Leukemia (2007) 21, 1977–1983 & 2007 Nature Publishing Group All rights reserved 0887-6924/07 $30.00 www.nature.com/leu ORIGINAL ARTICLE

Efficient nucleofection of primary human B cells and B-CLL cells induces apoptosis, which depends on the microenvironment and on the structure of transfected nucleic acids

M Seiffert1, S Stilgenbauer2,HDo¨hner2 and P Lichter1

1Department for Molecular Genetics, German Cancer Research Center, Heidelberg, Germany and 2Innere Medizin III, Universita¨t Ulm, Ulm, Germany

Accumulation of neoplastic cells in B-cell chronic lymphocytic under culture conditions that support the growth of human leukemia (B-CLL) is thought to be due to intrinsic defects in the Bcelllines.12 On the other hand, B-CLL cells can be rescued from apoptotic machinery of the leukemic cells or to an altered, apoptosis ex vivo, when cocultured with BM-derived stromal survival-stimulating microenvironment in vivo. Despite their 13 long survival in vivo, B-CLL cells undergo rapid spontaneous cells. B-CLL cell survival in these cocultures is associated with apoptosis ex vivo. To maintain survival in vitro, we established an active phosphatidylinositol-3 kinase/Akt signaling pathway a coculture system using the human bone marrow-derived and with increased expression of anti-apoptotic proteins like stromal cell line HS-5. The microenvironment in these cocul- Bcl-2.14–16 tures lead to B-CLL cell survival for at least several months and Probably owing to the quiescent nature of circulating B-CLL therefore provided a tool for valid in vitro analysis, mimicking cells, B-CLL tumor cell lines have never been obtained unless the in vivo situation. Although primary B lymphocytes are 17,18 notoriously resistant to most gene transfer techniques, we infected in vitro or in vivo by Epstein–Barr virus. On the achieved high transfection efficiency and cell viability in this other hand, primary B-CLL cells can be obtained easily from the coculture system by using a nucleofection-based strategy. PB of patients and represent the ideal system to study the Surprisingly, the introduction of circular plasmid DNA into molecular mechanisms of B-CLL pathogenesis. Major draw- B cells and B-CLL cells induced rapid apoptosis, which was backs for experiments with these cells have been the rapid independent of the type of transgene used, but dependent on the DNA concentration. However, transfection of these cells spontaneous apoptosis of B-CLL cells ex vivo and their with mRNA was highly efficient and resulted in sustained cell resistance to transient gene delivery with most currently viability and potent transgene expression. The described available gene transfer systems. procedure represents a new approach to study gene function In the present study, we established a coculture system using in primary B cells and B-CLL cells. the human BM-derived stromal cell line HS-5, which enables Leukemia (2007) 21, 1977–1983; doi:10.1038/sj.leu.2404863; long-term survival of primary human B cells and B-CLL cells published online 19 July 2007 Keywords: B-cell chronic lymphocytic leukemia; B cells; ex vivo. In addition, we developed a strategy to efficiently microenvironment; nucleofection; apoptosis; coculture transfect primary B cells and B-CLL cells, and document that introduction of circular plasmid DNA, but not mRNA or siRNA, into these cells causes apoptosis.

