Vox Sang. 29: 440 449 (1975)

Cell Electrophoretic, Membrane Sialic Acid and Quantitative Hemagglutination Studies on some MN Variants' STEPHEN J. LUNER,PHILLIPSTURGEON, DOROTHY SZKLAREK and DOROTHY T. MCQUISTON Gwynn Hazen Cherry Memorial Laboratories, Department of Pediatrics, Hematology Division, UCLA School of Medicine, Los Angeles, Calif.

Abstract. The group of conditions exhibiting diminished M N antigenicity, increased saline agglutinability, decreased electrophoretic mobility and reduced membrane content of sialic acid includes enzyme-treated cells, the hereditary MNSs variants MI< and Mn, En(a-) and the acquired condition of persistent mixed-field polyagglutinability. Here we report our studies on the above serological, chemical and biophysical properties of Mn, Mk and EnaEn? cells and on two additional hereditary variants, Miltenberger 111 and V, (Mi-I11 and Mi-V). The latter clearly fits into this groups of conditions. On the other hand, Mi-I11 shows its kinship to the broad group of abnormalities of membrane glycophorin but it deviates from normal in the opposite direction. That is we find evidence of decreased saline agglutinability, increased electrophoretic mobility and of increased sialic acid content. Moreover, in the rare MsMi-I** Mk phenotype, the opposing effects evident in the heterozygotes tend to balance out their serologic and physicochemical expressions in the double heterozygote.

Introduction

Enzymatic treatment of erythrocytes results serologically in increased saline agglutinability by incomplete antibodies and, depending on the enzyme used, in specific antigenic transformations. The latter include loss 1 This investigation was supported by USPHS research grant No. AM 10722 from the National Institutes of Arthritis and Metabolic Disease, by research grant No. H L 11160 from the Heart and Lung Institute, and by research grant No. C A 13955 from the National Cancer Institute, National Institutes of Health, Bethesda, Md.

Received: November 8, 1974; accepted: February 19, 1975.

LUNER/STLJRGEON/MCQLJI~TON/SZKLAREK

44 1

of MN and Duffy antigens and the exposure of T and other antigens easily detectable with lectin reagents. Physicochemically such treatment is associated with loss of both membrane sialic acid and of electrostatic charge. Recent reports from our laboratories deal with a possible biochemical basis for these changes and give supporting specific and general references [ 10, 111. Naturally occurring MN variants, both hereditary and (probably) acquired, have been shown in recent complete reports from other laboratories [l, 5 , 61, as well as in preliminary 191 and complete [lo, 111 reports from these laboratories, to be associated with similar physicochemical membrane alterations along with a number of the serological transformations mentioned above. Recently we have applied all of the above methods, including the newly developed method of endless fluid belt electrophoresis to the study of normal human erythrocytes [9], to several MN variants, and to the abnormality of persistent mixed-field polyagglutinability (PMFP) [lo, 111. Generally our findings confirm those previous reports and add new information on two antigenic types of the Miltenberger complex, Mi-I11 and Mi-V. In particular, the method of electrophoresis employed gives a photographic presentation showing how the two antigens in heterozygotes of type Mi-I11 and Mk interact in the double heterozygote, Mi-III/Mk, to neutralize their independent electrophoretic characteristics. The method also graphically portrays the electrokinetic abnormality in other MN variants. It is the purpose of this report to present these pictorial electrokinetic records and to compare them with corresponding sialic acid and quantitative hemagglutination data.

Methods and Materials Fluid belt electrophoresis. In brief, a 2-5% cell suspension is injected through a capillary tube into a layer of low ionic strength buffer solution through which passes an electric current and which additionally is set in motion around a central core by a magnetic field. Within a few seconds these forces generate in the buffer solution a helical streak of cells that is macroscopically visible and which may be easily photographed in dark-field illumination through a viewing window. A mixture of two samples of cells having different electrostatic surface charge density resolves almost immediately into two streaks. Precise details of the method and for calculating relative and absolute mobilities are given in previous publications from these laboratories [2, 91. Relative electrophoretic mobility is determined by mixing equal quantities of unknown cells with cells from a normal individual. Of 120 medical students studied

