Realization of a four-step molecular switch in scanning tunneling microscope manipulation of single chlorophyll-a molecules
- Quantitative Biology Institute, Nanoscale and Quantum Phenomena Institute, and Department of Physics and Astronomy, Ohio University, Athens, OH 45701
-
Edited by Robert H. Austin, Princeton University, Princeton, NJ, and approved July 20, 2006 (received for review May 5, 2006)
Abstract
Single chlorophyll-a molecules, a vital resource for the sustenance of life on Earth, have been investigated by using scanning tunneling microscope manipulation and spectroscopy on a gold substrate at 4.6 K. Chlorophyll-a binds on Au(111) via its porphyrin unit while the phytyl-chain is elevated from the surface by the support of four CH3 groups. By injecting tunneling electrons from the scanning tunneling microscope tip, we are able to bend the phytyl-chain, which enables the switching of four molecular conformations in a controlled manner. Statistical analyses and structural calculations reveal that all reversible switching mechanisms are initiated by a single tunneling-electron energy-transfer process, which induces bond rotation within the phytyl-chain.
Since the beginning of 1990, the scanning tunneling microscope (STM) has been demonstrated to be a useful tool to manipulate single atoms/molecules on supporting substrates with atomic-scale precision (1–4). To date, most STM manipulation experiments are performed on single atoms or small molecules, and the applications of this technique are concentrated mainly in the materials science research area. In this article, we extend the STM manipulation procedures to biology-related research areas and investigate a relatively large plant molecule known as chlorophyll-a.
Chlorophyll-a induces green color in plant leaves and is a key ingredient in photosynthesis, one of the most important biological processes, which converts sunlight into chemical energy in plants (5–9). Chlorophyll-a is also important from the evolutionary standpoint because photosynthesis played a central role in the early development of life on Earth (6). But, just as chlorophyll has been vital in the development and sustenance of plants and life forms, it may prove to be just as essential to the advancement of “green” energy research and nanotechnology. Because of their nontoxic nature and their abundance in the natural world, plant molecules like chlorophyll-a are given special interest in the quest for green energy resources and for the development of environment-friendly nanoscale devices (10–12). Chlorophyll-a consists of two main components: a porphyrin unit as the “head” and a long carbon-chain as the “tail.” In the light-harvesting reaction centers found in plant leaves, chlorophyll-a conforms into various shapes by bending the phytyl tail. Molecular conformation is a key process in many biological functions, and controlling the conformational changes of biological molecules with submolecular precision is a dream for many scientists. Here, we are not only able to resolve the structure of single chlorophyll-a molecules but also to reversibly switch four molecular conformations in a controlled manner; the detailed switching mechanisms are explained by means of both experimental analyses and theoretical calculations.
Results and Discussion
Chlorophyll-a is weakly bound to the Au(111) surface, and the molecules are easily displaced during imaging. The STM images of chlorophyll-a on Au(111) show both single molecules and regions of self-assembled molecular clusters preferentially located at the elbows of Au(111) herringbone reconstruction (Fig. 1 a). To understand the adsorption geometry of chlorophyll-a, we will first discuss the structure of the molecule in the clusters. Chlorophyll-a clusters grow epitaxially on Au(111) and form a close-packed structure with a unit cell length of 1.6 nm (Fig. 1 a). Inside the clusters, the molecules position in pairs with their “heads” facing each other. In a single row along the long molecular axis direction, the molecules assemble in an alternating “head–tail–tail–head” arrangement (Fig. 1 b). From the atomically resolved STM images, the orientation of the molecules with respect to the Au(111) surface is determined. The long molecular axis is aligned along the [211] surface directions of Au(111). Such a close-packed self-assembly of chlorophyll-a is significant, mimicing the in vivo packing of chlorophyll-a in the photosynthetic membrane and possibly having important applications in solar cells and medical devices (10, 12).
