Aβ1-42 Accumulation Accompanies Changed Expression of Ly6/uPAR Proteins, Dysregulation of the Cholinergic System, and Degeneration of Astrocytes in the Cerebellum of Mouse Model of Early Alzheimer Disease

Round 1

Reviewer 1 Report

The manuscript submitted by Bychkov et al. describes the response of the cholinergic system in the cerebellum of an Alzheimer’s disease (AD) mouse model. The authors’ goal was to characterize the influence of β-amyloid (Aβ) on proteins known to modulate nAChR function. They find increased Aβ in the cerebellum is associated with decreased levels of the anti-inflammatory transcription factor KLF4 and many of its downstream targets, the Ly6/uPAR proteins, resulting in altered interactions between Aβ and α7-nAChR. This is accompanied by changes in astrocyte morphology. The experiments appear to be well-performed and are appropriate. There are a few items that would strengthen the manuscript for publication. General and specific comments follow below.

 

General Comments:

The data presented is relevant to the field and the manuscript is well structured, starting with Western blots and qPCR for Ly6/uPAR proteins, moving through localization and interaction studies, and ending with astrocyte morphology and locomotor activity.

 

The cited references are mostly older publications, with more than half (60%) being published more than 5 years ago. Despite their age, the publications are relevant, though more recent references can be found (e.g. astrocyte activation in AD). The authors cite themselves 10 times (18% of all references).

 

The manuscript appears to be scientifically sound, with the experimental design appropriate to test the hypothesis.

 

The methods section lacks important details that make it hard to know if the results are reproducible and could impact the conclusions drawn from the results. See specific comments below.

 

The figures are appropriate. They are easily read, understood, and interpreted. The data is interpreted appropriately and consistently throughout the manuscript. The images of GFAP immunostaining would benefit from being brighter as it is difficult to see the processes in the “initial image” panels.

 

The conclusions are consistent with the evidence and arguments presented.

 

The statements regarding ethics and data availability are adequate.

  

 

Specific Comments:

In the “Animals” section (4.1) of the Materials and Methods, the authors state 12 mixed sex adult mice were purchased (line 388). It appears these may have been the same mice that were used for experiments, as line 385 refers to their age and weight at the end of the experiments. However, line 390 says they were bred, so please clarify whether the original mice purchased were used to establish a colony and breed experimental mice, or if the experimental mice were purchased. If purchased, please clarify how long the mice were permitted to adapt after arrival before the experiments began.

 

In the “AnimalsPlease include details on the experimental mice – were the sexes balanced? How old were they at tissue harvest? Are all analyses being performed at the same age/timepoint? How was the number of animals used determined – please include a power analysis and/or a rationale.

 

Please include details on the euthanasia and tissue harvest procedures. Include how much RNA was reverse transcribed, whether there were protease inhibitors included in the RIPA buffer, and how much protein was loaded for the western blots.

 

In the “Immunohistochemistry” and “Image Analysis” sections, the following details are needed for each set of experiments: orientation of cryosections (sagittal/coronal), how many slices were processed and analyzed for each mouse, where those slices were located (medial/lateral, etc), what region(s) if not the entire cerebellum was imaged/quantified. These details are significant due to known regional heterogeneity of glial cells and the potential for sampling bias.

In the “4.7. Statistical Analysis” section of the Methods, there is no explanation for sample size/replicate numbers or power analysis. Please include justification for the experiments being performed – why 3 mice per group in the binding assays versus 5 or 6 mice per group for most other analyses (including behavior which is known to require well more than 6 mice per group to attain significant results due to interanimal variability), and why 3-4 astrocytes per mouse in GFAP staining.

 

Section 2.5 addresses changes in astrocyte morphology, with the authors stating that astrocytes degenerate. This seems to be an overstatement as all they examine is morphology, which can also indicate changes in activation status. Rather, counting the number of astrocytes and/or performing double-labeling with an apoptotic marker would be required to know if they are losing astrocytes. Other markers of astrocyte health could be used as well, if there isn’t a loss of astrocytes but they are becoming unhealthy/dysfunctional. Please include additional experiments to back up the statement or temper the conclusion.

 

Lines 303-313 in the discussion address the discordance between mRNA and protein levels. This type of expression pattern observed here is understood to occur in instances of high/altered protein turnover where mRNA levels rise to maintain needed protein levels. This should be included in the discussion as well.

 

Figure 7 does a beautiful job illustrating the findings and proposed mechanism. Please include in the discussion potential mechanisms/reasons why Lynx1 protein levels would decrease after Aβ increases. The rest of the mechanism is clearly stated, so this seems a large omission.  

