|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
First published online March 21, 2007
doi: 10.1242/10.1242/jcs.005348
Correspondence |
1 Molecular Oncology Unit, 22 E-28040 Madrid, Spain
2 Epithelial Biomedicine Division CIEMAT. Av. Complutense, 22 E-28040 Madrid, Spain
* Author for correspondence (e-mail: jesusm.paramio{at}ciemat.es)
The existence of differential keratin protein functions is a matter of controversy. Keratin genes display a differential regulation in the various epithelia of the body and inherited mutations affecting keratin genes are responsible for a variety of epithelial fragility syndromes. In recent years, in addition to this shared family-wide function, new findings have suggested that keratins might play cell-type-dependent and context-dependent functions that include protection against metabolic stress, modulating apoptotic signals, cell cycle progression, and promotion of specific epithelial cytoarchitecture (Koch and Roop, 2004
; Paramio and Jorcano, 2002). However, despite the recent progress in this field, our understanding of the relationship between keratin proteins and the differentiation of epithelial cells remains poor. In their recent paper reporting the generation of K10K14 chimeric mice (Chen et al., 2006), Koch and coworkers present several observations that are hard to reconcile with previous work.
Our group has been actively investigating the possible existence of such specific keratin functions focusing on keratin K10. This protein is expressed in differentiating postmitotic epidermal keratinocytes, and its expression is severely reduced and even absent in conditions of hyperproliferation, including skin tumors. Our work in cultured cells has shown that the expression of this keratin inhibits cell proliferation in a pRb-dependent process through the inhibition of Akt and PKC
kinases (Paramio et al., 1999; Paramio et al., 2001). More recently, we extended these studies to in vivo situations through the generation of transgenic mice in which human keratin K10 gene expression was targeted to the basal layer of the epidermis (bK5hK10 transgenic mice) (Santos et al., 2002). These animals displayed epidermal hypoplasia and reduced susceptibility to tumor formation upon chemical skin carcinogenesis protocols (Santos et al., 2002). These data suggested that keratin K10 might control proliferation in cells committed in the differentiation program. This hypothesis seemed to be initially confirmed by the epidermal hyperplasia displayed by adult keratin K10-null mice (Reichelt and Magin, 2002). However, this was not the case because in these animals hyperproliferation was unexpectedly confined to basal cells, indicating that K10, a suprabasal keratin, can mediate effects in the neighboring basal cells (Reichelt and Magin, 2002). However, the mechanism involving the signaling between suprabasal and basal keratinocytes remains unknown. A similar juxtacrine signaling was found in the thymus of bK5K10 transgenic mice, affecting T-cell differentiation through Notch interference (Santos et al., 2005). Also surprisingly, Magin's group found that in K10 knockout mice, in spite of hyperproliferation and increased Myc and CycD1 expression, there is no change in pRb phosphorylation (Reichelt and Magin, 2002). Moreover, there are no changes in Akt activity, in apparent disagreement with our previous data in vitro and in vivo. In this regard, the possibility that keratins can play a role in the regulation of mTOR/Akt signaling has been reinforced by the findings of Coulombe's group who recently reported that in K17 knockout mice there is a reduced activity of this pathway, in part mediated by the altered distribution of 14-3-3
between the cytoplasm and nucleus (Kim et al., 2006). Of note, this protein, also called stratifin, is overexpressed in K10 knockout mouse epidermis (Reichelt and Magin, 2002). Our work also suggested that the possible functions of K10 might reside in the non-helical domains of K10 based on the following data: (1) a protein lacking head and tail causes no effects in transfected cells; (2) increased expression of aminoterminus competes with and reverts the effects of K10; and (3) two-hybrid experiments showed interaction between head domain and Akt (Paramio et al., 2001). In this regard, the activity of K17 modulating mTOR/Akt signaling also seems to be dependent on the amino terminus (Kim et al., 2006).
