Construction of implantable optical fibers for long-term optogenetic manipulation of neural circuits

Abstract

In vivo optogenetic strategies have redefined our ability to assay how neural circuits govern behavior. Although acutely implanted optical fibers have previously been used in such studies, long-term control over neuronal activity has been largely unachievable. Here we describe a method to construct implantable optical fibers to readily manipulate neural circuit elements with minimal tissue damage or change in light output over time (weeks to months). Implanted optical fibers readily interface with in vivo electrophysiological arrays or electrochemical detection electrodes. The procedure described here, from implant construction to the start of behavioral experimentation, can be completed in approximately 2–6 weeks. Successful use of implantable optical fibers will allow for long-term control of mammalian neural circuits in vivo, which is integral to the study of the neurobiology of behavior.

INTRODUCTION

The use of optogenetics for manipulation of neural tissue has become an essential component in the array of tools used in contemporary neuroscience to study brain function and behavior. Studies conducted in organisms such as zebrafish and nematodes^1–4, rodents^5–12 and nonhuman primates^13,14 have successfully used optogenetic manipulations to study neural circuit function. As genetically engineered opsin proteins and gene expression systems continue to improve, it is essential that optogenetic hardware components be further refined in order to complement these biological achievements. Furthermore, a comprehensive optogenetic tool set is essential for examining the neural circuits required for behavioral processes such as complex learning and decision-making tasks. Here we provide a protocol for constructing implantable optical fibers that provide precise control of neural tissue over many behavioral sessions and that also allow for the integration of other recording techniques such as in vivo electrophysiology and methodologies for the electrochemical detection of neurotransmitters. Although similar devices are commercially available, such as the fiber optic cannula systems offered by Doric Lenses (http://www.doriclenses.com/), these systems may not be fully customizable to meet specific requirements for all types of optogenetic experiments. In contrast, understanding the basic principles of optical fiber construction allows immediate experimental adjustments and full customization if necessary. The following protocol is based on techniques used by our laboratory¹⁵.

Advantages of implantable optical fibers

The majority of studies combining behavioral tasks and in vivo optogenetic studies are based on experiments in which the optical fiber is inserted through a guide cannula immediately before the start of the behavioral task^5–11; however, limitations to this method exist. Notably, during complex behavioral paradigms using acutely implanted optical fibers, mice can typically be used for only a few behavioral sessions, because of breakage of the optical fiber inside the guide cannula or tissue damage as a result of repeated fiber insertion. However, implantable optical fibers are permanently inserted into brain tissue and affixed to the skull, thus markedly reducing tissue damage and greatly enhancing experimental throughput. Implantable optical fibers also ensure that the same tissue is repeatedly stimulated or inhibited over multiple behavioral sessions, whereas optical fibers inserted acutely can potentially stimulate or inhibit different regions of tissue on the basis of movement inside the guide cannula. Furthermore, as each acute experiment requires a new fiber per behavioral session, there is the possibility that brain tissue can be damaged with repeated insertion of an optical fiber. In comparison, we have observed less tissue damage at the implantation site of optical fibers on the basis of histological examination after prolonged behavioral testing¹⁵. Light emission from the implantable fibers is not substantially diminished over time (tested up to 4 months after implantation; Fig. 1a). In addition to the experimental benefits listed above, implantable optical fibers can be interfaced with multielectrode arrays used in in vivo electrophysiology experiments. Combining multielectrode arrays with optical fibers for light delivery is relatively straightforward and can be used for in vivo recordings to determine how modulation affects the spiking activity and local field potentials during animal behavior^16,17. In short, implantable optical fibers remain patent for extended periods of time and can be used for multiple behavioral sessions without any changes in activation patterns caused by fiber movement or emission degradation.

Figure 1.

Light transmission from implantable optical fibers before and after implantation. (a) Optical fibers before implantation transmitted 85.2 ± 2.8% of the light from the patch cable. After 6 weeks in brain tissue, the same optical fibers transmitted 81.1 ± 3.1% of the light from the patch cable (n = 10 implanted optical fibers). (b) Photomicrograph of the ventral tegmental area depicting typical damage associated with the implantation of an optical fiber that was secured in the brain for 8 weeks. Cell nuclei are shown in blue (DAPI ) and neurons expressing opsin proteins are shown in green.

Limitations

Although implantable optical fibers have many advantages over acutely implantable optical fibers, there are some caveats associated with this technique. For example, we have observed that light emission can be visibly noticeable at the interface between the patch cable and the implanted optical fiber at high light-output intensities. If this is left unattended, it may serve as an unwarranted cue to the animal during behavioral sessions and may alter responses^18,19. To reduce this salient light source, it is possible to diminish extraneous light between the implant and the optical patch cable using light-impermeable tubing, which encases the optical patch cable (Box 1). Finally, although it is possible to recover the vast majority of implantable fibers to test light emission at the conclusion of experiments, approximately 1–10% of fibers break upon removal from brain tissue. Despite these limitations, implantable optical fibers allow for long-lasting control of neural circuits in vivo, with minimal drawbacks when compared with acute optical fiber procedures.

Box 1. Stereotaxic adapters for chronic implantable fibers ● TIMING 1 d.

1. Use a Dremel tool to cut a piece of 20-gauge stainless steel tubing to a final length of 4 cm.

2. By using a hemostat, bend ~1 cm of the steel tubing 90° in one direction. On the opposite end of the tubing, bend ~1 cm of the tubing 90° in the opposite direction.

3. Prepare the heat-curable epoxy (see REAGENT SETUP).

4. Fill the 1-ml syringe with epoxy. Add a blunted 25-gauge syringe tip to syringe. Insert the bent tubing into the vice. Add approximately one drop of epoxy to end of the exposed tubing. Use a cotton-tip applicator to spread the epoxy around evenly. Do not add more than one drop of epoxy.

