Introduction

According to the ACCF/AHA statement, they define heart failure as “a complex clinical syndrome that results from any structural or functional impairment of ventricular filling or ejection of blood” [1]. It has been estimated that 5.7 million Americans aged over 18 years have heart failure (NHLBI tabulation). Heidenreich et al. in their study drew the conclusion that HF rate will increase by 46 % over a time stretch from 2012 to 2030; thus, talking numbers, more than 8 million people with age greater than 18 years will incur in HF [2, 3]. Following these percentages in a survey run in 2009 in the USA, the risk of developing HF for people over 40 years of age is 20 % [4]. Two years later, according to what NCHS and NHLBI declared in 2011, one death out nine has heart failure mentioned on the death certificate. Furthermore, despite the outstanding progresses achieved in HF treatment, the morbidity, the mortality and the medical costs still remain elevated; thus, researchers are even more motivated to further dig into the subject in order to lower the overall expense for HF treatments. In 2012, total cost for HF has been estimated to be $30.7 billion. Out of this, 68 % was attributable to direct medical costs; projections show that by 2030, the total cost of HF will increase almost of 127 % to $69.7 billion from 2012. This equals ≈$244 for each US adult [3].

These provisions tell us that during the next 15 years we will not encounter diminishing costs related to heart failure treatments as we would expect (technological advance and new discoveries).

Different factors could induce HF; life style, food and geographical areas are demonstrated to influence the development of this disease [5].

HF severely worsens patient life quality by recurring to frequent hospitalizations. This condition is the end stage of many compensative mechanisms due to the inability of the heart itself to maintain the necessary blood pressure for organ perfusion. These mechanisms lead to a vicious cycle resulting in maladaptive structural and functional modifies in failing hearts [1, 6, 7].

S100 proteins appear to be involved in a large number of cellular activities including signal transduction, cell differentiation, regulation of cell motility, transcription and cell cycle progression [8]. Apart from these intracellular functions, some S100 proteins can be secreted from cells and exhibit cytokine-like extracellular functions. These include chemotactic activities related to inflammation [9, 10].

The incoming of contractility dysfunction and the presence of a pro-inflammatory state lead to a weakened cardiac performance. These are the basis of the impairment of failing heart. Since S100 proteins play a crucial role in these processes, these alarmins have been object of several studies to assess their prognostic and therapeutic value in heart failure.

The aim of this systematic review is to examine latest advances on S100 proteins related to heart failure. On the basis of last years’ findings which were a combination of tests conducted on both animals and humans, our group focused on a pending potential clinical application of this family of proteins. For this purpose, we reviewed studies published during a time period which starts in 2008 until May 2015 involving S100 family proteins in animals and humans failing hearts.

Many serum markers were identified as risk factors for incident in heart failure: BNP, urinary albumin-to-creatinine ratio, elevated serum γ-glutamyltransferase and higher levels of haematocrit [11–13]. Also increased circulating concentrations of resistin were associated with incident HF independent of prevalent coronary disease, obesity, insulin resistance and inflammation [14]. What Djousse et al. [5] found was that lifestyle factors (normal weight, not smoking, regular exercise, moderate alcohol intake, consumption of breakfast cereals and consumption of fruits and vegetables) were related to a lower risk of HF. On the other hand, inflammatory markers such as IL-6 and TNF-alpha, serum albumin levels and cigarette smoking exposure were also associated with HF risk [15–17].

Our purpose is to lower those costs, of course assuming that our first priority is helping people.

In last advances, the attention focused on the sarcoplasmatic reticulum and on the Ca handling process; this resulted in giving a main role to this intracellular organelle in cardiomyocytes [18].

