This technique uses a complementary DNA or RNA strand to detect a certain sequence of mRNA or DNA in the tissue, which for instance may be an entire brain section. Labeling can be radioactive or antigen based, using fluorophores or chromophores (Gall and Pardue 1969; Jin and Lloyd 1997). The method is suitable for determining cell-specific and region-specific changes in gene expression directly on brain sections.

One important variation of classical polymerase chain reaction (PCR) is quantitative PCR (qPCR), which precisely quantifies DNA sequences (Bustin et al. 2005). A second very important variation broadly used for gene expression detection is RTq-PCR. Instead of DNA, this method amplifies mRNA sequences using reverse transcriptase, polymerizing cDNA sequences as a product of the reaction. This method is very sensitive and also quantitative, allowing for the assessment of the amount of mRNA of a certain gene even if it is expressed at very low levels (Shiao 2003; Bustin et al. 2005); however, the method is performed on tissue homogenates and therefore allows for less regional specificity.

This method uses a plate with multiple spots that contain different DNA sequences for genes of interest. These sequences are hybridized with cDNA or cRNA extracted from a tissue sample. Hybridization levels can be detected by chemiluminescence or by the use of fluorophores. Typically, an important step before hybridization is the labeling of target sequences of the sample; mRNA is purified and submitted to RT-PCR to synthesize cDNA, which is labeled afterward with a fluorescent component (Freeman et al. 2000). The main advantage of this method is the ability to assess a very large number of genes for a single tissue sample.

This method is used to detect and sort specific proteins in tissue homogenates using gel electrophoresis. After being separated, proteins are transferred to a membrane (e.g., nitrocellulose), treated with a protein-rich blocking reagent to avoid non-specific labeling, incubated with a primary antibody (IgG) to detect a specific protein of interest, washed and then incubated with a secondary antibody (anti-IgG) associated with a reporter molecule (e.g., horseradish peroxidase or alkaline phosphatase). Detection is performed using a chromogenic or luminescent substrate (Gallagher et al. 2004). This method detects proteins and therefore provides only an indirect measurement of gene expression.

This method uses the same principle of antibody detection described above, but in tissue sections instead of homogenate samples (Jin and Lloyd 1997). This method also detects proteins and therefore provides only an indirect measurement of gene expression.

The immediate early gene c-fos gives rise to the fos protein that acts as a rather general transcription factor. Expression of the fos gene is upregulated by membrane depolarization and therefore widely used as a general marker of neuronal activity (Kovacs 1998; Cirelli and Tononi 2000a, b, c). The expression of c-fos is upregulated in several brain areas by spontaneous wakefulness (Pompeiano et al. 1994) as well as forced sleep deprivation (Cirelli et al. 1995). Both the mRNA and the protein levels of c-fos are elevated in rats when the sleep cycle is arrested, with a wide anatomical distribution that comprises the cerebral cortex, portions of the hypothalamus, the medial preoptic area, thalamic nuclei, and nuclei of the dorsal pontine tegmentum (Cirelli et al. 1995). Somewhat similar results were obtained in cats using immunohistochemistry, which revealed increased c-fos levels in the preoptic area and lateral hypothalamus of sleep-deprived animals (Ledoux et al. 1996). In mice subjected to sleep deprivation for 6h, c-fos mRNA protein levels were increased in the cerebral cortex, basal forebrain, thalamus, and cerebellum (Terao et al. 2003). A study of sleep deprivation in neonate rats found increased c-fos mRNA levels in the cerebral cortex and hippocampus at postnatal days 16, 20, and 24, with a return to basal levels after a 2-h recovery sleep period (Hairston 2004).

An anatomically comprehensive study of sleep-deprived mice used laser microdissection and cDNA microarrays to show upregulation of c-fos mRNA levels in visual and caudal somatosensory cortex, agranular insular cortex, orbitofrontal cortex, piriform cortex, amygdala, dorsal caudate putamen, and cerebellum, with downregulation in the ventral posteromedial nucleus of the thalamus. C-fos expression was largely coincident with expression of the excitatory neuron-specific vesicular glutamate transporter 1 (Slc17a7) and overlapped with Arc and Nr4a1 expression in visual and agranular insular cortices (Thompson 2010).

Sleep deprivation of non-mammalian vertebrates by and large yields similar results. In the white-crowned sparrow, c-fos mRNA levels are increased in several portions of the telencephalon after 6 h of sleep deprivation (Jones et al. 2007). In the zebra fish, sleep deprivation for 3 h or 6 h leads to c-fos mRNA upregulation in the olfactory bulb, telencephalon, diencephalon, and rhombencephalon, with a return to basal levels after recovery sleep (Appelbaum et al. 2010).

