Action potential propagation in the rodent myelinated optic nerve does not trigger neurovascular coupling

Action potential propagation in the rodent myelinated optic nerve does not trigger neurovascular coupling | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Action potential propagation in the rodent myelinated optic nerve does not trigger neurovascular coupling David Esteban Suarez-Baquero, Davide Boido, Ahlem Assali, Tatiana V. Esipova, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9086876/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract In brain gray matter, neurovascular coupling (NVC) maintains brain metabolism homeostasis by modulating blood flow according to neuronal activity. In white matter, the energy cost of information transmission along myelinated axons is reduced and the need for NVC is unknown. Here, we used two-photon imaging through chronically-implanted GRIN lenses in mice and high-field BOLD fMRI (17.2T) in rats to investigate NVC along the entire length of the optic nerve, a unique model of a myelinated axonal tract. We found that flickering light and drifting grating stimulations increased blood flow in the retina, the unmyelinated optic nerve head, and at the level of the nerve synaptic terminals. However, it did not affect blood flow and oxygenation in the myelinated part of the optic nerve, i.e., the intracranial optic nerve and the optic tract. We conclude that during natural visual stimulation, action potential propagation in activated myelinated axons does not trigger NVC. Cellular & Molecular Neuroscience NVC neurovascular coupling fmri Two-photon microscopy GRIN lens white matter optic nerve Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction In brain gray matter, neuronal activation triggers an increase in blood flow (known as functional hyperemia) via neurovascular coupling (NVC), an ensemble of processes that involves the cooperation of neurons, astrocytes, smooth muscle cells, pericytes, and endothelial cells. The mechanisms and the timing by which these cell types interact and contribute to coupling are still debated 1 but there is a consensus that, in awake rodents, functional hyperemia is characterized by a hemodynamic response function (HRF) peaking at about ~1-2 s 2, 3 and is tightly correlated with transmitter release and postsynaptic activation. Little is known about NVC in white matter, but some recent blood-oxygen-level-dependent functional MRI (BOLD-fMRI) studies in humans have reported task-evoked responses 4, 5 : the BOLD signals measured were smaller than signals typically measured in gray matter, characterized by an HRF with different features 6 , but still occurred with a timing compatible with standard NVC, i.e., within a few seconds from stimulation onset. Since most white matter tracts contain both myelinated and unmyelinated axons, e.g., 30-40 % of mouse corpus callosum axons are unmyelinated 7, 8 , it remains unclear as to whether the mechanisms underlying NVC are similar for both axon types. Understanding the functional vascular dynamics and metabolism of white matter is an important issue, given widespread white matter lesions and lacunar infarcts are two key signs of cerebral small vessel diseases 9 . Whether NVC is triggered by local axonal release of glutamate and some feed-forward mechanisms, or conversely, is principally regulated by the energy demand resulting from action potential propagation remains an open question. Theoretical calculations indicate that the energy budget of firing is much lower than that of synaptic transmission 10 , raising the hypothesis that, at least in myelinated nerves characterized by their paucity of synapses, NVC may not be required for action potential propagation. To address these questions, the rodent optic nerve provides a unique model of white matter that, in contrast to the corpus callosum, contains only myelinated axons 11, 12 , oligodendrocytes and their precursor cells, vessels of all types, astrocytes and few specific axo-myelinic synapses 13 . Apart from the unmyelinated nerve head, which can be functionally observed through the animal’s eye lenses, the functional properties of the extracerebral myelinated part of the optic nerve - whether electrical, metabolic or vascular- have only been studied in vitro. During propagation of the compound action potential, there is a tight coupling between axons and oligodendrocytes. This process involves calcium influxes in both cell types and a complex cascade of events, including the release of glutamate, K+ ions, lactate, and the activation of glycolysis and lipid metabolism in oligodendrocytes 14-17 . With regard to NVC, a specific form of axo-vascular coupling was recently reported, which did not fulfill the properties of NVC, as it was only marginally measurable upon strong synchronous electrical stimulation and occurred with dynamics of minutes rather than seconds 18 . Moreover, the in vitro experimental conditions did not allow for the ability to: 1) predict what may occur during natural light stimulation, since light and arousal change the balance of the ganglion cell type activated 19, 20 with an unknown effect on the nerve metabolism, and, 2) correctly assess NVC, which requires vessels with the adequate pressure tone. Therefore, in order to characterize NVC in the whole optic nerve, a non-invasive multiscale functional imaging approach in vivo was used, which allows for measuring signals related to blood flow and blood oxygenation in response to light, with high spatio-temporal resolution. Our results reveal that while visual stimulation triggers robust NVC in the retina, the unmyelinated optic nerve head and at the level of the nerve synaptic terminals, it does not elicit any measurable hemodynamic or oxygenation changes in the myelinated portions of the optic nerve and optic tract. These findings demonstrate that action potential propagation in activated myelinated axons does not trigger NVC. Results fMRI BOLD responses along the entire optic nerve In order to non-invasively investigate NVC along the entire length of the optic nerve, we first used BOLD fMRI and measured functional hyperemia from the retina to the nerve terminals in response to visual stimulation. Such a mesoscopic approach is non-invasive, and the strong magnetic field (17.2T) and gradients (1 T/m) used 21 , 22 ensured acquisition of images with high sensitivity and spatial resolution. Note that ganglion cell axons are fully myelinated in both the extracerebral part of the optic nerve (ON) and in the intracerebral part, or optic tract (OT). Data acquisition was performed in sedated rats, rather than mice, to maximize the number of voxels containing the optic nerve and thus improve the signal-to-noise ratio of the BOLD signal. At the beginning of each session, multi-slice gradient echo-planar imaging (GRE-EPI) functional BOLD acquisitions were performed to verify that visual stimulation evoked functional responses in the dorsal lateral geniculate nucleus (dLGN) or in the superior colliculus (SC). A visual flickering light paradigm was used, which consisted of 6 s LED “ON” (blue light flashing at 2 Hz) and 12 s LED “OFF” (repeated 20 times per session). To precisely target the ON, we first used diffusion weighted spin echo (DW-SE) acquisitions with 6 horizontal slices, positioned from the retina and the ON head down to the chiasm. Anatomical T 2 weighted images were then acquired, followed by BOLD acquisitions, during 7 consecutive sessions. Post hoc co-registration of DW-SE and T 2 weighted images allowed for measurement of functional responses located exclusively in the voxels containing the ON, from the retina to the most ventral part of the OT. Averaged trials improved the image SNR and the analysis was then performed without any assumptions about the timing and shape of the responses (4). Similar to what has been shown previously 23 , 24 , visual stimulation generated a significant BOLD signal at the level of the retina and the ON head (Fig. 1 ), a part of the nerve which is unmyelinated. In contrast, no signal was detected in the intracranial ON, the chiasm or the most ventral part of the optic tract (OT). In a single case, a minor increase in signal was detected rostrally in the initial segment of the ON, the total length of which is estimated to be approximately 8–9 mm, which can be attributed to a venous signal draining the activated retina and ON head or a vascular signal backpropagating along the feeding arteriole 25 . This signal also appeared in some functional activation maps (Supplementary Figs. 2 and 3) computed with GLM analysis, using a boxcar function representing the stimulation time and duration, shifted by 2 seconds, and a very low statistical threshold (p < 0.01). Even with such threshold, NVC was not detected in the main part on the ON and the chiasm. Next, to further investigate NVC in the OT, BOLD responses were acquired in 7 coronal slices (800 µm thick) from the optic chiasm back to the superior colliculus. Anatomical, T 2 weighted acquisitions were performed allowing for determination of voxels containing the OT, which appeared as a well delimited dark shadow in the first 2–3 slices caudal to the optic chiasm. In these slices, no BOLD response was detected in the ventral OT (Fig. 2 ), a part of the tract that is dense in axons and can be assumed to contain exclusively ganglion cell axons. In the caudal slices (> 4th slice) where the OT could not be sorted out anatomically from the T 2 weighed images, BOLD signals were detected in the dLGN and the vLGN. Functional activation maps similarly showed that BOLD signals were not detectable in the ventral OT (Supplementary Fig. 3). Taken together, BOLD fMRI data suggest that in response to visual stimulation, action potential propagation along the entire myelinated part of the optic nerve does not require NVC. To confirm this finding, additional studies were performed to exclude the possibility that the absence of a BOLD response was due either to the unlikely possibility that part of the nerve lacks a major vascular structure necessary to support NVC, or alternatively, due to insufficient signal-to-noise ratio. ON and OT vascular structure We used 3D light sheet microscopy and tissue clearing to reconstruct and compare the vascular architecture of the extracerebral and intracerebral portions of the nerve, by immunolabelling whole adult mouse brains for CD31, Podocalyxin and Sm22 with iDISCO+, as in 26 . Figure 3 a,b shows that a large arteriole enters the ON midway between the chiasm and the eye, sending small collaterals in the ON and runs rostrally to the ON head. In contrast, a large vein runs on the side of the ON, from the eye to the chiasm. We suggest that this vein underlies the BOLD fMRI signal that can occasionally be observed rostrally in the initial segment of the ON. A moderate density of capillaries, quantified by the number of branching points, was observed (Fig. 3 e, f), in line with Restrepo et al. 18 , who additionally reported a pericyte density almost similar to that of the hippocampus or visual cortex. The ventral part of the OT (Fig. 3 c, d) is traversed by a few arterioles (fed by the thalamic artery) and venules, and also has a moderately dense capillary network, as shown in the ON. Quantitative analysis also revealed (Fig. 3 g) that the vessel density of the ventral OT is comparable to that of the neighbouring gray matter region in the cortical sub-plate, the postero-dorsal part of the medial amygdala nucleus, but much less than the highly vascularized lateral thalamus 26 . These results indicate that the strength of NVC is unrelated to capillary density and that the ON and OT vascular structures do not explain the lack of NVC in response to a natural visual stimulation. Two-photon imaging of NVC in the ON In contrast to BOLD fMRI, two-photon microscopy is an approach that can be used to ascertain that NVC does not occur in the ON: in a small capillary, it detects single red blood cells (RBCs) and consequently, RBC velocity or flow changes can be monitored with a 100 ms temporal resolution 27 , 28 . We thus developed a preparation in which a long gradient-index lens (GRIN lens, length: 9 mm, diameter: 0.5 mm) was chronically implanted over the caudal ON in head-fixed mice, sedated or awake. Figure 4 a illustrates the experimental setup using the GRIN lens, which was fixed to the cranium with its lower end placed at the contact of the dura over the right optic nerve, about 1.5 mm rostral to the optic chiasm. This approach left the ON intact and two-photon imaging was performed at least 3 weeks after surgery, allowing for full recovery. Due to the working distance of the GRIN lens (250 µm or 50 µm), the imaging plane (or sample plane) was systematically located within the ON. By moving the objective focus up and down, we could change the imaging plane in z and image vessels labelled with Texas Red, from the center to the surface of the ON (Supplementary Fig. 1a). Note that this occurred at the expense of spatial resolution, as the two-photon point spread function increased with distance from the imaging (sample) plane (Supplementary Fig. 1b and 29 ) and that the pixel size had to be corrected according to the z plane (Supplementary Fig. 1c and 30 ). Such corrections were required to measure the real value of RBC velocity. To ensure that GRIN lens implantation did not caused damage that could alter the ON, i.e., local action potential propagation, we used functional ultrasound imaging, a mesoscopic approach that can be used to record changes in the cerebral blood volume (CBV) fraction flowing along the axial axis 31 , and investigated whether vascular responses to visual stimulation could be observed downstream of the ON, i.e., in the dLGN or visual cortex of mice implanted with phantom lenses (0.5 mm metal round bar) or standard GRIN lenses (n = 3). Figure 4 c shows that visual stimulation triggered a CBV increase in the contralateral dLGN and visual cortex, demonstrating that the ON was functional. Finally, to control that we were able to measure RBC velocity changes, we verified that in sedated mice, a brief inhalation of isoflurane–which increases blood flow– (5 s, 5%, Fig. 4 d) efficiently triggered an increased in RBC velocity in ON capillaries imaged with the implanted GRIN lens. Two-photon measurements in small capillaries (diameter 2–5 µm) revealed that the mean resting velocity of RBCs was 0.46 ± 0.05 mm/s (STD 0.20, n = 19 capillaries) ranging from 0.16 mm/s to 0.86 mm/s (Fig. 5 a, b). Visual stimulation did not elicit a change in RBC velocity in any of the vessels in the ON. We then measured resting partial pressure of oxygen (pO 2 ) and pO 2 changes in ON capillaries and larger vessels using two-photon phosphorescence lifetime microscopy using the oxygen sensor Oxyphor 2P 32 . In the gray matter, hyperemia has been shown to always be accompanied by an increase in pO 2 in individual capillaries 33 – 35 , which results in an oxygenation rise in the surrounding tissue 36 , 37 . In the ON (Fig. 2 c, d), a mean resting pO 2 of 59.8 ± 12.8 mmHg was measured (ranging from 46.5 mmHg to 79.4 mmHg). As with blood flow, visual stimulation did not evoke any changes in capillary pO 2 demonstrating that during sedation, propagation of action potentials in the ON does not trigger NVC. Because anesthesia and sedation affect both neuronal activity and neurovascular coupling mechanisms 38 , we then tested whether NVC could be detected in the awake brain state. To do this, mice were habituated to head fixation and visual stimulation for 10–14 days, after which repeated RBC velocity and pO 2 measurements in ON capillaries were taken in awake mice, including some which were previously tested during sedation. Figure 4 e, g shows the resting values of RBC velocity and pO 2 in small capillaries of awake mice. As during sedation, visual stimulation did not induce any functional hyperemia or oxygenation changes, from 3 weeks to 8 months post-surgery. Note that NVC