Many DELLA mutants are constitutive repressors of the GA response and they have been made famous by the socalled “Green Revolution” which included wheat and rice plants that were dwarfed by their inability to respond to GA. Both D8 and D9 maize mutants do not respond to application of GA3 that causes an increase in height in normal siblings. Neither D8 nor D9 are as short as plants treated with PBZ, nor as short as GA biosynthetic mutants like d1, which suggests there are other GA receptor pathways in addition to D8 or D9 in maize.The additivity of reduced height in the D9;fun plants points to fun being outside of the DELLA GA response pathway. Since blocking GA biosynthesis with PBZ also had an additive effect on height of fun plants, the reduced height phenotype of fun is unlikely to be due to deffects in GA biosynthesis or perception in the fun mutant. The feminisation of D9;fun tassels, as well as the epistasis of the fun leaf phenotype support the hypothesis that the fun phenotype is not due to deffects in the GA pathway. Similarly, since blocking GA with PBZ did not affect the feminisation of the tassel, nor did it affect the leaf phenotype, GA itself is not required for fun feminisation nor the leaf phenotype.FUN clearly operates at some point in the sex determination pathway of Zea mays since the loss of function mutant is feminised in the tassel. This observation leads to the hypothesis that FUN has some role in the masculinisation of maize inflorescences. A pictorial summary of the double mutant tassel phenotypes is given in Figure 4-22,vertical farming racks and a textual summary and interpretations of observations of the fun double mutants follow. One observation is that fun mutants consistently have fewer branches than siblings with a wild-type copy of FUN.
This is also true for double mutants with JA biosynthetic mutant ts1 , JA elimination mutant sk1 , BR perception mutant bri1::RNAi , BR hyperresponding mutant bin2::RNAiand GA perception mutant D9 . The loss of FUN causes reduction in tassel branches in all of these mutant backgrounds. Branches are one of the earliest visible sex specific characters of a maize inflorescence. Branches occur at the base of the inflorescence and due to the basipetal development of a maize inflorescence must therefore make the decision on how to develop long before the spikelet pair meristem develops. Since both na2 and sk1 have less branching in the tassel, it would appear that BR promotes tassel branching while JA inhibits it – this is surprising since although BR is associated with maleness, JA is also associated with maleness and branching is a male trait. This observation also prompts further study, since although we have some understanding of how the various hormones work at the floral level to define sex in maize , the link between hormones, tassel branching and inflorescence sex has not been examined. fun, with both branch number and sex determination deffects, could be operating at this stage of development. A cross to the ramosa branching mutants might help explore this link. Another sex trait that is feminised in fun mutants is tassel length. Though there is variability, fun mutants on average have shorter tassels than normal siblings. This is also true for double mutants with ts1 , bin2::RNAiand D9 . However, in the sk1 background, tassel length is not consistently shorter in combination with the fun mutation . Further, addition of JA may rescue tassel length in fun plants . Thus the presence of JA can limit the reduction of tassel length associated with the fun mutation, in this way FUN may be linked to the JA pathway. While the BR pathway double mutants point to involvement in the BR pathway, it is not yet possible to rule out involvement in the GA pathway for FUN. A GA biosynthetic mutant abolish a BR biosynthetic mutant’s ability to feminise a tassel, but chemical blocking of GA by PBZ did not disrupt the feminisation of the fun tassel .
Further fun’s additivity with na2 for height mirrors the additive height phenotype of the d5;na2 double31. In summary, I believe FUN sits at an intersection between all of these discussed hormones and may function to regulate maize development at an early stage.The FUN transcript, including all three exons, was cloned from B73 shoot tissue cDNA by Phusion using primers AV2 and AV2. This reaction used an annealing temperature of 60°C, an extension time of 1 minute 30 seconds, and 40 cycles. This PCR fragment was incubated with the pENTR mix provided by Thermo Fisher Scientific in a 3:1 insert:vector ratio for 5 minutes, before this mixture was used to transform Escherichia coli C3040 cells which were plated on LB agar + Kanamycin. The resulting purified plasmid was sequenced using flanking primers and the FUN insert was found to be in the correct orientation and without mutations. 2000ng of plasmid was then cut using restriction enzyme MluI and gel purified. The cut plasmid was then used in an LR clonase reaction with an empty pEARLEYGATE-104 plasmid in a 3:1 insert:vector ratio and the entire reaction was used to transform Escherichia coli C3040 cells which were plated on LB agar + Kanamycin. The resulting purified plasmid was sequenced using flanking primers and the FUN insert was found to be in the correct orientation. The plasmid was then transformed into Agrobacterium GV3103 cells which were plated on LB agar + Kanamycin + Gentamycin. 3ml cultures were grown overnight from resulting colonies. These cultures were spun down and the pellet resuspended in 3ml MES Competency Media . After shaking at room temperature for 3 hours, this suspension was diluted to 0.05 at OD600nm. The diluted suspension was then infiltrated into young N. benthamiana leaves and the plants allowed to recover for 3 days.
