Interactive models of communication at the nanoscale using nanoparticles that talk to one another


‘Communication’ between abiotic nanoscale chemical systems is an almost-unexplored field with enormous potential. Here we show the design and preparation of a chemical communication system based on enzyme-powered Janus nanoparticles, which mimics an interactive model of communication. Cargo delivery from one nanoparticle is governed by the biunivocal communication with another nanoparticle, which involves two enzymatic processes and the interchange of chemical messengers. The conceptual idea of establishing communication between nanodevices opens the opportunity to develop complex nanoscale systems capable of sharing information and cooperating.


Nanotechnology has undergone a remarkable growth in recent years, and a large number of nanodevices such as nanomemories1, nanobatteries2,3, nanocontainers4 and nanomotors5 have already been developed. Nevertheless, communication between human-made nanodevices remains almost unexplored. Experts in telecommunication and computer engineering have already envisioned the interconnection of nanodevices in ‘nanonetworks’ and their virtually-unlimited applications in different fields6,7,8. At the nanoscale, traditional communication technologies are not applicable given the large size and power requirements of classical transceivers, receivers and other components9. In this context, an interesting approach for establishing communication at the nanometric level is to mimic how nature communicates. Chemical or molecular communication10, based on transmitting and receiving information by means of molecules (chemical messengers), is the communication form used by living organisms. For instance, cells communicate with neighbours by exchanging chemicals11,12,13; neurons propagate, share and process information by using neurotransmitters14,15; and insects, bacteria and many mammals use pheromones to communicate with members of their same species. Inspired by biological organisms and cells, scientists have reported, for instance, the use of DNA systems and enzymatic cascades as a tool for information processing and computing16,17. Moreover, chemical logic systems based on individual molecules18,19 and bio-molecular networks20,21,22 have also been developed23,24,25,26.

However, despite these interesting advances, communication between nanoparticles has barely been explored. In this scenario, there are many unanswered questions about the construction of nanonetworks that integrate nanoparticles and molecules, such as: Which molecules should be used to encode information? How will these molecules be recognized? How can recognition be converted into propagation of information, and how can this information be reported? Addressing these questions is no trivial matter, and the experimental realization of these systems is still to come.

In communication theory terms, communication can be defined as the process of establishing a connection between two points for information exchange27. At one point of the communication process, there is the sender of information, which receives a stimulus to send a message across. The sender converts information into a code and transmits the message through an appropriate medium. The receiver perceives the message and decodes it. Communication models are systematic representations of the process that helps to understand how communication can be done. For example, in the linear model of communication, the message flows in a straight line from the sender to the receiver, and there is no feedback concept. In the more complex interactive model, the sender channels a message to the receiver and the receiver then sends feedback and channels a message to the original sender28. In this context, the interactive model is like two linear models piled on top of each other (Fig. 1). Communication is considered effective if it receives the desired result, response or reaction.

Figure 1: Illustration of an interactive model of communication.

The sender receives a stimulus and encodes a message for the receiver. The receiver interprets the message and returns feedback to the first point.

To design human-made nanodevices capable of communicating on the nanoscale, here we show a chemical communication process between gated nanoparticles, which mimics the interactive communication model shown in Fig. 1. It employs Janus Au-mesoporous silica gated nanoparticles containing a mesoporous face, loaded with a cargo and capped with stoppers that can be opened in the presence of a specific stimulus29,30,31,32,33,34,35; and an Au face that is functionalized with different bio-molecules36,37. Cargo delivery from one nanoparticle only occurs after biunivocal communication with the second nanoparticle through two enzymatic processes and the exchange of two chemical messengers.


Communication system design and operation

A representation of the communication process we report herein is depicted in Fig. 2. The first nanomachine (S1gal) is loaded with (Ru(bpy)3)2+, capped with β-cyclodextrin (β-CD) attached through disulfide bonds to the mesoporous face and functionalized with enzyme β-galactosidase on the Au face. The second nanomachine (S2gox) is loaded with N-acetyl-L-cysteine, capped with a pH-responsive β-CD:benzimidazole supramolecular nanovalve on the mesoporous face and functionalized with glucose oxidase (GOx) on the Au face. When S1gal and S2gox are placed together in an aqueous medium, addition of lactose triggers the communication process. First, lactose is hydrolysed by β-galactosidase into galactose and glucose. The produced glucose is transmitted through the aqueous medium towards the Au face of S2gox, where it is recognized by glucose oxidase and hydrolysed into gluconic acid (pKa=3.6). The generation of gluconic acid induces a local drop in pH that causes the protonation of benzimidazole groups (pKa=5.55)38 on the mesoporous face of S2gox and the dethreading of the supramolecular nanovalve. S2gox uncapping results in the delivery of entrapped N-acetyl-L-cysteine, which diffuses as feedback toward S1gal. Finally, N-acetyl-L-cysteine induces the rupture of the disulfide linkages39 on the mesoporous face of S1gal, and (Ru(bpy)3)2+ is released into the medium. Delivery of (Ru(bpy)3)2+ (output) is expected to occur only when the two nanoparticles communicate.

Figure 2: Representation of the interactive communication process between two Janus gated nanodevices.