442

LUNER/STURGEON/MCQUISTON/SZKLAREK

in this way with one arbitrary reference cell (S.L.), no separation or even increase in streak width was observed, except in two instances (3). In both there was distinct separation, but the samples showed no changes in agglutinability or sialic acid content. In one case, cells from other members of the family also exhibited abnormal electrophoretic mobility. Sialic acid. The thiobarbituric acid method, as described previously [2], was used on RBC ghosts. The membrane content was expressed as a ratio of nanomoles of sialic acid to micrograms of membrane protein determined using the Folin phenol reagent. Our normal result for membrane sialic acid content determined on fresh blood from healthy donors is a mean value of 100 t 4 (SD) nmol/mg. There is no association with M N type. Quantitative hernagglutinatiorz. A quantitative hemagglutination system based on the automated continuous flow blood typing machines developed in these laboratories [9] was used to measure the reactivity in a n anti-hr” (e) reagent serum. The cells were not enzyme-treated, and macromolecular substances (eg. PVP) were not used to augment agglutinability. Anti-hr” was used because all of the experimental cells were hr” positive, although their phenotypes differed (table I). Because this reagent, a t low dilutions, exhibits the serological properties of a saline agglutinating serum and of greater reactivity with rhi (Ce) than hr (f, ce) complexes, it was used only a t higher dilutions. At a dilution of 1/32 or greater, it behaves as an incomplete serum. Regardless of cell type, it no longer agglutinates cells in a saline medium but only reacts with enzyme-treated cells or cells in a medium of high protein concentration. The studies reported here were done with the serum diluted 1/32 using cell samples 7-16. Agglutinability was expressed as a percentage [9]. Serological methods. Traditional manual blood typing methods were used to determine the phenotypes of the control and experimental cells. In many instances, the latter came from individuals who had been the subject of extensive blood grouping studies in conjunction with previous reports (see below). Blood samples. Six experimental samples and two controls collected at the Zurich Red Cross Blood Transfusion Center were generously provided by M.N. METAXAS.The initials of the donors and their blood types are given in table 1. Type MgM, MgMP, EnaEn? [4] and Miltenberger class V were represented. In addition, seven controls and seven experimental bloods were obtained in Los Angeles. The latter included examples of Mk, MMi-111, and MkMsMi-111; all came from three generations of a large Chinese familiy which had been the subject of a previous report from these laboratories [8]. Additionally, two members of this Chinese familiy, who did not have variant M N types, were studied by electrophoresis only.

Results

Data from determinations of electrophoretic mobility, sialic acid content, and agglutinability for the MN variants (No. 1-9) and for the control cells (No. 11-18) are listed in table I. The two type Mg

Physicochemical Aspects of M N Variants

443

Table I. Mobility, sialic acid content and agglutinability of MN variants and control cells

No.

1

2 3 4

5 6 7 8

9 10

11 12 13 14 15 16 17 18

Initials’

G.G. K.G. H.R. K.F. R.C. E.B. E.C., 11-5 C.S., 11-4 H.Q., 11-9 P.S., ficin L.H.S. D.McQ. A.M.H. H.M.F. P.J. P.S. E.D. M.C.

Types

Electrophoretic mobility * Membrane sialic acid decrease, O/O protein 1st test 2nd test ratio 3

Rh

MNSs

Rh,rh Rh,Rh Rh,rh Rhgh Rh,rh rh Rh,Rh, Rh,Rh, Rh,Rk, Rh,rh Rh,Rh, Rh,rh Rh,Rh, rh Rh,rh Rh,rh Rh,Rh, Rh,rh

2.6 (M.C.) 4.9 (M.C.) 7.3 (M.C.) 8.6 (M.C.) 6.6 (M.C.) 11.3 (M.C.) NsMk 9.8 (S.L.) MsMsMi-111 (S.L.) 4 MkMsMi-111 1.4 (S.L.) MSNS 62.0 (P.S.) MsNs NsNs -

MSMgs MgsMgs EnaEn? EnaEn? MS/NsMi-v NsNsMi-v

-

MNSs MSNS -

4.0 (M.C.) 2.6 (E.D.) 5.3 (E.D.) 6.6 (E.D.) 10.6 (E.D.) 11.3 (S.L.) (S.L.)4 2.8 (S.L.) -

98.9,98.1 96.2 86.8 75.5 80.8 75.3 71.8 105.2 79.0 10.0

-

-

-

-

-

-

-

102.0 99.5 102.7

0.0 0.0 0.0

-

-

Agglutination with 1/32 dilution of anti-hr”, O/o 30.5 0 0.8 86.1 0 0 1.2 0 0 0 -

