TRANSFUSION Vol. 19
No. 5
September- October 1979
Scientific Articles
Augmentation of Hemagglutination by Low Ionic Conditions R. E. ROSENFIELD, S. H. SHAIKH, F. INNELLA, Z. KACZERA A N D S. KOCHWA From ihe Deparimeni of Paihology, ihe Mount Sinai School of Medicine. City Universiiy of N e w York, N e w York, N e w York
Short incubation at 37 C, 80 per cent reduction in ionic concentration and removal of liquid phases after each reaction step, provided the basis for the construction of four new serologic tests for alloantibodies to human erythrocytes. In the first, the incubation fluid was replaced with protamine sulfate to aggregate intensely the evaluated red blood cells. After dispersal by phosphate buffer, residual antibody mediated agglutination could be discerned. As a second method, this low ionic polycation (LIP) test was followed by a normal ionic IgG antiglobulin test (LIP-ACT). A third method employed low ionic washing of erythrocytes and low ionic antiglobulin serum (LIAGT). Finally, a modified LIP test was conducted entirely under low ionic conditions and followed by a low ionic antiglobulin test (modified LIP-ACT). LIP, LIP-ACT and LIACT were successfully employed for all routine blood bank serology tests. Their sensitivity and impact on blood bank performance are described.
PRETRANSFUSION serologic tests performed to detect an immunologic reaction between the red blood cells of a donor and the serum of a patient cannot be fully depended upon to reveal evidence of incompatibility. The existence of a problem in this essential area of laboratory service was first reported by Fudenberg and Allen5 and a recent s ~ r v e y , ' ~indicated *'~ that well over a half of all hemolytic transfusion Received for publication May 12, 1979; accepted June 1, 1979.
reactions were associated with failure of pretransfusion tests to allow recognition of immunologic imcompatibility between patient's serum and donor red blood cells. It is obvious, therefore, that test procedures such as those currently recommended by the American Association of Blood Banks* are inadequate and that other tests with more sensitivity are urgently needed. . ~ defined the Hughes-Jones et ~ 1 clearly capacity of low ionic incubation conditions to increase greatly both the rate and total amount of anti-Rh binding by Rh-positive red blood cells. Rosenfield et d . 1 9 and Lalezari" described AutoAnalyzer channels that incorporated a combination of low ionic incubation and polycation aggregation of red blood cells to achieve highly sensitive direct hemagglutination by a wide variety of human alloantibodies. Moore and Mollison13 described low ionic (LISS) incubation of red blood cells with alloantibodies to increase the sensitivity of detecting an antigen-antibody reaction by IgG antiglobulin test. The present report describes several procedures that possess all of the qualitative advantages of both Aut~Analyzer"*~~ and LISS13 tests. All of these tests employ 80
0041-1132/79/O9OO/0499 $01.10 0 J. B. Lippincott Co. Transfusion September-October 1979
499
Volume 19 Number 5
ROSENFIELD ET AL.
500
per cent reduction in ionic concentration to increase the rate and amount of antibody binding by red blood cells. Two of the tests use the polycation, protamine sulfate, to achieve direct hemagglutination by antibodies, while the others are based on IgG antiglobulin reactions. These procedures are uniquely applicable for compatibility testing, blood typing and identification of antibodies. Materials and Methods Glycine is used as a 1.9% solution in distilled water with the pH adjusted to 7.0. Protamine for LlPis Protamine Sulfate Injection, U.S.P., lOmg activity per ml, (Ely Lilly, Indianapolis, IN, or Upjohn, Kalamazoo, MI). For Modified LIP Upjohn powder is dissolved in 3 ml instead of 5 ml of diluent (16.7 mg activity per ml), and this solution further diluted 1:5 with glycine. Phosphate Buffer for LIP consists of an equal mixture (v/v) of 0.2 M phosphate, pH 7.3, and 0.9% NaCI. Heparin Phosphate for Modified LIP consists of an equal mixture (v/v) of 0.2 M phosphate, pH 6.0, and 0.9% NaCI, to which 100 units heparin are added ( e . g . , 9 parts phosphate-saline plus 1 part heparin, 1000 units/ml). Phosphate Glycine is a mixture (v/v) of 80 parts glycine, 18 parts 0.9% NaCI, and two parts 0.2 M phosphate buffer, pH 7.3. Low Ionic Antiglobulin Serum is rabbit anti-human Fraction 11diluted to contain 7 pg antibody N per ml. The diluent is a solution of 0.5% bovine albumin, 1.5% glycine, and 0.2% NaCl with pH adjusted to 7.0. The bovine albumin, originally 22% (Ortho Diagnostics, Inc.), was first dialyzed against glycine. In these studies, one drop is 0.033 ml ( i . e . , 30 drops/ml) unless otherwise designated. Low Ionic Polycation (LIP) Augmentation of Hemagglutination Procedure. This test consisted of placing one drop of serum containing (or suspected of containing) hemagglutinins in a standard 75 mm blood typing test tube having either 8 or 10 mm I.D. (10 or 12 mm O.D.). To this, one drop of washed or unwashed 3% red blood cell suspension in 0.9% NaCl was added. The mixture of serum and cells was warmed to 37 C, after which eight drops of prewarmed glycine were added and mixed to reduce ionic concentration by 80 per cent. Low ionic incubation at 37 C was
Transfusion
September-October 1979
