Andy Pachikerl, Ph.D
Dog erythrocyte antigens are responsible for initiating approximately 70% to 80% of immune-mediated transfusion reactions in the dog. As with other species, the red blood cell antigens found in the dog have various immunogenicities. In health, these antigens are participants in cell recognition-self versus nonself. In disease, they may serve as antigens for antibody or markers in disease occurrence. Little is known about their biochemical properties. Currently their description is reliant on polyclonal antibody serology. This reliance has limited the progress of transfusion practice in the dog.
Historically, the study of canine blood groups and their importance in transfusion began in the 1600s through a physician, Richard Lower. He is credited with the first canine-to-canine transfusion. The efforts of doctors Lower and Denis in heterologous transfusion using lamb, dog, and human subjects introduced the basic transfusion premise “like transfuses like.” In 1910, Von Dungem and Hirszfeld documented the presence of four hemolysins and agglutinins based on canine alloimmunization (Swisher & Young, 1961). Further work by Ottenburg, Kaliski, and Friedman in 1913 confirmed these findings. From 1937 to 1949, Wright, Whipple, and Eyquem further defined the presence of six canine blood groups (Colling & Saison, 1980).
However, not until 1961 was the importance of these antigens in transfusion and disease investigated by Swisher, Young, and Trabold (Swisher, et al., 1962). To date, the work submitted by Swisher and Young remains the most current published information of the importance of canine blood groups in transfusion (Swisher, 1954; Swisher, et al., 1962; Young, et al., 1951). Additional blood groups have been identified by Rubenstein (1968), Suzuki/5 Colling and Saison (Colling & Saison, 1980; Colling & Saison, 1980), and Symons and BelLl9 (Symons & Bell, 1992). Of this latter group, only the antigen first noted by Rubenstein (Colling & Saison, 1980) was evaluated in regard to transfusion significance by Bull (Bull, 1976).
The importance of canine blood groups in veterinary transfusion medicine is based on three factors: the incidence of the antigen in the dog population, the incidence of naturally occurring antibody within the population, and the effect of the antibody against the antigen in vivo. Current blood typing schemes identify six erythrocyte antigens with possible importance.
The dog erythrocyte system
Blood groups are defined by glycolipids and glycoproteins on the surface of the red blood cell membrane. Current blood typing schemes identify six dog erythrocyte antigens (DEAs): 1.1, 1.2, 3, 4, 5, and 7 (Table 1). Blood groups are independently inherited. Simple Mendelian laws of dominance govern their inheritance. These antigens are defined by using polyclonal antibodies generated through canine alloimmunization. Polyclonal antibody recognition may be dependent on multiple recognition sites to define the “antigens” currently accepted. Biochemically, little is known about the DEA system.
Table 1. Dog erythrocyte antigens established as international standards: classification, occurrence, and significance
This blood group system has been defined with multiple alleles. They include the antigens 1.1, 1.2, 1.3, and a null type. An individual dog may show only one of the four phenotypes. Family studies suggest a Mendelian type of autosomal dominance. Table 2 describes the current phenotypic and genotypic information on this blood group system.
1.1- and 1.2-positive dogs have been studied for transfusion significance. Naturally occurring antibody to these alleles has not yet been found. Therefore, first-time transfusion reactions do not occur. However, if a negative dog is exposed to 1.1- positive erythrocytes, a strong hemolysin can result. On second exposure, an immune-mediated hemolytic transfusion reaction results causing removal of transfused cells in less than 12 hours. Hemoglobinuria and hyperbilirubinemia frequently occur. In addition to uncross matched, untyped transfusion, pregnancy can cause production of antibody against DEA 1.1 25% of the time. For these reasons 1.1-positive dogs are excluded as transfusion donors.
1.2-positive dogs can cause a problem as both the transfusion donor and recipient. A previously sensitized negative type dog undergoes permanent red blood cell removal and loss 12 to 24 hours after the administration of 1.2-positive red blood cells. Thus, 1.2-positive dogs are poor erythrocyte donors. If a 1.2-positive dog is sensitized with DEA 1.1 red blood cells, it will produce a potent anti-DEA 1.1 antibody. Administration of DEA 1.1 red blood cells to a sensitized 1.2 dog results in an immediate hemolytic transfusion reaction. Therefore, 1.2-positive dogs are at risk after sensitization for immediate transfusion. 1.3-positive dogs have not been evaluated for transfusion significance. Future study is limited because of the unavailability of typing sera for DEA 1.3.
This red blood cell antigen is the most controversial among the six antigens discussed. Published reports of naturally occurring antibody to this antigen suggest that this antibody has a natural prevalence as high as 50% in DEA 7-negative dogs. Recent reports by Giger fail to support the presence of naturally occurring anti-DEA 7. Observations by the author suggest that naturally occurring antibody does exist in 20% to 50% of all DEA 7-negative dogs. However, the naturally occurring antibody is quite weak, rarely producing a titter greater than 1:8. In the presence of naturally occurring antibody, as in the cat, immunemediated transfusion reaction can occur during a first transfusion.
Sensitized DEA 7-negative dogs, when transfused with DEA 7-positive erythrocytes show a delayed transfusion reaction. Hemolysis does not occur; however, an irreversible sequestration and loss of red blood cells occurs in 72 hours. This type of delayed transfusion reaction is only significant if the regenerative ability of the transfusion recipient is compromised. Because of the presence of naturally occurring antibody in the DEA 7-negative population and because of the delayed loss of erythrocytes in sensitized dogs, DEA 7-positive dogs are not recommended as donors.
This antigen has not been considered significant because of its low incidence in the dog population of the United States. However, recent evaluation of DEA type by breed suggests that it may be more important. Only 6% of the general population has DEA 3-positive cells. Yet 23% of the Greyhounds typed from 1990 to 1995 in our laboratory had DEA 3-positive cells (unpublished data). Increasing numbers of our transfusion donors in the United States are Greyhounds. Naturally occurring antibody is found in 20% of the DEA 3-negative dogs in the United States. Swisher reported permanent red blood cell sequestration and loss within 5 days of administration in sensitized dogs transfused with DEA 3-positive cells. Once again, this type of delayed transfusion reaction is only significant if the regenerative ability of the transfusion recipient is compromised. This information suggests that use of DEA 3-positive donors should be avoided for the same reasons as use of DEA 7-positive dogs.
DEA 4 is a high-incidence antigen in the dog population. Upwards of 98% of dogs have this antigen. Dogs typed as DEA-4 positive that are negative to all other blood group systems are considered “universal donors.” Antibody does not occur naturally to DEA 4. Sensitized DEA 4-negative dogs do not show red blood cell loss or hemolysis when transfused with DEA 4-positive cells. The antibody produced through exposure is considered benign. Yet less than five subjects were used to generate the information available on transfusion with DEA4-sensitized dogs; additional study is warranted.
The incidence of DEA 4-negative dogs may be higher in certain controlled breeding situations. Studies on a related group of Dobermans on the East Coast of the United States documented a 25% incidence (unpublished data). The likelihood of generating harmful antibody in a DEA 4-negative dog when exposed to another blood group system is close to 100%. Transfusion donors for these DEA 4-negative dogs should be “universal” to avoid the production of harmful antibodies to DEA 1.1, 3, 5, or 7. A DEA 4-negative dog with no past red blood cell sensitization is an excellent donor. Dogs either positive or negative for DEA 4 are safe transfusion donors.
As with DEA 3-positive dogs, the significance of this antigen/antibody reaction was tempered by the low incidence of the antigen in the general population of dogs in the United States. Recent evaluation of DEA type by breed, in the author’s laboratory, suggests that the incidence of DEA 5-positive Greyhounds is 30%, 7% higher than noted in the general population (unpublished data). Naturally occurring antibody is found in 10% of the DEA 5-negative dogs in the United States. Swisher reported permanent red blood cell sequestration and loss within 3 days in sensitized dogs transfused with DEA-5 cells, thus demonstrating a delayed transfusion reaction similar to that occurring in DEA 7-sensitized dogs. DEA 5-positive dogs are not recommended as transfusion donors for the same reasons that dogs positive for DEA3 or 7 are not used.
Table 2. Genotypic and phenotypic description for the dog erythrocyte antigen-1 blood group system.
International standardization included two additional antigens, DEA 6 and DEA 8. Information regarding DEA 6 suggested a high incidence antigen. Only one DEA 6-negative dog was studied, and it did not have natural antibody. After sensitization, this dog showed moderately rapid removal of DEA 6-positive cells. DEA 8 was present in 40% of the population tested in 1962. This blood group was not evaluated for transfusion significance. Eleven other antigens of the dog have been described. In addition, Colling & Saison, 1980 suggested the presence of another blood group system involving multiple alleles. Unfortunately, these antigens were not evaluated for transfusion significance. Antisera does not exist for the antigens described. Therefore, comparative studies are impossible.
DEA typing and donor selection
Based on the available information, donor selection must be influenced by the antigens listed in Table 1. The goal of each red blood cell transfusion is to provide a maximum number of viable erythrocytes while minimizing the number of erythrocytes lost to storage injury and immune-mediated destruction. Many of our transfusions in veterinary medicine are performed between untyped and uncross matched animals. Sensitization and later exposure to any of the other DEA antigens, apart from DEA 4, results in the premature loss of red blood cells by the recipient. Ideally, avoidance of all nonself antigens should be practiced. Recipients of whole blood and red blood cell products should be typed and crossmatched. If typing is unavailable, recipients should be crossmatched with “universal” donors before transfusion to avoid sensitization. A “universal” donor is defined as a dog expressing only the DEA-4 antigen. Many authors suggest that avoiding exposure to DEA 1.1 is adequate. This belief is since DEA 1.1 antibody is the only known hemolysin (Swisher, 1954; Swisher, et al., 1962; Swisher & Young, 1961). Therefore, avoidance of a DEA 1.1 antibody-antigen interaction avoids acute haemolytic transfusion reaction. If the goal of red blood cell transfusion is to provide the maximum number of red blood cells to the recipient with minimal cell loss through immune-mediated pathways, then exposure to DEA 1.2, 3, 5, or 7 should also be avoided. Recommendations through which “universal” donors may be selected are given (Fig. 1).
