Vairimorpha (Nosema) ceranae is a fungal parasite that causes nosemosis, a chronic disease that is frequently observed in bee colonies worldwide (1). In 1996, Fries et al. (2). first identified V. ceranae in A. cerana colonies, and established A. cerana as the original host of this parasite. Since its initial discovery in Europe in 2006, V. ceranae has rapidly spread to other parts of the world, North America, South America, Asia, Africa, and likely other regions as well, likely facilitated by the global trade of bees, bee products, and the movement of beekeeping equipment (3).

V. ceranae is associated with a decline in bee health, reduced colony productivity, and colony collapse disorders (CCD) (4, 5). Over the past two decades, a substantial number of studies have been conducted to investigate the effect of V. ceranae infection on Apis mellifera, demonstrating that fungal invasion negatively influences various aspects of honeybee metabolism, behavior, physiology, and health (6–12). For instance, studies have shown that the V. ceranae infection cause metabolic stress, upregulates genes encoding α-glucosidase and trehalose transport enzymes (9), while downregulating trehalase and Glucose-Methanol-Choline oxidoreductase genes (11). These alterations, as confirmed by proteomics, resulted in reduced levels of energy supply proteins (8). The behavioral consequences of V. ceranae infection in A. mellifera are also noteworthy, as infected bees often display increased hunger and consume more sucrose to compensate for their impaired metabolic state. However, this increased food intake does not fully alleviate the underlying nutritional and energy deficits (6), and accelerated lipid loss in bees suggests that lipids may be mobilized as an emergency energy source (12). V. ceranae infection can damage the midgut epithelial cell structure, inhibit apoptosis and immunosuppression, and shorten the lifespan of A. mellifera workers (7, 10).

Through extensive interactions and long-term co-evolution, A. cerana and V. ceranae have evolved to coexist (13). Previous studies have investigated the effects of V. ceranae infections on A. cerana (14–17). Sinpoo et al. (14) found stronger immune responses and lower loads of Nosema apis and V. ceranae spores in workers of A. cerana compared to A. mellifera, suggesting a higher tolerance of A. cerana against microsporidian infection. Chantaphanwattana et al. (18) reported that V. ceranae proliferated more in A. mellifera, whereas A. cerana exhibited a stronger immune response and lower spore load, indicating a superior defense mechanism of A. cerana at the individual level. By contrasting the similarities and differences in V. ceranae infection of A. cerana and A. mellifera, a deeper understanding of the complex interactions between V. ceranae and bee hosts, as well as the factors responsible for coaptation between A. cerana and V. ceranae will be gained. In addition, novel and valuable insights from the aforementioned comparisons could guide the creation of targeted strategies to alleviate the adverse effects of fungal parasites on different bee species.

A. c. cerana is a bee species that is found in China and is widely used in beekeeping practice in several Asian countries (19). Previously, our team identified the miRNA response and miRNA-regulated network of A. c. cerana workers following V. ceranae invasion (20), and performed a comprehensive investigation of the profiles of highly expressed genes in both the host and microsporidian (21, 22). Recently, our group investigated the immune response of A. c. cerana workers to V. ceranae infection using a transcriptomic investigation and revealed that different cellular and humoral immune responses were used by A. c. cerana and A. mellifera ligustica workers to defend themselves against infection by V. Ceranae (23). Moreover, this study provides an opportunity for a deeper comparison of the effect of the same fungal parasite (V. ceranae) on two different bee species (A. cerana and A. mellifera), which will enrich our understanding of bee nosemosis and facilitate the development of biological drugs for nosemosis control.

Many aspects of the effect of V. ceranae on A. cerana are yet to be explored. Pathogenic infections in insects result in the generation of reactive oxygen species (ROS) that can kill and eliminate pathogens (24). However, excessive production of ROS in vivo induces oxidative stress, predominantly damaging various biological macromolecules, and ultimately resulting in cell death (25). Antioxidant enzymes reduce the oxidative response to balance ROS production and maintain homeostasis (26). Superoxide dismutase (SOD) catalyzes the disproportionation of superoxide to produce hydrogen peroxide and molecular oxygen, whereas catalase (CAT) participates in the decomposition of hydrogen peroxide into water and oxygen (27–29). Glutathione-S-transferase (GST) has been shown to exert function in metabolizing lipid peroxides (27, 30). Polyphenol oxidase (PPO) is an oxidoreductase that uses molecular oxygen to catalyze the oxidation of a wide range of phenolic compounds (31). Insects appear to use PPO in immune responses such as melanization to physically limit the spread of microbes (32). Whether V. ceranae infection affects the activities of major antioxidant enzymes in A. cerana remains largely unknown.