was not detectable whether vessels were considered independent (Fig. 5 b-h) or animal-dependent (Supplementary Fig. 4, mixed-effect model). Overall, these findings validate BOLD fMRI observations and demonstrate that in response to a visual stimulation, propagation of action potentials in the extracerebral or intracerebral axonal tract depends on a specific energy budget 16 , 17 that does not require NVC. Discussion Using two complementary functional imaging approaches, our study demonstrates that in rodents and upon light stimulation, action potentials can propagate in the ON and OT without triggering NVC. This absence of NVC appears to be restricted to the myelinated portion of the ganglion cell axonal tract. In the rodent retina, which contains a complex unmyelinated neuronal network that includes ganglion cell dendrites, somata and axons, light generates strong vascular responses that have been studied in vivo with BOLD fMRI 23 , 24 , fUS 39 , and with a microscopic resolution with confocal and two-photon microscopy. In such studies, NVC has been shown to occur diversely in the three retina vascular layers 40 , and with an involvement of astrocyte calcium that differs at the capillary and the arteriole level 41 , 42 . Concerning pericytes, since the initial demonstration that they can control capillary diameter 43 , most investigations have been performed in vitro (for reviews, see 44 – 46 ), with an in vivo study proposing that interpericyte tunneling nanotubes coordinate light-evoked responses between adjacent capillaries 47 . At the level of the ON head, ganglion cell axons are still unmyelinated, but light stimulation triggers vascular responses that are altered in a rat model of Alzheimer’s disease 39 and in caveolin-1 knockout mice 39 , 48 . In the dLGN where the nerve axons end, i.e., in the gray matter, NVC occurs 49 , 50 , but the differential weight of axon terminal firing and glutamate release, versus postsynaptic cell depolarization, in triggering of coupling has not yet been investigated. All this is in accordance with our BOLD fMRI observations which show that under the same experimental conditions, visual stimulation produces strong BOLD responses in the retina, the ON head and in the nerve targets, the dLGN, vLGN and SC. In contrast, no BOLD fMRI signals were detected in the ON, the chiasm nor the rostro-ventral OT, anatomically identified either by DW-SE (horizontal acquisitions) or T 2 weighted vertical acquisitions. Although we recognize that it is difficult to draw a definitive conclusion from a lack of BOLD responses, we believe that the absence of NVC demonstrated with two-photon microscopy in the ON can be extended to the ventral OT. What occurs at the level of the caudal (i.e. dorsal) part of the OT remains technically unclear because we estimate that at the level of the dLGN and vLGN, it is impossible to distinguish the thin dorso-lateral layer containing OT fibers from the neighboring nuclei per se. Indeed, due to the low rostro-caudal spatial resolution of BOLD fMRI, the voxels forming this layer most certainly include thalamic neurons in addition to OT axons 51 as well as other types of axons crossing the OT. Therefore, the caudal OT cannot be used as a model to assess a NVC that would be attributed exclusively to myelinated axons, although it is likely that it shares the same properties as the ventral OT. Our study also shows that the ON, and most probably other cranial nerves, can be functionally imaged in anesthetized or awake mice using a chronically-implanted GRIN lens. Moving the imaging plane in z, away from the working distance of the GRIN lens, allowed for scanning through the thickness of the ON. The displacement in z changes the imaging point spread function, i.e., the spatial resolution, and reduces the field of view, however, we showed that with a simple pixel size correction, GRIN lens imaging still allows for accurate RBC velocity measurements in small capillaries. This allowed us to demonstrate, by monitoring RBC velocity and pO 2 , that the ON transmits visual information without changes in blood flow and oxygenation. This observation may seem at odds with a previous in vitro study 18 , i.e. without physiological blood pressure tone, reporting that synchronous electrical stimulation of ON axons elicits a very small, slow, and delayed dilation of large vessels, a response that was proposed to be a form of NVC. The dynamics of the response reported, however, was too slow to account for NVC, which occurs within a few seconds of light stimulation in both the retina and the ON head, whether in vitro or in vivo 39 , 40 , 47 , 52 . It is important to underline an additional difference between the in vitro and in vivo ON models with respect to stimulation. In vitro, ON electrical stimulation activates axons synchronously, and from a presumably very low basal activity and metabolism. In vivo, there is a constant dark current in photoreceptors and a constant activity in various types of “OFF” ganglion cells. As a result, stimulation with flickering or drifting gratings changes the type and weight of activated ganglion cells 20 making it difficult to estimate the change in overall metabolic cost associated with a visual stimulation. Our two-photon data indicate that action potential propagation in the myelinated ON during light stimulation is not accompanied by any measurable NVC, despite its standard vascular organization. Could the absence of NVC result from our preparations or technical limitations? Arousal did not reveal NVC, even though ON responses in mice exhibit faster kinetics, a larger dynamic range, and higher firing activity in the awake state 20 . This supports the idea that the ON relies on a specific metabolism in which oligodendrocytes play a major role 16 , 53 , independently of brain state. The general scheme is that axonal firing increases extracellular potassium, which in turn triggers calcium influx and glycolysis in oligodendrocytes. This leads to lactate production and release, which is then taken up by axons to fuel ATP production. Our data suggest that this metabolic loop does not require NVC, whether in the myelinated ON of sedated or awake mice and rats. For both BOLD fMRI and two-photon experiments, it is impossible to rule out a minute vascular response that could be extracted from noise only after ~ 1000 stimulation repetitions. However, if such a response is masked by noise and by spontaneous fluctuations of blood flow, its physiological significance is questionable. The absence of a blood flow response also ensures that two-photon oxygen measurements and BOLD fMRI signals were not biased by the dual dependence of these signals on oxygen delivery and consumption. Finally, to process BOLD responses, we analyzed signals at the level of individual voxels selected anatomically, without any spatiotemporal smoothing and without any assumption on the response shape and timing. This should have allowed to extract small signals that would not pass conventional statistical thresholds. Note that using GLM analysis and a boxcar function representing the stimulation time and duration, shifted by 2 seconds, the activation maps did not show any BOLD response in the intracranial ON, even with statistical threshold of p < 0.01) (see Supplementary Fig. 2–3). We believe that our data may be extrapolated to humans because the human ON also contains exclusively myelinated fibers—unlike most other white-matter tracts—and because a recent BOLD fMRI study 54 indicates that visual stimulation does not elicit a response in the human ON. In gray matter, Harris and Attwell 10 calculated the proportion of ATP used by cellular processes to be 43% for synaptic transmission, 17% for action potentials, 15% for maintaining the resting membrane potential, and 25% for housekeeping (i.e., processes not related to signaling). In axons, the properties of Na + and K+ channels are tuned to minimize the energy cost of action potential generation 55 , and in the adult ON with sparse synapses, Harris and Attwell 10 estimated that only 0.1% and 0.4% of ATP is used for synaptic transmission and action potentials, respectively. It is therefore likely that the overall change in firing induced by natural visual stimulation consumes little energy, certainly less than in response to an electrical stimulation. This energy consumption change may not require any local blood flow regulation through NVC. The visual system is thus unique in that in the retina, it is also assumed that NVC does not occur in the layer containing the external segment of photoreceptors for an opposite reason: it consumes so much oxygen and glucose that energy is constantly delivered without NVC by the choroid vasculature 56 . Declarations Data availability Source data are provided with this paper. Code availability The codes used for the analyses are available from the corresponding author upon request. Acknowledgments: We thank Stephane Fouquet for image data segmentation in SYGLASS, Manon Omnes for the management of mice colonies and performing the surgeries for fUS, Emmanuelle Chaigneau for software support, and Melissa Glatigny for support with the experiments in rats. Financial support was provided by the Institut National de la Santé et de la Recherche Médicale (INSERM), the Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA), the Agence Nationale de la Recherche (CE14-0026-01), the Fondation Leducq Transatlantic Networks of Excellence program (16CVD05) and the IHU FOReSIGHT (IHU NVC intracranial). The 17.2T MRI system was supported through the Ile-de-France SESAME Large Equipment program. The synthesis of Oxyphor 2P was supported by the grant U24 EB028941 from the NIH USA (PI: Dr. Sergei A. Vinogradov). Author contributions: S.C. and L.C. designed the study. DESB, SC, DB and LC acquired BOLD fMRI data, which were analyzed by DESB and DB. A.A. and N.R. clarified and labeled the brains, which vascular structure was quantified by DESB. DESB developed the mouse preparation for two-photon imaging of the ON and acquired RBC velocity and blood oxygenation data. TV.E. synthetized Oxyphor 2P. SC and DESB interpreted two-photon data. All authors edited the manuscript. Competing Interests: The authors declare no competing interests. References Iadecola C (2017) The Neurovascular Unit Coming of Age: A Journey through Neurovascular Coupling in Health and Disease. Neuron 96:17–42 Aydin AK et al (2020) Transfer functions linking neural calcium to single voxel functional ultrasound signal. Nat Commun 11:2954 Rungta RL et al (2021) Diversity of neurovascular coupling dynamics along vascular arbors in layer II/III somatosensory cortex. Commun Biol 4:855 Schilling KG et al (2023) Whole-brain, gray, and white matter time-locked functional signal changes with simple tasks and model-free analysis. Proc. Natl. Acad. Sci. U. S. A 120, e2219666120 Wang H et al (2023) White matter BOLD signals at 7 Tesla reveal visual field maps in optic radiation and vertical occipital fasciculus. NeuroImage 269:119916 Schilling KG et al (2022) Anomalous and heterogeneous characteristics of the BOLD hemodynamic response function in white matter. Cereb Cortex Commun 3:tgac035 Waxman SG, Swadlow HA (1977) The conduction properties of axons in central white matter. Prog Neurobiol 8:297–324 Sturrock RR (1980) Myelination of the mouse corpus callosum. Neuropathol Appl Neurobiol 6:415–420 Markus HS, Joutel (2025) A. The pathogenesis of cerebral small vessel diseases and vascular cognitive impairment. Physiol Rev Harris JJ, Attwell D (2012) The energetics of CNS white matter. J Neurosci 32:356–371 Forrester J, Peters (1967) A. Nerve fibres in optic nerve of rat. Nature 214:245–247 Honjin R, Sakato S, Yamashita T (1977) Electron microscopy of the mouse optic nerve: a quantitative study of the total optic nerve fibers. Arch Histol Jpn 40:321–332 Ransom BR, Orkand RK (1996) Glial-neuronal interactions in non-synaptic areas of the brain: studies in the optic nerve. Trends Neurosci 19:352–358 Lev-Ram V, Grinvald A (1987) Activity-dependent calcium transients in central nervous system myelinated axons revealed by the calcium indicator Fura-2. Biophys J 52:571–576 Zhang CL, Wilson JA, Williams J, Chiu SY (2006) Action potentials induce uniform calcium influx in mammalian myelinated optic nerves. J Neurophysiol 96:695–709 Looser ZJ et al (2024) Oligodendrocyte-axon metabolic coupling is mediated by extracellular K(+) and maintains axonal health. Nat Neurosci 27:433–448 Asadollahi E et al (2024) Oligodendroglial fatty acid metabolism as a central nervous system energy reserve. Nat Neurosci 10:1934–1944 Restrepo A et al (2022) Axo-vascular coupling mediated by oligodendrocytes. BioRxiv 1–34. 10.1101/2022.06.16.495900 Schroder S et al (2020) Arousal Modulates Retinal Output Neuron 107:487–495 Boissonnet T, Tripodi M, Asari H (2023) Awake responses suggest inefficient dense coding in the mouse retina. Elife. 12 Boido D et al (2019) Mesoscopic and microscopic imaging of sensory responses in the same animal. Nat Commun 10:1110 Abe Y, Tsurugizawa T, Le BD, Ciobanu L (2019) Spatial contribution of hippocampal BOLD activation in high-resolution fMRI. Sci Rep 9:3152 De La Garza BH, Muir ER, Li G, Shih YY, Duong TQ (2011) Blood oxygenation level-dependent (BOLD) functional MRI of visual stimulation in the rat retina at 11.7 T. NMR Biomed 24:188–193 Duong TQ, Ngan SC, Ugurbil K, Kim SG (2002) Functional magnetic resonance imaging of the retina. Invest Ophthalmol Vis Sci 43:1176–1181 Rungta RL, Chaigneau E, Osmanski BF, Charpak S (2018) Vascular Compartmentalization of Functional Hyperemia from the Synapse to the Pia. Neuron 99:362–375 Kirst C et al (2020) Mapping the Fine-Scale Organization and Plasticity of the Brain Vasculature. Cell 180:780–795 Kleinfeld D, Mitra PP, Helmchen F, Denk W (1998) Fluctuations and stimulus-induced changes in blood flow observed in individual capillaries in layers 2 through 4 of rat neocortex. Proc Natl Acad Sci U S A 95:15741–15746 Chaigneau E, Oheim M, Audinat E, Charpak S (2003) Two-photon imaging of capillary blood flow in olfactory bulb glomeruli. Proc Natl Acad Sci U S A 100:13081–13086 Wang C, Ji N (2013) Characterization and improvement of three-dimensional imaging performance of GRIN-lens-based two-photon fluorescence endomicroscopes with adaptive optics. Opt Express 21:27142–27154 Piantadosi SC et al (2024) Holographic stimulation of opposing amygdala ensembles bidirectionally modulates valencespecific behavior via mutual inhibition. Neuron 112:593–610 Mace E et al (2011) Functional ultrasound imaging of the brain. Nat Methods 8:662–664 Esipova TV et al (2019) Oxyphor 2P: A High-Performance Probe for Deep-Tissue Longitudinal Oxygen Imaging. Cell Metabol 29:736–744 Sakadzic S et al (2010) Two-photon high-resolution measurement of partial pressure of oxygen in cerebral vasculature and tissue. Nat Methods 7:755–759 Lecoq J et al (2011) Simultaneous two-photon imaging of oxygen and blood flow in deep cerebral vessels. Nat Med 17:893–898 Sencan I et al (2020) Optical measurement of microvascular oxygenation and blood flow responses in awake mouse cortex during functional activation. J Cereb Blood Flow Metab 42:510–525 Parpaleix A, Houssen YG, Charpak S (2013) Imaging local neuronal activity by monitoring PO(2) transients in capillaries. Nat Med 19:241–246 Aydin AK, Verdier C, Chaigneau E, Charpak S (2022) The oxygen initial dip in the brain of anesthetized and awake mice. Proc. Natl. Acad. Sci. U. S. A 119, e2200205119 Masamoto K, Kanno I (2012) Anesthesia and the quantitative evaluation of neurovascular coupling. J Cereb Blood Flow Metab 32:1233–1247 Morisset C et al (2022) Retinal functional ultrasound imaging (rfUS) for assessing neurovascular alterations: a pilot study on a rat model of dementia. Sci Rep 12:19515 Kornfield TE, Newman EA (2014) Regulation of blood flow in the retinal trilaminar vascular network. J Neurosci 34:11504–11513 Biesecker KR et al (2016) Glial Cell Calcium Signaling Mediates Capillary Regulation