Additionally, many free‐ending veins were observed in bip0663 and ‘Lukullus’ leaves , which can promote sugar export . Leaf veins, similar to leaf shapes, also have a remarkable diversity in architecture within and across species. Numerous studies on mutant phenotypes have shown auxin signaling networks play an important role in patterning leaf vasculature . Polar auxin transport , as a positional cue, determines the sites of vascular cell differentiation, and inhibition of auxin transport can lead to excessive leaf vein growth in Arabidopsis . Since auxin also patterns blade outgrowth, leafmorphogenesis and vascular pattern formation can be coupled . Genetic and molecular studies have identified many important genes that are involved in regulating vascular pattern formation that are related to auxin transport or biosynthesis . For example, MONOPTEROS/AUXIN RESPONSE FACTOR 5 and TARGET OF MONOPTEROS 5 are essential for provascular establishment , AT HOMEOBOX 8 defines preprocambial cell state that is a prelude to vein formation, and ACAULIS5 and BUSHY AND DWARF 2 , which regulate polyamine synthesis, can inhibit auxin-induced xylem differentiation , while PHLOEM INTERCALATED WITH XYLEM and CLAVATA 3/ESR-RELATED 41/44 regulate vascular organization . Moreover, few genes such as VASCULAR-RELATED NAC-DOMAIN 6/7 , NAC SECONDARY WALL THICKENING PROMOTING FACTOR 1/2 and ALTERED PHLOEM DEVELOPMENT are also reported to be related to leaf vascular patterning though regulation of phloem or xylem differentiation .

In addition, raspberry plant pot using auxin transport inhibitor induced vascular overgrowth, Wenzel and coworkers identified many vascular related genes that were previously not known to have a role in vascular differentiation in Arabidopsis. These include At1g61800 , At3g16360, At1g07430 , and At5g17220 . These genes had vascular-related expression in transgenic Arabidopsis plants and were notably upregulated in tissue with vascular overgrowth. In our study, GPT2 was significantly down regulated in low leaf vein density genotypes such as bip0663and ‘Lukullus’, suggesting correlation between leaf vein density and expression of GPT2. Recent studies have shown that GPT2 functions in promoting cell proliferation, which would be one way to link GPT2 with leaf development and vascular patterning. Mutants in gpt2 may have accelerated chloroplast differentiation . GPT2 is also involved in starch biosynthesis . In leaves, GPT2 is responsible for transporting of Glucose-6-phosphate into plastids where G6P can be used as a carbon source for starch biosynthesis. Therefore, down regulation of GPT2 might decrease transport of glucose into plastids and increase transport of glucose out of the leaf in bip0663 and ‘Lukullus’ compared with bip2 and M82. Our study showed decreased leaf starch and sugar in bip0063 when compared to ‘Lukullus’, which may also be indicative of increased export of sugars from the leaf. There are two main explanations for the relationship between leaf morphology and fruit sugar content. The first is the purely developmental link between the leaves and the fruit. This is borne out by the fact that the morphological development of fruits and leaves is regulated by a similar gene regulatory network .

The second is that either genetic changes in overall metabolism alter leaf development or changes in carbon metabolism induce morphological changes in leaves . For example, ARF4 not only regulates the morphology of leaves and fruits , the accumulation of chloroplasts and the greening of fruits , but was also recently shown to have influence on fruit sugar levels . Because GPT2 has been shown to regulate cell proliferation and is also involved in starch biosynthesis, it could be a common regulator for leaf vein density and carbonallocation . In addition, our analyses show that DEGs enriched in “transcription and development” GO terms were highly correlated with those enriched in “carbohydrate and bio-synthetic” GO terms. In our gene co-expression network, GPT2 was in the core network involved in bio-synthetic and development processes and co-expressed with several highly connected genes , suggesting that it might be an important and conserved gene that connects leaf vein development and carbohydrate bio-synthetic process.Tomato seeds were treated with 50% bleach for 10 min and rinsed 3-5 times with water, then placed on water dampened Phytatrays . Seeds were moved to the dark and incubated at room temperature for 3 days, then transferred to a growth chamber set at 25°C with 16:8 photoperiod until seedlings had expanded cotyledons . The seedlings were then transplanted to 72 Seedling Propagation trays and grown in the chamber for 7 days. After that, seedlings were transferred to the greenhouse or grown for 2 weeks and then transplanted to field. The greenhouse plants were watered from the top to encourage hardening.

