RBM7 deficiency promotes breast cancer metastasis by coordinating MFGE8 splicing switch and NF-kB pathway

Reviewer #1 (Public Review):

Summary:
Fang Huang et al found that RBM7 deficiency promotes metastasis by coordinating MFGE8 splicing switch and NF-kB pathway in breast cancer by utilizing clinical samples as well as cell and tail vein injection models.

Strengths:
This study uncovers a previously uncharacterized role of MFGE8 splicing alteration in breast cancer metastasis, and provides evidence supporting RBM7 function in splicing regulation. These findings facilitate the mechanistic understanding of how splicing dysregulation contributes to metastasis in cancer, a direction that has increasingly drawn attention recently, and provides a potentially new prognostic and therapeutic target for breast cancer.

Weaknesses:
This study can be strengthened in several aspects by additional experiments or at least by further discussions. First, how RBM7 regulates NF-kB, and how it coordinates splicing and canonical function as a component of NEXT complex should be clarified. Second, although the roles of MFGE8 splicing isoforms in cell migration and invasion have been demonstrated in transwell and wound healing assays, it would be more convincing to explore their roles in vivo such as the tail vein injection model. Third, the clinical significance would be considerably improved, if the therapeutic value of targeting MFGE8 splicing could be demonstrated.

Reviewer #2 (Public Review):

Summary:
In this manuscript, the authors reported the biological role of RBM7 deficiency in promoting metastasis of breast cancer. They further used a combination of genomic and molecular biology approaches to discover a novel role of RBM7 in controlling alternative splicing of many genes in cell migration and invasion, which is responsible for the RBM7 activity in suppressing metastasis. They conducted an in-depth mechanistic study on one of the main targets of RBM7, MFGE8, and established a regulatory pathway between RBM7, MFGE8-L/MFGE8-S splicing switch, and NF-κB signaling cascade. This link between RBM7 and cancer pathology was further supported by analysis of clinical data.

Strengths:
Overall, this is a very comprehensive study with lots of data, and the evidence is consistent and convincing. Their main conclusion was supported by many lines of evidence, and the results in animal models are pretty impressive.

Weaknesses:
However, there are some controls missing, and the data presentation needs to be improved. The writing of the manuscript needs some grammatical improvements because some of the wording might be confusing.

Specific comments:
(1) Figure 2. The figure legend is missing for Figure 2C, which caused many mislabels in the rest of the panels. The labels in the main text are correct, but the authors should check the figure legend more carefully. Also in Figure 2C, it is not clear why the authors choose to examine the expression of this subset of genes. The authors only refer to them as "a series of metastasis-related genes", but it is not clear what criteria they used to select these genes for expression analysis.

(2) Line 218-220. The comparison of PSI changes in different types of AS events is misleading. Because these AS events are regulated in different mechanisms, they cannot draw the conclusion that "the presence of RBM7 may promote the usage of alternative splice sites". For example, the regulators of SE and IR may even be opposite, and thus they should discuss this in different contexts. If they want to conclude this point, they should specifically discuss the SE and A5SS rather than draw an overall conclusion.

(3) In the section starting at line 243, they first referred to the gene and isoforms as "EFG-E8" or "EFG-E8-L", but later used "EFGE8" and "EFGE8-L". Please be consistent here. In addition, it will be more informative if the authors add a diagram of the difference between two EFGE8 isoforms in terms of protein structure or domain configuration.

(4) Figure 7B and 7C. The figures need quantification of the inclusion of MFGE exon7 (PSI value) in addition to the RT-PCR gel. The difference seems to be small for some patients.

Minor comments:
The writing in many places is a little odd or somewhat confusing, I am listing some examples, but the authors need to polish the whole manuscript more to improve the writing.

(1) Line 169-170, "...followed by profiling high-throughput transcriptome by RNA sequencing", should be "followed by high-throughput transcriptome profiling with RNA sequencing".

(2) Line 170, "displayed a wide of RBM7-regulated genes were enriched...", they should add a "that" after the "displayed" as the sentence is very long.

(3) Line 213, "PSI (percent splicing inclusion)" is not correct, PSI stands for "percent spliced in".

(4) Line 216-217, the sentence is long and fragmented, they should break it into two sentences.

(5) Line 224, the "tethering" should be changed to "recognizing". There is a subtle difference in the mechanistic implication between these two words.

(6) Line 250, should be changed to "..in the ratio of two MFGE8 isoforms".