Peripheral inflammation is associated with alterations in brain biochemistry and mood: evidence from in vivo proton magnetic resonance spectroscopy study

Abdul Nashirudeen Mumuni; Rajeev Krishnadas; Jonathan Cavanagh

doi:10.1515/pac-2022-1119

Article Publicly Available

Peripheral inflammation is associated with alterations in brain biochemistry and mood: evidence from in vivo proton magnetic resonance spectroscopy study

Abdul Nashirudeen Mumuni , Rajeev Krishnadas and Jonathan Cavanagh

Published/Copyright: March 22, 2023

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal Pure and Applied Chemistry Volume 95 Issue 3

Abstract

Tumor necrosis factor (TNF)-α is a protein implicated in the prognosis of psoriatic arthritis (PsA), an autoimmune disease associated with episodes of clinical depression (CD). CD has been linked to abnormalities in the forebrain and subcortical structures, which can be probed noninvasively using proton magnetic resonance spectroscopy (¹H-MRS). Thus, this ¹H-MRS study assessed biochemical differences between healthy subjects and PsA patients in the frontal and bilateral hippocampal brain regions. Biochemical and mood responses in the anterior cingulate cortex and bilateral hippocampi following the administration of anti-TNF-α medication to the PsA patients were also assessed. Fifteen volunteers in each of the PsA and control groups participated in the study. Patients underwent MRS examination before and after 6–8 weeks of anti-TNF-α medication. Mood was assessed at baseline and after medication using the Beck’s Depression Inventory (BDI). Psychiatric symptoms were ruled out in all volunteers using the 12-item General Health Questionnaire. MRS studies were conducted using the PRESS sequence at 3 T. Spectral and statistical analyses were conducted using the LC Model and Minitab software, respectively. Metabolite levels were expressed as ratios, relative to the creatine peak. The PsA patients showed higher left hippocampal baseline Choline/Creatine ratio than controls (p = 0.014). After medication, decrease in frontal brain Choline/Creatine ratio was associated with decrease in BDI scores (r = 0.628; p = 0.025), while frontal brain NAA/Creatine ratio was significantly lower in the PsA patients compared to the controls (p = 0.023). No other significant differences were found in all comparisons, including metabolite ratios between the left and right hippocampi. The results of this study provide some evidence for biochemical alterations in the mood regulating segments of the brain, which has consequences for mood states of the affected patients.

Keywords: Brain; choline; creatine; hippocampus; magnetic resonance spectroscopy; major depression; psoriatic arthritis; tumor necrosis factor; VCCA-2022

Introduction

Inflammation refers to the body’s reaction towards the elimination of harmful agents or pathogens [1]. Even though inflammation is not an infection, infectious pathogens could trigger the inflammatory response of the body [1, 2]. Even in the absence of a stimulus, the inflammatory process may be initiated, releasing tissue-damaging pro-inflammatory chemicals [3]. Through a series of biochemical events, an acute response gradually leads to a chronic condition, which is associated with progressive alterations of the nature of cells at the inflammation site [3].

In the chronic phase, often known as autoimmune disease, tissues simultaneously undergo destruction and healing processes [1, 4]. The commonest inflammatory diseases are multiple sclerosis and arthritis. While multiple sclerosis affects the brain directly by causing plaques, arthritis attacks the extremities of the body. The commonest types of arthritis are rheumatoid (affects the joints) and psoriatic (affects the skin) arthritis.

This study focused on psoriatic arthritis. Notable symptoms of psoriatic arthritis include progressive growth of silvery-white or red colored skin at affected body parts, scaly patches on the skin, joints (elbows and knees), and other body parts such as the scalp, fingernails and genitals. The condition usually develops between the ages of 30 and 50 years, with varied symptoms. Even though the exact cause of psoriatic arthritis is unclear, it may be associated with genetics, and is worsened by lifestyle habits such as smoking and alcoholism [5, 6]. There are many different treatment options for psoriatic arthritis, yet it is still recurrent after each of these regimens [7].

In the development of psoriatic arthritis, pathogen-specific receptors are activated to induce the release of proinflammatory molecules namely, cytokines [e.g., tumor necrosis factor (TNF), interleukin-1 (IL-1) and interleukin-6 (IL-6)] and chemokines. These molecules facilitate the inflammatory process by opening up pores through vascular endothelial tissues including the blood-brain barrier [1, 6]. It has been suggested from animal studies of chronic inflammation that peripheral organ and central immune systems communication takes place, mediated by TNF-α which consequently affects the mood regulatory function of the brain in the affected individual [6, 8], [9], [10].

Peripheral TNF-α induces microglia to release harmful protein molecules that cross the blood-brain barrier [6, 8]. Thus, any intervention that can stop microglial activation or block the microglia receptors could alter the link between the central nervous system and the periphery; this may consequently improve the mood regulatory function of the brain [8].

