Cerebrospinal Fluid Sulfatide Levels Lack Diagnostic Utility in the Subcortical Small Vessel Type of Dementia

Abstract

Background:

Sulfatides (STs) in cerebrospinal fluid (CSF), as well as magnetic resonance imaging (MRI)-detected white matter hyperintensities (WMHs), may reflect demyelination. Here, we investigated the diagnostic utility of CSF ST levels in the subcortical small vessel type of dementia (SSVD), which is characterized by the presence of brain WMHs.

Objective:

To study the diagnostic utility of CSF ST levels in SSVD.

Methods:

This was a mono-center, cross-sectional study of SSVD (n = 16), Alzheimer’s disease (n = 40), mixed dementia (n = 27), and healthy controls (n = 33). Totally, 20 ST species were measured in CSF by liquid chromatography-mass spectrometry (LC-MS/MS).

Results:

CSF total ST levels, as well as CSF levels of hydroxylated and nonhydroxylated ST species, did not differ across the study groups. In contrast, CSF neurofilament light chain (NFL) levels separated the patient groups from the controls. CSF total ST level correlated with CSF/serum albumin ratio in the total study population (r = 0.64, p < 0.001) and in all individual study groups. Furthermore, CSF total ST level correlated positively with MRI-estimated WMH volume in the total study population (r = 0.30, p < 0.05), but it did not correlate with CSF NFL level.

Conclusion:

Although there was some relation between CSF total ST level and WMH volume, CSF ST levels were unaltered in all dementia groups compared to the controls. This suggests that CSF total ST level is a poor biomarker of demyelination in SSVD. Further studies are needed to investigate the mechanisms underlying the marked correlation between CSF total ST level and CSF/serum albumin ratio.

Keywords

Alzheimer’s disease cerebrospinal fluid mixed dementia sulfatide species sulfatides subcortical small vessel type of dementia white matter hyperintensities

INTRODUCTION

Sulfatides (STs; 3^′-O-sulfogalactosylceramide) are glycosphingolipids consisting of a hydrophobic moiety called ceramide (sphingoid base and a fatty acid) and a sulfated galactose. STs are together with galactosylceramides the most typical lipids in myelin [1], where the predominant ST species are characterized by long-chain fatty acids (C22–C26 carbon atoms) [2]. STs are required for the stability and maintenance of the myelin sheet [3 –5], but STs are also expressed in the brain gray matter, where they are involved in various functions and cellular responses [6].

STs in cerebrospinal fluid (CSF) have been proposed as markers of demyelination, myelin damage, and myelin turnover [7 –11]. Therefore, CSF measurements of STs could possibly be useful in the subcortical small vessel type of vascular dementia (SSVD). SSVD is often under-diagnosed [12, 13], but there are estimations that SSVD could comprise about half of the patients with vascular dementia (VaD) [14]. The manifestations of SSVD (speed/attention deficits, executive dysfunction, memory loss, motor and mood disturbances, and eventually functional disability) correlate with subcortical vascular changes like white matter hyperintensities (WMHs) on magnetic resonance imaging (MRI) [13, 15]. The underlying pathologies include arteriolosclerosis, lipohyalinosis, fibroid necrosis, edema and damage to the blood-brain barrier (BBB), resulting in chronic leakage of fluid and macromolecules in the white matter, and subsequent demyelination [12 , 15]. Furthermore, vascular disease of the SSVD type might accelerate the progression of AD [12 , 16]; the state of combined AD and vascular pathologies could be denominated as mixed dementia [12, 13].

In non-demented elderly, the baseline CSF ST level was positively associated with the baseline white matter lesion volume [17], but it was inversely associated with the progression of white matter lesions during a 3-year follow-up [9]. Some previous studies have measured STs in CSF from VaD populations partly consisting of SSVD patients [8 , 19]. One study demonstrated that STs were progressively increased in the postmortem temporal cortex gray matter from control to subcortical ischemic VaD to mixed dementia [20]. However, to our knowledge, only one previous study have compared CSF ST levels in VaD patients with those in controls [7]. In this study, VaD was defined as dementia and history of macrovascular cerebral events and/or severe vascular diseases, and the results showed increased CSF ST levels in the VaD group compared to that in the Alzheimer’s disease (AD) and control groups [7]. Brain levels of STs have been reduced in AD in rodent models and in postmortem human brain tissue samples [21 –30]. CSF ST levels were decreased in incipient human AD [31], whereas in manifest AD, CSF ST levels have been similar to those in controls [7, 32]. However, it has not been evaluated whether CSF ST levels are altered in a population consisting only of SSVD patients, or in mixed forms of AD and SSVD, compared to controls.

Recently, a highly sensitive method was developed to measure STs in CSF based on ultra-performance liquid chromatography coupled to tandem mass spectrometry (UPLC-MS/MS) [32]. This method also provides quantitative measurements of selected sulfatide species in CSF. In the present study, we used this method to measure CSF ST levels in a single-center memory clinic population of SSVD patients and compared the results with those in controls. Furthermore, we also included study groups of AD patients and patients with mixed dementia (combined AD and SSVD). We hypothesized that CSF ST levels would be increased in SSVD as a consequence of white matter axonal degeneration and demyelination.

