Testing the combined effects of probiotics and prebiotics against neurotoxic effects of propionic acid orally administered to rat pups

Abstract The present study investigated the combined effects of mixed probiotic and bee pollen on brain intoxication induced by propionic acid (PPA) in rat pups. Thirty western albino rats were divided into five groups, six animals each: (1) Control group receiving phosphate‐buffered saline; (2) Probiotic and bee pollen‐treated group being administered at the same dose with 200 mg/kg body weight; (c) PPA‐treated group receiving a neurotoxic dose 250 mg/kg body weight of PPA for 3 days; (d) Therapeutic group being administered the neurotoxic dose of PPA followed by probiotic and bee pollen treatment 200 mg/kg body weight; (e) Protective group receiving probiotic and bee pollen mixture treatment followed by neurotoxic dose of PPA. Selected biochemical parameters linked to oxidative stress, energy metabolism, and neurotransmission were investigated in brain homogenates from all the five groups. PPA treatment showed an increase in oxidative stress markers like lipid peroxidation coupled with a significant decrease in glutathione level. Impaired energy metabolism was ascertained via the alteration of creatine kinase (CK) and lactate dehydrogenase (LDH) activities. Dramatic increase of Na+ and K+ concentrations together with a decrease of GABA and IL‐6 and an elevation of glutamate levels in PPA‐treated rat's pups confirmed the neurotoxicity effect of PPA. Interestingly, the mixed probiotic and bee pollen treatment were effective in restoring the levels of glutamate, GABA, and IL‐6 in addition to normalizing the levels of lipid peroxidation and glutathione and the activities of CK and LDH. The present study indicates that mixed probiotic and bee pollen treatment can improve poor detoxification, oxidative stress, and neuroinflammation as mechanisms implicated in the etiology of autism.


| INTRODUC TI ON
Scientific research highlighted the effectiveness of antioxidants, vitamins, minerals, polyunsaturated fatty acids, prebiotics, and probiotics as inducer of optimistic effects on the body. Among these "active ingredients," the prebiotics play a mainly important role because of their ability to selectively stimulate the growth and/ or the metabolic activity of one or more beneficial bacterial species in the host gut. Probiotics, on the other hand, as "live active ingredients" are food supplements that are made up of live microorganisms that react favorably on the host improving the intestinal microbial balance. The probiotic microorganisms mostly belong to Lactobacillus and Bifidobacterium families (Cagnasso et al., 2002;Prasad et al., 1998;Rubio et al., 2014).
Lactobacillus is the largest of lactate bacteria genera, comprising microaerophilic, no spore-forming, catalase-negative, and Grampositive bacteria. Lactobacilli are commonly found in various environment such as plants and soil, human, and animal mucosal surfaces as well as in dairy products (Lahtinen et al., 2011). Selection of potential probiotics requires numerous characteristics (Guo et al., 2010).
In fact, the microorganism should be nonpathogenic and could tolerate the physiological concentrations of bile and the acidic pH in stomach and, therefore, to survive in the gastrointestinal tract; must present antagonistic activity against intestinal pathogens and should display desirable surface hydrophobicity for colonization (Mishra & Prasad, 2005).
It is all the more impressive to realize how some probiotics succeed in improving their "performance" by cooperating together.
Combined effects between prebiotics and probiotics have been demonstrated to create the optimum substrate necessary for the growth of several bacteria species, in particular, Bifidobacterium which is well known to be lower in autism spectrum disorder (ASD) patients. Tomova and collaborators (2015) investigated the impact of mixed probiotic (Children Dophilus) administration for 4 months on gut microbiota (GM) composition in ASD patients. Authors were able to observe an increase in bifidobacterial numbers with a modulation of the Bacteroidetes/Firmicutes ratio (Tomova et al., 2015). Prebiotics as food ingredients are selectively metabolized by indigenous beneficial bacteria therefore positively modulating GM. However, their impacts are not well documented in autism. MacFabe et al. (2007) and El-Ansary et al. (2012) demonstrated that intraventricular infusion or oral administration of PPA can modify both behavior and brain in the laboratory animals in a way that is consistent with human ASD symptoms. The neuropathological, behavioral, and biochemical results in the MacFabe PPA model provide further support for the hypothesis that autism could be a systemic metabolic encephalopathic process affecting the brain. The similarities in oxidative stress and innate neuroinflammatory changes between their animal model and human ASD cases could exhibit comparable immune metabolic or mediated processes (Moritz & Ayus, 2010) indirectly or directly linked to PPA. Particularly are their observations of broad alterations in lipid peroxides and glutathione (GSH) levels which may provide a common mechanism for elevated environmental sensitivity to diverse environmental compounds and increased oxidative stress (Hiratani et al., 1997). El-Ansary et al. (2017) evidenced the induction of imbalanced excitatory/inhibitory function in PPA rodent model. This information motivates our interest to test the combined effects of mixed probiotic and bee pollen in ameliorating the impaired biochemical features induced with PPA in rat pups through the measurement of K + , Na + , CK, LDH, interleukin-6 (IL-6), glutamate, ɣ-aminobutyric acid (GABA) in tissue homogenates of PPA intoxicated, and treated or protected with a symbiotic (probiotic+bee pollen).

