Artificial Chromosomes for Disease Modeling

Synthetic chromosomes are ideal delivery systems for ferrying large sections of human DNA into cells. In contrast to viral vectors, human artificial chromosomes (HAC) carry more genetic material and are less likely to trigger an immune response. So far, however, technical problems have prevented HAC from reaching their full potential.¹ Now, in a paper published in Science, researchers described an improved technique for engineering HAC that sidesteps previous barriers.²

Earlier methods to synthesize HAC relied on linking shorter DNA constructs into a larger chromosome within the cell in a process called multimerization. However, the genetic fragments tended to connect in unpredictable sequences of varying lengths, making it difficult to anticipate how the genes would behave. Furthermore, the constructs often attached to natural chromosomes, potentially disrupting the host cell’s genome.

In the new study, researchers at the University of Pennsylvania circumvented this problem, known as uncontrolled multimerization, by synthesizing larger strands of DNA so that HAC could be formed from a single construct. Instead of creating 200 kilobase pair sequences, the researchers increased the size of the DNA construct to approximately 750 kilobases, removing the need for multimerization.

“We hypothesized that to [avoid] multimerization, we would have to start big,” said University of Pennsylvania biochemist Ben Black, who led the research.

Within each HAC, the researchers designed sequences to bind centromeric proteins, which ensured that daughter cells would inherit the chromosome when the original cell divided. Specifically, they added hundreds of DNA binding sites for the bacterial Lac repressor protein (LacI). By engineering the host cells to express a fusion protein—LacI fused to another protein that binds to a key centromeric protein—the team recruited proteins to form the centromere on the DNA construct.

Instead of the linear chromosomes found in mammals, the team crafted a circular chromosome, which is compatible with DNA replication in yeast. They then assembled the HAC inside budding yeast. By removing the yeast cell wall and applying a chemical that stimulates membrane fusion, the team transferred the HAC from yeast cells into human cells.

To confirm that the HAC had passed to human cells in the correct configuration, the team lysed the cells and separated the DNA from the rest of the cellular contents. Unlike linear DNA, circular chromosomes cannot move through the gel during electrophoresis, so they remain trapped in the sample well. This electrophoresis test indicated the HAC had stayed circular. A separate technique called chromatin stretching, in which circular chromosomes produce a characteristic signal when stretched and stained with antibodies, provided further evidence that the HAC had retained its original shape and size.

Analyzing overall shape and size doesn’t rule out rearrangements within the chromosome sequence, said Bill Earnshaw, a cell biologist at the University of Edinburgh who was not involved in the study. “The HAC seem to be single copy, but they haven’t done a rigorous analysis to show that they’re not rearranged.”

Black agreed that more research is needed to confirm the integrity of the new HAC. However, the absence of an enlarged HAC suggest that the constructs aren’t rearranged by linking together, or sticking to the host’s chromosomes, he said.

The new method may one day be used for gene therapies that express working copies of genes that function independently of the host's chromosomes. The technique may be particularly useful for conditions that require replacement of a large gene. For certain cancers, researchers may use HAC to ferry a panel of protective genes into patient cells.

The most immediate application is likely to be for creating cell lines or animal models that more accurately recapitulate human disease. Animal models don’t always faithfully mirror human conditions, so their use in drug testing is sometimes unreliable. However, by expressing groups of genes—for an entire metabolic pathway, for example—scientists might engineer better animal models, Black said.

“As a synthetic biologist, you could be a kid in the candy store if you could deliver all sorts of genetic payloads into cells,” he said.