Altered intestinal macrophage phenotype in ovine paratuberculosis

Researchin VeterinaryScience1997,63, 139-143

l~lv~

Altered intestinal macrophage phenotype in ovine paratuberculosis H. M. ALZUHERRI, D. LITTLE, C. J. CLARKE, Department of Veterinary Pathology, University of Edinburgh, Easter

Bush, Roslin, Midlothian EH25 9RG

SUMMARY The expression of macrophage surface markers that are likely to be important in antigen presentation and cell interactions was examined in normal sheep and those with clinical paratuberculosis. Immunohistological studies demonstrated that intestinal macrophages in diseased sheep expressed MHC class II, LFA-1 and CR4 antigens weakly compared with normal tissues. Reverse transcriptase-polymerasechain reaction analysis of MHC class II mRNAin intestinal whole tissue samples showed no significant difference between control and diseased groups. A reduction in molecules such as MHC class II and LFA-1 on the surface of infected macrophages could have implications for survival of the intracellular mycobacteria and the persistence of infection.

PARATUBERCULOSIS (Johne's disease) is a chronic disease of ruminants caused by Mycobacterium avium subspecies paratuberculosis (Map). In susceptible animals, ingested mycobacteria appear to evade the host defences and persist and proliferate in intestinal macrophages. Recruitment of monocytes and other cells to the gut leads to further infection of macrophages and the development of a severe, extensive, granulomatous enteritis. The mechanisms of mycobacterial persistence and pathogenesis are incompletely understood but in other mycobacterioses, infection has profound influences on the events that should eliminate organisms from the tissues, such as antigen presentation, killing mechanisms of the macrophage, cytokine production and effector functions of lymphocytes. We have already reported the elevated expression of proinflammatory cytokines in the intestine of diseased sheep, demonstrating the link between infection and pathology (Alzuherri et al 1996). The current study examines the phenotype of intestinal macrophages in naturally-infected paratuberculosis cases. The macrophage is the target cell for infection and therefore has a pivotal role in the uptake of organisms and the presentation of mycobacterial antigens to the immune system. The identification of infectionrelated changes in the macrophage surface molecules involved in these mechanisms will assist our comprehension of the pathogenesis of paratuberculosis. In an effective immune response, surface major histocompatibility complex (MUG) class II molecules play a critical role in the presentation of mycobacterial antigens to T cells that are then able to upregulate and amplify a specific immune response (Ehlers et al 1994). MI-ICclass I molecules are also important in some mycobacterioses for presenting antigens and provoking T cell-mediated cytotoxic responses (Flynn et al 1992). Infection-related diminution of antigen presentation by maerophages to lymphocytes may therefore dislocate a potentially effective host immune response and allow the survival of intracellular mycobacteria. Poor presentation of antigen could result in a lack of T cell effector mechanisms such as interferon (IFN)-y mediated intracellular killing, and have been recognised in other mycobacterioses (Desai et al 1989, Tweardy et al 1984). Other molecules that may be of importance for cell-cell interactions include adhesion 0034-5288/97/050139 + 05 $18.00/0

molecules such as the leukocyte function-associated molecules (LFA)-I and LFA-3. Receptors for complement components and lipopolysaccharide may also play a role in the uptake of mycobacteria by macrophages (Schlesinger and Horwitz 1990).

MATERIALS A N D M E T H O D S

Sheep All sheep were adult (two- to five-year-old) Scottish Blackface ewes. Diseased sheep (n=14) were selected from flocks with an established paratuberculosis problem, and had clinical signs of progressive weight loss and serological evidence of Map infection (positive results from agar gel immunodiffusion (ACXD) test and absorbed enzyme-linked immunoabsorbent assay (ELISA) (Sherman et al 1984, Milner et al 1990). Control sheep (n=7) were from flocks without paratuberculosis, clinically normal and gave negative results with serological testing.

