The P Blood Group System in Pigeon Breeders and Pigeon Breeder's Disease*

Transcription

1 The P Blood Group System in Pigeon Breeders and Pigeon Breeder's Disease* Dean Effler; Fredy Roland, M.D.;t and Ralph A. Redding, M.D., F.C.C.P.t A blood group P 1 -like antigen bas been found in gramnegative bacilli isolated from pigeon droppings, in pigeon serum, and on pigeon red blood ceds. As a probable resident of the environment of the pigeon loft, the antigen stimulated formation of anti-p 1 antibody in eight of 11 pigeon breeders who belonged to the P 2 blood group. Only one of 11 random hospitalized patients with the P 2 blood phenotype had a detectable titer of anti-p 1 antibody (P <.1). The titer of anti-p 1 antibody was not significantly different in symptomatic vs asymptomatic breeders. The presence or absence of anti-p 1 antibody could not be correlated wicb any band of immunoprecipitates formed between pigeon breeder's serum and crude extract of pigeon droppings. The P antigen was identified in pigeon-dropping extracts of breeders with the P 2 blood phenotype by Inhibition of hemagglutination. We conclude that anti-p 1 antibodies appear to be a component of the immunologic response to avian antigeil'i of pigeon breeders but are probably unrelated to the pathogenesis of pigeon breeder's disease. pigeon breeder's disease is a pneumonitis due to hypersensitivity which apparently occurs in a small minority of pigeon breeders in response to one or several inhaled antigens found in the environment of the pigeon loft. The nature of the inham- For editorial comment, see page 694 matory response in the lung appears to be an Arthus phenomenon 1-3 or delayed hypersensitivityl'-4 or both. In the environment of the pigeon loft, there are many antigenic substances to which asymptomatic as well as symptomatic breeders can produce an antibody r e s p o 2 7 n~ ~ -sthe e. nature of the specific antigen or antigens that may be responsible for the pulmonary inflammatory response has continued to provoke interest. In fact, symptomatic breeders do appear to have several unique immunologic responses. Utilizing immunoelectrophoresis, Fink et al 8 demonstrated an IgG response directed against pigeon serum y-globulin that was unique to ill breeders. Peeters et al 9 found a precipitin antibody directed against pigeon feces which appeared to be present exclusively in those breeders with pigeon breeder's disease. Finally, Berrens and Maesen 7 From the Brown University Program in Medicine and the Memorial Hospital, Pawtucket, Rl. work performed in partial fulfillment of requirements for a Master of Medical Science degree at Brown University. t Assistant Professor of Medicine {Clinical) and Consultant in Infectious Diseases. tassistant Professor of Medical Science and Director, Pulmonary Division. Manuscript received March 1; revision accepted May 5. Reprint requests: Dr. Redding, Memorial Hospital, Pawtucket, Rhode Island 286Q found that the low-molecular-weight fraction of a nondialyzable lyophilized extract of pigeon droppings contained antigen to which only symptomatic breeders produced a precipitin antibody. This fraction contained a protease, glycoproteins, and polysaccharides. The antigenic material appears to be found in the dusts or aerosols that are distributed in the loft. Pigeon serum proteins, 1 a protease, 7 and other antigens not yet identified 1 are present in pigeon droppings. Other antigenic materials include feathers and internal secretory and excretory fluids, but these sources apparently only have antigens which can be isolated from droppings as well. 1 Fungal spores have not been implicated in pigeon breeder's disease. Flooring materials might be implicated as agents initiating the inflammatory process, but this has not been examined in any previous publication. Lastly, bacterial products which are present in the droppings are a potential source of antigens. The P blood group in humans consists of P1 and P2 phenotypes. Only in individuals with the P2 phenotype, normally about one quarter of the population, can anti-p1 antibodies develop. Within the United States, this antibody response, when observed, usually follows a blood transfusion of P1 red blood cells. Since titers of anti-p1 antibody have been found in a number of P2 individuals with avian contact, and since a blood group P1-like substance has been found on cells of gram-negative bacilli isolated from pigeon droppings, 13 the following clinical and immunologic studies were performed. P BLOOD GROUP SYSTEM IN PIGEON BREEDERS 719

