Maize endosperm-specific transcription factors O2 and PBF network the regulation of protein and starch synthesis

Significance

Nutritional quality and yield are equally important considerations in crop breeding, although they sometimes appear at odds. In this work we made the discovery that these traits are linked through regulation by two transcription factors. Mutations that affect the expression of these transcription factors can improve the nutritional quality of the seed but also can reduce kernel yield and hardness. Therefore future corn-breeding programs should silence zein genes directly, not by blocking transcription factors.

Abstract

The maize endosperm-specific transcription factors opaque2 (O2) and prolamine-box binding factor (PBF) regulate storage protein zein genes. We show that they also control starch synthesis. The starch content in the PbfRNAi and o2 mutants was reduced by ∼5% and 11%, respectively, compared with normal genotypes. In the double-mutant PbfRNAi;o2, starch was decreased by 25%. Transcriptome analysis reveals that >1,000 genes were affected in each of the two mutants and in the double mutant; these genes were mainly enriched in sugar and protein metabolism. Pyruvate orthophosphate dikinase 1 and 2 (PPDKs) and starch synthase III (SSIII) are critical components in the starch biosynthetic enzyme complex. The expression of PPDK1, PPDK2, and SSIII and their protein levels are further reduced in the double mutants as compared with the single mutants. When the promoters of these genes were analyzed, we found a prolamine box and an O2 box that can be additively transactivated by PBF and O2. Starch synthase IIa (SSIIa, encoding another starch synthase for amylopectin) and starch branching enzyme 1 (SBEI, encoding one of the two main starch branching enzymes) are not directly regulated by PBF and O2, but their protein levels are significantly decreased in the o2 mutant and are further decreased in the double mutant, indicating that o2 and PbfRNAi may affect the levels of some other transcription factor(s) or mRNA regulatory factor(s) that in turn would affect the transcript and protein levels of SSIIa and SBEI. These findings show that three important traits—nutritional quality, calories, and yield—are linked through the same transcription factors.

Maize (Zea mays) is one of the most important food sources on earth. Its endosperm is composed of ∼70% starch and 10% protein. Although starch is the main source of calories consumed, protein provides the critical nutrients for our food supply. However, maize’s nutritional quality is poor because its main storage protein, zein, is devoid of the essential amino acids lysine and tryptophan, and corn-based animal feed must be supplemented with soy protein and synthetic methionine. The classic mutant opaque2 (o2) was found to improve the seed nutritional value by reducing the synthesis of zein proteins (1). O2 encodes an endosperm-specific bZIP transcription factor and mainly regulates the expression of α- and β-zein genes by recognizing the O2 box in their promoters (2–4). For practical applications, however, the o2 mutant has multiple agronomic defects, i.e., soft texture, susceptibility to disease, and yield drop. Prolamine-box binding factor (PBF), another endosperm-specific transcription factor, belongs to the DOF family and regulates the expression of zein genes by recognizing the prolamine (P) box in their promoters (5, 6).

Starch, the main contributor of kernel weight, is synthesized and assembled into semicrystalline starch granules by a suite of well-characterized enzymes in the starchy endosperm cells, including sucrose synthase (SUS), ADP-glucose pyrophosphorylase (AGP), soluble starch synthase (SS), granule-bound starch synthase (GBSS), starch branching enzyme (SBE), and starch debranching enzyme (DBE) (7, 8). Waxy (GBSSI) is involved in the synthesis of amylose, whereas SSs are mainly involved in the synthesis of amylopectin, which is required for starch granule formation. Three starch synthases, i.e., SSI, SSIIa (Sugary2), and SSIII (Dull1) are preferentially expressed at the filling stage of the endosperm and are thought to be primarily responsible for amylopectin synthesis in the amyloplasts (7–9). The functions of SSIIa and SSIII have been genetically proven by mutant analysis (10, 11), but null mutants for SSI have not yet been identified. Interestingly, biochemical studies demonstrate that starch biosynthetic enzymes and proteins from multiple metabolic pathways associate with each other to form high-molecular-weight complexes in wheat and maize endosperm amyloplasts (12–14). Almost all SSIII and SSIIa exist in the complex form and are considered regulators of starch biosynthesis as well as of enzymatic functions (14). Pyruvate orthophosphate dikinase (PPDK) is a key enzyme for CO₂ fixation, which catalyzes pyruvate (PYR) to phosphoenolpyruvate (PEP) conversion in C4 photosynthesis (15). This protein is also abundant in the nonphotosynthetic tissue of endosperm in C3 and C4 cereal grasses (16, 17). Although the exact biological function of endosperm PPDK is still unclear, a small percentage of PPDK that exists in amyloplasts can associate stably with starch biosynthetic enzymes (14), suggesting that endosperm PPDK might be involved in starch or other reserve synthesis. In addition to the starch biosynthetic pathway, the oxidative pentose phosphate pathway (oxPPP) is also thought to play an important role in endosperm starch synthesis (18). A recent report showed that loss of 6-phosphogluconate dehydrogenase (PGD3) in oxPPP leads to severely reduced grain-fill phenotypes with reduced starch accumulation in maize (19).

Mutations of these starch biosynthetic genes generally cause a reduction in starch content and, in turn, the kernel yield (20), but rare reports show that the transgenic manipulation of starch biosynthetic genes is able to increase them. A modified maize, Shrunken-2 (Sh2), encoding the large subunit of AGP (HS33/Rev6 Sh2) was transformed to enhance yield by increasing the seed number rather than the starch mass (21). These facts suggest that increasing endosperm starch content by transgenic manipulation of starch biosynthetic genes is difficult to achieve because of the complexity of starch synthesis in endosperm. On the other hand, thus far there have been only a few reports regarding transcriptional regulation of starch synthesis in cereals. Among them, one barley WRKY transcription factor, SUSIBA2, and three maize transcription factors, ZmNAC36, ZmbZIP91, and ZmEREB156, are thought to regulate the expression of starch-synthetic genes based on transcriptional activation and/or EMSA assays of their promoters (22–25). Recently, a study in rice showed that OsbZIP58, the closest homologous protein of maize O2, regulates the expression of multiple genes in the starch biosynthetic pathways, and its null mutants cause abnormal seed morphology with decreased amounts of seed starch (26). O2 and PBF control kernel nutritional quality by regulating the synthesis of storage zein proteins. Consistent with the temporal pattern of zein gene expression, the expression of the major starch-synthetic genes and enzymatic activities increases sharply and later decreases gradually between 10 and 35 days after pollination (DAP) (8, 9, 27), suggesting that the transcriptional regulation of storage protein and starch synthesis in endosperm may be coordinated temporally by some common factors. Here, we show that O2 and PBF also regulate the endosperm starch synthesis and, in turn, kernel weight.

