Firstly, I hope you all had a great Christmas break and are enjoying the beginning of 2020. As the last year of the FSP project starts, some conclusions start to arise from the lab work carried out in the previous 2 years. In my case I would like to share with you my thoughts on which biorecognition element to use in a biosensor that aims to detect microbiological contaminants. You might remember that my previous work revolved around the detection of Campylobacter jejuni using a DNA-based approach and Mycobacterium bovis using an antibody-based approach. Both systems had advantages that made us chose them as the preferred method, but also obstacles that we knew in advance or realised during the lab work that would need to be overcome. So let’s start unpacking the information for both approaches.
DNA-based techniques are described as sensitive, reliable and specific approaches for detection. It is interesting to note that they allow for microorganism subtyping as well, which is very important for identifying and distinguishing pathogenic bacteria. This way, if the food regulation states that no pathogenic E. coli can be present in 25 g of sample, our detection system must be able to distinguish them from other E. coli strains that are harmless and might be widely spread in the environment, thus avoiding false positives. Of course this all comes to the oligonucleotides specificity, the molecular tools that we must design to delimit what will be positive in our assay and what will be negative. It is important in this sense that one can identify a DNA sequence specific from the bacteria (or strain) that needs to be detected. Detecting the DNA sequence that encodes for virulence factors can be the answer sometimes for targeting only the pathogenic strains. Sometimes, however, amplifying and detecting the 16S ribosomal RNA is the preferred approach, as it contains highly conserved primer binding sites as well as hypervariable regions that can provide species-specific signature sequences useful for identification of certain bacteria. Thus, depending on what microorganism we want to detect and with what degree of specificity, we can decide what biorecognition element to design. This versatility together with the relatively low price for primers synthesis provides a clear advantage over other approaches: you can design 10 primer pairs that showed good results in silico, and then test them all experimentally to determine which is the optimum pair. However, these approaches also have some disadvantages that must be mentioned: such as being negatively affected by the complex structure of food or any other real matrix that needs to be tested. Very often this is caused by the presence of PCR inhibitors in the medium that is being analysed, representing a diverse group of substances with different properties and mechanisms of action, reason why is difficult to predict their presence.
All in all, DNA-based approaches account for a big part of the commercial kits available for detection of microbiological contamination in food. This includes real-time PCR and DNA hybridization, commercialised by well-known companies such as Bio-Rad or Neogen. The figure below summarizes the approach, which usually starts with the extraction of DNA by breaking the cell wall and often follows by amplifying the DNA to increase the signal generated.
On the other hand, we find systems based in the antibody-antigen specificity. The main advantages in this case are being rapid, robust and reliable techniques. However, the time needed for the assay is usually related to the sensitivity. This means that antibody-based approaches will find troubles to reach sensitivity limits as good as the ones that can be found with DNA approaches after an amplification step. For this reason, antibody-based systems often follow a pre-enrichment step, in which the bacteria are incubated for a variable period of time (usually between 12-48 hours) so it can replicate and the system reach the sensitivity limits commonly demanded by EU or otherwise regulations. Obviously this step increases enormously the assay duration, so the sensitivity needed for each situation governs the decision of including an enrichment step or not. In any case, this lack of sensitivity is one of the main drawbacks of antibody-based approaches, together with the apparent lack of specificity when compared to DNA-based assays. This, however, does not mean that antibodies are not specific enough by themselves, but instead it points out the difficulties that might be faced to develop an antibody able to recognise a surface analyte only present in a certain strain. And this is the second main obstacle that one might find when working with antibodies as biorecognition elements: the production of specific and sensitive antibodies is very time-consuming and offers very little room for manoeuvre once it is produced. This means that the goodness of your antibody-antigen pair will determine the potential of the assay. And in agreement with what was explained in a previous blog (see Where do antibodies come from?) there is little control in the antibody design beyond the specificity for the target antigen (in our case a whole bacterium!).
However, there is something that plays in antibodies favour: lateral flow tests. These are one of the main detection systems in which antibodies are used commercially. Due to their simplicity, they are ready to use by any non-expert and require no equipment. In the area of food pathogens, these systems have been widely commercialised by important companies such as Neogen, Romer Labs, Thermo Fisher, DuPont or Zeulab. LFD include antibodies already immobilized in a test line and coloured nanoparticles that generate a coloured line if the target analyte is captured. Thus the end-user only needs to soak the paper strip with the sample to be measured and wait for the reagents to flow along the strip.
So what is best for my assay? It all comes down to this and it is a valid question that might still remain unanswered after reading the above. So you might need to consider the following points, especially if you aim to detect microorganisms:
Does your assay need to be specific towards a certain bacteria strain and able to distinguish it from similar strains? If this is the case, it might be difficult to obtain an antibody with the desired characteristics. The process is likely to be time-consuming and there are no guarantees that the resulting assay will show no cross-reactivity between similar strains. In this situation, maybe you should consider following a DNA-based approach if that is a possibility that your assay enables.
However, if you have access to an antibody that suits your needs, this might become your preferred approach.
So it can be summarised that DNA approaches offer more versatility during the design and production of biorecognition elements, which can be done easily from your computer. By contrast you will need to break the cell wall to access the target DNA. On the other hand, antibodies production is much more complex, but if you manage to obtain a good one, its application in your assay will be much easier.
I hope this blog shed some light on what are the pros and cons of each approach and helps you to pick up side.
See you soon!