 5B. in a concentration-related way that reached a plateau at 40 to 100 nM. Release induced by 40 nM of TPA was 5.6 " 0.4% and 8.2 " 0.9% for dispersed MC and BR cells, respectively ŽFig. 5B.. There were no significant differences between the two types of cells. 5.3. Substance P Histamine release induced by SP was observed at concentrations of 0.1 to 100 m M and increased in a concentration-related manner. When higher SP concentrations were used, cytotoxicity was observed Žresults not included.. Maximum histamine release was observed at 100 m M of SP from both dispersed MC and BR cells Ž23.3 " 0.7% and 18.8 " 0.7%, respectively. ŽFig. 5C..
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
331
Fig. 5. Histamine release from dispersed skin MC and BR cells induced by A23187, SP, TPA and compound 48r80. Cells were incubated with the stimuli at 378C for 20 min. Open circles: BR cells, closed circles: dispersed skin MC. Ž ns8 or 10. aCytotoxicity was assessed Žsee Table 1.. ) P - 0.05.
Fig. 6. Synergistic effect of TPA and A23187 on histamine release from BR cells. Cells were incubated for 20 min at 378C with various concentrations of A23187 in the presence or absence of TPA Ž40 nM.. Open circles: stimulated with TPA and A23187, closed circles: stimulated with A23187 only. Ž ns8 or 10. ) P - 0.05.
332
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
Table 1 Cytotoxicity of canine skin MC and the BR cell line induced by different concentrations of compound 48r80 Mast cell type
Canine skin MC
BR cell line
Concentration of 48r80
Histamine release
Cytotoxicity
Ž m grml.
Ž%.
Ž%.
10 30 100 300 1000 3 10 30 100
12.5"0.3 13.3"0.5 14.7"0.7 15.1"0.8 23.2"1.2 6.1"0.8 12.1"0.9 31.4"1.4 49.2"1.5
0.0"0.0 0.5"0.3 3.9"0.5 9.6"1.1 30.6"2.3 0.0"0.0 0.2"0.5 10.1"0.3 34.9"1.5
Each result is the mean"S.E.M. of 8 or 10 experiments.
5.4. Compound 48 r 80 The synthetic polyamine, compound 48r80, induced histamine release from dispersed MC at concentrations of 0.1 to 10 m grml. Histamine release Ž12.5 " 0.3%. peaked at 10 m grml ŽFig. 5D. and higher concentrations led to rapid increases in cytotoxicity ŽTable 1.. In BR cells, similar histamine release Ž12.1 " 0.9% at 10 m grml. was observed after addition of the same range of stimulus concentrations. BR cells were more sensitive to the toxic effects of compound 48r80 ŽTable 1..
6. Histamine release induced by Fc ´RI-mediated stimulation 6.1. Antigen Asc S 1 Antigen Asc S 1 at concentrations of 100 to 3000 m grml induced histamine release from dispersed sensitized MC, and a maximum response of 12.2 " 0.9% was observed at a concentration of 1000 m grml. Sensitized BR cells did not release histamine upon Asc S 1 stimulation ŽFig. 7A.. 6.2. ConcanaÕalin A The lectin Con A at concentrations of 0.1 to 1000 m grml markedly released histamine from dispersed MC but no effect on BR cells was observed ŽFig. 7B.. The maximum peak of 27.7 " 0.6% was reached at 1000 m grml, and higher doses of Con A did not increase histamine release Žnot shown.. 6.3. Anti-canine IgE Histamine was released from dispersed sensitized MC in a dose-dependent manner after addition of anti-canine IgE at concentrations of 0.0001 to 0.01 m grml and the
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
333
Fig. 7. Histamine release from dispersed skin MC and BR cells induced by Antigen Asc S 1, Anti-IgE and Con A. Cells were incubated for 20 min with Con A and 30 min with Asc S 1 and Anti-IgE. Open circles: BR cells, closed circles: dispersed skin MC. Ž ns8 or 10. ) P - 0.05.
maximum effect of 30.9 " 1.7% was observed at the 0.003 m grml. In contrast, anti-IgE did not induce histamine release from sensitized BR cells at any concentration ŽFig. 7C..
7. Discussion In the present study we have found both similarities and differences between the BR cell line and isolated canine skin MC. The two cell types share similar patterns of histamine releasability after non-immunological stimulation and have some common morphological characteristics. They differ, however, in their releasability after immunological activation. Presumably, the later is due to the absence of Fc ´ RI on the BR cells surface. Both the analogies and differences provide insights on canine mast cells biology and developmental processes respectively. Besides the absence of detailed information on BR previously to the present work, this cell line offered advantages vs. similarly developed canine MC lines Že.g. C 2 .. BR are a source of non-adherent fast growing cells that have the ability to divide in FCS-free medium permitting more accurate functional studies on their capability to liberate immune mediators. The present preliminary data on
334
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
their functional and morphological characteristics are a required step to further advance in the analysis of BR cells morphofunctional characteristics under diverse conditions. Our results on non-activated canine MC ultrastructure are similar to those published by Calonico et al. Ž1985.. This resemblance is extremely interesting in the light of the slightly different experimental approaches undertaken. Calonico et al. Ž1985. reported the ultrastructural analysis of freshly isolated mastocytoma cells rather than cell lines, our data emphasize that, after propagation in mice, the mastocytoma cells structure does not significantly change and therefore that the conversion from freshly isolated tumoral cells to a cell line ŽBR. has little effect on their phenotype. Furthermore the comparison between mature dermal MC and mastocytoma cells was done with frozen sections instead of cell suspensions which might provide interesting but potentially different functional and morphological information from the cells suspensions used in the present study. In our work, canine skin MC appear mature, highly differentiated cells with cytoplasm that is occupied mainly by granules and few organelles. Unlike human MC, canine skin MC granules are electrodense and homogeneous and contain no crystals ŽWhitaker-Menezes et al., 1995.. BR cells, on the other hand, have an immature and undifferentiated appearance. The, nuclei of BR cells are richer in euchromatin, and the nucleoli are more evident than those of canine skin MC. Moreover, cytoplasmic granules are scarce, less dense and surrounded by a more electron-lucent area, indicating that these granules are less mature than those with an electron-dense homogeneous appearance without an electron-lucent halo observed in canine skin MC ŽCalonico et al., 1985.. Among the cell lines established from neoplastic mast cell lines to date ŽC 