of Blood Flow in the Retina. J Neurosci 36:9435–9445 Mishra A et al (2016) Astrocytes mediate neurovascular signaling to capillary pericytes but not to arterioles. Nat Neurosci 19:1619–1627 Peppiatt CM, Howarth C, Mobbs P, Attwell D (2006) Bidirectional control of CNS capillary diameter by pericytes. Nature 443:700–704 Gonzales AL et al (2020) Contractile pericytes determine the direction of blood flow at capillary junctions. Proc. Natl. Acad. Sci. U. S. A 117, 27022–27033 Mughal A, Nelson MT, Hill-Eubanks D (2023) The post-arteriole transitional zone: a specialized capillary region that regulates blood flow within the CNS microvasculature. J Physiol 601:889–901 Pfeiffer T, Li Y, Attwell D (2021) Diverse mechanisms regulating brain energy supply at the capillary level. Curr Opin Neurobiol 69:41–50 Alarcon-Martinez L et al (2020) Interpericyte tunnelling nanotubes regulate neurovascular coupling. Nature 585:91–95 Loo JH et al (2021) Loss of Caveolin-1 Impairs Light Flicker-Induced Neurovascular Coupling at the Optic Nerve Head. Front Neurosci 15:764898 Mace E et al (2018) Whole-Brain Functional Ultrasound Imaging Reveals Brain Modules for Visuomotor Integration. Neuron 100:1241–1251 Sanganahalli BG et al (2022) Thalamic activations in rat brain by fMRI during tactile (forepaw, whisker) and non-tactile (visual, olfactory) sensory stimulations. PLoS ONE 17:e0267916 Shanks JA et al (2016) Corticothalamic Axons Are Essential for Retinal Ganglion Cell Axon Targeting to the Mouse Dorsal Lateral Geniculate Nucleus. J Neurosci 36:5252–5263 Metea MR, Newman EA (2006) Glial cells dilate and constrict blood vessels: a mechanism of neurovascular coupling. J Neurosci 26:2862–2870 Funfschilling U et al (2012) Glycolytic oligodendrocytes maintain myelin and long-term axonal integrity. Nature 485:517–521 Gao Y et al (2025) Myelination selectively modulates BOLD signal in white matter. Res Sq Alle H, Roth A, Geiger JR (2009) Energy-efficient action potentials in hippocampal mossy fibers. Science 325:1405–1408 Shih YY et al (2013) Quantitative retinal and choroidal blood flow during light, dark adaptation and flicker light stimulation in rats using fluorescent microspheres. Curr Eye Res 38:292–298 Methods Mice All animal care and experimentations were performed in accordance with the INSERM Animal Care and Use Committee guidelines and approved by the ethical committee (Charles Darwin, comité national de réflexion éthique sur l’expérimentation animale n°5; protocol number #27135 2020091012114621). Mice were fed ad libitum and housed in a 12-hour light-dark cycle at 22°C and 50% humidity. A total of 17 adult mice were used in this study: 10 mice TPLSM and 7 mice for iDISCO. Both males and females, from 4 to 14 months of age were included. The following mouse lines were used and bred in our animal facility: C57/BL6J and Thy1-GCaMP6s (GP5.1). Rats All animal procedures were approved by the Comité d’Ethique en Expérimentation Animale, Commissariat à l’Energie Atomique et aux Énergies Alternatives, and by the Ministère de l’Education Nationale, de l’Enseignement Supérieur et de la Recherche (France) under reference #55527-2025051411223097 v2 and were conducted in strict accordance with the recommendations and guidelines of the European Union (Directive 2010/63/EU) and the French National Committee (Décret 2013–118). Rats were fed ad libitum and housed in a 12-hour light-dark cycle at 22°C and 50% humidity. A total of seven female Sprague Dawley from 2 to 6 months of age were included in this study. Surgery for GRIN lens implantation above the right ON Dexamethasone (6 mg/kg, s.c) was administered 24h and 1h before the surgery to prevent brain edema and reduce inflammation. Buprenorphine (0.3 mg/kg s.c.) was injected 2 hours before and 24h after surgery. Mice were anaesthetized with isoflurane (3-5% for induction and 1.5-2% for maintenance) and positioned in a stereotaxic frame (Model 1900 Stereotaxic Alignment System - Kopf Instruments) with ear bars. Body temperature was maintained at 36.5±0.5ºC using a feedback-controlled heating pad with a rectal probe. Lidocaine was injected below the skin (32 mg/kg s.c.). The head of the mouse was shaved with scissors and depilatory cream was applied from the neck to the eye level. The skin was disinfected, and a midline incision was made in the skin from the neck to the level of the eyes with a scalpel blade and the skin was maintained on the sides with clamps. Connective tissues over the skull were removed using a small scalpel blade. A mixture of 50% air and 50% oxygen was delivered through a nose cone to maintain blood oxygenation during the anesthesia. The skull was disinfected with betadine solution, cleaned with a sterile NaCl buffer, and prepared for the adhesion of light-curing cement using a primer solution. The mouse’s head was then aligned with the aid of a stereotaxic alignment indicator (Model 1905), positioning the indicator at bregma and lambda to ensure proper orientation. Alignment was confirmed in both the anteroposterior and coronal directions, after which the centering scope was positioned at bregma, and the coordinates set to zero. A stereotaxic drill equipped with a 0.7 mm burr was carefully lowered to create a hole in the skull for the GRIN lens insertion at the following coordinates from bregma: anterior–posterior: 1.4 mm; medial-lateral: 1.4 mm. A coronal rotation of 11° (counterclockwise) and a sagittal rotation of 10° (clockwise) were then performed. The burr was replaced by a 25G needle, which was inserted in the hole and gradually lowered in the direction of the right optic nerve to a depth of -4.7 mm while continuously applying cortical buffer to limit brain displacement. The needle was subsequently removed and replaced with a custom-designed holder containing the GRIN lens. The lens was carefully lowered to -4.9 mm and its upper part was fixed to the bone with a thin layer of UV-cured dental cement. A layer of primer solution was applied on the skull and the upper part of the GRIN lens was protected by a small metal tube (5 mm in diameter and 7 mm in height) secured on the skull with UV-cured dental cement. To prevent light contamination from the visual stimulation screen, a mixture of Unifast TRAD solvent/powder, colored with non-toxic black acrylic paint, was applied over the UV-cured dental cement. A horizontal titanium bar was then attached and sealed with dental cement on the posterior part of the skull, maintaining orthogonality between the head bar and the GRIN lens. Finally, surgical glue was applied to close the skin around the cemented area. In total, the surgical procedure for GRIN lens implantation took approximately 90 minutes to complete. Post-surgery, the animal was placed in a heating box until fully awake, provided with gel boost, and housed overnight in a warm cage. A post-operative scoring was performed for three days after the surgery. In 8 out of 10 mice, the accurate targeting of the GRIN lens was verified postmortem. Keeping the head fixed, we performed a craniotomy, slowly and carefully removed the left cortex and diencephalon to uncover the two ONs. This allowed to ensure that the chronic GRIN lens had been correctly positioned over the right ON. Sedation for two-photon imaging A recovery period of 2-3 weeks minimum was respected before imaging experiments were performed. During imaging sessions, mice were sedated with continuous perfusion of dexmedetomidine as follows: induction of anesthesia was done with 3% isoflurane and then decreased by 0.5% every 5 min down to 0% within 30 minutes, and dexmedetomidine was administered with a bolus (0.025 mg/kg s.c.) at the beginning of the session and a s.c. perfusion (0.1 mg/kg/h) during the entire session. Texas Red (70 kDa dextran, D1830, LifeTechnologies) was administered intravenously (i.v.) with a retro-orbital injection (left eye) prior to the decrease in isoflurane (under 3-2.5% isoflurane). Recordings started 20 minutes after the isoflurane cutoff. Mice were allowed to freely breathe air supplemented with oxygen (30% final concentration). Body temperature was controlled with a rectal probe and maintained at 36.5°C with a feedback-controlled heating pad. The animal was monitored throughout the imaging session using an infrared webcam (DCC3240N, Thorlabs). At the end of each experiment, the mouse was injected with atipamezole to antagonize the effects of medetomidine and accelerate recovery of physiological functions. Training for two-photon imaging in awake mice For TPLSM training, mice were head-fixed to a frame with a running wheel, and gradually habituated to the experimental setup, which included regular delivery of visual stimulation, for >10 days before experimental sessions began. Locomotion and movement were monitored with a velocity encoder connected to the running wheel and the infrared camera. Mice were briefly (1 hour prior to initiating experimental sessions. Only trials for which the mouse remained still throughout visual stimulation were analyzed. Visual stimulations for two-photon and fUS imaging experiments. For mice, visual stimuli were delivered via a 13-inch screen controlled by custom MATLAB software and placed 9 cm in front of the mouse’s eye. Mice received oxygenated air to a final concentration of 30% O 2 continuously through a nose cone. Visual stimulation consisted of 10s of drifting gratings (frequency 0.04 cpd, 50% contrast, intensity 76 ± 1 lux) with a contrast of 50% after a rest period varying from 10s to 30s. To avoid light pollution from the screen monitor onto the light collection path, the upper ring of a black circular and foldable curtain was slid along the microscope objective and the lower ring along a small metal tube protecting the upper part of the GRIN lens (Fig. 4a). Two-photon laser scanning microscopy and data analysis TPLSM imaging was performed with a custom-built microscope, previously described 34 and data were acquired with a customized LabVIEW software (National Instruments). Two-Photon excitation was obtained using a femtosecond laser, either from Coherent (Mira, 120-fs pulses) or SpectraPhysics (Insight, 70-fs pulses). Laser power was modulated with an acousto-optic modulator (AA Optoelectronic, MT110B50-A1.5-IR-Hk). Galvanometric mirrors (Cambridge Technology) were used to target the sample at the desired points. The excitation beam was focused onto the sample by using a 10X/0.3NA (Olympus) and a GRIN lens from GRINTECH (NEM-050-05-10-860-DM or NEM-250-05-10-860-DM). GCaMP6f and Texas Red were excited at 920 nm, OXYPHOR-2P at 960 nm, and the emission light was separated from the excitation light with a dichroic mirror (DXCR 875, Chroma Technology Corp [Chroma]). Emitted photons were than divided by a dichroic mirror (cutoff wavelength = 560 nm). The green channel was lowpass and bandpass filtered (E800, HQ 525/50 nm, Chroma Technology Corp) and light was collected onto a GaAsP (Hamamatsu) photomultiplier tube. The red channel had one lowpass filter (E800, Chroma), one bandpass filter (FF01 794/ 160, Semrock), and a red-sensitive photomultiplier tube (R6357, Hamamatsu). Post-processing analysis for the RBC velocities and PO2 measurements was done in Matlab (version R2018a, MathWorks) with a custom-made software. fUS imaging and data analysis CBV measurements were acquired as previously 21 using an ultrasound scanner (Iconeus One, Iconeus). Briefly, we used a linear ultrasound probe (128 elements, 15 MHz central frequency, Vermon) with an ultrasound sequence consisting of transmitting 11 different tilted plane waves (from −10° to 10° in 2° increments) with a 5500-Hz pulse repetition frequency (500 Hz frame rate of reconstructed images). Post-processing analysis for the sagittal sections was done in Matlab (version R2018a, MathWorks) with a custom-made software: the first 40 singular value decomposition (SVD) were removed and the power Doppler (PD) signal was further filtered with a Butterworth filter (fifth order). Data are shown as ΔPD/PD, with a baseline of 9 s (from 1 to 10 s). CBV flowing at low (10-30 Hz, 0.5-1.5 mm/s) and high (> 60 Hz, > 3 mm/s) axial velocities were separated by filtering as previously described 21 . For the coronal sections, post-processing analysis of CBV changes and identification of the dLGN, SC and visual areas were done using the ICOstudio software. All experiments were performed in sedated mice. BOLD fMRI and data analysis The animals were anesthetized initially with isoflurane and a subcutaneously administered bolus of 0.05 mg/kg medetomidine (Domitor, Pfizer, Karlsruhe, Germany), and were freely breathing an air–oxygen gas mixture (36 % O 2 ). For 4 rats, a small volume of Aflunox (VWR International, Rosny-sous-Bois, France) was injected in the right middle ear in order to decrease the susceptibility artifacts on the right ventral side of the brain. The rats were placed in the cradle and the head was immobilized using a bite bar and ear pins. The body temperature of the animals was maintained in the range of 36–37.5°C using a circulating hot water blanket. An optical fiber was placed in front of the left eye. The visual stimulation paradigm consisted of 6 s LED-on (flashing at 2 Hz) and 12 s LED-off and was delivered using a blue light emitting diode (LED) controlled by an Arduino One (ArduinoCham, Switzerland). All MRI acquisitions were performed on a 17.2 Tesla MRI system (Bruker BioSpin, Ettlingen, Germany) with a 30 mm TX/RX coil (Bruker BioSpin, Ettlingen, Germany). Anatomical images of the whole brain were acquired using a Rapid Acquisition with Relaxation Enhancement (RARE) pulse sequence. Functional BOLD acquisitions were performed using a gradient echo echo-planar imaging (GE-EPI). Two series of acquisitions with two different orientation were acquired. The common parameters for the two acquisitions were the following: TE = 11.5 ms, TR = 1000 ms, number of slices = 9, number of repetitions = 372. The spatial resolution varied depending on the orientation as follows: coronal: slice thickness 0.8 mm, in plane resolution 0.2 mm x 0.2 mm; axial: slice thickness 0.5 mm, in plane resolution 0.24 mm x 0.24 mm. A diffusion weighted acquisition was also acquired in the axial orientation (TE = 16 ms, TR = 1500 ms, slice thickness 0.5 mm, in plane resolution 0.2 mm x 0.2 mm, b value = 1000 s/mm 2 , direction = [0,0,1], NA =2) in order to precisely identify the optic nerve. For all acquisitions, shimming was performed on a slab encompassing the slices of interest using ‘mapshim’ (Bruker). Co-registration of DW and BOLD images allowed to measure functional responses exclusively in the voxels containing the ON, from the retina to the most ventral part of the OT. Signal averaging was performed over trials and over animals in order to improve the SNR. After averaging, the analysis was performed without any assumption on the timing and shape of responses. This approach avoided imposing assumptions about the HRF and improved sensitivity to small effects. It was similar to that used with two-photon acquisitions of blood flow and oxygenation, providing a more accurate characterization of subtle BOLD responses. Functional activation maps were however also computed using GLM analysis and a boxcar function representing the stimulation time and duration, shifted by 2 seconds. Using a very low statistical threshold p<0.01, allowed to detect some activated voxels in the ON initial portion, as with our primary analysis. Tissue Clearing and Immunolabeling Mice underwent intracardiac perfusion with a solution of 4% PFA (EMS) in PBS, then the brains were dissected and post-fixed in the same solution for 2h at room temperature. Mouse brain tissues, including the optic nerve and optic tract, were processed using the previously described iDISCO+ tissue clearing protocol (24). Samples were immunolabelled with primary antibodies against SM22 at 1/1000 th dilution from stock (Abcam; #ab14106), directly conjugated with AlexaFluor 555 (Invitrogen), and podocalyxin (Podxl)(1/300 th dilution from stock) (R&D