GPN does not release lysosomal Ca2+ but evokes Ca2+ release from the ER by increasing the cytosolic pH independently of cathepsin C

Abstract

The dipeptide glycyl-l-phenylalanine 2-naphthylamide (GPN) is widely used to perturb lysosomes because its cleavage by the lysosomal enzyme cathepsin C is proposed to rupture lysosomal membranes. We show that GPN evokes a sustained increase in lysosomal pH (pH_ly), and transient increases in cytosolic pH (pH_cyt) and Ca²⁺ concentration ([Ca²⁺]_c). None of these effects require cathepsin C, nor are they accompanied by rupture of lysosomes, but they are mimicked by structurally unrelated weak bases. GPN-evoked increases in [Ca²⁺]_c require Ca²⁺ within the endoplasmic reticulum (ER), but they are not mediated by ER Ca²⁺ channels amplifying Ca²⁺ release from lysosomes. GPN increases [Ca²⁺]_c by increasing pH_cyt, which then directly stimulates Ca²⁺ release from the ER. We conclude that physiologically relevant increases in pH_cyt stimulate Ca²⁺ release from the ER in a manner that is independent of IP₃ and ryanodine receptors, and that GPN does not selectively target lysosomes.

Keywords: Ca2+ signals; Cathepsin C; Cytosolic pH; Endoplasmic reticulum; GPN; Lysosome.

Fig. 1.

GPN changes pH and [Ca²⁺]_c without rupturing lysosomes. (A) GPN is proposed to disrupt lysosomes because its cleavage by cathepsin C (blue arrow) causes osmotic lysis. (B) HEK cells loaded with LysoTracker Red, or with dextran conjugates of Oregon Green or fluorescein report an increase in pH_ly after addition of GPN (200 µM for 200 s). Increased pH causes fluorescence to decrease for LysoTracker Red and increase for the other indicators. (C) Time courses of GPN-evoked changes in fluorescence (F/F₀) of the pH_ly indicators. Each trace shows mean±s.d. from 3–4 ROIs in a single cell (summarised in Fig. 3F,H). (D) BAPTA (2.5 mM) was added to chelate extracellular Ca²⁺, and cyclopiazonic acid (CPA, 20 µM) to inhibit SERCAs, before the addition of GPN (200 µM) to fluo 8-loaded HEK cells. Results show mean±s.d. for 3 replicates. (E) Summary results (mean±s.e.m., n=3) from analyses similar to those in panel D show peak increase in [Ca²⁺]_c (Δ[Ca²⁺]_c) evoked by GPN. *P<0.05, Student's t-test. (F) Wild-type HEK cells (WT) or HEK cells without IP₃Rs (IP₃R-KO) kept in Ca²⁺-free HBS were either stimulated with 200 µM GPN alone or treated with 1 µM thapsigargin (TG) for 15 min followed by addition of 200 µM GPN. Results (mean±s.e.m., n=4, with 3 replicates) show Δ[Ca²⁺]_c. The green symbols are obscured by the overlying blue symbols. (G) Initial responses of HEK cells to carbachol (CCh, 1 mM), GPN (200 µM) or thapsigargin (TG, 1 µM) in Ca²⁺-free HBS (mean±s.d. of 3 replicates). A summary of these data is shown in Fig. S1C,D. (H) Effect of GPN (200 µM) on pH_cyt determined using the pH indicator SNARF-5F in populations of HEK cells. (I,J) Effects of bafilomycin A₁ (Baf A₁, 1 µM, 1 h) on basal pH_cyt (I) and on the ΔpH_cyt evoked by GPN (200 µM, 200 s) (J). Results (mean±s.e.m., n=5, each with 3 replicates) show no significant effect of BafA₁ (Student's t-test). (K) Representative confocal images show that GPN (200 µM, 10 min) had no effect on the punctate distribution of endocytosed Lucifer Yellow (M_r 447) or Alexa Fluor 488-coupled dextran (M_r ∼10,000). (L) Number of lysosomes identified before (0 min) and 5 min after GPN addition in at least 3 cells per coverslip. Mean±s.e.m., n=3 independent coverslips (Alexa Fluor 488-Dx, top graph) and n=4 independent coverslips (Lucifer Yellow, bottom graph).

