Functional significance of the activation-associated receptors CD25 and CD69 on human NK-cells and NK-like T-cells

Immunobiol. 207, 85 ± 93 (2003) ¹ Urban & Fischer Verlag http: // www.urbanfischer.de/journals/immunobiol

Functional significance of the activation-associated receptors CD25 and CD69 on human NK-cells and NK-like T-cells Johannes Clausen, Birgit Vergeiner, Martina Enk, Andreas L. Petzer, G¸nther Gastl, Eberhard Gunsilius Tumor Biology & Angiogenesis Laboratory, Division of Hematology & Oncology, University Hospital, Innsbruck, Austria Received: November 14, 2001 ¥ Accepted: July 29, 2002

Abstract The application of autologous ex-vivo expanded cytotoxic lymphocytes to cancer patients may help to control minimal residual disease. However, the number of effector cells and the resulting antitumoral activity that can be generated in vitro are remarkably variable. Thus, we separately assessed the proliferative and cytotoxic potential of CD56‡CD3 natural killer (NK) and CD56‡CD3‡ T-cells in relation to their expression of CD25, CD69, and CD16 in vitro. Two-week lymphocyte cultures from peripheral blood (n ˆ 51) and from G-CSFmobilized progenitor cell harvests (n ˆ 11) were performed repeatedly from 14 women with breast cancer throughout conventional- and high-dose chemotherapy. A large proportion of CD25‡ cells on day 7 of the culture predicted high expandability (r ˆ 0.69, p < 0.00001), while elevated expression of CD69 predicted augmented cytotoxicity (r ˆ 0.72; p ˆ 0.00001) and low expandability (r ˆ 0.69, p < 0.00001). CD25 and CD69 expression were inversely correlated (r ˆ 0.8, p < 0.0001). CD16 expression was not suited to predict functional properties. Additionally, NK-cells were sorted by FACS according to CD25 versus CD69 expression. In a [3H]thymidine incorporation assay the CD25‡ NK-cell fraction exhibited a higher proliferation rate than did the CD69‡ fraction in all of three experiments. Together, our data suggest that CD69 is a useful marker for cytotoxic activity of NK cells, whereas proliferative potential is indicated by CD25 expression. These findings should help optimizing the ex-vivo generation of large numbers of cytotoxic effector cells for immunotherapy. Abbreviations: Ab ˆ antibody; CIK ˆ cytokine induced killer; IL-2 ˆ interleukin-2; FITC ˆ fluorescein isothiocyanate; HDCT ˆ high-dose chemotherapy; NK ˆ natural killer; PBL ˆ peripheral blood lymphocytes; PE ˆ Phycoerythrin

Introduction The application of ex-vivo expanded autologous CD56‡ cytotoxic lymphocytes is a feasible immunotherapeutic approach to control or even eradicate

residual tumor cells after chemotherapy in cancer patients (Rosenberg et al., 1985, 1993). Spontaneous cytotoxicity against tumor cells can be mediated in vitro and in vivo by both, activated CD56‡CD3 natural killer (NK) cells and

Corresponding author: Eberhard Gunsilius, MD, Division of Hematology & Oncology, Innsbruck University Hospital, Anichstr. 35, 6020 Innsbruck, Austria. Phone: ‡ 43 51 25 04 86 68; Fax: ‡ 43 51 25 04 86 74; E-mail: [email protected]

0171-2985/03/207/02-085 $ 15.00/0

86

Johannes Clausen et al.

