Abstract
Keywords
Introduction
The reversible phosphorylation of proteins is an important posttranslational modification in eukaryotes1–3 and an essential mechanism regulating functions of more than 30% total cellular proteins.4,5 The protein phosphatase-2A (PP-2A) is one of the major phosphatases in eukaryotes, contributing to more than 50% serine/threonine phosphatase activity and participating in many cellular processes such as signal transduction, gene expression, neurotransmission, cell cycle control, cell transformation and senescence.1–7 In this regard, our recent studies have indicated that PP2A is associated with carcinogenesis8,9 and is highly regulated in ocular tissues.10–12
The holoenzyme of PP-2A is a heterotrimer, consisting of a scaffold A subunit, a catalytic C subunit and a regulatory B subunit.13–16 While the A and C subunits exist in two isoforms encoded by different genes, the B subunits exist in approximately 26 different isoforms and are encoded by four subfamilies of genes (B or PR55, B’ or PR61, B” or PR72, and B’” or PR93/PR110) and each family consists of multiple genes, with each gene generating multiple splice variants.13,14 These B subunits exhibit differential subcellular localization as well as tissue-specific and developmentally-regulated expression patterns. Variations in their expression pattern and cellular localization of B subunits provide substrate specificity, which is thought to be the molecular basis for the appropriate regulation of numerous cellular processes.1,2,5–7,13,14
The major function of the regulatory subunits for PP-2A is to provide specific PP-2A activity in different cellular compartments and different tissues of organisms.1,2,5–7,13,14 In addition, these different regulatory subunits may have other functions independent of PP-2A. For example, SG2NA, a member of the B’” or PR93/110 family, has been shown to act as a molecular scaffold to promote localization of the estrogen receptor to the plasma membrane and organize the ER-eNOS membrane signaling complex in endothelial cells.
17
In addition, it is found that the
Materials and Methods
Animals
The goldfish samples at the age of 6 months to one year were collected from the Experimental Fish Culture Facility of the Key Laboratory of the Educational Ministry of China in Hunan Normal University. And the fertilization was conducted at the laboratory.
Chemicals
The RNA extraction kit was purchased from Omega, the reverse transcription kit from Invitrogen, Inc, the protein size marker from Fermentas. The 5' and 3' RACE cloning kit was obtained from the Clontech, Inc. The PCR Taq polymerase and the PMD18-T vector were purchased from Takara Inc. The antibodies used for this study were purchased from Santa Cruz Biotechnology and from Sigma, Inc. Gel purification kit and all the oligo primers were provided by Sangon, Inc.
Collection of tissues and embryos
Goldfish were sacrificed through removal of the gill tissue. Various tissues including liver, spermary, ovary, brain, kidney, heart, muscle, gill and fin were quickly dissected out on ice and then frozen under liquid nitrogen for homogenization, first with a mortar and then with 1 ml syringe (18.5 G and 23.5 needles passed). Artificial fertilization was conducted in Hoff's solution (0.1 g CaCl2, 0.05 g KCl, 3.5 g NaCl dissolved in 1000 ml distilled H2O). The fertilized egg membranes were removed with 0.4% pancreatic protease and the de-membraned eggs were allowed to develop at 22 °C in Hoff's solution. Under microscopic examination, the developing embryos at stages of 2-cell, multiple-cell, blastula, gastrula, neurula, optical vesicle, brain differentiation, muscle differentiation, heart beat, eye pigmentation, body pigmentation and hatching larvae were collected and frozen under liquid nitrogen. The frozen embryos were homogenized for extraction of total RNA and proteins as described below.
Molecular cloning of the PR55/B family of PP2A
The two cDNAs for PR55/Bβ/δ were cloned using 5'-RACE and 3'-RACE as previously described.9,20 Briefly, the specific primers used to clone these cDNAs were designed using Jellyfish and prime 5.0 softwares and were shown in Table 1. The homology-based reverse transcriptase polymerase chain reaction (RT-PCR) cloning was used to isolate partial B subunit cDNAs from total adult goldfish brain RNA. Additional 5' sequences for B subunits were obtained by 5' rapid amplification of cDNA ends (5'-RACE) from goldfish brain RNA according to instructions supplied with the Marathon cDNA amplification kit (Clontech, Inc.). 3' Race was performed using 3'-RACE kit.
