- Email: [email protected]
Population response characteristics of intrinsic signals in the cat somatosensory cortex following canine mechanical stimulation
Neuroscience 329 (2016) 254–263
POPULATION RESPONSE CHARACTERISTICS OF INTRINSIC SIGNALS IN THE CAT SOMATOSENSORY CORTEX FOLLOWING CANINE MECHANICAL STIMULATION JIANXIANG TAO, ay JIAN WANG, by ZHONG LI, b JIANJUN MENG b AND HONGBO YU b*
dentures) such as tissue reactions, biomechanics, and occlusal features have been studied extensively. However, neural changes that occur when the stomatognathic system is altered (e.g., tooth loss, rehabilitative processes, and tooth pain) have been seldom addressed. Recently, attention has been focussed on these neural changes (Henry et al., 2005; Jantsch et al., 2005; Avivi-Arber et al., 2010) and the cortical adaptive processes underlying these rehabilitative processes (Yan et al., 2008; Habre-Hallage et al., 2012). Changes in areas of the central nervous system (CNS) elicited by alterations to the intraoral tissues are thought to be crucial for a person’s ability to learn and for the sensorimotor system to adapt to the new intraoral environment (Calford, 2005; Abarca et al., 2006; Trulsson et al., 2012). Thus, evaluating neural responses in the cortex in response to tooth stimulation in vivo would be extremely useful. For half a century, most knowledge of cortical function was obtained from single-unit recordings, which provided detailed firing information of individual neurons. Due to extremely intricate cortical circuitry, it is difficult to evaluate the relative contribution of each neuron to oral function and related mapping information in a large cortical area. Therefore, brain imaging techniques, such as functional magnetic resonance imaging (fMRI), have been used to demonstrate neural changes or neurop lasticity of the sensorimotor cortex for cross-sectional studies in human subjects and animals (Yan et al., 2008; Kimoto et al., 2011; Habre-Hallage et al., 2012). However, the spatial and temporal resolution of fMRI is low, and it cannot provide detailed and insightful information. In the past two decades, intrinsic signal optical imaging, which is based on changes in blood oxygen consumption, has been widely used to establish detailed functional maps in the sensory cortex (Grinvald et al., 1999). Intrinsic signal optical imaging is a valid, reliable, and a unique technique to investigate neuronal responses of millions of cells in the visual cortex (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014). Moreover, since it requires no chemicals or virus injections for labeling, this technique has no toxic effect and can record functional signals in the cortex from the same animal for an extended period. Thus, it is suitable for detecting cortical neuroplasticity for longitudinal studies (Kim and Jun, 2013; Bauer et al., 2014). Optical imaging with a voltage-sensitive dye was used to identify cortical responses evoked by electrical stimulation of the periodontal ligaments and dental pulps
a
Shanghai Engineering Research Center of Tooth Restoration and Regeneration, Department of Prosthodontics, School of Stomatology, Tongji University, Shanghai 200072, China b
Vision Research Laboratory, School of Life Sciences, The State Key Laboratory of Medical Neurobiology, Collaborative Innovation Center for Brain Science, Fudan University, Shanghai 200433, China
Abstract—Intrinsic signal optical imaging has been widely used to measure functional maps in various sensory cortices due to better spatial resolution and sensitivity for detecting cortical neuroplasticity. However, application of this technique in dentistry has not been reported. In this study, intrinsic signal optical imaging was used to investigate mechanically driven responses in the cat somatosensory cortex, when punctate mechanical stimuli were applied to maxillary canines. The global signal and its spatial organization pattern were obtained. Global signal strength gradually increased with stimulus strength. There was no significant difference in response strength between contralateral and ipsilateral mechanical stimulation. A slightly greater response was recorded in the sigmoidal gyrus than in the coronal gyrus. The cat somatosensory cortex activated by sensory inputs from mechanical stimulation of canines lacks both topographical and functional organization. It is not organized into columns that represent sensory input from each tooth or direction of stimulation. These results demonstrate that intrinsic signal optical imaging is a valid tool for investigating neural responses and neuroplasticity in the somatosensory cortex that represents teeth. Ó 2016 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: neuroimaging, cuspid, physical stimulation, brain, evoked potentials, optical devices.
INTRODUCTION Many characteristics of prosthodontics rehabilitative procedures (e.g., implants, crowns, bridges, and *Corresponding author. Address: Vision Research Laboratory, School of Life Sciences, The State Key Laboratory of Medical Neurobiology, Collaborative Innovation Center for Brain Science, Fudan University, Shanghai 200433, China. Fax: +86-65650149. E-mail address: [email protected] (H. Yu). y Authors contributing equally to this work. Abbreviations: BOLD, blood oxygenation level-dependent; fMRI, functional magnetic resonance imaging; ROI, regions of interest. http://dx.doi.org/10.1016/j.neuroscience.2016.04.052 0306-4522/Ó 2016 IBRO. Published by Elsevier Ltd. All rights reserved. 254
J. Tao et al. / Neuroscience 329 (2016) 254–263
in rats (Horinuki et al., 2015; Nakamura et al., 2016). However, examination of both topographical and functional organizations representing different individual teeth has been seldom reported. Unlike previous electrophysiological single-cell recordings, intrinsic optical imaging is a good approach for investigating neuronal responses of the cortex at the population level (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014). The purpose of this study was to use intrinsic signal optical imaging to examine population response characteristics of the cat primary somatosensory cortex following mechanical stimulation of canines and to evaluate the application of intrinsic signal optical imaging in dentistry.
EXPERIMENTAL PROCEDURES Animal preparation Seven healthy adult cats of either gender weighing between 1.5 and 2.5 kg were used in the study (Shanghai Yingen Farm). The experimental protocol used in this study was approved by the Animal Experimentation Committee at Fudan University. Animal treatments were performed in strict accordance with the Guide for the Care and Use of Laboratory Animals described by Fudan University and the U.S. National Institutes of Health. Animals were prepared as described in our previous studies (Tong et al., 2011; Chen et al., 2014). Craniotomy was performed over the coronal sulcus (Fig. 1B, C). A stainless steel imaging chamber was positioned and fixed with dental cement on the skull, covering the oral area portion of the somatosensory cortex. After careful removal of the dura, the chamber was filled with warm (37 °C) silicone oil and sealed with a transparent glass window. Punctate mechanical stimulation Square planes (2 2 mm) were fabricated in the cusps of maxillary canine teeth using dental filling resin. Punctate
255
mechanical stimuli were applied on either the left or right canine tooth in the anterior-posterior direction using von Frey filament (North Coast Medical, Inc., Gilroy, CA, USA) to impact the square resin plane. The von Frey filament was attached to a rail electrode (JF-1683, Leqing, China) (Fig. 1A). The stimulus was presented for 4 s (a constant frequency of 1 Hz, square pulse with 0.5 s on and 0.5 s off) and then repeated with 18 s interstimulus intervals. The strength of mechanical stimulation increased gradually from10 g (filament N.14), 15 g (N.15), 26 g (N.16), 60 g (N.17), to 100 g (N.18). Optical data acquisition A custom-made optical imaging system was employed. As in our previous studies (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014), a slow scan CCD camera (1600 1200 pixels, 6.7 6.7 lm/pixel; PCO, Inc., PCO, AG, Germany) was used to record optical images of intrinsic signals from the exposed portion of the somatosensory cortex (Fig. 1A). Data acquisition started 3 s before the mechanical stimulus and continued 3 s after the stimulus. Therefore, 10 frames at a frame rate of 1 Hz were recorded. To allow relaxation of activitydependent microvascular changes, the period of stimulus presentation was followed by an 18-s rest interval. Each stimulus was presented 24 times, and the evoked signals were averaged. The details of optical data acquisition have been described in our previous studies (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014) Data analysis The data were analyzed as described in our previous studies (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014). The resultant frames of the ratio map (dR/R = R/ Ra1) were defined as the global signal maps. In the image of the global signal map, the cortex with more reflectance reduction had the stronger signal and a more negative value. To calculate the strength of the global sig-
Fig. 1. The experimental setup for intrinsic signal optical imaging of the tooth-related area of the somatosensory cortex. (A) Diagram of the experimental setup. Intrinsic signal optical imaging was performed in the exposed primary somatosensory cortex of an anesthetized cat when punctate mechanical stimuli were applied to the left or right canine tooth. (B) Photograph of the cortical surface of the right hemisphere. Craniotomy was performed over the outlined region. Abbreviations: Cor.S., coronal sulcus; Cru.S., cruciate sulcus; Sups.S., suprasylvian sulcus; Lat.S., lateral sulcus. (C) The blood vessel map of a brain imaged via a transparent window.
256
J. Tao et al. / Neuroscience 329 (2016) 254–263
nal, the exposed somatosensory cortex region was outlined as regions of interest (ROI), and dR/R of all pixels in the ROIs were averaged as the signal strength for each stimulus. The mean and standard error were calculated across different pairs, and Student’s t-test was performed.
