- Email: [email protected]
The production of large ‘signalling competent’ myeloid cells from circulating CD34+ cells in neonatal blood
JOUMAL OF ELSEVIER
Journal of Immunological Methods 179 (1995) 187-192
The production of large ‘signalling competent’ myeloid cells from circulating CD34+ cells in neonatal blood G.M. Roberts a, T. Hoy ‘, M.B. Hallett ‘3* ’ Department of Surgery, UniL:ersityof Wales College of Medicine. Heath Park, CardiffCF4 4XN, UK h Department of Haematology, University of Wales College of Medicine, Heath Park, CardiffCF4 4XN. UK Received 15 September 1994; accepted 31 October 1994
Abstract A method is described for the production of large myeloid cells, expressing functional formylated peptide receptors. CD34+ cells were isolated from neonatal cord blood by a two stage cell sorting method. Inclusion of SCF, together with IL-3, GM-CSF or both cytokines stimulated growth of these cells over 14 day period. The resultant cells, which ranged from 30 pm to 100 pm in size, were maintained in culture for up to 5 weeks, during which time the cell population increasingly displayed myeloid characteristics, including expression of formylated peptide receptors, phagocytosis and oxidase activation. These large cells had a functioning Ca’+ signalling system in response to the chemotactic peptide, f-Met-Leu-Phe. The large size of the cells enabled the rise in cytosolic free Ca2+ which resulted from either transmembrane influx of extracellular Ca” and release of Ca” from intracellular stores to be visualised. These large myeloid cells thus provide a model system for investigating the spatial characteristics of Ca’+ signalling by formylated peptide receptors on human myeloid cells. Keywords: Myeloid
cell; CD34; Signalling
competent;
Umbilical
1. Introduction Although much is known about the signalling events which occur within myeloid cells, such as neutrophils and monocytes, progress is hampered by the small size of these cells (diameter: lo-20 pm). This limits visualisation of the spatial char-
Abbreviations: IMDM, Iscove’s modified Dulbecco’s medium; SCF, stem cell factor: GM-CSF, granulocyte-macrophage colony stimulating factor * Corresponding author. Tel.: 0222-74274X: Fax: 0222. 761623. 0022-1759/95/$09.50 0 1995 Elsevier SSDI 0022-I 759(94)00284-3
Science
cord blood; Cell sorting
acteristics of CaZC signals and prevents micro-injection and other intracellular manipulations. Larger myeloid cells exist in some other species. For example, the new eosinophil (Koonce et al., 1984) has a length of approximately 100 pm, and has been successfully used to probe the nature of Ca2+ signalling during chemotaxis (Brundage et al., 1993). Unfortunately, no equivalently sized human myeloid cell exists in the circulation. However, umbilical cord blood cells, which differentiate during culture into myeloid cells and into dendritic cells (Santiago-Schwartz and Fleit, 1988; Santiago-Schwartz et al., 19891, can produce large cells (up to 120 pm diam.). These large cells
B.V. All rights reserved
188
G.M. Robertset al. /Journal of ImmunologicalMethods 179 (1995) 187-192
express formylated peptide receptors which trigger a rise in cytosolic free Ca2+ concentration (Roberts et al., 19941, in a similar way to neutrophils (Hallett et al., 1990,1991). Their large size permits visualisation of Ca2+ waves and allows the Ca2+ rise at a focal point within the cell to be spatially resolved from the rise in Ca2+ at the cell edge (Roberts et al., 1994). Although these events, which result from release of Ca*+ from an intracellular store and then from transmembrane Ca2+ influx, are thought to occur in mature neutrophils, they have not been distinguishable in these cells (Hallett et al., 1991). These large human myeloid cells, thus represent a useful model system in which to investigate Ca2+ signalling in myeloid cells. The aim of the method described here was to produce large ‘signalling competent’ myeloid cells reproducibly and with known lineage. The method involved the production of CD34+ positive cells from human cord blood, and induction of proliferation by the inclusion of specific growth factors, SCF, IL-3, and GM-CSF.