The abaxial sides of the leaves were examined using a Leica DM4000B on the bright field, and 405nm and 514nm fluorescence channels . ImageJ was used to merge the resulting photos to line up YFP expression and DAPI stain in order to confirm nuclear expression.The 3rd exon of FUN was cloned into pENTR and recombined using LR clonase into pDEST15, which encodes an Nterm 6xHis tag. FUN-pDEST17 was then used to transform Rosetta cells and a one-litre culture was grown, induced with IPTG and grown for 6 hours. The culture was then spun down and the pellet collected. Nickel beads were used to purify protein from the lysed cells and the protein was resuspended in 6M urea. This protein suspension was sent to Cocalico Biologicals where it was injected into guinea pigs to produce antibody-containing sera. Concurrently, a FUN-pDEST17 plasmid was produced to create a GST-fusion of the FUN protein and antigenic FUN protein fragments. This was also transformed into Rosetta cells and the resulting protein was run over an anti-GST column to purify. Agarose beads were bound to the GST-fusion proteins and these were used to make columns to purify the guinea pig sera. The purified sera was used in a immunoblot on 12mm normal and fun tassels as well as for Western Blot on crude,vertical racking system nuclear and cytoplasmic extracts run on acrylamide gels.At the beginning of my project, FUN had already been mapped to a 638,541bp region on chromosome 3S containing 22 gene models by Thant Niang under the supervision of George Chuck. RNA-seq was carried out on two pools of ten 4-week-old SAMs collected from B73 and fun homozygotes grown in the greenhouse. The RNA-seq data for this region was examined in IGV and a C>T transition in GRMZM2G323353 was revealed. This mutation results in 233Q>STOP1 in the translated amino acid sequence . This mutation was confirmed by PCR amplification and sequencing of this region by Illumina-seq. The only other mutation detected in this region was a C>T transition in an intron of the zinc-finger protein ZmDDB3 . A similar phenotype to fun was found in an EMS screen in the A619 inbred background . Sequencing of GRMZM2G323353 in these individuals showed a C>T transition which would result in 503Q>STOP** . An allelism test showed no rescue of the fun phenotype. This second allele was thus designated fun-2. Updating of the MaizeGDB database to version 4 of the maize genome renamed GRMZM2G323353 as Zm00001d039435. Zm00001d039435 contains two more exons upstream of the single exon gene GRMZM2G323353 detailed in v3 of the maize genome. PCR of cDNA confirmed that these exons are present in the mRNA transcript .BLASTing the non-redundant maize genome with the FUN protein found partial homology to a region on chromosome 6, where chromosome 3 is known to have duplicates104. This region does not translate into a continuous peptide chain, which could mean that duplicates of this gene are punished by selection. BLAST retrieved closely conserved homologues to FUN across the grasses. With PSI-BLAST it was possible to collect homologues of FUN throughout the Plant Kingdom, including Amborella trichopoda, Rosids, Asterids and nongrass monocots. The resulting tree assembled by MAFFT from these proteins fell into the same distribution as the APG IV phylogeny105, and is therefore assumed to be reliable . Alignment of the diverse family of FUN proteins revealed conserved regions , including one of the regions predicted by NucPred to be involved in nuclear localisation, which may be important to the function of FUN. The conservation of FUN makes it likely to be an important gene in plant development in general.
Though BLASTing of ZmFUN only found one hit in the Brassicales , using the C. papaya gene as a BLAST query returned more Brassicale hits, allowing the retrieval ofan Arabidopsis thaliana homologue. The Brassicales were found to have retained the FRWW, MRLM and KKR motifs, but not the GAKHIL motif . RNA-seq datasets published by Stelpflug et al. tracking different stages of maize development106 showed that the FUN transcript is highly expressed in the developing seed and endosperm. The next highest peaks were: developing leaves, especially the base; immature and meiotic tassels; and the immature and pre-pollination cob. FUN transcript was detected in all samples examined in Stelpflug et al.’s dataset . Walley et al. collected transcript and protein from 23 developing maize tissues and mapped them to the v4 genome107. FUN was also examined in this data set, where the highest transcript reads were found indeveloping leaves, and were much lower in mature leaves. Endosperm levels were still high, but not as extreme as seen in the dataset of Stelpflug et al. Unfortunately, immature tassels were not sampled, but strikingly, no FUN transcript was detected in mature pollen, unlike all other tissues sampled. Relatively high levels of FUN transcript were found in the female spikelet, as well as some in the silk . The proteomic data mapped to v4 did not return anything for FUN, but mapped to v3, high levels of protein were found in the endosperm and the developing leaves, but not the mature leaf, nor any other tissues sampled . No phosphorylated peptides were detected in any tissue sampled. Since these datasets were highly generalised, a dataset created specifically to query auricle development in maize was also used.The entire FUN protein amino acid sequence was entered into NucPred, an online tool for prediction of nuclear proteins as well as PSORT, which predicts subcellular localisation given a protein sequence. The entire FUN protein sequence was entered into the PSIPRED prediction tool , which runs multiple protein folding and interaction predictions to come to a synthetic prediction for the structure and function of a given protein sequence. NucPred gave the FUN amino acid sequence a score of 0.98 on a scale of 0.1-1, with a score of 1 indicating certainty that the protein is nuclear . The k-NN Prediction provided by the online program PSORT lists FUN as 91.3% likely to be nuclear and 8.7% likely to mitochondrial. FFPred, part of the PSIPRED package, also agreed that FUN is likely to nuclear. FFPred also predicted that FUN is likely to be involved in DNA and RNA interactions, is likely to be involved in cytoskeletal and DNA binding, and has a high percentage of serines and glutmamic acid .