Materials and methods Introduction Primary cells and cell lines B-cell chronic lymphocytic leukemia (B-CLL) is the most 1 PB samples were obtained from 17 B-CLL patients (see common leukemia in adults in the western world. Unlike other Supplementary Information A) and five healthy donors after leukemias, there is only a small proportion of proliferating informed consent. All 17 B-CLL cases matched the standard neoplastic cells, which are localized in the so-called ‘pseudo- diagnostic criteria for B-CLL. The human BM-derived stromal follicles’ in the lymph nodes and are scattered in the bone 2–4 cell line HS-5 was purchased from American Type Culture marrow (BM) of the patients. The majority of leukemic cells Collection (Manassas, VA, USA). in B-CLL are nonproliferating cells arrested in the G0/G1 phase, which accumulate gradually in lymphoid organs, BM and peripheral blood (PB), due to a selective survival advantage 5 Cell isolation and culture enjoyed by these cells. Both, intrinsic defects in the cells PB mononuclear cells were isolated by Ficoll density gradient. regulation of programmed cell death (apoptosis) and an altered, Thereafter, B-CLL preparations consisted of 480% CD19 þ cells survival-stimulating microenvironment are discussed as major 6–11 as measured by flow cytometry and were used without further pathogenic factors of B-CLL. enrichment. Fractions of B-CLL cells were cryopreserved in cell The importance of external survival factors is exemplified by culture medium containing 10% fetal calf serum (FCS) and 10% the fact that B-CLL cells rapidly undergo spontaneous apoptosis dimethylsulfoxide. After thawing, B-CLL cells were cocultured for 1–2 days with HS-5 stromal cells before experiments were Correspondence: Professor Dr P Lichter, Department for Molecular performed. We reassessed the percentage and viability of B-CLL Genetics, German Cancer Research Center, Im Neuenheimer Feld cells thereafter, which were 490 and 470%, respectively. 280, Heidelberg 69120, Germany. þ E-mail: [email protected] CD19 cells from healthy donors were enriched by magnetic Received 30 October 2006; revised 23 May 2007; accepted 11 June bead-activated cell sorting (MACS) using CD19-MicroBeads and 2007; published online 19 July 2007 following the instructions of the manufacturer (Miltenyi Biotec, Apoptosis of B cells and B-CLL cells in coculture after nucleofection M Seiffert et al 1978 Bergisch Gladbach, Germany). The purity of the CD19 þ cell staining solution (BD Biosciences, Heidelberg, Germany). After preparations was 495% as measured by flow cytometry. All an incubation time of 15 min at 41C, stained cells were analyzed cells were maintained in Dulbecco’s modified Eagle’s medium by flow cytometry, gating on lymphocytes. Double-negative supplemented with 10% FCS, 4 mML-glutamine, 4.5 g/l glucose, cells were counted as viable cells. The results were confirmed 100 U/ml penicillin and 100 mg/ml streptomycin, and were on the basis of changes in forward light-scattering properties of cultured at 371C in a 10% CO2 humified incubator. For dead cells that have decreased cell size. cocultures, HS-5 cells were seeded at a density of 3 Â 105 cells/ml in 24-well plates 3 h before the addition of equal numbers of primary B cells or B-CLL cells. After the indicated Flow cytometry incubation times, B cells and B-CLL cells were separated from GFP expression after nucleofection with various GFP constructs the adherent HS-5 cells by shaking and carefully rinsing the was measured by flow cytometry, and transfection efficiency wells of the plate. Purity of the cell fractions after coculture was was calculated as percentage of 100% viable cells by excluding quantified by flow cytometry. For experiments with conditioned 7-AAD-positive cells from the analysis. Percentages of B cells in medium of HS-5 cultures, the medium was removed from more PB MACS preparations were quantified using PE-conjugated than 70% confluent stromal cell layers, freed from cells by monoclonal antibodies specific to CD19 (clone LT19; Miltenyi centrifugation for 10 min at 1000 g and added immediately to Biotec) and CD20 (clone LT20; Miltenyi Biotec). Purity of cell B-CLL cells in culture. In survival studies over several days, the fractions after separating cocultured cells was quantified on the conditioned medium was replaced daily. basis of different forward light-scattering properties of large HS-5 cells compared to small lymphocytes. All flow cytometry analyses were