engagement of the leukocyte-associated ig-like receptor-1 induces programmed cell death and prevents...

0014-2980/00/1010-2751$17.50+.50/0© WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000

Engagement of the leukocyte-associated Ig-likereceptor-1 induces programmed cell death andprevents NF- ‹ B nuclear translocation in humanmyeloid leukemias

Alessandro Poggi1, Fabio Pellegatta2, Biagio E. Leone3, Lorenzo Moretta1, 4 andM. Raffaella Zocchi5

1 Laboratory of Immunology, National Institute for Cancer Research, Genoa, Italy2 Laboratory of Cardiovascular Pathophysiology, Department of Cardiology, Scientific Institute

San Raffaele, Milan, Italy3 Department of Pathology, Scientific Institute San Raffaele, Milan, Italy4 Department of Experimental Medicine, University of Genoa, Genoa, Italy5 Laboratory of Tumor Immunology, Department of Internal Medicine, Scientific Institute San

Raffaele, Milan, Italy

Leukocyte-associated Ig-like receptor-1 (LAIR-1) is a surface molecule that functions as aninhibitory receptor on natural killer cells, T lymphocytes and monocytes. Here, we provideevidence that occupancy of LAIR-1 on human myelomonocytic leukemic cell lines inhibitsproliferation and leads to programmed cell death (PCD), evaluated by propidium iodidestaining and transmission electron microscopy. Interestingly, PCD elicited via LAIR-1 wasnot blocked by different caspase inhibitors, at variance with apoptosis induced via CD95/Fas, which was prevented by the caspase-1 and caspase-8 specific inhibitors. In addition,we show that the p65 subunit of the nuclear factor ‹ B (NF- ‹ B), constitutively expressed inthe nucleus of these cell lines, was retained in the cytoplasm upon engagement of LAIR-1.This was evident already 8 h after LAIR-1 occupancy, when apoptosis was not yet detect-able by fluorometric or ultrastructural analysis. Moreover, a reduction in inhibitor ‹ B § phos-phorylation was observed after LAIR-1 engagement. As blocking of NF- ‹ B activation hasbeen shown to rescue sensitivity to anti-cancer drugs in solid tumors, we suggest thatLAIR-1 may represent a possible target for pharmacological approaches aimed to potentiateanti-leukemic therapy.

Key words: Leukocyte-associated Ig-like receptor-1 / Myeloid leukemia / Apoptosis / NF- ‹ B /Proliferation

Received 10/4/00Accepted 23/6/00

[I 20858]

Abbreviations: LAIR-1: Leukocyte-associated Ig-likereceptor-1 PCD: Programmed cell death NF- ‹ B: Nuclearfactor ‹ B I ‹ B: Inhibitor ‹ B PI: Propidium iodide GAM:Goat anti-mouse HRP: Horseradish peroxidase GAR: Goatanti-rabbit

1 Introduction

Apoptosis is a form of programmed cell death (PCD)characterized by reduction of cell size and by definedultrastructural changes [1]. Cells undergo apoptosis bothduring ontogeny, where PCD is a way of remodelling tis-sues, and in pathological conditions, such as viral infec-tions and neoplasias, where apoptosis avoids or delaysthe release of noxious contents from dying cells [1].Apoptosis is generally mediated by the activation of a

series of proenzymes, the caspases, responsible for thecleavage of critical cellular substrates, leading to theabove-mentioned morphological changes and cell death[2–4].

PCD can be regulated by the NF- ‹ B/Rel transcriptionfactors (p50, p65 or RelA, c-Rel and RelB), which stimu-late the expression of proteins that inhibit apoptosis[5–7]. In normal cells, NF- ‹ B is retained in the cytoplasm,due to the binding to unphosphorylated inhibitor ‹ B (I ‹ B)proteins, and translocate to the nucleus upon activationwith stimuli such as mitogens or cytokines, which caninduce I ‹ B phosphorylation and degradation [5, 8–11]. Inthe nucleus, the NF- ‹ B dimers bind to target DNA ele-ments and activate transcription of genes encoding pro-teins involved in immune or inflammation responses andin cell growth [9–11]. In neoplastic cells, constitutive

Eur. J. Immunol. 2000. 30: 2751–2758 LAIR-1-mediated apoptosis of human myeloid leukaemias 2751

Fig. 1. LAIR-1 engagement prevents the proliferation ofTHP1, MM6 and U937 cell lines. THP1 (panel A) or MM6(panel B) or U937 (panel C) cells were cultured in theabsence [squares, control (CTR)] or presence (5 ? g/ml) ofanti-LAIR-1 (diamonds) or anti-CD58/LFA3 mAb (circles), asindicated. [3H] Thymidine uptake was evaluated at differenttime points (24, 48 or 72 h). Results are expressed as cpm ×10–3 ± SD of triplicate wells for each culture and are themean of ten independent experiments.

nuclear NF- ‹ B is found, resulting in the imbalancebetween mitosis and growth control [12–16]. In particu-lar, in juvenile myelomonocytic leukemia nuclear NF- ‹ B/Rel proteins are thought to be responsible for the spon-taneous proliferation of these cells [17]. Thus, mecha-nisms that result in the modulation of NF- ‹ B activitymight be expected to regulate both cell growth and PCD.

