Spontaneous apoptosis in neutrophils is

Spontaneous apoptosis in neutrophils is associated with downregulation of HLA Class I and is prevented by ligation of Class I Guido Frumento, Luciano Ottonello,* Maria Bertolotto,* Silvia Franchello, Giovanni Melioli,† Franco Dallegri* Immunogenetics Laboratory and †Flow Cytometry Unit, National Institute for Cancer Research, Genoa, Italy; and *Department of Internal Medicine, University of Genoa, Genoa, Italy

Abstract: In many types of cells, ligation of human leukocyte antigens (HLA) Class I molecules with specific mAbs results in the transduction of signals that trigger different cell functions. We have investigated the effects of Class I ligation in human neutrophils. After several hours in culture, neutrophils split spontaneously into two subpopulations, one with normal and the other with reduced levels of Class I. The latter subpopulation displayed high binding capacity for Annexin V, showed a hypodiploid peak, electrophoretic DNA fragmentation, and morphological features of apoptotic cells. The addition of drugs known to delay apoptosis (GM-CSF or cAMP) resulted in a reduction of Class I modulation. Furthermore, ligation of surface Class I with F(abⴕ)2 fragments of the anti-Class I mAb W6/32 resulted in a delay in the progression of apoptosis. These data indicate that this surface Class I molecule is a marker of agerelated apoptosis, and the ligation of these molecules results in the transduction of a signal that inhibits apoptosis. Thus, the downregulation of HLA Class I molecules in aging neutrophils prevents their halting the apoptotic process. J. Leukoc. Biol. 68: 873– 880; 2000. Key Words: human 䡠 neutrophils 䡠 age-related 䡠 apoptosis

fibroblasts [5, 12], have been shown to be capable of transducing signals as well. Similar behavior of HLA Class I molecules in granulocytes has not been investigated so far. Neutrophilic granulocytes play a first-line-of-defense role in the inflammation process, being recruited from circulation in the early phases of infection. Neutrophils have the potential to cause tissue injury by means of degradative enzymes and cytotoxic products. In particular, the lack of timely removal of these cells from the inflamed site can result in tissue damage or progression to a chronic, inflammatory state. Thus, it may be hypothesized that not only their functions but also their fate should be carefully regulated. For these reasons, we studied whether ligation of Class I molecules could result in the modification of neutrophil survival. During the investigation, we observed that after several hours in culture, neutrophils downregulate the membrane expression of HLA Class I molecules. Infection with Herpes simplex virus [13], treatment of melanoma cells with interleukin (IL)-10 [14], growth of epithelial cells at confluency [15], and other experimental conditions cause the downregulation of HLA Class I to occur. Nevertheless, to our knowledge, a similar phenomenon has never been described in neutrophils. Notably, neutrophils aging in culture undergo marked reduction in CD16 expression, a phenomenon strictly associated with the onset of spontaneous apoptosis [16, 17]. All these considerations prompted us to check whether Class I downregulation could be regarded also as a marker of apoptosis in aging neutrophils and whether ligation of Class I molecules could affect the progression of cells to programmed cell death.

INTRODUCTION Human leukocyte antigens (HLA) Class I molecules play a central role in the presentation of endogenously derived peptides to cytotoxic T cells (CTLs). Recent data also suggest that intracellular signals can be delivered by ligation of HLA Class I molecules at the cell membrane level [1] by monoclonal antibodies (mAbs). Following this interaction, different responses can be elicited, ranging from the induction to the inhibition of activation and proliferation [2–12], depending on the cell type [2–5], the Class I determinant recognized [2, 6, 7], the degree of costimulation [8], and the type of stimulus [9, 10]. Although these experiments were performed primarily on lymphoid cells, HLA Class I molecules expressed by other cell types, such as macrophages, mast cells, endothelial cells, and

MATERIALS AND METHODS Reagents RPMI 1640 with 25 mM HEPES (Irvine Scientific, Santa Ana, CA), supplemented with 10% fetal calf serum (FCS; Irvine), was used as culture medium. Mono-Poly Resolving Medium and Ficoll-Hypaque were from ICN Pharmaceuticals (Costa Mesa, CA). The W6/32 hybridoma, secreting an immunoglob-

Correspondence: Guido Frumento, M.D., Immunogenetics Laboratory, National Cancer Institute c/o Advanced Biotechnology Center A2, Largo Rosanna Benzi 10, 16132 Genova, Italy. E-mail: [email protected] Received January 4, 2000; revised June 19, 2000; accepted June 20, 2000.

Journal of Leukocyte Biology Volume 68, December 2000 873

ulin G (IgG)2a mAb directed against an HLA Class I monomorphic epitope, was purchased from the American Type Culture Collection (ATCC; Rockville, MD). The mAb was purified by affinity chromatography using Protein A Sepharose CL-4B (Pharmacia Biotech, Uppsala, Sweden). The fluorescein isothiocyanate (FITC)-conjugated W6/32 mAb was obtained by incubating the mAb for 2 h at room temperature in slow rotation at a concentration of 1 mg/mL in 0.5 M carbonate buffer in the presence of 50 ␮g/mL FITC (Sigma Chemical Co., St. Louis, MO). Afterward, the solution was passed through a Sephadex G-50 (Pharmacia) column, and the fractions containing the first fluorescent peak were pooled and dialyzed overnight in phosphate-buffered saline (PBS). The following mouse mAbs were also used as primary antibodies in indirect immunofluorescence: anti-human CD15 (Boehringer Mannheim, Mannheim, Germany), anti-human CD16/Fc␥RIII (Coulter-Immunotech, Westbrook, MA), and anti-human CD32/Fc␥RII (Coulter-Immunotech). The mouse anti-human CD8 Leu-2b mAb (Becton Dickinson, San Jose, CA) was used in some tests as an irrelevant first antibody. The secondary antibodies, anti-mouse IgG1/PE (phycoerythrin) and anti-IgG2a/PE, were purchased from Southern Biotechnology (Birmingham, AL), and the anti-mouse IgG ⫹ IgM/FITC was from Pierce (Rockford, IL). All antibodies were titrated before use. To obtain the F(ab⬘)2 fragments, the W6/32 IgG was digested by pepsin. Briefly, the antibody was incubated overnight at a concentration of 1 mg/mL in 100 mM citrate buffer, pH 3.5, in the presence of 700 ␮g/mL pepsin (Calbiochem-Novabiochem, San Diego, CA). Afterward, neutrality was restored by adding one-tenth the volume of Tris-HCl buffer (pH 8.8), the solution was passed through a Protein A Sepharose CL-4B column six times and centrifuged in Centricon 30 concentrators (Amicon, Beverly, MA), and the high molecularweight fraction was recovered and dialyzed in PBS. An aliquot of the sample was run in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) without finding undigested IgGs. The F(ab⬘)2 fragments of the antiCD16 3G8 mAb were purchased from Medarex (Annandale, NJ). Human recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF) was purchased from Genzyme Co. (Cambridge, MA). The Annexin V/FITC kit was purchased from Boehringer Ingelheim (Heidelberg, Germany). Propidium iodide, dibutyryl adenosine 3⬘,5⬘-cyclic phosphate (cAMP), and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma. Proteinase K and RNAse A were from Boehringer Mannheim.

Cell isolation and culture Heparinized (heparin 10 units/mL) venous blood was obtained from healthy male volunteers after informed consent. Anticoagulated blood (5 mL) was layered on a discontinuous density gradient of Ficoll-Hypaque (5 mL) over Mono-Poly Resolving Medium (3 mL) in 15 mL conical-bottom polystyrene tubes (Falcon, Becton Dickinson, Franklin Lakes, NJ). The tubes were centrifuged at 500 g for 60 min. Thereafter, the neutrophil fraction was recovered and layered on 1 mL Mono-Poly Resolving Medium for subsequent centrifugation (30 min). The resulting neutrophils were washed three times and checked for purity by May Grunwald-Giemsa staining and viability by ethidium bromide-fluorescein diacetate test. Cells were then resuspended in the above-mentioned medium at a concentration of 2 ⫻ 106/mL, and 0.5 mL of the cell suspension was incubated in tissue culture tubes (17⫻100 mm, Falcon) at 37°C in a 5% CO2 atmosphere for the length of time indicated in the text. Some experiments were carried out in the presence of GM-CSF (final concentration, 10 ng/mL), cAMP (1 mM), or PMA (1 and 5 ng/mL). The effect of Class I ligation was tested by culturing freshly isolated neutrophils for 24 h in culture medium supplemented with 20 ␮g/mL solution of W6/32 F(ab⬘)2 fragments or 3G8 F(ab⬘)2 fragments. PBS (50 ␮L) was added to the control samples.

Flow cytometric analysis For direct immunofluorescence, 2 ⫻ 107 cells were incubated for 30 min at 4°C with the appropriate concentration of FITC-conjugated W6/32 in ice-cold PBS plus 5% mouse serum, washed twice, and resuspended for cell sorting. The medium used for the other samples was ice-cold PBS supplemented with 3% FCS and 0.1% sodium azide. For indirect immunofluorescence, 5 ⫻ 105 cells were washed, and the pellet was incubated for 30 min at 4°C with appropriate concentrations of primary antibody, washed twice, incubated for 30 min at 4°C with the appropriate concentration of the secondary antibody, washed twice, and resuspended for flow cytometric analysis. When indicated,

874

Journal of Leukocyte Biology Volume 68, December 2000

cells were incubated with an irrelevant first antibody followed by the abovementioned secondary antibody from Pierce. For dual labeling of CD16 and Class I, neutrophils were incubated with 3G8, washed, incubated with antimouse IgG1/PE, washed, incubated with FITC-conjugated W6/32, washed again, and resuspended for flow cytometric analysis. For dual labeling with use of Annexin V/FITC, an indirect immunofluorescence with W6/32 followed by anti-mouse IgG2a/PE was initially performed. Then Annexin V/FITC was added with minor changes to the manufacturer’s instructions. Briefly, cells were resuspended in 100 ␮l isotonic binding buffer, and 3 ␮l Annexin V/FITC was added. After 15 min, cells were washed and resuspended in the buffer for flow cytometric analysis. The method of Nicoletti et al. [18] was used, with minor changes, to evaluate propidium iodide incorporation. Briefly, cells were washed and resuspended in 0.5 mL PBS, the cell suspension was added dropwise to 4.5 mL ice-cold 80% ethanol while vortexing, and kept at ⫺20°C for 24 h. Afterward, cells were washed twice, propidium iodide was added to a final concentration of 10 ␮g/mL, and the sample was analyzed by flow cytometry after incubation overnight. Flow cytometry analysis was performed on an EPICS XL flow cytometer (Coulter, Hialeah, FL). Briefly, living granulocytes were gated on the basis of physical properties (forward vs. side-light scatter), and at least 2000 living cells were analyzed per sample.

