Masahito Nemoto

Linear and Nonlinear Relationships between Neuronal Activity, Oxygen Metabolism, and Hemodynamic Responses

We investigated the relationship between neuronal activity, oxygen metabolism, and hemodynamic responses in rat somatosensory cortex with simultaneous optical intrinsic signal imaging and spectroscopy, laser Doppler flowmetry, and local field potential recordings. Changes in cerebral oxygen consumption increased linearly with synaptic activity but with a threshold effect consistent with the existence of a tissue oxygen buffer. Modeling analysis demonstrated that the coupling between neuronal activity and hemodynamic response magnitude may appear linear over a narrow range but incorporates nonlinear effects that are better described by a threshold or power law relationship. These results indicate that caution is required in the interpretation of perfusion-based indicators of brain activity, such as functional magnetic resonance imaging (fMRI), and may help to refine quantitative models of neurovascular coupling.

Evaluation of coupling between optical intrinsic signals and neuronal activity in rat somatosensory cortex.

We investigated the coupling between perfusion-related brain imaging signals and evoked neuronal activity using optical imaging of intrinsic signals (OIS) at 570 and 610 nm. OIS at 570 nm reflects changes in cerebral blood volume (CBV), and the 610 nm response is related to hemoglobin oxygenation changes. We assessed the degree to which these components of the hemodynamic response were coupled to neuronal activity in rat barrel, hindpaw, and forepaw somatosensory cortex by simultaneously recording extracellular evoked field potentials (EPs) and OIS while varying stimulation frequency. In all stimulation paradigms, 10 Hz stimulation evoked the largest optical and electrophysiological responses. Across all animals, the 610 late phase and 570 responses correlated linearly with sigmaEP (P < 0.05) during both whisker deflection and electrical hindpaw stimulation, but the 610 early phase did not (whisker P = 0.27, hindpaw P = 0.28). The signal-to-noise ratio (SNR) of the 610 early phase (whisker 3.1, hindpaw 5.3) was much less than that for the late phase (whisker 14, hindpaw 51) and 570 response (whisker 11, hindpaw 46). During forepaw stimulation, however, the 610 early phase had a SNR (17) higher than that during hindpaw stimulation and correlated well with neuronal activity (P < 0.05). We conclude that the early deoxygenation change does not correlate consistently with neuronal activity, possibly because of its low SNR. The robust CBV-related response, however, has a high SNR and correlates well with evoked cortical activity.

Columnar Specificity of Microvascular Oxygenation and Volume Responses: Implications for Functional Brain Mapping

Cortical neurons with similar properties are grouped in columnar structures and supplied by matching vascular networks. The hemodynamic response to neuronal activation, however, is not well described on a fine spatial scale. We investigated the spatiotemporal characteristics of microvascular responses to neuronal activation in rat barrel cortex using optical intrinsic signal imaging and spectroscopy. Imaging was performed at 570 nm to provide functional maps of cerebral blood volume (CBV) changes and at 610 nm to estimate oxygenation changes. To emphasize parenchymal rather than large vessel contributions to the functional hemodynamic responses, we developed an ANOVA-based statistical analysis technique. Perfusion-based maps were compared with underlying neuroanatomy with cytochrome oxidase staining. Statistically determined CBV responses localized accurately to individually stimulated barrel columns and could resolve neighboring columns with a resolution better than 400 μm. Both CBV and early oxygenation responses extended beyond anatomical boundaries of single columns, but this vascular point spread did not preclude spatial specificity. These results indicate that microvascular flow control structures providing targeted flow increases to metabolically active neuronal columns also produce finely localized changes in CBV. This spatial specificity, along with the high contrast/noise ratio, makes the CBV response an attractive mapping signal. We also found that functional oxygenation changes can achieve submillimeter specificity not only during the transient deoxygenation (“initial dip”) but also during the early part of the hyperoxygenation. We, therefore, suggest that to optimize hemodynamic spatial specificity, appropriate response timing (using ≤2-3 sec changes) is more important than etiology (oxygenation or volume).

