Long-term culture of primary B cells and in vitro expression of an exogenous gene

Immunology Letters. 47 (1995) 193-197 0165-2478/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved IMLET 2429

Long-term culture of primary B cells and in vitro expression of an exogenous gene Abhay

Kumar,

Emily

C. Guido,

Ru-Shya

Liu and Mohammad

S. Saedi

l

Molecular Biology Group, Hybritech Incorporated, P.O. Box 269006, San Diego, CA 92196. USA (Received

21 February

1995; revised 6 June 1995; accepted

Key words: B cell culture; Retroviral

infection; Human growth hormone;

1. Summary Efficient introduction and expression of exogenous genes into primary B cells is very important to study B-cell biology and is essential for gene therapy. These efforts have often been impeded by the lack of availability of a simple culture condition for growth and proliferation of primary B cells as well as the lack of vehicles for efficient introduction of genes of interest. In this communication, we have developed a culture condition that supports the growth of primary B cells from /?-gal-immunized mouse spleen for 30 days without the aid of feeder layers. During this period, B cells secreted polyclonal antibodies into the medium. To study expression of an exogenous reporter gene, human growth hormone (hGH) was introduced into cultured B cells using retroviral vectors. hGH was expressed up to 21 days in the absence of drug selection and the infected cells continued to secrete immunoglobulins into the medium.

2. Introduction Gene transfer technology has been employed to study the cell and molecular biology of hematopoietic cells. These studies require culture conditions to support the growth and proliferation of cells as well as vehicles for efficient introduction and long-term expression of exogenous genes. Gene transfer studies on B cells have

’ Corresponding author: Mohammad S. Saedi, Hybritech Incorporated, P.O. Box 269006, San Diego, CA 92196, USA. Tel.: (619) 535-8706; Fax: (619) 457-5308. SSDI 0165.2478(95)00092-5

12 June 1995) Gene delivery

been limited primarily to transgenic animal models [l-3] and DNA transfections into established B cell lines [4]. In the latter studies, drug selection has been used to enrich the number of transfected cells and to maintain the expression of the exogenous genes for more than 4 weeks [4]. The long-term expression of exogenous genes in primary B cells has not been well described. Buschle et al. [5] and Overell et al. [6] successfully transfected chloramphenicol acetyltransferase (CAT) into primary B cells [5,6]. However, CAT expression was transient and lasted only for 48-72 h. Kuo et al. [7] used retroviral vectors to transduce adenosine deaminase (ADA) gene into primary B cells. They monitored the expression of ADA gene by reconstituting SCID mice with infected cells up to 3 months, thus eliminating the need for drug selection but still relying on an in vivo system. The study presented here was conducted with two objectives: (1) to culture primary B cells in vitro for an extended period of time and (2) to examine the suitability of the culture conditions for studying the expression of exogenous genes in cultured primary B ceils. In this report, we describe a simple in vitro condition for culturing primary murine B cells without the need for incubation with feeder layers or addition of exogenous cytokines. Cultured B cells secreted polyclonal antibodies for up to 1 month. Human growth hormone (hGH) was transduced into these B cells using a retroviral delivery system. We chose hGH as a test gene because it is secreted into the medium enabling us to monitor its expression for an extended period of time without affecting other cell parameters. In addition, hGH is stable at -2O”C, making it possible to collect and assay samples at separate times. The culture conditions described here allowed us to study the hGH expression for three weeks without drug selection. The infected cells continued to secrete immunoglobulins into the medium. 193

