Fractalkine and Neuroinflammation in Autism Spectrum Disorder: A Novel Perspective

Abstract

Objective

To investigate the role of neuroinflammation in the etiopathogenesis of autism spectrum disorder (ASD), we investigated the role of fractalkine and tumour necrosis factor alpha (TNF-α), which may be potential biomarkers for ASD. This study aimed to evaluate the serum levels of interleukin-1beta (IL-1β), interleukin-6 (IL-6), and high-sensitivity CRP (hs-CRP) and to investigate the relationship between fractalkine, TNF-α, IL-1β, IL-6, and hs-CRP and the severity of symptoms in ASD.

Methods

In this cross-sectional study, 44 children between the ages of 24−72 months diagnosed with ASD constituted the research group, and 44 healthy children of similar age and sex constituted the control group. Detailed mental status examinations were performed in both groups. Symptom severity of children diagnosed with ASD was evaluated using the Childhood Autism Rating Scale, Autism Behaviour Checklist and Repetitive Behaviours Scale-Revised Turkish Version. Peripheral venous blood samples were obtained from children in both groups and serum fractalkine, TNF-α, IL-1β, IL-6 and hs-CRP levels were measured by ELISA method.

Results

Serum fractalkine and IL-1β levels of children in the ASD group were significantly lower than those in the control group. No significant difference was found between the groups in serum TNF-α, IL-6 and hs-CRP levels. There was no correlation between ASD severity and fractalkine, TNF-α, IL-1β, and IL-6 levels.

Conclusion

Our study is the first to evaluate serum fractalkine levels in ASD in early childhood. Our findings suggest that fractalkine may play a role in the etiopathogenesis of ASD in early life and may be a potential biomarker for ASD.

INTRODUCTION

Autism spectrum disorder (ASD) is a neurodevelopmental disorder that begins in the first years of life and manifests itself with difficulties in social relationships and communication, repetitive behaviours and restricted interests [1]. According to the United States Centers for Disease Control and Prevention data for 2023, the prevalence of ASD was reported as 1/36 [2]. Although the etiology of ASD is not clearly known, the role of immune system-related problems in the etiopathogenesis of ASD has recently attracted attention [3]. Studies have reported that maternal infection in the prenatal period increases the risk of ASD by causing maternal immune activation [4], the prevalence of autoimmune diseases is increased in children with ASD compared to the control group [5], and children with ASD have irregularities in cytokine profiles [6].

Irregularities in the cytokine profile are among the immunological changes associated with ASD [7]. Tumour necrosis factor alpha (TNF-α) is a proinflammatory cytokine that regulates the immune response. Synthesised by neurons and glial cells in the central nervous system (CNS), TNF-α has a critical role in neurogenesis, neuronal cell death and synaptic plasticity. Thus, it regulates the normal embryonic development of the brain. There are many studies in the literature investigating the relationship between ASD and TNF-α. Studies have shown that TNF-α levels are increased in brain tissue and cerebrospinal fluid of individuals with ASD and that there is a relationship between TNF-α increase and ASD symptom severity [7,8]. Tsilioni et al. [9] reported that serum TNF-α levels of children with ASD were higher compared to the control group. On the other hand, studies are reporting no difference in TNF-α levels of children with ASD compared to the control group [10] and lower TNF-α levels [11].

Interleukin-1 beta (IL-1β), a neuromodulator in normal brain development, plays a role in the activation of microglial cells and recruitment of peripheral leukocytes in neuroinflammation [12]. When we look at the studies examining the relationship between ASD and IL-1β, Li et al. [13] found no difference in IL-1β levels in brain tissues of individuals with ASD. IL-1β levels in amniotic fluid and peripheral blood samples of individuals with ASD were similar to controls [10,14]. There are also studies in the literature showing that peripheral IL-1β levels are increased in individuals with ASD [15]. In a study conducted by Abdallah et al. [16] using neonatal blood samples of 359 individuals with autism and 741 healthy controls, IL-1β levels were found to be lower in individuals with autism compared to controls. The low neonatal IL-1β levels in individuals with ASD were interpreted to be related to decreased or insufficient immune cell activity in the neonatal period.