threshold trigger on fluorescence, such that only positive cells were identified. For a sample with 1% positive cells, a trigger rate of 270/s was equivalent to 2.7 x lo4 cells/s or 10s cells/h, which places approximately 1 cell/droplet using a 70 pm nozzle. A three droplet sort was used, such that each positive cell was accompanied on average by two unwanted cells. The expected purity of this initial sort stage was thus approximately 33% positive cells. The second stage was by conventional sorting, triggering on forward scattering, for a few minutes to achieve the required high purity. 2.2. Culturing of CD34 + cells Between 5 x lo3 and lo4 CD34+ cells/ml were cultured in 24 well plates in Iscove’s modified Dulbecco’s medium (Gibco, Paisley, UK). The medium was supplemented with various combinations of SCF, GM-CSF and IL-3 (kind gifts from Genetics Institute, MA, USA). 2.3. Cytosolic free Ca2 ’ imaging
2. Materials and methods 2.1. Isolation of neonatal CD34 + cells Umbilical cord blood either from normal deliveries or from caesarian sections was collected into heparinised containers, diluted with an equal volume of phosphate buffered saline (PBS) and centrifuged through Ficoll-Hypaque. The resulting mononuclear cells (MN0 were subjected to hypotonic lysis of red cells and were incubated with phycoerythrin-conjugated CD34 antibody (HPCA-2 CD34-PE, Becton Dickinson, San Jose, CA) on ice for 30 min. The cells were then washed twice in PBS for fluorescent activated cell sorting on a FACS440 (Becton Dickinson, San Jose, CA). Since umbilical cord blood contained only approximately 1% CD34 positive cells, conventional sorting at lO’/h would produce no more than lo5 CD34’ cells/h. However, a two stage strategy was used here which achieved the same recovery level but at a faster rate (Baines et al., 1994). The first stage was achieved by setting the
After 2-3 weeks in culture the attached CD34 daughter cells were loaded with fura- as previously described. Excitation wavelengths were achieved using a Spex Fluorolog dual wavelength fluorimeter (Glen Spectra, Stanmore, UK), and ratio images acquired by an ISIS-M intensified CCD camera (Photonic Science, Tunbridge, UK) coupled to a Spex IMlOl analysis system (Hallett et al., 1990,1991). 2.4. Characterisation of myeloid ce1l.s NSE staining of cells was performed as described by Yam et al. (1971). The expression of formylated peptide receptors was demonstrated and quantified using of a fluorescently labelled N-formylated peptide, the tetramethyl derivative of f-Nle-Leu-Phe-Nle-Tyr-Lys (Molecular Probes, Oregon, USA). Phagocytosis was assessed microscopically after incubation with opsonised zymosan particles (0.4 mg/ml) for 30 min at 37”C, by staining the methanol fixed cells with Diffquick (Merz-Dade, Dudingen, Switzerland). Oxi-
GM
Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
189
Fluor rescence
> Forward Scatter
0
Forward Scatter
255
Fig. 1. The two stage cell sorting separation of CD34+ cells. The distributions of CD34+ positive cells within (A) the initial mononuclear cell population from cord blood, (B) the population after the initial enrichment and (C) the final population after resorting the enriched population. The initial population contains approximately only 1% positive cells, whereas the enriched population contains approximately 30% and the final population 98% positive cells.
dase activation was detected by the reduction of nitroblue tetrazolium (NBT) with superoxide anion to form insoluble formazan. Myeloid cells were stimulated with opsonized zymosan particles (0.4 mg/ml) in the presence of NBT (1 mg/ml) for 30 min at 37°C. Activation of the oxidase was assessed by the presence of a black precipitated formazan under bright field illumination.
0.5-1.5% CD34+ cells (Fig. 1A). After the initial enrichment sorting procedure, this percentage increased to approximately 30% (Fig. 1B). The final sorting stage produced populations of CD34+ cells which were approximately 98% pure (Fig. 10.
2.5. DNA measurement After 14 days the cultures were harvested by centrifuging the plates at 45°C for 30 min and decanting the supematant. The plates were frozen at -20°C thawed, water added, refrozen and re-thawed before allowing them to reach room temperature. Hoechst 33258 (10 pg/ml) in Tris (10 mM), EDTA (1 mM), NaCl (2 M), pH7.4 was added and the resulting fluorescence was measured in a fluorescence plate reader (Wellfluor, Denley Instruments) as described by Rago et al. (1990).
3. Results 3.1. Production of CD34 + cells Flow cytometry showed that the initial mononuclear fraction of cord blood contained
IMDM
only
SCF
SCF GMCSF
SCF+ILJ
SCF+IL3 GMCSF
Fig. 2. Proliferation of CD34+ cells induced by cytokines. The total DNA content of the cell population was determined at 14 days in the absence of added cytokines (IMDM only), SCF (20 ng/ml), SCF plus GM-CSF (20 ng/ml plus 5 ng/ml respectively), SCF plus IL-3 (20 ng/ml plus S ng/ml respectively) and SCF plus both GM-CSF and IL-3 (20 ng/ml, 5 ng/ml and 5 ng/ml respectively). The data shown. as fold increase DNA content, are the mean and standard error of four separate experiments.
190
G.M. Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
3.2. Amplification of stem cells
3.3. Characterisation of cultured cells
After 14 days culture in the absence of cytokines, there was no significant increase in total DNA. However, cell proliferation was enhanced by the inclusion of SCF (20 ng/ml) alone or together with G-MCSF (5 ng/ml), IL-3 (5 ng/ml) or both cytokines (Fig. 2). The DNA content was further increased by a combination of these cytokines, a combination of all three cytokines giving approximately a 75fold increase. The cells were maintained in culture for up to 37 days.
After approximately 1 week in culture, round cells with occasional colonies were seen. However, after 2 weeks, most cells were large, ranging in length from 30-100 pm, and varying in shape from round to elongated and dendritic (Fig. 3a). Time lapse video microscopy showed that individual cells could change from round to elongate over 2-3 h periods, although dendritic cells may have represented a separate class. The cells had
b
bright
field
d Fig. 3. Characterisation of cultured cells. a: phase contrast view of cultured cells at 2 weeks, showing cells of different morphologies (magnification x40 vertical dimension = 150 pm). b: staining for non-specific esterase. showing a high percentage of positive cells. c: phagocytosis of zymosan particles, visualised in the large picture by non-staining of the particles within the cell cytoplasm. Oxidase activation accompanying phagocytosis was detected by NBT staining at the sites of zymosan uptake, in the upper smaller picture and in the absence of NBT at the same magnification (in the lower small picture). d: formylated peptide receptors demonstrated by binding of fluorescently labelled formylated peptide.
G.M. Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
the following myeloid cell characteristics: (i) they stained positively for non-specific esterase (Fig. 3b), up to 87% of the population by day 33, (ii) they phagocytosed zymosan particles, up to 84% of the population by day 35 (Fig. 3c) with an accompanying oxidative burst, detectable by NBT reduction (Fig. 3c) and (iii) they expressed formylated peptide receptors (Fig. 3d). Time-lapse video has also shown that the cells form pseudopods and occasionally move bodily, towards a local source of the chemotactic peptide f-MetLeu-Phe within a micro-pipette positioned within 10 pm of the cell edge. 3.4. Ca” + signailing in c&wed
ceils
Ca*+ imaging of fura-Zloaded cultured cells demonstrated that in six out of ten cells tested, the cytosolic free Ca2+ concentration rose in response to the addition of f-Met-Leu-Phe (1 ,uM). The magnitude of the response and the subcellular distribution was similar to that described previously (Roberts et al., 1994) for cord blood cell derived cells (Fig. 4). Two phases of the Ca2+ rise were observed, an initial localised rise, followed by an elevated region near the plasma membrane which spread across the cell in a slow ‘wave’.
4. Discussion The work reported here demonstrates that myeloid cells with functioning signalling pathways for phagocytosis, oxidase activation and Ca*+ signalling can be derived from the CD34+ popula-
191
tion of umbilical cord blood. After 2-3 weeks of culture, giant cells, with dimensions of 100 pm, are produced which will be useful model cells for the investigation of signalling pathways. Not only are they large enough to visualise subcellular localisation of Ca2+ signals (Roberts et al., 1994), but they will also be sufficiently large for microinjection. Neither of these outcomes has yet been achieved with mature neutrophils (diameter: 10 Km). Previous reports of CD34+ derived myeloid cells grown in vitro (Saeland et al., 1988, Haylock et al., 1992, Santiago-Schwartz et al., 1992, Huang and Terstappen, 1994) have not included evidence for functional signalling. These groups have demonstrated that isolated CD34+ cells proliferate in vitro in response to various combinations of cytokines such as SCF, IL-3, GM-CSF, GSF, IL-6, IL-1 and TNF-a (Haylock et al., 1992, Caux et al., 1992). The data here show that over a 2-3 week period, cultured with combinations of either SCF, IL-3 or GM-CSF, the resultant cells increasingly attach and display myeloid characteristics and retain functional activity. In a previous study, we have shown that cultured unfractionated cdrd blood white cells have similar characteristics (Roberts et al., 1994). This raises the possibility that in the previous study the CD34+ cell was also the major proliferative cell. However, in the previous study, the production of these cells was not predictable and yields were low. Here the isolation of the CD34+ cells by a two stage cell sorting method, enabling sorting to be speeded up considerably (Baines et al., 19941, has permitted the starting cells to be precisely defined. Together
Fig. 4. Ca 2c signalling in large myeloid cells. The figure shows pseudo 3D ‘maps’ of cytosolic free Ca’+ concentration, horizontal plane represents the dimensions of the cell, and the vertical plane the cytosolic free Ca’+ concentration. the horizontal base is 100 pm, and the equivalent height represents 1 PM Cazf. The same cell is shown (a) before addition of f-Met-Leu-Phe (1 PM).
in which the Each side of and (b) after
192
G.M. Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
with the use of cytokines which amplify this cell population, a reliable method of producing high yields of pure populations of these cells has been developed.