systems; #MAB1556) and CD31 (PECAM-1)(1/1000 th diluation from stock) (R&D systems; #AF3628), both directly conjugated to AlexaFluor 647 (Invitrogen). The tissues were processed with iDISCO+ as previously described ( 24 ), with a 2 weeks incubation time at 37°c for primary antibodies (no secondary antibodies were used), and stored and imaged in DiBenzyl Ether (Sigma). Light-Sheet Imaging and Preprocessing The imaging of cleared samples was conducted using a light-sheet fluorescence microscope (Blaze, LaVision BioTec) at a voxel resolution of [1.62 × 1.62 × 2.5 µm], using the lightspeed acquisition mode. 4X NA0.35 objective was used for light collection, and a light sheet NA of 0.1 was used. Imaging channels included 488nm excitation for autofluorescence, 541nm for Sm22 and 639nm for CD31 and Podocalyxin. The 3 channels were acquired in the same imaging sequence to obtain a perfect alignment between them. The imaging sequence was set as z-drive -> filter wheel -> x-drive -> y-drive. The raw image tiles were imported into ClearMap for stitching 26 and channel alignment, using the CD31/Podocalyxin channel as the stitching reference, and its layout to stitch the other 2 channels. The volumetric datasets were processed and subsequently exported in multi-channel TIFF format for the purpose of subsequent analysis in Imaris (Oxford Instrument) and SyGlass (IstoVision). Vascular Segmentation and Quantification Image volumes were converted to a format compatible with syGlass using the syGlass Data Converter. Segmentation and analysis were performed using syGlass (IstoVision), allowing manual annotation within the volumetric data.The vascular networks were then segmented based on the expression of the markers CD31 and PODO31. SM22 expression was used to identify vessels containing smooth muscle cells. The segmentation of venous identity was based on the distribution of markers and the morphology of vessels (low expression of SM22, and large radii).The branching points, which are defined as discretevascular bifurcations, were manually annotated across the ON and OT in SyGlass, each point being assigned a single count. Statistics and onset analysis We computed some of TPLSM data in z score because of fluctuations in the baseline due to vasomotion 3 . Z score was calculated for each trial as follows: z score = (x – μ baseline ) / σ baseline with 10 s long baseline, with μ baseline the mean of the baseline, and σ baseline the standard deviation (SD) of the baseline. Trials from the same vessel were averaged (with a 0.1 s interpolation) for analysis. Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryInfoNCOMMS2559632A.pdf Supplementary Info Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9086876","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":604581759,"identity":"5fa70fda-d76c-4024-b9af-1be1f16553a0","order_by":0,"name":"David Esteban Suarez-Baquero","email":"","orcid":"https://orcid.org/0009-0008-6783-2326","institution":"Sorbonne Université, INSERM, CNRS, Institut de la Vision","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"Esteban","lastName":"Suarez-Baquero","suffix":""},{"id":604581760,"identity":"8d3b754d-46ec-4a6c-8a4c-82fa6d456155","order_by":1,"name":"Davide Boido","email":"","orcid":"https://orcid.org/0000-0002-1823-0555","institution":"Université Paris-Saclay, NeuroSpin CEA Saclay","correspondingAuthor":false,"prefix":"","firstName":"Davide","middleName":"","lastName":"Boido","suffix":""},{"id":604581761,"identity":"f8cc8c12-c245-4c5f-bfa5-35b6f1ffb453","order_by":2,"name":"Ahlem Assali","email":"","orcid":"https://orcid.org/0000-0002-5863-3682","institution":"Sorbonne Université, Institut du Cerveau et de la Moelle Epinière","correspondingAuthor":false,"prefix":"","firstName":"Ahlem","middleName":"","lastName":"Assali","suffix":""},{"id":604581762,"identity":"baf998f7-53f7-42e5-9edc-88543ecca0ca","order_by":3,"name":"Tatiana V. Esipova","email":"","orcid":"https://orcid.org/0000-0002-1172-419X","institution":"Department of Chemistry and Biochemistry, Loyola University","correspondingAuthor":false,"prefix":"","firstName":"Tatiana","middleName":"V.","lastName":"Esipova","suffix":""},{"id":604581763,"identity":"058cfda3-fa4c-4cc8-9418-8846712f480c","order_by":4,"name":"Luisa Ciobanu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1ElEQVRIiWNgGAWjYBACPgbGBhDNwyABoipABOMDvFrYULWcARHMBgS0wABIC2MbMVokkps//GCwk+GXbn72mHeenRx/AzPbB/xaEtskexiSeSTnHDM35t2WbCxxgJl5BiEtQI8c4DG4kWAmzbvtQGLDAf7DBByW2PzxD1hL+jdp3jkH6ucDbSGkpUEaYksO0JaGAwkGBLXwPGyTljEA+eVMGdA/yYYbDxPQws+e/vjjmwo7e37p9m0Sb2rs5OWON+PXAgEoEUGMhlEwCkbBKBgF+AEAcLU7+DijrGsAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-6932-6859","institution":"Université Paris-Saclay, NeuroSpin CEA Saclay, CNRS","correspondingAuthor":true,"prefix":"","firstName":"Luisa","middleName":"","lastName":"Ciobanu","suffix":""},{"id":604581764,"identity":"fee49af8-fd7e-40d2-a1cb-0edda93cad06","order_by":5,"name":"Serge Charpak","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIiWNgGAWjYBCDBBDB2FAhAeXbEK3lDExLGrFaGttgfDxa5Nt7H7/4wGCXxz/77MGPM+dZ5AFFnm5gSLiHU4vBmeNmljMYkoslzuUlS27cJlEMErnBkFCMW4tEGpsxD8OBxIYzPAaSD7dJJG4Aitxg/JGA22Hzn7EZ/wFqmX+Gx/jnwzkSifOBIkBbcGthuMHG/JgBqGXDGR4zyY0NEokNN9jwazE4k8bG2GOQnLjxDF+a5YxjQIcBRW4k4NEi336M+cOPCrvEeWd4D9/sqalLnN9+jO3GB3wOY2Bgk2AwANE8SGJ4NTAwMH+A0Dz4lY2CUTAKRsHIBQDx41eZf9W1dgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5516-1245","institution":"Sorbonne Université, INSERM, CNRS, Institut de la Vision","correspondingAuthor":true,"prefix":"","firstName":"Serge","middleName":"","lastName":"Charpak","suffix":""}],"badges":[],"createdAt":"2026-03-10 18:03:15","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-9086876/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9086876/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104582745,"identity":"fd5bcb7c-7699-42ea-bd2a-3846b8fadbce","added_by":"auto","created_at":"2026-03-13 15:14:20","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":250034,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVisual stimulation does not evoke a BOLD fMRI signal in the myelinated part of the ON.\u003c/strong\u003e \u003cstrong\u003e(a)\u003c/strong\u003e Co-registered anatomical (left column), diffusion weighted (DW, middle column) and functional BOLD horizontal acquisitions from the retina down to the optic chiasm. Functional ROIs (groups of voxels) were manually selected from the DW-SE images: ROI 1, the retina; ROI 2, the ON head; ROI 3, the initial oblique segment of the intracranial ON; ROI 4, the ON caudal segment, just preceding the chiasm; ROI 5, the chiasm and the most ventral part of the OT. Right column, with the BOLD selected voxels. On the bottom right, the enlargement shows ROI 4 and ROI 5 selected voxels containing the ON. \u003cstrong\u003e(b)\u003c/strong\u003e Average responses from the 5 ROIs to visual stimulation (in gray, 20 stimulations/session, 7-10 sessions per animal). \u003cstrong\u003e(a,b) \u003c/strong\u003eData are represented as mean ± SD (shading). \u003cstrong\u003e(c)\u003c/strong\u003e Box plots of the number of voxels/ROI for \u003cem\u003en=\u003c/em\u003e7 rats.\u003cstrong\u003e \u003c/strong\u003eIn the box plot, the center line represents the median, the box limits represent the 25th and 75th percentiles, and the whiskers represent the minimum and maximum values. \u003cstrong\u003e(d)\u003c/strong\u003e Average responses for \u003cem\u003en=\u003c/em\u003e7 rats. BOLD responses were only detected in the retina and the unmyelinated ON head. Data are represented as mean ± S.E.M. Source data are provided as a Source Data file.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/0ce13184c32fe15f80e9f996.jpg"},{"id":104582750,"identity":"3579023d-8fa8-4942-9eed-ce95846326f4","added_by":"auto","created_at":"2026-03-13 15:14:20","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":146903,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVisual stimulation induces BOLD fMRI responses in the dLGN and vLGN but not in the ventral OT\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e. \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e(a)\u003c/strong\u003e Co-registered anatomical T\u003csub\u003e2\u003c/sub\u003e weighted (left column) and functional BOLD acquisitions from the chiasm back to the SC. 4 ROIs (4 groups of voxels) were manually selected from the anatomical images where the chiasm and the ventral OT appear as well-defined dark shadows. Signal was only measured on the side on which Aflunox was injected into the middle ear to reduce susceptibility artefacts (white arrow). \u003cstrong\u003e(b)\u003c/strong\u003e Visual stimulation did not evoke a BOLD response in the chiasm and ventral OT. \u003cstrong\u003e(c)\u003c/strong\u003e In the same animal, 5 ROIs (groups of voxels) were selected using the Waxholm Space Atlas, co-registered with the anatomical acquisitions: ROI 1, the superior colliculus; ROI 2, the dorsal lateral geniculate nucleus; ROI 3, the ventral lateral geniculate nucleus; ROI 4, the dorsal OT; ROI 5, the ventral OT. \u003cstrong\u003e(d)\u003c/strong\u003e Clear BOLD responses to visual stimulation were observed in the dLGN, the vLGN and the SC. No response was detected in the ventral OT. \u003cstrong\u003e(b,d)\u003c/strong\u003e Data are represented as mean ± SD (shading). \u003cstrong\u003e(e)\u003c/strong\u003e Box plots of the number of voxels/ROI manually selected for the OT/chiasm as in \u003cstrong\u003e(a,b)\u003c/strong\u003e or selected automatically from the atlas as in \u003cstrong\u003e(c,d)\u003c/strong\u003e.\u003cstrong\u003e \u003c/strong\u003eIn the box plots, the center line represents the median, the box limits represent the 25th and 75th percentiles, and the whiskers represent the minimum and maximum values. \u003cstrong\u003e(f)\u003c/strong\u003e BOLD responses were detected in the gray matter (dLGN, vLGN or SC) but not in the ventral OT. Note that a delayed signal was detected in the dorsal OT trace, due to a few voxels in the upper part of the dorsal OT of a single animal. (n= 4 rats injected with Aflunox for ROIs: chiasm \u0026amp; OT (manual selection), OT dorsal, OT ventral and vLGN; n=10 rats for ROIs dLGN and SC). Data are represented as mean ± S.E.M. Source data are provided as a Source Data file.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/f1eed087b0715c6d2c68096a.jpg"},{"id":104582749,"identity":"f21d2d36-235e-443f-a2e0-ac94e9190d3b","added_by":"auto","created_at":"2026-03-13 15:14:20","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":177572,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVascular organization in the cleared ON and OT.\u003c/strong\u003e \u003cstrong\u003e(a)\u003c/strong\u003e Cleared chiasm and ONs, immunolabelled for CD31 and Podocalyxin (green) and Sm22 (red). A large vein (in green) runs along the surface of the right and left ON (arrows). A major artery (in red) enters the ON midway between the chiasm and the eye (arrowhead). \u003cstrong\u003e(b)\u003c/strong\u003e A cleared whole brain showing the ventral OT in grey (segmentation). \u003cstrong\u003e(c)\u003c/strong\u003e Enlargement of the left ON in \u003cstrong\u003e(a)\u003c/strong\u003e to show the penetration (arrowhead) of the arteriole (in red) in the nerve and the arteriole branches before and after penetration. \u003cstrong\u003e(d)\u003c/strong\u003e Isolation of the OT in \u003cstrong\u003e(b)\u003c/strong\u003e. Left, the arterioles (in red when outside) are shown in white when crossing the tract (grey). Middle, segments of veins (in cyan) crossing the tract (grey) are similarly shown in white. Right, all large vessels crossing the tract (arteries in pink, veins in green). \u003cstrong\u003e(e)\u003c/strong\u003e Further magnification of the ON (top) and OT (bottom) to show the density of capillaries (in green) in both parts of the nerve. \u003cstrong\u003e(f)\u003c/strong\u003eNumber of branch points in the ON and OT. For the ON, \u003cem\u003en\u003c/em\u003e=5 optic nerves from 3 mice were analyzed (both nerves from two mice and one nerve from a third mouse) with similar results. For the OT, \u003cem\u003en\u003c/em\u003e=4 hemi-brains from 4 independent mice were analyzed. The black dot indicates the value reported by Kirst et al. (2021). Data are represented as mean ± SD. \u003cstrong\u003e(g)\u003c/strong\u003e Top, a 100 µm-thick brain slice showing the left ventral OT. Bottom, the region of interest magnified. The density of labelled capillaries (in white) in the OT (arrowhead) is similar than in the neighboring amygdala (arrow). Source data are provided as a Source Data file.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/88ba799765c29cc7d2156756.jpg"},{"id":104582751,"identity":"0ed54019-be50-420d-80ba-6b29bcae2a51","added_by":"auto","created_at":"2026-03-13 15:14:20","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":120982,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTwo-photon imaging of the ON through a long GRIN lens.\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e(a) \u003c/strong\u003eSchematics of the GRIN lens implantation. The upper part is protected by a small metal tube sealed on the skull and a curtain isolates the light path from the screen light. \u003cstrong\u003e(b)\u003c/strong\u003e Left, moving up and down the objective allows to scan several imaging (sample) planes in the ON. Right, a series of images at several depths with vessels labelled with Texas Red (Injected I.V.). Images are representative of the vascular distribution observed in the 10 independent mice used for the results shown in Figure 5. \u0026nbsp;Scale bar (90 µm) is indicated for each depth panel. \u0026nbsp;\u003cstrong\u003e(c)\u003c/strong\u003e Power Doppler activation map acquired with functional ultrasound imaging (high-pass filter selecting the CBV flowing with an axial velocity \u0026gt; 2.5 mm/s). Visual stimulation (10 s of drifting gratings, 50% contrast, 75 lumens) increased CBV in the dLGN and visual cortex contralateral to the ON imaged through the GRIN lens, validating the GRIN lens innocuity. The 5-month-old sedated mouse was imaged through the skull (average trace superimposed on 4 trials). \u003cstrong\u003e(d)\u003c/strong\u003e An ON capillary imaged with two-photon microscopy through the GRIN lens. Left, A line was used in the linescan acquisition mode to monitor RBC velocity, which increased (right panel) in response to a brief 5% isoflurane inhalation (average trace superimposed on 3 trials). All data traces are represented as mean ± SD (shading). Source data are provided as a Source Data file.