- Research article
- Open Access
Transcriptional markers of sub-optimal nutrition in developing Apis mellifera nurse workers
- Vanessa Corby-Harris1, 2Email author,
- Beryl M Jones†1, 3,
- Alexander Walton†1, 4,
- Melissa R Schwan1 and
- Kirk E Anderson1, 2
© Corby-Harris et al.; licensee BioMed Central Ltd. 2014
- Received: 21 September 2013
- Accepted: 4 February 2014
- Published: 15 February 2014
Honey bees (Apis mellifera) contribute substantially to the worldwide economy and ecosystem health as pollinators. Pollen is essential to the bee’s diet, providing protein, lipids, and micronutrients. The dramatic shifts in physiology, anatomy, and behavior that accompany normal worker development are highly plastic and recent work demonstrates that development, particularly the transition from nurse to foraging roles, is greatly impacted by diet. However, the role that diet plays in the developmental transition of newly eclosed bees to nurse workers is poorly understood. To further understand honey bee nutrition and the role of diet in nurse development, we used a high-throughput screen of the transcriptome of 3 day and 8 day old worker bees fed either honey and stored pollen (rich diet) or honey alone (poor diet) within the hive. We employed a three factor (age, diet, age x diet) analysis of the transcriptome to determine whether diet affected nurse worker physiology and whether poor diet altered the developmental processes normally associated with aging.
Substantial changes in gene expression occurred due to starvation. Diet-induced changes in gene transcription occurring in younger bees were largely a subset of those occurring in older bees, but certain signatures of starvation were only evident 8 day old workers. Of the 18,542 annotated transcripts in the A. mellifera genome, 150 transcripts exhibited differential expression due to poor diet at 3d of age compared with 17,226 transcripts that differed due to poor diet at 8d of age, and poor diet caused more frequent down-regulation of gene expression in younger bees compared to older bees. In addition, the age-related physiological changes that accompanied early adult development differed due to the diet these young adult bees were fed. More frequent down-regulation of gene expression was observed in developing bees fed a poor diet compared to those fed an adequate diet. Functional analyses also suggest that the physiological and developmental processes occurring in well-fed bees are vastly different than those occurring in pollen deprived bees. Our data support the hypothesis that poor diet causes normal age-related development to go awry.
Poor nutrition has major consequences for the expression of genes underlying the physiology and age-related development of nurse worker bees. More work is certainly needed to fully understand the consequences of starvation and the complex biology of nutrition and development in this system, but the genes identified in the present study provide a starting point for understanding the consequences of poor diet and for mitigating the economic costs of colony starvation.
- Apis mellifera
The European honey bee (Apis mellifera) contributes substantially to the worldwide economy and ecosystem health by pollinating a wide variety of plants . Honey bee colonies function as a unit or superorganism, where facultatively sterile female workers support the queen and her developing brood through a progression of tasks under temporal, genetic, environmental, and physiological control [2, 3]. Age is intimately tied to development, and shifts in worker behavior over time are accompanied by dramatic physiological, anatomical, and physiological changes [4–6]. Despite the dramatic nature of these shifts, this network is flexible, where behaviors can accelerate, slow, or revert in response to the changing needs of the colony, such as in times of stress or changes in colony demography [3, 7].
Pollen is essential to the diet of larvae and young adult workers [8, 9]. Larvae acquire protein through worker jelly, a rich secretion from nurse workers’ hypopharyngeal glands (HGs). After eclosion, very young worker bees (less than ~3 days of age) receive a diet of regurgitated pollen and worker jelly . Between 3 and 8 days of age, workers transition to a diet of beebread, stored pollen mixed with sugars and microbes [2, 10, 11]. Much of the developing adults’ protein, lipids, and micronutrients come from beebread, and these nutrients stimulate the growth of HGs and the fat body (reviewed in ) in preparation for the physiologically demanding nurse (~8-14 days) and forager (~14-21 days) roles. Nurses support developing brood and young adults through their HG secretions, and foragers rely primarily on nutrient stores to supplement the small amount of nutrients consumed via trophallaxis and while foraging for nectar, water, and pollen to support the hive [10, 12]. Worker pollen consumption increases until nursing age and then subsequently decreases (reviewed in ), so the protein assimilation that precedes nursing is critical to colony growth and ontogeny.
Managed honey bees experience long periods of nutritional stress. The majority of honey bees in the U.S. are managed by migratory beekeepers whose livelihood depends on moving hundreds to thousands of colonies to crops requiring pollination. Spring buildup is a crucial time when honey bee colonies that have survived the winter are prepared for pollination of the first crop of the season. Colonies typically come out of winter with little to no pollen stores because any stores from the previous summer and fall were used to sustain the colony through the harsh winter. During spring buildup or throughout the pollination season beekeepers supplement their hives with homemade combinations of protein, sugars, micronutrients, phagostimulants, and antimicrobials that are an incomplete replacement for natural pollen [10, 13]. Commercially available pollen substitutes are also used and bees consume and colonies grow in response to these substitutes as well as patties containing natural pollen [14, 15]. Nonetheless, recent surveys of beekeepers in small and large commercial operations rank starvation as a major cause of colony loss [16, 17], and so there is still a great need for understanding the basic biology of starvation and for developing markers to assess the sublethal effects of poor diet in young adult honey bees at this critical life stage.
Two recent studies on bees kept in cages away from their hive illustrate the utility of high-throughput methods such as oligonucleotide microarrays and messenger RNA sequencing (mRNA-Seq) for studying the biological signatures of dietary stress in nurse worker honey bees. Using mRNA-Seq in whole abdomens (including the digestive tract), Alaux et al.  found that metabolic pathways involved in nutrient-sensing, metabolism, aging, and immunity were up-regulated in nurses fed pollen, while pathways involved in stress response and the regulation of gene expression were down-regulated. Ament et al.  used oligonucleotide microarrays to study gene expression in the abdominal carcasses (excluding the digestive tract) of nurses and foragers and found that pollen intake caused the activation of nutrient metabolism processes and reduced expression of transcripts involved in glycolysis, growth and development, neurogenesis, reproduction, and muscle contraction. These two studies used different approaches, yet both find that genes controlling transcriptional regulation and cell signaling are down-regulated in bees fed pollen while those controlling biosynthesis and lipid metabolism are up-regulated ([18, 19] and see Additional file 1: Table S1, Additional file 2: Table S2 and Additional file 3: Figure S1). However, it is still unclear (1) whether these responses apply to bees in a natural hive setting and (2) how these signatures of incipient starvation accompany the developmental and physiological changes that occur with age. More work must be done to understand how young adult bees respond to poor diet and to understand the basic biology of nutrition in this system.
