Caitlin R Siu1, Kathryn M Murphy1,2. 1. McMaster Integrative Neuroscience Discovery and Study (MiNDS) Program, McMaster University, Hamilton, ON, Canada. 2. Department of Psychology, Neuroscience & Behaviour, McMaster University, Hamilton, ON, Canada.
Abstract
The primary visual cortex (V1) is the first cortical area that processes visual information. Normal development of V1 depends on binocular vision during the critical period, and age-related losses of vision are linked with neurobiological changes in V1. Animal studies have provided important details about the neurobiological mechanisms in V1 that support normal vision or are changed by visual diseases. There is very little information, however, about those neurobiological mechanisms in human V1. That lack of information has hampered the translation of biologically inspired treatments from preclinical models to effective clinical treatments. We have studied human V1 to characterize the expression of neurobiological mechanisms that regulate visual perception and neuroplasticity. We have identified five stages of development for human V1 that start in infancy and continue across the life span. Here, we describe these stages, compare them with visual and anatomical milestones, and discuss implications for translating treatments for visual disorders that depend on neuroplasticity of V1 function.
The primary visual cortex (V1) is the first cortical area that processes visual information. Normal development of V1 depends on binocular vision during the critical period, and age-related losses of vision are linked with neurobiological changes in V1. Animal studies have provided important details about the neurobiological mechanisms in V1 that support normal vision or are changed by visual diseases. There is very little information, however, about those neurobiological mechanisms in human V1. That lack of information has hampered the translation of biologically inspired treatments from preclinical models to effective clinical treatments. We have studied human V1 to characterize the expression of neurobiological mechanisms that regulate visual perception and neuroplasticity. We have identified five stages of development for human V1 that start in infancy and continue across the life span. Here, we describe these stages, compare them with visual and anatomical milestones, and discuss implications for translating treatments for visual disorders that depend on neuroplasticity of V1 function.
The human brain has >20 cortical areas that receive strong visually driven activity and process that information to support all aspects of our visual perceptions. Changes in any of those cortical areas can affect visual perception, and abnormal visual experience, especially in childhood, often disrupts the maturation of visual cortical circuits causing poor vision. The role of the visual cortex in processing visual perception and plasticity has been well studied in animal models,1–9 but there are few studies about the neurobiology of human visual cortex10–19 and even fewer examine how it develops and changes across the life span.20–24 Brain imaging studies are beginning to address structural and functional development of the human cortex,25 but the lack of information about cellular and molecular mechanisms has slowed the translation of biologically inspired treatments for visual disorders.Over the past decade, our laboratory has focused on studying the neurobiology of human visual cortex by measuring the expression of molecular markers that regulate neural function and plasticity, and characterizing a series of neurobiological milestones. Perhaps the most striking finding from our studies has been the prolonged development of those markers in human primary visual cortex (V1). We have found that development of the human V120–22,24 mirrors the long process of visual maturation and age-related changes in perception.26–29 In this review, we will focus on the five stages that we identified for human V1 and link them with visual and anatomical milestones. Finally, we will discuss the implications of these new findings for translation of biologically inspired neuroplasticity-based therapeutic approaches for treating visual disorders.To study the development of human visual cortex, we chose to measure the expression of synaptic and non-synaptic proteins because these mechanisms link structure and function.20–22,24 Imagine a Venn diagram with the anatomical structure of V1 on one side and the physiological and visual functions on the other side. The neural proteins sit at the interface joining structure with function, they regulate how synapses and circuits develop, they respond to plasticity, and they control neural communication. Furthermore, measuring neural proteins in postmortem tissue from human cortex is a robust methodology that provides high-quality reliable data about these rare and valuable human tissue samples.