terminated after five minutes, the test tube was centrifuged to sediment the red blood cells, and all supernatant fluid was aspirated and discarded. The button of sedimented red blood cells was fully resuspended in two drops of glycine and one drop of protamine, and the test was reincubated at 37 C for two minutes. The test tube was now centrifuged very briefly ( e . g . , 1,000 x g for 1 to 5 seconds) to sediment the protamine aggregated red blood cells, and the supernatant fluid was aspirated and discarded. Upon addition of two or three drops of phosphate buffer, the test tube was agitated to determine whether the red blood cells would fully disperse in the buffer or remain clumped. Judgment as to whether LIP tested red blood cells would disperse in buffer usually depended upon having a suitable comparison with a nonantibody containing control test. This consisted of testing the same red blood cells under identical conditions but substituting type AB serum or plasma, known to be free of allohemagglutinins, for the serum known (or suspected) to contain antibody. Bovine albumin, 7 to 7.5% in 0.9% NaCl could also be substituted, but lower concentrations of albumin or plain saline supported “stickiness” of cells to glass and small aggregates of cells that were difficult to break up. To minimize the problem of agitating two or three test tubes under reasonably identical conditions, the test tubes were placed in a rack attached to a suitable rotator such as is used for a V.D.R.L. test for syphilis. The rack held the test tubes snugly so that rotation in position was insignificant. The amounts of buffer added depended on the I.D. of the test tube: with drop quantities of 0.03 ml each, three drops was suitable for 10 mm 1.D. (12 mm O.D.) test tubes whereas only two drops was best in a test tube of 8 mm 1.D. (10 mm O.D.). When the test tubes with buffer were agitated at 180 rpm for one to two minutes, controls tended to be dispersed without undue loss of residual antibody-mediated hemagglutination. Interpretation of’Results. When LIP tests were evaluated microscopically, tiny clumps of residually aggregated cells were often seen. A 6 to 7X hand lens was more suitable for evaluation. The question asked in interpreting a LIP test was whether the actual test displayed significantly more residual agglutination than the control. If it did, the test was highly likely to reflect an antigen-antibody reaction. Test ji) r Compatibility by LIP. LIP tests of the red blood cells of a prospective donor and the serum of a prospective recipient required two
50 1
LOW IONIC HEMAGGLUTINATION
Volume 19 Number 5
Table 1. Interpretation of LIP Compatibility Tests Donor Cells Patient Serum
Donor Cells AB Serum
Patient Cells Patient Serum
Interpretation
Auto-antibody; if donor cell agglutination exceeds patient cell agglutirlation, alloantibody may be present neg
Test invalid because donor cells are spontaneously agglutinated
POS
POS
POS
Tests inadequately dispersed
neg
POS
POS
Not observed
controls, one of the donor cells in type AB serum or plasma and an autocontrol of the patient's cells in the patient's serum. With red blood cells in three test tubes to be evaluated for residual agglutination after agitation in buffer, six results were possible. All six results are summarized in Table 1 where it should be noted that one result was never observed and another generally reflected inadequate dispersion of red blood cells. Antibody Titrations by LIP. LIP tests were designed to be performed on whole serum or plasma. Therefore, normal type AB serum or plasma was used as the diluent for construction of antibody dilutions. Whole serum, however, was an unsatisfactory diluent for anti-Lewis unless the serum was obtained from a donor known to be Le(a-b-). Type AB serum was also unsatisfactory for cold agglutinin titrations because of its usual content of weak cold agglutinins. For both anti-Lewis and cold agglutinin titrations, 7% bovine albumin in 0.9% NaCl was used as diluent. Cold agglutinin titrations were incubated in an ice bath for one hour or longer, and the glycine, protamine, and buffer reagents used were prechilled. Indirect IgG Antiglobulin Test Following LIP (LIP-ACT) Following performance and interpretation of a LIP test, the tested red blood cells were washed with saline and tested with commercial IgG antiglobulin serum. This test resembled a LISS testtg except that incubation was conducted at 80 per cent rather than 50 per cent reduction in ionic concentration, and the tested cells were exposed to protamine. Since supernatant fluid was removed twice in performance of the LIP test, only two instead of three washes in saline were required before antiglobulin serum was added.
Indirect LOWIonic Antiglobulin Test (LIAGT) This test was initiated exactly like LIP with a warmed mixture of one drop 3% cells and one drop serum, followed by eight drops of prewarmed glycine and incubation at 37 C for five minutes. After this low ionic incubation, however, protamine was not added. Instead, the red blood cells were washed three times with phosphate glycine and low ionic antiglobulin serum was used as the antiglobulin reagent. Controls with AB serum were not needed for LIAGT, but autocontrols were critically important for compatibility tests. Indirect Low Ionic Antiglobulin Test Following ModiJied LIP (LIP-LIAGT) A modified LIP test was conducted entirely with 80 per cent reduction in ionic concentration. To accomplish this, a low ionic protamine solution was prepared from Upjohn powdered protamine so that three drops corresponded precisely to the one drop protamine and two drops glycine employed for the standard LIP test. To disperse the protamine aggregated cells, heparin phosphate was used instead of phosphate buffer. This modified LIP test provided results reasonably comparable to LIP tests performed as described above. Heparin phosphate at pH 6.0 fully neutralized the cell-bound protamine of the LIP test without increasing ionic concentration. Such LIP tested cells were washed two times in phosphate glycine and then tested with low ionic antiglobulin serum as in LIAGT.
