sign in sign up
<<
Home , Keratin 15, Keratin 16, Keratin 17, Keratin 21, Keratin 5, Lamin

Developmental Essay

Keratins Are Going Nuclear

Ryan P. Hobbs,1,6 Justin T. Jacob,1,6 and Pierre A. Coulombe1,2,3,4,5,* 1Department of and Molecular Biology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD 21205, USA 2Department of Biological Chemistry, School of Medicine, Johns Hopkins University, Baltimore, MD 21205, USA 3Department of Oncology, School of Medicine, Johns Hopkins University, Baltimore, MD 21205, USA 4Department of Dermatology, School of Medicine, Johns Hopkins University, Baltimore, MD 21205, USA 5The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, MD 21205, USA 6Co-first author *Correspondence: [email protected] http://dx.doi.org/10.1016/j.devcel.2016.07.022

Previously thought to reside exclusively in the cytoplasm, the cytoskeletal 17 (K17) has been recently identified inside the nucleus of tumor epithelial cells with a direct impact on cell proliferation and expression. We comment on fundamental questions raised by this new finding and the associated significance.

Everything should be made as regulator autoimmune regulator (AIRE) levels of mRNAs that code for inflamma- simple as possible, but not simpler. (Hobbs et al., 2015) or the cell-cycle inhib- tory and immune cytokines in skin and itor p27KIP1 (CDKN1B) (Escobar-Hoyos cervical tumor paradigms. A key observa- —Albert Einstein et al., 2015)? Whereas a number of intri- tion in the study by Hobbs et al. was cate mechanisms could account for this that manipulating K17 expression had a Intermediate filaments (IFs) are 10-nm- finding, a simple explanation is that K17 marked impact on the distribution of the wide fibrous polymers that, alongside would itself be present and function inside AIRE protein inside the nucleus, i.e., , F-, and associated the nucleus, a possibility that had not punctate versus diffuse, in response to , form the in multi- been considered until recently. The two relevant stimuli. Escobar-Hoyos et al. (Es- cellular organisms. Remarkably, IFs are aforementioned studies (Hobbs et al., cobar-Hoyos et al., 2015) found K17 to formed by a large and diverse group of 2015; Escobar-Hoyos et al., 2015), in occur in the nucleus of cervical tumor >70 proteins (mass range 20–240 kilo- addition to an unbiased screen to iden- epithelial cells, following up on the obser- daltons) that are expressed and regulated tify nucleocytoplasmic shuttling proteins vation of a slower proliferation rate and in a tissue-, differentiation-, and context- (Kumeta et al., 2013), have separately increased incidence of G1 cell-cycle ar- specific fashion (Pan et al., 2013; Schwe- converged in demonstrating the presence rest when K17 expression is silenced via izer et al., 2006). IFs are abundant of endogenous K17 and other keratin pro- RNAi. In all three studies, uncovering the intracellular elements that partake in all teins inside the nucleus of tumor epithe- presence of keratin proteins inside the nu- basic cellular functions, including growth, lial cells. This unexpected finding raises cleus of cultured tumor cells necessitated death, and virtually everything in between, several questions and calls for new ideas the use of a potent inhibitor of CRM1/Ex- and are necessary to maintain cellular regarding the rationale underlying the portin 1-mediated nuclear export, namely integrity and function in the face of me- context-dependent regulation and signifi- Leptomycin B (LMB), so as to ‘‘trap’’ the chanical and other forms of stress (Pallari cance of K17 and other IF proteins for protein of interest inside the nucleus and and Eriksson, 2006; Pan et al., 2013; cells, tissues, and organs, in both health thus facilitate its detection. This same Toivola et al., 2010). Genetically deter- and disease. strategy had been used before to un- mined in IF proteins account The latest developments regarding nu- cover the presence of other cytoskeletal- for >100 diseases to date (Omary et al., clear are as follows. Kumeta associated proteins, e.g., LPP (lipoma- 2004; Szevereneyi et al., 2008; see the et al. (Kumeta et al., 2013) identified preferred partner; Petit et al., 2000), Zyxin Human Database several cytoskeletal proteins, including (Nix et al., 2001), ZO-2 (zona occludens at http://www.interfil.org for an up-to- the actin-binding protein a-, the protein 2; Islas et al., 2002) and KEAP1 date account), rendering their study versatile cellular cytolinker , and (Velichkova and Hasson, 2005), inside immensely relevant to medicine. select keratins (e.g., K7, K8, K17, and the nucleus. No such inhibitory treatment With the notable exception of the K18), in the context of a screen for nuclear is needed, however, to detect nuclear K17 nuclear , all other IFs (n = 68) matrix components that dynamically in BT-20-cultured cells (Figure 1A) and in are believed to reside and function exclu- shuttle in and out of the nucleus in biopsy samples of human BCC skin tu- sively in the cytoplasm. With this in mind, cultured HeLa cells. Hobbs et al. (Hobbs mors by microscopy (Hobbs et al., 2015) how could one account for the puzzling et al., 2015) found K17 to occur in the (Figure 1B), or to biochemically detect observation, independently made by two nucleus of human and mouse tumor the presence of K17 in subcellular frac- laboratories, that the type I IF protein ker- epithelia as part of a quest to decipher tions enriched for nuclear proteins (Esco- atin 17 (K17) impacts the nuclear localiza- the mechanisms underlying the ability of bar-Hoyos et al., 2015; Hobbs et al., 2015; tion and function of the transcriptional this keratin to regulate the induction and Kumeta et al., 2013), which altogether

