Transfer and stable transgene expression of a mammalian artificial chromosome into bone marrow-derived human mesenchymal stem cells.

Mammalian artificial chromosomes (ACEs) transferred to autologous adult stem cells (SCs) provide a novel strategy for the ex vivo gene therapy of a variety of clinical indications. Unlike retroviral vectors, ACEs are stably maintained, autonomous, and nonintegrating. In this report we assessed the delivery efficiency of ACEs and evaluated the subsequent differentiation potential of ACE-transfected bone marrow-derived human mesenchymal stem cells (hMSCs). For this, an ACE carrying multiple copies of the red fluorescent protein (RFP) reporter gene was transferred under optimized conditions into hMSCs using standard cationic transfection reagents. RFP expression was detectable in 11% of the cells 4-5 days post-transfection. The RFP-expressing hMSCs were enriched by high-speed flow cytometry and maintained their potential to differentiate along adipogenic or osteogenic lineages. Fluorescent in situ hybridization and fluorescent microscopy demonstrated that the ACEs were stably maintained as single chromosomes and expressed the RFP transgenes in both differentiated cultures. These findings demonstrate the potential utility of ACEs for human adult SC ex vivo gene therapy.

Efficient in-vitro transfer of a 60-Mb mammalian artificial chromosome into murine and hamster cells using cationic lipids and dendrimers.

Non-integrating artificial chromosomes represent a potentially promising approach to ex-vivo and in-vivo gene therapy applications. These large vectors require an efficient means for delivery to target cells. We have evaluated a panel of twenty-one commercially available transfection agents for their ability to mediate the in-vitro transfer of a 60-Mb murine artificial chromosome consisting of mouse major satellite DNA and a payload including a marker gene (hygromycin B) and a reporter gene (lacZ). A rapid screening procedure utilizing iododeoxyuridine-incorporated artificial chromosomes facilitated the assessment of different transfection conditions. The results were confirmed by cytogenetic analysis of positively transfected clones. By transfecting both hamster lung fibroblast cells (V79-4) and murine connective tissue cells [L-M(TK-)], the best results were obtained using either Superfect (cationic dendrimer) or LipofectAMINE 2000 (cationic lipid) with protocols adapted for metaphase chromosome preparation. Transfection efficiencies of 10(-4)-10(-2) (0.01-1%) were routinely observed, and recipient cells were able to maintain expression of the reporter gene over the total length of the experiment. This represents a significant advance over our previous attempts at mass-transfection of artificial chromosomes using microcell fusion, where we routinely achieved efficiencies at least two orders of magnitudes less than reported here. These data are particularly noteworthy given that lipid-mediated gene transfer typically involves transfecting millions of plasmids (1 microg of DNA from a 5 kb plasmid is approximately 1.2 x 10(11) copies) to each cell whereas the much larger artificial chromosomes comprise only a one-to-one ratio, yet achieve transfection efficiencies of (10(-2)-10(-1)), that is, comparable to our results. These data suggest that artificial chromosomes containing therapeutic genes can be successfully delivered to target cells in vitro using well-established transfection agents.

A flow cytometry technique for measuring chromosome-mediated gene transfer.

