Effects of papain on the agglutination of canine red cells with serum autoantibodies

Research in Veterinary Science 1993, 55, 156-161

Effects of papain on the agglutination of canine red cells with serum autoantibodies R. N. BARKER, Department of Pathology and Microbiology, Medical School, University of Bristol, Bristol BS8 1TD, D. R. E. JONES, Department of Immunology, Queen's Medical Centre, Nottingham

The papain test, which detects incomplete antired blood cell (RBC) autoantibodies in serum, was positive in 12 out of 16 anaemic dogs. Positive results were significantly correlated (Z2=II.1, P<0.001) with increased levels of RBCbound immunoglobulin, but in three of these cases it was considered that a diagnosis of autoimmune haemolytic anaemia could not be justified. Furthermore, enzymellinked antiglobulin test measurements of RBC-reactive serum IgG were increased in only three of the dogs with a positive papain test. Papainised canine RBC did not consistently take up more serum IgG than untreated cells in an indirect enzyme-linked antiglobulin test. However, the zeta potential of the RBC was reduced after enzyme treatment, and electrophoretic analysis revealed that glycophorins, which bear a strong negative charge, were cleaved from the cell membrane. It is concluded that a positive papain test alone is not reliable in the diagnosis of canine autoimmune haemolytic anaemia and that the enzyme increases the agglutinability of RRC by reducing the mutually repulsive electrostatic forces between the cells, rather than by increasing the amount of autoantibody bound. THE unequivocal diagnosis of autoimmune haemolytic anaemia (AIHA) in the dog requires the demonstration of red blood cell (RBC) specific autoantibodies (Halliwell 1978). RBC-bound autoantibodies have traditionally been detected using the direct antiglobulin test (DAT), or Coombs' test (Miller et al 1954, Slappendel 1979, Jones et al 1990). However, this technique has been estimated to record false negative results in 14 to 42 per cent of dogs with AIHA (Slappendel 1979, Switzer and Jain 1981, Jackson and Kruth 1985). Although this shortcoming of the DAT has 156

been overcome recently with the development of a more sensitive direct enzyme-linked antiglobulin test (DELAT) (Jones et al 1987, 1992, Barker et al 1992, 1993), fresh RBC are required for these techniques, and such samples might not be readily available in all cases. Therefore, assays that can detect unbound RBC autoantibody in serum remain potentially useful tools in the diagnosis of AIHA. The indirect antiglobulin test (IAT) is the standard technique for the detection of serum RBC autoantibodies in human medicine (Petz and Garratty 1980). The method is based on the ability of antiglobulin reagent to agglutinate normal donor RBC which have been incubated with autoantibody-containing serum from the patient. Unfortunately, the IAT is almost always negative in cases of AIHA in the dog (Quimby et al 1980, Jones 1986, Day 1989), stimulating the development of more sensitive agglutination-based tests for RBC autoantibodies in canine sera (Jones and Darke 1975, Feldman 1982, Jones 1986). For example, treatment of RBC with the proteolytic enzyme papain has been found to render the cells agglutinable by serum autoantibody, without the need for an antiglobulin reagent (Jones and Darke 1975). This papain test is apparently sensitive, being able to detect cases of DAT-negative AIHA, and has the advantage that it requires no speciesspecific antisera (Jones and Darke 1975). A development of the papain test, using low ionic strength suspending solution to augment the effects of the enzyme, has been described for use in the dog (Feldman 1982), although similar modified tests have been found to have no additional advantages in man (Holburn and Prior 1987). To date, the reliability of the papain test has not been evaluated fully in canine anaemia, and the mechanism by which the enzyme enhances agglutination is also unclear. It has been suggested that papain modifies

Effects of papain on canine red cells the RBC membrane to expose further antigenic sites, thereby allowing more autoantibody to bind (Jones and Darke 1975, Feldman 1982). However, the effects of proteolytic enzymes on the human RBC membrane are complex (Mollison et al 1987), and include a reduction of the zeta potential of RBC, the electrostatic force that causes the cells to repel one another (Stratton et al 1973, Luner et al 1975). It is possible, therefore, that papain increases the agglutinability of canine RBC by reducing the mutually repulsive forces between them. The first aim of the current work was to ascertain the reliability of the papain test (Jones and Darke 1975) in the diagnosis of AIHA in the dog. The second aim was to test the hypotheses that the enzyme acts either by increasing the amount of autoantibody bound to RBC, or by decreasing the zeta potential of the cells.