MATERIALS AND METHODS

Study participants

In this cross-sectional study, we evaluated CSF ST levels in SSVD, AD, mixed dementia, and cognitively healthy controls. The participants were recruited from the Gothenburg MCI study, a mono-center study performed at the memory clinic at Sahlgrenska University Hospital [33, 34]. The Gothenburg MCI study was designed to exclude somatic and psychiatric conditions associated with increased risk of cognitive impairment. The inclusion criteria comprised age≥50 and≤79 years, Mini Mental State Examination (MMSE) score > 19, and self- or informant-reported cognitive decline with a duration≥6 months. The exclusion criteria included severe somatic disease (e.g., subdural hemorrhage, brain tumor, untreated hypothyroid state, encephalitis, and unstable heart disease), psychiatric disorder (e.g., major affective disorder or schizophrenia), substance abuse, and confusion. The healthy controls were primarily recruited through senior citizen organizations, i.e., information meetings on dementia, and some were relatives of the patients. Present, or history of, cognitive decline was an exclusion criterion in the controls, otherwise the study procedures were similar as those applied for the patients.

The patients were classified using the global deterioration scale (GDS), in which GDS 1 equals no subjective or objective cognitive decline and GDS 4 equals possible mild dementia [35]. The classification into GDS groups was based on medical history, checklists and instruments for cognitive symptoms [33]. The neuropsychological tests used were: 1) Stepwise Comparative Status Analysis (STEP) [36] (variables 13–20); 2) I-FLEX, a short form of the Executive Interview (EXIT) [37]; 3) MMSE [38]; and 4) Clinical Dementia Rating (CDR) [39]. The CDR assessment was based on information from both the subject and an informant. Guidelines for GDS 4 were: STEP > 1, I-FLEX > 3, CDR > 1.0, MMSE≤25. However, a consensus decision among the physicians at the clinic was made to determine the appropriate GDS score.

In patients with GDS 4, the clinician who determined the specific dementia diagnoses had access to clinical symptomatology and MRI data, but was blinded to neuropsychological test results and CSF biomarker data. AD was diagnosed using the NINCDS-ADRDA criteria [40]. SSVD was defined according to the Erkinjuntti criteria [41], which included predominant frontal lobe symptoms and MRI-detected cerebral white matter hyperintensities (WMHs; moderate or severe according to Fazekas classification [42]). Mixed dementia was diagnosed if AD patients also exhibited MRI findings of cerebral WMHs (moderate or severe according to Fazekas classification) [42] with no predominant frontal lobe syndrome, or alternatively, if AD patients exhibited mild degree of WMHs in combination with a marked frontal lobe syndrome.

Totally, 33 controls and 83 patients (SSVD, n = 16; AD, n = 40; and mixed dementia, n = 27) were included. Other forms of dementia (cortical VaD, primary progressive aphasia, Lewy body dementia, frontotemporal dementia, or unspecified dementia) were excluded.

The study was approved by the regional ethical committee in Gothenburg. Oral and written consent was obtained from all study participants. The study was performed in compliance with the Declaration of Helsinki.

MRI procedures and white matter hyperintensities

A 1.5 T MRI scanner (Siemens Symphony, Erlangen, Germany) was used as described previously [43]. WMH volumes were estimated using the FreeSurfer automated segmentation software (version 5.3.0; https://surfer.nmr.mgh.harvard.edu/). To reduce head size variability, the individual values were corrected for intracranial volume (ICV) [44]. First, a regression analysis was performed in the control group between the raw WMH volume and the raw ICV volume to obtain the regression coefficient β. Then, the regression coefficient β was applied to the entire study sample, and WMH volumes (in cm³) were calculated according to the formula: adjusted WMH volume = raw WMH volume –β(raw ICV volume –mean ICV volume).

Neuropsychological tests

In addition to the tests used for GDS classification, we used the delayed recall from the Rey Auditory Verbal Learning Test (RAVLT) [45] to assess episodic memory and the Trail Making Test A (TMT-A) and B (TMT-B) [46] to evaluate visual scanning and complex attention.

Cerebrospinal fluid and blood samples

CSF samples were drawn at the lumbar vertebrae L3/L4 or L4/L5 interspace; the first portion of CSF was discarded to avoid blood contamination. In all, 20 ml of CSF was collected in polypropylene tubes, gently mixed by inverting the tube, and centrifuged at room temperature at 2,000 x g for 10 min. Blood samples were also drawn in fasted state between 08.00 and 10.00. The CSF and blood samples were then stored at –80°C pending analyses [33].

Biochemical methods

All biochemical analyses were performed with the analyst being blinded to clinical information. ST species in CSF were quantified as previously described in detail [32]. Briefly, STs were automatically extracted from 100μL CSF using the robot-assisted butanol:methanol (BUME) method [47], and further quantified by UPLC-MS/MS [32]. ST C19:0 was used as an internal standard and quantification was made using a native ST mixture with known total ST content and composition of ST species. The native ST mixture was also used to monitor inter-day variations of chromatographic behavior (such as retention times and signal intensities). Twenty ST species, including hydroxylated species, were quantified: ST C16:0, ST C18:0, ST C18:1, ST C20:0, ST C22:0, ST C24:0, ST C24:1, ST C25:1, ST C26:0, ST C26:1, ST C18:0-OH, ST C20:0-OH, ST C22:0-OH, ST C23:0-OH, ST C24:0-OH, ST C24:1-OH, ST C25:0-OH, ST C25:1-OH, ST C26:0-OH, and ST C26:1-OH. The lower limit of quantification (LLOQ) was 0.1 nmol/L. Intra- and inter-assay precisions were 10%for all ST species except for ST C18:0, ST C18:0-OH, and ST C26:0, which had interday precisions of 12%, 17%, and 27%, respectively.

