5. Discussion

Rearing honey bee larvae in vitro is of great importance for research on pathogens and risk assessment. Recent studies have developed protocols that gained success in producing unambiguous workers with low mortalities and facilitated the use of readily available materials (Vandenberg and Shimanuki, 1987; Aupinel et al., 2005). This enables laboratories to rear high numbers of workers compared to the significant efforts from pioneer studies (Rhein, 1933). However, the quality of food, incubation conditions and occurrence of infections can also result in high mortality, and we therefore present some recommendations to reduce undesirable effects.

The production of honey bee workers in the laboratory is still a challenge, in contrast to queen breeding. Queen breeding (see the BEEBOOK paper on rearing queens (Büchler et al., 2013) has been employed by beekeepers for decades. It is a process in which larvae grafted into queen cups are raised to queens in a colony, and has also been employed in the laboratory, where larvae are grafted on royal jelly in vitro. In all described methods royal jelly is diluted to reduce its queen-potency and produce workers. Alternatively, the recent findings regarding ‘Royalactin’ may provide another possibility. According to Kamakura (2011), ‘Royalactin’, a 57 kDa protein, is the queen-determining factor in royal jelly. He demonstrated that not only worker larvae, but even Drosophila larvae fed on the synthesized protein exhibit features normally associated with the honey bee queen: shortened developmental time, enhanced longevity and increased egg production. Even more intriguing is the fact that this protein seems to degrade with storage, and honey bee larvae reared on royal jelly stored at 40°C for 30 days all develop with a full worker morphotype. The potential of these findings to allow mass rearing of workers on degraded royal jelly needs further attention and investigation.

We gave much detailed information on what must or can be taken into account to make in vitro rearing of honey bee larvae comparable among different laboratories. Researchers also need the freedom to adopt the given recommendations as needed but in accordance with their research purpose and time management. In the end, researches must carefully consider a method (e.g. larval grafting stage, single or group rearing, ad libitum or limited feeding of diets or quantity and quality of manipulations) that is justified by the research topic.

In conclusion, it is absolutely essential for the successful rearing of honey bee larvae to use fresh materials of high quality and to gain experience in the chosen method before applying it to any scientific question or to routine testing of compounds and pathogens. However, even with the best of all methods, in vitro-reared larvae will always be in vitro-reared larvae and never become indistinguishable from larvae reared in a healthy colony. Therefore, researchers have to remain critical regarding results obtained with in vitro-reared larvae. Researchers need to decide how close to ‘natural’ (meaning reared in a colony, in vivo) bees they want to come with their in vitro efforts and adjust efforts accordingly. In general, the similarity of bees produced by current protocols to natural bees has been proven. This is a strong argument for the use of these methods in research and as the official testing method for plant protection products. Like with all other in vitro methods, these larvae are, although a valuable research tool, just a model for reality rather than reality itself.