1 , C 2 , G and BR., the BR line appears to be the least differentiated. The C 1 and C 2 lines, on the other hand, are the most highly differentiated, and the differentiation of the G line is intermediate ŽLazarus et al., 1986.. Such variation is understandable given that canine mastocytomas, from which they have been derived, can be either highly or scarcely differentiated ŽPatnaik et al., 1984.. BR cell stimulation, however, leads to a similar level of degranulation than the one observed for non-neoplastic MC. The histamine content of the BR cell line remained stable over different culture passages, ensuring that the functional studies would not be altered over time. BR cell histamine content is less than that of dispersed canine skin MC and is clearly related to the low number of granules observed in their cytoplasm. This is consistent with what has been found for other mastocytoma cell lines and in immature MC Ž0.46 to 0.07 pgrcell. ŽDeVinney and Gold, 1990.. Lazarus et al. Ž1986. working with other dog mastocytoma cell lines ŽC 2 and G. suggested that differences in histamine content could reflect the neoplastic origin of the cell line and some dedifferentiation in conjunction with neoplastic transformation in the original tumor. The release of preformed mediators is one of the major functions of MC. We examined the histamine releasability of BR cultured cells for comparison with that of dispersed skin MC using a variety of non-immunological and immunological secretagogues. As expected, calcium ionophore A23187 induced a significant amount of histamine release in both cell populations. However, quantitative differences were observed between BR cells and dispersed canine MC, a finding that can be accounted for by differences in membrane composition or in sensitivity of calcium channels as suggested by Bottjer et al. Ž1994.. In the BR cells, calcium ionophore A23187 histamine ¨
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
335
release was potentiated by incubation with TPA, whereas TPA or ionophore A23187 alone induced low histamine release, indicating that protein kinase C ŽPKC. activation by TPA interacts synergistically with a rise in intracellular calcium induced by calcium ionophore A23187 to produce histamine secretion from BR cells. This result is consistent with previous reports for a variety of MC stimulated with this combination of compounds ŽCantwell and Foreman, 1986; Sekino et al., 1990.. Accumulating evidence suggests that SP and compound 48r80 cause secretion from human and canine skin MC both in vitro and in vivo ŽTausk and Undem, 1995; Garcıa ´ et al., 1997a.. Some cell lines, however, such as C 2 , fail to respond to 48r80 stimulation ŽDeVinney and Gold, 1990.. In our study of the BR line 48r80 or SP stimuli did effectively induce histamine release, supporting a hypothesis that tumoral BR cells keep the particular membrane characteristics of canine skin MC in this respect, given that SP and compound 48r80 reactivity is thought to result from a more fluid or negatively charged membrane ŽTainsh et al., 1992.. Inase et al. Ž1993. likewise described similar results for freshly isolated BR mastocytoma cells stimulated by compound 48r80 at concentrations ranging from 0.1 to 10 m grml, although they observed higher histamine release than we did at the highest stimulus concentration, while finding no cell cytotoxicity at that level of stimulation. Immunologic stimulation failed to elicit significant histamine release from BR cells with any of the three stimulus types although all three induced release from dispersed canine skin MC. These results, indicate that BR cells would be unlikely to have any functional IgE receptors or that any such receptors would be present only at low density. The absence of such receptors appears to be an intrinsic property of BR cells given that passive sensitization did not change the response to immunologic stimulation. Receptors may have been lost during the different passages in nude mice or neoplastic transformation may have taken place. It is also possible to speculate that the mastocytoma cells obtained directly from the dog were also refractory to sensitization. Lazarus et al. Ž1986. reported heterogeneity in ragweed-pollen extract histamine release after passive sensitization with IgE-rich serum for 1 h, demonstrating that the C 2 cell line had functional IgE receptors and suggesting that the G cell line may have lost its IgE receptors as a result of neoplastic transformation leading to their inability to react passively with IgE. In agreement with data obtained by other investigators histamine release after anti-canine IgE and antigen Asc S 1 stimulation never occurred without preincubation with IgE ŽMita et al., 1995; Tunon de Lara et al., 1995.. Previous results from our laboratory showed that passive sensitization with IgE-enriched serum is not absolutely necessary for inducing histamine release from dispersed canine skin MC upon stimulation with Con A ŽDeMora et al., 1996; Garcıa ´ et al., 1997a.. However, the lack of Fc ´ RI receptors in the BR cell line produced a lack of response with Con A stimulation. A possible hypothesis, to explain these results, it is that Con A probably act by cross-link Fc ´ RI receptors direct rather than by a cross-link Fc region from IgE, as has been suggested by other authors ŽCarnuccio et al., 1989.. In conclusion, BR cells provide a pure and abundant source of canine MC for the study of mediators released by these cells through non-immunologic mechanisms, and this cell line is indeed a useful tool for analyzing non-specific-MC mediators and signal transduction pathway mechanisms. Furthermore, as BR cells show an array of functional
336
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
