Ese analyses we could not detect any changes in K8 expression in homozygous K7 knockout mouse SPDB site tissues as compared to tissues from wildtype mice (Table 1 and Figure S3). Immunofluorescence staining suggested reduced K18 expression within the collecting tubules of the kidney of K7 knockout mice (Figure S4) whereas other tissues showed no difference in K18 expression (Table 1). Despite extensive co-expression with K7 inwildtype tissues, there was no difference in K19 expression in the tissues of homozygous K7 knockout mice as compared to wildtype tissues (Table 1 and Figure S5). K20 is even more restricted in its pattern of expression as compared to K8 or K19, and in wildtype tissues where it is expressed along with K7, it did not appear to colocalise with K7 except for in the bladder where there was colocalisation of K7 with K20 at the apical cell 256373-96-3 site membranes of superficial urothelial cells (Figure S2). However, in the absence of K7, K20 still remained localised at the apical cell membranes of urothelial cells (Figure S2). In other K7 knockout tissues, K20 expression was unchanged (Table 1).DiscussionIn this paper we have described the generation and characterisation of K7 knockout mice, one of the remaining so-called “simple” epithelial keratin genes to be knocked out in the mouse using conventional gene targeting. Although 16985061 the absence of K7 protein was not associated with any pathological phenotype this is not an unexpected result given that K18 knockout mice, which also show secondary loss of K7, only developed a mild late-onset phenotype that was restricted to hepatocytes with no other associated pathology [11]. Compensation for the loss of K7 by other type II keratins, in particular K8 whose expression closely overlaps with K7 [1], provides the most likely reason for the lack of any overt phenotype in K7 knockout mice but overcoming this problem of functional redundancy, through the generation of a K7/K8 double knockout mouse for example, would be difficult since both genes are closely associated within the keratin gene cluster on mouse chromosome 15 and based on previous studies embryos lacking both K7 and K8 are likely to be non-viable anyway [12,13]. Despite the absence of any pathology associated with the loss of K7, homozygous K7 knockout mice showed increased proliferation of the bladder urothelium. Hyperproliferation is a feature of several keratin mouse knockouts and includes those which affect internal epithelial such as K8 [10] and K4 [17] as well as certain epidermally-expressed keratins such as K10 [18]. However, unlike these keratin knockout mice where hyperproliferation was associated with hyperplasia, there was no apparent urothelial hyperplasia in the bladder of K7 knockout mice. Unlike other types of epithelia such as the colon and epidermis, urothelial cells are characterised by their low proliferation index and longevity [16,19]. Therefore the five-fold increase in urothelial cell proliferation that we observed in K7 knockout mice, although similar to the four-fold increase in keratinocyte proliferation observed in the hyperplastic epidermis of K10 knockout mice [18], may simply have been insufficient to produce a phenotype in this particular type of epithelium. It is not clear how the absence of K7 led to stimulation of the normally quiescent urothelium since histologically there was no obvious disruption to the urothelium such as loss of the superficial umbrella cell layer, nor any evidence of apoptotic urothelia.Ese analyses we could not detect any changes in K8 expression in homozygous K7 knockout mouse tissues as compared to tissues from wildtype mice (Table 1 and Figure S3). Immunofluorescence staining suggested reduced K18 expression within the collecting tubules of the kidney of K7 knockout mice (Figure S4) whereas other tissues showed no difference in K18 expression (Table 1). Despite extensive co-expression with K7 inwildtype tissues, there was no difference in K19 expression in the tissues of homozygous K7 knockout mice as compared to wildtype tissues (Table 1 and Figure S5). K20 is even more restricted in its pattern of expression as compared to K8 or K19, and in wildtype tissues where it is expressed along with K7, it did not appear to colocalise with K7 except for in the bladder where there was colocalisation of K7 with K20 at the apical cell membranes of superficial urothelial cells (Figure S2). However, in the absence of K7, K20 still remained localised at the apical cell membranes of urothelial cells (Figure S2). In other K7 knockout tissues, K20 expression was unchanged (Table 1).DiscussionIn this paper we have described the generation and characterisation of K7 knockout mice, one of the remaining so-called “simple” epithelial keratin genes to be knocked out in the mouse using conventional gene targeting. Although 16985061 the absence of K7 protein was not associated with any pathological phenotype this is not an unexpected result given that K18 knockout mice, which also show secondary loss of K7, only developed a mild late-onset phenotype that was restricted to hepatocytes with no other associated pathology [11]. Compensation for the loss of K7 by other type II keratins, in particular K8 whose expression closely overlaps with K7 [1], provides the most likely reason for the lack of any overt phenotype in K7 knockout mice but overcoming this problem of functional redundancy, through the generation of a K7/K8 double knockout mouse for example, would be difficult since both genes are closely associated within the keratin gene cluster on mouse chromosome 15 and based on previous studies embryos lacking both K7 and K8 are likely to be non-viable anyway [12,13]. Despite the absence of any pathology associated with the loss of K7, homozygous K7 knockout mice showed increased proliferation of the bladder urothelium. Hyperproliferation is a feature of several keratin mouse knockouts and includes those which affect internal epithelial such as K8 [10] and K4 [17] as well as certain epidermally-expressed keratins such as K10 [18]. However, unlike these keratin knockout mice where hyperproliferation was associated with hyperplasia, there was no apparent urothelial hyperplasia in the bladder of K7 knockout mice. Unlike other types of epithelia such as the colon and epidermis, urothelial cells are characterised by their low proliferation index and longevity [16,19]. Therefore the five-fold increase in urothelial cell proliferation that we observed in K7 knockout mice, although similar to the four-fold increase in keratinocyte proliferation observed in the hyperplastic epidermis of K10 knockout mice [18], may simply have been insufficient to produce a phenotype in this particular type of epithelium. It is not clear how the absence of K7 led to stimulation of the normally quiescent urothelium since histologically there was no obvious disruption to the urothelium such as loss of the superficial umbrella cell layer, nor any evidence of apoptotic urothelia.