Acknowledgements
We are grateful to the Maternity Unit, UHW for the supply of cord blood, to J. Fisher for assistance with cell sorting and to Dr. P. Baines for generous advice.
References Baines, P., Truran, L., Bailey-Wood, R., Hoy, T., Lake, H., Poynton, C.H. and Burnett, A. (1994) Haernopoietic colony-forming cells from peripheral blood stem cell harvests: cytokine requirements and lineage potential. Br. J. Haematol., in press. Brundage, R.A., Fogarty, K.E., Tuft, R.A. and Fay, F.S. (19931: Chemotaxis of newt eosinophils: calcium regulation of chemotactic response. Am. J. Physiol. 265, C1527c1543. Caux, C., Dezutter-Dambuyant, C., Schmitt, D. and Banchereau, J. (1992) GM-CSF and TNF-o cooperate in the generation of dendritic Langerhans cells. Nature 360, 258-261. Hallett, M.B., Davies, E.V. and Campbell, A.K. (1990) Oxidase activation in individual neutrophils is dependent on the onset and magnitude of the calcium signal. Cell Calcium 11, 655-663. Hallett, M.B., Davies, E.V. and Campbell, A.K. (1991) Spatial and temporal analysis of Ca2+ signals in individual neutrophils using ratio imaging of fura-2. In: E. Reid, G.M.W. Cook and J.P. Luzio (Eds.), Surveys in Subcellullar Methodology, Vol. 21, Cell SignaIling: Experimental
Strategies. Royal Society of Chemistry, London, pp. 273286. Haylock, D.N., To, B., Dawse, T.L., Juttner, C.A. and Simmons, P.J. (1992) Ex-vivo expansion and maturation of peripheral blood CD34+ cells into the myeloid lineage. Blood 80, 1405-1412. Huang, S. and Terstappen, W.M.M. (1994) Lyrnphoid and myeloid differentiation of single human CD34+, HLADR+, CD38-hematopoietic stem cells. Blood 83, 1.5151526. Koonce, M.P., Cloney, R.A. and Berns, M.W. (1984) Laser irradiation of centrosomes in newt eosinophils: evidence of centriole role in motility. J. Cell. Biol. 98, 1999-2010. Rago, R., Mitchen, J. and Wilding, G. (1990) DNA fluormetric assay in 96-well tissue culture plates using Hoechst 33258 after cell lysis by freezing in distilled water. Anal. Biochem. 191, 31-34. Roberts, G.M., Davies, E.V. and Hallett, M.B. (1994) Slow calcium waves imaged in myeloid cells derived from neonatal blood. J. Leuk. Biol. 55, 461-466. Saeland, S., Caux, C., Favre, C., Aubry, J.P., Mannoni, P., Pebusque M.J.. Gentilhome, O., Otusuka, T., Yokota, T., Arai, N., Arai, K., Banchereau, J. and De Vries, J.E. (1988) Effects of recombinant human interleukin-3 on CD3Cenriched normal hematopoietic progenitors and on myeloblastic leukemia cells. Blood 72, 1580-1588. Santiago-Schwartz, F. and Fleit, H.B. (1988) Identification of nonadherent mononuclear cells in human cord blood that differentiate into macrophages. J. Leuk. Biol. 43, 51-59. Santiago-Schwartz, F., McHugh, D.M. and Fleit, H.B. (1989) Functional analysis of monocyte-macrophage derived from nonadherent cord blood progenitor cells: correlation with the ontogeny of cell surface proteins. J. Leuk. Biol. 46, 230-238. Santiago-Schwartz, F., Belilos, E., Diamond, B. and Carsons, S.E. (1992) TNF-a in combination with GM-CSF enhances the differentiation of neonatal cord blood stem cells into dendritic cells and macrophages. J. Leuk. Biol. 52, 274281. Yam, L.T., Li, C.Y. and Crosby, W.H. (1971) Cytochemical identification of monocytes and granulocytes. Am. J. Clin. Pathol. 55, 283-290.
