Alterations of aqueous humor Aβ levels in Aβ-infused and transgenic mouse models of Alzheimer disease.

Alzheimer's disease (AD) is an ageing-related neurodegenerative disease characterized and diagnosed by deposition of insoluble amyloid-β (Aβ) plaques in the brain. The plaque accumulation in the brain directly affects reduced levels of Aβ in cerebrospinal fluid (CSF) and blood, as Aβ can freely transport the blood-brain barrier, and clinical investigations have suggested these two biofluids as promising samples for in vitro diagnosis. Given that the human eye structurally resembles the brain and Aβ accumulation often observed in the ocular region of AD patients, in this study, we examined aqueous humor Aβ as another possible surrogate biomarker. First, using the acute Aβ-infused AD mouse model by injecting Aβ to the CSF in intracerebroventricular region of normal ICR mice, we investigated whether Aβ concentration in the aqueous humor in AD models is positively correlated with the concentration in the CSF. Then, we examined the correlation of aqueous humor Aβ levels with increased plaque deposition in the brain and reduced Aβ levels in both CSF and blood in adult and aged 5XFAD Alzheimer transgenic mice. Collectively, the synthetic Aβ injected into CSF immediately migrate to the aqueous humor, however, the age-dependently reducing pattern of Aβ levels in CSF and blood was not observed in the aqueous humor.

Introduction

Abnormally increased production and deposition of the amyloid-β (Aβ) peptide in human nervous system is a typical characteristic of Alzheimer disease (AD) [1]. During the pathological progression of AD, the amyloid precursor protein (APP) on the membrane of neurons is sequentially cleaved by β- and γ-secretases and releases excessive Aβ to the extracellular regions. Although the human brain has efficient clearance systems to remove toxic Aβ such as protein degradation, blood-brain barrier (BBB) efflux, glymphatic system clearance, and meningeal lymphatic vessel transport [2], the Aβ peptide in high concentration shows misfolding behaviors and begins to accumulate in the brain of AD patients, even before the onset of cognitive deficits [3]. Unfolded monomeric Aβ is reported to participate in the physiological synaptic processes [4].

The definitive diagnosis of AD has required the detection of Aβ deposits in the brain either by biopsy, autopsy, or positron emission tomography along with the signs of neurodegeneration [5]. Recently, cerebrospinal fluid (CSF) Aβ(1–42) was suggested as an alternative biomarker for the amyloid concentration measurement by the 2018 revision of AD diagnostic criteria by National Institute on Aging and Alzheimer’s Association (NIA-AA) [5-8]. Measurements of CSF Aβ(1–42) show high diagnostic accuracy [9, 10]. It is notable that, while Aβ concentration increase and soluble oligomers and insoluble plaques build up in the brain, the alteration of Aβ levels in CSF shows a proportionally inverse behavior. The concentration of CSF Aβ(1–42) in AD patients is reduced compared to normal adults, inversely indicating the Aβ deposition in the brain [11-13]. Despite the stacked evidence, CSF Aβ(1–42) level is not routinely used in the clinical AD cases due to the complicated sample collection procedure [14, 15]. Clinical investigations searching for the less invasive biomarkers focused on blood Aβ for its clear BBB transporting mechanism through low density lipoprotein receptor-related protein 1 [16] and, thus, anticipated role to directly reflect the Aβ alterations in CSF. Since the analytical results have the discrepancy between the studies [17, 18], the usage of plasma Aβ(1–42) level as a biomarker has not been consolidated in medical practice [19]. It is attributed to the systemic circulation nature of plasma, where the protein level can be affected by the peripheral cleavage of APP or vascular risk factors [20-22]. Therefore, it is essential to explore the novel biofluid to accurately reflect the pathologic changes of AD.

The human eye has neural similarities with the brain containing high-density of neurons and glia cells and has blood barrier [23]. Given the shared functional and structural features of brain and ocular tissues, it is no surprise that the eye has been studied as a window of the brain [24]. Previously, the lens and retina regions were reported to excessively produce Aβ and show accumulation of the soluble and insoluble aggregates of the peptide [25-31]. For the ease of in vitro diagnosis, among many ocular regions, we focused on the eye fluid, the aqueous humor, in the anterior chamber [23, 32]. Aqueous humor shares similar characteristics with CSF and plasma to contain a complex mixture of proteins [33]. As the eye lack efficient amyloid clearance systems compared to the brain, the aqueous humor might not directly mirror Aβ level alteration in CSF [27]. A previous study reported that the higher Aβ(1–40) level was detected in aqueous humor of AD patients [26].

