Cyclin-dependent kinase inhibitors in maize endosperm and their potential role in endoreduplication

Abstract

Two maize (Zea mays) cyclin-dependent kinase (CDK) inhibitors, Zeama;KRP;1 and Zeama;KRP;2, were characterized and shown to be expressed in developing endosperm. Similar to the CDK inhibitors in Arabidopsis (Arabidopsis thaliana) and tobacco (Nicotiana tabacum), the maize proteins contain a carboxy-terminal region related to the inhibitory domain of the mammalian Cip/Kip inhibitors. Zeama;KRP;1 is present in the endosperm between 7 and 21 d after pollination, a period that encompasses the onset of endoreduplication, while the Zeama;KRP;2 protein declines during this time. Nevertheless, Zeama;KRP;1 accounts for only part of the CDK inhibitory activity that peaks coincident with the endoreduplication phase of endosperm development. In vitro assays showed that Zeama;KRP;1 and Zeama;KRP;2 are able to inhibit endosperm Cdc2-related CKD activity that associates with p13(Suc1). They were also shown to specifically inhibit cyclin A1;3- and cyclin D5;1-associated CDK activities, but not cyclin B1;3/CDK. Overexpression of Zeama;KRP;1 in maize embryonic calli that ectopically expressed the wheat dwarf virus RepA protein, which counteracts retinoblastoma-related protein function, led to an additional round of DNA replication without nuclear division.

Figure 1.

Comparison of maize CKI protein sequences with those of the mammalian Cip/Kip family of CKIs (p21, p27, and p57), Arabidopsis (KRP1–7), and tobacco (NtKIS1a and NsKIS1a) KRPs. A, Protein sequence alignment of the Zeama;KRP;1 and Zeama;KRP;2 domains that share sequence similarity with the mammalian Cip/Kip inhibitors. B, Alignment of the amino acid sequences deduced from Zeama;KRP;1 and Zeama;KRP;2 with polypeptides deduced from the Arabidopsis and tobacco KRP cDNA sequences. The scale on the right illustrates the percentage of amino acid identity of the regions compared. C, Alignment of the conserved CDK inhibitory motifs among plant KRPs. The accession numbers of the sequences shown are as follows: KRP1(AF079587), KRP2 (AJ251851), KRP3 (AJ301554), KRP4 (AJ301555), KRP5 (AJ301556), KRP6 (AJ301557), KRP7 (AJ301558), NtKIS1a(AJ297904), and NsKIS1A(AJ297905); similarity of the amino acids among the proteins is indicated by the color code on the right.

Figure 2.

RT-PCR detection of Zeama;KRP;1 and Zeama;KRP;2 RNA transcripts during maize endosperm development. Total RNA was extracted from endosperm at 7 to 21 DAP and 50 ng were used as template for RT-PCR. Gene-specific primers were used to amplify Zeama;KRP;1 (A), Zeama;KRP;2 (B), and Actin1 (C), the latter of which was used as a loading control. The CKI cDNAs and genomic DNA were used as positive controls. Each assay was repeated at least three times, and the results illustrated are representative.

Figure 3.

Immunodetection of Zeama;KRP;1 and Zeama;KRP;2 during maize endosperm development. Protein was extracted from 7- to 21-DAP endosperms as described in “Materials and Methods,” and 50 μg were loaded per lane. A, Zeama;KRP;1 antibodies detected a single protein of approximately 21 kD (arrow). B, Zeama;KRP;2 antibodies detected a protein of approximately 26 kD (arrow), which corresponds to the molecular mass calculated for Zeama;KRP;2, and a protein of 20 kD that may represent the Zeama;KRP;1 or an unknown protein. Actin1 (C) and Ponceau S staining (data not shown) were used to standardize sample loading.

Figure 4.

Effect of Zeama;KRP;1 and Zeama;KRP;2 on p13^Suc1-bound CDK activities from maize endosperm. GST-p13^Suc1 agarose beads were used to pull down CDK activity from 9-DAP endosperm extracts. Subsequently, the kinase activity was incubated for 1 h with varying amounts (shown above each lane of the gel) of GST (control), GST-Zeama;KRP;1 (A), or GST-Zeama;KRP;2 (B) fusion proteins. Histone H1 (arrow) was used as substrate for the kinase assay, and the radioactive protein was analyzed by SDS-PAGE and autoradiography. Agarose beads (lane A) were used as a negative control.

Figure 5.

Effect of Zeama;KRP;1 on endosperm cyclin A1;3- and cyclin B1;3-associated CDK activity. Affinity-purified cyclin A1;3 (A) and cyclin B1;3 (B) antibodies were used to immunoprecipitate CDK activity from 9-DAP endosperm extracts. The immunoprecipitated cyclin A1;3- and cyclin B1;3-containing CDKs were incubated for 1 h with the amount of GST (control) or GST-Zeama;KRP;1 fusion proteins shown above the lanes of each gel. Histone H1 (arrow) phosphorylation was used to assay kinase activity, and protein A-bound agarose beads were used as a negative control (lane A). ImageQuaNT software was used to measure radioactive histone H1; the histograms show the average and sd of the percentage of kinase activity inhibited in each assay.

Figure 6.

