Patent Application
20210017611
This present disclosure relates generally to the field of biotechnology, specifically to genetic biomarkers that are associated with human cancers, and more specifically to methods and kits for identifying hepatitis B virus (HBV)-host junction sequences in tissue or body fluid samples and their use in screening, diagnosis, monitoring, management, and therapy of for hepatocellular carcinoma (HCC).
Chronic hepatitis B virus (HBV) infection remains a global health burden despite the availability of a preventive vaccine, affecting more than 240 million people worldwide and associated with more than 600,000 deaths annually. HBV is the major etiology of hepatocellular carcinoma (HCC), associating with over 50% of HCC cases worldwide and up to 70-80% of cases in HBV-endemic areas such as sub-Sahara Africa and Asian countries. HCC is the fifth most common cancer worldwide and the most frequent cancer in certain parts of the world. HCC surveillance programs have been implemented to screen high-risk populations, including HBV-infected individuals, for the early detection of HCC. Regardless of these efforts, most cases of HCC remain undetected until late stages, resulting in poor prognosis. The current lack of a sensitive and convenient screening method provides an urgent need for improved early detection strategies of HCC.
During the course of infection, the HBV genome can integrate into the host chromosome. Integrated DNA was detected in more than 85% of HBV related HCC cases (HBV-HCC). Although it is known that viral breakpoints predominately occur in the DR1-2 region of the HBV genome, the integrated sites in the host DNA vary. Thus, each HBV integration event generates a unique HBV-host junction sequence (HBV-JS) that essentially creates a fingerprint of each infected hepatocyte. Thus, HBV-JSs can be used as a unique marker to trace for the HBV-HCC DNA that is released into the circulation and filtered into urine as fragmented LMW DNA.
Circulating cell-free DNA (cfDNA) has been identified in biological fluids. For example, in urine, two species are seen: a high-molecular-weight (HMW) DNA, greater than 1 kb, derived mostly from sloughed off cell debris from the urinary tract, and a low-molecular-weight (LMW) DNA, approximately 150 to 250 base pairs (bp), derived primarily from apoptotic cells.
Methods for analysis of HBV-JS fingerprints (integration sites) from genomic DNA have become readily accessible to researchers through the increasing availability of high-throughput next generation sequencing (NGS). As tools to identify viral integration sites have emerged, they are not entirely appropriate to the majority of scientific researchers, as they are not packaged in an intuitive interface, are time intensive, and are not entirely accurate. Thus, there remains a need to provide widespread accessibility to a method that enables users to accurately identify HBV-JSs in a time sensitive manner.
In a first aspect, the present disclosure provides a method for identifying at least one HBV-host junction sequence (HBV-JS) from a biological sample of a subject.
The method includes the following steps: (1) preparing a DNA sample from the biological sample; and (2) performing at least one round of enrichment over the DNA sample. Each round of enrichment in step (2) includes a sub-step of capturing HBV DNA sequence-containing DNA molecules from the DNA sample by means of an HBV probe set. The HBV probe set includes a plurality of HBV primers (also called HBV probes) having sequences thereof selectively and respectively corresponding to different regions of an HBV genome, and each HBV primer is labelled with an immobilization portion configured to allow immobilization onto a solid support.
Herein, the subject can be a primate such as a human, a monkey, a chimpanzee, a gorilla, etc. The biological sample can be a tissue sample such as a tissue biopsy sample or a liver cell line sample, and the biological sample can be a fluid sample, selected from a group consisting of a saliva sample, a nasopharyngeal sample, a blood sample, a serum sample, a plasma sample, gastrointestinal fluid, a bile sample, a cerebrospinal fluid sample, a pericardial sample, a vaginal fluid sample, a seminal fluid sample, a prostatic fluid sample, a peritoneal fluid sample, a pleural fluid sample, a synovial fluid sample, an interstitial fluid sample, an intracellular fluid sample, a cytoplasm sample, a lymph sample, a bronchial secretion sample, a mucus sample, a vitreous tumor sample, an aqueous humor sample, saliva sample, and a urine sample. Preferably the biological sample is a plasma sample, and more preferably, it is a urine sample, and under this latter circumstance, the method disclosed in this application allows for non-invasive detection of HBV-JSs so to provide important information regarding the screening, diagnosis, maintenance, prognosis, and management of HBV-associated HCC.
Herein, the plurality of HBV primers are configured to contain sequences therein that selectively and respectively corresponding to different regions of an HBV genome. To be more specific, each HBV primer can be designed to have a sequence that correspondingly matches with a particular HBV genomic region (e.g. having a sequence that may be at least 90% homologous with a sense strand or an anti-sense strand of the HBV genomic region) while having minimum homology with any host genomic region such that the each HBV primer can selectively hybridize with a sequence of a DNA molecule that corresponds to the HBV genomic region, thereby providing a means to selectively capture the HBV DNA sequence-containing DNA molecule. It is noted that the sequence homology between one HBV primer and its target HBV genomic sequence does not have to be 100% identical, as long as the hybridization therebetween is secure and strong enough to allow the specific capture of the target DNA molecule under an appropriate condition.
Herein, the HBV DNA sequence-containing DNA molecules can include DNA molecules that harbor a chimeric polynucleotide that includes both a host genomic DNA portion and an HBV genomic DNA portion (i.e. a host genome-integrated HBV genomic DNA), and can also include a polynucleotide whose sequence is purely HBV's.
In the method, the sub-step of capturing, by means of an HBV probe set, HBV DNA sequence-containing DNA molecules from the DNA sample can be through a primer extension capture (PEC) assay, which comprises:
denaturing the DNA sample to thereby obtain a denatured DNA sample by, e.g., heating at 95° C. for several minutes;
contacting the plurality of HBV primers with the denatured DNA sample for annealing by, e.g., incubating at an appropriate temperature;
performing a primer extension reaction by, e.g., polymerization;
immobilizing the DNA molecules captured by the plurality of HBV primers; and
eluting the DNA molecules.
According to some embodiments of the method, each round of enrichment can further include a sub-step of amplifying the DNA molecules, which can be realized by PCR-based approach using appropriate primers:
In any of the embodiments of the method described above, each of the plurality of HBV primers comprises a sequence selected from a group consisting of SEQ ID NOS: 49-175. In other words, the HBV probe set or HBV probe panel includes a set of HBV primers that represent part of a whole list of the SEQ ID NOS: 49-175. More preferably, the HBV probe set include all of the 127 sequences in SEQ ID NOS: 49-175 to thereby provide a comprehensive coverage to substantially cover the entire HBV genome. Furthermore, each of the plurality of HBV primers in the HBV probe set is configured to selectively target a different region of the HBV genome, such that this particular HBV primer can hybridize with a corresponding HBV DNA fragment integrated to the host genome while having minimum level of off-target effect to the host genome so as to provide a means for the specific capture and enrichment of the DNA molecules containing the HBV DNA sequence.
According to some embodiments of the method, the step (1) of preparing a DNA sample from the biological sample comprises: constructing a DNA library from the biological sample. Herein, the DNA library can optionally be a double-stranded DNA (dsDNA) library, yet according to some other more preferred embodiments, the DNA library is an ssDNA library, allowing the capture and enrichment of not only both ssDNA and dsDNA molecules, but also the short fragmented DNA molecules (e.g. <150 bp), which are commonly found in cell-free DNA samples obtained from a liquid biopsy sample such as a urine sample or a plasma sample.
Optionally for the method disclosed herein, a number of the at least one round of enrichment can be more than one. In other words, in the method described above, more than one round of enrichment (i.e. step (2)) can be performed so as to increase the enrichment efficiency.
In the method, in step (1) of preparing a DNA sample from the biological sample, each DNA molecule obtained thereby comprises a pair of adaptors flanking a DNA fragment from the subject. Accordingly, in the sub-step of capturing, by means of an HBV probe set, DNA sequences comprising the at least one HBV-JS through a primer extension capture (PEC) assay, the DNA sequences are captured in presence of adaptor blockers which are configured to hybridize with the pair of adaptors so as to minimize off-target capture.
In the method, the PEC assay relies on the immobilization portion labelled on each of the plurality of HBV primers for the capture and enrichment of target DNA molecules, such that the immobilization portion can form a stable binding with a coupling partner conjugated onto surface of the solid support.
Such binding can optionally be non-covalent. For example, the immobilization portion can comprise a biotin moiety, and correspondingly, the coupling partner conjugated onto surface of the solid support can comprise at least one of streptavidin, avidin, or an anti-biotin antibody. Other examples of the immobilization portion-coupling partner pair can include, but is not limited to, a carbohydrate-lectin pair, an antigen-antibody pair and a negative charged group-positive charged group static interacting pair.
