British Iournal ojHaernatology, 1991. 79, 6-1 3
Productive infection of in vitro generated haemopoietic progenitor cells from normal human adult peripheral blood with parvovirus B19: studies by morphology, immunocytochemistry, flow-cytometry and DNA-hybridization STEFANSERKE,TINOFRANCISSCHWARZ,* HERRADBAURMANN, ANDREASKIRSCH, BARBARA HOTTENTRAGER,* ALBRECHT V O N BRUNN,*MICHAEL ROGGENDORF,* DIETERHUHNA N D FRIEDRICH DEINHARDT* Universitiitsklinikurn Rudolf Virchow-Charlottenburg, Abteilung Innere Medizin rnit Schwerpunkt Hiimatologie-Onkologie, Freie Universitiit Berlin, and *Max von Pettenkofer lnstitut f u r Medizinische Mikrobiologie und Hygiene, Abteilung Virologie, Ludwig-Maximilians Universitiit, Munchen, Germany
Received 26 February 1991; accepted for publication 16 M a y 1991
Summary. Parvovirus B19 exerts a highly selective cytopathic effect on erythroid progenitor cells. Studies so far on the pathogenesis of B19-infection have been performed using bone marrow samples providing large amounts of erythroid progenitor cells. Extensive study, however, has been hampered by the limited access to bone marrow samples. We have designed a liquid culture method allowing the generation of large numbers of erythroid progenitor cells, initiating cultures with CD3- and CD14-poor peripheral blood mononuclear cells. Following a 12 d preincubation in liquid cultures containing recombinant human interleukin 3 (rhII3 ) and recombinant human erythropoietin (rhEpo), cells harvested from the liquid cultures were exposed to B19-
containing plasma, followed by a further cultivation in liquid culture for up to 9 6 h. Cells expressing the CD13 and the glycophorin A (GlyA) antigens, respectively. were monitored sequentially by flow-cytometry, demonstrating a selective inhibition of GlyA-positive cells following B19-inoculation. Typical morphological changes were observed on cytocentrifuge-spots. and typical giant-cells were identified as staining for GlyA. Productive infection by B19 was demonstrable, as B19-DNA increased by about x 100 after 72 h of culture. The liquid culture method generating erythroid target cells for effective infection by B19 virus promises to be a useful and easily accessible tool for further research on B19 infection of haemopoietic cells.
With regard to the haemopoietic system, human parvovirus B19 exerts cytopathic effects almost exclusively upon cells of the erythroid lineage, leading to transient aplastic episodes in subjects with increased erythropoiesis (Pattison et a/. 1981). and to persistent long-standing anaemia in immunocompromised patients (Kurtman et al, 1987). In vitro studies have shown that B19-infection leads to inhibition of both erythroid burst-forming units (BFU-E) and erythroid colonyforming units (CFU-E). with no effect on granulocytic or monocytic precursors (Mortimer et a/, 1983: Ozawa et al, 1986; Takahashi et a / . 1990). Only recently, however,
Srivastava et al (1990) have shown that megakaryocytic colony formation was also inhibited by B19 infection, but no viral DNA replication was detected in bone marrow cells highly enriched for megakaryocytic progenitor cells. In vitro studies on B19 infection, so far. have relied on infection of fetal liver cells, or of bone marrow cells, both types of tissues not available o n a routine base. Moreover, no permanent cell line has been identified as yet, capable of productive infection by B19 (Mortimer et a/, 1983). In order to facilitate further the in vitro research on B19caused alterations of haemopoiesis, we have developed a liquid culture method allowing the generation of large numbers of erythroid progenitors capable of productive infection by B19, initiating cultures with peripheral blood mononuclear cells from normal human adult donors (Serke e t a / , 1991).
Correspondence: Dr med. Stefan Serke. UKRV-Charlottenburg, Hamatologie. Spandauer Damm 1 30. 1000 Berlin 19. Germany.
6
Eflective lnfection of PBMC Progenitors by BI 9
7
Number of anti-CDl3-~os. cells ( x 10000)
60
.
.
-.
50 40
30
20 10
0
111
I Number of anti-GlvA-Dos. cells ( x 10000) 60
i-1
........ ........ ........ ......... ......... ......... ......... ......... ......... ......... ......... ......... .........
