Acta Tropica, 49(1991)45-55
45
Elsevier ACTROP 00130
Histopathology of Brugia malayi-infected nude mice after immune-reconstitution A.C. Vickery 1, K.H. Albertine 2, J.K. N a y a r 3 a n d B . H . K w a 1 1College of Public Health, University of South Florida, Tampa, FL, U.S.A., 2Department of Medicine, Jefferson Medical College, Philadelphia, PA, U.S.A., and 3Florida Medical Entomology Laboratory, University of Florida, Vero Beach, FL, U.S.A. (Received 24 January 1990; accepted 10 September 1990) Long term ( > 200 day) Brugia malay#infected nude mice with grossly dilated lymphatics were reconstituted with 108 primed spleen cells from heterozygous donors. Histological and ultrastructural examination at 7, 14, 21 and 28 days post-reconstitution revealed progressive fbrosis, obliterative lymph thrombus formation, interstitial infiltrates and extensive perilymphangitis. Formation of lymph thrombi/granulomas was associated with killing of adult worms and microfilariae, and the predominant cell types involved were large granular macrophages, Langhan's giant cells and eosinophils. Thus, the ability to initiate the formation of obstructive lesions in the dilated lymphatics of chronically parasitized nude mice by immunological reconstitution, suggests that several complex mechanisms might operate in stages to cause filarial elephantiasis. Key words: Brugia malayi; Filariasis; Lymphatics; Nude mice; Reconstitution
Introduction Filariasis exists in nature as a broad spectrum of clinical manifestations varying from inapparent infection to obstructive lymphatic pathology, culminating in some cases in elephantiasis (Ottesen, 1980). It is generally accepted that variations in host immune responses to the parasite reflect the range of clinical disease observed (Ottesen, 1980; Nelson, 1981; Haque et al., 1983; Kwa and Mak, 1987). Cell mediated immune responses to adult worm antigens are thought to be involved in the severe granulomatous inflammation characteristic of elephantiasis (Nelson, 1966; Ottesen, 1980; Partono, 1982). However, it remains unclear whether the somatic antigens released by dead and disintegrating worms, or perhaps excretory/secretory products of living worms, are the cause of the granulomatous reaction (von Lichtenberg, 1957; Warren, 1971). Previous studies with the athymic nude mouse model have shown that long-term Brugia malayi infection results in progressive lymphedema of affected limbs and massive lymphangiectasis (Vickery et al., 1985) in the absence of thymus-dependent immune responses. The present study was conducted to determine whether reconstiCorrespondence address." A.C. Vickery, College of Public Health, University of South Florida, Tampa, FL 33612, U.S.A. 0001-706X/91/$03.50 © 1991 Elsevier Science Publishers B.V.
46 tution of similarly infected mice by adoptive transfer of primed spleen cells, would result in killing of adult worms and development of histopathological lesions which mimic those described in elephantiasis.
Materials and Methods
Animals Male congenitally athymic nude (nu/nu) and phenotypically normal heterozygotic (nu/+) C3H/HeN mice, maintained in a closed breeding colony under barrier conditions, were used in this study. Mice were progeny of sibling matings of male nude and female heterozygotes.
Parasitology Jirds, microfilaremic with Brugia malayi, were used to feed Aedes aegypti mosquitoes for production of infective larvae (L3), as previously described (Vincent et al., 1982). Fifty L 3 were inoculated subcutaneously into the left groin of 7 to 10 week old nude or heterozygous mice (Vickery and Vincent, 1984).
Pathology All mice used in this study had been parasitized for 200 to 250 days and exhibited enlarged regional lymph nodes and dilated subcutaneous lymphatics. Additionally, approximately 10% of these parasitized nude mice had swollen, edematous limbs and skin changes. Mice which exhibited these conditions were called elephantoid (Vickery et al., 1985). We used five groups of 3-4 nude mice each for histologic and ultrastructural examination: one group of parasitized controls and four groups of reconstituted nude mice. Reconstitution was accomplished by intraperitoneal injection of 108 primed spleen cells obtained from heterozygous mice exposed to 50 L 3 B. malayi 21 days previously (Vickery et al., 1983). Periods of reconstitution were 7, 14, 21, and 28 days. Nude mice were sacrificed by cervical dislocation, then the left and right superficial inguinal lymph nodes and associated lymphatic vessels were initially fixed in situ by instilling 2% glutaraldehyde/l% paraformaldehyde in Millonig's phosphate buffer (pH 7.4, 320 mosM) at room temperature. Excised tissues were stored in fixative for at least 24 h then those for histologic observation were dehydrated in a graded ethanol series, placed in chloroform to elute the fat, and embedded in glycol methacrylate. 2-gm thick sections were stained with 1% toluidine blue or hematoxylin and eosin. Tissues for ultrastructural viewing were post-fixed in 1% osmium tetroxide, dehydrated in a graded acetone series and embedded in Polybed 812 (Ladd Research Industries, Burlington, VT, U.S.A.). Thin sections (80-90 nm), cut with a diamond knife, were counterstained with uranyl acetate and lead citrate, and observed with a Philips 301 transmission electron microscope.
