Abstract
Protein kinases regulate protein activity through phosphorylation, and
many have been reported to participate in brain development. Among
them, serine/threonine-protein kinase 24 (STK24) is believed to
influence apoptosis, spinal synaptogenesis, and neuronal migration.
Despite its recognized roles, the functions of STK24 in the brain
remains insufficiently explored. Here, we present an in vivo study of
brain-specific Stk24 conditional knockout mice. We investigate the
impact of Stk24 deletion through histological analysis, behavior
assays, and the molecular changes. In our results, Stk24 deletion
disrupts the hippocampal formation during development and decreased
subsequent adult hippocampal neurogenesis whilst neuronal morphology is
relatively unaffected. Additionally, Stk24-deficient mice exhibit
anxiety-like behavior and altered stress responses, featuring increased
hippocampal neuronal activity, dysregulated HPA axis reactivity, and
modified expression patterns of glucocorticoid receptor
signaling-related genes. In conclusion, our findings highlight the
involvement of Stk24 in brain development, adult hippocampal
neurogenesis, as well as anxiety and stress responses.
Subject terms: Adult neurogenesis, Limbic system, Genetics of the
nervous system, Neurophysiology
__________________________________________________________________
Brain-specific Stk24 deletion in mice disrupts hippocampal development,
impairs adult neurogenesis, and alters stress responses including
anxiety-like behavior and HPA axis reactivity.
Introduction
Protein kinases regulate the biological activity of proteins via
phosphorylation. It has been reported that numerous kinases participate
in brain development including promoting axonal growth and altering
neurogenesis. Serine/Threonine-protein kinase 24 (STK24), also known as
Mammalian sterile 20-like kinase-3 (MST3), is one of those kinases. It
is well known for its role as an upstream regulator of the
mitogen-activated protein kinase (MAPK) signaling pathway, mediating
multiple biological processes including apoptosis, morphogenesis, and
cell migration^[32]1. The encoded protein, MST3, consists of a shorter
MST3a and a longer MST3b isoform, which differ by 12 amino acids^[33]2.
MST3a is expressed universally and is believed to regulate apoptosis
and neuronal migration during brain development^[34]3. MST3b,
demonstrated to participate in axonal outgrowth, is selectively
expressed in the brain; particularly highly in the hippocampus and
cerebral cortex^[35]2,[36]4. According to the Hippocampus RNA-seq Atlas
([37]https://hipposeq.janelia.org), Stk24 is expressed throughout the
dorsal and ventral hippocampus with equivalent expressions in CA1, CA2,
CA3, and DG regions. Additionally, as reported by the Human Protein
Atlas ([38]https://www.proteinatlas.org), STK24/MST3 is expressed in
excitatory neurons, inhibitory neurons, and glial cells.
So far, many studies have revealed the impact of STK24/MST3 on the
central nervous system. Among them, it is reported that shRNA-induced
Stk24 silencing reduced spinal synapses but increased filopodia in
developing hippocampal cultures^[39]5. In utero silencing of Stk24 also
disrupted neuronal positioning and dendritogenesis by modulating RhoA
activity in mice^[40]6. Moreover, knocking down Stk24 in rat dorsal
root ganglions was shown to severely impair the regeneration of injured
axons^[41]7. Previous studies have provided new insight into the role
of Stk24 in brain development. However, most of them used shRNA to
evaluate the roles of Stk24 in neuronal cells. The influence of Stk24
deletion on the brain and animal behavior is currently lacking. Here,
we present an in vivo study of brain-specific Stk24 conditional KO
(cKO) mice to reveal the roles of Stk24 in the brain, ranging from
histological analysis to behavioral phenotypes. Our results showed that
brain-specific Stk24 cKO mice exhibited ectopic neurons in the
hippocampus during brain development, along with impaired adult
hippocampal neurogenesis. Increased anxiety and elevated stress-induced
plasma corticosterone were also discovered. Our study demonstrates that
Stk24 plays a role in hippocampal architecture and mediates animal
behavior.
Results
Brain-specific conditional knockout of Stk24 disrupted postnatal hippocampal
development
A brain-specific Stk24 conditional knockout (cKO) mouse model was
generated by crossing Stk24 floxed mice with Sox1::Cre mice, which
express Cre throughout the neural tube from E9.5^[42]8,[43]9. This
conditional knockout model allowed us to investigate the functions of
Stk24 in the brain. Knockout efficiency was validated at both the mRNA
and protein levels using real-time PCR and western blot, respectively,
both of which showed a significant reduction in Stk24 expression
compared to the control wild-type (WT) group (PCR: t[(16)] = 10.37,
p < 0.001; western blot: t[(4)] = 4.88, p < 0.01) (Fig. [44]1a). The
specificity of this conditional knockout model was confirmed by the
significant reduction of Stk24 in the brain, while the peripheral
regions such as liver levels remained unchanged (OB: mRNA
t[(5)] = 10.73, p = 0.001; Protein t[(5)] = 5.58, p < 0.01; CTX: mRNA
t[(6)] = 8.80, p < 0.001; Protein t[(5)] = 7.12, p < 0.001; HIP: mRNA
t[(6)] = 14.66, p < 0.001; Protein t[(5)] = 7.05, p < 0.001; THAL: mRNA
t[(6)] = 6.68, p < 0.001; Protein t[(4)] = 8.06, p < 0.01; CB: mRNA
t[(6)] = 9.46, p < 0.001; Protein t[(5)] = 8.55, p < 0.001; Liver: mRNA
t[(6)] = 0.44, p = 0.67; Protein t[(5)] = 1.59, p = 0.17)
(Supplementary Fig. [45]1).
Fig. 1. Brain-specific conditional knockout of Stk24 caused disrupted
postnatal hippocampal development.
[46]Fig. 1
[47]Open in a new tab
a Stk24 cKO mice were obtained by crossing Sox1::Cre mice with
Stk24^f/f mice. Stk24 expression was significantly decreased in cKO
mice at the mRNA (left, n = 8–10) and protein (right, n = 3) levels.