We aimed to further understand the transcriptional changes associated with starvation during early adult development. To this end, we assayed 3 day old and 8 day old bees kept in normally functioning hives and caged over only honey (poor diet) or honey and bee bread (rich diet) and allowed trophallaxis with the rest of their colony. Whole transcriptomes from abdominal carcasses (i.e., the fat body, the reduced reproductive organs, exoskeleton, trachea, and muscles) were analyzed to focus on changes in the A. mellifera abdomen exclusive of the digestive tract. We hypothesized that (1) starvation and aging would greatly impact gene expression and physiology, (2) that starvation would impact young bees differently than older bees, and that (3) early adult development (aging) would be affected by diet. We also aimed to provide testable predictions for research on development and nutrition in A. mellifera and develop targets for improving the productivity of bee colonies under nutritional stress. Here, we report that starvation greatly affects the expression of genes related to nurse physiology and development and that the processes that occur as adults develop into nurse bees are altered by sub-optimal diet.
Bees fed pollen had larger hypopharyngeal glands (HG) compared to those that were not fed pollen
Due to the significant interaction between diet and colony in the full model discussed above, the effect of diet and age on HG size was also investigated for each colony separately. For colony A, the effect of diet (F1,4=217.65, p=0.0001) and the interaction between age and diet (F1,2=34.25, p=0.0280) were significant but the effect of age (F1,4=6.80, p=0.0596) was not (Additional file 4: Figure S2). For colony B, the effect of diet (F1,4=189.38, p=0.0002) and the interaction between age and diet (F1,2=14.05, p=0.0200) were significant but the effect of age (F1,4=2.85, p=0.1667) was not (Additional file 4: Figure S2). For colony C, the effect of age (F1,4=72.60, p=0.0010) and diet (F1,4=26.49, p=0.0068) were significant, but the effect of the age × diet interaction (F1,2=0.50, p=0.5511) was not (Additional file 4: Figure S2).
The HG sizes of bees that were used to evaluate the predictive power of the mRNA-Seq results with qPCR (but that were not part of the mRNA sequencing experiment), were also measured. Only 8d old bees were used in this phase of the experiment, yielding 6 different classes of bees (2 diets × 3 colonies). For each of these 6 different treatment combinations, the HGs of 5 bees were dissected and 10 randomly selected acini were measured for each gland. The HGs from 8d old bees fed pollen were significantly larger than those from 8d old bees fed only honey (8d rich diet mean acinus area: 0.015 mm2±0.002 s.e., 8d poor diet mean acinus area: 0.007 mm2±0.0003 s.e.; F2,4=24.48, p=0.0057).
Sequencing statistics and broad patterns
Across all twelve libraries (2 ages × 2 diets × 3 host colonies), approximately 149.4 million paired end reads passed the initial quality control filters and approximately 103 million paired end reads mapped to the A. mellifera genome (Additional file 5: Table S3). Across the twelve libraries, an average of 8.58 million (67.5%) paired end reads mapped to the A. mellifera genome per library. In total, the expression of 67,002 exons (non-singleton exons; see Methods) mapping to 12,340 transcripts (containing ≥2 exons; see Methods) was differentially expressed (Additional file 6: Table S4 ). The interaction between diet and age did not significantly impact the expression of any exons, and so results are presented below for exons and transcripts that changed due to the main effect of diet, the main effect of age, the effect of diet at either 3d or 8d, or the effect of age in bees fed a rich or poor diet.
Starvation caused drastic differences in gene expression, and these differences are most evident when the effect of starvation is assessed for each age class separately
Differentially expressed exons, transcripts, and orthologues for each category investigated
A. mellifera transcripts placed with D. melanogaster orthologuesD
Percent of total transcripts identified in the present study with an orthologueE
Transcript-based analysis of expression with longer periods of starvation or increasing age
% Transcripts decreasedB
Difference in % decrease?
Biological process (BP) terms and annotation clusters down-regulated in pollen deprived bees at 3d and 8d
BP term or cluster
X 2 adj A
Cell adhesion (BP)C
FBgn0034660, FBgn0000547, FBgn0031879, FBgn0000464, FBgn0000117, FBgn0026562
Cell adhesion (BP)C, regulation of cell shape (BP)C
FBgn0000547, FBgn0000464, FBgn0000117
Dorsal closure (BP)C
FBgn0000299, FBgn0000547, FBgn0000117
Lateral inhibition (BP)C
FBgn0010473, FBgn0000547, FBgn0243514, FBgn0053196
Low-density lipoprotein (LDL) receptor class A repeat
FBgn0031879, FBgn0053196, FBgn0261341
EGF-like region, conserved site
FBgn0003137, FBgn0026403, FBgn0031879, FBgn0243514, FBgn0053196
Calcium ion binding (MF)
FBgn0026403, FBgn0031879, FBgn0000079, FBgn0031464, FBgn0037963, FBgn0053196, FBgn0026562
Immunoglobulin subtype, Immunoglobulin-like 6
FBgn0003137, FBgn0010473, FBgn0000547, FBgn0005666, FBgn0000464, FBgn0004369
IPR003599, IPR007110, IPR003961, IPR008957
Immunoglobulin subtype, Immunoglobulin-like, Fibronectin, type III, Fibronectin type III domain
FBgn0010473, FBgn0000547, FBgn0005666, FBgn0000464, FBgn0004369
IPR001660, IPR010993, IPR011510, IPR021129
Sterile alpha motif domain, Sterile alpha motif homology, Sterile alpha motif, type 2, Sterile alpha motif, type 1
FBgn0000182, FBgn0031762, FBgn0039075
EF-Hand 1, calcium-binding site
FBgn0031464, FBgn0013809, FBgn0026562
Serine-type endopeptidase inhibitor activity (MF)
FBgn0003137, FBgn0026721, FBgn0053196
GO:0006030, GO:0008061, IPR002557
Chitin metabolic process (BP)C, chitin binding (MF), Chitin binding domain
FBgn0027600, FBgn0051973, FBgn0261341
Glycoside hydrolase, superfamily
FBgn0053080, FBgn0000079, FBgn0026415
Actin cytoskeleton organizationD
Motor axon guidanceD
Carbohydrate metabolic processD
Homophilic cell adhesionD
Response to oxidative stressD
The genes were down-regulated upon starvation exclusively at 3d or 8d of age (Figure 2, left side, rich>poor) were assessed to determine whether the processes associated with starvation differed in young versus old bees. Transcripts down-regulated due to starvation in young (3d old) but not old (8d old) bees included apidermin 3, cuticular proteins, and ecdysis triggering hormone (Additional file 6: Table S4 ) but no biological processes or annotation clusters. Transcripts down-regulated in pollen deprived bees at 8d but not at 3d included glutathione S-transferases, major royal jelly (MRJP) proteins 1 through 9, hexamerins, DNA methyltransferases, cytochrome p450s, and immune genes (Additional file 6: Tables S4 ) but were not associated with any biological processes or gene annotation clusters.