Pathology and sampling Sheep were killed by intravenous barbiturate administration and subject to a full necropsy. Selected tissues from the terminal ileum, mesenteric lymph nodes and major organ systems were fixed in buffered formalin and processed routinely to histological sections. These were stained with haematoxylin and eosin and by the ZiehlNeelsen technique. Fresh samples of ileum were also mounted in OCT, frozen on a bed of dry ice with isopentane and stored at -70°C until used for immunohistochemical staining.

Immunohistochemistry and monoclonal antibodies Immunostaining methodology was based on the technique of Andrew and Jasani (1997), using the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA). Monoclonal antibodies against ovine major histocompatibility complex (MHC) class lI DQo~ (VPM 36; Dutia et al © 1997 W. B. Saunders CompanyLtd

140

H. M. Alzuherri, D. Little, C. J. Clarke

1990) and DRc~ antigens (VPM 54; Dutia et al 1994), CD14 (VPM 61, 63 and 65; Gupta et al 1996), CR4 (ON1, Pepin et al 1992), leukocyte function-associated moleculel(LFA-1)(F10-150-39), LFA-3 (L180/1, Hunig et al 1987) and an uncharacterised macrophage marker (VPM 32; Gonzalez 1989) were used. Cryostat sections (61am) of ileum were prepared, air-dried for two hours, fixed in cold acetone for five minutes and washed in phosphate buffered saline (PBS). Endogenous peroxidase was blocked with 100 units m1-1 glucose oxidase at 37°C and sections washed in PBS. Non-specific binding was blocked with normal horse serum (0.2 per cent in PBS) for 15 minutes at room temperature. Primary monoclonal antibody (as neat hybridoma saturated supernatant fluids) or control normal mouse serum (2 per cent in PBS) were added for overnight incubation at 4°C. Sections were washed and incubated with biotinylated horse anti-mouse IgG antibody (1/200 dilution)(Vector, USA) for 30 minutes at room temperature. After washing, streptavidin compound (avidin-biotin-horseradish peroxidase complex) was added for 30 minutes at room temperature and the sections were washed. Sections were incubated for 10 minutes with 3-amino-9-ethylcarbazole (AEC) substrate, washed in tap water and counterstained with haematoxylin. Stained sections were examined microscopically at high power (x40 objective) and macrophages over the entire section were scored for staining intensity on a scale from '-' for no staining detected to '++' for strong positive staining. RNA

extraction and polymerase chain reaction

Total cellular mRNA was isolated from frozen ileum and mesenteric lymph node tissues as described (Alzuherri et al 1996). RNA was reverse-transcribed into complementary DNA (CDNA) using murine leukaemia virus reverse transcriptase (Gibco BRL) and stored at -70°C until use. Oligonucleotide primer pairs for ovine MHC class II DQ and DR and Map IS900 were used to amplify 181bp, 222bp and 1200bp products respectively. For the PeR assay, 3lal cDNA was added to a 47~1 reaction premix consisting of 100raM Tris-HC1,500raM KCI (xl0 Taq buffer), 50raM MgC12, 20mM of a mixture of dATP, dGTP, dCTP, dTTP and W1 detergent (Gibco BRL). This was overlaid with mineral oil and amplification proceeded in a thermal cycler with incubation at 95°C for five minutes, then 80°C with addition of 0.3 llal (1.5 units) Taq polymerase. Samples were submitted to-29 cycles of denaturation at 95°C for one minute, annealing at 50°C and primer extension at 72°C for two minutes, with a final cycle extension for five minutes. Appropriate positive and negative controls were run in each assay. Aliquots of 71al PCR product were loaded onto 2 per cent agarose gels and subjected to electrophoretic separation. Gels were stained with ethidium bromide and examined under u v light for bands. Gels were Southern blotted onto Gene Screen plus Nylon membrane (NEN)by alkaline transfer. Membranes were then u v cross-linked and hybridised at 65°C with specific radiolabelled internal primers for both MHC class II DQ and DR. Membranes were autoradiographed with Kodak X-OMATfilm and bands were scored on a scale from 0 (undetectable) ~- 5 (strong). Samples for ATPase were run in parallel and levels were found to be consistent between samples. Statistics

The RT-PCR data was not normally distributed so non-

FIG 1: Ileum of adult sheep with paratuberculosis. A lepromatous lesion in which the lamina propria is infiltrated by large macrophages ('epithelioid cells'). H & E, x 1800

parametric tests (Kruskall-Wallis and Mann-Whitney tests) were used. Significance was considered to be P<0.05.

