Genejet gel extraction

Oligonucleotide primers were
synthesized by Eurofins Genomics Germany GmbH (Ebersberg, Germany)
or Integrated DNA Technologies (Coralville, IA, USA). GeneJET Gel
Extraction and Plasmid Miniprep kits were used for DNA purification
(Thermo Fisher Scientific, Waltham, MA, USA). The Gibson assembly
master mix was purchased from New England Biolabs (Ipswich, MA, USA).
DNA fragments for plasmid construction were amplified by PCR using
Phusion HF (New England Biolabs, Ipswich, MA, USA) or PrimeStar HS
DNA polymerase (Takara Bio, Kusatsu, Shiga, Japan). DreamTaq DNA polymerase
(Thermo Fisher Scientific, Waltham, MA, USA) was used for colony PCR. CouR, RtMatB, and SpMae1 gene sequences were codon-optimized and synthesized by Doulix (Explora,
Venice, Italy). 4CL from Arabidopsis
thaliana was amplified from pCfB854.^{58 (link)} The genes CHS from Rhododendron
simsii and CHI from Paeonia suffruticosa were codon-optimized and synthesized
by GenScript Biotech (Piscataway Township, NJ, USA). Analytical standards
of naringenin (≥95%, TLC), malonic acid (≥98.5%, GC),
phloretic acid (≥97.5%, HPLC), and p-coumaric
acid (≥98%, HPLC) were obtained from Sigma-Aldrich/Merck KGaA
(Darmstadt, Germany). Phloretin (≥99%, HPLC) was obtained from
Extrasynthese (Lyon, France).

Primers used for amplifying candidate gene sequences were listed in Table 3. PCR reaction was conducted by using Phusion Flash high-fidelity PCR master mix (Thermo Fisher Scientific). PCR product was first visualized in 1% agarose gel under UV light, recovered using GeneJET gel extraction and DNA cleanup micro kit (Thermo Fisher Scientific), and used for Sanger sequencing (GenScript, Piscataway, NJ, USA).
SNPs were identified using NovoSNP program [37 (link)], which could detect heterozygous sequence data based on a cumulative scoring scheme. Multiple alignments were performed by default value, and SNP positions and the heterozygous alleles were detected and manually filtered with threshold of FScore = 10. SNPs in each candidate gene were then grouped based on three physiological parameters (Chl content, Fv/Fm, or EL), respectively.
Two-way clustering analysis was performed using PermutMatrix 1.9.3 [38 (link)], with the following setting: Clustering method: McQuitty’s criteria; Tree seriation rule: Multiple-fragment heuristic; Dissimilarity: Euclidean distance.

The limit of detection (LOD) of the multiplex reaction was determined based on a standard curve. The standard template was prepared as follows. Two large (874 bp) fragments of the TK gene and 1315 bp of the Or01 gene were amplified using the primer sets shown in Appendix A Table A1. The large fragments contained binding sites for 5ILTV.233F/R and 6ORT.467F/R primers. The 874 bp and 1315 bp PCR fragments were excised from the agarose gel and purified using GeneJET Gel Extraction (K0691, Thermo Fisher Scientific, Waltham, MA, USA). The DNA concentration of the purified fragments (37 ng/µL for 874 bp fragments; 31 ng/µL for 1315 bp fragments) was determined using a QuickDrop Micro-Volume Spectrophotometer (Molecular Devices). The template copy number was calculated using the following equation: template copies/µL = [amount of DNA (ng/µL) × (6.022 × 10²³)]/[fragment length (bp) × 10⁹ × 650], which is available from http://cels.uri.edu/gsc/cndna.html (accessed on 1 October 2022). Purified PCR fragments with known copy number (3.92 × 10¹⁰ for 874 bp fragment; 2.18 × 10¹⁰ for 1315 bp fragment) were serially diluted ten-fold from 10⁻¹ to 10⁻¹⁰ to create standard curves. The LOD of the multiplex PCR assay was determined under optimized conditions of the annealing temperature and primer concentration.