Laser-generated bubbles take aim at a cell membrane

Ever since World War I, when Lord Rayleigh laid the theoretical foundations for the dynamics of collapsing bubbles that were eroding the propeller blades on the Royal Navy’s ships and submarines, experiments have borne out the violent nature of the cavitation process. During a bubble’s collapse, the gas inside it can reach temperatures of more than 15 000 K—as hot as the surface of a star—and the energy can be released in the form of shock waves, heat, light, or turbulent vortices and high-speed jets of fluid. (See the article by Detlef Lohse in Physics Today, February 2003, page 36 .)

For decades, medical researchers have worked to harness that energy release for therapeutic applications such as the disintegration of cancerous tumors using focused ultrasound (see the article by Gail ter Haar in Physics Today, December 2001, page 29 ) and the delivery of drugs or genes into living cells (see Physics Today, December 2005, page 22 ). Although studies have demonstrated that cavitating bubbles can rupture nearby cells, thanks to the high shear and pressure forces that their expansion and collapse generate in surrounding fluid, a complete understanding of the bubble-cell interaction and how best to control it remains elusive.

Duke University researchers Georgy Sankin, Fang Yuan, and Pei Zhong have now developed an experimental approach to opening a cell’s membrane that entails carefully manipulating the fluid dynamics around it. ¹ The key to their approach is the use of two laser-generated bubbles that act in concert to direct a tiny jet of fluid into a target cell.

This high-speed sequence of photographs captures the process. The researchers focus two 5-ns laser pulses, each about 30 µJ, near one of thousands of cancer cells held in a microfluidic chamber. Heat from the first laser pulse vaporizes a pocket of liquid, which expands adiabatically into a microbubble (B₁). After 4 µs the second pulse creates another microbubble (B₂), 40 microns away, whose own rapid expansion causes the collapse of the first by pressing against it. Both bubbles lose their spherical symmetry, with B₂ drawn into the wake left by the collapsing B₁. The asymmetric collapse of both bubbles gives rise to two localized microjets that shoot in opposite directions along the bubbles’ axis: the first, between 6 and 7 µs, a downward-directed thin fluid spike that pricks a 2-um pore in the cell, and the second, between 8 and 11 µs, an upward-directed spike.

As one camera captures the bubbles’ interaction, a second records the fluid dynamics from tiny polystyrene tracers caught up in the flow. The particle image velocimetry (PIV) indicates that, as B₁ collapses, fluid rushes in from both sides. Zhong argues that the higher pressure there, along with a higher surface tension due to greater local curvature, causes the lower edge of B₂ to then collapse faster than the rest of it at 8 us. Indeed, the resulting microjet is so forceful that a nipple appears on the opposite side of the bubble.

According to the PIV measurements, a microjet typically flows at 10 m/s and sets up a pair of vortices—one clockwise, the other counterclockwise in the image plane—that can spin as rapidly as 350 000 s^-1 and persist for hundreds of microseconds. The jet-induced flow and vorticity generate a shear stress of about 1 kPa, which stretches and bends the cell membrane.

“The beauty of the Duke experiment,” says Nanyang Technological University’s Claus-Dieter Ohl, “is the level of control and precision it offers in a microfluidic setting.” The researchers can adjust the bubbles’ positions, separation, and orientation relative to any cell of interest. What’s more, the flexibility of their setup allows them to photograph the bubble dynamics simultaneously either with the PIV measurements—at about a million frames per second—or with measurements that capture the subsequent uptake and diffusion of dye molecules into the ruptured cell over a minute’s time.

Afterward, the team can chemically preserve the cell, mark its location with a laser spot burned on the Petri dish, and easily find the pore for later microscopy, as shown in the final image.