By David N. Leff

Can you name the commonest genetic disease on earth?

Alzheimer's? Nope. Parkinson's? Nope. Gaucher's? Nope.

The widest-spread inherited disorder in mankind is thalassemia. So says molecular geneticist Michel Sadelain, who heads the gene transfer and expression laboratory at Memorial Sloan-Kettering Cancer Center, in New York.

"From the perspective of thalassemia and sickle cell disease (SCD) - genetic disorders affecting the beta-globin gene -" he observed, "here's where we stand: In the case of the thalassemia gene, more than 125 mutations have been identified over the years, all of which result ultimately in insufficient production of beta-globin." This is part of the master protein in the blood that fuels the body with oxygen and nutrients.

"In the case of SCD," Sadelain continued, "there's a normal amount of beta-globin expressed, but it carries a mutation in its coding sequence, resulting in a protein that is abnormal. Mutations affecting the alpha- and beta-globin genes are highly prevalent in parts of the world where there is malaria. This means the Mediterranean basin, Africa, India and the Far East. In the U.S., thalassemia occurs mainly through immigrant populations of Greek and Italian origin.

"And there are 80,000 patients with symptomatic SCD in the U.S.," he continued, "mostly of African descent, who - with mutant genes inherited from both parents - are homozygous for the mutation. Another 2.5 million heterozygous people carry only one mutated SCD gene.

"For many years," Sadelain observed, "those two related disorders have been thought to be ideal model diseases for gene therapy. But they've faded away from mainstream gene therapy because of the complexity of the challenge.

"Over the years," he recalled, "a number of groups have tried to model gene therapy for these diseases in mice, using retroviruses as vectors to deliver the beta-globin gene. And the general outcome has been, they could achieve tissue-specific expression, but it was low - on the order of one or two percent of the normal level. Furthermore, these vectors tend to be silenced. Meaning, you observe increased expression for the same amounts of vectors to get less and less protein over time."

Challenge To Hit Therapeutic Levels

"We and others showed about five years ago," Sadelain went on, "that incorporating additional genetic elements - referred to as the core components of the locus control region - we could increase expression of the gene by another five- to eight-fold. That's still not enough to be therapeutically relevant. So the challenge over recent years was how to engineer a smarter, better vector that encompasses sufficient genetic information to control expression of the gene.

"Now the challenge for therapy is that we're still bound to use viruses to efficiently transfer these genes, because they are so much more efficient than current non-viral vectors. Retroviral vectors," he pointed out, "such as the conventional murine leukemia viruses, have a limited packaging capacity, so you can't just produce the large segments that are used to generate transgenic mice. So that brings up the issue of how to generate a stable vector structure - which cannot be taken for granted in the case of retroviruses.

"The reason is," Sadelain explained, "that these are RNA viruses, and once you've introduced genomic sequences into the vector, all of those sequences will now exist in single-stranded RNA form, to be packaged into virions - viral particles. So we turned to the lentiviruses two years ago - to HIV-1 in particular - because of its different mechanisms for controlling its own RNA."

Sadelain is senior author of a paper in the July 6, 2000, issue of Nature, titled "Therapeutic hemoglobin synthesis in b-thalassemic mice expressing lentivirus-encoded human b-globin."

"We show in this paper," he told BioWorld Today, "that we can get efficient gene transfer in mouse hematopoietic stem cells, that the structure is intact in those cells, that the vector expresses its cell lines certainly at the vast majority of integration sites. And in mice we showed that the beta-globin level is increased by another five- to six-fold over what we could previously achieve. And that expression is stable over six months.

"With that," Sadelain went on, "we felt that we might now be in the therapeutic range of expression levels. So we turned to a standard mouse model of beta thalassemia, originally engineered by targeted disruption of the adult murine globin gene. These heterozygous animals," he pointed out, "have a disease very similar to what is called in humans beta thalassemia intermedia. This is a form of the disease that is not quite as severe as the most severe form - thalassemia major, or Cooley's anemia.

"The homozygous mice in this model are not viable," Sadelain pointed out. "It's lethal to them; they die around birth. But the heterozygous mice have a severe anemia. Their hemoglobin is at around 8 to 9 grams per deciliter, whereas in their normal littermates it's around 15 or 16. And their blood smear is absolutely characteristic of human thalassemic patients, with deformed red cells of abnormal sizes, and hypochromic, because their hemoglobin content is low."

Sadelain made the bottom-line point that "What we showed in five out of five of those gene-transferred mice is that we could substantially and durably ameliorate their anemia. Their hemoglobin levels increased to 12 to 13 grams per deciliter. The morphology of their red cells was vastly improved, demonstrating that this vector is now tissue-specific and provides relatively elevated levels of human beta-globin expression. It's not up to the level of the protein expressed by one endogeous gene in its natural locus. It's around 15 percent of that level, but that makes it to the therapeutic range.

If Monkeys Can Cut It, Humans Next

"Beyond the relevance for beta-thalassemia, and eventually SCD," he added, "this is the first time that a gene has been successfully inserted into hematopoietic stem cells, and then regulated to this degree in the progeny cells. So we think there's something else to learn from this story, of broader interest beyond these two diseases, about how one goes about manufacturing a smarter vector that can be lineage-restricted and yet give high-level expression sustained over time. That demonstration is well under way in mice.

"Next," Sadelain concluded, "we want to embark on in vivo experiments in nonhuman primates. And the prospect of human trials will hinge on the results we obtain in these preclinical studies."