Sure F , Bertog M , Afonso S , Diakov A , Rinke R , Madej MG , Wittmann S , Gramberg T , Korbmacher C , Ilyaskin AV .

             serine protease inhibitor complex

	Figure 1. Stimulatory effect of TMPRSS2 on ENaC requires proteolytic activity of TMPRSS2. A, representative whole-cell current traces are shown for oocytes expressing human wildtype αβγ-ENaC alone (left trace, ENaC) or coexpressing αβγ-ENaC with human wildtype TMPRSS2 (middle trace, ENaC + TMPRSS2) or catalytically inactive TMPRSS2 (right trace, ENaC + TMPRSS2S441A). HA tag epitope was attached to the C terminus of TMPRSS2. HA tag neither disturbed the proteolytic activity of TMPRSS2 nor affected the stimulatory effect of TMPRSS2 on ENaC (Fig. S1). Amiloride (ami, 2 μM) and chymotrypsin (chym, 2 μg/ml) were present in the bath solution as indicated by black and gray bars, respectively. Dashed lines indicate zero current level. B, ENaC-mediated ami-sensitive whole-cell currents (ΔIami) were determined from similar experiments as shown in (A) by subtracting the baseline current in the presence of ami from the current level reached in its absence before (−) or after (+) chym application. Lines connect data points obtained in an individual oocyte. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ns, Kruskal–Wallis with Dunn’s post hoc test (51 ≤ n ≤ 52, N = 5). C, relative stimulatory effect of chym on ΔIami summarized from data shown in (B). Dashed line indicates normalized ΔIami value of one (no effect). Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ns, one-way ANOVA with Bonferroni post hoc test. D, in parallel experiments to those shown in (A–C), trypsin-like proteolytic activity at the cell surface was detected in the same batches of oocytes. Progress curves of proteolytic activity (RFU = relative fluorescent unit; mean ± SD) are shown. In each individual recording RFU values were normalized to the initial RFU value at the beginning of the measurement. ∗∗∗p < 0.001; Kruskal–Wallis with Dunn’s post hoc test (at the time point 190 min; 41≤ n ≤ 46, N = 5). E, representative western blots showing intracellular (left upper panel) or cell surface (right upper panel) expression of HA-tagged wildtype TMPRSS2 or mutant TMPRSS2S441A in oocytes from one batch. No specific signal was detected with the anti-HA antibody in oocytes expressing ENaC alone. TMPRSS2 zymogen (∼65 kDa) and TMPRSS2 in its activated cleaved form (catalytic chain, ∼27 kDa) are indicated by open and filled arrowheads, respectively. To validate separation of cell surface proteins from intracellular proteins, blots were stripped and reprobed using an antibody against β-actin (lower panels). Similar results were obtained in three additional repeats (n = 4). ENaC, epithelial sodium channel; HA, hemagglutinin; ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure 2. Coexpression of ENaC with TMPRSS2 largely increases average channel open probability. A, representative whole-cell current traces are shown for oocytes expressing wildtype α-ENaC and γ-ENaC together with a mutant β-ENaC subunit carrying a single-point mutation (S520C) without TMPRSS2 (left traces, αβS520Cγ-ENaC), with coexpression of wildtype TMPRSS2 (middle traces, αβS520Cγ-ENaC + TMPRSS2) or catalytically inactive TMPRSS2 (right traces, αβS520Cγ-ENaC + TMPRSS2S441A). In each individual oocyte, current measurement was performed before and after 5 min of incubation in ND96 bath solution containing MTSET (1 mM) and amiloride (ami; 2 μM). For the current measurement, an oocyte was clamped at a holding potential of −60 mV. The oocyte was unclamped during the incubation time in the presence of MTSET to minimize sodium loading of the oocytes. Before the second current measurement, MTSET was washed out with ND96 containing 2 μM ami. Impaling microelectrodes were not removed from the oocyte until the end of the experiment. The presence of ami (2 μM) in the bath solution is indicated by filled bars. Dashed lines indicate zero current level. B, summary of ΔIami values obtained in similar experiments as shown in (A). Lines connect data points obtained in an individual oocyte. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ∗∗p < 0.01; ns, one-way ANOVA with Bonferroni post hoc test (n = 20, N = 3). C, relative stimulatory effect of MTSET on ΔIami summarized from data shown in (B). Dashed line indicates normalized ΔIami value of one (no effect). Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ns, Kruskal–Wallis with Dunn’s post hoc test. ENaC, epithelial sodium channel; MTSET, 2-(trimethylammonium)ethyl methanethiosulfonate bromide; ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure 3. Aprotinin abolishes the proteolytic activity of TMPRSS2 at the cell surface but not its stimulatory effect on ENaC. A, representative whole-cell current traces are shown for oocytes expressing ENaC without or with coexpression of TMPRSS2, as indicated. After cRNA injection, oocytes were incubated for 48 h in standard incubation solution with aprotonin (two left panels, + aprotinin, 100 μg/ml) or without aprotinin (two right panels, − aprotinin). In experiments with aprotinin-pretreated oocytes, aprotinin (100 μg/ml) was also present in the bath solution throughout the whole-cell current recordings as indicated by gray bars. Amiloride (ami, 2 μM) was present in the bath solution as indicated by black bars. Dashed lines indicate zero current level. B, summary of ΔIami values obtained in similar experiments as shown in (A). Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ∗p < 0.05; ns, one-way ANOVA with Bonferroni post hoc test (n = 40, N = 3). C, progress curves of trypsin-like proteolytic activity at the cell surface (RFU = relative fluorescent unit; mean ± SD) were obtained as described for Figure 1D in parallel experiments with oocytes from the same batches as shown (A and B). ∗∗∗p < 0.001; ns, Kruskal–Wallis with Dunn’s post hoc test (at the time point 190 min, n = 19, N = 3). cRNA, complementary RNA; ENaC, epithelial sodium channel; ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure 4. TMPRSS2-dependent ENaC stimulation was associated with the appearance of fully cleaved γ-ENaC in the intracellular and cell surface protein fraction. A, schematic diagram showing γ-ENaC cleavage fragments, which can be detected using an antibody (in blue) raised against a C-terminal γ-ENaC epitope. The expected molecular weights of the corresponding C-terminal γ-ENaC cleavage fragments are given below. B, representative western blots showing cell surface (left upper panel) or intracellular (right upper panel) expression of γ-ENaC in oocytes from one batch expressing ENaC alone or coexpressing ENaC with TMPRSS2. Specific signal of γ-ENaC was detected using an antibody against the C-terminal epitope of γ-ENaC. To increase ENaC expression and improve γ-ENaC detection in western blot experiments, oocytes were injected with more than the usual amount of cRNA for ENaC (1 ng/subunit/oocyte) and TMPRSS2 (5 ng/oocyte). Noninjected oocytes served as a control (n.i.). Uncleaved (∼87 kDa), partially cleaved (∼76 kDa), and fully cleaved γ-ENaC (∼67 kDa) are indicated by open, gray, and black-filled arrowheads, respectively. A putative γ-ENaC degradation product (∼18 kDa) is indicated by an open arrowhead with dot pattern. To validate separation of cell surface proteins from intracellular proteins, blots were stripped and reprobed using an antibody against β-actin (lower panels). Similar results were obtained in four additional repeats (n = 5). C, in parallel experiments to those shown in (B), ΔIami values were measured to confirm the stimulatory effect of TMPRSS2 on ENaC in these batches of oocytes (n = 37, N = 5). Note that the relative stimulatory effect of TMPRSS2 on ΔIami was similar to that for Figure 1, Figure 2, Figure 3, but the absolute current values were higher, which reflects increased ENaC expression because of the larger amount of cRNA injected in these experiments. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; two-tailed unpaired Student’s t test. cRNA, complementary RNA; ENaC, epithelial sodium channel; TMPRSS2, transmembrane serine protease 2.