RESULTS Pathology and diagnostic tests

Diseased animals fell into two distinct categories based on serological tests and pathology (Clarke and Little 1996). Seven had clear positive ELISA and AGID test results and necropsy showed these sheep to be in poor condition with typical gross lesions of thickened intestinal mucosa affecting particularly the ileum, and enlarged regional lymph nodes. Histological examination of tissues showed lesions of severe chronic granulomatous enteritis with villous atrophy. The mucosa of affected gut was markedly thickened and infiltrated by sheets of macrophages with abundant pale eosinophilic cytoplasm ('epithelioid' cells) (Fig 1). Ziehl-Neelsen staining revealed large numbers of acid-fast bacteria within the cytoplasm of macrophages in the gut and lymph nodes. These animals were classified as having lesions of the lepromatous type. The remaining seven diseased sheep had weaker ELISA results and negative AGID test results. Gross lesions were similar to the lepromatous group but histological examination of the gut showed a granulomatous enteritis with multifocal small granulomata composed of aggregates of epithelioid cells amongst a

FIG 2: Ileum of adult sheep with paratuberculosis. A tuberculoid lesion in which a small, focal granuloma is surrounded by lymphocytes, granulocytes and plasma cells. H & E, × 1800

Macrophage phenotype in paratuberculosis

141

TABLE 1: Immunohistological scoring for macrophage staining in control and diseased ileum Monoclonal antibody specificity MHC IIDQ MHC II DR VPM 32 CR4 LFA-1 LFA-3 VPM 61 VPM 63 VPM 65

Controls (n=7)

Diseased lepromatous (n=7)

Diseased tuberculoid (n=7)

++ ++ ++ ++ ++

+/- (++ at villous tips) +/- (++ at villous tips) +/- (++ at villous tips) +/- (+/- in lymphatics) +/- (+/- in lymphatics)

+/- (++ at villous tips) +/- (++ at villous tips) +/- (++ at villous tips) +/- (+/- in lymphatics) + (++ in lymphatics)

+/+/-

+/+/-

+/+/-

Results represent a scale from no staining (-) up to very strong staining (++) TABLE 2: Scoring of autoradiograph band intensities from RT-PCR for MHC class II DQ and DR in ileum

MHC II DQ MHC II DR

Controls (n=6)

Lepromatous (n=6)

Tuberculoid (n=6)

3 (2-5) 2 (0-3)

5 (3-5) 3 (1-5)

4.5 (3-5) 4 (1-5)

Results are given as the median and, in brackets, the range for each group. Individual bands scored as 0 (undetectable) - 5 (strong band)

FIG 4: Ileum of control adult sheep. Normal tissue stained by an immunohistological method with antibodies against LFA-I. Proprial macrophages show strong positivity (arrow). x 1800

marked lymphocyte infiltrate (Fig 2). Ziehl-Neelsen staining in these cases showed only very few intracellular acidfast organisms. These lesions were classified as tuberculoid in type. Samples of ileum from all the lepromatous cases and most of the tuberculoid cases produced an IS900 PCN product of the correct size, indicating the presence of Map. Control sheep were in good condition, gave negative results with the serological and IS900 PeR tests and showed no significant gross or histological lesions.

