- Email: [email protected]
Monocyte-to-macrophage transition in vitro
Journal of Immunological Methods, 118 (1989) 9-16
9
Elsevier JIM 05078
Monocyte-to-macrophage transition in vitro A systematic study using human cells isolated by fractionation on Percoll Diane E. Davies and John B. Lloyd Department of Biological Sciences, University of Keele, Keele, Staffordshire ST5 5BG, U.K. (Received 3 August 1988, revised received 12 October 1988, accepted 18 October 1988)
Improved density-gradient methods, using Percoll or Nycodenz, have recently been introduced for the isolation of human monocytes, but the capacity of cells thus isolated to differentiate into macrophages has not been systematically studied. We have compared Percoll and Nycodenz methods for the isolation of monocytes from human blood. The Nycodenz method yielded a monocyte population of high purity, but the yield was low. The Percoll method gave almost quantitative yield of monocytes, and the contaminating cells, mostly lymphocytes, were readily washed away after allowing the monocytes to adhere to a plastic surface. The Percoll method was then successfully scaled up, providing a simple method to obtain the monocytes from 180 ml blood. These monocytes were maintained in culture and their capacity to mature into macrophages was studied, using the following criteria: increase in cell size and protein content, increase in specific activity of hexosaminidase, differential hexosaminidase release on exposure to opsonized zymosan and unopsonized polystyrene beads, loss of peroxidase activity, and development of fluoride-insensitivity by the cells' cytochemically demonstrable esterase. The cells also displayed morphological changes typical of the monocyte-to-macrophage transition. The procedures reported constitute a simple and reliable method for the production of human macrophages in increased yield. Key words: Percoll; Nycodenz; Monocyte; Macrophage; Differentiation
Introduction
Studies on human macrophages are severely hampered by the problems of obtaining these cells. An attractive alternative source of macrophages is the monocyte population of peripheral blood, followed by monocyte-to-macrophage differentiation in vitro. This approach requires a satisfactory method for obtaining human monocytes in good yield and culture conditions that reliably lead to monocyte maturation. Correspondence to: J.B. Lloyd, Department of Biological Sciences, University of Keele, Keele, Staffordshire ST5 5BG, U.K.
There are many differences, morphological, histochemical and biochemical, between monocytes and macrophages, and these can be used as criteria to monitor monocyte maturation in vitro. Thus human monocytes, isolated mostly by methods using Ficoll-Hypaque gradients, were shown to differentiate into macrophages over a period of 5-7 days, when cultured under suitable conditions (Johnson et al., 1977; Zuckerman et al., 1979). More recently alternative density-gradient media, Percoll (Ulmer and Flad, 1979; Pertoft et al., 1980) and Nycodenz (B~Syum, 1983), have been used for monocyte isolation; it is claimed that these avoid the considerable disadvantages of the earlier Ficoll-Hypaque methods.
0022-1759/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
10
Although there are references to the in vitro maturation of human monocytes isolated using Percoll or Nycodenz (Fluks, 1981; B~Syum, 1983), little or no evidence for this differentiation has been presented. This may account for the apparent reluctance of many investigators to adopt these newer methods. In the present paper we first compare Percoll and Nycodenz methods and conclude that the former is a better source of monocytes for subsequent culture in vitro. We then describe a scale-up of the Percoll method of Pertoft et al. (1980) and use a range of criteria, chosen from those used by previous workers, to monitor the in vitro maturation of the monocytes isolated. The work described constitutes the first systematic study of the monocyte-to-macrophage transition in vitro of human monocytes isolated by a Percoll method: we conclude that the methods described are suitable for studies of cell function during and after this differentiation process.
Methods
Isolation of a 'mononuclear cell' population from human peripheral blood Mononuclear cells were obtained by either the one-step Percoll method of Pertoft et al. (1980) or the Nycodenz method of B6yum (1983). When using the latter method, 3 ml portions of blood (using EDTA as anticoagulant) were layered over Nycodenz of density 1.078 g / m l or 1.069 g/ml. The monocyte yield was calculated (Leb et al., 1983) as yield (%) number of leucocytes present x monocyte abundance (%) number of monocytes in original blood sample
The total number of leucocytes in 'mononuclear cell' preparations was determined using a Neubauer counting chamber; and the abundance of monocytes, expressed as a percent of total leucocytes, by microscopic examination of Giemsastained smears. The number of monocytes in the original blood sample was taken as 5% of its total leucocyte population, as estimated with a Coulter counter.
Cell viability was determined by trypan blue exclusion (Tullis, 1953).
Scale-up of the Percoll method The Percoll method was scaled-up over ten-fold, as follows. Blood (180 ml) was collected from healthy volunteers and transferred immediately into a Duran bottle (500 ml capacity) containing approximately ten bent metal paperclips, and defibrinated by gentle rotation of the bottle for approximately 15 min (fibrin adheres to the clips, thus preventing clot formation). Into each of five conical 50 ml polystyrene tubes were placed 18 ml of Percoll (Pharmacia, diluted to a density of 1.076 g / m l ) in 0.15 M NaC1. Into each tube defibrinated blood (31.5 ml) was carefully layered over the Percoll. The tubes were centrifuged (15 ° C, 800 x g, 30 min), and the band of 'mononuclear cells' collected, washed and resuspended in culture medium (for composition, see below), as described by Pertoft et al. (1980). Culture of monocytes in vitro 'Mononuclear cell' preparations were suspended in prewarmed culture medium 199 (Gibco Europe, Uxbridge, U.K.) containing penicillinstreptomycin 200 U / m l ; Fungizone 2.5 /zg/ml; gentamycin 200 tzg/ml; fetal calf and horse serum, heat-inactivated, 10% v / v of each (Flow Laboratories, Hertfordshire, U.K.), to a density of 3-4 x 10 6 cells/ml. Portions (1 ml) were placed into wells of 24-well Linbro plates. In some experiments cells were seeded into Leighton tubes containing glass coverslips. Incubation was at 3 7 ° C in an atmosphere of air: CO z (95:5). After a 2.5-3 h period, non-adherent cells were removed by gentle aspiration of the culture medium, using a long-form Pasteur pipette attached to a mechanical pump. The cells remaining attached were washed three times with culture medium (1 ml portions, 37 ° C). Finally 1.0 ml of culture medium was added to each well (or Leighton tube), and incubation continued; the durations of incubation subsequently quoted were timed from this point. Cells remaining in culture beyond 96 h had half of their culture medium replaced by 0.5 ml of (warmed) fresh culture medium at this time and, if necessary, at weekly intervals thereafter.
