Cytometry 11:261-271 (1990)

0 1990 Wiley-Liss, Inc.

Flow Cytometric Analysis of Erythrocyte Populations in Tn Syndrome Blood Using Monoclonal Antibodies to Glycophorin A and the Tn William L. Bigbee: Richard G. Langlois, Larry H. Stanker, Martin Vanderlaan, and Ronald H. Jensen Sciences Division, Lawrence Livermore National Laboratory, University of California, Livermore, California 94550 Received for publication January 20, 1989; accepted August 29, 1989

Flow cytometric analysis employing the Gal-GalNAc disaccharide. The permonoclonal antibodies to the Tn antigen centages of Tn-positive red cells in samand glycophorin A was used to charac- ples from six unrelated Tn donors ranged terize the erythrocyte populations pre- from 28 to 99%.Binding of the glycophorin sent in blood samples from individuals A-specificmonoclonal antibodies showed with Tn syndrome. Four monoclonal an- that the erythrocytes composing the Tntibodies specific for the Tn antigen, Gal- negative fraction presented normal NAc monosaccharide, on human erythro- amounts of the M and N epitopes on glycytes were obtained from a fusion of cophorin A. The presumed somatic mutasplenocytes from a Biozzi mouse immu- tional origin of Tn-positive cells was nized with red cells from a Tn individual. tested in blood samples from five normal These monoclonal antibodies specifically donors; three possible Tn cells were obrecognize GalNAc monosaccharide sites served after analysis of a total of 1.1 x lo' located on the erythrocyte cell surface erythrocytes, suggesting that the fresialoglycoproteins, glycophorin A and quency of such cells in normal individuals glycophorin B, and do not bind to fixed is < 1 x normal red cells presenting the NeuNAca2-3Gal~l-3(NeuNAca2-6)GalNAcalKey terms: ELISA, hemagglutination, 0-Ser(Thr) tetrasaccharide or to fixed immunofluorescent labeling, somatic muneuraminidase-digested cells presenting tation

The Tn antigen has been shown to be an incompletely synthesized 0-linked carbohydrate on cell surface glycoproteins, resulting in the presentation of normally cryptic N-acetylgalactosamine (GalNAc) [4,14-16,24,42,51]. The defect is a loss of 3-P-D-galactosyltransferase activity in Tn-positive (Tn + ) human erythrocytes, granulocytes, lymphocytes, and platelets [ll-13,311. The Tn syndrome is defined by the presence of polyagglutinable erythrocytes [MI. It has been observed in a number of asymptomatic subjects and has also been associated with various hematologic abnormalities including hemolytic anemia, leukopenia, thrombocytopenia, and leukemia [1,4,8,9,12,14,16,23, 24,31,32,41-43,52,57]. As originally proposed by Sturgeon et al. [51], the syndrome is characterized by a heterogeneous population of Tn + peripheral cells that lack the transferase activity and normal Tn-negative (Tn-) cells [ll-131. As suggested by Sturgeon et al. [50] and Haynes et al. [26], the defect has been inferred

'Supported in part by grant No. R811819-01 from the U.S. Environmental Protection Agency and by grant No. 2S07-RR05917-04 from the U S . National Institutes of Health under the auspices of the U S . Department of Energy by the Lawrence Livermore National Laboratory under contract No. W-7405-ENG-48. 'DISCLAIMER. This document was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the University of California nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owner rights. Reference herein to any specific commercial products, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or the University of California. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or the University of California, and shall not be used for advertising or product endorsement purposes. "Address reprint requests t o Dr. William L. Bigbee, Biomedical Sciences Division, Lawrence Livermore National Laboratory, University of California, P.O. Box 5507, L-452, Livermore, CA 94550.