RESULTS Determining the cerebral cortex driven by canine inputs
be similar (Fig. 3C). Therefore, in this study, the camera was focused on the plane at 500 lm below the pial surface. We then examine intrinsic signals evoked by stimulations in various directions. A 60-g mechanical stimulus was applied to canines from four directions including axial (bottom-up), and angled directions (lateral to medial, medial to lateral, and straight anterior–posterior). The exposed cortex could be activated from all four directions (Fig. 3D, E). The spatial patterns of global signal maps for different directions seemed to be similar, and the differential map did not show very obvious spatial specificity for a certain direction (Fig. 3G). However, the global signal for anterior-posterior stimulation exhibited the highest strength (Fig. 3F). Moreover, the force of anteriorposterior stimulation was more accurate and easier to apply than the other directions. For this reason, anteriorposterior stimulation was determined.
The cerebral cortex driven by the cat canine periodontium was reported to be in the vicinity of the coronal sulcus (Landgren and Olsson, 1980; Taira, 1987; Watanabe et al., 1991) as shown in Fig. 1B. We exposed this area (Fig. 1C), made a transparent window over it, and examined the stimulus preference of the recording area by intrinsic signal optical imaging (Fig. 1A). The global signal map elicited by 60-g mechanical stimuli on the contralateral canine was compared with that elicited by 60-g mechanical stimuli on the contralateral lip (Fig. 2). Due to increased oxygen consumption, the exposed somatosensory cortex became darker after the onset of canine mechanical stimulation, while the unexposed skull region remained unchanged (Fig. 2B, right). In contrast, the exposed somatosensory cortex showed almost no changes after onset of identical mechanical stimuli to the lip (Fig. 2B, left), which suggests that 60-g von Frey filament stimulation only of the canine could evoke responses in the related cortical region. It is possible that this stimulation (60 g) was not optimal for the lip or that the region representing the lip was located at a different site. We selected a ROI (Fig. 2A, red box), and the averaged values of all selected pixels demonstrated a significant decrease after onset of the canine stimuli (Fig. 2C, the 4–7th frames, 1 frame/s, n = 24 trials); the peak appeared at the 6th frame. Clearly, this global signal decreased after canine stimulation, but remained flat after lip stimulation (Fig. 2C, blue vs. red curves). The global signal map was not uniform, and the contour lines showed a gradual decrease of the responses from the center (Fig. 2D, E). The global signals for 6 ROIs (Figs. 2D, 6 small boxes) were examined further, and the center area exhibited the greatest global signal with an increasingly weaker response in the surrounding area (Fig. 2F), which is consistent with a previous study that used field potential recordings (Watanabe et al., 1991). Therefore, the exposed somatosensory cortex close to the coronal sulcus was confirmed to be driven specifically by the canine periodontium.
We then investigated the effect of stimulus strength on the global signal. In a typical case, a 10-g mechanical stimulation did not evoke a response above threshold, and the global signal map (Fig. 4B, top) showed almost no changes after onset of mechanical stimulation. The cortex representing teeth became darker after the onset of the 15-g stimulus. The map became darker as stimulus strength increased (Fig. 4B). The global signal of a ROI (Fig. 4A, 2 3 mm) demonstrated a significant decrease after onset of the stimulus (Fig. 4C. n = 24 trials for each mechanical stimulus). It is important to note that signal strength gradually increased with stimulus strength and saturated at the 60 g of mechanical stimulation (Fig. 4D, n = 5 animals). It has been shown that two groups of mechanoreceptors (pain and mechanical sensation) lie in the periodontal ligament, and very strong forces might evoke noxious response. In our experiment, stimulations less than 100 g were applied to avoid this possible effect. It is notable that weak (15 or 26 g) and stronger (60 or 100 g) stimulation evoked similar spatial activation patterns (Fig. 4B), and no obvious spatial segregation could be identified in the differential maps (Fig. 4E, left: 60–15 g, and right: 60–26 g maps), which suggests a non-noxious response. In fact, axial stimulation (bottomup) was also used in an attempt to activate specifically pure mechanical receptors in the apex region of the tooth root, and a similar activation pattern was obtained (Fig. 3D, left panel).
Determining focal plane of the camera and direction of mechanical stimulation
Equivalent bilateral sensory inputs to the toothrelated cortex
The signals were obtained from 500 lm below the pial surface, as suggested in most of the intrinsic signal optical imaging experiments (Grinvald et al., 1999). To examine the possible effect of focal plane, we changed the focal plane of the camera from 300 to 500 lm below the pial surface. The map patterns of global signals remained unchanged at 300, 400, and 500 lm (Fig. 3B). The time-course curves of the global signal seemed to
Peripheral sensory inputs tend to project to the contralateral hemisphere of the cortex. For example, the visual field (divided by the central azimuth line) is represented in the primary visual cortex of the contralateral hemisphere, and left and right eye (the retina responsive to the contralateral visual field) dominated cortical regions are interleaved in the same area of the cortex to form the well-known ocular
Measuring the linear relationship between stimulus strength and global intrinsic signal
J. Tao et al. / Neuroscience 329 (2016) 254–263
257
Fig. 2. Determination of the cerebral cortex driven by canine inputs. (A) Blood vessel map of an imaged brain located near the lateral end point of the coronal sulcus. The outlined region is the ROI used for analysis of global signals. L, lateral; A, anterior. (B) Global signal maps elicited by 60-g mechanical stimuli on the contralateral canine or the neighboring lips. Data acquisition started 3 s before the mechanical stimulus; 10 frames (1 frame/s) were obtained and consecutively presented. Arrow: the stimulus onset time point. (C) The time course curve of the global signal in a full data acquisition trial for canine and lip mechanical stimuli. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (D) The blood vessel map and 7th frame of the global signal map elicited by canine stimulation, with 6 ROIs (6 small boxes) on the exposed somatosensory cortex. (E) The contour lines of the global signal map at the 7th frame with 6 ROIs. (F) The time course curve of the global signal for 6 ROIs (n = 24 trials). Center area exhibited the greatest global signal. Standard errors are shown in C and F. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
dominance map (Hubel et al., 1978). It is of interest to examine whether the tooth-related cortex has a similar contralateral dominance and whether inputs from different teeth are interleaved as well. In this study, ipsilateral and contralateral canines were stimulated using a 60 g von Frey filament, and the elicited global signal map and time course curves of the global signal were examined (Fig. 5B, C). It appeared that there was no significant difference between contralateral and ipsilateral canine-dominated maps. We further calculated the tooth dominance index for each pixel [(Contra. Ipsi.)/(Contra. + Ipsi.)], and the distribution of all pixels was a Gaussian curve that peaked at approximately 0, which suggests no bias for contralateral dominance (Fig. 5E). To examine the
possible spatial segregation for ipsilateral and contralateral canine inputs, the 7th frames of evoked global maps for each canine were subtracted and illustrated in Fig. 5D. However, the differential tooth dominance map was gray in color and spatially scattered without the black and white mosaic structure, and the map was different from a typical ocular dominance map. It suggests that inputs from the left or right maxillary canine dominate the same cortical regions. The global signals for 10 g, 15 g, 26 g, 60 g, and 100 g were further compared between contralateral and ipsilateral mechanical stimuli, and no significant difference was found (Fig. 5F). Statistically, the mean tooth dominance indexes for various stimuli were all close to 0 (Fig. 5G, n = 5 animals, p > 0.05 for all
258
J. Tao et al. / Neuroscience 329 (2016) 254–263
Fig. 3. Determination of the focal plane of the camera and the direction of mechanical stimulation. (A) The blood vessel map of an imaged brain. The dotted outlined region is the ROI used for further analysis. (B) Global signal maps obtained at different focal planes for a 60-g mechanical stimulus on ipsilateral canines. The map patterns remained unchanged. Red arrows indicate blood vessels. (C) The time course curve of the global signal in a full data acquisition trial for different focal planes. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (D) Global signal maps evoked by stimulations in various directions. Abbreviations: AP, anterior-posterior; LM, lateral to medial; ML, medial to lateral; BU, bottom-up. (E) The time course curve of the global signal in a full data acquisition trial for various stimulation directions. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (F) The mean strength and standard error of the global signal at the 7th frame for various stimulation directions (n = 24 trials). (G) Differential maps between different stimulation directions. Spatial specificity for a certain direction could not be identified. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
points, t test), which suggests a significant bilateral dominance. Comparing the global signal between the sigmoidal gyrus and coronal gyrus Previous sparse sampling by single-unit recordings have been inconsistent regarding the detailed location of the regions representing teeth (at the coronal gyrus or sigmoidal gyrus). We investigated the specific projection area driven by the canine periodontium. We exposed a large cortical region including the coronal and sigmoidal gyri (Fig. 6A, a ROI with two parts), and examined the global signal map for a 60-g ipsilateral mechanical stimulation. Results demonstrated that both gyri were activated, and the strongest region was close to their border (Fig. 6B). The top 5%, 10% and 15% of the strongest responsive pixels in the ROI were marked on the blood vessel maps (Fig. 6C), and it appeared that more pixels tended to locate in the sigmoidal gyrus
(Fig. 6D, red column vs. blue column). We then examined the mean responsiveness of all pixels in these two regions. In the time course curves of global signals, the response in the sigmoidal gyrus was slightly greater than that in the coronal gyrus (Fig. 6E), and this tendency could be found in both contralateral-evoked (Fig. 6F, n = 8, p < 0.05, paired t test) and ipsilateral-evoked (Fig. 6G, n = 8, p < 0.05, paired t test) global signal maps. These results demonstrated that the sensory information from the canine periodontal membrane was projected to the sigmoidal gyrus and coronal gyrus (with a slight bias toward the sigmoidal gyrus).
