Cloak Partly Lifted on Tiny Chlamydia

Direct Damage from Radiation May Be Passed to Neighboring Cells
Cancer Research:
Surprise Route Found for Breast Cancer Spread
Dignitaries Pitch in at Groundbreaking for New Building
International Health:
HMI Partnerships

Valentines from the Bench: Studies Show New Players and Patterns in Vertebrate Heart Development

Fat Cells Tied to Whole-Body Insulin Resistance

Report Figures Child Health Care Costs, Variations Across Country

How Cells Get and Slip Their Grip While Circuiting the Body

New Appointments to Full and Named Professorships

Kogan Joins Longwood Symphony in Benefit for Dimock Health Center

HSPH Children's Health Award

Honors and Advances

News Brief

Having 'Box Checked' Signals Final Lap of Graduate Education

Front Page


Valentines from the Bench: Studies Show New Players and Patterns in Vertebrate Heart Development

HMS cell biologists have isolated the first proteins capable of inducing development of a heart—at least in frogs and birds. They report that the proteins Dkk-1 and Crescent, which inhibit regulatory proteins of the Wnt family, can induce formation of a beating heart in tissue culture from Xenopus laevis and chick embryos. Other Wnt antagonists lacked this activity. The findings indicate that diffusion of Dkk1 and Crescent from organizer centers in the embryos initiates cardiogenesis in adjacent mesoderm by establishing a zone of low Wnt activity.

This histological section shows a heart tube induced by treating noncardiogenic mesoderm with the Wnt-inhibiting protein Dkk1. In the image, the heart tube cells are red and their nuclei are pink; the surrounding blue spots are the remaining nuclei in the tissue. Courtesy of Mark Mercola

"These results are the first demonstration of factors that initiate cardiogenesis in Xenopus," write authors Valerie Schneider, former HMS graduate student and now a postdoc at the University of Pennsylvania, and Mark Mercola, HMS associate professor of cell biology, in the Feb. 1 Genes and Development. In a companion paper in the same issue of the journal, Andrew Lassar, HMS professor of biological chemistry and molecular pharmacology; first author Martha Marvin, research fellow in that department; and colleagues demonstrate an identical role for Dkk1 and Crescent proteins in the chick.

Though heart-inducing factors in amphibians have remained unknown until now, the process of heart formation is generally understood. Induction of the embryonic heart in all vertebrates occurs in paired regions of the dorsoanterior mesoderm in response to signals from the organizer region and underlying endoderm. In Xenopus, these tissues together suffice to induce heart formation in noncardiogenic ventral marginal-zone mesoderm, which lies in the embryo's equatorial region. Likewise, in birds, the underlying endoderm induces cardiogenesis in precardiac mesoderm.

Although certain bone morphogenetic proteins and fibroblast growth factors can mimic aspects of cardiogenesis in birds, neither group of proteins accounts for the full range of activities elicited by the inducing tissues. In frog development, inhibition of bone morphogenetic protein and Wnt seem to be complementary. Yet the frog study reveals that inhibition of bone morphogenetic protein signaling did not promote cardiogenesis as did Wnt inhibition. Analysis of Wnt proteins expressed during the gastrulation stage of embryogenesis showed that Wnt3A and Wnt8, but not Wnt5A or Wnt11, inhibited heart development. Consequently, the antagonists Dkk1 and Crescent appear to inhibit Wnt proteins 3A and 8.

The chick data of Marvin et al. corroborate the report by Schneider and Mercola. They show that inhibition of Wnt signaling in chick embryos promotes heart formation in the anterior lateral mesoderm while Wnt signaling on the opposite side, the posterior lateral mesoderm, promotes blood development. The authors propose a model in which two activity gradients, one for Wnt along the anterior–posterior axis and the other for bone morphogenetic protein along the dorsal–ventral axis, intersect to induce heart formation in an area of high bone morphogenetic protein and, reinforcing the companion paper, low Wnt activity.