The Ca2+ sensor protein S100A1 has emerged as an attractive target for genetically targeted HF therapy in various in vivo HF models because of its molecular profile [19]. As shown in Fig. 1, S100A1 regulates a network in cardiomyocytes controlling sarcoplasmic reticulum Ca2+ cycling and mitochondrial function through the interaction with the ryanodine receptor (RyR2), sarcoplasmic reticulum Ca2+-ATPase (SERCA2) and mitochondrial F1-ATPase activity, causing anti-hypertrophic, positive inotrope and anti-arrhythmic effects and reducing energy depletion in HF [20–26]. In addition, S100A1 in myofilaments facilitates diastolic Ca2+ dissociation in order to ameliorate relaxation during diastole [27]. Importantly, the S100A1 protein is significantly down-regulated in human end-stage HF, rendering S100A1 an appropriate target for cardiac gene therapy [28, 29].

Fig. 1
figure 1

S100A1 action sites in cardiomyocytes. a Interaction of S100A1 with RyR2 and SERCA. These interactions modulate calcium flux in sarcoplasmic reticulum. b S100A1 in myofilaments facilitates diastolic Ca2+ dissociation in order to ameliorate relaxation during diastole. c S100A1 increases mitochondrial ATP retrieval during augmented energy request

Full size image

In cardiac muscles, S100A1 is the predominant S100 family member, while S100A4, S100A6 and S100B are also expressed with much less abundance [30]. These other S100 proteins family’s members have been studied during last decades either for prognostic or for therapy purpose.

We describe here the protocol of a systematic review to analyse latest advances on S100 Proteins role, utilization and potential application in the heart failure disease. For this purpose, we reviewed experimental studies that investigated these proteins as markers as well as potential treatments in failing hearts. This systematic review has been registered with PROSPERO (the NIHR International Prospective Register of Systematic Reviews) under registration number CRD42015027932.

Experimental design

Research strategy

This systematic review has been conducted according to PRISMA guidelines [31] employing PubMed and ScienceDirect databases. On these websites, we searched for articles from January 2008 through May 2015 using two key terms related to heart diseases, “heart” or “heart failure” and two key terms related to alarmins, “S100 proteins” and “S100A1”. The electronic search strategy for PubMed is given in Table 1.

Table 1 Search terms and strategy for the systematic review
Full size table

Articles have been selected by title and abstract; the entire article was read if title/abstract indicated that the association between heart and S100 proteins was evaluated and that the article potentially met the inclusion criteria. Lastly, we reviewed and searched references of the selected articles and the ones whose titles suggested that could have researched the association between heart and S100 proteins in order to identify additional studies that met the inclusion criteria.

Study selection

Articles were included in the present review according to the following inclusion criteria: English language, publication in peer reviewed journals, year of publication at least 2008. Articles were excluded by title, abstract or full text for irrelevance to the topic in question. Further exclusion criteria were: review articles, editorial comments. Furthermore, we arbitrarily decided to start our research from 2008 to give a more recent view of heart failure findings.

Data extraction

Two authors (MC, SQ) performed the initial search and independently reviewed and selected the references based on the inclusion and exclusion criteria. The data derived from our research of articles include: study author names, publication dates, study designs (i.e. case–control, cross-sectional, longitudinal), groups studied, clinical and biological variables and outcome of interest of the study. Principal outcome of interest included studies in advanced gene therapies on animals and clinical practice prognostic advances in human models. Given the considerable heterogeneity in the study designs and subjects of the selected studies (in terms of biological and clinical variables and targets), the characteristics of the studied populations and protocols were summarized and the study outcome was reported using descriptive statistics without conducting any meta-analyses.

Results

Although S100A1 has a well-known and probably main role in HF, recent findings also suggest the involvement of other S100 proteins in this disease. Consistently, we decided to focus our systematic review on S100 family in its ensemble.

In Fig. 2, the flow of articles retrieved for the review is reported. As given in Table 2, a total of 14 studies assessing the association between S100 proteins (S100) and heart failure (HF) were identified. Out of these 14, 4 evaluated S100 studies in human patients with HF and 10 evaluated S100 studies in animals affected by HF. The search on PubMed and ScienceDirect provided a total of 46 citations (S100 e Heart). An extra amount of three articles was added by matching S100A1 and heart. Eight studies were excluded because they did not meet the criteria as they were reviews or editorial comments. Some articles (n = 27) were excluded by reading the full text because they were clearly not relevant to key purpose of this review. The remaining 14 citations were examined in a more detailed way. All these articles met the inclusion criteria and were included in the systematic review. No additional studies that met the criteria for inclusion were identified by checking the references of selected papers.