5. Insert the ceramic split sleeve ~2 mm over the bent stainless tubing that is covered with epoxy. Do not insert more than 2 mm or epoxy will occlude the sleeve.

6. Using the heat gun, cure the epoxy. Epoxy should harden within 1–2 min of constant heat application. Make at least one adapter per implantable optical fiber (Fig. 5a). For bilateral implantations, you will need to have two stereotaxic adapters.

Experimental design

Subjects

Although the techniques described in this protocol are optimized for use in mice, they can be readily adapted for use in other mammalian species, such as rats and primates^13,14, by increasing the diameter of the optical fibers and ferrules used.

When conducting these experiments in rodents, subjects should be singly housed after implantation in order to minimize damage to the implantable optical fiber from interactions with cage mates. Before the start of experimentation, all procedures must first be approved by all relevant institutional animal care and use committees.

Materials for constructing implantable optical fibers

This protocol details the construction of implantable optical fibers using 200-µm multimode optical fibers with a numerical aperture (NA) of 0.37. Patch cables are constructed using a 50–62.5-µm multimode fibers with a 0.22 NA. For some applications, such as targeting a smaller or larger volume of neural tissue, optical fibers with different diameters and NAs should be used. To minimize the loss of light at ferrule interfaces, the two optical fibers at the coupling junction should match in diameter and NA. Alternatively, light can be transmitted from a smaller-diameter and lower-NA fiber to a larger diameter and larger NA fiber with minimal attenuation as described in the PROCEDURE.

Controls

For all in vivo optogenetic experiments, light output of the implanted optical fibers should be measured before and after experimentation. Although we report a minimal change in light output through implantable optical fibers (4–6 weeks post implantation; Fig. 1a), data from animals in which implanted fibers show severe light loss after experimentation ( >30%) should be excluded from data analysis. In addition to implanting optical fibers into subjects that are expressing opsin proteins, naive mice or those only expressing genetically encoded fluorescent proteins should also be implanted to ensure that any behavioral outcomes observed are not due to tissue damage from the implanted optical fibers.

MATERIALS

REAGENTS

Construction of implantable optical fibers

• Heat-curable epoxy, hardener and resin (Precision Fiber Products, cat. no. ET-353ND-16OZ)

Construction of patch cables for use in in vivo optogenetic experiments

• Five-min epoxy (Devcon; Grainger Industrial Supply, cat. no. 5A462)

Implantation of optical fibers

• Titan Bond or other cyanoacrylic adhesive (Horizon Dental Products, cat. no. 1301)

• Ethanol (70% (vol/vol), Sigma, cat. no. 459836) ! CAUTION Ethanol is flammable; avoid exposure to flame.

• Jet denture repair powder and Ortho-Jet BCA liquid, package (Lang Dental, cat. no. 1234PNK) ! CAUTION Inhalation of powder and fumes may be harmful.

• Ketamine (Ketaset; Butler-Schein, cat. no. 010177) ! CAUTION Ketamine is a controlled substance and should be handled according to institutional guidelines.

• Xylazine (AnaSed; Butler-Schein, cat. no. 033198)

• Adeno-associated viral constructs coding for opsin proteins (University of North Carolina Vector Core Facility; http://genetherapy.unc.edu/services.htm)

• Betadine solution (Fisher Scientific, cat. no. 19-027136)

• Lidocaine hydrochloride monohydrate (Sigma, cat. no. L5647)

• Ophthalmic ointment (Butler-Schein, cat. no. 008897)

• Vetbond tissue adhesive (Fisher Scientific, cat. no. NC9259532)

• Phosphate-buffered saline (Sigma, cat. no. P4417)

• Paraformaldehyde (4%, wt/vol) (Sigma, cat. no. P6148-1kg)

• Sodium chloride (NaCl; Fisher Scientific, cat. no. 5271-3)

• Food coloring

EQUIPMENT

Construction of implantable optical fibers

• Standard hard cladding multimode fiber (low OH, 200-µm core, 0.37 NA; Thorlabs, cat. no. BFL37-2000)

• Heat shrink tubing, 1/8 inch (Allied Electronics, cat. no. 689-0267)

• Multimode ceramic zirconia ferrule (1.25-mm outer diameter (o.d.)) with 230-µm inner diameter (i.d.) bore (Precision Fiber Products, cat. no. MM-FER2007C-2300)

• Fiber stripping tool, clad/coat (200 µm/300 µm; Thorlabs, cat. no. T10S13)

• Carbide-tip fiber optic scribe (Precision Fiber Products, cat. no. M1-46124)

• Straight, serrated hemostat (Fine Science Tools, cat. no. 13014-14)

• Fiber polishing/lapping film (aluminum oxide/silicon carbide: 0.3, 1, 3, 5 µm grits; Thorlabs, cat. nos. LFG03P, LFG1P, LFG3P, LFG5P)

• Polishing pad (9 inch × 13 inch, 50 durometer; Thorlabs, cat no. NRS913)

• Heat gun (250 W; 750–800 °F; Allied Electronics, cat. no. 972-6966)

• Needles (25 gauge, 5/8 inch; BD, Fisher Scientific, cat. no. 305122)

• Luer-tip tuberculin syringe (1 ml; Kendall, Fisher Scientific, cat. no. 22-257-154)

• Vise with weighted base (Panavise; Altex Electronics, cat. no. PAN381)

• Cotton-tip applicators, wood shaft, nonsterile (Fisher Scientific, cat. no. 23-400-119)

