Innate Defense Mechanisms Against Nosema ceranae in Hygienic Honey Bee ( Apis mellifera ) Colonies

16 The honey bee colony (Apis mellifera) acts as a superorganism, with a dual immune system 17 that operates at the individual and social level. However, the linkages between immune 18 mechanisms across the two levels remain poorly understood, despite the relevance for 19 developing effective breeding strategies to improve honey bee disease resistance. Hygienic 20 behavior involving the removal of unhealthy brood is a key component of honey bee social 21 immunity and is highly effective at limiting parasites and pathogens in the colony. While 22 this form of hygienic behavior can reduce brood diseases, parasites infecting adult bees 23 primarily, such as Nosema ceranae, are not directly impacted by the behavior. However, 24 when using the Unhealthy Brood Odor (UBeeO) assay to quantify hygienic behavior 25 performance, hygienic colonies have been shown to maintain lower Nosema spp. loads over 26 time and overall compared to non-hygienic colonies. To investigate the mechanisms driving 27 reduced Nosema spp. in hygienic colonies, we conducted a series of field and lab 28 experiments to test the innate immune performance of individual bees. We evaluated 29 several factors across hygienic and non-hygienic bees including (1) differences in N. 30 ceranae infection levels, (2) survival probability, (3) Vitellogenin and Hymenoptaecin gene 31 expression, and (4) amount of N. ceranae inoculant consumed. We found that hygienic bees 32 consumed less of the inoculant, exhibited upregulated Vitellogenin gene expression at peak 33 N. ceranae infection, showed a positive relationship between Hymenoptaecin gene 34 expression and N. ceranae infection levels, and had greater survivability when infected 35 with N. ceranae, compared to non-hygienic bees. Here, we present new findings that link 36 colony hygienic behavior performance to individual-level resistance and tolerance 37 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint mechanisms in response to N. ceranae, suggesting broader implications for the success of 38 selective breeding programs targeting hygienic traits. 39 40 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint

41 Pests and pathogens are a primary threat to honey bees (Apis mellifera) impacting the 42 health of brood and adult bees and contributing to overall colony decline. In response to 43 intruders, the honey bee colony acts as a superorganism, with a dual immune system that 44 operates at the individual and social level [1–3]. Honey bees rely on their innate immune 45 system (e.g. physical barriers, cellular and humoral immunity) to defend against infection 46 as well as complex social behaviors that reduce the impacts of parasites and pathogens in 47 colonies. Understanding the linkages between immune function at the social and individual 48 level is essential for informing effective selective breeding strategies aimed at improving 49 honey bee disease resistance and colony survival. 50 51 Hygienic behavior refers to the enhanced ability of worker bees to respond to chemical 52 odorants emitted by diseased or dead brood (developing larvae and/or pupae) by uncapping 53 and removing pupae from the nest [4,5]. The form of hygienic behavior involving the 54 removal of unhealthy brood from the hive should be distinguished from other forms of 55 honey bee hygiene, such as known auto- and allo-grooming behaviors performed by adults 56 [1,6]. As a heritable genetic trait, hygienic behavior is among the most important social 57 behaviors for conferring colony-level resistance against brood diseases [7–9] and in recent 58 years has become a major focus in honey bee breeding programs. Previously developed 59 assays used to quantify hygienic behavior (e.g. pin prick, freeze-killed brood) are based in 60 necrophoric activity and have shown to confer reduced levels of Foulbrood, Chalkbrood, 61 and Varroa infestations [10–13]. As an improved method for quantifying hygienic 62 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint behavior, the Unhealthy Brood Odor (UBeeO) assay challenges bees with synthetic 63 pheromones mimicking the natural odors emitted by live, parasitized brood. In addition to 64 predicting a low incidence of brood disease, the UBeeO assay has been shown to predict 65 lower spore loads of Nosema spp. over time and overall, in hygienic colonies compared to 66 non-hygienic colonies [14]. 67 68 Nosema ceranae is a common microsporidian endoparasite that infects the midgut 69 epithelial cells [15,16] of adult bees [17,18]. Nosema ceranae infection negatively impacts 70 honey bee health at the individual level—causing nutritional and energetic stress [19,20], 71 immunosuppression [21,22], altered behavior [23,24], reduced lifespan, and inhibition of 72 host cell apoptosis [25–27]—which can reduce colony fitness by lowering brood numbers 73 and honey production, and in severe cases, lead to colony death [15,28]. With limited 74 viable treatment options available to beekeepers, effective prevention and colony 75 management remain essential for controlling the pathogen [28]. Moreover, the risk of target 76 pests and pathogens building resistance to chemicals and rendering treatments ineffective— 77 as seen in global Varroa populations resistant to several well-known acaricides [29–31]— 78 further underscores the need for more sustainable interventions to control honey bee pests 79 and diseases. 80 81 Nosema ceranae is primarily an adult bee pathogen [32,33] that has only been found to 82 infect brood through manual inoculation in lab studies [34,35] or at extremely low 83 prevalences (1-3%) in natural hive settings [36,37]. While N. ceranae infection in 84 developing brood has not been thoroughly evaluated, many studies have reported an 85 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint absence of N. ceranae infection in emerging adults [17,38,39], suggesting that brood does 86 not normally become infected in the hive. Since hygienic behavior acts on infected brood, it 87 has not been shown to directly inhibit Nosema spp. transmission, aside from reducing its 88 prevalence at the apiary level [14,40]. In a hygiene-based breeding program in Turkey [40], 89 researchers reported that average apiary-level hygienic behavior increased from 43% to 90 93% (n = 123), while Nosema spp. levels declined consistently from 61% to 19% in only 91 three years. Therefore, recent findings by Alger et al. [14] are not the first to demonstrate 92 an association between high hygienic behavior and low Nosema spp. incidence in a field 93 study. Nevertheless, it remains unclear how hygienic colonies maintain low Nosema spp. 94 loads and whether colony-level resistance arises from social immunity in the form of 95 pleiotropic effects on brood and adult bee hygiene, innate immune mechanisms, or a 96 combination of both. 97 98 Several co-occurring mechanisms at the individual and social level may contribute to 99 colony-level resistance to N. ceranae. At the social level, adults in hygienic colonies may 100 communicate their diseased state through stronger chemical signals, prompting detection 101 and removal by nestmates, similar to the process of removing brood [7,41]. Adults in 102 hygienic colonies may, in turn, be more sensitive to atypical odors and better able to detect, 103 isolate, and/or discard of N. ceranae-contaminated individuals and/or food sources in the 104 hive. To better understand the social dynamics of N. ceranae-infected bees in hygienic 105 colonies, it is necessary to first evaluate their innate performance against N. ceranae. 106 Previous studies have shown no genetic tradeoffs between hygienic behavior and innate 107 immunity [42]. In fact, hygiene-performing bees have been associated with modifications 108 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint to the gut microbiome [43] and higher expression of antimicrobial peptides [44], which 109 may lead to more efficient immune responses against invading pathogens. Therefore, 110 individuals in hygienic colonies may exhibit stronger innate immunity and improved 111 performance under pathogen stress overall. Specifically, they might lessen the negative 112 health impacts of N. ceranae infection by upregulating immune-related genes in cellular or 113 humoral pathways, which could help limit pathogen invasion or tissue damage [12,42]. 114 115 Two major immune genes that are likely responsible for enhancing innate performance 116 against N. ceranae are Hymenoptaecin and Vitellogenin. Hymenoptaecin (Hym) is an 117 antimicrobial peptide activated by the humoral immune system (Imd pathway) that directly 118 resists pathogens by attacking their cell membranes [45]. Vitellogenin (Vg) is an egg yolk 119 precursor protein that can repair tissue damage [46,47] and perform immunological defense 120 functions against pathogens and reactive oxygen species [48,49]. Vitellogenin also 121 influences multiple physiological functions in honey bees including behavioral maturation 122 [50], social organization [51], longevity [52], and egg development [49]. Both Hym [22,53] 123 and Vg are commonly downregulated in honey bees infected with N. ceranae or other 124 parasites [54,55]; therefore, may play a central role in resisting N. ceranae infection in bees 125 from hygienic colonies. 126 127 In this study, we investigated innate immune mechanisms that may enhance individual 128 performance against N. ceranae infection and help explain the reduced N. ceranae loads 129 observed in hygienic colonies in previous field studies. We compared bees from hygienic 130 and non-hygienic colonies by evaluating (1) N. ceranae infection levels, (2) survival 131 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint probability, (3) Vitellogenin and Hymenoptaecin gene expression, and (4) amount of N. 132 ceranae inoculant consumed. Our objective was to identify individual-level traits 133 associated with bees from hygienic colonies, which could suggest broad-spectrum disease 134 management potential in hygiene-based breeding scenarios. 135

