Proliferating Cell Nuclear Antigen—A Marker for Ovarian Follicle Counts

Abstract

Enumerating ovarian follicles is an effective way to estimate the extent of ovarian toxicity in female rodents exposed to xenobiotics. Differential follicle counts are useful in safety assessment bioassays and in interspecies extrapolation of ovarian toxicity. Counting the follicles in H&E-stained sections is labor intensive, tedious, and costly. In the present study we demonstrated that in rat formalin-fixed, paraffin-embedded ovary sections follicles of all degrees of maturity can be visualized by the use of antibody directed against proliferating cell nuclear antigen (PCNA). Follicles are easily distinguished from ovarian background with the ability to detect and identify primordial follicles being enhanced. This translates into a significant decrease in variability of follicle counts, labor, and cost. Specifically, variability dropped from 11% to 0.2%, the counting time was reduced by 46%, and the cost by 48%.

Introduction

The ovary is a major target of xenobiotics that affect female fertility (Mattison and Thomford, 1989). U.S. and internationally harmonized Health Effects Test Guidelines for Reproduction and Fertility Effects include enumeration of primordial and developing ovarian follicles as endpoints of safety tests (OPPTS Harmonized Test Guidelines, 1998). The number of these structures is also of interest for other aspects of reproductive biology.

Performing ovarian counts microscopically on representative hematoxylin and eosin-stained (H&E) sections of ovary is routinely done by technicians and is tedious and error-prone (Bolon et al., 1997; Bucci et al., 1997). This occurs primarily because primordial follicles are easy to overlook. Recently we described the immunohistochemical method to facilitate counting of ovarian follicles (Muskhelishvili et al., 2002), in which we identified a rabbit polyclonal antibody directed against human cytochrome P450 1B1 (CYP1B1) that marks rodent oocyte nuclei in addition to nuclei of some ovarian granulosa and theca cells. However, the antibody is not marketed and can only be custom prepared upon request. This process is time-consuming and costly, and thus, inconvenient for investigators willing to obtain the antibody. This prompted a further search for a more easily available antibody that can be used for labeling ovarian follicles.

Immunohistochemical labeling of oocytes in rat ovaries with antibody directed against proliferating cell nuclear antigen (PCNA) has been reported by Oktay et al. (1995). According to this study, PCNA immunoreactivity coincided with the earliest sign of follicle growth, appearing in pregranulosa cells of early primary follicles just beginning to grow. In primordial follicles, neither granulosa cells nor oocytes stained for PCNA. PCNA immunoreactivity in oocytes first appeared in primary follicles, preceding oocyte enlargement, and was observed in all following stages of follicle development. These results are in close agreement with the data obtained by Wandji et al. (1996), who used cultured pieces of bovine ovarian cortex, rich in primordial follicles. In this study, primordial follicles in Bouin’s-fixed preculture tissue sections also did not stain for PCNA. After 2, 4, and 7 days in culture, PCNA was expressed intensely in the oocytes and granulosa cells of primary follicles. Taken together, results of these 2 studies suggested that expression of PCNA in granulosa cells and oocytes coincides with the initiation of follicle growth.

However, in our laboratory, weak PCNA immunoreactivity in primordial oocytes of rodent ovaries was commonly observed during routine proliferation studies. It could be hypothesized that in oocytes of primordial follicles PCNA is expressed at lower levels compared to the more mature follicles and therefore is harder to visualize. In the present study we show that PCNA is strongly visualized by the use of heat-induced epitope retrieval (HIER) technique and high concentration of primary antibody in oocytes of primordial follicles of rat formalin-fixed, paraffin-embedded ovaries.

Materials and Methods

Animals

80-day-old female Sprague–Dawley rats (n = 6) were sacrificed by CO₂ asphyxiation. Ovaries were removed, fixed in neutral buffered 10% formalin for 48 hours and embedded in paraffin. Following the most recent and internationally harmonized Health Effects Test Guidelines (1998), sections (5 μm) were taken from 5 levels 100 μm apart in the middle third of each ovary. Specifically, both ovaries were butterflied to sample the central plane of each and embedded in paraffin. Two sections were cut serially at 5 consecutive 100 μm intervals. Each specimen was placed on a separate slide resulting in two sets, each composed of 5 slides with nearly identical ovarian sections for follicle counting. One set of sections was stained with H&E and the other was stained immunohistochemically to detect PCNA.

Immunohistochemistry

For immunohistochemical demonstration of PCNA deparaffinized tissue sections were placed in an antigen retrieval solution (0.01 M citrate buffer, pH 6.0) for 15 minutes in a microwave oven at 100°C at 600 W. Endogenous peroxidase was inhibited by incubation with freshly prepared 3% hydrogen peroxide with 0.1% sodium azide for 10 minutes at room temperature (RT). Nonspecific staining was blocked with 0.5% casein for 20 minutes at RT. The sections were then incubated with mouse monoclonal anti-PCNA (clone PC10, DAKO, Carpinteria, CA) at a dilution of 1:800 (0.5 μg/ml) for 1 hour at RT. After incubation with primary antibody tissue sections were incubated with biotinylated goat anti-mouse IgG F(ab)₂ fragments (Rockland, Gilbertsville, PA) at a dilution of 1:600 for 30 minutes at RT, and later with streptavidin-conjugated horseradish peroxidase (ExtrAvidin Kit, Sigma, St. Louis, MO) at 1:30 dilution for 30 minutes at RT. Staining was developed with diaminobenzidine (DAB, Sigma) substrate for 5 minutes at RT, sections were counterstained with hematoxylin, and mounted with Permount (Fisher Scientific, Pittsburgh, PA). For the negative control, 0.5 μg/ml mouse IgG (Jackson Immunoresearch, West Grove, PA) or phosphate-buffered saline (PBS) replaced the primary antibody. For preparation of working dilutions of the primary antibody, biotin-conjugated secondary antibody, streptavidin-conjugated horseradish peroxidase label, and mouse IgG, 1% bovine serum albumin (Sigma) in PBS was used as a diluent. Between the steps of the staining procedure slides were washed in PBS for 5 minutes at RT.