Chemical input (lactose) is hydrolysed by β-galactosidase on S1gal to glucose (messenger 1), which is transformed into gluconic acid on S2gox. This induces a local drop in pH, which induces the dethreading of the β-CD:benzimidazole supramolecular nanovalve and cargo N-acetyl-L-cysteine release (messenger 2), which is the feedback that finally induces the delivery of the (Ru(bpy)3)2+ reporter from S1gal.

In terms of the interactive model of communication shown in Fig. 1, S1gal nanoparticles (the sender) receive a stimulus (lactose) and transmit information to the receiver (S2gox) via a messenger (glucose). At this point the receiver (S2gox) perceives the message and operates as a new sender by channelling the message to the original sender (S1gal) by using a second chemical messenger (N-acetyl-L-cysteine). Finally the new receiver (S1gal) interprets the message, which results in a final desired response (that is, delivery of (Ru(bpy)3)2+).

Synthesis of the nanodevices

In order to prepare S1gal and S2gox, mesoporous silica (MS) nanoparticles were first obtained by the hydrolysis and condensation of tetraethyl orthosilicate in basic media using n-cetyltrimethylammonium bromide as a template. The surfactant was removed by calcination in air at a high temperature, which yielded the starting mesoporous support. In a second step, gold nanoparticles were prepared according to the Turkevich-Frens method40,41. Then, MS nanoparticles were confined at the interface of Pickering emulsion between paraffin and an aqueous face in order to achieve their partial functionalization with (3-mercaptopropyl)trimethoxysilane, on which Au nanoparticles were attached by the formation of S-Au bonds. Paraffin wax was removed by washing the solid with CHCl3, which yielded the initial Janus Au-MS nanoparticles (S0). To prepare S1gal, the mesoporous face of S0 was loaded with (Ru(bpy)3)2+ and was first functionalized with 3-mercaptopropionic acid on the Au face, and later with (3-mercaptopropyl)trimethoxysilane on the mesoporous face. This solid was capped by the attachment of previously modified β-CDs through disulfide linkages, which yielded solid S1. Finally, the β-galactosidase enzyme was covalently immobilized by crosslinking the primary amine groups in the enzyme with carboxylic acid moieties on the Au face, which yielded the final nanoparticles S1gal. To obtain S2gox, firstly a fraction of the S0 was functionalized with (3-iodopropyl)trimethoxysilane on the mesoporous face. Then benzimidazole moieties were attached to the iodopropyl moieties through a nucleophilic substitution reaction and the Au-face was functionalized with 3-mercaptopropionic acid. Next, pores were loaded with N-acetyl-L-cysteine and capped by the formation of inclusion complexes between benzimidazole groups and β-CDs (K11=104 M−1)42, which resulted in solid S2. The glucose oxidase enzyme was then anchored to the Au face using a crosslinking reaction to yield S2gox. Furthermore, in order to demonstrate that S1gal and S2gox are essential for observing communication, S2blank was synthesized, which contained the same components as S2gox, but lacked the cargo inside the pore voids. In order to assess the capping/uncapping performance of the second nanodevice, we synthesized S2dye, which contained the same components as S2gox but was loaded with (Ru(bpy)3)2+ which facilitated cargo release monitoring by fluorometric techniques.

Characterization of the nanodevices

The different nanoparticles were characterized by standard methods (see Methods and Supplementary Methods for details). The mesoporous morphology of the MS nanoparticles (81±18 nm) and the presence of the Au nanoparticles (19±4 nm) in Janus colloids S0 was confirmed by transmission electron microscopy (TEM) analysis (see Fig. 3a,b). The powder X-ray diffraction (PXRD) pattern of the starting MS nanoparticles showed the characteristic mesoporous reflection peak (100) at around 2.4°, which thus confirmed the ordered mesoporous structure (see Fig. 3c). The preservation of this typical peak in the following solids (S0, S1, S2, S1gal, S2gox) (Supplementary Fig. 1) clearly confirmed that the surface functionalization and cargo loading processes did not damage the mesoporous scaffolding. The diffraction pattern at high angles for all the Janus colloids showed the characteristic cubic gold nanocrystals (111), (200), (220) and (311) peaks43, which confirmed the Au-MS architecture observed by TEM. The starting gold colloid shows a single absorption band at 520 nm, whereas there is a redshift of the absorbance maximum (535 nm) for S0 (Supplementary Fig. 2). Regarding the N2 adsorption-desorption isotherms, starting MS and Janus nanoparticles S0 showed an adsorption step at intermediate P/P0 values (0.1–0.3), which indicates the presence of empty pores in the solid structure (Supplementary Fig. 3). As a result of cargo loading and capping the N2 adsorption-desorption isotherms for the S1 and S2 nanoparticles led to a considerable reduction in the N2 volume absorbed, and curves were flat compared to the parent solids (Supplementary Fig. 3). The Brunauer–Emmett–Teller (BET) total specific surface area, pore volumes and pore sizes were calculated are summarized in Table 1. From the thermogravimetric and elemental analysis studies, the contents of (Ru(bpy)3)2+ and β-CD on S1 were determined as 119 and 78 mg per gram of solid, respectively (Supplementary Table 1, Supplementary Equations 1–5, and Supplementary Fig. 4). For S2, N-acetyl-L-cysteine and benzimidazole contents were determined as 31 and 58 mg per gram of solid, respectively. Furthermore, the presence of immobilized enzymes on S1gal and S2gox was confirmed by running specific glucose oxidase and β-galactosidase activity assays on each nanodevice (Supplementary Fig. 5 and Supplementary Equations 6–11).