Black dots represent male mice, while red dots represent females.
Unpaired t-test. b The spine densities were equivalent between the
groups (n = 3–5). Black dots represent male mice, while red dots
represent females. Unpaired t-test. Scale bar = 10 µm. c There was no
significant difference in the intersections, total intersections, and
the length of the longest dendrites (n = 10 neurons from each group).
WT: 2 males, 3 females; cKO: 2 males, 1 female. Repeated two-way ANOVA
and unpaired t-test. Scale bar = 10 µm. d Stk24 cKO mice exhibited
significantly more ectopic granule cells (PAX6, TBR2, PROX1: n = 4;
NeuroD: n = 2). The enclosed regions represent ROI. Black dots
represent male mice, while red dots represent females. Unpaired t-test.
Scale bar = 100 µm. e Volcano plot of the DE genes with a
p-value < 0.05 and a ∣log[2](fold change) ∣> 0.5 (n = 3). WT: 2 males,
1 female; cKO: 3 males. f Enriched BP pathways targeted by GO analysis.
The data were presented as mean ± SEM for each group. *p < 0.05,
**p < 0.01, ***p < 0.001.
Since Stk24 is known to influence synaptic morphology in developing
hippocampal cultures^[48]5, we first asked whether Stk24 deletion
affected the neuronal outgrowth and postnatal neurogenesis in
developing brains. To this end, Golgi staining was performed on
postnatal day 13 (p13) mouse brains to examine the morphology of
hippocampal neurons. Dendritic spine density was quantified, and there
was no significant difference in the densities between the groups
(t[(6)] = 1.28, p = 0.24) (Fig. [49]1b). We further conducted Sholl
analysis for a more detailed investigation, and our results showed that
WT and Stk24 cKO mice exhibited equivalent dendritic branching
(F[(1,18)] = 0.44, p = 0.51), total intersections (t[(18)] = 0.66,
p = 0.51), and length of the longest dendrite (t[(18)] = 1.83,
p = 0.08) (Fig. [50]1c), suggesting unaffected neuronal morphology in
the absence of Stk24. Next, we studied the postnatal neurogenesis by
assessing the expression of neurogenesis markers including PAX6, TBR2,
NeuroD, and PROX1. Immunohistochemical (IHC) staining was performed on
p13 mouse brains, and the result showed disrupted hippocampal
neurogenesis in Stk24 cKO mice, featuring ectopic granule cells in the
DG of the hippocampus (Fig. [51]1d). Given that hippocampal
neurogenesis occurs in the sub-granular layer of the DG under normal
conditions, we calculated the antibody-labeled cells in the hilus for
the quantification of ectopic cells, and there was an overall
significant increase in Stk24 cKO group (PAX6: t[(6)] = 4.61, p < 0.01;
TBR2: t[(6)] = 5.15, p < 0.01; NeuroD: t[(2)] = 13.16, p < 0.01; PROX1:
t[(6)] = 7.93, p < 0.001). Taken together, despite unaffected neuronal
morphology, Stk24 deletion disrupted hippocampal neurogenesis during
brain development.
To investigate gene expression differences between WT and cKO mice, we
performed RNA-sequencing on hippocampal tissues from P13 mice.
Differential expression analysis was conducted using DESeq2, with the
following criteria for identifying differentially expressed (DE) genes:
p-value < 0.05; |Log2 fold change (FC)| >0.5; baseMean >20. A total of
114 DE genes were identified, of which 74 were significantly
up-regulated and 40 were down-regulated (Fig. [52]1e). Gene ontology
(GO) enrichment analysis was further conducted via WebGestalt
([53]https://www.webgestalt.org, accessed on 2025-02-11) focusing on
biological process (BP). This analysis revealed 10 up-regulated and 8
down-regulated pathways from these DE genes (Fig. [54]1f). The enriched
GO categories compassed a broad range of biological processes including
hormone response, differentiation and development, cell division,
apoptosis and autophagy. Interestingly, the down-regulated genes were
strongly associated with neurogenesis, neuron migration, and pattern
specification, which is in line with our previous results where Stk24
cKO mice exhibited disrupted postnatal neurogenesis and hippocampal
formation.
Moreover, since Stk24 is well recognized for its role as an upstream
regulator of the MAPK signaling pathway, we specifically examined the
expression of MAPK pathway-related genes^[55]1. The gene list was
obtained from the mouse gene set “MAPK Family Signaling Cascades”
(R-MMU-5683057), provided by the Reactome database through the
Molecular Signatures Database (MSigDB)
([56]https://www.gsea-msigdb.org, accessed on 2025-02-12). Among the
303 MAPK pathway-related genes, 27 genes (12 up-regulated genes and 15
down-regulated genes) exhibited significant expression changes with a
p-value < 0.05 (Supplementary Fig. [57]2a). However, when we further
conducted the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
enrichment analysis via WebGestalt ([58]https://www.webgestalt.org,
accessed on 2025-02-11) using gene set enrichment analysis (GSEA), MAPK
pathway was not targeted as a significantly enriched pathway
(Supplementary Fig. [59]2b). In summary, although several MAPK
pathway-related genes were significantly affected by Stk24 deletion,
the overall MAPK signaling pathway did not appear to be influenced. It
indicated that MAPK pathway may not be the major pathway affected by
Stk24 knockout, or the genetic expression changes was not sufficient to
impact the pathway-level regulation.
