Introduction

Papaya (Carica papaya L.) is a productive fruit crop cultivated in tropical and sub-tropical regions worldwide. The consumed fruit, rich in vitamins A and C, has been recommended for the prevention of vitamin A deficiency in some countries 1, 2. Two flesh colors, yellow and red, of papaya fruit are controlled by a single gene with yellow color as dominant 3, 4. It has been previously documented that yellow-fleshed fruits have high amounts of the carotenoid β-cryptoxanthin and ζ-carotene but lack lycopene, whereas red-fleshed fruits have high amounts of lycopene along with β-cryptoxanthin, β-carotene, and ζ-carotene 2, 5, 6. The red color of papaya fruit is due to the accumulation of lycopene; the yellow color is the result of converting lycopene to β -carotene and β-cryptoxanthin 6, 7, 8.

Pigments such as lycopene and β-carotene are fat-soluble carotenoids essential to all higher plant species. Lycopene gives many red-fleshed fruits their brilliant color 9 and is found in the chloroplasts and chromoplasts of tomatoes, watermelon, guava, pink grapefruit, and papaya 10. Owing to lycopene storage in tight protein-carotenoid complexes, cooking and processing actually makes it more available in animal diets 11. Lycopene is a direct precursor of β-carotene, one of the most active and essential carotenoids. Conversion of lycopene to β-carotene is catalyzed by lycopene β-cyclase (LCY-B), as described below. Accumulation of lycopene during fruit ripening is due to down-regulation of the LCY-B gene activity 12.

Carotenoid pigments in the chloroplast of green tissues function to block the high-light lethal effects of oxygen and free radicals resulting from the transfer of electrons from triplet chlorophyll to molecular oxygen 12, 13, 14. Carotenoid pigments are synthesized in the isoprenoid biosynthetic pathway within the chloroplasts of algae and plants. Lycopene is formed from four desaturation or dehydrogenation reactions that begin with phytoene 15. Two further steps in the pathway result in the conversion of red lycopene to yellow cryptoxanthin, first converting lycopene to β-carotene by LCY-B and then to β-cryptoxanthin by β-ring hydroxylase 16, 17, 18, 19, 20.

Two cyclases, encoded by single genes in Arabidopsis, work within the carotenoid biosynthetic pathway to convert the linear lycopene structure to other carotenoids by the addition of rings on the end of the molecule 18. LCY-B adds a β-ring to each end of the lycopene molecule, resulting in the production of the yellow β-carotene. With the addition of one ε-ring to lycopene, lycopene ε-cyclase creates δ-carotene, which then is converted to α-carotene through a single β-ring addition catalyzed by LCY-B (Figure 1) 16, 18, 19, 20, 21, 22. It is through this pathway that one carotenoid is converted to another, giving the fruit its specific flesh color.

Figure 1
figure 1

Pathway of carotenoid biosynthesis in plants 8, 22, 23, 26. PSY, phytoene synthase; PDS, phytoene desaturase; ZDS, ζ-carotene desaturase; LCy-B, lycopene β-cyclase; LCY-E, lycopene ε-cyclase; CyC-B, chromoplast-specific lycopene β-cyclase; and CRTR-B, β-ring hydroxylase.

Full size image

While all plants have lycopene ε- and β-cyclases, some plants have additional cyclases. In capsicum peppers, the enzyme capsanthin-capsorubin synthase (CCS) converts violoxanthin and antheraxanthin to capsanthin and capsorubin, which give red color to the fruits. In addition, CCS also has some LCY-B activity 12, 18, 24, 25. In tomato, a second lycopene β-cyclase specific to the chromoplast has been identified 26 and named CYC-B 8. CYC-B, which is homologous to pepper CCS (86.1%), produces β-carotene in tomato fruits 27. Another enzyme in the carotenoid pathway, neoxanthin synthase (NXS), shows 98% sequence homology to CYC-B, in tomato 27. Function, gene sequences, and map position suggest that CCS and CYC-B are orthologs arising from LCY-B by gene duplication 8.

Kapoho and SunUp are parental cultivars of the hybrid Rainbow that accounts for majority of the papaya fruit produced in Hawaii. Kapoho is a popular commercial cultivar that is pear-shaped with a short neck and yellow flesh. SunUp is a transgenic red-fleshed cultivar that was developed to be resistant to papaya ringspot virus 28. SunUp was developed from the original red-fleshed papaya, Sunset, and is homozygous for the coat protein gene. Both SunUp and Kapoho play key roles in the papaya industry as commercial cultivars and as parents of the transgenic hybrid cultivar Rainbow.

