Bees under stress: sublethal doses of a neonicotinoid pesticide and pathogens interact to elevate honey bee mortality across the life cycle
Vincent Doublet1*, Maureen Labarussias1, Joachim R. de Miranda², Robin F. A. Moritz1,3, Robert J. Paxton1,3,4
1 Institut für Biologie, Martin-Luther-Universität Halle-Wittenberg, Hoher Weg 8, 06120 Halle (Saale), Germany
2 Department of Ecology, Swedish University of Agricultural Sciences, Uppsala, 750-07 Sweden
3 German Center for Integrative Biodiversity Research (iDiv), Halle-Jena-Leipzig, Leipzig, Germany.
4 School of Biological Sciences, Queen’s University Belfast, Belfast, UK.
* Corresponding author:
Martin-Luther-Universität Halle-Wittenberg
Institut für Biologie/Allgemeine Zoologie
Hoher Weg 8
D-06120 Halle (Saale), Germany
Phone: +493455526500
Fax: +493455527428
Email:
Running title: Pesticide-pathogen interactions on honey bees
Figure S1: Estimated LD50 of BQCV concentration on larval honey bees calculated from a linear mixed model using colony as random effect. The estimated LD50 is 1.53 x 108 genome equivalents (95% ci: = 6.99 x 107/ 1.35 x 109), at eleven days post-infection. Linear regression R² = 0.714.
Figure S2: Mean sugar consumption per bee and per day (±sem) for treatments with and without pesticide, during Experiment 3 (N. ceranae, BQCV and thiacloprid). Presence of thiacloprid had a negative effect on sugar consumption (linear mixed model using colony as random effect; t = -3.998, df = 18, P < 0.001), while infection with pathogens did not: N. ceranae (t = -1.042, df = 18, P = 0.3114) and BQCV (t = -1.833, df = 18, P = 0.0834). The decrease in sugar consumption due to the presence of thiacloprid in the food is of the order of 15% (calculated from the median values of sugar consumption for both groups, without (40.1 µl/bee/day) and with thiacloprid (34.0 µl/bee/day)). The green line represents the sugar consumption of honey bees from the treatments without thiacloprid and the blue line represents the mean consumption of honey bees from the treatments with thiacloprid. Thiacloprid was mixed into the 50% sucrose solution (0.1% acetone) provided ab libitum to honey bee workers across the whole experiment.
Figure S3: Estimated LD50 following chronic exposure of larval honey bees to thiacloprid, calculated from a linear mixed model using colony as random effect. The estimated LD50 is 76.9 mg/kg thiacloprid (95% ci: 60.5 / 99.8), 7 days after the first ingestion. Linear regression R² = 0.913.
Figure S4: Design of the three experiments.
Table S1: Instantaneous risk of death (hazard ratio, ± 95% c.i.) for adult honey bees in each treatment of Experiment 2 (N. ceranae and BQCV) compared to the control treatment, calculated from a Cox proportional hazard mixed model using treatment as fixed effect and colony and cages as nested random effects. In bold are the treatments with a hazard ratio statistically different to the control treatment.
Treatments / Hazard ratio / -95% CI / +95% CI / Z / Pcontrol / - / - / - / - / -
BQCV / 0.642 / 0.112 / 3.668 / -0.50 / 0.620
N. ceranae / 4.389 / 1.081 / 17.821 / 2.07 / 0.039 / *
BQCV + N. ceranae / 16.216 / 4.219 / 62.335 / 4.05 / <0.001 / ***
Table S2: Instantaneous risk of death (hazard ratio, ± 95% c.i.) for adult honey bees in each treatment of Experiment 3 (N. ceranae, BQCV and thiacloprid) compared to the control treatment, calculated from a Cox proportional hazard mixed model using treatment as fixed effect and colony and cages as nested random effects. In bold are the treatments with a hazard ratio statistically different to the control treatment.
