Semen rheology and its relation to male infertility

Abstract

Infertility affects 15% of couples of reproductive age worldwide. In spite of many advances in understanding and treating male infertility, there is still a number of issues that need further investigation and translation to the clinic. Here, we review the current knowledge and practice concerning semen rheology and its relation with pathological states affecting male infertility. Although it is well recognized that altered rheological properties of semen can impair normal sperm movement in the female reproductive tract, routine semen analysis is mostly focused on number, motility and morphology of spermatozoa, and includes only an approximate, operator-dependent measure of semen viscosity. The latter is based on the possible formation of a liquid thread from a pipette where a semen sample has been aspirated, a method that is sensitive not only to viscosity but also to elongational properties and surface tension of semen. The formation of a liquid thread is usually associated with a gel-like consistency of the sample and changes in spermatozoa motility in such a complex medium are still to be fully elucidated. The aim of this review is to point out that a more quantitative and reliable characterization of semen rheology is in order to improve the current methods of semen analysis and to develop additional tools for the diagnosis and treatment of male infertility.

1. Introduction

Fertility refers to the reproductive capacity of living organisms by sexual reproduction, which occurs through the unification of male and female gametes. In mammals, this requires that at least one sperm cell, among the hundreds of millions that flow toward the egg driven by biochemical signals, reaches the site of fertilization. Before standing a chance of fertilizing an egg cell, there is a long and difficult journey for sperms, that should travel through the fallopian tubes, where they find a complex winding path and a likely hostile environment, due to the presence of immune cells and a viscoelastic fluid as suspending medium. And that is not all. The few sperms able to overcome these first obstacles should be subjected to morphological and biochemical changes and, as the last step, should interact with the egg in the oviduct [1,2]. Thus, the success of reproduction is based on a fragile balance between biochemical and physical processes that, in turn, can be due to and affected by genetic and lifestyle factors. When the balance is not reached or is broken, we refer to infertility. According to the definition by the World Health Organization, infertility is ‘the failure to achieve a clinical pregnancy after 12 months or more of regular unprotected sexual intercourse’, and many other definitions have been used in clinical practice as well as in demographic research [3]. The heterogeneity of the criteria used to define infertility results in a difficult estimate of the prevalence of infertility worldwide. Moreover, the discrepancy between data from large-scale population surveys on infertility and the ones from epidemiological studies as well as the fact that infertility can be vaguely referred to women, men, couples or individuals make the determination of the number of people affected by this problem even more complicated [4]. Despite this, a recent study based on findings of the global burden of disease has assessed the burden of infertility in 195 countries and territories from 1990 to 2017, reporting that infertility affects 15% of couples of reproductive age worldwide [5–7], with increasing prevalence rates that are, in many areas, such as the USA and Europe, below the rate required to sustain the current population levels (see maps in figure 1). In more detail, almost 50% of infertility cases are only due to female factors, 30% are related to male factors and 20% to the combination of both male and female factors [7,8]. In particular, a decrease in male fertility as well as in semen quality, in terms of sperm concentration and motility, has been registered in the last few years. However, the male contribution to infertility could be underestimated, since in many countries, due to cultural issues and patriarchal societies, infertility is not registered as a male-related problem, hindering the collection of reliable data [7].

Figure 1.

Total fertility rate, measured as the number of children that would be born to a woman if she were to live to the end of her childbearing years and bear children in accordance with age-specific fertility rates of the specified year, for 1990 (top) and 2017 (bottom) (https://ourworldindata.org/fertility-rate).

Good semen quality is commonly expressed in terms of number, motility and morphology of spermatozoa as well as presence of leucocytes and immature cells [9], and is fundamental for child health as well as for the success of assisted reproductive technology treatments, that are proceeding at the same speed as the decreasing rate of male infertility. However, the success rate is approximately 33% per cycle, resulting in the need of multiple fertilization cycles with severe emotional and financial burdens for the couples [8].

The worrying reduction of semen quality and male fertility can be ascribed to several factors, that can be either environmental or genetic or a combination of them [10], as discussed in the following sections. Beyond being a clinical disease, infertility negatively affects life quality, also causing psychological distress, social stigmatization, economic effort and marital problems, with considerable psychological, social and economic implications. Therefore, all these issues ask for urgent attention, highlighting the need of a deeper understanding of the factors responsible of infertility, and their intercorrelations, and of the development of advanced strategies to make both diagnosis and treatment affordable and accessible and, eventually, personalized.