In this study, we examined the potent surrogate biomarker role of the aqueous humor Aβ(1–42) to reflect the AD manifestation. To investigate the correlation of Aβ(1–42) level in the aqueous humor with that in the brain, the CSF, and the blood, we conducted a series of in vivo experiments using two, Aβ-infused and transgenic (TG), Alzheimer mouse models. The Aβ-infused mouse model bypasses the ageing and APP processing steps and allow us to control the region-specific concentration changes of Aβ. After the injection of monomeric Aβ(1–42) directly into the intracerebroventricular (ICV) of the mouse brain in time- and dose-dependent manner, we measured levels of CSF, blood plasma, and aqueous humor Aβ(1–42) to examine if Aβ is transported from CSF to the aqueous humor. To further investigate the surrogate biomarker role of aqueous humor Aβ, we used the 5XFAD TG mouse model expresses human Aβ in its central and peripheral nervous system. We collected brain, CSF, blood plasma, and aqueous humor samples of adult and aged 5XFAD in both male and female genders and compared changes of Aβ(1–42).

Materials and methods

Animals models

Transgenic mouse (strain name; B6SJL-Tg(APPSwFlLon,PSEN1*M146L*L286V) 6799Vas/Mmjax) carrying five mutations associated with early onset familial Alzheimer’s disease (FAD) was used in the experiment. The 5XFAD mice were obtained from Jackson Laboratory (USA) and have been maintained by mating with C57BL/6 X SJL wild type mice. Institute of Cancer Research (ICR) mice (strain name; Crl:CD1, male, six-week-old) were purchased from Orientbio Inc. (Seoul, Korea). The strain is a fertile albino mouse that is widely used for the disease modeling studies. All mice were bred in a laboratory animal breeding room at Yonsei University (Seoul, Korea). They were housed in groups of five per cage with a controlled temperature, humidity, and a 12/12 hour light/dark cycle. Water and food were available ad libitum. All animal experiments were carried out in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. The research protocol was approved by the Institutional Animal Care and Use Committee of Yonsei University, Seoul, Korea (IACUC-A-201806-744-01).

ICV injection of Aβ(1–42) peptide

Aβ(1–42) peptides were synthesized using solid-phase peptide syntheses as previously reported [34]. Synthetic Aβ(1–42) peptides were dissolved in 10% dimethyl sulfoxide (DMSO) in distilled water at 0.5, 1, 2, and 4 nmol (5 μL of 100, 200, 400, and 800 μM). The mice were anesthetized with 4% avertin by intraperitoneal injection. Aβ(1–42) solutions were injected into the cerebral ventricle of mouse brain according to the previously reported protocol [35]. The injection site was 1.0 mm posterior to bregma, 1.8 mm lateral to the sagittal suture, and 2.4 mm in depth. Hamilton syringe with a 26-gauge stainless-steel needle was used to inject the Aβ(1–42) solutions.

Intravenous (IV) injection of Aβ(1–42) peptide

The 26.5-gauge syringe was prepared for Aβ(1–42) injections. To clearly see the lateral veins on both sides of a tail, the heat was applied to make the veins dilated using a 200 W lamp. Since the mice were not anesthetized, the restraining device was required to gain access to the mice veins. The mice were given stress until their lateral veins visible, followed by the administration of 100 μL of 40 μM Aβ(1–42) diluted in 1X phosphate-buffered saline (PBS).

Collection and sample preparation of brain, CSF, aqueous humor, and plasma

To obtain CSF and aqueous humor samples from mice, PYREX glass capillary tubes with a diameter of 1.5 mm were used. The capillary tube was flame-polished to obtain a diameter of 0.5 mm. The capillary tube was used to collect CSF from cisterna magna [36]. Then, we obtained aqueous humor with inserting capillary tubes to the center of cornea, enabling to reach the anterior chamber [37]. All mice were sacrificed by cervical dislocation after sampling. Collected CSF and aqueous humor samples were frozen immediately. A blood sample from the vena cava was transferred to EDTA tube and was centrifuged (3,000 rpm, 15 minutes, 4°C) to separate plasma. Protease inhibitor cocktail (Roche Diagnostics, Switzerland, cat# 11836170001) was then added to the plasma. The CSF, aqueous humor, and plasma samples were stored at −80°C in the freezer. For cryosection of the brain, each brain was initially fixed in 4% paraformaldehyde (pH 7.4) and was transferred to 30% sucrose after 24 hours. Then, the brain was cut into 35-μm-thick slices using a Cryostat (Leica, CM1860).

Analysis of Aβ(1–42) levels by sandwich-ELISA

Levels of Aβ(1–42) in biofluids were quantified by using human Aβ(1–42) ultrasensitive ELISA kit (Invitrogen, cat# KHB3544). CSF and plasma samples were 1,000-fold diluted. In time-dependent measurements, aqueous humor samples were 200-fold diluted. To detect the differences in low injection concentrations, aqueous humor samples were diluted 100-fold in dose-dependent measurements. When analyzing CSF, aqueous humor and plasma samples of 5XFAD were diluted 100-, 15-, and 5-fold respectively. The sandwich-ELISA was performed according to the manufacturer’s instructions using the diluted samples.

Immunohistochemistry assay

Brain slides were washed with 1X PBS 3 times, 5 minutes each, followed by the antigen was retrieved using 1% SDS in PBS for 10 minutes. The slides were washed with PBS, and 20% horse serum in PBS was used as a blocking reagent. We incubated the slides with 6E10 antibody (1:200, Covance) at 4°C, overnight. Then, the slides were incubated with goat anti-mouse IgG conjugated with Alexa Fluor Plus 488 (1:200, Life Technologies) for 1 hour at room temperature. Image were taken using a Leica DM2500 fluorescence microscope. The number of amyloid plaques was quantified using ImageJ software.

Statistical analysis

Statistical analysis was conducted with GraphPad Prism 7 using Student’s unpaired t-test comparisons and repeated-measures analysis of one-way ANOVA, followed by Tukey’s post hoc comparisons (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; other comparisons were not significant). Data were presented as mean ± SEM of each group.

Results

Time-dependent transport of Aβ(1–42) from CSF to aqueous humor

APP are on the membrane of neurons and, thus, their enzymatic cleavages releasing Aβ peptides are found in both central and peripheral nervous systems [38]. To eliminate the possibility that the Aβ found in the aqueous humor is produced in the eye, instead of being transported from the brain, we used Aβ-infused AD mouse. Previously, we reported an in vivo technique to acutely induce Alzheimer-like symptoms by ICV injection of Aβ [35]. This model is a useful tool to investigate Aβ-dependent pathology of AD by allowing researchers to control amyloid in a region-, a time-, and a dose-dependent manner. To verify that Aβ(1–42) migrates from the brain to the aqueous humor, 4 nmol of Aβ(1–42) was injected into the ICV regions of the brain of 6-week-old normal male ICR mice. We prepared five groups of Aβ-infused mice (male, n = 3 per group) each for separate time points and, in 15, 30, 60, 120, and 240 minutes since the ICV injection, we collected aqueous humor samples. We then used human Aβ(1–42) ultrasensitive ELISA kits to measure the biomarker concentration in each sample with triplicates (). As a result, in the aqueous humor samples of the subject mice, the artificially injected synthetic Aβ(1–42) was detected with the maximal peptide concentration at 30 minutes (). This finding is consistent with previous studies reporting the half-time of Aβ efflux from CSF to blood to be 34.63 minutes [34, 39].

Time-dependent measurements of Aβ(1–42) levels in aqueous humor after ICV injection.

ICR mice were prepared to make acute AD model (male, n = 3 per group). (A) Upper: the scheme of time-dependent experiment of aqueous humor sampling (15, 30, 60, 120, and 240 minutes). Left: ICV injection of Aβ(1–42) 4 nmol. Center part: aqueous humor collection using a capillary tube. Right: the measurement of Aβ(1–42) concentration by sandwich-ELISA. (B) Concentrations of Aβ(1–42) in aqueous humor were analyzed by ELISA. The data was analyzed by one-way ANOVA followed by Tukey’s post hoc comparisons tests. (****P < 0.0001).

Dose-dependent measurements of Aβ(1–42) levels in the aqueous humor upon ICV Aβ injection

Interim results support our hypothesis that Aβ(1–42) can transport from CSF to the aqueous humor in the mouse model. However, it is still uncertain whether the Aβ(1–42) level in the aqueous humor reflects that in CSF and how this biomarker transports from CSF to the aqueous humor. Although the blood is suspected, the migration route from CSF to aqueous humor is unclear yet. To assess the correlation of Aβ(1–42) levels in the aqueous humor with those in CSF and blood, the Aβ(1–42) peptide in various concentrations (0.5, 1, 2, and 4 nmol) was injected into the ICV region of normal ICR mice (male, n = 5 per each dose group). At the peak time of the maximal Aβ(1–42) in aqueous humor, 30 minutes from ICV injection, we collected CSF, blood, and aqueous humor samples of each (). First, the increase of CSF Aβ(1–42) levels was confirmed in a dose-dependent manner and the result supports that the ICV injection of Aβ(1–42) was successfully performed (). Secondly, as the concentration of injected Aβ(1–42) in CSF increased, the Aβ(1–42) levels in aqueous humor also increased (). This result indicates that acute changes in the concentration of Aβ(1–42) in the CSF can be reflected in the aqueous humor. In addition, we observed that the Aβ(1–42) level in the blood plasma also elevated depending on the increase concentration of injected peptide in the ICV ().

Dose-dependent measurements of Aβ(1–42) levels in aqueous humor, CSF and blood plasma after ICV injection.

