Neurons and neuronal activity control gene expression in astrocytes to regulate their development and metabolism.
Hasel P, Dando O, Jiwaji Z, Baxter P, Todd AC, Heron S, Márkus NM, McQueen J, Hampton DW, Torvell M, Tiwari SS, McKay S, Eraso-Pichot A, Zorzano A, Masgrau R, Galea E, Chandran S, Wyllie DJA, Simpson TI, Hardingham GE. · 2017
Read Neurons and neuronal activity control gene expression in astrocytes to regulate their development and metabolism. by Hasel P, Dando O, Jiwaji Z, Baxter P, Todd AC, Heron S, Márkus NM, McQueen J, Hampton DW, Torvell M, Tiwari SS, McKay S, Eraso-Pichot A, Zorzano A, Masgrau R, Galea E, Chandran S, Wyllie DJA, Simpson TI, Hardingham GE. (2017) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5418577.txt.
Introduction
Hasel and co-authors study how neurons shape the genes that astrocytes switch on. Their research article appeared in Nature Communications. Astrocytes are support cells in the brain. They clear transmitters and help feed neurons. How much neurons direct that work was unclear. Sorting mixed cells for study can distort the results. The team found a cleaner way to listen in.
They grew rat neurons beside mouse astrocytes. RNA reads could then be split by species on a computer. No physical sorting was needed. Neurons switched on a wide program in astrocytes. Notch signaling drove much of it. The program kept astrocytes mature and able to take up transmitters. Synaptic activity also tuned hundreds of astrocyte genes. It did so through a cAMP and CREB path. That tuning raised glucose use and lactate export to neurons. Both gene programs fade with age in humans.
Catalog record
| CREATOR | Hasel P, Dando O, Jiwaji Z, Baxter P, Todd AC, Heron S, Márkus NM, McQueen J, Hampton DW, Torvell M, Tiwari SS, McKay S, Eraso-Pichot A, Zorzano A, Masgrau R, Galea E, Chandran S, Wyllie DJA, Simpson TI, Hardingham GE. |
|---|---|
| YEAR | 2017 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 96 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Neurons and neuronal activity control gene expression in astrocytes to regulate their development and metabolism By Philip Hasel, Owen Dando, Zoeb Jiwaji, Paul Baxter, Alison C Todd, Samuel Heron, Nóra M Márkus, Jamie McQueen, David W Hampton, Megan Torvell, Sachin S Tiwari, Sean McKay Published in: Nature Communications DOI: 10.1038/ncomms15132 ABSTRACT Abstract The influence that neurons exert on astrocytic function is poorly understood. To investigate this, we first developed a system combining cortical neurons and astrocytes from closely related species, followed by RNA-seq and in silico species separation. This approach uncovers a wide programme of neuron-induced astrocytic gene expression, involving Notch signalling, which drives and maintains astrocytic maturity and neurotransmitter uptake function, is conserved in human development, and is disrupted by neurodegeneration. Separately, hundreds of astrocytic genes are acutely regulated by synaptic activity via mechanisms involving cAMP/PKA-dependent CREB activation. This includes the coordinated activity-dependent upregulation of major astrocytic components of the astrocyte–neuron lactate shuttle, leading to a CREB-dependent increase in astrocytic glucose metabolism and elevated lactate export. Moreover, the groups of astrocytic genes induced by neurons or neuronal activity both show age-dependent decline in humans. Thus, neurons and neuronal activity regulate the astrocytic transcriptome with the potential to shape astrocyte–neuron metabolic cooperation. Signalling between neurons triggers programmes of gene expression that mediate specific functions in development and maturity. Signalling between neurons and glia is also implicated in diverse processes, however, insight into the scope and mechanism by which different cell types in the neuro-glial unit influence each other is lacking. A notable example is neuron-to-astrocyte signalling: neurons induce the classic stellate morphology in astrocytes, resembling their appearance in vivo1. However, knowledge of the transcriptional changes that accompany this transformation are restricted to three astrocytic genes (Slc1a2, Slc1a3 and Gja1 (refs 2, 3)) induced by neurons. Similarly, despite astrocytes playing a critical role in the uptake of synaptically released neurotransmitters, whether synaptic activity controls astrocytic gene expression is unclear. This contrasts with the wealth of knowledge concerning the mechanisms and the roles of activity-dependent gene expression in neurons4,5,6,7. Understanding non-cell-autonomously regulated gene expression poses challenges associated with the physical separation of cell types before transcriptome analysis by methods including RNA-seq. Immunopanning or FACS are powerful approaches, but are subject to off-target cell type contamination, and cause aberrant induction of transcriptional stress responses, or the loss of material from delicate subcellular regions8,9,10. Theoretically, if RNA-seq reads could be unambiguously attributed to one cell type or another, physical sorting could be avoided. We investigated the possibility of achieving this by deriving the distinct cell types from different species11. We chose rat and mouse, two species that we reasoned may be closely enough related to enable