Neurofilament light chain (NFL) concentration in CSF was measured using a sensitive sandwich enzyme-linked immunosorbent assay (ELISA) (NF-light^® ELISA kit, UmanDiagnostics AB, Umeå, Sweden). Intra- and inter-assay coefficients of variation were < 10%. The LLOQ was 31 ng/L. CSF levels of total (T)-tau, phosphorylated (P)-tau 181, and Aβ amino acids 1 to 42 (Aβ_1–42) were determined using sandwich ELISAs (INNOTEST, Fujirebio, Gent, Belgium). Two or more internal control CSF samples (aliquots of pooled CSF) were analyzed each run as internal quality controls to minimize between-assay variability [33]. Serum and CSF albumin were measured using immunonephelometry on a Beckman Immage immunochemistry system (Beckman Instruments, Beckman Coulter, Brea, CA). The ratio between CSF albumin (mg/L) and serum albumin (g/L) was used as a measure of the BBB function [48]. APOE genotyping was performed by minisequencing [33].

Statistical analysis

We used SPSS version 25.0 for Windows (IBM Corp., Armonk, NY). Unless otherwise stated, the descriptive statistical results are presented as the median and interquartile range (25th–75th percentiles). Differences across all four study groups were assessed using the non-parametric Kruskal-Wallis test for multiple variables. For variables that differed significantly across all the study groups, the Mann-Whitney U test was used for pair-wise comparisons. Chi-square tests were used for categorical variables. As CSF NFL level is dependent on age, we in addition performed ANCOVA analyses using logarithmically transformed CSF NFL data and having age as a covariate; all differences across groups in CSF NFL level remained in these ANCOVA analyses. Correlations were sought using the Spearman rank order correlation test. The significance level was set to p < 0.05.

RESULTS

Baseline characteristics

Clinical characteristics are presented in Table 1. Male sex tended (non-significant) to be more frequent in the SSVD group than in the other study groups. Patients with SSVD and mixed dementia had higher age than the controls, and in addition, AD patients had lower age than patients with mixed dementia. Education level did not significantly differ across groups. The scores of all neuropsychological tests (MMSE, TMT-A, TMT-B, and RAVLT delayed recall) were impaired in all patient groups compared with the controls. CSF levels of AD biomarkers (Aβ_1–42, T-tau, and P-tau) were altered in the AD and mixed dementia groups compared with the SSVD and the control groups. CSF NFL level was higher in all patient groups compared with the controls. CSF/serum albumin ratio did not differ across groups although there was a tendency to a higher CSF/serum albumin ratio in SSVD patients. WMH volume as estimated using MRI was increased in the SSVD and mixed dementia groups compared with the controls, and additionally, SSVD patients had higher WMH volume than AD patients. Presence of the APOE ɛ4 allele non-significantly tended to be more common in AD and mixed dementia.

Table 1

Baseline characteristics in the study population of 83 patients with dementia and 33 healthy controls

	SSVD (n = 16)	AD (n = 40)	Mixed dementia (n = 27)	Controls (n = 33)	p
Men/women (n, %)	11/5 (69/31)	12/28 (30/70)	9/18 (33/67)	14/19 (42/58)	0.051
Age (y)	69 (65–75)^a	68 (62–72)^b	70 (67–75)^c	64 (57–68)	<0.001
Education (y)	10.5 (8.3–12.8)	10.5 (8.0–14.8)	10.0 (7.0–13.0)	12.0 (10.5–13.5)	0.34
MMSE score	25 (24–27)^c	25 (22–27)^c	25 (23–27)^c	30 (29–30)	<0.001
TMT-A (s)	81 (47–105)^c	61 (50–102)^c	55 (39–91)^c	32 (27–39)	<0.001
TMT-B (s)	217 (151–291)^c	173 (120–275)^c	162 (127–280)^c	80 (69–106)	<0.001
RAVLT delayed recall	2.0 (1.0–6.5)^a	2.5 (1.0–4.8)^c	2.0 (0–5.0)^c	8.0 (6.0–11.0)	<0.001
Aβ_1–42 (ng/L)	640 (473–778)^d,e	390 (320–451)^c	363 (240–450)^c	651 (500–940)	<0.001
T-tau (ng/L)	331 (221–424)^d,f	606 (441–844)^c	670 (480–940)^c	292 (170–394)	<0.001
P-tau (ng/L)	58 (36–61)^d,f	86 (65–99)^c	89 (70–130)^c	49 (32–63)	<0.001
NFL (ng/L)	1125 (908–2230)^c	1293 (901–1545)^c	1481 (1077–1729)^c	741 (541–1023)	<0.001^#
CSF/serum albumin ratio	7.0 (5.0–8.9)	5.2 (4.6–8.4)	6.5 (5.4–8.7)	6.4 (5.4–7.4)	0.33
WMH volume (cm³)	13.2 (6.4–27.9)^c,g	4.2 (2.3–6.3)^b	7.0 (4.9–10.6)^c	2.6 (1.8–3.9)	<0.001
APOE ɛ4	8 / 6 / 2	10 / 24 / 6	7 / 13 / 7	16 / 10 / 2	0.06
allele (0/1/2, %)	(50%/38%/12%)	(25%/60%/15%)	(26%/48%/26%)	(57%/36%/7%)

APOE genotyping was not performed in 5 control subjects. Values are given as the median (25th-75th percentile). The p-values in the right column refers to differences across all four groups using the Kruskal-Wallis test. Pair-wise comparisons were performed using the Mann-Whitney U test. ^#All differences across groups remained in ANCOVA analysis using logarithmically transformed CSF NFL data and having age as a covariate. ^ap < 0.01 versus controls; ^bp < 0.05 versus mixed dementia; ^cp < 0.001 versus controls; ^dp < 0.001 versus AD; ^ap < 0.01 versus mixed dementia; ^fp < 0.001 versus mixed dementia; ^gp < 0.01 versus AD.

Unchanged CSF concentrations of sulfatides

The CSF total ST concentration (Fig. 1A and Table 2), CSF concentrations of hydroxylated ST species (HFA) and nonhydroxylated ST species (NFA), and ST ratios (total HFA/total ST ratio, total NFA/total ST ratio, and total HFA/total NFA ratio) (Table 2), did not differ across the study groups. Furthermore, there were no differences across groups in any of the individual ST species, neither in the absolute CSF concentrations nor in the percentages of total CSF ST levels (the abundant ST species C18, C20, C22, and C24 are given as percentages of total ST concentrations in Table 3; other ST species are not shown).

Fig. 1

CSF total sulfatide (ST) concentration lack diagnostic utility but is markedly correlated with CSF/serum albumin ratio. A) CSF total ST concentration in patients with the subcortical small vessel type of dementia (SSVD, n = 16), Alzheimer’s disease (AD, n = 40), mixed dementia (n = 27), and cognitively healthy controls (n = 33). B) The positive correlation between CSF total ST concentration and CSF/serum albumin ratio (total study population: r = 0.64, p < 0.001; SSVD: r = 0.71, p < 0.001; AD: r = 0.49, p = 0.001; mixed dementia: r = 0.66, p < 0.001; and control: r = 0.65, p < 0.001). In (A), the horizontal lines represent the median values.

Table 2

CSF total sulfatide (ST) concentrations, CSF concentrations of hydroxylated ST species (HFA) and nonhydroxylated ST species (NFA), and CSF ST ratios in 83 patients with dementia and 33 healthy controls

	SSVD (n = 16)	AD (n = 40)	Mixed dementia (n = 27)	Controls (n = 33)	p
Total ST (nmol/L)	97 (85–113)	84 (77–110)	99 (82–129)	94 (77–108)	0.30
Total HFA (nmol/L)	21 (18–24)	19 (16–24)	22 (17–27)	21 (16–25)	0.47
Total NFA (nmol/L)	74 (65–89)	67 (61–87)	78 (65–100)	72 (62–88)	0.29
Ratios
Total HFA/total ST (%)	21 (20–24)	21 (19–24)	21 (19–24)	22 (19–25)	0.69
Total NFA/total ST (%)	79 (76–81)	79 (76–81)	79 (77–81)	79 (76–81)	0.80
Total HFA/total NFA (%)	26 (24–31)	27 (24–31)	26 (23–30)	29 (24–33)	0.77

Values are given as the median (25th–75th percentile). The p-values in the right column refers to differences across all four groups using the Kruskal-Wallis test for multiple comparisons.

Table 3

CSF concentrations of the abundant sulfatide (ST) species C18, C20, C22, and C24 (presented as percentages of total ST levels) in 83 patients with dementia and 33 healthy controls

	SSVD (n = 16)	AD (n = 40)	Mixed dementia (n = 27)	Controls (n = 33)	p
ST C18:0-OH (%of total ST)	2.5 (1.8–3.6)	2.7 (2.3–3.2)	2.8 (2.4–3.1)	2.8 (2.3–3.3)	0.88
ST C20:0-OH (%of total ST)	2.8 (2.4–3.0)	2.7 (2.5–2.9)	2.5 (2.4–2.9)	2.5 (2.4–3.0)	0.53
ST C22:0-OH (%of total ST)	4.0 (3.8–4.3)	3.9 (3.5–4.1)	3.6 (3.4–4.1)	3.8 (3.6–4.2)	0.34
ST C24:0-OH (%of total ST)	3.1 (2.9–3.3)	3.1 (2.8–3.4)	3.0 (2.8–3.3)	3.2 (2.9–3.4)	0.65
ST C24:1-OH (%of total ST)	5.4 (5.0–7.1)	5.5 (4.8–6.8)	5.8 (4.7–6.6)	5.7 (4.8–6.9)	0.97
ST C18:0 (%of total ST)	10.5 (9.7–12.1)	11.1 (10.2–12.0)	10.7 (9.8–11.5)	11.2 (10.6–12.6)	0.46
ST C18:1 (%of total ST)	9.8 (7.7–10.4)	9.6 (8.7–11.4)	10.0 (8.0–11.1)	9.8 (8.6–11.3)	0.68
ST C20:0 (%of total ST)	3.3 (3.1–3.6)	3.3 (3.1–3.4)	3.3 (3.1–3.5)	3.2 (3.1–3.4)	0.38
ST C22:0 (%of total ST)	4.4 (3.9–5.0)	4.3 (3.9–4.7)	4.4 (3.9–4.9)	4.1 (3.7–4.5)	0.15
ST C24:0 (%of total ST)	9.5 (8.9–10.8)	10.1 (8.9–11.1)	10.1 (9.2–10.8)	9.8 (8.6–11.1)	0.77
ST C24:1 (%of total ST)	34.3 (30.6–35.3)	32.5 (31.5–34.2)	33.2 (32.1–34.5)	32.3 (31.2–34.3)	0.52

Values are given as the median (25th–75th percentile). There were no differences across groups in ST C18, C20, C22, or C24 species. The p-values in the right column refers to differences across all four groups using the Kruskal-Wallis test for multiple comparisons.

Correlations

In the total study population (n = 116), we investigated whether CSF total ST levels were correlated with age, WMH volume as estimated using MRI, CSF NFL levels, AD biomarkers, CSF/serum albumin ratio, and neuropsychological test results. For correlations that were significant in the total study cohort, we assessed whether they were present also in the individual study groups.

