The authors have declared that no competing interests exist.
Conceived and designed the experiments: APW AMB. Performed the experiments: AMB CS SK. Analyzed the data: APW AMB CS SK DA GRC. Contributed reagents/materials/analysis tools: APW CS TD. Wrote the paper: APW AMB CS TD JH.
Because most human stroke victims are elderly, studies of experimental stroke in the aged rather than the young rat model may be optimal for identifying clinically relevant cellular responses, as well for pinpointing beneficial interventions.
We employed the Affymetrix platform to analyze the whole-gene transcriptome following temporary ligation of the middle cerebral artery in aged and young rats. The correspondence, heat map, and dendrogram analyses independently suggest a differential, age-group-specific behaviour of major gene clusters after stroke. Overall, the pattern of gene expression strongly suggests that the response of the aged rat brain is qualitatively rather than quantitatively different from the young, i.e. the total number of regulated genes is comparable in the two age groups, but the aged rats had great difficulty in mounting a timely response to stroke. Our study indicates that four genes related to neuropathic syndrome, stress, anxiety disorders and depression (
We suggest that a multi-stage, multimodal treatment in aged animals may be more likely to produce positive results. Such a therapeutic approach should be focused on tissue restoration but should also address other aspects of patient post-stroke therapy such as neuropathic syndrome, stress, anxiety disorders, depression, neurotransmission and blood pressure.
Stroke is a devastating condition afflicting mostly the elderly for which no viable medication exists to improve neurorehabilitation. In particular, great clinical benefit may accrue from deciphering and targeting basic neurobiological mechanisms underlying post-stroke CNS recovery both in structural and functional terms. Studies of stroke in experimental animals have identified a variety of interventions with marked neuroprotective effects, but most of these approaches have failed to benefit aged human stroke victims, perhaps because such therapies have been developed in stroke models using young animals. Indeed, recent studies of experimental stroke in the aged animal reveal age differences that may have more clinical relevance, both for understanding cellular responses to stroke and for identification of beneficial interventions
Recent advances in genomics and DNA array technology may lead to a comprehensive insight into the mechanisms underlying differences between aged and young animals in rate and extent of brain repair and regeneration after stroke. Several such studies have employed these techniques with the aim of identifying new therapeutic targets for stroke treatment. Some of these studies revealed changes in transcriptional activity of a variety of genes related to stress response, inflammation, acute- and delayed cell death in young rats
The present study expands and extends this work by taking advantage of recent developments in rat genomics, performing a whole-genome transcriptomic analysis of the perilesional infarct during the acute- and recovery phases following stroke. Further, through data mining and one-by-one gene function search in the context of the pathophysiology of stroke, we assigned 161 newly identified genes to stroke-relevant processes such as those mentioned above.
Thirty (30) young (3 to 4 mo of age) and 45 aged (19 to 20 mo) male Sprague-Dawley rats, bred in-house, were used. The two age groups were divided randomly into 3-day and 14 day post-stroke survival groups. In addition, nineteen (19) young (3 to 5 mo of age) and 24 aged (18 to 20 mo) male Sprague-Dawley rats were used as controls. Body weights ranged from 290 to 360 g for the young rats and from 520 to 600 g for the aged rats. The rats were kept in standard cages in a temperature- (22
Cerebral infarction was induced by transcranial interruption of blood flow by transiently lifting the middle cerebral artery with a tungsten hook as previously described
After the tissue was homogenized, total RNA was extracted from microdissected tissue using TRIzol reagent (Invitrogen life technologies, Karlsruhe, Germany). Genomic DNA was removed using the Rneasy Plus kit (Qiagen).
Prior to sample preprocessing, the integrity of RNA pools was assessed with the RNA 6000 nano kit using the Bioanalyzer 2100 instrument (Agilent, Böblingen, Germany). RNA integrity numbers ranged between 6.5 and 8.2. 200 ng of each sample were processed with the whole transcript (WT) expression kit (Ambion, Darmstadt, Germany), i.e. subjected to RNA amplification via reverse transcription to double-stranded cDNA and subsequent
Constantly high quality of microarray data was ensured by visual inspection of scanned images for hybridization artifacts and correspondence analysis of raw and normalized microarray data. Normalizations were performed with the Quantiles method
Each array reflected the expression of 19–24 pooled animal samples. This drastically reduces gene expression variance that is otherwise observed between individually hybridized animal samples. Hence, the power loss due to the smaller array sample size is at least partly compensated for.
