How optical super-resolution is achieved (7): PALM

After the accomplishment of STORM blog, there is nothing new to write about the super-resolved technology besides “switch-localization”.
But there are two twin techniques invented almost at the same time that I must mention. First of all, the index of the two articles:

Eric Betzig*, George H. Patterson, Rachid Sougrat, O. Wolf Lindwasser, Scott Olenych, Juan S. Bonifacino, Michael W. Davidson, Jennifer Lippincott-Schwartz, Harald F. Hess*, “Imaging Intracellular Fluorescent Proteins at Nanometer Resolution”, Science 2006 Vol. 313 no. 5793 pp. 1642-1645  (Received for publication 13 March 2006. Accepted for publication 2 August 2006. Published Online August 10 2006) *Equally contributed. 

 Samuel T. Hess, T. P. K. Girirajan, and M. D. Mason, “Ultra-High Resolution Imaging by Fluorescence Photoactivation Localization Microscopy,” Biophysical Journal, vol. 91, no. 11, pp. 4258-4272, 2006. Submitted June 12, 2006, and accepted for publication August 28, 2006. Published 1 December 2006

There can be a lot of coincidence, but this two in scientific history can be called a miracle. The first one of these two techniques is called PALM, while the second one is called FPALM; the inventor of the second technique is Hess, while one of the inventors, the co-first author of the first one is also known as Hess (If “s” is straightened into “l”, Hess will become Hell, i. e., the inventor of STED, which is even more surprising); the first technique was invented in March 2006, while the second one in June 2006.

Eric Betzig

Harald Hess

Samuel Hess

 First of all, let’s talk about Eric Betzig. I was fortunate enough to read his paper which was published in Science in 2006, and surprised to found that he lived in Michigan, USA Lansing where I was. When I introduced his work to everyone in the group meeting, my boss was surprised, “Oh, is it him? He had studied the near-field optics. We had dinner together and discussed some issues about optical imaging a few years ago, and then he opened a company and we lost contact. Wow, good for him that he is now in HHMI!”

So I day-dreamed freely that, maybe I have the luck to fish with Eric in the same river, just not at the same time point. Even, maybe, the fish escaped from his fishhook was just picked up by me in another time spot. There are only a few fishing places in Lansing so this could be real.

The story of Eric Betzig is a typical struggling story about what we called in Chinese “the rich second generation”. The following stories are partly from the personal introduction of HHMI and the editor interview of Nature Photonics in 2008.

Eric graduated from Caltech and received his doctorate at Cornell University. Then, he entered the Bell Labs. In the six years there, he studied the near-field optics and used it for biological cell imaging (in my last posts, readers asked me to write some about the super-resolution with perfect lens, in which the capture of evanescent wave is essential). He registered a company named New Millennium Research, but soon was called back by his father to manage their machining industry empire. After all, in Michigan the automobile industry is thriving, therefore machining industry holds a fair stake.

There, a problem had troubled the machining industry for many years, perhaps because it was too difficult, so people turned to accept it: a big heavy machine tool must be moved or stopped, in order to process a small component with high precision. Thus, much time and energy were consumed by moving the big heavy tool.
Eric skillfully made the processing machinery moving at a high speed without sacrificing the machining precision, so the processing efficiency was improved greatly.

However, how can a sparrow understand the ambition of a swan? Numerous nights, he dreamed that he returned to the familiar microscope, and there, the cells are swimming, the mysterious life signal is transmitting. However, the fuzzy eyes can not see all this…  He woke up with a start! Because he has figured out a way to see things clearly.

Nevertheless, Eric is realistic: when applying for jobs, there are ten years of academic blank on his resume. “If you want the world to listen to your theory again, you must take out some good convincing things.” he said.

Eric’s father seems to be not so supportive for him to leave the family business.Before go to HHMI, he didn’t have his own laboratory.In my opinion, his father could easily equip a thirty square meter room as an optical lab just to keep him.

So, most of their experiments were completed in a laboratory in the condo of his good friend Harold Hess.There, they put the optical bench in the living room to instrument their new microscope system.The whole winter, Eric Betzig, Harald Hess and Lippincott-Schwartz the biologist had been working in that tiny laboratory without heater.

A team of three dedicated persons, without the Christmas holidays, without the luxury experimental equipment. There is only a strong faith supporting them to keep moving: as it is feasible theoretically, then there is no excuses, we will prove it experimentally!