3. Materials and Methods 3.1. Immunization primary B cells

of mice and isolation and growth of

Balb/c mice (Charles River Laboratory, Raleigh, NC) were immunized i.p. with 50 pg of Escherichia coli p galactosidase ( P-gal, Sigma, Saint Louis, MO) in 10% Alum (Sigma) on days 1 and 14. Mice were finally boosted 3 days prior to each experiment with 50 pg of P-gal in PBS (pH 7.4). B cells were prepared essentially by the method described by Kruisbeek [8] and Hathcock [9]. Briefly, single-cell suspensions were prepared from the spleens of P-gal-immunized mice by teasing with a hypodermic needle. Red blood cells (RBCS) were lysed with 0.84% ammonium chloride and the dead cells were removed by passing the cells over fetal bovine serum gradient. T cells were removed from this population using the complement fixation method [9] and B cells were recovered by centrifugation and analysed by flow cytometry. This analysis showed that approximately 90% of cells expressed IgG and/or IgM on their surface. Typically, a mouse spleen yielded 4-5 x lo7 B cells. B cells were cultured in B-cell medium that is composed of RPMI, 20% fetal calf serum, 2 mM L-glutamine, 15 mM HEPES, 12.5 pg/ml lipopolysaccharides (LPS, E. coli serotype 055 : B5, Sigma, St. Louis, MO), 20 pg/ml dextran sulfate (DS, Pharmacia, Uppsala, Sweden) and 20% spleen condition medium (SCM). SCM was prepared by culturing the spleen cells of Balb/c mice at IO6 cell/ml in HH4 medium (JRH Biosciences, Lenexa, KS) supplemented with 10% horse serum, harvesting and filtering (0.45 pm) the conditioned medium 48 h later. The B-cell medium was used throughout the course of study unless otherwise mentioned. Cell viability was assessed by Trypan Blue exclusion. Proliferation of B cells was monitored by [ 3H] thymidine incorporation as described

’

lkb

’

pLNCX-hGH Fig. 1. Schematicrepresentationof the retroviral vector pLNCX-hGH. Arrows indicate the transcription initiation sites.

mM) and HEPES (15 mM), and transfected with pLNCX-hGH using the lipofection method (Gibco-BRL, Gaithersburg, MD). The transfectants were selected in G418 (Gibco-BRL, 0.6 mg/ml) for lo-15 days. The viral titer from individual clones was assessed by transducing NIH3T3 cells and a high-titer clone (+2hGH) was selected. Recombinant virus (v-hGH) was obtained by allowing 4?-2 hGH to reach near confluence, replacing the spent medium with fresh medium, harvesting and filtering (0.45 pm) the conditioned medium 24 h later. The virus was used immediately or stored at - 80°C. Viral infections were performed by culturing the B cells (4 X 106) in B-cell medium and infecting them 2 days later with v-hGH (4 X 10’ c.f.u.) in the presence of polybrene (4 pg/ml, Sigma, St. Louis, MO). Cells were incubated overnight at 37”C, then spun and resuspended in fresh B-cell medium, and analyzed at various times for growth, hGH production and immunoglobulin secretion. hGH in the culture supematant was measured in duplicate using Tandem R hGH assay@ kit (Hybritech, San Diego, CA) according to the manufacturer’s directions.

ml.

3.4. Measurement

3.2. Vector construction

Total IgG and IgM were measured by ELISA. Ninety-six well plates (Becton Dickinson Labware, NJ) were coated overnight at 37°C with 50 ,ul of goat anti mouse-IgG or IgM monoclonal antibodies (5 pg/ml, Fisher Scientific, Pittsburgh, Pa>. The wells were washed with PBS and incubated for 1 h with 50 ~1 of culture supematants. The wells were washed again with PBS and incubated for 1 h with 50 ~1 of goat anti mouse-IgG or anti-IgM monoclonal antibodies coupled with horseradish peroxide (0.4 pg/ml, Fisher Scientific). The wells were washed with PBS, incubated with ophenylenediamine dihydrochloride (OPD, Sigma, St. Louis, MO) for 5 min, and the calorimetric reaction was measured at A,, with an ELISA reader (Biotek Instru-

pLNCX-hGH (Fig. 1) was constructed by inserting the entire hGH coding sequence of 1.6 kb from p0GH (Nichols Institute, San Juan Capistrano, CA) into the HindIII/Clal sites of pLNCX [ll]. The orientation of hGH was confirmed by DNA sequencing. In this construct, the hGH expression is driven by the cytomegalovirus (CMV) promoter. 3.3. Cell lines, virus production and infection of B cells Ecotropic virus packaging cell line e-2 was grown in RPM1 plus 10% FCS supplemented with L-glutamine (2 194

of immunoglobulins

ments, model EL310, VT). All samples were assayed in duplicate.

