Getting one’s Fak straight

Abstract

Focal adhesion kinase (FAK) is a pivotal regulator of integrin signaling and responses to cell adhesive dynamics. In this issue, Ashton et al. demonstrate that Fak is critical for intestinal oncogenesis and regeneration after injury but not for day-to-day homeostasis, providing novel insights into intestinal biology and colorectal cancer therapy.

Large, long-lived organisms like vertebrates demand three things of their most environmentally exposed tissues - that they self-assemble reliably and consistently during development, that they maintain themselves homeostatically (at least through reproductive life), and that they repair themselves with adequate fidelity if damaged or disrupted. Nonetheless, unfettering the capacity of those same tissues to regenerate must be tempered by the risk of individual somatic cells’ going rogue and expanding into life-threatening tumors. The resulting compromise entails a complicated and dynamic balancing act between social control and individual liberty that would make the authors of the US constitution blush with humility. To make matters worse, dramatic differences in architecture, turnover, and risk of neoplasia, damage and infection amongst differing tissues mean that the rules governing homeostasis and repair must vary greatly between tissue types, raising the fascinating question of how evolution has managed to balance such diverse dynamics in so many different tissue types using only a limited, common toolkit of signaling and effector molecules. In this regard, intestinal epithelium is an especially informative adult tissue for study. Its distinctive epithelial architecture, supporting the continuous and swift conveyer belt that shuttles newborn cells out of the crypt to their demise at the villus tip provides, at once, both a spatial and temporal map of the tissue’s dynamics and an unparalleled opportunity to establish cause-and-effect relationships between underlying signaling pathways and consequent biology – especially when, as in the accompanying paper by Ashton et al., combined with state-of-the-art switchable mouse genetics.

The Wnt/β-catenin and Notch pathways together govern the development, maintenance and repair of the intestinal epithelium. However, the Wnt/β-catenin axis is pre-eminent in maintaining the proliferation of both the crypt stem cell and adjacent transit amplifying compartments, acting in great part through the proto-oncogene c-myc (He et al., 1998). It is also essential for the appropriate spatial disposition of the different cell types that comprise the crypt-villus unit and for migration of epithelial cells up the crypt-villus axis (Batlle et al., 2002). Constitutive signaling through the Wnt/β-catenin/TCF pathway effectively locks intestinal cells in a crypt progenitor-like state and drives the precocious proliferation and suppressed differentiation that underpins most colorectal cancers (van de Wetering et al., 2002).

Ashton et al. focus their attention on the potential role in intestinal homeostasis and regeneration played by Focal Adhesion Kinase (FAK), the integrin-activated cytoplasmic tyrosine kinase that articulates the highly nuanced responses of epithelial and other cell types to changes in inter-cellular adhesion status. Although there is as yet no definitive evidence linking FAK kinase activity with tumorigenesis, FAK is upregulated in colorectal cancers from its earliest stages and its role in integrin signaling places it center stage in many processes that drive colorectal cancer – cell adhesion, proliferation, invasion, migration and survival. Conversely, little is known of the role FAK plays in normal, adult intestinal biology. Ashton et al. use conditional, intestine-specific deletion of FAK and its upstream and downstream signaling partners to define the role of FAK signaling in intestinal homeostasis and regeneration. The first surprise is that FAK is dispensable for quotidian intestinal homeostasis - quite unlike the Wnt/β-catenin target gene c-myc, whose ablation precipitates immediate failure of intestinal crypt maintenance (Muncan et al., 2006; Soucek et al., 2008). Of note, a major limitation of conditional cre-lox technology is that gene deletion is never quantitative throughout a tissue – unrecombined cells always remain. In the case of c-myc deletion, such residual c-myc competent cells rapidly outgrow those crypts deficient in c-myc, presently obscuring the c-myc deletion phenotype (Bettess et al., 2005; Muncan et al., 2006) and confirming the critical role c-myc plays in crypt-villus homeostasis. No analogous outgrowth of FAK-competent crypts occurs after conditional deletion of FAK, belying even the most subtle of homeostatic roles for FAK.

