Funded projects

Magnetogenetic control of cell migration

2013 - D. Lisse, M. Dahan (UMR168)

Migration and invasion by cancer cells plays an essential role in the development of metastasis. It is thus important to study the mechanisms by which cancer cells move, individually or collectively, in different tissue environments. Investigations are usually based on the (passive) visualization of the cell movement, coupled to genetic or pharmacological perturbations of the molecular actors involved in cell migration. However, dissecting the subtle interplay between malignant and healthy cells requires versatile experimental techniques for controlling and manipulating migration in cell cultures and in tissue-like environments down to the level of individual cells. For this purpose, techniques for manipulating cellular signals with subcellular spatial resolution are required. While this has been demonstrated by optogenetic approaches, maintaining precisely localized optical signals within multiple cells over extended time periods is not practical.

In a recent work, we demonstrated an alternative approach, referred to as magnetogenetics, in which magnetic nanoparticles are used to control intracellular signaling activity with high spatial and temporal resolution. In our assay, a few magnetic nanoparticles (MNPs) with a diameter ~500 nm were injected into the cytoplasm of living cells. When functionalized with small GTPases of the Rho family or the corresponding guanine nucleotide exchange factor (GEF), they acted as signaling platforms that could be displaced by magnetic forces and enabled the manipulation of intracellular machineries at a subcellular scale (Figure 1 A and B).1 For instance, it was possible to bring the bio-functionalized MNPs to the plasma membrane to locally control the actin dynamics and the formation of protrusion. Importantly, magnetic gradients are much easier to maintain than optical gradients, and thus much more promising for long term experiments and the parallelized manipulation of multiple cells.

In this project, we aim to develop magnetogenetics towards precisely controlling the migration of cells in 2D and 3D in vitro assays. For this purpose, small MNPs for improved delivery will be engineered and adapted for the application in cancer cells and carcinoma-associated fibroblasts. Our goal is to apply these tools to the study of the migration of cancer cells, in relation to their cellular environment. The project, which is based on a new collaboration between the groups of Maxime Dahan (UMR168) and Danijela Vignevic (UMR144), is organized around three tasks of increasing difficulty.

Figure 1 Magnetic manipulation of endogenous protein activities. (A) Concept of magnetogenetics: Injection of functionalized MNPs (I) allows the self-assembly of an active signaling platform on the surface of MNPs by specific recruitment of effector protein (II). MNPs can be brought to different parts of the cell through the application of a magnetic field by means of a small magnetic-tip (III). In this approach, the signaling platform propagates a signal to cellular functions by locally activating a pool of endogenous proteins. (B) Activation of the small GTPase Rac1 at the plasma membrane using GEF-functionalized MNPs caused the formation of a protrusion (blue arrow) followed by the formation of an actin cloud (red arrow). The location of MNPs is indicated by the red dots. The scale bar corresponds to 1 μm in all images.

Task 1: Magnetogenetic manipulation with small nanoparticle

Flexible application of magnetogenetics in cell culture and tissue requires improved tools for intracellular delivery of MNPs at sufficiently high levels. In order to simplify nanoparticle uptake into cells and to activate a larger pool of intracellular proteins, much smaller MNPs (~ 40 nm compared to currently 500 nm) will be employed. Particles of this size can be loaded at high concentrations in the cytosol of living cells by means of pinocytotic influx techniques and thus do not require laborious and time-consuming mechanical injection. In preliminary experiments, we have shown that, it was readily possible to generate gradients of these MNPs inside the cytoplasm by simple magnetic devices (Figure 2). We will establish surface functionalization of these MNP in order to minimize recognition by intracellular degradation machineries of the nanoparticles2 and for efficient capturing of effector proteins in the cytoplasm of living cells.3

Figure 2: A- Principle of the formation of an activity gradient with small MNPs (in orange). B- Magnetic manipulation of small MNPs inside living cells. Left: Bright field image of a HeLa cell loaded with MNPs by pinocytosis. A magnetic tip was approached from the bottom left as indicated. Center: Fluorescence image of the MNPs. Accumulation of MNPs towards the magnet is highlighted by the white circle. Right: graded distribution of MNPs along the axis adjoining the tip and the cell, shown in blue in the fluorescence image. Exponential fit of the distribution is shown in red.

In parallel, we will implement fully genetically encoded functionalized MNPs, thus eliminating the need to deliver exogenous nanoparticles into cells. To this end, we will use the iron storage protein ferritin, which naturally form small magnetic nanoparticles.4 In recent experiments, a fusion protein of light chain ferritin and heavy chain ferritin (that maximizes iron binding) was stably overexpressed in eukaryotic cell and loaded with exogenously applied iron without evidence of toxic effects.5 Furthermore, the ferritin nanoparticles were successfully employed to control the production and release of insulin by means of radio-wave heating. Here, we will express the ferritin nanoparticles (fused to a fluorescent marker), and manipulate them with magnetic gradients. We will first check if the magnetic response of these nanoparticles is sufficient for their manipulation inside cells. This project will profit from improved magnetic manipulation using enhanced magnetic gradients, which are ongoing developments in the group of M. Dahan. Once the manipulation of ferritin nanoparticles is established, functionalization with GTPases or GEFs will be readily achieved by means of genetic fusion.

Task 2: Controlling the migration of isolated cells

The next task will be to control the migration of isolated cells in culture. Specific activation of a genetically modified small GTPase Rac1 has been shown to be sufficient to control the direction of cell movement.6 For magnetic control of cell migration, MNPs will be functionalized with TIAM (TIAMMNPs) a specific guanine nucleotide exchange factor activating Rac1. The functionalization will be performed either in vitro before loading the MNPs into cells (to work with genetically unmodified cells) or in situ using TIAM fused to HaloTag. In view of the goals of task 3, the ability to initiate cell migration will be studied in cancer cells that are known to migrate collectively (HT29 and HT116) or as single cells (CT26) and carcinoma-associated fibroblasts (CAFs).

The main goal of this task is to determine the stimulation conditions required to achieve and maintain cell migration over extended time periods. We will thus address the following questions: what are the minimal intracellular gradient steepness and the minimal gradient amplitude that can be detected by the cell? For how long should the stimulation be applied? Answers to these questions are not only important for task 3 but also provide important systems-level information on the sensitivity and robustness with which individual cells process intracellular graded information.

Task 3: Controlling the migration of cancer cells in multicellular environment.

In the third task, we will investigate how cells migrate when they are part of a multicellular environment. The long-term motivation is to investigate how CAFs and cancer cells interact during invasion. So far, it is not clear whether CAFs promote cancer cell invasion by secreting pro-invasion molecules or their physical presence is required. It has been shown that CAFs secrete hepatocyte growth factor and the ECM protein tenascin-C that stimulate invasion.7 On the other hand, others have shown that CAFs prepare the stage for cancer cell migration by remodeling the ECM, and that soluble long distance chemoattractants cannot induce cancer cell invasion.8 Since they also observed that CAFs are always the leading cells of invasive cohorts and that intrinsically non-invasive cancer cells are followers, the question remains whether physical contacts between CAFs and cancer cells are important for invasion; and if both cell types use cellular protrusions such filopodia as guidance organelles?

As a first and simple step, we will use 2D cultures of cancer cells (HT29, HT116 or CT26) on glass substrates. By magnetically stimulating individual cells within the 2D multicellular systems, we will observe how the migration of one cell affects the others and whether coordinated motion of many cells can be induced (Figure 3 A). We will also co-culture CAFs with cancer cells and investigate whether migration of selected and individual CAFs adherent to cancer cells guide their migration. Furthermore, a more physiological situation will consist in using 3D models. We will take advantage of the spheroid technology which has been actively developed at the Institut Curie over the past years.9 By growing cells inside capsules (size ~150 μm), we will test in particular how the 3D cell migration, induced by the field, depends on the density of cells. Finally, we will embed aggregates of cancer cells in a collagen matrix containing CAFs (Figure 3 B). In this co-culture system, we will test whether the magnetically-induced migration of CAFs guide the migration of cancer cells.

Figure 3: Investigation of cancer cell migration in vitro. (A) 2D concept of cancer cell migration: Co-culture of CAFs and cancer cells on glass substrates. Upon microinjection bfMNPs and applying of a magnetic field is thought to stimulate CAF migration. (B) 3D in vitro model of cancer cell migration: Co-culture of CAFs and cancer cells within a collagen matrix.


In this project, we propose to develop and apply a novel methodology to magnetically control intracellular signaling pathways that govern the migration of cells. Compared to currently existing techniques such as mechanical, chemical or optogenetic stimulation, magnetogenetics uniquely allows maintaining stimuli within single cells over extended time periods and with high spatial resolution, as well as the possibility of working with genetically unmodified cells. Note that, if successful, our approach can be applied to a variety of challenges relevant to the scientific themes of the Labex.

We are aware that all the goals of this new and ambitious project might not be reached in a single year. Thus, the candidate will apply to other fellowships (EMBO, DFG, Leopoldina foundation, HFSP, FRM, ARC…) in order to fully address the proposed tasks.


  1. Etoc, F. et al. Subcellular control of Rac-GTPase signalling by magnetogenetic manipulation inside living cells. Nat Nano 8, 5 (2013).
  2. Liße, D., Richter, C., Drees, C., You, C. & Piehler, J. Mono-functional stealth nanoparticles enables unbiased tracking of individual membrane proteins inside living cells. in preparation (2013).
  3. Liße, D., Wilkens, V., You, C., Busch, K. & Piehler, J. Selective Targeting of Fluorescent Nanoparticles to Proteins Inside Live Cells. Angewandte Chemie International Edition 50, 9352-9355 (2011).
  4. Sana, B., Johnson, E., Sheah, K., Poh, C.L. & Lim, S. Iron-based ferritin nanocore as a contrast agent. Biointerphases 5, FA48-52 (2010).
  5. Stanley, S.A. et al. Radio-wave heating of iron oxide nanoparticles can regulate plasma glucose in mice. Science 336, 604-8 (2012).
  6. Machacek, M. et al. Coordination of Rho GTPase activities during cell protrusion. Nature 461, 99-103 (2009).
  7. De Wever, O. et al. Tenascin-C and SF/HGF produced by myofibroblasts in vitro provide convergent pro-invasive signals to human colon cancer cells through RhoA and Rac. FASEB J 18, 1016-8 (2004).
  8. Gaggioli, C. et al. Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells. Nat Cell Biol 9, 1392-400 (2007).
  9. Alessandri, K. et al. Cellular capsules as a tool for multicellular spheroid production and for investigating the mechanics of tumor progression in vitro. Proc Natl Acad Sci U S A 110, 14843-8 (2013).