Document OJKw7z6BGe49X7bO8kKOrnnk1

I r :i i tn huf >!,il ! I'tUi'iil,. I . <' o V Quantitative Immunohistologic Assessment of Lymphocyte Populations in the Pulmonary Inflammatory Response to Intratracheal Silica Rakesh K. Kumar froni the hebool ofParholog)-, ! 'nuvrsity of \'ew houtb Wiilei. huhiey. Australia Immunogold-silver staining was used to identify T lymphocytes. T lymphocyte subsets, and B lympho cytes in lung tissue from mice injected intratracheally with silica, titanium dioxide, or saline atone. Morphometric quantitation revealed a marked influx ofTlymphocytes in the silica-treated animats during thefirst 3 weeks after injection. The relative numerical density of these cells remained elevated when compared with saline-treated con trols throughout the 12 weeks of the experiment. Cells expressing the CD4 and CD8 antigens were both increased in number, with the former ac countingfor approximately twothirds ofthe T lym phocytes. An increased number of B lymphocytes was also apparent from 6 weeks after treatment with silica. The Tlymphocyte response preceded the development of significant pulmonary fibrosis by several weeks. No lymphocyte response was ob served in the lungs of mice injected with nonfibrogenic titanium dioxide. These obsesrvations are consistent with the hypothesis that tympbokines se creted by T lymphocytesplay a role in thepathogen esis ofsilicotic inflammatory lesions and theirpro gression to fibrosis. (Am f Pathol 1989, 135-605614) Aw pathogenetic mechanisms involved in the develop'"rient of pulmonary fibrosis have not been completely deSnd- Histopathologically, there is usually evidence of in'firstrtial and often intra-alveolar proliferation of fibroblasts, "ith deposition and rearrangement of collagen, an accmpanying inflammatory response dominated by monoXidear cells, and hypertrophy and hyperplasia of type 2 >0edmocytes.' Possible interactions between these cell PoPdlations and the role of various cytokines in mediating >uch interactions have been the focus of a number of ,5cent studies. Much information has accumulated that supports the belief that cytokines secreted by cells that participate in the inflammatory response have the poten tial to stimulate fibroblast chemotaxis, proliferation, and collagen synthesis.2'13 The biological role of such mole cules in vivo, however, has not yet been elucidated. Most studies of fibroblast-stimulating cytokines in pul monary fibrosis have concentrated on the macrophage as the cellular source of such mediators. These have es tablished that, in both experimental and clinical disease, macrophages are able to release factors that may pro mote fibroblast growth, collagen synthesis, or both.'*'19 However, studies of the induction of secretion of macrophage-denved growth factors in vitro suggest that these cells exhibit responses of comparable magnitude to both fibrogemc and nonfibrogenic stimuli.2 ,9 Therefore, if mac rophage-derived growth factors have relevance in the de velopment of fibrosis in vivo, mechanisms must exist that regulate the secretion of growth factors, so that sustained secretion of biologically active cytokines occurs only in response to a fibrogenic stimulus Such regulatory signals may be provided by the secre tion of lymphokines that could activate pulmonary macro phages to secrete growth factors. Lymphocytes are fre quently present in the interstitial inflammation associated with the development of pulmonary fibrosis and may par ticipate in an immunologic response to alterations in the structural components of lung tissue induced by the fi brogenic stimulus.20 21 However, there have been few studies of the secretion of mediators by pulmonary lym phocytes. There is some evidence supporting both the hypothesis that lymphocytes secrete mediators that di rectly stimulate fibroblast proliferation22 and the possibility that lymphocyte-macrophage interactions play a role in the induction of growth factor secretion by macro phages.23 Supported by grants from me National Health and Medical Research Counc* of Australia and from the Australian Tobacco Research Founda tion. Accepted tor publication May 26.1969 Address repnnt requests to Or R K Kumar. Senior Lecturer m Pathol ogy. The University of New South Wales. PO Box I. Kensington. NSW, Australia 2033. 605 r 5- 606 Kumar .<(/' 1>, mut i 'ii i r-; . Ongoing studies in this laboratory are directed to Intratracheal Injection . lympnocvtei. L3T4 wards evaluating the possible contribution of lymphokines to the development of particle-induced pulmonary Mice were anesthetized by intraoentoneal inject; lamed from SeraLa - I anti-mouse IgM was fibrosis initiated by intratracheal injection of silica in mice. mixture of 100 mg/kg of ketamine (Ketalar. Parke .. _.,s Affinity-pun fiec goat This experimental system is considered a model of acute Sydney) and 3.3 mg/kg of xylazme (Rompun. Bayer, S/G dal gold particles (A or accelerated silicosis.2,1 That lymphocytes participate in ney). The ventral aspect of the neck was shaved and the hancement reagent the inflammatory response in human silicosis has long