Apr 14, 1998 - TR cDNA. Size is indicated by a ladder of markers of 1600, 1000, ..... 24 Snapper, C. M., Yamaguchi, H., Moorman, M. A. and Mond, J. J.. 1994.
International Immunology, Vol. 10, No. 8, pp. 1093–1101
© 1998 Oxford University Press
Phenotypic and functional characterization of a panel of cytotoxic murine NK cell clones that are heterogeneous in their enhancement of Ig secretion in vitro Quirijn Vos, John R. Ortaldo2, Michelle Conan-Cibotti3, Michele D. Vos4, Howard A. Young2, Steven K. Anderson2, Kim Witherspoon, Ilan Prager, Clifford M. Snapper1 and James J. Mond Departments of Medicine and 1Pathology, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, USA 2Laboratory of Experimental Immunology, Division of Basic Sciences, National Cancer Institute–FCRDC, Frederick, MD 21702, USA 3Experimental Immunology Branch, National Cancer Institute, Bethesda, MD 20892, USA 4Department of Cell and Cancer Biology, Division of Clinical Sciences, National Cancer Institute, Rockville MD 20850, USA
Keywords: B cell, cytotoxicity, granzyme, Ig secretion, NK cell, TI-2 response
Abstract NK cells not only function as cytotoxic effector cells, but also have immunoregulatory roles including the enhancement of Ig secretion. To have a stable and uniform population of NK cells to study their role in Ig secretion, we generated murine NK clones. Thus, culture of splenocytes from mice that were homozygous for a mutation in the p53 tumor suppressor gene (p53-KO) with IL-2 and poly(IC) resulted in a long-term NK line, from which four stable clones were derived. This approach also yielded a long-term NK line from splenocytes of normal C57BL/6 mice. Identification of the clones as members of the NK lineage was based on large granular morphology, expression of NK-TR and absence of TCR gene rearrangement. Flow cytometry revealed that all clones expressed IL-2R α and β, chains and B220, but no CD3, NK1.1, DX5 or Ly-49. RT-PCR analysis showed heterogeneity in NK1.1 gene expression, and demonstrated expression of perforin and several granzymes in all clones. Three out of four clones lysed YAC-1, but not P815 target cells, corresponding to a pattern of NK specificity. All NK clones enhanced Ig secretion in an in vitro model for T cell-independent type 2 antigens, albeit to varying degrees. We found no correlation between the degree of helper activity of the NK clones and the level of their cytotoxic activity on YAC-1 targets. Thus, we established murine NK clones, and show that they mediate both cytotoxicity and enhancement of Ig secretion.
Introduction NK cells were defined originally as large granular lymphocytes that lyse tumor and virally infected cells in the absence of MHC restriction and without previous sensitization (1,2), and were shown to mediate the rejection of incompatible bone marrow allografts in lethally irradiated mice (3). However, studies employing freshly explanted NK or cells resulting from short-term in vitro cultures have shown that NK cells are not limited to a role of cytotoxic effector cells, but have also been
implicated as regulatory cells in the immune response to bacterial (4,5), protozoan (6,7) and viral infections (8,9) of both conventional and T cell-deficient mice. This immunoregulatory function of NK cells is based in part on the secretion of various cytokines including CSF (10), IFN-γ (11), IL-1 (12), IL2 (13), IL-8 (14) and tumor necrosis factor-α (15). From these cytokines IFN-γ has emerged as a critical NK-derived inducer of Th1 differentiation in the immune response to protozoan
Correspondence to: Q. Vos Transmitting editor: Z. Ovary
Received 11 March 1998, accepted 14 April 1998
1094 Characterization of murine NK clones infection (16–18). Moreover, NK cells have been shown to stimulate Ig secretion both in vivo (19, 20) and in vitro (21– 24), possibly through the release of IFN-γ (25), a potent stimulator of Ig secretion (26) and inducer of isotype-switching to IgG2a (27). The characterization of NK cell surface receptors (28) and several aspects of their specificity (29,30) have resulted in the description of heterogeneity among NK subpopulations that display potentially different functions (31–34). To extend these findings from freshly explanted NK cells and cells from short-term in vitro cultures, two groups have recently described methods for the establishment of long-term murine NK lines: one using splenocytes from mice that were homozygous for a mutation in the p53 tumor suppressor gene (p53-KO) (35), the other employing culture of fetal liver cells from normal mice (36). These lines provide important tools for the study of functionally heterogeneous NK subpopulations and a variety of aspects of NK cell biology, while avoiding the influence of contaminating cells with phenotypic similarity like natural T cells (37). In recent studies we have reported that sort-purified B cells stimulated in an in vitro model for T cell-independent type 2 (TI-2) antigens, employing a multivalent mIg cross-linking agent and IL-2, can be induced to secrete Ig by addition of freshly explanted NK cells or of cells from short-term NK cultures (23,24). In order to carefully dissect the role of NK cells in the process of induction of Ig secretion in B cells and to determine whether there is any relationship between NK cytotoxicity and NK-mediated help to B cells, we generated murine NK clones. We here describe the establishment and characterization of a murine NK panel: one line and four clones from p53-KO spleen, and one line from normal C57BL/ 6 spleen, that display both cytotoxic effector function and enhance Ig secretion in stimulated B cells. Methods
Mice C57BL/6, CBA.C57BL/6 F1 and 129.C57BL/6 F1 mice homozygous for a mutation in the p53-KO (38) were obtained from Jackson Laboratories (Bar Harbor, ME) and were used at 8– 10 weeks of age. The experiments were conducted according to the principles set forth in the Guide for the Care and Use of Laboratory Animals, Institute of Animal Resources, National Research Council, Department of Health, Education, and Welfare publication (National Institutes of Health) 78-23.
Culture medium RPMI 1640 (Biowhittaker, Walkersville, MD) supplemented with 10% heat-inactivated FCS (Gibco, Grand Island, NY), Lglutamine (2 mM), HEPES (25 mM), 2-mercaptoethanol (50 mM), non-essential amino acids (100 µM), penicillin (100 U/ ml), pyruvate (1 mM) and streptomycin (100 µg/ml) was used for culturing cells.