Interleukin-6 (IL-6) is involved in processes such as neuronal survival, synapse formation and plasticity in the CNS and affects many stages of neurodevelopment [17]. Many studies have shown IL-6 dysregulation in individuals with ASD. In one study, circulating IL-6 levels were increased in individuals with autism compared to typical controls [18]. IL-6 has also been shown to be increased in postmortem brain tissues of individuals with autism, with more IL-6 staining, especially in the cerebellar region [19]. It is thought that increased IL-6 levels in ASD may play a role in the maintenance of the disorder or may be a biomarker of infectious and toxic environmental exposures and altered biological homeostasis [20].

C-reactive protein (CRP) is an acute phase reactant produced by IL-1, IL-6, and TNF-α stimulation from the liver [21]. Studies have shown that high CRP levels in peripheral blood are associated with depression, bipolar disorder and schizophrenia [22-24]. In animal studies, inflammation associated with peripheral CRP levels has been reported to be associated with autism-like behaviours in mice [25]. In the Finnish national cohort study, increased maternal CRP was associated with an increased risk for ASD [26]. High-sensitivity CRP (hs-CRP) refers to CRP that can be measured at levels as low as 0.30 mg/L and is more sensitive than CRP [27]. In literature the association of hs-CRP and ASD has been relatively less investigated and results of studies are inconsistent [28].

In ASD, abnormal expression of cytokines in the brain [29], peripheral blood [18,30] and gastrointestinal system [31] has been shown in studies. While many studies have focussed on the role of cytokines in the pathophysiology of ASD, the related immune protein family, chemokines, has been relatively neglected. Fractalkine, which is a member of the CXXXC chemokine family and is expressed more in the CNS than in the immune system and peripheral tissues, provides communication between neurons and microglia by binding to its receptor CX3CR1 and mediates the regulation of brain functions [32]. This interaction effectively regulates neural networks, synapse maturation and synaptic plasticity; regulates cognitive functions and controls immune processes [33]. In addition, fractalkine and CX3CR1 interaction play a role in the effective recovery of the inflammatory response in the damaged brain region by regulating the balance between proinflammatory and anti-inflammatory cytokines such as TNF-α, IL-1β, IL-6, IL-10 [34]. Recent studies have emphasised the role of fractalkine and CX3CR1 interaction disruption in the etiopathogenesis of CNS-related disorders [33,35]. To the best of our knowledge, no study in the literature investigates the relationship between serum fractalkine levels and ASD.

In light of this information, the results of clinical studies on TNF-α, IL-1β and IL-6 levels in individuals with ASD are inconsistent. It is thought that this may be due to methodological differences in the studies. In the literature, there is no study evaluating the serum levels of fractalkine and other cytokines together in patients with ASD and examining their relationship with autism symptom severity. Our study aimed to examine fractalkine, TNF-α, IL-1β, IL-6 and hs-CRP levels in children at risk for ASD, considering the importance of biomarker studies that can help diagnose ASD with minimally invasive methods. Our study suggests that investigating the role of neuroinflammation in ASD and determining immune profile abnormalities may lead to a better understanding of the pathophysiology of autism and the determination of reliable diagnostic methods and treatment strategies in autism.

METHODS

Study Participants

Patients admitted to Ankara University Faculty of Medicine, Child and Adolescent Psychiatry Outpatient Clinic between December 2022 and March 2023. Those diagnosed with ASD, according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), were included in the study. Inclusion criteria were as follows: being between 24 to 72 months of age, not having any medical disease, not having any psychiatric disease other than cognitive developmental delay, not using any medication and not having any infection in the last month. Exclusion criteria were the presence of any medical disease, taking any medication or immunomodulatory supplements, and having an infection in the last month. Age- and sex-matched children aged 24 to 72 months without any medical and psychiatric comorbidity were recruited as the control group. The same exclusion criteria were applied in the control group.

