| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 19, 2004 as doi:10.1096/fj.04-1683fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Epithelial Damage, Repair and Tissue Engineering, CIEMAT, Madrid, Spain;
Department of Animal Pathology, Veterinary School. University of Santiago de Compostela, Lugo, Spain;
Hospital General de Móstoles, Madrid, Spain; and
University of Texas M. D. Anderson Cancer Center Science Park, Smithville, Texas 78957, USA
1 Correspondence: Epithelial Damage, Repair and Tissue Engineering, CIEMAT, Avenida Complutense 22, 28040 Madrid, Spain. E-mail: llanos.casanova{at}ciemat.es
SPECIFIC AIMS
Although ectopic expression of keratin 8 in human and mouse tumors of different origin has been described, it has not been discerned whether this aberrant expression is the cause or the consequence of the development of the cancerous state. Aiming to answer this question, we have generated a mouse model of ectopic expression of the keratin 8 in epidermis and hair follicles. Our model allowed us to study the carcinogenesis response of epidermal cells "endowed" with the preexistence of K8 to determine a putative causal role of this keratin in malignancy. Our studies aim to assess the influence of keratin 8 aberrant expression both in normal and transformed skin.
PRINCIPAL FINDINGS
1. Expression of the HK8 transgene in mouse epidermis
We analyzed two lines of transgenic mice (TGHK8-4 and TGHK8-8) containing a 12 kb genomic fragment of the human K8 locus. Although the transgene showed mostly tissue-specific expression (i.e., in simple epithelial cell type), these lines (TGHK8 or transgenic mice) also express K8 in the skin, as determined by Northern blot. Immunohistochemical analyses of newborn mice skin showed transgenic keratin confined mainly to the basal layer of the epidermis and the outer root sheath (ORS) of hair follicles (Fig. 1
B) of TGHK8 mice. In a subset of transgenic mice, however, interfollicular K8 expression was progressively lost as they grew and was restricted to hair follicles and interfollicular epidermis adjacent to areas of hair emergence.
|
We next examined whether HK8 expression affects the endogenous cytoskeleton of keratins of epidermal cells (Fig. 1C-F
) and normal physiology of the skin (Fig. 1I-N'
). The expression pattern of HK8 in the skin of transgenic mice (Fig. 1B
) paralleled that of K5 in wild-type animals (Fig. 1C
). Expression of HK8 in TGHK8 mice partially replaced that of K5 in the bulb of hair follicles (Fig. 1B-D
). Mouse K14 (mK14), the normal K5 partner, did not differ between wild-type and transgenic skin, as revealed by immunostaining (Fig. 1E, F
). Double immunofluorescence analysis showed that HK8 in basal layer keratinocytes appeared to colocalize with mK14(Fig. 1G, H
). Similar results were obtained using simultaneous immunohistochemistry staining. Since K14 is the only type I keratin expressed in the basal layer of interfollicular epidermis, a K8/K14 pair is the most plausible filament composition. HK8 appears in a filamentous state since we did not detect keratin aggregates. This was confirmed by TEM studies showing an absence of aggregates or dots in the transgenic epidermal cells and the presence of normal filamentous keratin throughout the cytoplasm.
2. Epidermal hyperplasia and dysplastic hair follicles in TGHK8 transgenic mice
Except for a delay in hair follicle development, histological studies of neonatal skin showed no major differences between TGHK8 transgenic mice and their wild-type littermates. This difference disappeared when the mice reached 7–10 days of age, when a few dysplastic hair follicles were noticeable (Fig. 1J
) and scattered areas of epidermal hyperplasia (involving an expansion of spinous and granular layers) were seen in regions of dysplastic hair emergence (Fig. 1J
). At 22 days of age, transgenic mice showed a severe skin phenotype. Epidermal hyperplasia affected extensive areas of skin and was associated with an increased number of dysplastic hair follicles with twisted and irregular contours that were sometimes encysted and showed a lack of differentiation between inner and outer root sheaths (Fig. 1K, L
). Worsening of the skin phenotype was clear in 31 day-old and older TGHK8 mice in which extensive areas of severe epidermal hyperplasia associated with orthokeratotic hyperkeratosis (Fig. 1M, N
) and eosinophilic focal points of parakeratosis (Fig. 1
, inset N’) were evident. TGHK8 mice showed an important delay in the hair cycle in areas of hyperplastic epidermis, where hair follicles remained in the telogen phase instead of proceeding through the anagen phase of the second hair cycle (Fig. 1M, N
). Although the severity of the phenotypic features varied among different animals of the same line, progressive worsening was always seen with aging. No alterations in desmosomes or other epithelial components were found in the TGHK8 epidermis examined by TEM procedures.
3. Terminal differentiation is altered in the transgenic mouse epidermis
Immunohistological analysis of the expression of early and late terminal differentiation markers (including K10, filaggrin, loricrin, and involucrin, as well as K6 expression) in wild-type and TGHK8 mice revealed that the terminal differentiation program is altered in keratinocytes expressing HK8. The most striking aberrations were found in dysplastic hair follicles. No differences were found in the proliferation rate measured as bromodeoxiuridine incorporation.