Reagents Anti-IgD–dextran conjugates (αδ–dex), of anti-IgD a allotype specificity, Hδa/1 (39), and anti-IgD b allotype specificity, AF3.33 (40), were prepared by conjugation of the respective
mAb to a high mol. wt dextran (23106 Da) as previously described (41). For both conjugates approximately six mAb were conjugated to each dextran molecule. The concentration of αδ–dex reflects the amount of mAb and not the entire dextran conjugate. In the flow cytometry analysis of the NK panel the following mAb to Ly-49 were used: Yel/48 (anti-Ly-49A, rat IgG2c), SW5E6 (anti-Ly-49C, mouse IgG2a) and 4D11 (anti-Ly-49G, rat IgG2a). These mAb, and mAb specific for murine B220, NK1.1, DX5, CD3, IL-2Rα and IL-2Rβ were purchased from PharMingen (San Diego, CA). Murine recombinant IL-2 was a kind gift from Dr Maurice Gately (Hoffmann-La Roche, Nutley, NJ). BSA, FITC, poly(IC) and Tween 20 were purchased from Sigma (St Louis, MO). RA3-6B2 rat anti-mouse B220 (42) was conjugated to FITC using a standard protocol. Affinity purified goat anti-mouse κ light chain, goat anti-mouse IgM–alkaline phosphatase and rat-anti-mouse IgD–phycoerythrin (PE) were purchased from Southern Biotechnology Associates (Birmingham, AL).
Establishment of a panel of murine NK cell lines and clones Spleen cell suspensions were prepared from C57BL/6 and 129.C57BL/6 F1 p53-KO mice in culture medium. Red blood cells were lysed with 0.14 M NH4Cl, 0.017 M Tris, pH 7 and the splenocytes were washed twice with culture medium. Next, the cell suspensions were incubated with anti-Thy 1.2 (43) and anti-MHC class II (44) mAb in culture medium for 30 min on ice, washed, and incubated with complement for 45 min at 37°C. Live cells were isolated by centrifugation on a density gradient (Lympholyte-M; Cedarlane, Westbury, NY) and cultured at 106 cells/ml in 24-well flat bottom plates (Costar, Cambridge, MA) in culture medium containing IL-2 (50 U/ml) and poly(IC) (10 mg/ml) in a humidified atmosphere containing 5% CO2 at 37°C. After 14 days of culture the cell suspensions of the individual wells were pooled, centrifuged, resuspended at 105 cells/ml, transferred to culture flasks and cultured for 7 days. Because of the adherence of the resulting cell population, media was removed, and cells were detached using trypsin–versene solution (Biowhittaker, Walkersville, MD), washed twice and cultured for an additional 7 days. This procedure resulted in the establishment of two cell lines: one derived from C57BL/6 splenocytes, B6 NK, the other originating from 129.C57BL/6 F1 p53-KO, PKO. After the initial 28 days of culture, PKO cells were isolated and seeded in flat-bottom 96-well plates (Costar, Cambridge, MA) at a density of 0.3 cells/well and cultured for 14 days. From the initial 120 clones, four clones designated PKO 2, PKO 34, PKO 56 and PKO 101 were utilized for further study.
Analysis of a murine NK cell panel for NK-TR gene expression Total cellular RNA was isolated according to Chomcyznsky and Sacchi (45), and subjected to reverse transcription using NK-TR-specific primers as described previously (46). cDNA was subjected to PCR using NK-TR-specific primers according to Chambers et al. (47). One-fifth of the PCR product was run on 1% agarose gels and transferred onto nylon membranes (GeneScreen; NEN Research, Boston, MA). The filter was probed with the 1.4 kb EcoRI fragment of NK-TR cDNA (46).
Characterization of murine NK clones 1095 Analysis of cell surface molecules on a murine NK cell panel Expression of cell surface molecules was analyzed as previously described (48) using a FASCist flow cytometer (Becton Dickinson, Mountain View, CA). Cells were stained directly using PE- and FITC-labeled primary mAb, or indirectly stained using either a combination of a primary mAb followed by an isotype-specific FITC- or PE-conjugated secondary antibody, or a combination of a biotinylated primary mAb followed by streptavidin–PerCp or avidium–PerCP (Becton Dickinson).
Analysis of a murine NK cell panel for NK1.1 gene expression RNA isolation and RT-PCR were performed as described above. The following primers were used for PCR: sense 59CTACCTCGGTTTAAAGCCACC-39, anti-sense 59-GAAGCACAGCTCTCAGGAGTCAC-39. This primer pair generates a 575 bp PCR product. After gel electrophoresis and transfer to a nylon membrane PCR products were probed using 59GTTTCTCAAGTTTCCAACACTTG-39.
Analysis of a murine NK cell panel for gene expression of cytolytic granule components Total RNA was extracted using TRIzol (Life Technologies, Gaithersburg, MD) and cDNA was synthesized employing Super Script II reverse transcriptase (Life Technologies) according to the manufacturer’s instructions. mRNA expression of murine granzymes was analyzed using a novel RTPCR method. This method allows the simultaneous detection of all granzymes by using a single pair of consensus oligonucleotide primers, which correspond to two conserved motifs of the granzyme subfamily of serine proteases. These motifs encoding the PHSRPYMA amino acid sequence near the Nterminus, and the VAGWG motif located in the middle of the proteins were amplified using the sense primer 59-CCICA(C/ T)TCI(C/A)GICCITA(C/T)ATGGC-39 and the antisense primer 59-CCCCAICCAGC(T/C)(A/G)C-39 respectively. I is the neutral base deoxyinosine and (X/Y) indicates a mix of two bases for one position. Because of heterogeneity within the granzyme family these two primers generate amplified cDNA products of different sizes for granzyme A, B, CEFG (single size) and D, as illustrated by the analysis of B6 NK shown in Fig. 3. In addition, separate PCR were set up for perforin and the ‘housekeeping’ gene HPRT, which was used for normalization of the data. The following primers were used for PCR: perforin, sense primer 59-CACAAGTTCGTGCCAGGTGTA-39, antisense primer 59-CCACACGCCAGTCGTTATTGA-39; HPRT, sense primer 59-GCTGGTGAAAAGGACCTCTC-39, antisense primer 59-GATTCAACTTGCGCTCATCTTAGG C-39. The following number of cycles were used in the three PCR amplifications: granzyme 26, perforin 24 and HPRT 22. To facilitate the detection of the individual PCR products, each of the sense oligonucleotide primers was labeled at the 59 end with a different fluorescent phosphoramidite dye (Applied Biosystems, Foster City CA): the granzyme primer with 6carboxy-fluorescein, the perforin primer with tetra-chlorinated fluorescein and the HPRT primer with hexa-chlorinated fluorescein. The products of the three PCR were pooled, and the mixture was separated and analyzed on a high resolution 5.6% polyacrylamide–urea gel in the presence of fluorescent polynucleotide size standards using a 374A DNA sequencer
(Applied Biosystems). Genescan software (Applied Biosystems) was used to determine the size and peak area for each of the separated fragments.