The parents of the participants were informed about the study, and their written informed consent was obtained. The study was approved by the Ankara University Faculty of Medicine Human Research Ethics Committee (Decision No: İ10-605-22, 10 November 2022). It was supported by the Ankara University Scientific Research Projects Coordination Office with the project code TTU-2023-2845.

The sample size was calculated using the G-Power 3.1.9.4 programme, considering the significance level of the hypothesis and the effect size. Since the mean TNF-α pg/ml (12.15 ± 4.62) of the ASD group and TNF-α pg/ml (8.77 ± 3.16) of the control group obtained in the study by Xie et al. [8] were taken as basis, the effect size was found to be 0.85 (high effect level). In order to find a significant difference between the groups, the error was 0.005 (α = 0.05), the test’s power was 95% (1 − β = 0.95), and the sample size was calculated as at least 31 people in each group.

Procedures

A clinician assessed participants using a diagnostic interview based on DSM-5. All children included in the study were administered a standardized developmental or intelligence test appropriate to their developmental level. All participants underwent further diagnostic assessment by experienced child psychiatrists and psychologists in the infant observation room. After all these assessments, parents of children who met the inclusion/exclusion criteria for the ASD and control groups were interviewed, and a sociodemographic data form was completed. After confirmation of the diagnosis, Childhood Autism Rating Scale (CARS), Autism Behaviour Checklist (ABC) and Repetitive Behaviour Scale-Revised Form (RBS-R) were used to determine the severity of ASD symptoms. The clinician completed CARS in the ASD group through clinical observation, and the ABC and RBS-R were completed by interviewing the parents.

Data Collection Tools

Childhood Autism Rating Scale

This 15-item scale is frequently used in the differential diagnosis and screening of ASD. The total score ranges from 15 to 60, with a score of 30 and above indicating the presence of ASD. A score range of 30 to 36.5 points is considered ‘mild-moderate autistic,’ and a score range of 37 to 60 points is considered ‘severe autistic.’ The scale was developed by Schopler et al. [36].

Autism Behaviour Checklist

Developed by Krug et al. [37], the scale used to determine the level of autism symptoms consists of 57 items and 5 subscales as ‘sensory,’ ‘social association,’ ‘body and object use,’ ‘language skills,’ and ‘self-care skills.’ The total score is in the range of 0 to 159.

Repetitive Behaviours Scale-Revised Turkish Version

The scale developed by Bodfish et al. [38] to assess repetitive behaviours and their severity consists of 6 subscales as ‘stereotypic behaviours,’ ‘self-destructive behaviours,’ ‘compulsive behaviours,’ ‘ritualistic behaviours,’ ‘sameness/uniformity behaviours,’ ‘limited interest,’ and 43 items. As the score increases, the severity of repetitive behaviour is considered to increase.

Biochemical Assesments

Serum fractalkine, TNF-alpha, IL-1 beta, IL-6 and hs-CRP levels were measured in all participants. Blood samples from all participants were collected in the morning between 8:00 AM and 9:00 AM (after 12 hours of fasting). Each participant’s venous blood samples (10 cc) were collected in biochemistry tubes. Biochemistry tubes were centrifuged at 3,000 ×g for 15 minutes after 30 minutes of incubation. Serum samples were stored at −80°C until biochemical analysis. Serum fractalkine, TNF-alpha, IL-1 beta, IL-6 and hs-CRP levels were analysed by Enzyme-linked Immunosorbent Assay (ELISA) Kit For Chemokine C-X3-C-Motif Ligand 1 (Cloud-Clone Corp.), TNF-α -ELISA Kit (DIAsource), IL-1 Beta Human ELISA Kit (DIAsource), IL-6 Human ELISA Kit (DIAsource), HsCRP Human ELISA Kit (DIAsource) were measured by ELISA method.

