Characterizing the meiotic spindle configuration and chromosome complement of in vivo matured equine oocytes

1. Introduction

The incidence of embryonic loss between days 12 and 60 of gestation is approximately 10% for mares <12 years old and then increases to between 20% and 30% (or higher) in mares > 18 years old (Vanderwall, 2008). Thus, early embryonic failure is a major cause of reproductive inefficiency and considerable economic loss. Therefore, understanding the factors that contribute to early pregnancy loss, particularly in aged mares, has important implications for reproductive management. There is compelling evidence that oocytes from mares >18 years of age have a high incidence of inherent defects that result in early embryonic loss (Vanderwall, 2008). In older women, decreased oocyte quality and developmental potential are associated with several factors, including abnormal chromosome–microtubule interactions at the meiotic spindle that lead to chromosome segregation errors and a dramatically increased incidence of aneuploidy (Battaglia et al., 1996). We hypothesize that similar age-related alterations in meiotic spindle organization and chromosome interactions occur in the oocytes of aged mares, and that these contribute to the high incidence of early embryonic loss in older mares. The objective of this study was to evaluate three fluorescent antibody probes for the analysis of meiotic spindle organization and chromosome complement in equine oocytes from young fertile mares, preparatory to comparing these characteristics in young and aged mares.

2. Materials and methods

Mares (n = 8) used in this study ranged from 4 to 10 years old and weighed 300–500kg. The reproductive tract of each mare was examined by transrectal palpation and ultrasonography. When mares developed an ovarian follicle ≥35mm in diameter in conjunction with prominent endometrial edema, they were injected with 2500 IU human chorionic gonadotropin IV, and subsequently used for oocyte collection via transvaginal ultrasound-guided follicle aspiration (TVA) 28–32 h later. Transvaginal ultrasound-guided follicle aspiration was performed as previously described (Vanderwall et al., 2006) except that mares received n-butylscopal-ammonium bromide (Buscopan™, 0.1–0.4mg/kg, IV) to induce relaxation of the rectum, and the follicle was aspirated and then irrigated with ViGro Complete Flush Solution (Bioniche Animal Health USA, Inc., Pullman, WA) containing 10 U/mL heparin using a 60cm, 12-gauge double-lumen ovum pickup set (Specialized Medical Products, Inc., Fruit Heights, UT). On completion of the TVA the effluent from the follicle was immediately transported to the laboratory for oocyte identification and evaluation.

The oocyte was transferred to minimal essential medium (MEM) supplemented with 3mg/mL bovine serum albumin for removal of cumulus cells by brief exposure to 1mg/mL hyaluronidase and vigorous pipetting. For fixation, the oocyte was transferred to a pre-warmed 4% solution of paraformaldehyde for 10 min, followed by permeabilization in a 2% paraformaldehyde solution supplemented with 0.2% Triton X for 10 min and subsequently blocked overnight in 10% fetal calf serum in PBS and 0.2% Tween-20 (PBT). Oocytes were immuno-stained with a rabbit polyclonal antibody against histone H3 tri-methylated on lysine 9 (H3K9me3) for chromosome labeling as well as with a mouse monoclonal anti-β-tubulin antibody to stain the meiotic spindle microtubules. To count chromosomes, surface spread metaphase II oocytes were simultaneously stained with H3K9me3 and CREST antiserum to detect chromosomes and centromere–kinetochore complexes, respectively (De La Fuente et al., 2004). To reduce background fluorescence, oocytes were washed several times in PBT before exposure to a 1:1000 dilution of an appropriate combination of Alexa fluor 555 goat anti-rabbit, Alexa fluor 488 goat anti-human or an Alexa fluor 488 goat anti-mouse IgG. Detailed meiotic spindle and chromosome analyses were conducted using a Leica TCS-SP5 laser scanning confocal microscope.

3. Results

Fourteen follicles were aspirated from which 11 oocytes were recovered (78.6%). Ten oocytes were analyzed immediately after collection and one, with a compact cumulus, was incubated in vitro for 30 h (De La Fuente et al., 2004) before surface spread preparation for chromosome analysis. The meiotic progression of the 10 fresh oocytes revealed that 4 were at the germinal vesicle/germinal vesicle breakdown stage, 2 at metaphase I, 1 at telophase I, and 3 at metaphase II. In the 9 oocytes analyzed immunochemically, antibodies against H3K9me3 (red) reliably labeled meiotic chromosomes, antibodies against β-tubulin (green) detected meiotic spindle microtubules, and the two antibodies in combination allowed fine-structural analysis of the chromatin configuration at different stages of meiotic progression (Fig. 1A and B). In addition, chromosome alignment at the metaphase II plate in 3D reconstructions of Z-stack confocal micrographs (Fig. 1C) was found to be appropriate in all metaphase I (n = 2) and metaphase II oocytes (n = 3).

Fig. 1.

(A) A metaphase I equine oocyte showing spindle microtubules (green) and chromosomes (red). Numerous granulosa cells surround the oocyte. Scale bar = 10 μM. (B) A metaphase II equine oocyte showing spindle microtubules (green) and chromosomes (red). The arrow indicates the position of the meiotic spindle (see inset). The position of the polar body (pb) is indicated. Several granulosa cells surround the oocyte. Scale bar = 10 μM. (C) 3D reconstruction and digital rotation of the metaphase II spindle to facilitate the analysis of chromosome alignment (normal in this oocyte). Text beneath each image in the panel indicates the specific rotational position number in the series (i.e., 1, 4, 7, 10, 13, 16 and 19). (D) High resolution analysis of chromosome spreads (two images of the same oocyte). The site of microtubule attachment (centromere) is labeled in green (arrowheads). The presence of 64 CREST signals (two per chromosome) indicates this oocyte has the correct (haploid-1N/2C) number of chromosomes (32).

In the 2 oocytes used to count chromosomes, H3K9me3 immunochemistry (red) revealed histone methylation throughout the chromatids with a prominent enrichment at centromeric domains (Fig. 1D). CREST antiserum (green) detected two sister kinetochore domains (two CREST signals per chromosome, Fig. 1D). Both oocytes had the appropriate haploid chromosome complement (64 CREST signals, 32 chromosomes, Fig. 1D).

4. Discussion

Both nuclear and cytoplasmic maturation are required to confer the mammalian oocyte with full developmental potential. Functional differentiation of chromatin structure during oocyte maturation is required for proper chromosome segregation and maintenance of chromosome stability during embryonic development (De La Fuente, 2006). Our results indicate that equine meiotic chromosomes exhibit prominent histone methylation at centromeric domains, as in mouse metaphase II oocytes (De La Fuente et al., 2004). To the best of our knowledge, this is the first demonstration that a chromatin modification associated with centromere function is conserved, and that the centromere–kinetochore marker CREST can be reliably used for analyzing chromosome–microtubule interactions, in the equine oocyte. In agreement with recent work by Siddiqui et al. (2009), the majority of in vivo matured oocytes at the metaphase I and metaphase II stages in the present study exhibited chromosomes aligned at the equatorial region of a bipolar spindle characterized by an abundant microtubular network.

The three chromosomal markers proved applicable to the assessment of equine oocyte quality and chromatin configuration by laser scanning confocal microscopy. They should facilitate future studies of oocyte structure and quality after in vivo or in vitro maturation, and the etiology of declining fertility in older mares. In particular, the markers will allow us to compare microtubule interactions at the meiotic spindle, and the chromosome complement, in oocytes from young and aged mares.