ICR mice were prepared to make acute AD model (male, n = 5 per group). (A) Left: ICV injection with various Aβ(1–42) concentrations (0.5, 1, 2, and 4 nmol). Center: the collection of CSF, aqueous humor, and plasma at 30 minutes after ICV injection. Right: the analysis of biofluids Aβ(1–42) using sandwich ELISA. Aβ(1–42) levels in each (B) CSF, (C) aqueous humor, and (E) plasma were shown. (D) Scatter plot data of Aβ(1–42) levels in CSF and aqueous humor. The color of each circle represents the injection concentration of Aβ(1–42). Aβ(1–42) was analyzed in (F) plasma and (G) aqueous humor samples obtained at various time intervals (5, 15, 30, and 60 minutes) after IV injection. Data is presented as mean ± SEM against the highest concentration group in all biofluids. Standard deviation values for each group are listed in the supporting information (. Significance was tested by one-way ANOVA followed by Tukey’s post hoc comparisons tests. (**P < 0.01, ***P < 0.001, ****P < 0.0001).

To investigate the migration route of Aβ, we injected the synthetic Aβ(1–42) intravenously into the tail vein and collected aqueous humor samples with various time intervals (5, 15, 30, and 60 minutes) since the injection. As a result, we observed that the Aβ(1–42) levels in plasma significantly decreased over the time, which is consistent to the previous plasma and serum stability study [40] (). Aβ(1–42) levels in aqueous humor also decreased in the time-dependent manner which is similar to the plasma result. However, the progression was slower and Aβ(1–42) were highly detected in aqueous humor of 5 and 15 minutes group compared to the blood case of 30 and 60 minutes group (). Thus, the migration route of Aβ from the CSF to the aqueous humor is probably through the blood stream and related Aβ influx/efflux systems such as receptor for advanced glycation end products and low-density lipoprotein receptor-related protein [41].

Ageing- and gender-dependent alterations of Aβ(1–42) in brain, CSF, blood, and aqueous humor

The Aβ-infused mice utilized in the former set of experiments can be useful when the study needs to control Aβ and bypass its upstream cascades. However, in most pathophysiological phenomenon, the Aβ-infused model is less close to human AD cases compared to transgenic models. Thus, transgenic mice with human APP mutation genes are useful models to investigate amyloid cascade and related pathology of AD patients. First, to test our hypothesis that Aβ can be detected in the aqueous humor, we extracted eyes of 5XFAD mice and measured the concentrations of Aβ(1–42) in aqueous humor samples in an age-dependent manner. In this experiment, following sample sizes were used for the various age groups: 3.7-month-old (n = 5) for adult female mice, 14-month-old (n = 8) for aged female mice, 5-month-old (n = 5) for adult male mice, and 12-month-old (n = 10) for aged male mice. Each group was classified according to the disease progression. The tendency was analyzed by human Aβ(1–42) ultrasensitive ELISA kits between the levels of Aβ(1–42) detected in biofluids by gender and age differences. To determine whether Aβ peptide is progressively accumulated in the brain during the aging, immunohistochemistry assay was performed using an antibody capable of specifically detecting Aβ (). Compared to the adult mice, the higher levels of Aβ aggregates were found in the brains of aged mice (). In the CSF, Aβ(1–42) concentration decreased as ages increased (). Plasma Aβ(1–42) level also showed a decreasing tendency along the ages (). In consistent with previous reports in AD-related transgenic mouse models, the decrease of Aβ(1–42) levels in CSF and plasma was reproduced in this study [42]. Interestingly, the Aβ(1–42) levels in aqueous humor was not decreased with ages. In the female, particularly, the levels of Aβ(1–42) in aged 5XFAD mice was increased compared with those in adult 5XFAD mice. Moreover, the two male groups did not show any significant difference in Aβ(1–42) levels ().

Comparison of Aβ(1–42) levels in brain, CSF, plasma, and aqueous humor between adult mice and aged mice.

Analysis of Aβ levels in the brain and biofluids (CSF, plasma, and aqueous humor) in male and female. (n = 5 for 3.7-month-old TG female; n = 5 for 5-month-old TG male; n = 8 for 14-month-old TG female; n = 10 for 12-month-old TG male). (A) Representative brain hemisphere images stained with 6E10 antibody with age and gender differences. Scale bar = 1 mm. (B) Comparisons of the plaque numbers between adult and aged TG mice in each gender. The analysis of Aβ(1–42) levels in (C) CSF, (D) plasma, and (E) aqueous humor samples was individually shown. The differences were analyzed by unpaired t-test. (*P < 0.05, **P < 0.001, ***P < 0.001, ****P < 0.0001).

Discussion

Collectively, in AD animal mice, we observed that (1) Aβ(1–42) monomers infused into CSF in the ICV region acutely transport to the blood and the aqueous humor, (2) artificially injected Aβ(1–42) levels in CSF are proportionally reflected in the aqueous humor in mice without human APP expression, and (3) the progressive decrease of Aβ(1–42) levels in representative fluid biomarkers, CSF and blood, is not observed in aqueous humor. In this study, we infused the Aβ(1–42) monomer into CSF by ICV injection in normal ICR mice, thereby observed the corresponding increase in the aqueous humor. This acute AD model showed that ICV injected Aβ reaches the aqueous humor by excluding the other possible origins of Aβ detected in aqueous humor. As a result, the level of Aβ(1–42) in aqueous humor reached a peak concentration after 30 minutes of ICV injection. Moreover, as the injection concentration increased, detected Aβ(1–42) level in aqueous humor has also displayed an increasing trend.

