Ultrashort Single-Walled Carbon Nanotubes Insert into a Pulmonary

Article pubs.acs.org/JPCB

Ultrashort Single-Walled Carbon Nanotubes Insert into a Pulmonary Surfactant Monolayer via Self-Rotation: Poration and Mechanical Inhibition Tongtao Yue,*,†,‡ Yan Xu,‡ Shixin Li,‡ Zhen Luo,‡ Xianren Zhang,§ and Fang Huang†,‡ †

State Key Laboratory of Heavy Oil Processing and ‡Center for Bioengineering and Biotechnology, China University of Petroleum (East China), Qingdao 266580, China § State Key Laboratory of Organic-Inorganic Composites, Beijing University of Chemical Technology, Beijing 100029, China S Supporting Information *

ABSTRACT: It has been widely accepted that longer single-walled carbon nanotubes (SWCNTs) exhibit higher toxicity by causing severe pneumonia once inhaled, yet relatively little is known regarding the potential toxicity of ultrashort SWCNTs, which are of central importance to the development of suitable vehicles for biomedical applications. Here, by combining coarse-grained molecular dynamics (CGMD), pulling simulations, and scaling analysis, we demonstrate that the inhalation toxicity of ultrashort SWCNTs (1.5 nm < l < 5.5 nm) can be derived from the unique behaviors on interaction with the pulmonary surfactant monolayer (PSM), which is located at the air−water interface of alveoli and forms the frontline of the lung host defense. Molecular dynamics (MD) simulations suggest that ultrashort SWCNTs spontaneously insert into the PSM via fast self-rotation. Further translocation toward the water or air phase involves overcoming a high free-energy barrier, indicating that removal of inhaled ultrashort SWCNTs from the PSM is difficult, possibly leading to the accumulation of SWCNTs in the PSM, with prolonged retention and increased inflammation potentials. Under certain conditions, the inserted SWCNTs are found to open hydrophilic pores in the PSM via a mechanism that mimics that of the antimicrobial peptide. Besides, the mechanical property of the PSM is inhibited by the deposited ultrashort SWCNTs through segregation of the inner lipid molecules from the outer phase. Our results bring to the forefront the concern of the inhalation toxicity of ultrashort SWCNTs and provide guidelines for future design of inhaled nanodrug carriers with minimized side effects.

■

INTRODUCTION Single-walled carbon nanotubes (SWCNTs) are unique nanomaterials that have inimitable physicochemical properties, which suggest possible applications in different areas of science and technology.1 For example, it has been proposed that SWCNTs might be used as agents for cancer therapy2 and as vehicles for targeted drug delivery.3,4 To both promote safe biomedical diagnostics and therapies and reduce the health and environmental impacts of SWCNTs, we need to precisely control their fate once inside the body.5 Therefore, a systematic understanding of the detailed interactions between SWCNTs and a variety of biomolecules is urgently required.6 Among the possible portals, the lung provides a large surface area for inhaled SWCNT deposition.7 Because of the small size, a large portion of the inhaled SWCNTs can pass through the airway and reach the deep gas-exchange surface in alveoli, which is covered by a thin pulmonary surfactant monolayer (PSM). The PSM is a detergent-like substance composed of approximately 90% lipids, mostly phospholipids, and 10% proteins.8,9 The major function of the PSM is maintaining a low surface © 2017 American Chemical Society

tension at the alveolar air−water interface and modulating the innate immune defense.10 Therefore, the detailed interaction of the SWCNTs with the PSM critically influences their subsequent fate once inhaled and is an important consideration in pulmonary nanodrug delivery and diagnosis. Besides, both the ultrastructure and mechanical property of the PSM may be perturbed on interacting with SWCNTs, although this is likely to depend on many factors, like SWCNT size and concentration and PSM tension. It has been experimentally evidenced that longer SWCNTs display a higher toxicity than shorter ones.11,12 For example, long SWCNTs were found to undergo incomplete or frustrated phagocytosis,13 which might cause prolonged generation of reactive oxygen species, leading to the release of inflammatory mediators, cell death,14 and DNA and chromosomal damage in target cells of the lung.15 Comparatively, ultrashort SWCNTs, Received: January 11, 2017 Revised: March 13, 2017 Published: March 14, 2017 2797

DOI: 10.1021/acs.jpcb.7b00297 J. Phys. Chem. B 2017, 121, 2797−2807

Article

The Journal of Physical Chemistry B

Figure 1. Simulated first attack event of one ultrashort SWCNT on a PSM. (a−f) Time sequence of typical snapshots showing the detailed pathway of PSM insertion of the SWCNT; (g) time evolution of the interaction energy between the SWCNT and lipids; (h) time evolution of the COM position of SWCNT along the monolayer normal direction; (i) schematic diagram showing the entry angle between the tube radial surface and PSM plane; and (j) time evolution of the SWCNT orientation during the insertion process. The tube diameter and length were set to 5.3 and 1.8 nm, respectively. The PSM tension was fixed at 30 mN m−1.

with lengths ranging from 2 to 16 nm, are of central importance to the development of suitable vehicles for therapeutic and biosensing applications.16,17 Apart from the optimal balance between drug loading and drug release, more importantly,

ultrashort SWCNTs can bypass the macrophages and directly interact with the alveolar epithelium to play biomedical roles.18,19 Although ultrashort SWCNTs are advantageous in biomedical applications, to the best of our knowledge, relatively little is 2798


Article


Figure 2. SWCNT−PSM interaction under compression: (a−c) Three typical snapshots representing the different interaction states: horizontal adhesion, tube distortion, and vertical insertion; (d) time evolution of the interaction energy between the SWCNT and lipid molecules; and (e) time evolution of tube orientation during the insertion process. The PSM surface tension was fixed at 0 mN m−1.