skin incised in the midline. A pediatric ocular speculum | Sciences (Olen. Bet been recognized, and accompanying alterations in immu nologic responsiveness have also been documented.25 was inserted to expose the strap muscles, which were i Immunogold-silv. separated by blunt dissection to reveal the trachea. Usinr; I and B lymphocytes However, there is little quantitative information available regarding the involvement of lymphocytes in pulmonary a 1 ml tuberculin syringe attached to an angled #26 nee I son staining svsten jdie, 0.05 ml of a suspension of particles or of vehicle aicne nents for the manual inflammation induced by siftca and virtually no studies of were injected into the tracheal lumen. The mcis - ,vas ' obtained from Fish< their possible pathogenetic role have been per formed.2526 Therefore, as a first step towards assessing the functional capacity of lymphocytes to influence the development of silicotic fibrosis, this study was under closed with interrupted silk sutures. Experimental Design ; I frozen sections weri i with 0 5% gelatin ir I Sections were blocf j rum, incubated with taken to quantitate lymphocyte populations in the pulmo nary inflammatory response induced by silica, compared with the response induced by nonfibrogemc titanium diox ide or by saline alone. T lymphocytes. T lymphocyte sub sets. and B lymphocytes were identified by immunogoldsilver staining, and their relative numerical densities were estimated by appropriate morphometric methods The in Animals were divided into 13 groups of six each and re ceived intratracheal injections of either 2 mg of silica. 2"--. of titanium dioxide, or saline alone. Silica-treated amraii were killed 1, 2. 3, 4, 6, 8, and 12 weeks after mecticr Animals receiving titanium, dioxide or saline ah .ver; killed 1,4, and 12 weeks after treatment. I antibody for 45 mu gold-con|ugated go; ' er washing and po formalin, the gold la silver around the par vete counterstainet catch of slides that v volvement of these cells was correlated with the develop of normal mouse st i ment of histologically recognizable pulmonary fibrosis. Preparation of Tissues GSS procedure Materials and Methods Animals Twelve-week-old female specific pathogen-free BALB/c mice were obtained from the animal breeding facility at the Australian Nuclear Science and Technology Organisa tion, Sydney, Australia. Animals were housed in filter-top cages on autoclaved bedding in a bamer environment. Irradiated food and autoclaved water were allowed ad libi tum. All experimental procedures complied with the re quirements of the Animal Ethics Committee of the Univer sity of New South Wales (ref. no. 88/3). Particulates Silica with a particle size less than or equal to 5 titn (MinU-Sil, Pennsylvania Glass Sand Co., Pittsburgh, PA) (a generous gift from Dr. G. E. R. Hook, NIEHS. Research Triangle Park, NC) was prepared as previously descnbed by washing in hydrochlonc acid.27 For intratracheal injec tion. particles were suspended in 0.9% saline at 40 mg/ ml and ultrasomcated for 60 seconds to break up aggre gates. Titanium dioxide (Sigma. St. Louis, MO) was sim ilarly dispersed at 40 mg/ml by ultrasomcation. Particles were resuspended by vigorous synngmg before injection. Animals were anesthetized with 75 mg/kg of pentobarb tal (Nembutal, Ceva Chemicals, Sydney) and killed b. transection of the aorta. The pulmonary trunk was cannu lated with a #21 needle and the lungs perfused with 0 9* saline under a pressure of 60 cm HzO for 6C conds to remove blood from the pulmonary capillary ,,ed The trachea was then cannulated with a blunted #19 neetfe and the lungs were inflated to total lung capacity with t n of frozen section embedding medium (Tissue-Tek, Miles Scientific. Naperville, IL). Three horizontal slices, each acproximately 1 -mm thick, were cut from the midzone of the single-lobed left lung. One slice was fixed in 10% phcs phate-buffered formalin for 24 hours before embedding ' paraffin. Four-^m-thick sections from the paraff " -meed ded tissue were stained either with hematoxylir c eosr or with a modified Masson trichrome stain in wmen ire nuclear counterstain was omitted to improve contras'. The other two slices of lung were frozen in isopentare precooled in liquid nitrogen and stored at -70 C. Six-nm thick sections were cut on a Reichert Frigocut 280C (Reichert-Jung, Nussloch, West Germany), fixed in ace tone for 10 minutes, and stored at -20 C until use. Immunostaining Rat monoclonal antibodies to the mouse T lymphoc)-; surface membrane differentiation antigens Thy-1 (pah T Morphometry 4 systematic samplii employed as recorr mal, the blocks and chosen randomly. S a Zeiss microscope *hich a light video ,vas attached, and monitor (48 cm diag. coherent point-coun cm apart, of which c coeit. To examine e; scaled at one cornr kchteved by using ; :r to select one of io that the tissue unc "'Posed on the toea :-ter. fields were sy misusing the vernier ^on was evaluat atety 15 fields per: Quantitation of th Jlcyfes, B lymphoc -Wormed using an ' a*-31 Sections werr ciective at a final m; a I Lymphocyte Populations in Silicosis 607 t/fOuiiivr t'.W9, IV).' ;,,(v .Vo -I I lymphocyte). L3T4 (C04). and Ly-2 (CD8)28 were ob tained from SeraLab (Crawley Down, UK). Monoclonal t 5 j anti-mouse IgM was obtained from Serotec (Oxford, UK). is j Affinity-purified goat anti-rat IgG conjugated to 5 nm colloi- /C J dal gold particles (AuroProbe LM) and IntenSE II silver en- Te ' nancement reagent were purchased from Janssen Life urr Sciences (Olen, Belgium). e'a tmmunogold-silver staining (IGSS) of T lymphocytes r-j and B lymphocytes was performed using a capillary acee ! tion staining system as previously descnbed.29 Compo- e nents for the manual Probe-On slide staining system were .