2. Seminal fluid and its characterization

Seminal fluid, also known as sperm or semen, is composed of spermatozoa, i.e. male reproductive cells, suspended in a liquid medium called seminal plasma. Spermatozoa represent 2–5% of the overall volume of sperm, corresponding to about 15–30 million per millilitre [11]. Spermatozoa synthesis takes place within the seminiferous tubules of the testes via spermatogenesis, a tricky process that can produce malfunctioning or unripe spermatozoa if even only one step of it fails. Once formed, spermatozoa are kept viable thanks to the help of seminal plasma, a complex fluid composed of secretions of seminal vesicles (60–70%), prostate (20–30%) and epididymis (less than 1%) [12]. The former are rich in proteins [9], prostaglandins [13], cytokines, and fructose [14], while secretions by prostate are rich in proteolytic enzymes [15], citrate [16], lipids [17] and zinc [18,19]. Each component of seminal plasma has a specific function and is essential for the maturation, metabolism and life of spermatozoa, as well as for their survival after ejaculation. Among others, semenogelins (SEMGs) I and II, the major protein constituents of seminal vesicle secretions, have crucial functions, being involved in several key steps preceding fertilizations. In particular, immediately after ejaculation, SEMGs together with fibronectin, form a gel-like, cross-linked structure of SEMGs, the so-called coagulum, which hinders any spermatozoa movement and pre-activation (so-called capacitation) [20,21]. The subsequent cleavage action of prostate-specific antigen (also known as kallikrein-related peptidase 3 or KLK3) and other prostate proteins breaks the SEMG-based structure that re-assembles into amyloid fibrils. As a consequence of this process, called liquefaction, semen becomes watery and spermatozoa gain their motility [22] (figure 2).

Figure 2.

Cartoon of the liquefaction process. Adapted from [22].

Zinc [18,19] plays an important role in semen liquefaction, has antioxidative and bactericide functions, and acts as stabilizer of the chromatin of spermatozoa. Fundamental are also enzymes and citric acid, which intervene in the coagulation–liquefaction process of the sperm as well, by bicarbonate [23], which modulates sperm motility and neutralizes the acidity of the vaginal environment, and by fructose which provides the necessary energy for normal sperm functions. In addition to sperm and seminal plasma, the ejaculate comprises also immature cells from spermiogenesis and flaking epithelial cells.

Despite the complexity of such a fluid, from both the chemical and physical points of view, routine semen analysis, beyond male detailed medical history and physical examination, focuses on the evaluation of a few parameters [24,25], as shown in table 1. The resulting report, called spermiogram, contains information from two types of tests: macroscopic and microscopic.

Table 1.

World Health Organization reference values for human semen [26]. PR = progressive; NP = non progressive.

type of examination	parameter (unit)	lower reference limit (range)
macroscopic	semen volume (ml)	1.5 (1.4–1.7)
	pH	>7.2
	viscosity (semen wire length in cm)	<2
	time of liquefaction after ejaculation (min)	30 (15–60)
microscopic	total sperm number (106 per ejaculate)	39 (33–46)
	sperm concentration (106 per ml)	15 (12–16)
	progressive motility (PR %)	32 (31–34)
	total motility (PR + NP %)	40 (38–42)
	vitality (live spermatozoa)	58 (55–63)
	sperm morphology (normal form %)	4 (3–4)
	agglutination (spermatozoa per cluster)	<10

Macroscopic examination starts soon after semen liquefaction, the spontaneous process by which the gel-like ejaculate fluid is broken up, resulting in a thinner, homogeneous and watery fluid. The precise measurement of semen volume is fundamental for the right calculation of sperm concentration; pH measurements are important since pH can affect sperm capacitation and motility [27]; abnormal values of viscosity can affect sperm motility and could lead to the incorrect determination of some parameters (i.e. sperm concentration, biochemical biomarkers) during routine tests.

Microscopic examination is carried out by using a phase-contrast microscope with a high-magnification objective and focuses on the measurement of sperm number, motility, morphology and agglutination. Sperm motility is categorized into: (i) progressive (P)—ideal—motility, i.e. active movement of spermatozoa, which swim in a straight line; (ii) non-progressive (NP) motility, i.e. absence of progressive movement of spermatozoa, which tend to travel in a curved or crooked motion; and (iii) immotility, i.e. when spermatozoa move only their tail or are immotile (figure 3).

