A tissue-engineered jellyfish with biomimetic propulsion

Journal name:
Nature Biotechnology
Year published:
Published online

Reverse engineering of biological form and function requires hierarchical design over several orders of space and time. Recent advances in the mechanistic understanding of biosynthetic compound materials1, 2, 3, computer-aided design approaches in molecular synthetic biology4, 5 and traditional soft robotics6, 7, and increasing aptitude in generating structural and chemical microenvironments that promote cellular self-organization8, 9, 10 have enhanced the ability to recapitulate such hierarchical architecture in engineered biological systems. Here we combined these capabilities in a systematic design strategy to reverse engineer a muscular pump. We report the construction of a freely swimming jellyfish from chemically dissociated rat tissue and silicone polymer as a proof of concept. The constructs, termed 'medusoids', were designed with computer simulations and experiments to match key determinants of jellyfish propulsion and feeding performance by quantitatively mimicking structural design, stroke kinematics and animal-fluid interactions. The combination of the engineering design algorithm with quantitative benchmarks of physiological performance suggests that our strategy is broadly applicable to reverse engineering of muscular organs or simple life forms that pump to survive.

At a glance


  1. Key concepts of fluid transport in jellyfish and in vitro implementation.
    Figure 1: Key concepts of fluid transport in jellyfish and in vitro implementation.

    (a) Schematics of jellyfish stroke cycle generating thrust during the power stroke, and feeding currents during the recovery stroke. (b) Controlled configurational change. Symmetric, complete bell contraction is mediated by anisotropic striated muscle tissue, a functional syncytium synchronized by a system of distributed neuronal pacemaker centers (jellyfish, top). This mechanism can be approximated by electrical field stimulation of electromechanically coupled, anisotropic cardiac muscle (medusoid, bottom). In either case, lobed geometry facilitates circumferential constriction of the bell. (c) Stroke kinematics. In the bilayered design of jellyfish (top) and medusoid (bottom), a flexible elastomer opposes an actuator, which promotes asymmetric stroke patterns: active, fast contraction and passive, slow recoil. (d) Fluid dynamics. Fluid velocity gradients, so-called boundary layers, extend effective reach of lobes (top). Overlapping boundary layers close interlobate gaps to oncoming flow. This prevents leakage and inefficient fluid transport despite the presence of gaps (center). Optimized medusoid body geometry favors the formation of boundary layer overlap and thus efficient fluid transport (bottom). (e) Body design of jellyfish (top) and free-swimming medusoid construct (bottom). Comparison demonstrates similar geometry and dimensions but also illustrates that the medusoid constitutes a simplified version of a jellyfish, reduced to elements necessary for propulsive function. (f) Jellyfish 2D muscle architecture (top) was reverse-engineered in medusoids (bottom). Left: Composite brightfield image overlaid with F-actin stain (green) of muscle cell monolayer. Square inset: Close-up on muscle organization at lobe-body junction; F-actin stain (green). Note that jellyfish muscle tissue consists of a single layer of myofibrils, here in focus, whereas engineered medusoid muscle tissue contains a stack of myofibrils, most of them being out of focus and blurring the image at this resolution. Circular inset: microstructure of single myofibril layer; F-actin stain (green), sarcomeric α-actinin (gray). (g) Distribution of actin fiber orientation angles within single myofibril layer (centered on zero). Quantitative analysis of multiple fields of view revealed no significant difference in the orientation organization parameter (OOP) (P = 0.61, n = 10; two-sample t-test).

  2. Medusoids were engineered to exhibit jellyfish-like stroke kinetics.
    Figure 2: Medusoids were engineered to exhibit jellyfish-like stroke kinetics.

