Lack of Acrosome Formation in Hrb-Deficient Mice

See allHide authors and affiliations

Science  16 Nov 2001:
Vol. 294, Issue 5546, pp. 1531-1533
DOI: 10.1126/science.1063665


The sperm acrosome is essential for sperm-egg fusion and is often defective in men with nonobstructive infertility. Here we report that male mice with a null mutation in Hrb are infertile and display round-headed spermatozoa that lack an acrosome. In wild-type spermatids, Hrb is associated with the cytosolic surface of proacrosomic transport vesicles that fuse to create a single large acrosomic vesicle at step 3 of spermiogenesis. Although proacrosomic vesicles form in spermatids that lack Hrb, the vesicles are unable to fuse, blocking acrosome development at step 2. We conclude that Hrb is required for docking and/or fusion of proacrosomic vesicles during acrosome biogenesis.

The acrosome is a unique membranous organelle that is formed during spermiogenesis through an integrated process of transport vesicle production, trafficking, and fusion (1, 2). In the first period of acrosome biogenesis, the Golgi phase, numerous small proacrosomic vesicles accumulate in the medulla, a cytoplasmic compartment located between the juxta-nuclear Golgi apparatus and the nuclear surface. There, they coalesce into a single spherical acrosomic vesicle that attaches to the nucleus. During the subsequent Cap phase, the acrosomic vesicle expands by means of sustained vesicular membrane transport and fusion. At the onset of sperm elongation, the acrosomic vesicle stops growing in size and starts undergoing a series of highly complex morphological changes that shape the sperm cell. Little is known about the molecular mechanisms that regulate the targeting and fusion of transport vesicles during acrosomic vesicle formation, although the understanding of this process is highly relevant because disturbances of acrosomal development and function perturb the fertilizing capacity of spermatozoa (3,4).

Here we identify the Asn-Pro-Phe (NPF) motif-containing protein Hrb (also called Rab or hRip) (5–7) as a component that is essential for acrosome formation. We addressed the physiological role of Hrb through targeted inactivation of the Hrb gene in the mouse (8, 9). Hrb−/− mice were indistinguishable from Hrb+/− and Hrb+/+ mice, had a normal life-span, and showed no apparent histological abnormalities in brain, colon, heart, kidney, liver, lung, spleen, or thymus. Western blot analysis confirmed that the mutant Hrb allele was null (8, 9). Despite normal sexual behavior and copulation, Hrb−/− males derived from two independently targeted embryonic stem cell (ES) clones were infertile. By contrast, Hrb−/− females displayed normal fertility. The average testicular weight in 10-week-old Hrb−/− mice [0.079 g ± 0.1 (n = 12 testes)] was not significantly different from that in wild-type mice [0.082 g ± 0.09 (n = 12 testes)]. However, Hrb−/− mice contained on average ∼15-fold less spermatozoa in the cauda epididymis than wild-type males (8). The spermatozoa from mutant males showed severely reduced motility and several structural abnormalities. The spermatozoa displayed a globic head (Fig. 1A) with a round nucleus and a tail midpiece lacking the mitochondrial sheath (Fig. 1B) (8). This sheath is composed of mitochondria that generate adenosine triphosphate (ATP) for movement of the flagellum and is formed in the final steps of normal spermatid elongation (2). In vitro fertilization experiments revealed that Hrb−/−spermatozoa were unable to establish sperm-egg attachment and fertilization (Fig. 1C) (8).

Figure 1

Analysis of spermatozoa from Hrb-deficient mice. (A) Scanning electron micrographs of sperm cells in the cauda epididymis of wild-type and knockout mice. (B) Mitochondrial sheath configurations in Hrb+/+ and Hrb−/− spermatozoa. Mitochondria (arrows) were stained with MitoTracker Green (green). Nuclei were stained with Hoechst (blue) to visualize the sperm heads. (C) In vitro fertilization experiments with zona pellucida–intact eggs from wild-type females and sperm from Hrb+/+ or Hrb−/− males. Sperm-egg interaction and fertilization were determined by microcopy at 2 and 24 hour post-insemination, respectively.

β-galactosidase staining of the testis of Hrb+/− mice indicted that the Hrb gene was abundantly transcribed during spermiogenesis (9). Histological analysis of testis sections showed that seminiferous-tubule diameters in Hrb+/+ and Hrb−/− mice were comparable (9). The number and distribution of mitotic (spermatogonia) and meiotic (spermatocytes) germ cells in these tubules also appeared to be similar (Fig. 2, A and B) (8). The appearance of interstitial cells was also normal. However, the process of transformation of round spermatids into testicular spermatozoa (spermiogenesis) was clearly disrupted in Hrb−/− mutants. Typically, mutant tubules were devoid of elongated spermatids and contained exclusively round-headed spermatids (Fig. 2B) (9). Although acrosomes were easily detectable in wild-type round spermatids by light microscopy and by transmission electron microscopy (TEM) (Fig. 2, A and C) (8), no acrosome structures were visible in mutant spermatids (Fig. 2, B and D). Extensive ultrastructural analysis of testis sections by TEM showed normal spermatogenesis in knockout mice up through the second step of spermiogenesis, when numerous small proacrosomic vesicles accumulate in the medulla. However, these vesicles failed to fuse with each other to yield the single large acrosomic vesicle found in normal spermiogenesis (Fig. 2, E through J). This stage-specific defect was consistent in both lines of Hrb−/− mice (n = 5 mice). These results suggest that the primary spermatogenic defect in Hrb−/− mice is an arrest in acrosomic vesicle biogenesis at step 2 of spermiogenesis caused by defective docking and/or fusion of proacrosomic vesicles.

Figure 2

Hrb−/− round spermatids lack an acrosome. (A and B) Cross sections through seminiferous tubules from 10-week-old Hrb+/+ (A) and Hrb−/− (B) males stained with toluidine blue. Indicated are positions of interstitial cells (I), spermatogonia (Sg), spermatocytes (Sc), round-headed spermatids (RS), and elongated spermatids (ES). Arrowheads point to acrosomes of round spermatids. (C through J) Transmission electron micrographs of round spermatids from 10- or 12-week-old Hrb+/+and Hrb−/− mice. (C) Section through a step 6 Hrb+/+ spermatid with a flat, cap-like acrosome (*). Note the thin layer of electron dense material on the inner aspect of the nuclear envelope below the acrosome. This layer is typically present throughout steps 3 through 12 of spermiogenesis, but not at steps 1 and 2. (D) Arrested Hrb−/− round spermatid without an acrosome. (E) Medulla of a Hrb+/+ spermatid at step 2 with many proacrosomic vesicles. (F) Medulla of a step 2 Hrb−/− spermatid showing a high concentration of proacrosomic vesicles. (G) Step 4 Hrb+/+ spermatid with a newly formed acrosome (hemispherical). (H) Hrb−/−spermatid with small proacrosomic vesicles that target to the NE, but fail to fuse. (I) Step 6 Hrb+/+ spermatid containing a cap-like acrosome. (J) Hrb−/− spermatid with proacrosomic vesicles along the NE. Abbreviations are as follows: G, Golgi apparatus; Nu, nucleus; NE, nuclear envelope; AC, acrosome; M, medulla; P, large proacrosomic vesicle with an electron dense core. Arrowheads point out the NE.

To further characterize the role of Hrb in acrosome formation, we determined the intracellular localization of the protein in wild-type spermatids by immunofluorescence confocal microscopy (8). Hrb-specific immunoreactivity appeared in several small punctae throughout the cytoplasmic compartment of round spermatids. A high concentration of punctae consistently appeared in the medulla, as revealed by double labeling of round spermatids for Hrb and Golgi-membrane 58-kD protein (Fig. 3A). As expected, no Hrb labeling was detected in Hrb−/− round spermatids (data not shown). We further defined these Hrb-positive punctae by immunoelectron microscopy (IEM) (8). Labeling for Hrb was associated with the cytosolic surface of proacrosomic vesicles of early round spermatids (Fig. 3, B and C). In round spermatids with a developing acrosomic vesicle, gold particles were also found at distinct positions along the outer membrane of the acrosomic vesicle (Fig. 3D). These findings suggest that Hrb may function in vesicle-to-vesicle docking and/or fusion during acrosome biogenesis.

Figure 3

Hrb is associated with membranous organelles in the medulla of round spermatids. (A) Confocal images of a wild-type round spermatid after immunofluorescent staining with Hrb and Golgi 58 K protein. Bar, 5 μm. (B through D) Immunoelectron micrographs of Hrb+/+ round spermatids. We visualized Hrb by incubating a testicular germ cell suspension first with affinity-purified Hrb antibody and then with antibody to rabbit immunoglobulin G (IgG) covalently linked to 5-nm gold particles. (B) Detail of a step 2 spermatid, showing Hrb at the surface of a relatively large proacrosomic vesicle. (C) A step 3 spermatid with newly formed acrosomic vesicle settling at nuclear envelope. (D) A round spermatid with a cap-like acrosome exhibiting Hrb labeling at the cytoplasmic surface of the acrosomic vesicle. Arrowheads highlight the gold particles. Abbreviations are as indicated in Fig. 2.

Next, we investigated the relation between Hrb and Eps15, a transport vesicle adaptor protein with a role in endocytosis and vesicle sorting that has been shown to interact with Hrb and several other NPF proteins by virtue of its Eps15 homology (EH) domains (7, 10). We demonstrated that Hrb and Eps15 form a complex in testis by a coimmunoprecipitation approach (8) (Fig. 4K). We then used immunofluorescence confocal microscopy to determine whether Eps15 and Hrb colocalize during the process of acrosome biogenesis. Eps15 appeared prominently in small punctae in the medulla that generally coincided with Hrb. Measurements revealed that of 177 punctae in 30 round spermatids, 150 (85%) labeled for both Hrb and Eps15. Only 20 and 7 punctae were only positive for Hrb or Eps15, respectively. Triple labeling (8) of round spermatids for Hrb, Eps15, and Ap1, a heterotetrameric adaptor protein complex that is associated with Golgi-derived transport vesicles, showed that more than half (62%) of the Hrb+/Eps15+ punctae in the medulla also contained Ap1 (Fig. 4, A through D). Furthermore, we observed that a relatively small number of punctae [43 of 220 punctae (20%);n = 30 round spermatids] contained only Ap1. Unlike Ap1, Ap2, a marker for endocytic vesicles, did not exhibit any punctate labeling in round spermatids. These results suggest that the medulla of round spermatids contains distinct subsets of proacrosomic vesicles that are Hrb+/Eps15+/Ap1+, Hrb+/Eps15+, or Ap1+. When we labeled Hrb-deficient round spermatids for Ap1 and Eps15, both adaptor proteins exhibited a punctate staining in the medulla, but there was no appreciable overlap between the Eps15 and Ap1 staining patterns (Fig. 4, E through J). Specifically, of 345 punctae in 37 spermatids, only 7 (2%) contained both Eps15 and Ap1, whereas 159 (46%) were positive only for Eps15 and 179 (52%) only for Ap1. This strictly separate distribution of Eps15 and Ap1 suggests that Eps15-positive proacrosomic vesicles lack the ability to merge with Ap1-positive proacrosomic vesicles in the absence of Hrb and further supports the idea that Hrb is a critical component of the vesicle-to-vesicle docking and/or fusion machinery that forms the acrosomic vesicle.

Figure 4

The medulla of Hrb−/−spermatids lacks Ap1/Eps15 double-positive vesicles. (Athrough D) Confocal images of a representative Hrb+/+ round spermatid triple-labeled with Hrb (A), Eps15 (B), and Ap1 (C) antibodies. Note that the medulla section shown in the large images contains punctae that are Hrb+/Eps15+/Ap1+ or Ap1+. The insets show a distinct medulla section with several Hrb+/Eps15+ punctae. (D) Merged images. (E through J) Confocal images of representative Hrb−/− round spermatids double-labeled with Eps15 (E) and (H) and Ap1 (F) and (I) antibodies. (G) and (J) Merge images. Bar, 5 μm. (K) Interaction between Eps15 and Hrb in mouse testis. Immunoprecipitations were performed with affinity-purified antibodies to Hrb(386-562) (lane 2) or to control antibody [Rae1(145-368)] (lane 1) (8). Testis lysate was loaded to mark the position of Eps15 on the blot (lane 3).

Our studies demonstrate the critical role of Hrb in acrosomic vesicle formation during early mouse spermiogenesis. The primary defect in proacrosomic vesicle fusion seems to affect male fertility at the levels of sperm-egg fusion, sperm movement, and sperm number. The sperm acrosome is known to be critical for sperm-egg fusion (1), and the complete lack of this organelle in Hrb−/− spermatozoa explains why no sperm-egg fusion occurred in our in vitro fertilization experiments. The poor motility of Hrb−/− spermatozoa is apparently due to lack of the mitochondrial sheath in the tail midpiece. Formation of this tail structure is dependent on sperm elongation, which, in turn, depends on acrosomic vesicle formation (2). Hence, poor motility of Hrb−/− spermatozoa is probably a secondary consequence of the step 2 arrest in acrosome biogenesis. A decrease in sperm count appears to be a common phenotype for a variety of genetic defects perturbing the normal differentiation program of mouse spermiogenesis (11). Therefore, the reduced sperm count of Hrb−/− males is probably also a secondary effect of the acrosomal defect. Because healthy heterozygous and homozygous female and heterozygous male mice can easily transmit Hrb null alleles to the progeny, similar mutations in humans may also be highly transmissive. They may, in fact, account for some of the idiopathic human infertility syndromes associated with reduced sperm count and production of round-headed spermatozoa (3). The developmentally abnormal round spermatids from Hrb−/− males may be valuable for further testing and improvement of human infertility treatments that are based on round spermatid injection (12).

  • * To whom correspondence should be addressed. E-mail: vandeursen.jan{at}


View Abstract

Navigate This Article