Protein Phosphorylation during Coconut Zygotic Embryo Development^{<a href="#FN1" rid="FN1" class=" fn">1</a>}

Plant Material

Seeds were collected from coconut (Cocos nucifera L.) groves in the Yucatán Península, México, and transported to the laboratory. The embryo developmental stages were arbitrarily defined, taking the first pollinated inflorescence as stage 0; stages 1 to 16 correspond to inflorescences of increasing maturity. Embryos were excised from the seeds in the laboratory and immediately processed.

Tissue Homogenization

Embryos were excised from seeds and immediately homogenized in buffer containing: 50 mm Tris-HCl, pH 7.4, 10 mm sodium pyrophosphate, 50 mm NaCl, 250 mm Suc, 10% glycerol, 1 mm EGTA, 0.2 mm orthovanadate, 1 mm PMSF, 1 mm β-mercaptoethanol, and 1 μg/mL leupeptin and aprotinin (extraction buffer). Homogenates were centrifuged at 14,000g for 30 min. Supernatants were further centrifuged at 100,000g for 45 min at 4°C. The supernatants (5–7 mg protein/mL), referred as the soluble fraction, were snap-frozen and stored in liquid nitrogen prior to analysis.

Protein Quantification

Protein concentration of samples was measured according to the method of Smith et al. (1985), using the bicinchorinic acid protein assay reagent and BSA as the standard (Pierce).

In Vitro Phosphorylation Assay

Aliquots of soluble fractions (100 μg of protein) were incubated at 30°C with 20 mm Hepes, pH 7.0, 10 mm MgCl₂, 0.1 mm orthovanadate, and 50 μm ATP plus 2 μCi/nmol [γ-^{P]ATP (NEN-Dupont). After 15 min, the reaction was stopped with Laemmli buffer (}Laemmli, 1970) and heated for 5 min at 85°C. The samples were analyzed by 10% SDS/PAGE (40 V, overnight) and electroblotted onto Nitroplus transfer membrane (Micron Separations, Westborough, MA) for 5 h at 50 V. Phosphoprotein bands were visualized by autoradiography.

Immunoblot Analysis

Proteins in the soluble fractions were phosphorylated (or not) in vitro as described above, except that [γ-^{P]ATP was not added. After 15 min, the reaction was stopped with Laemmli buffer and heated for 5 min at 85°C. The samples were analyzed by 10% SDS/PAGE (40 V, overnight) and either stained with Coomassie blue or electroblotted onto Nitroplus transfer membrane for 5 h at 50 V. Filters were blocked in 5% defatted milk in TBS-T buffer (50 m}m Tris-HCl, pH 7.5, 150 mm NaCl, and 0.1% Tween 20) for 60 min. Membranes were washed in TBS-T buffer twice for 15 min and then probed in a 1:1000 dilution of recombinant anti-PY coupled to horseradish peroxidase RC20H (Transduction Laboratories, Lexington, KY) for 1 h at room temperature. The membranes were washed twice for 15 min in 100 mL (each time) of TBS-T buffer. Immunodetection was revelated in a dark room by exposing membranes for 1 min in enhanced chemiluminescence reagent (Amersham).

Protein Tyr Kinase Assay

This assay was done as described for in vitro phosphorylation in the presence of 10 μg of RR-SRC (GIBCO-BRL), and the reaction was stopped by precipitation with a cold solution of 10% TCA. The mixture was centrifuged at 14,000g for 10 min at 4°C. The supernatant was loaded onto phosphocellulose filters (Life Technologies). Filters were washed twice with 1% acetic acid and water and dried, and the radioactivity was determined by Cerenkov counting. Two controls were used in this assay: one with the extract alone in the absence of RR-SRC and another in the presence of RR-SRC but in the absence of protein extract. In both cases the activity was less than 1 pmol and was subtracted.