questions of non-cell-autonomous signalling to be answered. We used this system as a starting point to investigate how and to what extent neurons and neuronal activity control astrocytic gene expression, as well as the underlying mechanisms and functional consequences. RESULTS Unsorting rat and mouse RNA-seq reads in silico We first established the feasibility of our species-specific sorting (SSS) RNA-seq workflow. In the vast majority of genes, the majority of mouse RNA-seq paired-end reads can be unambiguously attributed to mouse (as opposed to rat), using both a simulated set of reads (Supplementary Fig. 1a; Supplementary Data 1) as well as a ‘real-world’ RNA-seq data set of mouse cortical neurons (Supplementary Fig. 1b; Supplementary Data 2). Also, compared to normal read mapping, the SSS workflow does not substantially affect differential gene expression analysis (DGE) when performed on a single-species (mouse) RNA-seq data-set taken from neurons treated±bicuculline+4-aminopyridine (BiC/4-AP) to induce synaptic activity via network disinhibition12: we observed a tight correlation in fold-induction, comparing normal and SSS approaches (Supplementary Fig. 1c; Supplementary Data 3). There was also little change in the adjusted P values: of the 4,632 genes altered by BiC/4-AP treatment, only 33 were rendered insignificant by the SSS approach (Supplementary Fig. 1d; Supplementary Data 4). Large P value deviations were more likely to occur in the small number of genes that had lost a lot of reads due to the SSS approach. However, the tiny number of genes affected leads us to conclude that the SSS workflow does not substantially impact on DGE analysis. Neurons transform astrocytic transcriptome and function To profile the influence of neurons on the astrocytic transcriptome, we cultured primary mouse cortical astrocytes, onto which we co-cultured primary rat cortical neurons. The mouse cortical astrocytes are >96% GFAP positive13, which we confirmed with another marker (Aldh1l1, 99.4±0.2% positive (n=4), Supplementary Fig. 2a). The astrocytes were <0.1% Neuro-Chrom+ neurons and <0.1% Iba+ microglia. Compared to parallel astrocytic mono-cultures, co-cultured astrocytes developed the expected stellate morphology over 9 days (Fig. 1a,b). At this point, RNA was extracted (3 biological replicates), RNA-seq performed and SSS employed to identify the mouse (that is, astrocyte) reads. DGE analysis revealed widespread up- and down-regulation of astrocytic gene expression due to the presence of neurons (Fig. 1c; Supplementary Data 5), including previously identified genes (Slc1a2, Slc1a3 and Gja1 (refs 2, 3)). Four hundred and ten genes were induced more than twofold and 353 genes repressed more than twofold (Adjusted P value <0.05, Supplementary Data 5). [Figure: Mixed-species RNA-seq uncovers neuronally induced astrocytic gene expression.(a) The mixed-species co-culture system recapitulates characteristic neuron-induced changes to astrocyte morphology. Upper: example picture of a GFP-transfected mouse cortical astrocyte cultured for 10 days with (right) or without (left) rat cortical neurons. Neurons are identified by immunofluorescent staining with pan-neuronal antibody cocktail Neuro-Chrom. Lower: Staining of mouse astrocyte monoculture (left) and mixed-species co-culture (right) for astrocyte-specific gene Gfap (plus Neuro-Chrom). Scale bar, 20 μm. (b) Perimeter:area ratio of GFP+ astrocytes described in a at the indicated times (days in vitro). 6–16 cells were analysed for each culture type, for each day. *P<0.05 compared to astrocytic monoculture on that day (two-way analysis of variance (ANOVA), plus Holm-Sidak’s post hoc test). Example outlines are shown on the right. Scale bar, 20 μm. (c) Neurons promote widespread gene expression changes in astrocytes. RNA from mixed-species mouse astrocyte/rat neuron co-cultures (n=3 biological replicates) was subjected to RNA-seq, followed by SSS workflow to identify reads that were unambiguously mouse (that is, astrocytic) in origin. The same workflow was applied to mouse astrocytic mono-cultures. Expression of genes (FPKM) in astrocytes in the presence or absence of neurons is plotted for all genes expressed >0.5 FPKM average across mono- and co-cultures. Red and black crosses indicate the astrocytic genes induced or repressed by neurons more than twofold or >1.3, ≤2-fold respectively (DESeq2 P_adj<0.05). See Supplementary Data 5. The table summarizes read depth, species sorting and number of differentially expressed genes. (d) Neurons promote increased glutamate transporter capacity in astrocytes. Mouse astrocytes cultured with or without neurons (as in a–c) were subject to whole-cell voltage-clamp recording of glutamate transporter currents, measured upon application of 200 μM aspartate (see Methods). Currents were inhibited by the glutamate transporter blocker TFB-TBOA (20 μM). *P<0.05 (unpaired t-test, n=13 (mono), 22 (co-culture)). (e) Example traces. Scale bar, 5 s, 10 pA. All error bars represent s.e.m.] A notable cluster of upregulated genes is involved in the uptake and metabolism of neurotransmitters, including GABA (Slc6a1, Slc6a11 and Abat), NAAG (Slc17a5), biogenic amine neurotransmitters (Slc29a2 and Maob) and glutamate (Slc1a2, Slc1a3, Glul and Glud1; Supplementary