Rapid methods in microbiology. An opportunity for everyone in the company.

Until very recently, the question we were asking ourselves was what information we would be willing to sacrifice in order to obtain results more quickly. All the methods available on the market were admittedly fast, but also destructive. We could detect the presence of microorganisms relatively quickly, but we could no longer identify them.

This can be troublesome when determining the microbial load of a product, or when using these techniques for a sterility test. Unless duplicating the tests, to the chagrin of the financiers, (a first test for rapid counting and the second for identification in the event of contamination), these methods were not very interesting. Not to mention the reliability between the two replicates! If we add to this the very low levels of contamination that we are supposed to locate, the statistics would not be favorable to us and the results of two tests conducted jointly would surely be different.

What does "fast" mean in this context? An almost instant response as for the MALDI-TOF identification technique described below is not open to discussion. But what about non-destructive counting methods, which for the most part involve a growth phase of microorganisms. Are we ready to wait, certainly half the time, but still several days, before quenching our scientific thirst? However, we would like to leave less and less room for the subjectivity related to this science, whether it is during counting, the appearance of colonies, the interpretation of Gram colorations or the bends of colored indicators. This article is not intended to be limiting or exhaustive, its sole purpose is to share a long experience in the field of microbiology whether traditional or fast.

The regulatory point of view
Industrialists are still very cautious about the use of rapid methods, they feel little encouraged by the pharmacopoeia who insist on their use, but in still shy terms, proposing the introduction of these new technologies during proactive corrective intervention "or with a view to" significantly improving the quality of controls ". Some of them even offer a guide to help us choose an alternative method to supplement or replace conventional approaches.

More shared enthusiasm could only benefit the development of these rapid methods. Their comparability with the traditional techniques which we fully understand remains the rule, although we all acknowledge some superior performances with the new technologies. Whether it is for determining the presence or absence of microorganisms or for counting them, the essential verification criteria are aligned with those of our colleagues in chemistry (accuracy, fidelity, specificity, limit of detection and/or quantification, robustness, linearity and measurement interval). This represents an apparently insurmountable challenge for some of our fellow biologists, who are more or less resistant to change in this field, but one which is perfectly surmountable, in fact, if one thinks from both a qualitative and quantitative viewpoint of the benefits brought to us by the comfort of having a rapid, properly validated method, which also meets all the very precise regulatory requirements cited in the pharmacopoeias (European Pharmacopoeia: 5.6.1 Alternative methods of control of microbiological quality. - U.S.P. <1223> Validation of alternative microbiological methods.) We can also count on the skill, involvement, assistance and total support of the manufacturers of rapid microbiology equipment whose proactivity has brought us to the current situation in which we have microbiological results that are more reliable and more rapidly obtained than in the past.


The techniques

1. Counting methods
As regards qualitative methods for assessing the presence/absence of microorganisms, the difficulty of establishing a precise relationship between relative light units and the number of microorganisms present for a low number of microorganisms, makes them perfect candidates for validating alternative methods.
For quantitative counting methods, the companies that market rapid microbiology equipment prefer to stick closely to the pharmacopoeias to avoid seeing their methods classed as alternative. Consequently, ‘rapid’ methods are increasingly based on the detection, by physical and optical artefacts, of microcolonies that are made visible by technology long before they can be detected with the naked eye. In this last case, a more traditional validation, whose principle we all understand perfectly, and which facilitates the change, is applied. In the field of quantification, the reference method (probably wrongly, as the future will tell us) is still the traditional technique of counting colonies on agar by eye. Rapid methods enable the time required for the detection of microorganisms to be reduced by half compared with those that rely on the human eye.
Compared to the human eye, rapid methods to halve microorganisms detection time.

