- Email: [email protected]
Microbial regulation of hippocampal miRNA expression: Implications for transcription of kynurenine pathway enzymes
Behavioural Brain Research 334 (2017) 50–54
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Short communications
Microbial regulation of hippocampal miRNA expression: Implications for transcription of kynurenine pathway enzymes
MARK
Gerard M. Moloneya,d, Olivia F. O’Learya,d, Eloisa Salvo-Romerob, Lieve Desbonneta, Fergus Shanahand, Timothy G. Dinanc,d, Gerard Clarkec,d,⁎, John F. Cryana,d a
Department of Anatomy and Neuroscience, University College Cork, Cork, Ireland Laboratory of Neuro-Immuno-Gastroenterology, Digestive Diseases Research Unit, Vall d’Hebron Institut de Recerca, Department of Gastroenterology, Hospital Universitario Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain c Department of Psychiatry and Neurobehavioural Science, University College Cork, Ireland d APC Microbiome Institute, University College Cork, Cork, Ireland b
A R T I C L E I N F O
A B S T R A C T
Keywords: Microbiota microRNA Hippocampus Kynurenine Germ-free Sex
Increasing evidence points to a functional role of the enteric microbiota in brain development, function and behaviour including the regulation of transcriptional activity in the hippocampus. Changes in CNS miRNA expression may reflect the colonisation status of the gut. Given the pivotal impact of miRNAs on gene expression, our study was based on the hypothesis that gene expression would also be altered in the germ-free state in the hippocampus. We measured miRNAs in the hippocampus of Germ free (GF), conventional (C) and Germ free colonised (exGF) Swiss Webster mice. miRNAs were selected for follow up based on significant differences in expression between groups according to sex and colonisation status. The expression of miR-294-5p was increased in male germ free animals and was normalised following colonisation. Targets of the differentially expressed miRNAs were over-represented in the kynurenine pathway. We show that the microbiota modulates the expression of miRNAs associated with kynurenine pathway metabolism and, demonstrate that the gut microbiota regulates the expression of kynurenine pathway genes in the hippocampus. We also show a sex-specific role for the microbiota in the regulation of miR-294-5p expression in the hippocampus. The gut microbiota plays an important role in modulating small RNAs that influence hippocampal gene expression, a process critical to hippocampal development.
Postnatal assembly of the gut microbiota controls the growth and functionality of the host immune system [1] along with brain development and behaviour [2]. Moreover, changes in microbiota-mediated gut-brain axis signalling have been associated with the stress response and anxiety-like behaviours [3]. The hippocampus appears to be particularly receptive to microbiome-gut-brain axis signalling with alterations noted in neuronal morphology, neurogenesis and serotonergic neurotransmission in germ free animals [4–6]. Behaviours associated with the hippocampus such as anxiety, depression and cognitive function are also abnormal in germ free mice. Alterations in the gut microbiota are known to effect the expression of neurotrophic genes such as Brain-derived neurotrophic factor (Bdnf) [7]. Cooperation of thousands of genes is required for normal neurodevelopment, thus, dysregulation of complicated gene networks are thought to be implicated in many neurodevelopmental and psychiatric disorders [8]. Activity-related transcriptional pathways in the CNS are
⁎
heavily influenced by the gut microbiome [9]. microRNAs (miRNAs) are single-stranded non-coding RNAs that regulate eukaryotic gene expression post-transcriptionally. Hippocampal changes in miRNA expression in the rat are influenced by early life stress and the immune system, and antidepressants have been shown to reverse these changes [10]. Moreover, inhibition of miRNAs in the hippocampus results in changes in cognitive behaviour in rodents [11]. Little is known regarding a microbial influence on miRNA expression in the brain and the role the gut microbiota plays in miRNA expression in the hippocampus has yet to be fully investigated. Given the pivotal impact of miRNAs on gene expression, our study was based on the hypothesis that this would also be altered in the germ free state in the hippocampus. Thus, we hypothesised that the presence of a gut microbiota is required for appropriate miRNA control of gene expression in the hippocampus. All animal experiments were performed as previously described [2]
Corresponding author at: Department of Psychiatry and Neurobehavioural Science, Biosciences Institute, University College Cork, Ireland. E-mail addresses: [email protected] (G.M. Moloney), [email protected] (O.F. O’Leary), [email protected] (E. Salvo-Romero), [email protected] (L. Desbonnet), [email protected] (F. Shanahan), [email protected] (T.G. Dinan), [email protected] (G. Clarke), [email protected] (J.F. Cryan). http://dx.doi.org/10.1016/j.bbr.2017.07.026 Received 24 May 2017; Received in revised form 17 July 2017; Accepted 18 July 2017 Available online 20 July 2017 0166-4328/ © 2017 Elsevier B.V. All rights reserved.
