When you are ill, your behaviour changes. You sleep more, eat less and are less likely to go out and be active. This behavioural change is called the ‘sickness response’ and is believed to help the immune system fight infection.

An area of the brain called the hypothalamus helps to regulate sleep and appetite. Previous research has shown that when humans are ill, the immune system sends signals to the hypothalamus, likely initiating the sickness response. However, it was not clear which brain cells in the hypothalamus are involved in the response and how long after infection the brain returns to its normal state.

To better understand the sickness response, Lemcke et al. infected mice with influenza then extracted and analysed brain tissue at different timepoints. The experiments showed that the major changes to gene expression in the hypothalamus early during an influenza infection are not happening in neurons – the cells in the brain that transmit electrical signals and usually control behaviour. Instead, it is cells called glia – which provide support and immune protection to the neurons – that change during infection. The findings suggest that these cells prepare to protect the neurons from influenza should the virus enter the brain.

Lemcke et al. also found that the brain takes a long time to go back to normal after an influenza infection. In infected mice, molecular changes in brain cells could be detected even after the influenza infection had been cleared from the respiratory system.

In the future, these findings may help to explain why some people take longer than others to fully recover from viral infections such as influenza and aid development of medications that speed up recovery.

In humans as well as other species, Influenza A virus (IAV) infection causes respiratory tract disease, evoking seasonal epidemics associated with increased morbidity and mortality rates (Macias et al., 2021; Paget et al., 2019). Beyond seasonal epidemics, H1N1 IAV strains have been identified as the causal agent for the pandemics in 1918 and 2009, leading up to 50 million and 18,000 deaths worldwide, respectively (Simonsen et al., 2013; Taubenberger and Morens, 2008). Beyond respiratory tract disease, IAV infections have also been associated with neurological consequences such as encephalitis lethargica and narcolepsy. Evidence from the Spanish flu pandemic in 1918 linked H1N1 influenza infections to epidemic encephalitis lethargica (Hoffman and Vilensky, 2017; Maurizi, 2010). Similarly, increased numbers of encephalitis cases were reported during the 2009 H1N1 pandemic (Goenka et al., 2014; Gu et al., 2013; Surana et al., 2011). A peak in new narcolepsy cases was also seen following the 2009 H1N1 pandemic, and while this was linked with the Pandemrix vaccine in Europe, in China it was unrelated to vaccinations and thus might instead be linked to infection with the virus itself (Ahmed et al., 2014; Han et al., 2011). A better understanding of the direct and indirect effects of IAV infection on the brain is needed to shed light on these observations.

Some IAV strains are considered neurotropic, meaning that they hold the machinery to infiltrate the CNS and replicate, but evidence suggest that even non-neurotropic influenza strains can infiltrate the brain and cause neurological disturbances (Ekstrand, 2012; Hodgson et al., 2012; Macias et al., 2021; Majde et al., 2007; Zielinski et al., 2013). When non-neurotropic viruses are found in the CNS, they are either not able to replicate at all or not efficiently in the brain resident cells (Hosseini et al., 2018; Zielinski et al., 2013). The H1N1 strain from the 2009 pandemic (H1N1 pdm09) is non-neurotropic (Sadasivan et al., 2015), so the mechanism behind the neurological consequences seen after that H1N1 infection must be caused by indirect effects rather than direct viral infection of brain cells. It is well known that cytokines produced during a peripheral infection can reach the brain and cause functional disturbances (Dantzer and Kelley, 2007). It has also been shown that microglia become activated during peripheral IAV infections with non-neurotropic strains. Whether this microglia activation is a neuroprotective mechanism or contributes to neuroinflammation and -degeneration remains elusive (Dusedau et al., 2021; Sadasivan et al., 2015; Wiley et al., 2015).

The hypothalamus is a functional integrator of peripheral signals and regulates several physiological and behavioural functions via several specialised nuclei (Berthoud and Münzberg, 2011; Bonnavion et al., 2016; Stuber and Wise, 2016). The different hypothalamic neuron populations control among other things feeding, body temperature, sleep-wake states, and circadian rhythms (Bonnavion et al., 2016; Timper and Brüning, 2017; Yousefvand and Hamidi, 2021). The lateral hypothalamic area (LHA) contains well described neuron populations like hypocretin (HCRT) or melanin-concentrating hormone releasing (MCH) neurons. These neurons are coupled to several biological functions including metabolism and the regulation of sleep and wake states (Oesch and Adamantidis, 2021). Recent studies showed that the LHA is also involved during inflammatory processes and regulates pain perception (Bonnavion et al., 2016; Brown et al., 2015; Fakhoury et al., 2020; Stuber and Wise, 2016). The dorsomedial hypothalamus (DMH) receives circadian input from the suprachiasmatic nucleus to regulate downstream behavioural and physiological changes including food intake, locomotion, and core body temperature (Gooley et al., 2006). Recent findings have linked neuronal subpopulations in the DMH to changes in thermoregulation and metabolism during immune activation with the bacterial cell wall component lipopolysaccharide (LPS; Osterhout et al., 2022). Several single-cell/nuclei-RNA sequencing (scRNA-seq, snRNA-seq) datasets have been published during the past years, shedding more light on the cell heterogeneity of the different hypothalamic regions, but little is known about how the transcriptome of these cell types change during environmental perturbations (Campbell et al., 2017; Chen et al., 2017; Mickelsen et al., 2019; Mickelsen et al., 2020; Romanov et al., 2017).

In the current study, we used intranasal infections with H1N1 pdm09 in female mice and snRNA-seq to follow changes in the transcriptional landscape of cell populations in the medial hypothalamus. We find changes in all populations during the peak of disease. A change in cell state is seen in microglia at early infection and later strong effects are detected in astrocytes and oligodendrocytes. Interestingly, after the mice have recovered from the acute infection, all non-neuronal cell types return to their resting state, while in neurons we detect long lasting changes of the transcriptome even after full recovery.

Mice were infected intranasally with influenza virus A/California/07/2009 (H1N1 pdm09) and monitored for a period of 23 days post infection for weight changes. Groups of mice were euthanized at 3, 7, and 23 days post infection (dpi) to collect lung, olfactory bulb (OB) and brain tissue for detection of viral abundance and snRNA-seq. The rationale for choosing these time points of sampling were as follows: (i) 3 dpi as a time point of initial disease onset before the start of weight loss – at this time point innate immunity is peaking; (ii) 7 dpi marks the point of strong weight loss and is where the adaptive immune response takes over, and before potential lethal outcome for some of the animals and (iii) 23 dpi as a recovery group where animals had completely regained their initial body weight (Uddbäck et al., 2016).

IAV respiratory tract disease progression and viral infiltration into lung and olfactory bulbs

Infected mice started losing weight around 4 dpi and had the highest weight loss at 10 dpi (Figure 1B). Animals in the recovery group had regained their body weight at 23 dpi. Lungs and OB were tested for the abundance of the M1 gene of the virus, to test for viral infection of those tissues. Viral mRNA was detected at 3 and 7 dpi in the lungs but was not detected at 23 dpi (Figure 1C). In the OB, viral mRNA was detected in one animal at 3 dpi, in all animals at 7 dpi, and in four out of five animals at 23 dpi. Three of the animals did not loose weight due to the infection (7 dpi: 7.3; 23 dpi: 23.3 and 23.4) instead the presence of viral mRNA in lungs and/or OB confirmed a successful infection (Figure 1—figure supplement 2).

Figure 1 with 4 supplements see all

Download asset Open asset

Identified NonN cell types

Among the non-neuronal cell types, we were able to identify 9 distinct populations, discriminated by their sets of cell-type-specific markers (Figure 2A and B). We used published datasets of different regions of the hypothalamus to confirm the identity of the distinct cell clusters (Campbell et al., 2017; Chen et al., 2017; Mickelsen et al., 2019; Mickelsen et al., 2020; Romanov et al., 2017). Two of the clusters belonged to the oligodendrocyte (OL) lineage, including oligodendrocyte precursor cells (OPCs, NonN_3) and mature oligodendrocytes (MOLs, NonN_1). The main marker genes in these two clusters were Lhfpl3 and Slc24a2 but they also expressed other typical marker genes as Pdgfra, C1ql1, Fyn (OPCs) or Olig1, Anln, Plp1 in MOLs. Further, we resolved a cluster of astrocytes (NonN_2), which was defined by the differential expression of Slc4a4 and which was also positive for Sox9, Gja1 and Agt. The nearby cluster NonN_7 expressed similar characteristics as the astrocytes and exhibited a high expression of Slit2, Col23a1, and Slc16a2 which have been described to be enriched in tanycytes (Chen et al., 2017; Figure 2B and D, and Supplementary files 2 and 4, Figure 2—figure supplement 1). The identity of this cluster was confirmed by label transfer of other snRNA/scRNA-seq datasets from the hypothalamus, which showed a high prediction score of this cluster as tanycytes (Supplementary files 2 and 3, Figure 2—figure supplement 2). The distinct clusters NonN_5 (Ranbp3l+, Cped1+, Slc6a20a+) and NonN_9 (Bnc2+, Slc47a1+, Slc26a7+) were identified as vascular and leptomeningeal cells, respectively. The NonN_8 cluster expressing Slco1a4 is expressing additionally Ebf1, Flt1, Rgs5, which are described as pericyte marker genes (Pagani et al., 2021; Zeisel et al., 2018). The cluster NonN_6 is a microglia cluster, with expression of Hexb, C1qa, and C1qc - known microglia marker genes. Best predictions scores for cell type label transfer were found for non-neuronal cells, with mean prediction scores 0.8 comparing the HypoMap C7 cell type levels (Figure 2—figure supplement 1; Steuernagel et al., 2022). Lastly, even though the cluster NonN_4 (Syt1+) is expressing general neuron markers, we could not identify any discriminatory markers for inhibitory or excitatory neuron populations (Figure 2D). Discriminatory marker genes such as Slc17a7, Crym, Ptk2b, Itcam5, Slit3, Crym indicate a progenitor/neuroblast identity (Kostin et al., 2021; Zeisel et al., 2018). Aligning this dataset with a dataset from the developing dentate gyrus including cells from the neuroblast lineage, indicates the presence of an immature granule subcluster strengthening the idea of NonN_4 as a neuroblast cluster (Figure 2—figure supplement 2; Hochgerner et al., 2018). However, because of the general low confidence and weak alignment we have kept this cluster unassigned.

Figure 2 with 2 supplements see all

Download asset Open asset

Heterogeneous populations of inhibitory and excitatory neurons

To subcluster the neurons, we first examined the expression signature of neuronal markers: Vesicular glutamate transporter 2 (VGLUT2; encoded by Slc17a6); Vesicular GABA Transporter (VGAT, encoded by Slc32a1), glutamate decarboxylase 67 (GAD67; encoded by Gad1); and histidine decarboxylase (encoded by Hdc) and separated the neuronal cells in GABAergic and glutamatergic cells (Figure 3A). We identified a total of 26 GABAergic and 30 glutamatergic clusters. Within the glutamatergic cells we found also a small population of histaminergic (Glut_24; Hdc+) and dopaminergic neurons (Glut_27; Th+, Ddc+). The histaminergic neurons were left for visualisation and analysis purposes in this cell type, despite them often being described as GABAergic (Figure 3B). Among the glutamatergic cells, we were able to detect two small clusters of HCRT expressing neurons (Glut_29; Hcrt+) and MCH expressing neurons (Glut_25; Pmch+). The cellular resolution aligns well with the composition of other cell types seen in other studies of the hypothalamus (Campbell et al., 2017; Chen et al., 2017; Mickelsen et al., 2019; Mickelsen et al., 2020; Romanov et al., 2017; Steuernagel et al., 2022). Among well-described neuronal cell cluster, we found a good overlap with Hcrt+, Pmch+, Hdc +and Prkch +neurons (Figure 3—figure supplement 1) and Pomc+, Nkx2−4+GABAergic neurons (Supplementary file 7). Additionally, we identified a Argp+/Npy +co-expressing cell cluster as a sub-cluster of GABA_1 (Figure 8—figure supplement 2).