When spring came, they finally met the Luck Fairy. Recalling the circumstances at that time, Lippincott-Schwartz said: “They were very excited. I still remember when we got the first microscopic image, we cannot distinguish what it is. I can’t believe that we succeed until I saw the result of overlaying the fluorescence image with the electron microscopic image.I think all this is really amazing.”

 Eric Betzig, Harald Hess and Lippincott-Schwartz published their PALM research results in the Science Journal in 2006. The cell adhesive plaques and the proteins in specific organelles can be seen clearly by PALM. The whole process of publishing: submission: 13th March, acception: 2nd August, publication: 10th August.

The lysosomal transmembrane protein visualized by PALM.

Next, the story of FPALM. Samuel T. Hess received his undergraduate and doctoral degrees at Yale University and Cornell University, respectively. He served as an assistant professor at University of Maine at that time. There, he had a project and the deadline was approaching. Samuel was indulged in some discussions between the chemical engineers and biological engineers of their university: how to improve the resolution of the observation of living cells lipid raft structure?

Recall our previous blog that, in confocal, the pinhole functions to block the noise from neighborhood. Hess was awakened by the burst of electronic percussion in a summer night of 2005. Neighbors held the dance party once again. Helpless! Pleasure-loving people, who knows that our scientific researchers are sleepless monsters? Half asleep, Hess walked downstairs. He wanted to ask them to be quiet, but hesitated (readers who have watched the “The Big Bang Theory” may know that our scientists are too often not good at this). OK, just let it go. He drew a design sketch, in which the cell morphology can be visualized more clearly with the aid of specific on-off fluorescent marker.

I often tell my students: when your thought is not clear, draw it!

The next morning, when he looked at the scratchy design sketch drawn in half-asleep state again, Samuel could not help laughing: it was so simple, but solved the problem of “blurry”. Is it correct? Looking at it again, it seems that surprisingly the design sketch did not violate any physical principle. He discussed it to his fellow colleagues to review the concept, and no problem was found.

Next, Samuel began to make the microscope according to his design sketch.
Then he assembled the microscope and did the test as soon as possible in hoping that it can be finished before the deadline. Meanwhile, in less than two days, his colleague from Surface Science and Technology Laboratory had prepared the sapphire crystal samples as a test sample for the new microscope. As you can see, an efficient cooperative team is always crucial to your success. The research results of Samuel Hess group were published in Biophysical Journal at the end of 2006.
Hess group proved that the protein clusters on the cell membrane lipid rafts can be distinguished with FPALM in 2007.

Open access is a blessing to the advance of science. It gives us back the privilege that ‘Human knowledge belongs to human’. And in PALM, I will take you first to the night of Effel tower:

This movie clip is localized with QuickPALM, an open access project hosted by Google Codes. You can find it readily in Fiji now, or download and install it in your ImageJ.

In New Testament of Bible, Jesus came to the disciples for the second time, but the disciples didn’t recognize their resurrected Savior. Jesus let the disciples to see and touch his hands. Seeing the holes by the nails, the disciples recognized him immediately, and cheered: “It is the Lord!” Since then, a gang of uneducated countrymen, who had been frightened and prepared to flee, became re-union and spread the belief of Jesus to all over the world. Our life today has been deeply changed by them.

PALM is precisely based on the on-off effect of the protein within the palm (determined by the diffraction limit). As there is only one hole of the nail in the palm, it can be localized, and memorized by the disciples.

Works from Xi group:

1. Liu, Y., et al., Achieving λ/10 Resolution CW STED Nanoscopy with a Ti: Sapphire Oscillator. PLoS ONE, 2012. 7(6): p. e40003.
   
2. Ding et al., Laser oblique scanning optical microscopy, Optics Express 2012 20(13) 14100-14108.
   

How optical super-resolution is achieved: Addendum

Right after this post being announced in Confocal Mailing List and LinkedIn groups, I received several emails pointing out a few works that are also very interesting. These may be a great resource for those who wishes to extend their knowledge on the rich history and explosive expansion of optical nanoscopy.

Albert Diaspro:

I love The Lukosz work, i think that One Could also considering works by Toraldo di Francia on super resolution starting from this fundamental reading on The resolution definition:
Toraldo di Francia (1955) Resolving power and information. JOSA 47(7), 497-501.