11
Identification of cells after culture for 24 h Morphological identification. Cell monolayers
determined, using a stage micrometer to calibrate the graticule eyepiece.
in Linbro wells were air-dried, fixed in methanol (100%, 15 min) and stained with Giemsa solution (one drop of Giemsa stain in 1 ml water; 20 min). The monolayers were washed with distilled water and examined microscopically ( × 1500); 200 cells were examined and classified by their morphological appearance as monocytes, lymphocytes or polymorphonuclear cells. Cytochemical identification. Cells grown on glass coverslips in Leighton tubes were scored for positive a-naphthyl acetate esterase activity, using the Sigma research kit 90-A1. Functional identification. Phagocytic ability was assessed using polystyrene beads (1.1 /~m diameter), according to the method and criteria of Michl et al. (1976).
Lysosomal enzyme release during phagocytic challenge. The experimental procedure described
Criteria used to monitor differentiation in vitro Cell DNA and protein. Cells were washed three times with medium 199, and water (250/~1) added to each well before inducing lysis by sonication at 4 ° C (amplitude 10 /~m, 15 s), using an M.S.E. probe-sonicator. Trichloroacetic acid (10% w / v , 250 #1) was added to the sonified suspensions to precipitate acid-insoluble material. After storage, overnight at 4 ° C, the samples were centrifuged (20 min, 4°C, 8 0 0 × g ) . The precipitates were resuspended in trichloroacetic acid (5%, 4 ° C, 500 /~1) and hydrolysed at 9 0 ° C for 30 min, then cooled on ice. Samples were centrifuged (800 × g, 20 min, 4 ° C ) and the supernatants containing solubilized D N A aldehydes reserved at 4 ° C for D N A determinations by the fluorometric method of Erwin et al. (1981); the precipitates were retained at 4 ° C for protein determination by the method of Lowry et al. (1951). Cell number. The method of Unkeless and Eisen (1975) was used to obtain lysates containing cell nuclei. Lysates (100/~1) were diluted in Isoton (Coulter Electronics, 10 ml) and the number of nuclei determined using a model ZBI Coulter counter. Cell-size. Cells growing in Linbro wells were examined microscopically (x300), using an inverted microscope fitted with an eyepiece graftcule. The sizes of 40 randomly selected cells was
by Knight and Soutar (1982) was used in order to measure hexosaminidase release from cells challenged with either opsonized zymosan or (unopsonized) polystyrene beads (1.1 /tm diameter). A minor modification was opsonization of zymosan with fetal calf serum rather than human serum; the latter has been said to yield a product that may be toxic to cells (Leoni, 1985). A method (Griffiths et al., 1978), using a 4-methylumbelliferyl substrate at p H 4.3 and 37 ° C, was used to determine N-acetyl-fl-glucosaminidase activity in medium and in cell samples; the cells were washed with medium 199 (twice) and lysed with Triton X-100 (0.05% v / v , 0.5 ml). Cell protein was determined as described by Lowry et al. (1951) (with the bovine serum albumin standards diluted in Triton X-100). Peroxidase activity. A colorimetric method based on oxidation of 4 - a m i n o a n t i p y r i n e (Worthington Diagnostics) was used. Cell samples were washed three times with medium 199 (prewarmed at 37 ° C), water (250/~1) was added and the cells lysed by sonication as described above. Samples (100/zl) were assayed within two hours. Equivolume sodium hydroxide (1 mol/1) was added to the remaining portions of the sonicated cell samples and these were retained for protein determinations. Samples (100/~1) of culture medium were also assayed for peroxidase. Esterase activity. Cells, grown on glass coverslips (in Leighton tubes) rather than in plastic wells, were examined on day 1 and day 7 to determine the extent to which their cytochemically detectable a-naphthyl acetate esterase was inhibited by sodium fluoride (1.5 mg/ml).
Results
Table I compares the yield and purity of 'mononuclear cell' preparations obtained from blood fractionated on Percoll or Nycodenz. Cell viability was high in all cases, indicating that the fractionation media are non-toxic. The Nycodenz method yielded 'mononuclear cell' fractions of
12 TABLE I A COMPARISON OF METHODS FOR THE PRODUCTION OF A CULTURE OF HUMAN MONOCYTES Each value represents the mean ± SE from six donors Percoll density gradient Viable leucocytes in 'mononuclear cell' preparations (% of total number of leukocytes present) Monocyte abundance in ' mononuclear cell' preparations (% of total leucocytes) Yield of monocytes in ' mononuclear cell' preparations (% of total monocytes present in blood sample) Number of monocytes in monolayers after 24 h culture (% of total cells present) as assessed by: morphology in Giemsa-stained preparations a-naphthyl esterase activity phagocytic ability (polystyrene beads)
Nycodenz density gradient Density 1.076 g / m l
Density 1.069 g/ml
98.4± 0.5
99.3±0.7
99.3±0.5
21.8± 0.6
90.0±2.4
97.5±0.7
115.4±27.5
15.9±3.1
46.0±5.8
95.0± 0.6 97.0± 0.6 95.0± 1.3
94.0±0.4 96.0±0.3 95.0±0.4
95.0±1.1 98.0±0.8 95.0±0.5
high purity, but the yield was low, particularly with the Nycodenz of higher density. The Percoll fractionation yielded a 'mononuclear cell' fraction that apparently contained the entire monocyte population, but it was heavily contaminated with other leucocytes, chiefly lymphocytes. However, only the monocytes are adherent and culture for 2.5-3 h followed by washing and a further 24 h culture yielded a monocyte monolayer of high purity (Table I). The Percoll method was therefore the method chosen for scale-up. The scaled-up Percoll method yielded monocyte monolayers of equivalent purity to those described above (results not shown). In subsequent work this method routinely yielded enough cells to seed approximately 48 Linbro wells and the entire
procedure could be completed in 4 h from collection of the blood. Table II shows data on monolayers of monocytes maintained in culture for periods up to 7 days. There was a good correlation, as expected, between cell number and DNA content, and the only major cell loss occurred in the first 2 days of culture. The protein content per cell was constant for the first 2 days and then rose. Day 0 cells were typically rounded in appearance, whereas day 7 cells displayed a variety of shapes and sizes - rounded, triangular, epitheloid, fusiform - and a number of the cells produced long thin cytoplasmic projections. Table III shows the size of cells in the monolayers. The increase in size after 2 days paralleled the in-
TABLE II CELL NUMBER, DNA CONTENT AND PROTEIN CONTENT OF HUMAN MONOCYTES DIFFERENTIATING IN VITRO Each value (mean + SE) represents data from three donors, cells from each donor grown in duplicate wells Time in culture (days)
Cell number ( × 10- 5) per well
DNA (/zg) per well
Protein (/~ g) per well
Protein : DNA ratio
0 1 2 5 7
7.1±0.1 5.0±0.5 3.9±0.4 3.6±0.3 3.6±0.3
7.1±0.4 4.8±0.4 3.9±0.3 3.8±0.2 3.8±0.2
37±1 28±2 22±3 28±1 37±1
5.3±0.3 5.8±0.2 5.6±0.3 7.6±0.5 9.9±0.5
13 T A B L E III SIZE O F H U M A N M O N O C Y T E S D I F F E R E N T I A T I N G IN VITRO Cells were obtained from a single donor, a typical example. Each value represents the mean of 40 measurements ± standard error.