For qualitative real time PCR (qPCR), we synthesized cDNA from large pools (n = 19−24) of total RNA with the High-Capacity cDNA reverse transcription kit (Applied Biosystems, USA). The qPCR was performed in 96-well 0.1-ml thin-wall PCR plates (Applied Biosystems) in the Step One Plus System (Applied Biosystems). Each 20 µl reaction contained 10 µl iQ SYBR Green Master Mix (BioRad Laboratories, Hercules, CA), 2 µl gene-specific forward and reverse primer mix (Qiagen, Alameda, CA) and 8 µl pre-diluted cDNA. No template controls contained nuclease-free water instead. The cycling conditions were 3 min 95°C to activate iTaq DNA polymerase followed by 45 cycles with 30 s denaturation at 95°C, 30 s annealing at 58°C and 30 s elongation at 72°C. At the end of amplification cycles, melting curves were used to validate PCR product specificity. All samples were amplified in triplicates. Data were analyzed using the ΔΔCt method
After raw data normalization and probe set summary, we employed empirical Baysian methodology to analyse expression values of 28,826 transcript clusters for differential expression between post-stroke samples of young and aged rats and their respective controls. This revealed in total 1,658 differentially expressed genes with a two-fold or greater change (up or down) of the transcription rate. Intensities of differentially expressed probe sets from all samples were subjected to agglomerative hierarchical clustering (AHC) and results were displayed as a heat map. The dendrogram shows that relative expression values clearly distinguish naïve rats from their post-stroke littermates (
Scaled expression values of all 1,658 differentially expressed genes are shown for each group with light red being the lowest and light green the highest expression level. The depicted dendrograms cluster samples (top) and genes (left) employing average agglomeration and euclidian distance measure.
For further examination of the data, we performed a correspondence analysis (COA) for the genes that were differentially expressed according to the
The left panel depicts the Eigenvalues of the correspondence analysis and shows that the major factors contributing to the variance of stroketomics analysis were stroke (52%), post-stroke time (25%) and age (12%). (Right panel): The first two sources of variability, stroke and post-stroke time formed the coordinates of the right panel. The graph shows the distribution of transcripts (black dotes) as a function of treatment (stroke) and post-stroke time. Samples from young (green) and aged (red) animals particularly differ in their post-stroke response (illustrated by ellipses that form non-parallel planes). Transcripts with characteristic expression in naive samples are encircled in black.
At day 3 post-stroke, we observed changes in 916 transcript-specific probe sets in young rats, indicating changes in the expression of the corresponding genes. Of these, 218 displayed age-specific upregulation and 52 showed decreased age-specific down regulation. By day 14 post-stroke, fewer probe sets show changes (n = 862), with 138 indicating increased expression levels and 115 showing age-specific decreased mRNA levels (
Note that at 14 days post-stroke, the differences between the age groups were more pronounced.
In aged rats, a similar number of genes was found to be differentially expressed on day 3 post-stroke. Of 874 probe sets, 149 were increased only in aged rats, while 79 showed decreased expression. By day 14 post-stroke, the aged rats had clearly a greater number of differentially expressed genes as compared to young rats at the same time point (n = 927) with 266 more genes being upregulated but fewer genes being down regulated than in young animals (n = 43; 46%) (
Most importantly, the Venn diagrams showed the divergent age-related gene expression with increasing time, ie. the number of upregulated genes in young animals decreased by 39% while the number of downregulated genes increased by 223%. The aged rats showed a mirrored gene expression, ie the number of upregulated genes increased by 187% while the number of downregulated genes
The kinetics of gene expression over a longer time period gives us clues as to what processes in the long run could be defective at the level of transcription in aged rodents. We distinguished several different patterns of gene regulation, as depicted schematically in
There were several distinct patterns of gene regulation: persistently upregulated (black line), transiently upregulated, (orange line), “late-upregulated” (red line), “late-downregulated” (yellow line), transiently downregulated (blue line), and persistently downregulated (green line). Aged animals showed larger numbers than young of genes that were late-upregulated, persistently upregulated and persistently downregulated. The young rats, in contrast, had a much larger number of transiently upregulated and delayed downregulated genes. Note that this representation does not take into account the fold changes for individual genes but the relative change in gene expression at days 3 and 14 post-stroke.