Alkaline Treatment

Proteins in the soluble fraction (100 μg) were phosphorylated in vitro in the presence of 10 μCi/nmol of [γ-^{P]ATP (NEN-Dupont) for 15 min, and the reaction was stopped with Laemmli buffer and heated for 5 min at 60°C. Samples were analyzed by 10% SDS/PAGE (40 V, overnight). Proteins were blotted onto a Hybond-PVDF membrane (Amersham) at 50 V for 5 h. Membranes were divided into two identical fractions that contained the same samples. One membrane was incubated in 1} m KOH at 56°C for 90 min, and the control membrane was maintained in water in the same conditions as the KOH-treated membrane. Membranes were independently washed twice (5 min each time) in 100 mL (each time) of TN buffer (10 mm Tris-HCl, pH 7.4, and 150 mm NaCl) followed by two washes in 1 m Tris-HCl, pH 7.0, and distilled water. Membranes were dried and exposed to autoradiography for 48 h at −80°C.

Protein Hydrolysis

[^{P]-labeled proteins were excised from PVDF membranes as strips (0.2 cm wide). The strips were rewet in methanol for 1 min and then rewet in 0.5 mL of water and placed into a glass ampule containing 300 μL of 6} n HCl (Sigma). The ampules were sealed and incubated at 110°C for 2 h in an oven (Duo-Vac, Lab-Line, Melrose Park, IL). At the end of the hydrolysis, the ampules were cooled at room temperature and centrifuged at 14,000g for 1 min. Then seals were broken and the hydrolysates were transferred to 13- × 100-mm glass tubes (Pyrex) and dried for 30 min in a Speedvac (Savant, Fullerton, CA). The dried hydrolysates were resuspended in 50 μL of distilled and deionized water and dried in the Speedvac as above. This step was repeated three times, and the hydrolysates were resuspended in 10 μL of distilled and deionized water.

Phosphoamino Acid Analyses

Protein of hydrolysate aliquots (7 μL) was loaded onto 20- × 20-cm cellulose TLC plates (Aldrich precoated TLC plates: cellulose on glass with 254-nm fluorecent indicator, 250-μm layer thickness, and fiber length of 2–20 μm). In each sample, 10 μg of internal phosphoamino acid standards (phosphoserine, phosphothreonine, and phosphotyrosine; Sigma) were included, and the phosphoamino acid standards were also loaded onto parallel lanes singly and as a mixture. The plates were developed first with n-propanol:0.5 n HCl (2:1, v/v) followed by n-butanol:acetic acid:water (100:22:50, v/v) in the same direction. The phosphoamino acid standards were located by staining with ninhydrin spray reagent (Sigma) and pencil marked. The plates were dried before autoradiography at −80°C with an intensifier.

Extracts of Cell Line A431

Extracts of the human epidermoid carcinoma cell line A431 were used as a postive control, since it is well documented (Sorkin et al., 1991; Rodríguez-Zapata and Hernández-Sotomayor, 1998) that this line contains kinase activity for Ser, Thr, and Tyr residues and substrates for these kinases. One of these substrates, the EGF receptor, has Ser, Thr, and Tyr residues that can be phosphorylated, and it also has Tyr kinase activity. A431 cells were treated with EGF and extracted with Triton X-100 as described previously (Hernández-Sotomayor et al., 1991). A431 cells were kindly provided by Dr. Graham Carpenter (Vanderbilt University, Nashville, TN).

Embryo Development

Changes in size and form were observed during embryo development. Form changed from round (stages 10–12) to cylindrical (stages 14–16). Length, base width, and fresh weight increased as the embryo developed, reaching maximum values, 6.6 mm, 3.7 mm, and 90 mg respectively, at stages 13 to 14 (Fig. ​(Fig.1). 1).

Figure 1

A, Developing zygotic embryos excised from coconuts at different developmental stages. Grid square side = 1 mm. B through D, Evaluation of embryo growth: B, length, C, base width, and D, fresh weight. Bars denote sd; n = 50.

