It's never a good thing when donated human blood – or even the blood in our bodies – is infected with bacteria. Scientists at the University of California San Diego, however, are developing a means of removing such blood-borne microbes using tiny ultrasound-powered robots.

The base "nanorobots" were created by a team led by Prof. Joseph Wang, and are made of microscopic lengths of gold nanowire. Via the external application of ultrasound, they can be propelled through liquids including blood, causing them to get thoroughly mixed with it.

In a process developed by Prof. Liangfang Zhang and colleagues, these nanorobots were coated in a hybrid of platelet and red blood cell membranes. This coating was created by first separating the membranes from their respective cells, using high-frequency sound waves to fuse the two different types of membranes together, and then applying them to the gold nanowires via a chemical process.

The resulting coated bots are approximately one twenty-fifth the size of the width of a human hair, and when subjected to ultrasound can travel through blood at a rate of 35 micrometers per second. As they do so, pathogens such as MRSA (methicillin-resistant Staphylococcus aureus) bacteria are bonded to surface proteins in the platelet membranes, while toxins created by those bacteria are neutralized by the red blood cell membranes.

In lab tests, the nanobots were used to treat MRSA-tainted blood samples. Within five minutes, the treated samples had only one third the amount of bacteria and toxins as their untreated counterparts. Plans now call for tests to be performed on live animals, plus the scientists are looking into substituting non-toxic biodegradable materials for the gold.

"By integrating natural cell coatings onto synthetic nanomachines, we can impart new capabilities on tiny robots such as removal of pathogens and toxins from the body and from other matrices," says Wang. "This is a proof-of-concept platform for diverse therapeutic and biodetoxification applications."

A paper on the research was recently published in the journal Science Robotics.