Researchers have found that amoebas called Dictyostelium discoideum and cancer cells are able to navigate mazes within 30-120 minutes using self-generated chemotaxis (Tweedy et al., 2020) which allows them to procure information regarding their surroundings. D. discoideum are cellular slime molds and are regularly studied as an example of cell communication (2020a). These cells also use diffusion of attractants to get through mazes as quickly as possible, straying from the correct path but eventually finding the glucose at the end of the maze. Chemotaxis refers to the directional migration of cells in response to these attraction gradients, moving towards higher concentrations in their environment (2020b). Examples of chemotaxis would be: bacteria finding food sources (glucose) like the amoebas in this paper or swimming away from poisons, sperm swimming towards an egg during fertilization, and finally leukocytes seeking out an injury or infection.

     In this experiment the researchers used an acidic solution called adenosine monophosphate (Specktor, 2020) in which the cells utilized to navigate such maps as Henry VIII’s Hedge Maze and trident mazes, moving from areas of a lower to higher concentration gradient of attraction. This chemoattractant is called cyclic adenosine monophosphate (cAMP for short) and is important in many biological processes: intracellular signal transduction, regulating ion channels, and activating protein kinases (2020c). When D. discoideum are in stress they secrete cAMP, and all amoebas move using filose pseudopods to higher concentration of cAMP that are produced by amoebas. This results in the amoebas like finding glucose at the end of the maze.

     Cells using self-generated gradients migrate aptly. They can make unerring choices concerning navigating through pathways they have not come across. In this way these amoebas can find their way through easy and hard mazes no matter the distance. Models are the best way to test how cellular organisms detect attractant gradients (Tweedy et al., 2020). Dictyostelium discoideum find chemotaxis easier in self-generated gradients rather than imposed or passive gradients. They reach the end of the pathway in less time in both simulation and experiment.

     What is also a key factor in self-generated gradients is diffusion. In shorter models of mazes cells will not enter shorter branches as they clear the attractant from that section by diffusion. This is relative to the length of the branch. If the length of the branch is less than 250 micrometers, then less than 10% of cells will explore it. If the length of the branch is longer than 600 micrometers, then 40% of cells will explore it. In simples mazes where there were two exits with a large attract reservoir, cells split equally when encountering the divide to the attractants. Either way a majority of the amoebas found the large attractant reservoir using chemotaxis toward cAMP without directional cues, but when encountering longer branches cells chose correctly 50% of the time (Specktor, 2020). Even D. discoideum cells steered themselves through long mazes with self-generated chemotaxis (2020b). They knew where the dead ends were and avoided them and were faster than cells using imposed gradients. This was true for both computer simulation and experiment. They were not convolving their own cAMP signals as the scientists created a cAMP mutant in a wild-type parent, which is healthier and migrates better (Tweedy et al., 2020) and still the amoebas found the attractant gradient at the end of the maze with notable accuracy.

     There were three mazes that were made to test the accuracy and duration of the amoebas navigation. Researchers found that length and complexity of pathways and branches was detrimental to the amount of time the cells took to find the attractant, thus completing the maze. What the researchers dubbed simple mazes had three short dead ends half the length of their live channels. What researchers dubbed long mazes had live channels and dead channels that were the same length, with the sole difference between the two being the connection to the large attractant reservoir at the end of the maze. What was found in simple mazes was that the short dead ends were mostly avoided by both amoebas tested, D. discoideum and pancreatic cancer cells, which was predicted in simulations and was true in experiments. When in came to the longer mazes the cells made more incorrect decisions, going down branches that were dead ends. This changed as the cells reached the end of the maze. Their decisions became increasingly accurate showing that a similarity in behavior between these different cell types indicates that these are general features of the chemotactic response to self-generated gradients (Tweedy et al., 2020).

     It is apparent that dead ends that are shorter in length as well as live ends that are shorter in length provide cells with more accurate decisions although in T-shaped mazes cells more often than not made the wrong decision. Decisions were more accurate when cells took more time to make them. The cancer cells took much more time (two days) to solve the maze compared to D. discoideum at two hours, which was consistent throughout the research. They both performed identically showing that as many cells that went the wrong way, near the same amount went the right way towards the attractant. Simulations predicted Dictyostelium correctly, while in experiments cancer cells did better. If high diffusivity was present, excellent decisions were made. Cells in the short maze did worse than a 50:50 split in slow-diffusing attractants. Diffusivity controls decisions cells make.

     Cells followed mirages which is a large reservoir with no attractant in it. Most cells went into the dead-end reservoir than to the attraction gradient, as short dead ends are more attractive if they are wider in length. Longer dead ends with a mirage were less likely to have cells go down their branch as the attractant created a stronger signal for the cell. Simulations predicted more cells would go into the reservoir but in experiments not as many did. This shows that attractant flux is more important than attractant quantity. The final mazes used in the study were designed with many dead ends in pairs of three called tridents. Simulations matched experiments. Easy mazes had short branches while longer mazes had long branches. The hard maze was vastly explored while the easy maze had branches that no cells visited, showing that the longer the branches, the harder the maze for chemotactic(2020b) cells.

Sources:

Tweedy, L., Thomason, P.A., Paschke, P.I., Martin, K., Machesky, L.M., Zagnoni, M., and Insall, R.H. (2020). Seeing around corners: Cells solve mazes and respond at a distance using attractant breakdown. Science 369, eaay9792.

Wikipedia contributors. (2020a). Dictyostelid, https://en.wikipedia.org/w/index.php?title=Dictyostelid&oldid=986442674

Wikipedia contributors. (2020b). Chemotaxis, https://en.wikipedia.org/w/index.php?title=Chemotaxis&oldid=984346046

Specktor, B. (2020). Cells solved Henry VIII’s infamous hedge maze by ‘seeing around corners,’ video shows. LiveScience, https://www.livescience.com/cells-solve-mazes-chemotaxis.html

Wikipedia contributors. (2020c). Cyclic Adenosine Monophosphate, https://en.wikipedia.org/w/index.php?title=Cyclic_adenosine_monophosphate&oldid=958699223