Alkaline Treatment

[γ-^{P]ATP-phosphorylated proteins from the soluble fraction of developmental stages 11 and 15 were transferred to Hybond-PVDF membranes, treated with KOH, and autoradiographed (48 h at −80°C). Both the control membrane that was not treated with KOH and cell line A431 showed a complex pattern of bands from the developing embryo samples, some showing high-intensity signals (Fig.} ​(Fig.2A).2A). After alkaline treatment a high content of ^{P was removed, as shown by the decrease in signal intensity of bands (Fig.} ​(Fig.2B).2B). Several of the bands disappeared, but some remained, although with a weaker signal, such as those at 175, 156, 125, 62, 44, 37, and 33 kD (stage 11), and 37 kD (stage 15). In contrast, some bands remained with a strong signal, such as those at 41 kD (stages 11 and 15). In line A431 the signal of several bands remained after alkaline treatment (Fig. ​(Fig.2B).2B). The signal of the EGF receptor (170 kD) was not detected before or after treatment.

Figure 2

Effect of alkaline treatment on in vitro [γ-^{P]ATP-labeled proteins of soluble fractions from coconut zygotic embryos of developmental stages 11 and 15. [P]-Labeled proteins were separated by 10% SDS-PAGE and electroblotted onto nylon membranes. Membranes were left without treatment (A) or treated with 1} m KOH (B), and bands were visualized by autoradiography. Figures with arrowheads on the left show the molecular mass markers. The arrowheads on the right show the protein bands that remained phosphorylated after the KOH treatment. A431 denotes a carcinoma cell line that was included as a positive control. The amount of protein per lane was 100 μg. The results shown are representative of four separate experiments.

Phosphoamino Acid Analysis of [^{P]-Labeled Protein Hydrolysates}

TLC of ^{P-labeled protein from coconut hydrolysates of the full complement proteins (developmental stages 11 and 15) showed two signals that comigrated with authentic phosphothreonine and phosphoserine, but phosphotyrosine was not detectable (Fig.} ​(Fig.3A).3A). In cell line A431 the three phosphoamino acids were detectable. However, phosphoamino acid analysis of blotted proteins that remained [^{P] labeled after treatment with KOH not only revealed the occurrence of phosphothreonine and phosphoserine, but also of phosphotyrosine, which was the most abundant phosphoamino acid (Fig.} ​(Fig.3B).3B). In hydrolysates of proteins from A431 cells the three phosphoamino acids were detected, as well as three additional bands that were not identified.

Figure 3

TLC analysis of phosphoamino acids of in vitro [γ-^{P]ATP-labeled proteins of the soluble fraction from embryos at developmental stages 11 and 15. Proteins were hydrolyzed and 7-μL aliquots of the hydrolysates were loaded onto phosphocellulose plates and developed. A, Results of analysis of all of the [P]-labeled proteins. B, Results of the 41-kD protein only. Plates were autoradiographed after development, and the phosphoamino acids were identified by comparison with authentic, unlabeled phosphoamino acid standards. The A431 cell line hydrolysate was included as a positive control. The results are representative of three separate experiments.}

Tyr Kinase Assay

Soluble fractions from embryos of stages 10 to 16 were able to phosphorylate RR-SRC (Pike et al., 1982). This peptide is specific for Tyr kinases, because in its sequence it has a single phosphorylation site, a Tyr residue (RRLIEEDAE[Y]AARG), and it is extensively used to characterize Tyr kinase activity (Pike et al., 1982). The results showed two peaks of activity, one during early development (stages 11–12) and another during late development (stages 15–16) (Fig. ​(Fig.4A).4A). Extracts of A431 cells were able to phosphorylate RR-SRC as well. The ability of embryo-soluble fractions (stages 11 and 15) to phosphorylate RR-SRC was reduced with the addition of genistein (extensively used in animal cells as a specific Tyr kinase inhibitor: O'Dell et al., 1991; Fry and Bridges, 1995) to about 50% (stages 11 and 15) when used at 200 μm, and to 25% (stage 11) and 5% (stage 15) when used at 400 μm (Fig. ​(Fig.4B).4B). Similarly, the addition of genistein to extracts of A431 cells also reduced their ability to phosphorylate RR-SRC (Fig. ​(Fig.4B). 4B).

