ABO bias in natural immunity
Peter D'Adamo ND
J. Naturopath. Med. 1991;2:11-17
ABBREVIATIONS: DTT: dithiothrehol; AHG: antihuman-globulin; BGD: blood group degrading enzymes
Sera from sixteen group O subjects reporting either urticaria, history of anaphylaxis, endometriosis, eczema or asthma were evaluated for the presence of isohemagglutinin-A (anti-A) by saline titration and human specific murine antiglobulins IgG1-4. Especially high saline titers and strong antiglobulin reactions were noted with urticaria, endometriosis and anaphylaxis when plotted against controls. Urticaria and anaphylaxis were also found to correlate with the broadest spectrum of IgG subclasses, with IgG 2 and IgG 3 appearing most consistently. Polymorphic factors may factor in pathogenesis, as differences in "natural immunity" observed with group O individuals appear to result in a less restricted, poorly behaved response (when T-cell dependent) to substances possessing A-like antigenicity.
From the age of six months normal subjects show "naturally occurring" antibody to non-self ABO antigens. These antibodies are largely IgM, but wider certain circumstances, particularly in group O, large amounts of "immune" IgG antibody may be produced. Almost thirty years ago, Rawson and Abelson () qualified physicochemical differences between isoanti-A,B and isoanti-A or B in attempting to explain the preponderance of group O women among mothers of infants with ABO hemolytic disease of the newbom. This observance, coupled with evidence that patients with acquired hemolytic anemias are more likely to belong to group O than to groups A or B, stimulated interest in further qualifying the differences between isohemagglutinin subclasses and bloodgroup.
By a combination of chromatography and ultracentrifugation, they separated the isohemagglutinin and isohemolysins into molecular types (i) gamma I and 2-globulins with S7 sedimentation coefficients (IgG) and (ii) gamma 1 -globulins with S19 sedimentation coefficients (IgM). In general, isohemagglutinins of S7 class showed maximal activity (titration endpoint) in the presence of antiglobulin serum, whereas S19 isohemagglutinins; are maximally active in sodium chloride solution. IgM class antibodies, although having relatively low affinity for single antigenic determinants, bind with great affinity to antigens with multiple epitopes and are efficient agglutinating and cytolytic agents. Membrane bound monomeric IgM is the major antibody receptor used by B-lymphocytes to recognize antigen, a very potent initiator of classical complement ABO fixation, and an effective first line of defense against bacteremia. The major class of isohemagglutinins (anti-A, anti-B) and many of the "natural" antibodies to micro-organisms are usually IgM. These have been postulated to arrive out of inapparent microbial inununization,() or are acquired characteristics resulting from oral immunization with animal foods containing A and B antigens.() Moreover, under some circumstances, particularly in subjects of group 0, large amounts of "immune" IgG class anti-A and anti-B can also be produced.
This heterogenicity of isohemagglutinin subclass, characteristic of group O, coupled with the wide spectrum of naturally occurring substances in the environment that possess blood group antigenicity, stimulated interest in investigating the activities of isohemagglutinins found in subjects of this blood group presenting with clinical symptoms of hypersensitivity, a correlation first advanced by Mourant,() who speculated that the presence of both anti-A and anti-B antibodies (in addition to a third hemolysin, anti-A,B) might render group O individuals more susceptible to allergic reactions.
|Strength Scale||Grade||Value|| Appearance of reaction|
|4+||Complete ||12||A single agglutinate. No free cells detected.|
|3 1/2+||4+w or 3+s||11|
|3+||3+||10||Strong reaction. A number of large agglutinates.|
|2-1/2+||3+w or 2+s||9|
|2+||2+||8||Large agglutinates in a sea of smaller clumps, no free cells. |
|1-1/2||1+s||6||Many medium and small agglutinates and a background of free cells|
|1+||1+||5||Many small agglutinates and a background of free cells|
|1+w||1+w||4||Many very small agglutinates; with many free cells|
|�+ or -||+/ macro||3||Weak granularity in cell suspension. A few macroscopic agglutinatesbut numerous agglutinates microscopically|
|Trace or micro||(+) macro||2||Appears negative macroscopically. A few agglutinates of 6-8 cells inmost fields|
|Questionable or micro||0||1||Rare agglutinates observed microscopically|
|0||0||0||An even cell suspension. No agglutinates observed microscopically.|
INTERPRETATION OF AGGLUTINATION REACTIONS. S-STRONG, W-WEAK. (SOURCE: AABB TECHNICAL MANUAL, 10TH EDITION. AMERICAN ASSOCIATION OF BLOOD BANKS, 1117 NORTH 19TH ST. STE 600 ARLINGTON VA, 22209)
In order to test this hypothesis, saline titration endpoints of group O individuals with a past or current history of allergic or inflammatory disorders were measured and compared to controls. When elevated saline titers were encountered, thiol neutralization and antiglobulin filtration were performed via standard technique.() Strong antiglobulin reactions were additionally subjected to further analysis via specific murine derived human-specific antiglobulin directed at subset groups IgG1 ,IgG 2 , IgG 3 and IgG 4 in a technique previously utilized by Kay and De Locke.(12)
Serum from sixteen subjects, all group O (3 men, 13 women) reporting clinical symptoms suggestive of immune hypersensitivity were examined for elevated titers of anti-A via saline agglutination endpoint technique. None of these subjects had ever been transfused, nor had any of the women reported heterospecific pregnancies.