Figure 3 with 4 supplements see all

Download asset Open asset

General transcriptional disease responses within the medial hypothalamus across all cells

To understand the nature of the molecular changes between the control group and the infected groups at different time points of disease progression, we first used a pseudo-bulk approach to compare their transcriptional signatures of all cells per time point against each other. A principal component analysis based on the pseudo-bulk dataset showed a clear clustering of the biological replicates per time point (Figure 4A). We found a clear separation of the samples taken at 3 dpi from the other time points. Seven dpi is clustering separately between 3 dpi and the control samples, whereas the samples from the recovery group at 23 dpi are spread between the 7 dpi and the control group cluster. This indicates a clear transcriptional change during the first 7 days of infection. The recovery group is mostly overlapping with the control group, but one sample clusters together with the 7 dpi group, suggesting a slower disease or recovery process in this animal (Animal 23.4, Figure 4A). The overall gene expression counts show a pattern of lower counts per cell at 3 dpi in both GABAergic and glutamatergic cell populations. In glutamatergic cells this is a lasting but less pronounced effect until 23 dpi (Figure 4B, Supplementary file 5). Next, we performed a differential gene expression (DGE) analysis comparing the control samples to samples from different time points during disease progression, disregarding any differentiation between cell subpopulations. We tested, based on previous studies, gene sets know to be involved in neuroinflammatory and immune regulatory process within the CNS for possible changes in gene expression (Dusedau et al., 2021; Hosseini et al., 2018; Sadasivan et al., 2015; Zielinski et al., 2013). We primarily detected a down-regulation of these gene categories, which was strongest at 3 dpi with only very few significant changes lasting until 23 dpi (Figure 4C).

Figure 4

Download asset Open asset

The only up-regulated genes were detected at 3 and 7 dpi belonging to the category of neuron-related factors. Genes in the other categories were consistently down-regulated or not significantly regulated at all. This also included the virus-related cytokines and IFN-response genes (i.e. Il1a, Ifih1, Figure 4C) at 3 dpi. Toll-like receptors (TLRs) are central in initiating the innate immune response by orchestrating signalling pathways leading to the release of inflammatory cytokines (Kawai and Akira, 2011). In our dataset, we see a general downregulation of TLRs (Tlr2, Tlr3, and Tlr4, Figure 4C). These results show that a peripheral H1N1 pdm09 infection cause decreased gene expression in non-neuronal, GABAergic, and glutamatergic cell populations in the hypothalamus affecting several immune related processes.

Specific molecular regulation in non-neuronal, glutamatergic, and GABAergic cell populations

Next, we resolved the molecular disease responses in better detail, by performing a differential gene expression (DGE) analysis on the three overall cell populations (GABAergic, glutamatergic, and NonN). Similarly, to the previous pseud-bulk approach, we detected the strong down-regulation of genes at 3 dpi in all the three distinct cell populations (Figure 5A). The overall number of down-regulated genes decreased at 7 and 23 dpi, but more differential expressed genes were detected in the GABAergic cells at 23 dpi, than at 7 dpi (Figure 5A and B). Of all highly regulated genes at 3 dpi (False discovery rate (FDR) ≤ 0.05, log fold change () >1 or logFC < –1), approximately a quarter (27%) of the genes were shared between the 3 cell populations, and 30% were shared between the two neuronal clusters. Around 10% of the genes expressed in the GABAergic and glutamatergic cell clusters at 3 dpi were found expressed only in these cells (Figure 5B). In contrast, 20% of the DEGs in the NonN cells at 3 dpi were unique to these cells only. At 7 dpi, only 4 genes were shared between all 3 clusters, and 59 genes between the two neuronal clusters (Figure 5B). At 23 dpi no more DEGs were detected in the NonN cells. Most differentially expressed genes (DEGs) at this time point belonged to the GABAergic cluster. To exclude if persistent transcriptional effects reported in neurons but not in non-neurons are caused by the different amount of captured transcripts between neurons and non-neurons (Figure 1G), we created a down-sampled dataset with similar transcripts per cell in neurons and non-neurons (Figure 5—figure supplement 1B). Here we reported the same effect, of high down-regulation of genes in all cells (Figure 5—figure supplement 1A) at 3 dpi, with a less pronounced pattern at 7 dpi. In the recovery group at 23 dpi, we found again a transcriptional response only in the neuronal cells but nit in non-neuronal cells (Figure 5—figure supplement 1A). To study what functional effects these detected changes in gene expression might have, we performed a gene ontology enrichment analysis. Figure 5C shows all the significantly regulated pathways. Of special interest, we detect an enrichment of genes involved in immune responses at 3 dpi in all three clusters, while this is restricted to the neuronal clusters at 7 dpi and only the Glut cluster at 23 dpi (Figure 5C). Further, enrichment of neuropeptide signalling, and hormone activity pathways are also present in all three clusters at 3 dpi, while lasting until 23 dpi only in the neuronal cells. These data suggest that long term effects in the hypothalamus after a peripheral IAV infection affects primarily neurons.

Figure 5 with 1 supplement see all

Download asset Open asset

Dissecting immune system processes in microglia

We tested all identified NonN and neuronal cell populations for differential gene expression. A cell cluster was excluded from analysis if their cells were originating from less than three of the biological replicates and if they were composed of less than two cells per cluster per biological replicate. Similarly, we used for further downstream analysis only strongly expressed genes with high up- or down-regulation (FDR ≤ 0.05, logFC >1 or logFC<-1). As shown before, the strongest regulatory effect on the different clusters’ transcriptomes were detected at 3 dpi, which lasted partially until 23 dpi (Figure 6A). Several GABAergic clusters down-regulated more genes at 23 dpi than at 7 dpi. These cell populations (GABA 2, 4, 6, 8, 10, 12) are aligning close to each other in the dimensional reduction (Figure 3B) and are overlapping in the centre of the UMAP. The glutamatergic neuron populations show in general a similar pattern, with the strongest transcriptional response at 3 dpi and less differentially regulated genes at 7 and 23 dpi.

Figure 6 with 6 supplements see all

Download asset Open asset

As described in the previous paragraph, we identified immune response pathways as one of the most enriched gene ontologies (GO) terms in the dataset. Therefore, we selected only the genes annotated in these immune-related GO terms, leaving us with 22 genes. Interestingly, all these 22 immune related genes are regulated in microglia (NonN_6) at 3 and 7 dpi (Figure 6C). In other non-neuronal clusters less than half of this genes are differentially regulated at 3 dpi and none at 7 or 23 dpi. Similarly, neuronal cell populations showed fewer genes regulated at 3 dpi than microglia, however especially some GABAergic cell types show a transcriptional response of immune genes until 7 and even 23 dpi (i.e. GABA_7, 16 and 17). The here reported highest regulation of immune genes indicate that microglia are also in the medial hypothalamus the main immune orchestrating cell population.

We next aimed to describe the transcriptional changes in the microglia during disease progression. Raw DEG numbers emphasize the microglia cluster as the only cell cluster with a strong upregulation of genes at 3 dpi (75 genes), and a long-lasting down-regulation of genes until 23 dpi (52 and 12 genes at 7 and 23 dpi respectively; Figure 6A, cluster NonN_6), indicating the importance of microglia for immune homeostasis in the medial hypothalamus. A closer look at the highest up-regulated genes reveals a strong induction of genes involved in cell adhesion, signalling and differentiation, synapse organization as well as apoptosis signalling. Down-regulated genes participate in the complement system, immune and antigen processing, microglia state genes and ribosomal genes contributing to protein synthesis (Figure 6E).

Typical markers like cell surface proteins or intracellular markers are often used to identify pro- (M1 state; i.e. Il1b, Ifng, Tfna, Fcgr3) or anti-inflammatory (M2 state; i.e. Tgfb, Mrc1, or Arg1) states of microglia (Jurga et al., 2020; Moehle and West, 2015). These markers were not regulated during disease progression in our dataset. Discrimination of the activation state of microglia using snRNA-seq has been reported to be challenging, due to a depletion of certain marker genes in the nucleus especially in human tissue (Thrupp et al., 2020). Because of this lack of typical M1 and M2 markers, we identified alternative DEGs in microglia at different the time points related to activation states. Two long non-coding RNA (lncRNA) genes: Maternally Expressed 3 (Meg3), Imprinted in Prader-Willi Syndrome (Ipw) and the gene Fibroblast Growth Factor 14 (Fgf14) which have been recently connected to immune activation and pro-inflammatory microglia states (Liu et al., 2021; Meng et al., 2021; Senatorov et al., 2019), were found in our dataset enriched in the microglia at 3 and partly at 7 dpi (Figure 6D). In contrast, genes connected to an anti-inflammatory state (F13a1, Mrc1, Arhgap14), were expressed in less microglia cells during acute infection than in control and 23 dpi. The results show that microglia are also in the medial hypothalamus a main immune orchestrating cell population, shifting from their homeostatic state towards a pro-inflammatory microglia population during 3 and 7 dpi. This shift is completely resolved after recovery at 23 dpi.

Identification of oligodendrocyte and astrocytes subclusters at 7 dpi

In oligodendrocytes (NonN_1) and astrocytes (NonN_2) between 144–244 genes were up- or down-regulated (logFC >0 or logFC <0) at 3 and/or 7 dpi (Figure 6A). This was accompanied by a shift of the cells in the dimensional reduction plot of the not integrated data, especially at 7 dpi (Figure 7A). In the oligodendrocytes this was accompanied by an up-regulation of genes involved in stress-related and glucocorticoid pathways (Figure 7B and C). The same was not seen in the astrocyte sub-cluster. We identified significantly regulated GO terms associated with transport pathways in oligodendrocytes, with a strong up-regulation of genes in these terms at 3 and 7 dpi (Figure 7D). Only mitochondrial transport processes and transmembrane transport included down-regulated genes. Genes associated with GABAergic transport was up-regulated in oligodendrocytes uniquely at 7 dpi. Marker genes for the astrocytic 7-dpi cluster were identified as Aldh1a1, Lrrc8a (annotated here as ENSMUSG00000099041) and Zfhx3. Zfhx3 showed a very distinct expression in this cluster (Figure 7C). For astrocytes, the GO term oxidoreductase activity was highly and uniquely enriched at 7 dpi (Supplementary file 8). Accordingly, we found the oxidoreductase genes to be highly up-regulated in the 7 dpi cluster, with only a few being also regulated a 3 dpi and non at 23 dpi (Figure 7E). Together these results suggest that an IAV infection induces slowly evolving transcriptional changes in oligodendrocytes and astrocytes, leading to the appearance of distinct sub-populations of these cell types at 7 dpi. This altered cell stated disappear again after recovery at 23 dpi.

Figure 7 with 1 supplement see all

Download asset Open asset

Long-term transcriptional response in neurons following as IAV infection

Even though the most prominent feature of the hypothalamic transcriptional response to H1N1 pdm09 was the downregulation of genes at 3 dpi across all cell clusters, some neuronal cell populations showed long-term differential expression of genes. As described in previous paragraphs, we detected overall changes in both GABAergic and glutamatergic neurons at 23 dpi (Figure 5B). When looking at neuronal subclusters, we found that 6 out of 26 GABAergic cell clusters and 17 out of 30 glutamatergic cell clusters contained significantly regulated genes at 23 dpi when compared to the control group (Figure 6A). The cluster based case-control analysis (Cacoa) showed a similar dysregulation of neuronal and non-neuronal cell populations at 3 and 7 dpi (Figure 6—figure supplement 2). However, the long-lasting transcriptional changes described before, were not reproduced with this cluster-based analysis. Though, analysis of cluster-free expression shifts shows dysregulation of neuronal and non-neuronal cells at the acute phase of the infection (Figure 6—figure supplements 4–6), however detects mainly transcriptional changes in neurons at 23 dpi (Figure 6—figure supplements 3 and 6). Our initial GO term enrichment analysis (Figure 5C) showed several different pathways being affected at all timepoints, so we aimed at dissecting this further by analysing individual neuronal cell clusters. In this analysis we included glutamatergic and GABAergic cell clusters, with a high DGE profile and at least 10 genes exhibiting a logFC greater than 1 or less than –1. The 15 neuronal clusters fulfilling this criterion were: Gaba_2, 4, 6, 8, 10, 12 and Glut_1, 4, 6, 9, 11, 15, 17, 18 and 20. In general, we found the same regulatory processes to be regulated across many of the neuronal cell types. When looking at the pathways that remain altered at 23 dpi these include translational processes in the cytosol, metabolic processes in the mitochondria, as well as apoptotic processes (i.e. mitochondrial respiration and TCA cycle) (Figure 8, Supplementary files 7 and 8).

