The axoneme: the propulsive engine of spermatozoa and cilia and associated ciliopathies leading to infertility
- 1.7k Downloads
KeywordsBasal body Fibrous sheath Intraflagellar transport Microtubules Primary cilia Outer dense fibers
Central pair microtubule apparatus
Outer dense fiber(s)
- ODFs #1, #3, #5, #6, and #8
Refer to specific ODFs
One of the major proteins of ODFs
Polycystic kidney disease
Primary ciliary dyskinesia
To van Leeuwenhoek who first examined spermatozoa, ca. 1677 , and to Gray in 1955  who began to study invertebrate sperm motility, the sperm cell appeared seemingly simple—a head (containing the condensed haploid nucleus) and a flagellum that propels the head to the egg by the propagation of bending waves at nearly 100 Hz in water. By contrast, mammalian spermatozoa are astonishingly complex in their morphology and development from germ cells in the seminiferous epithelia, under the direction of Sertoli cells [3, 4], into their fully formed but functionally inactive state. The inactive spermatozoa then pass through a series of ducts lined by ciliated epithelia followed by maturation in the epididymis. In the female tract, spermatozoa undergo capacitation  and self-propulsion through the ciliated oviduct. For reproductive biologists, clinicians, genetic counselors, and general readers, this article will review the advances in our understanding of sperm flagellar and ciliary engines, the axoneme, and some of the defects that cause certain forms of infertility.
Basic structure of the axoneme from 1888 to the present
The flagellum of the mammalian spermatozoon
Molecular structure of the axonemal microtubules
To understand the structural arrangement of the molecular machinery of motility, it is important next to review the structure of the underlying microtubules. In 1973, Tilney et al.  demonstrated by thin-section EM that singlet microtubules and sperm flagellar A-tubules of nearly all species are composed of precisely 13 longitudinal protofilaments (Fig. 2). In rapid succession but not chronological order, microtubule protofilaments were shown to be end-to-end polymers of heterodimers of GTP-binding α- and β-tubulin, each with a mass of approximately 50 kDa [36, 37], that could be polymerized into synthetic microtubules in vitro, but with variable numbers of protofilaments ranging from 11 to 17 [38, 39], and off the ends of isolated centrioles . Based on the evolutionarily conserved primary sequence of tubulin , the 3D structure was determined for the αβ-tubulin dimer from zinc-tubulin sheets . Tubulin is posttranslationally modified in functionally important ways, e.g., acetylation, glutamylation, glycylation, and tyrosination, which function in the formation of flagellar doublet microtubules —see below. Microtubules are functionally polar structures, as shown by Allen and Borisy , with the distal end (subsequently referred to as the “plus” end of the microtubule) assembling tubulin faster in vitro than the proximal (minus) end. The accepted structural polarity of the microtubule is such that the α-tubulin subunit of a dimer is oriented toward the minus end, and the β-subunit toward the plus end, as inferred from Fan et al. , where the minus ends, but not the plus ends, can be labeled with a phage display antibody specific to α-tubulin.
Ciliary and flagellar doublet microtubules are significantly more complex than singlet microtubules both in structure and protein composition (Figs. 2 and 3). Early on, they were noted to be the most stable class of microtubules , and in nearly all cases, sperm flagellar doublet microtubules are irreversibly assembled into long, elastic elements. Each doublet microtubule is composed of a complete, 13-protofilament A-microtubule (like most cytoplasmic, singlet microtubules) and a partial 10-protofilament B-tubule, which assembles off the ends of the A-B-tubules of the basal body triplet microtubules [21, 35]. The numbering of these protofilaments is formally specified (Fig. 2d), due to their unique positions for the attachments of different motor and regulatory proteins . Using the technique of optical diffraction of electron micrographs, Amos and Klug  and Linck and Amos  analyzed the arrangement (or lattice) of tubulin subunits in flagellar doublet microtubules, describing the 4- and 8-nm axial repeats of the monomers and dimers, respectively, the 5-nm lateral spacing between protofilaments, the 3-start left-handed helix of monomers, and the lattice of tubulin dimers in the incomplete B-tubule. The lattice of the A-tubule remained uncertain. Recently, Maheshwari et al.  showed that the A-tubule has the same lattice as the B-tubule and cytoplasmic microtubules (Fig. 2c); currently, all microtubules in vivo are believed to have the “B-lattice.” Like cytoplasmic singlet microtubules, the A-tubule contains a longitudinal “seam” or helical discontinuity in the lattice (Fig. 2c), with the seam positioned at the outer junction of the A- and B-tubules, i.e., between protofilaments A9 and A10 or between A10 and A11 . The potential functions of the seam in the assembly of microtubules had been discussed earlier , but they have not been directly investigated. Doublet microtubules are highly stable and contain a high level of acetylated tubulin (cf. ). In addition, the formations of the inner and outer junctions of the A- and B-tubules are affected by posttranslational glutamylation and glycylation of tubulin . In another example, mutations in Arl13b, a G-protein in the Sonic hedgehog signaling pathway, directly or indirectly inhibits the closure of the inner A-B junction in embryonic primary cilia [50, 51]. It is presently unknown to us, whether mutations in these modifications and pathways affect sperm axoneme structure and motility and lead to infertility, in part because the phenotypes are usually lethal.