Chlorophyll-a structure. (a) A self-assembled chlorophyll-a cluster having a rhombic unit cell of 1.6 nm length, and scattered single molecules on Au(111) surface with the Au(111) herringbone reconstruction in the background (30 × 28 nm2 area, V t = 1.5 V, I t = 280 pA). (b) Chlorophyll-a molecules located inside the cluster show three bright triangular units (shown with red circles) contributed from the phytyl-chain. The carbon atoms of these CH groups are labeled (2.8 × 2.5 nm2 area, V t = 0.3 V, I t = 280 pA). (c) The calculated geometry of chlorophyll-a using the parametric method shows the porphyrin head and phytyl tail positions. (d Left) A side view of the phytyl charge-density plot. The red arrows indicate the locations of the three triangular units seen in b. A yellow STM line scan (1.6 nm length) taken along the green line in b reveals a similar profile as the calculated density. (d Right) The top view of the phytyl-chain structure shows the elevated CH groups seen in b.
The calculated structure of a free-standing chlorophyll-a molecule using the parameterized method PM3 (ArgusLab software; Planaria Software, Seattle, WA) shows that the phytyl-chain is folded on the porphyrin unit (Fig. 1 c) (15). The PM3 method is based on a simplified Hartree–Fock theory using NDDO (neglect of differential diatomic overlap) integral approximation. It is a self-consistant field method that takes in account electrostatic repulsion and exchange stabilization. Chlorophyll-a adsorbs on Au(111) by maintaining a similar structure: The porphyrin unit lies flat on the surface while the phytyl-chain is folded on top and elevated from the surface by the support of four CH3 groups. In the high-resolution STM images, the phytyl appears as a chain of three triangular units, which are assigned as the elevated C-H groups corresponding to the following carbon atoms: (C4, C5, C6), (C8, C9, C10), and (C12, C13, C14), respectively (Fig. 1 c and d). The carbon atoms C3, C7, C11, and C15 are attached to the CH3 groups and are located slightly lower than their neighbors in this conformation (Fig. 1 d).
Isolated chlorophyll-a molecules on Au(111) show a similar structure as in the clusters (Fig. 2 a) with less detailed features: Instead of three triangular units, they appear as three lobes in the STM images. This effect might be partly due to a higher mobility of the phytyl-chain in isolated molecules during STM imaging as opposed to the ones in the clusters, which are locked-in by the neighboring molecules. In this adsorption geometry, the binding of phytyl-chain to the surface is weak, and thus it is relatively easy to change its conformation by using an inelastic tunneling spectroscopy scheme (2).
Four-step molecular conformation switching. STM images (Left) and calculated structures (Right) using the PM3 method. (a) A straight-tail conformation (1) can be switched into a bent-tail conformation (2; b) by changing the tail angle at A. This switching is caused by a counterclockwise rotation of phytyl-chain at the location shown with a red arrow in the calculated image. Another 60° bending at B in the STM image (Left) changes conformation 2 to conformation 3 (c). This tail-bending is due to the rotation of a part of the phytyl indicated with a blue arrow in the calculated image (Right). Further 60° bent at A changes the conformation 3 to conformation 4 (d). This switching is realized by a clockwise rotation of the phytyl shown with an orange arrow in the calculated image (Right). The STM tip-height profile below is taken along the long molecular axis of chlorophyll-a from a. The height of the molecule remains more or less the same upon switching of the conformation. The cartoons demonstrate the STM tip-induced switching procedure, and the tail bending locations, A and B.
Four conformations of chlorophyll-a, marked as 1, 2, 3, and 4 (Fig. 2), have been selectively switched by injecting tunneling electrons from the STM tip. Each switching step involves 60° bending of the phytyl-chain by positioning the STM tip at one of the two locations indicated as A and B in Fig. 2. For instance, to switch between conformations 1 and 2 (Fig. 2 a and b), the STM tip is placed at a fixed height above the phytyl-chain near A, and tunneling electrons are injected into the molecule by using a 1.5-V bias. Switching the conformation 2 to 3 and 3 to 4 are also realized by using a similar process. The 2-to-3 switching (Fig. 2 b and c) involves 60° bending at B, whereas the phytyl-chain angle at A remains unchanged. The 3-to-4 switching is done by bending the phytyl at A again. When the molecule is in conformation 2, it can be switched either to conformation 1 or 3 by bending of the phytyl at separate locations A and B. Similarly, 3-to-2 and 3-to-4 switching processes can be selectively performed by bending the phytyl at separate locations, B and A, respectively.