Results Low lonic Polycation (LIP) Hemagglutination Ionic Concentration. Specific hemagglutina-
tion was observed with maximal sensitivity when
ROSENFIELD ET AL.
the ionic concentration of the incubated reaction mixture of cells and serum was diluted 1 5 with glycine. Any additional increase in reaction volume with glycine resulted in loss of sensitivity. With 80 per cent reduction in concentration of both reactants and ions, incubation time at 37 C was found to be optimal at five minutes. Many antigen-antibody reactions could be incubated for longer periods without either adverse effect or gain in sensitivity, but some could not without at least a slight loss in avidity and titration value. These effects of longer incubation time were antibody source dependent and not related to specificity. They applied equally to LIP, LIP followed by IgG antiglobulin test (LIP-AGT), and low ionic IgG antiglobulin test (LIAGT). Polycation Aggregation. With removal of unwanted serum proteins following low ionic incubation, a single concentration of polycations was found to be satisfactory for all tested erythrocytes. One drop of protamine sulfate, 10 mg/ml, added to two drops of glycine supported the most satisfactory results. This concentration of polycations did not appear to depress the capacity of bound antibody to crosslink erythrocytes, while five times as much protamine did. Removal of unwanted serum proteins and, subsequently, of unbound polycations provided minimal opportunity for spontaneous elution of antibody and allowed protamine aggregation of erythrocytes to be dispersed with only a small amount of ions at a relatively neutral pH. Disaggregation was assisted by very brief centrifugation of the protamine treated cells to minimize “stickiness” of cells to glass. Such “stickiness” was more marked if low ionic incubation was conducted with too little or no serum protein. Disaggregation by rotator rather than by hand significantly improved the observation of agglutination by some anti-Kell antibodies. Specificities included KI, K2, K3, K4, and K6. Polybrene was not fully evaluated but, at the same concentration as protamine, it was found to be as effective providing that citrate was employed for deaggregation. Concentration of’ Tested Erythrocytes. The sensitivity of LIP tests was susceptible to concentration of erythrocytes as well as antibody. At a given concentration of antibody, greatest sensitivity was observed with tests of the smallest numbers of red blood cells. For practical purposes, however, from 3 to 5 x lo6 cells were needed to obtain a reasonable pellet that could be inspected carefully for agglutination, and lo7 or even 2 x lo7 cells was easier for most tech-
Transfusion
September-October 1979
nicians to use. One 0.05 ml droplet that contains lo7 cells is obtained from a 2% cell suspension (2 x 10”cellslml), while a 3% cell suspension will deliver lo7 cells in a drop of 0.033 ml. LIP Followed by I g C Antiglobulin Test (LIP-ACT) The vast majority of human alloantibodies to red blood cell antigens are IgG, and it was desirable to have an IgG antiglobulin test to confirm the positive or negative results obtained by LIP. This was accomplished by washing the LIP assayed cells with 0.9% NaCI, adding appropriate IgG antiglobulin serum, and inspecting the cells for agglutination after light centrifugation. On titration of IgG alloantibodies, the sensitivity of this IgG antiglobulin test (LIP-AGT) was never less and sometimes two to three times more than results observed with LISS.’Z For LIP-AGT tests, the cells were dispersed with phosphate buffer and only normal ionic solutions ( e . g . , 0.9% NaCI) could be used for washing the dispersed cells. These LIP tested cells, including controls, reaggregated spontaneously when they were washed in eitherglycine or phosphate glycine. With polybrene, reaggregation of dispersed cells occurred even when the assayed cells were washed with 0.9% NaCI. Washing LIP tested cells in saline was occasionally associated with “sticking” of red blood cells to the wall of the test tube. This problem, when severe, could be surmounted only by increasing the number of tested cells ( e . g . , 2 to 3 x lo7). Residual protamine on LIP tested cells was not important for sensitivity. If protamine was dislodged by heparin phosphate (polyanions, and reduction in pH), no loss of sensitivity was observed. However, full strength 0.2 M phosphate buffer, pH 7 . 3 , sometimes had a deleterious effect on sensitivity, while dilution of 0.2 M phosphate buffer with 0.9% NaCl provided consistently effective sensitivity when used ( v h ) from 20 to 80 per cent. The antiglobulin serum used after a LIP test had to be free of antibodies to complement components. Although sera from most normal donors without antibody did not put C3 and C4 on red blood cells under these low ionic conditions at pH 7.0, the sera of many patients did. LIP and LIP-AGT tests were designed to assay whole serum or plasma, and observed results did not depend on complement activation. Identical sensitivity was seen in comparative tests of fresh serum, heat inactivated serum, EDTA chelated serum, and heparinized plasma.
