Abstract
Hierarchical and parallel networks are fundamental structures of the mammalian brain1,2,3,4,5,6,7,8. During development, lower- and higher-order thalamic nuclei and many cortical areas in the visual system form interareal connections and build hierarchical dorsal and ventral streams9,10,11,12,13. One hypothesis for the development of visual network wiring involves a sequential strategy wherein neural connections are sequentially formed alongside hierarchical structures from lower to higher areas14,15,16,17. However, this sequential strategy would be inefficient for building the entire visual network comprising numerous interareal connections. We show that neural pathways from the mouse retina to primary visual cortex (V1) or dorsal/ventral higher visual areas (HVAs) through lower- or higher-order thalamic nuclei form as parallel modules before corticocortical connections. Subsequently, corticocortical connections among V1 and HVAs emerge to combine these modules. Retina-derived activity propagating the initial parallel modules is necessary to establish retinotopic inter-module connections. Thus, the visual network develops in a modular manner involving initial establishment of parallel modules and their subsequent concatenation. Findings in this study raise the possibility that parallel modules from higher-order thalamic nuclei to HVAs act as templates for cortical ventral and dorsal streams and suggest that the brain has an efficient strategy for the development of a hierarchical network comprising numerous areas.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The datasets generated during the current study are available as downloadable files at https://figshare.com/projects/Modular_strategy_for_the_development_of_hierarchical_visual_networks_in_the_mice/116580. Several representative instances of raw data are also available at this URL. Other raw data are available from the corresponding authors on reasonable request. Reference images of cortical sections of the mouse brain are available as downloadable files at the 3D coronal adult brain atlas of the Allen Institute (https://mouse.brain-map.org/static/atlas). Source data are provided with this paper.
Code availability
Code is available from the corresponding authors on reasonable request.
References
Zeki, S. & Shipp, S. The functional logic of cortical connections. Nature 335, 311–317 (1988).
Felleman, D. J. & Van Essen, D. C. Distributed hierarchical processing in the primate cerebral cortex. Cereb. Cortex 1, 1–47 (1991).
Kaas, J. H. & Hackett, T. A. Subdivisions of auditory cortex and processing streams in primates. Proc. Natl Acad. Sci. USA 97, 11793–11799 (2000).
Kaas, J. H. & Lyon, D. C. Pulvinar contributions to the dorsal and ventral streams of visual processing in primates. Brain Res. Rev. 55, 285–296 (2007).
Wang, Q., Sporns, O. & Burkhalter, A. Network analysis of corticocortical connections reveals ventral and dorsal processing streams in mouse visual cortex. J. Neurosci. 32, 4386–4399 (2012).
Park, H. & Friston, K. Structural and functional brain networks: from connections to cognition. Science 342, 1238411 (2013).
Markov, N. T. et al. Anatomy of hierarchy: feedforward and feedback pathways in macaque visual cortex. J. Comp. Neurol. 522, 225–259 (2014).
Bennett, C. et al. Higher-order thalamic circuits channel parallel streams of visual information in mice. Neuron 102, 477–492 (2019).
López-Bendito, G. & Molnár, Z. Thalamocortical development: how are we going to get there? Nat. Rev. Neurosci. 4, 276–289 (2003).
Guillery, R. W. Is postnatal neocortical maturation hierarchical? Trends Neurosci. 28, 512–517 (2005).
Luo, L. & O’Leary, D. D. M. Axon retraction and degeneration in development and disease. Annu. Rev. Neurosci. 28, 127–156 (2005).
Bourne, J. A. & Morrone, M. C. Plasticity of visual pathways and function in the developing brain: is the pulvinar a crucial player? Front. Syst. Neurosci. 11, 3 (2017).
Chomiak, T. & Hu, B. Mechanisms of hierarchical cortical maturation. Front. Cell. Neurosci. 11, 272 (2017).
Godement, P., Salaün, J. & Imbert, M. Prenatal and postnatal development of retinogeniculate and retinocollicular projections in the mouse. J. Comp. Neurol. 230, 552–575 (1984).
Auladell, C., Pérez-Sust, P., Supèr, H. & Soriano, E. The early development of thalamocortical and corticothalamic projections in the mouse. Anat. Embryol. 201, 169–179 (2000).
