Application of a Sulfoxonium Ylide Electrophile to Generate Cathepsin X-Selective Activity-Based Probes

Abstract

Cathepsin X/Z/P is cysteine cathepsin with unique carboxypeptidase activity. Its expression is associated with cancer and neurodegenerative diseases, although its roles during normal physiology are still poorly understood. Advances in our understanding of its function have been hindered by a lack of available tools that can specifically measure the proteolytic activity of cathepsin X. We present a series of activity-based probes that incorporate a sulfoxonium ylide warhead, which exhibit improved specificity for cathepsin X compared to previously reported probes. We apply these probes to detect cathepsin X activity in cell and tissue lysates, in live cells and in vivo, and to localize active cathepsin X in mouse tissues by microscopy. Finally, we utilize an improved method to generate chloromethylketones, necessary intermediates for synthesis of acyloxymethylketones probes, by way of sulfoxonium ylide intermediates. In conclusion, the probes presented in this study will be valuable for investigating cathepsin X pathophysiology.

Graphical Abstract

INTRODUCTION

Cathepsin X (also referred to as cathepsin Z or P) is a cysteine cathepsin protease that is unique among its family members in that it exhibits strict carboxypeptidase activity. It is one of the most recently discovered cysteine cathepsins, and its functions during health and disease are still incompletely understood. Cathepsin X contributes to adhesion and maturation of macrophages and dendritic cells and suppresses clathrin-dependent phagocytosis through cleavage of profilin.^1,2 Cathepsin X regulates hormone signaling, where its cleavage of bradykinin, kallidin, or angiotensin leads to alterations in specificity toward their cognate receptors and divergent downstream signaling.³ Cathepsin X is also expressed by neurons, where its cleavage of α-enolase regulates survival and the outgrowth of neurites.⁴ Furthermore, cathepsin X expression is enriched in amyloid plaques, where it may have a protective effect against neurodegenerative disorders such as Alzheimer’s disease,^5,6 and in the spinal cord during neuropathic pain.⁷ Upregulation of cathepsin X mRNA has been reported in pathology-free regions of multiple sclerosis-affected brains,⁸ and it has been implicated in the generation of IL-1μ^9,10 and in mediating neuroinflammation.⁹ It is also upregulated in the microenvironment of breast,¹¹ pancreatic,¹² prostate,¹³ and gastric cancers,^14,15 where it likely promotes tumor invasion. Thus, cathepsin X holds promise as a clinical biomarker and therapeutic target in diverse diseases.

Like most cathepsins, cathepsin X is synthesized as a zymogen that becomes activated in the acidic environment of endolysosomes. Once activated, it may also be negatively regulated by endogenous inhibitors, though likely not cystatin C or stefin A.^16,17 In addition to its proteolytic functions, cathepsin X can also promote integrin-mediated signaling through an Arg-Gly-Asp (RGD) motif in its pro-domain.¹² As a result of these complex modes of post-translational regulation, traditional biochemical methods that survey total protein levels rarely reflect the pool of proteolytically active enzyme. The ability to specifically measure cathepsin X activity in its native environment is therefore required to define its precise proteolytic functions during health and disease.

To this end, efforts have been focused on developing fluorescent activity-based probes (ABPs) for cathepsin X. ABPs are small molecules that contain an electrophilic moiety (warhead), a recognition sequence that confers selectivity, and a fluorophore for detection.^18–20 When active, the protease initiates a nucleophilic attack on the warhead, resulting in the formation of a covalent, irreversible bond. Assessment of probe labeling can then be used to quantify protease activity by SDS-PAGE (in-gel fluorescence), fluorescent microscopy, flow cytometry, or optical imaging of whole tissues or organisms. Importantly, the covalent nature of probe binding allows for target confirmation by immunoprecipitation with specific antibodies or affinity purification followed by proteomic analysis.