    (a) Time lapse of stroke cycle in jellyfish (top) and medusoid paced at 1 Hz (bottom); t, time (sec) elapsed since start of stroke cycle; T, duration of stroke cycle; here: jellyfish, T = 0.3 s; medusoid, T = 1.0 s. (b) Average trace of angular velocity of individual bell lobes throughout stroke cycle in juvenile jellyfish and medusoids (n = 9 lobes each). Inset illustrates characteristic parameters of stroke cycle (top) and velocity-time graph (bottom). Tpower/recovery, duration of power/recovery stroke; Upower/recovery, velocity of power/recovery stroke; Ûpower/recovery, peak velocity of power/recovery stroke; tpower/recovery, time point of peak power/recovery stroke velocity. (c,d) Box-plot representation. Bull's eyes, median; lower edge of box, 25th percentile; upper edge of box, 75th percentile; whiskers, extreme data points. (c) Relative asynchrony of lobe contraction. Asynchrony did not differ significantly in jellyfish and medusoids (P = 0.7, Wilcoxon rank sum test, n = 4 lobe pairs each). Δtpower, difference between time points of peak power stroke velocities in pair of lobes. (d) Ratio of maximal lobe velocities during power and recovery stroke did not differ significantly in jellyfish and medusoids (P = 0.7, Wilcoxon rank sum test, n = 9 lobes each).

  3. Medusoids generated jellyfish-like flow fields.
    Figure 3: Medusoids generated jellyfish-like flow fields.

    (ac) Velocity field at end of power stroke reveals similar thrust generation in a, jellyfish and b, medusoids, and reduced thrust in c, sieve-designed medusoids. White arrows, lobe motion. (df) Vorticity field during recovery stroke reveals similar formation of stopping vortices in jellyfish (d), medusoids (e) and sieve-designed medusoid (f). Blue and red contours, counter-rotating cross-sections of stopping vortex ring; gray arrows, lobe motion.

  4. Medusoids achieved functional performance of jellyfish.
    Figure 4: Medusoids achieved functional performance of jellyfish.

    (a) Box-plot representation of relative propulsion performances in BL/S. Performance in jellyfish (n = 7, red) and optimally designed medusoids (n = 11, black) spans similar range. Compared to optimal conditions, jellyfish in sieve conditions (n = 7, blue) and sieve-designed medusoids (n = 6, gray) performed significantly worse (P = 0.02 and P = 0.00015, respectively; Wilcoxon rank sum test). Bull's eyes, median; lower edge of box, 25th percentile; upper edge of box, 75th percentile; whiskers, extreme data points not considered outliers; circle, outlier. Asterisks denote statistically significant difference, P < 0.05. (b) For each stage of the recovery stroke, flow profile (i) and volume flow rate (ii) across subumbrellar reference section reveal similar fluid transport, that is, 'feeding current' in jellyfish and medusoids, and reduced performance—including flow reversal—in sieve-designed medusoids. Data shown for one representative sample each. Left: reference section at each stage of recovery stroke. Measured quantities are nondimensionalized to facilitate comparison. R, bell radius (mm); jellyfish: R = 9 mm; medusoid: R = 9 mm; sieve-designed medusoid: R = 9 mm; r, radial distance (mm) from center. Trecovery, duration (s) of recovery stroke; jellyfish: T = 0.2 s, medusoid, T = 0.48 s; sieve-designed medusoid, T = 0.55 s; t, time (s) elapsed since start of recovery stroke. U, flow velocity [mm/s] normal to reference section; jellyfish: Umax = 9 mm/s; medusoid: Umax = 2.4 mm/s; sieve-designed medusoid: Umax = 1.6 mm/s. Q, volume flow rate (mm3/s) passing across reference area (circle with diameter reference section); jellyfish: Qmax = 100 mm3/s; medusoid: Qmax = 40 mm3/s; sieve-designed medusoid: Qmax = 15 mm3/s.