4. Results and Discussion To establish a culture medium that supported the growth and proliferation of B cells, B cells were cultured in medium containing HH4 or RPM1 and different combinations of horse serum, fetal bovine serum, erythrogen, IL-6, SCM, LPS and DS. The medium formulation that best supported the sustained growth and proliferation of B cells consisted of RPMI, 20% FCS, 2 mM L-glutamine, 15 mM HEPES, 12.5 pg/ml LPS, 20 pg/ml DS and 20% SCM (B-cell medium). Also, in contrast to the other medium that we had tested, the B-cell medium did not support the growth of adherent cells. We found LPS and DS to be necessary for B-cell growth and proliferation probably because they are non-specific B-cell mitogens driving B cells towards division rather than differentiation [ 121. We also found SCM to be necessary, probably because it provided all the essential growth factors and cytokines to support B-cell growth in culture. To monitor the growth of B cells, B cells were cultured in B-cell medium and at various times the number of viable cells were counted by Trypan Blue exclusion (Fig. 2). After an initial loss, B cells from immunized mice multiplied rapidly. The number of viable cells peaked on day 4 and steadily declined until day 9. After day 9, the number of viable cells increased again until day 18 and stayed relatively unchanged until day 23. At day 23, the number of viable cells declined until the end of the study. Interestingly, B cells from the spleens of non-immunized mice did not survive for more than 3 days in B-cell medium (data not shown). To monitor the proliferation of B cells, B cells were cultured in B-cell medium and at various times assessed for [ 3Hlthymidine incorporation. Results showed that cells started to proliferate immediately and peaked after 5 days in culture (Fig. 2, inset). This peak coincided with the peak observed in the growth study. B cells did not proliferate in the absence of LPS and DS, indicating that these two components were essential for the proliferation of B cells (Fig. 2, inset). Growth and proliferation patterns of cultured B cells similar to the ones described above were observed in at least three independent experiments. To study whether these growth conditions would support the expression of exogenous genes, we used various transfection techniques to deliver human growth hormone (hGH) gene into cultured B cells. Lipofection and electroporation of hGH DNA into these B cells were not effective in our hands (data not shown).

0

10

20 Days inCulture

30

40

Fig. 2. Growth of primary B cells in B-cell medium. B cells were prepared by T-cell depletion and were brought into culture immediately. At indicated time points, viable B cells from duplicate wells were counted using Trypan Blue exclusion, and medium was replaced with fresh medium. 13, uninfected B cells; +, B cells infected with v-hGH. Inset: proliferation of B cells from the spleens of immunized mice measured by [‘Hlthymidine uptake. Cells were cultured in B-cell medium (0) or B-cell medium lacking LPS and DS ( l). Medium was supplemented with [‘Hlthymidine (1 &i/well) during the final 24 h before harvesting. Cells were harvested with PHD cell harvester (Cambridge Technology, Cambridge, MA) and counted. Each data point is an average of duplicate wells. This is a representative of three independently conducted experiments.