According to some other embodiments of the method, the immobilization portion can be configured to be able to form a covalent connection (or crosslinking) with a coupling partner conjugated onto surface of the solid support. As such, the immobilization portion and the coupling partner can respectively be one and another of a cross-linking pair. Examples of the cross-linking pair include an NHS ester-primary amine pair, a sulfhydryl-reactive chemical group pair (e.g. cysteines, or other sulfhydryls such as maleimides, haloacetyls, and pyridyl disulfides), an oxidized sugarhydrazide pair, photoactivatable nitrophenyl azide's UV triggered addition reaction with double bonds leading to insertion into C—H and N—H sites or subsequent ring expansion to react with a nucleophile (e.g., primary amines), or carbodiimide activated carboxyl groups to amino groups (primary amines), etc. The solid support can comprise at least one of a magnetic bead, a filter, a resin bead, a nanosphere, a plastic surface, a microtiter plate, a glass surface, a slide, a membrane, a microfluidic channel, a chip, or a matrix. Preferably, the immobilization portion labelled on each HBV primer in the HBV probe set is a biotin moiety; and the solid support comprises streptavidin magnetic beads.
The method may further include, after the at least one enrichment in step (2), steps of: (3) sequencing the DNA sequences; and (4) identifying the at least one HBV-JS. Herein, step (4) of identifying the at least one HBV-JS can be done through ChimericSeq.
In a second aspect, the present disclosure further provides a kit for identifying at least one HBV-host junction sequence (HBV-JS) from a biological sample of a subject, which can be utilized in implementing the method as described above.
The kit includes an HBV probe set, which comprises a plurality of HBV primers having sequences thereof selectively and respectively corresponding to different regions of an HBV genome, and each HBV primer is labelled with an immobilization portion. The kit further includes a solid support, which is conjugated with a coupling partner on a surface thereof, wherein the coupling partner is configured to form a secure coupling to the immobilization portion of each HBV primer to thereby allow immobilization of HBV DNA sequence-containing DNA molecules to the solid support.
According to some embodiments of the kit, each of the plurality of HBV primers comprises a sequence selected from a group consisting of SEQ ID NOS: 49-175. More preferably, the HBV primers included in the HBV probe set include HBV primers that cover all of the 127 HBV sequences as set forth in SEQ ID NOS: 49-175.
According to some embodiments, the kit can further include a pair of adaptors, which are configured to be ligated to two ends of each DNA molecule in the biological sample to thereby obtain a DNA library from the biological sample. Further optionally, the kit can further include at least one adaptor blocker, which is configured to hybridize with sequences corresponding to the pair of adaptors in the each DNA molecule in the DNA library so as to minimize off-target capture.
Herein, the DNA library can be a double-stranded DNA library, but more preferably can be a single-stranded DNA library.
Optionally, the kit can further include at least one pair of amplifying primers, configured to amplify the HBV DNA sequence-containing DNA molecules.
In the kit, the immobilization portion can comprise a biotin moiety, and the coupling partner comprises at least one of streptavidin, avidin, or an anti-biotin antibody. Preferably, the solid support comprises streptavidin magnetic beads.
The kit can further include a software for identifying the at least one HBV-JS from data obtained from a sequencing assay, and the software is preferably ChimericSeq.
In a third aspect, the present disclosure further provides a method for de novo identification of HBV-JS. The method comprises:
constructing a DNA library from a biological sample collected from a subject;
applying the kit and the method according to the various embodiments as described above to enrich for HBV DNA sequence-containing DNA molecules;
sequencing the enriched DNA molecules and analyzing a sequencing result; and
if the sequencing result shows that a particular HBV-JS does not match with re-curated HBV-JS in a database, depositing the HBV-JS in the database.
In a fourth aspect, the present disclosure further provides a method for identification of an HBV-related HCC driver gene, or to be more specific, for determining if a candidate HBV-JS is a potential HCC driver. The method comprises:
applying the kit and method as described above to enrich and sequence HBV DNA sequence-containing DNA molecules from a DNA sample obtained from a population of subjects;
determining, if a sequencing result indicates that an HBV-JS is recurrent, that the HBV-JS is a candidate HBV-related HCC driver.
In any of the above methods, the biological sample can be a tissue sample or a liquid sample (e.g. urine sample), and the DNA library is preferably an ssDNA library.
In a fifth aspect, the present disclosure further provides a method for evaluate a risk of a subject for HBV-associated HCC. The method comprises:
collecting a biological sample from the subject;
constructing a DNA library from a biological sample;
applying the kit and method as described above to enrich and sequence HBV DNA sequence-containing DNA molecules in the DNA library;
identifying all HBV-JSs based on the sequencing result to thereby establish an HBV-JS profile for the subject; and
evaluating the risk of the subject for HCC based on the HBV-JS profile.
Herein, the biological sample can be any sample, but preferably a urine sample. The DNA library can be any type, but preferably an ssDNA library. The evaluating step can be based a multivariable analysis which includes, in addition to the HBV-JSs, other independent variables such as age, family history, pre-condition, etc.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art pertinent to the methods and compositions described. As used herein, the following terms and phrases have the meanings ascribed to them unless specified otherwise.
Various embodiments will be described in detail through the displayed figures. Reference to these embodiments does not limit the scope of the claims attached hereto. Provided examples are not meant to limit the scope of methods and claims herein, but rather describe example uses of the embodiments of the claims.
The terms “a,” “an,” and “the” as used herein include plural referents, unless the context clearly indicates otherwise.
The term “genome” and “genomic” refer to any nucleic acid sequences (coding and non-coding) originating from any living or non-living organism or single-cell. These terms also apply to any naturally occurring variations that may arise through mutation or recombination through means of biological or artificial influence. An example is the human genome, which is composed of approximately 3×109 base pairs of DNA packaged into chromosomes, of which there are 22 pairs of autosomes and 1 allosome pair.
The term “nucleotide sequence” as used herein indicates a polymer of repeating nucleic acids (Adenine, Guanine, Thymine, and Cytosine, and Uracil) that is capable of base-pairing with complement sequences through Watson-Crick interactions. This polymer may be produced synthetically or originate from a biological source.
The term “nucleic acid” refers to a deoxyribonucleotide (DNA) or ribonucleotide (RNA) and complements thereof. The size of nucleotides is expressed in base pairs “bp”. Polynucleotides are single- or double-stranded polymers of nucleic acids and complements thereof.
The term “deoxyribonucleic acid” and “DNA” refer to a polymer of repeating deoxyribonucleic acids.
The term “ribonucleic acid” and “RNA” refer to a polymer of repeating ribonucleic acids.
The term “disease” or “disorder” is used interchangeably herein, and refers to any alteration in state of the body or of some of the organs, interrupting or disturbing the performance of the functions and/or causing symptoms such as discomfort, dysfunction, distress, or even death to the person afflicted or those in contact with a person. A disease or disorder can also relate to a distemper, ailing, ailment, malady, disorder, sickness, illness, complaint, or affectation.
As used herein, “cancer” refers to any stage of abnormal growth or migration of cells or tissue, including precancerous and all stages of cancerous cells, including but not limited to adenomas, metaplasias, heteroplasias, dysplasias, neoplasias, hyperplasias, and anaplasias.
As used herein, “cancer progression” refers to any measure of cancer growth, development, and/or maturation including metastasis. “Cancer progression” includes increase in cell number, cell size, tumor size, and number of tumors, as well as morphological and other cellular and molecular changes and other characteristics. As an example, one measure of cancer progression is the use of staging characteristics. As an additional example, one measure of cancer progression is the use of detecting expression, whether at the protein or mRNA level, of certain genes
The term “diagnosing” means any method, determination, or indication that an abnormal or disease condition or phenotype is present. Diagnosing includes detecting the presence or absence of an abnormal or disease condition, and can be qualitative or quantitative.
The term “gene” is well known in the art, and herein includes non-coding region such as promoter or other regulatory sequences or proximal non-coding region.
The terms “express” and “produce” are used synonymously herein, and refer to the biosynthesis of a gene product. These terms encompass the transcription of a gene into RNA. These terms also encompass translation of RNA into one or more polypeptides, and further encompass all naturally occurring post-transcriptional and post-translational modifications. The expression/production of an antibody or antigen-binding fragment can be within the cytoplasm of the cell, and/or into the extracellular milieu such as the growth medium of a cell culture.
