The Sperm Centrioles

Abstract

Centrioles are eukaryotic subcellular structures that produce and regulate massive subcellular cytoskeleton superstructures. They form centrosomes and cilia, regulate new centriole formation, secure cilia to the cell, and regulate cilia function. These basic centriolar functions are executed in sperm cells during their amplification from spermatogonial stem cells during their differentiation to spermatozoa, and finally, after fertilization, when the sperm fuses with the egg. However, sperm centrioles exhibit many unique characteristics not commonly observed in other cell types, including structural remodeling, centriole-flagellum transition zone migration, and cell membrane association during meiosis. Here, we discuss five roles of sperm centrioles: orchestrating early spermatogenic cell divisions, forming the spermatozoon flagella, linking the spermatozoon head and tail, controlling sperm tail beating, and organizing the cytoskeleton of the zygote post-fertilization. We present the historic discovery of the centriole as a sperm factor that initiates embryogenesis, and recent genetic studies in humans and other mammals evaluating the current evidence for the five functions of sperm centrioles. We also examine information connecting the various sperm centriole functions to distinct clinical phenotypes. The emerging picture is that centrioles are essential sperm components with remarkable functional diversity and specialization that will require extensive and in-depth future studies.

Graphical Abstract

1. Introduction

Sperm centrioles were discovered in the late 19^th century during the early studies on fertilization, yet even now, two centuries later, much remains unclear about their roles. It is accepted that sperm centrioles have essential functions in mitotic and meiotic cell divisions early in sperm formation (spermatogenesis), in forming the sperm tail, in linking the sperm tail to the head, and in establishing the microtubule organization center of the zygote. Centrioles are also thought to control sperm movement and aid in zygote cleavage after fertilization, but the mechanisms surrounding these functions are unclear. Here, we aim to address these suggested functions of the sperm centrioles, the clinical phenotypes associated with their maladies, and recent evidence produced using genetic studies involving humans, model animals, and microscopy. We will start with an overview of the major events of sperm centriole research. Then, we will discuss five main functions of sperm centrioles and their various clinical significance.

2. Historical Perspective

Eduardo van Beneden and Theodor Boveri initially identified the centriole during their studies of egg fertilization in sea urchin and Ascaris (a parasitic nematode worm) between 1876 and 1900 (Boveri, 1887; Van Beneden, 1876). Boveri states in his classic work, “About the contribution of the spermatozoon to the division of the egg”, that “the spermatozoon introduces a centrosome into the egg. After the two nuclei have fused, the simple central body divides (i.e., the centriole), as in every cell division, into two daughter centrosomes, which give rise to the formation of the first furrow spindle (Page 151).” Boveri adds, “I believe that the observations available so far, decisively speak for the derivation of both poles (i.e., the centrioles) from the spermatozoon (Page 157).”

In these classical studies, cell staining suggested that the sperm provides a dot-like structure inside the centrosome. This dot was later named the centriole (Boveri, 1900). It was concluded that the centriole is a permanent cell organelle, self-replicates, is passed on from the mother cell to the two daughter cells, and acts as an organizational center for cell division. These original findings are still apparent today and apply to centriole inheritance during the sexual reproduction of most animals with some notable exceptions, such as mice and other murines.

Around the same time, Henneguy (1898) discovered in insect spermatocytes that basal bodies are equivalent to centrioles but exist in a different functional state. A similar discovery was made by Von Lenhossék (1898) (von Lenhossék, 1898). This observation was consistent with the original observation that sperm’s basal body, which forms the flagellum, becomes the centriole of the fertilized egg. This idea was overlooked for many years, but ultimately it is correct.

It was later discovered using cell staining that early spermatids of vertebrates have two centrioles, which were named the proximal centriole and distal centriole (Ritter, 1919; THOMAS HARRISON MONTGOMERY, 1912) (Fig 1A). Initially, the distal centriole is attached to the cell membrane, but as the flagellum grows, the centrioles are pushed back and invaginated into the nucleus (Alieva et al., 2018; Phillips, 1974). In the spermatozoa, the proximal centriole is near the nucleus and farther away from the cell membrane, and the distal centriole is farther away from the nucleus and closer to the cell membrane – thus giving the centrioles their names. The distal centriole exclusively forms the flagellum that will become the sperm tail (Fig 1A–B).

Fig 1: Centriole Remodeling During Early Spermatid Differentiation to Spermatozoa.

A) Early round spermatid encompassed by the cell membrane (CM), with a round nucleus (N), a pair of canonical centrioles (PC and DC), a flagellum made of the axoneme (Ax) cilia membrane (CiM), and the transition zone (TZ) which separates the centriole from the flagellum. The proximal centriole gives rise to a centriolar adjunct (AC).

B) Spermatozoon with a flat elongated nucleus (N), a canonical centriole (PC), an atypical centriole (DC), a flagellum made of the axoneme (Ax) cilia membrane (CiM), and decreased amounts of the cell membrane (CM). The annulus/transition zone (An/TZ) separates the flagellum into two parts: a mid-piece and a principal-piece. An inset enlarges the neck with the two centrioles, the capitulum (Ca), and striated columns (SC).

C) The Centriolar Complex of early spermatids contains a pair of canonical centrioles (PC and DC) and an axoneme (Ax), pictured from a side view. A cross-sectional view of the proximal centriole consisting of nine triplet microtubules; each triplet microtubule is made up of an A, B, and C tubule. The triplet microtubules are connected by A-C linkers, which attach the A tubule to an adjacent C tubule. The upper part of the PC and DC has a scaffolding structure that connects the microtubules of the centriole. The axoneme is comprised of nine doublet microtubules; each doublet is made of an A and B tubule. Outer and inner dynein arms connect the A tubule of one doublet to the B tubule of an adjacent doublet. The proximal centriole gives rise to a centriolar adjunct (AC).

D) The Centriolar Complex of the spermatozoon contains the canonical centriole (PC) atypical centriole (DC), and an axoneme (Ax), pictured from a side view. A cross-section of the proximal centriole made of nine doublet microtubules; each doublet microtubule is made of an A and B tubule. The doublet microtubules are splayed out and have rods in the lumen. The upper part of the PC and DC has a scaffolding structure that connects the microtubules of the centriole. The axoneme is also made of nine doublet microtubules; each doublet is made up of an A and B microtubule. Outer and inner dynein arms connect the A tubule of one doublet to the B tubule of an adjacent doublet.

Please note that the figures are not to scale and are intended to point out the approximate shape and location of highlighted features.

