For pathogen detection, we used the following experimental procedure: First, infected mosquitoes are grinded in order to expose pathogen proteins. Then, mosquitoes proteins and pathogen proteins of the sample are bind to horse-radish peroxidase (HRP). The conjugated proteins are deposited on are the patch precoated with specific antibodies. The patch is then washed, pathogen proteins are retained by antibodies, and peroxidase activity is revealed using a specific substrat. The principle of that method had to be tested before running experiment. In particular, specificity and sensitivity had to be evaluated. For those tests, we used purified viral proteins (E-protein from Yellow fever virus or E2 protein from chikungunya virus) and specific antibodies (4G2 for E-protein of Yellow fever virus and 3E4 antibody for E2 protein from chikungunya virus). We first checked that we were able to bind viral proteins with HRP, then that we could detect viral protein on a membrane, either purified or diluted in a Bovine Serum Albumine (BSA) solution in order to mimic mosquitoes proteins. Next step was to test the method when viral proteins were diluted with mosquitoes proteins. Finally, we tested the patches ability to detect viral pathogens from infected mosquitoes. For this last experiment, all steps involving infectious materials were performed by one of the coaches in a BSL3. Materiel: - PVDF or nitrocellulose membrane - YFV E protein, CHIKV E2 protein - 4G2 antibody - Secondary antibody (anti-mouse alexafluor 488), - BSA - PBS - Tween - Milk 5% in PBS-Tween - EZ-link kit - Rocker agitator Protocol: I- Sample preparation : The sample is tagged using EZ-Link kit a. Add 500mL of ultrapure water to the dry-blend Phosphate Buffered Saline (PBS). b. Prepare 1mg of antibody in 1.0ml of PBS(+/-BSA1%) c. Reconstitute 1mg of lyophilized EZ-Link Plus Activated Peroxidase with 100μ L of ultrapure water and add it to the antibody solution or add the protein sample directly to the lyophilized activated. d. In a fume hood, immediately add 10μl of Sodium Cyanoborohydride to the reaction and incubate for 1 hour at room temperature e. Add 20μl of Quenching Buffer and react at room temperature for 15 minutes. The conjugate can be store at 4°C for up to 4 weeks. II- Membrane a. Coating: put 1µl of a non-diluted antibody on the membrane, let the membrane dry for 5 minute at room temperature b. Saturation: a. Put the membrane in PBS-Tween 0,5%-milk 5% on a rocker, 1 hour at room temperature b. Replace the washing solution with a fresh one c. Binding: Add the sample in the PBS-Tween 0,5%-milk 5% solution at the appropriate dilution, incubate overnight at 4°C on a rocker d. Washing: Wash the membrane 3 time for 5 minutes on a rocker at room temperature e. Revelation: Reveal the membrane using approximately 1ml of Pierce ECL Western Blotting substrate with a my ECL imager machine To download the General protocol, please follow in this link
4G2 antibody dilutions: - Rows A, E: 1/1000 --> 2.2µL 4G2 + 2.2mL PBS - Rows B, F: 1/2000 --> 1mL of previous dilution + 1mL PBS - Rows C, G: 1/6000 --> 800µL of previous dilution + 1.6mL PBS - Rows D, H: 1/12000 --> 1mL of previous dilution + 1mL PBS - No antibody in the column 12 Incubate overnight at 4°C.
Dilutions: E(YFV): - starting concentration: 3.46mg/mL - Target concentration: 1mg/mL (according to EZ-link kit) So we made a 1/3,46 dilution E(YFV): Ci.Vi = Cf Vf --> Vi=(Cf.Vf/Ci) --> 1x1/3,46 = 289µL We mixed 289µL of E(YFV) with 711µL H2O to have a 1mg/mL solution into 1mL = 100 dilution. We have 2 conditions x 8 wells of dilution.
We added 10µL of sodium cyanoborohydride per dilution, incubated them at room temperature during 1h15, and added 9µL or 5µL of quenching solution into 100 and other solutions, respectively during 15min à room temperature. NB: 10µL sodium cyanoborohydride was a mistake, should have added 4.5µL or 2.5µL into 100 and other dilutions, respectively (in order to dilute 10 times the sodium cyanoborohydride into the sample). ELISA: - The plate has been washed by 3 x 100µL PBS-Tween 0.9% - Saturation by 100µL of PBS-BSA 0.1% during 1h at room temperature - Staining with 25µL of sample (dilutions 100 to 10-10) per well, overnight at 4°C, according to the conjugated E-protein dilution scheme
For those experiments using ELISA plates, we have never been able to get any signal. We decided to move to a more relevant system that would mimic the patches we were developing and allow us to get better results.
Conclusion: The principle of the technique is validated.
Aim: Sensitivity of the detection. To evaluate the sensitivity, we proceed as described before, and we used YFV E protein at 1mg/ml or serially diluted into PBS + BSA 1% to mimic the presence of mosquito proteins in the sample. 3 dilutions were used : 1mg/ml, 10-4mg/ml, 10-8mg/ml. Results: No signal was visible.
Aim: Sensitivity of the detection. The previous experiment was repeated, we made new reagents and used nitrocellulose membrane in addition to PVDF membrane. Results: No signal was visible.
Aim: Sensitivity of the detection The previous experiment was repeated, we used a new antibody (anti-chik 3E4) together with CHIKV E2 envelope protein. Results:
Aim: Sensitivity of the detection with mosquito protein. 100 mosquitoes were grinded and CHIKV E2 envelope proteins were diluted into in order to have approximately the same amount of viral proein we would find into one infected mosquito alone and in one infected mosquito together with 100 non-infected mosquitoes in order to mimic the fact that not all mosquitoes trapped will be infected with a pathogen. Results:
Signal differences cannot be evaluated with naked eye, so we used My Image Analysis software to quantify the signal. We concluded that our technic is able to detect viral antigen specifically, although we have to greatly improve sensitivity and specificity.
Aim: Proof of concept. In this experiment, we used 8 prototype patches to check their ability to detect pathogens in mosquitoes. 4 patches were used to repeat the previous experiments in real conditions. The 4 patches remaining were used to test infected mosquitoes. Aedes aegypti were experimentally infected with the vaccinal strain of Yellow fever virus. 14 days after infection, they were grinded in 1,5 ml tubes. Tubes were centrifuged and supernatant were inactivated by placing the tubes under a UV lamp for 30 minutes. All steps using infectious materials were performed by a coach in a BSL3. . Samples were then tested on the patch using the same protocol as decribed previously with the patch as a solid support for antibody (in remplacement of membranes).
Results: Results are unconclusive for this experiment, but a tendancy seems to appears. Indeed, patch number 3, with specific antibody and viral protein displays the highest signal but the background seems to be elevated as well, as shown by the signal in membrane number 2. Thus, no statistical differences could be shown with a single experiment. Anyway, the sensitivity has to be improved for detecting pathogens at a lower level. Using our technic, we were not able to detect viral protein in infected mosquitoes. This experiment needs to be completed with more test, using different antibody and protein dilution. Sadly, we hadn’t enough protein to produce more patch, so we were not able to repeat the experiment.