Figure 3: TEM images and PXRD of nanoparticles.

TEM of (a) calcined MS nanoparticles and (b) Janus Au-MS nanoparticles (S0) showing the typical porosity of the MCM-41 matrix. Scale bars represent 50 nm. Powder X-ray diffraction pattern of calcined MS nanoparticles, S1gal and S2gox at low (c) and high (d) angles. Note the presence of the characteristic gold peaks for Janus colloids around 38°, 44° and 65° and 78°.

Table 1 The BET specific surface values, pore volumes and pore sizes calculated from the N2 adsorption-desorption isotherms for the selected materials.

To investigate the cooperative communication between S1gal and S2dye, we confirmed the optimal capping/uncapping behaviour of the molecular gates with the aid of dye-loaded nanoparticles S1gal and S2dye and ultraviolet-visible spectroscopy in the presence or absence of the corresponding messengers. The delivery studies with S1gal revealed that nanoparticles remained capped and displayed no dye release in aqueous solution, whereas the presence of the reducing agent N-acetyl-L-cysteine induced the opening of pores and dye release (Supplementary Fig. 6). With S2dye, it was confirmed that the (Ru(bpy)3)2+ cargo was released only after adding glucose, whereas the solid remained capped in the absence of the messenger (Supplementary Fig. 7).

Nanoparticles that talk to one another

Having characterized the single nanodevices, next we addressed the actual interactive communication process between them. In this complex system, the final release of the reporter (Ru(bpy)3)2+ from nanocarrier S1gal was expected to be related with the information shared between S1gal and S2gox via the exchange of encoding molecules (glucose and N-acetyl-L-cysteine). In a typical experiment, S1gal and S2gox were suspended in aqueous solution at pH 7.5 and shaken over time at 25 °C in the presence or absence of 5 mM lactose, which acted as the input signal (see Methods for details). At scheduled times, aliquots were taken, centrifuged to remove nanoparticles, and the absorbance at 453 nm (maximum of the absorption band of (Ru(bpy)3)2+) was measured. Figure 4 shows the time course of the (Ru(bpy)3)2+ delivery from S1gal in the presence and absence of lactose. When there was no input (black curve), no communication between the nanodevices occurred and no output signal was observed. In contrast, when lactose was introduced into the system (red curve), the biunivocal communication in the nanonetwork was triggered, which resulted in the clear (Ru(bpy)3)2+ reporter release. In the presence of lactose, a total amount of 81 μM of (Ru(bpy)3)2+ was released after 210 min which corresponds to a 39% release efficiency (maximum theoretical release efficiency was calculated by dissolving of S1gal in 20% NaOH solution). The final release of (Ru(bpy)3)2+ was ascribed to an effective interactive model of communication between S1gal and S2gox, as indicated in Fig. 2.

Figure 4: Cargo release from the interactive communication between nanoparticles.

(Ru(bpy)3)2+ release in aqueous solution at pH 7.5 that contained S1gal and S2gox in the absence (black curve) and presence (red curve) of lactose (5 mM). Error bars correspond to the s.d. from five independent experiments.

For communication to take place, all the individual nanocomponents act together and cooperatively to produce collective behaviour. To demonstrate the crucial role played by the messengers and enzymes in the nanonetwork, additional experiments were carried out with S1 (lacking β-galactosidase), S2 (lacking glucose oxidase) and S2blank (lacking the N-acetyl-L-cysteine messenger). If in the community S1gal/S2gox the information was shared and the final desired response (that is, delivery of (Ru(bpy)3)2+) was observed, mixtures S1/S2gox, S1gal/S2 and S1gal/S2blank should not be able to communicate, which would result in no dye release occurring. The delivery experiments with S1/S2gox, S1gal/S2 and S1gal/S2blank were performed by suspending nanoparticles in an aqueous solution at pH 7.5 in the presence of lactose. In the three uncompleted communities, communication was broken at a certain stage, and the information loop could not close. Therefore, no output signal ((Ru(bpy)3)2+ delivery) was observed (see the release profiles shown in Fig. 5). These experiments clearly stress the essential role played by the different system components. Moreover, the operation in the presence of lactose of the community S1gal/S2gox was compared with the system S1/S2/free enzymes, in which enzymes are freely dissolved in the bulk solution. In these experiments, the enzyme-free solids S1 and S2 were placed in a solution containing β-galactosidase (2 U l−1) and glucose oxidase (0.8 U ml−1) (enzymes were dissolved in the bulk solution at an equivalent concentration to that found in the community S1gal/S2gox). As can be seen in Fig. 5 (purple curve), in the community S1/S2/free enzymes, the response (that is, delivery of the dye from S1) was very low and far from that found for the S1gal/S2gox system. This indicated that in order to have an effective communication, enzymes, in particular glucose oxidase, must be placed in the proximity of the β-CD:benzimidazole complex to be able to generate a local pH drop around the nanoparticle. If the enzymes are in the solution, communication was broken and dye delivery from S1 was not observed.