The combination of Roman and Arabic numerals refer to the previously published pedigree of the kinship from which these specimens came. The P.S. cells treated with ficin in sample No. 10. 2 Initials of reference cell donor are given in parentheses. 9 Value for reference cell (S.L.) is 96.5. 4 Increased streak width (see figure 2) but not sufficient to ascribe a definite percent change in mobility.

bloods (No. 1 and 2, G.G. and K.G.) show slight but distinct decreases in relative electrophoretic mobility of from 2.6 to 4.9%. The magnitude of this difference is seen graphically in figure 1. The first two views are from the comparably collected and transported control bloods No. 17 and 18 (M.C. and E.D.) run separately. Even with five passages, as each of the control cells follows its helical path from right to left past the viewing window, there is but one streak. But in the 3rd view, where M.C. is used as a reference and is mixed with sample 1, the MgM blood (G.G.), a relatively broad streak is seen at each passage.

444

LVNER/STURGEON/MCQUISTON/SZKLAREK

M. C.(NN Control)

E. D. (MN Control) M. C.-1-

G.G. MY M

M. C.i K. G.Mg Mg

-+

M. C, E. B. NN" M. C.4 R. C,MN" M. C.t H. R . Ena En? M. C.-k K. F. Ena En? Fig. 1. Dark-field photographs of streaks generated in the endless belt electrophoresis apparatus. Each streak or pair of streaks traverses the viewing window five times, initially entering the field of view at the right and spending an additional 20 sec i n the electric field before each successive appearance from right to left. Thus, the greatest separation attained in the apparatus appears at the far left. The first two views show the streaks formed by suspensions of control cells, and the remaining six are mixtures of these control cells with MN variant cells.

With sample 2, the type MgMg (fourth view), more separation is evident, and with Mi-V and EnaEn? almost complete separation is present on the fourth passage. The percent decrease in electrophoretic mobility, calculated from measurements made on these photographs, is significantly greater with the latter variants (No. 3-6; table I). With the variant samples 7, 8 and 9 of table I, the aberrations in electrophoretic mobility vary from little change with 8 and 9, the MsMsMi-"' and MkMsMi-II1,to approximately 10% reduction with NsMk (No. 7). This is shown photographically in figure 2. In the third view, the type NsMk is clearly separated on its second passage from its

j

Physicochemical Aspects of MN Variants

-

~

1-

445

S.L.+II-5 NMk S.L.+II-9

M

I U k

M

Fig. 2. Streak pattern showing the effect on electrophoretic mobility of Mk and of Miltenberger class I11 traits separately and combined in the same individual.

reference cell while, at best, with sample 8, MsMsMi-II1(11-4, second view) increased streak width is suggested. From the sialic acid value of 105.2 for sample 8 (table 1) the possibility is raised that this increased streak width represents increased mobility of Mi-I11 rather than, as has been the case heretofore, decreased mobility. Finally with type MkMsMi-I1I (sample No. 9, 11-9) the separation seen with Mk alone is almost completely absent, again suggesting that Mi-I11 has increased mobility. Figure 3 illustrates an electrophoretic study of four Mk members from three generations of the Q family showing clearly the hereditary nature of the electrophoretic anomaly. The mixture of the reference cells (S.L.) with the Mk negative members, 11-1 and 111-3, shows no separation. In these studies the instrument was operated to give only three passages across the viewing window. The sialic acid data for the MN variants (No. 1-9; table I) correspond remarkably well with the electrophoretic studies. Sialic acid measurements were made on control bloods 16, 17 and 18 ( P A , E.D., and M.C.), and fell within our normal range. Relative electrophoretic mobilities of these control bloods, determined against reference cells S.L., showed no evidence of heterogeneity. The sialic acid value of 98.1-98.9 at best, for the MSMrs blood, falls on the low side of normal while 96.2 for MgsMgs is at the lower limits. EnaEn?, Mi-V and Mk heterozygotes (No. 3-7) range from 86.8 down to 71.8 for NsMk, all well below the lower limits of our normal range, while the value of 105.2 for Mi-I11 is at the upper limits of normal. The value of 79.0 for MkMsMi-III lies intermediate between the two heterozy-

