Barrett’s esophagus, often a precursor to esophageal cancer, results from residual, embryonic cells

This research will be published online in the June 24th issue of Cell.

Decades of cancer research tells us that most of the common cancers begin with genetic changes that occur over a period of 15 to 20 years, in some cases leading to aggressive cancers. However, for a subset of cancers that appear to be linked to chronic inflammation, this model might not hold.

Barrett's esophagus, which was first described by the Australian surgeon Norman Barrett in 1950, affects two to four million Americans. In this condition, tissue forms in the esophagus that resembles the intestinal tissue normally located much farther down the digestive tract. As a result, a person's chances of developing a deadly esophageal adenocarcinoma increase by 50- to 150-fold. Late stage treatment is largely palliative, so it is important to understand how acid reflux triggers it in the first place.

Research from the laboratory of Frank McKeon, Harvard Medical School professor of cell biology, together with Wa Xian, a postdoctoral researcher at Brigham and Women's Hospital and the Institute of Medical Biology, Singapore, along with an international consortium including Christopher Crum, director of Women's and Perinatal Pathology at Brigham and Women's Hospital, has shown that Barrett's esophagus originates from a minor population of non-esophageal cells left over from early development.

For the past decade, McKeon and his laboratory have been using mouse models to investigate the role of p63, a gene involved in the self-renewal of epithelial stem cells including those of the esophagus. McKeon joined forces two years ago with Wa Xian, an expert in signal transduction in cancer cells, to tackle the vexing problem of the origin of Barrett's esophagus.

At that time, the dominant hypothesis for Barrett's was that acid reflux triggers the esophageal stem cells to make intestine cells rather than normal esophageal tissue. However, McKeon and Xian felt the support for this concept was weak. Taking a different track, they studied a mouse mutant lacking the p63 gene and mimicked the symptoms of acid reflux. As a result, the entire esophagus was covered with a Barrett's-like tissue that proved to be a near exact match with human Barrett's at the gene expression level.

The researchers were particularly surprised by the sheer speed with which this Barrett's esophagus appeared in the mice.

"From the speed alone we knew we were dealing with something different here," said Xia Wang, postdoctoral fellow at Harvard Medical School and co-first author of this work.

Yusuke Yamamoto, a postdoctoral fellow at the Genome Institute of Singapore and also co-first author, added that, "we just had to track the origins of the Barrett's cells back through embryogenesis using our markers from extensive bioinformatics."

In essence, the investigators tracked the precancerous growth to a discrete group of leftover embryonic cells wedged between the junction of the esophagus and the stomach--precisely where endoscopists have argued Barrett's esophagus begins. As predicted by the mouse studies, the researchers identified a group of embryonic cells exactly at the junction between the esophagus and the stomach in all normal humans.

"Barrett's arises from this discrete group of pre-existing, residual embryonic cells present in all adults that seemingly lie-in-wait for a chance to take over when the esophagus is damaged," said McKeon. Added Xian, "We know these embryonic cells have different gene expression patterns from all normal tissues and this makes them inviting targets for therapies to destroy Barrett's before it progresses to cancer."

The therapeutic opportunities of this work are potentially immense.

"We are directing monoclonal antibodies to cell surface markers that can identify these precursor cells, so we may have a new opportunity to intervene therapeutically and prevent Barrett's esophagus in at-risk patients," said Wa Xian.

"Additionally," noted McKeon, "we are cloning the stem cells for both these precursors and for Barrett's esophagus itself, and these should represent critical targets for both monoclonal antibodies and small molecule inhibitors."

Finally, there is reason to believe that this unusual mechanism might apply to a subset of other lethal cancers with unsure origins.

Crum noted that "some very aggressive cancers arise at junctions of two tissues and these deserve closer scrutiny to get at their origins if we are to surmount these diseases."

This work was supported by the National Institutes of Health.