Conditional immortalization of human B cells by CD40 ligation

Abstract

It is generally assumed that human differentiated cells have a limited life-span and proliferation capacity in vivo, and that genetic modifications are a prerequisite for their immortalization in vitro. Here we readdress this issue, studying the long-term proliferation potential of human B cells. It was shown earlier that human B cells from peripheral blood of healthy donors can be efficiently induced to proliferate for up to ten weeks in vitro by stimulating their receptor CD40 in the presence of interleukin-4. When we applied the same stimuli under conditions of modified cell number and culture size, we were surprised to find that our treatment induced B cells to proliferate throughout an observation period of presently up to 1650 days, representing more than 370 population doublings, which suggested that these B cells were immortalized in vitro. Long-term CD40-stimulated B cell cultures could be established from most healthy adult human donors. These B cells had a constant phenotype, were free from Epstein-Barr virus, and remained dependent on CD40 ligation. They had constitutive telomerase activity and stabilized telomere length. Moreover, they were susceptible to activation by Toll-like receptor 9 ligands, and could be used to expand antigen-specific cytotoxic T cells in vitro. Our results indicate that human somatic cells can evade senescence and be conditionally immortalized by external stimulation only, without a requirement for genetic manipulation or oncoviral infection. Conditionally immortalized human B cells are a new tool for immunotherapy and studies of B cell oncogenesis, activation, and function.

Figure 1. Conditions for the establishment of CD40-stimulated B cells.

B cell cultures were established from 10⁵ PBMC (left) or 2×10⁶ PBMC (right). For each condition, four replicate cultures were set up from each of five donors. CD19+CD3– B cell and CD3+CD19– T cell content was assessed by flow cytometry. The increase in total B cell numbers (A), the proportions of B cells and T cells in the cultures on day 27 (B), and total increases in cell number on days 27 and 40 (C) were calculated.

Figure 2. Overview of a protocol to establish long-term CD40-stimulated B cell cultures.

Figure 3. Long-term culture of CD40-stimulated B cell lines.

(A) Culture periods and minimal proliferation doubling levels of Epstein-Barr virus–free CD40-stimulated B cell lines that were cultured for more than 200 days. Each cell line shown is from a different donor. Solid dots represent B cell lines that are currently proliferating in culture or were cryoconserved as proliferating cultures. Open dots indicate B cell lines that ceased to proliferate at the indicated time. (B) Proliferation of established CD40-stimulated B cell lines during a 30-day period. At the start of analysis (day n), B cell lines LXL5 and HXL7 were at day 884 of culture, the other B cell lines were at day 570, 512, 430, 251 (two B cell lines) or 155 (five B cell lines). (C) Dependence of B lymphoblast proliferation on CD40L and IL-4. B cells HXL7 or LXL5 (day 586 of culture) were grown for 11 days on CD40L-expressing murine fibroblasts or control fibroblasts in the presence or absence of IL-4, before cell counting and viability assessment by flow cytometry.

Figure 4. Long-term CD40-stimulated B cells remain EBV-free.

(A) PCR was used to detect the EBV EBNA1 gene in CD40-stimulated B cells from eleven donors after various culture periods (the number of days is indicated in parentheses for each cell line). EBV DNA and EBV-infected B lymphoblastoid cell lines GXLK and LXB2 were used as positive controls. (B) The PCR protocol allowed the detection of one EBV-infected cell equivalent per reaction. (C) In this example of a routine PCR analysis of early-passage CD40-stimulated B cell cultures, EBV was detected in one of 17 cultures from three different EBV-positive donors. B cell cultures had been set up with 100,000 PBMC each and were tested on day 80–90 of culture.

Figure 5. Phenotype of long-term CD40-stimulated B cells.

(A) Five CD40-stimulated B cell lines from different donors after 364 to 818 days of culture were stained for CD19, HLA-ABC and HLA-DR, the costimulatory molecules CD80 and CD86, and the adhesion molecules CD54 and CD11a, and were analyzed by flow cytometry. Thick lines represent staining with specific antibody, thin lines with a matched isotype control. Peripheral blood lymphocytes gated on forward and sideward scatter (top left diagram) or additionally gated for CD19 expression (top row, marked with an asterisk) and an EBV-infected B lymphoblastoid cell line, BXB4, were used for comparison. (B) One CD40-stimulated B cell line, HXL7, was stained for several of the above markers after various periods of culture, ranging from 55 to 735 days. (C) CD40-stimulated B cells, EBV-transformed B lymphoblastoid cell lines, and PBL were analyzed for surface expression of a panel of B cell differentiation and activation markers. Specific stainings and controls are represented as in (A).

Figure 6. Immunoglobulin light chain expression patterns.