Figure 3.

Sperm motility categorization.

Normal morphology could be more difficult to quantify, due to the large number of shapes that spermatozoa can exhibit. Hence, potential fertilizing spermatozoa are taken as normal if they appear similar in shape (both head and leg curvature are taken into account) to those reported in the literature [28] as recovered from endocervical mucus after intercourse. Regarding agglutination, it refers to the sticking of motile spermatozoa to each other, which could hinder the journey of sperm toward the egg [29].

3. Causes of male infertility

In the majority of male infertility cases (figure 4), the cause can be found in the testes, which are the male reproductive glands that produce sperm and androgens (i.e. male sex hormones, primarily testosterone). Damage to the testes, resulting in altered sperm production, can be caused by many different conditions, including inflammatory states and infections, trauma, treatments for cancer (i.e. radiation/chemotherapy) and surgery. Abnormal sperm production can also be caused by genetic diseases [30–34] (either inherited or acquired), which can lead to decreased motility. Sperm production can also be affected by heat stress [35–37], environmental [38–41] and lifestyle factors (i.e. obesity [42,43], smoking [44–46], alcohol abuse [36,47] and use of recreational drugs [36,48]) and, rarely, by hormone deficiency [49,50].

Figure 4.

Most common male infertility causes.

In the 30% of male infertility cases of unknown aetiology we talk of idiopathic or unexplained male infertility. The former is characterized by normal physical examination and endocrine laboratory testing and no evident history of fertility problems, but sperm abnormalities of idiopathic causes are found from semen analysis by established standard testing protocols. The term unexplained infertility, instead, refers to those cases in which spermiogram values are normal and a female infertility factor could not be identified [51].

Recently, a relation has been suggested between idiopathic male infertility and oxidative stress that, in turn, is related to excessive levels of reactive oxygen species (ROS) and depleted antioxidant capacity [52,53]. ROS overproduction can be due to intratesticular and post-testicular factors, like seminal leucocytes [54], residual cytoplasm or cytoplasmic droplets and abnormal spermatozoa, as well as to external factors, like alcohol, cigarette smoking, environment pollutants and several pathological conditions (e.g. varicocele and diabetes) [55]. Indeed, while oxygen is fundamental for aerobic metabolism of spermatogenic cells, it can also be dangerous being one of the main causes of DNA damages. The latter can be of different nature, from nuclear and mitochondrial DNA defects to telomere attrition, Y chromosomal microdeletions [56–59] and epigenetic alterations [60–63]. DNA damages are due to downregulation of DNA repair systems during late spermatogenesis. Indeed, oxidative stress can induce sperm chromatin/DNA damage, as demonstrated by recent studies that have correlated sperm DNA fragmentation and high levels of ROS in semen [47,64,65], while DNA integrity in germ cells is essential, since they have the task of passing the genome to the next generation [66]. In turn, DNA fragmentation can accelerate the process of germ cell apoptosis, resulting in low sperm count and in the deterioration of semen quality, compromising the potential of sperm to fertilize an oocyte and develop into a healthy embryo [67–69].

Additional factors affecting male fertility can lead back to oxidative stress and testicular and post-testicular damages. Indeed, it has been shown that aged men may present higher DNA fragmentation [70] than younger subjects, likely due to oxidative stress [71] and testicular apoptosis of germ cells, which leads to the decline of both potential daily sperm production and Leydig cell function [72]. However, other sperm parameters, such as concentration, motility and morphology, do not change significantly with age. Systemic diseases, such as chronic renal and liver failure, may also impair male fertility. In particular, renal failure may affect either hypothalamic–pituitary–gonadal axis or testes, leading to both endocrine and exocrine testicular dysfunction [73,74], while cirrhosis can entail testicular atrophy and sexual dysfunctions [75]. In both cases, transplantation of kidney and liver can be resolutive.

Regarding post-testicular defects, acquired or congenital obstruction arising from the epididymis, vas deferens, or ejaculatory duct can be the basis of obstructive azoospermia, present in about 40% of men with azoospermia [76]. By contrast with testicular damages, post-testicular abnormalities are treatable, and it is often possible to restore the fertility potential.

