Metal complexes throw new light on cells

Fluorescence microscopy is useful for imaging the internal structures within cells. This technique mostly uses organic molecules as the light-emitting agents, but recently heavy-metal complexes have started to be applied. The wavelengths and lifetimes of the emissions from metal complexes make it easy to separate the signal from the noise of natural fluorescence. Metal complexes can also act as sensors of their chemical environment, and they can enable concurrent use of other imaging modalities. The toxicity of heavy metals is tamed by using very stable molecular structures, but controlling the uptake of metal complexes into cells remains a challenge that is solved one case at a time.

Fluorescence and optical image of cells
Figure 1. Fluorescence microscopy (top) and ordinary optical microscopy (bottom) images of human adenocarcinoma cells. In the fluorescence image, an organic stain emits green light and a rhenium complex emits red light. The bright orange dots are nucleoli, where both agents have their highest intensity. (Source: Michael Coogan.)

In biology and medicine, fluorescence microscopy is used  to reveal subcellular detail of tissue and organisms. The cells are stained with chemical agents and then irradiated with light to excite ‘fluorophores’ in the agents to emit their own characteristic light (see Figure 1). Multicolored images can be generated by using multiple fluorophores with different properties. In addition to imaging the structures within cells, intracellular chemical sensing is possible because some fluorophores respond to their chemical environment (e.g., pH) by changing the wavelength (color) of light emitted. Although other imaging techniques, such as MRI and computed tomography, allow deeper tissue penetration and whole body scans, they lack the resolution to look inside the cell and at its structures and biochemistry on this scale. Ideally, a single chemical can be active in both fluorescence imaging and another technique, allowing bimodal imaging, combining deep penetration and subcellular detail.

Most fluorophores currently used are organic molecules, but these often have emission properties similar to the autofluorescence produced by naturally occurring chemicals in cells. This problem has led to a search for other systems with emission properties that will allow easy separation of signal from noise. Researchers have employed two allied approaches. First, one can change the fluorophore’s Stokes shift (the difference in wavelength between absorbed and emitted light) in order to make the emission signal easy to distinguish from autofluorescence by wavelength filtering. Second, one can increase the luminescence lifetime of the agent so that signal is emitted long after autofluorescence has died away, allowing clear images to be obtained.

Certain heavy-metal complexes, particularly those of late transition metals from the second and third rows and the lanthanides, have large Stokes shifts and long lived luminescence (hundreds of nanoseconds to milliseconds, compared to less than 10ns for most metal-free systems), but they have been applied in fluorescent cell imaging only relatively recently. The development of metal complexes for fluorescence imaging has required overcoming problems of cytotoxicity, solubility, cell uptake, and stability, and it has opened up possibilities for advanced techniques, such as chemical sensing and bimodal imaging.

The use of metal complexes as fluorophores in cell imaging has been widely reported,[1][2] but recently applications have emerged that showcase their advantages. A prime example is work by the group of Gareth Williams of the University of Durham, which did a time-resolved study using an iridium complex that is also a pH sensor.[3][4] Under acidic conditions (low pH), a nitrogen in the complex becomes protonated, causing a change not only in emission wavelength, but interestingly also in luminescence lifetime (840ns when protonated, 3200ns when basic). These long lifetimes allowed cell images to be recorded in a co-staining experiment using the iridium complex together with Hoechst 33342, which stains nuclei and has a short luminescence lifetime (3.6ns). Usually co-staining experiments require differentiation between agents by wavelength, but in this case the lifetimes were exploited. The emission from the Hoechst was obtained with constant irradiation. When a 10ns delay was introduced between irradiation and image recording, however, only the iridium-based emission was observed because all of the signal from the Hoechst had died away. This work represents the first example of co-staining by lifetime, and the first lifetime study of a pH sensor based on a transition metal.