Figure 5: Cargo release from uncompleted communities of nanoparticles.

(Ru(bpy)3)2+ release in aqueous solution at pH 7.5 in the presence of lactose (5 mM) that contained communities of nanoparticles S1+S2gox (lacking β-galactosidase, blue curve, circles), S1gal+S2 (lacking glucose oxidase, green curve, triangles), S1gal+S2blank (lacking the N-acetyl-L-cysteine messenger, grey curve, squares) and S1+S2+free enzymes (purple curve, circles), (error bars correspond to the s.d. from two independent experiments). Delivery from the full-equipped S1gal+S2gox system (red curve) is also displayed for comparative purposes. Error bars correspond to the s.d. from five independent experiments. Communication was achieved only when no component was lacking.

A highly desirable characteristic for a communication system is to be selective to a certain input, and it should not respond to other similar inputs that may be in the surroundings. To demonstrate the specificity of the interactive communication between S1gal and S2gox triggered by lactose, the performance of aqueous suspensions of the S1gal/S2gox community in the presence of disaccharides maltose and lactulose was also studied. In these studies, no (Ru(bpy)3)2+ delivery was observed (see Fig. 6). Maltose is not recognized by β-galactosidase on S1gal, whereas lactulose is hydrolysed by β-galactosidase into galactose and fructose, but these species are not recognized by glucose oxidase on S2gox. In both cases, communication is disrupted and no output signal was found.

Figure 6: Specificity of the interactive communication.

Release of dye in aqueous solution at pH 7.5 for the full-equipped nanonetwork S1gal+S2gox after 2 h in the presence of three different disaccharides: lactose, maltose and lactulose. Error bars correspond to the s.d. from two independent experiments.

Interactive communication is illustrated in Table 2 in a Boolean logic table that indicates how the final output (delivery of (Ru(bpy)3)2+ from S1gal) is found only when using the complete community of gated nanoparticles (S1gal/S2gox mixture) in the presence of lactose, but not in its absence or when the sender/receiver is not complete (S1, S2, S2blank, nanoparticles that lack β-galactosidase, glucose oxidase or N-acetyl-L-cysteine, respectively).

Table 2 Summary of the response of the communication system when using the full-equipped (S1gal and S2gox) or partially-equipped (S1, S2, S2blank) nanoparticles and the presence or absence of input (lactose).

In conclusion, we have developed an example of communication between nanodevices based on Janus Au-mesoporous silica nanoparticles which mimics an interactive model of communication, in which a sender nanoparticle receives a stimulus and encodes a message for a receiver nanoparticle that interprets the message and returns feedback to the first nanoparticle. In our communication nanoscale system, delivery from one nanoparticle is governed by an interactive biunivocal communication with another nanoparticle, and involves two enzymatic processes and the use of two chemical messengers. We believe that, in communication terms on the nanoscale, the system we report herein would allow advances to be made in the knowledge of how the recognition of individual molecules (via simple chemical or biochemical reactions) can be used to encode information, and how to convert molecular recognition into information propagation. The idea of establishing communication between nanodevices embraces an enormous potential for the design of more advanced and complex nanoscale systems governed by communication between individual nanocomponents. Inspired by how biological and human communities communicate, the development of such nanodevice communities may open new directions in a number of different areas44,45,46,47.


Synthesis of MCM-41 mesoporous silica nanoparticles

Approximately 1.00 g (2.74 mmol) of n-cetyltrimethylammonium bromide (CTABr) was dissolved in 480 ml of deionized water. Then, the pH was basified by adding 3.5 ml of a 2 mol l−1 NaOH solution and the temperature was increased to 80 °C. Afterward, TEOS (5.00 ml, 22.4 mmol) was added dropwise to this solution. Magnetic stirring was kept for 2 h to give a white precipitate. Finally, the solid was isolated by centrifugation, washed several times with water and dried at 70 °C overnight (as-synthesized MCM-41). To obtain the final mesoporous nanoparticles (MCM-41), the as-synthesized solid was calcined at 550 °C using an oxidant atmosphere for 5 h in order to remove the surfactant.

Synthesis of gold nanoparticles

Gold nanoparticles were synthesized based on the Turkevich–Frens method40,41. Briefly, 100 ml of a 3 μM HAuCl4·3H2O solution was heated to boiling under stirring and refluxing. Then, 750 μl of a 3.9 μM trisodium citrate solution was added to synthesize 20 nm gold nanoparticles. The initially faint yellow colour turned to blue-black and finally red-wine in 10 min. After this, the colloidal suspension was let to cool at room temperature.

Synthesis of Janus Au-MS nanoparticles (S0)

Janus nanoparticles were synthesized following a method recently reported by us36,37. MCM-41 mesoporous silica nanoparticles (200 mg) were dispersed in 10 ml of aqueous solution (6.7% ethanol) and n-cetyltrimethylammonium bromide (CTABr) was added for a 1 μM final concentration. The mixture was heated at 75 °C, and then 1 g of paraffin wax was added. Once the paraffin was melted, the mixture was vigorously stirred at 25,000 r.p.m. for 10 min using an Ultra-Turrax T-10 homogenizer (IKA, Germany). Afterward, the mixture was further stirred for 1 h at 4,000 r.p.m. and 75 °C using a magnetic stirrer. The resulting Pickering emulsion was then cooled to room temperature, diluted with 10 ml of methanol and reacted with 200 μl of (3-mercaptopropyl) trimethoxysilane. After 3 h under magnetic stirring, the solid was collected by filtration and washed with methanol. For gold attachment, the partially mercapto-functionalized MCM-41 nanoparticles were dispersed in 75 ml of methanol and added over 400 ml of the as-synthesized gold nanoparticles. The mixture was stirred overnight. Then, the solid was isolated by filtration and exhaustively washed with ethanol and with chloroform. The solid was dried and ground. This process finally yielded the Janus Au-MS nanoparticles (S0).