Overell et al. [6] also reported limited success with lipofection and electroporation of plasmid DNA in primary B cells, Therefore, we attempted retroviral infection that was successfully used to efficiently transduce gene(s) into other cells [ 13,141. hGH was cloned into the retroviral vector pLNCX resulting in the recombinant vector pLNCX-hGH (Fig. 1). hGH gene was cloned under the CMV promoter which was shown to be very active promoter in lymphoid cells [151. Ecotropic virus (v-hGH) was prepared by transfecting pLNCX-hGH into I)-2 cells. v-hGH was titered on NIH3T3 cells and the titer was estimated to be in excess of 2 x lo5 colony forming units/ml. After 48 h in culture, B cells (4 X 106) were infected with 4 X 10’ c.f.u. of v-hGH in presence of polybrene. L3H]thymidine uptake results indicated that at this stage B cells were actively proliferating (Fig. 2, inset), a requirement for efficient retroviral integration and expression [16]. Retroviral infection of target cells can be accomplished by co-cultivation with packaging cell line [20]. However, to avoid potential problem of hGH secretion by I,!J-2hGH cells producing artifactually high readings, primary B cells 195

*/

0

5

10 Days

15

20

25

in Culture

Fig. 3. Time course of retrovirus-mediated hGH expression in primary B cells. Cells were infected with v-hGH. At indicated time points, expression of hGH was monitored in supematants using Tandem R hGH assay@ kit and medium was replaced with fresh medium. Values are expressed as ng of hGH/ml of supematant/106 ceils. 0, uninfected B cells; +, B cells grown for 2 days and infected with v-hGH. Each data point is an average of duplicate samples. This is a representative of three independently conducted experiments.

were infected with virus supematant rather than by cocultivation with $-2 hGH cells. To monitor the growth of v-hGH infected B cells, B cells were cultured in B-cell medium and the number of viable cells were counted by Trypan Blue exclusion (Fig. 2). The number of infected cells peaked at day 4 and steadily declined until day 12. After day 12, the number of viable cells increased again until day 16. After day 16, the number of viable cells started to decline until the end of study. Although differences were observed for individual time points, the overall growth pattern observed for the infected B cells was similar to that of uninfected B cells in three independently conducted experiments. Initially, both infected and uninfected cell populations multiplied rapidly, followed by a drop in viable cell counts. Viable cell numbers increased in both polpulations again by day 16 (for infected cells) and day 18 (for uninfected cells) followed by a steady drop until the end of the study. Supematants from infected cells were assayed for hGH expression (Fig. 3). Supematants collected at 24 h post-infection had high levels of hGH (23.06 ng/ml/106 cells). Thereafter, hGH expression declined and stabilized. Since the medium was changed periodically, this indicates that hGH was continuously being secreted for 21 days post-infection. Similar pattern for hGH expression was observed in three independent experiments. Archer et al. [17] have also reported a 196

transiently high expression of hGH in goat’s milk after infection of mammary secretory epithelial cells with hGH carrying replication defective retrovirus. The reason for the transiently high expression of hGH is not clear but it may be that within a few days after infection a percentage of integrated MoMLV LTRs were repressed, which accounted for the decrease in hGH production. Alternatively, LTR enhancer elements may be suppressed at several levels with inactivation occurring after the LTR has started to function [ 181. These findings indicate that the conditions described above provide a good system to study the effect of exogenous gene expression in primary B cells in vitro, without the need for drug selection. In the studies published thus far, the exogenous genes in primary B cells were expressed for only a short period of time, 24-72 h [5,6]. The expression of exogenous genes for longer period of time was achieved either by developing immortal cell lines using EBV [4,19] or by using established cell lines [20]. To monitor the expression of immunoglobulins by the infected and uninfected cells, cell supematants were periodically collected and the medium was replaced with fresh medium. The levels of IgM and IgG were measured by ELISA (Fig. 4). IgM levels in the supernatants of uninfected and v-hGH-infected cells were not 250

I

II

I fection

Y

0

io

20

30

40

DaysinCultm

Fig. 4. Antibody production by primary B cells in culture. Supcrnatants were collected at indicated time points and the medium was replaced with fresh medium. IgG ( 0, uninfected B cells; l , B cells infected with v-hGH) and IgM (0, uninfected B cells; A, B cells infected with v-hGH) production in the collected supematants were assayed using ELISA. Concentration (rig/ml) of samples was determined by comparing the absorbance to standards of known concentrations in the same assay. Each data point is an average of duplicate wells. This is a representative of three independently conducted experiments.