Fig. 2.

CICR mediated by TMCO1 does not contribute to GPN-evoked increases in [Ca²⁺]_c. (A) TMCO1 has been proposed to oligomerise into a functional Ca²⁺-permeable channel as the ER loads with Ca²⁺ (Wang et al., 2016). Hence, TMCO1 might be able to mediate CICR because Ca²⁺ released by other intracellular channels fuels ER Ca²⁺ uptake by SERCAs, leading to Ca²⁺ overload and opening of TMCO1. (B) Typical western blot showing expression of TMCO1 in HEK cells in which CRISPR/Cas9 was used to achieve a partial knockdown of TMCO1 (67±14%, n=4, mean±s.d.). Molecular mass markers (kDa) and protein loadings (μg) are shown. (C,D) Peak Ca²⁺ signals evoked by a maximally effective concentration of carbachol (C) or the indicated concentrations of GPN (D) in Ca²⁺-free HBS (mean±s.e.m., n=3). *P<0.05, Student's t-test.

Fig. 3.

Effects of GPN on pH_ly do not require cathepsin C. (A) Western blot shows expression of cathepsin C and β-actin in wild-type (WT) HEK cells and after CRISPR/Cas9-mediated disruption of the cathepsin C genes (CTSC-KO). Protein loadings (µg) and positions of molecular mass markers (kDa) are shown. A representative for five similar blots is shown. (B) Cathepsin C activity, determined using a substrate (Gly-Arg-AMC) that fluoresces after proteolysis, was determined using whole-cell lysates from WT or CTSC-KO cells, alone or after treatment of cells with the cathepsin C inhibitor Gly-Phe-DMK (10 μM, 72 h, 37°C). Results show mean±s.d. for duplicate determinations. The green symbols are obscured by the overlying blue symbols. (C) Summary results (mean±s.e.m., n=3) from analyses similar to those in panel B show fluorescence recorded after 7 min. *P<0.05, one-way ANOVA, with Tukey's multiple comparison test (see panel B for colour key). (D) Lysosomes of WT or CTSC-KO HEK cells were loaded with pH indicators by endocytosis of dextran conjugates (Oregon Green and fluorescein) or by incubation with LysoTracker Red (100 nM, 20 min). Fluorescence was recorded before or 200 s after addition of GPN or d-GPN (300 µM each). Scale bars: 10 μm. (E) Time courses of the changes in LysoTracker Red fluorescence (F/F₀) after addition of GPN or d-GPN (solid bars) in WT and CTSC-KO cells. (F-H) Summary results (mean±s.e.m., n=5) from analyses similar to those in panel E show the fluorescence changes (F/F₀) before and 200 s after addition of the indicated GPN (300 µM). *P<0.05, Student's t-test relative to before GPN-treatment.

Fig. 4.

Effects of GPN on pH_cyt, pH_ly and [Ca²⁺]_c do not require cathepsin C. (A,B) Effects of the cathepsin C inhibitor Gly-Phe-DMK (10 µM, 72 h) and comparison of WT with CTSC-KO cells on pH_cyt (A) and [Ca²⁺]_c (B) of unstimulated HEK cells. Matched controls (Control or WT) are shown for each comparison (inhibitor or CTSC-KO). (C,D) Effects of the cathepsin C inhibitor (Gly-Phe-DMK, 10 µM, 72 h) on the changes in pH_cyt (ΔpH_cyt, measured 30 s after GPN addition) and the peak increase in [Ca²⁺]_c (Δ[Ca²⁺]_c) evoked by GPN. (E,F) Results of analyses similar to those shown in C,D, comparing the effects of GPN in WT and CTSC-KO cells. (G,H) Results of analyses similar to those shown in C,D, comparing the effects of GPN and d-GPN. Results (A-H) show means±s.e.m., n=3, each with 3 replicates.

Fig. 5.

