Small-molecule inhibition of BRDT for male contraception

Abstract

A pharmacologic approach to male contraception remains a longstanding challenge in medicine. Toward this objective, we explored the spermatogenic effects of a selective small-molecule inhibitor (JQ1) of the bromodomain and extraterminal (BET) subfamily of epigenetic reader proteins. Here, we report potent inhibition of the testis-specific member BRDT, which is essential for chromatin remodeling during spermatogenesis. Biochemical and crystallographic studies confirm that occupancy of the BRDT acetyl-lysine binding pocket by JQ1 prevents recognition of acetylated histone H4. Treatment of mice with JQ1 reduced seminiferous tubule area, testis size, and spermatozoa number and motility without affecting hormone levels. Although JQ1-treated males mate normally, inhibitory effects of JQ1 evident at the spermatocyte and round spermatid stages cause a complete and reversible contraceptive effect. These data establish a new contraceptive that can cross the blood:testis boundary and inhibit bromodomain activity during spermatogenesis, providing a lead compound targeting the male germ cell for contraception.

Graphical abstract

Figure 1

BRDT Inhibition by the BET Bromodomain Inhibitor JQ1 (A) Domain diagram of BRDT. Sequence boundaries for recombinant BRDT(1) are shown in bold. (B) Structure of the active (+)-JQ1 enantiomer. (C) Protein alignment reveals high sequence identity between homologous and orthologous domains. Identical (pink) and similar (blue) residues are highlighted. Major helical elements are depicted above the sequence. The conserved asparagine mediating acetyl-lysine recognition is depicted with a blue star. Contacts between (+)-JQ1 and BRDT(1) are depicted with a black star. (D) Competitive inhibition of human (squares) and mouse (circles) BRDT(1) binding to synthetic biotinylated H4Kac4 by (+)-JQ1 using proximity detection assays (hBRDT(1) IC₅₀ = 11 nM; mBrdt(1) IC₅₀ = 10 nM). (E) ITC data for titration of H4Kac4 into hBRDT(1) (black line) or into a 1:0.8 molar mixture of hBRDT(1) and (+)-JQ1 (red line). The inset shows normalized binding enthalpies corrected for heat of dilution as a function of binding site saturation. Solid lines represent a nonlinear least-squares fit using a single-site binding model. (F) Equilibrium binding constants and binding energies of (+)-JQ1 to human and mouse BRDT bromodomains measured by ITC. See also Data S1 and S2 and Table S1.

Figure 2

Molecular Recognition of BRDT by JQ1 (A) Crystal structure of the complex of (+)-JQ1 with hBRDT(1). The ligand is shown as a Corey-Pauling-Koltun (CPK) model, and the hydrogen bond formed to the conserved asparagine (N109) is shown as yellow dots. Main secondary structural elements are labeled. (B) The acetyl-lysine binding pocket of the N-terminal bromodomain of hBRDT is shown as a semitransparent surface with contact residues labeled and depicted in stick representation. Carbon atoms of (+)-JQ1 are colored blue to distinguish them from protein residues. N109 is labeled in blue, and a unique N-terminal residue is shown in red. (C) The surface structure of the acetyl-lysine recognition pocket of mBrdt(1) is shown in blue, with key residues highlighted in black, N109 highlighted in blue, and a distinct residue shown in red. (D and E) Competitive binding of JQ1 to the acetyl-lysine recognition pocket of hBRDT(1) illustrated by structural alignment to an acetylated histone peptide binding to mBrdt(1) (PDB: 2WP2). (D) Human BRDT(1) is shown as pink ribbons. JQ1 is shown as blue sticks with colored heteroatoms. N109 is highlighted in blue. The diacetylated histone H4 peptide is shown in yellow with colored heteroatoms. (E) The diacetyl-lysine recognition site on hBRDT is shown as a pink translucent surface over stick representations of critical binding residues. Key contact residues are highlighted in black. JQ1 is shown as blue sticks with colored heteroatoms. N109 is highlighted in blue text. The diacetylated histone H4 peptide is shown in yellow with colored heteroatoms. (D and E) JQ1 completely occupies the hydrophobic pocket which engages the diacetylated peptide. See also Table S2.

