Stimulus-responsive light-harvesting complexes based on the pillararene-induced co-assembly of β-carotene and chlorophyll

The locations and arrangements of carotenoids at the subcellular level are responsible for their designated functions, which reinforces the necessity of developing methods for constructing carotenoid-based suprastructures beyond the molecular level. Because carotenoids lack the binding sites necessary for controlled interactions, functional structures based on carotenoids are not easily obtained. Here, we show that carotene-based suprastructures were formed via the induction of pillararene through a phase-transfer-mediated host–guest interaction. More importantly, similar to the main component in natural photosynthesis, complexes could be synthesized after chlorophyll was introduced into the carotene-based suprastructure assembly process. Remarkably, compared with molecular carotene or chlorophyll, this synthesized suprastructure exhibits some photocatalytic activity when exposed to light, which can be exploited for photocatalytic reaction studies of energy capture and solar conversion in living organisms.


Supplementary Discussion
Size matching plays an important role in WP5 and -CAR complexation. The molecular size in the energy minimization fits well within the cavity of WP5. Considering the maximum length of the extended bulky terpene is slightly larger than the width of the cavity, -CAR might exhibit the twisted conformation in the cavity. Importantly, during the construction process, the water/ethanol binary solvents increase the solubility of -CAR, which, in turn, facilitates the hostguest interaction between WP5 and -CAR. The formation of the WP5-CAR complex is supported by the 1 H-NMR spectra, which show that the binding site was in the terpene section of -CAR based on the hydrophobic interactions. In contrast to the chemical shift of the bonded terpene caused by solubilization and host-guest interactions, there was no obvious signal of unbounded terpene that might be caused by the obstruction of movement in the binary solvent.
Thus, the stoichiometry should be determined through further characterization. The association constant of WP5 and -CAR was obtained using a non-linear curve-fitting method and fluorescence titration experiments. As shown in Supplementary Fig. 3g, upon addition of -CAR, the fluorescence intensity of WP5 (monitored at 653 nm) was gradually quenched, and the association constant of WP5⊃-CAR was calculated to be (2.34 ± 2.06) × 10 5 M −1 ( Supplementary   Fig. 3j).
However, the obstruction of the movement of Chl-b caused by WP5 is weaker than that of-CAR in WP5-CAR complexation, which could also be demonstrated by the similar peak intensities before and after complexation. which not only showed that Chl-b could participate in the construction of small sphere walls but also indicated that the insertion of Chl-b did not destroy the hierarchical structure. Supplementary   Fig. 18b shows the EDX mapping, which further demonstrated the existence of Chl-b in HMS with diameters of 400 nm. Supplementary Fig. 18c shows the EDX mapping of an LHC-b with a diameter of 900 nm, which further demonstrated the existence of Chl-b in hollow spheres with larger diameters; the black region in the image was caused by damage to the internal structure of hollow sphere. Supplementary Fig. 18d shows the EDX mapping of an LHC-b with a diameter of 1000 nm, which not only demonstrated the existence of Chl-b in HMS with larger diameters but also demonstrated the more homogeneous wall thickness. Interestingly, for the LHC-b with diameters larger than 1500 nm ( Supplementary Fig. 18e), a fusion phenomenon was observed, which provided the possibility of integration and growth.
In order to calculate the -CAR/Chl-b ratio in a single LHC-b and further analyse these data, a series of LHCs-b were selected randomly. Subsequently, the atomic percentage of Na (1 -CAR ~ 1 WP5 ~ 10 Na) and Mg (1 Chl-b ~ 1 Mg) in LHCs-b was chosen to be investigated ( Supplementary Fig. 19). It should be noted that the light elements 3 C, N, and O with an atomic number less than 10 result in their peaks areas overlapping with each other at a large scale, which makes it difficult to distinguish the characteristic peaks. Thus, Na and Mg (with atomic numbers more than 10) were selected as the characteristic elements of WP5 and Chl-b, respectively, for The generation of an HMS must be associated with the structural characteristics of the WCC; when either WP5 or -CAR was removed from the water or ethanol, no HMS formation was observed ( Supplementary Fig. 22 a-b). In addition, a model compound M (Supplementary Fig. 22 c) was used to prove the importance of the host-guest interactions in HMS construction; however, no notable hierarchical nanostructure ( Supplementary Fig. 22 d) was observed when the -CAR was added to the solution of M. These phenomena indicated that the host−guest-based WCC complexation plays an essential role in the formation of HMS, mainly driven by hydrophobic interactions and hydrogen bonds. The above results demonstrate that hydrogen bonding (CH· · · · · ), CH· · · · · interactions, hydrophobic interactions, and − stacking interactions reinforce the complex stability.
An HMS exhibits stability against the external environment. No structural damage was observed after the addition of magnesium chloride, copper dichloride, and glutamic acid ( Supplementary Fig. 23). An HMS exhibits stability over time. Supplementary Fig. 24 a-c shows optical microscopy images of WCC-based HMSs on glass slides over the course of more than six months. The preservation of the hollow microspherical morphology demonstrates the stability of the HMSs on the supramolecular level. In addition, the stability beyond the molecular level of HMSs is also supported by in situ polarized microscopy observations. The Maltese cross observed in Supplementary Fig. 24 d-f provides another important piece of evidence for the existence of WP5 and -CAR building blocks based on ordered arrays. More importantly, the HMS wall remains orange in colour, implying the stability of -CAR on the molecular level. The in situ Raman peaks at 1529 cm -1 , 1528 cm -1 , and 1528 cm -1 , which correspond to the surface information regarding the HMSs in a, b, and c, suggest that the stability to light and oxygen species was increased in favour of the protection of WP5 in the HMS ( Supplementary Fig. 24 g-i).
The HMS were observed to display characteristics that differ from those of traditional synthetic molecule-based suprastructures. As shown in Supplementary Fig. 25, two small spheres can attach to each other spontaneously, resulting in what could be considered the intermediate of the fusion process. The interactions between two HMSs might be ascribed to the conformational change of the unsaturated carbon chain in -CAR, which leads to the fluidity between the layers and further promotes the growth of microspheres. Furthermore, the peak in the SAXS profile is weak as a consequence of the low concentration of HMSs (the final concentration of WP5 and -CAR is 75 M) and the power limitation of the SAXS instrument (30 W). Thus, to overcome these disadvantages and further obtain convincing data, the acquisition time for detection was increased.
As shown in Supplementary Fig. 26, the intensity of the peak increased with increasing acquisition time.
The purpose of our work is to provide a platform for the construction of diverse artificial biological cells. Thus, Chl-a was introduced to verify the feasibility of the proposed method. As shown in Supplementary Fig. 27, LHC containing Chl-a could be obtained based on WCC-based HMS.