The English is of a sufficient quality. There are a few typographical errors, but nothing major. 

Author Response

Reviewer#1

The manuscript submitted by Bychkov et al. describes the response of the cholinergic system in the cerebellum of an Alzheimer’s disease (AD) mouse model. The authors’ goal was to characterize the influence of β-amyloid (Aβ) on proteins known to modulate nAChR function. They find increased Aβ in the cerebellum is associated with decreased levels of the anti-inflammatory transcription factor KLF4 and many of its downstream targets, the Ly6/uPAR proteins, resulting in altered interactions between Aβ and α7-nAChR. This is accompanied by changes in astrocyte morphology. The experiments appear to be well-performed and are appropriate. There are a few items that would strengthen the manuscript for publication. General and specific comments follow below.

 General Comments:

The data presented is relevant to the field and the manuscript is well structured, starting with Western blots and qPCR for Ly6/uPAR proteins, moving through localization and interaction studies, and ending with astrocyte morphology and locomotor activity.

 

Reviewer 1:

The cited references are mostly older publications, with more than half (60%) being published more than 5 years ago. Despite their age, the publications are relevant, though more recent references can be found (e.g. astrocyte activation in AD). The authors cite themselves 10 times (18% of all references).

 Answer:

In first, we are grateful the reviewer for the all comments and suggestions. All changes in the manuscript are shown by yellow. As for references, we added new ones (colored in yellow) and reduced self-citation rate, many thanks.

 

Reviewer 1:

The figures are appropriate. They are easily read, understood, and interpreted. The data is interpreted appropriately and consistently throughout the manuscript. The images of GFAP immunostaining would benefit from being brighter as it is difficult to see the processes in the “initial image” panels.

 Answer:

We changed the figures according to the reviewer’s suggestion.

 

Specific Comments:

Reviewer 1:

In the “Animals” section (4.1) of the Materials and Methods, the authors state 12 mixed sex adult mice were purchased (line 388). It appears these may have been the same mice that were used for experiments, as line 385 refers to their age and weight at the end of the experiments. However, line 390 says they were bred, so please clarify whether the original mice purchased were used to establish a colony and breed experimental mice, or if the experimental mice were purchased. If purchased, please clarify how long the mice were permitted to adapt after arrival before the experiments began.

 Answer:

Thank you for your comment. Initially, animals were bred and housed under the standard conditions of the Animal Breeding Facility, BIBCh, RAS, from which genotyped animals were purchased by the authors for the experiments. Animals were purchased at the age of 6 months and kept in SPF zone of vivarium of IBCh RAS until 9 months of age. Experiments were carried out on 9-month-old mice. We added this information in Animals section of Materials and Methods.

 

Reviewer 1:

In the “AnimalsPlease include details on the experimental mice – were the sexes balanced? How old were they at tissue harvest? Are all analyses being performed at the same age/timepoint? How was the number of animals used determined – please include a power analysis and/or a rationale.

 Answer:

We thank reviewer for this valuable point.  Tg- group consisted of 6 males and 7 females, 2xTg-AD group consisted of 7 males and 7 females, we added the details in the Methods section. All mice were 9-month-old, and analyses were performed at the same time point.

Our study is preliminary, studying the possible role of three-finger proteins in the cerebellum in AD onset. Determination of proper group size was quite a challenging task, especially given the small amount of material and the variety of experimental procedures performed. We have chosen the minimal n for our investigations as n =3 for affinity purification and n =5 for other experiments based on our previous researches [10.1111/jnc.15018]. We increased the number of animals where the amount of experimental material allowed this. Statistical analysis gave us significant difference between some groups of the data, but to answer the question properly, we also performed the post-hoc power analysis of our data (Sample Size Calculator 1.060), which showed power greater than 50% for all significant results. This was in accordance with statistical literature because in post-hoc power analysis p value and power are connected [10.1198/000313001300339897]. The size effect in significantly different groups was estimated as “medium” (> 0.5) and large (> 0.8) [10.4300/JGME-D-12-00156.1], however the power for some experiments with significantly different data was less than 80% usually sought when determining the group size. We suppose, that even when the power did not reach 80% we still caught the significant differences between the groups even with the numbers of animals we used. Please, look at Table R1 enclosed. It should be noted that post-hoc power analysis of conducted study sometimes can be irrelevant [10.1198/000313001300339897; 10.1136/gpsych-2019-100069]. Nevertheless, we are extremely grateful to the reviewer for valuable remark and will increase the number of animals in accordance with the power analysis in our further studies.

 

Table R1. Post-hoc power analysis of obtained experimental data

panel	Type of test	Difference	SEM	SD	n	power	significance (showed on the picture)
FIGURE 1
PCR experiments
1a	t	0.001036	0.0002066	0.000506	6	0.89	***
1b	t	0.1162	0.0126	0.030864	6	0.99	****
1c	u	0.001131		0	6	0.85	##
1d	t	0.007381	0.002098	0.005139	6	0.86	**
1e	u	0.9759		0	6	0.81	#
	t	0.008857	0.005853	0.014337	6	0.16	n.s.
WB quantification
1a	t	1.1805	0.1677	0.410779	6,5	0.99	**
1b	u	0.2481			6,5	0.53	##
1c	t	0.1332	0.03408	0.083479	6,5	0.75	**
1d	t	0.2581	0.0683	0.1673	6,5	0.74	**
1f	t	0.05341	0.02245	0.054991	6,5	0.6	*
	t	0.009936	0.01639	0.040147	6,5	0.14	n.s.
	t	1.273	0.5038	1.234053	6,5	0.62	*
FIGURE 2
Imaris quantification
2b	t	3336	235	664.6804	5,8	0.99	****
	t	224.7	52.71	149.0864	5,8	0.95	**
2c	t	108.2	60.12	170.045	5,8	0.29	n.s.
	t	32.69	64.13	181.387	5,8	0.06	n.s.
2d	t	3.956	1.69	4.780042	5,8	0.54	*
Pearson's colocalization
2f	t	0.1383	0.06079	0.17194	5,8	0.58	*
FIGURE 3
Lynx1/a7-nAChR quantification
3c	t	4.863	1.525	2.641377	3,3	0.65	*
FIGURE 4
Ab1-42 quantification
4b	t	0.4375	0.112	0.274343	6,5	0.77	**
Imaris quantification
4d	t	108.2	60.12	170.045	5,8	0.29	n.s.
	t	32.96	64.13	181.387	5,8	0.06	n.s.
4e	t	2455	146	412.9504	5,8	0.99	****
	u	556.5			5,8	0.83	###
4f	u	18.41			5,8	0.81	##
Pearson's colocalization
4h	t	0.1885	0.06786	0.191937	5,8	0.82	*
FIGURE 5
PCR experiments
5a	t	0.673	0.07245	0.177466	6,5	0.99	****
5b	t	0.093	0.02134	0.052272	6,5	0.98	***
WB quantification
5a	t	0.0002495	4.16E-05	0.000102	6	0.99	***
FIGURE 6
Astrocyte count
6c	t	253.5	43.21	122.2	5,8	0.98	***
6d	t	660.4	152.7	431.9	5,8	0.90	**
6e	t	0.196	0.33	0.93	5,8	0.07	n.s.
	t	6.533	2.065	5.84	5,8	0.52	*
6g	t	0.4251	0.04878	0.136	5,8	0.99	****
Imaris quantification
6i	t	3.424	1.187	3.357343	5,8	0.59	*

* - fot t-test

# - for Mann-Whitney U-test

 

Reviewer 1:

Please include details on the euthanasia and tissue harvest procedures. Include how much RNA was reverse transcribed, whether there were protease inhibitors included in the RIPA buffer, and how much protein was loaded for the western blots.

 Answer:

 All the details were added in the proper sections of Methods. Thank you.

 

Reviewer 1:

In the “Immunohistochemistry” and “Image Analysis” sections, the following details are needed for each set of experiments: orientation of cryosections (sagittal/coronal), how many slices were processed and analyzed for each mouse, where those slices were located (medial/lateral, etc), what region(s) if not the entire cerebellum was imaged/quantified. These details are significant due to known regional heterogeneity of glial cells and the potential for sampling bias.

 Answer:

Many thanks for the remark, we added details in the Methods section.

 

Reviewer 1:

In the “4.7. Statistical Analysis” section of the Methods, there is no explanation for sample size/replicate numbers or power analysis. Please include justification for the experiments being performed – why 3 mice per group in the binding assays versus 5 or 6 mice per group for most other analyses (including behavior which is known to require well more than 6 mice per group to attain significant results due to interanimal variability), and why 3-4 astrocytes per mouse in GFAP staining.

 Answer:

The number of animals for affinity extraction, Western blotting and PCR was calculated by us based on the previous studies [10.1111/jnc.15018] and based on material availability. For behavioral tests, the number of animals was 13 in the Tg- group and 9 in the 2xTg-AD group. The number of sections for immunohistochemistry was selected based on the quality of the sections, as well as the number of cells that could be adequately reconstructed in the fields of view. The necessary details have been added to the Methods section and figure legend (6h). The post hoc power analysis did not provide the 80 % power for all the experiments performed, but our study was preliminary, pointing on the influence of three-finger proteins on the Aβ-pathology onset in the cerebellum, we will increase the number of animals during further studies.

 

Reviewer 1:

Section 2.5 addresses changes in astrocyte morphology, with the authors stating that astrocytes degenerate. This seems to be an overstatement as all they examine is morphology, which can also indicate changes in activation status. Rather, counting the number of astrocytes and/or performing double-labeling with an apoptotic marker would be required to know if they are losing astrocytes. Other markers of astrocyte health could be used as well, if there isn’t a loss of astrocytes but they are becoming unhealthy/dysfunctional. Please include additional experiments to back up the statement or temper the conclusion.

 Answer:

 We re-imaged slices with the marker of cell death suitable for IHC (cleaved cytokeratin 18, magenta) and found that in 2x-Tg-AD mice the amount of cleaved cytokeratin 18 is higher than in Tg- mice, pointing on astrocytic death within Aβ pathology. We thank reviewer for valuable comment.

 

Reviewer 1:

Lines 303-313 in the discussion address the discordance between mRNA and protein levels. This type of expression pattern observed here is understood to occur in instances of high/altered protein turnover where mRNA levels rise to maintain needed protein levels. This should be included in the discussion as well.

 Answer:

 We have re-written corresponding part of the discussion. Thank reviewer for providing us such an interesting point of discussion.

 

Reviewer 1:

Figure 7 does a beautiful job illustrating the findings and proposed mechanism. Please include in the discussion potential mechanisms/reasons why Lynx1 protein levels would decrease after Aβ increases. The rest of the mechanism is clearly stated, so this seems a large omission. 

 Answer:

Thank you for this valuable remark. We changed Fig 7 in accordance with the reviewer comments, and added the text about possible reasons of Lynx1 down-reegulation upon Ab pathology in the discussion section.

Reviewer 2 Report

This manuscript presents new insights into the role of the cerebellar nACh receptor in the early stages of Alzheimer's Disease (AD) development. The study explores the underlying mechanisms of dysregulation by characterizing interactions with subunits like Lynx1 and identifying the relevant downstream intracellular signaling proteins. Given the scarcity of foundational research on the cerebellum in the context of AD, this study introduces an innovative approach that is likely to attract significant attention in the field.

 

Before publishing, it's essential to consider the following factors.

 

F5. (b) Explain why the Western blot is missing in TNF-alpha.

 

F6. The IHC contains no regional information of cerebellum where the astrocytes are located in which layer. They also lack the preparation detail in method section for which part of cerebellum was taken like vermis or hemisphere. In addition, the anatomical information of cutting direction (ex. Horizontal, vertical, sagittal) is missing. The astrocyte cell counting number is too small (< 8). Even though GFAP is a marker for an astrocyte, they designated it as for astrocytic progenitors which cannot be valid. A proper marker like CSPG4 as for the astrocytic progenitor should be applied. Otherwise, it would be proper to replace astrocytic progenitor to astrocyte. Accordingly, all the related descriptions (ex. L362. accompanied by the reduced number of the co-localized Lynx1and α7-nAChR clusters in the vicinity of the astrocytic progenitors) should be rewritten.

 

The locomotor activity data in F6 (h) doesn’t contain the information about the number of animals neither in the legend nor method section (4.6). Generally speaking, If the number is less than, the behavior study would not be valid to induce any conclusive result. (ex. L365. Although, no changes in the locomotor activity in the Rotarod test usually associated with the cere-bellum function were revealed (Figure 6h).’

 

Therefore, the number should be noted in both the legend and method. If the number is insufficient to support the evidence, they need to address scientific justification or even remove the data itself.

L54. These peptides bind to the nicotinic receptor type α7 (α7-nAChR), important for synaptic transmission, learning and memory, and induce neuronal death after endocytosis.

Refine the sentence

L56. Potentiation of α7-nAChR leads to neuroprotection, moreover agonists of α7-nAChR can modulate the Aβ1-42/α7-nAChR pathogenic signaling mechanisms in the cortex of AD patients

Replace ‘modulate’ into a more functionally descriptive word.