We inferred the possible migration route of Aβ was through the blood. The Aβ peptide generated in the brain can be released into blood by the low-density lipoprotein receptor-related protein [43]. Aqueous humor is secreted from the ciliary body, the circulation of which produces aqueous humor by blood ultrafiltration [44]. As injecting Aβ(1–42) to the normal ICR mice intravenously, spiked Aβ concentration was reflected in aqueous humor. We concluded that the Aβ monomer passes blood-brain barrier and ultimately enters the aqueous humor. The exact molecular mechanism of how Aβ that was originally in the CSF can be detected in the aqueous humor needs further investigation.

Through examining the aqueous humor of adult and aged 5XFAD TG mice with similar AD pathophysiology to human, we found the Aβ(1–42) level elevation in the aqueous humor can be a surrogate biomarker for the AD progression. The tendency of the aqueous humor was different with that of CSF or plasma, both of which showed decreasing trend along the ages. It can be presumed that the amount of Aβ(1–42) in the aqueous humor is more derived from APP peripheral cleavage in the ocular tissues than that from the CSF or plasma [33, 45]. In female TG mice, Aβ(1–42) levels of the aqueous humor were more closely associated with the amyloid plaque burden in the brain rather than those of the CSF, independent of the migration that occurs. Unlike the clear tendency in females, no significant differences between adult and aged mice were identified in males. This result may be due to the gender-based difference of composition ratio in the aqueous humor proteome [46].

Regarded as ocular AD, primary open-angle glaucoma (POAG) shares similar underlying etiology with AD, both of which are age-related and cause neurodegeneration [47-50]. The elevation of intraocular pressure (IOP) caused by the imbalanced flow of aqueous humor is a key risk factor of POAG. The neuronal cell deaths often continue to occur even after modulating IOP to normal levels, and Aβ is likely to mediate the development of retinal ganglion cells (RGC) apoptosis which implicates neurotoxic effect [38, 51]. Since these ophthalmic biochemical changes occur earlier than the onset of AD, aqueous humor Aβ(1–42) can be analyzed in the preclinical stage of AD [52-54].

As a preliminary stage, we revealed the link of Aβ(1–42) level between CSF and aqueous humor in acute AD mouse. Intriguingly, the propensity of aqueous humor Aβ(1–42) level in TG mice was a bit different from the CSF Aβ(1–42) level. The limitation of our research is a lack of comparison with the non-AD having genetically expressed APP protein, due to the experimental disease model displaying only human Aβ(1–42). We suggest the role for aqueous humor Aβ(1–42) that indirectly reflecting the AD-related pathology. Our future direction would be the longitudinal study in clinics using aqueous humor samples to measure Aβ(1–42) levels and observe the incidence of AD. A critical factor for this approach is how to collect the aqueous humor from the patients. Seeing the increasing frequency of relatively common ophthalmic surgeries, aqueous humor samples could be easily obtained through the surgeries in elderly population [55, 56].

Statistical analyses of data in Figs 1, 2, and 3.

(TIF)

Click here for additional data file.

4 Oct 2019

PONE-D-19-24249

Correlations of amyloid-β levels between CSF and aqueous humor in transgenic and acute Alzheimer disease mouse models

PLOS ONE

Dear Professor Kim,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

All the critical issues have been raised by the Reviewers, please address all the points.

We would appreciate receiving your revised manuscript by Nov 18 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Madepalli K. Lakshmana, Ph.D

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Synopsis

In this study the authors investigate the relationship of Aβ42 in CSF and aqueous humor fluid (AHF) in two animal models, one double transgenic (dTG) and one acute model by ICV injection of synthetic Aβ42. They show increased levels of Ab42 in AHF of dTG mice at 12 and 24 months of age, they also report a slight increase between the age group. Using ICV model the author show that injected Aβ reaches the AHF in a time- and dose-dependent manner.

Comments:

This could have been a very interesting paper if the authors have had executed the experiment in whole and presented all their data. Currently it appears they just show a selection of the data that support their claim. The data on CSF Aβ and the Aβ deposits in the brain of TG animals is missing. They hence do not show if Aβ is accumulated in AHF in relation to CSF and the brain Aβ (levels/deposits), allowing them to examine if AHF Aβ can be useful as a surrogate marker for AD diagnosis. The ICV experiment is also not complete as they should have tested iv injection of Aβ in circulation to perhaps be able to substantiate the route of Aβ to AHF. These shortcomings allow the authors to make just one single conclusion that ICV injected Aβ can reach the AHF. This is insufficient however to establish that AHF Aβ can be used or is useful as a surrogate biomarker for CSF Aβ for AD diagnosis.

