The shrinkage of dental cement during curing displaces the baseplate. This protocol minimizes the problem by creating an initial foundation of the dental cement that leaves space to cement the baseplate. Weeks later, the baseplate can be cemented in position on this scaffold using little new cement, therebyreducing shrinkage.
Neuroscientists use miniature microscopes (miniscopes) to observe neuronal activity in freely behaving animals. The University of California, Los Angeles (UCLA) Miniscope team provides open resources for researchers to build miniscopes themselves. The V3 UCLA Miniscope is one of the most popular open-source miniscopes currently in use. It permits imaging of the fluorescence transients emitted from genetically modified neurons through an objective lens implanted on the superficial cortex (a one-lens system), or in deep brain areas through a combination of a relay lens implanted in the deep brain and an objective lens that is preanchored in the miniscope to observe the relayed image (a two-lens system). Even under optimal conditions (when neurons express fluorescence indicators and the relay lens has been properly implanted), a volume change of the dental cement between the baseplate and its attachment to the skull upon cement curing can cause misalignment with an altered distance between the objective and relay lenses, resulting in the poor image quality. A baseplate is a plate that helps mount the miniscope onto the skull and fixes the working distance between the objective and relay lenses. Thus, changes in the volume of the dental cement around the baseplate alter the distance between the lenses. The present protocol aims to minimize the misalignment problem caused by volume changes in the dental cement. The protocol reduces the misalignment by building an initial foundation of dental cement during relay lens implantation. The convalescence time after implantation is sufficient for the foundation of dental cement to cure the baseplate completely, so the baseplate can be cemented on this scaffold using as little new cement as possible. In the present article, we describe strategies for baseplating in mice to enable imaging of neuronal activity with an objective lens anchored in the miniscope.
Fluorescent activity reporters are ideal for imaging of the neuronal activity because they are sensitive and have large dynamic ranges1,2,3. Therefore, an increasing number of experiments are using fluorescence microscopy to directly observe neuronal activity1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16. The first miniaturized one-photon fluorescence microscope (miniscope) was designed in 2011 by Mark Schnitzer et al.5. This miniscope enables researchers to monitor the fluorescence dynamics of cerebellar cells in freely behaving animals5 (i.e., without any physical restraint, head restraint, sedation, or anesthesia to the animals). Currently, the technique can be applied to monitor superficial brain areas such as the cortex6,8,15,16; subcortical areas such as the dorsal hippocampus8,11,13,14 and striatum6,17; and deep brain areas such as the ventral hippocampus14, amygdala10,18, and hypothalamus8,12.
In recent years, several open-source miniscopes have been developed4,5,6,7,11,13,17,19. The miniscope can be economically assembled by researchers if they follow the step-by-step guidelines provided by the University of California, Los Angeles (UCLA) Miniscope team4,7,11,13. Because optical monitoring of neural activity is restricted by the limitations of light transmission7 to and from the neuronal population of interest, a miniscope was designed that requires an objective gradient refractive index (GRIN) lens (or objective lens) to be preanchored at the bottom of the miniscope to magnify the field of view that is relayed from a relay GRIN lens (or relay lens)6,7,8,10,16,17. This relay lens is implanted into the target brain region such that the fluorescence activity of the target brain region is relayed onto the surface of relay lens6,7,8,10,16,17. Approximately 1/4 of a full sinusoidal period of light travels through the objective GRIN lens (~ 0.25 pitch) (Figure 1A1), resulting in a magnified fluorescence image6,7. The objective lens is not always fixed at the bottom of the miniscope nor is the implantation of the relay lens necessary6,7,11,13,15. Specifically, there are two configurations: one with a fixed objective lens in the miniscope and a relay lens implanted in the brain8,10,12,14,16 (Figure 1B1) and another with just a removable objective lens6,7,11,13,15 (Figure 1B2). In the design based on the fixed objective and implanted relay lens combination, the fluorescence signals from the brain are brought to the top surface of the relay lens (Figure 1A1)7,8,10,12,14,16. Subsequently, the objective lens can magnify and transmit the visual field from the top surface of the relay lens (Figure 1A2). On the other hand, the removable objective GRIN lens design is more flexible, which means preimplantation of a relay lens into the brain is not mandatory (Figure 1B2)6,7,11,13,15. When using a miniscope based on a removable objective lens design, researchers still need to implant a lens into the target brain region but they can either implant an objective lens6,7,11,13,15 or a relay lens in the brain6,7. The choice of an objective or a relay lens for implantation determines the miniscope configuration that the researcher must use. For instance, the V3 UCLA Miniscope is based on a removable objective GRIN lens design. Researchers can choose either to directly implant an objective lens in the brain region of interest and mount the "empty" miniscope onto the objective lens6,7,11,13,15 (a one-lens system; Figure 1B2) or to implant a relay lens in the brain and mount a miniscope that is preanchored with an objective lens6,7 (a two-lens system; Figure 1B1). The miniscope then works as a fluorescence camera to capture livestream images of neuronal fluorescence produced by a genetically encoded calcium indicator1,2,3. After the miniscope is connected to a computer, these fluorescence images can be transferred to the computer and saved as video clips. Researchers can study neuronal activity by analyzing the relative changes in fluorescence with some analysis packages20,21 or write their codes for future analysis.