50
40
30
20 10
Fig 1. Time courses of cell-subsets following
0
I
I1 Experiments
MATEKIAIS AND METHODS Cells. Peripheral blood was drawn from three healthy volunteers (non-members of the laboratory staff) and mononuclear cells were prepared by centrifugation on lymphoprep. CD3 + and CDl4+ cells then were removed by means of immunomagnetic purging using CD3 and 0 1 4 monoclonal antibodies (moAbs) (Immunotech, France) followed by goatanti-mouse IgG-coupled dynabeads (Dynal. Norway) and separation using a cobalt-samarium magnet (MPC-I : Dynal. Norway). Liquid culture for generation o$ turgrt cells. The method has been described in detail recently (Serke rt al, 1991). Briefly, purged mononuclear cells were cultured at 2 x 10j/ml tinal culture volume in Iscove's modified Dulbecco's medium (IMDM Gibco. U.K.) supplemented with 30% (v/v) human AB-serum, 1% (w/v) bovine serum albumin (Boehringer. Germany), 1.-glutamine. and antibiotics. RhII-3 (Amgen, [J.S.A.) and rhEpo (Cilag, Germany) were included at 1 0 0 U/ml. and at 4 U/ml, respectively (Serke & Huhn, 1990). Cultures were re-fed by 100 pl of complete medium without
111
control-plasma-incubation:(a) anti-CD13positive cells: (b) anti-GlyA-positive cells. Numbers of cells per 1 . 0 ml of culture volume.
AB-serum and with rhEpo added after 3 , 6 and 9 d of culture. After 12 d of culture. cells were harvested by pelletting. cell counts were made, aliquots were taken for phenotyping. and other aliquots were taken for B19 infection, or for control plasma incubation, respectively. Source o$ BI 9-contuiningplastna. Plasma was collected from a blood donor during the viraemic stage of infection, and it contained about 10"' B19 particles per 1 ml as estimated by electron-microscopy (Schwarz rt al. 1988). Virus-inoculation. Prior to inoculation, GlyA-positive cells were determined by flow cytometry (see below) and a volume of B1 %containing plasma was added to cells resulting in a ratio of (GlyA-positive cells/B1 9-particles) about 1/100. Control cells were incubated with a n equal volume of ABO/ Kh-identical normal plasma. Cells were kept at 4°C for 4 h with occasional shaking. Thereafter, cells were pelleted. and the supernatant stored. In order to remove adsorbed virus, the cell pellet was resuspended and washed once again. The washing solution was stored, and the cells were resuspended in IMDM supplemented as described above, but omitting rhII3 . Subsequently, cells were cultured again for up to 9 6 h.
Stefan Serke et a1
8
I-=---
Number of anti-CD13-pos. cells ( x 10000)
60
~~
50 -
40
-
30 -
20 10 -
0-
I
I1 I
Number of anti-GlyA-pos. cell8 (x 10000)
b,
I
I11 Experiments
Harvest ofcells and supernatants. Cells were counted directly from the dishes, spun down, and used for further analysis after 24. 48. 72 and 9 6 h. Flow-cytometry. Analyses were performed on a Facscan flow-cytometer (Becton-Dickinson, U.S.A.) with instrument settings as described (Neubauer et al, 1989). Indirect immunofluorescence was performed employing anti-CD13moAb (MY-7: Coulter, U.S.A.). and anti-GlyA-moAb. followed by goat anti-mouse immunoglobulin G-fluorescein isothiocyanate conjugate (IgG-FITC. An der Grub, Austria). Control staining was done using MPCl 1-moAb as described previously (Neubauer et al. 1989). The cut-off from unstained versus stained cells was adjusted to 2% of cells staining positive with control moAb. 5000-10000 cells were analysed on the Facscan research software, and a FSC-live-gate was set to exclude cell debris from analysis. Data were further processed by the LYSYSII software (Becton-Dickinson, U.S.A.). lmmunocy tochemistry. Cytocentrifuge spots were prepared and an alkaline phosphatase anti-alkaline phosphatase (APAAP)-staining was employed as described previously
Pig 2. Time courses of cell-subsets following B-19-plasma-incubation: (a) anti-CD13positive cells (b) anti-GlyA-positive cells. Numbers of cells per 1 .O ml of culture volume.
(Serke et al, 1989) after fixation in ethanol (70% v/v) for 6 0 s at - 20°C. Anti-GlyA-moAb (Immunotech. France) and MPCl1 (control moAb) were employed. DNA hybridization. Cell pellets, washing solutions and supernates were collected after 4 h of inoculation with B19containing plasma (or with control plasma), and the respective samples were also collected 72 h after culture inoculation. All samples were stored at - 70°C until DNA extraction. All samples were digested by proteinase K (Boehringer. Germany) and DNA was extracted following phenol/chloroform extraction. Subsequently, DNA was precipitated in ethanol, lyophilized and dissolved in Tris-EDTA buffer. 5 pl aliquots of all samples in serial dilutions were spotted onto a nitrocellulose membrane followed by hybridization using a 3zP-labelledB19 (700 base-pairs) genome fragment (Pst-1) as described (Schwarz et al. 1988). Preliminary hybridization studies with samples from experiments not shown revealed a maximum of B19-DNA after 72 h of culture, with assays done after 24. 48, 72. 9 6 and 120 h. Hybridization studies as performed in Fig 6 were performed with samples from experiment 11.