Functional signal- and paradigm-dependent linear relationships between synaptic activity and hemodynamic responses in rat somatosensory cortex

Linear relationships between synaptic activity and hemodynamic responses are critically dependent on functional signal etiology and paradigm. To investigate these relationships, we simultaneously measured local field potentials (FPs) and optical intrinsic signals in rat somatosensory cortex while delivering a small number of electrical pulses to the hindpaw with varied stimulus intensity, number, and interstimulus interval. We used 570 and 610 nm optical signals to estimate cerebral blood volume (CBV) and oxygenation, respectively. The spatiotemporal evolution patterns and trial-by-trial correlation analyses revealed that CBV-related optical signals have higher fidelity to summed evoked FPs (ΣFPs) than oxygenation-derived signals. CBV-related signals even correlated with minute ΣFP fluctuations within trials of the same stimulus condition. Furthermore, hemodynamic signals (CBV and late oxygenation signals) increased linearly with ΣFP while varying stimulus number, but they exhibited a threshold and steeper gradient while varying stimulus intensity, suggesting insufficiency of the homogeneity property of linear systems and the importance of spatiotemporal coherence of neuronal population activity in hemodynamic response formation. These stimulus paradigm-dependent linear and nonlinear relationships demonstrate that simple subtraction-based analyses of hemodynamic signals produced by complex stimulus paradigms may not reflect a difference in ΣFPs between paradigms. Functional signal- and paradigm-dependent linearity have potentially profound implications for the interpretation of perfusion-based functional signals.

Spatiotemporal Evolution of Functional Hemodynamic Changes and Their Relationship to Neuronal Activity

Brain imaging techniques such as functional magnetic resonance imaging (fMRI) have provided a wealth of information about brain organization, but their ability to investigate fine-scale functional architecture is limited by the spatial specificity of the hemodynamic responses upon which they are based. We investigated the spatiotemporal evolution of hemodynamic responses in rat somatosensory cortex to electrical hindpaw stimulation. We combined the advantages of optical intrinsic signal imaging and spectroscopy to produce high-resolution two-dimensional maps of functional changes in tissue oxygenation and blood volume. Cerebral blood flow changes were measured with laser-Doppler flowmetry, and simultaneously recorded field potentials allowed comparison between hemodynamic changes and underlying neuronal activity. For the first 2 to 3 secs of activation, hemodynamic responses overlapped in a central parenchymal focus. Over the next several seconds, cerebral blood volume changes propagated retrograde into feeding arterioles, and oxygenation changes anterograde into draining veins. By 5 to 6 secs, responses localized primarily in vascular structures distant from the central focus. The peak spatial extent of the hemodynamic response increased linearly with synaptic activity. This spatial spread might be because of lateral subthreshold activation or passive vascular overspill. These results imply early microvascular changes in volume and oxygenation localize to activated neural columns, and that spatial specificity will be optimal within a 2- to 3-sec window after neuronal activation.

Cortical spreading depression produces long-term disruption of activity-related changes in cerebral blood volume and neurovascular coupling

Cortical spreading depression (CSD) is a pronounced depolarization of neurons and glia that spreads slowly across the cortex followed by a period of depressed electrophysiological activity. The vascular changes associated with CSD are a large transient increase in blood flow followed by a prolonged decrease lasting greater than 1 h. Currently, the profile of functional vascular activity during this hypovolemic period has not been well characterized. Perfusion-based imaging techniques such as functional magnetic resonance imaging (fMRI) assume a tight coupling between changes in neuronal and vascular activity. Under normal conditions, these variables are well correlated. Characterizing the effect of CSD on this relationship is an important step to understand the impact acute pathophysiological events may have on neurovascular coupling. We examine the effect of CSD on functional changes in cerebral blood volume (CBV) evoked by cortical electrophysiological activity for 1 h following CSD induction. CBV signal amplitude, duration, and time to peak show little recovery at 60 min post-induction. Analysis of spontaneous vasomotor activity suggests a decrease in vascular reactivity may play a significant role in the disruption of normal functional CBV responses. Electrophysiological activity is also attenuated but to a lesser degree. CBV and evoked potentials are not well correlated following CSD, suggesting a breakdown of the neurovascular coupling relationship.