Based on previous reports, lycopene appeared to be the determining factor in the red color of papaya flesh. The objectives of this project were to clone the LCY-B gene in papaya and to test its expression pattern in yellow and red papaya cultivars.

Materials and Methods

Plant materials

Young leaf tissues of SunUp and Kapoho were collected from Kunia Station, Oahu for isolating genomic DNA. Fruits of these two cultivars were collected at three fruit size expansion and three fruit ripening stages (color break, 30%, and 90% ripe) for isolating RNA to test the levels of target gene expression.

DNA isolation

Bacterial artificial chromosome (BAC) DNA was isolated from the papaya BAC library 29 using Qiagen REAL kits (Qiagen, Valencia, CA, USA). Genomic DNA was isolated from young leaf tissues using a method described previously with minor modifications 30. Concentrations of the isolated DNA were checked on 0.9% agarose gels, displaying an average DNA concentration of 30 ng/μl.

Designing degenerate primers

cDNA sequences of lycopene β-cyclase of navel orange, grape- fruit, Capsicum, tomato, tobacco, Adonis, marigold, and Arabidopsis were obtained using the Basic Local Alignment Search Tool (BLAST) available at National Center for Biotechnology Information (NCBI). All LCY-B sequences were aligned using the software Seqweb (Accelrys Inc., San Diego, CA, USA) to produce a consensus sequence. A 29-bp forward-degenerate primer and 22-bp reverse-degenerate primer were designed. The primers were checked for dimer and hairpin formation using the software Oligoanalyzer 3.0 from Integrated DNA Technologies Inc. (Coralville, IA, USA). A polymerase chain reaction (PCR) consisting of 35 cycles of denaturation at 90 °C, annealing at 50 °C, and elongation at 72 °C was carried out using genomic DNA from papaya cultivar SunUp.

In a similar manner, degenerate primers were designed for a consensus sequence obtained from cDNA sequences of CYC-B gene from tomato, CCS gene from Capsicum, and navel orange and NXS from tomato and potato, to clone a CYC-B gene in papaya. As this CYC-B gene from tomato also does not have any intron 26, genomic DNA from papaya cultivar SunUp was used. Amplification was done with tomato DNA as a control. Annealing temperatures ranging from 50-54 °C were tried.

Superpool and matrix pool construction

Matrix technology (Amplicon Express, Pullman, WA, USA) was utilized to allow location identification of specific BAC clones within the 102 384-well plates in our BAC library. A total of nine superpools (SP1-9), eight (SP1-8) containing 12 384-well plates, and one (SP9) containing six 384-well plates were assigned to cover the entire library. Each superpool contains 4608 individual clones and the entire BAC library covers 13.7× haploid-genome equivalents 29. Each individual superpool was then organized into matrix pools, positioning the clones in a matrix of plate, row, column, and diagonal pools. This matrix setup ensures that each of the clones is represented in two of the matrix pools.

Round I PCR was then performed on all of the superpools, screening the entire BAC library. The reaction was carried out in 24 μl volume consisting of approximately 30 ng/μl of the template DNA, 0.2 mM dNTPs, 1× PCR buffer, 1.5 mM MgCl2 and 2 UI of Taq polymerase (Promega, Madison, WI, USA). The PCR reaction consisted of denaturation for 3 min at 94 °C and then 35 cycles of denaturation for 30 s at 94 °C, annealing for 45 s at 55 °C, and elongation for 1 min at 72 °C, followed by an extension time of 7 min at 72 °C. The positive PCR results from Round I then underwent Round II PCR, which was performed on the matrix pools (plate, row, column, and diagonal pools) of each superpool. A third, confirmation PCR reaction was used to identify the source plates within our BAC library that contained the desired clone. All PCR confirmation was determined by running the PCR product on 0.9% agarose gel.

Southern hybridization

Concurrent with the BAC library screening using the superpool technology, both the BAC library and cDNA library of the papaya hermaphrodite and female flowers were screened for positive BACs by Southern hybridization as described previously 29.

Individual BAC clone isolation and PCR confirmation

BAC stocks were streaked onto LB agar plates containing chloramphenicol. The single colonies were inoculated and grown in LB liquid medium containing chloramphenicol. The individual BAC DNA was isolated using Qiagen Midiprep (Qiagen, Valencia, CA, USA). The positive BAC clones were confirmed by a PCR reaction consisting of a 3 min denaturation at 94 °C and then 35 cycles of denaturation at 94 °C 30s, annealing at 55 °C for 45 s, and elongation at 72 °C for 1 min, followed by an extension time at 72 °C for 7 min. The reaction was carried out in 24 μl volume consisting of either 20 ng/μl or 40 ng/μl of the template DNA, 0.2 mM dNTPs, 1× PCR buffer, 1.5 mM MgCl2 and 2 U of Taq polymerase (Promega, Madison, WI, USA).

