Tumor necrosis factor-alpha convertase (ADAM17) mediates regulated ectodomain shedding of the severe-acute respiratory syndrome-coronavirus (SARS-CoV) receptor, angiotensin-converting enzyme-2 (ACE2)

Abstract

Angiotensin-converting enzyme-2 (ACE2) is a critical regulator of heart function and a cellular receptor for the causative agent of severe-acute respiratory syndrome (SARS), SARS-CoV (coronavirus). ACE2 is a type I transmembrane protein, with an extracellular N-terminal domain containing the active site and a short intracellular C-terminal tail. A soluble form of ACE2, lacking its cytosolic and transmembrane domains, has been shown to block binding of the SARS-CoV spike protein to its receptor. In this study, we examined the ability of ACE2 to undergo proteolytic shedding and investigated the mechanisms responsible for this shedding event. We demonstrated that ACE2, heterologously expressed in HEK293 cells and endogenously expressed in Huh7 cells, undergoes metalloproteinase-mediated, phorbol ester-inducible ectodomain shedding. By using inhibitors with differing potency toward different members of the ADAM (a disintegrin and metalloproteinase) family of proteases, we identified ADAM17 as a candidate mediator of stimulated ACE2 shedding. Furthermore, ablation of ADAM17 expression using specific small interfering RNA duplexes reduced regulated ACE2 shedding, whereas overexpression of ADAM17 significantly increased shedding. Taken together, these data provided direct evidence for the involvement of ADAM17 in the regulated ectodomain shedding of ACE2. The identification of ADAM17 as the protease responsible for ACE2 shedding may provide new insight into the physiological roles of ACE2.

Fig. 1

Shedding of ACE2 involves loss of its cytoplasmic tail. HEK293 cells were stably transfected with an expression vector encoding full-length ACE2 as described under “Materials and Methods.” OptiMEM containing 0.1 μm PMA or an equal volume of Me₂SO carrier was added to exponentially growing cells and collected after 1 h. Following sedimentation of cells, the media were concentrated 10-fold, and 40 μg of media proteins (M) were separated by SDS-PAGE (6% v/v) alongside 20 μg of corresponding detergent cell extract (C) and immunoblotted with an antibody raised to the ectodomain of ACE2 (left panel, ectodomain) or the cytosolic tail of ACE2 (right panel, cytosolic). Immunoreactive bands were visualized with enhanced chemiluminescence as described under “Materials and Methods.”

Fig. 2

Shed ACE2 occurs as two distinct glycoforms. HEK-ACE2 cells were incubated in OptiMEM containing 0.1 μm PMA or an equal volume of Me₂SO carrier for 1 h, and the media were collected and concentrated as described. Media proteins (M; 40 μg) or detergent cell extracts (C; 20 μg) were incubated at 37 °C for 16 h in the presence or absence of endoglycosidase H (Endo H) or PNGase F and subsequently separated by SDS-PAGE. Following electrotransfer, immunoblotting was carried out using an antibody to the ectodomain of ACE2 as described under “Materials and Methods.”

Fig. 3

Shedding of ACE2 is inhibited by broad spectrum hydroxamate-based metalloprotease inhibitors. HEK-ACE2 cells were incubated in OptiMEM containing various concentrations of the secretase inhibitors TAPI-1 or GM6001 or an equal volume of carrier (Me₂SO). After 15 min, PMA (0.1 μm) or an equal volume of Me₂SO was added, and incubation was continued for 1 h. The medium was subsequently harvested and concentrated as described; cells were pelleted and detergent extracts were collected as described under “Materials and Methods.” A, media proteins (upper panel, 40 μg) and cell lysates (lower panel, 20 μg) were separated by SDS-PAGE and immunoblotted for ACE2. Immunoreactive bands were visualized by enhanced chemiluminescence. B, graphical representation of results of densitometric analysis of three such experiments, ± S.E. Black shading, –PMA; gray shading, +PMA.

Fig. 4

PMA-stimulated ACE2 shedding is sensitive to ADAM17 inhibition. HEK-ACE2 cells were incubated for 15 min in the presence of the ADAM10 inhibitor GI254023X or the mixed ADAM10/ADAM17 inhibitor GW280264X or an equal volume of Me₂SO. Subsequently, incubation was continued in the presence of PMA (0.1 μm). Media were harvested and concentrated as described under “Materials and Methods.” A, media proteins (40 μg) were separated by SDS-PAGE and immunoblotted for ACE2. B, graphical representation of results of densitometric analysis of three such experiments, ± S.E.