The spermatid centrioles have a unique structure that resembles but is distinct from that of the sperm tail axoneme (the spine-like structure of the sperm’s flagellum; Fig 1C–D). Like centrioles of most other cells, they have a cylindrical shape, made up of a wall consisting of 9 microtubule triplets, and appear in the electron microscope as electron-dense material surrounding an electron light lumen. This structural characteristic became the defining feature of a centriole. Early studies of spermatids by electron microscopy have confirmed the presence of two centrioles in the spermatids of many species (Burgos and Fawcett, 1956). The two centrioles differ from each other, both in somatic cells and in germ cells (for a more comprehensive review of general centriole biology, please see (Avidor-Reiss and Gopalakrishnan, 2013; Winey and O’Toole, 2014)). But the sperm of many animals, including mammals and insects, undergo a process termed centrosome reduction, or remodeling leaves one centriole or both centrioles unrecognizable using the aforementioned structural definition.

In mammals, the spermatid inherits two canonical centrioles at the end of meiosis, but they become more and more atypical throughout spermiogenesis; after the extension of the distal centriole to form an axoneme, the distal centriole (and sometimes also the proximal centriole) is rendered unrecognizable. While the centriole remodeling process and the centriole structure differs greatly in insects when compared to humans, most mammals and insects studied have an atypical centriole of some sort (review in Avidor-Reiss, 2018). The seemingly independent evolution of atypical centrioles in two groups of animals, with such great evolutionary distance between them, suggests that this strict structural definition of centrioles limited previous research regarding the number of centrioles in the sperm cells in mammals as well as insects. Only more recently, it was understood that some of the centrioles do not adhere to this structural definition.

In mammalian spermatozoa (e.g., guinea pig and bat), only one centriole was observed using the above structural definition (Fawcett, 1965; Fawcett and Ito, 1965). Don Wayne Fawcett (1970), a pioneer electron microscopist and a leader in the study of vertebrate spermatozoa structure, stated that “The distal centriole, which served as the basal body of the flagellum during its formation in early spermiogenesis, is absent in the mature mammalian sperm. Thus, the traditional belief that an essential component of all motile flagella is a centriole at the base, serving as the kinetic center and site of origin for the flagellar wave now has to be modified” (Fawcett, 1970). This statement created the dogma that mammalian sperm lack a distal centriole, although it was not for almost 30 more years till a potential mechanism of disappearance emerged for mammals (Manandhar et al., 1998) (Manandhar and Schatten, 2000) (Manandhar et al., 1999).

Almost simultaneously, the sperm of some insects were similarly observed to be missing one centriole or lack them altogether. Phillips, D. M (1970) (Phillips, 1970), a pioneer of electron microscopy, who studied the structure of insect spermatozoa, stated: “In all species of insects that we have examined, both centrioles disappear during spermiogenesis” (Pages 259–260). But later, it was claimed that some insects have one centriole in both the spermatid and spermatozoa (Riparbelli et al., 2010). The missing centriole in insects was the proximal centriole, which differed from observations in the mammalian sperm, in which the distal centriole was undetected. Together these discoveries in mammals and insects set a precedent for sperm centriole number to be an anomaly when compared to those of cycling somatic cells.

Around the same time, studies of mice and rats suggested that no centrioles are present in the spermatozoon or the zygote (Fawcett, 1975; Schatten et al., 1986; Woolley and Fawcett, 1973). These studies methodically characterized the disappearance of both centrioles throughout spermiogenesis and referred to the process as centrosome reduction or centriole degeneration (reviewed by Manandhar et al., 2005; Manandhar et al., 2000).

Centrosome reduction, as described in mice by Manandhar (1998) using immunofluorescence and electron microscopy, began with the elimination of gamma-tubulin (Manandhar et al., 1999). Gamma tubulin is a prominent member of a protein group that surrounds the centriole and has a role in microtubule aster assembly, known as Peri-Centriolar Material, or PCM. Next, some core centriolar proteins were reduced, such as centrin. And then finally, as the spermatozoa entered the epididymis, the well-organized microtubules of both the proximal and distal centrioles were modified, rendering them undetectable using traditional electron microscopy techniques. Similar, but less extreme, findings were shown in Rhesus (Manandhar and Schatten, 2000), and thus emerged the dogma that mammalian sperm centrioles were reduced to different extents – leaving humans, monkeys, cows, and pigs with one intact centriole, the proximal, and leaving mice and other murine animals lacking them completely. This dogma pervaded the field for the subsequent decade.

A decade later, centrioles with an atypical structure were discovered in insects; since 2009, consistent evidence has accumulated, suggesting that two centrioles are present in insect spermatids, but their presence in spermatozoa is still controversial (Blachon et al., 2009) (Blachon et al., 2014) (Gottardo et al., 2015) (Khire et al., 2016; Khire et al., 2015) (Dallai et al., 2017) (Fishman et al., 2017) (Hou et al., 2020) (Mottier-Pavie and Megraw, 2009). Since mammals, like insects, were thought to have no more than one centriole, disrupting the centrosome reduction dogma in insects motivated re-investigating the centrioles of mammalian spermatozoa.

As a result of reopening this line of investigation in mammals, in 2018, it was also shown that two centrioles are present in the spermatozoa of non-murine-mammals (Fishman et al., 2018a)(Fig 1B and D). These findings led to the departure from the Centriole Degeneration (or Reduction) Hypothesis and proposing a new hypothesis: the Centriole Remodeling Hypothesis. The Centriole Remodeling Hypothesis contends that one or both spermatid centrioles are remodeled and are functional in the spermatozoon and/or the fertilized egg (reviewed by Avidor-Reiss, 2018) (Fig 1). The Centriole Remodeling Hypothesis is further supported by findings that despite the variable and atypical structure, these centrioles can function in the sperm and the zygote (Avidor-Reiss, 2018)

While evidence supporting the presence of atypical, yet functional centrioles in the sperm of most mammals and insects is convincing, the Centrosome Remodeling Hypothesis is not universally accepted, mainly because the spermatozoa of mice and other murine animals appear to lack centrioles altogether. Furthermore, mouse embryos seem to have adopted different biology that suggests that centrioles are not essential [reviewed in \Avidor-Reiss, 2019 #4287]. Still, there are several potential explanations for the lack of centrioles in mice. They range from the idea that they exist, and we just have not discovered them yet due to lack of technology/resolution to the idea that murine animals are an anomaly and have evolved exceptional biology to accommodate the absence of centrioles. While the sperm centrioles of mice are intriguing, there is not enough known to speculate on their existence, let alone their structure, protein composition, functions, or their implications on our understanding of other sperm centrioles. For this reason, we will omit any further discussion of mice sperm centrioles, but please reference (Avidor-Reiss, 2018; Avidor-Reiss and Fishman, 2018) for more speculation on this topic.

3. Sperm Centriole Biology is Different than in Other Cell Types

Available information indicates that the biology of sperm centrioles deviates from canonical centriolar biology established in other cell types, namely in cycling cell culture cells.

