Molecular Tissue Profiling Reveals Origins of Infantile Hemangioma

In a novel application of molecular tissue profiling techniques to test a biological hypothesis, researchers have gained new insights into the source of infantile hemangioma. Their findings in the Dec. 27, 2005 Proceedings of the National Academy of Sciences support the proposition that hemangiomas derive from placental endothelial cells.

Image courtesy of Carmen Barnés

Pictures of health and disease. Four GEDI portraits of different tissue types (including scleroderma [Sc] and the pulmonary tumors small cell carcinoma [SmC] and squamous carcinoma [Sq]) reveal similarities between derived cells and their origins as well as contrasts with unrelated tissues. Each colored pixel in a portrait represents the expression of a minicluster of genes, red for strong, yellow for weak, and blue if absent.

Each year, approximately one percent of full-term and 25 percent of premature babies develop a hemangioma, an abnormally dense collection of capillaries in the skin, liver, or other organs. These tumors normally start fading on their own after a year, but for a small percentage of afflicted infants the tumors can be blinding or life-threatening.

The origin of the hemangioma cells has remained uncertain, with several competing hypotheses. Recent research favored a placental origin, showing that hemangioma and placental endothelial cells share the unique expression of several proteins. Using microarray-based gene expression profiling, lead author Carmen Barnés, HMS instructor in surgery at Children’s Hospital Boston, confirmed this hypothesis at the molecular level.

Barnés examined transcriptomes — genomewide profiles of RNA — from muscle, skin, lung, placenta, hemangioma, and brain tissues, analyzing more than 7,800 genes from each tissue type to get holistic portraits of the gene expression patterns. Using both hierarchical and non-hierarchical clustering analysis, Barnés and colleagues analyzed the gene expression profiles, a kind of “complex finger print” of the RNA makeup of each tissue. GEDI software, developed by co-author Sui Huang, HMS assistant professor of surgery at Children’s, rendered the non-hierarchical images. “The beauty of GEDI is that you actually get a visual portrait,” said Barnés, “so differences and similarities become apparent right away.”

Patterns emerged in the profiles revealing close relationships between cells and their source tissues. Scleroderma tissue closely resembled the normal skin from which it originated. Hemangioma and placental profiles showed striking similarities and clear differences from other samples. “The endothelial genes were almost identical, which nails down that it’s the same tissue,” said Judah Folkman, the Julia Dyckman Andrus professor of pediatric surgery at Children’s and corresponding author on the paper.

Because most hemangiomas eventually disappear, researchers suspect the cells are programmed to stop growing, the same way that the placenta starts to break down at term. Future research will focus on understanding this program along with determining how placental cells are transported to the fetus.

This unusual use of genome-scale bioinformatics to test a biological hypothesis (rather than to screen for new genes) and the emphasis on distinctive images opens the door for other novel applications. According to Folkman, clinicians may someday read molecular profiles for visual diagnosis of genetic diseases the same way pathologists can see patterns in tissue microsections.

—Elizabeth Dougherty

Success of Depression Treatment Tied to Socioeconomic Status

Individuals with low socioeconomic status who are suffering from depression have worse treatment outcomes than those with middle and high socioeconomic status who are depressed, even when given the best care, according to a study in the January 2006 Archives of General Psychiatry.

“[Socioeconomic] disparities in health are usually explained through access to and quality of care. What this study suggests—preliminarily and with one specific age group—is that even when everyone gets the same care and it is of the highest quality, there are still disparities,” said lead author Alex Cohen, HMS assistant professor of social medicine.

Evidence for this comes from two clinical trials that assess the efficacy of combined pharmacologic and psychosocial treatments for depression in adults age 59 and older. Cohen and colleagues reanalyzed the data to examine how household income and education affected the subjects’ responses to treatment. The results demonstrated that individuals in the low socioeconomic group were significantly less likely to respond to treatment than those in the middle and high socioeconomic groups. They were also twice as likely to report suicidal thoughts as those in the middle and two and a half times as likely as those in the high socioeconomic group.

Previous research showing that those with low socioeconomic status (SES) have more severe courses of depression may help explain why individuals with low socioeconomic status were less likely to respond to treatment. “If individuals with lower SES tend to have depressive episodes of greater duration, rates of difficult-to-treat depression may also be elevated in this population,” the authors write. As to why this group also reports relatively high rates of suicidal feelings, Cohen said, “We can’t answer that definitively, but I would guess that it has something to do with diminished access to resources and less control over their lives.”

One unexpected finding was the absence of a correlation between education and response to treatment. Cohen offers a historical explanation. The individuals in the clinical trials worked in an economic environment in which individuals did not necessarily need education to achieve economic and social success.

To better understand how socioeconomic status influences treatment outcomes, Cohen said, medical researchers conducting clinical trials should collect socioeconomic data and recruit subjects from a wide cross-section of socioeconomic groups. “Investigators need to be aware of the importance of socioeconomic status and how it may influence the effectiveness of interventions.”

—Elizabeth Dougherty

New Diabetes Drug Acts Through Key Player in Insulin Pathway

A new class of drug for treating type 2 diabetes debuted last year with the Food and Drug Administration’s approval of exenatide (Byetta), a synthetic version of exendin-4. The medication triggers pancreatic beta cells to secrete more insulin, overcoming some of the effects of insulin resistance. A recent study in mice by Morris White, a Howard Hughes investigator and HMS professor of pediatrics at Children’s Hospital Boston, and his collaborators explored exendin-4’s mechanism of action, discovering that it works through the insulin receptor substrate 2 (IRS2) branch of the insulin/IGF signaling pathway.

The researchers had observed that exendin-4 yielded the same effects—increased insulin production and glucose tolerance—in people as they had seen in mice that in previous work they had genetically engineered to produce the Irs2 protein. They examined the correspondence by administering exendin-4 to mice lacking Irs2 and found, as reported in the Jan. 13 Journal of Biological Chemistry, that these animals did not experience the long-term benefits of exendin-4, though the drug did temporarily delay the onset of diabetes.

In addition to regulating brain function, reproduction, and satiety, Irs2 ensures the survival and function of beta cells. Without Irs2, mice develop diabetes because their beta cells do not release enough insulin to compensate for insulin resistance, even when treated with exendin-4.

“These results reveal an unexpected link between a new treatment for type 2 diabetes in people and our genetic experiments in mice,” said White. “We can conclude that you need IRS2 to get the effects of exendin-4.”

Genetic mutations such as those carried by the experimental mice in White’s lab do not cause type 2 diabetes in humans, however. “In fact,” said White, “such genetic defects would be devastating for people.” Rather, environmental conditions such as chronic stress, inflammation, obesity—many of the conditions associated with the onset of type 2 diabetes—degrade or shut down IRS2 functions in human tissues. The authors show that exendin-4 boosts IRS2 signaling in isolated human pancreatic islets, explaining in part how exendin-4 may work in people to prevent diabetes.

Since IRS2 signaling plays a central role in regulating appetite, energy balance, and insulin action, finding other ways to enhance it might be a starting point for improved diabetes therapy.

— Elizabeth Dougherty

top