DISCUSSION Detailed mapping information provided by intrinsic signal optical imaging Since its first application in early 1990s, blood oxygenation level-dependent (BOLD) fMRI has become
J. Tao et al. / Neuroscience 329 (2016) 254–263
259
Fig. 4. Influence of stimulus strength on the global signal. (A) The blood vessel map and global signal map of an imaged brain. The outlined region is the ROI (2 3 mm) used in further analysis. (B) Global signal maps for different strengths of mechanical stimuli. The tooth-responsive cortex became darker (blue in color map) as stimulus strength increased. Arrow: the time point of stimulus onset. (C) The time course curve of the global signal in a full data acquisition trial for different strengths of mechanical stimuli. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (D) The mean strength and standard error of the global signal for different strengths of mechanical stimuli are presented. (n = 5 animals; * indicates p < 0.05). (E) Differential maps between different stimulus strengths. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
one of the most powerful mapping techniques for localizing brain function and has revolutionized cognitive neuroscience. In dentistry, fMRI has been used to demonstrate neural changes or neuroplasticity of the sensorimotor cortex for cross-sectional studies in human subjects (Yan et al., 2008; Kimoto et al., 2011; HabreHallage et al., 2012). However, the low spatial (1 mm) and temporal (1 s) resolution of fMRI cannot provide detailed sub-columnar information of brain function, and different techniques are needed for mechanism studies. In contrast to BOLD fMRI, the intrinsic signal optical imaging technique (Grinvald et al., 1999) offers higher spatial (100 lm) and temporal (100 ms) resolution and has been widely used to study cortical function in animals (Chen et al., 2003, 2014; Yu et al., 2008), and humans (Schwartz, 2005). In addition to better resolution, since signal strength of intrinsic signal optical imaging is influenced by stimulus strength, the relationship between stimulus strength and response strength can be further investigated. Intrinsic signals are slow and lack precise temporal resolution compared with electrophysiological recordings. However, the spatial resolution of the camera used for measuring changes in intrinsic signals uniquely allows for simultaneous evaluation of mapping
and functional information of a large area of the cortex containing millions of neurons (Yu et al., 2008; Tong et al., 2011; An et al., 2012; Pan et al., 2012; Chen et al., 2014). Recent studies have proven that data obtained from intrinsic signal optical imaging closely matches data obtained from electrophysiology combining single-neuron recordings (Arieli and Grinvald, 2002; Thompson et al., 2003), local field potential recording, Doppler flowmetry (Sheth et al., 2004), and in vivo two photon calcium imaging (Yu et al., 2011). In this study, intrinsic signal optical imaging was used to enable sampling of a large area of the somatosensory cortex (our setup provides a 6.7 6.7 lm/pixel resolution of a maximal 10 10 mm area). This permitted data collection during a full session of mechanical stimulation, providing reliable population data from a large region of the cortex that contains millions of neurons as well as detailed mapping information regarding cortical function. Comparison with previous studies In this study, the strength of the von Frey filament was chosen to provide stimuli well above the mechanical detection threshold of the teeth but below pain thresholds. Forces ranging from 100 g to 10 g were applied to the canine according to previous studies
260
J. Tao et al. / Neuroscience 329 (2016) 254–263
Fig. 5. Bilateral sensory inputs to the tooth-responsive cortex. (A) The blood vessel map of an imaged brain. (B) Global signal maps for a 60-g mechanical stimulus on the contralateral and ipsilateral canines. (C) The time course curves of the global signal for a 60-g mechanical stimulus on the contralateral and ipsilateral canines. (D) The differential map for a 60-g mechanical stimulus (map of ipsilateral stimulus – map of contralateral stimulus). The outlined regions are the ROIs used in further analysis. (E) The distribution of tooth dominance index in ROIs. (F) The time course curves of the global signal for different strengths of mechanical stimuli on the contralateral and ipsilateral canines (n = 24 trials). (G) The mean tooth dominance index across five different mechanical stimuli (n = 5 animals, each symbol is the mean of one animal, lack of some points for certain stimuli).
(Taira, 1987; Watanabe et al., 1991). Noxious stimuli larger than 100 g were not used. These forces applied to canines were able to elicit clear cortical activations in the cat somatosensory cortex for the oral area. The cortical responses (global signals) increased with stimulus strength and peaked at 60 g for contralateral and ipsilateral mechanical stimulation. Our findings by intrinsic signal optical imaging are consistent with previous studies that used the electrophysiological technique by Taira (1987). In his study, when the stimulus amplitude increased, given a constant rate of stimulus application, the initial peak values of the neuronal spike discharge for periodontal mechanosensitive units increased and then plateaued at a maximum level. Because the mechanical stimulus of 60 g almost elicited the greatest global signal, 60 g may be the optimal force to stimulate canines in future experiments. In human subjects, greater
force (180 g) was applied to canines to elicit a clear and constant signal in fMRI (Habre-Hallage et al., 2010). In our data, the bilateral canine inputs to 1 hemisphere of the cortex differ from some other sensory systems. The contralateral receptive field in one hemisphere of the cerebral cortex was found in the visual cortices (Nikara et al., 1968) as well as the somatosensory cortices (Clemo and Stein, 1982), but in the cortex representing teeth, our data clearly show a bilateral canine-driven response. Hubel et al. (1978) found that inputs from different eyes (but optically corresponding to the contralateral receptive field) dominated different columns in the primary visual cortex. This pattern is also found in the whisker barrel cortex (Lee et al., 2009) and frequency-specific auditory columns (Aitkin et al., 1986), but fails in the caninerelated cortex in our study. The cortical regions driven by the left or right maxillary canine seemed to be identical,
J. Tao et al. / Neuroscience 329 (2016) 254–263
261
Fig. 6. A comparison of global signals from the sigmoidal gyrus and coronal gyrus. (A) The blood vessel map of an imaged brain. The dotted outlined region is the ROI used in further analysis for the sigmoidal gyrus (bottom) and the coronal gyrus (top). (B) Global signal maps for 60 g mechanical stimulus on ipsilateral canines. (C) The distribution of the top 5%, 10% and 15% responsive pixels in the ROI including the sigmoidal gyrus and coronal gyrus. It is notable that only the pixels in the outlined ROIs were analyzed and presented as color dots. (D) The individual percentage of the top 5%, 10% and 15% in the sigmoidal gyrus and coronal gyrus. (E) The time course curves of the global signal for the sigmoidal gyrus and coronal gyrus when a 60 g mechanical stimulus was applied to the ipsilateral canines (n = 24 trials). (F) The comparison of mean global signal strength between the sigmoidal gyrus and coronal gyrus ROI for 60-g or 100-g contralateral mechanical stimuli (n = 8; * indicates p < 0.05, paired t test). (G) The comparison of the mean global signal strength between the sigmoidal gyrus and coronal gyrus ROI for 60 g or 100 g ipsilateral mechanical stimuli (n = 8; * indicates p < 0.05, paired t test). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and there were no significant differences in the global signals between contralateral and ipsilateral canine stimulation. It seems that integration of bilateral teeth information occurs at the very beginning of the sensory cortex. It is well known that binocularity in the visual cortices is related to perception of 3-dimensional vision. Analogously, bilateral teeth representation in the cortex may help with stereognosis of masticated foods. Masticatory jaw movements are smoothly carried out in cooperative contractions and extensions of the bilateral muscles of mastication. The bilateral somatosensory information from the periodontal membrane may play a key role in regulating masticatory jaw movements. The bilateral inputs from oral structures to the somatosensory cortex have been also reported in previous studies using electrophysiological technique (Lund and Sessle, 1971; Toda and Taoka, 2004, 2006). To understand how the bilateral information is processed, further studies are needed at the single-cell level. Sensory information from the periodontal membrane has been reported to project to the cortical area around the coronal sulcus in the cat (Landgren and Olsson, 1980; Taira, 1987; Watanabe et al., 1991). However, the specific location remained controversial. Representation of the periodontal membrane was found in the
anterior coronal gyrus by multi-unit recording and singleunit recording by Taira (1987). In contrast, Watanabe et al. recorded large field potentials not only in the coronal gyrus but also in the sigmoidal gyrus when mechanical stimulation was applied to the contralateral upper canine (Watanabe et al., 1991). The results of our study support the finding by Watanabe et al. It is interesting that, using simultaneous intrinsic signal optical imaging, a previous study revealed different strengths of cortical population responses across multiple visual cortices following an identical visual stimulus (Pan et al., 2012). Similarly, in our study, a slightly greater response strength in intrinsic signal optical imaging was also recorded in the sigmoidal gyrus than in the coronal gyrus while punctate mechanical stimulation was applied to the upper canine. This simultaneous population-based study in a large cortical area provides the precise location of the canine-related cortex. In summary, we used intrinsic signal optical imaging to examine response characteristics of the cat primary somatosensory cortex following forces applied to canines. The global signal map was obtained, and we found that global signal strength gradually increased with stimulus strength and the time course curve of the global signal for mechanical stimulation of canines was similar to that for visual stimuli. Therefore, intrinsic
262
J. Tao et al. / Neuroscience 329 (2016) 254–263
signal optical imaging may be used to investigate neural changes and neuroplasticity of the cortex representing teeth in longitudinal studies. In our results, there was no significant difference in response strength between contralateral and ipsilateral mechanical stimulation. A slightly greater response was recorded in the sigmoidal gyrus compared with the coronal gyrus, and the cat somatosensory cortex is not organized into columns to represent sensory inputs from each tooth or direction of stimulation. Acknowledgments—This study was supported by grants from 863 Program of the Ministry of Science and Technology of China (No. 2015AA020508), the National Natural Science Foundation of China (91520203, 31571076, 31171054), and the Shanghai Key Discipline Foundation (B111) to H. Yu. The authors declare no potential conflicts of interest with respect to the authorship and/ or publication of this article.
REFERENCES Abarca M, Steenberghe D, Malevez C, Jacobs R (2006) The neurophysiology of osseointegrated oral implants. A clinically underestimated aspect. J Oral Rehabil 33:161–169. Aitkin LM, Merzenich MM, Irvine DRF, Clarey JC, Nelson JE (1986) Frequency representation in auditory cortex of the common marmoset (Callithrix jacchus jacchus). J Comp Neurol 252:175–185. An X, Gong H, Qian L, Wang X, Pan Y, Zhang X, Yang Y, Wang W (2012) Distinct functional organizations for processing different motion signals in V1, V2, and V4 of macaque. J Neurosci 32:13363–13379. Arieli A, Grinvald A (2002) Optical imaging combined with targeted electrical recordings, microstimulation, or tracer injections. J Neurosci Methods 116:15–28. Avivi-Arber L, Lee JC, Sessle BJ (2010) Effects of incisor extraction on jaw and tongue motor representations within face sensorimotor cortex of adult rats. J Comp Neurol 518:1030–1045. Bauer AQ, Kraft AW, Wright PW, Snyder AZ, Lee JM, Culver JP (2014) Optical imaging of disrupted functional connectivity following ischemic stroke in mice. NeuroImage 99:388–401. Calford MB (2005) Degree of adaptability of the somatosensory cortex to change: prospects for integration of bone-mounted dental prostheses. Clin Exp Pharmacol Physiol 32:115–118. Chen LM, Friedman RM, Roe AW (2003) Optical imaging of a tactile illusion in area 3b of the primary somatosensory cortex. Science 302:881–885. Chen Y, Li H, Jin Z, Shou T, Yu H (2014) Feedback of the amygdala globally modulates visual response of primary visual cortex in the cat. NeuroImage 84:775–785. Clemo HR, Stein BE (1982) Somatosensory cortex: a ’new’ somatotopic representation. Brain Res 235:162–168. Grinvald A, Shoham D, Shmuel A, Glaser D, Vanzetta I, Shtoyerman E, Slovin H, Wijnbergen C, Hildesheim R, Sterkin A (1999) Modern techniques in neuroscience research. In: Windhorst U, Johansson H, editors. In-vivo optical imaging of cortical architecture and dynamics, chapter 34. Springer: SpringerVerlag, Berlin. p. 893–969. Habre-Hallage P, Hermoye L, Gradkowski W, Jacobs R, Reychler H, Grandin CB (2010) A manually controlled new device for punctuate mechanical stimulation of teeth during functional magnetic resonance imaging studies. J Clin Periodontol 37:863–872. Habre-Hallage P, Dricot L, Jacobs R, van Steenberghe D, Reychler H, Grandin CB (2012) Brain plasticity and cortical correlates of osseoperception revealed by punctate mechanical stimulation of
osseointegrated oral implants during fMRI. Eur J Oral Implantol 5:175–190. Henry EC, Marasco PD, CataniaK C (2005) Plasticity of the cortical dentition representation after tooth extraction in naked mole-rats. J Comp Neurol 485:64–74. Horinuki E, Shinoda M, Shimizu N, Koshikawa N, Kobayashi M (2015) Orthodontic force facilitates cortical responses to periodontal stimulation. J Dent Res 94:1158–1166. Hubel DH, Wiesel TN, Stryker MP (1978) Anatomical demonstration of orientation columns in macaque monkey. J Comp Neurol 177:361–379. Jantsch HH, Kemppainen P, Ringler R, Handwerker HO, Forster C (2005) Cortical representation of experimental tooth pain in humans. Pain 118:390–399. Kim SA, Jun SB (2013) In-vivo optical measurement of neural activity in the brain. Exp Neurobiol 22:158–166. Kimoto K, Ono Y, Tachibana A, Hirano Y, Otsuka T, Ohno A, et al. (2011) Chewing-induced regional brain activity in edentulous patients who received mandibular implant-supported over dentures: a preliminary report. J Prosthodont Res 55: 89–97. Landgren S, Olsson KA (1980) Low threshold afferent projections from the oral cavity and the face to the cerebral cortex of the cat. Exp Brain Res 39:133–147. Lee SB, Beak SK, Park SH, Waterhouse BD, Lee HS (2009) Collateral projection from the locus coeruleus to whisker-related sensory and motor brain regions of the rat. J Comp Neurol 514:387–402. Lund JP, Sessle BJ (1971) Oral-facial and jaw muscle afferent projections to neurons in cat frontal cortex. Exp Neurol 45:314–331. Nakamura H, Shirakawa T, Koshikawa N, Kobayashi M (2016) Distinct excitation to pulpal stimuli between somatosensory and insular cortices. J Dent Res 95:180–187. Nikara T, Bishop PO, Pettigrew JD (1968) Analysis of retinal correspondence by studying receptive fields of binocular single units in cat striate cortex. Exp Brain Res 6:353–372. Pan Y, Chen M, Yin J, An X, Zhang X, Lu Y, Gong H, Li W, Wang W (2012) Equivalent representation of real and illusory contours in macaque V4. J Neurosci 32:6760–6770. Schwartz TH (2005) The application of optical recording of intrinsic signals to simultaneously acquire functional, pathological and localizing information and its potential role in neurosurgery. Stereotact Funct Neurosurg 83:36–44. Sheth SA, Nemoto M, Guiou M, Walker M, Pouratian N, Toga AW (2004) Linear and nonlinear relationships between neuronal activity, oxygen metabolism, and hemodynamic responses. Neuron 42:347–355. Taira K (1987a) The representation of the oral structures in the first somatosensory cortex of the cat. Brain Res 409:41–51. Taira K (1987b) Characteristics of periodontal mechanosensitive neurons in the first somatosensory cortex of the cat. Brain Res 409:52–61. Thompson JK, Peterson MR, Freeman RD (2003) Single-neuron activity and tissue oxygenation in the cerebral cortex. Science 299:1070–1072. Toda T, Taoka M (2004) Converging patterns of inputs from oral structures in the postcentral somatosensory cortex of conscious macaque monkeys. Exp Brain Res 158:43–49. Toda T, Taoka M (2006) Postcentral neurons with covert receptive fields in conscious macaque monkeys: their selective responsiveness to simultaneous two-point stimuli applied to discrete oral portions. Exp Brain Res 168:303–306. Tong L, Zhu B, Li Z, Shou T, Yu H (2011) Feedback from area 21a influences orientation but not direction maps in the primary visual cortex of the cat. Neurosci Lett 504:141–145. Trulsson M, van der Bilt A, Carlsson GE, Gotfredsen K, Larsson P, Mu¨ller F, Sessle BJ, Svensson P (2012) From brain to bridge: masticatory function and dental implants. J Oral Rehabil 39:858–877.
J. Tao et al. / Neuroscience 329 (2016) 254–263 Watanabe M, Tabata T, Karita K (1991) Facilitatory effect of jaw opening on somatosensory (S1) cortical neurons sensitive to tooth pressure in the cat. Archs Oral Biol 36:899–903. Yan C, Ye L, Zhen J, Ke L, Gang L (2008) Neuroplasticity of edentulous patients with implant-supported full dentures. Eur J Oral Sci 116:387–393.
263
Yu H, Chen X, Sun C, Shou T (2008) Global evaluation of contributions of GABAA, AMPA and NMDA receptors to orientation maps in cat’s visual cortex. NeuroImage 40:776–787. Yu H, Majewska AK, Sur M (2011) Rapid experience-dependent plasticity of synapse function and structure in ferret visual cortex in vivo. Proc Natl Acad Sci 108:21235–21240.
(Accepted 30 April 2016) (Available online 6 May 2016)
POPULATION RESPONSE CHARACTERISTICS OF INTRINSIC SIGNALS IN THE CAT SOMATOSENSORY CORTEX FOLLOWING CANINE MECHANICAL STIMULATION JIANXIANG TAO, ay JIAN WANG, by ZHONG LI, b JIANJUN MENG b AND HONGBO YU b*
dentures) such as tissue reactions, biomechanics, and occlusal features have been studied extensively. However, neural changes that occur when the stomatognathic system is altered (e.g., tooth loss, rehabilitative processes, and tooth pain) have been seldom addressed. Recently, attention has been focussed on these neural changes (Henry et al., 2005; Jantsch et al., 2005; Avivi-Arber et al., 2010) and the cortical adaptive processes underlying these rehabilitative processes (Yan et al., 2008; Habre-Hallage et al., 2012). Changes in areas of the central nervous system (CNS) elicited by alterations to the intraoral tissues are thought to be crucial for a person’s ability to learn and for the sensorimotor system to adapt to the new intraoral environment (Calford, 2005; Abarca et al., 2006; Trulsson et al., 2012). Thus, evaluating neural responses in the cortex in response to tooth stimulation in vivo would be extremely useful. For half a century, most knowledge of cortical function was obtained from single-unit recordings, which provided detailed firing information of individual neurons. Due to extremely intricate cortical circuitry, it is difficult to evaluate the relative contribution of each neuron to oral function and related mapping information in a large cortical area. Therefore, brain imaging techniques, such as functional magnetic resonance imaging (fMRI), have been used to demonstrate neural changes or neurop lasticity of the sensorimotor cortex for cross-sectional studies in human subjects and animals (Yan et al., 2008; Kimoto et al., 2011; Habre-Hallage et al., 2012). However, the spatial and temporal resolution of fMRI is low, and it cannot provide detailed and insightful information. In the past two decades, intrinsic signal optical imaging, which is based on changes in blood oxygen consumption, has been widely used to establish detailed functional maps in the sensory cortex (Grinvald et al., 1999). Intrinsic signal optical imaging is a valid, reliable, and a unique technique to investigate neuronal responses of millions of cells in the visual cortex (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014). Moreover, since it requires no chemicals or virus injections for labeling, this technique has no toxic effect and can record functional signals in the cortex from the same animal for an extended period. Thus, it is suitable for detecting cortical neuroplasticity for longitudinal studies (Kim and Jun, 2013; Bauer et al., 2014). Optical imaging with a voltage-sensitive dye was used to identify cortical responses evoked by electrical stimulation of the periodontal ligaments and dental pulps
a
Shanghai Engineering Research Center of Tooth Restoration and Regeneration, Department of Prosthodontics, School of Stomatology, Tongji University, Shanghai 200072, China b
Vision Research Laboratory, School of Life Sciences, The State Key Laboratory of Medical Neurobiology, Collaborative Innovation Center for Brain Science, Fudan University, Shanghai 200433, China
Abstract—Intrinsic signal optical imaging has been widely used to measure functional maps in various sensory cortices due to better spatial resolution and sensitivity for detecting cortical neuroplasticity. However, application of this technique in dentistry has not been reported. In this study, intrinsic signal optical imaging was used to investigate mechanically driven responses in the cat somatosensory cortex, when punctate mechanical stimuli were applied to maxillary canines. The global signal and its spatial organization pattern were obtained. Global signal strength gradually increased with stimulus strength. There was no significant difference in response strength between contralateral and ipsilateral mechanical stimulation. A slightly greater response was recorded in the sigmoidal gyrus than in the coronal gyrus. The cat somatosensory cortex activated by sensory inputs from mechanical stimulation of canines lacks both topographical and functional organization. It is not organized into columns that represent sensory input from each tooth or direction of stimulation. These results demonstrate that intrinsic signal optical imaging is a valid tool for investigating neural responses and neuroplasticity in the somatosensory cortex that represents teeth. Ó 2016 IBRO. Published by Elsevier Ltd. All rights reserved.
Key words: neuroimaging, cuspid, physical stimulation, brain, evoked potentials, optical devices.
INTRODUCTION Many characteristics of prosthodontics rehabilitative procedures (e.g., implants, crowns, bridges, and *Corresponding author. Address: Vision Research Laboratory, School of Life Sciences, The State Key Laboratory of Medical Neurobiology, Collaborative Innovation Center for Brain Science, Fudan University, Shanghai 200433, China. Fax: +86-65650149. E-mail address: [email protected] (H. Yu). y Authors contributing equally to this work. Abbreviations: BOLD, blood oxygenation level-dependent; fMRI, functional magnetic resonance imaging; ROI, regions of interest. http://dx.doi.org/10.1016/j.neuroscience.2016.04.052 0306-4522/Ó 2016 IBRO. Published by Elsevier Ltd. All rights reserved. 254
J. Tao et al. / Neuroscience 329 (2016) 254–263
in rats (Horinuki et al., 2015; Nakamura et al., 2016). However, examination of both topographical and functional organizations representing different individual teeth has been seldom reported. Unlike previous electrophysiological single-cell recordings, intrinsic optical imaging is a good approach for investigating neuronal responses of the cortex at the population level (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014). The purpose of this study was to use intrinsic signal optical imaging to examine population response characteristics of the cat primary somatosensory cortex following mechanical stimulation of canines and to evaluate the application of intrinsic signal optical imaging in dentistry.
EXPERIMENTAL PROCEDURES Animal preparation Seven healthy adult cats of either gender weighing between 1.5 and 2.5 kg were used in the study (Shanghai Yingen Farm). The experimental protocol used in this study was approved by the Animal Experimentation Committee at Fudan University. Animal treatments were performed in strict accordance with the Guide for the Care and Use of Laboratory Animals described by Fudan University and the U.S. National Institutes of Health. Animals were prepared as described in our previous studies (Tong et al., 2011; Chen et al., 2014). Craniotomy was performed over the coronal sulcus (Fig. 1B, C). A stainless steel imaging chamber was positioned and fixed with dental cement on the skull, covering the oral area portion of the somatosensory cortex. After careful removal of the dura, the chamber was filled with warm (37 °C) silicone oil and sealed with a transparent glass window. Punctate mechanical stimulation Square planes (2 2 mm) were fabricated in the cusps of maxillary canine teeth using dental filling resin. Punctate
255
mechanical stimuli were applied on either the left or right canine tooth in the anterior-posterior direction using von Frey filament (North Coast Medical, Inc., Gilroy, CA, USA) to impact the square resin plane. The von Frey filament was attached to a rail electrode (JF-1683, Leqing, China) (Fig. 1A). The stimulus was presented for 4 s (a constant frequency of 1 Hz, square pulse with 0.5 s on and 0.5 s off) and then repeated with 18 s interstimulus intervals. The strength of mechanical stimulation increased gradually from10 g (filament N.14), 15 g (N.15), 26 g (N.16), 60 g (N.17), to 100 g (N.18). Optical data acquisition A custom-made optical imaging system was employed. As in our previous studies (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014), a slow scan CCD camera (1600 1200 pixels, 6.7 6.7 lm/pixel; PCO, Inc., PCO, AG, Germany) was used to record optical images of intrinsic signals from the exposed portion of the somatosensory cortex (Fig. 1A). Data acquisition started 3 s before the mechanical stimulus and continued 3 s after the stimulus. Therefore, 10 frames at a frame rate of 1 Hz were recorded. To allow relaxation of activitydependent microvascular changes, the period of stimulus presentation was followed by an 18-s rest interval. Each stimulus was presented 24 times, and the evoked signals were averaged. The details of optical data acquisition have been described in our previous studies (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014) Data analysis The data were analyzed as described in our previous studies (Yu et al., 2008; Tong et al., 2011; Chen et al., 2014). The resultant frames of the ratio map (dR/R = R/ Ra1) were defined as the global signal maps. In the image of the global signal map, the cortex with more reflectance reduction had the stronger signal and a more negative value. To calculate the strength of the global sig-
Fig. 1. The experimental setup for intrinsic signal optical imaging of the tooth-related area of the somatosensory cortex. (A) Diagram of the experimental setup. Intrinsic signal optical imaging was performed in the exposed primary somatosensory cortex of an anesthetized cat when punctate mechanical stimuli were applied to the left or right canine tooth. (B) Photograph of the cortical surface of the right hemisphere. Craniotomy was performed over the outlined region. Abbreviations: Cor.S., coronal sulcus; Cru.S., cruciate sulcus; Sups.S., suprasylvian sulcus; Lat.S., lateral sulcus. (C) The blood vessel map of a brain imaged via a transparent window.
256
J. Tao et al. / Neuroscience 329 (2016) 254–263
nal, the exposed somatosensory cortex region was outlined as regions of interest (ROI), and dR/R of all pixels in the ROIs were averaged as the signal strength for each stimulus. The mean and standard error were calculated across different pairs, and Student’s t-test was performed.
