Mechano-Sensing and shear stress-shielding by endothelial primary cilia: structure, composition, and function
1Beijing Advanced Innovation Center for Biomedical Engineering, Key Laboratory for Biomechanics and Mechanobiology of Ministry of Education, School of Biological Science and Medical Engineering, Beihang University, Beijing, 100083, China
2Department of Biomedical Engineering, The City College of the City University of New York, New York, NY 10031, USA
*Address correspondence to: Bingmei M. Fu, email@example.com
Received: 09 April 2021; Accepted: 20 May 2021
Abstract: Primary cilium is an antenna-like and non-motile structure protruding from the apical surface of most mammalian cells including endothelial cells lining the inner side of all the blood vessels in our body. Although it has been over a century since primary cilia were discovered, the investigation about their mechano-sensing and other roles in maintaining normal functions of cardiovascular system has just started in recent years. This focused review aims to give an update about the current literature for the role of endothelial primary cilia in blood flow mechano-sensing and shear stress-shielding. To do this, we first summarized the characteristic features of endothelial primary cilia in terms of structure, dimension, molecular composition, and mechanical properties (e.g., bending rigidity), which are the dominant factors for their functions in mechano-sensing and transduction, as well as vascular protection from the blood flow-induced wall shear stress. We also described the experimental techniques and mathematical models for determining the dimension and mechanical properties of the primary cilium. Then we reviewed the molecular mechanisms underlying mechano-sensing and transduction by endothelial primary cilia and the mathematical model prediction for their roles in redistribution and reduction of wall shear stresses. Finally, we briefly discussed the common cardiovascular diseases, e.g., atherosclerosis, hypertension, and aneurysm, due to defects and malfunction of endothelial primary cilia and suggested potential targets for therapeutic treatments.
Keywords: Dimension; Mechanical properties; Mathematical modelling; Cardiovascular diseases
Primary cilia are microtubule-formed organelles discovered in many types of mammalian cells including epithelial cells (Grisanti et al., 2016; Han et al., 2018), osteocytes (Lee et al., 2015; Spasic and Jacobs, 2017a), chondrocytes (Zhan et al., 2017; Mansini et al., 2018), neurons (Norris and Santoro, 2018), and endothelial cells (Luu et al., 2018; Follain and Goetz, 2018). They were previously thought to be functionless until Praetorius and Spring (2001) observed the influx of Ca2+ in response to both micropipette bending and fluid flow towards primary cilia. Since then, primary cilia have been identified to be one of the most vital mechano-sensors and transducers in many tissues and organs. Other roles of endothelial primary cilia in blood vessel development and cardiovascular functions have also been discovered in recent years. Defects in primary cilia could induce various diseases, named ciliopathies, such as mental retardation, polycystic kidney disease, liver disorders, obesity and cardiovascular diseases including hypertension, aneurysm, and atherosclerosis (Luu et al., 2018; Pala et al., 2018; Wang et al., 2021).
Vascular endothelial cells (ECs) lining the inner wall of our blood vessels are continuously exposed to the blood flow. In order to maintain proper functions of the cardiovascular system, ECs should have a variety of mechano-sensors and transducers to sense the blood flow change and adjust the blood perfusion rate (e.g., through vessel diameter) and transport across the vessel wall accordingly. So far, at least a dozen EC mechano-sensors and transducers have been identified either on the EC surface, or at the intra- and trans-EC membrane, or within the EC cytoskeleton. Several prominent ones are endothelial surface glycocalyx, caveolae, integrins, VE-cadherins, PECAM-1, G-protein-coupled receptors and G-proteins, actin filaments, nesprins, and primary cilia (Fu and Tarbell, 2013; Jung et al., 2018; Morikis and Simon, 2018; Liu et al., 2019; Pala et al., 2018). Among these EC mechano-sensors and transducers, the primary cilium is attracting more and more attention due to its unique location, one per cell in the middle of the apical surface of ECs, and its long and flexible microtubule-based structure (Fig. 1). It behaves like an antenna, the frontier of the EC that detects the blood flow and transmits the signals to the EC via various signaling pathways along its length. In addition to a sensor and transducer, the primary cilium can serve as a mechanical shield which can reduce and redistribute the wall shear stress acting on the EC surface in pulsatile flows as predicted by a mathematical model (Cui et al., 2020). The aforementioned functions of endothelial primary cilia are determined by their structure, molecular composition, dimension, and mechanical properties (stiffness or bending rigidity), we therefore started the review with these features.