Brain-specific Stk24 cKO mice exhibited hyperactivity and increased anxiety
Deletion of Stk24 was demonstrated to affect the hippocampal
development. Since the hippocampal formation is widely believed to be
involved in emotional responses and cognitive abilities^[60]10, we
further investigated the effects of Stk24 on animal behaviors,
particularly those related to the hippocampal functions. Before the
mice were subjected to a series of behavioral tests, the spontaneous
locomotor activity was assessed. Stk24 cKO mice displayed a
significantly longer travel distance as compared to WT mice during
active phase (F[(1,26)] = 8.80, p < 0.01), but not inactive phase
(F[(1,26)] = 0.07, p = 0.78), suggesting a tendency to hyperactivity
(Fig. [61]2a). Grip strength and motor coordination were also measured
by grip test and rotarod test, respectively. Both showed no significant
difference between the groups (grip: t[(26)] = 1.60, p = 0.12; rotarod:
t[(26)] = 1.32, p = 0.19), suggesting normal motor skills
(Supplementary Fig. [62]3a, [63]b).
Fig. 2. Brain-specific Stk24 cKO mice exhibited hyperactivity and increased
anxiety.
[64]Fig. 2
[65]Open in a new tab
a Stk24 cKO exhibited longer travel distance during the active phase
(ZT19-ZT6) as compared to WT mice (n = 14). WT: 8 males, 6 females;
cKO: 8 males, 6 females. Repeated two-way ANOVA. b In the open field
test, Stk24 cKO mice spent longer time in the center zone and had
higher frequency of entering the center area (n = 14). Black dots
represent male mice, while red dots represent females. Unpaired t-test.
c In the elevated O-maze, no significant difference was detected in the
time mice spent on the open arms and the number of transitions
(n = 14). Black dots represent male mice, while red dots represent
females. Unpaired t-test. d In the light-dark box, Stk24 cKO mice spent
longer time in the light box and a greater number of transitions
(n = 14). Black dots represent male mice, while red dots represent
females. Unpaired t-test. The data were presented as mean ± SEM for
each group. *p < 0.05, **p < 0.01, ***p < 0.001.
Open field, elevated O-maze, and the light-dark box were used to
evaluate anxiety in mice. Although there was no significant difference
in the results of the elevated O-maze (relative distance:
t[(26)] = 1.41, p = 0.17; duration: t[(26)] = 1.10, p = 0.27;
transition: t[(26)] = 1.07, p = 0.29) (Fig. [66]2c; Supplementary Fig.
[67]3d), both of the open field and light-dark box tests showed
remarkable differences between cKO and control WT group. In open field
test, despite equivalent relative distance in the center zone
(t[(26)] = 1.72, p = 0.09), Stk24 cKO mice exhibited less time in the
center area (t[(26)] = 2.09, p = 0.04) and decreased number of
transitions into the center (t[(26)] = 2.58, p = 0.01) (Fig. [68]2b;
Supplementary Fig. [69]3c). Light-dark box test showed a consistent
tendency, featuring less time (t[(25)] = 2.81, p < 0.01) and decreased
frequency (t[(26)] = 4.12, p < 0.001) to the light box (Fig. [70]2d).
In summary, brain-specific Stk24 cKO mice displayed a higher level of
anxiety as compared to control WT mice.
No significant difference was detected in cognitive abilities and
electrophysiological characteristics between WT and Stk24 cKO mice
Given that the hippocampus plays a key role in learning and
memory^[71]10, we asked whether Stk24 deletion influenced cognition
since abnormal hippocampal formation was detected during development. A
battery of behavioral tests was conducted, including Morris water maze,
Y-maze, and novel object recognition, each of which tested for a
specific set of cognitive abilities. The Morris water maze was widely
used for measuring spatial learning and memory in rodents. No
significant difference was detected in the learning curves between WT
and cKO mice (F[(1,26)] = 0.70, p = 0.40), showing similar spatial
learning ability (Fig. [72]3a). Seven days after the last training
trial, a memory test was carried out. The time in the target quadrant
(U = 80.00, p = 0.42) and the number of transitions (t[(26)] = 0.93,
p = 0.35) were also equivalent between the groups, showing unaffected
spatial memory in cKO mice (Fig. [73]3b). The Y-maze was used to test
working memory. WT and Stk24 cKO mice showed equivalency in the
percentage of correct choices (t[(25)] = 0.24, p = 0.80), indicating no
difference in working memory (Fig. [74]3c). The novel object
recognition test was used to assess recognition memory, and the results
were calculated as discrimination index (D.I.). Both WT and cKO group
exhibited a D.I. of ~0.6, showing a tendency to novel objects; yet no
difference was found in the recognition memory between the two groups
(t[(26)] = 0.82, p = 0.41) (Fig. [75]3d). Taken together,
brain-specific Stk24 deletion did not significantly affect the tested
cognitive abilities.
Fig. 3. No significant difference was detected in cognitive abilities and
electrophysiological characteristics between WT and Stk24 cKO mice.
[76]Fig. 3
[77]Open in a new tab
a No significant difference was found in the learning curves between WT
and Stk24 cKO mice during Morris water maze test (n = 14). WT: 8 males,
6 females; cKO: 8 males, 6 females. Repeated two-way ANOVA. b In the
memory test, WT and cKO mice exhibited equivalency in the time spent in
target quadrant and the platform crosses (n = 14). Black dots represent
male mice, while red dots represent females. Unpaired t-test. c WT and
cKO mice showed comparable percentage of alterations in Y-maze
(n = 14). Black dots represent male mice, while red dots represent
females. Unpaired t-test. d In novel object recognition test, WT and
cKO mice displayed similar percentage of D.I (n = 14). Black dots
represent male mice, while red dots represent females. Unpaired t-test.
e There was no significant difference in the pattern of LTP between the
groups (n = 7 recording sites from 3 male mice for each group).
Repeated two-way ANOVA. f There was no significant difference in the
paired-pulse ratio between WT and Stk24 cKO mice (n = 10 recording
sites from 3 male mice for each group). Repeated two-way ANOVA. The
data were presented as mean ± SEM for each group.