CD56‡CD3‡ T-cells. CD56‡CD3 NK-cells constitute a minor, yet highly potent effector cell population among interleukin-2 (IL-2) activated killer (LAK)-cells (Rosenberg et al., 1985). CD56‡ T-cells, a population that may be related to the NK receptor protein 1 (NKR-P1)‡, Va24‡ natural killer T (NKT) cells (Bix & Locksley, 1995; Exley et al., 1997), can be preferentially expanded with interferon-g and anti-CD3 Ab in addition to IL-2. These cells are then referred to as cytokine induced killer (CIK) cells (Lu & Negrin, 1994; Schmidt-Wolf et al., 1993). Among the surface antigens that are involved in CD56‡ cell activation and target-cell binding are adhesion molecules, HLA class I specific receptors, the highaffinity IL-2 receptor (IL-2R) including the IL-2R achain CD25 (Siegel et al., 1987; Caligiuri et al., 1990; Nagler et al., 1990), and the low-affinity receptor for the Fc portion of immunoglobulin-G (FcgRIII, CD16) (Anegon et al., 1988; Hoshino et al., 1991). IgG coated target-cells may trigger antibody-dependent cytotoxicity of NK-cells via CD16 (Trinchieri, 1989). Another cell surface molecule present on activated NK-cells that triggers their spontaneous cytotoxicity is the very early antigen CD69 (Moretta et al., 1991). An increase in CD69 expression is accompanied by an enhanced cytotoxicity against various target-cells (Lanier et al., 1988). Yet, so far the natural ligand for CD69 remains unknown. In cancer patients the cytolytic capacity of CD56‡ lymphocytes against tumor target cells may be diminished by tumor-induced immunosuppression and by treatment procedures such as radiation, chemotherapy or surgery (Beitsch et al., 1994; Sewell et al., 1993; McCulloch & MacIntyre, 1993). In women with breast cancer we have previously shown considerable fluctuations of the in vitro expandability of cytotoxic CD56‡ cells throughout conventional and high-dose chemotherapy, being mostly impaired during peripheral blood progenitor cell (PBPC) mobilization and after autografting (Clausen et al., 2001). Thus, the identification of cell surface markers on CD56‡ cells which are indicative for their proliferative and cytotoxic capacity might be helpful for establishing optimal culture conditions to achieve maximal numbers of CD56‡ cells with high spontaneous cytotoxicity for adoptive immunotherapy. Therefore we analyzed the relationship of CD16, CD25 and CD69 expression on NK-cells and CD56‡ T-cells with their proliferative and cytotoxic potential in vitro.

Materials and methods Patients and treatment After obtaining informed consent, blood samples were collected from 14 females (median age, 49 years; range, 29 ± 57 years), with locally advanced (stage II/III, n ˆ 11) or metastatic breast cancer (stage IV, n ˆ 3) undergoing chemotherapy (CT). An intensified chemotherapy regimen consisting of epirubicin and paclitaxel was administered every 21 days for a total of 3 cycles (in 11 patients later receiving HDCT) or 6 cycles (in 3 patients not receiving HDCT). For PBPC mobilization, G-CSF (Filgrastim, Amgen, Thousand Oaks, CA, USA) was administered to the HDCT patients (10 mg/kg/day s.c.) starting day 5 after the 2nd CTcycle. PBPC were collected by leukapheresis (Fenwal CS3000‡, Baxter, Munich, Germany) and were cryopreserved in liquid nitrogen. After the 3rd CT cycle, HDCT (cyclophosphamide, thiotepa, and carboplatin, CTC) (Antman et al., 1992) was administered and the autologous PBPC were reinfused. The treatment protocols have been approved by the local ethics committee. Blood sampling 10 ml samples of heparinized peripheral blood were obtained the day before each CT cycle and repeatedly after PBPCT from day ‡ 12 through day ‡ 275. Aliquots of the PBPC harvests (250 ml) were processed immediately after collection (n ˆ 5) or after cryopreservation (n ˆ 6). Mononuclear cells were enriched by light density (1,077 g/l) centrifugation (Lymphoprep, Nycomed, Oslo, Norway). Monocytes were depleted by plastic adherence for 1 hour at 37 8C. No significant difference between fresh and cryopreserved samples was found regarding antigen expression, expandability and cytotoxicity (data not shown). Lymphocyte cultures Lymphocytes (monocyte-depleted mononuclear cells) were cultured for 14 days in T-25 tissue culture flasks in the presence of 1 000 U/ml recombinant human interleukin-2 (rhIL-2, Laevosan, Linz, Austria). On day 7 the cells were enriched for CD56‡ cells by MACS, as previously described (Clausen et al., 2001). The culture medium consisted of DMEM (PAA, Linz, Austria) plus Ham's F12 (Gibco, Paisley, Scotland) supplemented with 10% pooled human serum (provided by the local blood bank), 2 mM L-glutamine (Gibco), 0.05 mM 2-