carried out using a FACSCalibur flow cytometer DNA and mRNA preparations equipped with CellQuest software (BD Biosciences). pmax green fluorescent protein (GFP) was obtained as part of the Human B cell Nucleofector Kit from Amaxa (Cologne, Germany). pcDNA-GFP was produced by excising maxGFP Western blotting out of pmaxGFP and cloning it into pcDNA6/V5-His cloning Transfected primary B-CLL cells were harvested and lysed in vector (Invitrogen, Karlsruhe, Germany) using KpnI and XhoI phosphate-buffered saline containing 0.5% NP-40, 1 mM MgCl2, restriction sites. CD79b was PCR amplified from B-CLL cells 2 mg/ml aprotinin and 200 mg/ml 4-(2-aminoethyl)-benzenesul- using the following primers: 50-GAGAGAGAATTCGTGAC fonyl fluoride (ABSF), for 1 h on ice. Cell lysates were separated CATGGCCAGGCTGGCGTTGTC-30 and 50-TCTCTCCTCGA through 10% sodium dodecyl sulfate–polyacrylamide gel GCTCCTGGCCTGGGTGCTCACCTACAG-30. The product was electrophoresis and transferred to nitrocellulose membranes, cloned into pcDNA6/V5-His cloning vector using EcoRI and which were incubated with anti-V5-horseradish peroxidase XhoI restriction sites. Plasmid DNA was prepared using a (HRP) antibodies or anti-b-actin-HRP antibodies (Invitrogen). QIAGEN Plasmid Kit or a QIAGEN EndoFree Plasmid Kit CD79b-V5 fusion protein or b-actin was visualized using ECL (QIAGEN, Hilden, Germany). For linearization of plasmid DNA, Blotting Detection Reagents (Amersham Biosciences, Bucking- pmaxGFP was digested with XhoI, PvuII or SspI, whereas hamshire, England). pcDNA6 constructs were digested with SphI or DraIII restriction enzymes, followed by ethanol precipitation. Complete digestion was confirmed by agarose gel electrophoresis. Capped and Results and discussion polyA-tailed mRNA were produced using an mMESSAGE mMACHINE T7 Ultra Kit (Ambion, Austin, TX, USA) and HS-5 stromal cells support survival of B-CLL cells following the instructions of the manufacturer. One microgram ex vivo of linearized pcDNA-GFP or pcDNA-CD79b served as the DNA Primary B cells from healthy donors and primary B-CLL cells died template to produce GFP-specific or CD79b-specific mRNA. within 1 week by spontaneous apoptosis when cultured in The quality of the mRNA preparations was examined with an medium that allows survival of various B cell lines (Figure 1a). Agilent Bioanalyzer using an RNA 6000 Nano Chip (Agilent, The rate of apoptosis was more variable and in general slower in Waldbronn, Germany). B-CLL cells compared to normal B cells. To establish an in vitro assay for B-CLL cells and B cells, we tested the ability of the human BM-derived stromal cell line HS-5 to support the survival Nucleofection of B-CLL cells and B cells ex vivo. As shown in Figure 1a, both A total 1–3 Â 106 primary B cells or B-CLL cells were transfected B-CLL cells and normal B cells were rescued from apoptosis using an Amaxa Human B cell Nucleofector Kit and following when cocultured with HS-5 cells. Survival of B-CLL cells could be the instructions of the manufacturer (Amaxa, Cologne, Germany). maintained for several months under these conditions (data not For both cell types, the nucleofector program U-15 was used. shown). By measuring the DNA content of the cells, we did not After nucleofection, cells were either added to pre-seeded HS-5 observe any cell-cycle activity of B cells or B-CLL cells cultured in cells or cultured without stroma. the presence of HS-5 cells. We further observed no increase in size of B cells under these conditions, as detected in activated B cells after B-cell receptor stimulation (unpublished observation). Cell viability To evaluate the importance of soluble factors in providing Apoptotic cell death was detected by flow cytometry using survival signals for B-CLL cells in vitro, we performed Annexin V and propidium iodide (PI) or 7-amino-actinomycin experiments with conditioned medium of HS-5 stromal cell (7-AAD) staining. Cells were harvested and resuspended in cultures. B-CLL cells were rescued from spontaneous apoptosis Annexin V-binding buffer (10 mM HEPES (4-(2-hydroxyethyl)-1- when cultured in conditioned medium, but survival was piperazineethanesulfonic acid)/NaOH, pH 7.4, 140 mM NaCl significantly lower compared to B-CLL cells cultured in the and 2.5 mM CaCl2) containing 10% Annexin V-FITC and 10% PI presence of HS-5 cells (Figure 1b). In addition, survival rates of staining solution, or 10% Annexin V-PE and 10% 7-AAD B-CLL cells in conditioned medium dropped considerably after