Recently, the leukocyte-associated Ig-like receptor-1(LAIR-1) (previously termed p40), belonging to the immu-noglobulin superfamily, has been identified and cloned[18, 19]. It inhibits the proliferation of T lymphocytes inresponse to T cell receptor engagement and of naturalkiller (NK) cells in response to IL-2 [20]. LAIR-1 is ex-pressed on the large majority of peripheral blood leuko-cytes, including cells of myelomonocytic origin [18–21].We have reported that the engagement of LAIR-1 onCD34+ peripheral blood precursors interferes with theirdifferentiation to monocytes and dendritic cells inducedby granulocyte-monocyte colony-stimulating factor(GM-CSF) [22].

In this report, we provide evidence that occupancy ofLAIR-1 on human myelomonocytic leukemic cell linesinhibits proliferation and leads to PCD via a caspase-independent pathway. In addition, we show that the p65subunit of NF- ‹ B, which is constitutively expressed inthe nucleus of these cell lines, is mainly retained in thecytoplasm upon engagement of LAIR-1, together with areduction in I ‹ B § phosphorylation.

2 Results

2.1 Engagement of LAIR-1 on human myeloidleukemia cell lines prevents proliferation andinduces apoptosis

Since we have shown that LAIR-1 down-regulates signaltransduction and inhibits the differentiation of mono-cytes and dendritic cells from CD34+ peripheral bloodprecursors [22], we asked whether this molecule coulddeliver a negative signal on myeloid leukemia cells aswell. Thus, the myelomonocytic THP1 (Fig. 1 A), MM6(Fig. 1 B) or U937 (Fig. 1 C) cell lines were cultured in thepresence or absence of anti-LAIR-1 or anti-LFA3/CD58(used as an isotype-matched control) mAb. [3H]-Thymidine uptake was then evaluated at different timepoints (24, 48 or 72 h). As indicated in Fig. 1, engage-ment of LAIR-1, but not of CD58/LFA-3, could preventproliferation of the three cell lines. Analysis of the differ-ent phases of the cell cycle, performed by propidiumiodide (PI) staining and FACS analysis, revealed that thepercentage of MM6 cells in the G0/G1 phase increasedduring the 24–36 h with a concomitant decrease of cells

in the S or G2/M phases (Fig. 2). Similar results wereobtained with THP1 or U937 cell lines (not shown).

Interestingly, fragmentation of DNA, as assessed by PIstaining, was observed in THP1 (Fig. 3 upper panels),MM6 (Fig. 3, central panels) and U937 (Fig. 3 lower pan-els) cells upon engagement of LAIR-1. Conversely,cross-linking of LFA3/CD58 did not exert any significanteffect (Fig. 3). This finding support the idea that LAIR-1can induce an apoptotic signal in these cell lines. To fur-ther confirm PCD, MM6 (Fig. 4), THP1 and U937 (notshown) cells were evaluated by transmission electronmicroscopy at different time points after occupancy ofLAIR-1. Fig. 4 shows that untreated cells have a cyto-

2752 A. Poggi et al. Eur. J. Immunol. 2000. 30: 2751–2758

Fig. 2. Leukemic cells are blocked in G0/G1 phase uponLAIR-1 engagement. MM6 cells collected at the indicatedtime points after culture in the absence [white squares,control (CTR)] or presence (5 ? g/ml) of anti-LAIR-1 (blacksquares) mAb, were stained with 50 ? g/ml PI solution toevaluate the G0/G1, S or G2/M phases of the cell cycle.Samples were anlyzed ungated on a FACSort (BectonDickinson). Results are expressed as percentage of cells inthe different phases of the cell cycle and are the mean ± SDof three independent experiments.