262

BIGBEE ET AL.

to be a genetic alteration in bone marrow precursor cells [45,511. The somatic mutation-induced clonal origin of the disorder has been recently confirmed by the demonstration of T n + and Tn- clones derived from burst-forming-unit erythroid (BFU-E), colony-formingunit eosinophil (CFU-E,), colony-forming-unit granulocyteimacrophage (CFU-GM),and mixed colony-forming-unit (CFU-GEMM) colonies grown from blood samples obtained from two Tn syndrome patients [ 10,55,56]. These data suggest that the Tn phenotype may be a useful marker for in vivo somatic cell mutations in hemopoietic stem cells expressed in peripheral blood cells. Here, we report the partial characterization of four monoclonal antibodies (mAbs) specific for the Tn antigen on human erythrocytes, their application to the flow cytometric analysis of the expression of Tn determinants on erythrocytes, the cellular composition of Tn blood samples, and the frequency of Tn + red cells in normal individuals.

lin or dimethylsuberimidate as described previously [331. Comparable results were obtained with both fixation methods. The blood sample from Tn donor BM was used a s a source of white blood cells and platelets. Leukocytes were obtained from the buffy coat and platelets from low-speed centrifugation of the plasma.

Antibodies GPA-specific mAbs 6A7,9A3,10F7, MN1, "3, and NN5 have been described previously [2,3,331. MAbs 6A7 and 9A3 are specific for the M form of GPA (GPA(M); NN3 and NN5 bind only to GPA(N) and GPB; and 10F7 and MN1 are GPA specific, but do not distinguish the M and N forms. The hybridomas, as well a s the Tn-specific hybridomas described below, were grown in ascitic form as previously described [31 and the mAbs purified by ammonium sulfate precipitation followed by chromatography over Sephacryl S300 (Pharmacia, Inc., Piscataway, NJ) for the IgM mAbs or DEAE-Sephacel (Pharmacia) for the IgGl mAbs. The preparations were > 95% immunoglobulin as judged by SDS-PAGE. Antibody concentrations were estimated from absorbance with EiFo = 14. The mAbs were conjugated with fluorescein or biotin as described previously [33], except that some biotin-mAb conjugates were prepared using the biotin-carbohydrate method described by O'Shannessy et al. [441. Alkaline phosphatase-conjugated goat anti-mouse IgG and IgM were obtained from Sigma Chemical Co., St. Louis, MO; f luoresceinated goat anti-mouse IgG (GAMY-F), fluoresceinated goat anti-mouse IgM (GAMp-F), and peroxidase-conjugated goat anti-rabbit IgG, from Cappell Laboratories, Malvern, PA. TexasRed avidin (Av-TR) was obtained from Molecular Probes, Inc., Eugene, OR. Mouse isotype-specific rabbit antibodies were purchased from Miles Laboratories, Elkhart, IN.

MATERIALS AND METHODS Blood Samples Blood samples from normal human donors were obtained with informed consent from healthy volunteers. Samples from six unrelated Tn individuals were provided by the following sources: J. Hardman, Community Blood Center, Kansas City, MO (donor CL); K. Sigmund, American Red Cross, Southeast Michigan Chapter, Detroit, MI (donor DOS); Dr. T. Kickler, The Johns Hopkins Hospital, Baltimore, MA (donor NH); Dr. C . Hodson, National Blood Transfusion Service, Lancaster, Great Britain (donor BA); R.L. McShine, Red Cross Blood Bank, Groningen-Drenth, The Netherlands (donor SC); and P. Pilkington, American Red Cross, Penn-Jersey Region, Philadelphia, PA (donor BM). All samples, except for donor BM, were reported to contain > 90% T n + erythrocytes, donor BM 50% T n + . Samples were drawn in one of the standard anImmunization ticoagulants (ACD, EDTA, or heparin) and stored a t A month-old male Biozzi mouse was immunized by 4°C. All analyses were performed within 2 weeks of three intraperitoneal (IP) injections at 2-week intersample collection. vals of 0.2 ml of a 1 0 ~ / suspension ~1 in sterile saline of washed erythrocytes from donor CL. Ten days after the MN Blood Group Typing last IP injection, 0.1 ml of the suspension was given The blood samples were screened for MN blood group intravenously (IV) for 2 successive d. The spleen was activity using commercial rabbit anti-M,N typing sera removed 3 d later. (Ortho Diagnostics, Raritan, NJ). The majority of the red cells from the Tn donors CL, DOS, NH, and BA Hybridoma Isolation were not agglutinated with either sera. Cells from doSplenocytes were fused with SP210 myeloma cells nor BM were partially agglutinated by both sera; cells and the hybridomas cloned as described previously [21. from donor SC were unreactive with the anti-M serum but were agglutinated normally by the anti-N serum. On d 11following the fusion, media from each well was assayed for Tn-specific antibody production by screening against glutaraldehyde-fixed Tn (donor CL) and Cell Preparation normal red cells using a whole red cell ELISA [21. Neuraminidase and trypsin red cell digestions were Thirty-two wells were selected based on positive bindperformed as described earlier [2,3]. Flow cytometric ing of the antibodies to the Tn cells and negligible bindanalyses were performed on red cells fixed with forma- ing to normal cells. Supernatants from these cultures