L58. New insights recently showed the implication of the cerebellum in AD pathogenesis [10,11]. Although, the role of cerebellar α7-nAChR in regulation of early onset of AD nowadays remains unclear.

Please clarify the paragraph to eliminate any ambiguity.

Author Response

Reviewer#2

This manuscript presents new insights into the role of the cerebellar nACh receptor in the early stages of Alzheimer's Disease (AD) development. The study explores the underlying mechanisms of dysregulation by characterizing interactions with subunits like Lynx1 and identifying the relevant downstream intracellular signaling proteins. Given the scarcity of foundational research on the cerebellum in the context of AD, this study introduces an innovative approach that is likely to attract significant attention in the field.

Before publishing, it's essential to consider the following factors.

 Reviewer 2:

F5. (b) Explain why the Western blot is missing in TNF-alpha.

 Answer:

 In first, we are grateful the reviewer for the all comments and suggestions. All changes in the manuscript are shown by yellow. Thank you for your remark. The TNFα is a secreted protein, which we cannot detect by Western blotting analysis of the cerebellum homogenate. However, previously we performed ELISA of the serum of Tg- and 2Tg-AD mice that showed up-regulation of TNFα upon Aβ pathology [10.31857/S2686738922600881].

 

Reviewer 2:

F6. The IHC contains no regional information of cerebellum where the astrocytes are located in which layer. They also lack the preparation detail in method section for which part of cerebellum was taken like vermis or hemisphere. In addition, the anatomical information of cutting direction (ex. Horizontal, vertical, sagittal) is missing. The astrocyte cell counting number is too small (< 8). Even though GFAP is a marker for an astrocyte, they designated it as for astrocytic progenitors which cannot be valid. A proper marker like CSPG4 as for the astrocytic progenitor should be applied. Otherwise, it would be proper to replace astrocytic progenitor to astrocyte. Accordingly, all the related descriptions (ex. L362. accompanied by the reduced number of the co-localized Lynx1and α7-nAChR clusters in the vicinity of the astrocytic progenitors) should be rewritten.

 Answer:

Thank you for this valuable remark. The necessary details have been added to the Methods section. The number of sections for immunohistochemistry was selected based on the quality of the sections, as well as the number of cells that could be adequately reconstructed in the fields of view. We agree with insufficiency of GFAP alone to identify the astrocytic progenitors, so we have re-written the corresponding text within Results and Discussion sections of the manuscript. In the results section, we left term “major astrocytic progenitors” but also wrote that all astrocyte could not be stained solely by GFAP antibody.

 

Reviewer 2:

The locomotor activity data in F6 (h) doesn’t contain the information about the number of animals neither in the legend nor method section (4.6). Generally speaking, If the number is less than, the behavior study would not be valid to induce any conclusive result. (ex. L365. Although, no changes in the locomotor activity in the Rotarod test usually associated with the cere-bellum function were revealed (Figure 6h).’

Therefore, the number should be noted in both the legend and method. If the number is insufficient to support the evidence, they need to address scientific justification or even remove the data itself.

 Answer:

For behavioral tests, the number of animals was 13 in the Tg- group 1 and 9 in the 2xTg-AD group. We added the details in the figure legend and Methods section. We are very grateful for valuable remark.

 

Reviewer 2:

Comments on the Quality of English Language

L54. These peptides bind to the nicotinic receptor type α7 (α7-nAChR), important for synaptic transmission, learning and memory, and induce neuronal death after endocytosis.

Refine the sentence

Answer:

Refined, many thanks.

 

Reviewer 2:

L56. Potentiation of α7-nAChR leads to neuroprotection, moreover agonists of α7-nAChR can modulate the Aβ1-42/α7-nAChR pathogenic signaling mechanisms in the cortex of AD patients

Replace ‘modulate’ into a more functionally descriptive word.

Answer:

Thank you, the sentence was re-written.

 

Reviewer 2:

L58. New insights recently showed the implication of the cerebellum in AD pathogenesis [10,11]. Although, the role of cerebellar α7-nAChR in regulation of early onset of AD nowadays remains unclear.

Please clarify the paragraph to eliminate any ambiguity.

Answer:

We changed that paragraph, many thanks.

Round 2

Reviewer 1 Report

The reviewer thanks the authors for addressing all of the concerns. While it would be optimal to have used more than 1 slice per mouse for IHC due to regional variability, it appears the one used was relatively well matched. 

The manuscript is now suitable for publication. 

Reviewer 2 Report

All the comments have been sufficiently answered.