Effective Injection of PBMC Progenitors by B 1 9
9
10
Stefan Serke et al
Fig 5. (a)Cytocentrifuge spot from cells as generated in vitro during a 12 d liquid culture from normal human peripheral blood mononuclear cells after CD3 and CD14 purging: abundant erythroblasts with mild signs of dyserythropoiesis and monomyeloid cells at different stages of differentiation: Wright-stain: original magnification (om.) x 480. (b) Cytocentrifuge spot from cells as grown in liquid culture for 72 h following control plasma incubation: APAAP-stain using anti-GlyA-monoclonal antibody: o m . x 360. (c) Cytocentrifuge spot from cells as grown in liquid culture for 96 h following incubation with B19-containing plasma: two prominent giant-cells, and in the upper quadrant, two erythroblasts with
Eflective lnfection of PBMC Progenitors by B 1 9
11
extremely uncondensed chromatin and perinuclear halo: Wright-stain: o m . = x 480. (d) Cytocentrifuge spot from cells as grown in liquid culture for 72 h following incubation with B19-containing plasma: APAAP-stain using anti-GlyA-monoclonal antibody: among cells staining for GlyA and exhibiting normal appearance of erythroblasts. in the centre two large cells staining for GlyA and exhibiting uncondensed chromatin and intranuclear inclusions: o m . = x 360.
12
Stefan Serke et al
4 h
4 h
72 h
Plasma MI1
A
B19-uninfected cells
B
B 1 9-infected cells
C
Supernatant Washing solution
E
B 19-infected cells
F
Washing solution
G
Supernatant
72 h
B 1 9-uninfected cells
I
Supernatant
J
Dilution
-
Fig 6. Dot-blot hybridization using a '2P-labelledPst-1-probe olBl9-DNA: serial dilutions ( 1 , 1 0 , 50, 100, 500.1000)of the samples as indicated.
RESULTS
In vitro generation of target cells Cultures were initiated with an average number of 4 x 10" purged mononuclear cells, and an average number of 1 3 x 10"cells was obtained after 12 d of culture. Percentages of anti-GlyA+-cells.as present after 12 d of culture among all cells obtained, showed a broad inter-individual range (experiment I: 50%: experiment 11: 28%: experiment 111: 24%). Time courses of the anti-CD13+ and of the anti-GlyA+ cells following control plasma/B19 plasma incubation Fig 1 shows the course of both these subsets in cultures set up subsequent to control plasma incubation. Within the time chosen for observation, i.e. up to 96 h following plasma incubation, there is a steady increase of both anti-CD13 and ofanti-GlyA+ cells. In contrast, Figs 2(a) and 2(b) show that, while the course of anti-CD13+ cells is almost identical in these cultures, set up after B19 plasma incubation, a marked alteration of the erythroid cells is noted. Whereas in experiments I and I1 a decrease of anti-GlyA+cells was observed, in experiment 111 the increase of anti-GlyA+cells was markedly reduced when compared to control cultures. Figs 3 and 4 show representative histograms depicting fluorescence-1 staining of control and of B19 cultures, and the relative increase of anti-CD13+ cells in B19 cultures is evident. +
Morphology and immunocytochemistry Cells obtained after 12 d in liquid culture were composed by monomyeloid cells at various stages of differentiation, and of erythroblasts, mainly of polychromatic erythroblasts. Fig 5(a) shows cells as obtained for plasma incubation, applying Wright stain. Following incubation with control plasma, morphology of cells did not change at all, and Fig 5(b)shows cells as grown for 72 h subsequent to control plasma incubation, applying APAAP stain with anti-GlyA-moAb. In contrast, morphology of cells cultured subsequent to P' 9 plasma incubation was altered markedly. Fig 5(c) shows cells as grown following B19 plasma incubation, and giant cells with uncondensed chromatin, lobulated nuclei, eosinophilic inclusions, perinuclear halo and intracytoplasmatic vacuoles are visible. Applying APAAP stain with anti-GlyA-moAb. these giant cells could be identified as erythroid. by means of staining for GlyA (Fig 5d). Molecular biology Following incubation with B19-containing plasma. B19DNA was detectable in trace amounts in the supernate at a dilution of 1: 10 (Fig 6D). and in the washing solution at a dilution of 1: 100 (Fig 6E). In contrast, analysing samples from cultures after 72 h following B19 plasma incubation, B19-DNA was detected in a dilution of 1: 10 in the cell pellet (Fig 6F), and in a dilution of 1: 1000 in the supernatant (Fig 6H). No B19-DNA was detected in the washing solution (Fig