Analysis of a murine NK cell panel for cytolytic activity NK killing activity was determined in a 51Cr-release assay as previously described (46). The targets cells used were P815 and YAC-1, ATCC nos TIB-64 and TIB-160 respectively.
Preparation and culture of highly purified splenic B cells Enriched populations of B cells were obtained from CBA.C57BL/6 F1 spleen cells from which T cells were eliminated by treatment with rat anti-Thy-1.2 (43), followed by complement. Cells were then fractionated by centrifugation over a discontinuous Percoll gradient (Pharmacia Fine Chemicals, Piscataway, NJ), washed in cold clear HBSS staining buffer containing 3% heat-inactivated FCS, penicillin (100 U/ ml) and streptomycin (100 µg/ml), incubated with anti-B220– FITC in staining buffer for 30 min on ice and electronically sorted on an Epics Elite cytometer (Coulter, Hialeah, FL) for a B220bright phenotype. Reanalysis of sorted cells, that were subsequently stained with anti-IgD–PE, consistently revealed the presence of .99% IgD1B220bright B cells. These highly purified B cells were incubated in flat-bottom 96-well plates (Costar) at 105 cells/ml in a total volume of 200 µl culture media containing IL-2 (50 U/ml), αδ–dex against both a and b haplotype (10 ng/ml) and the indicated percentage of NK cells at 37°C in a humidified atmosphere containing 5% CO2 for 5 days.
Quantitation of IgM-secreting cells in the cultures of stimulated B cells IgM-secreting cells were detected by spot-ELISA according to a modified version of a previously described protocol (51). Cellulose ester membrane 96-well plates (Millipore Multiscreen; Millipore, Molsheim, France) were coated with 50 ml of a 5 mg/ml solution of goat anti-mouse κ light chain antiserum in PBS by incubating overnight at 4°C. The plates were postcoated using culture medium, and cell suspensions from the stimulated B cell cultures were added and incubated for 6 h in a humidified atmosphere containing 5% CO2 at 37°C. Goat anti-mouse IgM alkaline phosphatase was added and incubated overnight at 4°C. Next, the plates were developed using BCIP/NBT substrate solution (Kirkegaard & Perry, Gaithersburg, MD) and spots representing individual IgM-secreting cells were enumerated using an inverted microscope. Results
Generation of a panel of murine NK cell lines and clones We have previously reported that the addition of freshly explanted NK cells or cells from short-term NK cultures to highly purified splenic B cells enhanced the Ig secretory response in an in vitro model for TI-2 antigens (23,24). The recent description of two different strategies for generating murine NK cell lines that display long-term stability (35,36) has provided new opportunities to study the NK-mediated enhancement of Ig secretion in more detail. Thus, we modified
1096 Characterization of murine NK clones Table 1. Expression of cell surface molecules on a panel of murine NK cells Antigen
Fig. 1. All members of a murine NK cell panel express the NK-TR gene product. Total cellular RNA was isolated and subjected to reverse transcription employing NK-TR-specific primers. cDNA was amplified by PCR, separated by gel electrophoresis, transferred onto a nylon membrane and probed with a 1.4 kb EcoRI fragment of NKTR cDNA. Size is indicated by a ladder of markers of 1600, 1000, 558, 470, 396, 344 and 298 bp, marked as M. This method results in two different RT-PCR products: the active NK-TR gene product represented by a 500 bp band, the inactive form by a 528 bp band. The lanes correspond to the following cells: 1, a negative control of murine thymocytes; 2, PKO; 3, PKO 2; 4, PKO 34; 5, PKO 56; 6, PKO 101; and 7, B6 NK.
the method described by Karlhofer et al. (35), that employed culture of NK-enriched spleen cells from p53-KO mice (38), by addition of poly(IC) to the culture medium and omitting anti-NK1.1 coating of the tissue culture plates. Our method yielded one long-term line, PKO, from which four stable murine NK clones, PKO 2, PKO 34, PKO 56 and PKO 101, were derived. Moreover, the same culture conditions also resulted in the establishment of a long-term NK line from normal C57BL/6 spleen. All the members of this NK panel are adherent cells with a large granular morphology.
B220 IL-2Rα IL-2Rβ CD3 NK1.1 DX5 Ly-49A Ly-49C Ly-49D Ly-49G2
NK cells PKO PKO 2 PKO 34 PKO 56 PKO 101 B6NK
ALAK
1 1 1 – – – – – – –
1 1 1 – 1 1 1 1 1 1
1 1 1 – – – – – – –
1 1 1 – – – – – – –
1 1 1 – – – – – – –
1 1 1 – – – – – – –
1 1 1 – – – – – – –
Members of the NK panel were analyzed for the expression of cell surface molecules using flow cytometry. Cells of a short-term culture of activated polyclonal murine NK cells, ALAK, were included as a control population. These data are representative of three individual experiments.
Fig. 2. Heterogeneity of NK1.1 gene expression in a murine NK cell panel. Total cellular RNA was isolated from each of the NK lines and clones, and subjected to reverse transcription. cDNA was amplified by PCR, separated by gel electrophoresis, transferred onto a nylon membrane and the 575 bp product was probed with an NK1.1specific oligonucleotide probe. The lanes correspond to the following NK cells: 1, PKO; 2, PKO 2; 3, PKO 34; 4, PKO 56; 5, PKO 101; and 6, B6 NK.
Analysis of a murine NK cell panel for NK-TR gene expression As a first step in the characterization of our murine NK panel, the expression of the NK-marker NK-TR (46,47,51,52) was analyzed by RT-PCR. The NK-TR gene encodes a 150 kDa type II transmembrane protein, that is highly conserved between species and that constitutes one of the functional requirements for NK killing. The NK-TR protein has been postulated to be part of the putative NK target recognition complex, important for NK effector function. Figure 1 shows the results of the RT-PCR analysis. The data show the presence of two bands in all members of the NK panel (Fig. 1, lanes 2–7). The top band represents an inactive transcript, the bottom band the active one. No NK-TR transcripts were detected in the negative control consisting of murine thymocytes (Fig. 1, lane 1). These data demonstrate the presence of a marker that is required for NK cytotoxic effector function in all members of the NK cell panel. However, non-MHCrestricted killing has also been observed for some mouse and human T lymphocytes (52–55), where it also depends on the expression of NK-TR (47). To examine the possibility that the members of the NK cell panel belonged to the T cell lineage, TCR gene rearrangement was analyzed. Thus, Southern blot analysis of genomic DNA revealed that all members of the NK panel retained the TCR β locus in germline config-
uration (data not shown). Together with the large granular morphology and the NK-TR expression, the lack of TCR rearrangement supports the fact that our panel consists of NK cells.
Analysis of cell surface molecules on a murine NK cell panel Flow cytometry analysis of surface molecules has revealed the existence of heterogeneous NK subpopulations (31–34) and differences between NK lines derived from fetal liver (36). In order to characterize the cells in our NK panel, to address possible heterogeneity among them and to make a comparison with the phenotypes of the previously described long-term NK lines (35,36), flow cytometry analysis of surface antigens was performed. Table 1 summarizes the results of this analysis. These data show that all of the NK cells express the IL-2R α and β chains, and have low levels of B220. No CD3 was expressed, in accordance with the observed lack of T cell receptor rearrangement, and the cells had a uniform Ly-49 negative phenotype, in accordance with both previously described NK panels (35, 36). None of the cells expressed DX5 or NK1.1 on the cell surface. Additional analysis of CD4, CD8, TCR αβ, TCR γδ, FasL and sIg expression resulted in uniformly negative results.
Characterization of murine NK clones 1097
Fig. 3. Expression of cytolytic granule component genes in the murine NK line B6 NK. Total cellular RNA was extracted and subjected to reverse transcription. In separate PCR reactions the cDNA was amplified using primer pairs specific for the ‘housekeeping’ gene HPRT and for the gene encoding the cytolytic granule component perforin, resulting in products of 190 and 293 bp respectively. Transcripts of genes of the members of the granzyme (Gz) family of cytolytic granule components were amplified using a pair of consensus primers that recognize two conserved sequence motifs, that result in products of the following size: granzyme A, 350 bp; granzyme B, 359 bp; granzyme C, E, F, F and G, referred to as granzyme CEFG, all 362 bp; granzyme D, 374 bp. The detection of all of the RT-PCR products was facilitated by the addition of differently fluorescence-labeled oligonucleotide primers to each of the three PCR reactions. After the amplification the mixtures of each of the three reactions were pooled and separated by high resolution gel electrophoresis in the presence of fluorescent polynucleotide size standards (solid line). Individual bands were analyzed simultaneously in one lane using a DNA sequencing set up.
Analysis of a murine NK cell panel for NK1.1 gene expression Whereas previous analyses demonstrated heterogeneity in NK1.1 cell surface expression on long-term NK lines (35,36), flow cytometry failed to detect significant levels of this marker on our NK panel. To analyze whether the members of the panel express NK1.1 at levels that are beyond the detection limit of flow cytometry, RT-PCR analysis was performed. The results of this analysis shown in Fig. 2 demonstrate that NK 1.1 mRNA was present in PKO (lane 10), PKO 2 (lane 2), PKO 34 (lane 3) and PKO 101 (lane 5), but undetectable in PKO 56 (lane 4) and B6 NK (lane 6).
PCR analysis of our NK cell panel for the rat NK granzymes RNKP1 (56), RNKT2 (59) and Met-ase 1 (60) revealed the absence of these products, that were detected in the rat large granular leukemia cell line RNK-16 (data not shown). To quantitate the levels of gene expression of the granule components, the fluorescence peak areas for each of their RTPCR products were calculated and divided by the peak area obtained for HPRT. The results of this analysis for all members of our NK cell panel and a negative control of murine Hda/1 hybridoma cells (39) are shown in Table 2. The data demonstrate that all of the NK cells expressed similar message levels for granzyme A–G and the pore-forming protein perforin.
Analysis of a murine NK cell panel for gene expression of lymphocyte granule components
Analysis of a murine NK cell panel for cytolytic activity
To further characterize the NK cells, the expression of RNA for lymphocyte granule components was analyzed using a novel RT-PCR. The murine granzymes A, B, C, D, E, F and G, and perforin are stored in the granules of cytotoxic cells and mediate the lysis of target cells (56–58). In the granzyme RT-PCR one set of consensus oligonucleotide primers was used to generate different size products for the various members of this sub-family of serine proteases. In addition, separate amplifications with specific primers for perforin and the ‘housekeeping gene’ HPRT were performed. Figure 3 shows the results of the analysis of B6 NK. Since the RT-PCR products for granzyme C, E, F and G are of an identical size, they result in one single peak, that is referred to as granzyme CEFG. The presence of each of the individual granzymes was verified by analyzing the PCR products generated using specific anti-sense primers (data not shown). In addition, RT-
Since the previous analysis demonstrated the expression of various cytolytic granule components in each of the NK cells, their cytolytic activity was analyzed in a 51Cr-release assay using YAC-1 and P815 as targets, and activated NK cells from a short-term polyclonal culture in a high dose of IL-2, ALAK, as a positive control. A representative example of three individual experiments, shown in Fig. 4, demonstrates that YAC-1, a classic NK target, was lysed by all NK cells, with the exception of PKO 101. Both PKO 34 and B6 NK displayed a high level of cytotoxicity, whereas PKO, PKO 2 and PKO 56 mediated moderate lysis. ALAK and PKO displayed moderate cytotoxicity to P815, while all other NK cells did not kill this target. This pattern of reactivity is in agreement with NK specificity. It should be noted that ALAK represents the reactivity of activated NK cells that lyse YAC-1 at a level that is an order of magnitude higher than freshly explanted NK
1098 Characterization of murine NK clones cells (data not shown). The unusual shape of the curves observed for PKO 2, PKO 34 and B6 NK in the lysis of YAC1 is most probably due to the overcrowding conditions at the highest effector:target ratios caused by the cell size of the cloned NK cells.