Synthesis of β-CD-S-SO2CH3

NaSSO2CH3 (70 mg, 0.52 mmol) was added to a solution of mono 6-iodo-6-deoxy-β-cyclodextrin (0.5 g, 4.0 × 10−4mmol) in anhydrous DMF (5 ml) under Argon and warmed to 50 °C while stirring48. After 18 h, the solution was cooled and the solvent evaporated. The residue was washed with EtOH (2 × 3 ml) and acetone (2 × 3 ml), yielding 470 mg of a white solid (94%).

Synthesis of S1

Around 50 mg of Janus Au-MS nanoparticles (S0) were suspended in a concentrated solution of (Ru(bpy)3)Cl2 (25 mg) in acetonitrile (5 ml), and stirred during 24 h in order to achieve the loading of the pores. Then, the suspension was treated with 50 μl of 3-mercaptopropionic acid for 1 h, filtered, and washed with toluene. Afterwards, this solid was suspended in 6 ml of toluene and reacted with an excess of (3-mercaptopropyl) trimethoxysilane (50 μl) for 24 h. The thiol-functionalized solid was treated with 100 mg of potassium tert-butoxide, isolated by centrifugation and washed with toluene and dimethylformamide (DMF). Finally, the solid was suspended with 50 mg of the as-synthesized β-CD-S-SO2CH3 in 10 ml of DMF under inert atmosphere for 24 h in order to cap the pores. Afterwards, the solid was isolated by centrifugation, washed with DMF and acetronitrile and dried under vacuum. This process resulted in the capped Janus solid S1.

Synthesis of S1gal

Around 20 mg of Janus S1 were suspended in 10 ml of 50 mM sodium phosphate buffer (pH 7.5) and then 5 mg of β-galactosidase, 5 mg of EDC and 5 mg of N-hydroxysuccinimide (NHS) were added. The mixture was stirred in an ice bath overnight. The immobilization is based on the coupling reaction between the amine primary groups of the enzyme and the carboxylic groups on the Au surface. The solid was isolated by centrifugation, washed several times with a cold solution of 50 mM sodium phosphate buffer (pH 7.5) and kept wet in the refrigerator until use. This process yielded the final nanomachine S1gal.

Synthesis of S2

Around 100 mg of S0 were first suspended in 10 ml of anhydrous acetonitrile under stirring, and then treated with an excess of (3-iodopropyl) trimethoxysilane (100 μl, 0.5 mmol). The suspension was stirred overnight and then the solid was isolated by centrifugation, washed with acetonitrile and dried at 70 °C overnight. To functionalize the surface with benzimidazole moieties, the resulting solid was ground and suspended in 8 ml of a saturated solution of benzimidazole in toluene at 80 °C and 24 ml of triethylamine were then added (toluene and triethylamine in 1:3v/v ratio). The suspension was stirred and heated at 80 °C for three days. After this, the resulting solid was filtered off, washed with acetonitrile and dried at 70 °C overnight. To protect the gold face, 100 mg of the benzimizadole-functionalized solid was suspended in 8 ml of EtOH and reacted with an excess of 3-mercaptopropionic acid (100 μl) for 1 h. The solid was centrifuged, rinsed with ethanol and with water and let to dry a room temperature for 1day. 100 mg of this solid was suspended in 10 ml aqueous solution of N-acetyl-L-cysteine (0.5 g) at pH 7. After 12 h, 10 ml of 50 mM sodium phosphate buffer containing β-cyclodextrin (1.2 mg) and N-acetyl-L-cysteine (0.250 g) at pH 7.5 were added to the solid suspension and stirred overnight. Finally, the solid was filtered off, washed thoroughly with 50 mM phosphate buffer at pH 7.5 and dried under vacuum for 12 h. This process finally yielded the final solid S2.

Synthesis of S2gox

Around 20 mg of S2 were suspended in 10 ml of 50 mM sodium phosphate buffer at pH 7.5. Then, 5 mg of EDC, 5 mg of NHS and 5 mg of glucose oxidase were added and the suspension was stirred overnight at 0 °C. The solid was isolated by centrifugation and washed several times with cold 50 mM sodium phosphate buffer (pH 7.5). The resulting S2gox was kept wet in refrigerator until use.

Synthesis of S2blank

Solid S2blank was prepared following the same procedure described for S2gox but the mesoporous container was not loaded. First, the mesoporous surface on S0 was modified with benzimidazole moieties and the gold surface was protected with 3-mercaptopropionic acid as described above. 10 mg of this solid was suspended in 5 ml of 50 mM sodium phosphate buffer at pH 7.5 containing β-cyclodextrin (1.2 mg ml−1) and stirred overnight. Then, the solid was filtered off, washed with 50 mM phosphate buffer (pH 7.5) and dried under vacuum for 12 h. To functionalize this solid with the enzyme, we followed the same procedure, as described above. The solid was suspended in 5 ml of 50 mM sodium phosphate buffer (pH 7.5) and 2.5 mg of EDC, 2.5 mg of NHS and 2.5 mg of glucose oxidase were added. The mixture was stirred overnight at 0 °C. Finally, the nanoparticles were isolated by centrifugation and washed several times with cold 50 mM sodium phosphate buffer (pH 7.5). The resulting S2blank was kept wet in the refrigerator until use.

Synthesis of S2dye

Solid S2dye was prepared following the same procedure described for S2gox but the mesoporous container was loaded with (Ru(bpy)3)2+. First, the mesoporous surface on S0 was modified with benzimidazole moieties and the gold surface was protected with 3-mercaptopropionic acid, as described above. 10 mg of this solid was suspended in 5 ml of aqueous solution of (Ru(bpy)3)2+ (5 mg). After 12 h, 10 ml of 50 mM sodium phosphate buffer at pH 7.5 containing β-cyclodextrin (1.2 mg ml−1) were added to the solid suspension and stirred overnight. Then, the solid was filtered off, washed thoroughly with 50 mM phosphate buffer at pH 7.5 and dried under vacuum for 12 h. Finally, we follow the same procedure, as described above in order to attach the enzyme glucose oxidase to the Au face. This process finally yields the solid S2dye that was kept wet in refrigerator until use.

Chemical communication studies

To demonstrate the performance of the nanonetwork, the refrigerated supensions of nanoparticles were aliquoted, washed separately with aqueous solution (20 mM Na2SO4) at pH 7.5 and placed together in the same recipient. In a typical experiment, S1gal and S2gox (4 mg ml−1 and 1 mg ml−1, respectively) were suspended together and shaken overtime at 25 °C in the presence or absence of 5 mM lactose, which acts as the input signal. Aliquots were taken at scheduled times, centrifuged to remove the nanoparticles and the absorbance at 453 nm corresponding to the (Ru(bpy)3)Cl2 released was measured. In order to demonstrate the crucial role played by the components of the system, the same procedure was followed with suspensions of S1/S2gox, S1gal/S2, and S1gal/S2blank. For the release experiments with the community S1/S2/free enzymes, S1 and S2 (4 mg ml−1 and 1 mg ml−1 respectively) were placed in a recipient with Gal (2 U l−1) and GOx (0.8 U ml−1). For specificity experiments, the procedure was the same but maltose (5 mM) and lactulose (5 mM), instead of lactose, were added as the inputs to suspensions of S1gal/S2gox.

Data availability

The authors declare that data supporting the findings of this study are available within the paper and its Supplementary information files. All other relevant data are available from the corresponding authors on request.

Additional information

How to cite this article: Llopis-Lorente, A. et al. Interactive models of communication at the nanoscale using nanoparticles that talk to one another. Nat. Commun. 8, 15511 doi: 10.1038/ncomms15511 (2017).

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1

    Tseng, R., Huang, J., Ouyang, J., Kaner, R. & Yang, Y. Polyaniline nanofiber/gold nanoparticle nonvolatile memory. Nano Lett. 5, 1077–1080 (2005).

    CAS  ADS  Article  Google Scholar 

  2. 2

    Liu, R. & Sen, A. Autonomous nanomotor based on copper-platinum segmented nanobattery. J. Am. Chem. Soc. 133, 20064–20067 (2011).

    CAS  Article  Google Scholar 

  3. 3

    Valov, I. et al. Nanobatteries in redox-based resistive switches require extension of memristor theory. Nat. Commun. 4, 1771 (2013).

    CAS  Article  Google Scholar 

  4. 4

    Tarn, D. et al. Mesoporous silica nanoparticle nanocarriers: biofunctionality and biocompatibility. Acc. Chem. Res. 46, 792–801 (2013).

    CAS  Article  Google Scholar 

  5. 5

    Kline, T. & Paxton, W. Catalytic nanomotors: remote-controlled autonomous movement of striped metallic nanorods. Angew. Chem. Int. Ed. 117, 754–756 (2005).

    Article  Google Scholar 

  6. 6

    Akyildiz, I. F., Brunetti, F. & Blázquez, C. Nanonetworks: a new communication paradigm. Comput. Netw. 52, 2260–2279 (2008).

    Article  Google Scholar 

  7. 7

    Suda, T., Moore, M., Nakano, T., Egashira, R. & Enomoto, A. Exploratory research on molecular communication between nanomachines. Nat. Comput. 25, 1–30 (2005).

    Google Scholar 

  8. 8

    Malak, D. & Akan, O. B. Molecular communication nanonetworks inside human body. Nano Commun. Netw. 3, 19–35 (2012).

    Article  Google Scholar 

  9. 9

    Akyildiz, I. F., Jornet, J. M. & Pierobon, M. Nanonetworks: a new frontier in communications. Commun. ACM 54, 84–89 (2011).

    Article  Google Scholar 

  10. 10

    Nakano, T., Moore, M. J., Wei, F., Vasilakos, A. V. & Shuai, J. Molecular communication and networking: opportunities and challenges. IEEE Trans. Nanobiosci. 11, 135–148 (2012).

    Article  Google Scholar 

  11. 11

    Waters, C. M. & Bassler, B. L. Quorum sensing: cell-to-cell communication in bacteria. Annu. Rev. Cell Dev. Biol. 21, 319–346 (2005).

    CAS  Article  Google Scholar 

  12. 12

    Dickschat, J. S. Quorum sensing and bacterial biofilms. Nat. Prod. Rep. 27, 343–369 (2010).

    CAS  Article  Google Scholar 

  13. 13

    Kerényi, Á., Bihary, D., Venturi, V. & Pongor, S. Stability of multispecies bacterial communities: signaling networks may stabilize microbiomes. PLoS ONE 8, e57947 (2013).

    ADS  Article  Google Scholar 

  14. 14

    Gotti, C. & Clementi, F. Neuronal nicotinic receptors: from structure to pathology. Prog. Neurobiol. 74, 363–396 (2004).

    CAS  Article  Google Scholar 

  15. 15

    Betke, K. M., Wells, C. A. & Hamm, H. E. GPCR mediated regulation of synaptic transmission. Prog. Neurobiol. 96, 304–321 (2012).

    CAS  Article  Google Scholar 

  16. 16

    Qian, L., Winfree, E. & Bruck, J. Neural network computation with DNA strand displacement cascades. Nature 475, 368–372 (2011).

    CAS  Article  Google Scholar 

  17. 17

    Benenson, Y. Biomolecular computing systems: principles, progress and potential. Nat. Rev. Genet. 13, 455–468 (2012).

    CAS  Article  Google Scholar 

  18. 18

    Ball, P. Chemistry meets computing. Nature 406, 118–120 (2000).

    CAS  Article  Google Scholar 

  19. 19

    de Silva, A. P. & McClenaghan, N. D. Molecular-Scale Logic Gates. Chem. Eur. J. 10, 574–586 (2004).

    Article  Google Scholar 

  20. 20

    Condon, A. Automata make antisense. Nature 429, 351–352 (2004).

    CAS  ADS  Article  Google Scholar 

  21. 21

    Seelig, G., Soloveichik, D., Zhang, D. Y. & Winfree, E. Enzyme-free nucleic acid logic circuits. Science 314, 1585–1588 (2006).

    CAS  ADS  Article  Google Scholar 

  22. 22

    Douglas, S. M., Bachelet, I. & Church, G. M. A logic-gated nanorobot for targeted transport of molecular payloads. Science 335, 831–834 (2012).

    CAS  ADS  Article  Google Scholar 

  23. 23

    Angelos, S., Yang, Y. W., Khashab, N. M., Stoddart, J. F. & Zink, J. I. Dual-controlled nanoparticles exhibiting AND logic. J. Am. Chem. Soc. 131, 11344–11346 (2009).

    CAS  Article  Google Scholar 

  24. 24

    Liu, H. et al. Dual-responsive surfaces modified with phenylboronic acid-containing polymer brush to reversibly capture and release cancer cells. J. Am. Chem. Soc. 135, 7603–7609 (2013).

    CAS  Article  Google Scholar 

  25. 25

    Lee, J. W. & Klajn, R. Dual-responsive nanoparticles that aggregate under the simultaneous action of light and CO2 . Chem. Commun. 51, 2036–2039 (2015).

    CAS  Article  Google Scholar 

  26. 26

    Liu, D. et al. Resettable, multi-readout logic gates based on controllably reversible aggregation of gold nanoparticles. Angew. Chem. Int. Ed. 50, 4103–4107 (2011).

    CAS  Article  Google Scholar 

  27. 27

    Chitode, J. S. Communication Theory Technical Publications (2010).

  28. 28

    Wood, J. T. Communication in Our Lives Wadsworth (2009).

  29. 29

    Guardado-Alvarez, T. M., Sudha Devi, L., Russell, M. M., Schwartz, B. J. & Zink, J. I. Activation of snap-top capped mesoporous silica nanocontainers using two near-infrared photons. J. Am. Chem. Soc. 135, 14000–14003 (2013).

    CAS  Article  Google Scholar 

  30. 30

    Baeza, A., Guisasola, E., Ruiz-Hernández, E. & Vallet-Regí, M. Magnetically triggered multidrug release by hybrid mesoporous silica nanoparticles. Chem. Mater. 24, 517–524 (2012).

    CAS  Article  Google Scholar 

  31. 31

    Zhang, Z. et al. Biocatalytic release of an anticancer drug from nucleic-acids-capped mesoporous SiO2 using DNA or molecular biomarkers as triggering stimuli. ACS Nano 7, 8455–8468 (2013).

    CAS  Article  Google Scholar 

  32. 32

    Tang, F., Li, L. & Chen, D. Mesoporous silica nanoparticles: synthesis, biocompatibility and drug delivery. Adv. Mater. 24, 1504–1534 (2012).

    CAS  Article  Google Scholar 

  33. 33

    Li, Z., Barnes, J. C., Bosoy, A., Stoddart, J. F. & Zink, J. I. Mesoporous silica nanoparticles in biomedical applications. Chem. Soc. Rev. 41, 2590–2605 (2012).

    CAS  Article  Google Scholar 

  34. 34

    Coll, C., Bernardos, A., Martínez-Máñez, R. & Sancenón, F. Gated silica mesoporous supports for controlled release and signaling applications. Acc. Chem. Res. 46, 339–349 (2013).

    CAS  Article  Google Scholar 

  35. 35

    Aznar, E. et al. Gated materials for on-command release of guest molecules. Chem. Rev. 116, 561–718 (2016).

    CAS  Article  Google Scholar 

  36. 36

    Díez, P. et al. Toward the design of smart delivery systems controlled by integrated enzyme-based biocomputing ensembles. J. Am. Chem. Soc. 136, 9116–9123 (2014).

    Article  Google Scholar 

  37. 37

    Villalonga, R. et al. Enzyme-controlled sensing-actuating nanomachine based on Janus Au-mesoporous silica nanoparticles. Chem. Eur. J. 19, 7889–7894 (2013).

    CAS  Article  Google Scholar 

  38. 38

    Jerez, G., Kaufman, G., Prystai, M., Schenkeveld, S. & Donkor, K. K. Determination of thermodynamic pKa values of benzimidazole and benzimidazole derivatives by capillary electrophoresis. J. Sep. Sci. 32, 1087–1095 (2009).

    CAS  Article  Google Scholar 

  39. 39

    Sheffner, A. L. The reduction in vitro in viscosity of mucoprotein solutions by a new mucolytic agent, N-acetyl-L-cysteine. Ann. N. Y. Acad. Sci. 106, 298–310 (1963).

    CAS  ADS  Article  Google Scholar 

  40. 40

    Turkevich, J., Stevenson, P. C. & Hillier, J. A study of the nucleation and growth processes in the synthesis of colloidal gold. Discuss. Faraday Soc. 11, 55–75 (1951).

    Article  Google Scholar 

  41. 41

    Frens, G. Controlled Nucleation for the Regulation of the Particle Size in Monodisperse Gold Suspensions. Nature 241, 20–22 (1973).

    CAS  ADS  Google Scholar 

  42. 42

    Yousef, F. O., Zughul, M. B. & Badwan, A. A. The modes of complexation of benzimidazole with aqueous β-cyclodextrin explored by phase solubility, potentiometric titration, 1H-NMR and molecular modeling studies. J. Incl. Phenom. Macrocycl. Chem. 57, 519–523 (2007).

    CAS  Article  Google Scholar 

  43. 43

    Sánchez, A., Díez, P., Martínez-Ruíz, P., Villalonga, R. & Pingarrón, J. M. Janus Au-mesoporous silica nanoparticles as electrochemical biorecognition-signaling system. Electrochem. Commun. 30, 51–54 (2013).

    Article  Google Scholar 

  44. 44

    Akyildiz, I. F., Pierobon, M., Balasubramaniam, S. & Koucheryavy, Y. The internet of Bio-Nano things. IEEE Commun. Mag. 53, 32–40 (2015).

    Article  Google Scholar 

  45. 45

    Sancenón, F., Pascual, L., Oroval, M., Aznar, E. & Martínez-Máñez, R. Gated silica mesoporous materials in sensing applications. ChemistryOpen 4, 418–437 (2015).

    Article  Google Scholar 

  46. 46

    Akyildiz, I. & Jornet, J. The Internet of nano-things. IEEE Wirel. Commun. 17, 58–63 (2010).

    Article  Google Scholar 

  47. 47

    Giménez, C. et al. Towards chemical communication between gated nanoparticles. Angew. Chem. Int. Ed. 53, 12629–12633 (2014).

    Google Scholar 

  48. 48

    Davis, B. G., Lloyd, R. C. & Jones, J. B. Controlled site-selective glycosylation of proteins by a combined site-directed mutagenesis and chemical modification approach. J. Org. Chem. 63, 9614–9615 (1998).

    CAS  Article  Google Scholar 

Download references


A.L.-L. is grateful to ‘La Caixa’ Banking Foundation for his PhD fellowship. We wish to thank the Spanish Government (MINECO Projects MAT2015-64139-C4-1, CTQ2014-58989-P and CTQ2015-71936-REDT and AGL2015-70235-C2-2-R) and the Generalitat Valenciana (Project PROMETEOII/2014/047) for support. The Comunidad de Madrid (S2013/MIT-3029, Programme NANOAVANSENS) is also gratefully acknowledged.

Author information




A.L.-L. and P.D. performed the experiments. A.L.-L., P.D., A.S., R.V. and R.M.-M. designed and conceived the experiments. A.L.-L. and R.M.-M. wrote the manuscript. F.S. revised and helped in the elaboration of the manuscript. M.D.M. and P.M.-R. helped in useful discussions and technical issues. All authors discussed the results and commented on the manuscript.

Corresponding authors

Correspondence to Reynaldo Villalonga or Ramón Martínez-Máñez.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Information

Supplementary Figures, Supplementary Table, Supplementary Methods and Supplementary References. (PDF 844 kb)

Peer Review File (PDF 270 kb)

Rights and permissions

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Llopis-Lorente, A., Díez, P., Sánchez, A. et al. Interactive models of communication at the nanoscale using nanoparticles that talk to one another. Nat Commun 8, 15511 (2017).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing