Plasminogen activator inhibitor‐1 serum levels in frontotemporal lobar degeneration

Abstract Plasminogen activator inhibitor‐1 (PAI‐1) impedes brain plasmin synthesis. Reduced plasmin activity facilitates cumulation of amyloid beta (Aβ) in Alzheimer's disease (AD). Since plasmin also regulates the synaptic activity, it is possible that altered PAI‐1 is present in other neurodegenerative disorders. We investigated whether PAI‐1 and its counter‐regulatory tissue plasminogen activator (tPA) are altered in serum of patients with dementia due to frontotemporal lobar degeneration (FTLD). Thirty five FTLD patients (21 in mild cognitive impairment stage (MCI) and 14 in dementia stage) and 10 cognitively healthy controls were recruited. Serum tPA and PAI‐1 protein levels were measured by anova. Correlation between biochemical and demographic data were explored by measuring Pearson correlation coefficient. Serum PAI‐1 levels were elevated in the FTLD dementia group as compared to FTLD MCI and controls. tPA serum levels and PAI‐1/tPA ratio did not significantly differ among groups. There was a negative correlation between PAI‐1 serum levels and disease severity measured by MMSE score. No correlations of tPA serum levels and PAI‐1/tPA ratio with MMSE were found. Increased PAI‐1 serum levels may serve as a marker of dementia in FTLD, suggesting that, besides Aβ pathway, the plasmin system may affect cognition through synaptic activity.

CNS conditions, including ischaemic stroke, 2 severe traumatic brain injury 8 and also in neurodegenerative disorders such as Parkinson's 9 and Alzheimer's 10 diseases.
As for neurodegenerative diseases, previous studies indicate that in Alzheimer's disease (AD) there is an imbalance of tPA and PAI-1 that leads to reduced plasmin synthesis and activity.Low plasmin levels were observed in 67% of AD APOE4 brains. 11rthermore, in two transgenic AD mouse models (Tg2576 and TgCRND8), it was observed that accumulation of amyloid beta (Aβ) peptide in the brain was associated with the upregulation of PAI-1 and inhibition of tPA. 12 In another brain tissue post-mortem study, it was shown that both PAI-1 and tPA were elevated in AD patients, suggesting that there is not alteration of the synthesis of plasmin. 13wever, it was suggested that the increase in PAI-1 could have consequences in neuronal activity, including synaptic deregulation and excitotoxicity.In another study, plasmin and its precursor plasminogen were not altered in post-mortem AD versus the control brain tissue. 10However, reduced plasmin activity was observed in AD brains, 10 indirectly favouring the accumulation of Aβ in aggregate or soluble forms. 11 a previous recent study, we have shown that while tPA serum levels were unaffected, PAI-1 levels increased in patients with dementia due to AD and its prodromal stage amnestic mild cognitive impairment (aMCI). 14In addition, there was a negative correlation between PAI-1 serum levels and Mini-Mental State Examination (MMSE) score.In fact, PAI-1 levels were higher in patients with dementia due to AD than in aMCI patients and even more than in cognitively healthy controls. 14Similarly, but to a lesser extent, the ratio between PAI-1 and tPA gradually increased from controls to patients with aMCI and those with AD dementia. 14e role of plasmin in AD appears to be related to the ability to reduce the Aβ in its soluble forms. 15,16These data suggest that in AD dementia, an imbalance of these two enzymes may favour the accumulation of Aβ and therefore AD. 10,11 Despite this, we do not know if defects in the plasminogen/plasmin system are also associated with dementia in other non-AD pathologies and, therefore, unrelated to the accumulation of Aβ in neurons.
Frontotemporal dementia (FTD) represents a group of clinically defined brain disorders characterized by prominent behavioural/ cognitive symptoms like disinhibition, apathy, executive deficits and language disorders.Several clinical forms have been recognized based on the localisation of neurodegeneration and subsequent clinical manifestation. 17FTD is associated with several underlying neurodegenerative diseases characterized by frontotemporal lobar degeneration (FTLD). 18FTLD is characterized by deposits of tau or other proteins but, contrary to AD, intracerebral deposition of Aβ is not a typical hallmark of FTLD. 19,20Thus, subjects affected by FTLD represent a group of patients with dementia unrelated to the Aβ pathophysiological accumulation present in AD.
Thus, in this study we investigated these enzymes regulating plasmin synthesis in FTLD patients in mild cognitive impairment (FTLD-MCI) and dementia (FTLD dementia) and in cognitively healthy controls.Our aim was to establish whether serum levels of tPA and PAI-1 were altered in FTLD patients and whether there is a correlation with MMSE score.

