NEWSWISE — HOUSTON — ( Nov. 7, 2014 ) — By transforming human scar cells into blood vessel cells, scientists at Houston Methodist may have discovered a new way to repair damaged tissue. The method, described in an upcoming issue of Circulation (early online), appeared to improve blood flow, oxygenation, and nutrition to areas in need.

Cardiovascular scientists at Houston Methodist, with colleagues at Stanford University and Cincinnati Children’s Hospital, learned that fibroblasts — cells that causes scarring and are plentiful throughout the human body — can be coaxed into becoming endothelium, an entirely different type of adult cell that forms the lining of blood vessels.

“To our knowledge, this is the first time that trans-differentiation to a therapeutic cell type has been accomplished with a small molecules and proteins,” said Houston Methodist Research Institute Department of Cardiovascular Sciences Chair John Cooke, M.D., Ph.D., the study’s principal investigator. “In this particular case, we’ve found a way to turn fibroblasts into ‘shapeshifters’ nearly on command.”

Cooke said the regenerative medicine approach provides proof-of-concept for a small molecule therapy that could one day be used to improve the healing of cardiovascular damage or other injuries.

Other research groups have managed to generate endothelial cells cells using infectious virus particles specially engineered to deliver gene-manipulating DNA to cells. The DNA encodes proteins called transcription factors to alter gene expression patterns in such a way that cells behave more like endothelial cells.

“There are problems with using viruses to transfer genes into cells,” Cooke said. “This gene therapy approach is more complicated, and using viral vectors means the possibility of causing damage to the patient’s chromosomes. We believe a small-molecule approach to transforming the cells will be far more feasible and safer for clinical therapies.”

The new method described by Cooke and his coauthors starts with exposing fibroblasts to poly I:C (polyinosinic:polycytidylic acid), a small segment of double-stranded RNA that binds to the host cell receptor TLR3 (toll-like receptor 3), tricking the cells into reacting as if attacked by a virus. Cooke and coauthors reported to Cell in 2012 that fibroblasts’ response to a viral attack — or, in this case, a fake viral attack — appears to be a vital step in diverting fibroblasts toward a new cell fate. After treatment with poly I:C, the researchers observed a reorganization of nuclear chromatin, allowing previously blocked-off genes to be expressed. The fibroblasts were then treated with factors, such as VEGF, that are known to compel less differentiated cells into becoming endothelial cells.

Cooke and his colleagues reported to Circulation that about 2 percent of the fibroblasts were transformed from fibroblasts into endothelial cells, a rate comparable to what other research groups have accomplished using viruses and gene therapy. But Cooke said preliminary, as-yet-unpublished work by his group suggests they may be able to achieve transformation rates as high as 15 percent.

“That’s about where we think the yield of transformed cells needs to be,” Cooke said. “You don’t want all of the fibroblasts to be transformed — fibroblasts perform a number of important functions, including making proteins that hold tissue together. Our approach will transform some of the scar cells into blood vessel cells that will provide blood flow to heal the injury.”

In a second part of the study, the scientists introduced the transformed human cells into immune-deficient mice that had poor blood flow to their hind limbs. The human blood vessel cells increased the number of vessels in the mouse limb, improving circulation.

“The cells spontaneously form new blood vessels — they self assemble,” Cooke said. “Our transformed cells appear to form capillaries in vivo that join with the existing vessels in the animal, as we saw mouse red blood cells inside the vessels composed of human cells.”

Cooke, who is also the director of the Houston Methodist Center for Cardiovascular Regeneration, said that figuring out how to manipulate adult cells of one type into becoming a completely different type of cell will be an important part of the development of regenerative medicine as a scientific and clinical field. Humans are generally unable to regenerate heavily damaged tissue, whereas other animals, such as some newts and flat worms, can regenerate entire lost limbs — even entire heads.

“It is likely that modifications of this small molecule approach may be used to generate other body cells of therapeutic interest,” Cooke said. “What we are seeing is evidence of the fluidity of cell fate with the proper stimulation. If we can understand the underlying pathways and how to manipulate them, we may very well learn how reawaken primordial mechanisms for regeneration that are active in lower vertebrates such as newts.”

Cooke said more animal model studies are needed before his group begins clinical trials.