Journal of Immunological Methods 179 (1995) 187-192
The production of large ‘signalling competent’ myeloid cells from circulating CD34+ cells in neonatal blood G.M. Roberts a, T. Hoy ‘, M.B. Hallett ‘3* ’ Department of Surgery, UniL:ersityof Wales College of Medicine. Heath Park, CardiffCF4 4XN, UK h Department of Haematology, University of Wales College of Medicine, Heath Park, CardiffCF4 4XN. UK Received 15 September 1994; accepted 31 October 1994
Abstract A method is described for the production of large myeloid cells, expressing functional formylated peptide receptors. CD34+ cells were isolated from neonatal cord blood by a two stage cell sorting method. Inclusion of SCF, together with IL-3, GM-CSF or both cytokines stimulated growth of these cells over 14 day period. The resultant cells, which ranged from 30 pm to 100 pm in size, were maintained in culture for up to 5 weeks, during which time the cell population increasingly displayed myeloid characteristics, including expression of formylated peptide receptors, phagocytosis and oxidase activation. These large cells had a functioning Ca’+ signalling system in response to the chemotactic peptide, f-Met-Leu-Phe. The large size of the cells enabled the rise in cytosolic free Ca2+ which resulted from either transmembrane influx of extracellular Ca” and release of Ca” from intracellular stores to be visualised. These large myeloid cells thus provide a model system for investigating the spatial characteristics of Ca’+ signalling by formylated peptide receptors on human myeloid cells. Keywords: Myeloid
cell; CD34; Signalling
competent;
Umbilical
1. Introduction Although much is known about the signalling events which occur within myeloid cells, such as neutrophils and monocytes, progress is hampered by the small size of these cells (diameter: lo-20 pm). This limits visualisation of the spatial char-
Abbreviations: IMDM, Iscove’s modified Dulbecco’s medium; SCF, stem cell factor: GM-CSF, granulocyte-macrophage colony stimulating factor * Corresponding author. Tel.: 0222-74274X: Fax: 0222. 761623. 0022-1759/95/$09.50 0 1995 Elsevier SSDI 0022-I 759(94)00284-3
Science
cord blood; Cell sorting
acteristics of CaZC signals and prevents micro-injection and other intracellular manipulations. Larger myeloid cells exist in some other species. For example, the new eosinophil (Koonce et al., 1984) has a length of approximately 100 pm, and has been successfully used to probe the nature of Ca2+ signalling during chemotaxis (Brundage et al., 1993). Unfortunately, no equivalently sized human myeloid cell exists in the circulation. However, umbilical cord blood cells, which differentiate during culture into myeloid cells and into dendritic cells (Santiago-Schwartz and Fleit, 1988; Santiago-Schwartz et al., 19891, can produce large cells (up to 120 pm diam.). These large cells
B.V. All rights reserved
188
G.M. Robertset al. /Journal of ImmunologicalMethods 179 (1995) 187-192
express formylated peptide receptors which trigger a rise in cytosolic free Ca2+ concentration (Roberts et al., 19941, in a similar way to neutrophils (Hallett et al., 1990,1991). Their large size permits visualisation of Ca2+ waves and allows the Ca2+ rise at a focal point within the cell to be spatially resolved from the rise in Ca2+ at the cell edge (Roberts et al., 1994). Although these events, which result from release of Ca*+ from an intracellular store and then from transmembrane Ca2+ influx, are thought to occur in mature neutrophils, they have not been distinguishable in these cells (Hallett et al., 1991). These large human myeloid cells, thus represent a useful model system in which to investigate Ca2+ signalling in myeloid cells. The aim of the method described here was to produce large ‘signalling competent’ myeloid cells reproducibly and with known lineage. The method involved the production of CD34+ positive cells from human cord blood, and induction of proliferation by the inclusion of specific growth factors, SCF, IL-3, and GM-CSF.
