Sorting nexin 10 regulates lysosomal ionic homeostasis via ClC-7 by controlling PI(3,5)P₂

Lysosomes are degradative compartments that effectively break down macromolecules taken up by endocytosis. They also serve to recycle components of effete organelles designated for autophagy (De Duve and Wattiaux, 1966; Perera and Zoncu, 2016; Pillay et al., 2002). In addition, lysosomes are also indispensable for immunity, digesting foreign particulate matter engulfed by phagocytosis, as well as supporting antigen processing and presentation (Delamarre et al., 2005; Honey and Rudensky, 2003; Styrt and Klempner, 1982). To these ends, the lysosomal lumen is replete with a vast array of hydrolytic enzymes. Their defective expression or activity leads to accumulation of incompletely degraded cargo, often causing lysosomal distention and disturbing cellular homeostasis (Chadwick et al., 2021b; Farfel-Becker et al., 2019; Platt et al., 2018; Xu and Ren, 2015). More recently, lysosomes have also been appreciated for participating in nutrient sensing and contributing to intracellular ionic homeostasis (Chadwick et al., 2021a; Mony et al., 2016; Rebsamen et al., 2015; Shin and Zoncu, 2020).

Critical to the maintenance of the degradative milieu inside lysosomes is the activity of the vacuolar-type ATPase (V-ATPase), which utilizes ATP to drive the influx of protons (H⁺) into the organellar lumen (Ohkuma et al., 1982). The resulting luminal acidification (pH 4.5–5.0) not only provides an optimal environment for the activity of many of the hydrolases (Pillay et al., 2002), but the consequent establishment of a large H⁺ gradient across the lysosomal membrane drives the coupled transport of a variety of organic and inorganic solutes. Co-transport with H⁺ facilitates the export of amino acids (Löbel et al., 2022; Metzner et al., 2006), anionic sugars (Hu et al., 2023), and possibly dipeptides (Song et al., 2018), while counter-transport drives the accumulation of inorganic ions (Melchionda et al., 2016; Wartosch et al., 2009). The exchange of cytosolic chloride anions (Cl⁻) for intralysosomal H⁺ best illustrates the latter modality. ClC-7, an integral protein of the lysosomal-limiting membrane, has been convincingly established to catalyze the uptake of two Cl⁻ in exchange for a single H⁺, a notably rheogenic process (Leisle et al., 2011; Wartosch et al., 2009; Wu et al., 2023; Zhang et al., 2020). By virtue of the large transmembrane H⁺ gradient, ClC-7 drives the accumulation of Cl⁻ in the lysosomal lumen, where the concentration of the anion markedly exceeds the cytosolic concentration. Of note, the elevated luminal [Cl⁻] was previously shown to be essential to establish an optimal degradative milieu inside lysosomes (Wartosch et al., 2009; Zhang et al., 2023) and phagolysosomes (Wu et al., 2023). The inherent chloride dependence of several degradative enzymes is thought to account, at least in part, for these observations (Aghajari et al., 2002; Cigic and Pain, 1999).

Loss-of-function mutations in ClC-7 have been characterized and linked to several diseases (e.g., neurodegeneration and osteopetrosis) (Kasper et al., 2005; Kornak et al., 2001; Majumdar et al., 2011). The neurodegeneration and accompanying susceptibility to blindness have been attributed to the impaired ability of phagocytic cells, including microglia and macrophages, to clear extracellular debris, while the osteopetrosis is due to defective bone resorption by osteoclasts, which are developmentally and functionally related to phagocytes. Interestingly, osteopetrosis is also caused by mutation (Pangrazio et al., 2013; Stattin et al., 2017) or deletion (Ye et al., 2015; Zhou et al., 2016) of Snx10, a small phosphoinositide-binding protein implicated in membrane traffic (Barnea-Zohar et al., 2021). Mutations in Snx10 are also associated with presentations of blindness and neurological symptoms (Pangrazio et al., 2013; Stattin et al., 2017). Moreover, a CRISPR screen revealed that, similar to ClC-7 and its chaperone, OSTM1, Snx10 confers sensitivity to apilimod, a selective inhibitor of PIKfyve, the enzyme responsible for phosphatidylinositol (PI) 3,5-bisphosphate (PI[3,5]P₂) biosynthesis (Gayle et al., 2017). These similarities prompted us to study if, like ClC-7, Snx10 plays a role in (phago)lysosomal function and whether its effects are attributable to changes in Cl⁻ transport and degradative capacity.

While mutations in Snx10 were reported to lead to the development of disease (Aker et al., 2012; Pangrazio et al., 2013; Stattin et al., 2017), the underlying molecular mechanisms remain obscure. To better understand its mode of action, we initially edited Snx10 to generate Snx10 knockouts (KO) in RAW 264.7 macrophages, a phagocytic cell line. The effective deletion of Snx10 was validated by immunoblotting: as illustrated in Fig. 1 A, the protein was undetectable in two separate KO lines, which were selected for further study. To compare the degradative capacity of these cells, WT and Snx10 KO macrophages were challenged with IgG-opsonized sheep RBCs (sRBCs), which were taken up and contained within phagosomes. The extent of target degradation after 1 and 4 h was assessed by monitoring the sRBC volume, as described previously (Wu et al., 2023). In stark contrast to WT macrophages, which degraded the internalized sRBC almost entirely over 4 h, Snx10 KO macrophages were unable to break down the targets over the same time period (Fig. 1, B and C).

The failure of Snx10-deficient cells to degrade internalized sRBC could be attributable to a loss of hydrolytic activity within phagolysosomes, as reported for ClC-7 KO macrophages (Wu et al., 2023). To test this possibility, WT and Snx10 KO macrophages were challenged with IgG-opsonized silica beads that were coated with dye-quenched BSA (DQ-BSA), an indicator of proteolytic activity. Cleavage of DQ-BSA by proteases delivered to the phagosome through fusion with lysosomes results in an increase in the fluorescence signal. As illustrated in Fig. 1 D and quantified in Fig. 1 E, the proteolytic effectiveness of phagolysosomes was reduced by ≈50% in cells lacking Snx10, which resembles the phenotype of ClC-7 KO cells (Wu et al., 2023).

Most (phago)lysosomal proteases, including those of the cathepsin family, are synthesized as catalytically inactive precursors (zymogens) that only become active following an (auto)proteolytic cleavage transition (Muno et al., 1993). To determine whether Snx10 is important for the proteolytic maturation of cathepsins, lysates of WT and Snx10 KO macrophages were probed with anti-cathepsin C antibody. Snx10 KO cells displayed elevated levels of unprocessed, immature cathepsin C (∼55 kDa) and reduced levels of the mature cathepsin C band (∼25 kDa), compared with the parental WT cells (Fig. 1, F and G). Thus, defective conversion of the zymogens to their active counterparts contributes to the reduced proteolytic activity of Snx10-deficient phagolysosomes.