Several comprehensive reviews have been published so far related to the causes of male infertility [77] and the most recent ones have focused on genetic-related causes [30,59]. It is hence not the aim of this work to provide another detailed review of male infertility aetiology, but rather to focus on semen rheology and its possible correlation with impaired sperm motility and fertility.

4. Semen viscosity and viscoelasticity

About 20–30 min after ejaculation the gel-like semen starts the proteolytic process known as liquefaction which makes semen less viscous and watery [78]. In the female reproductive tract, this is a fundamental step allowing increased sperm motility and successful transport to the fertilization site. The total or partial failure of the liquefaction process is likely related to what is regarded as semen hyperviscosity, a condition that is considered an index of male impaired fertility. Some pathological conditions [79], either genetic, such as cystic fibrosis [80], or acquired, such as sex gland dysfunction, oxidative stress, inflammation and infections and leucocytospermia, have been correlated to the presence of hyperviscous semen, all resulting in decreased semen quality, in particular reduced sperm motility [81], and impaired fertility rate [82] (figure 5). Concerning genetic causes, CFTR (cystic fibrosis transmembrane regulator) gene mutation that causes cystic fibrosis pathology results in defects in the regulation of chloride and sodium ion exchange across epithelial membranes that, in turn, affect the fluidity of many body fluids, such as blood, amniotic fluid, synovial fluid and mucus [83]. Regarding other causes, in many cases, a correlation has been found between semen hyperviscosity and dysfunction of periprostatic venous plexus [84], prostate and seminal vesicles, hypofunction of the latter leading to an abnormal amount of zinc [85,86].

Figure 5.

Causes and effects of semen hyperviscosity.

The decrease of zinc uptake in the case of hyperviscosity can also lead to oxidative stress (remembering the antioxidative function of zinc), with all the related known negative effects. Oxidative stress itself has been recognized as one of the main factors correlated to semen hyperviscosity, even if the exact mechanism is not yet understood.

Seeking the aetiology and evaluating the consequences of semen hyperviscous nature are made more difficult by the fact that, currently, the routine semen viscosity measurement is carried out by only qualitative, operator-dependent and non-reproducible methods, consisting of the aspiration of the semen into a pipette and allowing the sample to drop by gravity. The length of the thread is used to classify viscosity as low, normal and high, considering normal viscosity when thread length is less than 2 cm. It is worth noting that such a procedure, more than a measure of viscosity, likely evaluates the combined effect of the elastic and viscous properties of the fluid and is also affected by surface tension. Despite the aforementioned issues, many studies focused on the correlation between semen hyperviscosity and the possible causes/effects of it by using the semen thread length evaluation [78]. Several investigations associated leucocytospermia in infertile patients to semen hyperviscosity, patients with hyperviscous semen showing higher percentage of leucocytes compared with controls, and reduced sperm motility and vitality [85]. Other studies [87,88] reported a correlation between semen hyperviscosity and severe impairment of low- and high-molecular-weight seminal antioxidants (e.g. enzymatic ROS scavengers that have the function to protect spermatozoa against ROS toxicity). In turn, ROS overproduction can be due to high levels of leucocytes present in infections or inflammations. Semen hyperviscosity has been also associated dysfunction of the male accessory glands. In particular, it has been observed that the quality of semen, in terms of sperm motility, vitality and fructose level, was impaired in highly viscous samples from donors with hypofunction of seminal vesicles [86].

So far, only few papers reporting the use of quantitative methods have been proposed for the rheological characterization of human semen, mostly exploiting glass capillary viscometers [84]. Tjioe & Oentoeng [89] used a capillary viscometer to correlate semen viscosity and motility, finding a decrease of sperm motility with increasing viscosity (figure 6a). Nag et al. [91,92] studied the correlation between spermatozoa concentration/motility and semen viscosity, observing a general negative correlation between sperm count and motility and viscosity for normal, asthenozoospermic (semen with reduced sperm motility) and oligozoospermic (semen with reduced count of spermatozoa) groups. Moreover, they found much higher viscosities in oligoasthenospermic (reduced number and motility of spermatozoa) and necrozoospermic (non-vital spermatozoa) groups, suggesting that the macromolecules present in plasma semen can play a role in the determination of semen viscosity, beyond sperm concentration and motility. Another possible explanation of this result can be the decrease of viscosity of semen due to the presence of swimming, pusher cells, as discussed in the following [93]. Dunn & Picologlou [94,95] used a multiple-point capillary viscometer for the evaluation of the time-dependent viscoelastic behaviour of semen, even if it was an only one-donor study. They showed that semen behaves as a viscoelastic fluid immediately after ejaculation and as a Newtonian fluid after full liquefaction. Aydemir et al. [90] studied the effect of malondialdehyde (MDA), an unsaturated carbonyl product of oxidative stress, on semen viscosity, reporting a significant positive relationship of seminal fluid viscosity with seminal plasma MDA and sperm MDA [90] (figure 6b).