threshold trigger on fluorescence, such that only positive cells were identified. For a sample with 1% positive cells, a trigger rate of 270/s was equivalent to 2.7 x lo4 cells/s or 10s cells/h, which places approximately 1 cell/droplet using a 70 pm nozzle. A three droplet sort was used, such that each positive cell was accompanied on average by two unwanted cells. The expected purity of this initial sort stage was thus approximately 33% positive cells. The second stage was by conventional sorting, triggering on forward scattering, for a few minutes to achieve the required high purity. 2.2. Culturing of CD34 + cells Between 5 x lo3 and lo4 CD34+ cells/ml were cultured in 24 well plates in Iscove’s modified Dulbecco’s medium (Gibco, Paisley, UK). The medium was supplemented with various combinations of SCF, GM-CSF and IL-3 (kind gifts from Genetics Institute, MA, USA). 2.3. Cytosolic free Ca2 ’ imaging
2. Materials and methods 2.1. Isolation of neonatal CD34 + cells Umbilical cord blood either from normal deliveries or from caesarian sections was collected into heparinised containers, diluted with an equal volume of phosphate buffered saline (PBS) and centrifuged through Ficoll-Hypaque. The resulting mononuclear cells (MN0 were subjected to hypotonic lysis of red cells and were incubated with phycoerythrin-conjugated CD34 antibody (HPCA-2 CD34-PE, Becton Dickinson, San Jose, CA) on ice for 30 min. The cells were then washed twice in PBS for fluorescent activated cell sorting on a FACS440 (Becton Dickinson, San Jose, CA). Since umbilical cord blood contained only approximately 1% CD34 positive cells, conventional sorting at lO’/h would produce no more than lo5 CD34’ cells/h. However, a two stage strategy was used here which achieved the same recovery level but at a faster rate (Baines et al., 1994). The first stage was achieved by setting the
After 2-3 weeks in culture the attached CD34 daughter cells were loaded with fura- as previously described. Excitation wavelengths were achieved using a Spex Fluorolog dual wavelength fluorimeter (Glen Spectra, Stanmore, UK), and ratio images acquired by an ISIS-M intensified CCD camera (Photonic Science, Tunbridge, UK) coupled to a Spex IMlOl analysis system (Hallett et al., 1990,1991). 2.4. Characterisation of myeloid ce1l.s NSE staining of cells was performed as described by Yam et al. (1971). The expression of formylated peptide receptors was demonstrated and quantified using of a fluorescently labelled N-formylated peptide, the tetramethyl derivative of f-Nle-Leu-Phe-Nle-Tyr-Lys (Molecular Probes, Oregon, USA). Phagocytosis was assessed microscopically after incubation with opsonised zymosan particles (0.4 mg/ml) for 30 min at 37”C, by staining the methanol fixed cells with Diffquick (Merz-Dade, Dudingen, Switzerland). Oxi-
GM
Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
189
Fluor rescence
> Forward Scatter
0
Forward Scatter
255
Fig. 1. The two stage cell sorting separation of CD34+ cells. The distributions of CD34+ positive cells within (A) the initial mononuclear cell population from cord blood, (B) the population after the initial enrichment and (C) the final population after resorting the enriched population. The initial population contains approximately only 1% positive cells, whereas the enriched population contains approximately 30% and the final population 98% positive cells.
dase activation was detected by the reduction of nitroblue tetrazolium (NBT) with superoxide anion to form insoluble formazan. Myeloid cells were stimulated with opsonized zymosan particles (0.4 mg/ml) in the presence of NBT (1 mg/ml) for 30 min at 37°C. Activation of the oxidase was assessed by the presence of a black precipitated formazan under bright field illumination.
0.5-1.5% CD34+ cells (Fig. 1A). After the initial enrichment sorting procedure, this percentage increased to approximately 30% (Fig. 1B). The final sorting stage produced populations of CD34+ cells which were approximately 98% pure (Fig. 10.
2.5. DNA measurement After 14 days the cultures were harvested by centrifuging the plates at 45°C for 30 min and decanting the supematant. The plates were frozen at -20°C thawed, water added, refrozen and re-thawed before allowing them to reach room temperature. Hoechst 33258 (10 pg/ml) in Tris (10 mM), EDTA (1 mM), NaCl (2 M), pH7.4 was added and the resulting fluorescence was measured in a fluorescence plate reader (Wellfluor, Denley Instruments) as described by Rago et al. (1990).
3. Results 3.1. Production of CD34 + cells Flow cytometry showed that the initial mononuclear fraction of cord blood contained
IMDM
only
SCF
SCF GMCSF
SCF+ILJ
SCF+IL3 GMCSF
Fig. 2. Proliferation of CD34+ cells induced by cytokines. The total DNA content of the cell population was determined at 14 days in the absence of added cytokines (IMDM only), SCF (20 ng/ml), SCF plus GM-CSF (20 ng/ml plus 5 ng/ml respectively), SCF plus IL-3 (20 ng/ml plus S ng/ml respectively) and SCF plus both GM-CSF and IL-3 (20 ng/ml, 5 ng/ml and 5 ng/ml respectively). The data shown. as fold increase DNA content, are the mean and standard error of four separate experiments.
