Protocols - Optimization

Troubleshooting and Optimization of Retroviral Transduction Parameters

One of the most frequent laments of using recombinant retroviruses is that the titer does not appear to be sufficient to infect the cells of interest at a high enough frequency to suit your purposes.

When you begin to use the Phoenix lines you must optimize your approach and transfections. There are a series of steps you should carry out before you despair and give up. This web site has many pages of hints and protocols to help you out. Follow the procedures outlined on this and other pages.

DO NOT contact our lab for information or to help us solve your problems if you have not read and tried ALL the associated optimization suggestions and protocols. Owing to the number of investigators who are using the lines and plasmids we have provided for transient retroviral approaches (several hundred) we obviously CAN NOT help you troubleshoot every problem you might encounter.

The purpose of this web site is to serve the community and we have spent a lot of effort setting it up. If you do have an intractable problem you may e-mail us and we will try to help you through e-mail as much as we can. If the problem is something we have not addressed in this web site we will add the solution, if there is one, to the site and credit your help.

To begin the standardization and optimization after you receive the Phoenix cells

Freeze down multiple aliquots of the cell lines.
You should use a control vector containing a marker gene such as lacZ or GFP to standardize your virus production.
Make HBS and check ability to form precipitates. Play with cell number, HBS precipitation time, plasmid concentrations and check transfection efficiency of the Phoenix cells. Phoenix cells should be dark blue if you use a lacZ vector such as MFG-lacZ and X-gal staining. Nearly 60% of the population should be transfected.
ALWAYS start by infecting NIH 3T3 cells.
They are the gold standard. Vary the time of application of the virus and the polybrene concentrations. Check transduction efficiency into 3T3 cells. When you can achieve nearly 100% infection of 5x10^5 3T3 cells on a 60mm petri dish using 1 ml or less of viral supernatent you have optimized the virus production conditions. While it is true that you might have found the near-best conditions, different inserts in different viral backbones or different target cells might require some modifications of the system and approach. You are on your own here, but good luck!

Protocols can be found at the following pages:

Helper-dependant protocol for transient retrovirus production. Helper-free protocol for transient retrovirus production.

Troubleshooting: A number of factors that can contribute to low efficiency of virus production are listed below. I expect that the fixes will be self-explanatory.

Phoenix cells might have lost their expression of gag-pol or env. Reselect Phoneix cells in 1 ug/ml diptheria toxin to increase Envelope expression or 500 ug/ml hygromycin to increase Gag-Pol expression. Two weeks of selection is sufficient for each selection; drug selection can be carried out simultaneously (One can also check for expression of Gag-Pol by the translationally-linked CD8 surface marker (IRES-linked) or for envelope expression using antibody to the retroviral envelope protein. Each marker should by FACS staining give a relatively tight peak of expression).
Phoenix cells might have become infected with Mycoplasma. Treat Phoenix cells with anti-mycobacterial agents ciprofloxacin for at least two weeks.
You might have low efficiency transfection of Phoenix cells. Density of Phoenix cells was too high or too low at the time of transfection. Too high a density and cells don't transfect well. Too low a density and cells do not survive transfection--cells need cross-feeding to survive transfection.
Chloroquine might not have been washed away properly, resulting in toxicity of cells. Chloroquine is toxic over the long term. Ensure you have washed cells sufficiently well.
Phoenix cells might have gone lost expression of the genes (GASP!). Did you freeze down multiple early passages as aliquots? Reselect cells or take out an early freeze and reselect again.
You didn't place the producer cells at 32 degrees centigrade for the last 24 hours of virus production. The virus half-life at 37 is much lower than at 32 degrees.

You are trying to infect hematopoietic cells with pBabe-derived vectors. We have experienced SIGNIFICANT problems trying to get high efficiency infection of B and T cells with the pBabe vectors.

I strongly recommend turning to and MFG-derived backbone. We suspect there is something in the packaging region of MFG that affords high level efficiency infection of hematopoietic cells. MFG and SFG were created in the Mulligan Lab.
In our hands the order of infectivity of vectors is MFG = MSCV > pBabe.
MSCV vectors seem to have trouble with certain IRES and double gene systems.

DNA for transfection might be bad, contaminated with RNA, or is not what you think it is. Check the purity of your DNA
Titre of virus might not be high enough because the insert is toxic or the gods are against you. Are you sure the cDNA is not toxic to cell growth?
There is a cryptic splice donor/acceptor somewhere in your vector and you are getting splice-out of your insert. Sometimes putting the gene of interest in the opposite orientation (with a promoter and polyA) can help this. This problem, by the way, is rare.
CaPO4 precipitation might not have worked. Make a new batch of HBS or try lipofectamine.
You didn't use polybrene or other polycationic substance to enhance infection.
You don't know how to calculate the titer properly or your X-gal or other reporter system for titer calculation is not working properly.

Target cells were not dividing at the time of infection (needed for efficient infection.) Ensure that the cells are in optimal growth media at time of infection.
Target cells might not have been healthy.
Target cells were not murine cells and you used an ecotropic envelope producer line.
Target cells ARE murine cells, but have a resident murine "wild-type retrovirus" expressing an envelope that blocks the receptor thereby preventing entry of new virus particles (interference).
Target cells are dead. Doh...
Target cells might have been killed by polybrene. This is especially a problem with some B cells and certain T cell lines. Lower the concentration of polybrene.
Target cells are murine T cell clones, which are sometimes particularly difficult to infect with murine retroviruses.
Target cells are EBV-derived human B cells, which also are sometimes difficult to infect efficiently.
Target cells are infected with Mycoplasma (we have seen that this can block infection).

Concentration of Infectious Virus:

New ways to find unaccessed virus. Retroviruses are titred generally by applying retrovirus-containing supernatant onto indicator cells, such as NIH3T3 cells, and then measuring the percentage of cells expressing phenotypic consequences of infection. The concentration of the virus is determined by multiplying the percentage of cells infected by the diltion factor involved, and taking into account the number of target cells available. Thus, a relative titre is obtained. If the retrovirus contains a reporter gene, such as lacZ, then infection, integration, and expression of the recombinant virus is measured by histological staining for lacZ expression or by flow cytometry (8, 34). In general, retroviral titres generated from even the best of producer cells do not exceed 10^7 per ml, unless concentrated by relatively expensive or exotic apparatus. The general thought has been that it is not possible to concentrate retrovirus by centrifugation or over sucrose gradients due to the lability of the retroviral envelope. Concentration has also thought to have been hindered by the production of "defective" virion particles, containing incomplete genomes or otherwise impaired core structures or envelope proteins that co-purify and compete with active retroviral particles.

Recently, two findings have emerged which challenge the view that retroviral titres are limited by concentration techniques or defective particle formation.

The first involves the biophysical consideration that a particle as large as a retrovirus will not move very far by brownian motion in liquid. Using pure physicodynamic considerations of movement in fluids it was postulated that in fact, the half-life for the time of an average retrovirus to move 3 mm in a theoretical standing solution would be 11.5 days-- a time much longer than the 1/2-life of the virus stability in solution (Palsson et al). Thus, although more viral supernatant can be added to cells, fluid dynamics would predict that much of that virus never comes in contact with cells to initiate the infection process. Therefore the effective "infective" volume around cells is quite small, and is rapidly depleted in an environment free of fluidic current flow. Of course, there is no such thing as a theoretical standing solution. One might try rocking the cells in virus supernatent, but experiments in our lab have shown this actually LOWERS the infectivity. However, if cells are grown or placed on a porous filter and retrovirus is allowed to move past cells by gradual gravitometric flow, one can effectively maintain a high concentration of virus around cells at all times. Experiments in our laboratory confirm this finding, as we observe up to ten-fold higher infectivity by infecting cells on a porous membrane and allowing retrovirus supernatent to slowly flow past them over a period of hours. Thus, it should be possible where we readily obtain viral titres of 10^7 by standard protocols to achieve titres of 10^9 per ml after concentration. This can alleviate the need to buy expensive concentrating devices or devise other more complicated means to generate high-titre virus.