Currently, little is known regarding the effect of V. ceranae infection on A. cerana, hindering the understanding of the mechanisms underlying the interaction between V. ceranae and A. cerana and the development of novel strategies for the control of bee nosemosis. Here, to evaluate the impact of V. ceranae infection on A. c. cerana workers, V. ceranae spores were purified and used to inoculate newly emerged workers to evaluate the effects of V. ceranae infection. This was followed by an in-depth investigation of V. ceranae spore load and host sucrose solution consumption. Activities of four major antioxidant enzymes (SOD, PPO, CAT, and GST) were determined. Paraffin sections of the host midgut tissue were prepared and subjected to microscopic observation. The survival rates of V. ceranae-inoculated and uninoculated workers were analyzed. The findings from this study provide valuable experimental evidence for V. ceranae infection of A. c. cerana workers and contribute to the dissection of the host response, microsporidian infection, and host–microsporidian interaction.

There is an ongoing debate and uncertainty surrounding the precise lifecycle duration of V. ceranae across different hosts. In A. mellifera, all intracellular stages of the life cycle of V. ceranae were observed at 3 dpi inside the worker’s epithelial cells using transmission electron microscopy, suggesting that the time for V. ceranae to complete the developmental cycle was 3 d (39). Based on other lines heterologous lepidopteran cell line IPL-LD-65Y in vitro, Gisder et al. reported that the life cycle of N. apis, the sister species of V. ceranae, is approximately 4 d (40). Although A. cerana is the original host of V. ceranae, its life cycle of V. ceranae in the midgut epithelium cells of A. cerana is unknown. Here, we observed a continuous increase in spore load from 1–12 dpi ( Figure 2 ), indicative of sustained proliferation of V. ceranae within host cells. Additionally, the number of spores increased 10-fold every 4 d ( Figure 2 ), suggesting that the life cycle of V. ceranae in the midgut epithelial cells of A. cerana workers was approximately 4 d, which differs from reports on A. mellifera. Intriguingly, we observed the presence of spores (3.5 ± 0.87 × 10⁶) as early as 1 dpi. Similar phenomenon was observed in previous studies, namely that by 3 h post inoculation, mature spores can be observed in the ventricular lumen of infected A. mellifera workers (39). We hypothesized that the detected V. ceranae spores at 1 dpi contained spores that did not infect the bee host cells or empty spores. However, further studies are needed to verify this hypothesis.

Nosema spp. have lost mitochondria during evolution, and thus must acquire energetic substrates directly from hosts for maintenance, growth, and reproduction (41, 42). Intestinal lesions caused by V. ceranae proliferation may decrease the absorptive capacity of honeybees and cause starvation symptoms, such as impoverishment of hypopharyngeal protein secretions (8). Chronic nutrient and energy stress suppress the innate immune system (43). After V. ceranae infection, workers of A. mellifera consumed a significantly higher amount of sucrose solution over the 24-h period tested (6). Here, we observed that the sucrose solution consumption of V. ceranae-inoculated workers was nearly always higher than that of uninoculated workers from 1–20 dpi, which indicated that the workers attempted to compensate for the imposed energy stress by consuming more sucrose solution. This is consistent with the results of a previous study on A. mellifera workers infected with V. ceranae (44). In summary, these results demonstrate that V. ceranae infection results in energy stress in both A. mellifera and A. cerana workers. In addition, the chronic stress caused by V. ceranae infection is likely to negatively influence foraging behavior, nutritional balance, immune response, and other important aspects of A. cerana workers, thus deserving more attention.

Insects maintain homeostasis in the body by neutralizing ROS and xenobiotics through the activation of various metabolic pathways that help protect cellular components from oxidative stress (26, 45). The antioxidant enzyme system, comprising SOD, PPO, CAT, and GST, is a critical physiological system for resisting oxidative stress in insects (46, 47). After infection with pathogens, the insect immune system stimulates the production of ROS, including superoxide radicals, to kill the invaders (24). CAT efficiently breaks down toxic hydrogen peroxide into harmless water and oxygen molecules (48). GST protects cells from oxidative stress by conjugating glutathione to ROS and other electrophilic compounds (49). It has been previously reported that GST is linked to oxidative stress response in the midgut of Drosophila melanogaster larvae (50). In the present study, CAT activity in the V. ceranae-inoculated group increased significantly at 5, 6, and 7 dpi ( Figure 5A ), whereas GST activity was significantly elevated at 6 dpi ( Figure 5B ). We inferred that the host enhanced the levels of CAT and GST activities in response to V. ceranae-caused oxidative stress in the middle stage of the infection process, thereby maintaining homeostasis of the internal environment. SOD catalyzes the dismutation of superoxide radicals into hydrogen peroxide and oxygen, thereby regulating ROS levels and preventing oxidative damage to cells (51). A significant decline in the activities of both SOD and GST was detected at 10 dpi, and CAT activity was reduced at both 9 and 10 dpi ( Figures 4A , 5 ). These results suggest that these three critical antioxidant enzymes are inhibited in the host midgut. This phenomenon may be attributed to the escalating spore load in the late infection stage, leading to a prolonged intensification of V. ceranae stress in the host and consequently suppressing the activities of SOD, CAT, and GST. PPO is involved in the melanization process, a common immune response in insects that involves the production of dark and pigmented structures that can trap and kill invading pathogens or parasites. In addition to its role in melanization, it simultaneously induces cellular and humoral immunity (52). Roberts and Hughes (53) found a negative correlation between PPO levels in the midguts of A. mellifera workers and V. ceranae infection intensity. Intriguingly, although the level of PPO activity in the V. ceranae-inoculated group was always lower than that in the uninoculated group, there was no significant difference between the groups ( Figure 4B ), indicating that PPO activity in the host midguts was not affected by fungal invasion. It is inferred that, as the host of origin of V. ceranae, A. cerana workers potentially evolved a strategy to counteract the negative influence of microsporidians on PPO activity. Further studies are required to explore these underlying mechanisms. Insects exhibit dynamic changes in GST and CAT in response to pathogen infection. For instance, in the larvae of Holotrichia parallela, following joint infection of entomopathogenic nematode and Bacillus thuringiensis, there was an initial increase in GST and CAT activity during the early stage, followed by a decrease in activity in the later stage (54). In this study, the contrasting alterations observed in GST and CAT levels between the middle and late stages of V. ceranae infection could potentially be ascribed to the insects’ initial augmentation of antioxidant enzyme activity as a mechanism of defense against the fungal invasion. Conversely, a decline in enzyme activity may occur in the later stage as a result of energy depletion or other physiological adjustments.