157

hydrochloride (Sigma) to a final concentration of 0-0125M, the papain solution was incubated for one hour at 37°C. The resulting activated papain was stored in small samples at -20°C until required for use. Ten volumes of washed, packed RBC were mixed with one volume of activated papain plus nine volumes of isotonic phosphate buffered saline pH 7.2 (PBS) and incubated for 15 minutes at 37°C. The cells were washed a further three times and made up as a 5 per cent suspension in pBS. Sera were tested for RBC-reacrive autoantibody by incubation for one hour at 37°C with equal volumes of the papain treated RBC suspension in glass tubes. A positive result was recorded if there was evidence of macroscopic agglutination using both autologous RBC and RBC samples from four normal dogs. Known negative and positive sera were included as controls in each test.

Materials and methods

Animals

Direct enzyme-linked antiglobulin test (DELAT)

Dogs with anaemia (haemoglobin under 12.0 g d1-1, Hinton and Jones 1977) were presented to the Department of Veterinary Medicine at the University of Bristol, or to neighbouring veterinary practices. AIHA was diagnosed in anaemic dogs on the basis of a number of factors, including previous history, clinical signs, haematology, a positive DELAT (Jones et al 1987) and the demonstration of RBC-specific antibody in RBC eluates (Barker et al 1991, 1993). Blood samples were collected by cephalic venepuncture into EDTA anticoagulant for haematology, citrate or EDTA for RBC immunoassays and plain glass tubes for serum separation (Becton Dickinson). Greyhound blood donors maintained at the University of Bristol, and healthy dogs admitted for routine screening or surgery, were used as a source of normal blood.

RBC-bound immunoglobulins were measured using the DELAT as described previously (Barker et al 1992).

Testing specificity of RBC-bound immunoglobulin The specificity of RBC-bound immunoglobulin for the erythrocyte membrane was tested as previously described (Barker et al 1991, 1993). Briefly, immunoglobulin was eluted from RBC by ether treatment (Rubin 1963) and the eluates were assayed for anti-RBC membrane antibody in an ELISA.

Detection of RBC-reactive IgG in sera by EL1SA Sera were assayed for erythrocyte membranereactive IgG by an ELISA, essentially as described by Barker et al (1991, 1993). Before testing, sera were diluted 1/50 in PBS containing 1 per cent bovine serum albumin and 0.5 per cent Tween 20.

Papain test RBC-reactive autoantibody in serum was detected by the papain test (Jones and Darke 1975). L6w's papain (Dacie and Lewis 1984) was prepared from a 1 per cent solution of freshly ground papain (BDH) in S6rensen's 0.067 M phosphate buffer pH 5-4. After centrifugation to remove particulate matter and the addition of cysteine

lndirect enzyme-linked antiglobulin test (IELAT) Serum IgG reactive with normal, intact RBC was measured with the reLAY (Barker et al 1993).

RBC membrane preparations RBC ghosts (Dodge et al 1963) were prepared by

158

R. N. Barker, D. R. E. Jones

hypotonic lysis (Barker 1991). After collection, RBC ghosts were frozen immediately and stored at -70°C until use. Protein concentrations of the preparations were determined by the method of Bradford (1976).

Results

Reliability of the papain test in diagnosis of canine AIHA

The papain test was performed on samples from seven AIHA cases and from eight dogs with other anaemias. These findings were compared with the results of the DELATfor RBC-bound immunoglobulins, and with ELISA measurements of RBC membrane-reactive IgG (Table 1). There was a significant association between positive results in the papain test and the DELAY (Z2= 11.1, P<0.001). It should be noted that the papain test was positive in all cases of AIHA, but also in five dogs with other anaemias. In three of these five cases a diagnosis of secondary AIHA could not be justified, since anti-erythrocyte membrane antibody could not be detected in ~BC eluates, indicating that immunoglobulin detected by the DELAT was non-specifically reactive with the cells. The four dogs with a negative papain test included only one DELATpositive animal, which was also not considered to have AtriA. Only three of the 12 sera yielding a positive papain test had increased RBC-reactive IgG in the ELISA.

Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE)

SDS-PAGE was performed in 12 per cent poly~ acrylamide gels according to the method of Laemmli (1970), using the Mini Protean II (BioRad) as previously described (Barker 1991). Polyacrylamide gels were stained for protein or glycoprotein with Coomassie brilliant blue or periodic acid-Schiff stains, respectively, using the methods of Barker (1991).

Measurement of RBC zeta potential The zeta potential of canine RBC was determined by measuring their electrophoretic mobility with the Malvem Zetasizer II (Malvern Instruments). Human RBC were used to standardise the assay. As a control, canine P~C were also tested after removal of negatively charged sialic acid residues from the cell surface by neuraminidase treatment. A 10 per cent suspension of RBC was incubated with 1 mg m1-1 neuraminidase (Sigma) for 30 minutes at 37°C and washed three times in

Effects of papain on the canine RBC membrane It has been proposed that papain treatment of RBC may increase their susceptibility to agglutination with autoantibodies by increasing the number

PBS.

TABLE 1 : Results of the papain test in anaemic dogs Case

Result of

number

papain test

89/51 90/3 89/8 89/9 90/21 90/22 90/30 89/38 89/50 90/32 90/37 90/39 90/13 89/13 90/34 90/41 Normal (n=6)

RBO-bound

IgG/IgM/IgA

RBC membrane-

RBC membrane-

reactive IgG in

reactive IgG in

(DELAT)*

RBC eluate (ELISA)*

serum (ELISA)*

+

+

+

+

+

+

+

+

+ +

+ +

NT NT

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

* OO > normal mean + 3SD NT Not tested

NT

+ + NT NT NT

+

Diagnosis

Primary AIHA Primary AIHA Primary A[HA Primary AIHA Primary AIHA Primary AIHA Primary AIHA MyelodysplasiaJALHA Heartworm Myocarditis Pyometra Thrombocytopenic purpura Intestinal neoplasia Thrombocytopenic purpura Thrombocytopenic purpura Splenic neoplasia

159

Effects of papain on canine red cells TABLE 2: Effect of papain treatment of RBC on the binding of RBc-reactive serum IgG

TABLE 3: Effect of papain treatment on the zeta potential of RBC

Case number

RBC source

88/13 89/7 Normal Mean (n=4)

Papain test result

+ +

RBc-reactive serum 10G (IELAT OD) Untreated RBC

Papainised RBC

% Change

0-129 0-125 0.112

0.153 0"115 0-117

+18-6%* -8%* +4.5%

Electrophoretic mobility of RBC after enzyme treatment (ps-1 V-1 cm) Untreated

Normal dog 1 Normal dog 2 Human

Papain

1-2 1.2 1.07

1.04 1.05 NT

Neuraminidase 0.67 0.68 NT

NT Not tested

* Greater than intra-assay coefficient of variation (6%)

of antigenic sites exposed on the cell surface (Jones and Darke 1975, Feldman 1982). To test this hypothesis, the amount of IgG taken up by RBC f r o m canine sera was measured by the IELAT, using papainised or untreated normal RBC as the source of antigen (Table 2). When two anaemic dogs with a positive papain test were investigated, enzyme treatment of RBC slightly increased the IELAT measurement of RBe-reactive IgG in one dog, but reduced the reading in the second. Papainisation of RBC had little effect on the mLAT results when normal sera were tested. To determine the effects of papain on the protein and glycoprotein components of the RBC BAND

Mr

"1&2

220"~

*3 "4.1 "4.2

117 ""~" 83 75

*5 6 7 8 9

44 38"~ 35 32 "+ 25

A

B

membrane, erythrocyte ghosts made from papainised or untreated cells were analysed by SDS-PAGE (Fig 1). When gels were stained with Coomassie blue, the apparent molecular weight and relative staining intensity of many of the major protein bands appeared unaffected by papain treatment (compare lanes A and B). However, band 3 was replaced by a less intensely stained doublet and two additional proteins could be visualised between bands 4.2 and 5. In contrast to these limited effects on the appearance of the protein bands, PAS development of gels revealed that the staining intensity of the six major canine RBC glycophorins was markedly reduced by papainisation (compare lanes C and D). PAS BAND

