A new branch

One of the wonders of development is the process of branching morphogenesis. Characteristic of many organs such as salivary glands and kidney, complex three-dimensional epithelial branching structures arise from repetitive cleft and bud formation, but the exact mechanisms behind this are not clear. Ken Yamada's group, though, now reports on the essential role of the extracellular matrix (ECM) protein fibronectin in epithelial branching.

ECM components are already known to be required for salivary gland branching, but the authors were interested in identifying locally synthesized regulatory proteins that might influence branching. Because collagen type III accumulates at clefts and fibronectin can regulate collagen III formation, Yamada's group proposed that local fibronectin production might be important in branching morphogenesis, so they searched for differences in fibronectin messenger RNA expression in salivary glands. Quantitative reverse-transcription polymerase chain reaction showed that fibronectin mRNA was expressed at 16-fold higher levels in cleft epithelial cells than in bud epithelium. A closer look showed that cells next to nascent clefts expressed fibronectin mRNA, and high levels of fibronectin fibrils were seen in clefts of branching epithelium. Furthermore, an even closer look showed that the levels of the cell–cell adhesion molecule E-cadherin were lower in the regions right next to the fibronectin fibrils.

So if the accumulation of fibronectin fibrils promoted cleft formation during branching morphogenesis, could inhibiting fibronectin function block branching? Anti-fibronectin antibodies did indeed prevent salivary cleft formation and branching in a dose-dependent manner. Furthermore, antibodies against β₁, α₅ or α₆ integrins inhibited salivary branching. α₅β₁ is a key fibronectin receptor, but the fact that combining antibodies against α₅ and α₆ more effectively inhibited branching indicated that fibronectin together with laminin (α₆β₁ is its main receptor in this tissue) might be necessary for branching.

Similar to the results seen when protein function was inhibited, small interfering RNA (siRNA)-mediated knock-down of fibronectin expression in salivary-gland organ culture also markedly decreased cleft formation. Yamada's group then looked at the effects of using exogenous fibronectin to replace the fibronectin. Not only did it successfully restore branching in siRNA-treated salivary glands, but it also stimulated branching in control, non-siRNA-treated cultures in a dose-dependent manner — cleft formation was accelerated and clefts deepened. Similar observations were made in other organs — fibronectin accumulated at sites of indentation, blocking fibronectin function inhibited branching, and exogenous fibronectin promoted branching.

The authors then returned to their previous observation that cadherin levels decreased near presumptive cleft regions, and tested whether fibronectin might induce this decrease. Treating cultured salivary gland epithelial cells with fibrillar fibronectin induced local cell–matrix adhesions, but next to these sites, cadherin localization was suppressed. The predicted resultant loss of cell–cell adhesion would then provide a way for deep clefts to form. And, as well as being responsible for creating the clefts, fibronectin might also maintain them through its ability to regulate collagen, which is thought to be essential for stabilizing clefts.