=. obtained from Fisher Scientific (Pittsburgh, PA). Briefly, frozen sections were mounted on Probe-On slides coated t /nth 0.5% gelatin in 0.05% chromic potassium sulfate. J Sections were blocked with 1:10 diluted normal goat se- I mm, incubated with appropnately diluted monoclonal rat | antibody for 45 minutes, washed, and incubated with ' gold-conjugated goat anti-rat antibody for 45 minutes. Af- : :er washing and postfixing the sections in 1 % ethanolic '3 formalin, the gold label was visualized by precipitation of 'r silver around the particles with IntenSE it reagent. Sections ` ' .vere counterstained with methyl green and eosin. Each batch of slides that was immunostained included sections of normal mouse spleen that served as controls for the ' GSS procedure. irti Morphometry "u A systematic sampling procedure with a random start was -cs I employed as recommended by Weibel.30 For each ani1 nal, the blocks and the block faces to be sectioned were thosen randomly. Stained sections were examined with i Zeiss microscope (Carl Zeiss, Oberkochen, FRG) to "iS "rich a light video camera (Sony Corp., Tokyo. Japan) =C- *as attached, and images were projected on a color Ti ~'nitor (48 cm diagonal). The screen was overlaid with a tbherent point-counting lattice of 7 X 5 points, spaced 5 i Th apart, of which one point was designated as the test 30nt- To examine each slide, the test quadrat was initially ''cated at one comer of the section. A random start was achieved by using a computer-generated random numas: to select one of the 35 points and moving the stage othat the tissue underlying the selected point was super""Posed on the location marked by the test point. There-`,er, fields were systematically sampled at 1 -mm inter is using the vernier stage movement. Every field in each *ri was evaluated, yielding an average of approxi- a,e|y 15 fields per section. ' Quantitation of the relative numerical density of T lymJ'cytes, B lymphocytes, and T lymphocyte subsets was formed using an approach similar to that of Chandler ^ a-3' Sections were examined with a X16 magnification Elective at a final magnification of X1667. Profiles of pos- itively immunostained cells with a visible nucleus were counted in each field, applying the rules discussed by Weibel,32 and the number of points falling on the section was recorded. The cell count was then expressed as number of cells/mm2 of tissue section. The volume den sity of collagen in Masson trichrome-stained sections was determined by relative point-counting of positive-staining areas in parenchymal tissue using a X40 objective. Statistical Methods Data from the cell counts were loganthmically trans formed before analysis. Differences between control and particle-treated groups of animals were assessed by one way analysis of variance. The effect of intratracheal injec tion of particles at each time point was compared with the combined means of the saline-treated control groups by the multiple contrasts procedure of Scheffe s S test13 Results Intratracheal injection was successful in all 78 animals, and none of the animals died during the course of the experiment. Histopathologic Findings Animals injected with silica exhibited a florid pulmonary inflammatory response. In sections stained with hematox ylin and eosin, lesions were most often found in the pos teromedial portion of the lung. As early as 1 week after injection of silica, the most striking component of the re sponse was the development of granulomas, consisting mostly of modified macrophages with the morphology of epithelioid cells. These were usually distributed in relation to respiratory bronchioles. Particles of silica were often identifiable within the granulomas by polarization micros copy. These particles were frequently surrounded by small clusters of neutrophils. A granulomatous response was also noted in hilar lymph nodes at this stage. Be tween the granulomas, many alveoli contained foamy, granular eosinophilic material. Particles of free silica were often identified.in these alveoli..which also contained nu merous free alveolar macrophages, many of which had highly vacuolated cytoplasm. Perivascular aggregates of i 1 .. *- -- ; : --> j--- 608 Kumar -\l!'"O'' / ' .7 \ Figure 2. Frozen section oflungjrom an animal . i tt.. silica 2 weeks preciously, demonstrating peru a> gates ofT lymphocytes, which exhibit membrane the Thy-1 antigen (IGSS-metbyl green