JEM: Enhances neurogenesis to save Alzheimer's disease

Guide:

Hippocampal nerve damage
occurs in Alzheimer's disease (AD) patients and mouse models of familial Alzheimer's disease (FAD).
However, whether neonatal neurons play a causal role in memory impairment is unclear
.
Recently, Orly Lazarov's team published a research paper
in the JOURNAL OF EXPERIMENTAL MEDICINE "Augmenting neurogenesis rescues memory impairments in Alzheimer's disease by restoring the memory-storing neurons" 。 The authors found that after the hippocampus-dependent memory task, immature neural individuals were actively recruited into the
engram.
However, this recruitment is severely inadequate
in FAD mice.
Recruited immature neurons exhibit impaired ridge density and altered transcriptional profiles
.
Directed enhancement of neurogenesis in FAD mice restored the number of new neurons in the engram, dendritic spike density, and transcriptional signatures of immature and mature neurons, ultimately leading to memory recovery
.
The chemical genetic inactivation of immature neurons after neurogenesis enhancement in AD reversed the performance of the mice and reduced memory
.
AD-linked App, ApoE, and Adam10 are the highest differentially expressed genes in engrams, suggesting that neurogenic defects lead to memory impairment
in AD patients.

More new neurons in DG in FAD mice led to better performance of hippocampus-dependent memory tasks

To investigate whether enhancing hippocampal neurogenesis in adults can save the learning and memory deficits of FAD, the authors designed a mouse model
of FAD with inducible neurogenesis.
5xFAD mice and wild-type mice were combined with NestinCreERT2; B a xfl/fl mouse hybridization (Figure 1A).

Female NestinCreERT2; Baxfl/fl(NB) or NestinCreERT2; Baxfl/fl; 5xFAD (NBF) mice are treated with tamoxifen (T-NB or T-NBF) or corn oil (C-NB or C-NBF, respectively; Figure 1 B).

Quantitative polymerase chain reaction confirmed a recombination-induced bax gene deletion of 75% (Figure S1A).

To support previous reports, deletion of the Bax gene enhances the survival of neural precursor cells (NPCs) and leads to an increase
in neurogenesis.
There is a tendency to increase the number of proliferative neuroblasts after recombination, but not significantly (Figure S1B), probably because the loss of Bax improves the survival rate of these cells, not their proliferation rate, so it is mainly manifested as an increase
in the number of immature neurons and neonatal neurons.
Notably, the number of proliferating neuroblasts in C-NBF mice was significantly reduced compared to C-NB mice, supporting the view that AD patients had reduced neurogenesis (Figure S1B).

Compared to age- and sex-matched C-NB mice, immature neurons of T-NB (bicortisin [DCX]+; Figure 1 C) is significantly higher in number and has a higher survival rate for new neurons (BrdU+NeuN+; Figure 1D), and more immature neurons (DCX+NeuN+; Figure 1, E and F).

Compared with C-NB, there were significantly fewer immature neurons and a reduced survival and number of new neurons (Figure 1, C-F), validating this model to reproduce the damage
that occurs in the hippocampal nerve in 5xFAD.
Importantly, T-NBF mice have significantly more immature and new neurons compared to C-NBF (Figure 1, C-F).

To investigate whether increased neurogenesis could save FAD mice from memory impairment and whether immature neurons were actively involved in memory malformations in the disease, four groups of mice underwent a new object localization (NOL) test
.
C-NB and T-NB behave similarly, spending 60% of their time exploring objects in new locations (Figure 1, G and H).

Again, their discrimination indices are similar (Figure 1I).

This may suggest an upper limit effect, or that behavioral paradigms have limited sensitivity
in detecting discrete improvements in memory performance.
Notably, C-NBF mice fail the test, preferably exploring the old location, exhibiting a negative discriminant index (Figure 1, H and I).

Importantly, T-NBF mice showed significantly improved performance, increased preference for new positions, and had a larger discrimination index similar to that of NB mice (Figure 1, H, and I).

To further characterize the effect of enhanced neurogenesis on AD memory, another group of C-NB, T-NB, C-NB, and T-NBF mice underwent different behavioral tests examining associative memory using the hippocampus-dependent CFC test (Figure 1J).

Comparable to their performance in the NOL trial, C-NB and T-NB mice were able to correlate the environment similarly well to shock, freezing an average of 40% on the trial day (Figure 1K).

C-NBF mice performed poorly in this trial, freezing an average of 20% on the trial day (Figure 1K).

Importantly, T-NBF mice had significantly better shock-related contexts when frozen on average 30% (Figure 1K).

To further confirm the above findings, we characterized anxiety levels
in mice by performing light and dark tests.
Four groups of mice showed a considerable degree of anxiety (Figure S1C).

Taken together, these results strongly suggest that enhanced hippocampal neurogenesis in FAD mice significantly improves spatial recognition and background memory
.

Fig.
1 Rescue of hippocampus-dependent memory after neurogenesis enhancement in FAD mice

Neonatal neurons play a role in the memory acquisition of FAD

To explore whether immature neurons rescue memories in FAD by participating in memory formation, the authors sought to track neurons
that were activated after memory acquisition.
To do this, the authors used a Tet-off-based viral engram labeling kit consisting of
a mixture of AAV9-cFos-tTA and AAV9-TRE-ChR2-eYFP.
Mice are started on a doxycycline diet 1 week before stereotactic injection of the mixture into DG and removed 18 h before giving a foot shock to allow enhanced expression of yellow fluorescent protein (eYFP) in neurons activated during memory acquisition (Figures 2, A and B).

Combined immunostaining with anti-c-fos antibodies validated eYFP+ neuronal expression of c-fos, indicating that these neurons are recruited into neuronal collections during memory acquisition and reactivated during memory recovery (Figures 2, B and D).

A previous study reported that certain AAV serotypes induced mass death of BrdU-labeled cells within 18 hours of injection, and there was no evidence of adult neurological recovery
within 3 months of injection.
To examine whether the viral engram labeling kit affects the degree of neurogenesis in the current study, the authors compared the number of surviving immature neurons (NeuN+BrdU+) in mice pre-injected with viral vectors (Figure 1D) and mice without injection (Figure S1D), and observed that the number of mice in the two groups was comparable
。 To further verify the specificity of the authors' labeling method, the cell numbers of EYFP+, Eg r-1+, E Y F P+Egr-1+, DCX+EYFP+Egr-1+, DCX+EYFP+Egr-1+ cells housed in caged mice were examined
.
The authors observed a significantly lower number of cells activated in DG in these mice compared to mice receiving CFC, suggesting that the observed recruitment of these cells into memory circuits is highly specific for
CFC.
To investigate whether new neurons are involved in memory acquisition, and whether increased neurogenesis in FAD mice leads to new neurons being recruited more into memory circuits, the authors first quantified the total number
of eYFP+ cells in the brains of four groups of mice.
Consistent with the behavior, the authors observed a similar number of cells in C-NB and T-NB recruited into the
blotting.
Therefore, further quantification focuses on the engram and its post-neurogenesis enhancement (T-NBF) state
in FAD mice with impaired behavior.
In C-NB, C-NB, and T-NBF mice, the total number of engram cells (i.
e.
, EYFP+ cells) recruited to engram after CFCs training is similar (Figure 2C).

To detect the total number of EYFP+ neurons, the authors costained
brain slices with anti-NeuN antibodies.
The number of NeuN+eYFP+ was observed to be similar to the total number of eYFP+ cells in all groups, suggesting that the cells recruited after DG acquired memory were neurons
.
The total number of NeuN+EYFP+ cells was similar between experimental groups (Figure 2E).

Interestingly, significant differences
were observed between the two groups in the number of new neurons.
The number of newborn neurons in C-NBF group mice (Neun+DCX+EYFP+) was significantly reduced compared to C-NB mice (Figure 2, F, I).

And the number of new neurons Neun+DCX+EYFP+ activated by T-NBF mice increased significantly (Figure 2F).

Similar results were observed when analyzing the survival of new neurons (Neun+BrdU+EYFP+) recruited into memory circuits in different groups (Figure 2G).

And compared with the C-NBF group and the T-NBF group, the percentage of new neurons activated in the C-NBF group to the total number of activated neurons was significantly reduced (%(Neun+DCX+EYFP+)/total EYFP+ cells; Figure 2 H).

Together, these results suggest that fewer immature neurons are recruited into the engram during memory acquisition in C-NBF mice, while enhanced neurogenesis results in recruitment of more immature neurons, comparable
to the number in C-NB mice.
In addition, these results suggest that a decrease in the number of immature neurons recruited to the engram is associated with impaired behavior in C-NBF, while in T-NBF mice, an increase in the number of immature neurons is associated
with intact behavioral performance.

Fig.
2 Enhancing neurogenesis to rescue immature neurons from re-entering the memory circuit

In FAD mice, neurons that were recruited when acquiring memories were less reactivated during retrieval

To further elucidate the role of new neurons in memory imprinting, the authors continue to explore whether immature neurons recruited during contextual memory acquisition are reactivated when that memory is reretrieved
.
To determine this, the authors examined levels of Egr-1 (Zif268), an immediate early gene
previously thought to be memory recovery.
This analysis showed that the total number of cells expressing the immediate early gene Egr-1 (Egr-1+ cells) after test-induced activation was similar in all three groups, suggesting that the total number of neurons recruited during memory recovery did not change due to the level of FAD genes or hippocampal neurogenesis (Figure 3A).

Notably, the number of DCX+Egr-1+ cells in C-NBF mice is significantly reduced compared to C-NB mice (Figure 3B).

In T-NBF mice, this number increases significantly with increased neurogenesis, suggesting that increased neurogenesis leads to the recruitment of more immature neurons during memory retrieval (Figure 3B).

The total number of EYFP+Egr-1+ cells recruited during memory acquisition and extraction was reduced in C-NBF mice and increased in T-NBF mice compared to C-NBF mice (Figure 3, C and H).

The number of immature neurons recruited during memory and retrieval (DCX+EYFP+Egr-1+) showed a decrease in the number of C-NBFs compared to C-NB, but not statistically significant, while the T-NBF group increased significantly (Figure 3, D and I).

Changes in the number of immature neurons (DCX+eYFP+Egr-1+) recruited during memory retrieval between the three groups were directly related to
their recruitment during acquisition.
Apparently, the number of immature neurons (DCX+EYFP−Egr-1+) that were activated during retrieval but not during acquisition showed no therapeutic or genotype effects (Figure 3E).

This result suggests that enhancing neurogenesis in particular increases the number of
new neurons involved in memory acquisition.
Notably, the number of mature neurons recruited during acquisition and the number of reactivations during extraction were similar between the three experimental groups (Figure 3F).

An examination of the proportion of immature and mature neurons participating in the neuronal combination in each experimental group showed that the number of immature neurons in the memory circuit of C-NBF was reduced compared to C-NBF, with only 11% of new neurons in DG of C-NBF and 20%
in C-NBF.
With the enhancement of neurogenesis, this ratio increases to 35% in T-NBF (Figure 3G).

Taken together, the study showed that in FAD mice, the number of immature neurons recruited into the collection of neurons for memory formation was affected
.
As a result, the total number of neurons recruited into the engram by FAD mice decreased, leading to memory loss
.
The enhancement of neurogenesis leads to an increase in the number of new neurons involved in memory formation, so that the memory is restored
.

Fig.
3 The number of immature neurons reactivated in episodic memory retrieval in FAD mice increases after enhanced neurogenesis

The density of synaptic spines of immature neurons in FAD mice participating in engram is enhanced after neurogenesis enhancement

Synaptic pathology is one of the earliest lesions in Alzheimer's disease and is associated with
memory impairment.
To elucidate whether enhanced recruitment of new neurons in engrams after neurostimulation in T-NBF mice is associated with memory recovery, the authors looked at whether the spinal density of these cells changed
.
To answer this question, the authors quantified the density
of dendritic spines in DCX+EYFP+Egr-1+ cells in C-NB, C-NB, and T-NBF mouse DGs.
Consistent with the idea that dendritic spines play an important role in memory formation, dendritic spines are damaged in Alzheimer's disease, and the authors observed that newborn neurons within C-NBF have lower synaptic densities compared to C-NB (Figures 4, A and B).

And the synaptic density of new neurons within the T-NBF has been restored (Figures 4, A and B).

