Mark Krasnow

But Metzger's medical studies and his scholarly research skills made him “the perfect person to set the whole project in motion,” Krasnow says. “Because he was new to biological research, he had the freedom to think big and not feel constrained by the everyday challenges that sometimes sour people who have been working in the field a long time.”

Metzger developed an antibody stain to make the airways glow bright green under the microscope. He studied embryonic mouse lungs from the 11th to the 15th day of development, a period when the first branches budding off the trachea bloom into thousands, filling out the shape of the lungs' five lobes.

“Even within the same litter of embryos, the lungs are not all at the same stage at the same time,” Metzger says. “They may differ by a single branch.” He stared at the eerie green airways under the microscope day after day, searching hundreds of specimens for these tiny differences in order to place the lungs in a developmental sequence. (“It does help that they're beautiful,” Metzger says.) Gradually, over many years of work, the branching pattern started to emerge, and its elegance was startling.

Metzger and Krasnow deduced that the complex network of airways arises from just three simple types of branching. During domain branching, airways sprout in rows from the sides of an existing branch like bristles on a bottlebrush. But the domains don't grow haphazardly; the branches start from the near end and move toward the far end within each domain, with new domains spiraling around the circumference. Domain branching sets out a scaffold for each lobe of the lung, determining its overall shape.

From these domains, the airways branch in two different ways. During planar bifurcation, an airway forks into two in the plane of the original branch, helping to form the thin edges of the lobes. During orthogonal bifurcation, the airway splits perpendicular to the plane, which helps form the surfaces of the lobes and fill out their shape. The researchers describe the process in the June 5, 2008, issue of Nature.

As if in a computer program, each of these types of branching forms a “subroutine” that gets called on at predictable stages. “What we'd like to do now is start assigning genes and gene products to translate the computational model into a genetic and molecular model,” Krasnow says. Their strategy is to look at mice with mutations in genes known to be expressed in the lungs. Before, such mutants might not have shown obvious lung defects, but perhaps the defects were so subtle that they escaped notice. Metzger is also working with another group to do computer simulation of the branching using the rules discovered in mapping the normal mouse lung.

“We're hoping that understanding the process in such detail will not only reveal interesting developmental principles but will also lead to novel types of treatments for lung disease,” Krasnow says. Premature infants often have problems with lung function, and studies have found that differences in lung development might predispose people to respiratory disease later in life. The scholars' Torah of the lung took a decade to write; now, scroll in hand, Metzger and Krasnow say they have enough to fuel their research for decades to come.

Photo: Tim Archibald

	PAGE 2 OF 2	Back