“One of the next steps will be to see if we can rescue an animal from an injury,” Cooke said. “We want to know if the therapy enhances healing by increasing blood flow to tissues that may have been damaged by a loss of blood because of ischemia.”

###Also contributing to the Circulation paper were Nazish Sayed, M.D., Ph.D., Wing Tak Wong, Ph.D., Frank Ospino, and Shu Meng, Ph.D. (Houston Methodist Research Institute), Jieun Lee, Ph.D. and Arshi Jha (Stanford University School of Medicine), and Phillip Dexheimer and Bruce J. Aronow, Ph.D. (Cincinnati Children’s Hospital Medical Center). Work was funded by the National Institutes of Health (U01HL100397, RC2HL103400, HL098049-01A1, and PCBC_JS_2012/1_02) and the American Heart Association (13SDG17340025 and 13SDG15800004).

To speak with John Cooke, please contact David Bricker, Houston Methodist, at 832-667-5811 or dmbricker@houstonmethodist.org.

By Sarah Yang, Media Relations | October 14, 2014

BERKELEY — UC Berkeley scientists have taken proteins from nerve cells and used them to create a “smart” material that is extremely sensitive to its environment. This marriage of materials science and biology could give birth to a flexible, sensitive coating that is easy and cheap to manufacture in large quantities.

The work, published today (Tuesday, Oct. 14) in the journal Nature Communications, could lead to new types of biological sensors, flow valves and controlled drug release systems, the researchers said. Biomedical applications include microfluidic devices that can handle and process very small volumes of liquid, such as samples of saliva or blood, for diagnostics.

“This work represents a unique convergence of the fields of biomimetic materials, biomolecular engineering and synthetic biology,” said principal investigator Dr. Sanjay Kumar, UC Berkeley professor of bioengineering. “We created a new class of smart, protein-based materials whose structural principles are inspired by networks found in living cells.”

Kumar’s research team set out to create a biological version of a synthetic coating used in everyday liquid products, such as paint and liquid cosmetics, to keep small particles from clumping together. The synthetic coatings are often called polymer brushes because of their bristle-like appearance when attached to the particle surface.

To create the biological equivalent of a polymer brush, the researchers turned to neurofilaments, pipe cleaner-shaped proteins found in nerve cells. By acting as tiny, cylindrical polymer brushes, neurofilaments collectively assemble into a structural network that helps keep one end of the nerve cell propped open so that it can conduct electrical signals.

“We co-opted this protein and turned it into a polymer brush by cloning a portion of a gene that encodes one of the neurofilament bristles, re-engineering it such that we could attach the resulting protein to surfaces in a precise and oriented way, and then expressing the gene in bacteria to produce the protein in large, pure quantities,” said Kumar. “We showed that our ‘protein brush’ had all the key properties of synthetic brushes, plus a number of advantages.”

Kumar noted that neurofilaments are good candidates for protein brushes because they are intrinsically disordered proteins, so named because they don’t have a fixed 3-D shape. The size and chemical sequence of these hair-like proteins are far easier to control when compared with their synthetic counterparts.

“In biology, precision is critical,” said Kumar. “Proteins are generally synthesized with the exact same sequence every time; the length and biochemical order of the protein sequence affects all of its properties, including structure and the ability to bind to other molecules and catalyze biochemical reactions. This kind of sequence precision is difficult if not impossible to achieve in the laboratory using the tools of chemical synthesis. By harnessing the precision of biology and letting the bacterial cell do all the work for us, we were able to control the exact length and sequence of the bristles of our protein brush.”

The researchers showed that the protein brushes could be grafted onto surfaces, and that they dramatically expand and collapse in reaction to changes in acidity and salinity. Materials that are environmentally sensitive in this way are often referred to as “smart” materials because of their ability to adaptively respond to specific stimuli.

Other co-authors on the paper are Nithya Srinivasan, Maniraj Bhagawati, and Badriprasad Ananthanarayanan, all of whom are postdoctoral fellows in Kumar’s lab.

This work was supported by a Presidential Early Career Award for Scientists and Engineers from the Army Research Office and a Hellman Faculty Fund Award, both given to Kumar.