E’ective 6 G ) , nor was it detected in any of the samples from control cultures (Fig hB. I. J). Considering the amount of B19-DNA as detected in the supernatant following the 4 h incubation with B19-containing plasma as the base value, there was about a 100-fold increase of B19-DNA after 72 h of culture, as B19-DNA was detected at a I : 1 0 0 0 dilution then, opposed to a dilution of 1 : 10, at which B19-DNA was detected in the supernatant following 4 h incubation with B l 9 plasma. DISCUSSION Studies on the effects of parvovirus B19-induced alterations of the haemopoietic system are hampered by the limited access to human bone marrow or human fetal liver, types of tissues used so far exclusively for the productive infection by B19. Making use of the in vitro liquid culture system described by us recently (Serke rt a / , 1991), we were able to generate large amounts of erythroid cells for B19 infection, initiating cultures with purged mononuclear cells from normal adult subjects. In agreement with previous studies (Mortimer et a/, 198 3 : Oxawa et a / . 1986: Srivastava c>t a / . 1990: Takahashi cr al. 1990). incubation of cells with B19-containing plasma was followed by inhibition of erythroid, i.e. anti-GlyA+ cells, only. There was no effect on myeloid cells, as identified by anti-CI) 1 3-moAb. Our findings, indicating a pronounced effect of B19 on erythroid cells only. are well in keeping with recent data of Takahashi et a/ ( 1990). They showed, using human bone marrow cells. that the susceptibility oferythroid cells to B1 9 infection increased with ongoing differentiation. The cells generated in vitro in our system represent almost synchronized cells, and at the time of inoculation chosen, i.e. 1 2 days of culture, the majority of erythroid cells are at the stage of normoblasts (see Fig 5a). To further characterize the stage of differentiation, at which susceptibility of erythroid cells to B19 infection is acquired. additional experiments, however, are mandatory. In conclusion. the method described herein should greatly facilitate further research on B19 infection, obviating the need for tissues with limited availability, such as bone marrow and fetal liver.
lnfection of PBMC Progenitors by B 1 9
13
REFERENCES Kurtzman.C.J.. Ozawa. K.. Cohen. B.. Hanson. G.. Oseas. R. & Young, N.S. (1987) Chronic bone marrow failure due to persistent B19 parvovirus infection. New England journal of Mediriw, 31 7. 28 7294. Mortimer. P.P., Humphries. R.K.. Moore. J.C.. Purcell. R.H. & Young. N.S. ( I 98 3) A human parvovirus-likevirus inhibits hematopoietic colony-formationin vitro. Nature. 302. 426-427. Neubauer. A.. Serke. S.. Siegert.W.. Kroll. W.. Musch. R. & Huhn. D. ( 1 989) A flow-cytometric assay for the determination of cell proliferation with a monoclonal antibody directed against DNAmethyltransferase. British journal oJ Haemutology. 72. 492-496. Ozawa. K., Kurtzman, C. & Young. N. (1986)Replication of the B19 parvovirus in human bone marrow cell cultures. Science. 233, 883-886. Pattison. J.R.. Jones, S.E.. Hodgson. 1.. Davis, L.R.. White. J.M.. Stroud. C.E. & Murtaza. L. (1981) Parvovirus infections and hypoplastic crises in sickle cell disease. Lancet. i, 664-665. Schwarz. T.F.. Roggendorf. M. & Deinhardt. P. (1988) Human parvovirus B 1 9: ELISA and imrnunoblot assays. lournnl of Virological Methods. 20, 155-168. Serke. S. & Huhn. D. (1990) Effects of various recombinant human haemopoietic growth factors (rhEpo. rhCM-CSF, rhC-CSF. rh11-3) on the growth of peripheral blood progenitor cells (BFU-E, CFUGM). Blut. 61, 25-29. Serke. S.. Neubauer. A. & Huhn. D. (1989) The expression of the transferrin receptor parallels the state of proliferation in HL-60 cells. British journal oJHaematology. 72, 297-299. Serke. S.. Sauberlich. S. & Huhn. D. (1991) A liquid culture method for the in vitro growth of hemopoietir progenitor cells from normal human adult peripheral blood allowing for analysis by multiparameter flow-cytometry. European Iournal oJ Haematology, 46, 8 592. Srivastava. A.. Bruno. E.. Briddell. R.. Cooper. R.. Srivastava.C.. van Besien. K. & Hoffman. R. (1990)Parvovirus B19-induced perturbation of human megakaryopoiesis in vitro. Blood. 76. 19972004. Takahashi. T.. Ozawa. K.. Takahashi. K.. Asano. S. & Takaku. F. (1990)Susceptibility of human erythropoietic cells to B19 parvovirus in vitro increases with differentiation. Blood. 75. 603-61 0.
