As discussed in the preceding chapters, it is now very clear now that the adult vertebrate retina undergoes marked modifications in its synaptic and cellular architecture, primarily in response to pathological states. Remarkable gains have been obtained towards understanding the molecular, cellular and biochemical events underlying retinal rewiring in a number of experimental paradigms, as described in detail in Chapters 2, 3 and 4. A major question in the field of retinal plasticity, however, is the extent of circuit reorganization that occurs in response to normal visual experience in the retina. Pioneering experiments conducted between the late 1950's and early 1970's addressed this particular issue by investigating the effects of light deprivation and stimulation on the organization of the mammalian retina. One of the main advantages in using the paradigms of sensory deprivation and stimulation is that it is rarely associated with retinal cell damage. In 1958, Weiskrantz reported that cats that had been deprived of normal visual information (form-deprived) displayed a marked reduction in thickness of the inner plexiform layer (IPL) (Weiskrantz, 1958). A few years later, in 1961, Rasch and colleagues provided evidence for considerable modifications in retinal circuitry as a function of light deprivation: it was demonstrated that cats that had been dark-reared for an extended period of time also underwent a significant reduction in the thickness of the IPL. In addition, chimpanzees that had also been maintained in darkness exhibited marked retinal ganglion cell degeneration (Rasch et al., 1961).
Subsequent electron microscopic studies provided the first detailed account for the modifications in synaptic architecture observed in animals that had been light-deprived. This series of experiments provided evidence for significant increases in the number of amacrine cell synapses in the rat retina, particularly those between amacrine and ganglion cells, these possibly
R.Pinaud, L.A. Tremere, P. De Weerd(Eds.), Plasticity in the Visual System: From Genes to Circuits, 79-95, ©2006 Springer Science + Business Media, Inc.
accounting for the gross alterations observed in IPL thickness in response to light-deprivation (Sosula and Glow, 1971; Fifkova, 1972a, b; Chernenko and West, 1976). Together these early anatomical data provided the first set of evidence suggesting that the circuitry of the intact mammalian retina is plastic and its cytoarchitecture exhibits malleability and likely undergoes reorganization based on a previous history of activation. In addition, these findings suggest that visual experience might play an active role in triggering the refinement of retinal connectivity.
Although these early works provided evidence that visual experience actively influenced the organization of the normal, adult, retinal circuitry, it became a commonly accepted view to contemporary neuroscientists that the adult retina is capable of extremely limited, if any, plasticity (Chernenko and West, 1976). Central to this argument was the notion that high levels of plasticity, and consequently the high possibility of circuit rewiring, was inversely correlated to the degree of fidelity of information processing and transfer of any given population of neurons. Given that accurate representations of the visual world are required for appropriate computations by neural ensembles in higher cortical areas, this dogma gained strength despite a complete lack of experimental data that supported (or refuted) these claims.
The dogma that the normal adult retina displays extremely limited plasticity was the reason for our interest in this field. Our interest in retinal plasticity occurred indirectly, in unrelated sets of experiments studying the rat striate cortex. We were interested in investigating genes that might be involved in triggering the cascade of genomic events leading to circuit rewiring in the mammalian brain, in a paradigm known to induce widespread cortical plasticity: the enriched environment protocol (EE; details below). At that time, we had preliminary evidence indicating that two immediate early genes were likely involved in the induction of the early stages of plasticity in the brain. Based on the highly plastic nature of the mammalian brain, we reasoned that if these genes were truly involved in the induction of plasticity in the adult central nervous system (CNS), they would exhibit a differential expression pattern in highly plastic CNS regions (e.g., primary sensory areas), as compared to regions where plasticity is very limited (e.g., retina, according to the fidelity hypothesis discussed above). Thus, in order to support our claims regarding the involvement of these genes in the induction of plasticity, we decided that we would use the retina as an internal control for our data collected in the brain. Our working hypothesis was that while brain sites would exhibit high induction of these genes in response to exposure to the EE, the retina of the same animals analyzed would display low gene expression levels. To our surprise, the results observed did not support our working hypothesis, as it will be discussed below.
The experimental conditions termed "enriched environment" refers to the creation of a housing and/or play facility where animals are expected to undergo an increase in the complexity and, perhaps, relative rates of sensory processing. In addition to changes in brain mass, studies on the effects of EE in laboratory animals, such as rats, revealed marked phenotypical and neuro-chemical changes, which have been forwarded as the physical signs of CNS plasticity, associated with the chronic or acute exposure to a complex environment (for reviews see Rosenzweig et al., 1972; van Praag et al., 2000). For example, rearing animals in an EE increased the number of dendritic arborizations, soma volume, levels of neurotrophic factors and the neuron-to-glia ratio in all main sensory systems (Rosenzweig et al., 1972; Volkmar and Greenough, 1972; Globus et al., 1973; Falkenberg et al., 1992; van Praag et al., 2000). Furthermore, this procedure led to a marked enlargement of synaptic terminals, a phenomenon that has been associated with increased numbers of quanta and, therefore, increased neurotransmitter release per action potential (Rosenzweig et al., 1972; van Praag et al., 2000).
Our research group has strongly advocated the use of the EE paradigm to invoke behaviorally-associated CNS plasticity (Pinaud et al., 2001; Pinaud et al., 2002a; Pinaud, 2004). One of the main advantages of the EE setup, as compared to other paradigms used to investigate plasticity, such as sensory deafferentation, is the complete absence of injury. In addition, visual stimulation associated with EE exposure is not associated with dramatic changes in light intensity, but rather involve, theoretically, changes in the complexity of the visual environment.