Despite much information about the biochemistry, structure, and developmental expression of tektins in ciliogenesis and spermatogenesis [60, 61, 62], their function remains elusive. Nevertheless, a thorough investigation by Tanaka and colleagues  showed that, in mice with a mutation in one of the tektin genes, the males were infertile and the sperm had defective motility and lacked some or all of the dynein arms (see below). There are also several reports of tektin mutations or a reduction in the amount of tektin in mouse  and human sperm [65, 66, 67, 68] being linked to asthenozoospermia (reduced sperm motility) or oligoasthenozoospermia (asthenozoospermia with a low sperm count), whereas one report found no proteomic difference in tektin levels in normozoospermic and asthenozoospermic humans . Several studies have also reported tektins to be localized to several mammalian sperm structures (refer to Fig. 4) by immunofluorescence microscopy, including (i) the principal piece containing the ODFs and FS and postacrosomal region , (ii) the principle piece and the basal body region but absent from the midpiece, and (iii) the endpiece containing only the axoneme, where presumably tektins should be found , and from the neck region to the tail piece but stronger in the midpiece (containing the mitochondrial sheath and ODF) . In fact, our own observations (unpublished) showed an intense staining of the FS of rat spermatozoa, using affinity-purified antibodies against sea urchin tektins. While tektins may be components of some of the periaxonemal elements, it seems highly unlikely that tektins would be present in all of the sperm structures. There remains the possibility of strongly binding, nonspecific anti-tektin staining of the accessory structures, and only weak or masked (nondetected) anti-tektin staining of the axoneme, or the presence of other proteins with similar epitopes as tektins (e.g., coiled-coil domains). A proteomic analysis also suggested that tektins may be present in the SDS-insoluble fraction of sperm flagella (e.g., ODFs and FS), but tektin filaments are known to be highly insoluble in metazoan cilia .
Axonemal doublet microtubules are the scaffold upon which the protein machinery for motility is attached and with which other proteins interact. Nevertheless, doublet microtubules are not passive elements: they undergo dynamic bending and twisting at acute angles, requiring enormous conformational changes in tubulin, and its associated proteins. There are generally two classes of the motile protein machinery: microtubule motors and regulatory structures and proteins.
Axonemal dynein motor proteins
The first class of microtubule motors, the dynein ATPases, was discovered and named by Gibbons  and shown to form the outer and inner arms observed earlier by Afzelius . Many cases of male infertility involving loss of sperm motility are caused by mutations in the genes for dynein and dynein-associated polypeptides [75, 76, 77, 78]. Dynein arms are essentially permanently anchored to each A-tubule and directed to the B-tubules of their next doublet microtubules in a clockwise direction to a viewer looking from the flagellar base to the tip  (Fig. 2), giving the axoneme enantiomorphic asymmetry or handedness. The importance of this asymmetry to the direction of ciliary motion (to be discussed) and its role in embryonic development may sometimes be misunderstood, as in articles where the handedness is incorrectly diagrammed as being clockwise when viewed from the tip to the base, e.g., . For species in which the central pair microtubule apparatus does not rotate (see below), this clockwise handedness allows for the unambiguous numbering of doublet microtubules around the axoneme (see also Fig. 4), with a line perpendicular to the plane of the fixed central pair passing through doublet #1, and between doublets #5 and #6. Since their discovery, dyneins have been studied to the greatest degree in Chlamydomonas, which possesses 16 genes for dynein heavy chain polypeptides [79, 80], with the masses of the heavy chains being approximately 500 kDa. In this organism, mutants with motility defects (typically called pf-mutants, for paralyzed flagella) can readily be generated and analyzed biochemically and structurally to identify the molecular defects, and therefore, their functions. The outer arms appear to be identical, with each arm (with a mass of ~1.2 mDa) consisting of three dynein heavy chain subunits in protist flagella and two heavy chains in vertebrate sperm, arranged along the A-tubule (for the sea urchin axoneme, see Fig. 3) . Each dynein arm also contains several intermediate and light chain polypeptides. The inner arms are more complex, consisting of a series of 11 dynein heavy chains and associated polypeptides, arranged in a complex but repeating pattern along the A-tubule (Fig. 3 and below).
