Author + information
- Received May 2, 2018
- Revision received June 14, 2018
- Accepted June 25, 2018
- Published online September 17, 2018.
- Daniela Pedicino, MD, PhDa,b,∗,
- Ramona Vinci, PhDa,b,∗,
- Ada F. Giglio, MDa,b,
- Eugenia Pisano, PhDa,b,
- Italo Porto, MD, PhDa,b,
- Rocco Vergallo, MD, PhDa,b,
- Giulio Russo, MD, PhDa,b,
- Aureliano Ruggio, MDa,b,
- Alessia D’Aiello, MDa,b,
- Davide Flego, PhDa,b,
- Gianmarco Annibali, MDa,b,
- Francesco Trotta, MDa,b,
- Roberto Piacentini, PhDc,
- Giampaolo Niccoli, MD, PhDa,b,
- Giovanna Liuzzo, MD, PhDa,b,∗,† ( and )
- Filippo Crea, MDa,b,†
- aDepartment of Cardiovascular and Thoracic Sciences, Catholic University of the Sacred Heart, Rome, Italy
- bFondazione Policlinico Universitario A. Gemelli, IRCCS, Rome, Italy
- cInstitute of Human Physiology, Catholic University of the Sacred Heart, IRCCS-Fondazione Policlinico Universitario A. Gemelli, Rome, Italy
- ↵∗Address for correspondence:
Dr. Giovanna Liuzzo, Department of Cardiovascular and Thoracic Sciences, Catholic University of the Sacred Heart, Largo A. Gemelli, 1 - 00168, Rome, Italy.
Background Superficial erosion currently causes at least one-third of acute coronary syndromes (ACS), and its incidence is increasing. Yet, the underlying mechanisms in humans are still largely unknown.
Objectives The authors sought to assess the role of hyaluronan (HA) metabolism in ACS.
Methods Peripheral blood mononuclear cells were collected from ACS (n = 66), stable angina (SA) (n = 55), and control (CTRL) patients (n = 45). The authors evaluated: 1) gene expression of hyaluronidase 2 (HYAL2) (enzyme degrading high-molecular-weight HA to its proinflammatory 20-kDa isoform) and of CD44v1, CD44v4, and CD44v6 splicing variants of HA receptor; and 2) HYAL2 and CD44 protein expression. Moreover, they compared HYAL2 and CD44 gene expression in ACS patients with plaque erosion (intact fibrous cap and thrombus) and in ACS patients with plaque rupture, identified by optical coherence tomography analysis.
Results Gene expression of HYAL2, CD44v1, and CD44v6 were significantly higher in ACS as compared with SA (p = 0.003, p < 0.001, and p = 0.033, respectively) and CTRL subjects (p < 0.001, p < 0.001, and p = 0.009, respectively). HYAL2 protein expression was significantly higher in ACS than in SA (p = 0.017) and CTRL (p = 0.032), whereas no differences were found in CD44 protein expression. HYAL2 and CD44v6 gene expression was significantly higher in patients with plaque erosion than in those with plaque rupture (p = 0.015 and p = 0.029, respectively).
Conclusions HYAL2 and CD44v6 splicing variants seem to play an important role in ACS, in particular when associated with plaque erosion. After further validation, HYAL2 might represent a potentially useful biomarker for the noninvasive identification of this mechanism of coronary instability.
The causes of acute coronary syndromes (ACS) are multiple and incompletely understood (1,2). A sizable proportion of patients presenting with ACS has obstructive atherosclerosis and systemic evidence of inflammation, supporting the hypothesis that an inflammatory outburst may lead to plaque rupture with thrombosis (3–5). However, >40% of patients with ACS have low levels of systemic markers of inflammation (6), suggesting that other mechanisms play a pathogenic role (1,7). In particular, in the era of statin treatment, we are facing a gradual shift in the histological features of unstable plaque, from rupture to erosion (8,9).