Oligo primers used for RT-PCR analysis to detect expression of PR55/Bβ/δ.
Reverse transcription-linked polymerase chain reaction (RT-PCR)
The reverse transcription was conducted with a kit from Invitrogen (Invitrogen #18085-019) as previously described.8,9,20–22 Briefly, 2 µg of total RNA were used in a total reaction volume of 20 µl and 2 µl of the reverse transcription reaction mixture were used for PCR reaction. To detect the mRNA expression of PR55/Bβ/δ, three pairs of specific primers as well as the β-actin primers (Table 1) were used. For PCR amplification, both specific primers and β-Actin primers were added into the same reaction at the beginning of PCR, and the PCR reaction was continued 30 cycles. At the end of each reaction, the PCR products were separated by agarose gel (1.5%) electrophoresis and photographed under UV illumination.
Western blot analysis
Western blot analysis was conducted as previously described.8–10,23 Briefly, 50 or 100 µg of total proteins from various tissues and each developmental stage of embryos were separated by 10% SDS-polyacrylamide gel electrophoresis and transferred into supporting nitrocellulose membranes (Bio-Rad). The protein blots were blocked with 5% milk in TBS (10 mM Tris, pH 8.0; 150 mM NaCl) for 60 minutes at normal room temperature. Then, each blot was incubated with the anti-B55β/δ antibodies (Santa Cruz Biotechnology) at a dilution of 1:100 in 5% milk prepared in TBS overnight at 4 °C with mild shaking. After washing 3 times with TBS-T (TBS with 0.05% Tween-20), 15 minutes for each, the blot was incubated with a secondary antibody (anti-rabbit IgG from Santa Cruz Biotechnology) at a dilution of 1 to 1000 for 45 minutes. After washing twice with TBS-T and once with TBS (15 minutes each), the PR55 proteins were detected with an enhanced chemiluminescence detection kit according to the instruction manual from Amersham.
As reference, after stripping the previous antibody, the blot was re-hybridized with the anti-β-actin primary antibody (1:2000 from Sigma, Inc.). After washing with TBST 3 times, the blot was incubated with the anti-mouse IgG (secondary antibody from GE Health Care, Inc. diluted in 1:1000). After washing with TBST twice times and TBS one time, the β-actin level was detected as described above.
Quantitation of RT-PCR and Western blot results
After RT-PCR, the relative density of each specific band verse β-actin control band was quantitated as described before. 24 Both RT-PCR and Western blot results in the x-ray films were analyzed with the Automated Digitizing System from the Silk Scientific Corporation. The relative expression levels (fold) were calculated by dividing the total pixel from each band under investigation by the total pixel from the corresponding β-actin band. The quantitative data averaged from three independent experiments and statistics were analyzed by students’ t-test.
Preparation of antisense expression construct for PR55/Bδ
The full-length cDNA for PR55/Bδ was cloned into the pEGFP vector in a reverse direction so that the anti-sense strand will be expressed under the direction of the viral promoter as previously described. 25 The pEGFP vector alone was used as mock.
Injection of plasmids and observation of the injected embryos development
Both vector and anti-sense expression vector were amplified in DH-5α, and then purified through maximal plasmid purification kit (Qiagen) according to the instruction manual. The 500 ng of purified plasmids in 0.05 µl were injected into each fertilized egg using a microinject developed by Shanghai Instrument, Inc. The vector-injected embryos and the antisense expression construct-injected embryos were allowed to develop at 22 °C in Hoff's solution. The wound embryos were removed from the experiments. The phenotypes of each developmental stage were recorded with microscopy (Table 2).
Statistical results of abnormality of development derived from inhibition of P55/Bδ expression from four independent experiments.
Results
Molecular cloning of the PR55/Bβ/δ cDNAs of PP-2A from goldfish
Using a molecular strategy as previously described,9,20 the full-length cDNAs for PR55/Bβ and PR55/Bβ were isolated. These cDNA sequences were deposited to gene bank database with the access numbers of FJ356012 and FJ356011 for PR55β and PR55δ, respectively.