BACKGROUND: Using artificial chromosome expression systems (ACes), we have developed a unique and rapid screening technique to quantify delivery of foreign DNA into cells in vitro. Delivery was measured within 24 h after transfection, using flow cytometry to detect the transfer of ACes labeled with thymidine analogue. This technique can be used to optimize delivery parameters of ACes and heterologous DNA into cells and eventually tissue. METHOD: Chinese hamster ovary (CHO) cells carrying artificial chromosomes were grown in media supplemented with iododeoxyuridine (IdUrd). The 60-mb artificial chromosome was purified by flow cytometry sorting and transfected into Chinese hamster lung fibroblast cells (V79-4) or mouse connective tissue cells [LM(tk-)] using LipofectAMINE 2000trade mark, a cationic lipid, and Superfecttrade mark, a cationic dendrimer. The cells were incubated with an FITC-conjugated anti-bromodeoxyuridine (BrdUrd) antibody and analyzed by flow cytometry. IdUrd-incorporated artificial chromosome expressing green fluorescent protein (GFP) was transfected into V79-4 cells. Delivery was measured at 24 h and GFP expression was detected at 48 h. RESULTS: The delivery of intact artificial chromosomes into V79-4 and LMtk- cells was detected within 2 h and up to 48 h post-transfection. Maximum delivery rates of 20% and 14% were observed using LipofectAMINE 2000 and Superfect, respectively. Flow cytometry data correlated with microscopic observations. IdUrd incorporation resulted in less quenching after staining with Hoechst 33258 and chromomycin A3 than BrdUrd incorporation. The fluorescence intensity of the FITC-conjugated anti-BrdUrd antibody was greater with IdUrd-incorporated chromosomes than with BrdUrd-incorporated chromosomes. CONCLUSION: The results indicate that IdUrd-labeled artificial chromosomes can be detected 24 h after transfection. This efficient, sensitive, high-throughput detection technique is being used to evaluate and optimize other transfer technologies (e.g., electroporation and sonoporation), different delivery reagents, and protocols in a variety of cells in vitro. This work represents the first step in utilizing artificial chromosomes as nonviral vectors for gene therapy.

Increase in the fraction of necrotic, not apoptotic, cells in SiHa xenograft tumours shortly after irradiation.

BACKGROUND AND PURPOSE: Approximately 18% of the cells recovered by rapid mechanical dissociation of SiHa xenograft tumours contain large numbers of DNA strand breaks. The number of damaged cells increases to 30-40% 4-6 h after exposure to 5 or 15 Gy, returning to normal levels by 12 h. This observation is reminiscent of the rate of production of apoptotic cells in other murine and human xenograft tumours. The nature of this damage, rate of development and relation to cell proliferation rate were therefore examined in detail. MATERIALS AND METHODS: SiHa human cervical carcinoma cells were grown as xenograft tumours in SCID mice. Single-cell suspensions were prepared as a function of time after irradiation of the mouse and examined for DNA damage using the alkaline comet assay. Cell cycle progression was measured by flow cytometry evaluation of anti-bromodeoxyuridine-labelled tumour cells. RESULTS: Significant numbers of apoptotic cells could not be detected in irradiated SiHa tumours using an end-labelling assay, electron microscopy, or histological examination of thin sections. Instead, xenograft cells exhibiting extensive DNA damage in the comet assay were predominantly necrotic cells. The increase in the proportion of heavily damaged cells 4-6 h after irradiation could be the result of an interplay between several factors including loss of viable cells and change in production or loss of necrotic cells. Analysis of the progression of BrdUrd-labelled cells confirmed that while 35% of cells from untreated SiHa tumours had divided and entered G1 phase by 6 h after BrdUrd injection, none of the labelled cells from tumours exposed to 5 or 15 Gy had progressed to G1. CONCLUSIONS: The increase in the percentage of SiHa tumour cells with extensive DNA damage 4-6 h after irradiation is attributable to necrosis, not apoptosis. Cell cycle progression and cell loss are likely to influence the kinetics of appearance of both apoptotic and necrotic cells in irradiated tumours.

Resistance to DNA denaturation in irradiated Chinese hamster V79 fibroblasts is linked to cell shape.