Supporting this concept of “unaccessed” virus in retroviral supernatants was the recent finding that retroviruses could be effectively concentrated using calcium-mediated precipitation (63). Again, the concept was that much of the retrovirus in supernatant never accesses cells, so if one could ”precipitate” the virus onto cells from the supernatant (increase the local density of virus around cell membranes) then an in crease in the concentration of retrovirus and its infectivity might be obtained. Studies show this is in fact the case, and we have confirmed this in our labora tory. Interestingly the experiments by Morling and Russell (63) show that not only is it possible to precipitate the virus in this manner but that the host range of the retrovirus, whether ecotropic or amphotropic, is not altered. Also, it is possible to chelate away the calcium with EDTA, without affecting retrovirus infectivity, and obtain calcium-free retrovirus concentration by a simple series of centrifu gation-chelation steps (Morling et al). By concentrating virus in this manner we observe a remarkable 100-fold increase in effective titres on 3T3 cells.

Palsson, B., Clarke., M.F., Chuck, A.S.Y. Methods of increasing rates of infection by directing motion of vectors. Int. Pat. App., Pub. WO95/10619. April 20, 1995.

Morling, F.J. and Russell, S.J. Enhanced transduction efficiency of retroviral vectors coprecipitated with calcium phosphate. 1995. Gene Therapy. 2: 504-508

Spin Infection.

We and many others have found that "spinning" the virus onto cells can result in up to a ten-fold increase in effective titer of the virus. "Spin-fection" is achieved by placing up to 10^6 suspension cells into a 24 well plate and overlying with virus supernatent in polybrene. Plates are sealed and placed in a microtiter rotor and spun at 1800 rpm for up to 45 minutes at room temperature. Fresh virus can be applied and the cells spun again for 45 minutes.

For adherant cells the same protocol is followed, usually in 6-well plates.

Obviously, spinning cells at 1800 rpm for 45 minutes is not enought to sediment free virus. It is thought that virus on membrane fragments is spun onto cells in a manner which effects greater infection. All we know is that it works and I strongly recommend you try it to increase infection.

Other suggested reading:

Makino M; Ishikawa G; Yamaguchi K; Okada Y; Watanabe K; Sasaki-Iwaki Y; Manabe S; Honda M; Komuro K. Concentration of live retrovirus with a regenerated cellulose hollow fiber, BMM. Archives of Virology, 1994, 139(1-2):87-96. (UI: 95126787)

Abstract. A concentrated live retrovirus is required for in vitro experiments. A cuprammonium-regenerated cellulose hollow fiber, termed BMM, originally developed for biohazardous viral removal, was used to concentrate two different retroviruses, an ecotropic murine leukemia virus (MuLV) and human immunodeficiency virus (HIV). The BMM was useful for concentrating live virus suspension 10- to 30-fold from 500-1000 ml of culture supernatant. The ecotropic MuLV concentrated by BMM was demonstrated to be viable and biologically intact by XC plaque-forming assay and reverse transcriptase assay. The concentrated MuLV reached a much higher titer in the spleen in mice than the original one. The virus concentration assessed by p24 antigen for HIV was clearly higher than that of the original culture supernatant of HIV-infected cell lines. Since BMM hollow fibers trapped viruses by the sieving mechanism but not by adsorption, the viral particles were recovered by washing and the total live virus recovery rate was high, about 50%. Furthermore 60 min sufficed to handle 1000 ml of supernatant in the case of a filtration area of 0.03 m2. These results show that the BMM provides us with a rapid, safe and efficient method for concentrating live retroviruses.

Morgan JR; LeDoux JM; Snow RG; Tompkins RG; Yarmush ML. Retrovirus infection: effect of time and target cell number. Journal of Virology, 1995 Nov, 69(11):6994-7000. (UI: 96013801)

Abstract. Using a model amphotropic recombinant retrovirus encoding the Escherichia coli lacZ gene and quantitative assays to measure virus infection, we have determined the effects of time and target cell number on infectivity. Infection of various numbers of NIH 3T3 fibroblasts showed that the extent of lacZ virus infection was dependent on virus concentration and independent of target cell number. These results demonstrate that multiplicity of infection is not an accurate predictor of the efficiency of retroviral infection. Varying the time of viral infection revealed that maximal infection occurred after greater than 24 h of exposure of the cells to the lacZ virus. Half-maximal infection occurred after 5 h of exposure. After 2 h of adsorption at 37 degrees C, the majority of infectious virus was not adsorbed to cells but was unbound and able to infect other cells. These results are discussed in terms of both their relevance to the fundamental biology of retrovirus infection and the use of recombinant retroviruses for retrovirus-mediated gene transfer with purposes of gene therapy.

Related links: