DNA barcoding reveals diverse growth kinetics of human breast tumour subclones in serially passaged xenografts.

Long V Nguyen,David J H F Knapp,Martin Hirst,Sohrab Shah,Claire L Cox,Samuel Aparicio,Annaick Carles,Connie J Eaves,Nagarajan Kannan,Davide Pellacani,Peter Eirew,Michelle Moksa,Sneha Balani

doi:10.1038/ncomms6871

Long V Nguyen, David J H F Knapp + Show 11 more

Open Access

https://doi.org/10.1038/ncomms6871

Copy DOI

Abstract

Genomic and phenotypic analyses indicate extensive intra- as well as intertumoral heterogeneity in primary human malignant cell populations despite their clonal origin. Cellular DNA barcoding offers a powerful and unbiased alternative to track the number and size of multiple subclones within a single human tumour xenograft and their response to continued in vivo passaging. Using this approach we find clone-initiating cell frequencies that vary from ~1/10 to ~1/10,000 cells transplanted for two human breast cancer cell lines and breast cancer xenografts derived from three different patients. For the cell lines, these frequencies are negatively affected in transplants of more than 20,000 cells. Serial transplants reveal five clonal growth patterns (unchanging, expanding, diminishing, fluctuating or of delayed onset), whose predominance is highly variable both between and within original samples. This study thus demonstrates the high growth potential and diverse growth properties of xenografted human breast cancer cells.

Highlights

Genomic and phenotypic analyses indicate extensive intra- as well as intertumoral heterogeneity in primary human malignant cell populations despite their clonal origin
We demonstrate the highly complex clonal growth dynamics that barcoding reveals in serially passaged xenografts of widely used cell lines originally derived from established human mammary cancers and related complexity in barcoded transplants of cells more immediately derived from primary human breast cancers
The controls allowed us to set a conservative threshold for clone detection at a fractional read value (FRV) equivalent to 300 cells, based on the finding that this FRV allowed clones containing more than 500 cells to be detected at 90% sensitivity (9 of 10 control samples detected, with an inferred detection efficiency of 100% for clones of more than 5,000 cells) and an associated specificity of 99.7%

Summary

Introduction

Genomic and phenotypic analyses indicate extensive intra- as well as intertumoral heterogeneity in primary human malignant cell populations despite their clonal origin. Cellular DNA barcoding offers a powerful and unbiased alternative to track the number and size of multiple subclones within a single human tumour xenograft and their response to continued in vivo passaging Using this approach we find clone-initiating cell frequencies that vary from B1/10 to B1/10,000 cells transplanted for two human breast cancer cell lines and breast cancer xenografts derived from three different patients. One source of heterogeneity is believed to derive from an incomplete suppression of the process of differentiation that takes place in the normal adult mammary gland to regulate its composition and size[1] This process includes the exclusive retention by rare subsets of cells of the property of self-renewal; that is, the ability to generate daughter cells that are not destined to die nor become unable to proliferate further by the activation of a terminal differentiation programme[2]. We demonstrate the highly complex clonal growth dynamics that barcoding reveals in serially passaged xenografts of widely used cell lines originally derived from established human mammary cancers and related complexity in barcoded transplants of cells more immediately derived from primary human breast cancers

Methods

Results

Conclusion