Expression of retinoic acid receptor alpha in the germline is essential for proper cellular association and spermiogenesis during spermatogenesis

Abstract

Signaling through vitamin A metabolites is indispensable for spermatogenesis, and disruption of retinoic acid receptor alpha (RARalpha) function resulted in male sterility and aberrant spermatogenesis, which resembled vitamin A deficiency. Here we investigated the lineage- and cell-specific role of RARalpha-mediated signaling during spermatogenesis using germ-cell transplantation and genetically manipulated mouse models. We demonstrated that RARalpha-deficient germ-cell stem cells were able to repopulate germ-cell-depleted wild-type testes and initiate spermatogenesis; however, improper cellular associations and abnormal sperm formation were observed. We further generated RARalpha-deficient mice that expressed RARalpha-EGFP fusion protein uniquely in haploid germ cells. Strikingly, spermatid orientation, alignment and release, as well as sperm morphology, were normal and there was a partial rescue of sterility. These data provide the first direct evidence for a distinct requirement of RARalpha-mediated retinoid signaling specifically in germ cells.

Fig. 1.

Rara^-/- spermatogonial stem cells repopulated and initiated spermatogenesis in germ-cell-depleted recipient testis. (A-F) Spermatogonial stem cells from Rara^+/+; Actb-EGFP⁺ (A) and Rara^-/-; Actb-EGFP⁺ (B-F) mice were transplanted into the tubules of busulfan-treated nude mice. Representative sections of the testes at 6 months after transplantation viewed under a fluorescent microscope at low (B) and higher (C) magnification showed repopulated green-fluorescing Rara^-/- cells in some but not all tubules, indicating the successful colonization of transplanted cells. The adjacent serial section, stained with Hematoxylin, is shown at higher magnification (D). C represents a higher magnification of the insert in B; E and F show higher magnification of the inserts in C and D. Arrows in E and F indicate elongated spermatids. (G,H) Abnormal sperm heads were observed in the epididymis after transplantation (H) compared with normal sickle-shaped spermatozoa in control epididymis (G). Magnification: B, ×10; A,C,D, ×20; E-H, ×60. (I) Graph showing the percentages of tubules in recipient testis with restored spermatogenesis. Individual counts from each testis are indicated with squares and triangles; the mean of each data set is plotted with a horizontal bar. The number of repopulated tubules from the Rara^-/- transplantation was similar to that from the Rara^+/+.

Fig. 2.

Rara^-/- germ cells supported by Rara^+/+ somatic cells exhibit improper cellular associations. (A-F) Representative fluorescent sections of testes at 5 months after transplantation showing repopulated green-fluorescing Rara^-/- germ cells in some but not all tubules, as well as the repopulated endogenous non-fluorescing Rara^+/+ germ cells (A,E). The same sections, counterstained with DAPI, are shown (B,F, respectively). C and D are higher magnifications of the tubules in B, illustrating various spermatogenic cell layers. The outer dashed white line encircles the outline of seminiferous tubules, spermatogonia, pre-leptotene and/or leptotene spermatocytes; the red dashed line encircles pachytene spermatocytes; the green dashed line encircles round spermatids; and the inner dashed white line encircles elongated spermatids. (C) Stage VIII tubules containing four layers of cells, with an outer layer of pre-leptotene spermatocytes, pachytene spermatocytes, step 8 round spermatids and step 16 elongated spermatids at the innermost layer. The round spermatids contained characteristic large and densely stained nucleoli (insert in C). An adjacent tubule (at the end of stage VIII and the beginning of stage IX) contained three layers of cells: pre-leptotene/leptotene spermatocytes at the basal lamina, pachytene spermatocytes in the middle, and step 8-9 spermatids at the innermost layer. Roman numerals in this figure and in Figs 3, 4, 5, 6 indicate the stage of the tubules (Russell et al., 1990). (G) Spermatogenic cell distribution in repopulated tubules transplanted with wild-type germ cells (WT GC) and Rara^-/- germ cells (Rara^-/-GC). The total number of various cell types [pachytene spermatocytes and diplotene spermatocytes (PS and D); step 1-8 round spermatids (Step 1-8)] per 40 repopulated tubules were counted. Error bars represent the mean±s.d. of the counts. ^***, P<0.01 and ^**, P<0.05. Magnification: A,B, ×20; C-F, ×40. ES, elongated spermatids; L, leptotene spermatocytes; P, pachytene spermatocytes; PL, pre-leptotene spermatocytes; PL/L, pre-leptotene/leptotene spermatocytes; RS, round spermatids.

Fig. 3.