2 Questionnaire MATERIALS AND M E T H o ~ The Rhode Island Pigeon Racing Club in Pawtucket, RI, was visited, and their help in this study was solicited. Thirtynine out of approximately 75 breeders volunteered. Each pigeon breeder filled out a short questionnaire which covered respiratory and cardiac symptoms and history. At the same time a more lengthy questionnaire was taken home by each breeder. This latter questionnaire was concerned with the breeder's history of exposure to pigeons and the presence or absence of symptoms usually associated with pigeon breeder's disease. In an attempt to evaluate the validity of the breeders' responses, eight randomly selected asymptomatic breeders and all breeders ( 12 of 39) with respiratory complaints were intensively questioned during a personal interview about their answers. All individuals with a P 2 blood group and formation of anti-p 1 antibody were asked about previous blood transfusions. Servm from Pigeon Breeders Serum was collected from each of the 39 breeders. In addition, out of 43 random admissions to the Memorial Hospital, Pawtucket, Rl, 11 patients with the P 2 blood phenotype were identified. Serum from these patients was collected as control samples. Determination of Anti-P 1 Antibody Reagent red blood cells (human) (Identigen and Selectogen ) were utilized to screen and identify antibodies in pigeon breeders' and control sera. Testing for the red blood cell agglutination by P 2 pigeon breeder's serum was accomplished at room temperature and at 4 c as warm and cold panels, according to the protocol provided by the supplier.u Modification of the cold panel, as described by the supplier, 14 was done as follows: (1) four drops of a 2-percent red blood cell suspension was washed by using three aliquots (about 1 ml) of buffered saline solution (ph= 7.4), followed by centrifugation and decanting; ( 2) two drops of pigeon breeder's serum was added to a resuspended red blood cell button after the final wash, centrifugation, and decanting; and ( 3) the washed combination of red blood cells and serum was treated as described in the supplier's literature that describes the 4 c cold panei.u The Identigen panel of reagent red blood cells was used only if the Selectogen cells showed agglutination with the serum at room temperature or in the cold. Samples of serum were sent to the supplier of the reagent red blood cells for further evaluation if the samples contained cold antibodies which could not be identified with the two standard red blood cells reagents ( Identigen or Selectogen). Red Blood CeU Twing Whole blood was collected from all breeders, anticoagulated with disodium edetate (EDTA), and stored at 4"C until used. Red blood cells from the random population of inpatients at the Memorial Hospital were taken from day-old clots left over after routine admission laboratory tests. These red blood cells and the respective sera were used as controls. Typing of the cells for P antigenicity was accomplished with anti-p 1 serum (Dade Divis:on lot p-23ac) using the procedure suggested in the supplier's literature.u Ortho Diagnostics Inc., Raritan, NJ. 72 EFFLER, ROLAND, REDDING Typing of the cells used in the absorption techniques was accomplished using blood-bank reagents ( Ortho Diagnostics Inc) for slide-test typing of A, B,, and Rh antigens.16,17 Reverse typing using the donor's serum was done using known reagent red blood cells (human) ( Affirmagenll ) for blood group A and blood group B and following the supplier's procedure.ts Pigeon-Dropping Extract Pigeon droppings from the lofts of each of the 39 breeders were dried in a hot air oven at 75 C for eight to ten hours. Ten grams of this dried material was mixed with 1 ml of.85-percent sodium chloride solution and periodically agitated by hand