The term “biomarker” is an agent used as an indicator of a biological state. It can be a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention. A biomarker can be a fragment of genomic DNA sequence that causes disease or is associated with susceptibility to disease, and may or may not comprise a gene.
The term “low molecular weight” or LMW nucleic acid refers a nucleic acid, such as DNA, of less than 1000 base pairs, usually less than 300 base pairs.
The term “nucleotide amplification reaction” refers to any suitable procedure that amplifies a specific region of polynucleotides (target) using primers.
A “protein” is a macromolecule comprising one or more polypeptide chains. A protein may also comprise non-peptidic components, such as carbohydrate groups. Carbohydrates and other non-peptidic substituents may be added to a protein by the cell in which the protein is produced, and will vary with the type of cell. Proteins are defined herein in terms of their amino acid backbone structures; substituents such as carbohydrate groups are generally not specified, but may be present nonetheless.
The terms “amino-terminal” and “carboxyl-terminal” are used herein to denote positions within polypeptides. Where the context allows, these terms are used with reference to a particular sequence or portion of a polypeptide to denote proximity or relative position. For example, a certain sequence positioned carboxyl-terminal to a reference sequence within a polypeptide is located proximal to the carboxyl terminus of the reference sequence, but is not necessarily at the carboxyl terminus of the complete polypeptide.
The term “chimeric reads” herein refers to a nucleotide sequence obtained from next generation sequencing, whereby the length of the read contains genomic material from two separate biological entities or chromosomes joined covalently through integration. For example, viruses can integrate viral nucleotide sequences into the genomic nucleotide sequence of a human host.
Throughout the disclosure, the terms “probe set”, “probe panel”, or alike, are considered to be exchangeable, and the term “HBV primer” mentioned in the HBV probe set is also considered to be exchangeable to “HBV probe”.
Due to the imprecision of standard analytical methods, molecular weights and lengths of polymers are understood to be approximate values. When such a value is expressed as “about” X or “approximately” X, the stated value of X will be understood to be accurate to ±10%.
Provided herein include methods and kits that can provide a sensitive, specific, and noninvasive platform for detecting HBV-JS in circulating nucleic acid sequences from a biological sample including body fluid or HBV-infected liver tissue DNA. Any HBV-JS DNA found in cell-free DNA isolated from a patient's body fluid can be used because it is representative of liver-derived DNA. The methods use a biotinylated HBV primer extension to enrich for HBV sequences of library DNA. The enriched libraries were analyzed for HBV-JS by NGS. As shown in the following examples, the methods are useful for HCC screening and monitoring of HBV-infected individuals. This method is particularly useful for high-risk HCC individuals and individuals with occult HBV infection to undergo frequent noninvasive screening to monitor disease progression, as they are often asymptomatic.
The present disclosure features at least the following three components used in developing an integrative HBV-JS analysis platform. First, a biotinylated HBV primer extension enrichment was used to enrich DNA samples for HBV DNA sequences that may contain HBV-JSs. Second, the enriched libraries are amplified by primers targeting all DNA template and sequenced by Illumina next generation sequencing platform. Lastly, the NGS data can be analyzed by ChimericSeq for identifying HBV-JSs, where the analysis results were successfully confirmed for an 87% validation rate (13/15).
Throughout the disclosure, the term “biological sample” can be deemed to comprise a tissue sample, such as a biopsy sample or a tissue culture sample. A biological sample may as well comprises biological fluids (i.e. liquid sample) including, but not limited to, saliva, nasopharyngeal, blood, plasma, serum, gastrointestinal fluid, bile, cerebrospinal fluid, pericardial, vaginal fluid, seminal fluid, prostatic fluid, peritoneal fluid, pleural fluid, urine, synovial fluid, interstitial fluid, intracellular fluid or cytoplasm and lymph, bronchial secretions, mucus, or vitreous or aqueous humor. Biological samples can also include cultured medium. In certain embodiments, the preferred biological fluid is urine, and in such cases, the method disclosed in this present application can be used to non-invasively detect HBV-JSs for HCC screening, cancer progression, and for HBV-HCC disease monitoring.
In certain embodiments, the platform uses biological samples containing fragmented circulation derived DNA known as “low molecular weight” (LMW) DNA. The DNA is low molecule weight because it is generally less than 300 base pairs in size. This LMW DNA is released into circulation through necrosis or apoptosis by both normal and cancer cells. It has been shown that LWM DNA is excreted into the urine and can be used to detect tumor-derived DNA, provided a suitable assay, such as a short template assay for which detection is available (Su Y H et al. 2008).
The inventions disclosed herein have the advantage that the procedures provided are capable of screening for HBV-related hepatocellular carcinoma, where unique major HBV-JSs serve as a marker of uncontrolled clonal expansion.
The methods described herein can be used to determine the status of an existing disease identified in a subject. For example, 19 HCC, 21 hepatitis and 19 cirrhosis urine samples were evaluated for HBV-JSs, and all HCC urine samples with HBV-JSs contained only integrated HBV sequences in the DR1-2 region, a higher load of HBV-JS, and a reduced HBV-JS complexity compared to non-HCC patients. Thus, the HBV-JS load and HBV-JS species detected in urine can be used to screen for HBV-HCC and monitor HBV related disease.
The methods described herein can be used to identify subject patients for treatment and to determine risk factors associated with HBV-JSs. Such methods can include, for example, determining whether an individual has relatives who have been diagnosed with a particular disease. Screening methods can also include, for example, conventional work-ups to determine familial status for a particular disease known to have a heritable component. Screening may be implemented as indicated by known patient symptomology, age factors, related risk factors, etc. These methods allow the clinician to routinely select patients in need of the methods described herein for treatment. In accordance with these methods, screening may be implemented as an independent program or as a follow-up, adjunct, or to coordinate with other treatments. Thus, the methods of the present inventions can be used for cancer screening, particularly for early detection, monitoring of recurrence, disease management, and to develop a personalized medicine regime for a cancer patient.
It is to be understood that the above described embodiments are merely illustrative of numerous and varied other embodiments which may constitute applications of the principles of the inventions disclosed herein. Other embodiments may be readily devised by those skilled in the art without departing from the spirit or scope of this invention and they shall be deemed within the scope of the disclosure.
The inventions provided in the disclosure is further illustrated by the following non-limiting examples.
In order to be able to reliably detect major HBV-JSs in urine samples, a biotinylated HBV primer extension enriched NGS assay was first developed. The following protocol was used: approximately 50-200 ng of tissue DNA was fragmented by sonication and subjected to NGS library DNA preparation as described by Ding et al. 2012 with minor modifications including 10 cycles of library DNA amplification (SEQ ID NO: 1, 2, 3) using Herculase II Fusion polymerase (Agilent Technologies, Santa Clara, Calif.). All the oligo sequences and reaction conditions for library preparation are listed in Table 1. To enrich for DNA that contains HBV DR1-2 DNA sequences, a multiplex biotin HBV primer extension reaction was performed using amplified library DNA in a reaction containing 1× Herculase II Buffer, 250 μM dNTP, and 20 pmol of biotinylated HBV primers as listed in Table 2. The reaction was held at the condition of 95° C. 2 mins, then 55° C. for 5 hrs with rotation. After a 5 hr incubation, 0.2 μl of heat inactivated Herculase II Fusion polymerase was added to each reaction and incubated at 55° C. for another 30 mins, followed by 72° C. for 90 s. The primer extended DNA was collected by using hydrophilic streptavidin magnetic beads (New England Biolabs, Ipswich, Mass.) as described by Gnirke et al. 2009 and used as the template in an indexing PCR (SEQ ID NO: 4 and 5) to add a unique barcode to each patient sample. Each indexed library was quantified and pooled accordingly for one NGS. NGS was performed to generate 150 bp paired-end reads on the Illumina MiSeq platform (Penn State Hershey Genomics Sciences Facility at Penn State College of Medicine, Hershey, Pa.). Sequences were analyzed using the ChimericSeq software to identify HBV-JSs.
Using the above developed approach, detection of major HBV-JSs derived from HBV-HCC tissue in matched tissue and urine samples was carried out.
Furthermore, the urine samples of HBV-infected patients are tested for the detection of HBV-JSs.
In order to be able to efficiently detect chimeric reads in the nucleotide sequence data, a software package ChimericSeq is developed.