Figure 8 with 2 supplements see all

Download asset Open asset

Our analysis shows that many of the neuronal subpopulations respond with a similar pattern when the organism is challenged by an infection. To determine whether there would also be differences in response pattern between neuronal subtypes, we selected six hypothalamic neuronal subtypes with known regulatory functions in metabolism: Agrp⁺/Npy⁺, Lepr⁺, and POMC⁺ neurons; and sleep: Hdc⁺, Pmch⁺, and Hcrt⁺ neurons. With both feeding and sleep patterns changing during influenza, we would expect these neuronal subtypes to change their gene expression during the experiment. This is indeed what we find. While most of the subtypes respond with overall downregulation of gene expression (Figure 8C–D), the POMC⁺ neurons show a different pattern with equal and strongly both up and downregulated genes.

Taken together, neuronal populations were the only cell populations that were affected by the H1N1 pdm09 infection until recovery, with changes including inhibition of genes contributing to metabolic and translational processes in the neurons. While most neuronal subpopulations follow this same regulatory pattern, POMC⁺ neurons show bigger and more unique changes in gene regulation following a peripheral influenza infection.

The compiled snRNA-seq dataset presented above, to our knowledge, represents the first comprehensive transcriptome profiling on single-nuclei level of the hypothalamus during a peripheral IAV infection. We detected 26 GABAergic, 30 glutamatergic and 9 non-neuronal cell types, aligning closely with previously published datasets of the hypothalamus (Campbell et al., 2017; Chen et al., 2017; Mickelsen et al., 2019; Mickelsen et al., 2020; Romanov et al., 2017; Zeisel et al., 2018). In addition, we deconvoluted the molecular changes of distinct cell populations in the hypothalamus during a peripheral IAV infection and detected a whole range of transcriptional changes across both NonN and neuronal cell types. In addition, we believe our dataset is a valuable resource for the scientific community to explore unsolved questions of hypothalamic responses during peripheral infections.

We were able to report viral infection of the olfactory bulb using H1N1 pdm09. This is comparable to what has been seen in previous studies using the non-neurotropic H1N1 strain PR8 (Majde et al., 2007; Zielinski et al., 2013). The same studies found replication intermediates of the viral nucleoprotein (NP1), suggesting the potential ability of the virus to replicate further within the OB. However, another mechanism has been suggested, in which the virus continuously replicate in the nasal epithelium and then gets transported into the OB (Zielinski et al., 2013). Other areas deeper in the CNS, like the hippocampal region, did not show any viral infiltration by PR8. (Hosseini et al., 2018; Jurgens et al., 2012). Similar results were reported for infections with H1N1 pmd2009, where the virus could not be detected in the hippocampus, striatum or cortex, supporting the hypothesis that this virus is not replicating in the CNS (Sadasivan et al., 2015).

In contrast to our results, studies using the PR8 disease model reported elevated mRNA expression levels of cytokines related to viral infections 24 hr and 6 days post infection (Zielinski et al., 2013). We found in general a strong down-regulation of genes at 3 dpi, including also virus-related cytokines and IFN-response genes (i.e., Tlr3, Il1a, Ifih1) at 3 dpi with no differential regulation of these transcripts at 7 dpi. This indicate that we do not have neuroinflammatory processes in the hypothalamus in our model, or that the here chosen time point of 3 dpi might be too late to capture innate immune activation in the CNS.

Even though we used different published datasets to annotate the here-in identified cell types or compare gene expression changes in a disease context, the comparability of these datasets is challenging due to differences in the experimental setups. We identified a lack of RNA sequencing studies focusing solely on molecular changes in the hypothalamus during non-neurotropic IAV challenges. Different molecular changes based on viral strain or between brain regions are expected and further studies are needed to resolve this.

A prominent feature of our dataset is a long-lasting down-regulation of gene transcripts until 23 dpi in many neuronal cell populations. Gene set enrichment analysis showed these to be mainly related to processes in mitochondria (i.e. citric acid cycle (TCA), ATP synthesis) or to protein translation in the cytosol. This could be interpreted as a defensive mechanism, for the neurons to escape viral entry or replication in the cell and hence protect the cells from apoptosis. Different intermediates of the citric acid (TCA) cycle can be important for viral replication within the cell. In the context of IAV, acetyl-CoA and ATP are known to be beneficial for viral replication. Acetyl-CoA is also used for viral acetylation and ATP is needed for budding of the virus (Hatakeyama et al., 2018; Hui and Nayak, 2001; Sánchez-García et al., 2021). The downregulation of genes contributing to the translational machinery could be another attempt to inhibit translation of viral proteins in case of a potential entry of the virus into the cell. Another explanation for our findings could be that the downregulation of metabolic and translational processes in the cell might be caused by the excessive weight loss over the disease course and the lack of energy resources (Figure 9). However, we did see these changes also after full weight regain at 23 dpi. In addition, our results indicate that the virus is present in the OB until complete recovery at 23 dpi, suggesting this as one potential trigger for the here reported long-lasting transcriptional changes on hypothalamic neuron populations. Further studies are needed to determine for how long the virus remains in the system and when/if the neurons return fully to the healthy control condition.

Figure 9

Download asset Open asset

A subanalysis of Agrp⁺/Npy⁺, Lepr⁺, POMC⁺, Hdc⁺, Pmch⁺, and Hcrt⁺ neurons showed a particularly strong response of the POMC⁺ neurons compared to the other neuronal subtypes. POMC⁺ neurons have been shown before to be activated during inflammation in various models, a mechanism that is speculated to be important for development of sickness behaviours (Jin et al., 2016; Kang et al., 2020). This fits well with our observation that the POMC gene it self is strongly upregulated at 3 and 7dpi. Microglia, the resident immune cells of the brain, can have based on their activation state different immunological functions and properties. Interestingly, our results demonstrate the activation of pro-inflammatory genes in the microglia population, whereas anti-inflammatory transcripts are depleted. Further, general microglia signature genes (i.e, Hexb, Cx3cr1 and Selplg) were found down-regulated at 3 and partly at 7 dpi, indicating a clear shift in microglia homeostasis at 3 dpi. This assumption is supported by the activation of genes contributing to cell adhesion, signalling and differentiation as well as synapse signalling. A pro-inflammatory microglia state (M1) is induced during an acute immune response and provide acute defence and repair mechanisms. For this, a tight control via constant cell-to-cell communication is necessary to avoid excessive long-term activation and eventual neurodegenerative consequences (Rodríguez-Gómez et al., 2020). In contrast, anti-inflammatory microglia (M2) are regulating the phagocytosis of pathogen and cell debris, inhibit inflammation and restore tissue homeostasis (Franco and Fernández-Suárez, 2015; Jurga et al., 2020). While our data does not support the presence of an acute inflammatory process in the hypothalamus, what we instead find is an increased alert state of the resident immune cells (Figure 9).

The most striking shift in gene expression in our dataset was detected in oligodendrocytes and astrocytes at 3 and 7 dpi, which caused the separation of the cell populations into distinct sub-populations at 7 dpi for both cell types. In oligodendrocytes, we detect a distinct expression of Serum/glucocorticoid Regulated Kinase 1 (Sgk1) in the 7-dpi population. This protein belongs to the family of glucocorticoid receptors (GCR) and has been associated with several environmental pertubations, including its up-regulation in the CNS during stress or during fasting in obese individuals (Avey et al., 2018; Hinds et al., 2017; Miyata et al., 2015; Nonogaki et al., 2006). Further, the activation of SGK1 in oligodendrocytes is believed to lead to morphological changes, potentially caused by abnormal swelling of cytoplasm in oligodendrocytes or arborization of their processes (Miyata et al., 2015; Miyata et al., 2011). Fkbp5, encoding for FK506-binding protein 5, was another strongly induced transcript in oligodendrocytes at 7 dpi. It has previously been shown to be induced by glucocorticoids (Hähle et al., 2019) and involved in myelination in the prefrontal cortex (Choi et al., 2021). One of the main functions of oligodendrocytes is myelin formation around neuronal axons and the metabolic support of these to ensure neuronal signalling and information processing (Lee et al., 2012; Moore et al., 2020; Saab et al., 2016). In addition, oligodendrocytes are involved in orchestrating stress responses mediated by GCRs. This is relevant for our model, as glucocorticoids are released following immune activation. This release usually occurs in a bi-phasic pattern, with an initial peak at 2–3 dpi caused by pro-inflammatory cytokines and a second peak at 7–8 dpi driven by T-cell-specific cytokines (Hermann et al., 1994; Silverman et al., 2005). Since the disease-specific subpopulation of oligodendrocytes, we detect at 7-dpi shows markers related to GC signalling, this sub-cluster is likely driven by T–cell specific cytokines.

For the astrocytes, one of the most pronounced discriminatory factors for the 7-dpi specific population was the expression of the Leucine-Rich Repeat-Containing Protein 8 A (Lrrc8a, also Swell1, annotated as ENSMUSG00000099041) transcript. Lrrc8a has putative regulatory transmembrane transporter functions and is one essential subunit of volume-regulated anion channels (VRAC). Hence, it plays an important role in cell volume changes and regulate synaptic transmission via glutamate release from astrocytes (Osei-Owusu et al., 2018; Yang et al., 2019). The upregulation of Lrrc8a suggests changes in astrocytic morphology in the hypothalamus in our model, comparable to what has been reported in the hippocampus during peripheral IAV infections (Hosseini et al., 2018). However, the functional importance of this change in our study stays elusive and needs further investigation. In addition, we showed that the zinc finger homeobox 3 (Zfhx3) transcript is specifically up-regulated in astrocytes at 7 dpi. Its genomic locus has been associated with body mass index and it is further known to influence circadian rhythm and sleep (Balzani et al., 2016; Parsons et al., 2015; Turcot et al., 2018). The latter two are believed to be also regulated via astrocytic released glutamate (Brancaccio et al., 2017). Further, we identified a distinctive expressed gene in a part of the astrocyte population - aldehyde dehydrogenase 1 A1 (Aldh1a1), which is contributing to retinol metabolism. The inhibition of Aldh1a1 leads to reduced weight gain (Haenisch et al., 2018). Aldh1a1 is known to mediate the synthesis of GABA in astrocytes (Kwak et al., 2020), which is then secreted by Best1-channels and GABA transporters (GAT, Slc6a1 and Slc6a11) transporters (Mederos and Perea, 2019). These two transporters were found robustly expressed in the here analysed astrocyte population but not differentially expressed at any time point. The release of glutamate by astrocytes is regulated via the excitatory amino acid transporters (glutamate transporter 1; GLT-1, also known as Slc1a2 and Glutamate Aspartate Transporter 1; GLAST, also known as Slc1a3) (Delgado, 2013), which we found highly expressed in the here analysed astrocytes (Figure 7—figure supplement 1). Additionally, we identified an up-regulation of genes of the oxidoreductase system. Such modulations of the antioxidant gene expression system has been shown to regulate microglia activation states (Shih et al., 2006), so what we see might be a regulatory feature enabling astrocytes to fine tune the transitions between microglia homeostatic and inflammatory states. Taken together, the data suggest activation of two different transport processes in astrocytes. The expression of VRAC might facilitate glutamine release by astrocytes which can be taken up by neurons and converted to glutamate or GABA (McKenna and Ferreira, 2016; Waagepetersen et al., 2003). This could be a supportive mechanism to maintain base level of neuronal signalling, which might be affected by the down-regulation of genes in the citric acid cycle in neurons and the accompanied loss of downstream products as neuropeptides (Figure 9). At the same time, the expression of transmembrane transporters might contribute to the swelling of astrocytes by the uptake of glutamate. Morphological changes in astrocytes are connected to appetite regulation by redirecting glutamate away from neurons and hence reducing glutamatergic signalling (Delgado, 2013). This might be a fundamental factor for sickness related weight loss and gain, categorizing the here identified 7-dpi astrocyte population as regulating neuroendocrine processes. Further studies are needed to understand and confirm the functional identity of the 7 dpi subclusters.

In summary, the data presented here shows how the cellular state of non-neuronal and neuronal cell populations in the hypothalamus change during peripheral IAV. Our data suggests the existence of tight co-regulation where one cell type affects the state of another (Figure 9). This connectivity on the one hand potentially ensures protection of the tissue by surveillant microglia but also counterbalances the microglia activity by astrocytic regulation of their antioxidant machinery. In addition, we found evidence that non-neuronal cell types like astrocytes and oligodendrocytes are potentially supporting neurons with their metabolic and signalling integrity, by enabling more nutrient transport as well as neuropeptide supply. Altogether, our study provides the first evidence of complex hypothalamic gene expression changes at the single cell level during a peripheral viral infection.

Raw data filtered count matrices and metadata are available on GEO under accession number: GSE226098.

