A cortical rat hemodynamic response function for improved detection of BOLD activation under common experimental conditions.

For a reliable estimation of neuronal activation based on BOLD fMRI measurements an accurate model of the hemodynamic response is essential. Since a large part of basic neuroscience research is based on small animal data, it is necessary to characterize a hemodynamic response function (HRF) which is optimized for small animals. Therefore, we have determined and investigated the HRFs of rats obtained under a variety of experimental conditions in the primary somatosensory cortex. Measurements were performed on animals of different sex and strain, under different anesthetics, with and without ventilation and using different stimulation modalities. All modalities of stimulation used in this study induced neuronal activity in the primary somatosensory cortex or in subcortical regions. Since the HRFs of the BOLD responses in the primary somatosensory cortex showed a close concordance for the different conditions, we were able to determine a cortical rat HRF. This HRF is based on 143 BOLD measurements of 76 rats and can be used for statistical parametric mapping. It showed substantially faster progression than the human HRF, with a maximum after 2.8 ± 0.8 s, and a following undershoot after 6.1 ± 3.7 s. If the rat HRF was used statistical analysis of rat data showed a significantly improved detection performance in the somatosensory cortex in comparison to the commonly used HRF based on measurements in humans.

Introducing Specificity to Iron Oxide Nanoparticle Imaging by Combining 57Fe-Based MRI and Mass Spectrometry.

Iron oxide nanoparticles (ION) are highly sensitive probes for magnetic resonance imaging (MRI) that have previously been used for in vivo cell tracking and have enabled implementation of several diagnostic tools to detect and monitor disease. However, the in vivo MRI signal of ION can overlap with the signal from endogenous iron, resulting in a lack of detection specificity. Therefore, the long-term fate of administered ION remains largely unknown, and possible tissue deposition of iron cannot be assessed with established methods. Herein, we combine nonradioactive 57Fe-ION MRI with ex vivo laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) imaging, enabling unambiguous differentiation between endogenous iron (56Fe) and iron originating from applied ION in mice. We establish 57Fe-ION as an in vivo MRI sensor for cell tracking in a mouse model of subcutaneous inflammation and for assessing the long-term fate of 57Fe-ION. Our approach resolves the lack of detection specificity in ION imaging by unambiguously recording a 57Fe signature.

Anesthesia differentially modulates neuronal and vascular contributions to the BOLD signal.

Most studies involving BOLD fMRI in basic neuroscience research are conducted with anesthetized animals. This study investigates neural and hemodynamic activity through a combination of experiments comprising BOLD fMRI, optical calcium recordings and ASL in vivo. Patch clamp experiments of neurons were conducted to evaluate electrophysiological correlates of neural activity in vitro. Various anesthetic conditions embracing numerous anesthetic depths evoked by different concentrations of isoflurane (ISO) and different degrees of hypercapnia under a constant stimulus were investigated. We observed that different anesthetic conditions had major impact on the results obtained, particularly that anesthesia could cause a massive divergence of different experimental modalities. In ventilated animals, robust BOLD responses were detectable even with relatively deep anesthesia, while in non-ventilated animals, BOLD responses were not detectable under these conditions. This was most likely due to hypercapnia caused by respiratory depression, as in ventilated animals administered CO2 had the same effect. This observation agreed with measurements of perfusion, which showed that inhaled CO2 increased perfusion significantly, while ISO did not. In optical calcium measurements, higher concentrations of ISO decreased spontaneous neural activity, but not stimulus-evoked responses. This observation was explained by a generally lower excitability of neurons under ISO, which suppressed spontaneous activity, and consequently left more neurons available to fire synchronously in response to a stimulus. Interpreting this phenomenon as an integrated signal of independent single neurons was supported by patch clamp experiments as the number of action potentials (APs) per stimulus was decreased by addition of CO2. Addition of ISO on the other hand had no significant effect. Our results provide an explanation on the cellular level for anesthesia-dependent observations in previous studies of task-induced BOLD and resting state connectivity. They further inform selection of the adequate anesthetic regimen for a given combination of modalities.

Line scanning fMRI reveals earlier onset of optogenetically evoked BOLD response in rat somatosensory cortex as compared to sensory stimulation.

The combination of optogenetic control and fMRI readout in the brain is increasingly used to assess neuronal networks and underlying signal processing. However, how exactly optogenetic activation or inhibition reproduces normal physiological input has not been fully unraveled. To assess details of temporal dynamics of the hemodynamic response, temporal resolution in rodent fMRI is often not sufficient. Recent advances in human fMRI using faster acquisition schemes cannot be easily translated to small animals due to smaller dimensions, fast physiological motion, and higher sensitivity to artefacts. Here, we applied a one dimensional line scanning acquisition with 50ms temporal resolution in rat somatosensory cortex. We observed that optogenetic activation reproduces the hemodynamic response upon sensory stimulation, but shows a 160 to 340ms earlier onset of the response. This difference is explained by direct activation of all opsin-expressing and illuminated cortical layers, while hemodynamic response to sensory stimulation is delayed during intracortical transmission between cortical layers. Our results confirm that optogenetic activation is a valid model for physiological neuronal input, and that differences in temporal behavior of only a few hundred milliseconds can be resolved in rodent fMRI.