As this method is semi-quantitative, titration results were assigned a number based on observed strength of reaction (Table 1). The sum of these values represent the 11 score - a measurement of antibody reactivity. A difference of 10 or more between different test samples has arbitrarily been deemed significant.
Saline titration is considered a good indicator of IgM isohemagglutinin activity. Test tubes were labeled according to serum dilution (I in 1, 1 in 2 etc.). Serum samples were number coded to eliminate observer bias. Serum (1 ml) was delivered to the first tube, and .5 ml of saline delivered to all tubes except the first. An equal volume (.5 ml) of serum was added to the second tube to make the 1 in 2 dilution. The tubes were agitated and one-half the contents of tube 2 transferred to tube 3 to make the I in 4 dilution. This was continued until the final dilution of serum (1 in 5096).
Reagent group Al cells were washed three times and a suspension of 4% cells in saline produced. Reagent cells were type Rh- and grouped in 3 banks according to MN blood groups. This was done to eliminate the only other major hemolysins (anti-Rh, anti-M and anti-N) which might have compromised the results. Reagent A1 cells (30ul) were then added via calibrated pipette. The samples were mixed and incubated at 371C for 15 minutes. Centrifugation at 1000 rpm for 45 seconds produced a cell button at the bottom of each tube which was then dislodged and graded by two observers. No effort was made to distinguish reactions which would require microscopic analysis (reaction less than 1/w; score less than 4).
Serum sample with strong (>1+) reactions, and high (>128) titration endpoints were subjected to further analysis for IgG subset activity. Thiol reagents possess the ability to cleave the intersubunit bonds characteristic of IgM, while leaving intact the disulphide bonds also characteristic of IgG and IgA monomers. Treating sera with thiol reagents abolishes both the agglutinating and complement-activating activity of serum IgM. Thiol neutralization is useful in determining immunoglobulin classes within the serum by allowing detection of co-existing IgG class antibodies otherwise masked by the higher avidity of IgM class antibodies.
In one tube .5 ml of serum was mixed with an equal volume of 0.01 M dissolved dithiothreitol (DTT) in phosphate buffered saline (PBS) and incubated at 37oC for two hours. Thus all indirect subclass determinations were performed at a 1:2 dilution. No effort was made to determine titration endpoint on the individual samples.
IgG subclass determination
IgG subclass specific antihuman-globulin (AHG) were obtained from Zymed Immunochemicals Inc. Each sample was tested for the presence of incomplete IgG anti-A subclasses using each of the four subclass (anti IgG,-,) specific AHG reagents. The treated serurn was thoroughly mixed with. I ml group Al cells and incubated for 15 minutes at 371C, after which the cells were washed using the Simwash System (Ortho Diagnostics). The appropriate IgG subclass specific AHG (. 1 nil) was then added, mixed and spun for 30 seconds at 1000 rpm. Each indirect subclass specific AHG test was inspected macroscopically and microscopically, Reaction strength was determined by the assessment score system previously described.
We found that individuals of group O blood reporting a previous urticaria or anaphylaxis showed high residual titers of anti-A isohemagglutinins. When a score value was assigned to each agglutination reaction to allow for characterization, it was found that individuals with these disorders showed remarkably high titration scores when measured against controls of the same blood group (Figures 1a and b). In several cases their scores showed almost threefold increases when measured against controls. A mild increase was noted for group O subjects with severe eczema or asthma but total scores were only marginally greater than controls and in both cases the numbers studied were quite small. However, a striking association was shown for group O women suffering from endometriosis.
Analysis of IgG subset characteristics showed consistent strongly positive reactions in the anaphylaxis and urticaria group when measured with murine derived humanspecific antiglobulin sera directed against individual IgG fractions. While almost all subjects had some IgG subset activity, the anaphylaxis and urticaria group had the most diverse and generally stronger (3 or 4+) reactions. What is particularly interesting in these groups was the unexpectedly high strength reactions consistently seen against both anti-A IgG 2 and anti-A IgG 3* This unique characteristic was also noted in both endometriosis subjects
TABLE 2. SUMMARY OF SALINE TITER AGGLUTINATION ENDPOINTS FOR 16 GROUP 0 INDIVIDUAL S REPORTING A VARIETY OF HYPERSENSITIVITY CONDITIONS. EACH REACTION IS RATED IN TERMS OF MACROSCOPIC AGGLUTINATION PATTERN (4-1 1W; UPPER NUMBER, SEE TABLE 1) AND INDIVIDUAL SCORE VALUE (12-4; LOWER NUMBER). CUMULATIVE SCORE VALUES (SCR) REPRESENT THE TOTAL OF INDIVIDUAL SCORE NUMBERS. SUMMARY OF ANTI-A IGG SUBSETS AS MEASURED BY INDIRECT ANTIGLOBULIN REACTION. WHEN NOTED, REACTIONS WERE GRADED FROM 4 (STRONG) TO 1 (VERY WEAK). NOTES. 1 ALSO REPORTED ASTHMATIC SYMPTOMS. 2 REPORTED PREVIOUS URTICARIA. 3REPORTED PREVIOUS ANAPHYLAXIS. I ALSO REPORTED FOOD AND AIRBORNE ALLERGIES. ALSO REPORTED POSITIVE RHEUMATOID FACTOR. CONTROLS: (NOT SHOWN, N= 16, GROUP 0) WERE RANDOMLY SELECTED FROM A BANK OF LYSOPHILIZED, FROZEN SERA AND TESTED FOR SALINE LITER SCORES. OBSERVER BIAS WAS CONTROLLED BY ASSIGNING RANDOM IDENTIFICATION NUMBERS TO BOTH THE HYPERSENSITIVITY SUBJECTS AND THE CONTROLS.