~r-4
Time in culture (days)
Size ( ~ m )
0 1 2 3 4 5 6 7
29.4±0.7 14.2±0.6 17.3±1.2 22.0±2.0 35.1±1.9 45.0±2.4 58.7±2.5 59.0±3.4
• ,--I N
|
~.~ •O
o I
I
3
6
Time i n c u l t u r e
crease in protein content already described. Cells immediately after isolation were 2-3-fold larger than after 24 h in culture: this contraction was not accompanied by any change in protein content and was presumably an artefact caused by osmotic factors. Fig. 1 shows the hexosaminidase content of monocyte monolayers cultured for periods up to 12 days. A three-fold increase was regularly observed. Fig. 2 shows the hexosaminidase released into the culture medium from cells challenged for 4 h with opsonized zymosan or unopsonized poly-
•,~ ~
10
~o •~
5
o ~
0
u~
0
•
I
a
I
I
3
6
9
12
Time in culture (days) Fig. I. Specific activity( U / m g cellprotein) of N-acctyl-fl-glucosaminidasc in Pcrcoll-isolatedhuman monocytcs maintained in culture for periods up to 12 days. The results of a typical experiment, using cellsfrom a singlehuman subject,arc shown. Each point represents the enzyme activity of the cells in one well. One unit of enzyme activitycorresponds to thc release of 1/x tool of 4-methylumbelliferone/h.
B
8
I
I
9
12
(days)
Fig. 2. Release of N-acetyl-fl-glucosaminidase from Percoll-isolated h u m a n monocytes maintained in culture for periods up to 12 days. Cells at a given age were incubated for 4 h with 0.5 ml of either opsonized zymosan; 0.38 m g / m l in culture medium ( 0 ) or polystyrene beads; 4 × 108 b e a d s / m l in culture medium (O) or culture m e d i u m alone (o). The results of a typical experiment, using cells from a single h u m a n subject, are shown. Each point represents the enzyme activity released by the cells in a single well, expressed as units per m g cell protein. One unit of enzyme activity corresponds to the release of 1 # mol of 4-methylumbelliferone/h.
styrene beads at various stages of maturation. Enzyme release using zymosan increased with the age of the cells (and thus with the enzyme content of the cells), some 50% being released at all stages of maturation. With polystyrene beads, enzyme release was lower than with opsonized zymosan, particularly after day 5. Little enzyme was released in the absence of a phagocytic stimulus. Table IV shows the peroxidase activity of cells after different periods in culture. Activity was progressively lost from the cells, becoming undetectable beyond 3 days of culture. No measurable release of peroxidase into the culture medium was observed. 97% of the cells showed cytochemically demonstrable esterase activity on day 1, but this value decreased to 6% if sodium fluoride (1.5 m g / m l ) was present. 98% of the cells were esterase-positive
14 TABLE IV PEROXIDASE C O N T E N T OF CELLS AT VARIOUS AGES Each value represents the mean _+standard error of nine determinations (triplicate wells for each of three donors) Time in
. Specific activity of peroxidase in
culture (days)
U/well a
U / m g cell-protein
0 1 2 3 4 7
0.018+0.001 0.009+0.001 0.007+0.001 0.004+0.0005 < 0.004 < 0.004
0.24+0.02 0.18+0.02 0.16+0.02 0.13_+0.01
a 1 U = the decomposition of 1 /tmol hydrogen peroxide/min at 25 o C, pH 7.0.
on day 7, but fluoride was ineffective at abolishing the activity.
Discussion
We have shown that it is possible to scale up the Percoll method of Pertoft et al. (1980) for isolating monocytes from human blood. Over 95% of the cells present in the monolayers after an initial 24 h incubation were recognizably monocytes, using as criteria their characteristic morphology, the presence of a-naphthyl esterase activity, and the ability to capture polystyrene beads by phagocytosis. The scaled-up method proved to be very reproducible as a source of human monocytes in high yield. We have also demonstrated that monocytes isolated by the scaled-up Percoll method differentiate into macrophages when incubated in vitro. In assessing this we have monitored a range of maturation markers used by previous workers. The most obvious indication that monocytes are maturing in vitro is their change in size. Lewis (1925), using hanging-drop cultures of human blood, noted an increase in monocyte size, which occurred after 3 - 4 days. Johnson et al. (1977), culturing monocytes isolated on Ficoll-Hypaque gradients, reported a rapid decrease (up to 50%) from the initial cell diameter of 30-40/~m within the first day, followed by a doubling in cell diameter between days 3 and 7. Johnson et al. (1977) also demonstrated a correlation between cell n u m -
ber and protein content in a monocyte population cultured in vitro. Both parameters decreased by approximately 50% of the initial values by day 4, after which the cell number continued to decrease (very slowly), whereas the protein content steadily increased. These data suggest that the protein content per cell remains stable up to day 4, after which it increases. Similar data have been reported by other authors (Musson et al., 1980; Nakagawara et al., 1981; O'Dorisio et al., 1984). The Percoll-isolated cells that we studied displayed this characteristi~ behaviour, consistent with maturation into macrophages. Zuckerman et al. (1979) described in detail the morphological changes seen in monocyte populations differentiating into macrophages in vitro. When introduced into culture monocytes were rounded in appearance, whereas after 3-6 days in culture the population was very heterogeneous, containing large flat cells that were triangular, round, spindle-shaped or epitheloid. The older cells had very ruffled membranes and sometimes filopodia and long cytoplasmic projections extending from one pole. Multinucleate giant cells were often observed after day 3 in culture. Our cultures displayed the same characteristics as those of Zuckerman et al. (1979). As monocytes are maintained in culture, the specific activity of N-acetyl-fl-ghicosaminidase increases. Epstein et al. (1981) and Knight and Soutar (1982) have each reported a 2-3-fold increase, using dextran sedimentation and FicollHypaque isolation methods respectively. In our experiments an increase in the enzyme activity of the same magnitude was observed. N-acetyl-fl-glucosaminidase is also released from monocytes in culture, and this release is greatly enhanced by a phagocytic stimulus (Musson et al., 1980; Knight and Soutar, 1982). Both investigations showed that the amount of enzyme released increased with the duration of culture and, moreover, beyond day 5 the response to opsonized zymosan (a phagocytic stimulus with chronic inflammatory potency) was much greater than that to unopsonized polystyrene beads. In our experiments the divergence of response to these two particulate stimuli was also seen at time points beyond day 5. Johnson et al. (1977) reported a progressive decrease in biochemically assayed peroxidase ac-
