KLK5

KLK5 is a secreted trypsin-like serine protease that initiates the epidermal desquamation cascade by directly degrading the corneodesmosome adhesion proteins corneodesmosin (CDSN), desmoglein 1 (DSG1), and desmocollin 1 (DSC1) at the acidic pH of the stratum corneum and by converting pro-KLK7 to its active form (PMID:15140227). Its activity is spatially confined by inhibitors whose action is pH-tuned: LEKTI (SPINK5) fragments bind and irreversibly inhibit KLK5 at neutral pH and release active enzyme at acidic pH, restricting proteolysis to the outer cornified layers (PMID:17596512), while SPINK9 provides palmo-plantar-specific inhibition through a reactive-loop His48 protonation mechanism that slows dissociation from the enzyme (PMID:22505519). KLK5 and its inhibitor are packaged separately into lamellar granules, and loss of LEKTI in Netherton syndrome skin produces unrestrained KLK5/KLK7 activity and premature loss of corneocyte cohesion (PMID:15675955). Consistent with this, allosteric antibody inhibition of KLK5 restores barrier integrity in mouse models of Netherton syndrome and atopic dermatitis (PMID:36516271), and STAT3 acts as an upstream transcriptional regulator that represses KLK5 while inducing SPINK5 (PMID:34378233). Genetic studies reveal that KLK5's role is context-dependent rather than purely pro-desquamatory: in corneodesmosin-deficient epidermis its deletion paradoxically worsens proteolysis and barrier loss, indicating a regulatory function (PMID:40943523). Beyond skin, KLK5 cleaves DPP4 from the surface of Th17 cells to generate soluble plasma DPP4 (PMID:29107298), primes and activates betacoronavirus spike proteins at both the S1/S2 and S2' sites to promote pulmonary infection (PMID:39163389), and influences cancer signaling—both as a tumor suppressor that suppresses the mevalonate pathway to inactivate RhoA in breast cancer (PMID:24158494) and, with KLK7, as a driver of KLK14-dependent PAR2/RhoA/NF-κB signaling in cervical carcinoma (PMID:40753921); its mRNA is stabilized by IGF2BP3 in an m6A-dependent manner in gallbladder carcinoma (PMID:36313631).

1. 2004 High

   Established KLK5 as the initiating protease of epidermal desquamation by showing it directly cleaves corneodesmosome adhesion proteins and activates a downstream protease.

   Evidence In vitro cleavage assays of recombinant/epidermal CDSN, DSG1, DSC1 and pro-KLK7 zymogen activation at acidic pH

   PMID:15140227

   Open questions at the time

   • Did not establish how KLK5 activity is restricted in intact skin
   • Endogenous KLK5 zymogen activation mechanism not addressed
2. 2005 High

   Connected KLK5 spatial regulation to disease by showing separate lamellar-granule packaging of protease and inhibitor and unrestrained activity in LEKTI-deficient skin.

   Evidence Confocal and immunoelectron microscopy of normal and Netherton syndrome skin

   PMID:15675955

   Open questions at the time

   • Localization is correlative; did not quantify enzymatic activity in situ
   • Relative contributions of KLK5 vs KLK7 to the split not resolved
3. 2007 High

   Defined the biochemical mechanism for zone-restricted desquamation: pH-dependent, irreversible LEKTI inhibition of KLK5 that releases active enzyme in the acidic outer stratum corneum.

   Evidence Inhibition kinetics, pH-titration binding studies, and LEKTI fragment characterization

   PMID:17596512

   Open questions at the time

   • In vitro kinetics may not capture in vivo concentrations/microenvironment
   • Did not address other endogenous KLK5 inhibitors
4. 2012 High

   Identified a tissue-restricted second inhibitor (SPINK9) and resolved the structural/charge basis for its pH-dependent, KLK5-selective inhibition.

   Evidence In vitro inhibition with reactive-loop mutants, binding kinetics, and computational modeling

   PMID:22505519

   Open questions at the time

   • Modeling not confirmed by experimental KLK5–SPINK9 structure
   • Physiological importance in palmo-plantar desquamation not tested in vivo
5. 2014 Medium

   Revealed a tumor-suppressive function distinct from desquamation, placing KLK5 upstream of mevalonate-pathway-dependent RhoA activation in breast cancer.

   Evidence Stable KLK5 re-expression in MDA-MB-231 cells with transcriptomics, lipid synthesis and RhoA activity assays, GGPP rescue, and xenografts

   PMID:24158494

   Open questions at the time

   • Whether the effect requires KLK5 catalytic activity not established
   • Single cell line model
   • Apparent contrast with pro-oncogenic roles in other tissues unresolved
6. 2017 Medium

   Extended KLK5 substrate repertoire beyond skin by identifying DPP4 as a surface substrate shed from Th17 cells, linking KLK5 to elevated plasma DPP4 in metabolic disease.

   Evidence Co-IP, in vitro/ex vivo cleavage assays, surface DPP4 flow cytometry, and docking in CD4+ T cells

   PMID:29107298

   Open questions at the time

   • Reciprocal validation of the KLK5–DPP4 interaction limited
   • Causal contribution to diabetes pathology not established in vivo
7. 2021 Medium

   Placed KLK5 under transcriptional control of STAT3, which coordinately represses KLK5 and induces its inhibitor SPINK5 in keratinocytes.

   Evidence Keratinocyte-specific STAT3 knockout mice plus STAT3 gain/loss-of-function in keratinocytes with qRT-PCR and transcriptomics

   PMID:34378233

   Open questions at the time

   • Direct STAT3 binding to KLK5/SPINK5 promoters not demonstrated
   • Single lab study
8. 2022 High

   Provided a structurally validated allosteric inhibition strategy and demonstrated therapeutic benefit of KLK5 blockade in barrier disease models.

   Evidence X-ray crystallography of a KLK5–Fab complex with bispecific antibody treatment in Netherton and atopic dermatitis mouse models

   PMID:36516271

   Open questions at the time

   • Antibody is bispecific (KLK5/KLK7), so KLK5-specific contribution to rescue is convolved
   • Human efficacy not addressed
9. 2022 Medium

   Identified post-transcriptional regulation of KLK5 via m6A-dependent mRNA stabilization driving PAR2/AKT oncogenic signaling.

   Evidence RIP, MeRIP, RNA stability and luciferase assays with IGF2BP3 depletion rescue in gallbladder carcinoma cells

   PMID:36313631

   Open questions at the time

   • Performed in a single tumor type
   • Whether secreted KLK5 protease activity mediates the PAR2 effect not directly shown
10. 2024 High

    Defined KLK5 as a host protease that primes and activates betacoronavirus spike proteins, making it a determinant of pulmonary infection and a druggable antiviral target.

    Evidence In vitro spike cleavage assays, infection of differentiated HBECs, KLK5 inhibitor treatment, and MERS-CoV/SARS-CoV-2 mouse models

    PMID:39163389

    Open questions at the time

    • Inhibitor (ursolic acid) selectivity for KLK5 in vivo not fully delineated
    • Relative in vivo contribution of KLK5 vs KLK12/KLK13 not quantified
11. 2025 Medium

    Demonstrated that KLK5's role is context-dependent and protective in corneodesmosin-deficient skin, where its loss aggravates rather than rescues over-desquamation.

    Evidence shRNA CDSN knockdown in human epidermal equivalents and Klk5/Cdsn double-knockout mice with ultrastructural and barrier assays

    PMID:40943523

    Open questions at the time

    • Identity of the compensating proteases not established
    • Molecular basis of the protective/regulatory function unknown
12. 2025 Medium

    Established KLK5 (with KLK7) as a driver of cervical carcinogenesis through KLK14 activation feeding PAR2/RhoA/NF-κB signaling.

    Evidence KLK5/KLK7 double-knockout mice with RNA-seq, NF-κB/RhoA reporter assays, and human biopsy analysis

    PMID:40753921

    Open questions at the time

    • Double-knockout design does not isolate KLK5-specific contribution
    • Direct KLK5–KLK14 activation in this context inferred rather than reconstituted

• How KLK5 switches between pro-desquamatory, protective, antiviral, and opposing oncogenic roles across tissues, and how its zymogen is endogenously activated, remains unresolved.
• Context-dependent tumor-suppressor vs oncogenic behavior unreconciled
• Endogenous activator(s) of pro-KLK5 not identified
• Compensating proteases in CDSN-deficient skin unknown