CD25 and CD69 predict distinct NK-cell functions

mercaptoethanol (Merck, Darmstadt, Germany) and antibiotics. Flow cytometry 2  105 MNC were incubated at 4 8C for 30 min with saturating concentrations of FITC-conjugated anti-human CD56 monoclonal antibody (CD56FITC, Becton Dickinson/BD, San Jose¬, CA, USA), CD3-RPE-Cy5 (DAKO, Glostrup, Denmark) and either CD16-PE (n ˆ 45), CD25-PE (n ˆ 42), CD69PE (n ˆ 40), or irrelevant IgG1a-PE as negative control (all purchased from BD). To exclude debris and residual erythrocytes, freshly isolated MNCs were additionally labeled with a combination of CD56-FITC, CD3-PE (BD) and CD45-RPE-Cy5 (DAKO). On day 7 and 14, cells were gated according to side scatter (SSC) and forward-scatter (FSC) properties. FITC-conjugated IgG2b (BD) was used as control for CD56 expression. CD56‡CD3 NK-cells and CD56‡CD3‡ T-cells, as well as the CD56bright and CD56dim NK-cell subpopulations were separately analyzed for CD16, CD25 and CD69 expression. The threshold of the CD56bright population of unstimulated NK-cells was set one log above the fluorescence-intensity of the control antibody. Since the CD56dim and CD56bright NK population were not clearly distinguishable on day 7 and 14, the percentage of CD56bright cells was evaluated only for unstimulated NK-cells. Cytotoxicity assays MCF-7 and Daudi cells were used as target-cells to assess the cytotoxic activity of the cultured CD56‡ lymphocytes. The flow cytometric cytotoxicity assay was performed as originally described (Hatam et al., 1994), with some modifications (Clausen et al., 2001). Spontaneous cell lysis (Daudi, 9.4  0.7%; MCF-7, 8.7  0.8; mean  SEM), as determined in the absence of effector cells, was subtracted from total lysis to calculate the amount of specific lysis.

87

mAbs, as well as anti-CD3 (APC) for the exclusion of residual T-cells, were purchased from BD. FACS sorted CD25‡ cells and CD69‡ cells were seeded in triplicates each at 20 000 cells into 96-well-plates, and were cultured for another 5 days with IL-2 (1 000 U/ml). [3H]thymidine was added for the last 18 hours (2 mCi per well; Amersham, Arlington Heights, IL). Cells were harvested using a semiautomated device, and [3H]thymidine uptake expressed in counts per minute (cpm) was measured in a liquid scintillation counter (Beckmann LS 1801, Galway, Ireland). Data analysis and statistics The expansion rates of NK-cells and CD56‡ T-cells were calculated on the basis of total cell counts and the relative proportion of each subset as determined by flow cytometry. If not otherwise stated, results are expressed as mean  SEM. Correlation coefficients were calculated by Pearson's linear correlation statistics.

Results Proportion of CD56‡CD3 NK-cells and CD56‡ T-cells during culture During the first week of culture the percentage of CD56‡CD3 NK-cells increased from 11.3  0.9% to 34.5  2.1% (Fig. 1). The CD56‡ T-cell fraction increased from 4.1  0.4% to 11.2  1.1%. Prior to activation in vitro a clearly distinguishable subpo-

Proliferation assay MNC from healthy donors (n ˆ 3) were cultured as described above in the presence of 1 000 U/ml IL-2. On day 7 NK-cells were purified by depletion of CD3‡ T-cells and subsequent positive selection of CD56‡ cells by MACS as described above. Purified NK-cells were then proceeded to separation of CD25‡ cells (CD69‡/ ) and CD69‡ cells (CD25 ) by fluorescence activated cell sorting (FACSVantage, Becton Dickinson). CD25 (FITC) and CD69 (PE)

Fig. 1. Proportion (%) of CD56‡CD3 NK-cells and CD56‡ T-cells before culture, before and after immunomagnetical enrichment on day 7, and on day 14 of culture. Results are expressed as mean  SEM from 51 experiments initiated with peripheral blood lymphocytes.