Figure 4

Tyr kinase activity in the soluble fractions of coconut zygotic embryos at developmental stages 10 to 16 (A) and the effect of (200 and 400 μm) genistein on this activity in soluble fractions of embryos at stages 11 and 15 (B). Tyr kinase activity was assayed in vitro with the synthetic substrate RR-RSC. The specific activity was calculated as picomoles of ^{P incorporated into RR-SRC min mg protein. The A431 carcinoma cell line was included as a positive control. The data shown are the averages of from three separate experiments. Bars denote} sd.

Immunoblot Analysis

Soluble fractions of developing embryos (stages 10–16) were incubated (or not for controls) for in vitro phosphorylation prior to SDS-PAGE and protein blotting. The occurrence of basal phosphorylation in Tyr residues, which was detected as anti-PY, revealed the presence of several bands in the untreated controls (Fig. ​(Fig.5A).5A). The intensity of some of the bands changed during development. For bands at 51, 53, and 41 kD, it was greater during early development (stages 10–12) and decreased during late development (stages 13–16). In contrast, the bands at 37, 35, and 24 kD showed low intensity during the early stages, which increased notably later (at stage 14). In samples subjected to in vitro phosphorylation, the intensity of those bands observed in the controls increased and additional bands were detectable (Fig. ​(Fig.5B).5B). Both basal and enhanced Tyr phosphorylation showed a differential pattern, depending on the developmental stage. The intensity of the bands at 175, 151, 125, 108, 75, 67, 62, 51, 44, and 41 kD was greater during early development (stages 10–13) than in later stages of development. In contrast, two additional major phosphotyrosine-containing bands of approximately 31 and 24 kD were observed only during late development (stages 14–16). When 1 mm phosphotyrosine was added to the antibody preparation, with the exception of two bands at 47.5 and 37 kD, the band patterns were not revealed (Fig. ​(Fig.5C).5C). These bands did not appear when the concentration of phosphotyrosine was increased to 2 mm (not shown). When extracts from tissue surrounding the embryo and tissues of vegetative parts (root, stem, and leaf) were assayed, no signal was observed in the immunoblots (not shown).

Figure 5

Patterns of Tyr-phosphorylated proteins from extracts of coconut zygotic embryos at developmental stages 10 to 16. Proteins (100 μg) were separated by 10% SDS-PAGE, electroblotted onto nitrocellulose membranes, and immunodetected using the anti-PY RC20 H. Proteins were assayed directly (A) or subjected before electrophoresis to in vitro phosphorylation without (B) or with (A) 1 mm phosphotyrosine. Figures and arrowheads on the left denote molecular mass markers. Arrowheads on the right denote bands that change their intensity with development (A and B) or to those that remained after treatment with phosphotyrosine (C). The results shown are representative of five separate experiments.

Immunodetection of the extract of A431 cells revealed bands at 107, 86, and 37 kD, and a major one at 170 kD corresponds to the EGF receptor (Fig. ​(Fig.5,5, A and B). Surprisingly, the EFG receptor Tyr phosphorylation was detected by immunoblot and not after alkaline treatment (Fig. ​(Fig.2).2). A possible explanation for this discrepancy is the fact that to have signal after the alkali treatment, the proteins had to incorporate ^P-PO₄ during the in vitro phosphorylation. If those proteins, like the EGF-receptor band, were already Tyr phosphorylated in vivo, the incorporation of ^{P expected in vitro would not be the same as for a protein that was not totally Tyr phosphorylated in vivo. It is important to remember that in this case, A431 cells were incubated in vivo with EGF to activate Tyr phosphorylation. With the addition of phosphotyrosine the pattern was not revealed (Fig.} ​(Fig.55C).