Results

Reduction of Seed Weight and Starch Content in PbfRNAi and o2 Mutants.

The PbfRNAi and o2 mutants and their double mutant PbfRNAi;o2 mutant are all in the W64A background (28). They all exhibit an opaque and soft endosperm phenotype. In addition, PbfRNAi;o2 also produces mildly shrunken kernels (Fig. 1A). Measurement of 1,000-kernel weight (KW) and test weight (TW) showed that PbfRNAi alone had less effect than o2. In the PbfRNAi mutant the KW was not overtly altered, but the TW was reduced by 9%, whereas in the o2 mutant the KW and TW were reduced by 20% and 13%, respectively. Moreover, in the PbfRNAi;o2 double mutant the two yield parameters were reduced by 43% and 23%, respectively (Fig. 1B). These results demonstrate that PBF and O2 affect kernel-weight traits additively.

Fig. 1.

Kernel phenotypes and carbohydrate determination in NG (W64A), PbfRNAi, o2, and PbfRNAi;o2 seeds. (A) Ear (Upper) and kernel (Lower) phenotypes of W64A, PbfRNAi, W64Ao2, and PbfRNAi;o2 seeds. e, embryo; en, endosperm. (Scale bars, 1 mm.) (B) The KW and TW of W64A, PbfRNAi, W64Ao2, and PbfRNAi;o2 seeds. Error bars represent the SD of three independent replicates. (C) SEM images of starch granules in the mature kernels. (Top) Kernel transverse sections. (Middle) SEM images of the kernel peripheral regions indicated by the white-outlined boxes in the top panel; (Lower) SEM images of the kernel central regions indicated by the black-outlined boxes in the top panel. (Scale bars, 10 μm.) (D) Carbohydrate contents in the mature kernels. Error bars represent the SD of five independent replicates. (E) Dosage of O2 in W64A, W64Ao2, and their reciprocal crosses. (F) Dosage effect of O2 on KW and starch content. Data represent the mean ± SD of three independent replicates. In B, D, and F, asterisks indicate a significant difference from W64A (Student’s t test, P < 0.05).

To investigate the reduction in kernel weight further, we used SEM to examine starch granules in PbfRNAi, W64Ao2, and PbfRNAi;o2 mutants, which became gradually downsized in the peripheral and central regions of endosperm as compared with normal genotypes (NG) (Fig. 1C), similar to the effects seen in some starch biosynthetic mutants (e.g., dull1) (20). We then quantitatively measured the starch and soluble sugars content in mature dry mutant seeds (Fig. 1D). Compared with NG, which contained around 65 mg starch per 100 mg dry seed flour, the starch content was reduced by ∼5% and 11% in PbfRNAi and W64Ao2, respectively. In PbfRNAi;o2 the starch content dropped more severely, by 25%, significantly lower than in the single mutants. In contrast to the reduction in starch, the levels of their soluble sugars increased in these mutants by 58%, 55%, and 86%, respectively, compared with NG, which contained 2.8 mg soluble sugars per 100 mg dry seed flour. This pattern of inverse accumulation of starch and soluble sugars in the three mutants was reminiscent of the patterns seen in starch biosynthetic mutants (20), implying that PbfRNAi and o2 probably decrease kernel weight by affecting starch synthesis.

Moreover, O2 appears to have a dosage effect on the starch content and kernel weight. Taking advantage of triploid endosperm, we were able to create seeds with two dosages or one dosage of O2 by reciprocally crossing W64A and W64Ao2 (Fig. 1E). Apparently, the starch contents of kernels with two dosages, one dosage, and zero dosage of O2 were decreased gradually by 2%, 4%, and 12%, respectively, compared with NG with three dosages (Fig. 1F). Consequently, the KWs were accordingly decreased by 4%, 13%, and 28%, respectively (Fig. 1F). This result differs from the regulation of storage protein transcription, in which the opaque phenotype is not dosage dependent but is recessive. (The seeds used in the experiment shown in Fig. 1F were propagated in 2014, and those used in the experiment shown in Fig. 1 A–D were grown in 2015.) This observation was consistent with a previous report that starch synthetic enzymes also have a dosage effect on the starch content in maize endosperm (29).

Down-Regulation of PPDKs, SSIII, SSIIa, and SBEI in PbfRNAi and o2 Mutants.

To investigate the downstream gene network that PBF and O2 cooperatively regulate in the developing endosperm, endosperms of NG, PbfRNAi, o2, and PbfRNAi;o2 seeds were harvested at 16 DAP, and transcriptome analysis was performed by RNA-sequencing (RNA-seq) as described in Materials and Methods. About 82% of the raw reads from each sample were mapped to the annotated gene-coding regions (Table S1). Based on the global FPKM-expressing values, principal component and hierarchical cluster analysis showed that the three biological replicates of each sample were clustered into one group (Fig. S1 A and B). On average 21,788 expressed genes were detected for each sample (Fig. S1C), consistent with the previous transcriptome analysis of developing endosperms (9, 28). A total of 1,296 up-regulated and 1,513 down-regulated genes were detected on average in the three mutants compared with NG (Fig. S1D and Dataset S1). Furthermore, carbohydrate- and amino acid metabolism-related Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were preferentially enriched in the differentially expressed genes of o2 and PbfRNAi;o2 mutants (Dataset S2). For example, lysine degradation, which is related to the nutritional quality and is under regulation by O2 (30), was significantly enriched in the down-regulated genes of the o2 and PbfRNAi;o2 mutants. The Gene Ontology (GO) terms associated with ribosome, translation, protein folding, and so forth were also enriched in the differentially expressed genes in the three mutants, especially in PbfRNAi;o2 (Dataset S2), consistent with the recent study of the O2 regulatory network (31). Among them, the nutrient reservoir activity (GO: 0045735), mainly containing zein genes, was significantly enriched in the down-regulated genes of all of the three mutants. In general, these results indicate that sugar and protein metabolism were significantly affected in the three mutants.

Table S1.