Fig. 2
figure 2

PRISMA flow diagram. Studies selected for inclusion with number of studies included after each stage of the screening process

Full size image
Table 2 Studies about S100 proteins in heart failure from January 2008 to May 2015
Full size table

Finally, 14 studies have been acknowledged as assessing the association between S100 proteins (S100) and heart failure (HF). Of these, three studies explored the potential employment of S100 as marker in HF while the other studies evaluated the potential therapeutic effects of S100 in both human and animal failing hearts. Table 1 summarizes the studies selected with a particular emphasis on the type of S100 protein in examination, the type of subjects, in vivo or in vitro dosage, the principal outcomes and potential application of the protein. The table shows literature results of the association between S100 family proteins and HF. The search gave a total of 14 results including S100 A1, A2, A4, A6, A8–A9, B protein. These studies had either a prognostic or a therapeutic purpose and were conducted either on animals or on humans.

S100 proteins in animal models

Table 3 shows a total of nine studies about S100 proteins in small and large size animals.

Table 3 Studies conducted on animals
Full size table

According to the evidence reached during the period before 2008, during last years the attention of the authors focused on a potential therapeutic effect of S100 in failing hearts.

Only one study focused on the confirmation of S100A1 as biomarker. Gupta and his collaborators evaluated the restoration of S100A1 proteins together with other HF markers in dogs with failing heart after chronic therapy with cardiac contractile modulation (CCM). The elements observed in this study confirm the role of S100A1 as a marker of therapy efficacy in HF [32].

Six studies evaluated S100A1 as a treatment by using gene therapy.

Pleger et al. in 2008 demonstrated that S100A1 could potentially improve endothelial function by the generation of NO. This result led to a potential therapeutic strategy by restoring S100A1 levels which are lessened in HF [33].

Both Most and Pleger et al. made a study aimed to determine long-term therapeutic efficacy, safety and feasibility of S100A1 gene therapy with the usage of adeno-associated viral-9 as vector in pigs affected with HF condition. After 12 weeks, animals showed improved cardiac performance and no alteration in blood levels and life parameters. Thanks to these results, they incited a human trial [34, 35].

Yamaguchi et al. proved that S100A1 prevented the development and the progression of HF in small animals. They were led to this conclusion by studying mice with RyR2 mutation that cancelled the inhibitory effect of S100A1 on the ryanodine receptor. Mice with this mutation developed heart failure more quickly than normal mice [36].

Weber and his collaborators compared vector specificity in S100A1 gene therapy. In a study comparing AAV6-S100A1 versus AAV9-S100A1 proteins treating failing pig hearts, they found that S100A1 was significantly over-expressed in pigs treated with AAV6. After these results, Weber et al. focused on treatment efficacy confirming the rescuing ability of this treatment on failing hearts. Their last target was answering authors’ concern on the safety of this kind of therapy. Data obtained confirmed the absence of side-effects such as ventricular tachyarrhythmia or impairment of contractile function. This last therapeutic study confirms once more the efficacy of S100A1 gene therapy on animals encouraging the beginning of human trials [37].

Ritteroff et al. unveiled novel mechanisms explaining the therapeutic effect of this S100 protein by an in vitro/in vivo study on rats analysing S100A1 interaction with RyR2. According to their findings, sarcoplasmic reticulum calcium leak could be recovered in failing cardiomyocytes by restoring the S100A1-RyR2 link (by gene therapy). S100A1, in fact, improved RyR2 closure during diastole allowing a better relaxing and avoiding futile SERCA2a activity compensating for the arrhythmic sarcoplasmic reticulum calcium leak [38].