CSF total ST level did not correlate with age in the total cohort (r = –0.04). Furthermore, in the total study population, CSF total ST level correlated positively with WMH volume (r = 0.30, p < 0.05). There was a non-significant tendency that CSF total ST level correlated with WMH volume in the controls (r = 0.44, p = 0.08), but there was no correlation in the patient groups.

CSF total ST level was markedly correlated with CSF/serum albumin ratio in the total study population (r = 0.64, p < 0.001; Fig. 1B) and in all the individual study groups (SSVD: r = 0.71, p < 0.001; AD: r = 0.49, p = 0.001; mixed dementia: r = 0.66, p < 0.001; and control: r = 0.65, p < 0.001). Finally, the CSF total ST level did not correlate with CSF NFL level, CSF AD biomarkers (Aβ_1–42, T-tau, and P-tau), or the neuropsychological test scores (MMSE, TMT-A, TMT-B, and RAVLT delayed recall) in the total study cohort.

DISCUSSION

This is the first study that have evaluated CSF ST levels in a VaD population consisting only of SSVD patients and compared the results with those in cognitively healthy controls. We also included patients with AD and mixed dementia (combined AD and SSVD). All patients and controls were recruited at a single memory clinic. We found that manifest SSVD was not associated with CSF total ST levels, ST ratios, or ST species. CSF ST levels were also unaltered in patients with AD and mixed dementia compared with the controls. In contrast, CSF NFL level was increased in all patient groups. In the total study population, there was a marked positive correlation between CSF total ST level and CSF/serum albumin ratio and a relatively weak positive correlation with WMH volume. CSF total ST levels did not correlate with age, CSF NFL levels, AD biomarkers, or neuropsychological test scores.

Our hypothesis was that CSF ST levels would be increased in SSVD as this disease is characterized by brain small vessel dysfunction, reduced BBB function, myelin loss reflected as WMHs using MRI, and a specific cognitive profile [12 , 34]. However, this hypothesis has to be rejected as the SSVD group did not differ from the other study groups in terms of CSF total ST levels, ST ratios, or ST species. Previously, in manifest VaD, several studies have measured STs in CSF [8 , 19], but only one study have compared CSF ST levels in VaD patients with those in controls [7]. This study found increased CSF ST levels in the VaD group, defined as history of macrovascular cerebral events and/or severe vascular diseases, as compared with the AD and control groups [7]. The reasons for the discrepant results between our and the earlier study [7] are unknown but could be related to differences in VaD classification, disease activity, or assay performance (LC-MS/MS in our study vs. thin layer chromatography overlay technique in the previous study [7]).

ST depletion has been observed in the AD brain in rodent models and in postmortem human brain tissue samples [21 –30]. CSF ST levels were decreased in very early human AD [31], whereas in manifest AD, CSF ST levels have been similar to those in controls [7, 32]. In the present study, we found unchanged CSF total ST level, ST ratios, and ST species in the AD group, confirming unaltered CSF ST levels in manifest AD. Previously, using the same LC-MS/MS methodology, we studied de-identified leftover CSF samples from the clinical routine at the Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital [32]. These CSF samples were defined as having an AD biomarker profile if levels of Aβ_1–42, T-tau, and P-tau were abnormal, or as being control samples if all three biomarkers were normal [32]. Although total ST level was similar in both groups of CSF, the amount of hydroxylated ST species in relation to nonhydroxylated ST species was increased in the CSF samples with an AD biomarker profile [32]. In the present study of a mono-center memory clinic population, we could not replicate this finding as the ratio in CSF between hydroxylated and nonhydroxylated ST species was similar in patients with a clinical diagnosis of AD as that in the other study groups (SSVD, mixed dementia, and cognitively healthy controls). Moreover, in some accordance with the results of the present study, another study found a higher proportion of hydroxylated STs in the brain gray matter than in the white matter, but ST distribution did not differ between AD and control brains [49].

Little is known about CSF ST levels in mixed dementia. One study assessing the postmortem temporal cortex gray matter demonstrated increased ST content in mixed dementia compared with controls [20]. However, in our study, CSF ST levels were unchanged in the mixed dementia group. Furthermore, in the Leukoaraiosis and Disability in the Elderly (LADIS) study, the baseline CSF ST level was positively associated with the baseline white matter lesion volume [17], but it was inversely associated with the progression of white matter lesions during a 3-year follow-up in non-demented elderly [9]. In studies of manifest VaD, the association between CSF ST level and WMH volume has not been investigated or been inconsistent [8 , 19]. In our study, CSF total ST level correlated positively with WMH volume in the total study cohort and tended to correlate with WMH volume in the controls, but there was no correlation in the patient groups. Thus, the association between CSF total ST level and brain WMH volume appears to be relatively weak in the studied dementia disorders. Moreover, we did not find any correlation between CSF total ST level and a CSF biomarker of subcortical axonal neurodegeneration, NFL [50], further illustrating that CSF total ST level is a poor biomarker of demyelination. However, CSF ST levels could be useful in other conditions, i.e., to identify altered ceramide composition in fatty acid 2-hydroxylase disorders [51] or pathological brain ST accumulation in lysosomal storage diseases such as metachromatic leukodystrophy [52].

The CSF/serum albumin ratio is to date the most established measure of BBB function [48]. We observed a strong positive correlation between CSF total ST level and CSF/serum albumin ratio, which is in line with the results of one previous study [32], but in contrast to another study [7]. Although most of cerebral STs are produced locally in the brain, a minor portion of CSF STs could be derived from blood [7, 32]. It is therefore possible that reduced BBB function could result in increased passage of STs into the central nervous system, resulting in at least somewhat increased CSF ST level. Another possibility is that reduced BBB function could result in chronic leakage of fluid and macromolecules into the brain white matter with subsequent white matter damage and demyelination [12 , 15]. However, in our study, CSF total ST level was only weakly associated with WMH volume and CSF NFL level, arguing against the latter hypothesis. Therefore, further studies are needed to explore the mechanisms underlying the marked association between CSF total ST level and CSF/serum albumin ratio found in both patients and controls.

Strengths of the present study, in addition to the extensive characterization of the patients, include the mono-center design. We excluded all patients with stroke-related VaD (cVaD), thereby having one study group consisting only of SSVD patients. In addition, we had an AD group, a mixed dementia group, and a control group. We used a validated LC-MS/MS method to measure CSF ST levels [32, 53], with improved sensitivity and reproducibility as compared to the previously used thin-layer chromatography technology. However, the number of study participants was limited, which could have reduced the statistical power, especially in terms of subanalyses. Another limitation is the cross-sectional design, and changes over time could therefore not be followed.

In conclusion, in a well-defined study population from a single memory clinic, there were no differences across groups in CSF total ST level, ST ratios, or ST species. Therefore, CSF ST levels lacked the ability to separate the patient groups (SSVD, AD, and mixed dementia) from healthy controls. CSF total ST level was weakly associated with brain WMH volume, but there was no relation to CSF NFL level, suggesting that CSF total ST level is a poor biomarker of demyelination in SSVD. We found a strong correlation between CSF total ST level and CSF/serum albumin ratio, and further studies are required to investigate the mechanisms underlying this association. Future studies also need to investigate whether CSF ST levels could be altered in the preclinical stages of SSVD.

Footnotes

ACKNOWLEDGMENTS

The authors thank Eva Bringman and Marie C. Johansson at the Department of Psychiatry, Sahlgrenska University Hospital, Mölndal, for excellent technical assistance. This work was supported by the Swedish state under the agreement between the Swedish government and the county councils, the ALF-agreement (ALFGBG-722371, ALFGBG-720661, and ALFGBG-724331).

Authors’ disclosures available online ().

References

Norton

, Cammer

(1984) Isolation and characterization of myelin. In Myelin, 2nd ed. Morell P, ed. Plenum, New York, pp. 147–180.

Pintado-Sierra

, García-Álvarez

, Bribián

, Medina-Rodríguez

, Lebrón-Aguilar

, Garrido

, de Castro

, Fernández-Mayoralas

, Quintanilla-López

(2017) A comprehensive profiling of sulfatides in myelin from mouse brain using liquid chromatography coupled to high-resolution accurate tandem mass spectrometry. Anal Chim Acta 951, 89–98.

Schmitt

, Castelvetri

, Simons

(2015) Metabolism and functions of lipids in myelin. Biochim Biophys Acta 1851, 999–1005.

McGonigal

, Barrie

, Yao

, McLaughlin

, Cunningham

, Rowan

, Willison

(2019) Glial sulfatides and neuronal complex gangliosides are functionally interdependent in maintaining myelinating axon integrity. J Neurosci 39, 63–77.

Palavicini

, Wang

, Chen

, Ahmar

, Higuera

, Dupree

, Han

(2016) Novel molecular insights into the critical role of sulfatide in myelin maintenance/function. J Neurochem 139, 40–54.

Eckhardt

(2008) The role and metabolism of sulfatide in the nervous system. Mol Neurobiol 37, 93–103.

Fredman

, Wallin

, Blennow

, Davidsson

, Gottfries

, Svennerholm

(1992) Sulfatide as a biochemical marker in cerebrospinal fluid of patients with vascular dementia. Acta Neurol Scand 85, 103–106.

Tullberg

, Månsson

, Fredman

, Lekman

, Blennow

, Ekman

, Rosengren

, Tisell

, Wikkelsø

(2000) CSF sulfatide distinguishes between normal pressure hydrocephalus and subcortical arteriosclerotic encephalopathy. J Neurol Neurosurg Psychiatry 69, 74–81.

Jonsson

, Zetterberg

, Rolstad

, Edman

, Gouw

, Bjerke

, Lind

, Blennow

, Pantoni

, Inzitari

, Wallin

(2012) Low cerebrospinal fluid sulfatide predicts progression of white matter lesions: The LADIS study. Dement Geriatr Cogn Disord 34, 61–67.

10.

Haghighi

, Lekman

, Nilsson

, Blomqvist

, Andersen

(2012) Myelin glycosphingolipid immunoreactivity and CSF levels in multiple sclerosis. Acta Neurol Scand 125, 64–70.

11.

Moyano

, Pituch

, Li

, van Breemen

, Mansson

, Givogri

(2013) Levels of plasma sulfatides C18: 0 and C24: 1 correlate with disease status in relapsing-remitting multiple sclerosis. J Neurochem 127, 600–604.

12.

Rosenberg

, Wallin

, Wardlaw

, Markus

, Montaner

, Wolfson

, Iadecola

, Zlokovic

, Joutel

, Dichgans

, Duering

, Schmidt

, Korczyn

, Grinberg

, Chui

, Hachinski

(2016) Consensus statement for diagnosis of subcortical small vessel disease. J Cereb Blood Flow Metab 36, 6–25.

13.