The transcription factor Zif-268 (zinc finger protein 268) is also known as NGFI-A (nerve growth factor-induced protein A), EGR-1 (early growth response protein 1), Krox-24, and ZENK (acronym of the initial letters of Zif-268, Egr-1, NGFI-A, and Krox-24). The Zif-268 protein regulates the expression of many genes involved with plasticity, mitogenesis, and differentiation, including synapsins I and II, and synaptobrevin II (Knapska and Kaczmarek 2004). Zif-268 was found to be upregulated after 6 h of sleep deprivation in the rat cerebral cortex assessed through Northern blotting (O’Hara et al. 1993). In rats submitted to sleep deprivation for up to 10 days using the disk-over-water method (Landis et al. 1993), immunocytochemistry and Northern blotting showed elevated Zif-268 expression in the dorsal raphe, lateral habenula, superior colliculus, and ventral periaqueductal gray. These neuroanatomical locations were not corroborated by a subsequent study, which applied in situ hybridization to brains of animals subjected to sleep deprivation by gentle handling for 3, 6, 12, and 24 h (Pompeiano et al. 1997). Increased mRNA levels were found in the cerebral cortex and caudate–putamen (Pompeiano et al. 1997). The same authors found that 3 h of sleep deprivation was associated with increased Zif-268 levels in the cerebral cortex but not in the locus coeruleus, while 12 or 24 h of sleep deprivation induced Zif-268 expression in the locus coeruleus, the parabrachial nuclei, and the superior and inferior colliculi (Tononi et al. 1994). A plausible explanation for the somewhat divergent findings may reside in the different amounts of sleep deprivation applied. Longer periods of deprivation may trigger self-regulation mechanisms such as the negative feedback enabled by the Zif-268 response, in which the protein recognizes consensus sequences in the promoter region of the gene, allowing for downregulation of its own expression (Christy and Nathans 1989).

More recently, a study in mice showed that sleep deprivation for 6 h followed by a recovery period of 4 h leads to the upregulation of Zif-268 mRNA levels in the cerebral cortex and basal forebrain (Terao et al. 2003). The deprivation-related increase of Zif-268 levels seems to obey a very similar dynamics in the avian brain. In white-crowned sparrows, telencephalic Zif-268 mRNA levels were reported to increase 117 % after 6 h of sleep deprivation, with a 90 % increase from spontaneous waking to control (Jones et al. 2008).

Another immediate early gene involved in neuronal plasticity and affected by sleep deprivation is Arc (also known as Arg3.1), encoded by mRNA transported to post-synaptic dendritic terminals for local translation (Lyford et al. 1995). Arc participates in synaptic remodeling by regulating the dynamics of F-actin, AMPA receptors, and Ca2+/calmodulin-dependent protein kinase II (CaMKII) (Bramham et al. 2010; Okuno et al. 2012).

Arc levels are upregulated in rats after both acute (8 h) (Cirelli and Tononi 2000a, b, c) and chronic (1-week) sleep deprivation, when assessed by microarrays and confirmed by real-time quantitative polymerase chain reaction (RT-qPCR) and RNase protection assay (RPA) (Cirelli et al. 2006). Another study found Arc mRNA levels to be upregulated in the cerebral cortex of mice and rats subjected to sleep deprivation for 6 h (Terao et al. 2006). Similar results were obtained in the cerebral cortex and hippocampus after 8 h of sleep deprivation in rats, with recovery to basal Arc mRNA levels after 2 h of sleep (Taishi et al. 2001). Corroborating these findings, a study with C57BL/6J mice showed higher levels of Arc mRNA in the hippocampus, amygdala, neocortex, and piriform cortex of sleep-deprived animals, in comparison with controls. However, increased Arc levels were detected even after a 4-h sleep rebound period (Thompson 2010). Another study in rats using gene chip analysis and 24 h of sleep deprivation by gentle handling found a 1.2-fold decrease of Arc mRNA levels in the hypothalamus and a twofold increase in the hippocampus, in comparison with controls (Conti et al. 2006). A meta-analysis of datasets comprising 91 different genes found Arc mRNA to be upregulated in the cerebral cortex, hippocampus, olfactory bulb, and striatum of mice deprived of sleep for 6 h (Wang et al. 2010). Similar effects occur in the brain of the white-crowned sparrow, with a 66 % increase in Arc mRNA levels after 6 h of sleep deprivation (Jones et al. 2008).