A comparison of the global pattern of gene indicated that aged animals showed larger numbers than young of genes that were late-upregulated (133 genes), persistently upregulated (115 genes) and persistently downregulated (20 genes). The young rats, in contrast, had a much larger number of transiently upregulated (177 genes) and delayed downregulated genes (62 vs 9) (
Gene Symbol | Function | Drug availability | Y/3d | Y/14d | A/3d | A/d 14d |
|
post-stroke inflammation | inhibitor available |
|
|
|
|
|
post-stroke excitotoxicity | agonist not available |
|
|
|
|
|
calcium-channel disregulation | agonist not available |
|
|
|
|
|
calcium-channel disregulation | agonist not available |
|
|
|
|
|
neuropathic syndrome; anxiety; depression | drugs available |
|
|
|
|
|
post-stroke blood pressure control | drugs available |
|
|
|
|
|
limits inflammatory reaction | drugs available |
|
|
|
|
|
limits inflammatory reaction | drug available |
|
|
|
|
|
post-stroke fibrosis | drug available |
|
|
|
|
|
limits inflammatory reaction | drugs available |
|
|
|
|
|
post-stroke inhibitory neurotransmission | agonist available |
|
|
|
|
|
pain; sleep homeostasis | agonist available |
|
|
|
|
|
post-stroke anxiety disorders and stress | under development |
|
|
|
|
|
post-stroke anxiety disorders and stress | under development |
|
|
|
|
|
post-stroke blood pressure control | drugs available |
|
|
|
|
|
blood vessel morphogenesis | drugs available |
|
|
|
|
|
post-stroke inflammation | drugs available |
|
|
|
|
|
neuropathic syndrome; anxiety; depression | drugs available |
|
|
|
|
|
calcium retrieval disregulation | agonist available |
|
|
|
|
|
extracellular matrix component | drugs available |
|
|
|
|
|
tissue remodeling | drug available |
|
|
|
|
|
tissue remodeling | drug available |
|
|
|
|
|
neuroprotection | modulator available |
|
|
|
|
|
neuropathic syndrome; anxiety; depression | agonists available |
|
|
|
|
|
blood pressure control | drugs available |
|
|
|
|
|
sodium homeostasis | modulators available |
|
|
|
|
|
calcium retrieval disregulation | modulators available |
|
|
|
|
|
neuroprotection | modulator available |
|
|
|
|
Legend
↔ no changes vs contralateral side.
↑ upregulated vs contralateral side.
↓ downregulated vs contralateral side.
y/3d Young, 3d post-stroke.
y/14d Young, 14d post-stroke.
A/3d Aged, 3d post-stroke.
A/14d Aged, 14d post-stroke.
Confirmation of arrays data for new stroke-related genes was done by RT-PCR. Modulation of gene/protein activity by the indicated drugs may, in combination or alone, improve post-stroke recovery in aged animals. For most of the upregulated genes there is a therapeutic option but not for downregulated genes.
Based on known biological functions, as indicated by GOs, we grouped into nine major stroke-relevant processes: “
Most of the stroke-related genes (39%) that were involved in “
Most classes of these new genes were upregulated, with the exception of “
The hallmark of the genes involved in “
While the expression pattern gives us clues as to what processes might be de-regulated after stroke in aged animals, it does not yield specific genes, the ultimate goal of stroke genomics. In the following section we make a specific age-related analysis of the genes involved in stroke-relevant processes. RT-PCR was used to verify the target genes.
Not surprisingly, the genes displaying persistent upregulation that were shared by the both age groups included some implicated in the pain response to stroke (
A number of genes that were upregulated at day 3 returned to contralateral expression levels by day 14. These included genes implicated in pain (
Aged rats had clearly more downregulated genes at day 14 than young rats. Most of the genes that displayed persistent (Type F) or delayed downregulation (Type D) were involved in responses to pain (
Persistent upregulation specific to young rats was displayed by
Strikingly, there were very few genes showing delayed downregulation or persistent downregulation for “
Genes that were persistently upregulated in both young and aged rats included those associated with oxidative stress (
Apoptosis and cell death are intricately linked to the inflammatory response and to phagocytosis. The vast majority of genes related to these processes were persistently upregulated (86%). All newly identified genes are shown in
Upregulated genes included genes whose expression is required for phagocytosis (
Among the genes persistently upregulated were those involved in energy metabolism (
Like those involved with apoptosis and cell death, the vast majority of genes associated with these processes were persistently upregulated. All newly identified genes and five confirmatory genes are shown in
First, we noted the early upregulation in aged rats of the gene encoding the receptor for the proinflammatory cytokine interleukin-6 (IL-6r). Among those genes showing persistent upregulation we found
We also observed the early upregulation of non-canonical I-kappaBkinase/NF-kappaB cascade via
Microglia/macrophage-specific responses to stroke included the persistent upregulation of the genes encoding the microglia activating factors
Scar build-up after stroke is a complex process involving recruitment of glial, fibroblast, and immune cells which function to seal the wounded tissue from the surrounding environment by promoting production of a fibrotic scar. The gene pattern for this process is similar to that described above for “
