If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Keratin cytoskeletal proteins protect the epidermis against mechanical stress, strengthen keratinocyte cohesion, and regulate cell growth in keratin isotype-specific ways. Expression of keratins is modulated upon differentiation and injury and in disease settings, raising the question of how individual keratin proteins are degraded during differentiation and pathological processes including epidermolysis bullosa simplex (EBS) (
). Recently, gain-of-function mutations abolishing the start codon in the ubiquitin ligase substrate adaptor KLHL24 were reported to target K14 for degradation, weakening the keratin cytoskeleton in basal keratinocytes and thereby leading to EBS (
). Subsequent articles identified the same KLHL24 mutations, accompanied by altered keratin organization, cytolysis, and skin blistering, but reported persistent or even elevated K5 and K14 protein levels (
). These contrasting findings suggest a more complex mechanism and raise the issue whether K14 represents the prime target of KLHL24.
Missense mutations in KLHL16 (also named gigaxonin), a widely expressed ubiquitin ligase substrate adaptor, cause giant axonal neuropathy, characterized by extensive cytoplasmic aggregates of neurofilaments, vimentin, and GFAP (
), suggests that KLHL16 may be involved in the regulation of keratin turnover.
To shed light on the respective contribution of KLHL24 and KLHL16 to keratin degradation, we compared their activity upon forced overexpression in a panel of mouse and human keratinocytes, expressing intact and EBS-associated mutant keratin isotypes (
). At 72 hours after transfection of mouse and HaCaT keratinocytes (efficiency ∼30%), KLHL24-GFP was detectable, frequently in small aggregates, throughout the cytoplasm (Figure 1b and d ; transfectants are marked with stars in a and c). However, KLHL24-GFP did not cause loss of K14-containing cytoskeletons (Figure 1a and c). In contrast, overexpression of KLHL16-GFP in mouse and human HaCaT keratinocytes and in normal human keratinocytes (efficiency ∼90%) induced partial loss of K14 in 44% and complete loss of K14 in 37% of all transfectants (Figure 1e–j, o, and t). To examine whether the nature of tags used affected the outcome (GFP, Flag), we repeated the experiments presented in Figure 1 with both Kelch proteins tagged with Flag, cloned in lentiviral expression vectors. Immunofluorescence analysis of wild-type mouse keratinocytes at 48, 72, and 96 hours after transduction (efficiency ≥ 90%) documented the same result as the experiments described in Figure 1. At 96 hours after transduction of Flag-KLHL16, keratins were degraded completely in most cells, as shown by immunofluorescence and Western blotting (see Supplementary Figure S1a–f and m online), whereas keratins remained unaltered in KLHL24-Flag transfectants (see Supplementary Figure S1g–m). At the same time, the tubulin and the actin cytoskeletons were not affected by overexpression of KLHL16 in mouse keratinocytes (see Supplementary Figure S2a–f online), similar to overexpression of KLHL16 in fibroblasts (
Figure 1Overexpression of KLHL16, but not of KLHL24, causes degradation of normal and mutant keratins. (a–d) Immunostaining for K14/GFP shows unaltered presence of the keratin cytoskeleton in WT mouse and human HaCaT keratinocytes 72 hours after transfection with KLHL24-GFP (efficiency ∼30%). (e–h) Immunostaining for K14/Flag shows extensive degradation of K14 in mouse, HaCaT, and EBS keratinocytes and corresponding controls 72 hours after transfection with Flag-KLHL16 (efficiency ∼90%). (i–n) Immunostaining for K14/Flag-KLHL16 in cells expressing K14 p-R125C and K5 p.E477K in EBS keratinocytes and corresponding healthy controls (NHKs) and (p–s) for K14/Flag-KLHL16 in KtyI–/– mouse keratinocytes expressing WT K14 and K14 p.R131P shows degradation of keratin filaments and aggregates, 72 hours after overexpression of Flag-KLHL16. Scale bars in a–s = 10 μm. Asterisks mark transfected cells. (t) Quantification of KLHL16-mediated keratin expression in transfectants: K14 present and partial and complete loss of K14. Cells analyzed: WT, 703; KtyI–/–, 464; KtyI–/–K.p14R131P, 204. EBS, epidermolysis bullosa simplex; NHK, normal human keratinocytes; WT, wild type.
