In this review, we will discuss the unique biological properties of the Xenopus egg extract system and the advantages it offers for screening small molecular modulators of complex biological pathways. We will discuss molecular pathways that have been reconstituted using Xenopus egg extracts and small molecule screens that have been performed using these assays. Finally, we will describe our recent studies using Xenopus egg extracts to identify small molecule modulators of the Wnt pathway and how this approach could be similarly adapted to other embryonic signaling pathways such as Hedgehog and Notch.
Over the past decade, there has been a great expansion in drug discovery efforts within academia as evidenced by the emergence of a new discipline known as Chemical Biology [1
]. With the advent of high throughput approaches and the development of faster and cheaper technologies, industry and academia have tremendous resources and opportunities to identify novel drugs at an accelerated pace. Traditionally, pharmacological agents have been identified through enzymatic assays using purified components [2
]. Such “targeted screening” approaches, which have been the focus of drug discovery over the past decade or so, offer a number of advantages. First, targeted biochemical screens represent the most direct way to identify drugs based on our current understanding of a molecular target or event. Second, enzymatic assays using purified components can often be readily adapted for high throughput screening (HTS); combined with well-established colorimetric, fluorescent, and luminescent readouts, experimental set-up is often relatively straightforward. Third, with purified components, drug target identification is not a limitation. Fourth, the recent advent of increasingly sophisticated technologies has allowed investigators to acquire very precise kinetic data. Such information can facilitate the grouping of drugs into subclasses based on mechanism even at initial stages of screening and can reveal subtle drug effects as well.
Targeted screening, however, has some major limitations. The most obvious is that the investigator is restricted to the originally hypothesized target; thus, there is no potential for uncovering novel targets. Furthermore, it is exceedingly difficult to predict which compounds will work in vivo
or even whether the targeted molecule will be an effective therapeutic target in the first place. An alternative approach to targeted drug screening is to screen for a specific phenotype [1
]. Phenotypic screens can be performed in systems ranging from cultured cells to whole organisms. In fact, in recent years, a majority of compounds receiving FDA approval were discovered through phenotypic screens [5
]. These require no a priori
knowledge of the direct target of compounds, only knowledge of the desired phenotype. Representative phenotypic screens include those based on C. elegans
morphology, gastrulation in zebrafish, and mitosis in cultured mammalian cells [6
]. Screening in such complex systems pre-selects for compounds that are cell/organism permeable, reach their target, and induce a desired effect in vivo
; as such, hits are proven to be active in vivo
from the start. The identification of novel drug targets can lend to important insights into the biological process in question.
As with targeted screening, there are also major drawbacks to phenotypic screening. Living systems are much more difficult to manipulate in HTS format and require significantly more manual set-up. Phenotypes can often require time-consuming manual inspection or sophisticated algorithms for image analysis. Drug incubation times are longer (i.e. on the order of hours to days) such that screens require significantly more time to perform. The longer time course also increases the likelihood that the desired phenotypes may occur via non-direct mechanisms. Finding the optimal dose of drug to screen is another challenge: too low of a drug concentration can give false negative results, whereas too high of a drug concentration can produce toxicity with a valid drug candidate. Perhaps the major limitation of phenotypic screens, however, is the difficulty of drug target identification, which can represent a formidable challenge.