Second-harmonic generation (SHG) microscopy brings the imaging of select biomolecular assemblies into 3D focus. It highlights biomolecules that follow specific structural organization within biological tissues, and complements two-photon excited fluorescence (TPEF) microscopy, as both imaging modalities can operate simultaneously on the same laser scanning microscope. Although SHG and TPEF have a lot in common as microscopy techniques, there remain some key differences in their fundamental emission properties. Users must be mindful of such differences between these techniques to ensure successful and effective imaging. Understanding SHG SHG is a process in which two identical photons within a material are converted into a single photon without losing energy. Conservation of energy during this process means the frequency of the new photon is exactly double the frequency (i.e., a second harmonic) of the original photons, and the wavelength is exactly half. For example, incident light at an 800-nm wavelength will be converted by SHG to 400-nm emissions. Collagen-rich dermis in chicken caruncle (500 × 500 µm). Second-harmonic generation (SHG) at 520 nm from the collagen in the dermis is captured simultaneously in two-photon excited fluorescence (TPEF) (autofluorescence, 575 to 647 nm) from the epidermis with an excitation wavelength of 1040 nm. Courtesy of Olympus America Inc. Nonlinear scattering effects such as SHG are driven by the interaction between incident photons and induced charge polarization in a material. In turn, polarizability is determined by the underlying structure of the material at size scales comparable to the incident photon wavelength. Second harmonics are generated if the material possesses a noncentrosymmetric structure. Noncentrosymmetric — lacking inversion symmetry — means that reversing the structure along the three spatial axes would result in a different form. Helical structures, for example, are noncentrosymmetric because they change in handedness from right to left (or vice versa) if the X, Y, and Z axes are reversed. Several biomolecular assemblies satisfy the structural conditions for SHG, including starch granules, collagen fibers, myosin filaments, and microtubules. SHG microscopy A powerful technique, SHG microscopy highlights specific structures in biological specimens without external labels such as dyes and fluorescent proteins. It can be used with live cells and tissue with little or no preparation, but is also applicable to fixed specimens. SHG is an important source of contrast for in vivo and intravital imaging, largely because of the strong SHG signal from fibrillar collagen in skin, connective tissue, vasculature, and many internal organs. Collagen content and organization can be an important marker of tissue development or disease progression. Muscle network of a 4-day-old zebra fish embryo (400 × 300 µm) captured on the Olympus FLUOVIEW FVMPE-RS multiphoton microscope. The label-free SHG signal is emitted by the myosin filaments in the striated skeletal muscle. High-resolution imaging allows measurement of the sarcomere length, a possible indicator of muscle injury. Courtesy of Olympus America Inc. The ubiquity of collagen fibers in connective tissue makes SHG a useful tool for studying tissues such as bone, cartilage, and tendon. Many open questions remain regarding the fundamental relationships between collagen organization and tissue function. High-resolution 3D images of collagen networks via SHG microscopy provide detailed data on fiber size, density, and alignment. These, combined with biomechanical testing, help inform models that attempt to tease out the correspondence of structure to function. Many forms of cancer alter collagen distribution in the extracellular matrix around malignant cells. Skin, breast, and ovarian cancers in particular show increased collagen density, as well as distinct changes in the spatial organization of collagen fibers in the tissue stroma. Collagen network in chicken heart tissue (250 × 250 × 36 µm). SHG microscopy captures the 3D organization of collagen fibers with submicron resolution. Information such as fiber size, density, and directionality can be useful for studying tissue infarction and subsequent scarring. Courtesy of Olympus America Inc. SHG microscopy provides a highly resolved view of such alterations, potentially enabling earlier detection and even helping identify tumors with a greater risk of metastasis. This technique is minimally invasive to the tissue being imaged, so it can be used to repeatedly image the same tumors over time, whether in animal or 3D organoid models. Such longitudinal data can be very powerful in the search for more effective cancer treatments. The study of other noncancerous fibrotic diseases — such as liver cirrhosis, kidney fibrosis, and atherosclerosis — stands to benefit similarly from SHG microscopy SHG vs. TPEF Similar to TPEF, SHG requires very high incident light intensity and relies on femtosecond pulsed lasers focused with high numerical aperture (NA) objectives to achieve observable responses. SHG also has the benefit of intrinsic optical sectioning because of the nonlinear optical response. Because TPEF and SHG microscopy have overlapping instrumentation requirements, these two imaging modalities can be implemented on the same laser scanning multiphoton microscope to achieve high-resolution, 3D imaging at deep sites in highly scattering specimens. However, there are some significant differences between SHG and TPEF that should be considered when applying multimodal imaging. Such differences are found in four categories: spectral response, emission direction, polarization dependence, and concentration dependence. TPEF is an absorption-emission process with a spectral emission profile determined by the