Spermatogenesis starts with the cell division of spermatogonial stem cells that are found near the basal lamina of the seminiferous tubule. The resulting daughter cells begin to differentiate (Reviewed in de Rooij, 2017). The differentiated spermatogonia undergo several rounds of mitotic division to amplify the number of differentiated spermatogonia. Then, differentiated spermatogonia become spermatocytes as they begin the meiotic divisions. By the end of meiosis I and II, many spermatids are formed from a single original spermatogonial stem cell. Interfering with this process is expected to result in reduced or no spermatozoon, such as when observed in the Sertoli cell-only syndrome (SCOS) (Reviewed in Zhang et al., 2019). Sertoli cells are somatic cells (non-sperm cells) that support sperm cells during sperm differentiation (Reviewed in Wu et al., 2020).

Sperm centrioles are present in the earliest sperm progenitor to the spermatozoa; like many other cell types, two centrioles are present in Drosophila spermatogonial stem cells (Tates, 1971). However, outside of the centriole number, little is known about these centrioles. Spermatogonia centrioles were not shown to form a primary cilium like many other dividing cells. However, human and rabbit spermatogonium centrioles appear to be similar to other canonical centrioles and are located internally, near the Golgi region (Nicander and Plöen, 1969; Paniagua et al., 1985).

When the spermatogonia become spermatocytes, it has been shown using electron microscopy in two mammals (hyrax and squirrel; both are not murine species and their spermatozoa centrioles more closely resemble those of humans than mice), that two periods of centriole duplication take place during sperm meiosis such that each meiotic division has a centriolar pair at each spindle pole (Rattner, 1972). The first centriole duplication occurs in the zygotene/pachytene primary spermatocytes; the second duplication occurs in the secondary spermatocytes. In the secondary spermatocytes of hyrax and squirrels, one of the four centrioles, the older mother centriole, was connected at its tip with the cell membrane (Rattner, 1972). This finding is interesting because the two older centrioles of the secondary spermatocyte are also attached to the cell membrane in Drosophila (Tates, 1971). Therefore, this attachment may represent an evolutionarily conserved property.

Rattner (1972) reports that, in hyrax and squirrels, the attachment of the mother centriole to the cell membrane persisted until the completion of telophase II where each centriole pair assumed a characteristic orientation, adjacent to the nucleus in the early spermatid; however this attachment to the membrane is sometimes lost (Rattner, 1972). This point is interesting as it was recently claimed that spermatids differ from other cell types in that the identity of the centriole that forms the flagellum is not the mother centriole. In all cells studied that have a single cilium, the older centriole forms the cilium since only the older centriole is structurally and compositionally mature to do so (Bowler et al., 2019). However, Alieva et al. (2018) has recently suggested that in spermatids, the daughter centriole forms the flagellum simultaneously, or immediately after the mother centriole forms another axoneme-like structure, known as the centriolar adjunct (Alieva et al., 2018). They conclude this because the distal centriole base is attached to the proximal centriole’s side, as daughter centrioles typically do to mother centrioles. Their hypothesis would require that in the spermatid, centrioles are not reoriented through detachment and reattachment before flagellum formation.

Marko Marjanović (2015) stated that centrosome duplication and separation in mouse spermatocytes are delayed when compared to most somatic cells (Marjanovic et al., 2015). The duplication is initiated in early prophase I (instead of S phase in other cell types), and duplicated centrosomes are not separated until the end of prophase I (instead of G2 in other cell types).

Ultimately, in all mammals studied, including mice and humans, meiosis ends with the production of a spermatid with two centrioles. Notably, insects diverge at this point; in Drosophila, the second centriole duplication in the secondary spermatocyte presumably is stunted, producing a highly atypical centriole-like structure, the Proximal Centriole Like structure, or PCL. The PCL has only recently been observed robustly in the spermatid but does undergo centrosome remodeling like other sperm centrioles (Blachon et al., 2009; Gottardo et al., 2015; Khire et al., 2016). However, this type of atypical structure has only been described in sperm and therefore is further evidence of unique sperm centriole biology. While the discovery of the PCL in insects, which disrupted the preexisting dogma that insect sperm possessed only one centriole, is reminiscent of the discovery of a second centriole in humans, which similarly disrupted the one-centriole dogma, it is worth noting that the PCL is not homologous to the human’s atypical centriole. The PCL is homologous to the human proximal centriole, which is canonical and easily recognizable in humans. Contrastingly, the canonical centriole of Drosophila sperm is more similar to the atypical distal centriole in human sperm.

In insects and mammals, when the spermatid begins spermiogenesis, the distal centriole (or the giant centriole in Drosophila) begins to nucleate the axoneme to form the flagellum. At this point, Drosophila centrioles develop yet another unique feature: their size. In many eukaryotes, centrioles measure is up to ~500nm (Winey and O’Toole, 2014). However, the centriole is much longer in Drosophila sperm (1,800nm, (Basiri et al., 2013; Tates, 1971)), Ostrich (~3000 nm (Soley, 1993)), and Side-Necked Turtles (−4000–5000 nm (Hale et al., 1989)).

During spermiogenesis, sperm of insects and possibly some or all mammals undergo a unique event called transition zone migration. The transition zone is usually found at the junction of the centriole and cilium (Malicki and Avidor-Reiss, 2014) (Fig 1A). The transition zone acts as a gate that controls the traffic in and out of the cilium. As such, the transition zone is critical to isolate and compartmentalize the cilium or flagellum from the cytoplasm. In contrast, in mammals and insect sperm cells, at least part of the axoneme is embedded in the cytoplasm and is referred to as cytosolic cilia or flagella (Avidor-Reiss and Leroux, 2015; Avidor-Reiss et al., 2004). For example, the midpiece of mammalian sperm includes a motile axoneme that is surrounded by mitochondria; these cytoplasmic organelles are not found inside compartmentalized flagella. In insects, the mitochondria are found all along the sperm tail axoneme. In these two animal groups, the cytosol surrounds the axoneme because the gatekeeper, the transition zone, is moved to another location in a process refed to as transition zone migration (Avidor-Reiss et al., 2017; Basiri et al., 2014) (Fig 1B). In Drosophila, the transition zone that formed between the centriole and cilium migrates away along the microtubules to the tip of the axoneme. In mammals, the transition zone migrates to form the boundary between the midpiece and principal piece and becomes known as the annulus (Kwitny et al., 2010; Toure et al., 2011).