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/1aac25241fdefe2791ce5766.jpg"},{"id":104582747,"identity":"68a979f2-ce10-43ee-988f-6b6f24320e38","added_by":"auto","created_at":"2026-03-13 15:14:20","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":211402,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVisual stimulation does not induce a change in blood velocity or in oxygenation in the myelinated part of the ON\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e. \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e(a)\u003c/strong\u003e Left, an image showing a capillary labeled with Texas Red in the ON of a sedated mouse. A linescan acquisition was used to monitor red blood cell (RBC) velocity. Right, 10 s visual stimulation (grey shadow) had no effect on RBC velocity (\u003cem\u003en\u003c/em\u003e=7 stimulations, mean trace 0.14 ± 0.01 mm/s). \u003cstrong\u003e(b)\u003c/strong\u003e Left, mean response for 19 vessels (n=6 sedated mice), expressed as percentage change and as z-score. Right, box plots showing the distribution of resting RBC velocity in true capillaries (\u0026lt;5 µm) and larger vessels (\u0026gt;5 µm). The lack of a response to visual stimulation was independent of resting RBC velocity. \u003cstrong\u003e(c)\u003c/strong\u003e Left, a capillary labeled with fluorescein dextran and containing the oxygen sensor OXYPHOR-2P. Right, visual stimulation had no effect on resting pO\u003csub\u003e2\u003c/sub\u003e measured from the phosphorescence lifetime of OXYPHOR-2P (\u003cem\u003en\u003c/em\u003e=6 stimulations, mean trace 47.8±4.5 mmHg). \u003cstrong\u003e(d)\u003c/strong\u003e Left, mean responses for 7 vessels (\u003cem\u003en\u003c/em\u003e=3 sedated mice), expressed as percentage change and as z-score. Right, box plots showing the distribution of resting pO\u003csub\u003e2\u003c/sub\u003e and the absence of response to visual stimulation. \u003cstrong\u003e(e)\u003c/strong\u003e A capillary in the ON of an awake mouse. Visual stimulation did not affect RBC velocity (\u003cem\u003en\u003c/em\u003e=6 stimulations, mean trace 0.8 ± 0.1 mm/s). \u003cstrong\u003e(f)\u003c/strong\u003e Left, mean RBC velocity responses for 28 vessels (\u003cem\u003en\u003c/em\u003e=9 awake mice), reported as percentage change and z-score. Right, box plots showing the distribution of resting RBC velocity in true capillaries (\u0026lt;5 µm) and larger vessels. The lack of a response to visual stimulation was independent of resting RBC velocity. \u003cstrong\u003e(g)\u003c/strong\u003e Left, a large vessel in which visual stimulation did not affect resting pO\u003csub\u003e2\u003c/sub\u003e (\u003cem\u003en\u003c/em\u003e=9 stimulations, mean trace 57.4±5.1 mmHg). \u003cstrong\u003e(h)\u003c/strong\u003e Left, mean pO\u003csub\u003e2\u003c/sub\u003e responses for 31 vessels (\u003cem\u003en\u003c/em\u003e=7 awake mice), as percentage change and as z-score. Right, box plots showing the distribution of resting pO\u003csub\u003e2\u003c/sub\u003e. The absence of a response to visual stimulation was independent of vessel size. All data traces are represented as mean ± SD (shading).\u003c/p\u003e\n\u003cp\u003eIn all box plots \u003cstrong\u003e(b, d, f, h)\u003c/strong\u003e, the center line represents the median, the box limits represent the upper and lower quartiles (25th and 75th percentiles), and the whiskers extend to the minimum and maximum values. All individual data points are overlaid; red crosses indicate values exceeding 2 standard deviations (SD) from the mean. All data traces are represented as mean ± SD (shading). Source data are provided as a Source Data file.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/23a91754d364b05ace20e25b.jpg"},{"id":104784631,"identity":"6db4bb9a-b282-489a-aee0-f889fb53ffec","added_by":"auto","created_at":"2026-03-17 08:08:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1877394,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/d699c3ad-8b67-4484-908b-036602472edf.pdf"},{"id":104782481,"identity":"b4b967ac-01b4-4fc0-9763-01c639dd836f","added_by":"auto","created_at":"2026-03-17 07:57:23","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2226241,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Info\u003c/p\u003e","description":"","filename":"SupplementaryInfoNCOMMS2559632A.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9086876/v1/62fa19d9fce019d802ccf2ea.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eAction potential propagation in the rodent myelinated optic nerve does not trigger neurovascular coupling\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIn brain gray matter, neuronal activation triggers an increase in blood flow (known as functional hyperemia) via neurovascular coupling (NVC), an ensemble of processes that involves the cooperation of neurons, astrocytes, smooth muscle cells, pericytes, and endothelial cells. The mechanisms and the timing by which these cell types interact and contribute to coupling are still debated\u0026nbsp;\u003csup\u003e1\u003c/sup\u003e but there is a consensus that, in awake rodents, functional hyperemia is characterized by a hemodynamic response function (HRF) peaking at about ~1-2 s \u003csup\u003e2, 3\u003c/sup\u003e and is tightly correlated with transmitter release and postsynaptic activation. Little is known about NVC in white matter, but some recent blood-oxygen-level-dependent functional MRI (BOLD-fMRI) studies in humans have reported task-evoked responses \u003csup\u003e4, 5\u003c/sup\u003e: the BOLD signals measured were smaller than signals typically measured in gray matter, characterized by an HRF with different features \u003csup\u003e6\u003c/sup\u003e, but still occurred with a timing compatible with standard NVC, i.e., within a few seconds from stimulation onset. Since most white matter tracts contain both myelinated and unmyelinated axons, e.g., 30-40 % of mouse corpus callosum axons are unmyelinated \u003csup\u003e7, 8\u003c/sup\u003e, it remains unclear as to whether the mechanisms underlying NVC are similar for both axon types. Understanding the functional vascular dynamics and metabolism of white matter is an important issue, given widespread white matter lesions and lacunar infarcts are two key signs of cerebral small vessel diseases \u003csup\u003e9\u003c/sup\u003e. Whether NVC is triggered by local axonal release of glutamate and some feed-forward mechanisms, or conversely, is principally regulated by the energy demand resulting from action potential propagation remains an open question. Theoretical calculations indicate that the energy budget of firing is much lower than that of synaptic transmission \u003csup\u003e10\u003c/sup\u003e, raising the hypothesis that, at least in myelinated nerves characterized by their paucity of synapses, NVC may not be required for action potential propagation.\u003c/p\u003e\n\u003cp\u003eTo address these questions, the rodent optic nerve provides a unique model of white matter that, in contrast to the corpus callosum, contains only myelinated axons \u003csup\u003e11, 12\u003c/sup\u003e, oligodendrocytes and their precursor cells, vessels of all types, astrocytes and few specific axo-myelinic synapses \u003csup\u003e13\u003c/sup\u003e. Apart from the unmyelinated nerve head, which can be functionally observed through the animal’s eye lenses, the functional properties of the extracerebral myelinated part of the optic nerve - whether electrical, metabolic or vascular- have only been studied in vitro. During propagation of the compound action potential, there is a tight coupling between axons and oligodendrocytes. This process involves calcium influxes in both cell types and a complex cascade of events, including the release of glutamate, K+ ions, lactate, and the activation of glycolysis and lipid metabolism in oligodendrocytes \u003csup\u003e14-17\u003c/sup\u003e. With regard to NVC, a specific form of axo-vascular coupling was recently reported, which did not fulfill the properties of NVC, as it was only marginally measurable upon strong synchronous electrical stimulation and occurred with dynamics of minutes rather than seconds \u003csup\u003e18\u003c/sup\u003e. Moreover, the in vitro experimental conditions did not allow for the ability to: 1) predict what may occur during natural light stimulation, since light and arousal change the balance of the ganglion cell type activated\u003csup\u003e19, 20\u003c/sup\u003e with an unknown effect on the nerve metabolism, and, 2) correctly assess NVC, which requires vessels with the adequate pressure tone. Therefore, in order to characterize NVC in the whole optic nerve, a non-invasive multiscale functional imaging approach in vivo was used, which allows for measuring signals related to blood flow and blood oxygenation in response to light, with high spatio-temporal resolution. Our results reveal that while visual stimulation triggers robust NVC in the retina, the unmyelinated optic nerve head and at the level of the nerve synaptic terminals, it does not elicit any measurable hemodynamic or oxygenation changes in the myelinated portions of the optic nerve and optic tract. These findings demonstrate that action potential propagation in activated myelinated axons does not trigger NVC.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003efMRI BOLD responses along the entire optic nerve\u003c/h2\u003e \u003cp\u003eIn order to non-invasively investigate NVC along the entire length of the optic nerve, we first used BOLD fMRI and measured functional hyperemia from the retina to the nerve terminals in response to visual stimulation. Such a mesoscopic approach is non-invasive, and the strong magnetic field (17.2T) and gradients (1 T/m) used \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e ensured acquisition of images with high sensitivity and spatial resolution. Note that ganglion cell axons are fully myelinated in both the extracerebral part of the optic nerve (ON) and in the intracerebral part, or optic tract (OT). Data acquisition was performed in sedated rats, rather than mice, to maximize the number of voxels containing the optic nerve and thus improve the signal-to-noise ratio of the BOLD signal. At the beginning of each session, multi-slice gradient echo-planar imaging (GRE-EPI) functional BOLD acquisitions were performed to verify that visual stimulation evoked functional responses in the dorsal lateral geniculate nucleus (dLGN) or in the superior colliculus (SC). A visual flickering light paradigm was used, which consisted of 6 s LED \u0026ldquo;ON\u0026rdquo; (blue light flashing at 2 Hz) and 12 s LED \u0026ldquo;OFF\u0026rdquo; (repeated 20 times per session). To precisely target the ON, we first used diffusion weighted spin echo (DW-SE) acquisitions with 6 horizontal slices, positioned from the retina and the ON head down to the chiasm. Anatomical T\u003csub\u003e2\u003c/sub\u003e weighted images were then acquired, followed by BOLD acquisitions, during 7 consecutive sessions. Post hoc co-registration of DW-SE and T\u003csub\u003e2\u003c/sub\u003e weighted images allowed for measurement of functional responses located exclusively in the voxels containing the ON, from the retina to the most ventral part of the OT. Averaged trials improved the image SNR and the analysis was then performed without any assumptions about the timing and shape of the responses (4). Similar to what has been shown previously \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, visual stimulation generated a significant BOLD signal at the level of the retina and the ON head (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), a part of the nerve which is unmyelinated. In contrast, no signal was detected in the intracranial ON, the chiasm or the most ventral part of the optic tract (OT). In a single case, a minor increase in signal was detected rostrally in the initial segment of the ON, the total length of which is estimated to be approximately 8\u0026ndash;9 mm, which can be attributed to a venous signal draining the activated retina and ON head or a vascular signal backpropagating along the feeding arteriole \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. This signal also appeared in some functional activation maps (Supplementary Figs.\u0026nbsp;2 and 3) computed with GLM analysis, using a boxcar function representing the stimulation time and duration, shifted by 2 seconds, and a very low statistical threshold (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Even with such threshold, NVC was not detected in the main part on the ON and the chiasm.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eNext, to further investigate NVC in the OT, BOLD responses were acquired in 7 coronal slices (800 \u0026micro;m thick) from the optic chiasm back to the superior colliculus. Anatomical, T\u003csub\u003e2\u003c/sub\u003e weighted acquisitions were performed allowing for determination of voxels containing the OT, which appeared as a well delimited dark shadow in the first 2\u0026ndash;3 slices caudal to the optic chiasm. In these slices, no BOLD response was detected in the ventral OT (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), a part of the tract that is dense in axons and can be assumed to contain exclusively ganglion cell axons. In the caudal slices (\u0026gt;\u0026thinsp;4th slice) where the OT could not be sorted out anatomically from the T\u003csub\u003e2\u003c/sub\u003e weighed images, BOLD signals were detected in the dLGN and the vLGN. Functional activation maps similarly showed that BOLD signals were not detectable in the ventral OT (Supplementary Fig.\u0026nbsp;3). Taken together, BOLD fMRI data suggest that in response to visual stimulation, action potential propagation along the entire myelinated part of the optic nerve does not require NVC. To confirm this finding, additional studies were performed to exclude the possibility that the absence of a BOLD response was due either to the unlikely possibility that part of the nerve lacks a major vascular structure necessary to support NVC, or alternatively, due to insufficient signal-to-noise ratio.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eON and OT vascular structure\u003c/h2\u003e \u003cp\u003eWe used 3D light sheet microscopy and tissue clearing to reconstruct and compare the vascular architecture of the extracerebral and intracerebral portions of the nerve, by immunolabelling whole adult mouse brains for CD31, Podocalyxin and Sm22 with iDISCO+, as in \u003csup\u003e26\u003c/sup\u003e. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea,b shows that a large arteriole enters the ON midway between the chiasm and the eye, sending small collaterals in the ON and runs rostrally to the ON head. In contrast, a large vein runs on the side of the ON, from the eye to the chiasm. We suggest that this vein underlies the BOLD fMRI signal that can occasionally be observed rostrally in the initial segment of the ON. A moderate density of capillaries, quantified by the number of branching points, was observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee, f), in line with Restrepo et al. \u003csup\u003e18\u003c/sup\u003e, who additionally reported a pericyte density almost similar to that of the hippocampus or visual cortex. The ventral part of the OT (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec, d) is traversed by a few arterioles (fed by the thalamic artery) and venules, and also has a moderately dense capillary network, as shown in the ON. Quantitative analysis also revealed (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eg) that the vessel density of the ventral OT is comparable to that of the neighbouring gray matter region in the cortical sub-plate, the postero-dorsal part of the medial amygdala nucleus, but much less than the highly vascularized lateral thalamus \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. These results indicate that the strength of NVC is unrelated to capillary density and that the ON and OT vascular structures do not explain the lack of NVC in response to a natural visual stimulation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eTwo-photon imaging of NVC in the ON\u003c/h3\u003e\n\u003cp\u003eIn contrast to BOLD fMRI, two-photon microscopy is an approach that can be used to ascertain that NVC does not occur in the ON: in a small capillary, it detects single red blood cells (RBCs) and consequently, RBC velocity or flow changes can be monitored with a 100 ms temporal resolution \u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. We thus developed a preparation in which a long gradient-index lens (GRIN lens, length: 9 mm, diameter: 0.5 mm) was chronically implanted over the caudal ON in head-fixed mice, sedated or awake. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea illustrates the experimental setup using the GRIN lens, which was fixed to the cranium with its lower end placed at the contact of the dura over the right optic nerve, about 1.5 mm rostral to the optic chiasm. This approach left the ON intact and two-photon imaging was performed at least 3 weeks after surgery, allowing for full recovery.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDue to the working distance of the GRIN lens (250 \u0026micro;m or 50 \u0026micro;m), the imaging plane (or sample plane) was systematically located within the ON. By moving the objective focus up and down, we could change the imaging plane in z and image vessels labelled with Texas Red, from the center to the surface of the ON (Supplementary Fig.