Plaque erosion is associated with ACS in at least one-third of patients, and it is characterized by thrombus formation at the site of a de-endothelialized atherosclerotic plaque. The in vivo detection of eroded plaques is based on optical coherence tomography (OCT) assessment, which shows coronary thrombosis in the absence of plaque rupture (10–13). Previous histological studies documented intense immunostaining positivity for hyaluronan (HA) and its receptor CD44 along the interface between the endothelium and luminal thrombus in eroded plaque, and abundance of myeloperoxidase (MPO)-positive cells within the overlying thrombus (14). Accordingly, ACS patients with plaque erosion on OCT exhibit higher MPO levels in peripheral blood as compared with those showing plaque rupture (15). Moreover, coculture of polymorphonuclear leukocytes (PMNs) with endothelial cells induces endothelial injury and apoptosis, at least in part mediated by toll-like receptor-2 (TLR2) activation and flow perturbation (8,16,17). Of note, HA is an endogenous activator for TLR2 (18). Systematic OCT studies documented a higher prevalence of plaque erosion associated with non–ST-segment elevation ACS (about 60%), as compared with ST-segment elevation myocardial infarction (STEMI) that conversely shows a prevalence of plaque rupture around 70% (19,20). The higher prevalence of eroded plaques occurs in women under 50 years of age and in current smokers, whereas other coronary risk factors such as diabetes, hypertension, dyslipidemia, and chronic kidney disease are less common. Furthermore, plaque erosion frequently involves the left anterior descending coronary artery and coronary bifurcations, probably due to local flow perturbation (21,22).
HA is one of the major constituents of the extracellular matrix and, under physiological condition, consists of a large glycosaminoglycan chain that undergoes fragmentation during infection and tissue damage. At homeostasis, extracellular HA is found predominantly in its high-molecular-weight form (HMW-HA) of over 1,000 kDa, and its production by stromal cells is balanced by its cellular uptake and degradation (23). During local inflammation, HA fragments of low molecular weight (LMW-HA) accumulate extracellularly and propagate the inflammatory response, whereas the restoration of HMW-HA is associated with the resolution of the inflammatory burst (24). Interestingly, fragments derived from HA metabolism are involved in PMN recruitment and activation (25,26).
CD44 is the major cell-surface HA binding protein (27,28). It is a polymorphic type I transmembrane glycoprotein encoded by a single gene, but expressed in over 20 isoforms as a result of cell-specific glycosylation and alternative splicing of at least 10 variable exons. The expression of variant isoforms is mostly detectable in hemopoietic cells, particularly in peripheral blood mononuclear cells (PBMCs) and in lymph nodes (29). Under homeostatic conditions, immune cells barely bind HA. Activation of immune cells induces HA binding by CD44 in a process that is usually accompanied by an increase in CD44 expression. Most of the functions ascribed to CD44 can be attributed to its ability to bind and internalize HA (30,31).
Cellular HA synthases (HAS-1, -2, and -3) and hyaluronidases (HYAL1, -2, -3, and 4, PH-20, and HYALP1) mediate the turnover of HA (32). During inflammation, an increase in HA is paralleled by a decrease in chain length (below 500 kDa), possibly due to modulation of HAS and HYAL activities (33) or to cleavage induced by locally released reactive oxygen and nitrogen species (34). The cellular uptake of HA, mostly mediated by macrophages, stems from CD44–HA binding and partial degradation by HYAL2 on the cell surface, with subsequent internalization (28,35). The HYAL2 enzyme hydrolyzes HMW-HA to intermediate-sized HA fragments of approximately 20 kDa, which are further hydrolyzed to small oligosaccharides by other hyaluronidases (mainly HYAL1). Interestingly, both HYAL1 and HYAL2 contribute to catabolism of HA in a CD44-dependent manner (36). Recent studies showed that HYAL2 protein expression on endothelial cells depends at least in part from shear stress and that low shear stress–induced endothelial impairment is partly mediated by increased HYAL2 activity (37). Moreover, by the analysis of carotid specimens of patients undergoing endarterectomy, a correlation has been found between plaque instability and both HYAL2 activity and CD44v3 (a splicing variant of CD44) expression (38).