DNA sequencing

Direct sequencing of the positive BAC DNA was started with primers TIGR994C and TIGR994D (GeneBank Accession No. CG026994). Further sequencing of the upstream and downstream regions was conducted by primer walking. After each sequencing result was obtained, a new primer was designed to allow further reading of both upstream and downstream regions.

Sequencing was done using the ABI 3700 Analyzer and the Big Dye Terminator Sequencing kit (Applied Biosystems, Foster City, CA, USA). The sequencing reaction consisted of 1 min of 96 °C denaturation and then 25 cycles of denaturation at 96 °C for 10 s, annealing at 50 °C for 5 s, and extension at 60 °C for 4 min. The sequencing reaction was carried out in a 20 μl volume consisting of 1.4 μg of template DNA, 4 μM primer, 1/16× Big Dye Reaction Premix, 5× Sequencing Buffer (Applied Biosystems), and water.

RNA isolation and preparation

Total RNA was isolated from SunUp and Kapoho fruits of three size expansion and three ripening stages of 5%, 30%, and 90% yellowing using hot phenol extraction. RNA was then treated with RNase-free RQ1 DNase I for 30 min at 37 °C to eliminate potential genomic contamination, the RQ1 DNase Stop Solution was added, and the mixture was incubated for another 10 min at 65 °C (Promega, Madison, WI, USA). Total RNA concentration was measured on a 37% formaldehyde DNase-RNase free gel.

Quantitiative real-time PCR (qPCR)

To assess lycopene β-cyclase expression at different organs and six fruit developmental stages of SunUp and Kapoho, qPCR was performed on the 5× diluted RT reaction using reagents from Applied Biosystems (Foster City, CA, USA) and primers designed from the SunUp papaya lycopene sequence (forward, 5' ACT TGC TAG ACT GCC TTG AC; reverse 5' ACA GTG GCC TGA ATC GTA AC). Papaya actin was included as the positive control. PCR was performed in a 50 μl reaction using a master mix containing 25 μl Platinum SYBR Green qPCR Supermix-UDG, 1 μl RXX reference dye, and 13 μl water. Eight microliters of the diluted RT reaction, 3 μl of 5 μM primer, and 39 μl master mix were combined and run on an Opticon2 Continuous Fluorescence Detection System (MJ Research, Waltham, MA, USA). The PCR program was as follows: 50 °C for 2 min, 95 °C for 10 s, 40 cycles of: 95 °C for 15 s, 58 °C for 20 s, 72 °C for 30 s, followed by 95 °C for 15 s, 60 °C for 15 s, and 95 °C for 15 s. Three replications for each sample were used for qPCR analyses. Analysis of variance (ANOVA) was used to test the significance of expression level.

Phylogenetic analysis

Multiple alignments of conceptual amino acid sequences were generated by using Clustal W (1.81) (http://www.cmbi.kun.nl/bioinf/tools) with a gap open penalty of 10.00 and a gap extension penalty of 0.05. Phylogenetic analyses were conducted using the neighbor-joining method as implemented by the PHYLIP program package. BLOSUM 30 protein weight matrix was used to generate distance. Bootstrap analysis (1000 replicates) was performed to assess the support of individual branches.

Results

A DNA fragment of 332 bp, which was the expected size, was amplified from SunUp using the degenerate primers designed from LCY-B orthologs (Table 1). When the primer sequences were excluded, the fragment length was 281 bp. The sequence was submitted to GenBank (Accession No. AY753202). A papaya genomic clone obtained from GenBank (Accession No. CG026994) with a size of 818 bp from cv. SunUp submitted by TIGR showed homology with papaya LCY-B.

Table 1 Degenerate primers designed from multiple LCY-B and CYC-B sequences of diverse plant species
Full size table

A pair of sequence-specific primers designed from the TIGR sequences was used to screen the superpools of papaya BAC library (see Materials and Methods). Targeted fragments that contained the clone of interest were amplified from superpools SP3, SP4, SP5, SP6, and SP8. Amplified fragments from SP3 and SP6 showing the most robust band were selected for targeted amplification of matrix pools and resulted in the identification of two positive BAC clones, 32N04 and 68A06. Southern hybridization of the targeted fragment yielded six positive BACs, including 68A06. Direct amplification of BAC DNA confirmed that BAC 68A06 contained the gene of interest. The insert size of BAC clone 68A06 was estimated, through CHEF gel analysis, to be approximately 70 kb in size.

