Generation of App knock-in mice reveals deletion mutations protective against Alzheimer’s disease-like pathology

Although, a number of pathogenic mutations have been found for Alzheimer’s disease (AD), only one protective mutation has been identified so far in humans. Here we identify possible protective deletion mutations in the 3′-UTR of the amyloid precursor protein (App) gene in mice. We use an App knock-in mouse model carrying a humanized Aβ sequence and three AD mutations in the endogenous App gene. Genome editing of the model zygotes using multiple combinations of CRISPR/Cas9 tools produces genetically mosaic animals with various App 3′-UTR deletions. Depending on the editing efficiency, the 3′-UTR disruption mitigates the Aβ pathology development through transcriptional and translational regulation of APP expression. Notably, an App knock-in mouse with a 34-bp deletion in a 52-bp regulatory element adjacent to the stop codon shows a substantial reduction in Aβ pathology. Further functional characterization of the identified element should provide deeper understanding of the pathogenic mechanisms of AD.


1)
The authors show abeta levels using ELISA for the final targetting, but not for the majority of data interrogating the 400-900 bp deletions. ELISA is a quantitative measure that should be shown for all data. The methods state the ELISA data were acquired for both the soluble and insoluble fractions -both should be shown. It is not clear what is shown in figure 3. The correlation coefficient to genome editing efficiency should be shown for both fractions.

2)
The methodology used for quantifying % cleavage of targeting is not clear and needs to be clearly delineated since this is a major aspect of the study. The method used is unclear in both the methods, main text, and figure legends. 3) The figure legends need more information: a) 3E -is this from a WB? There should be error bars for combining data from more than one well to show variability in the measure, and error around the red mean line for controls, to show variability across the different (untargeted) controls -perhaps could use gray shading. Also should show representative WB in main figure.

b)
2C -this is WB data -for N or C terminal antibody? c) 3F -is this qPCR data? Normalized to what? Again, add error around red mean line for controls 4) Figure 2E,F has a big problem -the text states an n=14, but there are 13 dots in E and only 8 in F. Also, the genome editing efficiency doesn't match for the dots in E and F!! There must have been an error in the analysis or graphing.

5)
Protein-level data should be shown for delta 400 animals in Figure 2. Other points: The title is somewhat confusing -consider rewording. Perhaps "Use of single App knock-in mice for protective deletion mutations against the Alzheimer's disease pathology".
In abstract, should specify the type of APP knock in mouse line used (human, mutant).
The first sentence of the main text is not accurate. AD is not pathologically characterized by plaque deposition alone. High plaque burden can occur in non-AD indiviuals as well.

Responses to reviewers
<Reviewer #2> Comments (Major points) Comment 1: The authors show abeta levels using ELISA for the final targetting, but not for the majority of data interrogating the 400-900 bp deletions. ELISA is a quantitative measure that should be shown for all data. The methods state the ELISA data were acquired for both the soluble and insoluble fractions -both should be shown. It is not clear what is shown in figure 3. The correlation coefficient to genome editing efficiency should be shown for both fractions.
Response: We appreciate the reviewer's suggestion. We newly performed ELISA to quantitatively measure Aβ40 and Aβ42 levels in both NL-G-F Δ400 mouse brains (Fig. 2i) and NL-G-F ΔUTR mouse brains ( Fig. 2d-g). To clearly display what is shown in figure 3, we have divided the previous version into two different parts: scatter plots showing the relationship between the genome editing efficiency and Aβ levels or APP expression levels in eight NL-G-F Δ100 mice with heterozygous deletions (Supplementary Fig. 9) and bar graphs showing reduced Aβ levels and APP expression levels in a NL-G-F mouse with biallelic 34 bp deletions (Fig. 3e, g). We believe that the new version is easier to understand than previous ones. All ELISA data have been shown for both the soluble and insoluble fractions.

Comment 2: The methodology used for quantifying % cleavage of targeting is not clear and needs to be clearly delineated since this is a major aspect of the study. The method used is unclear in both the methods, main text, and figure legends.
Response: We appreciate the reviewer's suggestion. We realized that the previous version was confusing for the following reasons. First, our definition of deleted band was ambiguous. Second, the explanation for the quantifying method was often too simple. Third, electrophoresis data used for calculation was shown only in the NL-G-F ΔUTR mouse brains. In this study, deletion efficiency was calculated as previous studies (e.g. Ousterout et al., Nat Commun. 2015, 6:6244). After the amplification of genomic DNA with specific primer sets, parental band and deleted fragments can be clearly distinguished by agarose electrophoresis. Then, we measure the intensity of the deleted bands as well as parental band using ImageJ. The deletion efficiency of each sample was represented as the ratio of the deleted band intensity to total band intensity (i.e. cleaved band intensity plus parental band intensity). We have accordingly modified the methods (p.21, line 11-14), main text (p.6, line 6-8), a figure (Fig. 1c) and the figure legend (p.17, line 5-8). Also, we have newly included the electrophoresis data from NL-G-F Δ400 mouse brains (Supplemental Fig.5) and NL-G-F Δ100 mouse brains (Supplemental Fig.7).

Comment 3: The figure legends need more information: a) 3E -is this from a WB? There should be error bars for combining data from more than one well to show variability in the measure, and error around the red mean line for controls, to show variability across the different (untargeted) controls -perhaps could use gray shading. Also should show representative WB in main figure. b) 2C -this is WB data -for N or C terminal antibody? c) 3F -is this qPCR data? Normalized to what? Again, add error around red mean line for controls
Response: a) 3E was data from multiple WB. We have added the information of technical replicate in method section (p.24, line 9¬-10). In accordance with the reviewer's advice, we have plotted the standard deviation for controls to show variability in the measurement for Aβ levels (Fig.  3e) as well as APP expression levels (Fig. 3g). Also, we have added the representative WB data in main figure (Fig. 3f). b) This is WB data for N terminal antibody. The information has been added in the figure legend. c) Fig. 3F is qPCR data normalized to GAPDH mRNA.
Comment 4: Figure 2E,F has a big problem -the text states an n=14, but there are 13 dots in E and only 8 in F. Also, the genome editing efficiency doesn't match for the dots in E and F!! There must have been an error in the analysis or graphing.
Response: As we stated in the text, we plotted 14 dots in previous Figure 2E. But, two overlapping dots might have been seen as one dot. Whereas previous figure 2E indicated the result of NL-G-F Δ400, previous figure 2F represented the data for wild-type Δ400. We apologize that we wrote a wrong strain name (NL-G-F Δ400) in the figure legend. The number of dots were correct. To avoid confusion of the readers, the data for wild-type Δ400 have been moved to supplementary Figure (Supplementary Fig.6) Comment 5: Protein-level data should be shown for delta 400 animals in Figure 2.
Response: We appreciate the reviewer's suggestion. We have added a scatter plot showing the correlation between Aβ plaque area and APP protein level in NL-G-F Δ400 mouse brains (Fig.  2j).