Summary of RNA-seq read mapping results

Sample ID	Raw high-quality reads	Coding region reads		5′ UTR reads		3′ UTR reads		Intron reads
		Number	% mapped*	Number	% mapped*	Number	% mapped*	Number	% mapped*
W64A_1†	52,478,380	37,050,843	70.60	17,170	0.03	23,118	0.04	4,318,513	8.23
W64A_2†	66,224,270	48,814,990	73.71	19,975	0.03	22,297	0.03	5,768,940	8.71
W64A_3†	52,012,184	38,891,576	74.77	15,114	0.03	17,247	0.03	4,673,418	8.99
PbfRNAi_1	59,701,484	48,773,223	81.70	23,572	0.04	35,326	0.06	5,133,531	8.60
PbfRNAi_2	53,525,698	42,413,587	79.24	20,529	0.04	24,832	0.05	4,962,937	9.27
PbfRNAi_3	47,772,696	36,971,619	77.39	18,468	0.04	22,046	0.05	3,920,614	8.21
W64Ao2_1	55,532,508	46,658,384	84.02	27,689	0.05	45,464	0.08	3,953,315	7.12
W64Ao2_2	46,934,234	40,423,146	86.13	26,144	0.06	28,767	0.06	2,377,039	5.06
W64Ao2_3	58,023,490	50,797,985	87.55	33,219	0.06	35,888	0.06	2,899,130	5.00
PbfRNAi;o2_1	49,164,098	44,252,647	90.01	28,787	0.06	52,330	0.11	2,024,548	4.12
PbfRNAi;o2_2	52,298,914	47,580,947	90.98	26,776	0.05	30,891	0.06	2,835,284	5.42
PbfRNAio2_3	48,351,186	42,646,296	88.20	24,086	0.05	27,715	0.06	2,371,960	4.91

^*The percentage of mapped reads in raw reads.

^†The three biological replicates per each genotype.

Fig. S1.

Global transcriptome analysis based on RNA-seq data in 16-DAP endosperms of NG, PbfRNAi, o2, and PbfRNAi;o2 seeds. (A) Principal component analysis. (B) Hierarchical cluster analysis. The color scale represents the Pearson correlation. (C) Venn diagram exhibiting the number of expressed genes in the four genotypes. (D) The number of genes differentially expressed in the single and double mutants compared with WT.

To examine whether starch reduction in the three mutants is directly related to the changes in the expression of genes involved in starch synthesis, we analyzed the transcript levels of starch biosynthetic pathway genes preferentially expressed in the filling stage of endosperm and PPDK as a component of starch synthetic complexes (9, 14) and also the transcript levels of oxidative pentose phosphate pathway genes closely related with starch synthesis (Tables S2 and S3) (18, 19). The most recent study of the o2 transcriptome showed that O2 regulated the expression of endosperm PPDK1 and PPDK2 (31). More specifically, we found that their expression was mildly down-regulated in PbfRNAi but was significantly reduced in o2 and in the double mutant PbfRNAi;o2 compared with NG (Fig. 2A). In addition, the expression of 11 starch biosynthetic genes, SSIII, SSI, Su1, and SSIIa was down-regulated in o2 and/or PbfRNAi;o2, whereas Su1 and Zpu1 levels were up-regulated in PbfRNAi (all based on t test analysis) (Fig. 2A). To confirm the RNA-seq data, we performed quantitative RT-PCR to analyze the expression of these genes individually in the three mutants and found overall correlation between the two datasets (Fig. 2B).

Table S2.

The FPKM-expressing value of the starch biosynthetic pathway and pentose phosphate pathway in the 16-DAP endosperms of the four genotypes

Gene product	Gene ID (gene name)	W64A_rep1	W64A_rep2	W64A_rep3	PbfRNAi_rep1	PbfRNAi_rep2	PbfRNAi_rep3	W64Ao2_rep1	W64Ao2_rep2	W64Ao2_rep3	PbfRNAi;o2_rep1	PbfRNAi;o2_rep2	PbfRNAi;o2_rep3
Starch biosynthetic pathway
Sucrose synthase	GRMZM2G089713 (Sh1)	1,561.94	1,811.71	1,886.77	1,901.40	2,153.72	2,419.70	2,217.65	2,974.07	3,115.50	1,835.95	2,953.02	3,042.03
ADP-glucose pyrophosphorylase	GRMZM2G068506 (Bt2)	642.68	893.67	872.21	565.60	1,062.85	970.92	1,179.28	2,716.87	2,596.26	1,314.10	1,551.17	1,563.60
	GRMZM2G429899 (Sh2)	1,267.43	1,738.36	1,620.42	1,393.88	1,984.70	1,876.37	2,243.02	4,013.90	4,229.17	2,443.40	3,001.60	,2913.36
Granule-bound starch synthase	GRMZM2G024993 (GBSSI)	1,084.62	1,613.43	1,689.23	1,447.76	1891.00	1,991.09	1,424.76	2,201.58	2,127.19	1,490.89	2,004.02	2,067.30
Starch synthase	GRMZM2G129451 (SSI)	1,26.38	151.27	146.73	134.98	203.02	185.51	150.85	225.94	221.69	205.72	190.22	188.46
	GRMZM2G348551 (SSIIa)	52.69	58.76	57.26	53.73	77.05	79.00	72.49	109.63	104.61	76.22	92.16	92.15
	GRMZM2G141399 (SSIII)	59.24	74.29	68.88	69.26	107.65	88.84	55.05	108.01	105.21	70.29	74.46	70.01
Starch branching enzyme	GRMZM2G088753 (SBEI)	106.57	150.99	148.19	193.32	307.47	292.64	175.96	328.20	322.73	126.13	205.40	210.75
	GRMZM2G032628 (SBEIIb)	515.75	645.63	641.82	771.57	1,001.82	953.84	1,016.41	1,898.75	1,877.34	970.73	1,363.32	1,335.92
Starch debranching enzyme	GRMZM2G138060 (Su1)	117.92	114.86	119.31	203.95	197.11	194.63	173.99	202.58	189.85	218.30	258.83	252.34
	GRMZM2G158043 (Zpu1)	28.34	36.15	35.29	52.94	68.36	61.82	85.79	71.74	62.99	53.34	93.36	84.79
Pentose phosphate pathway
Glucose-6-phosphate 1-dehydrogenase	GRMZM2G031107	12.32	11.10	11.26	11.84	10.69	10.94	15.55	16.28	16.57	17.86	17.66	15.67
	GRMZM2G130230	1.70	2.59	2.42	1.56	2.40	2.54	0.95	1.39	1.41	1.06	2.03	2.48
	GRMZM2G177077	11.01	8.82	8.37	11.46	9.87	9.37	13.20	17.56	17.09	8.84	8.18	8.77
	GRMZM2G426964	2.58	1.50	1.40	2.85	2.39	2.19	2.52	2.00	1.52	4.06	2.91	2.28
6-Phosphogluconolactonase	GRMZM2G122126	0.70	1.03	1.32	7.10	4.25	4.73	1.34	1.18	1.55	3.62	2.88	3.26
	GRMZM2G136918	87.52	81.26	81.84	116.56	100.66	108.72	152.94	173.42	178.26	228.72	202.16	205.28
	GRMZM2G148769	20.22	24.64	25.93	29.59	24.74	31.32	24.59	27.48	29.01	23.18	31.02	32.78
6-Phosphogluconate dehydrogenase	GRMZM2G127798 (PGD1)	5.92	7.83	8.61	6.98	7.44	8.66	7.34	7.54	7.76	9.95	8.26	8.89
	GRMZM2G145715 (PGD2)	19.11	19.41	19.37	18.10	15.95	20.87	15.39	14.67	16.42	15.07	13.91	16.65
	GRMZM2G440208 (PGD3)	14.77	11.65	11.93	25.71	16.19	18.47	15.26	12.39	13.09	21.67	26.33	27.74
Ribose 5-phosphate isomerase	GRMZM5G874903	0.82	1.12	0.80	3.11	1.27	1.64	1.74	0.84	0.81	0.88	1.36	1.30
	GRMZM2G104070	3.78	2.97	2.43	3.62	2.66	3.53	4.90	4.45	4.57	4.57	3.46	3.49
	GRMZM5G891282	1.76	1.02	0.93	1.36	2.05	3.36	5.54	2.37	1.95	10.12	3.42	4.57
Ribulose-phosphate 3-epimerase	GRMZM2G026807	7.63	7.42	7.71	10.16	8.81	10.10	11.34	11.97	10.95	12.29	11.20	10.44
	GRMZM2G083102	4.81	3.41	4.10	6.53	5.65	6.48	3.59	3.23	2.82	5.87	4.28	4.04
	GRMZM2G178960	1.01	1.37	1.44	2.04	1.28	1.42	2.69	2.89	3.10	2.54	1.51	1.32
Transketolase	GRMZM2G033208	5.89	5.82	6.42	9.37	9.77	11.06	7.18	6.96	6.90	20.32	17.43	17.52
Transaldolase	GRMZM2G134256	47.83	49.15	49.01	43.18	78.52	87.33	73.05	99.10	97.12	86.33	24.03	23.52
	GRMZM2G139550	3.85	3.79	3.21	5.28	4.93	4.18	3.56	5.20	5.25	5.87	3.87	4.05

