Team:NAWI Graz/design
The Bacterium Paenibacillus larvae (P.larvae) is the causative agent of American foulbrood (AFB), which is currently the most destructive bacterial disease in honeybees. In europe, the disease outbreaks started to escalate rapidly and hence became one of highest priority problems to solve for the beekeeper community. Additionally, prohibition of antibiotic treatment in Europe makes dealing with the disease even harder[1]. Interestingly, studies in 2014 showed that P.larvae acquired tetracycline-resistance via native plasmids, later followed with tylosin and lincomycin resistance[2][3]. Therefore, in America, alternative treatment methods are needed. We have noticed the importance of this problem, since it doesn’t affect only Europe but the whole world.
We considered different approaches and concluded that beekeepers need a quick and sensitive on-field AFB sensor in order to screen for ABF more often. The beekeepers could use these sensitive measuring devices as a prevention screening meeting to prevent the AFB outbreaks[4]. Beekeepers could use these sensitive devices for P. larvae screening, thus preventing AFB outbreaks in their hives. By doing so, they will not face quarantine and honey production loss.
Currently available immunobased field-on methods are not sufficiently specific nor sensitive to be used for prevention of AFB outbreaks. Laboratory diagnostic procedures are sensitive, but it can take weeks to obtain reliable results. Moreover, immunoassays and PCR do not distinguish between viable and nonviable bacterial cells, which can potentially lead to false-positive results and consequently inaccurate estimation of bacteria concentration in samples.
Biosensors require a specific recognition element and initially we focused on already existing antibodies against P. larvae. The specificitly of these antibodies was not very high, therefore we searched for an alternative, which brought us to bacteriophages.
The impedance spectra of various screen-printed electrodes can be used to determine surface properties at the electrode double layer. To bind Paenibacillus larvae to electrode surface and increase impedance, electrode surface was functionalized and treated with different compounds. The general process for achieving this is applied to every electrode and shall be shortly described here as well as some theoretical background.[5][6]
After we defined the goal of the project - to develop a cheap and fast diagnostic test for American foulbrood - we prepared a list of desired properties of such a diagnostic device. In many cases, specifications were defined only after consulting literature data or interviews with experts and beekeepers. Based on specifications, a draft version of the design was prepared.
Understanding the binding mechanism
Commission Regulation (EU) No 37/2010 of 22 December 2009 on pharmacologically active substances and their classification regarding maximum residue limits in foodstuffs of animal origin A lippi, Leon, and Lopez, Tetracycline-Resistance Encoding Plasmids from Paenibacillus Larvae, the Causal Agent of American Foulbrood Disease, Isolated from Commercial Honeys Krongdang et al., Multilocus Sequence Typing, Biochemical and Antibiotic Resistance Characterizations Reveal Diversity of North American Strains of the Honey Bee Pathogen Paenibacillus Larvae Gende et al., Searching for an American Foulbrood Early Detection Threshold by the Determination of Paenibacillus Larvae Spore Load in Worker Honey Bees Tlili, C. et al. Bacteria Screening, Viability, And Confirmation Assays Using Bacteriophage-Impedimetric/Loop-Mediated Isothermal Ampli fi cation Dual-Response Biosensors. (2013). Boonkaew, S. et al. An origami paper-based electrochemical immunoassay for the C-reactive protein using a screen-printed carbon electrode modified with graphene and gold nanoparticles. Microchim. Acta 186, 1–10 (2019). T. S. Brady, B. D. Merrill, J. A. Hilton, A. M. Payne, M. B. Stephenson, and S. Hope, ‘Bacteriophages as an alternative to conventional antibiotic use for the prevention or treatment of Paenibacillus larvae in honeybee hives’, J. Invertebr. Pathol., vol. 150, pp. 94–100, Nov. 2017. I. Takeuchi, K. Osada, A. H. Azam, H. Asakawa, K. Miyanaga, and Y. Tanji, ‘The Presence of Two Receptor-Binding Proteins Contributes to the Wide Host Range of Staphylococcal Twort-Like Phages’, Appl. Environ. Microbiol., vol. 82, no. 19, pp. 5763–5774, Oct. 2016. D. J. Simpson, J. C. Sacher, and C. M. Szymanski, ‘Development of an Assay for the Identification of Receptor Binding Proteins from Bacteriophages’, Viruses, vol. 8, no. 1, Jan. 2016. Bello-Gil, D. et al. Specific and Reversible Immobilization of Proteins Tagged to the Affinity Polypeptide C-LytA on Functionalized Graphite Electrodes. PLoS One 9, (2014).
Problem
To design an impactful project, we found it important to understand the beekeepers perspective of the problem. It helped us a lot in finding a realistic solution for a detection system.
Through extensive research and discussions with beekeeper experts, we identified that not only the treatment methods are very extreme (burning infected beehives) but the beekeepers are also obligated to report the outbreak to the beekeeper center. As a result, his/her bee garden and a 3 km (1.8 mi) surrounding area will be under a quarantine for a couple of months which results in a massive loss of honey production for the beekeeper.
Lack of robust, fast, sensitive and quick methods demotivates beekeepers to check for the presence of P. larvae in their hives.
Solution: Beeosensor
1. Selection of the Detection Method
In a search of a low cost, high sensitivity and user friendly detection method, we came across electrochemical biosensors. Among them, EIS-based (electrochemical impendance spectroscopy) sensors represent a powerful tool for performing high sensitivity measurements. The general approach can be extended to develop biosensor for a wide variety of bacteria, and the methodology can easily be applied in a relatively short period of time at low costs, with high sensitivity and specificity.
2. Paenibacillus Bacteriophage as Biorecognition Element
Compared to other biological agents that provide similar specific interaction for pathogen capture/detection (e.g., antibodies), phages offer the advantage of having a longer shelf life and being cheaper to be produced than antibodies. Most lytic phages also offer the dual functionality of host capture and destruction. This could be used in the long run in combination with our detection device. Compared to other bactericidal agents (e.g. antibiotics, antimicrobial peptides, silver nanoparticles, chemicals, etc.), phages offer specificity beside other advantages.
In order to prepare Paenibacillus larvae bacteriophage for biosensor application, our first task was to find an appropriate phage strain, to amplify and purify it in sufficient amount. HB10c2 paenibacillus bacteriophage (from Germany) was chosen as our biorecognition bacteriophage because it was the most practical for us to get it delievered to us. This demanded that we optimize bacteriophage enrichment protocols and its purification steps. The recommended bacteriophage titer for the EIS sensor is 1010 plaque forming units/ml[5]. The purity of the bacteriophage samples plays also a major role since the impurities such as proteins would decrease the sensitivity of our sensor.
3. Electrode-Bacteriophage-Spore Binding
The functionalization process [5][6] is described below.
As the first segment of our assembly, which should end with Paenibacillus larvae, L-Cysteine was used. This amino acid has two functional groups we can utilize. The thiol group binds strongly to gold-covered electrode surface, while the carboxyl group can be attacked by a nucleophile on the phage surface.
After the L-Cysteine has formed a self-assembly monolayer (SAM) on our electrode, we treat our electrode with a mixture of EDC/NHS crosslinker. The crosslinker is used to bind to our amino groups of our HB10c2 bacteriophage, which then forms our second chain segment.
Electrode surface that eventually remained free of phages could efficiently bind bacteria or reactive chemical species, thereby falsifying results. To prevent this, the electrodes were treated with BSA (Bovine serum protein) protein, thereby blocking any remaining active sites on the electrode.
To test the functionality of the biosensor, Paenibacillus larvae is applie. When bacteria bind to the immobilized cognate phage, charge transfer resistance increases substantially.
4. Device Development
Our design is based on the assumption that the concentration of Paenibacillus larvae spores correlates logaritmically with the charge transfer resistance of our electrode system. We decided to use the method of electrochemical impedance spectroscopy (EIS) for the measurement of the charge transfer resistance, because the studies on which we based our research upon used the same method with great success. We did literature research to find all possible alternatives with the following criteria in mind: accuracy, flexibility and price. Two different approaches were identified: buying a complete device for EIS measurement or building the whole device on our own with chemical ICs (Integrated circuits). The first alternative has less risk but can’t meet the crucial requirement of low price. So after comparing different ICs, we decided to use ADuCM355 on-chip system of our own circuit board. It is a cheap and versatile chip on which many companies rely on.
See more information about the design process on our development page
5. Further approaches
The core element of Beeosensor is the specific binding of the phage to Paenibaciullus larvae. This binding has not been a subject of detailed studies yet and almost nothing is known, not even the gene sequence of the receptor binding protein leave alone the receptor or kinetics.
However, it is known that different Paenibacillus phages bind different P. larvae strains very differently[7], although they are genetically very similar. The RBP of Siphiviridae is located at the tip of the phages, embedded in a structure called the baseplate. Potentially the structure the baseplate is responsible for this variety of binding characteristics. Maybe there are even more than one RBP in different variants of Peanibacillus phages, as it is the case with some other phages who bind differently to substrains of their host bacterium[8].
For a complete understanding of the topic and in order to be able to fulfill the beekeepers high demands for reliability and accuracy (see our survey) experiments towards identification and characterization of the receptor binding protein on the phage surface would be needed.
In order to identify the RBP we used a combined in-silico and experimental approach. The in-silico approach was meant to identify the RBP on the published but not further examined genome sequence without need to do more complex experiments designed to identify phage RBPs. You can find more on our modeling page.
If the in-silico approach would not have resulted in sufficiently satisfying results, we would have chosen the full experimental approach of Simpson et al. 2016.[9] With a candidate protein we conducted a part of the assay (data not shown).
C-LytAm7
In searching for different ways how to extend our scope of our project, we searched to ways how to express, purifiy and immobilize antibodies, proteins to the electrode surface on the most efficient way. The C-LytAm7 polypeptide tag was the perfect candidate for that kind of purpose. It can be washed off the surface where it is bound (e.g. column) with choline. The C-LytA method presents some advantages over other affinity tags, such as a higher buffer compatibility, non-interference with metalloproteins and a complete reversibility of binding[10]. This could allow for an easier cleaning process of the reusable electrodes.
Experiments with the tag and the electrode to determine weaknesses of our electrode are under way. Likewise, the experiments with producing and purifying the protein for further tests with the C-Lyt polypeptide tag are being planned.
Usage of this tag for our project will be the topic of the further research.
Cell free system
The classical way of producing phages is to grow their host bacteria and add vital phages which will then infect the bacteria and multiply. This is easy for some phages but can be challenging for others and it might require more complex or time-consuming. To eliminate this difficulty we apply a new method to our phage production - the cell free system. Phage-DNA, lysed cells and a special buffer with many additional ingredients are mixed and phages will be produced in this solution. We started experiments for this in October and will further investigate this method as an alternative to the classical production.
References