True and apparent optogenetic BOLD fMRI signals.

PURPOSE: Optogenetic fMRI (ofMRI) is a novel tool in neurophysiology and neuroimaging. The method is prone to light-induced artifacts, two of which were investigated in this study. METHODS: ofMRI was performed in rats using two excitatory opsins (ChR2 and C1V1TT ) virally transduced in somatosensory cortex or thalamus. Heat-induced apparent BOLD activation at the site of the optical fiber and stimulation light-induced activation of the visual pathways were investigated, and control experiments for these two artifacts were established. RESULTS: Specific optogenetic BOLD activation was observed with both opsins, accompanied by BOLD in the visual pathways. Unspecific heat-induced BOLD was ruled out by a control experiment employing low-level constant illumination in addition to pulsed optogenetic stimulation. Activation of the visual pathways was confirmed to be physiological by direct visual stimulation of the eyes and was suppressed by additional low-level constant light to the eyes. Light inside the brain was identified as one source of the BOLD signal observed in the visual pathways. CONCLUSION: ofMRI is a method of tremendous potential, but unspecific activations in fMRI not caused by the activation of opsins must be avoided or recognized as such. The control experiments presented here allow for validating the specificity of optogenetic stimulation. Magn Reson Med 77:126-136, 2017. © 2016 Wiley Periodicals, Inc.

Disentangling the impact of cerebrospinal fluid formation and neuronal activity on solute clearance from the brain.

BACKGROUND: Despite recent attention, pathways and mechanisms of fluid transposition in the brain are still a matter of intense discussion and driving forces underlying waste clearance in the brain remain elusive. Consensus exists that net solute transport is a prerequisite for efficient clearance. The individual impact of neuronal activity and cerebrospinal fluid (CSF) formation, which both vary with brain state and anesthesia, remain unclear. METHODS: To separate conditions with high and low neuronal activity and high and low CSF formation, different anesthetic regimens in naive rat were established, using Isoflurane (ISO), Medetomidine (MED), acetazolamide or combinations thereof. With dynamic contrast-enhanced MRI, after application of low molecular weight contrast agent (CA) Gadobutrol to cisterna magna, tracer distribution was monitored as surrogate for solute clearance. Simultaneous fiber-based Ca2+-recordings informed about the state of neuronal activity under different anesthetic regimen. T2-weighted MRI and diffusion-weighted MRI (DWI) provided size of subarachnoidal space and aqueductal flow as surrogates for CSF formation. Finally, a pathway and mechanism-independent two-compartment model was introduced to provide a measure of efficiency for solute clearance from the brain. RESULTS: Anatomical imaging, DWI and Ca2+-recordings confirmed that conditions with distinct levels of neuronal activity and CSF formation were achieved. A sleep-resembling condition, with reduced neuronal activity and enhanced CSF formation was achieved using ISO+MED and an awake-like condition with high neuronal activity using MED alone. CA distribution in the brain correlated with the rate of CSF formation. The cortical brain state had major influence on tracer diffusion. Under conditions with low neuronal activity, higher diffusivity suggested enlargement of extracellular space, facilitating a deeper permeation of solutes into brain parenchyma. Under conditions with high neuronal activity, diffusion of solutes into parenchyma was hindered and clearance along paravascular pathways facilitated. Exclusively based on the measured time signal curves, the two-compartment model provided net exchange ratios, which were significantly larger for the sleep-resembling condition than for the awake-like condition. CONCLUSIONS: Efficiency of solute clearance in brain changes with alterations in both state of neuronal activity and CSF formation. Our clearance pathway and mechanism agnostic kinetic model informs about net solute transport, solely based on the measured time signal curves. This rather simplifying approach largely accords with preclinical and clinical findings.

Retrosplenial Cortex Contributes to Network Changes during Seizures in the GAERS Absence Epilepsy Rat Model.