Thus many of conclusions and statements of the authors are either misgiving and or unjustified, as follows below:

1)- in 230-232, as this sentence gives a generalized expression rather than what the authors can claim in this particular experiment. Rephrase perhaps as follows: “This specific animal model showed that ICV injected Aβ reaches the AHF fluid”. You cannot rule out other sources at all as the ICR animals are not TG and your Aβ ELISA is human specific. Even if the animals were TG the experiment could not rule out peripheral sources by mere ICV injection and concentration changes, since even in TG there should be a pre-established equilibrium and ICV Aβ injection just add to this level.

2)- Also the conclusion in line 240-241 has to be carefully revised as it is, it is greatly misgiving. If you inject Aβ in circulation and then measure it in AHF vs CSF, you could perhaps make such a conclusion but as it is this claim is just one of several possibilities. In addition, such an information is useless (in the context of your manuscript) unless you can show that it is accumulated in AHF. Otherwise, why measure Aβ in AHF or CSF (for that matter) if you can measure it in plasma?

3)- The sentence in line 242-244 (“It is assured … TG mice”) is also not correct/justified in the present study. CSF Aβ is a biomarker of AD because of its altered levels in AD vs control and its inverse association with the amount of Aβ deposits in the brain but not merely because Aβ can be measured in CSF (as control subjects have much higher levels of Aβ in the CSF than AD).

4)- Also consider the following shortcoming of the study: You might have been able to provide some clues if you had measured changes in AHF Aβ in relation to CSF Aβ and the Aβ deposits in the brain of the TG animal to establish that levels of Aβ in AHF like that in CSF reflect the extend of Aβ deposits in the brain. Thus currently, the Fig 1 results does not really say much as the comparison between TG vs WT has limited value since your Aβ ELISA can only measure human Aβ. The main interesting point in Fig. 1 is the Aβ level at 12 vs 24 months but since you do not show the levels of plaques and the levels of Aβ in CSF you cannot establish a proper relationship to point out AHF Aβ as a surrogate biomarker.

5)- In Abstract, line 26-28. The author claim that they “… investigated whether the Aβ in the aqueous humor in AD models is directly correlated with the misfolding protein of the central nervous system.” You did not do such an investigation in this study as far as I can deduct from your experiment. As a matter of fact your ICV model reject your statement as you cannot claim (and it is very unlikely) that the synthetic Aβ monomers were all misfolded (not to mention that even in TG or human, one cannot claim that the soluble Aβ peptides are misfolded even though many believe so mainly because of confusion that aggregation can only occur due to misfolding of Aβ peptides and neglect that aggregation may also occur simply because of an excessive accumulation in e.g. parenchymal fluid in the brain.

Other issues:

1. ICR (mice) is not defined in abstract or the rest of the text.

2. Is the Age-dependent increase in AHF Aβ significant? Provide p-values!

3. Why didn’t you measure Aβ in serum, CSF and brain extracts of dTG?

4. To ascertain route of Aβ to AHF, why not also inject Aβ iv in circulation, which is much easier to do and would provide additional information?

5. In Fig 3, plasma Aβ42 at 4 nmole ICV injection is missing.

6. In Fig 3, it is difficult to appreciate the importance/significant of the comparisons with the highest conc group (i.e. 4 nmol group), particularly since there is no difference between 0.5 vs 1 nmol-groups or 2 vs 4 nmol groups (in 3B)? It is important here that direct correlation graphs plus their r- and p-values are shown, as well.

Reviewer #2: Reviewer Comments to Editor:

Recommendation: Accept after major revision.

Reviewer Comments to Author:

Globally, Alzheimer’s disease (AD) is a growing health and economic challenge that has no effective cure. AD is a progressive neurodegenerative disorder and pathological hallmarks include presence of hyperphosphorylated tau and amyloid protein deposition. Currently, these pathological biomarkers are detected either through cerebrospinal fluid analysis or brain imaging. Though effective, these methods are not widely available due to issues such as the difficulty in acquiring samples, lack of infrastructure or high cost associated.

The authors have previously reported a direct correlation of Aβ (1–42) levels between CSF and plasma in AD mouse model by injecting monomeric Aβ (1–42) directly into the intracerebroventricular (ICV) region of normal adult mouse brains to induce AD-like phenotypes. In this manuscript, Kwak et al. reported that presence of Aβ (1-42) in the aqueous humor can be used as a surrogate bio-marker of perturbed Aβ levels in central nervous systems of AD.

This manuscript is well written, concise, easy-to-follow and provides a new insight on the potential bio-markers for AD patients.

I support its acceptance provided the major revisions as noted below are adequately addressed:

Abstract:

• Well-organized and well written.

• Since the word limit is 300, I would suggest authors to include few sentences of background information.

• Preferable to be consistent with either Aβ (1-42) or amyloid beta.

Introduction:

• In the first paragraph, few sentences are written without any citations. Citations would strengthen use point of emphasis.

• It is advisable to revise last paragraph where the author’s discuses about the experiment protocol instead of focusing on the importance of their study.

• Please include one sentence showing the novelty and importance of your study.