4. HK8 expression in the skin of transgenic mice induces preneoplastic transformation of epidermis in aging mice
In aging transgenic mice (1 year and younger, Fig. 2
B), the epidermal hyperplasia evolved to preneoplastic changes. Figure 2
shows regions of pseudocarcinomatous hyperplasia (Fig. 2B
) or areas with severe dysplastic changes and skin atypia associated with an accumulation of aberrant hair follicles in skin of TGHK8 mice (Fig. 2C
).
|
5. Increased malignant progression of skin tumors in TGHK8 micexHa-ras mice
We examined the susceptibility of TGHK8 mice to develop skin cancer. Tumors were induced by TPA treatment in F1 crosses of TGHK8 mice with a transgenic strain (Tg.AC) carrying a mutant Ha-ras transgene that triggers the classic initiation event. We found that both groups of mice developed papillomas with a similar latency period (5–7 wk) and the percentage of animals developing tumors was similar (Fig. 2D
), although tumor multiplicity (number of tumors per mouse) was significantly reduced in TGHK8xHa-ras mice (Fig. 2E
). When the tumors first appeared, they were of similar size and displayed an exophytic growth in both TGHK8xHa-ras and control x Ha-ras mice. At 12 wk of promotion, tumors of the TGHK8xHa-ras mice often began to show an endophytic growth not seen in the papillomas of wild-type x Ha-ras mice, which continued to grow in an exophytic way (Fig. 2G
). Histopathological examination of tumors from wild-type x Ha-ras and transgenic xHa-ras mice 20–27 wk after the promotion protocol showed a significantly higher incidence of malignant progression in TGHK8xHa-ras mice (Fig. 2F
), measured as the frequency of appearance of focal invasive areas, microinvasive infiltration (Fig. 2I
), and in situ carcinomas. In TGHK8xHa-ras mice, a few tumors left to develop until wk 32–33 became very aggressive, undifferentiated, squamous cell carcinomas. Conversion had not occurred in wild-type mouse tumors by this time. 
6 integrin expression was used as a marker of malignancy (Fig. 2J, K
). Immunostaining of wild-type x Ha-ras and TGHK8xHa-ras tumors shows a disseminated expression of 6 integrin in basal and suprabasal layers of the TGHK8xHa-ras mouse-derived tumor (Fig. 2J
) vs. the basal localization of 6 integrin in wild-type x Ha-ras tumors (Fig. 2K
). The disseminated expression of 6 integrin corroborates the malignant phenotype of TGHK8xHa-ras tumors found by histopathological analysis. HK8 staining with a human K8-specific antibody of TGHK8xHa-ras tumors shows mosaic expression of the transgenic keratin (Fig. 2L, M
). This is in accordance with that obtained from staining of adult interfollicular skin, where we observed discontinuous HK8 expression. HK8 expression is detected more extensively in undifferentiated basal regions of the tumor (Fig. 2L
). An extreme case of mosaicism showing the absence of HK8 in more differentiated, benign regions of the tumor and a robust staining in the invasive area is presented in Fig. 2M
.
CONCLUSIONS AND SIGNIFICANCE
Keratin K8 is considered to be a marker of malignancy in skin tumors, including human cancer. Increased levels or ectopic expression of simple epithelium keratins in invasive tumors of different origin have been understood to be a consequence of malignancy. No causative role for K8/K18 in tumorigenesis had been identified. Here we report the study of a mouse model of transgenic mice that allows us to discern whether the aberrant expression of keratin 8 in skin and skin tumors is the cause or the consequence of malignization. To our knowledge, this is the first time a suitable in vivo model to analyze these questions has been developed.
We have found that the ectopic expression of keratin 8 in skin causes major alterations in its morphology, including epidermis and hair follicles hyperplasia, dysplasia, and ultimately preneoplastic changes of differentiated epidermal cells in aging mice (Fig. 3
). Our findings also show that keratinocyte terminal differentiation is altered in TGHK8 mice.
|
The dysplastic changes observed in hair follicles expressing K8 are self-renewing and contain reservoirs of multipotent stem cells capable of regenerating the epidermis. These are the origin of many neoplasms, including carcinomas and pilomatricomas, arising through the inappropriate activation of signaling pathways that regulate hair follicle morphogenesis and the hair cycle.
The use of our model of transgenic mice in chemical skin carcinogenesis assays has revealed that the aberrant keratin 8 expression in epidermis and hair follicles dramatically promotes malignant progression of benign skin tumors (Fig. 3)
. While the diminished number of papillomas developed by TGHK8 transgenic mice might be explained by the presence of their many aberrant hair follicles, the enhanced susceptibility progression toward malignancy seen in the transgenic tumors could be due to alterations in the pattern of hair follicle differentiation seen in TGHK8 mice. It is conceivable that the number of hair follicle keratinocyte populations with high growth/differentiation potential is increased in the skin of transgenic mice.
Our results support the hypothesis that K8 expression itself or any kind of perturbation originated in other molecules/pathways by inappropriate K8 expression in skin is far from innocuous, because it impairs the normal regulation of epidermal morphogenesis and is ultimately responsible for the invasive behavior of transformed epidermal cells.
This report provides evidences that K8 may function as a promoter of malignancy. Further research is required to elucidate the signaling pathways involved.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-1683fje;
This article has been cited by other articles:
|
|
 |
|
  K. M. Newkirk, A. E. Parent, S. L. Fossey, C. Choi, H. L. Chandler, P. J. Rajala-Schultz, and D. F. Kusewitt Snai2 Expression Enhances Ultraviolet Radiation-Induced Skin Carcinogenesis Am. J. Pathol., November 1, 2007; 171(5): 1629 - 1639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. F. Basil, Y. Zhao, K. Zavaglia, P. Jin, M. C. Panelli, S. Voiculescu, S. Mandruzzato, H. M. Lee, B. Seliger, R. S. Freedman, et al. Common cancer biomarkers. Cancer Res., March 15, 2006; 66(6): 2953 - 2961. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