Based on previous studies involving a RAGE-mediated inflammation damage in HF [39, 40], Volz et al. speculated a S100A8/A9-RAGE driven feed-forward loop. In fact, they treated both mice RAGE-knockout and normal mice with recombinant S100A8/A9. This resulted in a worsening of cardiac performance only in normal mice. According to their findings, inhibiting the ligation of S100A8/A9 with RAGE could be a potential therapeutic target [41].

Another S100 protein, the A4 was analysed by Tamaki’s group basing their study on the known role in cell proliferation and collagen expression through p53. They agreed on the fact that S100A4 modulates p53 function in cardiac fibroblasts and myocardial interstitial fibrosis through cell proliferation control and collagen production. Tamaki et al. did not explain how S100A4 modulates p53. According to them, the inhibition of p53 by blocking S100A4 may have a potential therapeutic role in heart failure by controlling cardiac fibrosis [42].

Wang and his group focused for the first time on S100A2 and A6 on rodent. Data obtained revealed that S100A2 (but not S100A6) enhances Ca2+ cycling and improves contraction and relaxation performance in normal and failing cardiomyocytes. They obtained ectopic expression of these proteins by using gene transfer [43].

S100 proteins in humans

Data obtained in animal studies during last years encouraged studies on humans. Only four studies given in Table 4 were conducted during the period analysed. Three of them aimed to find a biomarker of HF. Only one study aimed to confirm animal results of gene therapy in human cardiomyocytes.

Table 4 Studies conducted on humans
Full size table

Ma et al. speculated that S100A8 and A9 could be additional markers in elderly patient with failing hearts. A8/A9 complex, as well as IL-6 and IL-8, was found to be increased in enrolled patients (54-year-old patients with chronic HF). These findings suggested a potential use of S100A8 and A9 as pro-inflammatory biomarkers adding sensitivity to the already existing markers [44].

S100B was evaluated as a potential biomarker by Li et al. by assessing S100B protein levels in patients affected by chronic HF. They reported not only increased levels of the alarmin in patients with failing heart, but also that S100B was even more increased if they had chronic kidney disease. By analysing other disease markers such as hsCRP, TNF-alpha, NT-proBNP levels together with echocardiographic assessment, they concluded that S100B is an independent risk factor for chronic HF and for major cardiac events. Increased levels of this protein were correlated with poorer prognosis [45].

In a study published in 2014, Bennett et al. aimed to evaluate whether unloading human failing hearts by applying LVAD (left ventricular assist device) also S100A1 levels would have been recovered. This hypothesis was tested in order to have a better comprehension of the mechanism in rescuing failing hearts by this protein. Indeed, after the support therapy, S100A1 levels were not restored in these patients. According to this study, in fact, the improvement in heart function obtained by unloading the heart by application of LVAD does not correspond to an augmentation of S100A1 levels [46].

Only one study with therapeutic target was conducted on humans in vitro (ventricular cardiomyocytes obtained by failing hearts). Brinks et al. decided to evaluate the potential of gene target therapy of S100A1 in ventricular myocardium of 27 patients with chronic HF [29]. They observed that even modest changes in S100A1 protein by genetically targeted gene addition are actually therapeutic in failing human myocardium in vitro. These findings are consistent with those obtained in animal models [34, 35, 37, 47].

S100 proteins as biomarkers

Since S100A1 is highly expressed in cardiomyocytes, this protein is hypothesized to have a pivotal role in the sarcoplasmic reticulum function of the cardiomyocyte, also due to its ability to interact with the ryanodine receptor and with the SERCA/phospholamban complex [48]. Several years ago, a significant down-regulation of S100A1 protein in heart failure has been already shown [28]. According to these results, many studies aimed to focus on the level of proteins of S100 family in animals and patients with failing hearts. As given in Table 5, four studies evaluated S100 proteins as biomarkers from 2008.

