Immunoglobulin isotype distribution of antinuclear antibodies in dogs with leishmaniasis

Research

in Veterinary

Science

1998, 65,205-207

Immunoglobulin

isotype distribution of antinuclear in dogs with leishmaniasis

anti

ROSARIO LUCENA, PEDRO J. GINEL*, Departamento de Patologia Clinica Veterinaria, Hospital Clinico Veterinario, Campus de Rabanales, Ctra. Madrid-CLidiz km 396, 14014 Grdoba, Spain SUMMARY A modified indirect immunofluorescence method, using rat liver as substrate, was developed to determine the immunoglobulin isotypes forming antinuclear antibodies in sera from 12 antinuclear antibody-positive dogs out of 121 dogs with natural Leishmania infection. Immunoglobulin M was found to be the most frequent component of antinuclear antibodies (91.7 per cent), followed by IgG (41.7 per cent) and IgA (33.2 per cent). When these immunoglobulin isotypes were titrated, Ighi antinuclear antibodies showed higher titres (1:200) than IgM and IgA antinuclear antibodies (150 and 1:20 respectively). Most of the antinuclear antibody-positive dogs simultaneously had two immunoglobulin isotypes, whereas none had all three immunoglobulin isotypes at the same time. Spearman rank correlation analysis showed no significant correlation between antinuclear antibody titres and circulating immune complexes or immunoglobulin levels. The low incidence of antinuclear antibodies and the absence of a clear relationship between isotype titres and clinical signs suggest a minor pathogenic role of antinuclear antibodies in canine leishmaniasis.

CANINE leishmaniasis is a serious systemic disease caused by protozoa of the genus Leishmania. This parasite has a dixenic life cycle and its mammalian host is located within organs of the mononuclear-phagocytic system, thus affecting mainly the liver, spleen, bone marrow and lymph nodes (Galvao Castro et al 1984, Brandonisio et al 1990). The infection has been reported to impair the host’s immune system leading to a polyclonal B cell response and an associated hypergammaglobulinemia, characterised by immune complex and antinuclear antibody formation (Slappendel 1988, Brandonisio et al 1990). The term antinuclear antibodies is used in a generic way to describe autoantibodies against different nuclear components such as DNA, RNA, nuclear proteins and their molecular complexes (Tan 1983). Positive antinuclear antibody titres are observed in numerous inflammatory and infectious diseases and form part of the immune complexes found in canine leishmaniasis. Thus, antinuclear antibodies may contribute to vascular immune complex deposition and subsequent classical type III hypersensitivity reactions usually manifested clinically by glomerulonephritis, uveitis, dermatitis or arthritis (Argov et al 1989). The incidence of antinuclear antibody-positive titres in Leishmania-infected dogs has been recently established (Lucena et al 1996) but there is no data on the immunoglobulin isotypes forming these antinuclear antibodies. The distribution of immunoglobulin isotypes and subisotypes in response to nuclear antigens has been investigated in human and canine autoimmune diseases, mainly systemic lupus erythematosus, a disease which mimics some clinical manifestations of canine leishmaniasis. A significant correlation between isotype of antinuclear antibody and clinical activity was reported (Gonzalez and Rothfield 1966). These findings and the well-known different pathogenic capability of each immunoglobulin isotype * Corresponding author Tel:+3457218713 Fax: +34 57 211 093 00345288/98/060205 + 03 $18.00/O

prompted us to determine the different isotypes of immunoglobulins forming antinuclear antibodies in naturally acquired canine leishmaniasis.