Developmental Cell 38, August 8, 2016 ª 2016 Elsevier Inc. 227 Developmental Cell Essay

A D Human epithelial tumor cell line Head Rod Tail

BT-20 (breast) HeLa (cervix) A431 (vulva) 1A 1B 2A 2B Type I K17/DAPI K9 K10 B K12 K13 K14 K15 Veh K16 K17 B K18 B K19 B K20 K23 M B K24 K25 B K26 B K27 K28 K31 B K32 B B LMB K33a K33b K34 B K35 B B K36 B K37 K38 K39 B K40 B Human BCC biopsy Type II K1 K17/DAPI K2 K3 K4 B M K5 K6a B K6b B K6c B K7 B K8 K71 B K72 B K73 K74 B K75 B K76 B K77 B K78 B K79 K80 B K81 K82 K83 C 399 400 K84 B K17 (Homo sapiens) -EIATYRRLLEGEDAHLTQYE LT K-KEPVTTRQVRTI-I K85 K86 K17 (Pan troglodytes) -EIATYRRLLEGEDAHLTQYK-KEPVTTRQVRTI- K17 (Mus musculus) NLS -EIATYRRLLEGEDAHLTQYKPKEPVTTRQVRTI- Type III K17 (Xenopus laevis) -EIATYRRLLEGEDAHLSQSQKDGSRTTVQVRTI- B GFAP B B K17 (Danio rerio) -EIAEYRRLLDGE-ATSVSTSSSKTSTTRKVVTI- B B Conservaon B B -555455555455454434333333355445455- Vimenn B

Type IV Head Rod Tail α- B B NF-H B B NF-M B NF-L B B Nesn -α Synemin-β 194 197 199 204 206 ###*** Type V K17 (Homo sapiens) A B M M -ADINGLRRVLDELTLARADLEMQIENLKEEL- M B K17 (Pan troglodytes) -ADINGLRRVLDELTLARADLEMQIENLKEEL- B M Lamin C1 B M M B NES K17 (Mus musculus) -ADINGLRRVLDELTLARADLEMQIENLKEEL- Lamin C2 B M M B K17 (Xenopus laevis) -ADINGLRRVLDDLTIARSDLEIQIESLKEEL- Type VI K17 (Danio rerio) -ADISGLRKVLDELTMTRSDLEMQIEGLKEEL- Phakinin/CP49 B Conservaon -5554555455545534535554555355555- Filensin/CP115

Figure 1. Nuclear Localization and Nuclear Export and Import Sequences for K17 and All IF Proteins (A and B) Indirect immunostaining for K17 (green) in representative single-plane confocal micrographs of (A) cultured human epithelial tumor cells derived from three tissue types (breast, BT-20 [left]; cervix, HeLa [middle]; vulva, A431 [right]), treated with 40 nM LMB (bottom row) or vehicle (Veh, 70% methanol) (top row), (legend continued on next page) 228 Developmental Cell 38, August 8, 2016 Developmental Cell Essay