446

s.L.+ 1-1

M M ~

S.L.+ I [ - 1

MN

m-1

N M ~

s.L.+a-2

M M ~

s.L.+

S.L.4-

III - 3 MN

s.L.+a-4 N

M ~

Fig. 3. Electrophoretic streak patterns of cells Erom Mk-positive and Mknegative family members. Here only three passages across the viewing window are shown.

gotes, corresponding to the position it holds in the mobility and agglutinability studies. For comparison, ficinized normal cells have a sialic acid value of only 10 (see sample 10, table I). The percent agglutinability of untreated control bloods in relation to Rh type is shown by samples 11 through 16 of table I. At a 1/32 dilution of the anti-hr” serum, there is no agglutination except for No. 13 which gave a 1.2% reading. This is insignificant since it falls well within the precision of the machine’s ability for hemoglobinometry. To illustrate that there is activity at the 1/32 dilution, enzyme treated P.S. cells (No. 10) gave very strong (86%) agglutination. The three MN variants collected in Los Angeles (No. 7-9) show clear-cut differences in saline agglutinability from the controls of appropriate Rh type (table I). Type NsMkRh,Rh, (No. 7) gives distinct agglutination, 30.5%. In contrast, the Mi-I11 Rh,Rh, blood (No. 8) shows no agglutination. The MkMsMi-II1 double heterozygote (No. 9) is also Rh,Rh, but its absence of reactivity distinguishes it from the single Mk heterozygote, NsMk (No. 7). In the original investigation of the Chinese family from which the current samples were obtained, automated quantitative agglutinability

Physicochemical Aspects of MN Variants

441

studies were performed using an incomplete anti-s serum. At that time, the observations appeared contradictory and were not included in the report. With the better understanding of the nature of these variants, it is now apparent that the two sets of observations are entirely consistent. In brief, at 1/40, 1/20 and 1/10 dilutions of the anti-s, all seven Mk members gave from 70 to 83% agglutination, while the three members with normal MN types and the two type Mi-I11 members showed less than 25% agglutination at all three dilutions. The three MkMsMi-IIr members ranged between 35 and 63%.

Discussion

Combined serological and physicochemical studies of hereditary MN variants comparable to those performed here were first reported by FURUHJELM et al. [ l ] and NORDLINC et al. in 1969 [6]. MYLLYLA et al. [ 5 ] also made similar investigations of what is probably an acquired condition PMFP. Our findings, for the most part, are confirmatory of theirs for the MN variants and, as we recently reported, for PMFP as well [lo, 111. In the latter case we stressed that the abnormal (A) cells were clearly aberrant in their MN type. Thus it has been demonstrated and independently confirmed that Mk, Mg, En(a-) and PMFP all share various degrees of membrane (‘Envelope’) glycopeptide defects leading to a lower sialic acid content than in normal cells. As a consequence of this biochemical deficiency, electrostatic charge and zeta potential are diminished and, serologically, saline agglutinability is increased. That the agglutinability is essentially independent of the former has been clearly demonstrated by reports from other laboratories [7] and independently confirmed and extended in a report of our own [2]. This was shown by studying normal cells after removal of glycopeptide components, including sialic acid, by various enzymes. Neuraminidase treatment produces only slight to moderate increase in saline agglutinability but is associated with the largest decrease in electrophoretic mobility and in sialic acid content. Protease treatment brings about substantially greater agglutinability but with much smaller changes in electrophoretic mobility and in sialic acid content. The MN variants studied here have, in their sialic acid content and in surface charge and agglutinability, the characteristics of lightly protease-treated normal cells rather than those of neuraminidase-treated cells.