• Wire stripper (Ideal Industries, cat. no. 45–121)

• Polyethylene tubing (1.57 mm o.d. × 1.14 mm i.d., PE160; Warner Instruments, cat. no. 64-0756)

• Scotch tape (3M)

Construction of patch cables for use in in vivo optogenetic experiments

• Multimode fiber (low OH, 50-µm core, 0.22 NA, Vis-IR; Thorlabs, cat. no. AFS50/125Y)

• Coupler (50/50, 250 µm; Fiber Instrument Sales, cat. no. MMC28550122C)

• Multimode ceramic zirconia ferrule (1.25 mm o.d.) with 127 µm i.d. bore (Precision Fiber Products, cat. no. MM-FER2007C-1270)

• Furcation tubing (900 µm o.d., 250 µm i.d.; Precision Fiber Products, cat. no. FF9-250)

• Multimode FC MM ferrule assembly (length 3.95–4 mm, bore 0.127–0.128 mm; Precision Fiber Products)

• Connector boot-fits, SC/FC connector type (900 µm; Precision Fiber Products, cat. no. BOT-FCSC)

• Fiber stripping tool, clad/coat (125 µm/250 µm)

• Polishing disc (Thorlabs, cat. no. D50FC)

• Fiber microscope (Thorlabs, cat. no. CL-200)

Measuring light output of implantable optical fibers and patch cables

• Patch cables

• Implantable optical fibers

• Laser, driver and laser to fiber coupler, such as a 100-mW, 473-nm diode-pumped solid-state laser (OEM Laser Systems, http://www.oemlasersystems.com/, cat. no. BL-473-00100-CWM-SD-xx-LED-0). See Zhang et al.¹¹ for details.

• Compact power and energy meter console with digital display (Thorlabs, cat. no. PM100D)

• Ceramic split sleeve (1.25 mm i.d.; Precision Fiber Products, cat. no. SM-CS125S)

Stereotaxic adapters for implantable optical fibers

• Stainless steel tubing (20 gauge; McMaster Carr, cat. no. 8987K421)

• Dremel with rotary tool kit (Grainger Industrial Supply, cat. no. 5EEU8)

Implantation of optical fibers

• No.11 scalpel blade (Fisher Scientific, cat. no. S17302)

• Graefe forceps (0.8 mm curved; Fine Science Tools, cat. no. 11651)

• Curved, serrated hemostat (Fine Science Tools, cat. no. 13015-14)

• Stereotaxic adapters for implantable fiber (Box 1)

• Polypropylene microtubes (1.7 ml; Genesse Scientific, cat. no. 22-281)

• Plates, 24-well (Falcon; Fisher Scientific, cat. no. 353047)

• Double-ended micro-spatula (8 inch; Fisher Scientific, cat. no. 21-401-10)

• Needles (18 gauge, 1 inch; (BD, Fisher Scientific, cat. no. 305195)

• Optical fiber protectors

• Anchoring screws (Antrin Miniature Specialties, cat. no. AMS90/1B-100)

Behavioral procedures using implantable optical fibers

• Patch cables

• Mouse implanted with optical fibers

• Optical fiber protectors

• Furcation tubing (900 µm o.d., 250 µm i.d., white Hytrel; Precision Fiber Products, cat. no. FF9-250-WHI-100)

• Laser, driver and laser to fiber coupler, see Zhang et al.¹¹ for details.

• Ceramic split sleeve (1.25 mm i.d.; Precision Fiber Products, cat. no. SM-CS125S)

Interfacing implantable optical fibers with in vivo electrophysiological arrays

• Absorbable gelatin sponge (Harvard Apparatus, cat. no. 599863)

• Electrode microarrays (4 × 4) with 35-µm tungsten electrodes, 150-µm electrode spacing, 150-µm row space, 3.55-mm electrode length, low-impedance reference electrode, micropolished electrode tips, 0.0008-inch silver ground wires (Innovative Neurophysiology)

• Hypodermic stainless steel tubing (thin wall, gauge 20; A.M. Systems, cat. no. 843200)

REAGENT SETUP

Heat-curable epoxy Add 1 g of resin to 100 mg of hardener.

▲ CRITICAL Prepared heat-curable epoxy will be usable for ~3 h.

Headcap cement In a well plate, mix seven parts Jet denture repair powder to three parts Jet denture repair liquid. Mix evenly. ▲ CRITICAL Headcap cement should have a semi-viscous consistency to avoid improper coverage of skull. Prepared headcap mixture will be usable for 2–5 min.

EQUIPMENT SETUP

Sterotactic adapters for implantable optical fibers The adapters are constructed as described in Box 1.

PROCEDURE

Construction of implantable optical fibers

1| Strip ~25 mm of 200-µm core fiber using a micro-stripper while keeping the fiber attached to the fiber spool. This makes it easier to completely strip the coating off the optical fiber.

2| Cut the fiber from the spool using wire cutters or a diamond knife. Leave ~10 mm of unstripped fiber. This will leave you with ~35 mm of fiber (25 mm stripped, 10 mm unstripped; Fig. 2).

Figure 2.

Construction of implantable optical fibers. (a) A stripped and cut piece of optical fiber for construction of implantable optical fiber. (b) The optical fiber is inserted into the ferrule and secured in place by epoxy. (c) The optical fiber is scored to cleave it at the convex side of the ferrule. (d) The convex side of the ferrule is polished using progressively finer grades of polishing paper. (e) The optical fiber is scored to break at the appropriate length for implantation into the brain. (f) The finished implantable optical fiber.

3| Prepare heat-curable epoxy (see REAGENT SETUP).

4| Fill a 1-ml syringe with the prepared heat-curable epoxy mixture. Add a blunted 25-gauge needle to the syringe.

5| Insert the ferrule into the vice with the convex side pointing down. To test the viability of the ferrule, insert the stripped end of the fiber into the ferrule. The fiber should slide in to the ferrule bore smoothly without obstruction. Remove the stripped fiber from the ferrule.

▲ CRITICAL STEP If resistance is encountered when inserting the fiber, use a new ferrule.

6| Add one drop (enough to fill the ferrule) of heat-curable epoxy to the flat end of the ferrule.

▲ CRITICAL STEP Wipe any excess epoxy off the sides of the ferrule. If epoxy dries to the side of the ferrule, interfacing the implanted optical fiber to the patch cable will be difficult.

7| Insert the stripped end of the fiber into the ferrule and thread it through, leaving an extra 5 mm of stripped fiber exposed (e.g., if your desired length is 5 mm, leave an additional 5 mm of unstripped fiber exposed, resulting in a final length of 10 mm).

8| Cure the epoxy bead using a heat gun placed ~10 mm away from the epoxy bead. Epoxy will turn black/dark purple when fully cured. Epoxy should cure in approximately 30–40 s (Fig. 2b).

9| Score the fiber at the convex end of the ferrule with diamond knife. Gently tap the fiber off with your finger or a glass rod (Fig. 2c).

▲ CRITICAL STEP Score the fiber exactly at interface of the ferrule. Leaving exposed fiber could lead to a damaged fiber core. Do not score the fiber at the flat end of the ferrule. The flat end of the ferrule will be inserted into the brain. See Steps 11 and 12 for proper scoring of the flat end of the ferrule.

10| Polish the convex end of the ferrule using a hemostat to hold the ferrule securely in place (Fig. 2d). To polish, make ~20 rotations on each grade of polishing paper (four grades total). Polish in this order: 5, 3, 1, 0.3 µm.

▲ CRITICAL STEP The ferrule needs to be perpendicular to the polishing paper or polishing will be uneven and will result in poor light output.

▲ CRITICAL STEP Polishing in a vigorous motion may damage the fiber core. Use even pressure to polish the fiber.

11| Secure the remaining unstripped fiber under a piece of Scotch tape secured to a flat surface. Score the stripped fiber with a diamond knife to the desired length (1–2 mm greater than the ventral coordinates of the brain region of interest; Fig. 2e). When scoring the fiber, hold the diamond knife perpendicular to the fiber. Score in one direction in a single motion.

▲ CRITICAL STEP Do not cut the fiber completely with the diamond knife. This will damage the fiber core.

12| Use a hemostat to pull on the ferrule with the stripped end remaining secured in the Scotch tape. To ensure an even break, apply pressure to the end of the Scotch tape. The fiber should snap at the scored length. Alternatively, scoring the fiber with the diamond knife at the desired length and then gently tapping the remaining fiber typically results in clean and even fiber breaks with larger fiber diameters (≥ 200 µm). For further precision, examine the scored end of the optical fiber under the surgical microscope to determine whether there is any damage to the fiber core.

▲ CRITICAL STEP If the fiber does not break, make another implantable fiber.

13| Cut a 5-mm piece of polyethylene tubing to serve as an optical fiber

protector to place over the implantable fiber. Heat one end of the polyethylene tubing piece with a soldering iron and gently fuse it together using a hemostat. Mark the sealed end of the optical fiber protector with a permanent marker. This will be used in Step 57.

■ PAUSE POINT Store the implantable fibers in any sealed container in which the fiber is unobstructed. Keep the implantable fibers stored until you are ready to test their output (Steps 29–35; Fig. 2f).

Construction of patch cables for use in in vivo optogenetic experiments ● TIMING 1 d

14| Cut the 50-µm core multimode optical fiber and furcation tubing to the desired length (approximately 50–100 cm for patch cables to test the implantable optical fibers; approximately 200–300 cm for most behavioral experiments)

▲ CRITICAL STEP Cut the furcation tubing 50 mm shorter than the length of the optical fiber.

15| Thread the optical fiber through the furcation tubing (Fig. 3a).

Figure 3.

Construction of patch cables for use in in vivo optogenetic experiments. (a) The unstripped optical fiber is threaded through furcation tubing. (b) The protective coating lining the optical fiber is stripped off using a fiber stripper tool. (c) The stripped optical fiber is inserted into the epoxy filled FC connector. (d) The optical fiber protruding from the FC ferrule is scored and cleaved. The protective boot is secured to the FC connector. (e) The ferrule side of the FC connector is inserted into the polishing disc. The ferrule is polished using progressively finer grades of polishing paper. (f) The 1.25-mm ferrule is epoxied and polished at the other end of the patch cable. The exposed fiber is epoxied to the ferrule and furcation tubing to provide additional structural stability. Heat-shrink tubing slides over top of the epoxy bead and the ferrule to prevent light transmission that could be visible to the subject. (g) Implantable optical fiber is interfaced with the patch cable using a ceramic split sleeve. (h) Light transmission through the implantable optical fiber should be uniform and concentric. Light output from the implant should be > 70% of the input light from the patch cable. (i) Nonconcentric, uneven light output from the implant indicates a nonviable implantable optical fiber. These should be discarded.

▲ CRITICAL STEP The procedure is designed for unilateral stimulation/inhibition experiments. However, a bilateral patch can be created in a similar manner. The procedure to construct the FC end of the ferrule is identical to Steps 16–25 and the procedure to construct the ferrule end of the patch cable that will interface with the implantable optical fiber is identical to Steps 26–28. However, these steps will have to be performed twice, as the cable is split into two bare fiber ends.

16| Strip ~25 mm of the optical fiber at both ends (Fig. 3b).