to dissolve as much of the material as possible. After 3 minutes the fluid was decanted and centrifuged at 2,5 rpm ( 8 g) at room temperature for four minutes. The supernatant liquid was decanted and stored at -25"C as pigeon-dropping extract. Extracts used for absorption experiments were filtered through a.22,u. sieve ( Millipore) for sterilization. Immediately before use, the extracts were titrated with.1 N sodium hydroxide solution until the ph was between 6 and 8. Immunodiffusion Agar plates (Hyland Laboratory Immunoplate, pattern C) were obtained, and ten >. micropipettes were used to place the serum and pigeon-dropping extract into the agar wells. Two types of controls were used. The first-control was serum from a breeder (No. 39) whose findings from lung biopsy were consistent with pigeon breeder's disease and with known precipitin antibodies in his serum. The second control was a pigeon-dropping extract which offered clear and distinct patterns of precipitins with most pigeon breeder's sera. For each of the 39 breeders, the agar plate was arranged with the pigeon breeder's serum, his own extract, and the two eontrols. The agar plates were incubated at room temperature in a moist chamber and read at 24 and 48 hours. Plates were visualized before and after a fixation procedure 19 which helped identify the immunoprecipitates. The bands of precipitins were read with an indirect light source and a dark background. Absorption Techniques Pigeon-Dropping Absorbed Servm. A mixture of.4 ml of pigeon breeder's serum and.2 ml of the dropping extract from the breeder's own pigeons was allowed to sit at room temperature for six hours and then at 4 c overnight. On the next morning, this sample was spun at 4,5 rpm ( 3,5 g) at room temperature for n minutes. The supernatant liquid was removed and added to.2 ml of pigeon-dropping extract. The same incubation procedure was repeated again. On the next morning, centrifugation yielded.6 ml of pigeon-dropping absorbed serum, which was concentrated by incubation with.6 gm polyacrylamide gel ( Lyphogel) at 4"C. Twentyfour hours later, the resultant solution was used as pigeondropping absorbed serum. A control was simultaneously run, using.85-percent sodium chloride solution instead of pigeondropping extract; this was called the saline control. Other controls (control A and control B) were established using serum from two patients with nonspecific cold agglutinins; these sera were incubated with pigeon-dropping extract as described previously. IIOrtho Diagnostics Inc., Raritan, NJ.

3 Packed Red Blood Cells Absorbed Serum. A mixture of.4 ml of pigeon breeder's serum and.2 ml of washed packed red blood cells (, Rh-, P 1 ) was incubated at 4 c for one hour. The mixture was centrifuged at 2, rpm ( 7 g) at room temperature for one minute, and the supernatant liquid was quickly removed. To the supernatant was added.2 ml of washed packed red blood cells (, Rh-, P 1 ), followed by incubation and centrifugation as described previously. The final supernatant was called packed red blood cell absorbed serum. A paired control was run simultaneously, using saline-washed packed red blood cells (, Rh-, P 2 ). Both packed red blood cells absorbed serum and pigeondropping absorbed serum were tested for hemagglutination of P1 and P2 (, Rh-) cells. Both absorbed sera were also tested for irnmunoprecipitation with the dropping extract from the breeder's own pigeons and with a known pigeon-dropping extract control When testing for immunoprecipitation, serum from a patient with known pigeon breeder's disease was also included in one of the agar wells (see section on immunodihusion). All statistical comparisons in this study were done using x 2 analysis and Yates' correction factor where appropriate. Questionnaire REsULTS In their questionnaires, none of the 39 breeders indicated classic acute or subacute symptoms which could have been solely attributed to pigeon breeder's disease. The personal interview in regard to 2 breeders' written answers ( 12 symptomatic and eight asymptomatic) demonstrated a high degree of validity in their responses; there appeared to be full understanding of the acute or