Dong, H., Wang, Q., Valkova, K., Gonchar, Y. & Burkhalter, A. Experience-dependent development of feedforward and feedback circuits between lower and higher areas of mouse visual cortex. Vis. Res. 44, 3389–3400 (2004).
Berezovskii, V. K., Nassi, J. J. & Born, R. T. Segregation of feedforward and feedback projections in mouse visual cortex. J. Comp. Neurol. 519, 3672–3683 (2011).
Shipp, S. The functional logic of cortico–pulvinar connections. Philos. Trans. R. Soc. B 358, 1605–1624 (2003).
Sherman, S. M. Thalamus plays a central role in ongoing cortical functioning. Nat. Neurosci. 19, 533–541 (2016).
Zhou, N., Maire, P. S., Masterson, S. P. & Bickford, M. E. The mouse pulvinar nucleus: organization of the tectorecipient zones. Vis. Neurosci. 34, E011 (2017).
Rokszin, A. et al. Visual pathways serving motion detection in the mammalian brain. Sensors 10, 3218–3242 (2010).
Murakami, T., Matsui, T. & Ohki, K. Functional segregation and development of mouse higher visual areas. J. Neurosci. 37, 9424 (2017).
Feldheim, D. A. & O’Leary, D. D. M. Visual map development: bidirectional signaling, bifunctional guidance molecules, and competition. Cold Spring Harb. Perspect. Biol. 2, a001768 (2010).
Huberman, A. D., Feller, M. B. & Chapman, B. Mechanisms underlying development of visual maps and receptive fields. Annu. Rev. Neurosci. 31, 479–509 (2008).
Erzurumlu, R. S. & Gaspar, P. Development and critical period plasticity of the barrel cortex. Eur. J. Neurosci. 35, 1540–1553 (2012).
Arcaro, M. J. & Livingstone, M. S. A hierarchical, retinotopic proto-organization of the primate visual system at birth. eLife 6, e26196 (2017).
Homman-Ludiye, J. & Bourne, J. A. Mapping arealisation of the visual cortex of non-primate species: lessons for development and evolution. Front. Neural Circuits 8, 79 (2014).
Warner, C. E., Kwan, W. C. & Bourne, J. A. The early maturation of visual cortical area MT is dependent on input from the retinorecipient medial portion of the inferior pulvinar. J. Neurosci. 32, 17073–17085 (2012).
Chou, S. et al. Geniculocortical input drives genetic distinctions between primary and higher-order visual areas. Science 340, 1239–1242 (2013).
Vue, T. Y. et al. Thalamic control of neocortical area formation in mice. J. Neurosci. 33, 8442–8453 (2013).
Moreno-Juan, V. et al. Prenatal thalamic waves regulate cortical area size prior to sensory processing. Nat. Commun. 8, 14172 (2017).
Antón-Bolaños, N. et al. Prenatal activity from thalamic neurons governs the emergence of functional cortical maps in mice. Science 364, 987–990 (2019).
Batardière, A. et al. Early specification of the hierarchical organization of visual cortical areas in the macaque monkey. Cereb. Cortex 12, 453–465 (2002).
Baldwin, M. K. L., Kaskan, P. M., Zhang, B., Chino, Y. M. & Kaas, J. H. Cortical and subcortical connections of V1 and V2 in early postnatal macaque monkeys. J. Comp. Neurol. 520, 544–569 (2012).
Ackman, J. B., Burbridge, T. & Crair, M. C. Retinal waves coordinate patterned activity throughout the developing visual system. Nature 490, 219–225 (2012).
Matsui, T., Murakami, T. & Ohki, K. Transient neuronal coactivations embedded in globally propagating waves underlie resting-state functional connectivity. Proc. Natl Acad. Sci. USA 113, 6556–6561 (2016).
Yamada, J. et al. Ephrin-As and neural activity are required for eye-specific patterning during retinogeniculate mapping. Nat. Neurosci. 8, 1022–1027 (2005).
Monavarfeshani, A., Sabbagh, U. & Fox, M. A. Not a one-trick pony: diverse connectivity and functions of the rodent lateral geniculate complex. Vis. Neurosci. 34, E012 (2017).
