|Home | About | Journals | Submit | Contact Us | Français|
Human NK-like cell lines are difficult to transfect using standard mammalian expression vectors and conventional transfection protocols, but they are susceptible to retroviral transduction as a means to introduce cDNAs. Our lab has exploited this technique to study a number of receptors in human NK cell lines. The method utilizes a bicistronic retroviral vector that co-expresses either drug resistance or enhanced green fluorescent protein (EGFP) in parallel with the gene of interest. After a single infection with recombinant retrovirus, transduced NK cells can be sorted for expression of EGFP or the transduced cell surface marker. Alternatively, cells expressing the transduced cDNAs can be selected for by treatment with neomycin, puromycin or hygromycin. Using this method, the sorted/selected cells uniformly express the gene of interest and the expression is stable for many weeks of culture.
A number of transformed human natural killer (NK)-like cell lines have been adapted to culture and provide valuable models for studying NK cell function and signal transduction. Most of these cell lines lack expression of many normal NK cell surface receptors, especially killer cell Ig-like receptors (KIR), CD94/NKG2 heterodimers, and CD16. Therefore, it is very attractive to express these receptors in the transformed NK cell lines to examine molecular functions. Unfortunately, the available NK cell lines are highly resistant to transfection with standard mammalian expression vectors.
In the late 1990’s, several groups successfully expressed cDNAs in NK-like cell lines by retroviral transduction. Amphotropic retroviral transduction was first successfully used to introduce the IL-2 cDNA into the human NK-92 cell line (1). Cohen et al. subsequently introduced the ecotropic receptor into the human YTS cell line, which permitted susceptibility to transduction with mouse ecotropic retrovirus ((2) and further described in the chapter by Mandelboim in this volume). Amphotropic retroviruses can infect most mammalian cells, including human, and can therefore be biohazardous to laboratory personnel. On the other hand, ecotropic virus can only infect mouse and rat cells, and hence, working with ecotropic-sensitive YTS cells has the advantage of avoiding some biosafety issues. Nonetheless, many versions of replication-incompetent amphotropic retrovirus have been engineered, and these strains are not particularly dangerous if carefully handled under BSL2 biosafety conditions, which are achievable in most modern biology laboratories (see Note 1). Our lab has exploited amphotropic retroviral transduction to introduce a number of cDNAs into a variety of NK-like cell lines, including NK-92, NKL, NK3.3, and KHYG-1 ((3–7) and our optimized transduction protocol is detailed in this chapter (see Note 2).
Retroviral vectors derived from Moloney murine leukemia virus (MMLV) are the most widely used and allow the delivery of genes to dividing mammalian cells. The expression of a cloned gene of interest is strongly promoted by the flanking long-terminal repeat (LTR) elements within these vectors, and the vectors integrate into the cell’s chromosomes, thereby establishing long-term, stable protein expression after a single transduction procedure. Retroviral infection generates a polyclonal transduced population, since the distinct random chromosomal integration events occur in multiple clones. The polyclonal nature of the transduced population thereby dilutes potential bias that may be introduced by influences an integrated vector’s promoter on genes adjacent to the integration site when studying monoclonal transfected populations.
To allow purification of transduced cells that express the gene of interest, retroviral vectors may also encode selectable markers, such as neomycin-, puromycin- or hygromycin-resistance genes or a fluorescent marker, especially enhanced green fluorescence protein (EGFP). Standard mammalian expression vectors contain independent transcription units for the selectable marker and the gene of interest. A number of bicistronic retroviral vectors have been developed, however, that contain an internal ribosome entry site (IRES), which allows both the marker and the gene of interest to be expressed independently from a single transcript.
Our retroviral transduction of NK cell lines has utilized a system developed and made readily available by Dr. Garry Nolan (Stanford University, Stanford, CA). Detailed information about this system can be found at: www.stanford.edu/group/nolan/. This system utilizes the retroviral vector pBMN-IRES-EGFP, which co-expresses EGFP and the Phoenix-amphotropic packaging cell line. The Phoenix packaging lines encode three major retroviral elements: 1) pol, which functions as a reverse transcriptase, RNase H, and integrase, 2) gag, which is a large protein that is processed into viral matrix and core structures, and 3) the envelope (env) protein, which exists in the lipid layer and determines viral tropism. When Phoenix cells are transfected with the pBMN plasmid, these elements package a replication-incompetent retrovirus that is secreted into the culture medium and used to infect the NK cell lines.