Sequence analysis revealed that the full-length PR55/Bβ cDNA consists of 1940 bp, with an open reading frame of 1332 nucleotides coding for a deduced protein of 443 amino acids (Fig. 1A). Amino acid sequence alignment showed that goldfish PR55β shares high levels of identity to those from African clawed frog (88.4%), mouse (92.2%) and human (92.5%) (Figs. 1B and 3B). The full-length PR55/Bδ cDNA contains 2163 bp with an open reading frame of 1347 nucleotides, which encodes a deduced protein of 448 amino acids (Fig. 2A). The amino acid sequence alignment demonstrated that the goldfish PR55δ protein shared a sequence identity of 98.4%, 87.7%, 86.9%, 86.9% and 86.9% with those from zebrafish, western clawed frog, chicken, mouse and Norway rat, respectively. (Figs. 2B and 3C).

A) The full length cDNA of PR55/Bβ for PP-2A and the deduced protein sequences. The initiation and termination codons are underlined. The predicted amino acid sequence is shown in the one-letter code below the nucleotide sequences. The WD-40 repeats were high-lighted by open box. B) Alignment of the deduced amino acid sequence from goldfish PR55/Bβ of PP-2A with the known PR55/Bβ sequences of PP-2A from human (Homo sapiens, NM_004576), mouse (Mus musculus, NM_088979) and frog (Xenopus laevis, BC130184).

A) The full length cDNA of PR55/Bδ for PP-2A and the deduced protein sequences. The initiation and termination codons are underlined. The predicted amino acid sequence is shown in the one-letter code below the nucleotide sequences. The WD-40 repeats were high-lighted by open box. B) Alignment of the deduced amino acid sequence from goldfish PR55/Bδ of PP-2A with the known PR55/Bδ sequences of PP-2A from Rat (Rattus norvegicus, NM_144746), Mouse (Mus musculus, NM_026391), Chicken (Gallus gallus, NM_001006507), Frog (Xenopus tropicalis, NM_001006696), and Zebrafish (Danio rerio, NM_199776).

A) Amino acid sequence alignment of the PP2A-PR55/B family members, β/γ/δ in goldfish (The partial amino acid sequence for PR55/Bβ is non-published data from Zhao et al). The completely conserved region among the three isoforms is marked by black shadow. The less conserved region is marked by grey shadow and the non-conserved region is revealed by white background. B) and C) the corresponding phylogenetic trees of the PR55/BβB and PR55/Bδ (C) from four (B) or six (C) vertebrates. The phylogenetic tree for PR55/Bβ (B) was generated through comparative analysis of the coding sequences in human, mouse, frog and the present study using UPGMA calculation and the MEGA3.1 software. The phylogenetic tree for PR55/Bδ (C) was generated using the same strategy and software through comparative analysis of the coding sequences from mouse, rat, chicken, frog, zebrafish and the present study.
Analysis of the amino acid sequence in the deduced protein PR55/Bβ/γ/δ of PP-2A through both ExPASy and the conserved domain architecture retrieval tool (DART) revealed presence of the WD-40 tandem repeats in all three isoforms (Boxed regions in Figures 1A and 2A, and data not shown), indicating the functional importance for their binding to the scaffold subunits of PP-2A.12,15,16,21,22,26 Moreover, an 80.8% of sequence identity between PR55/Bβ and PR55/Bδ were found (Fig. 3A). In addition, the N-termini in PR55/Bβ and PR55/Bδ are significantly diversified (Fig. 3A).
Tissue specific expression of PR55/B family members
To explore the potential functions of PR55/B family members in various tissues of the lower vertebrates, we examined the mRNA levels for PR55/Bβ/δ of PP-2A in liver, spermary, ovary, brain, kidney, heart, muscle, gill and fin from goldfish using reverse transcription-linked polymerase chain reaction (RT-PCR) analysis. As shown in Figure 4A, a band of 370 bp cDNA was amplified using specific primers for PR55/Bβ in two tissues: high level of expression in brain and low level of expression in heart. Similarly, a single band of 372 bp was amplified in all tissues examined for PP2A/Bδ (Fig. 4B). Among these tissues, brain, ovary and kidney contained the highest levels of PR55/Bδ mRNA expression (Fig. 4B). In comparison with brain, ovary and kidney, muscle and heart displayed a slight decrease in PR55/Bδ mRNA expression, and fin, gill, spermary and liver showed further decrease.