Stage 1: the first year, early maturation of vision and the structure of V1 neurobiology
Visual milestones
Early visual development is characterized by progressive improvements in functions such as acuity,30,31 contrast sensitivity,32 orientation selectivity,33 and motion sensitivity.34 None of those visual abilities, however, attain adult levels at this early stage. In contrast, binocular functions such as fusion, stereopsis, and stereoacuity emerge abruptly around 3 months of age.35 By 2 months of age, infants can discriminate some color from white light,36 and by 3 months evidence for trichromacy emerges.37,38 Infants develop the ability to individuate objects by shape and size by 4.5 months,39 while the ability to integrate contours or edges emerges later around 6 months.40,41 By 5 months most infants have fusion and stereopsis, followed by rapid development to reach adult levels by 6–7 months of age;35 meanwhile, the development of spatial acuity continues to improve well past infancy (Figure 1).31,42 Normal time course for the development of spatial visual functions is not preprogrammed but instead is experience-dependent and abnormal vision can have a profound effect on the maturation of these functions.43,44 Infancy marks the onset of the sensitive period for developing amblyopia, as the average age of diagnosis is about 1.2 years.45 Early treatment of cataracts in infancy shows rapid improvement in visual acuity even within 1 hour of cataract removal.46 Many studies and clinical experience have shown that early treatment for amblyopia, even starting in infancy, improves the chance of developing normal acuity.47
Figure 1
Summary chart for development of human visual milestones.
Notes: A summary of the development of key visual perceptual milestones across the life span. The top panel shows the stages of human development (infants, young children, older children, teens, young adults, older adults), and associated ages in months and years (as presented by Siu et al22). The rows below illustrate the approximate timing of onset and emergence (green arrows), adult-like levels (gray shade with black arrows), and loss of function (red arrows). References linked to each milestone are provided in the right column.
Many of the anatomical features of human V1 develop prenatally. Neurogenesis begins around embryonic day 33 and is complete by birth,48–50 while the thalamic input to layer 4 in V1,51,52 bipolar, and pyramidal cells with long and thin dendritic spines forms distinct laminar patterns at around 20–30 weeks gestation.16,53 Other aspects of cortical development that begin prenatally continue to mature after birth. Cytochrome oxidase expression is present at 26 weeks gestation, and is organized into clearly visible “puffs” by 24 days postnatal, and becomes well organized by 4 months postnatal.19 Vertical interlaminar connections form between 26 and 29 weeks gestation, while long-range horizontal connections in layers 4B and 5 emerge at around 37 weeks gestation, and show adult-like patchiness by 8 weeks postnatal. Layer 2/3 horizontal connections emerge later at around 16 weeks postnatal and become adult-like by 15 months of age.15 By 4 months of age, feedforward connections from V1 to extrastriate area V2 have formed mature connections, while feedback connections are still immature only reaching adult levels at around 2 years of age.14 Synaptogenesis in human V1 increases to reach a peak between 8 months and 2 years and is followed by a longer period of synaptic pruning to reach adult levels later in childhood.13 The number of dendritic spines in V1 follows a similar trajectory that peaks at around 5 months of age and then decreases to adult levels by 2 years.54 Many anatomical features are already adult-like by the end of this stage (Figure 2); however, vision continues to mature well beyond the first year of life.
Figure 2
Summary chart for development of human V1 anatomical milestones.
Notes: A summary of the development of key neuroanatomical milestones in human V1 across the life span. The top panel shows the stages of human development (prenatal, infants, young children, older children, teens, young adults, older adults), and associated ages in months and years (as presented by Siu et al22). The rows below illustrate the approximate timing of onset and emergence (green arrows), adult-like levels and structure (gray shade with black arrows), and loss of expression (red arrows). Black dots refer to anatomical milestones that are completed before birth. References linked to each milestone are provided in the right column.
We have found that the first stage of human V1 development is characterized by rapid changes in neurobiological mechanisms that will support the emergence of visual function and synaptic plasticity. There are some early changes to both excitatory glutamatergic and inhibitory gamma-aminobutyric acid (GABA)ergic synaptic receptors (Figure 3),22,24 and a shift toward a balance between these excitatory and inhibitory (E-I) receptors.20 The immature GABAA receptors subunits, GABAAα2 and GABAAα3, dominate expression in the first year, but quickly show signs of maturation as those subunits are replaced by GABAAα1 (Figure 4) so that there is relatively more GABAAα1 by about 1 year of age, that peaks later in adolescence (Figure 3).24 For glutamatergic synapses, the N-methyl-d-aspartate (NMDA) receptor subunit GluN1 is highly expressed at birth, then rapidly declines to reach adult levels at around 1 year. That loss is balanced by an increase in GluA2 containing AMPA receptor (AMPAR) expression, and the shift from more NMDA to more AMPA signals the loss of NMDA-dominated silent glutamatergic synapses to be replaced by active AMPA containing synapses (Figure 3).22 The maturation of both GABAA receptor subunits and the AMPA:NMDA balance speed up responses at those receptors and trigger an environment that supports experience-dependent plasticity (Figure 4).55–57 For example, the maturation of GABAA receptor regulates the critical period for plasticity,56 as the mature α1 subunit is necessary for ocular dominance plasticity.55 Moreover, the insertion of AMPAR is driven by visual experience and is an important step in initiating the critical period.58
Figure 3
Summary chart for development of human V1 neurobiological milestones.
Notes: A summary of the development of key neurobiological milestones in human V1 across the life span. The top panel shows the stages of human development (infants, young children, older children, teens, young adults, older adults), and associated ages in months and years (as presented by Siu et al22). The rows below illustrate the approximate timing of onset and emergence (green arrows), peak expression (gray shades), adult-like levels (gray shade with black arrows), and loss of expression (red arrows). References linked to each milestone are provided in the right column.
Summary chart of glutamatergic and GABAergic receptor subunits.
Notes: This figure presents a summary of some key glutamate (AMPAR and NMDAR) and GABA (GABAA) receptor subunit compositions that regulate neuroplasticity in the primary visual cortex. The columns represent functional significance of the balance of NMDA:AMPA (top), GluN2A:GluN2B (middle), and GABAAα1:GABAAα3 (bottom). More juvenile synapses are dominated by more NMDAR, GluN2B containing NMDAR, and GABAAα3 containing GABAA receptors that allow for LTP in excitatory synapses127 and slower kinetics through the receptors. More mature synapses are dominated by more AMPAR, GluN2A containing NMDAR, and GABAAα1 containing GABAA receptors that allow for more LTD in excitatory synapses,127 and faster kinetics through the receptors.
Stage 2: preschool children have high variability in V1 development (1–4 years)
Many aspects of visual perception continue to improve through the first few years of development (Figure 1). Young children have experience-dependent improvements in visual acuity,42 biological motion perception,59 and contrast sensitivity,60,61 but those abilities are still not adult-like.62 During the first 2 years of this stage (~1–3 years), children are most susceptible to abnormal binocular vision63 that can cause amblyopia. Alternatively, if abnormal vision is identified and treated in children under 5 years of age, there is the greatest likelihood for recovery,64 for both high and low spatial frequencies.65 For example, binocular iPad treatment for amblyopia shows improvement in visual acuity at this stage for amblyopic children.66In young children, V1 undergoes synaptic and dendritic refinement to reach adult appearance at around 2 years of age (Figure 2).54 Other aspects of human cortical development are characterized as “adult-like” by this stage, including cortical thickness67 and the appearance of feedback connections from extrastriate areas to V1.14Although this period of experience-driven visual development points to significant increases in visual plasticity, we have found little evidence that neural plasticity mechanisms complete maturation during this stage (Figure 3). Instead, we found a novel aspect of human cortical development that is characterized by waves of high interindividual variability in the expression of neural plasticity markers in human V1.20–23 Interindividual variability in human V1 can be characterized across development using the variance-to-mean ratio of protein expression across a moving window of three age-adjacent cases. Using this, we found a period of high interindividual variability, or a “wave”, during young childhood for many of