Review Reports - Phosphatidylserine-Dependent Clearance of Damaged Red Blood Cells by Liver Sinusoidal Endothelial Cells in Alcohol-Related Liver Disease

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this manuscript, Li et all investigated the role of liver sinusoidal endothelial cells (LSECs) in the clearance of oxidatively damaged and ethanol-primed red blood cells (RBCs) in alcohol-related liver disease (ALD). The study is of research interest and the authors have used a combination of in vitro experiments, murine models, and human liver samples to addresses a relevant and understudied aspect in ALD. However, several conceptual and methodological limitations reduce the strength of the investigation and should be addressed before the manuscript can be considered for publication.

Scale bar is missing from figure 1B. Please use arrow to better indicate the uptake of the single RBC.
Figure 2 and 3. Are there morphological changes of SK-HEP1 cells under treatment with different doses of oxiRBC over time as the authors observed biphasic changes of HO-1, Bmp6 and ferritin. Did higher concentrations of oxiRBC suppress HO-1 and Bmp6? Also, for figure 2B and 3A-C, indication of significance is missing from the bar graph. Additionally, line 323, there was no data on HO-1 expression in figure 3. Did the authors refer to the data in figure 2?
Figure 4D needs clarification. Was the measurement of PS on oxiRBC done with stabilin-1 KO cells? Nuclear staining and data quantification are suggested. In addition, did the authors directly quantify RBC uptake following Stabilin-1 knockdown using Time-lapse imaging method?
The quality of figure 5A is poor, therefore it is hard to evaluate the data accuracy. Please add nuclear staining to indicate whole cells. Scale bar should be added.
The authors used ethanol concentrations up to 1000 mM, which may not be physiologically relevant. Discussion regarding physiologically relevant blood alcohol concentrations and how they relate to the findings from this study is suggested.
SK-HEP1 cells are described as "LSEC-like" cells. However, they are not widely accepted as bona fide primary LSECs. In the manuscript, several conclusions are phrased as if they could be directly apply to in vivo LSECs. The authors should double check on the conclusions throughout the manuscript to clearly distinguish between “LSEC-like" cells. and primary LSECs.
Figure 7A-B. Please make sure all images are presented under the same setting. Some images are brighter when compare with other images. Also, H&E staining is suggested to show hepatic pathological changes within different groups. Scale bars should be included in the image and in figure legend.
For the ALD mouse model, additional analysis, such as serum ALT, AST, GGT and hepatic inflammatory-related markers, should be included to demonstrate the successful induction of ALD.
Figure 8A-B and 9. Please make sure all images are presented under the same setting. Some images are brighter when compare with other images. Scale bars should be included in the image and in figure legend.
In the discussion, the authors mentioned that CD206 staining was used to identify LSECs, while also acknowledging that CD206 is expressed by macrophages and other cell types. Are there additional endothelial markers that the authors could used for LSEC identification?
Please double check and make sure all images are consistent and professionally aligned in the manuscript.

Comments on the Quality of English Language

Please read over carefully to check for grammatical errors and consistency. The authors should also improve clarity and readability in the Introduction and discussion by reducing sentence length and redundancy. English editing is suggested.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This study demonstrates that liver sinusoidal endothelial cells (LSECs) contribute to the clearance of damaged red blood cells (RBCs) in alcohol-related liver disease (ALD), alongside the well-known roles of macrophages and hepatocytes. Using SK-HEP1 cells as an LSEC surrogate, the authors show that oxidatively damaged (CuSO₄/ascorbate-treated) or ethanol-primed RBCs externalize phosphatidylserine (PS) and are rapidly internalized via the scavenger receptor Stabilin-1. This PS-Stabilin-1–dependent efferocytosis triggers a coordinated transcriptional response involving Nrf2 activation, HO-1 induction, BMP6 upregulation, and ferritin expression, consistent with heme degradation and intrahepatic iron sequestration. Ethanol concentrations as low as 25 mM were sufficient to render RBCs “efferocytosis-competent.” Complementary in vivo data from murine models of chronic ethanol feeding plus hemolysis (phenylhydrazine) and from human ALD liver biopsies revealed hemoglobin-derived signals within CD206⁺ sinusoidal structures with the diffuse distribution pattern characteristic of LSECs. The authors conclude that LSEC-mediated RBC clearance represents an additional, previously underappreciated pathway that may drive hepatic iron overload in ALD.

The identification of a PS–Stabilin-1 axis in LSEC efferocytosis of ethanol-damaged RBCs is a clear advance. The combination of live-cell imaging, siRNA knockdown (showing ~60 % reduction in HO-1 induction), dose–response curves, and PS staining provides robust, internally consistent evidence. The manuscript nicely integrates this with the known Keap1–Nrf2–HO-1 axis and places it in the context of ALD-associated hemolysis and iron dysregulation. This multicellular view of erythrocyte clearance (macrophages + hepatocytes + LSECs) is conceptually important and should be highlighted more prominently in the discussion as a paradigm shift.

I have a few comments that should be addressed:

SK-HEP1 cells are repeatedly described as “LSEC-like,” yet they originate from a hepatic adenocarcinoma. While useful as a surrogate, they lack the full fenestrated phenotype and zonation of primary human or murine LSECs. The manuscript would be substantially strengthened by repeating key experiments (PS-dependent uptake, Stabilin-1 induction, HO-1 response) in primary human LSECs or freshly isolated murine LSECs, even if only at a single time/dose point. If primary cells are unavailable, the authors should at minimum discuss the known differences (e.g., expression levels of Stabilin-1, CD206, and fenestrae) and cite prior studies that have validated SK-HEP1 for scavenger functions.
The use of CD206 to identify the “sinusoidal scavenger compartment” is reasonable, but the authors correctly note that CD206 is also expressed on macrophages. Reliance on “diffuse linear distribution pattern” alone is somewhat subjective. I recommend co-staining with additional LSEC-specific markers (e.g., LYVE-1, CD31, or Stabilin-1 itself) and macrophage exclusion markers (e.g., F4/80 or CD163) on the same sections. Quantification of hemoglobin autofluorescence co-localized with LSEC markers (% area overlap or Manders’ coefficient) would turn qualitative observations into quantitative data.
The in vitro ethanol priming (up to 1000 mM) is acknowledged as a model system, yet the lowest effective dose (25 mM) is physiologically relevant. However, the 24 h exposure time and the high concentrations used for the positive-control oxidized RBCs could induce non-specific membrane damage. I suggrest including a time-matched vehicle control at each ethanol concentration and reporting actual blood alcohol levels achieved in the Lieber-DeCarli mouse model for direct comparison. A short discussion of how these priming conditions mimic the chronic low-grade oxidative stress seen in human ALD drinkers would improve translational clarity.
Most experiments report n = 3 technical replicates and use one-way ANOVA with Tukey’s test, which is appropriate; however, several key panels (e.g., Fig. 2B dose–response, Fig. 3A–C, Fig. 4B) show large error bars and borderline significance at lower doses. The authors should clarify whether these are biological replicates from independent experiments or technical replicates, and consider presenting individual data points rather than (or in addition to) bar graphs.