Cell sorting For cell sorting, neutrophils were cultured for 24 h to achieve a bimodal distribution of HLA Class I molecule expression. Cells were then labeled with FITC-conjugated W6/32 to detect granulocytes with normal expression of HLA Class I and those with low HLA Class I expression. Cell sorting was performed on an EPICS Elite cell sorter (Coulter), following identification of the correct drop delay using the “dot matrix” procedure. A mean frequency of 1500 –2500 cells/sec was used, and the coincident granulocytes were excluded. During and after sorting, cells were kept at 4°C. The purity of the two sorted populations (either high HLA or low HLA Class I-expressing granulocytes) was immediately checked on the above-mentioned flow cytometer. The sorted subpopulations were further used for the evaluation of propidium iodide incorporation, DNA fragmentation, and cell morphology.

DNA electrophoresis Neutrophils (2⫻106) from the two sorted populations were washed twice in PBS. Pellets were incubated overnight at 37°C in 0.5 mL lysis buffer (1 mM EDTA, 10 mM Tris-HCl, pH 8.0, 200 ␮g/mL proteinase K, 4% SDS, 200 mM NaCl). After adding 15 ␮l of 10 mg/mL heat-treated RNAse A, each sample was incubated for 1 h at 37°C. After addition of 15 ␮l of 10 mg/mL proteinase K and another 1 h incubation at 37°C, samples were treated twice with 1 vol phenol:chloroform:isoamyl alcohol (25:24:1) and once with 1 vol chloroform: isoamyl alcohol (24:1). After overnight precipitation with 0.1 vol 5 M NaCl and 2 vol ethanol at ⫺20°C, pellets were air-dried and resuspended in TBE buffer (89 mM Tris, 89 mM borate, 2 mM EDTA). DNA concentration was then evaluated by spectrophotometric (260 nm) analysis, samples (5 ␮g DNA/lane) were added to loading buffer (2.5% Ficoll, 0.04% bromophenol blue, 0.04% xylene cyanole), heated at 60°C for 10 min, and separated at 10 V in 1.5% agarose gel containing 1% ethidium bromide.

Light-microscopy assessment of neutrophil apoptosis Cells from the two sorted populations were cytocentrifuged, fixed, and stained with May-Gru¨nwald-Giemsa. Thereafter, slides were read blindly by two independent observers using oil-immersion light microscopic examination of at least 500 cells/slide (1000⫻ magnification). Cells showing apoptotic morphology were identified, according to the following criteria: cell shrinkage, nuclear condensation and fragmentation, and plasma membrane ruffling and blebbing.

RESULTS Aging neutrophils downregulate Class I Freshly isolated neutrophils, investigated for Class I expression by indirect immunofluorescence, showed the unimodal http://www.jleukbio.org

Fig. 1. Time course of Class I expression in aging neutrophils. Indirect immunofluorescence was performed using the anti-HLA Class I MoAb W6/32 on freshly isolated neutrophils (A) and after 8 h (B), 24 h (C), and 36 h (D), in culture. Dotted curves show fluorescent intensity of indirect immunofluorescence with an irrelevant first MoAb. A representative experiment of the two performed is shown. After 24 h in culture, 51% of neutrophils expressed low levels of Class I in this experiment, and 69% did so after 36 h in culture.

distribution typical of a homogeneous cell population (Fig. 1A). However, when cells were tested during culture, a clear bimodal distribution of surface Class I expression was detected. The subpopulation with low expression increased over time (Fig. 1 B, C, and D). Many controls were carried out to elucidate the mechanisms involved in this phenomenon. First, it could not be related to contamination by cells other than neutrophils, because the purity of the population was consistently ⱖ97%, and viability was ⬎98% at the beginning of the culture. The possibility that necrotic cells or cellular debris could account for Class I downregulation was ruled out also by gating and analyzing only cells having the physical properties (in terms of size and lateral lightscatter) of living neutrophils. Furthermore, the hypothesis that modulation of HLA Class I expression could result from binding the Fc fragment of the W6/32 mAb with the Fc receptors on the membrane was evaluated. Indirect immunofluorescence assays were thus performed with an irrelevant IgG2a mAb. These experiments showed that nonspecific binding was low and did not change during culture (dotted lines, Fig. 1 A and C). The hypothesis that the modulation of Class I expression could be related to neutrophil activation during in vitro culture was verified also. In fact, activation with PMA at 1 and 5 ng/mL did not alter the expression of Class I molecules on cell membranes (unpublished results). Thus, our results suggest that separation of cultured neutrophils

into two subpopulations, one with normal and one with reduced Class I expression, is an apparently spontaneous phenomenon.

Downregulation of Class I is related to downregulation of CD16 To evaluate the selectivity of the modulation of HLA Class I expression, together with Class I, a panel of surface molecules expressed on granulocytes was studied by indirect immunofluorescence. These included CD16, downregulated in apoptotic neutrophils [16, 17]; CD32, which does not change [16]; and CD15. The modulation of CD15 has never been investigated previously. At the beginning of culture, the four proteins were characterized by a unimodal distribution (unpublished results). Although the expression levels of CD15 and CD32 remained unchanged during culture, granulocytes analyzed for the expression of Class I and CD16 split into two subpopulations of normal or low-expressing cells (Fig. 2). After 24 h of incubation, the reduction in mean fluorescence of the low-expressing population was ⫺77.9 ⫾ 4.7% (mean⫾1 SD in seven different experiments) for HLA Class I molecules. In the same cell population, the reduction of CD16 expression was ⫺90.2 ⫾ 4.9%. Although CD16 was reduced more drastically than Class I, remarkably, the curve shapes for Class I and CD16 were similar (Fig. 2 A and C), thus suggesting that neutrophils downregulating Class I could be the very same cells that also downregulate CD16.