This study also exemplifies the techniques used to overcome problems of applying metal complexes in biology, such as toxicity and cell uptake. The toxicity of heavy metals is usually associated with the interaction of the heavy-metal ion with biological molecules such as DNA. The iridium complex used by the Williams group, however, is a stable complex in which the metal ion is encapsulated by ligands, preventing the iridium from interacting with biological material. As long as the complex remains intact, the metal is detoxified. Furthermore, the complex is very stable thanks to its electronic nature—the metal ion is of a type called d6 low spin, meaning its outermost six electrons occupy just three of the d orbitals, with their spins paired up for a total spin of zero. The complex is also cationic (overall positive charge) and lipophilic (i.e., oily), which assists cell uptake. An electronic potential that exists across cell membranes encourages cations to accumulate within the membrane, and, because cell membranes are constructed from lipids, diffusion through them requires lipophilic properties.

Trimeric metal cage
Figure 2. Molecular cage built from three copies of a rhenium (yellow atoms) complex, which can hold another metal ion such as silver or copper (large white atom). Imaging of the rhenium’s fluorescent emission could be combined with other modalities based on the caged ion. Blue: Carbon. Green: Nitrogen. Red: Oxygen. Gray: Hydrogen. (Source: Michael Coogan.)

We have recently taken advantage of the same principles, applying a d6 low-spin (i.e., stable and therefore nontoxic), lipophilic complex, this time of the transition metal rhenium.[5][6] Unlike most of the complexes used in imaging, which have only one metal ion, our agent is a trimer of metal complexes, which forms a triangular cross-sectioned tube and acts as a cup or cage to hold another metal ion inside (see Figure 2). In this way, the agent can transport ions of the 64Cu isotope of copper, which is a positron emission tomography (PET)-active isotope, into cells, enabling bimodal PET/fluorescence imaging. The system switches between the neutral, empty cage, which is not taken up by cells, and a cationic loaded form, which is taken up well. Furthermore, colocalization with a known stain shows that the cationic form accumulates in nucleoli (see Figure 1), a completely fortuitous and very interesting outcome. Radiotherapy using 64Cu could be helped by localizing the radioisotope in the cells’ nucleus, assisting DNA damage aimed at killing cancerous cells.

A more complicated system of cell imaging agents involves complexes of the lanthanides, or rare earth metals, which require organic molecules to act as antennas to absorb light, and large ring systems as ligands. Lanthanides do not form stable complexes with simpler ligands, so the ligand system must wrap around the central ion with multiple points of attachment to stabilize the complex and prevent the loss of the toxic metal ions. Lanthanides can emit with very long lifetimes and in the near-infrared (NIR) region of the spectrum. NIR light has better tissue penetration than visible light, which can allow whole-body imaging, at least in small mammals. Many important ions and molecules in the cell will interact with the lanthanide ion in the complex and subtly change the properties of the emitted light, making these complexes good sensors of their environment.

Recently, David Parker’s group from the University of Durham developed a lanthanide complex that localizes in the mitochondria of human cells and responds to the bicarbonate ion with an increase in brightness.[7][8] Bicarbonate is linked to production of carbon dioxide by oxidation of glucose in the mitochondria, so this imaging agent allows this vital cellular process to be monitored as it happens. The intensity of the fluorescent signal, however, depends not only on the concentration of bicarbonate, but also the concentration of the imaging agent. The researchers therefore developed two complexes with identical ligands but different lanthanides (europium and terbium) to calibrate the system. Because the red-emitting europium complex is sensitive to bicarbonate, whereas the green-emitting terbium analogue is not, the ratio of the emission of each color serves to monitor changes in bicarbonate levels.

The large Stokes shifts of metal complexes have proven to be very useful in separating fluorophore emission from autofluorescence, with a large number of complexes based on mid- to late-transition metals and lanthanides giving excellent signal-to-noise ratios. Lifetime gating is also proving to be extremely powerful for separating signal from background autofluorescence, and for separating short- and long-lived luminescence from different fluorophores. More studies will undoubtedly use this approach now that it has been shown to be successful. Heavy-metal toxicity has been addressed through techniques such as the use of stable d6 low-spin complexes. In general, the ligands prevent loss of metal ions, taming the toxicity. However, the techniques used to control the biological properties of metal complexes have been less successful. Some complexes are not taken up well by cells, and for others it is difficult to control where the complexes go once they have entered the cells. So far, there have not been enough systematic studies to draw conclusions about controlling uptake to cells and behavior once inside cells, except for a few specific metal-ligand systems. More of these systematic studies are required if progress at controlling biological behavior is to be more than trial and error.