Most of the genes associated with these processes were involved in cell division & proliferation and were mostly transiently or persistently upregulated in both age groups. Persistently uregulated genes included
Astrocyte proliferation is a major event associated with wound healing. Genes persistently upregulated in both age groups included those coding for vimentin (
Transforming growth factor-beta (TGF-beta) signalling mechanisms at the wound site during the acute phase performs functions central to tissue scarring by mediating astrocytes migration, mesenchymal, fibroblasts and macrophage recruitment, extracellular matrix deposition, re-epithelization, and wound contraction
One gene involved in collagen processing showed delayed upregulation (
Several genes involved in fibrosis, the second major scarring process, were persistently upregulated (
Stroke induces a profound remodelling of the brain vasculature involving mostly permanently upregulated genes (60%) and those undergoing delayed upregulation (17%). All newly identified genes in this category and one confirmatory gene are shown in
Genes that were persistently upregulated in both age groups included transcripts of the TGF-beta signalling pathway involved in CNS vascular remodelling and angiogenesis including,
Two genes,
Of the genes showing an age effect, we noted the increased expression in aged rats of the matrix-degrading encoding proteases genes
We also observed very high levels of expression of the chemokine
In recovery from stroke, adult CNS tissue may require re-activation of pathways implicated in development. Aged rats had 50% more genes showing a temporal disregulation of the expression for several genes that are closely associated with brain development including
Several additional genes implicated in ECM and cytoskeleton remodelling during brain development (
Neurogenesis is controlled by genes in several important GOs, including neurite development, corticogenesis, axonogenesis, dentritogenesis and glial differentiation. Most of the genes belonging to this group were downregulated on day 14 post-stroke in both age groups. Very few genes, mostly those required for synaptic activity/turnover (
Genes related to axonal plasticity were either transiently downregulated in both age groups (
Clinical trials aimed at improving functional recovery after stroke have uniformly failed. One reason for this may be that very little genetic information is available describing the post-stroke events. By comparative transcriptomics we identified 161 new stroke-related genes some of which may be used as new therapeutic targets that may help stabilize the aged organisms soon after stroke and facilitate tissue and behavioral recovery after 2 weeks post-stroke.
We showed that in the gene expression dendograms naive rats are clearly separated from post-stroke littermates. Furthermore, observed changes are clearly occlusion time-dependent. Most of the changes observed are active responses, only a smaller group shows inhibitory responses and reduced expression. The gene expression data overview further suggests a specific group of genes with increased expression (compared to untreated animals) 3 days after stroke and with reduced expression later (14 days). Age differences in this pattern, were revealed by the heat map. This latter effect is strongly supported by the correspondence analysis. The first two principal coordinates in this analysis, the effect of stroke and the post-stroke time on gene expression in the perilesional cortex together explained 77% of the variance of stroketomics in aged rats and clearly separate young and older animals. These results (together with the specific annotation and function of the genes) support our notion of an active response to the stroke event at day 3 in the young rats balanced by an inhibitory effect on a roughly similar number of genes (about 1,400) which decays over time (so that at day 14 mostly decreased mRNA levels are seen). However, the aged rats show a set of up-regulated set of genes not observed in the young rats at both 3 and 14 days post-stroke. These genes are involved in wound healing, scar formation, vascular remodelling, and angiogenesis (examples discussed below).
The correspondence, heat map and dendrogram analyses thus independently suggest a differential, age-group-specific behaviour of major gene clusters after stroke. Furthermore, key genes identified were independently verified by RT-PCR.
Gene expression in the young and aged rats diverged with increased time following stroke, as reflected by a decreased number of upregulated genes in young rats and an increased number of upregulated genes in aged rats. A similar pattern of age-group divergence was seen in the downregulated genes, with young animals returning closer to control levels while aged rats showed a greater number of genes with persistently decreased expression. Genes that changed in a similar pattern in both age groups remained the same at both time points.
Overall, the pattern of gene expression strongly suggests that the response of the aged rat brain is qualitatively rather than quantitatively different from the young, i.e. the total number of regulated genes is comparable in the two age groups, but the aged rats had great difficulty in mounting a timely response to stroke.