At the same time, during spermiogenesis, in insects and all mammals studied, the sperm centrioles undergo centrosome reduction, which eliminates the astral microtubules, the PCM, some core proteins, and some microtubules (Manandhar and Schatten, 2000; Manandhar et al., 2005; Manandhar et al., 2000; Manandhar et al., 1998; Schatten, 1994). The extent of this reduction is species-specific. However recently, it was proposed that sperm centrioles are not merely reduced, they are remodeled, resulting in the formation of Atypical Centrioles (Avidor-Reiss et al., 2015; Fishman et al., 2017; Khire et al., 2016) (Fishman et al., 2018a). In human and bovine, spermatozoa have two centrioles. The proximal centriole maintains a canonical structure made of 9 interconnected triplet microtubules; the distal centriole is atypical and is made of doublet microtubules that are disconnected from each other. The distal centriole’s microtubules are splayed, and, between them, there are two unique rod-like structures made of centriole lumen proteins (Fishman et al., 2018a). This centrosome remodeling process, which includes both reduction of some features and enrichment and reshaping of others, has been observed only in sperm centrioles. However, the centrioles of other terminally differentiated cells should be examined to determine if they also undergo a similar, but perhaps less extreme form of centrosome reduction.

4. Essential Functions and Expected Clinical Manifestations

The sperm centrioles are essential for human fertility, but a comprehensive understanding of their vital functions in sperm and during embryonic development is lacking. Sperm centrioles have five presumed necessary functions; however, only some of them are well demonstrated, and others are controversial (Fig 2): (1) A role in mediating mitosis and meiosis early in spermatogenesis is well accepted, but rarely tested. (2) A role in forming the sperm tail during early spermiogenesis is well accepted and demonstrated. (3) A role in linking the sperm head and tail in the spermatozoon is well accepted and demonstrated. (4) A role in controlling the sperm tail movement is controversial. (5) The role in organizing cytoskeleton of the zygote to mediate pronuclei migration and cell cleavage has aspects that are well accepted and others that are controversial.

Fig 2: The Five Functions of Sperm Centrioles.

A) Two pairs of centrioles mediate cell division (both mitosis and meiosis) during early spermatogenesis; each pair is comprised of a mother centriole (MoC) and a daughter centriole (DoC) surrounded by pericentriolar material (PCM). The two pairs of centrioles act as two centrosomes located at the spindle pole. Each centrosome nucleates the astral microtubules (A-M), which orients the spindle (Sp) relative to the axis of cell division. The chromosomes (Ch) are organized at the central plate during metaphase.

B) Distal centriole (DC) forms the sperm tail during spermiogenesis. In the early spermatid, the distal centriole docks to the cell membrane (CM), nucleates the axoneme (Ax), and initiates the formation of the cilia membrane (CiM). The proximal centriole (PC) is attached to the distal centriole perpendicularly through its sidewall, and it contains an axoneme-like structure that is usually only found during the early spermatid stage called the centriolar adjunct (CA). In the spermatozoon, the annulus/transition zone (An/TZ) migrates to the tail.

C) The centrioles play an essential role in linking the head and tail together in the spermatozoon. There are three known subtypes of breaks in this link: a subtype one break is a separation between the head containing the proximal centriole and tail that contains the distal centriole; a subtype two break is a separation between the head containing both the proximal and distal centrioles and the tail that contains most of the axoneme; a subtype three break is a separation between the head and the tail that contains both the proximal and distal centrioles (Nie et al., 2020).

D) The centrioles perform a critical role in controlling the movement of the sperm tail. The sperm tail beating starts at the proximal and distal centriole found in the neck, and a wave is propagated down the flagellum.

E) The centrioles have a role in organizing the zygote’s cytoskeleton - they form the astral microtubules (A-M) that mediate pronuclei migration and assist in cell cleavage.

Because of the centrioles’ essential functions and many components, it is expected that defects in sperm centrioles will result in a spectrum of phenotypes, depending on the function affected. Because of their essential role in early spermatogenesis, some centriolar defects are expected to result in either reduced or no spermatozoon in semen (aka azoospermia and oligospermia, respectively). Because centrioles are essential for flagellum formation, some centriolar defects are expected to result in abnormal morphology (aka teratozoospermia). For the crucial role of the centriole in linking the tail to the head, some centriolar errors are expected to result in head-tail disengagements (asthenozoospermia with teratozoospermia or asthenoteratozoospermia). If an essential role is present during sperm tail movement, some centriolar errors are expected to result in abnormal sperm movement that may result in reduced motility (aka asthenozoospermia) and/or reduced progressive movement (teratozoospermia). Finally, if an essential role is present in the zygote in orchestrating the cytoskeleton, some centriolar defects are expected to result in unexplained infertility or reoccurring pregnancy failure. We also expect a combination of the aforementioned phenotypes, such as oligoasthenoteratozoospermia. This expectation is because spermatozoon centriole is derived by duplication from early sperm cells, and mild centriole defects in early staged sperm can manifest in later stages.

A comprehensive understanding of the centriole is essential for an effective diagnosis. Centriole dysfunction in early spermatogenesis, flagellum formation, and sperm tail movement can be diagnosed by a semen analysis and can potentially be treated as long there are some remaining haploid sperm cells. For that, there is a spectrum of technologies, including intracytoplasmic sperm injection (ICSI), testicular sperm aspiration (TESA), and testicular sperm extraction (TESE) (Rajfer, 2006). In these treatments, a spermatozoon from semen, epididymis or testis, or alternativity spermatids from the testis, is injected directly into a retrieved oocyte and develop in vitro to an embryo that later is transferred to the uterus (Palermo et al., 1992). However, these treatments may be ineffective if the centriole also has an essential function in the zygote.

If indeed, the centriole plays an essential role in the zygote, it would be important to identify male infertility cases with centriole maladies. For that, there is a need to develop a technology that can assess the sperm centriole rapidly in a clinical setting.

Studying sperm centrioles during early spermatogenesis has the benefit that it can be conducted in all model animals, and there are minimal restrictions on the use of discarded human material. While an in vitro system to culture human spermatozoa has been reported, its complexity and inefficiency remain a major limitation (Perrard et al., 2016) (reviewed in Ibtisham et al., 2017). There are two major challenges to our understanding of human centrioles during fertilization and development. First, the sperm of mice, the dominant genetic mammalian model, has a dramatically different structure, composition, and function in both the spermatozoa and zygote. Second, because of ethical limitations on destroying a human embryo, most of the studies on the role of the centrioles in fertilization are rare and of small scale. Available literature includes about a dozen case studies and case-control studies, with a few retrospective studies and no prospective studies reported (Garanina et al., 2019a; Moretti et al., 2019; Sha et al., 2017) (Emery et al., 2004; Moretti et al., 2017; Terada et al., 2009) (Porcu, 2003) (Rawe et al., 2002) (Nakamura et al., 2002) (Alosilla Fonttis et al., 2002) (Chemes et al., 1999) (Sathananthan et al., 1991) (Garanina et al., 2019b) (Kai et al., 2015). Most of these reports analyze less than a handful of infertile patients, with a sample size of 10 patients or fewer. Also, only in a few of these studies is the centriole directly examined, and in many, centriolar defects are only inferred from gross cell phenotypes. Therefore, there is a need for more extensive studies that directly assess the sperm centriole.