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
other regions of the brain, we wanted to understand what genes were targeted by the miRNAs listed in Table 2, and what potential pathways they may be involved in. Specifically, we investigated if those miRNAs that displayed a significant interaction in relation to germ free status and sex (Table 2) were overrepresented in the tryptophan catabolic process to kynurenine pathway (GO: 0019441), as recent data from our group has shown components of this pathway to be altered in germ free mice in a sex dependent manner [6]. Using this in-silico approach we found that these miRNA target genes were significantly overrepresented in the tryptophan-kynurenine pathway compared to what would be expected randomly (Fig. 2A). We performed qRT-PCR on hippocampal tissue using molecular probes for Tdao2, Kat1, Kynu, Haao and Kmo, (Table 1), the enzymes of the tryptophan-kynurenine pathway (GO: 0019441). Additionally, we measured expression of biologically relevant and predicted target genes of mmu-miR-294-5p, Brd2 and Slit3kr. Brd2 was increased following colonisation in female mice, the increase in Brd2 expression in recolonized animals may indicate that the expression of miR-294-5p is linked to immune system development with the immune-responsive miRNA leading to transcriptional activation following colonisation. That this is more pronounced in females suggests that they be more sensitive to recolonisation and its effect on transcription. All target genes measured showed no interaction between sex and colonisation status and a significant effect of germ free status was seen in kynurenine pathway genes (Fig. 2B–E). Of the kynurenine pathway genes measured, only Kat1 and Tdao2 amplified sufficiently. In male mice, Kat1 was significantly decreased in germ free mice (Fig. 2B, p < 0.05) compared with conventional mice, this decrease was not reversed by colonisation. In contrast, the gene which encodes Tdao2 was significantly increased in germ free male mice and in recolonized mice compared with conventional male mice (Fig. 2C, p < 0.05), (Fig. 2D, p < 0.05). These results demonstrate, to our knowledge, that the gut microbiota can regulate the expression of miRNAs in the hippocampus, specifically miR-294-5p. Changes in expression of miR-294-5p under germ free conditions were only evident in male mice, suggesting as we and other groups have previously shown that the microbiota can impact the CNS in a sex specific manner. The expression of some miRNAs was normalised by colonisation whereas others were not, suggesting that the interaction between the microbiota and miRNA expression in the hippocampus is temporally complex. We have demonstrated that the presence of the gut microbiota in mice controls the expression of hippocampal miR-294-5p (Fig. 1). Little is known about the function of miR-294 but it is part of a cluster of miRNAs that are expressed in embryonic stem cells and plays an important role in embryogenesis. We found a significant increase in the expression of miR-294-5p only in the hippocampus of male mice compared to conventionally raised mice but this did not result in alterations in Brd2. The increase in Brd2 expression in colonised female animals occurs despite an increase in miR-294-5p. This suggests that factors other than miRNA expression may be responsible for Brd2 gene expression in line with studies showing that there are multiple mechanisms in Brd2 misexpression [18]. One possibility being that ovarian hormones may influence the expression of Brd2. As neurodevelopment proceeds, the complexity of miRNA involvement in the regulation of gene expression increases. The gut microbiota and embryonic development overlap significantly. Although the foetus may not be completely sterile postnatally, early life is a crucial period in the initial colonisation of the gut with maternal bacteria being the primary source [19]. Our results suggest that the crucial role of both miRNAs and the gut microbiota in neurodevelopment overlap and open up the possibility of targeting the gut microbiome to influence hippocampal miRNA expression. Disorders of the CNS are dimorphic in regards gender and even at the molecular level there is evidence for sex-biased gene expression [20]. Few studies have looked at sex differences in relation to miRNA
Table 1 PrimeTime qRT-PCR Probes used in this study. Gene (Full Name)
Assay ID
Tdao2 (Tryptophan-2-dioxygenase) Kat1 (Kynurenine-oxoglutarate transaminase 1) Kynu (Kynureninase) Haao (3-Hydroxyanthranilate 3,4-dioxygenase) Kmo (Kynurenine 3-monooxygenase) Brd2 (bromodomain containing 2) Slitrk3 (SLIT and NTRK-like family, member 3)
Mm.PT.58.13890594 Mm.PT.58.29589633 Mm.PT.58.41853210 Mm.PT.58.5340426 Mm.PT.58.9602069 Mm01271171_g1 Mm01221729_m1
and samples were processed as in [12]. MicroRNA Array Profiling was performed at Exiqon Services, Denmark and hybridized to the miRCURY LNA™ microRNA Array which contains capture probes targeting all microRNAs registered in the mirBase 18.0. All qRT-PCR experiments were performed using microRNA LNA™ PCR primer sets (mmu-miR342-3p, MIMAT0000753, mmu-miR-294-5p, MIMAT0004574, mmumiR-16-5p, MIMAT0000069). Microarray data was validated using qRT-PCR, carried out in triplicate (n = 8–11 per group). qRT-PCR data was analysed using the ΔΔCt method [13]. For qRT-PCR, 1 μg of total cellular RNA was reverse transcribed and complimentary DNA was amplified as previously described [14]. All PCR probes were specific PrimeTime® qPCR assays (Table 1) or TaqMan assays where indicated (Thermo). Gene expression was visualised using Genesis software [15]. In silico analysis of miRNA binding sites relative to gene ontology terms (tryptophan catabolic process to kynurenine, GO: 0019441) was performed using miRWalk as per platform instructions [16] and predicted miRNAmRNA binding sites were visualised using target scan [17]. Data was expressed as mean ± standard error of the mean. Statistical analysis was carried out as previously described [12]. 1. Sex and germ free status interact to alter the expression of miRNA’s Microarray analysis was performed on hippocampi removed from both male and female conventional (CON), germ free (GF) and mice that had previously been germ free but were colonised (exGF) postweaning (day 21), (Fig. 1A). Microarray analysis revealed that microbial colonisation status and sex affected the expression of hippocampal miRNAs (Fig. 1B). A two-way analysis of variance (ANOVA) analysis revealed a significant interaction between sex and colonisation status in 15 miRNAs (Table 2). 