Aging is a complex process characterized by a steady decline in an organism’s ability to perform life-sustaining tasks. In the present study, two cages of approximately 12,000 mated
Aging is a general characteristic of life occurring across a great range of life forms [
Extrinsic factors, external threats to survival, can play a major role in senescence [
Previous research has been conducted on genome-wide gene expression in the context of aging using model systems for genetic research, especially the worm
Our motivation for conducting the present research using
The present study was based on simultaneous establishment of two large populations of
The
The flies were watched closely once pupation was evident and at the time of eclosion, they were lightly etherized, separated by sex, and counted. At each collection, 75 females or 75 males were placed into individual 8 oz bottles with food, until approximately 25,000 flies were collected (12,500 males and 12,500 females). The females were allowed to mature for 3 days and the males were allowed to mature for a minimum of 2 days. After this time period, sets of 75 females and males were allowed to mate for 24 hours. After mating, the flies were gently etherized, separated by sex, and counted and males were discarded. Approximately 12,000 mated females were released into a 3′ × 2′ × 1′ Plexiglas cage. Two cages were initialized in this manner. Each cage had two holes on either side covered with Tubigrip (ConvaTec, Princeton, NJ) to allow access into the cage without the loss of flies. The cages each contained six large Petri dishes of media and an additional two large Petri dishes of cotton balls moistened with Nanopure water. The cages were held in a laboratory at ~22–24°C with a diurnal light cycle. The media Petri dishes were changed every day, the water was checked every day, and water was replaced every other day. The cages had their positions changed every day with respect to top or bottom cages as they were stacked on top of each other.
Each day, the dead flies were collected by aspiration and tallied. Mortality curves comparing the number of total dead flies over time were constructed. Control time point sexually mature female flies were collected at 6 days old (4 days old before being released in the boxes and after two days residency in the large cages that were sampled for the present study). Flies from this time point were used for the standard sample in the two-sample microarrays used in the present study. Over the course of this study, twenty-two samples of 24 females each were collected by aspiration, gently etherized, counted, and allowed to recover for two hours in vials containing fly food. After two hours, the females were flash-frozen in liquid nitrogen, transferred to dry ice, and stored at −80°C. Every seven days after the collection of the control females, four samples of 12 females each were collected following the same protocol as the control females. The flies were collected at 1:00 pm CST and frozen at 3:00 pm CST. Collection lasted until day 79 after introduction of the flies into the each cage. At this time point, there were only enough surviving females for this last collection.
Total RNA was extracted from all sets of female flies collected at days 2, 9, 16, 23, 30, 37, 44, 51, 58, 65, 72, and 79 in the cages utilizing the standard TRIzol protocol following manufacturer’s instructions (Invitrogen, Carlsbad, CA). The RNA was cleaned using the Qiagen RNeasy Mini Kit per manufacturer’s instructions (Qiagen, Valencia, CA). The quality and integrity of the RNA were assessed at the UNL Genomics Core Facility by using an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc., Palo Alto, CA). The RNA was quantified and cDNA microarray analyses were performed by the University of Nebraska Medical Center (UNMC) Microarray Core Facility. Two-color Version 2 DGRC oligonucleotide microarrays (
Indirect labeling with Cy3/Cy5 fluorescent dyes was performed using 12
Cy3 (532 nm) and Cy5 (635 nm) scans were performed using a ScanPix 4000B slide reader as per manufacturer’s suggested conditions (Molecular Devices, Sunnyvale, CA). Care was given during the scanning procedure to carefully adjust the photomultiplier tubes (PMTs) such that the overall intensity from both the Cy3 and Cy5 channels was equalized. Following image capture, the overall images, as well as the individual spots, were assessed for uniformity of hybridization and individual integrity. Problematic spots (i.e., problematical morphology or those with aberrant hybridization properties) were flagged for subsequent removal from the final data set. The intensity assessment for gene spots from 16 bit TIFF files was performed with the GenePix image analysis software (Molecular Devices).
