Children with Autism Spectrum Disorders (ASD) have impairments in three core domains: socialization, communication, and restricted interests and repetitive behaviors [1–4]. Researchers have reported that psychiatric comorbidity in ASD ranges from 41% to 70% [5, 6].

Although the etiology of the disorder is unknown, recent studies have suggested that the susceptibility to autism is clearly attributable to genetic factors [7, 8]. In addition, emerging evidence points to inflammatory and apoptotic mechanisms being responsible for certain neuropsychiatric disorders including autism. Vargas et al. [9] suggested neuroinflammatory processes are present in the autistic brain by showing that transforming growth factor (TGF)α1, macrophage chemoattractant protein (MCP) 1, interleukin (IL)6 and IL10 are increased in the brain of autistic subjects. A number of studies have also shown that inflammatory cytokines including tumor necrosis factor (TNF)α, interferon (IFN)α, IL1α, IL6, IL8 and IL12 are elevated in blood mononuclear cells, serum, plasma and cerebrospinal fluid (CSF) of autistic subjects [9–16].

The mechanisms of apoptosis induction are complex and not fully known, but some key events are identified that appear essential for the cell to enter apoptosis. The role of specific ions in the apoptotic process is slowly being revealed. Changes in intracellular Ca²⁺ have long been associated with apoptotic neuronal cell death. Ca²⁺ ionophores have been shown to induce ultrastructural changes, such as cell shrinkage, chromatin condensation, and DNA fragmentation, consistent with apoptosis [17–20]. Increased Ca²⁺ has been linked to processes occurring during apoptosis including caspase activation.

One key event in apoptosis is loss of intracellular potassium ions (K⁺). Depletion of K⁺ is necessary for cells to shrink, activate caspases and degrade DNA [21–23], events that in turn lead to further characteristic apoptotic changes such as membrane blebbing and formation of apoptotic bodies. Apoptosis due to forced loss of intracellular K⁺ can be induced by ionophores or K⁺ channel activators [24–26]. In addition, Yu et al. [25, 27] have also shown that the outward K⁺ current that ensues from N-methyl-D-aspartate receptor activation has also been shown to induce apoptotic changes in cultured hippocampal neurons.

Just as with increased Ca²⁺and K⁺ efflux, the importance of sodium (Na⁺) entry in inducing neuronal injury and death in response to pathophysiologic conditions, such as hypoxia, has been well established [28–34]. Moreover, Banasiak et al. [35] proved that blocking Na⁺ entry in hypoxia-exposed neurons reduced the proportion of DNA fragmentation and reduced apoptotic cell.

Magnesium (Mg²⁺) has a profound effect on neural excitability; the most characteristic signs and symptoms of Mg²⁺ deficiency are produced by neural and neuromuscular hyperexcitability [36]. Iotti and Malucelli [37] clarify the functional relationship between energy metabolism and free [Mg²⁺], providing evidence that brain cells cytosolic [Mg²⁺] is regulated to equilibrate any changes in rapidly available free energy. Moreover, it has also been shown that the measurement of brain Mg²⁺ can help in the differential diagnosis of neurodegenerative diseases sharing common clinical features.

The immune system has been postulated to play an important role in the etiology of autism. Investigators have proposed infectious, autoimmune, and cytokine-related etiologies.

These information initiate our interest to measure concentrations of Na⁺, K⁺, Ca²⁺, Mg²⁺ together with caspase3 as a proapoptotic marker, IL6 and TNFα as proinflammation markers in the plasma of autistic patients from Saudi Arabia in an attempt to understand the role and relationship of these biochemical parameters in the etiology of autism and its commonly related psychiatric conditions.

Protection of the brain from injury during the fetal, neonatal and postnatal periods is of major importance owing to the significant number of infants who now survive early brain injury but develop neurodevelopmental and motor disabilities.

Table 1 and Figures 1 and 2 show the unexpected lower concentrations of caspase3, TNFα and IL6. This could be interpreted on the basis that the etiology of the fetal brain damage inflammation will involve many factors and is likely to include an increase in circulating cytokine concentrations. Rees et al. [43] have shown, for example that TNFα [44] and IL6 concentrations [45] increase within the early 6 hours of lipopolysaccharide (LPS) exposure. It has been proposed that circulating cytokines might act on cerebralendothelial cells or periventricular cells to upregulate prostaglandin synthesis, resulting in increased permeability of the blood-brain barrier [46]; thus the administration of LPS to fetal sheep results in the extravasation of plasma proteins and macrophages into the brain [46].

TNFα and IL6 are cytokines involved in cell-mediated immune response and their production has been shown to be associated with tissue inflammation and necrosis [47]. Based on these information, the recorded lower plasma concentrations of these two cytokines does not oppose with the neuroinflammatory model recently proved for autism [48]. This could help us to suggest that localized inflammation of the central nervous system may contribute to the pathogenesis of autism and that elevation of plasma cytokines could be an early event followed by infiltration of macrophages, cytokines and proapotic factors across the BBB to the brain. The lower recorded concentration of caspase3 in autistics compared to control subjects could be easily related to the decrease in TNFα. This could be supported through considering the previous report of Mundle et al. [49] which demonstrated a link between TNFα and the major effectors of its apoptotic signal, i.e. Caspase1 and 3. They identify the downstream effectors of TNFα apoptotic signalling and show a positive correlation of TNFα with Caspase3.