U-M-led team shows key role of protein called NT3 in ear-to-brain communication, and as a target for future treatments

Released: 20-Oct-2014 3:00 PM EDT
Source Newsroom: University of Michigan Health System

Newswise — ANN ARBOR, Mich. — Scientists have restored the hearing of mice partly deafened by noise, using advanced tools to boost the production of a key protein in their ears.

By demonstrating the importance of the protein, called NT3, in maintaining communication between the ears and brain, these new findings pave the way for research in humans that could improve treatment of hearing loss caused by noise exposure and normal aging.

In a new paper in the online journal eLife, the team from the University of Michigan Medical School’s Kresge Hearing Research Institute and Harvard University report the results of their work to understand NT3’s role in the inner ear, and the impact of increased NT3 production on hearing after a noise exposure.

Their work also illustrates the key role of cells that have traditionally been seen as the “supporting actors” of the ear-brain connection. Called supporting cells, they form a physical base for the hearing system’s “stars”: the hair cells in the ear that interact directly with the nerves that carry sound signals to the brain. This new research identifies the critical role of these supporting cells along with the NT3 molecules that they produce.

NT3 is crucial to the body’s ability to form and maintain connections between hair cells and nerve cells, the researchers demonstrate. This special type of connection, called a ribbon synapse, allows extra-rapid communication of signals that travel back and forth across tiny gaps between the two types of cells.

“It has become apparent that hearing loss due to damaged ribbon synapses is a very common and challenging problem, whether it’s due to noise or normal aging,” says Gabriel Corfas, Ph.D., who led the team and directs the U-M institute. “We began this work 15 years ago to answer very basic questions about the inner ear, and now we have been able to restore hearing after partial deafening with noise, a common problem for people. It’s very exciting.”

Using a special genetic technique, the researchers made it possible for some mice to produce additional NT3 in cells of specific areas of the inner ear after they were exposed to noise loud enough to reduce hearing. Mice with extra NT3 regained their ability to hear much better than the control mice.

Now, says Corfas, his team will explore the role of NT3 in human ears, and seek drugs that might boost NT3 action or production. While the use of such drugs in humans could be several years away, the new discovery gives them a specific target to pursue.

Corfas, a professor and associate chair in the U-M Department of Otolaryngology, worked on the research with first author Guoqiang Wan, Ph.D., Maria E. Gómez-Casati, Ph.D., and others in his former institution, Harvard. Some of the authors now work with Corfas in his new U-M lab. They set out to find out how ribbon synapses – which are found only in the ear and eye – form, and what molecules are important to their formation and maintenance.

Anyone who has experienced problems making out the voice of the person next to them in a crowded room has felt the effects of reduced ribbon synapses. So has anyone who has experienced temporary reduction in hearing after going to a loud concert. The damage caused by noise – over a lifetime or just one evening – reduces the ability of hair cells to talk to the brain via ribbon synapse connections with nerve cells.

Targeted genetics made discovery possible

After determining that inner ear supporting cells supply NT3, the team turned to a technique called conditional gene recombination to see what would happen if they boosted NT3 production by the supporting cells. The approach allows scientists to activate genes in specific cells, by giving a dose of a drug that triggers the cell to “read” extra copies of a gene that had been inserted into them. For this research, the scientists activated the extra NT3 genes only into the inner ear’s supporting cells.

The genes didn’t turn on until the scientists wanted them to – either before or after they exposed the mice to loud noises. The scientists turned on the NT3 genes by giving a dose of the drug tamoxifen, which triggered the supporting cells to make more of the protein. Before and after this step, they tested the mice’s hearing using an approach called auditory brainstem response or ABR – the same test used on humans.

The result: the mice with extra NT3 regained their hearing over a period of two weeks, and were able to hear much better than mice without the extra NT3 production. The scientists also did the same with another nerve cell growth factor, or neurotrophin, called BDNF, but did not see the same effect on hearing.

Next steps

Now that NT3’s role in making and maintaining ribbon synapses has become clear, Corfas says the next challenge is to study it in human ears, and to look for drugs that can work like NT3 does. Corfas has some drug candidates in mind, and hopes to partner with industry to look for others.

