Optical imaging

Broadly speaking, optical imaging can be divided into:

• fluorescence imaging
• bioluminescence imaging

The instrumentation required is largely identical for both methods, with three central components:

1. light source 
2. filters to allow selection of the appropriate wavelengths 
3. detection system 

While both techniques rely upon detection of photons (light) generated selectively in the cells or tissue of interest, the mechanisms by which this light is produced differ.

Fluorescence imaging

Fluorescence is a physical process in which absorption of light at a specific wavelength stimulates the formation of a high energy, excited state in a susceptible molecule.

This high energy excited state is energetically unfavourable. For the excited molecule to re-enter its energetically favourable ground state, energy must be released. 

This energy may be released in a number of ways, but the most common way for an excited state to release energy and revert to the ground state is through emission of a photon. 

Crucially, the emitted photon is at a longer wavelength than the absorbed photon, allowing distinction between the exciting and emitted photons with relative ease.

Energy level diagram for fluorescence

Fluorescence occurs naturally in a range of biological materials, including NADH, FAD, porphyrins, collagen, and elastin.

Fluorescence from these endogenous materials is termed tissue autofluorescence. Tissue autofluorescence is of limited use in fluorescence imaging as it is seen in all tissues, i.e. is non-specific, and when observed is usually treated as an undesirable interferant.

High specificity imaging is achieved either through the transfection of target cells with reporter genes that code for optical contrast agents or through the selective introduction of exogenous optical contrast agents.

The most common reporter genes code for naturally occurring fluorescent proteins or versions of these proteins that have been engineered to improve their fluorescence characteristics (for example to increase the quantum yield or to tune the emission wavelength to a more desirable spectral region).

These naturally occurring fluorescent proteins include green and red fluorescent proteins (GFP and RFP respectively).

For in vivo imaging studies, the choice of reporter gene is governed by a number of factors, including availability of vectors, ease of transfection, efficiency of transcription and in vivo quantum yield, excitation wavelength and emission wavelength.

In addition to transfection with reporter genes, high specificity fluorescence imaging can be achieved through the injection of reporter molecules.

Reporter molecules may be macromolecules such as antibodies, or low molecular weight materials such as small peptides, oligonucleotides or carbohydrates labelled with organic dyes or inorganic quantum dots.

The only requirements for a reporter molecule are:

1. It can be labelled with a fluorescent material
2. It selectively reports on the property of interest

Bioluminescence imaging

Light is produced biochemically due to the presence of an enzyme that has been artificially introduced into the cells/tissues of interest.

This type of imaging makes use of the light-generating properties of the enzyme luciferase. Luciferases may be isolated from a variety of sources including fungi, bacteria, insects, and marine invertebrates. The substrates, and so light-producing mechanisms, vary by species.

For example, firefly luciferase catalyzes the oxidation of D-luciferin to oxyluciferin in a multistage, ATP dependent reaction. It is important to note that as the reaction catalyzed by firefly luciferase requires ATP, firefly luciferase can be used to distinguish between living and dead cells.

Renilla luciferase catalyzes the oxidation of coelenterazine to coelenteramide. Bacterial luciferases catalyze a reaction in which a flavin mononucleotide oxidizes a long chain aliphatic aldehyde to an aliphatic carboxylic acid.

As with red and green fluorescent proteins, luciferase is not expressed in mammalian tissues, and bioluminescence studies therefore require incorporation of the gene for the appropriate luciferase into the target system.

Anti-tumour agents

One of the most common uses of optical imaging is preclinical studies of anti-tumour agents. In the example shown here, fluorescence imaging was used to study the binding of a novel antitumour antibody, NovoMab-G2-scFv, to tumour cells. NovoMab-G2-scFv was labelled with Cy5.5.18, a fluorescent dye with an absorption maximum at 670 nm and emission maximum at 720 nm.

This figure shows visible 670 nm and 720 nm images obtained two hours after injection of NovoMab-G2-scFv-Cy5.5.18. Fluorescence from the tumour site can clearly be seen in the 720 nm image, confirming that the antibody does indeed bind to tumour cells. In addition, strong fluorescence is seen in the kidneys due to non-specific binding of the antibody. Visible (A), 670 nm (B), and 720 nm (C) images of a tumour-bearing nude mouse two hours after injection of NovoMab-G2-scFv-Cy5.5.18. Exposure time: 60 seconds.

The non-invasive nature of fluorescence imaging coupled to the brief time it takes to acquire images (seconds to minutes, depending upon signal strength) makes it ideally suited to pharmacokinetic studies.

This example shows fluorescence images from a tumour-bearing mouse acquired 2, 12, 24 and 48 hours post injection of NovoMab-G2-scFv-Cy5.5.18. 

Fluorescence intensity through the region of greatest fluorescence intensity for each image is plotted below each image. Fluorescence clearly increases to a maximum value at 24 hours post-injection, with the antibody almost eliminated from the tumour site at 48 hours post injection.

Images acquired at 720 nm from a tumour-bearing mouse 2 (A), 12(B), 24 (AC) and 48 (D) hours after injection of NovoMab-G2-scFv-Cy5.5.18.

Fluorescence intensity at each pixel on the line through the region of greatest fluorescence intensity for each image (shown in white) is plotted below each image.

A more robust kinetic analysis may be obtained by drawing a region of interest (ROI) around the tumour site and dividing the integrated fluorescence in the ROI by the number of pixels in the ROI to generate mean fluorescence per pixel.

This normalizes data with respect to tumour size and allows data from multiple mice to be pooled and analyzed.

A. Mean fluorescence intensity per pixel in the region of interest (ROI) defining tumours as a function of time after injection.

An exponential decay of total fluorescence with time is observed, decreasing to approximately 5% of the fluorescence observed TWO hours after injection at 72 hours after injection.

The exponential fit to the data is described by Y=417.22e-x/6.107 + 61.124

Based on this fit, the calculated half time of the antibody-dye complex at the tumour site is 10.3 hours.