RESULTS Determining the cerebral cortex driven by canine inputs
be similar (Fig. 3C). Therefore, in this study, the camera was focused on the plane at 500 lm below the pial surface. We then examine intrinsic signals evoked by stimulations in various directions. A 60-g mechanical stimulus was applied to canines from four directions including axial (bottom-up), and angled directions (lateral to medial, medial to lateral, and straight anterior–posterior). The exposed cortex could be activated from all four directions (Fig. 3D, E). The spatial patterns of global signal maps for different directions seemed to be similar, and the differential map did not show very obvious spatial specificity for a certain direction (Fig. 3G). However, the global signal for anterior-posterior stimulation exhibited the highest strength (Fig. 3F). Moreover, the force of anteriorposterior stimulation was more accurate and easier to apply than the other directions. For this reason, anteriorposterior stimulation was determined.
The cerebral cortex driven by the cat canine periodontium was reported to be in the vicinity of the coronal sulcus (Landgren and Olsson, 1980; Taira, 1987; Watanabe et al., 1991) as shown in Fig. 1B. We exposed this area (Fig. 1C), made a transparent window over it, and examined the stimulus preference of the recording area by intrinsic signal optical imaging (Fig. 1A). The global signal map elicited by 60-g mechanical stimuli on the contralateral canine was compared with that elicited by 60-g mechanical stimuli on the contralateral lip (Fig. 2). Due to increased oxygen consumption, the exposed somatosensory cortex became darker after the onset of canine mechanical stimulation, while the unexposed skull region remained unchanged (Fig. 2B, right). In contrast, the exposed somatosensory cortex showed almost no changes after onset of identical mechanical stimuli to the lip (Fig. 2B, left), which suggests that 60-g von Frey filament stimulation only of the canine could evoke responses in the related cortical region. It is possible that this stimulation (60 g) was not optimal for the lip or that the region representing the lip was located at a different site. We selected a ROI (Fig. 2A, red box), and the averaged values of all selected pixels demonstrated a significant decrease after onset of the canine stimuli (Fig. 2C, the 4–7th frames, 1 frame/s, n = 24 trials); the peak appeared at the 6th frame. Clearly, this global signal decreased after canine stimulation, but remained flat after lip stimulation (Fig. 2C, blue vs. red curves). The global signal map was not uniform, and the contour lines showed a gradual decrease of the responses from the center (Fig. 2D, E). The global signals for 6 ROIs (Figs. 2D, 6 small boxes) were examined further, and the center area exhibited the greatest global signal with an increasingly weaker response in the surrounding area (Fig. 2F), which is consistent with a previous study that used field potential recordings (Watanabe et al., 1991). Therefore, the exposed somatosensory cortex close to the coronal sulcus was confirmed to be driven specifically by the canine periodontium.
We then investigated the effect of stimulus strength on the global signal. In a typical case, a 10-g mechanical stimulation did not evoke a response above threshold, and the global signal map (Fig. 4B, top) showed almost no changes after onset of mechanical stimulation. The cortex representing teeth became darker after the onset of the 15-g stimulus. The map became darker as stimulus strength increased (Fig. 4B). The global signal of a ROI (Fig. 4A, 2 3 mm) demonstrated a significant decrease after onset of the stimulus (Fig. 4C. n = 24 trials for each mechanical stimulus). It is important to note that signal strength gradually increased with stimulus strength and saturated at the 60 g of mechanical stimulation (Fig. 4D, n = 5 animals). It has been shown that two groups of mechanoreceptors (pain and mechanical sensation) lie in the periodontal ligament, and very strong forces might evoke noxious response. In our experiment, stimulations less than 100 g were applied to avoid this possible effect. It is notable that weak (15 or 26 g) and stronger (60 or 100 g) stimulation evoked similar spatial activation patterns (Fig. 4B), and no obvious spatial segregation could be identified in the differential maps (Fig. 4E, left: 60–15 g, and right: 60–26 g maps), which suggests a non-noxious response. In fact, axial stimulation (bottomup) was also used in an attempt to activate specifically pure mechanical receptors in the apex region of the tooth root, and a similar activation pattern was obtained (Fig. 3D, left panel).
Determining focal plane of the camera and direction of mechanical stimulation
Equivalent bilateral sensory inputs to the toothrelated cortex
The signals were obtained from 500 lm below the pial surface, as suggested in most of the intrinsic signal optical imaging experiments (Grinvald et al., 1999). To examine the possible effect of focal plane, we changed the focal plane of the camera from 300 to 500 lm below the pial surface. The map patterns of global signals remained unchanged at 300, 400, and 500 lm (Fig. 3B). The time-course curves of the global signal seemed to
Peripheral sensory inputs tend to project to the contralateral hemisphere of the cortex. For example, the visual field (divided by the central azimuth line) is represented in the primary visual cortex of the contralateral hemisphere, and left and right eye (the retina responsive to the contralateral visual field) dominated cortical regions are interleaved in the same area of the cortex to form the well-known ocular
Measuring the linear relationship between stimulus strength and global intrinsic signal
J. Tao et al. / Neuroscience 329 (2016) 254–263
257
Fig. 2. Determination of the cerebral cortex driven by canine inputs. (A) Blood vessel map of an imaged brain located near the lateral end point of the coronal sulcus. The outlined region is the ROI used for analysis of global signals. L, lateral; A, anterior. (B) Global signal maps elicited by 60-g mechanical stimuli on the contralateral canine or the neighboring lips. Data acquisition started 3 s before the mechanical stimulus; 10 frames (1 frame/s) were obtained and consecutively presented. Arrow: the stimulus onset time point. (C) The time course curve of the global signal in a full data acquisition trial for canine and lip mechanical stimuli. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (D) The blood vessel map and 7th frame of the global signal map elicited by canine stimulation, with 6 ROIs (6 small boxes) on the exposed somatosensory cortex. (E) The contour lines of the global signal map at the 7th frame with 6 ROIs. (F) The time course curve of the global signal for 6 ROIs (n = 24 trials). Center area exhibited the greatest global signal. Standard errors are shown in C and F. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
dominance map (Hubel et al., 1978). It is of interest to examine whether the tooth-related cortex has a similar contralateral dominance and whether inputs from different teeth are interleaved as well. In this study, ipsilateral and contralateral canines were stimulated using a 60 g von Frey filament, and the elicited global signal map and time course curves of the global signal were examined (Fig. 5B, C). It appeared that there was no significant difference between contralateral and ipsilateral canine-dominated maps. We further calculated the tooth dominance index for each pixel [(Contra. Ipsi.)/(Contra. + Ipsi.)], and the distribution of all pixels was a Gaussian curve that peaked at approximately 0, which suggests no bias for contralateral dominance (Fig. 5E). To examine the
possible spatial segregation for ipsilateral and contralateral canine inputs, the 7th frames of evoked global maps for each canine were subtracted and illustrated in Fig. 5D. However, the differential tooth dominance map was gray in color and spatially scattered without the black and white mosaic structure, and the map was different from a typical ocular dominance map. It suggests that inputs from the left or right maxillary canine dominate the same cortical regions. The global signals for 10 g, 15 g, 26 g, 60 g, and 100 g were further compared between contralateral and ipsilateral mechanical stimuli, and no significant difference was found (Fig. 5F). Statistically, the mean tooth dominance indexes for various stimuli were all close to 0 (Fig. 5G, n = 5 animals, p > 0.05 for all
258
J. Tao et al. / Neuroscience 329 (2016) 254–263
Fig. 3. Determination of the focal plane of the camera and the direction of mechanical stimulation. (A) The blood vessel map of an imaged brain. The dotted outlined region is the ROI used for further analysis. (B) Global signal maps obtained at different focal planes for a 60-g mechanical stimulus on ipsilateral canines. The map patterns remained unchanged. Red arrows indicate blood vessels. (C) The time course curve of the global signal in a full data acquisition trial for different focal planes. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (D) Global signal maps evoked by stimulations in various directions. Abbreviations: AP, anterior-posterior; LM, lateral to medial; ML, medial to lateral; BU, bottom-up. (E) The time course curve of the global signal in a full data acquisition trial for various stimulation directions. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (F) The mean strength and standard error of the global signal at the 7th frame for various stimulation directions (n = 24 trials). (G) Differential maps between different stimulation directions. Spatial specificity for a certain direction could not be identified. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
points, t test), which suggests a significant bilateral dominance. Comparing the global signal between the sigmoidal gyrus and coronal gyrus Previous sparse sampling by single-unit recordings have been inconsistent regarding the detailed location of the regions representing teeth (at the coronal gyrus or sigmoidal gyrus). We investigated the specific projection area driven by the canine periodontium. We exposed a large cortical region including the coronal and sigmoidal gyri (Fig. 6A, a ROI with two parts), and examined the global signal map for a 60-g ipsilateral mechanical stimulation. Results demonstrated that both gyri were activated, and the strongest region was close to their border (Fig. 6B). The top 5%, 10% and 15% of the strongest responsive pixels in the ROI were marked on the blood vessel maps (Fig. 6C), and it appeared that more pixels tended to locate in the sigmoidal gyrus
(Fig. 6D, red column vs. blue column). We then examined the mean responsiveness of all pixels in these two regions. In the time course curves of global signals, the response in the sigmoidal gyrus was slightly greater than that in the coronal gyrus (Fig. 6E), and this tendency could be found in both contralateral-evoked (Fig. 6F, n = 8, p < 0.05, paired t test) and ipsilateral-evoked (Fig. 6G, n = 8, p < 0.05, paired t test) global signal maps. These results demonstrated that the sensory information from the canine periodontal membrane was projected to the sigmoidal gyrus and coronal gyrus (with a slight bias toward the sigmoidal gyrus).