Probes with absolute specificity for cathepsin X have not been previously reported. BMV109, a fluorescently quenched ABP with a tetrafluorophenoxymethyl ketone warhead, is a pan-cathepsin probe that targets X, B, S, and L.²¹ Because cathepsin X is a similar size as cathepsin B, one of the most abundant and ubiquitously expressed cathepsins, it can be difficult to clearly resolve these two proteases by SDS-PAGE, which precludes accurate quantification by in-gel fluorescence. MGP140 is an epoxide-based probe that exhibits greater specificity for cathepsin X than BMV109 but also potently reacts with cathepsin B.²² If mice are pretreated with GB11-NH₂, an inhibitor of cathepsin B, S, and L, prior to MGP140 injection, specific labeling of cathepsin X can be achieved. However, this manipulation of the system results in hyperactivation of cathepsin X, possibly a compensatory response due to the loss of cathepsin B activity. Thus, it is crucial to develop probes with improved specificity for cathepsin X to allow for a more detailed investigation of its physiological activity.

Herein, we describe a series of ABPs containing a novel sulfoxonium ylide warhead that exhibit previously unseen selectivity for cathepsin X. We applied these probes to measure cathepsin X activity in lysates and live cells and in live mice. We also used the sulfoxonium ylide as a stepping stone to access chloromethylketones, which are intermediates in the synthesis of acyloxymethylketones (AOMK), warheads commonly used in probes for cathepsins and other cysteine proteases. This new method does not require generation of diazomethanes to access chloromethylketones and is thus a safer alternative to the previously used methods. By comparison to the sulfoxonium ylide probes, AOMK probes bearing identical recognition sequences exhibited unique specificity profiles and low reactivity with cathepsin X. Thus, sulfoxonium ylide probes represent a clear advancement in the tools that are available to study cathepsin X function.

RESULTS AND DISCUSSION

Design and Characterization of a Sulfoxonium Ylide Probe.

To explore new potential warheads for cysteine cathepsins, we designed and synthesized an ABP containing a dimethyl sulfoxonium ylide electrophile. This design was initially inspired by a dimethyl sulfonium salt reported to inhibit cathepsin B in 1988 by Shaw.²³ To increase the electrophilicity of this warhead, and thus its reactivity with the catalytic cysteine residue, we modified the dimethyl sulfonium salt to a dimethyl sulfoxonium ylide. We also incorporated a valine residue as the P1 recognition sequence and a sulfo-Cyanine 5 (sCy5) fluorophore to yield our initial probe, sCy5-Val-SY (17; Figure 1), synthesized according to Scheme 1.

Figure 1.

Design of a sulfoxonium ylide-based activity-based probe. (A) Structure of sCy5-Val-SY (17) probe. (B) The proposed mechanism by which sCy5-Val-SY binds to an activated cysteine protease.

Scheme 1. Synthesis of sCy5-AA-SY Probes^a.

^a(i) 4-Nitrophenylchloroformate, Et₃N, DMAP, CH₂Cl₂, 0°C, 6 h. (ii) SOMe₃+ I −, KOtBu, THF, reflux, then cool to 0°C and add nitrophenyl ester. (iii) TFA/CH₂Cl₂ (1:1), (iv) sulfo-Cy5, PyClock, DIPEA, DMF, rt, 18 h. The synthetic routes for related compounds Cbz-Lys(sCy5)-SY (7, 15, 23) and sCy5-Phe-Val-SY (8, 16, 24) can be found in Scheme S1 and S2, respectively).

To determine its reactivity profile, we first incubated sCy5-Val-SY (17) with protein lysates prepared from RAW264.7 cells, an immortalized mouse macrophage line that contains high levels of active cysteine cathepsins.²¹ Cells were lysed in citrate buffer (pH 5.5) to provide optimal conditions for preserving cathepsin activity, and the probe was added at 1 μM for 20 min. We then resolved the lysates by SDS-PAGE and scanned the gel for sCy5 fluorescence using a flatbed laser scanner. We observed exclusive, concentration- and time dependent labeling of a ~35-kDa protease (Figures 2A and S1A,B). This labeling was prevented by pretreatment of the lysates with JPM-OEt, a pan-cysteine cathepsin inhibitor, confirming that this protease was a member of the cysteine cathepsin family (Figure 2A). In contrast, MDV-590, a specific inhibitor for cathepsin S,²⁴ did not compete for sCy5-Val-SY (17) binding. We compared the labeling profile to that of BMV109, the pan-cathepsin probe and found that the sCy5-Val-SY (17)-labeled protease was the same molecular weight as BMV109-labeled cathepsin X.²¹ We confirmed that this protease was indeed cathepsin X by immunoprecipitating sCy5-Val-SY (17)-labeled lysates with a cathepsin X-specific antibody (Figure 2B).