resonant energy levels of the fluorophore and generally independent of the excitation wavelength. In contrast, SHG spectral emission is entirely determined by the incident laser. It is a nonabsorptive, nonlinear scattering process where emission is always half of the incident wavelength with a narrow bandwidth, which is also dependent on the incident laser bandwidth. The Olympus FLUOVIEW FVMPE-RS multiphoton microscope is a turnkey solution for nonlinear imaging applications such as TPEF and SHG. The system delivers circularly polarized light at the sample plane for unbiased SHG imaging. A two-channel transmission side detection module can efficiently capture forward emissions, while up to four detectors are positioned on the reflection side. Courtesy of Olympus America Inc. This tunable emission profile is useful when seeking to spectrally separate SHG signal from other emissions coming from the specimen. It can also be used to confirm that a captured signal truly represents SHG emissions. In other situations, it can enable simultaneous excitation of TPEF and SHG with a single laser wavelength. Unlike fluorescence, which is emitted uniformly in all directions, SHG emission is anisotropic. For very thin specimens, such as cell monolayers, SHG is most often stronger in the forward direction. In such situations, a detector placed on the transmission side of the microscope can capture SHG images with a higher signal-to-noise ratio than the standard reflection side detectors. SHG has the benefit of intrinsic optical sectioning because of the nonlinear optical response. For best results, the transmission detector should be coupled with a high NA condenser that is matched to the NA of the focusing objective lens. Forward SHG emission tends to be strongest along a cone determined by the NA of the focused incident laser. In thicker specimens, multiple scattering can redirect forward SHG to the backward direction, thus reducing directional bias. In fact, for whole mount tissue and intravital microscopy, detection of backscattered SHG is very efficient and transmission detectors are not necessary. Under linearly polarized laser illumination, SHG intensity is stronger for specimen features that are parallel to the axis of polarization and weaker for features that are perpendicular. For example, when imaging a network of collagen fibers, SHG intensity will be modulated according to the fiber’s orientation angle with respect to the polarization axis of the laser. Fibers that are parallel to the laser polarization will appear the brightest, while perpendicular fibers will exhibit the weakest emission. This orientation angle bias can be avoided by using circularly polarized light for the laser illumination. TPEF emission is incoherent, so there is a simple linear relationship between fluorophore concentration and fluorescence intensity. On the other hand, SHG is a coherent scattering process subject to constructive and destructive interference. Constructive interference between SHG photons leads to a quadratic scaling of SHG intensity with the concentration of SHG emitters. However, this can be modulated by destructive interference, depending on the orientation and organization of emitters within the excitation region. Thus, the quantitative treatment of SHG intensity is often not as straightforward as is quantitative fluorescence microscopy. One limitation worth noting is that the presence of noncentrosymmetric molecules in a medium is not a sufficient condition for observing SHG. Wider spatial organization that supports constructive interference is necessary for SHG intensity to reach a detectable signal-to-noise ratio. For example, SHG can be observed in intracellular microtubules, but not in solutions of the constituent tubulin heterodimers. Although tubulin is noncentrosymmetric, heterodimers are randomly oriented when in solution. As another illustration, collagen I in skin and tendon exhibits strong SHG emissions, thanks to a highly organized hierarchical structure from collagen molecule and fibril-to-fiber bundles. In contrast, collagen IV in extracellular basement membranes is not SHG active because these collagen molecules are arranged with more random orientations in a meshlike network. SHG microscopy — a label-free technique for 3D imaging of specific cellular and tissue structures — is not only chemically selective but also specific to the underlying organization of biomolecular assemblies. SHG can be applied on the same microscope as TPEF microscopy, but key differences should be considered to ensure successful imaging. SHG emission wavelength is always half of the incident laser wavelength. These wavelengths can be selected for minimum crosstalk with other spectral channels, or for simultaneous excitation with other channels. Such emission is generally biased in the forward direction, thus transmission side detectors should be utilized when working with thin specimens. Reflection side detectors can be applied for thicker specimens that exhibit significant backscattering, while circularly polarized laser illumination can be used to avoid angle orientation bias in SHG images. Although very useful for identifying structure and organization, the coherent nature of SHG makes quantitative interpretation of intensity versus concentration more complicated than with TPEF. Meet the author Carlo Alonzo is the product manager for multiphoton microscopy and customized solutions at Olympus America Inc. In this role, he assists scientists in identifying and understanding enabling technologies that can support their research goals. Alonzo has a Ph.D. in physics from the University of the Philippines and received postdoctoral training at the Technical University of Denmark. His background is in optics, multiphoton microscopy, biomedical optics, and lasers.