subacute symptoms mentioned. The presence of chronic bronchitis, chronic obstructive pulmonary disease, congenital heart disease, asthma, allergies, and hypochondriasis made it impossible to diagnose pigeon breeder's disease in the absence of symptoms temporally related to exposure. Interestingly, the one breeder (No. 39) who had both abnormalities of pulmonary function and a histologic picture on lung biopsy which were consistent with chronic pigeon breeder's disease also recalled no history of acute symptoms. As mentioned previously, there were 12 of 39 breeders who offered some respiratory complaints. Table I shows that significantly more breeders with the P2 blood phenotype had respiratory symptoms ( P <.1). This occurred despite the fact that anti PI antibodies and positive immunodiffusion patterns did not have a positive correlation with respiratory symptoms. The duration of the breeders' history of raising pigeons and clearing lofts varied from six months to 5 years, with a mean duration of 18 years. The number of pigeons owned varied from 3 to 2, with a mean of 84 birds. The time spent cleaning Table l ~ o m p aof rbreeder i a o n with or without Reapiratory Symptoma Sympto- Asymptomatic matic p Total No. of breeders 12 (31) 27 (69) P, blood phenotype 5 (42) 6 (22) p < O.ot Anti-Pt antibodies 2 (4) 5 (84) p >.2** Precipitins 4 (33) 14 (52) p >.2 *Table values are numbers of breeders; numbers within parentheses are percentages. **Yates' correction factor. the lofts varied from three times per year to greater than 45 minutes per week. Over 6 percent of the 39 breeders spent greater than 45 minutes cleaning the loft each week. The 39 breeders used a variety of flooring materials in their pigeon lofts (with some using combinations of materials), as indicated by the following list showing types of flooring materials and the number of breeders using each: lime, 23; sugar cane (bagasse), four; cotton seed waste, two; wood shavings, two; dry droppings, two; cedar chips, one; straw, one; tobacco, one; sand, one; Cabola (phenol, coal tar disinfectant), one; and nothing, nine. Determination of Anti-P1 Antibody The titers of anti-p1 antibody found in the breeders with the P2 blood phenotype are listed in Table 2. Samples of two breeders' sera and one control serum were sent to the supplier of the reagent ( Ortho Diagnostics Inc.) for further evaluation of a cold antibody. One of the two breeders had an anti PI antibody, while the other had an anti-1 antibody. The control serum from the hospitalized control patient had an anti-1 antibody and a very weak anti PI antibody that was only picked up using trypsinized reagent red blood cells. Eight ( 73 percent) of the II pigeon breeders with the P2 blood phenotype and one ( 9 percent) of the II random hospitalized controls with the P2 blood phenotype had titers of anti-p1 antibody (I= 9.2; P <.1). None of these nine subjects with titers Table 2--Titera of Anti-P 1 Antibody ir the Pi«eon Breeder with the P 2 Blood Phenotype No. of Titer of Anti-Pt Breeders Subjects Antibody 1 Breeder 24 1:32 2 Breeders 19 and 39 1:16 2 Breeders 7 and 9 1:8 1 Breeder 14 1:4 1 Breeder 22 1:2 1 Breeder 12 1:1 3 Breeders 15, 26, and 3 P BLOOD GROUP SYSTEM IN PIGEON BREEDERS 721

4 Of anti-p1 antibody (eight breeders and one control) had ever received a blood transfusion. CeU Typing Eleven ( 28 percent) of the 39 breeders and 11 ( 26 percent) of the 43 random hospitalized control subjects had the blood type, P2; 28 ( 72 percent) of the 39 breeders and 32 (74 percent) of the 43 random hospitalized control subjects had the blood type, P1. Immunodiffusion Eighteen ( 46 percent) of the 39 pigeon breeders and none of the 14 random hospitalized control subjects had at least one immunoprecipitin band to their own or a control pigeon-dropping extract (x 2 = 9.7; P <.1). Seven (64 percent) of the 11 breeders with the P2 blood phenotype and 11 (39 percent) of the 28 breeders with the P1 blood p h ~ notype had visible precipitin reactions in the immunoprecipitation system ( x 2 = 1.9, P <.2). Seven ( 88 percent) of the eight breeders with the P2 blood phenotype who had titers of anti-p1 antibody also had precipitin bands, while none of the three breeders with the P2 blood phenotype who did not have titers of anti-p1 antibody had precipitin antibodies against pigeon-dropping extract ( P <.5). Absorption of Pigeon Breeders' Sera Pigeon-Dropping Absorbed Serum. In regard to immunoprecipitation, absorption with pigeon-dropping extract from the breeder's own loft removed all of the precipitation antibodies to the droppings and to the pigeon-dropping extract control. The paired saline controls for the pigeon-dropping absorbed serum yielded the same banding pattern in agar gel as the unadulterated serum. In regard to hemagglutination, the pigeon-drop- Table 3--Pi«eo,...Droppin«Ab orbed Serum and Saline Controb: Hema««lulinalion of P 1,, Rh- Red Blood Cell at 4 c Phenotype and Pigeon-Dropping Paired Saline Subjects Absorbed Serum Control P 2 blood phenotype Breeder 7 3+ Breeder 14 I+ Breeder Breeder 22 I+ Breeder Breeder Breeder 9 Breeder I2 P, blood phenotype Breeder I mx Breeder 6 I+mx Breeder 1 Breeder 23 l+mx Breeder 28 I+mx Breeder 33 mx mx, Microscopic hemagglutination. ping absorbed sera did not agglutinate P1,, Rhred blood cells, yet all btrt two of the paired saline controls had hemagglutinating activity. Some of the sera from individuals with the P1 blood phenotype had a nonspecific hemagglutinating activity at 4 C using P1,, Rh- cells (Table 3) that was not present if these sera were incubated with pigeondropping extracts in the same manner as pigeondropping absorbed sera. Additional serum was not available to evaluate this finding. Processing the two control sera (controls A and B) that had nonspecific cold agglutinin activity in the same manner as pigeon-dropping absorbed serum did not significantly affect the ability of these sera to agglutinate red blood cells in the cold (Table 4). Packed Red Blood CeU Absorbed Serum. In regard to immunoprecipitation, the absorption with Table 4-Effeet of lneubali,.. with Pi«eo Droppi,.. E. traet on Speeifo; (anli-p 1 ) and Non peeifo; Hema««lulination at 4 c Titer 1:2 1:4 1:8 1:16 1:32 1:64 1:128 1:256 Breeder 39 (anti-p, antibody) Pigeon-dropping absorbed serum Saline control Control A.. Pigeon-dropping absorbed serum Saline control 2+ mx Control B** Pigeon-dropping absorbed serum Saline control mx mx mx, Microscopic hemagglutination. **Serum with a nonspecific cold agglutinin. 722 EFFLER, ROLAND, REDDING

5 Table r.--hemogglufinotion of, Ria-, P 1 Red Blood Cell Serum Absorbed Serum Absorbed with, Rh =, PI with, Rh =, P2 Subject Cells Cells Breeder 7 2+ Breeder 9 I+mx Breeder I2 I+ Breeder 14 I+ Breeder I9 Breeder 22 I+ Breeder Breeder mx, Microscopic hemagglutination. P1,, Rh- packed red blood cells had no affect on the banding pattern of pigeon breeders' sera against the pigeon-dropping extract from the breeders' own loft or against the pigeon-dropping extract control. The paired controls using P2,, Rh- packed red blood cells for absorption provided the same precipitin pattern as unadulterated sera. In regard to hemagglutination, the packed red blood cell absorbed sera had no ability to agglutinate P1,, Rh- cells. The controls prepared using P2,, RH- packed red blood cells retained the ability to agglutinate P1,, Rh - cells (Table 5). CoNCLUSIONs AND DIScussioN We are aware of the potential sampling error of this study, in that only volunteers were utilized in this study. Breeders with respiratory symptoms related to exposure to pigeons may avoid volunteering their serum and clinical history because they fear having to give up pigeon racing. Thus, the failure to identify breeders with acute symptoms related to exposure to pigeons in a group of 39 breeders may not accurately rehect the true incidence of the disease. 5 Utilization of questionnaires to identify previously unknown cases of pigeon breeder's disease was complicated by the presence of other underlying respiratory and cardiac diseases. These diseases may have been acting alone or in combination with the chronic form of pigeon breeder's disease. For example, one breeder with a progressive chronic obstructive pulmonary disease, exertional dyspnea, prolonged exposure to pigeons, strong titers of precipitin and anti-p1 antibody, pulmonary function consistent with pigeon breeder's disease, a lung biopsy which was consistent with pigeon breeder's disease of the chronic form, and no history for acute or subacute symptoms also smoke cigars. Without the lung biopsy and studies of pulmonary function, it would have been difficult to make a diagnosis other than chronic obstructive pulmonary disease. The questionnaire failed to identify any breeder with the acute or subacute symptoms of pigeon breeder's disease. Those with bird fancier's lung but without acute symptoms cannot be identified without specific in vitro tests, tests of pulmonary function and diffusing capacity, lung biopsy, proof of clinical improvement with avoidance of the antigenic material, or a positive provocative test in the laboratory. The discovery that a number of breeders use potentially allergenic materials in their loft which are not directly linked to the pigeons was enlightening. The use of the bagasse of sugar cane is of particular interest. H the use of bagasse is widespread nationally, some of the individuals presently diagnosed as having pigeon breeder's disease may, in fact, be suffering from bagassosis. The nature of the inert ingredients in Cabola, a disinfectant, was not determined. All materials that are used in the lofts of pigeon breeders and any substances that can be whipped into the air during cleaning or by agitated pigeons should arouse suspicion when searching for the etiologic agent of a pneumonitis due to hypersensitivity in a bird fancier. The presence of titers of anti-p1 antibody in pigeon breeders has been described, and the prevalence was recently investigated. 12 While the prevalence of titers of anti-pt antibody in breeders with the P2 blood phenotype in our study was double that found by Radermecker et al 13 (namely, 73 percent vs 34 percent), this difference was not significant because of small numbers (x 2 = 3.36;.1 > P >.5). The increased frequency of anti-p, antibody in those breeders with the P2 phenotype who were exposed to avian antigens was possibly related to inhaled pigeon serum proteins, pigeon cellular antigens, 13 gram-negative bacteria, 11 or autolysed fragments of bacterial cell walls. Serum proteins and autolysed fragments of bacterial cell walls seem most likely, since they would explain the inhibition of hemagglutination caused by the sterile pigeon-dropping extracts used in the absorption experiments. The anti-p1 antibody which was measured by hemagglutination appears to be in the IgM class. 13 It has been suggested that human isoautologous hemagglutinin antibodies may be stimulated by blood transfusions, ingested materials, intestinal bacterial Bora, or inhalation. 2 We believe that the stimulation of titers of anti-p1 antibody in pigeon breeders via the respiratory route is the first time isoautologous antibodies have been shown to be produced by antigens entering through the lungs in the absence of a clinical infection.. Although no one was identified as having the P BLOOD GROUP SYSTEM IN PIGEON BREEDERS 723

6 acute manifestations of pigeon breeder's disease, the previously mentioned breeder (No. 39) who has chronic pigeon breeder's disease did have an a n t i ~ Pt cold agglutinin titer ( 1:16) that was not significantly different from asymptomatic breeders. One cannot conclude from the fact that breeders with the P2 blood phenotype have more respiratory complaints that there is any relationship between the P blood group and pigeon breeder's disease. In fact, finding asymptomatic breeders with titers of anti-pt antibody and knowing that titers of anti-p1 antibody are not correlated significantly with respiratory symptoms strongly suggest that there is no relationship between the P blood group system and the symptoms of pigeon breeder's disease; however, it remains to be understood why the breeders with the P2 blood phenotype have more nonspecific pulmonary symptoms. It certainly has not been explained on the basis of length of exposure, smoking, or any risk factor examined in this study. The speculation of Radermecker et ap 3 that breeders with the P2 blood develop a more intense hypersensitivity response to avian antigens may indeed be true, since we did find a greater prevalence of respiratory complaints in breeders with the P2 blood phenotype. The frequency of occurrence of individuals with the P2 blood phenotype (namely, 28 percent) was similar to that found by others. 21 Although the incidence of anti-pt antibodies in the control population ( 1/11 or 9 percent) was radically different from the 56 percent offered by Henningsen, 21 the difference may be explained by different techniques, the strength of the Pt antigen on the reagent red blood cells, and the population sample. The important finding to be noted was that there was a significant difference between breeders with the P2 blood phenotype and a control population with the P2 blood phenotype when the same experimental procedure was utilized for both groups. The frequency ( 46 percent) of IgG precipitin antibodies 13 to avian antigens in the 39 breeders studied was similar to that found by others There was no significant difference in the incidence of precipitin antibodies or the immunoprecipitation patterns between the breeders with the Pt blood phenotype and breeders with the P2 blood phenotype; however, the frequencies of 39 percent and 64 percent, respectively, might have turned out to be significant with a larger population of breeders. A curious finding is the higher frequency of precipitin antibodies in those breeders with the P2 blood phenotype who had anti-pt antibodies than in those without anti-pt antibodies ( x 2 = 3.88; p <.5). There were no obvious differences between these groups relating to exposure. This relationship may be due to an unknown ejq>osure factor or to a genetic trait which provides different i m m u n o l o ~ c responsiveness. 13 Alternatively, it might only be a reflection of the fact that precipitin antibody and anti-pt antibody are stimulated simultaneously by antigens in the pigeon loft. As was expected, the pigeon-dropping absorbed sera contained no detectable precipitin antibody to pigeon-dropping extracts, while the controls which were incubated with saline solution performed as unadulterated sera. Hemagglutination with pigeondropping absorbed serum demonstrated that six of the eight sera with anti-pt antibody had the titers abolished by absorption, while the cold agglutinin activity in the saline control was retained. Sera from two of eight breeders lost the titer of anti-pt antibody in the saline control as well. The explanation for this second result is unknown, and no more serum was available to repeat the experiment. Curiously, control sera from individuals of the Pt blood phenotype showed weak nonspecific agglutination of the strong Pt red blood cells that were used to measure the titer of anti-pt antibody. After treating these sera as pigeon-dropping absorbed serum, this minimal agglutination disappeared. The following question had to be asked: does the pigeon-dropping extract have nonspecific antihemagglutinating factors? To evaluate this question, sera with nonspecific cold agglutinins were incubated with pigeon-dropping extracts (Table 3) as control absorbed sera. The weak nonspecific cold agglutinin activity from serum A (titer, 1:2) was lost after incubation with pigeon-dropping extract. This was analogous to the loss of the weak nonspecific agglutination of the Pt,, Rh- cells by Pt serum. Evidently the changes in the physical and chemical properties of the serum after mixing with pigeon-dropping extract results in some inhibition of hemagglutination; however, this was only noted if the agglutination was weak at the outset. For example, the nonspecific cold agglutinin in control serum B (titer, 1:128) displayed no alteration in cold agglutinin titer after incubation with pigeon-dropping extract. The nonspecific antihemagglutination effect of the pigeon-dropping extract would not be appreciated if the hemagglutinin titer was significant. Thus, it would appear that the failure of P2 pigeon-dropping absorbed sera to agglutinate Pt,, Rh - cells is the result of apt-like antigen in the pigeon-dropping extract. This could be explained by free Pt antigen liberated by metabolically active gramnegative bacilli or cell-wall fragments produced by bacterial autolysis. 724 EFFLER, ROLAND, REDDING