In order to evaluate the detection efficiency of integration events with defined lengths of HBV insert, random HBV fragments of specified lengths (0-100 bp) were joined to random human genomic DNA of 100 bp. As shown in
A primer extension capture (PEC) approach for the HBV enrichment has been developed, whose schematic for only one target HBV-host junction sequence (HBV-JS, i.e. a chimeric DNA sequence containing a human genomic DNA and an integrated HBV DNA fragment) is illustrated in
It is noted that
With a purpose to provide a broader coverage, an optimized probe panel is further developed, which includes a total of 127 probes (Table 3) covering the most frequent four genotypes (A-D) of HBV and covering the entire HBV genome, is further developed. Briefly, to design an HBV probe panel with high specificity and sensitivity for application in an HBV primer-extension capture (PEC) approach, a human micro-homology analysis was first performed to identify regions within the HBV genome that are highly homologous to the human genome. The analysis was done by performing an NCBI BLAST query to the human genome for every 50 bp increments of HBV DNA along the entire 3.2 kb genome. The analysis uncovered 142 human micro-homologous stretches of HBV DNA ranging from 10-30 bp (average size of 19.6 bp) with melting temperatures (Tm) as high as 65° C. A total of 127 HBV probes were next designed to target the antisense strand along the entire HBV genome for genotypes A-D that avoided these human micro-homologous stretches. When it was not possible to avoid human micro-homologous stretches containing a Tm of 55° C. or less, the HBV primer was designed to target the HBV sense strand to ensure full HBV genome coverage during the enrichment.
GACRAG
In Examples 1-3, all the HBV enrichment experiments, if any, were performed based on the double-stranded DNA (dsDNA) library construction. Out of curiosity, a similar enrichment experiment based on the single-stranded DNA (ssDNA) library construction, was also carried out, and compared with a parallel enrichment experiment based on dsDNA library construction from the same biological sample. Briefly, cell-free DNA (cfDNA) samples isolated form liquid biopsy specimens (urine) from different patient samples, was utilized for both ssDNA and dsDNA library construction, which then underwent HBV enrichment, and NGS sequencing analysis. For ssDNA library construction, the ClaretBio SRSLY™ PicoPlus DNA NGS Library Preparation Dual UMI Index kit was utilized where a critical DNA denaturing step is performed as the initial step. All other subsequent steps were performed in accordance with the manufacturer's protocol. For library construction of double-stranded DNA, the Takara SMARTer® ThruPLEX® Tag-seq kit was utilized and performed according to the manufacturer's protocol.
Unexpectedly, a significantly improved HBV (on-target) enrichment was observed in urine samples utilizing single-strand DNA library construction compared with the same urine samples utilizing double-strand DNA library construction (Table 4). While both methods have obtained a similar level of total NGS reads (
In order to evaluate the performance of the optimized HBV probe panel (n=127, shown in Table 10) relative to the initial HBV probe panel (n=43, shown in Table 1), enrichment analysis was carried out using reconstituted PLC HCC cell-line DNA containing known integrated HBV sequences, where normal DNA samples containing 1%, 0.5%, and 0.1% PLC genomic DNAs were compared for sensitivity and specificity evaluation, and the results are shown in Table 5. After two sequential primer-extension capture (PEC), both panels demonstrate ˜105-fold enrichment compared to whole genome sequence of 100% PLC (no enrichment).
The optimized HBV panel was also examined for its performance in detecting known HBV-junctions (such as HBV junction at TERT, CCDCl57 and MVK). As shown in Table 6, the optimized panel showed a better performance, and can detect additional junction reads compared to the initial panel when the number of NGS reads are similar.
In order to further evaluate whether an increased number of PEC enrichment can improve the enrichment result, a comparison experiment was carried out, which compare the two sequential PEC enrichment with three sequential PEC enrichment. Briefly, the workflow of a sequential PEC enrichment is illustrated in
First, reaction containing buffer, blockers, dNTP and library DNA was incubated at 95° C. for 5 mins to denature double-strand library DNA and facilitate binding of adapter blockers to prevent daisy chaining during enrichment. Next, the reaction was held at 72° C. for 5 mins before adding the biotinylated HBV primer mix to the reaction. The entire reaction was incubated at 60° C. for 1 hr. Lastly, 0.1 μl of heat inactivated Herculase II Fusion polymerase was added to each reaction and incubated at 72° C. for 90 s. The captured DNA was collected by using hydrophilic streptavidin magnetic beads (New England Biolabs, Ipswich, Mass.), washed twice at 55° C. using 5 mM TrisHCl pH 7.5, 0.5 mM EDTA, 1M NaCl buffer. Captured library DNA was eluted using 10 μl 0.1N NaOH and neutralized with 40 μl 1M Trish-HCl pH7.5. Prior to post-enrichment amplification, eluted library DNA was purified using 1.8× AMPure XP beads. Library DNA amplification post-enrichment utilized 1× Herculase II Buffer, 250 μM dNTP, and 30 pmol of P5/P7 Illumina adapter primers, and 0.3 μl of Herculase II Fusion polymerase. Reaction was performed at 98° C. 2 mins, 98° C. 30 s, 60° C. 30 s, 72° C. 1 min for 10 cycles followed by 72° C. extension for 10 mins. Amplified library DNA was purified using 1.8× AMPure XP beads. Following purification, subsequent enrichments can be performed by repeating the above procedures or library DNA can be quantified and sequenced. The comparison results are shown in Table 7.
Chronic hepatitis B virus (HBV) infection is a major etiology of hepatocellular carcinoma (HCC), associated with over 50% of cases worldwide and up to 70-80% of cases in HBV-endemic areas. High mortality of HCC is mainly due to late detection and limited treatment options. HCC surveillance programs have been implemented to screen HBV-infected individuals, to facilitate earlier detection of HCC. Unfortunately, most cases of HBV-related HCC (HBV-HCC) remain undetected until late stages resulting in poor prognosis, due to lack of a sensitive and convenient screening method. In the past years, over 100 clinical trials for HCC therapy failed, Sorafenib, with a limited efficacy, remains the only available chemotherapy after its approval 9 years ago. Identification of HCC drivers has been suggested to be important for drug development and patient selection in clinical trial design due to high heterogeneity of the diseases (REF).
During the course of infection, HBV can integrate into the host chromosome, and this integrated viral DNA was detected in more than 85% of HBV-HCC. Although it is known that viral breakpoints predominately occur in the DR1-2 region of the HBV genome, the integration sites in the host DNA have been observed to vary. Thus, each HBV integration event generates a unique HBV-host integration site, which creates a specific fingerprint of each infected hepatocyte. During the tumorigenesis, uncontrolled clonal expansion can amplify this molecular signature becomes a major, most abundant, over other host junctions found in other noncancerous infected hepatocytes. Thus, the merging of this uncontrolled, clonally expanded major HBV-host junction can be a biomarker for carcinogenesis, and can be a biomarker for early detection of HCC if this major HBV-host junction can be detected in periphery.
In order to test the feasibility to detect HBV-host junctions in circulation, urine was resorted since it is limited, if any of virions thus facilitating detection of integrated HBV DNA. It has been shown that urine contains DNA from circulation that can be used for cancer detection if a tumor is present. Although HBV DNA has been detected in urine, it has not been entirely clear if HBV DNA detected in urine was derived from fragmented integrated DNA from infected liver. In this proof-of-concept study, a method is developed to prepare a DNA library for NGS enriched for HBV integration. Using this approach, identical, major HBV integration sites from matched HCC tissue and urine are detected, providing evidence that clonally expanded, integrated HBV DNA derived from the infected liver is present in the urine. Combining this data with other reports of HBV integration, it was found the recurrently targeted genes are mostly associated with carcinogenesis suggesting potential approach for HBV-HCC driver identification. In particular, the TERT gene seems to be highly targeted within a narrow range of the promoter region. Together, these results not only suggest the utility of urine as a body fluid to study HBV integration sites in circulation, but also describe a noninvasive means for potential HCC screening and genetic characterization.
Study subjects: the HCC tissue and urine samples used were obtained with written informed consent from patients at the National Cheng-Kung University Medical Center, Taiwan, in accordance with the guidelines of the Institutional Review Board. Detailed sample information is provided in Table 8.
DNA isolation, urine collection, and low molecular weight (LMW) urine DNA fractionation: Tissue DNA was isolated using the Qiagen DNeasy Tissue kit (Valencia, Calif.) according to the manufacturer's instructions. Urine samples were collected and total urine DNA was isolated as previously described (Su Y H et al. 2004). Cell-free DNA (<1 kb) was obtained from total urine DNA using carboxylated magnetic beads, as previously developed (Su Y H et al. 2008).