The initial cDNA microarray analysis was to determine pairwise comparisons of each time point to the control (2 days in the cage, at which point the females were 6 days post-eclosion). The sample of females collected at 2 days in the cage was used as a common reference. The later-age samples from the cages were taken at 11 time points: 9, 16, 23, 30, 37, 44, 51, 58, 65, 72, and 79 days in the cages. Analyses were conducted with Linear Models for Microarray Analysis (LIMMA) package in Bioconductor [
The genes identified as differentially expressed across all the time points were subjected to clustering analyses. Clustering is a powerful exploratory technique for the analysis of gene expression data. The underlying biological assumption for clustering of genes is that genes participating in the same biological process are expected to exhibit similar expression patterns. A self-organizing map (SOM) clustering algorithm [
Light intensity observations from the scanned image of five replicates were subjected to quality assurance as implemented in various Bioconductor packages [
High-level overviews of the biological processes affected by the transcriptional dynamics of aging were obtained by using comprehensive classifications systems. These systems include the KEGG (Kyoto Encyclopedia of Genes and Genomes) Database of Biochemical Pathways [
Transcript level patterns across the seventy-nine day time span of the experiments were assessed by
Variation in gene expression was calculated for two sets of genes. Immune function genes were a focus for the measurement of variance as they were relatively frequently represented among the genes that exhibited differential expression as a function of age in the present study. Immune function gene samples were selected from a website in a publication describing such genes for
Reverse transcription was performed using Taqman Gene Expression Assay kits and the 7500 Real Time PCR system (Applied Biosystems, Foster City, CA) according to manufacturer’s instructions. The primer and probe sets used were
Each day, dead flies were collected by aspiration, separated by sex, and tallied. Mortality curves comparing the number of total dead flies over time are shown in Figure
Differential expression results for cDNA microarray compared to qRT-PCR. Gene expression data from qRT-PCR was normalized using
Day 16 | Day 79 | |||
---|---|---|---|---|
Microarray | qRT-PCR | Microarray | qRT-PCR | |
|
−3.01 | −4.59 | −6.45 | −18.23 |
|
−2.25 | −5.07 | −4.67 | −15.48 |
|
4.80 | 4.51 | 12.19 | 20.32 |
|
3.58 | 5.46 | 7.25 | 18.86 |
Percent survivorship females from cages A and B over the course of the experiment (day 79 in the cages). Peak of death occurs near day 31. The
The 1,581 genes differentially regulated at one or more time points (see Supplementary Table 1 in Supplementary Material available online at
PANTHER analysis of significantly differentially expressed genes across times per SOM cluster across all time points. The SOM clusters are shown from clusters 0 to 8 (left to right) across the top of the table. The numbers in the columns indicate the total number of genes found in this group from the gene identification numbers (IDs) able to be mapped in PANTHER. Some genes may be represented in more than one GO category.
Category name (gene ontology accession number) | Cluster | ||||||||