MATERIALS

AND METH

Animals The presence of antinuclear antibodies was investigated by immunofluorescence as previously described (Lucena et al 1996) in 121 dogs with natural leishmania infection. Serum samples from 12 dogs were found to have elevated antinuclear antibody titres (cut-off titre 120) and were included in the study. Diagnosis of leishmaniasis was made by direct observation of the parasite in Giemsa stained popliteal lymph node aspirates. Clinical signs most commonly associated with immune complex deposition and positive antinuclear antibody titres such as glomerulonephritis, skin lesions, uveitis and arthritis were assessed. Physical and occuIar examinations, haemogram, serum biochemistry and urinalysis were performed in all dogs. Additional diagnostic tests such as radiology, synovial fluid tapping and skin biopsy were performed when indicated during the differential diagnosis work-out.

Serum samples A blood sample was obtained from the cephalic vein of each animal. After being allowed to clot at room temperature, the blood was centrifuged (3000 x g, 15 minutes, room temperature) and the resulting serum samples were stored at -20°C in 0.5 ml aliquots until use. All samples were obtained after the definitive diagnosis was made and before initiating any specific treatment. Serum immunoglobulin isotypes and circulating immune complexes in these dogs were determined as previously described (Ginel et al 1996, Lopez et al 1996). 0 1998 W. B. Saunders company Ltd

R. Lucena, P. J. Ginel

206

~rn‘rn~~o~~o~~~in isotype of antinuclear antibodies Immunoglobulin isotypes forming antinuclear antibodies were determined using a modification of the indirect immunofluorescence method described by Lucena et al (1996). Briefly, 7 pm cryosections of a Wistar white rat liver were prepared at -20°C and fixed by temperature difference on 10 well teflon-coated microscope slides (10 wells of 7 mm diameter, Nutacon, The Netherlands). Slides were stored at -70°C until use. Using this method, a dilution of 1:20 was considered a significatitly positive titre as determined by the negativity of 30 normal sera. Thus, each antinuclear antibody-positive. sample was serially dilutkd in phosphate-buffered saline (0.01 M PBS, pH 7.4) and 25 ~1 of each dilution were added in triplicate to the corresponding well. Twenty-five microlitres of PBS (as negative control) and 25 1.11of a positive control serum were included on each slide. After incubation for 30 minutes at room temperature, the slides were washed by immersion in PBS for five minutes, dried and washed again for a further 10 minutes. Slides were finally dried using absorbent paper without touching the liver sections. To detect each immunoglobulin isotype, rabbit antiserum specific for canine IgG, IgM and IgA heavy chains (ICN Biomedicals, Costa Mesa, USA) were diluted 1:50 in PBS and 25 ~1 were added separately to each sample. After incubation for 30 minutes at room temperature, slides were washed and dried again. Bound antibodies were detected by adding 25 pl of fluorescein isothiocyanate goat antiserum specific for rabbit IgG (whole molecule) (Sigma Immunochemicals, St Louis, USA) diluted 1:50 in PBS to each well. Slides were incubated for 30 minutes at room temperature in a dark moist chamber. Slides were then washed again, dried and fixed with an aqueous medium (Immu-mount, Shandon, Pittsburg, USA). The fluorescence was detected using an epifluorescence microscope with a mercury lamp (Axioscop, Zeiss, Germany). Statistic& analysis The possible relationship between serum antinuclear antibody isotypes and circulating immune complexes or immunoglobulin levels was evaluated by a Spearman rank correlation test. RESULTS Four of the 1’2 dogs studied (33.2 per cent) had only one antinuclear antibody isotype (IgM in three cases and IgG in one case). The other eight dogs (66.7 per cent) had two antinuclear antibody isotypes, four of these dogs showed IgG and IgM antinuclear antibodies and the remaining four dogs had IgM and IgA antibodies. None of the dogs studied showed antinuclear antibodies of all three immunoglobulin isotypes at the same time. Immunoglobulin M antinuclear antibodies were present in 11 dogs (91.7 per cent), whereas five dogs (41.7 per cent) showed IgG antibodies and four dogs (33.2 per cent) showed IgA antinuclear antibodies (the sum of these percentages is higher than 100 per cent as most dogs showed more than one antinuclear antibody isotype). Most dogs with IgG antinuclear antibodies showed positive at 1:20 dilution; in only two cases IgG titres were >l:lOO. In approximately half the dogs showing IgM antinuclear antibodies, the highest positive titre was 1:20 whereas in the remaining dogs the titre only reached 1:50. In dogs showing IgA antinuclear antibodies the highest positive