glycerol (PG), and cholesterol.22 Four kinds of proteins, including SP-A, SP-B, SP-C, and SP-D, are associated with the PSM.23,24 As the simplest model of PSM, a DPPC monolayer provides an ideal model to easily explore both surface activity and its interactions with inhaled nanoparticles (NPs).25−31 Thus, our simulation setup consists of two symmetric DPPC monolayers confining a water slab (Figure S1). To model different conditions of respiration, several DPPC monolayers were constructed and equilibrated under different values of surface tension. SWCNT Model. In the coarse-grain (CG) model, the honeycomb atomic structure of the SWCNTs was reduced to a triangular lattice of SC4-type beads (SC4 is a ring-type hydrophobic parameter in the Martini force field), where every three carbons in the all-atom SWCNTs are modeled as one particle. The angular force constant and equilibrium angle along the axial direction are κangle = 700 kcal mol−1rad−2 and θ = 60°, respectively. Along the radial direction, the equilibrium angle is not constant but is determined by the tube diameter. The bond force constant and equilibrium bond length are κbond = 700 kcal mol−1 and l = 0.3684 nm, respectively. The dihedral angle force constant, multiplicity, and equilibrium angle are κχ = 3.1 kcal mol−1, n = 2, and δ = 180°, respectively. Here, we fixed the SWCNT diameter, d, to 5.3 nm and varied the tube length, l, from 1.8 to 5.5 nm (Figure S1). The specific CG model comes from the previous simulation work of Titov et al. and has been

known about the inhalation toxicity of ultrashort SWCNTs, especially their perturbation on the PSM. The urgency to develop guidelines for regulating their occupational and environmental exposure,20 as well as for manufacturing safer biomedical diagnostics and therapeutics,21 justifies the efforts to clarify the mechanism of interaction of the PSM with inhaled ultrashort SWCNTs. Limited by the available experimental technologies, it is still difficult to directly probe and visualize the whole process of interaction. A high resolution can be achieved by computer simulations, which, over the past decades, have substantially increased the accessible length and time scales. In this work, we combine coarse-grained molecular dynamics (CGMD), pulling simulations, and scaling analysis to systematically investigate the interactions between inhaled ultrashort SWCNTs and the PSM. We mainly focus on the perturbation of ultrashort SWCNTs on PSM and relate it to the potential inhalation toxicity.

■

MODEL AND SIMULATION METHOD PSM Model. The composition of the real PSM is quite complex. Besides lipids, which account for 90% of the surfactant monolayer, proteins make up the other 10% of the surfactant monolayer. Among the lipids, the most abundant components are dipalmitoylphosphatidylcholine (DPPC) along with a few other lipids, namely, phosphatidylcholine (PC), phosphatidyl2799


Article


Figure 3. Passive translocation of inhaled ultrashort SWCNTs across the PSM: (a) Time evolution of pulling force revealing spontaneous insertion and restrained further translocation; (b) time evolution of the orientation angle of the SWCNT revealing the staged self-rotation; and (c) typical snapshots during the translocation process. The inset figures are the enlarged evolutions highlighting the short retention of the SWCNT horizontally adhering on the PSM.

to be 5 × 10−5 bar−1 in the x−y plane and 0 bar−1 in the zdirection. The neighboring list for nonbonded interactions was updated every 10 steps. Periodic boundary conditions were imposed in all three dimensions of the simulation box.

successfully used to simulate interactions of graphene nanosheets with the lipid bilayer.32 MD Simulations. CGMD simulations were performed using the GROMACS software package,33 and the Martini force field was used for all simulations.34 Here, each simulation started with two ultrashort SWCNTs placed in vacuum above and below the pre-equilibrated PSM consisting of 2 × 2268 DPPC lipid molecules. After energy minimization, CGMD simulations with constant particle number, surface tension, and temperature were carried out. For each simulation, a cutoff of 1.2 nm was used for van der Waals interactions. The Lennard-Jones potential was smoothly shifted to zero between 0.9 and 1.2 nm to reduce the cutoff noise. The Coulombic potential, with a cutoff of 1.2 nm, was smoothly shifted to zero from 0 to 1.2 nm. The temperature was kept at 310 K with the Berendsen weak coupling algorithm, with a time constant of 2.5 ps. The Berendsen barostat was applied for constant surface tension. The compressibility was set

■

RESULTS AND DISCUSSION First Attack under Deep Inspiration: Spontaneous Insertion via Fast Self-Rotation. We started our simulations by exploring the first attack of one inhaled ultrashort SWCNT (d = 5.3 nm, l = 1.8 nm) on the PSM under deep inspiration. The PSM tension was fixed at 30 mN m−1 to model the deep inspiration condition. To visualize how the inhaled ultrashort SWCNT attacks the PSM, we provide in Figure 1a−f a typical simulated insertion event of an ultrashort SWCNT into the PSM. A more detailed process can be found in Video S1. The interaction energy between the SWCNT and lipid molecules is given in Figure 1g, which shows that the whole insertion process 2800


Article


Figure 4. Influence of PSM tension on passive translocation of ultrashort SWCNTs across the PSM: (a) Time evolutions of the interaction energy between the SWCNT and lipid molecules, confirming the staged translocation process; (b, c) final snapshots revealing the distinct internalization pathways under different respiration conditions.