Frumento et al. Class I molecules and apoptosis in human neutrophils

875

Fig. 2. Effect of culturing on membrane antigen expression in human neutrophils. After 24 h in culture, neutrophils were tested by indirect immunofluorescence using an anti-Class I (A), anti-CD15 (B), anti-CD16 (C), and anti-CD32 (D) MoAb, respectively, as the first antibody. A representative experiment of the two performed is shown.

To verify this hypothesis, a double-staining immunofluorescence assay of these two molecules was performed on aged neutrophils. As shown in Figure 3, downregulation of Class I and CD16 occurs coincidentally in the same cells.

Fig. 3. Coincidence of Class I and CD16 downregulation in aged neutrophils. Dual labeling was performed on neutrophils cultured for 24 h by incubating the cells with 3G8 MoAb, followed by anti mouse IgG1/PE, and then by FITCconjugated W6/32. Histograms are plotted with PE fluorescence on the x axis and with FITC fluorescence on the y axis. Relative percentages of the gated populations are indicated. A representative experiment of the two performed is shown.

876


Neutrophils downregulating Class I are undergoing age-related apoptosis Aging neutrophils undergo spontaneous apoptosis [19] and concomitantly downregulate surface CD16 expression [16, 17]. Thus, CD16 can be considered a marker for apoptosis in neutrophils. Because CD16 is downregulated coincidentally with Class I, it can be hypothesized that low Class I expression in cultured neutrophils could be related to apoptosis. To evaluate this hypothesis, we measured the binding of Annexin V, which binds to phosphatidylserine molecules expressed in apoptotic cells on the outer leaflet of the membrane [20]. When neutrophils cultured for 24 h were examined for their Class I expression and Annexin V binding capacity by means of double-color flow cytometry, it was found that the two parameters were correlated inversely. Neutrophils avidly binding Annexin V barely expressed Class I, and cells that bound low levels of Annexin V expressed normal amounts of Class I (Fig. 4). A more sophisticated approach was then used to corroborate these findings. Aged neutrophils were separated by cell sorting into two subpopulations, one consisting mostly of cells expressing low levels of Class I and the other, containing mostly cells expressing normal levels of Class I (Fig. 5). Each subpopulation was then tested for apoptosis by using three different approaches, namely PI incorporation, DNA fragmentation, and morphological features. Although the latter subpopulation did not display evident signs of apoptosis, a significant amount of the cells in the former had a hypodiploid DNA content, showed discretely fragmented DNA, and exhibited vacuolization and nuclear condensation, suggesting that these cells were apoptotic neutrophils indeed (Fig. 5). http://www.jleukbio.org

alone or in the presence of F(ab⬘)2 fragments of the anti-Class I W6/32 mAb or F(ab⬘)2 fragments of the anti-CD16 3G8 mAb. After 24 h of incubation, apoptosis was evaluated in each sample by morphological criteria, the Annexin V binding technique, and the measurement of PI incorporation. In all three cases, neutrophils undergoing ligation of their surface Class I molecules showed a marked reduction in the percentage of apoptotic cells (Fig. 7). On the contrary, no prevention of apoptosis was observed in cells that underwent ligation of membrane CD16. These data suggest that some functional

Fig. 4. Time course of Class I expression and Annexin V binding in aging neutrophils. Dual labeling was performed on freshly isolated neutrophils (A), and after 18 (B) and 24 (C) h of culture. After indirect immunofluorescence with W6/32 and anti-mouse IgG2a/PE was performed, cells were incubated with FITC-conjugated Annexin V. Histograms in the left column are plotted with forward scatter on the x axis and with side scatter on the y axis. Histograms in the right column are plotted with FITC fluorescence on the x axis and with PE fluorescence on the y axis. The gating set for analyzing the cells in the right column is indicated in the left column. For each histogram, the relative percentage of the cells in the respective field is indicated in the right column. A representative experiment of the three performed is shown.

Prevention of age-related apoptosis also results in prevention of Class I downregulation Because age-related apoptosis in neutrophils can be delayed by incubating the cells with GM-CSF or cAMP [21, 22], we examined the possibility that these substances might affect the expression of Class I molecules also. After 24 h of incubation with 10 ng/mL GM-CSF, we found that, together with a reduction in the number of apoptotic cells, the amount of cells downregulating Class I was sharply reduced (Fig. 6). Similar results were obtained when 1 mM cAMP was used instead of GM-CSF, thus confirming the close association between the onset of apoptosis and the reduction of Class I expression.