2. mir-294-5p expression in the hippocampus is regulated by the gut microbiota To validate our microarray results, we performed qRT-PCR on hippocampal RNA from the same mice from the microarray experiment. Based on the results of the two-way ANOVA and using specific primers for 3 miRNAs mmu-miR-342-3p, mmu-miR-294-5p and mmu-miR-165p (Table 1), we found a significant increase in expression of miR-2945p only in the hippocampus of male mice compared to conventionally raised male mice (Fig. 2C, p < 0.05). This increase in gene expression was reversed upon colonisation of male germ free mice. Expression of miR-16-5p was significantly increased in the germ free male mice also, though this increase was not reversed by colonisation. In males also, miR-342-3p displayed a similar pattern of expression to miR-294-5p. Collectively, these data suggests that individual miRNAs may differ in their expression patterns in the hippocampus based on sex and colonisation status. 3. mir-294-5p targets the kynurenine pathway enzymes Considering these results and the role the gut microbiota plays in 51
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
Fig. 1. Microbiota modulates miRNA expression in the hippocampus of male germ free mice. Experimental Design. (B) Differentially expressed genes between the 6 groups, red indicates increased expression and blue represents decreased expression. qRT-PCR validation of microarray data-set with selected miRNA’s. qRT-PCR analysis of hippocampal miRNA’s isolated from conventional (C, white), germ free (GF, black) and recolonized germ free mice (exGF, red). Relative expression levels of miR-294-5p (C), miR-16 (D) and miR-342-3p (E) standardized to Sno110 in male and female mice. N = 8–11 mice per group, (* = p < 0.05, Two-way ANOVA with Tukey’s HSD), $, t-test versus Control mice, (p < 0.05). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
genes, which demonstrates the importance of analysing data in different ways and also that relying solely on predicted target genes is unreliable. The prediction of miRNA target genes is possible through a large number of computational algorithms, each with differing approaches to measuring seed region complementarity and calculating mRNA: miRNA folding. These methods produce a large number of false positives and as we have shown here, biological plausibility should also
expression, though some have reported differences in expression in peripheral blood. In the hippocampus, we have demonstrated a modest sex specific increase in a miRNA primarily involved in embryonic development, suggesting that future work in the area of miRNA research in the brain should incorporate both male and female cohorts and that studies of the microbiota should also consider this phenomenon. No expression changes were evident in in-silico predicted target
Table 2 Two-way ANOVA of miRNA expression in the hippocampus showing a significant interaction between sex and colonisation status. miRNA
Type III Sum of Squares
df
Mean Square
F
p-value
Partial Eta Squared
mmu-miR-149-5p mmu-miR-186-5p mmu-miR-212-5p mmu-miR-26b-5p mmu-miR-294-5p mmu-miR-300-3p mmu-miR-3068-5p mmu-miR-31-5p mmu-miR-3473a mmu-miR-383-5p mmu-miR-409-5p mmu-miR-410-3p mmu-miR-491-3p mmu-miR-5105 mmu-miR-693-5p
0.016 0.024 0.024 0.055 0.017 0.034 0.112 0.089 0.016 0.081 0.050 0.087 0.021 0.030 0.032
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
0.008 0.012 0.012 0.027 0.008 0.017 0.056 0.045 0.008 0.040 0.025 0.044 0.010 0.015 0.016
4.175 4.036 4.528 3.452 3.912 3.353 3.981 5.010 3.570 4.380 7.251 6.419 3.538 3.568 7.639
0.032 0.036 0.026 0.054 0.039 0.058 0.037 0.019 0.049 0.028 0.005 0.008 0.051 0.050 0.004
0.317 0.310 0.335 0.277 0.303 0.271 0.307 0.358 0.284 0.327 0.446 0.416 0.282 0.284 0.459
52
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
Fig. 2. Microbiota and sex modulates expression of predicated pathway enriched genes but not predicted target genes in the hippocampus of germ free mice. (A) Microbiota and sex modulated miRNA’s (Table 2) were found to be significantly over-represented in the kynurenine metabolic pathway (GO: 0019441) using in-silico analysis. qRT-PCR expression of hippocampal kynurenine pathway genes (B and C) and predicted target genes (D and E) in male and female mice. mRNA was isolated from conventional (C, white), germ free (GF, black) and recolonized germ free mice (exGF, red). Relative expression of Kat1 (kynurenine aminotransferase) Tdao2 (tryptophan 2,3-dioxygenase), Brd2 (Bromodomain Containing 2) and Slitrk3 (SLIT and NTRK like family member 3) were measured relative to Actb (beta actin). There was no significant interaction between germ free status and sex (Two-way Anova, Table 2). Kat1 expression was significantly decreased in GF males compared with conventional mice. Tdao2 was increased in GF and exGF male mice compared to conventional mice. In female exGF mice, Kat1 expression was decreased and Tdao2 was increased. N = 8–11 per group. One-way ANOVA with Bonferroni’s Multiple Comparison Test, * = p < 0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
chain fatty acid butyrate is known to induce the expression of miR-375) or via the regulation of tryptophan metabolism [6]. Colonisation status alone may not be the exclusive determinant of miRNA regulation, factors such as diet and exercise should also be considered. Further work is needed to examine species-specific interactions using mono-association studies along with the potential role of bacterial metabolites in regulating miRNAs in the CNS. Our results show that the gut microbiota influences miRNA-associated mRNA expression patterns in the hippocampus of germ free mice. In addition, these transcriptional changes are sex-dependant pointing towards a divergence between molecular pathways that control the gut-brain axis. These changes at the molecular level may also influence the degradation of tryptophan and the generation of kynurenine pathway metabolites. In conclusion, our data underscore the role the environment plays in modulating small RNAs that influence hippocampal gene expression, a process critical to hippocampal development.