In this study, we sought to investigate the role of HA, HYAL2, and CD44 in the pathogenesis of ACS and their possible implications for plaque erosion.
The Online Appendix includes a detailed description of patient characteristics and methods. We enrolled 3 groups of patients: 1) 66 patients with a first diagnosis of ACS, admitted to our coronary care unit with non-STEMI; 2) 55 chronic stable angina (SA) patients; and 3) 45 control patients (CTRL) without overt coronary artery disease screened in our out-patients clinic for cardiovascular prevention (Figure 1).
Cellular and molecular analyses were performed on isolated PBMCs, under basal condition and after 16 h of incubation with HMW-HA, LMW-HA (R&D Systems by Lifecore Biomedical, Chaska, Minnesota) and with Escherichia coli lipopolysaccharide (LPS) (Sigma-Aldrich, St. Louis, Missouri), as positive control. The following parameters were then assessed before and after stimulation: 1) mRNA expression of HYAL2 (26 ACS, 20 SA, and 15 CTRL) by real-time–quantitative polymerase chain reaction (RT-qPCR); 2) mRNA expression of 3 different CD44 splicing variants, CD44v1 and CD44v6 (22 ACS, 20 SA, and 20 CTRL), and CD44v4 (10 for each group) by RT-qPCR; 3) HYAL2 (25 ACS, 25 SA, and 20 CTRL) and total CD44 (20 for each group) protein expression before and after stimulation (10 for each group) on monocytes by flow cytometry; 4) CD44v6 basal protein expression in ACS and SA (5 for each group); 5) platelet–monocyte binding by flow cytometry in 5 ACS and 5 SA patients, after 16 h of coincubation, with and without LMW-HA stimulation; and 6) moreover, HYAL2 and CD44 protein colocalization, before and after stimulation, was analyzed by confocal immunofluorescence microscopy in ACS and SA patients (3 for each group). Finally, in 15 of 66 ACS patients who underwent OCT analysis of culprit coronary plaques before stent implantation, it was possible to clearly identify the feature of the culprit plaques as plaque erosion characterized by an intact fibrous cap (IFC) with thrombus (6 patients) or plaque rupture containing a large necrotic core under a ruptured fibrous cap (RFC) (9 patients). We compared these 2 ACS subgroups for HYAL2 and CD44 alternative splicing variants gene expression. After a mean follow-up period of 12 ± 4 months, we reassessed 13 patients in order to investigate HYAL2 and CD44 gene expression over time.
Continuous variables that were normally distributed as assessed by Shapiro Wilk test were described as mean ± SD. In this case, 1-way analysis of variance (ANOVA) and 1-ANOVA for repeated measures, with Bonferroni correction, were used for multiple comparisons. Data that did not follow a normal distribution were described as median and interquartile range (IQR), and they were analyzed by using a nonparametric test (the Kruskal-Wallis test and the Dunn’s multiple pairwise comparisons for comparisons among groups, and the Friedman test for pairwise comparison within a group). Whenever the overall comparison between the 3 groups had a p value <0.01, we performed unpaired Student's t-test (for continuous variables that were normally distributed) or Mann-Whitney test (for nonparametric variables) to compare 2 groups, as appropriate. We used the nonparametric Spearman rank test for correlation analyses. For all the experimental assays performed, a 2-tailed p value ≤0.01 was considered statistically significant. Statistical analyses were performed with GraphPad Prism version 5.00 for Windows (GraphPad Software, La Jolla, California) and with SPSS software v22.0 (IBM, Armonk, New York).
Baseline characteristics of patient population are detailed in Table 1. The flow chart in Figure 1 summarizes the patient enrollment for each experiment, and the Online Appendix includes additional figures.
Expression of HYAL2 in basal condition and after stimulation with HMW-HA, LMW-HA, and LPS
HYAL2 gene was overexpressed in unstimulated PBMCs of ACS patients (median 8.3; IQR: 10.2, relative expression) as compared to SA patients (median 4.9; IQR: 4.2) and CTRL (median 3.6; IQR: 5.0) (ANOVA for trend p = 0.0002; p = 0.003 and p = 0.0003 ACS vs. SA and CTRL, respectively) (Figure 2A). Online Figure 1 shows receiver-operating characteristic curves for HYAL2 gene expression.