The situation is, however, quite different during intestinal regeneration after injury. Intestinal epithelium is prey to damage and disruption by pathogens, parasites, injury and immune attack but must nonetheless steadfastly maintain its critical barrier and absorptive functions. Consequently, intestinal epithelium enjoys significant regenerative capacity. When intestinal regeneration is experimentally triggered by radiation injury, FAK is induced. Likewise, super-activation of the Wnt/β-catenin pathway through acute ablation of the Apc gene up-regulates FAK. Moreover, induction of FAK by either radiation injury or elevated Wnt/β-catenin signaling is abrogated upon deletion of c-myc, consistent with the notion that FAK expression is induced by a Wnt/β-catenin/c-myc pathway. However, this doesn’t actually prove that c-Myc is the bona fide downstream Wnt/β-catenin effector that transcriptionally elicits FAK expression because basal c-myc activity might instead be serving as a permissivity factor that renders the FAK gene competent for induction by some other Wnt/β-catenin-induced transcription factor. Ashton et al. elegantly refute this latter possibility by showing that constitutive expression of c-myc at physiological levels, in this case from the weak but Wnt/β-catenin-insensitive Rosa26 promoter (Murphy et al., 2008), cannot compensate for deletion of the endogenous c-myc gene. Hence, c-myc mediates intestinal regeneration and FAK up-regulation only if it is responsive to Wnt/β-catenin signaling. Case closed. As icing on the cake, they also show that FAK function is required for overt intestinal tumorigenesis induced in vivo by inactivation of Apc. Given that FAK loss has no discernible impact on intestinal homeostasis, this opens the way for both prophylactic and therapeutic use of FAK inhibitors in colorectal cancer. Indeed, the pivotal role played by FAK in both regenerative and oncogenic Wnt/β-catenin-induced intestinal regeneration, but not in intestinal homeostasis, is especially provocative: it indicates the existence of a discrete, evolved, intestinal regenerative program distinct from normal intestinal homeostasis and, in turn, implies that colorectal oncogenesis is a hijacked version of this dedicated regenerative program.

FAK is implicated in many biological functions, most notably resistance to detachment-induced cell death (anoikis) and, ipso facto, tumor invasion and metastasis. But what might FAK do to support intestinal regeneration? Here, the situation is more murky. Ashton et al. show that FAK status has no effect on the immediate induction of intestinal apoptosis by radiation - no great surprise since FAK is not a generic inhibitor of apoptosis such as that induced by DNA damage. On the other hand, functional FAK does ameliorate the more delayed apoptosis that attends the actual regenerative process, perhaps by blunting the transient anoikis that inevitably accompanies the regenerative remodeling. Proliferation of injured FAK-deficient intestine is also compromised, suggesting an additional role for FAK in orchestrating regenerative cell proliferation. Activated FAK engages several signaling modalities, including the canonical Ras-Erk, PLCγ and PI-3K/Akt/mTOR pathways, so it is potentially intriguing that that FAK deletion blocks activation of mTOR during intestinal regeneration and that, in turn, pharmacological inhibition of either PI-3K or mTOR blocks radiation-induced intestinal regeneration. However, such inferences are less than watertight because PI-3K/Akt/mTOR signaling is generically involved in suppressing apoptosis: hence, it is not unexpected that pharmacological inhibition of this crucial pathway degrades intestinal regenerative potential. Similarly, while Ashton et al.’s restoration of regenerative capacity in the intestines of FAK-deficient intestines by systemic administration of IGF-1 is intriguing, activation of the IGF type I receptor engages a plenitude of downstream effector pathways, making this observation provocative but not yet compelling. That notwithstanding, Ashton et al. craft a convincing and elegantly comprehensive picture of a fascinating and unexpected regenerative program that offers new opportunities for the treatment of a dreadful disease.