Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Mar 21;38(12):3060-3080.
doi: 10.1523/JNEUROSCI.1740-17.2018. Epub 2018 Feb 27.

Long-Term Neuroinflammation Induced by Influenza A Virus Infection and the Impact on Hippocampal Neuron Morphology and Function

Shirin Hosseini  1   2 , Esther Wilk  3 , Kristin Michaelsen-Preusse  1 , Ingo Gerhauser  4 , Wolfgang Baumgärtner  4 , Robert Geffers  5 , Klaus Schughart  3   6   7 , Martin Korte  8   2
Affiliations

Long-Term Neuroinflammation Induced by Influenza A Virus Infection and the Impact on Hippocampal Neuron Morphology and Function

Shirin Hosseini et al. J Neurosci. .

Abstract

Acute influenza infection has been reported to be associated with neurological symptoms. However, the long-term consequences of an infection with neurotropic and non-neurotropic influenza A virus (IAV) variants for the CNS remain elusive. We can show that spine loss in the hippocampus after infection with neurotropic H7N7 (rSC35M) and non-neurotropic H3N2 (maHK68) in female C57BL/6 mice persists well beyond the acute phase of the disease. Although spine number was significantly reduced at 30 d postinfection (dpi) with H7N7 or H3N2, full recovery could only be observed much later at 120 dpi. Infection with H1N1 virus, which was shown previously to affect spine number and hippocampus-dependent learning acutely, had no significant long-term effects. Spine loss was associated with an increase in the number of activated microglia, reduced long-term potentiation in the hippocampus, and impairment in spatial memory formation, indicating that IAV-associated inflammation induced functional and structural alterations in hippocampal networks. Transcriptome analyses revealed regulation of many inflammatory and neuron- and glia-specific genes in H3N2- and H7N7-infected mice at day 18 and in H7N7-infected mice at day 30 pi that related to the structural and functional alterations. Our data provide evidence that neuroinflammation induced by neurotropic H7N7 and infection of the lung with a non-neurotropic H3N2 IAV result in long-term impairments in the CNS. IAV infection in humans may therefore not only lead to short-term responses in infected organs, but may also trigger neuroinflammation and associated chronic alterations in the CNS.SIGNIFICANCE STATEMENT In the acute phase of influenza infection, neuroinflammation can lead to alterations in hippocampal neuronal morphology and cognitive deficits. The results of this study now also provide evidence that neuroinflammation induced by influenza A virus (IAV) infection can induce longer-lasting, virus-specific alterations in neuronal connectivity that are still detectable 1 month after infection and are associated with impairments in spatial memory formation. IAV infection in humans may therefore not only lead to short-term responses in infected organs, but may also trigger neuroinflammation and associated chronic alterations in the CNS.