	Figure 5. Preventing the release of the inhibitory tract from γ-ENaC abolishes the stimulatory effect of TMPRSS2 on ENaC. A, ribbon diagram of extracellular domains of human αβγ-ENaC generated using atom coordinates from PDB entry 6WTH (8, 77). The putative location of unresolved transmembrane domains is indicated with a box placed within the plasma lipid bilayer (outer and inner leaflets), which is schematically depicted with dipalmitoylphosphatidylcholine (DPPC) molecules in stick representation. The inset shows the location of the specific binding site of the key inhibitory amino acid sequence (γ-11, in yellow) of the γ-inhibitory tract on an expanded scale. Serine (S155) and glutamine (Q426) residues, which were substituted by cysteines to introduce a disulfide bond between γ-inhibitory tract and its binding site, are indicated with arrows and shown in stick representation with side-chain carbons in orange for S155 or green for Q426, nitrogen in blue, and oxygens in red. Hydrogen atoms are omitted for clarity. B, representative western blots showing intracellular (left upper panel) or cell surface (right upper panel) expression of γ-ENaC in oocytes expressing wildtype (αβγ) or mutant (αβγS155C;Q426C) ENaC without or with TMPRSS2 coexpression. Specific signal of γ-ENaC was detected using the same antibody as for Figure 4. Noninjected oocytes served as control (n.i.). Positions of uncleaved (∼87 kDa), partially cleaved (∼76 kDa), and fully cleaved γ-ENaC (∼67 kDa) are indicated by open, gray, and black-filled arrowheads, respectively. To validate separation of cell surface proteins from intracellular proteins, blots were stripped and reprobed using an antibody against β-actin (lower panels). Similar results were obtained in another repeat (n = 2). C and E, representative whole-cell current traces are shown for oocytes expressing the mutant ENaC (αβγS155C;Q426C, C) or wildtype ENaC (αβγ, E) without (left panels) or with TMPRSS2 coexpression (right panels). In each individual oocyte currents were measured before and after 5 min incubation in ND96 bath solution containing DTT (30 mM) and amiloride (ami; 2 μM). The experimental protocol was similar to that described for Figure 2. The presence of ami (2 μM) is indicated by filled bars. Dashed lines indicate zero current level. D and F, summary of ΔIami values obtained in similar experiments as shown in (C) and (E). Lines connect data points obtained in an individual oocyte. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ns, Kruskal–Wallis with Dunn’s post hoc test (16≤ n ≤ 18, N = 3). ENaC, epithelial sodium channel; ns, not significant; PDB, Protein Data Bank; TMPRSS2, transmembrane serine protease 2.
	Figure 6. Prediction of putative TMPRSS2 cleavage sites distal to the γ-inhibitory tract using a molecular docking approach. A, the primary sequence of human γ-ENaC (amino acid residues 134–192) is shown in the region including the γ-inhibitory tract (underlined in yellow). The key inhibitory amino acid sequence (γ-11) is highlighted with a yellow rectangle. Candidate TMPRSS2 cleavage sites (arginine and lysine residues) distal to the γ-inhibitory tract are in bold and numbered. The length of the amino acid sequence released by cleavage may slightly vary depending on the cleavage site used in the region indicated by the dashed yellow line. Sequences of three 6-mer peptides, which were docked to the catalytic domain of TMPRSS2 in computer simulations, are shown below the γ-ENaC sequence. The numbering of amino acid residues in the simulated peptides is the same as in the γ-ENaC sequence. B and C, a homology model of human TMPRSS2 generated based on the crystal structure of human homologous protease hepsin (PDB accession no.: 1Z8G) is shown in ribbon (B) or electrostatic potential molecular surface representation (C). In (B), amino acid residues forming the catalytic triad (histidine H296, aspartate D345, and serine S441) and the aspartate residue D435 at the bottom of the S1 pocket are shown in stick representation with carbons in tan, nitrogens in blue, and oxygens in red. In (C), a representative binding mode of the GKARDF-peptide to the TMPRSS2 catalytic domain (in the inset on an expanded scale), which fulfills the selection criteria described in Figure S2A, is depicted in stick representation with carbons in the same color as the corresponding amino acid residue of the peptide sequence given in the lower left corner of the inset, nitrogens in blue, and oxygens in red. Hydrogen atoms are omitted for clarity. Scissile peptide bond is marked in the inset with a scissors symbol (✄). D–F, all binding modes that fulfill the selection criteria are shown for the GKARDF-peptide (D), GRKRKV-peptide (E), or IHKASN-peptide (F). Peptide backbone carbons and nitrogens (in the same color as the corresponding amino acid residue of the peptide sequence given in the lower left corner) and the side chains of arginine or lysine residues occupying the S1 pocket (with carbons in white and nitrogens in blue) are shown. Bar diagrams demonstrate the percentage of binding modes, which fulfill the selection criteria out of the total number of binding modes (90) generated for each peptide, and indicate the arginine or lysine residue that occupies the S1 pocket. ENaC, epithelial sodium channel; PDB, Protein Data Bank; TMPRSS2, transmembrane serine protease 2.