5. The Sperm Centriole is Essential in Early Spermatogenesis

Centrioles are classically thought to play a role in forming the centrosomes that organize the cytoskeleton in animal cells (Bornens, 2012). Significantly, centrioles regulate cell division by participating in spindle assembly and positioning (Tang and Marshall, 2012), including the regulation of asymmetric cell division in stem cells (Yamashita et al., 2003; Yamashita et al., 2007). While this data was only shown in Drosophila spermatogonial stem cells and murine neural progenitors (Paridaen et al., 2013b; Wang et al., 2009), the establishment of a biological mechanism to control this suggests that it may be in use in many other cell types. In both flies and mice stem cells, it was proposed that centrosomes orient the spindle such that stem cell divisions produce a cell that is connected to the stem cell niche and becomes a stem cell, and a cell that is further away from the niche, which differentiates. Mutations in centriole proteins result in abnormal division axes that leads to depletion of the stem cell population (Paridaen et al., 2013a). This type of mechanism was shown in fly testes stem cells (Cheng et al., 2008). However, this has not yet been described in the sperm stem cells of humans or other mammals.

Recently, a specific type of mutation in a centriole specific protein was shown to affect spermatogonia: PLK4 and POC1A in mammals. PLK4, a well-studied centriole protein, is the master regulator of centriole duplication and is essential for centriole formation in all animals studied (reviewed in Arquint and Nigg, 2016; Bettencourt-Dias et al., 2005; Habedanck et al., 2005). Male mice with a heterozygous 13-bp deletion in the Ser/Thr kinase domain of the Plk4 gene, which results in a frameshift with a premature stop codon, show azoospermia associated with germ cell loss and Sertoli cell-only syndrome (Miyamoto et al., 2016). Male mice with a heterozygous missense mutation in the kinase domain (I242N) leads to patchy germ cell loss (Harris et al., 2011). Plk4 haploinsufficiency causes mitotic infidelity and carcinogenesis in mice; however, both mutations may function as dominant-negatives (Ko et al., 2005). The reason why a mutation in such a critical centriole protein has a specific effect on spermatogenesis is unclear. Still, it may be due to the specific effect of the mutation. Other PLK4 mutations were shown to have a tissue-specific phenotype, such as microcephaly in humans (Tsutsumi et al., 2016).

POC1A, a highly conserved centriole protein, is one of the two paralogs of the POC1 protein family in mammals (the other is POC1B). The POC1 protein participates in centriole structural stability and is not essential for centriole duplication in many species including protists, insects, and mammals (Keller et al., 2009) (Pearson et al., 2009) (Fourrage et al., 2010) (Fishman et al., 2017; Shaheen et al., 2012) (Venoux et al., 2013) (Jo et al., 2019). POC1A mutations in humans cause the SOFT syndrome that is characterized by short stature, onychodysplasia, facial dysmorphism, hypotrichosis, and variable skeletal abnormalities (Saida et al., 2019). Male mice with an insertion mutation of a LINE-1 retrotransposon in the Poc1a gene show defective meiosis and progressive germ cell loss (Geister et al., 2015). The insertion causes exon skipping but retains the Poc1a reading frame. Why the mutation causes exclusively male infertility (and skeletal dysplasia) is unclear, but POC1A may have a domain that is essential only in some cell types.

CEP63 is one of two paralogs in the CEP63 protein family in vertebrates (the other one is Deup1). CEP63 is a centrosomal (PCM) protein that facilitates centriole duplication along with CEP57 and CEP152 in Xenopus laevis (Lukinavicius et al., 2013). Like in many centrosomal mutations, its depletion results in microcephaly in humans (Sir et al., 2011). Male mice with no CEP63 show impaired meiotic recombination, leading to profound male infertility (Marjanovic et al., 2015).

Usually, centrioles duplicate once per cell cycle in spermatogonia and spermatocytes to generate spermatids with precisely two centrioles. It was recently reported that this number control mechanism is abnormal in null mice of the tubulin deglutamylase, CCP5 (AKA AGBL5 (Giordano et al., 2019)(Kastner et al., 2015)(Rogowski et al., 2010)).

These studies show that despite centrioles being essential for the development of live animals or humans, some mutations in centriolar components can have a relatively specific effect on male fertility.

6. The Sperm Centriole is Required for Flagellum Formation and Linking the Head to the Tail

The two functions of the centriole in forming the tail and anchoring it to the head are related, and we discuss them here together. Centrioles are nucleators of cilia, and the extension of their microtubules form the cilium or flagellum’s axoneme (reviewed in Wang and Stearns, 2017). The terms “cilium” and “flagellum” are used interchangeably throughout the field, although there are structural and motility differences between the two depending on species and cell type; “flagellum” is used most universally in sperm, despite it being greatly different from the bacterial flagellum. Throughout the remainder of this review, we use the term flagellum to refer to the sperm tail and cilium where applicable across broader cell types.

The centriole’s role in axoneme formation is preserved in spermatids; the centriole pair attaches to the cell membrane and forms a primary cilium that later differentiates into the flagellum of the sperm tail (Rattner, 1972) (Fawcett and Phillips, 1969). In Drosophila, multiple mutations in centriole genes have shown defects in sperm flagellum formation (Blachon et al., 2009; Blachon et al., 2008; Khire et al., 2016; Reina et al., 2018). A few similar mutations have been described in humans and other mammals (Lv et al., 2020; Sha et al., 2017).

The centrioles are components of the spermatozoon neck (aka connecting piece) where they are attached to the nucleus and form the flagellum, thus connecting the head to the tail (Fig 1b). In the neck, the centrioles are integral to two specialized cytoskeletal structures: (1) A dense fibrous plate-like structure named the capitulum and several highly organized bundles of pale and dark bands called the segmented columns or striated columns. (2) The centrioles, capitulum, and striated columns form a structural anchor connecting the tail to the head and serve as a mechanical link between them.

Recently two centriolar protein CEP135 and DZIP1 were identified in patients with Multiple morphological abnormalities of the sperm flagella (MMAF) (Lv et al., 2020; Sha et al., 2017). CEP135 is a conserved centriole core protein located in the cartwheel and centriole wall (Blachon et al., 2009; Hiraki et al., 2007; Kraatz et al., 2016; Mottier-Pavie and Megraw, 2009). A homozygous missense mutation (p.D455V) in CEP135 induced severe MMAF (Sha et al., 2017). The CEP135 D455V variant mislocalizes in the sperm and leads to the formation of aggregates in the sperm neck. Only 60% of the sperm had flagellum, and 45% of them were short. This data is consistent with the idea that sperm centrioles are essential for flagellum formation in the sperm.