Fig. 3. THP1, MM6 and U937 cell lines undergo apoptosisafter LAIR-1 engagement. THP1 (upper panels) or MM6(central panels) or U937 cells (lower panels) were collected48 h after engagement of LAIR-1 or CD58/LFA-3, as indi-cated, obtained using the corresponding mAb at 5 ? g/ml,and stained with 50 ? g/ml PI solution to evaluate DNA frag-mentation. Nil: untreated cells. Samples were analyzedungated on a FACSort (Becton Dickinson). Results are plot-ted as histograms of red fluorescence intensity (PI-DNA)(arbitrary units, a.u., Log scale) vs. number of cells. Apop-totic cells are identified by a decreased PI-DNA content, i.e.decreased red fluorescence intensity.

plasm rich in organelles and a bilobate nucleus contain-ing evident nucleoli, this morphology being conservedup to 8 h after LAIR-1 engagement (not shown). By 12 h,chromatin condensation was detectable, with periphe-ralization of chromatin evident at 18–24 h and blebbingof the nucleus by 24–36 h; eventually, shrinking of thecell, an apoptotic nucleus and a lack of cytoplasmicorganelles became evident (48 h).

2.2 Comparison between LAIR-1- and CD95/Fas-mediated apoptosis

It is well established that the engagement of CD95/Fason leukocytes leads to PCD [23, 24] and that this effectis mediated by the activation of cytoplasmic enzymestermed caspases [3, 4]. Thus, we compared LAIR-1 andCD95/Fas-mediated apoptosis in THP-1 and MM6 celllines in terms of kinetics and sensitivity to different cas-pase inhibitors.

First, we found that LAIR-1 was expressed at higher lev-els than CD95/Fas on THP-1 and MM6 cell lines (Fig. 5,panels A and C). Kinetics analysis showed that the per-centage of apoptotic cells was about 20 % 24 h afterLAIR-1 occupancy on both THP1 and MM6 cell lines,increased to 40–50 % at 36 h, reaching the 80–85 % at48 h (Fig. 5 B and D). PCD induced via LAIR-1 was con-sistently higher and was delayed compared to that trig-gered via CD95/Fas in THP1 cells (Fig. 5 B). Indeed, theengagement of CD95/Fas led to a maximum of apop-totic cells (15–30 %) after 24 h incubation, whereas thepeak of LAIR-1-induced apoptosis was detected at 48 h

(Fig. 5 B). That CD95/Fas triggering elicited apoptosis inMM6 cells to a lesser extent than in THP1 cells is not sur-prising, since the former cell line expresses very low lev-els of CD95/Fas (Fig. 5 C).

Interestingly, none of the caspase inhibitors used couldblock LAIR-1-induced PCD, both in MM6 and THP1 celllines, at variance with apoptosis elicited via CD95/Faswhich was prevented by the caspase-1 inhibitors Z-VADand Y-VAD and the caspase 8 inhibitor IETD-fmk(Table 1). Taken together, these findings would suggestthat LAIR-1- and CD95/Fas induce apoptosis in thesemyeloid cell lines via different molecular mechanisms.

2.3 LAIR-1 engagement prevents nucleartranslocation of NF- ‹ B p65 subunit and thephosphorylation of I ‹ B >

In light of the reported involvement of NF- ‹ B in preven-tion of PCD in several cell types [5–7], we analyzed thepossibility that a defective nuclear translocation of NF-‹ B subunits p50 and p65 might follow LAIR-1 occu-

pancy. Fig. 6 A shows that the p50 subunit of NF- ‹ B is


Fig. 4. Ultrastructural changes in MM6 cells upon LAIR-1 engagement. Apoptosis was evaluated by transmission electronmicroscopy at the indicated time points after engagement of LAIR-1 molecule obtained using the corresponding mAb at 5 ? g/ml.Cells were fixed in 2 % glutaraldehyde for 20 min at 4 °C and post fixation was performed in 1 % osmium tetroxide. After dehyd-ratation and embedding in epon-araldite resin, thin sections (80 nm) were stained with uranyl acetate and lead citrate and ana-lyzed under a Zeiss CEM 902 electron microscope. Chromatin condensation is evident at 18 h, nuclear blebbing at 24–36 h andaptotic nucleus by 48 h.

Table 1. Effects of caspase inhibitors on the apoptosis induced by LAIR-1 engagement in THP1 or MM6 cell lines: comparisonwith CD95/Fas

% Apoptotic cellsa)

THP1 MM6

Caspase inhibitors LAIR-1 CD95/Fas LAIR-1 CD95/Fas

None 80 ± 3 28 ± 4 79 ± 5 14 ± 2

Caspase 1/inhibitor I (Y-VAD-cmk) 78 ± 5 3 ± 2 78 ± 3 2 ± 1

Caspase 1/inhibitor V (Z-VAD-fmk) 79±3 4 ± 3 80 ± 4 2 ± 1

Caspase 3/inhibitor I (DEVD-CHO) 77 ± 6 n.d. 76 ± 5 n.d.

