Abstract
Biological systems have evolved to create a structural and dynamic continuum of biomacromolecular assemblies for the purpose of optimizing their functions. The formation of these dynamic higher-order assemblies are precisely controlled by biological cues. However, controlling self-assembly of synthetic molecules spatiotemporally in or on live cell is still a big challenge, especially for performing functions. This concept article introduces the use of in-situ reactions as a spatiotemporal control to form the assemblies of small molecules that induce cell morphogenesis or apoptosis. After briefly introducing a representative example of natural dynamic continuum of the higher-order assemblies, we describe enzyme-instructed self-assembly (EISA) for constructing dynamic assemblies of small molecules, then discuss the use of EISA for controlling cell morphogenesis and apoptosis. Finally, we provide a brief outlook to discuss the future perspective of this exciting new research direction.
Keywords: self-assembly, dynamic continuum, peptide, cell spheroid, instructed-assembly
Graphical Abstract

Instructed noncovalent synthesis to form dynamic continuum of molecular assemblies in complex condition (e.g., cells) for controlling cell fates.
Introduction
Consisting of higher-order assemblies of biomacromolecules, molecular machineries developed by nature through evolution carry sophisticated biological functions (e.g., cell signaling, regulation, and macromolecular synthesis) in live organism.[1] Understanding of the assemblies of the macromolecules at molecular level in such a machinery has provided new insights of life, which is resulted from the processes of forming higher-order nanostructures. The formed nanostructures depend on assembly and disassembly of the components in a spatiotemporally regulated way to perform functions. Dynamic higher-order structures are prevalent in signal transduction for transmitting receptor activations to cellular responses. In classical concept of signal transduction, ligand induced conformation changes through the formation of trimers or dimers of proper receptors (e.g., tumor necrosis factor receptor (TNFR) and toll-like receptor (TLR) superfamily) are the key factors for the transmission of receptor activation signals to intracellular components. In the higher-order assembly modes, receptors and signaling proteins together form the higher-order oligomeric signalosomes, which activate the signaling cascade to perform functions.[2] For example, dynamic higher-order structures of death domain (DD) superfamily share a common structural feature that comprises of six antiparallel alpha helices,[3] which mediate homotypic interactions within each subfamily and play critical roles in the formation of oligomeric signaling complexes and perform different functions.[3b, 4] Although the concept of dynamic continuums is becoming a norm in cell biology, the applications of such a concept in supramolecular chemistry, chemical biology, and materials science are only at the beginning.
In the past one and half decades, we have integrated the process of self-assembly of small molecules with enzymatic reaction for manipulating the properties of supramolecular assemblies and for exploring new strategies for treating human diseases.[5] Although the formation of nanoscale assemblies by enzyme-instructed self-assembly (EISA) in-situ (i.e., on or inside live cells) can selectively cause the death of cancer cells[6] or profile enzyme activities,[7] only recently, we discovered that partially enzymatic dephosphorylation of precursors can generate the dynamic and hierarchical assemblies by two components to enable cell spheroid formation.[8] Moreover, by controlling the activity of enzymes of live cells, we also are able to control cell morphogenesis or apoptosis in a context-dependent manner.[9] In this short article, we describe the discovery, the design principle, and potential applications of the concept of dynamic continuum of molecular nanostructures.
