Are Plasma Levels of Vascular Adhesion Protein-1 Associated Both with Cerebral Microbleeds in Multiple Sclerosis and Intracerebral Haemorrhages in Stroke?

Cerebral microbleeds (CMBs) are defined as small and hypointense areas which could correspond to clusters of hemosiderin-laden macrophages resulting from small self-limiting haemorrhages.[1] [2] CMBs have been associated with aging, traumatic brain injury, stroke, and neurodegenerative disorders, among them the cerebral amyloid angiopathy. CMBs are potentially a radiological biomarker of the cerebral small vessel disease prone to bleeding and developing spontaneous intracerebral haemorrhages (ICHs).[3] [4] Recently, in patients with atrial fibrillation anticoagulated after ischaemic stroke or transient ischaemic attack, the presence of CMBs was independently associated with symptomatic ICH risk, and could be used to inform anticoagulation decisions.[5] [6]

Failure of blood–brain barrier integrity leading to focal extravascular leakage of blood components is a decisive event in the pathogenesis of multiple sclerosis (MS), a disease characterized by multifocal demyelinated lesions within the central nervous system.[7] Extravascular leakage of blood may have the features of radiologically measurable CMBs.[8] Adhesion molecules participating in blood–brain barrier disruption and in inflammatory responses, supported by fibrinogen extravasation and promoting tissue factor expression,[9] [10] [11] [12] in turn could be involved in the formation of CMBs.

Vascular adhesion protein-1 (VAP-1) facilitates leukocyte infiltration into inflamed tissue[13] through an enzymatic activity that mediates cell binding to the vessel wall. VAP-1 catalyzes formation of free radicals from its substrates on leukocytes, providing an inflammatory microenvironment and causing expression of additional adhesion molecules.[14] VAP-1 is both a cell surface and a circulating protein, released through cleavage mechanisms that are only partially defined.[14] [15]

Higher plasma activity level of VAP-1 has been found both in consecutive patients with spontaneous ICH[16] and in patients with stroke treated with tissue plasminogen activator, who subsequently experienced ICH.[17] In animal models, VAP-1 inhibition decreased both immune cell infiltration after ICH and micro-vascular dysfunction.[18] [19]

Taking advantage of MS as a disease model,[2] [20] [21] we aimed at extending our knowledge about the association between plasma levels of VAP-1 and occurrence of CMBs in MS patients, assessed by magnetic resonance imaging (MRI) measures in a cross-sectional study ([Table 1] and [Supplementary Material], available in the online version). In turn, this could underline potential biological relations between CMBs and ICH.

Table 1 Cerebral microbleeds in MS patients

	All MS	RR-MS	P-MS	HI
Sample size, n	138	85	53	42
Female, n (%)	100 (72.5)	60 (70.6)	40 (75.5)	31 (73.8)
Age, y	54.3 ± 10.8	50.1 ± 10.7	60.9 ± 7.2	51.0 ± 14.3
Cerebral microbleeds (CMBs)
Individuals with CMB, n (%)	12 (9.6)	5 (6.3)	7 (15.2)	3 (7.3)
Number of CMBs
NA	13	6	7	1
0	113	74	39	38
1	9	4	5	2
2	2	−	2	−
≥ 3	1	1	−	1
CMB volume, mm3	22.9 ± 21.04	33.4 ± 29	15.4 ± 9.8	14.8 ± 13.65
Disease-modifying treatment (DMT)
No DMT	27 (19.6)	15 (17.6)	12 (22.6)	−
Interferon-β	45 (32.6)	30 (35.3)	15 (28.3)
Glatiramer acetate	42 (30.4)	23 (27.1)	19 (35.9)
Natalizumab	5 (3.6)	4 (4.7)	1 (1.9)
Other DMT[a]	19 (13.8)	13 (15.3)	6 (11.3)

Abbreviations: HI, healthy individuals; MS, multiple sclerosis; NA, not available; P-MS, progressive multiple sclerosis; RR-MS, relapsing remitting multiple sclerosis.

Note: 10 CMB patients were under the following DMTs: 6, glatiramer acetate; 2, interferon-β; 1, natalizumab; 1, fingolimod. Age and CMB volume are reported as mean ± standard deviation.