47 Results
Histopathologic examination of parasitized lymphatics taken from nude mice infected with 50 L 3 of Brugia malayi for more than 200 days showed large, redundant sacculations in which small, non-obstructive lymph thrombi were infrequently located. The lymph thrombi were composed primarily of small mononuclear cells, macrophages and occasional large monocuclear cells. Although in random tissue sections some lymph thrombi appeared free-floating in the lymphatic lumen, in fact most, if not all, lymph thrombi were anchored to the vessel wall (Fig. 1). These attachments were detected by examination of serial sections. Despite the presence of lymph thrombi, interstitial cellular infiltrates were rarely evident outside dilated lymphatics in which viable adult B. malayi were found in the sacculations (Fig. 2). Cellular components of the perilymphatic interstitium were interstitial cells and mast cells. Perilymphatic granuloma formation was rarely seen. As reported previously, we found no histologic evidence of mast cell degranulation not even when mast cells flanked lymphatic varicosities harboring viable adult worms (Vickery et al., 1985). Primed spleen cell reconstitution of chronically parasitized nude mice for periods of 7-28 days was associated with gradual formation of massive lymph thrombi and interstitial infiltrates. These phenomena were observed only in close approximation to adult worms and not in unparasitized tissues. At 7 days post-reconstitution luminal aggregates of lymphocytes, some of which were pavemented, large granular macrophages and multinucleated Langhan's giant ceils, as well as interstitial infiltrates of lymphocytes, were evident (Fig. 3). Further-
Fig. 1. Tortuous and distended lymphaticvessel (L) of a mouse infected for 210 days, containing small and isolated lymph thrombi (T) comprised of lymphocytesand macrophages.
48
Fig. 2. Section through an adult worm (B) in the lumen of a lymphatic vessel (L) of a mouse infected for 210 days. There is a paucity of inflammatory/immune cells in the vessel and mast cells (arrowheads) are present in the interstitium.
Fig. 3. Section through a lymphatic vessel of a mouse 7 days after reconstitution containing large and densely packed lymph thrombi (T). Cellular infiltrates (I) in the perilymphatic tissue are predominantly lymphocytes.
49 more, the lymph stained intensely and contained swirls of amorphous lamellae, both indicative of high lymph protein concentration. The histopathologic response at 14 days post-reconstitution consisted of massive endolymphatic granulomas obliterating some lymphatic saccular lumina. Perilymphatic granuloma formation became more expansive compared to lymphatics at 7 days post-reconstitution. There was evidence that the formation of endolymphatic granulomas was at least partly associated with destruction of adult worms (Fig. 4). Frequently, remnants of necrotised adult worms were seen within aggregates of large mononuclear phagocytes and eosinophils or more rarely, apparently within the cytoplasmic syncytium of a single giant cell. Microfilariae were rarely seen either trapped within clusters of eosinophils and mononuclear phagocytic cells or partially necrotised within granulomatous aggregates of mononuclear phagocytes, Langhan's foreign body giant cells and eosinophils. At 21 days post-reconstitution lymph thrombus formation was patchy and found around necrotized worms. In some instances the lymphatic saccules were obliterated by thrombi (Fig. 5) but in others this was not the case. Interstitial infiltrates became more widespread than at the two earlier reconstitution time points. Perilymphangitis, ostensibly composed of small lymphocytes, were prevalent at 21 days post-reconstitution. Much the same histopathologic picture was evident at 28 days post-reconstitution. Both obliterative lymph thrombus formation and expansive perilymphangitis were seen. Evidence of disintegrating parasites within granulomas was seldom seen at this stage, suggesting that phagocytosis of microfilariae and adult worms first seen
Fig. 4. Large and denselypacked lymph thrombi (T) attached to the endothelium of a lymphatic saccule (L) of a mouse 14 days after rcconstitution.One thrombus contains fragments of an adult worm (B) and interstitial cellular infiltrates (I) are visible.