17| Insert the multimode FC MM ferrule assembly, ferrule end down, into the vice.

18| Fill a 1-ml syringe with heat-curable epoxy mixture and add a 25-gauge syringe tip. Wipe off excess epoxy before inserting the 25-gauge syringe tip.

19| Insert the 25-gauge syringe tip into the metal end of the ferrule. Apply epoxy until the back end of the ferrule is filled to the top (Fig. 3c). Before adding the epoxy to the ferrule, insert the stripped fiber into the ferrule to ensure that the bore diameter is correct.

? TROUBLESHOOTING

20| Thread the stripped end of the fiber through the ferrule opening until it stops (where the unstripped portion of the fiber starts; Fig. 3c). You will observe the stripped fiber sticking out of the ferrule end.

21| Heat-cure the epoxy with the heat gun until it turns a brownish-black color. Heat curing should take approximately 2–3 min.

▲ CRITICAL STEP The ferrule will become extremely hot during the curing process. Wait ~15 min before removing it from the vice.

22| Once the ferrule is cool, slide the protective boot over the opposite end of the optical fiber and attach it to the ferrule.

23| Score the excess fiber at the end of the FC connector with a diamond knife and gently tap the excess fiber off with your finger or a glass rod (Fig. 3d).

▲ CRITICAL STEP Score exactly at the opening of the ferrule’s FC end. Leaving exposed fiber will damage the fiber core.

24| Polish the FC end of the ferrule. To polish, insert the FC end of the ferrule into the polishing disc and make ~20 figure-eight patterns on each grade of polishing paper (four grades total). Polish in this order: 5, 3, 1, 0.3 µm (Fig. 3e).

▲ CRITICAL STEP Polishing in a vigorous motion may damage the fiber core. Use even pressure to polish the fiber.

25| Insert the FC end of the ferrule into the fiber inspection scope to check for any excess epoxy or cracks to the fiber core. The fiber core should look like a black concentric circle. If there is excess epoxy (excess epoxy looks filmy) on the core, repolish with the 1- and 0.3-µm grades of polishing paper.

? TROUBLESHOOTING

26| Insert a 127-µm ferrule into the vice with the convex side facing down. To test the viability of the ferrule, insert the stripped end of the fiber into the ferrule. The fiber should fit in smoothly without obstruction.

▲ CRITICAL STEP If resistance is observed, use a new ferrule.

27| Attach and polish the ferrule as described in Steps 6–10. Place a 2-cm piece of heat-shrink tubing over the interface of the furcation tubing and the ferrule to prevent light emission.

▲ CRITICAL STEP Do not shrink the heat-shrink tubing; it should readily slide over the ferrule, the sleeve and the ferrule on the implanted optical fiber.

28| To increase structural integrity of the patch cable, place a bead of 5-min epoxy on the exposed fiber between the ferrule and the furcation tube (Fig. 2f).

▲ CRITICAL STEP 5-min epoxy has a working time of 5 min, however, it takes approximately 20–30 min to completely cure.

■ PAUSE POINT Store patch cables until you are ready to test the light output of the patch cable (Steps 29–35).

Measuring light output of implantable optical fibers and patch cables ● TIMING 1 d

29| Connect the FC connector of the patch cable into the FC port on the coupler of the laser.

▲ CRITICAL STEP Make sure the FC connector notch fits into the driver.

30| Turn the laser driver on and wait ~15 min for the laser to stabilize. The patch cable should emit a bright, concentric circle of light (Fig. 2h,i). If the light is dim or not concentric, adjust the coupling between the optical fiber and the laser.

▲ CRITICAL STEP If no improvement in light output is observed with the coupler adjustment, make a new patch cable.

31| Measure the intensity of the light output through the patch cable using the optical power meter. Set the wavelength on the power meter to match that of the laser being used. Position the ferrule attached to the end of the patch cable directly above the center of the sensor. An ideal 50-µm-core patch cable should have light loss not exceeding 30% and should emit a concentric circle of light. For example, if the laser output is 20 mW, the patch cable output should be between > 14 mW. If you are using a bilateral patch cable, output should be equivalent on both sides.

▲ CRITICAL STEP If you do not have equivalent light output on both sides of the bilateral patch cable, remake the bilateral patch cable.

32| Once an acceptable patch cable is constructed, decrease the intensity of the laser driver until 10 mW of light output through the fiber is maintained.

33| Insert approximately 3–4 mm of the 127-µm ferrule of the patch cable into the ceramic split sleeve.

34| Insert the implantable optical fiber into the remaining space available of the ceramic split sleeve so that the two ferrules are in physical contact with each other (Fig. 2g). Make sure the implantable optical fiber makes direct contact with the end of the 127-µm ferrule patch cable. Use the window on the sleeve to determine connectivity.

? TROUBLESHOOTING

35| Measure the light intensity through the implantable optical fiber using the optical power meter (Fig. 2h,i). An ideal implantable optical fiber should have < 30% light loss and emit a concentric circle of light. As the patch cable is set to 10 mW, an acceptable implantable optical fiber should emit between 7 and 9 mW of light. Record the percent of light transmission for each implantable optical fiber.

▲ CRITICAL STEP This value is needed to set the proper light intensity delivered to brain tissue during experiments (see Step 66).

Implantation of optical fibers ● TIMING 1 d

36| Anesthetize mice (approximately 6 weeks old or 20–25 g) using an i.p. injection of 100 mg kg^{− 1} ketamine and 10 mg kg⁻¹ xylazine. By using this procedure, it should take ~15 min for mice to become fully anesthetized.