This state caused severe tube distortion and a local PSM perturbation around it (Figure 2b, inset). To release the high tube-distorting energy, the SWCNT underwent shape adjustment, which corresponds to a fluctuation of the orientation angle for 10 ns (Figure 2e). At t = 13.5 ns, the composite reached equilibrium, where the SWCNT vertically inserted into the PSM, with no tube distortion or PSM perturbation (Figure 2c). Accordingly, the interaction energy further decreased to the minimized value of −4000 kJ mol−1 (Figure 2d). Vertical PSM Insertion of Ultrashort SWCNTs is Robust. The above simulations have shown that the inhaled ultrashort SWCNTs spontaneously insert into the PSM, regardless of the PSM tension (Figures 1 and 2). More importantly, no further translocation toward the water or air phase was observed for a long time. This behavior indicates that once inhaled, the ultrashort SWCNTs might be trapped in the PSM and become hard to remove or internalize. In reality, this may lead to accumulation of inhaled ultrashort SWCNTs in the PSM, with prolonged retention and increased inflammation potentials. Here, we want to elucidate why vertical PSM insertion is the preferred mode of interaction for ultrashort SWCNTs. To understand this behavior, we performed pulling simulations that cause an inhaled ultrashort SWCNT in vacuum to cross the PSM and reach the water phase. The pulling simulations were based on the fact that the inhaled SWCNTs have nonzero speed and thus may accomplish translocation across the PSM once the membrane resistance is overcome.35 By taking the average inspiratory reserve volume (∼3.0 L), human respiration rate (12−20 breaths per minute), terminal bronchiole diameter (∼0.5 mm), and average bronchiole number (∼60 000) into account, we estimated the inhalation velocity to be about 0.01−0.1 nm ns−1. To accomplish translocation, here an external spring force with a constant of 2000 kJ mol−1 nm−2 was exerted on the COM of the SWCNT to pull it downward and across the PSM with a constant velocity of 0.1 nm ns−1. The time evolutions of pulling force and tube orientation angle are given in Figure 3, from which several typical interaction states were captured and further confirmed. The first state that featured by horizontal alignment of the SWCNT adhering on the PSM corresponded to a sharp increase in the pulling force from 0 to 150 kJ mol−1 nm−1 (insets of Figure 3a). Shortly after, a further

was driven by maximizing the interaction between the SWCNT and lipid molecules. More detailed information about the SWCNT deposition and rotation during this process can be revealed by examining its position along the PSM normal direction and angle between the tube radial surface and the surface parallel to the PSM (Figure 1h−j). We combined the above information and divided the whole interaction process roughly into three successive stages. Before deposition, the SWCNT was not captured by the PSM and thereby vibrated, translated, and rotated freely in vacuum under thermal fluctuation (Figure 1a, Video S1). Once captured by the PSM in a finite tilt angle (Figure 1b), an instantaneous but predominant rotation of the SWCNT toward a horizontal alignment was observed (Figure 1c). This rotation was to temporarily increase the contact between the SWCNT and lipids. However, the horizontal alignment was not stable under the monolayer undulation. Shortly after, a second rotation toward vertical alignment was observed, and the SWCNT inserted itself into the PSM to maximize the interaction with the lipid molecules (Figure 1d,e). In the last stage, the SWCNT kept inserting into the PSM to form a stable composite (Figure 1f). Influence of PSM Tension. External SWCNTs attack the PSM most probably during inspiration. In fact, under different inspiration depths, the tension of the PSM dynamically changes from near zero to tens of mN m−1. Thus, it is of equal importance to explore how the surface tension influences both the pathway and mechanism of the interaction of PSM with SWCNTs. To probe this effect, further CGMD simulations were carried out under a fixed tension of 0 mN m−1. The time sequence of typical snapshots representing distinct interaction states is given in Figure 2a−c. The first state was featured by horizontal alignment of the SWCNT adhering on the PSM surface (Figure 2a), just like that during PSM expansion (Figure 1), but was maintained for a longer time (Video S2). We analyze that it was the ordered lipid packing under compression that helped stabilize the SWCNT horizontally adhering on the PSM surface. After that, a sudden decrease of the interaction energy from −1500 to −3200 kJ mol−1 was observed (Figure 2d). In this stage, the preformed contact between the SWCNT and PSM was temporarily maintained, whereas the free segment bent downward to further increase the contact with the lipids (Figure 2b). 2801


Article


length-dependent.40 For example, increasing the length of the SWCNTs makes them more inflammogenic.11 However, relatively little is known regarding the inhalation toxicity of ultrashort SWCNTs, especially their perturbation on the PSM. The above simulations have indicated that the deposition of ultrashort SWCNTs on the PSM can induce PSM perturbation (Figure 2). Once the composite reaches equilibrium, the steady state of the interaction may induce permanent PSM damage. Figure 5a shows the time evolution of the local order parameter of lipids adjacent to the inserting SWCNT (see the

increase in the positive pulling force and decrease in the orientation angle reveal that the PSM insertion via tube rotation is an energetically favorable process (Figure 3a,b). After insertion completed, further translocation led to a roughly linear decrease in the pulling force, suggesting an energetically unfavorable process. At t = 125 ns, a reverse increase in the pulling force from −350 to −150 kJ mol−1 nm−1 was observed, and the orientation angle increased from 0 to 90°, suggesting that a self-rotation toward horizontal alignment is required to reduce the free energy and accomplish translocation (Video S3). This behavior can be understood as that of other anisotropic NPs, like graphene nanosheets36−38 and nanorods,39 the translocation of which across a lipid bilayer is accomplished via a corner or tip-first mechanism. In the last stage, the translocated SWCNT detached from the PSM, with a small disklike lipid aggregate encapsulated in the SWCNT (Figure 3c). The pulling force further increased and finally reached and fluctuated around the value of zero (Figure 3a). Note that the fast pulling speed in our simulations could lead to nonequilibrium rotational dynamics of the SWCNT, thus locking it at random high-energy orientations. To further prove the importance of self-rotation in both insertion and translocation, we performed two more pulling simulations, in which the SWCNTs were restrained from rotating and translocated across the PSM vertically and horizontally. The time evolutions of the pulling forces and typical snapshots are given in Figure S2. Once captured by the PSM that corresponds to a sudden increase in the pulling force, the vertical insertion was found to generate a larger positive pulling force, suggesting that the vertical insertion into the PSM is more energetically favorable. After insertion, the subsequent translocation of the vertical SWCNT led to a more striking decrease in the pulling force, whereas that of the horizontal SWCNT generated a mild decrease. This result further confirmed that the inhaled ultrashort SWCNTs insert into the PSM via fast rotation to vertical alignment, whereas subsequent translocations require further rotation back to the horizontal alignment (Figure 3) To probe how the PSM tension influences the PSM translocation of ultrashort SWCNTs, we repeated the pulling simulation under a fixed tension of 0 mN m−1. The time evolutions of the interaction energy between the SWCNT and lipids are given in Figure 4, from which two major differences were identified. First, the state of horizontal alignment of the SWCNT adhering on the PSM was maintained for a much longer time under PSM compression (Figure 4a and Video S4), confirming the ordered lipid packing under compression that helped stabilize the horizontal alignment of the SWCNT (Figure 2). The second difference was in the stage of final translocation. Under expansion (30 mN m−1), the PSM translocation of the SWCNT was accompanied by a continuous increase in the interaction energy. The final value of −2000 kJ mol−1 was derived from the disklike aggregate of lipid molecules that was encapsulated inside the tube (Figure 4b). Comparatively, under compression, the entry of the SWCNT is more energetically favorable. After a short energy increase by tube self-rotation, subsequent entry was accomplished via a shape transformation of the PSM from a monolayer to a bilayer, in which the SWCNT was embedded. Thus, we surmise that the pulmonary internalization of ultrashort SWCNTs might be possible only under expiration via both tube self-rotation and PSM shape-transformation. Perturbation of Inserting Ultrashort SWCNTs on the PSM. It is well-known that the cytotoxicity of SWCNTs is