be considered during miRNA target gene selection. MiRNAs that were altered by colonisation status were over represented in the kynurenine metabolic pathway and this was confirmed by qRT-PCR, in which gene expression of kynurenine pathway enzymes were changed in germ free mice, although colonisation did not reverse these expression changes. Kynurenines have many important physiological roles in the brain and have been implicated in multiple psychiatric disorders [21]. In contrast, germ free mice display increased levels of plasma tryptophan and reduced kynurenine concentrations that are normalised upon colonisation with a normal gut flora [22]. This work suggests that miRNA: gene networks in the hippocampus can be potentially targeted for interventions by altering the gut microbiota with an expansive therapeutic range given the scope of influence of kynurenine pathway metabolites [23]. Tdao2, which was increased in germ free mice (irrespective of sex), has previously been shown to be increased in the hippocampus of a mouse model of Alzheimer’s disease [24] while Kat1, which is altered in a number of neurological disorders, [25] was decreased in germ free mice. However, caution is required until supporting evidence from enzyme activity assays or actual metabolite levels is obtained. This is especially salient given that many of the enzymes assessed, including those that may have a more prominent role in normal CNS kynurenine pathway metabolism such as Haao and Kmo, did not have measurable expression levels. It is possible that many of the behavioural effects mediated by the gut microbiota are controlled by miRNAs given the perceived overlap between the function of miRNAs and the microbiota in brain-gut communication. Mechanisms through which the gut microbiota regulate brain function remain poorly understood but there is an important intersection between components of the gut-brain axis and miRNAs including the immune system, HPA axis and the vagus nerve [3]. The impact may also be mediated by microbial metabolites (e.g. the short
Funding The APC Microbiome Institute is a research centre funded by Science Foundation Ireland, through the Irish Government’s National Development Plan Dr Clarke reported that his research is supported by Science Foundation Ireland (SFI) (grant number SFI/12/RC/2273) and by the Health Research Board (HRB) through Health Research Awards (grant number HRA-POR-2-14-647; GC). GC is supported by a NARSAD Young Investigator Grant from the Brain and Behaviour Research Foundation grant number (20771). Prof Cryan reported that his research is supported by SFI and by the Health Research Board (HRB) through Health Research Awards and through EU GRANT 613979. Prof Dinan reported that his research is supported by SFI and by the Health 53
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
[9] R.M. Stilling, et al., Microbes & neurodevelopment – absence of microbiota during early life increases activity-related transcriptional pathways in the amygdala, Brain Behav. Immun. 50 (2015) 209–222. [10] R.M. O’Connor, et al., microRNAs as novel antidepressant targets: converging effects of ketamine and electroconvulsive shock therapy in the rat hippocampus, Int. J. Neuropsychopharmacol. 16 (8) (2013) 1885–1892. [11] J. Malmevik, et al., Distinct cognitive effects and underlying transcriptome changes upon inhibition of individual miRNAs in hippocampal neurons, Sci. Rep. 6 (2016) 19879. [12] A.et al. Gururajan, MicroRNAs as biomarkers for major depression: a role for let-7b and let-7c, Transl. Psychiatry 6 (8) (2016) e862. [13] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) method, Methods 25 (4) (2001) 402–408. [14] A.E. Hoban, et al., Regulation of prefrontal cortex myelination by the microbiota, Transl. Psychiatry 6 (2016) e774. [15] A. Sturn, J. Quackenbush, Z. Trajanoski, Genesis: cluster analysis of microarray data, Bioinformatics 18 (1) (2002) 207–208. [16] H. Dweep, et al., miRWalk–database: prediction of possible miRNA binding sites by walking the genes of three genomes, J. Biomed. Inform. 44 (5) (2011) 839–847. [17] V. Agarwal, et al., Predicting effective microRNA target sites in mammalian mRNAs, eLife 4 (2015) e05005. [18] E. Shang, et al., The bromodomain-containing gene brd2 is regulated at transcription, splicing, and translation levels, J. Cell. Biochem. 112 (10) (2011) 2784–2793. [19] M.G. Dominguez-Bello, et al., Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns, Proc. Natl. Acad. Sci. 107 (26) (2010) 11971–11975. [20] D. Trabzuni, et al., Widespread sex differences in gene expression and splicing in the adult human brain, Nat. Commun. 4 (2013). [21] G. Clarke, T.W. Stone, R. Schwarcz, The kynurenine pathway: towards metabolic equilibrium, Neuropharmacology 112 (Pt. B) (2017) 235–236. [22] P.J. Kennedy, et al., Kynurenine pathway metabolism and the microbiota-gut-brain axis, Neuropharmacology 112 (Pt. B) (2017) 399–412. [23] G. Clarke, T.W. Stone, R. Schwarcz, The kynurenine pathway: towards metabolic equilibrium, Neuropharmacology 112 (Part B) (2017) 235–236. [24] W.et al. Wu, Expression of tryptophan 2,3-dioxygenase and production of kynurenine pathway metabolites in triple transgenic mice and human Alzheimer’s disease brain, PLoS One 8 (4) (2013) e59749. [25] Q. Han, et al., Structure: expression, and function of kynurenine aminotransferases in human and rodent brains, Cell Mol. Life Sci. 67 (3) (2010) 353–368.
Research Board (HRB) through Health Research Awards and through EU GRANT 613979. Acknowledgements We acknowledge the contribution of Ms Frances O’Brien and Patrick Fitzgerald in this study. This manuscript results in part from collaboration and network activities promoted under the frame of the international network GENIEUR (Genes in Irritable Bowel Syndrome Research Network Europe), which has been funded by the COST program (BM1106, www.GENIEUR.eu) and is currently supported by the European Society of Neurogastroenterology and Motility (ESNM, www. ESNM.eu). Eloisa Salvo-Romero received a Short Term Scientific Mission grant. References [1] M. Lyte, Microbial endocrinology in the microbiome-gut-brain axis: how bacterial production and utilization of neurochemicals influence behavior, PLoS Pathog. 9 (11) (2013) e1003726. [2] L. Desbonnet, et al., Microbiota is essential for social development in the mouse, Mol. Psychiatry 19 (2) (2014) 146–148. [3] N. Sudo, et al., Postnatal microbial colonization programs the hypothalamic-pituitary-adrenal system for stress response in mice, J. Physiol. 558 (1) (2004) 263–275. [4] E.S. Ogbonnaya, et al., Adult hippocampal neurogenesis is regulated by the microbiome, Biol. Psychiatry 78 (4) (2015) e7–e9. [5] P. Luczynski, et al., Adult microbiota-deficient mice have distinct dendritic morphological changes: differential effects in the amygdala and hippocampus, Eur. J. Neurosci. (2016). [6] G. Clarke, et al., The microbiome-gut-brain axis during early life regulates the hippocampal serotonergic system in a sex-dependent manner, Mol. Psychiatry 18 (6) (2013) 666–673. [7] R.D. Heijtz, et al., Normal gut microbiota modulates brain development and behavior, Proc. Natl. Acad. Sci. U. S. A. 108 (7) (2011) 3047–3052. [8] M.C. Oldham, et al., Functional organization of the transcriptome in human brain, Nat. Neurosci. 11 (11) (2008) 1271–1282.