Transgenic model to overexpress RARα/EGFP in round and elongated spermatids. (A) Diagram of the Prm1-Rara/EGFP transgene. Prm1, a spermatid-specific promoter, was used to drive expression of an Rara cDNA fused in-frame to EGFP coding sequences (construct: Prm1-Rara/EGFP-Prm1 poly A and intron 1). (B) Expression of EGFP in the testis of a Prm1-Rara/EGFP transgenic mouse at 6 weeks of age. Intact testes of transgenic and wild-type mice were observed under fluorescent light at 6 weeks (a,b, respectively). Representative sections examined under a fluorescent microscope (c,d) revealed that green-fluorescing cells were detected in some but not all tubules, suggesting the transgene expression is specific to particular stages of spermatid differentiation (a,c) and that there is no expression in the wild-type testis (b,d). Immunohistochemical detection of EGFP on histological sections of 6-week-old testes from transgenic mice (e-g). EGFP was detected in nuclei of round spermatids at stage VII (f), elongating spermatids at stage IX-XII (e-g), and elongated spermatids at stage I-VIII (e-g) of adult transgenic but not control (h) testes. Magnification: c-h, ×40. Arabic numerals indicate the step of spermatids shown. D, diplotene spermatocytes; L, leptotene spermatocytes; P, pachytene spermatocytes; PL, pre-leptotene spermatocytes; Z, zygotene spermatocytes.

Fig. 4.

Expression of RARα in round and elongated spermatids can rescue spermiogenesis in Rara^-/- testes. (A-H) Histological sections of testes from wild-type (A), Rara^-/- (C), rescued Rara^-/-; Prm1-Rara/EGFP (E) and non-rescued Rara^-/-; Prm1-Rara/EGFP (G) mice at 4 months of age are shown with their corresponding epididymides (B,D,F,H, respectively). Inserts in B, D, F and H show higher magnifications. (I,J) Immunostaining of rescued Rara^-/-; Prm1-Rara/EGFP (I) and non-rescued Rara^-/-; Prm1-Rara/EGFP (J) testes are shown using Brdt (I,J) antibodies. Asterisks mark vacuoles in the tubules. (K) Spermatogenic cell distribution in rescued Rara^-/-; Prm1-Rara/EGFP and non-rescued Rara^-/-; Prm1-Rara/EGFP mice. The total number of various cell types [step 1-8 round spermatids (Step 1-8); pachytene spermatocytes and diplotene spermatocytes (PS and D)] per 100 seminiferous tubules was counted. Error bars represent the mean±s.d. of the counts. ^**, P<0.05. Arabic numerals indicate the step of spermatids shown. Magnification: A-J, ×40. ES, elongated spermatids; L, leptotene spermatocytes; MI/II, meiosis I/II; non-res, non-rescued mice; P, pachytene spermatocytes; PL, pre-leptotene spermatocytes; res, rescued mice; RS, round spermatids; Z, zygotene spermatocytes.

Fig. 5.

Successful spermatid orientation, alignment and release at stage VIII and IX of the spermatogenic cycle in rescued adult testes and morphology of isolated spermatozoa from rescued adult mice. (A,B) Histological sections from the testis and epididymis of 4-month-old rescued mice. Successful spermatid alignment and release at stage VIII (A) and XI (B). (C-G) Spermatozoa from wild-type (C), mutant (D-F) and rescued (G) mice were examined after one step Eosin-Nigrosin staining. Defective sperm heads in mutant mice were noted (D-F), with bent or blunted sharp (E), or round or ovoid (F) heads. By contrast, normal sickle-shaped heads with apical hooks were found in rescued males, similar to wild type (G versus C). Magnification: ×60. The arrow points to the thinned mid-piece of the tail. Arabic numerals indicate the step of spermatids shown. P, pachytene spermatocytes; PL/L, pre-leptotene/leptotene spermatocytes.

Fig. 6.

Sloughing of spermatogenic cell layers in the testes of older rescued mice. (A,B) Histological sections of testes and epididymis of control male at 8 months. (C,D) As early as 8.5 months, striking sloughing of spermatogenic cell layers was detected in non-rescued testes (C) and no spermatozoa were found in the corresponding epididymis (D). (E-H) Histological sections from testes of rescued Rara^-/-; Prm1-Rara/EGFP mice at 8.5 months (E) and 15.5 months (G) are shown with their corresponding epididymides (F and H, respectively). (I,J) Immunostaining of testes of rescued but now infertile Rara^-/-; Prm1-Rara/EGFP mice using GFP antibody. J represents a higher magnification of the insert in I. Asterisks in C, E and G mark the vacuoles in the tubules. Magnification: A-I, ×40; J, ×60. ES, elongated spermatids; non-res, non-rescued mice; P, pachytene spermatocytes; res, rescued mice; RS, round spermatids.

Fig. 7.

RARα in the somatic-cell lineage is essential for normal donor germ-cell stem cells to colonize the Rara^-/- germ-cell-depleted recipient seminiferous tubules. (A-H) Green-fluorescing cells were detected in some but not all tubules of Rara^+/+ (A) and Rara^+/- (C,E) germ-cell-depleted recipient testes, as examined 3.5 months after transplantation, indicating the successful colonization of transplanted cells. The right panel showed the same tubules as the corresponding left panel but counterstained with DAPI (B, D and F, respectively). No green-fluorescing cells were noted in the Rara^-/- germ-cell-depleted recipient testes examined 3.5 months after transplantation (G), and Sertoli-cell-only tubules were noted after the same tubules were counterstained with DAPI (H). A-H, ×20. ES, elongated spermatids; P, pachytene spermatocytes; RS, round spermatids.