Changes in brain function associated with inflammation have been observed by brain imaging studies (specifically, functional magnetic resonance imaging [11], positron emission tomography [12] and single-photon emission computerized tomography [13]) to follow the pattern of reduced activity in the frontal brain structures (notably the dorsolateral prefrontal and anterior cingulate cortices) and increased activity in the limbic structures (notably the hippocampus, amygdala, insula and thalamus) [11, 12]. This pattern of activity has been reported by previous proton magnetic resonance spectroscopy studies to be associated with alterations in neurochemistry [14], [15], [16], [17], [18]. The observed mood changes experienced over time by affected individuals may be explained by the physiological alterations in the mood regulating brain regions. These findings suggest that inflammation has a molecular basis initiated by biochemical processes within cells.

Since the brain is affected in peripheral inflammation, normal cerebral metabolism could be affected in psoriatic arthritis. The subtle neurochemical changes may not be observed in structural diagnostic images of the patients. Meanwhile, proton magnetic resonance spectroscopy (¹H-MRS) detects these subtle neurochemical changes earlier prior to the manifestation of their associated effects in magnetic resonance images and other diagnostic imaging modalities. ¹H-MRS is a nuclear magnetic resonance technique that records biochemical profiles (displayed as peaks) of a region of interest non-invasively and in vivo. So far, a total of about thirty-five metabolites are detectable in vivo and in vitro by ¹H-MRS [19]. Large scale, longitudinal and multicenter studies are therefore required to establish neuro-metabolite correlates of psoriatic arthritis, with high degree of certainty. There are currently few studies reporting on alterations in neuro-metabolite levels in patients diagnosed with multiple sclerosis [20], [21], [22], rheumatoid arthritis [7, 9], and psoriatic arthritis [7]. To the best of our knowledge, no previous study has investigated the link between psoriatic arthritis and mood of patients, in the context of anti-TNF-α (anti-inflammatory) medication.

In this study, it was hypothesized that peripherally increased amount of TNF-α in psoriatic arthritis will signal the brain, causing an increase in neuronal cell membrane breakdown. This will reflect in an increase in choline concentrations. Blocking TNF-α with medication is expected to cause a reversal of the above process.

The objectives of this study were therefore to use the ¹H-MRS technique to study the anterior cingulate cortex and bilateral (left and right) hippocampi in order to assess the differences in neuro-metabolite levels between healthy controls and psoriatic arthritis patients at baseline and following anti-TNF-α medication; to compare the effect of anti-TNF-α treatment on neuro-metabolite levels in psoriatic arthritis patients; and to assess the association between alterations in neuro-metabolite levels and mood scores of psoriatic arthritis patients.

Methods

Study participants

The protocol for this study was approved by the West of Scotland Research Ethics Committee 3 (WoSREC3, reference number 09/S0701/77).

Psoriatic arthritis patients [males/females = 7/8, mean age (SD) = 44 (10) years] seeking medical care at the Rheumatology departments of the NHS Greater Glasgow and Clyde Hospitals were recruited into the study. No patient was on medication for the treatment of psoriatic arthritis at study entry. Following a baseline ¹H-MRS scan, each patient received Etanercept (for treatment of inflammation) intravenously for 6–8 weeks before the next ¹H-MRS scan.

Healthy volunteers [males/females = 7/8, mean age (SD) = 37 (12) years] were drawn from Glasgow city through e-mail contacts and phone calls.

All participants gave written consent prior to study entry after being taken through the procedures of the study. Participants were screened for neurological and/or psychiatric conditions using the 12-item General Health Questionnaire (GHQ-12) [23]. They were also screened for MRI contraindications using a standard MRI safety questionnaire. Brain MR images were reviewed by a consultant neuroradiologist.

Assessment of mood scores

Each psoriatic arthritis patient answered the Beck’s Depression Inventory (BDI) prior to each of the baseline and post-medication ¹H-MRS scans. The BDI is a 21-item self-administered screening questionnaire (https://www.ismanet.org/doctoryourspirit/pdfs/Beck-Depression-Inventory-BDI.pdf); each item on the questionnaire has four questions ordered in worsening mood conditions which are scored from 0 to 3.

The current mood score of the patient is marked in response to each question, and then all the scores are summed up by tallying the recorded scores. A score of 3 on all the 21 questions results in the highest possible score of 63 on the test; the lowest possible score of 0 results from a score of 0 on each item of the 21 questions. Severity of depression is interpreted (Table 1) from the total mood score of a patient. If a patient persistently scores 17 or greater, immediate medical attention is required.

Table 1:

Interpretation of the Beck’s depression inventory.

Total BDI score	Level of depression
1–10	These ups and downs are considered normal
11–16	Mild mood disturbance
17–20	Borderline clinical depression
21–30	Moderate depression
31–40	Severe depression
Over 40	Extreme depression

MRI/MRS acquisition

The ¹H-MRS experiments were conducted on a 3.0 T General Electric Signa HD MRI/MRS scanner (software version 12.5; Milwaukee, WI, USA) equipped with an eight-channel receive-only head coil (Fig. 1).

Fig. 1:

MRI/MRS machine (a) connected to an eight-channel receive-only head coil (b) for acquisition of structural brain MRI (c) and localization of MR spectra to the anterior cingulate cortex in the MRI (d). The patient’s head rests on a head coil lining (e).

