Calcium imaging as I understand it:-)

(Here's  a picture of GCaMP from Wikipedia, in case you haven't seen it;-)-- read on to find out about what it is and what it does!)

\

 

 

Calcium is a crucial signaling molecule. It plays a role in neurotransmitter vesicle release from pre-synaptic terminals. Its entry into the post-synaptic cell is correlated with action potential firing. It enters muscle cells to initiate signaling pathways that lead to muscle contraction. It is involved in the regulation of neuronal pathways that lead to changes in synaptic strength between neurons, the basis of biological learning, which allows us to adapt to a changing environment just in time, faster than evolutionary timescales (Harvard course). Calcium imaging is therefore important for elucidating the cellular basis of many biological phenomena:-) But how do you image it, i.e. observe its spatial and temporal molecular fluxes during in vivo conditions?

Fluorescent probes are molecules that are used to "tag" other molecules of interest and they are the basis of fluorescence microscopy. In the case of calcium imaging, calcium is the ion to be tagged and a fluorescent indicator can either be a dye or a genetically encoded protein indicator, a subset of which is genetically encoded calcium indicators (GECIs). As Helmchen in the "Handbook of Neural Activity Measurement" puts it so well: "Being familiar with the principles of how calcium indicators work and with the mathematical description of calcium dynamics is essential for the design and interpretation of calcium imaging experiments." But first, what is fluorescence?

We start our review with dye molecules because it is simpler to explain some basic principles using these molecules. The first thing to remember is that light can control the quantum states of molecules and in turn, molecular states control the light that is emitted when the photons of the light ray and these molecules come into contact. When you shine light of a special frequency on a dye, it evokes the process of excitation/emission. What happens is that an electron in the molecule gets excited into a higher orbital state by absorbing a photon of a particular frequency and (sometimes, see the following paragraph) falls back to an intermediate level, dissipating a tiny bit of heat (this means that there is an energy loss during the excitation process) and finally decays into its ground state. During these processes the dye molecule emits light with a different frequency than the original light you exposed it to. In general, the emission light is of a longer wavelength (lower frequency, lower energy) than the excitation light due to the energy loss during the intermediate process, as described above (Stuurman, 2013). The excitation process happens very fast! It happens in femto-seconds, which is a time period that it takes a light ray to travel 0.3 microns (Stuurman, 2013). In general, the emission light is of a longer wavelength (lower frequency, lower energy) than the excitation light (Stuurman, 2013). The intermediate process happens on the timescale of pico-seconds, during which a light ray would travel 0.3 mm. And only then the electron falls back to its ground state and emits a photon. This process takes a comparatively long time, a light ray would travel 3 m using this time. The time lag between absorbing a photon and emitting one is called the LifeTime and it is usually on the order of nanoseconds for most dyes (Stuurman, 2013).

The electron will not always emit a photon when it absorbs one. It can transfer the absorbed energy to another molecule instead or it can fail to emit a photon in a probabilistic process. This is where the concept of quantum efficiency comes in. The quantum efficiency of the dye is the ratio between the absorbed and emitted photons.

The fluorescent molecules have spectra (Fourier analysis, hooray!-- spectra are basically histograms that show how much of each frequency a ray of light contains). There is an excitation spectrum and emission spectrum, which captures the distribution of frequencies of light that cause a dye molecule to fluoresce and the spectrum of the population of photons that it emits, respectively. Stokes shift quantifies the distance between the excitation maximum and emission maximum in these spectra (Stuurman, 2013). Different dyes have different Stokes shifts. 

Microscopes work by differentiating between absorption and emission light using filters:-) I will cover different microscopy techniques in a later post, so stay tuned:-)

We can now take the first step and describe why we can measure calcium concentrations using indicators. The thing is that dyes can bind calcium and when calcium binds to a fluorescent indicator it induces a measurable change in the molecule's fluorescence that can be read out using microscopes (e.g. optical filters) (Helmchen, 2012). An example of a small organic molecule that binds calcium is the Oregon Green BAPTA-1. Binding of calcium to the "pocket" in this molecule leads to a reconfiguration of the conjugated electron system, which then induces changes in the fluorescence properties (Helmchen, 2012). The binding of a calcium indicator can change some of the parameters that we talked about in the previous paragraphs: absorption and emission spectra, fluorescence yield and LifeTime.

After talking about dyes we can talk about more modern methods that are much more versatile and arguably more powerful. These are genetically encoded calcium indicators. As mentioned before, these are protein indicators. Generally, "GECIs are frequently used in one of two different modes: to track activity in large populations of neuronal cell bodies, or to follow dynamics in sub-cellular compartments such as axons, dendrites and individual synaptic compartments." (Dana et al, 2019). Thus, compared to dyes the general advantages of GECIs are the possibilities of long-term expression and of targeting them to specific subtypes of neurons or subcellular locations (Helmchen, 2012).

There is a very thorough and extensive lecture on Youtube from a pioneering Howard Hughes Medical Insitute scientist-- Tsien, 2013. He talks about all kinds of protein indicators, but here I will try to integrate his material with other material I found online and filter out material that is relevant to the measurement of calcium concentrations. I just have to say: Man, these protein indicators work through mechanisms that are so much more sophisticated than dyes!

The biological literature is permeated with all sorts of protein indicators with their confusing acronyms but which of them bind calcium (e.g. are termed GECIs)? And more importantly: how do they work?

There are two main classes of GECIs: those consisting of two protein molecules and those consisting of a single protein molecule (Helmchen, 2012). In the first case, the main mechanism that is exploited for imaging is FRET, but please don't fret, it's just fluorescence resonance energy transfer. It's a mechanism describing energy transfer between two light sensitive molecules which are called chromophores (Wikipedia). One of them is called the donor and the other acceptor. The donor is excited directly. The acceptor reemits. Energy flows from the donor to the acceptor. What happens is that the donor absorbs a photon but if the acceptor is nearby, it will "steal" the energy and the donor won't emit a photon but the acceptor will. Donors and acceptors have two different colors! The donor is typically the shorter wavelength (higher energy, higher frequency) chromophore. Thus, FRET depends on the distance of the donor and acceptor and their relative orientation. Here's the catch: the two molecules can be bound by a calcium-binding linker that changes their distance when bound. To phrase the previous chat more formally: this distance decrease (to a few nanometers) leads to a decrease in donor fluorescence and an increase in acceptor fluorescence (e.g. the energy transfer) that can be detected and used for reading out information exactly because the two have different colors (Helmchen, 2012). There's also the possibility of irreversibly cleaving the linker by a protease, so that the donor and acceptor drift apart and start fluorescing autonomously. This is akin to one-shot activity detection, so the photons that are emitted before the cleaving are a marker that they used to be together, so it can be localized in time (Tsien, 2013).

From Helmchen, 2012: "The second major class of GECIs comprises single-protein indicators rather than tandem pairs of proteins (Figure 10.1B). Example proteins are Inverse Pericam or members of the GCaMP family. In these indicators, the protein has been modified by insertion of a calcium-binding domain such that calcium-binding leads to a conformational change that either increases or decreases the fluorescence yield of the chromophore (the light sensitive molecule itself)."

Here are some GECI's.

1. Single fluorescent proteins: Classical YFP and GFP proteins that have been created through mutagenesis to monitor things like pH, Cl and also Ca(!) (Tsien, 2013). I couldn't find much about these proteins being used to monitor calcium online,  but the very famous, all-over the place GCaMP (pair of fluorescent proteins and a linker) is engineered from GFP.  GCaMP is created from a fusion of green fluorescent protein (GFP), calmodulin, and M13, a peptide sequence from myosin light chain kinase (Wikipedia). Tsien 2013 has an excellent  A recent Nature paper (Dana et al, 2019) reports: "Using structure-guided mutagenesis and neuron-based screening, we optimized the green fluorescent protein-based GECI GCaMP6 for different modes of in vivo imaging. The resulting jGCaMP7 sensors provide improved detection of individual spikes (jGCaMP7s,f), imaging in neurites and neuropil (jGCaMP7b), and may allow tracking larger populations of neurons using two-photon (jGCaMP7s,f) or wide-field (jGCaMP7c) imaging." An interesting zebrafish connectomics study that used GFP not for calcium imaging, but for tracing neurons is Kunst et al, 2019. This is an example of the last use case of GECIs from the quote in the previous paragraph. Oh, and pulling away from the calcium, pH can also be used to understand synaptic transmission. (Tsien, 2013) has an excellent explanation. Synaptic vesicles are more acidic than the cellular membrane, so if you have GFP in them and they fuse with the pre-synaptic membrane (exocytosis) and thus change environments, the special GFP synaptofluorin will fluoresce (see the lecture for an example involving drosophila and odors-- around 8 minutes):-)

2. Pairs of fluorescent proteins: GCaMP is engineered through modifying GFP (as previously mentioned) through a circular permutation of the DNA encoding it, see Tsien, 2013, around 10 mins into the lecture. Tsien, 2013 goes on to give an example of measuring calcium when parallel Purkinje fibers are stimulated with current, which opens up voltage-gated calcium ion channels and lets calcium into the cell which flashes up intracellular GCaMP indicators. 

I've been working on this post for 7 hours now and I think I'm done! I hope this little compilation of material will help someone navigate the jungle. I've worked a lot with calcium imaging data but in the form of deidentified numbers, so it was fascinating for me to find out how these processes work on a molecular and quantum level. The paper by Helmchen is also very helpful for trying to make sense of how calcium imaging experiments work. This stuff is super sophisticated and cool, fortunately I'm engaged to a molecular biologist:-D

Sources

Helmchen, 2012, "Calcium Imaging", from "Handbook of Neural Activity Measurement", https://www.cambridge.org/core/books/handbook-of-neural-activity-measurement/AA580182A688ED698B6A2768A53B2B55

 Dana et al, 2019, "High-performance calcium sensors for imaging activity in neuronal populations and microcompartments", https://www.nature.com/articles/s41592-019-0435-6

Kunst et al, 2019, "A Cellular-Resolution Atlas of the Larval Zebrafish Brain", Neuron, https://www.cell.com/neuron/fulltext/S0896-6273(19)30391-5

 Harvard course on Fundamentals of Neuroscience, part 2, EdX, videos on Synaptic Plasticity,  https://learning.edx.org/course/course-v1:HarvardX+MCB80.2x+2T2020

Wikipedia 

https://en.wikipedia.org/wiki/F%C3%B6rster_resonance_energy_transfer

https://en.wikipedia.org/wiki/GCaMP

Freely available on Youtube:

Tsien, 2013, Fluorescent protein indicators, Howard Hughes Medical Institute https://www.youtube.com/watch?v=QoRTv2qZXbY 

Stuurman, 2013, Introduction to fluorescence microscopy, Howard Hughes Medical Institute https://www.youtube.com/watch?v=AhzhOzgYoqw