Boosting NT3 production through gene therapy in humans could also be an option, he says, but a drug-based approach would be simpler and could be administered as long as it takes to restore hearing.

Corfas notes that the mice in the study were not completely deafened, so it’s not yet known if boosting NT3 activity could restore hearing that has been entirely lost. He also notes that the research may have implications for other diseases in which nerve cell connections are lost – called neurodegenerative diseases. “This brings supporting cells into the spotlight, and starts to show how much they contribute to plasticity, development and maintenance of neural connections,” he says.

In addition to Corfas, Wan and Gómez-Casati, who now works in Argentina, the research was performed by Angelica R. Gigliello, and M. Charles Liberman, Ph.D. director of the Eaton-Peabody Laboratories of the Massachusetts Eye and Ear Infirmary. The research was supported by the National Institute on Deafness and Other Communication Disorders (DC004820, DC005209) and by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD18655), both part of the National Institutes of Health, and by the Hearing Health Foundation.

Reference: eLife, http://elifesciences.org/content/early/2014/10/17/eLife.03564

Oct 24, 2014

The technology for creating new tissues from stem cells has taken a giant leap forward. Two tablespoons of blood are all that is needed to grow a brand new blood vessel in just seven days. This is shown in a new study from Sahlgrenska Acadedmy and Sahlgrenska University Hospital published in EBioMedicine.

Just three years ago, a patient at Sahlgrenska University Hospital received a blood vessel transplant grown from her own stem cells.

Suchitra Sumitran-Holgersson, Professor of Transplantation Biology at Sahlgrenska Academy, and Michael Olausson, Surgeon/Medical Director of the Transplant Center and Professor at Sahlgrenska Academy, came up with the idea, planned and carried out the procedure.

Missing a vein

Professors Sumitran-Holgersson and Olausson have published a new study in EBioMedicine based on two other transplants that were performed in 2012 at Sahlgrenska University Hospital. The patients, two young children, had the same condition as in the first case – they were missing the vein that goes from the gastrointestinal tract to the liver.

“Once again we used the stem cells of the patients to grow a new blood vessel that would permit the two organs to collaborate properly,” Professor Olausson says.

Stroke of genius

This time, however, Professor Sumitran-Holgersson, found a way to extract stem cells that did not necessitate taking them from the bone marrow.

“Drilling in the bone marrow is very painful,” she says. “It occurred to me that there must be a way to obtain the cells from the blood instead.”

The fact that the patients were so young fueled her passion to look for a new approach. The method involved taking 25 milliliter (approximately 2 tablespoons) of blood, the minimum quantity needed to obtain enough stem cells.

Blood willingly cooperates

Professor Sumitran-Holgersson’s idea turned out to surpass her wildest expectations – the extraction procedure worked perfectly the very first time.

“Not only that, but the blood itself accelerated growth of the new vein,” Professor Sumitran-Holgersson says. “The entire process took only a week, as opposed to a month in the first case. The blood contains substances that naturally promote growth.”

More groups of patients can benefit

Professors Olausson and Sumitran-Holgersson have treated three patients so far. Two of the three patients are still doing well and have veins that are functioning as they should. In the third case the child is under medical surveillance and the outcome is more uncertain.

They researchers have now reached the point that they can avoid taking painful blood marrow samples and complete the entire process in the matter of a week.

“We believe that this technological progress can lead to dissemination of the method for the benefit of additional groups of patients, such as those with varicose veins or myocardial infarction, who need new blood vessels,” Professor Holgersson says. “Our dream is to be able to grow complete organs as a way of overcoming the current shortage from donors.”

The article In vivo application of tissue-engineered veins using autologous peripheral whole blood: A proof of concept study was published in EBioMedicine on 22 October.