After 16-h stimulation, HYAL2 gene expression was higher in ACS patients (median 16.9; IQR: 18.3, relative expression) as compared with SA patients (median 2.9; IQR: 2.0) and CTRL (median 2.7; IQR: 3.9) (ANOVA for trend p = 0.011; p = 0.045 and p = 0.048 ACS vs. SA and CTRL, respectively) for LMW-HA stimulation (Figure 2B). Interestingly, HYAL2 gene expression after stimulation with HMW-HA was similar in the 3 groups.
In keeping with the results of stimulation, at flow cytometry, unstimulated CD14+ monocytes from ACS patients showed a higher HYAL2 protein expression (median mean fluorescence index [MFI] 2.2; IQR: 2.1) as compared with SA patients (median MFI 0.6; IQR: 1.1) and CTRL (median MFI 1.0; IQR: 0.8) (ANOVA for trend p = 0.002; p = 0.032 and p = 0.017 ACS vs. SA and CTRL, respectively) (Figure 3A). These data were further confirmed in 5 ACS and 5 SA samples by using a different HYAL2 polyclonal antibody for flow cytometry (p = 0.017) (Online Figure 2). After 16 h of stimulation, HYAL2 protein expression did not differ among groups (ANOVA for trend p = 0.02), although there was a significant increase in LPS-stimulated cells of ACS as compared with HMW-HA (Figure 3B).
In order to verify whether an increased HA catabolism in the basal condition was related also to an increased HA production, we tested the basal gene expression of hyaluronan synthase-3 (HAS3) in unstimulated PBMCs. No differences were observed among the groups (median for ACS 2.7; IQR: 1.6; for SA 2.2; IQR: 2.3; for CTRL 4.8; IQR: 2.6; ANOVA for trend p = 0.05) (Online Appendix, Online Figure 3).
Finally, we analyzed serum levels of soluble HYAL2 by enzyme-linked immunosorbent assay. No differences were observed among the 3 groups (ANOVA for trend p = 0.79) (Online Figure 4).
Gene expression of HYAL2 and cardiovascular risk factors: Role of cigarette smoking
Among the patients tested for HYAL2 gene expression, HYAL-2 was significantly higher in smoker ACS patients (n = 10) (median 33.1; IQR: 44.9) compared with nonsmoker ACS patients (n = 16) (median 7.5; IQR: 4.0; p = 0.0004), to smoker SA patients (n = 7) (median 7.0; IQR: 7.8; p = 0.001) and smoker CTRL (n = 6) (median 5.2; IQR: 3.5; p = 0.03) (ANOVA for trend p = 0.001) (Figure 4A). HYAL2 protein expression showed a similar behavior (median MFI for smoker ACS 3.4; IQR: 1.7, for nonsmokers ACS 1.1; IQR: 1.0; for smoker SA 0.9; IQR: 0.9; for smoker CTRL 0.7; IQR: 0.6; ANOVA for trend 0.008) (Figure 4B). After revising other common cardiovascular risk factors among groups, HYAL2 gene expressions was not related to obesity, dyslipidemia, diabetes, or positive family history.