Figure 5. Transient local perturbation of inserting an ultrashort SWCNT on a PSM: (a) Time evolution of the order parameter of lipids adjacent to the inserting SWCNT and (b) transient distribution of the lipid order parameter along the Y direction at t = 4.0 ns. The inset of (a) illustrates the calculation of the lipid order parameter. The inset of (b) is the snapshot illustrating the detailed lipid arrangement around the SWCNT. The PSM surface tension was fixed at 0 mN m−1. The tube diameter and length were 5.3 and 1.8 nm, respectively.

inset for detailed calculations of the order parameter). The first interaction state that featured by horizontal alignment of the SWCNT and a trivial PSM perturbation is reflected by a slight decrease in the lipid order parameter at t = 2.5 ns (Figure 5a). As the contact was increased by tube distortion (Figure 2d,e), a further decrease in the order parameter was observed (Figure 5a). On a molecular level, adjacent lipid molecules underwent rearrangement around the SWCNT to adapt to its distortion (Figure 5b). As the simulation proceeded, the lipid rearrangement rehealed and the SWCNT gradually recovered to a deformed cylinder and vertically inserted into the monolayer with minor distortion and PSM perturbation (Figure 2c, Video S2). Once the composite reached equilibrium, it seemed that the PSM recovered to its natural form and displayed normal functions (Figure 2). However, further simulations suggested that there exists novel PSM damage underlying the mild SWCNT−PSM interaction. Here, we fixed the tube diameter at 5.3 nm and increased the tube length from 1.8 to 5.4 nm. To model different respiration conditions, we varied the PSM tension from 0 to 40 mN m−1. The final steady states of interaction under different conditions are summarized in the phase diagram (Figure 6a). Interestingly, under certain conditions the inserting SWCNTs induced lipid rearrangement 2802


Article


Figure 6. Influence of SWCNT length and PSM tension on the final interaction configuration: (a) Simulated phase diagram as a function of tube length and PSM tension, summarizing the conditions under which PSM undergoes poration inside the tube: the filled blue circle represents no PSM poration, the open red circle represents PSM poration, and the green circled plus represents two monolayers with one opening pore inside the tube and the other not inside the tube and (b) Normalized local tension inside the tube calculated by the scaling analysis as a function of tube length and PSM tension. The right panel of (a) provides two typical snapshots of the PSM−SWCNT composite with and without poration.

embedded in the lipid tails, leaving the bottom segment immerged in water (Figure S3c). To reduce unfavorable water exposure, inner lipid molecules were partially extracted from the monolayer and covered the inner tube surface to protect it from water.44 The longer the SWCNT was, the more the lipids extracted. Interestingly, no lipids were found to be extracted to cover the outer tube surface (Figure S3c), confirming the preferential lipid extraction for a concave surface and weaker or no extraction for a convex surface.45 Collectively, both tight lipid packing and lipid extraction caused severe deficiency of lipids inside the tube, thus increasing the inner monolayer tension. Once the local tension exceeded a critical value, the monolayer would rupture and open a hydrophilic pore inside the tube to release the local high surface energy. We would like to note that the porelike structure was quite stable during our simulations. Even under further PSM compression, no pore closure was observed (Video S5), indicating that the destructive PSM poration on insertion of SWCNTs might be lasting. In reality, the PSM poration may cause efflux of alveolar contents, finally leading to respiratory failure. We thus relate this behavior to that of antimicrobial

inside the tube, thus opening hydrophilic pores in the PSM. Besides, the PSM poration is sensitive to both the SWCNT length and PSM tension. For example, no pore was generated under the current range of tube length if the PSM was supercompressed (e.g., γ < 10 mN m−1). Under a higher PSM tension, longer SWCNTs can easily generate pores, whereas shorter ones simply insert into the PSM with no poration. Finally, under the expanded phase (e.g., γ = 40 mN m−1), a slight increase of the tube length could induce a remarkable PSM perturbation via opening water pores. We analyze the molecular mechanism of PSM poration by inserting ultrashort SWCNTs as follows. Both top and crosssectional views of the final configurations illustrate the distinct lipid arrangements inside and outside the tube (Figure S3a−c). It has been demonstrated that inserting ultrashort SWCNTs into a lipid bilayer can “strengthen” ambient lipids, thus generating periodic spatial distributions of lipid properties as a function of distance from the SWCNT.41−43 Similarly, for the monolayer system, we observed a tighter packing of ambient lipids around the vertically inserted SWCNTs, as indicated in the calculated radial distribution functions (RDFs) (Figure S3d−f). For SWCNTs longer than the PSM thickness, the top segment was 2803