	Figure 7. γ-ENaC constructs generated by site-directed mutagenesis to identify TMPRSS2 cleavage sites distal and proximal to the γ-inhibitory tract. Comparison of the primary sequence of wildtype γ-ENaC with those of γ-ENaC mutants generated and experimentally tested in the current study. The γ-inhibitory tract is labeled as described in Figure 6A. The putative TMPRSS2 cleavage sites (arginine and lysine residues) are marked in bold black. Putative cleavage sites eliminated by alanine substitutions are marked in bold red. Results from corresponding functional experiments are shown in Figure 8, Figure 9, Figure 10. ENaC, epithelial sodium channel; TMPRSS2, transmembrane serine protease 2.
	Figure 8. Eliminating putative TMPRSS2 cleavage sites distal to the γ-inhibitory tract significantly reduces the stimulatory effect of TMPRSS2 coexpression on ENaC and prevents the appearance of fully cleaved γ-ENaC in whole-cell lysates. A, representative whole-cell current traces recorded in individual oocytes from the same batch injected with 1 ng/subunit/oocyte of cRNA for wildtype (αβγ) or mutant ENaC (αβγRKRK178AAAA, αβγK168A;K170A;R172A;K189A or αβγRKRK178AAAA;K168A;K170A;R172A;K189A) either alone (−) or in combination with 5 ng/oocyte cRNA for TMPRSS2 (+). Amiloride (ami, 2 μM) was present in the bath solution as indicated by black bars. Dashed lines indicate zero current level. Summary data obtained in similar experiments are shown to the right of the representative traces. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; two-tailed unpaired Student’s t test (22≤ n ≤ 61, 3 ≤ N ≤ 9). B, ΔIami values of individual oocytes obtained in the same experiments as shown in (A) were normalized to the corresponding mean ΔIami recorded in oocytes from the same batch expressing wildtype (αβγ) or mutant ENaC (αβγRKRK178AAAA, αβγK168A;K170A;R172A;K189A, or αβγRKRK178AAAA;K168A;K170A;R172A;K189A) without TMPRSS2 coexpression. Dashed line indicates a normalized ΔIami value of one (no effect). Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ∗p < 0.05; ns; Kruskal–Wallis with Dunn’s post hoc test. C, representative western blots showing whole-cell expression of γ-ENaC detected using the same antibody as for Figure 4 in oocytes expressing wildtype (αβγ) or mutant ENaC (αβγRKRK178AAAA, αβγK168A;K170A;R172A;K189A, or αβγRKRK178AAAA;K168A;K170A;R172A;K189A) either without (−) or with (+) TMPRSS2 coexpression. Noninjected oocytes served as control (n.i.). Uncleaved (∼87 kDa), partially cleaved (∼76 kDa), and fully cleaved γ-ENaC (∼67 kDa) are indicated by open, light gray, and dark gray–filled arrowheads, respectively. Similar results were obtained in additional repeats shown in Fig. S3. cRNA, complementary RNA; ENaC, epithelial sodium channel; ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure 9. TMPRSS2 cleaves γ-ENaC at sites proximal to the γ-inhibitory tract. A, representative whole-cell current traces recorded in individual oocytes from the same batch injected with 1 ng/subunit/oocyte of cRNA for wildtype (αβγ) or mutant ENaC (αβγR138A) either alone (−) or in combination with 5 ng/oocyte cRNA for TMPRSS2 (+). V5-tag epitope was attached to the N terminus of γ-ENaC. Amiloride (ami, 2 μM) was present in the bath solution as indicated by black bars. Dashed lines indicate zero current level. Summary data obtained in similar experiments are shown to the right of the representative traces. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; two-tailed unpaired Student’s t test (27≤ n ≤ 31, N = 5). B, ΔIami values of individual oocytes obtained in the same experiments as shown in (A) were normalized as described for Figure 8B. Dashed line indicates a normalized ΔIami value of one (no effect). Mean ± SD and data points for individual oocytes are shown; ns; one-way ANOVA with Bonferroni post hoc test. C, schematic diagram showing γ-ENaC cleavage fragments, which can be detected using an antibody raised against a C-terminal γ-ENaC epitope (in blue) or an anti-V5 antibody (in green). The expected molecular weights of the corresponding C-terminal and N-terminal γ-ENaC cleavage fragments are given below in the respective color. D, representative western blots showing whole-cell expression of γ-ENaC detected using the C-terminal anti-γ-ENaC (upper panel) or N-terminal anti-V5 antibody (lower panel) in oocytes expressing wildtype (αβγ) or mutant ENaC (αβγR138A) either without (−) or with (+) TMPRSS2 coexpression. Noninjected oocytes served as control (n.i.). Uncleaved γ-ENaC (∼87 kDa), cleaved in the proximal (∼76 kDa and ∼15 kDa) or distal (∼67 kDa and ∼21 kDa) regions of the γ-inhibitory tract, are indicated by open, gray, and black-filled arrowheads, respectively. A putative N-terminal γ-ENaC degradation product (∼12 kDa) is indicated by an open arrowhead with dot pattern. Similar results were obtained in three additional repeats (n = 4). cRNA, complementary RNA; ENaC, epithelial sodium channel; ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure 10. The arginine residue R153 at the proximal end of γ-11 may serve as another TMPRSS2-cleavage site mediating partial ENaC activation. A, representative whole-cell current traces recorded in individual oocytes from the same batch injected with 1 ng/subunit/oocyte of cRNA for wildtype (αβγ) or mutant ENaC (αβγRKRR138AAAA or αβγRKRR138AAAA + R153A) either alone (−) or in combination with 5 ng/oocyte cRNA for TMPRSS2 (+). Amiloride (ami, 2 μM) was present in the bath solution as indicated by black bars. Dashed lines indicate zero current level. Summary data obtained in similar experiments are shown to the right of the representative traces. Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ns, two-tailed Mann–Whitney test (18 ≤ n ≤ 25, N = 3). B, ΔIami values of individual oocytes obtained in the same experiments as shown in (A) were normalized as described for Figure 8B. Dashed line indicates a normalized ΔIami value of one (no effect). Mean ± SD and data points for individual oocytes are shown; ∗∗∗p < 0.001; ∗p < 0.05; Kruskal–Wallis with Dunn’s post hoc test. C, summary of ΔIami values obtained in oocytes injected with 1 ng/subunit/oocyte of cRNA for wildtype (αβγ) or mutant ENaC (αβγRKRK178AAAA;K168A;K170A;R172A;K189A + R153A) either alone (−) or in combination with 5 ng/oocyte cRNA for TMPRSS2 (+). Mean ± SD and data points for individual oocytes are shown; ∗∗p < 0.01; ns, Kruskal–Wallis with Dunn’s post hoc test. cRNA, complementary RNA; ENaC, epithelial sodium channel; TMPRSS2, transmembrane serine protease 2.