54
Contents lists available at ScienceDirect
Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr
Short communications
Microbial regulation of hippocampal miRNA expression: Implications for transcription of kynurenine pathway enzymes
MARK
Gerard M. Moloneya,d, Olivia F. O’Learya,d, Eloisa Salvo-Romerob, Lieve Desbonneta, Fergus Shanahand, Timothy G. Dinanc,d, Gerard Clarkec,d,⁎, John F. Cryana,d a
Department of Anatomy and Neuroscience, University College Cork, Cork, Ireland Laboratory of Neuro-Immuno-Gastroenterology, Digestive Diseases Research Unit, Vall d’Hebron Institut de Recerca, Department of Gastroenterology, Hospital Universitario Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain c Department of Psychiatry and Neurobehavioural Science, University College Cork, Ireland d APC Microbiome Institute, University College Cork, Cork, Ireland b
A R T I C L E I N F O
A B S T R A C T
Keywords: Microbiota microRNA Hippocampus Kynurenine Germ-free Sex
Increasing evidence points to a functional role of the enteric microbiota in brain development, function and behaviour including the regulation of transcriptional activity in the hippocampus. Changes in CNS miRNA expression may reflect the colonisation status of the gut. Given the pivotal impact of miRNAs on gene expression, our study was based on the hypothesis that gene expression would also be altered in the germ-free state in the hippocampus. We measured miRNAs in the hippocampus of Germ free (GF), conventional (C) and Germ free colonised (exGF) Swiss Webster mice. miRNAs were selected for follow up based on significant differences in expression between groups according to sex and colonisation status. The expression of miR-294-5p was increased in male germ free animals and was normalised following colonisation. Targets of the differentially expressed miRNAs were over-represented in the kynurenine pathway. We show that the microbiota modulates the expression of miRNAs associated with kynurenine pathway metabolism and, demonstrate that the gut microbiota regulates the expression of kynurenine pathway genes in the hippocampus. We also show a sex-specific role for the microbiota in the regulation of miR-294-5p expression in the hippocampus. The gut microbiota plays an important role in modulating small RNAs that influence hippocampal gene expression, a process critical to hippocampal development.
Postnatal assembly of the gut microbiota controls the growth and functionality of the host immune system [1] along with brain development and behaviour [2]. Moreover, changes in microbiota-mediated gut-brain axis signalling have been associated with the stress response and anxiety-like behaviours [3]. The hippocampus appears to be particularly receptive to microbiome-gut-brain axis signalling with alterations noted in neuronal morphology, neurogenesis and serotonergic neurotransmission in germ free animals [4–6]. Behaviours associated with the hippocampus such as anxiety, depression and cognitive function are also abnormal in germ free mice. Alterations in the gut microbiota are known to effect the expression of neurotrophic genes such as Brain-derived neurotrophic factor (Bdnf) [7]. Cooperation of thousands of genes is required for normal neurodevelopment, thus, dysregulation of complicated gene networks are thought to be implicated in many neurodevelopmental and psychiatric disorders [8]. Activity-related transcriptional pathways in the CNS are
⁎
heavily influenced by the gut microbiome [9]. microRNAs (miRNAs) are single-stranded non-coding RNAs that regulate eukaryotic gene expression post-transcriptionally. Hippocampal changes in miRNA expression in the rat are influenced by early life stress and the immune system, and antidepressants have been shown to reverse these changes [10]. Moreover, inhibition of miRNAs in the hippocampus results in changes in cognitive behaviour in rodents [11]. Little is known regarding a microbial influence on miRNA expression in the brain and the role the gut microbiota plays in miRNA expression in the hippocampus has yet to be fully investigated. Given the pivotal impact of miRNAs on gene expression, our study was based on the hypothesis that this would also be altered in the germ free state in the hippocampus. Thus, we hypothesised that the presence of a gut microbiota is required for appropriate miRNA control of gene expression in the hippocampus. All animal experiments were performed as previously described [2]
Corresponding author at: Department of Psychiatry and Neurobehavioural Science, Biosciences Institute, University College Cork, Ireland. E-mail addresses: [email protected] (G.M. Moloney), [email protected] (O.F. O’Leary), [email protected] (E. Salvo-Romero), [email protected] (L. Desbonnet), [email protected] (F. Shanahan), [email protected] (T.G. Dinan), [email protected] (G. Clarke), [email protected] (J.F. Cryan). http://dx.doi.org/10.1016/j.bbr.2017.07.026 Received 24 May 2017; Received in revised form 17 July 2017; Accepted 18 July 2017 Available online 20 July 2017 0166-4328/ © 2017 Elsevier B.V. All rights reserved.
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
other regions of the brain, we wanted to understand what genes were targeted by the miRNAs listed in Table 2, and what potential pathways they may be involved in. Specifically, we investigated if those miRNAs that displayed a significant interaction in relation to germ free status and sex (Table 2) were overrepresented in the tryptophan catabolic process to kynurenine pathway (GO: 0019441), as recent data from our group has shown components of this pathway to be altered in germ free mice in a sex dependent manner [6]. Using this in-silico approach we found that these miRNA target genes were significantly overrepresented in the tryptophan-kynurenine pathway compared to what would be expected randomly (Fig. 2A). We performed qRT-PCR on hippocampal tissue using molecular probes for Tdao2, Kat1, Kynu, Haao and Kmo, (Table 1), the enzymes of the tryptophan-kynurenine pathway (GO: 0019441). Additionally, we measured expression of biologically relevant and predicted target genes of mmu-miR-294-5p, Brd2 and Slit3kr. Brd2 was increased following colonisation in female mice, the increase in Brd2 expression in recolonized animals may indicate that the expression of miR-294-5p is linked to immune system development with the immune-responsive miRNA leading to transcriptional activation following colonisation. That this is more pronounced in females suggests that they be more sensitive to recolonisation and its effect on transcription. All target genes measured showed no interaction between sex and colonisation status and a significant effect of germ free status was seen in kynurenine pathway genes (Fig. 2B–E). Of the kynurenine pathway genes measured, only Kat1 and Tdao2 amplified sufficiently. In male mice, Kat1 was significantly decreased in germ free mice (Fig. 2B, p < 0.05) compared with conventional mice, this decrease was not reversed by colonisation. In contrast, the gene which encodes Tdao2 was significantly increased in germ free male mice and in recolonized mice compared with conventional male mice (Fig. 2C, p < 0.05), (Fig. 2D, p < 0.05). These results demonstrate, to our knowledge, that the gut microbiota can regulate the expression of miRNAs in the hippocampus, specifically miR-294-5p. Changes in expression of miR-294-5p under germ free conditions were only evident in male mice, suggesting as we and other groups have previously shown that the microbiota can impact the CNS in a sex specific manner. The expression of some miRNAs was normalised by colonisation whereas others were not, suggesting that the interaction between the microbiota and miRNA expression in the hippocampus is temporally complex. We have demonstrated that the presence of the gut microbiota in mice controls the expression of hippocampal miR-294-5p (Fig. 1). Little is known about the function of miR-294 but it is part of a cluster of miRNAs that are expressed in embryonic stem cells and plays an important role in embryogenesis. We found a significant increase in the expression of miR-294-5p only in the hippocampus of male mice compared to conventionally raised mice but this did not result in alterations in Brd2. The increase in Brd2 expression in colonised female animals occurs despite an increase in miR-294-5p. This suggests that factors other than miRNA expression may be responsible for Brd2 gene expression in line with studies showing that there are multiple mechanisms in Brd2 misexpression [18]. One possibility being that ovarian hormones may influence the expression of Brd2. As neurodevelopment proceeds, the complexity of miRNA involvement in the regulation of gene expression increases. The gut microbiota and embryonic development overlap significantly. Although the foetus may not be completely sterile postnatally, early life is a crucial period in the initial colonisation of the gut with maternal bacteria being the primary source [19]. Our results suggest that the crucial role of both miRNAs and the gut microbiota in neurodevelopment overlap and open up the possibility of targeting the gut microbiome to influence hippocampal miRNA expression. Disorders of the CNS are dimorphic in regards gender and even at the molecular level there is evidence for sex-biased gene expression [20]. Few studies have looked at sex differences in relation to miRNA
Table 1 PrimeTime qRT-PCR Probes used in this study. Gene (Full Name)
Assay ID
Tdao2 (Tryptophan-2-dioxygenase) Kat1 (Kynurenine-oxoglutarate transaminase 1) Kynu (Kynureninase) Haao (3-Hydroxyanthranilate 3,4-dioxygenase) Kmo (Kynurenine 3-monooxygenase) Brd2 (bromodomain containing 2) Slitrk3 (SLIT and NTRK-like family, member 3)
Mm.PT.58.13890594 Mm.PT.58.29589633 Mm.PT.58.41853210 Mm.PT.58.5340426 Mm.PT.58.9602069 Mm01271171_g1 Mm01221729_m1
and samples were processed as in [12]. MicroRNA Array Profiling was performed at Exiqon Services, Denmark and hybridized to the miRCURY LNA™ microRNA Array which contains capture probes targeting all microRNAs registered in the mirBase 18.0. All qRT-PCR experiments were performed using microRNA LNA™ PCR primer sets (mmu-miR342-3p, MIMAT0000753, mmu-miR-294-5p, MIMAT0004574, mmumiR-16-5p, MIMAT0000069). Microarray data was validated using qRT-PCR, carried out in triplicate (n = 8–11 per group). qRT-PCR data was analysed using the ΔΔCt method [13]. For qRT-PCR, 1 μg of total cellular RNA was reverse transcribed and complimentary DNA was amplified as previously described [14]. All PCR probes were specific PrimeTime® qPCR assays (Table 1) or TaqMan assays where indicated (Thermo). Gene expression was visualised using Genesis software [15]. In silico analysis of miRNA binding sites relative to gene ontology terms (tryptophan catabolic process to kynurenine, GO: 0019441) was performed using miRWalk as per platform instructions [16] and predicted miRNAmRNA binding sites were visualised using target scan [17]. Data was expressed as mean ± standard error of the mean. Statistical analysis was carried out as previously described [12]. 1. Sex and germ free status interact to alter the expression of miRNA’s Microarray analysis was performed on hippocampi removed from both male and female conventional (CON), germ free (GF) and mice that had previously been germ free but were colonised (exGF) postweaning (day 21), (Fig. 1A). Microarray analysis revealed that microbial colonisation status and sex affected the expression of hippocampal miRNAs (Fig. 1B). A two-way analysis of variance (ANOVA) analysis revealed a significant interaction between sex and colonisation status in 15 miRNAs (Table 2). 2. mir-294-5p expression in the hippocampus is regulated by the gut microbiota To validate our microarray results, we performed qRT-PCR on hippocampal RNA from the same mice from the microarray experiment. Based on the results of the two-way ANOVA and using specific primers for 3 miRNAs mmu-miR-342-3p, mmu-miR-294-5p and mmu-miR-165p (Table 1), we found a significant increase in expression of miR-2945p only in the hippocampus of male mice compared to conventionally raised male mice (Fig. 2C, p < 0.05). This increase in gene expression was reversed upon colonisation of male germ free mice. Expression of miR-16-5p was significantly increased in the germ free male mice also, though this increase was not reversed by colonisation. In males also, miR-342-3p displayed a similar pattern of expression to miR-294-5p. Collectively, these data suggests that individual miRNAs may differ in their expression patterns in the hippocampus based on sex and colonisation status. 3. mir-294-5p targets the kynurenine pathway enzymes Considering these results and the role the gut microbiota plays in 51
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
