Clonal transitions and phenotypic evolution in Barrett esophagus

James A. Evans, Emanuela Carlotti, Meng‐Lay Lin, Richard Hackett, Adam M. Passman, Lorna Dunn, George Elia, Ross J Porter, Mairi H McClean, Frances Hughes, Joanne Chin‐Aleong, Philip Woodland, Sean Preston, S M Griffin, Laurence Lovat, Manuel Rodriguez‐Justo, Nicholas A. Wright, Marnix Jansen, Stuart A.c. McDonaldmedRxiv 2021 · DOI

Abstract

Barrett esophagus (BE) is a significant risk factor for esophageal adenocarcinoma, yet the mechanisms of its phenotypic evolution remain poorly understood. This study demonstrates that dynamic clonal phenotypic changes occur at the gland level and that increased glandular diversity is associated with progression to dysplasia. By analyzing mitochondrial DNA (mtDNA) mutations as clonal markers, the researchers established common ancestry between gastric-type (oxynto-cardiac) and intestinal-type (specialized) glands. These findings suggest that BE phenotypes represent an evolutionary process where the cardiac gland serves as a basic unit that diversifies in response to the esophageal microenvironment.

Introduction: The Challenge of Barrett Esophagus

Barrett esophagus is a condition characterized by the replacement of the normal squamous lining of the esophagus with a metaplastic columnar epithelium. This transformation significantly increases the risk of developing esophageal adenocarcinoma (EA). While genetic diversity—the accumulation of mutations—is a known predictor of cancer risk, the "phenotypic" evolution of the tissue has been less clear. Specifically, researchers have long debated whether the various gland types observed in BE (some resembling stomach tissue, others resembling intestinal tissue) arise independently or evolve from one another. Understanding this "niche succession" is vital for identifying patients at high risk of progression.

Methodology: Tracking Lineages with Mitochondrial DNA

To map the evolutionary "family tree" of BE cells, the research team employed a multi-modal approach. They identified five distinct gland phenotypes using histology and immunohistochemistry (IHC) for markers such as MUC2 (goblet cells) and H+/K+-ATPase (parietal cells).

The study's innovative "secret sauce" was the use of mitochondrial DNA (mtDNA) as a molecular clock. Because mtDNA mutations are inherited by all daughter cells, they act as natural barcodes. By using laser capture microdissection (LCM) to isolate individual glands and sequencing their mitochondrial genomes, the team could determine if different-looking glands shared rare mutations, thereby proving a common ancestry. Additionally, the Shannon Index—an ecological tool for measuring species diversity—was used to quantify the variety of gland types within patient biopsies.

The Evolutionary Tree of Glands

The most striking finding of the study was the confirmation of shared ancestry between seemingly unrelated gland types. The researchers discovered that "oxynto-cardiac" glands, which contain acid-producing parietal cells, and "specialized" Barrett glands, which contain intestinal goblet cells, can share identical mtDNA mutations. This provides definitive evidence that intestinal-type tissue in the esophagus can evolve directly from gastric-type tissue.

Evolutionary pathways of Barrett esophagus glands and the correlation between phenotypic diversity and progression risk.

As illustrated in the model above, the authors propose that the simple cardiac-type gland acts as the "fulcrum" or basic unit of BE. From this starting point, glands evolve and diversify to adapt to the harsh, acid-reflux-heavy environment of the esophagus. This expansion likely occurs through "gland fission," where a single gland splits into two, preserving its phenotypic and genetic identity.

The Clinical Significance of Phenotypic Diversity

The study revealed that phenotypic diversity is not merely random variation but a significant indicator of disease state. By analyzing three patient cohorts—non-dysplastic BE, BE adjacent to dysplasia/cancer, and "neo-BE" (early-stage BE post-esophagectomy)—the team found that diversity increases over time and with disease progression.

Quantitative analysis showed that the Shannon Index for BE adjacent to dysplasia was significantly higher than for non-dysplastic or early-stage BE (p < 0.001). Interestingly, this phenotypic richness was independent of the patient's age or the total length of the Barrett’s segment, suggesting that diversity is driven by internal evolutionary dynamics rather than simply the duration of the condition.

Conclusion and Future Directions

This research provides a new framework for understanding Barrett esophagus as a dynamic evolutionary process. By demonstrating that gastric and intestinal lineages share common ancestors and that glandular diversity correlates with cancer risk, the study opens new avenues for clinical diagnostics.

Measuring phenotypic diversity could serve as a powerful clinical biomarker. If routine surveillance biopsies reveal high "glandular chaos" (a high Shannon Index), these patients may require more intensive monitoring or earlier intervention, even in the absence of obvious genetic drivers of cancer. Future work will need to validate these findings in larger, longitudinal cohorts to determine the precise predictive value of phenotypic diversity in clinical practice.

Summary updated

Abstract

Abstract Background &amp; Aims Barrett esophagus (BE) is a risk factor for the development of esophageal adenocarcinoma, however our understanding of how Barrett esophagus evolves is still poorly understood. We demonstrate that dynamic clonal phenotypic changes occur at the gland level, the mechanism by which these changes evolve, and how diversity may play a role in progression. Methods We analyzed the distribution and diversity of gland phenotype between and within BE biopsies and the background mucosa of those that had progressed to dysplasia or developed BE post-esophagectomy, using immunohistochemistry and H&amp;E analysis. Clonal relationships between distinct gland types were determined by laser capture microdissection sequencing of the mitochondrial genome. Results Five different non-dysplastic gland phenotypes were identified in a cohort of 64 patients biopsies taken at the same physical location in the esophagus; most non-dysplastic patients showed a single gland phenotype per biopsy, but some showed two or three gland types. We reveal a shared common ancestor between parietal cell-containing oxynto-cardiac glands and goblet cell-containing specialized Barrett glands through a shared somatic mtDNA mutation. We also reveal a similar relationship between specialized and cardiac-type glands, and specialized and Paneth cell-containing glands. The diversity of gland types was significantly increased adjacent to dysplasia compared to non-dysplastic BE and patients with post-esophagectomy BE, suggesting that gland diversity evolves in BE patients over time. Conclusions We have shown that the range of BE phenotypes represent an evolutionary process and that changes in gland diversity may play a role in progression. Furthermore, we demonstrate common ancestry between gastric and intestinal glands in BE. Graphic abstract A) The cardiac gland as the basic unit of Barrett esophagus that can evolve into phenotypes that adapt to the esophageal microenvironment. B) Phenotypic diversity of non-dysplastic glands is associated with the presence of dysplasia or cancer in patients with BE .
BibTeX
@misc{evans2021clonal,
  author = {James A. Evans and Emanuela Carlotti and Meng‐Lay Lin and Richard Hackett and Adam M. Passman and Lorna Dunn and George Elia and Ross J Porter and Mairi H McClean and Frances Hughes and Joanne Chin‐Aleong and Philip Woodland and Sean Preston and S M Griffin and Laurence Lovat and Manuel Rodriguez‐Justo and Nicholas A. Wright and Marnix Jansen and Stuart A.C. McDonald},
  title = {{Clonal transitions and phenotypic evolution in Barrett esophagus}},
  year = {2021},
  doi = {10.1101/2021.03.31.21252894},
  howpublished = {medRxiv},
}
Figure