	Figure 11. TMPRSS2 knockdown in H441 human airway epithelial cells does not affect normal epithelial monolayer formation in culture. A, left panel, representative western blot showing endogenous expression of TMPRSS2 in H441 cells without (wildtype) or with TMPRSS2 knockdown (models 1 and 2) detected using a TMPRSS2-specific antibody. TMPRSS2 in its activated cleaved form (catalytic chain, ∼24 kDa) is indicated by a filled arrowhead. Right panel, densitometric evaluation of TMPRSS2 expression from similar blots as shown in left panel. In each blot, the density value of the ∼24 kDa TMPRSS2 band obtained for TMPRSS2-knockdown model 1 (n = 13) or model 2 (n = 9) was normalized to that of the corresponding TMPRSS2 band obtained in wildtype. Dashed line indicates a normalized density value of one (no effect). Mean ± SD and individual data points are shown; ∗∗∗p < 0.001; one-sample Student’s t test compared with wildtype (1.0). B, transepithelial electrical resistance (TEER) values recorded in wildtype or two different TMPRSS2-knockdown H441 cell models on day 8 after seeding cells on permeable supports. Mean ± SD and individual data points are shown; ns, one-way ANOVA with Bonferroni post hoc test (49 ≤ n ≤ 56). C, immunofluorescence staining for the tight junction Zonula occludens-1 protein (ZO-1, in green) in control or two different TMPRSS2-knockdown H441 cell models was performed on day 9 after seeding cells on permeable supports. One representative image is shown (n = 3). ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure 12. TMPRSS2 is involved in proteolytic ENaC activation in H441 cells. A and C, representative equivalent short circuit current (ISC) recordings are shown from wildtype H441 cells (wildtype, left traces) or TMPRSS2-knockdown H441 cells from model 1 (TMPRSS2 knockdown, right traces). B and D, summary data obtained from similar experiments as shown in (A and C). Trypsin (20 μg/ml) and amiloride (ami, 10 μM) were present in the apical bath solution as indicated by open and filled horizontal bars, respectively. Initial parts of recordings (∼30 min) corresponding to the equilibration phase after transferring the cells into Ussing chambers and applying Ringer’s solution to the apical compartment are omitted for clarity. Effect of trypsin on ISC (ΔISC trypsin; open data points) in the absence (B) or the presence (D) of apical ami determined in each individual recording by subtracting the ISC measured before trypsin application from the current level reached in the presence of trypsin before ami application (B) or at the end of the recording (D). Effect of ami on ISC (ΔISC ami; filled data points) in the presence (B) or the absence (D) of apical trypsin was determined in each individual recording by subtracting the ISC measured before ami application from the current level reached in the presence of ami at the end of the recording (B) or immediately before trypsin application (D). Absolute ISC values obtained in these experiments, which were used to calculate ΔISC trypsin and ΔISC ami, are shown in Figure S6. Mean ± SD and individual data points are shown; ∗∗∗p < 0.001; ∗∗p < 0.01; ns, two-tailed Mann–Whitney test (14 ≤ n ≤ 20). ENaC, epithelial sodium channel; ns, not significant; TMPRSS2, transmembrane serine protease 2.
	Figure S1. TMPRSS2 without C-terminal HA-tag also proteolytically activates αβγ-ENaC. (A) representative whole-cell current traces recorded in a human αβγ-ENaC expressing oocyte without (left trace, ENaC) or with (right trace, ENaC+TMPRSS2) human TMPRSS2 coexpression. In these control experiments a TMPRSS2 construct was used without HA-tag attached to its C-terminus. Amiloride (ami, 2 µM) and chymotrypsin (chym, 2 µg/ml) were present in the bath solution as indicated by black and grey bars, respectively. Dashed lines indicate zero current level. (B) ENaC-mediated amiloride-sensitive whole-cell currents (ΔIami) were determined as described in Fig. 1B from similar experiments as shown in (A). Lines connect data points obtained in an individual oocyte. Mean ± SD and data points for individual oocytes are shown; ***p < 0.001; **p < 0.01; n.s., not significant, Kruskall-Wallis with Dunn’s post hoc test (n=17, N=3). (C) relative stimulatory effect of chymotrypsin on ΔIami summarized from data shown in (B). Dashed line indicates a normalized ΔIami value of one (no effect). Mean ± SD and data points for individual oocytes are shown; ***p < 0.001; two-tailed unpaired Student’s t test. (D) In parallel experiments to those shown in (A-C), trypsin-like proteolytic activity at the cell surface (RFU = relative fluorescent unit; mean ± SD) was detected in these batches of oocytes as described in Fig. 1D. ***p < 0.001; two-tailed Mann-Whitney test (at the time point 190 min; n=22, N=3).