1. 1. Kornum BR

   (2023) NCBI Gene Expression Omnibus

   ID GSE226098. Molecular consequences of peripheral influenza a infection on cell populations in the murine hypothalamus.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE226098

1. 1. Ahmed SS
   2. Schur PH
   3. MacDonald NE
   4. Steinman L

   (2014) Narcolepsy, 2009 A(H1N1) pandemic influenza, and pandemic influenza vaccinations: what is known and unknown about the neurological disorder, the role for autoimmunity, and vaccine adjuvants

   Journal of Autoimmunity 50:1–11.

   https://doi.org/10.1016/j.jaut.2014.01.033

   • PubMed
   • Google Scholar
2. Software

   1. Alexa A

   (2022)

   Enrichment analysis for gene Ontology, version R package version 2.48.0

   topGO.
3. 1. Avey D
   2. Sankararaman S
   3. Yim AKY
   4. Barve R
   5. Milbrandt J
   6. Mitra RD

   (2018) Single-Cell RNA-Seq Uncovers a Robust Transcriptional Response to Morphine by Glia

   Cell Reports 24:3619–3629.

   https://doi.org/10.1016/j.celrep.2018.08.080

   • PubMed
   • Google Scholar
4. 1. Balzani E
   2. Lassi G
   3. Maggi S
   4. Sethi S
   5. Parsons MJ
   6. Simon M
   7. Nolan PM
   8. Tucci V

   (2016) The Zfhx3-Mediated Axis Regulates Sleep and Interval Timing in Mice

   Cell Reports 16:615–621.

   https://doi.org/10.1016/j.celrep.2016.06.017

   • PubMed
   • Google Scholar
5. 1. Berthoud HR
   2. Münzberg H

   (2011) The lateral hypothalamus as integrator of metabolic and environmental needs: from electrical self-stimulation to opto-genetics

   Physiology & Behavior 104:29–39.

   https://doi.org/10.1016/j.physbeh.2011.04.051

   • PubMed
   • Google Scholar
6. 1. Bonnavion P
   2. Mickelsen LE
   3. Fujita A
   4. de Lecea L
   5. Jackson AC

   (2016) Hubs and spokes of the lateral hypothalamus: cell types, circuits and behaviour

   The Journal of Physiology 594:6443–6462.

   https://doi.org/10.1113/JP271946

   • PubMed
   • Google Scholar
7. 1. Brancaccio M
   2. Patton AP
   3. Chesham JE
   4. Maywood ES
   5. Hastings MH

   (2017) Astrocytes control circadian timekeeping in the suprachiasmatic nucleus via glutamatergic signaling

   Neuron 93:1420–1435.

   https://doi.org/10.1016/j.neuron.2017.02.030

   • PubMed
   • Google Scholar
8. 1. Brown JA
   2. Woodworth HL
   3. Leinninger GM

   (2015) To ingest or rest? Specialized roles of lateral hypothalamic area neurons in coordinating energy balance

   Frontiers in Systems Neuroscience 9:9.

   https://doi.org/10.3389/fnsys.2015.00009

   • PubMed
   • Google Scholar
9. 1. Campbell JN
   2. Macosko EZ
   3. Fenselau H
   4. Pers TH
   5. Lyubetskaya A
   6. Tenen D
   7. Goldman M
   8. Verstegen AMJ
   9. Resch JM
   10. McCarroll SA
   11. Rosen ED
   12. Lowell BB
   13. Tsai LT

   (2017) A molecular census of arcuate hypothalamus and median eminence cell types

   Nature Neuroscience 20:484–496.

   https://doi.org/10.1038/nn.4495

   • PubMed
   • Google Scholar
10. 1. Chen R
    2. Wu X
    3. Jiang L
    4. Zhang Y

    (2017) Single-Cell RNA-Seq Reveals Hypothalamic Cell Diversity

    Cell Reports 18:3227–3241.

    https://doi.org/10.1016/j.celrep.2017.03.004

    • PubMed
    • Google Scholar
11. 1. Choi K
    2. Lee J
    3. Kang HJ

    (2021) Myelination defects in the medial prefrontal cortex of Fkbp5 knockout mice

    FASEB Journal 35:e21297.

    https://doi.org/10.1096/fj.202001883R

    • PubMed
    • Google Scholar
12. 1. Dantzer R
    2. Kelley KW

    (2007) Twenty years of research on cytokine-induced sickness behavior

    Brain, Behavior, and Immunity 21:153–160.

    https://doi.org/10.1016/j.bbi.2006.09.006

    • PubMed
    • Google Scholar
13. 1. Delgado TC

    (2013) Glutamate and GABA in Appetite Regulation

    Frontiers in Endocrinology 4:103.

    https://doi.org/10.3389/fendo.2013.00103

    • PubMed
    • Google Scholar
14. 1. Dusedau HP
    2. Steffen J
    3. Figueiredo CA
    4. Boehme JD
    5. Schultz K
    6. Erck C
    7. Korte M
    8. Faber-Zuschratter H
    9. Smalla KH
    10. Dieterich D
    11. Kroger A
    12. Bruder D
    13. Dunay IR

    (2021) Influenza A Virus (H1N1)

    Infection Induces Microglial Activation and Temporal Dysbalance in Glutamatergic Synaptic Transmission. mBio 12:e0177621.

    https://doi.org/10.1128/mBio.01776-21

    • Google Scholar
15. 1. Ekstrand JJ

    (2012) Neurologic complications of influenza

    Seminars in Pediatric Neurology 19:96–100.

    https://doi.org/10.1016/j.spen.2012.02.004

    • PubMed
    • Google Scholar
16. 1. Fakhoury M
    2. Salman I
    3. Najjar W
    4. Merhej G
    5. Lawand N

    (2020) The lateral hypothalamus: an uncharted territory for processing peripheral neurogenic inflammation

    Frontiers in Neuroscience 14:101.

    https://doi.org/10.3389/fnins.2020.00101

    • PubMed
    • Google Scholar
17. 1. Franco R
    2. Fernández-Suárez D

    (2015) Alternatively activated microglia and macrophages in the central nervous system

    Progress in Neurobiology 131:65–86.

    https://doi.org/10.1016/j.pneurobio.2015.05.003

    • PubMed
    • Google Scholar
18. 1. Goenka A
    2. Michael BD
    3. Ledger E
    4. Hart IJ
    5. Absoud M
    6. Chow G
    7. Lilleker J
    8. Lunn M
    9. McKee D
    10. Peake D
    11. Pysden K
    12. Roberts M
    13. Carrol ED
    14. Lim M
    15. Avula S
    16. Solomon T
    17. Kneen R

    (2014) Neurological manifestations of influenza infection in children and adults: results of a National British Surveillance Study

    Clinical Infectious Diseases 58:775–784.

    https://doi.org/10.1093/cid/cit922

    • PubMed
    • Google Scholar
19. 1. Gooley JJ
    2. Schomer A
    3. Saper CB

    (2006) The dorsomedial hypothalamic nucleus is critical for the expression of food-entrainable circadian rhythms

    Nature Neuroscience 9:398–407.

    https://doi.org/10.1038/nn1651

    • PubMed
    • Google Scholar
20. 1. Gu Y
    2. Shimada T
    3. Yasui Y
    4. Tada Y
    5. Kaku M
    6. Okabe N
    7. Sambhara S

    (2013) National surveillance of influenza-associated encephalopathy in Japan over six years, before and during the 2009-2010 influenza pandemic

    PLOS ONE 8:e54786.

    https://doi.org/10.1371/journal.pone.0054786

    • PubMed
    • Google Scholar
21. 1. Haenisch M
    2. Treuting PM
    3. Brabb T
    4. Goldstein AS
    5. Berkseth K
    6. Amory JK
    7. Paik J

    (2018) Pharmacological inhibition of ALDH1A enzymes suppresses weight gain in a mouse model of diet-induced obesity

    Obesity Research & Clinical Practice 12:93–101.

    https://doi.org/10.1016/j.orcp.2017.08.003

    • PubMed
    • Google Scholar
22. 1. Hähle A
    2. Merz S
    3. Meyners C
    4. Hausch F

    (2019) The Many Faces of FKBP51

    Biomolecules 9:35.

    https://doi.org/10.3390/biom9010035

    • PubMed
    • Google Scholar
23. 1. Han F
    2. Lin L
    3. Warby SC
    4. Faraco J
    5. Li J
    6. Dong SX
    7. An P
    8. Zhao L
    9. Wang LH
    10. Li QY
    11. Yan H
    12. Gao ZC
    13. Yuan Y
    14. Strohl KP
    15. Mignot E

    (2011) Narcolepsy onset is seasonal and increased following the 2009 H1N1 pandemic in China

    Annals of Neurology 70:410–417.

    https://doi.org/10.1002/ana.22587

    • PubMed
    • Google Scholar
24. 1. Hao Y
    2. Hao S
    3. Andersen-Nissen E
    4. Mauck WM

    (2021) Integrated analysis of multimodal single-cell data

    Cell 184:e3529.

    https://doi.org/10.1016/j.cell.2021.04.048

    • Google Scholar
25. 1. Hatakeyama D
    2. Shoji M
    3. Yamayoshi S
    4. Yoh R
    5. Ohmi N
    6. Takenaka S
    7. Saitoh A
    8. Arakaki Y
    9. Masuda A
    10. Komatsu T
    11. Nagano R
    12. Nakano M
    13. Noda T
    14. Kawaoka Y
    15. Kuzuhara T

    (2018) Influenza A virus nucleoprotein is acetylated by histone acetyltransferases PCAF and GCN5

    The Journal of Biological Chemistry 293:7126–7138.

    https://doi.org/10.1074/jbc.RA117.001683

    • PubMed
    • Google Scholar
26. 1. Hermann G
    2. Tovar CA
    3. Beck FM
    4. Sheridan JF

    (1994) Kinetics of glucocorticoid response to restraint stress and/or experimental influenza viral infection in two inbred strains of mice

    Journal of Neuroimmunology 49:25–33.

    https://doi.org/10.1016/0165-5728(94)90177-5

    • PubMed
    • Google Scholar
27. 1. Hinds LR
    2. Chun LE
    3. Woodruff ER
    4. Christensen JA
    5. Hartsock MJ
    6. Spencer RL

    (2017) Dynamic glucocorticoid-dependent regulation of Sgk1 expression in oligodendrocytes of adult male rat brain by acute stress and time of day

    PLOS ONE 12:e0175075.

    https://doi.org/10.1371/journal.pone.0175075

    • PubMed
    • Google Scholar
28. 1. Hochgerner H
    2. Zeisel A
    3. Lönnerberg P
    4. Linnarsson S

    (2018) Conserved properties of dentate gyrus neurogenesis across postnatal development revealed by single-cell RNA sequencing

    Nature Neuroscience 21:290–299.

    https://doi.org/10.1038/s41593-017-0056-2

    • PubMed
    • Google Scholar
29. 1. Hodgson NR
    2. Bohnet SG
    3. Majde JA
    4. Krueger JM

    (2012) Influenza virus pathophysiology and brain invasion in mice with functional and dysfunctional Mx1 genes

    Brain, Behavior, and Immunity 26:83–89.

    https://doi.org/10.1016/j.bbi.2011.07.238

    • PubMed
    • Google Scholar
30. 1. Hoffman LA
    2. Vilensky JA

    (2017) Encephalitis lethargica: 100 years after the epidemic

    Brain 140:2246–2251.

    https://doi.org/10.1093/brain/awx177

    • PubMed
    • Google Scholar
31. 1. Hosseini S
    2. Wilk E
    3. Michaelsen-Preusse K
    4. Gerhauser I
    5. Baumgärtner W
    6. Geffers R
    7. Schughart K
    8. Korte M

    (2018) Long-term neuroinflammation induced by influenza a virus infection and the impact on hippocampal neuron morphology and function

    The Journal of Neuroscience 38:3060–3080.

    https://doi.org/10.1523/JNEUROSCI.1740-17.2018

    • PubMed
    • Google Scholar
32. 1. Hui EK
    2. Nayak DP

    (2001) Role of ATP in influenza virus budding

    Virology 290:329–341.

    https://doi.org/10.1006/viro.2001.1181

    • PubMed
    • Google Scholar
33. 1. Jin S
    2. Kim JG
    3. Park JW
    4. Koch M
    5. Horvath TL
    6. Lee BJ

    (2016) Hypothalamic TLR2 triggers sickness behavior via a microglia-neuronal axis