Figure 3

BRDT Inhibition with a Testis-Permeable Small Molecule Reduces Testis Size and Sperm Counts and Motility (A) Pharmacokinetic analysis of JQ1 in plasma (black lines) and testis (red lines) following a single administration of JQ1 (50 mg/kg IP) to male mice. Data represent the mean ± SD. (B) Gross analysis of testes from 9-week-old mice that received control or JQ1 (50 mg/kg daily) for 3 weeks. (C) Testis weights from mice treated with control or JQ1 (50 mg/kg QD) for 3–6 weeks, 6–9 weeks, or 6–12 weeks of age. Data represent the mean ± SEM and are annotated with P values obtained from a two-tailed t test (^∗p < 0.05). (D) Graphical representation of sperm counts obtained from the entire epididymides of males treated with JQ1 or control from 6–9 weeks of age or the tail of the epididymis of males treated from 6–12 weeks of age with vehicle or JQ1 (50 mg/kg QD). Data represent the mean ± SEM (^∗p < 0.05). (E) Motility of mature sperm obtained from the cauda epididymis of adult males treated with JQ1 (50 mg/kg daily) from 6–12 weeks of age. Data represent the mean ± SEM (^∗p < 0.0001). (F) Pharmacokinetic analysis of JQ1 in plasma (black lines) and testis (red lines) following twice-daily (BID) administration of JQ1 (50 mg/kg IP) to male mice. Data represent the mean ± SD. (G) Testis weights (mg) from mice treated with control or JQ1 (50 mg/kg BID) from 9–12 weeks of age. Data represent the mean ± SEM (^∗p < 0.05). (H) Sperm counts obtained from the cauda epididymis of males treated from 9–12 weeks of age with vehicle or JQ1 (50 mg/kg BID). Data represent the mean ± SEM (^∗p < 0.05). (I) Motility of mature sperm obtained from the cauda epididymis of adult males treated with JQ1 (50 mg/kg BID) or vehicle from 9–12 weeks. Data represent the mean ± SEM (^∗p < 0.001). (J) Cross-sectional area of seminiferous tubules from JQ1-treated mice (50 mg/kg QD for 3–6 weeks or 6–12 weeks, or 50 mg/kg BID for 9–12 weeks, as shown) or control (^∗p < 0.05). (K, L, and M) Male mice treated with JQ1 or vehicle from 6–12 weeks exhibit statistically similar serum levels of (K) testosterone, (L) luteinizing hormone (LH), and (M) follicle-stimulating hormone (FSH). See also Figure S1 and Tables S3 and S4.

Figure 4

Histological Analysis of the Antispermatogenic Effects of JQ1 (A and B) Histology of stage VII seminiferous tubules of testes of 6-week-old mice treated with (A) vehicle control or (B) JQ1 (50 mg/kg QD) from 3–6 weeks of age (40× magnification). Red arrows indicate intertubular Leydig cell islands. (C and D) Histology of stage VII seminiferous tubules of 12-week-old mice treated with (C) control or (D) JQ1 (50 mg/kg QD) from 6–12 weeks. A large mass of sloughed epithelium is observed in the lumen of the tubule in (D). (E–H) Histological analysis of testis tubules from 12-week-old mice treated with vehicle control (E, 40×; G, 100×) or JQ1 (50 mg/kg BID) (F, 40×; H, 100×) for 3 weeks. Whereas the stage VII tubules of the control show an abundance of round spermatids (red arrowheads, G), only a few normal appearing round spermatids are evident in (H) after JQ1 treatment. Abnormal spermatids with large nuclei and abundant cytoplasm (blue arrowheads) and symplasts (black arrows in F and H) are also observed.

