British Journal of Haematology, 2002, 118, 885–892
Human natural killer cell development in a xenogeneic culture system Isabel Bara˜ o, 1 Fa´ tima Vaz, 2 Grac¸ a Almeida-Porada, 1 Edward F. Srour, 3 Esmail D. Zanjani 1 and Joa˜ o L. Ascensa˜ o 1 1University of Nevada School of Medicine, V.A. Medical Center, Reno, NV, USA, 2Portuguese Institute of Oncology, Lisbon, Portugal, and 3Indiana University School of Medicine, Indianapolis, IN, USA Received 20 December 2001; accepted for publication 20 March 2002
Summary. In vivo and in vitro xenogeneic models have shown the ability of a non-human environment in supporting human haemopoiesis. In the present study, we evaluated the effect of fetal sheep thymic stroma in the in vitro development of natural killer (NK) cells from human haemopoietic progenitors. CD34+HLA-DR+ (CD34+ DR+)Lin– and CD34+DR–Lin– bone marrow (BM) progenitors were cultured for 3 weeks with or without interleukin 2 (IL-2), in fetal sheep thymic stroma contact and transwell cultures. Both progenitors gave rise to NK cells, defined as CD45+CD56+ cells, in the presence or absence of IL-2; however, the percentage of NK cells originated in cultures with IL-2 was significantly higher. Direct contact with stroma seemed to be required for the most immature
progenitors, CD34+DR– Lin–, to differentiate along the NK cell lineage. Functional assays revealed that only cells grown in the presence of IL-2 were cytolytic against K562 targets and, curiously, NK cells derived from CD34+DR– Lin– progenitors were more cytotoxic that NK cells derived from CD34+DR+Lin– progenitors. These studies suggest that the ability of fetal sheep thymic stroma in promoting the generation of human NK cells from haemopoietic progenitors may have relevance in terms of NK cell ontogeny and induction of tolerance in transplantation.
Human natural killer (NK) cells are CD56+CD3– lymphocytes (Trinchieri, 1989) characterized by non-major histocompatibility complex (MHC)-restricted cytotoxicity against a wide variety of target cells, including tumour cells, and cells infected with bacteria and viruses (Herberman et al, 1979; Welsh, 1981). NK cells can be generated in vitro from human BM progenitors in stroma-dependent cultures (Miller et al, 1992, 1994; Tjonnfjord et al, 1995; Vaz et al, 1998a) and in stroma-independent cultures, when supplemented with appropriate growth factors (Lotzova & Savary, 1993; Silva et al, 1994). The stromal microenvironment plays an important role in the differentiation and proliferation of haematopoietic progenitors either by the production of growth factors, and/or by providing a functional support for cell–cell contact between progenitor and stromal cells or their extracellular matrix (Cashman et al, 1985). Xenogeneic models of long-term bone marrow cultures (LTBMC) have demonstrated that stroma or stromal cell lines
of non-human origin are capable of supporting the differentiation of human haemopoietic stem cells (HSC) in vitro (Paul et al, 1991; Issaad et al, 1993; Almeida-Porada et al, 1996; Miller et al, 1999). The ability of a non-human haematopoietic microenvironment to support human haemopoiesis in vivo is demonstrated by the successful engraftment and multilineage differentiation of human HSC in adult severe combined immunodeficient (SCID) mice (McCune et al, 1988), normal mice (Pallavicini et al, 1992; Pixley et al, 1994) and sheep (Zanjani et al, 1991, 1992a,b, 1995a,b; Srour et al, 1992, 1993). In the sheep model, the resulting chimaeric recipients exhibit long-term multilineage (erythroid, myeloid, lymphoid) expression of donor cells and remain responsive to human-specific cytokines. Several lines of evidence support the notion that NK cells can develop in the thymus. The demonstration of the existence of bipotential T/NK progenitors, committed NK precursors, and mature NK cells in the thymus (Phillips & Lanier, 1987; Ramsdell & Golub, 1987; Michon et al, 1988; Denning et al, 1991; Hori & Spits, 1991; Lanier et al, 1992: Rodewald et al, 1992; Matsuzaki et al, 1993; Sanchez et al, 1993, 1994; Barcena et al, 1994; Spits et al, 1995) and the generation of NK cells in murine and human thymic organ
Correspondence: Joa˜o L. Ascensa˜o, M.D., Ph.D., George Washington University Medical Center, Department of Hematology/Oncology, 2150 Pennsylvania Avenue, NW, Suite 3-428, Washington, DC 20037, USA. E-mail:
[email protected] 2002 Blackwell Science Ltd
Keywords: human haemopoietic progenitors, sheep, thymic stromal cells, NK cell development, tolerance.