Exponentially growing Chinese hamster V79-171b lung fibroblasts seeded at high density on plastic (approximately 7 x 10(3) cells/cm2) flatten, elongate, and produce significant amounts of extracellular fibronectin. When lysed in weak alkali/high salt, the rate of DNA denaturation following exposure to ionizing radiation is exponential. Conversely, cells plated at low density (approximately 7 x 10(2) cells/cm2) on plastic are more rounded 24 h later, produce little extracellular fibronectin, and display unusual DNA denaturation kinetics after X-irradiation. DNA in these cells resists denaturation, as though "constraints" to DNA unwinding have developed. Cell doubling time and distribution of cells in the growth cycle are identical for both high and low density cultures as is cell survival in response to radiation damage. The connection between DNA conformation and cell shape was examined further in low density cultures grown in conditioned medium. Under these conditions, cells at low density were able to elongate, and DNA denaturation of low density cultures was identical to that of high density cultures. Conversely, cytochalasin D, which interferes with actin polymerization causing cells to "round up" and release fibronectin, allowed development of constraints in high density cultures. These results suggest that DNA conformation is sensitive to changes in cell shape which result when cells are grown in different environments. However, these changes in DNA conformation detected by the DNA unwinding assay do not appear to play a direct role in radiation-induced cell killing.

Schedule dependence for cisplatin and etoposide multifraction treatments of spheroids.

Cisplatin and etoposide are both widely used as antineoplastic chemotherapeutic agents. Two approaches are generally adopted when an attempt is made to maximize the efficiency of these drugs: concurrent use (where synergism is expected) or sequential administrations (exploiting the antiproliferative effects of etoposide). To differentiate between these approaches in a quantitative manner, we exposed an in vitro tumor model (V79 multicell spheroids) to the drugs, using treatment regimens with a constant weekly dose intensity. Some treatment schedules suggested the development of drug resistance, but this resulted from a changing growth fraction in the spheroids rather than from selection for or induction of cellular resistance. Fewer administrations of larger doses were generally less satisfactory than multiple, smaller treatments. The most effective sequence was an alternating regimen, by which the cytoreductive effects of cisplatin resulted in recruitment of quiescent cells into active proliferation, enhancing in turn the efficiency of subsequent etoposide treatment.

Sequencing radiation and adriamycin exposures in spheroids to maximize therapeutic gain.

Chinese hamster V79 spheroids were used as a tumor model to investigate sequence effects for combination treatments with adriamycin and radiation. Using cell sorting techniques, the subpopulation of cells most resistant to each modality as a single agent was identified, and combination treatments were then evaluated with the intent of identifying protocols leading to enhanced efficacy against the cell subpopulations that normally limited the success of single-agent treatments. Pre-irradiation drug exposure had the greatest therapeutic potential, not due to drug/radiation interactions, but rather, to drug-induced spheroid reoxygenation. Conversely, post-irradiation drug exposure was judged to have little potential for therapeutic gain, since the same cell subpopulations were killed (or spared) by the two agents.

Tumor resistance to therapy: a genetic or kinetic problem?

Chinese hamster V79 spheroids exposed to cisplatin for 2 hr daily for 3 weeks responded very similarly to tumors undergoing chemotherapy. Initially, the number of viable cells per spheroid decreased in a dose-dependent fashion but, after several treatments, the apparent effectiveness of the cisplatin decreased. At low doses, spheroid regrowth eventually occurred despite continued therapy. A detailed examination of the cellular basis for this response showed that the so-called acquired resistance was not due to a change in cellular responsiveness to cisplatin but, rather, was the result of a marked increase in the cellular growth fraction in the treated spheroids.

Response of cell subpopulations in spheroids to radiation-drug combinations.

Chinese hamster V79 cells grown as multicell spheroids were treated with radiation and/or selected antineoplastic drugs (5-fluorouracil, cisplatin, doxorubicin, mitomycin, and carmustine) administered at equitoxic levels in defined preirradiation and postirradiation sequences. Clonogenicity of the treated cells was determined, and the results from conventional assays were compared with those from cell-selection experiments, using a fluorescence-activated cell sorter, to determine the degree of interaction and regions of activity of the agents. Marked sequence-dependent differences in overall toxicity were observed; drug toxicity, spheroid reoxygenation, and direct sensitization appeared to be operative to a different extent with each drug and sequence. Maximal interaction between all drugs and radiation was, however, observed when drug exposure immediately preceded irradiation.