The mechanism of dynein-based motility was determined in a series of studies beginning with Satir, who showed that in ciliated epithelia, that were rapidly fixed during active beating, the doublet microtubules are preserved in positions where they have slid past each other . Gibbons and colleagues then demonstrated that this sliding was actively driven by dynein ATPase motors, by first demembranating the sperm with mild Triton detergent [82, 83] and digesting the axonemes with trypsin to break the presumably elastic connections between doublet microtubules . Sale and Satir  followed by showing that the sliding was unidirectional, with the arms tightly anchored to their A-tubule moving in a minus-end direction along the adjacent B-tubule of the next doublet tubule; hence, the definition of dyneins as being minus-end-directed motors. Brokaw elegantly demonstrated sliding and measured the sliding oscillations in demembranted, reactivated sea urchin sperm by attaching gold particles to the exposed doublet microtubules and observing the oscillations of the particles as the sperm were swimming . Another dynein, involved in a different form of flagellar motility, will be discussed later.
Dyneins are mechanochemical force-transducing enzymes that convert the energy from ATP hydrolysis into a shear force between adjacent doublet microtubules. The dynein cross-bridge cycle has been studied biochemically  and by cryo-electron tomography , which captures the conformational states of the dynein arm in its cycle. In general, dynein arms anchored along one A-tubule bind ATP, which dissociates them from their prior attachments (cross-bridges) to the B-tubule of the adjacent doublet tubule. After hydrolysis, the dynein arms rebind to the B-tubule. The release of the products (ADP and inorganic phosphate) causes a conformational change in the dynein arm (acting as a lever arm), leading to a shear force, driving the A-tubule in a minus direction along the adjacent B-tubule. In this manner, dynein acts in a manner analogous to muscle myosin interactions with actin filaments. In a mechanism not fully understood but regulated by other axonemal structures (see below), dynein cross-bridges are presumably sequentially activated along the A-tubule from the base/minus to tip/plus ends of the microtubules (like falling dominos).
Both outer and inner dyneins arms are attached along their A-tubule with unique spacings, and they have unique functions. The outer arms form a single row with arms spaced at 24-nm intervals (their axial periodicity). The inner arms consist of several morphologically distinct subunits arranged in a slightly staggered pattern along the A-tubule, with subrepeats of an overall 96-nm axial periodicity (Fig. 3) . These long repeats and different configurations could explain why dynein arms (especially inner arms) sometimes appear absent or reduced in numbers even in normal human sperm. One of these inner arm subunits is the dynein regulatory complex, discovered by Piperno  and studied in great detail since then [12, 90]. The inner dynein arms are the primary motors that generate the amplitude and waveform of the propagated bends, whereas the outer arms appear only to provide additional power to increase the speed of wave propagation [83, 91].
Axonemal bend formation and propagation
First distinctly observed by Hopkins , radial spokes each consist of a spoke head and stalk and have complex axial spacings [13, 96, 97] (Fig. 3). The absence of radial spokes leads to axonemal paralysis and in sperm, infertility. In Chlamydomonas spokes consist of approximately 16 polypeptides . In Tetrahymena protozoa, invertebrate cilia, and rat spermatozoa, radial spokes are arranged in a single row with a triplet repeat along the A-tubules, in a polar orientation from the minus to the plus end of the A-tubule,; i.e., a 32-nm spacing between spokes #1 and #2, 24 nm between spokes #2 and #3, and 40 nm between spokes #3 and #1 of the next spoke triplet, all adding up to an overall axial repeat of 96 nm (Fig. 3). In Chlamydomonas, spoke #3 appears to be missing, yielding an alternate spacing of 32 nm between the two remaining spokes and 64 nm between spoke pairs; however, it was recently discovered  that Chlamydomonas does have the basal structure in place of spoke #3 (that was presumably lost in the evolution of that species). Thus, the fundamental axial periodicity of all spoke “triplets” is 96 nm, with radial spoke #2 being in register with the nexin-DRC , implying that this registration is essential for the flagellar machinery to work correctly. Radial spokes and their interactions with the central pair microtubules (see below) have been studied in great detail at the molecular level [99, 100, 101, 102, 103, 104].