885
886
I. Bara˜o et al
culture systems (Tjonnfjord et al, 1995; Aiba et al, 1997; Vaz et al, 1998a) indicate that the thymic microenvironment supports NK cell development. In the present study, we examined the ability of fetal sheep thymic stroma to support the in vitro development of human NK cells from human CD34+DR+Lin– and CD34+DR–Lin– bone marrow (BM) progenitors, in stroma contact and transwell cultures, with or without interleukin 2 (IL-2). Our data show that fetal sheep thymic stroma promotes the in vitro generation of human NK cells from haemopoietic progenitors; however, the addition of IL-2 to the cultures resulted in a substantial and significant increase in the number of cytotoxic NK cells. In this xenogeneic culture system, direct contact with stroma seemed to be required for the most immature progenitors, CD34+DR–Lin–, to differentiate along the NK cell lineage. Additional studies investigating the role of the thymus in NK development will be important to understand the mechanisms of tolerance induction in transplantation. MATERIALS AND METHODS Isolation of human BM progenitors. BM was aspirated from the posterior iliac crest of normal adult volunteers after informed consent was obtained. BM mononuclear cells were isolated by Ficoll–Hypaque (Sigma, St. Louis, MO, USA) centrifugation. CD34+ cells were purified by positive immunomagnetic selection using a CD34 isolation kit according to the manufacturer’s instructions (Miltenyi Biotec, Auburn, CA, USA). Cell sorting was performed on a Becton Dickinson FACScan (Becton Dickinson, San Jose, CA, USA), and appropriate controls, including fluorescein isothiocyanate (FITC)-, phycoerythrin (PE)- and PC-5-conjugated isotype-matched immunoglobulins (Igs), were used. BM progenitors were sorted using the following cocktail of monoclonal antibodies (mAbs): FITC-conjugated anti-CD2, CD3, CD4, CD7, CD8, CD14, CD15, CD16, CD19, CD56, CD57 (all from Becton Dickinson) and glycophorin A (Amac, Westbrook, ME, USA), PE-conjugated anti-HLA-DR (Becton Dickinson) and PC-5-conjugated anti-CD34 (Beckman Coulter, Palantine, IL, USA). Re-analysis of the sorted CD34+DR+Lin– and CD34+DR–Lin– populations was done to ensure their purity. CD56+ cells were not detected in these cell populations. Establishment of thymic stromas. Thymic stromal cell cultures were initiated by mincing thymic samples obtained from fetuses of Dorsett Merino ewes (age range, 54–72 d old; term: 145 d). Small thymic fragments were washed twice in Iscove’s-modified Dulbecco’s medium (IMDM; Gibco, Grand Island, NY, USA) supplemented with 5% heat-inactivated fetal bovine serum (FBS; Sigma) and cultured in 25 cm2 flasks with IMDM supplemented with 10% heat inactivated FBS, 10)10 mol/l cholera toxin (Gibco) and 1% penicillin G sodium (10 000 U/ml)-streptomycin sulphate (10 000 mg/ml) (Gibco) hereafter referred to as thymic media (TM). Cultures were maintained in a humidified atmosphere at 37C and 5% CO2 and the culture medium was replaced weekly with fresh TM. Upon 75% confluence, non-adherent cells were removed by repeated
washing with 5% FBS in phosphate-buffered saline (PBS; Sigma), and adherent cells detached with trypsin-EDTA (Sigma) and irradiated at 1500 cGy with a cobalt source. After irradiation, stromal cells were subcultured in 12 (transwell) and 96 (stroma contact) well plates (Costar, Cambridge, MA, USA) at 1Æ5 · 105 and 1Æ25 · 104 cells/ well respectively. Immunofluorescence microscopy. To identify the cell types present in the adherent cell layer, adherent cells grown in Lakslides (Costar) were fixed in methanol, blocked by incubation with 5% bovine serum albumin (BSA) (Sigma) for 30 min at 4C and incubated with the following monoclonal antibodies for 30 min: anticytokeratin (Becton Dickinson), antilaminin, antivimentin, anticellular fibronectin, anti-smooth-muscle actin (Sigma), collagen IV and antifactor VIII (Caltag, Burlingame, CA, USA). After twice washing with PBS containing 5% FBS, the slides were incubated in the dark with fluorescein-conjugated antimouse polyvalent immunoglobulins (IgA, IgG, IgM) (Sigma) for 30 min at room temperature. After duplicate washes, the slides were examined using an Olympus IX70 fluorescent microscope. Culture conditions. Sorted haemopoietic BM progenitors (2Æ5 · 104 and 1 · 104) were plated directly onto the irradiated adherent thymic stromal cells or separated by a Transwell membrane (Costar), in 12 and 96 well plates respectively. IMDM was used as culture media supplemented with 10% heat-inactivated human AB serum (North American Biologicals, Miami, FL, USA) and 1% penicillin G sodium (10 000 U/ml)-streptomycin sulphate (10 000 mg/ml), with or without 1000 U/ml of recombinant human IL-2 (Peprotech, Rocky Hill, NJ, USA). The remainder of the cells were cultured without cytokines. For each type of culture, three separate experiments were set up in triplicate. Cultures were subjected to half media changes twice weekly. After 3 weeks of culture, cells were recovered by repeated washing, counted and their viability determined by trypan blue exclusion. Proliferation of cultured cells. The proliferation index (PI) was calculated by dividing the total number of cells at time of analysis by the number of cells plated onto stroma or on transwell membranes on d 0 as shown in the following equation: PI ¼ total number of cells at time of analysis/ number of progenitors plated at d 0 Flow cytometric analysis of fetal sheep thymic stromal cells. Cell-surface antigen expression was evaluated by flow cytometry using a FACScan flow cytometer (Becton Dickinson). To analyse if fetal sheep thymic stromal cells crossreacted with human CD markers, samples of stromal cells were washed twice with 5% FBS in PBS and blocked with 5% BSA in PBS at 4C for 30 min. After washing, direct staining of the samples was performed with the following mAbs purchased from Becton Dickinson: FITC-conjugated anti-CD3, CD4, CD5, CD7, CD10, CD16 and CD25, and PE-conjugated anti-CD2, CD8, CD14, CD33, CD45, CD56 and CD62L. Appropriate controls included FITC- and
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