Figure 6.

(a) Graph reporting the percentage of motile spermatozoa as a function of semen viscosity (cP) for 1111 semen samples. Adapted from [89]. (b) Graph reporting plasma semen viscosity (mPa s) as a function of plasma semen MDA levels (nmol ml⁻¹) for 60 infertile males. Adapted from [90].

However, capillary-based approaches are not the right tools for the assessment of viscosity of a fluid as complex as human semen. In fact, it has been observed that semen viscosity is not constant with the imposed shear rate, but decreases, showing a behaviour typical of viscoelastic fluids [96,97]. In fact, viscoelastic behaviour is common in biological microstructured fluids, such as blood [98] and mucus [83]. Thus, more reliable viscosity measurements have been performed by using rotational viscometers. Hubner et al. [97] used a Couette viscometer to measure the viscosity of previously frozen semen samples at three different shear rates, classifying the samples according to the spermatozoa concentration, and found a non-Newtonian behaviour for all the samples but no correlation with sperm count. Similar results have been reported in the work of Lin et al. [81], except for the higher viscosity found for oligospermic patients. More recently, Mendeluk et al. [96] reported a power law behaviour for semen (figure 7), providing shear stress versus shear rate curves for both normal and hyperviscous samples, and attributed hyperviscosity to the presence of a highly organized network of macromolecules.

Figure 7.

Shear stress versus shear rate curve reporting the comparison between normal (19 samples) and hyperviscous (67 samples) semen as estimated by the thread length method. Adapted from [96].

Regarding the effect of semen abnormal physical properties, it has been found that altered semen viscosity could strongly affect spermatozoa progressive motility [99], as happens for many swimming microorganisms, such as bacteria and algae [100]. In fact, it has been observed that physical environmental factors have a role in shaping the evolution of spermatozoa flagellar kinematics.

In particular, the increase of suspending medium viscosity decreases the wavelength and the frequency of spermatozoa flagellar wave [8,101,102] (figure 8), while larger uniform flagellar beating amplitude and localized regions of high curvature optimize swimming efficiency [103,104]. In turn, fluid viscosity can be influenced by the presence of actively swimming microorganisms, such as sperm cells.

Figure 8.

(a) Images of human spermatozoa swimming in low-viscosity (top) and high-viscosity (bottom) media. (b) Spermatozoa flagellar profiles. (c) Waveform curvature maps of flagellar movement as a function of time t and arc length position along the flagellum. Adapted from [101].

It has been found that viscosity in elongational flow of suspensions of pullers (e.g. algae) is higher than that of a passive suspension (i.e. a suspension of inactive cells of the same size, shape and concentration), while viscosity of suspensions of pushers (i.e. spermatozoa of mice and bacteria) is lower than that of a passive suspension [93].

Furthermore, possible elastic effects in semen rheology should be considered as well: in fact, the forces experienced by a flexible swimmer in a viscoelastic medium are different from those experienced in a Newtonian fluid [105]. Most of the literature, however, is concerned with measurements of viscosity and essentially no information is currently available on the elastic properties of semen. As shown in Creppy et al. [106] on a work on ram semen, even after strong dilution, both semen and plasma semen show the same shear-thinning trend that could likely be ascribed to the presence of proteins (figure 9). This means that the motility of spermatozoa in such a complex medium could be altered in a way that has not been fully elucidated.

Figure 9.

Flow curve (viscosity versus shear rate) as a function of shear rate for semen (filled symbols) and seminal plasma (open symbols) at different dilution (green for pure fluid, red for ×10 dilution, black for ×100 dilution). Adapted from [106].