normal sheep, the surface expression of MHC class II (DQ and DR) was strong on macrophages in the villous lamina propria, particularly cells at the villous tips (Fig 3). Class II MHC antigens were also strongly expressed by the macrophages scattered throughout the deeper lamina propria and the submucosa. Expression of LFA-1 (CDllaJCD18), CR4 and the VPM 32-reactive molecule on macrophages was also strong in these same areas (Fig 4). Class II MHC, CR4, CD14 positive, and VPM 32-reactive, cells were identified in organised gut-associated lymphoid tissue (6ALT) follicles. Reactivity of proprial macrophages with VPM 63 and VPM 65 was very weak. In contrast to the above, in diseased ileum, MHC class II DQ and DR antigens were expressed only weakly or undetectably on the many large, 'epithelioid' -type macrophages in the mucosa and submucosa (Fig 5). However, a few cells at the villous tips expressed these antigens moderately. Occasional, isolated, positively-staining cells were also scattered through the lamina propria but the morphology of these suggested that they were neither infected nor of the epithelioid cell type. Staining with antibodies against LFA-1 and CR4, and the VPM 32 antibody, gave a pattern of weak or negative

Immunohistochemistry Immunostaining of sections gave consistent results for each monoclonal antibody for animals in each group (Table 1). The macrophage staining patterns in the lepromatous and tubercnloid groups were very similar. In sections from

FIG 3: Ileum of control adult sheep. Normal tissue stained by an immunehistological method with antibodies against MHC class II antigen. Proprial macrophages show strong positivity (arrow). x 1800

FIG 5: Ileum of adult sheep with paratuberculosis. A lepromatous lesion stained by an immunohistological method with antibodies against MHC class II DR antigen. Infiltrating macrophages show only very weak staining (arrow). Other positively-staining lymphocytes are visible, x 1800

142

H. M. Alzuherri, D. Little, C. J. Clarke

FIG 6: Autoradiograph of MHC Class II and DQ RT-PCR products from samples of ileum. C, control group (n=6); L, lepromatous group (n=6); T, tuberculoid group (n=6); DQ and DR, positive controls for each product; N, negative control. Arrows show level of probed product

expression of antigens on epithelioid cells in the lamina propria and submucosa in lepromatous lesions. A few cells near the villous tips stained moderately strongly with VPM 32. Expression of LFA-1 was slightly stronger on macrophages in the tuberculoid lesions. In areas of granulomatous lymphangitis in diseased gut, intraluminal macrophages expressed surface LFA-1 strongly and reacted with VPM 32 weakly. VPM 32-reactive cells (uninfected) were also observed in GALT follicles. Expression of CD14 (VPM 61 and VPM 63) was very weak to undetectable on macrophages. Multinucleate giant cells found occasionally in diseased ileum had a similar staining pattern to epithelioid cells. In control and diseased tissues VPM 61 and antiLFA-3 reagents failed to stain any cells. Class II MHC and LFA-1 expression was also noted on most intestinal lymphocytes in both control and diseased tissues. Reverse-transcriptase polymerase chain reaction for class 11 antigens

MHC

Analysis of autoradiographs probed for MHC class II DQ and DR sequences (Fig 6) showed no significant difference in expression between control and diseased groups (Table 2).

DISCUSSION The immunohistological findings indicate that the infected, epithelioid-type macrophages of the diseased intestine had a different phenotypic profile to macrophages in the normal intestine. Although results were not quantifiable, a weaker staining pattern for MHC class II surface expression was clearly and consistently noted in diseased animals and suggests a reduced potential for presentation of mycobacterial antigens. Intestinal levels of MHC class II mRNA were