Table S3.

PPDK and actin

Gene product	Gene ID (gene name)	W64A_rep1	W64A_rep2	W64A_rep3	PbfRNAi_rep1	PbfRNAi_rep2	PbfRNAi_rep3	W64Ao2_rep1	W64Ao2_rep2	W64Ao2_rep3	PbfRNAi;o2_rep1	PbfRNAi;o2_rep2	PbfRNAi;o2_rep3
Pyruvate orthophosphate dikinase	GRMZM2G306345 (PPDK1)	636.14	468.95	463.83	705.61	500.20	501.47	237.52	200.56	198.90	121.12	97.57	100.24
	GRMZM2G097457 (PPDK2)	1,313.17	1,351.74	1,300.36	1,460.71	1,411.15	1,377.97	1,024.67	1,539.59	1,518.23	798.00	1,374.66	1,361.23
Reference gene	GRMZM2G126010 (Actin)	144.91	135.92	139.34	183.83	158.21	178.39	230.67	273.63	284.33	293.88	258.12	275.25

Fig. 2.

Expression of 11 starch-synthetic genes and two PPDKs in the NG, PbfRNAi, o2, and PbfRNAi;o2 seeds. (A) Relative expression of these genes in the mutants relative to NG based on the RNA-seq data in Tables S2 and S3. (B) Relative expressions of these genes in the mutants relative to NG based on quantitative RT-PCR. The data shown in A and B are from three biological replicates per sample and are presented as ± SD. Asterisks indicate a significant difference from W64A (Student’s t test, P < 0.05). (C) Protein accumulations of these genes in the NG, PbfRNAi, o2, and PbfRNAi;o2 seeds. Non-zein proteins from 18-DAP endosperms were separated by SDS/PAGE and subjected to immunoblot analysis. Each lane was loaded with 12 μg total protein. ACTIN and a piece of replicated gel stained with Coomassie Brilliant Blue (CBB) were used as loading controls.

Consistent with the RNA transcript levels, the protein levels of PPDK, SSIII, and SSIIa in 18-DAP endosperms were reduced in the o2 mutant and were decreased further in the double mutant (Fig. 2 B and C). Although the RNA transcript level of SBEI was not significantly decreased in the o2 mutant, its protein level was lower than those in NG and PbfRNAi. Like PPDK, SSIII, and SSIIa, its protein content was most severely affected in the PbfRNAi;o2 double mutant (Fig. 2C). Furthermore, immunoblotting analysis of starch synthetic enzymes in 24-DAP and 32-DAP endosperms demonstrated that protein levels of PPDKs, SSIII, SSIIa, and SBEI were more reduced in the double mutant than in NG or in the two single mutants (Fig. S2). These results indicate a strong correlation between the reduction of starch synthesis in the o2 and PbfRNAi;o2 mutants and the down-regulation of genes encoding starch biosynthetic enzymes.

Fig. S2.

Immunoblotting analysis of protein levels of starch synthetic enzymes in 24- and 32-DAP endosperms of NG and the three mutants. A piece of replicated gel stained with Coomassie Brilliant Blue (CBB) was used as a loading control. The loading protein concentration of the eight samples is listed in the table. Each lane was loaded with 3 μL proteins.

Activation of the PPDK and SSIII Promoters by PBF and O2.