Expression of CD44v1, CD44v6, and CD44v4 alternative splicing variants in basal conditions and after incubation with LMW-HA, HMW-HA, and LPS
We investigated gene expression of CD44 and transcript variants 1 (CD44v1), 6 (CD44v6), and 4 (CD44v4) in unstimulated PBMCs. CD44v1 gene expression was higher in ACS patients (median 29.5; IQR: 42.6, relative expression) compared with SA patients (median 4.2; IQR: 5.7) and CTRL (median 9.9; IQR: 6.0) (ANOVA for trend p = 0.0003; p < 0.0001 and p = 0.0003 ACS vs. SA and CTRL, respectively) (Figure 5A). CD44v6 gene expression also was significantly higher in ACS patients (median 12.4; IQR: 8.4, relative expression) as compared with SA patients (median 4.2; IQR: 12.5) and CTRL (median 7.7; IQR: 4.9) (ANOVA for trend p = 0.009; p = 0.033 and p = 0.009 ACS vs. SA and CTRL, respectively) (Figure 5B). In sharp contrast, CD44v4 gene expression was higher in SA patients (median 4.5 IQR: 3.4, relative expression) compared with both CTRL subjects (median 1.7; IQR: 0.8) and ACS patients (median 2.3; IQR: 1.9) (ANOVA for trend p = 0.003; p = 0.005 and p = 0.043 SA vs. CTRL and ACS, respectively) (Figure 5C). No significant differences were found for CD44v1, v6, and v4 splicing variants after 16-h stimulation with LMW-HA, HMW-HA, and LPS (Online Figure 5).
Of note, a positive correlation was observed between basal gene expression of HYAL2 and of CD44v6 in ACS patients (r = 0.63; p = 0.016) (Figure 5D), although no correlation was found between HYAL2 and CD44v1, and between HYAL2 and CD44v4 (Online Figure 6).
CD44 protein expression was assessed on the surface of CD14+ monocytes by flow cytometry, using a CD44 antibody clone that recognizes an epitope common to the different isoforms of this molecule, in order to include in the analysis all the splicing variants. Moreover, in 5 ACS and in 5 SA patients a monoclonal antibody corresponding to the splicing variant 6 was tested. In basal conditions, no difference was found, neither in total CD44 protein expression comparing ACS (median MFI 35.1; IQR: 5.2) to SA patients (median 36.5; IQR: 8.6) and to CTRL (median 36.8; IQR: 7.3) (ANOVA for trend p = 0.49) (Figure 6A), nor in CD44v6 protein expression (Online Figure 7). After 16 h of stimulation, in ACS patients only, total CD44 protein expression was significantly increased in LPS-treated samples (MFI fold increase: median 1.17; IQR: 0.22) as compared with HMW-HA (median 0.96; IQR: 0.08) and LMW-HA (median 0.94; IQR: 0.08; ANOVA for trend p = 0.001; p = 0.004 and p = 0.009 LPS vs. HMW-HA and LMW-HA, respectively) (Figure 6B). After revising all the common cardiovascular risk factors among groups, CD44 splicing variant gene expressions were not related to smoking (Online Figure 8), obesity, dyslipidemia, diabetes, or family history.
Online Figure 1 shows receiver-operating characteristic curves for CD44v1 and CD44v6 gene expression.
Effect of HA on platelet–monocyte binding in ACS
We compared platelet–monocyte binding (expressed as % of CD14/CD42b-positive cells) by flow cytometry in 5 ACS and 5 SA patients after 16 h of coincubation, with and without LMW-HA stimulation. Despite the low number of patients, our data showed an increase in platelet–monocyte binding after LMW-HA treatment in patients presenting with ACS (median MFI fold increase 31.42; IQR: 31.08) as compared with SA (median 3.0; IQR: 3.19) (Figure 7). On confocal microscopy analyses, we found a trend of increased CD44 and HYAL2 protein expression in ACS patients after stimulation (Online Figure 9). Interestingly, although CD44 and HYAL2 colocalized on cell membranes in SA patients also after stimulation, in ACS, this colocalization was lost, thus suggesting an intense alteration of the CD44/HYAL2 pathway due to a failure of synergism (Figure 8).
HYAL2, CD44v6, and CD44v1 gene expression according to OCT analysis of the unstable plaques: Acute phase and follow-up
HYAL2 and CD44v6 gene expression were significantly higher in ACS patients with plaque erosion (IFC group) as compared with those with plaque rupture (RFC group) (for HYAL2: median 15.2; IQR: 17.6 vs. median 7.3; IQR: 4.9; p = 0.015; for CD44v6: median 17.5; IQR: 35.3 vs. median 11.0; IQR: 2.3; p = 0.029) (Figures 9A and 9B). In sharp contrast, CD44v1 gene expression was higher in RFC group as compared with the IFC group (for CD44v1: median 48.6; IQR: 59.5 vs. median 10.2; IQR: 5.5; p = 0.029) (Figure 9C).