Laing, R. J., Lasiene, J. & Olavarria, J. F. Topography of striate–extrastriate connections in neonatally enucleated rats. BioMed Res. Int. 2013, 592426–592429 (2013).
Cang, J. et al. Development of precise maps in visual cortex requires patterned spontaneous activity in the retina. Neuron 48, 797–809 (2005).
Dye, C. A., Abbott, C. W. & Huffman, K. J. Bilateral enucleation alters gene expression and intraneocortical connections in the mouse. Neural Dev. 7, 5 (2012).
Frangeul, L. et al. A cross-modal genetic framework for the development and plasticity of sensory pathways. Nature 538, 96–98 (2016).
Smith, I. T., Townsend, L. B., Huh, R., Zhu, H. & Smith, S. L. Stream-dependent development of higher visual cortical areas. Nat. Neurosci. 20, 200–208 (2017).
Tsukano, H. et al. Reconsidering tonotopic maps in the auditory cortex and lemniscal auditory thalamus in mice. Front. Neural Circuits 11, 14 (2017).
Vanni, M. P., Chan, A. W., Balbi, M., Silasi, G. & Murphy, T. H. Mesoscale mapping of mouse cortex reveals frequency-dependent cycling between distinct macroscale functional modules. J. Neurosci. 37, 7513–7533 (2017).
Adibi, M. Whisker-mediated touch system in rodents: from neuron to behavior. Front. Syst. Neurosci. 13, 40 (2019).
Phillips, J. W. et al. A repeated molecular architecture across thalamic pathways. Nat. Neurosci. 22, 1925–1935 (2019).
Pouchelon, G. et al. Modality‐specific thalamocortical inputs instruct the identity of postsynaptic L4 neurons. Nature 511, 471–474 (2014).
Gorski, J. A. et al. Cortical excitatory neurons and glia, but not GABAergic neurons, are produced in the Emx1-expressing lineage. J. Neurosci. 22, 6309–6314 (2002).
Taniguchi, H. et al. A resource of Cre driver lines for genetic targeting of GABAergic neurons in cerebral cortex. Neuron 71, 995–1013 (2011).
Zariwala, H. A. et al. A Cre-dependent GCaMP3 reporter mouse for neuronal imaging in vivo. J. Neurosci. 32, 3131–3141 (2012).
Madisen, L. et al. Transgenic mice for intersectional targeting of neural sensors and effectors with high specificity and performance. Neuron 85, 942–958 (2015).
Peirce, J. W. PsychoPy—psychophysics software in Python. J. Neurosci. Methods 162, 8–13 (2007).
Vincent, J. L. et al. Intrinsic functional architecture in the anaesthetized monkey brain. Nature 447, 83–86 (2007).
Fratzl, A. et al. Flexible inhibitory control of visually evoked defensive behavior by the ventral lateral geniculate nucleus. Neuron 109, 3810–3822 (2021).
Acknowledgements
We thank members of the Ohki laboratory for discussions; A. Honda, Y. Sono, A. Hayashi, T. Inoue, A. Ohmori, M. Saito, Y. Kato, M. Taki, S. Uto and N. Yamamoto for animal care and genotyping; T. Ohmine and N. Yamamoto for lab management; and H. Takeuchi and K.M. Hagihara for comments on the manuscript. We acknowledge support from The Research Support Center, Kyushu University Graduate School of Medical Sciences, and Imaging Core of International Research Center for Neurointelligence (IRCN), The University of Tokyo, Virus Vector Core of the Brain/MINDS-AMED (JP21dm0207111), and GENIE project, the Janelia Farm Research Campus. This work was supported by Brain/MINDS-AMED (14533320, 15653077, JP16dm0207034 and JP20dm0207048 to K.O.); the Institute for AI and Beyond (to K.O.); JSPS KAKENHI (25221001, 19H05642 and 20H05917 to K.O., 17K14931, 18H05116 and 21H0516513 to T. Matsui, and 19K21207 and 21K15181 to T. Murakami); Brain/MINDS-beyond-AMED (JP20dm0307031 to T. Matsui); and JST PRESTO (JPMJPR19M9 to T. Matsui).