Evolutionarily conserved axonemal spacings
The evolutionary significance of the longitudinal spacings of the different structures associated with A-tubules is that they are all multiples of the tubulin dimer repeat of 8 nm, i.e., outer dynein arms (24 nm), inner dynein arms (subrepeats of 96 nm), the nexin-DRC (96 nm), radial spokes (32 + 24 + 40 = 96 nm), and in addition, several proteins located along the lumen of the A- and B-tubules [10, 11] repeating at 8 and 16 nm. This means that they must all fit onto the tubulin lattice (Fig. 3); however, Chlamydomonas expresses only one α-tubulin isoform and one β-tubulin isoform ; thus, tubulin alone probably cannot account for the complex axial spacings nor for the unique pattern of protein structures attached (circumferentially) around the A-tubules. The circumferential asymmetry may be generated by the basal body template but may also be determined by a unique position of asymmetry in the A-tubule itself. Docking proteins have been identified for outer dynein arms, but that only begs the question of how the docking proteins “know” where to attach around the A-tubule. The only known structural-chemical features of the A-tubule wall that might set up a circumferentially asymmetric scaffold are the “seam” positioned at the outer A-B junction  and the stable 3-protofilament Ribbon with which the tektin filament is associated , but the seam and the position of the Ribbon and tektin filament could in turn be determined by the basal body. As for the complex longitudinal spacings along A-tubules, a ruler protein complex (comprised of FAP59 and FAP172) has been discovered that in Chlamydomonas specifies the 32/64-nm axial spacing of the radial spokes and the 96-nm spacing of the nexin-DRC . Curiously, tektins and tektin filaments have many of the axonemal repeats (e.g., 4, 8, 16, and 48 nm) and were speculated to be a ruler , but tektins are absent in Chlamydomonas mutants that retain the 96-nm radial spoke periodicity .
The central pair microtubule apparatus
The central pair microtubules function to regulate doublet microtubule sliding and potentially taxis and reversal of beat. The absence of the central pair from normally motile cilia and flagella leads to axonemal paralysis and in sperm, infertility; however, they are not essential for motility in all species, because in sperm flagella and motile cilia of some species, the central pair is absent altogether or is replaced by a nonmicrotubule core [20, 107]. Earlier studies of the ultrastructure of the central pair in squid and rat spermatozoa are best summarized elsewhere [29, 97], showing that the two microtubules are structurally different, and that each has two different rows of projections repeating along their respective microtubule with repeats of 16 and 32 nm (again, multiples of the tubulin dimer repeat). The central microtubules and their assemblage of projections act as a fixed unit and are thus referred to as central pair microtubule apparatus (CPMA). More advanced studies with cryo-electron tomography reveal the structure in great detail [104, 108]. As doublet tubule sliding takes place, the radial spoke heads move past and transiently interact with the CPMA, involving phosphorylation and dephosphorylation events that regulate interdoublet sliding, bend formation, and bend propagation [96, 100, 101, 102, 103, 104]. Remarkably, the CPMA of protozoa have been shown to actually rotate (i.e., spin) within the nine doublet microtubules, perhaps signaling which and when specific doublet tubules slide , and this function could be related to these cilia and flagella being able to reverse the direction of beat. However, in metazoan 9 + 2 cilia (e.g., ctenophore ciliary comb plates which do reverse direction) and spermatozoan 9 + 2 flagella, the CPMA does not rotate  and the axonemes initiate their bends and generally beat within a plane that is perpendicular to a line passing through the central pair microtubules, as originally observed by Fawcett and Porter ; this feature would seem to produce a more effective propulsive force. Although spermatozoa do not reverse the direction of their beat, they do undergo taxis (see below).
Regulation of oscillatory bends
The axoneme is a curious and intricate machine. The axoneme appears at first glance to be symmetrical, but as mentioned earlier, it has enantiomorphic asymmetry or handedness, where viewed from the base to the tip, the dynein arms point in a clockwise direction toward their adjacent doublet microtubule (Figs. 2 and 3). In some species, certain of the nine doublet microtubules also have specialized structures that the other doublets lack, e.g., a fixed “dynein” bridge between doublet tubules #5 and #6. Furthermore, as mentioned earlier, mammalian spermatozoa possess major accessory structures [20, 27], e.g., ODFs and the FS (Fig. 4), with doublet tubules #3 and #8 anchored to the longitudinal columns of the FS, making doublet tubules #3 and #8 incapable of being moved by their adjacent doublet tubules #2 and #7, respectively (see Fig. 6). These asymmetries and accessory structures manifest themselves in the oscillatory motions of cilia and flagella, which have a principal bend (or effective stroke) and reverse bend (or recovery stroke) . Many cilia and flagella beat in planar waves, but in several species, they beat out of the plane with a helical or semi-helical waveform [17, 111]. Some authors have related the left-handed form of this helical rotation, and the clockwise orientation of the dynein arms, to the direction of fluid flow, e.g., embryonic nodal cilia . However, the correlation between dynein arm orientation, the helical handedness of rotation, and the ultimate effect these might have on ciliary and flagellar function have been called into question in a recent analysis by high-speed holographic microscopy of malaria parasites . Regardless, it is impossible for all doublet microtubules around the axoneme to slide in the same direction simultaneously, and it has been partially demonstrated that the doublets on one side of the axoneme (i.e., doublets 1–4, and their accompanying dense fibers in sperm) slide to produce bending in one direction (while the other doublets are inactive), and that the other doublets (i.e., 6–9) slide to produce bending in the other direction . According to the Geometric Clutch hypothesis , these bends appear to generate a transverse force leading to a compression of one side of the axoneme, bringing those doublet microtubules close enough together for their dyneins to actively slide, while the doublets on the other side of the axoneme remain too far apart for dynein-microtubule interactions to occur (Fig. 6); this transverse force oscillates between the two sides of the axoneme to produce propagated bending waves. Computer models take into account as many factors as possible and closely simulate the physical dynamics of flagellar wave propagation [93, 114].