The use of traditional culture media (U.S.P./EP/JP), the absence of added reagents and effective counts in CFU (colony forming units) mean that this equipment can be or is automated.
Automation of this technology has recently enabled a company specialized in rapid microbiology to market a device based on this principle. It allows for the running of microorganism counting tests (or bioburden testing to check water or pharmaceutical solutions before sterilizing filtration) using suitable media (TSA, R2A, SDA etc.), tests to monitor the production environment (TSA with lecithin and Tween etc.) and sterility tests which are likely to be of interest to many parenteral production units which are subject to these regulatory controls.

These methods represent definite progress, as much from the point of view of quality and the reduction in storage costs, as speed to market. It also enables the testing of pharmaceutical products with a short shelf-life for which sterility testing with a minimum duration of 14 days is not practicable.

2. Identification methods
In this field too, besides the phenotypic methods, the choice remains relatively limited. The reference method (although debatable and controversial) is DNA sequencing ("gold standard"). This technique is considered to be the most reliable, the most accurate and the most reproducible, but it is undeniably the most expensive. Its implementation requires not only experienced analysts, but also specialists in phylogenetic classification for organisms with a non-reportable gross result, which in our industries represents no less than 20% of cases when monitoring production environments . To this will be added the important costs of materials, reagents and labor. Although in many cases, identification with the genus and species allows us to carry out in-depth investigations during production incidents, it often comes down to recommendations for thorough cleaning of the areas in which these microorganisms were found. This last point, a little luxury that can be offered in times of fat cows, is quickly challenged in the current context. Out of this little note of humor, it is certainly very important, for production environments, a fortiori parenteral, to know the flora present to the genus and the species in order to eradicate or at least control it. . Fast techniques allow us to achieve these results while making substantial savings compared to the reference method, without having to sacrifice important information for this application.
The installation of these identification methods is already well advanced as there are few regulatory restraints in this area, or at least no additional constraints relative to the phenotypic methods that we have used until now.
The proteotypic MALDI-TOF method (Matrix Assisted Laser Desorption/Ionization-Time Of Flight) is one of these methods which allows high-quality identifications to be obtained at lower cost.

The method uses mass spectrophotometry. It is extremely simple to implement, and only requires a few minutes from a trained operator and a few euros from the laboratory to obtain a protein spectrum which will be compared to a validated database.

A portion of the colony to be identified is placed on a plate then mixed with the reagent (matrix). The deposit (or spot) formed is called the target. A laser source is directed onto the target to ionize the molecules in the sample. The ions may be positively or negatively charged according to their nature. Proteins and peptides have proton acceptor groups and are positively ionized. The ions are subsequently detected by measuring the time taken by the different particles to reach the detector. The speed of each particle depends on the mass/charge ratio. The largest molecules take longer to reach the detector, whilst the smallest molecules reach it more quickly.

Once the ion reaches the detector, the signal is amplified and sent to a computer which processes the data and returns the results in the form of a spectrum. The recorded data are calculated in order to transpose the results into a spectrum in which each peak corresponds to a molecule type.

The equipment integrates the different peaks recorded and looks for the identification of the corresponding bacterium in the reference database of microorganisms.
It is at the level of this database that it is important to position oneself. Whatever the manufacturer of the mass spectrophotometer (they are 2 to share the market), it will be appropriate to choose this reference base of microorganisms (3 major industrial market them). Some of these marketed bases are oriented towards the "clinic" and are essentially solicited by hospital laboratories and biological analyzes to identify the microorganisms responsible for human pathologies. Other industry-based bases will enable pharmaceutical and other production units to accurately monitor the evolution of their flora in production environments. Attention will also be paid to the size of the reference microorganism base. It will be readily understood that the more microorganisms a reference library will have, the greater the chances of identifying an unknown microorganism. Whatever the technique used - genotypic, genotypic or proteotypic - the evolution of microorganisms and their classification being permanent, it will be necessary to regularly update its reference library (1x per year in view of the current evolution of this science) which of course leads to its revalidation. To perform this tedious work, one of the major laboratories specialized in this technique and having the largest reference base in the field of industry offers a comparison of the spectra obtained, not to an internal database in your laboratory, but to its own databases, regularly updated and validated, so as to dispense with this tedious work.