the synaptic proteins, but not during the other stages of development.22 This variability may signify either interindividual differences in the rate of development, or it may identify intraindividual fluctuations in the expression of plasticity mechanisms.22 Nevertheless, this large dynamic range in protein expression likely contributes to increased plasticity or learning for optimal behavioral performance.68 Interestingly, this stage of interindividual variability comes just after the E-I balance has been reached in human V1.20 Balanced excitation and inhibition in the cortex establish cortical criticality, defined as a dynamic range of spontaneous activity that maximizes the processing of input activity.69 Thus, the waves of variability in cortical development may be an important stage of development when visual circuits “learn” complex processing by using the variability to fine-tune optimal neural circuits.22,70
Stage 3: experience-dependent visual development in school aged children (5–11 years)
Many visual abilities finish maturation in older children (Figure 1). However, the precise age of maturation may vary significantly, and usually, depends on the type of measure used to assess vision, or the type of patterned visual input that a child has experienced.71 For example, visual acuity can mature between 5 and 15 years, while contrast sensitivity can mature anywhere between 6 and 19 years of age.26,72 While some studies suggest that motion perception can mature in older children,73–76 others suggest that it continues to improve beyond childhood into adolescence and adulthood.34,77–79Children aged 6–10 years are just beyond the period of susceptibility for developing amblyopia65,80,81 that is associated with the end of the critical period for ocular dominance plasticity in animal models.20 Despite this, there is evidence for significant visual plasticity at this stage in both children with amblyopia and children with normal vision.82–84 The neurobiological mechanisms that regulate this experience-dependent plasticity, including triggers proteins that promote neuroplasticity, and brakes that limit it are well studied in animal models.5,85,86The expression of some neural plasticity mechanisms peaks during this stage of development (Figure 3). These include peaks in the expression of the glutamatergic receptor scaffolding protein PSD-95 and AMPA receptor subunit GluA2.22 Peaks in the expression for both of these proteins have been linked with ending the critical period for experience-dependent plasticity in the visual cortex.58,87 The maturation of those proteins is experience-dependent87 and contributes to stabilizing synapses.88 Furthermore, GluA2 is necessary for a form of plasticity, homeostatic synaptic scaling, which regulates synaptic strength over fluctuations in synaptic activity.89 The homeostatic scaling up or down of AMPAR expression is dependent on synaptic activity and cooperates with NMDA-dependent Hebbian plasticity to refine cortical connectivity and promote synaptic stability during the development of V1.90–92In human V1, expression of the GABA receptor scaffolding protein gephyrin also matures during late childhood development (Figure 3).20,24 Gephyrin is directly related to the strength and stability of inhibitory synapses93 and this peak suggests a developmental balance between excitatory (eg, PSD-95, GluA2) and inhibitory synaptic mechanisms.20 It is interesting to note that the AMPAR subunit GluA2 is highly expressed on parvalbumin-positive (PV+) inhibitory interneurons,94 and PV+ cell activity regulates critical period plasticity.95 It will be important for future experiments to address the development of PV+ inhibitory interneurons in human V1 to fully understand the maturation of these plasticity mechanisms during this important stage of childhood visual development.
Stage 4: prolonged visual development in adolescence and adulthood (12–55 years)
A series of studies characterizing visual development have shown that “higher-order” visual abilities continue to mature through the teen and young adult years (Figure 1). For example, global and biological motion34,77,96–98 and spatial integration of contours27 mature during adolescence (eg, 14–15 years of age). Face perception has an even slower pace of maturation, with continuous improvements into adulthood, as face learning and recognition improve into the third decade of life.29,99,100 Expertise in face perception depends on visual experience, and abnormal early vision has a “sleeper effect” on the development of the neural circuits and perceptual processing that support normal face perception.101–103Children older than 7 years of age are less responsive to amblyopia treatment,104 but some forms of treatment may be effective in teens and adults and suggest that plasticity persists in the visual cortex.105–107 For example, perceptual training for low-level perceptual abilities like contrast sensitivity and letter-recognition can improve