Leukemia Apoptosis of B cells and B-CLL cells in coculture after nucleofection M Seiffert et al 1979

Figure 1 Survival of primary B cells and B-CLL cells in vitro.(a) Primary B cells (solid lines) or B-CLL cells (dashed lines) were isolated from PB samples and seeded at a density of 3 Â 105 cells/ml in 24-well plates, either in the presence or in the absence of 3 Â 105 pre-seeded HS-5 cells. B cells and B-CLL cells were harvested at the indicated time points and cell viability was quantified by Annexin V-FITC and PI staining. Data were normalized to 100% viable cells at day 0. Vertical bars represent the s.d. from seven B-CLL samples (B-CLL 1–7) and three B cell samples, which were analyzed in duplicates. (b)3Â 105 primary B-CLL cells (B-CLL 1, 3, 5, 7 and 14–17) were cultured in 0.5 ml conditioned medium of HS-5 cells for 1–3 days, while adding fresh conditioned medium daily. Survival rates in these cultures were compared with B-CLL cells cultured either in the presence or in the absence of HS-5 stromal cells. Circles represent data points after 1 day, and triangles after 3 days of culture. B-CLL, B-cell chronic lymphocytic leukemia; PB, peripheral blood; PI, propidium iodide.

3 days of culture although fresh conditioned medium was added reduction of spontaneous apoptosis of B-CLL cells in condi- daily (data not shown). Therefore, soluble factors seem to play tioned medium of stromal cells, that can maintain survival in an important role in inducing survival of B-CLL cells, but either cocultures.13,19 An explanation for this contradiction could be these factors have very short half-life or additional signals are the differences in quality or quantity of soluble factors produced required that are only transduced through direct cell–cell by the stromal cells used in these studies compared to the contact between B-CLL cells and stromal cells. Our results are conditioned medium of HS-5 cells. A high sensitivity of the in contrast to previously reported data, which show no putative factors in the conditioned medium is corroborated by

Leukemia Apoptosis of B cells and B-CLL cells in coculture after nucleofection M Seiffert et al 1980 the fact that survival with conditioned medium was only interest by gating on either HS-5 cells or B-CLL cells. Since these observed when the medium was used absolutely fresh and two cell types differ enormously in size, they can be easily was refreshed daily. distinguished in forward/side scatter plots. In addition, there are a When using B-CLL/HS-5 cocultures, B-CLL cells had to be number of cell surface markers that can be used to distinguish or separated from the stroma for subsequent analysis. This was carried sort the two cell fractions if higher purities are required. out by simply removing the suspended leukemic cells from the Our data show that the stromal cell line HS-5 provides a long- adherent HS-5 cell layer, which led to cell purities of approxi- term, survival-stimulating microenvironment for B-CLL cells and mately 90% in the case of HS-5 cells, and of 75% in B-CLL cells. B cells. Therefore, these cocultures can be used as an in vitro Flow cytometry analyses were focused on the cell population of model system mimicking the in vivo behavior of these cells.

Figure 2 Transfection efficiency and cell viability after nucleofection of primary B cells and B-CLL cells. (a) A total 1 Â 106 freshly isolated B cells or B-CLL cells, as well as cryopreserved B-CLL cells were transfected with 2 mg pmaxGFP. Twenty-four hours after nucleofection, cell viability in the absence of HS-5 cells and GFP expression were quantified by flow cytometry. Transfection efficiency is given as fraction of 100% viable, 7-AAD-negative cells. Representative examples of three to five independently performed experiments are shown. (b) Survival of primary B cells and B-CLL cells transfected with 2 mg pmaxGFP (circles) was compared with ‘no DNA’ transfections (squares) and ‘no program’ control samples (triangles) after 3 h and after 1 and 4 days. Cell viability in the presence or absence of HS-5 cells was quantified after Annexin V-PE and 7-AAD staining. One representative example (B-CLL 10) out of three B-CLL and two B cell samples is shown. (c) Viability (black bars) and GFP expression (grey bars) were analyzed 24 h after transfection of 1 Â 106 primary B-CLL cells with various amounts of pmaxGFP DNA by flow cytometry. One representative example (B-CLL 12) out of three independently performed experiments is shown. (d) The percentages of early and late apoptotic cells were quantified 24 h after nucleofection with 2 mg pmaxGFP or without DNA by Annexin V-PE and 7-AAD staining and flow cytometry. One representative example (B-CLL 10) of the experiments listed in Table 1 is shown. 7-AAD, 7-amino-actinomycin; B-CLL, B-cell chronic lymphocytic leukemia; GFP, green fluorescent protein.

Leukemia Apoptosis of B cells and B-CLL cells in coculture after nucleofection M Seiffert et al 1981 Table 1 Cell viability of B-CLL cells after nucleofection with various DNA and mRNA preparations

Survival on stromal cells (%) B-CLL 5 B-CLL 6 B-CLL 9 B-CLL 10 B-CLL 11 B-CLL 1 (cryopreserved) B-CLL 4 (cryopreserved)

No program control 82.9 83.2 74.7 77.7 73.7 86.9 88.0 No DNA/RNA control 73.9 62.4 66.7 68.1 34.7 ND 76.2 2 mg pmaxGFP 43.0 29.2 46.5 38.8 26.7 48.5 ND 2 mg pcDNA-GFP 26.0 15.5 ND 20.2 ND 29.7 63.1 2 mg linearized pmaxGFP ND ND ND 68.2 ND 76.1 ND 2 mg linearized pcDNA-GFP 70.4 55.8 60.8 ND ND 70.5 ND 2 mg GFP mRNA 68.9 56.6 64.9 72.7 44.5 76.7 79.9 Abbreviations: B-CLL, B-cell chronic lymphocytic leukemia; GFP, green fluorescent protein; ND, not determined.