Papaya hermaphrodite and female flower cDNA libraries were also screened using the DNA fragment amplified by primers TIGR994A and TIGR994B by Southern hybridization. This procedure resulted in the identification of a positive cDNA clone. This clone was sequenced and the obtained sequence was confirmed to be the papaya gene coding for lycopene β-cyclase, named cpLCY-B, through BLAST search at GenBank database. The total length of the cDNA sequence is 1731 bp, including 102 bp 5' untranslated region (UTR) and 342 bp 3' UTR. To obtain the complete genomic sequence of cpLCY-B, direct sequencing of positive BAC clone 68A06 through primer walking started from primers designed from the TIGR sequence. The total genomic sequence from BAC 68A06 of red-fleshed genotype SunUp reached 4118 bp, including 1281 bp of coding sequence, 2311 bp of upstream, and 526 bp of downstream sequences (GenBank Accession No. DQ415894). The full promoter region was likely covered in the 2311 bp upstream sequence. Two possible sites of the cpLCY-B promoter were predicted using RegSite Plant DB (www.softberry.com). The first predicted promoter site is located at −471 upstream of the start code and its related TATA box is at −506 upstream of the start code. The second predicted promoter is located at −896 upstream of the start codon and its related TATA box is at −925 upstream of the start codon. An almost equal length of genomic sequence was obtained from the yellow-fleshed genotype Kapoho. No sequence difference was detected between these two genotypes.

The cDNA sequence showed an 85% sequence identity to orthologs of citrus species and an 81% identity to Lycopersicon esculentum mRNA LCY-B. At the amino acid level, cpLCY-B shared 80% identity with citrus LCY-B, 78% identity with tomato LCY-B, and 54% identity with tomato CYC-B. Phylogenetic analysis also confirmed that cpLCY-B was most closely related to citrus LCY-B and then tomato LCY-B (Figure 2).

Figure 2
figure 2

(A) Phylogenic relationship of LCY-B orthologs. Bootstrap analysis (1000 replicates) was performed to assess the support of each branch. Bootstrap values (percentages) are shown on branches when over 50. Branch lengths are proportional to the number of amino acid substitutions. GenBank accession numbers for the sequences are: citrus LCY-E (BAD07293); Arabidopsis LCY-E (NP200513); tomato LCY-E (CAA74745); cyanobacteria CRTL (NP896821); Arabidopsis LCY-B (NP187634); tomato LCY-B (CAA60170); papaya LCY-B (AAV28693); citrus LCY-B (AAF44700); pepper CCS (Q42435); and tomato CYC-B (AF254793). (B) Sequence comparison of LCY-B proteins. GenBank accession numbers for the sequences are: papaya LCY-B (DQ415894); citrus LCY-B (AAF44700); Arabidopsis LCY-B (NP187634); tomato LCY-B (CAA60170); tobacco LCY-B (S66349); rice LCY-B (XP464409); corn LCY-B (AAO18661).

Full size image

qPCR was conducted using fruit samples from both SunUp and Kapoho immature fruit at three different sizes and three ripening stages at color break, 30%, and 90% ripening stages with three replications (Figure 3). The papaya endogenous actin gene was used as a control. The expression of cpLCY-B was at the same level between the yellow- and red-fleshed genotypes and among all fruit developing and ripening stages as indicated by ANOVA (Figure 4).

Figure 3
figure 3

(A) Papaya fruits at six developmental stages used for qPCR analysis from cultivars: SunUp and Kapoho. (B) Flesh color of SunUp (red) and Kapoho (yellow).

Full size image
Figure 4
figure 4

qPCR analyses of LCY-B expression in leaves, flowers, and six developmental stages of papaya fruit. Actin gene was used as positive control.

Full size image

Further expression analyses of cpLCY-B by qPCR using papaya leaves, flowers, and six different ripening stages of fruit tissues revealed significant differences among leaves, flowers, and fruits (P=5 × 10−10). The expression of cpLCY-B in leaves and flowers was seven- and three-fold higher, respectively, than in fruits (Figure 4).

The degenerate primers for the CYC-B gene did not yield any PCR products using papaya genomic DNA as a template. Different DNA concentrations and annealing temperatures were tested without success, whereas the positive control of tomato genomic DNA was amplified in every test (data not shown).