Mr

*3

117 "~

A B

62 59 ' ~

DC

45"'="

E F

30.-*, 28"*

43"*

C

D

FIG 1 : The effect of papain treatment on canine RBC membrane components. RBC ghosts were prepared from papainised and untreated RBC and analysed by SDS-PAGE: lanes A and B, Coomassie blue protein stain (20 pg protein/track); lane A, untreated RI3C; lane B, papainised RBC; fanes C and D, PAS glycoprotein stain (30 gg protein/track); lane C, untreated RBC; lane D, papainised RBC. Bands marked * have been labelled according to the nomenclature for human RBC (Fairbanks et al 1971, Steck 1974), others have no obvious equivalent in man and have been designated as proposed by Barker (1991). The apparent molecurar mass of the major bands is indicated in kD. Lanes A + B and C + D are from different gels

160

R. N. Barker, D. R. E. Jones

stained gels also demonstrated the replacement of band 3 by a fainter doublet after papain treatment. An alternative hypothesis to account for the increased agglutinability of papainised RBC is that the enzyme reduces the zeta potential of the cells. The electrophoretic mobility of canine RBC was therefore measured before and after papain treatment (Table 3). It can be seen that papainisation clearly reduced the zeta potential of RBC, but not as effectively as neuraminidase treatment. It should also be noted that untreated canine RBC had a higher zeta potential than the human RBC which were included to standardise the assay.

papainisation of RBC, even when the treated cells were agglutinated. However, it is possible that some epitopes were destroyed by the enzyme as other cryptic antigens were exposed, since papain treatment has been reported to have both effects on the human RBC membrane (Bell et al 1973, Mollison et al 1987). The exact nature of such changes to the erythrocyte surface may be important. SDS-PAGEanalysis revealed that papainisation of canine RBC resulted in a dramatic loss of the glycophorins from the membrane and the partial breakdown of band 3, components which may bear antigens bound by pathogenic autoantibodies in some cases of canine AIHA (Barker et al 1991). In man cryptic epitopes exposed on papain treated Discussion RBC are recognised by naturally occurring Using the papain test, canine AIHA is diagnosed haemolysins found in some normal sera (Bell et al by the demonstration of incomplete serum autoan- 1973). One hypothesis to explain anomalous tibodies which can agglutinate enzyme-treated results of the papain test in the diagnosis of canine ~BC (Jones and Darke 1975). In the current work, AIHA is that proteolysis may destroy autoantigens a positive papain test was shown to be significant- important in haemolysis, while exposing determily correlated with elevated gBC-bound immuno- nants reactive with normal serum antibodies. globulins measured by the DELAT in a series of Papain treatment of canine RBC was shown to anaemic dogs. However, the group of DELAT- reduce their zeta potential. This finding is consispositive dogs included a high proportion of cases tent with the observation that there was also protewhere a diagnosis of AIHA was not justified, since olytic breakdown of the glycophorins, which are the immunoglobulin was bound non-specifically heavily sialated and therefore highly charged to RBC (Barker et al 1993). Therefore, although the (Barker 1991). The zeta potential of normal papain test may be useful as a simple screen for RBC provides a mutually repulsive force between AIHA, positive results must be interpreted with cells which can prevent cross-linking by anticaution in conjunction with clinical findings and bodies (Van Oss and Absolom 1983). Thus, it the results of further immunohaematological tests. seems likely that in the papain test incomplete Additional evidence that the papain test yields a antibodies are able to cause agglutination primarihigh proportion of false positive results is pro- ly because the electrostatic forces between RBC vided by the finding that most samples showing have been reduced. In man most proteolytic agglutination in this test did not have increased enzymes also reduce the zeta potential of ~BC by ELISA readings for RBC membrane-reactive serum splitting glycophorins off the membrane (Mollison IgG. However, it could also be argued that this et al 1987). However, this effect alone is not result suggests that the papain test is more sensi- thought sufficient to explain the ability of papain tive than the ELISA in detecting serum autoanti- to potentiate the agglutination of human RBC with alloantibodies (Stratton et al 1973, Luner et al bodies. The implications of papain test results in AreA 1975). diagnosis would be clearer if the effects of the The zeta potential of RBC was found to be enzyme on canine erythrocytes were understood. greater in the dog than in man. It may be speculatIt has previously been suggested that papain treat- ed that the relatively high zeta potential of canine ment of RBC enables these cells to be agglutinated RBC is one factor which contributes to the insensiwith incomplete antibodies by exposing further tivity of the IAT (Quimby et al 1980, Jones 1986, antigenic sites on the membrane, thus allowing Day 1989) and DAT (Switzer and Jain 1981, more antibody to bind (Jones and Darke 1975, Jackson and Kruth 1985, Jones et al 1990) in this Feldman 1982). The results of the current work do species, since a higher surface charge may not support this hypothesis, since increased bind- increase the resistance of the cells to agglutination. In conclusion, the papain test detects incoming of serum IgG did not necessarily follow

Effects of papain on canine red cells

plete RBC autoantibodies because the enzyme cleaves highly charged glycophorins from the cells, thus lowering their zeta potential and increasing their agglutinability. However, positive results do not provide unequivocal evidence of AIHA in the dog. Acknowledgements The authors wish to thank the staff in the department of veterinary medicine for their cooperation in the collection of blood samples and Miss M. Graham for performing the papain tests. We are also grateful to Dr A. W. Preece for his use of the Malvern Zetasizer. The work was supported by the Wellcome Trust. References BARKER, R. N. (1991) Electrophoretic analysis of erythrocyte membrane proteins and glyeoproteins from different species. Comparative Haematology International 1, 155-160 BARKER, R. N., GRUFFYDD-JONES, T. J. & ELSON, C. J. (1993) Red cell-bound immunoglobulins and complement measured by an enzyme-linked antiglobulin test in dogs with autoimmune heamolysis or other anaemias. Research in Veterinary Science 54, 170-178 BARKER, R. N., GRUFFYDD-JONES, T. J., STOKES, C. R. & ELSON, C. J. (1991) Identification of autoantigens in canine autoimmune haemolytic anaemia. Clinical and Experimental Immunology 85, 33-40 BARKER, R. N., GRUFFYDD-JONES, T. J., STOKES, C. R. & ELSON, C. J. (I992) Autoimmune haemolysis in the dog: Relationship between anaemia and the levels of red blood cell bound immunoglobulins and complement measured by an enzyme-linked antiglobulin test. Veterinary Immunology and lmmunopathology 34, 1-20 BELL, C. A., ZWICKER, H. & NEVIUS, D. B. (1973) Nonspecific warm hemolysins of papain-treated cells: serologic characterization and transfusion risk. Transfusion 13, 207-213 BRADFORD, M. M. (1976) A rapid and sensitive method for the quantitation of microN'am quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248-254 DACIE, J. V. & LEWIS, S. M. (1984) Red-cell blood groups and identification of antigens. In Practical Haematology, 6th edn, Edinburgh, Churchill Livingstone. pp 343-357 DAY, M. J. (1989) Immune-mediated blood dyscrasias. In University of Sydney Postgraduate Committee in Veterinary Science, Proceedings number 118, pp 79-104 DODGE, J. T., MITCHELL, C. & HANAHAN, D. J. (1963) The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Archives of Biochemistry and Biophysics 100, 119-130 FAIRBANKS, G., STECK, T. L. & WALLACH, D. F. H. (1971) Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry ! 0, 2606-2617 FELDMAN, B. F. (1982) Use of low ionic strength solution in combination with papain treated red blood cells for the detection of canine erythrocyte autoantibodies. Journal of the American Animal Hospital Association 18, 653-658 HALLIWELL, R. E. W. (1978) Autoimmune disease in the dog. Advances in Veterinary Science and Comparative Medicine 22, 221263

