Skip to main content
NCBI home page
Journal of Biochemistry logoLink to Journal of Biochemistry
. 2025 Jul 9;178(4):267–275. doi: 10.1093/jb/mvaf043

UNC5B is an isoform-dependent target for ectodomain shedding

Kotaro Sugimoto 1, Eichi Watabe 2,3, Mio Takuma 4, Kaname Nagahara 5, Toshinori Sawano 6, Mihoko Kajita 7, Junichi Takagi 8, Hidehito Kuroyanagi 9,10, Kyoko Shirakabe 11,12,✉
PMCID: PMC12480741  PMID: 40631620

Abstract

Ectodomain shedding (shedding) is a processing mechanism that cleaves the juxtamembrane region of membrane proteins and solubilizes almost the entire extracellular domain. Shedding irreversibly regulates the localization and function of membrane proteins; however, its physiological role is not fully understood. Previously, we showed that the shedding susceptibility of multiple membrane proteins is altered by skipping or inclusion of skipping exon(s) that encode their juxtamembrane region. In this study, we screened the skipping exon encoding the juxtamembrane region of membrane proteins and found that the shedding susceptibility of UNC5B, a Netrin-1 receptor, is altered by skipping or inclusion of the skipping exon encoding its juxtamembrane region. These results raise the possibility that the biological phenomena involving UNC5B, including neural circuit formation, angiogenesis and cancer development, are regulated by shedding in a splice isoform-dependent manner.

Keywords: alternative splicing, ectodomain shedding, skipping exon, transmembrane domain, UNC5B

Graphical Abstract

Graphical Abstract.

Graphical Abstract

Open in a new tab

Ectodomain shedding, hereafter referred to as shedding, is a processing mechanism in which transmembrane (TM) proteins are cleaved at their juxtamembrane region, and almost the entire extracellular domain is solubilized (1). Shedding is mainly executed by a disintegrin and metalloprotease (ADAM) family membrane-anchored metalloproteases (2, 3), which convert a single membrane protein into a soluble extracellular domain that are released extracellularly and a membrane-anchored protein with almost no extracellular domain. Many of the membrane-anchored shedding products undergo intramembrane processing executed by γ-secretase (4), which can only cleave membrane proteins with almost no extracellular domain, and the resulting products are released both intra- and extracellularly. In other words, shedding is a rate-limiting step in the two-step processing of membrane proteins that irreversibly converts a membrane protein into multiple soluble proteins with different cellular localizations. To date, membrane proteins with various functions, such as growth factors, inflammatory cytokines, their receptors and adhesion molecules, have been reported to undergo shedding (1–3), and their functions are thought to be regulated by the two-step processing; however, the physiological role played by this processing is not fully understood.

To gain a better understanding of the role of membrane protein processing, we developed a screening system for targets of metalloprotease-dependent shedding (5, 6), as shedding is the rate-limiting step of the processing. We screened a macrophage cell line stimulated with an endotoxin and identified multiple membrane proteins as shedding targets (6). Through analysis of the identified shedding targets, we found that the shedding susceptibility of three of them, cell adhesion molecule 1 (CADM1), signal-regulatory protein alpha (SIRPα) and activated leukocyte cell adhesion molecule (ALCAM), is regulated by post-transcriptional modification, alternative splicing (6, 7). Both shedding-susceptible and shedding-resistant isoforms are generated by alternative splicing of their genes. Among them, the shedding susceptibilities of CADM1 and ALCAM, two immunoglobulin superfamily adhesion molecules, are altered by skipping or inclusion of single skipping exon encoding the extracellular juxtamembrane region, which is less than 3% of the full-length protein (6, 7). These results not only indicate that the shedding susceptibility of these membrane proteins is precisely regulated by alternative splicing but also raise the possibility that alternative splicing regulates the shedding susceptibility of other membrane proteins.

To verify whether there are other membrane proteins whose shedding susceptibility is regulated by alternative splicing, we screened skipping exons that encode the juxtamembrane region of type I TM proteins. By examining the identified skipping exons, we found that UNC5B, a Netrin-1 receptor (8), is an isoform-dependent shedding target, although its shedding has not been previously reported. These results suggest that the function of UNC5B, assumed to be involved in neural circuit formation, angiogenesis and cancer development (9–12), is regulated by shedding in a splice isoform-dependent manner.

Materials and Methods

Screening of skipping exons encoding juxtamembrane region of type I TM proteins

GTF annotation files for the RefSeq genes were downloaded from the UCSC Genome Browser (13). We used versions 109.20211119 for hg38 (human) and 109.20201027 for mm39 (mouse). To focus only on protein-coding genes, we filtered the GTF files to include only genes containing ‘NM’ or ‘XM’ transcripts. Next, we ran SUPPA v2.3 (14) using the command ‘python suppa.py generateEvents -i input.gtf -o output -e [SE,SS,MX,RI,FL] -f ioe’ to generate alternative splicing events.

Bed files for TM domains and signal peptides were downloaded from the UniProt FTP server (15). Both hg38 and mm39 used the version released in 2022_05. The exon-TM domain overlaps and relationships between the generated alternative splicing exons were obtained using bedtools v2.31.1 (16). Overlapping exons were identified using the ‘intersect’ command in bedtools, whereas adjacent exons were identified using the ‘closest’ command. We then filtered for cases in which the upstream exon adjacent to the TM domain was labelled as skipping exon. Additionally, only genes with a single TM domain and signal peptide at the N-terminus were classified as type I TM proteins. For gene duplication checks, mouse homologs of the screened genes in hg38 were obtained using homologene (17) and compared with the results from mm39. All filtering and analyses were performed using bedtools v2.31.1 (16) and custom in-house R scripts.