Of note, after 12 ± 4 months of follow-up in which all patients received similar pharmacological treatments, HYAL2 and CD44v6 gene expression dramatically decreased in patients with plaque erosion (IFC group) as compared with the same patients analyzed during the acute phase, with expression levels becoming similar to those observed in the RFC group. Thus, in the IFC patient subset, median HYAL2 expression was 2.0; IQR: 0.7 at follow-up (p = 0.004; follow-up vs. baseline). In the same IFC patient subset, median CD44v6 expression was 5.2; IQR: 1.6 at follow-up (p = 0.004 follow-up vs. baseline) (Figures 9A and 9B). Similarly, a significant decrease over time was observed for CD44v1 gene expression in patients with plaque rupture (RFC median 2.8; IQR: 2.2; p = 0.016 follow-up vs. baseline) (Figure 9C). See also Online Table 1 for raw data and Online Figure 10.
Finally, even though the number of patients examined was very small, and we could not perform any statistical analysis, we found the highest HYAL2 gene expression in smoker ACS patients with IFC on OCT (Online Figure 11).
Plaque erosion is associated with ACS in at least one-third of patients. Recent studies documented a higher prevalence of plaque erosion in non-STEMI, in women under 50 years of age, and in current smokers (21,22). Experimental studies proposed an exciting model of plaque instability associated with erosion, where blood flow perturbation determines endothelial activation by HA-induced TLR2 stimulation leading to de-endothelialization, PMN recruitment and thrombus formation (8,9,16). These data, together with the notion of an intense immunostaining for CD44 on eroded plaques and for MPO on the overlying thrombus at histological examination of coronary lesions of sudden coronary death victims, led to the intriguing hypothesis that the accumulation of HA along the plaque may be involved, through CD44 receptor engagement, in this pathogenic mechanism of ACS (10,13,14). Adjunctive experimental evidence recently highlighted the role of HYAL2 as a possible mediator of endothelial dysfunction in the presence of perturbed flow (8,37). This enzyme might be involved also in the generation of a local inflammatory response, given its ability to cut HMW-HA in proinflammatory fragments of 20 kDa, that are involved inter alia in PMN chemotaxis and activation (25).
HYAL2 expression in ACS: Role of HA pathway in plaque erosion
In our study, for the first time in the clinical setting, we show an altered HA metabolism in ACS patients, characterized by enhanced HYAL2 gene expression in circulating PBMCs that is involved in the degradation of HMW-HA to 20-kDa LMW-HA fragments. Indeed, in ACS patients HYAL2 gene expression was significantly higher when compared with that observed in SA patients and in CTRL in the basal condition; furthermore, in ACS patients, but not in the 2 other groups, HYAL2 gene expression significantly increased after PBMC incubation with LMW-HA to levels that paralleled those observed with a potent proinflammatory stimulus (E. coli LPS). Enhanced HYAL2 expression was prevalent in patients presenting with ACS and OCT evidence of IFC. Of note, we found that HYAL2 gene expression was significantly enhanced in smoker versus nonsmoker ACS patients, whereas no difference was found between smokers and nonsmokers among SA patients and CTRL subjects. Even more interestingly, we found that smoker ACS patients presenting with IFC on OCT analysis had the highest HYAL2 gene expression and that also nonsmoker ACS patients with IFC showed higher levels than RFC (smokers or nonsmokers) (Online Appendix). Although these are very preliminary data, due to the small number of patients, they suggest hyper-reactivity to this environmental factor during the acute phase of coronary artery disease, in keeping with post mortem data suggesting that smoking is a risk factor for erosion (22).
In the general population, HYAL2 protein expression, assessed by flow-cytometry, paralleled what was observed for gene expression by RT-qPCR mRNA assessment.