Productive infection of in vitro generated haemopoietic progenitor cells from normal human adult peripheral blood with parvovirus B19: studies by morphology, immunocytochemistry, flow-cytometry and DNA-hybridization STEFANSERKE,TINOFRANCISSCHWARZ,* HERRADBAURMANN, ANDREASKIRSCH, BARBARA HOTTENTRAGER,* ALBRECHT V O N BRUNN,*MICHAEL ROGGENDORF,* DIETERHUHNA N D FRIEDRICH DEINHARDT* Universitiitsklinikurn Rudolf Virchow-Charlottenburg, Abteilung Innere Medizin rnit Schwerpunkt Hiimatologie-Onkologie, Freie Universitiit Berlin, and *Max von Pettenkofer lnstitut f u r Medizinische Mikrobiologie und Hygiene, Abteilung Virologie, Ludwig-Maximilians Universitiit, Munchen, Germany
Received 26 February 1991; accepted for publication 16 M a y 1991
Summary. Parvovirus B19 exerts a highly selective cytopathic effect on erythroid progenitor cells. Studies so far on the pathogenesis of B19-infection have been performed using bone marrow samples providing large amounts of erythroid progenitor cells. Extensive study, however, has been hampered by the limited access to bone marrow samples. We have designed a liquid culture method allowing the generation of large numbers of erythroid progenitor cells, initiating cultures with CD3- and CD14-poor peripheral blood mononuclear cells. Following a 12 d preincubation in liquid cultures containing recombinant human interleukin 3 (rhII3 ) and recombinant human erythropoietin (rhEpo), cells harvested from the liquid cultures were exposed to B19-
containing plasma, followed by a further cultivation in liquid culture for up to 9 6 h. Cells expressing the CD13 and the glycophorin A (GlyA) antigens, respectively. were monitored sequentially by flow-cytometry, demonstrating a selective inhibition of GlyA-positive cells following B19-inoculation. Typical morphological changes were observed on cytocentrifuge-spots. and typical giant-cells were identified as staining for GlyA. Productive infection by B19 was demonstrable, as B19-DNA increased by about x 100 after 72 h of culture. The liquid culture method generating erythroid target cells for effective infection by B19 virus promises to be a useful and easily accessible tool for further research on B19 infection of haemopoietic cells.
With regard to the haemopoietic system, human parvovirus B19 exerts cytopathic effects almost exclusively upon cells of the erythroid lineage, leading to transient aplastic episodes in subjects with increased erythropoiesis (Pattison et a/. 1981). and to persistent long-standing anaemia in immunocompromised patients (Kurtman et al, 1987). In vitro studies have shown that B19-infection leads to inhibition of both erythroid burst-forming units (BFU-E) and erythroid colonyforming units (CFU-E). with no effect on granulocytic or monocytic precursors (Mortimer et a/, 1983: Ozawa et al, 1986; Takahashi et a / . 1990). Only recently, however,
Srivastava et al (1990) have shown that megakaryocytic colony formation was also inhibited by B19 infection, but no viral DNA replication was detected in bone marrow cells highly enriched for megakaryocytic progenitor cells. In vitro studies on B19 infection, so far. have relied on infection of fetal liver cells, or of bone marrow cells, both types of tissues not available o n a routine base. Moreover, no permanent cell line has been identified as yet, capable of productive infection by B19 (Mortimer et a/, 1983). In order to facilitate further the in vitro research on B19caused alterations of haemopoiesis, we have developed a liquid culture method allowing the generation of large numbers of erythroid progenitors capable of productive infection by B19, initiating cultures with peripheral blood mononuclear cells from normal human adult donors (Serke e t a / , 1991).
Correspondence: Dr med. Stefan Serke. UKRV-Charlottenburg, Hamatologie. Spandauer Damm 1 30. 1000 Berlin 19. Germany.
6
Eflective lnfection of PBMC Progenitors by BI 9
7
Number of anti-CDl3-~os. cells ( x 10000)
60
.
.
-.
50 40
30
20 10
0
111
I Number of anti-GlvA-Dos. cells ( x 10000) 60
i-1
........ ........ ........ ......... ......... ......... ......... ......... ......... ......... ......... ......... .........
