Genetic evidence that the two isozymes of sucrose synthase present in developing maize endosperm are critical, one for cell wall integrity and the other for starch biosynthesis

In maize, two paralogous genes, Sh1 and Sus1, encode two biochemically similar isozymes of sucrose synthase, SS1 and SS2, respectively. Previous studies have attributed the mild starch deficiency of the shrunken1 (sh1) endosperm to the loss of the SS1 isozyme in the mutant. Here we describe the first mutation in the sucrose synthase1 (Sus1) gene, sus1-1, and the isolation of a double recessive genotype, sh1 sus1-1. Combined data from diverse studies, including Northern and Western analyses, RT-PCR and genomic PCR, cloning and sequencing data for the 3′ region, show that the mutant sus1-1 gene has a complex pattern of expression, albeit at much reduced levels as compared to the Sus1 gene. Endosperm sucrose synthase activity in sh1 sus1-1 was barely 0.5% of the total activity in the Sh1 Sus1 genotype. Significantly, comparative analyses of Sh1 Sus1, sh1 Sus1 and sh1 sus1-1 genotypes have, for the first time, allowed us to dissect the relative contributions of each isozyme to endosperm development. Starch contents in endosperm of the three related genotypes were 100, 78 and 53%, respectively. Anatomical analyses, which confirmed the previously described early cell degeneration phenotype unique to the sh1 Sus1 endosperm, revealed no detectable difference between the two sh1 genotypes. We conclude that the SS1 isozyme plays the dominant role in providing the substrate for cellulose biosynthesis, whereas the SS2 protein is needed mainly for generating precursors for starch biosynthesis.     

Genetic evidence that the two isozymes of sucrose synthase present in developing maize endosperm are critical, one for cell wall integrity and the other for starch biosynthesis

In maize, two paralogous genes, Sh1 and Sus1, encode two biochemically similar isozymes of sucrose synthase, SS1 and SS2, respectively. Previous studies have attributed the mild starch deficiency of the shrunken1 (sh1) endosperm to the loss of the SS1 isozyme in the mutant. Here we describe the first mutation in the sucrose synthase1 (Sus1) gene, sus1-1, and the isolation of a double recessive genotype, sh1 sus1-1. Combined data from diverse studies, including Northern and Western analyses, RT-PCR and genomic PCR, cloning and sequencing data for the 3′ region, show that the mutant sus1-1 gene has a complex pattern of expression, albeit at much reduced levels as compared to the Sus1 gene. Endosperm sucrose synthase activity in sh1 sus1-1 was barely 0.5% of the total activity in the Sh1 Sus1 genotype. Significantly, comparative analyses of Sh1 Sus1, sh1 Sus1 and sh1 sus1-1 genotypes have, for the first time, allowed us to dissect the relative contributions of each isozyme to endosperm development. Starch contents in endosperm of the three related genotypes were 100, 78 and 53%, respectively. Anatomical analyses, which confirmed the previously described early cell degeneration phenotype unique to the sh1 Sus1 endosperm, revealed no detectable difference between the two sh1 genotypes. We conclude that the SS1 isozyme plays the dominant role in providing the substrate for cellulose biosynthesis, whereas the SS2 protein is needed mainly for generating precursors for starch biosynthesis. Received: 22 January 1998 / Accepted: 30 March 1998     

Evidence for plasma membrane-associated forms of sucrose synthase in maize

Plasma membrane fractions were isolated from maize (Zea mays L.) endosperms and etiolated kernels to investigate the possible membrane location of the sucrose synthase (SS) protein. Endosperms from seedlings at both 12 and 21 days after pollination (DAP), representing early and mid-developmental stages, were used, in addition to etiolated leaf and elongation zones from seedlings. Plasma membrane fractions were isolated from this material using differential centrifugation and aqueous two-phase partitioning. The plasma membrane-enriched fraction obtained was then analyzed for the presence of sucrose synthase using protein blots and activity measurements. Both isozymes SS1 and SS2, encoded by the lociSh1 andSus1, respectively, were detected in the plasma membrane-enriched fraction using polyclonal and monoclonal antisera to SS1 and SS2 isozymes. In addition, measurements of sucrose synthase activity in plasma membrane fractions of endosperm revealed high levels of specific activity. The sucrose synthase enzyme is tightly associated with the membrane, as shown by Triton X-100 treatment of the plasma membrane-enriched fraction. It is noteworthy that the gene products of bothSh1 andSus1 were detectable as both soluble and plasma membrane-associated forms.     