Having a propulsive engine is only useful if it can be guided. Sperm and most ciliated and flagellated protists and metazoans have mechanisms to alter the motion of their axonemes in order to move in a favorable direction, i.e., taxis. These mechanisms include chemotaxis, rheotaxis, thermotaxis, and phototaxis, with the first three occurring in mammalian sperm in order to move toward the ovum, but all four share similar signaling pathways. These topics have been discussed elsewhere and will only be mentioned here briefly. The most well understood are phototaxis in Chlamydomonas  and chemotaxis from studies of marine invertebrate (e.g., sea urchin) sperm , where chemotaxis operates at distances of less than a millimeter and is recapitulated in mammalian sperm [116, 117, 118, 119]. When a chemical attractant isolated from eggs binds to receptors on the sperm flagellar membrane, it causes an influx of calcium into the sperm cytoplasm through membrane calcium channels (polycystins?—see below), activated in mammals by progesterone and prostaglandins; mutations in the channel protein result in mammalian infertility . The rate of increase in intracellular calcium (from pCa 8 to pCa4) is followed by a G-protein signaling cascade in conjunction with a brief depolarization of the sperm membrane and a rise in cAMP levels. These signals are transmitted to, and stimulate the axoneme, and evidence implicates phosphorylation of radial spoke/CPMA interactions that relay to the phosphorylation of inner dyneins and nexin-DRC, which according to Lindemann and colleagues  may determine which doublet microtubules on which side of the axoneme will slide in a given sequence (Fig. 6); dephosphorylation presumably returns the axoneme to it prestimulated state. Less well understood is how these events alter the direction of sperm swimming. The excellent article by Seifert, Kaupp, Strünker and colleagues  suggests that sperm chemotaxis is analogous to bacterial chemotaxis, where in favorable conditions (steady or increasing concentrations of attractant) sperm swim in straight lines, and in less favorable conditions (lower attractant concentrations), they briefly randomize their swimming direction, and repeat this process until a favorable conditions allows them to again swim linearly, e.g., toward the egg. Nearly similar mechanisms are involved in thermotaxis (in the range of 31 to 37 °C) and rheotaxis (movement against a fluid flow), both of which have been identified and tested in mammals, principally mice and rabbits, and in vitro for human sperm .
In addition to the mechanism of propulsive force for sperm movement (axonemal bends and wave propagation), flagella and cilia have a second form of motility termed intraflagellar transport (IFT), as first reported by the laboratories of Witman, Pazour, Rosenbaum, and Yoder [120, 121, 122, 123], and studied more recently in its astonishingly widespread effects on vertebrate development (see below). IFT is independent of propulsive axonemal motility (i.e., it occurs in nonmotile cilia and flagella) but does require the axoneme to be present. IFT involves both a unique form of dynein and a second class of microtubule motors, the kinesins. Kinesin motors (which are plus-end motors) move their specialized protein cargo (“IFP particles” containing ciliary precursors and signaling molecules) from the cell body to the flagellar or ciliary distal tip along the doublet microtubules in the region between them and the membrane, while IFT-dyneins (minus-end motor) move other cargo from the distal tip back to the cell body. IFT, however, has not so far explained how the massive amount of tubulin required for axoneme assembly is transported to the growing distal tip or for tubulin turnover in terminally differentiated axonemes [124, 125]. IFT has been thoroughly investigated in Chlamydomonas flagella and vertebrate (mouse) cilia, and is essential for ciliary and flagellar assembly and for signaling pathways (e.g., Hedgehog, PDGF, and Wnt pathways). To our knowledge, IFT has not been studied in mature mammalian spermatozoa; however, it is entirely expected that IFT would be essential for the formation of the mammalian sperm tail during spermatogenesis, and IFT could perhaps function in signaling events during capacitation or later during guided propulsion toward the egg. We are not aware of such studies in mammalian sperm, but in Drosophila with a disrupted polycystin-2 gene (see below), the sperm are motile but effectively sterile, because they fail to enter the female sperm storage organ . Finally, it is important to understand that entry of molecules from the cytoplasm into the ciliary/flagellar environment, and their exit from it, are regulated by a cytoplasmic compartment termed the “ciliary gate” or “ciliary pore,” located between the basal body and the plasma membrane .