50
40
30
20 10
Fig 1. Time courses of cell-subsets following
0
I
I1 Experiments
MATEKIAIS AND METHODS Cells. Peripheral blood was drawn from three healthy volunteers (non-members of the laboratory staff) and mononuclear cells were prepared by centrifugation on lymphoprep. CD3 + and CDl4+ cells then were removed by means of immunomagnetic purging using CD3 and 0 1 4 monoclonal antibodies (moAbs) (Immunotech, France) followed by goatanti-mouse IgG-coupled dynabeads (Dynal. Norway) and separation using a cobalt-samarium magnet (MPC-I : Dynal. Norway). Liquid culture for generation o$ turgrt cells. The method has been described in detail recently (Serke rt al, 1991). Briefly, purged mononuclear cells were cultured at 2 x 10j/ml tinal culture volume in Iscove's modified Dulbecco's medium (IMDM Gibco. U.K.) supplemented with 30% (v/v) human AB-serum, 1% (w/v) bovine serum albumin (Boehringer. Germany), 1.-glutamine. and antibiotics. RhII-3 (Amgen, [J.S.A.) and rhEpo (Cilag, Germany) were included at 1 0 0 U/ml. and at 4 U/ml, respectively (Serke & Huhn, 1990). Cultures were re-fed by 100 pl of complete medium without
111
control-plasma-incubation:(a) anti-CD13positive cells: (b) anti-GlyA-positive cells. Numbers of cells per 1 . 0 ml of culture volume.
AB-serum and with rhEpo added after 3 , 6 and 9 d of culture. After 12 d of culture. cells were harvested by pelletting. cell counts were made, aliquots were taken for phenotyping. and other aliquots were taken for B19 infection, or for control plasma incubation, respectively. Source o$ BI 9-contuiningplastna. Plasma was collected from a blood donor during the viraemic stage of infection, and it contained about 10"' B19 particles per 1 ml as estimated by electron-microscopy (Schwarz rt al. 1988). Virus-inoculation. Prior to inoculation, GlyA-positive cells were determined by flow cytometry (see below) and a volume of B1 %containing plasma was added to cells resulting in a ratio of (GlyA-positive cells/B1 9-particles) about 1/100. Control cells were incubated with a n equal volume of ABO/ Kh-identical normal plasma. Cells were kept at 4°C for 4 h with occasional shaking. Thereafter, cells were pelleted. and the supernatant stored. In order to remove adsorbed virus, the cell pellet was resuspended and washed once again. The washing solution was stored, and the cells were resuspended in IMDM supplemented as described above, but omitting rhII3 . Subsequently, cells were cultured again for up to 9 6 h.
Stefan Serke et a1
8
I-=---
Number of anti-CD13-pos. cells ( x 10000)
60
~~
50 -
40
-
30 -
20 10 -
0-
I
I1 I
Number of anti-GlyA-pos. cell8 (x 10000)
b,
I
I11 Experiments
Harvest ofcells and supernatants. Cells were counted directly from the dishes, spun down, and used for further analysis after 24. 48. 72 and 9 6 h. Flow-cytometry. Analyses were performed on a Facscan flow-cytometer (Becton-Dickinson, U.S.A.) with instrument settings as described (Neubauer et al, 1989). Indirect immunofluorescence was performed employing anti-CD13moAb (MY-7: Coulter, U.S.A.). and anti-GlyA-moAb. followed by goat anti-mouse immunoglobulin G-fluorescein isothiocyanate conjugate (IgG-FITC. An der Grub, Austria). Control staining was done using MPCl 1-moAb as described previously (Neubauer et al. 1989). The cut-off from unstained versus stained cells was adjusted to 2% of cells staining positive with control moAb. 5000-10000 cells were analysed on the Facscan research software, and a FSC-live-gate was set to exclude cell debris from analysis. Data were further processed by the LYSYSII software (Becton-Dickinson, U.S.A.). lmmunocy tochemistry. Cytocentrifuge spots were prepared and an alkaline phosphatase anti-alkaline phosphatase (APAAP)-staining was employed as described previously
Pig 2. Time courses of cell-subsets following B-19-plasma-incubation: (a) anti-CD13positive cells (b) anti-GlyA-positive cells. Numbers of cells per 1 .O ml of culture volume.
(Serke et al, 1989) after fixation in ethanol (70% v/v) for 6 0 s at - 20°C. Anti-GlyA-moAb (Immunotech. France) and MPCl1 (control moAb) were employed. DNA hybridization. Cell pellets, washing solutions and supernates were collected after 4 h of inoculation with B19containing plasma (or with control plasma), and the respective samples were also collected 72 h after culture inoculation. All samples were stored at - 70°C until DNA extraction. All samples were digested by proteinase K (Boehringer. Germany) and DNA was extracted following phenol/chloroform extraction. Subsequently, DNA was precipitated in ethanol, lyophilized and dissolved in Tris-EDTA buffer. 5 pl aliquots of all samples in serial dilutions were spotted onto a nitrocellulose membrane followed by hybridization using a 3zP-labelledB19 (700 base-pairs) genome fragment (Pst-1) as described (Schwarz et al. 1988). Preliminary hybridization studies with samples from experiments not shown revealed a maximum of B19-DNA after 72 h of culture, with assays done after 24. 48, 72. 9 6 and 120 h. Hybridization studies as performed in Fig 6 were performed with samples from experiment 11.