To gain insight into the types of spinous processes observed in the dendrites of Engram, the authors quantified the number of mushroom-like spinous processes and the density
of fine spinous processes.
The authors observed a decrease in the number of mushroom-like spines in the dendritic processes of C-NBF mice compared to C-NB, while the number of mushroom-like spines recovered in T-NBF mice (Figure 4C).

Similar trends were observed in the fine spinous process; However, the difference between C-NBF and T-NBF was not statistically significant (Figure 4D).

Quantification of the density of spines to the distance of the cell body shows that the densities of C-NBF and T-NBF are comparable, while the density of the spinous process of C-NBF is consistently low, significantly lower than that of C-NBF (Figure 4, E-G).

Interestingly, the authors observed that the enhancement of neurogenesis rescued the number of tertiary dendrites in the engram (EYFP+Egr-1+ cells; Figures A-C).

Together, these results support the role of
newborn neurons in memory recovery in T-NBF mice.

Fig.
4 Recovery of synaptic density in FAD mice after enhanced neural regeneration

Enhancement of neurogenesis restores synaptic spine density in mature granule neurons involved in engram in FAD

Given the effect of enhanced neurogenesis on the density of spinous processes in immature neurons participating in engrams, the authors wondered whether this process affects the density of spinous processes in mature granule neurons in DG that function in
engrams.
To examine this, the authors quantified the density and morphology of Neun+EYFP+Egr-1+ dendritic spines in the granular cell layer in brain slices of C-NB, C-NB, and T-NBF mice (Figure 5A).

Quantification of spinous density showed that Neun+EYFP+Egr-1+ neurons in DG in C-NBF group had lower spike density than in C-NB mice (Figure 5B).

The vertebral density of Neun+EYFP+Egr-1+ neurons in T-NBF mouse DG was observed to be comparable to that of C-NB mice (Figure 5B).

This result suggests that enhancing neurogenesis can involuntarily restore synaptic plasticity
in mature neurons within DG.
To examine whether specific forms of dendritic spinous processes are modulated in engrams, the authors quantified the density
of mushroom-like spinous processes, coarse spinous processes, and fine spinous processes.
The data showed that in C-NBF mice, the density of mushroom spines in Neun+EYFP+Egr-1+ neurons was significantly insufficient (Figure 5C).

Mushroom spinous density recovery in T-NBF mice (Figure 5C).

The density of coarse and fine spinous processes showed a similar trend, although not statistically significant (Figure 5, D and E), and examining the density of spines as a function of distance to the cell body compared to C-NBF and T-NBF found that the total vertebral density of mouse neurons in the C-NBF group was consistently damaged, independent of distance from the cell body compared to the C-NB and T-NBF groups (Figure 5F).

Compared to C-NB and T-NBF, defects in mushroom spinous density in C-NBF are more pronounced at > 10 μm from the cell body (Figure 5G).

Detection of the ratio of spinous process types in granule neurons in three groups of mice showed that most of the spinous processes in C-NB and T-NBF mouse granule neurons were mushroom-shaped (∼40%), while most of the C-NBF group were fine spinous processes (Figure 5H).

Next, the authors compared changes in spinous process density between immature and mature neurons after enhanced neural regeneration in
T-NBF.
The data showed that although both immature and mature neurons had an increase in spike density in both immature and mature neurons in T-NBF mice compared to C-NBF, the overall spike density of immature neurons was higher than that of mature neurons (Figure 5I).

While all three spinous process types in mature and immature neurons in T-NBF were increased compared to C-NBF (Figure 5, J-L), the density of short spinous processes in immature neurons in T-NBF mice increased significantly compared to C-NBF, but not in mature neurons (Figure 5 L).

Taken together, these results suggest that enhancing neurogenesis in FAD can restore spinous process density defects
in mature granular neurons in DG.

Fig.
5 Enhanced regulation of neurogenesis involved in synaptic plasticity of mature granule neurons in engram in DG

Immature neurons are necessary for the formation of normal memories

To address this, T-NBF mice were injected with a retroviral vector expressing a GI designer receptor exclusively activated by the designed drug (DREADD) RV-hM4Di-EGFP to specifically inactivate newly mature neurons 4 weeks prior to CFC (Figure 6A).