Now that the principles of using metal complexes in cell imaging have been established, and studies are turning to real-life applications, practical issues must be addressed for metal complexes to find a useful role. Many organic molecules are simply brighter than most metal complexes, and a huge range of organics is available for imaging every cellular region. Therefore, I doubt that metal complexes will replace organics as everyday imaging agents. However, I expect that the ability of metal complexes to change their emission properties in the presence of other molecules and ions will ensure their future in more specialist applications. Time-resolved methods exploiting the long fluorescence lifetimes of metal complexes also have great potential, but they might have limited popularity if the complicated imaging equipment that they require does not become cheaper and more common. Still, long lifetimes can also be used in simpler imaging techniques in which one fluorophore acts as an energy donor to measure distance to an acceptor. A donor with a long lifetime can move around more during the experiment than if the lifetime is short. Other potential specialist areas for metal complexes include NIR imaging and bimodal imaging, in which the same metal can be used for fluorescence imaging together with radio or magnetic resonance imaging.


  1. V. Fernández-Moreira, F. L. Thorp-Greenwood, and M. P. Coogan, Application of d6 transition metal complexes in fluorescence cell imaging, Chem. Commun. 48, pp. 186–202, 2010. doi:10.1039/B917757D.
  2. F. L. Thorp-Greenwood, An introduction to organometallic complexes in fluorescence cell imaging: current applications and future prospects, Organometallics 31, pp. 5686–5692, 2012. doi:10.1021/om3004477.
  3. S. W. Botchway, M. Charnley, J. W. Haycock, A. W. Parker, D. L. Rochester, J. A. Weinstein, and J. A. G. Williams, Time-resolved and two-photon emission imaging microscopy of live cells with inert platinum complexes, Proc. Nat'l Acad. Sci. USA 105, p. 16071, 2008. doi:10.1073/pnas.0804071105.
  4. L. Murphy, A. Congreve, L.-O. Pålsson, and J. A. G. Williams, The time domain in co-stained cell imaging: time-resolved emission imaging microscopy using a protonatable luminescent iridium complex, Chem. Commun. 46, p. 8743, 2010. doi:10.1039/C0CC03705B.
  5. F. L. Thorp-Greenwood, V. Fernández-Moreira, C. O. Millet, C. F. Williams, J. Cable, J. B. Court, A. J. Hayes, D. Lloyd, and M. P. Coogan, A ‘sleeping Trojan Horse’ which transports metal ions into cells, localises in nucleoli, and has potential for bimodal fluorescence/PET imaging, Chem. Commun. 47, p. 3096, 2010. doi:10.1039/C1CC10141B.
  6. V. Fernández-Moreira, F. L. Thorp-Greenwood, A. J. Amoroso, J. Cable, J. B. Court, V. Gray, A. J. Hayes, et al., Uptake and localisation of rhenium fac-tricarbonyl polypyridyls in fluorescent cell imaging experiments, Org. Biomol. Chem. 8, p. 3888, 2010. doi:10.1039/C004610H.
  7. D. G. Smith, G. Law, B. S. Murray, R. Pal, D. Parker, and K.-L. Wong, Evidence for the optical signalling of changes in bicarbonate concentration within the mitochondrial region of living cells, Chem. Commun. 47, pp. 7347–7349, 2011. doi:10.1039/C1CC11853F.
  8. C. P. Montgomery, E. J. New, R. Pal, and D. Parker, Cell-penetrating metal complex optical probes: targeted and responsive systems based on lanthanide luminescence, Acc. Chem. Res. 42, p. 925, 2009. doi:10.1021/ar800174z.