^a Other DMTs included intravenous immunoglobulin, mitoxantrone and methotrexate.

Concentration of soluble VAP-1 molecule ([Fig. 1]) was measured in ethylenediaminetetraacetic acid plasma samples using a multiplex assay (see [Supplementary Material], available in the online version), which permitted the parallel investigation of several adhesion molecules.

Fig. 1 (A) Soluble vascular adhesion protein 1 (VAP-1) concentration in multiple sclerosis (MS) patients with cerebral microbleeds (CMBs) (MS (+)CMB, n = 12), MS without CMB (MS(–)CMB, n = 113). Mann–Whitney U test showed a trend for difference p = 0.076. (B) VAP-1 levels among patients on disease-modifying treatments (MS-DMTs, n = 111), patients without treatment (MS-none, n = 27) and healthy individuals (HIs, n = 42) groups. The adjusted p-values from Dunn's multiple comparison test are provided for comparisons between groups where Kruskal–Wallis test was significant. (C) VAP-1 levels in MS cohort according to the DMT (IFNb: Interferon-β,, n = 45; GA: glatiramer acetate, n = 42; Other: other DMTs, n = 19; None: no DMT, n = 27). Since very few patients ([Table 1]) had been treated with natalizumab, we did not include them in this panel of analysis. The adjusted p-values from Dunn's multiple comparison test are provided for comparisons between groups where Kruskal–Wallis test resulted significant. (D) VAP-1 concentration in MS patients without treatment (none DMTs, n = 25) grouped for presence of CMBs (MS(+)CMB-none, n = 2 and MS(–)CMB-none, n = 23). The significant p-value from Mann–Whitney U test is shown. In all panels, median values and interquartile ranges are shown.

Analysis of VAP-1 concentration by non-parametric Mann–Whitney U test revealed a trend for higher VAP-1 in MS(+)CMB versus MS(–)CMB (median = 300.9, interquartile range [IQR] =192.2–401.5 ng/mL vs. median = 237.2, IQR = 195.8–276.1 ng/mL, p = 0.076, [Fig. 1A]). VAP-1 levels were not associated with CMB volume (p = 0.86, linear regression). Interestingly, mean levels of this soluble adhesion molecule were numerically 32% higher in MS patients with CMBs than in healthy individuals (HIs) (300 ± 105 ng/mL vs. 227.3 ± 50 ng/mL), and 35% higher in ICH patients than in controls,[16] a very similar proportion.

None of the other MRI measures (T2-LV, T1-LV, normalized brain, normalized cortical, lateral ventricular, deep grey matter and thalamus volumes) in MS patients were associated with VAP-1 concentration by regression analyses (data not shown) adjusting for age, sex and type of disease-modifying treatment (DMT).

Taking into account that DMTs could potentially influence VAP-1 levels, the proportional distribution of DMT's use between MS(+)CMB versus MS(–)CMB was also assessed (p = 0.761, chi-square). Analysis among MS patients on DMTs, patients without treatment and HI groups provided a significant variation in VAP-1 levels (p = 0.001, Kruskal–Wallis test, [Fig. 1B]). In particular, MS patients on DMTs had lower VAP-1 levels than those without treatment (median = 231.1, IQR = 190.3–276.3 ng/mL vs. median = 272.7, IQR = 251.6–324 ng/mL; p = 0.002, Dunn's multiple comparison test). Further, VAP-1 levels in MS patients without treatment were higher than in HI (median = 272.7, IQR = 251.6–324 ng/mL vs. median = 233.2, IQR = 181.1–263.5 ng/mL; p = 0.002, Dunn's multiple comparison test, [Fig. 1B]).

Modulation of VAP-1 levels by these drugs ([Table 1]) was further investigated in MS patients (p = 0.002, Kruskal–Wallis test, [Fig. 1C]). This comparison showed that the decrease in VAP-1 levels was mostly influenced by glatiramer acetate treatment compared with MS without treatment (median = 225.3, IQR = 181.9–253.3 ng/mL vs. median = 272.7, IQR = 251.6–324 ng/mL; p = 0.002, Dunn's multiple comparison test, [Fig. 1C]).