Although Stk24 cKO mice did not show impaired learning nor memory, we
further examined the synaptic plasticity of hippocampal neurons, as
these are believed to underpin cognitive functions. For this purpose,
we investigated the electrophysiological characteristics of hippocampal
neurons. Field excitatory postsynaptic potential (fEPSP) was recorded
in the CA1 region of hippocampus to measure the long-term potentiation
(LTP) and the paired-pulse facilitation (PPF). Our data revealed that
WT and cKO mice displayed similar patterns of LTP (F[(1,12)] = 0.0001,
p = 0.99) and PPF (F[(1,18)] = 1.47, p = 0.24) (Fig. [78]3e, f),
indicating unaffected neuronal plasticity in the absence of Stk24.
Brain-specific Stk24 deletion decreased adult hippocampal neurogenesis
Stk24 cKO mice had showed disrupted hippocampal neurogenesis during
brain development. Adding that emotional behavior has been proposed to
share a link with adult hippocampal neurogenesis^[79]11,[80]12,
although more research is needed, we were interested if Stk24 altered
adult hippocampal neurogenesis as well considering the increased
anxiety and abnormal hippocampal development in Stk24 cKO mice. To this
end, mice were given 5-bromo-2’-deoxyuridine (BrdU) for two injections
every other day (200 mg kg^−1 per injection, i.p.) to label newly born
cells. Twenty-eight days after the injections, mice were sacrificed and
the brains were harvested for the assessment of adult hippocampal
neurogenesis. Several neurogenesis markers were selected for IHC
staining including KI67, TBR2, NeuroD, and DCX. Brain-specific Stk24
cKO mice were found to have decreased numbers of Ki67^+
(t[(16)] = 1.81, p = 0.08), TBR2 (U = 10.00, p < 0.01), NeuroD^+
(U = 0.00, p < 0.01), and DCX^+ (t[(16)] = 3.19, p < 0.01) cells in the
dentate gyrus of the hippocampus, where the neurogenesis occurs,
indicating impaired adult hippocampal neurogenesis (Fig. [81]4a). To
further confirm whether cKO mice had fewer newly born neurons, we
double-labeled the NeuN and BrdU by immunofluorescent (IF) staining,
and a significant reduction of co-localized NeuN^+/BrdU^+ cells was
observed in Stk24 cKO mice (t[(18)] = 3.29, p < 0.01) (Fig. [82]4b),
confirming the decreased adult hippocampal neurogenesis.
Fig. 4. Brain-specific Stk24 deletion decreased adult hippocampal
neurogenesis.
[83]Fig. 4
[84]Open in a new tab
a Significant decreases were found in the number of TBR2^+ (n = 8–10),
NeuroD^+ (n = 5–6), and DCX^+ (n = 8–10) cells in the dentate gyrus of
Stk24 cKO mice as compared to WT, while no difference was found in the
number of KI67^+ cells (n = 8–10). Black dots represent male mice,
while red dots represent females. Unpaired t-test for KI67, TBR2, and
DCX. Mann–Whitney U test for NeuroD. Scale bar = 100 µm. b Stk24 cKO
mice exhibited a significantly lower level of colocalized NeuN^+/BrdU^+
cells than control WT (n = 9–11). Black dots represent male mice, while
red dots represent females. Unpaired t-test. Scale bar = 10 µm. The
data were presented as mean ± SEM for each group. *p < 0.05,
**p < 0.01, ***p < 0.001.
Brain-specific Stk24 deletion altered HPA reactivity but not the downstream
GR signaling pathway
Dysregulated plasma cortisol and altered stress response are important
features of anxiety disorders^[85]13. Since Stk24 cKO mice exhibited
increased anxiety-like behavior, we further investigated the
hypothalamus-pituitary-adrenal (HPA) axis reactivity to see whether the
regulation of corticosterone had been altered. As the two most major
stress hormones, corticosterone and cortisol are considered as
indicators for stress response and are regulated by the HPA axis, with
the former being the main stress hormone in rodents and the latter
being predominant in humans^[86]14,[87]15. In fact, plasma
corticosterone is one of the major factors to influence adult
hippocampal neurogenesis and anxiety^[88]16,[89]17. We measured the
plasma corticosterone levels at three different time points: under
basal condition, after 30 min of restraint stress, and after 60 min of
recovery from restraint stress. Our results showed significantly higher
corticosterone levels in Stk24 cKO mice under the stressed condition as
compared to control WT (t[(16)] = 4.61, p < 0.001). However, there was
no significant difference under basal (t[(16)] = 0.40, p = 0.69) nor
recovered (t[(16)] = 0.23, p = 0.81) conditions between the groups
(Fig. [90]5a). Moreover, we also assessed the hippocampal neuronal
activation under basal and stressed conditions, respectively, since the
hippocampus was the upstream of HPA axis and could mediate the HPA axis
reactivity^[91]18,[92]19. We performed IHC staining for the detection
of c-FOS, an immediate early gene known as the molecular marker of
neural activity, in the hippocampus^[93]20. Consistent with the
corticosterone levels, IHC staining showed an increased number of
c-FOS^+ cells in Stk24 cKO mice under stressed condition (U = 7.00,
p = 0.02), while equivalent numbers were measured between WT and cKO
groups under the basal condition (U = 16.50, p = 0.55) (Fig. [94]5b,
c). In summary, brain-specific Stk24 deletion caused increased neural
activity in the hippocampus under stressed conditions, along with a
higher plasma corticosterone level. Yet no significant difference was
found under basal nor recovered states at the time points measured.
Fig. 5. Brain-specific Stk24 deletion altered HPA reactivity but not the
downstream GR signaling pathway.
[95]Fig. 5
[96]Open in a new tab
a Stk24 cKO mice had a significantly higher level of plasma
corticosterone under stress, while there was no significant difference
between WT and cKO mice under the basal and recovered conditions
(n = 8–10). WT: 4 males, 6 females; cKO: 6 males, 2 females. Unpaired
t-test for each condition. b The c-FOS expressions were comparable
between WT and Stk24 cKO mice under basal condition (n = 6–7). Black
dots represent male mice, while red dots represent females.
Mann–Whitney U test. Scale bar = 100 µm. c Stk24 cKO mice displayed
increased number of c-FOS^+ cells under stress as compared to WT
(n = 6–8). Black dots represent male mice, while red dots represent
females. Mann–Whitney U test. Scale bar = 100 µm. d Either under basal
condition (n = 6–8) or stressed condition (n = 6–8), Stk24 was
confirmed deleted in cKO mice; Crh was found upregulated in cKO mice in
both basal and stressed states, while most of the other gene
expressions were equivalent between WT and cKO mice, except for Sgk1.
Basal WT: 4 males, 4 females; Basal cKO: 2 males, 4 females; Stress WT:
2 males, 4 females; Stress cKO: 4 males, 4 females. Mann–Whitney U test
for each gene. The data were presented as mean ± SEM for each group.
*p < 0.05, **p < 0.01, ***p < 0.001.
In our data, Stk24 exerted a remarkable influence on anxiety and HPA
reactivity. To further understand the possible mechanism at the
molecular level, we investigated the glucocorticoid receptor (GR)
signaling pathway in the hippocampus. Although the GR signaling is
famous for its connection with the HPA axis reactivity in the
paraventricular nucleus (PVN) of the hypothalamus, the GR signaling in
the hippocampus is more likely associated with anxiety and stress, and
it is also where the highest levels of glucocorticoid binding
occured^[97]21–[98]24. Hippocampal tissues were collected from basal
group and stress group. Real-time PCR was then performed to quantify
the mRNA expression levels of GR signaling pathway-related genes
including Crh and Crhr1, which were demonstrated to play important
roles in emotional responses, and the GR downstream genes GR (Nr3c1),
Sgk1, Fkbp5, and Gilz^[99]25,[100]26 (Fig. [101]5d). As the Stk24
expression was confirmed deleted (basal: U = 0.00, p < 0.001; stress:
U = 0.00, p < 0.001), a great increase in the Crh, but not Crhr1,
expression was detected in Stk24 cKO mice in both basal and stress
groups (basal Crh: U = 2.00, p < 0.01; basal Crhr1: U = 12.00,
p = 0.14; stress Crh: U = 0.00, p < 0.001; stress Crhr1: U = 20.00,
p = 0.63). Nevertheless, no significant difference was found in most of
the GR signaling-related genes (basal GR: U = 24.00, p > 0.99; basal
Fkbp5: U = 14.00, p = 0.22; basal Gilz: U = 22.00, p = 0.83; stress GR:
U = 18.00, p = 0.47; stress Fkbp5: U = 11.00, p = 0.10; stress Gilz:
U = 15.00, p = 0.27), except for Sgk1 (basal: U = 5.00, p = 0.01;
stress: U = 14.00, p = 0.22). These results suggested an alteration in
the Crh and Sgk1 signaling caused by the Stk24 deletion. Put together,
Stk24 altered the neural activity in the hippocampus under stressed
conditions and may consequently mediated the HPA axis reactivity and
the downstream gene expressions.
Discussion
In this study, we set out to investigate the roles of Stk24 in the
hippocampus by testing brain development, adult hippocampal
neurogenesis, animal behavior, and stress response. Brain-specific
Stk24 cKO mice were used in this research. When examining the neuronal
morphology and postnatal neurogenesis, ectopic neurons were found in
Stk24 cKO mice during postnatal brain development, while the neuronal
morphology including dendritic spine density, dendrite branching, and
dendrite length, remained unaffected. The decreased adult hippocampal
neurogenesis was also detected during their adulthood, and these
phenotypes were in consistent with the RNA-sequencing findings, where
neurogenesis, neuronal migration, and pattern specification were
indicated as the down-regulated pathways in Stk24 cKO mice. As for
animal behavior, Stk24 cKO mice exhibited increase anxiety, but not
impaired cognitive abilities. When we further looked into the possible
mechanisms behind the increased anxiety, a higher level of
stress-induced plasma corticosterone was measured, indicating altered
HPA axis reactivity. The neural activity in the hippocampus, which is
known to mediate the HPA axis, was also shown to be elevated after
acute stress. Meanwhile, altered expressions of the GR
signaling-related genes, Crh and Sgk1, were found in the hippocampus of
cKO mice. In conclusion, brain-specific Stk24 deletion disrupted the
hippocampal formation during brain development and adult hippocampal
neurogenesis. The absence of Stk24 also increased anxiety-like behavior
and altered the stress response including neural activity, HPA axis
reactivity, and GR signaling-related gene expressions.
It has become increasingly evident that Stk24 plays an essential role
in neuronal development. Several previous studies have demonstrated
that Stk24 is important for cortical neuronal migration and hippocampal
neuronal filopodia. It is also reported to regulate axonal regeneration
in both central nervous system and peripheral nervous systems by
shRNA^[102]5–[103]7. In our study, Stk24 cKO mice featured ectopic
granule cells in the dentate gyrus during hippocampal formation.
Intriguingly, unlike previous result^[104]5, we failed to detect a
significant reduction in the spine density in cKO mice. No significant
difference was detected in neuronal morphology by Sholl analysis
either. While many confounding factors may lead to such inconsistency,
the biggest difference is that we used “gene knockout mice” instead of
“shRNA-mediated knockdown”. Moreover, for adult hippocampal
neurogenesis, there was no evident histological changes between adult
WT and Stk24 cKO mice, unlike what we had observed during brain
development. Other factors might be involved and probably compensate
for the effects of Stk24 deletion after brain development. Despite
normal hippocampal anatomy, decreased level of adult hippocampal
neurogenesis was found, including reduced number of KI67^+, TBR2^+,
NeuroD^+, DCX^+, and NeuN^+/BrdU^+ cells. This suggested that the
reduced number of proliferating cells may be the cause of neurogenesis
deficit in Stk24 cKO mice. Past studies have suggested a connection
between hippocampal neurogenesis and cognitive functions and/or
anxiety^[105]12,[106]27; yet some suggested otherwise^[107]28. In our
results, brain-specific Stk24 cKO mice were found to exhibit
anxiety-like behavior, while the cognitive abilities and the
electrophysiological characteristics were not significantly affected.
Since we deleted Stk24 in the whole brain, it is unclear whether the
behavioral phenotypes were resulted from the altered hippocampal
neurogenesis caused by Stk24 deletion. We could not rule out the
possibility that other brain regions such as cortex or amygdala may
also play a role in it. Further investigation is needed to address
this.
Although the link between anxiety-like behavior and altered adult
hippocampal neurogenesis remains unclear, we examined other mechanisms
known to be related. To this end, we measured the stress response, from
the HPA axis reactivity to the molecular changes of GR signaling. Under
basal conditions, mice with Stk24 deletion exhibited an upregulation of
Crh and a downregulation of Sgk1 in the hippocampus. A past study has
indicated that mice with corticotropin-releasing hormone (CRH)
overproduction displayed anxiogenic behavior^[108]29. The elevated Crh
expression in Stk24 cKO mice might explain the anxiety-like behavior;
however more evidence is required. Sgk1 is a serine/threonine kinase,
which has been shown to play a role in neurogenesis, memory, neuronal
plasticity and dendritic growth^[109]30–[110]33. The activity of SGK1
is down-regulated in post-mortem brains in post-traumatic stress
disorder, and inhibition of SGK1 results in anhedonic–like behavior in
rats^[111]34, suggesting a role of Sgk1 in emotional response. In our
result, we detected lower levels of hippocampal Sgk1 expression in cKO
mice. While no significant impairment was observed in the cognitive
abilities and neuronal plasticity, it remained unclear whether the
decreased Sgk1 level was involved in anxiety-like behavior. The role of
Sgk1 in Stk24 deleted mice remains to be answered. Furthermore, when
exposed to acute stress, Stk24 cKO mice displayed significantly
increased Crh expression and increased neural activity in the
hippocampus. Consistently, the stress-induced corticosterone was higher
in cKO mice as compared to WT, while no significant differences were
detected in the downstream regulation of GR. Based on current data,
Stk24 influences the expression of Crh under basal and stress
conditions, and the deletion of Stk24 could disrupt the stress
response. However, the specific mechanism remains undetermined.
In conclusion, several previous studies have demonstrated the
importance of Stk24 in neuronal development, migration, and
regeneration. Herein, we used brain-specific Stk24 cKO mice to provide
evidence that Stk24 played a role in modulating anxiety, hippocampal
neurogenesis, and stress response including HPA axis reactivity, Crh
expression, and Sgk1 expression. Further understanding of the
underlying cellular mechanisms of Stk24 is needed to interrogate how
the brain Stk24 plays its role.
Materials and methods
Animals
Stk24 conditional knockout mice were obtained by crossing Stk24 floxed
mice (National Laboratory Animal Center, Taiwan) with Sox1::Cre mice
(Access No. CDB0525K)^[112]8. Mice were housed in individually
ventilated cages (IVC) in the AAALAC certified animal facility, where
they were maintained in a 12:12 h light-dark cycle at a temperature of
22 °C and a humidity level of 60–70%. Animals had ad libitum access to
food and water. We have complied with all relevant ethical regulations
for animal use. All procedures were carried out in accordance with the
local regulations and approved by the Institutional Animal Care and Use
Committee at Chang Gung University (Permit Number: CGU112-116). All
following experimental details were previously
described^[113]35,[114]36. The sex of mice for each experiment were
indicated in different colors in the figure, or were described in the
figure captions.
mRNA quantification
Hippocampi were isolated and homogenized in TRIzol^TM reagent
(Invitrogen, 15596026, CA, USA), followed by Phenol-Chloroform
extraction of total RNA. cDNA was then synthesized using Modified MMLV
Reverse Transcriptase (Protech, Taipei, Taiwan). Gene expression levels
were calculated with the ΔΔCt method and normalized against a Gapdh
control. Experiments were performed in duplicate.
The table below shows the primer sequences used for real-time PCR.
Forward Reverse
Stk24 CAGCTGACGGATACCCAGAT CAGTGTGGGAGGGTTGTTCT
Crh AGCCCTTGAATTTCTTGCAG AGCCCTTGAATTTCTTGCAG
Crhr1 AGCCCTTGAATTTCTTGCAG CTGCCATCCGGAAGAGGT
GR AGGCGATACCAGGATTCAGA GCAAAGCATAGCAGGTTTCC
Sgk1 CTGCTCGAAGCACCCTTACC TCCTGAGGATGGGACATTTTCA
Fkbp5 AACGGAAAGGCGAGGGATAC ACACCACATCTCGGCAATCA
Gilz AACACCGAAATGTATCAGACCC GTTTAACGGAAACCAAATCCCCT
Gapdh TGCACCACCAACTGCTTAGC GGCATGGACTGTGGTCATGAG
[115]Open in a new tab
Western blot
Hippocampi were isolated and homogenized in 1% SDS solution at 100 °C
for 5 min, followed by the quantification of protein extracts using a
Bradford protein assay. Equal amounts of protein samples were separated
by 10% SDS–polyacrylamide gel electrophoresis and were transferred to a
polyvinylidene difluoride (PVDF) membrane (Millipore, Cork, Ireland).
Membranes were incubated with 5% non-fat milk for 1 h at room
temperature and then with diluted primary antibodies (STK24, A10576,
ABclonal, MA, USA; GAPDH, sc-32233, Santa Cruz, TX, USA) in 5% non-fat
milk at 4 °C overnight. After being washed with TBST, membranes were
incubated with diluted horseradish peroxidase (HRP)-conjugated
secondary antibody (goat anti-rabbit IgG, bs-0295G-HRP, Bioss, MA, USA;
goat anti-mouse IgG, bs-0296G-HRP, Bioss, MA, USA) for 1 h at room
temperature. Blots were visualized using Immobilon Western
Chemiluminescent HRP Substrate (WBKLS0500, Millipore, MA, USA) and were
detected with a ChemiDoc XRS+ imaging system (Bio-Rad). The protein
signals were quantified and normalized against GAPDH control using
Image-J software.
Golgi staining
Brains were harvested from postnatal day 13 (P13) mice and were
impregnated with Golgi-Cox solution for 7 days, followed by 2 days of
dehydration in sucrose. Brain sections at a thickness of 150 µm were
collected using a microtome and were then stained using a superGolgi
Kit (003010, Bioenno Tech, CA, USA) in accordance with the
manufacturer’s instructions. For the quantification of spine density,
we selected 36 dendrites per animal from the granule cells in the
dentate gyrus of the hippocampus and calculated the number of dendritic
spines. Spine density was then measured as the number of dendritic
spines divided by the length of chosen dendrites. Sholl analysis was
performed using the Neurolucida software (MicroBrightField Bioscience,
Williston, VT, USA).
Immunohistochemistry, immunofluorescence, and quantification
Brains were harvested and fixed in 4% paraformaldehyde overnight,
followed by dehydration in 25% sucrose at 4 °C. Brain cryosections at a
thickness of 40 µm were collected and stored in 60% glycerol before
staining.
For immunohistochemical staining, brain sections were mounted on
SuperFrost Plus slides (Thermo, PA, USA) and air-dried overnight. The
slides were incubated with 0.01 M citric acid buffer (pH = 6.4) for
20 min at 95 °C (40 min in the case of KI67), 3% H[2]O[2] for 5 min,
and then diluted primary antibody (PAX6, AB2237, Millipore Sigma, MO,
USA; TBR2, ab23345, Abcam, Cambridge, UK; PROX1, AB5475, Millipore
Sigma, MO, USA; NeuroD, sc-1084, Santa Cruz, TX, USA; KI67, ab16667,
Abcam, Cambridge, UK; DCX, ab18723, Abcam, Cambridge, UK; c-FOS:
sc-8047, Santa Cruz, TX, USA) at room temperature overnight. Slides
were rinsed with PBS for 15 min (3 times) between each step.
Subsequently, a standard IgG ABC kit (PK-6101, Vector Labs, CA, USA)
was used according to the manufacturer’s instructions, and the slides
were incubated with 3,3’-Diaminobenzidine (DAB) tablets (D4293-50SET,
Millipore Sigma, MO, USA). Sections were then counterstained with
hematoxylin and mounted with a DPX mountant (06522, Millipore Sigma,
Madrid, Spain). Experimental details were described previously^[116]37.
For the immunofluorescent staining for NeuN/BrdU double labeling, brain
sections were incubated with 2 N HCl for 15 min at 37 °C, neutralized
with boric acid (pH = 8.5) for 5 min at room temperature, and washed
with PBS for 15 min before the incubation with diluted primary antibody
(NeuN, MAB377, Millipore Sigma, MO, USA; BrdU, ab6326, Abcam,
Cambridge, UK) for 8 hr at room temperature. After being rinsed with
PBS for 15 min, brain sections were incubated with the diluted
fluorescent secondary antibody (Alexa Fluor 488: A21121, Invitrogen,
OR, USA; Texas Red: A11007, Invitrogen, OR, USA) for 2 h at room
temperature. Sections were then washed with PBS for 15 min and mounted
on SuperFrost Plus slides (Thermo, PA, USA), air-dried for 20 min, and
mounted with Fluoromount-G (100-20, SouthernBiotech, AL, USA).
For quantification, all sections were examined under a microscope with
a magnification of 200x. Antibody-stained cells were counted in the
bilateral dentate gyrus of the hippocampus every eight sections through
the entire extent of the granule cell layer (six sections per mouse).
The number of cell counts was then multiplied by eight to obtain an
estimate of total antibody-stained cells in the region of interest
(ROI). The ROI area was measured using ImageJ and then multiplied by
the brain slice thickness to estimate its volume. Data represented as
the number of antibody-stained cells normalized against the ROI volume.
RNA-sequencing
Library preparation and sequencing
Purified RNA was used to prepare sequencing libraries using the TruSeq
Stranded mRNA Library Prep Kit (Illumina) following the manufacturer’s
instructions. Briefly, mRNA was purified from 1 µg of total RNA using
oligo(dT)-coupled magnetic beads and fragmented at elevated
temperatures. First-strand cDNA was synthesized with reverse
transcriptase and random primers, followed by second-strand cDNA
synthesis to generate double-stranded cDNA. The cDNA fragments were
adenylated at their 3’ ends, and adaptors were ligated. The library
products were enriched via PCR and purified using the AMPure XP system
(Beckman Coulter). The libraries were validated with the Qsep400 System
(Bioptic Inc.) and quantified using a Qubit 2.0 Fluorometer (Thermo
Scientific). Sequencing was performed on the Illumina NovaSeq X Plus
platform with 150 bp paired-end reads generated by Genomics, BioSci &
Tech Co., Taiwan.
Behavioral testing
Behavioral tests were carried out on mice at the age of 8–12 weeks and
were conducted during the mouse active phase. Mice were randomized
between WT and cKO to minimize the current environmental effects. The
experimental details were described below.
Grip
The grip strength meter (47200, UGO BASILE, VA, Italy) consisted of a
grasping bar which was fitted to a forced transducer connected to the
peak amplifier. Mice were allowed to grip the grasping bar while their
tails were held and slightly pulled by the experimenters. The maximal
grip force was recorded. Each mouse was tested three times and the
average grip force was recorded.
Rotarod
Mice were placed on the rotarod apparatus (UGO BASILE, VA, Italy) and
the latency for them to fall off the rod was recorded manually. The rod
rotated at an initial speed of 4 rpm and an accelerated speed of 9 rpm
per minute. Each mouse was tested three times and the average falling
latency was recorded.