NK Cell Development in Xenogeneic Culture PE-conjugated isotype-matched Igs. After incubation, the cells were washed three times in PBS with 0Æ5% sodium azide and fixed with 1% formaldehyde. Cells were considered positive for the tested antigen if fluorescence intensity was above the upper limit of the negative control. Flow cytometric analysis of cultured cells. For cultured cells, direct staining for 15 min was performed with the following mAbs from Becton Dickinson: FITC-conjugated anti-CD2, CD3, CD4, CD7, CD16, CD19 and CD56; PE-conjugated mAbs CD3, CD8 and CD45, and the peridinin chlorophyll (PerCP)-conjugated mAbs anti-CD3 and CD45. As before, controls included FITC-, PE- and PerCP-conjugated isotype-matched Igs. After twice washing the samples in IMDM supplemented with 5% FBS they were labelled with a saturating concentration of mAb for 15 min at room temperature in the dark, washed twice in PBS with 0Æ5% sodium azide and fixed with 1% formaldehyde. At least 10 000 cells per aliquot were analysed and gated on the presumptive lymphocyte region as defined by forward and side scatter. NK cells were phenotypically defined as CD45+CD56+ cells. Cytotoxicity assays. Cells harvested from our cultures were tested for cytotoxicity against the NK-sensitive erythroleukaemia cell line K562 in a standard 4 h 51Cr release assay. A total of 1 · 106 target cells were washed and incubated for 90 min at 37C with Na2 51CrO2 (NEN Life Science, Pittsburgh, PA, USA) at 3Æ7 MBq/106 target cells. The cells were washed five times in IMDM supplemented with 5% FBS and counted. Effector cells harvested from the cultures on the day of analysis were washed, counted, assessed for their viability by Trypan blue exclusion and seeded in V-shaped microwell plates (Applied Scientific, San Francisco, CA, USA) at effector:target (E:T) ratios that ranged from 5:1 to 0Æ6:1. The plates were then centrifuged at 120 g for 3 min and incubated for 4 h at 37C in a 5% CO2-humidified air atmosphere. After this period, the plates were centrifuged at 200 g, and 0Æ1 ml of the supernatants was removed from each well and withdrawn into aliquots of 1 ml of liquid scintillation cocktail (Wallac, Gaithersburg, MD, USA). Radioactivity was measured in a scintillation counter 1450 Microbeta (Wallac). All determinations were done in triplicate, and percentage lysis was determined using the following equation: Specific lysis percentage ¼ Experimental mean cpm spontaneous release mean cpm 100=Total release mean cpm spontaneous release mean cpm Total 51Cr release was determined by adding 0Æ1 ml of 1% sodium dodecyl sulphate solution (SDS; Sigma) to labelled target cells. Spontaneous 51Cr release, as determined by adding 0Æ1 ml of supplemented medium to target cells, averaged 15%. Statistical analysis. Data are presented as mean ± SEM. To evaluate the statistical significance of the difference between two groups, a Student’s t-test unequal variance was used. A P-value of < 0Æ05 was considered to be statistically significant.
887
RESULTS Characterization of fetal sheep thymic stromal cells The proliferation and differentiation of haematopoietic progenitor cells is dependent on the interactions between the stem cells and the matrix components of the microenvironment (Cashman et al, 1985). Microscopic evaluation of fetal sheep thymic stromas has shown that most adherent cells had an elongated, fibroblastoid morphology; seeded among these, we observed nests of polygonal, epithelioid cells with a large nucleus. Using immunofluorescence, we examined if stromal cell proteins, which play a role in the maintenance of human haemopoiesis in vitro, were present in fetal sheep thymic stromas. Vimentin, fibronectin, cytokeratin and a-smooth muscle actin were present in sheep stromas, but not collagen IV, laminin or Factor VIII. Flow cytometric analysis using antihuman monoclonal antibodies revealed that fetal sheep thymic stromal cells did not cross-react with antibodies to human CD2, CD3, CD4, CD5, CD7, CD8, CD10, CD14, CD16, CD25, CD33, CD45, CD56 and CD62L antigens. However, we observed that fetal sheep thymic stromal cells cultured for 3 weeks in IMDM supplemented with 10% of heat-inactivated human AB and 1000 U/ml of IL-2, cross-reacted with the human antigen CD56 (2–25%), showing the existence of some antigenic similarity with the human counterpart. Proliferation of cultured cells As seen in Fig 1, CD34+DR+Lin– and CD34+DR–Lin– BM progenitors cultured for 3 weeks with or without 1000 U/ml of IL-2 in fetal sheep thymic stroma contact and transwell cultures did not exhibit a high cellular proliferation. Proliferation indexes (PIs) ranged between 1Æ77 and 12Æ36, and no significant differences were observed in the PIs between both progenitors or between stroma contact vs transwell cultures. Phenotype of cultured cells The phenotype CD45+CD56+ was used as a criterion to identify human NK cells, as fetal sheep thymic stromal cells cross-react with the human antigen CD56 when stimulated with IL-2, but do not cross-react with the human-specific mAb CD45. After 3 weeks of culture, CD34+DR+Lin– and CD34+DR–Lin– BM progenitors gave rise to CD45+CD56+ cells in the presence or absence of IL-2 (Table I and Fig 2A and B), however, the percentage of NK cells originated in cultures with IL-2 was significantly higher. The purity of the initial cell populations was verified by a second analysis, and no CD56+ cells were detected. When CD34+DR+Lin– progenitors were separated from stromas by a transwell membrane, NK cell development was not significantly different, in contrast to cultures of CD34+DR–Lin– progenitors, which only gave rise to high numbers of CD45+CD56+ cells when cultured in direct contact with stroma and in the presence of IL-2. CD45+ cells originated in this xenogeneic culture system were shown to express low levels of other markers of lymphoid lineage, including CD2 (0Æ8–5%), the T-cell marker CD3 (1–7%), CD7 (2–12%), CD8 (1–4%) and
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
888
I. Bara˜o et al
Fig 1. Growth ability of CD34+DR+Lin– and CD34+DR–Lin– BM progenitors cultured for 3 weeks, with or without 1000 U/ml of IL-2, in fetal sheep thymic stroma contact or transwell cultures. None of the culture types exhibited a significant cellular proliferation. Results are shown as the mean ± SEM of proliferation indexes of three experiments performed in triplicate.
