**1. Introduction**

Honey bees play a vital role in agricultural crop production and ecosystem stability due to their pollination services [1–6]. Despite their importance there has been a global decline in bee health around the world at unsustainable rates [7–9]. The health decline can be attributed to a number of health factors such as pesticide exposure and poor nutrition, with parasitic infections as one of the major contributors [10,11]. *Nosema* disease or nosemosis is one of the most widespread parasitic infections of adult honey bees and has been implicated to play a major role in the most recent global bee health decline [12–14]. Nosemosis is caused by three species of microsporidia, *Nosema apis*, *N. ceranae* and *N. neumanni*, but the most prevalent strain found in honey bees that has emerged is *N. ceranae* displacing much of the *N. apis* infections worldwide [15,16]. Nosemosis is considered to be a chronic infection that does not exhibit obvious external disease symptoms, but can cause a poor nutrient and energy absorption leading to a suppressed immune function and ultimately a shortened life span [17–21]. Infected bees have evidence of lower trehalose and lipid levels, and a reduced hypopharyngeal gland resulting from the poor nutrient absorption across the gut lining [22–24]. *N. ceranae* primarily lives and reproduces in the gut lining which is likely the cause for the poor nutrient absorption in infected bees [25,26]. Consequently, infected individuals suffer from energetic stress, which results in increased bee mortality on the individual and colony level [24,27–30].

There are only a few treatment options on the market for controlling Nosemosis. The antibiotic Fumagillin has been on the market for a long time, but it is unable to kill the

**Citation:** Naree, S.; Ponkit, R.; Chotiaroonrat, E.; Mayack, C.L.; Suwannapong, G. Propolis Extract and Chitosan Improve Health of *Nosema ceranae* Infected Giant Honey Bees, *Apis dorsata* Fabricius, 1793. *Pathogens* **2021**, *10*, 785. https:// doi.org/10.3390/pathogens10070785

Academic Editor: Giovanni Cilia

Received: 25 May 2021 Accepted: 20 June 2021 Published: 22 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

mature spore form of the parasite [31], so reinfections can occur [32,33]. Moreover, its use has been banned in the European Union because it has been shown to contaminate honey and could possibly lead to the buildup of antibiotic resistance in humans [34]. A natural product that is completely safe and environmentally friendly is desirable, especially for organic beekeepers. There have been a number of recent developments in this area which include using phytochemicals, Bee Cleanse, zeolite clinoptilolite, plant extracts, and propolis extract that have been documented to be effective alternative treatments [30,35–38]. These studies generally show lowered parasite loads and improved survival of the treated bees, but very few assess the health of the bee to determine how the survival of the treated bees are being increased. Among the various promising substances to control *N. ceranae-* infection, ApiHerb® and Api-Bioxal®, commercial dietary supplements were used as treatments against *N. ceranae*-infection effectively in both laboratory and colony level [39]. The commercial probiotics, Vetafarm Probotic, Protexin Concentrate single-strain (*Enterococcus faecium)*, and Protexin Concentrate multi-strain (*Lactobacillus acidophilus*, *L. plantarum*, *L. rhamnosus*, *L. delbrueckii*, *Bifidobacterium bifidum*, *Streptococcus salivarius*, and *E. faecium* [40], and *Parasaccharibacter apium* (PC1 sp.) and *Bacillus* sp. (PC2 sp.) [41] also were used to reduce spore loads and mortality in *N. ceranae* infected-honey bees. Currently, the control of *N. ceranae*-infections involves the use of the natural compounds to stimulate the immunity of honey bee, *A. mellifera* by inducing resistance against pathogens. Chitosan and peptidoglycans were used to reduce *N. ceranae*-infection, and increase survivorship of *N. ceranae* infected-bees. In addition, peptidoglycan and chitosan promoted the gene expression of hymenoptaecin and defensin2 [42]. Another example of natural compounds being effective at reducing *N. ceranae* loads involves using Brassicaceae defatted seed meals (DSMs) containing antimicrobial and antioxidant properties [43]. Besides fumagillin, sulforaphane was used to control *N. ceranae*-infection in laboratory. It was reported that 1.25 mg/mL of sulforaphane showed 100% reduction of spore counts, but also caused 100% bee mortality. The antimicrobial properties of this new alternative treatment may be promising, however reducing its toxicity is required before it can be considered as an alternative treatment for controlling *N. ceranae* [44].