\u0026nbsp;1a). Note that this occurred at the expense of spatial resolution, as the two-photon point spread function increased with distance from the imaging (sample) plane (Supplementary Fig.\u0026nbsp;1b and \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e) and that the pixel size had to be corrected according to the z plane (Supplementary Fig.\u0026nbsp;1c and \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e). Such corrections were required to measure the real value of RBC velocity. To ensure that GRIN lens implantation did not caused damage that could alter the ON, i.e., local action potential propagation, we used functional ultrasound imaging, a mesoscopic approach that can be used to record changes in the cerebral blood volume (CBV) fraction flowing along the axial axis \u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, and investigated whether vascular responses to visual stimulation could be observed downstream of the ON, i.e., in the dLGN or visual cortex of mice implanted with phantom lenses (0.5 mm metal round bar) or standard GRIN lenses (n\u0026thinsp;=\u0026thinsp;3). Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec shows that visual stimulation triggered a CBV increase in the contralateral dLGN and visual cortex, demonstrating that the ON was functional. Finally, to control that we were able to measure RBC velocity changes, we verified that in sedated mice, a brief inhalation of isoflurane\u0026ndash;which increases blood flow\u0026ndash; (5 s, 5%, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed) efficiently triggered an increased in RBC velocity in ON capillaries imaged with the implanted GRIN lens. Two-photon measurements in small capillaries (diameter 2\u0026ndash;5 \u0026micro;m) revealed that the mean resting velocity of RBCs was 0.46\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05 mm/s (STD 0.20, n\u0026thinsp;=\u0026thinsp;19 capillaries) ranging from 0.16 mm/s to 0.86 mm/s (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea, b). Visual stimulation did not elicit a change in RBC velocity in any of the vessels in the ON. We then measured resting partial pressure of oxygen (pO\u003csub\u003e2\u003c/sub\u003e) and pO\u003csub\u003e2\u003c/sub\u003e changes in ON capillaries and larger vessels using two-photon phosphorescence lifetime microscopy using the oxygen sensor Oxyphor 2P \u003csup\u003e32\u003c/sup\u003e. In the gray matter, hyperemia has been shown to always be accompanied by an increase in pO\u003csub\u003e2\u003c/sub\u003e in individual capillaries \u003csup\u003e\u003cspan additionalcitationids=\"CR34\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, which results in an oxygenation rise in the surrounding tissue \u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. In the ON (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec, d), a mean resting pO\u003csub\u003e2\u003c/sub\u003e of 59.8\u0026thinsp;\u0026plusmn;\u0026thinsp;12.8 mmHg was measured (ranging from 46.5 mmHg to 79.4 mmHg). As with blood flow, visual stimulation did not evoke any changes in capillary pO\u003csub\u003e2\u003c/sub\u003e demonstrating that during sedation, propagation of action potentials in the ON does not trigger NVC. Because anesthesia and sedation affect both neuronal activity and neurovascular coupling mechanisms \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, we then tested whether NVC could be detected in the awake brain state. To do this, mice were habituated to head fixation and visual stimulation for 10\u0026ndash;14 days, after which repeated RBC velocity and pO\u003csub\u003e2\u003c/sub\u003e measurements in ON capillaries were taken in awake mice, including some which were previously tested during sedation. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee, g shows the resting values of RBC velocity and pO\u003csub\u003e2\u003c/sub\u003e in small capillaries of awake mice. As during sedation, visual stimulation did not induce any functional hyperemia or oxygenation changes, from 3 weeks to 8 months post-surgery. Note that NVC was not detectable whether vessels were considered independent (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb-h) or animal-dependent (Supplementary Fig.\u0026nbsp;4, mixed-effect model). Overall, these findings validate BOLD fMRI observations and demonstrate that in response to a visual stimulation, propagation of action potentials in the extracerebral or intracerebral axonal tract depends on a specific energy budget \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e that does not require NVC.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eUsing two complementary functional imaging approaches, our study demonstrates that in rodents and upon light stimulation, action potentials can propagate in the ON and OT without triggering NVC. This absence of NVC appears to be restricted to the myelinated portion of the ganglion cell axonal tract. In the rodent retina, which contains a complex unmyelinated neuronal network that includes ganglion cell dendrites, somata and axons, light generates strong vascular responses that have been studied in vivo with BOLD fMRI \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, fUS \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, and with a microscopic resolution with confocal and two-photon microscopy. In such studies, NVC has been shown to occur diversely in the three retina vascular layers \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e, and with an involvement of astrocyte calcium that differs at the capillary and the arteriole level \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. Concerning pericytes, since the initial demonstration that they can control capillary diameter \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e, most investigations have been performed in vitro (for reviews, see \u003csup\u003e\u003cspan additionalcitationids=\"CR45\" citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e), with an in vivo study proposing that interpericyte tunneling nanotubes coordinate light-evoked responses between adjacent capillaries \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. At the level of the ON head, ganglion cell axons are still unmyelinated, but light stimulation triggers vascular responses that are altered in a rat model of Alzheimer\u0026rsquo;s disease \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e and in caveolin-1 knockout mice \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. In the dLGN where the nerve axons end, i.e., in the gray matter, NVC occurs \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e, but the differential weight of axon terminal firing and glutamate release, versus postsynaptic cell depolarization, in triggering of coupling has not yet been investigated. All this is in accordance with our BOLD fMRI observations which show that under the same experimental conditions, visual stimulation produces strong BOLD responses in the retina, the ON head and in the nerve targets, the dLGN, vLGN and SC. In contrast, no BOLD fMRI signals were detected in the ON, the chiasm nor the rostro-ventral OT, anatomically identified either by DW-SE (horizontal acquisitions) or T\u003csub\u003e2\u003c/sub\u003e weighted vertical acquisitions. Although we recognize that it is difficult to draw a definitive conclusion from a lack of BOLD responses, we believe that the absence of NVC demonstrated with two-photon microscopy in the ON can be extended to the ventral OT. What occurs at the level of the caudal (i.e. dorsal) part of the OT remains technically unclear because we estimate that at the level of the dLGN and vLGN, it is impossible to distinguish the thin dorso-lateral layer containing OT fibers from the neighboring nuclei per se. Indeed, due to the low rostro-caudal spatial resolution of BOLD fMRI, the voxels forming this layer most certainly include thalamic neurons in addition to OT axons \u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e as well as other types of axons crossing the OT. Therefore, the caudal OT cannot be used as a model to assess a NVC that would be attributed exclusively to myelinated axons, although it is likely that it shares the same properties as the ventral OT.\u003c/p\u003e \u003cp\u003eOur study also shows that the ON, and most probably other cranial nerves, can be functionally imaged in anesthetized or awake mice using a chronically-implanted GRIN lens. Moving the imaging plane in z, away from the working distance of the GRIN lens, allowed for scanning through the thickness of the ON. The displacement in z changes the imaging point spread function, i.e., the spatial resolution, and reduces the field of view, however, we showed that with a simple pixel size correction, GRIN lens imaging still allows for accurate RBC velocity measurements in small capillaries. This allowed us to demonstrate, by monitoring RBC velocity and pO\u003csub\u003e2\u003c/sub\u003e, that the ON transmits visual information without changes in blood flow and oxygenation. This observation may seem at odds with a previous in vitro study \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, i.e. without physiological blood pressure tone, reporting that synchronous electrical stimulation of ON axons elicits a very small, slow, and delayed dilation of large vessels, a response that was proposed to be a form of NVC. The dynamics of the response reported, however, was too slow to account for NVC, which occurs within a few seconds of light stimulation in both the retina and the ON head, whether in vitro or in vivo \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. It is important to underline an additional difference between the in vitro and in vivo ON models with respect to stimulation. In vitro, ON electrical stimulation activates axons synchronously, and from a presumably very low basal activity and metabolism. In vivo, there is a constant dark current in photoreceptors and a constant activity in various types of \u0026ldquo;OFF\u0026rdquo; ganglion cells. As a result, stimulation with flickering or drifting gratings changes the type and weight of activated ganglion cells \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e making it difficult to estimate the change in overall metabolic cost associated with a visual stimulation.\u003c/p\u003e \u003cp\u003eOur two-photon data indicate that action potential propagation in the myelinated ON during light stimulation is not accompanied by any measurable NVC, despite its standard vascular organization. Could the absence of NVC result from our preparations or technical limitations? Arousal did not reveal NVC, even though ON responses in mice exhibit faster kinetics, a larger dynamic range, and higher firing activity in the awake state \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. This supports the idea that the ON relies on a specific metabolism in which oligodendrocytes play a major role \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e, independently of brain state. The general scheme is that axonal firing increases extracellular potassium, which in turn triggers calcium influx and glycolysis in oligodendrocytes. This leads to lactate production and release, which is then taken up by axons to fuel ATP production. Our data suggest that this metabolic loop does not require NVC, whether in the myelinated ON of sedated or awake mice and rats. For both BOLD fMRI and two-photon experiments, it is impossible to rule out a minute vascular response that could be extracted from noise only after ~\u0026thinsp;1000 stimulation repetitions. However, if such a response is masked by noise and by spontaneous fluctuations of blood flow, its physiological significance is questionable. The absence of a blood flow response also ensures that two-photon oxygen measurements and BOLD fMRI signals were not biased by the dual dependence of these signals on oxygen delivery and consumption. Finally, to process BOLD responses, we analyzed signals at the level of individual voxels selected anatomically, without any spatiotemporal smoothing and without any assumption on the response shape and timing. This should have allowed to extract small signals that would not pass conventional statistical thresholds. Note that using GLM analysis and a boxcar function representing the stimulation time and duration, shifted by 2 seconds, the activation maps did not show any BOLD response in the intracranial ON, even with statistical threshold of p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) (see Supplementary Fig.\u0026nbsp;2\u0026ndash;3). We believe that our data may be extrapolated to humans because the human ON also contains exclusively myelinated fibers\u0026mdash;unlike most other white-matter tracts\u0026mdash;and because a recent BOLD fMRI study\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e indicates that visual stimulation does not elicit a response in the human ON.\u003c/p\u003e \u003cp\u003eIn gray matter, Harris and Attwell \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e calculated the proportion of ATP used by cellular processes to be 43% for synaptic transmission, 17% for action potentials, 15% for maintaining the resting membrane potential, and 25% for housekeeping (i.e., processes not related to signaling). In axons, the properties of Na\u0026thinsp;+\u0026thinsp;and K+ channels are tuned to minimize the energy cost of action potential generation \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, and in the adult ON with sparse synapses, Harris and Attwell \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e estimated that only 0.1% and 0.4% of ATP is used for synaptic transmission and action potentials, respectively. It is therefore likely that the overall change in firing induced by natural visual stimulation consumes little energy, certainly less than in response to an electrical stimulation. This energy consumption change may not require any local blood flow regulation through NVC. The visual system is thus unique in that in the retina, it is also assumed that NVC does not occur in the layer containing the external segment of photoreceptors for an opposite reason: it consumes so much oxygen and glucose that energy is constantly delivered without NVC by the choroid vasculature \u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSource data are provided with this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe codes used for the analyses are available from the corresponding author upon request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e We thank Stephane Fouquet for image data segmentation in SYGLASS, Manon Omnes for the management of mice colonies and performing the surgeries for fUS, Emmanuelle Chaigneau for software support, and Melissa Glatigny for support with the experiments in rats. Financial support was provided by the Institut National de la Santé et de la Recherche Médicale (INSERM), the Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA), the Agence Nationale de la Recherche (CE14-0026-01), the Fondation Leducq Transatlantic Networks of Excellence program (16CVD05) and the IHU FOReSIGHT (IHU NVC intracranial). The 17.2T MRI system was supported through the Ile-de-France SESAME Large Equipment program. The synthesis of Oxyphor 2P was supported by the grant U24 EB028941 from the NIH USA (PI: Dr. Sergei A. Vinogradov).