136 Pupal evaluations 137 Pupal samples were collected from Nosema spp.–infected honey bee colonies in St. Albans, 138 Vermont and analyzed for spore presence to determine whether developing pupae 139 experience infection under natural hive conditions. Since hygienic behavior targets 140 unhealthy brood, it was important to determine whether pupae serve as a source of Nosema 141 spp. infection in our target honey bee population and whether removing infected brood 142 could help reduce pathogen loads in the colonies. Thirty pupae were collected in composite 143 samples from each of 28 colonies with detectable Nosema spp. loads in nurse bees ranging 144 from 5×10⁴ to 1.4×10⁶ spores per bee. Pink to purple-eyed pupae were extracted from their 145 wax cell with forceps and stored at -80ºC until processing [36]. 146 147 To conduct spore counts on pupae, composite pupal samples were rinsed in phosphate 148 buffered saline and pulverized in a plastic bag using a rolling pin. One mL of distilled water 149 per pupa was added and allowed to settle for 45 s. Ten µL was transferred from the stock 150 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint solution onto two haemocytometer (Improved Neubauer) counting chambers. Spores were 151 counted under 40× magnification and converted to spores per pupae [56]. 152 Incubation and N. ceranae isolation 153 To compare hygienic and non-hygienic bees in our innate immune-response trials, we 154 obtained newly emerged adults from hygienic and non-hygienic colonies. Throughout the 155 text, we use the terms hygienic bees and non-hygienic bees to refer to individuals 156 originating from hygienic and non-hygienic colonies, respectively, rather than bees actively 157 performing hygienic behaviors. We constructed frame cages to house deep hive frames of 158 emerging brood for 1-3 days. The frame cages consisted of a wooden frame and lid with 159 8 mm mesh screened sides that provided adequate ventilation. Adult bees were transferred 160 to hoarding cages upon emergence (within ~6 hours) to avoid consumption of contaminated 161 food stores from their frames. We constructed hoarding cages to house adult workers for 162 12-14 days during N. ceranae spore inoculation and infection period. Each hoarding cage 163 consisted of a 473 mL plastic cup with ventilation holes encircling the upper and lower rim. 164 Feeders consisted of 5 mL plastic pipettes severed at the base of the bulb and secured in the 165 straw hole of the plastic cup lid. A small piece of wax foundation served as a ramp to the 166 feeder [56]. 167 168 Adult workers in hoarding cages were maintained in two separate incubators to segregate 169 N. ceranae-infected bees from non-infected control bees, both in complete darkness at 30°C 170 and approximately 60-70% RH. A thermometer/humidity gauge was used to monitor the 171 interior environmental conditions each day. Adult workers were fed a diet of 50% (v/v) 172 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint sugar syrup that was administered via a pipette feeder. Fresh sugar syrup was replaced 173 every other day during the infection period [56]. 174 175 To obtain active N. ceranae spores for inoculation, we collected foragers [38] from live 176 colonies with existing Nosema spp. loads of 11-13 million spores per bee. Foragers were 177 collected in a package cage, fed 50% (v/v) sugar syrup, and placed in an incubator until 178 processed. Nosema ceranae spores were isolated from 100 forager bees by homogenizing 179 one bee per 0.5-1 mL of water. The solution was strained through 70 μm mesh, evaluated 180 for concentration using standard microscopy and hemocytometer (Improved Neubauer), 181 and diluted into 50% (v/v) sugar syrup to achieve spore concentrations of 104 spores per 182 0.04 mL (low dose) or 5 x 104 spores per 0.04 mL (high dose). Final inoculants were fed to 183 bees the same day. Control bees received pure 50% (v/v) sugar syrup. 184 Determining Nosema spp. inoculation method 185 To determine the most effective Nosema spp. inoculation strategy for our individual 186 immune-response trials, we conducted a pilot study examining how Nosema spp. load and 187 its variability are influenced by (1) individual versus group feeding and (2) the number of 188 bees per cage under group-feeding conditions. Newly emerged adult bees were randomly 189 assigned to one of the two feeding methods and, if assigned to group-feeding, cages of 30 190 or 10 bees. All bees were starved for 2-4 hours before administering the Nosema spp. 191 inoculant. The Nosema spp. inoculant was administered to group-fed bees via ~3 mL of 192 sugar syrup containing 5 x 104 spores per 0.04 mL, ad libitum, for 24 hours [57]. We used a 193 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint pipette to administer 5 µL of sugar syrup containing 5 x 104 spores to individually-fed bees, 194 then returned the bees to their hoarding cages after 30 minutes (10 bees per cage) [33]. All 195 bees were expected to consume 5 x 104 spores by the end of the inoculation period. Bees 196 were maintained in their hoarding cages for a 10-day infection period, at which time, bees 197 were extracted from their cages and stored individually at -20ºC until processed. 198 199 To conduct spore counts on adult bees using standard microscopy, the abdomens of 200 individual bees were dissected, rinsed in phosphate buffered saline, and pulverized using a 201 1.5mL pestle for 90 s. One mL of distilled water was added and allowed to settle for 45 s. 202 Ten µL was transferred from the stock solution onto two haemocytometer (Improved 203 Neubauer) counting chambers. Spores were counted under 40× magnification and 204 converted to spores per bee [56]. 205 Unhealthy Brood Odor (UBeeO) assays 206 To identify hygienic and non-hygienic colonies from which to source newly emerged adults 207 for our innate immune-response trials, we tested 30 honey bee colonies located in Northern 208 Vermont, which were part of an existing three-year program designed to select for hygienic 209 behavior. Queens were reared and overwintered in Vermont and tested prior to the 210 experiment in early June. The queens were primarily Carniolan (Apis mellifera carnica) 211 and were not sourced from a designated “hygienic” line. No official permits were required 212 to conduct hygienic behavior testing or pathogen sampling on live colonies, other than 213 permission from Michael Palmer of French Hill Apiaries and Bianca Braman of Vermont 214 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint Bees LLC for apiary access. As per the manufacturer’s instructions, 0.5 ml of UBeeO was 215 applied to a small circular region of capped, non-emerging honey bee brood cells, and 216 hygienic response was quantified after two hours. Assay scores were calculated as the 217 percentage of the capped cells at T0 that were manipulated (any uncapping including 218 piercing) at T2. Colonies that tested over 60% were considered hygienic [58]. 219 Innate immune-response trials 220 To evaluate differences in innate immune response, behavior, and mortality between 221 hygienic and non-hygienic bees infected with N. ceranae, we selected four hygienic 222 colonies (scoring >60% on UBeeO assay) and four non-hygienic colonies (scoring <60% 223 on UBeeO assay) from which to source newly emerged adults. Ten newly emerged adults 224 per colony were collected from frame cages and tested for Nosema spp. using standard 225 microscopy (methods above) to ensure an absence of infection at the start of the 226 experiment. Replicate hoarding cages (30 bees per cage) from each colony were randomly 227 assigned a high dose (5 x 104 spores per 0.04 mL), low dose (104 spores per 0.04 mL), or 228 control (sugar only) inoculant, consumed ad libitum, for 24 hours [57]. To determine the 229 amount of sugar syrup inoculant consumed, feeders were weighed before and after they 230 were administered to each hoarding cage. Mortality was recorded for each cage, and four 231 adults were extracted every two days post-inoculation for 10-14 days to assess innate 232 immune response. Samples were stored at -80º C until processed [53]. 