Uptake of circular plasmid DNA induces apoptosis of containing protein-coding transgenes, two plasmids containing primary B cells and B-CLL cells independently of the non-coding transgenes and one plasmid lacking a functional type of transgene used eukaryotic promoter sequence, as well as plain pcDNA6 vector. Transient gene delivery or knockdown studies offer a rapid Nucleofection of all constructs resulted in high apoptotic rates. strategy to study gene function in human cells. Primary To exclude the possibility that impure DNA preparations are the lymphocytes are, in general, difficult to transfect by methods reason for the apoptotic effect after transfection, we purified such as electroporation, lipofection or ballistic gene transfer. plasmid DNA using Qiagen’s EndoFree Plasmid Kit. However, Using the recently introduced Amaxa nucleofection technology, the rate of apoptosis was not reduced when endotoxin-free which is a non-viral, electroporation-based transfection method plasmid DNA preparations were used for transfections (Supple- that transports DNA or RNA molecules directly into the cell mentary Information B). We further tested whether the amount nucleus, primary B cells and B-CLL cells can be transfected with of transfected DNA has an influence on the survival rates of the very high efficiency. We established this technique for CD19 þ cells and could show that by decreasing the input of plasmid B cells, freshly isolated, as well as cryopreserved B-CLL cells DNA from 4 to 0.125 mg per transfection, the viability of the using the Human B cell Nucleofector Kit and pmaxGFP. With all cells increased, but as expected the expression of GFP three cell types, we achieved transfection efficiencies of decreased as well (Figure 2c). In addition, we repeatedly 50–70% (Figure 2a). GFP expression could be detected as early observed a positive correlation between the size of the plasmid as 3 h after nucleofection and was detectable for up to 5 days. In and the percentage of apoptosis, where large plasmids (e.g. addition, we quantified cell viability after nucleofection by pcDNA6 constructs) induced higher apoptotic rates compared Annexin V and 7-amino-actinomycin D (7-AAD) staining. to smaller plasmids (e.g. pmaxGFP) (Table 1 and Supplementary Survival rates of B cells or B-CLL cells 24 h after nucleofection Information C). Therefore, we conclude that the introduction of with pmaxGFP showed some variability (between 20 and 60%), plasmid DNA into primary B cells and B-CLL cells induces but were generally much lower compared to non-transfected apoptosis in these cells, even in the presence of a survival- cells. stimulating microenvironment provided by HS-5 cells. To elucidate the reason for the low viability of B cells and To learn more about the two types of apoptosis, that is, B-CLL cells after nucleofection, we compared survival rates of spontaneous apoptosis due to the lack of a survival-stimulating cells transfected with pmaxGFP, ‘no DNA’-transfected cells, microenvironment and apoptosis induced by uptake of plasmid which were treated equally to the transfected samples, but no DNA after nucleofection of B-CLL cells, we quantified DNA was added to the sample, and ‘no program’ control cells, percentages of early apoptotic cells, which are Annexin where 2 mg of pmaxGFP was added to the cells in the V-positive, but 7-AAD-negative and Annexin V and 7-AAD nucleofector solution, but the nucleofector program was not double-positive late apoptotic cells, in transfected and non- performed. Transfected cells were cultured over 3 days in the transfected samples. Our results revealed that 24 h after presence or absence of HS-5 stromal cells (Figure 2b). Cell nucleofection, introduction of plasmid DNA induced an survival 3 h after nucleofection with or without DNA was similar increase in late apoptotic cells, whereas the lack of HS-5 cells to the ‘no program’ control. The nucleofection procedure itself is led to an increase in early apoptotic cells but not in late obviously not disruptive to B cells or B-CLL cells, but we apoptotic cells after pmaxGFP transfection (Figure 2d). These repeatedly observed an increase in apoptotic cell death after data indicate that apoptosis induced by the introduction of 24 h as a result of the nucleofection procedure (‘no DNA’ plasmid DNA into B cells or B-CLL cells is more rapid and control versus ‘no program’ control). The viability of cells could independent of spontaneous apoptosis triggered by the lack of be extremely improved by transferring them immediately after survival-stimulating stromal cells. nucleofection to HS-5 stromal cells or to their conditioned medium. Therefore, it is highly recommendable to combine nucleofection of primary B cells or B-CLL cells with an in vitro Nucleofection of mRNA results in high transfection culture system that has long-term, survival-stimulating activity, efficiency and cell viability like HS-5 cells. To investigate if the structure of the transfected nucleic acids has Surprisingly, the viability of cells transfected with pmaxGFP an effect on the apoptotic rate of B-CLL cells after nucleofection, was repeatedly lower compared to the ‘no DNA’ control, and we performed experiments using transfer of linearized plasmid the presence of HS-5 cells only partly improved cell survival DNA and in vitro transcribed mRNA into primary B cells and (Figure 2b and Table 1). This observed effect was independent of B-CLL cells. For linearization of plasmid DNA, we used different the plasmid preparation and transgene expression. Transfections restriction enzymes (XhoI, PvuII and SspI for pmaxGFP, and SphI with a second GFP-containing vector, pcDNA-GFP, resulted in and DraIII for pcDNA6 constructs) to ensure that the observed lower transfection efficiencies, but in higher apoptosis rates effects are independent of the restriction site and of the resulting compared to pmaxGFP. We further tested four plasmids template length downstream of the coding sequence. The

