Did your images impress the audience during your latest presentation? Help secure your latest grant or solidify your tenure review? If not, did you know that in high resolution (not super-res) imaging, it’s image contrast, NOT resolution, that is often the difference between an image that sticks with the audience or garners a journal cover and one that is soon forgotten? Considering that out-of-focus light is one of the factors that can significantly decrease contrast in fluorescence imaging, most of the high resolution microscopy techniques available are based upon ways to manage that out-of-focus light and increase image contrast. For example:
- Confocal microscopy techniques use a physical barrier (a pinhole) to reject out-of-focus light such that out-of-focus fluorescence never makes it to the detector, resulting in an image with increased contrast.
- Deconvolution is a computational technique often used in conjunction with widefield microscopy. With this method, all in-focus and out-of-focus fluorescence is collected. Then a deconvolution algorithm is used to restore blurred light to its original point source thus increasing contrast.
- Total Internal Reflection Fluorescence (TIRF) microscopy uses an optical technique to limit the excitation illumination to a very thin section at the coverslip. Since TIRF does not excite fluorophores more than ~200nm into the sample, contrast of structures very close to the coverslip is increased.
Since contrast is critical to success in all modes of microscopy, it is essential to understand some of the basic principles of light microscopy to maximize image contrast.
Principles of Light Microscopy
If a perfect lens existed, all the light passing through that lens would converge to a single focal point on a central axis. However, when light rays pass through a convex lens, of which there can be dozens in a microscope light path, the rays near the edge of the lens are refracted, or bent, to a higher degree than those rays that pass through the center of the lens. As a result, the rays from the edge of the lens converge to a focal point that is closer to the front of the lens than those rays that pass through the center of the lens. This property of light is called spherical aberration (SA) and it leads to increased blur through the optical path and, therefore, decreased contrast in image data.
To measure the blur in a given microscope light path, one can image a sub resolution ~100nm fluorescent latex bead. This measurement is known as the Point Spread Function (PSF). Viewing the PSF, as in Figure 1 below, shows how the fluorescence from the bead spreads out and widens in concentric circles, called airy disks, as focus moves further from the bead in either direction. The size and shape of the PSF is affected by every part of the optical path including a number of sample specific properties and environmental conditions.
When left uncorrected, spherical aberration (SA) produces Z-axis asymmetry within the shape of the PSF resulting in a loss of signal intensity and resolution. The objective lens can introduce this type of aberration but most high quality objectives used today are very well corrected for SA. In practice, the most common cause of SA is a mismatch between the refractive index (RI) of the objective lens immersion oil and the mounting medium of the biological specimen.
The RI is a numerical value without units and is determined by a material’s physical and optical properties. The RI characterizes the extent to which light is refracted, or bent, when it passes through that material. For example, air has an RI of 1.0 which makes sense if we consider that clean dry air does not bend light when it travels through it. Glass, and therefore most microscope slides and coverslips, has an RI of 1.515. Not coincidentally, many of the available commercial immersion oils also have an RI of 1.515, however, this may not be optimal to fully or properly correct SA for your specific sample and microscope environment. So how do we correct SA?
Experimental Factors Affecting Spherical Aberration and the PSF
Coverslips are available in a range of thicknesses and numbers from No. 0.0 (~100µm) to No. 4.0 (~500µm). The correct coverslips to use for high resolution microscopy are No. 1.5 (170µm). Confusingly, No. 1.0 coverslips (150µm) are readily available for purchase but should not be used for high resolution imaging. All commercially available objective lenses are corrected for coverslips that are 170µm thick, as specified by the 0.17 marking on the barrel of the objective. Even on a well-aligned microscope, using a coverslip with the incorrect thickness will introduce significant SA into the system. Optimal contrast requires adjusting for those variations by moving away from RI 1.515 oil.
The RI of any immersion oil is specified for an optimal working temperature. Many of the commercial immersion oils are designed to work at an average room temperature of 23°C. Temperature variation in the microscope environment will affect the viscosity of the immersion oil which will then change the effective RI of the oil. If you routinely image at a temperature other than 23°C, for example 37°C for live cell imaging, it is important to consider the actual temperature when selecting the correct immersion oil RI.
Temperature fluctuation within the microscope working environment can also impact imaging results. Changes of as little as 1°C during your imaging session can be enough to notice a shift in the amount spherical aberration present in your images. To properly adjust for every 3°C change in temperature, you should adjust the immersion oil RI by 0.002 in the same direction your temperature moved (i.e. go up in oil RI for increase in temperature). Additionally, if you are planning a long-term live-cell imaging experiment or a long imaging session of many different samples, temperature stability is an important consideration for optimal imaging.
The specific wavelength of light affects how it travels through optical components, like an objective lens, and different wavelengths will exhibit slight variations in SA and shape of the PSF. In addition to the SA, resolving power is also related to the excitation wavelength used to image a sample – the shorter the wavelength of light, the greater the resolution we can expect. The shortest wavelength used in light microscopy is the near ultra-violet, giving the highest resolution. Knowing both of these relationships to wavelength, a scientist should consider choice of wavelength when designing experiments as well as when finalizing their selection of an immersion oil RI for the microscopy.
Depth into the Sample
Most standard calculations of SA are done imaging directly at the coverslip so they place the symmetric PSF at the coverslip. This is ok for very thin samples, but if you would like to image part of a biological sample 5, 10 or even 20 µm away from the coverslip, you will see significant spherical aberration at those depths. Therefore, in practice, it is important to consider how far into the sample (in Z) you need to image to ensure your PSF will be focused and symmetric in the middle of your sample.
RI of the Mounting Medium
There are many types of sample medias utilized in microscopy depending on the specific biological application. Live samples remain in an aqueous media while many fixed samples are mounted in a glycerol based media. The RI of these aqueous mountants, glycerol-based mountants, and biological specimen all differ, and most significantly, they differ from the RI of the coverslip. Live cell imaging, for example, requires culture medium, water, or PBS with an RI of 1.33, while fixed samples utilize mounting mediums with RIs ranging from 1.45 to 1.60. These variations in mounting medium RI will affect how light is refracted at the coverslip/sample interface therefore the sample mountant RI must be accounted for when choosing an appropriate oil RI to minimize spherical aberration. All of these variations will introduce and influence the amount of SA visible in your images.
Why is it important to choose the correct immersion oil?
As discussed above, there are many factors that contribute to how light travels through the optical path of the microscope and the sample, and each of these factors will affect the level of SA introduced in different ways. If we consider all these factors when choosing an immersion oil with a well matched RI, we can minimize SA in our imaging, and produce raw data with significantly improved contrast.
In addition to improving image contrast, minimizing SA is important if you plan to deconvolve your images. An integral part of most deconvolution algorithms (with the exception of blind deconvolution algorithms) is a model of the imaging PSF. This model PSF assumes that your imaging PSF does not exhibit significant spherical aberration. For best results after deconvolution, you should do everything possible to make sure you minimize spherical aberration in your raw data, thus matching the experimental PSF to the model PSF as closely as possible.
How the GE Healthcare Immersion Oil Calculator helps you choose the correct immersion oil
Taking into consideration all of the variables from sample preparation to objective characteristics to environment can be overwhelming. One thing should begin to stand out about adjusting the RI of the oil you are using; it’s the only adjustment you can make in a matter of seconds! Finding the right immersion oil for your unique imaging application and microscope has never been easier. With the free Immersion Oil Calculator app from GE Healthcare, you simply enter your experimental details and it will automatically generate a recommended RI oil to use!
There are drop-down menus for all of the following:
- Objective working distance. A choice of six common objectives and their respective working distances are provided, as well as the option to input a custom working distance.
- Coverslip thickness. As discussed previously, all oil immersion objectives are optimized for #1.5 (170 µm thick) coverslips, so those should be used whenever possible. However, the app contains entries for #1.0 (150 µm) coverslips and a custom thickness field as well.
- Specimen RI. In this field you should enter the RI of the mounting medium. Five very common mountants are preloaded in the dropdown for convenience and a custom option is available.
The Immersion Oil Calculator also takes into account both temperature and excitation wavelength with easy to use sliding scale bars. The temperature range is from 15°C to 50°C and the wavelength from 390 nm to 650 nm; both cover practically all applications for fluorescence imaging of biological specimens. The App also takes into consideration sample thickness by asking for the typical distance between the coverslip and the middle of your specimen, or in other words, half the thickness of your specimen.
Once you have selected all of the parameters of your experiment, the App will calculate the recommended immersion oil RI for the specific conditions you entered.
Using this App will take into consideration all of the parameters above, ensuring that spherical aberration in your imaging is minimized, thus significantly increasing contrast of your raw data!
The App is freely available on a number of platforms including a Web App, an Android App (available at Google Play) and an iPhone/iPad App (available at the iTunes App Store).