CD16 (2–5%). We did not detect any expression of CD4 or the B-cell marker CD19. Cytotoxicity of cultured cells Although CD45+CD56+ cells were also obtained in cultures without factors, only cells grown in the presence of IL-2 were cytolytic against K562 targets (Fig 3). Stroma contact cultures of CD34+DR–Lin– progenitors gave rise to NK cells that exhibited significantly more cytotoxicity than NK cells grown from CD34+DR+Lin– progenitors (P < 0Æ05) at E:T ratios 5:)0Æ6:1. No significant differences were found in the NK cell cytolytic activity between stroma contact vs transwell cultures, in contrast to cultures of CD34+DR–Lin– progenitors, which only generated cytotoxic NK cells when cultured in direct contact with stroma (P < 0Æ05). Fetal sheep thymic stromal cells cultured with 1000 U/ml of IL-2 for 3 weeks did not show cytolytic activity against K562 target cells (0Æ8–3Æ2%). Table I. CD45+CD56+ cells grown in culture from haemopoietic BM progenitors.*
Type of culture
Factors
CD34+DR+Lin–
CD34+DR–Lin–
Stroma contact Transwell Stroma contact Transwell
IL2
76Æ5 74Æ6 5Æ8 1Æ85
67Æ1 ± 10Æ1§– 5Æ6 ± 2Æ2– 3Æ0 ± 0Æ7§ 5Æ1 ± 0Æ9
None
± ± ± ±
15Æ8 13Æ8 2Æ3 1Æ0
*Percentage of CD45+CD56+ cells generated from CD34+DR+Lin– and CD34+DR–Lin– BM progenitors, after 3 weeks of culture in fetal sheep thymic stroma contact or transwell cultures, with or without IL-2. P ¼ 0Æ022. P ¼ 0Æ016. §P ¼ 0Æ011. –P ¼ 0Æ001. Data are shown as the mean ± SEM of three different experiments.
DISCUSSION Adult human bone marrow CD34+DR– cells possess functional properties commonly associated with HSC (Andrews et al, 1990; Brandt et al, 1990; Verfaillie et al, 1990; Srour et al, 1991). These HSC populations contain primitive haematopoietic progenitor cells capable of initiating and sustaining long-term multilineage haemopoiesis in vitro (Briddell et al, 1992; Verfaillie, 1992; Srour et al, 1993). Srour et al (1993) demonstrated that CD34+DR– BM progenitors give rise to different human lymphohaematopoietic lineages, including NK cells, when transplanted into sheep fetuses in utero. In the present study, we evaluated the effects of fetal sheep thymic stroma on the in vitro NK cell development from human haemopoietic progenitors. In this xenogeneic culture system, CD34+DR+Lin– and CD34+ DR–Lin– BM progenitors gave rise to CD45+CD56+ cells in the presence or absence of IL-2 in fetal sheep thymic stroma contact and transwell cultures; however, IL-2 was required to obtain high numbers of cytotoxic NK cells. The interactions of haematopoietic progenitor cells with both stromal cells and components of the extracellular matrix, in conjunction with signals mediated by growth factors, are important in the differentiation, expansion and survival of haematopoietic progenitors. Early acting cytokines, including stem cell factor (SCF) and fetal liver kinase ligands (flk2L/flt3L), act on HSCs to commit them to the NK lineage with simultaneous induction of expression of the b subunit of the IL-15 receptor (IL2Rb chain) (Williams et al, 1998; Yu et al, 1998). Subsequently, triggering of the IL-2Rb/cc (common gamma chain) receptor complex by either IL-15 or IL-2 is required for the expansion and differentiation of these cells to fully mature NK cells (Williams et al, 1997). Our group (Silva et al, 1994; Vaz et al, 1998b) and other investigators (Lotzova & Savary, 1993; Mrozek et al, 1996; Yu et al, 1998; Miller et al, 1999; Parrish-Novak et al, 2000; Asao et al, 2001) have studied the effects of IL-2, IL-7, IL-15 and IL-21 [cytokines which receptors share the common c (cc) chain of the IL-2R] in the differentiation and cytotoxicity
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
NK Cell Development in Xenogeneic Culture
889
Fig 2. Representative two-colour flow cytometric analysis of cells originated from CD34+DR+Lin– (A) and CD34+DR–Lin– (B) BM progenitors. After 3 weeks of culture, both progenitors gave rise to NK cells defined as CD45+CD56+ cells in fetal sheep thymic stroma contact or transwell cultures, with or without 1000 U/ml of IL-2. Negative controls are shown.
Fig 3. Cytotoxicity of IL-2 grown cells derived from CD34+DR+Lin– and CD34+DR–Lin– BM progenitors was tested against the NK-sensitive target K562. Stroma contact cultures of CD34+DR–Lin– progenitors gave rise to NK cells significantly more cytotoxic than NK cells grown from CD34+DR+Lin– progenitors (P < 0Æ05) at E:T ratios 5:1–0Æ6:1. No significant differences were found in the NK cell cytolytic activity between stroma contact and transwell cultures, with the exception of cultures of CD34+DR–Lin– progenitors, which only generated cytotoxic NK cells when cultured in direct contact with stroma (P < 0Æ05). Results from three experiments performed in triplicate are shown as the mean ± SEM of K562 lysis at each E:T ratio.
of NK cells. Our data indicate that besides IL-2 and IL-15, IL-7 is also able to induce NK cell differentiation from haematopoietic progenitors; however, IL-7 differentiated NK cells are functionally immature. None of the mentioned cytokines or growth factors are capable of supporting NK cell development by themselves (Mrozek et al, 1996; Williams et al, 1997, 1999; Parrish-Novak et al, 2000). Stromal cell cultures of human or non-human origin support differentiation of haematopoietic progenitors into NK cells. Primitive CD34+DR– progenitors can be induced to differentiate along the NK cell lineage when cultured with adult marrow allogeneic stroma (Miller et al, 1992) or with human thymic stroma (Vaz et al, 1998a) in the presence of IL-2. In the absence of stroma, these haematopoietic progenitors can differentiate into NK cells, if soluble factors, such as stem cell factor (SCF), IL-1a and IL-2 are added to the cultures (Silva et al, 1994). Studies of Miller et al (1999) have shown that co-culture of CD34+DR– progenitors on AFT024 (a murine fetal liver stromal cell line) in the presence of SCF, flt3L and IL-7 at culture initiation and IL-2 throughout culture resulted in NK cell progeny. In our xenogeneic cultures, fetal sheep thymic stroma and IL-2 were efficient in promoting the development of phenotypic and functionally mature human NK cells from CD34+DR–Lin– and CD34+DR+Lin– progenitors. Curiously, our stroma-based system was also able to generate a small percentage of CD45+CD56+ cells in the absence of IL-2. It is possible that fetal sheep thymic stroma produce a soluble and/or membrane-bound sheep homologue for flt3L and/or for SCF that interact synergistically with other sheep cytokines and with human cytokines, such as IL-2, to stimulate NK cell development. The limited capacity of fetal
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
890
I. Bara˜o et al
sheep thymic stroma in promoting, by itself, optimal NK cell development is probably due to the lack of specific cytokines, as demonstrated by the addition of IL-2, the cytokine selected to perform these studies, which was essential to enhance NK cell differentiation. IL-2 may exert a direct effect on haematopoietic progenitors or may induce the production of differentiation-promoting factors in the stromal microenvironment. Significant differences were not found in the generation of NK cells between stroma contact vs transwell cultures, in contrast to cultures of CD34+DR–Lin– progenitors, which only gave rise to high numbers of CD45+CD56+ cells when cultured in direct contact with stroma and in the presence of IL-2. This observation is in agreement with studies published by Miller et al (1992, 1994) showing that direct contact with stroma is required for the most primitive CD34+DR– and CD34+ CD7–Lin– progenitors to differentiate along the NK cell lineage. The stage of differentiation and maturation of haematopoietic progenitors appears to define their responsiveness to growth factors and their need for stroma. Phenotypic differences between CD34+DR–Lin– and CD34+ DR+Lin– progenitors, namely in the expression of receptors involved in the initial differentiation steps of NK cell lineage, can account for the need for stroma by the most primitive CD34+DR–Lin– progenitors. The direct contact with stromal cells or their extracellular matrix may be essential for the activation of signalling pathways that lead to NK cell development. Alternatively, direct haematopoietic cell–stroma contact may abrogate any negative influence from soluble factors present in the culture medium and/or in the serum. Functional assays revealed that only NK cells grown in the presence of IL-2 were cytolytic against K562 target cells. The reduced cytolytic activity of NK cells generated in cultures without IL-2 could be due to the low numbers of NK cells grown in these cultures or alternatively these cells were functionally immature. In the presence of IL-2, stroma contact cultures of CD34+DR–Lin– progenitors gave rise to significantly more cytotoxic NK cells than these grown from CD34+DR+Lin– progenitors at E:T ratios 5:1–0Æ6:1. Silva et al (1994) also observed that NK cells derived from CD34+DR– progenitors were more cytotoxic that NK cells grown from CD34+DR+ progenitors. NK cell-mediated cytotoxicity depends on the recognition of and attachment to target cells by the effector cell and the subsequent induction of target cell apoptosis via perforin/ granzymes release or FasL/FasR interactions (Dustin & Springer, 1989; Singer, 1992; Zinkernagel et al, 1996). Differences in the levels of expression of adhesion molecules and lytic proteases, such as perforin and granzyme B, can account for the high cytolytic activity observed on NK cells derived from CD34+DR–Lin– progenitors; however, these analyses were not performed as a result of insufficient numbers of cells. Alternatively, other NK inhibitory cell populations may not have developed from the CD34+DR–Lin– populations. No significant differences were found in the cytolytic activity of NK cells derived from CD34+DR+Lin– progenitors, between stroma contact and transwell cultures.