Ligation of Class I delays age-related apoptosis Upon ligation, Class I molecules on lymphocytes transduce signals that eventually result in the induction of apoptosis [6, 23, 24]. Therefore, we investigated the effects of ligation of Class I molecules on the progression of age-related apoptosis in neutrophils. Freshly isolated cells were cultured in medium

Fig. 5. Evaluation of apoptosis in neutrophils expressing normal levels of Class I and low levels of Class I. After 24 h of culture, 2 ⫻ 107 neutrophils were labeled with FITC-conjugated W6/32 MoAb (upper panel). By means of a fluorescence-activated cell sorter, the cells were divided into two subpopulations of low expression of Class I (A) and normal expression of Class I (B). Each sorted sample was then divided into three aliquots for measuring apoptosis. Propidium iodide incorporation was evaluated after cold-ethanol fixation and overnight incubation with propidium. DNA fragmentation was checked by gel electrophoresis of DNA extracted from the two samples. Lanes 1 indicate the relative sample; lanes 2 indicate the molecular weight markers. Cell morphology was studied on cytocentrifuge preparations stained with May Grunwald-Giemsa. A representative experiment of the two performed is shown. Neutrophils (54%) showed hypodiploid DNA content, and 64% displayed morphological features of apoptosis in the sorted sample, expressing low levels of Class I (A), compared with 12% and 31%, respectively, in the sorted sample expressing normal levels of Class I.


877

choalveolar lavage from babies with airway inflammation [25]. Their number is increased in blood from patients with Systemic Lupus Erythematosus [26], although apoptotic neutrophils can be found in small amounts in the peripheral blood of normal subjects. Macrophages or semiprofessional phagocytes are involved directly in a rapid, specific, and nonflogistic recognition of aged neutrophils [27]. This in vivo apoptosis is mimicked in vitro by the spontaneous apoptosis displayed by neutrophils in culture [16, 17, 19]. In this study, using the experimental conditions that induce apoptosis, we observed a marked reduction in Class I expression. In our experiments, all basic conditions that could account for this finding, such as contamination by cells other than granulocytes, the presence of interfering necrotic cells or cellular debris, and binding to membrane Fc receptors, have been ruled out. Similarly, the hypothesis that surface Class I modulation could be strictly related to cell activation could not be responsible, because a powerful activator like PMA did not alter the membrane expression of Class I molecules. Because the subpopulation with reduced expression increases over the time, a mechanism involving the cells progressively could be responsible. As a matter of fact, our data suggest a close relationship between Class I downregulation and neutrophil senescence. The temporal changes in Class I expression closely resemble those shown by CD16 [16], a membrane molecule whose expression is reduced in aging neutrophils [16, 17]. Notably, in our experiments, CD16 is more downregulated than HLA Class I. The difference in the degree of downregulation, together with a marked individual variability in the time course of agerelated apoptosis, could account for the fact that Dransfield and

Fig. 6. Effect of drugs preventing apoptosis on Class I downregulation. Freshly isolated neutrophils were incubated for 24 h in the absence (A) or presence of 10 ng/mL GM-CSF (B) or 1 mM cAMP (C). Afterward, Class I expression was evaluated by indirect immunofluorescence with the W6/32 mAb, and the amount of apoptosis was measured by the Annexin/FITC technique. The percentage of apoptotic cells was 50.4% in A, 26.5% in B, and 25.8% in C. A representative experiment of three that were performed with GM-CSF and two that were performed with cAMP is shown.

signal is delivered through Class I molecules, preventing apoptosis in aging neutrophils.

DISCUSSION Neutrophilic granulocytes are short-lived cells endowed with highly dangerous histotoxic potential related to their ability to release oxidizing radicals and proteolytic enzymes. Not only their function, but also their disposal, is strictly regulated to prevent unwanted spreading of potentially harmful substances by dying neutrophils. To achieve this goal, neutrophils are constitutively programmed to undergo apoptosis. The removal of neutrophils is particularly relevant at the inflammation site. For example, apoptotic neutrophils are detectable in bron878


Fig. 7. Effect of surface Class I ligation on age-related apoptosis in neutrophils. Freshly isolated neutrophils were incubated for 24 h in medium alone (hatched bars), medium plus F(ab⬘)2 fragments of the 3G8 mAb (open bars), and medium plus F(ab⬘)2 fragments of the W6/32 mAb (solid bars). The percentages of living and apoptotic cells were measured in each sample by evaluating cell morphology, Annexin V/FITC binding, and propidium iodide incorporation. Values are expressed as % living cells/% apoptotic cells. Results shown are the mean determination ⫾ 1 SD from two separate experiments for morphology and propidium iodide and from four separate experiments for Annexin V. In the morphology and propidium iodide assays: medium alone vs. medium plus W6/32, p ⬍ 0.05; medium alone vs. medium plus 3G8, p ⬎ 0.05. In Annexin V assay: medium alone vs. medium plus W6/32, p ⬍ 0.01; medium alone vs. medium plus 3G8, p ⬎ 0.05. One-way analysis of variance (ANOVA) with Bonferroni multiple comparison post test was used.