The defective resolution of the contents of phagosomes in Snx10-deficient macrophages may be caused by improper (incomplete) maturation (i.e., failure to merge with lysosomes) or by their inability to fully acidify their lumen; the latter mechanism is compatible with the reported association of Snx10 with the D subunit of the V-ATPase (Chen et al., 2012), which could conceivably be required for optimal H⁺ pumping. These possibilities were examined in turn. The rate and extent of phagosome maturation were assessed by visualizing the acquisition of LAMP1, a widely used marker of late (endo)lysosomes. As anticipated, nascent phagosomes (formed by WT cells within 3 min of adding the targets) were largely devoid of LAMP1; after maturing for 30 min, the phagosomes became richly endowed with this marker (Fig. 2 A). Importantly, Snx10 KO macrophages behaved similarly (Fig. 2 A), implying that delayed or incomplete maturation to phagolysosomes was not responsible for their failure to degrade the engulfed targets. This conclusion was further supported by ratiometric determinations of the phagosomal pH. Consistent with earlier findings, the luminal pH of phagosomes formed following ingestion of sRBC labeled with FITC—a pH-sensitive fluorophore—was highly acidic (pH ≤ 5.0) and was indistinguishable between WT and Snx10 KO macrophages (Fig. 2 B). A profound acidification was also observed when phagosomes contained FITC-labeled opsonized zymosan particles and was identical in WT and Snx10 KO cells (Fig. 2 C). In both instances the luminal acidification was dissipated by addition of concanamycin A (CCA), a specific inhibitor of the V-ATPase. Therefore, we concluded that its reported interaction with Snx10 is not essential for proton pumping by the (phago)lysosomal V-ATPase.

Phagosomes and lysosomes contain high luminal [Cl⁻], which was recently shown to directly support the activity of resident hydrolases, fostering the degradation of phagosomal targets (Wu et al., 2023; Zhang et al., 2023). Importantly, the accumulation of Cl⁻ in these organelles is driven by ClC-7 or its homologue, and deletion of this antiporter impaired phagosomal resolution without affecting phagosome maturation or acidification (Wu et al., 2023), closely resembling the phenotype of Snx10-deficient cells. We therefore considered the possibility that Snx10 deletion might impair Cl⁻ accumulation in (phago)lysosomes. Two separate approaches were used to estimate the Cl⁻ concentration in lysosome-related organelles in WT and Snx10 KO macrophages. The first method involves the colorimetric determination of chloride in cell extracts using mercury thiocyanate and ferric nitrate. Two strategies were applied to maximize the contribution of lysosomes in collected cell extracts to the Cl⁻ determinations: first, the lysosomal volume was greatly increased by overnight incubation with sucrose (Fig. 3 A), which accumulates in the lysosomal lumen as it cannot be hydrolyzed or transported across the limiting membrane (Bright et al., 2016; Cai et al., 2024). As a result, the enlarged lysosomes (sucrosomes) occupy a sizeable fraction of the total cellular volume, which becomes readily apparent when trapping fluorescent dextrans within their lumen (Fig. 3 B). Note that, as is the case for intact lysosomes, ClC-7 is distinctly present in the sucrosomal-limiting membrane (Fig. S1 A). Moreover, sucrosomes display robust proteolytic activity that is markedly depressed when ClC-7 or Snx10 are deleted (Fig. S1 B), indicating that they are suitable mimics of lysosomes. To further increase the contribution of lysosomes/sucrosomes to the Cl⁻ determinations, the cytosolic Cl⁻ was leached from the cells by permeabilizing the plasma membrane briefly (30 s) while bathing the cells in Cl⁻-depleted (gluconate-substituted) medium; this was accomplished by activating plasmalemmal P₂X₇ receptors with ATP (Fig. 3 A). The effectiveness of the permeabilization step was verified using FM1–43, a small (m.w. 611) solvochromic dye that stains only the outer leaflet of the plasma membrane of intact cells, but readily enters ATP-permeabilized cells, staining endomembranes (Fig. 3 C). The integrity of the enlarged lysosomal membrane is preserved under the gentle permeabilization conditions used, as indicated by the persistence of the transmembrane H⁺ gradient, detected by the accumulation of cresyl violet (Ostrowski et al., 2016). Using this approach and considering the total sucrosomal volume determined by integrating the fluorescence of the trapped dextran, the sucrosomal [Cl⁻] was determined to be 129.7 + 2.9 mM in WT cells (Fig. 3 D), which agrees with previous measurements of (phago)lysosomal [Cl⁻] (Riazanski et al., 2021; Tan et al., 2013; Wu et al., 2023). Sucrosomal accumulation of Cl⁻ occurs in exchange for luminal H⁺, a process mediated by ClC-7. This is indicated by the reduced [Cl⁻] detected following treatment with CCA, as well as in sucrosomes formed in ClC-7 KO cells. Strikingly, sucrosomal Cl⁻ was also greatly reduced in Snx10 KO cells (Fig. 3 D).

To validate the results obtained in sucrosomes, we also estimated [Cl⁻] in normal lysosomes loaded with 10,10′-bis[3-carboxypropyl]-9,9′-biacridinium dinitrate (BAC)—a fluorophore that is quenched by high [Cl⁻] (Sonawane et al., 2002)—which had been covalently conjugated to 10 kDa amino dextran to ensure its retention within lysosomes. To account for the varying size and shape of lysosomes, the fluorescence of BAC was normalized to that of AlexaFluor-647–labeled dextran and, because precise calibration of BAC inside lysosomes is difficult and unreliable, the results are presented relative to untreated, WT lysosomes. In agreement with the sucrosome data, we found the relative lysosomal [Cl⁻] to be markedly reduced following the inhibition of the V-ATPase and also in the absence of ClC-7, compared with the untreated WT (Fig. 3 E). More importantly, lysosomal [Cl⁻] was also depressed in the two Snx10 KO clones, confirming that Snx10 is required for the accumulation of lysosomal Cl⁻, presumably by enabling the activity of ClC-7.

The preceding results suggest that Snx10 controls the uptake of Cl⁻ into lysosomes by modulating ClC-7 activity. Because sorting nexins have been implicated in membrane traffic, we considered the possibility that Snx10 may be required to deliver ClC-7 to lysosomes and phagolysosomes. To this end, WT, ClC-7 KO, and Snx10 KO macrophages were challenged with IgG-opsonized targets (silica beads) and, after 1 h, were fixed, permeabilized and immunostained with anti–ClC-7 antibody. As expected, ClC-7 was clearly detectable in the limiting membrane of WT phagosomes but not in those formed by ClC-7 KO macrophages (Fig. 4, A and B), attesting to the specificity of the antibodies used. Phagosomes formed by Snx10 KO macrophages displayed ClC-7 levels comparable with those of WT phagosomes (Fig. 4, A and B), indicating that the sorting nexin is not required for targeting the antiporter to (phago)lysosomes. To determine whether endolysosomal morphodynamics were more broadly altered upon loss of Snx10, WT, ClC-7 KO, and Snx10 KO macrophage lysosomes were visualized by pulse labeling with 10 kDa tetramethylrhodamine dextran overnight. Loss of Snx10 had no discernible effect on the tubulovesicular morphology of lysosomes in RAW 264.7 cells (Fig. S1 C). Nevertheless, endolysosomal maturation dynamics could conceivably be affected by the loss of Snx10. To assess this, we first labeled the lysosomes of WT and Snx10 KO macrophages with an overnight pulse chase of 10 kDa tetramethylrhodamine dextran, followed by an acute incubation with 10 kDa AlexaFluor-647 dextran for 10–30 min to observe bulk fluid uptake and transit along the endocytic pathway. Loss of Snx10 had no noticeable effect on the transit kinetics of fluid phase markers from early to late endocytic compartments (Fig. S1 D).