Field plants were watered with furrow irrigation once weekly.Mature fully expanded leaves from adult nodes were used for leaf complexity and shape analysis, and at least five leaves were collected from each plant for analysis. Leaf complexity is defined as the number of all leaflets present on the leaf. For leaf shape analysis, intercalary and secondary/tertiary leaflets were ignored due to their irregular shapes. Leaf shape was analyzed using a method previously described . After leaf complexity was measured, the leaflet images were used for shape and size analysis. Leaflets were imaged using Epson Perfection V600 Photo scanner , and each individual leaflet image was saved as a binary image so that the leaflet was black on a white background. The binary images were then processed in R using MOMOCS . After importing and aligning along their axes, leaflet images were then processed using elliptical Fourier analysis based on the number of harmonics calculated from the MOMOCS package . Traditional leaflet shape measures such as leaflet area , solidity, circularity, and roundness were measured based on figure pixel. PCA analysis was performed on eFourier results and statistical correlations between Principal Components and traditional leaflet shape measures were used to determine the leaf characteristics captured by each PC.For leaf vein analysis, leaf discs with an area of 0.28 cm2 and were collected by a hole puncher from the second lateral primary leaflets of each plant . Leaf discs were cleared using a modified method from the Ainsworth lab . Leaf discs were heated in 80% EtOH for 20 minutes at 80℃ and this process repeated twice or until leaf discs turned white. Leaf discs were then placed in 5% NaOH and heated to 80℃ for 5 minutes and were cooled by incubating at room temperature for 10 minutes. After that NaOH was removed and leaf discs were treated with 50% bleach for approximately 30 seconds. Bleach treatment was repeated until leaf discs were clear white. Then leaf discs were washed by ddH2O and vacuum infiltrated with 50% glycerol for 20 minutes. Cleared leaf discs were placed on slides and leaf veins of leaf disc were imaged using Eclipse C1 plus microscope at a fixed magnification . Leaf veins of images were traced and measured by LEAF GUI , a tool that facilitates improved empirical understanding of leaf veins structure . Leaf vein density is measured as leaf vein length per observed leaf area.The grafting assay used self-grafting as controls and reciprocal grafting of the two genotypes. ‘Lukullus’ grafted on bip0663 , and bip0663 grafted on ‘Lukullus’ at the first internode as the treatments, while ‘Lukullus’ grafted on ‘Lukullus’ and bip0663 grafted on bip0663 at the first internode served as controls . Thirty-day old plants grown in the chamber were used for grafting. The grafted plants were allowed to recover for about 7 days before being transferred outside into 2-gallon pots. Once the grafts start flowering, leaves from scion and flowers from the stock were continuously removed to ensure that growth of fruit on scion was only driven by root stock leaves as source. The vegetative biomass, leaf shape, leaf complexity and leaf veins , fruit yield, blueberry production and fruit sugar were measured.To determine the sugar and starch content of bip0663 and ‘Lukullus’, leaf discs with an area of 0.28 cm2 were collected with a hole puncher from the second lateral primary leaflets of each plant . These samples were taken at dusk and immediately placed into microcentrifuge tubes containing 500 L of 100% EtOH. Samples were heated at 80°C for 20 minutes.

Supernatant was immediately removed, stored at -20°C, and used for determination of sugar content . Sugar content was determined as previously described . Leaf discs were kept for further processing and starch extraction. Residual sugar was removed from leaves by two additional rounds of heating in 500 L 100% EtOH, discarding the supernatant each time. Leaf discs were resuspended in 500 L 5% NaOH and heated to 80°C for 20 minutes. Samples were cooled and neutralized with 125 L 5M HCl. The supernatant was removed, and leaf discs were rinsed 2X with ddH2O. Discs were resuspended in 500 μL 50 mM Sodium Acetate Buffer and bead beaten by hand for 30 seconds after the addition of a small metal bead. Samples were centrifuged for 1 minute at 13K rpm. 25−L of a starch degradation mixture, containing a final concentration of 1.3U Amyloglucosidase and 220U α-Amylase, was added. Samples were incubated at RT for 1 hour, followed by 65°C for 23 hours. Enzymes were deactivated by heating to 80°C for 5 minutes. These samples, now containing glucose were quantified using the same protocol as for sugar samples above.Thissue used for RNA-seq libraries for Illumina sequencing were collected using the method previously described . RNA-seq libraries were prepared from collected tissues using the BrAD-seq method . Libraries were prepared from 5 replicates of each type of tissue, collected at fruiting stage . These RNA-seq libraries were sequenced at the Vincent Coates Genomics facility at University of California, Berkeley on a single lane of the Illumina Hi-Seq 4000 platform and 50-bp single-end reads were generated. A total of 326 M raw paired-end 100 bp reads were generated, ranging from 5.7 to 16.3 M reads per library.Tomato is one of the most extensively used vegetable crops in the world. The wild relatives of tomato have diverse morphology, physiology, life history, and biochemistry, and are adapted to environments ranging from temperate deserts to wet tropical rainforests . The rich phenotypic and genetic diversity in the tomato genus has been used to successfully breed yield, sugar content and disease resistance into commercial tomato cultivars . Intensive studies on tomato and the wild tomato relatives provide a unique opportunity to understand phenotypic diversity, adaptation and yield at the genomic level. For example, S. pennellii, a more distant wild relative of the domesticated tomato has adapted to dry rocky hillsides and sandy areas . Cultivated tomatoes, along with their wild-relatives, harbor broad genetic diversity and large phenotypic variability . Comparative transcriptomics for cultivated and wild species has been used to identify transcript abundance variation underlying trait differences between them . In tomato, wide interspecific crosses can bring together divergent genomes and such hybridizations can cause extensive gene expression alterations compared to either parent . Introgression lines have been developed by crosses between wild-relatives and the cultivated tomato and prove to be a useful genetic resource for genomics and molecular breeding studies. The size of the introgressed region can vary and gene content may range from a few to more than a thousand genes. Crosses between the wild desert-adapted species Solanum pennellii and domesticated Solanum lycopersicum cv. M82 have been used to generate a very useful genetic resource in introgression lines and back-crossed introgression lines . Numerous QTLs for metabolites, enzymatic activity, yield, fitness traits, and developmental features, such as leaf shape, size, and complexity have been mapped in this population . This is usually followed by traditional identification of genes regulating a trait, utilizing QTL analysis followed by fine mapping. Several genes regulating tomato yield and fruit size have been identified using this approach. During domestication cultivated tomato was selected to produce fruit that can be 1,000 times larger than those seen in the wild progenitors . A relatively small number of loci regulate this change in fruit size and two processes controlled by these loci include cell cycle, regulated by fw2.2 , and organ number, regulated by fasciated . Although great strides have been made in identification of fruit size related QTL, QTL relating to fruit quality , the interactions between these, and the role of the vegetative shoot in generating fruit yield and quality is as yet relatively unexplored.