Comparing the Sequences of Keratins Expressed in "Soft" Epithelia

In addition to the type I-type II classification, keratin genes can be discriminated based on whether they are expressed in so-called hard epithelia (hair, nail, filiform papillae of the tongue) or in soft epithelia (all others). Here we compare the amino acid sequences of keratin proteins expressed in human soft epithelial tissues (ranging in complexity from liver to epidermis). Individual sequences were obtained from GenBank and were compared using the ClustalW software. The outcome of the comparison is illustrated by mean of a tree display, in which sequence relatedness in inversely related to the length of the lines connecting the various sequences, as well as the number and position of the branch points. Note that K6a is the only K6 isoform sequence included in this comparison - the K6a, K6b, K6c, K6d, and K6e sequences are so similar that they would all be superimposed on the same branch.

What is the outcome of this analysis? Two major branches can be identified in this tree, and they correspond exactly to the known partition into type I (top; K9-K20) and type II (bottom; K1-K8) sequences. Within each keratin subtype, two major branches occur as well. These correspond to the sequences expressed primarily in simple epithelia (K7 and K8 as type II; K18 and K20 as type I) and in complex epithelia (all others) (Note: K19 represents an exception, in that it is best related to the K13-K17 group whereas many of the epithelia in which it is expressed are of the simple type). Note that the tree structure would be essentially the same if it had involved mouse keratin protein sequences.

Clicking on any of the individual entries identified with an arrow will lead you to a page in which information about the sequence, its regulation, and function (in human and mouse) are provided.