Article


Figure 7. Mechanical inhibition of the PSM by an ultrashort SWCNT: (a) The compression−expansion isotherm of the PSM in the absence of SWCNT; (b) the compression−expansion isotherm of PSM in the presence of an SWCNT; (c) local conformations before and after PSM compression, revealing the indirect compression of inner lipids via SWCNT lateral shrinking under outer compression; and (d) local conformations before and after PSM compression, revealing the distinct lipid arrangement inside and outside the tube.

and tube diameter. Fourth, the number of ambient lipids that tightly pack on the inner tube surface can be estimated via locally integrating the calculated RDFs (Figure S3). The detailed implementation of the analysis process is given in the Supporting Information. Finally, the normalized monolayer tension inside SWCNT, calculated as a function of tube length and PSM tension, is given in Figure 6b. Clearly, it is the bulk PSM tension and SWCNT length that collectively determine whether PSM poration could take place. Expectedly, a higher surface tension destabilizes the monolayer inside the SWCNT and promotes its poration. Under fixed tension, longer SWCNTs further increase the inner monolayer tension by extracting more lipids, thus easily

peptides, which are thought to embed into the cell membrane and generate porelike structures that kill the target.46−48 Scaling Analysis. To establish a plausible physical mechanism for PSM poration by insertion of ultrashort SWCNTs, we present a scaling analysis, which is based on four simple facts or hypotheses. First, and most important, ultrashort SWCNTs vertically insert into the PSM and increase the local monolayer tension, which provides the driving force for PSM poration. Second, the tension of the small monolayer patch inside the tube roughly follows a linear relation with the effective lipid area.26 Third, the inner lipids are segregated from the outer phase by inserting SWCNTs, meaning that the total lipid number inside the tube can be approximated by the bulk PSM tension 2804


Article


along the radial direction of the SWCNTs can be tuned, for example, by varying the tube size,53 surface roughness or defects,54 and chirality.55 We thus highlight the crucial role of the intrinsic mechanical strength of ultrashort SWCNTs on inhibition of PSM biofunction.

opening pores inside them. Our calculations are well consistent with the simulated diagram in Figure 6a. Compression−Expansion Isotherm Indicates Mechanical Inhibition of the PSM on Insertion of Ultrashort SWCNTs. Besides the destructive PSM poration on insertion of ultrashort SWCNTs via a mechanism that mimics that of the antimicrobial peptide, the inhalation toxicity of ultrashort SWCNTs is also reflected in the mechanical inhibition of PSM.49 Here, we first calculated the surface tension of the PSM as a function of lipid area to construct a simulated breath cycle. A small hysteresis loop was observed (Figure 7a). Note that a similar hysteresis was also observed in recent experiments.31,50,51 Next, we placed one ultrashort SWCNT with a diameter of 5.3 nm and length of 1.8 nm inserting into the PSM and reconstructed the compression−expansion isotherm to simulate respiration with deposited SWCNTs. Apparently, the mechanical property of the PSM is inhibited by inserting ultrashort SWCNTs, as indicated by the increasing hysteresis area of the compression−expansion loop (Figure 7b). Besides, the mechanical inhibition is mainly reflected in the expansion process, whereas the compression is less affected. To gain a molecular level understanding of the mechanical inhibition of the PSM on insertion of ultrashort SWCNTs, we provide in Figure 7c,d the local conformations of the composite before and after compression and expansion. Once the SWCNT inserted into the PSM under expansion, interestingly, subsequent compression was accompanied by tube buckling (Figure 7c). In other words, the SWCNT was radially deformed under PSM compression, just like the case of tube deformation under compressive, torsional, or bending stress. In fact, SWCNTs should be much softer in the radial direction than along the axial direction. TEM observations suggested that even van der Waals forces can deform two adjacent SWCNTs.52 The direct consequence of tube buckling was that the inner lipid molecules were no longer thoroughly isolated from the outer phase but indirectly compressed via tube shrinkage. This behavior was verified by calculating the distribution of the lipid order parameter after compression. Indeed, no obvious difference in the lipid order parameter was observed between the inside and outside of the tube (Figure S5a). This explains why the compression was less affected by ultrashort SWCNTs in the breath cycle. On the contrary, if the SWCNT inserted into the PSM under compression, subsequent expansion was inhibited, as indicated by the fact that a smaller average lipid area was needed to reach the required PSM tension (Figure 7b). As shown in the typical snapshots, the cylindrical shape of the SWCNT was well maintained during the expansion process (Figure 7d). This is because a cylinder has the largest cross-sectional area compared to that of other shapes. Therefore, inner lipid molecules were well isolated from the outer phase and restricted in expansion. Knowing that the minor effect of ultrashort SWCNTs on PSM compression is ascribed to its buckling under bulk compression, we surmise that if we increase the radial bending rigidity of the SWCNT its inhibition on PSM compression should be enhanced. To test this, we increased the angular force constant, κangle, along the radial direction from 700 to 1800 kcal mol−1 rad−2. Indeed, by comparing the compression isotherms, the restraining effect on PSM compression was enhanced by increasing the buckling resistance of the SWCNTs (Figure S6a). Both typical snapshots and the distribution of the lipid order parameter show that the cylindrical shape of the SWCNT was maintained, thus immunizing inner lipid molecules from bulk compression (Figure S6b,c). In fact, the bending stiffness