448

LUNER/STURGEON/MCQUISTON/SZKLAREK

In this report we add two more MN variants to the list that have been studied in the above manner. Mi-V fits in with the desialized class of variants. Mi-111, although a member of the Miltenberger group, is suggestively changed in the opposite direction. Sialic acid is in the high normal range, a slight increase in electrophoretic mobility is apparent in the heterozygote, and saline agglutinability is definitely not increased. Also in this study we report on the only family in which the two rare MN genes, Mk and M~~i-111 interact [8]. Here it is clear that Mi-I11 does in fact supress saline agglutinability as the type MkMsMi-IT1 sibs are substantially less reactive than are the Mk heterozygous sibs. The net physicochemical consequence of this genetic interaction is also a cell with values that lie intermediate between those of the two kinds of heterozygotes. Acknowledgement We are grateful to Dr. M. N. METAXASfor the generous gift of several M N variants as mentioned in the text and for a critical reading of the manuscript.

References 1 FURUHJELM, U.; MYLLYLA, G.; NEVANLINA, H. R.; NORDLING, S.; PIRKOLA, A.; GAVIN,J . ; GOOCH,A,; SANGER, R., and TIPPEIT,P.: The red cell phenotype En(a-) and anti-En&:serological and physicochemical aspects. Vox Sang. 17: 256 (1969). 2 LUNER,S. J.; STURGEON, P.; SZKLAREK, D., and MCQUISTON, D. T.: Effects of proteases and neuraminidase on RBC surface charge and agglutination - a kinetic study. Vox Sang. 28: 184 (1975). 3 LUNER, S. J. and SZKLAREK, D. : Comparison of electrophoretic mobility and membrane sialic acid content of erythrocytes from adult and umbilical cord blood. Pediat. Res. (in press). 4 METAXAS, M. N. and METAXAS-BUHLER, M.: The detection of MNSs ‘variants’ in serial tests with incomplete Rh antisera in saline. Vox Sang. 22: 474 (1972). 5 MYLLYLA, G . ; FURUHJELM, U.; NORDLING, S . ; PIRKOLA, A.; TIPPETT,P.; GAVIN, J., and SANGER, R. : Persistent mixed-field polyagglutinability : electrokinetic and serological aspects. Vox Sang. 20: 7 (1971). 6 NORDLING, S. ; SANGER, R. ; GAVIN,J. ; FURUHJELM, U. ; MYLLYLA, G., and METAXAS, M. N. : Mk and Mg. Some serological and physicochemical observations. Vox Sang. 17: 300 (1969). 7 STRATTON, F.; RAWLINSON, V. I. ; GUNSON,H. H., and PHILLIPS, P. K. : The role of zeta potential in Rh agglutination. Vox Sang. 24: 273 (1973). 8 STURGEON, P.; METAXAS-BUHLER, M. ; METAXAS, M. N . ; TIPPETT, P., and IKIN, E. W.:

Physicochemical Aspects of MN Variants

449

An erroneous exclusion of paternity in a Chinese family exhibiting the rare MNSs gene complexes Mk and MsIII. Vox Sang. 18: 395 (1970). 9 STURGEON, P.; KOLIN,A.; KWAK,K. S., and LUNER,S. J.: Studies of human erythrocytes by endless belt electrophoresis: I. A comparison of electrophoretic mobility with serological reactivity. Haematologia 6: 93 (1972). 10 STURGEON, P.; MCQUISTON, D. T.; TASWELL, H. F., and ALLAN,C.: Permanent mixedfield polyagglutinability (PMFP). I. Serological observations. Vox Sang. 25: 481 (1973). 11 STURGEON, P.; LUNER,S. J., and MCQUISTON,D. T.: Permanent mixed-field polyagglutinability (PMFP). 11. Hematological, biophysical and biochemical observations. Vox Sang. 25: 498 (1973).

Addendum Subsequent to the submission of this paper, we were informed by Dr. M.MEthat electrophoretic mobility and sialic acid determinations had been performed in 1971 by Dr. STIGNORDLING, 3rd Department of Pathology, University of Helsinki, on red cells of types MgMg, EnaEn?, Mi-I and Mi-V. Electrophoretic mobility was reduced below control values by 13V0 for Mi-V, 8010 for MgMs and 12010 for EnaEn? Reductions in sialic acid content were 27 O/O for Mi-V, 23 O/O for MgMg and 19Vo for EnaEn? Mi-I cells showed control values. These unpublished findings agree well with those reported in the present paper. TAXAS

Dr. STEPHENJ. LUNER,The Gwynn Hazen Cherry Memorial Laboratories, Department of Pediatrics, Hematology Division, UCLA School of Medicine, Los Angeles, C.4 90024 (USA)