In a previous study, Higes et al. (39) detected that, 7 d after V. ceranae infection, the majority of epithelial cells showed evidence of degeneration, such as the presence of vacuoles in the cytoplasm or a brightly stained nucleus. Few unaltered cells were also observed, and either at the tips or at the bottom of the epithelial folds, there were cells containing V. ceranae intracellular stages. Here, we observed that the number of V. ceranae in the A. cerana worker midgut epithelial cells increased from 7 to 10 dpi ( Figure 6 ), which is consistent with the findings of Higes et al. (39) Collectively, these results offer a solid basis for further investigations of V. ceranae infection in A. cerana workers, such as functional studies on fungal virulence factor-associated genes and the development of an RNAi-based control strategy.

V. ceranae infection damaged the structure of midgut epithelial cells in A. mellifera workers (7). Based on microscopic observations by optical microscopy and transmission electron microscopy, Fries et al. (2) first reported that V. ceranae-infected ventriculus cells of A. cerana workers were shed into or burst open into the gut lumen where mature spores were released, and described the size of the spores. Darkly stained microsporidia were clearly detected in the midgut epithelial cells of V. ceranae-inoculated workers at 7–10 dpi ( Figure 6 ). Additionally, the boundaries of the midgut epithelial cells in V. ceranae-inoculated workers were blurred, and the nuclei almost disappeared, with nucleic acid substances dispersed around the cytoplasm ( Figure 6 ). These results suggest that V. ceranae infection also causes serious damage to the structure of the midgut epithelial cells of A. c. cerana workers. Collectively, the findings from previous studies (2) and this study offer experimental evidence for the damage caused by V. ceranae infection to the structure of A. c. cerana worker midgut epithelial cells, laying the foundation for further investigation of the pathogenesis of bee nosemosis.

Paris et al. (55) assessed the survival rate and eating behavior of A. mellifera workers infected with V. ceranae for 1–22 d, and the results showed that the survival rate of infected workers was significantly lower than that of uninfected workers. We previously observed that the mortality rate of A. mellifera workers inoculated with V. ceranae spores gradually increased over time, and that the mortality rate of V. ceranae-inoculated workers was significantly higher than that of uninoculated workers at 7 and 10 dpi (33). It has been documented that V. ceranae infection significantly increased the mortality rate of A. cerana workers (56). In this study, the survival rates of workers in both the V. ceranae- and un-inoculated groups were high at 1–5 dpi ( Figure 7 ), implying that V. ceranae had little influence on host longevity at the early stage of infection. However, at 5–20 dpi ( Figure 7 ), the survival rate of V. ceranae-inoculated workers was always lower than that of uninoculated workers, and the difference was significant at 11–20 dpi, indicating that V. ceranae infection shortened the lifespan of A. cerana workers, similar to the findings of Huang et al. (56). In conclusion, V. ceranae can negatively affect the midgut epithelial cell structure and lifespan of A. c. cerana workers during infection.

Collectively, the findings of the present study indicate that the spore load continuously increases as V. ceranae proliferates in the midgut of A. c. cerana workers, giving rise to continuous energetic stress for the host, apparent damage to the host midgut epithelial cell structure, induction of CAT and GST activities in the host midgut at the middle stage, and inhibition of the activities of SOD, CAT, and GST at the late stage of infection, which ultimately shortens the host lifespan ( Figure 8 ).