Agri-inputs & biocontrol
Bee health biopreparations
Insect innate immunity, vitellogenin as a carrier, the unusually simple honeybee gut microbiome — and why any product's effect is hard to measure at colony level.
A honeybee cannot be immunised in the sense a mammal is. Insects have no antibodies, no B cells and no somatic recombination of receptors; defence rests on an innate response — recognition of conserved microbial patterns, synthesis of antimicrobial peptides, phagocytosis by haemocytes, melanisation. Anything sold as a bee vaccine therefore rests on a different phenomenon: trans-generational immune priming.
Priming through the queen
The mechanism turns on vitellogenin, the egg-yolk storage protein that in bees also binds fragments of bacterial cells. A killed whole-cell preparation of Paenibacillus larvae, the agent of American foulbrood, is fed to the colony in queen feed; nurse bees process it and pass it to the queen in royal jelly, vitellogenin carries the bacterial fragments into developing oocytes, and the larva that hatches meets the pathogen with innate defences already raised. This is the route granted a conditional licence by USDA-CVB in 2023 — the first veterinary biologic for an insect.
The same mechanism sets the ceiling. Priming is not immunological memory: there is no clonal population holding specificity, the effect decays across later brood cycles, and protection is partial. It reduces larval mortality on challenge; it does not clear the colony, and it does not touch the pathogen’s defining property — P. larvae spores stay viable in comb and equipment for decades, so quarantine and destruction of infected material remain part of the scheme.
The gut microbiome, and why it is hard to join
The second line of products is probiotic. The honeybee gut is unusually simple: roughly a dozen stable phylotypes, among them Snodgrassella alvi, Gilliamella apicola, the Firm-4 and Firm-5 lactobacilli and Bifidobacterium asteroides. This community ferments pollen polysaccharides, lowers gut pH and competitively excludes Vairimorpha (Nosema) ceranae and opportunists. It is acquired by trophallaxis and contact with comb rather than from food, and that is the constraint: an introduced strain must establish in a niche that is already occupied. Usually it passes through, so the effect depends on repeated dosing and fades when dosing stops.
Why field results are smaller than laboratory ones
Colony loss is multifactorial. Varroa destructor does its damage less by parasitism than as a vector for deformed wing virus, and to that are added protein-poor forage in monoculture landscapes and sublethal pesticide exposure. A product that removes one factor delivers only a modest gain in survival within that system.
Then there is the statistics. The unit of observation is the colony, not the bee: thousands of individuals in a hive are not independent replicates, between-colony and between-apiary variance is large, and overwintering adds its own. Separating a real effect from noise takes tens of hives per treatment and observation through a winter — which is why well-controlled field trials against an untreated control are scarce in this field, and why cautious wording is the honest wording.