Success of Selection in Terms of Hygienic Behavior in Struggle Against DWV and Varroa

The Honey bee (Apis mellifera L.) is the most important pollinator for the scientific world because of its role in pollination (Ryabov et al., 2014; McMenamin & Genersch, 2015; Tantillo et al., 2015) and they produce such economically important products as honey, wax, pollen, propolis, royal jelly and bee venom (Popovska et al., 2018; Sforcin, Bankova, & Kuropatnicki, 2017).

Hygienic behavior is a potential struggle mechanisms against V. destructor (Boecking, Bienefeld, & Drescher, 2000; Ibrahim & Spivak, 2006), a resistance mechanism against viruses or mites and a natural response of workers’ (Harbo & Harris, 2001; Cheruiyot et al., 2018), and is also a heritable trait, not a learned behavior (Al Toufailia et al., 2018). Resistant honey bees remove mites from the brood before bees are infected with the diseases (Gramacho & Spivak, 2003; Al Toufailia et al., 2018), and over the last five decades this trait has been extensively studied and improved in several research and breeding programs (Spivak 1996; Harbo & Harris, 2001). Breeding programs have shown that selection for a specific trait might also impact other traits. For example, Ibrahim et al. (2007) found fewer mites in hybrid honey bee colonies (hygienic “HYG” X suppression of mite reproduction “SMR”) compared to HYG colonies but no difference in the fertility and number of viable female offspring among the lines. These results suggest that breeding for SMR may improve multiple traits rather than just hygienic behavior. On the other hand, it was reported that bees raised for SMR show hygienic behavior and adult bees selectively remove infected pupae (Ibrahim & Spivak, 2006). However, the benefits of improving hygienic behavior have yet to be clarified along with whether there are any costs associated with the trait at the colony level (Leclercq et al., 2017). This is also a source of concern primarily for honey-trading countries including Turkey. For over ten years, there have been numerous reports of significant losses of honey bee colonies in twenty-four worldwide (Neumann & Carreck, 2010; Spleen et al., 2013; Steinhauer et al., 2014; McMenamin & Genersch, 2015) including Turkey (Giray et al., 2007; van der Zee et al., 2012; Desai, Kumar, & Currie, 2016), and the latest and unresolved crisis has been associated with bee viruses, including DWV (Organtini et al., 2017), reported up to (de Miranda, Cordoni, & Budge, 2010; Remnant et al., 2017). Viruses considerably affect the health of honey bees and shorten their lifespan by infecting the different life stages including eggs, larvae, pupae and adults (Chen & Siede, 2007) and also can be transmitted vertically or horizontally between colony members in a short time (Chen, Evans, & Feldlaufer, 2006). V. destructor is the most harmful parasite for A. mellifera (Solignac et al., 2005), although its natural host is Apis cerana (Le Conte, Ellis, & Ritter, 2010). Importantly, honey bee viruses and V. destructor are often thought to be related to colony losses in recent years (Le Conte, Ellis, & Ritter, 2010). In the absence of V. destructor, many viruses of honey bee cause latent infections which have no clinical symptoms in infected bees or colonies (Gisder, Aumeier, & Genersch, 2009; de Miranda & Genersch, 2010) because of a viral vector. The presence of V. destructor is associated not only with the prevalence but also with the course of DWV infection (Francis, Nielsen, & Kryger, 2013), and in its absence the virus usually cause latent infection (Lanzi et al., 2006; Yue et al., 2007; McMenamin et al., 2018) due to its synergy with V. destructor (Chen & Siede, 2007; Simeunovic et al., 2014; Ryobov et al., 2014; Khongphinitbunjong et al., 2015; Di prisco et al., 2016; Nazzi & Pennacchio, 2018; Evans & Cook, 2018).

Although several diagnostic techniques have been used to detect honeybee viruses, it is very difficult or impossible to differentiate and quantify virus infections through field symptoms (Desai, Kumar, & Currie, 2016). Reverse transcriptase-quantitative PCR (RT-qPCR), which is 1000 times more sensitive than ELISA and 100 times more sensitive than conventional non-nested RT-PCR (de Miranda, Cordoni, & Budge, 2010), can be rapidly implemented in independent laboratories after the preparation of primary protocol and primer sequences (Simeunović et al., 2014). RT-qPCR has been preferred more recently with TaqMan (Cirkovic et al., 2018) due to highly reliable, fast and accurate results (Chantawannakul et al., 2006) even if a little bit expensive. In the present study, we investigated the success of the selection in terms of hygienic behavior in the struggle against DWV and Varroa in colonies which are under breeding programmes for hygienic behavior.

To determine whether the selection in terms of hygienic behavior is effective on V. destructor and DWV, A. m. anatoliaca honey bee workers and pupae samples were collected from the Mugla region showing four generations of hygienic and unhygienic behavior of the colony in Turkey. DWV was detected in most of the studied samples, and the DWV loads and Cq values in RT-qPCR for both worker bees and pupae in the control and hygienic colonies were summarized graphically in the Fig. 1a, 1b. Fig. 1a shows that the DWV loads in the worker bees and pupae taken from the control group are a little higher than in the hygienic colonies. Moreover, as seen clearly in Fig. 1b both the worker bees and the pupae taken from the control group gave signals earlier in terms of DWV in RT-qPCR than the samples taken from the hygienic group. As seen from Tab. 1, DWV loads (p<0.01) and Cq values (p<0.05) in worker bees are significantly different between the hygienic and control groups. Interestingly, the prevalence of DWV in pupae (69.0%) taken from the hygienic colonies is higher than the pupae taken from control colonies (60.8%). However, the number of worker bees (66.0%) infected with DWV from the control colonies was higher than that from the hygienic colonies (41.3%), and these differences with regard to virus prevalence between hygienic and control colonies were calculated to be significant by using Chi-square test (p<0.01). According to t-test results, the averages of Varroa counts for the hygienic colonies (28.92) was significantly (p<0.01) lower than that of the control colonies (108.90). However, we could not find any correlation between Varroa loads and DWV loads of both worker and pupae in this study.

Fig. 1

In recent decades in Turkey, Gülmez, Bursalı, & Tekin (2009) and Tozkar et al. (2015) have studied Deformed Wing Virus, and Muz & Muz (2009, 2017) studied how both DWV and V. destructor interaction caused colony collapse in Hatay and Tekirdağ regions. There have been more studies about the coexistence of DWV and Varroa world-wide (Martin 2001; Berényi et al., 2006; Yue et al., 2007; Nielsen, Nicolaisen, & Kryger, 2008; Gisder, Aumeier, & Genersch, 2009; Locke, Forsgren, & de Miranda, 2014). DWV has been shown to have different concentrations in healthy colonies in the highest percentage of pooled samples of worker bees and at high levels in unhealthy colonies (Desai, Kumar, & Currie, 2016). DWV was the most prevalent virus in Austria, present in 91% of samples (Berényi et al., 2006), and Nielsen et al., (2008) reported that hives with a high rate of winter mortality had 57% DWV infection. Baker & Schroeder (2008) informed of a 97% prevalence of DWV in the A. mellifera L. colonies without any indication of infection. RT-qPCR has been preferred in many studies due to its being reliable, specific, and sensitive to determining honey bee viruses (Cirkovic et al., 2018; Nazzi, & Pennacchio, 2018). Simeunovıć et al. (2014) obtained a high frequency of DWV (76.4%) using TaqMan-based real-time RT-PCR technique in asymptomatic colonies. Although, there was no sign of any disease in the colonies examined in our study, the number and prevalence of colonies infected with DWV were calculated to be considerably high using TaqMan-based real-time RT-PCR (Tab. 1).