50
Fig. 5. Section through a lymph thrombus (T) in the lymphatic lumen (L) of a mouse 21 days after reconstitution. Extensiveinterstitial infiltrate (I) is visible. at 14 days might have completely destroyed the parasites by 28 days post-reconstitution. Transmission electron microscopy was used to ascertain the cytopathologic responses in these experiments. The mere presence of a living adult worm in an unreconstituted nude mouse was insufficient to cause perilymphatic interstitial fibrosis (Fig. 6A). At 7 days post-reconstitution, however, marked fibrosis was seen (Fig. 6B). By day 14 post-reconstitution, perilymphangitis was present. Both granular and agranular leucocytes comprised the granulomatous tissue in the lymphatic adventitia (Fig. 6C). At day 21 post-reconstitution, not only were interstitial cellular infiltrates seen, there also was edema formation which was ultrastructurally visible as a washed-out, distended interstitial compartment (Fig. 6D). Often large granular macrophages with cytoplasmic inclusions were rosetted with lymphocytes and smaller macrophages. A phenomenon that captured our attention was the formation of specialized cellcell contacts between activated lymphocytes and macrophages (Fig. 7). Dense bands of filamentous material were condensed along the contact surfaces, a condition reminiscent of gap junctions. Although we do not know the significance of these specialized cell contacts, it seems reasonable to suggest a transduction or antigen presentation mechanism designed to augment the immune response.
Discussion One of the classic unresolved questions in lymphatic filariasis is what causes elephantiasis? It has been suggested by some workers that the pathological changes seen in
51
the chronic obstructive lesions which probably result in elephantiasis, are associated with hypersensitive immunopathology to dead worms and antigenic material released by the disintegrating parasites (von Lichtenberg, 1957; Warren, 1971). Others have suggested that living worms cause mechanical damage or secrete metabolic products which induce cellular immune responses which kill the worms. Thus, parasiticidal
52
Fig. 6. Transmission electron micrographs showing the structural changes resulting from reconstitution in the lymphatic (L) wails of mice infected for 210 days. Panel A: Unreconstituted. Outside the continuous endothelium (E) and fragmented basal lamina is loose connective tissue (C). Panel B: 7 day reconstitution. A macrophage (M) with a ruffled surface is adherent to the lymphatic wall. The endothelial cells (E) are scalloped and folded. Substantial amounts of collagen have been deposited in the perilymphatic connective tissue (C) space. Panel C: 14 day reconstitution. Lymphatic endothelial cells (E) are scalloped and cellular infiltrates (I) are visible in the interstitial space. Panel D: 21 day reconstitution. The interstitium is distended by edema fluid (*).
53
Fig. 7. Cell to cellcontact betweena lymphocyte(LC) and macrophage(M) in a dilatedlymphaticfrom a mouseinfectedfor 210 days. Two densejunctionalcomplexes(arrowheads)are locatedwherethe cells are in contact.
host reactions to living parasites might ultimately give rise to the lesions seen in chronic obstructive disease and elephantiasis (Denham and Nelson, 1976, Partono, 1982). The B. malayi/athymic nude mouse model developed in this laboratory has enabled us to study the pathogenesis of chronic filarial infections, and serves as a model to study the pathological changes associated with early non-obstructive and late obstructive lymphatic lesions. Previous studies have shown that congenitally athymic nude mice with chronic (> 200 day) infection with B. malayi developed persistent lymphedema and progressive lymphangiectasis resulting in massive lymphatic dilations (Vickery et al., 1985). This condition in nude mice was termed 'elephantoid' and could be reversed by removal of the worms. That model served to mimic the early stages of human filariasis where similarly dilated tortuous lymphatics were revealed by lymphograms (Cohen et al., 1961). In those patients observed, dilation and tortuosity was not accompanied by obliterative cellular reactions or mechanical blockage of the lymphatics and could be reversed by chemotherapy. The present study has revealed that adoptive transfer of primed spleen cells to nude mice with dilated lymphatics results in a progressive buildup of massive lymph thrombi and interstitial infiltrates over a period of 7-28 days post-reconstitution. By 14 days, obliterative lymph thrombus formation and expansive perilymphangitis were evident. The development of these lesions was associated with concurrent destruction of both adult worms and microfilariae in massive thrombi/granulomas composed of large mononuclear cells, multinucleated Langhan's giant cells and eosinophils. These obliterative granulomas appeared to reach a peak at 14 days, and destruction of