Genes that were up-regulated upon starvation exclusively at 3d or 8d of age were next assessed (Figure 2, right side, poor>rich) to determine whether the increases in gene expression that occurred differed in young and old bees. One transcript encoding transient receptor potential-gamma protein-like (XM_001122469) was up-regulated in pollen deprived bees exclusively at 3d (Additional file 6: Tables S4 ) but this transcript was not orthologous to a D. melanogaster gene. Transcripts up-regulated in pollen deprived bees exclusively at 8d included immune recognition genes, cytochrome P450s, cuticular proteins, MRJPs, hexamerins, DNA methyltransferases, and one microRNA (Additional file 6: Tables S4 ) but were not associated with any biological processes or annotation clusters.
Starvation alters normal age-associated nurse development
Biological process (BP) terms and annotation clusters significantly down-regulated due to the main effect of age
X 2 adj A
ATP binding (MF)
FBgn0260990, FBgn0261014, FBgn0032243, FBgn0025743, FBgn0026059, FBgn0000723, FBgn0030343, FBgn0015615, FBgn0036486, FBgn0051729
FBgn0031390, FBgn0038927, FBgn0031266
Dorsal closure (BP)C
FBgn0010434, FBgn0000723, FBgn0044323
IPR001680, IPR011046, IPR019775
WD40 repeat, WD40 repeat-like-containing domain, WD40 repeat, conserved site
FBgn0037758, FBgn0029903, FBgn0044323, FBgn0038927, FBgn0043362
IPR001680, IPR011046, IPR019775, IPR020472, IPR019782, IPR017986, IPR019781
WD40 repeat, WD40 repeat-like-containing domain, WD40 repeat, conserved site, G-protein beta WD-40 repeat, WD40 repeat 2, WD40-repeat-containing domain, WD40 repeat, subgroup
FBgn0029903, FBgn0044323, FBgn0038927, FBgn0043362
WD40 repeat-like-containing domain
FBgn0037758, FBgn0029903, FBgn0024698, FBgn0044323, FBgn0038927, FBgn0043362
Actin binding (MF)
FBgn0035347, FBgn0010434, FBgn0029903
FBgn0260990, FBgn0030674, FBgn0043362, FBgn0031266
Mushroom body developmentD
Transcripts that were up- or down-regulated with age were investigated in pollen deprived and well-fed bees separately to determine whether the signatures of normal age-associated development differed in underfed versus well-fed bees. Transcript expression decreased with age more frequently in bees fed a poor diet (77 out of 91; 79.3%) compared to bees fed a rich diet (39 out of 62; 62.9%; X21=8.33, p=0.0039; Table 2 and Additional file 6: Table S4 ). Exon-based analyses also supported this trend: 71.9% of exons were down-regulated with age in pollen deprived bees compared to the 49.2% down-regulated with age in bees fed pollen (X21=22.59, p<0.0001; Additional file 7: Table S5).
The overlap between transcripts differentially regulated due to age in bees fed either diet was investigated to determine what processes that were up- or down-regulated with age were common between underfed and well-fed bees. Only one transcript (with no orthologue), encoding a hypothetical protein (Additional file 6: Table S4 ), was down-regulated with age in bees fed both diets. Transcripts up-regulated with age in bees fed either diet included apidaecin, cytochrome P450 342A1, and hexamerin 110 (Additional file 6: Table S4 ). These transcripts were associated with one orthologue that was not associated with a biological process or annotation cluster.
Biological process (BP) terms down-regulated with age only in bees fed the rich diet
Χ 2 adj A
Open tracheal system development
Chitin metabolic process
FBgn0051973, FBgn0260653, FBgn0261341
Regulation of tube length, open tracheal system
Biological process (BP) terms and annotation clusters down-regulated with age only in pollen deprived bees
BP term or cluster
X 2 adj A
Dorsal closure (BP)C
FBgn0044323, FBgn0010434, FBgn0000723
FBgn0031266, FBgn0021847, FBgn0030674
Lateral inhibition (BP)C
FBgn0029685, FBgn0015615, FBgn0025743
FBgn0031266, FBgn0043362, FBgn0260990, FBgn0030674
ATP binding (MF)
FBgn0051729, FBgn0036486, FBgn0261014, FBgn0032243, FBgn0015615, FBgn0025743, FBgn0026059, FBgn0260990, FBgn0000723
GO:0005524, GO:0006468, IPR002290, IPR000719
ATP binding (MF),protein phosphorylation (BP), Serine/threonine-protein kinase domain, Protein kinase, catalytic domain
FBgn0025743, FBgn0260990, FBgn0000723
WD40 repeat-like-containing domain
FBgn0024698, FBgn0044323, FBgn0027518, FBgn0043362, FBgn0037758, FBgn0029903, FBgn0038927
Actin binding (MF)
FBgn0010434, FBgn0029903, FBgn0035347
Nuclear mRNA splicing, via spliceosome (BP)C
FBgn0031266, FBgn0034002, FBgn0038927
FBgn0031266, FBgn0038927, FBgn0031390
FBgn0025743, FBgn0260990, FBgn0000723
Mushroom body developmentD
Compound eye developmentD
Potassium ion transportD
Biological processes up-regulated with age only in bees fed the rich diet
Χ 2 adj A
Innate immune response
Peptidoglycan catabolic process
Targeted rt-PCR of differentially expressed genes confirmed the mRNA-sequencing results
Pollen is a critical component of the diet for young adult bees and can greatly impact their development. Surveys of beekeepers suggest that starvation is a major cause of colony losses and our understanding of how starvation affects bees at the critical young adult stage is incomplete. To understand the basic biology of pollen deprivation and to develop transcriptional markers of starvation in developing nurse worker bees, we used mRNA-seq to measure changes in gene expression in 3d and 8d old bees fed a diet of honey and stored pollen within the hive compared to a restricted diet containing honey but no pollen. Many interesting results became apparent when diet-related transcripts were studied separately for bees aged 3d or 8d and by determining which transcripts were differentially expressed at either 3d or 8d of age. Substantial changes in gene expression were observed when bees were deprived of pollen and these changes were more numerous as bees were deprived of pollen for longer periods of time. Additionally, normal patterns of gene expression that accompanied age-related changes in development were altered in underfed bees. The data suggest that nutrients in pollen regulate the key developmental transitions that honey bees experience during early adult life as they prepare for critical brood care tasks.