Structural MRI acquisitions involved T₁-weighted FLAIR sagittal and T₂-weighted FSE propeller axial scans, each with the following image acquisition parameters: TR = 25000 ms, FOV = 24 × 24 cm², 21 oblique slices parallel to the hippocampus, each of slice thickness 4 mm with no slice gaps.

MR spectra were acquired with the PRESS localization sequence involving CHESS water suppression and 128 signal averages. Voxels were planned on the anterior cingulate cortex (Fig. 2) using TE/TR = 35/2000 ms with variable voxel sizes ranging between 8.0 and 14.5 mL specific to each patient’s head. The voxels covered the dorso-ventral region of the anterior cingulate cortex, including parts of the two halves of the frontal brain near the anterior portion of the corpus callosum (Fig. 2).

Fig. 2:

Voxel positioning in the anterior cingulate cortex, in the sagittal (a) and axial (b) views.

Fig. 3:

Voxel positioning in the right (a) and left (b) hippocampi of the same volunteer.

Spectra were acquired from the right (Fig. 3a) and left (Fig. 3b) hippocampi using TE/TR = 144/2000 ms. The voxel sizes were altered between 5.3 and 13.2 mL to fit each hippocampal size, and were placed such that the longest side covered the length of the hippocampus.

Varying the voxel sizes among study participants in a study is an established procedure [24]. The echo-time of 144 ms for the hippocampal acquisitions was long enough to remove unwanted signals from cerebrospinal fluid and lipids, which are absent in the frontal brain. At the relative TE values for the two regions of interest, whereas five metabolite ratios (Table 3) were reliably measured from the anterior cingulate cortex, only two metabolite ratios (Tables 4, 5) were reliably measured from the hippocampal regions.

Spectral analysis

Spectral analysis was performed using the Linear Combination (LC) Model software package [25], which expresses metabolite peak areas relative to the peak area of creatine (Cr). The software is a commercially available user-independent spectral fitting program.

The raw data is automatically eddy-current corrected using the unsuppressed-water signal, and is then phase-corrected before fitting each metabolite spectra with a linear combination of model spectra of established concentrations, called the basis set. The transmitter and receiver gains recorded in the raw data are further used to adjust the estimated spectral areas to the areas of the model spectra.

The components of the basis set saved in the LC Model version (6.2-4A) used in this study included sixteen metabolites, including those of interest in this study. The molecular simulation library had two basis sets of very high signal-to-noise ratio; one simulated for TE = 35 ms, TR = 5000 ms, NSA = 128, and another for TE = 144 ms, TR = 5000 ms, NSA = 128. The metabolite ratio estimates were considered reliable at standard deviations of 20 % or less, according to previous reports [26, 27].

Estimation of changes in BDI scores and metabolite ratios

For only the psoriatic arthritis patient group, percentage differences (%ΔX) between baseline (X_pre) and post-medication (X_post) estimates of BDI scores and metabolite ratios were calculated from:

(1) % Δ X = [ ( X post − X pre ) / X pre ] × 100 %

From eq. 1, a negative percentage change in BDI score is indicative of mood improvement, whereas a positive percentage change shows that the patient’s mood has rather worsened after medication.

Statistical analysis

Comparisons between baseline and post-medication BDI scores and metabolite ratios in the patient group, and between the left and right hippocampal metabolite ratios of both patients and healthy controls were performed using the paired t-test. Comparisons between patients and healthy controls, and between males and females for metabolite ratio and age differences were performed using the two-sample t-test. Association between changes in BDI scores and metabolite concentrations was assessed by Pearson correlation. All analyses outcomes were significant at a critical value of p < 0.05. Statistical analyses were performed using the Minitab software package (version 17; Minitab Inc., State College, Pennsylvania, USA).

Results

There were no significant age differences between the healthy controls and psoriatic arthritis patients (p = 0.093), and between males and females in the patient (p = 0.686) and healthy control (p = 0.904) groups. Consistent voxel placements were ensured between the two ¹H-MRS acquisitions on each volunteer by referring to screenshots captured during the baseline scans and voxel position coordinates.

Mood assessment outcomes

The sex-age distribution, BDI scores before (BDI_pre) and after (BDI_post) medication, and percentage changes in BDI scores (%Δ BDI) of the psoriatic arthritis patients are shown in Table 2. Names of the patients are represented by identifier codes (P01, P02, P03, etc.).

Table 2:

Mood scores of patients before and after receiving medication.

