Synthetic mammalian signaling circuits for robust cell population control

Yitong Ma, Mark W. Budde, Michaëlle N. Mayalu, Junqin Zhu, Richard M. Murray, Michael B. Elowitz

Preprint posted on September 03, 2020

Cells learn a new language! Auxin confers new synthetic communication-channel for cells to communicate with.

Selected by Pavithran Ravindran

Categories: synthetic biology


One of synthetic biology’s greatest challenges is to engineer cells that can learn from both their environment as well as their neighboring cells to make decisions about what program to execute1. Previous studies have been extremely successful in developing methods to engineer cells to integrate endogenous signaling ligands, such as Sonic Hedgehog2. The development of synthetic Notch receptors that provide completely orthogonal/synthetic communication channels made great progress to the eventual goal, but is limited to cell-to-cell contact-mediated signaling3,4. Very recently, synthetic Notch receptors were used with synthetic ligands (GFP) to program cell decisions, but still required two cells with two separate receptors such that the GFP becomes anchored and displayed on one cell and then activates the adjoining receiver cell5. Therefore, there is still work to be done to develop ligand-based orthogonal communication channels that synthetic biologists can use to engineer cells.

The authors of this preprint developed a method that repurposes the plant hormone auxin to control the expression of user-defined genetic programs. They then utilize this system to perform synthetic quorum sensing, a signaling phenomenon present in bacteria that allows for bacterial cells to learn about local cell density and aids in their decision to divide or not. The authors find that a simple circuit from auxin ligand leading to cell death only transiently limits the size of the population; they instead need a paradoxical circuit where the auxin ligand allows for both cell growth and cell death to ensure escape mutations do not occur in their circuit.

Key Findings

In this preprint, the authors set out to develop a method to use the plant hormone auxin to control the expression of genes in an engineered cell. To do so, they found that they only needed two components: (1) ectopic expression of F-box transport inhibitor response 1 (TIR1) from rice (osTIR1) which recruits E2 ubiquitin ligase for targeted degradation and (2) the auxin inducible degron (AID) attached to genes that the user wants regulated. As a first test, they expressed osTIR1 in mammalian CHO-K1 cells as well as mCherry fused to AID and Blasticidin resistance. With this setup, addition of auxin to the media decreases the expression of mCherry and Basticidin resistance (Figure 1). The latter results in the susceptibility of cells to blasticidin at higher doses of auxin hormone. With this, the authors had set up their receiver cell.

Figure 1. Schematic of initial auxin induced degradation system. Auxin enters the cells, binds to osTIR1 that initiates ubiquitylation of proteins with auxin-inducible degron (AID). Image taken from Figure 1.


The next step was to develop a way for cells to produce auxin hormone. To do so, they propose that the amino acid tryptophan can be oxidized to a precursor of auxin and that can be hydrolyzed to the final auxin hormone with the help of a hydrolase. After performing a small screen to identify hydrolases that work in mammalian cells, the authors find that by expressing this enzyme, known as iaaH, in their receiver cells, they can simply add precursor to the media and still get degradation of mCherry and Blasticidin (Figure 2). Furthermore, by plating cells containing iaaH (‘sender cells’) on one end of a 60mm dish and filling the rest with receiver cells, the authors find that they get a gradient of mCherry expression in receiver cells where cells near the senders have little to no mCherry fluorescence, whereas those further away have close to unperturbed levels of fluorescence. The authors thus have developed cells that could now send auxin hormone.

Figure 2. Schematic of auxin induced degradation system with auxin production mechanism. Cells express osTIR1 and iaaH, the latter of which takes precursor molecules and creates auxin that can bind to osTIR1 (shown in fig. 1). Image taken from figure 2.


As a proof-of-principle to show the utility of such a system, the authors wanted to allow the cells to sense their own density and regulate their growth based on it. The idea was that by plating sender-receiver cells in the presence of the precursor molecule and Blasticidin, once cells achieved a high enough density, the cells would increase the concentration of auxin in the media resulting in Blast resistance degradation and thus cell death due to susceptibility to Blasticidin. After finding that they needed to express another transport channel (PIN2) to get efficient auxin transport to the media, the authors saw that when they plated the cells at higher density, mCherry fluorescence would decrease. However, when they tried a long time-course of this experiment by plating cells at low density and imaging over time, they found that cells would start initially regulating their confluency in the well, but then start to uncontrollably grow. Upon analysis of the resulting cells that had grown out, the authors found mutations that resulted in auxin-insensitivity and thus constitutive expression of Blasticidin resistance.

Cheater mutations that result in the destruction of genetic circuits are a major problem in synthetic biology. Taking inspiration from previous literature on theoretical circuits that could combat this6, the authors hypothesized that placing an AID on an inducible caspase (iCasp9) and allowing the cells to grow in the small molecule that activates the apoptosis-inducing enzyme will get them around this. With this, the authors have created a paradoxical circuit. At high cell densities with high auxin (a) if the cells have a functioning circuit then Blasticidin resistance is downregulated and cell growth is halted but (b) if cells have broken the circuit (cheater mutation) then iCasp9 is not downregulated leading to the apoptosis of the cheater cell (Figure 3). With this circuit in place the authors show that indeed fewer cheater mutations occur and the cells are better able to control their population size.

Figure 3. Schematic of synthetic instantiation of a paradoxical circuit for population control. Auxin causes both the degradation of iCasp9, which prevents cell death upon binding to AP1903, and blasticidin, which prevents the ability to grow in blasticidin. Thus, auxin is both activating cell death as well as cell growth, a paradoxical circuit. Image taken from figure 5.


Why I chose this preprint

The development of synthetic channels by which cells can communicate with one another to perform user-defined decisions is a major goal in biology. The use of synNotch juxtacrine signaling has already found many uses in synthetic immunology, but is still limited to cell-to-cell contacts4. Here, the authors of the preprint have found a way to use a small molecule plant hormone (auxin) to drive their genetic circuits. This could open up the possibility for more flexible cellular engineering. Finally, the preprint itself is extremely straightforward to read with each figure containing schematics that clearly delineate the purpose and setup of each experiment.

Questions for the authors

  1. In most of the current setup, you need to add precursor molecule to drive production of the auxin and turn on the circuit, although you do show in supplementary figure 1 that one can express iaaM to produce the precursor molecule in the sender-receiver cells. Can you comment on why you decided to keep using precursor molecule instead of this setup?
  2. Is auxin safe in humans? If not, what ideas do you have to port this system for engineered cells for therapeutic cases?


  1. Bashor, C. J. & Collins, J. J. Understanding Biological Regulation Through Synthetic Biology. Annu. Rev. Biophys. 47, (2018).
  2. Li, P. et al. Morphogen gradient reconstitution reveals Hedgehog pathway design principles. 0645, 1–10 (2018).
  3. Morsut, L. et al. Engineering Customized Cell Sensing and Response Behaviors Using Synthetic Notch Receptors. Cell 164, 780–791 (2016).
  4. Roybal, K. T. et al. Engineering T Cells with Customized Therapeutic Response Programs Using Synthetic Notch Receptors. Cell 167, 1–14 (2016).
  5. Toda, S., Mckeithan, W. L., Hakkinen, T. J. & Lopez, P. Engineering synthetic morphogen systems that can program multicellular patterning. Science (80-. ). 331, 327–331 (2020).
  6. Hart, Y. et al. Paradoxical signaling by a secreted molecule leads to homeostasis of cell levels. Cell 158, 1022–1032 (2014).

Tags: communication channel, synthetic biology

Posted on: 3rd November 2020


Read preprint (No Ratings Yet)

Author's response

Author team shared

  1. This is a great question. In fact, for future in vivo applications, we do anticipate using iaaM for complete auxin biosynthesis, thus freeing the system from the need for added substrates. In this paper, however, omitting iaaM from the circuit was useful. It allowed us to control auxin production rate by varying the concentration of the added auxin precursor. This enabled us to operate different circuits within the same cell line, by including different combinations of auxin precursor, AP1903 and blasticidin, making for better controlled comparisons. For instance, the analysis in Figure 6C and 6E would be confounded by iaaM-dependent auxin production.
  2. Although a safety study is still needed, based on existing evidence we expect the auxin levels used in our experiments to be safe in humans. Auxin is present in the plants that we eat. In the 1950s, people were fed about 100x the concentration (based on body weight) of auxin that we used without obvious problems (Mirsky & Diengott, 1956). Additionally, auxin (IAA more specifically) is produced at a low level by human cells as a byproduct of tryptophan metabolism. While usually present at low levels, patients with kidney disease can accumulate auxin in the serum at similar levels as we used in our experiment, indicating that these levels are tolerable.. In addition , we don’t expect that auxin will accumulate throughout the entire body in a therapeutic context, but would instead act semi-locally, further improving its safety. A related issue is that the current circuit uses blasticidin, which is toxic to mammalian cells. To implement a similar circuit in a cell therapy context, we anticipate replacing the role of blasticidin resistance with coupling of auxin to a cell-cycle regulator.

Have your say

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here