Adjusting to a different culture

Most researchers desire a better protein yield from their microorganisms, but how can it be improved? Antje Neubauer urges scientists to reconsider their culture conditions

A recent survey of over 1200 European and American research scientists producing recombinant proteins from bacterial cultures frequently mentioned low yields, poor solubility and/or low activity as their major challenges – no different from those challenges apparently solved by their counterparts in production!

Using bacteria to express proteins may seem just a small, routine step – yet the ability to produce sufficient quantity and quality in research-scale cultures is a basic requirement that supports a vast number of research projects. Examples include expression of proteins for structure/function characterisation, for assay components, or production of biomolecules such as antibodies. However, history shows that while scientists have developed, and continue to discuss, methodologies for small- and large-scale production of mammalian cells, improvements for growing bacterial cultures have focused heavily only on large-scale production issues.

[caption id="attachment_34725" align="alignright" width="200" caption="Figure 1: Typical bacterial growth curve in conventional growth media"][/caption]

Figure 1 represents a typical bacterial growth curve. When supplied with a suitable carbon source such as glucose, bacteria initially grow and divide extremely rapidly at their maximum growth rate. However, as uncontrolled growth and fast aerobic metabolism begin to deplete oxygen and glucose levels and alter the pH of their surroundings, the bacteria switch over, within a matter of hours, to anaerobic metabolism. Consequently, not only are protein yields reduced due to uncontrolled nutrient supply, accumulation of harmful metabolites such as acetate, oxygen depletion and poor pH control, but the time window for induction of protein expression can be extremely short. The quality of proteins expressed and modified under sub-optimal metabolic conditions is also compromised – for example normal folding and other post-translational machinery, namely chaperones, foldases, phosphorylases or hydroxylases – may be unable to match the fast rate of synthesis. Insufficient levels of these accessory proteins, together with oxygen depletion, increase the risk of poor solubility or low activity of the end product.

In the world of large-scale bioproduction, precise control of bacterial growth has long been recognised as the key to successfully maximising the yield and quality of proteins from bacterial cultures. Bacteria are kept under a constant volumetric growth rate in large scale bioreactors, constantly stirred and monitored, among other parameters, for pH, oxygen and glucose. This enables bacteria such as E. coli to reach very high cell densities (up to OD=200) and to express proteins under optimal conditions over a longer time period thereby significantly increasing the final yield.

Clearly, equipment complexity – requiring feed pumps, control and monitoring units, a consistent oxygen supply and constant stirring – makes it impossible to use the same ‘large-scale’ approach for those wishing to routinely produce ‘research-scale’ amounts of high quality protein. Even before the days of genetic engineering, there was an ever-expanding range of growth media, each one offered with or without specific components such as phosphate buffers, organic acids and salts that could improve growth conditions. So called ‘high density’ growth media, including vitamins, trace metals and minerals such as potassium, magnesium, and calcium to serve as prosthetic groups, co-factors, or ligands, were developed to improve the solubility of the final recombinant protein product.

[caption id="attachment_34726" align="alignleft" width="200" caption="Figure 2: New approach to controlling growth to increase yield and improve functionality (EnPresso growth systems, BioSilta)"][/caption]

More recently, scientists have tended to spend considerable time and effort ‘re-engineering’ their expression host and constructs in the hope of changing growth or expression characteristics. Using the pET expression system with the T7 lac promoter ensures a high level of protein expression, but may generate a higher amount of insoluble protein. An alternative, the pBAD expression system, controls expression through a complex system that regulates specific carbon sources. However, although particularly beneficial for toxic proteins, general expression levels are compromised in exchange for improved solubility. Co-expression of selected amino acid transporters or helper proteins to increase protein yield are additional options that add further complexity. The introduction of auto-induction growth media designed for pET and other isopropyl-D-thiogalactopyranoside (IPTG)-inducible bacterial expression systems could be seen as a merger of these two approaches i.e. genetic manipulation and modification of growth media. In these media, protein expression is automatically induced when glucose levels have been depleted in the hope of maximising protein yield when cells have reached a maximum density and just before they change to anaerobic metabolism. Laboratory equipment manufacturers also offer practical solutions to maintain cells under optimal growing conditions ‘for just that bit longer’: conventional Erlenmeyer flasks are offered with baffled surfaces to improve mixing, studies are presented recommending optimal shaking speeds for baffled or non-baffled flasks or microplates, a range of closures, from silicone sponge closures or AirOTop seals for shake flasks to sandwich covers for microplates, ensure better oxygenation levels.

Yet the main tenet remains: at any scale, if bacteria are supplied with the key elements to support growth, control pH and regulate glucose availability, cells will grow steadily to reach higher densities before induction and continue in linear growth for many hours after induction.

[caption id="attachment_34727" align="alignright" width="200" caption="Figure 3a"][/caption]

Clearly, production scale processes demonstrate that control of glucose delivery (feeding the bacteria) is essential to enable cells to reach higher cell densities and thereby increase protein yields. But how can this be implemented in typical shake flask or microplate cultures?

Figure 2 illustrates the theory behind the use of new bacterial

[caption id="attachment_34730" align="alignright" width="200" caption="Figure 3 >30-fold increase in active enzymes, haloalkane dehalogenases(Fig.3a) and >3.3-fold increase in specific activity (Fig.3b). Specific activity determined with 1,2-dibromoethane (LinB, DhaA and DbjA) and 1-bromobutane (Spur). Comparison of an EnPresso B growth system with a commercially-available LB growth media. Results courtesy of Lukáš Chrást, Masaryk University, Czech Republic"][/caption]

growth systems that take a different approach to controlling the feeding and thereby growth, of bacterial cultures in the research lab. Put simply, the growth systems contain a glucose-releasing agent that breaks down long chains of polysaccharides, releasing glucose units to feed the bacteria. Control of glucose release controls the rate of growth keeping the bacteria in linear growth. Buffering components maintain an optimal pH. The systems have been optimised and used successfully in commonly used growth formats i.e. Erlenmeyer shake flasks and 24 deep well plates as well as Ultra Yield shake flasks.

These growth systems maintain bacteria under optimal growth conditions, cells remaining in linear growth while proteins are expressed and modified under optimal metabolic conditions. Results from independent laboratories (see Figure 3) indicate that, for many research scientists, simply changing from conventional growth media such as LB, TB or M9, would easily resolve the problems of low yield, poor solubility and low activity currently encountered. Over 50-fold Increases in yield have been reported – well above the two- or three-fold improvement that many scientists claim would solve their problems! Perhaps the significant role played by growth conditions in increasing the yield of high quality recombinant proteins in lab-scale cultures has been underestimated and ignored for too long?

Author: Dr Antje Neubauer, Customer Application Specialist, BioSilta AB