Follicle Counts

All sections were examined by light microscopy (BX40, Olympus, Japan). Follicles in the 2 slide sets (H&E and anti-PCNA) were counted using a 20× objective. The counts included the total number of follicles categorized as either: (1) small follicles—primordial follicles with oocyte and one or more flattened pregranulosa cells at the periphery of the follicle and primary follicles with oocyte having one complete ring of cuboidal granulosa cells; (2) growing follicles—enlarging oocyte in an enlarging follicle having more than one complete ring of granulosa cells; or (3) large (antral) follicles—full-sized oocyte, several hundred granulosa cells, and fluid-containing antral space. Only follicles in which the intact oocytic nucleus was observed were counted. This method, coupled with the procedure of sections being taken 100 μm apart, minimized likelihood of counting the same follicle twice or counting a follicle in which the oocyte had degenerated. Counts were performed by two technicians independently. Each of them recorded time spent on counting the follicles in each ovary section stained with H&E and anti-PCNA, separately.

Results

The results demonstrated strong nuclear immunostaining for PCNA in all oocytes of small, growing, and large follicles, and in proliferating granulosa and theca cells (Figure 1). We were unable to identify any primordial follicle with an unstained oocyte. In some primordial follicles, PCNA immunoreactivity was present only in oocytes, while squamous pregranulosa cells appeared stain-free (Figure 1, insert). This suggests that PCNA is expressed by oocyte before the follicle is selected to grow. No immunostaining was observed in negative controls.

There were no statistically significant differences between the numbers of ovarian follicles counted in PCNA- and H&E-stained serial sections of the same animal ( p > 0.1, with n = 5 for each ovary). However, the technicians spent significantly less time doing ovarian follicle counts in the PCNA-immunostained sections, compared to the H&E-stained sections. Specifically, counting time was reduced by 46%. Variability in counts between the two technicians in the H&E-stained sections was 11%, and it was minor (0.2%) in the PCNA-immunostained sections. The total cost of the counts (including cost of reagents, time spent on slide preparation and counting) was decreased by 48% by the use of PCNA compared to H&E (Table 1).

Discussion

It is known that detectable levels of PCNA can vary significantly among different cell types, as well as depending on the fixatives and the antigen retrieval solutions used (Morris and Mathews, 1989; Coltrera and Gown, 1991; Schwarting, 1993; Scholzen and Gerdes, 2000; Muskhelishvili et al., 2003). It appears that in oocytes of primordial follicles, PCNA is expressed at lower levels compared to the more mature follicles and was not detected in the study by Oktay et al. (1995), due to the following: (a) ovaries were fixed in Bouin’s solution, which is known to sometimes render subsequent immunohistochemical staining difficult, and b) none of currently available antigen retrieval procedures were used.

In the present study, we fixed rat ovaries in formalin and used the HIER technique. HIER is commonly applied to facilitate PCNA immunostaining of proliferating cells in formalin-fixed, paraffin-embedded tissues (Foley et al., 1993; Muskhelishvili et al., 2003). In order to maximize the signal in primordial oocytes we used approximately a 6-fold higher concentration of primary antibody compared to that usually used for proliferation assay in our laboratory (0.5 μg/ml vs 0.08 μg/ml IgG, respectively). Clearly these adjustments increased the intensity of the immunostaining and as a result oocytes of primordial follicles appeared strongly labeled.

Although the use of a high concentration of primary antibody renders substantial background staining in proliferating granulosa and theca cells, strong PCNA immunostaining of oocyte nuclei in combination with their large size makes oocytes easy to distinguish from other labeled cells; ability to detect and identify primordial follicles is particularly enhanced. Thus, in PCNA-stained sections, follicles of all degrees of maturity are much easier to distinguish from ovarian background than in H&E-stained sections (Figure 1).

These results demonstrate that PCNA is a useful marker for ovarian follicle counts in rats. The ability to mark oocyte nuclei distinctly with PCNA antibody significantly increases speed and accuracy of counting, that results in decreased labor and cost. The labor and costs will be decreased even more when counts are done using semi-automated image analysis.

PCNA is an auxiliary protein of DNA polymerases δ and ɛ, enzymes necessary for DNA synthesis and is thus a marker of proliferation (Kurki et al., 1986; Bravo et al., 1987; Wood and Shivji, 1997). However, PCNA expression in oocytes cannot be attributed to cell proliferation since the oocyte is arrested meiotically (Hirshfield, 1991). The significance of the PCNA expression in the oocyte is unknown and needs to be elucidated. It has been demonstrated that PCNA is involved in DNA repair (Celis and Madsen, 1986; Toschi and Bravo, 1988; Shivji et al., 1992; Wood and Shivji, 1997), which suggests that PCNA may be expressed by cells that are not proliferating. Correspondingly, it has been hypothesized that although no new DNA synthesis takes place in the growing oocyte, it is possible that DNA polymerases are activated to repair potential damage to the genetic material in the oocytes selected to grow (Oktay et al., 1995). The presence of PCNA in oocytes of primordial follicles observed in our study suggests a role for this protein even in earlier stages of folliculogenesis. PCNA expression levels in follicles of different maturity can be evaluated precisely by the use of laser capture microdissection and real time RT-PCR approach.