Volume 19 Number 5
LOW IONIC HEMAGGLUTINATION
Low Ionic Antiglobulin Test (LIAGT) With the development of LIP-AGT as a supplemental test for LIP, spontaneous elution of bound antibody in phosphate buffer, in the 0.9% NaCl used for washing, and in the applied IgG antiglobulin reagent posed a problem that required evaluation. Accordingly, 80 per cent reduction in ionic concentration was employed for incubation, for washing cells, and for the IgG antiglobulin serum, and both protamine and phosphate buffer were eliminated from the test. The immediate result was observation of weak agglutination of control cells. This was eliminated by adding 2% (v/v) amount of 0.2 M phosphate buffer to the glycine solution used for washing; 1% phosphate buffer was insufficient while 5% or more decreased the avidity and titer of anti-red blood cell antibody mediated agglutination. In this test, however, avidity and titer value were almost invariably reduced if normal ionic IgG antiglobulin serum was substituted for low ionic. With one patient thus far, an Rh-negative mother whose baby's erythrocytes were negative by normal ionic direct antiglobulin test, LIAGT provided slightly increased sensitivity over what was observed with LIP and LIP-AGT. In general, LIAGT was slightly less sensitive. However, numerous examples of antisera stored either frozen or at 4 C displayed more antibody activity by LIAGT than by either LIP or LIPAGT, and some displayed residual antibody only by LIAGT. Modified LIP Followed by LIAGT Modified LIP tested cells dispersed by heparin phosphate could be washed in phosphate glycine without undergoing reaggregation. They were, however, more difficult to evaluate than LIP cells dispersed by phosphate buffer because the tested cells tended to adhere more to glass, controls required more agitation for dispersal, and sensitivity was sometimes less. In addition, LIAGT following modified LIP was not quite as sensitive as LIAGT alone, and this was apparently related to antibody elution. If the LIAGT procedure was sufficiently protracted to allow antibody elution, it lost sensitivity. Evaluation o j LIP, LIP-ACT and LIAGT A serological test to detect red blood cell antigen-antibody reactions can be evaluated with known anti-red blood cell antibodies, and its sensitivity can be matched with that of other procedures by comparing titration data. This was done for low ionic polycation (LIP) augmentation of hemagglutination, for LIP tested cells
503
further assayed by IgG antiglobulin test (LIPAGT), and for low ionic IgG antiglobulin tests (LIAGT). For titration at 37 C, all sera were diluted in normal type AB serum except anti-Lea and anti-Le* which were diluted in 7% bovine albumin in saline. The latter diluent was also used for cold agglutinins. A summary comparison with standard normal ionic tests in saline, tests with ficin treated cells, and IgG antiglobulin tests is given in Table 2. From Table 2, it is clear that LIP and LIPAGT not only detected known anti-red blood cell antibodies but usually did so with sharply enhanced sensitivity. Indeed, the results with LIP often resembled working sensitivities previously experienced only with AutoAnalyzer test^.'^.'^ However, unlike the AutoAnalyzer ~ of tested tests of Berkman et ~ 1 all. examples antibodies were found to be effective by LIP. LIP and LIP-AGT failures, along with failures of normal ionic tests, were encountered with some antisera that had deteriorated in storage, and some of these alloantibodies were detected by LIAGT. LIP and LIP-AGT failures with fresh clinical materials have not been observed thus far. It is noteworthy that LIP was extraordinarily sensitive with all examples of Rh and Duffy antibodies, and was usually significantly more sensitive with Kidd antibodies. A mechanical exception to LIP superiority occurred with Kell where LIP sometimes was only one-tenth as sensitive as LIP-AGT unless a rotator was used for disruption of protamine aggregated red blood cells. With a rotator, Kell was observed essentially as well by LIP as by LIP-AGT. LIP tests were designed to be effective with IgM as well as IgG antibodies. The titer value of IgM anti-I at 4 C was the same by LIP method as by titration against ficin treated cells. At 37 C, IgM anti-Rh isolated by sucrose density gradient ultracentrifugation, native IgM anti-S, and native IgM anti-Lu" were also found to be fully active by LIP titration. All of these IgM alloantibodies tended not to support a positive IgG antiglobulin test performed after LIP. Ordinary cold agglutinins having a 1: 1000 titer at 4 C were not discerned by LIP at 37 C, and yet multiple examples of anti-M, anti-PI, anti-Le", and anti-Le* were recognized. Thus far, all of these usually cold-reactive alloantibodies that were detected by LIP at 37 C were also detected by LIP-AGT, suggesting that the sera contained at least some IgG antibodies. Compatibility Tests Despite the use of two control tests and a demand for more maneuvers than are required
ROSENFIELD ET AL.