88

Johannes Clausen et al.

pulation, containing 12.8  1.4% of the NK-cells, expressed CD56 at a high density (CD56bright). While the mean fluorescence intensity for CD56 increased during culture, a sharp discrimination between the CD56bright and the CD56dim cells became unfeasible. The CD56‡ cell purity was 94  1% after immunomagnetical selection on day 7 (72  2% NK-cells and 22  2% CD56‡ T-cells) and 82  2% on day 14 (62  3% NK-cells and 20  2% CD56‡ Tcells). Expression of CD25, CD69 and CD16 on CD56‡CD3 NK-cells and CD56‡ T-cells In PBL cultures the CD56‡CD3- NK-cells expressed CD25 at 1.1  0.2% on d0, 25.4  4.0% (d7), and 4.9  1.0% (d14). CD25 expression on CD56‡ Tcells was similar on day 0 (1.3  0.3%), but higher than on NK-cells on day 7 (44.9  5.5%) and day 14 (18.0  2.9%; Fig. 2A). To determine whether the observed decrease in CD25 expression after day 7 resulted from the CD56‡ cell purification step, PBL cultures were performed in duplicates, each with and without a CD56‡ selection step on day 7 (n ˆ 5). On day 14, the CD56‡ lymphocytes in the non-purified cultures expressed CD25 consistently higher than the corresponding purified CD56‡ cells in all of five experiments (30  9% versus 14  2%, p ˆ 0.05). During culture, CD69 expression on PBL-derived NK-cells increased from 11.1  2.3% (d0) to 92.4  1.2% (d7) and 92.2  2.1% (d14). On the CD56‡ Tcells, CD69 expression increased from 8.4  1.4 (d0) to 75.8  3.5% (d7), and 70.4  3.4% (d14; figure 2B). On NK-cells derived from PBPC collections (G-PBL), CD69 expression was elevated even prior to culture, compared with steady-state PBL-derived NK-cells (36.5  4.7 vs. 11.1  2.3; p < 0.001). CD16 expression on NK-cells in PBL cultures declined from 83.7  3.1% (d0) to 41.6  4.6% (d7), and subsequently increased to 61.2  3.8% on day 14 (Fig. 2C), thus showing an expression pattern inverse to that of CD25. CD16 expression on NK-cells from G-PBL was not different from that in PBL cultures. Expression of CD25, CD69 and CD16 on CD56bright and CD56dim NK-cells Prior to culture (d0), CD25 expression was higher on the CD56bright than on the CD56dim NK-cell population (p ˆ 0.001; Fig. 3A). However, on day 7 the CD56dim cells expressed CD25 at a higher proportion than the CD56bright cells (p < 0.0001; figure 3A). Vice versa, CD69 and CD16 were initially higher expressed on the CD56dim NK-cells than on the

Fig. 2. Expression of CD25 (A), CD69 (B), and CD16 (C) on CD56‡CD3 NK-cells and CD56‡ T-cells during 14-day cultures. The phenotype on day 7 was determined after immunomagnetical selection of CD56‡ cells. CD16 expression is exclusively shown for NK-cells. Results are expressed as mean  SEM.

CD56bright NK-cells (p < 0.0001; figure 3B and 3C). After activation with rhIL-2 for 7 days, there was no significant difference between the CD56bright and the CD56dim NK-cells regarding the expression of both markers.

CD25 and CD69 predict distinct NK-cell functions

89

Table 1. Prediction of cytotoxicity against MCF-7

Increased cytotoxicity * Proportion of NK-cells * CD69 expression (day 7) * CD69 expression on NK-cells (day 7) Decreased cytotoxicity * Proportion of CD56‡ T-cells * CD25 expression (day 7)

r ˆ 0.52 *** r ˆ 0.66 **** r ˆ 0.78 **** rˆ rˆ

0.54 **** 0.57 ***

If not otherwise indicated, the expression of surface markers was analyzed on the total CD56‡ population. Significance levels, **p < 0.01, ***p < 0.001, ****p < 0.0001

(specific lysis, 37  3% and 26  2%, respectively, at an E : T ratio of 8 : 1; data not shown for E : T ratios 1 : 1, 2 : 1, and 4 : 1). Table 1 shows the parameters with the strongest association with either increased or decreased cytotoxicity against MCF-7. Importantly, day-14 cytotoxicity could be estimated seven days in advance by CD69 expression on NK-cells (Fig. 4A). A trend towards an inverse correlation was found for the CD56‡ cell expansion rate and cytotoxic activity against MCF-7 (r ˆ 0.31; p < 0.05). The association of CD69 expression with cytotoxic activity against MCF-7 cells also proved significant for Daudi cell lysis, however in some cases a lower r-value was found (data not shown). Prediction of the in vitro expandability of NK-cells and CD56‡ T-cells by the expression of CD25, CD69, and CD16

Fig. 3. Expression of CD25 (A), CD69 (B), and CD16 (C) on the CD56dim (closed symbols) and the CD56bright (open symbols) NK-cell subpopulation. Each symbol represents the surface marker expression (% positive) for day 0 and day 7 in one individual experiment. Median values are shown as horizontal bars.