Table 2. All members of the murine NK cell panel express cytolytic granule components Cells
PKO PKO 2 PKO 34 PKO 56 PKO 101 B6 NK Hda/1
Granzyme
Perforin
A
B
CEFG
D
1 1 1 1 1 11 –
1 1 1 1 1 1 –
1111 111 1111 1111 11111 111 –
1 1 1 1 1 1 –
11 11 111 11 111 111 –
The murine NK cell panel and a negative control population of Hda/1 B cell hybridoma cells were analyzed as described in Fig. 3 for B6 NK. The fluorescence peak areas of the individual granule components were divided by the peak area obtained for HPRT in order to calculate the relative level of gene expression, which is represented by the following semi-quantitative scale: 1, 1–5; 11, 5– 10; 111, 10–15; 1111, 15–20; and 11111, .20. These data are representative of three individual experiments.
Analysis of NK cell mediated enhancement of Ig secretion in purified B cells We have previously described an in vitro model for responses to TI-2 antigens, as exemplified by polysaccharides, that employs a dextran-conjugated anti-IgD mAb (αδ–dex) (61). This reagent provides a multivalent anti-Ig complex, which is a powerful B cell activator (41). Using this TI-2 model it was demonstrated that depletion of asialo-GM11 cells from a B cell-enriched population of T cell-depleted spleen cells eliminated the Ig secretion induced by αδ–dex and IL-2 (23). Moreover, it was shown that addition of cells from short-term NK cultures was able to induce Ig secretion in αδ–dex 1 IL2-stimulated highly purified B cells (23,24). To test the ability of the members of our NK panel to induce Ig secretion, B cells were highly purified by electronic cell sorting and cultured with αδ–dex 1 IL-2 in the presence of 3% NK cells. Figure 5 shows the results of a spot-ELISA of IgM-secreting cells induced after 5 days of culture. The data demonstrate that all NK cells were able to enhance Ig secretion in αδ– dex 1 IL-2-stimulated highly purified B cells, with the best stimulation observed for PKO 101, 18-fold background, and the weakest for PKO 56 and B6 NK, 3-fold. This pattern, of PKO 101 being the strongest inducer among the clones and PKO 56 the weakest, has proven consistent in five individual experiments. Discussion This paper describes the generation of a panel consisting of one long-term NK line from p53-KO mice, four NK clones
Fig. 4. Members of a murine NK cell panel lyse YAC-1 target cells, but display poor cytotoxicity to P815 targets. YAC-1 and P815 target cells were labeled with [51Cr]sodium chromate and incubated with the NK cells at the indicated effector:target cell ratios for 4 h. Supernatants were analyzed using a γ-counter and the percent lysis was calculated. These data are representative of three individual experiments.
Characterization of murine NK clones 1099 derived from this line and a long-term line generated from normal murine splenocytes. The members of this NK panel augment Ig secretion in an in vitro model for TI-2 responses and, with the exception of PKO 101, display NK cytotoxicity. The strategy that we used to produce our NK panel was based on a method to generate long-term murine NK lines from p53-KO mice, recently described by Karlhofer et al. (35). Our success in isolating murine NK clones from limiting dilution cultures seeded with 0.3 cells/well and in generating a long-term NK line from normal spleen may be based on the use of the NK stimulator poly(IC). Although high concentrations of IL-2 are capable of inducing NK cells in short-term cultures, Karlhofer et al. (35) described that IL-2 by itself was insufficient in generating stable long-term NK lines. Thus, in the process of generating our NK panel several culture conditions were probed: culture with IL-2 by itself, or in addition to fucoidan, in the presence of chondroitin C or in combination with poly(IC). Fucoidan was chosen because it has been described to bind the NK surface marker Ly-49A (62) and chondroitin C because of its affinity for NKR-P1 (28). Although all of these culture conditions resulted in the induction of large granular lymphocytes after 2 weeks of culture and generated enough cells for limiting dilution cloning, it was only those cultures that were set up in the presence of both IL-2 and poly(IC) that resulted in NK cells with longterm in vitro stability. Another factor that we found of critical importance in the generation of our NK panel was the density of the cell culture during the initial polyclonal establishment of the lines. Both lines resulted from cultures that were seeded with a high density of cells, that caused ‘overcrowded’ conditions prior to the limiting dilution cloning. The addition
Fig. 5. Members of a murine NK cell panel show panel varying degrees of enhancement of Ig secretion in an in vitro model for TI-2 responses. Highly purified B cells were isolated by electronic cell sorting from preparations of B cell-enriched CBA.C57BL/6 F1 splenocytes. B cells were cultured at a concentration of 105 cells/ml in a final volume of 200 µl culture medium containing IL-2 (50 U/ml) and αδ–dex (10 ng/ml). NK cells were added at a final concentration of 600 NK cells/well. The B cells were harvested after 5 days of culture and analyzed for IgM-secreting cells (IgMSC) using a spotELISA. These data are representative of five individual experiments.