	Figure S2. Prediction of putative TMPRSS2 cleavage sites distal to the γ-inhibitory tract using a molecular docking approach. (A) Schematic diagram illustrating the molecular docking strategy used to predict binding modes of 6-mer peptides corresponding to different segments of the distal region of the γ-inhibitory tract to the catalytic domain of TMPRSS2 which are compatible with proteolysis, i.e. fulfill the selection criteria. (B, C) All binding modes that fulfill the selection criteria for the GAARDF- (B) and GKAADF-peptide (C) are shown. (D) All generated binding modes are shown for the GAAADF-peptide, which lacks positively charged arginine or lysine residues. Peptide backbone carbons and nitrogens (in the same color as the corresponding amino acid residue of the peptide sequence given in the lower left corner) and the side chains of arginine or lysine residues occupying the S1 pocket (with carbons in white and nitrogens in blue) are shown. Bar diagrams demonstrate the percentage of the binding modes which fulfill the selection criteria out of the total number of binding modes (90) generated for each peptide, and indicate the arginine or lysine residue that occupies the S1 pocket. (E) Binding modes for each simulated peptide were subdivided into groups of binding modes that do not fulfill (−) or fulfill (+) the selection criteria. Average values of the root-mean-square deviation (RMSD) were calculated for each individual binding mode k (��������௞ തതതതതതതതതത) according to the following equation: ��������௞ തതതതതതതതതത = 1 �� − 1෍��������௞௝ ௡ ௝ୀଵ where ��������௞௝ is the RMSD value calculated between all backbone carbon and nitrogen atoms of a peptide binding mode k and the corresponding atoms of a peptide binding mode j (k≠j) from the same group, n is the total number of peptide binding modes in the corresponding group. ***p<0.001, **p<0.01, Kruskall-Wallis with Dunn’s post hoc test (7≤N≤90).
	Figure S3. Eliminating putative TMPRSS2 cleavage sites distal to the γ-inhibitory tract prevents the appearance of fully cleaved γ-ENaC in whole-cell lysates. All western blots obtained in similar experiments as shown in Fig. 8C are depicted. Parts of blots shown in Fig. 8C are framed by dotted rectangles. Whole-cell expression of γ-ENaC was detected using an antibody against a C-terminal γENaC epitope in oocytes expressing wild-type (αβγ) or mutant ENaC (αβγRKRK178AAAA, αβγK168A;K170A;R172A;K189A or αβγRKRK178AAAA;K168A;K170A;R172A;K189A) either without (−) or with (+) TMPRSS2 coexpression. Non-injected oocytes served as control (n.i.). Expression of γ-ENaC was analysed for αβγRKRK178AAAA- (A) and αβγK168A;K170A;R172A;K189A- (B) in four, for αβγRKRK178AAAA;K168A;K170A;R172A;K189A- (C) in six, and for αβγ-ENaC (A-C) in nine different repeats indicated by circles of different colours. Positions of uncleaved (~87 kDa), partially cleaved (~76 kDa) and fully cleaved γ-ENaC (~67 kDa) are indicated by open, light grey and dark grey filled arrowheads, respectively.
	Figure S4. Validation of a TMPRSS2-specific antibody using HA-tagged TMPRSS2 heterologously expressed in Xenopus laevis oocytes. Western blot analysis of intracellular and cell surface expression of C-terminally HA-tagged TMPRSS2 in oocytes from one batch was performed as described in Fig. 1E. The blot was first probed with anti-HA antibody (left two panels) and then stripped and re-probed using a commercial TMPRSS2-specific antibody (right two panels). Importantly, both antibodies specifically detected the mature catalytic chain of TMPRSS2 (~27 kDa band) as indicated by filled arrowheads. The ~3 kDa higher molecular weight observed for the catalytic chain of heterologously expressed TMPRSS2 compared to that of TMPRRS2 endogenously expressed in H441 cells (see Figs. 11A, S8A) corresponds to the expected molecular weight of the HA-tag. The detection of the zymogen form of TMPRSS2 (~65 kDa band, indicated by open arrowheads) by the TMPRSS2- specific antibody was compromised by the presence of a faint non-specific band of similar molecular weight which was also detected in oocytes without TMPRSS2 expression. Nevertheless, these control experiments confirmed the usefulness of the commercial TMPRSS2-specific antibody to detect TMPRSS2 knockdown in H441 human airway epithelial cells.
	Figure S5. Similar expression of endogenous prostasin (CAP1, PRSS8) in wild-type and TMPRSS2-knockdown H441 cells. Left panel, representative western blot showing endogenous whole-cell expression of PRSS8 in H441 cells without (wild-type) or with TMPRSS2-knockdown (model 1 and model 2) detected using a PRSS8-specific antibody. PRSS8 in its activated cleaved form (catalytic heavy chain, ~38 kDa) with hydrolyzed disulfide bonds between the light and heavy chain under reducing WB conditions is indicated by an open arrowhead. Right panel, densitometric evaluation of PRSS8 expression from similar blots as shown in left panel. In each blot the density value of the ~38 kDa PRSS8 band obtained for TMPRSS2-knockdown model 1 (n=9) or model 2 (n=5) was normalized to that of the corresponding PRSS8 band obtained in wild-type H441 cells. Dashed line indicates the normalized density value of one (no effect). Mean ± SD and individual data points are shown; *p < 0.05; n.s., not significant, one sample Student’s t test compared to wild-type (1.0).
	Figure S6. Absolute ISC values obtained in the same experiments as shown in Fig. 12. (A) Summary of baseline equivalent short circuit current (ISC) values are shown, which were measured before trypsin (same values as in B) or amiloride (same values as in C) application in wild-type H441 cells (blue points) or TMPRSS2-knockdown H441 cells (model 1, red points). Mean ± SD and individual data points are shown. 32≤n≤35; n.s., not significant; two-tailed unpaired Student’s t test. (B, C) Absolute ISC values used to calculate ΔISC trypsin and ΔISC ami shown in Fig. 12 are depicted. Lines connect data points obtained from an individual H441 cell monolayer. Mean ± SD and individual data points are shown. 14≤n≤20; ***p < 0.001; n.s., not significant; repeated measures ANOVA with Bonferroni post hoc test.