found to be significantly different. IgG levels of infected cells were 50% of that of uninfected cells for 2 days following infection. However, the IgG levels of infected cells increased by day 4 post-infection and found not to be significantly different from that of uninfected cells throughout the rest of the experiment. The reason for this initial difference is not known. IgG levels varied from 100 to 200 ng/ml of supematant during the course of study. IgM levels were slightly higher than IgG. Antigen-specific ( p-gal) antibody production was also measured. Again, no significant difference was observed between infected and uninfected cells (data not shown). In summary, we have described culture conditions to support primary B cells for 30 days without the aid of feeder layer or cytokines. We also show that we could use retroviral vectors to deliver genes into these B cells. These growth conditions support the expression of an exogenous gene, hGH, without drug selection.

Acknowledgements We thank Dr. K. Kuus-Reichel for valuable suggestions and critical reading of the manuscript. Thanks are also due to C. Fenstermaker and D. Ta for excellent technical assistance.

[2] Goodnow. CC. (1992) Ann. Rev. lmmunol. 10. 489. [3] Taveme, I. (1993) Int. J. Exp. Path. 74. 525. [4] Maly, F.E., Schuerer-Maly, C.C., Quilliam. L., Cochrane, C.G., Newburger, P.E., Cumutte, J.T., Gifford, M. and Dinauer, M.C. (1993) J. Exp. Med. 178, 2047. [5] Buschle, M., Brenner, M.K., Chen, I.S.Y.. Drexler, H.G.. Gignac, S.M. and Rooney, C.M. (1990) J. lmmunol. Methods 133, 77. [6] Overell, R.W., Weisser, K.E., Hess, B.W.. Ziegler, S.F.. Pleiman, CM., Maliszewski, C. and Grabstein. K.H. (1991) J. Immunol. Methods 141, 53. [7] Kuo, M.-L., Sutkowski, N., Ron Y. and Dougherty, J.P. (1993) Blood 82, 845. [8] Kmisbeek, A.M. (1991) in: Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies. E.M. Shevach, and W. Strober, eds.,) pp. 3.1.1-3.1.5. John Wiley, New York. [9] Hathcock, K.S. (1991) in: Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies. E.M. Shevach, and W. Strober, eds.,) pp. 3.4.1-3.4.5. John Wrley. New York. [IO] Mond, J.J. and Brunswick, M. (19911 m: Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies. E.M. Shevach, and W. Stroker, eds.,) pp. 3.10.1-3.10.8. John Wiley, New York. [I11 [I21 [I31 [I41 [15] [I61 [17] [I81

References [l] Palmiter, 465.

[I91 R.D. and Brinster,

R.L. (1986) AM. Rev. Genet. 20,

[20]

Miller. A.D. and Rosman, G.J. (1989) Biotechniques 7, 980. Wetzel, G.D. and Kettman, J.R. (1981) J. lmmunol. 126, 723. Friedmann, T. (1989) Science 244, 1275. Miller, A.D. (1992) Nature 357, 455. Saedi, M.S., Rao, S., Ill, C.R. and Guido, E.C. (1994) Immunol. Lett. 40, 79. Miller, D.G., Adam, M.A. and Miller. A.D. (1990) Mol. Cell. Biol. 10, 4239. Archer, J.S., Kennan, W.S., Gould, M.N. and Bremel, R.D. (1994) Proc. Natl. Acad. Sci. USA 91, 6840. Hoeben, R., Migchielsen, A., van der Jagt. R.. van Ormondt, H. and van der Eb, A. (1991) J. Viral. 65. 904. Thrasher. A.. Chetty. M.. Casimir, C. and Segal. A.W. (1992) Blood 80, 1125. Krauss, J.C., Bond, L.M., Todd Ill, R.F. and Wilson, J.M. (1991) Hum. Gen. Ther. 2. 221.

197