*2.4. Honey Bee Survival Rates*

Kaplan–Meier curves showed that *A. dorsata* workers infected with *N. ceranae* dosed with 10<sup>6</sup> spores per bee (NO-0P) had significantly lower survival in comparison to the infected bees that received propolis treatment (*χ*<sup>2</sup> = 17.33, df = 3, *p =* 0.0005, Figure 10). The control bees treated with propolis extract (CO-50P) had the highest survival, followed by CO-0P and NO-50P, respectively. A similar trend was found in bees treated with 0.5 ppm COS, except in this case there was no significant difference between the control bees and the control bees treated with COS (*χ*<sup>2</sup> = 16.08, df = 3, *p =* 0.0010, Figure 11).

**Figure 10.** Kaplan–Meier survivorship curves of *A. dorsata* workers after being infected with 106 *N. ceranae* spores (NO-0P), versus infected bees that received a propolis treatment (NO-50P) or no infection and a propolis treatment (control: CO-0P and CO-50P). Survivorship curves with different letters within treatments are significantly different (Kruskal–Wallis test: *χ*<sup>2</sup> = 17.33, df = 3, *p =* 0.0005).

**Figure 11.** Kaplan–Meier survivorship curves of *A. dorsata* workers after *N. ceranae* infection at 106 spores (NO-0COS), NO-0.5COS or no infection (control: CO-0COS and CO-0.5COS). Survivorship curves with different letters within treatments are significantly different (Kruskal–Wallis test: *χ*<sup>2</sup> = 16.08, df = 3, *p =* 0.0010).

#### **3. Discussion**

The increase of hemolymph trehalose levels, protein content of the hypopharyngeal glands, and enhanced acini diameters of hypopharyngeal glands of the infected giant honey bee, *A. dorsata*, all indicate improved health after 50% propolis extract and 0.5 ppm COS treatment. Although the levels after treatment were not to the same level of the uninfected control bees, except for the acini diameter from the propolis treatment, there was still a significant increase for all health measures in comparison to the infected bees without any treatment. Based on our results, both stingless bee propolis extract and COS are effective treatments in improving the health of the honey bee. Propolis extract however may be a slightly better treatment as indicated by the full recovery of the acini diameter distance of the hypopharyngeal gland, which was not the case for the COS treatment. For both treatments there were no detrimental effects in the uninfected control bees for hemolymph trehalose levels, protein content in the hypopharyngeal gland, and the acini diameters in the hypopharyngeal gland, which suggests that these treatments are not having any damaging side-effects for these health parameters measured. Whether the increased trehalose levels and protein content of the hypopharyngeal gland is due to a lower parasite load or the improved nutrient absorption across the gut lining in treated bees—which results in higher tolerance of the parasite—remains to be investigated.

The higher trehalose levels in the uninfected bees treated with propolis extract in comparison to the uninfected bees without any treatment suggests that propolis might be affecting the sugar metabolism of the honey bee. This is interesting to note because hemolymph trehalose levels are central to buffering against the energetic stress suffered from the infected bees [24]. In addition, increased trehalose levels were found to be a key difference in honey bees selected to better tolerate *N. ceranae* infections [56]. Therefore, the increased hemolymph trehalose levels from propolis consumption could be one way in which increased survival results from infected bees treated with this [30]. Propolis extract also positively increased the protein contents of hypopharyngeal glands, and the acini diameters of hypopharyngeal glands of honey bee. This suggests that the increased health measures may also be due to the lowering of the *N. ceranae* load in the treated bees. This corresponds to the results of the previous study which showed the potential of propolis extracted from stingless bee to control *N. ceranae* infection in the red dwarf honey bee, *A. florea* [30]. Recently, it has also been demonstrated that propolis extract can increase the survival of *A. mellifera* [45,46]. Taken together, these results suggest that propolis extract may have a general positive impact on bee health, across all honey bee species. Further supporting this notion is the fact that stingless bee propolis extracts are known to specifically have antifungal properties [57]. Moreover, a previous study demonstrated abnormal structure of *N. ceranae* spores, inside *A. cerana* bees, after being treated with propolis extract, which corresponded with the interference of spore growth and development [47].