indicate that nuclear keratins occur in the non-lamin IF proteins have been shown sequence, a highly conserved helix-ter- natural context of tumor cells in culture to associate with or localize to the nu- minating segment that corresponds to and tissues in situ. cleus in various experimental settings. residue Glycine 390 or Gluta- The idea of nuclear-localized keratins Early studies found that nuclear elements mate 391 in human K17. Interestingly, may not be so far-fetched. Is it not the (e.g., , matrix, DNA) therefore, the predicted NLS for K17 case, for instance, that other cytoskel- could physically associate with non-lamin spans the boundary between the a-heli- etal proteins previously thought to reside IF proteins added to cells or nuclear ex- cal, coiled-coil-forming rod domain, and function exclusively in the cytoplasm tracts as purified products in vitro, or after which represents the main driving force (e.g., b-, actin, , and even subjecting cells or tissues to harsh chal- toward assembly into 10-nm filaments motors) have later on been found lenges (e.g., Bastos et al., 1992; Djabali (Herrmann and Aebi, 2004), and the non- to both occur and fulfill important roles et al., 1991; Georgatos and Blobel, 1987; helical tail domain located at the C termi- in the nucleus (McCrea and Gottardi, Tolstonog et al., 2002; Ward et al., nus of K17 (Figure 1C) (Hobbs et al., 2016; Pellegrini and Budman, 2005; 1984). Nuclear-localized keratin proteins 2015). Such a location for a subcellular Philimonenko et al., 2004)? Certainly, the have also been seen in intact culture cell localization signal within the IF protein notion of IF proteins localizing in the models, e.g., when transiently expressing backbone is very intriguing. On one nucleus is not unprecedented, because tail-less truncation mutants of human K8, hand, the surface accessibility and func- the A-, B-, and C-type lamins are bona K18, or K19 (the latter being a natural tail- tionality of the bipartite NLS, the amino- fide nuclear proteins responsible for the less keratin) in murine fibroblasts (Bader terminal moiety of which is embedded in formation of a dense meshwork of 10-nm et al., 1991), when transiently expressing a coiled-coil-forming region of the rod, filaments covering the inner surface of the the epidermal-specific K1 in HeLa cells may well depend upon the polymerization nuclear envelope (Aebi et al., 1986), as or murine fibroblasts (Blessing et al., status of the protein. It is easy to envision well as a diffuse, less well-defined struc- 1993), or when human K18 or the temper- how either a cryptic or intrinsically ‘‘weak’’ ture in the nucleoplasm (Bridger et al., ature-sensitive Xenopus were NLS may be masked or ineffective in the 1993). In fact, IF proteins are believed to modified to include a ‘‘strong’’ NLS and setting of 10-nm filaments (or smaller- have initially appeared as nuclear-localized were transiently expressed in SW13 cells sized IF subunits) but be kept accessible elements in more ‘‘primitive’’ organisms (Herrmann et al., 1993; Reichenzeller when a given IF protein takes on an alter- (Peter and Stick, 2015), as inferred from et al., 2000). Finally, nuclear-localized native conformation. On the other hand, the notion that the two IF-encoding keratins have been observed in intact tis- the carboxy-terminal moiety of the K17 found in Drosophila melanogaster and the sues of transgenic mice engineered to NLS occurs in its tail domain, a preferred lone IF-encoding gene in Dictyostelium ectopically express human K1 and/or site of regulation of IF proteins via either discoidum encode lamin-like proteins that K10 in pancreatic islet cells (Blessing post-translational modifications (PTM) or localize to the nucleus (Erber et al., 1998; et al., 1993) or to overexpress human protein-protein interactions (Snider and Kru¨ ger et al., 2012). The loss of a small K16 in skin (Takahashi et al., 1994). A Omary, 2014). Large-scale proteomics- coding for a classical nuclear locali- key distinction between these early based surveys have identified K17 to zation signal (NLS) along with a mem- studies and the most recent findings harbor PTMs on 398 (phosphor- brane-targeting CAAX motif likely facili- regarding nuclear K17 resides in the lat- ylation), Lysine 399 (acetylation, ubiqui- tated the appearance of lamin-like IFs in ter’s assessment of endogenous protein tination), Lysine 400 (ubiquitination), the cytoplasm at some point during meta- and the assignment of a functional role and Tyrosine 410 (phosphorylation) (see zoan evolution (Peter and Stick, 2015). for a nuclear-localized keratin. http://www.phosphosite.org; Hornbeck The element of intrigue here is that at least In the case of K17 protein, current-day et al., 2015). As is the case for the conven- some, and perhaps many, of the remaining in silico sequence analysis algorithms tional polymeric state of K17 or any other 68 non-lamin polypeptides forming the readily identify a classical bipartite NLS IF protein (Snider and Omary, 2014), the modern IF superfamily may not exhibit beginning at Glutamate 380 and ending functionality of the NLS is likely subject an exclusively cytoplasmic localization in at Isoleucine 411 (Figure 1C). Based on to tight regulation by modifications and/ interphase cells, as previously thought. available crystal structure data for the or interactions with key partners in vivo. As part of early efforts geared toward K5/K14 heterodimer (Lee et al., 2012) The predicted bipartite NLS for human understanding the basic properties of and the vimentin homodimer (Chernya- K17 is highly conserved (given >50% IF proteins and the significance of their tina et al., 2012), the central rod domain sequence identity) in primates, rodents, tissue- and context-specific expression, ends with either the ‘‘LLEG’’ or ‘‘LLEGE’’ amphibians, and fish (Figure 1C). Of and (B) human biopsy of basal cell carcinoma in situ (image at right is a zoomed view of the dotted box in image at left). Nuclei are stained with DAPI (blue). All scale bars for the xy plane are 10 mm. Scale bars for the z planes are 5 mm. BT-20 insets in (A) highlight diffuse pattern for nuclear K17. Arrows denote nuclear punctae containing K17. For detailed methods on cell culture, tissue acquisition, immunostaining, and imaging, see Hobbs et al. (2015). (C) Schematic of K17 protein (head in red, rod in green, tail in blue) highlighting the conservation of NLS (above) and NES (below) motifs across orthologs. Conserved residues are highlighted (red-to-yellow color gradient reflects most to least conserved). Numbers refer to the human K17 primary amino acid sequence (NCBI reference sequence NP_000413.1). Bolded residues are experimentally validated functional elements. # denotes NES residues predicted with ValidNESs (Fu et al., 2013). * denotes NES residues predicted with NetNES1.1 (la Cour et al., 2004). Alignment of K17 orthologs to the human K17 sequence was done using the Molecular Evolutionary Genetics Analysis (MEGA) software tool. (D) Schematic of an IF protein (head in red, rod in green, tail in blue) highlighting the domain location for predicted monopartite (M) or bipartite (B) NLS sequences across the entire IF superfamily. ‘‘B’’ located on a dotted line indicates an NLS spanning the boundary between rod and tail domains. Classical importin- a-dependent NLS sequences were predicted using the cNLS Mapper online software (Kosugi et al., 2009).

Developmental Cell 38, August 8, 2016 229 Developmental Cell Essay

note, the di-lysine motif located in the tail NLS motifs, including the type I IFs K23, pase-mediated cleavage, e.g., in breast domain moiety of K17’s predicted NLS K32, and K35, the type II K4, the type III epithelial cells in culture (Badock et al., is required for the nuclear localization GFAP, peripherin, syncoilin, and, finally, 2001). Additionally, the number and size and function of both the human (Lysine the type IV a-internexin, NF-H, and of K17-containing nuclear punctae per 399, Lysine 400) and mouse (Lysine 399, NF-L. A recent study demonstrated the cell, as well as the fraction of cells Lysine 401) orthologs (Escobar-Hoyos presence of desmin in the nucleus of car- showing intra-nuclear keratins, vary de- et al., 2015; Hobbs et al., 2015). This motif diac stem cells, where it occurs at the pro- pending on the tumor cell line being appears to be specific to mammals, the moter of actively transcribed genes (just cultured. For instance, we find that significance of which is unclear. In addi- as is the case for K17; see Fuchs et al., following exposure to a relevant stimulus, tion, in silico analyses using ValidNESs 2016). Also, has been observed in significantly fewer A431 cells show nu- (Fu et al., 2013) or NetNES1.1 (la Cour the nucleus of neurogenic tumor cell lines clear K17 relative to HeLa and BT-20 et al., 2004) reveal that K17 orthologs (Krupkova et al., 2011). The type II IF cells, and the latter shows a diffuse rather feature a nuclear export signal (NES; see K8 represents an example of a ‘‘cyto- than punctate pattern for intra-nuclear Figure 1C) that is conserved in many other plasmic’’ IF protein known to occur in K17 (Figure 1A). Thus, it follows that type I keratins (unpublished data). This the nucleus without the apparent benefit various factors likely specify endogenous motif comprises three leucine residues— of a recognizable NLS (Kumeta et al., keratins against polymerization inside the Leucine 194, Leucine 197, and Leucine 2013). In this case—and likely others—it nucleus. Kumeta et al. (2013) elegantly 199—that were shown, via mutagenesis, could be that the remarkable propensity established that nuclear- and cytoplas- to markedly impact the nuclear localiza- of IF proteins to multimerize with compat- mically localized K8 in HeLa cells tion of K17 (Escobar-Hoyos et al., 2015). ible IF partners and/or interact with pro- show a differential reactivity with various For K17 protein, therefore, cis-acting de- teins that shuttle in and out of the nucleus monoclonal antibodies that recognize terminants affecting nuclear import and plays a role in their nuclear import. In all conformation-specific epitopes, along export have been identified in silico but a few of these cases, the nuclear with a differential sensitivity to detergent and experimentally verified, though the occurrence of the said IF protein and extraction. From this they inferred that partners, pathways, and regulatory mo- functionality of predicted NLS motifs the form adopted by keratin inside the nu- dalities involved await further investiga- have yet to be experimentally validated. cleus is fundamentally distinct from that tion. The presence of a classical NLS What form do keratins take while in occurring in the cytoplasm (10-nm IFs). does not exclude the possibility that K17 the nucleus? Prior studies showed that Other cytoskeletal proteins, including could also enter the nucleus via a non- mutated forms of IF proteins, such as actin (Schoenenberger et al., 2005) and classical mechanism. K17 interacts with tail-less human K8/K18 and K8/K19 tubulin (Akoumianaki et al., 2009), have a growing list of proteins that shuttle in and Xenopus NLS-vimentin, can occur been shown to adopt a distinct conforma- and out of the nucleus, including AIRE as conventional 10-nm filaments in addi- tion while inside the nucleus, correlating (Hobbs et al., 2015), 14-3-3s (Kim et al., tion to punctae while inside the nucleus with their association with a distinct set 2006), hnRNP K (Chung et al., 2015), (Bader et al., 1991; Herrmann et al., of protein partners. Going forward, identi- and CDKN1B/p27KIP1 (Escobar-Hoyos 1993; Reichenzeller et al., 2000), suggest- fying these regulatory factors (i.e., post- et al., 2015). Together with a classical ing that conventional polymerization in- translational modifications and/or unique NLS-based, importin-mediated mecha- side the nucleus is possible under spe- interacting protein partners) represents a nism, this potential for alternative nuclear cific circumstances. However, the three fundamentally important though techni- import as a piggyback element may allow recent reports discussed here are in full cally challenging aspect of understand- for greater adaptability and/or broader agreement in finding that endogenous ing the significance of nuclear-localized regulation, depending on physiological K17 occurs as discrete punctae and/or keratins. circumstances. in a diffuse pattern while in the nucleus What is the functional significance of In silico analyses of all other human IF of tumor epithelial cells (Kumeta et al., nuclear-localized keratins? The limited sequences reveal that a sizable number 2013; Hobbs et al., 2015; Escobar-Hoyos data available originate from two studies of non-lamin IFs exhibit one or even two et al., 2015). Neither study found nuclear focused on K17 in cervical (Escobar- monopartite or bipartite predicted NLS K17 to exist in conventional 10-nm fila- Hoyos et al., 2015) and skin (Hobbs motifs. In fact, 37 out of the 68 non-lamin ments, despite the observations that nu- et al., 2015) tumors and already hint at human IF sequences exhibit a predicted clear K17 can co-localize with the type II the prospect that this keratin may exert NLS (Figure 1D) from the outcome of an IF (K5), one of its natural assem- multiple roles in the nucleus, as it does in silico survey using NCBI consensus bly partners, in intra-nuclear punctae in the cytoplasm (Kim et al., 2006; McGo- coding sequences together with the (Hobbs et al., 2015) and can occur in its wan et al., 2002; Tong and Coulombe, cNLS Mapper freeware (Kosugi et al., full-length version, as can be inferred 2006). Though functionally important in 2009). The vast majority of these NLSs from the use of GFP fusion proteins and both settings, expression of K17 impacts occur either in the non-helical head of antibodies to known epitopes (Hobbs cervical and skin tumor epithelia in domain (e.g., K10, K6, desmin, GFAP, et al., 2015; Kumeta et al., 2013). At this distinct ways. In skin tumor , CP49) or in the tail domain (e.g., K25, time, one cannot comment on the addi- K17 plays a key role in promoting the cell- K78, GFAP, vimentin, NF-L, NF-H, tional possibility of cleaved K17 (or other autonomous expression of genes encod- NF-M). Akin to the lamin IFs, a few cyto- keratins) inside the nucleus, though K17 ing inflammatory and immune cytokine plasmic IF proteins feature two predicted (and other keratins) is (are) subject to cas- effectors (e.g., Cxcl5, Cxcl10, Cxcl11,

230 Developmental Cell 38, August 8, 2016 Developmental Cell Essay

Cytoplasm Nuclear importInside the nucleus Nuclear export (KIF assembly) (Forms and functions)

12 4 punctaeppu ctttaae 1 6 8 Proteasome

2 9 11 cell cycle (p27KIP1) 10 (Aire, p65) 13 7 diffuseuse 5 3

14

PTM for Importin and/or Ribosome Chromatin accessory protein targeting Exportin complex Keratin mRNA Accessory protein for nuclear Diffuse nuclear K17 Newly synthesized keratin translocation Ubiquitin chain Keratin dimer Importin complex Punctate nuclear K17 Proteasome Oligomeric keratin subunit complex PML Mature keratin filament

Figure 2. Model for Keratin Localization and Function in the Nucleus (1) Newly synthesized keratin proteins in the cytoplasm are (2) assembled into oligomeric subunits, which become either (3) incorporated into mature fully polymerized keratin filaments or (4) specified for nuclear targeting, likely by a post-translational modification (PTM). Appropriately specified keratin oligomers can then be imported into the nucleus by either (5) classical importin-a-mediated NLS recognition or (6) non-classical nuclear localization mechanisms. Once inside the nucleus, keratins may localize (7) in a diffuse pattern and/or (8) in small rounded punctae adjacent to PML bodies and in chromatin-sparse regions (Hobbs et al., 2015). The two known functions for nuclear K17 include roles in (9) cell-cycle regulation (via p27KIP1 association [Escobar-Hoyos et al., 2015]) and (10) transcription (via Aire and p65 association [Hobbs et al., 2015]). To be exported, (11) nuclear keratin oligomers are recognized by the exportin machinery and shuttled to the cytoplasm through the nuclear pore complex. Newly exported keratin oligomers may then be ‘‘re-specified’’ and (12) targeted for degradation, (13) become incorporated into mature fully polymerized keratin filaments, or (14) be targeted for another round of nuclear import.

Ccl2, Ccl19, Ifng, Mmp9). Nuclear K17 could be related to nuclear K17’s role in and export of K17 to and from the nucleus has been found to associate with pro- cell-cycle progression (Escobar-Hoyos and its significance as a nuclear protein is moter regions of cytokine genes, as well et al., 2015). Importantly, K17’s ability to presented in Figure 2. Its main points are as the transcriptional regulator AIRE and regulate immune and inflammatory cyto- as follows. First, we posit that IF subunits the p65 subunit of NF-kB, altogether indi- kine in skin tumor kerati- destined for nuclear import are likely cating roles in chromatin binding and nocytes, and CDKN1B/p27KIP1 activity to be small, and, whether they are newly transcription (Hobbs et al., 2015). Addi- and cell-cycle progression in cervical tu- synthesized elements or derived from tionally, K17 acts with the ribonucleo- mor keratinocytes, depends in part on its the existing (e.g., polymer-bound) pool protein hnRNP K to stabilize its target presence in the nucleus (Escobar-Hoyos of IFs, they have to be specified for cytokine transcripts in skin tumor kerati- et al., 2015; Hobbs et al., 2015). The differ- this fate via post-translational modifica- nocytes, suggesting a possible role for ential significance of nuclear K17 in skin tions (e.g., phosphorylation, sumoylation, K17 in RNA export and/or processing versus cervical tumor keratinocytes likely acetylation) and/or interactions with (Chung et al., 2015). In cervical tumor reflects distinct sites of expression for accessory proteins. Second, import of epithelia, K17 has been shown to pro- K17 in these two tissues in relation to non-lamin IF proteins into the nucleus mote the nuclear export and deactiva- the origin of tumor cells in the paradigms can occur via classical NLS/importin- tion of CDKN1B/p27KIP1, a key negative exploited (Hobbs et al., 2016). High- mediated and/or non-canonical pathways regulator of the G1 to S transition (Esco- throughput screens focused on the K17- (Wagstaff and Jans, 2009). Third, nuclear- bar-Hoyos et al., 2015). Furthermore, in associated genome, transcriptome, and localized non-lamin IF proteins likely cervical lesions in situ, K17 acts to sup- nuclear proteome are examples of next partake in multiple processes in the nu- press inflammation and immune cell steps likely to shed light on the ultimate cleus: so far, there is evidence for partici- recruitment, in addition to impacting p63 significance of nuclear-localized K17 pation of keratins in transcription and cell- expression and the balance between protein. cycle regulation, depending on the tissue epithelial cell proliferation and differentia- A model that incorporates what we context and tumor paradigm. Fourth, tion (Hobbs et al., 2016), a notion that have learned so far regarding the import export of non-lamin IF proteins from

Developmental Cell 38, August 8, 2016 231 Developmental Cell Essay

the nucleus can occur via a canonical, Bastos, R., Engel, P., Pujades, C., Falchetto, R., Hobbs, R.P., DePianto, D.J., Jacob, J.T., Han, NES-mediated path via direct associa- Aligue´ , R., and Bachs, O. (1992). Increase of cyto- M.C., Chung, B.M., Batazzi, A.S., Poll, B.G., keratin D during liver regeneration: association Guo, Y., Han, J., Ong, S., et al. (2015). Keratin- tion with exportins or indirectly via piggy- with the nuclear matrix. Hepatology 16, 1434– dependent regulation of Aire and gene expression backing, and/or that loss could also entail 1446. in skin tumor keratinocytes. Nat. Genet. 47, 933–938. degradation in the nucleus (von Mikecz, Blessing, M., Ru¨ ther, U., and Franke, W.W. (1993). 2006). Ectopic synthesis of epidermal in Hobbs, R.P., Batazzi, A.S., Han, M.C., and In conclusion, endogenous keratins pancreatic islet cells of transgenic mice interferes Coulombe, P.A. (2016). Loss of induces with cytoskeletal order and insulin production. tissue-specific cytokine polarization and cellular have now been found to occur and func- J. Cell Biol. 120, 743–755. differentiation in HPV16-driven cervical tumorigen- tion inside the nucleus. The available evi- esis in vivo. Oncogene. http://dx.doi.org/10.1038/ dence suggests that the nuclear pool of Bridger, J.M., Kill, I.R., O’Farrell, M., and Hutchi- onc.2016.102. son, C.J. (1993). Internal lamin structures within keratin proteins is dynamically regulated G1 nuclei of human dermal fibroblasts. J. Cell Hornbeck, P.V., Zhang, B., Murray, B., Korn- and can exist in conformations that Sci. 104, 297–306. hauser, J.M., Latham, V., and Skrzypek, E. are distinct from canonical cytoplasmic (2015). PhosphoSitePlus, 2014: mutations, PTMs Chernyatina, A.A., Nicolet, S., Aebi, U., Herrmann, and recalibrations. Nucleic Acids Res. 43, D512– 10-nm IFs. In the case of K17, classically H., and Strelkov, S.V. (2012). Atomic structure of D520. defined cis-acting determinants medi- the vimentin central a-helical domain and its impli- cations for intermediate filament assembly. Proc. Islas, S., Vega, J., Ponce, L., and Gonza´ lez-Maris- ating nuclear import and export have Natl. Acad. Sci. USA 109, 13620–13625. cal, L. (2002). Nuclear localization of the tight junc- been elucidated, and specific leads tion protein ZO-2 in epithelial cells. Exp. Cell Res. regarding functional significance secured Chung, B.M., Arutyunov, A., Ilagan, E., Yao, N., 274, 138–148. Wills-Karp, M., and Coulombe, P.A. (2015). Regu- from studies focused on skin and cervical lation of C-X-C chemokine gene expression by ker- Kim, S., Wong, P., and Coulombe, P.A. (2006). A tumors indicate active roles in regulating atin 17 and hnRNP K in skin tumor keratinocytes. keratin cytoskeletal protein regulates protein syn- gene expression and cell cycle. These J. Cell Biol. 208, 613–627. thesis and epithelial cell growth. Nature 441, 362–365. developments add to the notion that Djabali, K., Portier, M.M., Gros, F., Blobel, G., and cytoskeletal proteins play a fundamental Georgatos, S.D. (1991). Network antibodies iden- Kosugi, S., Hasebe, M., Tomita, M., and Yana- role in facilitating dynamic and tightly tify nuclear lamin B as a physiological attachment gawa, H. (2009). Systematic identification of cell site for peripherin intermediate filaments. Cell 64, cycle-dependent yeast nucleocytoplasmic shut- regulated signaling processes and create 109–121. tling proteins by prediction of composite motifs. a new dimension to the challenge of Proc. Natl. Acad. Sci. USA 106, 10171–10176. understanding the diversity and context- Erber, A., Riemer, D., Bovenschulte, M., and Weber, K. (1998). Molecular phylogeny of meta- Kru¨ ger, A., Batsios, P., Baumann, O., Luckert, E., dependent regulation of keratins and all zoan intermediate filament proteins. J. Mol. Evol. Schwarz, H., Stick, R., Meyer, I., and Gra¨ f, R. other IF proteins, along with the signifi- 47, 751–762. (2012). Characterization of NE81, the first lamin- like nucleoskeleton protein in a unicellular organ- cance of their dysregulation in chronic Escobar-Hoyos, L.F., Shah, R., Roa-Pen˜ a, L., ism. Mol. Biol. Cell 23, 360–370. diseases such as cancer. Vanner, E.A., Najafian, N., Banach, A., Nielsen, E., Al-Khalil, R., Akalin, A., Talmage, D., and Krupkova, O., Jr., Loja, T., Redova, M., Neradil, J., Shroyer, K.R. (2015). Keratin-17 promotes Zitterbart, K., Sterba, J., and Veselska, R. (2011). Analysis of nuclear nestin localization in cell lines ACKNOWLEDGMENTS p27KIP1 nuclear export and degradation and offers potential prognostic utility. Cancer Res. 75, derived from neurogenic tumors. Tumour Biol. 32, 3650–3662. 631–639. The authors are grateful to laboratory and depart- mental colleagues for stimulating discussions and Fu, S.C., Huang, H.C., Horton, P., and Juan, H.F. Kumeta, M., Hirai, Y., Yoshimura, S.H., Horigome, to Drs. Carole Parent and Michael Matunis for (2013). ValidNESs: a database of validated T., and Takeyasu, K. (2013). Antibody-based anal- reading the manuscript. Relevant efforts in the leucine-rich nuclear export signals. Nucleic Acids ysis reveals ‘‘filamentous vs. non-filamentous’’ and Coulombe laboratory are supported by grants Res. 41, D338–D343. ‘‘cytoplasmic vs. nuclear’’ crosstalk of cytoskeletal AR044232 and CA160255 from the National Insti- proteins. Exp. Cell Res. 319, 3226–3237. tutes of Health. Fuchs, C., Gawlas, S., Heher, P., Nikouli, S., Paar, H., Ivankovic, M., Schultheis, M., Klammer, J., la Cour, T., Kiemer, L., Mølgaard, A., Gupta, R., Gottschamel, T., Capetanaki, Y., and Weitzer, G. Skriver, K., and Brunak, S. (2004). Analysis and (2016). Desmin enters the nucleus of cardiac prediction of leucine-rich nuclear export signals. REFERENCES stem cells and modulates Nkx2.5 expression by Protein Eng. Des. Sel. 17, 527–536. participating in transcription factor complexes Aebi, U., Cohn, J., Buhle, L., and Gerace, L. (1986). that interact with the nkx2.5 gene. Biol. Open 5, Lee, C.H., Kim, M.S., Chung, B.M., Leahy, D.J., The is a meshwork of intermediate- 140–153. and Coulombe, P.A. (2012). Structural basis for type filaments. Nature 323, 560–564. heteromeric assembly and perinuclear organiza- Georgatos, S.D., and Blobel, G. (1987). Two tion of keratin filaments. Nat. Struct. Mol. Biol. 19, Akoumianaki, T., Kardassis, D., Polioudaki, H., distinct attachment sites for vimentin along the 707–715. Georgatos, S.D., and Theodoropoulos, P.A. plasma membrane and the nuclear envelope in (2009). Nucleocytoplasmic shuttling of soluble avian erythrocytes: a basis for a vectorial assem- McCrea, P.D., and Gottardi, C.J. (2016). Beyond tubulin in mammalian cells. J. Cell Sci. 122, bly of intermediate filaments. J. Cell Biol. 105, b-catenin: prospects for a larger catenin network 1111–1118. 105–115. in the nucleus. Nat. Rev. Mol. Cell Biol. 17, 55–64.

Bader, B.L., Magin, T.M., Freudenmann, M., Herrmann, H., and Aebi, U. (2004). Intermediate McGowan, K.M., Tong, X., Colucci-Guyon, E., Stumpp, S., and Franke, W.W. (1991). Intermediate filaments: molecular structure, assembly mecha- Langa, F., Babinet, C., and Coulombe, P.A. filaments formed de novo from tail-less cytokera- nism, and integration into functionally distinct (2002). Keratin 17 null mice exhibit age- and tins in the cytoplasm and in the nucleus. J. Cell intracellular Scaffolds. Annu. Rev. Biochem. 73, strain-dependent alopecia. Genes Dev. 16, 1412– Biol. 115, 1293–1307. 749–789. 1422.

Badock, V., Steinhusen, U., Bommert, K., Witt- Herrmann, H., Eckelt, A., Brettel, M., Grund, C., Nix, D.A., Fradelizi, J., Bockholt, S., Menichi, B., mann-Liebold, B., and Otto, A. (2001). - and Franke, W.W. (1993). Temperature-sensitive Louvard, D., Friederich, E., and Beckerle, M.C. induced cleavage of and keratin 17 in a intermediate filament assembly. Alternative struc- (2001). Targeting of zyxin to sites of actin mem- human breast epithelial cell line. Cell Death Differ. tures of Xenopus laevis vimentin in vitro and in vivo. brane interaction and to the nucleus. J. Biol. 8, 308–315. J. Mol. Biol. 234, 99–113. Chem. 276, 34759–34767.

232 Developmental Cell 38, August 8, 2016 Developmental Cell Essay

Omary, M.B., Coulombe, P.A., and McLean, W.H. Reichenzeller, M., Burzlaff, A., Lichter, P., and Toivola, D.M., Strnad, P., Habtezion, A., and (2004). Intermediate filament proteins and their Herrmann, H. (2000). In vivo observation of a nu- Omary, M.B. (2010). Intermediate filaments take associated diseases. N Engl J Med. 351, 2087– clear channel-like system: evidence for a distinct the heat as stress proteins. Trends Cell Biol. 20, 2100. interchromosomal domain compartment in inter- 79–91. phase cells. J. Struct. Biol. 129, 175–185. Pallari, H.M., and Eriksson, J.E. (2006). Intermedi- Tolstonog, G.V., Sabasch, M., and Traub, P. ate filaments as signaling platforms. Sci. STKE Schoenenberger, C.A., Buchmeier, S., Boerries, (2002). Cytoplasmic intermediate filaments are 2006, pe53. M., Su¨ tterlin, R., Aebi, U., and Jockusch, B.M. stably associated with nuclear matrices and poten- (2005). Conformation-specific antibodies reveal tially modulate their DNA-binding function. DNA Pan, X., Hobbs, R.P., and Coulombe, P.A. (2013). distinct actin structures in the nucleus and the Cell Biol. 21, 213–239. The expanding significance of keratin intermediate cytoplasm. J. Struct. Biol. 152, 157–168. filaments in normal and diseased epithelia. Curr. Opin. Cell Biol. 25, 47–56. Schweizer, J., Bowden, P.E., Coulombe, P.A., Tong, X., and Coulombe, P.A. (2006). Keratin 17 Langbein, L., Lane, E.B., Magin, T.M., Maltais, L., modulates hair follicle cycling in a TNFalpha- Pellegrini, F., and Budman, D.R. (2005). Review: Omary, M.B., Parry, D.A., Rogers, M.A., and dependent fashion. Genes Dev. 20, 1353–1364. tubulin function, action of antitubulin drugs, Wright, M.W. (2006). New consensus nomencla- and new drug development. Cancer Invest. 23, ture for mammalian keratins. J. Cell Biol. 174, 264–273. Velichkova, M., and Hasson, T. (2005). Keap1 reg- 169–174. ulates the oxidation-sensitive shuttling of Nrf2 into Peter, A., and Stick, R. (2015). Evolutionary and out of the nucleus via a Crm1-dependent Snider, N.T., and Omary, M.B. (2014). Post- aspects in intermediate filament proteins. Curr. nuclear export mechanism. Mol. Cell. Biol. 25, translational modifications of intermediate filament Opin. Cell Biol. 32, 48–55. 4501–4513. proteins: mechanisms and functions. Nat. Rev. Petit, M.M., Fradelizi, J., Golsteyn, R.M., Ayoubi, Mol. Cell Biol. 15, 163–177. T.A., Menichi, B., Louvard, D., Van de Ven, W.J., von Mikecz, A. (2006). The nuclear ubiquitin-pro- and Friederich, E. (2000). LPP, an actin cytoskel- Szeverenyi, I., Cassidy, A.J., Chung, C.W., Lee, teasome system. J. Cell Sci. 119, 1977–1984. eton protein related to zyxin, harbors a nuclear B.T., Common, J.E., Ogg, S.C., Chen, H., Sim, S.Y., Goh, W.L., Ng, K.W, et al. (2008). The human export signal and transcriptional activation capac- Wagstaff, K.M., and Jans, D.A. (2009). Importins intermediate filament database: comprehensive ity. Mol. Biol. Cell 11, 117–129. and beyond: non-conventional nuclear transport information on a gene family involved in many hu- mechanisms. Traffic 10, 1188–1198. Philimonenko, V.V., Zhao, J., Iben, S., Dingova´ , H., man diseases. Hum Mutat. 29, 351–360. Kysela´ , K., Kahle, M., Zentgraf, H., Hofmann, W.A., de Lanerolle, P., Hoza´ k, P., and Grummt, I. (2004). Takahashi, K., Folmer, J., and Coulombe, P.A. Ward, W.S., Schmidt, W.N., Schmidt, C.A., and Nuclear actin and myosin I are required for (1994). Increased expression of causes Hnilica, L.S. (1984). Association of RNA polymerase I transcription. Nat. Cell Biol. 6, anomalies in cytoarchitecture and keratinization in p39 with DNA in intact Novikoff hepatoma cells. 1165–1172. transgenic mouse skin. J. Cell Biol. 127, 505–520. Proc. Natl. Acad. Sci. USA 81, 419–423.

Developmental Cell 38, August 8, 2016 233