Fat Cells Tied to Whole-Body Insulin Resistance

Though obesity is a well-known risk factor for Type II diabetes, the molecular basis of the link has remained unclear. What process might tie a disease of adipose tissue to one thought to be initiated primarily by the insulin resistance of muscle and liver? Research done by Barbara Kahn, professor of medicine at Beth Israel Deaconess Medical Center, and colleagues now shows that glucose uptake by adipose tissue is important for maintaining the body's ability to respond to insulin. Their results also point to a mechanism by which an abnormality in adipocytes may trigger insulin resistance and ultimately diabetes.

Skeletal muscle is the primary tissue for insulin-stimulated glucose uptake. Yet in obesity and Type II diabetes, expression of the primary insulin-stimulated glucose transporter, GLUT4, is normal in muscle but reduced in adipose tissue. This paradox led Kahn and colleagues to ask what role adipose-expressed GLUT4 may play in the development of insulin resistance. The group generated mice whose GLUT4 gene was selectively disrupted in adipose tissue. The mice are insulin resistant and glucose intolerant. The results, reported in the Feb. 8 Nature, clearly show that "fat is important for whole-body insulin action," said Kahn.

As predicted, skeletal muscle from these mice was capable of normal glucose uptake ex vivo. Surprisingly, the muscle was impaired in glucose uptake in vivo. "We think fat releases a molecule that circulates to muscle and liver and impairs insulin signaling in those cells," said Kahn. A previously unidentified molecule or molecules are likely to be at work.

The link between adipose GLUT4 levels and the ability to maintain normal blood glucose levels may explain why obesity is a risk factor for Type II diabetes.

—Heather Ettinger

Report Figures Child Health Care Costs, Variations Across Country

A new report by researchers at the Harvard Center for Children's Health at HSPH and the federal Agency for Healthcare Research and Quality provides the first nationwide data on children's health expenditures in more than a decade.

Annual medical costs for a single child averaged $1,019 in 1996, with nearly 21 percent paid out of pocket by families. For children with major health problems, costs rose to $11,000. About two thirds of American children are covered by private insurance, 19 percent by public sources, and 15 percent are uninsured.

The researchers also compared hospital care across 22 states, revealing wide differences in average length of hospitalization (ranging from 2.7 days in Oregon to 4.0 days in New York) and proportion of hospital admissions through the emergency department (from 9 percent in Utah and Oregon to 23 percent in New York).

"There are large and unexplained differences in state-by-state utilization," said Marie McCormick, director of the center and lead author of the report. "Some are explained by historic and geographical differences such as reliance on large county hospitals. These differences beg for further analysis. The availability of the data should encourage more extensive study by people involved in children's care." The second in an annual series on children's access to health care, the report is published in the inaugural January/February issue of Ambulatory Pediatrics (www.ambpeds.org).

How Cells Get and Slip Their Grip While Circuiting the Body

Cells shooting through the vessels of the body face a mighty challenge when attempting to come to a stop. Not only must they form a bond with a molecule on the vessel substrate, they must maintain this bond if they are to remain stationary.

"Cells experience large forces in the body which can break them away from their proper location. So they have to resist lots of force," said Timothy Springer, the Latham Family professor of pathology.

To see how force affects the rate at which receptor–ligand bonds dissociate, researchers have devised a variety of equations. But most of these had remained mere abstractions. "Nobody had a data set that was good enough to compare these theories to see which of five different equations fit the best," Springer said.

He and Shuqi Chen, HMS instructor in pathology, set about to remedy the situation. They sent neutrophils through a flow chamber lined with a P-selectin substrate and observed the duration of the bonds formed between the two. Then they correlated this information with the amount of force exerted in the chamber. Force was determined by the rate of flow and the viscosity of the buffer inside the flow chamber.

According to their results, which are presented in the Jan. 30 Proceedings of the National Academy of Sciences, a particular equation, the Bell model, best fits the data. In addition, by varying the two components of force—the rate of flow and viscosity—they uncovered an intriguing distinction. They found that while changing the rate of flow affected the rate at which bonds both form and dissociate, variations in viscosity affected only the rate at which bonds break.