The work by Chen et al. (Chen et al., 2006), helps to give new light on these aspects. They generated K10K14 chimeric mice in which the head and tail of K14 have been replaced by those of K10, thus encoding a chimeric protein K10K14 consisting of the K14 rod fused to the K10 head and tail domains. Unexpectedly, these mice develop no apparent gross phenotype. Their keratinocytes do not display altered proliferation, as our results would predict, nor altered cell migration as Magin's group indicated. However, when they are challenged by two-stage skin carcinogenesis protocols they develop an increased number of tumors that also appear earlier than in controls. This observation points to altered apoptosis. In agreement, reduced apoptotic susceptibility was observed in these mutant mice when they are challenged by TNF
treatment in vitro. Of note, we also observed increased production of TNF and an altered response to this cytokine in our transgenic mice (Santos et al., 2003). However, probably owing to reduced NF
B signaling, the bK5HK10 transgenic keratinocytes display several ultrastructural alterations characteristic of increased apoptosis (Santos et al., 2003).
Putting together all these observations is not an easy task. One is always tempted to ascribe the differences to the levels of transgenic protein expression, genetic backgrounds (enriched C57Bl in K10K14 chimeric mice, Balb/c in K10 knockout mice or DBA/C57 hybrids in bK5hK10 transgenics) and different skin tumorigenesis protocols (different DMBA/TPA doses or crosses with TgAC mice), which are also affected by the genetic background. Obviously, these are important aspects to take into consideration and that might vary the results.
One might expect that in K10K14 chimeric mice both Akt and MAPK pathways can be altered, at least upon Ras mutations in carcinogenesis experiments, to confer the observed increased tumor susceptibility (Segrelles et al., 2002). However, one would expect increased rather than reduced activity to explain the augmented tumor susceptibility (Segrelles et al., 2002). How can the differences between the two mouse models be explained? Arguing in favor of possible effects of transgenic protein level is the fact that the phenotype in the bK5hK10 transgenic model is dose dependent, although only a moderate increase in expression of transgenic K10 was observed compared with endogenous keratins (Santos et al., 2002). Another possibility is the proposal that the composition of the keratin cytoskeleton might affect its activities. The basal cell layer cytoskeleton in bK5hK10 transgenic mice is composed of K5, K14 and transgenic K10, whereas in K10K14 chimeric mice there are only K5 and K14/K10 hybrids. If we assume that the non-helical ends protrude from the core of the filaments, the binding abilities of K10 for possible targets (i.e. Akt and other signaling molecules) can be altered by the presence of those ends protruding from K14, which are absent in K10K14 chimeric mice. On the other hand, we also have to consider that the rod domain is not neutral regarding the ability to bind and sequester apoptotic signaling molecules. In this regard, both K14 and K17 rods have been shown to interact with TNF receptor 1 (TNFR1)-associated death domain protein (TRADD), a death adaptor essential for TNFR1-dependent signal relay (Tong and Coulombe, 2006; Yoneda et al., 2004).
Along the same line of evidence, even considering the binding of molecules to the protruding ends, the rod domains can also affect the functionality of the keratin filaments. Our previous data have indicated that K10 ends interact with and sequester relevant signaling targets to keratin filaments, precluding their translocation process and consequent activation. This implies that the dynamic behavior of the keratin cytoskeleton can affect keratin function as it can modulate the localization of bound signaling molecules. In this regard it is worth mentioning that, in cells expressing K14 and K10, these two proteins display a very different, or even opposite, dynamic behavior (Paramio et al., 1997). Although, this is in part controlled by phosphorylation (Paramio, 1999), the main player in this dynamics is the ability to form soluble precursors, i.e. tetramers, in which the rod domain exerts dominant effects for pairing (Paramio et al., 1997). Consequently, we can speculate that the presence of the K14 rod domain alters the function of factors bound to K10 end domain in two ways: (1) because of proteins that specifically bind to it and (2) modulating the ability of K10 end domain-bound proteins to change their localization. Another aspect that might contribute to the unexpected phenotype of the K10K14 chimeric mice might depend on the cell context. We have previously shown that K10 causes a cell cycle arrest only if the cells bear a functional pRb pathway (Paramio et al., 1999). The BrdU incorporation observed in the K10K14 chimeric mouse epidermis, at least in hair follicles (from where most epidermal cells are derived) suggests that in these cells pRb function is decreased or inactivated. Although the molecular mechanisms underlying this alteration in K10K14 chimeric mice are not known, they could be related to protein levels and/or be a consequence of the time point of expression of the transgene (i.e. during embryogenesis, in cells and at times where K10 is not normally expressed). If this reasoning is correct, this pRb inactivation in K10K14 chimeric mice, according to our data, would predict that the cell cycle blocking effect of K10 does not take place, or is diminished in these mice. If this is the case, it is worth mentioning that, in mouse bearing the epidermal-specific depletion of Rb1 gene (Ruiz et al., 2004), the expression of bK5hK10 transgene does not cause cell cycle arrest, but increased proliferation (M.S. and J.M.P., unpublished results), which even exceeds that displayed by Rb-deficient mouse skin keratinocytes (Ruiz et al., 2004). Therefore, the functional consequences of expressing K10 ends could depend on the cellular contest and under different settings different functions can emerge.
In the same context, it is worth considering the possible non cellautonomous effects. Indeed, the absence of K10 in knockout mice affects MAPK signaling (Reichelt et al., 2004) in neighboring basal cells, whereas in the epithelial cells of the thymus the expression of the bK5hK10 transgene leads to the modulation of Notch signaling in adjacent T lymphocytes (Santos et al., 2005). Interestingly, both pathways are involved in epidermal differentiation and can also modulate skin tumor susceptibility (Koster et al., 2002; Koster and Roop, 2004; Lefort and Dotto, 2004).
Unfortunately, Chen et al. (Chen et al., 2006) do not provide biochemical data in their report that might help to define the existence of altered signaling and/or to discriminate among the different possibilities, but certainly the K10K14 chimeric mice will become essential tools in the endeavor to solve the problem of keratin end functions in cell physiology.
References
Chen, J., Cheng, X., Merched-Sauvage, M., Caulin, C., Roop, D. R. and Koch, P. J. (2006). An unexpected role for keratin 10 end domains in susceptibility to skin cancer. J. Cell Sci. 119, 5067-5076.
Kim, S., Wong, P. and Coulombe, P. A. (2006). A keratin cytoskeletal protein regulates protein synthesis and epithelial cell growth. Nature 441, 362-365.[CrossRef][Medline]
Koch, P. J. and Roop, D. R. (2004). The role of keratins in epidermal development and homeostasis - going beyond the obvious. J. Invest. Dermatol. 123, x-xi.[Medline]
Koster, M. I. and Roop, D. R. (2004). Genetic pathways required for epidermal morphogenesis. Eur. J. Cell Biol. 83, 625-629.[CrossRef][Medline]
Koster, M. I., Huntzinger, K. A. and Roop, D. R. (2002). Epidermal differentiation: transgenic/knockout mouse models reveal genes involved in stem cell fate decisions and commitment to differentiation. J. Investig. Dermatol. Symp. Proc. 7, 41-45.[CrossRef][Medline]
Lefort, K. and Dotto, G. P. (2004). Notch signaling in the integrated control of keratinocyte growth/differentiation and tumor suppression. Semin. Cancer Biol. 14, 374-386.[CrossRef][Medline]
Paramio, J. M. (1999). A role for phosphorylation in the dynamics of keratin intermediate filaments. Eur. J. Cell Biol. 78, 33-43.[Medline]
Paramio, J. M. and Jorcano, J. L. (2002). Beyond structure: do intermediate filaments modulate cell signalling? BioEssays 24, 836-844.[CrossRef][Medline]
Paramio, J. M., Casanova, M. L., Alonso, A. and Jorcano, J. L. (1997). Keratin intermediate filament dynamics in cell heterokaryons reveals diverse behaviour of different keratins. J. Cell Sci. 110, 1099-1111.[Abstract]
Paramio, J. M., Casanova, M. L., Segrelles, C., Mittnacht, S., Lane, E. B. and Jorcano, J. L. (1999). Modulation of cell proliferation by cytokeratins K10 and K16. Mol. Cell. Biol. 19, 3086-3094.
Paramio, J. M., Segrelles, C., Ruiz, S. and Jorcano, J. L. (2001). Inhibition of protein kinase B (PKB) and PKCzeta mediates keratin K10-induced cell cycle arrest. Mol. Cell. Biol. 21, 7449-7459.
Reichelt, J. and Magin, T. M. (2002). Hyperproliferation, induction of c-Myc and 14-3-3sigma, but no cell fragility in keratin-10-null mice. J. Cell Sci. 115, 2639-2650.
Reichelt, J., Furstenberger, G. and Magin, T. M. (2004). Loss of keratin 10 leads to mitogen-activated protein kinase (MAPK) activation, increased keratinocyte turnover, and decreased tumor formation in mice. J. Invest. Dermatol. 123, 973-981.[CrossRef][Medline]
Ruiz, S., Santos, M., Segrelles, C., Leis, H., Jorcano, J. L., Berns, A., Paramio, J. M. and Vooijs, M. (2004). Unique and overlapping functions of pRb and p107 in the control of proliferation and differentiation in epidermis. Development 131, 2737-2748.
Santos, M., Paramio, J. M., Bravo, A., Ramirez, A. and Jorcano, J. L. (2002). The expression of keratin k10 in the basal layer of the epidermis inhibits cell proliferation and prevents skin tumorigenesis. J. Biol. Chem. 277, 19122-19130.
Santos, M., Perez, P., Segrelles, C., Ruiz, S., Jorcano, J. L. and Paramio, J. M. (2003). Impaired NF-kappa B activation and increased production of tumor necrosis factor alpha in transgenic mice expressing keratin K10 in the basal layer of the epidermis. J. Biol. Chem. 278, 13422-13430.
Santos, M., Rio, P., Ruiz, S., Martinez-Palacio, J., Segrelles, C., Lara, M. F., Segovia, J. C. and Paramio, J. M. (2005). Altered T cell differentiation and Notch signaling induced by the ectopic expression of keratin K10 in the epithelial cells of the thymus. J. Cell. Biochem. 95, 543-558.[CrossRef][Medline]
Segrelles, C., Ruiz, S., Perez, P., Murga, C., Santos, M., Budunova, I. V., Martinez, J., Larcher, F., Slaga, T. J., Gutkind, J. S. et al. (2002). Functional roles of Akt signaling in mouse skin tumorigenesis. Oncogene 21, 53-64.[CrossRef][Medline]
Tong, X. and Coulombe, P. A. (2006). Keratin 17 modulates hair follicle cycling in a TNFalpha-dependent fashion. Genes Dev. 20, 1353-1364.
Yoneda, K., Furukawa, T., Zheng, Y. J., Momoi, T., Izawa, I., Inagaki, M., Manabe, M. and Inagaki, N. (2004). An autocrine/paracrine loop linking keratin 14 aggregates to tumor necrosis factor alpha-mediated cytotoxicity in a keratinocyte model of epidermolysis bullosa simplex. J. Biol. Chem. 279, 7296-7303.
This article has been cited by other articles:
|
|
 |
|
  J. Chen, X. Cheng, M. Merched-Sauvage, C. Caulin, D. R. Roop, and P. J. Koch Reply to: The ends of a conundrum? J. Cell Sci., April 1, 2007; 120(7): 1147 - 1148. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