504
Transfusion September-October 1979
Table 2. Normalized' Reciprocal Dilutions of Average Normal and Low Ionic Titrations Normal Ionic Titers Specificity
Rh Rh Rh MN (rabbit)
ssu
Kell Duffy Kidd Lutheran xg" P1 A and B li (4C) li (37C) Lewis (human) Lewis (goat)
Saline
Low Ionic Titerst
Ficin
AGT
LIP
1 0.1 1
0.1 1 1
10 1* 1-5 5-10 1-10 1* 10-20 1-5 1-5 2-4 1-3 3 10-20 1 1-2 300
1 1 1 1 1 1 1
1 1 1
No. 5
September- October 1979
Scientific Articles
Augmentation of Hemagglutination by Low Ionic Conditions R. E. ROSENFIELD, S. H. SHAIKH, F. INNELLA, Z. KACZERA A N D S. KOCHWA From ihe Deparimeni of Paihology, ihe Mount Sinai School of Medicine. City Universiiy of N e w York, N e w York, N e w York
Short incubation at 37 C, 80 per cent reduction in ionic concentration and removal of liquid phases after each reaction step, provided the basis for the construction of four new serologic tests for alloantibodies to human erythrocytes. In the first, the incubation fluid was replaced with protamine sulfate to aggregate intensely the evaluated red blood cells. After dispersal by phosphate buffer, residual antibody mediated agglutination could be discerned. As a second method, this low ionic polycation (LIP) test was followed by a normal ionic IgG antiglobulin test (LIP-ACT). A third method employed low ionic washing of erythrocytes and low ionic antiglobulin serum (LIAGT). Finally, a modified LIP test was conducted entirely under low ionic conditions and followed by a low ionic antiglobulin test (modified LIP-ACT). LIP, LIP-ACT and LIACT were successfully employed for all routine blood bank serology tests. Their sensitivity and impact on blood bank performance are described.
PRETRANSFUSION serologic tests performed to detect an immunologic reaction between the red blood cells of a donor and the serum of a patient cannot be fully depended upon to reveal evidence of incompatibility. The existence of a problem in this essential area of laboratory service was first reported by Fudenberg and Allen5 and a recent s ~ r v e y , ' ~indicated *'~ that well over a half of all hemolytic transfusion Received for publication May 12, 1979; accepted June 1, 1979.
reactions were associated with failure of pretransfusion tests to allow recognition of immunologic imcompatibility between patient's serum and donor red blood cells. It is obvious, therefore, that test procedures such as those currently recommended by the American Association of Blood Banks* are inadequate and that other tests with more sensitivity are urgently needed. . ~ defined the Hughes-Jones et ~ 1 clearly capacity of low ionic incubation conditions to increase greatly both the rate and total amount of anti-Rh binding by Rh-positive red blood cells. Rosenfield et d . 1 9 and Lalezari" described AutoAnalyzer channels that incorporated a combination of low ionic incubation and polycation aggregation of red blood cells to achieve highly sensitive direct hemagglutination by a wide variety of human alloantibodies. Moore and Mollison13 described low ionic (LISS) incubation of red blood cells with alloantibodies to increase the sensitivity of detecting an antigen-antibody reaction by IgG antiglobulin test. The present report describes several procedures that possess all of the qualitative advantages of both Aut~Analyzer"*~~ and LISS13 tests. All of these tests employ 80
0041-1132/79/O9OO/0499 $01.10 0 J. B. Lippincott Co. Transfusion September-October 1979
499
Volume 19 Number 5
ROSENFIELD ET AL.
500
per cent reduction in ionic concentration to increase the rate and amount of antibody binding by red blood cells. Two of the tests use the polycation, protamine sulfate, to achieve direct hemagglutination by antibodies, while the others are based on IgG antiglobulin reactions. These procedures are uniquely applicable for compatibility testing, blood typing and identification of antibodies. Materials and Methods Glycine is used as a 1.9% solution in distilled water with the pH adjusted to 7.0. Protamine for LlPis Protamine Sulfate Injection, U.S.P., lOmg activity per ml, (Ely Lilly, Indianapolis, IN, or Upjohn, Kalamazoo, MI). For Modified LIP Upjohn powder is dissolved in 3 ml instead of 5 ml of diluent (16.7 mg activity per ml), and this solution further diluted 1:5 with glycine. Phosphate Buffer for LIP consists of an equal mixture (v/v) of 0.2 M phosphate, pH 7.3, and 0.9% NaCI. Heparin Phosphate for Modified LIP consists of an equal mixture (v/v) of 0.2 M phosphate, pH 6.0, and 0.9% NaCI, to which 100 units heparin are added ( e . g . , 9 parts phosphate-saline plus 1 part heparin, 1000 units/ml). Phosphate Glycine is a mixture (v/v) of 80 parts glycine, 18 parts 0.9% NaCI, and two parts 0.2 M phosphate buffer, pH 7.3. Low Ionic Antiglobulin Serum is rabbit anti-human Fraction 11diluted to contain 7 pg antibody N per ml. The diluent is a solution of 0.5% bovine albumin, 1.5% glycine, and 0.2% NaCl with pH adjusted to 7.0. The bovine albumin, originally 22% (Ortho Diagnostics, Inc.), was first dialyzed against glycine. In these studies, one drop is 0.033 ml ( i . e . , 30 drops/ml) unless otherwise designated. Low Ionic Polycation (LIP) Augmentation of Hemagglutination Procedure. This test consisted of placing one drop of serum containing (or suspected of containing) hemagglutinins in a standard 75 mm blood typing test tube having either 8 or 10 mm I.D. (10 or 12 mm O.D.). To this, one drop of washed or unwashed 3% red blood cell suspension in 0.9% NaCl was added. The mixture of serum and cells was warmed to 37 C, after which eight drops of prewarmed glycine were added and mixed to reduce ionic concentration by 80 per cent. Low ionic incubation at 37 C was
Transfusion
September-October 1979