Because the targeting of ClC-7 to (phago)lysosomes appeared to be unaffected by Snx10, we instead considered that the transport activity of the antiporter may be regulated by the sorting nexin, whether directly or indirectly. It was previously reported that ClC-7 and its Arabidopsis homologue ClC-a are strongly inhibited by PI(3,5)P₂, a rare signaling phosphoinositide (Carpaneto et al., 2017; Leray et al., 2022). Importantly, PI(3,5)P₂ has been detected in late endocytic compartments (Takatori et al., 2016), where ClC-7 is known to reside. Given that Snx10 has been demonstrated to bind the precursor of PI(3,5)P₂, PI 3-phosphate (PI[3]P) (Barnea-Zohar et al., 2021; Yao et al., 2009) (Fig. S1 E), which it requires for both its vesicular localization (Fig. S1 F) and vacuolating effect upon overexpression (Fig. S1 G), we inquired whether the expression of the sorting nexin influences the levels of these phosphoinositides. To distinguish PI(3,5)P₂ from its more abundant bisphosphorylated regioisomers (e.g., PI[4,5]P₂ and PI[3,4]P₂), we employed a combination of chiral column chromatography and triple-stage quadrupole tandem mass spectrometry (see Morioka et al. [2022] for details). This analysis revealed a greater than twofold increase in the PI(3,5)P₂ content of Snx10 KO cells, compared with their WT counterparts (Fig. 5 A). The increase was evident in multiple replicates of the two Snx10 KO clones. A more modest, yet significant increase in PI(3)P content was also detected in the Snx10 KO cells (Fig. 5 B). By contrast, all other phosphoinositides, including PI(4,5)P₂ and PI(3,4,5)P₃, as well as other common phospholipid species analyzed (phosphatidylcholine [PC], phosphatidylethanolamine [PE], phosphatidylserine [PS], phosphatidylglycerol [PG], and phosphatidic acid [PA]), were all found at similar levels in WT and Snx10 KO macrophages (Fig. S2, A–I), confirming that equivalent quantities of cells were analyzed in all instances.

The above findings suggest that Snx10 influences the activity of ClC-7 by controlling the availability of PI(3,5)P₂. This model predicts that pharmacologically ablating PI(3,5)P₂ should preclude the effect of Snx10 KO on the activity of the antiporter. This prediction was tested experimentally by comparing the sucrosomal [Cl⁻] of WT and Snx10 KO macrophages that had been incubated in the presence of YM201636, a potent PIKfyve inhibitor (Jefferies et al., 2008). Contrary to untreated sucrosomes, where the [Cl⁻] of WT cells is markedly higher than that of Snx10 KO cells (see Fig. 3 D and also Fig. 5 C), sucrosomes of cells that were treated with YM201636 for 30 min showed no significant differences; [Cl⁻] was elevated regardless of the deletion of Snx10 (Fig. 5 C).

Based on the finding that PI(3,5)P₂ depresses the activity of ClC-7 (Leray et al., 2022) and ClC-a (Carpaneto et al., 2017), we predicted that inhibition of PIKfyve would unleash Cl⁻ transport across the lysosomal membrane via the antiporter. Accordingly, and in good agreement with previous observations (Gayle et al., 2017; Ikonomov et al., 2001; Jefferies et al., 2008; Kim et al., 2014), pharmacological ablation of PI(3,5)P₂ caused the overactivation of ClC-7 that manifested as a profound swelling of the late endolysosomal (ClC-7-positive; Fig. S3 A) compartments, which was readily visualized by preloading with fluorescent fluid-phase markers (Fig. 5, D and E; Fig. S1 C; and Fig. S5 H). In sharp contrast, endolysosomal swelling resulting from PIKfyve inhibition failed to occur in ClC-7–deficient cells, supporting the notion that the antiporter mediates the volume gain induced by inhibition of PIKfyve (Fig. 5, D and E; and Fig. S1 C). Importantly, swelling persisted in the absence of Snx10 (Fig. 5, D and E; and Fig. S1 C), implying that the sorting nexin exerts its effect on ClC-7 by modifying PI(3,5)P₂.

The enlargement of the (endo)lysosomes induced by pharmacological inhibition of PIKfyve enabled us to measure their Cl⁻ content colorimetrically, as described for Fig. 3, but without having to resort to sucrose preloading (Fig. S3 B). As documented in Fig. S3 C, the swelling caused by ablation of PI(3,5)P₂ was accompanied by lysosomal accumulation of Cl^-, which occurred independently of Snx10.

Lysosomal swelling was also recently reported to occur in patient fibroblasts expressing hyperactive variants of ClC-7 (ClC-7^Y715C or ClC-7^K285T) that cause a novel type of lysosomal storage disorder (Nicoli et al., 2019; Polovitskaya et al., 2024). Interestingly, the excessive activity conferred by these mutations in ClC-7 is associated with—and seemingly attributable to—desensitization of the exchanger to PI(3,5)P₂ (Leray et al., 2022); as such, expression of these variants affords a convenient model to probe the mode of action of Snx10. As shown in Fig. 5 F, the swelling phenotype described in fibroblasts expressing ClC-7^Y715C could be readily replicated when this variant was heterologously expressed in RAW 264.7 macrophages. Revealingly, the lysosomal swelling induced by expression of ClC-7^Y715C persisted in cells lacking Snx10, consistent with the notion that the sorting nexin alters the availability of phosphoinositides, rather than targeting the H⁺/Cl⁻ antiporter directly. Treatment with YM201636, a PIKfyve inhibitor, did not further enlarge the swollen compartments of WT and Snx10 KO macrophages expressing ClC-7^Y715C, suggesting that these compartments were maximally swollen (Fig. S3 D).

The elevated PI(3,5)P₂ and depressed ClC-7 activity detected in Snx10-deficient cells suggest that Snx10 limits the generation and/or delivery of PI(3,5)P₂ to lysosomes. According to this model, overabundance of Snx10 is anticipated to have the opposite effect, namely activation of ClC-7 resulting from reduced PI(3,5)P₂ in lysosomes. This prediction was tested through the heterologous overexpression of Snx10. Transient transfection (overexpression) of Snx10-GFP led to profound endomembrane swelling in both RAW 264.7 macrophages and HeLa cells (Fig. 5 G). The swollen compartment was both Rab7- and ClC-7-positive (Fig. S3, E and F), indicative of its lysosomal identity. Moreover, the swelling caused by Snx10 overexpression required the presence of ClC-7 and was precluded by inhibition of the V-ATPase (Fig. 5 G). Jointly, these observations indicate that overexpression of Snx10 results in hyperactivation of ClC-7.

We interpreted the preceding results to mean that the overexpression of Snx10 stimulated ClC-7 activity by reducing the availability of PI(3,5)P₂ in the lysosomal compartment, where the antiporter resides. Because there are currently no reliable means to quantify subcellular PI(3,5)P₂—specifically in lysosomes—we instead tested our hypothesis by attempting to reverse the stimulation of ClC-7 induced by Snx10 overexpression by forcibly repleting PI(3,5)P₂ in lysosomes. This was accomplished using a newly developed approach by which an engineered and highly active catalytic fragment of PIKfyve (catalytic core; CC) is selectively recruited to an organelle of choice, where, if the requisite PI(3)P substrate is present, PI(3,5)P₂ is generated de novo (Pemberton et al., 2025). Targeted recruitment was accomplished using rapamycin-induced heterodimerization. As illustrated in Fig. 6 A, the kinase fusion protein (mRFP-FKBP-PIKfyve^CC) can be rapidly recruited to late (endo)lysosomes endowed with Rab7 using the iRFP-FRB-Rab7 targeting construct, causing accumulation of PI(3,5)P₂ in this compartment. The localized production of PI(3,5)P₂ was demonstrated in live cells using an optimized variant of a recently described PI(3,5)P₂-specific biosensor (Vines et al., 2023) termed NES-mNG-NES-(2x)DdSnxA^PX (Pemberton et al., 2025), which, while failing to reveal any steady-state enrichment of PI(3,5)P₂ in subcellular membranes, effectively detects the excess phosphoinositide generated upon recruitment of the engineered kinase construct to the PI(3)P-containing endosomal compartments.

The heterodimerization system enabled us to test whether ectopic generation of PI(3,5)P₂ on Rab7-positive compartments could revert the swelling phenotype induced by Snx10 overexpression. Strikingly, the large vacuoles formed by cells overexpressing Snx10 began shrinking shortly after PI(3,5)P₂ was synthesized by the recruited kinase, virtually disappearing within an hour (Fig. 6, B and D). Importantly, the vacuoles persisted throughout this period when an inactive form of the kinase (mRFP-FKBP-PIKfyve^CC K1877A) was recruited (Fig. 6, C and D), indicating that neither the transfection procedure nor the addition of rapamycin were themselves responsible for the vacuolar shrinkage, which must instead be attributed to the de novo formation of PI(3,5)P₂.

While these data corroborated the notion that Snx10 modulates ClC-7 activity by controlling the level of PI(3,5)P₂ in lysosomes, it remained unclear how Snx10 could function to impact the levels of the phosphoinositide. Interestingly, Snx10 has been reported to physically interact with PIKfyve (Bao et al., 2021), raising the possibility that it could directly inactivate the kinase. Alternatively, Snx10 might delay the delivery of PI(3)P to lysosomes or activate/recruit a phosphatase(s) that can reduce the levels of either PI(3)P or PI(3,5)P₂. These possibilities were analyzed by taking advantage of the heterodimerization-induced recruitment of the active FKBP-PIKfyve^CC fragment. As shown in Fig. 6 A, recruitment of the kinase to Rab7-containing membranes rapidly increased their PI(3,5)P₂ content, as revealed by the NES-mNG-NES-(2x)DdSnxA^PX biosensor. Importantly, the increase in PI(3,5)P₂ was more pronounced in the Snx10 KO cells than in their WT counterparts, despite comparable levels of recruitment of the active kinase (Fig. 6 E; and Fig. S4, A and B). This was not the case when the kinase was recruited to Rab5-containing vesicles (Fig. 6 F; and Fig. S4, C and D), implying that Snx10 selectively affects phosphoinositide metabolism in Rab7-positive, late (endo)lysosomal compartments. The greater accumulation of PI(3,5)P₂ in the Rab7-positive compartments of Snx10-deficient cells (Fig. 6 E; and Fig. S4, A and B) implies that the sorting nexin either promotes clearance of the phosphoinositide or delays the delivery of PI(3)P, the substrate of PIKfyve, from upstream compartments.

The latter explanation is consistent with the observations reported in Fig. 6, G and H (illustrated in Fig. S4, E and F), where the presence of PI(3)P was assessed using a probe (Kanai et al., 2001) based on the PX domain of p40_phox (p40_phox^PX). A small but significant fraction of the Rab7-containing vesicles were decorated by the p40_phox^PX probe, indicative of the presence of PI(3)P in late (endo)lysosomal compartments. Compared with WT macrophages, the fraction of p40_phox^PX- and Rab7-positive (endo)lysosomes was considerably greater in both Snx10 KO clones (Fig. 6 G and Fig. S4 E), while the fraction of p40_phox^PX- and Rab5-positive (endo)lysosomes was modestly decreased (Fig. S5, A and B). Conversely, fewer late (endo)lysosomes recruited p40_phox^PX in HeLa cells, where Snx10 had been heterologously overexpressed, although the difference did not reach statistical significance (Fig. 6 H and Fig. S4 F). The fraction of late (endo)lysosomes labeled with p40_phox^PX increased markedly when PIKfyve was inhibited by YM201636 and, under these circumstances, the effect of Snx10 became more noticeable and statistically significant (Fig. 6 H and Fig. S4 F). It is important to note that under these conditions, there was no observed overlap between Rab5 and Rab7 compartments, ruling out the possibility that PIKfyve inhibition increased PI(3)P on Rab7 compartments via the formation of hybrid compartments (Fig. S5 G). To determine whether this difference in PI(3)P localization along the endocytic pathway had an effect on the distribution of PI(3)P-binding proteins, we compared the localization of early endosome antigen 1 (EEA1), a known PI(3)P effector, in WT and Snx10 KO cells. We found that loss of SNX10 had no appreciable effect on the localization of EEA1 to early Rab5-positive vesicles but did led to a slight increase in the percentage of Rab7-positive compartments that associated with EEA1 (Fig. S5, C–F), which agrees with the observations made in Fig. 6 G and Fig. S4 E. Jointly, these data favor the notion that Snx10 restricts the delivery of PI(3)P to late (endo)lysosomes and hence its conversion to PI(3,5)P₂ by PIKfyve. Deletion of the sorting nexin enhances the availability and levels of the substrate (PI(3)P) (Fig. 5 B) and, as a consequence, also the enzymatic product of PIKfyve, PI(3,5)P₂, which ultimately restrains the activity of ClC-7.

Given the regulatory relationship between Snx10 and ClC-7, the defects in phagosome resolution observed in Snx10 KO macrophages may be ascribed to altered regulation of the antiporter by PI(3,5)P₂. If this were the case, uncoupling the activity of ClC-7 from its regulation by the phosphoinositide would be anticipated to overcome the effects of deleting Snx10. This prediction was tested by heterologously expressing the PI(3,5)P₂-insensitive variant ClC-7^Y715C in Snx10 KO macrophages. These cells were then challenged with sRBC, and their ability to degrade the phagocytic cargo was monitored. While as reported in Fig. 1, A and B, Snx10-deficient macrophages failed to break down their phagocytic cargo; expression of the phosphoinositide-insensitive form of ClC-7 restored their ability to degrade sRBC over 4 h, as illustrated in Fig. 7, A and B. Thus, Snx10 acts to control the activity of ClC-7 in cells by restricting the abundance of PI(3,5)P₂ on mature (phago)lysosomes.

We report that macrophages lacking Snx10 fail to degrade their engulfed cargo contained within phagolysosomes. In principle, this impaired degradation could result from an inability of Snx10-deficient phagosomes to acidify their lumen; accordingly, inhibition of the V-ATPase responsible for (phago)lysosomal acidification has been repeatedly demonstrated to inhibit cargo degradation (Lo and Zeng, 2023; Lukacs et al., 1990; Moriwaki et al., 2024; Wu et al., 2023; Zhang et al., 2022). This possibility is lent credence by the reported physical interaction between Snx10 and the V-ATPase (Chen et al., 2012). However, our measurements indicate that the pH of Snx10-deficient phagolysosomes is highly acidic, indistinguishable from that of WT cells (Fig. 2, B and C), thereby ruling out impaired acidification as the mechanism underlying defective phagolysosome resolution.

Instead, our results indicate that altered [Cl⁻] homeostasis is responsible for the inability of Snx10 KO macrophages to degrade their cargo (Fig. 3). Cl⁻ is required for the optimal activity of multiple lysosomal hydrolases (Aghajari et al., 2002; Cigic and Pain, 1999; Wu et al., 2023), and failure to accumulate this anion was earlier shown to account for the defective degradation of lysosomal and (phago)lysosomal content in ClC-7 KO cells (Wu et al., 2023; Zhang et al., 2023). Furthermore, we find that target degradation can be rescued in Snx10 KO macrophages by expression of a constitutively active variant of ClC-7. Jointly, these observations indicate that the effect of Snx10 is exerted through regulation of the activity of ClC-7.

Abnormal delivery of ClC-7 could in theory explain the inability of Snx10 KO phagolysosomes to accumulate Cl⁻. However, direct visualization and quantitation of (phago)lysosomal ClC-7 content (Fig. 4, A and B) compelled us to dismiss this possibility. Rather, our data indicate that Snx10 affects the regulation of ClC-7 through spatial changes to membrane phosphoinositide composition. Specifically, deletion of Snx10 causes accumulation of PI(3,5)P₂ in (phago)lysosomes, which in turn depresses the activity of the antiporter. Several conceivable mechanisms could account for the observed alteration in PI(3,5)P₂ levels. First, the sorting nexin, which has been shown to associate with PIKfyve (Bao et al., 2021), could directly inhibit its lipid kinase activity. Multiple lines of evidence argue against this mechanism. First, the catalytic domain of PIKfyve (mRFP-FKBP-PIKfyve^CC) increased PI(3,5)P₂ to a greater extent when recruited to the Rab7 compartment in Snx10 KO than in WT cells, yet phosphoinositide accumulation was comparable when recruited to the Rab5 compartment, where most of the Snx10 resides. Secondly, deletion of Snx10 was associated with increased total PI(3)P (Fig. 5 B) and with increased recruitment of the p40_phox^PX probe to Rab7 compartments (Fig. 6 G and Fig. S4 E), while the opposite would be expected had the activity of PIKfyve been increased. This contrasts with observations made when expressing a hyperactive mutant of PIKfyve, in which case a drop in PI(3)P accompanied the overproduction of PI(3,5)P₂ (McCartney et al., 2014). Together, these data suggest that direct modulation of PIKfyve kinase activity alone cannot fully explain the effects of Snx10.

Two additional modes of action can be contemplated. First, Snx10 may prevent accumulation of PI(3,5)P₂ in late (endo)lysosomal compartments by stimulating its degradation, conceivably by activating lipid phosphatase(s). The reported association of Snx10 with PIKfyve (Bao et al., 2021) may in fact occur via a binding partner, Fig4, which is a phosphoinositide 5-phosphatase, that is also an integral part of the conserved PIKfyve heteromeric complex (Lees et al., 2020). While regulation of PI(3,5)P₂ phosphatase activity in lysosomes cannot be ruled out, we regard this mechanism as unlikely because Snx10 is primarily present in earlier endosomal compartments and because overexpression of Snx10 reduced the lysosomal accumulation of PI(3)P when PIKfyve was inhibited (Fig. 6 H and Fig. S4 F); this observation cannot be readily explained by inhibition of a PI(3,5)P₂ phosphatase, inasmuch as this phosphoinositide was absent following treatment with YM201636.

The second alternative, which we favor, implicates the availability of PI(3)P in late (endo)lysosomal compartments, where it determines the rate of formation of PI(3,5)P₂. In this scenario, Snx10 restricts the delivery of PI(3)P to (Rab7-positive) late compartments. This could result from reduced availability of PI(3)P in early, upstream compartments, or from its impaired delivery to the lysosomal compartment. Indeed, given its ability to bind PI(3)P through its PX domain, Snx10 may directly scavenge a fraction of the phosphoinositide in the early endosomes. However, the vacuolating effect of Snx10 overexpression required not only its PX domain but also its C-terminal region (Yao et al., 2009), ruling out mere PI(3)P scavenging as the responsible mechanism. Direct scavenging of PI(3)P would also be expected to limit the ability of mRFP-FKBP-PIKfyve^CC to generate PI(3,5)P₂ when recruited to the early Rab5 compartment; this, however, was not the case (Fig. 6 F; and Fig. S4, C and D). Moreover, scavenging fails to explain the observed net accumulation of PI(3)P detected by mass spectrometry in Snx10 KO cells (Fig. 5 B). Alternatively, Snx10 may negatively regulate the generation of PI(3)P in early compartments by modulating the activity and/or the traffic of the class III PI3-kinase Vps34, which is the major source of vesicular PI(3)P in our cells (Fig. S1 H). However, we find both possibilities to be unlikely. An inhibitory role for Snx10 on Vps34 kinase activity conflicts with the observation that loss of Snx10 led to a modest decrease in PI(3)P levels on early Rab5 compartments (Fig. S5, A and B), where the opposite effect would be expected. Additionally, recruitment of mRFP-FKBP-PIKfyve^CC to Rab5 compartments resulted in virtually complete solubilization of the PX-GFP probe in both WT and Snx10 KO cells (Fig. S5, I and J), which suggests that early Rab5-positive endosomes are the source of all vesicular PI(3)P detected in the Snx10 KO cells, and that Vps34 resides exclusively in that compartment. These observations together argue against Snx10 negatively regulating Vps34 activity and/or traffic. It is also noteworthy that the lysosomal swelling observed in Snx10-overexpressing cells (e.g., Fig. 5 G) was not replicated in cells where PI(3)P in the early endosomal compartments was acutely depleted by inhibition of Vps34 (see Fig. S5 H).

For these reasons, we feel that rather than altering the availability of PI(3)P in early compartments, Snx10 may control its delivery to late endosomes and lysosomes. Our data appear consistent with this mode of action: First, more PI(3)P, as detected by the p40_phox^PX probe, was found in Rab7-positive compartments when Snx10 was deleted (Fig. 6 G and Fig. S4 E). Snx10 overexpression also affected the amount of PI(3)P accumulated in Rab7-positive compartments in response to inhibition of PIKfyve (Fig. 6 H and Fig. S4 F). In addition, when inducing recruitment of FKBP-PIKfyve^CC, the deletion of Snx10 altered the amount of PI(3,5)P₂ formed in Rab7-, but not in Rab5-positive compartments (Fig. 6, E and F; and Fig. S4, A–D).

The exact mechanism whereby Snx10 controls PI(3)P delivery to late compartments is still unclear. Snx10 may be directly involved in the generation or maintenance of putative intermediate compartments that delay the delivery of early endosomal PI(3)P to late compartments, or could instead indirectly control delivery by recruiting PI(3)P phosphatases—such as myotubularins—to early and/or intermediate compartments and accelerate PI(3)P turnover. Indeed, loss of Snx10 leads to an overall increase in PI(3)P in our macrophages (Fig. 5 B). Future follow-up studies could differentiate these alternatives.

In summary, we found that Snx10 regulates the activity of ClC-7 indirectly by influencing the availability of PI(3,5)P₂ in the lysosomal compartment. Collectively, our data suggest that Snx10 controls the rate of delivery of PI(3)P to late compartments, where it is converted to PI(3,5)P₂ by PIKfyve. This interpretation is consistent with the proposed function of sorting nexins in the regulation of membrane traffic (Carlton et al., 2004; Elson et al., 2021; Worby and Dixon, 2002). The well-established roles of Snx10 in bone and neuronal homeostasis as well as development (Aker et al., 2012; Pangrazio et al., 2013; Stattin et al., 2017) may very well result from similar effects on phosphoinositide traffic and metabolism.

RAW 264.7 and HeLa cells were obtained from and authenticated by the American Type Culture Collection. ClC-7 KO RAW 264.7 macrophages were generated previously (Wu et al., 2023). Cell lines were grown in DMEM + 5% FBS at 37°C under 5% CO₂.

RAW 264.7 cells were seeded on 6-well tissue culture plates for 24 h and then transfected with a Guaranteed Predesigned CRISPR gRNA p01 plasmid (Sigma-Aldrich) against Snx10 (MMPD0000084302) containing Cas9 and GFP to indicate transfection. Cells were then incubated with propidium iodide, and live, transfected cells (GFP positive and propidium iodide negative) were sorted by FACS into 96-well tissue culture plates. After expansion, KO clones were verified by immunoblotting.

All plasmids utilized in this study are summarized in Table S1. The generation of ClC-7^Y715C-GFP from rat ClC-7-GFP was accomplished with site-directed mutagenesis using the In-Fusion HD EcoDry Cloning Kit (121416; Takara), and the following forward and reverse primers (5′ – 3′): 5′-AAA​GAC​TTT​CGC​GAT​GCC​TGCCC​ACG​CTT​C-3′, 5′-GAA​GCG​TGG​GCA​GGC​ATC​GCG​AAA​GTC​TTT-3′ (point mutation in bold). Snx10-mCherry was constructed from Snx10-GFP (MG51516-ACG; Sinobiological) by PCR amplifying the Snx10-GFP ORF with the following primers (5′ – 3′): 5′-GAT​CTC​GAG​CTC​AAG​CTT​CGA​TGT​TCC​CAG​AAC​AGC​AGA​AAG-3′, 5′-GGT​ACC​GTC​GAC​TGC​AGA​ATT​TTG​GAC​TCC​TGC​AGA​GCT​GGG​CTC-3′, and inserted into a pmCherry N1 vector (Clontech) linearized with EcoRI by Gibson assembly (New England Biolabs). Snx10^R51Q-GFP was generated from Snx10-GFP by site-directed mutagenesis using the Phusion High-Fidelity DNA Polymerase (F530S; Thermo Fisher Scientific), and the following forward and reverse primers (5′ – 3′): 5′-CAG​AGA​AGG​TAT​AGA​GAG​TTC​GTG-3′, 5′-TAC​ACA​AGA​TGT​TTT​CAT​TGT​AA-3′. The design, assembly, and validation of the mRFP-FKBP-PIKfyve^CC and mRFP-FKBP-PIKfyve^CC K1877A recruitable enzymes, as well as the NES-mNG-NES-(2x)DdSnxA^PX probe, are detailed in a complementary study (Pemberton et al., 2025), while plasmids iRFP-FRB-Rab5 (#51612; Addgene) and iRFP-FRB-Rab7 (#51613; Addgene) have been described previously (Hammond et al., 2014).

Commercially available reagents utilized in this study are summarized in Table S2. Antibodies utilized in this study are summarized in Table S3. All antibodies were resuspended in 1X TBS solution with 0.05% Tween 20 (TBST) supplemented with 5% BSA. All fluorescent secondary antibodies conjugated with AlexaFluor-488, -555, -647, or -HRP were obtained from Jackson ImmunoResearch.

IgG-opsonized sRBCs were prepared as described previously (Wu et al., 2023). Briefly, 30 μl of a 10% suspension of sRBCs were resuspended in 500 μl of PBS. Cells were washed thrice with PBS, then coated with IgG by incubating for 30 min with 1.5 μl of anti-sRBC IgG fraction, shaking at 300 rpm at 37°C. sRBCs were then washed thrice with PBS and utilized.

WT, ClC-7 KO, and Snx10 KO RAW 264.7 macrophages were grown on 6-well tissue culture plates for 18 h prior to harvesting. Cells were subsequently collected on ice and lysed in 100 μl radioimmunoprecipitation assay buffer containing protease and phosphatase inhibitors. Protein concentrations were measured using the BCA assay (Pierce), and samples were diluted in Laemmli buffer containing 10% 2-mercaptoethanol and boiled for 5 min at 95°C. Samples were then separated on 10% SDS-PAGE gels, transferred onto a polyvinylidene difluoride membrane, and blocked overnight in TBST and 5% BSA at 4°C. The blocked membranes were then incubated with primary antibodies in TBST containing 5% BSA for 1 h at room temperature; unless otherwise indicated all antibodies were used at a 1:1,000 dilution. The membrane was then washed thrice with TBST and subsequently incubated for 1 h at room temperature with secondary antibody conjugated to HRP (diluted 1:2,000 unless otherwise indicated), before being washed thrice with TBST. Blots were developed using the ECL Prime western blot detection reagent and visualized with a ChemiDoc MP Imaging System (BioRad) and Image Lab software 5.2.1 (BioRad).

WT, ClC-7 KO, and Snx10 KO RAW 264.7 macrophages were seeded on 18-mm coverslips in 12-well tissue culture plates 18 h prior to experimentation. After experimental treatment, cells were fixed at room temperature in 3% paraformaldehyde for 10 min, permeabilized in 0.1% Triton X-100 for 10 min, and blocked in 5% BSA for 30 min. Subsequently, the cells were incubated with the indicated primary antibodies in 5% BSA for 1 h, washed thrice with 1X PBS, and then incubated with appropriate fluorescent secondary antibodies in 5% BSA for 1 h. Cells were then washed thrice with PBS and imaged.

WT and Snx10 KO RAW 264.7 cell lysates were separated by SDS-PAGE and probed with anti-cathepsin C and anti-vinculin antibodies. The percentage of mature cathepsin C was determined by calculating the ratio of the mature cathepsin C (≈25 kDa) over the total cathepsin C (sum of mature cathepsin C and pro-cathepsin C [≈55 kDa] bands). Signals were normalized to the amount of vinculin, used as loading control. Densitometric measurements were quantified using ImageJ.

RAW 264.7 macrophages were seeded on 18-mm coverslips in 12-well tissue culture plates 18 h prior to experimentation. Cells were then incubated with IgG-opsonized DQ-BSA–coated silica beads for 30 min prior to imaging. To generate labeled beads, the silica beads were incubated with DQ-BSA red (Molecular Probes) shaking at 950 rpm at room temperature for 2 h. Beads were subsequently washed thrice with 1X PBS to remove excess unbound DQ-BSA, then opsonized with human IgG at 37°C, shaking at 950 rpm for 30 min. Beads were then washed thrice more with excess 1X PBS prior to addition to cells. The fluorescence of the DQ-BSA signal from beads that were not internalized was used to subtract the background (baseline DQ-BSA signal) from all experimental samples.

RAW 264.7 macrophages were seeded on 18-mm coverslips in 12-well tissue culture plates 18 h prior to experimentation. Subsequently, cells were challenged for 1 or 4 h with IgG-opsonized sRBC. In all cases, the cells were washed for 20 min after initial incubation with sRBC to ensure all unbound particles were removed. For visualization, cells were fixed, permeabilized, and stained for 30 min prior to imaging with fluorescent secondary antibodies against the opsonizing antibody and with fluorescent phalloidin to visualize the sRBC target and macrophage actin, respectively. Stained actin was used to identify cell outlines for quantification and segmentation purposes and is outlined in representative images.

RAW 264.7 macrophages were seeded on 18-mm coverslips in 12-well tissue culture plates at a concentration of ∼2 × 10⁵ cells 18 h prior to transfection. Cells were then transfected with the desired plasmid DNA using FuGENE HD (Promega) transfection reagent at a ratio of 3:1 (FuGENE to DNA), using 1 μg of total plasmid DNA per condition. Medium was replaced 4 h after transfection, and cells were imaged 8–10 h later.

HeLa cells were seeded on 18-mm coverslips in 12-well tissue culture plates at a concentration of ∼1.5 × 10⁵ cells 18 h prior to transfection. Cells were then transfected with the desired plasmid DNA using FuGENE 6 (Promega) reagent at a ratio of 3:1 (FuGENE to DNA), with 1 μg of total plasmid DNA per condition. Medium was replaced 5 h after transfection, and cells were imaged 12–18 h later.

RAW 264.7 macrophages seeded on 18-mm coverslips in 12-well tissue culture plates were incubated (“pulsed”) for 16 h with 100 μg/ml of fluorescently labeled 10 kDa dextran, then washed thrice (“chased”) with fresh medium 1 h prior to experimentation to allow for the delivery of the labeled dextran from earlier endocytic compartments to lysosomes.

WT and Snx10 KO RAW 264.7 macrophages were seeded on 18-mm coverslips in 12-well tissue culture plates 18 h prior to experimentation. Macrophages were incubated with FITC-conjugated IgG-opsonized sRBC or zymosan for 1 h, which were prepared by reacting sRBC or zymosan with FITC as described previously (Wu et al., 2023), then subsequently imaged to determine the steady-state phagosomal pH. FITC was excited at 490 nm, which emits at 520 nm in a pH-dependent manner, as well as at 440 nm, which emits at 520 nm in a manner that is much less sensitive to pH. Cells were then sequentially bathed in standard pH buffers (143 mM KCl, 5 mM glucose, 1 mM MgCl₂, 1 mM CaCl₂, and 20 mM Hepes) of pH 7.5, 6.5, 5.5, and 4.5 containing also 10 μM nigericin and 5 μM monensin for 10 min each. The 490 nm/440 nm fluorescence ratio was collected at each pH to construct a calibration curve, fitted by the least squares method, correlating the mean fluorescence ratio (background subtracted) against the known pH values. This was ultimately utilized to convert the recorded fluorescence ratio values to pH values. A standard pH curve was generated for each experimental condition.

RAW 264.7 macrophages were incubated overnight in culture medium containing 50 mM sucrose prior to experimentation. Cells were visually confirmed to have formed sucrosomes by the formation of large, distended vacuoles clearly observable by brightfield microscopy.

Colorimetric determination of Cl⁻ was adapted from a previous report (Zall et al., 1956). Sucrosome-containing WT, ClC-7 KO, and Snx10 KO cells were incubated in either chloride-containing buffer (130 mM NaCl, 3 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂, 10 mM glucose, and 20 mM Hepes, pH 7.4) or in nitrate containing buffer (130 mM NaNO₃, 3 mM KNO₃, 1 mM Mg[NO₃]₂, 1 mM Ca[NO₃]₂, 10 mM glucose, and 20 mM Hepes, pH 7.4) for 2 h to deplete cell chloride. Where indicated, the cells were treated with 500 nM CCA for 30 min. Cells were subsequently washed thrice in ice-cold gluconate buffer (140 mM Na-gluconate, 10 mM glucose, and 20 mM HEPES, pH 7.4), then treated with 3 mM ATP for 30 s, and washed rapidly thrice in ice-cold gluconate buffer prior to overnight lysis in 1% HNO₃. Lysates were then centrifuged at 10,000 × g for 10 min to remove particulate material, and the supernatant was mixed in equal parts with a mercuric solution (1 part 0.417% mercuric thiocyanate in methanol, 1 part 20.2% ferric nitrate solution, and 13 parts water) in a 96-well plate. The absorbance at 490 nm is indicative of the concentration of Cl⁻ in the sample and was measured using a microplate reader (Biotek, Epoch Microplate Spectrophotometer). [Cl⁻] standards were used to construct a standard curve. To determine sucrosomal [Cl⁻], cell number was determined in parallel samples using a Coulter counter (Beckman Coulter Life Sciences, Multisizer 4), and sucrosomal volume was determined by imaging parallel samples that had been pulsed overnight with AlexaFluor-647–conjugated 10 kDa dextran that was chased for 1 h with fresh medium to ensure delivery to sucrosomes. Sucrosomal volume was then calculated using Imaris (Oxford Instruments) software. To ensure their predominant contribution, the mean sucrosomal volume was verified to be >60% of the mean total cell volume of each replicate prior to [Cl⁻] determinations.

RAW 264.7 macrophages were co-incubated with 100 nM YM201636 for 30 min prior to experimentation to enlarge endolysosomes, which was confirmed by the formation of large vacuoles in all cells by brightfield microscopy. Cells were subsequently processed exactly as described above with the determinations of sucrosomal [Cl⁻].

The succinimidyl ester of the [Cl⁻]-sensitive fluorescent molecule BAC-succinimidyl ester was conjugated to 10 kDa amino-dextran by co-incubating both reagents in borate buffer (50 mM borate, pH 8.5) for 2 h at room temperature (25°C) while shaking at 600 rpm away from light. The resulting mixture was subsequently dialyzed for 24 h (Slide-A-Lyzer Dialysis Cassettes, 2K MWCO) against PBS at 4°C with gentle stirring, changing the PBS every 8 h, to generate 10 kDa BAC-dextran.

To estimate their lysosomal [Cl⁻], WT, ClC-7 KO, or Snx10 KO RAW 264.7 macrophages were pulsed overnight with 10 kDa BAC-dextran and chased for 1 h prior to experimentation; where indicated, WT, ClC-7 KO, and Snx10 KO cells were treated with CCA or bathed in nitrate medium devoid of chloride. To normalize for differential uptake of the indicator, cells were simultaneously pulsed with AlexaFluor-647–conjugated 10 kDa dextran, which is insensitive to [Cl⁻] changes. The fluorescence ratio of the 647-dextran to the BAC-dextran was quantified microscopically as described below.

Cell pellets were frozen and thawed once and then mixed with 50 μl of chloroform/methanol (1:9) solution containing 1 nmol of C8:0/C8:0 PI(4,5)P₂ (as an absorption inhibitor) and 10 pmol of heptadecanoyl/arachidonoyl (C17:0/C20:4) species of each phosphoinositide class (as an internal standard). 450 μl of formic acid was added and vortexed, followed by the addition of 4.5 ml of chloroform/methanol (1:1) and vortexing. A 50 μl aliquot was transferred to a separate tube for measurement of other glycerophospholipids. The rest was applied to a DEAE-Sepharose column that had been equilibrated with chloroform for the preparation of samples for phosphoinositide measurement. The column was washed with 3 ml of chloroform/methanol (1:1) solution, followed by 2 ml of chloroform/methanol/28% ammonium hydroxide/glacial acetic acid (200:100:3:0.9) solution. Phosphoinositides were eluted with 2 ml of chloroform/methanol/concentrated HCl/ultrapure water (12:12:1:1) solution. The eluate was collected in a new tube, and 1,130 μl of 120 mM NaCl was added and vigorously vortexed. After centrifugation (1,200 × g for 3 min), the lower phase was transferred to another tube. The derivatization of phosphoinositides by methylation was performed as described in our previous publication (Morioka et al., 2022). Phosphoinositide regioisomer measurement by chiral column chromatography and mass spectrometry was conducted using a triple quadrupole mass spectrometer QTRAP 6500 (SCIEX) combined with a Nexera XR HPLC/Autosampler system (Shimadzu). A sample in acetonitrile (10 μl) was injected, and phosphoinositides were separated using a CHIRALPAK IC-3 column (cellulose tris[3,5-dichlorophenylcarbamate], 2.1 × 250 mm, 3 μm; DAICEL) at 23°C. The LC was operated at a flow rate of 100 μl/min with a gradient as follows: 40% mobile phase A (methanol/5 mM ammonium acetate) and 60% mobile phase B (acetonitrile/5 mM ammonium acetate) were held for 1 min, linearly increased to 70% mobile phase A over 2 min and held at 70% mobile phase A for 7 min. Table S4 lists the MRM transitions (pairs of m/z values of precursor and product ions) used for the measurements of phosphoinositide molecular species. Analyst 1.6.3 (SCIEX) was used for data acquisition and processing. MultiQuant (SCIEX) was used for manual data evaluation for peak integration. No background subtraction was performed. Gaussian smoothing width was 1.0 points. The sample peak area value was divided by the corresponding internal standard peak area value for quantification.

For general phospholipid analysis, PA, PG, PC, PE, and PS were separated by reverse-phase column chromatography and subjected to mass spectrometry for measurement. To the 50 μl of cellular lipid extract solution that had been previously separated as described above, 10 pmol each of C15:0/C18:1-d7 PC, C15:0/C18:1-d7 PE, C15:0/C18:1 PA, C12:0/C13:0 PG, and C17:0/C14:1 PS were added as internal standards. The samples were mixed with 1.5 ml of chloroform/methanol/concentrated HCl/ultrapure water (12:12:1:1) solution and 850 μl of 120 mM NaCl and vigorously vortexed. After centrifugation (1,200 × g for 3 min), the lower phase was transferred to another tube and subjected to a methylation reaction as described previously, except that the resultant dried phospholipids were dissolved in 400 μl of methanol/ultrapure water (3:1) solution, and 10 μl was used for measurement. A Nexera X2 HPLC system (Shimadzu) equipped with an HTC PAL autosampler (CTC Analytics) was used for column chromatography. Separation of the phospholipid classes used a COSMOCORE 2.6C18 column (2.1 × 150 mm, 2.6 µm; Nacalai Tesque) at 60°C, with the mobile phase containing isopropanol/acetonitrile/1 M ammonium acetate (160:40:1) (A) and acetonitrile/ultrapure water/1 M ammonium acetate (160:40:1) (B). The gradient program was as follows: 0–1 min = 30% A, 1–3 min = 30–90% A, 3–7.5 min = 90% A,and 7.5–12 min = 30% A. A QTRAP5500 (SCIEX) was used for electrospray ionization MS/MS analysis. Measurement of the phospholipid species was achieved through multiple reaction monitoring using a preset list of mass-to-charge ratio values shown in Table S5.

Microscopy was performed on a spinning-disk system (Quorum Technologies, Inc.), comprised of an Axiovert 200 M microscope (Zeiss), cooled CCD camera (ORCA-Fusion BT, Hamamatsu), five-line laser module (Spectral Applied Research) with 405-, 443-, 491-, 561- and 655-nm lines, and a filter wheel (MAC5000; Ludl). The microscope was operated using Volocity v6.3 software (Perkin Elmer). Images were acquired with a 63×/1.4 NA oil objective (Zeiss). Ratiometric pH imaging was performed on a microscopy system that consists of an Axio Observer microscope (Zeiss), cooled CCD camera (Evolve 512; Photometrics), and a fluorescent light source (HXP 120V; Zeiss). Ratiometric imaging of FITC was performed using 470/40-nm and 436/20-nm excitation filters. The microscope was operated using ZEN 2 blue edition software (Zeiss), and images were acquired with a 63×/1.4 NA oil objective (Zeiss).

Image processing and analyses were performed using Volocity v6.3 software (Perkin Elmer). Fluorescence intensities of microscopy images were quantified using Volocity v6.3 and ImageJ. Volume of dextran-labeled sucrosomes was determined with Imaris (Oxford Instruments). Pseudocolor representations were created in ImageJ using the “Fire” look-up table function and were used to visualize the relative fluorescence intensity of DQ-BSA–coated silica beads contained within phagosomes and the inverted “Magenta hot” look-up table to visualize phagosomal resolution.

Data are presented as means ± SD. Statistical significance of data was determined using unpaired Student’s t tests with P < 0.05 considered significant. Statistical analyses were performed in GraphPad Prism 9.5.1 and Microsoft Excel. Each experiment was performed at least three times.

Fig. S1 characterizes the identity and morphology of endolysosomes from macrophages that were subjected to the various treatments and genetic ablations used in this study. Fig. S2 details the content of PC, PE, PS, PG, PA, PI, PI4P, PI(4,5)P₂, and PIP₃ in WT and Snx10 KO macrophage lysates, determined by phosphoinositide regioisomer measurement by chiral column chromatography and mass spectrometry. Fig. S3 characterizes the identity, morphology, and chloride content of maximally swollen endolysosomes in macrophages treated with PIKfyve inhibitors. Fig. S4 encompasses the representative microscopy images of the experiments used to quantify Fig. 6, E–H. Fig. S5 characterizes the localization of PI(3)P and of the PI(3)P-effector EEA1 relative to early and late endosomes in WT and Snx10 KO macrophages, as well as illustrating the depletion of PI(3)P in early endocytic compartments upon recruitment of active PIKfyve^CC. Table S1 shows the plasmids. Table S2 shows the reagents. Table S3 shows the antibodies. Table S4 shows the MRM transitions for phosphoinositides. Table S5 shows the MRM transitions for PC, PE, PA, PG, and PS.

Original data are available from the corresponding authors upon reasonable request.

The authors thank Dr. Thomas Jentsch (Leibniz-Forschungsinstitut für Molekulare Pharmakologie and Max-Delbrück-Centrum für Molekulare Medizin, Berlin, Germany) for the anti–ClC-7 antibody.

J.Z. Wu is funded by a Natural Sciences and Engineering Research Council of Canada (NSERC) Scholarship. S. Grinstein is supported by a Project Grant PJT-169180 from CIHR. S.A. Freeman is supported by a Discovery Grant RGPIN-2022-04485 from NSERC and holds a Canada Research Chair from CIHR.

Author contributions: J.Z. Wu: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, validation, visualization, and writing—original draft, review, and editing. J.G. Pemberton: conceptualization, methodology, resources, and writing—review and editing. S. Morioka: investigation. J. Sasaki: investigation. P. Bablani: investigation and resources. T. Sasaki: investigation, methodology, resources, supervision, and writing—original draft. T. Balla: funding acquisition, project administration, resources, and writing—review and editing. S. Grinstein: conceptualization, formal analysis, funding acquisition, project administration, supervision, and writing—original draft, review, and editing. S.A. Freeman: conceptualization, funding acquisition, methodology, project administration, resources, supervision, visualization, and writing—original draft, review, and editing.