Patient	Sex	Age (years)	BDI_pre	BDI_post	ΔBDI (%)
P01	Female	49	17	3	−82.4
P02	Male	36	16	22	37.5
P03	Female	33	2	1	−50.0
P04	Female	34	11	4	−63.6
P05	Male	56	5	2	−60.0
P06	Male	34	22	23	4.5
P07	Female	62	16	16	0
P08	Male	49	9	6	−33.3
P09	Female	49	7	5	−28.6
P10	Male	42	14	10	−28.6
P11	Male	53	5	4	−20.0
P12	Female	47	25	29	16.0
P13	Male	31	7	6	−14.3
P14	Female	56	23	18	−21.7
P15	Female	32	13	24	84.6

At study entry, before medication, the mood scores of the patients (males/females = 7/8) were normal (males/females = 4/2), mild mood disturbances (males/females = 2/3), borderline clinical depression (males/females = 0/1), and moderate depression (males/females = 1/2). At follow-up, the recorded mood conditions were normal (males/females = 5/4), mild mood disturbances (males/females = 0/1), borderline clinical depression (males/females = 0/1), and moderate depression (males/females = 2/2). The same number of males and females (5 each) showed improved mood conditions, but the decrease in mood scores were generally higher among females than among males even though this was not statistically significant (p = 0.229). Diagnosis of mood conditions were done according to Table 1.

BDI scores did not differ significantly between pre-medication and post-medication (p = 0.407), males and females before medication (p = 0.411), males and females after medication (p = 0.685), males before and after medication (p = 0.599), and females at baseline and post-medication (p = 0.527). Mood scores were not significantly associated with age before medication (r = 0.427, p = 0.112) and post-medication (r = 0.016, p = 0.954). Percentage change in mood scores was also not significantly associated with age (r = −0.281, p = 0.310). However, females generally presented with poorer mood conditions compared to males (3/6 and 2/4 in the various depression sub-types for males/females at baseline and after medication, respectively).

Proton MRS findings

This study did not establish sex and age dependence of metabolite ratios in the anterior cingulate cortex, left and right hippocampi in both the healthy control and psoriatic arthritis patient groups. Tables 3 –5 show the metabolite ratios for the anterior cingulate cortex (Table 3), left hippocampus (Table 4) and right hippocampus (Table 5).

Table 3:

Comparison of anterior cingulate cortex metabolite ratios between the control and patient groups before and after medication.

Metabolite ratio	Average measurement (SE)			p-Value for comparison
Metabolite ratio	Controls (C)	Baseline (B)	Post-medication (P)	C versus B	C versus P	B versus P
NAA/Cr	1.22 (0.05)	1.20 (0.13)	1.07 (0.03)	0.856	0.023	0.366
Cho/Cr	0.32 (0.02)	0.32 (0.01)	0.31 (0.02)	0.847	0.847	0.600
^aGlx/Cr	1.84 (0.07)	2.09 (0.16)	2.13 (0.17)	0.169	0.137	0.699
Glu/Cr	1.35 (0.06)	1.33 (0.12)	1.46 (0.12)	0.906	0.420	0.545
Ins/Cr	1.01 (0.07)	1.02 (0.07)	1.01 (0.09)	0.941	0.989	0.939

^aThe Glx complex comprises of glutamate, glutamine, and γ-aminobutyric acid.

Table 4:

Comparison of left hippocampus metabolite ratios between the control and patient groups before and after medication.

Metabolite ratio	Average measurement (SE)			p-Value for comparison
Metabolite ratio	Controls (C)	Baseline (B)	Post-medication (P)	C versus B	C versus P	B versus P
NAA/Cr	1.34 (0.05)	1.22 (0.05)	1.26 (0.06)	0.120	0.311	0.557
Cho/Cr	0.35 (0.01)	0.40 (0.01)	0.38 (0.01)	0.014	0.110	0.296

Table 5:

Comparison of right hippocampus metabolite ratios between the control and patient groups before and after medication.

Metabolite ratio	Average measurement (SE)			p-Value for comparison
Metabolite ratio	Controls (C)	Baseline (B)	Post-medication (P)	C versus B	C versus P	B versus P
NAA/Cr	1.37 (0.06)	1.29 (0.07)	1.42 (0.05)	0.405	0.558	0.290
Cho/Cr	0.38 (0.02)	0.38 (0.02)	0.40 (0.02)	0.820	0.590	0.497

From Table 3, only the NAA/Cr ratio comparison between the healthy controls and patients after medication was statistically significant, showing a significantly lower ratio post-medication (p = 0.023). Percentage decrease in Cho/Cr ratio significantly correlated positively with percentage decrease in BDI scores (r = 0.628, p = 0.025). Inspection of the spectral analysis output of the data showed a relatively elevated choline peak and decreased NAA peak in the patient group (Fig. 4). No other finding was significant for the anterior cingulate cortex.

Fig. 4:

Proton MR spectra acquired from the anterior cingulate cortex of a healthy control (a) and psoriatic arthritis patient (b), showing increased choline and decreased NAA peaks in the spectra of the patient.

According to the results for the left hippocampus shown in Table 4, baseline Cho/Cr ratio was significantly higher in the patient group compared to the healthy control group (p = 0.014). No other metabolite ratio was found to be significant in the comparisons; percentage changes in concentrations were also not associated with percentage change in the BDI scores.

As shown in Table 5, no significant difference was found in the comparisons between the healthy controls and patients, both at baseline and after medication. There was also no significant link between the percentage changes in metabolite concentrations and BDI scores.

Metabolite concentrations did not differ significantly between the left hippocampus and right hippocampus in both the healthy control and patient groups, and before and after medication in the patient group (p > 0.5 in all comparisons).