Other weak bases evoke ER-dependent Ca²⁺ signals. (A) Structures and mechanisms of action of some lysosomotropic drugs. (B) Confocal images of HEK cells loaded with LysoTracker Red (100 nM, 20 min) and then treated (200 s) with NH₄Cl (20 mM) or fluoxetine (300 μM). (C) Quantification of fluorescence (F/F₀) from images similar to those in panel B. Results show mean±s.e.m., n=7. F₀ and F are fluorescence recorded before and 200 s after the addition of NH₄Cl or fluoxetine. (D) Effects of NH₄Cl (20 mM) or fluoxetine (300 μM) on pH_cyt recorded in populations of SNARF-5F-loaded HEK cells. Results show mean±s.d. of n=3 determinations. (E) Summary results (mean±s.e.m., n=3, each with 3 replicates) show peak increases in pH_cyt (ΔpH_cyt). (F-H) Effects of thapsigargin (TG, 1 μM, 15 min) in Ca²⁺-free HBS on the Ca²⁺ signals evoked by d-GPN (200 μM) (F), fluoxetine (300 μM) (G) or NH₄Cl (20 mM) (H). Results show mean±s.d. of n= 3 determinations. (I-K) Summary results from analyses similar to those in panels F-H show the effects of d-GPN (I), fluoxetine (J) or NH₄Cl (K) on Δ[Ca²⁺]_c alone or after treatment with thapsigargin (TG, 1 μM, 15 min). Results show mean±s.e.m., n=3, each with 3 replicates. (L) A requirement for ER Ca²⁺ for GPN to evoke an increase in [Ca²⁺]_c might reflect a need for the ER to fuel lysosomal Ca²⁺ uptake (option 1) or a direct action of GPN on the ER (option 2).

Fig. 6.

Buffering the GPN-evoked increase in pH_cyt abolishes Ca²⁺ release from the ER. (A) Effects of bafilomycin A₁ (Baf A₁, 1 μM, 1 h) on the Ca²⁺ signals evoked by GPN (200 μM) in HEK cells in Ca²⁺-free HBS. Mean±s.d. of 3 replicates. (B) Summary results (mean±s.e.m., n=4, each with 3 replicates) from analyses similar to those in panel A show peak increases in [ΔCa²⁺]_c evoked by GPN (Δ[Ca²⁺]_c). *P<0.05, Student's t-test. (C) Effects of pre-treating HEK cells with LLOMe (Leu-Leu-OMe; 1 mM, 1 h) on the Ca²⁺ signals evoked by GPN (200 μM) in Ca²⁺-free HBS. Mean±s.d. of 3 replicates. (D) Summary results (mean±s.e.m., n=4, each with 3 replicates) from analyses similar to those in panel C show peak increases in [Ca²⁺]_c evoked by GPN (Δ[Ca²⁺]_c). *P<0.05, Student's t-test. (E) HEK cells in Ca²⁺-free HBS were treated with sodium proprionate (NaP, 30 mM) before addition of GPN (200 μM). Results show pH_cyt (mean of 3 replicates). (F) Summary results (mean±s.e.m., n=3, each with 3 replicates) from analyses similar to those in panel E show pH_cyt measured 30 s or 2 min after GPN addition. *P<0.05, Student's t-test. (G) Typical images of HEK cells showing the effects of NaP (30 mM, 30 min) and then GPN (200 μM, 30 s) on LysoTracker Red staining. (H) Time-course (mean±s.d., n= 3 cells). (I) Analysis of the effects of NaP and GPN on [Ca²⁺]_c (mean of 3 replicates). (J-L) Summary results [mean±s.e.m., n=4 (J) or n=3 (K,L), each with 3 replicates] from analyses similar to those in panel E show peak Ca²⁺ signals (Δ[Ca²⁺]_c) evoked by GPN (200 μM), carbachol (CCh, 1 mM) or ionomycin (5 μM) in Ca²⁺-free HBS. *P<0.05, Student's t-test.

Fig. 7.

GPN evokes Ca²⁺ release from the ER through an increase in pH_cyt. GPN, a weak membrane-permeant base, causes pH_cyt to transiently increase which directly stimulates Ca²⁺ release from the ER through a mechanism that requires neither IP₃Rs nor RyRs. Some of the Ca²⁺ released by this pH-regulated mechanism, similarly to Ca²⁺ released by IP₃Rs, is then accumulated by lysosomes (LY). Many physiological stimuli that increase pH_cyt, including those that stimulate the Na⁺/H⁺ antiporter (NHE) through protein kinase C (PKC), evoke increases in [Ca²⁺]_c through the same pathway.