15 tivity as monocytes were m a i n t a i n e d in culture; b y day 3 the enzyme activity was less t h a n 20% of the value o n day 0. N o peroxidase was f o u n d in the m e d i u m . The disappearance of the enzyme paralleled the disappearance of cytochemically detectable peroxidase-positive granules in the cells. The cells in our study showed similar characteristics: peroxidase activity decreased rapidly over the first 3 days in culture, a n d was u n d e t e c t a b l e on day 4 a n d beyond. A q u a n t i t a t i v e c o m p a r i s o n of results was not possible, since J o h n s o n et al. (1977) expressed their results in arbitrary units. M u s s o n (1983) reported that when m o n o c y t e s were cultured in vitro, their esterase staining activity developed a resistance to fluoride inhibition. The cytochemically d e m o n s t r a b l e esterase in our cells showed fluoride sensitivity o n day 1 b u t n o t o n day 7. I n s u m m a r y we have shown that the m o n o c y t e p o p u l a t i o n isolated b y the Percoll m e t h o d is able to differentiate reliably into macrophages when cultured in vitro u n d e r suitable conditions. The criteria used to c o n f i r m m o n o c y t e - t o - m a c r o p h a g e m a t u r a t i o n encompassed widely accepted m o r p h o logical, cytochemical, biochemical a n d f u n c t i o n a l tests, and each gave a positive result. Since we have also b e e n able to scale up the Percoll method, the procedures described in this paper constitute a reliable technique for the p r o d u c t i o n a n d study of h u m a n macrophages in increased yield. The cont i n u e d use of the older F i c o l l - H y p a q u e m e t h o d for studies o n m o n o c y t e m a t u r a t i o n p r o b a b l y derives from the previous lack of evidence for m a t u r a t i o n of Percoll-isolated cells, and also perhaps from fears a b o u t the possible toxicity of Percoll (Wakefield et al., 1982). We believe that the results presented here justify more widespread use of the rapid and simple Percoll method.
Acknowledgement D . E D . thanks the Science a n d E n g i n e e r i n g Research Council for a Research Studentship.
References BOyum, A. (1983) Isolation of human blood monocytes with Nycodenz, a new non-ionic iodinated gradient medium. Scand. J. Immunol. 17, 429.
Epstein, L.B., Yu, K., Chong, L.P. and Reese, C.C. (1981) Culture of human monocytes in microplates and enzymatic assays for following their maturation. In: P.J. Edelson, H. Koren and D.O. Adams (Eds.), Methods for Studying Mononuclear Phagocytes. Academic Press, New York, p. 49. Erwin, B.G., Stoscheck, C.M. and Florini, J.R. (1981) A rapid fluorometric method for the estimation of DNA in cultured cells. Anal. Biochem. 110, 291. Fluks, A.J. (1981) Three-step isolation of human blood monocytes using discontinuous density gradients of Percoll. J. Immunol. Methods 41,225. Griffiths, P.A., Milson, J.P. and Lloyd, J.B. (1978) Plasma acid hydrolases in normal adults and children, and in patients with some lysosomal storage diseases. Clin. Chim. Acta 90, 129. Johnson, W.D., Mei, B. and Cohn, Z.A. (1977) The separation, long-term cultivation, and maturation of the human monocyte. J. Exp. Med. 146, 1613. Knight, B.L and Soutar, A.K. (1982) Changes in the metabolism of modified and unmodified low-density lipoproteins during the maturation of cultured blood monocyte-macrophages from normal and homozygous familial hypercholesterolaemic subjects. Eur. J. Biochem. 125, 407. Leb, L., Crusberg, T., Fortier, N. and Snyder, L.M. (1983) Evaluation of methods using adherence to substrate and density gradient for the isolation of human monocytes. J. Immunol. Methods 58, 309. Lewis, M.R. (1925) The formation of macrophages, epitheloid cells and giant cells from leucocytes in incubated blood. Am. J. Pathol. 1, 91. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265. Michl, S., Ohlbaum, D.J. and Silverstein, S.C. (1976) 2-Deoxyglucose selectively inhibits Fc and complement receptor-mediated phagocytosis in mouse peritoneal macrophages. I. Description of the inhibitory effect. J. Exp. Med. 144, 1465. Musson, R.A. (1983) Human serum induces maturation of human monocytes in vitro. Am. J. Pathol. 111, 331. Musson, R.A., Shafran, H. and Henson, P.M. (1980) Intracellular levels and stimulated release of lysosomal enzymes from human peripheral blood monocytes and monocyte-derived macrophages. J. Reticuloendothel. Soc. 28, 249. Nakagawara, A., Nathan, C.F. and Cohn, Z.A. (1981) Hydrogen peroxide metabolism in human monocytes during differentiation in vitro. J. Clin. Invest. 68, 1243. O'Dorisio, M.S., Fertel, R., Finkler, E., Brooks, R. and Vassalo, L. (1984) Characterization of cyclic nucleotide metabolism during human monocyte differentiation. J. Leukocyte Biol. 35, 617. Pertoft, H., Johnsson, A., Warmegard, B. and Seljelid, R. (1980) Separation of human monocytes on density gradients of Percoll R j. Immunol. Methods 33, 221. Tullis, J.L (1953) Preservation of leukocytes. Blood 8, 563. Ulmer, A.J. and Flad, H.D. (1979) Discontinuous density gradient separation of human mononuclear leucocytes using Percoll R as gradient material. J. Immunol. Methods 30, 1.
16 Unkeless, J.C. and Eisen, H.N. (1975) Binding of monomeric immunoglobulins to Fc receptors of mouse macrophages. J. Exp. Med. 142, 1520. Wakefield, J.St.J., Gale, J.S., Berridge, M.V., Jordan, T.W. and Ford, H.C. (1982) Is Percoll innocuous to cells? Biochem. J. 202, 795.
Zuckerman, S.H., Ackerman, S.K. and Douglas, S.D. (1979) Long-term human peripheral blood monocyte cultures: establishment, metabolism and morphology of primary human monocyte-macrophage cell cultures. Immunology 38, 401.
