INTRODUCTION

Many aspects of cellular function are regulated by the concentration of free Ca²⁺ in the cytoplasm. Intracellular organelles play major roles in the regulation of cytosolic free Ca²⁺ levels due to their ability to sequester and release Ca²⁺ following cell activation. The release of Ca²⁺ from stores can occur in response to a first messenger such as neurotransmitters, hormones and sperm (Berridge, 1987; Parrington et al., 1996). In some cases, Ca²⁺ that binds intracellular Ca²⁺-binding proteins may be released.

Calreticulin (CLT) has a high capacity to bind Ca²⁺ in the endoplasmic reticulum (ER); it binds 20 moles of Ca²⁺ per mole of protein at low affinity (Baksh and Michalak, 1991; Nakamura et al., 1993). Full length amino acid sequences of numerous CLTs have been deduced from cDNA (Fliegel et al., 1989; Smith and Koch, 1989; Michalak et al., 1992; Yamamoto et al., 1996). Recently, CLT (Parys et al., 1994) and inositol 1,4,5-trisphosphate receptor (IP₃R) (Callamaras and Parker, 1994) were found in X. laevis eggs. Repetitive Ca²⁺ release is induced by IP₃R activation (Parker and Yao, 1991). In addition, overexpression of CLT inhibits repetitive IP₃-induced Ca²⁺ waves (Camacho and Lechleiter, 1995). In view of these findings, CLT is thought to be one of the key factors in signal transduction of frog eggs. In order to understand its function in eggs, it is urgent to study the gene expression and distribution of CLT in growing oocytes of frogs. Here we report that the CLT gene expression occurs at the early stage of growing oocytes of frogs, and that the CLT distribution is stage-dependent.

MATERIALS AND METHODS

RESULTS

Expression of the CLT gene during differentiation of gonads

The CLT gene expression in gonads of tadpoles of the frog, R. rugosa was examined. When tadpoles at stage IV-V were reared, they grew to be mostly at stage XIII-XVI in 4 weeks (Table 1). Using the total RNA prepared from gonads of those tadpoles, and from ovaries and testes of 2-month post-metamorphosis frogs and adult frogs, Northern blot analysis was carried out with CLT cDNA. The CLT gene expression with a 3.1-kb transcript increased gradually in gonads as tadpoles developed (Fig. 1A). The ovary of 2-month post-metamorphosis frogs produced an extremely strong signal of the CLT gene expression (Fig. 1A), while the β-actin gene expression with a 1.7-kb trabscript did not increase in the ovary before and after metamorphosis (Fig. 1C). Both CLT and β-actin gene expressions in the ovary of adult frogs declined to the level lower than those in the testis of adult frogs (Fig. 1A, C).

Table 1

Developmental stages and body weights of tadpoles

Fig. 1

Expression of the CLT mRNA in gonads. Total RNA was isolated from gonads which were excised surgically one (a), two (b), three (c) and four (d) weeks after tadpoles at stage IV–V were reared. Stages of tadpoles in each group are shown in Table 1. Total RNA was also isolated from testes (e) and ovaries (f) of 2-month post-metamorphosis young frogs, and from testes (g) and ovaries (h) of 2-year old adult frogs. Expressions of the CLT and β-actin genes are shown in the panels A and C, respectively. Total RNA stained with ethidium bromide was shown in the panel B. Alphabetical letters (a–h) on the top of panels A and C correspond to those in the panel B. The number on the right side of panels A and C indicates the size of a transcript. Arrows in the panel B indicate the position of the 28S and the18S ribosomal subunits.

Localization of CLT in oocytes and eggs

As shown in Fig. 1A, the CLT gene expression was extremely strong in the gonad of 2-month post-metamorphosis frogs. Then, the localization of CLT in the ovary was examined immunohistologically with the antibody of CLT. For this purpose, its quality was tested first by Western blot analysis. As shown in Fig. 2, a single band with Mr=52 kDa, estimated by SDS-PAGE, was detected with the antibody. The molecular weight of 52 kDa obtained in this study was the same when the antibody of rat CLT raised against rabbits was used (see Yamamoto et al., 1996). Thus, this antibody was used for immunohistological studies.

Fig. 2

The quality of an antibody of CLT. Ovaries of 2-month post-metamorphosis frogs were homogenized in 50 mM Tris-HCl (pH 7.4), mixed with the sample buffer (Laemmli, 1970) and boiled for 5 min. The sample was then subjected to SDS-PAGE. Fifty μg of protein were loaded on a lane and transferred to nitrocellulose membranes electrophoretically. Then, proteins were reacted for anti-CLT staining, as described in MATERIALS AND METHODS. S, Coomassie blue staining; W, immunostaining.

A faint, but significant staining was seen in the gonad of tadpoles at stage I (Fig. 3a-1), but no staining was observed with non-immune serum (Fig. 3a-2). In the ovary of tadpoles at stage XX, CLT was seen in the peripheral and perinuclear regions in oocytes at stage A (Fig. 3b-1), but there was no significant staining with non-immune sera (Fig. 3b-2). Oocytes at stage B in the ovary of 2-month post-metamorphosis frogs, as indicated with an arrow, showed intense staining in the whole cytoplasm (Fig. 3c). In an oocyte at stage C, a staining was evident in the thin rim of the cytoplasm and in the reticular network structure (Fig. 3d-1, d-2). Again, there was no significant staining in oocytes in the same ovary when non-immune serum was used (Fig. 3d-3). Furthermore, enriched localization of CLT was found in the cytoplasm free from yolk platelets of oocytes at stage F (Fig. 3e).

Fig. 3

Distribution of CLT in oocytes. Gonad of tadpoles at stage I (a). Oocytes at stage A (b), stage B (c), stage C (d), and stage F (e). Bars represent 50 (a, b, d-2, d-3) and 100 (c, d-1, e) μm, respectively.

Finally, the CLT level in the gonad and ovary was examined by Western blot analysis. As shown in Fig. 4A, it was low in gonads at stage VII where oocytes at stage A were not found and the formation of ovarian cavities was still poor (Fig. 4B-1). However, the CLT level increased significantly in gonads at stage XVI where oocytes at stage A were seen (Fig. 4B-2). It was the highest in the ovary of 2-month post-metamorphosis frogs containing oocytes at stage B (Fig. 4B-3), and then declined to a lower level in adult frogs (Fig. 4A).

Fig. 4

Immunoblot analysis (A) and histological aspects (B) of gonads and ovaries. (A) homogenates of gonads of tadpoles at stages VII and XVI, and ovaries of 2-months post-metamorphosis frogs and 2-year old frogs were run on SDS-PAGE. Then, nitrocellulose filters used for the detection of CLT were prepared as described in MATERIALS AND METHODS. One hundred μg of protein were loaded on each lane. (B) gonads and ovaries were fixed in a paraformaldehyde solution, embedded in paraffin and serially sectioned as described in MATERIALS AND METHODS. Sections were then stained with Mayer's hematoxylin (Merck). Typical histological features of ovaries of tadpoles at stage VII (1) and at stage XVI (2), and frogs 2 months after metamorphosis (3) are shown. Bars represent 50 μm.

DISCUSSION

We reported previously that the CLT gene was expressed in embryos at stage 1 through 25 (Yamamoto et al., 1996). In this study, the CLT gene expression was shown to be at the moderate level in the gonad of metamorphosing tadpoles, and at the extremely high level in the ovary of 2-month post-metamorphosis frogs. The CLT mRNA level, then, decined to a much less extent in the ovary of adult frogs. As for CLT protein, its distribution in oocytes was stage-dependent. Immunostainings were observed in gonads of tadpoles at stage I, but was not restricted to any specific type of cells. We could not identify the type of cells because they were all indifferent. In oocytes at stage A, a weak staining was observed in the peripheral and perinuclear regions. However, intense staining was observed in the whole cytoplsm of oocytes at stage B. As oocytes grow, immunostaining was observed in the thin rim of cytoplasm and perinuclear regions, and in the reticular network structure in oocytes, in particular at stage D. The staining perhaps depends on the localization of the cytoplasm free from yolk platelets.

Western blottings revealed that the CLT level was low in the gonad of tadpoles at stage VII where no oocytes at stage A were found. However, its level increased significantly in the gonad containing oocytes at stage A (see Fig. 4B), and was higher in the ovary containing oocytes at stage B. These results suggest that the CLT synthesis starts in the early stage of growing oocytes. It becomes active in oocytes at stage A, and more active at stage B, and then less active in adult frogs. Since strong immunostaining was observed in the cytoplasm of oocytes at stage B, but not at stage D, the CLT synthesis may be the highest in oocytes at stage B to C. To prove this, it is necessary to purify oocytes in different stages, and to examine which oocytes can synthesize CLT at the highest level. Unfortunately, the method for purification of frog oocytes in different stages is not available at the present time because of small sizes of gonads in tadpoles and 2-month post-metamorphosis frogs. It should also be mentioned that CLT is actively synthesized during the growth of oocytes in frogs, suggesting that CLT may play an important role(s) in growing oocytes. But, its function is unclear presently. As far as we know, however, this is the first report showing the CLT distribution in growing oocytes in animals.

In a wide variety of eggs such as medaka fish (Gilkey et al., 1978), frog (Kubota et al., 1987) and hamster (Miyazaki, 1988), a transient increase in Ca²⁺ concentration occurs at fertilization and propagates as Ca²⁺ waves. Microinjection of IP₃ into eggs activates eggs and induces early developmental events (Busa et al., 1985). However, the IP₃-induced Ca²⁺ release was blocked, when an IP₃ antibody was injected into eggs (Miyazaki et al., 1992), or when overexpression of CLT was induced as described earier. Since CLT is an intraluminal Ca²⁺-binding protein of the endoplasmic reticulum (ER) and can bind a large amount of Ca²⁺, CLT might play an important role in the release of Ca²⁺ from the inside of the ER at fertilization. In addition, Fluck et al. (1991) recently studied the relationship of calcium to contractile band in the cleaving medaka fish eggs. They found two successive Ca²⁺ waves in those eggs; the first wave accompanies furrow elongation in each of the early cleavages of medaka eggs and comes from a high calcium region within the assembling contractile band, while the second one accompanies the slow zipping together of the separating cells. Coppolino et al. (1997) also showed that CLT is an essential modulator for calcium signalling and cell adhesion in embryonic stem cells. Taking together with these findings, CLT may be related to cell division and signal transduction in frog cells. To prove these, further investigations must be required. Nevertheless, it is very interesting to note that CLT is synthesized in growing oocytes and distributed in a stage-dependent manner.