To understand the structural changes during switching, we have performed geometrically relaxed calculations for several free-standing chlorophyll-a structures using the parameterized method PM3 (15). The calculations reveal that the three bent molecular conformations observed here are caused by rotations of parts of the phytyl-chain. The 1-to-2 switching involves bending of the phytyl at C12 joint. This is caused by the rotation of the σC11-C12 bond (indicated in Fig. 2 with a red arrow in 1), which rotates the last part of the phytyl (from C12 to C15) in a counterclockwise direction resulting in a 60° tail bending (Fig. 2). The 2-to-3 switching is induced by bending the phytyl at C8 joint by rotating the σC7-C8 bond (shown in Fig. 2 with a blue arrow in 2). This rotates a part of the phytyl, from C8 to C15, in a counterclockwise direction. A clockwise rotation of the end part of the phytyl at C13 in conformation 3 switches to the conformation 4 (indicated in Fig. 2 with an orange arrow in 3). Because the phytyl-chain is lifted up from the surface by the support of CH3 groups, such chain rotations can easily take place. The calculations reveal that chlorophyll-a can have a rich variety of conformations. We conclude that our ability to switch only four chlorophyll-a conformations on Au(111) is due to the limitation of the phytyl tail rotation imposed by the surface. By comparing the calculated results with the experiment, we find location A (Fig. 2) as the C12 and C13 region and B as the C8 joint.
During the tip-induced switching process, the tunneling current is recorded as a function of time and the conformational changes can be recognized from the abrupt changes in current intensity (Fig. 3 a). By increasing the process duration, the molecule can be switched between the two conformations back and forth, producing a two-step current signal (Fig. 3 b) (16–19), which functions like a toggle switch. The activation barrier for switching is determined by acquiring I–V spectroscopy data on isolated chlorophyll-a molecules, which reveal the current fluctuation above 0.8 V due to the conformational changes (Fig. 3 c) (21). In general, rotational excitations of a free molecule require lower energies than electronic excitations (20). The 0.8-eV barrier (1 eV = 1.602 × 10−19 J) determined here may include contribution from the surface. The I–V curve in Fig. 3 c shows that when the current increases at the larger voltages, the frequency of switching also increases (21). For a fixed bias, the average switching frequency can be adjusted by varying the tunneling current only. The increase or decrease in the current induces faster or slower switching rates, respectively.
Switching operation and mechanism. (a) An abrupt change in the tunneling current is associated with 1-to-2 switching. The current trace is recorded for a fixed voltage pulse of 1.5 V. (b) For a longer electron injection period, a two-step multiple-switching signal is observed. (c) An I–V curve of chlorophyll-a shows fluctuation of current above 0.8 V due to switching between 1 and 2. (d) A binned distribution of the switching events corresponding to 0.6 nA. (e) A logarithmic switching rate vs. current graph shows a unity slope. (f) The yield plots for 2-to-1 and 2-to-3 switching events. All of the data sets used for the plots are recorded with a 1.5-V bias.
To understand the tunnelling-electron-induced switching process, we have performed statistical analyses taken over 1,200 switching
events by using a 1.5-V bias and at different currents for the four conformations of the molecule. As described above, the
switching can already performed with a 0.8-V bias. 1.5 V is chosen here to ensure a single-electron tunnelling process. Fig. 3
d shows the plot of conformational changes vs. time for the 1-to-2 switching with a fixed current of 0.6 nA. The time constant
of this exponential decay has been determined as 0.97 ± 0.06 s. The switching rates are determined from the inverse of the
time constants. The linear dependence of switching rate on the tunneling current is illustrated in Fig. 3
e, where each data point is determined by plotting an exponential curve shown in Fig. 3
d. The rate, R, and tunneling current, I, are related as
where N is the number of inelastic tunneling electrons involved in the energy-transfer process (22–24) to the molecule that induces conformational switching. From the slope of this curve, N is determined as ≈1. Therefore, this switching process is initiated by a single-tunneling-electron energy-transfer process.
Similar results have been found for the 2-to-3 and 3-to-4 switching processes, and thus single-tunneling-electron energy transfer
causes all of the switching events described in this article. We have also determined the quantum yields for the switching
events. Fig. 3
f provides the yield plots for the 2-to-1 and 2-to-3 switching events. Here, the yield, Y, is related to the switching rate, R, and current, I, as
where e is the electronic charge. The determined yields are 3.2 × 10−10 for switching from 2 to 3 and 1.5 × 10−10 for the switching from 2 to 1. Almost zero slopes of the yield plots further verify the single-electron energy-transfer process
(25). The measured yields for 3-to-2 and 3-to-4 switching events are 3.8 × 10−10 and 3.3 × 10−10, respectively. To demonstrate our ability of control further, we present snapshot images from a STM movie (Movie 1, which
is published as supporting information on the PNAS web site) showing the switching of chlorophyll-a conformation between the
conformations 1, 2, 3, and 4 in Fig. 4.
Frames of STM images from a STM movie showing reversible switching of conformations 1, 2, 3, and 4. Each manipulation step involves taking the STM image of chlorophyll-a, inelastic tunneling spectroscopy excitation to switch the conformation, and then retaking the molecule image. Image parameters: 2.8 nm × 3.6 nm scan, 0.5 V, and 93 pA; switching parameters: 1.5 V and 0.6 nA. The images taken after each switching event are sequentially presented.
In summary, we report an extensive study of single chlorophyll-a molecules and provide detailed information about the structural properties of isolated molecules and self-assembled molecular structures at the atomic level. Our four-step molecular switching scheme demonstrates that chlorophyll-a may be useful in nanoscale biomechanical devices. Furthermore, this achievement opens up a route to investigate, and even control, the conformational changes of biological molecules including some proteins with submolecular resolution by using STM manipulation schemes.
Materials and Methods
Our experiments were performed by using a home-built low-temperature STM operated at 4.6 K in an ultra-high-vacuum environment (2). The Au(111) surface was chosen as a supporting substrate for the experiments because of its relative inertness. The sample surface was cleaned by repeated cycles of sputtering with neon ions and annealing. After confirming the cleanliness of the Au(111) sample by STM imaging, a submonolayer coverage of chlorophyll-a produced from spinach (97% purity; Sigma-Aldrich, St. Louis, MO) was deposited onto a surface held at room temperature via vacuum evaporation (13). The sample temperature was then lowered to 4.6 K for the measurements. The temperature of the base of the scanner, where the sample sits, is monitored precisely by using a silicon diode. The STM tip, an electrochemically etched tungsten wire, was prepared before the experiment by using a controlled tip-crash procedure described in ref. 14.
Acknowledgments
We thank Jessica Benson for initiation of the project and Aparna Deshpande for proofreading the manuscript. We gratefully acknowledge the Ohio University NanoBioTechnology Initiative and funding provided by the U.S. Department of Energy (Basic Energy Sciences Grant DE-FG02-02ER46012).
Footnotes
- *To whom correspondence should be addressed. E-mail: hla{at}ohio.edu
-
Author contributions: S.-W.H. designed research; V.I. performed research; V.I. and S.-W.H. analyzed data; and S.-W.H. wrote the paper.
-
Conflict of interest statement: No conflicts declared.
-
This paper was submitted directly (Track II) to the PNAS office.
- Abbreviation:
- STM,
- scanning tunneling microscope.
-
Freely available online through the PNAS open access option.
- © 2006 by The National Academy of Sciences of the USA
