Suspension Culture Infections: High MOI

Return to Protocol Index


Infection of suspension cultures is the most common way of producing protein in the baculovirus system. Suspension cultures are easily scaleable from about 10 mL to more that 100 L. Generally, the suspension cultures are maintained in serum free media. These media are available from several commercial sources, all of which work well, though there is variation in effectiveness dependent upon cell line and protein.

It is probably best to test your protein and cell line in a variety of media and pick the one that best produces your protein in your cells. Give your cells a chance to adapt to various media before testing expression or the expression test may not be representative of results you would get after doing so.

Media Sources (see vendor page)

NOTE: You must know the normal growth parameters of your cell line in order to optimize expression; however, I will present some common values in order to facilitate getting started. My favorite Sf9 cell line can be split easily to 2E5 cells/mL and recovers almost without a lag. It will double in under 20 hours in log phase, and grow to about 5E6 cells/mL without too much trouble. These are the growth parameters that I will be assuming in my protocol. If your line behaves differently, you should take this into consideration and make appropriate adjustments.

Introduction to High MOI Infections

High MOI infections (Multiplicity Of Infection = ratio of infectious virus particles to cells) are the best strategy for producing protein. This is for two reasons. You are not concerned about DIPs (Defective Interfering Particles that represent partial genomes packaged by complementation from intact genomes coinfected in the same cell) because you are only interested in the protein that is expressed in the cells or medium. High MOI infections are most easily and reproducibly controlled to produce high levels of protein expression.

Low MOI infections can also be used to produce protein, but they are more difficult to control and less reliable for producing maximal amounts of protein. They do have the advantage, however, of requiring much less virus stock. This can be a great advantage for large scale production, provided they can be reliably reproduced. This seems to be dependent upon the particular recombinant virus.

In general, the strategy with High MOI infections is to synchronously infect all the cells in the culture with at least one virus particle. The probabilities of this process are generally described by the Poisson distribution, but the short answer is that if you want all the cells infected simultaneously, infect the culture at an MOI near 3. Since the cells are not going to grow anymore once you've added the virus you can precisely control the cell density at infection to hold around one half of maximum stationary phase density. Depending on the cell line this will be in the vicinity of 1.5E6 to 2.5E6%nbsp;cells/mL.

It is important that the cells not overgrow the culture or there won't be sufficient medium resources to dedicate to virus or protein production. On the other hand, it is important that there not be so much virus present that the cells are killed before they can expand into the culture and utilize all its resources. Using a high MOI infection strategy this level can be optimized fairly easily by infecting the cells at a fixed ratio of cells to virus in a series of cultures at densities from 1E6 to 3E6 cells/mL. Simply screen for the density that gives you the most product in the supernatant or in the cell lysate depending upon where you're directing the protein.

Execution of this strategy requires that you infect the cells with enough virus to stop growth immediately. This can be verified by doing an infectivity assay on smaller volumes using an optimal cell density and varying the amount of virus inoculum that is added. The biggest disadvantage to this method is that, especially at large scale, you have to generate substantial volumes of high titer virus stock. For example, if your virus fails to produce good titer stock and you're stuck with titers in the neighborhood of 3E7 pfu/ml, you need to add 20% of the volume of your culture in virus inoculum. This would translate to adding 2 L of virus stock to infect 8 L of cells at 2.5E6 to produce 10 L of infected cells at 2E6 cells/mL. Your cells would have to be capable of growing to pretty high density to still be comfortably in log phase at the 2.5E6 cells/mL density, and you'd be adding 20% of the volume in spent media to the final culture, which is the limit of what I'm comfortable doing.

If, however, you can get the volumes of high titer virus stocks that are required to infect your cells efficiently, the reproducibility of this approach from protein to protein is much higher as a rule. I would, therefore, definitely recommend this as the place to start when looking at new proteins, and I would establish this as the gold standard for the expression level that you seek in trying to develop a low MOI strategy at large scale. When you're dealing with small volume cultures (< 500 mL) this is the only way to go.


Split your cells to a density of 2-8E5 cells/mL in a shake flask or spinner with a total volume between 20% and 40% of the flask or spinner capacity, remembering to leave some room under the limit for adding virus stock. I personally prefer shake flasks for most of my small scale work (less than 500 mL). I culture them at 27±2°C with shaking between 100 and 130 rpm, usually 120 rpm.

Day 0 Incubate the cells until they reach a density of approximately 1.5-2.5E6 cells/mL, that is until about half of maximal density. This may require more than one day, depending on the seed density. Add sufficient virus to infect all the cells simultaneously, but avoid adding more than 20% of the final volume in virus stock. I try to keep it under 10%. The amount depends on the titer of your virus stock. If your cells are at 2E6 cells/mL and your virus stock is at 5E7 pfu/mL, an MOI of 3 would be 12% by volume (e.g. add 6 mL of virus stock to a 50 mL culture in 250 mL flask to give 56 mL at a density of 1.8E6 cells/mL). Return the flask to a 27°C shaker for 24 hours.

1 DPI (Day Post Infection): Count your cells and estimate their size. If the infection is indeed a high MOI infection, then all the cells will be infected. They will not have increased in number. They will be swelling considerably with diameters around 20;nbsp;µL [4200 fl (1 fl = 1 µL^3)]. This is the time it is easiest to see the roughened cell surface attributed to the process of budding viruses. If the cell number has increased more than a few percent, the virus stock titer was probably not as high as estimated, and the cells weren't simultaneously infected. This is actually not common at an MOI greater than 3, because the typical baculovirus plaque assay underestimates the number of infective viruses. If this does occur, but the cell number increases less than 25% or so, it will probably only delay the optimal harvest time. The protein produced per cell may be lower.

2 DPI: Again, count your cells and estimate their size. At this point there should be no further increase in cell number relative to 1 dpi. If there is, your infection is not well distributed, most likely due to low titer virus stock. Your cells should have swollen considerably by now to 21-22 µL (4500-6000 fl). The nuclei will occupy most of the volume of the cell. Depending upon the particular protein you are expressing, it may be time to harvest the cells or medium. Most proteins stop accumulating at between 48 and 72 hrs. This should be empirically determined.

3 DPI:Again, count your cells and estimate their size. By 3 dpi the cells had better be thoroughly infected, swollen and growth arrested, or there were not sufficient virus in the infection to stop the culture before it reached stationary phase. Harvest the culture, spin down the cells, and store the medium at 4°C protected from light or frozen at -70°C.

Usually it is best to store the pellets frozen. My favorite method is to suspend the cells in a very small volume (ca. 1/4 vol) of PBS or TBS containing protease inhibitors at about 5x the usual concentration. After the cell pellet is uniformly resuspended, the slurry is dripped directly into a clean Dewar flask full of liquid nitrogen, usually from a pipet. The drops of suspension freeze instantly upon hitting the nitrogen and form little pellets. These pellets can be poured and weighed almost like granulated material as long as they're kept very cold. They can be aliquoted into 50 mL conical tubes, simply stored in Revco freezer boxes, or whatever. When it is time to purify protein from the cells, just add them gradually to 3 to 5 volumes of rapidly stirring lysis buffer at room temperature. They will thaw uniformly and very quickly and drop the buffer temperature to near 0°C. Individual pellets can be easily transferred with forceps to eppendorf tubes and analyzed in mini-scale experiments.

Return to Protocol Index
webmaster: Jim Litts
© Copyright 1999, 2000 Jim Litts