A rapid and efficient method for concentration of small volumes of retroviral supernatant

Many current gene therapy protocols use retroviruses as the vector to package and transfer genes to target cells. Cocultivation of target cells on a layer of irradiated retrovirus producing cells is an efficient method for exposure of target cells to high titers of virus. However, this technique has several adverse features, most notably the resulting contamination of target cells with virus producing cells, and the difficulty of carrying out quality control procedures on the virus stock in the time frame of a transduction procedure. Retroviral transduction, therefore, is usually carried out by culturing the target cells in the presence of cell free (sterile filtered) media collected over a period of time from virus producing cells. This supernatant can be stored for relatively long periods of time (>6 months) at –80 C with little loss in titer. The storage of viable virus allows exhaustive quality control of virus batches to be carried out before its use in transduction experiments. This is of obvious benefit in clinical application of retroviral gene transfer technology. The titer resulting from such a harvest can range from 10 2 to 106, depending on the type of producing cells and the structure of the retrovirus itself. Efficient retroviral transduction of target cells using viral supernatant can be difficult to accomplish. This is largely due to the relatively short half life of retrovirus particles at 37 C. A technique to overcome this problem is to add a great excess of virus particles to target cells several times in the transduction period. A common method is to add 10 viral particles for every target cell, repeating the addition once or twice a day. Unfortunately, the protocol of adding large numbers of virus to a small volume of cells means that the volume will be increased quickly. One problem with increasing the volume is that the efficiency of infection is reduced as the volume surrounding a small number of cells increases. Also, the optimal harvest medium for the virus is rarely the optimal growth medium for the target cells, and adding a significant amount of harvest medium may affect the growth of the target cells. We decided that increasing viral titer though concentration of viable virus would be the most obvious method of retaining the advantages of multiple infections, while reducing the disadvantages. Adding a very small volume of virus a number of times will have relatively little effect on the volume of the target cell culture, and therefore will not adversely effect the concentration of important growth factors within that culture or the efficiency of the virus finding and infecting a target cell. Concentrating has the added advantage that low titer viruses that are desirable for possessing other characteristics will also be usable. Concentration methods for retrovirus have been published previously. Paul et al. (1) reported in 1993 a method of concentrating a large volume of viral supernatant utilizing tangential flow hollow fiber filters with a molecular weight cut off of 500 000. We have attempted to modify this method for a small volume of supernatant, utilizing CentriCell 60 units (Polyscience product #19182). We had little success using these units as they quickly became blocked by protein contained in harvest media, and therefore the volume could not be reduced as far as desired. We also discovered that virus harvested in standard media was incompatible for concentrating, as some factor (we believe the phenol red) also concentrated and would cause bone marrow cells to lose viability if added. We have developed a successful retroviral concentration method utilizing Amicon ultrafiltration membranes with a 100 000 molecular weight cut off (YM100). The filter is prewashed with 100 ml deionized water and 50 ml phosphate buffered saline. Virus is harvested in DMEM media without phenol red (Gibco #13000-021), pooled and prefiltered through a 0.45 μm filter to remove any cells and debris. The filtrate is then added to an amicon stirred cell system (series 8000) with a YM100 membrane, and concentrated under nitrogen pressure (500 kpa) with very gentle stirring until the desired volume is reached. The concentration system is held at 4 C to minimize loss of viability of retrovirus. The final volume is normally 2 ml from a starting volume of 100–150 ml. The concentration time is normally 2.5 h.