190
G.M. Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
3.2. Amplification of stem cells
3.3. Characterisation of cultured cells
After 14 days culture in the absence of cytokines, there was no significant increase in total DNA. However, cell proliferation was enhanced by the inclusion of SCF (20 ng/ml) alone or together with G-MCSF (5 ng/ml), IL-3 (5 ng/ml) or both cytokines (Fig. 2). The DNA content was further increased by a combination of these cytokines, a combination of all three cytokines giving approximately a 75fold increase. The cells were maintained in culture for up to 37 days.
After approximately 1 week in culture, round cells with occasional colonies were seen. However, after 2 weeks, most cells were large, ranging in length from 30-100 pm, and varying in shape from round to elongated and dendritic (Fig. 3a). Time lapse video microscopy showed that individual cells could change from round to elongate over 2-3 h periods, although dendritic cells may have represented a separate class. The cells had
b
bright
field
d Fig. 3. Characterisation of cultured cells. a: phase contrast view of cultured cells at 2 weeks, showing cells of different morphologies (magnification x40 vertical dimension = 150 pm). b: staining for non-specific esterase. showing a high percentage of positive cells. c: phagocytosis of zymosan particles, visualised in the large picture by non-staining of the particles within the cell cytoplasm. Oxidase activation accompanying phagocytosis was detected by NBT staining at the sites of zymosan uptake, in the upper smaller picture and in the absence of NBT at the same magnification (in the lower small picture). d: formylated peptide receptors demonstrated by binding of fluorescently labelled formylated peptide.
G.M. Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
the following myeloid cell characteristics: (i) they stained positively for non-specific esterase (Fig. 3b), up to 87% of the population by day 33, (ii) they phagocytosed zymosan particles, up to 84% of the population by day 35 (Fig. 3c) with an accompanying oxidative burst, detectable by NBT reduction (Fig. 3c) and (iii) they expressed formylated peptide receptors (Fig. 3d). Time-lapse video has also shown that the cells form pseudopods and occasionally move bodily, towards a local source of the chemotactic peptide f-MetLeu-Phe within a micro-pipette positioned within 10 pm of the cell edge. 3.4. Ca” + signailing in c&wed
ceils
Ca*+ imaging of fura-Zloaded cultured cells demonstrated that in six out of ten cells tested, the cytosolic free Ca2+ concentration rose in response to the addition of f-Met-Leu-Phe (1 ,uM). The magnitude of the response and the subcellular distribution was similar to that described previously (Roberts et al., 1994) for cord blood cell derived cells (Fig. 4). Two phases of the Ca2+ rise were observed, an initial localised rise, followed by an elevated region near the plasma membrane which spread across the cell in a slow ‘wave’.
4. Discussion The work reported here demonstrates that myeloid cells with functioning signalling pathways for phagocytosis, oxidase activation and Ca*+ signalling can be derived from the CD34+ popula-
191
tion of umbilical cord blood. After 2-3 weeks of culture, giant cells, with dimensions of 100 pm, are produced which will be useful model cells for the investigation of signalling pathways. Not only are they large enough to visualise subcellular localisation of Ca2+ signals (Roberts et al., 1994), but they will also be sufficiently large for microinjection. Neither of these outcomes has yet been achieved with mature neutrophils (diameter: 10 Km). Previous reports of CD34+ derived myeloid cells grown in vitro (Saeland et al., 1988, Haylock et al., 1992, Santiago-Schwartz et al., 1992, Huang and Terstappen, 1994) have not included evidence for functional signalling. These groups have demonstrated that isolated CD34+ cells proliferate in vitro in response to various combinations of cytokines such as SCF, IL-3, GM-CSF, GSF, IL-6, IL-1 and TNF-a (Haylock et al., 1992, Caux et al., 1992). The data here show that over a 2-3 week period, cultured with combinations of either SCF, IL-3 or GM-CSF, the resultant cells increasingly attach and display myeloid characteristics and retain functional activity. In a previous study, we have shown that cultured unfractionated cdrd blood white cells have similar characteristics (Roberts et al., 1994). This raises the possibility that in the previous study the CD34+ cell was also the major proliferative cell. However, in the previous study, the production of these cells was not predictable and yields were low. Here the isolation of the CD34+ cells by a two stage cell sorting method, enabling sorting to be speeded up considerably (Baines et al., 19941, has permitted the starting cells to be precisely defined. Together
Fig. 4. Ca 2c signalling in large myeloid cells. The figure shows pseudo 3D ‘maps’ of cytosolic free Ca’+ concentration, horizontal plane represents the dimensions of the cell, and the vertical plane the cytosolic free Ca’+ concentration. the horizontal base is 100 pm, and the equivalent height represents 1 PM Cazf. The same cell is shown (a) before addition of f-Met-Leu-Phe (1 PM).
in which the Each side of and (b) after
192
G.M. Roberts et al. /Journal
of Immunological Methods 179 (1995) 187-192
with the use of cytokines which amplify this cell population, a reliable method of producing high yields of pure populations of these cells has been developed.