and morphological features typical of immature MC populations, the undifferentiated nature of this cell line may make it useful for studying certain aspects of MC differentiation. Acknowledgements Dog skin biopsies were kindly provided from the municipal pound ŽCentre de Zoonosi. of Barcelona ŽSpain. from dogs that had been sacrificed following city sanitation policy. This work was supported by a grant from the Direccion ´ General de Investigacion y Tecnologıa ´ Cientıfica ´ ´ of Spain ŽDGICYT, PB93-0905.. References Bergendorf, A., Uvnas, ¨ B., 1972. Storage of 5-hydroxytryptamine in rat mast cells. Evidence for an ionic binding to carboxyl groups in a granule heparin-protein complex. Acta Physiol. Scand. 84, 320–331. Bottjer, J., Amon, V., Wolff, H.H., 1994. Functional comparison of different histamine-containing IgE-recep¨ tor positive cells. Agents and Actions, Vol. 41, Special Conference Issue: C28–C29. Calonico, L.D., Phillips, M.J., McDonald, D.M., Gold, W.M., 1985. An ultrastructural analysis of dog mastocytoma cells and normal mast cells. Anat. Rec. 212, 399–407. Cantwell, M.E., Foreman, J.C., 1986. Characteristics of the effect of phorbol ester on rat mast cells: interaction with anti-IgE. Agents and Actions 18 Ž1r2., 77–80. Carnuccio, R., Di Rosa, M., Ialenti, A., Iuvone, T., Sautebin, L., 1989. Selective inhibition by vasocortin of histamine release induced by dextran and concanavalin A from rat peritoneal cells. Br. J. Pharmacol. 98 Ž1., 32–34. Caughey, G.H., Lazarus, S.C., Viro, N.F., Gold, W.M., Nadel, J.A., 1987. Tryptase and chymase: comparison of extraction and release in two dog mastocytoma lines. Immunology 63, 339–344. DeBoer, D.J., Ewing, K.M., Schultz, K.T., 1993. Production and characterization of mouse monoclonal antibodies directed against canine IgE and IgG. Vet. Immunol. Immunopathol. 37, 183–199. DeMora, F., Garcıa, ´ G., Ferrer, L., Arboix, M., 1993. Canine cutaneous mast cells dispersion and histamine secretory characterization. Vet. Immunol. Immunopathol. 39, 421–429. DeMora, F., Garcıa, ´ G., Puigdemont, A., Arboix, M., Ferrer, L., 1996. Skin mast cells releasability in dogs with atopic dermatitis. Inflamm. Res. 45, 423–427. DeVinney, R., Gold, W.M., 1990. Establishment of two dog mastocytoma cell lines in continuous culture. Am. J. Respir. Cell Mol. Biol. 3, 413–420. Galli, S.J., Costa, J.J., 1995. Mast-cell-leukocyte cytokine cascades in allergic inflammation. Allergy 50, 851–862. Garcıa, ´ G., DeMora, F., Ferrer, L., Puigdemont, A., 1997a. Effect of H 1-antihistamines on histamine release from dispersed canine cutaneous mast cells. Am. J. Vet. Res. 58 Ž3., 293–297. Garcıa, ´ G., DeMora, F., Ferrer, L., Puigdemont, A., 1997. Inhibition of histamine release from dispersed canine skin mast cells by cyclosporin A, rolipram and salbutamol, but not by dexamethasone or sodium cromoglycate. Vet. Dermatol., Žin press.. Inase, N., Schreck, R.E., Lazarus, S.C., 1993. Heparin inhibits histamine release from canine mast cells. Am. J. Physiol. 264, L387–L390, ŽLung Cell. Mol. Physiol. 8.. Kimura, I., Moritani, Y., Tanizaki, Y., 1973. Basophils in bronchial asthma with reference to reagin type allergy. Clin. Allergy 3, 195–202. Lazarus, S.C., DeVinney, R., Laurence, J.M., Walter, E.F., Darlene, J.E., Gold, W.M., 1986. Isolated canine mastocytoma cells: propagation and characterization of two cell lines. Am. J. Physiol. 251, C935–C944, ŽCell Physiol. 20.. Mita, H., Yasueda, H., Ishii, T., Akiyama, K., 1995. IgE-mediated basophil releasability is influenced by intrinsic factors and by IgE on the cell surface. Allergy 50, 952–958.
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
337
Patnaik, A.K., Ehler, W.J., MacEwen, E.G., 1984. Canine cutaneous mast cell tumor: morphological grading and survival in 83 dogs. Vet. Pathol. 21, 469–474. Sekino, H., Uchida, M.K., Suzuki-Nishimura, T., 1990. The modulatory effects of phorbol esters on histamine release from mast cells induced by a polyethylenimine and a polyallylamine. Gen. Pharmacol. 21, 535–539. Shore, P.A., Burkhalter, A., Cohn, V., 1959. A method for the fluorometric assay of histamine in tissues. J. Pharm. Exp. Ther. 127, 182–186. Tainsh, K.R., Liu, W.L., Lau, H.Y., Cohen, J., Pearce, F.L., 1992. Mast cell heterogeneity in man: unique functional properties of skin mast cells in response to a range of polycationic stimuli. Immunopharmacology 24, 171–180. Tausk, F., Undem, B., 1995. Exogenous but not endogenous substance P releases histamine from isolated human skin fragments. Neuropeptides 29, 351–355. Thomas, P.S., Schreck, R.E., Ruoss, S.J., Lazarus, S.C., 1991. Heterogeneity of intact granules purified from canine mastocytoma cell lines. Am. J. Physiol. 260, L153–L160, ŽLung Cell Mol.Physiol. 4.. Tunon de Lara, J.M., Okayama, Y., Savineau, J.-P., Marthan, R., 1995. IgE-induced passive sensitization of human isolated bronchi and lung mast cells. Eur. Respir. J. 8, 1861–1865. Turner, C.R., Darowski, M.J., Sampson, H.A., Spannahake, E.W., Hirshman, C.A., 1989. Dermal mast cell releasability and end organ responsiveness in atopic and nonatopic dogs. J. Allergy Clin. Immunol. 83, 643–648. Whitaker-Menezes, D., Schechter, N.M., Murphy, G.F., 1995. Serine proteinases are regionally segregated within mast cell granules. Lab. Invest. 72 Ž1., 34–41.
Comparative morphofunctional study of dispersed mature canine cutaneous mast cells and BR cells, a poorly differentiated mast cell line from a dog subcutaneous mastocytoma Gloria Garcıa ´ a, Pilar Brazıs ´ a, Natalia Majo´ b, Luis Ferrer b, Fernando de Mora a,1, Anna Puigdemont a,),1 b
a Departament de Farmacologia i Terapeutica UniÕersitat Autonoma de Barcelona, Barcelona, Spain ` ` Departament de Patologia i Produccio´ Animals, UniÕersitat Autonoma de Barcelona, Barcelona, Spain `
Accepted 18 November 1997
Abstract The dog mastocytoma BR cell line provides us with a permanent source of canine mast cells, allowing a characterization of secretory mediators that exert important effects in canine allergic and nonallergic diseases and in physiological processes. We studied the ultrastructural characteristics and histamine releasing activity after immunological and non-immunological stimuli of the dog mastocytoma BR cell line, and compared the cell line to normal skin mast cells enzymatically isolated from healthy dogs. The histamine content of BR cells was 0.04 " 0.002 pgrcell, approximately 100-fold less than that found in canine skin mast cells. Non-immunologic stimuli induced similar concentration-dependent histamine release from skin mast cells and BR cells: 29.3 " 0.9% vs. 12.7 " 0.7% Žcalcium ionophore A23187., 23.3 " 0.7% vs. 18.8 " 0.7% Žsubstance P. and 12.5 " 0.3% vs. 12.1 " 0.9% Žcompound 48r80., respectively. Immunologic stimulation, however, was only effective on canine skin mast cells, causing 30.9 " 1.7%, 27.7 " 0.6% and 12.2 " 0.9% histamine release in response to anti-canine IgE, concanavalin A, and antigen Asc S 1, respectively. The absence of functional IgE receptors in BR cells was
AbbreÕiations: MC, mast cells; Con A, concanavalin A; OPT, o-phthaldialdehyde; TPA, 12-O-tetradecanoylphorbol-13-acetate; MEM, Minimum essential medium; DMEM, Dulbecco’s modified eagle medium; X PBS, Phosphate buffered saline; HEPES, N-2-hydroxyethyl-piperazine-N -2-ethanesulphonic acid; FCS, foetal calf serum; HBSS, Hanks’ balanced salt solution; SP, substance P; TCA, trichloroacetic acid; IgE, immunoglobulin E; PKC, protein kinase C; Fc ´ RI, high-affinity IgE receptor ) Corresponding author. Facultat de Veterinaria, Universitat Autonoma de Barcelona, 08193 ` BellaterrarBarcelona, Spain. Fax: q34 3 581 2006; e-mail: [email protected] 1 This author has equally contributed to the elaboration of this project. 0165-2427r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 2 4 2 7 Ž 9 7 . 0 0 1 6 6 - 9