Link to article

Link to journal: http://www.ebiomedicine.com/

Contact:
Suchitra Sumitran-Holgersson, PhD, Professor of Transplantation Biology, Sahlgrenska Academy, Laboratory for Transplantation and Regenerative Medicine, University of Gothenburg, and Department of Surgery, Sahlgrenska University Hospital, phone +46 31-343 0021,
cell +46 72-749 08 08, suchitra.holgersson@gu.se

Michael Olausson, Surgeon/Medical Director, Transplant Center, Sahlgrenska University Hospital, Professor, Sahlgrenska Academy, University of Gothenburg, phone 46 31-342 70 25,
cell 46 70-543 43 60, michael.olausson@surgery.gu.se

Images above: Robert Emilsson

BY: Krister Svahn
+4631786 38 69

Released: 7-Nov-2014 1:00 PM EST
Source Newsroom: Rockefeller University

Newswise — In a significant technical advance, a team of neuroscientists at The Rockefeller University has devised a fast, inexpensive imaging method for probing the molecular intricacies of large biological samples in three dimensions, an achievement that could have far reaching implications in a wide array of basic biological investigations.

The new method, called iDISCO, optimizes techniques for deep tissue immunolabeling and combines them with recent technological innovations in tissue clearing and light sheet microscopy to achieve unprecedented deep labeling and imaging of molecular structures in the brain, the kidney, and other organs and tissues in experimental settings. A detailed report on iDISCO is in the November 6 issue of the journal Cell.

“What we did was optimize many different parameters of several existing techniques to create this powerful new labeling method,” says Marc Tessier-Lavigne, Rockefeller president, Carson Family Professor, head of the Laboratory of Brain Development and Repair, and senior author on the new study. “These optimization efforts paid off in spades, dramatically extending our ability to visualize molecular structures deep in intact complex tissues like the brain. Although developed in our laboratory to help us pursue neuroscience questions, we believe this new method will provide biological researchers in many disciplines with an important new tool for advancing their work.”

When executed as part of a coordinated protocol, the advance creates a powerful method for imaging molecular structures deep in tissues that is much simpler and more rapid than previous approaches used by biological researchers.

“For us as neuroscientists, one of the big applications of this new imaging method has been the ability to visualize axonal pathways in the developing and adult brain,” says Nicolas Renier, a postdoctoral associate in the Tessier-Lavigne laboratory and co-first author on the study. “We were surprised at just how well we were able to image these detailed structures in the context of the whole brain.”

“The fact that we can now visualize neural circuit formation in larger embryos allows us to study the developing nervous system when it is more well formed,” says Zhuhao Wu, also a postdoctoral associate in the Tessier-Lavigne laboratory and co-first author on the study. “This opens entire new avenues to our research.”

The team focused on three techniques, as they sought to maximize the effectiveness of deep tissue immunolabeling and combine it with innovations by other scientists in tissue clearing and light sheet microscopy.

Clearing is a chemical treatment process that renders biological samples transparent, enabling light to reach deep inside them, and it is in this area particularly that new frontiers have been established in a number of laboratories recently, making it possible to peer deeper into samples than ever before.

Immunolabeling is a longstanding laboratory technique for tagging molecules of interest in biological samples with selected antibodies, but usually in relatively small or thin samples. In this area, the Rockefeller group demonstrated their ability to introduce 28 widely used research antibodies much more deeply and more rapidly into a diversity of samples than had previously been possible. Immunolabeling differs from another commonly used technique in which fluorescent reporter molecules are introduced into a tissue transgenically. Immunolabeling instead attaches tags to endogenous molecules in a study sample.

“Being able to visualize molecules introduced transgenically by the experimenters is a very valuable tool,” says Tessier-Lavigne. “What drove us, however, was a desire to complement that tool with a greater ability to look at endogenous molecular processes. In this area, our new labeling method breaks a barrier we weren’t necessarily expecting to be able to break when we set out on this project.”

Light sheet microscopy has the power to scan whole organs or large tissues in relatively short amounts of time at very high resolutions. The data set captured is so detailed that it can be translated either into images offering a comprehensive view of the entire sample or a closer look at smaller structures within the sample.

Renier and Wu are co-lead authors on the Cell study, and Tessier-Lavigne is senior author. Their coauthors are: David I. Simon, Jing Yang and Pablo Ariel, also in the Tessier-Lavigne laboratory.

A new technique can light up cancer cells so doctors can see them sooner.

Originally published: Nov 11 2014 – 3:00pm
By: Marsha Lewis, Contributing Producer
(Inside Science TV) – Cancer is the second leading cause of death in the United States killing 585,000 Americans every year.

Susann Brady-Kalnay, a cancer biologist at Case Western Reserve University in Cleveland, Ohio wants to change that.