Preparation of HBV DR1-2 enriched library DNA for NGS: Tissue DNA was fragmented by sonication and subjected to Next-Generation Sequencing (NGS) library DNA preparation as described by Ding et al. 2012. This involved minor modifications, including 10 cycles of library DNA amplification using Herculase II Fusion polymerase (Agilent Technologies, Santa Clara, Calif.). To enrich for DNA that contains HBV DR1-2 sequences, a multiplex biotin HBV primer extension reaction was performed using amplified library DNA in a reaction containing 1× Herculase II Buffer, 250 μM dNTP, and 20 pmol of biotinylated HBV primers. The primer-extended DNA was collected, as described by Gnirke et al. 2009, subjected to three individual nested HBV DR1-2 PCR enrichment reactions, and followed by an indexing PCR. Each indexed library was quantified and pooled accordingly for one NGS. NGS was performed to generate 150 bp paired-end reads on the Illumina MiSeq platform (Penn State Hershey Genomics Sciences Facility at Penn State College of Medicine, Hershey, Pa.).
Identification and characterization of HBV-JS sequences: NGS data was analyzed using JBS ChimericSeq software (http://www.jbs-science.com/ChimericSeq.php, Jongeneel et al. manuscript submitted) to identify integration sites and major integration sites. For all the major integration sites identified, the software provided the annotation of breakpoints for both the HBV genome and human genome, human genes within 100 kb of the breakpoints, the number of overlapping viral and human nucleotides at the junction site and the Tm of the overlapping sequences.
Short amplicon PCR assays: Short amplicon junction PCR was performed using Hotstart Plus Taq Polymerase (Qiagen, Valencia, Calif.), junction primers, and the LMW urine DNA templates. Junction PCR products were visualized on a 2.2% FlashGel DNA Cassette (Lonza Group, Basel, Switzerland) and subsequently subjected to either a nested PCR reaction using a set of inner primers, or a restriction endonuclease (RE) digestion using RE obtained from New England Biolabs (Ipswich, Mass.), per the manufacturer's specifications to further compare the PCR products derived between tissue and urine.
Results:
Development of an NGS Library Enrichment Method for HBV Integrations:
To directly enrich for HBV integrated DNA, a primer extension capture (PEC) approach was adopted to the HBV DNA libraries. In short, this technique uses 5′-biotinylated oligonucleotide primers to capture targeted regions, and then uses a DNA polymerase to extend the primers (
Identification of Major HBV Integration Sites from HCC Tumor Tissue Using PEC of HBV DR1-2:
In order to test whether the PEC approach was effective at enriching HBV integrated DNA from a biological sample, this technique was applied to an adapter-ligated tissue DNA library of 23 patients with chronic HBV infection and hepatocellular carcinoma (HCC). With the assumption that sampled tumors contain HBV integrated DNA at 1:1 ratio with human genomic DNA, A 10E4-fold enrichment would be necessary to obtain 1% HBV reads out of total reads. Through improving the specificity by PEC, it is able to obtain an average of 3.5% HBV reads of total NGS reads (data not shown).
Tumors are clonally expanded and most HBV-HCC tumors contain integrated HBV DNA (Ref), thus should contain at least one major, clonally expanded, HBV integration junction. In this study a major integration junction is defined as a distinctively identified sequence supported by at least 10% of the total HBV junction reads (minimum of 3 reads) within each DNA tissue sample. Reads containing HBV junctions were efficiently identified using the recently developed software program, ChimericSeq as described in Methods. The major HBV integration junctions identified in the NGS data by ChimericSeq are summarized in Table 9.
GGGCTGGGAGGGCCCGGAGGGGGCTGG (SEQ
CCCTCCCCTTCCTTTCCGCGGCC (SEQ ID NO:
CAGCTTTCTCTTCTTCTCTCTGTTTTTGTCTTGTTT
GGTGTGTTTCCTTGGGGTCATGG (SEQ ID NO:
AATTGTTGATACTCCAATAATATTAATTGCTAAG
ttttcTTATGAATGTTTTCTATATTTCAAAGCCCTGCT
CAAACACCACCTCCTCCAGAAAGGCTCCTGGTAT
CCTCTTTCTTTTCTAACCTAGAAAAGA (SEQ ID
GACAGGCCACCCCGCCATCGGCCATCTTCCTGG
CTCGCCCGGCCGCCCGCGCGCA (SEQ ID NO:
ACCATGTTGCCCAGGCTGGTCTTGAACACCTGGC
CTCAAGGGACGCTCCCAGC (SEQ ID NO: 185)
ACCACCTCCAGGCAGC (SEQ ID NO: 186)
ATCATTGTGTCAAACCTGGCACCGTGCCTGAAAC
ACAGTAGCCTCTCAATAAATA (SEQ ID NO: 187)
GGGGAATCTTGAGCCTTTGGCCACAGACTGAAG
GCTGCACTGTCAGCTTCCCTACTTTTGAGGCTTT
CG (SEQ ID NO: 188)
GCTCTCATCACTGCTGTAGAACAAAGTCCCACAT
AGAGCCAATGGCCAAGAACCAGTTAATAAAA
AGTTTGTGGGGAGGGGGTGAAATCGGGACTTCT
TCTAGCTGCCACGG (SEQ ID NO: 190)
CTTGCTGACTTACTATGACCTACTGGTAG (SEQ ID
GACATACCAGC (SEQ ID NO: 192)
CAAAGAATGGACTCAGAGACCCAGAGAACAACGA
AAGTGACGGTTTGTTCTT (SEQ ID NO: 193)
AACTCAGGTTTTCAACTAGTCTTACCATTGAAAGA
ACTATTGTGGCAAAGACGGAATG (SEQ ID NO:
AAACTGACCTTTGAATATCCGGATGCACGAGATT
CCCTGAAAGGGGAACAATAAATGT (SEQ ID NO:
AGGCATGAGCAGGGCAGGAGAGAACGCTCCCCT
GACTCACCAGGAATGTCAGGCAATCATTG (SEQ
GATGGGGATGTGGCAGTTGTGGACTGAAGTTGTA
CTGAGTGGTG (SEQ ID NO: 197)
CCGCCCTTCTCTGCCCAGCACTTTTCTGCCCCCC
TCCCTCTGGAACACAGAGTGGCAGTTTCCACAAG
CACTAAGCATCCTCTTCCCAAAAGACCCAGC
GTGGCATTGCAAGTGTACTGTTTAA (SEQ ID NO:
gacctgcatcatCCGGACTCCATAC (SEQ ID NO: 200)
TCCCTCGGCATGATGGGGCTGCTCCGG (SEQ ID
CTATTCTGTTCC (SEQ ID NO: 202)
ctgtggaggGAAGACTAAGTAGAGACGCGGATGTTT
ATGGCAGTGAAACTGTTC (SEQ ID NO: 203)
GAGTAGGCTCTGGAAATTGGAAGTGATCTTAGTA
TTTAAGTTCAGTCACTCAACTACAATCTCTGAAAC
ATGGCTTCCTCGCTGTCTTCCTGTGGTGGCACAC
GCCTGTAGTCCCAGCTACTTGGGAGGCTGAGGC
AGGAGAATTGCTTGAACC (SEQ ID NO: 205)
TAAACACAGGAGATATTTTTAAGCTTCACTCATAC
AGAAAATACA (SEQ ID NO: 206)
GCGCATGCTGGGAGTGTTCAATAAATACAGGCTG
AATGAATGAATGAACTGATGCATCCAAAACTT
CATATCAGCCCTGAGCAAGACAGCCAAACCAAAA
CAACCACAGCGAGGGATTCTGATTCCTTTGACAG
ACTCTGTTTCT (SEQ ID NO: 208)
TCCCCTTTCCCCACTGGTACAGGGTGGAGAGGT
C (SEQ ID NO: 209)
AGAAGGA (SEQ ID NO: 210)
cCTTGACTAAAGCCCATGGGCCA (SEQ ID NO:
To confirm the junction sequences obtained from the NGS analysis, PCR primers were designed for the major HBV integration junctions of 15 patients and performed amplification from the corresponding tissue DNA for Sanger sequencing. The respective tissue NGS library DNA was used as a positive control (+) for the junction sequence identified by NGS and HepG2 cell line DNA as a negative control (−) for each DNA tissue sample. Encouragingly, it was able to generate PCR products for 13 out of 15 of the tissue DNA samples tested. Only 2 of the 15 samples (patients 7 and 8) were unable to generate a PCR product using custom primers (data not shown). Further Sanger sequencing of each PCR product revealed matching HBV integrated sequences to their corresponding NGS-identified integration sequence, thus confirming the 13 samples. In total, it was able to validate 87% (13/15) of the major NGS identified HBV integration sites.
