Introduction

Sulfatides, which are sulfuric esters of galactosylceramides (galactocerebrosides), are a group of sulfated glycosphingolipids [1]. They are located primarily on the plasma membrane, especially the myelin sheath. From the late 1980s, our team, as well as other research groups, have focused on the role of sulfatides in thrombosis and hemostasis [2, 3]. In 1987, Hara and Taketomi observed high amounts of sulfatides in serum lipoproteins, including low-density lipoprotein from Watanabe hyperlipidemic rabbits [4], and their accumulation in atheromatous aortae [5]. We previously reported that sulfatides induced large thrombi if they were injected into rats with intensely damaged vessels [6, 7]. Furthermore, we also demonstrated that sulfatides play a major role in inflammation in vascular injury and initiate neointimal thickening after bare-metal stent implantation [8]. Recently, Li et al. reported high peripheral serum sulfatide levels in patients with ST segment elevation myocardial infarction (STEMI) [9]. These observations suggest that sulfatides in sera and in atheromatous plaques in humans could contribute to the pathogenesis of acute coronary syndrome.

In this study, we directly measured sulfatide levels proximal and distal to a stenotic lesion before and after percutaneous coronary intervention (PCI) in patients with stable atherosclerotic coronary artery disease. This study aimed to determine whether plaque rupture affects local sulfatide levels across plaques.

Methods

Study Design

We included patients with stable atherosclerotic coronary artery disease who underwent planned single coronary stent implantation or balloon angioplasty alone using a drug-coated balloon (with the exception of three patients; see Table 1) from October 2014 to December 2015. In all of the patients (one-vessel disease, 27/41 (65.9%); two-vessel disease, 8/41 (19.5%); three-vessel disease, 6/41 (14.6%)), the target lesion was a type A or B lesion according to the American College of Cardiology/American Heart Association lesion classification [10] in the proximal left anterior descending artery. To reduce the heterogeneity of the study population, we excluded patients with chronic kidney disease (CKD) stage G3b or higher [11]. All of the patients received oral medications, including antiplatelet agents, statins, beta-blockers, renin–angiotensin system inhibitors, long-acting nitrates, or calcium channel blockers (Table 1).

Table 1 Characteristics of patients undergoing PCI
Full size table

The sampling procedure is shown in Fig. 1. Blood was sampled at four points in each patient. Before the procedure, point 1 was sampled proximal to the stenotic lesion and point 2 was sampled distal to the lesion. Point 3 was sampled proximally 15 min after the procedure and point 4 distally 15 min after the procedure. A microcatheter was used to sample blood distal to the lesion. All patients were divided into two groups according to the requirement for pre-balloon dilation (pre-BD) before sampling blood at point 2 because of a incapability to pass the microcatheter through the lesion.

Fig. 1
figure 1

Summary of PCI procedures. Blood was sampled at four points in each patient. Points 1 and 2 were sampled proximally and distally, respectively, to the stenotic lesion before the procedure. Points 3 and 4 were sampled proximally and distally, respectively, to the stenotic lesion 15 min after the procedure. Patients were principally treated by either a DES or DCB. All patients were divided into two groups according to the requirement for pre-BD before sampling blood at point 2 because of the inability to pass a microcatheter through the lesion. According to this additional procedure, the sampling time at point 2 was delayed approximately 15–20 min in the pre-BD (+) group compared with the pre-BD (−) group. DCB, drug-coated balloon; DES, drug-eluting stent; pre-BD, pre-balloon dilation

Full size image

Intravenous heparin was administered to maintain an activated clotting time of 200–250 s during the procedure. The blood samples were run off at one time in a blinded fashion. At each sampling point, whole blood was immediately collected into a plain tube or a tube containing acid citrate dextrose.

Measurement of Sulfatides

Sulfatides were extracted from 50 μL of serum with a ratio of n-hexane/isopropanol of 3:2 (v/v). Sulfatides were then analyzed as lyso forms using a matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (Autoflex™; Bruker Daltonik, Bremen, Germany) in the negative ion mode [12].