324
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
confirmed by the lack of response to anti-IgE and antigen Asc S 1 following passive sensitization with dog atopic serum and dog antigen sensitized serum. We conclude that BR cells are able to release histamine after non-immunologic stimulation in a similar manner to canine skin mast cells, but that there are morphological and functional differences possibly due to different states of maturity or differentiation. For this reason the study of the highly homogeneous BR cells could offer insights into dog mast cell biology in contexts where freshly isolated cells cannot be used because of low purity and recovery. q 1998 Elsevier Science B.V. Keywords: Skin mast cells; BR cells; Dog; Histamine
1. Introduction Canine cutaneous mast cells ŽMC. are considered key immune effector cells in the pathogenesis of canine atopic dermatitis, as suggested by increased skin MC number in biopsy specimens of lesions from dogs with atopic dermatitis, increased cutaneous histamine levels in lesional skin, and enhanced releasability both in vivo and in vitro of cutaneous MC from atopic dogs ŽTurner et al., 1989; DeMora et al., 1996.. We have studied skin MC isolated from normal and atopic dogs ŽDeMora et al., 1996. and evaluated the ability of several compounds to modulate MC histamine release ŽGarcıa ´ et al., 1997a,b., finding that while the dispersion of canine skin MC is a valuable tool, it nevertheless has limitations due to low recovery and low purity ŽDeMora et al., 1993.. The analysis of non-specific MC mediators that participate in the pathophysiology of dermal allergies and that might be released by other cells ŽGalli and Costa, 1995. requires 100% pure populations only obtainable through cell lines. In vitro dog MC lines maintained in continuous culture provide a large number of pure cells and allow the characterization of a broader range of MC secretory mediators. In recent years, Gold and colleagues established 4 cell lines derived from dog subcutaneous mastocytomas ŽC 2 , C 1 , G and BR. propagated subcutaneously in athymic mice ŽLazarus et al., 1986; DeVinney and Gold, 1990.. Because cell lines are functionally heterogeneous, specific lines, such as BR, can be studied to further our understanding of canine MC biology ŽThomas et al., 1991.. To date BR cell line has mainly been studied to characterize protease content from freshly isolated mastocytoma ŽCaughey et al., 1987. but as yet no histamine release or ultrastructural studies have been performed with cells passaged in long-term in vitro culture for comparison with normal canine skin MC. In the current work, we compare the ultrastructural characteristics and the histamine release activity of BR cells and cutaneous MC enzymatically isolated from healthy dogs after activation by immunological and non-immunological stimuli. 2. Materials and methods 2.1. Materials The following chemicals and enzymes were purchased from Sigma Chemical ŽSt. Louis, MO, USA.: calcium ionophore A23187, concanavalin A Žtype IV. ŽCon A.,
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
325
substance P ŽSP., compound 48r80, 12-O-tetradecanoylphorbol-13-acetate ŽTPA., ophthaldialdehyde ŽOPT., histamine diphosphate salt, histaminase, collagenase Žtype I., hyaluronidase Žtype I-S., protease Žpronase E, type XIV. and bovine albumin Žfraction V.. Penicillin-streptomycin, Eagle’s minimum essential medium ŽMEM. containing 10% foetal calf serum ŽFCS., Dulbecco’s modified eagle medium ŽDMEM., Hanks’ balanced salt solution ŽHBSS. containing N-2-hydroxyethyl-piperazine-N X-2-ethanesulphonic acid ŽHEPES. and 2% FCS, Phosphate buffered saline ŽPBS., glutamine, histidine, fungizone and gentamicin were obtained from Gibco ŽIzasa, Barcelona, Spain.. Spurr resin was purchased from Tash Laboratories Equipment ŽUK.. Anti-canine IgE was kindly supplied by Dr. D.J. DeBoer ŽUniversity of Wisconsin, Madison, USA. ŽDeBoer et al., 1993.. Antigen Asc S 1 from Ascaris suum ŽGreer Lab., Lenoir, NC, USA.. The BR canine mastocytoma cell line was kindly provided by Dr. W.M. Gold ŽCardiovascular Research Institute, University of California, San Francisco, CA, USA.. 2.2. Isolation of canine cutaneous MC Skin biopsies from healthy dogs were obtained at the Barcelona animal pound ŽCentre de Zoonosi., where dogs are sacrificed twice a week, for reasons related to city sanitation policy. The dogs were kept under observation by a veterinarian and any dog showing signs of disease was withdrawn from the study. Skin samples were obtained from the abdominal craniolateral region of healthy dogs, and the MC were isolated by an enzymatic procedure developed in our department and described elsewhere ŽDeMora et al., 1993.. Briefly, after subcutaneous fat tissue was removed, the skin was chopped into approximately 0.5–1 mm3 fragments, washed twice with MEM containing 10% FCS and incubated Ž180 min, 378C. in an enzyme mixture Ž34.5 mg collagenase, 18 mg hyaluronidase and 12 mg protease. supplemented with bovine albumin and antibiotics. Cells dispersed by this procedure were separated from undissociated tissue by filtration, washed once with MEM and incubated for 16 h at 378C with enriched DMEM in a humidified atmosphere with 5% CO 2 . Later, cells were washed twice with Ca2q- and Mg 2q-free HBSS ŽCMF-HBSS.. MC numbers was assessed in a Neubaeur hemocytometer after a Kimura metachromatic stain ŽKimura et al., 1973.. Cell viability was checked with trypan blue dye and was always greater than 95%. Before activation for histamine release, cells were resuspended in an appropriate volume of prewarmed HBSS with Ca2q and Mg 2q. 2.3. BR cell culture BR mastocytoma cells were seeded in 75-cm2 tissue culture flasks at a density of 10 6 cellsrml of enriched DMEM supplemented with 2 mM glutamine, 1.6 mM histidine, 25 mM HEPES ŽpH 7.4., 100 Urml penicillin, 100 m grml streptomycin, 2.5 m grml fungizone and 5% FCS. Cells were incubated at 378C in a humidified atmosphere with 5% CO 2 and fresh medium was added every 48 h. Cells maintained in this medium were split 1:4 and the cultured medium was changed at weekly intervals. BR cells were serially passaged to obtain the required number of cells.
