Cell Maze

Have you ever been to a hedge maze and got lost after a couple turns? What if I told you that scientists have found cells that are able to solve mazes by themselves? A study done by Tweedy and his colleagues found that cells can solve mazes through self-generated chemotaxis. Chemotaxis is the migration of cells based on the gradient of the chemoattractant. The gradient can become steeper due to it only being made in the vicinity of the cells. Self-generated chemotaxis is also the migration of cells, but the cell generates its own path by breaking down chemoattractant in their environment. In the study, Tweedy found that two amoebas, Dictyostelium discoideum and mouse pancreatic cancer, would move from areas of lower concentration to areas of higher concentration via self-generated chemotaxis eventually leading to the exit. To do this, the amoebas break down molecules that are in front of them which causes the chemoattractant nearby to diffuse towards them. The cells will then move forward, behind the chemoattractant which are depleting as the cells go. The way the amoebas move through chemotaxis is by ameboid movement. Ameboid movement is a crawling like movement used by cells that have no features for mobility. Since the cells do not have mobility features, cytoplasmic flow pushes the cell forward while pseudopodia which are “fake feet” are formed allowing them to move. The chemoattractant used to attract the amoebas are called cyclic adenosine monophosphate or cAMP. The chemical cAMP sends waves of signals that attract the amoeba (Singer, 2019). The waves can be sent over large distances which allows the cells to cover as much as several centimeters (Wikipedia contributors, 2020). Normally, cAMP is a second messenger that transfers effects of hormones into cells that typically are not capable of passing through the plasma membrane.

When it comes to the speed of success solving the maze, the Dictyostelium discoideum was faster than the mouse pancreatic cancer. This D. discoideum can complete mazes under an hour while the pancreatic cancers are able to complete mazes in one to two days depending on the complexity of the maze. The complexity of the maze can cause issues for the cells. During the experiment, there were three types of mazes used: simple, intermediate, and complex. The number of cells that are lost is similar in both cells. In the simple maze, the cells lose a very small number of its population before it completes the maze. This is because the simple maze contains only three dead ends for the cells to run into. In the intermediate maze, the cells lose up to twenty percent of their cells with more dead ends added. In the complex maze, the cells can lose up to half of their cells due to the maze’s split being similar on both sides. To add more complexity in the mazes, the dead ends were varied in length. The longer the dead end, the harder it is for the cells to navigate themselves to the end. Labyrinth mazes were also used as well to see if the cells can solve them. There were two difficulties of labyrinth mazes, easy and hard. Easy mazes had shorter dead ends, while the hard mazes had longer dead ends. The results were that about 60% of the cells were able to complete the easy maze, while 40% were able to complete the hard maze. This solidified the prediction of the longer the dead end, the harder the maze is to solve. Lastly, trident mazes were used to give cells a much more elaborate maze. The reason why they made a maze this way is to give the cells a more human like pathway. This helped to determine how cells can reach their destination in a human like pathway with how much cells are left, as well as their speed and accuracy as a result.

Before the experiments were done, computer generated simulations of the possible outcome were done to predict how well the cells will do in mazes. Comparing the computer generated to the actual experiment, they were spot on every time on how long it would take to complete the maze comparing their results to the D. discoideum. Comparing on how the simulation predicted the number of cells that are lost, the simulation was off, gradually increasing as the maze got more complex. In the simple maze, the simulation predicted one cell to be lost, while thirty plus were to succeed, but the D. discoideum loses five plus cells and has thirty cells completing the maze. In the intermediate maze, the simulation starts to get the numbers off slightly. Dictyostelium discoideum in the intermediate maze loses eight cells and thirty cells complete the maze, while the simulation predicted that ten cells will be lost and twenty-five would succeed. With the complex maze prediction, the simulation was way off. The simulation predicted that twenty-five or more cells would be lost while thirty-five cells would succeed, while the D. discoideum really lost ten cells and twenty-five cells succeeded. When comparing the succeeded cells to the lost cells, the size of the margins on the graph is a big giveaway to determine if the cells went through a simple, intermediate, or complex maze. Pancreatic cancer cells did not match with the simulation predictions because they are slow moving cells compared to D. discoideum which are rapid moving cells, leading to jagged lines in the graph.

What impacts the cell’s ability to solve a maze is the rate of diffusion. Diffusion allows the cells to gain information of its surroundings. The rate of diffusion is determined by the concentration of the substance. For this experiment, three similar mazes were made, with only one difference being that there are two concentrations of chemoattractant in the maze. The result is that the cells prefer to go towards the area of the maze that has a higher concentration, leading to a faster solve time and possibly better accuracy depending on the difference in concentrations. An example of this would be that if two concentrations are the same, the speed of completion is what is only affected. If the two chemoattractant differ in concentration levels, the higher concentration would affect the speed and accuracy of the cells.