 

Ultramicroscope:
Siedentopf, H., & Zsigmondy, R. (1902). Uber Sichtbarmachung und Größenbestimmung ultramikoskopischer Teilchen, mit besonderer Anwendung auf Goldrubingläser. Annalen der Physik, 315(1), 1–39.

 

Mark Canell:  Just a couple of examples, they are probably even earlier papers … the idea of a resolution ‘limit’  for self luminous objects was dismissed in the 1950′s -if not earlier. Here’s a few papers that may lead to more:

Exchanging time for spatial information to increase resolution:

LUKOSZ W. Optical systems with resolving powers exceeding the classical limit. II. JOSA. Optical Society of America; 1967;57(7):932–9.

General discussion about super resolution
McCutchen CW. Superresolution in microscopy and the Abbe resolution limit. JOSA. Optical Society of America; 1967;57(10):1190–0.

Using a grating to increase resolution;
BACHL A, LUKOSZ W. Experiments on Superresolution Imaging of a Reduced Object Field. J Opt Soc Am. 1967;57(2):163–&.

Resolution enhancement by non-linear optical effects:
Ehrlich DJ, Tsao JY. Nonreciprocal laser-microchemical processing: Spatial resolution limits and demonstration of 0.2-μm linewidths. Appl. Phys. Lett. 1984;44(2):267.

 

This post is intended to limit on the discussion of far-field nanoscopy, and consequently no near field optics is included. But Guy Cox has made a very brief list of near field microscopy: Synge proposed near-field microscopy in 1928 (Philosophical Magazine 6, 356).  It was first demonstrated in visible light in 1984 (D.W. Pohl, W. Denk, and M. Lanz (1984). “Optical stethoscopy: Image recording with resolution λ/20″. Appl. Phys. Lett. 44 (7): 651.) though it had been achieved in the microwave region as early as 1972.  This is all way before the concept of STED had even been proposed (1994) let alone demonstrated (2002).

 

STED experiment 1999 with 106nm resolution: Thomas A. Klar and Stefan W. Hell, Subdiffraction resolution in far-field fluorescence microscopy OPTICS LETTERS, 24(14) 954-956, 1999

 

Alexandr Egorov:

1. Optical Profile Restoration with a Superresolution Using Sampling Expansions.
2. Solution of the inverse problem for a heterodyne differential microscope for a stepwise profile of a microobject.
3. Microobject profile reconstruction in a heterodyne differential microscope.
See my profile: ru.linkedin.com/pub/alexandr-egorov/54/4b4/249 

 

 

 

 

How optical super-resolution is achieved (6): STORM

Concerning resolving, I cannot help to think of the Four Classis in China. One of them is “Journey to the West”, which tells a fairy story about how the Monkey King helps Monk Tang to fight over monsters to get to India for Buddha’s script, so that Buddhism can be spread in China. In this novel, there is a chapter which describes a six-ear monkey which is almost identical to the Monkey King. This even troubled the Jade Emperor, Monk Tang and the Goddess of Mercy, since the two monkeys are indistinguishable in appearance. Later on, they went to the Buddha, and Buddha told them apart. On their way to Buddha, do you know what Buddha was doing? Chengen Wu the author wrote:

“There’s being in nonbeing, nonbeing in being; there is body in non-body, and emptiness in non-emptiness. Non-body is body, non-emptiness is emptiness. Emptiness is emptiness, color is color. There is no static color in color, for color is emptiness; there is no static emptiness in emptiness, for emptiness is color. Knowing emptiness is not emptiness and color is not color, will make you wise.”

(In Buddhism, color refers to this colorful world and all the real exist in the world.)

 

People may think Chengen Wu wrote this simply for more copy money, since it seems meaningless. It is ABSOLUTELY meaningful!

 

 

A beautiful girl must have read this, and smiled.

 

She is Xiaowei Zhuang, a Chinese graduated from the genius class of Chinese university of science and technology, now a woman professor in Harvard, newly-elected academician in American Academy of Science.

 

I have to introduce another scientist, Paul R. Silven’s work here first. He said, “The trick to get around this problem was to realize that a single molecule’s position can be located arbitrarily well. It’s much like a mountain peak, which can be located to within a few yards, even though the mountain itself may be a a mile wide. Similarly, a single molecule’s center can be determined to be its width (~250 nm), divided by the square root of the numbers of photons collected.” Apparently he is not the first one who realizes localization, but I like the way he explain the concept.