Table 5 Studies evaluating S100 proteins as biomarkers
Full size table

Gupta and his collaborators evaluated S100A1 as HF marker in animals with failing hearts due to its well-known down-regulated expression in this heart disease. Elements observed in Gupta study confirm the role of S100A1 as a marker of therapy’s efficacy in HF [32]. The remaining three studies focused on S100 proteins as biomarkers of HF in humans. Ma et al. speculated that S100A8 and A9 could be additional markers in elderly patient with failing hearts. They confirmed a potential use of S100A8 and A9 as pro-inflammatory biomarkers adding sensitivity to the already existing markers [44]. Li et al. [45] noticed an increased level of S100B not only in patients with failing hearts but that it was even more increased if they had comorbidity. Bennett et al. demonstrated that LVAD therapy in patients with heart failure did not restore S100A1 levels [44]. Tamaki’s group observed that, as S100A4 modulates p53 function in myocardial interstitial fibrosis, the inhibition of p53 by blocking S100A4 could have a potential therapeutic role in heart failure by controlling cardiac fibrosis [42].

S100 proteins as potential therapy

The finding of low levels of S100A1 in animals and humans affected by heart failure encouraged to focus on potential therapies aimed to restore S100A1 levels in failing hearts. As given in Table 6, 10 studies are focused on the therapeutic role of S100 proteins in HF. Six studies evaluated S100A1 as a treatment by using gene therapy in animals. Most, Yamaguchi, Weber and Pleger in two different studies confirmed the efficacy of S100A1 gene therapy in rescuing failing hearts on animal models [33–38].

Table 6 Studies confirming the therapeutic role of S100 proteins
Full size table

Other authors decided to focus on potential treatment based on different S100 proteins. Volz et al. [41] found that inhibiting the ligation of S100A8/A9 with RAGE could improve cardiac performance. Wang et al. [43] revealed that S100A2 (but not S100A6) enhances Ca2+ cycling and improves contraction and relaxation performance in normal and failing cardiomyocytes.

Only one study with therapeutic target was conducted on humans in vitro. Brinks’ group observed that even modest increase in S100A1 by target gene therapy in failing myocardiocytes was therapeutic in vitro, consistently with data already obtained in previous animal studies [33, 34, 37, 47].

Discussion

In this review, 14 studies examining latest advances in S100 proteins related to heart failure were identified and systematically reviewed. Different S100 proteins were evaluated in different reviewed articles, due to the enormousness of the S100 proteins family. S100 proteins, in fact, constitute a sizeable family of Ca2+-binding proteins characterized by the EF-hand structural motif [9, 10]. These proteins modulate a multitude of processes, including enzyme activities, calcium handling, energy metabolism, contractility, secretion, genotype modifications [9, 49–51].

Among S100 family’s members, S100A1 is mainly situated in cardiomyocytes, while S100A4, S100A6 and S100B are less represented, but, however, reported in heart tissue [30]. On the other hand, S100A8/A9 complex has been acknowledged as a useful biomarker of inflammation [42].

S100A1 modulates cardiac Ca2+ cycling both by releasing and retaking, and it is also fundamental for the cardiac muscle contractility and in regulating mitochondrial metabolism [19, 27].

Several studies suggested that diminished levels of S100A1 correlate with heart failure severity [24, 35, 52–54], and an accelerated progression to heart failure in mice missing of S100A1 proteins was also confirmed [24, 55]; we found nine studies focused on the therapeutic role of S100A1; seven out of these evaluated S100A1 gene therapy. Six of these studies were conducted on animals, while one only on human models [29]. Four out of these studies reported that restoring S100A1 levels in animals HF models by cardiac-targeted S100A1 gene transfer improved cardiac function in the long term [24, 35, 52, 54]. Brinks et al. demonstrated that S100A1 gene therapy recovered the protein levels in human cardiomyocytes obtained from patients with severe HF undergoing heart transplantation. Obviously, the study was conducted in vitro evaluating only the cardiomyocyte microenvironment itself. Actually, there is no evidence of this kind of gene therapy in vivo on humans [29].