Structure, Molecular Composition, Dimensions and Mechanical Properties
The antenna-like primary cilium is considered as a cellular organelle, which mainly consists of a membrane, soluble compartment (cilioplasm), axoneme, basal body and ciliary tip. The structure of a general primary cilium is depicted in Fig. 2. It protrudes from a basal body complex with two centrioles. One is called the mother centriole, where the ciliary axoneme is rooted beneath the cell membrane. Another is called the daughter centriole which is perpendicular to the mother centriole. Both centrioles are connected with striated connectors (Hagiwara et al., 2008; Satir et al., 2010). The cilium structure containing the microtubular parts of the cytoskeletal core unit is named the ciliary axoneme. The ciliary axoneme, which is bounded by the ciliary membrane and protrudes from the basal body, consists of nine doublet microtubules distributing evenly along the perimeter (Hoey et al., 2012). Each doublet consists of two tubules named tubule A and tubule B. Tubule A contains 13 protofilaments and tubule B contains 10 protofilaments (Downing and Sui, 2007). Different from the well-known motile cilia, the primary cilia lack two central microtubules and motility (Young et al., 2012). The ciliary membrane is thought to be continuum of cytomembrane, however, the composition of the two are different. Decorated with various proteins, the ciliary membrane seems to coordinate the initiation and transmission of the extracellular stimuli to the interior of the cell. The region between the axoneme and the basal body is the transition zone, which forms a unique subcellular domain and acts as a barrier to prevent free exchange between the cytosol and cilium. At the base of the cilium, there are invaginations of the plasmalemma that abut the ciliary membrane to form the ciliary pocket (Zhang et al., 2019). Cilioplasm, the soluble compartment of the cilium, contains materials that are essential for the proper function of the cilium (AbouAlaiwi et al., 2009a). In the basal body complex, there are structural components termed alar sheet, basal feet, and striated rootlets. Alar sheets are fibrous triangular structures, which originate from the basal body to the transition zone and attach to the cell membrane (Hagiwara et al., 2008). They can regulate protein diffusing into and out of the primary cilium (Hoey et al., 2012). Basal feet are conical structures. Unlike motile cilia, which have only one basal foot, primary cilia have several basal feet protruding from the basal body and connecting with cytoskeleton microtubules to stabilize the basal body (Hagiwara et al., 2008). Striated rootlets are coiled filamentous structures, which extend from the basal body of the cilium towards the cell nucleus and connecting with the membrane structures (Hagiwara et al., 2008). Wheatley (2008) proposed that striated rootlets act like muscle contraction, which pulls the primary cilium towards the cell surface in response to specific stimuli.
Various molecules residing on the primary cilia contribute to the maintenance of the proper ciliary structure and function. Any alteration in the fluid pressure and shear stress can be detected by the sensory proteins localized on the cilia and these alterations are transduced into the interior of the cell via various signaling pathways. The signaling pathways organized by primary cilia are rather diverse and depend on the cell type. Here we only focus on several proteins that are widely described on endothelial primary cilia, which are illustrated in Fig. 2.
Intraflagellar transport (IFT) particles, including intraflagellar transport protein 88 (IFT88), are essential for building primary cilia, or ciliogenesis. During ciliogenesis, the centrosome migrates towards the cell surface when a cell enters G0 (resting state) and the mother centriole attaches to a Golgi-derived vesicle (Satir et al., 2010). Then ciliary axoneme projects from the mother centriole into the vesicle lumen. The axonemal microtubules polymerize at the tip of the projection where the cargo is delivered by IFT. The vesicle is ultimately exocytosed to expose the primary cilium at the cell surface, where the ciliary growth continues to form the mature cilium (Satir et al., 2010). IFT particles, along with several other molecules (kinesin 2, cytoplasmic dynein 2 and ciliary membrane proteins), contribute to delivering cargos to assemble and maintain the axoneme of the primary cilia, including kinesin-based anterograde and dynein-powered retrograde transport (Satir et al., 2010). Knockdown of IFT88 causes deficiency of primary cilia. This approach has been widely used to investigate the role of primary cilia. Tg737 is a mouse homologue gene of IFT88 and cells with mutation in Tg737 exhibit no or shorter cilia (Pazour et al., 2000; Yoder et al., 2002).
Polycystin-1 and -2 (PC1/2) have been thought to form a heteromeric ion channel complex on the primary cilia, regulating Ca2+ influx into the cell (AbouAlaiwi et al., 2009b; Theodorakopoulou et al., 2019). PC1, coded by PKD1, is an 11-transmembrane protein with a long extracellular N-terminus that contains multiple cell adhesion domains (Hughes et al., 1995; AbouAlaiwi et al., 2009b). PC2, coded by PKD2, belongs to a superfamily of transient receptor potential (TRP) ion channel with 6-transmembrane domains (AbouAlaiwi et al., 2009b). Although the interaction of PC1 and PC2 may be controversial (Liu et al., 2018; Ha et al., 2020; Ta et al., 2020), it has been widely accepted that this complex is necessary for normal vascular development and is required for endothelial cilia to sense the blood flow induced forces and promote normal intracellular signaling (AbouAlaiwi et al., 2009b; Diagbouga et al., 2018; Theodorakopoulou et al., 2019). AbouAlaiwi et al. (2009b) also demonstrated that in response to fluid shear stress, mouse endothelial cells with knockdown or knockout of PKD2 lose the ability to generate nitric oxide (NO). They thus proposed a new role for PC2 in transmitting extracellular shear stress to intracellular NO synthesis. In addition to PC1 and PC2, there are other Ca2+ flux mediating channels including transient receptor potential vanilloid 4 (TRPV4) and PIEZO1/2 (Spasic and Jacobs, 2017b; Lee et al., 2015; Praetorius, 2015).
The TRP channels (e.g., TRPV4, TRPC1, and TRPP2) also exist at the EC membrane and can be activated by mechanical stimuli such as stretch and flow induced-shear stress to regulate endothelial [Ca2+] and cell membrane potential (Kohler et al., 2006; Kwan et al., 2007). These stimuli can either directly activate the TRP channels on the EC membrane or through other mechano-sensors located at the EC surface such as endothelial glycocalyx (Dragovich et al., 2016). In contrast, the TRP channels on the primary cilium might only be activated by bending of the cilium through a force perpendicular to the cilium, similar to that for activating the PC1 channel (Nauli et al., 2008). In addition, the TRP channels on the cilium only regulate the Ca2+ influx, not the EC membrane potential.
Methods for determining the dimensions of primary cilia
To investigate dimensions (e.g., length and diameter) of primary cilia, various techniques have been employed including transmission electron microscopy, scanning electron microscopy (Collin and Collin, 2000; Van der Heiden et al., 2006; Geerts et al., 2011; Egorova et al., 2012) and fluorescent confocal microscopy (Van der Heiden et al., 2006; Geerts et al., 2011; Russellpuleri et al., 2017; Nauli et al., 2008; Antal et al., 2017; Vo and Bols, 2016; Iomini et al., 2004; Ki et al., 2020; Dummer et al., 2018; Goetz et al., 2014). Transmission electron microscopy can provide the detailed ultrastructure information of the cilium and scanning electron microscopy can provide information regarding cell surface morphology, cilia density and the contact of the cilia with the cell surface. But electron microscopy is expensive, time-consuming and may induce artifacts due to dehydration and other processes during sample preparation. Fluorescent confocal microscopy is chosen when the function and molecular mechanism of cilia are investigated in either live samples or fixed samples without dehydration. But its spatial resolution is not as good as that of electron microscopy.
Due to its convenience, easiness in sample preparation and capability in identifying molecular composition, fluorescent confocal microscopy is the most commonly used technique in determining the ciliary length (Dummer et al., 2016). Protruding from the apical surface of cells, primary cilia may appear in random orientations with some cilia standing perpendicular to or tilted (angled) to the cell surface while others lying in the cell surface (focus plane),which are named flat cilia (Dummer et al., 2016). Therefore, imaging only one focus plane may underestimate the ciliary length. The application of confocal Z-stacks of many imaging planes can better reconstruct the cilium. From the confocal Z-stacks, three methods have been used to determine the ciliary length: the maximum intensity projection (MIP), the Pythagorean theorem (PyT), and 3D alternative angled slicing (DAAS) (Dummer et al., 2016). MIP is the projection of the fluorescently labeled ciliary components with the maximum intensity from each slice of the collected confocal Z-stacks. PyT calculates the actual ciliary length c by using c2 = a2 + b2. Here a and b represent the length determined by the MIP and the height of the cilium provided by the number and thickness of Z-slices, respectively. In DAAS, the 3D reconstruction of the cilium from the collected Z-slices is separated by a bilinear interpolation to several sections and each section is measured. The total length of a cilium is the summation of all the section lengths. Each method has its advantages and disadvantages. In general, MIP is the easiest approach that can provide an accurate result for flat cilia but underestimates the length of angled cilia, whereas PyT is able to accurately determine both flat and angled straight cilia (Dummer et al., 2016). For irregularly shaped cilia, DAAS is the best, but it is time-consuming (Dummer et al., 2016). In addition to the measuring methods, the results may also be influenced by the artifact in the fixation and/or immune-staining procedure, and by the choice to label specific components of the cilia. The common antibodies used to label endothelial primary cilia are monoclonal acetylated-tubulin antibody (Russellpuleri et al., 2017; Dummer et al., 2018; Ki et al., 2020), anti–polycystin-1 antibody (Nauli et al., 2008), antibody to polaris18 (Nauli et al., 2008), monoclonal antibody IFT-71 ab (Iomini et al., 2004), and anti-ARL13b antibody (Ki et al., 2020).
Lengths of endothelial primary cilia
Endothelial primary cilia have been investigated in different species and in different types of ECs located at aortas, arteries, arterioles, veins, and microvessels, as well as under different physiological and pathological conditions. Although their lengths vary from 0.5 to over 15 μm, endothelial primary cilia have a constant diameter of ~0.2 μm (Jin et al., 2014). The lengths of primary cilia of ECs in various vessel types and in various species are summarized in Tab. 1. They are either measured from the images presented in the literature (Russellpuleri et al., 2017; Nauli et al., 2008; Antal et al., 2017; Vo and Bols, 2016; Iomini et al., 2004; Ki et al., 2020; Dummer et al., 2018; Goetz et al., 2014) or directly cited from the literature (Abdul-Majeed and Nauli, 2011; Lim et al., 2015b).