Prediction of the cytotoxic activity by the expression of CD25, CD69, and CD16 on NK-cells and CD56‡ Tcells MCF-7 cells and Daudi cells were lysed in a dosedependent manner by the expanded CD56‡ cells

During 14-day cultures with IL-2 the NK-cells and the CD56‡ T-cells expanded 70  15-fold and 64  13-fold, respectively. The expandability varied considerably throughout the course of chemotherapy, as previously reported in detail (Clausen et al., 2001). The parameters with the strongest association with either high or low expandability of CD56‡ cells are listed in Table 2. As shown, a high CD56‡ cell expandability was most accurately predicted by high CD25 expression (Fig. 4B) and low CD69 expression, respectively, on day 7. At this time CD25 and CD69 were found to be inversely correlated (r ˆ 0.80; p < 0.0001; figure 4C). The expansion of NK-cells correlated not only with their own CD25 expression, but also with CD25 expression on the CD56‡ T-cells (Table 2). Similarly, CD69 on the CD56‡ T-cells was inversely correlated with the expandability of NK-cells (Table 2). A weak association was also found for the percentage of CD56bright NK-cells on day 0 with the expansion of the CD56‡ cells during the first week of culture (p < 0.01). Furthermore, expression of CD16 on the NKcells at day 0 was associated with reduced NK-cell

90

Johannes Clausen et al.

expandability (Table 2), predominance of the CD56dim NK-cell phenotype (r ˆ 0.55, p < 0.001), and low CD25 expression (r ˆ 0.7; p < 0.001). In contrast to the predictive value of CD25 and CD69, the CD56‡ cell expansion rate during the first week of culture was not predictive for the expandability during the second week (r< 0.1, p ˆ 0.5). Proliferation of CD25‡ versus CD69‡ NK-cells CD25‡ NK-cells and CD69‡ NK-cells obtained after one week of culture were separately assessed for their proliferative potential. In all of three experiments the CD25‡ NK-cells exhibited a higher proliferative activity than did the CD69‡ NK-cell fraction (31 332  19 041 cpm vs. 12 501  8 401 cpm per 20 000 cells).

Discussion The efficacy of adoptive immunotherapy for cancer using ex-vivo expanded and activated autologous cytotoxic lymphocytes has been examined in animal models as well as in clinical trials (Rosenberg et al., 1985, 1993). The antitumoral activity of the transferred lymphocytes is dependent on the number of infused cells and their activation status (Basse et al., 1991). Moreover, the proliferative capacity of the transferred cells was shown to be essential for their efficacy in vivo, since an irradiation-mediated proliferation arrest abrogated their antitumoral activity in vivo despite preserving their cytotoxicity in vitro (Lu & Negrin, 1994). Thus, it is likely that for clinical purposes, both the cytotoxic potential of the effector cells and their capacity to proliferate in vivo after retransfusion are required for a sufficient antitumoral effect. Consequently, the culture conditions for the generation of cytotoxic lymphocytes should be targeted towards both, the enhancement of cytotoxic activity and sustained proliferation. In this study, elevated cytotoxic activity of the in vitro expanded CD56‡ cells was found to be associated with the expression of CD69 on a high percentage of cells. Determination of CD69 expression on NK-cells allowed the prediction of their cytotoxic activity even one week in advance, render3 Fig. 4. (A) Correlation of CD69 expansion on NK-cells (day 7) with MCF-7 lysis on day 14 (r ˆ 0.78; p < 0.0001); (B) Correlation of CD25 expression (day 7) with the expandability of CD56‡ cells (r ˆ 0.067; p < 0.0001). (C) Inverse correlation between CD25 and CD69 expression on CD56‡ cells on day 7 (r ˆ 0.80; p < 0.0001).