To investigate whether this correlation can be explained by direct transactivation of these starch-synthetic genes with PBF and O2, the dual luciferase transcriptional activity assay was used. In this system, the reporter construct contains two luciferase cassettes, one being the Renilla LUC (REN) reporter gene driven by the 35S promoter (35S-REN) that is used as an internal control, and the other being the firefly luciferase (LUC) driven by the target gene promoter (Fig. 3A). Here, the starch synthetic gene promoters were fused with LUC, giving rise to 13 reporter constructs. The effectors were 35S-O2 and 35S-PBF (Fig. 3A and Fig. S3). Consistent with previous reports that O2 regulates PPDK1 and PPDK2 (31, 32), coexpression of Pppdk1-LUC or Pppdk2-LUC with 35S-O2 resulted in a 17- or 44-fold increase in LUC activity, respectively, compared with the control. Although 35S-PBF alone had limited activation on the two promoters, the combination of 35S-PBF and 35S-O2 caused 50- and 80-fold increases in LUC signal, indicating that PBF and O2 provide additive activation of the PPDK1 and PPDK2 promoters (Fig. 3A). For the SSIII promoter, both 35S-PBF and 35S-O2 alone exerted strong activation on its transcription, yielding 15- and 35-fold enhancement in LUC activity, respectively. When P_SSIII-LUC was coexpressed with 35S-PBF and 35S-O2, the activation was apparently expressed as an additive action, resulting in a 64-fold increase in LUC activity (Fig. 3A). However, promoters of SSIIa and SBEI were not activated by PBF and O2 (Fig. 3A). Similar results were seen for the other eight starch synthetic gene promoters, none of which was activated by PBF and O2 (Fig. S3).

Fig. 3.

Transactivation of the PPDK1, PPDK2, SSIII, SSIIa, and SBEI promoters by PBF and O2. (A) Dual luciferase assays of the promoters of PPDK1, PPDK2, SSIII, SSIIa, and SBEI. (Upper) Schematic diagrams of the effector and reporter constructs. REN, Renilla luciferase; LUC, firefly luciferase; ter, terminator. (Lower) The ratios of LUC to REN activity in Arabidopsis protoplasts cotransformed with the reporter and effector. The numbers in the chart represent the fold-increase of activation compared with the negative control (empty effector). Data represent means ± SD of three independent replicates. (B and C) EMSA of PBF and O2 with probes containing the ACGT and AAAG elements in the promoters of PPDK1 and PPDK2 (B), and SSIII (C). The positions of probes in the promoters are indicated in Fig. S3. Comp, competing probes unlabeled with biotin; Mp, mutated probes with a deletion in the corresponding ACGT or AAAG core element.

Fig. S3.

The dual luciferase assay of the other eight starch-synthetic gene promoters. (Upper) The schematic diagram of the constructs. REN, Renilla luciferase; LUC, firefly luciferase; ter, terminator. (Lower) The ratio of LUC/REN activity in Arabidopsis protoplasts cotransformed with the reporters and effectors shown in the upper panel.

PBF and O2 Bind the P and O2 Boxes in the PPDK and SSIII Promoters.

PBF and O2 belong to the DOF and bZIP transcription factor families, which recognize a motif containing the AAAG (P box) and ACGT (O2 box) core element, respectively. We found that both the P and O2 box are present in the first 500 bp of the PPDK1, PPDK2, and SSIII promoters (Fig. S4). Interestingly, the core elements of the P and O2 boxes were side by side in the PPDK1 and PPDK2 promoters (Fig. S4). In the SSIII promoter, the two elements were separated by 25 bp, similar to the separation in the 22-kDa α-zein promoters (6). EMSA was used to test whether the two boxes are specifically recognized by PBF and O2. Because the P and O2 boxes are too close to one another in the two PPDK promoters, 55-bp (PPDK1) and 51-bp (PPDK2) oligonucleotides spanning the two boxes were used to examine the binding affinity. Binding of His-O2 and His-PBF fusion proteins to the probes could be visualized as retarded bands in the gel. The results reveal that the two probes are bound by O2 and PBF (Fig. 3B). It appears that O2 and PBF have a stronger affinity for the PPDK1 probe than for the PPDK2 probe. When the P and O2 boxes in the two probes were mutated, the corresponding retarded bands were abolished. The binding specificity was verified by the addition of unlabeled intact probes in the reaction, resulting in a gradual loss of all retarded bands (Fig. 3B). Similar results were obtained for the two probes from the SSIII promoter, each bearing only the P or the O2 box, showing that they were specifically bound by PBF and O2, respectively (Fig. 3C). This line of results demonstrates that PBF and O2 regulate PPDKs and SSIII directly by recognizing their P and O2 boxes, respectively.

Fig. S4.

The promoter sequences of PPDK1, PPDK2, and SSIII. PBF- and O2-binding sites are marked in the upstream 600-bp region beginning from their start codon. The P and O2 boxes are marked by red and black boldface, respectively.

The Pentose Phosphate Pathway Is Affected in PbfRNAi and o2 Mutants.

Because the pentose phosphate pathway is also important for starch synthesis in endosperm, we investigated the expression of genes in this pathway (Tables S2 and S3). Among these genes, the transcript levels of three glucose-6-phosphate dehydrogenases, one 6-phosphogluconolactonase, three 6-phosphogluconate dehydrogenases, one ribose 5-phosphate isomerase, two ribulose-phosphate 3-epimerases, one transketolase, and two transaldolases were significantly down-regulated based on t test analysis in PbfRNAi, o2, and/or PbfRNAi;o2 mutants compared with NG (Fig. S5A). We also conducted quantitative RT-PCR to validate 11 of the 13 down-regulated genes in the three mutants and found overall similar patterns, as revealed by the RNA-seq data (Fig. S5B), indicating that decrease in starch in PbfRNAi and o2 might be caused, at least in part, by the down-regulation of the pentose phosphate pathway. To investigate whether PBF and O2 could regulate the genes in this pathway directly, promoters of 12 down-regulated genes were amplified for the transcriptional activation assay. It revealed that several promoters could be transactivated by PBF or O2, but none of them additively (Fig. S6).

Fig. S5.

Relative expression of genes in the pentose phosphate pathway in NG, PbfRNAi, o2, and PbfRNAi;o2 seeds. (A) Expression of these genes in the mutants relative to NG based on the RNA-seq data in Tables S2 and S3. (B) Expression of these genes in the mutants relative to NG based on quantitative RT-PCR. The data in A and B are from three biological replicates per sample and are presented as ± SD. Asterisks indicate a significant difference from W64A (Student’s t test, P < 0.05).

Fig. S6.

The dual luciferase assay of the 12 genes in the pentose phosphate pathway. (Upper) A schematic diagram of constructs. REN, Renilla luciferase; LUC, firefly luciferase; ter, terminator. (Lower) The ratio of LUC/REN activity in Arabidopsis protoplasts cotransformed with reporters and effectors shown in the upper panel.