DISCUSSION Detailed mapping information provided by intrinsic signal optical imaging Since its first application in early 1990s, blood oxygenation level-dependent (BOLD) fMRI has become
J. Tao et al. / Neuroscience 329 (2016) 254–263
259
Fig. 4. Influence of stimulus strength on the global signal. (A) The blood vessel map and global signal map of an imaged brain. The outlined region is the ROI (2 3 mm) used in further analysis. (B) Global signal maps for different strengths of mechanical stimuli. The tooth-responsive cortex became darker (blue in color map) as stimulus strength increased. Arrow: the time point of stimulus onset. (C) The time course curve of the global signal in a full data acquisition trial for different strengths of mechanical stimuli. The mean dR/R of the ROI was calculated for each frame (n = 24 trials). (D) The mean strength and standard error of the global signal for different strengths of mechanical stimuli are presented. (n = 5 animals; * indicates p < 0.05). (E) Differential maps between different stimulus strengths. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
one of the most powerful mapping techniques for localizing brain function and has revolutionized cognitive neuroscience. In dentistry, fMRI has been used to demonstrate neural changes or neuroplasticity of the sensorimotor cortex for cross-sectional studies in human subjects (Yan et al., 2008; Kimoto et al., 2011; HabreHallage et al., 2012). However, the low spatial (1 mm) and temporal (1 s) resolution of fMRI cannot provide detailed sub-columnar information of brain function, and different techniques are needed for mechanism studies. In contrast to BOLD fMRI, the intrinsic signal optical imaging technique (Grinvald et al., 1999) offers higher spatial (100 lm) and temporal (100 ms) resolution and has been widely used to study cortical function in animals (Chen et al., 2003, 2014; Yu et al., 2008), and humans (Schwartz, 2005). In addition to better resolution, since signal strength of intrinsic signal optical imaging is influenced by stimulus strength, the relationship between stimulus strength and response strength can be further investigated. Intrinsic signals are slow and lack precise temporal resolution compared with electrophysiological recordings. However, the spatial resolution of the camera used for measuring changes in intrinsic signals uniquely allows for simultaneous evaluation of mapping
and functional information of a large area of the cortex containing millions of neurons (Yu et al., 2008; Tong et al., 2011; An et al., 2012; Pan et al., 2012; Chen et al., 2014). Recent studies have proven that data obtained from intrinsic signal optical imaging closely matches data obtained from electrophysiology combining single-neuron recordings (Arieli and Grinvald, 2002; Thompson et al., 2003), local field potential recording, Doppler flowmetry (Sheth et al., 2004), and in vivo two photon calcium imaging (Yu et al., 2011). In this study, intrinsic signal optical imaging was used to enable sampling of a large area of the somatosensory cortex (our setup provides a 6.7 6.7 lm/pixel resolution of a maximal 10 10 mm area). This permitted data collection during a full session of mechanical stimulation, providing reliable population data from a large region of the cortex that contains millions of neurons as well as detailed mapping information regarding cortical function. Comparison with previous studies In this study, the strength of the von Frey filament was chosen to provide stimuli well above the mechanical detection threshold of the teeth but below pain thresholds. Forces ranging from 100 g to 10 g were applied to the canine according to previous studies
260
J. Tao et al. / Neuroscience 329 (2016) 254–263
Fig. 5. Bilateral sensory inputs to the tooth-responsive cortex. (A) The blood vessel map of an imaged brain. (B) Global signal maps for a 60-g mechanical stimulus on the contralateral and ipsilateral canines. (C) The time course curves of the global signal for a 60-g mechanical stimulus on the contralateral and ipsilateral canines. (D) The differential map for a 60-g mechanical stimulus (map of ipsilateral stimulus – map of contralateral stimulus). The outlined regions are the ROIs used in further analysis. (E) The distribution of tooth dominance index in ROIs. (F) The time course curves of the global signal for different strengths of mechanical stimuli on the contralateral and ipsilateral canines (n = 24 trials). (G) The mean tooth dominance index across five different mechanical stimuli (n = 5 animals, each symbol is the mean of one animal, lack of some points for certain stimuli).
(Taira, 1987; Watanabe et al., 1991). Noxious stimuli larger than 100 g were not used. These forces applied to canines were able to elicit clear cortical activations in the cat somatosensory cortex for the oral area. The cortical responses (global signals) increased with stimulus strength and peaked at 60 g for contralateral and ipsilateral mechanical stimulation. Our findings by intrinsic signal optical imaging are consistent with previous studies that used the electrophysiological technique by Taira (1987). In his study, when the stimulus amplitude increased, given a constant rate of stimulus application, the initial peak values of the neuronal spike discharge for periodontal mechanosensitive units increased and then plateaued at a maximum level. Because the mechanical stimulus of 60 g almost elicited the greatest global signal, 60 g may be the optimal force to stimulate canines in future experiments. In human subjects, greater
force (180 g) was applied to canines to elicit a clear and constant signal in fMRI (Habre-Hallage et al., 2010). In our data, the bilateral canine inputs to 1 hemisphere of the cortex differ from some other sensory systems. The contralateral receptive field in one hemisphere of the cerebral cortex was found in the visual cortices (Nikara et al., 1968) as well as the somatosensory cortices (Clemo and Stein, 1982), but in the cortex representing teeth, our data clearly show a bilateral canine-driven response. Hubel et al. (1978) found that inputs from different eyes (but optically corresponding to the contralateral receptive field) dominated different columns in the primary visual cortex. This pattern is also found in the whisker barrel cortex (Lee et al., 2009) and frequency-specific auditory columns (Aitkin et al., 1986), but fails in the caninerelated cortex in our study. The cortical regions driven by the left or right maxillary canine seemed to be identical,
J. Tao et al. / Neuroscience 329 (2016) 254–263
261
Fig. 6. A comparison of global signals from the sigmoidal gyrus and coronal gyrus. (A) The blood vessel map of an imaged brain. The dotted outlined region is the ROI used in further analysis for the sigmoidal gyrus (bottom) and the coronal gyrus (top). (B) Global signal maps for 60 g mechanical stimulus on ipsilateral canines. (C) The distribution of the top 5%, 10% and 15% responsive pixels in the ROI including the sigmoidal gyrus and coronal gyrus. It is notable that only the pixels in the outlined ROIs were analyzed and presented as color dots. (D) The individual percentage of the top 5%, 10% and 15% in the sigmoidal gyrus and coronal gyrus. (E) The time course curves of the global signal for the sigmoidal gyrus and coronal gyrus when a 60 g mechanical stimulus was applied to the ipsilateral canines (n = 24 trials). (F) The comparison of mean global signal strength between the sigmoidal gyrus and coronal gyrus ROI for 60-g or 100-g contralateral mechanical stimuli (n = 8; * indicates p < 0.05, paired t test). (G) The comparison of the mean global signal strength between the sigmoidal gyrus and coronal gyrus ROI for 60 g or 100 g ipsilateral mechanical stimuli (n = 8; * indicates p < 0.05, paired t test). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and there were no significant differences in the global signals between contralateral and ipsilateral canine stimulation. It seems that integration of bilateral teeth information occurs at the very beginning of the sensory cortex. It is well known that binocularity in the visual cortices is related to perception of 3-dimensional vision. Analogously, bilateral teeth representation in the cortex may help with stereognosis of masticated foods. Masticatory jaw movements are smoothly carried out in cooperative contractions and extensions of the bilateral muscles of mastication. The bilateral somatosensory information from the periodontal membrane may play a key role in regulating masticatory jaw movements. The bilateral inputs from oral structures to the somatosensory cortex have been also reported in previous studies using electrophysiological technique (Lund and Sessle, 1971; Toda and Taoka, 2004, 2006). To understand how the bilateral information is processed, further studies are needed at the single-cell level. Sensory information from the periodontal membrane has been reported to project to the cortical area around the coronal sulcus in the cat (Landgren and Olsson, 1980; Taira, 1987; Watanabe et al., 1991). However, the specific location remained controversial. Representation of the periodontal membrane was found in the
anterior coronal gyrus by multi-unit recording and singleunit recording by Taira (1987). In contrast, Watanabe et al. recorded large field potentials not only in the coronal gyrus but also in the sigmoidal gyrus when mechanical stimulation was applied to the contralateral upper canine (Watanabe et al., 1991). The results of our study support the finding by Watanabe et al. It is interesting that, using simultaneous intrinsic signal optical imaging, a previous study revealed different strengths of cortical population responses across multiple visual cortices following an identical visual stimulus (Pan et al., 2012). Similarly, in our study, a slightly greater response strength in intrinsic signal optical imaging was also recorded in the sigmoidal gyrus than in the coronal gyrus while punctate mechanical stimulation was applied to the upper canine. This simultaneous population-based study in a large cortical area provides the precise location of the canine-related cortex. In summary, we used intrinsic signal optical imaging to examine response characteristics of the cat primary somatosensory cortex following forces applied to canines. The global signal map was obtained, and we found that global signal strength gradually increased with stimulus strength and the time course curve of the global signal for mechanical stimulation of canines was similar to that for visual stimuli. Therefore, intrinsic
262
J. Tao et al. / Neuroscience 329 (2016) 254–263
signal optical imaging may be used to investigate neural changes and neuroplasticity of the cortex representing teeth in longitudinal studies. In our results, there was no significant difference in response strength between contralateral and ipsilateral mechanical stimulation. A slightly greater response was recorded in the sigmoidal gyrus compared with the coronal gyrus, and the cat somatosensory cortex is not organized into columns to represent sensory inputs from each tooth or direction of stimulation. Acknowledgments—This study was supported by grants from 863 Program of the Ministry of Science and Technology of China (No. 2015AA020508), the National Natural Science Foundation of China (91520203, 31571076, 31171054), and the Shanghai Key Discipline Foundation (B111) to H. Yu. The authors declare no potential conflicts of interest with respect to the authorship and/ or publication of this article.