of poly(IC) to the culture medium proved only critical in the initial 3 months of the generation of the NK panel, since repeated functional tests on the established clones have revealed no phenotypic or functional differences between cells cultured in presence or absence of poly(IC). In a comparison of the surface phenotype of the cells in our NK panel with the KY lines described by Karlhofer et al. (35) there is similarity in the absence of LY-49 and CD3, but difference in the expression of NK1.1. Whereas the KY lines express NK1.1 at different levels, this marker was not detected on the surface of the cells of our NK panel. RT-PCR analysis revealed that some of the members of our panel did express NK1.1 mRNA, suggesting that the levels of surface expression of this antigen may be too low for detection by flow cytometry. This difference between our panel and the KY lines could be explained by a difference in culture conditions, since the KY lines were generated in wells coated with anti-NK1.1, which proved to be a crucial factor. In a comparison of our NK panel with the FT and FL lines that Manoussaka et al. (36) derived from mouse fetal liver, the most obvious difference again lies in the expression of NK1.1. Although it has become increasing clear that NK1.1 can no longer be considered as a marker expressed exclusively by NK cells (42), this report presents the first description of cells that possess phenotypical and functional characteristics of NK cells, but lack detectable NK1.1 cell surface expression. This finding can be interpreted in different ways. It is possible that the conditions that resulted in the establishment of our panel induced phenotypic changes, similar to the absence of Ly-49 observed in both other NK panels (35,36). However, it could also be speculated that our culture conditions selected for a relatively minor NK subpopulation. All of our NK cells express the NK-TR gene product, which is required for NK-type killing and a panel of cytolytic enzymes that mediate cytotoxicity. Except for the PKO 101 clone all members of our NK panel lysed the NK target YAC-1, whereas only the PKO line displayed cytotoxicity towards P815. The data of the YAC-1 lysis experiment shown in Fig. 4 indicate some level of heterogeneity in cytotoxicity, with PKO 34 and B6 NK as the best killers, but none of the analyses described in this communication indicate the cause of the observed differences. Our data support the findings by Karlhofer et al. (35) that NK cells do not need to express Ly-49 molecules in order to be cytotoxic. All members of the NK panel enhanced Ig secretion in an in vitro model for TI-2 responses (61), albeit with a distinct level of heterogeneity. These results extend our previous findings based on the use of polyclonal NK populations (23,24) and preliminary data from ongoing experiments aimed at identification of the mechanism responsible for the observed heterogeneity in the enhancement of Ig secretion suggest the importance of the level of IFN-γ production. The data generated by the PKO 2, PKO 34 and PKO 56 clones provide the first direct evidence that a single NK cell can be both a cytotoxic effector cell and have immunoregulatory activity. Moreover, our findings suggest no correlation between the degree of helper activity of the NK clones and the level of their cytotoxic activity on YAC-1 targets. The current number of clones does not allow us to draw any conclusions on the existence of heterogeneous subpopula-
1100 Characterization of murine NK clones tions among NK cells, which is also being obscured by the absence of members of the Ly-49 family, which have been described as distinguishing markers among NK subpopulations (31–34), but efforts are currently being made at establishing a more extensive NK panel. Manoussaka et al. (63) have recently demonstrated heterogeneity in the surface phenotype within clonal fetal NK cell populations and variation of the phenotype during culture. Both surface phenotype, level of cytotoxicity and level of enhancement of Ig secretion of the members of our NK panel were first analyzed at 3 months after their generation, and have proven stable over a 2 year period. The difference in the stability of the phenotype of both types of NK clones is most likely caused by their respective origin: we used adult spleen cells, whereas Manoussaka et al. derived their clones from fetal thymus, which may contain less differentiated NK precursor cells, that display the observed continuous variation of the phenotype during culture. The availability of a panel that mediates the observed range of Ig production in highly purified B cells stimulated with αδ– dex and IL-2 allows a detailed study of the mechanism responsible for the responses in this in vitro model, and may yield an improved understanding of the processes involved in the in vivo response to TI-2 antigens.
9
10
11 12
13
14 15 16
Acknowledgements The authors thank Dr Andrew Lees for providing selected reagents and Mark Moorman for electronic cell sorting. This work was supported by grants from the National Institutes of Health (AI 33411 and AI 32560). Opinions and assertions herein are the private ones of the authors and are not to be construed as official or reflecting the views of the Department of Defense or the Uniformed Services University of the Health Sciences.
17
18 19
Abbreviations αδ–dex PE pKO-53 TI-2
dextran-conjugated anti-IgD mAb phycoerythrin mice homozygous for a mutation in the p53 tumor suppressor gene T cell-independent type 2
References 1 Herberman, R. B., Reynolds, C. W. and Ortaldo, J. R. 1986. Mechanism of cytotoxicity by natural killer (NK) cells. Annu. Rev. Immunol. 4:651. 2 Trinchieri, G, 1989. Biology of natural killer cells. Adv. Immunol. 47:187. 3 Yu, Y. Y. L, Kumar, V. and Bennett, M. 1992. Murine natural killer cells and marrow graft rejection. Annu. Rev. Immunol. 10:189. 4 Wolfe, S. A., Tracey, D. E. and Henney, C. S. 1976. Induction of ‘natural killer cells’ by BCG. Nature 262:584. 5 Ojo, E., Haller, O., Kimuna, A. and Wigzell, H. 1978. An analysis of conditions allowing Corynebacterium parvum to cause either augmentation or inhibition of natural killer cell activity against tumor cells in mice. Int. J. Cancer. 21:444. 6 Eugui, E. M. and Allison, A. C. 1980. Differences in susceptibility of various mouse strains to heamoprotozoan infections: possible correlation with natural killer cell activity. Parasite Immunol. 2:277. 7 Hatcher, F. M. and Kuhn, R. E. 1981. Spontaneous lytic activity against allogeneic tumor cells and depression of specific cytotoxic responses in mice infected with Trypanosoma cruzi. J. Immunol. 126:2436. 8 Welsh, R. M. Jr., Zinkernagel, R. M. and Hallenbeck, L. A. 1979.