c-eostn. original w,js jicationXXO) Figure t. Pulmonary tnfiantmaron' response induced 1 week after intratracheal injection of silica. Xumerous epithelioid cell granulomas are present and many alreoli contain granu lar eosinophilic debris. Ousters of lymp(x>cytes surrounding small blootl ressels are also risible (arrows) (Hc-E, original magnification X 120). lymphocytes were frequently seen (Figure 1). In sections stained by IGSS, most of these cells were T-lymphocytes that stained positive for the Thy-1 antigen (Thy-1 ") (Figure 2), although a small proportion of B lymphocytes exhibit ing surface membrane staining for IgM (stgfrP) was also noted. Polanzable silica particles continued to be visible in the granulomas for the duration of the experiment. In later le sions, the inflammatory response was sustained and was accompanied by obvious alveolar septal thickening, with large numbers of plump cuboidal epithelial cells covenng the alveolar surfaces. The morphology of these cells sug gested that they were probably hypertrophic type 2 pneumocytes. Nodular aggregates of lymphocytes around blood vessels and bronchioles were quite prominent by approximately 6 weeks after inaction ot silica. IGSS re vealed that most of these cells were now slgM" B lympho cytes (Figure 3). Examination of sections stained with Masson's tri chrome demonstrated deposition of collagen fibers in the lung parenchyma from 4 weeks onwards, although some animals exhibited fibrosis as early as 1 week after injection of silica. Fine bundles of collagen fibers were seen in peri- vascular and alveolar interstitial regions, whereas nodus mtra-alveolar fibrosis occurred in relationship to the s. colic granulomas (Figure 4). rgure 4. Dei .'untilontas j .:< (Masson In contra ~als mjecte '"ammator urn dioxic -ages. In "onstrable -'?arances ^sficiesof t "ages in tf No mflar ` Kted witt ''orphom Figure 3. Frozen section of lung from a sthcatreitic`11< after 6 weeks, showing a nodular aggregate oflymph,]L' * ` jirt ent to a blood ressel. Most ofthe cells are B lympho*-1 exhibit surface membranestatntngforIgM flG$s metb\i 2 & eosm. original magnification X 20(>) 1 quaJitat "Uts Of in n'als, en ""arked es j miectir ! l|Jhgs < I elevj of the ^'andC "iher a& Lymphocyte Populations m Silicosis 609 {/!'< . -w.v, ; iy \,, , I rgure 4. Deposition of collagen fibers uitbm silica-induced ;cu i sunulomasfrom an animal 4 weeks after mtratraclyeal injec- u (Masson tricbrome. ortgnml magnification X2G0J I * cells Most T lymphocytes at each time point studied were CD4* cells The ratio of tne mean number of CD4' CD8* cells was similar to the ratio of 2 4.1 found in tne lungs of control animals In contrast, slgM' B lymphocytes were only modestly increased early in the inflammatory response, but were present in significantly increased numbers from 6 weeks onwards (Figure 8) Correspondingly, the ratio of the mean number of B T lymphocytes rose from 0.4 1 in the first 2 weeks to over 1.6:1 in the period 6-to-12 weeks after injection, compared with a mean ratio of 0 7.1 in the control animals. The volume fraction of stamable collagen increased during the early weeks of the expenment and reached statistical significance at 8 weeks after injection of silica (Figure 9). No significant changes in either the relative numerical densities of pulmonary lymphocytes or the volume frac tion of collagen were found in the animals injected with titanium dioxide (Figures 6, 8, and 9). Discussion The pulmonary response induced by intratracheal injec tion of silica has been previously described. Most studies In contrast to the response-observed with silica, am! "als injected with titanium dioxide exhibited little if any | 'lammatory response to the particles. Much of the tita; 'urn dioxide was phagocytized by alveolar macro| fages. In addition, free titanium dioxide was readily de- j 'onstrable in the alveolar lumma (Figure 5). These api tearances persisted throughout the period of study. 'Vticles of titanium dioxide were also found within macro phages in the hilar lymph nodes. , No inflammatory changes were observed in animals . '.acted with saline alone. Morphometry "e qualitative changes observed were validated by the ^ of the morphometric analysis. In the silica-treated -*r>als, enumeration of Thy-1'' T lymphocytes revealed 11ari(ed early influx of these cells dunng the first 3 weeks `hjection. Thereafter, the number of Thy-1* cells in * lungs decreased somewhat, but remained signifi- tevated relative to the saline-treated controls for st of the penod of study (Figure 6) The numbers of ^ and CD8" cells were both increased (Figure 7) and vether accounted for the increase in the number of Thy- Figure 5. Aggregated particles of titanium dioxide and partide-laden macrophages in the alveoli of an animal 4 weeks after intratracheal injection (H-E, original magnification XJOO) c* 610 Kumar MPOitoher /9tfO. I'./ lit .