Leukemia Apoptosis of B cells and B-CLL cells in coculture after nucleofection M Seiffert et al 1982 observe any induction of apoptosis after uptake of these molecules (data not shown). Therefore, we conclude that only transfer of circular plasmid DNA, but not linear DNA or RNA molecules, induces apoptosis in transfected B cells or B-CLL cells. Apoptosis caused by the uptake and accumulation of DNA in the cytoplasm has been described for other cell types as well, including dendritic cells, macrophages and monocytic cell lines.20–22 In addition, it has been shown that introduction of linearized plasmid DNA into monocytic cells resulted in reduced toxicity compared to supercoiled circular DNA,21 which is in accordance with our results. A possible explanation for apoptotic cell death induced by the uptake of DNA might be issued from immune functions of these cells after certain retroviral infections.23–25 Since B lymphocytes share with the cell types mentioned above the ability to present antigens, it is likely that they possess a similar response mechanism leading to apoptosis after cytoplasmic accumulation of foreign DNA. However, since the sensitivity of B lymphocytes for supercoiled circular DNA and linearized DNA differs considerably, the observed effect might be part of an uncharacterized viral defense system. We next evaluated the expression levels of the transgene after nucleofection with linearized pmaxGFP and GFP-specific mRNA. As shown in Figure 3b, nucleofection with linearized pmaxGFP resulted in decreased transgene expression compared to circular pmaxGFP (33% GFP þ cells compared to 58% GFP þ cells). However, nucleofection of GFP mRNA into primary B cells and B-CLL cells resulted in even higher transfection efficiencies, with 80–90% GFP-positive cells. The intensity of the GFP signal per cell was weaker in transfections with mRNA compared to pmaxGFP transfections. This lower level of transgene expression is, however, in most cases more similar to the endogenous expression level of proteins and might therefore be more suitable to reveal functions of transgenes. We observed GFP expression in nucleofected cells from 3 h to 4 days after nucleofection with GFP-specific mRNA (data not shown). One reason for this rather long duration of transgene expression could be the relatively high stability of maxGFP in human cells. The overall duration of transgene expression depends certainly on the transgene itself and needs to be Figure 3 Cell viability and transfection efficiency after nucleofection determined for each experiment individually. of B-CLL cells with various DNA and mRNA preparations. (a) A total 6 We analyzed further transgene expression after nucleofection 1 Â 10 freshly isolated or cryopreserved B-CLL cells were nucleo- of B-CLL cells with pcDNA-CD79b plasmid DNA, linearized fected with the indicated DNA and mRNA preparations. After nucleofection (18–41 h), cell viability was quantified after Annexin pcDNA-CD79b and CD79b-specific mRNA by western blotting. V-PE and 7-AAD staining. One representative example (B-CLL 10) out Figure 3c shows that transfection with mRNA results in potent of seven independently performed experiments (see Table 1) is shown. transgene expression 6 h after transfection. A weaker expression (b) GFP expression 24 h after transfection was analyzed by FACS of CD79b-V5 was detected after transfection with pcDNA- analysis and is indicated as a fraction of 100% viable, 7-AAD-negative CD79b plasmid DNA, which is in line with the lower cells. (c) Transgene expression was analyzed by western blotting 6 h transfection efficiency observed with pmaxGFP compared to after transfection of 3 Â 106 B-CLL cells with 2 mg of the indicated DNA or mRNA preparations. CD79b-V5 fusion protein was detected GFP-specific mRNA. A further explanation for this weaker using anti-V5-HRP antibodies and ECL detection reagents. Antibodies transgene expression is the observation that transfections with to b-actin were used to confirm equal protein loading. One pcDNA-GFP, with a size of 5.8 kb, repeatedly resulted in lower representative example (B-CLL 17) out of three independently efficiencies compared to transfections with pmaxGFP with a size performed experiments is shown. 7-AAD, 7-amino-actinomycin; of 3.5 kb. Therefore, for an adequate estimation of transfection B-CLL, B-cell chronic lymphocytic leukemia; GFP, green fluorescent efficiency, it is important to use GFP-containing constructs protein. within the same vector backbone as used for the gene of interest. We further observed even weaker expression levels of CD79b- in vitro-transcribed mRNA molecules were capped and polyA- V5 after transfections with linearized pcDNA-CD79b, which tailed for stabilization reasons. Interestingly, cell survival after again confirms our results with circular and linearized nucleofection with linearized plasmid DNA or mRNA was pmaxGFP. Our data therefore show that it is highly recommend- comparable to the control experiments, where no DNA or RNA able to use mRNA molecules instead of plasmid DNA when was added to the nucleofection protocol (Figure 3a and Table 1). transfecting primary B cells or B-CLL cells. In addition, we introduced siRNA molecules of selected In summary, we present a new approach for maintenance of specificities into primary B cells and B-CLL cells and did not primary B cells and B-CLL cells in an in vitro model system

Leukemia Apoptosis of B cells and B-CLL cells in coculture after nucleofection M Seiffert et al 1983 mimicking the in vivo situation of these cells. Moreover, this by different microenvironmental elements. Curr Top Microbiol model system allows for the detailed evaluation of transient Immunol 2005; 294: 135–145. expression of transgenes in primary B cells and B-CLL cells. The 11 de Totero D, Meazza R, Zupo S, Cutrona G, Matis S, Colombo M described nucleofection protocol leads to high transfection et al. Interleukin-21 receptor (IL-21R) is up-regulated by CD40 triggering and mediates proapoptotic signals in chronic lympho- efficiency and good cell viability of primary B cells and B-CLL cytic leukemia B cells. Blood 2006; 107: 3708–3715. cells by transferring mRNA into these cells with Amaxa’s 12 Collins RJ, Verschuer LA, Harmon BV, Prentice RL, Pope JH, Kerr Nucleofector technology and by coculturing transfected cells JF. Spontaneous programmed death (apoptosis) of B-chronic with the stromal cell line HS-5. This protocol provides a lymphocytic leukaemia cells following their culture in vitro. Br J powerful tool to study gene function in primary B cells and Haematol 1989; 71: 343–350. B-CLL cells. 13 Panayiotidis P, Jones D, Ganeshaguru K, Foroni L, Hoffbrand AV. Human bone marrow stromal cells prevent apoptosis and support the survival of chronic lymphocytic leukaemia cells in vitro. Br J Haematol 1996; 92: 97–103. Acknowledgements 14 Ringshausen I, Schneller F, Bogner C, Hipp S, Duyster J, Peschel C et al. Constitutively activated phosphatidylinositol-3 kinase (PI-3K) The authors thank Daniel Mertens for helpful discussions and is involved in the defect of apoptosis in B-CLL: association with suggestions and for his critical review of this paper and Stephan protein kinase Cdelta. Blood 2002; 100: 3741–3748. Wolf for technical advice. MS is supported by the German Cancer 15 Cuni S, Perez-Aciego P, Perez-Chacon G, Vargas JA, Sanchez A, Research Center and the Helmholtz Association of German Martin-Saavedra FM et al. A sustained activation of PI3K/NF- kappaB pathway is critical for the survival of chronic lymphocytic Research Centers. leukemia B cells. Leukemia 2004; 18: 1391–1400. 16 Sanz L, Garcia-Marco JA, Casanova B, de La Fuente MT, Garcia- Gila M, Garcia-Pardo A et al. Bcl-2 family gene modulation during References spontaneous apoptosis of B-chronic lymphocytic leukemia cells. Biochem Biophys Res Commun 2004; 315: 562–567. 1 Rozman C, Montserrat E. Chronic lymphocytic leukemia. New 17 Nilsson K. Human B-lymphoid cell lines. Hum Cell 1992; 5:25–41. Engl J Med 1995; 333: 1052–1057. 18 Stacchini A, Aragno M, Vallario A, Alfarano A, Circosta P, Gottardi 2 Lennert K. Malignant Lymphomas other than Hodgkin’s Disease: D et al. MEC1 and MEC2: two new cell lines derived from Histology, Cytology, Ultrastructure, Immunology. Springer-Verlag: B-chronic lymphocytic leukaemia in prolymphocytoid transforma- Berlin, 1978, 119–126. tion. Leukemia Res 1999; 23: 127–136. 3 Granziero L, Ghia P, Circosta P, Gottardi D, Strola G, Geuna M 19 Lagneaux L, Delforge A, Bron D, De Bruyn C, Stryckmans P. et al. Survivin is expressed on CD40 stimulation and interfaces Chronic lymphocytic leukemic B cells but not normal B cells are proliferation and apoptosis in B-cell chronic lymphocytic leuke- rescued from apoptosis by contact with normal bone marrow mia. Blood 2001; 97: 2777–2783. stromal cells. Blood 1998; 91: 2387–2396. 4 Messmer BT, Messmer D, Allen SL, Kolitz JE, Kudalkar P, Cesar D 20 Stacey KJ, Ross IL, Hume DA. Electroporation and DNA- et al. In vivo measurements document the dynamic cellular dependent cell death in murine macrophages. Immunol Cell Biol kinetics of chronic lymphocytic leukemia B cells. J Clin Invest 1993; 71: 75–85. 2005; 115: 755–764. 21 Shimokawa T, Okumura K, Ra C. DNA induces apoptosis in 5 Chiorazzi N, Rai KR, Ferrarini M. Chronic lymphocytic leukemia. electroporated human promonocytic cell line U937. Biochem New Engl J Med 2005; 352: 804–815. Biophys Res Commun 2000; 270: 94–99. 6 Dighiero G, Travade P, Chevret S, Fenaux P, Chastang C, Binet J. 22 Van De Parre TJL, Martinet W, Schrijvers DM, Herman AG, B-cell chronic lymphocytic leukemia: present status and future De Meyer GRY. mRNA but not plasmid DNA is efficiently directions. French Cooperative Group on CLL. Blood 1991; 78: transfected in murine J774A.1 macrophages. Biochem Biophys Res 1901–1914. Commun 2005; 327: 356–360. 7 Meinhardt G, Wendtner CM, Hallek M. Molecular pathogenesis of 23 Keshet E, Temin HM. Cell killing by spleen necrosis virus is chronic lymphocytic leukemia: factors and signaling pathways correlated with a transient accumulation of spleen necrosis virus regulating cell growth and survival. J Mol Med 1999; 77: 282–293. DNA. J Virol 1979; 31: 376–388. 8 Caligaris-Cappio F. Role of the microenvironment in chronic 24 Temin HM, Keshet E, Weller SK. Correlation of transient lymphocytic leukemia. Br J Haematol 2003; 123: 380–388. accumulation of linear unintegrated viral DNA and transient cell 9 Granziero L, Circosta P, Scielzo C, Frisaldi E, Stella S, Geuna M killing by avian leukosis and reticuloendotheliosis viruses. Cold et al. CD100/Plexin-B1 interactions sustain proliferation and Spring Harb Symp Quant Biol 1980; 44: 773–778. survival of normal and leukemic CD5+ B lymphocytes. Blood 25 Donahue PR, Quackenbush SL, Gallo MV, deNoronha CM, 2003; 101: 1962–1969. Overbaugh J, Hoover EA et al. Viral genetic determinants of 10 Ghia P, Circosta P, Scielzo C, Vallario A, Camporeale A, T-cell killing and immunodeficiency disease induction by the Granziero L et al. Differential effects on CLL cell survival exerted feline leukemia virus FeLV-FAIDS. J Virol 1991; 65: 4461–4469.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Leukemia