Making the Connection: Optics, Neural Activity and Behavior

Fiber photometry and fluorescence microscopy bolster optogenetic research into how the brain works.

An international army of neuroscientists is trying to better understand the inner workings of the brain by exploring the correlations between neural activity and behavior of freely moving or partially restrained lab animals. The best method for neural activity control, activation or silencing of opsin-expressing neurons is optogenetics. The prevalent optics-based methods for monitoring the neural activity of fluorophore-labeled neurons are fiber photometry and head-mounted fluorescence microscopy.

Neuroscientists appear equally divided over whether fiber photometry or fluorescence microscopy is the more appropriate method of monitoring the neural activity revealed by excitation of specific fluorophores. But there is no such division regarding optogenetics and its control over neural activity by light and optics.

Figure 1. Courtesy of Doric Lenses.

Optogenetics uses light to activate or silence cells labeled with genetically encoded light-sensitive proteins or opsins that are coupled to ion channels. When the light-gated ion channel absorbs the light of a specific wavelength range, it changes its configuration and lets ions flow into the cell, modulating its activity. By expressing opsins in a specific population of cells, those cells become responsive to light signals and can be controlled with light.

The most popular opsins are Channel rhodopsin-2 (ChR2) and Halorhodopsin-3.0 (NpHR3.0). When blue light (450 to 473 nm) activates ChR2, a nonspecific cation-permeable channel opens and the depolarization of the cell occurs. The light-gated chloride pump NpHR3.0 is activated by yellow light (595 nm), allowing chloride ions to enter the cell, causing the hyperpolarization of the neuron or its silencing. The opsin activation light power threshold is a few mW/mm² for blue and up to 20 mW/mm² for amber light.

The direct measurement of neural electrical activity in freely moving animals is very difficult because of its fast timescale and the low sensitivity of available sensors. For these reasons, it is beneficial to use an indirect approach.

Calcium fluctuations at the cellular level relate to underlying electrical activity and can be measured using fluorescent genetically encoded calcium indicators (GECIs). Variations of fluorescence indicate changes in calcium concentration, revealing cellular activity.

With GECIs, it is possible to target a specific cellular type. The most common GECI is GCaMP, a protein that absorbs blue light with peak absorption around 495 nm and emits green light with peak emission around 510 nm. The fluorescence emission of the GCaMP indicator is quenched when no calcium is present and increases with elevating calcium concentration.

Fiber photometry

To monitor neural activity in freely moving animals with fiber photometry, a chronically implanted optical fiber delivers excitation light to neurons tagged with a fluorescent calcium indicator(s) and collects the overall activity-induced fluorescence. The diameter of the light delivery and light collection fiber is typically 200 or 400 µm. The small size of the implant makes it possible to use fiber photometry to record from multiple brain regions.

Fiber photometry is compatible with multicolor and even FRET (fluorescence resonance energy transfer) experiments. The basic experimental setup consists of a fluorescence cube that separates excitation and emission light, a photodetector that measures fluorescence variation, and a data acquisition box that records the electrical signal coming from the photodetector. More sensitive photodetectors require less intense excitation light, reducing photo bleaching and photo damage in the tissue.

However, integrating all the fluorescence of a single nonmodulated excitation with a single detector makes it impossible to distinguish fluorescence variations coming from the calcium indicator from those induced by excitation light intensity instabilities, rotary joint optical coupling variations and brain movements.

The separation of real calcium variations from those artifacts is possible if the excitation light is modulated or if another calcium-independent excitation wavelength is used simultaneously.

When choosing components for a fiber photometry setup, select a stable light source that can be modulated, an optimized rotary joint minimizing variations upon rotation, a low-noise amplified photodetector, and low auto-fluorescence patch cords and cannulas. The connectorized fluorescence cubes prevent light leakage and eliminate optical alignment by the end user.

Different applications require cubes with a corresponding number of ports and internal configuration. Data acquisition hardware and software make it easy to set up fiber photometry recording sessions with different parameters for excitation and lock-in detection capabilities (Figure 1).

Head-mounted microscopy

Head-mounted miniature fluorescence microscopy provides micron-scale resolution images of bursts of calcium activity within GECI-labeled neurons that are excited by light in the brain of a freely moving animal.

The typical approach to head-mounted miniaturized fluorescent microscopes is to use a CMOS chip as an image sensor; an LED light source and a dichroic filter within the microscope body; and a gradient index (GRIN) relay lens within the implanted imaging cannula. The electronic signal from the image sensor is then electrically sent to a data acquisition box and a computer.

However, Doric Lenses uses external light sources that are connected to the microscope via an optical fiber and an assisted electro-optic rotary joint. This improves the chances of finding a suitable light source — an LED, a laser or other — with the appropriate spectral band and optical power level. In the case of internal light sources, the only suitable miniature light sources are LEDs, but they do not cover all spectral bands, might not have sufficient optical power and cannot be easily replaced.