Author information
Authors and Affiliations
Contributions
T. Murakami, T. Matsui and K.O. designed the research; T. Murakami and T. Matsui performed the experiments; T. Murakami analysed the data; M.U. designed and created several AAVs encoding GCaMP; and T. Murakami, T. Matsui and K.O. wrote the paper.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Schematics of the two hypotheses of visual network development.
a, In the sequential scenario, neural connections emerge along with a hierarchical structure from lower to higher-order areas. Magenta arrows indicate newly formed connections at each stage. Colour gradients indicate a retinotopic structure. dLGN: lateral geniculate nucleus, SC: superior colliculus, V1: primary visual area, LPN: lateral posterior nucleus, vHVA and dHVA: higher visual areas in ventral and dorsal streams, respectively. It is postulated that projections from LPN to HVAs is formed around the same time as those from V1 to HVAs. In addition, projections from V1 and LPN to HVAs are firstly formed with a part of HVAs. b, In the parallel modular scenario, projections from higher-order visual thalamus to HVA are formed as an independent module with the dLGN-V1 module. It is postulated that projections from LPN to HVAs are formed much earlier than those from V1 to HVAs. In this step, LPN projects to all HVAs.
Extended Data Fig. 2 Change in the spontaneous activity in HVAs by binocular enucleation.
a, Schematics representation of binocular enucleation at P5. b, Frequency maps of spontaneous activity in a representative mouse. c, Time course of V1 and HVAs of the mouse in (b). d, Changes in the frequency of spontaneous activity in V1 and individual HVAs by binocular enucleation (n = 6 mice). e, Comparison of the changes in spontaneous activity between ventral and dorsal HVAs. Wilcoxon signed-rank test, *: P = 0.031
Extended Data Fig. 3 Representative FC maps in each age and development of FCs between all areal pairs.
a–d, Representative FC maps of seeds in V1 and seven HVAs at P5–14. White crosses indicate seed regions of interests (ROIs). Scale bar: 1 mm. e–h, Boxplots of FCs between all areal pairs.
Extended Data Fig. 4 Developmental changes of HVA-HVA connections observed by anatomical DiI tracing and electrical stimulation.
a, Tangential sectioning of the brain injected DiI into LM at P5 and P10. Black crosses indicate DiI injection sites. b, Tangential sectioning of the brain injected DiI into PM at P5 and P10. In both cases of injections into LM and PM, retrogradely labelled neurons in other HVAs are observed at P10, but not at P5. c, Schematic of the experimental procedure. d, Representative response maps by electrical stimulation of LM at P5 and P9. e, Response amplitudes in V1 and PM by LM stimulation, which were normalized by signal amplitude of the stimulation site. *: P < 0.05. Student’s t-test with Bonferroni correction, V1: t(6) = –2.72, P = 0.065; PM: t(6) = –5.77, P = 5.42 × 10–4.
Extended Data Fig. 5 Anatomical and functional examinations of the development of projections from V1 to HVAs.
a, Representative examples of DiI injection site and retrogradely labelled neurons at P5. Red circles indicate retrogradely labelled neurons. b, c, Retrogradely labelled neurons in V1 by DiI injection into PM (b) or LM (c) at P5 and P10 (n = 4 mice at each age). No significant differences were observed between P5 and P10 mice. Student’s t-test with Bonferroni correction, PM: t(6) = –1.28, P = 1.0; LM: t(6) = –1.63, P = 1.0. P values were Bonferroni corrected, including the tests in Fig. 2. d, Schematic of the experimental procedure for labelling projections from V1 to HVAs. e, Top: Tangential sections of the flattened cortex at each age. Bottom: Contour plots of the sections above. The open triangle represents the peak of V1 injection site, and filled triangles represent peaks in HVAs. f, The number of projection peaks in HVAs (n = 5 mice for each age). One-way ANOVA, F(2,12) = 15.44, P = 4.80 × 10–4; post hoc Tukey HSD, *: P < 0.05. g, Schematic of experimental design. h, Representative response maps by electrical stimulation in V1 at P5, P9, and P14 (n = 6 mice for each age). i–k, Response amplitudes in LM (i), AL (j), and PM (k) by V1 stimulation, which was normalized by the signal amplitude of the stimulation site. One-way ANOVA with post hoc Tukey HSD, V1-LM: F(2,15) = 9.25, P = 0.002; V1-AL: F(2,15) = 7.90, P = 0.005; V1-PM: F(2,15) = 1.18, P = 0.334; *: P < 0.05.
Extended Data Fig. 6 Functional examination of V1-HVA projections by cortical removal of V1 at P5 and severing connections between V1 and HVA at P10.
a, Schematic of surgical lesion of V1 at P5. The V1 lesion site retinotopically corresponded to the seeds in HVAs. b, Time courses of V1 before and after V1 lesion. c, Representative FC maps of LM, AL, and PM seeds before and after V1 lesion. d, Time course of ROIs in the mouse in (a). e, Average FC matrix among V1 and HVAs before and after the V1 lesion (n = 5 mice). There was no significant difference between FCs before and after V1 lesion (all P values exceeded 0.05, Student’s t-test with Bonferroni correction). Statistics are shown in Supplementary tables. f, Schematic of the severing areal connection between V1 and lateral HVAs. g, Examples of spontaneous activity after cortical severing. h, Changes in frequency of activity in V1, LM, and AL. Wilcoxon signed-rank test with Bonferroni correction, V1: P = 1.0, LM: P = 1.0, AL: P = 1.0. i, Changes in the frequency of synchronous activity between V1 and HVAs. *: P < 0.05. Wilcoxon signed-rank test with Bonferroni correction, V1-LM: P = 0.019, V1-AL: P = 0.284, LM-AL: P = 1.0. P values were Bonferroni corrected.
Extended Data Fig. 7 Retrograde tracing of projections from the thalamic nuclei to HVAs.
a, b, Coronal brain slices show retrogradely labelled neurons in the thalamic nuclei by DiI injection into ventral (a, cyan) and dorsal HVAs (b, magenta). The gray in images indicate GCaMP signal. To perform wide-field imaging before DiI injection to determine locations of HVAs based on FC maps, GCaMP was expressed in cortical neurons and axons. Each slice was fitted to the 3D coronal adult brain atlas of the Allen Institute (n = 3 and 4 mice in ventral and dorsal HVAs, respectively). Retrogradely labelled neurons in thalamic nuclei were plotted in reference images in the bottom rows of brain slices. Brain slices from the posterior to anterior axis are arranged from left to right. The distance between slices was 100 µm. The white line indicates the dLGN. In All HVAs, retrogradely labelled neurons in LPN were much more numerous than in dLGN. *: P < 0.05. Paired t-test with Bonferroni correction, LM: t(2) = –26.46, P = 0.011; LI: t(2) = –19.7, P = 0.021; POR: t(2) = –13.91, P = 0.041; AL: t(3) = –7.27, P = 0.043; RL: t(3) = –6.86, P = 0.051; PM: t(3) = –11.16, P = 0.012; AM: t(3) = –10.31, P = 0.016; P values were Bonferroni corrected. These panels show that neurons in the posterior/dorsal LPN send projections to ventral HVAs, whereas those in the anterior/ventral LPN send projections to dorsal HVAs. These labelled structures are consistent with those observed in adults8. c, d, DiI and DiD were injected into two HVAs of the same stream [AL and RL (c) or LI and POR (d)]. Left panels of each image show retrogradely labelled neurons in LPN. DiI and DiD labelled regions in LPN mostly overlapped. White lines indicate the boundary between dLGN and LPN. Right panels of each image in c show the double labelled neurons. Only a very small number of double-positive neurons was observed in LPN.