Keywords: dendritic spines; hippocampus; influenza; microglia; neuroinflammation; structural plasticity.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Female C57BL/6J mice were infected intranasally with the indicated viruses and dosages. A, Body weight loss depicted as percentage of the starting weight of mice during the acute phase of IAV infection (n = 10–18 in each group). B, Brains of infected mice were tested for the presence of infectious virus in embryonated eggs. Positivity in a hemagglutination assay is displayed as number of positive samples/number of tested samples. Positive samples from the indicated days were titrated by determining EID50 (egg infectious dose 50)/ml. C, Representative sections from the immunohistochemical analysis, hippocampus (left), and medulla oblongata (right) of mice 7 d after intranasal infection with H3N2 (maHK68) (top row) and H7N7 (rSC35M) (bottom row) IAV subtypes. Immunohistochemistry did not reveal influenza NP in the hippocampus and medulla oblongata of H3N2-infected mice and hippocampus of H7N7-infected mice, whereas high numbers of virus infected cells in the medulla oblongata of H7N7-infected mice were detected. Scale bars, 200 μm. Sections were counterstained with Mayer's hematoxylin. Inserts, Higher magnifications of the respective images. Scale bars, 33 μm. D, In the medulla oblongata of a mouse at 9 d after intranasal H7N7 infection, severe lymphohistiocytic meningitis, few numbers of inflammatory cells in the parenchyma, and a moderate gliosis (center of the image) were observed. Scale bars, 80 μm. Top insert, Degenerating cells in higher magnification (arrow). Bottom insert, Viral antigen was found in one cell (arrow) using immunohistochemistry for influenza nucleoprotein and Mayer's hematoxylin as counterstaining. Scale bars, 33 μm. E, H&E staining from brains of H3N2- and H7N7-infected mice were scored semiquantitatively for signs of inflammation at the respective days (n = 3–5). F, Immunohistochemistry of viral NP was scored semiquantitatively (n = 4–5). Data are presented as mean ± SEM.
Figure 2.
Figure 2.
Long-term effect of IAV infection on general locomotion and willingness to explore in the open-field test and anxiety-like behavior in the elevated plus maze test. A, B, At 30 and 120 dpi, a total distance traveled [one-way ANOVA 30 dpi (n = 7–10): F(3,29) = 2.12, p = 0.11, and 120 dpi (n = 7–8): F(2,19) = 6.19, p = 0.08), average speed (one-way ANOVA - 30 dpi (n = 7–10): F(3,29) = 2.86, p = 0.054 and 120 dpi (n = 7–8): F(2,19) = 6.24, p = 0.08] and representative tracks of movement patterns of mice in an open-field box are presented. There was no significant difference between all tested groups. C, D, Activity percentage of mice in the periphery [one-way ANOVA 30 dpi (n = 7–10): F(3,29) = 2.15, p = 0.11 and 120 dpi (n = 7–8): F(2,19) = 2.73, p = 0.09] and center part [one-way ANOVA 30 dpi (n = 7–10): F(3,29) = 2.14, p = 0.11 and 120 dpi (n = 7–8): F(2,19) = 2.73, p = 0.09] of open-field arena and not show any significant changes. Therefore, no sickness behavior, locomotors deficiency or anxiety-like behavior was detectable in infected mice. E, F, Percentage of time spent in the open [one-way ANOVA 30 dpi (n = 7–10): F(3,29) = 3.80, p = 0.20 and 120 dpi (n = 7–8): F(2,19) = 1.00, p = 0.38), and closed arms (One-way ANOVA - 30 dpi (n = 7–10): F(3,29) = 1.61, p = 0.20 and 120 dpi (n = 7–8): F(2,19) = 0.18, p = 0.83] of elevated plus maze were similar in all groups tested at 30 and 120 dpi. Mice did not indicate elevated anxiety levels. Data are presented as mean ± SEM; ordinary one-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were performed.
Figure 3.
Figure 3.
Long-term effect of IAV infection on hippocampus-dependent spatial learning. A, During 8 d of acquisition training, the escape latency reduced significantly in each group of control and infected mice, at 30 dpi, the escape latency in H7N7-infected mice was significantly increased compared with control and non-neurotropic H1N1- and H3N2-infected mice. B, At 120 dpi, the escape latency in all control and IAV-infected mice did not reveal any significant differences. One single probe trial was performed after day 3, 6, and 9 of the training period. C, Percentage of time spent in the target quadrant (NE) by H1N1- and H3N2-infected mice at 30 dpi was increased similarly to control mice, whereas H7N7-infected mice showed a significantly reduce target quadrant preference on day 6 and 9 compared with the other groups tested. D, Quadrant preference during the probe trials 120 dpi was similar in all groups. Data are presented as mean ± SEM (n = 7–10), one-way and two-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were performed. *p < 0.05 and ***p < 0.001 compared with control. ++p < 0.01 and +++p < 0.001 compared with H1N1; ∧p < 0.05, ∧∧p < 0.01, and ∧∧∧p < 0.001 compared with H3N2.
Figure 4.
Figure 4.