The length of endothelial primary cilia varies among various types of ECs and at the different locations of the vascular system. As proposed by Resnick and Hopfer (2007), the longer the cilium, the more sensitive the cell becomes to flow changes. A longer cilium increases the torque, which is the cross product of the lever-arm vector and the force vector, leading to a renewed force equilibrium. The longer the cilium is, the less force is needed to bend the cilium and activate the Ca2+ influx (Dummer et al., 2016). In addition, with the length increase, the volume of the cilium increases, leading to potential concentration adaptations of ions, proteins, and signaling molecules, and resulting in biological responses. Increased length of the primary cilium results in higher strains on both the cilium structure and the intracellular components (Khayyeri et al., 2015). With the cilium elongation, the ciliary membrane also increases in its surface area. Since the composition of the ciliary membrane is highly regulated and contains a distinct population of receptors, increased membrane area might affect the number of available receptors for signaling. Taken together, differences in cilium length might regulate and fine-tune signaling in the cilium and in the cell.
The length of cilia can change during injury and inflammation. For example, the length of cilia increases when exposed to pro-inflammatory cytokines (Wann and Knight, 2012). This is consistent with the finding that the length of cilia in ECs of patients with pulmonary arterial hypertension (PAH) is usually longer than that of healthy ECs (Dummer et al., 2018). The EC seems to elongate its cilium to increase its mechano-sensitivity under low biophysical stimuli and to shorten its cilium to reduce the strains under the high stimuli (Khayyeri et al., 2015). High shear stresses may lead to endothelial cilia glutamylation and deciliation (Ki et al., 2020). The ciliary length varies in different types of ECs and also depends on the cultural conditions. Lim et al. (2015b) found that the length of the primary cilia on human umbilical vein endothelial cells (HUVECs) and that on human microvascular endothelial cells (HMECs) are distinct. Even on the same type of cells, the ciliary length varies when the culture media have different serum concentrations and under different culture periods.
Mechanical properties of primary cilia
Methods for determining mechanical properties of primary cilia
Besides structure, molecular composition and dimension, mechanical properties, e.g., stiffness or bending rigidity, are essential for the primary cilium to serve as a blood flow sensor and transducer. The mechanical properties of primary cilia can be determined by combining the bending behavior of the cilium recorded by the image-based techniques and the mathematical models for the force-deflection of cilia (Lim et al., 2015a). Briefly, imaging technology is used to quantify the deflection of primary cilia in response to the known flow. Then the presumed mathematical model for the cilia-fluid flow is employed to simulate the flow-deflection process and the mechanical properties of primary cilia (e.g., bending rigidity) can be determined by iteratively updating during each calculation in the model until the predicted bending shape matches the experimentally observed deflection (Lim et al., 2015a). Imaging technologies range from the application of light microscopes (Schwartz et al., 1997) to that of laser scanning confocal microscopes (Rydholm et al., 2010; Young et al., 2012; Downs et al., 2014), enabling the ciliary bending behavior to be observed from 2 to 3 dimensions. In addition to the in vitro sample, high-speed confocal microscopy has been applied to record the bending of the endothelial primary cilia under flow in the vascular system of zebrafish embryo in vivo (Goetz et al., 2014; Boselli et al., 2015). In their study, fluorescent transgenic zebrafish lines are used to provide simultaneous imaging of both the endothelial cilia and EC surface. The mechanical property, e.g., the bending rigidity of the cilium, was determined by matching the prediction of a mathematical model for the bending cilium with the observation by confocal microscopy (Goetz et al., 2014; Boselli et al., 2015).
The bending rigidity of the cilium can also be obtained by using an optical trap with a magnetic bead affixed to the primary cilium (Resnick, 2015). The optical trap can apply a compressive (buckling) force to the cilium, which relies on fewer parameters than the application of a shearing force (Resnick, 2016). Resnick (2015) measured the mechanical properties of a primary cilium by using an optical trap to induce resonant oscillation of the structure. Their data indicated that the primary cilium is not a simple cantilevered beam. Instead, the base of the cilium may be modeled as a nonlinear rotatory spring. Interestingly, their data implied that the ciliary base may be an essential regulator of mechano-transduction signaling. Compared to the static method, measuring dynamic responses of the cilium provides more information about the mechanical properties of the cilium as well as the cilium-fluid interaction. It also eliminates the need for the detailed shape fitting, which requires either side-on views of the cilium or insertion of fluorescent transmembrane proteins (Resnick, 2015).
