HomeLegsBrainsEyesNecksMouthsManeClaws/HoovesSexes

 

 

Claws/Hooves Development & Induction

 

 

 

Normal Development:

So much of development is due to epithelial-mesenchymal interactions, where almost always the mesenchyme provides the initial signal that induces the responding tissue, the epithelial. Hoof and claw development is a perfect example; they are formed because of a series of reciprocal exchanges of inductive signals. In the case of limb development it is the somatic mesoderm that is responsible for the initial signal to the limb dermis.  There is a limb morphogenic field that is committed to give rise to the limb and will do so upon receiving the right induction signal. Tbx4 is a transcription factor that is expressed in the hind limbs and in combination with a growth factor, FGF, the limb out grows and forms the hooves.   As the limb grows outward, the stylopod forms first, followed by the zeugopod and then the autopod; this shows how the tissues are differentiated in the proximal to distal order.  Each phase of limb development requires a particular expression of hox genes. BMPs are expressed in the distal area of the sclerotome and BMPs both induce apoptosis to form the digits but also help differentiate mesenchyme cells into cartilage.  In normal development BMPs are inhibited so that apoptosis can take place and digits and joints are formed. Claws form similar in which a claw field develops as an epithelial thickening on the dorsal side of each digit.  This epithelial thickening, or placode, is the first sign of induction for an epithelial appendage.  Next to this placode a transverse groove appears and forms a deep fold in the epidermal matrix and the cells of this matrix produce a keratinized layer that slides distally over the claw bed. (Hamrick 2001).

 

© 1999 by John Stear

(Click on Image to view Original Source)

How we differ:

Although much of our development appears to be very similar to this so-called chick creature, we do differ in some of the hooves and claw development.  It appears that our development is very similar for the stylopod and zeugopod, but differs at the autopod region.  Whereas the chick embryo secretes noggin, a BMP inhibitor, it is obvious that we lack this factor because it is the increase expression of BPM that converts the mesechyme of the digits to cartilage which will later form hard keratin and form the hooves (Gilbert 2003).  Our specific claw pattern develops through interactions with SHH which is expressed in the region of the limb bud called the zone of polarizing activity (ZPA), which is a region of mesoderm located between the limb bud and the body wall which affects the development of digits in the limb.  SHH induces a signal that affects the surrounding hox genes and this epithelial-mesenchyme interaction is what results in the claws of the hooves. We seem to display a Hoxd-13 mutation, (as those people on earth call it) which is why the claws on its hooves can fuse together.  Thereby, three claws form instead of the normal five found in most vertebrates on earth.  In creatures on earth, this is considered a deformity known as syndactyly, as seen below.  This syndrome is analogous to the our normal claw. 

 

© 2005 by Jon A. Baskin

(Click on Image to view Original Source)

 

 

                                                  

               © 2005 Florida Museum of Natural History                                       © 1996-2003 Nesssus

                                  (Click on Image to view Original Source)                                                      (Click on Image to view Original Source)

 

 

The central digit becomes increasingly stronger while the “side digits” become less important and are virtually lost in us.  

 

 

Sources:

 

Gilbert, Scott F. Development Biology. 7th Editon. Massachusetts: Sinauer Associates, Inc., 2003.

 

Hamrick, Mark W. “Development and evolution of the mammalian limb: adaptive diversification of nails, hooves, and claws.” Evolution and Development. Volume 3, Issue 5, Page 355. September-October 2001. 

 

Tosney, K. Lecture Material. Bio 208, Fall 2004