REFERENCES Abarca M, Steenberghe D, Malevez C, Jacobs R (2006) The neurophysiology of osseointegrated oral implants. A clinically underestimated aspect. J Oral Rehabil 33:161–169. Aitkin LM, Merzenich MM, Irvine DRF, Clarey JC, Nelson JE (1986) Frequency representation in auditory cortex of the common marmoset (Callithrix jacchus jacchus). J Comp Neurol 252:175–185. An X, Gong H, Qian L, Wang X, Pan Y, Zhang X, Yang Y, Wang W (2012) Distinct functional organizations for processing different motion signals in V1, V2, and V4 of macaque. J Neurosci 32:13363–13379. Arieli A, Grinvald A (2002) Optical imaging combined with targeted electrical recordings, microstimulation, or tracer injections. J Neurosci Methods 116:15–28. Avivi-Arber L, Lee JC, Sessle BJ (2010) Effects of incisor extraction on jaw and tongue motor representations within face sensorimotor cortex of adult rats. J Comp Neurol 518:1030–1045. Bauer AQ, Kraft AW, Wright PW, Snyder AZ, Lee JM, Culver JP (2014) Optical imaging of disrupted functional connectivity following ischemic stroke in mice. NeuroImage 99:388–401. Calford MB (2005) Degree of adaptability of the somatosensory cortex to change: prospects for integration of bone-mounted dental prostheses. Clin Exp Pharmacol Physiol 32:115–118. Chen LM, Friedman RM, Roe AW (2003) Optical imaging of a tactile illusion in area 3b of the primary somatosensory cortex. Science 302:881–885. Chen Y, Li H, Jin Z, Shou T, Yu H (2014) Feedback of the amygdala globally modulates visual response of primary visual cortex in the cat. NeuroImage 84:775–785. Clemo HR, Stein BE (1982) Somatosensory cortex: a ’new’ somatotopic representation. Brain Res 235:162–168. Grinvald A, Shoham D, Shmuel A, Glaser D, Vanzetta I, Shtoyerman E, Slovin H, Wijnbergen C, Hildesheim R, Sterkin A (1999) Modern techniques in neuroscience research. In: Windhorst U, Johansson H, editors. In-vivo optical imaging of cortical architecture and dynamics, chapter 34. Springer: SpringerVerlag, Berlin. p. 893–969. Habre-Hallage P, Hermoye L, Gradkowski W, Jacobs R, Reychler H, Grandin CB (2010) A manually controlled new device for punctuate mechanical stimulation of teeth during functional magnetic resonance imaging studies. J Clin Periodontol 37:863–872. Habre-Hallage P, Dricot L, Jacobs R, van Steenberghe D, Reychler H, Grandin CB (2012) Brain plasticity and cortical correlates of osseoperception revealed by punctate mechanical stimulation of
osseointegrated oral implants during fMRI. Eur J Oral Implantol 5:175–190. Henry EC, Marasco PD, CataniaK C (2005) Plasticity of the cortical dentition representation after tooth extraction in naked mole-rats. J Comp Neurol 485:64–74. Horinuki E, Shinoda M, Shimizu N, Koshikawa N, Kobayashi M (2015) Orthodontic force facilitates cortical responses to periodontal stimulation. J Dent Res 94:1158–1166. Hubel DH, Wiesel TN, Stryker MP (1978) Anatomical demonstration of orientation columns in macaque monkey. J Comp Neurol 177:361–379. Jantsch HH, Kemppainen P, Ringler R, Handwerker HO, Forster C (2005) Cortical representation of experimental tooth pain in humans. Pain 118:390–399. Kim SA, Jun SB (2013) In-vivo optical measurement of neural activity in the brain. Exp Neurobiol 22:158–166. Kimoto K, Ono Y, Tachibana A, Hirano Y, Otsuka T, Ohno A, et al. (2011) Chewing-induced regional brain activity in edentulous patients who received mandibular implant-supported over dentures: a preliminary report. J Prosthodont Res 55: 89–97. Landgren S, Olsson KA (1980) Low threshold afferent projections from the oral cavity and the face to the cerebral cortex of the cat. Exp Brain Res 39:133–147. Lee SB, Beak SK, Park SH, Waterhouse BD, Lee HS (2009) Collateral projection from the locus coeruleus to whisker-related sensory and motor brain regions of the rat. J Comp Neurol 514:387–402. Lund JP, Sessle BJ (1971) Oral-facial and jaw muscle afferent projections to neurons in cat frontal cortex. Exp Neurol 45:314–331. Nakamura H, Shirakawa T, Koshikawa N, Kobayashi M (2016) Distinct excitation to pulpal stimuli between somatosensory and insular cortices. J Dent Res 95:180–187. Nikara T, Bishop PO, Pettigrew JD (1968) Analysis of retinal correspondence by studying receptive fields of binocular single units in cat striate cortex. Exp Brain Res 6:353–372. Pan Y, Chen M, Yin J, An X, Zhang X, Lu Y, Gong H, Li W, Wang W (2012) Equivalent representation of real and illusory contours in macaque V4. J Neurosci 32:6760–6770. Schwartz TH (2005) The application of optical recording of intrinsic signals to simultaneously acquire functional, pathological and localizing information and its potential role in neurosurgery. Stereotact Funct Neurosurg 83:36–44. Sheth SA, Nemoto M, Guiou M, Walker M, Pouratian N, Toga AW (2004) Linear and nonlinear relationships between neuronal activity, oxygen metabolism, and hemodynamic responses. Neuron 42:347–355. Taira K (1987a) The representation of the oral structures in the first somatosensory cortex of the cat. Brain Res 409:41–51. Taira K (1987b) Characteristics of periodontal mechanosensitive neurons in the first somatosensory cortex of the cat. Brain Res 409:52–61. Thompson JK, Peterson MR, Freeman RD (2003) Single-neuron activity and tissue oxygenation in the cerebral cortex. Science 299:1070–1072. Toda T, Taoka M (2004) Converging patterns of inputs from oral structures in the postcentral somatosensory cortex of conscious macaque monkeys. Exp Brain Res 158:43–49. Toda T, Taoka M (2006) Postcentral neurons with covert receptive fields in conscious macaque monkeys: their selective responsiveness to simultaneous two-point stimuli applied to discrete oral portions. Exp Brain Res 168:303–306. Tong L, Zhu B, Li Z, Shou T, Yu H (2011) Feedback from area 21a influences orientation but not direction maps in the primary visual cortex of the cat. Neurosci Lett 504:141–145. Trulsson M, van der Bilt A, Carlsson GE, Gotfredsen K, Larsson P, Mu¨ller F, Sessle BJ, Svensson P (2012) From brain to bridge: masticatory function and dental implants. J Oral Rehabil 39:858–877.
J. Tao et al. / Neuroscience 329 (2016) 254–263 Watanabe M, Tabata T, Karita K (1991) Facilitatory effect of jaw opening on somatosensory (S1) cortical neurons sensitive to tooth pressure in the cat. Archs Oral Biol 36:899–903. Yan C, Ye L, Zhen J, Ke L, Gang L (2008) Neuroplasticity of edentulous patients with implant-supported full dentures. Eur J Oral Sci 116:387–393.
263
Yu H, Chen X, Sun C, Shou T (2008) Global evaluation of contributions of GABAA, AMPA and NMDA receptors to orientation maps in cat’s visual cortex. NeuroImage 40:776–787. Yu H, Majewska AK, Sur M (2011) Rapid experience-dependent plasticity of synapse function and structure in ferret visual cortex in vivo. Proc Natl Acad Sci 108:21235–21240.
(Accepted 30 April 2016) (Available online 6 May 2016)