Finding higher VAP-1 concentration in plasma of MS patients without any DMT and lower in those on DMTs, clearly indicated that DMTs were not responsible for the higher VAP-1 levels in patients with CMBs. Instead, DMTs could mask even higher VAP-1 levels in patients with CMBs as indicated by comparison of patients not on DMTs (MS(+)CMB no DMT, median = 400.8, IQR = 333.2–468.4 ng/mL vs. MS(–)CMB no DMT, median = 266.5, IQR = 238–315, p = 0.007, Mann–Whitney U test, [Fig. 1D]).

Of note, in patients with CMBs, plasma concentration differences were observed for VAP-1 and not for other soluble adhesion molecules (neural cell adhesion molecule, intercellular adhesion molecule 1 and vascular cell adhesion molecule 1), investigated in multiplex assays (unpublished data).

Our observations have some limitations: (1) CMBs are heterogeneous as suggested by their radiographic appearance,[22] and in MS they are likely smaller in size as compared with CMBs investigated in ICH[6]; (2) we have evaluated in peripheral blood the soluble VAP-1, the level and role of which in brain vessel endothelium are only inferred; (3) the VAP-1 activity values have the potential to reveal the presence of a functional enzyme in plasma,[16] [17] whereas we measured the VAP-1 protein concentration, which however corresponds well to the level of enzymatic activity found in serum or plasma[14]; (4) the number of patients with CMBs in our study was low; and (5) we are unaware of the effects and/or the biological implications of VAP-1 levels on CMBs over time. To confirm and detail this association, further investigation in prospective studies of VAP-1 levels in larger cohorts of patients with CMBs is needed, including cerebral amyloid angiopathy, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy and small vessel disease patients.

In light of the previously detected association between CMBs and ICH occurrence,[6] finding similarly increased plasma levels of VAP-1 both in patients with ICH[16] and, at least as a trend, in MS with CMBs, is hypothesis generating. VAP-1, present in endothelium and smooth muscle cells of the brain vessel,[23] promotes inflammatory cell recruitment,[24] which might be associated with the conveyance of pro-coagulant mediators to the sites of vascular injury.[25] It is tempting to speculate that VAP-1 contributes both to cerebral micro-vascular endothelial cells dysfunction and to small and self-limiting haemorrhages, revealed by MRI. Our data foster the investigation in prospective studies of VAP-1 and other molecules potentially bridging ICH and CMBs.

Ethical Approval

The study protocol was approved by the local Institutional Review Boards of University of Buffalo, USA (CEG-MS study; IRB ID: MODCR00000352) and of University/Hospital of Ferrara, Italy (IRB ID: 170585). All participants gave their written informed consent.

• References

• 1 Tanaka A, Ueno Y, Nakayama Y, Takano K, Takebayashi S. Small chronic hemorrhages and ischemic lesions in association with spontaneous intracerebral hematomas. Stroke 1999; 30 (08) 1637-1642
• 2 Zivadinov R, Ramasamy DP, Benedict RR. , et al. Cerebral microbleeds in multiple sclerosis evaluated on susceptibility-weighted images and quantitative susceptibility maps: a case-control study. Radiology 2016; 281 (03) 884-895
• 3 Wardlaw JM, Smith EE, Biessels GJ. , et al; STandards for ReportIng Vascular changes on nEuroimaging (STRIVE v1). Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 2013; 12 (08) 822-838
• 4 Miwa K, Tanaka M, Okazaki S. , et al. Multiple or mixed cerebral microbleeds and dementia in patients with vascular risk factors. Neurology 2014; 83 (07) 646-653
• 5 Paciaroni M, Agnelli G, Ageno W, Caso V. Timing of anticoagulation therapy in patients with acute ischaemic stroke and atrial fibrillation. Thromb Haemost 2016; 116 (03) 410-416
• 6 Wilson D, Ambler G, Shakeshaft C. , et al; CROMIS-2 collaborators. Cerebral microbleeds and intracranial haemorrhage risk in patients anticoagulated for atrial fibrillation after acute ischaemic stroke or transient ischaemic attack (CROMIS-2): a multicentre observational cohort study. Lancet Neurol 2018; 17 (06) 539-547
• 7 Frohman EM, Racke MK, Raine CS. Multiple sclerosis--the plaque and its pathogenesis. N Engl J Med 2006; 354 (09) 942-955
• 8 Fisher MJ. Brain regulation of thrombosis and hemostasis: from theory to practice. Stroke 2013; 44 (11) 3275-3285
• 9 Hoppe B. Fibrinogen and factor XIII at the intersection of coagulation, fibrinolysis and inflammation. Thromb Haemost 2014; 112 (04) 649-658
• 10 López ML, Bruges G, Crespo G. , et al. Thrombin selectively induces transcription of genes in human monocytes involved in inflammation and wound healing. Thromb Haemost 2014; 112 (05) 992-1001
• 11 Wang S, Reeves B, Pawlinski R. Astrocyte tissue factor controls CNS hemostasis and autoimmune inflammation. Thromb Res 2016; 141 (Suppl. 02) S65-S67
• 12 Petersen MA, Ryu JK, Akassoglou K. Fibrinogen in neurological diseases: mechanisms, imaging and therapeutics. Nat Rev Neurosci 2018; 19 (05) 283-301
• 13 Salmi M, Jalkanen S. A 90-kilodalton endothelial cell molecule mediating lymphocyte binding in humans. Science 1992; 257 (5075): 1407-1409
• 14 Salmi M, Jalkanen S. Vascular adhesion protein-1: a cell surface amine oxidase in translation. Antioxid Redox Signal 2017
• 15 Stolen CM, Yegutkin GG, Kurkijärvi R, Bono P, Alitalo K, Jalkanen S. Origins of serum semicarbazide-sensitive amine oxidase. Circ Res 2004; 95 (01) 50-57
• 16 Hernandez-Guillamon M, Solé M, Delgado P. , et al. VAP-1/SSAO plasma activity and brain expression in human hemorrhagic stroke. Cerebrovasc Dis 2012; 33 (01) 55-63
• 17 Hernandez-Guillamon M, Garcia-Bonilla L, Solé M. , et al. Plasma VAP-1/SSAO activity predicts intracranial hemorrhages and adverse neurological outcome after tissue plasminogen activator treatment in stroke. Stroke 2010; 41 (07) 1528-1535
• 18 Ma Q, Manaenko A, Khatibi NH, Chen W, Zhang JH, Tang J. Vascular adhesion protein-1 inhibition provides antiinflammatory protection after an intracerebral hemorrhagic stroke in mice. J Cereb Blood Flow Metab 2011; 31 (03) 881-893
• 19 Xu H, Testai FD, Valyi-Nagy T. , et al. VAP-1 blockade prevents subarachnoid hemorrhage-associated cerebrovascular dilating dysfunction via repression of a neutrophil recruitment-related mechanism. Brain Res 2015; 1603: 141-149
• 20 Ziliotto N, Bernardi F, Jakimovski D. , et al. Hemostasis biomarkers in multiple sclerosis. Eur J Neurol 2018; 25 (09) 1169-1176
• 21 Ziliotto N, Baroni M, Straudi S. , et al. Coagulation factor XII levels and intrinsic thrombin generation in multiple sclerosis. Front Neurol 2018; 9 (245) 245
• 22 Renard D. Cerebral microbleeds: a magnetic resonance imaging review of common and less common causes. Eur J Neurol 2018; 25 (03) 441-450
• 23 Airas L, Lindsberg PJ, Karjalainen-Lindsberg ML. , et al. Vascular adhesion protein-1 in human ischaemic stroke. Neuropathol Appl Neurobiol 2008; 34 (04) 394-402
• 24 Jalkanen S, Karikoski M, Mercier N. , et al. The oxidase activity of vascular adhesion protein-1 (VAP-1) induces endothelial E- and P-selectins and leukocyte binding. Blood 2007; 110 (06) 1864-1870
• 25 Ghasemzadeh M, Hosseini E. Intravascular leukocyte migration through platelet thrombi: directing leukocytes to sites of vascular injury. Thromb Haemost 2015; 113 (06) 1224-1235