▲ CRITICAL STEP Use the toe pinch reflex procedure to determine whether the mouse is anesthetized. If the mouse shows reflex activity in response to the toe pinch, administer a booster of ketamine.

▲ CRITICAL STEP For all surgical procedures, use clean and sterile surgical instruments that have been sterilized using an autoclave.

▲ CRITICAL STEP All experiments involving animals should adhere to relevant ethical guidelines for animal use and care.

▲ CRITICAL STEP Steps 37–59 describe the procedure for standard implantation of optical fibers for in vivo behavior experiments. To interface the implantable optical fiber with in vivo electrophysiological arrays, follow the procedure outlined in Box 2 and Figure 4.

Figure 4.

Interfacing implantable optical fibers with in vivo electrophysiological arrays. (a) An optical fiber mount that is constructed out of stainless steel tubing is epoxied to the multiunit electrode array. (b) The implantable optical fiber is mounted and secured to the optical fiber mount located on the array. (c) The finished optrode is connected to the stereotaxic adapter and is ready for implantation.

37| By using the forceps, pluck the hair from the top of the mouse’s head.

38| Place the mouse in the stereotaxic frame.

39| Inject 0.3 ml of 0.9% (wt/vol) NaCl subcutaneously to prevent dehydration during surgery.

40| Apply ophthalmic ointment to eyes to prevent drying.

41| To minimize the risk of infection, wipe the surgical area with Betadine and alcohol using certified aseptic techniques.

42| Use a scalpel to make a 1–1.5-cm incision through the midline of the scalp. Topically apply 30 µg of lidocaine to the top of the skull.

43| By using the forceps, pull back the skin into four quadrants to ensure a large working surgical surface.

44| Insert a cotton swab into a 70% (vol/vol) ethanol solution and swab the skull to identify bregma and lambda. Allow the ethanol to dry to allow for proper identification of skull sutures.

▲ CRITICAL STEP If the skull sutures are difficult to determine, apply a drop of food coloring. Wipe excess food coloring with gauze or a cotton swab.

45| Level the skull using the anterior-posterior tilt and medial-lateral tilt on the stereotax. To level, place the tip of drill on bregma and measure the dorsal-ventral (D/V) coordinates. Move the arm to lambda and record the D/V measurement of the drill back on bregma. Move the drill ± 1.0 mm laterally and record the D/V measurements. If the head is not level, you will have inaccurate placement of the implantable optical fiber. Correct placement is crucial for any mouse brain structure.

? TROUBLESHOOTING

46| Use the stereotaxic atlas to determine the brain region of interest. Move the drill to the appropriate coordinates and drill the skull. Once the skull is removed, use a 26-gauge syringe tip to break through the dura without damaging the cortex.

▲ CRITICAL STEP Do not drill through the dura mater.

Box 2. Interfacing implantable optical fibers with in vivo electrophysiological arrays ● TIMING 1 d.

One benefit of using implantable optical fibers for in vivo behavioral tasks is that they can readily be adapted and used in conjunction with other techniques such as in vivo electrophysiology. The following steps outline a procedure for interfacing an implantable optical fiber with a nondriveable chronically implanted multiunit electrode array (MEA). With the successful construction of an optical fiber to a MEA, selective control of neural circuit elements while recording spike activity as well as local field potentials is achievable.

▲ CRITICAL STEP The length of the optical fiber that will be interfaced with the MEA should be ~4 mm longer than the length of the electrodes on the MEA.

1. Use a Dremel tool to cut two 5-mm pieces of 20-gauge stainless steel hypodermic tubing.

2. Flatten one-half of each piece of stainless steel tubing using pliers, thereby resulting in the closure of one end of each piece of tubing.

3. Under a stereomicroscope, place one drop of 5-min epoxy on the Omnetics connector of the MEA and glue the two pieces of stainless steel tubing to the MEA with the flattened ends pointing toward the electrodes; let the glue dry (Fig. 4a). ▲ CRITICAL STEP This will allow for the positioning of the optical fiber within 1 mm of the MEA tips as well as creating an angle of ~20° between the MEA and the optical fiber; this allows sufficient spacing between the MEA and the optical fiber and enables the connection of the headstage and patch cable.

4. Place one drop of 5-min epoxy on the two stainless steel tubes attached to the MEA and position the optical fiber portion of the implantable optical fiber so that it rests between the stainless steel tubes (Fig. 4b). Use forceps to gently position the optical fiber so that the optical fiber tip terminates ~500 µm above the MEA tips. ▲ CRITICAL STEP This is important to allow the diffusion of light in the tissue surrounding the tips of the electrodes and to minimize photoelectric artifacts during recording. See Cardin et al.²² for additional details on minimizing optical artifacts during electrophysiological recordings.

5. Let the epoxy that is used to interface the optical fiber with the MEA fully cure. This will take approximately 10–15 min. The curing process should be carefully monitored through the microscope, as the MEA and optical fiber are not yet secured. This can result in misalignment of the implantable optical fiber during curing. Once the fiber is optimally positioned, place two additional drops of 5-min epoxy on the optical fiber and let cure. ▲ CRITICAL STEP If any misalignment occurs, gently reposition the optical fiber using fine forceps.

6. Immediately before implantation of the optical fiber coupled to the MEA, prepare the animal as described in Steps 37–45 of the main PROCEDURE. Deliver virus as described by Zhang et al.¹¹.

7. Make a square-shaped craniotomy (slightly larger than the optrode) directly above the brain regions of interest. Carefully remove excess bone fragments with the forceps. ▲ CRITICAL STEP Steps 7–9 of this box are best performed using a surgical microscope.