Detection of Tissue Identified Major HBV Integration Sites in Matched Urine:
Next it is examined whether major HBV integration sites can be detected in the circulation. As previously demonstrated, urine contains circulation derived DNA. The use of urine over serum collection is also advantageous, as it does not contain high amounts HBV DNA from virions in the circulation. In order to test the feasibility of detecting HBV integration junction sequences in the urine, seven patients (ID 9, 10, 11, 12, 13, 14, and 15) that have major HBV integration junctions identified by NGS study from this study were selected for this experiment based on the availability of matched urine DNA. For each major HBV integration site, primers were custom designed to amplify short products of less than 60 bp (illustrated in
Interestingly, for patient 7, the PCR product generated from urine DNA was larger than the one obtained from the tissue by PCR amplification (
Major HBV Integrations in HCC Recurrently Target TERT and CCNE1:
In HBV-infected individuals, integration into the host genome is thought to be random, having the potential to become oncogenic by insertional mutagenesis. The HBV DR1-2 sequences contain enhancer elements that may up-regulate host genes within a proximity of 100 kb, independently of position and orientation. With the identified locations of major HBV integration sites in HCC patients, host genes within 100 kb of these major sites were searched. ChimericSeq is used to identify the genes and positions of each breakpoint in both HBV and human genomes from the NGS data from tissue DNA. Out of the 34 major integration sites that were identified in 23 patients, 4 were not in a 100 kb proximity of a gene. Among these genes, TERT and CCNE1 were targeted in more than 1 patient; TERT was targeted in 5 of the 23 patients from this study, and CCNE1 was targeted in 3. Interestingly, both genes were found to be associated with carcinogenesis. Indeed, TERT is a suggested gatekeeper of hepatocarcinogenesis as the promoter region is frequently mutated in certain cancers. It thus was wondered whether identification of recurrent integration targeted genes could be a potential approach to identify drivers involved in hepatocarcinogenesis.
To explore this hypothesis, a meta-analysis of data reported from 15 studies, 446 patients, and 1554 HBV integrations was compiled. ChimericSeq was used again on this data set to identify genes within 100 kb of the integration site. From the 51 genes that were identified in at least 2 HCC patients, 12 were from at least two separate studies, defined as HBV integration recurrently targeted genes in HCC (
In alignment with this study of 23 HCC tumor tissues, TERT and CCNE1 were among the most common recurrent integration sites. Because of the presence of the most data for integrations near TERT, these 67 integration sites were compiled for further study. First, the location of the TERT integration breakpoints in the host genome was mapped against their locations in the HBV genome (
Promoter mutations and upstream rearrangements of the TERT gene including HBV integration, are known factors that drive carcinogenesis. It was of interest to investigate the distribution of these two events in the same tumor. The TERT promoter region of 20 of 23 tissue samples was successfully sequenced from the study, and identified 5 mutations of which 3 are of the major TERT hotspot mutation (˜124) (
Discussion:
This is the first study demonstrating that liver-derived HBV integration junction sequences can be detected in urine. This was enabled through the identification of the major integration site(s) in HCC tissue, followed by validation using tailored primers for these major sites from urine. The novel sequence created by HBV integration was taken advantage of, using it as a unique marker to trace for the HBV-integrated DNA that was released into circulation, and demonstrated the detection of identical integration sequences between the tumor tissues and corresponding urine samples. Detection of such unique sequences in the urine provides unambiguous evidence that HBV integrated DNA from the liver is released into circulation, and is filtered into urine as fragmented, cell-free DNA.
Two important features of HBV integration are foundations of this proof-of-concept study. First is the appearance of over-represented or major HBV integration sites in HCC due to uncontrolled clonal expansion, as demonstrated in earlier studies. While proliferation of infected hepatocytes can occur in non-HCC liver disease, mostly within 105 cells, clonal expansion observed in HCC tumors is uncontrolled. This results in expansion of ˜109 cells (1-3 cm tumor size), and results in preferentially abundant HBV integrated sequences in the infected liver or in the HCC nodule. This is shown in the supporting reads, which describes as the major HBV integrations in the NGS study (Table 9). Because of their high abundance, it was reasoned that these major HBV integration sites in the infected liver would most likely to be predominantly detected in urine. As predicted, major HBV integrations sites were detected in matching urine samples in six of nine HCC patients tested.
Second, the HBV integration events are random, and HCC-derived integration sites have previously been used as a cellular signature of the clonality of HBV-HCC tumors. Among over a thousand HBV integration sites identified in recent NGS-based studies, the most frequently reported recurrent integration targeted gene is TERT. Strikingly, with over 60 HBV-TERT junction sequences reported, no two are identical at both viral and host breakpoints. This further supports the hypothesis that HBV integration sites created by integration could serve as a molecular signature of the infected hepatocyte. Therefore, detection of an emerging, predominant integration site in the urine could be a potential biomarker for an early clonal expansion or HCC in a chronic HBV infected individual, as illustrated in
The mechanistic links between HBV integration and hepatocarcinogenesis have been suggested to include activation of oncogenic genes and induction of chromosomal instability. By analyzing 34 major integration sites from 23 HBV-HCC patients, five were targeted in proximity of the TERT gene, and three within range of the CCNE1 gene, both commonly recognized oncogenes. Three additional integration sites at TSHZ2, GPHN, and miR512-1 have also been reported to be associated with carcinogenesis. The integration site identified from patient #7 showed chromosomal rearrangement, a common event in cancer. This high frequency of integration in oncogenic genes and the evidence of chromosomal instability detected in this study led people to study and compare other reports. Therefore, a meta-analysis of data reported from 15 studies, 446 patients, and 1554 HBV integrations was carried out. In line with this study, it was found that TERT and CCNE1 are among the most frequently reported targeted genes by HBV. Interestingly, it was observed that 10 other genes were targeted by separate studies from different groups, and most had previously reported association with cancer while other two functions are unknown. This indicated that while HBV integration may be random, disruption of particular regions might have more of an impact on development of HCC. Since TERT was by far the most commonly targeted gene, both the human and HBV genomic locations of each integration site were mapped. Strikingly, it was found that HBV integration is frequently observed in a narrow region of the TERT promoter, despite every integration site being unique. Since TERT promoter mutations are recognized drivers of carcinogenesis and TERT promoter integrations are mutually exclusive with these mutations, it is suggested that HBV integrations have the potential to act as drivers of carcinogenesis. Of note, the cohort in this small study was mostly of HBV-HCC patients that were predominantly non-cirrhotic (77%). This could imply that HBV integration plays a more direct role in HCC carcinogenesis in non-cirrhotic patients.
In moving forward, a more thorough analysis of HBV integration sites is needed to better assess the role of integration with carcinogenesis. While disruptions in TERT and CCNE1 appear to be well implicated in connection with development of HCC, there are likely several other important genes that are less frequently targeted. It was previously reported for. The detection of circulation derived DNA in the urine, and it thus believed that urine will be the best source to profile HBV integrations of the liver because unlike blood, urine contains limited (if any) infectious HBV particles. Even though HBV integrated DNA in the urine makes up only a very small fraction of total cfDNA, with advance in sensitivity of technology of detecting cfDNA, detection of major HBV integration sites in urine is plausible. As 85% of HBV-HCC samples were found to contain integrated HBV DNA, detection of the major HBV integration sites in urine could serve as a specific and sensitive marker for HCC screening of the chronic HBV infected population.
Hepatocellular carcinoma (HCC) is the 2nd leading cause of cancer deaths worldwide [1-3], and suffers from poor prognosis in part due to lack of effective treatment options. The major etiology of this multifactorial disease is chronic hepatitis B virus (HBV) infection, which is associated with approximately 50% of HCC cases worldwide [4]. During the course of infection, HBV can integrate into the host genome. It has been believed that integration events mostly occur through non-homologous end joining (NHEJ) [5], as well as through micro-homologous recombination [6-9]. While HBV DNA integration into the host genome is considered rare, with an estimate of one integration event per ten thousand HBV-infected hepatocytes [10], the integrated viral DNA has been reported in more than 85% of HBV-related HCCs (HBV-HCC), suggesting a significant association of HBV integration in hepatocarcinogenesis. Mechanisms of HBV integration in HCC carcinogenesis could vary in patients and include insertional mutagenesis of HCC-associated genes, induction of chromosomal instability, and continuous expression of viral proteins [11,12]. Understanding the impact of integrated HBV DNA on carcinogenesis and potentially identifying HCC driver genes as personalized biomarkers could pave the way for precision disease management in HBV-HCC patients.