Discussion

Our attempt to identify and isolate the gene controlling papaya fruit flesh color led to the cloning of the cpLCY-B gene. The isolated cDNA clone of cpLCY-B confirmed that this gene is active on the carotenoid biosynthesis pathway in papaya. The robust expression of cpLCY-B in leaves may reflect the activities of this gene in converting δ-carotene to α-carotene in chloroplast 8. Since both yellow-fleshed Kapoho and red-fleshed SunUp have white flowers without carotenoid deposition in the petal chromoplast, the high level of cpLCY-B expression is likely from enzyme activities converting δ-carotene to α-carotene in the remaining chloroplasts of the cells. The constant low level of cpLCY-B expression during the ripening process of papaya fruit suggested that this gene was down-regulated at this developmental stage, similar to the situation with the tomato LCY-B ortholog 12.

The identical full-length genomic DNA sequence of the papaya LCY-B gene, including its promoter region from both yellow- and red-fleshed genotypes Kapoho and SunUp, ruled out the possibility that cpLCY-B was the target gene responsible for the flesh color difference among different genotypes. Additional evidence to support this possibility were the similar expression levels of cpLCY-B at six maturing stages of papaya fruits from both yellow- and red-fleshed cultivars as detected by RT-PCR and qPCR analyses. The complete conversion of lycopene to carotenoids in yellow-fleshed papaya and the partial conversion of lycopene to carotenoids in red-fleshed papaya indicated an alternative pathway for converting lycopene to β-carotene as has been noted in tomato 5, 26. The cpLCY-B is likely responsible for converting part of the lycopene to β-carotene in red-fleshed fruit and also contributing to the complete degradation of lycopene in yellow-fleshed fruit. The elusive target gene in yellow-fleshed papaya fruit, accounting for converting about most of the lycopene to β-carotene, could be either a second LCY-B gene as in tomato or a novel gene as Or in cauliflower encoding a non-carotenogenic enzyme that regulates the accumulation of carotenoids giving the orange color 26, 31.

To test for the existence of an alternative pathway β-carotene formation in papaya, we employed the approach of Ronen et al.26 for using degenerate primers to clone potential orthologs of the CYC-B gene in tomato. However, no papaya DNA fragment was amplified by the degenerate primers designed from orthologous sequences of five species. A single locus associated with fruit flesh color was previously mapped on linkage group 7 of a high-density map of papaya 32. Currently, two F2 populations segregating for flesh color were planted at the Kunia Station, Oahu, for map-based cloning of this target gene that one of its alleles causes the accumulation of lycopene in red-fleshed papaya fruit.

The Beta mutant in tomato has the phenotype of orange fruit due to the accumulation of β-carotene at the expense of lycopene. This B locus encodes CYC-B that has identical sequence in the coding region, but different short sequence elements in the promoter region between the mutant and the wild type 8, 26. Extensive upstream sequence of the cpLCY-B gene was obtained likely including the promoter region. No DNA sequence variation was found between yellow-fleshed genotype Kapoho and red-fleshed SunUp. The absence of differences in gene expression level among genotypes and six maturing stages discounts this gene from being the causal agent of the flesh color phenotype.

The molecular bases of fruit flesh color difference in tomato, bell pepper, and onion have been elucidated through cloning and characterization of major genes in the carotenoid and flavonoid pathways. To further characterize the carotenoid biosynthetic pathway in papaya, it will be necessary to use a number of approaches including cloning orthologs of lycopene ε-cyclase (LCY-E) and CCS, map-based cloning of the target gene, and examining the expression pattern of these genes in leaves, flowers, and different ripening stages of fruit. These genes are likely up-regulated or down-regulated in different tissues and at different developmental stages as demonstrated in tomato 12, 26, 33. Moreover, the efficiency of accumulating or degrading ζ-carotene could affect the lycopene metabolism in yellow and red fruits. Therefore, it might be necessary to clone the phytoene desaturase and ζ-carotene desaturase and test their expression levels in yellow and red genotypes to determine whether the activities of these two enzymes are the same in both genotypes.

cpLCY-B showed highest sequence homology with citrus LCY-B, even though papaya is phylogenetically closer to Arabidopsis. A similar situation was found in PFL, a papaya LFY ortholog 34 that was more closely related to sycamore than to Arabidopsis. The different rates of divergence among functional genes might result from variable selection pressure on particular traits relevant to their roles contributing to the fitness of the species.

Fruit flesh color is often a targeted trait for papaya improvement because this trait is linked to shelf-life and nutritional value. Red-fleshed papaya is preferred by some consumers, but it is associated with a faster rate of softening, thus a shorter shelf-life. While higher content of pro-vitamin A carotenoids is a direct consequence of carotenoid metabolism in the fruit cells, the different composition of carotenoids in chromoplasts of fruit cells would have no relevance to the rate of cell wall degradation causing softening of the fruit. The red flesh and short shelf-life are most likely controlled by linked genes and the linkage might be overcome by selection in sufficiently large segregating populations.