Extended Data Fig. 8 Retrograde and anterograde tracing of connections between V1 and thalamic nuclei.
a, Schematic representation of DiI injection into V1. b, Coronal brain slices showing the retrogradely labelled neurons in the thalamic nuclei by DiI injection into V1. The gray in images indicate GCaMP signal. To perform wide-field imaging before DiI injection to determine locations of HVAs based on FC maps, GCaMP was expressed in cortical neurons and axons. Each slice was fitted to the 3D coronal adult brain atlas of the Allen Institute (n = 4 mice). Retrogradely labelled neurons in thalamic nuclei were plotted in reference images in the bottom rows of brain slices. The brain slices from the posterior to anterior axis are arranged from left to right. The distance between each slice is 100 µm. The white line indicates the dLGN. c, Percentage of labelled neurons in dLGN or LPN in total labelled thalamic neurons within the brain section images. DiI labelled neurons in dLGN were significantly higher than that in LPN. *: P < 0.05. Student’s t-test with Bonferroni correction, t(3) = 27.48, P = 8.46 × 10–4, P value was Bonferroni corrected, including the tests in Extended Data Fig. 7a, b. Schematic of the experimental procedure for labelling projections from V1 to thalamic nuclei. e, f, Coronal sections at P5 (e) and P10 (f) showing V1 axons in the visual thalamic nuclei (dLGN, LPN, and vLGN). We found that although V1 already sent projections to dLGN, vLGN, and LPN at P5, the axon density was relatively weak and became strong at P10. In addition, spontaneous activity of HVAs at P5 did not change after surgical lesion of V1 (Extended Data Fig. 6), suggesting that spontaneous activity did not propagate through V1-LPN-HVA pathways.
Extended Data Fig. 9 Anterograde tracing of projections from SC to LPN and the pharmacological inactivation of SC.
a, Histological confirmation of tdTomato-expressing neurons in SC. b, Anterograde tracing of projections from SC to thalamic nuclei. White line indicates the dLGN. Brain slices from posterior to anterior axis are arranged from left to right. c, Population data of tdTomato-expressing projection from SC to LPN (n = 4 mice). The data were aligned to the reference images from the Allen 3D Brain Atlas in adults. We extracted the projections from SC to LPN and plotted on reference images. Warm coloured lines show LPN regions that received projections from SC in individual mice. SC sent projections to a posterior part of LPN. d, Experimental schema of injection of muscimol or TTX. e, Representative frequency maps of spontaneous activity before and after the SC inactivation. Red line indicates the V1 region. f, Time course of spontaneous activity in V1 and HVAs. g, Population data of the change ratio of activity frequency (n = 6 mice for each condition). Open circles mean the TTX injection data. *: p < 0.05. Student’s t-test with Bonferroni correction, V1: t(10) = –1.74, P = 0.902; POR: t(9) = –6.46, P = 0.001; LI: t(10) = –3.73, P = 0.031; LM: t(9) = –4.45, P = 0.013; PM: t(9) = –4.19, P = 0.019; AL: t(10) = –2.28, P = 0.365; AM: t(10) = –1.71, P = 0.944; RL: t(9) = –3.52, P = 0.052; P values were Bonferroni corrected.
Extended Data Fig. 10 Retrogradely labelled neurons in subcortical regions by DiI injection into aLPN.
a, Illustration of the brain and observation sites. The numbered lines indicate the positions of slices shown below. Right panel indicates the injection site included in slice #4. The gray in this image indicates GCaMP signal. To perform wide-field imaging before DiI injection to determine locations of LPN based on FC maps, GCaMP was expressed in thalamic neurons. b, Representative slices include the retrogradely labelled neurons. The brain slices from posterior to anterior axis are arranged from left to right. Bottom panels show retrogradely labelled regions by DiI injections into aLPN. The magenta square in slice #4 indicates the DiI injection site. SC: superior colliculus, APN: anterior pretectal nucleus, MGV: ventral part of medial geniculate complex, PO: Posterior complex, VPM: ventral posteromedial nucleus, TRN: thalamic reticular nucleus, LDN: lateral dorsal nucleus, CP: caudoputamen. c, Enlarged views of DiI labelled regions numbered in (b). In pLPN, vLGN, PO, LDN, and TRN (squares #3–7), several retrogradely labelled neurons were observed. In SC and APN (squares #1 and #2), DiI signal was observed in neurites, but not cells. d, Ratio of DiI labelled neurons in each nucleus of the total number of DiI labelled neurons in these areas (n = 3 mice). e, Schema of projections from subcortical regions to aLPN at P5. Although DiI labelled neurons were observed in a deep (medial and ventral) part of vLGN, this domain does not receive direct retinal innervation, and vLGN mainly consists of inhibitory neurons55. Thus, vLGN unlikely relays the retinal activity to aLPN. Because TRN is inhibitory thalamic area and PO is the higher-order thalamic nucleus in the somatosensory system, TRN and PO are also unlikely to send retinal activity to aLPN. LDN receives projections from other visual thalamic nuclei (dLGN, LPN, and vLGN) in adults (Allen Brain Atlas, Mouse Connectivity; https://connectivity.brain-map.org/). Although it is still possible that the LDN propagates retinal activity to aLPN, it is unclear whether these connections are formed by P5. Therefore, we consider that the pLPN is likely a relay point for retinal activity propagation to the aLPN. Spontaneous activity in dorsal HVAs at P5 were reduced to 39.0 ± 35.6 % and 58.0 ± 36.2 % by BE and SC inactivation, respectively (Extended Data Fig. 2 and 9), suggesting that retinal activity propagated to the aLPN largely through SC. Because the SC sent projections to the pLPN but not to the aLPN (Extended Data Fig. 9), the pLPN likely relayed retinal activity to the aLPN and dorsal HVAs.