Ciliopathies: pathologies of cilia and flagella
The original term suggested for these ciliary and flagellar pathologies was “immotile cilia syndrome” , but the term primary ciliary dyskinesia (PCD) was later adopted to cover a wider range of “ciliopathies” of motile cilia [137, 138], including cilia that are not completely immotile or that are abnormally motile. As the name implies, the term PCD is a subcategory of broader ciliopathies, because it does not refer to cases of the complete absence of cilia, or to defects in normally nonmotile cilia, e.g., auditory hair cell kinocilia (in deafness), olfactory cilia (in anosmia), retinal photoreceptor connecting cilia (in retinal degeneration), and kidney cilia (in polycystic kidney disease)—see below. What may confuse some readers is that “primary” in PCD does not refer to “primary cilia” (which are “short, single, nonmotile” ), but to the fact that PCD is a primary genetic defect, rather than a secondary, acquired condition (note: “nonmotile” refers to the fact that primary cilia do not propagate dynein-depending bending waves; it does not apply to IFP or to deflections of cilia produced by cytoplasmic actin-myosin attached to the basal bodies). To avoid confusion here, we will use the more inclusive term, ciliopathies.
Following the discovery of the cause-effect of missing dynein arms and dynein-associated mutations in nonmotile sperm, additional defects were found not only in human sperm flagella but in human respiratory (and other) epithelial cilia, illustrating common mechanistic denominators between cilia and flagella, but also differences between them. Additional axonemal defects (Fig. 7) include the lack of the central pair microtubule apparatus (CPMA), the lack of radial spokes, missing or transposed doublet microtubules, ciliary aplasia, and the abnormal migration of basal bodies to the apical cell surface [140, 141, 142, 143, 144]. The absence of the missing CPMA is noteworthy. While sperm require the CPMA for motility, embryonic nodal cilia have no central pair (i.e., a 9 + 0) but are motile, having a rotary motion that is responsible for left-side rotation of the developing cardiac sac. When nodal cilia are immotile or not present, heart orientation occurs randomly, with about 50 % of patients having their hearts positioned on the right . Thus, it would seem obvious that normally motile 9 + 0 axonemes possess a different mechanism of initiating and regulating axonemal motility than do normally motile 9 + 2 axonemes (e.g., sperm flagella), perhaps due to differences in their nexin-dynein regulatory complexes (DRC)—recalling that suppressor mutants of the nexin-DRC restore motility to nonmotile mutants lacking radial spokes [12, 89, 90]. Since the nexin-DRC complex holds the nine doublet microtubules together , it would be worth knowing if nexin-DRC also exists in normally nonmotile cilia (e.g., 9 + 0, nonmotile cilia of the kidney and retina).
Still, some defects present in sperm flagella are not apparent in cilia, and vice versa, reflecting differences between cilia and flagella (e.g., sperm flagella possess periaxonemal structures, while cilia do not), and/or differential regulation of axonemes in somatic and germ cells. Of the various sperm flagellar periaxonemal elements, the ODFs and FS (Fig. 4) do not appear to contribute actively to motility but are thought to dampen the amplitude of flagellar waves for movement through more viscous environments of the oviduct, by allowing longer stretches of dynein arms to generate and propagate bending waves . In addition, the relative orientations of the CPMA, the FS, and the especially thick ODFs #1, #5, and #6, limit the spermatozoan to beat largely in a plane perpendicular to a line through the central pair microtubules and the columns of the FS (Fig. 4); this plane of beat may be more efficient for the propulsion the spermatozoan through the oviduct. In addition, splice variants of ODF2 gene/protein homologous with cenexin are associated with centrioles and essential for centriologeneis and ciliogenesis [33, 34, 146]. While not necessarily active, motile elements, ODF and FS are essential for sperm motility, as demonstrated by ultrastructural defects that occur leading to male infertility [147, 148, 149, 150] (Fig. 7). Sperm tails may appeared thick, short, or irregular, or duplicated due to the failure of neighboring spermatids to separate their cytoplasmic bridges. Besides the lack of dynein arms and/or the CPMA (Fig. 7c, d), other anomalies include missing, misplaced, or aberrant doublet microtubules and their associated ODF, e.g., 8 + 0 axoneme (Fig. 7f), and widespread disarray of the axoneme and FS (Fig. 7e–g). The latter phenotype was characterized as dysplasia of the fibrous sheath or DFS [130, 131, 132, 133], owing to the remarkable participation of fibrous sheath abnormalities and its origin from a dysplastic development of the sperm tail cytoskeleton. DFS has family incidence, associates to classical forms of PCD, and does not respond to any therapies. A genetic origin was suggested for PCD-DFS (and its combinations) by reports of mutations-deletions in genes encoding axonemal and periaxonemal proteins [151, 152]. Sperm flagellar disorders are unique among ciliopathies, where the periaxonemal structures only appear in the adult, and axonemal defects in the embryo are often lethal or compromise the adult to the point that reproductive function is never reached. Certainly, some of the defects in the periaxonemal structures would be due to mutations in the genes for and regulation of the FS and ODFs, and other defects would be due to defects in the axoneme which acts as a scaffold for the FS and ODF assembly.
Given the structural and functional similarities between sperm flagella and epithelial cilia, and given that the male reproductive tract also contains ciliated epithelia in the rete testis and efferent ducts (as well as in their embryonic progenitor cells), one should ask how defects in these cilia might impact male infertility. There are at least two lines of investigations that address this point. The first of these relates male infertility to polycystic kidney disease (PKD). In the normal kidney specific epithelial cells of uriniferous tubules possess a single, primary, nonmotile, 9 + 0 cilium that projects into the lumens of the uriniferous tubules and collecting ducts. These cilia sense fluid flow and respond via calcium influx through channels containing the transmembrane proteins, polycystin-1, and polycystin-2 ; polycystin gene PKD1 also interacts with the Tgf-β/Bmp signal transduction pathway. PKD has been shown to be directly linked to mutations in polycystin genes . In PKD, cilia-based cell signaling is compromised, growth control is lost, and the normal tubular epithelia forms spherical cysts. There is a high correlation between PKD and male sterility. Some human PKD patients have 9 + 0, nonmotile sperm , some patients have cysts in the seminal vesicle , and in mice with disrupted PKD2 genes malformations occur in the testis, the mesonephric ducts (embryonic origin of the efferent ducts) and the epididymis, with resulting male sterility [157, 158]. The second line of investigation relates male infertility to retinal degeneration, which involves the retinitis pigmentosa GTPase regulator gene (RPGR) and results from a failure of intraflagellar (ciliary) transport (IFT) of newly synthesized material from the cell body along the photoreceptor connecting cilium (a primary, 9 + 0, nonmotile axoneme) to the outer rod segment (see ). In transgenic mice overexpressing RPGR, which were sterile, there was either a complete absence of flagella, or there were defects in the assembly and organization of the axoneme, FS and ODFs, indicating a role of RPGR and IFT in sperm tail formation . Oddly, ciliopathies that would be expected to affect oviduct cilia, only slightly impair the female reproductive system .
In conclusion, ciliopathies are widespread human diseases and disorders, in many cases affecting male fertility. Most if not all human embryonic cells possess cilia (motile and nonmotile) and most adult cells retain them and the testis makes (sperm) flagella. Currently, the manifestations of ciliopathies are known to include anosmia, bone, cartilage and tooth development, brain development and disease (e.g., hydrocephaly, juvenile myoclonic epilepsy) and many other neurological abnormalities, deafness, hyperphagia, male infertility, obesity, hepatic, pancreatic and splenic diseases, polycystic kidney disease, polydactyly, respiratory diseases, retinal degeneration, and situs inversus. Clearly, this information indicates a highly complex set of interacting genetic pathways controlling the assembly of cilia and flagella, their motile functions, and their signaling functions. Mutations occurring downstream may have only a single effect on development (e.g., the sperm flagellar defects and infertility), while other mutations upstream will have global effects (e.g., as in Kartagener syndrome and Bardet-Biedl syndrome), which will inevitably impact sperm development and fertility. The genetic pathways and cell signaling pathways associated with cilia are beyond the scope of this article and readers are referred to the cited literature and earlier reviews [26, 75, 76, 77, 127, 161, 162, 163, 164, 165].
For part of the work cited here, the authors gratefully acknowledge funding to R.W. Linck from the National Institutes of Health (USPHS Grants GM 21527 and GM35648) and the National Science Foundation (Awards DIR-9113444, DBI-9602237, and 1024963); to H. Chemes from the National Research Council of Argentina (CONICET, PIP 4584 and 5479) and ANPCyT (PICT 9591). For Fig. 4, we thank Daniel Stoddard and Dr. Jianfeng Lin from the laboratory of Dr. Daniela Nicastro (Brandeis University and the University of Texas Southwestern Medical Center). Figure 7 originally published in Chemes et al. (1998, Human Reproduction 13, 2521–2526) and Chemes and Rawe (2003, Human Reproduction Update 9, 405–428), European Society of Human Reproduction and Embryology, reproduced by permission of Oxford University Press/Human Reproduction. We are grateful to Don Fawcett who had provided his prints with which to compose Fig. 4. We especially wish to acknowledge his immense scientific contributions and for the special role that he played in our scientific development.
- 1.van Leeuwenhoek, A. http://lensonleeuwenhoek.net/.
- 2.Gray J. The movement of sea-urchin spermatozoa. J Exp Biol. 1955;32:775–801.Google Scholar
- 7.Grigg GW, Hodge AJ. Electron microscopic studies of spermatozoa. Aust J Sci Res. 1949;2:271–86.Google Scholar
- 17.Sleigh MA. Cilia and flagella. New York: Academic; 1974. p. 500.Google Scholar
- 20.Baccetti B, Afzelius BA. The biology of the sperm cell. Monogr Dev Biol. 1976;1976:1–254.Google Scholar
- 28.Fawcett DW. The cell. Philadelphia, London, Toronto: W.B. Saunders Company; 1981.Google Scholar
- 61.Norrander J, Larsson M, Stahl S, Hoog C, Linck R. Expression of ciliary tektins in brain and sensory development. J Neurosci Off J Soc Neurosci. 1998;18:8912–8.Google Scholar
- 62.Larsson M, Norrander J, Graslund S, Brundell E, Linck R, Stahl S, et al. The spatial and temporal expression of Tekt1, a mouse tektin C homologue, during spermatogenesis suggest that it is involved in the development of the sperm tail basal body and axoneme. Eur J Cell Biol. 2000;79:718–25.PubMedCrossRefGoogle Scholar
- 64.Roy A, Lin YN, Agno JE, DeMayo FJ, Matzuk MM. Absence of tektin 4 causes asthenozoospermia and subfertility in male mice. FASEB J Off Publ Fed Am Soc Exp Biol. 2007;21:1013–25.Google Scholar
- 66.Bhilawadikar R, Zaveri K, Mukadam L, Naik S, Kamble K, Modi D, et al. Levels of Tektin 2 and CatSper 2 in normozoospermic and oligoasthenozoospermic men and its association with motility, fertilization rate, embryo quality and pregnancy rate. J Assist Reprod Genet. 2013;30:513–23.PubMedCentralPubMedCrossRefGoogle Scholar
- 68.Zhang S, Zhang J, Xianping D, Zhang S, Chen H, Jing Y. Asssociation of polymorphisms in tektin-t gene with idiopathic asthernozoospermia in Sichuan, China. J Assist Reprod Genet. 2015. doi: 10.1007/s10815-015-0617-9.
- 73.Gibbons IR. Chemical dissection of cilia. Arch Biol. 1965;76:317–52.Google Scholar
- 78.Ben Khelifa M, Coutton C, Zouari R, Karaouzene T, Rendu J, Bidart M, et al. Mutations in DNAH1, which encodes an inner arm heavy chain dynein, lead to male infertility from multiple morphological abnormalities of the sperm flagella. Am J Hum Genet. 2014;94:95–104.PubMedCentralPubMedCrossRefGoogle Scholar
- 79.King SM, and Kamiya R. Axonemal dyneins: assembly, structure and force generation. The Chlamydomonas Sourcebook. 2nd ed. 3:131–208, 2009.Google Scholar
- 103.Yang P, and Smith EF. Radial spokes of Chlamydomonas flagella: genetic analysis of assembly and function. The Chlamydomonas Sourcebook. 2nd edn. Cell Motil Behav. 2009;3: 209–234.Google Scholar
- 139.Alberts B, Johnson A, Lewis J, Morgan D, Raff M, Roberts K, et al. Molecular biology of the cell. 6th ed. New York, NY: Garland Science; 2015. p. 1342.Google Scholar
- 143.de Santi MM, Gardi C, Barlocco G, Canciani M, Mastella G, Lungarella G. Cilia-lacking respiratory cells in ciliary aplasia. Biol Cell Auspices Eur Cell Biol Organ. 1988;64:67–70.Google Scholar
- 150.Escalier D, David G. Pathology of the cytoskeleton of the human sperm flagellum: axonemal and peri-axonemal anomalies. Biol Cell Auspices Eur Cell Biol Organ. 1984;50:37–52.Google Scholar
- 164.Tobin JL, Beales PL. The nonmotile ciliopathies. Genet Med Off J Am Coll Med Genet. 2009;11:386–402.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.