Optimising Protein Expression

Once you have finished making your recombinant baculoviruses expression vector you naturally want to maximize the amount of protein made in insect cells.  A point we have emphasized in a previous blog is the importance of determining the infectious titre of your virus stock accurately.  It is also prudent that you produce a stock of virus large enough that you can conduct a series of experiments to optimise protein production without the need to switch to a new one.  This minimises variation between experiments.  Virus stocks may be stored at 4°C or -70/80°C depending on your requirements.

The reason why we continually deliver this message about virus stocks and their infectious titre is that knowledge of this value is vital if you are to add the correct amount of virus to a cell culture to obtain efficient recombinant protein production.  Adding too much virus is not usually a problem (although see below), but adding too little can result in very poor protein yields.  This is because when adding virus inoculums you want to infect all cells simultaneously to synchronize protein expression.  If not all cells are infected soon after adding virus to the culture then the ones that escape infection continue to multiply.  The uninfected cell density increases and the culture approaches a stationary phase where virus replication is inhibited.  The result is that only a small proportion of your cells produce protein.  It also means that your protein is synthesised in a biphasic process.  This can affect the quality of your protein, particularly if it is prone to instability.

There are no rigid guidelines for the amount of virus to add to cells.  This value is often referred to as the multiplicity of infection (moi) and describes the added plaque forming units (pfu)/cell.  It can range between 0.1 and 10 pfu/cell.  However, it is a good idea to keep the difference between inoculums constant.  For example, using 1, 5 and 10pfu/cell might sound logical but if you think about it the fold difference between doses is 5 and 2 respectively.  If you use 1, 3 and 9 pfu/cell then the difference between doses is 3 and 3.  For a finer optimisation you could use 1, 2, 4 and 8 (work out the differences yourself!).  Beyond an moi of 10 pfu/cell you generally won’t see any increase in the number of cells that become virus-infected.

A further complication is that the amount of virus to add to a cell culture to obtain optimal protein production seems to vary between different genes.  It can also differ between cell lines.  Consequently, there is no substitute for spending a little time at the appropriate point in a project determining the optimal moi to use for your particular gene/virus/cell line.

While adding more virus than is necessary may not seem like a problem, it can potentially be detrimental to the growth of suspension cell cultures.  Unless you purify your virus stocks, you are essentially adding old/spent medium to your exponentially growing cells.  Although this might seem a trivial issue, if your virus stocks have low infectious titres you will be adding quite large volumes that may have a slightly detrimental effect on the growth of your culture.

When optimising recombinant protein synthesis in baculovirus-infected cells, you may perform many of the initial operations in small monolayer cultures of insect cells.  This can be very efficient of materials, but beware that the results you obtain in this format may not translate directly to a shake/suspension culture.  Therefore, if you do use monolayer cultures for your first check on protein production, follow up with tests in small scale (20ml) suspension cultures if that is to be your chosen format for larger scale production.  This may sound a bit cumbersome, but it will pay off in the long run.

Finally, a brief word about the best time to harvest recombinant protein from your virus-infected cell cultures.  Again this can vary depending on your particular protein.  Once you have tested the best moi to use, follow up with a time course to monitor when you should harvest your virus-infected cell culture.  Don’t just assume that 72 h p.i. is going to work for all proteins.  You may get a high yield, but if protein degradation has set in then the quality of the target may suffer.  This is where using either flashBAC GOLD or flashBAC ULTRA can help.  Both these variants of our technology lack the baculovirus-encoded cathepsin gene, which encodes a cysteine protease that can degrade recombinant proteins in the latter stages of infection.

Next time we will discuss the various options for scaling up recombinant protein production in virus-infected insect cells, before moving onto protein purification methods.