Pituitary-specific overexpression of porcine follicle-stimulating hormone leads to improvement of female fecundity in BAC transgenic mice

Abstract

Follicle-stimulating hormone (FSH) is a pituitary glycoprotein that, together with luteinizing hormone, plays a crucial role in ovarian folliculogenesis and female fertility. We previously found that FSH beta is a major gene controlling high prolificacy of Chinese Erhualian pigs. To directly study the biological effects on reproductive function of porcine FSH (pFSH) for polyovulatory species, we generated a novel gain-of-function mouse model using a bacterial artificial chromosome (BAC) system to jointly introduce 92 kb and 165 kb genomic fragments comprising the pFSH α- and β-subunit genes. These directed the physiological expression of pFSH with the same temporal and spatial pattern as endogenous FSH in female transgenic (TG) mice. Serum levels of biologically active pFSH heterodimers in independent TG lines ranged from 6.36 to 19.83 IU/L. High basal pFSH activity led to a significant reduction of serum LH and testosterone levels in TG females compared to wild-type (WT) littermates, yet endogenous FSH and estradiol levels were significantly elevated. Interestingly, ovarian histology showed that the number of corpora lutea was significantly higher at 14 and 28 weeks of age in TG females and breeding curves revealed that mean litter sizes of TG females were obviously larger than for WT littermates before 52 weeks of age. These findings indicate that pituitary-specific overexpression of pFSH within physiological boundaries can increase ovulation rate and litter size, but it does not cause reproductive defects. Therefore, our TG mouse model provides exciting insights for investigating the actions of pFSH in vivo.

Figure 1. Pulsed-field gel electrophoresis (PFGE) and schematic representation of BAC clones spanning pFSHα and pFSHβ.

(A) Lane 1: λ-HindIII marker; lanes 2–5: NotI-digested BAC183O11 DNA at different concentrations (3 µg, 0.2 µg, 1 µg, 0.1 µg); lane 6: MidRange II PFG marker; lanes 7–9: NotI-digested BAC412H8 DNA at different concentrations (0.2 µg, 0.8 µg, 0 µg); upper arrows, linear BAC bands; lower arrows, vector bands. (B–C) BAC clones were mapped by a combination of restriction endonuclease digestion, PFGE, direct DNA sequencing and alignment with the published gene sequence. Solid boxes represent gene exons and dashed lines represent two flanking sequences. The positions of the NotI restriction sites are indicated by the vertical bars.

Figure 2. Identification of TG mice and assessment of the integrity of BAC transgenes in TG mice.

(A) PCR for detecting TG mice. Two pairs of primers were used to amplify the 489 bp and 282 bp inserts. M, 100 bp DNA ladder; 10–37, TG founders; N, genomic DNA from WT mouse as a negative control; P, circular BAC DNA as a positive control. (B) Southern blot analysis of integrated pFSH transgenes. Purified genomic DNA (10 µg) from each TG founder was digested by BamHI overnight at 37°C and analyzed by Southern blotting using probes specific for pFSHα and pFSHβ. 10–37, TG founders; P, porcine genomic DNA as positive control; N, genomic DNA from a WT mouse as a negative control. Band intensity was quantified by densitometry using Quantity-One software (Bio-Rad). (C–D) The two ends of both the BAC412H8 and BAC183O11 transgenes were amplified separately from TG mouse genomic DNA using 5′- or 3′-specific primers.

Figure 3. Tissue expression of pFSHα and pFSHβ in adult TG mice.

(A–B) RT-PCR analysis of RNA extracted from the pituitary and other tissues from mice of the six different TG lines. Gapdh mRNA was used as a control. Ht, heart; Li, liver; Sp, spleen; Lu, lung; Ki, kidney; Br, brain; Pi, pituitary; Ov, ovary; Ut, uterus; Mu, muscle; In, intestine; M, 100 bp DNA ladder; P, upper and middle lanes, expression vector as a positive control and, lower lane, pituitary of a WT mouse; N, upper and middle lanes, pituitary of a WT mouse as a negative control and, lower lane, double-distilled water. (C) Northern blot analysis of transgene expression in adult tissues. Northern hybridization with the indicated cDNA probes was carried out using total RNA (20 µg) isolated from different tissues of TG and WT mice. Ethidium bromide (EtBr) staining of 28S, 18S, and 5S rRNA served as a control for RNA quality. Pituitary samples from four TG animals were detected. (D–E) pFSHα and pFSHβ mRNA levels in the pituitaries of TG and WT mice were analyzed using quantitative real-time PCR with specific primers and are expressed relative to β-actin (internal control). Data were combined from three independent experiments; bars represent means ± SEM (n = 7–8 mice per group).

Figure 4. Effects of pFSHα and pFSHβ expression on body and reproductive organ weights.

(A) Growth curves for TG and WT females from birth to 9 weeks of age. (B–C) Ovary- and uterus-to-body weight ratios (mg/g) for TG (n = 12) and age-matched WT (n = 11) females (19–21 weeks of age). No effect was detected. The results of the analysis of each of the two separate lines were found to be comparable to that of the combined analysis.

Figure 5. Reproductive performance of TG females.

Fertility data on liveborn pups from TG (n = 15) and WT (n = 15) female mice up to 450 days of age; the number of litters per mouse was approximately six. Total pups per litter plotted against age of female on the day of birth showing comparable breeding curves for the TG (solid line) and WT (dashed line) females. The results of the analysis of each of the two separate lines were found to be comparable to that of the combined analysis.

Figure 6. Effects of pFSHα and pFSHβ expression on the number of ovarian corpora lutea.

(A) Ovarian sections (5 µm) from WT and TG females at 14, 28, and 56 weeks of age were stained with H&E. Antral follicles, arrows; corpora lutea, asterisks; scale bars, 2.0 mm. (B) Total number of ovarian corpora lutea during the diestrous stage of the estrous cycle in TG and WT females at 14, 28, and 56 weeks of age; n = 6–8 mice per genotype; ***P<0.001. The results of the analysis of each of the two separate lines were found to be comparable to that of the combined analysis.

Figure 7. Serum FSH, LH, estradiol, and testosterone levels in TG and WT mice.

Blood was collected from TG and WT mice at 19–21 weeks of age during the diestrous stage of the estrous cycle for measurement of mouse FSH (A), LH (B), estradiol (C), and testosterone (D) (n = 18–20 mice per genotype). Data were combined from three independent experiments; bars represent means ± SEM; *P<0.05; **P<0.01. The results of the analysis of each of the two separate lines were found to be comparable to that of the combined analysis.

Figure 8. Pituitary FSHα, FSHβ and LHβ mRNA expression in TG and WT mice.

Mouse FSHα, FSHβ and LHβ mRNA were analyzed using quantitative real-time PCR with specific primers and were expressed relative to a β-actin (internal control) (n = 7–8 mice per genotype). Data were combined from three independent experiments; bars represent means ± SEM; **P<0.01. The results of the analysis of each of the two separate lines were found to be comparable to that of the combined analysis.