	Figure S7. Effect of apical application of trypsin and amiloride on ISC observed in the second TMPRSS2-knockdown H441 cell model (model 2). (A, B) Representative equivalent short circuit current (ISC) recordings from TMPRSS2-knockdown H441 cells (model 2) are shown on the left with corresponding summary data from similar experiments shown on the right. Experimental measurements and data analysis were performed as described in Fig. 12. Mean ± SD and individual data points are shown (16≤n≤20).
	Figure S8. Properties of luc-knockdown H441 cells are similar to those of wild-type H441 cells but differ from those of TMPRSS2-knockdown cells. TMPRSS2 protein expression detection (A, n=3) and transepithelial electrical resistance (TEER) measurements (B, 18≤n≤27) were performed and analyzed as described in Fig. 11. Equivalent short circuit current (ISC) recordings (C-F; 4≤n≤9) were performed and analyzed as described in Fig. 12. E and F summarize data from experiments similar to those shown in C and D, respectively. Data points from wild-type and TMPRSS2-knockdown H441 cells (model 1) shown in E and F are also included in the summary data shown in Figs. 11, 12 and S6.

Adebamiro, A segment of gamma ENaC mediates elastase activation of Na+ transport. 2007, Pubmed

Adebamiro, A segment of gamma ENaC mediates elastase activation of Na+ transport. 2007, Pubmed
Afar, Catalytic cleavage of the androgen-regulated TMPRSS2 protease results in its secretion by prostate and prostate cancer epithelia. 2001, Pubmed
Aguiar, Gene expression and in situ protein profiling of candidate SARS-CoV-2 receptors in human airway epithelial cells and lung tissue. 2020, Pubmed
Anand, ENaC activation by proteases. 2022, Pubmed , Xenbase
Andreasen, Activation of epithelial sodium channels by mouse channel activating proteases (mCAP) expressed in Xenopus oocytes requires catalytic activity of mCAP3 and mCAP2 but not mCAP1. 2006, Pubmed , Xenbase
Antalis, Membrane-anchored serine proteases in health and disease. 2011, Pubmed
Baker, Electrostatics of nanosystems: application to microtubules and the ribosome. 2001, Pubmed
Balchak, The epithelial Na+ channel γ subunit autoinhibitory tract suppresses channel activity by binding the γ subunit's finger-thumb domain interface. 2018, Pubmed , Xenbase
Bohnert, Aprotinin prevents proteolytic epithelial sodium channel (ENaC) activation and volume retention in nephrotic syndrome. 2018, Pubmed , Xenbase
Bohnert, Urokinase-type plasminogen activator (uPA) is not essential for epithelial sodium channel (ENaC)-mediated sodium retention in experimental nephrotic syndrome. 2019, Pubmed , Xenbase
Bowie, A method to identify protein sequences that fold into a known three-dimensional structure. 1991, Pubmed
Bruns, Epithelial Na+ channels are fully activated by furin- and prostasin-dependent release of an inhibitory peptide from the gamma-subunit. 2007, Pubmed , Xenbase
Butterworth, Modulation of the epithelial sodium channel (ENaC) by bacterial metalloproteases and protease inhibitors. 2014, Pubmed
Buzza, Prostasin is required for matriptase activation in intestinal epithelial cells to regulate closure of the paracellular pathway. 2013, Pubmed
Béliveau, Probing the substrate specificities of matriptase, matriptase-2, hepsin and DESC1 with internally quenched fluorescent peptides. 2009, Pubmed
Caldwell, Neutrophil elastase activates near-silent epithelial Na+ channels and increases airway epithelial Na+ transport. 2005, Pubmed
Carattino, Proteolytic processing of the epithelial sodium channel gamma subunit has a dominant role in channel activation. 2008, Pubmed , Xenbase
Carattino, Prostasin interacts with the epithelial Na+ channel and facilitates cleavage of the γ-subunit by a second protease. 2014, Pubmed , Xenbase
Chen, Transcriptomes of major renal collecting duct cell types in mouse identified by single-cell RNA-seq. 2017, Pubmed
Chraïbi, Protease modulation of the activity of the epithelial sodium channel expressed in Xenopus oocytes. 1998, Pubmed , Xenbase
Collier, Intersubunit conformational changes mediate epithelial sodium channel gating. 2014, Pubmed , Xenbase
Colovos, Verification of protein structures: patterns of nonbonded atomic interactions. 1993, Pubmed
Diakov, Cleavage in the {gamma}-subunit of the epithelial sodium channel (ENaC) plays an important role in the proteolytic activation of near-silent channels. 2008, Pubmed , Xenbase
Dolinsky, PDB2PQR: expanding and upgrading automated preparation of biomolecular structures for molecular simulations. 2007, Pubmed
Donaldson, Regulation of the epithelial sodium channel by serine proteases in human airways. 2002, Pubmed , Xenbase
Eaton, The contribution of epithelial sodium channels to alveolar function in health and disease. 2009, Pubmed
Essigke, Zymogen-locked mutant prostasin (Prss8) leads to incomplete proteolytic activation of the epithelial sodium channel (ENaC) and severely compromises triamterene tolerance in mice. 2021, Pubmed , Xenbase
Faller, Functional analysis of a missense mutation in the serine protease inhibitor SPINT2 associated with congenital sodium diarrhea. 2014, Pubmed , Xenbase
Frindt, Cleavage state of γENaC in mouse and rat kidneys. 2021, Pubmed
García-Caballero, ENaC proteolytic regulation by channel-activating protease 2. 2008, Pubmed
Garty, Epithelial sodium channels: function, structure, and regulation. 1997, Pubmed
Gentzsch, A Pathophysiological Model for COVID-19: Critical Importance of Transepithelial Sodium Transport upon Airway Infection. 2020, Pubmed