■

CONCLUSIONS Ultrashort SWCNTs, with lengths ranging from 2 to 16 nm, are advantageous as suitable vehicles for therapeutic and biosensing applications. Compared to that of longer SWCNTs, which are believed to exhibit a higher toxicity, the potential toxicity of ultrashort SWCNTs, however, has been relatively neglected. Especially, when designing ultrashort SWCNTs as pulmonary nanodrug carriers or if the ultrashort SWCNTs are inhaled accidentally, the first interaction between the SWCNTs and PSM crucially determines both the delivery efficiency and subsequent tendency to reach the secondary organs. Here, we have combined CGMDs, pulling simulations, and a simple scaling analysis to determine the detailed kinetics and fundamental mechanisms for the interaction of the PSM with inhaled ultrashort SWCNTs. Our simulations revealed spontaneous insertion of ultrashort SWCNTs into the PSM via fast self-rotation, regardless of the PSM tension. Further translocation toward either the water or air phase must overcome a high free-energy barrier, indicating that the inhaled ultrashort SWCNTs might be trapped in the PSM and be hard to remove or internalize, possibly leading to their accumulation in PSM and causing permanent lung damage. Under certain conditions, the inserted ultrashort SWCNTs were found to open hydrophilic pores via rearrangement of lipids inside the tube. This behavior mimics that of an antimicrobial peptide, which is thought to embed into the cell membrane and generate a porelike structure that kills the target. After the composite reached equilibrium, further simulations indicated that the mechanical property of the PSM was inhibited by inserting ultrashort SWCNTs. Comparatively, the mechanical inhibition is mainly reflected in the expansion process, whereas the compression is less affected but can be enhanced by increasing the intrinsic mechanical strength of the SWCNTs. In summary, our results demonstrate that the inhalation toxicity of ultrashort SWCNTs should not be neglected and can be reflected in both the ultrastructure perturbation and mechanical inhibition of the PSM. Thus, our simulations can provide practical guidelines for developing strategies to prevent PSM damage induced by inhaled SWCNTs. For example, decreasing the hydrophobicity of the SWCNTs by surface decoration can prevent their retention in the PSM, thus reducing the risk of accumulation and resultant inflammation. The phase diagram provided in Figure 6 also suggests that decreasing both the SWCNT length and PSM tension can effectively prevent the poration of the PSM by the insertion of SWCNTs. The intrinsic mechanical strength of the SWCNTs, which influences the extent to which the mechanical property of the PSM is inhibited, can be tuned in reality by changing both the aspect ratio and chirality of the SWCNTs. When designing SWCNTs for pulmonary drug delivery, our simulations can guide how the properties of the SWCNTs can be modified to reach the required therapeutic efficiency with minimized toxicity. Because the hydrophobic SWCNTs can be trapped in the PSM with no further translocation, decreasing the hydrophobicity of the SWCNTs and increasing the SWCNT length may help overcome this barrier and promote their transport across the PSM to secondary organs and tissues. 2805


Article


■

Translocation of Nanoparticles from the Lung Airspaces to the Body. Nat. Biotechnol. 2010, 28, 1300−1303. (8) Griese, M. Pulmonary Surfactant in Health and Human Lung Diseases: State of the Art. Eur. Respir. J. 1999, 13, 1455−1476. (9) Goerke, J. Pulmonary Surfactant: Functions and Molecular Composition. Biochim. Biophys. Acta, Mol. Basis Dis. 1998, 1408, 79−89. (10) Bachofen, H.; Schürch, S. Alveolar Surface Forces and Lung Architecture. Comp. Biochem. Physiol., Part A: Mol. Integr. Physiol. 2001, 129, 183−193. (11) Bhattacharya, K.; Andon, F. T.; El-Sayed, R.; Fadeel, B. Mechanisms of Carbon Nanotube-induced Toxicity: Focus on Pulmonary Inflammation. Adv. Drug Delivery Rev. 2013, 65, 2087−2097. (12) Kostarelos, K. The Long and Short of Carbon Nanotube Toxicity. Nat. Biotechnol. 2008, 26, 774−776. (13) Shi, X.; von dem Bussche, A.; Hurt, R. H.; Kane, A. B.; Gao, H. Cell Entry of One-Dimensional Nanomaterials Occurs by Tip Recognition and Rotation. Nat. Nanotechnol. 2011, 6, 714−719. (14) Sanchez, V. C.; Pietruska, J. R.; Miselis, N. R.; Hurt, R. H.; Kane, A. B. Biopersistence and Potential Adverse Health Impacts of Fibrous Nanomaterials: What Have We Learned from Asbestos? Wiley Interdiscip. Rev.: Nanomed. Nanobiotechnol. 2009, 1, 511−529. (15) Nagai, H.; Toyokuni, S. Biopersistent Fiber-Induced Inflammation and Carcinogenesis: Lessons Learned from Asbestos toward Safety of Fibrous Nanomaterials. Arch. Biochem. Biophys. 2010, 502, 1−7. (16) Sun, X.; Zaric, S.; Daranciang, D.; Welsher, K.; Lu, Y.; Li, X.; Dai, H. Optical Properties of Ultrashort Semiconducting Single-Walled Carbon Nanotube Capsules Down to Sub-10 nm. J. Am. Chem. Soc. 2008, 130, 6551−6555. (17) Liu, L.; Yang, C.; Zhao, K.; Li, J.; Wu, H.-C. Ultrashort SingleWalled Carbon Nanotubes in A Lipid Bilayer as a New Nanopore Sensor. Nat. Commun. 2013, 4, No. 2989. (18) Crapo, J. D.; Young, S. L.; Fram, E. K.; Pinkerton, K. E.; Barry, B. E.; Crapo, R. O. Morphometric Characteristics of Cells in the Alveolar Region of Mammalian Lungs 1−3. Am. Rev. Respir. Dis. 1983, 128, S42− S46. (19) Ruenraroengsak, P.; Chen, S.; Hu, S.; Melbourne, J.; Sweeney, S.; Thorley, A. J.; Skepper, J. N.; Shaffer, M. S.; Tetley, T. D.; Porter, A. E. Translocation of Functionalized Multi-Walled Carbon Nanotubes across Human Pulmonary Alveolar Epithelium: Dominant Role of Epithelial Type 1 Cells. ACS Nano 2016, 10, 5070−5085. (20) Brown, D. M.; Kinloch, I. A.; Bangert, U.; Windle, A.; Walter, D.; Walker, G.; Scotchford, C.; Donaldson, K.; Stone, V. An in vitro Study of the Potential of Carbon Nanotubes and Nanofibres to Induce Inflammatory Mediators and Frustrated Phagocytosis. Carbon 2007, 45, 1743−1756. (21) Liu, Z.; Tabakman, S.; Welsher, K.; Dai, H. Carbon Nanotubes in Biology and Medicine: in vitro and in vivo Detection, Imaging and Drug Delivery. Nano Res. 2009, 2, 85−120. (22) Lopez-Rodriguez, E.; Pérez-Gil, J. Structure-Function Relationships in Pulmonary Surfactant Membranes: from Biophysics to Therapy. Biochim. Biophys. Acta, Biomembr. 2014, 1838, 1568−1585. (23) McCormack, F. X.; Whitsett, J. A. The Pulmonary Collectins, SPA and SP-D, Orchestrate Innate Immunity in the Lung. J. Clin. Invest. 2002, 109, 707−712. (24) Whitsett, J. A.; Weaver, T. E. Hydrophobic Surfactant Proteins in Lung Function and Disease. N. Engl. J. Med. 2002, 347, 2141−2148. (25) Baoukina, S.; Monticelli, L.; Risselada, H. J.; Marrink, S. J.; Tieleman, D. P. The Molecular Mechanism of Lipid Monolayer Collapse. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 10803−8. (26) Baoukina, S.; Monticelli, L.; Marrink, S. J.; Tieleman, D. P. Pressure-Area Isotherm of a Lipid Monolayer from Molecular Dynamics Simulations. Langmuir 2007, 23, 12617−12623. (27) Lin, X.; Bai, T.; Zuo, Y. Y.; Gu, N. Promote Potential Applications of Nanoparticles as Respiratory Drug Carrier: Insights from Molecular Dynamics Simulations. Nanoscale 2014, 6, 2759−2767. (28) Yue, T.; Xu, Y.; Li, S.; Zhang, X.; Huang, F. Lipid Extraction Mediates Aggregation of Carbon Nanospheres in Pulmonary Surfactant Monolayers. Phys. Chem. Chem. Phys. 2016, 18, 18923−18933.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.7b00297. Scaling analysis method (Figures S1−S6) (PDF) Typical simulated insertion event of an ultrashort SWCNT into a PSM; horizontal alignment of the SWCNT adhering on the PSM surface; self-rotation toward horizontal alignment is required to reduce the free energy and accomplish translocation; the state of horizontal alignment of the SWCNT adhering on the PSM was maintained for a much longer time under PSM compression; destructive PSM poration by insertion of SWCNTs might be lasting (AVI) (AVI) (AVI) (AVI) (AVI)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Tongtao Yue: 0000-0002-8329-167X Xianren Zhang: 0000-0002-8026-9012 Fang Huang: 0000-0001-8735-0350 Author Contributions

