Cell Death and Differentiation (1998) 5, 231 ± 240 1998 Stockton Press All rights reserved 13509047/98 $12.00 Selective killing of CD8+ cells with a `memory' phenotype (CD62Llo) by the N-acetyl-D-galactosamine-speci®c lectin from Viscum album L 1,3 1 2 Arndt BuÈssing , Gerburg M. Stein and Uwe PfuÈller Introduction 1 Krebsforschung Herdecke, Department of Applied Immunology, Communal There is emerging evidence that lectins are dynamic Hospital Herdecke, D-58313 Herdecke. Tel: ++49-2330-623246; fax: ++49- contributors to tumour cell recognition, cell adhesion, signal 2330-624003; e-mail: [email protected] transduction across membranes, mitogenic stimulation, and 2 Institute of Phytochemistry, University Witten/Herdecke, D-58448 Witten, augmentation of host immune defence (reviewed by Mody et Germany 3 al, 1995). However, lectins were also recognised to induce corresponding author: Dr A BuÈssing, Krebsforschung Herdecke, Department of Applied Immunology, Communal Hospital Herdecke, D-58313 Herdecke, apoptosis in several tumour cell lines, epithelial cells, Germany macrophages, and human lymphocytes (Griffiths et al, 1987; Kim et al, 1993; Janssen et al, 1993; Khan and Waring, 1993; Received 1.6.97; revised 1.9.97; accepted 14.10.97 Kulkarni and McCulloch, 1995; Perillo et al, 1995; BuÈssing et Edited by M.L. Gougeon al, 1996a). A number of plant proteins such as the toxic lectins from Ricinus communis, Abrus precatorus,andViscum album Abstract have been identified that catalytically damage eukaryotic ribosomes making them unable to bind the elongation As reported previously by our group, among the toxic factor 2, and consequently unable to perform the elongation proteins from Viscum album L. only the mistletoe lectins step of protein synthesis (Stirpe et al, 1992). These (MLs) induce the apoptotic killing pathway in human `ribosome-inactivating proteins' (RIPs) possess carbohy- lymphocytes. Although one may expect a homogenous drate-binding B chains linked by hydrophobic bonds and distribution of carbohydrate domains on cell surface disulphide bridges to the catalytic A chain (type 2 and type receptors for the carbohydrate binding B chains of the 4 RIPs). The lectin domains of these RIPs can bind to any toxic protein, the sensitivity of cells to these B chains appropriate carbohydrate domain on cell surface receptors, obviously differ. Here we report a selective killing of CD8+ enabling the protein to enter the cell by receptor-mediated CD62Llo cells from healthy individuals by the galNAc- endocytosis (Stirpe et al, 1982, 1992; Endo et al, 1988). specific ML III (and RCA , which binds to gal and galNAc), Subsequently, the catalytic A chain of the misletoe lectins 60 (MLs) inhibit protein synthesis (Olsnes et al, 1989; Endo et while the gal-specific ML I was less effective. This selective al, 1988; Stirpe et al, 1992), and the cells undergo killing is not sufficiently explained by protein synthesis apoptosis (reviewed in BuÈssing, 1996b). It appears that inhibition alone, since this subset was not affected by other both, glycoproteins and -lipids on the cell surface may act ribosome inhibiting proteins such as the lectin from as binding sites for toxins. Ricinus communis (RCA120), lectin from Abrus precatorus The lectin chains of the various RIPs differ in their (APA), abrin A, and inhibitors of RNA, DNA and/or protein cellular interactions, as do the enzymic chains. This is synthesis such as actinomycin D, mitomycin C, and suggested by the different lesions each toxin causes in cycloheximide. We conclude that CD8+ cells with `memory' animals (Stirpe et al, 1992), with ricin at high concentra- phenotype (CD62Llo) are more sensitive to the ML III- tions damaging primarily Kupffer and other macrophagic mediated killing than their CD8+ CD62Lhi counterparts, cells, whereas modeccin and volkensin affect both CD4+ T cells, and CD19+ B cells. These cells probably parenchymal and non-parenchymal liver cells. Since the type 1 RIPs are lacking a lectin subunit, these single chain express a distinct receptor with galNAc domains that is + RIPs are less toxic than type 2 RIPs; however, they are missing or not active on CD8 cells with a `naive' highly toxic to some cells, for instance macrophages and phenotype. trophoblasts, possibly due to their high pinocytic activity (Stirpe et al, 1992). Thus, although one may expect a + Keywords: CD8 cells; cytotoxicity; apoptosis; cell surface homogenous distribution of `receptors' for the lectin B molecules; mistletoe lectins; ricin; abrin; protein synthesis inhibition chains on the cell surface, the sensitivity of cells to the lectin-mediated cytotoxicity obviously differ. An unexpected Abbreviations: RIPs: ribosome-inactivating proteins; ML: finding reported recently by our group was related to the mistletoe lectins; RCA: Ricinus communis agglutinin; APA: cytotoxicity mediated by the N-acetyl-D-galactosamine Abrus precatorus agglutinin; Mit C: mitomycin C; Act D: (galNAc)-specific ML III, which was more effective against actinomycin D; CHX: cycloheximide; galNAc: N-acetyl-D- CD8+ T cells than CD19+ BcellsandCD4+T cells (BuÈssing galactosamine; b-gal: b-galactose; WBCC: whole blood cell et al, 1997). Here we report a predominant killing of CD8+ cultures; IL: interleukin; TNF-a: tumour necrosis factor alpha CD62Llo cells by the galNAc-specific ML III, while the b- Killing of CD8+ `memory' cells by ML III ABuÈssing et al 232 galactose (b-gal)-specific ML I was less effective. We and lymphocytes from WBCC incubated with increasing conclude that CD8+ cells with `memory' phenotype concentrations of ML III. (CD62Ll0) are more sensitive to the ML III-mediated killing In lymphocytes from WBCC, ML III at final concentra- than CD4+ cells and CD19+ cells and suggest that they tions of 50 and 100 ng/ml decreased the amount of CD3+ may harbour `receptors' with galNAc domains not ex- T cells (Table 2), while CD19+ B cells were less sensitive pressed or less active on the surface of their CD8+ to the ML-mediated killing, as reported previously CD62Lhi counterparts. (BuÈssing et al, 1997). At a final concentration of 10 ng/ mL ML III, CD8+ cells declined, while the number of CD4+ T cells remained almost unchanged. Predominantly the Results CD287 CD8+ subset declined (Table 2). Within the CD8+ cells, the CD8lo CD38+ subset was eliminated, i.e. CD3+ Killing of cultured human lymphocytes by ML III CD8lo T cells and CD37 CD8lo natural killer cells (data The MLs induce an apoptotic cell death in cultured not shown). However, by increasing the concentration of lymphocytes (BuÈssing et al, 1996a). Within 72 h, the ML III, also the CD4+ T cells and the CD28+ CD8+ subset number of Annexin-V+ lymphocytes with low PI fluores- died. cence (PIlo) increased by the addition of ML III, however, Similar to lymphocytes from WBCC, in Ficoll-isolated even at a final concentration of 3 ng/ml (Table 1). The lymphocytes the number of CD8+ cells decreased by the number of Annexin-V+ PI7 apoptotic cells and Annexin-V+ addition of ML III, while the relative number of CD4+ T cells PIhi necrotic cells did not significantly change by the addition and CD19+ B cells increased at final concentrations of 3, 6 of ML III, or protein and RNA synthesis inhibitors such as and 10 ng/ml ML III (Table 3). At 30 and 50 ng/ml ML III, cycloheximide (CHX) and actinomycin D (Act D) that served also the number of CD4+ cells declined, while the relative as a positive control (Table 1). As shown in Figure 1, within number of CD19+ B increased. Addition of CHX at a final 6 h of incubation with ML III at 10 ng/ml, no apoptotic cells concentration of 10 mg/ml decreased the relative number of were detected by flow cytometry as compared to controls, CD19+ B cells, but did not affect the CD8+ subset, while Act i.e. the cells did not bind Annexin-V and did not stain for D killed CD19+ B cells and CD8+ cells at 1 mg/ml (Table 3). mitochondrial membrane protein Apo2.7 which is expressed These results indicate that the CD8+ cells are more in cells undergoing apoptosis (Zhang et al, 1996), while the sensitive to the ML III-mediated killing than the CD4+ T number of both, apoptotic and necrotic cells increased cells and CD19+ B cells. within 24 h. The cells undergoing apoptosis became permeable to PI within 48 h (Annexin-V+ PIlo) (Figure 1). Decrease of CD8+ cells with `memory' phenotype However, within 48 h the number of lymphocytes that stain lo for mitochondrial membrane protein Apo2.7 went beyond (CD62L ) by ML III the number of cells that bind Annexin-V, indicating that Using three-colour immunofluorescence staining, we ob- much more cells received a `death signal' than actually served that the CD8+ cells from WBCC dying after application change membrane configuration. of 10 ng/ml ML III were predominantly of CD287 CD62Llo phenotype (Figure 2), while in Ficoll-isolated lymphocytes, also the CD28+ CD62Llo subset decreased. However, after Killing of defined lymphocyte subsets by ML III CD8 enrichment using magnetic beads, no significant We wondered whether lymphocyte subsets may differ in numbers of CD8+ CD28+ CD62Llo cells were detected; thus, regard of sensitivity to the ML III-mediated killing and only the CD287 CD62Llo subset was affected by the toxic analyzed the surviving lymphocytes by flow cytometry. Since protein (Figure 2). Also in cultured WBCC from six HIV ML III binds to granulocytes, monocytes and erythrocytes (but infected individuals, addition of ML III at 10 ng/ml resulted in a shows no blood group specificity and no hemolysis but significant decline of CD8+ CD287 CD62Llo cells, while the agglutination), and thus the overall killing capacity towards the CD8+ cells with CD62Lhi phenotype were not affected by the lymphocytes from whole blood cell cultures (WBCC) may toxic proteins (A BuÈssing, C SchnuÈrer, U PfuÈller, unpublished decrease, we investigated both, Ficoll-isolated lymphocytes results).