To verify the inactivation effect of RV-hM4Di-EGFP on infected cells, mice injected with RV-hM4Di-EGFP were treated with clozapine N-oxide (CNO) and the expression
of EGFP and endogenous c-fos in their brain tissues was detected.
The authors observed no overlap in the expression of EGFP and endogenous c-fos, confirming that RV-hM4Di-EGFP can inactivate infected cells (Figure 6, B-E).

To further validate the specificity of DREADD in several mice, the authors tested the putative side effects
of the actuator treatment in mice not injected with the Gi fear receptor.
The authors observed no change in
their freezing levels (Figures S3, D, and E) or the effect of the actuator or injection site (Figures S3, F-I).
Next, the authors investigated the effect of
inactivating immature neurons on memory formation in CFCs in T-NBF mice.
For this, T-NBF mice are stereotactically injected with RV-hM4Di-EGFP or control virus (RV-EGFP)
4 weeks before CFC4wk.
5 days before the behavioral test, treat mice with an actuator or water (Figure 6A).

T-NBF mice injected with RV-hM4Di-EGFP treated with an actuator significantly reduced memory in these mice (Figure 6F).

Moreover, inhibition of the activity of new neurons during CFC in RV-hM4Di-EGFP-injected wild-type mice, C-NB, or T-NB mice does not affect the performance of the mice (Figure 6, G-I).

The results in wild-type and C-NB mice support previous observations in thymidine-kinase-expressing mice that the absence of new neurons does not affect the performance of
CFCs.
The lack of effectors in T-NB mice is consistent with C-NB and T-NB mice exhibiting similar performance on CFC tasks, and suggests that additional immature neurons in T-NB mice are not essential for functional CFC engrams, supporting the authors' hypothesis
that there is a ceiling effect in these mice.
These results suggest that normal memory formation in T-NBF mice requires more immature neurons, and their lack in C-NBF underlies
memory impairment.
In T-NBF mice, increasing their number saved memories, and in turn, their inactivation disrupted engram and memory formation
.
All in all, increasing neurogenesis in FAD mice increased the availability of new neurons, integrating them into situational engram cells, leading to correct engram formation and full performance
in memory tasks.
In addition, immature neurons are favorably reactivated during memory retrieval, which is necessary to
save the engram in FAD.
Notably, the authors show that in addition to immature neurons, enhancing neurogenesis affects synaptic plasticity
in mature granule neurons involved in memory engrams.

Figure 6 The formation of CFC memories in FAD requires the participation of new neurons

Enhancing the neurogenesis of FAD resulted in engram transcription features similar to wild-type

The results so far suggest that enhancement of neurogenesis regulates hippocampal function
in FAD (T-NBF) mice.
Therefore, the authors sought to study the signaling pathways of engrams in FAD compared to wild-type mice, as well as their alterations
after neurogenesis enhancement.
In addition, the authors investigated whether the enhancement of neurogenesis affects the profile
of granule cells in DG.
Therefore, the authors studied the transcriptional characteristics
of immature and mature neurons recruited in DG in C-NB, C-NB, and T-NBF mice.
To do this, C-NB, C-NB, and T-NBF mice were injected with an engram mixture, placed on a doxycycline diet, and received CFC, as previously described
.
After 45 minutes of the test phase of the CFC, the brains were removed from the rodent and the brain tissue was frozen
.
Coronal sections were analyzed by spatial transcriptomics (Figure 7A).

Simultaneous single-cell resolution sequencing of 159 genes of interest was performed using in situ sequencing (Figure 7, A-E).

To assign readings to individual cells, cell segmentation
is performed using a customized MatLab script (Figure 7, F-L) for images of brain sections stained with DAPI nuclei.
Quantification of the number of cells showed that the average total cell count in the three sets of brain slices was similar
.
The total number of cells in DG showed 6460±235 cells in the C-NB group±SE, 4619±325 in the C-NBF group, and 6450±313 in the T-NBF group
.
This may indicate that the number of cells in DG depends on the level of neural regeneration in
these mice.
To determine the engram in DG, the authors examined the expression of EYFP in a uniformly tracked region in DG, which includes the gate, subgranularity, and particle layer
of DG.
If a cell contains at least one EYFP read, it is defined as EYFP+
.
Fisher's Exact test (FET) is used to compare the proportion of
cells expressing each gene in a paired grouping of each cell type.
Figure 7M shows a random neighbor embedding (t-SNE) plot
of the t-distribution of expression patterns in DG for three experimental groups.

Figure 7 In situ sequencing of immature and mature engrams

To gain insight into the molecular mapping of engram neurons, the authors compared
the profiles of EYFP+ neurons with those of EYFP−neurons in the C-NB group.
In EYFP+ neurons, the expression of a series of genes critical for neuronal function, such as SYN1, NDNF, NCAM1, NPY2R, SLC6A5, OPRK1, MAPK3, and Gabra1, is upregulated in EYFP+ neurons (Figure 8A).

Interestingly, several AD-related genes such as App, ADAM10, and PSEN1 were also regulated (Figure 8A).

Some genes such as App, Fos, Npas4, Npy2r, Oprk1, Sst, Gul, Syn1, Slc17a8, ApoE
.
Mapk3, Adam10, Pvalb, and Gad2 are upregulated in eYFP+ mature neurons in C-NB mice than eYFP-mature neurons (Figure 8B).

In the new neurons of eYFP+, genes such as Neurod6, Src6a1, Src6a5, Src17a8, Ncam1, and Grin2b were upregulated compared to the new neurons of eYFP+ (Figure 8C).

Interestingly, the transcriptional profile of eYFP+ neurons in C-NBF mice is opposite to that of C-NB (Figure 8A).

In most cases, in the C-NBF engram, the expression of those genes that were substantially upregulated in the C-NB blot did not change significantly, while another group of genes that were not strongly regulated in C-NB, such as Fev, Wfs1, Map2, Vipr2, Ptprc, Arc, Egr-1, was upregulated in the C-NBF blot (Figure 8A).

It is worth noting that the distribution of engrams in T-NBF partially reflects the situation of C-NB (Figure 8A).

Examination of mature and new neuronal profiles in different groups showed similar trends (Figure 8, B, C).

Compared to eYFP-mature neurons in C-NBF, the characteristics of eYFP+ mature neurons are largely opposite those of C-NB mice; The expression of the T-NBF group was similar to that of the other groups (Figure 8B).

Similar to the expression of eYFP+ mature neurons, the expression of new eYFP+ neurons in C-NBF is the opposite of that in C-NB (Figure 8C).

Interestingly, gene expression of eYFP+ new neurons in T-NBF is similar to that of these cells in the C-NB group (Figure 8, C and D).

Genes that regulate neuronal function, such as Ncam1, have the lowest P values in new neurons in C-NB and T-NBF mice (eYFP+/eYFP-) (Figure 8E).

In summary, neurons recruited to Engram have unique gene expression profiles
compared to other peers.
In FAD mice, this feature varies greatly and is partially recovered
after neurogenesis enhancement.
After neurogenesis enhancement, the recovery of the Engram neuronal map is particularly pronounced
in new and total neurons.

Figure 8 The altered memory engram in FAD is restored after neurogenesis enhancement

AD-related signal conditioning engrams

Next, the authors compared the engram profiles between the three experimental groups, regardless of the profiles of the rest of the neurons in DG that were not recruited into the engram map
.
Studies of the distribution of DG and intragate EYFP+ neurons have shown a reduced distribution of EYFP+ neurons within C-NBF compared to C-NB and T-NBF (Figure 9, A-C).

Differential gene expression of EYFP+ neurons compared to C-NBF in C-NB (Figure 9, D and G) indicates that Mapk3 and ADAM10 are the most upregulated genes in C-NB relative to C-NBF; The largest downward revisions to c-NB relative to C-NBF were in Wfs1 and Nefh (Figure 9, D and G).

Compared to C-NBF, Camk2a, App, GLUL, Wfs1, Arc, and Lmo1 were the most downward in T-NBF (Figure 9, E and H).

Relative to C-NB, T-NB saw the most upward revisions in Grin1 and Syt6, while Camk2a, Mapk3, Rprm, PER1, and Pvalb made the most downward revisions (Figure 9, F and I).

The fractional expression in the total DG shows the random order of the experimental groups (Figures 9J and 5C).

And partial expression of EYFP+ cells showed similar expression patterns between T-NBF and C-NB groups, while C-NBF groups showed different expression patterns (Figure 9K and Figure 5C).

