Discoveries in DNA: What's New Since You Went to High School?

Thus far, the impact of molecular genetics on human disease has been primarily to identify specifically which genes are implicated in specific diseases. For single-gene disorders, mutations have been discovered in hundreds of genes. The current challenge is to identify which genes contribute to multifactorial conditions like obesity, heart disease, alcohol dependency, schizophrenia and autism. So far, “genome-wide association studies” have identified variant DNA sequences showing statistical association with these and other complex diseases, but demonstrating a mechanistic role for these variants has proven elusive.

Rapid and inexpensive genome sequencing, together with high-speed informatics and a large and expanding database of annotated human DNA sequence variants associated with disease risk, has made possible personalized medicine and customized genetic counseling. In a particularly famous case, actress Angelina Jolie elected to have a double mastectomy and her ovaries and fallopian tubes surgically removed when she learned that she carried a mutation in the BRCA1 gene that predicted an 87 percent risk of breast cancer and a 50 percent risk of ovarian cancer. Unfortunately, for the vast majority of genetic cancer associations, removing nonessential tissue is not an option. However, prior knowledge of an increased risk can result in increased surveillance, and cancer is most curable when caught early. For example, persons at increased risk for hereditary non-polyposis colorectal cancer should undergo frequent colonoscopies to identify and remove pre-cancerous colon polyps before they turn into full-blown colon cancer. Since 23andMe began offering direct-to-consumer marketing of genomic sequencing, over a dozen companies now offer various forms of genome sequencing and analysis.

A healthy intestine carries about 10 times as many microorganisms as the number of cells in the entire body. The metabolic activity of these microorganisms can significantly impact health. For example, their metabolic activity is an important source of biotin (vitamin B7), and the composition of gut microflora can shape the immune response, leading to sensitivity or resistance to allergies and autoimmunity. The availability of a large number of complete microbial genomes and the technology of high-volume DNA sequencing has enabled the genotyping of gut microbiomes under different dietary and health conditions, leading to new, detailed understanding of the differences between healthy and unhealthy gut microflora. Direct-to-consumer gut microflora sequencing services are currently available, although the benefits of this knowledge for otherwise healthy people are currently limited.

Rapid DNA sequencing has displaced more time-consuming, costly and less specific culture or antibody assays in the detection, classification and diagnosis of infectious disease. In 2003, DNA sequencing at the Michael Smith Genome Sciences Centre in Toronto and at the Centers for Disease Control in Atlanta showed that the virus causing an outbreak of severe acute respiratory syndrome (SARS) in China was a new variant of coronavirus. This made the epidemic easier to track and contain. More recently, sequences of 99 Ebola virus genomes taken from 78 confirmed patients at different points in their infections helped to map the origin and spread of the virus during the 2014 outbreak in Sierra Leone. This epidemiological information should aid in containment efforts in future outbreaks.

Most high school biology students learn some basic animal and plant taxonomy. The foundation of classical taxonomy is morphology. With the availability of genome sequences from representative species in whole phyla, rigorous quantitative measurements of genetic distance based on DNA sequence divergence has been used to test existing evolutionary trees and to re-classify organisms in all kingdoms. For example, large-scale taxonomic DNA sequence comparisons have established more rigorous relationship trees and taxonomic distances for the large and diverse class Aves (birds) and the phylum Arthropoda.

Human genome sequencing provides much more detailed and specific genealogical information. Several commercial services will provide information on likely ancestry based on combinations of DNA sequence variants known to be rare or prevalent among people originating from different regions of the world. However, it should be noted that one sometimes unwelcome outcome of genome sequences for pedigree or genealogical purposes is the discovery of non-paternity. While rates vary widely between different populations, they have ranged between 2 and 30 percent in specific studies.

Forensic science is increasingly turning to DNA sequencing to implicate or exonerate potential culprits and to identify remains. In such cases, it is usually sufficient to sequence only a subset of genomic DNA representing regions found to be most variable among individuals. This approach circumvents the much higher cost of sequencing and data management for whole genome sequencing, while providing sufficient specificity for forensic purposes.

High school biology students are taught Mendel’s laws of genetic inheritance. The first law states that a genetic trait is transmitted from one generation to the next without modification, even when it is recessive and not evident in carriers. The idea that a genetic trait could be modified by the life experience of its carrier is a violation of this law and is typically dismissed as Lamarckian fallacy in high school classes.

Biochemists have known that human DNA (as well as the DNAs of many microbes, plants and animals) contains other bases besides the canonical adenine, cytosine, guanine and thymine (ACGT). In human chromosomes, 3 to 5 percent of the cytosine bases are actually a modified form of cytosine called 5-methyl cytosine. This modification is usually associated with repression of genes in humans. Importantly, the extent of modifications can differ at the same gene in the sperm and egg, such that the expression of dad’s copy will be different from mom’s copy in the child that inherits them. This phenomenon is called “imprinting.” Parental imprinting is important for genetic health, as failure of this imprinting underlies syndromes such as Prader-Willi, Angelmans, Beckwith-Weidemann and Silver-Russell syndromes.

The transmission of different states of gene expression through multiple cell divisions and across generations has been termed “epigenetic,” since the underlying DNA sequence is identical in both states. While the four bases of DNA – adenine, cytosine, guanine and thymine – cannot be altered by a parent’s life experiences, scientists have discovered that a form of one of the base pairs, cytosine, can be expressed in different forms as a result of environmental factors. Techniques have been developed over the last decade to allow for “epigenomics,” the genome-wide characterization of methylation patterns. Epigenomics is presently being used to identify inherited pre-disposition to obesity, diabetes, cardiovascular disease, addiction and psychiatric disorders, as well as markers for aging and cancer progression.

Drugs that inhibit or stabilize epigenetic marks are in clinical use for cancer and are being tested for other indications like sickle cell disease.

For over 50 years, it has been known that among multicellular animals and plants, the size of genomes can vary over orders of magnitude in ways that are not explained by the apparent complexity of the organisms specified by those genomes. With the availability of whole genome sequencing, it has become evident that, for example, the number of genes in humans is not much greater than that of the fruit fly. Much of our genome consists of repetitive DNA sequences, transposable elements and non-functional relicts of genes and transposons with no discernible function. This DNA has been termed “junk DNA” to convey this apparent lack of function.

In the past 15 years, however, detailed analysis of which regions of DNA are transcribed into RNA copies has uncovered a significant amount of non-protein-coding RNAs that serve regulatory functions. So-called microRNAs have been shown to target specific protein-coding RNAs for destruction or inhibition of protein synthesis. Other RNAs, termed long noncoding RNAs, also appear to regulate expression of protein-coding RNAs, but the mechanisms are just now being worked out. That said, this still leaves a majority of our genome with no apparent function. It seems likely that the mechanisms by which DNA accumulates in genomes through transposon jumping and genome duplications is not balanced with a similar rate of sequence elimination, and that the burden of this unpurged DNA has little or no evolutionary cost.

A major goal for gene sequencing has always been as a platform for the design and implementation of gene therapy. The first gene therapy was bone marrow engraftment for the treatment of leukemia and other blood cancers. The first human bone marrow transplant was performed in 1956 by E. Donnall Thomas. In these therapies, the patient’s own blood-producing bone marrow cells are treated with radiation and chemotherapy, and the blood of a twin or closely matched donor is instilled. The idea is that the ablation will destroy any remnant of the cancer together with the healthy hematopoietic stem cells, and that donor’s blood stem cells will populate the patient’s marrow and regenerate the entire red and white blood cell repertoire from healthy cells. In effect, this is gene therapy, since the donor’s genes are replacing the patients genes in the blood cell lineages.

Targeted gene therapies, however, had to wait for (1) the identification of the genes to target, (2) the cloning and/or sequencing of the relevant genes and in some cases, the specific disease-causing variant, (3) a full understanding of the normal gene function and regulation, and (4) the development of efficient ways to deliver genes to the relevant tissues at therapeutic levels. The advent of molecular cloning, DNA sequencing and the many tools of molecular genetics and cell biology has given us sufficient knowledge of the basis for disease and the genes to target, but what has limited the application of gene therapy has been efficient gene delivery systems.

Scientists realized viruses could be the perfect tool to do the work of gene editing. They are already designed by nature to insert themselves into our DNA. The first successful targeted human gene therapy was reported in 2000. This was a virus-mediated therapeutic gene to treat X-linked severe combined immunodeficiency. The deck was stacked in favor of success in the choice of this particular disease since it was known that the therapeutic gene just had to be expressed in a modest number of blood cells to achieve therapeutic benefit.  The therapeutic gene was carried by an engineered retrovirus that was used to infect the patient’s blood cells before they were re-injected into the patient.

Thus far, targeted gene therapy successes have been very limited. Other blood disorders that have shown significant benefit from targeted gene therapy in small trials include hemophilia (specifically, factor IX deficiency), severe beta-thalassemia (deficiency for the adult beta-globin gene) and leukemia, where the patient’s immune cells were treated to enable them to recognize cancer cells and destroy them. Targeted gene therapy for degenerative blindness caused by Leber congenital amaurosis improved vision for a few years but failed to arrest the degeneration process.

The first approved commercial targeted gene therapy is Alipogene tiparovovec (trade name Glybera), a virus-mediated therapeutic delivery of human lipoprotein lipase to the muscle cells of lipoprotein lipase deficiency patients. It was approved for clinical use in Europe in 2012.

For many genetic disorders, the disease results from the expression of a defective gene product, not the complete absence of the product. For example, sickle cell disease and the major form of cystic fibrosis are both associated with abnormal proteins. In such cases, editing a patient’s own gene to the normal form should provide greater benefit than merely expressing the normal protein in the presence of the abnormal one.

For gene editing to work, it is essential to uniquely target a single site among the 3 billion nucleotides in the haploid (single set of unpaired chromosomes) human genome. In other words, therapeutic gene editing therapy must be able to efficiently edit the intended target without introducing off-target editing at sites that fortuitously resemble the intended target.

Two targeting strategies are currently under development—protein-based targeting and RNA-based targeting. In both cases, the idea is to target an enzyme that cuts both strands of the double helix at a specific site. If the goal is to inactivate the target gene, the creation of the break is sufficient to trigger cellular mechanisms that lead to error-prone repair and inactivating mutations. True editing—the replacement of bad sequence with good sequence—requires the simultaneous introduction of DNA fragments containing good sequence into the same cells.

Protein-based targeting strategies rely on custom-engineered modular proteins that recognize and bind to specific DNA sequences. One approach builds on the so-called zinc finger protein fold first described in sequence-specific transcription factors. The publicly traded biotech company Sangamo BioSciences was founded in 1995 to exploit zinc finger protein engineering for gene therapy and agricultural genetic engineering. In this approach, a series of zinc finger modules, each chosen to recognize a specific 3-nucleotide motif, are fused in tandem to one another and to a nuclease subunit.

Another protein-based targeting strategy, TALENs, is based on the transcription activator-like effector (proteins secreted by Xanthomonas bacteria). In this case, repeating 33-34 amino acid modules with specificity to each of the four bases in DNA are fused in tandem to create the targeting peptide.

More recently, the clustered regularly-interspaced short palindromic repeats (CRISPR) prokaryotic immunity mechanism has been exploited for gene editing. In this case, specific DNA sequences are targeted by RNA-DNA hybridization, which directs the Cas9 enzyme to cleave DNA at the target site. This mechanism was only worked out in 2007, but has already emerged as the front-runner technology for gene editing, due to its relative simplicity and high efficiency. In human cells, the efficiency of zinc-finger- and TALE-mediated editing achieve efficiencies of 1 to 50 percent, while CRISPR-Cas9 editing has been reported to have efficiencies of up to 78 percent in single-cell mouse embryos. Exciting clinical applications of gene editing include correcting the mutation in the bone marrow stem cells of patients with sickle cell disease or hemophilia.

One of the most exotic applications of genetic engineering to be proposed is the resurrection of woolly mammoths. A project underway at Harvard University, under the direction of geneticist Dr. George Church, seeks to edit the elephant genome to create a cold-tolerant mammoth-like chimera using CRISPR/Cas9 technology. This strategy is enabled by the complete sequencing of the mammoth and elephant genomes, and based on gene annotations that suggest the target genes most likely to program the mammoth adaptations to extreme cold into elephant DNA. The stated goal is to repopulate the tundra and boreal forest in Eurasia and North America, to protect endangered Asian elephants and to revive an ancient grassland in the tundra, with the hope of forestalling the melting of Siberian permafrost.