9
Elsevier JIM 05078
Monocyte-to-macrophage transition in vitro A systematic study using human cells isolated by fractionation on Percoll Diane E. Davies and John B. Lloyd Department of Biological Sciences, University of Keele, Keele, Staffordshire ST5 5BG, U.K. (Received 3 August 1988, revised received 12 October 1988, accepted 18 October 1988)
Improved density-gradient methods, using Percoll or Nycodenz, have recently been introduced for the isolation of human monocytes, but the capacity of cells thus isolated to differentiate into macrophages has not been systematically studied. We have compared Percoll and Nycodenz methods for the isolation of monocytes from human blood. The Nycodenz method yielded a monocyte population of high purity, but the yield was low. The Percoll method gave almost quantitative yield of monocytes, and the contaminating cells, mostly lymphocytes, were readily washed away after allowing the monocytes to adhere to a plastic surface. The Percoll method was then successfully scaled up, providing a simple method to obtain the monocytes from 180 ml blood. These monocytes were maintained in culture and their capacity to mature into macrophages was studied, using the following criteria: increase in cell size and protein content, increase in specific activity of hexosaminidase, differential hexosaminidase release on exposure to opsonized zymosan and unopsonized polystyrene beads, loss of peroxidase activity, and development of fluoride-insensitivity by the cells' cytochemically demonstrable esterase. The cells also displayed morphological changes typical of the monocyte-to-macrophage transition. The procedures reported constitute a simple and reliable method for the production of human macrophages in increased yield. Key words: Percoll; Nycodenz; Monocyte; Macrophage; Differentiation
Introduction
Studies on human macrophages are severely hampered by the problems of obtaining these cells. An attractive alternative source of macrophages is the monocyte population of peripheral blood, followed by monocyte-to-macrophage differentiation in vitro. This approach requires a satisfactory method for obtaining human monocytes in good yield and culture conditions that reliably lead to monocyte maturation. Correspondence to: J.B. Lloyd, Department of Biological Sciences, University of Keele, Keele, Staffordshire ST5 5BG, U.K.
There are many differences, morphological, histochemical and biochemical, between monocytes and macrophages, and these can be used as criteria to monitor monocyte maturation in vitro. Thus human monocytes, isolated mostly by methods using Ficoll-Hypaque gradients, were shown to differentiate into macrophages over a period of 5-7 days, when cultured under suitable conditions (Johnson et al., 1977; Zuckerman et al., 1979). More recently alternative density-gradient media, Percoll (Ulmer and Flad, 1979; Pertoft et al., 1980) and Nycodenz (B~Syum, 1983), have been used for monocyte isolation; it is claimed that these avoid the considerable disadvantages of the earlier Ficoll-Hypaque methods.
0022-1759/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
10
Although there are references to the in vitro maturation of human monocytes isolated using Percoll or Nycodenz (Fluks, 1981; B~Syum, 1983), little or no evidence for this differentiation has been presented. This may account for the apparent reluctance of many investigators to adopt these newer methods. In the present paper we first compare Percoll and Nycodenz methods and conclude that the former is a better source of monocytes for subsequent culture in vitro. We then describe a scale-up of the Percoll method of Pertoft et al. (1980) and use a range of criteria, chosen from those used by previous workers, to monitor the in vitro maturation of the monocytes isolated. The work described constitutes the first systematic study of the monocyte-to-macrophage transition in vitro of human monocytes isolated by a Percoll method: we conclude that the methods described are suitable for studies of cell function during and after this differentiation process.
Methods
Isolation of a 'mononuclear cell' population from human peripheral blood Mononuclear cells were obtained by either the one-step Percoll method of Pertoft et al. (1980) or the Nycodenz method of B6yum (1983). When using the latter method, 3 ml portions of blood (using EDTA as anticoagulant) were layered over Nycodenz of density 1.078 g / m l or 1.069 g/ml. The monocyte yield was calculated (Leb et al., 1983) as yield (%) number of leucocytes present x monocyte abundance (%) number of monocytes in original blood sample
The total number of leucocytes in 'mononuclear cell' preparations was determined using a Neubauer counting chamber; and the abundance of monocytes, expressed as a percent of total leucocytes, by microscopic examination of Giemsastained smears. The number of monocytes in the original blood sample was taken as 5% of its total leucocyte population, as estimated with a Coulter counter.
Cell viability was determined by trypan blue exclusion (Tullis, 1953).
Scale-up of the Percoll method The Percoll method was scaled-up over ten-fold, as follows. Blood (180 ml) was collected from healthy volunteers and transferred immediately into a Duran bottle (500 ml capacity) containing approximately ten bent metal paperclips, and defibrinated by gentle rotation of the bottle for approximately 15 min (fibrin adheres to the clips, thus preventing clot formation). Into each of five conical 50 ml polystyrene tubes were placed 18 ml of Percoll (Pharmacia, diluted to a density of 1.076 g / m l ) in 0.15 M NaC1. Into each tube defibrinated blood (31.5 ml) was carefully layered over the Percoll. The tubes were centrifuged (15 ° C, 800 x g, 30 min), and the band of 'mononuclear cells' collected, washed and resuspended in culture medium (for composition, see below), as described by Pertoft et al. (1980). Culture of monocytes in vitro 'Mononuclear cell' preparations were suspended in prewarmed culture medium 199 (Gibco Europe, Uxbridge, U.K.) containing penicillinstreptomycin 200 U / m l ; Fungizone 2.5 /zg/ml; gentamycin 200 tzg/ml; fetal calf and horse serum, heat-inactivated, 10% v / v of each (Flow Laboratories, Hertfordshire, U.K.), to a density of 3-4 x 10 6 cells/ml. Portions (1 ml) were placed into wells of 24-well Linbro plates. In some experiments cells were seeded into Leighton tubes containing glass coverslips. Incubation was at 3 7 ° C in an atmosphere of air: CO z (95:5). After a 2.5-3 h period, non-adherent cells were removed by gentle aspiration of the culture medium, using a long-form Pasteur pipette attached to a mechanical pump. The cells remaining attached were washed three times with culture medium (1 ml portions, 37 ° C). Finally 1.0 ml of culture medium was added to each well (or Leighton tube), and incubation continued; the durations of incubation subsequently quoted were timed from this point. Cells remaining in culture beyond 96 h had half of their culture medium replaced by 0.5 ml of (warmed) fresh culture medium at this time and, if necessary, at weekly intervals thereafter.