All 159 genes are listed in L-N in Figure 9 and sorted
by log2FC compared to C-NB/C-NBF (Figure 9 L), T-NBF/C-NBF (Figure 9M), and T-NBF/C-NB (Figure 9N).
Next, the authors sought to investigate whether the enhancement of neurogenesis affects the ratio
of excitatory neurons to inhibitory neurons in DG.
To do this, the authors quantified the number of mature and immature neurons that express genes
known to regulate inhibitory or excitatory activity.
The authors observed that the rates of inhibition and excitability were similar
in the three experimental groups.
Next, the authors looked at the distribution of immature neurons and mature neurons in these subpopulations, and whether these neurons played a role
in engram (EYFP+).
The data show that enhanced neurogenesis has the greatest
effect on excitatory immature and mature neurons.
In addition, the number of inhibitory eYFP-mature neurons in T-BN and C-BN is higher
compared to C-NBF.
However, all neurons defined as inhibitory express at least one indicator of
excitability.
Therefore, further experiments will be required to verify the function of
these neurons.

Fig.
9 Engram transcription profile after FAD and neurogenesis enhancement

To gain insight into the distribution of neonatal versus mature neurons in the engram map, the authors first examined their distribution based on their spatial transcription and observed that the distribution of both immature and mature neurons in the C-NBF group was reduced compared to the C-NB and T-NBF groups (Figure 10, A-C).

Next, the authors examined the average number of cells expressing the lowest P<0.
05 gene in each EYFP+ neuron type (Figure 10, D-L).

Among genes with P<0.
05, ADAM10, Mapk3, Gad1, 2, and N d Slc17a8 had the strongest regulatory effect in C-Nb/CNBF comparisons of EYFP+ neurons and EYFP+ mature neurons (Figure 10, D and E).

Among immature neurons with EYFP+, APOE, BDNF, Camk2a, Nefh, and n d NeuroD1 were most strongly expressed (Figure 10).

Interestingly, under T-NBF/C-NBF conditions, ADAM10 and App, Camk2a, GLUL, a d Lmo1 were most strongly expressed in EYFP+ total and mature neurons (Figure 10, G and H), while APOE, BDNF, Gabra1, Hmer 1, and MAPT were most strongly expressed in new neurons (Figure 10I).

Under T-NBF/C-NB conditions, Camk2a and Mapk3 are most regulated in mature and immature neurons (Figure 10, J-L).

Comparing the directionality of differentially expressed genes in C-NB/C-NBF and T-NBF/C-NBF engrams showed similar directionality (Figure 10, M-O).

。 Importantly, among the 20 genes with the lowest P values in the C-NB/C-NBF and T-NBF/C-NBF comparisons, eYFP+ neurons, eYFP+ mature neurons, and eYFP+ neonatal neurons had a statistically significant proportion of eYFP+ neurons in the C-NB/C-NBF and T-NBF/C-NBF comparisons, rather than being expected solely due to random chance (Figure 10P).

These results suggest that enhanced neurogenesis promotes similar genes
compared to wild-type (C-NB).

Figure 10 Enhancement of changes in new and mature engram transcription profiles in FAD DG after neurogenesis

Summary

The study provides several new observations
.
The first is direct evidence that immature neurons in DG play a role
in hippocampus-dependent engrams in wild-type and FAD mice.
Second, disorders of hippocampal neurogenesis lead to defects in the formation of engrams in FAD and lead to memory impairment
.
Third, more and more neurogenesis saves memories
by restoring memory engrams.
Fourth, in FAD, immature neurons are necessary for
the formation of normal memories.
Fifth, enhancing neurogenesis rescued the spinous process density deficit
of immature and mature engrams in FAD mouse DG.
Sixth, enhancing neurogenesis can modulate the characteristics of immature and mature engrams in DG, making them similar
to the transcriptional characteristics of engrams in wild-type mice.
Seventh, AD-related signals, particularly App, APOE, and ADAM10, function in the engram and are regulated
after neurogenesis enhancement and memory recovery.
Most importantly, this result suggests that the enhancement of neurogenesis can modulate the transcriptional profile of the engram in DG in FAD, making it similar
to one engram in DG in wild-type mice.
This suggests that enhancing neurogenesis can rescue memory deficits
in FAD by affecting the molecular profile of the blot in DG.

： ，
。   ， ，