Figure 5

Selective Depletion of Germ Cell Transcripts by JQ1 (A) Heatmap representation of the top 25 down- and upregulated genes (Q < 0.05) following treatment of male mice with JQ1 (50 mg/kg from 6–12 weeks). Plk1 and Ccna1 (bold) are downregulated by JQ1 exposure. (B) GSEA of three spermatogenesis gene clusters (Schultz et al., 2003) in the testicular transcriptional profile of JQ1-treated males. (C) GSEA of a canonical MYC-dependent gene set (Zeller et al., 2003) in the testicular transcriptional profile of JQ1-treated male mice. (D) Quantitative RT-PCR analysis of males treated from 6–12 weeks of age with JQ1 or vehicle. The mouse genes are Plzf (promyelocytic leukemia zinc-finger or Zbtb16), Stra8 (stimulated by retinoic acid gene 8), Brdt (bromodomain, testis-specific), Ccna1, Hist1h1t (histone cluster 1, histone 1, testis-specific), Msy2 (Y box protein 2 or Ybx2), Papolb (poly (A) polymerase β or Tpap), Klf17 (Kruppel-like factor 17 or Zfp393), and Prm1 (protamine 1). Data represent the mean ± SEM and are annotated with p values as obtained from a two-tailed t test (^∗p < 0.05; ^∗∗p < 0.001; the p value for Prm1 is 0.06). See also Tables S5 and S6.

Figure 6

Histological Evidence of Antispermatogenic Effects of JQ1 (A and B) Immunohistochemical analysis (40×) of ZPBP1 in stage VII seminiferous tubules of males treated daily from 6–12 weeks with (A) vehicle or (B) JQ1. (C and D) Immunofluorescence analysis (20×) of ZPBP1 in testis tubules of males treated BID from 9–12 weeks with (C) vehicle or (D) JQ1. (E and F) Immunohistochemical analysis (40×) of TNP2 in testis tubules of males treated BID from 9–12 weeks with (E) vehicle or (F) JQ1. (G and H) Exposure to JQ1 has no effect on mitotic progression or meiotic chromatin condensation, as determined by histone 3 phosphoserine 10 (pH3Ser10) staining of testis tubules of males treated daily from 6–12 weeks with (G) vehicle or (H) JQ1. See also Figures S2 and S3.

Figure 7

BRDT Inhibition with JQ1 Causes a Reversible Contraceptive Effect in Male Mice (A) Adult males were pretreated for 6 weeks with vehicle control (n = 7) or JQ1 (50 mg/kg QD; n = 3) and then caged continuously with two females each while continuing 50 mpk QD for month 1 and escalating to 75 mpk QD for month 2. JQ1 treatment was stopped at the end of month 2 of mating. Graphical representation of pups born in each month to the females reveals a contraceptive effect evident in months 1–3 (data represent mean ± SEM; ^∗p < 0.001) and durable restoration of fertility at month 4. (B) Adult males were pretreated for 6 weeks with vehicle (n = 7) or JQ1 (50 mg/kg QD; n = 4) and then caged continuously with two females each while continuing 50 mg/kg QD for month 1, escalating to 75 mg/kg QD for month 2, and further escalating to 50 mg/kg BID for month 3. JQ1 treatment was stopped at the end of month 3 of mating. Graphical representation of pups born in each month to the caged females reveals a complete contraceptive effect evident in months 3–5 (data represent mean ± SEM; ^∗p < 0.001) and durable restoration of fertility at month 6. (C–E) Graphical representation of (C) testis mass (mg), (D) sperm count, and (E) sperm motility of adult control males (n = 3) and JQ1-treated males (n = 3) following a 4 month recovery from the low-dose regimen shown in (A) (data represent mean ± SEM; NS, not significant). (F–H) Graphical representation of (F) testis mass, (G) sperm count, and (H) sperm motility of adult control males (n = 4) and JQ1-treated males after 3 weeks of 50 mg/kg BID treatment (n = 2), as well as following 10 days (n = 2), 30 days (n = 2), and 60 days (n = 2) of recovery from the high-dose (B) regimen (data represent mean ± SEM; ^∗p < 0.05). (I and J) Microscopic analysis of seminiferous tubules from (I) control and (J) JQ1-treated males following 4 months of recovery reveals compete histological recovery. (K) Litter sizes and pups born following JQ1 cessation are normal. See also Figure S4.