    Scientific Reports 6:29424.

    https://doi.org/10.1038/srep29424

    • PubMed
    • Google Scholar
34. 1. Jurga AM
    2. Paleczna M
    3. Kuter KZ

    (2020) Overview of general and discriminating markers of differential microglia phenotypes

    Frontiers in Cellular Neuroscience 14:198.

    https://doi.org/10.3389/fncel.2020.00198

    • PubMed
    • Google Scholar
35. 1. Jurgens HA
    2. Amancherla K
    3. Johnson RW

    (2012) Influenza infection induces neuroinflammation, alters hippocampal neuron morphology, and impairs cognition in adult mice

    The Journal of Neuroscience 32:3958–3968.

    https://doi.org/10.1523/JNEUROSCI.6389-11.2012

    • PubMed
    • Google Scholar
36. 1. Kang D
    2. Kim HR
    3. Kim KK
    4. Kim DH
    5. Jeong B
    6. Jin S
    7. Park JW
    8. Seong JY
    9. Lee BJ

    (2020) Brain-specific chemokine FAM19A5 induces hypothalamic inflammation

    Biochemical and Biophysical Research Communications 523:829–834.

    https://doi.org/10.1016/j.bbrc.2019.12.119

    • PubMed
    • Google Scholar
37. 1. Kawai T
    2. Akira S

    (2011) Toll-like receptors and their crosstalk with other innate receptors in infection and immunity

    Immunity 34:637–650.

    https://doi.org/10.1016/j.immuni.2011.05.006

    • PubMed
    • Google Scholar
38. 1. Kostin A
    2. Alam MA
    3. McGinty D
    4. Alam MN

    (2021) Adult hypothalamic neurogenesis and sleep-wake dysfunction in aging

    Sleep 44:zsaa173.

    https://doi.org/10.1093/sleep/zsaa173

    • PubMed
    • Google Scholar
39. 1. Kwak H
    2. Koh W
    3. Kim S
    4. Song K
    5. Shin JI
    6. Lee JM
    7. Lee EH
    8. Bae JY
    9. Ha GE
    10. Oh JE
    11. Park YM
    12. Kim S
    13. Feng J
    14. Lee SE
    15. Choi JW
    16. Kim KH
    17. Kim YS
    18. Woo J
    19. Lee D
    20. Son T
    21. Kwon SW
    22. Park KD
    23. Yoon BE
    24. Lee J
    25. Li Y
    26. Lee H
    27. Bae YC
    28. Lee CJ
    29. Cheong E

    (2020) Astrocytes control sensory acuity via tonic inhibition in the thalamus

    Neuron 108:691–706.

    https://doi.org/10.1016/j.neuron.2020.08.013

    • PubMed
    • Google Scholar
40. 1. Lee Y
    2. Morrison BM
    3. Li Y
    4. Lengacher S
    5. Farah MH
    6. Hoffman PN
    7. Liu Y
    8. Tsingalia A
    9. Jin L
    10. Zhang PW
    11. Pellerin L
    12. Magistretti PJ
    13. Rothstein JD

    (2012) Oligodendroglia metabolically support axons and contribute to neurodegeneration

    Nature 487:443–448.

    https://doi.org/10.1038/nature11314

    • PubMed
    • Google Scholar
41. 1. Liu H
    2. Sun Y
    3. Zhang Q
    4. Jin W
    5. Gordon RE
    6. Zhang Y
    7. Wang J
    8. Sun C
    9. Wang ZJ
    10. Qi X
    11. Zhang J
    12. Huang B
    13. Gui Q
    14. Yuan H
    15. Chen L
    16. Ma X
    17. Fang C
    18. Liu YQ
    19. Yu X
    20. Feng S

    (2021) Pro-inflammatory and proliferative microglia drive progression of glioblastoma

    Cell Reports 36:109718.

    https://doi.org/10.1016/j.celrep.2021.109718

    • PubMed
    • Google Scholar
42. 1. Macias AE
    2. McElhaney JE
    3. Chaves SS
    4. Nealon J
    5. Nunes MC
    6. Samson SI
    7. Seet BT
    8. Weinke T
    9. Yu H

    (2021) The disease burden of influenza beyond respiratory illness

    Vaccine 39:A6–A14.

    https://doi.org/10.1016/j.vaccine.2020.09.048

    • PubMed
    • Google Scholar
43. 1. Majde JA
    2. Bohnet SG
    3. Ellis GA
    4. Churchill L
    5. Leyva-Grado V
    6. Wu M
    7. Szentirmai E
    8. Rehman A
    9. Krueger JM

    (2007) Detection of mouse-adapted human influenza virus in the olfactory bulbs of mice within hours after intranasal infection

    Journal of Neurovirology 13:399–409.

    https://doi.org/10.1080/13550280701427069

    • PubMed
    • Google Scholar
44. 1. Maurizi CP

    (2010) Influenza caused epidemic encephalitis (encephalitis lethargica): the circumstantial evidence and a challenge to the nonbelievers

    Medical Hypotheses 74:798–801.

    https://doi.org/10.1016/j.mehy.2009.12.012

    • PubMed
    • Google Scholar
45. 1. McCarthy DJ
    2. Campbell KR
    3. Lun ATL
    4. Wills QF

    (2017) Scater: pre-processing, quality control, normalization and visualization of single-cell RNA-seq data in R

    Bioinformatics 33:1179–1186.

    https://doi.org/10.1093/bioinformatics/btw777

    • PubMed
    • Google Scholar
46. 1. McKenna MC
    2. Ferreira GC

    (2016) Enzyme complexes important for the glutamate-glutamine cycle

    Advances in Neurobiology 13:59–98.

    https://doi.org/10.1007/978-3-319-45096-4_4

    • PubMed
    • Google Scholar
47. 1. Mederos S
    2. Perea G

    (2019) GABAergic-astrocyte signaling: A refinement of inhibitory brain networks

    Glia 67:1842–1851.

    https://doi.org/10.1002/glia.23644

    • PubMed
    • Google Scholar
48. 1. Meng J
    2. Ding T
    3. Chen Y
    4. Long T
    5. Xu Q
    6. Lian W
    7. Liu W

    (2021) LncRNA-Meg3 promotes Nlrp3-mediated microglial inflammation by targeting miR-7a-5p

    International Immunopharmacology 90:107141.

    https://doi.org/10.1016/j.intimp.2020.107141

    • PubMed
    • Google Scholar
49. 1. Mickelsen LE
    2. Bolisetty M
    3. Chimileski BR
    4. Fujita A
    5. Beltrami EJ
    6. Costanzo JT
    7. Naparstek JR
    8. Robson P
    9. Jackson AC

    (2019) Single-cell transcriptomic analysis of the lateral hypothalamic area reveals molecularly distinct populations of inhibitory and excitatory neurons

    Nature Neuroscience 22:642–656.

    https://doi.org/10.1038/s41593-019-0349-8

    • PubMed
    • Google Scholar
50. 1. Mickelsen LE
    2. Flynn WF
    3. Springer K
    4. Wilson L
    5. Beltrami EJ
    6. Bolisetty M
    7. Robson P
    8. Jackson AC

    (2020) Cellular taxonomy and spatial organization of the murine ventral posterior hypothalamus

    eLife 9:e58901.

    https://doi.org/10.7554/eLife.58901

    • PubMed
    • Google Scholar
51. 1. Miyata S
    2. Koyama Y
    3. Takemoto K
    4. Yoshikawa K
    5. Ishikawa T
    6. Taniguchi M
    7. Inoue K
    8. Aoki M
    9. Hori O
    10. Katayama T
    11. Tohyama M

    (2011) Plasma corticosterone activates SGK1 and induces morphological changes in oligodendrocytes in corpus callosum

    PLOS ONE 6:e19859.

    https://doi.org/10.1371/journal.pone.0019859

    • PubMed
    • Google Scholar
52. 1. Miyata S
    2. Hattori T
    3. Shimizu S
    4. Ito A
    5. Tohyama M

    (2015) Disturbance of oligodendrocyte function plays a key role in the pathogenesis of schizophrenia and major depressive disorder

    BioMed Research International 2015:492367.

    https://doi.org/10.1155/2015/492367

    • PubMed
    • Google Scholar
53. 1. Moehle MS
    2. West AB

    (2015) M1 and M2 immune activation in Parkinson’s Disease: Foe and ally?

    Neuroscience 302:59–73.

    https://doi.org/10.1016/j.neuroscience.2014.11.018

    • PubMed
    • Google Scholar
54. 1. Moffitt JR
    2. Bambah-Mukku D
    3. Eichhorn SW
    4. Vaughn E
    5. Shekhar K
    6. Perez JD
    7. Rubinstein ND
    8. Hao J
    9. Regev A
    10. Dulac C
    11. Zhuang X

    (2018) Molecular, spatial, and functional single-cell profiling of the hypothalamic preoptic region

    Science 362:eaau5324.

    https://doi.org/10.1126/science.aau5324

    • PubMed
    • Google Scholar
55. 1. Moore S
    2. Meschkat M
    3. Ruhwedel T
    4. Trevisiol A
    5. Tzvetanova ID
    6. Battefeld A
    7. Kusch K
    8. Kole MHP
    9. Strenzke N
    10. Möbius W
    11. de Hoz L
    12. Nave KA

    (2020) A role of oligodendrocytes in information processing

    Nature Communications 11:5497.

    https://doi.org/10.1038/s41467-020-19152-7

    • PubMed
    • Google Scholar
56. 1. Nonogaki K
    2. Ohashi-Nozue K
    3. Oka Y

    (2006) Induction of hypothalamic serum- and glucocorticoid-induced protein kinase-1 gene expression and its relation to plasma des-acyl ghrelin in energy homeostasis in mice

    Biochemical and Biophysical Research Communications 344:696–699.

    https://doi.org/10.1016/j.bbrc.2006.03.196

    • PubMed
    • Google Scholar
57. 1. Oesch LT
    2. Adamantidis AR

    (2021) Sleep and metabolism: implication of lateral hypothalamic neurons

    Frontiers of Neurology and Neuroscience 45:75–90.

    https://doi.org/10.1159/000514966

    • PubMed
    • Google Scholar
58. 1. Osei-Owusu J
    2. Yang J
    3. Vitery MDC
    4. Qiu Z

    (2018) Molecular Biology and Physiology of Volume-Regulated Anion Channel (VRAC)

    Current Topics in Membranes 81:177–203.

    https://doi.org/10.1016/bs.ctm.2018.07.005

    • PubMed
    • Google Scholar
59. 1. Osterhout JA
    2. Kapoor V
    3. Eichhorn SW
    4. Vaughn E
    5. Moore JD
    6. Liu D
    7. Lee D
    8. DeNardo LA
    9. Luo L
    10. Zhuang X
    11. Dulac C

    (2022) A preoptic neuronal population controls fever and appetite during sickness

    Nature 606:937–944.

    https://doi.org/10.1038/s41586-022-04793-z

    • PubMed
    • Google Scholar
60. 1. Pagani F
    2. Tratta E
    3. Dell’Era P
    4. Cominelli M
    5. Poliani PL

    (2021) EBF1 is expressed in pericytes and contributes to pericyte cell commitment

    Histochemistry and Cell Biology 156:333–347.

    https://doi.org/10.1007/s00418-021-02015-7

    • PubMed
    • Google Scholar
61. 1. Paget J
    2. Spreeuwenberg P
    3. Charu V
    4. Taylor RJ
    5. Iuliano AD
    6. Bresee J
    7. Simonsen L
    8. Viboud C
    9. Global Seasonal Influenza-associated Mortality Collaborator Network and GLaMOR Collaborating Teams*

    (2019) Global mortality associated with seasonal influenza epidemics: New burden estimates and predictors from the GLaMOR Project

    Journal of Global Health 9:020421.

    https://doi.org/10.7189/jogh.09.020421

    • PubMed
    • Google Scholar
62. 1. Parsons MJ
    2. Brancaccio M
    3. Sethi S
    4. Maywood ES
    5. Satija R
    6. Edwards JK
    7. Jagannath A
    8. Couch Y
    9. Finelli MJ
    10. Smyllie NJ
    11. Esapa C
    12. Butler R
    13. Barnard AR
    14. Chesham JE
    15. Saito S
    16. Joynson G
    17. Wells S
    18. Foster RG
    19. Oliver PL
    20. Simon MM
    21. Mallon AM
    22. Hastings MH
    23. Nolan PM

    (2015) The Regulatory Factor ZFHX3 Modifies Circadian Function in SCN via an AT Motif-Driven Axis