Enrichment analysis

Pathway enrichment analysis was conducted using Enrichr (18). Background was defined as the complete set of type I TM protein genes filtered using the methods described above. From the Reactome 2022 (19) results, the top 10 terms (P-value <0.01) were used to create bubble plots using ggplot2 with custom in-house R scripts.

Cell line, transfection and sample preparation for western blotting

Chinese hamster ovary (CHO) cells were cultured in alpha-MEM supplemented with 10% fetal bovine serum (FBS) and antibiotics. Wild-type and ADAM17−/− mouse embryonic fibrobrasts (MEFs) (20) were kindly provided by Dr. Keisuke Horiuchi (National Defense Medical College Hospital, Saitama, Japan) and cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% FBS and antibiotics. Transfection was performed using PEI-MAX (Polysciences, Warrington, PA, USA) for CHO cells and NEPA21 (NEPA GENE, Chiba, Japan) for MEFs. For western blotting, cell extracts and proteins from the culture supernatants were prepared as previously described (5). All western blotting were performed at least three times, and one representative data are shown. The amount of soluble extracellular domain in the culture supernatant was quantified using Image Studio (LI-COR, Lincoln, NE, USA).

Antibodies, chemicals, expression plasmids and siRNAs

Antibodies were purchased from Promega (anti-HaloTag, G9211; Madison, WI, USA), Fujifilm Wako (anti-PA tag, NZ-1; Tokyo, Japan), Santa Cruz Biotechnology (anti-β-actin, sc-81,178; Santa Cruz, CA, USA), and Abcam (anti-ADAM10, ab39177; anti-ADAM17, ab39162; Cambridge, MA, USA). TPA and DAPT were purchased from Sigma-Aldrich (St. Louis, MO, USA). BB94 was purchased from Selleck Chemicals (Houston, TX, USA). Bortezomib was purchased from Cell Signaling Technology (Danvers, MA, USA). The coding sequences of human long AXL (#105932) and murine long UNC5B (#72195) were purchased from Addgene (Cambridge, MA, USA). Splice isoforms and substitution mutants were constructed using a PCR-based method. The cytoplasmic PA-tag was attached to the C-terminus of UNC5B isoforms using PCR. The coding sequences of UNC5B and AXL were subcloned into an N-terminally-Halo-tagged-type I TM protein expressing plasmid (21) after the removal of their signal sequences (Met1-Ala26 and Met1-Met25, respectively) using PCR. siRNAs (ADAM10, s61946; ADAM17, s61958) were purchased from Thermo Fisher Scientific (Waltham, MA, USA).

Cell surface staining of halo-tagged proteins

CHO cells expressing N-terminally Halo-tagged UNC5B isoforms and a red fluorescent protein (DsRed) were stained with a cell-impermeable HaloTag Alexa Fluor 488 ligand (Promega) according to the manufacturer’s instructions. Fluorescence images were captured using an APX100 Digital Imaging System (Evident, Tokyo, Japan).

Identification of shedding cleavage site of long UNC5B

CHO cells expressing N-terminally Halo-tagged and C-terminally PA-tagged long UNC5B were treated with 200 ng/ml TPA and 10 μM DAPT for 3 h and extracted in RIPA buffer (20 mM Tris–HCl (pH 7.4), 150 mM NaCl, 2 mM EDTA, 1% NP20, 0.1% SDS) supplemented with a protease inhibitor mixture (Sigma-Aldrich) and 10 μm BB94. The extract was incubated with anti-PA tag antibody beads (Fujifilm Wako) at 4°C overnight, washed with RIPA buffer, and eluted via boiling in sample buffer (56.25 mM Tris–HCl (pH 6.8), 1.8% SDS, 40 mM DTT, 9% glycerol). The eluted proteins were separated using SDS-PAGE, blotted onto a PVDF membrane and stained with Coomassie Blue. The approximately 65-kDa membrane-anchored shedding product of UNC5B was excised and subjected to automated Edman degradation using a protein sequencer PPSQ-53A (Shimadzu, Kyoto, Japan).

Results

Screening of skipping exons encoding juxtamembrane region of type I TM proteins

To comprehensively explore membrane proteins whose shedding susceptibility may be regulated by alternative splicing, we screened publicly available data (Fig. 1A). First, we listed the alternative splicing events for human and mouse protein-coding genes registered in RefSeq using SUPPA2 (14). This resulted in the identification of 98,975/66,620 alternative splicing events from 19,545/22,173 genes and 138,947/102,245 transcripts in the human/mouse genome (hg38/mm39). Among the alternative splicing events identified, the percentage of skipping exons was high in both human and mouse genomes (Fig. S1A). We then marked TM exons encoding TM domains registered in UniProt (15) and checked whether these TM exons, as well as the immediately upstream and downstream exons, were subject to alternative splicing. We found that alternative splicing events were most frequent in the exons immediately upstream of TM exons (Fig. S1B). From these ‘TM domain-associated alternative splicing events’, we decided to focus on the skipping exons immediately upstream of the TM exons of type I TM proteins based on the following rationale.

Fig. 1.

Fig. 1

Open in a new tab

Screening of skipping exons encoding juxtamembrane region of TM proteins. (A) An overview of the screening for the candidates for alternative splicing-regulated shedding targets. The number of genes, transcripts, alternative splicing events and skipping exons corresponding to each filtering step is shown. Numbers for human (hg38) and mouse (mm39) are listed separately, divided by a slash. In the gene structure diagram, TM domains are shown in light blue, and skipping exon is shown in light green. (B) Venn diagram of the skipping exons identified in humans (hg38) and mice (mm39). The homologs identified in both humans and mice on homologene is shown in overlapping. (C) Enrichment analysis of the identified candidates for alternative splicing-regulated shedding target genes using all type I TM protein genes as the background. The top 10 enriched pathway terms in Reactome 2022 are plotted. The x-axis and colour represent the P-value, and the size of the dots reflects the number of genes associated with each term.

Our previous proteomic screening of shedding targets revealed that 88.9% of the identified proteins were type I TM proteins (6). Similarly, secretome analysis showed that 87.3% of the proteins secreted in a metalloprotease-dependent manner were type I TM proteins (22). In addition, 71.3% of the shedding targets registered in SheddomeDB (23), a semi-automatically curated database based on publications, were type I TM proteins. Taken together, we concluded that the majority of shedding targets are type I TM proteins.

Regarding the position of the shedding cleavage sites, in the above-mentioned secretome analysis, metalloprotease-dependent cleavage sites were mainly located within 200 amino acids from the TM domain (22). In SheddomeDB (23), 1,160 of the 1,344 cleavage sites (86.3%) of the human shedding targets were located upstream of the TM domain (Fig. S1C), half of which were within 199 amino acid residues (Fig. S1D). A similar trend was observed in mice (Fig. S1C and D). Furthermore, in humans, half of the cleavage sites nearest to the TM domain are located within 45 amino acids (Fig. S1D). These observations suggest that shedding cleavage sites are generally close to the TM domain; thus, we focused on the exons immediately upstream of the TM exons of type I TM protein genes. Furthermore, among the multiple types of alternative splicing events, we focused on skipping exons, which occur most frequently in exons immediately upstream of TM exons (Fig. S1B) and is thought to affect the structure of proteins by inserting or deleting a certain length of the amino acid sequence.

We screened for skipping exons immediately upstream of the TM exons of type I TM protein genes, which have a signal sequence and one TM domain in UniProt, and obtained 174 and 105 skipping exons from the human and mouse genomes, respectively (Fig. 1A, Table S1A and B). Among these, 40 skipping exons were identified in both the human and mouse genomes (Fig. 1B). The genes containing identified skipping exons included not only CADM1 and ALCAM (6, 7), but also well-known shedding target genes, ERBB2, NTRK2, NRP1 and IL6R (1, 3, 24, 25), suggesting that this screening is suitable for clarifying the relationship between shedding susceptibility and alternative splicing. All extracted exons, genes, proteins and the corresponding SheddomeDB entries are summarized in Table S1A and B.

Next, to characterize the identified genes, candidates for alternative splicing-regulated shedding targets, we performed enrichment analysis using Enrichr (18), with all type I TM protein genes as the background, across various pathway databases. To date, nearly half of the FDA-approved cancer biomarkers are known to be shedding targets (26). Our screening also identified notable cancer-related genes such as ERBB2, ERBB4, NTRK2, NTRK3, FGFR2 and FGFR3 (27). Reactome (19) showed enrichment for pathways associated with cancer, including ‘MAPK Family Signalling Cascades’ and ‘Signalling By Receptor Tyrosine Kinases’ (Fig. 1C). Immune-related pathways, such as ‘immune system’, and neural differentiation-related pathways ‘axon guidance’ and ‘nervous system development’ were also enriched (Fig. 1C). Shedding plays a crucial role in immune regulation (28), and our screening identified important immune-related genes such as IL6R, IL2RA and TLR4. Additionally, several genes related to axon guidance, such as NRP1, NRP2, UNC5B and UNC5C (29) have been identified, suggesting a potential link between neural differentiation and alternative splicing-regulated shedding.

Splice isoforms of UNC5B, but not those of AXL, have different susceptibilities to shedding

To examine whether splice isoforms generated by skipping or inclusion of identified skipping exons have different shedding susceptibilities, we selected skipping exons for analysis according to the following criteria: (i) identified in both humans and mice. (ii) Less than 40 bases in length, similar to skipping exons of CADM1 and ALCAM (33 and 39 bases, respectively). (iii) No other alternative splicing events were listed on UniProt. As a result, the skipping exons of three membrane proteins were selected: AXL, a tyrosine kinase receptor (30); UNC5B, a netrin receptor (10); and PTPRA, a tyrosine phosphatase receptor (31). When the amino acid sequences of their juxtamembrane regions were compared, only PTPRA isoforms contained many O-glycosylatable amino acids that could interfere with shedding (6) (Figs S1E and 3B); thus, we focused on AXL and UNC5B for the analysis. As shown in Fig. 2A, the skipping exons of both AXL and UNC5B (27 and 33 bases, respectively) encode the stalk region between the extracellular domain structure and the TM domain. The coding sequences of both human long AXL and mouse long UNC5B were purchased, and those of their short isoforms were constructed using a PCR-based method. N-terminally Halo-tagged AXL and UNC5B isoforms were expressed in CHO cells, which were treated with or without 12-O-tetradecanoylphorbol 13-acetate (TPA), a strong shedding inducer (1). The cell extracts and culture supernatants were subjected to western blotting using an anti-Halo antibody. The Halo-tagged soluble extracellular domain of AXL was almost equally released into the culture supernatant from cells expressing the long or short isoforms in a TPA-dependent manner (Fig. 2B). In contrast, the soluble extracellular domain of UNC5B was released only from cells expressing the long UNC5B, and not from cells expressing the short UNC5B (Fig. 2B). Release of the extracellular domain of long UNC5B was potentiated by TPA (Fig. 2B and C) and completely inhibited by the metalloprotease inhibitor BB94 (Fig. 2C). These results indicate that the two UNC5B isoforms have different susceptibilities to metalloprotease-mediated shedding.

Fig. 3.

Fig. 3

Open in a new tab

Investigation of the shedding cleavage site and sheddase of long UNC5B. (A) CHO cells expressing N-terminally Halo-tagged and cytoplasmic PA-tagged long or short UNC5B were treated with or without 200 ng/ml TPA and/or 10 μM DAPT for 60 min, and cell extracts and culture supernatants were subjected to western blotting using indicated antibodies. Black triangle indicates the approximately 65 kDa membrane-anchored shedding product. (B) The amino acid sequences of juxtamembrane region of UNC5B isoforms. The amino acids encoded by the skipping exon are shown in red, and the amino acids of the TSP type 1 domain are shown in light blue. An arrowhead indicates the shedding cleavage site of long UNC5B. (C) MEF cells transfected with siRNAs and/or Halo-tagged long UNC5B expressing vector as indicated were treated with or without TPA, and cell extracts and culture supernatants were subjected to western blotting using indicated antibodies. The mean values of soluble extracellular domain quantities in three independent experiments are shown with SD. NS means not significant. (D) Wild-type and ADAM17−/− MEFs transfected with Halo-tagged long UNC5B expressing vector were treated with or without TPA, and cell extracts and culture supernatants were subjected to western blotting using indicated antibodies. The mean values of soluble extracellular domain quantities in three independent experiments are shown with SD. Asterisk indicates the statistical significance (P < 0.05) as calculated using the Student’s t-test. (E) CHO cells expressing Halo-tagged long UNC5B mutants were treated with TPA for 60 min, and cell extracts and culture supernatants were subjected to western blotting. The amino acid sequences of substitution mutants are indicated above. The mean values of soluble extracellular domain quantities in three independent experiments are shown with SD. Asterisk indicates the statistical significance (P < 0.05) as calculated using the Student’s t-test. (F) CHO cells expressing N-terminally Halo-tagged and cytoplasmic PA-tagged long UNC5B were treated with 200 ng/ml TPA and/or 1 μM bortezomib for 60 min, and cell extracts and culture supernatants were subjected to western blotting. Black and white triangles indicate the membrane-anchored shedding product and γ-fragment of long UNC5B, respectively.

Fig. 2.

Fig. 2

Open in a new tab

Two UNC5B isoforms have different susceptibility to shedding. (A) Schematic diagrams of molecular structure of splice isoforms of human AXL and mouse UNC5B. Exons are indicated by black rectangles and alternative exon are coloured in red. SP, signal peptide; Ig, immunoglobulin-like domain; FN III, fibronectin type III domain; TM, transmembrane domain; Kinase, tyrosine kinase domain; TSP, thrombospondin type 1 domain; ZU5, ZU5 domain; UPA, UPA domain; DD, Death domain. (B) CHO cells expressing N-terminally Halo-tagged splice isoforms of human AXL and mouse UNC5B were treated with or without 200 ng/ml TPA for 60 min, and cell extracts and culture supernatants were subjected to western blotting using anti-Halo antibody. The mean values of soluble extracellular domain quantities in three independent experiments are shown with SD. Asterisk indicates the statistical significance (P < 0.05) as calculated using the Student’s t-test. NS means not significant. (C) CHO cells expressing Halo-tagged long UNC5B were treated with or without 200 ng/ml TPA and/or 10 μM BB94 for 60 min, and cell extracts and culture supernatants were subjected to western blotting. The mean values of soluble extracellular domain quantities in three independent experiments are shown with SD. Asterisk indicates the statistical significance (P < 0.05) as calculated using the Student’s t-test. (D) CHO cells expressing Halo-tagged UNC5B isoforms were stained using cell-impermeable HaloTag ligand (Halo ligand). Transfected cells were indicated as DsRed-positive cells (DsRed). Bar, 20 μm.

As shedding often occurs at the cell surface, we investigated the cell surface localization of the two UNC5B isoforms. Staining of cells expressing N-terminally Halo-tagged long or short UNC5B with a membrane-impermeable Halo-ligand showed that both isoforms were present on the cell surface to the same extent (Fig. 2D). These results suggest that the two UNC5B isoforms differ in their susceptibility to metalloprotease(s).

Identification of the cleavage site of long UNC5B

Next, we attempted to determine the shedding cleavage site of the long UNC5B. N-terminal sequencing of the membrane-anchored shedding product is effective for identifying the shedding cleavage site (6, 7, 32). To detect membrane-anchored shedding product of long UNC5B, C-terminally PA-tagged UNC5B isoforms were expressed in CHO cells, the cells were treated with TPA and DAPT, a γ-secretase inhibitor, and cell extracts were subjected to western blotting using an anti-PA antibody. PA-tagged protein corresponding to the membrane-anchored shedding product (approximately 65 kDa) was detected in the extracts of long UNC5B-expressing cells and increased in response to TPA and DAPT (Fig. 3A, black triangle). This protein was barely detected in the extracts of short UNC5B-expressing cells (Fig. 3A), indicating that it was a membrane-anchored shedding product of long UNC5B. Edman degradation assay of this protein showed that the N-terminal sequence was Asp-Pro-Lys-Ser, indicating that long UNC5B was cleaved between asparagine 361 and aspartic acid 362 (Fig. 3B, arrowhead).

Investigation of the responsible sheddase of long UNC5B

Because shedding is mainly executed by two ADAM metalloproteases, ADAM10 and ADAM17 (1–3), we examined whether these ADAMs were responsible for the shedding of long UNC5B. Silencing using the siRNA of ADAM10 did not substantially inhibit the shedding of long UNC5B in mouse embryonic fibroblasts (MEFs), whereas that of ADAM17 tended to inhibit it (Fig. 3C). To further confirm the involvement of ADAM17, we used ADAM17 knockout MEFs (20). In ADAM17 knockout MEFs, TPA-dependent shedding of long UNC5B was observed, but its extent was reduced compared to that of wild-type MEFs (Fig. 3D). These results suggest that ADAM17, rather than ADAM10, is partially involved in the shedding of long UNC5B.

Construction of shedding resistant mutant of long UNC5B

We have previously shown that a cluster of negatively charged amino acids located around the cleavage site interfere with shedding (7). Therefore, we constructed substitution mutants of long UNC5B and examined their susceptibility to shedding. Substitutions of the four amino acids bracketing the identified cleavage site with negatively charged amino acids only partially attenuated the shedding of long UNC5B (Fig. 3E, EDDD), whereas substitutions of the six amino acids almost completely attenuated (Fig. 3E, DEDDDE). These results show that negatively charged amino acids can inhibit the shedding of long UNC5B and indicate the reliability of the identified shedding cleavage site.

Detection of UNC5B fragment generated by γ-secretase cleavage

The inhibition of γ-secretase resulted in the accumulation of membrane-anchored shedding product of long UNC5B (Fig. 3A), suggesting that the product is processed by γ-secretase and the entire cytoplasmic domain (γ-fragment) is released as in the case of the Notch receptor (33). To detect the γ-fragment of UNC5B, we used the proteasome inhibitor, bortezomib, because the γ-fragment of Neogenin has been reported to increase upon proteasome inhibition (34). CHO cells expressing C-terminally PA-tagged long UNC5B were treated with TPA and bortezomib, and cell extracts were subjected to western blotting using anti-PA antibody. PA-tagged protein smaller than membrane-anchored shedding product was detected in the extracts of cells treated with both TPA and bortezomib (Fig. 3F, white triangle). This protein was hardly detected in the extract of cells treated with TPA or bortezomib alone. These results suggest that it is a γ-fragment of UNC5B.

Discussion

In the present study, we identified UNC5B as an isoform-dependent target for shedding. To the best of our knowledge, this is the first report on the shedding of UNC5B and provides useful insights into the function of UNC5B. UNC5B is a receptor for Netrin-1 that has been identified as a neuronal axon guidance factor (8) and is assumed to play an important role in not only neural circuit formation but also angiogenesis and cancer development (9–12). This study suggests that these phenomena may be regulated by the shedding of UNC5B. The long UNC5B coding sequence purchased from Addgene was cloned from mouse retinal cDNA (35), suggesting that long UNC5B is sufficiently expressed in the retina, a neural tissue. In addition, RT-PCR using primers flanking exon 8 demonstrated that long UNC5B mRNA was expressed in adult mouse cerebellum at nearly the same level as short UNC5B mRNA (data not shown). These results suggest that shedding-susceptible long UNC5B is expressed in neural tissue and that UNC5B shedding may occur. UNC5B is also known to be a ‘dependence receptor’, which actively induces apoptosis in the ‘absence’ of the corresponding ligand, Netrin-1 (36). Recently, it has been reported that the UNC5B isoforms have different inducibilities of apoptosis; long UNC5B induces apoptosis only in the absence of Netrin-1, whereas short UNC5B induces apoptosis both in the presence and absence of Netrin-1 (37). The functional differences between UNC5B isoforms may be due to the differences in their shedding susceptibility. The report that the shedding-resistant short UNC5B does not respond to Netrin-1 suggests that shedding of UNC5B might be necessary to respond to Netrin-1. Many of the membrane proteins that undergo shedding are further processed by γ-secretase, resulting in the release of the γ-fragment, which contains the entire cytoplasmic domain, into the cytoplasm. In this study, we identified a protein presumed to be the γ-fragment of long UNC5B, which accumulates only when shedding is activated and proteasome is inhibited. It has been reported that the γ-fragments of Notch and Neogenin receptors translocate to the nucleus and regulate transcription (33, 34). We would like to investigate whether the γ-fragment of UNC5B translocates to the nucleus and regulates transcription in our future studies. Unfortunately, because of the absence of antibodies that recognize the extracellular domain of UNC5B, we have not yet succeeded in observing the shedding of endogenous UNC5B. Whether or not endogenous UNC5B is subject to shedding, and if so, whether the shedding is affected by the extracellular environment, such as the presence of Netrin-1, and what functions the processing products perform, are all questions that future research needs to address.

The skipping exon of UNC5B encodes the juxtamembrane region, which corresponds to approximately 1% of the full-length protein, indicating the accuracy of the regulatory mechanism of shedding susceptibility. By analyzing the shedding of CADM1 and ALCAM isoforms, we hypothesized that shedding susceptibility is determined by unfavourable factors for sheddases such as O-glycans and negatively charged amino acids (6, 7). Our hypothesis was that sheddase tends to cleave at a certain distance from the cell surface and that only membrane proteins with no unfavourable factors around there are susceptible to shedding. In the case of long UNC5B, the shedding cleavage site is located 16 amino acids from the cell surface, and short UNC5B has a TSP type 1 domain at that position. We believe that this study supports our hypothesis and proposes ‘domain structure’ as a new unfavourable factor for sheddases.

The expression ratio of UNC5B isoforms differs depending on the cell type and developmental stage (37); thus, even if cells express UNC5B, the responsiveness to Netrin-1 can be altered by alternative splicing. Netrin-1 expression is upregulated in various cancers (11, 12), and cancer therapy using an anti-Netrin-1 neutralizing antibody, which blocks the interaction between Netrin-1 and UNC5B, has been developed, and its efficacy has been verified (38, 39). If this therapy is only effective in cancers expressing shedding-susceptible, Netrin-1-responsive UNC5B, the released extracellular domain of UNC5B could be used as a marker to predict the efficacy of this therapy.

In this study, we revealed that ADAM17 is partially responsible for shedding of long UNC5B. Shedding of long UNC5B was completely inhibited using metalloprotease inhibitor but not completely inhibited in ADAM17 knockout MEFs, suggesting that metalloproteinases other than ADAM17 also contribute to the shedding of long UNC5B. It is difficult to identify the responsible metalloproteases from the cleavage sequence of long UNC5B because many metalloproteases lack a consensus sequence for cleavage (40). Experiments using MEFs lacking metalloproteases may make it possible to identify the responsible metalloproteases in future.

In this study, we only analyzed the skipping exons of AXL and UNC5B among many identified alternative exons; these skipping exons are less than 40 bases in length. As skipping or inclusion of skipping exons of SIRPα, which are more than 300 bases in length, generates SIRPα isoforms with different shedding susceptibilities (6), additional alternative splicing-regulated shedding targets should exist in our list. We argue that the discovery of additional isoform-dependent targets for shedding is a future challenge and that it would be informative to analyze the alternative splicing events of membrane proteins in relation to their susceptibility to shedding.

Supplementary Data

Supplementary Data are available at JB Online.

Supplementary Material

Web_Material_mvaf043
web_material_mvaf043.zip (567.4KB, zip)

Acknowledgements

We thank Mr. Naoki Matsuda and Mr. Yuta Kawai for technical assistance. The authors used DeepL for the Japanese–English translation. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. We thank Editage (www.editage.jp) for English language editing.

Contributor Information

Kotaro Sugimoto, Department of Biomedical Sciences, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan.

Eichi Watabe, Medical Research Institute, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, Japan; Business Development Department, Digital Healthcare Division, Hitachi Ltd. Healthcare Innovation Division, 1-17-1 Toranomon, Minato-ku, Tokyo 105-6412, Japan.

Mio Takuma, Department of Biomedical Sciences, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan.

Kaname Nagahara, Department of Biomedical Sciences, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan.

Toshinori Sawano, Department of Biomedical Sciences, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan.

Mihoko Kajita, Department of Biomedical Sciences, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan.

Junichi Takagi, Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565-0871, Japan.

Hidehito Kuroyanagi, Medical Research Institute, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, Japan; Department of Biochemistry, Graduate School of Medicine, University of the Ryukyus, 1076 Kiyuna, Ginowan, Okinawa 901-2725, Japan.

Kyoko Shirakabe, Department of Biomedical Sciences, College of Life Sciences, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan; Ritsumeikan Global Innovation Research Organization, Ritsumeikan University, 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan.

Funding

This work was financially supported by JSPS KAKENHI (JP18K06911 (to K.S.)), Takeda Science Foundation (to K.S.), Naito Foundation (to K.S.) and a research grant from the Astellas Foundation for Research on Metabolic Disorders (to K.S.).

Conflict of Interest

The authors declare no conflict of interest. E.W. is employed by Hitachi Ltd., which had no role in the design and execution of this study, the interpretation of its findings or the writing of this manuscript.

Author Contributions

K.S., E.W., M.T. and M.K. performed the experiments and data analysis. J.T., H.K. and K.S. designed the experiments. K.S. wrote the manuscript. All authors reviewed and approved the final version of the manuscript.

References

  • 1. Lichtenthaler, S.F., Lemberg, M.K., and Fluhrer, R. (2018) Proteolytic ectodomain shedding of membrane proteins in mammals-hardware, concepts, and recent developments. EMBO J.  37, e99456 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Blobel, C.P. (2005) ADAMs: key components in EGFR signalling and development. Nat. Rev. Mol. Cell Biol.  6, 32–43 [DOI] [PubMed] [Google Scholar]
  • 3. Reiss, K. and Saftig, P. (2009) The "a disintegrin and metalloprotease" (ADAM) family of sheddases: physiological and cellular functions. Semin. Cell Dev. Biol.  20, 126–137 [DOI] [PubMed] [Google Scholar]
  • 4. De Strooper, B., Iwatsubo, T., and Wolfe, M.S. (2012) Presenilins and γ-secretase: structure, function, and role in Alzheimer disease. Cold Spring Harb. Perspect. Med.  2, a006304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Shirakabe, K., Hattori, S., Seiki, M., Koyasu, S., and Okada, Y. (2011) VIP36 protein is a target of ectodomain shedding and regulates phagocytosis in macrophage raw 264.7 cells. J. Biol. Chem.  286, 43154–43163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Shirakabe, K., Omura, T., Shibagaki, Y., Mihara, E., Homma, K., Kato, Y., Yoshimura, A., Murakami, Y., Takagi, J., Hattori, S., and Ogawa, Y. (2017) Mechanistic insights into ectodomain shedding: susceptibility of CADM1 adhesion molecule is determined by alternative splicing and O-glycosylation. Sci. Rep.  7, 46174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Iwagishi, R., Tanaka, R., Seto, M., Takagi, T., Norioka, N., Ueyama, T., Kawamura, T., Takagi, J., Ogawa, Y., and Shirakabe, K. (2020) Negatively charged amino acids in the stalk region of membrane proteins reduce ectodomain shedding. J. Biol. Chem.  295, 12343–12352 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Leonardo, E.D., Hinck, L., Masu, M., Keino-Masu, K., Ackerman, S.L., and Tessier-Lavigne, M. (1997) Vertebrate homologues of C. Elegans UNC-5 are candidate netrin receptors. Nature.  386, 833–838 [DOI] [PubMed] [Google Scholar]
  • 9. Lu, X., Le Noble, F., Yuan, L., Jiang, Q., De Lafarge, B., Sugiyama, D., Bréant, C., Claes, F., De Smet, F., Thomas, J.L., Autiero, M., Carmeliet, P., Tessier-Lavigne, M., and Eichmann, A. (2004) The netrin receptor UNC5B mediates guidance events controlling morphogenesis of the vascular system. Nature.  432, 179–186 [DOI] [PubMed] [Google Scholar]
  • 10. Sun, K.L.W., Correia, J.P., and Kennedy, T.E. (2011) Netrins: versatile extracellular cues with diverse functions. Development.  138, 2153–2169 [DOI] [PubMed] [Google Scholar]
  • 11. Mehlen, P., Delloye-Bourgeois, C., and Chedotal, A. (2011) Novel roles for slits and netrins: axon guidance cues as anticancer targets?  Nat. Rev. Cancer  11, 188–197 [DOI] [PubMed] [Google Scholar]
  • 12. Nakayama, H., Kusumoto, C., Nakahara, M., Fujiwara, A., and Higashiyama, S. (2018) Semaphorin 3F and Netrin-1: the novel function as a regulator of tumor microenvironment. Front. Physiol.  9, 1662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Raney, B.J., Barber, G.P., Benet-Pagès, A., Casper, J., Clawson, H., Cline, M.S., Diekhans, M., Fischer, C., Navarro Gonzalez, J., Hickey, G., Hinrichs, A.S., Kuhn, R.M., Lee, B.T., Lee, C.M., Le Mercier, P., Miga, K.H., Nassar, L.R., Nejad, P., Paten, B., Perez, G., Schmelter, D., Speir, M.L., Wick, B.D., Zweig, A.S., Haussler, D., Kent, W.J., and Haeussler, M. (2024) The UCSC genome browser database: 2024 update. Nucleic Acids Res.  52, D1082–d1088 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Trincado, J.L., Entizne, J.C., Hysenaj, G., Singh, B., Skalic, M., Elliott, D.J., and Eyras, E. (2018) SUPPA2: fast, accurate, and uncertainty-aware differential splicing analysis across multiple conditions. Genome Biol.  19, 40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. The UniProt Consortium (2023) UniProt: the universal protein knowledgebase in 2023. Nucleic Acids Res.  51, D523–d531 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Quinlan, A.R. and Hall, I.M. (2010) BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics.  26, 841–842 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. NCBI Resource Coordinators (2018) Database resources of the National Center for biotechnology information. Nucleic Acids Res.  46, D8–d13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Xie, Z., Bailey, A., Kuleshov, M.V., Clarke, D.J.B., Evangelista, J.E., Jenkins, S.L., Lachmann, A., Wojciechowicz, M.L., Kropiwnicki, E., Jagodnik, K.M., Jeon, M., and Ma'ayan, A. (2021) Gene set knowledge discovery with Enrichr. Curr Protoc.  1, e90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Gillespie, M., Jassal, B., Stephan, R., Milacic, M., Rothfels, K., Senff-Ribeiro, A., Griss, J., Sevilla, C., Matthews, L., Gong, C., Deng, C., Varusai, T., Ragueneau, E., Haider, Y., May, B., Shamovsky, V., Weiser, J., Brunson, T., Sanati, N., Beckman, L., Shao, X., Fabregat, A., Sidiropoulos, K., Murillo, J., Viteri, G., Cook, J., Shorser, S., Bader, G., Demir, E., Sander, C., Haw, R., Wu, G., Stein, L., Hermjakob, H., and D'Eustachio, P. (2022) The reactome pathway knowledgebase 2022. Nucleic Acids Res.  50, D687–d692 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Horiuchi, K., Kimura, T., Miyamoto, T., Takaishi, H., Okada, Y., Toyama, Y., and Blobel, C.P. (2007) Cutting edge: TNF-alpha-converting enzyme (TACE/ADAM17) inactivation in mouse myeloid cells prevents lethality from endotoxin shock. J. Immunol.  179, 2686–2689 [DOI] [PubMed] [Google Scholar]
  • 21. Shirakabe, K., Shibagaki, Y., Yoshimura, A., Koyasu, S., and Hattori, S. (2014) A proteomic approach for the elucidation of the specificity of ectodomain shedding. J. Proteome  98, 233–243 [DOI] [PubMed] [Google Scholar]
  • 22. Tsumagari, K., Chang, C.H., and Ishihama, Y. (2021) Exploring the landscape of ectodomain shedding by quantitative protein terminomics. iScience.  24, 102259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Huang, W.Y. and Wu, K.P. (2023) SheddomeDB 2023: a revision of an Ectodomain shedding database based on a comprehensive literature review and online resources. J. Proteome Res.  22, 2570–2576 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Swendeman, S., Mendelson, K., Weskamp, G., Horiuchi, K., Deutsch, U., Scherle, P., Hooper, A., Rafii, S., and Blobel, C.P. (2008) VEGF-A stimulates ADAM17-dependent shedding of VEGFR2 and crosstalk between VEGFR2 and ERK signaling. Circ. Res.  103, 916–918 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Kreitman, M., Noronha, A., and Yarden, Y. (2018) Irreversible modifications of receptor tyrosine kinases. FEBS Lett.  592, 2199–2212 [DOI] [PubMed] [Google Scholar]
  • 26. Jørgensen, J.T. (2021) The current landscape of the FDA approved companion diagnostics. Transl. Oncol.  14, 101063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Lemmon, M.A. and Schlessinger, J. (2010) Cell signaling by receptor tyrosine kinases. Cell.  141, 1117–1134 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Khokha, R., Murthy, A., and Weiss, A. (2013) Metalloproteinases and their natural inhibitors in inflammation and immunity. Nat Rev Immunol.  13, 649–665 [DOI] [PubMed] [Google Scholar]
  • 29. Dickson, B.J. (2002) Molecular mechanisms of axon guidance. Science.  298, 1959–1964 [DOI] [PubMed] [Google Scholar]
  • 30. Lemke, G. (2017) Phosphatidylserine is the signal for TAM receptors and their ligands. Trends Biochem. Sci.  42, 738–748 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Tonks, N.K. (2006) Protein tyrosine phosphatases: from genes, to function, to disease. Nat. Rev. Mol. Cell Biol.  7, 833–846 [DOI] [PubMed] [Google Scholar]
  • 32. Hua, Z., Watanabe, R., Fukunaga, T., Matsui, Y., Matsuoka, M., Yamaguchi, S., Tanabe, S.Y., Yamamoto, M., Tamura-Kawakami, K., Takagi, J., Kajita, M., Futai, E., and Shirakabe, K. (2024) C-terminal amino acids in the type I transmembrane domain of L-type lectin VIP36 affect γ-secretase susceptibility. Biochem. Biophys. Res. Commun.  696, 149504. [DOI] [PubMed] [Google Scholar]
  • 33. Sprinzak, D. and Blacklow, S.C. (2021) Biophysics of notch signaling. Annu. Rev. Biophys.  50, 157–189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Goldschneider, D., Rama, N., Guix, C., and Mehlen, P. (2008) The neogenin intracellular domain regulates gene transcription via nuclear translocation. Mol. Cell. Biol.  28, 4068–4079 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Visser, J.J., Cheng, Y., Perry, S.C., Chastain, A.B., Parsa, B., Masri, S.S., Ray, T.A., Kay, J.N., and Wojtowicz, W.M. (2015) An extracellular biochemical screen reveals that FLRTs and Unc5s mediate neuronal subtype recognition in the retina. Elife.  4, e08149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Brisset, M., Grandin, M., Bernet, A., Mehlen, P., and Hollande, F. (2021) Dependence receptors: new targets for cancer therapy. EMBO Mol Med.  13, e14495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Pradella, D., Deflorian, G., Pezzotta, A., Di Matteo, A., Belloni, E., Campolungo, D., Paradisi, A., Bugatti, M., Vermi, W., Campioni, M., Chiapparino, A., Scietti, L., Forneris, F., Giampietro, C., Volf, N., Rehman, M., Zacchigna, S., Paronetto, M.P., Pistocchi, A., Eichmann, A., Mehlen, P., and Ghigna, C. (2021) A ligand-insensitive UNC5B splicing isoform regulates angiogenesis by promoting apoptosis. Nat. Commun.  12, 4872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Lengrand, J., Pastushenko, I., Vanuytven, S., Song, Y., Venet, D., Sarate, R.M., Bellina, M., Moers, V., Boinet, A., Sifrim, A., Rama, N., Ducarouge, B., Van Herck, J., Dubois, C., Scozzaro, S., Lemaire, S., Gieskes, S., Bonni, S., Collin, A., Braissand, N., Allard, J., Zindy, E., Decaestecker, C., Sotiriou, C., Salmon, I., Mehlen, P., Voet, T., Bernet, A., and Blanpain, C. (2023) Pharmacological targeting of netrin-1 inhibits EMT in cancer. Nature.  620, 402–408 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Cassier, P.A., Navaridas, R., Bellina, M., Rama, N., Ducarouge, B., Hernandez-Vargas, H., Delord, J.P., Lengrand, J., Paradisi, A., Fattet, L., Garin, G., Gheit, H., Dalban, C., Pastushenko, I., Neves, D., Jelin, R., Gadot, N., Braissand, N., Leon, S., Degletagne, C., Matias-Guiu, X., Devouassoux-Shisheboran, M., Mery-Lamarche, E., Allard, J., Zindy, E., Decaestecker, C., Salmon, I., Perol, D., Dolcet, X., Ray-Coquard, I., Blanpain, C., Bernet, A., and Mehlen, P. (2023) Netrin-1 blockade inhibits tumour growth and EMT features in endometrial cancer. Nature.  620, 409–416 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Hey, S. and Linder, S. (2024) Matrix metalloproteinases at a glance. J. Cell Sci.  137, jcs261898 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Web_Material_mvaf043
web_material_mvaf043.zip (567.4KB, zip)

Articles from Journal of Biochemistry are provided here courtesy of Oxford University Press

RESOURCES