Side nav buttonsContactFundingResourceFAQAbout


Research - Second Harmonic Generation



We have been developing the use of Second Harmonic Generation (SHG) as a new imaging nonlinear optical modality for obtaining high-resolution images of tissues including skin, bone, tendon, and muscle. This method is related to the more common modality of multiphoton excited fluorescence microscopy and allows for intrinsic optical section deep within tissues with minimal adverse phototoxic effects. However since the SHG contrast arises from the endogenous proteins themselves, it can provide more molecular and structural information on the assembly in tissues. Our long term goals are to use this imaging modality as a clinical diagnostic tool to differentiate normal and diseased tissues.

Overview of collagen-based tissues

Type I collagen is the most abundant protein in our bodies and comprises the matrix in bone, skin, and tendon. The figure shows representative SHG images of these tissues.

Many connective diseases including Osteogenesis imperfecta (brittle bone disease) and scleroderma arise from mutations in the collagen sequence. We are working on methods that compare the collagen fibril/fiber structure in these diseases. In these efforts, we compare the size of these structures, the resulting SHG intensity, the polarization anisotropy, as well as the scattering properties in the collagen matrix.


Overview of SHG muscle imaging

We are also using SHG imaging to examine the muscle sarcomere structure in an effort to understand the morphological changes that happens to our muscle tissue during aging as well as in diseases including muscular dystrophy. A representative SHG optical section of muscle is shown in the figure. Unlike collagen-based tissues where collagen is essentially the only component, muscle is comprised of acto-myosin complexes which also contain a large number of other proteins as well. A fundamental issue is identifying the SHG generating component.  We examined this question by antibody labeling and selective removal of the actin and myosin species. The image of myofibrils shows the result when myosin is extracted by addition of salt, where the left panel is the combined two-photon fluorescence of the actin-phalloidin (green) and SHG (violet) before salt extraction and the right panel shows that only the green channel remains after treatment,  proving that the myosin is essential for the SHG contrast.  

An additional aspect to be considered is that the myosin molecule has both a coiled-coil tail domain as well as two heads that are oriented at45 degrees with respect to the tail. It is important to identify which domain(s) give rise to the SHG contrast. We have examined is by inducing contraction of myofibrils and comparing the before and after SHG images. The B panel shows rapid line scan imaging of the contraction, and we observe no change in the image intensity. Since contraction only involved the myosin heads and there is no effect on the SHG image we conclude that the contrast arises from the myosin tails.    


Increasing imaging depth by optical clearing

A limiting aspect of tissue imaging is the achievable imaging depth. Most tissues have scattering lengths (1/e point) of about 100 microns. For clinical applications, we need to image deeper into tissues.  To this end, we have been treating tissues with hyper-osmotic agents such as glycerol to achieve optical clearing. In the 1950’s Huxley showed that myofibrils could be extracted with treatment of 50% glycerol solution. Here we use this idea and obtaina  2.5 fold increase imaging depth in muscle. The figure shows SHG  x-z projections of ex vivo muscle at timepoints from 10 minutes -24 hours.  We have performed measurements showing that the molecular level structure as well as the sarcomere periodicity are not effected by the treatment, indicating it is an effective, non-invasive method to significantly increase imaging depth.