Materials and Methods:

• In experiment 2, please elaborate how the aqueous humor at different time point from the same mice were collected.

Results:

• The authors have used male transgenic mice in their experiments whereas several studies (refer to the journals: PMID: 30820070, PMID: 26987699, PMID: 20442496) have reported that the incidence of AD is higher in females. So, I would suggest to repeat the experiments in female mice model to make the data more authenticate to the clinical data.

• Since hyperphosphorylated tau has been found in both AD and Glaucoma, suggesting a possible pathophysiologic link between the two diseases (PMID: 26425322), but no attempt was made to quantify hyperphosphorylated tau in this study.

• The authors measured the Aβ (1–42) concentration in CSF by injecting different concentration of Aβ (1–42) into the ICV region of normal adult ICR mice by following their previous study. But, the result showed significant differences (around 5-folds higher in this study compare to the previous result) between two studies. Why this discrepancy.

• In the description of figure 1 results, the authors claimed that human Aβ (1-42) was not found in the aqueous humor of WT mice, but in figure 1 bar data, it clearly showed the presence of some amount Aβ (1-42) in WT.

• The authors claimed that Aβ transport from the central nervous system to the eye is through blood. So, I would recommend to perform the experiment for Aβ (1–42) concentration in plasma also to see how much Aβ (1–42) transported to the aqueous humor through the blood.

• In figure 2B, the authors measured the half-time of Aβ efflux from CSF to blood and their finding is consistent with the work done by Shibata M. et al. (2000). In addition, Shibata M. et al. (2000) also mentioned that there was also a slow, time-dependent retention of Aβ in brain parenchyma with a t1/2 of 164.5 minutes. But, figure 2B shows after 34.63 minutes, release of Aβ is suddenly stopped from the brain which is not consistent with the Shibata study, i.e. there is no time-dependent retention of Aβ in brain.

• For figure 2B & 3C, the authors followed the same procedure (injecting monomeric Aβ (1–42) directly into the intracerebroventricular (ICV) region of normal adult mouse brains), used same concentration (4nmole), same time point (30 minutes). But, why two different Aβ (1-42) concentrations.

• More labels are necessary in Figs 2A & 3A.

Discussion:

• The authors should discuss more about how the Aβ (1–42) is transported into the aqueous humor through blood-brain barrier.

Reviewer #3: This is a generally well written and interesting paper about the significantly increased levels of Aβ in the aqueous humour of transgenic AD mouse model in comparison to WT, and the ability to measure acutely perturbated levels of CSF Aβ in both the plasma and the aqueous humour of ICR mice, demonstrating that Aβ introduced to CSF is transported into aqueous humour.

There are some grammatical errors, as well as some questions regarding the statistics and presentation of your data that I have listed in the sections below.

In general, where you say central nervous system, please change to CSF for better clarity, as you are referring to soluble Aβ(1-42).

Abstract

• Line 25: “Based on the clinical evidence…”

• Line 27: Please clarify what specifically is correlated; i.e. presence/concentration of Aβ in aqueous humour is positively correlated with the presence/concentration if Aβ in CSF

• Line 31: Consider changing wording, i.e. “To investigate the correlation between Aβ levels in CSF and aqueous humour, we…”

• Line 37: Consider changing conclusion wording, i.e. “Our results indicate that Aβ peptides are elevated in the aqueous humour in AD mice. In addition to this, acute elevation of Aβ peptides through injection into CSF can be detected in the aqueous humour, suggesting that

Introduction

• Line 47: Other limitations to the amyloid-PET scan e.g. cost, exposure to radiation?

• Line 59: “The eye has been suggested…”

• Section beginning with Line 59 is generally unclear; please clarify why you have included information regarding POAG, i.e. do you mean that Aβ in aqueous has been found to be elevated in POAG, and may play a role in neurodegenerative diseases in the eye, and so measuring Aβ in aqueous humour may be useful for AD as well as other neurodegenerative diseases?

• Line 61: Reference 6 does not directly measure levels of Aβ in aqueous humour of patients, so does not support your statement

• Line 62: Please clarify logic for this sentence, your meaning is unclear. Reference 10 is a review paper that discusses the role of Aβ in AMD, and how changes in Aβ may be involved in the early stages of AMD pathogenesis. Do you mean that changes in Aβ levels in aqeuosu humour may also be seen in the early stages of AD pathogenesis?

• Section beginning with Line 59

• Line 72: “Given that human aqueous humour samples have been shown to contain Aβ…” this statement needs a reference

Methods

• Consider adding section summarising statistical analyses used, for better clarity

• Line 99: “…as previously reported”

Results

• Line 140: Should be “Results”

• Line 150: You have stated that Aβ(1-42) was not found in the aqueous humour of WT mice, and you refer to Figure 1. However, Figure 1 shows that > 500 pg/mL of Aβ(1-42) was found in 12-month-old WT, and approx. 250 pg/mL of Aβ(1-42) was found in 24-month-old WT. In addition to this, you noted on line 125 that the analytical sensitivity of your sandwich-ELISA is < 1.0 pg/mL. Please clarify this discrepancy. Do you mean that Aβ(1-42) was present in significantly lower levels in WT compared to TG model?