Transcripts that were down-regulated in pollen deprived bees at both ages (i.e., due to the model main effect of age) included those with potential roles in development and muscle and motor function. These results agree with those of Ament et al., who also found that the transcription of growth, development and muscle contraction genes was reduced in poorly fed nurses. It appears that poor diet is costly to the expression of genes involved in muscle function and growth from a rather early age. Whether this affects the ability of nurse bees to perform in-hive tasks or whether the lasting effects of starvation are evident in foragers remains to be studied. However, it is possible that traits such as foraging ability may be affected by poor diet in early adulthood.
Many more transcripts changed due to starvation in older bees (20,376 transcripts) compared to younger bees (173 transcripts). Of the transcripts where expression changed due to poor diet, there was more frequent down-regulation of gene expression in younger bees (78.5%) compared to older bees (55.9%). Transcriptional changes occurring due to starvation at 3d were largely a subset of those occurring at 8d. Many more diet-related differences were observed only in bees deprived of pollen for 8d, but it is unclear whether the genes differentially expressed exclusively at 8d are so due to the longer period of starvation or because specific processes that are turned on only at later ages are also impacted by diet. The expression of MRJPs, immune genes, and genes encoding detoxification enzymes declined due to poor diet exclusively at later ages. This group of down-regulated transcripts included several members of the CYP4 and CYP6 cytochrome P450 families and one Delta GST, which have roles in insecticide detoxification , chitin-binding proteins that maintain the integrity of the peritrophic membrane, which plays a defensive role against chemicals and parasites , genes such as PGRP S3 and cytP450 4G11 whose down-regulation is associated with colony collapse disorder CCD; , and the trancripts dicer and argonaute that mediate immunity towards viruses that are sometimes prevalent in failing colonies [24, 25]. One potential consequence, therefore, of pollen deprivation in nurses but not newly eclosed bees is reduced resistance to disease and insecticides.
Aging was associated with the overall decline of processes such as development, transcription, and mRNA splicing, and this decline was exacerbated in pollen deprived bees. Age-related transcriptional decline was more frequent in underfed bees (Tables 2 and Additional file 7: Table S5), and the processes occurring as well-fed bees aged were distinct from bees deprived of pollen (Figure 3). Bees fed a rich diet up-regulated genes involved in immunity and down-regulated those involved in tracheal system development and chitin metabolism, possibly as part of a developmental trajectory that prepares them for tasks outside of the hive. Poorly fed bees did not differentially regulate any of the age-related processes occurring in bees fed a normal diet. Pollen deprived bees showed down-regulation of genes involved in transcription and mRNA splicing and a non-coding RNA potentially involved in transcriptional regulation AncR-1; . Loss of transcriptional regulation and the dysregulation of gene expression are associated with age-related functional decline in a variety of taxa ([27–30]; but see ). The data therefore suggest that pollen deprivation may cause normal age-related transcriptional regulation to go awry, leading to various downstream effects. For example, aging bees deprived of pollen exhibited the reduced expression of genes involved in mushroom body and compound eye development. If such signals from the fat body impact brain signaling and development, this decrease could be a sign of diet-induced early foraging, as (1) early foraging is associated with nutrient restriction , (2) reduced volume of Kenyon cells in the mushroom body is associated with the transition to later age tasks such as foraging [32–34] and (3) the mushroom body is where sensory signals are integrated .
Further support for the hypothesis that aging is dysregulated in pollen-deprived bees becomes evident from comparing our results to examples presented in the literature, such as juvenile hormone esterase (JHE; which degrades juvenile hormone), vitellogenin, and hexamerin 70a. Nurses have highest levels of JHE in order to prevent the switch to foraging , as high levels of juvenile hormone are associated with the onset of foraging. We observed that for both fed and pollen starved bees JHE expression is higher in 8d old bees compared to 3d old bees. However, the magnitude of this increase was greater in starved bees and this increased expression with age was only significant in starved bees. Vitellogenin is very high in newly emerged workers, decreases slightly at days 1 and 2 post-eclosion, and then peaks at 3 days of age . It then steadily decreases from this highest level until approximately 20 days of age [38, 39]. In the present study, vitellogenin expression decreased as expected in well-fed bees aging from 3 days to 8 days old, but the difference was not significant. However, starved bees showed the opposite pattern, with expression increasing as they aged from 3d to 8d, and the change in expression was significant. Cunha et al.  find that titres of the hexamerin 70a protein decrease a very small amount, if any between emergence and 7d of age. We saw no difference in the expression of hexamerin 70a gene in old versus young bees when they were well-fed. However, when deprived of pollen, expression was significantly lowered as the bees aged. The aberrant expression of these genes in starved but not well-fed bees again suggests that aging is dysregulated in starved bees.
The expression of relatively few genes changed with respect to the main effect of age, yet there was measurable overlap in the patterns of expression when we compared the genes that were up- or down-regulated with age in the poor versus rich diets (for each diet analyzed separately; Figure 3). The transcripts that differed in expression due to the main effect of age showed similar patterns of significant up- or down-regulation for both diets and the magnitude of this change was the same for the different diets. Because we were interested in comparing the aging process in starved versus well-fed bees, we took the additional approach of determining what genes were differentially expressed due to age considering each diet separately. Figure 3 shows the overlap between diets in the transcripts that changed with age and this change was in the same direction for each of the diets. The main reason that these genes were not also significantly affected by the main effect of age was that the magnitude of the age-related change differed between diets, particularly in genes that were down-regulated as bees aged (Additional file 8: Figure S3). This difference in the magnitude of the age-related change in expression that is diet-dependent provides additional support for our hypothesis that the aging process itself differs due to diet.
We found that substantial changes in gene expression occurred during early adult development and due to sub-optimal pollen intake in early adulthood. Diet-induced changes in gene transcription occurring in early age were largely a subset of those occurring at later age, but most signatures of starvation were only evident in older bees. The aging process itself differed when bees were deprived of pollen and the data are consistent with the hypothesis that poor diet causes normal age-related development to go awry. The preponderance of evidence from this study and other recent studies [18, 19] indicates that adequate diet is critical to early worker development.