Bennett et al. [46] investigated S100A1 tissue levels in failing left ventricular assist device (LVAD)-supported hearts, with the aim to evaluate whether the rescuing of cardiac performance is accompanied to an increase in S100; the findings appeared to be controversial, since although several indices suggested an improvement of left ventricle performance, the expected normalization of S100A1 levels was not confirmed. This mismatch was probably due to the use of a different LVAD technology from the one used in previous studies [56].

The remaining five studies evaluated other S100 family proteins. Two of them (one in animals and the other in humans) focused on the S100A8–A9 complex [41, 44]; based on these studies, a prognostic role was given to the complex levels. As seen in many other studies, these proteins could be also increased in many inflammatory diseases [57–61].

S100B was closely associated with HF stadiation and with the severity of coexisting kidney failure by Jin Ping Li and his group. According to these data, S100B over-expression could be considered a negative prognostic marker [45].

Only two studies evaluated the therapeutic role of different S100 proteins in animals [42, 43]. S100A2-6 and S100A4 in fact appear to have a potential in failing heart treatment.

During the last decades, the attention in HF mechanisms moved from the organic and functional to the molecular approach. The review of these 14 studies over the last 7 years gave us a better comprehension of advanced HF physiopathology. Data are quite consistent. Together with macroscopic alterations, including cardiac remodelling and fibrosis, that may result in a worse cardiac performance, a microscopic damage coexists that may lead to impaired cardiac function.

According to the evidence gathered, a main role in HF is attributed to calcium handling and to sarcoplasmic reticulum, and S100A1 protein has a key role in calcium homoeostasis through this organelle [62–66]. Advanced heart failure modifies cardiomyocyte expression of S100A1, resulting in lowering S100A1 tissue levels [67]. Data from animal studies are consistent: S100A1 gene therapy recovers tissue S100A1 levels and improves cardiac performance [33, 34, 37, 47]. This suggestion has been also confirmed by in vitro study on human cardiomyocytes [29].

S100B and S100A8–A9 complex appear to have a minor clinical impact, but their levels could add more specificity to the prognosis and stadiation in HF. Moreover, study results suggest that S100 proteins family has a main role also in local damage and inflammation.

An interesting aspect concerns the potential ability of S100 proteins as biomarkers in HF. C-reactive protein (CRP) has been already widely investigated in HF, already evaluating a potential pathophysiological role; several studies reported similar findings: patients affected by HF with higher levels of CRP have a worse prognosis [68–71]. However, CRP is not a cardiac-specific biomarker, and its levels are conditioned by many causes, since it is the most clinical-adapt marker of active inflammation.

S100B protein on the contrary is a tissue-specific protein (chondrocytes, adipocytes, skeletal myofibers, cardiomyocytes, dendritic cells, etc.); it is released after a damage (and the consequent remodelling) involving these tissues, increasing its plasma levels. These findings suggest S100B to be a more specific marker of heart disease [72, 73].

Limitations

Given the considerable heterogeneity in the study designs and subjects (sample sizes, proteins analysed, animal/human studies) of the selected studies, the characteristics of study populations, the objectives and the study results are reported using descriptive statistics without conducting meta-analyses.

Tables redundancy may appear as a limitation, but is voluntarily obtained to focus reader attention on the topic in examination.

Conclusions

New insight into molecular mechanisms in heart failure is a current challenge for a modern medicine approach. More evidence on efficacy and safety of gene therapy has to be collected in humans. Also, S100A1 safety profile has to be further investigated. Other kinds of target therapy (i.e. monoclonal antibodies versus S100A1 ligand-site like RyR2, SERCA and phospholamban) could be considered as an alternative. S100A8–A9 and S100B have a good potential and should be evaluated on a larger cohort of patients. In a near future, these proteins could be exploited as biomarkers in stadiation and prognosis of chronic heart failure, as well as therapeutic target to rescue failing heart.