233 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint Nosema ceranae, Vitellogenin, and Hymenoptaecin quantification 234 To quantify active N. ceranae infection, and Vitellogenin and Hympenoptaecin gene 235 expression in bees from our innate immune-response trials, relative qPCR analyses were 236 conducted. RNA was extracted from frozen bees using Trizol (Sigma Aldrich) following 237 manufacturers' instructions and resuspended in 50 ul molecular-grade water. RNA quantity 238 and quality was assessed with a Nanodrop800 spectrophotometer. DNase treatment with 239 amplification-grade DNAseI (Thermofisher) and cDNA synthesis was performed using 240 Invitrogen SuperScript II Reverse Transcriptase (Thermofisher) with dT and random 241 priming. One μl cDNA was amplified by qPCR using SsoAdvanced Universal SYBR 242 Green Supermix (Biorad) in a 20 ul reaction as per manufacturer’s protocol. Cycling 243 parameters were the same for all targets: 95ºC for 3 min, 50 cycles of: 95ºC for 5 sec, 60ºC 244 for 30 sec, followed by a melt curve to assess product specificity. Primers for all targets are 245 listed in S1 and S2 Tables. 246 Data Analysis 247 Data analysis was conducted in R (version 4.5.2). All mixed models were fit using the 248 LME4 package (v1.1.37;[59]). The significance of main effects and interaction terms was 249 assessed using Type II Wald chi-square tests conducted using the Anova function from the 250 CAR package (v3.1.3; [60]). We used alpha = 0.05 as the threshold of significance. All 251 outliers were retained in our datasets. Non-normality and zero-inflated data were handled 252 using log transformations or modeling with the appropriate distributions. 253 254 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint Colony hygienic status was determined by converting our continuous variable of UBeeO 255 score (0.00-100%) to a binary variable where colonies were considered ‘hygienic’ if 256 UBeeO score >= 60% and ‘non-hygienic’ if UBeeO score < 60%. Prevalence was 257 calculated from presence–absence data as the proportion of bees testing positive for N. 258 ceranae or Nosema spp., by dividing the number of infected individuals by the total sample 259 size within each experimental group. Differences in prevalence among groups were 260 evaluated using a chi-square test. The terms “load” or “average load” refer to the quantity 261 of N. ceranae spores per bee. The term “levels” refers to relative quantification using ΔΔCt 262 and does not imply an absolute unit of measurement. 263 264 Relative N. ceranae infection levels and Vg/Hym gene expression were quantified using 265 the ΔΔCt method [61]. Ct values for the target genes and N. ceranae were normalized to 266 our reference housekeeping gene (Actin) to obtain ΔCt values, then compared to the mean 267 ΔCt of the control group to calculate ΔΔCt (1). Relative N. ceranae infection, Vg, and Hym 268 expression levels were log-transformed to address non-normal distributions while 269 preserving true zeros. 270 ∆Ct = Ct (𝑡𝑎𝑟𝑔𝑒𝑡 𝑔𝑒𝑛𝑒) − Ct (ℎ𝑜𝑢𝑠𝑒𝑘𝑒𝑒𝑝𝑖𝑛𝑔 𝑔𝑒𝑛𝑒) 271 ∆∆Ct = ∆Ct (𝑠𝑎𝑚𝑝𝑙𝑒) − ∆Ct (𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑎𝑣𝑒𝑟𝑎𝑔𝑒) 272 (1) 273 Inoculant Consumption 274 To examine the main effect of colony hygienic status, N. ceranae dose, and their interaction 275 effect on the amount of sugar syrup inoculant consumed by bees, colony hygienic status, N. 276 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint ceranae dose (control (0 spores), low dose (104 spores per 0.04 mL), high dose (5 x 104 277 spores per 0.04 mL)), and an interaction term were included as predictor variables in a 278 linear model. An ANOVA Type II test was performed to compute significance of terms in 279 the linear model. Estimated marginal means (EMMs) were calculated for combinations of 280 colony hygienic status and N. ceranae dose using the EMMEANS package (v1.11.2.8) and 281 pairwise comparisons of colony hygienic status were performed within each N. ceranae 282 dose. To account for the small sample size of hoarding cages in the study, non-parametric 283 Kruskal-Wallis tests were performed to further compute overall significance of colony 284 hygienic status and N. ceranae dose as main effects. 285 286 Nosema ceranae Levels 287 To test whether relative N. ceranae infection levels were influenced by colony hygienic 288 status, N. ceranae dose, sampling day, and all possible two-way interactions, we 289 constructed a linear mixed effects model. To account for repeated measures and potential 290 non-independence among individuals reared in the same hoarding cage, cage identity was 291 included as a random effect. Estimated marginal means (EMMs) were calculated for 292 combinations of colony hygienic status, N. ceranae dose, and sampling day using the 293 EMMEANS package (v1.11.2.8). Pairwise comparisons among treatment groups were 294 performed using the Tukey method in the MULTCOMP package. 295 296 Immune Gene Expression 297 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint To assess whether relative Vitellogenin (Vg) or Hymenoptaecin (Hym) gene expression 298 were influenced by main effects of colony hygienic status and N. ceranae, four additional 299 linear mixed effects models were constructed (two models per target gene). For each target 300 gene, one model included N. ceranae dose, colony hygienic status, sampling day, and all 301 possible two-way interaction terms as predictors. Another linear mixed effects model 302 included N. ceranae infection levels and colony hygienic status as main effects with an 303 interaction term, and included sampling day as an additive main effect. Cage identity was 304 included as a random effect in all models. Estimated marginal means (EMMs) were 305 calculated for combinations of predictor variables, while estimated marginal trends were 306 used to compute the regression slopes between groups, using the EMMEANS package. 307 Pairwise comparisons among treatment groups were performed using the Tukey method in 308 the MULTCOMP package. 309 310 Survival Probability 311 To calculate probability of survival, we performed a Kaplan–Meier survival analysis using 312 the SURVIVAL package (v3.8.3). Kaplan–Meier curves were generated to visualize 313 survival probabilities over time for each factor independently. Statistical differences 314 between survival curves were tested using log-rank tests. To evaluate the relative risk of 315 mortality, we fitted a Cox proportional hazards model with colony hygienic status and N. 316 ceranae dose as fixed effects, as well as two Cox proportional hazards models with each 317 predictor variable as an independent fixed effect. Survival time in days was used as the 318 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint response, with individuals surviving the 14-day observation period treated as right-319 censored. 320

321 No N. ceranae spores found in pupae 322 Pupal samples were collected from 28 honey bee colonies with confirmed Nosema spp. 323 infections in nurse bees ranging in average spore load from 5×10⁴ to 1.4×10⁶ spores per 324 bee. None of the pupal samples collected were found to contain N. ceranae spores. These 325

indicate that the removal of brood through hygienic behavior would have no direct 326 effect on N. ceranae load of the colonies in our target population, since N. ceranae 327 infection is likely only present in the adult bee castes. 328 Group-fed bees experienced higher N. ceranae loads 329 At 10 days post-inoculation, N. ceranae prevalence did not differ between group-fed bees 330 (59.3%) and individually-fed bees (43.5%) (p = 0.24), nor between group-fed bees with 10 331 bees (69.6%) or 30 bees (57%) per cage (p = 0.38). Of the infected bees, group-feeding 332 resulted in significantly higher average N. ceranae loads (p < 0.001), but also greater 333 variance (p = 0.021) compared to individual-feeding, which can be found in S3 Fig. The 334 number of bees per hoarding cage (10 or 30 bees) had no significant effect on N. ceranae 335 loads (p = 0.89). Infected group-fed bees experienced an average N. ceranae load of 4.28 × 336 10⁶ ± 2.13 × 10¹³ spores per bee, whereas individually-fed bees experienced an average load 337 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint of 8.05 × 10⁵ ± 2.61 × 10¹² spores per bee. For subsequent experiments, we selected the 338 group-feeding method with 30 bees per cage to achieve the highest infection levels and 339 obtain a large sample size, despite the higher observed variation in N. ceranae loads. We 340 determined that the group-feeding approach better reflected natural N. ceranae transmission 341 dynamics in the hive, which are largely driven by a few highly infected individuals [62]. 342 Immune function and performance differ between hygienic and 343 non-hygienic bees 344 In N. ceranae-inoculated bees, N. ceranae infection levels increased over time, 345 demonstrating successful inoculation methods and effective contraction of the pathogen (p 346 < 0.001) (Fig 1). At day 12 post-inoculation, we found a N. ceranae prevalence of 50% in 347 bees that received the low N. ceranae dose and 81.5% in bees that received the high N. 348 ceranae dose. All N. ceranae-inoculated bees experienced significantly higher infection 349 levels than control bees (p < 0.001), but there was only a marginal difference between the 350 low (104 spores per bee) and high N. ceranae (5 x 104 spores per bee) dose groups (p = 351 0.052) (Fig 2). We attribute the low levels of N. ceranae observed in 17.2% of control bees 352 at day 12 post-inoculation to newly emerged adults ingesting spores on their original 353 frames before being transferred to hoarding cages for the experiment. 