Figure S1

Motility and Histological Effect of JQ1 Treatment on the Epididymides of Male Mice, Related to Figure 3 (A) Progressive motility of mature sperm obtained from the epididymis of males treated with JQ1 (50 mg/kg QD) or control from 6-12 weeks of age. Data was obtained using video analysis of sperm on a 37°C heated stage. (B) Treatment of male mice with JQ1 (50 mg/kg QD) for 8 weeks does not affect the mass of seminal vesicles. (C) Histology of the epididymides from male mice treated with control or JQ1 (50 mg/kg QD) from 6-12 weeks of age. Fewer spermatozoa and multiple large nucleated cells (black arrow) are observed in the epididymal lumen of the JQ1-treated mice compared to the control epididymal lumen, which is densely packed with mature spermatozoa.

Figure S2

Histological Characterization of the Testes of Mice Treated with JQ1 or Control, Related to Figure 6 (A and B) Immunohistochemical study (20x) of ZPBP1 in seminiferous tubules of male mice treated daily from 6-12 weeks with vehicle control (A) or JQ1 (B). (C and D) Immunohistochemical study (20x) of TNP2 in testis tubules of male mice treated BID from 9-12 weeks with vehicle control (C) or JQ1 (D). (E–H) Immunofluorescence study (40x) of GASZ in testis tubules of male mice treated BID from 9-12 weeks with vehicle control (E, G) or JQ1 (F, H). GASZ is expressed in nuage granules in pachytene spermatocytes.

Figure S3

BRDT Inhibition by JQ1 Does Not Affect Proliferating Gonocytes and Spermatogonia, Related to Figure 6 (A–C) (A and B) Staining of Cyclin D1 in the epididymides of male mice treated with control or JQ1 from 6-12 weeks of age reveals no effect on the number of stained spermatogonia by quantitative immunohistochemistry as shown in (C). Data represented as mean ± SEM. (D–F) (D and E) JQ1 treatment does not elicit a significant apoptotic response in seminiferous tubules of male mice, compared to vehicle-treated animals, as determined by TUNEL staining and quantitative immunohistochemistry as shown in (F). Data represented as mean ± SEM.

Figure S4

Reversal of Testicular Phenotypes following Withdrawal of JQ1, Related to Figure 7 (A–C) Analysis of (A) testicular mass, (B) sperm counts, and (C) sperm motility of adult male mice treated for 3 weeks with JQ1 (50 mg/kg QD), followed by euthanasia at the end of treatment (JQ1), 2 months following cessation of treatment, and 4 months following cessation of treatment. Normalization of all parameters is evident following drug withdrawal. Data represent mean ± SEM. (D and E) Testis histology (10X) of tubules from control and JQ1-treated male mice 4 months following cessation of JQ1 as treated in (A-C). (F) Graphical representation of tubule area calculated from histological preparations of testis tubules from male mice following four month recovery from the low-dose (Figure 6A) regimen. Data represent mean ± SEM. (G) Graphical representation of tubule area calculated from histological preparations of testis tubules from male mice during and following 10 day, 30 day, and 60 day recovery from a 3 week period of JQ1 BID treatment. Data represent mean ± SEM (^∗, significant at p < 0.05). (H) A single pup born from a JQ1-treated male after halting JQ1 treatment. Normal developmental and behavioral phenotypes are observed in offspring of mice with restored fertility following a period of contraception conferred by JQ1 treatment.

Figure S5

JQ1 Reduces Testis Size, Sperm Counts, and Motility in Male Rats, Accompanied by Histological Evidence of Impaired Spermiogenesis, Related to Figure 4 (A) Pharmacokinetic analysis of JQ1 in the plasma of male rats following a single administration of JQ1 (50 mg/kg IP). Data represent the mean ± SD. (B–D) Graphical representation of (B) testis mass, (C) sperm count and (D) sperm motility derived from adult male Sprague-Dawley rats treated with intraperitoneal JQ1 for 3 weeks, as described in the Experimental Procedures. Reduction of each parameter is consistent with an anti-spermatogenic effect of BRDT bromodomain inhibition by JQ1. Data represent the mean ± SEM (^∗, significant at p < 0.05). (E and F) Testis histology (40x) of seminiferous tubules from control and JQ1-treated male rats. Comparable depletion of maturing spermatocytes and round spermatids is evident in treated Sprague-Dawley rats compared to male mice (as shown in Figure 4).