Fig 1.
Generation of matriptase NTF monoclonal antibodies.
A. Matriptase NTF was expressed and purified. The purified proteins were analyzed by SDS-PAGE and visualized by Coomassie Blue (lane 1). B. The 50- and 35-kDa protein bands were identified by in-gel trypsinization and tandem mass spectrometry (MS/MS). The identified tryptic peptides from the 50-kDa protein matching matriptase are indicated in red and from the 35-kDa protein in blue. The SEA domain cleavage site Gly149-Ser-150 (GS) in green is also indicated. C. Two different amounts of 184A1N4 human mammary epithelial cell lysate were analyzed by western blot under non-reducing and non-boiled condtions for matriptase species using the mAbs PS6 (lanes 1 and 2), PS7 (lanes 3 and 4) and M24 (lanes 5 and 6). FL stands for full-length, NTF for N-terminal fragment, and CTF for C-terminal fragment. The immunoblot data presented are representative examples of at least two independent experiments.
Fig 2.
Matriptase is tethered on the plasma membrane in spite of the cleavage of the SEA domain.
HaCaT human keratinocyte cells lysates (lanes 1, L) were incubated with the matriptase NTF mAb PS6-Sepharose (PS6 IP, lanes 1–5) or CTF mAb M24-Sepharose (M24 IP, lanes 6–10). The unbound fraction (lanes 2 and 7, UB) was collected as the immunodepletion fraction. The mAb-captured proteins were sequentially eluted into three fractions (lanes 3–5 and 8–10, E1-E3). These samples were analyzed by western blot under non-reducing and non-boiled condtions for matriptase CTF by the mAb M24 (A, M24 WB) and matriptase NTF by the mAb PS6 (B, PS6 WB). The immunodepletion-immunoblot data presented are representative examples of at least two independent experiments.
Fig 3.
Matriptase shedding requires de novo proteolytic cleavages.
Six different human cell lines, as indicated, were grown in the medium containing no serum for 24 hrs. The conditioned media were collected and concentrated. The cell lysates (lanes 1) and conditioned medium (lanes 2) were analyzed by western blot under non-reducing and non-boiled conditions for matriptase species using the matriptase mAb M24. The samples were analyzed by loading various sample volumes in order to clearly demonstrate the sizes of cell-associated and shed matriptase. The analyses of matriptase species in the cell lysates and conditioned media from cells was conducted at least two times, and representative data are presented.
Fig 4.
Matriptase shedding is temporally coupled with matriptase zymogen activation.
A. LNCaP human prostate adenocarcinoma cells (lanes 1–3) or the insoluble fractions of the cells (lanes 4–6) were exposed to PBS as the non-activation control (lanes 1 and 4, Control) or phosphate buffer pH 6.0 for 20 min to induce matriptase zymogen activation (lanes 2 and 5, Activation). The conditioned buffer was collected as the shed fraction (lanes 3 and 6, Shed). The lysates and the conditioned buffer were analyzed by western blot for matriptase CTF using the mAb M24 (upper panel) and for matriptase NTF using the mAb PS6 (lower panel). Various matriptase (MTP) species were indicated, including the three matriptase-HAI-1 complexes with apparent masses of 120-, 110-, and 95-kDa, and full-length, cell-associated, and shed matriptase (MTP). B. HaCaT human keratinocytes were exposed to a phosphate buffer pH 6.0 for indicated time from 0 to 7 min or PBS as control (lanes C). The shed fractions were concentrated to the same volume as that of the cell lysate. Equal volume of the cell lysate (Cell-associated, upper panel) and the concentrated conditioned buffer (Shed fraction, lower panel) were analyzed by immunoblot under non-reducing and non-boiled conditions for matriptase species. The data presented are representative of that obtained in more than three independent experiments.
Fig 5.
Matriptase shedding depends on matriptase zymogen activation and cleavage at Arg or Lys residues in the link region between the SEA and the CUB domain.