20 21 22 23 24 25 26
27 28 29
Cytotoxic cells induced during choriomeningitis lymphocytic virus infection of mice. II. Specificities of the natural killer cells. J. Immunol. 122:475. Bukowski, J. F., Woda, B. A., Habu, S., Okumura, K. and Welsh, R. M. Jr. 1983. Natural killer cell depletion enhances virus synthesis and virus-induced hepatitis in vivo. J. Immunol. 131:1531. Cuturi, M. C., Anegon, I., Sherman, F., Loudon, R., Clark, S. C., Perussia, B. and Trinchieri, G. 1989. Production of hematopoietic colony-stimulating factors by human natural killer cells. J Exp. Med. 169:569. Young, H. A. and Ortaldo, J. R. 1987. One-signal requirement for interferon-γ production by human large granular lymphocytes. J. Immunol. 139:724. Scala, G., Allavena, P., Djeu, J. Y., Kasahara, T., Ortaldo, J. R., Herberman, R. B. and Oppenheim, J. J. 1984. Human large granular lymphocytes are potent producers of interleukin-1. Nature 309:56. Anegon, I., Cuturi, M. C., 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. Smyth, M. J., Zachaaiae, C. D., Norihisa, Y., Ortaldo, J. R., Hishinuma, A. and Matsushima, K. 1991. IL-8 gene expression and production in human PBL subsets. J. Immunol. 146:3815. Naume, B., Gately, M. and Espevik, T. 1992. A comparative study of IL-12, IL-2 and IL-7 induced effects on immunomagnetically purified CD561 NK cells. J. Immunol. 148:2429. Scharton, T. M. and Scott, P. 1993. Natural killer cells are a source of interferon γ that drives differentiation of CD41 T cell subsets and induces early resistance to Leishmania major in mice. J. Exp. Med. 178:567. Sher, A., Oswald, I. P., Hieny, S. and Gazzinelli, R. T. 1993. Toxoplasma gondii induces a T cell-independent IFN-γ response in NK cells that requires both adherent accessory cells and TNFα. J. Immunol. 150:3982. Laskay, T., Rollinghoff, M. and Solbach, W. 1993. Natural killer cells participate in the early defense against Leishmania major infection in mice. Eur. J. Immunol. 23:2237. Brenner, M. K., Vyakarnam, A., Reittie, J. E., Wimperis, J. Z., Grob, J. P., Hoffbrand, A. V. and Prentice, H. G. 1987. Human large granular lymphocytes induce immunoglobulin synthesis after bone marrow transplantation. Eur. J. Immunol. 17:43. Wilder, J. A., Koh, C. Y. and Yuan, D. 1996. The role of NK cells during in vivo antigen-specific antibody responses. J. Immunol. 156:146. Michael, A., Shao, A. and Yuan, D. 1991. Productive interactions between B and natural killer cells. Nat. Immun. Cell Growth Regul. 10:71. Yuan, D., Wilder, J., Dang, T., Bennett, M. and Kumar, V. 1992. Activation of B lymphocytes by NK cells. Int. Immunol. 4:1373. Snapper, C. M., Yamaguchi, H., Moorman, M. A., Sneed, R., Smoot, D. and Mond, J. J. 1993. Natural killer cells induce activated murine B cells to secrete Ig. J. Immunol. 151:5251. Snapper, C. M., Yamaguchi, H., Moorman, M. A. and Mond, J. J. 1994. An in vitro model for T cell-independent induction of humoral immunity. A requirement for NK cells. J. Immunol. 152:4884. Michael, A., Hackett, J. J., Bennett, M., Kumar, V. and Yuan, D. 1989. Regulation of B lymphocytes by natural killer cells: role of IFN-γ. J. Immunol. 142:1095. Snapper, C. M., Rosas, F., Moorman, M. A., Jin, L., Shanebeck, K., Klinman, D. M., Kehry, M. R., Mond, J. J. and Maliszewski, C. R. 1996. IFN-γ is a potent inducer of Ig secretion by sort-purified murine B cells activated through the mIg, but not the CD40, signaling pathway. Int. Immunol. 8:877. Snapper, C. M. and Paul, W. E. 1987. Interferon-γ and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 236:944. Yokoyama, W. M., 1995. Natural killer cell receptors specific for major histocompatibility complex class I molecules. Proc. Natl Acad. Sci. USA 92:3081. Bezouska, K., Yuen, C.-T., O’Brien, J., Childs, R. A., Chai, W.,
Characterization of murine NK clones 1101
30
31
32
33
34
35
36
37 38
39
40 41
42 43 44
45 46
Lawson, A. M., Drbal, K., Fiserova, A., Pospisil, M. and Feizi, T. 1994. Oligosaccharide ligands for NKR-P1 protein activate NK cells and cytotoxicity. Nature 372:150. Malnati, M. S., Peruzzi, M., Parker, K. C., Biddison, W. E., Ciccone, E., Moretta, A. and Long, E. O. 1995. Peptide specificity in the recognition of MHC class I by natural killer cell clones. Science 267:1016. Brennan, J., Mager, D., Jefferies, W. and Takei, F. 1994. Expression of different members of the Ly-49 gene family defines distinct natural killer cell subsets and cell adhesion properties. J. Exp. Med. 180:2287. Stoneman, E. R., Bennett, M., An, J., Chesnut, K. A., Wakeland, E. K., Scheerer, J. B., Siciliano, M., Kumar, V. and Mathew, P. A. 1995. Cloning and characterization of 5E6 (Ly-49C), a receptor molecule expressed on a subset of murine natural killer cells. J. Exp. Med. 182:305. Murphy, W. J., Raziuddin, A., Mason, L., Kumar, V., Bennett, M. and Longo, D. L. 1995. NK cell subsets in the regulation of murine hematopoiesis. I. 5E61 NK cells promote hematopoietic growth in H-2d strain mice. J. Immunol. 155:2911. Brennan, J., Lemieux, S., Freeman, J. D., Mager, D. L. and Takei, F. 1996. Heterogeneity among Ly-49C natural killer (NK) cells: characterization of highly related receptors with differing functions and expression patterns. J. Exp. Med. 184:2085. Karlhofer, F. M., Orihuela, M. M. and Yokoyama, W. M. 1995. Ly49-independent natural killer (NK) cell specificity revealed by NK cell clones derived from p53-deficient mice. J. Exp. Med. 181:1785. Manoussaka, M., Georgiou, A., Rossiter, B., Shrestha, S., Toomey, J. A., Sivakumar, P. V., Bennett, M., Kumar, V. and Brooks, C. G. 1997. Phenotypic and functional characterization of long-lived NK cell lines of different maturational status obtained from mouse fetal liver. J. Immunol. 158: 112. Bix, M., and Locksley, R. M. 1995. Natural T cells: cells that coexpress NKRP-1 and TCR. J. Immunol. 155:1020. Donehower, L. A., Harvey, M., Slagle, B. L., MacArthur, M. J., Montgomery, C. A., Jr, Butel, J. S. and Bradley, A. 1992. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumors. Nature 356:215. Zitron, I. M. and Clevinger, B. L. 1980. Regulation of murine B cells through surface immunoglobulin. I. Monoclonal anti-δ antibody that induces allotype-specific proliferation. J. Exp. Med. 152:1135. Stall, A. M. and Loken, M. R. 1984. Allotypic specificities of murine IgD and IgM recognized by monoclonal antibodies. J. Immunol. 132:787. Brunswick, M., Finkelman, F. D., Highet, P. F., Inman, J. K., Dintzis, H. M. and Mond, J. J. 1988. Picogram quantities of anti-Ig antibodies coupled to dextran induce B cell proliferation. J. Immunol. 140:3364. Coffman, R. L. and Weissman, I. L. 1981. B220: a B cell-specific member of the T200 glycoprotein family. Nature 289:681. Ledbetter, J. A. and Herzenberg, L. A. 1979. Xenogeneic monoclonal antibodies to mouse lymphoid differentiation antigens. Immunol. Rev. 47:63. Janeway, C. A., Jr, Conrad, P. J., Lerner, E. A., Babich, J., Wettstein, P. and Murphy, D. B. 1984. Monoclonal antibodies specific for Ia glycoproteins raised by immunization with activated T cells: possible role of T cell bound Ia antigens as targets of immunoregulatory T cells. J. Immunol. 132:662. Chomczynski, P. and Sacchi, N. 1987. Single step method of RNA isolation by acid guanidium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162:156. Anderson, S. K., Gallinger, S., Roder, J., Frey, J., Young, H. A. and Ortaldo, J. R. 1993. A cyclophilin-related protein involved in the function of natural killer cells. Proc. Natl Acad. Sci. USA 90:542.