T.Y. conceived and designed the project and designed and performed the simulations. The manuscript was written by T.Y. and revised by Y.X., S.L., Z.L., X.Z., and F.H., with contributions from all authors. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (no. 21303269), Science and Technology Major Project of the Shandong Province (2016GSF117033), and the Qingdao Science and Technology Project (no. 16-5-1-73-jch). Simulations were performed at the National Supercomputing Center in Shenzhen.

■

REFERENCES

(1) Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A. Carbon Nanotubes–The Route toward Applications. Science 2002, 297, 787− 792. (2) Kam, N. W. S.; O’Connell, M.; Wisdom, J. A.; Dai, H. Carbon Nanotubes as Multifunctional Biological Transporters and Nearinfrared Agents for Selective Cancer Cell Destruction. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 11600−11605. (3) Liu, Z.; Cai, W.; He, L.; Nakayama, N.; Chen, K.; Sun, X.; Chen, X.; Dai, H. In vivo Biodistribution and Highly Efficient Tumour Targeting of Carbon Nanotubes in Mice. Nat. Nanotechnol. 2007, 2, 47−52. (4) Shi Kam, N. W.; Jessop, T. C.; Wender, P. A.; Dai, H. Nanotube Molecular Transporters: Internalization of Carbon Nanotube-Protein Conjugates into Mammalian Cells. J. Am. Chem. Soc. 2004, 126, 6850− 6851. (5) Xue, X.; Yang, J.-Y.; He, Y.; Wang, L.-R.; Liu, P.; Yu, L.-S.; Bi, G.-H.; Zhu, M.-M.; Liu, Y.-Y.; Xiang, R.-W.; et al. Aggregated Single-Walled Carbon Nanotubes Attenuate the Behavioural and Neurochemical Effects of Methamphetamine in Mice. Nat. Nanotechnol. 2016, 11, 613− 620. (6) Zhao, Y.; Xing, G.; Chai, Z. Nanotoxicology: Are Carbon Nanotubes Safe? Nat. Nanotechnol. 2008, 3, 191−192. (7) Choi, H. S.; Ashitate, Y.; Lee, J. H.; Kim, S. H.; Matsui, A.; Insin, N.; Bawendi, M. G.; Semmler-Behnke, M.; Frangioni, J. V.; Tsuda, A. Rapid 2806


Article

The Journal of Physical Chemistry B (29) Yue, T.; Wang, X.; Zhang, X.; Huang, F. Molecular Modeling of Interaction between Lipid Monolayer and Graphene Nanosheets: Implications for Pulmonary Nanotoxicity and Pulmonary Drug Delivery. RSC Adv. 2015, 5, 30092−30106. (30) Hu, G.; Jiao, B.; Shi, X.; Valle, R. P.; Fan, Q.; Zuo, Y. Y. Physicochemical Properties of Nanoparticles Regulate Translocation across Pulmonary Surfactant Monolayer and Formation of Lipoprotein Corona. ACS Nano 2013, 7, 10525−10533. (31) Hu, Q.; Jiao, B.; Shi, X.; Valle, R. P.; Zuo, Y. Y.; Hu, G. Effects of Graphene Oxide Nanosheets on the Ultrastructure and Biophysical Properties of the Pulmonary Surfactant Film. Nanoscale 2015, 7, 18025−18029. (32) Titov, A. V.; Král, P.; Pearson, R. Sandwiched Graphene− Membrane Superstructures. ACS Nano 2010, 4, 229−234. (33) Hess, B.; Kutzner, C.; Van Der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 435−447. (34) Marrink, S. J.; Risselada, H. J.; Yefimov, S.; Tieleman, D. P.; De Vries, A. H. The MARTINI Force Field: Coarse Grained Model for Biomolecular Simulations. J. Phys. Chem. B 2007, 111, 7812−7824. (35) Arai, N.; Yasuoka, K.; Zeng, X. C. A Vesicle Cell under Collision With a Janus or Homogeneous Nanoparticle: Translocation Dynamics and Late-Stage Morphology. Nanoscale 2013, 5, 9089−9100. (36) Li, Y.; Yuan, H.; von dem Bussche, A.; Creighton, M.; Hurt, R. H.; Kane, A. B.; Gao, H. Graphene Microsheets Enter Cells through Spontaneous Membrane Penetration at Edge Asperities and Corner Sites. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 12295−12300. (37) Guo, R.; Mao, J.; Yan, L.-T. Computer Simulation of Cell Entry of Graphene Nanosheet. Biomaterials 2013, 34, 4296−4301. (38) Mao, J.; Guo, R.; Yan, L.-T. Simulation and Analysis of Cellular Internalization Pathways and Membrane Perturbation for Graphene Nanosheets. Biomaterials 2014, 35, 6069−6077. (39) Yang, K.; Ma, Y.-Q. Computer Simulation of the Translocation of Nanoparticles with Different Shapes across a Lipid Bilayer. Nat. Nanotechnol. 2010, 5, 579−583. (40) Poland, C. A.; Duffin, R.; Kinloch, I.; Maynard, A.; Wallace, W. A.; Seaton, A.; Stone, V.; Brown, S.; MacNee, W.; Donaldson, K. Carbon Nanotubes Introduced into the Abdominal Cavity of Mice Show Asbestos-like Pathogenicity in a Pilot Study. Nat. Nanotechnol. 2008, 3, 423−428. (41) Shityakov, S.; Dandekar, T. Molecular Dynamics Simulation of POPC and POPE Lipid Membrane Bilayers Enforced by an Intercalated Single-Wall Carbon Nanotube. Nano 2011, 6, 19−29. (42) Goose, J. E.; Sansom, M. S. Reduced Lateral Mobility of Lipids and Proteins in Crowded Membranes. PLoS Comput. Biol. 2013, 9, No. e1003033. (43) Garcia-Fandiño, R.; Piñeiro, A.; Trick, J. L.; Sansom, M. S. Lipid Bilayer Membrane Perturbation by Embedded Nanopores: A Simulation Study. ACS Nano 2016, 10, 3693−3701. (44) Tu, Y.; Lv, M.; Xiu, P.; Huynh, T.; Zhang, M.; Castelli, M.; Liu, Z.; Huang, Q.; Fan, C.; Fang, H.; Zhou, R. Destructive Extraction of Phospholipids from Escherichia coli Membranes by Graphene Nanosheets. Nat. Nanotechnol. 2013, 8, 594−601. (45) Luan, B.; Huynh, T.; Zhou, R. Complete Wetting of Graphene by Biological Lipids. Nanoscale 2016, 8, 5750−5754. (46) Illya, G.; Deserno, M. Coarse-Grained Simulation Studies of Peptide-induced Pore Formation. Biophys. J. 2008, 95, 4163−4173. (47) Zhang, M.; Zhao, J.; Zheng, J. Molecular Understanding of a Potential Functional Link between Antimicrobial and Amyloid Peptides. Soft Matter 2014, 10, 7425−7451. (48) Kabelka, I.; Vácha, R. Optimal Conditions for Opening of Membrane Pore by Amphiphilic Peptides. J. Chem. Phys. 2015, 143, No. 243115. (49) Arick, D. Q.; Choi, Y. H.; Kim, H. C.; Won, Y.-Y. Effects of Nanoparticles on the Mechanical Functioning of the Lung. Adv. Colloid Interface Sci. 2015, 225, 218−228. (50) Valle, R. P.; Wu, T.; Zuo, Y. Y. Biophysical Influence of Airborne Carbon Nanomaterials on Natural Pulmonary Surfactant. ACS Nano 2015, 9, 5413−5421.

(51) Valle, R. P.; Huang, C. L.; Loo, J. S.; Zuo, Y. Y. Increasing Hydrophobicity of Nanoparticles Intensifies Lung Surfactant Film Inhibition and Particle Retention. ACS Sustainable Chem. Eng. 2014, 2, 1574−1580. (52) Ruoff, R. S.; Tersoff, J.; Lorents, D. C.; Subramoney, S.; Chan, B. Radial Deformation of Carbon Nanotubes by Van der Waals Forces. Nature 1993, 364, 514−516. (53) Chang, T.; Gao, H. Size-dependent Elastic Properties of a SingleWalled Carbon Nanotube via a Molecular Mechanics Model. J. Mech. Phys. Solids 2003, 51, 1059−1074. (54) Falvo, M. R.; Clary, G. J.; Taylor, R. M.; Chi, V.; Brooks, F. P.; Washburn, S.; Superfine, R. Bending and Buckling of Carbon Nanotubes under Large Strain. Nature 1997, 389, 582−584. (55) Wang, Q. Effective in-plane Stiffness and Bending Rigidity of Armchair and Zigzag Carbon Nanotubes. Int. J. Solids Struct. 2004, 41, 5451−5461.

2807


Ultrashort Single-Walled Carbon Nanotubes Insert into a Pulmonary

Recommend Documents