8. Remove the dura completely using fine forceps or a 26-gauge needle. ▲ CRITICAL STEP It is crucial for successful optrode implantation that the dura is completely removed, because dura fragments may result in the bending of the electrodes when the optrode is lowered into the brain.

9. Cut a piece of absorbable gelatin sponge, approximately 2.5 mm × 2.5 mm × 2.5 mm, and moisten it with sterile 0.9% (wt/vol) saline solution. Place the sponge over the craniotomy to prevent swelling of the cortex and to prevent any bone fragment from entering the craniotomy.

10. Drill three holes in the skull for insertion of anchoring screws to help anchor the headcap. ▲ CRITICAL STEP Drill the holes ~5 mm away from optrode insertion site.

11. Mount screws into skull using fine forceps and a micro-screwdriver.

12. Apply an even layer of Titan Bond (~50 µl) to the skull to ensure proper anchoring for the headcap. Use a cotton-tip applicator to evenly apply Titan Bond over entire skull surface. ▲ CRITICAL STEP Wait approximately 10–15 min to allow the Titan Bond to dry completely. ? TROUBLESHOOTING

13. Mount the optrode on the stereotactic arm by connecting it to an Omnetics connector that has been epoxied to a 4-cm piece of 20-gauge stainless steel tubing that is held by the stereotactic arm. Gently position the optrode so that the MEA wires are perpendicular to the skull (Fig. 4c).

14. Remove the sponge piece from the craniotomy.

15. Lower the optrode through the craniotomy to the desired depth. ▲ CRITICAL STEP This should be done very slowly, (e.g., do not lower the optrode at rate of more than 500 µm min^{− 1}).

16. Wrap the ground wire connected to the MEA around each skull screw twice using fine forceps. ▲ CRITICAL STEP Do not disturb the optrode during this procedure.

17. Construct a headcap out of cement to fully secure the optrode to the skull. Refer to Steps 56–59 of the main PROCEDURE for headcap construction. ▲ CRITICAL STEP Make sure that the all of the black epoxy (i.e., that interfaces the MEA wires with the Omnetics connector) and the lower half of the ferrule connected to the optical fiber are fully encased in headcap cement.

18. Let the mouse recover in its home cage for approximately 3–6 weeks (see Step 61 of the PROCEDURE). Follow the same habituation procedures as outlined in Steps 66–68 of the PROCEDURE to habituate mouse to tethering before behavioral experimentation.

47| Thaw the virus on ice or at 4 °C. Deliver the virus as described in Zhang et al.¹¹.

▲ CRITICAL STEP Keep the virus on ice or at 4 °C throughout the procedure and return it to −80 °C promptly when you are finished. Avoid thawing virus more than two times.

48| Drill two holes into the skull to insert the anchoring screws that will be used later to secure the head cap.

▲ CRITICAL STEP Drill anchoring screw holes ~5 mm away from the implantable optical fiber site.

49| By using the forceps and a micro-screwdriver, drive the anchoring screws into the skull.

▲ CRITICAL STEP Use one or two turns of the screwdriver to attach the anchoring screws to the skull and to avoid damage to the brain.

50| Attach the stereotaxic adapter for implantable optical fibers to the stereotaxic arm (Box 1 and Fig. 5a,b). Insert ~2 mm of the implantable optical fiber into the stereotaxic adapter.

Figure 5.

Implantation of optical fibers. (a) The stereotaxic adapter for implantation of optical fibers. (b) The stereotaxic adapter interfaced with an implantable optical fiber is secured to the stereotaxic arm. (c) Anchoring screws are secured to the skull and the implantable optical fibers are slowly lowered into the brain tissue. (d) The implanted optical fibers are secured into place with headcap cement. (e) After surgery, optical fiber protectors are placed over the implanted optical fibers to prevent damage to the fiber cores. (f) Before behavioral experimentation, optical patch cables are connected to the implanted optical fibers.

51| Align the implantable optical fiber such that it is perpendicular to the skull. If the implantable optical fiber is misaligned (i.e., not perpendicular to the skull), bend the stereotaxic adapter with hemostats or pliers to the correct position.

52| Lower the stereotaxic arm with the implantable optical fiber into the brain region of interest (Fig. 5c). When inserting the optical fiber in the brain, advance the fiber ventrally at a rate of ~2 mm min^{− 1}.

53| Apply an even layer of Titan bond (~50 µl) to the skull to ensure proper anchoring for the headcap. Use a cotton-tip applicator to evenly apply Titan Bond over the entire skull surface.

▲ CRITICAL STEP Wait approximately 10–15 min to allow the Titan Bond to dry completely.

? TROUBLESHOOTING

54| Fill a 1-ml syringe with the headcap cement (see REAGENT SETUP) and attach an 18-gauge syringe tip. Wipe off excess mixture from the tip of the syringe.

55| Apply the headcap cement evenly around the implantable optical fiber. Use a spatula to smooth the headcap into a sphere (Fig. 5d). Leave at least 4–5 mm of the ferrule of the implantable optical fiber exposed (i.e., not covered by headcap cement) to ensure that the fiber can be interfaced with patch cable.

▲ CRITICAL STEP Do not let the headcap mixture come in contact with the skin of the mouse. This will lead to improper anchoring as well as irritation to the mouse.

56| Let the headcap dry for ~15 min or until the cement is completely cured. If the headcap does not completely cover the skull, use Vetbond to suture the skin so that the skull is not exposed.

57| Place the optical fiber protectors over the ferrule on the implantable fiber (Fig. 5e).

58| Place the mouse into a clean cage over a heating blanket until the mouse fully recovers from anesthesia. Administer postsurgery analgesics if necessary.