But whatever your choice of equipment and supplier, this rapid method will enable you to identify microorganisms not in 2 days, not even in 1 day as with the reference genotypic method, but in a few minutes, allowing you to improve your responsiveness when faced with a trend or an investigation relating to microorganisms.

3. The bacterial endotoxin test
Because a rapid variant of this test has been available for several years (result in 15 min.) which is time-saving relative to the traditional method (result in 1 h at best), and which moreover does not detect microorganisms but the residues produced at the moment of lysis (bacterial endotoxins), the LAL test is often forgotten when citing rapid methods.

Setting aside the question of horseshoe crab viability (*), the issues posed by this test remain the same as for strictly microbiological methods. Firstly, are we ready to use an alternative method (recombinant factor C) with its multitude of criteria to be validated (see §2.) or do we prefer to follow a more conservative approach and limit ourselves to verification by the LAL pharmacopoeia method (Limulus amoebocyte lysate test method and variants)?

Secondly, do we wish to install an automated or automatable method?

It should be known that there is currently a precalibrated cartridge system which uses an archived standard curve, does not require a calibration curve to be produced and can provide a result in less than 15 minutes.

This breakthrough can be used by laboratories that perform only a few tests. For the biggest "consumers", the question of the automation of the pharmacopoeial method or the alternative method will have to be evaluated on a case by case basis, knowing that the only automated method commercialized ("turnkey") is currently susceptible of result in a reduction in labor costs more or less equivalent to the increase in consumables costs. And this is what will lead us to the next point.

A changing profession
If the rapid methods such as these identification techniques are a godsend for the financial people, they nonetheless remain a source of apprehension.
Apprehension for investors, who currently have little perspective on this subject, as their use is still limited.
Apprehension for decision makers, who will be the scapegoats for the slightest teething problem with these techniques and the victims of their receptiveness to the techniques of the future.
Apprehension, especially for analysts, who are relieved of more than half of their working time, regardless of the mentioned method.
These methods are intended to "free up time for more rewarding activities" according to their promoters. And while it is true that these methods, most of them automatable, can handle routine and sometimes unrewarding tasks, a quick brainstorming exercise will be enough to identify the very real opportunities they offer to technicians. laboratory.
To fulfil our expectations, the equipment will need to be used by technicians who understand its functioning and its possibilities perfectly. This equipment will certainly bring its batch of new small problems to resolve. The profession will evolve towards posts which involve decision-making rather than execution. The more creative among us will use their knowledge to apply other existing laboratory methods to this new equipment (counts, AET tests, fertility tests etc.).
And it is free of suspicions of errors of dilutions or contaminations by the analyst, always easy but still too often put forward, free from the frustration of not being able to follow through the investigations resulting from their work by lack of time, that they will be able to project themselves into the future. No more raw data integrity issues either. This adaptation will require the participation of all actors (HR, management, technicians ...) in order to overcome this fear of innovation and to acquire the openness necessary to face the professional world of tomorrow.

Guy ROEHRIG NB

Guy ROEHRIG - LILLY

The replacement of rabbit pyrogen tests with the bacterial endotoxin test in the early 80 was Guy Roehrig's first encounter with alternative methods. After spending more than 30 years in the microbiology laboratory of Lilly France in the field of bacterial endotoxins and validations of methods, he uses his knowledge to demystify rapid microbiology and implement it in this laboratory.

guyr@lilly.com

References

(*) Atlantic States Marine Fisheries Commission Stock Assessments. http://www.asmfc.org/uploads/file//52a88db82013HSC_StockAssessmentUpdate.pdf
South Carolina Department of Natural Resources, Marine Resources Research Institute.

www.dnr.sc.gov/marine/mrri/SEAMAP/seamap.html