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u0026nbsp;\u003c/strong\u003eS.C. and L.C. designed the study. DESB, SC, DB and LC acquired BOLD fMRI data, which were analyzed by DESB and DB. A.A. and N.R. clarified and labeled the brains, which vascular structure was quantified by DESB. DESB developed the mouse preparation for two-photon imaging of the ON and acquired RBC velocity and blood oxygenation data. TV.E. synthetized Oxyphor 2P. SC and DESB interpreted two-photon data. All authors edited the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests:\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eIadecola C (2017) The Neurovascular Unit Coming of Age: A Journey through Neurovascular Coupling in Health and Disease. Neuron 96:17\u0026ndash;42\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAydin AK et al (2020) Transfer functions linking neural calcium to single voxel functional ultrasound signal. Nat Commun 11:2954\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRungta RL et al (2021) Diversity of neurovascular coupling dynamics along vascular arbors in layer II/III somatosensory cortex. Commun Biol 4:855\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchilling KG et al (2023) Whole-brain, gray, and white matter time-locked functional signal changes with simple tasks and model-free analysis. Proc. Natl. Acad. Sci. U. S. A 120, e2219666120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang H et al (2023) White matter BOLD signals at 7 Tesla reveal visual field maps in optic radiation and vertical occipital fasciculus. NeuroImage 269:119916\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchilling KG et al (2022) Anomalous and heterogeneous characteristics of the BOLD hemodynamic response function in white matter. Cereb Cortex Commun 3:tgac035\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWaxman SG, Swadlow HA (1977) The conduction properties of axons in central white matter. Prog Neurobiol 8:297\u0026ndash;324\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSturrock RR (1980) Myelination of the mouse corpus callosum. Neuropathol Appl Neurobiol 6:415\u0026ndash;420\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarkus HS, Joutel (2025) A. The pathogenesis of cerebral small vessel diseases and vascular cognitive impairment. Physiol Rev\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris JJ, Attwell D (2012) The energetics of CNS white matter. J Neurosci 32:356\u0026ndash;371\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForrester J, Peters (1967) A. Nerve fibres in optic nerve of rat. Nature 214:245\u0026ndash;247\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHonjin R, Sakato S, Yamashita T (1977) Electron microscopy of the mouse optic nerve: a quantitative study of the total optic nerve fibers. Arch Histol Jpn 40:321\u0026ndash;332\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRansom BR, Orkand RK (1996) Glial-neuronal interactions in non-synaptic areas of the brain: studies in the optic nerve. Trends Neurosci 19:352\u0026ndash;358\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLev-Ram V, Grinvald A (1987) Activity-dependent calcium transients in central nervous system myelinated axons revealed by the calcium indicator Fura-2. Biophys J 52:571\u0026ndash;576\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang CL, Wilson JA, Williams J, Chiu SY (2006) Action potentials induce uniform calcium influx in mammalian myelinated optic nerves. J Neurophysiol 96:695\u0026ndash;709\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLooser ZJ et al (2024) Oligodendrocyte-axon metabolic coupling is mediated by extracellular K(+) and maintains axonal health. Nat Neurosci 27:433\u0026ndash;448\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsadollahi E et al (2024) Oligodendroglial fatty acid metabolism as a central nervous system energy reserve. Nat Neurosci 10:1934\u0026ndash;1944\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRestrepo A et al (2022) Axo-vascular coupling mediated by oligodendrocytes. BioRxiv 1\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/2022.06.16.495900\u003c/span\u003e\u003cspan address=\"10.1101/2022.06.16.495900\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchroder S et al (2020) Arousal Modulates Retinal Output Neuron 107:487\u0026ndash;495\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoissonnet T, Tripodi M, Asari H (2023) Awake responses suggest inefficient dense coding in the mouse retina. Elife. 12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoido D et al (2019) Mesoscopic and microscopic imaging of sensory responses in the same animal. Nat Commun 10:1110\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbe Y, Tsurugizawa T, Le BD, Ciobanu L (2019) Spatial contribution of hippocampal BOLD activation in high-resolution fMRI. Sci Rep 9:3152\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe La Garza BH, Muir ER, Li G, Shih YY, Duong TQ (2011) Blood oxygenation level-dependent (BOLD) functional MRI of visual stimulation in the rat retina at 11.7 T. NMR Biomed 24:188\u0026ndash;193\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuong TQ, Ngan SC, Ugurbil K, Kim SG (2002) Functional magnetic resonance imaging of the retina. Invest Ophthalmol Vis Sci 43:1176\u0026ndash;1181\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRungta RL, Chaigneau E, Osmanski BF, Charpak S (2018) Vascular Compartmentalization of Functional Hyperemia from the Synapse to the Pia. Neuron 99:362\u0026ndash;375\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirst C et al (2020) Mapping the Fine-Scale Organization and Plasticity of the Brain Vasculature. Cell 180:780\u0026ndash;795\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKleinfeld D, Mitra PP, Helmchen F, Denk W (1998) Fluctuations and stimulus-induced changes in blood flow observed in individual capillaries in layers 2 through 4 of rat neocortex. Proc Natl Acad Sci U S A 95:15741\u0026ndash;15746\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChaigneau E, Oheim M, Audinat E, Charpak S (2003) Two-photon imaging of capillary blood flow in olfactory bulb glomeruli. Proc Natl Acad Sci U S A 100:13081\u0026ndash;13086\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang C, Ji N (2013) Characterization and improvement of three-dimensional imaging performance of GRIN-lens-based two-photon fluorescence endomicroscopes with adaptive optics. Opt Express 21:27142\u0026ndash;27154\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiantadosi SC et al (2024) Holographic stimulation of opposing amygdala ensembles bidirectionally modulates valencespecific behavior via mutual inhibition. Neuron 112:593\u0026ndash;610\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMace E et al (2011) Functional ultrasound imaging of the brain. Nat Methods 8:662\u0026ndash;664\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEsipova TV et al (2019) Oxyphor 2P: A High-Performance Probe for Deep-Tissue Longitudinal Oxygen Imaging. Cell Metabol 29:736\u0026ndash;744\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSakadzic S et al (2010) Two-photon high-resolution measurement of partial pressure of oxygen in cerebral vasculature and tissue. Nat Methods 7:755\u0026ndash;759\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLecoq J et al (2011) Simultaneous two-photon imaging of oxygen and blood flow in deep cerebral vessels. Nat Med 17:893\u0026ndash;898\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSencan I et al (2020) Optical measurement of microvascular oxygenation and blood flow responses in awake mouse cortex during functional activation. J Cereb Blood Flow Metab 42:510\u0026ndash;525\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParpaleix A, Houssen YG, Charpak S (2013) Imaging local neuronal activity by monitoring PO(2) transients in capillaries. Nat Med 19:241\u0026ndash;246\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAydin AK, Verdier C, Chaigneau E, Charpak S (2022) The oxygen initial dip in the brain of anesthetized and awake mice. Proc. Natl. Acad. Sci. U. S. A 119, e2200205119\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMasamoto K, Kanno I (2012) Anesthesia and the quantitative evaluation of neurovascular coupling. J Cereb Blood Flow Metab 32:1233\u0026ndash;1247\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorisset C et al (2022) Retinal functional ultrasound imaging (rfUS) for assessing neurovascular alterations: a pilot study on a rat model of dementia. Sci Rep 12:19515\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKornfield TE, Newman EA (2014) Regulation of blood flow in the retinal trilaminar vascular network. J Neurosci 34:11504\u0026ndash;11513\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBiesecker KR et al (2016) Glial Cell Calcium Signaling Mediates Capillary Regulation of Blood Flow in the Retina. J Neurosci 36:9435\u0026ndash;9445\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMishra A et al (2016) Astrocytes mediate neurovascular signaling to capillary pericytes but not to arterioles. Nat Neurosci 19:1619\u0026ndash;1627\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeppiatt CM, Howarth C, Mobbs P, Attwell D (2006) Bidirectional control of CNS capillary diameter by pericytes. Nature 443:700\u0026ndash;704\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonzales AL et al (2020) Contractile pericytes determine the direction of blood flow at capillary junctions. Proc. Natl. Acad. Sci. U. S. A 117, 27022\u0026ndash;27033\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMughal A, Nelson MT, Hill-Eubanks D (2023) The post-arteriole transitional zone: a specialized capillary region that regulates blood flow within the CNS microvasculature. J Physiol 601:889\u0026ndash;901\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePfeiffer T, Li Y, Attwell D (2021) Diverse mechanisms regulating brain energy supply at the capillary level. Curr Opin Neurobiol 69:41\u0026ndash;50\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlarcon-Martinez L et al (2020) Interpericyte tunnelling nanotubes regulate neurovascular coupling. Nature 585:91\u0026ndash;95\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoo JH et al (2021) Loss of Caveolin-1 Impairs Light Flicker-Induced Neurovascular Coupling at the Optic Nerve Head. Front Neurosci 15:764898\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMace E et al (2018) Whole-Brain Functional Ultrasound Imaging Reveals Brain Modules for Visuomotor Integration. Neuron 100:1241\u0026ndash;1251\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanganahalli BG et al (2022) Thalamic activations in rat brain by fMRI during tactile (forepaw, whisker) and non-tactile (visual, olfactory) sensory stimulations. PLoS ONE 17:e0267916\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShanks JA et al (2016) Corticothalamic Axons Are Essential for Retinal Ganglion Cell Axon Targeting to the Mouse Dorsal Lateral Geniculate Nucleus. J Neurosci 36:5252\u0026ndash;5263\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMetea MR, Newman EA (2006) Glial cells dilate and constrict blood vessels: a mechanism of neurovascular coupling. J Neurosci 26:2862\u0026ndash;2870\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFunfschilling U et al (2012) Glycolytic oligodendrocytes maintain myelin and long-term axonal integrity. Nature 485:517\u0026ndash;521\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao Y et al (2025) Myelination selectively modulates BOLD signal in white matter. Res Sq\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlle H, Roth A, Geiger JR (2009) Energy-efficient action potentials in hippocampal mossy fibers. Science 325:1405\u0026ndash;1408\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShih YY et al (2013) Quantitative retinal and choroidal blood flow during light, dark adaptation and flicker light stimulation in rats using fluorescent microspheres. Curr Eye Res 38:292\u0026ndash;298\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eMice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll animal care and experimentations were performed in accordance with the INSERM Animal Care and Use Committee guidelines and approved by the ethical committee (Charles Darwin, comité national de réflexion éthique sur l’expérimentation animale n°5; protocol number #27135 2020091012114621). Mice were fed ad libitum and housed in a 12-hour light-dark cycle at 22°C and 50% humidity. A total of 17 adult mice were used in this study: 10 mice TPLSM and 7 mice for iDISCO. Both males and females, from 4 to 14 months of age were included. The following mouse lines were used and bred in our animal facility: C57/BL6J and Thy1-GCaMP6s (GP5.1).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRats\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll animal procedures were approved by the Comité d’Ethique en Expérimentation Animale, Commissariat à l’Energie Atomique et aux Énergies Alternatives, and by the Ministère de l’Education Nationale, de l’Enseignement Supérieur et de la Recherche (France) under reference #55527-2025051411223097 v2 and were conducted in strict accordance with the recommendations and guidelines of the European Union (Directive 2010/63/EU) and the French National Committee (Décret 2013–118). Rats were fed ad libitum and housed in a 12-hour light-dark cycle at 22°C and 50% humidity. A total of seven female Sprague Dawley from 2 to 6 months of age were included in this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurgery for GRIN lens implantation above the right ON\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDexamethasone (6 mg/kg, s.c) was administered 24h and 1h before the surgery to prevent brain edema and reduce inflammation. Buprenorphine (0.3 mg/kg s.c.) was injected 2 hours before and 24h after surgery. Mice were anaesthetized with isoflurane (3-5% for induction and 1.5-2% for maintenance) and positioned in a stereotaxic frame (Model 1900 Stereotaxic Alignment System - Kopf Instruments) with ear bars. Body temperature was maintained at 36.5±0.5ºC using a feedback-controlled heating pad with a rectal probe. Lidocaine was injected below the skin (32 mg/kg s.c.). The head of the mouse was shaved with scissors and depilatory cream was applied from the neck to the eye level. The skin was disinfected, and a midline incision was made in the skin from the neck to the level of the eyes with a scalpel blade and the skin was maintained on the sides with clamps. Connective tissues over the skull were removed using a small scalpel blade. A mixture of 50% air and 50% oxygen was delivered through a nose cone to maintain blood oxygenation during the anesthesia. The skull was disinfected with betadine solution, cleaned with a sterile NaCl buffer, and prepared for the adhesion of light-curing cement using a primer solution. The mouse’s head was then aligned with the aid of a stereotaxic alignment indicator (Model 1905), positioning the indicator at bregma and lambda to ensure proper orientation. Alignment was confirmed in both the anteroposterior and coronal directions, after which the centering scope was positioned at bregma, and the coordinates set to zero. A stereotaxic drill equipped with a 0.7 mm burr was carefully lowered to create a hole in the skull for the GRIN lens insertion at the following coordinates from bregma: anterior–posterior: 1.4 mm; medial-lateral: 1.4 mm. A coronal rotation of 11° (counterclockwise) and a sagittal rotation of 10° (clockwise) were then performed. The burr was replaced by a 25G needle, which was inserted in the hole and gradually lowered in the direction of the right optic nerve to a depth of -4.7 mm while continuously applying cortical buffer to limit brain displacement. The needle was subsequently removed and replaced with a custom-designed holder containing the GRIN lens. The lens was carefully lowered to -4.9 mm and its upper part was fixed to the bone with a thin layer of UV-cured dental cement. A layer of primer solution was applied on the skull and the upper part of the GRIN lens was protected by a small metal tube (5 mm in diameter and 7 mm in height) secured on the skull with UV-cured dental cement. To prevent light contamination from the visual stimulation screen, a mixture of Unifast TRAD solvent/powder, colored with non-toxic black acrylic paint, was applied over the UV-cured dental cement. A horizontal titanium bar was then attached and sealed with dental cement on the posterior part of the skull, maintaining orthogonality between the head bar and the GRIN lens. Finally, surgical glue was applied to close the skin around the cemented area. In total, the surgical procedure for GRIN lens implantation took approximately 90 minutes to complete. Post-surgery, the animal was placed in a heating box until fully awake, provided with gel boost, and housed overnight in a warm cage. A post-operative scoring was performed for three days after the surgery.