Analysis of learning strategies reveals a spatial learning impairment for both neurotropic and non-neurotropic virus subtypes. With regard to the different searching strategies to locate the hidden platform during the acquisition phase of the Morris water maze experiment, hippocampus-independent searching strategies including random swimming, chaining, and scanning decreased over time, whereas the hippocampus-dependent strategy-directed search increased. The searching strategies (directed search, chaining, scanning, and random swimming) were color coded and the relative contribution of the respective strategy is presented for each day of the Morris water maze task. A, Hippocampus-dependent strategy was decreased after H7N7 infection compared with the other groups and H3N2-infected mice showed a reduction in the usage of direct search at 30 dpi. B, No significant differences in searching strategies between IAV-infected and control mice were observed at 120 dpi. Data are presented as mean ± SEM (n = 7–10), two-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were performed. *p < 0.05 and **p < 0.01 compared with control; +p < 0.05 compared with H1N1.
Figure 5.
Figure 5.
Infection with neurotropic and non-neurotropic virus subtypes impairs memory formation for a new platform position. A, During 3 d of training, the escape latency to a new position of the hidden platform (SW) decreased significantly in control, H1N1-infected, and H7N7-infected mice over days, however, not in H3N2-infected mice. H7N7-infected mice had a significantly elevated escape latency compared with control and H1N1-infected mice. B, At 120 dpi, the IAV-infected group did not show any significant differences in the escape latency compared with the control. A single probe trial test 24 h after the last day of reversal training was performed. C, Only control and H1N1-infected mice spent significantly more time in the new target quadrant (T) in comparison with the average time spent in nontarget quadrants (NT). D, All tested groups spent more time in T compared with NT at 120 dpi. Data are presented as mean ± SEM (n = 7–10). In A and B, repeated-measures one-way and two-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were used; in C and D, unpaired t test was used. **p < 0.01 compared with control; +p < 0.05 and +++p < 0.001 compared with H1N1; #p < 0.05, ##p < 0.01, and ###p < 0.001 compared with NT.
Figure 6.
Figure 6.
Long-term effect of IAV infection on the function of CA1 hippocampal neurons. A, Input–output curves of fEPSP slopes in hippocampal slices (n = 10) of control and infected mice at 30 and 120 dpi do not show any significant differences between groups. B, PPF of fEPSP slopes depicted as response to the second stimulation over the first at different interpulse intervals (10, 20, 40, 60, 80, and 100 ms) in hippocampal slices (n = 9–11) do not show any differences between the groups at 30 and 120 dpi. C, Hippocampal slices from H7N7-infected mice (n = 15) exhibit significantly lower induction and maintenance of LTP compared with control, whereas H3N2-infected mice (n = 13) showed a reduced maintenance of LTP compared with control (n = 17) at 30 dpi. D, At the induction phase of LTP (T 20–25), only hippocampal slices from H7N7-infected mice had a significantly reduced LTP; however, at the stable phase of LTP (T 75–80), both groups of slices from H3N2 and H7N7 influenza virus-infected mice revealed a significant reduction in LTP compared with control hippocampal slices. E, F, At 120 dpi, the induction and maintenance phases of LTP did not show any differences in control and infected groups (n = 11–15). Data are presented as mean ± SEM (n = 4–5). In A, B, C, and E, two-way ANOVA was used; in D and F, one-way ANOVA and post hoc Bonferroni's multiple-comparisons test were used. *p < 0.05 and **p < 0.01 compared with control; ∧p < 0.05 compared with H3N2. N, Number of mice; n, number of hippocampal slices in each group.
Figure 7.
Figure 7.
Long-term effect of IAV infection on dendritic spine density of hippocampal neurons. A, Representative images of Golgi-stained hippocampus sections. Scale bars, 200 μm, 2.5× hippocampal neurons; 20 μm, 20× and dendritic spines in hippocampal CA1 apical neurons after infection with IAV; 2 μm, 63×. B–E, After infection with H3N2 and H7N7 IAV, the spine density of apical dendrites of CA1 (B) and CA3 (C) hippocampal neurons decreased at 30 dpi; only H7N7 IAV infection reduced dendritic spine density of dentate granule cells located in the superior (D) and inferior (E) blade of the granule cell layer. At 60 dpi, a partial recovery occurred in the DG and CA3 hippocampal subregions of infected animals and, at 120 dpi, the dendritic spine density fully recovered in all regions of the hippocampus. Data are presented as mean ± SEM (n = 4–5 and number of dendrites in each group = 40–50), one-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were performed. *p < 0.05 and ***p < 0.001 compared with control; ##p < 0.01 and ###p < 0.001 compared with 30 dpi time point.
Figure 8.
Figure 8.
Long-term effect of IAV infection on glial cell density and activation status within the hippocampal subregions. A, Representative examples of IBA-1 immunostaining at 30 dpi. Scale bar, 100 μm. Inserts, Higher magnifications of the respective images. Scale bar, 10 μm. B, After infection with H3N2 IAV, microglia density in the CA3 region and inferior blade of the DG was increased significantly, whereas the neurotropic H7N7 IAV infection induced an increased microglia density in all hippocampal subregions at 30 dpi. At 60 dpi, a partial recovery occurred in the CA3 and DG regions of infected mice and, at 120 dpi, microglia density was fully recovered in all subregions of the hippocampus (n = 4 and number of ROIs in each group = 20). The activation status of microglia was assessed by counting the number of primary processes. C, After infection with H3N2 and H7N7 IAV, the number of primary processes of microglia in all subregions of the hippocampus decreased at 30 dpi, however, upon H7N7 infection, the strongest reduction became visible in the superior and inferior blade of the granule cell layer. Conversely, at 60 dpi, a partial recovery occurred in the CA3 and DG regions of infected mice and, at 120 dpi, microglia activation status was fully recovered in all subregions of the hippocampus (n = 4 and number of selected microglia in each group = 120–200). D, Representative examples of GFAP immunostaining at 30 dpi. Scale bar, 50 μm. Inserts, Higher magnifications of the respective images. Scale bar, 10 μm. E, Astrocyte density in all hippocampal subregions was increased at 30 dpi with H7N7 IAV, whereas only CA1 and CA3 were affected after H3N2 IAV infection. Interestingly, at 60 dpi and 120 dpi, a reduction of GFAP-positive cells to the level of controls was observed (n = 2–4 and number of ROIs in each group = 5–20). Data are presented as mean ± SEM, one-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were performed. *p < 0.05 and ***p < 0.001 compared with control; ∧∧∧p < 0.001 compared with H3N2; #p < 0.05, ## p < 0.01, and ### p < 0.001 compared with the 30 dpi time point.
Figure 9.
Figure 9.
Effect of IAV infection on BBB permeability and cytokine level. A, Injection of Evans blue dye for assessment of the BBB integrity upon infection with H3N2 and H7N7 IAV showed an increased Evans blue absorbance on 8 dpi in both H3N2- and H7N7-infected mice (n = 3–4 and number of samples in each group = 6–8). B, On 10 dpi, Evans blue dye was well visible macroscopically only in H7N7-infected mice, whereas in H3N2-infected mice, it was only weakly visible around the ventricle (black arrow). C–H, Levels of IFN-γ and TNF-α were significantly elevated in the blood serum, brain, and hippocampus of H7N7-infected mice. H, H1N1 and H3N2 non-neurotropic IAV infection led to significantly increased TNF-α level within the hippocampus of infected mice (n = 2–4 and number of samples in each group = 3–8). Data are presented as mean ± SEM, one-way ANOVA of data and post hoc Bonferroni's multiple-comparisons test were performed. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with control.
Figure 10.
Figure 10.
Whole genome microarray analysis from hippocampus of influenza-infected mice at 18 and 30 dpi. DEPs were identified based on an adjusted p-value of < 0.1 and exhibiting more than a 1.4-fold (log2 of 0.5) difference in expression levels. A, At 18 dpi, 487 and 174 DEPs were detected in the hippocampus of H3N2- and H7N7-infected mice respectively. However, at 30 dpi, DEPs (250) were only found in H7N7-infected mice. B, Overlap of differentially expressed genes (DEGs) that are represented by the DEPs is presented as Venn diagram. C, KEGG pathway analysis of DEGs after H3N2 and H7N7 IAV infection revealed significant pathways involved in local immune responses and cell adhesion molecules in the hippocampus of H3N2- and H7N7-infected mice at 18 dpi, which are more pronounced and continued until 30 dpi for H7N7 IAV infection. The diameter of the dots indicates the gene ratio; range of 0.05 (smallest dot) to 0.20 (biggest dot), colors show significance of DEG representation for each pathway. D, Relative changes (with reference to mock-infected mice) in expression levels of microglia signature and activation genes in the hippocampus after IAV infection. Data are presented as LogFC (fold change) mean in each groups compared with control group (n = 3–4 ad independent biological replicates). p-value is adjusted using Benjamini–Hochberg correction for multiple testing.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Bailey CH, Kandel ER, Harris KM (2015) Structural components of synaptic plasticity and memory consolidation. Cold Spring Harb Perspect Biol 7:a021758. 10.1101/cshperspect.a021758 - DOI - PMC - PubMed
    1. Banks J, Speidel E, Alexander DJ (1998) Characterisation of an avian influenza A virus isolated from a human–is an intermediate host necessary for the emergence of pandemic influenza viruses? Arch Virol 143:781–787. 10.1007/s007050050329 - DOI - PubMed
    1. Bannerman DM, Bus T, Taylor A, Sanderson DJ, Schwarz I, Jensen V, Hvalby Ø, Rawlins JN, Seeburg PH, Sprengel R (2012) Dissecting spatial knowledge from spatial choice by hippocampal NMDA receptor deletion. Nat Neurosci 15:1153–1159. 10.1038/nn.3166 - DOI - PMC - PubMed
    1. Bao Y, Gao Y, Shi Y, Cui X (2017) Dynamic gene expression analysis in a H1N1 influenza virus mouse pneumonia model. Virus Genes 53:357–366. 10.1007/s11262-017-1438-y - DOI - PubMed
    1. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful apporach to multiple testing. Journal of the Royal Statistical Society 57:289–300.

Publication types

-