354 355 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 356 Fig 1. Effect of colony hygienic status and N. ceranae dose on relative N. ceranae 357 infection levels in bees over time. 358 There was a significant effect of sampling day (χ²₁ = 42.87, p < 0.001) and N. ceranae dose 359 (χ²₂ = 79.11, p < 0.001) on relative N. ceranae infection levels (ΔΔCt, log-transformed), 360 with a significant interaction effect of sampling day and N. ceranae dose (χ²₂ = 15.56, p < 361 0.001). There was no effect of colony hygienic status (χ²₁ = 1.22, p = 0.27). Purple 362 points/lines represent hygienic bees; green points/lines represent non-hygienic bees. Panels 363 correspond to N. ceranae doses (High = 5 × 10⁴ spores/bee; Low = 1 × 10⁴ spores/bee; 364 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint Control = 0 spores/bee). Sample sizes are denoted at each time point as n = hygienic bees / 365 non-hygienic bees. 366 367 368 Fig 2 Effect of N. ceranae dose on relative N. ceranae infection levels in hygienic and 369 non-hygienic bees. 370 There was a significant main effect of N. ceranae dose on relative infection levels (χ²₂ = 371 79.11, p < 0.001), but no significant effect of colony hygienic status (χ²₁ = 1.22, p = 0.27) 372 or an interaction effect between colony hygienic status and N. ceranae dose (χ²₂ = 0.062, p 373 = 0.938). Purple boxes represent hygienic bees; green boxes represent non-hygienic bees. 374 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint Panels correspond to N. ceranae doses (High = 5 × 10⁴ spores/bee; Low = 10⁴ spores/bee; 375 Control = 0 spores/bee). Infection levels are shown as relative N. ceranae score (ΔΔCt, log-376 transformed). Sample sizes are denoted above each box and varied based on bee mortality. 377 Significance between pairs is denoted as ‘.’ < 0.1, ‘*’ < 0.05, ‘**’ < 0.01, ‘***’ < 0.001. 378 379 Hygienic and non-hygienic bees did not differ significantly by N. ceranae prevalence 380 (p = 1) nor in increasing N. ceranae infection levels observed over time (p =0.938). 381 However, we found that hygienic bees in hoarding cages consumed significantly less sugar 382 syrup inoculant in the 24hr inoculation period overall, compared to non-hygienic bees (p < 383 0.001). There was no significant effect of N. ceranae dose (p = 0.129) nor an interaction 384 effect between colony hygienic status and N. ceranae dose (p = 0.101). Broken out by N. 385 ceranae dose in a pairwise comparison of our linear model, we found a biologically 386 relevant trend that hygienic bees consumed less of the highest N. ceranae dose (5 x 104 387 spores per bee) compared to non-hygienic bees (p = 0.055), but otherwise no significant 388 differences were found between the volume of sugar syrup consumed by hygienic and non-389 hygienic bees among the control (0 spores per bee, p = 0.595) and low N. ceranae dose 390 groups (1 × 10⁴ spores/bee, p = 0.449) (Fig 3). 391 392 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 393 Fig 3. Effect of colony hygienic status and N. ceranae dose on the volume of sugar 394 syrup inoculant consumed by bees in hoarding cages. 395 There was a significant main effect of colony hygienic status on the volume of inoculant 396 consumed by bees in hoarding cages (Kruskal-Wallis test, χ² = 4.3104, df = 1, p = 0.038), 397 but no effect of N. ceranae dose (χ² = 4.0909, df = 2, p = 0.129) nor interaction between 398 colony hygienic status and N. ceranae dose (χ² = 9.228, df = 5, p = 0.101). There was a 399 marginal difference in the volume of inoculant consumed between hygienic and non-400 hygienic bees in the high N. ceranae dose group (p = 0.055). Purple boxes represent 401 hygienic bees; green boxes represent non-hygienic bees. Panels correspond to N. ceranae 402 doses (High = 5 × 10⁴ spores/bee; Low = 1 × 10⁴ spores/bee; Control = 0 spores/bee). 403 Inoculant consumption is shown in milliliters (mL). Sample sizes are denoted above each 404 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint box and varied based on bee mortality. Significance between pairs is denoted as ‘.’ < 0.1, 405 ‘*’ < 0.05, ‘**’ < 0.01, ‘***’ < 0.001. 406 407 Vitellogenin expression levels significantly differed by colony hygienic status (p = 0.006) 408 and sampling day (p < 0.001) (Fig 4). At peak N. ceranae infection (day 12 post-409 inoculation), hygienic bees showed upregulated Vitellogenin (Vg) expression while non-410 hygienic bees showed downregulated expression in all groups, including the low N. 411 ceranae dose (p < 0.001), high N. ceranae dose (p = 0.005), and control (p < 0.001) bees. 412 Additionally, we found a marginal difference in Vg expression on day 4 post-inoculation 413 between hygienic and non-hygienic bees that received the highest dose of N. ceranae (5 x 414 104 spores per bee) (p = 0.056). Conversely, we found that Hymenoptaecin levels did not 415 differ significantly by colony hygienic status, N. ceranae dose, sampling day, nor their 416 interactions, which can be found in S4 Figure. 417 418 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 419 Fig 4. Effect of colony hygienic status on Vitellogenin (Vg) gene expression across time 420 and N. ceranae doses. 421 There were significant main effects of colony hygienic status (χ²₁ = 7.62, p = 0.006) and 422 sampling day (χ²₄ = 22.78, p < 0.001), as well as a significant interaction between colony 423 hygienic status and day (χ²₄ = 19.79, p < 0.001) on Vitellogenin expression levels. 424 Vitellogenin expression levels are shown as the relative Vg score (ΔΔCt, log₁₀-transformed). 425 Purple bars indicate hygienic bees; green lines indicate non-hygienic bees. Panels 426 correspond to sampling days post-inoculation. Error bars represent standard errors of the 427 mean. Sample sizes are denoted above each bar pair as n = hygienic bees/non-hygienic 428 bees. Significance between pairs is denoted as ‘.’ < 0.1, ‘*’ < 0.05, ‘**’ < 0.01, 429 ‘***’ < 0.001. 430 431 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint As a more accurate measure of bees’ immune response against N. ceranae, we correlated 432 Vitellogenin (Vg)/Hymenoptaecin (Hym) expression levels with actual N. ceranae 433 infection. We found that N. ceranae infection levels (p = 0.028) and colony hygienic status 434 (p = 0.012) had a significant main effect on Vg expression levels, but no significant 435 interaction effect (p = 0.242). In all bees, Vg expression levels downregulated in response 436 to increasing N. ceranae infection levels; however, hygienic bees showed less of a negative 437 relationship, where Vg expression was relatively unaffected by increasing N. ceranae 438 infection levels (Fig 5). Evaluating Hymenoptaecin expression in response to N. ceranae 439 infection, we found a significant interaction effect between N. ceranae infection levels and 440 colony hygienic status (p = 0.016), where hygienic bees showed lower Hym expression 441 with mild infection followed by upregulation in response to increasing N. ceranae infection 442 levels. In contrast, non-hygienic bees showed consistent downregulation in Hym expression 443 in response to N. ceranae infection levels (Fig 6). 444 445 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 446 Fig 5. Effect of N. ceranae infection levels on Vitellogenin (Vg) gene expression by 447 colony hygienic status. 448 Vg gene expression levels (ΔΔCt, log-transformed) were impacted significantly by main 449 effects of colony hygienic status (χ²₁ = 6.27, p = 0.012), N. ceranae infection levels (χ²₁ = 450 4.81, p = 0.028), and sampling day (χ²₄ = 14.27, p = 0.006). There was no significant 451 interaction effect between colony hygienic status and N. ceranae infection levels (χ²₁ = 452 1.37, p = 0.242). Purple points/lines represent hygienic bees; green points/lines represent 453 non-hygienic bees. Sample sizes are denoted as n = hygienic bees / non-hygienic bees. 454 455 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 456 Fig 6. Effect of N. ceranae infection levels on Hymenoptaecin (Hym) gene expression 457 by colony hygienic status. 458 Hym gene expression levels (ΔΔCt, log₁₀-transformed) were impacted by an interaction 459 effect of colony hygienic status and N. ceranae infection levels (χ²₁ = 5.805, p = 0.016). No 460 main effects of N. ceranae infection levels (χ²₁ = 0.05, p = 0.831) or colony hygienic status 461 (χ²₁ = 0.00, p = 0.997) on Hym expression levels were detected. Purple points/lines 462 represent hygienic bees; green points/lines represent non-hygienic bees. Sample sizes are 463 denoted as n = hygienic bees / non-hygienic bees. 464 465 Hygienic and non-hygienic bees differed significantly in their probability of survival during 466 the experimental trials (p = 0.02). Among the N. ceranae-inoculated groups, hygienic bees 467 had significantly better survival odds than non-hygienic bees starting 8 days post-468 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint inoculation (p = 0.004), where non-hygienic bees had a 53% higher risk of death when 469 infected (Fig 7). Bees in the control group did not differ in survival probability based on 470 colony hygienic status (p = 0.50). Among all bees, N. ceranae inoculation significantly 471 affected the probability of bee survival (p < 0.001) starting four days post-inoculation. 