Bees and experimental manipulations
Three host colonies of Apis mellifera maintained for several months at the Carl Hayden Bee Research Center in Tucson, Arizona were used for dietary manipulations. The three host colonies were untreated (no miticides or antibiotics) 10 frame colonies with approximately 18,000 adults, 1 full frame face of brood, and 3.5 full frame faces of food stores. To collect newly emerged workers, sealed brood frames from approximately five colonies different from the host colonies used in the experiment were placed in a temperature-controlled room (approximately 34°C) and all adults that emerged in an 18 hour period were collected. These callow bees (≤ 18 hours of age) were marked using a small dot of non-toxic paint on their thorax and were added to the three host colonies, confined to each of two diets (a poor diet consisting of only stored honey or a rich diet of stored honey and bee bread) at a density of approximately one bee per cm2 (~40-60 bees per cage) using wire cages pushed into drawn comb over either diet. Bees were confined for 3 or 8 days in July and August of 2011. To sum, each of the three host colonies contained two dietary treatments, and the bees were sampled after being confined to that treatment for either 3 or 8 days, yielding three replicate (host) colonies for each diet (poor versus rich) by age (3d versus 8d) combination. Sampled bees were flash frozen in liquid nitrogen and maintained at -80°C until further processing.
Hypopharyngeal gland (HG) measurements
Hypopharyngeal glands (HGs) of bees typically grow in preparation for nursing duties and this growth is impacted by the amount and source of dietary protein [10, 41]. Our experimental design allowed trophallaxis (food sharing) with bees exterior to the treatment cage that had access to protein. To confirm that the bees caged over honey alone were indeed protein restricted compared to bees caged over the rich diet containing beebread and honey despite their access to other bees in the colony, the acini of HGs in bees aged 3d or 8d that were fed either diet were measured. HGs were dissected from approximately 5 bees per colony for each diet by age combination. 10 randomly selected acini per gland were visualized at 8X magnification. Acini area (mm2) was measured using the Leica Applications Suite v.3.8.0 software. Measurements were analyzed using a mixed model repeated measures ANOVA testing for the effect of diet, colony, age, the two-way interactions between colony and diet, age and diet, age and colony, and the three-way interaction between age, colony, and diet. Each acinus was a repeated measure taken on the same bee and a compound symmetry covariance structure was employed to model the correlation among measures taken on the same bee. A post-hoc analysis was performed using the Tukey HSD test to determine whether there were significant differences in acini area between diets (rich versus poor diet) for each age (3d and 8d old) and between ages for each diet.
RNA extraction and library preparation
Abdominal carcasses were dissected from three bees per colony per treatment by removing their head and wings before making a dorsal incision through their abdominal cuticle, exposing the abdominal organs and forming a cup structure where RNAlater (Life Technologies) was added. After removing the digestive tract, the remaining tissue (i.e., the fat body, reduced reproductive organs, exoskeleton, trachea, and muscles) was separated from the thorax and this abdominal carcass was preserved at -80°C for subsequent RNA extraction. Three bees per host colony were pooled and total RNA was extracted using TriReagent (Invitrogen) according to the manufacturer’s specifications. Total RNA integrity was confirmed using Agilent’s 2100 Bioanalyzer.
Twelve mRNA sequencing (mRNA-Seq) libraries (2 ages × 2 diets × 3 replicate (host) colonies) were prepared using Illumina’s TruSeq RNA Sample Preparation Kit according to the manufacturer’s protocols. Briefly, poly-A containing mRNA was purified using poly-T oligo-attached magnetic beads and was subsequently fragmented and primed for first-strand cDNA synthesis with random hexamer primers. After degrading the RNA template used for first-strand cDNA synthesis, synthesis of the second cDNA strand followed, yielding a double-stranded cDNA molecule. Adapters were ligated to both ends of the double-stranded cDNA molecules and adapter-ligated cDNAs were enriched using 15 cycles of PCR using adapter-specific primers. The libraries were validated according to the manufacturer’s protocol and the approximately 350 bp fragments were isolated from a 6% Tris Base -Boric Acid-EDTA (TBE) PAGE gel and run through the Illumina sequencer for 2 × 100 cycles at a starting concentration of 12 pM per library.
Analysis of paired end sequence data
Adaptor sequences were removed from each sequence and poor-quality reads were excluded using Trimmomatic  prior to the analysis of the ~150 base pair reads. The size distribution of the original transcripts for which sequence data was taken was first estimated by aligning the reads from each library separately to all publicly available A. mellifera RNA sequences listed in GenBank using the Burrows-Wheeler Aligner (BWA; ). Using the results from the BWA alignment, the expected mean and standard deviation of the inner distance between pairs was set for each library, and the reads were subsequently aligned to version 4.5 of the A. mellifera genome  with TopHat  version 1.4.0. The genome annotation contained NCBI reference sequence annotations and ab initio predictions based on the A. mellifera version 4.5 genome . First, we analyzed a full model to determine whether the effects of age, diet, and the interaction between age and diet significantly impacted gene expression using the edgeR package  and a Benjamini-Hochberg correction for multiple testing  at a 5% false discovery rate. To test whether the effect of diet was different for 3d old versus 8d old bees, the effect of diet was investigated for each age separately. To test whether the effects of aging on gene expression differed with respect to diet, the effect of aging was studied separately for bees fed pollen and those that were not fed pollen.
Statistical analyses of differential expression yielded a corrected significance value for each exon that mapped to each mRNA transcript in the Apis mellifera version 4.5 genome annotation . If ≥2 exons mapped to the same transcript, we reasoned that the entire mRNA transcript (irrespective of isoform) was differentially expressed. Therefore, single exon exon transcripts (i.e., the only significantly differentially expressed exon mapping to an annotated transcript) were eliminated from further analyses. This approach was conservative because it eliminated the occurrence of false positives but came at a cost because it also removed single-exon transcripts from further analyses. In addition, we did not analyze alternative splicing events, which were beyond the scope of the study.
The significantly differentially expressed transcripts (containing ≥2 exons) were subjected to further characterization. Drosophila melanogaster genes orthologous to the differentially expressed A. mellifera transcripts were identified by searching for reciprocal best BLAST hits between A. mellifera mRNA sequences and D. melanogaster proteins. A FlyBase gene ID (FBgn)  was assigned to all A. mellifera transcripts where a D. melanogaster orthologue was identified. GeneCoDis3 [50–52] was then employed to find enriched biological process gene ontology (BP GO) terms and gene annotation co-occurrence clusters (i.e., annotation clusters) that were significantly associated with the differentially expressed orthologues [50–52]. In addition to single BP GO annotations, annotation clusters represented significant relationships among biological process, molecular function, InterPro protein domain, and KEGG pathway annotations that were associated with the differentially expressed orthologoues [50–52]. A reference background list of all D. melanogaster genes orthologous to all A. mellifera RNAs was used to determine significant enrichment of BP GO terms and annotation clusters using a chi-square (X2) test corrected for multiple testing . Significantly enriched BP GO terms that contained only one orthologue were discarded.