Measurement of Coagulative, Fibrinolytic, and Coronary Atherosclerosis Markers

Thrombin–antithrombin complex (TAT), plasmin-α2-plasmin inhibitor complex (PIC), and high-sensitivity C-reactive protein (hs-CRP) levels were analyzed at SRL, Inc. (Tokyo, Japan) using a chemiluminescent enzyme immunoassay, latex photometric immunoassay, and nephelometry, respectively. Diacron-reactive oxygen metabolite (d-ROM) testing for oxidative stress was performed at Dokkyo Medical University Saitama Medical Center laboratory as previously described [13]. The results of the d-ROM test are expressed in arbitrary units (Caratelli Units [U.CARR]), where 1 U.CARR corresponds to 0.08 mg/100 mL H2O2.

Statistical Analysis

Values are expressed as the mean ± SD. Variables were evaluated by repeated-measures analysis of variance for intragroup comparison. For comparison between values at point 1 from different groups, the unpaired t test was used. A p value < 0.05 was considered to be significant. Analyses were performed using KaleidaGraph, version 4.5 (Synergy Software, Reading, PA, USA).

Results

Sulfatide Levels from Four Different Sampling Points

Serum sulfatide levels that were obtained from the four different sampling points are shown in Fig. 2. In the group without balloon dilation before PCI [(pre-BD (−)] (Fig. 2, left), mean sulfatide levels at points 2 (4.60 ± 4.04 nmol/mL), 3 (4.35 ± 3.76 nmol/mL), and 4 nmol/mL (4.53 ± 3.26) were significantly higher than that at point 1 (2.49 ± 1.11 nmol/mL) (vs point 2, p < 0.02; vs point 3, p < 0.03; vs point 4, p < 0.03). In contrast, in the group with pre-BD [pre-BD (+)] (Fig. 2, right), there was no significant difference in mean sulfatide levels between each sampling point (point 1, 4.29 ± 2.74 nmol/mL; point 2, 4.35 ± 3.97 nmol/mL; point 3, 3.57 ± 2.73 nmol/mL; and point 4, 4.55 ± 3.09 nmol/mL). These results suggested that the PCI procedure induced elevation of sulfatides across plaques and that the pre-BD procedure affected this change. Sulfatide levels at sampling point 1 were the most likely to be unaffected by the PCI procedure and showed the baseline value at this point. The mean sulfatide level in the pre-BD (+) group was significantly higher than that in the pre-BD (−) group at sampling point 1 (p < 0.001) (Fig. 2).

Fig. 2
figure 2

Changes in sulfatide levels caused by PCI. In the pre-BD (−) group (left), mean sulfatide levels at points 2 (4.60 ± 4.04 nmol/mL), 3 (4.35 ± 3.76 nmol/mL), and 4 (4.53 ± 3.26 nmol/mL) were significantly higher than that at point 1 (2.49 ± 1.11 nmol/mL; vs point 2, p < 0.02; vs point 3, p < 0.03; vs point 4, p < 0.03). In the pre-BD (+) group (right), there was no significant difference in mean sulfatide among the sampling points. The mean sulfatide level in the pre-BD (+) group (right) was significantly higher than that in the pre-BD (−) group (left) at sampling point 1 (p < 0.001). NS, no significant difference; pre-BD, pre-balloon dilation

Full size image

Levels of TAT, PIC, Hs-CRP, and d-ROMs

TAT and PIC are parameters that indicate coagulative and fibrinolytic states [14], respectively, while hs-CRP and d-ROM levels indicate vascular inflammation [15, 16]. Sequential changes in these variables during the observation period were not observed (Table 2), regardless of whether there was a pre-BD procedure. However, mean values of PIC and d-ROMs tended to be higher in the pre-BD (+) group than in the pre-BD (−) group. There was no correlation between the sulfatide level and these parameters at each corresponding point.

Table 2 TAT, PIC, hs-CRP, and d-ROM levels from four different sampling points
Full size table

Discussion

In our study, we measured serum sulfatide levels at the proximal (points 1 and 3) and distal (points 2 and 4) sites of atheromatous plaques before and after PCI. Patients in the pre-BD (−) group had a significantly lower mean sulfatide level at point 1 compared with the other points. This finding indicated that elevated sulfatide levels were probably derived from atheromatous plaques themselves, consistent with the finding of accumulated sulfatides in rabbit aortae [5]. In contrast, such an elevation in sulfatide levels in patients in the pre-BD (+) group was not observed. This finding is probably because of technical difficulty in collecting sulfatides from coronary arteries. These results suggest that a substantial elevation in sulfatide levels in PCI from plaques is caused by mechanical stimuli, such as chafing or injuring of vessels, rather than continuous production from sites. Released sulfatides from plaques because of the pre-BD procedure may have been diluted in the circulation before sample collection because of the longer sampling time (Fig. 1). Further exploration of sulfatides in the coronary circulation has yet to be undertaken. In general, until now, studies of glycosphingolipids (including sulfatides) have focused mainly on the plasma membranes of animal cells for their development and differentiation [17]. Therefore, information regarding the origins and metabolism of glycosphingolipids in body fluids is very limited.

In experiments using WHHL rabbits, sulfatides were accumulated in the aorta and these were assumed to come from serum lipoproteins. This situation could be applicable in humans. In support of this assumption, a positive correlation between serum sulfatide levels and carotid intima–media thickness in patients with familial hypercholesterolemia was reported previously [18]. In our study, the mean sulfatide level in the pre-BD (+) group was significantly higher than that in the pre-BD (−) group at sampling point 1. This finding suggests that higher serum sulfatide levels cause a higher plaque burden, resulting in promotion of atheromatous plaque formation.

We measured TAT, PIC levels in our study. Coagulation and fibrinolysis are causally associated with cardiovascular events [14]. Levels of hs-CRP are a biomarker for coronary atherosclerosis [15] and d-ROMs indicate oxidative stress, which damages vessels [16]. However, in our study, these variables showed no sequential changes within the observation period, emphasizing the significance of sulfatides.

Because of the adequate use of heparin, thrombotic incidents with the procedure are unusual. There were no significant changes in TAT and PIC levels in our study during the procedure. However, once a vulnerable plaque ruptures in the absence of adequate antithrombotics, sulfatides that are released from the site may make a powerful contribution to inducing plaque thrombosis. In the early 1980s, sulfatides were reported to activate blood coagulation factor XII (FXII) in vitro, suggesting a possible role of sulfatides in promoting coagulation [19]. The physiological importance of FXII was disputed until a recent report in which FXII was shown to participate in plaque thrombus formation in a mouse model [20]. Furthermore, sulfatides enhance aggregation of activated platelets [21]. These results suggest the possible contribution of sulfatides in plaques for not only enhancing atherosclerosis but also accelerating thrombosis under injured endothelium. Indeed, Li et al. reported high peripheral serum sulfatide levels in patients with STEMI [9].

While sulfatides probably promote atherothrombosis, their role in other conditions is controversial. Sulfatides have been contradictorily reported to show anticoagulant properties, including prolongation of the clotting time and bleeding time, probably disturbing fibrin formation by binding to fibrinogen [2, 3, 22, 23]. Cardiovascular disease (CVD) is a critical complication of chronic kidney disease (CKD) [11, 24]. We previously found a negative correlation between serum sulfatide levels and a high incidence of CVD in patients in the terminal stage of CKD, with end-stage renal failure (ESRF), who were treated with hemodialysis [25]. Such patients were excluded from this study. CVD in CKD comprises atheromatous and non-atheromatous diseases (heart failure, cardiac dysrhythmia, valvular diseases, etc.) [26]. Additionally, the pathological condition of patients with CVD and a terminal stage of CKD, especially in those with ESRF treated by hemodialysis, is different from that in patients in our study [27]. In fact, serum lipid quality and arteriosclerosis in patients with ESRF is distinct from those in patients with typical cardiac atherosclerosis without CKD [10, 27,28,29]. Therefore, serum sulfatides may be incorporated differently in lipoproteins in patients with ESRF, and their thrombotic activity may be modified. These issues remain to be clarified in further studies.

Conclusion

On the basis of our finding of changes in serum sulfatide levels throughout the PCI procedure, we speculate that sulfatides are rapidly released from ruptured atheromatous plaques, probably accelerating atherothrombogenesis. Sulfatides are candidate molecules targeted for exploration, both for their thrombotic as well as their atherosclerotic properties.