http://www.jleukbio.org

coworkers [16] failed to find Class I downregulation in their system. CD16 and Class I are not the only surface molecules whose expression is modified in aging neutrophils. Actually, several changes occur at the membrane level in these cells, with some molecules being upregulated, some being downregulated, and some remaining unchanged [16, 17, 28, and present paper]. These findings suggest that reshuffling the membrane molecules in polymorphonuclear neutrophils (PMNs) on the way to death is not only a precisely regulated, but also an apparently specific, phenomenon. At present, a fine analysis of the reasons as to why this surface reorganization occurs is lacking. Thus, it may only be hypothesized that it occurs because of particular behaviors of the different molecules, probably related to distinctive functions. From our experimental data, initial conclusions can be drawn about the functions of CD16 and Class I only. Neutrophils could downregulate the expression of these two molecules to ignore the signals delivered through them perturbing their orderly progression to apoptosis. Apoptosis is crucial for the removal of aged granulocytes. Neutrophils that have started their cell death program exist in a state of functional isolation from the external milieu, displaying a global impairment in their functions, such as chemotaxis, secretion, and metabolic burst [19, 29]. This condition is retained for more than 24 h before the disintegration and release of their contents, a period long enough to allow macrophages to be recruited and then to phagocytose the apoptotic neutrophils [30, 31]. Thus, to enter and maintain this standby condition, neutrophils have to switch off all signals that can induce activation or prevent apoptosis. Downregulation of CD16 in aging neutrophils is probably directed at this purpose. Its ligation results in granule release, reorganization of actin microfilaments, and priming of Fc␥RII-mediated phagocytosis, events that are associated with neutrophil activation and effector cell function [32, 33]. Thus, it is reasonable that a marked reduction in CD16 expression may render the cells less sensitive to stimuli that trigger effector functions. In fact, removal by phospholipase C of greater than 80% of surface CD16, a threshold well above the reduction observed in aged neutrophils, abolishes chemotactic responses to formyl-Met-Leu-Phe (fMLP) [34]. A similar model could be used to explain the reduction in Class I expression on aged neutrophils. Indeed, Class I downregulation could render neutrophils less sensitive to signals preventing apoptosis delivered by this molecule. From this perspective, we have analyzed the relationship linking Class I to age-related apoptosis. As shown, Class I downregulation parallels apoptosis. Even more interestingly, conditions inhibiting apoptosis (such as GM-CSF and cAMP) inhibit downregulation also. To further confirm the close relationship between the two phenomena, the effect of HLA class I ligation on programmed cell death was also studied. For this purpose, we used the Fab fragment of the W6/32 mAb, which recognizes a monomorphic determinant on all human HLA-A, -B, and -C subclasses and has been used widely, either soluble or crosslinked, in experiments on Class I ligation [2–5, 7–11]. As shown, ligation of Class I molecules delayed neutrophil pro-

gression to apoptosis markedly. Similar evidence was not achieved by ligation of CD16 molecules. To our knowledge, the “antiapoptotic” effect of HLA Class I ligation represents the first evidence so far of signal transduction via Class I molecules in neutrophils. This effect was unexpected partially, and only speculative explanations can be proposed. At present, two natural ligands of HLA Class I molecule have been described. One is the TcR/CD8 complex, of which two ways for neutrophilic Class I ligation to occur should be considered. One could occur through the specific recognition of a nonself peptide, and the other is based on low-affinity binding to Class I molecules in the absence of the antigen. The latter does occur peripherally and probably plays, in lymphocytes, a role in the maintenance of T cell memory [35–38]. At present, our data do not allow us to arrive at any conclusions on the significance or relevance of the two mechanisms. The other natural ligand for Class I molecule is represented by Killer Receptors, either activating or inhibitory, expressed by natural killer (NK) cells [39]. No data are available, to our knowledge, on the functional relationship between NK cells and aged neutrophils. However, from a speculative point of view, an increased susceptibility of neutrophils having downregulated Class I to NK cell lysis should result in the spreading of harmful substances by neutrophils. Also of importance to consider is that Class I has been shown to be structurally and/or functionally associated with peptide hormone receptors, some of which are also expressed by neutrophils, such as receptors for IL-2, insulin, insulin-like growth factor, and glucagon [40 – 47]. Thus, Class I molecules could be involved in signal transduction by modifying or fine-tuning signaling via other cell-surface molecules [11, 48].

ACKNOWLEDGMENTS This work was supported by grants from Programma di Ricerca Scientifica di Interesse Nazionale 9706117821_002; Consiglio Nazionale delle Ricerche n. 01273.49 and Target Project on Biotechnology 1998; AIRC; and Ministero della Sanita` Ricerca Finalizzata 1998.

REFERENCES 1. Skov, S. (1998) Intracellular signal transduction mediated by ligation of MHC Class I molecules. Tissue Antigens 51, 215–223. 2. Geppert, T. D., Wacholtz, M. C., Davis, L. S., Lipsky, P. E. (1988) Activation of human T4 cells by cross-linking class I MHC molecules. J. Immunol. 140, 2155–2164. 3. Taylor, D. S., Nowell, P. C., Kornbluth, J. (1987) Anti-HLA Class I antibodies inhibit the T cell-independent proliferation of human B lymphocytes. J. Immunol. 139, 1792–1796. 4. Walle´n-O¨hman, M., Larrick, J. W., Carlsson, R., Borrebaeck, C. A. K. (1997) Ligation of MHC class I induces apoptosis in human pre-B cell lines, in promyelocytic cell lines, and in CD40-stimulated mature B cells. Int. Immunol. 9, 599 – 606. 5. Bian, H., Harris, P. E., Mulder, A., Reed, E. F. (1997) Anti-HLA antibody ligation to HLA Class I molecules expressed by endothelial cells stimulates tyrosine phosphorylation, inositol phosphate generation and proliferation. Hum. Immunol. 53, 90 –97. 6. Sambhara, S. R., Miller, R. G. (1991) Programmed cell death of T cells signaled by the T cell receptor and the ␣3 domain of Class I MHC. Science 252, 1424 –1427.