• Figure 1: Please state whether you have plotted means ± SD or means ± SEM in the figure caption, include the sample size as a legend and add brackets to indicate which two groups you are referring to for each p-value

• For the data shown in Figure 1, you have used an F-test to determine the difference in Aβ(1-42) levels between WT and TG, which assumes both datasets are normally distributed, and have equal variances. Please include the statistical test for normality, and provide a table summarising the numerical mean ± SD for each group

• Please include the F-value, and degrees of freedom, for each p-value you have given for Figure 1

• Line 153: Please perform a statistical analysis to determine whether levels of Aβ(1-42) are significantly different between TG 12 months and TG 24 months

o If all datasets are normally distributed with equal variance, consider doing a one-way ANOVA for all 4 datasets, and then perform a Tukey post hoc test to determine which means are significantly different from each other amongst all 4 groups

• Line 154: Please change “subjects” to “in ageing TG mice”; also, please reword this sentence, the logic is not clear.

• Line 157: Please change “the eye” to “aqueous humour”.

• Line 187: In the caption for Figure 2, please change “central nervous system” to “CSF” and “eye” to “aqueous humour”

• Line 195: please change “central nervous system” to “CSF” and “eye” to “aqueous humour”

• Line 209: please change to “…that acute changes in the concentration of Aβ(1-42) in the CSF…”

• Line 212: This sentence needs a citation

• Please perform a correlation between levels of Aβ(1-42) in CSF and Aβ(1-42) in aqueous humour to show how strongly these are related, include the resulting r2 value and statistics to support statements regarding the “significant correlation” e.g. line 222, and generate a figure showing this data

• Please include the sample size for each time point as a legend, and state in the figure caption whether you have plotted mean ± SD or mean ± SEM

• In Figure 2, the level of Aβ(1-42) in aqueous humour after 4 nmole ICV injection was approx. 6000 pg/mL, however in Figure 3, the level of Aβ(1-42) in aqueous humour after 4nmole ICV injection appears to be half of this, at approx. 3000 pg/mL. Can you comment on why you think this may be the case, as the protocol was the same?

• For Figure 3, you state you have plotted mean ± SEM, however it appears you have used box plots. Please clarify what you have plotted. If they are box plots, this is a good indication of the spread of your data, but please also provide mean ± SD values in a table.

• You have used a one-way ANOVA to determine whether the levels of Aβ(1-42) in CSF, aqueous humour and plasma are significantly different following different injection concentrations, which assumes that datasets are normally distributed and have equal variances. Please include the statistical tests for normality.

• For the data provided in Figure 3, please report the F-values and degrees of freedom from your one-way ANOVA. Also, please clarify which pairwise comparison was significantly different, either on the figure with brackets, or in a table.

• Please include the data for level of Aβ(1-42) in plasma when 4 nmole of Aβ(1-42) was injected, or provide details regarding why this data point was excluded from Figure 3D

Discussion

• Line 228 should read “Aβ levels in CSF and aqueous humour”

• Line 230: Is this statement accurate? To my understanding, although ICR mouse does not specifically express human APP, this mouse model still expresses endogenous murine APP which may produce Aβ which may be detected by ELISA. In addition to this, your WT control had detectable levels of Aβ in aqueous. Better evidence of the movement of Aβ from CSF to aqueous humour would be the change in Aβ concentration over time, as well as the significant changes in Aβ levels in aqueous when you alter injected CSF Aβ concentrations.

• Line 239: You may also wish to add, to further support your argument, that it is known that aqueous humour is composed of blood plasma (REF), and it is believed that aqueous humour proteins are derived from blood plasma, rather than synthesised locally inside the eye (REF).

• Line 240: “Thus, it is possible…”

• Line 242: Consider discussing results in the order that they have been presented e.g. WT vs. TG, then ICV injection to ICR mice

• Line 245: “known to precede other processes like the…”

References

• Line 280 should be “References”

Minor Comments

• In my version, the figures look pixelated (e.g. text is hard to read in Figure 3). Please check the resolution of all figures.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Taher Darreh-Shori

Reviewer #2: Yes: Rajib Kumar Dutta

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

5 Dec 2019

We have a separate file upload for the response to reviewers.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

26 Dec 2019

Alterations of aqueous humor Aβ levels in Aβ-infused and transgenic mouse models of Alzheimer disease

PONE-D-19-24249R1

Dear Dr. Kim,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Madepalli K. Lakshmana, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

31 Dec 2019

PONE-D-19-24249R1

Alterations of aqueous humor Aβ levels in Aβ-infused and transgenic mouse models of Alzheimer disease

Dear Dr. Kim:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Madepalli K. Lakshmana

Academic Editor

PLOS ONE