Discussion

Storage protein and starch synthesis are temporally coordinated during maize endosperm development through a transcriptional regulatory network linking the two different synthetic processes (8, 9, 27). This coordination is now supported by genetic and molecular evidence. PBF and O2 are two endosperm-specific transcription factors that are specifically expressed from 8 to 10 DAP. The two transcription factors control 89% of zein transcription (28). Zein mRNAs then are translated on the polyribosomes of the rough endoplasmic reticulum (RER), and the resulting proteins are translocated into the lumen of the RER, where they are assembled into protein bodies (33). In contrast to storage proteins, which have no known enzymatic function, starch biosynthesis is far more complex because the respective genes encode enzymes that are organized in interacting protein complexes and assemblies to produce mature starch granules. In fact, starch is synthesized by a series of combined reactions of a fine-tuned enzyme complex and is assembled into semicrystalline starch granules as end products. In total, 11 starch-synthetic genes in maize endosperm have been cloned (7), although some of their functions are not fully understood. Because of the lack of null mutants of PPDKs in maize, their functions could not clearly be elucidated. Based on their activity in catalyzing reversible conversion between PYR and PEP, it has been hypothesized that they are involved in the synthesis and regulatory balancing of amino acids, protein, starch, and lipids. In addition, PPDKs also have been shown to be components of the starch synthetic complex and to modulate starch synthesis by interacting with other enzymes in the complex. In fact, RNA-seq transcriptome analysis of immature endosperm reveals that the expression of 13 genes increases sharply between 8 and 10 DAP, paralleling the temporal expression pattern of storage protein zein genes (9).

Previous studies showed that the level of zein proteins was reduced markedly in the o2 mutant and was reduced further in the double mutant PbfRNAi;o2. When measuring kernel weight, we found that the KW was reduced by 20% and 43% in the o2 and PbfRNAi;o2 mutants, respectively. The TW for the two mutants also was reduced, by 13% and 23%, respectively. However, it is not likely that the reduction of zein proteins contributes significantly to the loss of kernel weight, because in total zein proteins account for only about 6% of the dry seed flour in NG. Moreover, the decrease in zein protein accumulation would result in a compensatory increase in non-zein protein synthesis, which maintains the total protein content in a relatively constant range at maturity (28). In contrast, starch constitutes around 70% of the dry seed mass and therefore is the most preponderant substance in determining the kernel weight. Indeed, SEM showed that the starch granules were progressively smaller in the PbfRNAi, o2, and PbfRNAi;o2 mutants, correlating well with the KW value. Further chemical quantification showed that the starch contents in the three mutants were decreased by 5%, 11%, and 25% (Fig. 1D). We noted that the parameters of KW, TW, and starch content were affected far more severely in the PbfRNAi;o2 double mutant than in the o2 mutant; however, PbfRNAi alone had limited influence, indicating that O2 is the main factor and acts additively with PBF, affecting starch and protein composition in parallel. Still, elevated expression of PBF, when introgressed from teosinte, is associated with increased kernel weight in inbred W22 (34).

O2 regulates many other genes in addition to zein genes. Likewise, PBF could participate in a more extensive regulatory network because of the pleiotropic phenotype of seeds, which is not yet completely understood. Previous research indicated that O2 probably is involved in the regulation of starch synthesis (35–39). Here, we found that the starch content and kernel weight were positively correlated to the O2 dosage (Fig. 1 E and F). However, it had been unclear whether O2 directly regulated some, if not all, the starch-synthetic genes. Moreover, there previously was no genetic evidence for the regulatory role of PBF in this pathway. Taking advantage of RNA-seq transcriptome analysis, we found that more than 1,000 genes were down-regulated or up-regulated in PbfRNAi and o2 mutants and that the differentially expressed genes were mainly enriched in the sugar and protein metabolism pathways. Among the down-regulated genes, the RNA transcript and/or protein levels of endosperm PPDKs and three starch biosynthetic genes, SSIII, SSIIa, and SBEI, were reduced in developing endosperms in PbfRNAi and o2 background. Using the dual luciferase assay and EMSA, we then verified that PBF and O2 could cooperatively activate the PPDK1, PPDK2, and SSIII promoters by recognizing the respective P and O2 boxes. However, the two transcription factors failed to activate the SSIIa and SBEI promoters (Fig. 3). One possible explanation for this apparent discrepancy is that o2 and PbfRNAi may affect the levels of some other transcription factor(s) or other type of mRNA regulatory factor(s) and thereby affect the transcript and protein levels of SSIIa and SBEI. This study indicates that PbfRNAi and o2 mutations are able to cause direct and indirect defects in starch synthesis.

In starch synthesis, some of the enzymes are assembled into a well-coordinated machinery complex in which the physical association of SSIII with other enzymes, including PPDK, has been proposed to play a critical role in regulatory interactions. Although PPDKs themselves are not likely to be involved directly in any catalytic step of starch synthesis, it has been proposed that they function as a modulator of the complex. Coimmunoprecipitation assays also indicated that PPDK, SSIII, SSIIa, and SBEI are associated with one other (Fig. S7), as is consistent with previous reports (13, 14). Therefore we modeled the impact of PBF and O2 on the regulation of kernel nutritional quality and yield, as shown in Fig. 4. The reduction or absence of both transcription factors can raise the nutritional quality of maize but at the same time reduce starch synthesis and therefore yield. Zein proteins are devoid of lysine and tryptophan. Mutations in Pbf and O2 enhance the nutritional value because of the reduction of zein proteins and the compensatory increase of non-zein proteins. On the other hand, PBF and O2 directly regulate PPDKs and the starch synthetic gene SSIII and also indirectly regulate SSIIa and SBEI. In PbfRNAi and o2, the down-regulation of SSIII, SSIIa, and SBEI might directly cause decreased synthesis of starch. In addition, because PPDKs and SSIII are components of the enzyme complex and are associated with other enzymes, their reduction probably would affect the efficiency with which the components of the complex are assembled into an integral structure or the stability of the complex in o2 or PbfRNAi;o2 mutants (Fig. 2C) and consequently would result in reduced starch synthesis. Furthermore, O2 and PBF regulate more than 1,000 genes, at least indirectly, and their mutations have a pleiotropic effect on endosperm development. It is very likely that other affected genes and pathways related to central metabolism [e.g., the pentose phosphate pathway (Fig. S5)] contribute to the decrease in starch (Fig. 4) (19). Given these results, the use of RNAi specifically against the expression of zein genes in the production of Quality Protein Maize (QPM) may be more effective than the classic o2 mutation, not only because α-zein RNAi is dominant but also because it does not reduce yield, as results, at least in part, from the starch biosynthesis defect demonstrated here (40, 41).