Locomotor activity
Mice were individually placed in the cage and had ad libitum access to
food and water. The animals were maintained in the behavioral room in a
12:12 light-dark cycle under 24 h continuous recording. The total
travel distance was measured using Ethovision software.
Open field
Mice were allowed to explore in a circular-shaped area (radius = 60 cm)
for 5 min with a lamp illuminating the center. The total distance, the
time spent in the center area (radius = 40 cm), and the frequency of
entering the center area were recorded by Ethovision software.
Elevated O-maze
The O-shaped maze (radius = 55 cm) was elevated 60 cm above the floor
and consisted of the open arms and the closed arms. The closed arms
featured 15 cm-high walls. Mice were placed in the closed arm and
allowed to freely move for 5 min. The total distance, the time spent in
the open arms, and the frequency of entering open arms were measured by
Ethovision software.
Light-dark box
The light-dark box was divided into a covered dark chamber and a
coverless light chamber. Mice were put into the covered dark chamber
and allowed to freely move between the two chambers via a door for
5 min. The time spent in the light chamber and the frequency of
transitions were recorded by Ethovision software.
Morris water maze
A circular water tank (radius = 130 cm) was filled with 22 °C water,
and a circular platform (radius = 10 cm) was located inside the tank,
submerged 1.5 cm below the water surface. The water had been colored
with white acrylic turpentine to cover the platform. Mice were put into
the tank at three different starting positions, which refer to the
three quadrants other than the quadrant with the platform. Each mouse
was trained to search for the hidden platform and stayed on it for
30 s, three trials a day for five consecutive days. The time each mouse
used to find the platform was recorded using Ethovision software. Seven
days after the last training, a one-minute memory test was performed
with the platform removed. The time each mouse spent in the target zone
(quadrant where the platform was previously) was recorded by Ethovision
software.
Y-maze
Mice were placed in a Y-shaped maze and were allowed to freely move for
5 min. An alteration was defined as consecutive entries into three
different arms without any repeats, while the max alteration indicated
the maximal possible alterations according to the number of total
entries, both of which were analyzed using Ethovision software. The
percentage of alteration (%) was calculated as the number of
alterations divided by max alteration.
Novel object recognition
Mice were allowed to freely move in the empty chamber (30 × 30 × 25 cm)
for 10 min on day 1. Then on day 2, mice were exposed to two identical
objects in the chamber and were also allowed to freely explore for
10 min. On day 3, one of the objects was replaced with a novel one.
Mice were allowed to explore for 10 min. The time mice spent to explore
the objects were measured using Ethovision software, and the
discrimination index (D.I.) was calculated according to the below
formula.
[MATH:
D.I.=(Timeexploringnovelobject−Timeexploringfamiliarobject)(Timeexploringnovelobject+Timeexploringfamiliarobject)×100% :MATH]
For the objects used in the test, the two identical ones were white
plastic cylinders, with a diameter of 4 cm and a height of 6 cm. The
novel object introduced on day 3 was a yellow, dumbbell-shaped plastic
piece with equilateral triangles attached to each end, each triangle
having a side length of 4.5 cm and the total height of the object was
5.2 cm.
Electrophysiology
A bipolar stainless steel stimulating electrode (Frederick Haer
Company, ME, USA) (10 Meg-ohm impedance) and a glass pipette filled
with 3 M NaCl were positioned in CA1. For LTP experiments, field
excitatory postsynaptic potential (fEPSP) was evoked every 20 s for the
first 20 min as baseline, and then a high-frequency stimulation (HFS)
which composed of five trains of 100 pulses at 100 Hz with an
inter-train interval of 20 s was introduced to trigger LTP. fEPSP
activity was subsequently recorded for 1 h to assess LTP. For paired
pulse, it was determined for five inter-pulse intervals (50, 100, 150,
200, 500, 1000 milliseconds). Paired of pulses were repeated four times
for each inter-pulse interval, and the paired pulse ratio was
calculated as the second response divided by the first response.
Electrophysiological traces were amplified with an amplifier
(Multiclamp 700B: Axon Instruments, CA, USA). All signals were
low-pass–filtered at 1 kHz and digitized at 10 kHz using a CED Micro
1401 mKII interface (Cambridge Electronic Design, Cambridge, UK). Data
were collected using Signal software (Cambridge Electronic Design,
Cambridge, UK). Synaptic responses were normalized to the average of
the baseline.
Corticosterone assay
Blood samples were collected by facial vein puncture at three different
time points: under baseline (basal) conditions, after 30 min of
restraint stress, and after 60 min of recovery from restraint stress.
Plasma was separated from whole blood by centrifugation (3000 rpm,
4 °C, 15 min) and stored at −80 °C until used. Plasma corticosterone
concentration was measured using the Corticosterone ELISA kit
(ADI-900-097, Enzo Life Sciences, NY, USA) according to the
manufacturer’s instructions.
Statistics and reproducibility
The data were presented as mean ± SEM for each group. *p < 0.05,
**p < 0.01, ***p < 0.001. Statistical analysis was performed using
Graphpad Prism 6.0 software. All data were tested for normal
distribution using D’Agostino and Pearson omnibus normality test
(Supplementary Table [117]1) and were further analyzed via unpaired
t-test, Mann–Whitney U test, or an analysis of variance (ANOVA) as
appropriate. The sample size and the number of replications were
determined by referencing previous studies in the same field. All
experiments were replicated at least once or replicated on more than
one cohort of mice. Replication attempts were successful and the
observed results were consistent.
Reporting summary
Further information on research design is available in the [118]Nature
Portfolio Reporting Summary linked to this article.
Supplementary information
[119]Supplementary information^ (808.4KB, pdf)
[120]42003_2025_8035_MOESM2_ESM.pdf^ (28.1KB, pdf)
Description of Additional Supplementary Files
[121]Supplementary data^ (67.8KB, xlsx)
[122]Reporting summary^ (1.5MB, pdf)
Acknowledgements