326
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
Before activation, cells were washed twice in Ca2q- and Mg 2q-free PBS ŽCMF-PBS. at 238C. Cell number and viability were assessed by applying trypan blue stain. Viability was always greater than 95%. Finally, BR cells were resuspended in an appropriate volume of prewarmed PBS with Ca2q and Mg 2q before experimental use. 2.4. Canine skin MC and BR cell actiÕation To assess anti-canine IgE and antigen Asc S 1 induced histamine release, BR cells and dispersed skin MC were passively sensitized by incubation for 2 h at 378C in MEM containing 15% serum from either atopic dogs or Ascaris suum spontaneously sensitized dogs to achieve 100% occupancy of IgE receptors. Before stimulation, sensitized cells were washed twice in an appropriate buffer. Isolated canine skin MC and BR cells were distributed at a rate of 15 000 and 2 = 10 6 cells per 180 m l and stimulated by adding 20 m l of each stimulus Žanti-canine IgE, antigen Asc S 1, Con A, SP, calcium ionophore A23187, compound 48r80 or TPA. at increasing concentrations. Activation was halted at 08C and after centrifugation at 2630 rcf for 20 min at a 48C, the supernatant was collected. Twenty microliters of 55% trichloroacetic acid ŽTCA. was added to 180 m l of the supernatant to precipitate proteins before the histamine assay was performed. All conditions were performed in duplicate. Spontaneous histamine release was measured in samples incubated in the absence of any stimulus. Total histamine per MC was calculated by adding the histamine content remaining in the pellet after centrifuging to the histamine present in the supernatant and dividing by the number of cells. The cytotoxicity of each stimulus at different concentrations was assessed by trypan blue exclusion. 2.5. Histamine assay The classic spectrofluorimetric method of Shore et al. Ž1959., with the modifications of Bergendorf and Uvnas ¨ Ž1972., was used. One hundred and seventy microliters of the supernatant was taken and centrifuged Ž2000 rcf, 2 min. in the presence of 10% TCA. The medium was made alkaline Žto pH 12.5. with 0.3 M NaOH and the histamine fluorophore derivative was formed by incubating at room temperature for 4 min with OPT Ždiluted to 10 mgrml in methanol.. The histamine-OPT condensation reaction was halted at pH 3.5 by the addition of 6 N HCl. The spectrofluorometric reading conditions were excitation, 350 nm, and emission, 437 nm. At the concentrations used, none of the stimuli interfered with the fluorometric histamine assay. The presence of histamine in the samples was confirmed by degradation with histaminase Ž0.30 mgrml. for 3 h at 378C. Additionally, the histamine assay proved to have an excellent recovery rate Ž98%. and on 2- and 10-fold dilutions gave readings for histamine that were inversely proportional to the dilution rate. 2.6. Morphological study Stimulated and non-stimulated canine skin MC and BR cultured cells were treated identically in order to study their ultrastructure by transmission electron microscopy.
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
327
Canine skin MC were previously purified using a discontinuous Percoll gradient Ž40, 60 and 80%.. Freshly disgregated skin MC and cultured BR cells suspended in PBS were fixed in an equal volume of the fixative for 4 h at 48C. The fixative contained 5% glutaraldehyde in 0.1 M Sorensen’s phosphate buffer. Samples were then postfixed in 1% osmium tetroxide in 0.1 M Sorensen’s phosphate buffer for 1 h at 48C. After repeated washes with the same buffer, samples were dehydrated using a graded series of ethanol and embedded in Spurr resin following conventional procedures. Before every change of liquid or resin, suspensions were centrifuged for 10 min and the supernatant was removed. Sections 1 m m thick were stained with 1% toluidine blue for examination under a light microscope. Areas containing a higher MC density were selected for ultrathin sectioning. Sections 40 to 60 nm thick were cut using glass knives on a NOVA ultramicrotome and collected on 300 mesh copper grids. Thereafter, the sections were stained 10 min with lead citrate and examined using a Hitachi 7000 transmission electron microscope. 2.7. Data analysis Results are expressed as the arithmetical means " standard error Žmean " SEM. of 8 or 10 experiments. Differences between means were tested for significance using a Student’s t test for paired data with a level of significance of 0.05. Net histamine release after the addition of each stimulus was expressed as a percentage of total histamine corrected for spontaneous release.
Fig. 1. Non-stimulated canine dispersed skin MC. Numerous oval or round granules can be seen in the cytoplasm. The granules are filled with a highly electron-dense, homogeneous material with little or no lucent halo. Bar s1.5 m m.
328
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
3. Results The mean histamine content of BR cells was 0.04 " 0.002 pgrcell, approximately 100-fold less than that found in canine dispersed cutaneous MC Ž4.93 " 0.39 pgrcell.. Histamine content and histamine release from BR cells remained stable over time after 20 culture passages, allowing us to study their releasability adequately.
4. Electron microscopy Examination under electron microscope demonstrated significant differences between normal canine skin MC and the BR cells. Non-stimulated canine skin MC were generally oval or round and their diameters ranged from 7.5 to 12 m m. The nucleus was oval or irregular and usually located eccentrically. Nucleoli were observed in a few cells. The plasma membranes presented some folds or villous projections and some mitochondria, abundant microfilaments and numerous granules were observed in the cytoplasm. The granules of these cells were oval or round and highly uniform in size and appearance. They ranged from 255 to 410 nm in size and were filled with a highly electron-dense, homogeneous and amorphous material with little or no lucent halo ŽFig. 1.. Stimulated canine skin MC were rounder and their average diameter was similar to that of non-stimulated canine MC. The nuclei were generally more irregular and
Fig. 2. Stimulated canine dispersed skin MC. An irregular nucleus and a very prominent nucleolus are present. The plasma membrane has numerous folds and some degranulation channels are in the cytoplasm. No cytoplasmic granules are visible. Bar s 2 m m.
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
329
Fig. 3. Non-stimulated BR cells. Few moderately electron-dense granules with a visible lucent halo Žarrowheads. are present in the cytoplasm. Bar s 2.5 m m.