    Cell 162:607–621.

    https://doi.org/10.1016/j.cell.2015.06.060

    • PubMed
    • Google Scholar
63. Preprint

    1. Petukhov V
    2. Igolkina A
    3. Rydbirk R
    4. Mei S
    5. Christoffersen L
    6. Khodosevich K
    7. Kharchenko PV

    (2022) Case-Control Analysis of Single-Cell RNA-Seq Studies

    bioRxiv.

    https://doi.org/10.1101/2022.03.15.484475

    • Google Scholar
64. 1. Rodríguez-Gómez JA
    2. Kavanagh E
    3. Engskog-Vlachos P
    4. Engskog MKR
    5. Herrera AJ
    6. Espinosa-Oliva AM
    7. Joseph B
    8. Hajji N
    9. Venero JL
    10. Burguillos MA

    (2020) Microglia: Agents of the CNS Pro-Inflammatory Response

    Cells 9:1717.

    https://doi.org/10.3390/cells9071717

    • PubMed
    • Google Scholar
65. 1. Romanov RA
    2. Zeisel A
    3. Bakker J
    4. Girach F
    5. Hellysaz A
    6. Tomer R
    7. Alpár A
    8. Mulder J
    9. Clotman F
    10. Keimpema E
    11. Hsueh B
    12. Crow AK
    13. Martens H
    14. Schwindling C
    15. Calvigioni D
    16. Bains JS
    17. Máté Z
    18. Szabó G
    19. Yanagawa Y
    20. Zhang MD
    21. Rendeiro A
    22. Farlik M
    23. Uhlén M
    24. Wulff P
    25. Bock C
    26. Broberger C
    27. Deisseroth K
    28. Hökfelt T
    29. Linnarsson S
    30. Horvath TL
    31. Harkany T

    (2017) Molecular interrogation of hypothalamic organization reveals distinct dopamine neuronal subtypes

    Nature Neuroscience 20:176–188.

    https://doi.org/10.1038/nn.4462

    • PubMed
    • Google Scholar
66. 1. Saab AS
    2. Tzvetavona ID
    3. Trevisiol A
    4. Baltan S
    5. Dibaj P
    6. Kusch K
    7. Möbius W
    8. Goetze B
    9. Jahn HM
    10. Huang W
    11. Steffens H
    12. Schomburg ED
    13. Pérez-Samartín A
    14. Pérez-Cerdá F
    15. Bakhtiari D
    16. Matute C
    17. Löwel S
    18. Griesinger C
    19. Hirrlinger J
    20. Kirchhoff F
    21. Nave KA

    (2016) Oligodendroglial NMDA Receptors Regulate Glucose Import and Axonal Energy Metabolism

    Neuron 91:119–132.

    https://doi.org/10.1016/j.neuron.2016.05.016

    • PubMed
    • Google Scholar
67. 1. Sadasivan S
    2. Zanin M
    3. O’Brien K
    4. Schultz-Cherry S
    5. Smeyne RJ

    (2015) Induction of microglia activation after infection with the non-neurotropic A/CA/04/2009 H1N1 influenza virus

    PLOS ONE 10:e0124047.

    https://doi.org/10.1371/journal.pone.0124047

    • PubMed
    • Google Scholar
68. 1. Sánchez-García FJ
    2. Pérez-Hernández CA
    3. Rodríguez-Murillo M
    4. Moreno-Altamirano MMB

    (2021) The role of tricarboxylic acid cycle metabolites in viral infections

    Frontiers in Cellular and Infection Microbiology 11:725043.

    https://doi.org/10.3389/fcimb.2021.725043

    • PubMed
    • Google Scholar
69. 1. Senatorov VV
    2. Friedman AR
    3. Milikovsky DZ
    4. Ofer J
    5. Saar-Ashkenazy R
    6. Charbash A
    7. Jahan N
    8. Chin G
    9. Mihaly E
    10. Lin JM
    11. Ramsay HJ
    12. Moghbel A
    13. Preininger MK
    14. Eddings CR
    15. Harrison HV
    16. Patel R
    17. Shen Y
    18. Ghanim H
    19. Sheng H
    20. Veksler R
    21. Sudmant PH
    22. Becker A
    23. Hart B
    24. Rogawski MA
    25. Dillin A
    26. Friedman A
    27. Kaufer D

    (2019) Blood-brain barrier dysfunction in aging induces hyperactivation of TGFβ signaling and chronic yet reversible neural dysfunction

    Science Translational Medicine 11:eaaw8283.

    https://doi.org/10.1126/scitranslmed.aaw8283

    • PubMed
    • Google Scholar
70. 1. Shih AY
    2. Fernandes HB
    3. Choi FY
    4. Kozoriz MG
    5. Liu Y
    6. Li P
    7. Cowan CM
    8. Klegeris A

    (2006) Policing the police: astrocytes modulate microglial activation

    The Journal of Neuroscience 26:3887–3888.

    https://doi.org/10.1523/JNEUROSCI.0936-06.2006

    • PubMed
    • Google Scholar
71. 1. Silverman MN
    2. Pearce BD
    3. Biron CA
    4. Miller AH

    (2005) Immune modulation of the hypothalamic-pituitary-adrenal (HPA) axis during viral infection

    Viral Immunology 18:41–78.

    https://doi.org/10.1089/vim.2005.18.41

    • PubMed
    • Google Scholar
72. 1. Simonsen L
    2. Spreeuwenberg P
    3. Lustig R
    4. Taylor RJ
    5. Fleming DM
    6. Kroneman M
    7. Van Kerkhove MD
    8. Mounts AW
    9. Paget WJ
    10. GLaMOR Collaborating Teams

    (2013) Global mortality estimates for the 2009 Influenza Pandemic from the GLaMOR project: a modeling study

    PLOS Medicine 10:e1001558.

    https://doi.org/10.1371/journal.pmed.1001558

    • PubMed
    • Google Scholar
73. Website

    1. Srivastava AHY

    (2020) Feature Barcoding based Single-Cell Quantification with alevin

    Accessed April 20, 2020.

    https://combine-lab.github.io/alevin-tutorial/2020/alevin-features/
74. 1. Steuernagel L
    2. Lam BYH
    3. Klemm P
    4. Dowsett GKC
    5. Bauder CA
    6. Tadross JA
    7. Hitschfeld TS
    8. Del Rio Martin A
    9. Chen W
    10. de Solis AJ
    11. Fenselau H
    12. Davidsen P
    13. Cimino I
    14. Kohnke SN
    15. Rimmington D
    16. Coll AP
    17. Beyer A
    18. Yeo GSH
    19. Brüning JC

    (2022) HypoMap-a unified single-cell gene expression atlas of the murine hypothalamus

    Nature Metabolism 4:1402–1419.

    https://doi.org/10.1038/s42255-022-00657-y

    • PubMed
    • Google Scholar
75. 1. Stuber GD
    2. Wise RA

    (2016) Lateral hypothalamic circuits for feeding and reward

    Nature Neuroscience 19:198–205.

    https://doi.org/10.1038/nn.4220

    • PubMed
    • Google Scholar
76. 1. Surana P
    2. Tang S
    3. McDougall M
    4. Tong CYW
    5. Menson E
    6. Lim M

    (2011) Neurological complications of pandemic influenza A H1N1 2009 infection: European case series and review

    European Journal of Pediatrics 170:1007–1015.

    https://doi.org/10.1007/s00431-010-1392-3

    • PubMed
    • Google Scholar
77. 1. Taubenberger JK
    2. Morens DM

    (2008) The pathology of influenza virus infections

    Annual Review of Pathology 3:499–522.

    https://doi.org/10.1146/annurev.pathmechdis.3.121806.154316

    • PubMed
    • Google Scholar
78. 1. Thrupp N
    2. Sala Frigerio C
    3. Wolfs L
    4. Skene NG
    5. Fattorelli N
    6. Poovathingal S
    7. Fourne Y
    8. Matthews PM
    9. Theys T
    10. Mancuso R
    11. de Strooper B
    12. Fiers M

    (2020) Single-Nucleus RNA-Seq Is Not Suitable for Detection of Microglial Activation Genes in Humans

    Cell Reports 32:108189.

    https://doi.org/10.1016/j.celrep.2020.108189

    • PubMed
    • Google Scholar
79. 1. Timper K
    2. Brüning JC

    (2017) Hypothalamic circuits regulating appetite and energy homeostasis: pathways to obesity

    Disease Models & Mechanisms 10:679–689.