54 most of the parasites appeared to be accomplished by day 28 post-reconstitution. Interestingly, the time course and cellular composition of the reaction, as well as the time of parasite killing, was very similar to earlier reports of B. pahangi larval killing in immunologically normal heterozygous mice (Vincent et al., 1983). Transmission electron microscopy confirmed that cytopathological responses, such as extensive fibrosis, appeared as early as 7 days post-reconstitution. Perilymphangitis, granuloma formation, edema and interstitial infiltrates developed progressively over the 28 day post-reconstitution period studied. The predominant cell type involved in the thrombi were confirmed as large macrophages which appeared to be associated with the destruction of the worms. Multinucleated Langhan's giant cells were sometimes seen within the granulomas, but unlike the large macrophages, have not been seen directly adhering to adult worms or microfilariae. Eosinophils were commonly seen associated with the granulomas and cell aggregates around the worms, but whether they were involved in destruction of worm cuticle as reported by Yates and Higashi (1986) could not be confirmed in this study. Current studies in this laboratory which focus on histological and ultrastructural observations during the period between 7 and 14 days post-reconstitution may reveal the roles of the different cell types in the early phase of worm-killing. The present study suggests, therefore, that in the nude mouse/B, malayi model, the syndrome of obliterative lymph thrombi, fibrosis, interstitial infiltration and expansive perilymphangitis is caused by host cell-mediated and parasiticidal immune responses against living worms. In the mouse, a single such episode can be elicited by administration of immunocompetent cells from primed heterozygous donors. The presence of living B. malayi adult worms is essential to produce the initial lymphatic dilation in nude mice (Vickery et al., 1985). However, in the absence of immunologic reconstitution, worms which die fail to elicit obliterative phenomena (Vincent et al., 1984). These results are consistent with the suggestion that clinical obstructive disease and elephantiasis in man may be an ultimate consequence of active cell-mediated immune responses to the living parasite. In a fashion similar to the single episode induced in mice by reconstitution, perhaps repeated exposure to infective larvae results in exacerbations of cellular hypersensitivity which, by killing the worms, leads to clinical filariasis. Thus, obstructive cellular responses are superimposed upon dilated and tortuous, but still functioning, lymphatics. The 'elephantoid' nude mouse, chronically parasitized with B. malayi does not exhibit all of the characteristics of filarial elephantiasis. However, it's inability to mount thymus-dependent cell-mediated immune responses to filarial antigen, has allowed us to observe the phenomenon of lymphatic dilation in the presence of living parasites. That worm killing and obstructive lymphatic lesions can be induced by immune reconstitution, suggests that the characterization of the seemingly complex immunologic as well as non-immunologic mechanisms of lymphatic damage will be possible.
Acknowledgements We wish to thank the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (ID No. 870382) for their support of this study.
55
References Cohen, L.B., Nelson, G., Wood, A.M., Manson-Bahr, P.E.C. and Bowen, R. (1961) Lymph-angiography in filarial lymphoedema and elephantiasis. Am. J. Trop. Med. Hyg. 10, 843-848. Denham, D.A. and Nelson, G.S. (1976) Nematodes parasitic in the lymphatic system. In: Soulsby, E.J.L. (ed.), Pathophysiology of Parasitic Infection, Academic Press. Haque, A., Capron, A., Ouaissi, A., Kouemeni, L., Lejeune, J.P., Bonnel, B. and Pierce, R. (1983) Immune unresponsiveness and its possible relation to filarial disease. Contr. Microbiol. Immunol. 7, 9-21. Kwa, B.H. and Mak, J.W. (1987) Filarids (excluding D. immitis). In: Soulsby, E.J.L. (ed.), Immune Responses in Parasitic Infection: Immunology, Immunopathology and Immunoprophylaxis. Vol. I, Chapt. 10, CRC Press. Nelson, G. S. (1966) The pathology of filarial infections. Helminthol. Abstr. 35, 31 I. Nelson, G. S. (1981) Issues in filariasis - a century of enquiry and a century of failure. Acta Trop. 38, 197. Ottesen, E.A. (1980). Immunopathology of lymphatic filariasis in man. Springer Seminars. Immunopathology 2, 373-385. Partono, F. (I 982) Elephantiasis and its relation to filarial immunity. Southeast Asian J. Trop. Med. Publ. Health 13, 275. Vickery, A.C. and Vincent, A. L. (1984) Immunity to Brugia pahangi in athymic nude and normal mice: eosinophilia, antibody and hypersensitivity responses. Parasite Immunol. 6, 545-559. Vickery, A.C., Vincent, A.L. and Sodeman, W.A. (1983) Effect of immune reconstitution and resistance to Brugia pahangi in congenitally athymic nude mice. J. Parasitol. 69, 478-485. Vickery, A.C., Nayar, J.K., and Albertine, K.H. (1985) Differential pathogenicity of Brugia malayi, B. patei and B. pahangi in immunodeficient mice. Acta Trop. 42, 353-363. Vincent, A.L., Vickery, A.C., Lotz, M.J. and Desai, U. (1984) The lymphatic pathology of Brugia pahangi in nude (athymic) and thymic mice, C3H/HeN. J. Parasitol. 70, 48-56. Vincent, A.L., Vickery, A.C., Winters, A. and Sodeman, W.A. (1982) Life cycle of Brugia pahangi (Nematoda) in nude mice, C3H/HeN (nu/nu). J. Parasitol. 68, 553-561. Von Lichtenberg, F. (1957) The early phase of endemic bancroftian filariasis in the male. Pathological study. J. Mr. Sinai Hosp. 24, 983-1000. Warren, K.S. (1971) Worms. In: Samter, M. (ed.), Immunological Diseases, Little, Brown. Yates, J.A. and Higashi, G.I. (1986) Ultrastructural observations on the fate of Brugia rnalayi in jirds previously vaccinated with irradiated infective stage larvae. Am. J. Trop. Med. Hyg. 35, 982-987.