The successful engraftment and multilineage differentiation of human haematopoietic stem cells in adult SCID mice (McCune et al, 1988), normal mice (Pallavicini et al, 1992; Pixley et al, 1994) and sheep (Srour et al, 1992, 1993; Zanjani et al, 1992a,b, 1995a,b), and the efficiency of some murine stromal cell lines in supporting the maintenance, and the myeloid and lymphoid proliferation of HSC (Issaad et al, 1993; Aiba et al, 1997; Miller et al, 1999) demonstrated that a xenogeneic environment supports human haemopoiesis. The present study provides evidence that fetal sheep thymic stroma is able to support in vitro differentiation of NK cells from human haematopoietic progenitors. However, a significant responsiveness of human haematopoietic progenitors to regulatory factors and/or cell–cell interactions may be species restricted and/ or dependent on the expression of certain adhesion molecules, growth factor milieu, components of the extracellular matrix and/or the functional maturity of the stromal cells or progenitor cells. The sheep represents a large animal model of human haematopoiesis that was developed by taking advantage of the tolerant environment of the pre-immune sheep fetus. Several studies have addressed the involvement of the thymus in NK cell development. The existence of NK cells in the thymus was demonstrated by the ability to expand NK cells from mouse or human immature thymocytes cultured in IL-2 (Phillips & Lanier, 1987; Ramsdel & Golub, 1987; Michon et al, 1988; Poggi et al, 1990; Mingari et al, 1991). Our group (Vaz et al, 1998a) and others (Tjonnfjord et al, 1995) also showed that cytotoxic NK cells can be obtained by co-culture of immature haematopoietic progenitors with thymic stroma and IL-2. The present study, demonstrating the in vitro NK cell development over fetal sheep thymic stromas, correlates with the detection of human NK cells in sheep thymuses at different times post in utero transplantation with HSCs (unpublished results) and is an additional indication of the involvement of the thymus in NK cell development, either by cellular interactions and/or the production of growth factors. Alternatively, the thymus may function as a homing device during human NK cell ontogeny, allowing for survival and perhaps proliferation and maturation of these cells. It will be of interest to investigate the main role of the thymus during NK ontogeny and clarify its contribution to the development of specific tolerance in transplantation. ACKNOWLEDGMENTS These studies were supported, in part, by funds provided by The Veterans Administration Research Office to Joa˜o L. Ascensa˜o. Isabel Bara˜o is the recipient of a scholarship from FCT (Lisbon, Portugal); Fa´tima Vaz was the recipient of a scholarship from Invotan (Lisbon, Portugal); Esmail Zanjani is supported in part by funds from the VA Research Office and by NIH grants HL49042, HL52955, DK51427. We wish to thank Dorothy Hudig for her comments and Kimberly Higgins for helping with the preparation of this manuscript.
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
NK Cell Development in Xenogeneic Culture REFERENCES Aiba, Y., Hirayama, F. & Ogawa, M. (1997) Clonal proliferation and cytokine requirement of murine progenitors for natural killer cells. Blood, 89, 4005–4012. Almeida-Porada, G., Ascensao, J.L. & Zanjani, E. (1996) The role of sheep stroma in human haemopoiesis in the human/sheep chimaeras. British Journal of Haematology, 93, 795–802. Andrews, R.G., Singer, J.W. & Bernstein, I.D. (1990) Human hematopoietic precursors in long-term culture: single CD34+ cells that lack detectable T cell, B cell, and myeloid cell antigens produce multiple colony-forming cells when cultured with marrow stromal cells. Journal of Experimental Medicine, 172, 355–358. Asao, H., Okuyama, C., Kumaki, S., Ishii, N., Tsuchiya, S., Foster, D. & Sugamura, K. (2001) Cutting edge: the common gamma-chain is an indispensable subunit of the IL-21 receptor complex. Journal of Immunology, 167, 1–5. Barcena, A., Galy, A.H.M., Punnonen, J., Muench, M.O., Schols, D., Roncarolo, M.G., de Vires, J.E. & Spits, H. (1994) Lymphoid and myeloid differentiation of fetal liver CD34+ lineage cells in human thymic organ culture. Journal of Experimental Medicine, 180, 123– 132. Brandt, J.E., Srour, E.F., van Besien, K., Briddell, R.A. & Hoffman, R. (1990) Cytokine-dependent long-term culture of highly enriched precursors of hematopoietic progenitor cells from human bone marrow. Journal of Clinical Investigation, 86, 932–941. Briddell, R.A., Broudy, V.C., Bruno, E., Brandt, J.E., Srour, E.F. & Hoffman, R. (1992) Further phenotypic characterization and isolation of human hematopoietic progenitor cells using a monoclonal antibody to the c-kit receptor. Blood, 79, 3159– 3167. Cashman, J., Eaves, A.C. & Eaves, C.J. (1985) Regulated proliferation of primitive hematopoietic progenitor cells in long-term human marrow cultures. Blood, 66, 1002–1005. Denning, S.M., Jones, D.M., Ware, R.E., Weinhold, K.J., Brenner, M.B. & Haynes, B.F. (1991) Analysis of clones derived from human CD7+CD4–CD8–CD3– thymocytes. International Immunology, 3, 1015–1024. Dustin, M. & Springer, T.A. (1989) T-cell receptor cross-linking transiently stimulates adhesiveness through LFA-1. Nature, 341, 619–624. Herberman, R.B., Djeu, J.Y., Kay, H.D., Ortaldo, J.R., Ricardi, C., Bonnard, G.D., Holden, H., Fagmani, R., Santori, A. & Pucetti, P. (1979) Natural Killer cells: characteristics and regulation of activity. Immunological Reviews, 44, 43–70. Hori, T. & Spits, H. (1991) Clonal analysis of human CD4–CD8–CD3– thymocytes highly purified from postnatal thymus. Journal of Immunology, 146, 2116–2121. Issaad, C., Croisille, L., Katz, A., Vasinchenker, W. & Coulombel, L. (1993) A murine stromal cell line allows the proliferation of very primitive human CD34++/CD38– progenitor cells in long-term cultures and semisolid assays. Blood, 81, 2916–2924. Lanier, L.L., Spits, H. & Phillips, J.H. (1992) The developmental relationship between NK cells and T cells. Immunology Today, 13, 392–395. Lotzova, E. & Savary, C.A. (1993) Human natural killer cell development from bone marrow progenitors: analysis of phenotype, cytotoxicity and growth. Natural Immunity, 4–5, 209–217. McCune, J.M., Namikawa, R., Kaneshima, H., Shultz, L.D., Leiberman, M. & Weissman, I.L. (1988) The SCID-hu mouse: murine model for the analysis of human hematolymphoid differentiation and function. Science, 241, 1632–1639.