Effect of Zeama;KRP;1 on cyclin D5;1-associated CDK activity from 9-DAP endosperm and unfertilized ears. Affinity-purified cyclin D5;1 antibodies were used to immunoprecipitate CDK activity from endosperm (A) and unfertilized ear (B) extracts. Immunoprecipitated cyclin D5;1-bound CDK was incubated for 1 h with the amount of GST (control) or GST-Zeama;KRP;1 fusion proteins indicated above the lanes of the gels. Kinase activity was assayed as described in Figure 5. The histogram shows the average and sd of the percentage of kinase activity inhibited in each assay, which was measured by using ImageQuaNT software.

Figure 7.

Effect of Zeama;KRP;2 on cyclin A1;3- and cyclin B1;3-associated CDK activity from 9-DAP endosperm and cyclin D5;1-associated CDK activity from unfertilized ears. Affinity-purified cyclin A1;3 (A), cyclin B1;3 (B), and cyclin D5;1 (C) antibodies were used to immunoprecipitate CDK activity from 9-DAP endosperm and unfertilized ear extracts. The immunoprecipitated cyclin A1;3-, cyclin B1;3-, and cyclin D5;1-containing CDKs were incubated for 1 h with the amount of GST (control) or GST-Zeama;KRP;2 fusion proteins indicated above the lanes of each gel. Kinase activity was assayed as described in Figure 5.

Figure 8.

Characterization of CKI activity in developing maize endosperm. A, Human recombinant cyclin B/cdc2 was used to assay the partially purified (see “Materials and Methods”) endosperm CKI activity in 9- to 19-DAP endosperm using histone H1 (arrow) as substrate. NETT buffer was used as the negative control in the histone H1 kinase assays (lane c). B, Histogram showing quantification of the cyclin B/cdc2 inhibitor activity in 9- to 19-DAP endosperm.

Figure 9.

Association of Zeama;KRP;1 with CDK inhibitory activity in 15-DAP endosperm extract. A, CDK inhibitory activity present in fractions eluted from the G-100 column was assayed with human recombinant cyclin B/cdc2 using histone H1 (arrow) as substrate. B, Histogram showing the percent reduction of kinase activity, which was measured using ImageJ software (http://rsb.info.nih.gov/ij), compared to a noninhibited control (lane c). C, Immunodetection of Zeama;KRP;1 in the fractions shown in A. Human recombinant cyclin B/cdc2 was incubated for 10 min with fractions eluted from the Sephadex G-100 column containing 15-DAP endosperm extract. Kinase activity was assayed as described in Figure 8. NETT buffer was used as control (lane c). For the immunoblot, 25 μL of the protein present in the fractions shown in A were loaded per lane. Zeama;KRP;1 antibodies detected an approximately 21-kD protein (arrow).

Figure 10.

Effect of CDK inhibitory activity from 15-DAP endosperm on cyclin A1;3-, cyclin B1;3-, and cyclin D5;1-containing CDKs. Affinity-purified cyclin A1;3 (A), cyclin B1;3 (B), and cyclin D5;1 (C) antibodies were used to immunoprecipitate CDKs from 9-DAP endosperm (A and B) or unfertilized ear extracts (C). The immunoprecipitated cyclin A1;3-, cyclin B1;3-, and cyclin D5;1-associated CDK activities were incubated with alternate fractions (lanes 1–3) from the Sephadex G-100 column (see “Materials and Methods”), containing CDK inhibitory activity but lacking Zeama;KRP;1. Kinase activity was assayed based on histone H1 phosphorylation (arrow). NETT buffer (lane c) and a fraction that did not contain the CDK inhibitory activity (lane 4) were used as controls. The histograms show the average and sd of the percent reduction of kinase activity, compared to a noninhibited control, and were measured using ImageQuaNT software.

Figure 11.

Effect of Zeama;KRP;1 immunodepletion on the partially purified CDK inhibitory activity. A, CDK inhibitory activity from aliquots of the fractions shown in Figure 10 was assayed with human recombinant cyclin B/cdc2 after incubation with protein A-bound agarose beads; the Zeama;KRP;1 (arrow) in these fractions was subsequently detected by immunoblotting. B, CDK inhibitory activity from aliquots of the fractions shown in A was assayed after immunodepletion of Zeama;KRP;1. Kinase activity was assayed as described in Figure 11A, followed by immunodetection of Zeama;KRP;1. C, Histogram showing the percent reduction of kinase activity, measured by ImageJ software, before (gray bar) and after (white bar) immunoprecipitation of Zeama;KRP;1. Affinity-purified Zeama;KRP;1 antibodies were used to immunodeplete Zeama;KRP;1 protein from fractions as described in “Materials and Methods.” Protein A bound to agarose beads was used as a negative control.

Figure 12.

Effect of Zeama;KRP;1 overexpression on nuclear ploidy of maize embryonic calli expressing the wheat dwarf virus RepA protein. Calli were cobombarded with plasmids expressing the RepA and Zeama;KRP;1 genes under control of the ubiquitin promoter and nuclei were analyzed with a PARTEC flow cytometer. Flow cytometric analysis of nuclei from transgenic calli expressing RepA only (A and B) and RepA plus Zeama;KRP;1 (C–F). The percentage of endoreduplicated nuclei shown in each image was calculated by dividing the total number of nuclei with a ploidy equal to or greater than 8C by total number of nuclei and multiplying by 100. The histogram (G) shows the level of Zeama;KRP;1 RNA (white bar) and RepA RNA (gray bar) in these calli, relative to Actin1 RNA.