    https://doi.org/10.1242/dmm.026609

    • PubMed
    • Google Scholar
80. 1. Turcot V
    2. Lu Y
    3. Highland HM
    4. Schurmann C
    5. Justice AE
    6. Fine RS
    7. Bradfield JP
    8. Esko T
    9. Giri A
    10. Graff M
    11. Guo X
    12. Hendricks AE
    13. Karaderi T
    14. Lempradl A
    15. Locke AE
    16. Mahajan A
    17. Marouli E
    18. Sivapalaratnam S
    19. Young KL
    20. Alfred T
    21. Feitosa MF
    22. Masca NGD
    23. Manning AK
    24. Medina-Gomez C
    25. Mudgal P
    26. Ng MCY
    27. Reiner AP
    28. Vedantam S
    29. Willems SM
    30. Winkler TW
    31. Abecasis G
    32. Aben KK
    33. Alam DS
    34. Alharthi SE
    35. Allison M
    36. Amouyel P
    37. Asselbergs FW
    38. Auer PL
    39. Balkau B
    40. Bang LE
    41. Barroso I
    42. Bastarache L
    43. Benn M
    44. Bergmann S
    45. Bielak LF
    46. Blüher M
    47. Boehnke M
    48. Boeing H
    49. Boerwinkle E
    50. Böger CA
    51. Bork-Jensen J
    52. Bots ML
    53. Bottinger EP
    54. Bowden DW
    55. Brandslund I
    56. Breen G
    57. Brilliant MH
    58. Broer L
    59. Brumat M
    60. Burt AA
    61. Butterworth AS
    62. Campbell PT
    63. Cappellani S
    64. Carey DJ
    65. Catamo E
    66. Caulfield MJ
    67. Chambers JC
    68. Chasman DI
    69. Chen YDI
    70. Chowdhury R
    71. Christensen C
    72. Chu AY
    73. Cocca M
    74. Collins FS
    75. Cook JP
    76. Corley J
    77. Corominas Galbany J
    78. Cox AJ
    79. Crosslin DS
    80. Cuellar-Partida G
    81. D’Eustacchio A
    82. Danesh J
    83. Davies G
    84. Bakker PIW
    85. Groot MCH
    86. Mutsert R
    87. Deary IJ
    88. Dedoussis G
    89. Demerath EW
    90. Heijer M
    91. Hollander AI
    92. Ruijter HM
    93. Dennis JG
    94. Denny JC
    95. Di Angelantonio E
    96. Drenos F
    97. Du M
    98. Dubé MP
    99. Dunning AM
    100. Easton DF
    101. Edwards TL
    102. Ellinghaus D
    103. Ellinor PT
    104. Elliott P
    105. Evangelou E
    106. Farmaki AE
    107. Farooqi IS
    108. Faul JD
    109. Fauser S
    110. Feng S
    111. Ferrannini E
    112. Ferrieres J
    113. Florez JC
    114. Ford I
    115. Fornage M
    116. Franco OH
    117. Franke A
    118. Franks PW
    119. Friedrich N
    120. Frikke-Schmidt R
    121. Galesloot TE
    122. Gan W
    123. Gandin I
    124. Gasparini P
    125. Gibson J
    126. Giedraitis V
    127. Gjesing AP
    128. Gordon-Larsen P
    129. Gorski M
    130. Grabe HJ
    131. Grant SFA
    132. Grarup N
    133. Griffiths HL
    134. Grove ML
    135. Gudnason V
    136. Gustafsson S
    137. Haessler J
    138. Hakonarson H
    139. Hammerschlag AR
    140. Hansen T
    141. Harris KM
    142. Harris TB
    143. Hattersley AT
    144. Have CT
    145. Hayward C
    146. He L
    147. Heard-Costa NL
    148. Heath AC
    149. Heid IM
    150. Helgeland Ø
    151. Hernesniemi J
    152. Hewitt AW
    153. Holmen OL
    154. Hovingh GK
    155. Howson JMM
    156. Hu Y
    157. Huang PL
    158. Huffman JE
    159. Ikram MA
    160. Ingelsson E
    161. Jackson AU
    162. Jansson JH
    163. Jarvik GP
    164. Jensen GB
    165. Jia Y
    166. Johansson S
    167. Jørgensen ME
    168. Jørgensen T
    169. Jukema JW
    170. Kahali B
    171. Kahn RS
    172. Kähönen M
    173. Kamstrup PR
    174. Kanoni S
    175. Kaprio J
    176. Karaleftheri M
    177. Kardia SLR
    178. Karpe F
    179. Kathiresan S
    180. Kee F
    181. Kiemeney LA
    182. Kim E
    183. Kitajima H
    184. Komulainen P
    185. Kooner JS
    186. Kooperberg C
    187. Korhonen T
    188. Kovacs P
    189. Kuivaniemi H
    190. Kutalik Z
    191. Kuulasmaa K
    192. Kuusisto J
    193. Laakso M
    194. Lakka TA
    195. Lamparter D
    196. Lange EM
    197. Lange LA
    198. Langenberg C
    199. Larson EB
    200. Lee NR
    201. Lehtimäki T
    202. Lewis CE
    203. Li H
    204. Li J
    205. Li-Gao R
    206. Lin H
    207. Lin KH
    208. Lin LA
    209. Lin X
    210. Lind L
    211. Lindström J
    212. Linneberg A
    213. Liu CT
    214. Liu DJ
    215. Liu Y
    216. Lo KS
    217. Lophatananon A
    218. Lotery AJ
    219. Loukola A
    220. Luan J
    221. Lubitz SA
    222. Lyytikäinen LP
    223. Männistö S
    224. Marenne G
    225. Mazul AL
    226. McCarthy MI
    227. McKean-Cowdin R
    228. Medland SE
    229. Meidtner K
    230. Milani L
    231. Mistry V
    232. Mitchell P
    233. Mohlke KL
    234. Moilanen L
    235. Moitry M
    236. Montgomery GW
    237. Mook-Kanamori DO
    238. Moore C
    239. Mori TA
    240. Morris AD
    241. Morris AP
    242. Müller-Nurasyid M
    243. Munroe PB
    244. Nalls MA
    245. Narisu N
    246. Nelson CP
    247. Neville M
    248. Nielsen SF
    249. Nikus K
    250. Njølstad PR
    251. Nordestgaard BG
    252. Nyholt DR
    253. O’Connel JR
    254. O’Donoghue ML
    255. Olde Loohuis LM
    256. Ophoff RA
    257. Owen KR
    258. Packard CJ
    259. Padmanabhan S
    260. Palmer CNA
    261. Palmer ND
    262. Pasterkamp G
    263. Patel AP
    264. Pattie A
    265. Pedersen O
    266. Peissig PL
    267. Peloso GM
    268. Pennell CE
    269. Perola M
    270. Perry JA
    271. Perry JRB
    272. Pers TH
    273. Person TN
    274. Peters A
    275. Petersen ERB
    276. Peyser PA
    277. Pirie A
    278. Polasek O
    279. Polderman TJ
    280. Puolijoki H
    281. Raitakari OT
    282. Rasheed A
    283. Rauramaa R
    284. Reilly DF
    285. Renström F
    286. Rheinberger M
    287. Ridker PM
    288. Rioux JD
    289. Rivas MA
    290. Roberts DJ
    291. Robertson NR
    292. Robino A
    293. Rolandsson O
    294. Rudan I
    295. Ruth KS
    296. Saleheen D
    297. Salomaa V
    298. Samani NJ
    299. Sapkota Y
    300. Sattar N
    301. Schoen RE
    302. Schreiner PJ
    303. Schulze MB
    304. Scott RA
    305. Segura-Lepe MP
    306. Shah SH
    307. Sheu WHH
    308. Sim X
    309. Slater AJ
    310. Small KS
    311. Smith AV
    312. Southam L
    313. Spector TD
    314. Speliotes EK
    315. Starr JM
    316. Stefansson K
    317. Steinthorsdottir V
    318. Stirrups KE
    319. Strauch K
    320. Stringham HM
    321. Stumvoll M
    322. Sun L
    323. Surendran P
    324. Swift AJ
    325. Tada H
    326. Tansey KE
    327. Tardif JC
    328. Taylor KD
    329. Teumer A
    330. Thompson DJ
    331. Thorleifsson G
    332. Thorsteinsdottir U
    333. Thuesen BH
    334. Tönjes A
    335. Tromp G
    336. Trompet S
    337. Tsafantakis E
    338. Tuomilehto J
    339. Tybjaerg-Hansen A
    340. Tyrer JP
    341. Uher R
    342. Uitterlinden AG
    343. Uusitupa M
    344. Laan SW
    345. Duijn CM
    346. Leeuwen N
    347. van Setten J
    348. Vanhala M
    349. Varbo A
    350. Varga TV
    351. Varma R
    352. Velez Edwards DR
    353. Vermeulen SH
    354. Veronesi G
    355. Vestergaard H
    356. Vitart V
    357. Vogt TF
    358. Völker U
    359. Vuckovic D
    360. Wagenknecht LE
    361. Walker M
    362. Wallentin L
    363. Wang F
    364. Wang CA
    365. Wang S
    366. Wang Y
    367. Ware EB
    368. Wareham NJ
    369. Warren HR
    370. Waterworth DM
    371. Wessel J
    372. White HD
    373. Willer CJ
    374. Wilson JG
    375. Witte DR
    376. Wood AR
    377. Wu Y
    378. Yaghootkar H
    379. Yao J
    380. Yao P
    381. Yerges-Armstrong LM
    382. Young R
    383. Zeggini E
    384. Zhan X
    385. Zhang W
    386. Zhao JH
    387. Zhao W
    388. Zhao W
    389. Zhou W
    390. Zondervan KT
    391. Rotter JI
    392. Pospisilik JA
    393. Rivadeneira F
    394. Borecki IB
    395. Deloukas P
    396. Frayling TM
    397. Lettre G
    398. North KE
    399. Lindgren CM
    400. Hirschhorn JN
    401. Loos RJF
    402. CHD Exome+ Consortium
    403. EPIC-CVD Consortium
    404. ExomeBP Consortium
    405. Global Lipids Genetic Consortium
    406. GoT2D Genes Consortium
    407. EPIC InterAct Consortium
    408. INTERVAL Study
    409. ReproGen Consortium
    410. T2D-Genes Consortium
    411. MAGIC Investigators
    412. Understanding Society Scientific Group

    (2018) Publisher Correction: Protein-altering variants associated with body mass index implicate pathways that control energy intake and expenditure in obesity

    Nature Genetics 50:766–767.

    https://doi.org/10.1038/s41588-018-0082-3

    • PubMed
    • Google Scholar
81. 1. Uddbäck IEM
    2. Steffensen MA
    3. Pedersen SR
    4. Nazerai L
    5. Thomsen AR
    6. Christensen JP

    (2016) PB1 as A potential target for increasing the breadth of T-cell mediated immunity to Influenza A

    Scientific Reports 6:35033.

    https://doi.org/10.1038/srep35033

    • PubMed
    • Google Scholar
82. 1. Waagepetersen HS
    2. Sonnewald U
    3. Schousboe A

    (2003) Compartmentation of glutamine, glutamate, and GABA metabolism in neurons and astrocytes: functional implications

    The Neuroscientist 9:398–403.

    https://doi.org/10.1177/1073858403254006

    • PubMed
    • Google Scholar
83. 1. Wiley CA
    2. Bhardwaj N
    3. Ross TM
    4. Bissel SJ

    (2015) Emerging Infections of CNS: Avian Influenza A Virus, Rift Valley Fever Virus and Human Parechovirus

    Brain Pathology 25:634–650.

    https://doi.org/10.1111/bpa.12281

    • PubMed
    • Google Scholar
84. 1. Yang J
    2. Vitery M
    3. Chen J
    4. Osei-Owusu J
    5. Chu J
    6. Qiu Z

    (2019) Glutamate-Releasing SWELL1 Channel in Astrocytes Modulates Synaptic Transmission and Promotes Brain Damage in Stroke

    Neuron 102:813–827.

    https://doi.org/10.1016/j.neuron.2019.03.029

    • Google Scholar
85. 1. Yousefvand S
    2. Hamidi F

    (2021) The role of ventromedial hypothalamus receptors in the central regulation of food intake

    International Journal of Peptide Research and Therapeutics 27:689–702.

    https://doi.org/10.1007/s10989-020-10120-9

    • Google Scholar
86. 1. Zeisel A
    2. Hochgerner H
    3. Lönnerberg P
    4. Johnsson A
    5. Memic F
    6. van der Zwan J
    7. Häring M
    8. Braun E
    9. Borm LE
    10. La Manno G
    11. Codeluppi S
    12. Furlan A
    13. Lee K
    14. Skene N
    15. Harris KD
    16. Hjerling-Leffler J
    17. Arenas E
    18. Ernfors P
    19. Marklund U
    20. Linnarsson S

    (2018) Molecular architecture of the mouse nervous system

    Cell 174:999–1014.

    https://doi.org/10.1016/j.cell.2018.06.021

    • PubMed
    • Google Scholar
87. 1. Zielinski MR
    2. Souza G
    3. Taishi P
    4. Bohnet SG
    5. Krueger JM

    (2013) Olfactory bulb and hypothalamic acute-phase responses to influenza virus: effects of immunization

    Neuroimmunomodulation 20:323–333.

    https://doi.org/10.1159/000351716

    • PubMed
    • Google Scholar

The following is the authors’ response to the current reviews.

Reviewer #1 (Recommendations For The Authors):

The revision and rebuttal have addressed all concerns raised in the initial review. Upon review of the revised figures, however, it is unclear why Figure 8C shows many significant DEGs in POMC neurons (which according to Figure 8b is the "GABA_24" cluster), whereas Figure 6A shows few to no DEGs in the GABA_24 cluster. Same for Pmch neurons/Glut_25, which seem to be missing from Figure 6A.

Answer: In order to capture changes in these smaller cell population we performed an additional DEG analysis with modified and less strict parameters (compared to the first main analysis). We mention the different parameters in the methods part of the revised manuscript (Differential gene expression analysis and case-control based expression shifts (Cacoa)).

The following is the authors’ response to the original reviews.

Reviewer #1 (Recommendations For The Authors):

Major issues

1. A key conclusion of this study is that neurons show longer lasting infection-related changes in gene expression than do non-neuronal cells, suggesting that neurons are more persistently affected, which could potentially underlie persistent effects of infection on behavior or physiology. However, the authors also report that over twice as many transcripts were captured in neurons than in non-neuronal cells, and that neurons and non-neurons were not equal in number. The number of transcripts and cells per cell type can affect the likelihood of detecting a differentially expressed gene when comparing cell types. Thus, the difference in infection related DEGs between non-neuronal cells and neurons may be due in part to differences in the numbers of transcripts and cells in each group. How would the number of infection related DEG's compare if the same number of transcripts were detected in neurons as in non-neuronal cells? In addition, is there any relationship between thenumber of infection related DEGs detected and the number of cells in the respective groups?

We performed an additional analysis, down sampling the transcripts per cells to similar numbers (~1600 transcripts/cell), showing a similar pattern as shown in the original calculation of DEGs. High downregulation of genes in GABAergic, Glutamatergic and Nonneuronal cells at 3 and 7 dpi, but long-lasting dis-regulation at 23 dpi only in the neuronal subtypes. The analysis results can be found in Supplementary figure12 and on page 11 in the results section.

1. The rationale for focusing on the LH and DMH is unclear. While these regions do play important roles in control of body weight and wakefulness, the authors do not report whether the cell types relevant to these functions are among those affected by infection. For instance, the authors mention HCRT and MCH neurons in the introduction but do not comment on whether these neurons show any significant changes after H1N1 infection in their analysis. Also, what about the POMC neurons or the Lepr+ DMH neurons? Knowing whether and how these body weight associated cell types are affected could help to connect the phenotypic (e.g., body weight) and molecular changes observed.

We have added an additional analysis of some well know hypothalamic subtypes. What is interesting is that the different neuronal subtypes respond to the infection differently. While most neurons show the strongest response at 3dpi, POMC+ neurons show consistent changes across all three time points. This could point to different neuronal subtypes paying different roles in the sickness response to the influenza infection. The new data has been added to Figure 8 together with new text in the result section and discussion (Page 17 & 20).

1. For discriminating neurons and non-neuronal cells based on their expression of neuronal marker genes, was this performed at the single-cell level or the cluster level? Similarly, was the discrimination of GABAergic and glutamatergic neurons done at the cell or cluster level?