---|---|---|---|---|---|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
31 |
13 |
38 |
31 |
39 |
12 |
16 |
36 |
6 | |
Cell communication (GO:0007154) | 2 | 15 | 44 | 34 | 37 | 13 | 18 | 37 | 5 |
Cellular process (GO:0009987) | 3 | 20 | 53 | 43 | 58 | 2 | 25 | 48 | 6 |
Transport (GO:0006810) | 4 | 7 | 39 | 2 | 32 | 7 | 29 | 20 | 4 |
Cellular component organization (GO:0016043) | 1 | 7 | 6 | 8 | 19 | 7 | 6 | 7 | 0 |
System process (GO:0003008) | 2 | 9 | 29 | 35 | 30 | 9 | 15 | 26 | 3 |
Reproduction (GO:0000003) | 2 | 4 | 8 | 7 | 7 | 3 | 3 | 10 | 0 |
Response to stimulus (GO:0050896) | 2 | 6 | 11 | 8 | 18 | 8 | 9 | 22 | 3 |
Developmental process (GO:0032502) | 2 | 11 | 16 | 25 | 32 | 8 | 15 | 15 | 2 |
Metabolic process (GO:0008152) | 18 | 33 | 94 | 121 | 81 | 37 | 109 | 84 | 10 |
Immune system process (GO:0002376) | 4 | 6 | 24 | 14 | 26 | 12 | 16 | 30 | 2 |
Cell cycle (GO:0007049) | 0 | 5 | 9 | 8 | 15 | 6 | 9 | 13 | 1 |
Cell adhesion (GO:0007155) | 0 | 5 | 11 | 8 | 15 | 3 | 8 | 12 | 2 |
Apoptosis (GO:0006915) | 0 | 1 | 9 | 3 | 7 | 6 | 6 | 9 | 0 |
Generation of precursor metabolites & energy (GO:0006091) | 0 | 1 | 7 | 12 | 3 | 3 | 11 | 3 | 1 |
Homeostatic process (GO:0042592) | 0 | 0 | 1 | 3 | 2 | 1 | 0 | 0 | 0 |
Localization (GO:0051179) | 0 | 0 | 0 | 2 | 3 | 0 | 0 | 0 | 0 |
Three rows by three columns self-organizing map (3 × 3 SOM) for gene expression patterns of clusters. Each cluster is represented by the centroid (average pattern) for genes. The
The number of genes changing per time point for each cluster was determined and found to fluctuate to varying degrees, with the exception of SOM clusters 0 and 8 (Table
Number of genes significantly differentially expressed across time from SOM clustering analysis. The number in the upper left-hand corner of the SOM is the cluster number. The other number in the SOM is the number of genes in that cluster. To be included in the cluster, a gene only needs to be significantly different at any one time point versus the day 2 control. The numbers in the table represent the number of significantly differentially regulated genes at that time point.
Cluster | Days in box | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
9 | 16 | 23 | 30 | 37 | 44 | 51 | 58 | 65 | 72 | 79 | |
|
2 | 6 | 23 | 30 | 57 | 59 | 60 | 59 | 60 | 60 | 60 |
|
12 | 35 | 70 | 93 | 101 | 97 | 114 | 111 | 113 | 113 | |
|
1 | 45 | 25 | 47 | 24 | 28 | 67 | 42 | 58 | 68 | |
|
11 | 39 | 81 | 130 | 95 | 114 | 146 | 228 | 240 | 213 | |
|
1 | 26 | 26 | 10 | 51 | 12 | 14 | 48 | 50 | 11 | |
|
3 | 42 | 67 | 73 | 118 | 110 | 114 | 86 | 56 | 115 | |
|
1 | 2 | 1 | 10 | 1 | 4 | 5 | 74 | 187 | 55 | |
|
2 | 20 | 15 | 12 | 143 | 90 | 97 | 57 | 51 | 90 | |
|
9 | 37 | 41 | 42 | 42 | 42 | 42 | 42 | 40 | 42 | |
|
|||||||||||
Total | 2 | 46 | 269 | 356 | 474 | 634 | 557 | 658 | 746 | 855 | 767 |
Distribution of genes differentially regulated at day 79 from PANTHER analysis. The last time point sampled in the analysis was day 79, which represents the oldest living females. The numbers in the table are the number of genes for each SOM that are either upregulated, downregulated, or unaffected at day 79 compared to the day 2 control time point. The numbers in parentheses are the % of genes that are either upregulated, downregulated, or unaffected at day 79 compared to the day 2 control time point. The genes found in SOM clusters 0 and 8 are 100% affected at day 79.