45
Elsevier ACTROP 00130
Histopathology of Brugia malayi-infected nude mice after immune-reconstitution A.C. Vickery 1, K.H. Albertine 2, J.K. N a y a r 3 a n d B . H . K w a 1 1College of Public Health, University of South Florida, Tampa, FL, U.S.A., 2Department of Medicine, Jefferson Medical College, Philadelphia, PA, U.S.A., and 3Florida Medical Entomology Laboratory, University of Florida, Vero Beach, FL, U.S.A. (Received 24 January 1990; accepted 10 September 1990) Long term ( > 200 day) Brugia malay#infected nude mice with grossly dilated lymphatics were reconstituted with 108 primed spleen cells from heterozygous donors. Histological and ultrastructural examination at 7, 14, 21 and 28 days post-reconstitution revealed progressive fbrosis, obliterative lymph thrombus formation, interstitial infiltrates and extensive perilymphangitis. Formation of lymph thrombi/granulomas was associated with killing of adult worms and microfilariae, and the predominant cell types involved were large granular macrophages, Langhan's giant cells and eosinophils. Thus, the ability to initiate the formation of obstructive lesions in the dilated lymphatics of chronically parasitized nude mice by immunological reconstitution, suggests that several complex mechanisms might operate in stages to cause filarial elephantiasis. Key words: Brugia malayi; Filariasis; Lymphatics; Nude mice; Reconstitution
Introduction Filariasis exists in nature as a broad spectrum of clinical manifestations varying from inapparent infection to obstructive lymphatic pathology, culminating in some cases in elephantiasis (Ottesen, 1980). It is generally accepted that variations in host immune responses to the parasite reflect the range of clinical disease observed (Ottesen, 1980; Nelson, 1981; Haque et al., 1983; Kwa and Mak, 1987). Cell mediated immune responses to adult worm antigens are thought to be involved in the severe granulomatous inflammation characteristic of elephantiasis (Nelson, 1966; Ottesen, 1980; Partono, 1982). However, it remains unclear whether the somatic antigens released by dead and disintegrating worms, or perhaps excretory/secretory products of living worms, are the cause of the granulomatous reaction (von Lichtenberg, 1957; Warren, 1971). Previous studies with the athymic nude mouse model have shown that long-term Brugia malayi infection results in progressive lymphedema of affected limbs and massive lymphangiectasis (Vickery et al., 1985) in the absence of thymus-dependent immune responses. The present study was conducted to determine whether reconstiCorrespondence address." A.C. Vickery, College of Public Health, University of South Florida, Tampa, FL 33612, U.S.A. 0001-706X/91/$03.50 © 1991 Elsevier Science Publishers B.V.
46 tution of similarly infected mice by adoptive transfer of primed spleen cells, would result in killing of adult worms and development of histopathological lesions which mimic those described in elephantiasis.
Materials and Methods
Animals Male congenitally athymic nude (nu/nu) and phenotypically normal heterozygotic (nu/+) C3H/HeN mice, maintained in a closed breeding colony under barrier conditions, were used in this study. Mice were progeny of sibling matings of male nude and female heterozygotes.
Parasitology Jirds, microfilaremic with Brugia malayi, were used to feed Aedes aegypti mosquitoes for production of infective larvae (L3), as previously described (Vincent et al., 1982). Fifty L 3 were inoculated subcutaneously into the left groin of 7 to 10 week old nude or heterozygous mice (Vickery and Vincent, 1984).
Pathology All mice used in this study had been parasitized for 200 to 250 days and exhibited enlarged regional lymph nodes and dilated subcutaneous lymphatics. Additionally, approximately 10% of these parasitized nude mice had swollen, edematous limbs and skin changes. Mice which exhibited these conditions were called elephantoid (Vickery et al., 1985). We used five groups of 3-4 nude mice each for histologic and ultrastructural examination: one group of parasitized controls and four groups of reconstituted nude mice. Reconstitution was accomplished by intraperitoneal injection of 108 primed spleen cells obtained from heterozygous mice exposed to 50 L 3 B. malayi 21 days previously (Vickery et al., 1983). Periods of reconstitution were 7, 14, 21, and 28 days. Nude mice were sacrificed by cervical dislocation, then the left and right superficial inguinal lymph nodes and associated lymphatic vessels were initially fixed in situ by instilling 2% glutaraldehyde/l% paraformaldehyde in Millonig's phosphate buffer (pH 7.4, 320 mosM) at room temperature. Excised tissues were stored in fixative for at least 24 h then those for histologic observation were dehydrated in a graded ethanol series, placed in chloroform to elute the fat, and embedded in glycol methacrylate. 2-gm thick sections were stained with 1% toluidine blue or hematoxylin and eosin. Tissues for ultrastructural viewing were post-fixed in 1% osmium tetroxide, dehydrated in a graded acetone series and embedded in Polybed 812 (Ladd Research Industries, Burlington, VT, U.S.A.). Thin sections (80-90 nm), cut with a diamond knife, were counterstained with uranyl acetate and lead citrate, and observed with a Philips 301 transmission electron microscope.