The V3 UCLA Miniscope provides flexibility for users to determine whether to image neuronal activity with a one- or two-lens system7. The choice of the recording system is based on the depth and size of the target brain area. In brief, a one-lens system can only image an area that is superficial (less than approximately 2.5 mm deep) and relatively large (larger than approximately 1.8 x 1.8 mm2) because the manufacturers only produce a certain size of objective lens. In contrast, a two-lens system can be applied to any target brain area. However, the dental cement for gluing the baseplate tends to cause misalignment with an altered distance between the objective and relay lenses, resulting in a poor image quality. If the two-lens system is being used, two working distances need to be precisely targeted to achieve the optimal imaging quality (Figure 1A). These two critical working distances are between the neurons and bottom surface of the relay lens, and between the top surface of the relay lens and the bottom surface of the objective lens (Figure 1A1). Any misalignment or misplacement of the lens outside of the working distance results in imaging failure (Figure 1C2). In contrast, the one-lens system requires only one precise working distance. However, objective lens size limits its application for monitoring of deep brain regions (the objective lens that fits the miniscope is approximately 1.8 ~ 2.0 mm6,11,13,15). Therefore, implantation of an objective lens is limited for the observation of the surface and relatively large brain regions, such as the cortex6,15 and dorsal cornu ammonis 1 (CA1) in mice11,13 . In addition, a large area of the cortex must be aspirated to target the dorsal CA111,13. Because of the limitation of the one-lens configuration that prevents imaging of deep brain regions, commercial miniscope systems offer only a combined objective lens/relay lens (two-lens) design. On the other hand, the V3 UCLA miniscope can be modified into either a one-lens or two-lens system because its objective lens is removable6,11,13,15. In other words, V3 UCLA miniscope users can take advantage of the removable lens by implanting it in the brain (creating a one-lens system), when performing experiments involving superficial brain observations (less than 2.5 mm in depth), or by preanchoring it in the miniscope and implanting a relay lens in the brain (creating a two-lens system), when performing experiments involving deep brain observations. The two-lens system can also be applied for superficially observing the brain, but the researcher must know the accurate working distances between the objective lens and relay lens. The main advantage of the one-lens system is that there is a decreased chance of missing the working distances than with a two-lens system, given that there are two working distances that need to be precisely targeted to achieve optimal imaging quality in the two-lens system (Figure 1A). Therefore, we recommend using a one-lens system for superficial brain observations. However, if the experiment requires imaging in the deep brain area, the researcher must learn to avoid misalignment of the two lenses.
The basic protocol for two-lens configuration of miniscopes for experiments includes lens implantation and baseplating8,10,16,17. Baseplating is the gluing of a baseplate onto an animal's head so that the miniscope can be eventually mounted on top of the animal and videotape the fluorescence signals of neurons (Figure 1B). This procedure involves using dental cement to glue the baseplate onto the skull (Figure 1C), but the shrinkage of dental cement can cause unacceptable changes in the distance between the implanted relay lens and the objective lens8,17. If the shifted distance between the two lenses is too large, cells cannot be brought into focus.
Detailed protocols for deep brain calcium imaging experiments using miniscopes have already been published8,10,16,17. The authors of these protocols have used the Inscopix system8,10,16 or other customized designs17 and have described the experimental procedures for viral selection, surgery, and baseplate attachment. However, their protocols cannot be precisely applied to other open-source systems, such as the V3 UCLA Miniscope system, NINscope6, and Finchscope19. Misalignment of the two lenses can occur during the recording in a two-lens configuration with a UCLA Miniscope due to the type of dental cement that is used to cement the baseplate to the skull8,17 (Figure 1C). The present protocol is needed because the distance between the implanted relay lens and the objective lens is prone to shift due to the undesirable shrinkage of dental cement during the baseplating procedure. During baseplating, the optimal working distance between the implanted relay lens and the objective lens must be found by adjusting the distance between the miniscope and the top of the relay lens, and the baseplate should then be glued at this ideal location. After the correct distance between the objective lens and the implanted relay lens is set, longitudinal measurements can be obtained at cellular resolution (Figure 1B; in vivo recording). Since the optimal range of working distances of a relay lens is small (50 – 350 µm)4,8, excessive cement shrinkage during curing can make it difficult to keep the objective lens and the implanted relay lens within the appropriate range. The overall goal of this report is to provide a protocol to reduce the shrinkage problems8,17 that occur during the baseplating procedure and to increase the success rate of miniscope recordings of fluorescence signals in a two-lens configuration. Successful miniscope recording is defined as recording of a livestream of noticeable relative changes in the fluorescence of individual neurons in a freely behaving animal. Although different brands of dental cement have different shrinkage rates, researchers can select a brand that has been previously tested6,7,8,10,11,12,13,14,15,16,22. However, not every brand is easy to obtain in some countries/regions due to the import regulations for medical materials. Therefore, we have developed methods to test the shrinkage rates of available dental cements and, importantly provide an alternative protocol that minimizes the shrinkage problem. The advantage over the present baseplating protocol is an increase in the success rate of calcium imaging with tools and cement that can be easily obtained in laboratories. The UCLA miniscope is used as an example, but the protocol is also applicable to other miniscopes. In this report, we describe an optimized baseplating procedure and also recommend some strategies for fitting the UCLA miniscope two-lens system (Figure 2A). Both examples of successful implantation (n= 3 mice) and examples of failed implantation (n=2 mice) for the two-lens configuration with the UCLA miniscope are presented along with the discussions for the reasons of the successes and failures.