Paul R. Silven, Myosin V Walks Hand-Over-Hand: Single Fluorophore Imaging with 1.5-nm Localization, Science 2003.

 

This idea is so simple, yet, it will surely revolutionize the world. It is named FIONA (Fluorescence Imaging with One Nanometer Accuracy).

 

I will invite you to a cinema now. This time, the film “I, Robot” is on. Will Smith is aftering the intelligent robot who can dream (which resembles that he is revolutionized to have the same feeling as human), and the robot hide himself in the robot factory. When there are many identical objects, with the same appearance and act, identification is impossible. So Will put a gun out and threat to shoot, and looking for which robot will sense the fear and open its eye. It is then easy to localize it, if it has a different behavior (from close eye to open, or OFF to ON)!

Now let’s come back to how Xiaowei Zhuang realized discrimination from “Color is emptiness”.

 

Previously we discussed that two fluorophore cannot be distinguished by traditional methods when they are too closed. The idea behind her method is, if there are no two emitters close by at the same time, then we can use localization method to give the accurate position. So the key is to validate our assumption of “No two close by emitters at the same time”. How? We can let most of them “dead” by photobleaching (reversible). By giving a tiny portion of magic herbal, one comes back to life for localization. Then after that we photobleach it again. Save another one with herbal, and keep this loop until most of them were alive once for imaging.

 

Why Xiaowei Zhuang could publish in Nature Method? Except for the reason she find the proper issue, another key point is she found a Cy3-Cy5 pair, each of its components could emit light of different color, which had been used in FRER before. When researching in cold virus XIaowei Zhuang incidentally found this lovely protein could be switched by light to emit light or not.

 

With this switch super-resolution is realized. Switch has only two states: ON and OFF. It is therefore, binary. And Binary has formed our entire digital world. Binary is also termed Yin and Yang in Dao, which resembles the negative and positive, respectively. From the binary Yin and Yang combination, Bagua is formed, and so to the entire universe in the Daoism, the genuine Chinese religion and outlook on world.

Thank you Xiaowei, you show the beauty of Eastern philosophy to the Western technology world.

 

How optical super-resolution is achieved (5): SSIM

On the closing remark of 2011 Focus On Microscopy, Fred Brakenhoff , chairman of the conference, announced a news which shocked everyone: Professor Mats Gustafsson passed away for fighting with cancer a day ago, at the age of 51. The 2012FOM sub-theme is to memorize the great man, and new advances of SSIM, the nanoscopy technique that he brings to the world.

What is the SSIM? Its full name is Saturated Structured Illumination Microscopy. It is first SIM and then with saturated illumination, breaks the resolution barrier. So before introduction SIM, let me share a picture of my chair with you:

What can you see in the backrest of the chair? Irregular stripes, right?

If you magnify the picture, you will find that such stripe is caused by overlapping of two reticulated fabric layers. It is termed Moire pattern.

The reticulated grids are rather hard to be seen (high frequency), but the Moire pattern are easier to be visualized (low frequency). So, when two fine pattern A and B are overlapping, if we know one of the overlapped pattern of B, through detecting the Moire pattern A+B, we are able to derive the fine structure information that A has. This is the core of SIM, as indicated below.

Before proceed, we can review one well-known resolution enhancing technique: confocal. Through the combination of point-source illumination and pinhole detection, confocal achieves better resolution. More precisely, as in resolving two subjects, the neighborhood is your enemy (they always host parties and make noise). So, one can use a pinhole to block the noise due to neighborhood, to improve the resolution. This is how confocal works for resolution enhancing.

Confocal improve resolution through blocking the detection, whilst SIM improves resolution with modulating the illumination.

How much can the resolution be improved? For confocal, it is 1.4. This is because even the pinhole shrinks to a dimensionless dot, due to the reversibility of optical path, it will form the same Gaussian shape at the focal spot. Therefore, the result PSF is the multiple of two PSFs. If we set the modulation of excitation to the detection side through conjugation, then, SIM mimics a one-dimensional modulation. Further, one can limit the “pinhole slit” from multiple directions , and get a two dimensional resolution enhancing. As it is based on the sum or difference of two frequencies, it can obtain 2x resolution enhancement in theory.

If further resolution improvement is needed, one requires narrower lines. But as we have mentioned before, the width of such optical line is limited by diffraction limit (if the process is a linear one). This holds the limit also for optical lithography, data storage (yes, DVDs), and astronomy. The solution to this problem is: if there is a nonlinear mechanism, then, narrower line can be formed. Adding the 2x resolution enhancement by SIM, the new technique can surpass the diffraction barrier.

Let’s appreciate a beautiful result obtained with SIM:

A wide-field microscopy image (left) and a superresolution structured illumination microscopy (SR-SIM) image (right) each shows actin (green) and tubulin (red) cytoskeleton in a primary chicken fibroblast. (http://www.photonics.com/Article.aspx?AID=47750)

 

Now, I think many people may already eager to join this exciting field of optical super-resolution. Some readers may be hesitating that he/she doesn’t have relevant optical background. Interestingly, Mats also did not have formal optical training when he started this research in his postdoc period. His background is electronics. When he was interviewed, he mentioned: “my own strange perspective on things. For example, I like to think in frequency space, rather than in real space.”

If you are dedicated to a subject that you love, your lacking of background should not be an excuse. But rather, one cannot make up his mind, or have a goal but never put effort on it, will truly holds your dream from coming true.

Mats was diagnosed of cancer in 2005. But he is a naturally optimistic and kind man, with a persistent pursuit toward science. Once a scientific editor asked him, what is the driving force for you to be so dedicated on research? He replied: Suppose you lived in a world in which the ground was a crossword puzzle. Would you or would you not try to solve the puzzle?” He paused. “I can’t imagine not trying. That’s why I’m in science. Because we do live in a world like that.”

(http://www.hhmi.org/research/groupleaders/gustafsson_bio.html)

Mats Gustafsson(1960-2011). Does the fence pattern reminds you of SIM？

Let us memorize Mats, not only because he brings us a new method to benefit human beings with scientific advances, but also his modesty and persistence, which light up the road ahead of us.

Works from Xi group:

1. Liu, Y., et al., Achieving λ/10 Resolution CW STED Nanoscopy with a Ti: Sapphire Oscillator. PLoS ONE, 2012. 7(6): p. e40003.
   
2. Ding et al., Laser oblique scanning optical microscopy, Optics Express 2012 20(13) 14100-14108.
   

How optical super-resolution is achieved (4): RESOLFT

In the previous two chapters, we talked about excitation before de-excitation using STED, or de-excitation before excitation accomplished by GSD. Both methods are able to achieve optical super-resolution microscopy (now termed nanoscopy).

What is the key of these mechanisms? Are there other approaches to be developed?

The energy levels of STED are shown in the figure above (left). It is critical to distinguish the red arrow (stimulated emission) from the yellow arrow (spontaneous emission). Therefore, it is more distinguishable if the red arrow is tuned to the direction which is opposite to that of the yellow one. In experiments, it is feasible to keep the excited electrons going, thus de-excitation can be achieved by Excited State Absorption (ESA). In fact, Stefan Hell’s team has realized the super-resolution of manganese doped quantum dots by using ESA [1].

An arrow can upside down the whole universe.

Let us answer the question aforementioned. In essence, this kind of methods primarily aims to decrease the probability of the electrons staying in the excited state, i.e., prohibit the electrons from staying in S1 state. The electrons may be killed in the cradle (GSD), pulled down in the excited state (STED), or sent to the heaven (ESA).

Now, let us appreciate the beautiful dance on the energy levels:

All these methods can be concluded with RESOLFT (reversible saturable/switchable optical fluorescence transitions). Not only this, but also any kinds of reversible/saturable materials can be used successfully in super-resolution, such as the switching dyes [2] that we are going to mention in the next chapters.

In the next chapter, two other innovational super-resolution imaging modalities as well as their internal relations will be generally introduced.

1.    Irvine, S.E., et al., Direct Light-Driven Modulation of Luminescence from Mn-Doped ZnSe Quantum Dots. Angewandte Chemie, 2008. 120(14): p. 2725-2728.

2.    Grotjohann, T., et al., Diffraction-unlimited all-optical imaging and writing with a photochromic GFP. Nature, 2011. 478(7368): p. 204-208.

Works from Xi group:

1. Liu, Y., et al., Achieving λ/10 Resolution CW STED Nanoscopy with a Ti: Sapphire Oscillator. PLoS ONE, 2012. 7(6): p. e40003.
2. Ding et al., Laser oblique scanning optical microscopy, Optics Express 2012 20(13) 14100-14108.

How optical super-resolution is achieved (3):GSD

In Greek mythology, Antaeus was a very powerful half-giant. He is the son of sea god Poseidon and earth goddess Gaia. He would challenge all he came into contact with to wrestling matches in order to kill them and collect their skulls, hoping to one day build out of them a temple to his father. He was overwhelmingly strong as long as he remained in contact with the ground (his mother), but once he was lifted into the air he became as weak as any other man.

Antaeus had defeated most of his opponents until it came to his fight with Heracles. Upon finding that he could not beat Antaeus by throwing him to the ground, as he would regain his strength, Heracles discovered the secret of his power. Holding Antaeus aloft in the air, Heracles crushed him to death.

Figure 1 Hercules and Antaeus (1690), by Gregorio de Ferrari.

In the previous chapter, we mentioned that STED can narrow the point spread function (PSF) by removing the periphery of the large generated excitation dot, similar to an eraser for a thick pen.

Someone may say that, even without an eraser, I can still draw a fine line! By putting two pieces of paper close to each other to form a slit and then drawing a line following the slit, you may create a line that will be very thin.

Yes, you are absolutely right! In fact, plenty of marks in our daily life are done in this way.

Figure 2 Aerosol cans and masks for painting fine labels with coarse pens.

But, how is this related to super-resolution? Let’s go back to the energy level diagram:

On the right, the figure illustrates that if the peripheral granules were thrown away by a strong excitation initially (similar to Antaeus being lifted from the ground), the granules would come back to the dark triplet state slowly so that only the center particles can be excited, and the effective PSF is decreased!

By discriminating the central excitable molecules (powerful Antaeus) with the peripheral non-excitable (ground-away Antaeus), GSD, the abbreviation of ground state depletion, can achieve super-resolution.

As GSD excites the molecule to a higher energy level, the light level it requires can be much less than STED, in which the molecule is forced to do stimulated emission against the free jump
spontaneous emission that they like. In 2010, Hell’s group reported the realization of 12 nm resolution with carbon nanocrystal GSD effect, in which the light level was 3 orders of magnitude less than that of STED [1].

Figure 3 Comparison of the imaging result of confocal (left) and GSD[1].

 

So far, super resolution is obtained at the current point. But when you shift to the next point, miserable thing happens: the particle you threw away right now is still travelling. What can you do? You have several options:

(1) A stubborn way: wait until the particles come back. Suppose the particle lifetime in T₁ was 1 millisecond; in this case, the integral time per pixel should be at least 1 ms. In other words, if you want to get a 500×500 pixels image, you need at least 4 minutes.

(2) Enhance the speed by parallel imaging. Prof. Hell is also a pioneer of Multifocal Multiphoton Microscopy (MMM) [2], in which multiple foci are generated simultaneously to increase the imaging speed. Applying such a technique on STED or GSD will undoubtedly increase the imaging speed.

(3) Or, you can do simultaneous detection with assistance of a 2-D detector, such as a CCD. Okay, you will ask on how to manipulate a 2-D detector in a confocal system. No I am not talking about confocal; I mean localization nanoscopy which we will discuss in more detail later. The term is GSDIM, GSD followed by Individual Molecule return.

 

1.    Han, K.Y., et al., Metastable dark states enable ground state depletion microscopy of nitrogen vacancy centers in diamond with diffraction-unlimited resolution. Nano Letters, 2010. 10: p. 3199-3203.

2.    Bewersdorf, J., R. Pick, and S.W. Hell, Multifocal multiphoton microscopy. Optics letters, 1998. 23(9): p. 655-657.

 

 

Works from Xi group:

1. Liu, Y., et al., Achieving λ/10 Resolution CW STED Nanoscopy with a Ti: Sapphire Oscillator. PLoS ONE, 2012. 7(6): p. e40003.
   
2. Ding et al., Laser oblique scanning optical microscopy, Optics Express 2012 20(13) 14100-14108.
   

How optical super-resolution is achieved (2): STED

    Previously, we talked about the reason why optical super-resolution can’t be obtained directly within only one step. So, Stefan Hell was wondering whether this could be made by two steps.

    This problem can also be simplified as, how can you draw a thinner line with a thick marker pen?

In this case, you might consider finding an eraser. Then we can draw a thick line and erase the extra part to get a thinner one. This is exactly the same basis for STED.

STED is the abbreviation of Stimulated Emission Depletion, the principle of which can be depicted by the diagram below.

This is the schematic diagram of a typical STED system. In the upper right corner there are energy states of an organic fluorophore. Do you notice the red and yellow arrows among the states?

If so, I believe we can reach an agreement that these two arrows can be easily distinguished. Am I right?

Now, we need to introduce a little bit background information to help us understand this schema better. The green arrow represents the electrons excited from ground state S₀ to the excited state S₁. These excited electrons typically relax to a lower vibrational level in very short time. Then the electrons will smoke a cigarette and take a break here. Oh, wait, no smoking in public places is allowed. These particles can only have a tea or coffee break. The relaxation time lasts only a few nano-seconds. This temporary grand interval is called fluorescent lifetime. After that, the electrons can no longer stay in the upper class S₁ but drop to the bottom level S₀. (To be or not to be, that’s always a question.)

The majority generally make the same choice. Meanwhile, the electrons will drop to different vibrational sub-levels in the relaxed state S₀ subjecting a certain distribution. This statistic behavior can be depicted in the form of emission spectrum.

The upper figure contains the excitation spectrum (blue curve) and emission spectrum (red curve) of an organic dye ATTO647N. From this figure, we can see that the electrons can emit photons with wavelength from 620nm to 850nm, while the maximum possibility happened at 670nm.

This is the truth.

“But the truth isn’t cool,” suddenly, one student stood up and argued,”It doesn’t look orderly enough”. The boy was Albert Einstein.

“Has everyone seen the film Titanic? Isn’t the most touching part ‘You jump, I jump’ ?” he continued.

    “When Rose met Jack, they would like to jump together intimately.”

The romantic film brought little gold Oscar to the director James Francis Cameron, and also helped Chaels Townes, A. Prokhorov and N. Basov who believed in Albert’s prediction to win their Nobel Prizes. And letters “SE” among the word laser that was invented later stands for stimulated emission. This neat and tidy jump makes laser to be the strongest light source in power.

[Sometimes the Nobel Prize Committee could be ridiculous. For example, Einstein didn’t win his Nobel Prize on the General theory of relativity or stimulated emission but with the discovery on photoelectric effect. Striking similarity also happened to the first inventor of laser, Theodore Maiman, who is in every textbook of laser but not share the Nobel Prize.]

    Let’s turn back to the super-resolution question. It was a warm summer evening. The young post-doc Stefan Hell was lying on his bed in the dorm when the STED idea suddenly hit him, “Bingo! I can separate the stimulated emission from the spontaneous emission, to get super-resolution!”

Let’s turn back to the yellow and red arrows situation. If the minimum PSF is equal to the radius of the green circle in the fluorescent phenomenon raised by the green arrow, we overlap a red doughnut shape eraser to the green circle (the stimulation emission wavelength is the same as the pump beam), then only the molecules in the center will still fluoresce. The PSF is effectively reduced, thus super-resolution is achieved.

    Then the following experiment can be much easier: pick a proper fluorescent dye, and align the laser pulses in precise time sequence. After a 2ps excitation pulse to excite the electrons to the excited state, a depletion pulse come with the duration of 250ps. Then we distinguish the spontaneous emission from the stimulated one in the detection path. The stronger the depletion beam is, the smaller PSF, or higher resolution we will get. This is the pulse STED system. If you feel that the precise synchronization of two laser pulses and maintenance of the accurate time delay is too complicated and we can actually use continuous laser sources, since the DC mirror can separates the signals, only with a heavier background. This is the CW-STED system.

Now, let’s appreciate the beauty that STED brings to us:

(Image from Leica website).

Another interesting phenomenon: take a look at the image below. If you wear glasses, you will see Einstein, the famous scientist that we have mentioned many times. Now put your glasses off. What have you seen? Maria Monron the beauty. (What? You see constantly Einstein? Tat means your myopia is not deep enough, or you don’t like beauty… Move your head far away from the screen).

Einstein Monron hybrid image made by MIT.

Our group has visualized all three kinds of cellular skeleton with our STED, and we found that, insufficient resolution can sometimes make your illusion mistake, just like Einstein to Monron. See below the actin filaments.

As the resolution of confocal system is insufficient, the noise mimics a double helix structure of actin filaments, which is clearly in parallel as shown in STED.

Works from Xi group:

1. Liu, Y., et al., Achieving λ/10 Resolution CW STED Nanoscopy with a Ti: Sapphire Oscillator. PLoS ONE, 2012. 7(6): p. e40003.
   
2. Ding et al., Laser oblique scanning optical microscopy, Optics Express 2012 20(13) 14100-14108.
   

How optical super-resolution is achieved (1)

Preface

Shining!

The amazingly bright sunshine in the May of Beijing could not shelter the glory of optical microscopy for super-resolution imaging yet in scientific fields.

This month^*, Optical Nanoscopy, a new open access journal launched by Springer, invited Prof. Stefan W. Hell, founder of several optical nanoscopy methods, to be editor-in-chief.

In January 2009, Nature Methods picked fluorescence super-resolution microscopy as the Method of the Year 2008. Only half a year later, Nature Photonics could not help publishing a new issue on super-resolution imaging again.

Indeed, it is an unprecedented affair for Nature Publishing Group to publish new issues in such a frequent manner.

^*This series is originally written in Chinese, after I gave a related plenary talk in May 2012. The audiences were from all sorts of disciplines, from fresh graduate students to renowned professors. Therefore, I decided to make the talk interesting, and easy to follow by everyone. It was surprising to me that, afterwards, the audiences gave a long applause following with intuitive questions, which has prompted me to enlighten the audience by blogging more information about the subject. The number of Chinese reader hits is now over 10 thousand. The number is not important; what is important is that every effort is worthwhile if it helps you to stand on the shoulder of the giants, especially the giants of optical microscopy, which is a discipline that relates to so many others.

Chapter 1        Every Cloud Has a Silver Lining

Peng Xi and Yichen Ding

Biologists almost fell in love at first sight when optical microscopes were introduced. What other method can help us understand inner structures of live cells or tissues so clearly? Thus, this is why both Antonie van Leeuwenhoek and Robert Hooke applied their first handcrafted microscopes to biology above all.

Optics is essential in microscopy. Its historical development relies on three major aspects: theory, material and engineering. To our surprise, in the early twentieth century, these three came together in a small, beautiful town in Germany called Jena. Ernst Karl Abbe, an optical theory master; Otto Schott, a pioneer in optical materials; and Carl Zeiss, a great founder of optical engineering. They have made concerted efforts to lead optics to a higher level for the whole world. Nowadays, the enterprises founded by Schott and Zeiss, respectively, are still renowned around the world.

I will assume that readers are familiar with the Ten Commandments, which is God’s word engraved in stone and given to Moses. In the Bible there is another story (Daniel 5): a very proud king, who is the descendent of Nebuchadnezzar, does not worship God, but worships those idols made of gold and stone (Nebuchadnezzar was once turned into a bull and made to eat grass for 7 years as punishment by God for worshipping idols). One day when the king was playing in his palace, he saw a finger writing message on stone. The king was very surprised at this and asked many wise people to read it, but nobody could. He then went to Daniel, who was a prophet of God. Daniel told the king that
the message represents God’s plan to take the kingdom from him. That night, the proud king was murdered. Therefore, “words on the stone” sometimes refers to unchangeable facts.

To challenge the tradition, you need first courage, and then intelligence.

After careful research in traditional theory of optics, Abbe put forward an issue about the diffraction limit. That is, no matter how one improves the resolution of a microscope, it is hindered by diffraction all the time due to the finite diameter of objective. Following is its presentation in mathematics.

It is also called the point spread function (PSF). This formula reveals the limit of dissolving power of optical microscope, which basically states: if simultaneously lighting on all the molecules inside d, there is no way to differentiate them optically.

Figure 1 Ernst Abbe’s memorial sculpture in Jena University.

However, when most people were disheartened, someone burst out a revolutionary idea: “Why not try to switch some of them on purpose so that we can tell them apart?”

Einstein once said, “The mere formulation of a problem is far more essential than its solution”. Definitely, he’s right again. This time, the problem is raised by Stefan Hell, who was just 29 when he put forward the question above.

Works from Xi group:

1. Liu, Y., et al., Achieving λ/10 Resolution CW STED Nanoscopy with a Ti: Sapphire Oscillator. PLoS ONE, 2012. 7(6): p. e40003.
   
2. Ding et al., Laser oblique scanning optical microscopy, Optics Express 2012 20(13) 14100-14108.