Effective Injection of PBMC Progenitors by B 1 9
9
10
Stefan Serke et al
Fig 5. (a)Cytocentrifuge spot from cells as generated in vitro during a 12 d liquid culture from normal human peripheral blood mononuclear cells after CD3 and CD14 purging: abundant erythroblasts with mild signs of dyserythropoiesis and monomyeloid cells at different stages of differentiation: Wright-stain: original magnification (om.) x 480. (b) Cytocentrifuge spot from cells as grown in liquid culture for 72 h following control plasma incubation: APAAP-stain using anti-GlyA-monoclonal antibody: o m . x 360. (c) Cytocentrifuge spot from cells as grown in liquid culture for 96 h following incubation with B19-containing plasma: two prominent giant-cells, and in the upper quadrant, two erythroblasts with
Eflective lnfection of PBMC Progenitors by B 1 9
11
extremely uncondensed chromatin and perinuclear halo: Wright-stain: o m . = x 480. (d) Cytocentrifuge spot from cells as grown in liquid culture for 72 h following incubation with B19-containing plasma: APAAP-stain using anti-GlyA-monoclonal antibody: among cells staining for GlyA and exhibiting normal appearance of erythroblasts. in the centre two large cells staining for GlyA and exhibiting uncondensed chromatin and intranuclear inclusions: o m . = x 360.
12
Stefan Serke et al
4 h
4 h
72 h
Plasma MI1
A
B19-uninfected cells
B
B 1 9-infected cells
C
Supernatant Washing solution
E
B 19-infected cells
F
Washing solution
G
Supernatant
72 h
B 1 9-uninfected cells
I
Supernatant
J
Dilution
-
Fig 6. Dot-blot hybridization using a '2P-labelledPst-1-probe olBl9-DNA: serial dilutions ( 1 , 1 0 , 50, 100, 500.1000)of the samples as indicated.
RESULTS
In vitro generation of target cells Cultures were initiated with an average number of 4 x 10" purged mononuclear cells, and an average number of 1 3 x 10"cells was obtained after 12 d of culture. Percentages of anti-GlyA+-cells.as present after 12 d of culture among all cells obtained, showed a broad inter-individual range (experiment I: 50%: experiment 11: 28%: experiment 111: 24%). Time courses of the anti-CD13+ and of the anti-GlyA+ cells following control plasma/B19 plasma incubation Fig 1 shows the course of both these subsets in cultures set up subsequent to control plasma incubation. Within the time chosen for observation, i.e. up to 96 h following plasma incubation, there is a steady increase of both anti-CD13 and ofanti-GlyA+ cells. In contrast, Figs 2(a) and 2(b) show that, while the course of anti-CD13+ cells is almost identical in these cultures, set up after B19 plasma incubation, a marked alteration of the erythroid cells is noted. Whereas in experiments I and I1 a decrease of anti-GlyA+cells was observed, in experiment 111 the increase of anti-GlyA+cells was markedly reduced when compared to control cultures. Figs 3 and 4 show representative histograms depicting fluorescence-1 staining of control and of B19 cultures, and the relative increase of anti-CD13+ cells in B19 cultures is evident. +
Morphology and immunocytochemistry Cells obtained after 12 d in liquid culture were composed by monomyeloid cells at various stages of differentiation, and of erythroblasts, mainly of polychromatic erythroblasts. Fig 5(a) shows cells as obtained for plasma incubation, applying Wright stain. Following incubation with control plasma, morphology of cells did not change at all, and Fig 5(b)shows cells as grown for 72 h subsequent to control plasma incubation, applying APAAP stain with anti-GlyA-moAb. In contrast, morphology of cells cultured subsequent to P' 9 plasma incubation was altered markedly. Fig 5(c) shows cells as grown following B19 plasma incubation, and giant cells with uncondensed chromatin, lobulated nuclei, eosinophilic inclusions, perinuclear halo and intracytoplasmatic vacuoles are visible. Applying APAAP stain with anti-GlyA-moAb. these giant cells could be identified as erythroid. by means of staining for GlyA (Fig 5d). Molecular biology Following incubation with B19-containing plasma. B19DNA was detectable in trace amounts in the supernate at a dilution of 1: 10 (Fig 6D). and in the washing solution at a dilution of 1: 100 (Fig 6E). In contrast, analysing samples from cultures after 72 h following B19 plasma incubation, B19-DNA was detected in a dilution of 1: 10 in the cell pellet (Fig 6F), and in a dilution of 1: 1000 in the supernatant (Fig 6H). No B19-DNA was detected in the washing solution (Fig