The annotated transcripts (containing ≥2 exons) and orthologues that exhibited differential expression were compared. To address the hypothesis that diet and early adult development affect A. mellifera gene expression and physiology, annotated transcripts that increased or decreased due to starvation or as individuals aged from 3d to 8d old were compared and biological processes and annotation clusters corresponding to differentially expressed orthologues were analyzed for significant enrichment. To address the hypothesis that starvation affects younger and older workers differently, comparisons were made between annotated transcripts and orthologues that were affected by diet at either 3d or 8d of age. Venn Diagrams were constructed using the annotated transcripts that were differentially expressed due to diet at 3d and those that were differentially expressed due to diet at 8d. To address the last hypothesis that development and aging are affected by diet, comparisons were made between annotated transcripts and orthologues that changed with age for bees fed either a rich diet or those fed a poor diet. Venn Diagrams were constructed using the annotated transcripts that were differentially expressed due to age for the rich diet and those that were differentially expressed due to age in bees deprived of pollen.
rt-PCR validation of gene expression changes due to diet
To assess the predictive power of the mRNA-seq analyses, we measured the expression of six genes – vermiform (XM623720), cdc42 (XM_394608), E75 (NM_001080110), GTP-binding protein 10 (XM_396976), vitellogenin (NM_001011578), glucocerebrosidase transcript variant 1 (XM_393207; Additional file 11: Table S8) – that were significantly differentially expressed based on the previous mRNA-seq experiment. Fat bodies of 8d old bees from three replicate host colonies (N=3) were used. These bees were not used in the previous mRNA-seq experiment, but the colonies they came from shared similar population sizes, age distributions, and distribution of food stores to the three colonies used in the mRNA-seq experiments and to each other. Dietary manipulations, fat body dissections, and RNA isolations were as described above except RNA integrity was not assayed. To test that the dietary manipulations were successful, HGs were dissected and acini were measured in the manner described above. RNA was isolated from three pooled fat bodies per colony for each age and for each dietary treatment, yielding a total of six samples (3 host colonies x 2 diets). Isolated RNA was subjected to a DNase treatment (Ambion) and the lack of genomic DNA contamination was verified by performing a PCR on the RNA with A. mellifera actin (AB023025.1; Additional file 11: Table S8) and the following cycling conditions: 3 minutes at 94°C, 35 cycles of 45 seconds at 94°C, 45 seconds at 57°C, and 45 seconds at 72°C, and a final extension of 72°C for 10 minutes. cDNA was synthesized using the iScript cDNA synthesis kit (Bio-Rad) according to the manufacturer’s protocol. The amplification efficiencies of all genes ranged between 95% and 105%. For each gene, a two-step qPCR was performed on the cDNA using SsoFast EvaGreen Supermix (Bio-Rad) according to the manufacturer’s protocol and the following cycling conditions (annealing temperatures, n, are listed in Additional file 11: Table S8): 3 minutes at 95°C, 45 cycles of 10 seconds at 95°C and 10 seconds at n°C, followed by a melt curve from 55°C to 95°C to confirm the lack of contamination and/or non-specific amplification. For each of the three colonies per diet, the threshold cycle (CT) values from three technical replicates were averaged for each target gene and two standards: A. mellifera actin (AB023025.1) and GAPDH (XM_393605; Additional file 11: Table S8). The average CT values of each target gene were normalized relative to the mean CT of each standard separately, accounting for primer amplification efficiencies . Significant differences between the diets were determined using a Mann–Whitney/Wilcoxon test  and results were significant only if results were different from one for both of standards (actin and GAPDH).
Availability of supporting data
Additional file 6: Table S4 supporting the results of this article is available in the Dryad Digital Repository http://doi.org/10.5061/dryad.pg2kb. All reads were submitted to the NCBI Sequence Read Archive under accession number SRA056350.
The authors acknowledge Laurel Johnstone, Troy Comi, and Susan Miller for programming advice, the University of Arizona Genetics Core (Arizona Research Laboratories, Division of Biotechnology, http://uagc.arl.arizona.edu/), and the staff of the Carl Hayden Bee Research Center. We also thank Bruce Eckholm, Brendon Mott, Anna Himler, Kate Aronstein, Jay Evans and Gloria DeGrandi-Hoffman for helpful discussion and comments on previous versions of the manuscript. The ARS/USDA is an equal opportunity employer and provider.