TABLE 1: Serum circulating immune ccrn~~ex~e (CIC), expressed tpci a percentage of a reference standard pool, anti ~rnrnu~5~i~~~i~ cpncantrations (mg do-1) in those dogs showing pasittve a~ti~u~~~ar aflt~~~~~ (ANA) titres lmmunoglobulin

cc N

ANA ~--

W

IgM

MA

FG

k!M

igA

I@

KIM

w

1

163.7

201.5

178.6

10,000

159

22.5

200

20

-

2 3 4 5 6 7 8 9

299.1 208.2 186.8 254.5 302.6 117.4 147.7 201.1

83.3 403.7 241.5 199.5 203.5 352.0 291.1 489.2

91.3 196.6 178.6 224.6 160.5 205.3 188.6 146.0

5000 10,200 6700 8200 10,400 12,800 11,600 12,800

250 115 178 149 149 250 190 320

58 18 19 15 17 60 10 22.5

20 20 -

50 20 50 20 20 50 20 20

20 20 20 -

10 . 11 12

298.2 119.4 210.4

201.3 423.0 252.4

170.0 206.0 196.6

8200 9500 5900

181 194 260

22.5 10 13

100 20

50 50 -

20 -

titre found was 1:20 (Table 1). No correlation could be found between antinuclear antibody isotypes and serum circulating immune complexes or immunoglobulin levels. Only two of the 12 antinuclear antibody-positive dogs studied had developed glomerulonephritis and both cases had IgG and IgM antinuclear antibodies. However, two other dogs had the same isotypes of antinuclear antibodies and did not show signs of glomerulonephritis. Immunoglobulin M antinuclear antibodies were always associated with renal, occular, cutaneous or joint involvement whereas IgG and IgA were only variably present; one dog with IgG antinuclear antibodies showed only non-specific signs (Table 2).

DISCUSSION The fluorescent antinuclear antibody test is the most widely used and accepted technique for the detection of most antinuclear antibodies and can be easily adapted to determine their immunoglobulin isotypes. Reported positive titres in the fluorescent antinuclear antibody test for canine sera range from 1: 10 to 1: 100, but with rodent tissue substrates dilutions of 1:20-1:30 or lower are commonly used (McVey and Shuman 1991, Hansson et al 1996). In this study a titre of 21:20 was therefore considered significant. Leishmania infection consistently produces several abnormalities on the ho&s immune system, particularlv a polyclonal B cell activation (Slappendel 1988, Brando&i0 et al 1990). The resulting hypergammaglobulinemia is not protective and includes the production of circulating immune complexes, antinuclear antibodies and other autoantibodies. Clinical signs associated with immune complex deposition such as uveitis, arthritis and cutaneous vasculitis are frequently recognised and well characterised in canine leishmaniasis (Slappendel 1988, Nieto et al TABLE 2: Relationship between antinuclear some clinical signs in canine ieishmaniasis. of dogs in each category (most dogs showed