All procedures performed in this study were approved by the National Taiwan University Animal Care and Use Committee (Approval No.: NTU-109-EL-00029 and NTU-108-EL-00158).
1. Assessment of the volume alteration of dental cement
NOTE: Changes in the volume of dental cement occur during the curing process. Test the volume changes of dental cement before implantation and baseplating. Researchers can test any brand of dental cement and use the brand with the lowest volume change to cement the baseplate. An example is shown in Figure 3.
2. Anesthesia, surgical implantation, viral injection, and dummy baseplating
3. Baseplating
NOTE: Usually, the baseplating procedure (Figure 5) cannot be performed within 2-3 weeks of the initial procedure due to the recovery and viral incubation time. The induction procedures for baseplating are similar to those described in steps 2.1.1 to 2.1.8. However, baseplating is not an invasive procedure, and the animal requires only light sedation. Therefore, 0.8-1.2% isoflurane is sufficient, as long as the animal cannot move on the stereotaxic frame. In addition, relatively light isoflurane anesthesia can also help facilitate the expression of the fluorescence transients of neurons during monitoring8 (Supplementary video S2).
Assessment of the dental cement volume alteration
Since the volume of dental cement changes during the curing process, it may significantly impact the imaging quality, given that the working distance of a GRIN lens is approximately 50 to 350 µm4,8. Therefore, two commercially available dental cements were tested in this case, Tempron and Tokuso, before the implantation and baseplating procedure (Figure 5). Videos were first evaluated to determine whether the cements had shrunk and then the volumes of cements were compared when completely dried. The same concentration and volumes of the initial mixtures of dental cements were loaded into pipette tips and videotaped (Supplementary video S1).Video S1 demonstrated that both cements shrunk during the drying process. Tokuso performed better than Tempron (i.e, it had a larger volume than Tempron even after shrinking): Figure 3B; mean volume of Tempron: 0.93 ± 0.07 mL; mean volume of Tokuso: 1.18 ± 0.07 mL; t-test: p = 0.048, t(6) = 2.46, suggesting that it shrunk less. The densities of the cements were also tested 4 times after complete drying. Tokuso had a lower mean density than Tempron (mean density of Tempron: 1.07 ± 0.12 g/mL; mean density of Tokuso: 0.93 ± 0.08 g/mL; t-test: p = 0.38, t(6) = -0.94), implying that it had a lower shrinkage rate. Therefore, Tokuso was used for the following implantation and baseplating steps. The purpose in describing the experiment above is not to recommend any brands but rather to remind experimenters to find a dental cement that does not undergo a considerable volume change when curing.
Measurement of image shifting
We simulated the dummy baseplating and original baseplating procedures on a fixed object to investigate whether the dummy baseplating procedure results in a smaller shift in the baseplate location than the original procedure. A light-cure adhesive was used to fix a 23 G 2.5 cm syringe needle in the center of a baseplate with 1 cm sticking out of the baseplate (Figure 3C). Then, baseplating was simulated by holding the needles with the arm of a stereotaxic instrument. Instead of using laboratory animals, a sticker was attached onto the table of the stereotaxic instrument. The syringe needle pierced the sticker, thereby representing the original location. Then, the needle was elevated by 0.5 mm and the simulation surgeries were started. In a surgery mimicking the one-time baseplating procedure8, dental cement was applied until it covered the baseplate, which was 1 cm above the sticker. Then, there was a 2 h wait period for the cement to cure before gently pushing the needle to pierce the sticker again. Since the light-cure adhesive did not completely stabilize the needle, the needle could be pushed deeper to create another hole, representing the location of the baseplate after the dental cement had dried. The distance between the initial hole and the second hole was measured to quantify the location shift of the needle. The results demonstrated that baseplate 1 cm above the skull resulted in a 1.76 ± 0.14 mm location shift, but the dummy baseplating resulted in shift of only 0.75 ± 0.17 mm (Figure 3D; mean shift distance of one-time baseplating versus dummy baseplating; t-test: p < 0.01, t(8) = 6.03).
Additionally, we were curious about whether the height of the dental cement would influence the shift. Therefore, we further tested a shorter height (Figure 3E; 0.5 cm above) by using the one-time baseplating procedure. The baseplate that was anchored 0.5 cm above the skull shifted significantly less than the baseplate 1 cm above the skull (Figure 3F; mean shift distance: 0.43 ± 0.11 vs 1.76 ± 0.14 mm; t-test: p < 0.01, t(8) = 9.73). Data suggested that the height of the baseplate above the skull and the distance the baseplate shifted were positively correlated. Since the dummy baseplating procedure is followed by a second baseplating step that uses only a minimal amount of dental cement, this data also supports the hypothesis that dummy baseplating helps improve the precision of the baseplating procedure.