Most importantly, KLHL16 mediated degradation of mutant keratins (K5del, K5 p.E477K, K14 p.R125C, K14 p.M272T) in EBS keratinocytes (Figure 1k–n, and see Supplementary Figure S1n–q) and of K14 p.R131P in mouse keratinocytes, corresponding to the human K14 p.R125P mutation causing severe generalized EBS (former Dowling-Meara) (Figure 1p–s). Although mouse keratinocytes contain exclusively K14 p.R131P-positive keratin aggregates, their degradation was more efficient than that of wild-type filamentous keratins, indicated by 91% of keratin-free cells after KLHL16 expression (Figure 1t). The rapid and efficient degradation of several K5 and K14 mutations in human and mouse transfectants underscores its ability to degrade a range of normal and mutant keratins in a disease-relevant context (Figure 1k–n, r, and s, and see Supplementary Figure S1n–q). Whether certain keratin mutants resist KLHL16-mediated degradation, analogous to GFAP (
To address the range of KLHL16 targets among keratins, K5, K6, K16, and K17 were analyzed in mouse keratinocytes upon KLHL16 overexpression. At 72 hours after transfection, these keratins were degraded in most transfectants (Figure 2a–h ). To investigate whether K17 is degraded independently of K14, KLHL16 was transfected into mouse keratinocytes expressing either K14 or K17 as the sole type I keratin (KtyI–/–K14, KtyI–/–K17), in addition to the endogenous type II keratins K5 and K6 (
). The rapid and efficient degradation of K17 suggests that it is a KLHL16 target in addition to K14 (Figure 2 i–j). Comparison of keratin degradation over time in mouse keratinocytes indicated that K16 and K17 were degraded faster than K5, K6, and K14 (see Supplementary Figure S3a–t online). Whether this indicates preference for a subset of keratins or results from the different abundance of these keratins is subject to future analysis. The ability of ectopically expressed KLHL16 to degrade several differentiation-associated keratins is consistent with its similar activity toward vimentin, peripherin, neurofilaments, and GFAP in various cell types (
Figure 2Flag KLHL16 expression results in degradation of major type I and type II keratins. (a–h) Immunostaining of various keratins 72 hours after transfection with Flag KLHL16. Note that K5, K6, K16, and K17 are degraded in addition to K14 after KLHL16 overexpression. (i–j) Immunostaining for K17/Flag in KtyI–/– keratinocytes expressing K17 shows degradation of K17-positive filaments in the absence of K14 at 72 hours after overexpression of Flag KLHL16. (k–r) Immunostaining for K14/Flag in WT mouse keratinocytes 72 hours after transfection with Flag KLHL16. At 60 hours after transfection, cells were treated with the solvent DMSO or with 10 μmol/L MG132. Treatment with MG132 inhibits the degradation of K14 in Flag KLHL16 transfectants. Scale bars in a–r = 10 μm. Asterisks mark transfected cells. (s) Quantification of KLHL16-mediated keratin expression in transfectants: K14 present and complete loss. Number of cells analyzed = 180 in DMSO and 220 in MG132. h, hour; M, mol/L; WT, wild type.
To examine whether KLHL16 targets keratins for degradation via the ubiquitin-proteasome system, proteasome activity was inhibited by MG132. Treatment of KLHL16-transfected wild-type mouse keratinocytes with 10 μmol/L of MG132 prevented clearance of keratins in 84.6% of transfected cells, whereas only 17.9% of vehicle-treated cells retained their keratin cytoskeleton, as shown by K14 staining. These data suggest that KLHL16 targets keratins for proteasome degradation (Figure 2k–s). Finally, we examined the expression of KLHL16 mRNA in several mouse tissues including skin and in keratinocytes (see Supplementary Figure S2g). In line with
Our data clearly show that upon overexpression, KLHL16 degrades several keratins including K6, K16, and K17, associated with wound healing, migration, and inflammation, states that are known to require extensive keratin remodeling (
). At the same time, the disease mechanisms underlying KLHL24-related EBS still remain to be identified. The low levels of KLHL16 in most tissues including the epidermis and other epithelia indicate its tight control. The accumulation of cytoplasmic keratin aggregates in many keratinopathies raises the question of why KLHL16 fails to degrade them. Also, it is tempting to hypothesize that KLHL16 is involved in the rapid degradation of K1 and K10 upon wounding (
). How KLHL16 recognizes normal and mutant keratins and possibly other substrates crucial for skin integrity, whether its activity is exclusively linked to substrate degradation, and how its expression is regulated, are pressing questions. The efficient degradation of several disease-associated keratin mutants suggests that boosting the activity of KLHL16 may be suitable for the removal of keratin aggregates in EBS and additional keratinopathies.
Conflict of Interest
The authors state no conflict of interest.
Acknowledgments
We thank Aileen Wingenfeld, Max Bonsack, and Miriam Richter (Magin laboratory) and Juna Leppert for expert technical assistance and Yinghong He (Has lab). This work was partially supported by the DFG (MA1316-15, MA1316-17, MA1316-19, MA1316-21, INST 268/230-1) and Debra International (to CH). Skin samples were obtained for diagnostic purposes and from the left over material, cells were isolated. Patients or their relatives gave written informed consent. The study was approved the Ethics Committee of the University Freiburg.
30140-4&id=gr1.jpg){kind=link}
30140-4&id=gr2.jpg){kind=link}