Thursday, 17 November 2016

The origins of photoluminescence

The first reports of a green fluorescence associated with Aequeoria victoria are now over 50 years old. As a boy, Professor 下村 脩 (Shimomura Osamu) was living close to Nagasaki, when the atomic bomb devastated his city. After a very short period of blindness, he was covered in radioactive ash, but survived and went on to gain a degree in Chemistry and Pharmacology. In his laboratory work on the unusual glow produced by certain marine organisms, he went on to isolate and characterise two fluorescent proteins aequorin and green fluorescent protein (GFP). The remarkable glow of from these ocean going crustaceans, received the same level of wonder as the glow of the firefly, or the "jaws" of the Venus Fly trap. But what is fluorescence and who was the first to observe the phenomenon? 

In the middle of the 16th century, a Spanish physician, Nicolás Monardes and a Franciscan missionary, Bernardino de Sahagún, wrote of the "strange blue glow from a small Mexican tree" which was shortly after named Lignum Nephriticum (kidney wood) by the Botanist Charles de L’Écluse (the rather dashing, Black Adder like figure, shown in the engraving on the right). These were the first documented observations of fluorescence: but nobody had any idea what caused this phenomenon. Around 50 years later in 1612, Galileo Galilei, wrote of a similar phenomenon emitted by the Bolognia stone (probably Barium sulphate), but this turns out to be phosphorescence (of which more in another post).

I have written before about the molecular structure of GFP, and its applications in contemporary Biology, so here I thought I would say a few things about the Physics behind fluorescence. The word is derived from the rock that glows: a mineral form of calcium fluoride called fluorspar, shown on the left. 

Fluorescence and phosphorescence are two forms of what is called photoluminescence. From its name you can work out that it has something to do with light (and therefore photons) and, well yes, light again (lumen is Latin for light). So light that produces more light! Actually that's a pretty good description of fluorescence. Solutions of transition metal salts such as copper sulphate, or textile dies like indigo are coloured because they have extracted a discreet amount of energy from the incident light (the classic spectrum of the rainbow) leaving behind (in both these cases) a blue colour. The Physics of the process is explained by the transfer of energy from a photon of light to an electron, leading to an increase in thermal energy in the molecule, together with a reduction in the intensity of the emitted light wave. Absorption is a manifestation of the quantisation of energy, in this case the energy associated with a photon of light.

Fluorescence is not dissimilar to absorption, except that fluorescent molecules emit light (or in general, electromagnetic radiation) following absorption of a photon of light. In most cases, the emitted light has a longer wavelength, and therefore lower energy, than the absorbed radiation. The key difference in the two processes is the pathway followed by the energy within the fluorescent molecule. In the first of three stages, called excitation, a photon of energy is absorbed by the fluorophore creating an excited singlet state. [An explanation of the term singlet is not straight forward. It has a quantum mechanical definition which I understand as a set of linked or connected particles possessing a net angular momentum of zero (angular momentum is a quantum phenomenon and one that is key to explaining Nuclear Magnetic Resonance, NMR). However, this probably doesn't help many of you, without first year undergraduate Physics, so I suggest you accept, for now, that the term singlet is used to describe a set of particles produced by (in this case) electromagnetic irradiation, that match (or show a level of coherence with) one of the wavelenghts of the visible spectrum that gives a colour, such as blue, red, green etc]. In stage two, this excited state has a specific lifetime (typically around a few nanoseconds). During this time the electrons in the body of the molecule undergo a reorganisation, which is often influenced by the molecular environment (such as the solvent or neighbouring amino acid side chains in a protein such as GFP). It is at this stage that energy may be dissipated, yielding what is referred to as a relaxed singlet state, from which fluorescence emerges. Typically, only some of the excited molecules in the first stage return to the ground state by fluorescence, since other pathways exist, such as  quenching by molecular collision or the transfer of energy through a process called Fluorescence Energy Transfer, FRET, but these mechanisms require more space for a proper explanation. However, I should say that FRET is a very powerful phenomenon that has been applied very effectively in GFP technology. By now, you have probably worked out that the greater the number of these and other related mechanisms of energy dissipation, the lower the fluorescence energy output: which is known by the term quantum yield. A strongly fluorescent molecule has a high quantum yield and is often said to be quantum efficient. For the physicists among you, the quantum yield is the ratio of the number of fluorescence photons emitted to the number of photons absorbed.

The third and final stage in fluorescence is emission, after which the fluorescent molecule is returned to the ground state. As a consequence of the processes described in stage 2, the energy of the emitted photon is lower than the absorbed photon, and this is reflected (no pun intended) in the reduced wavelength of the emitted photon of light. Recall the standard equation that relates energy to the source of light (or electromagnetic radiation)

Again for the physicists among you, the difference in the two wavelengths (or energies) is called the Stokes shift, after the Irish physicist, George Stokes. This is the parameter that is measured by fluorescence detectors, in contrast, absorption spectrophotometers measure the difference in the intensity between the incident and transmitted light at a particular wavelength (which you will all be familiar with).

A few words now about a couple of fluorophores that you will come across in Life Sciences: GFP and intercalating dyes like ethidium bromide (shown left). The former is used mainly in cell biology and the latter in molecular biology. In years to come, I have no doubt ethidium bromide and GFP will be replaced by safer reagents with a higher quantum yield, and then by completely different biophysical technology. However, the principles will be very similar. In the case of ethidium bromide, this deep purple, aromatic molecule absorbs a photon of light in the near UV region (around 310nm) and ONLY in the presence of nucleic acid polymers (DNA and RNA) does it fluoresce. You will have read above that the phenomenon of fluorescence is influenced by the environment of the fluorophore. Agarose gels containing DNA of different sizes are revealed beautifully, when UV light is used to illuminate the gel that has been soaked in a weak solution of ethidium bromide. The dye intercalates between the bases in the DNA and its environment is now suitable for fluorescence. If there is no DNA, there is no intercalation and therefore no fluorescent event.

In the case of GFP, visible light, which is of a lower energy than UV light, is sufficient to induce singlet formation. And by the way, for those of you contemplating the use of sun beds, I would think seriously about the potential damage to your skin cells by prolonged exposure to UV light! Getting
back to GFP, a cluster of amino acids in the core of the basket shaped molecule absorb a photon of visible light, leading to singlet formation. When the molecule of GFP returns to the ground state, it does so by emitting light with a bright green fluorescence. Interestingly, a number of labs have managed to a retune this fluorescence, by modifying the amino acid side chains around the fluorophore, giving us yellow and red versions of GFP (see the gallery of GFP variants above, right)! As you might expect, Nature had already beat us to it, and recall that the discovery of GFP went alongside the discovery of aequorin, which is a Blue Fluorescent Protein. 

Finally, I hope you all enjoy the next few weeks and I look forward to Dr Dyer's stories of successful purification of your GFPs!


  1. What is the energy cost for the microorganism to produce "light"?

    And what is the benefit gained by such mono-celluar organisms by making themselves "Shiny"?

    1. An excellent question. The light emitting organisms that we so often see on programmes like David Attenborough's classic series range from microbes, marine creatures, some plants and insects. The two most popular light producing mechanisms are fluorescence and phosphorescence. The GFP protein does not require any biochemical energy input to fluoresce, but firefly luciferase, for exmplae utilses ATP in a two step reaction:

      luciferin + ATP → luciferyl adenylate + PPi
      luciferyl adenylate + O2 → oxyluciferin + AMP + light

      Therefore while GFP does not draw on the energy reserves of Aequoria victoria, the luciferases do and therefore the selective advantage of light emission comes at a price.

    2. Why glow in the dark? I can only assume that light emission serves as an attraction for those species that mate or attract other organisms for another purpose. Equally, light emission could send out a warning to deter predators. Maybe they could light the way for others through a symbiotic relationship? or a defense