It might be expected that since allergy is an immunologic disorder, allergic diseases would tend to show disturbances of blood group frequencies. It is therefore rather disappointing to find that apart from asthma (in 1684 cases the A/O relative incidence was 1/13 -just below significance; the B/ O ratio of 1-47 - highly significant) and eczema (no significant departure from controls), cases examined for blood groups have been rather few and for allergy as a whole rather inconclusive.(4) Although the goal of our study was to examine possible response variations and not blood group frequencies, the relative blood group frequency of previous anaphylaxis and group 0 should be further examined. Of patients presenting at our clinic in the last two years reporting urticaria, or previous anaphylaxis (n=11), two urticaria patients were not group 0. Several subjects reported more than one concurrent disorder (Table 2). Reports of previous anaphylaxis were noted to be exclusively from group O patients. Eczema and asthma were reported in relatively equal numbers regardless of ABO group.
The occurrence of particular subclasses cannot be accounted for by expected percentage as IgG 2 and IgG 3 account for less than 30% of the total IgG found in healthy serum.() IgG 3 is the most potent of all the subsets at fixing complement via the classic pathway, in addition to possessing a high avidity for monocyte binding.
Blood group associations are fraught with difficulty.() Concurrent factors such as race and geography can be great obstacles to analysis. This study should not be interpreted as implying direct causality exists between isohemagglutinin titer and the pathogenesis of any of the conditions reported. Rather, it would indicate that in group O subjects these isohemagglutinins and other "natural" antibodies of early mechanisms are interactive if not interdependent.
In general, isohemagglutinin studies on groups A and B are underreported. These antibodies are almost entirely IgM class and rarely show dramatic elevations in titer score. McDuffie and Kabat () studied the Coombs test behavior of anti-A and anti-B and reported that of 18 group O individuals the sera of 15 showed enhancement of anti-A and anti-B by the Coombs test as compared to only 2 of 10 sera from non-group O sources. Contreras et al showed that the IgG response is greater in group O subjects than those of groups A or B after deliberate immunization with appropriate antigen.() It was also noted that even before immunization the presence of some IgG antibody could be shown in all blood groups, although in a lower titer in those of groups A and B. In 1986 Kay and Locke () concurred with this finding and characterized particular IgG 1.4 subsets in those groups with elevated IgG titers. Their findings showed that group O subjects produced a wider spectrum of response and a greater strength of antibody within each subclass. Although only a few subjects in their study produced strong reactions within IgG, and IgG 3 subclasses, with the exception of one subject, all were group O.
In a previous report, Kay had speculated that group O subjects could produce IgG antibodies to non-self ABO system antigens more readily than those of groups A or B because there is a greater antigenic disparity between group O and groups A/B than between group A and group B. This was postulated to allow a greater ability to provide T cell help during the response to T-dependent (glycoprotein) forms of ABO system antigens in subjects of group 0.() ABO glycoprotein is the predominant environmental form observed in substances possessing A-like antigenicity. T cell help is known to allow responding B cells to change from IgM production to IgG, giving a less restricted response. Responses to T independent antigens are known to be more X: restricted, often to IgM alone.()
Glycoproteins with A antigenicity are ubiquitous in nature, occurring not only in humans, but in the blood and tissue of lower animals, () a variety of natural products,() and a wide variety of microbia. One study showed almost 50% of 282 aerobic Gram negative bacteria were found to be blood group active. () They are extensively degraded in the human colon ecosystem, mostly from heat sensitive obligate anaerobes.() Human feces contains many blood group degrading enzymes (BGD); most if not all glycosidases. BGD are absent from meconium and appear in infant stool after the first few months of life, perhaps indicating increasing induction from developing microflora. Interestingly, they have been found in achlorhydric gastric secretions but not in normal acidic gastric juice. The specificity of BGD is affected by ABO group and secretor status.( These observations should be viewed as interdependent: the large enteric population of fecal anaerobes are an important source of A antigen substrate for A-degrading bacteria. Exogenous and enterically developed A substance and its degradation byproducts present a non-A host with a constant, dynamic antigen burden, which appears to be of special significance to individuals of group O who possess the inherent ability to produce anti-A in both IgM and IgG classes, resulting in a less restricted, "poorly behaved" response to challenge by substances with A-antigenicity.