“A PCR test or a bee?” This question will surely never be spoken in a pharmacy. Nevertheless, it is possible that the insect will one day participate in the fight against covid-19. A team of Dutch researchers have demonstrated that honey bees (apis mellifera) can be trained to detect samples contaminated with SARS-CoV-2.
Bees “smell” the coronavirus
“The COVID-19 pandemic has illustrated the need for develop fast and reliable test methods for novel zoonotic viral diseases in humans and animals,” the scientists explain in their paper published in the journal Biology Open on May 3, 2022.
For them, this diagnosis can be made by animals. Indeed, pathologies cause detectable changes in volatile organic compound profile (VOC) emitted by living beings that are likely to be noticed by certain animals.
VOCs harbor a unique olfactory fingerprint according to its sex, age, genetic heritage, metabolic conditions or even its diet. Thus, the state of health of an individual can be known through it. If dogs have already been trained to identify patients infected with the coronavirus according to their smell, the bees could be a “viable alternative to dogs for detecting SARS-CoV-2 infection due to their minimal maintenance and availability costs,” the research says.
Bee and covid-19: a reliable and sensitive diagnosis
The authors used methods of Pavlovian conditioning (conditioned reflex) to train bees to identify mink infected with SARS-CoV-2. They tested two different training protocols to assess their performance in terms of learning speedof accuracy and of memory. “We have designed a non-invasive rapid test in which several bees are tested in parallel on the same samples”, they assure.
Each insect was subjected to three different scents : samples of mink odors recently cured of covid-19, those of former sick animals or contaminated subjects. Researchers have shown that after training, flying animals manage to distinguish them from each other. “Although the discriminating ability of bees decreases between 1 h and 24 h after conditioning, we observed that they were nevertheless still able to discriminate significantly between new infected and healthy samples 1 day after conditioning”, they say in their article. They also seemed to be able to differentiate between samples whether the viral load was high or not.
According to the researchers’ calculations, trained bees could be an effective COVID-19 screening tool with a expected sensitivity of approximately 92% and an specificity about 86%. “Once improved, a diagnostic test using the learning abilities of bees could therefore provide an important complement to the current system for monitoring zoonotic diseases in remote livestock systems,” they conclude.