Resting state-fMRI was performed to explore brain networks in Genetic Absence Epilepsy Rats from Strasbourg and in nonepileptic controls (NEC) during monitoring of the brain state by simultaneous optical Ca2+-recordings. Graph theoretical analysis allowed for the identification of acute and chronic network changes and revealed preserved small world topology before and after seizure onset. The most prominent acute change in network organization during seizures was the segregation of cortical regions from the remaining brain. Stronger connections between thalamic with limbic regions compared with preseizure state indicated network regularization during seizures. When comparing between strains, intrathalamic connections were prominent in NEC, on local level represented by higher thalamic strengths and hub scores. Subtle differences were observed for retrosplenial cortex (RS), forming more connections beyond cortex in epileptic rats, and showing a tendency to lateralization during seizures. A potential role of RS as hub between subcortical and cortical regions in epilepsy was supported by increased numbers of parvalbumin-positive (PV+) interneurons together with enhanced inhibitory synaptic activity and neuronal excitability in pyramidal neurons. By combining multimodal fMRI data, graph theoretical methods, and electrophysiological recordings, we identified the RS as promising target for modulation of seizure activity and/or comorbidities.

Functional MRI Readouts From BOLD and Diffusion Measurements Differentially Respond to Optogenetic Activation and Tissue Heating.

Functional blood-oxygenation-level-dependent (BOLD) MRI provides a brain-wide readout that depends on the hemodynamic response to neuronal activity. Diffusion fMRI has been proposed as an alternative to BOLD fMRI and has been postulated to directly rely on neuronal activity. These complementary functional readouts are versatile tools to be combined with optogenetic stimulation to investigate networks of the brain. The cell-specificity and temporal precision of optogenetic manipulations promise to enable further investigation of the origin of fMRI signals. The signal characteristics of the diffusion fMRI readout vice versa may better resolve network effects of optogenetic stimulation. However, the light application needed for optogenetic stimulation is accompanied by heat deposition within the tissue. As both diffusion and BOLD are sensitive to temperature changes, light application can lead to apparent activations confounding the interpretation of fMRI data. The degree of tissue heating, the appearance of apparent activation in different fMRI sequences and the origin of these phenomena are not well understood. Here, we disentangled apparent activations in BOLD and diffusion measurements in rats from physiological activation upon sensory or optogenetic stimulation. Both, BOLD and diffusion fMRI revealed similar signal shapes upon sensory stimulation that differed clearly from those upon heating. Apparent activations induced by high-intensity light application were dominated by T2 ∗-effects and resulted in mainly negative signal changes. We estimated that even low-intensity light application used for optogenetic stimulation reduces the BOLD response close to the fiber by up to 0.4%. The diffusion fMRI signal contained T2, T2 ∗ and diffusion components. The apparent diffusion coefficient, which reflects the isolated diffusion component, showed negative changes upon both optogenetic and electric forepaw stimulation. In contrast, positive changes were detected upon high-intensity light application and thus ruled out heating as a major contributor to the diffusion fMRI signal.

Multimodal Functional Neuroimaging by Simultaneous BOLD fMRI and Fiber-Optic Calcium Recordings and Optogenetic Control.

Recent developments of optogenetic tools and fluorescence-based calcium recording techniques enable the manipulation and monitoring of neural circuits on a cellular level. Non-invasive imaging of brain networks, however, requires the application of methods such as blood oxygen level-dependent (BOLD) functional magnetic resonance imaging (fMRI), which is commonly used for functional neuroimaging. While BOLD fMRI provides brain-wide non-invasive reading of the hemodynamic response, it is only an indirect measure of neural activity. Direct observation of neural responses requires electrophysiological or optical methods. The latter can be combined with optogenetic control of neuronal circuits and are MRI compatible. Yet, simultaneous optical recordings are still limited to fiber-optic-based approaches. Here, we review the integration of optical recordings and optogenetic manipulation into fMRI experiments. As a practical example, we describe how BOLD fMRI in a 9.4-T small animal MR scanner can be combined with in vivo fiber-optic calcium recordings and optogenetic control in a multimodal setup. We present simultaneous BOLD fMRI and calcium recordings under optogenetic control in rat. We outline details about MR coil configuration, choice, and usage of opsins and chemically and genetically encoded calcium sensors, fiber implantation, appropriate light power for stimulation, and calcium signal detection, to provide a glimpse into challenges and opportunities of this multimodal molecular neuroimaging approach.

Cortex-wide BOLD fMRI activity reflects locally-recorded slow oscillation-associated calcium waves.

Spontaneous slow oscillation-associated slow wave activity represents an internally generated state which is characterized by alternations of network quiescence and stereotypical episodes of neuronal activity - slow wave events. However, it remains unclear which macroscopic signal is related to these active periods of the slow wave rhythm. We used optic fiber-based calcium recordings of local neural populations in cortex and thalamus to detect neurophysiologically defined slow calcium waves in isoflurane anesthetized rats. The individual slow wave events were used for an event-related analysis of simultaneously acquired whole-brain BOLD fMRI. We identified BOLD responses directly related to onsets of slow calcium waves, revealing a cortex-wide BOLD correlate: the entire cortex was engaged in this specific type of slow wave activity. These findings demonstrate a direct relation of defined neurophysiological events to a specific BOLD activity pattern and were confirmed for ongoing slow wave activity by independent component and seed-based analyses.