Propolis extract of stingless bees is emerging to be an effective treatment to control *N. ceranae* across three of the four honey bee species, *A. cerana*, *A. mellifera*, and *A. florea*, [30,45–47]. Propolis is collected by bees and contains a number of plant resins, which are considered to be a natural product. In general, the plant resins are known to have antimicrobial effects and are used by bees to aid in sanitizing their hives. These plant resins also have recently been found to have a potential inhibitory effect on microsporidian development [30,45–48]. However, the propolis has to be fed to the honey bee in order to observe a reduction in the proliferation of *N. ceranae* in the midgut cells as the bees do not preferentially consume food containing propolis when infected. When fed, propolis extract treatment significantly enhances bee survival [30,45–47]. Another natural product, chito-oligosaccharides (COS) promotes antimicrobial activity and has been shown to stimulate the immune system thereby reducing *N. apis* infection in *A. mellifera* [48–51]. COS is a derivative of chitosan which is known as a biopolymer and polysaccharide found in the exoskeleton of insects and crustaceans. This water-soluble glycoprotein molecule has been used as a pre-biotic for gastrointestinal infections and diarrhea. COS is also known to aid in increased amino acid absorption across the gut lining, and also promote gut health including anti-inflammation activity through activation of 5' AMP-activated protein kinase (AMPK) [52–55]. We, therefore, hypothesize that this treatment can aid in treating the symptoms of a *N. ceranae* infection and consequently improve the health of the honey bee.

Whether the pathological effects from a *N. ceranae* infection is of the same magnitude across the honey bee species and can be generalized to the giant honey bee, *A. dorsata*, remains unknown. *A. dorsata* serves as a main pollinator for crop plants in Thailand and provides a substantial amount of honey for a number of Asian countries [6]. Thus, the first aim of this study is to investigate the pathological effects of a *N. ceranae* infection in *A. dorsata*. Secondly, we aim to determine the efficacy of propolis extract and COS, as

alternative treatment options for *N. ceranae* infections, by measuring hemolymph trehalose levels, protein contents in the hypopharyngeal gland, survival rates and acini diameters of the hypopharyngeal glands as health status indicators.

#### **2. Results**

#### *2.1. Hemolymph Trehalose Levels*

*N. ceranae*-infected bees without any treatment had the lowest hemolymph trehalose levels on day 14 p.i. compared to all other treatment groups (*χ*<sup>2</sup> = 34.52, df = 3, *p* < 0.0001, Figure 1). The highest levels of hemolymph trehalose were found in uninfected bees treated with propolis extract, CO-50P (273.2 ± 6.69 μg/bee) followed by the control group, CO-0P (250.8 ± 2.26 μg/bee). However, *N. ceranae*-infected bees treated with 50% propolis extract (NO-50P) showed higher levels of trehalose (204.2 ± 5.13 μg/bee) than that of *N. ceranae*-infected bees without propolis extract treatment, NO-0P (148.0 ± 5.79 μg/bee). Interestingly, similar trend was found in bees treated with COS where the highest hemolymph trehalose levels were found in the control group with 0.5 COS (CO-0.5COS) and without COS(CO-0COS) treatment 250.8 ± 2.26 μg/bee and 254.2 ± 1.73 μg/bee, respectively. The lowest hemolymph trehalose levels were found in the *Nosema* infected bees without treatment (NO-0COS) 148.0 ± 5.79 μg/bee, while there was a significant increase in the infected bees that received a COS treatment (NO-0.5COS) 184.2 ± 5.14 <sup>μ</sup>g/bee (*χ*<sup>2</sup> = 33.21, df = 3, *p* < 0.0001, Figure 2).

**Figure 1.** A box plot with the lines representing the median levels of hemolymph trehalose across the treatment groups of the propolis extract experiment. The control bees (CO-0P) (grey), propolis control bees (CO-50P) (green), *N. ceranae*-infected bees not treated with propolis extract (NO-0P) (red) and infected bees treated with 50% propolis extract (NO-50P) (light green) are represented by each box plot. The hemolymph trehalose levels are measured on 14 days p. i. The box indicates the inter-quartile range while the vertical bars indicate the range of the data. The different letters above each box represent significant differences (Kruskal–Wallis test: *χ*<sup>2</sup> = 34.52, df = 3, *p <* 0.0001).

**Figure 2.** A box plot showing the median hemolymph trehalose levels data across treatments from the COS experiment. The control bees not treated (CO-0COS) (grey), the treated COS control bees (CO-0.5COS) (purple), *N. ceranae*-infected bees not treated (NO-0COS) (red) and the infected bees treated with 0.5 ppm (NO-0.5COS) (light blue) are each indicated by a box plot. The hemolymph trehalose levels were measured on day 14 p.i.. The boxes indicate the interquartile range, while the vertical bars represent the range of the data. The different letters above each box plot represent significant differences (Kruskal–Wallis test: *χ*<sup>2</sup> = 33.21, df = 3, *p* < 0.0001).
