The Effect of Genetic Variation on Phenotype

Detecting Mutations Using Targeted Genomic Enrichment of Multiplex Barcoded Samples

Monitoring Phenotypes Using Next Generation Sequencing

Scientists in Prof Dr Edwin Cuppen’s laboratory (Hubrecht Institute, Royal Netherlands Academy of Arts and Sciences and the University Medical Center Utrecht, both in The Netherlands) use next generation sequencing (NGS) in a systematic genomics approach to understand the principles by which genetic variation affects phenotype. Specifically, molecular phenotypes, like quantitative and qualitative gene expression and epigenetic modifications, are monitored in inbred animal models for which complete genome sequences are available. The laboratory also works on human genetics projects driven by the phenotypes presented by patients. Where a genetic basis for disease is suspected, NGS techniques are used in an attempt to identify the genetic cause.

Whole Genome Sequencing of Human Samples

Whole genome sequencing can readily be performed on inbred animals, but it is less straightforward with human samples for two main reasons:

  1. It is relatively expensive as every individual is unique, and needs to be analyzed independently
  2. The majority of information obtained is irrelevant or cannot be interpreted because of our limited knowledge about genomic elements and the effects on genetic variation on them.

To overcome some of these hurdles, Prof Dr Cuppen’s group has spent several years developing techniques for targeted genomic enrichment (TGE), initially using microarrays and in-solution methods with DNA or RNA probes. However, these approaches were not efficient as each assay allowed enrichment of only a single sample. The group has recently outlined a more cost-effective method of TGE that enables multiplexing of samples [1 , 2]. Multiplexing patient samples speeds up the scientists’ research by enabling examination of cohorts of patients with the same phenotype, rather than just single patients. 

Targeted Genomic Enrichment Protocol

In their protocol, genomic DNA is isolated from samples using standard nucleic acid isolation procedures. The DNA is fragmented by sonication and end-repaired by blunting and phosphorylation for subsequent ligation to truncated adaptors. For the adaptors, the group uses oligonucleotides manufactured by IDT, which are preannealed before ligating to the fragmented DNA. The ligated DNA fragments are purified using AMPure beads (Agencourt). Individual oligonucleotide barcodes, also manufactured by IDT, are incorporated during the ligation mediated PCR step to prepare the sequencing libraries. The barcodes can be introduced by ligation, but the group chooses to use PCR because it is more flexible and convenient, and they make use of an obligatory PCR step in library construction. The barcoded libraries are then pooled before the researchers perform solution-based enrichment using Agilent SureSelect technology or microarray-based capture with custom Agilent SurePrint arrays.

“The quality of IDT oligos is constant, and I do not know of any situations in which we have experienced problems with specific primers or primer sets. When experiments fail in our hands we never doubt the quality of the oligos—we can always be certain there is another cause. It is typically one of the other solutions or a problem with the robotics.”

— Prof Dr Cuppen, on his laboratory’s experience with IDT oligonucleotides

To prevent hybridization of the long, barcoded 3’ adaptors to unrelated library molecules, which could reduce enrichment specificity, the scientists use degenerate primers (IDT) to block barcode and adapter sequences before and during hybridization-based enrichment. Both methods of enrichment are compatible with Applied Biosystems SOLiD™ 4 NGS sequencing, which is used in the Cuppen laboratory.

Using Barcode Blockers to Improve Target Read Rate 

SOLiD sequencing is performed following clonal amplification of library fragments according to the manufacturer’s instructions. Using a 10-base barcoding procedure, the group can perform multiplex amplification of up to 96 independent samples with an average of 59% of reads on target, which compares favorably with the 60–90% of reads on target that they routinely obtain for non-multiplexed enrichment. The use of degenerate barcode-blocking oligonucleotides that bind to all possible barcode decamers significantly increases the multiplexed enrichment efficiency, resulting in an efficient, flexible, and cost-effective means of targeted genomic enrichment for NGS. This technique is being applied to the human genetics research performed in the Cuppen laboratory.

Using Multiplex NGS as a Diagnostics Tool

Prof Dr Cuppen’s human genetics research is focused on the investigation of genetic variation associated with syndromal diseases and different types of cancer. Using conventional sequencing approaches to study syndromal diseases, his group was limited in the number of disease-related genes they could investigate, as they could only examine a single gene per patient at a time. Prioritizing, they would first investigate the most frequently mutated gene. If no mutation was found, they would move on to the second, and then the third. This was a lengthy and costly process and, typically, no more than three genes could be investigated even when there were other candidate genes. Using the multiplex NGS protocol they have developed, they are now able to screen all candidate genes for a given patient simultaneously, in one assay. Tests can be performed much faster and more comprehensively, resulting in a much shorter wait for patients eager to get their results.

The group’s goal is to move these multiplex protocols into routine diagnostic use in order to exploit the immense  benefits gained from NGS. “NGS has revolutionized genomics research,” says Prof Dr Cuppen, “and it enables faster, cheaper, and more comprehensive genetic analysis.” However, a higher quality of NGS data is required before NGS can become a diagnostic tool. Even though good quality data is being obtained by various laboratories, NGS data quality is generally not as high as that obtained using older, conventional techniques such as PCR-based dideoxy resequencing, which is less likely to miscall a base. For this reason, the lab still performs capillary sequencing as a downstream validation step to reconfirm findings observed by NGS. Because of the large amount of sequencing they perform, their validation is never for just one position, but hundreds that are identified by NGS. They typically order their IDT oligonucleotides in 96- or 384-well plates, which are convenient because these can be immediately incorporated into the robots that comprise their automated setups for capillary sequencing, enabling fast turnaround time from identifying candidate variants to verifying activating mutations.

Establishing a Personalized Cancer Treatment Center

With other researchers, Prof Dr Cuppen has recently established The Center for Personalized Cancer Treatment (CPCT), a collaboration of leading scientists from the 3 largest cancer centers in The Netherlands: UMC Utrecht, the Netherlands Cancer Institute, and the Erasmus MC/Daniel den Hoed Clinic. Many cancer drugs that target specific enzymes or pathways are tested in clinical trial, but most of them won't make it to the market as their effect is only slightly better than existing drugs. This does, however, not mean that these drugs don't work. Physicians involved with the center find individual patients who respond extremely well to drugs that otherwise do not perform consistently. Thus, some of these drugs can be effective treatments in specific circumstances.

Researchers at the CPCT believe that cancer drug resistance can be explained by mutations in the specific signal transduction pathways involved. In a bid to guide treatment decisions, they have set up a program to sequence up to 2000 candidate genes in biopsies from cancer patients with metastases in order to add the genetic information of these metastases to the pathological and radiological analysis of the patients’ tumors. This research is still in an early experimental phase, and will first be implemented as a way to stratify patients for specific clinical trials.

According to Prof Dr Cuppen, there are approximately 1000 targeted drugs in early stages of clinical trials, of which only a few will be brought to market if outdated methods continue to be used for validation. He hopes the research at the CPCT will allow them to stratify patient groups to enable selection of the right patients for treatment with drugs under development. The result will be a wider variety of drugs on the market and better patient care through increasing the likelihood that a specific drug will be effective, while avoiding overtreatment that results from giving patients ineffective therapies. In light of spiraling healthcare costs, Prof Dr Cuppen states that it is important to be able to select patients using relatively inexpensive diagnostic tests before administering very expensive targeted therapies that might be ineffective.

Prof Dr Edwin Cuppen obtained his PhD at the Radboud University in Nijmegen, The Netherlands. He pursued postdoctoral research at the Netherlands Cancer Institute in Amsterdam and the Hubrecht Institute in Utrecht, both in the Netherlands. In 2005, Prof Dr Cuppen was awarded the prestigious European Young Investigator Award for his work on naturally occurring and induced genetic variation in the laboratory rat. He was one of the first researchers to generate gene knock out models in this species. Since 2007 he has been professor of Genome Biology in the Biology Department of Utrecht University. In 2009 he was appointed professor of Human Genetics and head of the research section of the Medical Genetics Department of the University Medical Center Utrecht. Pictured here are Prof Dr Edwin Cuppen (back right) and members of his research group, on retreat in Scotland, UK.

References

  1. Harakalova M, Mokry M, et al. (2011) Multiplexed array-based and insolution genomic enrichment for flexible and cost-effective targeted next-generation sequencing. Nat Protoc. 6(12):1870–1886.
  2. Nijman IJ, Mokry M, et al. (2010) Mutation discovery by targeted genomic enrichment of multiplexed barcoded samples. Nature Methods 7(11):913–915.

Author: Nicola Brookman-Amissah is a Scientific Writer at IDT.