SOM class | Total number of genes | Upregulated | Downregulated | Unaffected |
---|---|---|---|---|
|
|
|
|
|
1 | 120 | 0 (0) | 113 (94.2) | 7 (5.8) |
2 | 244 | 1 (0.41) | 67 (27.5) | 176 (72.1) |
3 | 267 | 0 (0) | 213 (79.8) | 54 (20.2) |
4 | 234 | 11 (4.7) | 0 (0) | 223 (95.3) |
5 | 127 | 115 (90.6) | 0 (0) | 12 (9.4) |
6 | 233 | 1 (0.43) | 54 (23.2) | 178 (76.4) |
7 | 254 | 90 (35.4) | 0 (0) | 164 (64.6) |
|
|
|
|
|
|
||||
Total | 1581 | 260 (16.4) | 507 (32.1) | 814 (51.5) |
Clusters 0 and 8 are relatively consistent in number of genes differentially expressed across all time points (Tables
Similar to cluster 0, clusters 1, 2, 3, and 6, also demonstrated a relative trend of downregulation of gene expression over time (Figure
One cluster, cluster 7, exhibited a similar pattern of gene expression to cluster 8 (Figure
The only cluster that did not change in an obvious up- or downexpression pattern over time was cluster 4 (Figure
Gene Set Enrichment Analysis (GSEA) was used to classify overrepresented categories of genes for each time point (days 9, 16, 23, 30, 37, 44, 51, 58, 65, 72, and 79) relative to the control time point (day 2 in box), characterized by gene ontology. The entire set of genes that were differentially expressed in at least one time point was used for this analysis. Overrepresented gene categories were similar to the analysis previously done. An upregulation of the immune system GO categories was found from days 16 to 79 (Table
Gene ontologies (GO) derived from GSEA of temporal gene expression as compared to day 2 in the box. The first column is the normalized enrichment score (NES), the second column is the name of the GO group, and columns represented by 9–79 are days in the box. The numbers presented in the table are the median fold changes in gene expression. Missing fold change values were replaced by zeroes. The data is ranked and the False Discovery Rate (FDR) cutoff was set at 0.10.
Sum(NES) | Name | 9 | 16 | 23 | 30 | 37 | 44 | 51 | 58 | 65 | 72 | 79 |
---|---|---|---|---|---|---|---|---|---|---|---|---|
27.7624 | ANTIBACTERIAL_HUMORAL_RESPONSE_GO:0019731 | 0 | 1.93 | 2.412 | 2.374 | 2.586 | 2.443 | 2.544 | 2.609 | 2.581 | 2.701 | 2.806 |
26.4471 | DEFENSE_RESPONSE_TO_GRAM-POSITIVE_BACTERIUM_GO:0050830 | 0 | 2.056 | 2.219 | 2.224 | 2.481 | 2.491 | 2.368 | 2.564 | 2.528 | 2.395 | 2.572 |
25.3498 | DEFENSE_RESPONSE_TO_GRAM-NEGATIVE_BACTERIUM_GO:0050829 | 0 | 0 | 2.368 | 2.366 | 2.52 | 2.446 | 2.452 | 2.584 | 2.586 | 2.609 | 2.691 |
25.1411 | DEFENSE_RESPONSE_TO_FUNGUS_GO:0050832 | 0 | 2.091 | 2.289 | 2.1 | 2.349 | 2.441 | 2.438 | 2.375 | 2.173 | 2.292 | 2.27 |
23.1763 | DEFENSE_RESPONSE_TO_BACTERIUM_GO:0042742 | 0 | 0 | 2.204 | 2.278 | 2.318 | 2.193 | 2.258 | 2.372 | 2.342 | 2.437 | 2.411 |
22.4411 | DEFENSE_RESPONSE_GO:0006952 | 0 | 2.032 | 2.023 | 2.23 | 2.033 | 2.506 | 2.603 | 2.392 | 1.961 | 0 | 2.333 |
16.1288 | RESPONSE_TO_STRESS_GO:0006950 | 0 | 2.059 | 2.199 | 2 | 0 | 1.866 | 1.917 | 0 | 2.071 | 0 | 2.03 |
7.8628 | CELL_CYCLE_GO:0007049 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.001 | 1.974 | 1.912 |
7.5424 | ANTIFUNGAL_HUMORAL_RESPONSE_GO:0019732 | 0 | 0 | 0 | 0 | 0 | 1.858 | 0 | 0 | 0 | 1.945 | 1.847 |
5.9188 | INNATE_IMMUNE_RESPONSE_GO:0045087 | 0 | 0 | 0 | 0 | 0 | 2.05 | 0 | 0 | 0 | 0 | 1.919 |