\o -f increased with CD4* cells accounting tor approximates, two thirds of the T lymphocytes participating r e re sponse. Although the Thy-1 antigen is not unux 0 m nne T lymphocytes, being expressed to a significant ex tent on NK cells and dendritic cells,37 the CD4 ana CCc antigens are not markers for these populations.38 ana the observation that the increase in numbers of CD4* ar<CD8* cells accounted for the increase in Thy-1 - cells ccn firms that the Thy-1* cells enumerated were T lympho cytes. The T lymphocyte response, which precedes the development of significant fibrosis by 3 to 6 .v?eks. s sustained through the evolution of the pulmonr 3:0ns In the later stages, there is also a marked increase in trnumber of slgM* B lymphocytes in the lung Statistical evaluation of the morphometric data prota bly underestimates the magnitude of the lymphocyte re sponse within the silica-induced lesions The coefficient c: 60-i 50 - 40E E 30o w 9 a 3 2 20- 10- 12 3 4 6 8 12 Time (weeks) Rgure 6. Relative numerical density (mean SEM) of Thy-1* cells m the lungs of animals at various times after treatment with silica < -------- ), titanium dioxide (a-------- ), or sa line alone (M- --------Significant differencesfrom the combined means of the saline-treated control groups are sbou n as *, P < 0.05and P < 0.0/. have emphasized macrophage-dominated granulomatous inflammation,3435 the development of alveolar lipoprotemosis24 35 and accompanying hypertrophy, and hy perplasia of type 2 pneumocytes.35 36 In the experiments described here, the response was entirely consistent with that found in earlier studies. These reports also com mented on the presence of lymphocytes in the inflamma tory cell population, both in the interstitium surrounding the macrophage-dominated silicotic granulomas34 and in penbronchiolar aggregates.35 However, so far as the au thor is aware, quantitative immunohistologic analysis of the responding lymphocytes has not previously been un dertaken. The data from the present experiments establish that a significant influx of Thy-14 T lymphocytes occurs early in pulmonary inflammation induced by intratracheal injec tion of silica. CD4* and CD8* cells are correspondingly Time (weeks) Figure 7. Relative numerical density ofCD4* < ------- - * [ \ * CDS* (+%) cells in tint lungs ofanimals treated m ui} For the sake ofclarity, data for other experimental group* not shown Significant differences from the combined of the saline-treated control groups are shou n as *. P < and **, P < 0.01. 1 r 1 r9ure8. Retain I animats treat *------ A. ), or tv/* the comhtit Ire shown as I tnation for mo S'tg the con: 'before, was i 1 the number < ' the alveolar c ^ability of the 'Tuted to the 'Portant reast '-ttnbution of ''tWem associ '3cheal injectic ^ proportior Of uninvr '^cent block ^ng a muc cellular re: ''treason for lymohocyte Populations m Silicosis 611 Air >\:>inrr mv) i, >. ^5. -4 phocytes recorded lor several animals ot the group that had received silica 4 weeks previously Nevertheless, to maintain the integrity of the random sampling procedure, no sections were excluded from evaluation on the grounds of inefficient sampling. Recently. Struhar et al40 demonstrated that, following administration of silica to rats by intratracheal injection, there was an early increase in the number of T-lymphocytes obtained by bronchoalveolar lavage or by dissocia tion of lung tissue with collagenase. In agreement with the results reported here, the lymphocytic response was dominated by CD44 cells and was maximal at 14-30 days postexposure. It remains to be established whether hu man pulmonary disease induced by inhalation exposure to silica is also associated with a sequential increase in the number of T lymphocytes and 8 lymphocytes in the lung parenchyma, as has been demonstrated in this ex periment. However, it is noteworthy that perivascular and peribronchial aggregates of lymphocytes were described in mice exposed to silica dust by inhalation,4' although no assessment of lymphocyte populations was carried out. Similarly, the presence of lymphocytes surrounding the Time (weeks) Pgure 8. Relative numerical density ofslg.\r cells in the lungs Yanimals treated with silica < #-------- ), titanium dioxide {* - - - ), or saline atone Significant differences fmm the combined means of tlye saline-treated control groups jrr shown as *, P < 0 05 and * *, P < 0.01 ^nation for most groups was high; statistical significance -S'ng the conservative procedure of Scheffe's S test. ~erefore, was only achieved when the mean differences n the number of cells were large. Quantitative vanations n ihe alveolar delivery of silica and the inherent biological inability of the inflammatory response undoubtedly con'r**Jted to the high coefficient of variation. However, an 'Txxtant reason for the variability was the nonunitorm istr*XJtion of the lesions, which is a well-recognized ^oblem associated with the delivery of particles by intraacheal injection.39 In some of the sections examined, a *9 proportion of the microscopic fields evaluated con. sd of uninvolved lung tissue, when sections from an 'acent block from the same animal exhibited lesions ^vtvmg a much greater proportion of the section, a more 0n<1 cellular response, or both. This appeared to be the 4,1 reason for the low relative numerical densities of lym ~ ^ g tt ^ .2 w (0 3 ^ Time (weeks) Figure 9. Volumefraction ofcollagen in tlye parenchymal lung tissue oj animals treated u tth silica < ------ ), titanium di oxide l*.