Multifactorial diseases such as cancer, Alzheimer’s and diabetes usually originate from complex interactions of genetic and environmental factors. Research has shown that diet and physical exercise as well as exposure to toxins and pathogens can change the way genes are expressed without altering DNA structure, enhancing protection of the organism or its predisposition to the development of a given pathology.
Epigenetic changes, which are a set of biochemical processes triggered by environmental stimuli that modify the functioning of the genome and hence the phenotypic profile by switching specific genes on or off, were the focus of the Thematic Project “Chromatin structure and organization with aging and diabetes compared to induced changes in epigenetic markers”, coordinated by Maria Luiza Silveira Mello, a professor at the University of Campinas’s Biology Institute (IB-UNICAMP). In connection with this project, Marina Barreto Felisbino has investigated the use of certain drugs for epigenetic modulation for the treatment of diabetes, supported by a doctoral scholarship from FAPESP.
Among the known epigenetic mechanisms are DNA methylation – addition of a methyl group (carbon and hydrogen atoms) to the DNA base cytosine, potentially preventing expression of some genes – and histone modification, in which acetyl (carbon, oxygen and hydrogen), methyl or other groups are added to or subtracted from the amino acid residues in histones, a family of basic proteins that associate with DNA in the cell nucleus and help condense it into chromatin.
Just as these biochemical processes are often among the causes of certain diseases, they may also be part of a cure, scientists believe. Based on this hypothesis, several research groups have performed preliminary experiments to test the extent to which drugs can induce epigenetic changes for the treatment of complex pathologies.
“There’s evidence that DNA methylation and histone modification regulate the glucose and insulin metabolism pathways,” Felisbino said. “In the situation of hyperglycemia, these epigenetic markers are dysregulated. Our hypothesis is that we can restore the normal state by using an epigenetic drug.”
According to Mello, the line of research pursued by Felisbino aims to test drugs such as valproic acid and trichostatin, both of which are histone deacetylase inhibitors, for the treatment of diabetes.
“We decided to use liver cells as a model because of the liver’s overall importance to metabolism,” Mello said. “The liver normally releases glucose into the bloodstream only to compensate for a shortage of nutrients. But insulin resistance or deficiency can cause an increase in glucose production by the liver, and this is a central event in the development and progression of diabetes.”
One of the aims of the study was to understand how liver cell chromatin responds both to hyperglycemia and to exposure to histone deacetylase inhibitors.
“This knowledge helps us understand the consequences of treatment with drugs like valproic acid, widely used to treat epileptic seizures and more recently proposed as a histone deacetylase inhibitor,” Mello said.
A first experiment was performed to find out whether hyperglycemia alone induced modifications in liver cell chromatin without any interference by modulating drugs. According to Felisbino, chromatin was the main target of the analysis because “it’s the scene of epigenetic control.”
The group compared cells cultured in a medium with normal glucose levels and cells in a hyperglycemic medium using confocal microscopy and image analysis software to measure the nuclear area, perimeter, density and compactness.
Flow cytometry was used to gauge the abundance of certain histone modifications identified in previous research as being relevant to the modulation of gene expression.
“Some histone modifications are classic and play a well-understood role,” Felisbino said. “For example, H3K9 acetylation generally relates to more open and more active chromatin, meaning it’s associated with an increase in gene expression, whereas H3K9 dimethylation relates to more closed and compacted chromatin and hence to a decrease in gene expression.”
The results of this first experiment showed that hyperglycemia makes chromatin less compacted and more open to gene expression. At this stage, the analysis did not focus on specific genes but rather evaluated gene expression in general.
The next step was to treat liver cells in the normoglycemic and hyperglycemic media with valproic acid and trichostatin. Because these drugs are histone deacetylase inhibitors, they are associated with an increase in histone acetylation and hence with chromatin decompaction and increased gene expression.
“In normoglycemic cells, they did indeed act in this way, increasing the size of the nucleus and the abundance of epigenetic markers associated with more decompacted and active chromatin,” Felisbino said. “We expected this effect to be more pronounced in hyperglycemic cells but were proved mistaken. The difference between the two cultures was statistically non-significant.”
Thus, from a morphological standpoint, hyperglycemia and treatment with histone deacetylase inhibitors induced similar changes in liver cell chromatin. However, Felisbino explained, that does not mean both conditions increased gene expression in the same way: different genes could have been modified.
“We saw an example where this modulation was different when we analyzed expression of the gene for DNA methyltransferase, an enzyme responsible for catalyzing the transfer of methyl radicals to DNA and hence an important agent for chromatin remodeling. Valproic acid induced a decrease in expression of this gene, and hyperglycemia changed nothing,” she said.
This suggests that the level of DNA methylation may have decreased in the cells treated with valproic acid, according to Felisbino. If so, the decrease is also an epigenetic modification that can be linked to reduced gene expression. “It’s only a possibility. To be absolutely sure, we’d have to evaluate the entire cascade of enzymes involved in DNA methylation,” she said.
In a new stage of the research project, Felisbino is investigating the effect of chromatin remodeling induced by valproic acid on each of the protein-coding genes in the human genome.
“We plan to analyze the level of these histone modifications gene by gene and to see whether gene expression increases or decreases,” she said. “The goals are to identify key pathways that are altered and to understand the mechanism of the drug’s action.”
“I’m now analyzing the data, which is the part that takes the longest,” she said. “Once we’ve identified the action mechanism, we’ll test it in cultured cells and in mice.”
According to Mello, although valproic acid is frequently used in clinical medicine, its possible side effects in longer treatments and at higher doses are unknown.
Trichostatin is also used in humans, albeit experimentally, to treat cancer and prevent premature births.