-

FLOW ANALYSIS OF Tn SYNDROME ERYTHROCYTES

were re-assayed on d 14 using ELISA analysis against Tn erythrocytes (donor CL) and normal red cells from a second normal donor. Nine cultures, selected for antibody titer and specificity, were further expanded in T-25 flasks and subcloned twice by limiting dilutions in 96-well plates. Four hybridoma clones, ETnl.01, ETnl.02, ETnl.05, and ETnl.06 were obtained. ELISA, using Ig subclass- and Ig light chain-specific antibodies, revealed the following isotypes: ETnl.01, ETnl.02, ETnl.05, IgM(K); ETnl.06, IgGl(K).

263

RESULTS ELISA Binding Specificities of Tn MAbs Fusion of a spleen from a mouse immunized with erythrocytes from Tn donor CL resulted in the isolation of four hybridoma clones (ETnl.10, ETnl.02, ETnl.05, and ETnl.06) secreting mAbs of apparent Tn specificity as determined by ELISAs using Tn cells from the same donor and normal cells from six donors of differing MM, MN, NN, A, B, 0, and Rh phenotypes as antigens. ELISAs using erythrocytes from two additional Tn donors (DOS, NH) confirmed the Tn specificities of all of the mAbs. Results obtained using red cells from Tn donor NH and a normal donor are presented in Figure 1. ETnl.01 and ETnl.02 showed residual binding to trypsinized Tn erythrocytes; ETnl.05 and ETnl.06 exhibited no binding to trypsinized Tn cells over that observed to normal cells. Also as observed by ELISA, no binding of any of the Tn mAbs was observed to neuraminidase-treated cells from normal donors.

MAb Binding Assays For erythrocyte hemagglutination assays, whole blood was sedimented and the cells washed twice in Alsever's solution (2% w/v glucose in 0.07 M NaC1,0.03 M sodium citrate buffer, pH 6.0) then resuspended in PBS (0.15 M NaCl, 0.01 M sodium phosphate buffer, pH 7.2). Purified mAb solutions were serially diluted with PBS; 0.5 ml of each dilution was pipetted into a 12 x 75 mm glass tube. To each tube, 0.1 ml of washed erythrocyte suspension was added to produce a final Hemagglutination Assay Using Tn MAbs suspension of 0.5-1.0 x 10' erythrocytes/ml. After The specificity of two of the mAbs was observed to be mixing, suspensions were incubated for 3 h a t room different in hemagglutination assays than in ELISA temperature. Titer was defined as the reciprocal of the assays (Table 1). All four mAbs showed high hemaghighest dilution of mAb that produced at least a hemagglutination pattern as defined by Stavitsky glutination titers against Tn + cells, and ETnl.05 and ETnl.06 showed no reaction with normal erythrocytes. [491. However, both ETnl.01 and ETnl.02 showed high heFlow cytometric analyses were performed using the magglutination titers against normal red cells containLLNL Dual Beam Sorter [19] equipped with a n argon ing GPA(N). Trypsin digestion of normal cells removes ion laser and a n argon ion-pumped dye laser. Smalla n important part of the epitope that is recognized by angle light scatter from the 488 nm laser beam was ETnl.01 and ETnl.02. No hemagglutination of trypused to discriminate between debris and erythrocytes. sinized normal MM or NN cells was seen with any of Fluorescein fluorescence was detected after 1 W excithe mAbs. Trypsinized T n + cells (from donor BA) tation a t 488 nm through a 514 nm narrow band emisshowed a decrease in hemagglutination titer of at least sion filter. Texas Red fluorescence was measured after 20-fold for all four mAbs, but a n insufficient supply of 0.5 W excitation a t 590 nm through a 630 nm narrow Tn + blood prevented a precise determination of titer. band emission filter. Fluorescence intensities were Neuraminidase treatment also resulted in changes of standardized using fluorescent microspheres (fluoreshemagglutination with these mAbs. Using normal NN cein, 1.7 pm green spheres; Texas Red 1.6 pm red spheres; Polysciences, Warrington, PA). For precise red cells, the titer for ETnl.01 and ETnl.02 decreased fluorescence intensity measurements, linear amplifi- by 20-fold. As with untreated erythrocytes, neuraminication was used. To display fluorescence distributions dase-treated MM red cells were unreactive with any of of cell populations of widely varying intensity, loga- the mAbs. No change in titer of any of the mAbs was rithmic amplification, with a dynamic range of approx- evident using neuraminidase-treated Tn + erythrocytes from donor BA. The apparent hemagglutination imately three decades, was used. To search for T n + cells in normal bloods, two-color cross-reactivity of ETnl.01 with normal erythrocytes flow analyses were performed using formalin-fixed was confirmed by other laboratories in studies for a erythrocytes labeled with a Tn-specific mAb recent international workshop [as]. (ETnl.01+ GAMp-F) and a GPA(M) specific antibody Immunofluorescent Labeling and (6A7-B + Av-TR). A 1O:l mixture of normal MN cells Flow Cytometry and T n + cells (donor NH) was used to determine the MAb labeling and flow cytometric analyses were region in the flow histogram where T n + variants should be detected, i.e., a level of binding of ETnl.01 performed to quantify the specificity of binding of the characteristic of Tn + cells and no binding of6A7. Cells Tn mAbs to individual erythrocytes from Tn syndrome from five normal donors of MN blood type were ana- donors. Flow analysis of the Tn mAb-labeled red cells lyzed using the same instrumental settings. For each from four of the Tn donors (CL, DOS, NH, and BA) donor, 1-4 x lo6 cells were analyzed and objects with revealed two discrete populations of erythrocytes, the fluorescence intensities within the Tn + region were majority being brightly labeled and the minority barely detectable (Fig. 2A). Labeling with the GPA(M)sorted onto microscope slides.

*

BIGBEE ET AL.

264 I

I

I

I

I

2.o

1 .o

I

I

I

I

I

ETnl.06

Loglo [mAb] dilution FIG.1. Binding of Tn mAbs to Tn ( O ) , trypsinized Tn (01, and normal

(m) erythrocytes in the whole red cell ELISA. Purified mAbs were initially diluted 1:100,then serially fourfold. The activity at each dilution is plotted as the value of the linear rate of change with time

of the absorbance at 405 nm resulting from hydrolysis of the colorimetric substrate p-nitrophenyl phosphate by the bound alkaline phosphatase-conjugated goat anti-mouse immunoglobulin.

TABLE 1 Hemagglutination Titers of T n and Normal Erythrocytes W i t h Tn Specific MAbs MAb ETnl.01 ETnl.02 ETnl.05 ETnl.06

TniMM 200.000

TniNN 200.000

4,500 10,000

4,500 10,000

9o;ooo

9o;ooo

Red cell DhenotvDe NN 26.000 451000