| Participants
Thirty one FLTD subjects from the Czech Brain Aging Study, a longitudinal, memory clinic-based study on aging and cognitive impairment, 21 and 10 cognitively healthy participants were included in the study.
Among these subjects, 14 FLTD subjects were in the dementia stage (FTLD dementia-six patients with a semantic variant of primary progressive aphasia, five patients with a behavioural variant of FTD, two patients with FTD-amyotrophic lateral sclerosis and one patient with a non-fluent variant of primary progressive aphasia).
The remaining 21 patients were in MCI stage (FTLD MCI-5 patients with a semantic variant of primary progressive aphasia, 12 patients with the behavioural variant of FTD, 1 patient with FTDamyotrophic lateral sclerosis, and 5 patients with non-fluent variant of primary progressive aphasia).The main criterion to diagnose the dementia stage was based on the impairment of activities of daily living reported by the patient's informant during the structured interview.Diagnoses of bvFTD was based on Rascovski criteria 22 and language variants were diagnosed based on Gorno-Tempini criteria. 23e diagnosis of the MCI stage was based on Petersen's criteria for MCI, including cognitive complaints, evidence of memory dysfunction on neuropsychological testing, generally intact activities of daily living, and absence of dementia. 24All participants underwent standard neurological and laboratory evaluations, comprehensive neuropsychological examination, and 1.5-T brain magnetic resonance imaging (MRI) within 3 months from the initial visit.The majority of participants underwent cerebrospinal fluid (CSF) sampling with analysis of AD biomarkers (n = 40).FTLD patients (n = 33) had CSF negative for amyloid biomarkers (normal Aβ42), and healthy controls (n = 7) were negative for all three AD biomarkers (Aβ42, T-tau, P-tau181) based on internal cut-offs.Individuals who did not donate CSF (n = 5) underwent Aβ PET imaging with a negative Aβ load.The results of biomarker analyses are summarized in Table 1.All participants in this study signed written informed consent approved by the Motol University Hospital ethics committee.

| Exclusion criteria
The participants were not included in the study if they had a history of neurological or psychiatric disorders other than FTLD potentially causing cognitive deficit (i.e.history of stroke, Parkinson's or Alzheimer's disease, brain tumour, alcohol abuse), hearing difficulties, depressive symptomatology (≥6 points on the 15-item Geriatric Depression Scale) 25 or had significant vascular impairment on the brain MRI (Fazekas scale more than 2). 26

| Blood and CSF sampling
Blood samples were drawn by venipuncture.After collection of the whole blood, the blood was allowed to clot by leaving it undisturbed at room temperature for 15 min.Blood samples were then centrifuged at 1700 × g at 20°C for 5 min.All samples were centrifuged within 30 min of collection.Serum supernatant was collected, divided into 0.5 mL polypropylene aliquot tubes, and kept frozen at −80°C until further use.
CSF was obtained by lumbar puncture in a supine position.The pattern of CSF collection was the same for all study individuals, with the patient lying on their side, and CSF drawn at the L3/L4 or L4/L5 region and collected in 8-ml polypropylene tubes.Each sample was then combined, gently mixed, and centrifuged within 30 min of lumbar puncture at 1700 × g at 20°C for 5 min.CSF was aliquoted in polypropylene tubes of 0.5 mL and stored at −80°C until analysis.
Before biomarker analysis, serum and CSF samples stored at −80°C were thawed and vortexed for 15 s.

| Immunological assays
AD biomarkers in CSF (Aβ42, T-tau, and P-tau181) were analysed using ELISA (Innogenetics, Ghent, Belgium) in the Cerebrospinal Fluid Laboratory, Institute of Immunology and Department of Neurology, Second Faculty of Medicine, Charles University and Motol University Hospital.Unbiased cut-offs of less than 665 pg/ mL and more than 48 pg/mL and 358 pg/mL were used to define amyloid-b1-42, p-tau181, and total tau positivity, respectively.These predefined cut-offs 27,28 were based on internal receiver operating characteristic (ROC) analyses and were validated against amyloid PET status in the Czech Brain Aging Study with 79% agreement and areas under the ROC curves (AUCs) of 85%. 29rum levels of PAI-1 (Catalogue Number: DY1786) and tPA (Catalogue Number: DY7449) were measured with commercial ELISA kits from R and D Systems as previously described. 14All samples were tested in duplicate.Values were ng/mL for PAI-1 and pg/ mL for tPA.
The difference between FTLD MCI and controls was not significant.
There were no statistical differences in serum levels of tPA among the groups (p = 0.709).In addition, PAI-1/tPA ratio was not significantly different among groups (p = 0.745) (Figure 1).

| DISCUSS ION
This study investigated whether patients with dementia due to FTLD have altered serum levels of tPA and PAI-1 compared to those with MCI due to FTLD and controls.We found that FLTD dementia patients were characterized by increased levels of the inhibitor of plasmin PAI-1 versus FTLD MCI and controls.In addition, PAI-1 levels negatively correlated with the MMSE score.
These data align with our previous study showing that PAI-1 levels in serum are increased in patients with AD dementia compared to aMCI and healthy controls. 14The present findings support the idea that the syndrome of dementia is associated with high PAI-1 levels in serum, independently from the pathology causing dementia, which was AD in the mentioned study and FLTD in the present study.
High levels of PAI-1 could mean that plasmin synthesis is reduced in dementia.In line with these data, it has been demonstrated that plasmin activity is decreased in AD animal models and humans. 10,12rthermore, in animal models of AD 31 and cell cultures, 10,32,33 when PAI-1 is pharmacologically inhibited, plasmin increases its degrading activity on Aβ.
The present findings also suggest that increased PAI-1 serum levels characterize even dementia not associated with Aβ accumulation.This assumption implies that, besides Aβ degradation, the plasmin system may be involved in cognitive functions through other signalling pathways in the CNS.In line with this hypothesis, it has been shown that the plasmin system is important in processes regulating cognitive functions in the CNS, such as synaptic plasticity, 4 NMDA receptor-mediated signalling 34 and long-term potentiation. 34,358][39] It has been repetitively demonstrated that BDNF levels in the CNS and the periphery are altered during pathological conditions characterized by cognitive dysfunctions and dementia. 40,41It would be interesting to verify whether circulating and brain levels of BDNF and PAI-1 show some correlation.Besides this mechanistic hypothesis, these data also suggest that a pharmacological intervention aimed at increasing plasmin in the CNS, or reducing the level of its inhibitor PAI-1, could represent a new and alternative strategy to treat dementia associated with various neurological disorders.Nonetheless, before undertaking this type of study, the involvement of plasmin and its enzymes in dementia need to be investigated in larger cohorts of subjects affected by dementia of different origin.

| LI M ITATI O N S
There are some limitations in the interpretation of our results.The number of subjects included is small due to the difficulty to recruit patients with FTLD and healthy subjects into the study.Moreover, we only included a single time point measurement at admission, thus we do not know how serum levels of PAI-1 and tPA change during disease progression.Also, we did not measure tPA activity in serum and we do not know if the increase in PAI-1 could have affected such activity while tPA serum levels were unchanged.For these reasons, our data must be considered as preliminary observations.More

2. 6 |
Statistical analysisSerum levels of PAI-1 and tPA and PAI-1/tPA ratio among the groups were compared by anova and Fisher-protected least post hoc.Demographic data were analysed by Chi-squared test.Correlation between biochemical and demographic data were explored by measuring Pearson correlation coefficient.p <0.05 was considered statistically significant.The level of statistical significance was set at p < 0.05.Statistical analysis was performed by Statview (SAS Institute).TA B L E 1 Clinical and demographic characteristics of FTLD patients and cognitively healthy controls.