472 Nosema ceranae-inoculated bees had a 135-138% higher risk of death compared to control 473 bees, but the high and low N. ceranae dose groups did not differ significantly in survival 474 probability (p = 0.93) (Fig 8). 475 476 477 Fig 7. Survival probability of bees by colony hygienic status and treatment group over 478 time. 479 Among N. ceranae-inoculated groups, hygienic bees had significantly better survival odds 480 than non-hygienic bees (χ² = 8.3, df = 1, p = 0.004; HR = 1.53, 95% CI: 1.15–2.04). Among 481 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint control bees, there was no significant difference by colony hygienic status (χ² = 0.4, df = 1, 482 p = 0.50; HR = 0.83, 95% CI: 0.46–1.50). Purple lines indicate hygienic bees; green lines 483 indicate non-hygienic bees. Dashed lines represent control bees; solid lines represent N. 484 ceranae-inoculated bees. Sample sizes are denoted beside each line. 485 486 487 Fig 8. Effect of N. ceranae dose on survival probability over time. 488 There was a significant effect of N. ceranae dose on survival probability (χ²₂ = 28.3, df = 2, 489 p = 7 × 10⁻⁷), where N. ceranae exposure significantly increased the hazard of death 490 compared to control bees. Bees that received the low and high dose N. ceranae inoculant 491 did not differ significantly in survival probability (p = 0.93). Red line indicates high N. 492 ceranae dose (5 × 10⁴ spores/bee), orange line indicates low N. ceranae dose (10⁴ 493 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint spores/bee), and blue line indicates control group (0 spores/bee). Sample sizes are n = 240 494 for each group. 495

496 Our results suggest that individual bees originating from hygienic (high UBeeO) colonies 497 express innate defense mechanisms against Nosema ceranae. Despite hygienic bees 498 showing similar levels of N. ceranae infection to non-hygienic bees in this cage study, we 499 find that individual bees in hygienic colonies may actively mitigate N. ceranae infection by 500 1) limiting the amount of inoculant consumed, 2) upregulating Vitellogenin expression 501 during peak infection, 3) upregulating Hymenoptaecin expression in response to increasing 502 infection, and 4) experiencing greater survivability. Hygienic bees may also modulate 503 investment in innate immunity in response to infection severity while limiting the health 504 impacts of N. ceranae by maintaining Vg and Hym expression as infection increases. It is 505 important to note that, due to our study design, bees were limited in their ability to remove 506 unhealthy individuals from their cages—a key behavior that likely contributes to reducing 507 N. ceranae loads in hygienic colonies under natural hive conditions. As a result, our 508 measurements of individual N. ceranae levels may not fully reflect differences between 509 hygienic and non-hygienic colonies in the field since social immunity is known to play a 510 significant role in host-pathogen dynamics. 511 512 We found no evidence of N. ceranae spores in developing pupae of N. ceranae-infected 513 colonies, suggesting that brood does not likely experience N. ceranae infection under 514 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint natural hive settings in our target population. Therefore, we believe that hygienic behavior 515 would have no direct effect on the level of N. ceranae infection in the colony, since the 516 behavior acts solely on infected brood. Previous studies have shown that larvae can be 517 manually inoculated with N. ceranae and will develop N. ceranae-induced physiological 518 and metabolic impairments as adults [34,35]; however, pupae have only been shown to 519 experience rare infection under natural hive conditions [36,37]. Overall, Nosema ceranae 520 infections in brood have not been thoroughly investigated, and the extent of their 521 prevalence—and how it may vary geographically—remains unclear. Our findings are 522 consistent with previous studies showing an absence of N. ceranae infection in newly 523 emerged adults [17,53] and little direct impact of hygienic behavior on N. ceranae, besides 524 an observed overall reduction at the apiary level over time [14,40]. 525 526 Compared to non-hygienic bees, hygienic bees consumed less sugar syrup inoculant 527 overall. We found a biologically relevant trend that hygienic bees consumed less of the 528 high dose N. ceranae inoculant, compared to the low N. ceranae dose or control group. It 529 remains unclear whether hygienic colonies may be able to detect and avoid N. ceranae-530 contaminated food sources within the hive. N. ceranae transmission most often occurs 531 through the oral-fecal route, by consuming contaminated food stores [63,64], cleaning bee 532 excrement from the frames, or through engagement in trophallaxis with infected nestmates 533 [65]. The recognition and avoidance of N. ceranae spores on hive materials could have a 534 significant impact on reducing N. ceranae transmission in the colony. Our findings point to 535 a potentially heightened sensitivity of hygienic bees to atypical odors at high 536 concentrations, such as those associated with N. ceranae or other pathogens. Future studies 537 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint should investigate whether hygienic bees avoid N. ceranae–contaminated hive materials in 538 cage-choice experiments and how they respond to pathogen-related odors in olfactometer 539 assays. Observational studies should also assess whether hygienic bees exhibit additional 540 in-hive behaviors, such as the “entombment” of contaminated food stores [66] increased 541 attentiveness to infected adults, or higher rates of auto- and allo-grooming [1,6]. 542 543 Overall, N. ceranae infection substantially increased the risk of bee mortality, supporting 544 existing evidence of its negative impact to bee health [19–22,25,27]. However, hygienic 545 bees seem to be more tolerant to N. ceranae compared to non-hygienic bees. Despite 546 exhibiting similar N. ceranae infection levels in our cage study, hygienic bees survived 547 significantly longer than non-hygienic bees when infected with N. ceranae. Tolerance is 548 defined by an organism’s ability to minimize the damage caused by a pathogen, rather that 549 reducing or eliminating the pathogen itself. Our findings are consistent with the enhanced 550 survival observed in infected drones of a known N. ceranae-tolerant honey bee strain in 551 Denmark [26,67]. When infected with N. ceranae, tolerant drones showed normal rates of 552 apoptosis in the midgut epithelium, maintaining normal cell function and the ability to rid 553 damaged tissue. Limiting N. ceranae’s ability to inhibit apoptosis– a key mechanism in the 554 pathogenesis of N. ceranae infections– may therefore contribute to the enhanced 555 survivability observed in hygienic bees. If an altered apoptotic response to N. ceranae 556 infection in hygienic bees could facilitate defecation of infected cells outside of the hive, it 557 may also play a role in limiting transmission of the pathogen between nestmates and 558 explain the reduced loads observed at the colony level [3]. However, further evaluations to 559 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint compare the apoptosis rates between hygienic and non-hygienic bees with N. ceranae 560 infections are needed to support these hypotheses. 561 562 Improved survivability in hygienic bees may also be explained by an enhanced buffering 563 capacity that reduces the energetic stress caused by N. ceranae [19,20], as demonstrated in 564 N. ceranae-tolerant drones in Denmark [3]. Although we did not measure sugar syrup 565 consumption throughout the 12-day infection period, the reduced overall intake of inoculant 566 during the inoculation period may indicate a lower carbohydrate demand in hygienic bees. 567 Future studies should evaluate hemolymph trehalose levels in Nosema-infected hygienic 568 bees to better understand if their energy availability is preserved over time [68], which may 569 contribute to their improved survival and performance in the colony. Nevertheless, the 570 prolonged survival of N. ceranae-infected individuals in hygienic colonies may help to 571 alleviate colony-level impacts of N. ceranae (e.g. reduction in population, decreased honey 572 production [15]) by retaining the workforce over time. Conversely, surviving beyond seven 573 days old, when precocious foraging caused by N. ceranae infection is likely to occur [24], 574 may be an adaptation of hygienic colonies to lower pathogen transmission in the hive by 575 favoring the reduced homing ability of infected adults [23,69]. 576 577 We evaluated Vitellogenin and Hymenoptaecin gene expression to assess the innate 578 immune function of hygienic bees with N. ceranae infection. Hygienic bees exhibited 579 significantly upregulated Vitellogenin (Vg) gene expression at peak N. ceranae infection 580 (day 12 post-inoculation). Notably, the level of upregulation was independent of N. ceranae 581 dose, indicating that hygienic bees exhibit upregulated Vg expression at this time point 582 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint even in the absence of pathogen exposure. Non-hygienic bees showed downregulated Vg 583 expression as N. ceranae levels increased compared to hygienic bees, which maintained 584 relatively constant Vg expression, suggesting that non-hygienic bees may have a more 585 compromised or altered physiological response under infection compared to hygienic bees. 586 The pronounced increase in Vg expression observed at day 12 in hygienic bees has not 587 been reported in previous studies. In a healthy colony, Vg levels typically peak in nurse 588 bees around four days old and decline as workers transition from in-hive tasks to foraging 589 duties [70,71]. While N. ceranae infection can cause elevated Vg levels in younger bees, 590 the late spike observed in hygienic bees is unusual and suggests potential functional 591 implications that warrant further investigation. 592 593 The overexpression of Vg at day 12 post-inoculation (≈15 days old) could reflect changes 594 in normal behavioral maturation [50] and social organization [51] in hygienic colonies. 595 However, overexpression could also enhance immunological defenses and resilience 596 against pathogens. Vitellogenin has been shown to bind to pathogens, suppress microbial 597 growth, and contribute to tissue repair from oxidative stress [48]. At 15 days old, workers 598 typically transition to undertaker roles [72], or in hygienic colonies, perform hygienic 599 behaviors to remove dead or parasitized individuals [73]. Concurrent Vg upregulation may 600 protect these bees while performing risky duties. High levels of Vg are linked to the 601 prolonged lifespan of queen bees and stress resilient diutinus bees [52], suggesting that 602 upregulated Vg may underlie the enhanced survivability observed in hygienic bees. Further, 603 the potential for Vg to perform trans-generational immune priming functions [49,74] could 604 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint have important implications for the heritability of N. ceranae tolerance in hygienic 605 colonies. 606 607 While hygienic bees do not show clear differences in Hymenoptaecin expression when 608 compared to non-hygienic bees over time, there is a significant relationship between Hym 609 expression levels and N. ceranae infection severity. Hygienic bees exhibited lower Hym 610 expression at low infection intensities compared to non-hygienic bees and a stronger 611 upregulation of Hym expression in response to increasing N. ceranae infection levels. Our 612 findings suggest that hygienic bees may reduce investment in innate immunity under low 613 pathogen stress, but are better able to combat N. ceranae as infection increases. In contrast, 614 non-hygienic bees show a stronger immune response under low pathogen stress but weaken 615 in Hym expression as N. ceranae infection increases. The relationship between Hym 616 expression and N. ceranae infection in non-hygienic bees reflects previous studies showing 617 the pathogen’s immunosuppressing capabilities in infected bees [22,53]. While we do not 618 find higher Hymenoptaecin expression in hygienic bees overall, our findings are 619 comparable with previous work showing elevated Toll pathway–mediated immune gene 620 expression in N. ceranae–tolerant drones [67] and higher Hym expression in workers from 621 hygienic colonies [12]. Furthermore, the reduced initial investment in Hym expression in 622 hygienic bees may result in more energy availability and explain their reduced demand for 623 sugar syrup inoculant during the inoculation period. Future studies should examine 624 additional Toll pathway–mediated antimicrobial peptides to fully characterize innate 625 immunity in hygienic bees and its role in controlling N. ceranae. 626 627 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint Our study highlights that colony-level resistance to N. ceranae may emerge from the 628 cumulative effects of individual-level mechanisms. Disease resistance that confers reduced 629 levels of pest and pathogen infestation at the colony level has been a major focus in recent 630 honey bee breeding efforts [75–77]. However, selective breeding programs may also 631 consider targeting individual-level tolerance mechanisms, such as maintained apoptosis 632 rates and/or improved energetic buffering capacity under infection, that avoid antagonistic 633 host–parasite coevolution [78] and could promote colony-level resistance to pathogens 634 while circumventing the pitfalls of pure tolerance-based selection [79]. For example, 635 individual tolerance to Deformed Wing Virus is thought to contribute to the winter survival 636 of Varroa-resistant colonies [80]. In general, tolerance mechanisms are not usually 637 pathogen-specific and could offer protection against a broad range of pathogens in honey 638 bee colonies [77]. 639 640 Overall, our findings suggest that hygienic behavior in honey bee colonies, quantified by 641 the UBeeO assay, may be linked to individual-level defenses that function concurrently to 642 maintain low levels of N. ceranae at the colony level. These investigations advance our 643 understanding of how hygienic behavior performance can predict pathogen loads and have 644 important implications for selective breeding strategies, N. ceranae prevention, and disease 645 management. Further research is needed to explore potential social immune mechanisms 646 that combat N. ceranae and how nestmates interact with infected individuals in hygienic 647 colonies. While previous studies have shown that nestmates can exhibit behaviors ranging 648 from avoidance to aggression towards Nosema-infected individuals [81], it remains unclear 649 how these interactions differ in hygienic colonies and how social behaviors might 650 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint complement the individual-level traits of hygienic bees revealed in this study. Here, we 651 offer a valuable perspective on the abilities of individual workers to regulate N. ceranae 652 infection in hygienic colonies and contribute to ongoing efforts to improve honey bee 653 health. 654 Acknowledgments 655 We thank the USDA Beltsville technicians, Allison Shaulis and Kyle Grubbs, for their 656 assistance with laboratory work. We are especially grateful to Michael Palmer of French 657 Hill Apiaries and Bianca Braman of Vermont Bees LLC for providing access to their 658 apiaries and extensive support throughout the project. 659

660 1. Evans JD, Spivak M. Socialized medicine: Individual and communal disease barriers 661 in honey bees. J Invertebr Pathol. 2010;103: S62–S72. doi:10.1016/j.jip.2009.06.019 662 2. Evans JD, Aronstein K, Chen YP, Hetru C, Imler J-L, Jiang H, et al. Immune 663 pathways and defence mechanisms in honey bees Apis mellifera. Insect Mol Biol. 664 2006;15: 645–656. 665 3. Kurze C, Routtu J, Moritz RFA. Parasite resistance and tolerance in honeybees at the 666 individual and social level. Zoology. Elsevier GmbH; 2016. pp. 290–297. 667 doi:10.1016/j.zool.2016.03.007 668 4. Rothenbuhler WC. Behaviour genetics of nest cleaning in honey bees. I. Responses 669 of four inbred lines to disease-killed brood. Anim Behav. 1964;12: 578–583. 670 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 5. Wagoner KM, Spivak M, Rueppell O. Brood Affects Hygienic Behavior in the 671 Honey Bee (Hymenoptera: Apidae). J Econ Entomol. 2018;111: 2520–2530. 672 doi:10.1093/jee/toy266 673 6. Boecking O, Spivak M. Behavioral defenses of honey bees against Varroa jacobsoni 674 Oud. Apidologie. 1999;30: 141–158. 675 7. Spivak M, Danka RG. Perspectives on hygienic behavior in Apis mellifera and other 676 social insects. Apidologie. 2021;52: 1–16. doi:10.1007/s13592-020-00784-z 677 8. Spivak M, Gilliam M. Hygienic behaviour of honey bees and its application for 678 control of brood diseases and varroa: Part II. Studies on hygienic behaviour since the 679 Rothenbuhler era. Bee World. 1998;79: 169–186. 680 doi:10.1080/0005772X.1998.11099408 681 9. Lapidge KL, Oldroyd BP, Spivak M. Seven suggestive quantitative trait loci 682 influence hygienic behavior of honey bees. Naturwissenschaften. 2002;89: 565–568. 683 doi:10.1007/s00114-002-0371-6 684 10. Spivak M, Downey DL. Field Assays for Hygienic Behavior in Honey Bees 685 (Hymenoptera: Apidae). J Econ Entomol. 1998;91: 64–70. doi:10.1093/JEE/91.1.64 686 11. Medina-Flores CA, Medina LAM, Guzmán-Novoa E. Effect of hygienic behavior on 687 resistance to chalkbrood disease (Ascosphaera apis) in Africanized bee colonies 688 (Apis mellifera). Rev Mex Cienc Pecu. 2022;13: 225–239. 689 doi:10.22319/RMCP.V13I1.5907 690 12. Erez T, Bonda E, Kahanov P, Rueppell O, Wagoner K, Chejanovsky N, et al. 691 Multiple benefits of breeding honey bees for hygienic behavior. J Invertebr Pathol. 692 2022;193: 107788. doi:10.1016/J.JIP.2022.107788 693 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 13. Invernizzi C, Rivas F, Bettucci L. Resistance to Chalkbrood Disease in Apis 694 mellifera L. (Hymenoptera: Apidae) Colonies with Different Hygienic Behaviour. 695 Neotrop Entomol. 2011;40: 28–34. 696 14. Alger SA, Burnham PA, Miller MS, Amiri E, Jordan C, Wagoner K. Unhealthy 697 brood odor scores predict pathogen loads of several important honey bee diseases. 698 Frontiers in Bee Science. 2025;3: 1–15. doi:10.3389/frbee.2025.1509871 699 15. Botías C, Martín-Hernández R, Barrios L, Meana A, Higes M. Nosema spp. 700 infection and its negative effects on honey bees (Apis mellifera iberiensis) at the 701 colony level. Vet Res. 2013;44: 1–15. doi:10.1186/1297-9716-44-25 702 16. Gisder S, Mockel N, Linde A, Genersch E. A cell culture model for Nosema ceranae 703 and Nosema apis allows new insights into the life cycle of these important honey 704 bee-pathogenic microsporidia. Environ Microbiol. 2011;13: 404–413. 705 doi:10.1111/J.1462-2920.2010.02346.X 706 17. Smart MD, Sheppard WS. Nosema ceranae in age cohorts of the western honey bee 707 (Apis mellifera). J Invertebr Pathol. 2012;109: 148–151. 708 doi:10.1016/J.JIP.2011.09.009 709 18. Jabal-Uriel C, Albarracín VN, Calatayud J, Higes M, Martín-Hernández R. Age and 710 Season Effect the Timing of Adult Worker Honeybee Infection by Nosema ceranae. 711 Front Cell Infect Microbiol. 2022;11: 1–12. doi:10.3389/fcimb.2021.823050 712 19. Mayack C, Naug D. Energetic stress in the honeybee Apis mellifera from Nosema 713 ceranae infection. J Invertebr Pathol. 2009;100: 185–188. 714 doi:10.1016/J.JIP.2008.12.001 715 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 20. Martín-Hernández R, Botías C, Barrios L, Martínez-Salvador A, Meana A, Mayack 716 C, et al. Comparison of the energetic stress associated with experimental Nosema 717 ceranae and Nosema apis infection of honeybees (Apis mellifera). Parasitol Res. 718 2011;109: 605–612. doi:10.1007/s00436-011-2292-9 719 21. Antúnez K, Martín-Hernández R, Prieto L, Meana A, Zunino P, Higes M. Immune 720 suppression in the honey bee (Apis mellifera) following infection by Nosema 721 ceranae (Microsporidia). Environ Microbiol. 2009;11: 2284–2290. 722 doi:10.1111/j.1462-2920.2009.01953.x 723 22. Chaimanee V, Chantawannakul P, Chen Y, Evans JD, Pettis JS. Differential 724 expression of immune genes of adult honey bee (Apis mellifera) after inoculated by 725 Nosema ceranae. J Insect Physiol. 2012;58: 1090–1095. 726 doi:10.1016/J.JINSPHYS.2012.04.016 727 23. Dussaubat C, Maisonnasse A, Crauser D, Beslay D, Costagliola G, Soubeyrand S, et 728 al. Flight behavior and pheromone changes associated to Nosema ceranae infection 729 of honey bee workers (Apis mellifera) in field conditions. J Invertebr Pathol. 730 2013;113: 42–51. doi:10.1016/J.JIP.2013.01.002 731 24. Goblirsch M, Huang ZY, Spivak M. Physiological and Behavioral Changes in Honey 732 Bees (Apis mellifera) Induced by Nosema ceranae Infection. PLoS One. 2013;8: 1–733 8. doi:10.1371/journal.pone.0058165 734 25. Kurze C, Le Conte Y, Kryger P, Lewkowski O, Müller T, Moritz RFA. Infection 735 dynamics of Nosema ceranae in honey bee midgut and host cell apoptosis. J 736 Invertebr Pathol. 2018;154: 1–4. doi:10.1016/J.JIP.2018.03.008 737 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 26. Kurze C, Le Conte Y, Dussaubat C, Erler S, Kryger P, Lewkowski O, et al. Nosema 738 tolerant honeybees (apis mellifera) escape parasitic manipulation of apoptosis. PLoS 739 One. 2015;10: 1–8. doi:10.1371/journal.pone.0140174 740 27. Higes M, Juarranz Á, Dias-Almeida J, Lucena S, Botías C, Meana A, et al. 741 Apoptosis in the pathogenesis of Nosema ceranae (Microsporidia: Nosematidae) in 742 honey bees (Apis mellifera). Environ Microbiol Rep. 2013;5: 530–536. 743 doi:10.1111/1758-2229.12059 744 28. Marín-García PJ, Peyre Y, Ahuir-Baraja AE, Garijo MM, Llobat L. The Role of 745 Nosema ceranae (Microsporidia: Nosematidae) in Honey Bee Colony Losses and 746 Current Insights on Treatment. Vet Sci. 2022;9: 1–12. doi:10.3390/vetsci9030130 747 29. Bahreini R, González-Cabrera J, Hernández-Rodríguez CS, Moreno-Martí S, 748 Muirhead S, Labuschagne RB, et al. Arising amitraz and pyrethroids resistance 749 mutations in the ectoparasitic Varroa destructor mite in Canada. Sci Rep. 2025;15: 1-750 11. doi:10.1038/s41598-025-85279-6 751 30. Hernández-Rodríguez CS, Marín Ó, Calatayud F, Mahiques MJ, Mompó A, Segura 752 I, et al. Large-scale monitoring of resistance to coumaphos, amitraz, and pyrethroids 753 in Varroa destructor. Insects. 2021;12: 1–12. doi:10.3390/insects12010027 754 31. Elzen PJ, Eischen FA, Baxter JR, Elzen GW, Wilson WT. Detection of resistance in 755 US Varroa jacobsoni Oud. (Mesostigmata: Varroidae) to the acaricide fluvalinate. 756 Apidologie. 1998;30: 13–17. 757 32. Higes M, Martín-Hernández R, Garrido-Bailón E, García-Palencia P, Meana A. 758 Detection of infective Nosema ceranae (Microsporidia) spores in corbicular pollen of 759 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint forager honeybees. J Invertebr Pathol. 2008;97: 76–78. 760 doi:10.1016/j.jip.2007.06.002 761 33. Forsgren E, Fries I. Comparative virulence of Nosema ceranae and Nosema apis in 762 individual European honey bees. Vet Parasitol. 2010;170: 212–217. 763 doi:10.1016/j.vetpar.2010.02.010 764 34. Eiri DM, Suwannapong G, Endler M, Nieh JC. Nosema ceranae Can Infect Honey 765 Bee Larvae and Reduces Subsequent Adult Longevity. PLoS One. 2015;10: 1-17. 766 doi:10.1371/journal.plos.0126330 767 35. Benvau LR, Nieh JC. Larval honey bees infected with Nosema ceranae have 768 increased vitellogenin titers as young adults. Sci Rep. 2017;7: 1-8. 769 doi:10.1038/s41598-017-14702-4 770 36. Urbieta-Magro A, Higes M, Meana A, Gómez-Moracho T, Rodríguez-García C, 771 Barrios L, et al. The levels of natural Nosema spp. infection in Apis mellifera 772 iberiensis brood stages. Int J Parasitol. 2019;49: 657–667. 773 doi:10.1016/J.IJPARA.2019.04.002 774 37. Rodríguez M, Vargas M, Antúnez K, Gerding M, Castro FO, Zapata N. Prevalence 775 and phylogenetic analysis of honey bee viruses in the Biobío Region of Chile and 776 their association with other honey bee pathogens. Chil J Agric Res. 2014;74: 170–777 177. doi:10.4067/S0718-58392014000200007 778 38. Jack CJ, Lucas HM, Webster TC, Sagili RR. Colony level prevalence and intensity 779 of nosema ceranae in honey bees (Apis mellifera L.). PLoS One. 2016;11: 1-20. 780 doi:10.1371/journal.pone.0163522 781 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 39. Li W, Evans JD, Li J, Su S, Hamilton M, Chen Y. Spore load and immune response 782 of honey bees naturally infected by Nosema ceranae. Parasitol Res. 2017;116: 3265–783 3274. doi:10.1007/s00436-017-5630-8 784 40. Ivgin Tunca R, Oskay D, Erginoglu S. Türkiye’deki hijyenik bal arısı islah çalışması 785 süresince nosema enfeksiyon düzeyinin takibi. Kafkas Univ Vet Fak Derg. 2017;23: 786 521–526. doi:10.9775/kvfd.2016.17186 787 41. Wagoner KM, Millar JG, Schal C, Rueppell O. Cuticular pheromones stimulate 788 hygienic behavior in the honey bee (Apis mellifera). Sci Rep. 2020;10: 1–11. 789 doi:10.1038/s41598-020-64144-8 790 42. Harpur BA, Chernyshova A, Soltani A, Tsvetkov N, Mahjoorighasrodashti M, Xu Z, 791 et al. No genetic tradeoffs between hygienic behaviour and individual innate 792 immunity in the honey bee, Apis mellifera. PLoS One. 2014;9: 1–7. 793 doi:10.1371/journal.pone.0104214 794 43. Tola YH, Wagoner K, Strand MK, Rueppell O, Tarpy DR. The gut microbiome 795 differs between hygiene-performing and non-hygiene-performing worker honey 796 bees. Insectes Soc. 2025;72: 397–404. doi:10.1007/s00040-025-01029-x 797 44. Teixeira ÉW, de Paiva Daibert RM, Glatzl Júnior LA, da Silva MVGB, Alves 798 MLTMF, Evans JD, et al. Transcriptomic analysis suggests candidate genes for 799 hygienic behavior in African-derived Apis mellifera honeybees. Apidologie. 800 2021;52: 447–462. doi:10.1007/s13592-020-00834-6 801 45. Casteels P, Ampe C, Jacobs F, Tempst P. Functional and chemical characterization 802 of Hymenoptaecin, an antibacterial polypeptide that is infection-inducible in the 803 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint honeybee (Apis mellifera). Journal of Biological Chemistry. 1993;268: 7044–7054. 804 doi:10.1016/s0021-9258(18)53143-4 805 46. Havukainen H, Münch D, Baumann A, Zhong S, Halskau Ø, Krogsgaard M, et al. 806 Vitellogenin recognizes cell damage through membrane binding and shields living 807 cells from reactive oxygen species. Journal of Biological Chemistry. 2013;288: 808 28369–28381. doi:10.1074/jbc.M113.465021 809 47. Seehuus S-C, Norberg K, Gimsa U, Krekling T, Amdam G V. Reproductive protein 810 protects functionally sterile honey bee workers from oxidative stress. PNAS. 811 2006;103: 962–967. doi:10.1073/pnas.0502681103 812 48. Zhang S, Wang S, Li H, Li L. Vitellogenin, a multivalent sensor and an 813 antimicrobial effector. Int J Biochem Cell Biol. 2011;43: 303–305. 814 doi:10.1016/J.BIOCEL.2010.11.003 815 49. Harwood G, Amdam G, Freitak D. The role of Vitellogenin in the transfer of 816 immune elicitors from gut to hypopharyngeal glands in honey bees (Apis mellifera). 