- Morse R, Calderone N: The value of honey bee pollination in the United States. Bee Culture. 2000, 128 (18): 1-15.Google Scholar
- Winston ML: The Biology of the Honey Bee. 1987, Cambridge, Massachusetts: Harvard University PressGoogle Scholar
- Robinson GE: Regulation of division of labor in insect societies. Annu Rev Entomol. 1992, 37: 637-665. 10.1146/annurev.en.37.010192.003225.PubMedView ArticleGoogle Scholar
- Chan QW, Mutti NS, Foster LJ, Kocher SD, Amdam GV, Wolschin F: The worker honeybee fat body proteome is extensively remodeled preceding a major life-history transition. PLoS One. 2011, 6 (9): e24794-10.1371/journal.pone.0024794.PubMed CentralPubMedView ArticleGoogle Scholar
- Keller I, Fluri P, Imdorf A: Pollen nutrition and colony development in honey bees: part I. Bee World. 2005, 86 (1): 3-10.View ArticleGoogle Scholar
- Toth AL, Kantarovich S, Meisel AF, Robinson GE: Nutritional status influences socially regulated foraging ontogeny in honey bees. J Exp Biol. 2005, 208 (Pt 24): 4641-4649.PubMedView ArticleGoogle Scholar
- Page RE, Robinson GE, Britton DS, Fondrk MK: Genotypic variability for rates of behavioral development in worker honeybees (Apis mellifera L.). Behav Ecol. 1992, 3 (2): 173-180. 10.1093/beheco/3.2.173.View ArticleGoogle Scholar
- Haydak MH: Brood rearing by honeybees confined to a pure carbohydrate diet. J Econ Entomol. 1935, 28: 657-660.View ArticleGoogle Scholar
- Haydak MH: The influence of pure carbohydrate diet on newly emerged honeybees. Ann Entomol Soc Am. 1937, 30: 252-262.View ArticleGoogle Scholar
- Crailsheim K: The protein balance of the honey bee worker. Apidologie. 1990, 21: 417-429. 10.1051/apido:19900504.View ArticleGoogle Scholar
- Anderson KE, Sheehan TH, Eckholm BJ, Mott BM, DeGrandi-Hoffman G: An emerging paradigm of colony health: microbial balance of the honey bee and hive (Apis mellifera). Insect Soc. 2011, 58 (4): 431-444. 10.1007/s00040-011-0194-6.View ArticleGoogle Scholar
- Crailsheim K: Interadult feeding of jelly in honeybee (Apis mellifera L.) colonies. J Comp Physiol B: Biochem Syst Environ Physiol. 1991, 161: 55-60. 10.1007/BF00258746.View ArticleGoogle Scholar
- Standifer LN, Moeller FE, Kauffeld NM, Herbert EW, Shimanuki H: Supplemental Feeding of Honey Bee Colonies. 1977, United States: Deparment of AgricultureGoogle Scholar
- DeGrandi-Hoffman G, Wardell G, Ahumada-Segura F, Rinderer T, Danka R, Pettis J: Comparisons of pollen substitute diets for honey bees: consumption rates by colonies and effects on brood and adult populations. J Apic Res. 2008, 47: 265-270.View ArticleGoogle Scholar
- Mattila HR, Otis GW: Influence of pollen diet in spring on development of honey bee (Hymenoptera: Apidae) colonies. J Econ Entomol. 2006, 99 (3): 604-613. 10.1603/0022-0493-99.3.604.PubMedView ArticleGoogle Scholar
- van Engelsdorp D, Hayes J, Underwood RM, Pettis J: A survey of honey bee colony losses in the U.S., fall 2007 to spring 2008. PLoS One. 2008, 3 (12): e4071-10.1371/journal.pone.0004071.PubMedView ArticleGoogle Scholar
- van Engelsdorp D, Hayes J, Underwood R, Pettis JS: A survey of honey bee colony losses in the United States, fall 2008 to spring 2009. J Apiculture Res. 2010, 49 (1): 7-14. 10.3896/IBRA.1.49.1.03.View ArticleGoogle Scholar
- Alaux C, Dantec C, Parrinello H, Le Conte Y: Nutrigenomics in honey bees: digital gene expression analysis of pollen's nutritive effects on healthy and varroa-parasitized bees. BMC Genomics. 2011, 12: 496-10.1186/1471-2164-12-496.PubMed CentralPubMedView ArticleGoogle Scholar
- Ament SA, Chan QW, Wheeler MM, Nixon SE, Johnson SP, Rodriguez-Zas SL, Foster LJ, Robinson GE: Mechanisms of stable lipid loss in a social insect. J Exp Biol. 2011, 214 (Pt 22): 3808-3821.PubMed CentralPubMedView ArticleGoogle Scholar
- Corby-Harris V, Jones BM, Walton A, Schwan MR, Anderson KE: Transcriptional markers of sub-optimal nutrition in developing Apis mellifera nurse workers (Table S4. Table of differentially expressed genes). 2013,http://doi.org/10.5061/dryad.pg2kb,Google Scholar
- Claudianos C, Ranson H, Johnson RM, Biswas S, Schuler MA, Berenbaum MR, Feyereisen R, Oakeshott JG: A deficit of detoxification enzymes: pesticide sensitivity and environmental response in the honeybee. Insect Mol Biol. 2006, 15 (5): 615-636. 10.1111/j.1365-2583.2006.00672.x.PubMed CentralPubMedView ArticleGoogle Scholar
- Merzendorfer H, Zimoch L: Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases. J Exp Biol. 2003, 206 (24): 4393-4412. 10.1242/jeb.00709.PubMedView ArticleGoogle Scholar
- Johnson RM, Evans JD, Robinson GE, Berenbaum MR: Changes in transcript abundance relating to colony collapse disorder in honey bees (Apis mellifera). Proc Natl Acad Sci USA. 2009, 106 (35): 14970-14795.View ArticleGoogle Scholar
- Cornman RS, Tarpy DR, Chen Y, Jeffreys L, Lopez D, Pettis JS, van Engelsdorp D, Evans JD: Pathogen webs in collapsing honey bee colonies. PLoS One. 2012, 7 (8): e43562-10.1371/journal.pone.0043562.PubMed CentralPubMedView ArticleGoogle Scholar
- Cox-Foster DL, Conlan S, Holmes EC, Palacios G, Evans JD, Moran NA, Quan P-L, Briese T, Hornig M, Geiser DM, et al: A metagenomic survey of microbes in honey bee colony collapse disorder. Science. 2007, 318 (5848): 283-287. 10.1126/science.1146498.PubMedView ArticleGoogle Scholar
- Sawata M, Takeuchi H, Kubo T: Identification and analysis of the minimal promoter activity of a novel noncoding nuclear RNA gene, AncR-1, from the honeybee (Apis mellifera L.). RNA. 2004, 10 (7): 1047-1058. 10.1261/rna.5231504.PubMed CentralPubMedView ArticleGoogle Scholar
- Chambers SM, Shaw CA, Gatza C, Fisk CJ, Donehower LA, Goodell MA: Aging hematopoietic stem cells decline in function and exhibit epigenetic dysregulation. PLoS Biol. 2007, 5 (8): e201-10.1371/journal.pbio.0050201.PubMed CentralPubMedView ArticleGoogle Scholar
- Somel M, Khaitovich P, Bahn S, Pääbo S, Lachmann M: Gene expression becomes heterogeneous with age. Curr Biol. 2006, 16 (10): R359-R360. 10.1016/j.cub.2006.04.024.PubMedView ArticleGoogle Scholar