891
Matsuzaki, Y., Gyotoku, J.-I., Ogawa, M., Nishikawa, S.-I., Katsura, Y., Gachelin, G. & Nakauchi, H. (1993) Characterization of c-kit positive intrathymic stem cells that are restricted to lymphoid differentiation. Journal of Experimental Medicine, 178, 1283– 1292. Michon, J.M., Caligiuri, M.A., Hazanow, S.M., Levine, H., Schlossman, S.F. & Ritz, J. (1988) Induction of natural killer effectors from human thymus with recombinant IL-2. Journal of Immunology, 140, 3660–3667. Miller, J.S., Verfaillie, C. & McGlave, P. (1992) The generation of human natural killer cells from CD34+DR– primitive progenitors in long-term bone marrow cultures. Blood, 80, 2182–2187. Miller, J.S., Alley, K.A. & McGlave, P. (1994) Differentiation of natural killer (NK) cells from human primitive marrow progenitors in a stroma-based long-term culture system: identification of a CD34+7+ NK progenitor. Blood, 83, 2594–2601. Miller, J.S., McCullar, V., Punzel, M., Lemischka, I.R. & Moore, K.A. (1999) Single adult human CD34(+)/Lin–/CD38(–) progenitors give rise to natural killer cells, B-lineage cells, dendritic cells, and myeloid cells. Blood, 93, 96–106. Mingari, M.C., Poggi, A., Biassoni, R., Bellomo, R., Ciccone, E., Pella, N., Morelli, L., Verdiani, S., Moretta, A. & Moretta, L. (1991) In vitro proliferation and cloning of CD3– CD16+ cells from human thymocyte precursors. Journal of Experimental Medicine, 174, 21–26. Mrozek, E., Anderson, P. & Caligiuri, M.A. (1996) Role of interleukin-15 in the development of human CD56+ natural killer cells from CD34+ hematopoietic progenitor cells. Blood, 87, 2632–2640. Pallavicini, M.G., Flake, A.W., Madden, D., Bethel, C., Duncan, B., Gonzalgo, M.L., Haendel, S., Montoya, T. & Roberts, L. (1992) Hemopoietic chimerism in rodents transplanted in utero with fetal human hemopoietic cells. Transplantation Proceedings, 24, 524–543. Parrish-Novak, J., Dillon, S.R., Nelson, A., Hammond, A., Sprecher, C., Gross, J.A., Johnston, J., Madden, K., Xu, W., West, J., Schrader, S., Burkhead, S., Heipel, M., Brandt, C., Kuijper, J.L., Kramer, J., Conklin, D., Presnell, S.R., Berry, J., Shiota, F., Bort, S., Hambly, K., Mudri, S., Clegg, C., Moore, M., Grant, F.J., LoftonDay, C., Gilbert, T., Rayond, F., Ching, A., Yao, L., Smith, D., Webster, P., Whitmore, T., Maurer, M., Kaushansky, K., Holly, R.D. & Foster, D. (2000) Interleukin 21 and its receptor are involved in NK cell expansion and regulation of lymphocyte function. Nature, 408, 57–63. Paul, S.R., Yang, Y.-C., Donahue, R.E., Goldring, S. & Williams, D.A. (1991) Stromal cell-associated hematopoiesis: immortalization and characterization of a primate bone marrow-derived stromal cell line. Blood, 77, 1723–1733. Phillips, J.H. & Lanier, L.L. (1987) Acquisition of non-MHC restricted cytotoxic function by IL 2 activated thymocytes with an immature antigenic phenotype. Journal of Immunology, 139, 683–687. Pixley, J.S., Tavassoli, M., Zanjani, E.D., Shaft, D.M., Futamachi, K.J., Sauter, T., Tavassoli, A. & MacKintosh, F.R. (1994) Transplantation in utero of fetal human hematopoietic stem cells into mice results in hematopoietic chimerism. Pathobiology, 62, 238– 244. Poggi, A., Biassoni, R., Pella, N., Paolieri, F., Bellomo, R., Bertolini, A., Moretta, L. & Mingari, M.C. (1990) In vitro expansion of CD3/TCR- human thymocyte populations that selectively lack CD3 delta gene expression: a phenotypic and functional analysis. Journal of Experimental Medicine, 172, 1409–1418.