With the advent of next generation sequencing (NGS), thousands of HBV integration sites have been identified across the human genome. Over 15,000 HBV integration sites have been reported from PCR and NGS-based approaches from tumors [6,13-36]. While no known host sequence preference or specificity [5,37-41] was identified, integration can activate known HCC driver genes and has been reported in TERT, CCNE1, and MLL4 [42]. Integration in these genes has been reported in a recurrent manner (i.e. in more than one HCC patient) and have become known as recurrently targeted genes (RTGs). Interestingly, no RTG has been identified from non-HCC livers of chronically HBV-infected patients (n=90, 960 integration sites) [11, 27, 43, 44], suggesting its specificity for HBV-HCC. Similar to the approach of identifying BRAF V600E driver mutations by the identification of recurrent hotspot mutations, here we take advantage of the large amount of reported integration sites from literatures and our in-house study reported here to test the hypothesis that HCC drivers can be identified by characterizing RTGs.
In this study, we compared integrations sites identified in tumor and adjacent-to-tumor (adj-tumor) tissue and defined RTGs. By characterizing the top 10% most frequent RTGs, we demonstrate the potential of identifying HCC drivers for HCC precision medicine and drug development.
2.1. Identification of RTGs in 22 HBV-HCC Tumors
The HBV DR1-2 region is a known integration hotspot. To identify HBV integration sites in a cost-effective manner, we applied an HBV DR1-2 enrichment NGS assay, as described in Materials and Methods, to enrich for HBV DNA in the DR1-2 region. NGS libraries prepared from archived DNA isolated from a cohort of 22 HBV-HCC formalin-fixed paraffin-embedded (FFPE) tissue specimens were used. NGS reads were analyzed using ChimericSeq [45]. We aimed to detect HBV junction sequences (HBV-JS) in 1-10 million NGS reads. Table 10 summarizes the NGS results and the major HBV-JS identified. Major HBV-JSs were defined as the most abundant HBV-JS in each tested sample that has at least 2 supporting reads and having more than 10% of total junction sequences. Assuming a 1:1 copy ratio of HBV to human genomic DNA, we obtained at least 1,000-fold enrichment resulting in an average of 1.0±0.3% on-target HBV reads (Table 10). Encouragingly, integrated HBV DNA was detected in 91% of HBV-HCC tumors from a 1-10 million NGS reads per sample (Table 10). Interestingly, of 27 major HBV-JS identified, seven junctions were found in frequently reported HCC driver genes (TERT and CCNE1) [46]. Junction-specific PCR primers were designed for 16 junctions with the most supporting reads and amplified in respective tissue DNA. PCR products for 14 of 16 tissue DNA samples were obtained and the junction sequences were confirmed by Sanger sequencing for an 88% validation rate (data not shown).
2.2 Overview of the Studies for RTG Identification
The studies included in RTG identification are summarized in Table 11, where 19 studies utilize NGS-based and 8 studies utilize PCR-based approaches for HBV integration identification. For each study, the sample size and the number and percentage of HCC tumor or adj-tumor tissue that had detectable integration sites are listed. Note, most of the studies did not examine the DNA from the adj-tumor. Together, we compiled a total of 15,749 integration sites: 8,491 from tumor tissues and 7,258 from the adj-tumor, from 1,023 HCC patients. We found 80% of tumor tissues (n=1,276) and 50% of adj-tumor tissues (n=760) contained detectable integration sites. Of the seven studies that enriched for the whole HBV genome, on average 81% (range 57%-100%) of the tumors examined were found to have integrated HBV DNA (n=7) [6,22-24, 26, 27]. In two studies, 65% [28] and 91% (our study) of tumors examined were positive for integrated HBV DNA.
1HBV (+) HCC cohorts-only;
2three patients overlapping with Jiang 2012 [13] were removed, while the cumulative number of integration sites were compiled and considered unique integration sites due to different reported assay parameters;
3only human chromosome sequence position provided;
4cohorts of HBsAg (−)/occult (+) and HBsAg (+) HCC patients;
5out of 20 paired non-tumor tissue analyzed;
2.3 Clinical Characteristics of HBV-HCC Patients with Integrated HBV DNA
The major clinical factors associated with HCC, such as age, gender, HBV genotype, and whether the HCC arose in a cirrhotic liver, designated as “cirrhotic HCC”, are summarized in Table 12. We categorize HCC patients based on the detectability of integrated HBV DNA in tumor tissue. The general characteristics of the HBV-HCC population [4, 47, 48] are also summarized. Analysis of each parameter was performed as available. The sample sizes that were available for data analysis of each parameter in each cohort are noted in parentheses. Overall, there is no significant difference between the two cohorts as compared to the overall HBV-HCC population for age and gender. The male:female ratio across the cohorts was not significantly different. Of the three reported HBV genotypes, genotype C was the most frequently reported in the integration-detectable tumor cohort (73%), while the tumor cohort with no detectable integration had only 2 patients with genotype reported and both were genotype C. In this cohort, 62% of HCC was derived from the cirrhotic liver in the integration-detectable tumor cohort, which is less than the 70-80% range found in the HBV-HCC population, reported from the literature [4]. 47% of patients with cirrhotic HCCs in the tumor cohort with no detectable integration were reported from 15 patients with available cirrhosis information.
1, characteristics of the general HBV-HCC population obtained from the following references [4, 47, 48];
2.4 Recurrent Sites of HBV DNA Integration
Next, we identified RTGs in the compiled HCC cohort and explored their associations with carcinogenesis. Of the 15,749 integration sites examined, 6,249 integration sites were found within 150 kb of gene coding sequences in HCC tumors, and 2,800 genes were identified. Among these 2,800 genes, we considered an integrated gene as a RTG if it was detected from at least two HCC patients and from two independent studies, as described in Materials and Methods. A total of 358 genes were found in 556 HCC patients, constituting 54% of the HBV-HCC patients with detectable HBV integration (n=1,023) and 43% of all HBV-HCC patients (n=1,276) in this cohort. The top 10% of the most frequently recurrent genes (n=36) are listed with summaries of their counts, identified integration sites, and associations with carcinogenesis in Table 13. Interestingly, these 36 genes either have previously suggested associations with carcinogenesis (28/36, 78%) or have no known function (8/36, 22%). As expected, TERT and MLL4 are the two most recurrent genes.
Next, the 358 RTGs were queried for significantly enriched Gene Ontology (GO) pathways using Enrichr [96]. The top enriched biological pathway of the RTGs was chromatin-mediated maintenance of transcription with a combined score of 17.27 (p<0.05), suggesting possible links with oncogenesis (
2.5 Integration Breakpoints in the HBV Genome
To investigate the distribution patterns of the integration breakpoints in the HBV genomes, we analyzed the HBV breakpoints in tumors (n=3,052) and adj-tumors (n=5,259), where available. We omitted studies that enriched for HBV DR1-2 sequences to assess HBV breakpoints distribution in a non-biased manner. Consistent with previous reports, we observed that 37% of breakpoints were within nt. 1300-1900 region in tumors and 56% in adj-tumors. This region covers the 3′ end of the HBx gene and is where the initiation site of viral replication/transcription are located [6, 23, 27]. Also consistent with previous reporting [16], we observed a breakpoint hotspot in the HBV DR1-2 region, representing 15% for HCC tumors and 28% for adj-tumors of all HBV breakpoints (
2.6 Genomic Breakpoints of TERT, MLL4 and PLEKHG4B RTGs
As HBV integration is believed to be non-sequence-specific, it was of interest to examine all RTG coordinates for similarity to each other. To do so, we plotted the available human and HBV breakpoint coordinates of the three most frequent RTGs identified, TERT, MLL4, and PLEKHG4B (
For TERT, the most frequently recurring RTG, 219 of 415 junctions from 161 HCC patients have both human and HBV breakpoint coordinates available. As expected, most of these breakpoints were centered between DR2 and DR1 of the viral genome and were highly concentrated at the promoter region of the TERT gene (
MLL4 is the second most frequently reported RTG with 102 junctions identified from 178 HCC patients studied. Among them, 115 breakpoints from 64 HCC patients have both human and viral coordinates available and are plotted in
The third most reported RTG is PLEKHG4B. The reported breakpoints were interestingly all centered within a 3 kb region that is around 131 kb away from the PLEKHG4B coding region. A total of 47 of 116 breakpoints from eight HCC patients have both viral and human coordinates available, as shown in
TERT hotspot promoter mutations (−124, −146) are the most frequently reported mutations in HCC, found in about 50% of cases [99-104]. In HBV-HCC, up-regulation of TERT expression could also be caused by HBV integration at or near the TERT promoter region [14, 16, 22, 28, 29, 105]. Next, we compared the incidence of TERT promoter mutation and HBV integration. For our in-house cohort (n=22), shown in
In this study, we compiled and studied over 15,000 HBV DNA integration sites from 1,276 HCC patients reported from 26 previous studies and our in-house study, to test our hypothesis that frequent recurrently targeted genes (RTGs) by HBV integration are HCC driver gene candidates. By using three criteria for RTG identification, we identified 358 RTGs. Encouragingly, the top 10% of the most frequent RTGs (n=36) either have known involvement in carcinogenesis (28/36, 78%) or have unknown function (8/36, 22%). By gene ontology analysis, RTGs were mapped to functions related to carcinogenesis. Together, we demonstrate the potential of HCC driver identification by characterization of frequent RTGs. More studies are needed to define the association of carcinogenesis with the frequency of RTGs.
Three criteria were applied to identify 358 RTGs from HBV integration sites in this study: (1) gene annotation within 150 kb of the breakpoint, the distance previously reported where host genes can be impacted by integration [105,106], (2) reports from ≥2 HCC patients to define “recurrent”, and (3) by ≥2 independent laboratories to avoid the possibility of contamination within a laboratory. We are aware that identification of RTGs across multiple studies is complex in nature, with multi-faceted underlying variables such as integration detection methodologies and patient populations. For instance, some studies do not contain any of 358 RTGs that we identified [35,36], while others have a high detection rate of a particular RTG, such as MLL4 [50], and cMYC [23]. We are also aware that different methodologies for identifying integrations may have different sensitivities that can result in detection of different integration site profiles. Despite these limitations, that may result in missing some RTGs, detection of RTGs constitute a potential HCC driver gene identification that maybe clinically useful for HCC patients.
Encouragingly, the most frequent 10% of RTGs (n=36) identified using the three criteria defined in this study either have known involvement in carcinogenesis (28/36, 78%) or have no known function (8/36, 22%). Although more studies are needed to explore the association of the genes that have unknown functions in hepatocarcinogenesis, of the genes that have known functions, all have been associated with either liver cancer or other cancers. Together with RTG ontology analysis where a significant mapping of genes to functions related to carcinogenesis was observed, our data suggests the potential to not only identify known HCC drivers, but to discover new HCC driver genes by characterization of frequent RTGs for precision disease management. More studies are needed to define the degree of association of carcinogenesis with the frequency of RTGs.
By detailing the three most frequent RTG junction coordinates (TERT, MLL4, and PLEKHG4B), we reveal three important features. First, as expected, the majority of junction coordinates are different, confirming the non-sequence-specific integration in the host genome. The overlapping identical junctions identified in the TERT promoter region highlight the potential importance of the site on impairing the expression of the TERT gene. Second, an interesting pattern was observed in PLEKH4G4B junctions. Although a microhomology search did not suggest the homologous recombination was the cause of this interesting pattern, a highly repetitive sequence, satellite sequences, and a motif of TAAACCCTAAC were identified in these regions. Together suggest possible repeated breakpoints in the region. This supports a possibility of occasional homologous recombination in addition to the non-homologous end-joining mechanism of HBV integration. Since these unique integration pattern sequences was reported from one study and was not reported to be validated in the original tissue DNA, an artifact has not been excluded. Lastly, the mutually exclusive detection of TERT promoter mutations and TERT integration is shown by our small cohort of 22 HCC patients and confirmed by a larger compiled cohort of 151 HCC patients [24,26]. When describing the TERT genetic alterations as an HCC driver, TERT promoter mutations only account for 50% of alterations, indicating the importance of identifying TERT integration. This further emphasizes the need for analysis of frequent RTGs to better characterize HCC.
Most HCC cases develop in a cirrhotic background, though up to 30% of HBV-HCC cases were reported in the absence of cirrhosis (non-cirrhotic HCC) [4,48]. In our study cohort, we identified slightly (but not significant) lower rates (62%) of cirrhotic HCC when integration was detected. In the case of TERT-integrated HCC (n=257) in this study cohort, 51 had information to assess whether the HCC was rising in a cirrhotic background. We identified a significant association (p=0.01) of TERT integrations with cirrhotic HCCs compared to non-cirrhotic HCCs (data not shown). While this cannot be applied to the remaining 206 TERT-integrated HCC patients, in which there was no available information to assess the existence of cirrhosis, it is in line with the association of TERT hotspot promoter mutations with cirrhosis [107].
4.1. Data Mining/Search Strategy
We searched PubMed (2000-Dec. 1, 2018) databases using Medical Subject Heading (MeSH) terms “hepatitis B virus”, “HBV integration”, “hepatitis B integration sites” to identify the literature that have reported HBV integration sites by either NGS- or PCR-based approaches. Additional studies were obtained by cross-referencing from the literature. We included only studies in English and studies that included HCC subjects. We included all studies that identified HBV integration sites using NGS-based approaches. For the studies using PCR-based methods, we only included the studies that analyzed a study sample size of 10 or more HCC patients. HBV integration sites identified by RNA-seq or transcriptome NGS [7, 8, 109] were not included as expression of integrated sequences can be due to many host cellular factors that enable expression of integrated sequences and thus are not within the scope of this study. We filtered out repeated integration sites to ensure each integration site was included only once in our study, with the exception of two studies that utilized different methods on overlapping samples [13,19]. A total of 26 reported studies in addition to our study are included as summarized in Table 2.
4.2. In-House HCC Specimens and HBV Integration Analysis
Archived FFPE tumor tissue DNA (Table 14), as described previously [110,111], from stage I-IIIB patients (n=32) was obtained from the National Cheng-Kung University Medical Center, Taiwan, collected in accordance with the guidelines of the Institutional Review Board. An HBV enrichment NGS assay (JBS Science, Inc) was used. Briefly, NGS libraries were generated, enriched for HBV DR1-2 sequences through two rounds of a multiplex biotinylated HBV primer extension capture (PEC). Libraries were sequenced on the Illumina MiSeq platform (Penn State Hershey Genomics Sciences Facility at Penn State College of Medicine, Hershey, Pa.) and analyzed using ChimericSeq [45] to identify HBV-host junction sequences. Tailored junction-specific PCR-Sanger sequencing was designed and used to validate each HBV integration site of interest, identified by HBV-enriched NGS assay.
4.3 TERT Promoter Mutation Analysis by PCR-Sanger Sequencing
HCC tissue DNA was used to amplify a 163-bp region (Chr5:1295151-1295313) of the TERT promoter by using HotStart Plus Taq Polymerase (Qiagen, Valencia, Calif.) with forward primer 5′-CAGCGCTGCCTGAAACTC-3′ (SEQ ID NO: 212) and reverse primer 5′-GTCCTGCCCCTTCACCTT-3′ (SEQ ID NO: 213). The PCR products were sequenced at the NAPCore Facility at the Children's Hospital of Philadelphia (Philadelphia, Pa.) and analyzed using ClustalW software [112].
4.4 Identification of Integration Recurrently Targeted Host Genes (RTGs)
To identify host genes that maybe affected by HBV DNA integration in a universal manner across all studies, we identify the closest gene within 150 kb of the integration event, the distance previously reported where host genes can be impacted by integration [105,106]. To define the status of a RTG, we assessed whether the reported gene was identified in tumors from (A) two or more HCC patients and (B) two or more independent studies to avoid potential cross contamination within a study. The full list of identified RTGs can be provided upon request.
4.5 Gene Functional Enrichment Pathway Analysis
358 RTGs were subjected to enrichment pathway analysis using Enrichr (http://amp.pharm.mssm.edu/Enrichr), to identify significantly (p<0.05) enriched pathways as determined by gene ontology.
This HBV integration study using an in-house HBV-HCC cohort, in conjunction with previously reported HBV integration sites, allows us to test the hypothesis that HCC drivers can be identified by characterizing frequent recurrent targeted genes (RTGs) by HBV integration. By analyzing over 15,000 HBV integration sites, we bring forth a RTG consensus and demonstrate that characterization of frequent RTGs can be a novel approach to discover or identify HCC drivers for HBV-HCC precision medicine and drug development/discovery.
The present application claims priority to the U.S. provisional patent application No. 62/875,059, filed Jul. 17, 2019, whose content is hereby incorporated by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 62875059 | Jul 2019 | US |