Fig. 1. Microbiota modulates miRNA expression in the hippocampus of male germ free mice. Experimental Design. (B) Differentially expressed genes between the 6 groups, red indicates increased expression and blue represents decreased expression. qRT-PCR validation of microarray data-set with selected miRNA’s. qRT-PCR analysis of hippocampal miRNA’s isolated from conventional (C, white), germ free (GF, black) and recolonized germ free mice (exGF, red). Relative expression levels of miR-294-5p (C), miR-16 (D) and miR-342-3p (E) standardized to Sno110 in male and female mice. N = 8–11 mice per group, (* = p < 0.05, Two-way ANOVA with Tukey’s HSD), $, t-test versus Control mice, (p < 0.05). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
genes, which demonstrates the importance of analysing data in different ways and also that relying solely on predicted target genes is unreliable. The prediction of miRNA target genes is possible through a large number of computational algorithms, each with differing approaches to measuring seed region complementarity and calculating mRNA: miRNA folding. These methods produce a large number of false positives and as we have shown here, biological plausibility should also
expression, though some have reported differences in expression in peripheral blood. In the hippocampus, we have demonstrated a modest sex specific increase in a miRNA primarily involved in embryonic development, suggesting that future work in the area of miRNA research in the brain should incorporate both male and female cohorts and that studies of the microbiota should also consider this phenomenon. No expression changes were evident in in-silico predicted target
Table 2 Two-way ANOVA of miRNA expression in the hippocampus showing a significant interaction between sex and colonisation status. miRNA
Type III Sum of Squares
df
Mean Square
F
p-value
Partial Eta Squared
mmu-miR-149-5p mmu-miR-186-5p mmu-miR-212-5p mmu-miR-26b-5p mmu-miR-294-5p mmu-miR-300-3p mmu-miR-3068-5p mmu-miR-31-5p mmu-miR-3473a mmu-miR-383-5p mmu-miR-409-5p mmu-miR-410-3p mmu-miR-491-3p mmu-miR-5105 mmu-miR-693-5p
0.016 0.024 0.024 0.055 0.017 0.034 0.112 0.089 0.016 0.081 0.050 0.087 0.021 0.030 0.032
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
0.008 0.012 0.012 0.027 0.008 0.017 0.056 0.045 0.008 0.040 0.025 0.044 0.010 0.015 0.016
4.175 4.036 4.528 3.452 3.912 3.353 3.981 5.010 3.570 4.380 7.251 6.419 3.538 3.568 7.639
0.032 0.036 0.026 0.054 0.039 0.058 0.037 0.019 0.049 0.028 0.005 0.008 0.051 0.050 0.004
0.317 0.310 0.335 0.277 0.303 0.271 0.307 0.358 0.284 0.327 0.446 0.416 0.282 0.284 0.459
52
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
Fig. 2. Microbiota and sex modulates expression of predicated pathway enriched genes but not predicted target genes in the hippocampus of germ free mice. (A) Microbiota and sex modulated miRNA’s (Table 2) were found to be significantly over-represented in the kynurenine metabolic pathway (GO: 0019441) using in-silico analysis. qRT-PCR expression of hippocampal kynurenine pathway genes (B and C) and predicted target genes (D and E) in male and female mice. mRNA was isolated from conventional (C, white), germ free (GF, black) and recolonized germ free mice (exGF, red). Relative expression of Kat1 (kynurenine aminotransferase) Tdao2 (tryptophan 2,3-dioxygenase), Brd2 (Bromodomain Containing 2) and Slitrk3 (SLIT and NTRK like family member 3) were measured relative to Actb (beta actin). There was no significant interaction between germ free status and sex (Two-way Anova, Table 2). Kat1 expression was significantly decreased in GF males compared with conventional mice. Tdao2 was increased in GF and exGF male mice compared to conventional mice. In female exGF mice, Kat1 expression was decreased and Tdao2 was increased. N = 8–11 per group. One-way ANOVA with Bonferroni’s Multiple Comparison Test, * = p < 0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
chain fatty acid butyrate is known to induce the expression of miR-375) or via the regulation of tryptophan metabolism [6]. Colonisation status alone may not be the exclusive determinant of miRNA regulation, factors such as diet and exercise should also be considered. Further work is needed to examine species-specific interactions using mono-association studies along with the potential role of bacterial metabolites in regulating miRNAs in the CNS. Our results show that the gut microbiota influences miRNA-associated mRNA expression patterns in the hippocampus of germ free mice. In addition, these transcriptional changes are sex-dependant pointing towards a divergence between molecular pathways that control the gut-brain axis. These changes at the molecular level may also influence the degradation of tryptophan and the generation of kynurenine pathway metabolites. In conclusion, our data underscore the role the environment plays in modulating small RNAs that influence hippocampal gene expression, a process critical to hippocampal development.
be considered during miRNA target gene selection. MiRNAs that were altered by colonisation status were over represented in the kynurenine metabolic pathway and this was confirmed by qRT-PCR, in which gene expression of kynurenine pathway enzymes were changed in germ free mice, although colonisation did not reverse these expression changes. Kynurenines have many important physiological roles in the brain and have been implicated in multiple psychiatric disorders [21]. In contrast, germ free mice display increased levels of plasma tryptophan and reduced kynurenine concentrations that are normalised upon colonisation with a normal gut flora [22]. This work suggests that miRNA: gene networks in the hippocampus can be potentially targeted for interventions by altering the gut microbiota with an expansive therapeutic range given the scope of influence of kynurenine pathway metabolites [23]. Tdao2, which was increased in germ free mice (irrespective of sex), has previously been shown to be increased in the hippocampus of a mouse model of Alzheimer’s disease [24] while Kat1, which is altered in a number of neurological disorders, [25] was decreased in germ free mice. However, caution is required until supporting evidence from enzyme activity assays or actual metabolite levels is obtained. This is especially salient given that many of the enzymes assessed, including those that may have a more prominent role in normal CNS kynurenine pathway metabolism such as Haao and Kmo, did not have measurable expression levels. It is possible that many of the behavioural effects mediated by the gut microbiota are controlled by miRNAs given the perceived overlap between the function of miRNAs and the microbiota in brain-gut communication. Mechanisms through which the gut microbiota regulate brain function remain poorly understood but there is an important intersection between components of the gut-brain axis and miRNAs including the immune system, HPA axis and the vagus nerve [3]. The impact may also be mediated by microbial metabolites (e.g. the short
Funding The APC Microbiome Institute is a research centre funded by Science Foundation Ireland, through the Irish Government’s National Development Plan Dr Clarke reported that his research is supported by Science Foundation Ireland (SFI) (grant number SFI/12/RC/2273) and by the Health Research Board (HRB) through Health Research Awards (grant number HRA-POR-2-14-647; GC). GC is supported by a NARSAD Young Investigator Grant from the Brain and Behaviour Research Foundation grant number (20771). Prof Cryan reported that his research is supported by SFI and by the Health Research Board (HRB) through Health Research Awards and through EU GRANT 613979. Prof Dinan reported that his research is supported by SFI and by the Health 53
Behavioural Brain Research 334 (2017) 50–54
G.M. Moloney et al.