Dynamic continuum of molecular assemblies
Driving by noncovalent interactions, molecular self-assembly is an ingenious strategy used by nature to make functional structures from biomacromolecules. After its relatively slow progress in the early exploration,[10] self-assembly of small molecules in water, in recent years, has emerged as a research frontier of supramolecular chemistry. More and more scientists are active and have made considerable contribution in this field.[5b, 11] A variety of triggers can induce self-assembly of small molecules, including pH, sonication, ionic strength adjustments, temperature modulation, ligand-receptor interaction, light irradiation, and organic solvent assistance.[5a, 12] Inspired by that endogenous cellular assemblies are the functional products of enzyme controlled self-assembly in cellular environment, we have developed enzyme instructed self-assembly (EISA) of small molecules to form biocompatible materials (e.g., hydrogels) for biomedical applications, including smart drug delivery, assisting biomineralization, enzyme immobilization, inhibitor screening, and enzyme activity detection.[5b] During the exploration of EISA, we realized that most of cancer cells overexpress certain enzymes compared to their normal cell counterpart. Incubating such cells with the precursor, which is the substrate of the overexpressed enzyme, should be possible to form supramolecular structures in-situ (i.e., nanostructure localized with these cells) for functions. Bearing this idea in mind, we have used the overexpressed enzymes to catalyze the formation of anticancer nanofibers of peptides inside[6a] or on the cells.[6b] We also established the correlation among enzyme activity, self-assembling ability, and anticancer efficiency.[13] During these studies, we also found that the self-assembly processes in cell microenvironment are more important than the final state of assemblies, and the precursors that exist in the assemblies play an important role for defining their functions. This discovery leads to the concept of dynamic continuum of molecular assemblies, a consequence related to EISA[8] and inspired by the concept of high-order assemblies in biology.[4a]
Figure 1 illustrates the generation of higher-order assemblies of two components in the context of cells. Glycosylation and phosphorylation of proteins play important roles in the formation of higher-order structures that perform biological functions. However, it is difficult to synthesize glycoproteins and phosphoproteins by conventional covalent synthesis.[14] Thus, we used noncovalent synthesis as a facile route to generate the functional mimics of glycoproteins and phosphoproteins and explored the potential functions of the resulting supramolecular phosphoglycopeptides (sPGP). We also intended to combine ligand-receptor interactions and EISA because both are prevalent triggering mechanisms in nature for generating sophisticated functions in cell environment.[15] Vancomycin (as the glycopeptide) binds to its receptor 1P with high affinity to form nanoparticles in aqueous solution, which transform to uniform nanofibers upon the addition of alkaline phosphatase (ALP). Compared with 1P itself, introducing ligand of 1P prevents the complete dephosphorylation of 1P, resulted in higher-order assemblies with phosphorylation and glycosylation. Such morphology transition occurs in cultured cells that express moderate amount of ALP on their surface, which induces cell spheroids formation from 2D cell sheet (Figure 1D). This study, to the best of our knowledge, is the first discovery that uses enzymatic noncovalent synthesis to construct intercellular dynamic functional materials for controlling the morphogenesis of live cells.
Figure 1.

(A) Ligand–receptor interactions between vancomycin (Van, 2) and the d-Ala-d-Ala containing phosphopeptides (Nap-FFpYGGaa, 1P) result in a sPGP (1P:2). (B) Illustration of sPGPs, formed by noncovalent interactions, as a dynamic continuum in the cell milieu. (C) Transmission electron microscopic (TEM) images of 1P:2 (300 μM) before (up) and after (down) being treated by a phosphatase (ALP, 1 U/mL. 24 h). Scale bar is 20 nm. (D) The formation of cell spheroids from a 2D cell sheet upon the addition of sPGP and the reversibility of the process. Scale bar is 50 μm.
Context-dependent molecular assemblies for controlling cell fate
The phenomenon of context-dependent signals (or functions) is ubiquitous in cell biology,[16] but transferring this knowledge into the field of supramolecular chemistry remains rare. As an advancement of the dynamic continuum for controlling cell morphogenesis, we further developed the dynamic continuum of noncovalent assemblies formed by self-assembly of the sPGP as context-dependent signals for controlling death (e.g., apoptosis) and morphogenesis of live cells. The ectoenzyme (i.e., alkaline phosphatase) on Saos-2 cell surface triggers self-assembly of sPGP to form cytotoxicity nanofibers, which result in cell death. On the contrary, inhibiting the activity of the alkaline phosphatase induces the formation of cell spheroids.[9] More importantly, modulating the expression of ectoenzymes by controlling the ratio of stromal and cancer cells in a co-culture of Saos-2 and HS-5 cells, the sPGP also is able to generate the heterotypic cell spheroids (Figure 2), which mimic the tumor microenvironment that consists multiple cell types. These results validated the feasibility of controlling enzymatic noncovalent synthesis as a strategy to mimic the essence of context-dependent signaling for defining phenotypes.