47 Chambers, C. A., Gallinger, S., Anderson, S. K., Giardina, S., Ortaldo, J. R., Hozumi, N. and Roder, J. 1994. Expression of the NK-TR gene is required for NK-like activity in human T cells. J. Immunol. 152:2669. 48 Mason, L., Giardina, S. L., Hecht, T., Ortaldo, J. and Mathieson, B. J. 1988. LGL-1: a non-polymorphic antigen expressed on a major population of mouse natural killer cells. J. Immunol. 140: 4403. 49 Vos, Q., Claassen, E. and Benner, R. 1990. Substantially increased sensitivity of the spot-ELISA for the detection of anti-insulin antibody-secreting cells using a capture antibody and enzyme conjugated insulin. J. Immunol. Methods 126:89. 50 Frey, J. L., Bino, T. Kantor, R. R. S., Segal, D. M., Giardina, S. L., Roder, J., Anderson, S. and Ortaldo, J. R. 1991. Mechanism of target cell recognition by natural killer cells: characterization of a novel triggering molecule restricted to CD3– large granular lymphocytes. J. Exp. Med. 174:1527. 51 Young, H. A., Jenkins, N. A., Copeland, N. G., Simek, S., Lerman, M. I., Zbar, B., Glenn, G., Ortaldo, J. R. and Anderson, S. K. 1993. Localization of a novel natural killer triggering receptor locus to human chromosome 3p23–p21 and mouse chromosome 9. Genomics 16:548. 52 Hercend, T., Reinherz, E. L., Meuer, S. Schlossman, S. F. and Ritz, J. 1983. Phenotypic and functional heterogeneity of human cloned NK cell lines. Nature 301:158. 53 Brooks, C. G. 1983. Reversible induction of natural killer cell activity in cloned murine cytotoxic T lymphocytes. Nature 305:155. 54 Brooks, C. G., Holscher, M. and Urdal, D. 1985. Natural killer activity in cloned cytotoxic T lymphocytes: regulation by IL-2, IFN, and specific antigen. J. Immunol. 135:1145. 55 Shortman, K. and Wilson, A. 1986. Natural and unnatural killing by cytotoxic T lymphocytes. Curr. Top. Microbiol. Immunol. 126:111. 56 Trapani, J. A., Smyth, M. J., Apostolidis, V. A., Dawson, M. and Browne, K. A. 1994. Granule serine proteases are normal nuclear constituents of natural killer cells. J. Biol. Chem. 269:18359. 57 Smyth, M. J. and Trapani, J. A. 1995. Granzymes: exogenous proteinases that induce target cell apoptosis. Immunol. Today 16:202. 58 Shresta, S., MacIvor, D. M., Heusel, J. W., Russell, J. H. and Ley, T. J. 1995. Natural killer and lymphokine-activated killer cells require granzyme B for the rapid induction of apoptosis in susceptible target cells. Proc. Natl Acad. Sci. USA 92:5679. 59 Sayers, T. J., Wiltrout, T. A., Smyth, M. J., Ottaway, K. S., Pilaro, A. M., Sowder, R., Henderson, L. E., Sprenger, H. and Lloyd, A. R. 1994. Purification and cloning of a novel serine protease, RNKTryp-2, from the granules of a rat NK cell leukemia. J. Immunol. 152:2289. 60 Smyth, M. J., Wiltrout, T., Trapani, J. A., Ottaway, K. S., Sowder, R., Henderson, L. E., Kam, C.-M., Powers, J. C., Young, H. A. and Sayers, T. J. 1992. Purification and cloning of a novel serine protease, RNK-Met-1, from the granules of a rat natural killer cell leukemia. J. Biol. Chem. 267:24418. 61 Pec¸anha, L. M. T., Snapper, C. M., Finkelman, F. D. and Mond, J. J. 1991. Dextran-conjugated anti-Ig antibodies as a model for T cell-independent type 2 antigen-mediated stimulation of Ig secretion in vitro. I. Lymphokine dependence. J. Immunol. 146:833. 62 Daniels, B. F., Nakamura, M. C., Rosen, S. D., Yokoyama, W. M. and Seaman, W. E. 1994. Ly-49A, a receptor for H-2Dd, has a functional carbohydrate recognition domain. Immunity 1:785. 63 Manoussaka, M. S., Smith, R. J., Conlin, V., Toomey, J. A. and Brooks, C. G. 1998. Fetal mouse NK cell clones are deficient in Ly49 expression, share a common broad lytic specificity, and undergo continuous and extensive diversification in vitro. J. Immunol. 160:2197.