Bees treated with COS after infection with *N. ceranae* also had significantly higher trehalose levels, protein contents of hypopharyngeal glands, and increased acini diameters. However, it is more likely that these effects are resulting from indirect mechanisms such as enhanced bee immunity or increased nutrient absorption across the gut lining as opposed to directly reducing the reproduction and growth of the *N. ceranae* infection. COS is known to improve nutrient digestibility, gut functions and gut modifications in animals [50]. COS may affect *Nosema* development, but is more likely to achieve this through enhancing bee immunity, that will eventually result in higher hypopharyngeal gland protein contents, trehalose levels, and the increasing hypopharyngeal gland acini from decreased *N. ceranae* loads [58]. This is plausible because *N. ceranae* typically suppresses the immune system in infected bees in order for increased growth and reproduction inside the host [17,59].

It is important to note that *N. ceranae* can infect *A. dorsata* and develop well in this host. As previously shown, *A. mellifera*, *A. cerana* and *A. florea* can also be infected by *N. ceranae*. To date, all of the honey bee species have now been shown that not only can become infected with this parasite, but they are also suffering from the pathological effects of the infections as well. Based on our results *A. dorsata* is no exception, which raises concerns as the honey bee species may be suffering from some of the same behavioral and physiological changes that have been documented in *A. mellifera* from a *N. ceranae* infection [27]. Previous results show that the parasite develops well in each of the four honey bee species and that the intracellular life cycle is completed within three days p.i. [60–62]. Due to the successful reproduction in all four of the honey bee species there are opportunities for cross transmission between the species on a community level as they have overlapping foraging ranges and are known to share the same floral species when foraging [62–64].

The reduction in acini diameter of the hypopharyngeal glands of *N. ceranae*-infected bees might due to deficiency of amino acids used for secretory cell development. This is not surprising because previously it has been shown that the amino acid profiles in the hemolymph of infected bees is altered and feeding pollen can increase the survival of infected bees [65,66]. The recovery in the hypopharyngeal gland protein is important because it has been noted to play a role in protein synthesis of royal jelly production [6]. Metabolite dysregulation of royal jelly secretions has been documented in *N. ceranae* infected bee hives, which has implications for the antibacterial effectiveness of the secretions when feeding the brood [18]. Our findings suggest that COS and propolis extract treatment is likely to contribute to the increase of royal jelly productivity as well at the colony level, due to the increase of protein contents and the acini diameters of the hypopharyngeal glands. The lowering of the immune system is likely to be a result in the lack of protein nutrition resulting from the force feeding of spores that geminate and proliferate within the midgut epithelial cells where they disrupt host nutrient absorption [67,68]. Although infected bees do not exhibit obvious external disease symptoms, some of the main pathophysiological effects from an infection identified from omics studies have pointed to metabolic dysregulation [18,21,25,69]. Resulting from this metabolic dysregulation are the key symptoms of infection, which are lowered trehalose levels and reduced hypopharyngeal glands [22,24,62]. Therefore, we find that the measures used in this study to be accurate predictors of bee health and recovery from a *N. ceranae* infection. The effects of a *N. ceranae* infection on the colony level include lower colony population and the reduction of honey production [61,70]. Thus, we are interested if this treatment at the colony level might show an improvement in these colony level symptoms of infection.

The use of natural product such as propolis from stingless bees and COS will facilitate new strategies that can be used to control *Nosema* and improve honey bee health and beekeeping production. However, further experiments could be performed to determine the optimal doses to maximize the effect for each of the treatments. In addition, long term treatments could be investigated on a colony level to determine if they are great enough to effectively reduce the *N. ceranae* parasitic loads in a more natural setting. Perhaps using both treatments at the same time will synergistically improve the overall effectiveness in improving the health and survival of the honey bee. On one hand, *N. bombi* might also be another suitable target for this treatment which could lead to the health improvement of bumble bees as well, but on the other hand, the non-lethal side effects of these treatments should be investigated to determine if they pose any detriment to the health of bees. All of these would be interesting avenues to pursue in the future to further understand the practical use of stingless bee propolis extract and COS in terms of managing *N. ceranae* infections around the world.