A major endogenous antioxidant in mammalian cells is the enzyme superoxide dismutase (SOD), which catalyzes the dismutation of the superoxide anion (O₂ ^-) into hydrogen peroxide (H₂O₂) and molecular oxygen (O₂). Dimayuga et al. [50] show that overexpression of SOD1 in microglial cells leads to significant decreases in superoxide concentrations, with corresponding increases in H₂O₂ concentrations. They proved that the release of the proinflammatory cytokines TNFα and IL6 is significantly attenuated by overexpression of SOD1. With special consideration of the effect of population, the recorded lower concentrations of TNFα and IL6 in autistic patients as subjects of the present study compared to controls could be related to the overexpression of SOD previously reported as metabolic biomarker in Saudi autistic patients [51].

Table 1 and Figure 2 demonstrate that autistic patients from Saudi Arabia recorded lower concentrations of plasma Ca²⁺. This could find a support through considering the work of Shearer et al. [52] in which they observed lower Ca²⁺ concentrations in the hair of autistic population and that of Krey and Dolmetsch [53] in which they proved that some forms of autism are caused by failures in activity-dependent regulation of neural development due to mutations of several voltage-gated and ligand-gated ion channels that regulate neuronal excitability and Ca²⁺ signalling. On the other hand, the recorded lower concentration of Ca²⁺ is not in accordance with the recent work of Laura et al. (2011) [54] which reported higher Ca²⁺ concentrations in plasma of Italian autistic patients compared to age and gender matching controls. The reduced plasma Ca²⁺ concentrations of the present study could be associated with high intracellular brain Ca²⁺ in autistics compared to control subjects. This suggestion could be supported with the recent evidence from post-mortem studies of autistic brains which points toward abnormalities in mitochondrial function as possible downstream consequences of dysreactive immunity and altered Ca²⁺ signalling [55]. Low plasma Ca²⁺ and the speculated high brain Ca²⁺ concentration could be easily correlated to the oxidative stress previously recorded in Saudi autistic patients [56] as elevated brain Ca²⁺ is recently related to ROS generation. Mitochondrial aspartate/glutamate carrier (AGC1), isoform predominantly expressed in the brain, heart and skeletal muscle, is known to play a pivotal role in energy metabolism and is regulated by neurone intracellular Ca²⁺[57, 58]. This carrier was found to be approximately three-fold higher in brain homogenates from each of six autistic patients compared to their matched controls. This could support the lower plasma Ca²⁺ concentrations recorded in the present study. Moreover, direct fluorimetric measurements of Ca²⁺ concentrations in the post-mortem mitochondrial supernatant confirmed significantly higher Ca²⁺ concentrations in brain of autistics [55].

This suggested increased influx of blood-to-brain Ca²⁺ could be easily related to the loss of amyloid beta (Aβ) equilibrium between the brain and blood which may lead to failure of drawing out Aβ from the brain across the blood brain barrier (BBB) as a mechanism for Aβ accumulation in Saudi autistics [Al-Ayahdi L, Ben Bacha A, Kotb M, El-Ansary A: A novel study on amyloid β peptide 40, 42 and 40/42 ratio in Saudi autistics, Submitted]. Vitamin E which is known to attenuate A β-induced apoptosis despite Ca²⁺ accumulation in brain cells is significantly lower in Saudi autistic patients [51]. This could support the suggested mechanism relating Aβ and Ca²⁺- induced apoptosis in brain cells of Saudi autistics.

Table 1 and Figure 1 demonstrate K⁺ concentrations in plasma of autistic and control subjects. It could be easily noticed that autistic patients recorded raised concentrations of K⁺ compared to controls. This could be attributed to the altered Na⁺/K⁺ ATPase activity previously reported by El-Ansary et al. [56], which may represent an important neurotoxic mechanism for neurons.

The recorded higher plasma concentrations of K⁺ which reflect the remarkable higher rate of k⁺ efflux from brain to blood in autistic patients could be easily related to the significant lower Ca²⁺, the unchanged Na⁺, lower Ca²⁺/Na⁺ ratios and to the speculated higher brain caspase3 activity. Xiao et al. [59] showed previously that activation of the N-methyl-D-aspartic acid (NMDA) subtype of glutamate receptors in a low Ca² and Na⁺ condition induced apoptotic neuronal death, and that the K⁺ efflux via NMDA receptor channels was likely a key event in NMDA-induced apoptosis. This postulation could be supported by Pigozzi et al. [60] who proved that entry of Ca²⁺ into neuron cells can accelerate apoptosis by accelerating the expression of growth arrest and DNA Damage inducible gene 153 (GADD153) and by inducing a prolonged efflux of K⁺ out of the cell. This is in good agreement with the elevated K⁺ and the reduced Ca²⁺ concentrations in plasma of autistic patients compared to controls as a report of the present study. Moreover, the significantly impaired Ca²⁺ and K⁺ concentrations in plasma of autistic patients could be easily related to the postulated increase of brain cytokines (TNFα and IL6) after infiltration from plasma to brain. Experimental evidence demonstrates that ion channels are targeted by cytokines, which can specifically modulate their function [61] and TNFα was associated with the remarkable Ca²⁺ influx from blood to brain [62]. These suggested mechanisms of the alteration of the studied parameters could be supported through the obtained Pearson correlations presented in table 3 and Figure 3.

ROC analysis presented in Figure 4, support the previous discussion and suggestions which based on the obtained data. Most of the measured parameters recorded AUC near 1 and satisfactory levels of specificity and sensitivity and hence they could be used as biochemical markers for the early diagnosis of autism in Saudi population.