the vision of some amblyopic patients.108 These training-induced improvements are often small and not clinically significant.109 Perceptual learning studies have shown that there is plasticity in adulthood that can support recovery from amblyopia.110 Adults with amblyopia can improve visual acuity with extensive perceptual training,111 and succeed in refining contrast sensitivity,83 orientation selectivity,112 stereopsis,113 spatial discrimination,114 and face learning.115,116Structural imaging studies of humans show that intracortical myelin in the visual cortex continues to increase well into adulthood, peaking between 30 and 40 years of age (Figure 2).117 Anatomical analysis of postmortem human visual cortex also indicates the prolonged development of cortical myelination that continues into the third decade of life.118 Gray matter density mapping shows a slow linear decline of cortical thickness with age in the occipital cortex,119,120 and that regression appears to mature sequentially across the cortical areas, with primary areas maturing before higher-order association areas.121Our studies of neuroplasticity mechanisms in human V1 provide new evidence that many aspects of V1 continue to develop into the adult years (Figure 3). Our findings include measures of myelin expression as a brake on plasticity classic-myelin basic protein [MBP]),122 and of the balance between NMDA receptor subunits 2A and 2B that supports plasticity when 2B dominates and reduces plasticity when 2A dominates (2A:2B) (Figure 4).123–127 Myelin expression peaks in young adults21 and the shift from more 2B to more 2A also ends at around 35 years of age.22 Some GABAergic proteins also continue developing into adulthood. The enzyme that makes the on-demand pool of GABA (GAD65) and the GABAAα1 receptor subunit continues to increase into adulthood (Figure 3).24 Each of these late maturing mechanisms is important for regulating neural transmission and plasticity. For example, the bidirectional regulation of the 2A:2B balance can facilitate plasticity when 2B is favored, or reduce experience-dependent plasticity when 2A is favored.123–127 All of these findings show that the progressive shift in human V1 from a very plastic environment in childhood to a less plastic environment in adults continues into the third decade of life (Figure 3). That pace of neurobiological development is much slower than predicted by vision studies or animal research.20–22,24The very slow maturation of human V1 may keep a sliver of the plasticity window open that both normal visual development and some types of vision treatments can use.
Stage 5: loss of plasticity mechanisms during aging of human V1 (>55 years)
Certain visual losses in aging have been interpreted as part of normal aging that changes the receptive field properties of neurons in V1 (Figure 1).28 During normal aging, there is an increase in the population receptive field size for neurons in V1 and in the extrastriate area V2 that serve the foveal representations.128 These neural changes and others contribute to age-related losses of low-level visual functions like visual acuity,129 contrast sensitivity,130 and orientation selectivity.131 In addition, there are age-related losses for many higher-order visual perceptions,132 including face perception,99,133–135 motion processing,136–140 and reading speed.28There are also acquired causes for age-related vision loss that include diseases such as diabetic retinopathy, macular degeneration, cataracts, and glaucoma. All of these diseases affect the eye and either directly or indirectly reduce retinal functioning either because the cataract has degraded the image or the disease has caused degeneration of the retina.141 These retinal changes impact the information that is transmitted to the central visual pathway and many studies have shown changes in the visual areas of the brain.142 Often, these eye diseases are described as neurodegeneration spreading that starts in the eye and progresses to affect the visual cortex.143 Thus, it is likely that vision changes in normal aging and adult-acquired visual diseases may involve neurodegenerative processes in V1 in addition to optical changes.Aging in the visual cortex is characterized by specific microstructural changes (Figure 2). These include significant changes in the morphology of pyramidal cell bodies in human V1, a loss of dendrite number, and reduced complexity of the dendritic arborizations.10 In primate V1, aging changes the process of cortical demyelination and remyelination so that remyelinated axons have shorter segments and the myelin sheaths are less tightly compacted around the axons thereby affecting the efficiency of axonal conduction.144–146 In human V1, there is a progressive loss of intracortical myelin content in the stria of Gennari that begins around 30 years of age and continues to decline into the late 90s.147 Some animal studies have also found a loss of intracortical inhibition in V1 that