Tissue-specific expression of PR55/Bβ/δ mRNAs of PP-2A in adult goldfish. A) RT-PCR to detect the mRNA level of PR55/Bβ of PP-2A in 9 tissues of adult goldfish as indicated. Up panel: both the PR55/Bβ primers and the β-actin primers were added to the reaction at the beginning of PCR. The 370 bp DNA band for PR55/Bβ mRNA was detected in brain and heart. The 276 bp β-actin DNA band was detected in all tissues. Bottom panel: quantitative results of the PR55/Bβ mRNA levels from three independent experiments. The relative level of expression (fold) was calculated by dividing the total pixel from each PR55/Bβ mRNA band with the total pixel from the corresponding β-actin mRNA band. B) RT-PCR to detect the mRNA level of PR55/Bδ of PP-2A in the same 9 tissues of adult goldfish as indicated in A. Like the 276 bp β-actin DNA band, the 372 bp DNA band for the PR55/Bδ mRNA was also expressed in all the tissues examined. Bottom panel: quantitative results of the PR55/Bδ mRNA expression from three independent experiments. The relative level of expression (fold) was calculated as described above in A.
To further explore the tissue-specific distribution of PR55/Bβ/δ, we conducted western blot analysis. As shown in Figure 5A, PR55/Bβ protein was detected at relatively high level in the brain tissue but much attenuated in the heart. All other tissues have no detectable PR55/Bβ protein. In contrast to PR55/Bβ, the PR55/Bδ was highly expressed in the brain and heart, moderately expressed in liver, spermary, ovary, muscle, fin and gill, and the lowest level detected in kidney (Fig. 5B).

Western blot analysis of the PR55/Bβ and PR55/Bδ proteins in 9 tissues of the adult goldfish indicated. A) Up panel: 100 µg of total proteins extracted from the 9 different tissues of the adult goldfish were subjected to Western blot analysis as described in the Materials and Methods. Bottom panel: quantitative results of PR55/Bβ protein in the above 9 tissues of the adult goldfish from three independent experiments. Note that the highest expression levels of the PR55/Bβ protein was detected in the brain, and to a much less degree in the heart. B) Up panel: 100 µg of total proteins extracted from the 9 different tissues of the adult goldfish were subjected Western blot analysis as described in A. Bottom panel: quantitative results of PR55/Bδ protein in the above 9 tissues of the adult goldfish from three independent experiments. Note that the highest expression levels of the PR55/Bδ protein was detected in the brain and heart, and a reduced level of this protein was detected in liver, spermary, ovary, gin and gill. A much reduced PR55/Bδ/δ protein expression was found in kidney.
Developmental expression patterns of PR55/Bβ/δ
To explore the possible function of the PR55/Bβ/δ during goldfish development, we first determined their developmental expression patterns at both mRNA (Fig. 6) and protein (Fig. 7) levels. Through RT-PCR analysis, we demonstrated that PR55/Bβ mRNA level was relatively low from two-cell, multiple-cell to blastula stage. This mRNA level was substantially increased at the gastrula stage transiently, then dropped down at the neurula stage. From the optic vesicle, through brain and muscle differentiation, to heart beat, the PR55/Bβ mRNA became gradually increased. And it maintained relatively stable at this level in the next four different stages of development (Fig. 6A). Different from the expression pattern of PR55/Bβ mRNA (Fig. 6A), the PR55/Bδ mRNA, in the very first three stages of development, displayed the highest level, then slightly dropped down from gastrula to neurula stages, gradually increased from optical vesicle stage to brain differentiation stage, maintained at this level at muscle differentiation and heart beat stages, and became gradually decreased from eye pigmentation, to hatching larval stages (Fig. 6B).