Caspase 3/inhibitor II (Z-DEVD-fmk) 80 ± 3 n.d. 82 ± 2 n.d.

Caspase 8/FLICE inhibitor (IETD-fmk) 78 ± 2 4 ± 2 80 ± 2 3 ± 2

a) Apoptosis was evaluated in THP1 or MM6 cell lines by PI staining and FACS analysis as in Fig. 2, after LAIR-1 (48 h) or CD95/Fas (24 h) engagement (obtained using the corresponding mAb at 5 ? g/ml). In some experiments, cells were treated with theindicated caspase inhibitors at 10 ? M concentration before engagement of LAIR-1 or CD95/Fas. Results are expressed aspercentage of apoptotic cells and are the mean ± SD from six independent experiments.


Fig. 5. Comparison between LAIR-1 and CD95/Fas-mediated apoptosis. Left panels: THP1 (A) and MM6 (C)cells were stained with the anti-LAIR-1 or the anti-CD95/FasmAb, as indicated, followed by PE-conjugated anti-isotype-specific GAM serum. NIL: cells stained with an unrelatedmAb followed by PE-GAM. Samples were analyzed on aFACSort (Becton Dickinson) and results expressed as Logred fluorescence intensity (a.u.) vs. number of cells. Onerepresentative experiment out of six. Right panels: Apopto-sis was evaluated in THP1 (B) or MM6 (D) cell lines, at theindicated time points after engagement of LAIR-1 (dia-monds) or CD95/Fas (triangles), (obtained using the corre-sponding mAb at 5 ? g/ml), by PI staining and FACS analysisas in Fig. 2. CTR (squares): control untreated cells. Resultsare expressed as percentage of apoptotic cells and are themean ± SD from six independent experiments.

Fig. 6. LAIR-1 engagement prevents nuclear translocationof NF- ‹ B p65 and I ‹ B § phosphorylation. Panel A: cytosolic(C) or nuclear (N) extracts were prepared from MM6 cells, asdescribed in the Materials and methods section, at the indi-cated time points after engagement of LAIR-1 (obtainedusing the corresponding mAb at 5 ? g/ml). Thirty microgramsof protein/lane were loaded and run by SDS-PAGE (12 %gel) under reducing conditions. Samples were electrotrans-ferred onto nitrocellulose filters and incubated with the anti-NF- ‹ B p50 or p65 mAb at 10 ? g/ml, followed by HRP-conjugated GAM (1:10,000 dilution) and the immunoreac-tive bands were revealed by luminol reaction. g -actin isshown to confirm equal protein loading. Panel B: cells weretreated with the anti-LAIR-1 mAb and after 0.5, 1 or 2 h,were lysed and immunoprecipitated with an anti-I ‹ B § mAb.Samples were run and blotted as above and incubated witheither a rabbit polyclonal anti-I ‹ B § or a rabbit polyclonalanti-phosphoserine I ‹ B § (1 :400 dilution) antiserum (P-I ‹ B § ), followed by HRP-conjugated GAR (1:10,000 dilution)and the immunoreactive bands were revealed by luminolreaction. One representative experiment out of three isshown.

mostly present in the cytoplasm of untreated, proliferat-ing MM6 cells, whereas the p65 subunit of NF- ‹ B ismainly expressed in the nuclei. This is in keeping withother observations showing constitutive nuclear NF- ‹ Bin neoplastic cells [12–16], in particular in juvenile myelo-monocytic leukemia where activated NF- ‹ B is thoughtto be responsible for the spontaneous proliferation ofleukemic cells [17]. Noteworthy, upon engagement ofLAIR-1, most of p65 subunit was retained in the cyto-plasm (Fig. 6 A). This was evident already 8 h afterLAIR-1 occupancy, when apoptosis was not yet detect-able by PI staining or ultrastructural analysis. Similarresults were obtained with the THP1 and U937 cell lines(not shown). These findings further support that NF- ‹ Bnuclear translocation is needed for leukemic cell survivaland point to the involvement of NF- ‹ B in LAIR-1-inducedPCD.

As one of the reported mechanisms whereby NF- ‹ B isretained in the cytoplasm is the binding to unphosphoryl-ated I ‹ B § [5], we addressed the question of whetherLAIR-1 engagement on leukemic cells affected I ‹ B §phosphorylation. Cells were treated with the anti-LAIR-1mAb and after 30 min, 1 h or 2 h were lysed, immunopre-cipitated with a monoclonal anti-I ‹ B § and blotted witheither a polyclonal anti- ‹ B § or a polyclonal anti-phosphoserine I ‹ B § . Fig. 6 B shows that engagement ofLAIR-1 leads to a significant decrease in I ‹ B § phosphor-ylation.