Extended Data Fig. 11 Simultaneous imaging of the spontaneous activity of the cortical neurons and axonal projections from thalamic nuclei to the cortex.
a, Brain sections showing the thalamic neurons expressing tdTomato. The white line indicates the dLGN. b, The brain section including the visual cortex shows the axon clusters in the cortex. Right panels are enlarged views of the axon clusters. c, Schematics of the experimental procedure and gene information coded in AAVs. d, Cortical slices showing the GCaMP expressing neurons in the thalamic nuclei (1), and jRGECO1a expressing cortical neurons and GCaMP7b expressing thalamocortical axons in LM (2). e, Representative activity of cortical neurons and thalamocortical axons in LI observed using two-photon imaging. f, Time courses of the activity of cortical neurons and thalamocortical axons in V1 and HVAs. Each data was obtained in different scans. g, Correlation coefficients of the spontaneous activity between cortical neurons and thalamocortical axons in V1 and HVAs.
Extended Data Fig. 12 Spontaneous activity was altered by chronic removal of the retinal activity, but anatomical projection patterns from LPN to HVAs remained.
a, DiI and DiD injections into two sites in LM of BE0 and control mice at P10. Different subregions in LPN were labelled by injection into two sites of LM in both BE0 and control mice. This result suggests that retinotopic projections from LPN to LM were not largely affected by BE0. b, DiI and DiD injections into ventral and dorsal HVAs of a BE0 mouse. Injection sites were determined based on the FC map. Because the FC map in BE0 mouse were broader than control mice, we could not distinguish between LM and LI, and between AL and RL. Therefore, we injected DiI and DiD into their intermediate regions. As shown in Fig. 3, the posterior part of the LPN was labelled by injection into the putative ventral HVA region, whereas the anterior LPN was labelled by injection into the putative dorsal HVA region. This result suggests that pLPN-ventral HVA and aLPN-dorsal HVA pathways mostly remained in a BE0 mouse. c, Schematics of the experimental procedure. d, Example of spontaneous activity in a BE0 mouse at P5. e, f, Comparison of the size (e) and frequency (f) of spontaneous activity between control and BE0 mice (n = 7 and 6 mice in control and BE0 conditions, respectively). Student’s t-test, size: t(11) = –11.37, P = 4.05 × 10–7; frequency, t(11) = 0.85, P = 0.82. P values were Bonferroni corrected.
Extended Data Fig. 13 Possible mechanisms for decorrelation of spontaneous activity in parallel visual pathways at P5.
