Nucleosome Spacing--The Baggage Tag for Cell-Free DNA.
The field of liquid biopsies is still young, and a recent development in the laboratory of Dr. Jay Shendure at the University of Washington brings a new dimension to uncovering the tissue of origin of cfDNA (3). Clinical Chemistry spoke with Shendure about his latest work.
What Is the Innovation?
Every cell in the human body, with some exceptions, has the exact same package of DNA. How then is a hepatocyte so different in form and function from a neuron? One answer lies in epigenetics--the selective control of specific genes in the development of an organism. And to regulate genes that a tissue type may or may not need, its DNA is pack aged and bound in different ways by proteins called histones, with each unit of histones and tightly coiled DNA termed a nucleosome. It is the selective regulation of genes that gives the hepatocyte its unique identity. One of the fingerprints of this regulation is the spacing of nucleosomes.
As a cell dies, cleaving enzymes break up its DNA. But what about DNA guarded by a nucleosome? Could we expect a predominance of intact DNA sequences from nucleosome-bound DNA? Based on an observation that cfDNA samples cluster around certain fragment sizes graduate student Matthew Snyder brought that question one day to Shendure.
The state of the art in cfDNA analysis uses sequencing to profile the DNA for notable mutations, and to report on its origins. For example, a prenatal screen can use paternal sequences to probe for cell-free fetal DNA. Cancer diagnostics can use the unique sequences of tumors to label tumor cfDNA. This does not work for all applications of cfDNA. Sequencing is no good for detecting a pathological leak of healthy tissue into the circulation, for example after a myocardial infarction. Today, specialized antibodies are applied in laboratory assays to detect tissue-specific biomarkers. In the future, Shendure's work could lead to a versatile assay that detects cfDNA and notes its tissue of origin without the need for elaborate sequencing. The first step in that direction is to make an initial nucleosome map to serve as a reference.
How Do They Create Their Map?
Shendure and colleagues apply computational genomics to build an effective "bioinformatic particle sorter" that allows "selective analysis of tumor DNA," according to Dr. Cees Oudejans of the Department of Clinical Chemistry at the VU University Medical Center Amsterdam.
Shedure's team sequences cfDNA with a hypothesis that the fragments of DNA that are recovered intact are bound (and thereby protected) by histones or transcription factors (Fig. 1). An advantage with this approach is that it depends on endogenous DNA digestion, and so provides a physiological picture.
Shendure describes the sequence they generate to be an analog signal that is "fuzzy," with boundaries of fragments that are not precise. They apply 3 elegant steps to make sense of these data. First, they
compute a windowed protection score (WPS), then apply a fast Fourier transform (FFT), and finally generate a correlation rank using a reference data set. To calculate the WPS, they perform deep sequencing of the cfDNA and count the number of fragments that fit within a 120-bp window centered on a nucleotide. From this number they reduce the number of fragments with 1 end in the window. The WPS represents the protection that a DNA fragment could receive from being bound to a nucleosome or transcription factor. Quantitatively, a high WPS maps to the location of a potential nucleosome or, less commonly, a transcription factor.
The spacing between nucleosomes is broader, they found, when the nucleosomes are in DNase I hypersensitivity sites, presumably because of chromatin remodeling due to transcription factor binding. And here lies the key to their analysis. The enhancement in spacing depends on the cell type. Using a reference data set of 76 human cell lines and primary tissues, the scientists classify their cfDNA to a particular cell type. A practical challenge with cfDNA is that it contains DNA from multiple cell types. To separate the signals, they use an FFT that converts the WPS spacing data into the frequency domain. The peaks in the FFT correspond to different cells of origin with unique spacing repeats. Ultimately Shendure and his group created a dense nucleosome map represented by 14.5 billion fragments and 700-fold coverage.
Where Can This Fit in the Clinical Laboratory?
Shendure calls this a "proof of concept" and in his report (3) states that the "goal is not necessarily to outperform the sensitivity of mutation-based monitoring of circulating tumor DNA. Rather, we envision that a unique application of this approach may be to noninvasively classify cancers at time of diagnosis by matching the epigenetic signature of cfDNA fragmentation patterns against reference datasets corresponding to diverse cancer types." He suggests it could have use in classifying tumors of unknown origins.
Oudejans believes Shendure's work "is a major step forward in cancer diagnostics" and through its in silico cell sorting will lead to an "increase in assay specificity and sensitivity as seen in prenatal diagnostics."
"Most importantly," he says, "it explores an additional layer of biological information in the human (epi)genome that can be analyzed in a noninvasive, genome-wide manner. The biological information in the cell is a 3-dimensional layer of variations consisting of unique combinations of DNA sequences (polymorphism, mutation), DNA modifications (methylation), proteins bound (transcription factors), their modifications (acetylation, methylation), and RNA bound (R-loops)." He believes that in combination with assays exploring the additional layers (4), Shendure's work can generate "personalized, patient-specific, noninvasive diagnostic assays for cell dysfunction as in cancer."
What will it take to move beyond this proof of concept stage? "The time has come to find the hole in the boat before the boat starts sinking and is sending SOS signals," urges Oudejans. "Large, international prospective studies are needed with exploration of cfDNA as well as RNA in presymptomatic healthy individuals in their forties, long term follow-up, and the use of multiple genome-wide assays that address each layer of biological information."
He says that this work shows that "the technology needed is not the limiting factor anymore."
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.
(1.) Yu SC, Chan KC, Zheng YW, Jiang P, Liao GJ, Sun H, et al. Size-based molecular diagnostics using plasma DNA for noninvasive prenatal testing. Proc Natl Acad Sci U S A 2014; 111:8583-8.
(2.) Straver R, Oudejans CB, Sistermans EA, Reinders MJ. Calculating the fetal fraction for noninvasive prenatal testing based on genome-wide nucleosome profiles. Prenat Diagn 2016; 36:614-21.
(3.) Snyder MW, Kircher M, Hill AJ, Daza RM, Shendure J. Cell-free DNA comprises an in vivo nucleosome footprint that informs its tissues-of-origin. Cell 2016; 164:57-68.
(4.) Sun K, Jiang P, Chan KC, Wong J, Cheng YK, Liang RH, et al. Plasma DNA tissue mapping by genome-wide methylation sequencing for noninvasive prenatal, cancer, and transplantation assessments. Proc Natl Acad Sci U SA2015; 112:E5503-12.
Vikram S. Kumar  * and Molly Webster 
 Dimagi, Boston, MA; 2 Science writer and producer, Brooklyn, NY.
* Address correspondence to this author at: Dimagi, 390 Commonwealth Ave., Apt. 605,
Boston, MA, 02215. E-mail firstname.lastname@example.org.
Received August 26,2016; accepted August 29,2016.
 Nonstandard abbreviations: cfDNA, cell-free DNA; WPS, windowed protection score; FFT, fast Fourier transform.
Caption: Jay Shendure
Caption: Cees Oudejans
Caption: Fig. 1. cfDNA in plasma after endogenous chromatin digestion that spares DNA bound to nucleosomes and transcription factors.
Image courtesy of Matthew Snyder. Reproduced with permission.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Technology Corner|
|Author:||Kumar, Vikram S.; Webster, Molly|
|Date:||Nov 1, 2016|
|Previous Article:||Antimicrobial Resistance-Could Less Really Be More?|
|Next Article:||Of Doctors, Artists, and Academia: Andrzej Szczeklik.|