Fig. 4. Stimulated BR cells. Many degranulation channels can be seen in the cytoplasm. No cytoplasmic granules are visible. Bar s1.5 m m.
330
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
centrally situated. The plasma membranes had numerous folds, and many degranulation channels lined with common membranes were observed in the cytoplasm. Most MC did not show cytoplasmic granules. However, a few cells presented scarce cytoplasmic granules, similar to those described in the non-stimulated MC ŽFig. 2.. Non-stimulated BR cells were generally round and diameters ranged from 5.8 to 10 m m. The nucleusrcytoplasm ratio was much higher in these cells than in normal MC. The nucleus was usually oval and eccentric and a nucleolus was observed in many cells ŽFig. 3.. The plasma membrane had very few irregular folds and few granules were visible in the cytoplasm. These granules were homogeneous, moderately electron-dense and lucent halos were visible ŽFig. 3.. Stimulated BR cells were also round and had an average diameter similar to that of non-stimulated cells. The ultrastructural morphology of stimulated BR cells was very similar to that of non-stimulated ones, except for the presence of some degranulation channels in the cytoplasm and the absence of granules in stimulated cells ŽFig. 4..
5. Histamine release induced by Fc ´RI-independent stimuli 5.1. Calcium ionophore A23187 Dispersed skin MC and BR cells showed concentration-dependent histamine release in the presence of A23187 ŽFig. 5A.. At a concentration of 1 m M, A23187 induced a significantly lower percentage of net histamine release from BR cells than from dispersed MC Ž12.7 " 0.7% vs. 29.3 " 0.9%. ŽFig. 5A.. Histamine release from BR cells increased to 20.6 " 1.1% when TPA at 40 nM and ionophore were added simultaneously ŽFig. 6.. The 40 nM concentration of TPA had been determined previously to be the one that gave the highest synergistic effect with A23187. 5.2. TPA TPA at concentrations of 5 to 100 nM induced slight histamine release from both MC types ŽFig. 5B. in a concentration-related way that reached a plateau at 40 to 100 nM. Release induced by 40 nM of TPA was 5.6 " 0.4% and 8.2 " 0.9% for dispersed MC and BR cells, respectively ŽFig. 5B.. There were no significant differences between the two types of cells. 5.3. Substance P Histamine release induced by SP was observed at concentrations of 0.1 to 100 m M and increased in a concentration-related manner. When higher SP concentrations were used, cytotoxicity was observed Žresults not included.. Maximum histamine release was observed at 100 m M of SP from both dispersed MC and BR cells Ž23.3 " 0.7% and 18.8 " 0.7%, respectively. ŽFig. 5C..
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
331
Fig. 5. Histamine release from dispersed skin MC and BR cells induced by A23187, SP, TPA and compound 48r80. Cells were incubated with the stimuli at 378C for 20 min. Open circles: BR cells, closed circles: dispersed skin MC. Ž ns8 or 10. aCytotoxicity was assessed Žsee Table 1.. ) P - 0.05.
Fig. 6. Synergistic effect of TPA and A23187 on histamine release from BR cells. Cells were incubated for 20 min at 378C with various concentrations of A23187 in the presence or absence of TPA Ž40 nM.. Open circles: stimulated with TPA and A23187, closed circles: stimulated with A23187 only. Ž ns8 or 10. ) P - 0.05.
332
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
Table 1 Cytotoxicity of canine skin MC and the BR cell line induced by different concentrations of compound 48r80 Mast cell type
Canine skin MC
BR cell line
Concentration of 48r80
Histamine release
Cytotoxicity
Ž m grml.
Ž%.
Ž%.
10 30 100 300 1000 3 10 30 100
12.5"0.3 13.3"0.5 14.7"0.7 15.1"0.8 23.2"1.2 6.1"0.8 12.1"0.9 31.4"1.4 49.2"1.5
0.0"0.0 0.5"0.3 3.9"0.5 9.6"1.1 30.6"2.3 0.0"0.0 0.2"0.5 10.1"0.3 34.9"1.5
Each result is the mean"S.E.M. of 8 or 10 experiments.
5.4. Compound 48 r 80 The synthetic polyamine, compound 48r80, induced histamine release from dispersed MC at concentrations of 0.1 to 10 m grml. Histamine release Ž12.5 " 0.3%. peaked at 10 m grml ŽFig. 5D. and higher concentrations led to rapid increases in cytotoxicity ŽTable 1.. In BR cells, similar histamine release Ž12.1 " 0.9% at 10 m grml. was observed after addition of the same range of stimulus concentrations. BR cells were more sensitive to the toxic effects of compound 48r80 ŽTable 1..
6. Histamine release induced by Fc ´RI-mediated stimulation 6.1. Antigen Asc S 1 Antigen Asc S 1 at concentrations of 100 to 3000 m grml induced histamine release from dispersed sensitized MC, and a maximum response of 12.2 " 0.9% was observed at a concentration of 1000 m grml. Sensitized BR cells did not release histamine upon Asc S 1 stimulation ŽFig. 7A.. 6.2. ConcanaÕalin A The lectin Con A at concentrations of 0.1 to 1000 m grml markedly released histamine from dispersed MC but no effect on BR cells was observed ŽFig. 7B.. The maximum peak of 27.7 " 0.6% was reached at 1000 m grml, and higher doses of Con A did not increase histamine release Žnot shown.. 6.3. Anti-canine IgE Histamine was released from dispersed sensitized MC in a dose-dependent manner after addition of anti-canine IgE at concentrations of 0.0001 to 0.01 m grml and the
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
333
Fig. 7. Histamine release from dispersed skin MC and BR cells induced by Antigen Asc S 1, Anti-IgE and Con A. Cells were incubated for 20 min with Con A and 30 min with Asc S 1 and Anti-IgE. Open circles: BR cells, closed circles: dispersed skin MC. Ž ns8 or 10. ) P - 0.05.
maximum effect of 30.9 " 1.7% was observed at the 0.003 m grml. In contrast, anti-IgE did not induce histamine release from sensitized BR cells at any concentration ŽFig. 7C..