CD25 and CD69 predict distinct NK-cell functions

91

Table 2. Prediction of CD56‡ cell expandability in vitro

Increased expandability * CD25 expression (day 7)a * CD25 expression on CD56‡ T-cells (day 7)b * CD25 expression (day 7)c Deceased expandability * CD69 expression (day 7)a * CD69 expression on CD56‡ T-cells (day 7)a * CD16 expression on NK-cells (day 0)b * CD69 expression on NK-cells (day 7)c * CD69 expression (day 7)c

r ˆ 0.67 **** r ˆ 0.52 ** r ˆ 0.65 **** rˆ rˆ rˆ rˆ rˆ

0.69 0.63 0.55 0.59 0.76

**** *** ** *** ****

If not otherwise indicated, the expression of surface markers was analyzed on the total CD56‡ population. For significance levels, see legend of table 1: a referred to total CD56‡ cell expansion, b NK-cell expansion, c CD56‡ T-cell expansion.

ing this marker particularly valuable for the monitoring of LAK cultures. The level of cytotoxicity was also dependent on a high proportion of NK-cells in the culture, and was inversely correlated with the percentage of CD56‡ T-cells. This finding is in line with the superior cytotoxicity of activated CD56‡CD3 cells compared to CD56‡CD3‡ cells on a per-cell basis in vitro (Schmidt-Wolf et al., 1993). CD25 expression upon activation turned out to be a suitable parameter for estimating the expandability of CD56‡ cells. Conversely, at the same time a high expression of CD69 on CD56‡ cells was predictive for a low expandability of these cells. The resulting presumption, i.e. that CD25 but not CD69 is a suitable indicator of proliferative potential in NK-cells, was confirmed by a side-by-side comparison of separated CD25‡ and CD69‡ NKcells in a proliferation assay. These findings are particularly important since both CD25 and CD69 have been generally regarded as activation markers without discrimination of their significance for proliferation and cytotoxic activity, respectively. Indeed, previous studies demonstrated an association between CD69 and cytotoxic activity of NKcells, whereas the importance of CD69 for proliferation and cell survival remained unclear (Moretta et al., 1991; Lanier et al., 1988; Jewett & Bonavida, 1995; Lebow et al., 1993; Craston et al., 1997; Borrego et al., 1999; Lauzurica et al., 2000). Another intriguing finding is the clearly inverse correlation between CD25 and CD69 expression upon stimulation with IL-2. This finding is reflected by the inverse relationship between the proliferative capacity of CD56‡ cells and their cytotoxic activity. Furthermore, our observations point to an interdependence between NK-cells and CD56‡ T-cells in vitro, i.e. that an activated status (indicated by CD69 expression) of one cell population may affect the expansion of the other, while proliferative activity of

one population (indicated by CD25 expression) does not impair the expandability of the other. The expression pattern of CD25 on CD56bright versus CD56dim NK-cells was found to change substantially upon activation with IL-2. The resulting phenotype, CD56dimCD25‡CD16dim, is characteristic for activated plastic-adherent NK-cells (ANK-cells) which posses high proliferative and cytotoxic activity (Vujanovic et al., 1993). Our findings are based on CD56‡ cells from breast cancer patients obtained at various times throughout a complex treatment course. The impact of the sampling time on the cytotoxic and proliferative properties was previously reported in detail (Clausen et al., 2001). Still, it is not clear whether conclusions can be drawn as to NK-cell functions in other diseases. We found an increased expression of CD69 on NK-cells from PBPC harvests, compared with those from steady state hematopoiesis. This may reflect activation of the NK-cells by the mobilization and/or the apheresis procedure. However, since we and others have previously observed a severely impaired proliferative and cytotoxic capacity of NK-cells from PBPC collections (Clausen et al., 2001; Miller et al., 1997; Rondelli et al., 1998), the upregulation of CD69 by these cells is obviously not associated with functional superiority. In conclusion, our findings suggest a discrimination between CD25 and CD69 as activation markers on cytotoxic CD56‡ lymphocytes, i.e., CD25 should be viewed as an indicator for proliferative potential, and CD69 as a marker for cytotoxic activity. The ability to distinguish between the proliferative and cytotoxic potential of cultured effector cells should help optimizing the culture conditions to achieve a large number of fully active cytotoxic CD56‡ lymphocytes. Acknowledgements. We express our gratitude to Dr. H. Glassl for her excellent assistance in the flow-cytometry

92

Johannes Clausen et al.

analyses, and to Drs. R. Stauder, C. Marth, and R. Margreiter for their contribution to the treatment of the patients described in this article. J. C. received financial support from AMGEN, Austria; B. V. was supported by a grant from the Hans and Blanca Moser Foundation, Austria. This work was further supported by the Tiroler Verein zur Fˆrderung der Krebsforschung an der Universit‰tsklinik Innsbruck.