817 J Insect Physiol. 2019;112: 90–100. doi:10.1016/J.JINSPHYS.2018.12.006 818 50. Rodriguez Messan M, Page RE, Kang Y. Effects of vitellogenin in age polyethism 819 and population dynamics of honeybees. Ecol Modell. 2018;388: 88–107. 820 doi:10.1016/J.ECOLMODEL.2018.09.011 821 51. Nelson CM, Ihle KE, Fondrk MK, Page RE, Amdam G V. The gene vitellogenin has 822 multiple coordinating effects on social organization. PLoS Biol. 2007;5: 0673–0677. 823 doi:10.1371/journal.pbio.0050062 824 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 52. Amdam G V., Fennern E, Havukainen H. Honeybee Neurobiology and Behavior. 825 Galizia GC, Eisenhardt D, Giurfa M, editors. Springer Dordrecht Heidelberg; 2011. 826 doi:10.1007/978-94-007-2099-2 827 53. Li W, Chen Y, Cook SC. Chronic Nosema ceranae infection inflicts comprehensive 828 and persistent immunosuppression and accelerated lipid loss in host Apis mellifera 829 honey bees. Int J Parasitol. 2018;48: 433–444. doi:10.1016/J.IJPARA.2017.11.004 830 54. Doublet V, Poeschl Y, Gogol-Döring A, Alaux C, Annoscia D, Aurori C, et al. Unity 831 in defence: Honeybee workers exhibit conserved molecular responses to diverse 832 pathogens. BMC Genomics. 2017;18: 1-17. doi:10.1186/s12864-017-3597-6 833 55. Alaux C, Dantec C, Parrinello H, Le Conte Y. Nutrigenomics in honey bees: Digital 834 gene expression analysis of pollen’s nutritive effects on healthy and varroa-835 parasitized bees. BMC Genomics. 2011;12: 1-13. doi:10.1186/1471-2164-12-496 836 56. Fries I, Chauzat MP, Chen YP, Doublet V, Genersch E, Gisder S, et al. Standard 837

for Nosema research. Journal of Apicultural Research. 2013;52: 1-28. 838 doi:10.3896/IBRA.1.52.1.14 839 57. Berbeć E, Migdał P, Cebrat M, Roman A, Murawska A. Honeybee age and inoculum 840 concentration as factors affecting the development of Nosema ceranae infection. 841 European Zoological Journal. 2022;89: 1180–1190. 842 doi:10.1080/24750263.2022.2121009 843 58. Wagoner K, Millar JG, Keller J, Bello J, Waiker P, Schal C, et al. Hygiene-Eliciting 844 Brood Semiochemicals as a Tool for Assaying Honey Bee (Hymenoptera: Apidae) 845 Colony Resistance to Varroa (Mesostigmata: Varroidae). Journal of Insect Science. 846 2021;21: 4–5. doi:10.1093/jisesa/ieab064 847 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 59. Bates D, Mächler M, Bolker BM, Walker SC. Fitting linear mixed-effects models 848 using lme4. J Stat Softw. 2015;67: 1-48. doi:10.18637/jss.v067.i01 849 60. Fox J, Weisberg S. Companion to Applied Regression. 2025. Available: https://r-850 forge.r-project.org/projects/car/, 851 61. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time 852 quantitative PCR and the 2-ΔΔCT method. Methods. 2001;25: 402–408. 853 doi:10.1006/meth.2001.1262 854 62. Mulholland GE, Traver BE, Johnson NG, Fell RDF. Individual variability of 855 Nosema ceranae infections in Apis mellifera colonies. Insects. 2012;3: 1143–1155. 856 doi:10.3390/insects3041143 857 63. Sokół R, Michalczyk M. Detection of Nosema spp. in worker bees, pollen and bee 858 bread during the honey flow season. Acta Veterinaria Brno. 2016;85: 261–266. 859 doi:10.2754/avb201685030261 860 64. Chen Y, Evans JD, Smith IB, Pettis JS. Nosema ceranae is a long-present and wide-861 spread microsporidian infection of the European honey bee (Apis mellifera) in the 862 United States. J Invertebr Pathol. 2008;97: 186–188. doi:10.1016/J.JIP.2007.07.010 863 65. Fries I. Nosema ceranae in European honey bees (Apis mellifera). J Invertebr Pathol. 864 2010;103: S73–S79. doi:10.1016/J.JIP.2009.06.017 865 66. vanEngelsdorp D, Evans JD, Donovall L, Mullin C, Frazier M, Frazier J, et al. 866 “Entombed Pollen”: A new condition in honey bee colonies associated with 867 increased risk of colony mortality. J Invertebr Pathol. 2009;101: 147–149. 868 doi:10.1016/J.JIP.2009.03.008 869 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 67. Huang Q, Kryger P, Le Conte Y, Moritz RFA. Survival and immune response of 870 drones of a Nosemosis tolerant honey bee strain towards N. ceranae infections. J 871 Invertebr Pathol. 2012;109: 297–302. doi:10.1016/J.JIP.2012.01.004 872 68. Mayack C, Naug D. Parasitic infection leads to decline in hemolymph sugar levels in 873 honeybee foragers. J Insect Physiol. 2010;56: 1572–1575. 874 doi:10.1016/j.jinsphys.2010.05.016 875 69. Kralj J, Fuchs S. Nosema sp. influences flight behavior of infected honey bee (Apis 876 mellifera) foragers. Apidologie. 2010;41: 21–28. doi:10.1051/apido/2009046 877 70. Amdam G V., Norberg K, Hagen A, Omholt SW. Social exploitation of vitellogenin. 878 Proc Natl Acad Sci U S A. 2003;100: 1799–1802. doi:10.1073/pnas.0333979100 879 71. Tsuruda JM, Page RE. The effects of young brood on the foraging behavior of two 880 strains of honey bees (Apis mellifera). Behav Ecol Sociobiol. 2009;64: 161–167. 881 doi:10.1007/s00265-009-0833-3 882 72. Breed MD, Williams DB, Queral A. Demand for Task Performance and Workforce 883 Replacement: Undertakers in Honeybee, Apis mellifera, Colonies. J Insect Behav. 884 2002;15: 319–329. 885 73. Spivak M, Masterman R, Ross R, Mesce KA. Hygienic behavior in the honey bee 886 (Apis mellifera L.) and the modulatory role of octopamine. J Neurobiol. 2003;55: 887 341–354. doi:10.1002/neu.10219 888 74. Salmela H, Amdam G V., Freitak D. Transfer of Immunity from Mother to Offspring 889 Is Mediated via Egg-Yolk Protein Vitellogenin. PLoS Pathog. 2015;11: 1-12. 890 doi:10.1371/journal.ppat.1005015 891 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 75. Kaskinova MD, Gaifullina LR, Saltykova ES, Poskryakov A V., Nikolenko AG. 892 Genetic markers for the resistance of honey bee to Varroa destructor. Vavilovskii 893 Zhurnal Genet Selektsii. 2021;24: 853–860. doi:10.18699/VJ20.683 894 76. Bouuaert DC, De Smet L, de Graaf DC. Breeding for virus resistance and its effects 895 on deformed wing virus infection patterns in honey bee queens. Viruses. 2021;13: 1–896 9. doi:10.3390/v13061074 897 77. Guichard M, Dietemann V, Neuditschko M, Dainat B. Advances and perspectives in 898 selecting resistance traits against the parasitic mite Varroa destructor in honey bees. 899 Genetics Selection Evolution. 2020;52: 1–22. doi:10.1186/s12711-020-00591-1 900 78. Doeschl-Wilson AB, Kyriazakis I. Should we aim for genetic improvement in host 901 resistance or tolerance to infectious pathogens? Frontiers in Genetics. 2012;9: 1-2. 902 doi:10.3389/fgene.2012.00272 903 79. Sokolov NA, Boots M, Bartlett LJ. Avoiding the tragedies of parasite tolerance in 904 Darwinian beekeeping. Proceedings of the Royal Society B: Biological Sciences. 905 Royal Society Publishing; 2025;292: 1-12. doi:10.1098/rspb.2024.2433 906 80. Locke B, Forsgren E, De Miranda JR. Increased tolerance and resistance to virus 907 infections: A possible factor in the survival of Varroa destructor-resistant honey bees 908 (Apis mellifera). PLoS One. 2014;9: 1-7. doi:10.1371/journal.pone.0099998 909 81. Biganski S, Kurze C, Müller MY, Moritz RFA. Social response of healthy 910 honeybees towards Nosema ceranae-infected workers: care or kill? Apidologie. 911 2018;49: 325–334. doi:10.1007/S13592-017-0557-8/FIGURES/4 912 82. Boncristiani H, Underwood R, Schwarz R, Evans JD, Pettis J, Vanengelsdorp D. 913 Direct effect of acaricides on pathogen loads and gene expression levels in honey 914 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint bees Apis mellifera. J Insect Physiol. 2012;58: 613–620. 915 doi:10.1016/j.jinsphys.2011.12.011 916 917 Supporting Information 918 919 S1 Table. Primers used to determine relative quantification of Nosema ceranae [82]. 920 921 922 S2 Table. Primers used to determine relative quantification of Hymenoptaecin and 923 Vitellogenin expression [1]. 924 925 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 926 S3 Fig. Effect of inoculation strategy on Nosema spp. loads in worker bees. 927 Nosema spp. loads (spores per bee) differed significantly between group-fed and 928 individually fed bees (Welch’s t-test: t₃₅.₀₅ = 4.67, p < 0.001), with greater loads and 929 variance among group-fed bees (Levene’s test: F₁,₈₁ = 5.52, p = 0.021). Purple boxes 930 indicate group-fed bees, and yellow boxes indicate individually-fed bees. Sample sizes are 931 denoted above each box. 932 933 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint 934 S4 Fig. Effect of colony hygienic status on Hymenoptaecin (Hym) gene expression 935 across time and N. ceranae doses. 936 No significant main effects of colony hygienic status (χ²₁ = 0.012, p = 0.913), sampling day 937 (χ²₄ = 3.33, p = 0.504), N. ceranae dose (χ²₂ = 3.23, p = 0.199), nor interaction effects 938 between the predictor variables were detected. Hymenoptaecin expression levels are shown 939 as the relative Hym score (ΔΔCt, log₁₀-transformed). Purple bars represent hygienic bees; 940 green bars represent non-hygienic bees. Error bars represent standard errors of the mean. 941 Sample sizes are denoted above each bar pair as n = hygienic bees/non-hygienic bees. 942 Significance between pairs is denoted as ‘.’ < 0.1, ‘*’ < 0.05, ‘**’ < 0.01, ‘***’ < 0.001. 943 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted February 4, 2026. ; https://doi.org/10.64898/2026.02.02.693565doi: bioRxiv preprint