Discussion

Mood assessment findings

The BDI questionnaire used in this study has high sensitivity at picking up symptoms of depression in the general population [28]. From the BDI assessment, females generally presented with worse depressive symptoms than their male counterparts. This finding agrees with the observation that depression is a common phenomenon among women than among men [9, 29, 30], which implies that the female gender is a significant risk factor for depression [29], [30], [31]. Following medication, there was no significant difference in the mood scores between males and females, but the females generally showed more prospects of mood recovery than the males (Table 2). This outcome is supported by the observation elsewhere [30] that antidepressants show higher efficacy in females than in males. Our finding of higher positive response rate in mood among females may suggest that the anti-inflammatory medication may have some associated antidepressant effect among females.

Proton MRS findings

The structural MRI of all study volunteers did not show any visible abnormality, as reported by the consultant neuroradiologist. We did not find age or sex dependent variations in the metabolite ratios, even though there is an indication of age-dependence of NAA concentration in the brain [32]. The non-age dependence of the ratio of NAA/Cr in our study could be due to the narrow age range of our volunteers and the small sample size of the study.

Metabolite concentrations in this study were estimated as ratios of the spectral peak areas to the peak area of creatine. Other studies have used NAA [33] and choline [34] as denominators of metabolite ratios but creatine is the commonest denominator and has been used in the LC Model analysis. Regional and age variations in metabolite concentrations exist [35, 36], but sex-dependent concentrations are unclear. For instance, significantly higher NAA/Cho and lower Cho/Cr ratios in females compared to males in the parieto-occipital white matter have been previously reported [34]. Conversely, Cho/Cr and Ins/Cr ratios were not found to differ in the basal ganglia, frontal, temporal, thalamus and hippocampus in another study [36]. These two studies and our study did not find significant variation of NAA/Cr ratio between males and females in both groups.

Our study observed some tendencies toward, though not significant, increased choline and decreased NAA among the psoriatic arthritis patients (Fig. 4). The NAA decrease could either be due to loss of some brain tissue, or due to the conversion of some NAA to the Glx complex which is linked to the mood changes observed in the patients. The elevated choline may be explained by increased MRS-detectible mobile cell membrane components following the breakdown of the blood-brain barrier as part of the prognosis of psoriatic arthritis. The breakdown process of the blood-brain barrier takes place through active transport which is an energy-dependent process from the cell membrane; this could be linked to some marginal increase in creatine concentration. Indeed, a three-fold increase in Cr/glycine ratio in skin extracts of psoriatic plaques reported by Kim et al. [37] supports our finding of possible marginal increase in creatine levels in our patients, which is linked to significantly increased choline concentration.

Our study found significantly higher Cho/Cr ratio in the left hippocampus of the psoriatic arthritis patients at baseline, and significantly lower NAA/Cr ratio in the anterior cingulate cortex after medication. In a similar study of 35 rheumatoid arthritis patients on various anti-inflammatory medications focusing on the centrum semiovale brain area [9], significantly higher Cho/Cr and lower NAA/Cho ratios were found in active rheumatoid arthritis patients who also had increased erythrocyte sedimentation rates (ESR). The Cho/Cr and NAA/Cho ratios were associated with ESR, but treatment did not affect these correlations. The ambiguity associated with changes in metabolite ratios makes their interpretation quite challenging, especially when information is lacking about the nature of the effect of a disease process on the metabolite levels. However, our findings in comparison with the literature consistently implicate the Cho/Cr ratio in the disease process of peripheral inflammation. The possible explanation is that the prognosis of the disease may be associated with a cell membrane breakdown process which involves the use of energy from the cells involved. Therefore, absolute concentrations of both choline and creatine will be expected to increase, resulting in the elevated Cho/Cr ratio. The decreased NAA/Cr ratio in our study compared to the decreased NAA/Cho ratio reported by Emmer et al. [9] implies that absolute concentration of NAA may be decreased in peripheral inflammation.

Link between mood and brain biochemistry

MRS investigation involving multiple sclerosis, rheumatoid arthritis and psoriatic arthritis have generally focused on alterations in levels of NAA, Cho, Cr, Glu and Ins [7, 16, 38], [39], [40]. Changes in Glu and Ins have been observed to correlate with mood changes [38]. Our study however found a significant association between change in anterior cingulate cortex Cho/Cr ratio and mood change. There is an indication that the left brain hemisphere may be linked to the development of clinical depression [11, 41]. This is in sharp contrast with the lack of association between the significantly higher Cho/Cr ratio in the left hippocampus at baseline and the mood of patients in this study.

A published study [7] of 5 rheumatoid arthritis and 2 psoriatic arthritis patients used three different nuclear magnetic resonance (NMR) techniques focusing on the left centrum semiovale to observe treatment response and mood of the patients. The NMR techniques were magnetization transfer (MT) imaging, diffusion weighted imaging (DWI), and magnetic resonance spectroscopy (MRS). The patients were treated with various anti-TNF-α medications including Etanercept administered after the first scan and repeated after 3 days (which is sufficient time to allow maximal plasma concentration of Etanercept). Assessments of neuropsychological states and MR images did not show improvements after medication. Estimates of Apparent Diffusion Coefficient for white matter (WM) and gray matter (GM), and ratios of NAA/Cho, NAA/Cr and Cho/Cr were not also associated with medication use. However, MT ratio histogram peak heights (MTR-Pht) of WM and GM were significantly reduced with medication. This observed reduction in the MTR-Pht ratio with medication is suggestive of deteriorated structure of parenchyma tissues, which had no link to either inflammation or demyelination as shown by the MRS and DWI results. The decreased MTR-Pht did not correlate with decreased cognitive function. In our study however, decrease in anterior cingulate cortex Cho/Cr ratio correlated positively with decrease in mood score.

Study limitations and future directions

Even though it is quicker to estimate metabolite ratios, their interpretation may be quite ambiguous if little is known about the underlining cause of variations. Absolute metabolite concentration estimates on the other hand will require a cumbersome process of using various correction factors. The use of both schemes on the MRS data could help clarify precisely the specific metabolites implicated in the disease process.

Since mood scores of the control group were not assessed, there was no direct comparison of mood states of the patients with mood states of the healthy controls. Future studies should consider the inclusion of this data to help identify the minimum mood score at which psoriatic arthritis could have been implicated as an underlying cause of the mood changes.

Detailed clinical history of the patients was lacking, namely, duration and severity of the condition, present medication, stress assessment, smoking and alcoholism habits (if any), erythrocyte rate and cytokine levels, skin extracts, cerebrospinal fluid samples, salivary concentrations [5, 6, 42], and demographic details. Future MRS studies could increase the specificity of the findings by including these variables in the study design, some of which could be used as confounds in statistical modeling of the data. Specificity of the MRS findings could as well be increased by assessing the correlations between its results and the findings from other complementary techniques such as functional MRI [42], MT spectroscopy [43] and DWI.

Conclusion

The results of this study showed that the psoriatic arthritis patients had reduced frontal brain NAA/Cr ratio post-medication, but increased left hippocampal baseline Cho/Cr ratio compared to the healthy controls. After receiving anti-inflammatory medication, decreased frontal brain Cho/Cr ratio was associated with improved mood of the patients. The findings therefore provide some evidence for biochemical alterations in the mood regulating areas of the brain, with associated effects on the mood states of the affected patients which can be enhanced by anti-inflammatory medication.

Corresponding author: Abdul Nashirudeen Mumuni, Department of Medical Imaging, School of Allied Health Sciences, University for Development Studies, Tamale, Ghana, e-mail: mnashiru@uds.edu.gh

Article note: A collection of invited papers based on presentations at the Virtual Conference on Chemistry and its Applications 2022 (VCCA-2022) held on-line, 8–12 August 2022.

Funding source: University of Glasgow

Award Identifier / Grant number: 11145801

Funding source: Sackler Institute of Psychological Research

Award Identifier / Grant number: 32660

Funding source: Scottish Imaging Network, A Platform for Scientific Excellence

Award Identifier / Grant number: Scottish Funding Council HR07020

Research funding: This study was jointly funded by the Scottish Imaging Network, A Platform for Scientific Excellence (SINAPSE, ‘Scottish Funding Council HR07020’), University of Glasgow (grant number: 11145801), and Sackler Institute of Psychological Research (grant number: 32660).

References

[1] A. U. Ahmed. Front. Biol. 6(4), 274 (2011). https://doi.org/10.1007/s11515-011-1123-9.Search in Google Scholar

[2] L. Ferrero-Miliani, O. H. Nielsen, P. S. Andersen, S. Girardin. Clin. Exp. Immunol. 147(2), 227 (2007). https://doi.org/10.1111/j.1365-2249.2006.03261.x.Search in Google Scholar PubMed PubMed Central

[3] P. Libby. Nutr. Rev. 65(suppl_3), S140 (2007). https://doi.org/10.1111/j.1753-4887.2007.tb00352.x.Search in Google Scholar PubMed

[4] R. S. Cotran, V. N. Kumar, R. L. Stanley. Robbins Pathologic Basis of Disease, WB Saunders CompHny, Philadelphia, USA (2004).Search in Google Scholar

[5] T. L. Stedman (Ed.). in Stedman’s Medical Dictionary for the Health Professions and Nursing, Lippincott Williams & Wilkins, Philadelphia, United States (2005).Search in Google Scholar

[6] A. H. Miller, V. Maletic, C. L. Raison. Biol. Psychiatr. 65(9), 732 (2009). https://doi.org/10.1016/j.biopsych.2008.11.029.Search in Google Scholar PubMed PubMed Central

[7] A. E. Van der Bijl, B. J. Emmer, F. C. Breedveld, H. A. Middelkoop, C. K. Jurgens, M. A. van Buchem, T. W. Huizinga, J. van der Grond. Clin. Exp. Rheumatol. 25(2), 301 (2007).Search in Google Scholar

[8] J. D. Huber, K. A. Witt, S. Hom, R. D. Egleton, K. S. Mark, T. P. Davis. Am. J. Physiol. Heart Circ. Physiol. 280(3), H1241 (2001). https://doi.org/10.1152/ajpheart.2001.280.3.h1241.Search in Google Scholar PubMed

[9] B. J. Emmer, A. E. van der Bijl, T. W. Huizinga, F. C. Breedveld, S. C. Steens, G. P. Th. Bosma, M. A. van Buchem, J. van der Grond. Arthritis Rheum. 60(11), 3190 (2009). https://doi.org/10.1002/art.24932.Search in Google Scholar PubMed

[10] C. D’Mello, T. Le, M. G. Swain. J. Neurosci. 29(7), 2089 (2009). https://doi.org/10.1523/jneurosci.3567-08.2009.Search in Google Scholar

[11] G. J. Siegle, C. S. Carter, M. E. Thase. Am. J. Psychiatr. 163(4), 735 (2006). https://doi.org/10.1176/ajp.2006.163.4.735.Search in Google Scholar PubMed

[12] P. Videbech, B. Ravnkilde, A. R. Pedersen, A. Egander, B. Landbo, N. A. Rasmussen, F. Andersen, H. Stødkilde-Jørgensen, A. Gjedde, R. Rosenberg. Psychol. Med. 31(7), 1147 (2001). https://doi.org/10.1017/s0033291701004469.Search in Google Scholar PubMed

[13] H. Ito, R. Kawashima, S. Awata, S. Ono, K. Sato, R. Goto, M. Koyama, M. Sato, H. Fukuda. J. Nucl. Med. 37(3), 410 (1996).Search in Google Scholar

[14] T. R. Farchione, G. J. Moore, D. R. Rosenberg. Biol. Psychiatr. 52(2), 86 (2002). https://doi.org/10.1016/s0006-3223(02)01340-9.Search in Google Scholar PubMed

[15] S. Gruber, R. Frey, V. Mlynárik, A. Stadlbauer, A. Heiden, S. Kasper, G. J. Kemp, E. Moser. Invest. Radiol. 38(7), 403 (2003). https://doi.org/10.1097/01.rli.0000073446.43445.20.Search in Google Scholar PubMed

[16] N. J. Coupland, C. J. Ogilvie, K. M. Hegadoren, P. Seres, C. C. Hanstock, P. S. Allen. Biol. Psychiatr. 57(12), 1526 (2005). https://doi.org/10.1016/j.biopsych.2005.02.027.Search in Google Scholar PubMed

[17] D. R. Rosenberg, F. P. MacMaster, Y. Mirza, J. M. Smith, P. C. Easter, S. P. Banerjee, R. Bhandari, C. Boyd, M. Lynch, M. Rose, J. Ivey. Biol. Psychiatr. 58(9), 700 (2005). https://doi.org/10.1016/j.biopsych.2005.05.007.Search in Google Scholar PubMed

[18] V. Gabbay, D. A. Hess, S. Liu, J. S. Babb, R. G. Klein, O. Gonen. Am. J. Psychiatr. 164(12), 1881 (2007). https://doi.org/10.1176/appi.ajp.2007.06122032.Search in Google Scholar PubMed PubMed Central

[19] V. Govindaraju, K. Young, A. A. Maudsley. NMR Biomed. 13(3), 129 (2000). https://doi.org/10.1002/1099-1492(200005)13:3<129::aid-nbm619>3.0.co;2-v.10.1002/1099-1492(200005)13:3<129::AID-NBM619>3.0.CO;2-VSearch in Google Scholar

[20] R. Srinivasan, N. Sailasuta, R. Hurd, S. Nelson, D. Pelletier. Brain 128(5), 1016 (2005). https://doi.org/10.1093/brain/awh467.Search in Google Scholar

[21] K. T. Fernando, M. A. McLean, D. T. Chard, D. G. MacManus, C. M. Dalton, K. A. Miszkiel, R. M. Gordon, G. T. Plant, A. J. Thompson, D. H. Miller. Brain 127(6), 1361 (2004). https://doi.org/10.1093/brain/awh153.Search in Google Scholar

[22] L. Li, H. Zhang, M. Varrin-Doyer, S. S. Zamvil, A. Verkman. Faseb. J. 25(5), 1556 (2011). https://doi.org/10.1096/fj.10-177279.Search in Google Scholar

[23] M. del Pilar Sánchez-López, V. Dresch. Psicothema 20(4), 839 (2008).Search in Google Scholar

[24] D. P. Auer, B. Pütz, E. Kraft, B. Lipinski, J. Schill, F. Holsboer. Biol. Psychiatr. 47(4), 305 (2000). https://doi.org/10.1016/s0006-3223(99)00159-6.Search in Google Scholar

[25] S. W. Provencher. Magn. Reson. Med. 30(6), 672 (1993). https://doi.org/10.1002/mrm.1910300604.Search in Google Scholar

[26] R. F. Schulte, P. Boesiger. NMR Biomed. 19(2), 255 (2006). https://doi.org/10.1002/nbm.1026.Search in Google Scholar PubMed

[27] M. Walter, A. Henning, S. Grimm, R. F. Schulte, J. Beck, U. Dydak, B. Schnepf, H. Boeker, P. Boesiger, G. Northoff. Arch. Gen. Psychiatr. 66(5), 478 (2009). https://doi.org/10.1001/archgenpsychiatry.2009.39.Search in Google Scholar PubMed

[28] P. J. Moran, D. C. Mohr. J. Behav. Med. 28(1), 35 (2005). https://doi.org/10.1007/s10865-005-2561-0.Search in Google Scholar PubMed

[29] B. Saraceno. Epidemiol. Psychiatr. Sci. 11(2), 83 (2002). https://doi.org/10.1017/s1121189x00005546.Search in Google Scholar PubMed

[30] M. Paez-Pereda, M. Panhuysen. Strategies to identify biomarkers for depression. In Biomarkers for Psychiatric Disorders, pp. 299–314, Springer, Boston, MA (2008).10.1007/978-0-387-79251-4_13Search in Google Scholar

[31] M. Korostil, A. Feinstein. Mult. Scler. J. 13(1), 67 (2007). https://doi.org/10.1177/1352458506071161.Search in Google Scholar PubMed

[32] J. C. Brooks, N. Roberts, G. J. Kemp, M. A. Gosney, M. Lye, G. H. Whitehouse. Cerebr. Cortex 11(7), 598 (2001). https://doi.org/10.1093/cercor/11.7.598.Search in Google Scholar PubMed

[33] P. Sarchielli, O. Presciutti, R. Tarducci, G. Gobbi, A. Alberti, G. P. Pelliccioli, A. Orlacchio, V. Gallai. J. Neurol. Neurosurg. Psychiatr. 64(2), 204 (1998). https://doi.org/10.1136/jnnp.64.2.204.Search in Google Scholar PubMed PubMed Central

[34] I. D. Wilkinson, M. N. Paley, K. A. Miszkiel, M. A. Hall-Craggs, B. E. Kendall, R. J. Chinn, M. J. Harrison. Magn. Reson. Imag. 15(2), 243 (1997). https://doi.org/10.1016/s0730-725x(96)00334-7.Search in Google Scholar PubMed

[35] L. G. Kaiser, N. Schuff, N. Cashdollar, M. W. Weiner. Neurobiol. Aging 26(5), 665 (2005). https://doi.org/10.1016/j.neurobiolaging.2004.07.001.Search in Google Scholar PubMed PubMed Central

[36] R. A. Komoroski, C. Heimberg, D. Cardwell, C. N. Karson. Magn. Reson. Imag. 17(3), 427 (1999). https://doi.org/10.1016/s0730-725x(98)00186-6.Search in Google Scholar PubMed

[37] Y. H. Kim, E. K. Orenberg, K. F. Faull, N. G. Wade-Jardetzky, O. Jardetzky. J. Invest. Dermatol. 92(2), 210 (1989). https://doi.org/10.1111/1523-1747.ep12276738.Search in Google Scholar PubMed

[38] M. J. Taylor, S. Selvaraj, R. Norbury, P. Jezzard, P. J. Cowen. J. Affect. Disord. 119(1–3), 186 (2009). https://doi.org/10.1016/j.jad.2009.02.022.Search in Google Scholar PubMed PubMed Central

[39] H. C. Charles, F. Lazeyras, K. R. Krishnan, O. B. Boyko, M. Payne, D. Moore. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 18(7), 1121 (1994). https://doi.org/10.1016/0278-5846(94)90115-5.Search in Google Scholar PubMed

[40] G. Hasler, J. W. van der Veen, T. Tumonis, N. Meyers, J. Shen, W. C. Drevets. Arch. Gen. Psychiatr. 64(2), 193 (2007), https://doi.org/10.1001/archpsyc.64.2.193.Search in Google Scholar PubMed

[41] R. M. Dixon, K. M. Bradley, M. M. Budge, P. Styles, A. D. Smith. Brain 125(10), 2332 (2002). https://doi.org/10.1093/brain/awf226.Search in Google Scholar PubMed

[42] D. W. O’Connor, D. Ames, B. Gardner, M. King. Int. Psychogeriatr. 21(2), 225 (2009). https://doi.org/10.1017/s1041610208007588.Search in Google Scholar PubMed

[43] I. Catalaa, R. I. Grossman, D. L. Kolson, J. K. Udupa, L. G. Nyul, L. Wei, X. Zhang, M. Polansky, L. J. Mannon, J. C. McGowan. Radiology 216(2), 351 (2000). https://doi.org/10.1148/radiology.216.2.r00au16351.Search in Google Scholar PubMed

Published Online: 2023-03-22

Published in Print: 2023-03-28

© 2023 IUPAC & De Gruyter. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, please visit: http://creativecommons.org/licenses/by-nc-nd/4.0/

Articles in the same Issue

https://doi.org/10.1515/pac-2022-1119

Keywords for this article

Brain; choline; creatine; hippocampus; magnetic resonance spectroscopy; major depression; psoriatic arthritis; tumor necrosis factor; VCCA-2022