Wild-type (lanes 2, WT) or matriptase mutants with point mutation at Ser-805 (lanes 3, S805A), mutations at Lys-189 and Arg-186 together (lanes 4, K189A+R186A), mutations at Lys-204 and Arg-208 together (lanes 5, K204A+R208A), or mutations at all four positively charged amino acid residues together (lanes 6, combined) were expressed in 293T HEK cells. The parental cells (lanes 1, Control) and the transfected cells were exposed to a pH 6.0 buffer to induce matriptase zymogen activation. The cell lysates (A. and B. Cell-associated) and the conditioned buffers (C. and D. Shed) were analyzed by immunoblot under non-reducing and non-boiled conditions for matriptase species (A. C. and D., Total Matriptase) and HAI-1 species (B. Total HAI-1). In order to detect the trace matriptase-HAI-1 complexes and other matriptase species shed, the immunoblots were analyzed with both shorter exposure (C.) and longer exposure (D.). The shed fractions were concentrated to the same volume as that of the cell lysate. Equal volume of the cell lysate and the shed fraction were subjected to SDS-PAGE. Matriptase and HAI-1 species are indicated. FL stands for full-length and Zym for zymogen. The matriptase-HAI-1 complexes with slightly greater mass are indicated with *. The data are representative of more than 3 independent experiments.
Fig 6.
Arg-186 is the cleavage site responsible for matriptase shedding following matriptase zymogen activation.
The wild-type (lanes 1, WT) and matriptase mutants, including point mutation at Arg-186 (lanes 2, R186A), point mutation at Lys-189 (lanes 3, K189A), or mutations at Arg-186 and Lys-189 together (lanes 4, comb.) were expressed in 293T HEK cells. The cells were exposed to a pH 6.0 buffer to induce matriptase zymogen activation. The cell lysates (A. Cell-associated) and the conditioned buffers (B. Shed) were analyzed by immunoblot under non-reducing and non-boiled conditions for matriptase species (A. and B, Matriptase) and HAI-1 species (A. HAI-1). Matriptase and HAI-1 species are indicated. MTP stands for matriptase and FL for full-length. The shed fractions were concentrated to the same volume as that of cell lysate. Equal volume of the cell lysate and the concentrated were subjected to SDS-PAGE. C. The amino acid sequences of the linker region between the SEA domain and the CUB domain of matriptase are compared among various species, including Homo sapiens (human, Swiss-Prot accession no. Q9Y5Y6), Pan troglodytes (chimpanzee, Swiss-Prot accession no. H2Q547), Gorilla gorilla gorilla (gorilla, Swiss-Prot accession no. G3S550), Rattus norvegicus (rat, Swiss-Prot accession no. Q9JJI7), Mus musculus (mouse, Swiss-Prot accession no. P56677), Cavia porcellus (Guinea pig, Swiss-Prot accession no. H0VJD4), and Oryctolagus cuniculus (rabbit, Swiss-Prot accession no. G1SX25). The full protein sequences were obtained from Uniprot. Sequences were aligned by Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets using the MUSCLE method [41]. The data are representative of at least 3 independent experiments.
Fig 7.
Matriptase in complex HAI-1 is a shed species.
HaCaT human keratinocyte cells were induced to activate matriptase by pH 6.0 buffer treatment. Matriptase-HAI-1 complex in the lysate was immunodepleted with HAI-1 mAb-Sepharose. The cells lysate (lanes 1, 3, and 5) and the immunodepleted fraction (lanes 2, 4, and 6) were analyzed by western blot under non-reducing and non-boiled conditions for matriptase CTF species (lanes 1 and 2), HAI-1 species (lanes 3 and 4), and matriptase NTF (lanes 5 and 6). MTP stands for matriptase, CTF for C terminal fragment, and NTF for N terminal fragment. The data presented are representative of at least 2 independent experiments.
Fig 8.
The summary and schematic illustration of matriptase life cycle.
The matriptase life cycle has three distinct stages which are summarized by schematic illustrations of the changes in the protein domain structures. A description of matriptase life cycle can be found in the Discussion. SPD stands for the serine protease domain; LDLR for the LDL receptor class A domain; CUB for the CUB (complement C1r/C1s, Uegf, Bmp1) domain; SEA for the SEA (sea urchin sperm protein, enterokinase and agrin) domain; TM for the transmembrane domain; CD for the cytoplasmic domain; N for the N-terminus; and C for the C-terminus.