11
Identification of cells after culture for 24 h Morphological identification. Cell monolayers
determined, using a stage micrometer to calibrate the graticule eyepiece.
in Linbro wells were air-dried, fixed in methanol (100%, 15 min) and stained with Giemsa solution (one drop of Giemsa stain in 1 ml water; 20 min). The monolayers were washed with distilled water and examined microscopically ( × 1500); 200 cells were examined and classified by their morphological appearance as monocytes, lymphocytes or polymorphonuclear cells. Cytochemical identification. Cells grown on glass coverslips in Leighton tubes were scored for positive a-naphthyl acetate esterase activity, using the Sigma research kit 90-A1. Functional identification. Phagocytic ability was assessed using polystyrene beads (1.1 /~m diameter), according to the method and criteria of Michl et al. (1976).
Lysosomal enzyme release during phagocytic challenge. The experimental procedure described
Criteria used to monitor differentiation in vitro Cell DNA and protein. Cells were washed three times with medium 199, and water (250/~1) added to each well before inducing lysis by sonication at 4 ° C (amplitude 10 /~m, 15 s), using an M.S.E. probe-sonicator. Trichloroacetic acid (10% w / v , 250 #1) was added to the sonified suspensions to precipitate acid-insoluble material. After storage, overnight at 4 ° C, the samples were centrifuged (20 min, 4°C, 8 0 0 × g ) . The precipitates were resuspended in trichloroacetic acid (5%, 4 ° C, 500 /~1) and hydrolysed at 9 0 ° C for 30 min, then cooled on ice. Samples were centrifuged (800 × g, 20 min, 4 ° C ) and the supernatants containing solubilized D N A aldehydes reserved at 4 ° C for D N A determinations by the fluorometric method of Erwin et al. (1981); the precipitates were retained at 4 ° C for protein determination by the method of Lowry et al. (1951). Cell number. The method of Unkeless and Eisen (1975) was used to obtain lysates containing cell nuclei. Lysates (100/~1) were diluted in Isoton (Coulter Electronics, 10 ml) and the number of nuclei determined using a model ZBI Coulter counter. Cell-size. Cells growing in Linbro wells were examined microscopically (x300), using an inverted microscope fitted with an eyepiece graftcule. The sizes of 40 randomly selected cells was
by Knight and Soutar (1982) was used in order to measure hexosaminidase release from cells challenged with either opsonized zymosan or (unopsonized) polystyrene beads (1.1 /tm diameter). A minor modification was opsonization of zymosan with fetal calf serum rather than human serum; the latter has been said to yield a product that may be toxic to cells (Leoni, 1985). A method (Griffiths et al., 1978), using a 4-methylumbelliferyl substrate at p H 4.3 and 37 ° C, was used to determine N-acetyl-fl-glucosaminidase activity in medium and in cell samples; the cells were washed with medium 199 (twice) and lysed with Triton X-100 (0.05% v / v , 0.5 ml). Cell protein was determined as described by Lowry et al. (1951) (with the bovine serum albumin standards diluted in Triton X-100). Peroxidase activity. A colorimetric method based on oxidation of 4 - a m i n o a n t i p y r i n e (Worthington Diagnostics) was used. Cell samples were washed three times with medium 199 (prewarmed at 37 ° C), water (250/~1) was added and the cells lysed by sonication as described above. Samples (100/zl) were assayed within two hours. Equivolume sodium hydroxide (1 mol/1) was added to the remaining portions of the sonicated cell samples and these were retained for protein determinations. Samples (100/~1) of culture medium were also assayed for peroxidase. Esterase activity. Cells, grown on glass coverslips (in Leighton tubes) rather than in plastic wells, were examined on day 1 and day 7 to determine the extent to which their cytochemically detectable a-naphthyl acetate esterase was inhibited by sodium fluoride (1.5 mg/ml).
Results
Table I compares the yield and purity of 'mononuclear cell' preparations obtained from blood fractionated on Percoll or Nycodenz. Cell viability was high in all cases, indicating that the fractionation media are non-toxic. The Nycodenz method yielded 'mononuclear cell' fractions of
12 TABLE I A COMPARISON OF METHODS FOR THE PRODUCTION OF A CULTURE OF HUMAN MONOCYTES Each value represents the mean ± SE from six donors Percoll density gradient Viable leucocytes in 'mononuclear cell' preparations (% of total number of leukocytes present) Monocyte abundance in ' mononuclear cell' preparations (% of total leucocytes) Yield of monocytes in ' mononuclear cell' preparations (% of total monocytes present in blood sample) Number of monocytes in monolayers after 24 h culture (% of total cells present) as assessed by: morphology in Giemsa-stained preparations a-naphthyl esterase activity phagocytic ability (polystyrene beads)
Nycodenz density gradient Density 1.076 g / m l
Density 1.069 g/ml
98.4± 0.5
99.3±0.7
99.3±0.5
21.8± 0.6
90.0±2.4
97.5±0.7
115.4±27.5
15.9±3.1
46.0±5.8
95.0± 0.6 97.0± 0.6 95.0± 1.3
94.0±0.4 96.0±0.3 95.0±0.4
95.0±1.1 98.0±0.8 95.0±0.5
high purity, but the yield was low, particularly with the Nycodenz of higher density. The Percoll fractionation yielded a 'mononuclear cell' fraction that apparently contained the entire monocyte population, but it was heavily contaminated with other leucocytes, chiefly lymphocytes. However, only the monocytes are adherent and culture for 2.5-3 h followed by washing and a further 24 h culture yielded a monocyte monolayer of high purity (Table I). The Percoll method was therefore the method chosen for scale-up. The scaled-up Percoll method yielded monocyte monolayers of equivalent purity to those described above (results not shown). In subsequent work this method routinely yielded enough cells to seed approximately 48 Linbro wells and the entire
procedure could be completed in 4 h from collection of the blood. Table II shows data on monolayers of monocytes maintained in culture for periods up to 7 days. There was a good correlation, as expected, between cell number and DNA content, and the only major cell loss occurred in the first 2 days of culture. The protein content per cell was constant for the first 2 days and then rose. Day 0 cells were typically rounded in appearance, whereas day 7 cells displayed a variety of shapes and sizes - rounded, triangular, epitheloid, fusiform - and a number of the cells produced long thin cytoplasmic projections. Table III shows the size of cells in the monolayers. The increase in size after 2 days paralleled the in-
TABLE II CELL NUMBER, DNA CONTENT AND PROTEIN CONTENT OF HUMAN MONOCYTES DIFFERENTIATING IN VITRO Each value (mean + SE) represents data from three donors, cells from each donor grown in duplicate wells Time in culture (days)
Cell number ( × 10- 5) per well
DNA (/zg) per well
Protein (/~ g) per well
Protein : DNA ratio
0 1 2 5 7
7.1±0.1 5.0±0.5 3.9±0.4 3.6±0.3 3.6±0.3
7.1±0.4 4.8±0.4 3.9±0.3 3.8±0.2 3.8±0.2
37±1 28±2 22±3 28±1 37±1
5.3±0.3 5.8±0.2 5.6±0.3 7.6±0.5 9.9±0.5
13 T A B L E III SIZE O F H U M A N M O N O C Y T E S D I F F E R E N T I A T I N G IN VITRO Cells were obtained from a single donor, a typical example. Each value represents the mean of 40 measurements ± standard error.