47 Results
Histopathologic examination of parasitized lymphatics taken from nude mice infected with 50 L 3 of Brugia malayi for more than 200 days showed large, redundant sacculations in which small, non-obstructive lymph thrombi were infrequently located. The lymph thrombi were composed primarily of small mononuclear cells, macrophages and occasional large monocuclear cells. Although in random tissue sections some lymph thrombi appeared free-floating in the lymphatic lumen, in fact most, if not all, lymph thrombi were anchored to the vessel wall (Fig. 1). These attachments were detected by examination of serial sections. Despite the presence of lymph thrombi, interstitial cellular infiltrates were rarely evident outside dilated lymphatics in which viable adult B. malayi were found in the sacculations (Fig. 2). Cellular components of the perilymphatic interstitium were interstitial cells and mast cells. Perilymphatic granuloma formation was rarely seen. As reported previously, we found no histologic evidence of mast cell degranulation not even when mast cells flanked lymphatic varicosities harboring viable adult worms (Vickery et al., 1985). Primed spleen cell reconstitution of chronically parasitized nude mice for periods of 7-28 days was associated with gradual formation of massive lymph thrombi and interstitial infiltrates. These phenomena were observed only in close approximation to adult worms and not in unparasitized tissues. At 7 days post-reconstitution luminal aggregates of lymphocytes, some of which were pavemented, large granular macrophages and multinucleated Langhan's giant ceils, as well as interstitial infiltrates of lymphocytes, were evident (Fig. 3). Further-
Fig. 1. Tortuous and distended lymphaticvessel (L) of a mouse infected for 210 days, containing small and isolated lymph thrombi (T) comprised of lymphocytesand macrophages.
48
Fig. 2. Section through an adult worm (B) in the lumen of a lymphatic vessel (L) of a mouse infected for 210 days. There is a paucity of inflammatory/immune cells in the vessel and mast cells (arrowheads) are present in the interstitium.
Fig. 3. Section through a lymphatic vessel of a mouse 7 days after reconstitution containing large and densely packed lymph thrombi (T). Cellular infiltrates (I) in the perilymphatic tissue are predominantly lymphocytes.
49 more, the lymph stained intensely and contained swirls of amorphous lamellae, both indicative of high lymph protein concentration. The histopathologic response at 14 days post-reconstitution consisted of massive endolymphatic granulomas obliterating some lymphatic saccular lumina. Perilymphatic granuloma formation became more expansive compared to lymphatics at 7 days post-reconstitution. There was evidence that the formation of endolymphatic granulomas was at least partly associated with destruction of adult worms (Fig. 4). Frequently, remnants of necrotised adult worms were seen within aggregates of large mononuclear phagocytes and eosinophils or more rarely, apparently within the cytoplasmic syncytium of a single giant cell. Microfilariae were rarely seen either trapped within clusters of eosinophils and mononuclear phagocytic cells or partially necrotised within granulomatous aggregates of mononuclear phagocytes, Langhan's foreign body giant cells and eosinophils. At 21 days post-reconstitution lymph thrombus formation was patchy and found around necrotized worms. In some instances the lymphatic saccules were obliterated by thrombi (Fig. 5) but in others this was not the case. Interstitial infiltrates became more widespread than at the two earlier reconstitution time points. Perilymphangitis, ostensibly composed of small lymphocytes, were prevalent at 21 days post-reconstitution. Much the same histopathologic picture was evident at 28 days post-reconstitution. Both obliterative lymph thrombus formation and expansive perilymphangitis were seen. Evidence of disintegrating parasites within granulomas was seldom seen at this stage, suggesting that phagocytosis of microfilariae and adult worms first seen
Fig. 4. Large and denselypacked lymph thrombi (T) attached to the endothelium of a lymphatic saccule (L) of a mouse 14 days after rcconstitution.One thrombus contains fragments of an adult worm (B) and interstitial cellular infiltrates (I) are visible.