The mathematical models for the cilium-fluid flow (or other applied forces) are necessary in determining the mechanical properties of the cilia. Schwartz et al. (1997) first employed the Euler-Bernoulli beam formulation and laminar fluid velocity to simulate the primary cilium under fluid flows. Later, Liu et al. (2005) considered a more precise fluid flow profile and allowed a small rotation at the base of the cilium near the EC surface. Downs et al. (2014) modeled the primary cilium based on the Euler-Bernoulli beam theory but further included novel features such as the ciliary cap, initial base rotation, and initial cilium bending profile. Young et al. (2012) studied the dynamics of the cilium under the Stokes flow using slender-body theory, with the cilium base modeled as a nonlinear rotational spring. Resnick (2015) later determined the mechanical properties of the primary cilium by employing an optical trap to induce resonant oscillations of the structure. Their results revealed that the primary cilium is not a simple cantilevered beam, its base rather behaves like a nonlinear rotator spring. The linear and nonlinear spring constants determined for the base of the primary cilium of mCCD (a mouse cell line derived from the cortical collecting duct) are listed in Tab. 2. Resnick (2016) also employed the optical trap method and found that hypoxia-inducible factor (HIF) stabilization by a pharmaceutical agent significantly weakens the primary cilia on MDCK cells. Khayyeri et al. (2015) established a finite-element model for the cell-cilium system, which is composed of nucleus, cytoplasm, cortex, microtubules, and actin bundles. Their model suggested that the length and stiffness of the primary cilia are responsible for the transmission of mechanical stimuli to the cytoskeleton. Most recently, Flaherty et al. (2020) modeled the axonemal microtubule as an anisotropic elastic shell with actomyosin-driven stochastic basal body motion and showed that the longer cilia are stiffer than the shorter ones, indicating a length-dependent persistence length for primary cilia.
Bending rigidity and spring constants of primary cilia
So far, almost all the studies investigating the mechanical properties of the cilia that we can find are on the epithelial cells or other types of cells except one on the endothelial cells. As described above, the bending rigidity and spring constants are the parameters charactering the mechanical properties of primary cilia, we thus list in Tab. 2 the values of these parameters for the endothelial cells and for the epithelial cells. From Tab. 2, we can see that the bending rigidity of primary cilia is in the similar range, 0.5–5 × 10−23 Nm2, for the endothelial and epithelial cells.
Primary Cilium as a Mechano-Sensor and Transducer to Blood Flows
Protruding from the middle of the apical surface of an EC into the vessel lumen, the endothelial primary cilium has the best position to sense the blood flow. It has thus been widely recognized as one of the prominent mechano-sensors and mechano-transducers, which could sense and respond to the blood flow and convert the blood flow induced mechanical stimuli into biochemical signaling through its transmembrane proteins and other accessories. In fact, primary cilia are linked to the EC cytoskeleton via the microtubule-organizing center. As a result, the torque exerted on the primary cilium by the blood flow can be transmitted throughout the cell. Meanwhile, there are many other mechano-sensors located on the EC, which can be affected and activated by the alteration of cytoskeletons caused by bending of primary cilia. Although the exact molecular mechanism by which primary cilia act as a blood flow sensor and transducer is still under investigation, the proper mechanical properties (reviewed in the section above) and adequate mechanical strength of the primary cilium is essential for this duty.
Primary cilia in flow sensing
It has been found that endothelial primary cilia participate in regulating actin organization, focal adhesion formation, directional migration, vascular development, and cell permeability via flow sensing (Jones et al., 2012; Goetz et al., 2014; Vion et al., 2018; Chen et al., 2017; Ma and Zhou, 2020). Human induced pluripotent stem cell (hiPSC)-derived ECs lacking cilia do not align to shear stress and have aberrant calcium influx upon shear exposure (Smith et al., 2018). The flow sensing of primary cilia varies with flow magnitudes and patterns. In zebrafish development, through primary cilia, ECs can sense extraordinarily low flow forces and discriminate between the very subtle variations in flow regimes generated along the developing vascular network (Goetz et al., 2014). It is also suggested that endothelial cilia have a critical role in vascular maturation by transducing low-magnitude mechanical stimuli into the molecular signals (Vion et al., 2018; Chen et al., 2017).
In contrast, high shear stresses (HSS) may cause disassembly of primary cilia both in vitro and in vivo. Ki et al. (2020) found that HSS-exposed cilia show stronger glutamylation intensity, which would contribute to the disassembly of EC primary cilia. They also found that HUVECs exposed to a low shear stress (3.96 dyn/cm2) have similar number of cilia to those of static control cells, but cells exposed to a high shear stress (19 dyn/cm2) have fewer number of cilia (Ki et al., 2020). Vion et al. (2018) showed that in confluent monolayers of HUVECs, about 15% of cells demonstrate an apical primary cilium under static conditions. This percentage decreases with increasing shear stress levels, dropping to 5% under HSS (20 dyn/cm2). In an in vivo study using zebrafish, Goetz et al. (2014) found that endothelial cilia are almost completely abrogated in zebrafish trunk and caudal vasculature, concomitant with increases in flow velocity and shear stress levels.
The generation of primary cilia, or ciliation, is highly dependent on the flow pattern rather than the flow magnitude. Egorova et al. (2012) discovered that only oscillatory flows could cause ECs to regain their cilia upon deciliation under high shear stress. Venous ECs or atheroprone regions within the vasculature, which are exposed to oscillatory low shear forces, are highly ciliated in vivo (Smith et al., 2018). Primary cilia are not uniformly distributed in aortic and arterial vessels. They prefer to locate at the vascular bifurcation and branching points, and curved and arch-shaped sites, where the flow patterns are disturbed and the flow levels are relatively low (van der Heiden et al. 2008; Dinsmore and Reiter, 2016; Eisa-Beygi et al., 2018; Zhang et al., 2019).