“What happens now with cancer patients is they come in when they have symptoms, and it’s almost too late. The cancer has spread all over their body and caused symptoms,” said Brady-Kalnay.

She wants to catch cancer before symptoms start, with a method called molecular imaging.

“Molecular imaging means that you’re making an imaging agent that actually detects a specific event,” said Brady-Kalnay.

Her team has developed an imaging agent that can be injected into the body. It travels through the bloodstream, finds tumor cells and lights them up so they are visible on MRI scans.

“These probes light up those tumors in two minutes. They travel anywhere in the body,” Brady-Kalnay explained.

In an animation, green areas showed the main tumor mass and the color yellow showed tumor cells that had spread. While performing a procedure, surgeons would not normally be able to see the yellow cancer cells and where they have spread.

“Now, we’re able to allow the surgeon to actually see where the tumor cells are while they’re cutting,” Brady-Kalnay said.

In lab models, the agents detected more than 99% of brain tumor cells. The technique could be used to detect melanoma, prostate and lung cancers sooner.

“That’s our goal here to screen people, to do lifelong screening to find things early because we’ve always known that the best way to treat cancer is to find it early and take it out,” said Brady-Kalnay.

She hopes to begin human clinical trials with her agent in the next two to three years. She says the real key to coming up with this discovery was having a team of chemists, physicists, biomedical engineers and cancer biologists all work collaboratively.

---

Marsha Lewis is a freelance producer based in California. She has won 11 National Telly Awards and nine Regional Emmy Awards for her work in local and national syndicated news.

Will ultrasound-on-a-chip make medical imaging so cheap that anyone can do it?

By Antonio Regalado on November 3, 2014

A scanner the size of an iPhone that you could hold up to a person’s chest and see a vivid, moving, 3-D image of what’s inside is being developed by entrepreneur Jonathan Rothberg.

Rothberg says he has raised $100 million to create a medical imaging device that’s nearly “as cheap as a stethoscope” and will “make doctors 100 times as effective.” The technology, which according to patent documents relies on a new kind of ultrasound chip, could eventually lead to new ways to destroy cancer cells with heat, or deliver information to brain cells.

Rothberg has a knack for marrying semiconductor technology to problems in biology. He started and sold two DNA-sequencing companies, 454 and Ion Torrent Systems (see “The $2 Million Genome” and “A Semiconductor DNA Sequencer”), for more than $500 million. The profits have allowed Rothberg, who showed up for an interview wearing worn chinos and a tattered sailor’s belt, to ply the ocean on a 130-foot yacht named Gene Machine and to indulge high-concept hobbies like sequencing the DNA of mathematical geniuses.

The imaging system is being developed by Butterfly Network, a three-year old company that is the furthest advanced of several ventures that Rothberg says will be coming out of 4Combinator, an incubator he has created to start and finance companies that combine medical sensors with a branch of artificial-intelligence science called deep learning.

Rothberg won’t say exactly how Butterfly’s device will work, or what it will look like. “The details will come out when we are on stage selling it. That’s in the next 18 months,” he says. But Rothberg guarantees it will be small, cost a few hundred dollars, connect to a phone, and be able to do things like diagnose breast cancer or visualize a fetus.

Butterfly’s patent applications describe its aim as building compact, versatile new ultrasound scanners that can create 3-D images in real time. Hold it up to a person’s chest, and you would look through “what appears to be a window” into the body, according to the documents.

With the $100 million supplied by Rothberg and investors, which include Stanford University and Germany’s Aeris Capital, Butterfly appears to be placing the largest bet yet by any company on an emerging technology in which ultrasound emitters are etched directly onto a semiconductor wafer, alongside circuits and processors. The devices are known as “capacitive micro-machined ultrasound transducers,” or CMUTs.

Most ultrasound machines use small piezoelectric crystals or ceramics to generate and recieve sound waves. But these have to be carefully wired together, then attached via cables to a separate box to process the signals. Anyone who can integrate ultrasound elements directly onto a computer chip could manufacture them cheaply in large batches, and more easily create the type of arrays needed to produce 3-D images.

“The vision for this product has been around for many years. It remains to be seen whether someone can make it into a market-validated reality.”