Figure 2.

In vitro characterization of sCy5-Val-SY (17). (A) Labeling of RAW264.7 lysates with sCy5-Val-SY or BMV109 alone or after pretreatment with 10 μM MDV-590 (cathepsin S inhibitor) or JPM-OEt (pan cysteine cathepsin inhibitor). (B) Immunoprecipitation of sCy5-Val-SY-labeled samples in A with a cathepsin X-specific antibody. (C) Labeling of splenic lysates from wildtype or cathepsin X-deficient mice with sCy5-Val-SY or BMV109. (D) Labeling of living RAW264.7 cells with increasing doses of sCy5-Val-SY or BMV109 for 2 h. (E) Immunoprecipitation of sCy5-Val-SY-labeled samples in D with a cathepsin X-specific antibody. (F) Labeling of living RAW264.7 cells with and without overnight pretreatment with 10 μM MDV-590 with sCy5-Val-SY or BMV109 (1 μM, 2 h). Also refer to Figure S1.

Next, we tested the ability of sCy5-Val-SY (17) to label cathepsin X in mouse splenic lysates. As we observed in macrophage lysates, the probe exhibited exclusive reactivity with cathepsin X in splenic lysates from wildtype mice, and this labeling was absent in lysates prepared from spleens of cathepsin X-deficient mice (Figure 2C). By comparison, BMV109 strongly labeled cathepsin B and, to a lesser extent, cathepsin S and L.

Having observed unique specificity of sCy5-Val-SY (17) in cell and tissue lysates, we sought to assess the probe’s permeability and specificity profile in living RAW264.7 cells. After incubating the probe with live cells for increasing lengths of time (at 1 μM) or with increasing probe concentrations (for 2 h), we analyzed lysates by in-gel fluorescence as above. Here, we observed time- and concentration-dependent labeling of two proteases (Figures 2D and S1C), which we identified as cathepsin X and S by immunoprecipitation (Figure 2E) and competition with MDV-590 (Figure 2F), respectively. We were surprised to see cathepsin S labeling in live cells, given its lack of binding to sCy5-Val-SY (17) in cell lysates, where we had confirmed high levels of cathepsin S activity with BMV109. This suggests that the reactivity of cathepsin S with the sulfoxonium ylide is dependent on the labeling conditions. We attempted to explore this by lysing the cells in various buffers that might mimic the endosomal environment of cathepsin S but we were not able to improve the labeling of cathepsin S in lysates (not shown).

Nonetheless, the sulfoxonium ylide probe exhibited clear labeling of cathepsin X in lysates and live cells with considerably improved selectivity compared to BMV109 (Figure 2D,F). To our knowledge, it is the first covalent ABP for cathepsin X that does not also bind to cathepsin B or L. As observed in Figure 2, it is difficult to distinguish cathepsin X labeling from cathepsin B with BMV109 due to the similarity in size of the two proteases. However, sCy5-Val-SY (17) allows for clear delineation of cathepsin X activity.

Sulfoxium Ylide Library with Variable P1 Residues.