Molecular mechanism of flow sensing and transducing by primary cilia
As expected, many molecules should be coordinated for the primary cilia to perform their functions. By flow sensing, endothelial primary cilia can regulate actin organization, focal adhesion formation, directional migration, and cell permeability via heat-shock protein 27 (hsp27) pathway (Jones et al., 2012). Histone deacetylase 6 (HDAC6), a post-transcriptional regulator, can deacetylate α-tubulin and actin-remodeling protein cortactin (Kaluza et al., 2011). Inhibition of HDAC6 activity in human induced pluripotent stem cell (hiPSC)-ECs lacking cilia rescues cilia formation and restores mechanical sensing (Smith et al., 2018).
Fig. 3 demonstrates the major signaling pathway by which a primary cilium senses and transduces the blood flow induced forces (drag and shear) into ECs. Bending of a primary cilium by the drag and shear forces triggers the stretch-activated channels polycystin-1/polycystin-2 (PC1/2), inducing increase in [Ca2+] in EC cytoplasm, via Ca2+ influx and release from the internal store (AbouAlaiwi et al., 2009a; Besschetnova et al., 2010; Pala et al., 2018). Cytosolic Ca2+ activates calcium-dependent protein kinase (PKC) and forms complex with calmodulin (CaM) (Pala et al., 2018). The Ca2+-CaM complex can activate endothelial nitric oxide synthase (eNOS) through serine-threonine kinase/protein kinase B (Akt/PKB) signaling (Pala et al., 2018). The activated eNOS induces upregulation of nitric oxide (NO) (AbouAlaiwi et al., 2009a; Pala et al., 2018). NO derived from eNOS is an important vasodilator, which can regulate vascular tone and blood flow (Yen et al., 2015; Zhang et al., 2019). NO can also regulate endothelial permeability, control vascular smooth muscle proliferation, inhibit platelet aggression and adhesion, and inhibit leukocyte adhesion and vascular inflammation (Ma and Zhou, 2020; Yen et al., 2015; Zhang et al., 2016; Zhang et al., 2019). Using the ECs from the aorta of a mouse embryo in an in vitro experiment, Nauli et al. (2008) demonstrated that the wild type ECs responded to the shear stress with the increase in [Ca2+] and NO, whereas the ECs mutant in PKD1/PKD2 and Tg737 did not have the response (Nauli et al., 2008; AbouAlaiwi et al., 2009b). It is notable that increasing in [Ca2+] by tangential forces (drag and shear) is a cilia-specific mechanism, which is evidenced by the observation that cilia-genetic-deficient ECs still show increases in cytosolic Ca2+ in response to other types of mechanical forces (circumferential stretching force and manipulator touching of the EC apical surface), as well as pharmacological stimuli (acetylcholine treatment) (Nauli et al., 2008).
In addition to increasing cytosolic [Ca2+] acutely upon the flow, the antenna-like cilium is thought to amplify the cytoskeletal strain and exert a prolonged effect on the expression of shear responsive transcription factors, including Krüppel–like factor 2 (Klf2) and Krüppel–like factor 4 (Klf4) (Egorova et al., 2011; Egorova et al., 2012; Ten Dijke et al., 2012). In general, Klfs are induced by the HSS, and they coordinate a major part of the phenotypic response of ECs to shear forces (Egorova et al., 2011). It has been proposed that under HSS, the regulation of EC function and the establishment of a quiescent, anti-inflammatory, and anti-thrombotic phenotype are largely rendered through the family of Klfs (Shaaban and Duerinckx, 2000). Expression of Klfs is downregulated in areas where flows are low and disturbed (Egorova et al., 2012; Zhang et al., 2019), but the ciliated ECs exhibit more expression of Klfs than non-ciliated ECs (AbouAlaiwi et al., 2009a), indicating a potential role of endothelial cilia in the quiescent state of the healthy vessel.
Endothelial primary cilia also play a role in maintaining normal morphology of ECs lining vascular walls. Under fluid shear stresses (FSS), the wild-type ciliated ECs retain cobblestone morphology, while the non-ciliated ECs undergo endothelial-mesenchymal transition (EndoMT) and become spindle-shaped (Ten Dijke et al., 2012). It was found that Klf4 overexpression prevents flow-induced EndoMT in non-ciliated cells by maintaining the cobblestone morphology of ECs (Ten Dijke et al., 2012).
Besides morphological maintenance of ECs under blood flows, endothelial primary cilia contribute to EC directional migration and vascular stability. Vion et al. (2018) found that primary cilia strongly enhance EC sensitivity to bone morphogenic protein 9 (BMP9), particularly under low flows. Activin receptor-like kinase 1 (ALK1) is the receptor of BMP, which is essential for the EC polarization and migration against blood flows (Rochon et al., 2016). The BMP–ALK1–SMAD pathway is important for EC quiescence and the vascular stability (Rochon et al., 2016; Vion et al., 2018). ECs lacking primary cilia display reduced polarization against blood flows, especially at low and intermediate flows, and demonstrate a stronger migratory behavior and random regression of vessels (Vion et al., 2018).
