**4. Material and Methods**

#### *4.1. Pathogen Screening*

The number of bees present in each sample (around 300 bees) was counted, and the bees were examined individually to detect the presence of *Varroa* mites and collect them by means of sterile tweezers [14]. Each mite detected was analyzed macroscopically to confirm the species. A honey bee colony was considered infested with *V. destructor* when at least 1 *Varroa* mite was found in the sample. The rate of infestation of the bee colony was estimated by assessing the number of *Varroa* mites in relation to the number of adult bees in each sample, and it was expressed as the number of *Varroa* mites/100 bees/sample [69].

The presence of *Nosema* spp., Trypanosomatids, Neogregarines, and *Acarapis woodi* was evaluated in a sample (*n* = 60) from each colony. The remaining bees were kept frozen at −80 ◦C.

The presence of different viruses was only analyzed in the surviving colonies sampled because the viral RNA integrity could not be ensured in the dead colonies.

The subsample of each colony (*n* = 60) was macerated in 50% AL buffer (Qiagen GmbH, Hilden, Germany) before DNA and RNA were extracted, as detailed in [69–71].

Briefly, macerated bees were centrifuged at 3000 rpm for 10 min, and the resulting pellets were used for DNA extraction, and the supernatants for RNA extraction. Both the pellets and supernatants were stored at −80 ◦C prior to nucleic acid (DNA or RNA) extraction. For DNA extraction, the pellets were resuspended in 3 mL MilliQH2O, and a 400 μL aliquot was transferred to a 96-well plate (Qiagen) with glass beads (2 mm diameter, Merck KGaA, Darmstadt, Germany) using disposable Pasteur pipettes. After overnight preincubation with proteinase K (20 μL, Qiagen), the samples were then processed as described previously [71] following the BS96 DNA Tissue extraction protocol in a BioSprint station (Qiagen). The plates were then stored at −20 ◦C. For RNA extraction, 400 μL of the supernatant was incubated for 15 min with protease (20 μL, Qiagen) at 70 ◦C, and the nucleic acids were then extracted as described above (BioSprint 96 DNA in, BioSprint workstation, Qiagen). The total nucleic acids recovered were then subjected to DNA digestion with DNase I (Qiagen) to completely remove any genomic DNA, and the total RNA recovered was used immediately to generate firststrand cDNAs using the QuantiTect Reverse Transcription Kit (Qiagen) according to the manufacturer's instructions. The resultant cDNA was used for subsequent virus analysis with no further dilution [70].

Negative and positive controls were run in parallel for each step: bee maceration, DNA and RNA extraction, and reverse transcription [69].

Published PCR or RT-PCR protocols were considered to screen *Nosema apis* and *N. ceranae* [71], Trypanosomatids and Neogregarines [72,73], *Acarapis woodi* [74], LSV complex [75], AKI complex [76], DWV, and BQCV [77]. Table S2 includes the primers used.

In addition, the proportion of *Nosema* spp. infection was determined by PCR on 25 individual worker honey bees from each colony sampled [71].

#### *Test of* Varroa *Mite Resistance to Acaricides*

When possible, mite resistance to acaricides was determined using the respective marketed products, CheckMite® (a.m.: coumaphos), Apistan® (a.m.: tau-fluvalinate), and Apitraz® (a.m.: amitraz), according to the protocol described previously [17] with the following modifications:


The honey bees were kept at 35 ◦C during the test, and after incubation periods, a control test without treatment was used to determine how the basal conditions affected *Varroa* mite mortality.

### *4.2. Stored Pollen Analysis*

Beebread was extracted aseptically from the combs, removing the wax and preparing a composite sample for each colony. Finally, each pollen sample was divided into two 100 g aliquots, for chemical and palynological analyses, and stored at −80 ◦C.

A multiresidue chemical analysis of 60 substances was carried out following a method described elsewhere [59], assessing acaricides (AC), fungicides (FU), herbicides (HB), and insecticides (IN). In addition, 7 neonicotinoid INs (acetamiprid, clothianidin, dinotefuran, imidacloprid, nitenpyram, thiacloprid, and thiamethoxam) were measured as described previously [78].

Based on the results of the multiresidue analysis and the toxicity data reported previously (Table A2 in [14]), the toxic unit of the mixtures (TUm) was calculated following an approach given elsewhere [14] to assess the risk of the chemical mixture found in each hive sampled. Subsequently, the natural logarithm (Ln(TUm)) was estimated for comparison purposes with [14]. In that work, the possible relationships between TUm and the prevalence of pathogens were studied by using a factor analysis. The natural logarithm

of TUm, Ln(TUm), was derived to normalize data. Thus, the higher the value of Ln(TUm), the higher the toxicity risk.

Finally, the type of foraging flora was confirmed by analyzing beebread samples as described elsewhere [11,20,21] and estimating the proportion of pollen from wild (WP) and cultivated (CP) plants. The pollen of the beebread was extracted by diluting 0.5 g in 10 mL of acidulated water (0.5% sulfuric acid) and centrifuging at 2500 r.p.m. for 15 min. The pellet was washed with double-distilled water and centrifuged twice. The sediment was placed onto a glycerine jelly slide and examined microscopically in order to identify the pollen. The frequency of the pollen grains of each taxon is expressed as a percentage of the total pollen grains. Between 300 and 1200 pollen grains were counted in each sample.

The pollen grains were identified and classified on the basis of the identification keys [79,80] and the pollen slide reference collection available at the honey laboratory at the CIAPA.

### *4.3. Meteorological Data*

A Walter–Leith diagram [81] was developed with historical weather data obtained from the meteorological station of Brihuega (lat. 40.765, long. −2.874) from the network of the State Meteorological Agency (AEMET) to compare it with the data for the period of the study. This weather station is located 9 km far away from the studied areas.

Walter and Leith climate diagrams are brief summaries of average climatic variables and their time course. They illustrate precipitation and temperature changes throughout the year in 1 standardized chart. Originally aimed at visualizing those climatic variables and their dynamics, which are particularly important for vegetation, they have proven useful for a wide range of sciences. The diagrams were developed with the diagnostic tool of the Worldwide Bioclimatic Classification System, 1996–2021 [82].

### *4.4. Statistical Analysis*

A 1-tailed Mann–Whitney test (α = 0.05) was used to analyze possible differences between the dead and surviving colonies in terms of the different experimental parameters measured (pathogens, % wild pollen in beebread, chemical residues, and TUm). The analysis was carried out with Statgraphics Centurion 18©.

#### **5. Conclusions**

The veterinary inspection and analytical evidence presented here indicate that nosemosis C infection was the underlying cause of the colony weakness and collapse of the professional apiary studied, probably accelerated by the presence of high levels of miticides and unusual climatic conditions. In conjunction with the unchecked concentrations of acaricide that accumulated in honey bee hives, *N. ceranae* infection represents a real danger in honey bee colony survival. Therefore, in addition to the correct use of veterinary products to control *V. destructor*, appropriate wax renewal of the combs should be introduced to develop specific preventive strategies aimed at controlling possible infections from prevalent pathogens.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/pathogens10080955/s1, Table S1: Details of calculation of TUm values. Table S2: Primers used for each pathogen in PCR reactions. Figure S1: Main taxa identified of the pollen grains found in the beebread samples.

**Author Contributions:** Conceptualization, M.H., E.A.-P., R.M.-H.; methodology, M.H., E.A.-P., R.M.-H., J.L.B., J.B., A.-V.G.-P.; formal analysis, M.H., E.A.-P., R.M.-H., J.L.B., J.B., A.-V.G.-P.; writing original draft preparation, M.H., E.A.-P., R.M.-H., J.L.B., J.B., A.-V.G.-P.; writing—review and editing, M.H, E.A.-P., R.M.-H, J.L.B, J.B., A.-V.G.-P.; visualization, M.H., E.A-P., R.M.-H.; supervision, M.H., E.A.-P., R.M.-H.; project administration, M.H., J.L.B.; funding acquisition, M.H., J.L.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by INIA-FEDER (RTA2017-00004-C02).

**Institutional Review Board Statement:** In Europe, the EU Directive 2010/63/EU on the protection of animals used for scientific purposes laid the down the ethical framework for the use of animals in scientific experiments. The scope of this directive also includes specific invertebrate species, i.e., cephalopods, but no insects. Thus, according to European legislation no specific permits were required for the described studies.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** The authors wish to thank V. Albendea, T. Corrales, M. Gajero and C. Uceta at the honey bee pathology laboratory for their technical support. The authors also want to thank M.D. Moreno for his support in the visit to the apiaries.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