Figure 2.

(A) The illustration of the formation of 3D cell spheroids made of HS-5 and Saos-2 cells from a 2D cell sheet upon the addition of 1P:2. (B) Expression levels of ALPL on HS-5, Saos-2, and the mixture of HS-5 and Saos-2 at different ratios. The total cell numbers in each group are same. (C) CLSM images of the co-culture of HS-5 and Saos-2 (ratio is 1:1) cells at the density of 3×104 treated with mixture of 1P:2 (300 µM) for 48 h. HS-5 cells were treated with Hoechst 33342 (red) for 10 minutes prior to co-culture with the Saos-2 cells that were treated with membrane probe (green)[17] for 1 h.
Outlook
The aforementioned two examples of multiple components dynamic continuum for driving cell spheroid formation should help the development of a simpler strategy (e.g., using only one component) for tuning cell morphogenesis. We envision that incorporation of other established cell specific adhesive molecules[18] for the dynamic continuum will provide unique ways for controlling the intercellular interactions between multiple types of cells. When one develops dynamic continuum of molecular assemblies, it always would be useful by paying attentions to the aspects below listed in Figure 3.
Figure 3.

Potential directions of dynamic continuum of molecular assemblies of synthetic molecules in cell milieu
Aiming for functions.
The design of self-assembly of synthetic molecules in cell and in vivo setting should always aim for functions. Many works have reported beautiful nanostructures generated from biomacromolecules and their mimics (e.g., peptide, DNA, or RNA) in last three decades. Although these nanostructures have offered considerable understanding on the assemblies of biomacromolecules and even stimulated some extraordinary folding art at nanoscales, the future direction of noncovalent synthesis (or supramolecular chemistry) should focus more on functions.
Dynamic and spatiotemporal control.
Biological system have evolved to create a structural and dynamic continuum of biomacromolecular assemblies, which responses to biological cues precisely, for the purpose of optimizing their functions. To develop supramolecular assemblies for performing functions in biological system, we should take dynamics into consideration. Phosphorylation and dephosphorylation are an important posttranslational modification in biological system, and play important roles in signal transduction cascades. For example, the dynamics of P granules in C. elegans embryos are tightly regulated by phosphorylation and dephosphorylation—phosphorylation of certain proteins promotes granule disassembly and dephosphorylation promotes granule assembly.[19] Such dynamic processes control the division of embryo. Similarly, a more ideal in-situ self-assembly of peptides should aim to achieve similar, if not the same, properties.
Reaction-diffusion.
With more exploration of instructed self-assembly,[20] we now had more understanding on their behaviors in complex conditions (e.g., cells). For example, the different enzymatic reaction rates of tyrosine modified nanoparticles with different chirality result in different cell phenotypes, and enable cancer cell death.[21] Moreover, we also found that controlling the reaction-diffusion process can selectively sequester proteins in live cells with addition of ligand that binding to specific protein during instructed self-assembly.[22] These studies underscore the importance of assembling processes, rather than the final state, of the supramolecular assemblies in cellular environment. Following this line of inquiry may led to more exciting and unexpected discoveries in the future.
Potential biomedical applications.
Compared with the monolayer cells, multicellular cell spheroid produce higher level of certain extracellular matrix,[23] and the ECM profiles in tumor spheroids (e.g., osteosarcoma and glioma spheroids) exhibit more close characters to tumors in vivo.[24] For example, in certain tumor spheroids, cell aggregation through cell-cell contact could also induce cell survival rather than apoptosis, which could due to the compensatory signals that generate by cell-cell interaction.[25] These biological discoveries indicate that the way of forming tumor spheroids reported here may be useful for studying various aspects of cells, such as contact inhibition of locomotion, cell-cell (cell-matrix) interaction, mimicking microenvironment of tumors, and mechanical force driven cell adhesion, just name a few.
Acknowledgements
This work was partially supported by NIH (CA142746) and NSF (DMR-1420382). ZF thanks NIH (F99CA234746).