Fig. S7.

Coimmunoprecipitation assays of PPDK, SSIIa, SBEI, and SSIII. Immunoprecipitates with αPPDK, αSSIIINP, αSSIIa, and αSBEI were separated by SDS/PAGE and analyzed by immunoblot using the indicated protein antibodies. The minus sign indicates the negative control without the corresponding antibody.

Fig. 4.

A proposed model for PBF and O2 controlling kernel nutritional quality and yield in maize.

Methods and Materials

Genetic materials and molecular procedures, such as measurement of soluble sugars and starch in the mature dry seeds, RNA-seq analysis, quantitative RT-PCR, analysis of the relative expression of the starch-synthetic genes and PPDKs, generation of antibodies, immunoblot analysis, coimmunoprecipitation and the dual luciferase assays, and EMSA are described in SI Materials and Methods. Primers are listed in Table S4.

Table S4.

Primers used in this study

Gene ID	Primer sequences (5′ to 3′)
For quantitative RT-PCR
GRMZM2G306345	gaaggctgggctggattac
	aaagggagatgggattgtagc
GRMZM2G097457	cagggatgatgtggggaag
	cgtaatccagcccagtcttg
GRMZM2G141399	tgtcaacctggcgaataagc
	ggctcgttccttgtcattgtc
GRMZM2G129451	gtctgctttggctgccttg
	aggacaacaacacaggtaataatc
GRMZM2G348551	atcgtggtggctgctgaatg
	gttcacttctaggtcctgtcctgc
GRMZM2G024993	gtcgaaggcgaggagatc
	cgcttattaggttgtgcca
GRMZM2G089713	tgtttcaccgcaattcgca
	agacaggtgaacgagcaggc
GRMZM2G068506	actaatgggtgcggactactatg
	tacccgtctgtctccattgc
GRMZM2G429899	tgggagcggacacctatg
	tcaccacgattccagacctt
GRMZM2G032628	cgaaagcctggggtgtat
	cactggagcatagacgacacat
GRMZM2G088753	tgaaggggtgccaggg
	gcctccttgtcttctttgctac
GRMZM2G138060	cggtggttgttgggcttc
	cgcaatacaaggatgatggag
GRMZM2G158043	cgaagatgcacgaaatgatagg
	ctgtccttttcggcacggt
GRMZM2G126010	gctacgagatgcctgatggtc
	cccccactgaggacaacg
GRMZM2G145715	acacgtacgagagggttgac
	aatgggagagactttggactt
GRMZM2G127798	ctaccgcagggacaggct
	ttacaaagatcaatagacaagttcc
GRMZM2G440208	tagcacgccaggaatgact
	taacccacagagagaaatcacac
GRMZM2G031107	gccgcatgggaaatcttca
	ttagaaagggtcggcggaac
GRMZM2G130230	cgagagttggatacgtgcag
	atactgaatggaacacgcaa
GRMZM2G148769	acaagcaaacctcgtcagacc
	aaatcctggaggacgaatgc
GRMZM2G104070	cgccgttcctcaccgacaa
	tcacacacggacgcacacatac
GRMZM2G026807	accggagatgaaaggagacg
	gttacatcgtttaccagtctgaag
GRMZM2G083102	tgcggacccaggagccatc
	cgtgcagggtcgaagtactc
GRMZM2G134256	tggcatcgactgggaagag
	ctaagaggagagccagcaagc
GRMZM2G139550	cttgtgaagtgggaccagc
	ttaaacattgggaggtggccag
For construction of reporter vector in the duel luciferase assay
GRMZM2G306345	aaaaaagcttttgacacatgttcaaaggcatacac
	aaaaccatggtagcacagggaaacagcgctagcta
GRMZM2G097457	aaaagtcgacgcacaagtcaagacacgatcgga
	aaaaccatggtagcacggcaaagcaaagc
GRMZM2G141399	aaaaaagcttagtggcgcttgttgaggtttg
	aaaactgcaggggaagaaaagaagggtgaactc
GRMZM2G348551	aaaaaagcttgggctaggccaactggactc
	aaaaccgcggggcggcgggatcgatcgg
GRMZM2G129451	aaaaaagcttgaaggcactcattcatcggtctg
	aaaatctagatgcggagagggagagcagacag
GRMZM2G068506	aaaaaagcttcaatatgcccataatctctaaaccac
	aaaaggatccggtttgcaactcgagaattatgattg
GRMZM2G429899	aaaaaagcttctgcacctagggagctcgtatac
	aaaaccatggctcctccaactactagatacacctgc
GRMZM2G024993	aaaaaagctttgcccccaacgaattttatagaag
	aaaaccatgggccgattaatccactgcatagc
GRMZM2G032628	aaaaaagcttggatcggagggaattaaggtg
	aaaaccatggctcgccttcgcagccggatc
GRMZM2G088753	aaaaaagcttctctacactgcaaggaacccc
	aaaaccgcggcggcgtgtgagtcccatctc
GRMZM2G138060	aaaaaagcttcctccaaataaaacctgcaaataag
	aaaaccatggggagaggaagtggaggggag
GRMZM2G158043	aaaaaagcttctgggcggttttttgaacattg
	aaaaccatgggtcatccaatactccaggtagttg
GRMZM2G145715	aaaaaagcttagatagatattttccacccgcg
	aaaactgcagctcctcctacactgagaaaaaaaa
GRMZM2G127798	aaaaaagcttgatagtctgttggagtgaggtgc
	aaaactgcagctcctacctgaacacacaagcg
GRMZM2G440208	aaaaaagcttttgcgagtgagagatggtctgg
	aaaaccgcggcggtgggtgggtgggtag
GRMZM2G031107	aaaaaagcttggtttgaccgtctcagacagtcag
	aaaactgcagttgtatagtttcttcttccctcag
GRMZM2G130230	aaaaaagcttgcaacattaaggtggtgtttgg
	aaaactgcaggacaaatggcacgattctcttctac
GRMZM2G148769	aaaaaagcttgaaagatactttcagatggagc
	aaaactgcaggggtgtggtagtttgctg
GRMZM2G104070	aaaaaagcttcaattgaatacagttgaatggcttg
	aaaactgcagggtgaggggcggtggcgggagc
GRMZM2G026807	aaaaagcttctcctgtccggaataatctac
	aaaactgcaggcgcgcctctctcaatctatc
GRMZM2G083102	aaaaaagctttctcggtttatcaaaagtggtg
	aaaactgcaggtccgcttaccgctaggttg
GRMZM2G134256	aaaaaagcttagaatgaacctgtcaaagctgg
	aaaactgcagtttgctcgttggggttgttgc
GRMZM2G139550	aaaaaagcttctgctttcgtcttaaaatctaaac
	aaaactgcagggcagttgggatcggagcc

SI Materials and Methods

Genetic Materials.