The discrimination of the cell types was done on single cell level. This information has been added to the revised manuscript on page 25.

1. The authors mention that body weight did not change in some of the mice. Was there any difference in infection related DEGs between the mice that lost weight and the mice that didn't? Was there any correlation between the molecular and phenotypic (i.e., body weight) changes observed?

We agree that this could have been an interesting point to investigate, however, we can only say with certainty for 2 animals in the recovery group (23.7 and 23.8) that they didn’t lose weight (Supplement figure 2). In Figure 4A we show that overall the different time points group well together, with exception for animal 23.7 which seems to have a better overlap with 7 dpi, indicating that we possibly captured here a delayed disease response. However, to make any indepth analysis, we have to few animals without weight-loss.

5)The authors noted that the hypothalamic neurons continue to show infection-related changes in gene expression at 23dpi though body weight has returned to normal. In this H1N1 model, are there any persistent behavioral deficits at 23dpi that could be explained by the persistent changes in gene expression in DMH and LH neurons?

We did not test for long-lasting behavioural changes in these animals. Another study by Hosseini et al. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6596076/) focus on cognitive long term effects of viral infections. Even though they did not include the here used H1N1 model, they included the PR8 strain, but didn’t report any long lasting behavioural or cognitive changes. So far only cognitive deficits during the acute phase of the infection caused by the PR8 H1N1 model have been shown. This would be a very interesting follow up study to perform, but this, we believe, is out of scope for the current manuscript.

1. In Figure 1F, the 3dpi sample appears to differ from the other samples in terms of its neuron/non-neuron composition. The authors point this out but offer no discussion or further analysis. Was this difference driven by one or more cell types? Is this difference likely to be technical (e.g., less white matter in sample = fewer oligodendrocytes), or could this be related to the infection (e.g., glial death or neurogenesis at 3dpi)?

We have added the location of the punching within the hypothalamus for the different groups to the supplements (Supplementary Figure 3). The differences in neuron/non-neuron composition could originate from differences in the punching location, but we do not have data to support this conclusion. The difference could also stem from biological alterations during the infection.

1. Since influenza viruses replicate in the cell nuclei, did the authors capture any H1N1 RNA in their single-nuclei RNA-seq samples?

We mapped the single nuclei data against the viral genes, but could not detect any of the viral genes in the data set. We are still optimizing detecting of low amounts of viral genes in snRNA-seq data and have not included this information in the manuscript. We believe, that the virus did not manage to migrate in the hypothalamus and infiltrate the cells in the here captured area.

Minor Issues

1. Page 1. The abstract ends with the sentence: This is complemented by increased activity of microglia monitoring their surroundings. Presumably, the authors are basing this statement on the functions of genes altered in microglia by infection. However, saying that microglia behavior has changed is a bit of a stretch here, since the results suggest a change in the molecular phenotype of microglia but do not demonstrate a change in their behavior.

We agree that the phrasing of the end of the abstract was not accurate and didn’t reflect the outcome of the analysis. We adjusted the sentence to: “The change of microglia gene activity suggest that this is complemented by a shift in microglia activity to provide increased surveillance of their surroundings.” Which should provide a better idea that the findings we present are a suggestion based on the transcriptomic changes in the cell population (Page 1).

1. Page 8. The authors refer to Th+, Ddc+ neurons as dopaminergic. However, adrenergic/noradrenergic neurons also express these genes. How do the authors know the neurons are not adrenergic/noradrenergic?

There are to our knowledge no nor-adrenaline/adrenaline producing neurons in the hypothalamus. In contrast dopaminergic neurons have indeed been identified in this area.

1. In the Methods section, Slc17a6 and Slc32a1 are not "pan-neuronal markers" since they are only expressed by subsets of neurons.

We removed the glutamatergic and GABAergic marker genes (Slc17a6 and Slc32a1) from the list of neuronal markers. They are stated further down in the method section as glutamatergic and GABAergic markers. Find the changes on (Page 24/25).

1. Was the hashtagging antibody custom or commercial? If commercial, what was the source, catalog #, lot #? If custom, the authors should describe how it was made and validated.

We used commercial antibodies for hash-tagging. We added the missing information to the manuscript and can be found on Page 24 of the revised manuscript.

1. In the data processing section of the Methods, SCTransform is mentioned twice. Was normalization with SCTransform applied twice?

The data was only normalized once using the SCTransfrom method. We adjusted the part of the method section to make it more clear (Page 24).

1. In the section on gene set enrichment analysis, the first sentence includes this text: "(is a reference needed?)." The answer is yes - Alexa A, Rahnenfuhrer J (2022). topGO: Enrichment Analysis for Gene Ontology. R package version 2.50.0.

The missing reference was added (Page 26).

1. Page 4: "leaved" should be corrected to "left"

The wrong wording was corrected.

1. Figure 2D - gene is labeled as Slc31a1 on the figure and Slc32a1 in the figure legend

We provided a new Figure plate with the right marker genes.

1. Official gene IDs should be italicized

We checked the gene IDs again, and italicized wrongly formatted gene IDs.

1. It is not clear whether the authors are planning to share their code. However, their code would be needed to reproduce their results, since the methods section provides a summary of what was done but lacks key details (e.g., parameters and software packages used during data processing and analysis)

Code will be shared on request. We added this also to the revised manuscript (Page 1).

Reviewer #2 (Public Review):

The new work from Lemcke et al suggests that the infection with Influenza A virus causes such flu symptoms as sleepiness and loss of appetite through the direct action on the responsible brain region, the hypothalamus. To test this idea, the authors performed single-nucleus RNA sequencing of the mouse hypothalamus in controlled experimental conditions (0, 3, 7, and 23 days after intranasal infection) and analyzed changes in the gene expression in the specific cell populations. The key results are promising.

However, the analysis (cell type annotation, integration, group comparison) is not optimal and incomplete and, therefore should be significantly improved.

More specifically:

1. The current annotation of cell types (especially neuronal but also applicable to the group of heterogeneous "Unassigned cells") did not make a good link to existing cell heterogeneity in the hypothalamus identified with scRNA seq in about 20 recently published works. All information about different peptidergic groups can not be extracted from the current version (except for a few). There are also some mistakes or wrong interpretations (eg, authors assigned hypothalamic dopamine cells to the glutamatergic group, which is not true). This state is feasible to improve (and should be improved) with already existing data.

We repeated the cell label transfer with the newly published HypoMap and added additional information to the supplements about the cell type assignments. Additionally, we agree that the dopaminergic neurons do not belong to the group of glutamatergic neurons, however assigned them into this group based on the clustering. We changed the phrasing in the results, to make a better differentiation between the two groups (Page 8).

1. I am confused with the results shown in the label transfer (suppl fig 3 and 4; note, they do not have the references in the text) applied to some published datasets (authors used the Seurat functions 'FindTransferAnchors' and 'TransferData'). The final results don't make sense: while the dataset for the arcuate nucleus (Campbel et al) well covered the GABAergic neurons it is not the case for the whole hypothalamus datasets (Chen et al; Zeisel et al). Similarly, for glutamatergic neurons. Additionally, I could not see that the label transfer works well for PMCH cells which should be present in the dataset for the lateral hypothalamus (Mickelsen et al,2019).

We performed the additional label transfer of the hypothalamus data. Here we accepted a prediction score of 0.5 and transferred a cell type label to our annotated cluster IDs, if at least 10% of cells within a cluster were annotated with the 0.5 prediction score. We found that well defined neuron population types like Hcrt+, Pmch+ and Hdc+ neurons as well as Pomc+ neurons were tagged with a high predictions scores ( >= 0.9, Supplement Figures 6 and 7) and non-neuronal cell types (Supplement Figure 8) were well annotated. Additionally we identified an Agrp+ neuron population with the Gaba_1 neurons. This information has been added to the revised manuscript (Pages 6, 8).

1. There are newly developed approaches to check the shifts in the cell compositions and specific differential gene expression in the cell groups (e.g. Cacoa from Kharchenko lab, scCoda from Büttner et al; etc). Therefore, I did not fully understand why here the authors used the pseudo-bulk approaches for the data analysis (having such a valuable dataset with multiple hashed samples for each timepoint). Therefore it would be great to use at least one of those approaches, which were developed specifically for the scRNAseq data analysis. Or, if there are some reasons - the authors should argue why their approach is optimal

We performed an additional analysis comparing case-control studies (Cacoa). We perfomed both modalities, cluster-based and cluster-free expression shifts and cell type compositions We could partly confirm our findings using the pseudo-bulk approach. The clusterspecific density shift (Supplement Figure 15) identified only shifts in non-neuronal cell types between the Control group and 3 dpi. We believe, these composition shifts are caused by the lower number of non-neuronal cells in the 3 dpi time point. Cluster-specific expression shifts show similar results as in the pseudo-bulk approach, with significant expression shift identified at 3 and 7 dpi in neuronal and non-neuronal cell clusters (Supplement Figure 16). However, no significant expression shifts were identified in the recovery group at 23 dpi. Using the cluster-free expression shift approach, however we were able to identify a similar picture as described with the pseudo-bulk approach. In the recovery group at 23 dpi, we found mainly changed gene programs in neuronal cells, and no transcriptional changes in the non-neuronal cells (Supplement Figure 17-20). This new analysis has been added to the revised manuscript (Pages 4-6, 26) including supplementary figure and tables as stated.

1. When the authors describe the DGE changes upon experimental conditions (Figures 5 and 6), my first comment is again relevant: it is difficult to use the current annotation and cell type description as the reference for testing virus effects and shifts in the DGE in distinct neuronal subtypes.

The cell type annotations have been checked and additional label transfer has been performed. All figures in the manuscript has been updated.

I have to note that the experimental design is well done and logical. Therefore I believe that to strengthen the conclusions, the already obtained datasets can be used for improved analysis.

Reviewer #2 (Recommendations For The Authors):

I have some minor concerns:

1. For the quality check it would be good to see how different hashed samples for each timepoint cover the UMAP embeddings.

We added the UMAP embeddings to the supplement (Supplement Figure 4).

1. In Fig 1e colors are not optimal - it is impossible to assess it.

We separated the UMAPs for the different time points to make it easier to assess. See updated Figure 1E.

1. In the methods authors started "Single-nucleus RNA-sequencing cell population identification" from the description of using a Gaussian mixture model (GMM). However, I could not clearly understand how this model was used and which kind of result it provided.

We used an GMM model with known markers for neurons and in a second step for glutamatergic and GABAergic cells to sub-cluster the cells and then selected based on high and low expression of the marker genes in the cluster into their respective classes. This information has been added to the method section (Page 24/25).

1. Could the authors better clarify why "they calculated normalization factors using the scran function 'computeSumFactors'" when working with pseudobulk analysis?

This size factor normalization was recommended for single cell data by the authors of the DESeq2 packages.

http://bioconductor.org/packages/devel/bioc/vignettes/DESeq2/inst/doc/DESeq2.html

1. I didn't find logic in "a cell cluster was only included if it contained more than 2 nuclei in at least 3 individual animals" (page 24). Maybe I misinterpreted it.

The rationale for the selection methods was based on the findings that not all animals in the recovery group had the same effects in weight loss. The acute time points didn’t show enough weight loss to decide if all animals in these groups lost the same amount and were equally sick. Hence, in order to have biological robustness we decided to only analyse clusters where cells from at least 3 animals at a specific time point contributed to a cell type. In order to have enough cells per cell type for the calculation of DEGs, we decided to only include a cell type at a specific time point if it contained at least 3 cells from one individual. This selection method limits the analysis to cell types with at least 9 cells per time point.

Author details

1. René Lemcke

   Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   For correspondence

   rene.lemcke@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4903-4741

2. Christine Egebjerg

   Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Methodology, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0009-0006-2035-6145

3. Nicolai T Berendtsen

   Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

4. Kristoffer L Egerod

   Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

5. Allan R Thomsen

   Department of Immunology and Microbiology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Conceptualization, Resources, Writing – review and editing

   Competing interests

   No competing interests declared

6. Tune H Pers

   Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Resources, Software, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-0207-4831

7. Jan P Christensen

   Department of Immunology and Microbiology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Conceptualization, Resources, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4299-9479

8. Birgitte R Kornum

   Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Conceptualization, Supervision, Funding acquisition, Investigation, Writing – original draft, Project administration, Writing – review and editing

   For correspondence

   kornum@sund.ku.dk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2515-9451

• 202

  Page views

• 36

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.