Several changes in gene expression in the lesioned hemisphere were unique to aged animals: (i) a large increase in the number of late upregulated genes; a substantial increase in the number of persistently upregulated genes; (ii) an increase in the number of persistent downregulated genes. The unusually large number of transiently regulated genes in the lesioned hemisphere of young suggests the inability of aged rats to mount a controlled transcriptional response for these genes. Specifically, a large number of genes involved in “
“
In the present work we carried out a genome-wide, transcriptome analysis of post-stroke gene expression in young and aged animals and identified 161 new genes that are of interest to stroke research, as well as some that may be targets for development of stroke therapies (summarized in
Since the young rats recover unexpectedly well in terms of function but do not fully recover at tissue level after stroke, we expect the comparative analysis of gene expression will identify new therapeutic targets for stroke in general, as well as genes that could be therapeutic targets in aged subjects in particular. We distinguish two major approaches for therapeutic intervention, (i) to stabilize the infarct size during the early phase of stroke by controlling inflammation and apoptosis and, (ii) to enhance to repair the capacity of the CNS during the rehabilitation phase that starts at the end of the second week post-stroke on a background of stabilized physiology and system homeostasis. Because young rats recover behaviourally within one week after stroke, we assume that the gene expression is optimal in this age group. In selecting genes as therapeutic targets, preference should be given to those genes for which activity-modulating drugs are already available.
Further, the gene encoding CHRM3 (cholinergic receptor, muscarinic 3) was specifically and persistently downregulated in the aged rat brains. Because of its recent implication in atrial fibrillation/arrhythmia
Neglected aspects of patient post-stroke life quality are neuropathic syndrome, stress, anxiety disorders and depression. Our study indicates that four genes related to these processes may have impaired response to stroke in aged rats. These are activin A receptor, type IC (
The gene coding for the neuropeptide corticotropin-releasing hormone (
Our results strongly indicate a disregulation of post-stroke blood pressure, a condition often neglected in clinical settings. New therapeutic options suggested by our study for control of blood pressure include the genes
Perturbations in calcium balance, have received attention as major cause of axonal damage and neuronal dysfunction in many pathological conditions
Oxidative stress and DNA damage-related genes have been well studied in stroke models. However, none of the proposed drugs targeting these genes was efficient in reducing oxidative stress and DNA damage incurred cell death. Therefore further therapeutic targets are needed.
Post-stroke young rats were better protected against oxidative stress by persistently upregulating genes coping with the oxidative stress like the antioxidant zinc-binding proteinmetallothionein 1a (
Young rats also had better synchronized gene expression as was illustrated by the coordinated co-expression of genotoxic stress-induced genes
New genes whose modulation may help limit the inflammatory reaction in aged rats may also include
Scar build-up and post-stroke fibrosis are believed to create an obstacle toward axonal re-growth after stroke. Therefore other potential therapeutic target includes gene products that contribute to the post-stroke fibrosis and that were also strongly upregulated including
Both scar build-up and fibrosis are controlled mainly by astrocytes, microglia and fibroblasts and are intimately associated with genes required for cell proliferation. Genes of this type were frequently upregulated, either transiently or persistently, in both age groups. However, in order to be able to define further therapeutic targets, further studies are required to unequivocally identify the cellular specificity of these genes.
Genes involved in
Young rats on the other hand had very high levels of expression of the chemokine
Following stroke most of the developmental pathways are activated in both age groups. Identification of gene networks that become active early after stroke may help us to illuminate basic genetic mechanisms underlying post-stroke at the early stages of tissue recovery after stroke. It may also promote design of better pharmacological approaches to sustain developmental pathways that are do initiated shortly after stroke. Aged rats had 50% more genes showing both a temporal disregulation and surprisingly an increased gene expression related to brain development including
Retinoic is a strong morphogen during CNS development
Finally, the more complete tissue recovery in young rats may be due to increased levels of neuronal precursors, pro-survival genes, the helix-loop-helix (HLH) family of transcription factors,
Post-stroke axonal growth was compromised in both age groups
Genes whose expression was persistently upregulated in both age groups were quite rare. Among the genes in this class were
Genes that were persistently downregulated in the aged rats were more numerous than in the young group and included a group of genes involved in synaptic activity-induced neuroprotection (
Many genes implicated in the
To date, all monotherapeutic attempts to prevent or lessen brain damage following stroke have failed. In view of our findings that stroke impacts a wide range of systems in an age-dependent manner, from CNS physiology to CNS regeneration and plasticity, the failure of therapies aimed at only a single target system is perhaps inevitable. Our results suggest that a multi-stage, multimodal treatment in aged animals may be more likely to produce positive results. Such a therapeutic approach should be focused on tissue restoration but should also address other aspects of patient post-stroke therapy such as neuropathic syndrome, stress, anxiety disorders, depression, neurotransmission and blood pressure.
(XLSX)
(XLSX)
(XLSX)
(XLSX)
(XLSX)
(XLSX)
(XLSX)
(XLSX)
(XLSX)