leads to poor orientation selectivity,148–150 and treating V1 of old monkeys with the neurotransmitter GABA sharpens orientation selectivity of V1 neurons so they are similar to the selectivity found in young adults.150Our studies of human V1 have found that the expression of many synaptic and non-synaptic proteins decreases in older adults (Figure 3). There are losses for GAD65, the enzyme that makes the on-demand pool of GABA,24 Ube3A, an E3 ubiquitin ligase that is necessary for experience-dependent plasticity,23 classic-MBP that is necessary for normal axonal myelination,21 and GluN2A, the mature subunit of the NMDA receptor that regulates certain forms of plasticity.22 Not all synaptic proteins change on aging; for example, expression of the GABAA receptor subunit does not decline.24 Interestingly, most of the proteins that do change shift toward the more juvenile-like partner. For example, the age-related loss of GluN2A shifts the 2A:2B balance toward more 2B, and that may reinstate a more plastic environment.127 Also, the loss of classic-MBP shifts the composition of MBP to favor the immature oligodendrocyte protein Golli-MBP21 which can give rise to various developmental regulated isoforms of MBP.151 These shifts toward more juvenile-like neural proteins raise the possibility that there is support for a more plastic environment in aging. However, not all of the changes point to greater plasticity since there is a loss of Ube3A, a protein that is necessary for experience-dependent plasticity in the visual system.152 Perhaps, the age-related losses of these important neural proteins reflect an internal drive to maintain stable functioning of visual cortical circuits in the face of degraded inputs due to optical changes153,154 and neurodegenerative processes.143
Conclusions
We have demonstrated lifelong changes in the neurobiological mechanisms found in human V1 that support neural development, plasticity, and processing of visual information. These changes can be described in five stages: very early establishment of mechanisms for E-I transmission; a novel stage of variability in young children; maturation of low-level mechanisms in older children; continued fine-tuning through teens and young adults; and age-related losses. How far these five stages of V1 maturation will generalize across the 20 cortical areas that process visual information remains unknown. This is an important question to address, especially for developing new cortically inspired treatments for adult-acquired vision loss, since plasticity in the extrastriate area may prove to be important for supporting maintenance or recovery of vision caused by a retinal disease.The first three figures summarize key milestones for visual system development and illustrate compelling similarities between the timing of visual, anatomical, and neurobiological milestones in human V1. Tapping into these neurobiological mechanisms is going to be key for the next generation of treatments for visual disorders. For example, a wide range of potential new therapies has been developed in animal models for amblyopia. The treatments include everything from fine-tuning of traditional patching therapy155,156 to drug treatments,157 to novel visual stimulation paradigms158 and visual environments.159–161 Also, it is likely that normal age-related changes in human V1 interact with the spreading neurodegeneration caused by diseases like glaucoma. In contrast to the excitement from preclinical models, no new plasticity-based treatments have crossed the chasm into clinical practice. One of the impediments has been the lack of information about neurobiological mechanisms in the human visual cortex, but our studies are beginning to fill that gap. Although our approach to studying neurobiology in human V1 does not provide information about circuitry, synaptic function, cellular type, or laminar localization, these data are valuable as the first steps for identifying neurobiological mechanisms that underlie visual perception and plasticity in humans. In addition, these data will help to guide future human and animal studies by making it easier to make more direct links between the neurobiological developments of V1 in humans and animal models, that can pave the way for the translation of biologically inspired vision treatments.
Authors: Alison Gunn; Elizabeth Cory; Janette Atkinson; Oliver Braddick; John Wattam-Bell; Andrea Guzzetta; Giovanni Cioni Journal: Neuroreport Date: 2002-05-07 Impact factor: 1.837
Authors: Gábor Jandó; Eszter Mikó-Baráth; Katalin Markó; Katalin Hollódy; Béla Török; Ilona Kovacs Journal: Proc Natl Acad Sci U S A Date: 2012-06-18 Impact factor: 11.205
Authors: Giovanni Granato; Emilio Cartoni; Federico Da Rold; Andrea Mattera; Gianluca Baldassarre Journal: PLoS One Date: 2022-05-10 Impact factor: 3.752
Authors: Sendy Caffarra; Sung Jun Joo; David Bloom; John Kruper; Ariel Rokem; Jason D Yeatman Journal: Hum Brain Mapp Date: 2021-09-06 Impact factor: 5.038