879

7. De Felice, M., Turco, M. C., Carandente Giarrusso, P., Corbo, L., Pizzano, R., Martinelli, V., Ferrone, S., Venuta, S. (1987) Differential regulatory role of monomorphic and polymorphic determinants of histocompatibility leukocyte antigen Class I antigens in monoclonal antibody OKT3-induced T cell proliferation. J. Immunol. 139, 2683–2689. 8. Wacholtz, M. C., Patel, S. S., Lipsky, P. E. (1989) Patterns of costimulation of T cell clones by cross-linking CD3, CD4/CD8, and Class I MHC molecules. J. Immunol. 142, 4201– 4212. 9. Dasgupta, J. D., Cemach, K., Dubey, D. P., Yunis, E. J., Amos, D. B. (1987) The role of class I histocompatibility antigens in the regulation of T cell activation. Proc. Natl. Acad. Sci. USA 84, 1094 –1098. 10. Schwab, R., Weksler, M.E., Russo, C. (1988) CD3 pathway of T-cell activation. II. Role of HLA-Class I molecules in early events. Cell. Immunol. 115, 310 –324. 11. Bregenholt, S., Ropke, M., Skov, S., Claesson, M. H. (1996) Ligation of MHC Class I molecules on pheripheral blood T lymphocytes induces new phenotypes and functions. J. Immunol. 157, 993–999. 12. Tscherning, T., Claesson, M. H. (1994) Signal transduction via Class I molecules in T cells. Scand. J. Immunol. 39, 117–121. 13. Hill, A. B., Barnett, B. C., McMichael, A. J., McGeoch, D. J. (1994) HLA Class I molecules are not transported to the cell surface in cells infected with herpes simplex virus types 1 and 2. J. Immunol. 152, 2736 –2741. 14. Matsuda, M., Salazar, F., Petersson, M., Masucci, G., Hansson, J., Pisa, P., Zhang, Q. J., Masucci, M. G., Kiessling, R. (1994) Interleukin 10 pretreatment protects target cells from tumor- and allo-specific cytotoxic T cells and downregulates HLA Class I expression. J. Exp. Med. 180, 2371–2376. 15. Feralzzo, G., Semino, C., Quartone, G., Pietra, G., Melioli, G. (1997) Adherent neoplastic cells grown at confluence downregulate HLA Class I expression and enhance their susceptibility to lysis mediated by natural killer cells. Tissue Antigens 50, 459 – 465. 16. Dransfield, I., Buckle, A. M., Savill, J. S., McDowall, A., Haslett, C., Hogg, N. (1994). Neutrophil apoptosis is associated with a reduction in CD16 (Fc␥RIII) expression. J.Immunol. 153, 1254 –1263. 17. Homburg, C. H. E., de Haas, M., von de Borne, A. E. G. K., Verhoeven, A. J., Reutelingsperger, C. P. M., Roos, D. (1995) Human neutrophils lose their surface Fc␥RIII and acquire annexin V binding sites during apoptosis in vitro. Blood 85, 532–540. 18. Nicoletti, I., Migliorati, G., Pagliacci, M. C., Grignani, F., Riccardi, C. (1991) A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J. Immunol. Methods 139, 271–279. 19. Whyte, M. K. B., Meagher, L. C., MacDermot, J., Haslett, C. (1993) Impairment of function in aging neutrophils is associated with apoptosis. J. Immunol. 150, 5124 –5134. 20. Koopman, G., Reutelingsperger, C. P. M., Kuijten, G. A. M., Keehnen, R. M. J., Pals, S. T., van Oers, M. H. J. (1994) Annexin V for flow cytometric detection of phosphatidylserine expression on B cells undergoing apoptosis. Blood 84, 1415–1420. 21. Brach, M. A., de Vos, S., Gruss, H. J., Herrmann, F. (1992) Prolongation of survival of human polymorphonuclear neutrophils by granulocyte-macrophage colony-stimulating factor is caused by inhibition of programmed cell death. Blood 80, 2920 –2924. 22. Ottonello, L., Gonella, R., Dapino, P., Sacchetti, C., Dallegri, F. (1998) Prostaglandin E2 inhibits apoptosis in human polymorphonuclear leukocytes: role of intracellular cyclic AMP levels. Exp. Hematol. 26, 895–902. 23. Skov, S., Bregenholt, S., Claesson, M. H. (1997) MHC Class I ligation of human T cells activates the ZAP70 and P56lck tyrosine kinases, leads to an alternative phenotype of the TCR/CD3 zeta chain, and induces apoptosis. J. Immunol. 158, 3189 –3196. 24. Woodle, E. S., Smith, D. M., Bluestone, J. A., Kirkman, W. M., Green, D. R., Skowronsky, E. W. (1997) Anti-human class I MHC antibodies induce apoptosis by a pathway that is distinct from the Fas antigenmediated pathway. J. Immunol. 158, 2156 –2164. 25. Grigg, J. M., Savill, J. S., Sarraf, C., Haslett, C., Silverman, M. (1991) Neutrophil apoptosis and clearance from neonatal lungs. Lancet 338, 720 –722. 26. Courtney, P. A., Crockard, A. D., Williamson, K., Irvine, A. E., Kennedy, R. J., Bell, A. L. (1999) Increased apoptotic peripheral blood neutrophils in systemic lupus erythematosus: relations with disease activity, antibodies to double strand DNA, and neutropenia. Ann. Rheum. Dis. 58, 309 –314. 27. Savill, J. (1997) Apoptosis in resolution of inflammation. J. Leukoc. Biol. 61, 375–380.