Statistical Analysis

Statistical Package for Social Sciences (SPSS) 25.0 package program (IBM Co.) was used for data analysis. The normal distribution of continuous variables was tested by Kolmogorov-Smirnov and expressed as mean and standard deviation. Student ttest was used for normally distributed variables, and the Mann-Whitney Utest was used for nonparametric variables. The chi-square and Fisher’s exact tests were used to compare categorical variables. In the univariate analysis, logistic regression models were applied to evaluate the potential association between independent variables and ASD diagnosis. Variables with a pvalue less than 0.20 in the univariate analysis were included in the multivariate model. The results of both univariate and multivariate analyses were presented as odds ratios (ORs) with 95% confidence intervals (CIs). Pearson’s test or Spearman’s test was used to determine whether there was a relationship between continuous variables. A pvalue of < 0.05 was considered statistically significant.

RESULTS

The study included 44 patients and 44 healthy controls. There was no significant difference between the groups in terms of age and sex (p > 0.05). The clinical and demographic characteristics of all participants are shown in Table 1.

Table 1.

Sociodemographic and clinical characteristics of ASD and control group

	ASD (n = 44)	Control (n = 44)	t or χ2	p
Age (mo)	41.75 ± 12.27	45.11 ± 9.57	1.434a	0.155
Sex			0.604b	0.437
Male	36 (81.8)	33 (75.0)
Female	8 (18.2)	11 (25.0)
Scales (min.−max.)
CARS	38.20 ± 5.58 (29.50−50.50)	-	-	-
ABC	48.54 ± 26.88 (5.00−103.00)	-	-	-
RBS–R	15.30 ± 14.77 (1.00−55.00)	-	-	-

Values are presented as number (%) or mean ± standard deviation.

ASD, autism spectrum disorder; CARS, Childhood Autism Rating Scale; ABC, Autism Behaviour Checklist; RBS-R, Repetitive Behaviour Scale-Revised Form; -, not available.

^aIndependent groups ttest. ^bPearson’s chi-square test.

Serum fractalkine levels were significantly lower in the ASD group compared to the control group (p = 0.009). IL-1β levels were significantly lower in the ASD group compared to the control group (p = 0.003). There was no statistically significant difference between TNF-α, IL-6, and hs-CRP levels according to the groups (p > 0.05). Blood parameters of the patient and control groups are shown in Table 2.

Table 2.

Comparison of fractalkine, TNF-a, IL-1β, IL-6, hs-CRP levels in ASD and control groups

	ASD (n = 44)	Control (n = 44)	t or Z	p
Fractalkine (ng/ml)	0.41 ± 0.09 0.38 (0.10)	0.47 ± 0.13 0.44 (0.18)	2.695b	0.009*
TNF-a (ng/ml)	18.96 ± 5.82 18.00 (6.18)	19.90 ± 4.82 19.77 (7.45)	0.828b	0.410
IL-1β (ng/ml)	31.70 ± 10.25 28.16 (7.23)	38.98 ± 20.55 31.36 (12.05)	−2.946c	0.003*
IL-6 (pg/ml)	39.14 ± 19.60 33.17 (6.76)	40.05 ± 20.72 33.29 (5.61)	−0.430c	0.667
hs-CRP (ng/ml)a	1,757.68 ± 713.48 1,623.57 (1,126.84)	1,776.8 ± 709.39 1,756.24 (1,277.89)	0.126b	0.900

Values are presented as mean ± standard deviation or median (IQR).

ASD, autism spectrum disorder; TNF-a, tumour necrosis factor alpha; IL-1β, interleukin-1 beta; IL-6, interleukin-6; hs-CRP, high-sensitivity C-reactive protein; IQR, interquartile range.

^ahs-CRP > 3,000 ng/ml is excluded. ^bIndependent sample ttest. ^cMann-Whitney Utest.

*p < 0.05.

As seen in Table 3, fractalkine was found to be significant (p = 0.012) in the univariate logistic regression analysis and IL-1β variable was included in the multivariate logistic regression model according to the p = 0.200 rule. According to the model results, an increase in fractalkine values decreases the probability of being diagnosed with ASD by 99% (OR: 0.01, 95% CI: 0.00−0.24, p = 0.010). The variables in the model explain approximately 18% of the variance in ASD diagnosis.

Table 3.

Univariate and multivariate logistic regression analysis results of blood parameters

Variables	Univariate			Multivariate
	OR (95% CI)	p		OR (95% CI)	p
Fractalkine (ng/ml)	0.01 (0.00−0.30)	0.012*		0.01 (0.00−0.24)	0.010*
TFN-α (ng/ml)	0.96 (0.89−1.04)	0.409		-	-
IL-1β (ng/ml)	0.96 (0.92−1.01)	0.070		0.96 (0.92−1.00)	0.053
IL-6 (pg/ml)	0.99 (0.97−1.01)	0.830		-	-
hs-CRP (ng/ml)	1.00 (0.99−1.01)	0.899		-	-

−2 Log likelihood = 107.898, R² = 0.181.

CI, confidence interval; OR, odds ratio; TNF-a, tumour necrosis factor alpha; IL-1β, interleukin-1 beta; IL-6, interleukin-6; hs-CRP, high-sensitivity C-reactive protein; -, not available.

*p < 0.05.

When the relationship between the blood parameters and the scores of the CARS, ABC and RBS-R was analysed in the ASD group, there was no significant relationship between serum fractalkine, TNF-α, IL-1β, IL-6, hs-CRP levels and the scale scores (p > 0.05).

DISCUSSION

To the best of our knowledge, our study is the first to evaluate the serum levels of fractalkine and other cytokines in early childhood and examine their relationship with autism symptom severity. Our study showed that serum fractalkine and IL-1β levels were significantly lower in the ASD group compared to the control group and were not associated with ASD symptom severity.

Looking at the studies examining the relationship between fractalkine and psychiatric disorders, Hill et al. [39] showed that fractalkine levels were decreased in the prefrontal cortex of schizophrenia patients. In another study, serum fractalkine levels of schizophrenia patients were found to be lower compared to controls. In this study, it was thought that low fractalkine levels may be related to increased microglial activation and increased cytokine levels [40]. Based on the findings that schizophrenia is also a disease of neurodevelopmental origin [41] and that there are many common points with schizophrenia in the etiopathogenesis of ASD [42], the results of our study were found to be compatible with the results of studies conducted with schizophrenia patients. In a study conducted by Fernández de Cossío et al. [43] with maternal immune activation mouse models exhibiting ASD-like behaviours, decreased CX3CR1 expression was reported in the hippocampus of mice, while no change was found in fractalkine levels. Similar to this study, Bar and Barak [44] showed that CX3CR1-deficient mice exhibit ASD-like features such as impairment in social behaviours and functioning. It was thought that ASD-like features may be associated with a decrease in microglial cells, poor functional brain connectivity and impaired synaptic transmission, and it was emphasised that CX3CR1 may be of critical importance in neurodevelopment. Ishizuka et al. [45] reported that the CX3CR1 Ala55Thr gene variant increases the risk of schizophrenia and ASD. It was thought that the increased risk of schizophrenia and ASD in individuals with the CX3CR1 Ala55Thr gene variant may be related to decreased fractalkine/CX3CR1 interaction and impairment in this interaction, affecting microglial functions. When we look at the findings obtained from our study, the finding of lower fractalkine levels in individuals with ASD suggests that fractalkine may play a role in the etiopathogenesis of ASD. In order to better understand the role of fractalkine in the etiopathogenesis of ASD, further studies evaluating CX3CR1 receptor expressions and polymorphisms, as well as fractalkine levels, are required.

In our study, no difference was found between the groups in terms of serum TNF-α levels. There are many studies investigating the relationship between ASD and TNF-α in the literature. Tsilioni et al. [9] reported that serum TNF-α levels of children with ASD were higher compared to the control group. On the other hand, studies reporting that TNF-α levels were lower in children with ASD compared to controls are also available in the literature [11]. In a study conducted by Tonhajzerova et al. [10], similar to our study, no difference was found in TNF-α levels in children with ASD compared to controls. It is thought that the inconsistent results of the studies examining the relationship between ASD and TNF-α may be due to methodological differences such as sample selection and sample size in the studies, and more studies are needed in this field.

The results of studies investigating the relationship between ASD and IL-1β are inconsistent. In our study, serum IL-1β levels were found to be significantly lower in individuals with ASD compared to the control group. In a study by Abdallah et al. [16], researchers analyzed neonatal blood samples from 359 individuals with autism and 741 healthy controls. They found that levels of IL-1β were lower in individuals with autism compared to the controls. This suggests that low neonatal IL-1β levels in individuals with ASD may be linked to decreased or insufficient immune cell activity during the neonatal period [16]. Additionally, another study involving neonatal blood samples from preterm children indicated that, while not statistically significant, lower IL-1β levels were associated with reduced mental and psychomotor development scores [46]. Therefore, diminished immune activity in the early neonatal period may negatively impact neurogenesis, memory, and learning abilities [47]. Our study’s findings align with those of Abdallah et al. [16], supporting the connection between low IL-1β levels and immune activity in individuals with autism.

In studies investigating the relationship between ASD and IL-6, IL-6 levels increased in postmortem brain tissues of individuals with ASD and higher IL-6 levels were found mainly in the cerebellar region. Wei et al. [19] reported that increased IL-6 expression in the cerebellum of individuals with ASD leads to disruption in cell adhesion and migration and irregularities in synapse formation between cerebellar granular cells. In the literature, the results of studies examining IL-6 levels in peripheral blood samples of individuals with ASD are inconsistent [15,48]. Similar to the results of our study, Tsilioni et al. [9] found no difference between children with ASD and healthy controls in terms of serum IL-6 levels. However, they reported that serum IL-6 levels decreased and symptoms improved in children with ASD after treatment with luteolin, which has anti-inflammatory effects. This study suggests that IL-6 may be a target for promising therapeutic agents in ASD, and further studies are needed in this field.

In our study, no difference was found between the groups in terms of hs-CRP levels. In a meta-analysis including 9 studies with 592 children with ASD and 604 healthy children, CRP levels in peripheral blood were found to be significantly higher in children with ASD compared to healthy controls [28]. The fact that the results of our study are not compatible with the literature may be related to the evaluation of hs-CRP levels in our study, unlike the studies included in the meta-analysis and the insufficient sample size.

Although this study is the first study evaluating serum fractalkine levels in individuals with ASD in early childhood, it has some limitations. The limitations of our study include being a cross-sectional study, clinical sampling and small sample size. These limitations may prevent the results of our study from being generalisable to the whole population. Potential biases that may occur because our study is a single-centre study and random sampling was not used in sample selection are other limitations of our study. In our study, fractalkine and other cytokines were evaluated in peripheral blood. Since it is not yet known whether the values in peripheral blood accurately reflect the values in the CNS, the fact that fractalkine and other cytokines were not analysed in cerebrospinal fluid or brain tissue is also among the limitations of our study. Since psychiatric comorbidities were excluded from our study, the possible effects of other psychiatric disorders on the results could not be evaluated. Since our study was not a follow-up, the relationship between fractalkine and other cytokines and disease prognosis and treatment response could not be analysed. This study also has some strengths. One of them is that it is the first study in which fractalkine levels and other cytokines were evaluated together in individuals diagnosed with ASD in early childhood. Since age and sex may affect the levels of inflammatory biomarkers, the confounding effect of age and sex was eliminated by matching the groups in terms of age and sex in our study. In addition, the evaluation of hs-CRP as an indicator of inflammation and the exclusion of infection not only clinically but also biochemically is another strength of our study.

In conclusion, our findings highlight the significance of immunological changes in the pathophysiology of ASD. The reduced levels of fractalkine and IL-1β in children with ASD compared to those without suggest that these molecules could serve as potential biomarkers associated with the disorder. Conducting further studies in this area with larger sample sizes, including children of varying ages and those with different comorbid psychiatric disorders, may enhance our understanding of the roles of fractalkine and other cytokines in ASD. This could also aid in evaluating these molecules as diagnostic or prognostic biomarkers and pave the way for more research to clarify the relationship between ASD and inflammation.