Ultrasound is used more often by doctors than any other type of imaging test, including to view a baby during pregnancy, to find tumors in soft tissues like the liver, and more recently to treat prostate cancer by heating up cells with sound waves.

The idea for micromachined ultrasound chips dates to 1994, when Butrus Khuri-Yakub, a Stanford professor who advises Rothberg’s company, built the first one. But none have been a commercial success, despite a decade of interest by companies including General Electric and Philips. This is because they haven’t functioned reliably and have proved difficult to manufacture.

“The vision for this product has been around for many years. It remains to be seen whether someone can make it into a market-validated reality,” says Richard Przybyla, head of circuit design at Chirp Microsystems, a startup in Berkeley, California, that’s developing ultrasound systems that let computers recognize human gestures. “Perhaps what was needed all along is a large investment and a dedicated team.”

Rothberg says he got interested in ultrasound technology because his oldest daughter, now a college student, has tuberous sclerosis. It is a disease that causes seizures and dangerous cysts to grow in the kidneys. In 2011 he underwrote an effort in Cincinnati to test whether high-intensity ultrasound pulses could destroy the kidney tumors by heating them.

What he saw led Rothberg to conclude there was room for improvement. The setup—an MRI machine to see the tumors, and an ultrasound probe to heat them—cost millions of dollars, but wasn’t particularly fast, more like a “laser printer that takes eight days to print and looks like my kids drew it in crayon,” he says. “I set out to make a super-low-cost version of this $6 million machine, to make it 1,000 times cheaper, 1,000 times faster, and a hundred times more precise.”

Rothberg claims there’s a “secret sauce” to Butterfly’s technology, but he won’t reveal it. But it may have as much to do with clever device and circuit design as overcoming the physical limits and manufacturing problems that CMUT technology has faced so far.

One reason to think so is that the company’s cofounder, Nevada Sánchez, previously helped cosmologists design a much cheaper radio telescope with a signal-processing trick called a butterfly network, also the origin of the startup’s name. Also working with the company is Greg Charvat, who joined it from MIT’s Lincoln Laboratory, where he developed radar that can see human bodies even through thick stone walls (see “Seeing like Superman”).

During a visit to 4Combinator’s headquarters, which sits inside a marina in Guilford, Connecticut, Charvat and Sanchez showed off a picture of a penny so detailed you could read the letters and numbers on it. They’d taken the image this spring using a prototype chip. “The ultrasound [industry] is basically back in the 1970s. GE and Siemens are building on old concepts,” says Charvat. With chip manufacturing and a few new ideas from radar, he says, “we can image faster, with a wider field of view, and go from millimeter to micrometer resolution.”

Ultrasound works by shooting out sound and then capturing the echo. It can also create beams of focused energy—and chip-based devices could eventually lead to new systems for killing tumor cells. Small devices might also be used as a way to feed information to the brain (it was recently discovered that that neurons can be activated with ultrasonic waves).

“I think it will become better than a human in saying ‘Does this kid have Down syndrome, or a cleft lip?’ And when people are pressed for time it will be superhuman.”

Rothberg says his first goal will be to market an imaging system cheap enough to be used even in the poorest corners of the world. He says the system will depend heavily on software, including techniques developed by artificial intelligence researchers, to comb through banks of images and extract key features that will automate diagnoses.

“We want it to work like ‘panorama’ on an iPhone,” he says, referring to a smartphone function that steers a picture taker to pan across a vista and automatically assembles a composite image. But in addition to recognizing objects—body parts in the case of a fetal exam—and helping the user locate them, Rothberg says the system would also reach preliminary diagnostic conclusions based on pattern-finding software.

“When I have thousands of these images, I think it will become better than a human in saying ‘Does this kid have Down syndrome, or a cleft lip?’ And when people are pressed for time it will be superhuman,” says Rothberg. “I will make a technician able to do this work.”

Rothberg says his incubator has started three other companies in addition to Butterfly, and he’s given each of them between $5 million and $20 million in seed capital. They include a biotechnology firm, Lam Therapeutics, working on treatments connected to tuberous sclerosis; Hyperfine Research, a startup in stealth mode that hasn’t said what type of technology it is developing; and another company that’s unnamed.