~r-4
Time in culture (days)
Size ( ~ m )
0 1 2 3 4 5 6 7
29.4±0.7 14.2±0.6 17.3±1.2 22.0±2.0 35.1±1.9 45.0±2.4 58.7±2.5 59.0±3.4
• ,--I N
|
~.~ •O
o I
I
3
6
Time i n c u l t u r e
crease in protein content already described. Cells immediately after isolation were 2-3-fold larger than after 24 h in culture: this contraction was not accompanied by any change in protein content and was presumably an artefact caused by osmotic factors. Fig. 1 shows the hexosaminidase content of monocyte monolayers cultured for periods up to 12 days. A three-fold increase was regularly observed. Fig. 2 shows the hexosaminidase released into the culture medium from cells challenged for 4 h with opsonized zymosan or unopsonized poly-
•,~ ~
10
~o •~
5
o ~
0
u~
0
•
I
a
I
I
3
6
9
12
Time in culture (days) Fig. I. Specific activity( U / m g cellprotein) of N-acctyl-fl-glucosaminidasc in Pcrcoll-isolatedhuman monocytcs maintained in culture for periods up to 12 days. The results of a typical experiment, using cellsfrom a singlehuman subject,arc shown. Each point represents the enzyme activity of the cells in one well. One unit of enzyme activitycorresponds to thc release of 1/x tool of 4-methylumbelliferone/h.
B
8
I
I
9
12
(days)
Fig. 2. Release of N-acetyl-fl-glucosaminidase from Percoll-isolated h u m a n monocytes maintained in culture for periods up to 12 days. Cells at a given age were incubated for 4 h with 0.5 ml of either opsonized zymosan; 0.38 m g / m l in culture medium ( 0 ) or polystyrene beads; 4 × 108 b e a d s / m l in culture medium (O) or culture m e d i u m alone (o). The results of a typical experiment, using cells from a single h u m a n subject, are shown. Each point represents the enzyme activity released by the cells in a single well, expressed as units per m g cell protein. One unit of enzyme activity corresponds to the release of 1 # mol of 4-methylumbelliferone/h.
styrene beads at various stages of maturation. Enzyme release using zymosan increased with the age of the cells (and thus with the enzyme content of the cells), some 50% being released at all stages of maturation. With polystyrene beads, enzyme release was lower than with opsonized zymosan, particularly after day 5. Little enzyme was released in the absence of a phagocytic stimulus. Table IV shows the peroxidase activity of cells after different periods in culture. Activity was progressively lost from the cells, becoming undetectable beyond 3 days of culture. No measurable release of peroxidase into the culture medium was observed. 97% of the cells showed cytochemically demonstrable esterase activity on day 1, but this value decreased to 6% if sodium fluoride (1.5 m g / m l ) was present. 98% of the cells were esterase-positive
14 TABLE IV PEROXIDASE C O N T E N T OF CELLS AT VARIOUS AGES Each value represents the mean _+standard error of nine determinations (triplicate wells for each of three donors) Time in
. Specific activity of peroxidase in
culture (days)
U/well a
U / m g cell-protein
0 1 2 3 4 7
0.018+0.001 0.009+0.001 0.007+0.001 0.004+0.0005 < 0.004 < 0.004
0.24+0.02 0.18+0.02 0.16+0.02 0.13_+0.01
a 1 U = the decomposition of 1 /tmol hydrogen peroxide/min at 25 o C, pH 7.0.
on day 7, but fluoride was ineffective at abolishing the activity.
Discussion
We have shown that it is possible to scale up the Percoll method of Pertoft et al. (1980) for isolating monocytes from human blood. Over 95% of the cells present in the monolayers after an initial 24 h incubation were recognizably monocytes, using as criteria their characteristic morphology, the presence of a-naphthyl esterase activity, and the ability to capture polystyrene beads by phagocytosis. The scaled-up method proved to be very reproducible as a source of human monocytes in high yield. We have also demonstrated that monocytes isolated by the scaled-up Percoll method differentiate into macrophages when incubated in vitro. In assessing this we have monitored a range of maturation markers used by previous workers. The most obvious indication that monocytes are maturing in vitro is their change in size. Lewis (1925), using hanging-drop cultures of human blood, noted an increase in monocyte size, which occurred after 3 - 4 days. Johnson et al. (1977), culturing monocytes isolated on Ficoll-Hypaque gradients, reported a rapid decrease (up to 50%) from the initial cell diameter of 30-40/~m within the first day, followed by a doubling in cell diameter between days 3 and 7. Johnson et al. (1977) also demonstrated a correlation between cell n u m -
ber and protein content in a monocyte population cultured in vitro. Both parameters decreased by approximately 50% of the initial values by day 4, after which the cell number continued to decrease (very slowly), whereas the protein content steadily increased. These data suggest that the protein content per cell remains stable up to day 4, after which it increases. Similar data have been reported by other authors (Musson et al., 1980; Nakagawara et al., 1981; O'Dorisio et al., 1984). The Percoll-isolated cells that we studied displayed this characteristi~ behaviour, consistent with maturation into macrophages. Zuckerman et al. (1979) described in detail the morphological changes seen in monocyte populations differentiating into macrophages in vitro. When introduced into culture monocytes were rounded in appearance, whereas after 3-6 days in culture the population was very heterogeneous, containing large flat cells that were triangular, round, spindle-shaped or epitheloid. The older cells had very ruffled membranes and sometimes filopodia and long cytoplasmic projections extending from one pole. Multinucleate giant cells were often observed after day 3 in culture. Our cultures displayed the same characteristics as those of Zuckerman et al. (1979). As monocytes are maintained in culture, the specific activity of N-acetyl-fl-ghicosaminidase increases. Epstein et al. (1981) and Knight and Soutar (1982) have each reported a 2-3-fold increase, using dextran sedimentation and FicollHypaque isolation methods respectively. In our experiments an increase in the enzyme activity of the same magnitude was observed. N-acetyl-fl-glucosaminidase is also released from monocytes in culture, and this release is greatly enhanced by a phagocytic stimulus (Musson et al., 1980; Knight and Soutar, 1982). Both investigations showed that the amount of enzyme released increased with the duration of culture and, moreover, beyond day 5 the response to opsonized zymosan (a phagocytic stimulus with chronic inflammatory potency) was much greater than that to unopsonized polystyrene beads. In our experiments the divergence of response to these two particulate stimuli was also seen at time points beyond day 5. Johnson et al. (1977) reported a progressive decrease in biochemically assayed peroxidase ac-