| Animals
All animal experiments were approved by the Ethics Committee of the College of Science at King Saud University and were carried out according to the national guidelines for the use and care of animals. Thirty male Wister albino rats (60-70 g) were kept in cage (42.5 cm × 26.5 cm × 14.5 cm) under standard laboratory conditions (temperature 23°C, humidity 37%, and light for 12 hr) and the food supply was from Grain Silos and Flour mills organization. Rats were randomly divided into five groups of six animals as shown in Figure 1. Carbon dioxide-anesthetized rats were decapitated at the end of feeding trials, and their brains were removed from the skull and dissected into small pieces. After homogenization in bidistilled water (1:10, w/v), brain samples were stored at −80°C.

| Biochemical analyses
The activity of catalase was determined according to the method of Maehly and Chance (1954) Szasz et al. (1976) and Amador et al. (1963), respectively, by following the rate of NADH formation, which is directly proportional to the sample LDH or CK activity, at 340 nm.
The levels of GSH and lipid peroxidation in the brain samples were measured spectrophotometrically according to the protocols described by Beutler (1963) and Ruiz-Larrea et al. (1994) Quality control assays were performed to evaluate experimental reproducibility through the inter-and intra-assay coefficients of variability (%CV).

| Statistical analysis
The statistical package for the social sciences (SPSS) was used to analyze the data of the current study. Obtained results are shown as mean ± standard deviation (S.D). Statistical correlations and comparisons between parameters were performed using Pearson's correlation coefficient (r) and the independent t test, respectively. To evaluate brain neurotoxicity in animal modeling, the receiver operating characteristics (ROC) curve and the area under the ROC curve (AUC) were used as a fundamental tool. Only p values ≤0.05 were considered significant. Table 1 showed the brain homogenates levels of N + , K + , LDH, CK, IL-6, GABA, GSH, lipid peroxidation, glutamate, catalase, and GABA/ glutamate ratio in addition to their percentage change (also shown in Figure 2) relative to control of all the tested groups. Table 1 and  Table 2 showing either positive or negative correlations and may clarify the inter-relationship between different studied parameters as etiological mechanisms associated with neurotoxicity of the brain.

| RE SULTS
ROC analysis is also presented in Table 3 showing AUC, specificity, and sensitivity of all measured parameters.

| D ISCUSS I ON
Sodium (Na + ) and Potassium (K + ) are two vital cations, being the most abundant cations in the extracellular and the intracellular fluids, respectively. In neurons, the flow of sodium and potassium in and out of the cell through sodium-potassium pump creates an action potential (AP) (Forrest, 2014;Pohl et al., 2013).
The elevation of sodium and potassium levels in PPA-treated rats is seen in Table 1 and Figure 2a, and that could be explained by Good (2011) who hypothesized a strong connection between low sodium levels and autism. Autistic children suffer from hyponatremia caused by diarrhea. The author attributed hyponatremia to the increased testosterone that induce the release of vasopressin which makes the kidney reabsorb water and causing a dilution in sodium levels (Córdoba et al., 2010).
Lactate dehydrogenase is a cytoplasm enzyme present in brain, is released into the blood when the brain is injured, and the rise of its serum level is usually concomitant with decrease of brain LDH relative to the degree of brain damage. In the present study, LDH did not demonstrate any significant alterations either in response to PPA neurotoxicity or the therapeutic and protective effects of combined treatments (Table 1 & Figure 2a). This is not in good agreement with the study of Al-Orf et al. (2018), which recorded significant decrease F I G U R E 1 Schematic presentation of the animal model experimental design. Probiotics (PROTEXIN ® ), a product of Probiotics International Limited (UK). Bee pollen (NZ Bee Pollen Granules) was product of Happy Valley (New Zealand). Propionic acid was Sigma-Aldrich (USA) product.

Group 2 (Bee pollen and probiotics treated group)
Rats received probiotics (Protexin ®) & bee pollen mixture orally at same dose of 0.2 g/kg body weight/day for 4 weeks.

Group 3 (Propionic acid treated group)
Rats received neurotoxic dose of PPA orally (250 mg/kg body weight/day) in the last 3 days of the whole duration (4 weeks).

Group 4 (Therapeutic group)
Firstly, rats received neurotoxic dose of PPA orally (250 mg/kg body weight/day) for 3 days.
After that, they received probiotics & bee pollen mixture orally at the same dose (0.2 g/kg body weight/day) for 25 days.

Group 5 (Protective group)
Firstly, rats received probiotics & bee pollen mixture orally at the same dose (0.2 g/kg body weight/day) for 25 days.
Followed by a neurotoxic dose of PPA orally (250 mg/kg body weight/day) for 3 days.

TA B L E 1 Mean
± SD of all the measured parameters in brain homogenate of treated rats' pups compared with control group Parameters Groups N Min.
Max.  in brain homogenate of PPA-treated rats. This can be attributed to the efflux of IL-6 from brain to blood through the disrupted bloodbrain barrier (Banks et al., 1995;Chen et al., 1997). This can find support through multiple studies which prove elevation of IL-6 in plasma of autistic patients (Guloksuz et al. 2017;Inga Jácome et al. 2016;Saghazadeh et al. 2019).

Mean
In Table 1 and Figure 2b, the protective potency of the mixture was more effective compared with therapeutic, and the protective group showing p value at the margin of statistical significance (p < .075) close to being statistically significant (Dahiru, 2008). In Table 1  An increase of GSH levels is seen in therapeutic group (Table 1 and  Meldrum, 2000). Glutamate have been contributed in Ca 2+ homeostasis, developmental plasticity, and many neural physiological processes, directly and indirectly, as well as its relation with brain disease as epilepsy, stroke, Alzheimer's disease, and autism since many studies reported atypical of its signaling pathways (El-Ansary et al., 2017;Mattson, 2008).
In Table 1  ɣ-aminobutyric acid, an inhibitory neurotransmitter, is a product of glutamate decarboxylation by glutamic acid decarboxylase enzyme (Rowley et al., 2012). Unlike glutamate, it prevents Ca 2+ passage into the neurons and therefore inhibits the neural excitability (Li & Xu, 2008). It usually induces hyperpolarization of neuronal membranes, as an inhibitory signaling thus balance the depolarizing excitotoxic effect of glutamate. Aside from being an inhibitory neurotransmitter, GABA is also involved in cell's differentiation, proliferation and death, GI motility, and immune response. Its dysregulation is related to psychiatric disorders as depression and anxiety, TA B L E 2 Pearson's positive and negative correlations between the measured parameters Parameters
b Negative correlation.
*Correlation is significant at the .05 level.; **Correlation is significant at the .01 level.

TA B L E 3
Analysis of receiver operating characteristics (ROC) of the parameters measured in the brain homogenate of the treated rats' pups as well as other neurological disorders like seizure and autism (Gaetz et al., 2014;Rowley et al., 2012).
In Table 1 and Figure 2c, a significant decrease of GABA levels in PPA-treated rats pups (p < .005) was supported by El-Ansary et al. (2012) in which brain homogenates of PPA-treated rats demonstrated similar decrease in GABA levels compared with control. This is in good agreement with studies which showed a disturbance in GABAergic neurotransmission in autistic patients and that is relevant to their hyperactivity (Cochran et al., 2015;El-Ansary & Al-Ayadhi, 2014;Pretzsch et al., 2019).
In Table 1 and Figure  high content of flavonoid which was found in brain tissue of rodents after oral gavage therefore indicating its ability to pass BBB (Franco et al., 2010;Xue et al., 2012).
Balanced ratio between GABA as inhibitory neurotransmitter and glutamate as excitatory neurotransmitter is essential in maintaining a normal neural function, since disruption of their levels would cause neurological disorders and social impairments as autistic patients (Ford et al., 2017). According to the data obtained from glutamate and GABA (Table 1 and Figure 2c), their ratio (Table 1 and Figure 3c) is significantly decreased in PPA-treated rats pups compared with control (p < .005), and this is in good agreement with previous works in which they found low ratio of GABA/glutamate in PPA-treated rats (El-Ansary et al., 2017 as well as GABA/glutamate levels were decreased in frontal lobe of autistic patients compared with control (Harada et al., 2011). Studies have exhibited such a disturbance in GABA/glutamate in ASD patients which appears as social, memory-related, sensory, and emotional imbalance; and this may be explained by Casanova et al. (2006) work which demonstrated changes in the number of glutamatergic and GABAergic neurons via analysis of postmortem tissues from age-matched ASD patients (El-Ansary & Al-Ayadhi, 2014;Pizzarelli & Cherubini, 2011;Rosa et al., 2016). As previously mentioned how the mixture of bee pollen and probiotic was effective as protective and therapeutic in restoring the levels of glutamate and GABA, their GABA/glutamate ratio was elevated as well for the same reasons.
The significant correlations between the biochemical statuses are observed in Table 2. CK as energy replenishing marker is positively correlated with Na + and K + ; as two cations closely related to Na + / K + ATPase, as an energy-consuming enzyme, that is. the more active Na + /K + ATPase, the highly active is CK to replenish the depleted ATP (Al-Mosalim et al., 2009). However, CK is negatively correlated with GSH as oxidative stress marker and that can be elucidated by a study which observed that GSH supplementation has the ability to reduce the high levels of CK caused by impairment of plasma redox status in hypoglycemia, a risk factor in autism (Hoirisch-Clapauch & Nardi, 2019;Jiang et al., 2007). Na + is negatively correlated with IL-6, a neuroinflammation marker. This could find support through the study of Li et al., (2014) which demonstrated the IL-6 suppression effect in voltage-gated sodium channel (VGSC) currents which also led to suppression of spike amplitude in rat spinal cord neurons.
Likewise, Na + is negatively correlated with GSH. Contrarily, Clark et al. (1996) demonstrated that the GSH depletion in the brain of rats caused by hyponatremia is recovered when Na + levels are normalized in vivo.
Neuroinflammation marker (IL-6) and oxidative stress marker (GSH) are positively correlated as expected; due to their multiple neuroprotective effects besides the regulatory role of GSH in neuronal hippocampal cells (Schmidt et al., 2005). The obtained positive and negative correlations may clarify the inter-relationships between different studied parameters as etiological mechanisms associated with neurotoxicity of the brain. In conclusion, bee pollen and probiotic demonstrated remarkable combined effects in ameliorating the neurotoxic effect of PPA.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analyzed during this study are included in this published article.

ACK N OWLED G M ENTS
The authors would like to thank Deanship of scientific research in King Saud University for funding and supporting this research through the initiative of DSR Graduate Students Research Support (GSR).

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.