CD44 receptor expression in ACS: Differential expression of different splicing variants
Gene expression analysis of CD44 splicing variants revealed significantly higher levels of both CD44v1 and CD44v6 mRNA in ACS patients as compared with SA patients and CTRL subjects. Notably, patients with IFC showed higher gene expression of CD44v6, with a direct correlation between gene expression of this splicing variant and that of HYAL2. The differences observed in CD44v1, CD44v4, and CD44v6 gene expression between SA and ACS, and between IFC and RFC demonstrated that different isoforms of the CD44 receptor might be variously involved in different types of unstable plaques, with a specific commitment of CD44v6 in the HA metabolic pathway linked to plaque erosion. Of note, and in accord with our findings, previous experimental work has already documented a higher expression of other variants of CD44 receptor in unstable plaques (38). We did not find any difference in CD44 protein expression in circulating monocytes among ACS, SA, and CTRL under basal conditions. Yet, after incubation with LPS, CD44 protein levels increased in PBMCs from ACS patients, but not in those from SA patients, thus suggesting that enhanced CD44 gene expression translates into increased protein expression only under conditions of local inflammation, as might be the case in eroded plaques.
The data acquired at follow-up for both gene expression of HYAL2 and CD44 splicing variants documented a significant reduction of HYAL2 and CD44v6 expression in the IFC group and a significant reduction of CD44v1 for the RFC group, thus confirming that the alterations observed in our study were strictly limited to the acute phase of ACS (39,40).
Our data on CD44 splicing variants seem to suggest that what we observed in PBMCs might mirror what happens in endothelial cells at the site of the eroded plaque where enhanced expression of specific isoforms of the CD44 receptor might amplify HA-induced endothelial activation.
Taken together, these findings suggest that in plaque erosion, the enhanced expression of HYAL2 in PBMCs might provide a positive feedback promoting local LMW-HA formation, TLR2 stimulation, endothelial activation, and neutrophil accumulation, also enhanced by CD44 overexpression in neutrophils, eventually leading to thrombus formation. The latter might be also favored by LMW-HA accumulation resulting in enhanced monocyte-platelet binding (Central Illustration). This model derives from both the existing experimental and post-mortem studies on plaque erosion and the data emerging from our clinical study.
First of all, our study was a prospective observational analysis that included a limited number of patients. No power calculation could be performed because of a lack of previous studies in this setting; thus, the enrollment of 66, 55, and 45 subjects for each group and their allocation in the different experiments performed were arbitrary, and they were mainly driven by the number of available PBMCs. Moreover, ACS and SA patients and controls were not properly matched for age, sex, and risk factors; however, no significant differences were observed in this regard between the ACS and SA groups, and none of these variables were independently associated with HYAL2 and CD44 expression on univariate analysis, with the exception of smoking and HYAL2 gene expression. These limitations imply 2 dominant methodological issues that cannot be eluded. First, several variables other than the coronary disease state might explain the differences observed across these 3 populations. Thus, we cannot exclude the likelihood that some of our findings of statistical significance might be false positives. Second, it is impossible in this type of study to determine a cause–effect relationship. More specifically, for obvious limitations, in this study, we analyzed PBMCs and we did not collect data on the molecular composition of the culprit plaque. Thus, we cannot say whether these alterations might have a causal role or whether they need to be considered as a consequence of the plaque fate. Finally, the greater limitation in our study is the lack of OCT imaging analysis for all the patients enrolled, because we obtained intracoronary imaging only for clinical reasons.
In our study, for the first time to our knowledge, we describe different molecular signatures in vivo in humans connecting a similar clinical presentation (ACS) with culprit plaques that show different pathological features. This paves the way toward new immunomodulatory targets in ACS that might be of diagnostic and prognostic value by using specific biomarkers for a specific plaque phenotype, and that might be crucial in targeting new, personalized therapeutic approaches. Indeed, HA metabolism alteration could represent a novel therapeutic target in patients in whom ACS is caused by plaque erosion. More importantly, if confirmed in more robust and larger studies, HYAL2 might represent a noninvasive diagnostic tool allowing one to avoid the use of OCT imaging, which is quite expensive, time consuming, and burdened with risks. This might be even more valuable in centers in which OCT technology is not available, paving the path toward the noninvasive characterization of the mechanism of plaque instability in ACS patients (41).