terminated after five minutes, the test tube was centrifuged to sediment the red blood cells, and all supernatant fluid was aspirated and discarded. The button of sedimented red blood cells was fully resuspended in two drops of glycine and one drop of protamine, and the test was reincubated at 37 C for two minutes. The test tube was now centrifuged very briefly ( e . g . , 1,000 x g for 1 to 5 seconds) to sediment the protamine aggregated red blood cells, and the supernatant fluid was aspirated and discarded. Upon addition of two or three drops of phosphate buffer, the test tube was agitated to determine whether the red blood cells would fully disperse in the buffer or remain clumped. Judgment as to whether LIP tested red blood cells would disperse in buffer usually depended upon having a suitable comparison with a nonantibody containing control test. This consisted of testing the same red blood cells under identical conditions but substituting type AB serum or plasma, known to be free of allohemagglutinins, for the serum known (or suspected) to contain antibody. Bovine albumin, 7 to 7.5% in 0.9% NaCl could also be substituted, but lower concentrations of albumin or plain saline supported “stickiness” of cells to glass and small aggregates of cells that were difficult to break up. To minimize the problem of agitating two or three test tubes under reasonably identical conditions, the test tubes were placed in a rack attached to a suitable rotator such as is used for a V.D.R.L. test for syphilis. The rack held the test tubes snugly so that rotation in position was insignificant. The amounts of buffer added depended on the I.D. of the test tube: with drop quantities of 0.03 ml each, three drops was suitable for 10 mm 1.D. (12 mm O.D.) test tubes whereas only two drops was best in a test tube of 8 mm 1.D. (10 mm O.D.). When the test tubes with buffer were agitated at 180 rpm for one to two minutes, controls tended to be dispersed without undue loss of residual antibody-mediated hemagglutination. Interpretation of’Results. When LIP tests were evaluated microscopically, tiny clumps of residually aggregated cells were often seen. A 6 to 7X hand lens was more suitable for evaluation. The question asked in interpreting a LIP test was whether the actual test displayed significantly more residual agglutination than the control. If it did, the test was highly likely to reflect an antigen-antibody reaction. Test ji) r Compatibility by LIP. LIP tests of the red blood cells of a prospective donor and the serum of a prospective recipient required two
50 1
LOW IONIC HEMAGGLUTINATION
Volume 19 Number 5
Table 1. Interpretation of LIP Compatibility Tests Donor Cells Patient Serum
Donor Cells AB Serum
Patient Cells Patient Serum
Interpretation
Auto-antibody; if donor cell agglutination exceeds patient cell agglutirlation, alloantibody may be present neg
Test invalid because donor cells are spontaneously agglutinated
POS
POS
POS
Tests inadequately dispersed
neg
POS
POS
Not observed
controls, one of the donor cells in type AB serum or plasma and an autocontrol of the patient's cells in the patient's serum. With red blood cells in three test tubes to be evaluated for residual agglutination after agitation in buffer, six results were possible. All six results are summarized in Table 1 where it should be noted that one result was never observed and another generally reflected inadequate dispersion of red blood cells. Antibody Titrations by LIP. LIP tests were designed to be performed on whole serum or plasma. Therefore, normal type AB serum or plasma was used as the diluent for construction of antibody dilutions. Whole serum, however, was an unsatisfactory diluent for anti-Lewis unless the serum was obtained from a donor known to be Le(a-b-). Type AB serum was also unsatisfactory for cold agglutinin titrations because of its usual content of weak cold agglutinins. For both anti-Lewis and cold agglutinin titrations, 7% bovine albumin in 0.9% NaCl was used as diluent. Cold agglutinin titrations were incubated in an ice bath for one hour or longer, and the glycine, protamine, and buffer reagents used were prechilled. Indirect IgG Antiglobulin Test Following LIP (LIP-ACT) Following performance and interpretation of a LIP test, the tested red blood cells were washed with saline and tested with commercial IgG antiglobulin serum. This test resembled a LISS testtg except that incubation was conducted at 80 per cent rather than 50 per cent reduction in ionic concentration, and the tested cells were exposed to protamine. Since supernatant fluid was removed twice in performance of the LIP test, only two instead of three washes in saline were required before antiglobulin serum was added.
Indirect LOWIonic Antiglobulin Test (LIAGT) This test was initiated exactly like LIP with a warmed mixture of one drop 3% cells and one drop serum, followed by eight drops of prewarmed glycine and incubation at 37 C for five minutes. After this low ionic incubation, however, protamine was not added. Instead, the red blood cells were washed three times with phosphate glycine and low ionic antiglobulin serum was used as the antiglobulin reagent. Controls with AB serum were not needed for LIAGT, but autocontrols were critically important for compatibility tests. Indirect Low Ionic Antiglobulin Test Following ModiJied LIP (LIP-LIAGT) A modified LIP test was conducted entirely with 80 per cent reduction in ionic concentration. To accomplish this, a low ionic protamine solution was prepared from Upjohn powdered protamine so that three drops corresponded precisely to the one drop protamine and two drops glycine employed for the standard LIP test. To disperse the protamine aggregated cells, heparin phosphate was used instead of phosphate buffer. This modified LIP test provided results reasonably comparable to LIP tests performed as described above. Heparin phosphate at pH 6.0 fully neutralized the cell-bound protamine of the LIP test without increasing ionic concentration. Such LIP tested cells were washed two times in phosphate glycine and then tested with low ionic antiglobulin serum as in LIAGT.
Results Low lonic Polycation (LIP) Hemagglutination Ionic Concentration. Specific hemagglutina-
tion was observed with maximal sensitivity when
ROSENFIELD ET AL.
the ionic concentration of the incubated reaction mixture of cells and serum was diluted 1 5 with glycine. Any additional increase in reaction volume with glycine resulted in loss of sensitivity. With 80 per cent reduction in concentration of both reactants and ions, incubation time at 37 C was found to be optimal at five minutes. Many antigen-antibody reactions could be incubated for longer periods without either adverse effect or gain in sensitivity, but some could not without at least a slight loss in avidity and titration value. These effects of longer incubation time were antibody source dependent and not related to specificity. They applied equally to LIP, LIP followed by IgG antiglobulin test (LIP-AGT), and low ionic IgG antiglobulin test (LIAGT). Polycation Aggregation. With removal of unwanted serum proteins following low ionic incubation, a single concentration of polycations was found to be satisfactory for all tested erythrocytes. One drop of protamine sulfate, 10 mg/ml, added to two drops of glycine supported the most satisfactory results. This concentration of polycations did not appear to depress the capacity of bound antibody to crosslink erythrocytes, while five times as much protamine did. Removal of unwanted serum proteins and, subsequently, of unbound polycations provided minimal opportunity for spontaneous elution of antibody and allowed protamine aggregation of erythrocytes to be dispersed with only a small amount of ions at a relatively neutral pH. Disaggregation was assisted by very brief centrifugation of the protamine treated cells to minimize “stickiness” of cells to glass. Such “stickiness” was more marked if low ionic incubation was conducted with too little or no serum protein. Disaggregation by rotator rather than by hand significantly improved the observation of agglutination by some anti-Kell antibodies. Specificities included KI, K2, K3, K4, and K6. Polybrene was not fully evaluated but, at the same concentration as protamine, it was found to be as effective providing that citrate was employed for deaggregation. Concentration of’ Tested Erythrocytes. The sensitivity of LIP tests was susceptible to concentration of erythrocytes as well as antibody. At a given concentration of antibody, greatest sensitivity was observed with tests of the smallest numbers of red blood cells. For practical purposes, however, from 3 to 5 x lo6 cells were needed to obtain a reasonable pellet that could be inspected carefully for agglutination, and lo7 or even 2 x lo7 cells was easier for most tech-
Transfusion
September-October 1979
nicians to use. One 0.05 ml droplet that contains lo7 cells is obtained from a 2% cell suspension (2 x 10”cellslml), while a 3% cell suspension will deliver lo7 cells in a drop of 0.033 ml. LIP Followed by I g C Antiglobulin Test (LIP-ACT) The vast majority of human alloantibodies to red blood cell antigens are IgG, and it was desirable to have an IgG antiglobulin test to confirm the positive or negative results obtained by LIP. This was accomplished by washing the LIP assayed cells with 0.9% NaCI, adding appropriate IgG antiglobulin serum, and inspecting the cells for agglutination after light centrifugation. On titration of IgG alloantibodies, the sensitivity of this IgG antiglobulin test (LIP-AGT) was never less and sometimes two to three times more than results observed with LISS.’Z For LIP-AGT tests, the cells were dispersed with phosphate buffer and only normal ionic solutions ( e . g . , 0.9% NaCI) could be used for washing the dispersed cells. These LIP tested cells, including controls, reaggregated spontaneously when they were washed in eitherglycine or phosphate glycine. With polybrene, reaggregation of dispersed cells occurred even when the assayed cells were washed with 0.9% NaCI. Washing LIP tested cells in saline was occasionally associated with “sticking” of red blood cells to the wall of the test tube. This problem, when severe, could be surmounted only by increasing the number of tested cells ( e . g . , 2 to 3 x lo7). Residual protamine on LIP tested cells was not important for sensitivity. If protamine was dislodged by heparin phosphate (polyanions, and reduction in pH), no loss of sensitivity was observed. However, full strength 0.2 M phosphate buffer, pH 7 . 3 , sometimes had a deleterious effect on sensitivity, while dilution of 0.2 M phosphate buffer with 0.9% NaCl provided consistently effective sensitivity when used ( v h ) from 20 to 80 per cent. The antiglobulin serum used after a LIP test had to be free of antibodies to complement components. Although sera from most normal donors without antibody did not put C3 and C4 on red blood cells under these low ionic conditions at pH 7.0, the sera of many patients did. LIP and LIP-AGT tests were designed to assay whole serum or plasma, and observed results did not depend on complement activation. Identical sensitivity was seen in comparative tests of fresh serum, heat inactivated serum, EDTA chelated serum, and heparinized plasma.
