https://www.euro-online.org/enog/inoc2007/Papers/mac-slots.html https://www.euro-online.org/enog/inoc2007/Papers/m https://www.euro-online.org/enog/inoc2007/Papers/mac-slots.html

Has this growth media gone off?

Determining whether food has gone off or not can usually be done with a quick sniff or taste, but it's a bit trickier if that food is growth media for microbes. Now, however, German and Portuguese scientists have shown that NIR spectroscopy can offer an effective way to determine whether or not microbial growth media has gone past its 'best before' date.

Microbial growth media consists of a mixture of amino acids, vitamins, sugars and inorganic salts, and its precise composition can have a major effect on the growth rate of the microbes. Maximising this growth rate is particularly important if the microbes are being grown on an industrial scale to produce some of kind of biochemical, such as a biopharmaceutical.

Unfortunately, as with all food, the composition of the growth media tends to change over time, as a result of chemical reactions between the different components. So biopharmaceutical manufacturers could do with an objective method to determine the age of their growth media, in order to determine whether it's still good to feed microbes on.

Christian Hakemeyer at Roche Diagnostics in Penzberg, Germany, decided to see whether NIR spectroscopy and two-dimensional (2D) fluorescence spectroscopy could form the basis for such a method. Together with his colleagues, he used both techniques to analyse samples of growth media stored for up to 63 days. They then used several multivariate statistical techniques to see if there was any relationship between the spectral data and the age of the media.

As reported in the Biotechnology Journal, they found that sections of the spectral data produced by both techniques did indeed vary in line with the age of the growth media. When incorporated into a model, this spectral data could accurately predict the age of the media, even when the media was stored for longer than the samples used to produce the models (up to 84 days). What is more, models produced from NIR data turned out to be slightly more accurate than those produced from 2D fluorescence spectroscopy data.

Hakemeyer and his colleagues are now planning to use these techniques to help study exactly how the composition of the growth media changes over time.

Blog tags: