Histone deacetylases are best known for controlling gene expression by modifying chromosomal structure to make genes either more or less accessible to the cell's transcription machinery. That seat in the control tower has made them the focus of investigations into multiple processes in health and disease.

Now, in the Nov. 15, 2004, issue of Cell, researchers from the University of Texas Southwestern Medical Center in Dallas and Baylor College of Medicine in Houston add another job to the growing list: regulating bone formation. The findings are reported in a paper titled "Histone Deacetylase 4 Controls Chondrocyte Hypertrophy During Skeletogenesis."

HDACs are divided into three classes based on criteria that include size, cellular localization and mechanism of action. Research in the laboratory of Eric Olson, professor and chair of the department of molecular biology at the University of Texas Southwestern Medical Center and senior author of the study, focuses on Class II HDACs. His work previously had implicated two other HDACs in controlling the formation of cardiac cells.

So when Olson and his group created the HDAC4 knockout mouse, they had "absolutely no expectation" that that particular histone deacetylase would turn out to be involved in bone formation instead, he told BioWorld Today. "We expected it would control heart growth, like the others did. In fact, when the [HDAC4 knockouts] started dying, it took us quite a long time to figure out what was wrong with these animals."

Taking The Brakes Off The System'

Tissue-staining experiments finally showed what was wrong - too much bone.

In normal bone development, with the exception of the shoulder blade and the top of the head, bones ossify in several steps. Initially, mesenchymal cells known as chondrocytes form template cartilaginous bones. Those mesenchymal cells undergo hypertrophy - that is, they stop dividing, then they enlarge, and ultimately, undergo apoptosis. They are replaced by osteoblasts, or bone stem cells, which differentiate into bone cells.

In the HDAC4 knockouts, that process - specifically, the chondrocyte switch to hypertrophy - was out of control. Regular bones were prematurely hardening, and even parts of the skeleton that normally remain cartilaginous, such as the front of the rib cage and sutures of the skull, were turning into bone. Phenotypically, the net result was retarded growth, the inability to move, difficulty breathing and, finally, premature death.

Bone development already was known to be up-regulated by transcription factor Runx2; in fact, the phenotype of the HDAC was remarkably similar to that of previously generated mouse models that overexpress Runx2. Runx2 acts both by promoting chondrocyte hypertrophy and by binding to itself in a positive feedback loop. Olson described HDAC4 as "the governor of this system, needed to precisely control chondrocyte hypertrophy. So knocking it out is like letting a car roll downhill and then taking the brakes off," allowing bone formation to spiral out of control. The scientists showed that Runx2 was indeed up-regulated in rib-cage cartilage of HDAC4 knockouts.

The mechanism of that particular role of HDAC4 does not appear to be histone deacetylation, at least not directly. Instead, cell culture experiments showed that HDAC4 binds to Runx2, suggesting that it normally interacts with Runx2 to inhibit the latter's target genes.

"HDAC4 has a histone deacetylase domain, but interestingly, other Class II HDACs have been shown to repress gene expression independent of their own intrinsic deacetylase domain. In fact, there are splice variants of other Class II HDACs that don't even have a catalytic region and can still repress gene expression. One mechanism for such repression is through recruitment of Class I HDACs. Another may be through disruption of DNA binding, as we showed here for Runx2. I believe both mechanisms are operating in the case of HDAC4," Olson said.

One More Clinical Target For HDAC Inhibitors

Olson, who described himself as "keenly interested" in the clinical possibilities suggested by his work, said that there had been a lot of interest from industry, as well. He named various potential areas that HDAC4 manipulation in particular could ultimately prove to be of therapeutic benefit, including the regeneration of damaged cartilage (runners, take note) and the control of bone formation. In the short term, "one thing we have done is to create a mouse with a conditional HDAC4 allele." Turning HDAC4 off in adulthood might allow insights into osteoporosis and diseases, as well as possible treatments.

Olson said that part of the reason for industry's interest in the research is that "HDAC inhibitors are being aggressively pursued in the clinic now, and show great efficacy and tolerability." Indeed, HDAC inhibitors are the molecule of the moment; they are in all phases of clinical trials for the treatment of various cancers, and also are being investigated for their effects in other diseases, including spinal muscular atrophy, lupus-associated kidney damage, diabetes and, in Olson's own lab, cardiac disorders.

And in October, Cambridge, Mass.-based Gloucester Pharmaceuticals Inc. announced its plans to start a pivotal trial in cutaneous T-cell lymphoma with its HDAC inhibitor FK228. (See BioWorld Today, July 14, 2004; Sept. 22, 2004; and Oct. 15, 2004.)