antibody ~sot~pea an Figures represent number more than one lesion) Clinical

Ig lsotypes IgM W IgG + IgM IgA + IgM Total

Number of dogs 3 1 4 4 12

Dermatitis 2 0 3 4 9

Glomerulonephritis 0 0 2 0 2

signs

tlveiiis 3 0 1 2 6

Arthritis 1 0 0 I 2

Antinuclear

antibodies

1992). In contrast, the role of antinuclear antibodies and other autoantibodies in canine leishmaniasis remains controversial. Ferrer (1992) reported that antinuclear antibodies were elevated in more than 30 per cent of infected dogs, however in a recent study a lower percentage of 15.9 per cent (Lucena et al 1996) was obtained. In addition, only 12 out of 121 dogs (10 per cent) included in this study showed significant titres. Moreover, titres in these dogs tended to be low. Thus, although in two dogs IgG antinuclear antibody titres reached 1: 100 and 1:200 dilutions, the most frequently found IgM antinuclear antibodies showed a higher titre of I:50 whereas IgA factors were only present at the 1:20 dilution. No studies on the titration of antinuclear antibodies in canine leishmaniasis were found but these titres are lower or similar to those described in other canine autoimmune diseases (McVey and Shuman 1991, Hansson et al 1996). Isotype determination of antinuclear antibodies in association with circulating immune complexes and serum immunoglobulin levels was attempted in order to further investigate their pathogenic relevance in canine leishmaniasis. Antinuclear antibodies can be formed by each of the three main immunoglobulin isotypes and can cause tissue lesions by binding directly to tissue antigens or most commonly by forming circulating immune complexes (Svec and Veit 1967, Tan 1983, Kass et al 1985). However, each immunoglobulin isotype has a different pathogenic role with IgG and IgM having the highest potential tissue damage (Kass et al 1985, Tizard 1992). In our study, IgM associated with IgG or IgA was by far the most frequently detected type of antinuclear antibody. In most human (Gonzalez and Rothfield 1966, Tan 1983, Massa et al 1995) and canine (Lapras and Monier 197 1, Kass et al 1985) systemic autoimmune processes antinuclear antibodies of the IgG isotypes predominate, a smaller percentage belong to the IgM isotype, whereas IgA antinuclear antibodies are seldom encountered. Avila and Rojas (1990) most frequently found IgM (visceral leishmaniasis) and IgM associated with IgG (localised cutaneous leishmaniasis) autoantibodies, but these authors did not focus specifically on antinuclear antibodies. In contrast, Kager et al (1984) considered IgG as the major antinuclear antibody in human leishmaniasis. In this study, the high prevalence found of IgM antinuclear antibodies suggests they have a more relevant pathogenic role. However, no clear relationship could be made between a single clinical manifestation and immunoglobulin isotype of antinuclear antibodies (Table 2) a fact that may be partly due to the low number of infected dogs having significant tines. Similarly, Nieto et al (1992) could not establish any correlation between the isotype of immunoglobulin involved in circulating immune complex formation and glomerular lesions in dogs with leishmaniasis. Serum IgG levels were remarkably high in the antinuclear antibody positive dogs whereas IgM and IgA were within the normal range (Ginel et al 1996). Similarly, circulating immune complex levels of the three immunoglobulins studied but specially of the IgM isotype were variably increased in most dogs when expressed as a percentage of pooled reference sera. These findings are consistent with the polyclonal hypergammaglobulinemia classically reported in canine leishmaniasis (Ferrer 1992). Although some coefficients of correlation were relatively high, no significant correlation could be found between antinuclear anti-

207

in canine leishmaniasis

body and serum immunoglobulin or circulating inmune complex concentrations. It is perhaps more likely that the association observed represents merely downstream effects of the same B cell altered mechanisms. In conclusion, low titres of antinuclear antibodies were found in a relatively small percentage of dogs with leishmaniasis. Although IgM was the predominant antinuclear antibody isotype, most of the dogs had a combination of IgM and IgG or IgA antinuclear antibodies. The Iow incidence of antinuclear antibodies and the absence of a clear relationship between isotype titres and clinical signs suggests a minor pathogenic role of antinuclear antibodies in canine leishmaniasis.

FERENCES ARGOV. S., JAFFE, C. L., KRUPP, M., SLOR, H. & SHOENFELD, 1. (1989) Autoantibody production by patients infected with leishmania. Clinical Experimental Imm~mology 76, 190-197 AVILA, J. L. & ROJAS, M. (1990) Elevated cerebroside antibody !evels in human visceral and cutaneous leishmaniasis, trypanosoma rangeli infection and chronic chagas’ disease. American JoumaE of Tropical Medicine and H~~giene 43,52-60 BRANDONISIO, 0.; CARELLI, G., ALTAMURA, M., VARVARA; B. & CECI, L. (1990) Circulating immune complexes and autoantibodies in canine Ieishmaniasis. Parassitolo~ia 32.275-281 FERRER, L. (y992) Leishmaniasis. In Current Veterinary Therapy XI. Eds R. W. Kirk and J. E. Bonagura. Philadelphia, W. B. Saunders. pp 266-278 GALVAO-CASTRO, B., SA FERREIRA, J. A., MARZOCHI. K. F.. MARZOCHI, M. C., COUTINHO, S. G. & LAMBERT. P. H. (1984) Polyclonai B cell activation. circulatinrz immune conmlexes and autoimmunitv in human am&an viscer-al leishmaniasi;. Clinical E,&iment~~I ~~rzmunology $58-66 GlNEL, P. J., MARGARITO, J. M., MOLLEDA. J. M.. LOPEZ, R., NOVALES, M. & BERNADINA, W. E. (1996) Biotin-avidin amplified enzyme-linked immunosorbent assay (ELISA) for the measurement of canine serum IgA. IgG and IgM. Research in Veterinary Science 60,107-l 10 GONZALEZ, E. N. 6r ROTHFlELD, N. F. (1966) ImmunogIobulin class and pattern of nuclear fluorescence in systemic lupus erythematosus. New &$and /our& 0s Medicine 214,1333-1338 HANSSON. H., TROWALD-WIGH, G. & KARLSSON-PARRA, A. (1996) Detection of antinuclear antibodies by indirect immunofluorescence in dog sera: comparison of rat liver tissue and human epithelial-2 cells as antigenic substrate. Journal of Veterinary Internal Medicine 10, 199-203 KAGER, P. A., VAN DER PLAS-VAN DALEN, C., REES, I’. H.. HELMERHORST, F. M. & VON DEM BORNE, AEGKr. (1984) Red cell, white cell and platelet autoantibodies in visceral leishmaniasis. Tr+cal Geographzc Medicine 36,143.150 KASS, P. H., STROMBECK. D. R.; FARVER, T. B. & ARDANS, A. A. (1985) Application of the log-linear model in the prediction of the antinuciear antibody test in the dog. American Jourml of Veterinary Research 46,2336-2339 LAPRAS, M. & MONIER; J. C. (1971) Technique d’irnmunofluorescence appliquee au diagnostic des maladies auto-immunes chez Ie chren. Revlie de Midicine V&&inaire 122, 973.989 LOPEZ, R., LUCENA, R., NOVALES, M., GINEL, P. J., MARTIN, E. & MOLLEDA. J. M. 11996) Circulatine immune complexes and renal fnnction in canine leishmaniasis. Jo&al of V&inary Medic& L143,469-474 LUCENA, R., GlNEL, P. J., LOPEZ, R., NOVALES. M., MARTIN, E. & MOLLEDA, J. M. (1996) Antinuclear antibodies in dogs with leishmaniasis. Journal of Veterinary Medicine A 43,255.259 MASSA, M., BENEDETTI, F., PIGNATTI, P., ALBANI, S., MONESTIER, .M. & MARTINI, A. (1995) Lack of temporal association of iridocuclitis with IgG reactivities to core histones and nucleosome subparticles in pauciarticuiar juvenile chronic arthritis. Bn’risi? .lonmuZ ofRheumatology 34,507-511 McVEY, D. S. & SHUMAN, W. (1991) Use of multiple antigen substrates to detect Veterinaly Imnzuiznlogy and antinuclear armbody in canine sera. Immulzopathology 28, 37-43 NIETO, C. G., NAVARRETE, I., HABELA, M. A.. SERRANO. F. & REDONDO. E. (1992) Pathological changes in kidneys of dogs with natural Leishm