Guipponi, The transmembrane serine protease (TMPRSS3) mutated in deafness DFNB8/10 activates the epithelial sodium channel (ENaC) in vitro. 2002, Pubmed , Xenbase
Haerteis, The delta-subunit of the epithelial sodium channel (ENaC) enhances channel activity and alters proteolytic ENaC activation. 2009, Pubmed , Xenbase
Haerteis, An inhibitory peptide derived from the α-subunit of the epithelial sodium channel (ENaC) shows a helical conformation. 2012, Pubmed , Xenbase
Haerteis, Plasmin and chymotrypsin have distinct preferences for channel activating cleavage sites in the γ subunit of the human epithelial sodium channel. 2012, Pubmed , Xenbase
Haerteis, Proteolytic activation of the human epithelial sodium channel by trypsin IV and trypsin I involves distinct cleavage sites. 2014, Pubmed , Xenbase
Haerteis, Plasma kallikrein activates the epithelial sodium channel in vitro but is not essential for volume retention in nephrotic mice. 2018, Pubmed , Xenbase
Haerteis, Proteolytic activation of the epithelial sodium channel (ENaC) by the cysteine protease cathepsin-S. 2012, Pubmed , Xenbase
Harris, A novel neutrophil elastase inhibitor prevents elastase activation and surface cleavage of the epithelial sodium channel expressed in Xenopus laevis oocytes. 2007, Pubmed , Xenbase
Harris, Preferential assembly of epithelial sodium channel (ENaC) subunits in Xenopus oocytes: role of furin-mediated endogenous proteolysis. 2008, Pubmed , Xenbase
Hedstrom, Serine protease mechanism and specificity. 2002, Pubmed
Herter, Hepatocyte growth factor is a preferred in vitro substrate for human hepsin, a membrane-anchored serine protease implicated in prostate and ovarian cancers. 2005, Pubmed
Hobbs, Does epithelial sodium channel hyperactivity contribute to cystic fibrosis lung disease? 2013, Pubmed
Hoffmann, SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor. 2020, Pubmed
Hughey, Epithelial sodium channels are activated by furin-dependent proteolysis. 2004, Pubmed , Xenbase
Hummler, Importance of ENaC-mediated sodium transport in alveolar fluid clearance using genetically-engineered mice. 2010, Pubmed
Ilyaskin, Inhibition of the epithelial sodium channel (ENaC) by connexin 30 involves stimulation of clathrin-mediated endocytosis. 2021, Pubmed , Xenbase
Ilyaskin, Activation of the Human Epithelial Sodium Channel (ENaC) by Bile Acids Involves the Degenerin Site. 2016, Pubmed , Xenbase
Jacquinet, Cloning, genomic organization, chromosomal assignment and expression of a novel mosaic serine proteinase: epitheliasin. 2000, Pubmed
Jacquinet, Cloning and characterization of the cDNA and gene for human epitheliasin. 2001, Pubmed
Kashlan, Inhibitory tract traps the epithelial Na+ channel in a low activity conformation. 2012, Pubmed
Kashlan, Allosteric inhibition of the epithelial Na+ channel through peptide binding at peripheral finger and thumb domains. 2010, Pubmed , Xenbase
Kawabata, Highly sensitive peptide-4-methylcoumaryl-7-amide substrates for blood-clotting proteases and trypsin. 1988, Pubmed
Kellenberger, International Union of Basic and Clinical Pharmacology. XCI. structure, function, and pharmacology of acid-sensing ion channels and the epithelial Na+ channel. 2015, Pubmed
Kellenberger, An external site controls closing of the epithelial Na+ channel ENaC. 2002, Pubmed , Xenbase
Kim, Phenotypic analysis of mice lacking the Tmprss2-encoded protease. 2006, Pubmed
Kleyman, Epithelial Na+ Channel Regulation by Extracellular and Intracellular Factors. 2018, Pubmed
Kleyman, Regulating ENaC's gate. 2020, Pubmed
Ko, GalaxyWEB server for protein structure prediction and refinement. 2012, Pubmed
Kota, Energetic and structural basis for activation of the epithelial sodium channel by matriptase. 2012, Pubmed , Xenbase
Krueger, The phosphorylation site T613 in the β-subunit of rat epithelial Na+ channel (ENaC) modulates channel inhibition by Nedd4-2. 2018, Pubmed , Xenbase
Kunitz, ISOLATION FROM BEEF PANCREAS OF CRYSTALLINE TRYPSINOGEN, TRYPSIN, A TRYPSIN INHIBITOR, AND AN INHIBITOR-TRYPSIN COMPOUND. 1936, Pubmed
Laporte, Airway proteases: an emerging drug target for influenza and other respiratory virus infections. 2017, Pubmed
Laskowski, AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR. 1996, Pubmed
Lazrak, cAMP-induced changes of apical membrane potentials of confluent H441 monolayers. 2003, Pubmed
Lee, Deep Sequencing in Microdissected Renal Tubules Identifies Nephron Segment-Specific Transcriptomes. 2015, Pubmed
Lennon, The I.M.A.G.E. Consortium: an integrated molecular analysis of genomes and their expression. 1996, Pubmed
Leyvraz, The epidermal barrier function is dependent on the serine protease CAP1/Prss8. 2005, Pubmed
Loffing, Regulated sodium transport in the renal connecting tubule (CNT) via the epithelial sodium channel (ENaC). 2009, Pubmed
Lorenz, Heteromultimeric CLC chloride channels with novel properties. 1996, Pubmed , Xenbase
Lu, Small molecule activator of the human epithelial sodium channel. 2008, Pubmed , Xenbase
Lucas, The androgen-regulated protease TMPRSS2 activates a proteolytic cascade involving components of the tumor microenvironment and promotes prostate cancer metastasis. 2014, Pubmed
Lukassen, SARS-CoV-2 receptor ACE2 and TMPRSS2 are primarily expressed in bronchial transient secretory cells. 2020, Pubmed
Lüthy, Assessment of protein models with three-dimensional profiles. 1992, Pubmed
Malsure, Colon-specific deletion of epithelial sodium channel causes sodium loss and aldosterone resistance. 2014, Pubmed