59| Monitor the mouse’s body weight and visually inspect it daily to ensure proper recovery from surgery. The mouse will be ready for behavioral experimentation once its body weight returns to presurgery levels and when opsin expression is sufficient for the manipulation of neural circuits. Depending on the promoter used to drive opsin expression, this can take approximately 2–4 weeks for optimal expression in cell bodies and approximately 4–6 weeks for optimal expression in cell fibers and terminals.

Behavioral procedures for using implantable optical fibers ● TIMING 3–6 weeks

60| Construct a habituation tether, as described in Steps 61–64. Start by cutting ~50 cm of the furcation tubing.

61| Evenly apply a bead of heat-curable epoxy to one end of the cut furcation tube.

62| Insert approximately 2–3 mm of the ceramic split sleeve into the furcation tubing.

▲ CRITICAL STEP Do not insert more than 2–3 mm of the ceramic split sleeve. This will occlude the sleeve.

63| Using the heat gun to cure the epoxy.

▲ CRITICAL STEP Do not cure for more than 10 s. Overheating will melt the furcation tubing.

64| To habituate mice, remove the dust cap from the implantable optical fiber. Attach the habituation tether to the implantable optical fiber on the mouse. All mice should receive approximately 2–3 daily habituation sessions, which should last between 30 and 45 min.

▲ CRITICAL STEP Failure to habituate mice to tethering may result in disrupted/abnormal behavior.

65| Tape the excess furcation tube to the home cage.

▲ CRITICAL STEP Furcation tube should be long enough to give mouse full range of motion in the cage. Failure to do so will influence future experiments.

66| Once the mouse has been habituated to the tethering procedure, behavioral experiments can commence. Immediately before connecting the implanted optical fiber to the patch cable, adjust the light intensity at the end of the patch cable so that the value of light transmitted into brain tissue is sufficient to affect the desired tissue volume. This can be calculated using equations detailed by Aravanis et al.²⁰. For example, if the desired light output at the end of the implanted fiber is 10 mW and the percent transmission of the implantable fiber is 80%, then the light output at the end of the patch cable should be adjusted to 12.5 mW.

67| Restrain the mouse and connect the ceramic split sleeve on the patch cable to the implanted optical fiber (Fig. 5f). Use the window of the ceramic split sleeve to connect the patch cable to the implanted optical fiber.

▲ CRITICAL STEP Failure to do so will result in markedly diminished light output into the brain.

68| Begin behavioral experiments. Results show that chronic implantable fibers will work for a diverse array of behavioral tasks such as Pavlovian conditioned approach behavior, operant self-administration and conditioned place preference¹⁵. Furthermore, these methodologies should be adaptable to most rodent behavioral experiments.

69| At end of the experimental testing, perfuse the mouse with phosphate-buffered saline and 4% (wt/vol) paraformaldehyde and remove the headcap to test the implantable optical fiber light output. Store brains in paraformaldehyde for at least 24–48 h before slicing for histological verification of virus injection and optical fiber placement. Mice that are not euthanized for histological verification can be kept in their home cages until needed.

▲ CRITICAL STEP Try to remove the headcap in one motion by pulling up with pliers in one direction.

? TROUBLESHOOTING

Troubleshooting advice can be found in Table 1.

TABLE 1.

Troubleshooting table.

Step	Problem	Possible reason	Solution
19	The fiber does not fit into the ferrule	Improper ferrule bore diameter	Discard and use a new ferrule
25	The fiber core appears filmy or waxy; light emission is low and not concentric	Epoxy on the fiber core Cracked fiber core	Repolish with 1- and 0.3-μm polishing paper If the fiber core is cracked, remake the cable
34	Light emission is low	Poor connection between the implantable fiber and the patch cable	Use the window in the ferrule sleeve to determine whether the fiber is making direct contact with the cable
45	Inaccurate virus injection	Mouse head is not level	Adjust both anterior/posterior and medial/lateral tilt
53, Box 2	Headcap comes off during behavioral session	Improper anchoring of the implantable fiber to the skull	Allow Titan Bond to completely dry for 10 min. After headcap application, wait an additional 10–20 min to allow proper curing

? TIMING

Steps 1–35, Construction of implantable optical fibers, construction of patch cables for use in in vivo optogenetic experiments, measuring light output of optical fibers and patch cables, stereotaxic adapters for implantable optical fibers, implantation of optical fibers: 3 d

Steps 36–69, Behavioral procedures for using implantable optical fibers: 1 d plus 3–6 weeks

Box 1, Stereotaxic adapters for chronic implantable fibers: 1 d

Box 2, Interfacing implantable optical fibers with in vivo electrophysiological arrays: 1 d

ANTICIPATED RESULTS

With proper construction and implantation of optical fibers, consistent control of neural circuits over many behavioral sessions is achievable. Although we have examined subjects that have undergone optogenetic manipulations for at least 20 behavioral sessions without any damage to the implantable optical fiber, we predict that mice can be tested in longer-term behavioral studies ( >20 behavioral sessions) without any deterioration to the implantable optical fiber. Importantly, light transmission through the implantable optical fibers is relatively unchanged over many weeks to months ( > 6 months; Fig. 1a and D.R.S., unpublished observations) after its implantation into the brain. In addition, the damage to neural tissue associated with the implantation and long-term use of optical fibers in experiments is much less than that associated with acute optical fibers inserted through guide cannulas (Fig. 1b). Although we detail the use of chronically implantable optical fibers for optogenetic manipulations, with some modification, the use of these devices may be extended to monitor neural activity via excitation of genetically encoded calcium or voltage indicators²¹. Taken together, the use of implantable optical fibers provides unique advantages, such as enhanced experimental precision and throughput, thus allowing for sophisticated optogenetic manipulations during complex behavioral tasks.