CD19+ PBMCs ex vivo, EBV-transformed B lymphoblastoid cell lines, and CD40-stimulated B cells from various different donors were cultured for the indicated periods of time and analyzed for expression of Igκ and Igλ chains by flow cytometry.

Figure 7. Telomere length and telomerase activity in long-term CD40-stimulated B cells.

(A) Telomeric terminal restriction fragment (TRF) length and telomerase activity in two long-term CD40 B lymphoblast lines, HXL7 and LXL5, over time. Telomere-containing terminal restriction fragments were obtained by restriction hydrolysis of genomic DNA. Telomerase activity was semiquantitatively determined according to a modified telomeric repeat amplification protocol (TRAP) in an ELISA format. An internal PCR standard template was co-amplified in each reaction and detected with an independent probe. As a positive control, a template containing 8 telomeric repeats was PCR-amplified and analyzed in the same manner. Relative telomerase activities of samples with respect to the positive control were calculated. (B) Telomerase activities of various CD40-stimulated B lymphoblast lines. Relative telomerase activities of CD40 B lymphoblasts were in a similar range as standard immortalized cell lines (K562 and HEK293) and the artificial telomere product used as positive control. The maximum number of CD40L-expressing feeder cells that might have been present in the CD40 B lymphoblast preparations, corresponding to a dilution of 1/30, had negligible telomerase activity.

Figure 8. Long-term CD40-stimulated B cells are activated by oligodeoxynucleotides.

Long-term CD40 B cells were treated with TLR9-agonistic oligonucleotides in the presence of CD40L-expressing or non-expressing fibroblast feeder cells. After 4 days, secretion of IL-6 and IL-12p40 was analyzed in an ELISA. TLR9 stimuli and controls included a phosporothioate oligonucleotide (ODN 2006) containing 4 CpG motifs (“CpG”), an oligonucleotide of the same sequence except that each CpG dinucleotide was replaced by a GpC dinucleotide (“GpC”), an oligonucleotide composed of twenty phosphorothioate deoxycytidines (“CCC”), and a medium-alone control. For each condition, two CD40 B cell cultures from different donors were tested. At the time of the experiment, CD40 B cells had been cultured for 570 days (BXL12, top left), 652 days (GXL5, top left), 1074 days (line LXL5, bottom left), or 702 days (line GXL5, bottom right).

Figure 9. Specific expansion of cytotoxic T cells by stimulation with long-term CD40-stimulated B cells.

(A) Frequency of antigen-specific CD8+ T cells in four T cell lines after 33 days of repeated stimulation with peptide-loaded autologous long-term CD40 B cells. B cells had been cultured for 693 days at the beginning of the T cell stimulation. Either of four HLA-A2-restricted antigenic peptides derived from viral or melanoma antigens (abbreviated NLV, GLC, CLG, and ELA) was used for stimulation. The frequency of specific T cells was assessed by staining with the corresponding specific HLA/peptide multimer reagent. For control stainings, an HLA/peptide multimer containing an irrelevant peptide was used for staining (lower row): ELA for NLV-stimulated cultures and NLV for all other cultures. (B) Expansion of antigen-specific T cells after stimulation with peptide-loaded long-term CD40 B cells. Total cell numbers (top) in the T cell cultures were determined by microscopic counting. The proportion of antigen-specific CD8+ T cells, represented in linear (middle) and logarithmic scale (bottom), was determined by specific HLA/peptide multimer staining at various times of T cell culture. (C) The cytotoxic reactivity of two of these T cell cultures against cells endogenously presenting the target antigen was assessed in a calcein release assay. Top: T cells expanded with the peptide NLV (from CMV pp65) were tested against HLA-A2-matched mini-LCLs endogenously expressing pp65 or not, autologous CD40 B cells loaded with the target peptide NLV or the control peptide GLC, or K562 cells to test for natural killer-like reactivity. Bottom: T cells expanded with the peptide ELA from MelanA were tested against HLA-A2-matched melanoma cell lines either expressing MelanA (Mel624) or not (A375) that had either been additionally loaded with the ELA peptide or not. Further targets included autologous CD40 B cells loaded with the target peptide ELA or the control peptide NLV, and K562 cells.

Figure 10. Culture periods of CD40/IL-4-stimulated B cells reported in different studies.

For each study, the longest period during which B cell proliferation was observed is represented. A vertical bar indicates that proliferation was observed to terminate. A triangle indicates that the observation period ended at a time when B cells were still profilerating. An asterisk marks studies in which EBV infection of late-passage CD40/IL-4-stimulated B cell cultures was reported as a typical outcome of B cell cultivation, or was positively observed without information given about the frequency of such an event. Vertical dashed lines indicate proposed limits to the proliferation of non-immortalized human fibroblasts , T cells or EBV-transformed B cells , using an estimation that these cells divide twice a week on average.