3 Discussion

In this study we report that the transmembrane signaldelivered via LAIR-1 inhibits proliferation of human mye-lomonocytic leukemic cells, leading to PCD. This apop-tosis does not seem to be caspase dependent, indicat-ing that LAIR-1-induced PCD is due to the triggering of apathway different from that of Fas/FasL. Remarkably,engagement of LAIR-1 prevents the nuclear transloca-tion of the p65 subunit of NF- ‹ B.

Inducing or preventing PCD is a way of regulating tissuedevelopment during ontogeny [1, 2]. From this viewpoint,LAIR-1 seems to function as a regulator of both cellgrowth and differentiation. Indeed, we have previouslyreported that it inhibits the proliferation of T lymphocytesand NK cells [21] and the differentiation of blood precur-sors to monocytes and dendritic cells in vitro [22].Although the ligand(s) for LAIR-1 are still unknown, it istempting to speculate that such ligand(s) are expressedby stromal cells in the microenvironment where normalor neoplastic cell differentiation takes place. Moreover,the existence of mRNA coding for LAIR-1 isoform lack-ing the transmembrane domain [18], suggests that solu-ble LAIR-1, secreted in the extracellular milieu, can blockthe interaction of the transmembrane receptor with itsligands. This might represent a feed-back control ofLAIR-1-mediated effects on cell proliferation and differ-entiation.

Along this line, our present data provide evidence for theinvolvement of LAIR-1 in blocking cell growth of myeloidleukemia cell lines and inducing apoptosis by preventingNF- ‹ B nuclear translocation. Constitutive nuclear ex-pression of NF- ‹ B/Rel transcription factors has beenreported to contribute to myeloid leukemia cell prolifera-tion [17]. Proteins of NF- ‹ B/Rel family are usuallyretained in the cytoplasm in normal cells, due to theirbinding to unphosphorylated I ‹ B [5]; upon activation,I ‹ B proteins are dephosphorylated, and NF- ‹ B/Rel com-plexes translocate to the nucleus and bind to ‹ B DNAelements up-regulating the transcription of several pro-teins, including growth factors and proteins which pre-vent cells from apoptosis [8–10]. In this regard, it is ofnote that LAIR-1 engagement on myeloid leukemia celllines prevented nuclear translocation of NF- ‹ B p65 sub-unit and decreased the phosphorylation of I ‹ B § ; thismechanism might account both for the inhibition of cellproliferation and induction of apoptosis. This concept isalso supported by the observation that cytoplasmicretention of NF- ‹ B is evident early after (8 h) LAIR-1engagement, while ultrastructural changes and internu-cleosomal fragmentation, typical of the apoptotic pro-cess, occur after 24–48 h. Conversely, conventional pro-

apoptotic signals, such as those mediated via CD95/Fas, induce PCD soon after the delivery of the signalitself [23, 24]. Moreover, CD95/Fas-mediated apoptosisis caspase dependent [23, 24], whereas PCD whichfollows LAIR-1 triggering is appearently caspase in-dependent.

Taken together, these findings suggest that the role ofLAIR-1 goes beyond the originally described inhibition ofimmune cell functions [18]. In particular, the transmem-brane signal delivered via LAIR-1 and preventing NF- ‹ Bnuclear translocation resembles the inhibition of NF- ‹ Bactivation mediated by anti-neoplastic drugs such as flu-darabine [15]. The efficacy of anti-neoplastic therapiescorrelates to the growth rate of tumors and to the alteredexpression of apoptosis-regulating genes [1, 2]. In par-ticular, chronic leukemias with a low growth fraction aremore sensitive to pro-apoptotic drugs than to anti-mitotic agents and the extent of drug-induced apoptosiscorrelates with the therapeutic response in vivo [2]. Inturn, the apoptotic response to anti-neoplastic agentsdepends on NF- ‹ B/Rel nuclear activity [11–15], which isalso believed to play an important role in the resistanceto chemotherapy [25, 26]. Indeed, blocking of NF- ‹ Bactivation has been shown to rescue the tumor cell sen-sitivity to anti-cancer drugs [17, 27, 28]. From this view-point, LAIR-1 may represent a possible target for phar-macological approaches aimed to potentiate anti-tumortherapy.

In conclusion, LAIR-1 provides a novel pathway todeliver membrane signals which block proliferation andinduce PCD; this might implicate that it is able to turn offautocrine growth and switch on a suicide program inleukemic cells.