COMPETENCY IN MEDICAL KNOWLEDGE: OCT studies of the pathogenesis of ACS have identified changes in unstable plaques that include but are not limited to rupture. Molecular pathways leading to plaque erosion involve alterations in metabolism of HA and its receptor, CD44, at the interface between the coronary endothelium and intraluminal thrombus.
TRANSLATIONAL OUTLOOK: In the future, assays for HA metabolites and detection of specific splicing variants of CD44 may allow for non-invasive identification of plaque erosion and guide diagnostic algorithms and targeted management strategies for patients with ACS.
The authors thank the catheterization laboratory director Prof. C. Trani and all the colleagues, technicians, and nurses for their technical and scientific support. The authors strongly believe in translational research involving clinicians, interventional cardiologists, and molecular biologists to promote personalized medicine.
↵∗ Drs. Pedicino and Vinci contributed equally to this work as first authors.
↵† Drs. Liuzzo and Crea contributed equally to this work as senior authors.
This work was supported by the Catholic University of the Sacred Heart, Rome, Italy (R4124500458 LINEA D.1 2016) and IRCCS-Fondazione Policlinico Universitario “A. Gemelli.” The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute coronary syndrome
- analysis of variance
- control patients
- high-molecular-weight hyaluronan
- intact fibrous cap
- interquartile range
- low-molecular-weight hyaluronan
- mean fluorescence index
- optical coherence tomography
- peripheral blood mononuclear cell
- polimorphonuclear leukocyte
- ruptured fibrous cap
- real-time–quantitative polymerase chain reaction
- stable angina
- ST-segment elevation myocardial infarction
- Received May 2, 2018.
- Revision received June 14, 2018.
- Accepted June 25, 2018.
- 2018 American College of Cardiology Foundation
- Crea F.,
- Liuzzo G.
- Libby P.,
- Hansson G.K.
- Cristell N.,
- Cianflone D.,
- Durante A.,
- et al.,
- FAMI Study Investigators
- Crea F.,
- Libby P.
- Franck G.,
- Mawson T.,
- Sausen G.,
- et al.
- Bentzon J.F.,
- Falk E.
- Virmani R.,
- Kolodgie F.D.,
- Burke A.P.,
- Farb A.,
- Schwartz S.M.
- Prati F.,
- Uemura S.,
- Souteyrand G.,
- et al.
- Jia H.,
- Abtahian F.,
- Aguirre A.D.,
- et al.
- Kolodgie F.D.,
- Burke A.P.,
- Farb A.,
- et al.
- Ferrante G.,
- Nakano M.,
- Prati F.,
- et al.
- Mullick A.E.,
- Soldau K.,
- Kiosses W.B.,
- Bell T.A. III.,
- Tobias P.S.,
- Curtiss L.K.
- Kubo T.,
- Imanishi T.,
- Takarada S.,
- et al.
- Chandran S.,
- Watkins J.,
- Abdul-Aziz A.,
- et al.
- Dai J.,
- Xing L.,
- Jia H.,
- et al.
- Jinnouchi H.,
- Virmani R.,
- Finn A.V.
- Lu C.H.,
- Lin C.H.,
- Li K.J.,
- et al.
- Teder P.,
- Vandivier R.W.,
- Jiang D.,
- et al.
- Salles G.,
- Zain M.,
- Jiang W.M.,
- Boussiotis V.A.,
- Shipp M.A.
- Huebener P.,
- Abou-Khamis T.,
- Zymek P.,
- et al.
- Harada H.,
- Takahashi M.
- Kong X.,
- Chen L.,
- Ye P.,
- et al.
- Flego D.,
- Liuzzo G.,
- Weyand C.M.,
- Crea F.
- Partida R.A.,
- Libby P.,
- Crea F.,
- Jang I.K.