The limitation to the practice of good transfusion medicine in the dog is the lack of a clear, simple, and inexpensive typing assay for in clinic typing. Canine blood groups are typed, as are other species, with an agglutination assay. Reagents for typing are generated by canine alloimmunization. This process of producing polyclonal antibody in dogs takes several months to complete. Antisera must be screened for reactivity and verified against dogs of known type. Production of antisera by canine alloimmunization is difficult, time consuming, and costly. Currently, one consistent source of DEA typing sera exists. Dr. Robert Bull’s laboratory at Michigan State University has provided DEA typing reagents for almost two decades. Antisera for DEA 1.1 and 1.2 are also produced at Stormont Laboratories and the University of Guelph College of Veterinary Medicine. However, these two laboratories do not currently offer their typing sera for sale.
DEA typing based on agglutination and homolyses is performed by many laboratories throughout North America. In-clinic assays for all six DEA types are not commercially available. Table 3 provides a list of known typing centres. The performance of DEA typing on potential donors is now quite commonplace. Many laboratories type only for DEA 1.1 or for DEA 1.1 and 1.2. To date, only Dr. Robert Bull’s laboratory offers a complete DEA type, screening for 1.1, 1.2, 3, 4, 5, and 7.
Figure 1. Recommendations for minimizing the cost of identifying “universal donors” with current typing resources DEA = dog erythrocyte antigen.
Recently, Ejima and Andrews have produced monoclonal anti-bodies to DEA 3 and DEA 1.1, respectively (Hara, et al., 1991; Andrews, et al., 1992). Dr. Joseph Smith at Kansas State University offers a card typing kit for the recognition of DEA 1.1 based on the monoclonal antibody generated by Andrews. Monoclonal antibodies are the first step towards providing a consistent means of in clinic typing. The production of monoclonal antibodies may also lead to further biochemical definition of these antigens. Additional efforts to define the erythrocyte antigens of the dog on a molecular level are underway in several laboratories. Both advancements will allow for more rapid and uniform testing of the canine red blood cell or its DNA template.
A method is recommended through which “universal” donors may be selected at the lowest cost to the practitioner utilizing current resources (see Fig. 1). To lower the cost of mass screenings for the procurement of “universal” donors, field testing for DEA 1.1 through the card typing kit can be performed. Subjects negative for DEA 1.1 can then be tested for the remaining DEA types by an outside laboratory.
Table 3. Current listing of laboratories offering DEA typing in North America
- Andrews, G., Chavey, P. & Smith, J., 1992. Production, characterization and applications of a murine monoclonal antibody to dog erythrocyte 1.1. J Am Vet Med Assoc, Volume 10, p. 201.
- Bull, R., 1976. Canine Immunohematology. In The Animal Models of Thrombosis and Hemorrhagic Diseases, Washington, DC, US Dept. of Health, Education, and Welfare, Public Health Services: NIH.
- Colling, D. & Saison, R., 1980. Canine blood groups: I, description of new erythrocyte specificities. Animal Genetic, Volume 11, pp. 1-12.
- Colling, D. & Saison, R., 1980. Canine blood groups: II. Description of a new allele in the Tr blood group system. Animal Genetic, Volume 11, pp. 13-20.
- Hara, Y., Ejima, H. & Aoki, S., 1991. Preparation of monoclonal antibodies against dog erythrocyte antigen D1 (DEA-3). J Vet Med Sci, Volume 53, p. 1105.
- Swisher, S., 1954. Studies of the mechanisms of erythrocyte destruction initiated by antibodies. In The Transactions of the Association of American Physicians, Volume 97, p. 124.
- Swisher, S. & Young, L., 1961. The blood group systems of dogs. Physiol Rev, 14(3).
- Swisher, S., Young, L. & Trabold, N., 1962. In vitro and in vivo studies of the behavior of canine erythrocyte-isoantibody system. Ann NY Acad Sci, Volume 97, pp. 15-25.
- Symons, M. & Bell, K., 1992. Canine plasma alkaline phosphatase polymorphism and its relationship with canine Tr group system. Animal Genetic, Volume 23, pp. 315-324.
- Young, L., O’Brien, W. & Miller, G., 1951. Erythrocyte-isoantibody reactions in dogs. New York Academy of Science Series II, Volume 13, p. 209.