Mansley, Prostaglandin E2 stimulates the epithelial sodium channel (ENaC) in cultured mouse cortical collecting duct cells in an autocrine manner. 2020, Pubmed
Morris, AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. 2009, Pubmed
Myerburg, Airway surface liquid volume regulates ENaC by altering the serine protease-protease inhibitor balance: a mechanism for sodium hyperabsorption in cystic fibrosis. 2006, Pubmed
Myerburg, Acute regulation of the epithelial sodium channel in airway epithelia by proteases and trafficking. 2010, Pubmed
Nesterov, Trypsin can activate the epithelial sodium channel (ENaC) in microdissected mouse distal nephron. 2008, Pubmed
Noori, How SARS-CoV-2 might affect potassium balance via impairing epithelial sodium channels? 2021, Pubmed
Noreng, Structure of the human epithelial sodium channel by cryo-electron microscopy. 2018, Pubmed
Noreng, Molecular principles of assembly, activation, and inhibition in epithelial sodium channel. 2020, Pubmed
Notredame, T-Coffee: A novel method for fast and accurate multiple sequence alignment. 2000, Pubmed
Orce, Inhibition of short-circuit current in toad urinary bladder by inhibitors of glandular kallikrein. 1980, Pubmed
Paoloni-Giacobino, Cloning of the TMPRSS2 gene, which encodes a novel serine protease with transmembrane, LDLRA, and SRCR domains and maps to 21q22.3. 1997, Pubmed
Passero, Defining an inhibitory domain in the gamma subunit of the epithelial sodium channel. 2010, Pubmed , Xenbase
Passero, Plasmin activates epithelial Na+ channels by cleaving the gamma subunit. 2008, Pubmed , Xenbase
Passero, TMPRSS4-dependent activation of the epithelial sodium channel requires cleavage of the γ-subunit distal to the furin cleavage site. 2012, Pubmed , Xenbase
Patel, Tissue kallikrein activation of the epithelial Na channel. 2012, Pubmed , Xenbase
Pettersen, UCSF Chimera--a visualization system for exploratory research and analysis. 2004, Pubmed
Planès, ENaC-mediated alveolar fluid clearance and lung fluid balance depend on the channel-activating protease 1. 2010, Pubmed
Prulière-Escabasse, Effect of neutrophil elastase and its inhibitor EPI-hNE4 on transepithelial sodium transport across normal and cystic fibrosis human nasal epithelial cells. 2010, Pubmed
Reihill, Inhibition of Protease-Epithelial Sodium Channel Signaling Improves Mucociliary Function in Cystic Fibrosis Airways. 2016, Pubmed
Rossier, Epithelial sodium channel (ENaC) and the control of blood pressure. 2014, Pubmed
Rossier, Activation of the epithelial sodium channel (ENaC) by serine proteases. 2009, Pubmed
Rotin, Function and Regulation of the Epithelial Na+ Channel ENaC. 2021, Pubmed
Sali, Comparative protein modelling by satisfaction of spatial restraints. 1993, Pubmed
Sanjana, Improved vectors and genome-wide libraries for CRISPR screening. 2014, Pubmed
Sanner, Python: a programming language for software integration and development. 1999, Pubmed
Sanner, Reduced surface: an efficient way to compute molecular surfaces. 1996, Pubmed
Shalem, Genome-scale CRISPR-Cas9 knockout screening in human cells. 2014, Pubmed
Snyder, Gating induces a conformational change in the outer vestibule of ENaC. 2000, Pubmed , Xenbase
Svenningsen, Plasmin in nephrotic urine activates the epithelial sodium channel. 2009, Pubmed , Xenbase
Szabo, Membrane-anchored serine proteases in vertebrate cell and developmental biology. 2011, Pubmed
Szabo, Membrane-anchored serine proteases as regulators of epithelial function. 2020, Pubmed
Taly, Using the T-Coffee package to build multiple sequence alignments of protein, RNA, DNA sequences and 3D structures. 2011, Pubmed
Tan, Cleavage of endogenous γENaC and elevated abundance of αENaC are associated with increased Na⁺ transport in response to apical fluid volume expansion in human H441 airway epithelial cells. 2011, Pubmed
Tarran, Soluble mediators, not cilia, determine airway surface liquid volume in normal and cystic fibrosis superficial airway epithelia. 2006, Pubmed
Trott, AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. 2010, Pubmed
Vaarala, Expression of transmembrane serine protease TMPRSS2 in mouse and human tissues. 2001, Pubmed
Vallet, An epithelial serine protease activates the amiloride-sensitive sodium channel. 1997, Pubmed , Xenbase
Vallet, Cell-surface expression of the channel activating protease xCAP-1 is required for activation of ENaC in the Xenopus oocyte. 2002, Pubmed , Xenbase
Verghese, Prostasin regulates epithelial monolayer function: cell-specific Gpld1-mediated secretion and functional role for GPI anchor. 2006, Pubmed
Vuagniaux, Activation of the amiloride-sensitive epithelial sodium channel by the serine protease mCAP1 expressed in a mouse cortical collecting duct cell line. 2000, Pubmed , Xenbase
Vuagniaux, Synergistic activation of ENaC by three membrane-bound channel-activating serine proteases (mCAP1, mCAP2, and mCAP3) and serum- and glucocorticoid-regulated kinase (Sgk1) in Xenopus Oocytes. 2002, Pubmed , Xenbase
Webb, Protein structure modeling with MODELLER. 2014, Pubmed
Wichmann, Incorporation of the δ-subunit into the epithelial sodium channel (ENaC) generates protease-resistant ENaCs in Xenopus laevis. 2018, Pubmed , Xenbase
Wichmann, Evolution of epithelial sodium channels: current concepts and hypotheses. 2020, Pubmed
Wörner, Renal effects of the serine protease inhibitor aprotinin in healthy conscious mice. 2022, Pubmed
Zhao, Plasmin improves blood-gas barrier function in oedematous lungs by cleaving epithelial sodium channels. 2020, Pubmed , Xenbase