UC Berkeley

A remarkable feat of animal development is the precision with which organs grow to a

consistent and characteristic size. The lengths of the right and left arm of a human, for

instance, match with an accuracy of about 0.2%, and the size of adult mouse brains varies

by only about 5%. How a developing organ|and the individual cells within|can know

when it is time to stop growing has long-fascinated biologists.

Reciprocal organ transplantation experiments from the 1920s were among the rst to

demonstrate that, for an individual organ, growth is subject to both extrinsic and intrinsic

programs. Working with two closely related but dierent-sized species of salamanders, the

faster-growing yet smaller-sized Ambystoma punctatum and the slower-growing yet largersized

Ambystoma tigrinum, Harrison found that the growth rate of a limb transplanted from

one species to the other matched that of the host, but the overall size of the transplanted

limb matched that of the donor. He concluded, therefore, that while some circulating factor,

perhaps a hormone, dictated the rate of limb growth, the organ's size overall was determined

by some intrinsic "growth potential."

Now, close to a century later, we know the identities of many evolutionarily-conserved

factors that are critical to achieving a properly-sized organ, but our understanding of what

ultimately determines an organ's size remains far from complete. Much of what we do know

about the mechanisms regulating tissue growth comes from studies using the model organism

Drosophila melanogaster, whose genetic tractability and high reproducibility of organ size

allow us to easily evaluate the eects of genetic perturbations on tissue growth.

In Chapter 1, I provide an overview of some of these growth-regulatory mechanisms

and describe how genetic studies using a well-characterized tissue model of growth|the

Drosophila wing imaginal disc|have helped to provide some clues into how these mechanisms

might operate. I will highlight two signaling pathways|the Hippo pathway and the JNK

pathway|that have very dierent functions but seem to both impact tissue growth in unique

ways. Activities of both of these pathways have been areas of interest for my dissertation

research.

The Hippo pathway regulates growth in most multicellular organisms and is altered

(either directly or indirectly) in many human cancers. Since the elucidation of the core

pathway approximately a decade ago, a key aim for researchers has been to identify the

upstream signals that link this pathway to external cues and understand how these signals

are sensed. An important upstream regulator of Hippo signaling is the transmembrane

protein Fat (Ft), though the mechanistic link between the core pathway and this upstream

factor is not fully clear. We identied the Drosophila F-box protein, Fbxl7, as a downstream

eector of Ft activity that is important for regulating the cellular distribution of Dachs, a

protein that mediates much of Ft's eect on tissue growth. This work, described in Chapter

2, provides a more complete understanding of the functional link between Ft signaling and

the Hippo pathway.

The JNK pathway is a well-conserved pathway involved in several morphogenetic processes

during development and commonly activated in response to stress. JNK signaling

promotes apoptosis, yet|quite paradoxically|is also important in promoting tissue proliferation

during regeneration and tumorigenesis. The role this pathway plays in regulating

growth during organ development is less established. We found that JNK signaling is active

at low levels in the developing wing imaginal disc and regulates an enhancer of the

gene bantam (ban), which encodes a microRNA that promotes growth. This ban enhancer

activity is opposed by the transcriptional co-repressor CtBP, which we characterize as a

negative growth-regulator in Drosophila. These ndings, described in Chapter 3, support a

role for JNK signaling in promoting tissue growth and suggest that CtBP may help to direct

this broadly-functioning signaling pathway towards specic eects on growth during normal

development.

During the course of these studies, I made the surprising observation that a common

genetic technique used in Drosophila, namely shRNA-mediated gene knockdown using the

Gal4/UAS system, can lead to an unexpected result: clonal expression of shRNAs causes

knockdown in cells that do not express Gal4. Chapter 4 describes this phenomenon, which we

term "shadow RNAi," and shows how this eect can lead to erroneous conclusions regarding

cell-autonomous vs. non-autonomous genetic functions. We outline how shadow RNAi can

be mitigated, as well as how it can be exploited as an eective lineage-tracing tool.

Taken together, my work oers important clues for solving the mysteries of organ size

control and explores new applications of standard genetic techniques, while potentially providing

useful insights for the development of novel tools in cancer therapy and regenerative

medicine.