References Anegon I, Cuturi MC, Trinchieri G and Perussia B. 1988. Interaction of Fc receptor (CD16) ligands induces transcription of interleukin 2 receptor (CD25) and lymphokine genes and expression of their products in human natural killer cells. J. Exp. Med. 167: 452 ± 472. Antman K, Ayash L, Elias A, Wheeler C, Hunt M, Eder JP, Teicher BA, Critchlow J, Bibbo J, Schnipper LE et al. 1992. A phase II study of high-dose cyclophosphamide, thiotepa, and carboplatin with autologous marrow support in women with measurable advanced breast cancer responding to standard-dose therapy. J. Clin. Oncol. 10: 102 ± 110. Basse P, Herberman RB, Nannmark U, Johansson BR, Hokland M, Wasserman K and Goldfarb RH. 1991. Accumulation of adoptively transferred adherent, lymphokine-activated killer cells in murine metastases. J. Exp. Med. 174: 479 ± 488. Beitsch P, Lotzova E, Hortobagyi G and Pollock R. 1994. Natural immunity in breast cancer patients during neoadjuvant chemotherapy and after surgery. Surg. Oncol. 3: 211 ± 219. Bix M and Locksley RM. 1995. Natural T-cells: cells that co-express NKRP-1 and TCR. J. Immunol. 155: 1020 ± 1022. Borrego F, Robertson M, Ritz J, Pena J and Solana R. 1999. CD69 is a stimulatory receptor for natural killer cell and its cytotoxic effect is blocked by CD94 inhibitory receptor. Immunology. 97: 159 ± 165. Caligiuri MA, Zmuidzinas A, Manley TJ, Levine H, Smith KA and Ritz J. 1990. Functional consequences of interleukin 2 receptor expression on resting human lymphocytes. Identification of a novel natural killer cell subset with high affinity receptors. J. Exp. Med. 171: 1509 ± 1526. Clausen J, Petzer AL, Vergeiner B, Enk M, Stauder R, Gastl G and Gunsilius E. 2001. Optimal timing for the collection and in vitro expansion of cytotoxic CD56‡ lymphocytes from patients undergoing autologous peripheral blood stem cell transplantation. J. Hematother. Stem Cell Res. 10: 513 ± 521. Craston R, Koh M, McDermott A, Ray N, Prentice HG and Lowdell MW. 1997. Temporal dynamics of CD69 expression on lymphoid cells. J. Immunol. Methods. 209: 37 ± 45. Exley M, Garcia J, Balk SP and Porcelli S. 1997. Requirements for CD1d recognition by human