Temporal mRNA expression patterns of PR55/Bβ/δ for PP-2A during embryonic development of goldfish. A) Up panel: RT-PCR to detect the mRNA level of PR55/Bβ of PP-2A during 12 different developmental stages. Bottom panel: quantitative results of the PR55/Bβ mRNA expression from three independent experiments. B) Up panel: RT-PCR to detect the mRNA level of PR55/Bδ of PP-2A during the 12 stages. Bottom panel: quantitative results of the PR55/Bδ mRNA expression from three independent experiments. The method is the same as described in Figure 4

Western blot analysis of the protein for PR55/Bβ/δ in 8 developmental stages as indicated. A) Up panel: 100 micrograms of total proteins extracted from 8 developmental stages of the developing goldfish embryos were subjected to Western blot analysis. Note that no PR55/Bβ protein was detectable at any stage. B) Up panel: 100 micrograms of total proteins extracted from the 8 different stages were subjected to Western blot analysis as described in Figure 5 Bottom panel: quantitative results of PR55/Bδ protein in the 8 developmental stages as determined using the methods described in Figure 5
To further confirm the developmental expression of PR55/Bβ/δ at the protein level, we have conducted western blot analysis. As shown in Figure 7A, PR55/Bβ protein seems to be undetectable at any stage of development. In contrast, the PR55/Bδ protein was maintained at similar levels in the 8 different stages examined: multiple-cell, blastula, gastrula, neurula, optic vesicle, brain differentiation, eye pigmentation, body pigmentation and hatching (Fig. 7B).
Attenuation of PR55/Bδ protein expression led to severe abnormality in eye development of goldfish
To further confirm the role of PR55/B family subunit in regulating development of goldfish, we constructed an expression construct for the generation of the antisense strand RNA from the PR55/Bδ cDNA. Basically, the full length cDNA of PR55/Bδ was ligated into the pEGFP-C3 vector in the non-coding direction so that the anti-sense RNA would be generated when injected into fertilized eggs. The empty vector was used as mock injection. Expression of the antisense PR55/Bδ RNA substantially attenuated the translated level of PR55/Bδ protein at several developmental stages examined (Figs. 8A and 8B). When PR55/Bδ protein level was significantly downregulated, the development of the goldfish displayed severe phenotype in both trunk and eye (Fig. 8D) in comparison with those in the normal larvae (Fig. 8C). The trunk was severely bent and the eye appeared in much small size (microphthalmia) (Fig. 8D Table 2) in comparison with the vector-injected embryos (Fig. 8C). Thus, our results demonstrate that PR55/Bδ is important for goldfish organogenesis, especially the trunk and the eye.

Inhibition of PR55/Bδ protein expression through anti-sense blockage of translation led to severe phenotype in goldfish development. A) The relative expression levels of P55/Bδ in vector-injected (A-a) and anti-sense P55/Bδ expression con struct-injected embryos (A-b). B) Quantitation of comparative expression in vector-injected and antisense P55/Bδ expression construct-injected embryos. Note that antisense blockage of PR55/Bδ translation led to a substantial decrease in the expression level of the PR55/Bδ protein at 4 different stages. The down-regulation of the PR55/Bδ protein caused severe abnormality of organogenesis in the eye and trunk (D) in comparison with the vector-injected embryos (C).
Discussion
In the present study, we have demonstrated: 1) The goldfish PR55/Bβ/δ cDNAs contain ORFs of 1332 bp and 1347 bp, coding for the deduced proteins of 443 PR55/Bβ and 448 PR55/Bδ amino acids, respectively; 2) the deduced goldfish PR55/Bβ protein share an amino acid identity of 88.4%, 92.5% and 92.5% to that from frog, mouse and human, respectively; the deduced goldfish PR55/Bδ protein share an amino acid identity of 98.4%, 87.7%, 86.9%, 86.9% and 86.9% to that from zebrafish, frog, chicken, mouse and rat, respectively; 3) the PR55/Bβ mRNA is present in brain and heart only, and the PR55/Bδ mRNA is detected in all 8 tissues examined; 4) At the protein level, the PR55/Bβ and PR55/Bδ also displayed distinct patterns; 5) Inhibition of PR55/Bδ translation through anti-sense RNA blockage caused severe phenotype of the injected embryos including severe reduction of the eye size. Together, our results reveal that PR55/B family members have important functions controlling animal development and maintaining adult tissue homeostasis of goldfish.