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
892
I. Bara˜o et al
Ramsdell, F.J. & Golub, S.M. (1987) Generation of lymphokineactivated killer cell activity from human thymocytes. Journal of Immunology, 139, 1446–1453. Rodewald, H.R., Moingeo, P., Lucich, J.L., Dosiou, C., Lopez, P. & Reinherz, E.L. (1992) A population of early fetal thymocytes expressing Fc gamma RII/III contains precursors of T lymphocytes and natural killer cells. Cell, 69, 139–150. Sanchez, M.J., Spits, H., Lanier, L.L. & Phillips, J.H. (1993) Human natural killer cell committed thymocytes and their relation to the T-cell lineage. Journal of Experimental Medicine, 178, 1857–1866. Sanchez, M.J., Muench, M.O., Roncarolo, M.G., Lanier, L.L. & Philips, J.H. (1994) Identification of a common T/natural killer cell progenitor in human fetal thymus. Journal of Experimental Medicine, 180, 569–576. Silva, M.R.G., Hoffman, R., Srour, E.F. & Ascensao, J.L. (1994) Generation of human natural killer cells from immature progenitors does not require marrow stromal cells. Blood, 84, 841– 846. Singer, S.J. (1992) Intercellular communication and cell-cell adhesion. Science, 255, 1671–1677. Spits, H., Lanier, L.L. & Phillips, J.H. (1995) Development of human T and natural killer cells. Blood, 85, 2654–2670. Srour, E.F., Brandt, J.E., Briddell, R.A., Leemhuis, T., van Besien, K. & Hoffman, R. (1991) Human CD34+HLA-DR– bone marrow cells contain progenitor cells capable of self-renewal, multilineage differentiation and long term in vitro hematopoiesis. Blood Cells, 17, 287–295. Srour, E.F., Zanjani, E.D., Brandt, J.E., Leemhuis, T., Briddell, R.A., Heerema, N.A. & Hoffman, R. (1992) Sustained human hematopoiesis in sheep transplanted in utero during early gestation with fractionated adult human bone marrow cells. Blood, 79, 1404–1412. Srour, E.F., Zanjani, E.D., Cornetta, K., Traycoff, C.M., Flake, A.W., Hedrick, M., Brandt, J.E., Leemhuis, T. & Hoffman, R. (1993) Persistence of human multilineage, self-renewing lymphohematopoietic stem cell in chimeric sheep. Blood, 82, 3333–3342. Tjonnfjord, G.E., Steen, R., Vieby, O.P. & Egeland, T. (1995) Thymic stromal cells support differentiation of natural killer cells from CD34+ bone marrow cells in vitro. European Journal of Haematology, 54, 46–50. Trinchieri, G. (1989) Biology of Natural Killer cells. Advances in Immunology, 47, 187–376. Vaz, F., Srour, E.F., Almeida-Porada, G. & Ascensao, J.L. (1998a) Human thymic stroma supports human natural killer (NK) development from immature progenitors. Cellular Immunology, 186, 133–139. Vaz, F., Hoffman, R., Almeida-Porada, G. & Ascensao, J.L. (1998b) Definition of early progenitors and functional maturation of human natural killer cells: requirements for cytocidal activity. Pathobiology, 66, 41–48. Verfaillie, C. (1992) Direct contact between human primitive hematopoietic progenitors and bone marrow stroma is not required for long-term in vitro hematopoiesis. Blood, 79, 2821–2826.
Verfaillie, C., Blackolmer, K. & McGlave, P. (1990) Purified primitive human hematopoietic progenitor cells with long-term in vitro repopulating capacity adhere selectively to irradiated bone marrow stroma. Journal of Experimental Medicine, 172, 509–512. Welsh, R.M. (1981) Natural Killer cell mediated immunity during viral infection. Current Topics in Microbial Immunology, 92, 83– 106. Williams, N.S., Moore, T.A., Schatzle, J.D., Puzanov, I.J., Sivakumar, P.V., Zlotnik, A., Bennett, M. & Kumar, V. (1997) Generation of lytic natural killer 1.1+, Ly-49– cells from multipotential murine bone marrow progenitors in a stroma-free culture: definition of cytokine requirements and developmental intermediates. Journal of Experimental Medicine, 186, 1609–1614. Williams, N.S., Klem, J., Puzanov, I.J., Sivakumar, P.V., Schatzle, J.D., Bennett, M. & Kumar, V. (1998) Natural killer cell differentiation: insights from knockout and transgenic mouse models and in vitro systems. Immunological Reviews, 165, 47–61. Williams, N.S., Klem, J., Puzanov, I.J., Sivakumar, P.V., Bennett, M. & Kumar, V. (1999) Differentiation of NK1.1+, Ly49+ NK cells from flt3+ multipotent marrow progenitor cells. Journal of Immunology, 163, 2648–2656. Yu, H., Fehniger, T.A., Fuchshuber, P., Thiel, K.S., Vivier, E., Carson, W.E. & Caligiuri, M.A. (1998) Flt3 ligand promotes the generation of a distinct CD34 (+) human natural killer cell progenitor that responds to interleukin-15. Blood, 92, 3647–3657. Zanjani, E.D., MacKintosh, F.R. & Harrison, M.R. (1991) Hematopoietic chimerism in sheep and non-human primates by in utero transplantation of fetal hematopoietic stem cells. Blood Cells, 17, 349–363. Zanjani, E.D., Ascensao, J.L., Harrison, M.R. & Tavassoli, M. (1992a) Ex vivo incubation with growth factors enhances the engraftment of fetal hematopoietic stem cells transplanted in sheep fetuses. Blood, 79, 3045–3049. Zanjani, E.D., Pallavacini, M.G., Ascensao, J.L., Flake, A.W., Langlois, R.G., Reitsma, M., MacKintosh, F.R., Stutes, D., Harrison, M.R. & Tavassoli, M. (1992b) Engraftment and long-term expression of human fetal hemopoietic stem cells in sheep following transplantation in utero. Journal of Clinical Investigation, 89, 1178–1188. Zanjani, E.D., Almeida-Porada, G. & Flake, A.W. (1995a) Retention and multilineage expression of human fetal hemopoietic stem cells in human-sheep quimeras. Stem Cells, 13, 101–111. Zanjani, E.D., Srour, E.F. & Hoffman, R. (1995b) Retention of long-term repopulating ability of xenogeneic transplanted purified adult human bone marrow hematopoietic stem cells in sheep. Journal of Clinical and Laboratory Medicine, 126, 24–28. Zinkernagel, R.M., Bachmann, M.F., Kundig, T.M., Oehen, S., Pirchet, H. & Hengartner, H. (1996) On immunological memory. Annual Review of Immunology, 14, 333–367.
2002 Blackwell Science Ltd, British Journal of Haematology 118: 885–892