Figure 2. The body of the two-color miniature microscope. Courtesy of Doric Lenses.

Our approach has additional advantages when two or more colors are used as in an optogenetically synchronized fluorescence microscope, which combines fluorescence excitation and optogenetic stimulation, or as in a two-color microscope that observes calcium activity of two groups of neurons labeled with different GECIs.

The strong absorption and scatter of light within brain tissue limits imaging depth to approximately 150 microns. For that reason, the relay lens has to be as close as possible to the region of interest. In order to minimize the damage to the brain, the chronically implanted lens has to be as thin as possible. To collect the fluorescence, the lens has to have a big numerical aperture. So far, it seems that only the GRIN lens is a viable option.

However, GRIN lenses suffer from a high chromatic aberration that cannot be easily compensated. In its two-color fluorescent microscope, Doric compensates for this chromatic shift by using two CMOS chips and separate optical paths to image signals emitted from the same optical plane (Figure 2).

The illumination power requirement for optogenetic stimulation is roughly 10 times greater than the power needed for fluorescence excitation, preferably without any crosstalk. This requires fine-tuned optical filters and appropriate stray light management to prevent opsin stimulation from affection fluorescence image recording.

Meet the author

Sead Doric is the president and founder of Doric Lenses, which initially marketed gradient index lenses for collimation of laser diodes. The company got involved in optogenetics in 2007, making everything from fiber optic cannulas to optical rotary joints, fluorescence microscopes and fiber photometry systems; email:sead@doriclenses.com.

Hardware That Makes a Difference

Both fiber photometry and head-mounted fluorescence microscopy require an optical and/or electrical tether and an optical port of entry (fiber optic or imaging cannula) chronically implanted on the animal’s head.

An optical tether is made up of an optical patch cord with a flexible jacket. The jacket protects the cord from being chewed, and its strength depends greatly on the size and age of the lab animal and the type of behavioral experiments being conducted.

High numerical aperture (NA) large-core multimode fibers are preferred for the patch cord itself because they provide higher LED power light delivery and better collection of the tissue fluorescence. When laser diodes are used as a light source, smaller core and smaller NA fibers are a better fit for some optical rotary joints.

Two-color optogenetics stimulation with a Ce:YAG and LED fiber light source. Courtesy of Doric Lenses.

The part of the cannula secured to the top of the skull is simply an optical connector, albeit a very small one, while the side facing the brain interior has a predefined length of optical fiber or a rod lens pointing toward the selected region of interest. The scattering of the light in the brain tissue requires that the fiber tip be as close as possible to the region under investigation. Cannulas can be made in different forms as long as they are small, allow for precision guiding during implantation, and provide repetitive and strong connection with the optical patch cord.

Once the fiber tip of the cannula is placed close to a group of neurons labeled by fluorophores or photo sensitized by opsins, neural activity can be activated or silenced using optogenetics and monitored using either fiber photometry or fluorescence microscopy.

By combining an external camera to record the animal’s behavior, some modulated light sources with their drivers, a fiber-optic rotary joint, a bunch of fiber-optic patch cords, appropriate detectors and a console to synchronize all of this, one is set to do simple behavior experiments.

Things can get more complex if groups of neurons are labeled with different fluorophores or opsins and illuminated with corresponding excitation or activation light pulses. In this case, autofluorescence is subtracted from the signal and fluorescence resonance energy transfer, or FRET, is monitored.

On the photonics hardware side, all this is made possible thanks to extreme miniaturization of all involved components, especially those sitting on the animal’s skull.

Doric Lenses became involved in neuroscience when the Karl Deisseroth lab at Stanford University asked for a fiber optic rotary joint for its optogenetics experiments with freely moving optically tethered mice. This resulted in a miniature, low-friction and connectorized fiber optic rotary joint.

The early success with rotary joints led to the realization that optogenetics needs a complete photonics infrastructure from the modulated light source to chronic animal brain implants. Inspired by the optical fiber deployment in the telecommunication industry, the company also developed vertically integrated photonics hardware for optogenetics that provides control over the light sources and light delivery systems.

In the process, other critical tools and components were developed, including single and dual fiber optic cannula, a range of low-friction optical and hybrid rotary joints, fiber-coupled light sources including powerful yellow light sources, and a variety of fiber-optic patch cords.

Controlling brain activity with optogenetics and observing the related behavioral changes is a first step toward understanding the links between the two. To better understand the underlying brain complexity, it is important to monitor its activity at a cellular level too. Optogenetics hardware developments have prepared the ground for the implementation of the optical monitoring of neural activity with fiber photometry or fluorescence microscopy.