W64A, PbfRNAi, W64Ao2, and PbfRNAi;o2 seeds are all in W64A background and have been described elsewhere (6, 28). W64A and W64Ao2 seeds were reciprocally crossed to create progeny with two (W64A × W64Ao2) or one (W64Ao2 × W64A) dosage of O2. All materials used in the study were propagated in 2014 and 2015.

Measurement of Soluble Sugars and Starch in Mature Dry Seeds.

Twenty mature kernels of each material were ground into fine flour using a coffee grinder. One hundred milligrams of fine flour were weighed for the measurement of soluble sugars and starch. Soluble sugars were extracted with 80% (vol/vol) ethanol and were measured by the anthrone method (26). Starch content was measured with the Total Starch Assay Kit (K-TSTA; Megazyme) according to the manufacturer’s procedure.

RNA-Seq Analysis.

Total RNA was extracted from 16-DAP endosperms using TRIzol reagent (Invitrogen) and was purified with the RNeasy Mini Kit (Qiagen). RNA-seq libraries were prepared according to the Illumina Standard library preparation kit, and Illumina HiSeqTM 2500 was used as a platform for RNA-seq at Shanghai Oe Biotech Co, Ltd. FastQC combined with the NGS QC TOOLKIT v2.3.3 were used for quality control checks on raw sequencing data; then clean reads were aligned to the B73 reference genome (RefGen_v3) and the reference gene model dataset (FGS 5b) using TopHat/Bowtie2 (ccb.jhu.edu/software/tophat/). The gene-expression value was normalized as fragments per kilobase of transcript per million mapped reads (FPKM). A gene was considered expressed only when its FPKM value was greater than zero in the three biological replicates. The differentially expressed genes were produced with the threshold of a false-discovery rate (FDR) <0.05 by the DESeq Software Packages (bioconductor.org/).

Quantitative RT-PCR.

Total RNA extracted as described above was also used for reverse transcription with the SuperScript III First Strand Kit (Invitrogen). Quantitative RT-PCR was performed as previously described (28). Primers are listed in Table S4.

Analysis of the Relative Expression of the Starch-Synthetic Genes and PPDKs.

To calculate relative gene expression from the RNA-Seq data, the FPKM values of target genes were normalized to that of Actin for each sample. Then the ratio of mutant/NG normalized value was taken as the relative expression level of the target genes. To calculate relative gene expression from real-time PCR data, the relatively quantified value was produced by the ΔΔCT method with Actin as the internal reference gene. Then the ratio of mutant/NG relatively quantified value was taken as the relative expression level of the target genes. All data were produced from the averaged value of gene expression in three biological replicates.

Generation of Antibodies and Immunoblot Analysis.

The polyclonal antisera αSSIII, αSSI, αSSIIa, αBEI, and αBEIIb were raised against their synthetic peptides at ABclonal Biotech Co., Ltd as described in ref. 13. The antibodies of Bt2 and Sh1 were purchased from Agrisera (catalog nos. AS111739 and AS152830, respectively). The PPDK and GBSSI antibodies were gifts from Baichen Wang from the Institute of Botany, Chinese Academy of Sciences (CAS) and from Peng Zhang of the Institute of Plant Physiology and Ecosystem, CAS, respectively. Non-zein proteins of 18-, 24-, and 32-DAP endosperm were extracted, as described previously (42). For each sample, 20 μg of protein was loaded and separated by SDS/PAGE on the 4–20% (wt/vol) precast polyacrylamide gel (catalog no. 4561093; Bio-Rad) or 7% (wt/vol) acrylamide gel (37.5:1 acrylamide:bisacrylamide). The separated proteins were transferred to the nitrocellulose membrane for Western blotting, which was performed according to the manufacturer’s procedure. The signal was detected with the Novex ECL Chemiluminescent Substrate Reagent Kit (catalog no. WP20005; Thermo Fisher) and was visualized under the Tanon-5200 Chemiluminescent Imaging System (Tanon Science and Technology).

Coimmunoprecipitation.

The precipitating antiserum against αSSIIINP, αSSIIa, αSBEI, or αPPDK was mixed with Pierce Protein A Magnetic Beads (200 mL bed volume) at 4 °C for 2 h, and the beads then were washed in PBS. Endosperm extracts from W64A at 18 DAP were prepared, and the protein phosphatase inhibitor NaF was added to the cell extracts as described previously (13, 14). Total protein extract of five 18-DAP endosperms was incubated with the corresponding antibody/Protein A Magnetic Beads or only with protein A-Sepharose beads at 4 °C for 2–4 h with gentle agitation. Next, the pelleted beads were washed extensively with the PBS augmented with 1% Nonidet P-40. The beads were boiled in 200 μL SDS/PAGE loading buffer and were applied to gels for immunoblot analyses.

Dual Luciferase Assay and EMSA.

The isolation of Arabidopsis mesophyll protoplasts, PEG-calcium transfection of plasmid DNA, and protoplast culture were performed according to standard protocols (43). The foreign gene-expressing vector pRI101 (Clontech) was used for the expression of 35S-PBF and 35S-O2 constructs. The transient expression vector pGreenII 0800-LUC was used for the reporter constructs by inserting the target gene promoter into its multiple cloning sites. The ratio of LUC/LUC activity was measured using the dual luciferase reporter assay system (Promega). Primers for amplifying the promoters of these genes are listed in Table S4.

The oligonucleotide probes in the promoters of PPDK1, PPDK2, and SSIII were synthesized and labeled with the Pierce Biotin 3′ End DNA Labeling Kit. Expression and purification of recombinant proteins of His-PBF and His-O2 and the EMSA procedure have been described previously (28).