Imaging using a two-lens miniscope system
By following the surgical protocols for viral infusion, dummy baseplating, and baseplating (Figure 2, Figure 4, Figure 5), we observed fluorescence transients of individual neurons in three mice (n = 3/5) (Figure 6A, Supplementary video S3) during the baseplating procedure and after the baseplating procedure while the mice were freely moving, confirming that the working distance between the objective and relay lenses remained the same (Figure 6A1 versus Figure 6A3, only a slight shift in position occurred) after the dental cement had completely dried. Here, we provide videos of the baseplating of one mouse (Supplementary video S2; mouse #5) and subsequent calcium imaging while mice were moving freely (Supplementary video S3; mouse #3, #4, #5; raw data). Please note that the data acquisition software for V3 does not contain functions for the background subtraction, and the contrast of the video can be improved by offline processing). For Supplementary video S2, a manual search was performed for the fluorescence signals and then the miniscope was attached to the stereotaxic arm for fine adjustments. Some fluorescence transients were observed during the probing (Supplementary video S2; Figure 6A). A transient is a smooth shift in brightness in gray or white spots, not a firing pattern resembling a flashing light; Figures 6A1 に 6A2 display images of transients. The video of freely behaving mice showed an acceptable spatial shift compared with the region of interest during the baseplating procedure (Supplementary video S3). The margins of the cells could not be significantly improved at this stage because we could not adjust the distance between the bottom of the relay lens and the active neurons. Notably, some issues may frequently occur during the baseplating procedure, including an absence of anything except a uniform background (Figure 6B; Supplementary video S4) or visibility of white cell margins without no fluorescence transients (Figure 6C; Supplementary video S5). Two examples of negative results are also provided (n = 2/5) (Figure 6B,C, Supplementary videos S4, S5). The potential reasons for the failures are examined in the Discussion section. The surgical procedures for these two mice were performed without using a 1 mm diameter needle to clear the pathway; thus, the relay lens was directly implanted. We propose that in mouse #2, the relay lens may have compressed the neurons and dragged some neurons down, resulting in damaged neurons. Therefore, fluorescence transients were not found.
Figure 1: Mechanisms of the miniscope and the defects that usually cause imaging failure. (A) Cartoon diagram of the relay lens and objective lens displaying the mechanisms necessary for visualization of fluorescent cells in a two-lens system. The signaling waves of each lens emphasize the importance of and the mechanism involved in finding a working distance to visualize the fluorescent cells. (A1) The relay GRIN lens is placed above the target neurons within the working distance. Since the pitch of the relay lens is 0.5*N (where N is an integer), the lens enables visualization of target neurons by relaying light, bringing the original fluorescence signals to the top of the relay lens. Then, approximately 1/4 of a full sinusoidal period of light travels through the objective GRIN lens (~ 0.25 pitch) and results in a magnified image. (A2) The objective lens transmits the images to a complementary metal-oxide-semiconductor (CMOS) sensor located on the top (the green plate in the diagram) of the miniscope. (B) Cartoon diagram presenting the differences in the GRIN lens implantation procedures between a (B1) two-lens system and (B2) a one-lens system. In general, the major difference is that the one-lens system directly uses its objective lens to image the surface of the brain; as a result, the objective lens needs to be implanted on top of the target neurons. (C) In contrast, in the two-lens system, in comparison, an additional relay lens is implanted in the brain and the objective lens is preanchored in the miniscope. (C1) The shape of the baseplate for the V3 version of the UCLA Miniscope may cause an uneven shift after the dental cement completely dries. (C2) This shift causes misalignment or deviation from the working distance between the relay lens and the objective lens. Please click here to view a larger version of this figure.
Figure 2: Modified protocols and detailed surgical protocols to improve the success rate of the two-lens miniscope system. (A1) Modified surgical protocol for the two-lens system. The locations of the viral injection, and relay GRIN lens implantation are shown. Dummy baseplating allows the working distance to accurately encompass the neurons of interest and reduces the positional shifting associated with the (A2) baseplating procedure. Hence, (A3) the probability of imaging fluorescent cells in freely behaving mice is increased. (B) Cartoon image of the steps for viral injection, and relay GRIN lens placement. (B1) First, mark the aspiration needle with a pen approximately 1 mm (for mouse experiments) from the tip to aid in assessment of the cortex aspiration depth. Then, aspirate the connective tissue and cortex (approximately 1 mm deep in mice) from the brain. (B2) Use the ventral hippocampus as the target area to implant a relay lens; notably, the tough corpus callosum is the landmark for stopping aspiration of the cortical region. The fibrous texture of the corpus callosum can be seen under the stereomicroscope during aspiration. (B3) Second, puncture the point of interest with an inner needle of a 20 G catheter (approximately 1 mm in diameter; the diameter is similar to that of the relay lens) to clear a pathway before viral infusion. Under the stereomicroscope, a dent (dashed circle in the picture) created by the needle should be visible. Lower the microinjection needle and relay lens at the dent (or only 250 µm away from the center of the dent). Finally, place the relay GRIN lens (in the present study, we used a relay lens 1 mm in diameter). Please click here to view a larger version of this figure.
Figure 3: Measurement of the volume change of the cured dental cement. (A) An example with cured dental cement of two brands. The black arrows indicate the original levels after we mixed the powder and solution. The white arrows indicate the levels after 10, 20, and 40 min of curing. The red arrows point to the bottoms of the tips, which were sealed with light-cure glue. (B) Mean volumes of the remaining dental cements after they completely dried. The bars and error bars are the means ± SEMs. * denotes significance (P < 0.05) as determined by Student's t-test. (C) Illustrations and a photograph explaining the method used to measure the location shift of the baseplate. A needle was glued through the baseplate and was used to simulate the original baseplating procedure (one-step baseplating) and the dummy baseplating procedure. The red and black arrows represent the locations of the needle before and after the dental cement was cured. (D) Mean distance changes of the tip of the needle (E) Illustrations of the method used to measure the relation between the baseplate above skull level and the location shift. (F) Mean distance changes of the tip of the needle. The bars and error bars are the means ± SEMs. ** denotes significance (P < 0.01) as determined by Student's t-test. Please click here to view a larger version of this figure.
Figure 4: Detailed dummy baseplating procedure (executed immediately after relay lens implantation). (A1) Three-step cartoon of objective lens, baseplate, and paraffin film assembly on the miniscope. (A2) Photograph version of the assembly. (B1) Cartoon showing the dummy baseplating procedure. First, the assembled miniscope is aligned with the relay lens. The objective lens positioned is as close to the relay lens as possible. Second, dental cement is added to the parafilm surrounding the baseplate and applied onto the skull to make a female mold for future baseplating. Then, the baseplate is removed, and molding silicon rubber (pink) is added and topped off with dental cement to protect the female mold and relay lens. The animal is then allowed to recover from anesthesia. After at least 3 weeks, the baseplating procedure is performed. (B2) Photograph version of (B1). Please click here to view a larger version of this figure.
Figure 5: Steps for the baseplating procedure. (A1) Attach the baseplate and (A2) manually adjust the focus slider to 2.7 to 3 mm. A setting of 2.7 to 3 mm (at the position shown in the picture) seems to be the best place to start before manually searching for the fluorescent cells. (B1) The first cartoon diagram illustrates the importance of the length between the objective lens and the relay lens for visualization of the fluorescent cells (fluorescent cells are usually observed at a distance between 0.5 and 2 mm). Before the fluorescent cell emerges, the researcher should see the margin of the relay lens (arrows in the top picture). It is helpful to first manually adjust the lens to the optimal viewing position because it is easier to make quick adjustments by hand. (B2) Place reusable adhesive clay onto the stereotaxic arm probe, and then stabilize the miniscope by adhering it onto the arm with reusable adhesive clay. At this stage, the region of interest may have shifted, but this can be easily fixed by slightly adjusting the stereotaxic arm and the reusable adhesive clay. After checking for and finding the optimal region of interest, add a small amount of dental cement (represented in red). Glue the baseplate with as little dental cement as possible. Separate the miniscope and baseplate after the dental cement dries. If the baseplate is not stable enough, add additional dental cement. (C) This diagram illustrates the importance of every step for achieving success. Please click here to view a larger version of this figure.
Figure 6: Examples of success and failure during the baseplating procedure. (A) An example of successful imaging during the baseplating procedure. The relay lens targeted the ventral hippocampus. Fluorescence transients were noticed when the animal was under anesthesia with isoflurane (see also Supplementary video S2). (A2) The contrast was increased for better visibility. The red dashed lines indicate blood vessels. (A3) Fluorescence transients five days after baseplating when the mouse (#5) was freely behaving in its homecage (see also Supplementary video S3; the images of mice #3 and #4 are also provided in the Supplementary video S3). Only a slight shift in position occurred. (B1) An example of a failure. During baseplating, only homogeneous white/gray areas were noticed that were not cell-like in shape (see also Supplementary video S4). (B2) The relay lens missed the target area and was located above a region with no infected cells. (C1) Another example of a failure. In this mouse, although many round cells were observed, no fluorescence transients were observed during the baseplating (after three weeks of incubation and while the mouse was anesthetized/sedated). Furthermore, no fluorescence transients were onserved when the baseplating procedure was performed again (in the fifth week of incubation) or even when the animal was freely behaving. The image was obtained during baseplating at the fifth week of incubation (see also Supplementary video S5). (C2 and C3) The relay lens missed the pyramidal layer of the ventral hippocampus. The blue dots are nuclei that were stained with DAPI. Please click here to view a larger version of this figure.
Problem | Possible cause | Solution | ||
Had good imaging when baseplating but completely out of focus the next day | Volume change upon curing the dental cement between the subject’s skull and the baseplate | Proceed “dummy baseplating” after lens implantation procedure as suggested in the present report. Choose a dental cement that has less volume change. | ||
Has homogeneous white/gray areas without any cell-like shape during baseplating (also see supplementary video S4) | A. The relay lens missed the target area B. Bad expression of fluorescence indicator C. Forget to install an objective lens beneath the miniscope where the fluorescence indicator is actually expressed | A. Use dummy lenses for practicing surgeries. Check the coordinates of the dummy lens after the practicing implantation B. Find an optimal viral titer for calcium imaging experiments (referred to Resendez et al. 2016). If for some reason, pretesting cannot be conducted, we recommend that beginners try a higher titer first. (See Disscusion section for details) C. Screw on an objective GRIN lens (~ 0.25 pitch. ex: Edmund Optics; Stock #64-519) at the bottom of miniscope. | ||
Observe many cells but no fluorescence dynamics during baseplating | A. Unhealthy neurons or dead neurons B. Sparse fining type of neurons C. Deep anesthesia status | A. Be careful for every surgical procedure. Use a new syringing needle which is similar to the diameter of your relay lens to cut the infusion/implantation path first to release the pressure by lens implantation. Slowly infuse the virus and slowly implant the lens. B. Wait for a longer time to monitor the activities of neurons or pinch the tail a little bit to stimulate the mouse. C. Baseplating is not an invasive procedure, and the animal only needs sedation. Therefore, maintaining 1.2 ~ 0.8% isoflurane is sufficient, as long as the animal is unable to move on the stereotaxic frame. |
Table 1: Troubleshooting chart.
Supplementary video S1: Dental cement volume test. The difference between the volume were measured at the beginning of the dental cement preparation and that the volume after the cement dried. Two brands of dental cement (e.g., Tempron and Tokuso Curefast) were tested. To test the volume of the dental cement after it dried, 0.5 g of each dental cement powder were weighed and mixed with its appropriate solution (1 mL). Then, 0.1-10 µL pipette tips were loaded with 2.5 µL of the dental cement mixture and sealed with light-cure glue. Then, the tips were placed on a rack, the surfaces of the dental cement mixtures were marked, and videotaped for 40 minutes. Please click here to download this Video.
Supplementary video S2: Demonstrations of the baseplating procedure. The angle between the surface of the objective lens and relay lens was adjusted by turning the ear bar, tooth bar, or reusable adhesive clay on the z-axis. The margin of the relay lens appeared as a moon-like gray circle on the monitor. Successful viral expression and lens implantation should reveal at least one fluorescence dynamic (arrows) during baseplating. Please click here to download this Video.
Supplementary video S3: Demonstrations of fluorescence dynamics in freely behaving mice. Fluorescence transients from mice #3, #4, and #5. Please click here to download this Video.
Supplementary video S4: An example of a failure. A uniform background appeared during the baseplating procedure. Please note that the alteration in brightness was not because of the fluorescence transients of individual cells; it was caused by the search procedure. Please click here to download this Video.
Supplementary video S5: Demonstration of another failure example. A lack of fluorescence transients occurred during the baseplating procedure, but some round cell margins were still visible. Please note that the alteration in brightness was not caused by the fluorescence transients of individual cells; it was caused by the search procedure. Please click here to download this Video.
The present report describes a detailed experimental protocol for researchers using the two-lens UCLA Miniscope system. The tools designed in our protocol are relatively affordable for any laboratory that wishes to try in vivo calcium imaging. Some protocols, such as viral injection, lens implantation, dummy baseplating, and baseplating, could also be used for other versions of the miniscope system to improve the success rate of calcium imaging. Other than general problems with viral injection, and lens implantation that need to be fixed regardless of the miniscope version, the structure of the UCLA baseplate may cause a positional shift, which can lead to mismatched working distances between the objective and relay lenses (Figure 1C). A modified experimental protocol was developed to minimize the positional shift of the two lenses (Figures 2, Figure 4, Figure 5). Successful calcium imaging can be achieved through a combination of successful viral expression, lens implantation, and baseplating procedures (Figure 5C). Using a two-lens system to observe deep brain areas requires a high level of accuracy in the baseplating procedures because the relative positions of the implanted relay lens and the objective lens also need to be accurate. This report not only presents a protocol that solves baseplating issues but also discusses some tips for viral injection, and lens implantation. Below, we discuss the baseplating procedures first, followed by the viral injection, and lens implantation procedures, and we describe some examples of failures (Figure 6).
Baseplating
The V3 version of the UCLA Miniscope is currently the most commonly used design and can be ordered online. The baseplate of this version consists of two edges without walls (Figure 1C1), but its shape can cause an uneven shift (Figure 1C1, C2) after the dental cement is completely dry. The dummy baseplating procedures in our protocol (Figure 4) minimizes the misalignment and baseplate shifting issues because it uses paraffin film to reduce the available space for the dental cement during real baseplating (Figure 5B). The remaining space is still sufficient for finding an optimal imaging position during the real baseplating procedure. Therefore, the researcher can glue the baseplate with as little dental cement as possible (Figure 5B2). In addition, dummy baseplating procedure takes relatively little time (approximately 15 to 20 min) since it does not require a perfect distance between the objective lens and relay lens to be achieved. However, dummy baseplating is just one of the key factors for successful calcium imaging; for example, during monitoring of fluorescence signals, neurons may appear blurry if they are slightly off the working plane (depending on the working distance of the relay lens; often approximately 100 ~ 300 µm). The ability to improve the working distance is limited, since the working distance of the neurons is predetermined by the relay lens implantation procedure. An experiment was performed without the dummy baseplating procedure and the region of interest was always (in four out of four mice) missed after one-time baseplating. Therefore, these solutions were devised to reduce the shrinkage problem. Some light-cure dental cements may avoid the problem of shrinkage during curing because they cure very quickly; these cements may help researchers adjust the distance during the baseplating procedure. Although the range of wavelengths for exciting the calcium indicators and curing the dental cement are different, photobleaching during the baseplating procedure needs to be prevented. In order to prevent photobleaching, the range of wavelengths generated by the light source needs to be fairly narrow to avoid cross-reactions between the calcium indicators and the light-cure dental cement. Therefore, the use of light-cure dental cement is not recommended for the baseplating procedure. In addition, some specific brands of dental cements are frequently used for gluing baseplates by other research teams6,7,8,10,11,12,13,14,15,16,22. The low-shrinkage properties of these cements may solve the problem, but we did not have the opportunity to test these brands due to difficulties in purchasing (importing) them. Medical-grade products may be subject to import regulations depending on the country/region. If researchers have difficulty accessing the cements mentioned in the literature6,7,8,10,11,12,13,14,15,16,22, our protocol will solve the dental cement shrinkage problem.
In addition to the shrinkage problem, structural wear can occur in the miniscope itself and on the baseplate (especially the magnets in the baseplate) over the course of many experiments. Any small gap caused by wear reduces the stability of the distance between the two lenses. Again, accuracy over the whole optical path is key to imaging of calcium transients in freely behaving mice using a miniscope.
Troubleshooting of viral injection/lens implantation
Before infusing the virus into the brain region, it is recommended to mount a new needle onto the stereotaxic arm and slowly puncture the brain at the appropriate coordinates. Because the new needle is very sharp, this puncturing serves as a preparation step to clear the path (or cut a path) for the actual microinjection needle and relay lens and minimizes tissue damage, which can be caused by compression from the microinjection needle or relay lens. The sharp needle cuts a path for the blunt relay lens; thus, the tip of the needle should be lowered to the same coordinates with the relay lens. The inner needle of a 20 G I.V. catheter was chosen because it had a diameter of 1 mm, which was similar to the diameter of our relay lens. A different needle gauge can be used depending on the diameter of the relay lens.
Although the present study did not focus on the properties and titers of viral vectors, these factors are still critical for successful calcium imaging (Figure 5C). It is necessary to pretest the expression quality of the viral vector. Steps 1 to 5 of Resendez et al.'s protocol provide detailed methods for the identification of an optimal viral titer for calcium imaging experiments8. If pretesting cannot be conducted for some reason, it is recommended that beginners try a relatively high titer first. One of the disadvantages of using a high viral titer iscytotoxicity8. Dead neurons give rise to autofluorescence and are visible as bright white spots under the miniscope8. However, these dead neurons are good landmarks for beginners to find and practice the baseplating procedure. On the other hand, although infusion of a low-titer virus has a lower risk of killing cells, the fluorescence may be masked by the background if the cells do not show any fluorescence transients during the baseplating procedure. It is difficult to pinpoint the reasons for failure because the target neurons may be missed due to relay lens implantation or due to the poor expression by the viral vector. Determining the cause is very confusing without histological confirmation. Therefore, using relatively high-titer viruses is recommend if the viral vectors cannot be pretested.
Resendez et al. recommend inserting the microinjection needle 250 µm lateral and ventral to the placement of the lens. Our observations demonstrated that infusion of the virus at the same depth as the relay lens can also lead to the expression of good calcium signals. We propose that due to the possibility of infection and spread, infusing the virus at the same depth as the coordinate of the relay lens is optimal for reducing tissue damage and increasing the accuracy of the distance between the infected neurons and the relay lens. Resendez et al also recommend performing viral injection, and lens implantation during the same surgery given the narrow range of working distances between target cells and the relay lens8. However, some researchers follow a protocol in which the viral infusion step and lens implantation step are conducted two (or more) weeks apart18,26. Lengthening the time between steps reduces the individual operative time. But in this protocol these two steps were merged into one surgery, because in the one-surgery protocol, the surgeon can monitor the position between the microinjection needle and the relay lens when lowering them into the target area, which reduces the chance of failed targeting (Figure 2B3).
For two mice (mice #1 and #2), the use of a needle to create a path for the lens was also omitted, and in mouse #2, round cell-like gray spots were observed (Figure 6C). However, fluorescence dynamics (Supplementary video S5) were not observed. Upon examining the brain slice of this example (Figure 6C2, C3), we postulated that the relay lens dragged some cells down while it was being lowered. These dragged neurons were unhealthy, so no activity was noticed during the baseplating procedure or during the experiment.
We had one mouse with a completely homogenous gray signal and no visible cell-like images during the baseplating procedure (Figure 6B1). After sacrificing the mouse its brain slices were observed. It was noticed that the lens was on the medial side of the curved pyramidal layer; the target area (Figure 6B2) was completely missed. These unsuccessful attempts were critical experiences that led us to modify our surgery protocol and ultimately succeed with three mice (Supplementary video S3).
Microinjection needle
The corpus callosum is a tough fibrous tract that overlays dorsal hippocampus. Because of its stiffness, it resists penetration by a flat-tipped microinjection needle. Therefore, we recommend using a microinjection needle with a beveled tip. This type of needle not only reduces tissue damage but also reduces the possibility of failure to infuse the virus into the target area due to blockage by nearby fibers. We have summarized the abovementioned problems in a troubleshooting chart (Table 1).
Time frame
The time frame for the entire procedure is summarized in Figure 2A, with days as the units. In detail, the surgery part (viral injection, relay lens implantation, dummy baseplating) may take 3 h for beginners or 1.5 h for sufficiently trained researchers. Mice have a small body size and fast metabolism, and they are more susceptible to health problems caused by long periods under anesthesia than larger species27. Researchers should aim to keep the time spent in surgery under 3 h. Real baseplating can be performed after 3 to 6 weeks of calcium indicator expression. This procedure may take 1 h. The subsequent longitudinal calcium activity observation can be continued for more than 1 month.
Conclusions
The UCLA Miniscope is an open-source in vivo calcium imaging device. Without a ready-made, preinstalled baseplate or lens module, individuals conducting experiments need to be able to accurately achieve the optimal distance between the objective lens and the relay lens during the baseplating procedure. The increased difficulty of using the two-lens UCLA Miniscope system lies in the volume change of the dental cement. We have optimized the original protocols and minimized the defects caused by the problems mentioned above, thus increasing the success rate of calcium imaging with the two-lens UCLA Miniscope system.
The authors have nothing to disclose.
This work was supported by the Ministry of Science and Technology, Taiwan (108-2320-B-002 -074, 109-2320-B-002-023-MY2).
0.7-mm drill bit | #19008-07 | Fine Science Tools; USA | for surgery |
0.1–10 μl pipette tips | 104-Q; QSP | Fisher Scientific; Singapore | for testing dental cement |
20 G IV cathater | #SR-OX2032CA | Terumo Corporation; Tokyo, Japan | for surgery |
27 G needle | AGANI, AN*2713R | Terumo Corporation; Tokyo, Japan | for surgery |
AAV9-syn-jGCaMP7s-WPRE | #104487-AAV9; 1.5*10^13 | Addgene viral prep; MA, USA | for viral injection |
Atropine sulfate | Astart; Hsinchu, Taiwan | for surgery/dummy baseplating/baseplating | |
Baseplate | V3 | http://miniscope.org | for dummy baseplating/baseplating |
BLU TACK | #30840350 | Bostik; Chelsea, Massachusetts, USA | Reusable adhesive clay; for surgery/dummy baseplating/baseplating |
Bone Rongeur Friedman | 13 cm | Diener; Tuttlingen, Germany | for baseplating |
Buprenorphine | INDIVIOR; UK | for surgery | |
Carprofen | Rimadyl | Zoetis; Exton, PA | analgesia |
Ceftazidime | Taiwan Biotech; Taiwan | prevent infection | |
Data Acquisition PCB for UCLA Miniscope | purchased on https://www.labmaker.org/collections/neuroscience/products/data-aquistion-system-daq | for baseplating | |
Dental cement set | Tempron | GC Corp; Tokyo, Japan | for testing dental cement |
Dental cement set | Tokuso Curefast | Tokuyama Dental Corp.; Tokyo, Japan | for testing dental cement/surgery/dummy baseplating/baseplating |
Dual Lab Standard with Mouse and Rat Adaptors | #51673 | Stoelting Co; Illinois, USA | for surgery/dummy baseplating/baseplating |
Duratear ointment | Alcon; Geneva, Switzerland | for surgery/dummy baseplating/baseplating | |
Ibuprofen | YungShin; Taiwan | analgesia | |
Isoflurane | Panion & BF Biotech INC.; Taoyuan, Taiwan | for surgery/dummy baseplating/baseplating | |
Inscopix | nVista System | Inscopix; Palo Alto, CA | for comparison with V3 UCLA Miniscope |
Ketamine | Pfizer; NY, NY | for euthanasia | |
Normal saline | for surgery | ||
Micro bulldog clamps | #12.102.04 | Dimedo; Tuttlingen, Germany | for lens implantation |
Microliter Microsyringes, 2.0 µL, 25 gauge | #88400 | Hamilton; Bonaduz, Switzerland | for viral injection |
Molding silicone rubber | ZA22 Thixo | Zhermack; Badia Polesine, Italy | for dummy baseplating |
Objective Gradient index (GRIN) lens | #64519 | Edmund Optics; NJ, USA | for dummy baseplating/baseplating |
Parafilm | #PM996 | Bemis; Neenah, USA | for dummy baseplating |
Portable Suction | #DF-750 | Doctor's Friend Medical Instrument Co., Inc., Taichung, Taiwan | for surgery |
Relay GRIN lens | #1050-002177 | Inscopix; Palo Alto, CA, USA | for dummy baseplating/baseplating |
Stainless steel anchor screws | 1.00 mm diameter, total length 3.00 mm | for surgery | |
Stereo microscope | #SL720 | Sage Vison; New Taipei City, Taiwan | for surgery/dummy baseplating/baseplating |
Stereotaxic apparatus | #51673 | Stoelting; IL, USA | for surgery/dummy baseplating/baseplating |
UV Cure Adhesive | #3321 | Loctite; Düsseldorf, Germany | for testing dental cement |
V3 UCLA Miniscope | purchased on https://www.labmaker.org/products/miniscope-complete-set-of-components | for surgery/dummy baseplating/baseplating | |
Xylazine | X1126 | Sigma-Aldrich; St. Louis, MO | for euthanasia |
Xylocaine pump spray 10% | AstraZeneca; Södertälje, Sweden | for surgery |