Some insight on the effect of viscoelastic properties can be obtained from the work done on sperm motility when the suspending medium is mucus of the female reproductive tract [107], which plays a key role in spermatozoa selection as well as in spermatozoa swimming efficiency. Thus, as a way of mimicking the behaviour of mucus of the female reproductive tract, some studies focused on the swimming of spermatozoa in highly viscoelastic media. By numerical modelling, it has been found that the viscoelastic nature of the medium increases spermatozoa velocity by forming regions of highly strained fluid behind the tail [108]. However, the proposed model, although effective in capturing the elastic responses, did not include the shear thinning behaviour of the fluid nor the presence of macromolecules that induce viscoelasticity. In a recent study—on bull spermatozoa in a model viscoelastic fluid—it has been observed how the viscoelastic nature of the medium can increase the efficiency of spermatozoa swimming and can also induce dynamic clustering and collective swimming, likely contributing to effective fertilization [109] (figure 10). Furthermore, Creppy et al. [110] used livestock semen to analyse the swimming efficiency of concentrated suspensions of spermatozoa in confined conditions, mimicking the confined environment of the oviduct tract. They found a non-trivial spontaneous motion of spermatozoa, such as a phase transition at a critical volume fraction of cells. However, while the motion of swimming microorganisms in Newtonian suspending media has been investigated in some detail, what happens in viscoelastic fluids, where elasticity can either promote or suppress locomotion, is still an open issue. It can be expected that spermatozoa motility, promoted by flagellar beating at certain frequencies, could be influenced by the viscoelastic properties of the medium where they are swimming, as happens for both passive and active particles (i.e. bacteria and algae).

Figure 10.

Bull sperm motion in (a) standard medium for sperm suspension; (b) viscous medium and (c) viscoelastic medium. (d) Percentage of sperm found in clusters (including pairs) in standard medium (Std), Newtonian viscous medium (V) and viscoelastic medium (VE) [109].

In any case, while the above-mentioned studies have highlighted how relevant is the viscoelasticity of the suspending fluid for sperm motility, it remains to be understood whether human semen actually behaves as a viscoelastic fluid, even in the case of hyperviscosity. The study of semen rheology could play a key role not only for a deeper understanding of the effect of semen physical properties on male fertility and human reproduction, but also for bioengineering problems, such as semen storage [111], and design of sperm analysis and selection devices [8]. However, despite all the efforts devoted to this issue until now, comprehensive, quantitative rheological characterizations of human semen viscoelasticity are lacking in the literature, as well as a correlation between semen elasticity and sperm motility. The latter correlation will be relevant to elucidate the effect of semen rheology on male fertility, where the impairment of sperm motility in semen with altered viscoelastic properties can play a key role.

5. Perspectives

It is a fact that a significant percentage of male infertility cases are of unknown or not-well-understood aetiology, and it is a fact that semen physical properties, such as viscosity, are implicated in impaired sperm motility and DNA damages. However, the techniques currently in use for the evaluation of semen rheology are not reliable in terms of reproducibility and significance of the results, despite the physiopathological importance that quantitative measurements of semen viscoelastic properties could have in the understanding of some of the mechanisms, both inherited and acquired, at the base of impaired sperm motility, and the likely consequent advancements in the diagnosis and treatment of infertility.

Thus, the attention of further research should be directed toward a systematic quantitative assessment of the rheological characteristics of human seminal fluid, in terms of viscosity and elasticity, defining a quantitative index of semen viscoelasticity to be used in diagnosis routes, coupled to sperm motility tests. Such an approach could eventually inspire researchers and companies to develop cheap, fast and user-friendly tools for point-of-care analysis, for example, based on microfluidic techniques [8].

Furthermore, in order to tackle the growing problem of infertility, quantitative and complete investigations should be carried out on both male and female samples, associating molecular analysis [34] to the investigation of the interaction of seminal fluid with cervical mucus, a long-standing open problem in the study of human reproduction [112] in terms of viscoelastic properties of both fluids. The integration of genetic analysis and microfluidics experiments could lead to a reappraisal of the post-coital test [113] and of the sperm–cervical mucus penetration test [114] to study the semen–cervical mucus interplay and assess sperm quality.

Data accessibility

This article has no additional data.

Authors' contributions

G.T.: conceptualization, data curation, resources, writing—original draft, writing—review and editing; F.F.: data curation; V.P.: data curation; S.G.: writing—review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

The authors acknowledge financial support from PON Ricerca e Innovazione 2014–2020 (grant no. CCI 2014IT16M2OP005).