Can moths teach drones to navigate?

Written by: Sayan Roychowdhury

Original Article: Zhu et al. PLOS Computational Biology 2020
The Gist of it:
Autonomous aerial vehicles, such as drones, are carefully programmed and trained to move based on their surroundings. However, these controls sometimes react poorly when introduced to new or unknown environments. So, how can safe and reliable flying strategies be developed for use in unfamiliar territory? To answer this question, a group of scientists from Boston University and the University of Washington turned to hawkmoths for inspiration. Previous research shows that hawkmoths navigate largely based on visual cues just as drones do, making the moths’ flight decision-making very applicable to drones. In this study, the researchers gathered physical flight data from hawkmoths through a virtual forest projected on a screen. The researchers then created a moth decision-making model based on this data; this model could simulate how a hawkmoth would react to seeing a certain visual scenario. The researchers tested this model on a wide variety of forest arrangements differing in size and density to examine how well a hawkmoth would perform when encountering new surroundings. They also compared the performance of the hawkmoth to an optimal flight pattern for each virtual forest.
The researchers found that for each case, the moth decision-making model performed very well, suggesting that the moths’ strategy is quite versatile, even for unfamiliar terrain. In particular, the moth model worked best in dense forest scenarios, possibly indicating that hawkmoths have developed flight strategies best suited to the types of forests they encounter in their natural habitat. These results are exciting because researchers can study biologically evolved flight patterns to help program autonomous vehicles to perform better in new and unfamiliar environments.

Flight data from hawkmoths navigating a virtual forest can be used to train flight decision-making for autonomous drones.

The Nitty Gritty:
To study navigational patterns, the researchers connected eight hawkmoths (Manduca sexta) to a metal rod and placed them in front of a large screen displaying a virtual forest moving at a constant speed of 2 m/s. Each moth’s motion in the horizontal direction was measured through a torque meter, capturing changes in direction based on the view. Data were collected for each moth at 5 varying levels of fog density to simulate varying visibility conditions. The virtual forest and the moth’s flight trajectory were discretized in time and space, and a Markov Decision Process was used to model the flight. Logistic regression was used to estimate the parameters of the moth control policy. This policy was further refined using a reward function based on distance to trees – minimizing distance to simulate hiding from predators in the foliage, while maintaining enough distance to avoid collision. The original moth policy and an optimized one were then compared on a new set of virtual forests with varying size and density, where the optimal policy was tuned on a per forest basis. The original moth control policy was very robust and was within 30% of the performance of the refined policy for all of the dense forests. Thus, this work shows that hawkmoths can serve as a reliable starting point for successful bio-inspired control and decision rules for autonomous aerial vehicles exposed to new terrains.
Original Research Article: Zhu, H., et al. “Learning from animals: How to navigate complex terrains.PLoS Comput Biol 16.1 (2020): e1007452.

Paraquat herbicide poisoning: Antidote on the horizon?

Written by: Sravya Kotaru

 Original article: Liu et al. BioMed Research International 2020
The Gist of It:
 Paraquat is a common herbicide used to kill weeds in croplands. It is poisonous if consumed, breathed, or touched. Severe poisoning can cause multiple organ failure including lungs, kidneys, and liver, and can be fatal. Even mild poisoning leads to long-term respiratory issues from lung damage and scarring. The usage of paraquat is highly debated; it is banned in Europe but is still widely used in the United States and other parts of the world. The people at most risk of exposure are those using the herbicide and those living in areas near where it is used, but it is also possible for paraquat to contaminate food and water sources and reach consumers. Once paraquat enters the human body, it produces harmful molecules called reactive oxygen species (ROS), which cause inflammation and tissue damage. Specifically, in the lungs, paraquat damages the tiny air sacs called alveoli, which are like small balloons that fill with air during each breath. In the alveoli, oxygen from the air diffuses into the bloodstream. When paraquat is breathed in and reaches these lung alveoli, it causes inflammation and the build-up of scar tissue around these air sacs. This scarring restricts the alveoli’s ability to expand with each breath, causing shortness of breath and lung disease. So far, there is no antidote or treatment for paraquat poisoning. Excitingly, scientists at North Sichuan Medical College in China recently discovered more details about how paraquat causes lung damage, allowing them to identify a potential target for treatment. They found that the ROS produced by paraquat lung poisoning in rats damaged mitochondria, the energy factories in alveolar air sacs. This damage turns on two proteins called PINK1 and Parkin that are responsible for removal of the damaged mitochondria and cause scar tissue to develop around damaged alveoli, thus causing lung injury. The researchers also found that deleting the protein PINK1 from lung cells grown in the laboratory prevented the production of fibronectin and collagen, two molecules that form scar tissue. These results are good news because they suggest that PINK1 could be targeted to develop a paraquat antidote and prevent lung damage. This possibility must be further explored.

Paraquat is a poisonous herbicide that causes long-term lung injury by damaging alveolar air sacs and scarring surrounding lung tissue via a molecular mechanism involving PINK1 and Parkin proteins