161

HINTON, M. H. & JONES, D. R. E. (1977). Anaemia in tile dog: An analysis of laboratory data. Journal of Small Animal Practice 18, 701 706 HOLBURN, A. M. & PRIOR, D. (1987) The UK National External Quality Assessment Scheme in blood group serology. Compatibility testing 1983-1984: the influence of variables in test procedures on detection of incomplete antibodies. Clinical Laboratory Haematology 9, 33-48 JACKSON, M. L. & KRUTH, S. A. (1985) Immune-mediated hemolytic anemia and thrombocytopenia in the dog: a retrospective study of 55 cases diagnosed from 1969 through 1983 at the Western College of Veterinary medicine. Canadian Veterinary Journal 26, 245-250 JONES, D. R. E. (1986) Use of an enzyme indirect antiglobulin test for the diagnosis of autdimmune haemolytic anaemia in the dog. Research in Veterinary Science 41, 187-190 JONES, D. R. E. & DARKE, P. G. G. (1975) Use of papain for the detection of incomplete erythrocyte autoantibodies in autoimmune haemolytic anaemia of the dog and cat. Journal of Small Animal Practice 16, 273-279 JONES, D. R. E., GRUFFYDD-JONES, T. J., STOKES, C. R. & BOURNE, F. J. (1990) Investigation into factors influencing performance of the canine antiglobulin test. Research in Veterinary Science 48, 53-58 JONES, D. R. E., STOKES, C. R., GRUFFYDD-JONES, T. J. & BOURNE, F. J. (1987) An enzyme-linked antiglobulin test for the detection of erythrocyte-bound autoantibodies in canine autoimmune haemolytic anaemia. Veterinary Immunology and lmmunopathology 16, 11-21 JONES, D. R. E., GRUFFYDD-JONES, T. J., STOKES, C. R. & BOURNE, F. J. (1992) The use of a direct enzyme-linked antiglobulin test in the laboratory diagnosis of immune-mediated hemolytic mlemia in dogs. American Journal of VeterinaryResearch 53,457-465 LAEMMLI, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 LUNER, S. J., STURGEON, P., SZKLAREK, D. & McQUISTON, D. T. (1975) Effect of proteases and neuraminidase on RBC surface charge and agglutination. A kinetic study. Vox Sanguinis 28, 184-199 MILLER, G., SWISHER, S. N. & YOUNG, L. E. (1954) A case of autoimmune hemolytic disease in a dog. Clinical Research Proceedings 2, 60-61 MOLLISON, P. L., ENGELFRIET, C. P. & CONTRERAS, M. (1987) Red cell antibodies against self antigens, bound antigens and induced antigens. In Blood Transfusion in Clinical Medicine 8th edn, Oxford, Blackwell Scientific Publications. pp 410-452 PETZ, L. D. & GARRATTY, G. (1980) The serologic investigation of autoimmune hemolytic anemia. In The Acquired Hemolytic Anemias, New York, Churchill Livingstone. pp 139-184 QUIMBY, F. W., SMITH, C. S., BRUSHWEIN, M. & LEWIS, R. W. (1980) Efficacy of immunoserodiagnostic procedures in the recognition of canine immunologic diseases. American Journal of Veterinary Research 41, 1662-1666 RUBIN, H. (1963) Antibody elution from red blood cells Journal of Clinical Pathology 16, 70-73 SLAPPENDEL, R. J. (1979) The diagnostic significance of the direct antiglobulin test (DAT) in anemic dogs. Veterinary Immunology and lmmunopathology 1, 49-59 STECK, T. L. (1974) The organization of proteins in the human red blood ceil membrane. Journal of CellBiology 62, 1-19 STRATTON, F., RAWLINSON, V. I., GUNSON, H. H. & PHILLIPS, P. K. (1973) The role of zeta potential in Rh agglutination. Vox ganguinis 24, 273-279 SWITZER, J. W. & JAIN, N. C. (1981) Autdimmune hemolytic anemia in dogs and cats. Veterinary Clinics of North America: Small Animal Practice 11, 405-420 VAN oSS, C. J. & ABSOLOM, D. R. (1983) Hemagglutination and the closest distance approach of normal, neuraminidase and papain-treated erythrocytes. Vox Sanguinis 47, 250-256

Received September 7, J992 Accepted February 5, 1993