7 If pigeon-dropping extract absorbs out anti-pt antibody specifically, why are there no demonstrable changes in the immunodiffusion patterns using packed red blood cell absorbed sera? It is possible that if immunoelectrophoresis of crude pigeon-dropping extract or two-dimensional immunodiffusion of purified pigeon-dropping extract were used instead of the technique used in this present study, one might find a band which was consistently abolished as the titer of anti-pt antibody was removed by absorption. Alternatively, pigeon breeders' sera may produce a weak invisible anti-p.-(pt) band. A final possibility is that there may be soluble anti-p.-(pt) complexes in pigeon-dropping absorbed sera; therefore, anti-pt antibody would not be available for hemagglutination. In conclusion, anti-pt antibodies appear to be stimulated by the environment of the pigeon loft, along with a host of other antibodies and delayed hypersensitivity responses. Production of anti-pt antibody does not appear to be related to the manifestations of pigeon breeder's disease. As broad a scope as possible should prevail in searching for the etiology of the hypersensitivity response in symptomatic breeders, since a host of possible allergenic substances can be present in the atmosphere of the pigeon loft. Finally, a further area of research might be to elucidate why breeders With the P2 blood phenotype have significantly more respiratory complaints than breeders with the P. blood phenotype. REFERENCES 1 Pepys J: Pulmonary hypersensitivity to inhaled organic antigens. Ann Intern Med 52: , Fink J, Barboriak JJ, Sosman AJ: Immunologic studies of pigeon breeder's disease. J Allergy 39: , Calwell J, Pearce DE, Spencer C, et al: Immunologic mechanisms in hypersensitivity pneumonitis. J Allergy Clin Immunol 52:225-23, Mackaness G: Induction and expression of cell-mediated hypersentivity in the lung. Am Rev Respir Dis , Fink JN, Schlueter DP, Sosman AJ, et al: Clinical survey of pigeon breeders. Chest 62: , Fink J, Barboriak JJ, Sosman AJ, et al: Antibodies against pigeon serum proteins in pigeon breeders. J Lab Clin Med 71:2-24, 1968 'J Berrens L, Maesen FPV: An immunological study of pigeon breeder's disease. Int Arch Allergy Appl Immunol 43:289-34, , Fink J, Tebo T, Barboriak JJ: DiHerences in the immune responses of pigeon breeders to pigeon serum proteins. J Lab Clin Med 74:325-33, Peeters JABM, Brombacher PJ, Maesen FPV: Specific and non-specific precipitins in the serum of patients suffering from pigeon breeder's disease. Clin Chim Acta 34: , Edwards JH, Barboriak JJ. Fink JN: Antigens in pigeon breeder's disease. Immunology 19: , Roland F: Presence of the human blood group substance Pt in gram negative bacilli. Ann Microbioi124A:375-38, Roland F, EfHer D: Inhaled environmental avian bacterial antigens: Immunologic response in blood group P 2 persons. Chemosphere 4: , Radermecker M, Bruwier M, Francois C, et al: Anti-P 1 activity in pigeon breeder's serum. Clin Exp lmmunol 22: , Directions for Use of Identigen and Selectogen Reagent Red Blood Cells. Raritan, NJ, Ortho Diagnostics Inc, March Directions for Use of Anti-P 1 Serum. Miami, Fl, Dade Division of American Hospital Supply Corp, Dec Directions for Use of Anti-A, Anti-B, Blood Grouping Sera; Directions for Use of Anti-A, B Blood Grouping Serum; and Directions for Use of Anti-Rh Typing Serum Anti Rho (Anti-D). Raritan, NJ, Ortho Diagnostics Inc, July Directions for Use of Anti-Rh Typing Serum- Anti-Rh, Rh', rh" (Anti-CDE). Raritan, NJ, Ortho Diagnostics Inc, July Directions for Use of Affirmagen. Raritan, NJ, Ortho Diagnostics Inc, Aug Ouchterlony : ImmunodiHsuion and immunoelectrophoresis. In Weir DM (ed) : Handbook of Experimental Immunology. Oxford, England, Blackwell Scientific Publications, 1967, pp Goldsmith KLG: Blood transfusion serology. InGell PGH, Coombs RRA ( eds) : Clinical Aspects of Immunology. Oxford, England, Blackwell Scientific Publications, 1968, p Henningesn K: Investigations on the blood factor Pl' Acta Path Microbial Scand 26: , 1949 P BLOOD GROUP SYSTEM IN PIGEON BREEDERS 725