E’ective 6 G ) , nor was it detected in any of the samples from control cultures (Fig hB. I. J). Considering the amount of B19-DNA as detected in the supernatant following the 4 h incubation with B19-containing plasma as the base value, there was about a 100-fold increase of B19-DNA after 72 h of culture, as B19-DNA was detected at a I : 1 0 0 0 dilution then, opposed to a dilution of 1 : 10, at which B19-DNA was detected in the supernatant following 4 h incubation with B l 9 plasma. DISCUSSION Studies on the effects of parvovirus B19-induced alterations of the haemopoietic system are hampered by the limited access to human bone marrow or human fetal liver, types of tissues used so far exclusively for the productive infection by B19. Making use of the in vitro liquid culture system described by us recently (Serke rt a / , 1991), we were able to generate large amounts of erythroid cells for B19 infection, initiating cultures with purged mononuclear cells from normal adult subjects. In agreement with previous studies (Mortimer et a/, 198 3 : Oxawa et a / . 1986: Srivastava c>t a / . 1990: Takahashi cr al. 1990). incubation of cells with B19-containing plasma was followed by inhibition of erythroid, i.e. anti-GlyA+ cells, only. There was no effect on myeloid cells, as identified by anti-CI) 1 3-moAb. Our findings, indicating a pronounced effect of B19 on erythroid cells only. are well in keeping with recent data of Takahashi et a/ ( 1990). They showed, using human bone marrow cells. that the susceptibility oferythroid cells to B1 9 infection increased with ongoing differentiation. The cells generated in vitro in our system represent almost synchronized cells, and at the time of inoculation chosen, i.e. 1 2 days of culture, the majority of erythroid cells are at the stage of normoblasts (see Fig 5a). To further characterize the stage of differentiation, at which susceptibility of erythroid cells to B19 infection is acquired. additional experiments, however, are mandatory. In conclusion. the method described herein should greatly facilitate further research on B19 infection, obviating the need for tissues with limited availability, such as bone marrow and fetal liver.
lnfection of PBMC Progenitors by B 1 9
13
REFERENCES Kurtzman.C.J.. Ozawa. K.. Cohen. B.. Hanson. G.. Oseas. R. & Young, N.S. (1987) Chronic bone marrow failure due to persistent B19 parvovirus infection. New England journal of Mediriw, 31 7. 28 7294. Mortimer. P.P., Humphries. R.K.. Moore. J.C.. Purcell. R.H. & Young. N.S. ( I 98 3) A human parvovirus-likevirus inhibits hematopoietic colony-formationin vitro. Nature. 302. 426-427. Neubauer. A.. Serke. S.. Siegert.W.. Kroll. W.. Musch. R. & Huhn. D. ( 1 989) A flow-cytometric assay for the determination of cell proliferation with a monoclonal antibody directed against DNAmethyltransferase. British journal oJ Haemutology. 72. 492-496. Ozawa. K., Kurtzman, C. & Young. N. (1986)Replication of the B19 parvovirus in human bone marrow cell cultures. Science. 233, 883-886. Pattison. J.R.. Jones, S.E.. Hodgson. 1.. Davis, L.R.. White. J.M.. Stroud. C.E. & Murtaza. L. (1981) Parvovirus infections and hypoplastic crises in sickle cell disease. Lancet. i, 664-665. Schwarz. T.F.. Roggendorf. M. & Deinhardt. P. (1988) Human parvovirus B 1 9: ELISA and imrnunoblot assays. lournnl of Virological Methods. 20, 155-168. Serke. S. & Huhn. D. (1990) Effects of various recombinant human haemopoietic growth factors (rhEpo. rhCM-CSF, rhC-CSF. rh11-3) on the growth of peripheral blood progenitor cells (BFU-E, CFUGM). Blut. 61, 25-29. Serke. S.. Neubauer. A. & Huhn. D. (1989) The expression of the transferrin receptor parallels the state of proliferation in HL-60 cells. British journal oJHaematology. 72, 297-299. Serke. S.. Sauberlich. S. & Huhn. D. (1991) A liquid culture method for the in vitro growth of hemopoietir progenitor cells from normal human adult peripheral blood allowing for analysis by multiparameter flow-cytometry. European Iournal oJ Haematology, 46, 8 592. Srivastava. A.. Bruno. E.. Briddell. R.. Cooper. R.. Srivastava.C.. van Besien. K. & Hoffman. R. (1990)Parvovirus B19-induced perturbation of human megakaryopoiesis in vitro. Blood. 76. 19972004. Takahashi. T.. Ozawa. K.. Takahashi. K.. Asano. S. & Takaku. F. (1990)Susceptibility of human erythropoietic cells to B19 parvovirus in vitro increases with differentiation. Blood. 75. 603-61 0.