\u003c/p\u003e\n\u003cp\u003eIn 8 out of 10 mice, the accurate targeting of the GRIN lens was verified postmortem. Keeping the head fixed, we performed a craniotomy, slowly and carefully removed the left cortex and diencephalon to uncover the two ONs. This allowed to ensure that the chronic GRIN lens had been correctly positioned over the right ON.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSedation for two-photon imaging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA recovery period of 2-3 weeks minimum was respected before imaging experiments were performed. During imaging sessions, mice were sedated with continuous perfusion of dexmedetomidine as follows: induction of anesthesia was done with 3% isoflurane and then decreased by 0.5% every 5 min down to 0% within 30 minutes, and dexmedetomidine was administered with a bolus (0.025 mg/kg s.c.) at the beginning of the session and a s.c. perfusion (0.1 mg/kg/h) during the entire session. Texas Red (70 kDa dextran, D1830, LifeTechnologies) was administered intravenously (i.v.) with a retro-orbital injection (left eye) prior to the decrease in isoflurane (under 3-2.5% isoflurane). Recordings started 20 minutes after the isoflurane cutoff. Mice were allowed to freely breathe air supplemented with oxygen (30% final concentration). Body temperature was controlled with a rectal probe and maintained at 36.5°C with a feedback-controlled heating pad. The animal was monitored throughout the imaging session using an infrared webcam (DCC3240N, Thorlabs). At the end of each experiment, the mouse was injected with atipamezole to antagonize the effects of medetomidine and accelerate recovery of physiological functions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTraining for two-photon imaging in awake mice\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor TPLSM training, mice were head-fixed to a frame with a running wheel, and gradually habituated to the experimental setup, which included regular delivery of visual stimulation, for \u0026gt;10 days before experimental sessions began. Locomotion and movement were monitored with a velocity encoder connected to the running wheel and the infrared camera. Mice were briefly (\u0026lt;2 min) anesthetized with isoflurane in order to inject Texas Red dextran or OXYPHOR-2P in the left eye, and recovered for \u0026gt;1 hour prior to initiating experimental sessions. Only trials for which the mouse remained still throughout visual stimulation were analyzed.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVisual stimulations for two-photon and fUS imaging experiments.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor mice, visual stimuli were delivered via a 13-inch screen controlled by custom MATLAB software and placed 9 cm in front of the mouse’s eye. Mice received oxygenated air to a final concentration of 30% O\u003csub\u003e2\u003c/sub\u003e continuously through a nose cone. Visual stimulation consisted of 10s of drifting gratings (frequency 0.04 cpd, 50% contrast, intensity 76 ± 1 lux) with a contrast of 50% after a rest period varying from 10s to 30s. To avoid light pollution from the screen monitor onto the light collection path, the upper ring of a black circular and foldable curtain was slid along the microscope objective and the lower ring along a small metal tube protecting the upper part of the GRIN lens (Fig. 4a).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTwo-photon laser scanning microscopy and data analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTPLSM imaging was performed with a custom-built microscope, previously described \u003csup\u003e34\u003c/sup\u003e and data were acquired with a customized LabVIEW software (National Instruments). Two-Photon excitation was obtained using a femtosecond laser, either from Coherent (Mira, 120-fs pulses) or SpectraPhysics (Insight, 70-fs pulses). Laser power was modulated with an acousto-optic modulator (AA Optoelectronic, MT110B50-A1.5-IR-Hk). Galvanometric mirrors (Cambridge Technology) were used to target the sample at the desired points. The excitation beam was focused onto the sample by using a 10X/0.3NA (Olympus) and a GRIN lens from GRINTECH (NEM-050-05-10-860-DM or NEM-250-05-10-860-DM). GCaMP6f and Texas Red were excited at 920 nm, OXYPHOR-2P at 960 nm, and the emission light was separated from the excitation light with a dichroic mirror (DXCR 875, Chroma Technology Corp [Chroma]). Emitted photons were than divided by a dichroic mirror (cutoff wavelength = 560 nm). The green channel was lowpass and bandpass filtered (E800, HQ 525/50 nm, Chroma Technology Corp) and light was collected onto a GaAsP (Hamamatsu) photomultiplier tube. The red channel had one lowpass filter (E800, Chroma), one bandpass filter (FF01 794/ 160, Semrock), and a red-sensitive photomultiplier tube (R6357, Hamamatsu). Post-processing analysis for the RBC velocities and PO2 measurements was done in Matlab (version R2018a, MathWorks) with a custom-made software.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003efUS imaging and data analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCBV measurements were acquired as previously \u003csup\u003e21\u003c/sup\u003eusing an ultrasound scanner (Iconeus One, Iconeus). Briefly, we used a linear ultrasound probe (128 elements, 15 MHz central frequency, Vermon) with an ultrasound sequence consisting of transmitting 11 different tilted plane waves (from −10° to 10° in 2° increments) with a 5500-Hz pulse repetition frequency (500 Hz frame rate of reconstructed images). Post-processing analysis for the sagittal sections was done in Matlab (version R2018a, MathWorks) with a custom-made software: the first 40 singular value decomposition (SVD) were removed and the power Doppler (PD) signal was further filtered with a Butterworth filter (fifth order). Data are shown as ΔPD/PD, with a baseline of 9 s (from 1 to 10 s). CBV flowing at low (10-30 Hz, 0.5-1.5 mm/s) and high (\u0026gt; 60 Hz, \u0026gt; 3 mm/s) axial velocities were separated by filtering as previously described\u003csup\u003e21\u003c/sup\u003e. For the coronal sections, post-processing analysis of CBV changes and identification of the dLGN, SC and visual areas were done using the ICOstudio software. All experiments were performed in sedated mice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBOLD fMRI and data analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe animals were anesthetized initially with isoflurane and a subcutaneously administered bolus of 0.05 mg/kg medetomidine (Domitor, Pfizer, Karlsruhe, Germany), and were freely breathing an air–oxygen gas mixture (36 % O\u003csub\u003e2\u003c/sub\u003e). For 4 rats, a small volume of Aflunox (VWR International, Rosny-sous-Bois, France) was injected in the right middle ear in order to decrease the susceptibility artifacts on the right ventral side of the brain. The rats were placed in the cradle and the head was immobilized using a bite bar and ear pins. The body temperature of the animals was maintained in the range of 36–37.5°C using a circulating hot water blanket. An optical fiber was placed in front of the left eye. The visual stimulation paradigm consisted of 6 s LED-on (flashing at 2 Hz) and 12 s LED-off and was delivered using a blue light emitting diode (LED) controlled by an Arduino One (ArduinoCham, Switzerland).\u003c/p\u003e\n\u003cp\u003eAll MRI acquisitions were performed on a 17.2 Tesla MRI system (Bruker BioSpin, Ettlingen, Germany) with a 30 mm TX/RX coil (Bruker BioSpin, Ettlingen, Germany). Anatomical images of the whole brain were acquired using a Rapid Acquisition with Relaxation Enhancement (RARE) pulse sequence. Functional BOLD acquisitions were performed using a gradient echo echo-planar imaging (GE-EPI). Two series of acquisitions with two different orientation were acquired. The common parameters for the two acquisitions were the following: TE = 11.5 ms, TR = 1000 ms, number of slices = 9, number of repetitions = 372. The spatial resolution varied depending on the orientation as follows: coronal: slice thickness 0.8 mm, in plane resolution 0.2 mm x 0.2 mm; axial: slice thickness 0.5 mm, in plane resolution 0.24 mm x 0.24 mm. A diffusion weighted acquisition was also acquired in the axial orientation (TE = 16 ms, TR = 1500 ms, slice thickness 0.5 mm, in plane resolution 0.2 mm x 0.2 mm, b value = 1000 s/mm\u003csup\u003e2\u003c/sup\u003e, direction = [0,0,1], NA =2) in order to precisely identify the optic nerve. For all acquisitions, shimming was performed on a slab encompassing the slices of interest using ‘mapshim’ (Bruker).\u003c/p\u003e\n\u003cp\u003eCo-registration of DW and BOLD images allowed to measure functional responses exclusively in the voxels containing the ON, from the retina to the most ventral part of the OT. Signal averaging was performed over trials and over animals in order to improve the SNR. After averaging, the analysis was performed without any assumption on the timing and shape of responses. This approach avoided imposing assumptions about the HRF and improved sensitivity to small effects. It was similar to that used with two-photon acquisitions of blood flow and oxygenation, providing a more accurate characterization of subtle BOLD responses. Functional activation maps were however also computed\u0026nbsp;using GLM analysis and a boxcar function representing the stimulation time and duration, shifted by 2 seconds. Using a very low statistical threshold p\u0026lt;0.01, allowed to detect some activated voxels in the ON initial portion, as with our primary analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTissue Clearing and Immunolabeling\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMice underwent intracardiac perfusion with a solution of 4% PFA (EMS) in PBS, then the brains were dissected and post-fixed in the same solution for 2h at room temperature. Mouse brain tissues, including the optic nerve and optic tract, were processed using the previously described iDISCO+ tissue clearing protocol (24). Samples were immunolabelled with primary antibodies against SM22 at 1/1000\u003csup\u003eth\u003c/sup\u003e dilution from stock (Abcam; #ab14106), directly conjugated with AlexaFluor 555 (Invitrogen), and podocalyxin (Podxl)(1/300\u003csup\u003eth\u003c/sup\u003e dilution from stock) (R\u0026amp;D systems; #MAB1556) and CD31 (PECAM-1)(1/1000\u003csup\u003eth\u003c/sup\u003e diluation from stock) (R\u0026amp;D systems; #AF3628), both directly conjugated to AlexaFluor 647 (Invitrogen). The tissues were processed with iDISCO+ as previously described (\u003cem\u003e24\u003c/em\u003e), with a 2 weeks incubation time at 37°c for primary antibodies (no secondary antibodies were used), and stored and imaged in DiBenzyl Ether (Sigma).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLight-Sheet Imaging and Preprocessing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe imaging of cleared samples was conducted using a light-sheet fluorescence microscope (Blaze, LaVision BioTec) at a voxel resolution of [1.62 × 1.62 × 2.5 µm], using the lightspeed acquisition mode. 4X NA0.35 objective was used for light collection, and a light sheet NA of 0.1 was used. Imaging channels included 488nm excitation for autofluorescence, 541nm for Sm22 and 639nm for CD31 and Podocalyxin. The 3 channels were acquired in the same imaging sequence to obtain a perfect alignment between them. The imaging sequence was set as z-drive -\u0026gt; filter wheel -\u0026gt; x-drive -\u0026gt; y-drive. The raw image tiles were imported into ClearMap for stitching \u003csup\u003e26\u003c/sup\u003e and channel alignment, using the CD31/Podocalyxin channel as the stitching reference, and its layout to stitch the other 2 channels. The volumetric datasets were processed and subsequently exported in multi-channel TIFF format for the purpose of subsequent analysis in Imaris (Oxford Instrument) and SyGlass (IstoVision).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVascular Segmentation and Quantification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eImage volumes were converted to a format compatible with syGlass using the syGlass Data Converter. Segmentation and analysis were performed using syGlass (IstoVision), allowing manual annotation within the volumetric data.The vascular networks were then segmented based on the expression of the markers CD31 and PODO31. SM22 expression was used to identify vessels containing smooth muscle cells. The segmentation of venous identity was based on the distribution of markers and the morphology of vessels (low expression of SM22, and large radii).The branching points, which are defined as discretevascular bifurcations, were manually annotated across the ON and OT in SyGlass, each point being assigned a single count.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistics and onset analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe computed some of TPLSM data in z score because of fluctuations in the baseline due to vasomotion \u003csup\u003e3\u003c/sup\u003e. Z score was calculated for each trial as follows: z score = (x – μ\u003csub\u003ebaseline\u003c/sub\u003e) / σ\u003csub\u003ebaseline\u003c/sub\u003e with 10 s long baseline, with μ\u003csub\u003ebaseline\u003c/sub\u003e the mean of the baseline, and σ\u003csub\u003ebaseline\u003c/sub\u003e the standard deviation (SD) of the baseline. Trials from the same vessel were averaged (with a 0.1 s interpolation) for analysis.\u0026nbsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Institut de la Vision","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"NVC, neurovascular coupling, fmri, Two-photon microscopy, GRIN lens, white matter, optic nerve","lastPublishedDoi":"10.21203/rs.3.rs-9086876/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9086876/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn brain gray matter, neurovascular coupling (NVC) maintains brain metabolism homeostasis by modulating blood flow according to neuronal activity. In white matter, the energy cost of information transmission along myelinated axons is reduced and the need for NVC is unknown. Here, we used two-photon imaging through chronically-implanted GRIN lenses in mice and high-field BOLD fMRI (17.2T) in rats to investigate NVC along the entire length of the optic nerve, a unique model of a myelinated axonal tract. We found that flickering light and drifting grating stimulations increased blood flow in the retina, the unmyelinated optic nerve head, and at the level of the nerve synaptic terminals. However, it did not affect blood flow and oxygenation in the myelinated part of the optic nerve, i.e., the intracranial optic nerve and the optic tract. We conclude that during natural visual stimulation, action potential propagation in activated myelinated axons does not trigger NVC.\u003c/p\u003e","manuscriptTitle":"Action potential propagation in the rodent myelinated optic nerve does not trigger neurovascular coupling","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-13 15:14:15","doi":"10.21203/rs.3.rs-9086876/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e564360f-351e-435e-9bb2-6594c9649dee","owner":[],"postedDate":"March 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64341987,"name":"Cellular \u0026 Molecular Neuroscience"}],"tags":[],"updatedAt":"2026-03-13T15:14:15+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-13 15:14:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9086876","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9086876","identity":"rs-9086876","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}