invariant Va24‡ CD4 CD8 T-cells. J. Exp. Med. 186: 109 ± 120. Hatam L, Schuval S and Bonagura VR. 1994. Flow cytometric analysis of natural killer cell function as a clinical assay. Cytometry. 16: 59 ± 68. Hoshino S, Oshimi K, Teramura M and Mizoguchi H. 1991. Activation via the CD3 and CD16 pathway mediates interleukin-2-dependent autocrine proliferation of granular lymphocytes in patients with granular lymphocyte proliferative disorders. Blood. 78: 3232 ± 3240. Jewett A and Bonavida B. 1995. Interferon-alpha activates cytotoxic function but inhibits interleukin-2-mediated proliferation and tumor necrosis factor-alpha secretion by immature human natural killer cells. J. Clin. Immunol. 15: 35 ± 44. Lanier LL, Buck DW, Rhodes L, Ding A, Evans E, Barney C and Phillips JH. 1988. Interleukin 2 activation of natural killer cells rapidly induces the expression and phosphorylation of the Leu-23 activation antigen. J. Exp. Med. 167: 1572 ± 1585. Lauzurica P, Sancho D, Torres M, Albella B, Marazuela M, Merino T, Bueren JA, Martinez A and SanchezMadrid F. 2000. Phenotypic and functional characteristics of hematopoietic cell lineages in CD69deficient mice. Blood. 95: 2312 ± 2320. Lebow LT, Jewett A and Bonavida B. 1993. Killer cell recruitment and renewal capacity of purified cytolytic and noncytolytic human peripheral blood natural killer cell subsets. J. Immunol. 150: 320 ± 329. Lu PH and Negrin RS. 1994. A novel population of expanded human CD3‡CD56‡ cells derived from Tcells with potent in vivo antitumor activity in mice with severe combined immunodeficiency. J. Immunol. 153: 1687 ± 1696. McCulloch PG and MacIntyre A. 1993. Effects of surgery on the generation of lymphokine-activated killer cells in patients with breast cancer. Br. J. Surg. 80: 1005 ± 1007. Miller JS, Prosper F and McCullar V. 1997. Natural killer (NK) cells are functionally abnormal and NKcell progenitors are diminished in granulocyte colony-stimulating factor-mobilized peripheral blood progenitor cell collections. Blood. 90: 3098 ± 3105. Moretta A, Poggi A, Pende D, Tripodi G, Orengo AM, Pella N, Augugliaro R, Bottino C, Ciccone E and Moretta L. 1991. CD69-mediated pathway of lymphocyte activation: anti-CD69 monoclonal antibodies trigger the cytolytic activity of different lymphoid effector cells with the exception of cytolytic T lymphocytes expressing T-cell receptor alpha/ beta. J. Exp. Med. 174: 1393 ± 1398. Nagler A, Lanier LL and Phillips JH. 1990. Constitutive expression of high affinity interleukin 2 receptors on human CD16-natural killer cells in vivo. J. Exp. Med. 171: 1527 ± 1533. Rondelli D, Raspadori D, Anasetti C, Bandini G, Re F, Arpinati M, Stanzani M, Morelli A, Baccini C, Zaccaria A, Lemoli RM and Tura S. 1998. Alloanti-

CD25 and CD69 predict distinct NK-cell functions gen presenting capacity, T-cell alloreactivity and NK function of G-CSF-mobilized peripheral blood cells. Bone Marrow Transplant. 22: 631 ± 937. Rosenberg SA, Lotze MT, Muul LM, Leitman S, Chang AE, Ettinghausen SE, Matory YL, Skibber JM, E. Shiloni E, Vetto JT et al. 1985. Observations on the systemic administration of autologous lymphokineactivated killer cells and recombinant interleukin-2 to patients with metastatic cancer. N. Engl. J. Med. 313: 1485 ± 1492. Rosenberg SA, Lotze MT, Yang JC, Topalian SL, Chang AE, Schwartzentruber DJ, Aebersold P, Leitman S, Linehan WM, Seipp CA et al. 1993. Prospective randomized trial of high-dose interleukin-2 alone or in conjunction with lymphokine-activated killer cells for the treatment of patients with advanced cancer. J. Natl. Cancer Inst. 85: 622 ± 632. Schmidt-Wolf IG, Lefterova P, Mehta BA, Fernandez LP, Huhn D, Blume KG, Weissman IL and Negrin RS. 1993. Phenotypic characterization and identification

93

of effector cells involved in tumor cell recognition of cytokine-induced killer cells. Exp. Hematol. 21: 1673 ± 1679. Sewell HF, Halbert CF, Robins RA, Galvin A, Chan S and Blamey RW. 1993. Chemotherapy-induced differential changes in lymphocyte subsets and natural-killer-cell function in patients with advanced breast cancer. Int. J. Cancer. 55: 735 ± 738. Siegel JP, Sharon M, Smith PL and Leonard WJ. 1987. The IL-2 receptor beta chain (p70): role in mediating signals for LAK, NK, and proliferative activities. Science. 238: 75 ± 78. Trinchieri G. 1989. Biology of natural killer cells. Adv. Immunol. 47: 187 ± 376. Vujanovic NL, Rabinowich H, Lee YJ, Jost L, Herberman RB, and Whiteside TL. 1993. Distinct phenotypic and functional characteristics of human natural killer cells obtained by rapid interleukin 2induced adherence to plastic. Cell. Immunol. 151: 133 ± 157.