Germ of a new food microbiology

 Think of a food microbiologist you know.  I’m guessing that you now have in mind someone wearing a white laboratory coat, surrounded by food samples, Petri dishes and agar slopes, using an inoculating loop and making smears on microscope slides.  This might be accurate in most cases.

In the 80s, I read an article entitled “Germ of a New Food Microbiology”.  The author’s argument was that the so-called “Standard Plate Count” gives us less information on food safety than just about any other analysis.  There is no indication of how the microbial population will perform in the hands of the consumer;  the SPC is anything but total - many of the bacteria in the sample may not be able to grow on the count medium or at the incubation conditions; single cells and clumps will both be counted as one cell; if we use selective media to count specific types of bacteria, perhaps pathogens, they may not grow if they have been stressed during processing or storage; sample preparation, incubation and transfer to selective media may involve several days, meaning that the count may be obtained only after a significant proportion of the shelf life is over.  We also get no information on how the consumer might react after eating the food

The situation hasn’t changed very much.  Any modern analytical technique still has to be able to correlate with the plate count because of the way food safety is measured..  

The development of rapid microbiological methods now has a long history.
In many cases, the development has involved reducing the number of steps in the process, automating manipulative procedures and reducing the scale of operations to reduce costs.

Many so-called ‘rapid methods’ will give a result within hours of being set up.  However, they may involve significant technician time.

Modern methods often involve molecular techniques, particularly the polymerase chain reaction (PCR) and sequencing.  PCR relies on the ability of DNA polymerase to replicate a portion of a DNA molecule, using specific primers that bind to complementary strands of the target DNA.  The laboratory process usually involves a thermal cycler.  The replication of regions between the primer binding sites results in an exponential amplification of the target, so that within a few hours, millions of copies are produced.  

Recent developments have enabled real-time detection of the products by means of fluorescent reporter molecules that bind to the amplified products.  The progress of the amplification can be followed by monitoring the increase in fluorescence as the number of cycles increases.  The larger the number of target molecules in the sample, the fewer numbers of cycles are required to reach a detection threshold, so the number of cycles required indicates the level of contamination of the sample.

Though the PCR technique can detect a single molecule of DNA, there are some hurdles to overcome.  If we are looking to detect numbers of bacteria in the range <3 cfu/g to 10^2 cfu/g  either large samples or enrichment of the samples is required.  However, if we want to retain the relationship between the initial numbers in the food sample and enumeration by PCR, enrichment cannot be used.  In addition, the food matrix itself may interfere with the replication process.   

Taking this a step further, we can now analyse the population in particular parts of the processing equipment, using small portable real-time sequencers that allow microbiome-based monitoring of surfaces within the plant.  In turn, this may enable the identification of sources of the contamination and allow timely intervention by suitable control measures.

In order to get away from assessment of food safety based on plate counts, we need to formulate microbiological reference criteria based on these new methods and introduce them to our microbiological food standards.

The material presented here represents a great simplification of the techniques involved.  If the article has inspired you to seek further information, you will find hundreds of explanatory articles on the Internet - just search for RT-qPCR, and Microbiome-based environmental monitoring etc.

It's not CSI

Most of us have seen the odd episode, or at least been aware of the popular CSI-Miami or other incarnations of the crime drama TV series.  Those of us who are scientists are somewhat cynical about the ease with which samples from crime scenes can be analysed, seemingly within hours, by whizzy machines in gleaming laboratories.

In most cases, this is far from the truth; analysis of DNA samples, for example, requires painstaking care during the collection and processing of the material.  You only have to see the real world courtroom challenges to forensic laboratory evidence to realise that the whole process is much more complicated than the TV programmes would have us believe.

However, the detection of pathogenic bacteria in foods, or even the deliberate adulteration of beef burgers with other meats, is now benefiting from these molecular techniques.

The ability to sequence whole genomes of bacteria, coupled with the cheap synthesis of primers (probes) that will bind to specific parts of the bacterial DNA has enabled us to test for the presence of pathogenic bacteria in food samples. 

In principle, we extract the bacterial DNA and add our probes to bind to unique sequences in the DNA.  If the probe binds to the DNA, then we can use the Polymerase Chain Reaction** to amplify that piece of DNA and then detect it, separating it on a gel to produce a pattern of bands similar to what we see being examined on CSI.  If there is no binding, no amplification occurs and no detection, so the bacteria are absent from the food.

In practice, it's a bit more complicated and time-consuming.  We normally have to 'selectively enrich' our sample to increase the number of bacteria to a level at which we can detect them.  We do this by adding the food to a culture medium that inhibits most other bacteria and encourages our target bacteria to grow.  The actual preparation of the media, weighing the sample and putting them together takes only a couple of hours.  However, we need to incubate the mixture for up to 48 hours under controlled conditions before we can run the PCR.

Over the past few weeks, I have been working with my research assistant, testing a new PCR-based method of detecting Listeria.  It looks as though the method will be quicker and easier than existing methods and we'll publish our results in the near future.

Manufacturers continue to develop new rapid methods, many based on DNA and using automated equipment, making the detection of pathogens in food easier and quicker, allowing products to be checked and released to the market earlier.  These methods can also be used to track down sources of contamination, such as in the European E. coli O104:H4 outbreak of 2011.

**  For those readers keen to know more about PCR, I'll post a more complete description of the technique, trying to keep it relatively simple.