References
- [1].Steven AC, Baumeister W, Johnson LN, Perham RN, Molecular biology of assemblies and machines, Garland Science, 2016.
- [2].Wu H, Cell 2013, 153, 287–292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].(a) Huang B, Eberstadt M, Olejniczak ET, Meadows RP, Fesik SW, Nature 1996, 384, 638; [DOI] [PubMed] [Google Scholar]; (b) Ferrao R, Wu H, Curr. Opin. Struct. Biol 2012, 22, 241–247; [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Park HH, Logette E, Raunser S, Cuenin S, Walz T, Tschopp J, Wu H, Cell 2007, 128, 533–546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].(a) Wu H, Fuxreiter M, Cell 2016, 165, 1055–1066; [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Fu T-M, Li Y, Lu A, Li Z, Vajjhala PR, Cruz AC, Srivastava DB, DiMaio F, Penczek PA, Siegel RM, Mol. Cell 2016, 64, 236–250; [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Kohl A, Grütter MG, Biol CR. 2004, 327, 1077–1086; [DOI] [PubMed] [Google Scholar]; (d) Reed JC, Doctor KS, Godzik A, Sci. Signaling 2004, 2004, re9–re9. [DOI] [PubMed] [Google Scholar]
- [5].(a) Yang Z, Liang G, Xu B, Acc. Chem. Res 2008, 41, 315–326; [DOI] [PubMed] [Google Scholar]; (b) Du X, Zhou J, Shi J, Xu B, Chem. Rev 2015, 115, 13165–13307; [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Feng Z, Zhang T, Wang H, Xu B, Chem. Soc. Rev 2017, 46, 6470–6479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].(a) Yang ZM, Xu KM, Guo ZF, Guo ZH, Xu B, Adv. Mater 2007, 19, 3152–3156; [Google Scholar]; (b) Kuang Y, Shi JF, Li J, Yuan D, Alberti KA, Xu QB, Xu B, Angew. Chem., Int. Ed 2014, 53, 8104–8107; [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Wang HM, Feng ZQQ, Wu DD, Fritzsching KJ, Rigney M, Zhou J, Jiang YJ, Schmidt-Rohr K, Xu B, J. Am. Chem. Soc 2016, 138, 10758–10761; [DOI] [PMC free article] [PubMed] [Google Scholar]; (d) Feng ZQQ, Zhang TF, Wang HM, Xu B, Chem. Soc. Rev 2017, 46, 6470–6479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].(a) Gao Y, Shi JF, Yuan D, Xu B, Nat. Commun 2012, 3; [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Zhou J, Du XW, Berciu C, He HJ, Shi JF, Nicastro D, Xu B, Chem 2016, 1. [DOI] [PMC free article] [PubMed]
- [8].Wang HM, Shi JF, Feng ZQQ, Zhou R, Wang SY, Rodal AA, Xu B, Angew. Chem., Int. Ed 2017, 56, 16297–16301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Wang H, Feng Z, Xu B, Angew. Chem., Int. Ed 2019.
- [10].Estroff LA, Hamilton AD, Chem. Rev 2004, 104, 1201–1218. [DOI] [PubMed] [Google Scholar]
- [11].(a) Shigemitsu H, Hamachi I, Acc. Chem. Res 2017, 50, 740–750; [DOI] [PubMed] [Google Scholar]; (b) Razgulin A, Ma N, Rao JH, Chem. Soc. Rev 2011, 40, 4186–4216; [DOI] [PubMed] [Google Scholar]; (c) Branco MC, Sigano DM, Schneider JP, Current Opinion in Chemical Biology 2011, 15, 427–434; [DOI] [PMC free article] [PubMed] [Google Scholar]; (d) Rajagopal K, Schneider JP, Curr. Opin. Struct. Biol 2004, 14, 480–486; [DOI] [PubMed] [Google Scholar]; (e) Wang HM, Shi Y, Wang L, Yang ZM, Chem. Soc. Rev 2013, 42, 891–901; [DOI] [PubMed] [Google Scholar]; (f) Zhang PC, Cui YG, Anderson CF, Zhang CL, Li YP, Wang RF, Cui HG, Chem. Soc. Rev 2018, 47, 3490–3529; [DOI] [PubMed] [Google Scholar]; (g) Sato K, Hendricks MP, Palmer LC, Stupp SI, Chem. Soc. Rev 2018, 47, 7539–7551; [DOI] [PMC free article] [PubMed] [Google Scholar]; (h) Aida T, Meijer EW, Stupp SI, Science 2012, 335, 813–817; [DOI] [PMC free article] [PubMed] [Google Scholar]; (i) Fleming S, Ulijn RV, Chem. Soc. Rev 2014, 43, 8150–8177. [DOI] [PubMed] [Google Scholar]
- [12].Wang HM, Yang ZM, Adams DJ, Materials Today 2012, 15, 500–507. [Google Scholar]
- [13].(a) Zhou J, Xu B, Bioconjugate Chem 2015, 26, 987–999; [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Feng ZQQ, Wang HM, Chen XY, Xu B, J. Am. Chem. Soc 2017, 139, 15377–15384; [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Zhou J, Du XW, Xu B, Angew. Chem., Int. Ed 2016, 55, 5770–5775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].(a) Gamblin DP, Scanlan EM, Davis BG, Chem. Rev 2009, 109, 131–163; [DOI] [PubMed] [Google Scholar]; (b) Delbianco M, Bharate P, Varela-Aramburu S, Seeberger PH, Chem. Rev 2016, 116, 1693–1752. [DOI] [PubMed] [Google Scholar]
- [15].Lodish H, Darnell JE, Berk A, Kaiser CA, Krieger M, Scott MP, Bretscher A, Ploegh H, Matsudaira P, Molecular cell biology, Macmillan, 2008.
- [16].(a) Moon RT, Kohn AD, De Ferrari GV, Kaykas A, Nat. Rev. Genet 2004, 5, 691; [DOI] [PubMed] [Google Scholar]; (b) Grell M, Douni E, Wajant H, Löhden M, Clauss M, Maxeiner B, Georgopoulos S, Lesslauer W, Kollias G, Pfizenmaier K, Cell 1995, 83, 793–802; [DOI] [PubMed] [Google Scholar]; (c) Nowell CS, Radtke F, Nat. Rev. Cancer 2017, 17, 145. [DOI] [PubMed] [Google Scholar]
- [17].Wang HM, Feng ZQQ, Del Signore SJ, Rodal AA, Xu B, J. Am. Chem. Soc 2018, 140, 3505–3509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Du XW, Zhou J, Guvench O, Sangiorgi FO, Li XM, Zhou N, Xu B, Bioconjugate Chem 2014, 25, 1031–1035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Wang JT, Smith J, Chen B-C, Schmidt H, Rasoloson D, Paix A, Lambrus BG, Calidas D, Betzig E, Seydoux G, Elife 2014, 3, e04591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].He H, Xu B, Bull. Chem. Soc. Jpn 2018, 91, 900–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Du X, Zhou J, Wang J, Zhou R, Xu B, ChemNanoMat 2017, 3, 17–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Feng Z, Wang H, Xu B, J. Am. Chem. Soc 2018, 140, 16433–16437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Nederman T, Norling B, Glimelius B, Carlsson J, Brunk U, Cancer Res 1984, 44, 3090–3097. [PubMed] [Google Scholar]
- [24].(a) Davies C, Müller H, Hagen I, Gårseth M, Hjelstuen M, Anticancer Res 1997, 17, 4317–4326; [PubMed] [Google Scholar]; (b) Paulus W, Huettner C, Tonn JC, Int. J. Cancer 1994, 58, 841–846. [DOI] [PubMed] [Google Scholar]
- [25].(a) Kantak SS, Kramer RH, J. Biol. Chem 1998, 273, 16953–16961; [DOI] [PubMed] [Google Scholar]; (b) Day ML, Zhao X, Vallorosi CJ, Putzi M, Powell CT, Lin C, Day KC, J. Biol. Chem 1999, 274, 9656–9664. [DOI] [PubMed] [Google Scholar]