Contrast expression patterns of PR55/Bβ/δ are present in lower and higher vertebrate
The protein phosphatase-2A (PP-2A) is one of the major phosphatases in eukaryotes, and the holoenzyme of PP-2A is a heterotrimer, which contains a scaffold A subunit, a catalytic C subunit and a regulatory B subunit.1–7 Both A and C subunits exist in two isoforms which are encoded by different genes. In contrast, the B subunits exist in 26 or more isoforms and so far, four subfamilies of genes, PR55/B, PR61/B’, PR72/B”, and PR93/PR110/B’” have been identified to code for these different isoforms.1,2,6,15,16 In the present studies, we have isolated two members of the PR55/B family from goldfish. Although each member of the goldfish PR55/B family shares high levels of amino acid identity (from 70% to 98%) with the counterpart from other vertebrates (Fig. 3), their expression patterns may be substantially different in different vertebrates. In the present study, we demonstrate that the goldfish PR55/Bβ mRNA is mainly expressed in the brain and to a much less degree in the heart. However, in mouse and rat, it is mainly expressed in testis and to a less degree in murine brain.27,28 For the expression pattern of PR55/Bδ, there is significant difference between goldfish and murine. While in goldfish, PR55/Bδ mRNA is expressed in all 8 tissues examined with the highest level found in brain, ovary and kidney, and the lowest level in liver and testis (spermary), the high level of PR55/Bδ mRNA is only detected in mouse testis, the remaining tissues either have very little PR55/Bβ mRNA (kidney, muscle, liver and brain) or no PR55/Bδ mRNA (lung, spleen and heart).27–29 Thus, goldfish (lower vertebrate) and murine (higher vertebrate) display distinct difference in the tissue-specific expression patterns of PR55/Bβ/δ.
In the present study, we found that the PR55/Bβ mRNA was present at low levels at the first 3 stages and then became clearly upregulated at gastrulation stage. After a brief downregulation in neurula and optic vesicle stages, the PR55/Bβ mRNA were gradually increased from optic vesicle to heart beat stages and then maintained at this level with some slight fluctuations in the next three stages. Similar to the temporal mRNA expression pattern in goldfish, the PR55/Bβ mRNA was also detected in mouse embryo, as early as embryonic day 11 (ED11). This mRNA level became gradually increased from ED 11 to ED 17. 28 In contrast to the goldfish PR55/Bβ mRNA expression pattern, we hardly detected any PR55/Bβ protein expression at the 12 different developmental stages examined. Such results suggest that the PR55/Bβ mRNA may be non-translatable and the specific PP-2A activity with PR55/Bβ as regulatory subunit may be not necessary for goldfish development. Whether the PR55/Bβ mRNA in mouse embryo yields any detectable protein remains to be explored. On the other hand, we could not exclude the possibility that a low level of PR55/Bβ protein exists that cannot be detected with the antibody we used and in the presence of a large portion of yolk protein in goldfish embryo. Different from the PR55/Bβ the PR55/Bδ is highly expressed at both mRNA and protein levels from early to later developmental stages of goldfish. This temporal pattern is also different from that in mouse where no PR55/Bδ transcripts could be detected until ED17. 28 Such distinct difference in their temporal expression patterns between lower and higher vertebrates suggest that PR55/Bδ plays an important role in regulating development of goldfish embryo but not mouse embryo before EDI7.
The discrepancy of the mRNA and protein levels for PR55/Bδ in goldfish kidney (Figs. 5 and 6) suggests that the other isoform of PP2A PP55/B family regulatory subunit, PR55/Bα, may play an important role in regulating specific PP-2A activity in this tissue.
P55/Bδ of PP-2A has important functions in regulating development
Besides its regulatory function in the heterotrimeric holoenzyme of PP-2A, the regulatory subunits of PP-2A may have other functions. In
Disclosure
This manuscript has been read and approved by all authors. This paper is unique and is not under consideration by any other publication and has not been published elsewhere. The authors and peer reviewers of this paper report no conflicts of interest. The authors confirm that they have permission to reproduce any copyrighted material.