A Comparison of Two Sucrose Synthetase Isozymes from Normal and shrunken-1 Maize

The maize sucrose synthetase isozyme (SS2) present in sh1 endosperm, sh1 seedlings, and in suspension culture cells was purified to homogeneity from each of these tissues by sequential ammonium sulfate fractionation, diethylaminoethyl-cellulose chromatography, gel filtration chromatography, and affinity elution with UTP from a carboxymethyl-cellulose column. Cyanogen bromide digests were used to demonstrate that the SS2 isozymes in these different tissues are structurally identical and are therefore the product of the same gene. The sucrose synthetase produced by the Sh1 gene (SS1) was purified by modification of the SS2 procedure and was used in comparative analyses of the two isozymes. Ouchterlony assays demonstrated that SS1 and SS2 have partial antigenic identity. The two isozymes have similar enzyme kinetics in the sucrose cleavage reaction but differ in their relative activities with ADP and TDP. The amino acid compositions of SS1 and SS2 are similar, and proteolytic digests revealed that they share limited structural homologies.     

Gene expression studies on developing kernels of maize sucrose synthase (SuSy) mutants show evidence for a third SuSy gene

Previous studies have identified two tissue- and cell-specific, yet functionally redundant, sucrose synthase (SuSy) genes, Sh1 and Sus1, which encode biochemically similar isozymes, SH1 and SUS1 (previously referred to as SS1 and SS2, respectively). Here we report evidence for a third SuSy gene in maize, Sus3, which is more similar to dicot than to monocot SuSys. RNA and/or protein blot analyses on developing kernels and other tissues show evidence of expression of Sus3, although at the lowest steady-state levels of the three SuSy gene products and without a unique pattern of tissue specificity. Immunoblots of sh1sus1-1 embryos that are either lacking or deficient for the embryo-specific SUS1 protein have shown a protein band which we attribute to the Sus3 gene, and may contribute to the residual enzyme activity seen in embryos of the double mutant. We also studied developing seeds of the double mutant sh1sus1-1, which is missing 99.5% of SuSy enzyme activity, for evidence of co-regulation of several genes of sugar metabolism. We found a significant reduction in the steady-state levels of Miniature-1 encoded cell wall invertase2, and Sucrose transporter (Sut) mRNAs in the double mutant, relative to the lineage-related sh1Sus1 and sh1Sus1 kernels. Down-regulation of the Mn1 gene was also reflected in significant reductions in cell wall invertase activity. Co-regulatory changes were not seen in the expression of Sucrose phosphate synthase, UDP-glucose pyrophosphorylase, and ADP-glucose pyrophosphorylase.     

Identification of the Soluble Starch Synthase Activities of Maize Endosperm

This study identified the complement of soluble starch synthases (SSs) present in developing maize (Zea mays) endosperm. The product of the du1 gene, DU1, was shown to be one of the two major soluble SSs. The C-terminal 450 residues of DU1 comprise eight sequence blocks conserved in 28 known or predicted glucan synthases. This region of DU1 was expressed in Escherichia coli and shown to possess SS activity. DU1-specific antisera detected a soluble endosperm protein of more than 200 kD that was lacking in du1- mutants. These antisera eliminated 20% to 30% of the soluble SS activity from kernel extracts. Antiserum against the isozyme zSSI eliminated approximately 60% of the total soluble SS, and immunodepletion of du1- mutant extracts with this antiserum nearly eliminated SS activity. Two soluble SS activities were identified by electrophoretic fractionation, each of which correlated specifically with zSSI or DU1. Thus, DU1 and zSSI accounted for the great majority of soluble SS activity present in developing endosperm. The relative activity of the two isozymes did not change significantly during the starch biosynthetic period. DU1 and zSSI may be interdependent, because mutant extracts lacking DU1 exhibited a significant stimulation of the remaining SS activity.     

Subcellular localization of isozymes of NAD-dependent malate dehydrogenase in sugar beet <Emphasis Type="Italic">Beta vulgaris</Emphasis> L.

Subcellular localization of isozymes of NAD-dependent malate dehydrogenase (MDH) in sugar beet was studied. Isozymes ss and ll controlled by loci Mdh2 and Mdh3, respectively, were shown to locate in mitochondria, whereas isozyme pp controlled by locus Mdh1, in microbodies. All examined samples lack hybrid MDH isozymes, which could testify to the interaction between products of nonallelic Mdh genes. This can be explained by the localization of nonallelic isozymes in various compartments of the cell and organelles.     

Altered tissue specificity of the revertant shrunken allele upon Dissociation (Ds) excision is associated with loss of expression and molecular rearrangement at the corresponding non-allelic isozyme locus in maize

Summary Transposable element Activator (Ac) induced wild-type stable revertants, derived from McClintock's Dissociation (Ds) insertion shrunken (sh) mutant sh-m5933, have been examined for sucrose synthases, SS1 and SS2, encoded by the revertant (Sh) locus and the non-allelic gene Sus (previously designated as Ss2), respectively. A structurally normal Sh locus has been previously described in these revertants. Immuno-blot (Western) and Southern hybridization analyses reported here identify one of the nine alleles, Sh-r5, as unique for several features. It showed altered tissue specificity, as the SS1 protein encoded by the Sh-r5 allele was readily detectable in the immature embryo which is otherwise characterized by the Sus expression only. The level of Sh-r5 expression at the protein and enzyme level was marked by endosperm specific SS1 abundance and a significant down-regulation in the embryo similar to the standard Sh and Sus loci in endosperm and embryo, respectively. We infer that tissue specific levels of gene expression among maize Ss genes is significantly determined by trans-regulatory factors present in these two tissues. The Sh-r5 strain also exhibited a complete loss of the Sus expression in all tissues tested in the plant. Lack of any detectable phenotypic abnormality in the Sh-r5 strain due to the loss of SS2 protein indicated that either the SS2 protein is nonessential or that the two SS isozymes are functionally compensatory. Genomic filter hybridizations with the Sus cDNA clone indicated that the Sus locus in the Sh-r5 strain was not deleted and was, in fact, unique among these revertants. Together, these data provide an unusual insight into the regulation and function of the two SS isozymes in the maize plant.     

The Spatial Distribution of Sucrose Synthase Isozymes in Barley

The sucrose (Suc) synthase enzyme purified from barley (Hordeum vulgare L.) roots is a homotetramer that is composed of 90-kD type 1 Suc synthase (SS1) subunits. Km values for Suc and UDP were 30 mM and 5 [mu]M, respectively. This enzyme can also utilize ADP at 25% of the UDP rate. Anti-SS1 polyclonal antibodies, which recognized both SS1 and type 2 Suc synthase (SS2) (88-kD) subunits, and antibodies raised against a synthetic peptide, LANGSTDNNFV, which were specific for SS2, were used to study the spatial distribution of these subunits by immunoblot analysis and immunolocalization. Both SS1 and SS2 were abundantly expressed in endosperm, where they polymerize to form the five possible homo- and heterotetramers. Only SS1 homotetramers were detected in young leaves, where they appeared exclusively in phloem cells, and in roots, where expression was associated with cap cells and the vascular bundle. In the seed both SS1 and SS2 were present in endosperm, but only SS1 was apparent in the chalazal region, the nucellar projection, and the vascular bundle. The physiological implications for the difference in expression patterns observed are discussed with respect to the maize (Zea mays L.) model.     

Genome‐wide screen of genes imprinted in sorghum endosperm,and the roles of allelic differential cytosine methylation

Imprinting is an epigenetic phenomenon referring to allele‐biased expression of certain genes depending on their parent of origin. Accumulated evidence suggests that, while imprinting is a conserved mechanism across kingdoms, the identities of the imprinted genes are largely species‐specific. Using deep RNA sequencing of endosperm 14 days after pollination in sorghum, 5683 genes (29.27% of the total 19 418 expressed genes) were found to harbor diagnostic single nucleotide polymorphisms between two parental lines. The analysis of parent‐of‐origin expression patterns in the endosperm of a pair of reciprocal F₁ hybrids between the two sorghum lines led to identification of 101 genes with ≥ fivefold allelic expression difference in both hybrids, including 85 maternal expressed genes (MEGs) and 16 paternal expressed genes (PEGs). Thirty of these genes were previously identified as imprinted in endosperm of maize (Zea mays), rice (Oryza sativa) or Arabidopsis, while the remaining 71 genes are sorghum‐specific imprinted genes relative to these three plant species. Allele‐biased expression of virtually all of the 14 tested imprinted genes (nine MEGs and five PEGs) was validated by pyrosequencing using independent sources of RNA from various developmental stages and dissected parts of endosperm. Forty‐six imprinted genes (30 MEGs and 16 PEGs) were assayed by quantitative RT–PCR, and the majority of them showed endosperm‐specific or preferential expression relative to embryo and other tissues. DNA methylation analysis of the 5’ upstream region and gene body for seven imprinted genes indicated that, while three of the four PEGs were associated with hypomethylation of maternal alleles, no MEG was associated with allele‐differential methylation.