Even though it is widely accepted that experiences in an EE induce vast morphological and neurochemical modifications in central neurons, only a few studies have specifically addressed what functional benefits may correlate with time spent in a complex visual environment. One of these studies, conducted by Prusky and colleagues, showed that visual acuity in mice was increased by 18% following rearing in an EE from birth (Prusky et al., 2000). This finding supports the idea that exposure of animals to a complex visual environment leads to modifications in CNS neurons aimed at improved sensory system performance.
Gene Expression and Rewiring: The IEGs NGFI-A and arc
Phenotypic changes are dependent on gene expression. Changes in cell structure and physical interactions between neurons have been postulated to depend on a series of biochemical signals that first lead to the expression of immediate early genes (IEGs) (Hughes and Dragunow, 1995; Herdegen and Leah, 1998; Pinaud, 2004; see also chapter 8). Often the expression of IEGs is induced quickly after the cell stimulation and, in all cases, induction of members of this class of genes does not depend on de-novo protein synthesis (Hughes and Dragunow, 1995; Kaczmarek and Chaudhuri, 1997; Herdegen and Leah, 1998). Therefore, it has been argued that inducible IEGs provide the most immediate genomic response in central neurons to sensory input (Pinaud et al., 2002a; Pinaud, 2004).
As stated above, we were interested in studying the expression of IEGs in response to EE exposure as members of this family of genes are likely involved in the early phases of the induction of plasticity in the CNS. Two IEGs became the focus of our research efforts: the nerve growth factor-induced gene A (NGFI-A; also known as zif-268; egr-1; krox-24 and zenk) (Milbrandt, 1987; Christy et al., 1988; Lemaire et al., 1988; Sukhatme et al., 1988; Mello et al., 1992) and the activity-regulated cytoskeletal gene (arc; also known as arg3.1) (Lyford et al., 1995). The protein products encoded by these two genes shared interesting, and presented complimentary, characteristics that placed them well-positioned to mediate key aspects of the induction of plasticity in response to a complex sensory environment. For example, both NGFI-A and arc are activity-dependent genes (Murphy et al., 1991; Worley et al., 1991; Lyford et al., 1995; Herdegen and Leah, 1998; Steward et al., 1998). Furthermore, their expression depends on the activation of the NMDA receptors, which have been involved in various aspects of synaptic reinforcement and facilitation in a number of experimental systems (Cole et al., 1989; Wisden et al., 1990; Kunizuka et al., 1999; Steward and Worley, 2001b). Despite these critical similarities between NGFI-A and arc, the protein products encoded by these genes plays dramatically different roles for cell physiology. NGFI-A is a zinc-finger transcription factor, whose expression is coupled to the activation of the ERK/mitogen-activated protein kinase (MAP) kinase pathway, and thus plays a role in the control of the expression of other genes that possess the NGFI-A binding domain contained within their promoters (Herdegen and Leah, 1998; Dziema et al., 2003) (see also Chapter 8). For instance, it has been previously demonstrated that the NGFI-A protein is involved in the transcriptional regulation of synapsin I and synapsin II genes, as well as the gene that encodes for synaptobrevin II (Thiel et al., 1994; Petersohn et al., 1995; Petersohn and Thiel, 1996) in in-vitro essays. These proteins have been demonstrated to play a direct role in neuro-transmitter release. Other candidate proteins putatively regulated by NGFI-A include neurotransmitter-gated ion channels, the monoamine oxidase B gene, the alpha-7 subunit of the nicotinic acetylcholine receptor, neurofilament and the adenosine 5'-triphosphate binding cassette, sub-family A, transporter 2 (ABCA2)(Pospelov et al., 1994; Carrasco-Serrano et al., 2000; Wong et al., 2002; Davis et al., 2003; Mello et al., 2004). Thus, NGFI-A is well positioned to affect the expression of genes that are involved not only in neurotransmitter release and membrane transport, but also other fundamental neuronal functions such as excitability. The activity-dependent IEG arc, unlike NGFI-A, does not encode a transcription factor, but rather a growth factor (Lyford et al., 1995). Interestingly, after transcribed, arc mRNA is readily transported to dendrites where local protein synthesis takes place (Lyford et al., 1995; Steward et al., 1998; Steward and Worley, 2001a, b). This interesting feature of arc expression has led to the suggestion that the protein encoded by this gene might be involved in activity-dependent dendritic reconfiguration, such as spine rotation, retraction or elongation, in addition to dendrite sprouting or retraction (Pinaud et al., 2001; Pinaud, 2004), which typically occurs under conditions of altered sensory input, such as in the case of lesions or enhanced activity (Grutzendler et al., 2002; Nimchinsky et al., 2002; Trachtenberg et al., 2002).
Based on these expression characteristics, these two IEGs became highly attractive to our research group as possible regulators of some aspects of plasticity in CNS neurons. As mentioned above, assuming that these genes play a role in triggering the early phases of plasticity in CNS networks, it would be expected that regions that exhibit low plasticity levels, such as the adult retina, would also display low expression levels of both NGFI-A and arc. We have thus looked at IEG in the adult rat retina in response to a paradigm that reliably induces plasticity in the mammalian brain.
Was this article helpful?