*3.3. Immunohistochemical Analyses*

Varieties of apoptotic midgut cell deletions were observed in bees exposed to different HMF concentrations for 10 days. The amount of immunohistochemically positive cells was not dependent on HMF concentration between 5 and 30 days (Figure 4). However, in bees exposed to the lowest dose of HMF, cell death was observed in the epithelium of the midgut (sporadic positive cells) but at a level below that of bees fed higher concentrations of HMF. In contrast, the level of cell death in the midgut epithelium of bees fed with candy containing 500 mg/kg of HMF in Apifonda remained high throughout the bioassay (Table 1).

**Table 1.** Results of the semi-qualitative analyses of midgut cell death using the In Situ Cell Death Detection Kit (ISCDDK) immunohistochemical method. The results represent the percentage of positive cells in midgut tissue treated with ISDDK: (I) individual positive cells; (II) range between 0% and 20% positive cells; (III) range between 21% and 60% positive cells; (IV) range between 61% and 90% positive cells; (x) no samples available. Treatment groups: 1. Untreated control; 2. 100 mg/kg of HMF in Apifonda; 3. 500 mg/kg of HMF in Apifonda; 4. 1000 mg/kg of HMF in Apifonda; 5. 1500 mg/kg of HMF in Apifonda.


**Figure 4.** Midgut of formalin-fixed, paraffin-embedded tissue of worker bees exposed to HMF. Detection of programmed cell death by TUNEL using the In Situ Cell Death Detection Kit. Red azo-dye staining is localized in midgut epithelial cell nuclei (arrows). (**A**) 100 mg/kg of HMF in Apifonda diet (5 days), magnification 200×; (**B**) 500 mg/kg of HMF (5 days), magnification 200×; (**C**) 100 mg/kg of HMF (10 days), magnification 100×; (**D**) 1500 mg/kg of HMF (10 days), magnification 100×; (**E**) control group (15 days), magnification 200×. (**F**) 1000 mg/kg of HMF (15 days), magnification 200×; (**G**) 500 mg/kg of HMF (20 days), magnification 400×; (**H**) control group (5 days), magnification 400×. The dosage of HMF was incorporated to 1 kg of Apifonda candy.

Notably, the proportion of positive midgut digestive cells in the bees treated with 100 and 500 mg/kg HMF in Apifonda (Figure 4A,B) was much lower than in the groups exposed to 1000 or 1500'mg/kg HMF. In bees exposed to 100 mg/kg of HMF, a ffected midgut epithelial cells were vacuolated with apical cell fragments released into the lumen (Figure 4A). Similar vacuolization was observed in bees exposed to 500 mg/kg HMF in Apifonda (Figure 4B). Midgut cells with positive red reaction products in the epithelium apical region were also seen in bees fed with candy containing 100 or 1500 mg/kg of HMF for 10 days (Figure 4C,D) and also in bees fed with candy containing 1500 mg/kg of HMF for 15 days (Figure 4F). Midgut epithelial cells were still intact and attached to the basal membrane even at the highest HMF concentration over the course of the experiment. In the control group of untreated bees (without added HMF), sporadic midgut cells with specific red reaction products were found (Figure 3EH). In live bees exposed to any HMF dose for 25 or 30 days, no morphological alterations or increased levels of positive ISCDDK cells were seen (Figure 4G). Bees fed with the two highest HMF concentrations, 1000 or 1500 mg/kg of HMF, and surviving to 25–30 days displayed few midgut cells with positive reaction products, an observation similar to the reaction levels displayed by untreated control bees. Taken together, the highest levels of ISCDDK-positive midgut epithelial cells were found in bees fed with 500, 1000, or 1500 mg/kg of HMF in Apifonda from 15 to 20 days. After that time, there was a notable decrease in the proportion of midgut cells with specific red reaction products. The most persistent and high levels of apoptotic cells throughout the experiment were found in bees fed with 500 mg/kg of HMF. The proportion of apoptotic cells remained high when compared with all treatment groups, including the controls. We did not include *Nosema*-infested bees in ISCDDK assays in order to exclude the e ffect of parasites on midgut cell death. The lowest HMF dose e ffect at the first and second sampling dates (5 and 10 days) was demonstrated by red reaction products in midgut epithelial cells (Figure 4A,C).
