Collections of human DNA fragments are maintained for research purposes as clones in bacterial host cells. However, for unknown reasons, some regions of the human genome appear to be unclonable or unstable in bacteria. The team led by Jean-Michel Vos [University of North Carolina at Chapel Hill (UNCCH)] has developed a system using episomes (extrachromosomal, autonomously replicating DNA) that maintains large DNA fragments in human cells. This human artificial episomal chromosomal (HAEC) system may prove useful for coverage of these especially difficult regions. In the broader biomedical community, the HAEC system also shows promise for use in functional genomics and gene therapy. In the article below, Vos discusses some recent improvements to the HAEC system and its application to mapping, sequencing, and functionally studying human and mouse DNA.

Mapping and sequencing the human genome and model organisms are only the first steps in determining the function of various genetic units critical for gene regulation, DNA replication, chromatin packaging, chromosomal stability, and chromatid segregation. Such studies will require the ability to transfer and manipulate entire functional units into mammalian cells.

The development of extrachromosomal large-capacity cloning vectors for mammalian cells represents a powerful tool for functional genomic study.1 The UNCCH laboratory has developed the HAEC system to establish large DNA fragments as episomes in human cells, using the latent replication elements from the human herpes Epstein-Barr virus (EBV). A first-generation episomal vector based on the latent origin of replication oriP and its transactivator EBNA-1 from EBV was used to establish and maintain up to 350 kb of circular DNA in human cells.2-3 Such a system allowed the generation of a random clone library covering 10% to 20% of the human genome as extrachromosomal self-replicating episomes in human cells.

Current methods of transferring DNA into mammalian cells are particularly laborious and ineffective. To improve delivery of large genomic inserts into human cells, Vos's team developed a second-generation HAEC system based on engineering EBV.4-6 Such a mini-EBV vector carries the minimal cis elements for replicating and packaging the cloned DNA into infectious virus particles. In combination with a helper cell line, it was shown that up to 180 kb of non-EBV DNA can be packaged and delivered into target human cells as self-replicating HAECs.

Functional gene delivery by this mini-EBV HAEC system was demonstrated by introducing specific human genes coding for the hypoxanthine guanine phosphorybosyl transferase and Fanconi's anemia (FA) group C in lymphoblastoid cells from people suffering from Lesh-Nyhan and FA, respectively. In vitro correction of these defects supported the potential therapeutic usefulness of such a virus-based HAEC delivery system. By analogy to the bacteriophage lambda cosmid packaging system, the mini-EBV/HAEC technology represents an alternative to current DNA transfection techniques for studying and manipulating (within human cells) large human and model-organism chromosomal sections as nonintegrated and easily recoverable DNA clones.

The stable shuttling of any large human insert previously cloned in bacteria or yeast would be beneficial to future functional genomic analysis and interpretation of sequencing data. To establish specific clones isolated from current YAC and BAC-PAC large-insert libraries in human cells, the HAEC technology was upgraded recently by engineering the third-generation hybrid BAC-HAEC and retrofitting 'POP-IN' vectors, respectively.7-8 The POP-IN HAEC system converts any bacterial-based P1, BAC, or PAC into a HAEC by cre-loxP recombination, while the hybrid BAC HAEC system is based on subcloning large YAC inserts into the BAC-HAEC vector. Such vectors can shuttle DNA inserts of up to 260 kb from bacteria or yeast into human cells using bacteria as an intermediate host. In addition, the extrachromosomal HAEC DNA isolated from human cells can be easily transferred back into bacteria for further analysis.

The usefulness of these systems was illustrated by successfully shuttling expressed human housekeeping and tissue-specific genes as 100- to 200-kb genomic HAECs into human cells. Current efforts are focused on upgrading these respective technologies for high-throughput functional assaying in human cells of large human chromosomal regions.

The development of a technology analogous to the HAEC system for mouse cells, which are widely used as a surrogate system for studying human diseases, would enable additional levels of manipulation and analysis of the human genome. Mouse cells, however, have proven nonpermissive to the oriP EBNA-1 system. Through a strategy based on the transfer of HAECs as minichromosomes using microcell fusion, circular molecules carrying 100 to 200 kb of human DNA were stably shuttled into the mouse cells at low copy number9. The establishment of a first-generation mouse artificial episomal chromosomes (MAEC) system will facilitate future functional and evolutionary studies of the human genome in a murine genetic background. Current efforts focus on generating MAEC-based transgenic mice for future developmental, genetic, and therapeutic studies.