-------), or saline alone // Significant differencesfrom the combined means oftlje saline-treated con trol groups are stx>wn as**,? <0 01. 612 Ilf Kumar nrr 1`IS'I 1m1 1 ;; \m , granulomatous lesions in human silicosis has long been recognized, but no information about the phenotype of the cells involved at different stages of the disease is avail able.25 The mechanisms underlying the demonstrated pattern of T lymphocyte and B lymphocyte responses are un known. A plausible hypothesis might be that T lympho cytes are recruited as a result of the secretion of chemotactic factors by macrophages.42 Macrophages that phagocytize silica are stimulated to secrete interleukin 1,43 which has been shown to be chemotactic for T lym phocytes in vitro.*4 Disruption of the alveolar epithelial sur face by the initiating in|ury may lead to the generation of immunogenic fragments from components of the intersti tial matnx such as collagen and elastin,4546 which have been shown to be capable of evoking an immunologic response.20 2' T lymphocytes participating in such a re sponse might be activated to secrete lymphokmes. such as interleukins 2. 4. 5, or 6, or mterferon-7, which can trigger the recruitment and/or proliferation of B lympho cytes47 At this stage one can only speculate on what role, if any, these cells play in the progression of the silicotic le sion. However, the belief that T lymphocytes contribute to the development of pulmonary fibrosis is supported by studies of lung disease induced by bleomycin.4849 The observation that the influx of T lymphocytes in silica-in duced inflammation precedes demonstrable fibrosis by 3 to 6 weeks is also consistent with such a hypothesis. Although experiments employing lung slices in organ cul ture suggest that the development of pulmonary fibrosis does not necessarily require the presence of an inflam matory response,505' the possibility exists that lymphokmes secreted by T lymphocytes stimulate fibroblast pro liferation and collagen synthesis, either directly or by acti vation of macrophages to secrete growth factors for fibroblasts. A direct role for T lymphocyte-denved media tors is suggested by T cell-dependent models of fibrogenesis52 53 and is consistent with evidence of reduced fibrosis after bleomycin-induced pulmonary injury in T lym phocyte-deficient nude mice.54 The properties of a lym phocyte-derived fibroblast-activating factor are the sub ject of continuing investigation.'2'3-55 Mediators that might trigger growth factor secretion could include lymphokines such as interfeukm-4 and interferon-7, which are known to activate macrophages.47 There is some experi mental evidence for such a role for interferon-7.23 T lymphocyte-denved lymphokmes, which might cause a polyclonal proliferation of B lymphocytes as hy pothesized above, could also stimulate immunoglobulin synthesis by these cells. This could provide an explana tion for the disordered regulation of immunoglobulin syn thesis observed in human silicosis, which is charactenzed by circulating antinuclear antibodies and rheumatoid fac tors and an increased incidence of autoimmune rj,s eases.56 57 Local synthesis of immunoglobulins rr 1 ISo explain the long-recognized presence of immur tins within the hyaline material in a classical smcot'c nodule." The results of the present study provide a founaaticr. for future in vitro expenments focusing on the possible contribution of lymphokmes to the progression of the sw cotic inflammatory lesion and the ensuing fibrosis References 1. Flint A: Pathologic features ot interstitial lung Cisease. ir';- stitial Lung Disease. Edited by Ml Schwarz, TE King Ph ; delphia. B. C Decker Inc. 1988. pp 45-62 2. Bitterman PB. Rennard SI. Hunnmghake GW Crvstal R" Human alveolar macrophage growth tactor :or iibrociasts Regulation and partial characterization. J Clin invest 19c; 70.806-822 3. Seppa H. Grotend.orst GR. Seppa S. Schiffrr \iar- GR- Platelet-derived growth factor is chemclac. . or -,tr: blasts. J Cell 81011982. 92.584-588 4 Shimokado K. Raines EW, Madtes DK. Barrett T8. Bert - EP. Ross R: A significant part ot macroonage-der.e: growth factor consists of at least two forms of PDGF C- 1985. 43.277-286 5. Kumar RK, Bennett RA, Brody AR: A homologue of plated' derived growth factor produced by rat alveolar mac: phages. FASEB J 1988.2:2272-2277 6. PostlethwaiteAE. Keski-OiaJ, Moses HL. Kar- Siimua tion ot the chemotactic migration of human . .casts transforming growth factor-beta. J Exp Med 1987. 165.25' - 256 7. Assoian RK, Fleurdeiys E, Stevenson HC, Miller PJ. Mac"-?s DK, Raines EW, Ross R, Spom MB. Expression and sect- tion of type beta transforming growth factor by activated ~- man macrophages. Proc Natl Acad Sci USA 1987.84 6027 6024 8. Fine A, Goldstein RH The effect of transforming grewth -> ' or-beta on cell proliferation and collagen Ic " oy - ' fibroblasts. J Bid Chem 1987,262 3897-39 9. Baird A. Mormede P. Bohlen P Immunoreact.ve tcrcc >- growth factor in ceils of pentoneal exudate suggests :s tity with macrophage-denved growth factor. Siocrem : phys Res Commun 1985. 126:358-364 10. Vilcek J, PaJombella VJ, Hennksen-DeStefano D. S*ec~1 C, Fetnman R, Hirai M, Tsu|imoto M: Fibroblast grevn hancmg activity of tumor necrosis factor and its re*aW * to other polypeptide growth factors. J Exp Med 1 vco. 740-745 11. Singh JP, Adams LD. Bonin 0: Mode of enhancement by human interleukin-1. J Ce; - - : 813-819 12 Wahl SM, Gately CL: Modulation of fibroblast grew'h lymphokme of human T cell and continuous T cel i`re - J Immunol 1983. 130:1226-1230 -3 Lamm... SM Pr Simulacytes * ! Kovacs growt.n J Leuxc 5. Lemairc phage : duced ( '6. Weiderr hanced 'macroo 99-103 ' Bitterm; monary phage-c J Chn In 3 Martinet RG Ex; growth idicpath 209 !. Kumar F say for p the me: kmes an Powand ) Kravis T Patholoc duced rr mediater 1232 : Schner C n bleom 7 Cathcart sive help nary fibre fmmunop - Kovacs derived c Am J Pat : Gross KE changes snica. Am Sficosis ; Wed with Arch Patf Daws GS f'VDothes ethloff L intra- a '`posed Cobbold.' ^ Therap subse Kumar Rk *7 actior 37 913-9 I 3 uammie PJ. Monroe JG Micnael Ai, Jonnson GO. Phillies SM. Prystowsky M0- Partial characterization ot a fibroblast- i t Stimulating factor produced by cloned murine T lympno- cytes Am J Pathol 1988, 130:289-295 4 Kovacs EJ. Kelley J: Secretion of macrophage-denved growth factor during acute lung injury induced by bleomycin. jLeukocyte Biol 1985. 37:1-14 -; Lemaire I. Beaudom H, Masse S, Grondin C. Alveolar macro phage stimulation of lung fibroblast growth m asbestos-in duced pulmonary fibrosis. Am J Pathol 1986. 122:205-211 a weidermann J, Adamson IY, Pert CB. Bowden DH. En hanced secretion of immunoreactive bombesin by alveolar macropnages exposed to silica. J Leukocyte Biol 1988. 43: 99-103 Bitterman rB. Adelberg 3. Crystal RG: Mechanisms ot pul monary fibrosis: Spontaneous release of the alveolar macro phage-derived growth factor in the interstitial lung disorders, j On Invest 1983, 72:1801 -1813 1 Martinet Y. Rom WN, Grotendorst GR. Martin GR, Crystal RG. Exaggerated spontaneous release of platelet-denved growth factor by alveolar macrophages from patients with idiopathic pulmonary fibrosis. N Engl J Med 1987, 317:202- t 209 } Kumar RK. Bennett RA, Brody AR: An enzyme immunoas i say for platelet-denved growth factor (POOR) Application to i the measurement of macrophage-denved PDGF. Mono tt. kines and Other Non-Lymphocytic Cytokines. Edited by MC Powanda. New York, Alan R. Liss Inc., 1988, pp 393-396 .3 Kravis TC, Ahmed A, Brown TE. Fulmer JD. Crystal RG: i Pathologic mechanisms in pulmonary fibrosis: collagen-in duced migration inhibition factor production and cytotoxicity mediated by lymphocytes. J Ctm-lnvest 1976. 58:1223- 1232 Schrier DJ, Phan SH, Ward PA: Cellular sensitrvity to collagen n bleomycin-treated rats. J Immunol 1982, 129:2156-2159 2 Cathcart MK, Emdur U, Ahtiala-Stewart K, Ahmad M: Exces sive helper T-cell function in patients with idiopathic pulmo- ' r*Y fibrosis: Correlation with disease activity. Clin Immunol *TMtiunopathol 1987. 43:382-394 Kovacs EJ, Kelley J. Lymphokine regulation of macrophage ienved growth factor secretion following pulmonary in|ury. . AmjPathot 1985. 121.261-268 1 Gross KB. White HJ, Smiler KL: Functional and morphologic changes in the lungs after a single Intratracheal insulation of sAca. Am Rev Respir Dis 1984. 129:833-839 sAcosis and Silicate Disease Committee: Diseases associ- ned with exposure to silica and nonfibrous silicate minerals. js pathol Lab Med 1988. 112:673-720 . Caws GS. Pathogenesis of silicosis: Current concepts and i >'iPotheses. Lung 1986. 164:1339-1354 j ^miolf LA, Gilmore LB. Brody AR, Hook GER:Induction i -t ntra- and extra-cellufar phospholipids m the lungs of rats j , fPosed to silica. BiochemJ 1986.233:111-118 j ; -vPbotd SP. Jayasunya A, Nash A. Prospero TD. Waldmann ^ Therapy with monoclonal antibodies by elimination of T- , j* subsets m vivo. Nature 1984. 312.548-551 PK: Immunogold-silver cytochemistry using a capil- aofion staining system. J Histochem Cytochem 1989. *913-917 Lymphocyte Populations m Silicosis 613 ns*i vt'/ : ^ v., -# 30 A eiDel ER Sampling ot tissue Storeological Methods, voi 1 \e.v York, Academic Press. 1979. po 63-100 31 Chancier DB. Hyae DM. Gin SN Morphometric estimates of ihiiltrafive cellular changes during the development of bleomycm-induced pulmonary fiorosis in hamsters Am J Pathol' 1983.112 170-177 32 Weioel ER Point counting methods. Stereological Methods, voi 1 New York. Academic Press. 1979, pp 101-161 33. Zzr JH: BiostatisticaJ Analysis. Englewood Cliffs, PrenticeHail Inc. 1974 34 CauOer JH, Rossman MD. Daniele RP Characterization ot the inflammatory response in expenmentat silicosis. Health Erects of Synthetic Silica Particulates. Edited by DO Dunrcm. Philadelphia. American Society tor Testing and Materi al. 1981. pp 94-117 35. Reiser KM. Hesterberg TW. Haschek WM, Last JA: Expenmental silicosis I Acute effects of mtratracheally instilled cuartz on collagen metabolism and morphologic character istics of rat lungs. Am J Pathol 1982, 107:176-185 36 M ner BE. Detnlof! LA. Hook GER: Silica-induced hypertroDPv ot type II cells in the lungs ot rats. Lab Invest 1986, 55. 153-163 37 Mathieson 8J. Sharrow SO' Cell-surface differentiation anti gens expressed on thymocytes and T cells ot the mouse, Differentiation Antigens in Lymphohemopoietic Tissues. Ed ited by M Miyasaka, Z Tmka. New York. Marcel Dekker, 1988. pp 65-151 38. Lanier IX, Phillips JH. What are natural killer cells? ISI Atlas Immunol 1988. 1:15-19 39. Pntchard JN. Holmes A, Evans JC. Evans N. Evans RJ, Mor gan A: The distnbution ot dust in the rat lung following administranon by inhalation and by single intratracheal injection. Environ Res 1985,36 268-297 40 Struhar D. Harbeck RJ. Mason RJ: Lymphocyte populations in lung tissue, bronchoalveolar lavage fluid, and peripheral blood in rafs at various times during the development ot sili cosis. Am Rev Respir Ois 1989,139:28-32 41 Burns CA. Zarkower A, Ferguson FG Munne immunologicaf and histological changes in response to chronic silica expo sure. Environ Res 1980. 21:298-307 42 Kaelm RM, Center DM, Bernardo J. Grant M. Snider GL: The role of macrophage-denved chemoattractant activities in the early inflammatory events of bleomycin-induced pulmonary iniury Am Rev Respir Dis 1983, 128:132-137 43. Schmidt JA, Oliver CN, Lepe-Zuruga X. Green I, Gery I: Sili ca stimulated monocytes release fibroblast proliferation fac tors identical to interleukin 1: A potential role for interleukin 1 m the pathogenesis ot silicosis. J Clin Invest 1984, 73: 1462-1472 44. Hunmnghake GW, Glazier J. Monick MM. Dinarefio CA: Inter leukin-1 is a chemotactic (actor tor human T4ympbocytes. Am Rev Respir Dis 1987,135.66-71 45. Cate CC. Burrell R Lung antigen induced cell-mediated im mune m|ury in chronic respiratory diseases. Am Rev Respir Dis 1974. 109:114-123 46 Damule TV, Ukhite V. Turmo GM. Mandl I: Induction of cellmeoiated immune response to pepbdes produced by enry- i 614 l/fJ Kumar /'W> \'n! / > ? Yr. matic digestion of eiastin from human lung parencnyma Proc Soc Exp Bkd) Wed 1980. 165 413-419 47. O'Garra A. Umland S. De France T, Christiansen J B-cell factors' are pleiotropic. Immunol Today 1988. 9.45-54 48 Thrall RS, Barton RW: A comparison of lymphocyte popula tions in lung tissue and in bronchoalveolar lavage fluid of rats at various times during the development of bleomycininduced pulmonary fibrosis. Am Rev Respir Dis 1984, 129 279-283 49. Kumar RK. Watkins SG. Lykke AWJ: Pulmonary responses to bleomycin-induced injury: An immunomorphologic and electron microscopic study. Exp Pathol 1985. 28.33-43 50. Fisher GL. Placke ME. In vitro models of lung toxicity. Toxi cology 1987, 47:71-93 51. Adamson IY. Young L. Bowaen DH: Relationship of alveolar epithelial m|ury and repair to the induction of pulmonary fi brosis. Am J Pathol 1988, 130:377-383 52. Pollaco S. Nicholas WL. Mitchell GF. Stewart AC: T-cell de pendent collagenous encapsulating response m the mouse liver to Mesocestoides corti (cestoda) Int J Parasitol 1978. 8:457-462 53. Wahl SM. Hunt DA. Allen JB. Wilder RL. Paglia L. Hand AR. Bactenal cell wall-induced granulomas' An in vivo model of T cell-dependent fibrosis, J Exp Med 1986. 163.884-902 54. Schner Dj. Phan SH, McGarry BM The effer (nu/nu) mutation on txeomycin-induced pulrr Am Rev Respir Dis 1983. 127 614-617 55. Agelli M, Wahl SM: Synthesis of biologically active t.cr-activating factor (FAF) by xenopus oocytes m;ectec lymphocyte mRNA. Cell Immunol 1987. 110 183-190 56. Burrell R Immunological aspects of silica. Health ElSynthetic Silica Particulates. Edited by DO Qunnom : detphia, American Society tor Testing and Materials pp 82-93 57. Uber CL. McReynolds RA. immunotoxicolog-. - - :,Ca Cnt Rev Toxicol 1982. 10 303-319 58. Vigliam EC. Perms B An immunological apc-v s;r. - . sis. J Occup Med 1959, 1 319-328 Acknowledgment t J Ifnmunopht Characters I Mon-Hoc^ j ,V. E. Katzin,' M. D. pdR. R- Tubbst - m the Departments r.y I.'h'foiiv.t The Clcreic,1 ,: t land. Ohio I thank Mrs. Daphne Greening and Mrs Mignon Wegner tor. tance in the preparation of tissue sections, Mr --`rea Ckolbe for caring for the experimental animals. . -e 2-; ment of Medical Illustration. University of New S. .Va.es preparation of the figures '.fuse, mixed smal. ; DML) are a heteroi ! tin s lymphomas. Th l -imunophenotyptc < ; niL. and tbey indue. j a unclear whether . f `.anal, oiigoclonai, o i e? monoclonal anti 1 SI. L\2, and L26) . j imbedded, B5 fixed t . 'udied. This method < ! ::r>n between cell mot ' :'P* compared with f O' In addition, Soi. t-itlon was used to u : 'y neoplastic cell pop | -'ion oftarge B lympi 'nmunohistology. In t immunoglobulin t ~n;m section tmmui ~r<ytology, or both, si lymphocytes art ^tases(l2ofl3). l ded predominantly t-ohulin gene rearra 'r cases, but no T j ~mts were detected. j '* monoclonal lymp. | ' lnn neopIastic T ce i ",9- 135:615-621J ^-se mixed small and I, or diffuse mi> ^as, were described __ '^^Sling ct a prolifer ** fe,|culum cells" ("h