Volume 19 Number 5
LOW IONIC HEMAGGLUTINATION
Low Ionic Antiglobulin Test (LIAGT) With the development of LIP-AGT as a supplemental test for LIP, spontaneous elution of bound antibody in phosphate buffer, in the 0.9% NaCl used for washing, and in the applied IgG antiglobulin reagent posed a problem that required evaluation. Accordingly, 80 per cent reduction in ionic concentration was employed for incubation, for washing cells, and for the IgG antiglobulin serum, and both protamine and phosphate buffer were eliminated from the test. The immediate result was observation of weak agglutination of control cells. This was eliminated by adding 2% (v/v) amount of 0.2 M phosphate buffer to the glycine solution used for washing; 1% phosphate buffer was insufficient while 5% or more decreased the avidity and titer of anti-red blood cell antibody mediated agglutination. In this test, however, avidity and titer value were almost invariably reduced if normal ionic IgG antiglobulin serum was substituted for low ionic. With one patient thus far, an Rh-negative mother whose baby's erythrocytes were negative by normal ionic direct antiglobulin test, LIAGT provided slightly increased sensitivity over what was observed with LIP and LIP-AGT. In general, LIAGT was slightly less sensitive. However, numerous examples of antisera stored either frozen or at 4 C displayed more antibody activity by LIAGT than by either LIP or LIPAGT, and some displayed residual antibody only by LIAGT. Modified LIP Followed by LIAGT Modified LIP tested cells dispersed by heparin phosphate could be washed in phosphate glycine without undergoing reaggregation. They were, however, more difficult to evaluate than LIP cells dispersed by phosphate buffer because the tested cells tended to adhere more to glass, controls required more agitation for dispersal, and sensitivity was sometimes less. In addition, LIAGT following modified LIP was not quite as sensitive as LIAGT alone, and this was apparently related to antibody elution. If the LIAGT procedure was sufficiently protracted to allow antibody elution, it lost sensitivity. Evaluation o j LIP, LIP-ACT and LIAGT A serological test to detect red blood cell antigen-antibody reactions can be evaluated with known anti-red blood cell antibodies, and its sensitivity can be matched with that of other procedures by comparing titration data. This was done for low ionic polycation (LIP) augmentation of hemagglutination, for LIP tested cells
503
further assayed by IgG antiglobulin test (LIPAGT), and for low ionic IgG antiglobulin tests (LIAGT). For titration at 37 C, all sera were diluted in normal type AB serum except anti-Lea and anti-Le* which were diluted in 7% bovine albumin in saline. The latter diluent was also used for cold agglutinins. A summary comparison with standard normal ionic tests in saline, tests with ficin treated cells, and IgG antiglobulin tests is given in Table 2. From Table 2, it is clear that LIP and LIPAGT not only detected known anti-red blood cell antibodies but usually did so with sharply enhanced sensitivity. Indeed, the results with LIP often resembled working sensitivities previously experienced only with AutoAnalyzer test^.'^.'^ However, unlike the AutoAnalyzer ~ of tested tests of Berkman et ~ 1 all. examples antibodies were found to be effective by LIP. LIP and LIP-AGT failures, along with failures of normal ionic tests, were encountered with some antisera that had deteriorated in storage, and some of these alloantibodies were detected by LIAGT. LIP and LIP-AGT failures with fresh clinical materials have not been observed thus far. It is noteworthy that LIP was extraordinarily sensitive with all examples of Rh and Duffy antibodies, and was usually significantly more sensitive with Kidd antibodies. A mechanical exception to LIP superiority occurred with Kell where LIP sometimes was only one-tenth as sensitive as LIP-AGT unless a rotator was used for disruption of protamine aggregated red blood cells. With a rotator, Kell was observed essentially as well by LIP as by LIP-AGT. LIP tests were designed to be effective with IgM as well as IgG antibodies. The titer value of IgM anti-I at 4 C was the same by LIP method as by titration against ficin treated cells. At 37 C, IgM anti-Rh isolated by sucrose density gradient ultracentrifugation, native IgM anti-S, and native IgM anti-Lu" were also found to be fully active by LIP titration. All of these IgM alloantibodies tended not to support a positive IgG antiglobulin test performed after LIP. Ordinary cold agglutinins having a 1: 1000 titer at 4 C were not discerned by LIP at 37 C, and yet multiple examples of anti-M, anti-PI, anti-Le", and anti-Le* were recognized. Thus far, all of these usually cold-reactive alloantibodies that were detected by LIP at 37 C were also detected by LIP-AGT, suggesting that the sera contained at least some IgG antibodies. Compatibility Tests Despite the use of two control tests and a demand for more maneuvers than are required
ROSENFIELD ET AL.
504
Transfusion September-October 1979
Table 2. Normalized' Reciprocal Dilutions of Average Normal and Low Ionic Titrations Normal Ionic Titers Specificity
Rh Rh Rh MN (rabbit)
ssu
Kell Duffy Kidd Lutheran xg" P1 A and B li (4C) li (37C) Lewis (human) Lewis (goat)
Saline
Low Ionic Titerst
Ficin
AGT
LIP
1 0.1 1
0.1 1 1
10 1* 1-5 5-10 1-10 1* 10-20 1-5 1-5 2-4 1-3 3 10-20 1 1-2 300
1 1 1 1 1 1 1
1 1 1
