Nature has honed its ability to harvest energy from the sun for millions of years, which makes it no surprise that light-harvesting in biological systems is extremely optimized. Photosynthesis, as is taught in many early biology classes, doesn’t seem to have very many outstanding questions. We know how sunlight is turned into carbohydrates to power the cell, so I was shocked to learn how many steps of this process were still mysteries. For example, the crucial first step of photosynthesis, light-harvesting, remains difficult to imitate. In this step, the photon is absorbed and the resulting energy is transferred towards the reaction center, but nature has perfected this process with a quantum efficiency close to one. Even more impressively, this process occurs in disordered and fluctuating environments, so the efficiency must remain robust to the heterogeneity of the biological world. Artificial photovoltaics that attempt to mimic this efficient light harvesting have, to date, utilized very ordered, crystalline structures and yet fail to replicate the high efficiency maintained by natural light harvesters. One way to gain insight into the mechanisms behind this high efficiency is to study how the different microstates of the small protein subunits responsible for light harvesting affect energy transfer dynamics. 

When I say photosynthesis, you’re probably thinking of green plants, but it turns out single-cell bacteria were the true developers of photosynthesis. Not only were these bacteria the first organisms to perform oxygenic photosynthesis, they remain today as the most abundant. Their machinery is less complicated than other, higher-order organisms like those green plants we all first think of. The specific light-harvesting subunits chosen for this investigation are all from the light-harvesting complex in cyanobacteria. The subunits we chose, allophycocyanin (APC) and C-phycocyanin (CPC) contain only 6 and 9 pigments, respectively, embedded in their protein scaffold. Therefore, uncovering the interaction of the various pigments and how energy would likely be transferred between them was much simpler in APC and CPC and made them excellent initial biological samples.

How, though, can we measure these interactions of the pigments involved in the light-harvesting step? Energy is transferred super quickly, at a rate that is equivalent to a quadrillionth to a trillionth of a single second, a range we call ultrafast. Additionally, we can’t forget that the pigments transferring the energy are attached to a protein scaffold that is fluctuating. These changes in the local environment around the pigment should alter function, and this has not yet been thoroughly investigated. Typical investigations into light-harvesting focus on either resolving the ultrafast energy transfer steps with ultrafast spectroscopy, or probing individual molecules with single-molecule spectroscopy. Both have their drawbacks – ultrafast averages over the underlying distribution since it requires an ensemble sample and single-molecule is limited to longer timescales due to the fluorescence detection. So, to measure the energy transfer in APC and CPC, we needed a specialized tool that could measure really fast timescales while also focusing on a single molecule at a time. Luckily, a nascent technique had just been developed called single-molecule pump-probe (SM2P) which aimed to combine the pros of both ultrafast and single-molecule spectroscopy. This allowed us to start investigating how the different conformational states of the small subunits affect energy transfer.

SM2P uses two laser pulses shot into a confocal microscope, which we focus onto a single molecule and collect its fluorescence. The time between the two pulses is varied, which changes the intensity of the fluorescence we record if energy transfer occurs in that molecule during the measurement. When the delay between the two pulses is small, i.e. shorter than the amount of time it would take for the system to transfer energy away from the excited pigment, the second pulse simply stimulates emission and the fluorescence is low. When the delay time between the pulses is longer, the excited pigment has time to transfer energy to other pigments which will then fluoresce, leading to an increase in signal. Therefore, the edges of a typical SM2P trace have higher fluorescence and a “dip” in fluorescence in the middle. We can then extract the energy transfer rate from the curvature of this dip. The idea that this technique allows not only measurement of individual energy transfer rates from single molecules, but also direct observation of the effect of excitation moving from pigment to pigment was extremely exciting. Now we just had to get it to work for our sample.

In single-molecule experiments, you need to repeat the experiment on multiple molecules to build up statistics to completely understand the distribution. This meant many late nights collecting many different SM2P dips to be fit. Not only that, but the complexity in the experiment led to a low success rate. In order to observe a SM2P trace that displays the characteristic dip, the molecule must not photobleach or change state during the entirety of the measurement. This meant, during the crunch time of experiments for this paper, I would be stuck in the dark for hours at a time, every day, watching little molecules fluoresce. Soon, I started to see “dips” everywhere. I would be riding home from work and notice a crack in the sidewalk or a piece of graffiti and think – there’s a dip I could fit.  And that’s when I realized I had fallen in love with this experiment. When I started to see “dips” everywhere, I was in so deep with SM2P that it had permeated every part of my day…and I am so happy it did. SM2P helped us start to uncover how the protein scaffold affects the dissipation of energy in an excited pigment. We discovered that the median timescales of energetic relaxation in APC and CPC were longer than the timescale we observed in a simple dye pigment with no protein scaffold. Furthermore, now we have a specialized tool to answer our remaining questions about the details in light-harvesting, and I’m so excited to see which questions SM2P can help us answer next.

If you're interested in reading our thorough investigation of energy transfer in APC with SM2P, please read our full article here.