ABPP is a powerful new technique that provides an unbiased, functional analysis of the proteome. Many genomic, transcriptomic and proteomic approaches only identify changes in protein quantity, which does not reflect protein functional status. ABPP offers additional insights over metabolomics, since specific enzymes which are differentially active can be identified for further characterization. This study demonstrates the ease and utility of ABPP with its potential applicability to different renal disease models. This study also provides a unique in-depth activity and compositional analysis of serine hydrolases in normal urine. Serine hydrolase activity frequently, but not invariably, correlates with total protein, thus emphasizing the need for functional proteomic characterization. Interestingly, three active esterases were identified that have limited functional characterization and no previously specified role in the urine, and these may be important targets for further analysis. Equally importantly, we identified active serine proteases involved in key renal physiological processes and innate immunity, and these are discussed below.
Kallikrein-1 (tissue kallikrein) cleaves kininogen to produce vasoactive kinins and their activity is primarily mediated through the bradykinin 1 (B1) and bradykinin (B2) receptors . The pleiotropic effects of kinins include vasodilation, natriuresis, diuresis, anti-fibrotic and anti-hypertrophic actions . Bradykinin-dependent activation of the B2 receptor causes natriuresis by inhibiting sodium reabsorption in the collecting duct . However, tissue kallikrein also acts in a kinin-independent manner to regulate sodium reabsorption [16, 17]. Tissue kallikrein is a locally produced regulator that acts luminally on ENaC receptors of the principal cells by cleaving its γ-subunit, thus promoting increased sodium reabsorption [16, 17]. While it modulates sodium absorption, its actions are not essential, as tissue kallikrein deficient mice maintain normal blood pressure and extracellular fluid volume status.
Tissue kallikrein is also implicated in renal calcium homeostasis . Calcium is actively reabsorbed in the distal convoluted tubule through the apical transient receptor potential channel vanilloid subtype 5 (TRPV5), transported through the cytosol by calbindin-D28K, and then basolaterally transported via the Na+/Ca2+ exchanger and Ca2+ ATPase transporters . Luminal tissue kallikrein stimulates calcium reabsorption by activating the B2 receptor, which results in protein kinase C (PKC)-dependent phosphorylation of TRPV5 . This causes stabilization and accumulation of TRPV5 at the plasma membrane, thereby increasing net calcium reabsorption . Notably, regulation of the TRPV5 receptor was specific to luminal, not basolateral, tissue kallikrein .
Tissue kallikrein is also a unique aldosterone-independent kalliuretic factor that allows for rapid adaptation to a dietary potassium load in the cortical collecting duct . Luminal tissue kallikrein promotes potassium secretion by stimulating ENaC activity and sodium reabsorption in the principal cells, as described above [16, 20]. Furthermore, tissue kallikrein inhibits potassium reabsorption in intercalated cells by decreasing H+/K+-ATPase expression and activity, resulting in a net kalliuretic effect . Importantly, luminal tissue kallikrein demonstrated inhibition of H+/K+-ATPase activity by 70%, whereas basolateral tissue kallikrein had no effect .
In the kidney, tissue kallikrein is synthesized primarily in the connecting tubule cells, and to a lesser extent in the distal convoluted tubules and cortical collecting duct . Taken together, these data suggest that tissue kallikrein is synthesized proximally and released into the pro-urine of the tubular lumen to act distally in a paracrine fashion to regulate sodium, calcium and potassium handling. Our observation that tissue kallikrein is present in an active conformational state in normal human urine expands on the mouse and in-vitro work done thus far, and suggests that it may reflect real-time regulation of sodium, calcium and potassium handling. Shedding of active tissue kallikrein in the urine may also represent a rapid means to down-regulate its activity.
Kallikrein-related peptidase (KLK3), or prostate specific antigen, is located on the same gene locus as tissue kallikrein (KLK1), however its function relates primarily to the liquefaction of semen to allow sperm to move freely, and has no known renal physiological functions. Conversely plasma kallikrein (KLKB1) is located on a different gene locus but has very similar physiological functions as tissue kallikrein . Plasma kallikrein cleaves high molecular weight kininogens to release vasoactive kinins that activate the B2 receptor . Plasma kallikrein may potentially contribute to B2-receptor mediated natriuresis and calcium reabsorption, in a manner similar to tissue kallikrein.
Urokinase-type plasminogen activator cleaves the zymogen plasminogen into plasmin, a serine protease. Urinary plasmin directly activates ENaC by cleaving its γ-subunit to promote sodium reabsorption in nephrotic patients and mice [22–24], in a mechanism similar to tissue kallikrein. Svenningsen et al. postulated that a defective glomerular filtration barrier allowed for passage of plasmin to activate ENaC, thus contributing to hypertension and edema in nephrotic syndrome . However, the increased sensitivity of ABPP and mass spectrometry techniques utilized in our study permitted the identification of both plasmin and urokinase-type plasminogen activator in healthy individuals. These novel data raise the possibility that urinary plasmin plays a paracrine regulatory role via ENaC in normal renal physiology, not just nephrotic syndrome.
Urinary plasmin from nephrotic individuals also decreases calcium reabsorption by inhibiting the TRPV5 receptor . Urinary plasmin catalyzes protease-activated receptor-1, which promotes phosphorylation of TRPV5 at a different site from tissue kallikrein, resulting in decreased channel pore size and calcium reabsorption . Tudpor et al. did not identify plasmin in normal urine by Western blot, but they demonstrated plasmin activity with a plasmin-specific activity assay in normal urine , which is consistent with our finding that plasmin is present in an active conformational state in normal urine. Interestingly, Tudpor et al. demonstrated that plasmin inhibits calcium influx with 50% inhibitory concentration of ~3nM . Taken together, this suggests that urinary plasmin may play a physiological role at low concentrations, which is readily detectable with mass spectrometry-based techniques.
In a cardiac model, cathespin A localizes to atrial tissue myocytes where it cleaves angiotensin I to release angiotensin 1–9 [26, 27]. Angiotensin 1–9 is a bio-active peptide that enhances the kinin effect on the B2 receptor and augments arachidonic acid and nitric oxide release from endothelial cells [26, 27]. Grobe et al. used MALDI-TOF MS to evaluate bio-active peptides resulting from renal processing of angiotensin II , but there is no renal data regarding cathepsin A on angiotensin I. Our observation that active cathepsin A is present in normal urine suggests that it may play a physiological role, such as potentiating the effect of renal-derived bradykinin.
Several of the serine hydrolases detected with ABPP have been shown to be involved in innate immunity. Mannan-binding lectin serine protease 2 (MASP2) participates in activating the lectin pathway of the complement cascade when MASP2 cleaves C4 [29, 30]. Complement C1r subcomponent-like protein (C1r-LP), homologous to C1r, participates in activating the classical complement pathway via cleavage of pro-C1s . The role of C1r-LP in complement-mediated functions is still unclear with suggestions of both activating and inhibitory roles [31, 32]. Proteinase 3 acts in concert with neutrophil elastase to promote neutrophil activation by cleaving and inactivating the anti-inflammatory progranulin . Furthermore, proteinase 3 has been demonstrated to process human cathelicidin-18 into the antimicrobial peptide LL-37 . Taken together, these observations raise the intriguing possibility that this group of enzymes may play an active role in regulating the innate immunity of the urinary tract.
There are some limitations to this study. First, this is strictly an observational characterization of serine hydrolases in normal individuals. While there is literature to support their activities in urine, these observations are only hypothesis-generating since their physiological role cannot be validated in the current model. Secondly, the affinity-purification step was effective but did not fully retrieve all the FP-TAMRA labelled proteins in the sample. Given the ENaC regulators identified, we anticipated the identification of other ENaC regulators, such as furin and prostasin . Both furin and prostasin were present on the 2D LC-MS/MS compositional analysis, but their active species were not identified with ABPP. One possibility is that tissue kallikrein and plasmin are physiologically dominant ENaC regulators, and furin and prostasin play a limited role if any. However, since all the bands were not enriched with affinity purification we cannot categorically exclude their activity.
The strengths of this study relate to the novel ABPP methodology which provides a functional proteomic characterization that may be more physiologically relevant than a straightforward compositional analysis. Indeed, it provides insight into which enzymes may be active versus filtered peptide fragments that are detectable with highly sensitive MS/MS approaches but may have no function. ABPP allows for simultaneous assessment of activity for most of an enzyme family and this offers clear advantages for the discovery of novel species with as yet undetermined substrate specificities. The ease of visualization and comparison of labeled proteins in different samples offers the potential for rapid comparative analysis of samples. These properties, linked with the potential for affinity purification and MS-based identification of labeled enzymes, markedly enhance the utility of the approach. Critically, identification of specific enzyme activity with ABPP offers the potential for developing rapid colorimetric or fluorometric urine screening assays using immobilized substrates (e.g. urine dipstick).