To improve the specificity and potency of the probe for cathepsin X, we generated a small library of sulfoxonium ylide probes by varying the amino acids in the P1 position (Scheme 1, Table 1). In RAW264.7 lysates, probes bearing Ile (18), Leu (19), Nle (20), and Phe (21) all showed similar specificity for cathepsin X as sCy5-Val-SY (17) with sCy5-Leu-SY (19) and sCy5-Nle-SY (20) exhibiting a clear improvement in potency (Figure 3A). Cbz-Lys(sCy5)-SY (23), in which the sCy5 was attached via the lysine side chain, exhibited a loss of specificity, favoring cathepsin S over X and B. sCy5-Phe-Val-SY (24), in which a P2 Phe residue was incorporated, also exhibited a loss of specificity (Figures 3A and S2). The labeling profile of this probe was similar to BMV109, though it showed improved potencies for cathepsin X and S compared to BMV109. A hydrophobic S2 pocket is a feature of virtually all cysteine cathepsins, which may explain the increased affinity of a dipeptide probe for other members of the family.²⁵

Table 1.

Amino Acids Used in Compounds 1−24 (Scheme 1)

Boc-AA-OH	Boc-AA-ONp	Boc-AA-SY	sCy5-AA-SY
Boc-Val-OH	1	9	17
Boc-Ile-OH	2	10	18
Boc-Leu-OH	3	11	19
Boc-Nle−OH	4	12	20
Boc-Phe-OH	5	13	21
Boc-Trp-OH	6	14	22

Figure 3.

In vitro characterization of a sulfoxonium ylide library in lysates and live cells. Labeling of (A) RAW264.7 lysates (0.01, 0.05, 0.1, 0.5, 1, 5 μM), (B) kidney lysates (0.1, 0.5, 1, 5 μM), (C) live RAW264.7 cells (0.1, 0.5, 1, 5 μM), or (D) live MDA-MB-231HM cells (0.1, 0.5, 1, 5 μM) with the indicated SY probe or BMV109, as analyzed by in-gel fluorescence. Also refer to Figure S2.

In murine kidney lysates, Leu and Nle conferred the most potency and specificity for cathepsin X with Cbz-Lys (23), Phe (21), and Phe-Val (24) yielding broader reactivity and Val (17) and Ile (18) exhibiting weaker labeling (Figures 3B and S2).

To examine the potency and permeability of the sulfoxinium ylide probe series in living cells, we applied them to RAW264.7 cells for 2 h. Probes bearing Trp (22), Val (17), Ile (18), Leu (19), Nle (20), and Phe (21) labeled cathepsin X and S to similar extents and with similar potency, whereas Cbz-Lys (23) exhibited a preference for cathepsin S, and Phe-Val (24) labeled B and L in addition to X and S (Figures 3C and S2A,C–G). We confirmed the 25-kDa protease labeled by sCy5-Nle-SY (20) to be cathepsin S by competition with two cathepsin S-specific inhibitors, MDV-590 and Z-FL-COCHO (Figure S2B).

We tested the specificity of these probes for cathepsin X in a human breast cancer line known to express very low levels of cathepsin S, MDA-MB-231^HM.²⁶ These cells also allowed us to test whether the probes could bind to human cathepsin X (in addition to mouse cathepsin X shown previously). When we incubated the probes with MDA-MB-231 cells for shorter time periods, we observed very little labeling of cathepsin X (not shown); however, clear labeling was observed after overnight incubation (Figure 3D). This likely reflects differences in the rates of endocytosis between macrophages and tumor cells and suggests that the probes may be taken up directly into the endolysosmal pathway rather than by diffusion through membranes. The sulfoxonium ylide probe series generally shows specific labeling of cathepsin X in these cells with minimal cross-reactivity occurring only at 5 μM. Cbz-Lys(sCy5)-SY (23) and especially sCy5-Phe-Val-SY (24) exhibited the most cross-reactivity with cathepsin B and L.

In Vivo Characterization of sCy5-Nle-SY.

Taking into consideration all of the data from cell and tissue lysates and live mouse and human cells, sCy5-Nle-SY (20) emerged as the probe showing the highest potency and selectivity for cathepsin X. Thus, we elected to move forward with this probe for in vivo studies. We injected the probe into mice intravenously, and after 2 h of circulation tissues were harvested, lysed, and analyzed for probe labeling by fluorescent SDS-PAGE. We observed labeling of cathepsin X in liver, kidney, colon, stomach, and spleen (Figure 4A), and this was confirmed by immunoprecipitation with a cathepsin X antibody (Figure 4B). While some labeling of cathepsin S was also observed, the overall specificity profile was clearly improved compared to BMV109, which also strongly labels cathepsin B and L.

Figure 4.

In vivo characterization of sCy5-Nle-SY (20). (A) SDS-PAGE and in-gel fluorescence of tissue lysates prepared from mice that received no probe (NP), sCy5-Nle-SY, or BMV109. BMV109-labeled samples were cut from the same gel and are presented at the same gain setting as the other samples in the corresponding tissue. Gains for each tissue were set individually to display optimal contrast for cathepsin X labeling. An autofluorescent band was observed in the no-probe control (labeled as Auto). (B) Immunoprecipitation of liver and kidney samples from A with a cathepsin X-specific antibody. Also refer to Figure S2.

It is important to note that, in addition to cathepsin X and S, we also observed the labeling of additional species in vivo at 55 and 15 kDa with sCy5-Nle-SY (20). The 55-kDa species was weakly observed when kidney lysates were labeled but not the 15-kDa species. We synthesized a biotinylated Nle-SY probe in attempt to affinity purify these species; however, labeling with this probe was much weaker than the sCy5 probe suggesting that sCy5 contributes in part to selectivity (not shown). Efforts to develop new affinity probes are ongoing.

We then used confocal microscopy to image sCy5-Nle-SY (20) fluorescence in kidney cryosections after in vivo probe administration. We observed strong punctate sCy5 fluorescence reminiscent of endolysomal staining, and this signal largely overlapped with immunoreactive cathepsin X (Figure 5). Thus, we could use sCy5-Nle-SY (20) to distinguish active cathepsin X relative to total cathepsin X in tissues after in vivo administration.

Figure 5.

Confocal microscopy of cathepsin X labeling in kidney with sCy5-Nle-SY. Kidney sections from sCy5-Nle-SY-injected mice or no-probe control were analyzed for sCy5 fluorescence (red) or cathepsin X immunoreactivity (green) along with DAPI (blue) to visualize nuclei. The middle row is a zoomed-in image of the top row, as denoted by the white box. White arrowheads point to areas where the probe and immunoreactive cathepsin X are overlaid.

Sulfoxonium Ylides as a Route to Acyloxymethylketone Probes.

Many of the reported activity-based probes for cysteine proteases incorporate AOMK or phenoxymethylketone (PMK) warheads.^21,27–31 Synthesis of these electrophiles requires generation of chloromethylketone intermediates, a process that has historically been achieved, among other methods, through generation of diazomethane, an extremely explosive yellow gas.³² To avoid this potentially dangerous reaction, we utilized sulfoxonium ylides as key intermediates to make AOMK derivatives (Scheme 2, Table 2). Previous studies have shown that chiral integrity was maintained after both ylide formation and conversion to the chloromethylketone using these conditions.^33,34

Scheme 2. Synthesis of sCy5-AA-AOMK Probes via a Sulfoxonium Ylide Intermediate^a.

a(i) 1.15 eq HCl in dioxane, THF, reflux, 4 h. (ii) 2,6-dimethylbenzoic acid, KF, DMF, rt, 18 h. (iii) TFA/CH₂Cl₂ (1:1), (iv) sulfo-Cy5, PyClock, DIPEA, DMF, rt, 18 h. The synthetic route for the related compound Cbz-Lys(sCy5)-AOMK (27, 30, 33) can be found in Scheme S1.

Table 2.

Amino Acids Used in Compounds 25−33 (Scheme 2)

Boc-AA-SY	Boc-AA-CH2Cl	Boc-AA-AOMK	sCy5-AA-AOMK
Boc-Nle-SY 12	25	28	31
Boc-Phe-SY 13	26	29	32

Using this method, we successfully generated three AOMK probes bearing Nle (31), Phe (32), and Cbz-Lys (33), suggesting that this method could be broadly applied to the synthesis of diverse ABPs. We compared the reactivity of the new AOMK probes with the corresponding sulfoxonium ylide probes in living RAW2647 cells. The AOMK probes were much less potent than the ylide probes, suggesting reduced reactivity. These probes labeled cathepsin B and S but not X (Figure 6A,B), We also compared the labeling profile of sCy5-Nle-SY (20) and sCy5-Nle-AOMK (31) in RAW264.7 lysates. Here, we could only observe clear labeling of cathepsin B with the AOMK probe at 50 μM, whereas with the SY probe we observed cathepsin X labeling at 0.1 μM (Figure S3A,B). We also treated living RAW264.7 cells overnight, and again 5 μM sCy5-Nle-AOMK labeled cathepsin B and S but minimal cathepsin X (Figure S3C,D).

Figure 6.

In vitro and in vivo characterization of AOMK and sulfoxonium ylide probes. (A) Labeling of living RAW264.7 cells with the indicated AOMK and SY probes (0.1, 0.5, 1, 5 μM), as analyzed by in-gel fluorescence. In the top panel, gain settings are equal for all samples. In the bottom panel (B), gain settings were individually set to show optimal contrast for the AOMK probes. Also refer to Figures S3 and S4.

We compared the serum stability of sCy5-Nle-SY and sCy5-Nle-AOMK probes by preincubating them with fetal calf serum. No loss of activity was observed, compared to untreated probes, suggesting that both probes are stable in serum (Figure S4A). In splenic lysates from wildtype mice, sCy5-Nle-SY (1 μM) clearly labeled cathepsin X and to a lesser extent cathepsin S, but no labeling of cathepsin X was observed in splenic lysates from cathepsin X-deficient mice. As in RAW264.7 lysates, 1 μM sCy5-Nle-AOMK exhibited only weak labeling of cathepsin B splenic lysates, which did not differ between wildtype and cathepsin X-deficient mice (Figure S4B). Finally, we tested sCy5-Nle-AOMK in vivo and analyzed its labeling in tissues. Only weak labeling of cathepsin B and S was observed in the colon but not in other tissues examined (Figure S4C).

CONCLUSIONS

We have designed a new dimethyl sulfoxonium ylide warhead that exhibits unique selectivity toward cysteine cathepsin proteases in cell lysates, live cells, and in mouse and human tissues. Our best probe, sCy5-Nle-SY (20), is the most selective probe for cathepsin X to date, showing specificity in cell lysates and cells that express low levels of cathepsin S. While this probe does cross-react with cathepsin S in live macrophages and in vivo, it does not appreciably label cathepsin B or L, which is a clear improvement over the only other covalent probes that target cathepsin X (BMV109, MGP140, DCG-04). The use of sCy5-Nle-SY (20) allows for clear measurement of the activity of the cathepsin X by SDS-PAGE, whereas this was difficult with previous probes due to confounding levels of cathepsin B labeling.

Furthermore, we established that the sulfoxonium ylide warhead is stable enough for in vivo detection of cathepsin X activity. While the probe is most reliable in gel-based analyses of tissue lysates, sCy5-Nle-SY (20) signal was bright enough to detect by confocal microscopy. In conjunction with cathepsin X-specific antibodies, this method can distinguish active from inactive cathepsin X by cellular imaging and in the future could be applied to advance our understanding of the function of cathepsin X in animal models of disease.

Little is known about the preferred cleavage sequence for cathepsin X and this may be partially due to the difficulties in profiling carboxypeptidases with fluorogenic substrate libraries (i.e., its preference for a free carboxylic acid limits the choice and placement of the fluorophore). A study by Devanathan and colleagues used a fluorogenic substrate library based on the aminobenzoic acid-Phe(4-NO₂) fluorophore-quencher pair to explore the preferred P1 and P2 residues of cathepsin X.³⁵ In the P1 position, weak reactivity was observed with Met, Phe, Tyr, Thr, Gln, Glu, Lys, and Arg; however, Val, Ile, Leu, and Trp (among others) were not tolerated at all. By contrast, these residues were among the most potent in the P1 position of our sulfoxonium ylide library. In a similar study by Puzer and colleagues, in which aminobenzoic acid and Lys-(dinitrophenol) were used as the fluorophore quencher pair, Leu was well tolerated in the P1 position.³⁶ In both screens, most residues were well tolerated in the P2 position with the exception of proline. In direct contrast to this, cathepsin X has been shown to cleave natural substrates such as CXCL-12 with proline at the P2 position.³⁷ Collectively, these studies demonstrate the dependence of probe structure on specificity and warrant the development of larger sulfoxonium ylide libraries with greater diversity of P1, P2, and P3 residues. Given the observed crossreactivity of the current probes with other as yet unknown proteases (e.g., in the kidney), we anticipate that expanding the sulfoxonium ylide library will open the door to selective ABPs for other proteases in addition to cathepsin X.

In conclusion, our new sulfoxonium ylide-based probes will be valuable for understanding the contribution of cathepsin X to normal physiology and disease and for establishing cathepsin X and a drug target and diagnostic marker for cancer and other inflammatory and neurodegenerative diseases.

EXPERIMENTAL SECTION

Synthetic Methods and Key Resources.

Detailed synthetic methods and a table summarizing the source of all key reagents (antibodies, chemicals, biochemical assays, cell lines, and mouse strains) can be found within the Supporting Information.

Cell Culture.

RAW264.7 or MDA-MB-231^HM cells were cultured in DMEM containing 10% fetal bovine serum (v/v) and 1% antibiotic/antimycotic (v/v). RAW264.7 cells were passaged by scraping with a rubber policeman, while MDA-MB-231^HM cells were lifted with 0.02% EDTA (w/v) in phosphate-buffered saline (PBS).

Animals.

All experiments involving animals were approved by the Monash University Animal Ethics Committee. Male C57BL/6J mice were obtained from the Monash Animal Research Platform and used in accordance with the guidelines at 8–10 weeks of age. Snap-frozen spleens from wildtype and cathepsin X knockout mice, as described in ref 38 were obtained from the University of Calgary and used in accordance with the University of Calgary Animal Care and Use Committee.

Cell Lysate Labeling and SDS-PAGE Analysis.

Cells were harvested by scraping, washed once with PBS, and resuspended in lysis buffer containing 50 mM citrate [pH 5.5], 0.5% CHAPS (w/v), 0.1% Triton X-100 (v/v), and 4 mM DTT. Cells were incubated on ice for at least 10 min with intermittent vortexing followed by centrifugation (21×g at 4 °C for 5 min). Cleared supernatants were then transferred to a fresh tube and protein concentration was determined by BCA. Total protein (50 μg) was aliquoted into tubes in a final volume of 20 μL lysis buffer. Where indicated, JPM-OEt or MD-590 was added from a 100× DMSO stock and incubated at 37 °C for 20 min prior to probe addition. The indicated concentration of the probe was added from a 100× DMSO stock. Labeling was carried out at 37 °C for 20 min, and the reactions were quenched by the addition of 5× sample buffer (200 mM Tris-Cl [pH 6.8], 8% SDS (w/v), 0.04% bromophenol blue (w/v), 5% μ-mercaptoethanol (v/v), and 40% glycerol (v/v)). Samples were then boiled for 5 min and proteins were resolved on a 15% SDS-PAGE gel. The gels were scanned on a Typhoon 5 flatbed laser scanner at 633/670 nm excitation/emission to detect sCy5 fluorescence.

Live Cell Labeling.

RAW cells or MDA-MB-231^HM cells were plated in 12-well plates. Where indicated, MDV-590, a closely related analogue to the cathepsin S-specific inhibitor MIV-247,²⁴ Z-FL-COCHO³⁹ or DMSO (vehicle) was added at 10 μM or 20 μM, respectively, from a 1000× DMSO stock for overnight incubation. When the cell density reached 80%, the indicated probes were added at the indicated concentrations from a 1000× DMSO stock and allowed to incubate for the indicated time. Media was then removed and replaced with PBS. The cells were then scraped and transferred to tubes, and lysis and SDS-PAGE analysis were carried out as above, except skipping the probe addition step.

Serum Stability Test.

Probes (5 mM) were diluted 10-fold in fetal calf serum followed by 3 h incubation at 37 °C. Probe-containing serum was then diluted 10-fold in serum-free DMEM, followed by incubation with living RAW265.7 cells for 2 h (5 μM final probe concentration). Serum-free media containing 5 μM probe was used as a control. Cells were then harvested, lysed, and analyzed by in-gel fluorescence as above.

Tissue Analysis.

Tissues or biopsies were harvested from healthy mice or patients, respectively, and snap frozen. At the time of analysis, lysis buffer was added at 10× (v/w), and tissues were sonicated on ice. Cleared lysates were labeled with the indicated probe and analyzed as above. For in vivo labeled tissues, mice were first injected intravenously via the tail vein with sCy5-Nle-SY, BMV109, or sCy5-Nle-AOMK (50 nmol in 100 μL 10% DMSO/PBS (v/v) or vehicle control). Tissues were harvested after 2 h and analyzed as above except without further probe addition.

Immunoblotting.

After detection of in-gel fluorescence, human cancer samples were transferred to a nitrocellulose membrane using the TransBlot system (Bio-Rad). Loading and transfer efficiency were assessed by Ponceau Stain (Sigma). The membrane was then incubated overnight at 4 °C with a goat anticathepsin X antibody (1:1000) in Odyssey Blocking Buffer (LiCor) diluted by 50% in PBS (v/v) containing 0.05% Tween-20 (v/v; PBS-T). After washing the membrane three times with PBS-T, it was incubated with donkey antigoat-IRDYE800 (1:10 000) at rt for 1 h. After washing, binding was detected by scanning the membrane on a Typhoon 5 (IR-long filter).

Immunoprecipitation Assay.

Probe-labeled lysate from above (in sample buffer) was divided into input or pulldown (~50 g total protein each). The input sample was stored at −20 °C. The pulldown sample was diluted in 500 μL IP buffer (PBS [pH 7.4], 0.5% NP-40 (v/v), 1 mM EDTA). Goat anticathepsin X antibody (10 μL) was added along with 40 μL slurry of prewashed Protein A/G agarose beads. Samples were rotated overnight at 4 °C. Beads were then washed four times with IP buffer followed by a final wash in 0.9% NaCl (w/v). Beads were then resuspended in 2××sample buffer and boiled. The pulldown supernatants, alongside the input samples, were analyzed by fluorescent SDS-PAGE as above.

Confocal Microscopy.

Kidney tissues from mice that received sCy5-Nle-SY (or vehicle control) above were fixed overnight in 4% paraformaldehyde in PBS (w/v) followed by overnight cryoprotection in 30% sucrose (v/v). Tissues were embedded in OCT, frozen on dry ice, and sectioned at 10 μm. Immunostaining for cathepsin X was carried out according to standard protocols. In brief, sections were air-dried, fixed in cold acetone for 10 min, air-dried again, and then rehydrated in PBS. Sections were blocked in PBS containing 3% normal horse serum (v/v) with 0.1% Triton X-100 (v/v). Goat anticathepsin X was added at 1:100 in blocking buffer overnight at 4 °C. Sections were then washed, and a secondary antibody, donkey antigoat-AlexaFluor594, was added at 1:500 for 1 h at rt. Sections were stained with DAPI for 5 min, washed, and mounted with ProLong Diamond. Staining was analyzed using a Leica SP8 inverted confocal microscope.

Statistical Analysis.

All experiments were performed with at least three biological replicates. Data are reported as means ± SEM. Statistical significance was determined by comparing two groups using a Student’s t test, and p values of less than 0.05 were considered significant.

ABBREVIATIONS

ABP

activity-based probe

SY

sulfoxonium ylide

AOMK

acyloxymethylketone

PMK

phenoxymethylketone