Treatments / Hazard ratio / -95% CI / +95% CI / Z / Pcontrol / - / - / - / - / -
N. ceranae / 1.559 / 0.727 / 3.342 / 1.14 / 0.250
BQCV / 1.008 / 0.464 / 2.191 / 0.02 / 0.980
thiacloprid / 0.893 / 0.426 / 1.874 / -0.30 / 0.760
N. ceranae + BQCV / 2.622 / 1.233 / 5.577 / 2.50 / 0.012 / *
BQCV + thiacloprid / 1.568 / 0.761 / 3.232 / 1.22 / 0.220
N. ceranae + thiacloprid / 2.707 / 1.329 / 5.515 / 2.74 / 0.006 / **
N. ceranae + BQCV + thiacloprid / 2.779 / 1.353 / 5.705 / 2.79 / 0.005 / **
Table S3: Coefficient contrast comparisons, adjusted (or not) for multiple comparisons with FDR method, based on the hazard ratio from each treatment in Experiment 3 (N. ceranae, BQCV and thiacloprid). Double treatments were compared to single treatments, while the triple treatment (N. ceranae + BQCV + thiacloprid) was compared with the three doublet treatments. In bold is the comparison that appeared significant without correction for multiple analyses.
Treatment comparisons / Z / P (non-adjusted) / P (adjusted)N. ceranae + BQCV / vs. / N. ceranae BQCV / 2.247 / 0.0246 * / 0.0931
N. ceranae + thiacloprid / vs. / N. ceranae thiacloprid / 0.976 / 0.329 / 0.7876
BQCV + thiacloprid / vs. / BQCV thiacloprid / 1.671 / 0.0947 / 0.3204
N. ceranae + BQCV + thiacloprid / vs. / N. ceranae + BQCV N. ceranae + thiacloprid BQCV + thiacloprid / 0.814 / 0.416 / 0.8765
Table S4: List of primers used for quantification of viruses after propagation, using RT-qPCR (and efficiendy of qPCR). The qPCR efficiency for BQCV quantification after experimental infections in adults and larvae was of 103.3%.
Targets / Primers / Sequences / qPCR efficiency / ReferencesABPV / ABPV-F6548 / TCATACCTGCCGATCAAG / 93,7% / de Miranda et al. (2010a)
KIABPV-B6707 / CTGAATAATACTGTGCGTATC
BQCV / BQCV-qF7893 / AGTGGCGGAGATGTATGC / 97.7% / Locke et al. (2012)
BQCV-qB8150 / GGAGGTGAAGTGGCTATATC / (103.3%)
CBPV / CBPV1-qF1818 / CAACCTGCCTCAACACAG / 90,2% / Locke et al. (2012)
CBPV1-qB2077 / AATCTGGCAAGGTTGACTGG
DWV / DWV-F8668 / TTCATTAAAGCCACCTGGAACATC / 85.6% / Forsgren et al. (2009)
DWV-B8757 / TTTCCTCATTAACTGTGTCGTTGA
IAPV / IAPV-F6627 / CCATGCCTGGCGATTCAC / 95.9% / de Miranda et al. (2010a)
KIABPV-B6707 / CTGAATAATACTGTGCGTATC
KBV / KBV-F6639 / CCATACCTGCTGATAACC / 91.5% / de Miranda et al. (2010a)
KIABPV-B6707 / CTGAATAATACTGTGCGTATC
LSV-1 / qLSV1-F2569 / AGAGGTTGCACGGCAGCATG / 89.3% / Runckel et al. (2011)
qLSV1-R2743 / GGGACGCAGCACGATGCTCA
LSV-2 / qLSV2-F1722 / CGTGCTGAGGCCACGGTTGT / 94.1% / Runckel et al. (2011)
qLSV2-R1947 / GCGGTGTCGATCTCGCGGAC
VDV-1 / VDV-F1409 / GCCCTGTTCAAGAACATG / 86.0% / Locke et al. (2012)
DWV-B1806 / CTTTTCTAATTCAACTTCACC
SBPV / SBPV-F3177 / GCGCTTTAGTTCAATTGCC / 93,8% / de Miranda et al. (2010b)
SBPV-B3363 / ATTATAGGACGTGAAAATATAC
SBV / SBV-qF3164 / TTGGAACTACGCATTCTCTG / 92,1% / Locke et al. (2012)
SBV-qB3461 / GCTCTAACCTCGCATCAAC
N. ceranae / forward / CAATATTTTATTATTTTGAGAGA / 76.2% / vanEngelsdorp et al. (2009)
reverse / TATATTTATTGTATTGCGCGTGCA
References to table S4:
Forsgren E, de Miranda JR, Isaksson M, Wei S, Fries I (2009) Deformed wing virus associated with Tropilaeps mercedesae infesting European honey bees (Apis mellifera). Exp Appl Acarol 47:87-97
de Miranda JR, Cordoni G, Budge G (2010a) The Acute bee paralysis virus-Kashmir bee virus-Israeli acute paralysis virus complex. Journal of Invertebrate Pathology 103:S30-S47
de Miranda JR, Dainat B, Locke B, Cordoni G, Berthoud H, Gauthier L, Neumann P, Budge GE, Ball BV, Stoltz DB (2010b) Genetic characterisation of slow paralysis virus of the honey bee (Apis mellifera L.). Journal of General Virology 91: 2524-2530
Locke B, Forsgren E, Fries I, de Miranda JR (2012) Acaricide treatment affects viral dynamics in Varroa destructor-infested honey bee colonies via both host physiology and mite control. Applied and Environmental Microbiology 78: 227-235
Runckel C, Flenniken ML, Engel JC, Ruby JG, Ganem D, et al. (2011) Temporal analysis of the honey bee microbiome reveals four novel viruses and seasonal prevalence of known viruses, Nosema, and Crithidia. PLoS One 6(6): e20656
vanEngelsdorp D, Evans JD, Saegerman C, Mullin C, Haubruge E, et al. (2009) Colony collapse disorder: a descriptive study. PLoS One 4(8): e6481
Appendix S1: Effect of chronic exposure to a sub-lethal dose of thiacloprid (0.1 mg/kg) and three different doses of BQCV, alone or in combination, on larval honey bee mortality.
A: Instantaneous risk of death (hazard ratio, ± 95% c.i.) for larval honey bees in each treatment of Experiment 1 (BQCV and thiacloprid) compared to the control treatment, calculated from a Cox proportional hazard mixed model using treatment as fixed effect and colony as random effect. In bold are the treatments with a hazard ratio statistically different from the control treatment.
Treatments / Hazard ratio / -95% CI / +95% CI / Z / Pcontrol / - / - / - / - / -
thiacloprid / 1.707 / 0.559 / 5.218 / 0.94 / 0.350
1.4 x 104 BQCV / 0.596 / 0.142 / 2.491 / -0.71 / 0.480
1.4 x 104 BQCV + thiacloprid / 1.732 / 0.567 / 5.292 / 0.96 / 0.340
1.4 x 107 BQCV / 2.450 / 0.864 / 6.950 / 1.68 / 0.093
1.4 x 107 BQCV + thiacloprid / 3.449 / 1.254 / 9.482 / 2.39 / 0.017 / *
1.4 x 109 BQCV / 19.414 / 7.622 / 49.448 / 6.22 / <0.001 / ***
1.4 x 109 BQCV + thiacloprid / 27.385 / 10.752 / 69.750 / 6.95 / <0.001 / ***
B: Instantaneous risk of death (hazard ratio, ± 95% c.i.) for larval honey bees from Experiment 1, treated with 3 different doses of BQCV or thiacloprid, compared to the control treatment, calculated from a Cox proportional hazard mixed model using ‘treatment’ as fixed effect and ‘colony’ as random effect. In bold are the treatment with a hazard ratio statistically different to the control treatment.
Treatments / Hazard ratio / -95% CI / +95% CI / Z / Pcontrol / - / - / - / - / -
thiacloprid / 1.527 / 1.083 / 2.151 / 2.42 / 0.015 / *
1.4 x104 BQCV / 0.850 / 0.381 / 1.900 / -0.40 / 0.690
1.4 x 107 BQCV / 2.184 / 1.126 / 4.235 / 2.31 / 0.021 / *
1.4 x 109 BQCV / 17.514 / 9.652 / 31.780 / 9.42 / <0.001 / ***