Endothelial primary cilia also mediate Wnt, Sonic Hedgehog (SHH) activity and Notch signaling pathways during EC development and function (Guo and Wang, 2009). By detecting the expression and localization of several core proteins of the Wnt/β-catenin and Wnt/planar cell polarity (PCP) signaling pathways, Sheng et al. (2018) elucidated the relationship between the Wnt signaling pathways and ciliogenesis. By detecting the expression of PTCH1 and GLI1, reference genes for activating the Sonic Hedgehog (SHH) pathway, Vion et al. (2018) discovered that silencing either IFT88 or KIF3a leads to decrease in SHH activity. Chen et al. (2017) found that during vascular myogenesis in zebrafish, the primary cilia are responsible for recruiting vascular mural cell (vMC) through a foxc1b mediated Notch pathway in the arterial vessels of zebrafish embryos. Recently, Liu et al. (2019) demonstrated that via Notch signaling, the cilia modulate hematopoietic stem and progenitor cells (HSPC) development, especially in hemogenic endothelium (HE) specification in zebrafish embryos. HSPCs are capable of producing all mature blood lineages, as well as maintaining the self-renewal ability throughout life (Liu et al., 2019).
Redistribution and Reduction of WSS by Primary Cilia
Independent of mechano-sensing and transduction, due to its unique location, dimension, structure and mechanical properties, the primary cilium has recently proposed to redistribute and reduce the WSS under real physiological pulsatile flows, and thus play a protective role for the ECs lining the blood vessel walls. Cui et al. (2020) modeled the EC primary cilium base as a nonlinear rotational spring and applied an explicit immersed boundary-lattice Boltzmann method to numerically investigate the dynamics of primary cilium in pulsatile blood flows with two-way fluid–structure interaction. Their study predicted that due to the obstruction of the primary cilium, the WSS distribution on the cell surface no longer remains uniform as in the absence of cilia. The presence of the cilium reduces the overall level of the WSS, especially at the region near the cilium anchor point. Fig. 4 demonstrates this role of endothelial primary cilia in WSS shielding. In addition, primary cilia can do a periodic flapping during each cardiac cycle, which depends on the peak Reynolds numbers (Repeak) and Womersley numbers (Wo) (Cui et al., 2020). When subject to fluid flows with higher Repeak or lower Wo, the flapping amplitude, tip angular speed, basal rotation and maximum tensile stress all increase in cilia (Cui et al., 2020). Under lower Repeak and higher Wo, the flapping pattern in primary cilia transfers from two-side into one-side (Cui et al., 2020). They also predicted that under pulsatile flow conditions, the maximum tensile stress in the cilium can propagate periodically within a certain distance to the base instead of at the base all the time (Cui et al., 2020). Their findings show the potential protective role of cilia in vascular ECs by redistributing and reducing the blood-flow induced forces. The mild disturbance by the interaction between the cilia and fluid flow may also be good for mass transport near the vessel wall to bring the nutrients and remove the metabolic wastes.
The Role of Primary Cilia in Cardiovascular Diseases
Dysfunction of primary cilia induces many diseases including cardiovascular diseases. These diseases are collectively named as ciliopathies. Specifically, malfunction of endothelial primary cilia in shear forces (or tangential forces) sensing causes atherosclerosis, hypertension, and aneurysm formation. This section briefly summarizes the role of endothelial primary cilia in these diseases. Extensive reviews can be found in (Luu et al., 2018; Pala et al., 2018; Wang et al., 2021).
Atherosclerosis plaques preferentially occur at bifurcations, branch points and inner surfaces of arched arteries with relatively low and disturbed blood flows, where primary cilia are usually enriched (Hoi et al., 2011). Van der Heiden et al. (2006) investigated the correlation between the location and the frequency of endothelial primary cilia and atherogenesis. They found that primary cilia are located at the atherosclerotic predilection sites, where the flow is disturbed, in wild type mice and they occur on and around atherosclerotic lesions in apolipoprotein-E-deficient mice, which have significantly more primary cilia in the aortic arch than wild type mice. The reason for why there are more cilia in atherosclerosis-prone regions is unclear. But the lack of endothelial cilia in vascular branches results in significant upregulation of the expression of the genes encoding the pro-inflammatory cytokines, the inflammatory adhesion molecules, the macrophage marker, and the lymphocyte marker (Dinsmore and Reiter, 2016). These are the key factors causing atherosclerosis. They also reported that when the mice with genetic constitutive deletion of endothelial primary cilia were fed with a high-fat and high-cholesterol diet, compared to the wild type mice, they exhibited larger atherosclerotic plaques, accompanying with decreased levels of eNOS and Klfs but increased levels of pro-inflammatory cytokines. Some studies also show a protective role of primary cilia in atherosclerosis by attenuating vascular calcification in mice (Egorova et al., 2012).