50
Fig. 5. Section through a lymph thrombus (T) in the lymphatic lumen (L) of a mouse 21 days after reconstitution. Extensiveinterstitial infiltrate (I) is visible. at 14 days might have completely destroyed the parasites by 28 days post-reconstitution. Transmission electron microscopy was used to ascertain the cytopathologic responses in these experiments. The mere presence of a living adult worm in an unreconstituted nude mouse was insufficient to cause perilymphatic interstitial fibrosis (Fig. 6A). At 7 days post-reconstitution, however, marked fibrosis was seen (Fig. 6B). By day 14 post-reconstitution, perilymphangitis was present. Both granular and agranular leucocytes comprised the granulomatous tissue in the lymphatic adventitia (Fig. 6C). At day 21 post-reconstitution, not only were interstitial cellular infiltrates seen, there also was edema formation which was ultrastructurally visible as a washed-out, distended interstitial compartment (Fig. 6D). Often large granular macrophages with cytoplasmic inclusions were rosetted with lymphocytes and smaller macrophages. A phenomenon that captured our attention was the formation of specialized cellcell contacts between activated lymphocytes and macrophages (Fig. 7). Dense bands of filamentous material were condensed along the contact surfaces, a condition reminiscent of gap junctions. Although we do not know the significance of these specialized cell contacts, it seems reasonable to suggest a transduction or antigen presentation mechanism designed to augment the immune response.
Discussion One of the classic unresolved questions in lymphatic filariasis is what causes elephantiasis? It has been suggested by some workers that the pathological changes seen in
51
the chronic obstructive lesions which probably result in elephantiasis, are associated with hypersensitive immunopathology to dead worms and antigenic material released by the disintegrating parasites (von Lichtenberg, 1957; Warren, 1971). Others have suggested that living worms cause mechanical damage or secrete metabolic products which induce cellular immune responses which kill the worms. Thus, parasiticidal
52
Fig. 6. Transmission electron micrographs showing the structural changes resulting from reconstitution in the lymphatic (L) wails of mice infected for 210 days. Panel A: Unreconstituted. Outside the continuous endothelium (E) and fragmented basal lamina is loose connective tissue (C). Panel B: 7 day reconstitution. A macrophage (M) with a ruffled surface is adherent to the lymphatic wall. The endothelial cells (E) are scalloped and folded. Substantial amounts of collagen have been deposited in the perilymphatic connective tissue (C) space. Panel C: 14 day reconstitution. Lymphatic endothelial cells (E) are scalloped and cellular infiltrates (I) are visible in the interstitial space. Panel D: 21 day reconstitution. The interstitium is distended by edema fluid (*).
53
Fig. 7. Cell to cellcontact betweena lymphocyte(LC) and macrophage(M) in a dilatedlymphaticfrom a mouseinfectedfor 210 days. Two densejunctionalcomplexes(arrowheads)are locatedwherethe cells are in contact.
host reactions to living parasites might ultimately give rise to the lesions seen in chronic obstructive disease and elephantiasis (Denham and Nelson, 1976, Partono, 1982). The B. malayi/athymic nude mouse model developed in this laboratory has enabled us to study the pathogenesis of chronic filarial infections, and serves as a model to study the pathological changes associated with early non-obstructive and late obstructive lymphatic lesions. Previous studies have shown that congenitally athymic nude mice with chronic (> 200 day) infection with B. malayi developed persistent lymphedema and progressive lymphangiectasis resulting in massive lymphatic dilations (Vickery et al., 1985). This condition in nude mice was termed 'elephantoid' and could be reversed by removal of the worms. That model served to mimic the early stages of human filariasis where similarly dilated tortuous lymphatics were revealed by lymphograms (Cohen et al., 1961). In those patients observed, dilation and tortuosity was not accompanied by obliterative cellular reactions or mechanical blockage of the lymphatics and could be reversed by chemotherapy. The present study has revealed that adoptive transfer of primed spleen cells to nude mice with dilated lymphatics results in a progressive buildup of massive lymph thrombi and interstitial infiltrates over a period of 7-28 days post-reconstitution. By 14 days, obliterative lymph thrombus formation and expansive perilymphangitis were evident. The development of these lesions was associated with concurrent destruction of both adult worms and microfilariae in massive thrombi/granulomas composed of large mononuclear cells, multinucleated Langhan's giant cells and eosinophils. These obliterative granulomas appeared to reach a peak at 14 days, and destruction of