4 Materials and methods

4.1 Cells and reagents

The myelomonocytic leukemic cell lines THP1 and MM6(MonoMac-6) and U937 were kindly provided by D. Vercelli(HSR-Dibit, Milan, Italy). Cells were cultured in RPMI 1640medium supplemented with 2 mM L-glutamine, 100 IU/mlpenicillin and 100 ? g/ml streptomycin (Biochrom, Berlin,Germany) and 10 % heat-inactivated FCS (PAA Labour, Linz,Austria). Media were endotoxin free as shown by the Limuluslysate colorimetric assay (PBI, Milan, Italy).

The caspase 1 inhibitor I (ICE inhibitor, YVAD-cmk) andinhibitor V (Z-VAD-fmk), caspase 3 inhibitor I (DEVD-CHO)and II (Z-DEVD-fmkI and FLICE/caspase 8 inhibitor (IETD-fmk) were purchased from Alexis Biochemicals (Laufelfin-geln, Switzerland) and used at the concentration of 10 ? M.The anti-LAIR-1 (previously termed p40) (NKTA255, IgG1)


(mAb was produced in our laboratory, as described [22]; theanti-LFA3/CD58 (IgG1, clone TS2.9) was from the AmericanTissue Culture Collection (Rockville, MD) and the anti-Fas/CD95 mAb (CH-11, IgM) was from Diaclone (Besancon,France); the anti-NF- ‹ B p50 and p65, the monoclonal anti-I ‹ B § (H4, IgG2a), the rabbit polyclonal anti-I ‹ B § and theanti- g -actin mAb were from Santa Cruz Biotechnology(Santa Cruz, CA); the rabbit anti-phosphoserine I ‹ B § wasfrom Calbiochem (San Diego, CA). The affinity purified anti-Ig (H+L) goat anti-mouse serum (GAM) and horseradish per-oxidase (HRP) conjugated-GAM or goat anti-rabbit (GAR)were purchased from Zymed (San Francisco, CA).

4.2 Proliferation assay

THP1 or MM6 or U937 cells (104) were cultured in the pres-ence or absence of anti-LAIR-1 or anti-CD58/LFA-3 mAb inU-bottom 96-microwell plates in a final volume of 200 ? l. Atdifferent time points (24, 48 or 72 h) 1 ? Ci of [3H] thymidinewas added to the cultures during the last 16 h of incubation;then, cells were harvested on a filter paper, solubilized in ascintillation liquid for acqueous samples (Ultima Gold, Pack-ard, Sterling, VA) and analyzed in a g -counter (Packard).Results are expressed as cpm × 10–3 ± SD of triplicate wellsfor each culture and are the mean of ten independent experi-ments.

4.3 Indirect immunofluorescence and cytofluorimetricanalysis

Single fluorescence stainings were performed as described[20]. Briefly, 5 × 104 cells were stained with the correspond-ing mAb followed by PE-conjugated anti-isotype-specificGAM serum (Southern Biotechnology, Birmingham, AL).Control aliquots were stained with an unrelated mAb fol-lowed by the second reagent. Samples were analyzed on aflow cytometer (FACSort, Becton Dickinson) equipped withan argon ion laser exciting PE at 488 nm. Data were ana-lyzed using Lysis II (version 1.1) and results are expressed asLog red fluorescence intensity (arbitrary units, a.u.) vs. cellnumber.

4.4 Evaluation of apoptosis

THP1 or MM6 or U937 cells were collected at different timepoints after engagement of LAIR-1 or CD95/Fas, obtainedusing the corresponding mAb at 5 ? g/ml, and stained with50 ? g/ml PI solution containing 100 units/ml RNase type A,10 mM EDTA (to inactivate endogenous nucleases) and0.0015 % NP40 as previously described to evaluate DNAfragmentation [29]. In some experiments, cells were treatedwith the caspase inhibitors at 10 ? M concentration beforeoccupancy of LAIR-1 or CD95/Fas. Samples were analyzedungated on a FACSort (Becton Dickinson) equipped withlaser excitation at 488 nm and 610 nm longpass filter for the

PI fluorescence detection. Calibration was assessed withCaliBRITE particles (Becton Dickinson) using AutoCOMPcomputer program (version 2.1.2). At least 104 cells per sam-ple were analyzed and results plotted as histograms of sin-gle parameters (PI-DNA) (arbitrary units, a.u., Log scale)using Lysis II (version 1.1).