3.9333 | CHROMATIN_SILENCING_GO:0006342 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.979 | 1.954 | 0 |
3.7802 | CYTOKINESIS_GO:0000910 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.891 | 1.889 | 0 |
2.2342 | DNA_REPLICATION_GO:0006260 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.234 | 0 |
2.0984 | POSITIVE_REGULATION_OF_ANTIFUNGAL_PEPTIDE_BIOSYNTHETIC_PROCESS_GO:0006967 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.098 | 0 |
2.0353 | EGGSHELL_CHORION_GENE_AMPLIFICATION_GO:0007307 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.035 | 0 |
2.0003 | ESTABLISHMENT_OR_MAINTENANCE_OF_CELL_POLARITY_GO:0007163 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0 |
1.9647 | FEMALE_MEIOSIS_CHROMOSOME_SEGREGATION_GO:0016321 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.965 | 0 |
1.9552 | DNA-DEPENDENT_DNA_REPLICATION_INITIATION_GO:0006270 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.955 | 0 |
1.9529 | CELLULARIZATION_GO:0007349 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.953 | 0 |
1.9523 | CHROMOSOME_CONDENSATION_GO:0030261 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.952 | 0 |
1.9084 | PROGRAMMED_CELL_DEATH_GO:0012501 | 0 | 0 | 0 | 0 | 0 | 0 | 1.908 | 0 | 0 | 0 | 0 |
1.8628 | JAK-STAT_CASCADE_GO:0007259 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.863 | 0 |
1.861 | R8_CELL_FATE_SPECIFICATION_GO:0045464 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.861 | 0 | 0 |
1.8558 | POSITIVE_REGULATION_OF_TOLL_SIGNALING_PATHWAY_GO:0045752 | 0 | 0 | 0 | 0 | 0 | 1.856 | 0 | 0 | 0 | 0 | 0 |
1.852 | CELL_FATE_DETERMINATION_GO:0001709 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.852 | 0 |
1.8439 | CYTOPLASMIC_TRANSPORT,_NURSE_CELL_TO_OOCYTE_GO:0007303 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.844 | 0 |
1.8355 | OVARIAN_NURSE_CELL_TO_OOCYTE_TRANSPORT_GO:0007300 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.835 | 0 |
1.8353 | IMMUNE_RESPONSE_GO:0006955 | 0 | 0 | 0 | 0 | 0 | 1.835 | 0 | 0 | 0 | 0 | 0 |
1.8186 | PRE-REPLICATIVE_COMPLEX_ASSEMBLY_GO:0006267 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.819 | 0 |
1.8163 | RESPONSE_TO_PHEROMONE_GO:0019236 | 0 | 1.816 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
1.8026 | CILIUM_ASSEMBLY_GO:0042384 | 0 | 1.803 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
1.7815 | PEPTIDOGLYCAN_CATABOLIC_PROCESS_GO:0009253 | 0 | 0 | 0 | 0 | 0 | 1.781 | 0 | 0 | 0 | 0 | 0 |
−1.7325 | SENSORY_PERCEPTION_OF_CHEMICAL_STIMULUS_GO:0007606 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.73 | 0 |
−1.814 | OOGENESIS_GO:0048477 | 0 | 0 | 0 | 0 | 0 | 0 | −1.81 | 0 | 0 | 0 | 0 |
−1.8205 | CELL_WALL_MACROMOLECULE_CATABOLIC_PROCESS_GO:0016998 | 0 | 0 | 0 | 0 | 0 | 0 | −1.82 | 0 | 0 | 0 | 0 |
−1.8445 | MITOCHONDRIAL_ELECTRON_TRANSPORT,_CYTOCHROME_C_TO_OXYGEN_GO:0006123 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.85 | 0 |
−1.8479 | PROTON_TRANSPORT_GO:0015992 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.85 | 0 |
−1.8508 | SARCOMERE_ORGANIZATION_GO:0045214 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.85 | 0 | 0 | 0 |
−1.8693 | REGULATION_OF_RHO_PROTEIN_SIGNAL_TRANSDUCTION_GO:0035023 | 0 | 0 | 0 | 0 | 0 | −1.87 | 0 | 0 | 0 | 0 | 0 |
−1.9418 | REGULATION_OF_SYNAPSE_STRUCTURE_AND_ACTIVITY_GO:0050803 | 0 | −1.94 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−1.9713 | MITOTIC_SPINDLE_ELONGATION_GO:0000022 | −1.97 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−2.0548 | COURTSHIP_BEHAVIOR_GO:0007619 | 0 | 0 | −2.06 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−2.0802 | MESODERM_DEVELOPMENT_GO:0007498 | 0 | −2.08 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−2.2744 | MITOCHONDRIAL_ELECTRON_TRANSPORT,_NADH_TO_UBIQUINONE_GO:0006120 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −2.27 | 0 |
−3.5887 | TRANSPORT_GO:0006810 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.78 | −1.8 | 0 |
−3.7605 | PROTEOLYSIS_GO:0006508 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.82 | −1.94 | 0 |
−3.7839 | FLIGHT_BEHAVIOR_GO:0007629 | 0 | −1.94 | 0 | 0 | 0 | 0 | −1.84 | 0 | 0 | 0 | 0 |
−3.8231 | NEUROPEPTIDE_SIGNALING_PATHWAY_GO:0007218 | 0 | 0 | −1.85 | −1.98 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−3.981 | SYNAPTIC_VESICLE_EXOCYTOSIS_GO:0016079 | 0 | −2.09 | −1.9 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−4.7191 | TRANSLATION_GO:0006412 | −2.52 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −2.2 | 0 |
−5.5074 | VITELLOGENESIS_GO:0007296 | 0 | 0 | 0 | 0 | 0 | −1.86 | −1.86 | 0 | 0 | −1.79 | 0 |
−5.6526 | MYOFIBRIL_ASSEMBLY_GO:0030239 | 0 | −1.94 | −1.86 | −1.86 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−5.9978 | CALCIUM_ION_TRANSPORT_GO:0006816 | 0 | −2.04 | 0 | 0 | 0 | −1.94 | −2.02 | 0 | 0 | 0 | 0 |
−7.4059 | SEX_DIFFERENTIATION_GO:0007548 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.83 | −1.75 | −1.9 |
−7.5705 | LIPID_METABOLIC_PROCESS_GO:0006629 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.89 | −1.83 | −1.91 |
−11.3157 | DEACTIVATION_OF_RHODOPSIN_MEDIATED_SIGNALING_GO:0016059 | 0 | −1.98 | −2 | 0 | 0 | 0 | −1.84 | 0 | −1.79 | 0 | −1.85 |
−13.8505 | CHORION-CONTAINING_EGGSHELL_FORMATION_GO:0007304 | 0 | 0 | 0 | 0 | 0 | −2.08 | −1.97 | −1.92 | −2.05 | −1.79 | −2.04 |
−13.8792 | MULTICELLULAR_ORGANISMAL_DEVELOPMENT_GO:0007275 | 0 | 0 | 0 | 0 | 0 | −1.96 | −1.96 | −1.85 | −2.1 | −1.88 | −2.1 |
−17.7278 | VITELLINE_MEMBRANE_FORMATION_INVOLVED_IN_CHORION-CONTAINING_EGGSHELL_FORMATION_GO:0007305 | 0 | 0 | 0 | −1.89 | −1.89 | −1.95 | −1.97 | −1.93 | −1.99 | −2.04 | −2.07 |
−19.9978 | EGGSHELL_CHORION_ASSEMBLY_GO:0007306 | 0 | 0 | 0 | −2.06 | −2.15 | −2.32 | −2.23 | −2.17 | −2.32 | −2.17 | −2.29 |
−20.3805 | PHOTOTRANSDUCTION_GO:0007602 | 0 | −2.4 | −2.04 | −2.12 | −2.18 | 0 | −1.89 | −1.99 | −2.01 | −1.9 | −1.92 |
The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways was also considered using GSEA (Table
Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways identified using GSEA of temporal gene expression as compared to day 2 in the box. The first column is the normalized enrichment score (NES), the second column is the name of the KEGG pathway, and columns represented by 9–79 are days in the box. The numbers presented in the table are the median fold changes in gene expression. Missing fold change values were replaced by zeroes. The data is ranked and the False Discovery Rate (FDR) cutoff was set at 0.10.
Sum(NES) | Name | 9 | 16 | 23 | 30 | 37 | 44 | 51 | 58 | 65 | 72 | 79 |
---|---|---|---|---|---|---|---|---|---|---|---|---|
3.9903 | LIMONENE_AND_PINENE_DEGRADATION_00903 | 1.905 | 0 | 0 | 0 | 0 | 2.09 | 0 | 0 | 0 | 0 | 0 |
2.1169 | DNA_REPLICATION_03030 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.117 | 0 |
1.9396 | MISMATCH_REPAIR_03430 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.94 | 0 |
1.8761 | PORPHYRIN_AND_CHLOROPHYLL_METABOLISM_00860 | 1.876 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
1.8689 | PENTOSE_AND_GLUCURONATE_INTERCONVERSIONS_00040 | 1.869 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
1.8595 | BASE_EXCISION_REPAIR_03410 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.86 | 0 |
1.7939 | GLYCOSAMINOGLYCAN_BIOSYNTHESIS_-_HEPARAN_00534 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1.794 | 0 |
0.1672 | ASCORBATE_AND_ALDARATE_METABOLISM_00053 | 1.971 | 0 | 0 | 0 | −1.8 | 0 | 0 | 0 | 0 | 0 | 0 |
0.0946 | RETINOL_METABOLISM_00830 | 1.871 | 0 | 0 | 0 | −1.78 | 0 | 0 | 0 | 0 | 0 | 0 |
−1.7027 | FATTY_ACID_BIOSYNTHESIS_00061 | 0 | 0 | 0 | 0 | −1.7 | 0 | 0 | 0 | 0 | 0 | 0 |
−1.7171 | ECM-RECEPTOR_INTERACTION_04512 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.72 | 0 |
−1.7337 | GLYCOLYSIS_/_GLUCONEOGENESIS_00010 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.73 | 0 | 0 |
−1.741 | OTHER_GLYCAN_DEGRADATION_00511 | 0 | 0 | 0 | −1.74 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
−1.7658 | FOLATE_BIOSYNTHESIS_00790 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.77 | 0 |
−1.7669 | ALANINE_ASPARTATE_AND_GLUTAMATE_METABOL_00250 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.77 | 0 | 0 |
−1.8355 | NEUROACTIVE_LIGAND-RECEPTOR_INTERACTION_04080 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.84 |
−3.6284 | PYRUVATE_METABOLISM_00620 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.79 | −1.84 | 0 |
−3.6518 | ARGININE_AND_PROLINE_METABOLISM_00330 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.87 | −1.78 |
−4.318 | OXIDATIVE_PHOSPHORYLATION_00190 | −1.91 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −2.41 | 0 |
−6.4788 | RIBOSOME_03010 | −2.54 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | −1.82 | −2.12 | 0 |
−9.3465 | STARCH_AND_SUCROSE_METABOLISM_00500 | 0 | 0 | 0 | −1.75 | −2.04 | 0 | 0 | 0 | −1.85 | −1.82 | −1.89 |
−9.5827 | GALACTOSE_METABOLISM_00052 | 0 | 0 | 0 | −1.92 | −1.86 | 0 | 0 | 0 | −2.08 | −1.97 | −1.74 |
−18.1624 | PHOTOTRANSDUCTION_-_FLY_04745 | 0 | −2.28 | −2.21 | −2.09 | −2.07 | 0 | −1.95 | −2.02 | −1.87 | −1.9 | −1.77 |
Shared expression patterns of differentially expressed genes were investigated using biclustering analysis of gene expression and experiments/replicates. Days 9–79 in the box were compared to day 2 (5 days old) in the box. The data demonstrates the reproducibility of the experiments. One cluster of genes is upregulated at early time points, while being downregulated at the later time points. Another cluster of genes shows the opposite trend. The data are clustered into three clusters, early, midlife, and late. The early time points include four time points (9, 16, 23, and 30) with days 9 and 16 being closely related and days 23 and 30 being closely related. The midlife time points (days 37, 44, 51, and 58) demonstrate that days 44 and 51 are closely related and days 37 and 58 are closely related. The late time points (65, 72, and 79) are closely related. The significantly up- or downregulated genes across time points compared to day 2 in the box were determined. In order to be included, the gene had to be detected at only one time point. The GSEA detected 263 upregulated genes (Supplementary Table 2) and 355 downregulated (Supplementary Table 3) genes. Of these 618 differentially regulated genes, only 1,
Figure
Change in the variation of gene expression of immune function genes within cages (boxes A and B). This figure presents the average variation over time. The slope of gene expression change is shown on the
Change in the variation of gene expression of immune function genes among cages. For this analysis, samples from cages (boxes A and B) were mixed then the slope calculated. The slope of gene expression change is shown on the
Change in gene expression of a set of 200 genes randomly selected from the
In the present study, we conducted a fine-scaled temporal analysis of genome-wide gene expression in replicate laboratory populations. The only factor that obviously varied over the course of the study was age, but the environment could have varied from one cage to the other. The data in the present study was generated from replicate populations, initially very nearly 12,000 mated female
Our data set included 1,581 genes that were differentially expressed as a function of age as revealed by SOM clustering and PANTHER analysis. There was consistency among clusters in the pattern of gene expression. As one example, clusters 0, 1, 2, 3, and 6 exhibited a marked decline in gene expression starting the first week of life. Cluster 2 showed a leveling off of this decline starting the second week of life. Clusters 0 and 1 consisted of genes whose expression levels off late in life. Clusters 3 and 6 are apparently characterized by an upturn in level of gene expression late in life. As another example of consistency, clusters 5, 7, and 8 exhibited an increase in gene expression starting early in life. In these clusters, the level of gene expression levels off at approximately midlife and there was a trend of declining gene expression after the midlife leveling off, followed by an upturn of gene expression late in life. Another general pattern observed in the data was that there was an apparent tendency for an upturn in gene expression in five clusters (3, 6, 5, 7, and 8). Cluster 4 was unique in the tendency of genes in this group not to vary as a function of age. Although many genes may change expression as a function of age in
Within the present study data set, there were a number of SOM clusters of interest, for example, clusters 0 and 8, due to their consistent number of differentially expressed genes throughout the life span (Tables
In our study, a large number of immune-related genes were overrepresented in PANTHER analyses (Table
In addition to the present study, other studies have demonstrated upregulation of immune genes and downregulation of serine proteases, specifically the
Other candidate genes were identified in the present study using the gene lists generated from both SOM clustering (Supplementary Table 1) and GSEA for overrepresented genes (Supplementary Tables 2 and 3). These were compared to published literature and two databases: GenAge: The Ageing Gene Database (
There were a number of noteworthy patterns in gene expression in the present study. Two important patterns are described in this paragraph and the third is described in the next paragraph. Firstly, there is a general trend in the results in the present study in which gene clusters that change in mean level of expression as a function of age start to change relatively early in life (e.g., Figure
A third noteworthy pattern of gene expression in the present study is the change in variance in immune system genes as a function of age. The variance in gene expression as a function of age was analyzed in the present study for two sets of genes: immune function genes and a random set of genes. The random set of genes was intended to be representative of much of the genome apart from immune system genes. This comparison of focal interest (immunity versus random set of genes) considered the heterogeneity, specifically variance, of gene expression not mean values per se. No change in mean gene expression was observed for a random set of genes in the present study (Figure
The present study has provided evidence about the identity of differentially expressed genes that are associated with aging in
The authors declare they have no competing financial interests.
The authors thank Tiffany Schwasinger for collecting the flies, Jeff Kittrell for microarray data banking assistance, and Dr. Brad Ericson, Sarah Marshall, Rachel Hall, and Art Hasbrouck for technical assistance. The research was supported by grants from the National Center for Research Resources (5P20RR016469) and the National Institute for General Medical Science (8P20GM103427), a component of the National Institutes of Health, UNK URC grant, UNK RSC Collaborative Grant, and the UNK Biology Department.