Acknowledgements
We are grateful to the Maternity Unit, UHW for the supply of cord blood, to J. Fisher for assistance with cell sorting and to Dr. P. Baines for generous advice.
References Baines, P., Truran, L., Bailey-Wood, R., Hoy, T., Lake, H., Poynton, C.H. and Burnett, A. (1994) Haernopoietic colony-forming cells from peripheral blood stem cell harvests: cytokine requirements and lineage potential. Br. J. Haematol., in press. Brundage, R.A., Fogarty, K.E., Tuft, R.A. and Fay, F.S. (19931: Chemotaxis of newt eosinophils: calcium regulation of chemotactic response. Am. J. Physiol. 265, C1527c1543. Caux, C., Dezutter-Dambuyant, C., Schmitt, D. and Banchereau, J. (1992) GM-CSF and TNF-o cooperate in the generation of dendritic Langerhans cells. Nature 360, 258-261. Hallett, M.B., Davies, E.V. and Campbell, A.K. (1990) Oxidase activation in individual neutrophils is dependent on the onset and magnitude of the calcium signal. Cell Calcium 11, 655-663. Hallett, M.B., Davies, E.V. and Campbell, A.K. (1991) Spatial and temporal analysis of Ca2+ signals in individual neutrophils using ratio imaging of fura-2. In: E. Reid, G.M.W. Cook and J.P. Luzio (Eds.), Surveys in Subcellullar Methodology, Vol. 21, Cell SignaIling: Experimental
Strategies. Royal Society of Chemistry, London, pp. 273286. Haylock, D.N., To, B., Dawse, T.L., Juttner, C.A. and Simmons, P.J. (1992) Ex-vivo expansion and maturation of peripheral blood CD34+ cells into the myeloid lineage. Blood 80, 1405-1412. Huang, S. and Terstappen, W.M.M. (1994) Lyrnphoid and myeloid differentiation of single human CD34+, HLADR+, CD38-hematopoietic stem cells. Blood 83, 1.5151526. Koonce, M.P., Cloney, R.A. and Berns, M.W. (1984) Laser irradiation of centrosomes in newt eosinophils: evidence of centriole role in motility. J. Cell. Biol. 98, 1999-2010. Rago, R., Mitchen, J. and Wilding, G. (1990) DNA fluormetric assay in 96-well tissue culture plates using Hoechst 33258 after cell lysis by freezing in distilled water. Anal. Biochem. 191, 31-34. Roberts, G.M., Davies, E.V. and Hallett, M.B. (1994) Slow calcium waves imaged in myeloid cells derived from neonatal blood. J. Leuk. Biol. 55, 461-466. Saeland, S., Caux, C., Favre, C., Aubry, J.P., Mannoni, P., Pebusque M.J.. Gentilhome, O., Otusuka, T., Yokota, T., Arai, N., Arai, K., Banchereau, J. and De Vries, J.E. (1988) Effects of recombinant human interleukin-3 on CD3Cenriched normal hematopoietic progenitors and on myeloblastic leukemia cells. Blood 72, 1580-1588. Santiago-Schwartz, F. and Fleit, H.B. (1988) Identification of nonadherent mononuclear cells in human cord blood that differentiate into macrophages. J. Leuk. Biol. 43, 51-59. Santiago-Schwartz, F., McHugh, D.M. and Fleit, H.B. (1989) Functional analysis of monocyte-macrophage derived from nonadherent cord blood progenitor cells: correlation with the ontogeny of cell surface proteins. J. Leuk. Biol. 46, 230-238. Santiago-Schwartz, F., Belilos, E., Diamond, B. and Carsons, S.E. (1992) TNF-a in combination with GM-CSF enhances the differentiation of neonatal cord blood stem cells into dendritic cells and macrophages. J. Leuk. Biol. 52, 274281. Yam, L.T., Li, C.Y. and Crosby, W.H. (1971) Cytochemical identification of monocytes and granulocytes. Am. J. Clin. Pathol. 55, 283-290.