4.5 Electron microscopy

Apoptosis was also evaluated by transmission electronmicroscopy (EM). Cells were fixed in 2 % glutaraldehyde for20 min at 4 °C and washed three times in phosphate buffer.Post fixation was performed in 1 % osmium tetroxide. Sam-ples were then dehydrated in ethyl alcohol and propyleneoxide and embedded in epon-araldite resin. Thin sections(80 nm) were then obtained, stained with uranyl acetate andlead citrate and analyzed under a Zeiss CEM 902 electronmicroscope.

4.6 Western blot

Cytosolic (C) or nuclear (N) extracts were obtained asdescribed [24]. Briefly, cells were incubated in a buffer con-taining 50 mM Tris, 10 mM NaCl, 1 mM EDTA, 1 mM dithio-threitol, 1 ? M leupeptin, pepstatin, aprotinin and phenyl-methylsulfonylfluoride and 0.5 % NP40; nuclei were thenextracted at 4 °C in the same buffer as above, except NaCl0.4 M. buffer. The purity of the two fractions was checked byWestern blot using an anti-histone antibody (not shown).Protein dosage was performed with the Detergent-Compatible Bio-Rad kit based on the colorimetric Lowrymethod and 30 ? g of protein/lane were loaded and run bySDS-PAGE (12 % gel) under reducing conditions. Sampleswere then electrotransferred onto nitrocellulose filters(Hybond ECL, Amersham Italia S.r.l., Milan, Italy) asdescribed [22] and blocked overnight with 10 % non-fat drymilk in PBS. In some experiments cells were treated with theanti-LAIR-1 mAb, lysed, immunoprecipitated with the anti-I ‹ B § mAb and blotted as above. Membranes were thenincubated 1 h with the anti-NF- ‹ B p50 or p65 mAb at 10 ? g/ml, or with either the anti-phosphoserine-I ‹ B § antiserum orwith the rabbit anti-I ‹ B § polyclonal antiserum (1:400 dilu-tion, followed by HRP-conjugated GAM, or GAR, respec-tively, (1 :10,000 dilution) and the immunoreactive, bandswere revealed by luminol reaction (ECL, Amersham, LittleChalfont, GB).

Acknowledgements: This work was partially supported bygrants from the Italian Association for Cancer Research(AIRC) and Istituto Superiore di Sanita-Ministero della Sanita(PSN98–2000).


References

1 Earnshaw, W. C., Nuclear changes in apoptosis. Curr. Opin. Cell.Biol. 1995. 7: 337–343.

2 Martin, S. J. and Green, D. R., Protease activation during apop-tosis: death by a thousand cuts? Cell 1995. 82: 349–352.

3 Henkart, P. A., ICE family proteases: mediators of all apoptoticcell death? Immunity 1996. 4: 195–201.

4 Cohen, G. M., Caspases: the executioner of apoptosis. Bio-chem. J. 1997. 326: 1–16.

5 Baldwin Jr., A. S., The NF- ‹ B proteins: new discoveries andinsights. Annu. Rev. Immunol. 1996. 14: 649–681.

6 Lee, H., Arsura, M., Wu, M., Duyao, M., Buckler, A. and Sonen-shein, G., Role rel-related factors in control of c-myc gene tran-scription in receptor-mediated apoptosis of the murine B cellWEHI 231 line. J. Exp. Med. 1995. 181: 1169–1177.

7 Sarma, W., Lin, Z., Clark, L., Rust, B., Tewar, M., Noelle, R. andDixit, V., Activation of the B-cell surface receptor CD40 inducesA20, a novel zinc finger protein that inhibits apoptosis. J. Biol.Chem. 1995. 270: 12343–12346.

8 Siebenlist, U., Franzoso, G. and Bauer, K., Structure, regulationand function of NF- ‹ B. Annu. Rev. Cell Biol. 1994. 10: 405–455.

9 Bauerle, P. and Henkel, T., Function and activation of NF- ‹ B inthe immune system. Annu. Rev. Immunol. 1994. 12: 141–179.

10 Thanos, D. and Maniatis, T., NF- ‹ B: lesson in family values. Cell1995. 80: 529–532.

11 Finko, T. and Baldwin, A., Regulation of NF- ‹ B: the emergingroles of phosphorylation and proteolysis. Immunity 1995. 3:263–272.

12 Beg, A. and Baltimore, D., An essential role for NF- ‹ B in pre-venting TNF- § -induced cell death. Science 1996. 274: 782–784.

13 Wang, C.-Y., Mayo, M. W. and Baldwin Jr., A. S., TNF- and can-cer therapy-induced apoptosis: potentiation by inhibition of NF-‹ B. Science 1996. 274: 784–787.

