In this study, the relationship between cognitive functions and BDNF, GDNF, NGF and Klotho levels in 41 patients diagnosed with schizophrenia who had acute psychotic exacerbation and in 43 healthy controls was examined.
When the PANSS, CGI, IGD and BPRS scales of the patients on Day 1 and Day 20 were evaluated, the PANSS and BPRS scale scores on Day 20 were found to be significantly lower than the scale scores on Day 1. When the IGD scores of the patients were evaluated, the scale scores on Day 20 were higher than those on Day 1. The functionality levels of the patients were also shown to increase. When the side-effect sub-parameter was examined in the CGI scale, no statistically significant difference was found, but the severity of the disease decreased; an improvement in the sub-parameter scale score was observed. These scores suggest that the patients were in a recovery period and that they were benefiting from the treatments they were receiving. The clinical features of the patients and their differences from features of the control group participants were generally consistent with the typical clinical features and differences of schizophrenia patients known in the literature and demonstrated in previous studies.
Comparison of serum marker levels of the patient and control groups
When the BDNF, GDNF, NGF and Klotho levels of the patient and control groups were compared, the levels in the control group were significantly higher. When these levels were compared in patients on Day 1 and Day 20, it was determined that they were higher on Day 20, yet only BDNF levels reached statistical significance. This suggests that patients may experience an increase in neuroplasticity biomarkers with treatment, which may in turn slow down the neurodegenerative process in these patients. However, further studies with a larger sample are needed to verify this effect. In this area, BDNF especially has been the most studied biomarker with respect to the neurodevelopmental hypothesis [
34].
In this study, no correlation was found between neurotrophic levels, but various antipsychotics were used and the sample was too small to make subgroups based on one kind of antipsychotic; one type of antipsychotic may have a different impact than another so this impact may be a reason for the lack of correlation. In one study that investigated the effects of antipsychotic drugs on serum and plasma BDNF levels, a positive correlation was found between clozapine dose and BDNF levels in schizophrenia patients using clozapine but, similar to the present study, the same correlation was not found with equivalent doses of typical antipsychotics [
35]. While most studies on serum BDNF levels in patients with schizophrenia have found low serum BDNF levels [
9‐
13,
36] [
14], in a few studies, BDNF levels were found to be high [
15,
16]. In one meta-analysis, serum BDNF levels of patients with schizophrenia were found to be lower than those of healthy controls, regardless of drug use [
34]. Although the relationship between changes in the brain and serum BDNF levels due to BDNF expression is not fully understood, it is thought that a genetic predisposition towards schizophrenia may affect BDNF levels. In addition, a few studies have been conducted on the effect of chronic neuroinflammation on BDNF levels in patients with schizophrenia [
37]. In addition, it should not be overlooked that BDNF levels are affected by lifestyle factors such as drug use, age, physical exercise, diet and stress, which may be significant for schizophrenia patients [
38].
In a meta-analysis examining NGF levels in schizophrenia, decreased peripheral serum NGF levels were found to be associated with the psychopathology of schizophrenia (Qin, Wu et al. 2017). In studies conducted in this area, it has been reported that serum NGF levels are decreased in schizophrenia patients compared to healthy controls; NGF levels and NGF receptor levels have also been shown to decrease in cerebrospinal fluid (CSF) [
39]. In one study that investigated NGF levels and antipsychotic drug use, it was determined that antipsychotic drug use increased NGF serum levels [
40]. In the present study, serum NGF levels were found to be lower in schizophrenia patients compared to the control group participants, and an increase in serum NGF levels was observed on the 20th day of the treatment as compared to the 1st day, in accordance with the literature. But in the present study, the clinical scales of the patients and Klotho, BDNF, GDNF and NGF levels were assessed on the 1st and 20th days. When the relationship between the two was evaluated, no statistical significance was found. This calls into question the effect of these markers on clinical symptoms. The relationship between low serum NGF levels and structural changes in the brain in schizophrenia patients points to a decrease in gray matter volume, especially in the left midcingulate cortex, in addition to low hippocampal volume [
41].