Fig. 5

Stimulated ACE2 shedding is inhibited by NTIMP3 but not by NTIMP1. HEK-ACE2 cells were incubated for 15 min in the presence of NTIMP1, NTIMP3, or an equal volume of phosphate-buffered saline. Subsequently, incubation was continued in the presence of PMA (0.1 μm). Media were harvested and concentrated as described. A, media proteins (40 μg) were separated by SDS-PAGE and immunoblotted for ACE2. B, graphical representation of results of densitometric analysis of the immunoblots of three such experiments, ± S.E.

Fig. 6

Ablation of ADAM17 expression by siRNA reduces stimulated ACE2 shedding. HEK-ACE2 cells were transiently transfected with a mixture of double-stranded RNA derived from the coding sequence of ADAM10, ADAM17, or the control sequence GL2 as described under “Materials and Methods.” Twenty-four hours after transfection, cells were incubated in OptiMEM containing 0.1 μm PMA for 1 h. Media were concentrated as described, and cell lysates were prepared. A, media proteins (40 μg) and cell lysates (50 μg) were separated by SDS-PAGE and immunoblotted for ACE2, ADAM10, and ADAM17, as indicated. Mock, mock transfection. B, graphical representation of densitometric analysis of immunoblots of media ACE2 from three such experiments, ± S.E.

Fig. 7

Overexpression of ADAM17 increases PMA-stimulated ACE2 shedding. HEK-ACE2 cells were transiently transfected with an expression vector encoding ADAM9, ADAM10, or ADAM17, as described under “Materials and Methods.” Thirty-six hours after transfection, cells were incubated in OptiMEM containing 0.1 μm PMA for 1 h. Media were concentrated as described, and detergent cell extracts were harvested. A, media proteins (40 μg) and detergent cell extracts (50 μg) were separated by SDS-PAGE and immunoblotted for ACE2 (upper panel, media), ADAM9, ADAM10, or ADAM17 (lower panels, lysates). Membranes were stripped and reprobed for β-actin as a loading control. B, graphical representation of densitometric analysis of immunoblots of media ACE2 from three such experiments, ± S.E.

Fig. 8

PMA-stimulated endogenous ACE2 shedding is sensitive to ADAM17 inhibition.A, OptiMEM containing 0.1 μm PMA or an equal volume of Me₂SO carrier was added to exponentially growing Huh7 cells and collected after 6 h. Following sedimentation of cells, the media were concentrated, and 100 μg of media proteins (M) was separated by SDS-PAGE (4–12% v/v) alongside 50 μg of corresponding detergent cell extract (C) and immunoblotted with an antibody raised to the ectodomain of ACE2 (left panel, ectodomain) or the cytosolic tail of ACE2 (right panel, cytosolic). Immunoreactive bands were visualized with enhanced chemiluminescence as described under “Materials and Methods.” B, Huh7 cells were incubated for 15 min in the presence of 50 μm GM6001 or 1 μm the ADAM10 inhibitor GI254023X or the mixed ADAM10/ADAM17 inhibitor GW280264X or an equal volume of Me₂SO. Subsequently, incubation was continued for 4 h in the presence or absence of PMA (0.1 μm). Media were harvested and concentrated as described under “Materials and Methods.” Concentrated media samples (20 μg) were assayed for their ability to cleave an ACE2-specific fluorogenic substrate, as described under “Materials and Methods.” Black shading, –PMA; gray shading, +PMA.

Fig. 9

Ablation of ADAM17 expression by siRNA reduces regulated endogenous ACE2 shedding. Huh7 cells (50% confluent) were transfected with double-stranded RNA oligomers (as described under “Materials and Methods”) to ADAM10, ADAM17, or a negative control sequence. After 40 h, the media were aspirated and replaced with OptiMEM containing 100 nm PMA or an equal volume of Me₂SO. After a further 5-h incubation, the media were concentrated, and cell lysates were prepared as described under “Materials and Methods.” Cell lysates (50 μg) were separated by SDS-PAGE and immunoblotted for ADAM10 (A) and ADAM17 (B). Concentrated media samples (20 μg) were assayed for their ability to cleave an ACE2-specific fluorogenic substrate as described under “Materials and Methods.”

Fig. 10

Overexpression of ADAM17 increases PMA-stimulated endogenous ACE2 shedding. HEK-ACE2 cells were transiently transfected with an expression vector encoding ADAM9, ADAM10, or ADAM17, as described under “Materials and Methods.” Thirty-six hours after transfection, cells were incubated in OptiMEM containing 0.1 μm PMA for 4 h. Media were concentrated as described, and detergent cell extracts were harvested. A, detergent cell extracts (50 μg) were separated by SDS-PAGE and immunoblotted for ADAM9, ADAM10, or ADAM17. Membranes were stripped and reprobed for β-actin as a loading control. B, concentrated media samples (20 μg) were assayed for their ability to cleave an ACE2-specific fluorogenic substrate as described under “Materials and Methods.”