Targeted disruption of hormone-sensitive lipase results in male sterility and adipocyte hypertrophy, but not in obesity

Abstract

Hormone-sensitive lipase (HSL) is known to mediate the hydrolysis not only of triacylglycerol stored in adipose tissue but also of cholesterol esters in the adrenals, ovaries, testes, and macrophages. To elucidate its precise role in the development of obesity and steroidogenesis, we generated HSL knockout mice by homologous recombination in embryonic stem cells. Mice homozygous for the mutant HSL allele (HSL-/-) were superficially normal except that the males were sterile because of oligospermia. HSL-/- mice did not have hypogonadism or adrenal insufficiency. Instead, the testes completely lacked neutral cholesterol ester hydrolase (NCEH) activities and contained increased amounts of cholesterol ester. Many epithelial cells in the seminiferous tubules were vacuolated. NCEH activities were completely absent from both brown adipose tissue (BAT) and white adipose tissue (WAT) in HSL-/- mice. Consistently, adipocytes were significantly enlarged in the BAT (5-fold) and, to a lesser extent in the WAT (2-fold), supporting the concept that the hydrolysis of triacylglycerol was, at least in part, impaired in HSL-/- mice. The BAT mass was increased by 1.65-fold, but the WAT mass remained unchanged. Discrepancy of the size differences between cell and tissue suggests the heterogeneity of adipocytes. Despite these morphological changes, HSL-/- mice were neither obese nor cold sensitive. Furthermore, WAT from HSL-/- mice retained 40% of triacylglycerol lipase activities compared with the wild-type WAT. In conclusion, HSL is required for spermatogenesis but is not the only enzyme that mediates the hydrolysis of triacylglycerol stored in adipocytes.

Figure 1

Targeted disruption of the HSL gene. (a) Map of the HSL gene and targeting construct. The targeting vector was designed to delete HSL genomic sequences including a part of exon 5 and the entire exon 6 encoding the catalytic motif (GXSXG). Targeted events generate an enzymatically inactive truncated protein lacking the C-terminal part of HSL. Tall boxes represent exons. The probe used for the diagnostic Southern blot is indicated as a shaded box. Restriction sites: X, XbaI; E, EcoRI. (b) Southern blot of EcoRI-digested DNA. The wild-type genomic fragment is 5.9 kb, whereas the mutated gene is 2.6 kb. Genotypes are indicated below each lane. (c) Immunoblot analysis of white adipose tissue (WAT) and testes, showing the absence of immunoreactive HSL in −/− mice. Note that the testicular isoform (HSL_tes) is larger than the adipocyte isoform (HSL_adi) (130 kDa vs. 84 kDa). Molecular masses of the standards are indicated on the left of the gel (5–15% acrylamide). (d) NCEH activities of WAT, brown adipose tissue (BAT) and testes were abolished in HSL−/− (open bars), whereas those of mouse peritoneal macrophages (MPM) elicited by thioglycolate were comparable to wild-type (solid bars). Enzyme activity is expressed as picomoles of substrate hydrolyzed per milligram of protein per minute. Data are represented as mean ± SE. Numbers of mice used are as follows: 4 +/+ and 4 −/− for WAT; 6 +/+ and 5 −/− for BAT; 3 +/+ and 3 −/− for testes; and 7 +/+ and 8 −/− for MPM.

Figure 2

Histology of epididymis (×50) and testis (×100). HSL−/− epididymis contains no sperms but degenerated spermatocytes and spermatids. The thickness of the epithelial layers of the HSL−/− seminiferous tubules was greatly reduced from 12 to 5–7 layers. The number of mature spermatids was also markedly reduced. Extensive vacuolation was observed in the epithelial cells.

Figure 3

Enlarged lipid droplets in adipocytes from BAT (a) and epididymal fat (WAT) (b) in wild-type and HSL−/− mice. (×50.) Distribution of cytoplasmic area is shown in the graphs: ●, wild-type; ○, HSL−/−. Median cytoplasmic area was increased by about 5-fold in BAT and by 2-fold in WAT.

Figure 4

The redundancy of triacylglycerol lipase. (a) Triacylglycerol lipase activities in WAT and BAT from the wild-type (solid bar) and HSL−/− mice (open bar) (n = 5). (b) In vitro lipolysis in adipocytes isolated from epididymal fat pad of wild-type (+/+; n = 12) and HSL−/− mice (−/−; n = 8). Adipocyte suspensions were treated with 100 μM isoproterenol (solid bar; ISO) or without it (open bar; none) for 15 min. Glycerol and FFA concentrations in the media were determined. (c) In vivo lipolysis. After a 12-h fast, 0.3 mg/kg of isoproterenol (solid bar; ISO) or saline (open bar; Saline) was injected intraperitoneally into control (+/+; n = 7) and HSL−/− (−/−; n = 10) mice. After 15 min, blood was collected and plasma concentrations of glycerol and FFA were determined. Data are expressed as mean ± SE. Mice used in these experiments were 12–16 weeks old. *, P < 0.001; **, P < 0.05 by Student's t test.