Autosomal dominant polycystic kidney disease (ADPKD) is caused by mutations in PKD1 or PKD2 genes encoding the proteins polycystin-1 or polycystin-2. The patients with ADPKD also show many cardiovascular manifestations such as hypertension, cardiac valveabnormalities, and pericardial effusions (Luciano and Dahl, 2014; Lai et al., 2016; Diagbouga et al., 2018). Even with the transplanted kidney, the patients with ADPKD still suffer from cardiovascular complications (Luu et al., 2018), suggesting the essential role of primary cilia in maintaining normal cardiovascular functions. Impaired polycystins lead to abnormal intracellular signal transmission in ECs, resulting in reduced NO production and increased oxidative stress and asymmetric dimethylarginine (ADMA) release, followed by systemic vasoconstriction and hypertension. It also causes regional vasoconstriction, impaired renal blood flow, vascular remodeling, and renal hypoxia. These disorders increase the levels of hypoxia-induced factor 1α (HIF1α) and other angiogenetic factors and lead to increased apoptosis and fibrosis in the renal tissue in patients with ADPKD (Theodorakopoulou et al., 2019). Primary cilia further play a role in pulmonary arterial hypertension (PAH). Inflammatory cytokines could lead to a significant elongation of healthy endothelial cilia and anti-inflammatory cytokines, such as IL10, could block and reverse the cilia elongation (Dummer et al., 2018). However, compared to the healthy controls, primary cilia of ECs from the PAH patients are significantly longer. But they do not elongate further upon pro-inflammatory stimulation nor shorten under anti-inflammatory treatments (Dummer et al., 2018). Loss of ciliary length regulation upon cytokine stimulation is one of the endothelial dysfunctions in PAH (Dummer et al., 2018).
An aneurysm is a formation of a swelling in an area of a blood vessel that can rupture, leading to bleeding and possibly to death if occurs in vital organs such as brain. Aneurysm formation and rupture are one of the major complications associated with ADPKD because polycystin-1 and polycystin-2 at endothelial primary cilia are required in blood vessels for proper flow sensing and transduction. AbouAlaiwi et al. (2014) showed that abnormal function of mechano-sensory cilia leads to survivin downregulation, which is associated with abnormal ploidy formation and eventually contributes to vascular aneurysm phenotypes. Liu et al. (2018) found that in the mice model of intracranial aneurysm (IA), the number of primary cilia significantly decreases in the IA region.
Conclusions and Perspectives
The unique location, structure, mechanical properties and molecular compositions, all of them contribute to the function of endothelial primary cilia in maintaining normal cardiovascular system. As an antenna, the primary cilium can sense the changes in external blood flows and transduce them into ECs via various signaling pathways along the ciliary membrane, cilioplasm, axoneme, and at the basal body complex. Even without the signaling function, the endothelial cilium can still behave like a structural component to redistribute and reduce the flow generated wall shear stresses to protect the vessel wall. The disturbance near the vessel wall by the interaction between the cilium and fluid flow also favors the mass transport at the EC surface to bring nutrients from the bulk flow and remove the metabolic wastes. Lack of primary cilia or malfunction of cilia cause many cardiovascular diseases including atherosclerosis, hypertension, and aneurysm formation.
Although the molecular mechanism by which endothelial primary cilia perform their flow sensing and transducing has been partially elucidated, further investigation is highly expected. With the development of super high resolution optical microscopy, more detailed information for the structural (molecular) components of EC primary cilia can be revealed, especially when the ECs are alive, and the forces are known. As the force and stress distribution on a cilium and other structural and mechanical factors are beyond the capability of current experimental approaches due to the nanometer scales in its structure and the pulsatile nature of the real physiological flows, more sophisticated mathematical models and numerical approaches should be developed. By combining the model predictions with the experimental observations, the more detailed molecular mechanism underlying the cilium sensing-transducing can be elucidated under realistic physiological conditions. In addition, the interactions of primary cilia with other endothelial mechano-sensors and transducers also deserve investigation to give a more comprehensive understanding for how the ECs sense and transduce the external stimuli into internal signals for their functions.
The understanding for how the primary cilia function will also help to design proper therapies for cardiovascular diseases due to lack of cilia or malfunction of cilia. For example, targeted gene therapy for PKD1/PKD2 may reverse the ADPKD and related cardiovascular diseases such as hypertension and aneurysm. Pharmacological agents can be invented to either help in generating primary cilia or in informing alternative flow sensors and transducers on ECs in the absence of cilia. Since pulsatile flows at certain levels can help regenerate lost primary cilia, proper exercises can be designed for this purpose.
Author Contribution: The authors confirm contribution to the paper as follows: review outline by H. Yin and B. M. Fu; draft by H. Yin, revision by L. Wang, Y. Fan and B. M. Fu. All authors reviewed and approved the final version of the manuscript.
Funding Statement: This work was supported by Grants (11421202, 11572029) from National Natural Science Foundation of China and NIH 1UG3UH3TR002151.
Conflicts of Interest: The authors declare that there are no conflicts of interest to report regarding the present study.
|This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.|