  1. Mitragotri, S. & Lahann, J. Physical approaches to biomaterial design. Nat. Mater. 8, 1523 (2009).
  2. Place, E.S., Evans, N.D. & Stevens, M.M. Complexity in biomaterials for tissue engineering. Nat. Mater. 8, 457470 (2009).
  3. von der Mark, K., Park, J., Bauer, S. & Schmuki, P. Nanoscale engineering of biomimetic surfaces: cues from the extracellular matrix. Cell Tissue Res. 339, 131153 (2010).
  4. Kaznessis, Y.N. Models for synthetic biology. BMC Syst. Biol. 1, 47 (2007).
  5. Fu, P. A perspective of synthetic biology: assembling building blocks for novel functions. Biotechnol. J. 1, 690699 (2006).
  6. Bar-Cohen, Y. EAP as artificial muscles: progress and challenges. Proc. SPIE 5385, 1016 (2004).
  7. Bar-Cohen, Y. Biomimetics–using nature to inspire human innovation. Bioinspir. Biomim. 1, P1P12 (2006).
  8. Grosberg, A. et al. Self-organization of muscle cell structure and function. PLOS Comput. Biol. 7, e1001088 (2011).
  9. Alford, P.W., Feinberg, A.W., Sheehy, S.P. & Parker, K.K. Biohybrid thin films for measuring contractility in engineered cardiovascular muscle. Biomaterials 31, 36133621 (2010).
  10. Huang, A.H., Farrell, M.J. & Mauck, R.L. Mechanics and mechanobiology of mesenchymal stem cell-based engineered cartilage. J. Biomech. 43, 128136 (2010).
  11. Arai, M.N. A Functional Biology of Scyphozoa (Springer; 1997).
  12. Anderson, P.A.V. & Schwab, W.E. The organization and structure of nerve and muscle in the jellyfish Cyanea capillata (coelenterata; scyphozoa). J. Morphol. 170, 383399 (1981).
  13. Dabiri, J.O., Colin, S.P., Costello, J.H. & Gharib, M. Flow patterns generated by oblate medusan jellyfish: field measurements and laboratory analyses. J. Exp. Biol. 208, 12571265 (2005).
  14. Costello, J.H. & Colin, S.P. Flow and feeding by swimming scyphomedusae. Marine. Biol. 124, 399406 (1995).
  15. Feitl, K.E., Millett, A.F., Colin, S.P., Dabiri, J.O. & Costello, J.H. Functional morphology and fluid interactions during early development of the scyphomedusa Aurelia aurita. Biol. Bull. 217, 283291 (2009).
  16. Nawroth, J.C., Feitl, K.E., Colin, S.P., Costello, J.H. & Dabiri, J.O. Phenotypic plasticity in juvenile jellyfish medusae facilitates effective animal-fluid interaction. Biol. Lett. 6, 389393 (2010).
  17. Gladfelter, W.G. A comparative analysis of the locomotory systems of medusoid Cnidaria. Helgol. Wiss. Meeresunters. 25, 228272 (1973).
  18. Widmer, C.L. How to Keep Jellyfish in Aquariums: An Introductory Guide for Maintaining Healthy Jellies (Wheatmark; 2008).
  19. Gladfelter, W.B. Structure and function of the locomotory system of the Scyphomedusa Cyanea capillata. Marine Biol. 14, 150160 (1972).
  20. Passano, L.M. Pacemakers and activity patterns the medusae: homage to Romanes. Am. Zool. 5, 465481 (1965).
  21. Satterlie, R.A. Neuronal control of swimming in jellyfish: a comparative story. Can. J. Zool. 80, 16541669 (2002).
  22. Koehl, M.A. et al. Lobster sniffing: antennule design and hydrodynamic filtering of information in an odor plume. Science 294, 19481951 (2001).
  23. Koehl, M.A.R. Biomechanics of microscopic appendages: functional shifts caused by changes in speed. J. Biomech. 37, 789795 (2004).
  24. Chapman, D.M. Microanatomy of the bell rim of Aurelia aurita (Cnidaria: Scyphozoa). Can. J. Zool. 77, 3446 (1999).
  25. Blanquet, R.S. & Riordan, G.P. An ultrastructural study of the subumbrellar musculature and desmosomal complexes of cassiopea xamachana (Cnidaria: Scyphozoa). Trans. Am. Microsc. Soc. 100, 109119 (1981).
  26. Hayward, R.T. Modeling Experiments on Pacemaker Interactions in Scyphomedusae. PhD thesis, UNCW left fencehttp://libres.uncg.edu/ir/uncw/listing.aspx?id=1787right fence (2007).
  27. Grosberg, A., Alford, P.W., McCain, M.L. & Parker, K.K. Ensembles of engineered cardiac tissues for physiological and pharmacological study: heart on a chip. Lab Chip 11, 41654173 (2011).
  28. Böl, M., Reese, S., Parker, K.K. & Kuhl, E. Computational modeling of muscular thin films for cardiac repair. Comput. Mech. 43, 535544 (2008).
  29. Shim, J., Grosberg, A., Nawroth, J.C., Parker, K.K. & Bertoldi, K. Modeling of cardiac muscular thin films: pre-stretch, passive and active behavior. J. Biomech. 45, 832841 (2012).
  30. Feinberg, A.W. et al. Muscular thin films for building actuators and powering devices. Science 317, 13661370 (2007).
  31. Marder, M., Deegan, R. & Sharon, E. Crumpling, buckling, and cracking: elasticity of thin sheets. Phys. Today 60, 3338 (2007).
  32. Vogel, S. Life in Moving Fluids: the Physical Biology of Flow (Princeton University Press, 1996).
  33. Fedorov, V.V. et al. Application of blebbistatin as an excitation-contraction uncoupler for electrophysiologic study of rat and rabbit hearts. Heart Rhythm 4, 619626 (2007).
  34. Bayly, P.V. et al. Estimation of conduction velocity vector fields from epicardial mapping data. IEEE Trans. Biomed. Eng. 45, 563571 (1998).
  35. Bray, M.-A., Sheehy, S.P. & Parker, K.K. Sarcomere alignment is regulated by myocyte shape. Cell Motil. Cytoskeleton 65, 641651 (2008).
  36. Hamley, I.W. Liquid crystals. in Introduction to Soft Matter: Synthetic and Biological Self-Assembling Materials, pp. 221273 (John Wiley & Sons, 2007) left fencehttp://onlinelibrary.wiley.com/doi/10.1002/9780470517338.ch5/summaryright fence.
  37. Umeno, A. & Ueno, S. Quantitative analysis of adherent cell orientation influenced by strong magnetic fields. IEEE Trans. Nanobioscience 2, 2628 (2003).
  38. Satheesh, V.K., Chhabra, R.P. & Eswaran, V. Steady incompressible fluid flow over a bundle of cylinders at moderate Reynolds numbers. Can. J. Chem. Eng. 77, 978987 (1999).
  39. Masliyah, J.H. & Epstein, N. Steady symmetric flow past elliptical cylinders. Ind. Eng. Chem. Fundam. 10, 293299 (1971).
  40. Halliday, D., Resnick, R. & Walker, J. Fundamentals of Physics (John Wiley and Sons, 2010).
  41. Adrian, R.J. Particle-imaging techniques for experimental fluid mechanics. Annu. Rev. Fluid Mech. 23, 261304 (1991).
  42. Willert, C.E. & Gharib, M. Digital particle image velocimetry. Exp. Fluids 10, 181193 (1991).
  43. Abramoff, M., Magelhaes, P. & Ram, S. Image processing with ImageJ. Biophotonics Int. 11, 3642 (2004).

Download references

Author information


  1. Division of Biology, California Institute of Technology, Pasadena, California, USA.

    • Janna C Nawroth
  2. Disease Biophysics Group, Wyss Institute for Biologically Inspired Engineering and School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts, USA.

    • Hyungsuk Lee,
    • Adam W Feinberg,
    • Crystal M Ripplinger,
    • Megan L McCain,
    • Anna Grosberg &
    • Kevin Kit Parker
  3. Graduate Aeronautical Laboratories and Bioengineering, California Institute of Technology, Pasadena, California, USA.

    • John O Dabiri


K.K.P., J.O.D. and J.C.N. conceived the project, designed the experiments and prepared the manuscript. J.O.D. and J.C.N. developed the fluid model. J.C.N. did the experiments and analyzed the data. H.L., A.W.F., C.M.R., M.L.M. and A.G. supervised experiments, analyzed data and gave conceptual advice. M.L.M. isolated rat cardiomyocytes for experiments.

Competing financial interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to:

Author details

Supplementary information

PDF files

  1. Supplementary Text and Figures (6M)

    Supplementary Methods and Supplementary Figs. 1-8


  1. Supplementary Movie 1 (4M)

    This movie shows the stroke cycle in a free-swimming juvenile Moon jellyfish (Aurelia aurita)consisting of muscle-powered contraction (power stroke) and elastic recoil (recovery stroke)

  2. Supplementary Movie 2 (2M)

    This movie shows the failure of a biomimetic Medusoid design to propel itself. The Medusoidconstruct was paced at 1 Hz through an externally applied monophasic square pulse (2.5V/cm,10ms duration). Due to mismatch of muscle stresses and substrate compliance, musclecontraction does not result in sufficient bell deformation, and no thrust is generated

  3. Supplementary Movie 3 (6M)

    This movie demonstrates jellyfish-like body contraction and free-swimming of optimallydesigned Medusoid constructs. Medusoids were paced at 1 Hz through an externally appliedmonophasic square pulse (2.5V/cm, 10ms duration). The first scene shows a mature constructstill attached at its center to its casting mold, just prior to release. (Note that the striatedappearance of the casting mold is caused by its fabrication process; in particular, this striationdoes not reflect the alignment of the muscle tissue on top of the silicone membrane covering themold). Subsequent scenes show exemplary propulsion of free-swimming Medusoids

  4. Supplementary Movie 4 (3M)

    This movie demonstrates comparative propulsion efficacy (distance per stroke) in jellyfish andoptimal Medusoids, whereas suboptimal Medusoids (“sieve design”) exhibit inferiorperformance. While Medusoids were paced at 1 Hz through an externally applied electric field of6 V, here the jellyfish contracts at a frequency of ca. 2 Hz. In order to facilitate comparison, theframe rate of the jellyfish recording was halved to synchronize stroke phases

  5. Supplementary Movie 5 (3M)

    This movie shows a sequence of raw DPIV data for jellyfish, Medusoid and suboptimally (=sieve-) designed Medusoids. Here, fluid flow around the jellyfish/Medusoid bell is visualized by the displacement of neutrally buoyant particles suspended within the fluid and illuminatedwithin a single plane using laser light. The relative motion of the particles allows quantifying 2Dfluid flow within the plane

  6. Supplementary Movie 6 (6M)

    This movie shows the fluid flow field and the vorticity field of a juvenile jellyfish during thestroke cycle. The movie was generated from DPIV data. The power stroke is characterized bymaximal fluid velocities and formation of a starting vortex, generating thrust. The recoverystroke is characterized by reduced fluid velocities and the formation of a stopping vortex,generating feeding currents towards the subumbrellar side

  7. Supplementary Movie 7 (5M)

    This movie shows the fluid flow field and the vorticity field of an optimally designed Medusoidduring the stroke cycle. The movie was generated from DPIV data. As in the jellyfish, the powerstroke generates maximal fluid velocities and a starting vortex ring, resulting in thrust. Therecovery stroke generates a stopping vortex ring that draws feeding currents towards the subumbrella

  8. Supplementary Movie 8 (6M)

    This movie shows the fluid flow field and the vorticity field of a suboptimal Medusoid design.The movie was generated from DPIV data. In contrast to jellyfish and optimal Medusoids,suboptimal Medusoids with “sieve design” fail to sufficiently accelerate fluid during the powerstroke, resulting in poor thrust generation. Vorticity patterns are more diffuse compared to thoseobserved in jellyfish and optimal designs, and further flow analysis revealed that generation offeeding currents was inferior as well (Fig. 4)

Additional data