measured to assess if these changes in the macrophages were reflected in whole tissue samples. Levels were similar for all groups but these reflect a composite of all the cell types present, not only macrophages. Infiltrating lymphocytes in the diseased gut have been shown to express MHC class II at higher levels than normal (Little et al, unpublished data) and this may have influenced the results. In addition, the presence of mRNA does not necessarily correlate with expression of the protein product. , Studies have shown that M tuberculosis and M leprae are chpable of inhibiting MHC class II expression on cells of the macrophage series (Desai et al 1989, Gercken et al 1994, Wadee et al 1995). The mechanism of reduced MHC class II expression is unknown but it is interesting to note that mycobacteria have been isolated from monocytes in cattle with clinical paratnberculosis (Koenig et al 1993). Defective antigen presentation related to reduced MHC class II expression has been demonstrated in M tuberculosis infection of monocytes (Gercken et al 1994). Expression of the integrin LFA-1 was reduced on epithelioid cells in infected gut as determined by immunohistology. Such adhesion molecules play an essential part in the cell-cell and cell-endothelial interactions that are important in immune responses. LFA-1 was highly expressed on some macrophages within, intestinal lymphatics and MLN. However, cells here usually contained few, if any, organisms and an ability to migrate into such areas suggests a higher state of activation than the mucosal epithelioid cells. A higher LFA-1 expression was the only notable difference between macrophages in tuberculoid compared to lepromatous lesions and could be related to the lower mycobacterial load in tuberculoid cases. Putative receptors for mycobacteria such as CR4 and the CD14 antigens were poorly expressed on epithelioid cells but CR4 was strongly expressed on normal macrophages. A reduction of CD14 has previously been reported following mycobacterial infection of monocytes (Tsuyuguchi et al 1990) and epithelioid cells are also known to have low surface complement receptor expression (Papadimitriou and van Bruggen 1986). However, even if these were receptors for mycobacterial entry, a reduction of expression on an already-infected macrophage seems unlikely to disturb the pathogenesis of the disease, with newly-recruited monocyte-macrophages being more probable targets. It was of interest to note that the phenotype of epithelioid cells in lepromatous and tuberculoid lesions was similar (with the exception of LFA-1), despite the obvious difference in the degree of mycobacterial infection. This may imply a less di.stinct difference in intracellular killing between the two groups than was previously supposed. In both types of pathology the mycobacteria are persistent over long periods and there is no evidence of aggressive 'tubercle' formation as seen in tuberculosis, supporting the idea that macrophages from both types of lesion are inadequate defence cells. The mechanisms of depression of surface antigens are poorly characterised but mycobacterial wall components such as lipoarabinomannan (LAM) Can block surface expression of MHC class II and make macrophages refractory to activation by WN-7 (Sibley et al 1990). These findings indicate a reduced expression of surface molecules that are potentially important in antigen presentation and cell adhesion, on macrophage-type cells in the intestine of sheep with clinical paratuberculosis. This observation underpins the thesis that the persistent infection and resultant inflammation of the gut may be related to

Macrophage phenotype in paratuberculosis

poor antigen presentation by infected cells and a reduced efficacy of the host immune response.

ACKNOWLEDGEMENTS This work was funded by the Biotechnology and Biological Sciences Research Council.

REFERENCES ALZUHERRI, H. M., WOODALL, C. J. & CLARKE, C. J. (1996) Increased intestinal TNFo~, IL-1B and IL-6 expression in ovine paratnhercnhsis. Veterinary Immunology and lmmunopathology 49, 331-345 ANDREW, E. M. & JASANI, B. (1987) An improved method for the inhibition of endogenous peroxidase deleterious to lymphocyte surface markers. Application to immunoperoxidase studies on eosinophil rich tissue preparations. Histochemistry Journal 19, 425-430 CLARKE, C. J. & LITTLE, D. (1996) The pathology of "ovine paratuberculosis: gross and histological changes in the intestine and other tissues. Journal of Comparative Pathology 114, 419-437 DESAI, S. D., BIRDI, T. J. & ANTIA, N. H. (1989) Correlation between macrophage activation and bactericidal function and Mycobacerium leprae antis gen presentation in rnacrophages of leprosy patients and normal individuals, Infection and Immunity 57, 1311-1317 DUTIA, B. M., HOPKINS, J., ALLINGTON, M. P., BUJDOSO, R. & McCONNELL, I. (1990) Characterization of monoclonal antibodies specific for ~- and B-chains of sheep MHC class n. Immunology 70, 27-32 DUTIA, B. M., McCONNELL, I., BALLINGALL, K. T., KEATING, P. & HOPK/NS, J. (1994) Evidence for the expression of two distinct IvlI-/Cclass n DR like molecules in the sheep. Animal Genetics 25, 235-241 EHLERS, S., MIELKE, M. E. A. & HAHN, H. (1994) Progress in TB research; Robert Koch's dilemma revisited. Immunology Today 15, 1-4 FLYNN, J. L., GOLDSTE1N, M. M., TRIEBOLD, K. J., KOLLER, B. & BLOOM, B. R. (1992) Major histocompatibility complex class I restricted T cells are required for resistance to Mycobacterium tuberculosis infection. Proceedings of the National Academy of Sciences, USA89, 12013-12017 GERCKEN, J,. PRYMA, J., ERNST, M. & FLAD, H. D. (1994) Defective antigen presentation by Mycobacterium tuberculosis-infected mouocytes. Infection and Immunity 62, 3472-3478 GONZALEZ, C. N. (1989) Characterisation of sheep macrophage differentiation antigens by monoclonal antibodies. MPhil thesis, University of Edinburgh

143

GUPTA, V. K., McCONNELL, I., DALZIEL, R. G. & HOPKINS, J. (1996) Identification of the sheep homologue of the monocyte cell surface marker CD14. Veterinary Immunology and lmmunopathology 51, 89-99 HUNIG, T., TJEFENTHALER, G., MEYER ZUM BUSCHENFELDE, K. M. & MEUER, S. C. (1987) Alternative pathway activation of T cells by binding of CD2 to its cell surface ligand. Nature 326, 298-301 KOENIG, G. J, HOFFSIS, G. F., SHULAW, W. P., BECH-NIELSEN, S., RINGS, M. & ST-JEAN, G. (1993) Isolation of Mycobacterium paratuberculosis from mononuclear cells in tissues, blood, and mammary glands of cows with advanced paratnherculosis. American Journal of Veterinary Research 54, 14411445 MILNER, A. R., MACK, W. N., COATES, K. J., HILL, J., GILL, I. & SHELDRICK, P. (1990) The sensitivity and specificity of a modified ELISA for the diagnosis of Johne's disease from a field trial in cattle. Veterinary Microbiology 25, 193-198 PAPADIMITRIOUS J. M. & VAN BRUGGEN, I. (1986) Evidence that multinucleate giant cells are examples of mononuclear phagocyte differentiation. Journal of Pathology 148, 149-157 PEPIN, M., CANNELLA, D., FONTAINE, J-J., PITET, J-C. & LE PAPE, A. (1992) Ovine mononuclear phagocytes in sial: identification by monoclonal antibodies and involvement in experimental pyogranulomas. Journal of Leukocyte Biology 51, i88-198 SCHLESINGER, L. S. & HORWITZ, L. A. (1990) Phagocytosis of leprosy bacilfi is mediated by complement receptors CR1 and CR3 on human monocytes and complement component CR3 in serum. Journal of Clinical Investigation 85, 13041314 SHERMAN, D. M., MARKHAM, R. J. F. & BATES, F. (1984) Agar gel immunodiffusion test for diagnosis of clinical paratubercnlosis in cattle. Journal of the American Veterinary Medical Association 185, 179-182 SIBLEY, L. D., ADAMS, L. B. & KRAHENBUHL, J. L. (1990) Inhibition of interferon-gamma-mediated activation in mouse macrophages treated with lipoarabinomannan. Clinical and Experimental Immunology 80, 141-148 TSUYUGUCHI, I., KAWASUMI, H., TAKASHIMA, T., TSUYUGUCHI, T. & KISHIMOTO, S. (1990) Mycobacterium avium-Mycobacterium intracellulare complex-induced suppression of T-cell proliferation in vilxo by regulation of monocyte accessory cell activity. Infection andlmmunity 58, 1369-1378 TWEARDY, D. J, SCHACTER, B. Z. & ELLNER, J. J. (1984) Association of altered dynamics of monoctye surface expression of human leukocyte antigen DR with immunosuppression in tuberculosis. The Journal of Infectious Diseases 149, 31-37 WADEE, A. A., KUSCHKE, R. H. & DOOMS, T. G. (1995) The inhibitory effects of Mycobacterium tuberculosis on MHC class II expression by monocytes activated with riminophenazines and phagocyte stimulants. Clinical and Experimental Immunology 100, 434-439

Received November 12, 1996 Accepted March 26, 1997