[9] R.M. Stilling, et al., Microbes & neurodevelopment – absence of microbiota during early life increases activity-related transcriptional pathways in the amygdala, Brain Behav. Immun. 50 (2015) 209–222. [10] R.M. O’Connor, et al., microRNAs as novel antidepressant targets: converging effects of ketamine and electroconvulsive shock therapy in the rat hippocampus, Int. J. Neuropsychopharmacol. 16 (8) (2013) 1885–1892. [11] J. Malmevik, et al., Distinct cognitive effects and underlying transcriptome changes upon inhibition of individual miRNAs in hippocampal neurons, Sci. Rep. 6 (2016) 19879. [12] A.et al. Gururajan, MicroRNAs as biomarkers for major depression: a role for let-7b and let-7c, Transl. Psychiatry 6 (8) (2016) e862. [13] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) method, Methods 25 (4) (2001) 402–408. [14] A.E. Hoban, et al., Regulation of prefrontal cortex myelination by the microbiota, Transl. Psychiatry 6 (2016) e774. [15] A. Sturn, J. Quackenbush, Z. Trajanoski, Genesis: cluster analysis of microarray data, Bioinformatics 18 (1) (2002) 207–208. [16] H. Dweep, et al., miRWalk–database: prediction of possible miRNA binding sites by walking the genes of three genomes, J. Biomed. Inform. 44 (5) (2011) 839–847. [17] V. Agarwal, et al., Predicting effective microRNA target sites in mammalian mRNAs, eLife 4 (2015) e05005. [18] E. Shang, et al., The bromodomain-containing gene brd2 is regulated at transcription, splicing, and translation levels, J. Cell. Biochem. 112 (10) (2011) 2784–2793. [19] M.G. Dominguez-Bello, et al., Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns, Proc. Natl. Acad. Sci. 107 (26) (2010) 11971–11975. [20] D. Trabzuni, et al., Widespread sex differences in gene expression and splicing in the adult human brain, Nat. Commun. 4 (2013). [21] G. Clarke, T.W. Stone, R. Schwarcz, The kynurenine pathway: towards metabolic equilibrium, Neuropharmacology 112 (Pt. B) (2017) 235–236. [22] P.J. Kennedy, et al., Kynurenine pathway metabolism and the microbiota-gut-brain axis, Neuropharmacology 112 (Pt. B) (2017) 399–412. [23] G. Clarke, T.W. Stone, R. Schwarcz, The kynurenine pathway: towards metabolic equilibrium, Neuropharmacology 112 (Part B) (2017) 235–236. [24] W.et al. Wu, Expression of tryptophan 2,3-dioxygenase and production of kynurenine pathway metabolites in triple transgenic mice and human Alzheimer’s disease brain, PLoS One 8 (4) (2013) e59749. [25] Q. Han, et al., Structure: expression, and function of kynurenine aminotransferases in human and rodent brains, Cell Mol. Life Sci. 67 (3) (2010) 353–368.
Research Board (HRB) through Health Research Awards and through EU GRANT 613979. Acknowledgements We acknowledge the contribution of Ms Frances O’Brien and Patrick Fitzgerald in this study. This manuscript results in part from collaboration and network activities promoted under the frame of the international network GENIEUR (Genes in Irritable Bowel Syndrome Research Network Europe), which has been funded by the COST program (BM1106, www.GENIEUR.eu) and is currently supported by the European Society of Neurogastroenterology and Motility (ESNM, www. ESNM.eu). Eloisa Salvo-Romero received a Short Term Scientific Mission grant. References [1] M. Lyte, Microbial endocrinology in the microbiome-gut-brain axis: how bacterial production and utilization of neurochemicals influence behavior, PLoS Pathog. 9 (11) (2013) e1003726. [2] L. Desbonnet, et al., Microbiota is essential for social development in the mouse, Mol. Psychiatry 19 (2) (2014) 146–148. [3] N. Sudo, et al., Postnatal microbial colonization programs the hypothalamic-pituitary-adrenal system for stress response in mice, J. Physiol. 558 (1) (2004) 263–275. [4] E.S. Ogbonnaya, et al., Adult hippocampal neurogenesis is regulated by the microbiome, Biol. Psychiatry 78 (4) (2015) e7–e9. [5] P. Luczynski, et al., Adult microbiota-deficient mice have distinct dendritic morphological changes: differential effects in the amygdala and hippocampus, Eur. J. Neurosci. (2016). [6] G. Clarke, et al., The microbiome-gut-brain axis during early life regulates the hippocampal serotonergic system in a sex-dependent manner, Mol. Psychiatry 18 (6) (2013) 666–673. [7] R.D. Heijtz, et al., Normal gut microbiota modulates brain development and behavior, Proc. Natl. Acad. Sci. U. S. A. 108 (7) (2011) 3047–3052. [8] M.C. Oldham, et al., Functional organization of the transcriptome in human brain, Nat. Neurosci. 11 (11) (2008) 1271–1282.
54