54 most of the parasites appeared to be accomplished by day 28 post-reconstitution. Interestingly, the time course and cellular composition of the reaction, as well as the time of parasite killing, was very similar to earlier reports of B. pahangi larval killing in immunologically normal heterozygous mice (Vincent et al., 1983). Transmission electron microscopy confirmed that cytopathological responses, such as extensive fibrosis, appeared as early as 7 days post-reconstitution. Perilymphangitis, granuloma formation, edema and interstitial infiltrates developed progressively over the 28 day post-reconstitution period studied. The predominant cell type involved in the thrombi were confirmed as large macrophages which appeared to be associated with the destruction of the worms. Multinucleated Langhan's giant cells were sometimes seen within the granulomas, but unlike the large macrophages, have not been seen directly adhering to adult worms or microfilariae. Eosinophils were commonly seen associated with the granulomas and cell aggregates around the worms, but whether they were involved in destruction of worm cuticle as reported by Yates and Higashi (1986) could not be confirmed in this study. Current studies in this laboratory which focus on histological and ultrastructural observations during the period between 7 and 14 days post-reconstitution may reveal the roles of the different cell types in the early phase of worm-killing. The present study suggests, therefore, that in the nude mouse/B, malayi model, the syndrome of obliterative lymph thrombi, fibrosis, interstitial infiltration and expansive perilymphangitis is caused by host cell-mediated and parasiticidal immune responses against living worms. In the mouse, a single such episode can be elicited by administration of immunocompetent cells from primed heterozygous donors. The presence of living B. malayi adult worms is essential to produce the initial lymphatic dilation in nude mice (Vickery et al., 1985). However, in the absence of immunologic reconstitution, worms which die fail to elicit obliterative phenomena (Vincent et al., 1984). These results are consistent with the suggestion that clinical obstructive disease and elephantiasis in man may be an ultimate consequence of active cell-mediated immune responses to the living parasite. In a fashion similar to the single episode induced in mice by reconstitution, perhaps repeated exposure to infective larvae results in exacerbations of cellular hypersensitivity which, by killing the worms, leads to clinical filariasis. Thus, obstructive cellular responses are superimposed upon dilated and tortuous, but still functioning, lymphatics. The 'elephantoid' nude mouse, chronically parasitized with B. malayi does not exhibit all of the characteristics of filarial elephantiasis. However, it's inability to mount thymus-dependent cell-mediated immune responses to filarial antigen, has allowed us to observe the phenomenon of lymphatic dilation in the presence of living parasites. That worm killing and obstructive lymphatic lesions can be induced by immune reconstitution, suggests that the characterization of the seemingly complex immunologic as well as non-immunologic mechanisms of lymphatic damage will be possible.
Acknowledgements We wish to thank the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (ID No. 870382) for their support of this study.
55
References Cohen, L.B., Nelson, G., Wood, A.M., Manson-Bahr, P.E.C. and Bowen, R. (1961) Lymph-angiography in filarial lymphoedema and elephantiasis. Am. J. Trop. Med. Hyg. 10, 843-848. Denham, D.A. and Nelson, G.S. (1976) Nematodes parasitic in the lymphatic system. In: Soulsby, E.J.L. (ed.), Pathophysiology of Parasitic Infection, Academic Press. Haque, A., Capron, A., Ouaissi, A., Kouemeni, L., Lejeune, J.P., Bonnel, B. and Pierce, R. (1983) Immune unresponsiveness and its possible relation to filarial disease. Contr. Microbiol. Immunol. 7, 9-21. Kwa, B.H. and Mak, J.W. (1987) Filarids (excluding D. immitis). In: Soulsby, E.J.L. (ed.), Immune Responses in Parasitic Infection: Immunology, Immunopathology and Immunoprophylaxis. Vol. I, Chapt. 10, CRC Press. Nelson, G. S. (1966) The pathology of filarial infections. Helminthol. Abstr. 35, 31 I. Nelson, G. S. (1981) Issues in filariasis - a century of enquiry and a century of failure. Acta Trop. 38, 197. Ottesen, E.A. (1980). Immunopathology of lymphatic filariasis in man. Springer Seminars. Immunopathology 2, 373-385. Partono, F. (I 982) Elephantiasis and its relation to filarial immunity. Southeast Asian J. Trop. Med. Publ. Health 13, 275. Vickery, A.C. and Vincent, A. L. (1984) Immunity to Brugia pahangi in athymic nude and normal mice: eosinophilia, antibody and hypersensitivity responses. Parasite Immunol. 6, 545-559. Vickery, A.C., Vincent, A.L. and Sodeman, W.A. (1983) Effect of immune reconstitution and resistance to Brugia pahangi in congenitally athymic nude mice. J. Parasitol. 69, 478-485. Vickery, A.C., Nayar, J.K., and Albertine, K.H. (1985) Differential pathogenicity of Brugia malayi, B. patei and B. pahangi in immunodeficient mice. Acta Trop. 42, 353-363. Vincent, A.L., Vickery, A.C., Lotz, M.J. and Desai, U. (1984) The lymphatic pathology of Brugia pahangi in nude (athymic) and thymic mice, C3H/HeN. J. Parasitol. 70, 48-56. Vincent, A.L., Vickery, A.C., Winters, A. and Sodeman, W.A. (1982) Life cycle of Brugia pahangi (Nematoda) in nude mice, C3H/HeN (nu/nu). J. Parasitol. 68, 553-561. Von Lichtenberg, F. (1957) The early phase of endemic bancroftian filariasis in the male. Pathological study. J. Mr. Sinai Hosp. 24, 983-1000. Warren, K.S. (1971) Worms. In: Samter, M. (ed.), Immunological Diseases, Little, Brown. Yates, J.A. and Higashi, G.I. (1986) Ultrastructural observations on the fate of Brugia rnalayi in jirds previously vaccinated with irradiated infective stage larvae. Am. J. Trop. Med. Hyg. 35, 982-987.