A comparison of colonies showed that DWV prevalence was higher in pupae (69.0%) but in contrast lower in worker bees (41.3%) from hygienic colonies (Tab. 1). This supports a previous study by Gauthier et al. (2007) who suggested that pupae were more susceptible to DWV replication and found differences between DWV loads in adults and pupae, which might result from the fact that diseased adults have much shorter lifespans. Moreover, depending on the higher hygienic behavior of hygienic colonies, bees could remove the infected pupae from colonies, and by doing so, the number of healthy adults might be higher which would contribute to a lowering of virus concentration in hygienic colonies Gauthier et al. (2007). We also found the amount of DWV of hygienic colonies to be lower than the control (Tab. 1). Yañez et al., (2015) and Li et al., (2012) drew attention to the higher prevalence of DWV in A. mellifera compared to A. cerana, which Chanpanitkitchote et al. (2018) interpreted as a strong ability of A. cerana to resist both viruses and mites. Rsistance, similar to that of A. cerana against DWV and Varroa, can also be demonstrated by A. mellifera, which has been developed in terms of hygienic behavior.

Spivak & Reuter (1998) compared hygienically selected and unselected A. m. ligustica colonies and showed that hygienically behaving bees had been successfully removed from frozen slaughtered pupae and parasites and that brood diseases were at very low levels or not at all. It has been reported that honey bee colonies can be easily selected for hygienic behavior as a result of hygienic behavior tests conducted by land experiments (Spivak & Gilliam, 1998) or direct pathogen (Gilliam et al., 1988) and mite (Boecking & Drescher, 1990) practices. However, all the effects of hygienic behavior have not yet been fully explained.

The virus prevalence was unexpectedly higher at the pupae phase in the hygienic colonies in the present study. While hygienic colonies have fewer viruses in their worker bees and pupae, it is not known why the virus prevalence is higher in terms of pupae in hygienic colonies than the control (Tab. 1). While there are a few possible responses to explain this situation, the sperm pool may be used in artificial insemination to obtain hygienic colonies. Namely, when sperm is collected, the somatic cells of virus-infected drones might also be collected. Even if there is no virus in the sperm, somatic cells can carry the infection to the pool. If there is an infection even in a single drone, the sperm pool used to fertilize the queens may be contaminated. More populations may have been infected with viruses due to fertilized queens using the sperm pool in the hygienic colonies. Results that support this have been pointed out in previous studies that viruses can translate horizontally through individuals in the same generation and can be transferred from one generation to another vertically through their queens (Chen, Evans, & Feldlaufer, 2006; Yue et al., 2007; Tantillo et al., 2015). The venereal passing of DWV through artificial insemination with infected semen in DWV-free virgin queens has been already testified fully by de Miranda & Fries (2008). The key result of the present study may be an indication that hygienic colonies remove the infected pupae from the hive and thereby reduce the spread of the virus in adult bees, and if so, this is a success of hygienic colony breeding.

Like all other bee viruses, DWV is harmful for beekeeping and is considered the most serious secondary pathogen associated with varroosis (Yue & Genersch, 2005), although many studies have concluded that the Varroa mite is a vector for honeybee viruses including DWV (Tentcheva et al., 2006; Gauthier et al., 2007; Carreck, Ball, & Martin, 2010; Simeunovic et al., 2014). Because we could not reveal the exact number of mites in the colonies using the bottom-board method, we could not find any correlation between Varroa loads and DWV loads of both workers and pupae in this study. The selected colonies were shown to be better in terms of virus and Varroa due to increased hygienic behavior in this study, and as a result the selection for hygienic behavior is effective on viruses and mite loads. On the other hand, in this project, the surveyed region is an important beekeeping center, because Turkey is one of the world's largest honey producers (Can et al., 2015) and also because 90% of the world's pine honey is produced in the Muğla region (Miguel, Pukkala, & Yesil, 2014; Özkök et al., 2017). In addition, the distribution and relative concentration of viruses within bee colonies in Muğla are still poorly understood. For these reasons, these findings may be important for researchers and be evaluated as an indicator for practices.