880


28. Dransfield, I., Stocks, C., Haslett C. (1995) Regulation of cell adhesion molecule expression and function associated with neutrophil apoptosis. Blood 85, 3264 –3273. 29. Narayanan, P. K., Ragheb, K., Lawler, G., Robinson, J. P. (1997) Defects in intracellular oxidative metabolism of neutrophils undergoing apoptosis. J. Leukoc. Biol. 61, 481– 488. 30. Savill, J. S., Wyllie, A. H., Henson, J. E., Walport, M. J., Henson, P. M., Haslett, C. (1989) Macrophage phagocytosis of aging neutrophils in inflammation. Programmed cell death in the neutrophil leads to its recognition by macrophages. J. Clin. Invest. 83, 865– 875. 31. Meszaros, A. J., Reichner, J. S., Albina, J. E. (1999) Macrophage phagocytosis of wound neutrophils. J. Leukoc. Biol. 65, 35– 42. 32. Huizinga, T. W. J., Dolman, K. M., van der Linden, N. J., Kleijer, M., Nuijens, J. H., von dem Borne, A. E. G. K., Roos, D. (1990) PI-linked FcRIII mediates exocytosis of neutrophil granule proteins but does not mediate initiation of the respiratory burst. J. Immunol. 144, 1432–1437. 33. Salmon, J. E., Brogle, N. L., Edberg, J. C., Kimberley, R. P. (1991) Fc␥ receptor III induces actin polimerization in human neutrophils and primes phagocytosis mediated by Fc␥ receptor II. J. Immunol. 146, 997–1004. 34. Kew, R. R., Grimaldi, C. M., Furie, M. B., Fleit, H. B. (1992) Human neutrophil Fc␥RIIIB and formyl peptide receptors are functionally linked during formyl-methionyl-leucyl-phenylalanine-induced chemotaxis. J. Immunol. 149, 989-997. 35. Markiewicz, M. A., Girao, C., Opferman, J. T., Sun, J., Hu, Q., Agulnik, A. A., Bishop, C. E., Thompson, J. B., Ashton-Rickardt, P. G. (1998) Long term T cell memory requires the surface expression of self-peptide/major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 95, 3065–3070. 36. Kirberg, J., Berns, A., von Boehmer, H. (1997) Peripheral T cell survival requires continual ligation of the T cell receptor to major histocompatibility complex-encoded molecules. J. Exp. Med. 186, 1269 –1275. 37. Sandberg, J. K., Karre, K., Glas, R. (1999) Recognition of the major histocompatibility complex restriction element modulates CD8(⫹) T cell specificity and compensates for loss of T cell receptor contacts with the specific peptide. J. Exp. Med. 189, 883– 893. 38. Tanchot, C., Lemonnier, F. A., Perarnau, B., Freitas, A. A., Rocha, B. (1997) Differential requirements for survival and proliferation of CD8⫹ naive or memory T cells. Science 276, 2057–2062. 39. Moretta, A., Biassoni, R., Bottino, C., Moretta, L. (2000) Surface receptors delivering opposite signals regulate the function of human NK cells. Semin. Immunol. 12, 129 –138. 40. Hansen, N. Q., Tscherning, T., Claesson, M. H. (1991) T-cell activation. IV. Evidence for a functional linkage between MHC class I, interleukin-2 receptor, and interleukin-4 receptor molecules. Cytokine 3, 35– 41. 41. Giraud, D., Paquet, M. E., Paquin, R., Beaulieau, A. D. (1996) Differential effects of interleukin 15 (IL-15) and IL-2 on human neutrophils: modulation of phagocytosis, cytoskeleton rearrangement, gene expression, and apoptosis by IL-15. Blood 88, 3176 –3184. 42. Due, C., Simonsen, M., Olsson, L. (1986) The major histocompatibility complex class I heavy chain as a structural subunit of human cell membrane insulin receptor: implications for the range of biological functions of histocompatibility antigens. Proc. Natl. Acad. Sci. USA 83, 6007– 6011. 43. Spagnoli, A., Spadoni, G. L., Sesti, G., Del Principe, D., Germani, D., Boscherini, B. (1995) Effect of insulin on hydrogen peroxide production by human polymorphonuclear leukocytes. Studies with monoclonal anti-insulin receptor antibodies, and an agonist and an inhibitor of protein kinase C. Horm. Res. (Basel) 43, 286 –293. 44. Centrella, M., McCarthy, T. L., Canalis, E. (1989) Beta 2-microglobulin enhances insulin-like growth factor I receptor levels and synthesis in bone cell cultures. J. Biol. Chem. 264, 18268 –18271. 45. Fu, Y. K., Arkins, S., Wang, B. S., Kelley, K. W. (1991) A novel role for growth hormone and insuline-like growth factor I. Priming neutrophils for superoxide anion secretion. J. Immunol. 146, 1602–1608. 46. Lafuse, W., Edidin, M. (1980) Influence of the mouse major histocompatibility complex, H-2, on liver adenylate cyclase activity and on glucagon binding to liver cell membranes. Biochemistry 19, 49 –54. 47. Al-essa, L., Niwa, M., Kobayashi, M., Nozaky, M., Tsurumi, K. (1993) Glucagon modulates superoxide generation in human polymorphonuclear leukocytes. Life Sci. 53, 1439 –1445. 48. Olsson, L., Goldstein, A., Stagsted, J. (1994) Regulation of receptor internalization by the major histocompatibility complex class I molecule. Proc. Natl. Acad. Sci. USA 91, 9086 –9090.

http://www.jleukbio.org