The Nitty Gritty: 
Paraquat, a common herbicide, is a known oxidant that generates ROS and causes mitochondrial damage in alveolar epithelial cells, resulting in alveolitis and pulmonary fibrosis. Independently, it is known that damaged mitochondria accumulate the kinase PINK1 on their outer membrane; PINK1 in turn recruits Parkin to initiate mitophagy. The authors investigated whether PINK1-Parkin signaling and mitophagy are involved in paraquat-mediated poisoning. They observed that in rat models, paraquat exposure resulted in loss of alveolar architecture and a build-up of fibrous extracellular matrix in lungs over time, indicative of fibrosing alveolitis. The poisoned rat lungs also showed a time-dependent increase in the expression of the extracellular matrix fibers fibronectin (FN) and collagen (COL-1), which contribute to the formation of fibrous scar tissue. In addition, there was a time-dependent increase in PINK1 and Parkin expression and loss of mitochondrial membrane potential, indicative of mitophagy. The researchers observed similar results in the cultured A549 lung cancer cell line, where paraquat induced a time- and dose-dependent loss in cell viability and increase in the expression of PINK1, Parkin, FN, and COL-1. Moreover, silencing PINK1 expression in A549 cells negated the effects of paraquat-mediated upregulation of Parkin, FN, and COL-1 expression. Thus, mitophagy and fibrosis due to paraquat poisoning appear to be PINK1-dependent, and PINK1 is a possible molecular target for the development of a paraquat-specific antidote.
Original Research Article: Liu, K., et al. “Paraquat exposure induces pulmonary cell mitophagy by enhancing the PINK1/Parkin signaling.” BioMed Res Int 6 (2020): 1–6. doi: 10.1155/2020/7103105.

A surprising discovery of live mitochondria in human blood

Written by: Abby Stahl

Original Article: Al Amir Dache et al. The FASEB Journal 2020
The Gist of It
You may have heard that mitochondria are the “powerhouse” of the cell, but what is DNA from these tiny organelles doing circulating in the bloodstream?  That was the question a team of scientists at the Institut Regional du Cancer de Montpellier in France asked in response to several studies showing that DNA from mitochondria could be found in circulation.
Mitochondria are important producers of cellular energy that evolved millions of years ago through a symbiotic relationship between primitive eukaryotic cells (our ancestors) and free-living prokaryotes (bacteria). Mitochondria have retained many of their bacterial features today, including having their own DNA. The fact that mitochondrial DNA is in the bloodstream has been known by doctors for years, and its presence can even be used to help diagnose diseases and conditions such as diabetes, cancer, and heart disease. Surprisingly, this team of researchers found that there are 50,000 times more copies of mitochondrial DNA in the bloodstream than nuclear DNA, or the genomic DNA at the “core” of the cell. However, the purpose of this mitochondrial cell-free DNA (McfDNA) was still unknown.
By collecting blood plasma from healthy volunteers, the scientists found that McfDNA was longer and more stable than nuclear cell-free DNA (NcfDNA), suggesting the McfDNA might be protected or encapsulated by some larger structure. The researchers went on to find that spinning the plasma at a high speed caused sedimentation of most of the McfDNA. Spinning at high speed can also cause intact mitochondria to be spun down. In fact, using a dye that labels live mitochondria, they found an abundance of the organelles within the centrifuged blood plasma. More powerful electron microscopy confirmed that these mitochondria existed without a cell around them, and metabolic studies showed that these mitochondria were functional at consuming oxygen, an important step in energy production.
These research studies are the first to report the presence of intact and functional mitochondria in blood from healthy individuals. The scientists estimate that there are between 200,000 and 3.7 million cell-free, intact mitochondria per milliliter of blood plasma and that these mitochondria might have crucial biological roles still to be discovered!

Blood serum collected from healthy patients was found to contain live mitochondria capable of converting oxygen (O2) to energy (ATP) and encapsulating stable, circulating, cell-free mitochondrial DNA.

The Nitty Gritty
In this study, the authors began by comparing the stability of DNA found in the blood of healthy individuals. They hypothesized that NcfDNA would be more stable than McfDNA due to the presence of supporting histone proteins that mitochondrial DNA lacks. To their surprise, the McfDNA had a greater DNA Integrity Index, meaning the McfDNA had a mean size greater than 300 base pairs when assessed from 13 blood samples. Sequencing of cell-free DNA from 80 additional individuals showed that McfDNA contained two populations of DNA – small fragments between 30–300 base pairs (approximately 20%) and large pieces greater than 1000 base pairs (approximately 80%); these populations could not be sequenced due to limits of sequencing technology.
To further determine the structure of this stable McfDNA, the researchers centrifuged blood samples at increasing speeds to sediment different components. Interestingly, the McfDNA sedimented at 16,000 × g, whereas NcfDNA was retained in the blood plasma, suggesting McfDNA was larger and denser. Importantly, the researchers also verified these findings in cells that had been cultured in media to confirm that the McfDNA was not being released by platelets during the experimental handling of blood samples, showing an abundance of McfDNA pelleted at 16,000 × g in cell cultures. In all cases, the McfDNA was too large to pass through a 0.22 mm filter, suggesting it might be retained within mitochondria organelles themselves, which are typically 0.4–1.5 mm in size and are known to sediment at 7,000–20,000 × g .
To confirm these suspicions, the scientists used MitoTracker Green dye to visualize live mitochondria and found them in serum samples and cell media supernatants. Western blot analysis also confirmed the presence of membrane transport proteins, TOM22 and TIM23, in these McfDNA fractions. Finally, the researchers used electron microscopy and identified structures with a double membrane and similar size to mitochondria in the samples, although they noted some morphological differences.
After determining that mitochondria were present and surrounding the McfDNA in blood and cell samples, the scientists asked whether these mitochondria were metabolically functional and able to produce energy, a process known as respiration. Using a Seahorse XF flux analyzer, the researchers found that the isolated mitochondria consumed oxygen and were sensitive to inhibition and activation by known regulators of respiration.
These studies show that human blood contains a previously unknown and significant population of mitochondria that can be useful for diagnostics; however, the physiological roles of these cell-free organelles (if any) remain a mystery for now.
Original Research Article: Al Amir Dache Z, A., et al. “Blood contains circulating cell‐free respiratory competent mitochondria.” FASEB J (2020). doi: 10.1096/fj.201901917RR. [Epub ahead of print].