In one study that explored GDNF serum levels in different psychiatric disease groups, serum GDNF levels were found to be lower in schizophrenia patients compared to control group participants [
42]. Serum GDNF levels of first-episode psychosis patients who did not use medication were followed for 8 weeks and measured in the 2nd, 4th, 6th and 8th weeks; GDNF levels, which were initially found to be lower in schizophrenia patients than in healthy controls, increased with the use of antipsychotic drugs (Xiao, Ye et al. 2016). In a study evaluating GDNF serum levels and clinical scales, only female patients were included. In this study, no significant difference in serum GDNF levels were found in the healthy control group (Skibinska, Kapelski et al. 2017). In the present study, serum GDNF levels were found to be lower in schizophrenia patients than in the healthy controls.
Klotho is a biomarker that has recently been studied in psychiatric disorders, but there is a limited number of studies measuring these serum levels in schizophrenia patients. GDNF is secreted from glial cells, and morphological and functional defects of glial cells may be associated with the pathogenesis of schizophrenia. Finding low NGF serum levels in patients with schizophrenia can be considered as both a result and a treatment goal [
43]. The current study evaluated serum levels of biomarkers but it was shown that serum levels, CSF levels and brain mRNA levels and genes, and even different brain part properties of these molecules, may change in schizophrenia and are important to explaining the pathogenesis of disorders [
44]. As such, further studies on CSF and an investigation of their relation to brain areas may enhance understanding in this area.
In one study, patients with schizophrenia who were experiencing either acute exacerbation or remission were compared with healthy controls. Klotho levels were found to be higher in schizophrenia, but no statistical significance was found [
7]. However, in other studies, Klotho levels were found to be significantly higher in schizophrenia [
25]. In the present study, Klotho levels were found to be statistically significantly lower in patients with schizophrenia than in healthy controls, with a statistically insignificant increase on the 20th day of treatment. These results suggest a dynamic relation between schizophrenia and Klotho, but it remains difficult to determine the reason for the changes in the levels of Klotho with the current data. In this study, the standard deviation values of BDNF, GDNF, NGF and Klotho levels were higher than the mean values and demonstrated a wide range. The reason for this result may be due to the differences in the basal metabolic rates of the individuals included in the study. Basal metabolism is affected by many factors, such as age, gender, genetic factors, body mass index, body fat ratio, physical activity, hormonal changes and drugs [
45]. Patients and control group participants with different weights and physical activities were included in the study. Antipsychotics from different groups that affect metabolism were used in the patient group, and all of these factors may have caused the standard deviation of BDNF, GDNF, NGF and Klotho to be high.
Association with cognitive symptoms
All cognitive domains are commonly affected in schizophrenia. In meta-analysis studies conducted in this area, it was determined that attention, executive functions, memory functions, language functions and global cognitive functions were impaired in schizophrenia patients compared to healthy controls [
46,
47]. In the current study, the Stroop test and Weschler Visual Memory Test were applied to both the patient and healthy control groups, and correlation analyses were made with serum levels of BDNF, GDNF, NGF and Klotho. While no correlation was found between neurocognitive tests performed on the 1st and 20th days for patients and neuroplasticity biomarkers, a significant correlation was found in the control group. In a previous study, different from the present results, Klotho was found to be higher in patients with schizophrenia, and it was positively correlated with cognitive functions [
25]. In one study conducted in this area, animal modelling with BDNF was performed, and it was shown that BDNF is involved in higher-level functions of the brain, such as perception and regulation of emotions, learning, memory and other neurocognitive functions [
48]. In a recent meta-analysis, correlations between serum BDNF and neurocognitive functions were evaluated. Studies involving reasoning and problem solving, attention and all neurocognitive phenotypes were included. In this meta-analysis, while no significant difference was found between attention and serum BDNF levels, reasoning and problem solving and all neurocognitive tests were found to be positively correlated with serum BDNF levels [
49]. Studies in the literature have found a relationship between schizophrenia and BDNF in terms of cognitive functions, but this relationship was not confirmed in the current study. This may be related to the number of samples or to the fact that patients were evaluated in the acute phase.
In this study, when neuroplasticity biomarkers and neurocognitive tests were compared in the healthy control group, a negative correlation was found between the 1st and 40th minute test scores on the Weschler Visual Memory Test and GDNF levels. In one recent study, GDNF levels were found to be associated with working memory in healthy controls and attention deficit in schizophrenia patients [
17]. Working memory is called short or recent memory, and it refers to the entirety of processes in which necessary information is kept in mind and transferred to long-term memory [
50]. The Weschler Visual Memory Test also measures learning and memory functions. The increase in GDNF levels in the healthy control group may have adversely affected working memory. An alternative perspective might be compensatory increases in GDNF levels at the receptor level due to resistance.
In this study, when neuroplasticity biomarkers were compared with neurocognitive tests in the healthy control group, a positive correlation was found between all biomarkers measured between the Stroop test’s time to read coloured words and the time needed to say the colour of coloured words. This situation led to the conclusion that the healthy control group responded by allowing more time to think rather than responding impulsively. In the Stroop test, a positive correlation was found between misreading the colour of coloured words and GDNF, and it was thought that this might be due to distracted attention during the information processing process. It seems that GDNF works through different systems in patients with schizophrenia and in healthy controls.
Correlation and predictiveness between markers
When BDNF, GDNF, NGF and Klotho levels were evaluated in all groups separately, a significant correlation was found between all markers on the 1st and 20th days in the patients and in the control group participants. NGF is secreted endogenously by peripheral cells, such as immune and nerve cells, smooth muscle cells, melanocytes, fibroblasts and Schwann cells [
51].. Klotho is predominantly found in the kidneys but also in the brain, lungs, skeletal muscles, bladder, testes and ovaries [
21]. When these two molecules, Klotho and NGF, were examined, it was shown that they were produced by many different organs and cell systems in the peripheral system and played a role in many diseases from neuroinflammation to cancer, such as osteoporosis and aging [
20,
52]. These molecules, which are affected by many different systems, may negatively affect GDNF and BDNF via different mechanisms. BDNF and GDNF are mostly involved in central nervous system functions and affect each other positively. In light of all of this information and findings, BDNF seems to work in cooperation with all of the molecules studied.
In the regression analysis, when the healthy control group was evaluated, BDNF, GDNF, NGF and Klotho were found to be predictive of each other. This result implies a strong relationship between BDNF, GDNF, NGF and Klotho, but the mechanism and reasons for this relationship could not be explained. Considering the regression analysis of the patient group and the healthy control group, the studied molecules may be working with different mechanisms in each group.
Limitations and strengths
The current study had some limitations. Cognitive functions, for example, were evaluated using a limited number of tests. For this reason, it was possible to comment only on certain areas in terms of cognitive functions. It is thus important to repeat the study with the results of the tests performed by evaluating neurocognitive functions in different areas. Serum BDNF, GDNF, NGF and Klotho levels are affected by many variables, and special methods are therefore needed to study these neuropeptides, as they are otherwise difficult to use and have a high margin of error. Considering the prevalence of schizophrenia, increasing the sample size in future studies may increase the strength of the results of the current study.
The strengths of the present study are that, when compared to other studies in the literature, a different variety of neurocognitive tests were included, the diagnosis of schizophrenia was confirmed by an experienced psychiatrist according to DSM-V criteria, and a sample that could be considered homogeneous, in which confounding factors such as substance use and physical diseases were largely excluded, was employed. Serum levels of BDNF, GDNF and NGF in schizophrenia patients have been evaluated previously. In the current study, three neurotrophic factors were evaluated together with neurocognitive tests for the first time. Klotho has started to be studied as a new potential biomarker in psychiatric diseases, and it is thus important to examine its relationship with neurocognitive tests. To date, no previous study has evaluated four biomarkers together, and in this sense the present study makes valuable contributions to the literature.