Several different NK cell lines, including KHYG-1, NK-92, NK3.3 and NKL, can be cultured under the following conditions. Cultures should be replaced with freshly thawed stocks every 4–6 weeks to maintain biological uniformity that can drift upon long-term culture.
To assure the optimal viability of cell lines, they should be harvested from log phase cultures prior to freezing. NK cell lines grow on suspension, whereas Phoenix cells are adherent and easily detach from the tissue culture flask by gentle shaking.
The first step of retroviral transduction is to clone your gene of interest into the pBMN-IRES-EGFP vector (see Note 4) and use this engineered vector to prepare recombinant retrovirus by transfecting into the packaging cell line, Phoenix-ampho. The Phoenix-ampho cell line should be maintained at less than 80% confluence, and cultures should be replaced with freshly thawed stocks every 6–8 weeks (see Note 3).
This section describes the generation and manipulation of biohazardous retrovirus. Therefore, BSL2 biosafety procedures should be followed throughout the following steps. Upon finishing this section, incubate all disposed plasticware (including pipettes) under the UV light of a biohazard hood for at least one hour to destroy the retroviral contamination.
We would like to thank all previous members of the Campbell lab for establishing and optimizing this technique, Dr. Amanda Purdy for review of the manuscript, and Dr. Garry Nolan for reagents. Supported by National Institutes of Health grants R01-CA083859, R01-CA100226 (K.S.C.), T32-CA009035 (S.M.S.M.), and Centers of Research Excellence grant CA06927 (FCCC). The research was also supported in part by an appropriation from the Commonwealth of Pennsylvania. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.
1Importantly, constructs encoding potential oncogenes should be avoided when working with retroviral, adenoviral, or lentiviral expression systems.
2This is a general protocol for transducing human NK cell lines, but can be applied with little variation to other cell lines.
3An important feature of the Phoenix cell lines is high transfection efficiency using conventional transfection methods (e.g. including calcium phosphate or lipid-based techniques). In our hands, approximately 60–90% of Phoenix-ampho cells can be transiently transfected with Lipofectamine reagents, depending on the construct introduced.
4The Stbl2 bacterial cells are suitable for the cloning of unstable inserts such as LTR-containing retroviral sequences or direct repeats and for optimal performance, bacteria should be grown at 30°C.
5The transfection efficiency of Phoenix-ampho retroviral packaging cells depends on their health and growth status, which must be maintained by regular passage. If the cells are 100% confluent, transfection is very inefficient, so never let the cells reach confluence.
6The retroviral supernatant can be used immediately for transduction of target cells or kept on ice if used within several hours. Otherwise, retroviral supernatant may be frozen at −80°C, resulting in a minimal loss of viral titer.
7Our usual centrifugation period is 30 min, but increasing the centrifugation time up to 90 minutes can increase transduction efficiency in some cell lines.
8Depending on growth rate, cells are generally sorted 6–8 days after transduction. Transduction efficiency can depend upon the number of cells infected, the construct used and the NK cell line to be transduced. Starting with a higher number of cells usually requires relatively shorter times to be ready for sorting.
9We have successfully used this transduction protocol to express cDNAs in the following human NK-like cell lines at the indicated efficiency of transduction: KHYG-1 (20–50%), NK-92 (5–15%), NKL (15–30%), and NK3.3 (5–15%). KHYG-1 cells are highly susceptible to retroviral transduction and can be successfully transduced by adding polybrene (10 μg/ml) instead of using Lipofectamine reagents. To improve transduction efficiencies, more NK-92 or NK3.3 cells can be infected in steps 6–7 of section 3.2.2.
10Starting on day 2 after transduction with retrovirus containing an antibiotic-resistance gene, the transduced cells should be selected with antibiotics for 5 days.
11For testing any biological effect in transduced cells, the results should always be compared in cells from separate transduction procedures using the same construct. This will assure that the impact is not unique to the cells derived from a specific transduced population.