Another gene of interest is Homer1, which is widely expressed in the brain and other tissues with two main splice variants: a short form called Homer1a, which is induced by neuronal activity, and long forms called Homer1b and c, which have constitutive expression (Shiraishi-Yamaguchi and Furuichi 2007). Homer1a inhibits the scaffolding action of the long forms, decreasing glutamatergic signaling and diminishing dendritic spines (Tu et al. 1998). Like Arc (Sect. 4.3), Homer1a is broadly upregulated in the brains of sleep-deprived rats (Cirelli et al. 2006; Maret et al. 2007). Additionally, it is expressed in spontaneously awake rats, which suggests that Homer1a is generally associated with wakefulness and neuronal activity (Thompson 2010). Homer1a mRNA is upregulated in the somatosensory cortex of rats deprived of sleep for 6 h during the day (sleep period); in contrast, Homer1a gene expression is downregulated by sleep deprivation during the evening (active period); similar results of a lesser magnitude were obtained for Homer1bc (Nelson et al. 2004). Homer1a is significantly overexpressed in the cerebral cortex, hippocampus, and striatum in three different mouse genotypes (Maret et al. 2007). Another study using microarray analysis and the disk-over-water method presented compatible results: Rats deprived of sleep for 8h showed upregulated Homer1a levels in comparison with animals allowed to sleep, while sleep deprivation for 7 days yielded a non-significant tendency for transcriptional downregulation related to short-term sleep deprivation (8 h) and spontaneous waking (Cirelli et al. 2006). Homer1 upregulation was detected in the basal forebrain and cerebral cortex of rats submitted to sleep deprivation by introducing novel objects and tapping the cage during 6 h (Terao et al. 2006). Another study used in situ hybridization to detect Homer1 upregulation in rats deprived of sleep for 24 h (Conti et al. 2006). An investigation of Homer1 expression after 3, 6, 9, and 12 h of sleep deprivation found gene upregulation in the cerebral cortex for all these periods (Mackiewicz et al. 2007). Although these particular three studies (Conti et al. 2006; Terao et al. 2006; Mackiewicz et al. 2007) did not specify which Homer1 variation was measured, it is likely that it was Homer1a, since most of the other studies cited in this section specifically indicate Homer1a to be upregulated by sleep deprivation (Mackiewicz et al. 2007). A meta-analysis found Homer1a levels to be upregulated in mice after 6 h of sleep deprivation in the cerebral cortex, hippocampus, olfactory bulb, and striatum (Wang et al. 2010). Interestingly, experiments performed in drosophila showed Homer downregulation in sleep-deprived flies. Homer1a uncouples the metabotropic glutamate receptor from other Homer isoforms. Drosophila has only one Homer isoform, which acts in the opposite direction of Homer1a (Zimmerman et al. 2006; Mackiewicz et al. 2008).

Gene expression is highly influenced by nuclear receptors that act as transcription factors upon activation by steroid and thyroid hormones, among other ligands (Green and Chambon 1988; McKenna and O’Malley 2002). Some nuclear receptors are regulated as immediate early genes, such as the NR4A subfamily of orphan receptors that include the nerve growth factor IB (NGFIB/NR4A1), the nuclear receptor-related protein 1 (NURR1/NR4A2), and the neuron-derived orphan receptor 1 (NR4A3/NOR1) (Maxwell and Muscat 2006; Hawk and Abel 2011). Sleep deprivation affects Nr4a expression, which is activated by the same calcium-dependent factors that upregulate the memory-related IEG expression reviewed above (Hawk and Abel 2010). In 6-h sleep-deprived rats, Nr4a mRNA levels were found to be upregulated in the cerebral cortex, basal forebrain, and hypothalamus (Terao et al. 2006). Another study using gene chip analysis found a 100 % increase in Nr4a3 mRNA expression in the rat hypothalamus after 24 h of sleep deprivation (Conti et al. 2007). In white-crowned sparrows deprived of sleep for 6 h, Nr4a1 mRNA levels in the telencephalon were found to be increased by 60 %, in comparison with control animals allowed to sleep (Jones et al. 2008). An assessment of Nr4a1 gene expression in mice deprived of sleep for 6h using laser microdissection and microarray hybridization found elevated mRNA levels in the amygdala and in the primary visual, agranular insular, orbitofrontal and piriform cortices, as well as in the cerebellum and putamen; controls allowed to sleep showed increased levels of Nr4a1 and Arc in the rostral and ventral caudate–putamen, as well as in the motor and rostral somatosensory cortices; sleep-deprived mice showed Nr4a1 upregulation in the caudal dorsolateral putamen, overall a very similar profile found for Arc expression. (Thompson 2010). A meta-analysis of microarray data in mouse, rat, sparrow, and fly identified the Nr4a gene family as conserved genes affected by sleep deprivation (Wang et al. 2010).

The upregulation of immediate early gene expression by sleep deprivation requires the involvement of secondary messengers. An investigation of the levels of phosphorylated cAMP response element-binding protein (p-CREB) after 3 h of sleep deprivation detected a significant and widespread increase in the cerebral cortex (Cirelli and Tononi 2000a, b, c). Another study found the mRNA levels of CaMKII to be upregulated in the cerebral cortex and basal forebrain of rats deprived of sleep for 6 h (Terao et al. 2006). In contrast, other researchers found that 48 h of sleep deprivation reduces CREB mRNA levels to the levels found in controls kept in their home cages; similar results were obtained for CaMKII (Guzman-Marin 2006). Chemical lesions of the noradrenergic system showed that the expression of Zif-268 and c-fos during wakefulness depends critically on the integrity of the locus coeruleus (Cirelli et al. 1996). The implication of this finding is that calcium-dependent immediate early gene expression, and presumably upstream events such as increases in cAMP and p-CREB levels, also requires noradrenergic modulation.