Wallin

, Kapaki

, Boban

, Engelborghs

, Hermann

, Huisa

, Jonsson

, Kramberger

, Lossi

, Malojcic

, Mehrabian

, Merighi

, Mukaetova-Ladinska

, Paraskevas

, Popescu

, Ravid

, Traykov

, Tsivgoulis

, Weinstein

, Korczyn

, Bjerke

, Rosenberg

(2017) Biochemical markers in vascular cognitive impairment associated with subcortical small vessel disease - A consensus report. BMC Neurol 17, 102.

14.

Yoshitake

, Kiyohara

, Kato

, Ohmura

, Iwamoto

, K.

, Ohmori

, Nomiyama

, Kawano

, Ueda

, Sueishi

, Tsuneyoshi

, Fujishima

(1995) Incidence and risk factors of vascular dementia and Alzheimer’s disease in a defined elderly Japanese population: The Hisayama Study. Neurology 45, 1161–1168.

15.

Kalaria

(2016) Neuropathological diagnosis of vascular cognitive impairment and vascular dementia with implications for Alzheimer’s disease. Acta Neuropathol 131, 659–685.

16.

Toledo

, Arnold

, Raible

, Brettschneider

, Xie

, Grossman

, Monsell

, Kukull

, Trojanowski

(2013) Contribution of cerebrovascular disease in autopsy confirmed neurodegenerative disease cases in the National Alzheimer’s Coordinating Centre. Brain 136, 2697–2706.

17.

Jonsson

, Zetterberg

, van Straaten

, Lind

, Syversen

, Edman

, Blennow

, Rosengren

, Pantoni

, Inzitari

, Wallin

(2010) Cerebrospinal fluid biomarkers of white matter lesions - cross-sectional results from the LADIS study. Eur J Neurol 17, 377–382.

18.

Tarkowski

, Tullberg

, Fredman

, Wikkelsö

(2003) Correlation between intrathecal sulfatide and TNF-alpha levels in patients with vascular dementia. Dement Geriatr Cogn Disord 15, 207–211.

19.

Tullberg

, Hultin

, Ekholm

, Månsson

, Fredman

, Wikkelsø

(2002) White matter changes in normal pressure hydrocephalus and Binswanger disease: Specificity, predictive value and correlations to axonal degeneration and demyelination. Acta Neurol Scand 105, 417–426.

20.

Lam

, Wang

, Duan

, Wenk

, Kalaria

, Chen

, Lai

, Shui

(2014) Brain lipidomes of subcortical ischemic vascular dementia and mixed dementia. Neurobiol Aging 35, 2369–2381.

21.

Han

, Holtzman

, McKeel

, Kelley

, Morris

(2002) Substantial sulfatide deficiency and ceramide elevation in very early Alzheimer’s disease: Potential role in disease pathogenesis. J Neurochem 82, 809–818.

22.

Han

, Cheng

, Fryer

, Fagan

, Holtzman

(2003) Novel role for apolipoprotein E in the central nervous system. Modulation of sulfatide content. J Biol Chem 278, 8043–8051.

23.

Cheng

, Zhou

, Holtzman

, Han

(2010) Apolipoprotein E mediates sulfatide depletion in animal models of Alzheimer’s disease. Neurobiol Aging 31, 1188–1196.

24.

Cheng

, Wang

, Li

J-L

, Cairns

, Han

(2013) Specific changes of sulfatide levels in individuals with pre-clinical Alzheimer’s disease: An early event in disease pathogenesis. J Neurochem 127, 733–738.

25.

Couttas

, Kain

, Suchowerska

, Quek

L-E

, Turner

, Fath

, Garner

, Don

(2016) Loss of ceramide synthase 2 activity, necessary for myelin biosynthesis, precedes tau pathology in the cortical pathogenesis of Alzheimer’s disease. Neurobiol Aging 43, 89–100.

26.

Hong

, Kang

, Kim

, Baik

, Kim

, Shanta

, Jung

, Mook-Jung

, Kim

(2016) Global changes of phospholipids identified by MALDI imaging mass spectrometry in a mouse model of Alzheimer’s disease. J Lipid Res 57, 36–45.

27.

Mendis

, Grey

, Faull

, Curtis

(2016) Hippocampal lipid differences in Alzheimer’s disease: A human brain study using matrix-assisted laser desorption/ionization-imaging mass spectrometry. Brain Behav 6, e00517.

28.

Gónzalez de San Román

, Manuel

, Giralt

, Ferrer

, Rodríguez-Puertas

(2017) Imaging mass spectrometry (IMS) of cortical lipids from preclinical to severe stages of Alzheimer’s disease. Biochim Biophys Acta Biomembr 1859(9 Pt B), 1604–1614.

29.

Kaya

, Brinet

, Michno

, Syvänen

, Sehlin

, Zetterberg

, Blennow

, Hanrieder

(2017) Delineating amyloid plaque associated neuronal sphingolipids in transgenic Alzheimer’s disease mice (tgArcSwe) using MALDI imaging mass spectrometry. ACS Chem Neurosci 8, 347–355.

30.

Kaya

, Jennische

, Lange

, Tarik Baykal

, Malmberg

, Fletcher

(2020) Brain region-specific amyloid plaque-associated myelin lipid loss, APOE deposition and disruption of the myelin sheath in familial Alzheimer’s disease mice. J Neurochem 154, 84–98.

31.

Han

, Fagan

, Cheng

, Morris

, Xiong

, Holtzman

(2003) Cerebrospinal fluid sulfatide is decreased in subjects with incipient dementia. Ann Neurol 54, 115–119.

32.

Blomqvist

, Borén

, Zetterberg

, Blennow

, Månsson

, Ståhlman

(2017) High-throughput analysis of sulfatides in cerebrospinal fluid using automated extraction and UPLC-MS/MS. J Lipid Res 58, 1482–1489.

33.

Wallin

, Nordlund

, Jonsson

, Lind

, Edman

, Gothlin

, Stalhammar

, Eckerstrom

, Kern

, Borjesson-Hanson

, Carlsson

, Olsson

, Zetterberg

, Blennow

, Svensson

, Ohrfelt

, Bjerke

, Rolstad

, Eckerstrom

(2016) The Gothenburg MCI study: Design and distribution of Alzheimer’s disease and subcortical vascular disease diagnoses from baseline to 6-year follow-up. J Cereb Blood Flow Metab 36, 114–131.

34.

Wallin

, Nordlund

, Jonsson

, Blennow

, Zetterberg

, Ohrfelt

, Stalhammar

, Eckerstrom

, Carlsson

, Olsson

, Gothlin

, Svensson

, Rolstad

, Eckerstrom

, Bjerke

(2016) Alzheimer’s disease–subcortical vascular disease spectrum in a hospital-based setting: Overview of results from the Gothenburg MCI and dementia studies. J Cereb Blood Flow Metab 36, 95–113.

35.

Reisberg

, Ferris

, de Leon

, Crook

(1988) Global Deterioration Scale (GDS). Psychopharmacol Bull 24, 661–663.

36.

Wallin

, Edman

, Blennow

, Gottfries

, Karlsson

, Regland

, Sjogren

(1996) Stepwise comparative status analysis (STEP): A tool for identification of regional brain syndromes in dementia. J Geriatr Psychiatry Neurol 9, 185–199.

37.

Royall

, Mahurin

, Gray

(1992) Bedside assessment of executive cognitive impairment: The executive interview. J Am Geriatr Soc 40, 1221–1226.

38.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

39.

Morris

(1997) Clinical dementia rating: A reliable and valid diagnostic and staging measure for dementia of the Alzheimer type. Int Psychogeriatr 9(Suppl 1), 173–176; discussion 177-178.

40.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

41.

Erkinjuntti

, Inzitari

, Pantoni

, Wallin

, Scheltens

, Rockwood

, Roman

, Chui

, Desmond

(2000) Research criteria for subcortical vascular dementia in clinical trials. Neural Transm Suppl 59, 23–30.

42.

Wahlund

, Barkhof

, Fazekas

, Bronge

, Augustin

, Sjögren

, Wallin

, Ader

, Leys

, Pantoni

, Pasquier

, Erkinjuntti

, Scheltens

, European Task Force on Age-Related White Matter Changes (2001) A new rating scale for age-related white matter changes applicable to MRI and CT. Stroke 32, 1318–1322.

43.

Eckerström

, Eckerström

, Göthlin

, Molinder

, Jonsson

, Kettunen

, Svensson

, Rolstad

, Wallin

(2020) Characteristic biomarker and cognitive profile in incipient mixed dementia. J Alzheimers Dis 73, 597–607.

44.

Voevodskaya

, Simmons

, Nordenskjöld

, Kullberg

, Ahlström

, Lind

, Wahlund

, Larsson

, Westman

, Alzheimer’s Disease Neuroimaging Initiative (2014) The effects of intracranial volume adjustment approaches on multiple regional MRI volumes in healthy aging and Alzheimer’s disease. Front Aging Neurosci 6, 264.

45.

Geffen

, Butterworth

, Geffen

(1994) Test-retest reliability of a new form of the auditory verbal learning test (AVLT). Arch Clin Neuropsychol 9, 303–316.

46.

Reitan

(1985) The Halstead-Reitan Neuropsychological Test Battery. Neuropsychology Press, Tucson.

47.

Löfgren

, Ståhlman

, Forsberg

, Saarinen

, Nilsson

, Hansson

(2012) The BUME method: A novel automated chloroform-free 96-well total lipid extraction method for blood plasma. J Lipid Res 53, 1690–1700.

48.

Blennow

, Hampel

, Weiner

, Zetterberg

(2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol 6, 131–144.

49.

Yuki

, Sugiura

, Zaima

, Akatsu

, Hashizume

, Yamamoto

, Fujiwara

, Sugiyama

, Setou

(2011) Hydroxylated and non-hydroxylated sulfatide are distinctly distributed in the human cerebral cortex. Neuroscience 193, 44–53.

50.

Palermo

, Mazzucchi

, Della Vecchia

, Siciliano

, Bonuccelli

, Azuar

, Ceravolo

, Lista

, Hampel

, Baldacci

(2020) Different clinical contexts of use of blood neurofilament light chain protein in the spectrum of neurodegenerative diseases. Mol Neurobiol 57, 4667–4691.

51.

Zöller

, Meixner

, Hartmann

, Büssow

, Meyer

, Gieselmann

, Eckhardt

(2008) Absence of 2-hydroxylated sphingolipids is compatible with normal neural development but causes late-onset axon and myelin sheath degeneration. J Neurosci 28, 9741–9754.

52.

Rosengren

, Fredman

, Månsson

, Svennerholm

(1989) Lysosulfatide (galactosylsphingosine-3-O-sulfate) from metachromatic leukodystrophy and normal human brain. J Neurochem 52, 1035–1041.

53.

Novakova

, Singh

, Axelsson

, Ståhlman

, Adiels

, Malmeström

, Zetterberg

, Borén

, Lycke

, Cardell

, Blomqvist

(2018) Sulfatide isoform pattern in cerebrospinal fluid discriminates progressive MS from relapsing-remitting MS. J Neurochem 146, 322–332.