DZIP1 is a component of a part of the centrosome called the distal appendages/transition fibers, which functions in microtubule anchoring and anchoring the centrioles to the cell membrane during cilium formation (Fig 3) (Lapart et al., 2019; Wang et al., 2013; Wei et al., 2015; Zhang et al., 2017). A homozygous missense mutation (pArg63Gln) and a homozygous truncation mutation (p.Tyr230*) in DZIP1 induced asthenoteratospermia with severe MMAF (Lv et al., 2020). Immunofluorescent staining with the centriolar protein CETN1 finds abnormally reduced centriole number in half of Tyr230* patient’s spermatozoa. Dzip1-knockout male mice have non-motile spermatozoa with severe morphological abnormalities and are infertile. This data is consistent with the idea that sperm centrioles are essential for flagellum formation in the sperm.

Fig 3: Canonical Centriole Formation, Structure, and Function.

A) Canonical centriole formation. New centriole formation starts with the formation of a Torus structure around the pre-existing mother centriole (MoC). Next, a single cartwheel like structure forms at one spot along the Torus perpendicular to the wall of the mother centriole. Then, the cartwheel develops into a procentriole by forming microtubules (green). The procentriole continues to grow into a daughter centriole (DoC) with an upper scaffold and detaches from the mother wall. The daughter centriole is connected to the mother by a linker complex. Finally, the daughter centriole develops appendages, after which the daughter centriole recruits additional proteins, which stabilize its structure and provide additional functionality (e.g., forming pericentriolar microtubules, astral microtubules, and a flagellum). Proteins involved in centriole formation and are implicated in sperm function are marked in bold at the step they mediate. Please note that the distal appendages are placed on the same level of the MT triplets, and they differ from sub-distal appendages; the number of sub-distal appendages can be from 0 to 14 (Uzbekov and Alieva, 2018).

B) Centrosome structure. The centrosome consists of a pair of centrioles (a mother centriole (MoC) and an attached procentriole), centriolar satellites, and pericentriolar material (PCM). A cross-section at the upper part of the mother centriole depicts the presence of an upper scaffold. A cross-section at the procentriole depicts the presence of a cartwheel and A-C linker.

C) Flagellum structure. The flagellum consists of a pair of centrioles and the axoneme (Ax). A cross-section shows the upper part of the mother centriole (MoC) and depicts the presence of an upper scaffold. A cross-section at the upper part of the daughter centriole (DoC) depicts the presence of an upper scaffold. A cross-section at the axoneme depicts a central pair running through the axoneme, and inner and outer dynein arms decorate the 9-doublet microtubules.

Centrin (CETN) proteins are conserved centriolar proteins that fill the lumen of the mature centriole in mammals (reviewed in Dantas et al., 2012). However, CETN proteins are found in non-centriole sites in the cell, and their role is unclear in most cell types (Dantas et al., 2011; Paoletti et al., 1996; Salisbury et al., 2002; Sullivan et al., 1998). In mammals, there are four CETN paralogs, and CETN1 is very similar to CETN2 (84% aa identity) (Wolfrum and Salisbury, 1998). (Sawant et al., 2015). CETN1 is the only member that is specifically enriched in sperm cells (Sun et al., 2002). CETN2 is also expressed in sperm but is expressed earlier than CETN1 in spermatogenesis. Germline deletion of CETN1 in mice causes spermatids to lack tails (Avasthi et al., 2013). Centriole arrangement of the early spermiogenesis was almost normal, and the proximal centriole appears to form the centriolar adjunct (based on CETN3 labeling). Interestingly, in later spermatids, the striated columns surrounding the distal centriole and flagellum are missing, suggesting that CETN1 is essential for the formation or stability of both structures.

Azi1/Cep131/DILA localize to centriole and centriolar satellites (Fig 3B), another centrosome-associated substructure that has a role in microtubule organization (Mahjoub and Tsou, 2013). Azi/Cep131/DILA is also found in other compartments of the spermatids (e.g., acrosome and manchette) (Aoto et al., 1997). CEP131 is required for cilia formation in Danio rerio and Drosophila melanogaster (Ma and Jarman, 2011; Wilkinson et al., 2009). In mammals, CEP131 appears to have a sperm-specific phenotype; in CEP131 null mice, the only phenotype is a failure to form the sperm tail, resulting in male infertility (Hall et al., 2013).

Acephalic Spermatozoa

An exciting group of proteins, which works together with the centrioles to link the head to the tail, results in Head-neck defects or Acephalic spermatozoa when perturbed. Recently, it was proposed to classify acephalic sperm into three different subtypes based on the site of break inside the sperm neck. (Nie et al., 2020) (Fig 2B). In Subtype I, the hypothetical separation point is between the proximal and distal centrioles, but no mutations that caused this phenotype were identified. In Subtype II, the separation point is between the nucleus and proximal centriole; this phenotype is associated with mutations in HOOK1, SUN5 (Elkhatib et al., 2017; Shang et al., 2017), and PMFBP1 (Zhu et al., 2018). In Subtype III, the separation point is between the distal centriole and mitochondrial sheath and is associated with mutations in the genes TSGA10 and BRDT. This classification system is useful, as different subtypes appear to correlate with varying outcomes of ICSI. Subtype II achieved better ICSI outcomes than the other tested subtypes, potentially because the sperm head carries the two centrioles. Subtypes I and III are presumed to be more likely to result in failed clinical pregnancy because the sperm head is missing one or both centrioles.

Mice studies suggested additional, non-centrosomal, genes that can cause acephalic spermatozoa. Spata6 is required for the formation of the segmented columns, and the capitulum (Yuan et al., 2015) and a Spatc1l mutation leads to male sterility owing to the separation of sperm heads from tails (Kim et al., 2018).

Two centrosomal proteins are known to cause acephalic spermatozoa: CCDC42 and TSGA10, and (Pasek et al., 2016; Sha et al., 2018; Tapia Contreras and Hoyer-Fender, 2019). CCDC42 is a non-canonical centriolar protein found in eukaryotes with motile cilia. It is expressed in male germ cells, the brain, and in tissue culture cells (NIH3T3 cells), where it localizes to the centrosome, but its function is unclear (Pasek et al., 2016; Tapia Contreras and Hoyer-Fender, 2019). Also, CCDC42 localizes to the base of cilia in Tetrahymena thermophila, suggesting it is part of the basal body. However, it is also localized outside of the sperm neck. In spermatids, CCDC42 localizes to the manchette, the connecting piece, and the tail. CCDC42 interacts biochemically with ODF2, a prevalent protein involved in the formation of a spermatid-specific cytoskeletal structure known as ODF (Tapia Contreras and Hoyer-Fender, 2019). Mutations in Ccdc42 are associated with malformation of the mouse sperm flagella and formation acephalic sperm (Pasek et al., 2016; Tapia Contreras and Hoyer-Fender, 2019).

TSGA10 is a testis enriched protein (Modarressi et al., 2001). Mouse Tsga10 encodes a 65-kDa spermatid protein, which forms two proteins; (1) A 27-kDa N-terminal protein that localizes to the fibrous sheath of mature spermatozoa (Modarressi et al., 2004). (2) And a 55-KDa C-terminal protein, which localizes to the midpiece of mature spermatozoa (Behnam et al., 2015). Mutations in TSGA10 are associated with acephalic spermatozoa, and a patient with a homozygous deletion within TSGA10 (A71Hfs*12) resulted in 99% headless sperm in his ejaculate (Sha et al., 2018). Interestingly, TSGA10 is overexpressed in cancer, and in these cells, it localizes to the centrosome and may interact directly or indirectly with CEP135 (Behnam et al., 2015; Inanç et al., 2013).

CNTROB is a daughter-centriole specific protein with five isoforms. In typical somatic cells, it interacts with NIP2 to stabilize microtubules and disruption of this complex by knocking down NIP2 results in centrosomes that have a reduced number of, or complete lack centrioles (Jeong 2007), suggesting that CNTROB may have a role in centriole stability. A retroviral insertion of CNTROB produced a truncated allele in rats that exhibited sperm with severed heads and tails (Liska et al., 2009). It is worth noting that the tails of these mutants appear fully formed, indicating that the distal centriole was able to form the flagellum before reduction, but that the mutated protein performs a separate function. While this work could be perceived as having limited implications in humans due to the dramatic difference between human sperm, and centriole-lacking rat sperm, it could also be interpreted to suggest that centriolar proteins in murine sperm are not dispensable. Despite that centrioles have not been observed in spermatozoa, Liska et al’s work suggests that the proteins still perform a function, or that their function at earlier stages have lasting effects, even after the centrioles have been reduced. However, this research must be treated with caution because the phenotype is not rescued by overexpression of full-length CNTROB (Liška et al., 2013), suggesting this gene may not mediate this phenotype. Therefore CNTROB and its role in fertility require more research before making any further conclusions.

These mutations, both in humans and murine mammals, suggest that centrioles or centriolar proteins have a critical role in the attachment of the head to the tail, possibly due to their role in manchette formation, the capitulum, or the striated columns. This phenotype holds an exciting relationship between the centrioles and infertility. Teratozoospermia is present in about 3% of male infertility (Pandruvada et al., In preparation), and some simple physical manipulations could improve ICSI treatment in these men, including depositing the severed head and tail near each other (Emery et al., 2004).

7. Sperm Centriole Control Sperm Movement

It was recently found that human sperm beat asymmetrically, and progressive movement is achieved by the sperm rotating around its beating center [Gadêlha, 2020 #5199]. How sperm movement is generated, propagated, and regulated has been an important question in sperm biology [reviewed by \Ishijima, 2019 #4627; Simons, 2018 #5202]. There are two models for flagellum movement, and depending on the extent they apply, the centriolar defects would have different outcomes for sperm motility.

The classic model depicts an active sliding movement of axoneme microtubules restricted at its base by the centriole (King and Sale, 2018; Woolley et al., 2006). This model is consistent with the firm structure of the canonical centriole, where various structures stabilize the centriole wall. Two structural mechanisms stabilize mature centrioles: A-C linkers that connect the A microtubule of one triplet to the C tubule of a neighboring triplet (Li et al., 2019; Meehl et al., 2016), and a helical inner centriole scaffold (Greenan et al., 2018) (Li et al., 2012) (Vorobjev and Chentsov, 1980) (Dippell, 1968; Perkins, 1970) (Ibrahim et al., 2009) (Paintrand et al., 1992). These structures generate a structural cohesiveness that restricts the movement of microtubules at the flagellum base of canonical centrioles.

An alternative model known as the Basal Sliding model was explicitly proposed for mammalian sperm beating. In this model, the base of the mammalian flagellum is dynamic and allows limited sliding to modulate the tail-beating pattern (Vernon and Woolley, 2004; Woolley and Fawcett, 1973). The Basal Sliding model is based on the observation of a slight movement in the sperm’s neck using light microscopy and electron microscopy that shows varied width of the bands that constitute the striated columns. Also, mathematical modeling of the sperm tail waveform suggests that there is sliding of the microtubules at the base of the axoneme; this sliding is necessary to control the shape of the tail’s waveform (Gadelha et al., 2013). Finally, the distal centriole located at the base of the axoneme was recently proposed to be composed of splayed microtubule doublets. Because of the disconnected microtubular distribution in the atypical distal centriole, this centriole is likely to be missing the A-C linker and the helical inner centriole scaffold, as evident by the formation of rod-like structures of lumen proteins that, in other cell types, form discs (Fishman et al., 2018b; Sydor et al., 2018). As a result, the atypical distal centriole structure is consistent with the Basal Sliding model and may have evolved to permit it (Avidor-Reiss and Fishman, 2018). However, no direct evidence showing sliding at the axoneme base of the microtubules has been provided to date.

8. The Sperm Centriole is Essential for Organizing the Microtubule Cytoskeleton in the Zygote

The sperm is the sole contributor of centrioles to the zygote, and these centrioles are thought to be essential for fertility in humans (reviewed by Chemes and Sedo, 2012) (Schatten and Sun, 2010) (Palermo et al., 1997). The sperm centriole recruits maternal proteins and forms a large aster associated with the sperm inside the zygote (Meaders and Burgess, 2020; Nakamura et al., 2001; Terada et al., 2000; Van Blerkom and Davis, 1995). Schatten (1981) proposed that the sperm aster is involved in bringing together the male and female pronuclei (Schatten, 1981). However, this was not tested rigorously, and it is unclear what the aster is essential for if anything, but this idea is widely accepted.

A key question is what are the essential functions of sperm centrioles in the zygote if there are any. In the zygotes of some animals, sperm centrioles are accepted as essential for bipolar spindle formation during the first division (O’Connell et al., 2001; Varmark et al., 2007; Yabe et al., 2007). However, sperm centrioles are not essential in mice (Manandhar et al., 1999), and therefore, the essential role of the sperm centrioles may be species-specific. In this case, a need exists for a direct demonstration that human sperm centrioles are essential for embryo development and identifying, their precise role, mechanism of action, and clinical implications (reviewed by Avidor-Reiss et al., 2019).

Strong evidence for the essential role of centrioles in humans is obtained from the detailed analysis of ICSI with Acephalic sperm (see discussion above). Nie at al (2020) carefully examined the outcome of ICSI using Acephalic sperm and correlates the outcomes with the precise location of the break in the sperm neck. They found that breaks that generate a sperm head with two centrioles have a better ICSI outcome than breaks that generate a sperm head with one or no centrioles (Fig 2B).

Also, two interesting small case studies directly connect the sperm centrioles to infertility due to post-fertilization dysfunction (Garanina et al., 2019a; Sha et al., 2017).

Sha et al. (2017) found that ICSI using sperm with mutations in the centriole specific component CEP135 failed to result in pregnancy (Sha et al., 2017). Single sperm was injected into each of six oocytes, which resulted in six embryos that underwent cleavage and developed to morula. Still, the transfer of three of these embryos resulted in a failed pregnancy. This data suggests that sperm centrioles are not essential for zygote cleavage but are essential for subsequent development of the early embryo.

Garanina et al. (2019) recently suggested that zygote arrest may be related to an incompletely disassembled centriolar adjunct in a spermatozoon from a couple with unexplained infertility (Garanina et al., 2019a). The centriolar adjunct (Fig 1A) is an unusual extension of the proximal centriole, with unknown function, which is present in most mammals only earlier during spermiogenesis (Fawcett and Phillips, 1969). However, oddly enough, the centriolar adjunct is present in healthy ejaculated human spermatozoa (Zamboni and Stefanini, 1971). It was found that two patients with unexplained infertility had longer adjuncts in their mature sperm than those of donor sperm. One of these patients belonged to a couple that performed three unsuccessful IVF cycles after seven failed sperm inseminations. The second patient had a shorter, but still abnormally longer-than-normal adjunct. He belonged to a couple that had no pregnancies after two years of unprotected intercourse, and then two unsuccessful IFV cycles. These cases suggest that while the centriolar adjunct is surprisingly present in human spermatozoa, its length affects fertility. Because the adjunct is directly associated with the proximal centriole, it is possible that some of these adjunct defects could be a result of an underling centriolar defect.

9. The Sperm Centriole in Other Health Conditions

Sperm centrioles may be affected by other factors including, aging, pathogens, health conditions, cryopreservation, and environmental factors. Still, no direct assessment of the sperm centriole concerning these factors has been published. Aging is an important factor in fertility, as in many countries, the average age of paternity is rising. Older men have worse semen parameters, poor sperm, and their offspring have an increased incidence of health problems (Mazur and Lipshultz, 2018). Cheng et al. found that centrosome misorientation reduces sperm stem cell division in aging fruit flies (Cheng et al., 2008). Therefore, it would be important to test if sperm centrioles of aging men are compromised.

Pathogens may also affect sperm centrioles. Seyed Hamidreza Monavaria (2013) found that the herpes simplex virus (HSV) reduces male fertility and correlates with lower sperm count in the seminal fluid (Monavari et al., 2013). Vilma Jeršovienė et al. (2019) found that the Human papillomavirus infection (HPV) was more frequently detected in men with abnormal sperm parameters (Jeršovienė et al., 2019). Furthermore, human papillomavirus-16 E7 oncoprotein induces centriole multiplication by activating the promoter of the centriole duplication master regulator PLK4 in tissue culture explements (Duensing et al., 2009; Korzeniewski et al., 2011). This suggests that HSV and HPV, as well as other viruses, may target sperm centrioles, thereby affecting fertility.

An example of health conditions that may affect the sperm centriole is obesity. Obesity has the possibility of affecting male fertility by negatively effecting sperm count, sperm progressive motility, and increasing sperm head defects (Baydilli et al., 2020; Carette et al., 2019; Martins et al., 2020; Palmer et al., 2012). Recent differential proteomic analysis of obese mice, which were fed with a high-fat diet, contained reduced levels of the centriole protein Centrin 1 (CETN1), and the centrosomal protein CSPP1 in sperm (Peng et al., 2019). These proteins normally localize to the sperm centriole and sperm neck, increasing the probability that centriole quality is reduced in obesity. Another proteomic study comparing spermatozoa of fertile and infertile men identified under-expression of the microtubule cytoskeleton interacting protein CSPP1 in infertile men. Because of CSPP1’s known effect on the cytoskeleton, this suggests that CSPP1 may affect the centrosome in ejaculated spermatozoa (Samanta et al., 2019).

Cryopreservation of sperm in liquid nitrogen is used for the storage of sperm for agriculture, biomedical research, and gene banking (Barbas and Mascarenhas, 2009; Ugur et al., 2019), but what effect cryopreservation has on sperm centrioles is unclear. Sánchez-Partida et al. (2008) evaluated sperm centrioles using electron microscopy, after freeze-drying cryopreservation in Rhesus macaque (Sánchez-Partida et al., 2008). The proximal centriole was abnormal, depending on the composition of cryopreservation media. This observation suggests that cryopreservation has a potential negative impact on sperm centrioles, and this effect needs further examination.

10. conclusion

In conclusion, the centrioles are an essential component of the sperm with many mysteries surrounding both their structure and composition, as well as their function and role in the sperm and the resulting embryo. Because of their potential implications in fertilization and early development, more studies on centrioles during reproduction are critical. However, because of the limited application of studies in mice, due to species-specific differences when compared to humans, it is essential to conduct these studies in other systems. While ideally, these studies would be done in human embryos, ethical and legal considerations further limit this avenue. For this reason, we encourage a multidisciplinary, multi-species analysis of centrioles during development. Sperm centrioles should be studied primarily in humans since this material is easily accessible to most academic labs. Then, results should be confirmed in livestock species that model human embryonic development well. Then future embryonic studies can be done in livestock embryos. And finally, with the help of the medical community, the resulting hypotheses can be investigated efficiently in human embryos.

Furthermore, some of the proposed experiments regarding sperm motility are difficult because of the tiny size of the centrioles and the limited applications that allow for live visualization of the centrioles and the sperm movement. Because of this limitation, mathematical modeling and epidemiological studies are critical to assess hypotheses, while also developing new live imaging techniques.

Centrioles during gamete development, fertilization, and embryonic development are a research topic on the brink of major discoveries that will ultimately change our understanding of fertilization and eventually will lead to the development of more efficient diagnoses and IVF therapies.

Highlights.

• Spermatozoon have two centrioles: one canonical and one atypical

• Sperm centrioles have roles during spermatogenesis, including during cell divisions and flagellum formation

• Sperm centrioles play roles in the mature sperm, including linking the head and tail and controlling beating.

• Sperm centrioles have roles in post-fertilization, including in shaping the zygote cytoskeleton.

• Many aspects of sperm centriole biology are different than in other cell types.