Co-activation of the secondary somatosensory and auditory cortices facilitates frequency discrimination of vibrotactile stimuli

The contribution of the auditory cortex to tactile information processing was studied by measuring somatosensory evoked magnetic fields (SEFs). Three kinds of vibrotactile stimuli with frequencies of 180, 280 and 380 Hz were randomly delivered on the right index finger with a probability of 40, 20 and 40%, respectively. Twenty normal subjects participated in four kinds of tasks: a control condition to ignore these stimuli, a simple task to discriminate the 280-Hz stimulus from the other two stimuli (discrimination task for the vibrotactile stimuli, Ts task), a feedback task modified from the Ts task by adding acoustic feedback of the vibratory frequency at 1300 ms poststimulus (tactile discrimination with auditory clues, TA), and an easy version of the TA task (TA-easy) to discriminate the 280-Hz stimulus (20% target) from the 180- or 380-Hz stimuli (80% nontarget). The Ts and TA tasks required accurate perception of the vibrotactile frequencies to discriminate among the three kinds of stimuli. Under such a task demand, the post hoc auditory feedback in the TA task was expected to induce acoustic imagery for the tactile sensation. The SEFs for the nontarget stimuli were analyzed. A middle-latency component (M150/200) was specifically evoked by the three discrimination tasks. In the Ts and TA-easy tasks, the M150/200 source indicated inferior parietal cortical activities (SII area). In the TA task, 11 subjects showed activity in both the SII area and the superior temporal auditory region and increased accuracy of discrimination compared with the Ts task, in contrast with other subjects who showed activity only in the SII area and small changes in task accuracy between the Ts and TA tasks. Asynchronous auditory feedback for the vibrotactile sensation induced the auditory cortex activity in the SEFs in relation to the progress in tactile discrimination, which suggested an induction of acoustic imagery to complement the tactile information processing.

Intraoperative monitoring of depth-dependent hemoglobin concentration changes during carotid endarterectomy by time-resolved spectroscopy

By measuring the adult human head during carotid endarterectomy, we investigate the depth sensitivity of two methods for deriving the absorption coefficient changes (Dmu(a)) from time-resolved reflectance data to absorption changes in inhomogeneous media: (1) the curve-fitting method based on the diffusion equation (DE-fit method) and (2) the time-independent calculation based on the modified Lambert-Beer law (MLB method). Remarkable differences in the determined values of Dmu(a) caused by clamping the external carotid artery and subsequently clamping the common carotid artery were observed between the methods. The DE-fit method was more sensitive to mu(a) changes in cerebral tissues, whereas the MLB method was rather sensitive to mu(a) changes in the extracerebral tissues. Our results indicated that the DE-fit was useful for monitoring the cerebral blood circulation and oxygenation during neurosurgical operations. In addition, the combined evaluation of mu(a) changes with the DE-fit and MLB methods will provide us with more available information about the hemodynamic changes in the depth direction.

28. Acoustic imagery for tactile sensation: Contribution of the auditory cortex to vibrotactile frequency discrimination

Complementary interaction between the somatosensory and auditory cortices was studied by measuring the evoked magnetic fields. Fourteen normal volunteers were asked to discriminate three kinds of vibrotactile stimuli (180, 280 and 380 Hz) delivered to the right index finger under two conditions with and without the acoustic feedback of the vibratory frequencies at 1.3 s after the tactile stimuli (T and TA tasks, respectively). In the 10 subjects, the TA task was more accurately performed (the error rate of 9.6%) than the T task (31.9%), which was accompanied by induction of co-activation of the SII and supratemporal auditory cortices in a late SEF-component (140–250 ms) and the SII response to the feedback sound in a late AEF-component (130– 280 ms). Such cross-modal activities were not observed in other 4 subjects showing small changes in the error rate between the two tasks (2.5%; T–TA). In contrast, even with the feedback sound just like the TA task, oddball tasks (280 vs. either of 180 or 380 Hz-vibration) to detect the deviance of stimuli rather than the vibrotactile frequencies induced no cross-modal cortical activity. These results suggest a ’binding’ of asynchronous somatosensory and auditory inputs to complement vibrotactile frequency information processing.