14 Van Antwerp, D. J., Martin, S. J., Kafri, T., Green, D. R. andVerma, I. M., Suppression of TNF- § -induced apoptosis by NF-‹ B. Science 1996. 474: 787–789.

15 Romano, M. F., Lamberti, A., Tassone, P., Alfinito, F., Costan-tini, S., Chiurazzi, F., Defrance, T., P., Bonelli, P., Tuccillo, F.,Turco, M. C. and Venuta, S., Triggering of CD40 antigen inhibitsfludarabine-induced apoptosis in B chronic lymphocytic leuke-mia cells. Blood 1998. 92: 990–995.

16 Barinaga, M., Life-Death balance within cell. Science 1998. 274:780.

17 Kochetkova, M., Iversen, P. O., Lopez, A. F. and Shannon, M.F., Deoxyribonucleic acid triplex formation inhibits granulocytemacrofage colony-stimulating factor gene expression and sup-presses growth in juvenile myelomonocytic leukemic cells.J. Clin. Invest. 1996. 99: 3000–3008.

18 Meyaard, L., Adema, G. J., Chang, C., Woolat, E., Sutherland,G. R., Lanier, L. L. and Philips, J. H., LAIR-1, a novel inhibitoryreceptor expressed on human mononuclear leukocytes. Immu-nity 1997. 7: 283–290.

19 Meyaard, L., Hurenkamp, J., Clevers, H., Lanier, L. L. and Phil-lips, J. H., Leukocyte-associated Ig-like receptor-1 functions asan inhibitory receptor on cytotoxic T cells. J. Immunol. 1999. 162:5800–5804.

20 Poggi, A., Tomasello, E., Revello, V, Nanni, L., Costa, P. andMoretta, L., p40 molecule regulates NK cell activation mediatedby NK receptors for HLA-class I antigens and TCR-mediated trig-gering of T lymphocytes. Int. Immunol. 1997. 9: 1271–1279.

21 Poggi, A., Pella, N., Morelli, L., Spada, F., Revello, V., Sivori, S.,Augugliaro, R., Moretta, L. and Moretta, A., p40, a novel sur-face molecule involved in the regulation of the non-major histo-compatobility complex-restricted cytolytic activity in humans.Eur. J. Immunol. 1995. 25: 369–376.

22 Poggi, A., Tomasello, E., Ferrero, E., Zocchi, M. R. andMoretta, L., LAIR-1 regulates the differentiation of peripheralblood precursors to dendritic cells induced by granulocyte-monocyte colony-stimulating factor. Eur. J. Immunol. 1998. 28:2086–2091.

23 Itoh, N., Tsujimoto, Y. and Nagata, S., Effect of Bcl-2 on Fasantigen-mediated cell death. J. Immunol. 1993. 151: 621–627.

24 Ogasawara, J., Watanabe-Fukunaga, R., Adachi, M., Matsu-zawa, A., Kasugai, T., Kitamura, Y, Itoh, N, Suda, T. andNatata, S., Lethal effect of the anti-Fas antibody in mice. Nature1993. 364: 806–809.

25 Thompson, C. B., Apoptosis in the pathogenesis and treatmentof diseases. Science 1995. 267: 1456–1462.

26 Huang, P., Robertson, L. E., Wright, S. and Plunkett, W., Highmolecular weight DNA fragmentation: a critical event in nucleo-side analogous-induced apoptosis in leukemia cells. Clin. CancerRes. 1995. 1: 1005–1013.

27 Wang, C.-Y., Cusack Jr, J. C., Liu, R. and Baldwin Jr., A. S.,Control of inducible chemoresistance: enhanced anti-tumor ther-apy through increased apoptosis by inhibition of NF- ‹ B. Nat.Med. 1999. 5: 412–417.

28 Suyang, H., Phillips, R., Douglas, I. and Ghosh, S., Role ofunphosphorylated, newly synthesized I ‹ B g in persistent of NF ‹ B.Mol. Cell. Biol. 1996. 16: 5444–5449.

29 Zocchi, M. R., Poggi, A. and Rubartelli, A., Selective engulfe-ment of apoptotic bodies by dendritic cells via § v g 3 integrin.requirement of intracellular and extracellular calcium. Eur. J.Immunol. 1997. 27: 1893–1900.

Correspondence: Alessandro Poggi, Laboratory of Im-munology, National Institute for Cancer Research c/oAdvanced Biotechnology Center, L.go Rosanna Benzi 10,I-16132-Genoa, ItalyFax: +39 010 354123e-mail: poggi — ermes.cba.unige.it


engagement of the leukocyte-associated ig-like receptor-1 induces programmed cell death and prevents...

Documents