7. Discussion In the present study we have found both similarities and differences between the BR cell line and isolated canine skin MC. The two cell types share similar patterns of histamine releasability after non-immunological stimulation and have some common morphological characteristics. They differ, however, in their releasability after immunological activation. Presumably, the later is due to the absence of Fc ´ RI on the BR cells surface. Both the analogies and differences provide insights on canine mast cells biology and developmental processes respectively. Besides the absence of detailed information on BR previously to the present work, this cell line offered advantages vs. similarly developed canine MC lines Že.g. C 2 .. BR are a source of non-adherent fast growing cells that have the ability to divide in FCS-free medium permitting more accurate functional studies on their capability to liberate immune mediators. The present preliminary data on
334
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
their functional and morphological characteristics are a required step to further advance in the analysis of BR cells morphofunctional characteristics under diverse conditions. Our results on non-activated canine MC ultrastructure are similar to those published by Calonico et al. Ž1985.. This resemblance is extremely interesting in the light of the slightly different experimental approaches undertaken. Calonico et al. Ž1985. reported the ultrastructural analysis of freshly isolated mastocytoma cells rather than cell lines, our data emphasize that, after propagation in mice, the mastocytoma cells structure does not significantly change and therefore that the conversion from freshly isolated tumoral cells to a cell line ŽBR. has little effect on their phenotype. Furthermore the comparison between mature dermal MC and mastocytoma cells was done with frozen sections instead of cell suspensions which might provide interesting but potentially different functional and morphological information from the cells suspensions used in the present study. In our work, canine skin MC appear mature, highly differentiated cells with cytoplasm that is occupied mainly by granules and few organelles. Unlike human MC, canine skin MC granules are electrodense and homogeneous and contain no crystals ŽWhitaker-Menezes et al., 1995.. BR cells, on the other hand, have an immature and undifferentiated appearance. The, nuclei of BR cells are richer in euchromatin, and the nucleoli are more evident than those of canine skin MC. Moreover, cytoplasmic granules are scarce, less dense and surrounded by a more electron-lucent area, indicating that these granules are less mature than those with an electron-dense homogeneous appearance without an electron-lucent halo observed in canine skin MC ŽCalonico et al., 1985.. Among the cell lines established from neoplastic mast cell lines to date ŽC 1 , C 2 , G and BR., the BR line appears to be the least differentiated. The C 1 and C 2 lines, on the other hand, are the most highly differentiated, and the differentiation of the G line is intermediate ŽLazarus et al., 1986.. Such variation is understandable given that canine mastocytomas, from which they have been derived, can be either highly or scarcely differentiated ŽPatnaik et al., 1984.. BR cell stimulation, however, leads to a similar level of degranulation than the one observed for non-neoplastic MC. The histamine content of the BR cell line remained stable over different culture passages, ensuring that the functional studies would not be altered over time. BR cell histamine content is less than that of dispersed canine skin MC and is clearly related to the low number of granules observed in their cytoplasm. This is consistent with what has been found for other mastocytoma cell lines and in immature MC Ž0.46 to 0.07 pgrcell. ŽDeVinney and Gold, 1990.. Lazarus et al. Ž1986. working with other dog mastocytoma cell lines ŽC 2 and G. suggested that differences in histamine content could reflect the neoplastic origin of the cell line and some dedifferentiation in conjunction with neoplastic transformation in the original tumor. The release of preformed mediators is one of the major functions of MC. We examined the histamine releasability of BR cultured cells for comparison with that of dispersed skin MC using a variety of non-immunological and immunological secretagogues. As expected, calcium ionophore A23187 induced a significant amount of histamine release in both cell populations. However, quantitative differences were observed between BR cells and dispersed canine MC, a finding that can be accounted for by differences in membrane composition or in sensitivity of calcium channels as suggested by Bottjer et al. Ž1994.. In the BR cells, calcium ionophore A23187 histamine ¨
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
335
release was potentiated by incubation with TPA, whereas TPA or ionophore A23187 alone induced low histamine release, indicating that protein kinase C ŽPKC. activation by TPA interacts synergistically with a rise in intracellular calcium induced by calcium ionophore A23187 to produce histamine secretion from BR cells. This result is consistent with previous reports for a variety of MC stimulated with this combination of compounds ŽCantwell and Foreman, 1986; Sekino et al., 1990.. Accumulating evidence suggests that SP and compound 48r80 cause secretion from human and canine skin MC both in vitro and in vivo ŽTausk and Undem, 1995; Garcıa ´ et al., 1997a.. Some cell lines, however, such as C 2 , fail to respond to 48r80 stimulation ŽDeVinney and Gold, 1990.. In our study of the BR line 48r80 or SP stimuli did effectively induce histamine release, supporting a hypothesis that tumoral BR cells keep the particular membrane characteristics of canine skin MC in this respect, given that SP and compound 48r80 reactivity is thought to result from a more fluid or negatively charged membrane ŽTainsh et al., 1992.. Inase et al. Ž1993. likewise described similar results for freshly isolated BR mastocytoma cells stimulated by compound 48r80 at concentrations ranging from 0.1 to 10 m grml, although they observed higher histamine release than we did at the highest stimulus concentration, while finding no cell cytotoxicity at that level of stimulation. Immunologic stimulation failed to elicit significant histamine release from BR cells with any of the three stimulus types although all three induced release from dispersed canine skin MC. These results, indicate that BR cells would be unlikely to have any functional IgE receptors or that any such receptors would be present only at low density. The absence of such receptors appears to be an intrinsic property of BR cells given that passive sensitization did not change the response to immunologic stimulation. Receptors may have been lost during the different passages in nude mice or neoplastic transformation may have taken place. It is also possible to speculate that the mastocytoma cells obtained directly from the dog were also refractory to sensitization. Lazarus et al. Ž1986. reported heterogeneity in ragweed-pollen extract histamine release after passive sensitization with IgE-rich serum for 1 h, demonstrating that the C 2 cell line had functional IgE receptors and suggesting that the G cell line may have lost its IgE receptors as a result of neoplastic transformation leading to their inability to react passively with IgE. In agreement with data obtained by other investigators histamine release after anti-canine IgE and antigen Asc S 1 stimulation never occurred without preincubation with IgE ŽMita et al., 1995; Tunon de Lara et al., 1995.. Previous results from our laboratory showed that passive sensitization with IgE-enriched serum is not absolutely necessary for inducing histamine release from dispersed canine skin MC upon stimulation with Con A ŽDeMora et al., 1996; Garcıa ´ et al., 1997a.. However, the lack of Fc ´ RI receptors in the BR cell line produced a lack of response with Con A stimulation. A possible hypothesis, to explain these results, it is that Con A probably act by cross-link Fc ´ RI receptors direct rather than by a cross-link Fc region from IgE, as has been suggested by other authors ŽCarnuccio et al., 1989.. In conclusion, BR cells provide a pure and abundant source of canine MC for the study of mediators released by these cells through non-immunologic mechanisms, and this cell line is indeed a useful tool for analyzing non-specific-MC mediators and signal transduction pathway mechanisms. Furthermore, as BR cells show an array of functional
336
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
and morphological features typical of immature MC populations, the undifferentiated nature of this cell line may make it useful for studying certain aspects of MC differentiation. Acknowledgements Dog skin biopsies were kindly provided from the municipal pound ŽCentre de Zoonosi. of Barcelona ŽSpain. from dogs that had been sacrificed following city sanitation policy. This work was supported by a grant from the Direccion ´ General de Investigacion y Tecnologıa ´ Cientıfica ´ ´ of Spain ŽDGICYT, PB93-0905.. References Bergendorf, A., Uvnas, ¨ B., 1972. Storage of 5-hydroxytryptamine in rat mast cells. Evidence for an ionic binding to carboxyl groups in a granule heparin-protein complex. Acta Physiol. Scand. 84, 320–331. Bottjer, J., Amon, V., Wolff, H.H., 1994. Functional comparison of different histamine-containing IgE-recep¨ tor positive cells. Agents and Actions, Vol. 41, Special Conference Issue: C28–C29. Calonico, L.D., Phillips, M.J., McDonald, D.M., Gold, W.M., 1985. An ultrastructural analysis of dog mastocytoma cells and normal mast cells. Anat. Rec. 212, 399–407. Cantwell, M.E., Foreman, J.C., 1986. Characteristics of the effect of phorbol ester on rat mast cells: interaction with anti-IgE. Agents and Actions 18 Ž1r2., 77–80. Carnuccio, R., Di Rosa, M., Ialenti, A., Iuvone, T., Sautebin, L., 1989. Selective inhibition by vasocortin of histamine release induced by dextran and concanavalin A from rat peritoneal cells. Br. J. Pharmacol. 98 Ž1., 32–34. Caughey, G.H., Lazarus, S.C., Viro, N.F., Gold, W.M., Nadel, J.A., 1987. Tryptase and chymase: comparison of extraction and release in two dog mastocytoma lines. Immunology 63, 339–344. DeBoer, D.J., Ewing, K.M., Schultz, K.T., 1993. Production and characterization of mouse monoclonal antibodies directed against canine IgE and IgG. Vet. Immunol. Immunopathol. 37, 183–199. DeMora, F., Garcıa, ´ G., Ferrer, L., Arboix, M., 1993. Canine cutaneous mast cells dispersion and histamine secretory characterization. Vet. Immunol. Immunopathol. 39, 421–429. DeMora, F., Garcıa, ´ G., Puigdemont, A., Arboix, M., Ferrer, L., 1996. Skin mast cells releasability in dogs with atopic dermatitis. Inflamm. Res. 45, 423–427. DeVinney, R., Gold, W.M., 1990. Establishment of two dog mastocytoma cell lines in continuous culture. Am. J. Respir. Cell Mol. Biol. 3, 413–420. Galli, S.J., Costa, J.J., 1995. Mast-cell-leukocyte cytokine cascades in allergic inflammation. Allergy 50, 851–862. Garcıa, ´ G., DeMora, F., Ferrer, L., Puigdemont, A., 1997a. Effect of H 1-antihistamines on histamine release from dispersed canine cutaneous mast cells. Am. J. Vet. Res. 58 Ž3., 293–297. Garcıa, ´ G., DeMora, F., Ferrer, L., Puigdemont, A., 1997. Inhibition of histamine release from dispersed canine skin mast cells by cyclosporin A, rolipram and salbutamol, but not by dexamethasone or sodium cromoglycate. Vet. Dermatol., Žin press.. Inase, N., Schreck, R.E., Lazarus, S.C., 1993. Heparin inhibits histamine release from canine mast cells. Am. J. Physiol. 264, L387–L390, ŽLung Cell. Mol. Physiol. 8.. Kimura, I., Moritani, Y., Tanizaki, Y., 1973. Basophils in bronchial asthma with reference to reagin type allergy. Clin. Allergy 3, 195–202. Lazarus, S.C., DeVinney, R., Laurence, J.M., Walter, E.F., Darlene, J.E., Gold, W.M., 1986. Isolated canine mastocytoma cells: propagation and characterization of two cell lines. Am. J. Physiol. 251, C935–C944, ŽCell Physiol. 20.. Mita, H., Yasueda, H., Ishii, T., Akiyama, K., 1995. IgE-mediated basophil releasability is influenced by intrinsic factors and by IgE on the cell surface. Allergy 50, 952–958.
G. Garcıa ´ et al.r Veterinary Immunology and Immunopathology 62 (1998) 323–337
337
Patnaik, A.K., Ehler, W.J., MacEwen, E.G., 1984. Canine cutaneous mast cell tumor: morphological grading and survival in 83 dogs. Vet. Pathol. 21, 469–474. Sekino, H., Uchida, M.K., Suzuki-Nishimura, T., 1990. The modulatory effects of phorbol esters on histamine release from mast cells induced by a polyethylenimine and a polyallylamine. Gen. Pharmacol. 21, 535–539. Shore, P.A., Burkhalter, A., Cohn, V., 1959. A method for the fluorometric assay of histamine in tissues. J. Pharm. Exp. Ther. 127, 182–186. Tainsh, K.R., Liu, W.L., Lau, H.Y., Cohen, J., Pearce, F.L., 1992. Mast cell heterogeneity in man: unique functional properties of skin mast cells in response to a range of polycationic stimuli. Immunopharmacology 24, 171–180. Tausk, F., Undem, B., 1995. Exogenous but not endogenous substance P releases histamine from isolated human skin fragments. Neuropeptides 29, 351–355. Thomas, P.S., Schreck, R.E., Ruoss, S.J., Lazarus, S.C., 1991. Heterogeneity of intact granules purified from canine mastocytoma cell lines. Am. J. Physiol. 260, L153–L160, ŽLung Cell Mol.Physiol. 4.. Tunon de Lara, J.M., Okayama, Y., Savineau, J.-P., Marthan, R., 1995. IgE-induced passive sensitization of human isolated bronchi and lung mast cells. Eur. Respir. J. 8, 1861–1865. Turner, C.R., Darowski, M.J., Sampson, H.A., Spannahake, E.W., Hirshman, C.A., 1989. Dermal mast cell releasability and end organ responsiveness in atopic and nonatopic dogs. J. Allergy Clin. Immunol. 83, 643–648. Whitaker-Menezes, D., Schechter, N.M., Murphy, G.F., 1995. Serine proteinases are regionally segregated within mast cell granules. Lab. Invest. 72 Ž1., 34–41.