- Fraga MF, Ballestar E, Paz MF, Ropero S, Setien F, Ballestar ML, Heine-Suñer D, Cigudosa JC, Urioste M, Benitez J, et al: Epigenetic differences arise during the lifetime of monozygotic twins. Proc Natl Acad Sci USA. 2005, 102 (30): 10604-10609. 10.1073/pnas.0500398102.PubMed CentralPubMedView ArticleGoogle Scholar
- Finch CE: Longevity, senescence, and the genome. 1994, Chicago: University of Chicago PressGoogle Scholar
- Pletcher SD, Macdonald SJ, Marguerie R, Certa U, Stearns SC, Goldstein DB, Partridge L: Genome-wide transcript profiles in aging and calorically restricted Drosophila melanogaster. Curr Biol. 2002, 12 (9): 712-723. 10.1016/S0960-9822(02)00808-4.PubMedView ArticleGoogle Scholar
- Robinson GE: Regulation of honey bee age polyethism by juvenile hormone. Behav Ecol Sociobiol. 1987, 20 (5): 329-338. 10.1007/BF00300679.View ArticleGoogle Scholar
- Withers GS, Fahrbach SE, Robinson GE: Selective neuroanatomical plasticity and division of labour in the honeybee. Nature. 1993, 364 (6434): 238-240. 10.1038/364238a0.PubMedView ArticleGoogle Scholar
- Withers GS, Fahrbach SE, Robinson GE: Effects of experience and juvenile hormone on the organization of the mushroom bodies of honey bees. J Neurobiol. 1995, 26 (1): 130-144. 10.1002/neu.480260111.PubMedView ArticleGoogle Scholar
- Erber J, Homberg U, Gronenberg W: Functional roles of the mushroom bodies in insects. Arthropod brain: its evolution, development, structure, and functions. Edited by: Gupta AP. 1987, New York: Wiley, 588-xiGoogle Scholar
- Mackert A, Bitondi MM, Hartfelder K, Simoes ZL, do Nascimento AM: Identification of a juvenile hormone esterase-like gene in the honey bee, Apis mellifera L.--expression analysis and functional assays. Comp Biochem Physiol B Biochem Mol Biol. 2008, 150 (1): 33-44. 10.1016/j.cbpb.2008.01.004.PubMedView ArticleGoogle Scholar
- Barchuk AR, Bitondi MM, Simoes ZL: Effects of juvenile hormone and ecdysone on the timing of vitellogenin appearance in hemolymph of queen and worker pupae of Apis mellifera. J Insect Sci. 2002, 2: 1-PubMed CentralPubMedView ArticleGoogle Scholar
- Piulachs MD, Guidugli KR, Barchuk AR, Cruz J, Simoes ZL, Belles X: The vitellogenin of the honey bee, Apis mellifera: structural analysis of the cDNA and expression studies. Insect Biochem Mol Biol. 2003, 33 (4): 459-465. 10.1016/S0965-1748(03)00021-3.PubMedView ArticleGoogle Scholar
- Engels W: Reproduction and caste development in social bees. Chemistry and Biology of Social Insects. Edited by: Peperny VJ. 1987, Muchen, 275-281.Google Scholar
- Cunha AD, Nascimento AM, Guidugli KR, Simoes ZL, Bitondi MM: Molecular cloning and expression of a hexamerin cDNA from the honey bee, Apis mellifera. J Insect Physiol. 2005, 51 (10): 1135-1147. 10.1016/j.jinsphys.2005.06.004.PubMedView ArticleGoogle Scholar
- Crailsheim K, Stolberg E: Influence of diet, age, and colony condition upon intestinal proteolytic activity and size of the hypopharyngeal glands in the honeybee (Apis mellifera L.). J Insect Physiol. 1989, 35 (8): 595-602. 10.1016/0022-1910(89)90121-2.View ArticleGoogle Scholar
- Lohse M, Bolger AM, Nagel A, Fernie AR, Lunn JE, Stitt M, Usadel B: RobiNA: a user-friendly, integrated software solution for RNA-Seq-based transcriptomics. Nucleic Acids Res. 2012, 40 (Web Server issue): W622-627.PubMed CentralPubMedView ArticleGoogle Scholar
- Li H, Durbin R: Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009, 25 (14): 1754-1760. 10.1093/bioinformatics/btp324.PubMed CentralPubMedView ArticleGoogle Scholar
- Honey Bee Genome Sequencing Consortium: Finding the missing honey bee genes: lessons learned from a genome upgrade. BMC Genomics. in reviewGoogle Scholar
- Trapnell C, Pachter L, Salzberg SL: TopHat: discovering splice junctions with RNA-Seq. Bioinformatics. 2009, 25 (9): 1105-1111. 10.1093/bioinformatics/btp120.PubMed CentralPubMedView ArticleGoogle Scholar
- Kitts P: Genome assembly and annotation process. The NCBI Handbook. Edited by: McEntyre J, Ostell J. 2002Google Scholar
- McCarthy DJ, Chen Y, Smyth GK: Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Res. 2012, 40 (10): 4288-4297. 10.1093/nar/gks042.PubMed CentralPubMedView ArticleGoogle Scholar
- Hochberg Y, Benjamini Y: More powerful procedures for multiple significance testing. Stat Med. 1990, 9 (7): 811-818.PubMedView ArticleGoogle Scholar
- McQuilton P, St Pierre SE, Thurmond J: FlyBase 101–the basics of navigating FlyBase. Nucleic Acids Res. 2012, 40 (Database issue): D706-714.PubMed CentralPubMedView ArticleGoogle Scholar
- Carmona-Saez P, Chagoyen M, Tirado F, Carazo JM, Pascual-Montano A: GENECODIS: a web-based tool for finding significant concurrent annotations in gene lists. Genome Biol. 2007, 8 (1): R3-10.1186/gb-2007-8-1-r3.PubMed CentralPubMedView ArticleGoogle Scholar
- Nogales-Cadenas R, Carmona-Saez P, Vazquez M, Vicente C, Yang X, Tirado F, Carazo JM, Pascual-Montano A: GeneCodis: interpreting gene lists through enrichment analysis and integration of diverse biological information. Nucleic Acids Res. 2009, 37 (Web Server issue): W317-322.PubMed CentralPubMedView ArticleGoogle Scholar
- Tabas-Madrid D, Nogales-Cadenas R, Pascual-Montano A: GeneCodis3: a non-redundant and modular enrichment analysis tool for functional genomics. Nucleic Acids Res. 2012, 40 (Web Server issue): W478-483.PubMed CentralPubMedView ArticleGoogle Scholar
- Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001, 29 (9): e45-10.1093/nar/29.9.e45.PubMed CentralPubMedView ArticleGoogle Scholar
- Yuan JS, Reed A, Chen F, Stewart CN: Statistical analysis of real-time PCR data. BMC Bioinformatics. 2006, 7: 85-10.1186/1471-2105-7-85.PubMed CentralPubMedView ArticleGoogle Scholar
- Robinson MD, McCarthy DJ, Smyth GK: edgeR: a bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010, 26 (1): 139-140. 10.1093/bioinformatics/btp616.PubMed CentralPubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.