Notes on Engineering Health, May 2021: The Ends, Not the Means

In an unexpected outcome, scientists were able to train rats to create and maintain a large difference in temperature between their two ears. The training mechanism was fairly simple — the larger the temperature delta, the larger the reward in the form of electrical stimulation. After a few weeks of training, the rodents were able to induce and maintain a temperature difference of 5°C between their left and right ears.

Identifying the specific pathway in a rat to increase one ear’s temperature rather than the other, and then to act on this pathway precisely enough to achieve the intended output seems like an unreasonably difficult task. The somewhat surprising outcome of the experiment described above points toward the opportunity to guide biological systems at a high-level toward a desired output, rather than attempting  to control each step of a complex process at a fine-grain.

The idea that organisms, organs, tissues, and even cells can generate a desired outcome on their own if you create the proper incentives is a powerful one. It has led to important basic discoveries and could possibly be used in the near future to solve complex problems. Three variations of this idea serve as landmarks toward engineering biology in news ways:

1. Directed evolution is a method used in protein engineering that mimics the process of natural selection to steer proteins or nucleic acids toward a defined goal. This method appeared in the late 1960s and started with RNA. It quickly was used to evolve various proteins, and in particular enzymes able to catalyze challenging reactions. In 2018, Frances Arnold was awarded the Nobel Prize in Chemistry for her work on protein optimization via directed evolution. Directed evolution works on the principle that, when it comes to biological discovery, the path to optimization does not matter and should be left to the organism to work out. All that matters, and what engineers need to be focused on, is how best to set up biological systems to get them to find the preferred outcome.

2. Slime mold is quite a remarkable unicellular organism. A seminal paper from 2010 published in Science showed that the slime mold Physarum polycephalum is able to recreate complex networks with comparable efficiency, fault tolerance, and cost to those of real-world infrastructure systems such as the Tokyo rail system. The beauty of this work is that the mechanisms by which the unicellular organism creates the network can be captured in a biologically-inspired mathematical model—we have written about natural computing before and this is a prime example of it. Opening the black box on biological processes at the level of solutions produced (in this case adaptive network formation) should allow scientists and engineers to adapt these approaches to other domains.

3. Michael Levin’s work (featured in a recent issue of the The New Yorker if you want an expansive portrait of the man and his research) focuses on non-neurologic bioelectric signaling and identifies it as a key organizing principle in complex biological processes such as morphogenesis. One of the most striking examples showcasing the importance of bioelectric signaling is the distinction between growing a head or a tail in the planarian model organism. Solely by manipulating bioelectric circuits, Levin can control the regrowth of severed body parts while guiding the organism’s overall fate to outputs such as creating a two-headed animal. His novel strategy is to control morphogenesis not at the individual cellular level, but rather to trigger a master mechanism that governs the plan for the organism to make a head. Having done this, the organism retains the function of creating two-headed offspring, despite the fact that no changes have been made to its genome. This very unexpected outcome points toward the opportunity to guide biology to desired outcomes such as limb regeneration at the level of tissues, rather than by re-wiring systems from the molecular level on up.

These examples inspire us to think broadly about how to better guide biologic systems to take advantage of their already existing capabilities in engineering therapeutic outcomes.

– Jonathan Friedlander, PhD & Geoffrey W. Smith



First Five
First Five is our list of essential media for the month which spans a range of content including scientific papers, books, podcasts, and videos. For our full list of interesting media in health, science, and technology, updated regularly, follow us on Twitter or Instagram.

1/ Brains
Peer-reviewed advances in brain-computer interfaces (as opposed to science by press release advances) are beginning to emerge. Researchers at Stanford University have published a paper in Nature describing the design of a novel brain-computer interface, based on the imagined act of handwriting, that has achieved an unmatched 90 characters per minute. This article in PhysicsWorld provides a good overview of their work. This piece in The New Yorker provides a provocative patient-centric view of how neural devices can impact individuals and their loved ones, and reminds us it is not just about getting the technology right when it comes to creating an effective long-term treatment.

2/ Skin (and Bones)
While the brain may be the most important organ in the body, skin is the largest making up approximately 16% of your overall body mass. Our history of traumatic injuries can often be traced across our skin by the scars these events have left behind. Another Stanford University lab reported in Science recently that they have identified the physical and chemical signals that cause a specific population of skin fibroblasts to produce scars. Moreover, they describe a method to reprogram these cells into another cell type capable of regenerating normal tissue in mice. Amazingly, the recovery was so complete that an image-classifying algorithm could not separate healed wounds from the animals' healthy un-traumatized skin.

When trauma affects both skin and bones, such as in injuries to the face and skull, the repair challenge is even greater. This article in Advanced Functional Materials reports on the use of bioprinting intra-operatively (!) to create different craniomaxillofacial tissues including bone, skin, and composite tissues.

3/ Cancer Nutrition
Scientific consensus has held for over a hundred years that cancer cells consume glucose at an inordinately high rate in order to support rapid growth. More recently, PET scans and other imaging modalities were developed to detect high glucose levels in the body, visually identifying where tumors were consuming the simple sugars. A recent paper in Nature shows that the consensus on tumors and glucose has been wrong. Rather than cancer cells consuming the most glucose, immune cells within the tumor and its microenvironment are preferentially taking up the glucose while the cancer cells showed the highest uptake of glutamine. Understanding nutrient metabolism more accurately could open up new therapeutic approaches.

A recent review article in Science takes on another area where scientific consensus is shifting under the continued work in modern cancer biology. The review “delineate[s] between causal and complicit roles of microbes in cancer and trace[s] common themes of their influence through the host's immune system, … defined as the immuno-oncology-microbiome axis. We further review evidence for intratumoral microbes and approaches that manipulate the host's gut or tumor microbiome while projecting the next phase of experimental discovery.”

4/ Nutrition Nutrition
Nutrition science also continues to elucidate more fine-grained understanding of the impact of diet on health across a number of interesting recent studies:
– Selenium supplementation protects against obesity and may extend lifespan. eLife >
– Cocoa flavanol supplementation improves cognition in subjects with poor diets. Nature Scientific Reports >
– Eating unprocessed red meat and poultry seems OK, but stay away from processed meats. American Journal of Clinical Nutrition >
– Consumption of processed foods that are frequently rich in fats and simple carbohydrates but lacking in fiber are shown to be a significant contributor to the increased incidence of chronic inflammatory diseases that have accompanied industrialization. PLOS Pathogens >
– Microbial-targeted food intervention shows promise in treating the 30 million children worldwide who have moderate acute malnutrition. NEJM >

5/ Artificial Intelligence
The 2021 Artificial Intelligence Index Report is out and well worth reading to get a baseline on the continued development and impact of AI technologies.



Digitalis Commons
Public-Interest Technologies for Better Health

Digitalis Commons is a non-profit that partners with groups and individuals striving to address complex health problems by building public-interest technology solutions that are frontier-advancing, open-access, and scalable.

Bruce Schneier maintains a valuable repository of public-interest technology resources.