15 tivity as monocytes were m a i n t a i n e d in culture; b y day 3 the enzyme activity was less t h a n 20% of the value o n day 0. N o peroxidase was f o u n d in the m e d i u m . The disappearance of the enzyme paralleled the disappearance of cytochemically detectable peroxidase-positive granules in the cells. The cells in our study showed similar characteristics: peroxidase activity decreased rapidly over the first 3 days in culture, a n d was u n d e t e c t a b l e on day 4 a n d beyond. A q u a n t i t a t i v e c o m p a r i s o n of results was not possible, since J o h n s o n et al. (1977) expressed their results in arbitrary units. M u s s o n (1983) reported that when m o n o c y t e s were cultured in vitro, their esterase staining activity developed a resistance to fluoride inhibition. The cytochemically d e m o n s t r a b l e esterase in our cells showed fluoride sensitivity o n day 1 b u t n o t o n day 7. I n s u m m a r y we have shown that the m o n o c y t e p o p u l a t i o n isolated b y the Percoll m e t h o d is able to differentiate reliably into macrophages when cultured in vitro u n d e r suitable conditions. The criteria used to c o n f i r m m o n o c y t e - t o - m a c r o p h a g e m a t u r a t i o n encompassed widely accepted m o r p h o logical, cytochemical, biochemical a n d f u n c t i o n a l tests, and each gave a positive result. Since we have also b e e n able to scale up the Percoll method, the procedures described in this paper constitute a reliable technique for the p r o d u c t i o n a n d study of h u m a n macrophages in increased yield. The cont i n u e d use of the older F i c o l l - H y p a q u e m e t h o d for studies o n m o n o c y t e m a t u r a t i o n p r o b a b l y derives from the previous lack of evidence for m a t u r a t i o n of Percoll-isolated cells, and also perhaps from fears a b o u t the possible toxicity of Percoll (Wakefield et al., 1982). We believe that the results presented here justify more widespread use of the rapid and simple Percoll method.
Acknowledgement D . E D . thanks the Science a n d E n g i n e e r i n g Research Council for a Research Studentship.
References BOyum, A. (1983) Isolation of human blood monocytes with Nycodenz, a new non-ionic iodinated gradient medium. Scand. J. Immunol. 17, 429.
Epstein, L.B., Yu, K., Chong, L.P. and Reese, C.C. (1981) Culture of human monocytes in microplates and enzymatic assays for following their maturation. In: P.J. Edelson, H. Koren and D.O. Adams (Eds.), Methods for Studying Mononuclear Phagocytes. Academic Press, New York, p. 49. Erwin, B.G., Stoscheck, C.M. and Florini, J.R. (1981) A rapid fluorometric method for the estimation of DNA in cultured cells. Anal. Biochem. 110, 291. Fluks, A.J. (1981) Three-step isolation of human blood monocytes using discontinuous density gradients of Percoll. J. Immunol. Methods 41,225. Griffiths, P.A., Milson, J.P. and Lloyd, J.B. (1978) Plasma acid hydrolases in normal adults and children, and in patients with some lysosomal storage diseases. Clin. Chim. Acta 90, 129. Johnson, W.D., Mei, B. and Cohn, Z.A. (1977) The separation, long-term cultivation, and maturation of the human monocyte. J. Exp. Med. 146, 1613. Knight, B.L and Soutar, A.K. (1982) Changes in the metabolism of modified and unmodified low-density lipoproteins during the maturation of cultured blood monocyte-macrophages from normal and homozygous familial hypercholesterolaemic subjects. Eur. J. Biochem. 125, 407. Leb, L., Crusberg, T., Fortier, N. and Snyder, L.M. (1983) Evaluation of methods using adherence to substrate and density gradient for the isolation of human monocytes. J. Immunol. Methods 58, 309. Lewis, M.R. (1925) The formation of macrophages, epitheloid cells and giant cells from leucocytes in incubated blood. Am. J. Pathol. 1, 91. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265. Michl, S., Ohlbaum, D.J. and Silverstein, S.C. (1976) 2-Deoxyglucose selectively inhibits Fc and complement receptor-mediated phagocytosis in mouse peritoneal macrophages. I. Description of the inhibitory effect. J. Exp. Med. 144, 1465. Musson, R.A. (1983) Human serum induces maturation of human monocytes in vitro. Am. J. Pathol. 111, 331. Musson, R.A., Shafran, H. and Henson, P.M. (1980) Intracellular levels and stimulated release of lysosomal enzymes from human peripheral blood monocytes and monocyte-derived macrophages. J. Reticuloendothel. Soc. 28, 249. Nakagawara, A., Nathan, C.F. and Cohn, Z.A. (1981) Hydrogen peroxide metabolism in human monocytes during differentiation in vitro. J. Clin. Invest. 68, 1243. O'Dorisio, M.S., Fertel, R., Finkler, E., Brooks, R. and Vassalo, L. (1984) Characterization of cyclic nucleotide metabolism during human monocyte differentiation. J. Leukocyte Biol. 35, 617. Pertoft, H., Johnsson, A., Warmegard, B. and Seljelid, R. (1980) Separation of human monocytes on density gradients of Percoll R j. Immunol. Methods 33, 221. Tullis, J.L (1953) Preservation of leukocytes. Blood 8, 563. Ulmer, A.J. and Flad, H.D. (1979) Discontinuous density gradient separation of human mononuclear leucocytes using Percoll R as gradient material. J. Immunol. Methods 30, 1.
16 Unkeless, J.C. and Eisen, H.N. (1975) Binding of monomeric immunoglobulins to Fc receptors of mouse macrophages. J. Exp. Med. 142, 1520. Wakefield, J.St.J., Gale, J.S., Berridge, M.V., Jordan, T.W. and Ford, H.C. (1982) Is Percoll innocuous to cells? Biochem. J. 202, 795.
Zuckerman, S.H., Ackerman, S.K. and Douglas, S.D. (1979) Long-term human peripheral blood monocyte cultures: establishment, metabolism and morphology of primary human monocyte-macrophage cell cultures. Immunology 38, 401.