Highlights
- •
Ingested dsRNA spreads via hemolymph and is secreted in worker and royal jellies
- •
Functional dsRNA flows horizontally among honey bees by jelly consumption
- •
Worker and royal jellies harbor differential natural ssRNA and dsRNA populations
- •
Diverse pathogenic RNA fragments naturally occur in worker and royal jellies
Introduction
In eukaryotes, sequence-specific gene silencing pathways, generally termed RNA interference (RNAi), are induced and maintained by the presence of dsRNA (Fire et al., 1998). Through processing of base-paired RNA into small RNAs, these mechanisms regulate gene expression in both co-transcriptional and post-transcriptional levels (Bernstein et al., 2001). While RNA-mediated nascent transcript destabilization and heterochromatin remodeling inhibits gene transcription, post-transcriptional gene silencing downregulates gene expression through guiding target RNA degradation or repression of translation (Almeida and Allshire, 2005, Castel and Martienssen, 2013).
RNAi can be divided into cell-autonomous and non-cell autonomous (Whangbo and Hunter, 2008). In cell-autonomous RNAi, silencing is restricted to cells that produce or were exposed to the dsRNA trigger. Initiation of local RNAi can develop in some organisms into a non-cell autonomous silencing signal, affecting cells and tissues that originally did not generate or were not introduced to dsRNA (Jose and Hunter, 2007, Voinnet and Baulcombe, 1997). The mechanisms that facilitate RNA export from donor cells, extracellular spread, and import into acceptor cells are not fully elucidated, but are under ongoing investigation in diverse biological systems.
In 1998, Timmons and Fire (1998) were the first to report on gene silencing triggered by environmentally acquired dsRNA. In essence, their finding represents a form of horizontal regulatory RNA transfer. To date, susceptibility to environmental RNAi has been established in fungus and animals from different phyla, including Nematodes, Platyhelminthes and Arthropods (Knip et al., 2014). Environmental RNAi experiments mostly involve dsRNA ingestion, suggesting that dietary consumption is an effective RNA uptake pathway. Further supporting this, potent RNAi transmission from transgenic dsRNA-expressing plants to invertebrate herbivores has been widely reported (Mao et al., 2007, Zhang et al., 2015). Accordingly, host-to-parasite RNAi transfer (commonly termed “host-induced gene silencing” [HIGS]) has been applied to agriculture in recent years, demonstrating a potential practical strategy to control various viral and pest-related diseases (Garbian et al., 2012, Zotti et al., 2018).
One of the main questions in the field of environmental RNAi is whether natural and functional RNA transfer among organisms occurs. Recently, transmission of parasitic nematode-derived miRNA to its mammalian host has been shown to compromise immunity of infected mice (Buck et al., 2014). Similarly, pathogenic fungi exploit mobile small RNA signals to modulate plant immune responses via RNAi (Weiberg et al., 2013).
Reciprocally, Arabidopsis plants secrete and transfer vesicles containing small RNAs that could suppress virulence fungal genes (Cai et al., 2018). Furthermore, ingestion of pollen-derived plant miRNA induces worker bee sterility in a sequence-specific manner (Zhu et al., 2017).
Interestingly, while the aforementioned examples provide evidence that some organisms acquire, and are affected by foreign regulatory RNA, it is still puzzling why would they allow so? This evolutionarily maintained susceptibility to non-self-regulatory RNA is intriguing in light of the fact that the most well-known transmissible RNA are viruses.
One of the remarkable characteristics of honey bees is their environmentally mediated phenotypic plasticity (Winston, 1991). Female bee larvae can either develop into worker or queen; two castes with distinct morphology, physiology, reproductive capability, lifespan, and behavior. This developmental flexibility of genetically identical individuals is driven by differential diet consumption. A larva fed exclusively on royal jelly will develop into a queen, whereas larva fed on worker jelly will develop into a worker (Haydak, 1970). In other words, nutritional differences trigger one genome to direct two distinct phenotypic outputs in honey bees.
Epigenetic regulation has been shown to play a role in the honey bee’s caste differentiation (Kucharski et al., 2008). Consistently, the methylation imprint varies between the brain’s DNA of workers and queens, demonstrating unique epigenetic profiles among workers and queens (Lyko et al., 2010). Nonetheless, although the general involvement of epigenetics has been established, it is still not clear how caste-specific DNA methylation marking is directed.
Previously, we reported on an RNAi-based ingestion system for the control of Israeli Acute Paralysis Virus (IAPV) disease in honey bees (
Maori et al., 2009). Field trials, in which this environmental RNAi system was employed, indicated that the colony performance of virus-inoculated hives deteriorated following virus infection in control hives, whereas that of dsRNA-treated hives remained strong (
Hunter et al., 2010). Interestingly, dsRNA-treated hives also produced more honey, when the main honey flow was 3–4 months after the last dsRNA treatment. By that time, most of the originally treated bees would have been replaced by new generations. Honey bee viruses can be transmitted among individuals in the hive both horizontally and vertically (Chen and Siede, 2007). It was therefore expected that hives would become virus-affected once the new generation gradually replaced the previous, dsRNA-treated one. Potential persistence of disease protection raised the question of whether treated honey bees may serve as vectors for RNA.
Following this hypothesis, here we show that environmentally consumed dsRNA in honey bees is up-taken from the digestive system and systemically spread through the hemolymph associated with a protein complex. Then, these RNAi carrier bees transfer silencing-triggering molecules to the next generation via dsRNA secretion into the jelly. We demonstrate that jelly-secreted dsRNA is biologically active and triggers a long-lasting silencing effect in the recipient generation. Finally, we characterize diverse naturally occurring endogenous and exogenous RNA populations in royal and worker jellies. These findings demonstrate an environmentally mediated transmissible RNA in honey bees.