a, Decorrelation of retinal activity propagation suggested by this study. Projections from LPN to HVAs are formed by P5 but not from V1 to HVAs. Low transfer reliability in the SC-pLPN pathway reduced the activity correlation between dLGN and pLPN, further between V1 and ventral HVA. In addition, SC inactivation experiment suggested the existence of non-retinal activity in aLPN, which may decorrelate spontaneous activity between ventral and dorsal HVAs. b, c, Possible visual pathways in the sequential scenario based on the FC analysis, in which FCs between V1 and HVA were lower than those between HVAs at P5 (Fig. 1). (b) First possibility is that V1-HVA projections are formed around the same time as LPN-HVA projections and low transfer reliability between dLGN and V1 decorrelates spontaneous activity between V1 and HVAs. (c) Second possibility in the sequential scenario is that the transfer reliability from V1 to HVAs is low and the projection from the LPN to HVAs is not yet formed. By integrating results based on both anatomical and other functional experiments, we consider that both possibilities (b, c) are unlikely. Simultaneous imaging of spontaneous activities of thalamocortical axons and cortical neurons showed a high activity correlation between thalamic axons and cortical neurons in V1 and HVAs (Extended Data Fig. 11), indicating that the transfer reliability from the thalamic nuclei to the cortex is very high. Therefore, the possibility in (b) is unlikely. Anatomical tracing of V1 projections, electrical stimulation of V1, and V1 lesion experiments in this study have shown that V1-HVA projections are not effective at P5. On the other hand, simultaneous imaging of LPN-HVA projections and cortical neurons in HVAs (Extended Data Fig. 11) and pharmacological LPN inactivation (Fig. 5) showed that projections from the LPN to HVAs are already effective at P5. These results indicate the earlier formation of LPN-HVA pathways than the V1-HVA pathway, which is inconsistent with the possibility in (c).
Supplementary information
Supplementary Tables
Supplementary Tables 1–7.
Supplementary Video 1
Spontaneous activity before EO in the entire visual cortex. These examples represent, from left to right, spontaneous activity at P5, P8, P11 and P14. Red dashed lines indicate the V1 region. Spontaneous activity occurred in local spot-like patterns at P5–P11. This video is played at 5× speed.
Supplementary Video 2
Spontaneous activity at P5 before and after binocular enucleation. Left and right examples are before and after binocular enucleation, respectively. Red dashed lines indicate the V1 region. Spontaneous activities of V1 and HVAs decreased after binocular enucleation at P5. This video is played at 5× speed.
Supplementary Video 3
Spontaneous activity at P5 before and after V1 lesion. Left and right examples are before and after V1 lesion, respectively. Red dashed lines indicate the V1 region. The blue circle indicates the lesion site in V1. Even after V1 lesion, synchronous activity remained in HVAs, retinotopically corresponding to the lesion site. This video is played at 5× speed.
Supplementary Video 4
Spontaneous activity in thalamic nuclei at P5. This video shows activity in the same mouse as in Fig. 4b. The cortex above the thalamic nuclei was removed, and the spontaneous activity of the thalamic nuclei was directly observed using wide-field imaging. The red dashed line indicates the dLGN region. Wave-like activity propagating within dLGN was observed. This video is played at 5× speed.
Supplementary Video 5
Spontaneous activities of thalamocortical axons and cortical neurons at P5 observed simultaneously using two-photon imaging. Cortical neurons and thalamocortical axons expressing jRGECO1a and GCaMP7b, respectively, were observed in area LI (depth of 250 µm from the pia). Most cortical activities occurred synchronously with those of thalamic axons at P5. This video is played at 5× speed.
Supplementary Video 6
Spontaneous activity of thalamocortical axons at P5 observed using wide-field imaging. This video shows activity in the same mouse as in Fig. 4i. The red dashed line indicates the V1 region. Spontaneous activity of the thalamocortical axons at P5 was a local spot-like pattern similar to that of cortical neurons. This video is played at 5× speed.
Supplementary Video 7
Spontaneous activity at P5 before and after muscimol injection into LPN. Left and right examples are before and after muscimol injection into LPN, respectively. Red dashed lines indicate the V1 region. Spontaneous activity of the ventral and dorsal HVAs was sharply decreased by LPN inactivation. This video is played at 5× speed.
Source data
Rights and permissions
About this article
Cite this article
Murakami, T., Matsui, T., Uemura, M. et al. Modular strategy for development of the hierarchical visual network in mice. Nature 608, 578–585 (2022). https://doi.org/10.1038/s41586-022-05045-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-022-05045-w
This article is cited by
-
How deep is the brain? The shallow brain hypothesis
Nature Reviews Neuroscience (2023)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
