Skip to main content

The report of anthocyanins in the betalain-pigmented genus Hylocereus is not well evidenced and is not a strong basis to refute the mutual exclusion paradigm

A Research article to this article was published on 31 July 2020


Here we respond to the paper entitled “Contribution of anthocyanin pathways to fruit flesh coloration in pitayas” (Fan et al., BMC Plant Biol 20:361, 2020). In this paper Fan et al. 2020 propose that the anthocyanins can be detected in the betalain-pigmented genus Hylocereus, and suggest they are responsible for the colouration of the fruit flesh. We are open to the idea that, given the evolutionary maintenance of fully functional anthocyanin synthesis genes in betalain-pigmented species, anthocyanin pigmentation might co-occur with betalain pigments, as yet undetected, in some species. However, in absence of the LC-MS/MS spectra and co-elution/fragmentation of the authentic standard comparison, the findings of Fan et al. 2020 are not credible. Furthermore, our close examination of the paper, and re-analysis of datasets that have been made available, indicate numerous additional problems. Namely, the failure to detect betalains in an untargeted metabolite analysis, accumulation of reported anthocyanins that does not correlate with the colour of the fruit, absence of key anthocyanin synthesis genes from qPCR data, likely mis-identification of key anthocyanin genes, unreproducible patterns of correlated RNAseq data, lack of gene expression correlation with pigmentation accumulation, and putative transcription factors that are weak candidates for transcriptional up-regulation of the anthocyanin pathway.


In the plant kingdom, betalains occur only in the order Caryophyllales where they substitute the otherwise ubiquitous anthocyanin pigments [1, 2]. Although betalains are found in most families in Caryophyllales, several families have anthocyanin pigmentation and do not produce betalains. Betalains and anthocyanins have never been found in the same species and are widely held to be mutually exclusive at the organismal level [3, 4]. However, both pigments have been observed in a genetically engineered tomato plant [5], on transgenic heterologous production of betalains. The molecular basis of mutual exclusion is unclear, especially as betalain-pigmented species seem to retain all the genes encoding the necessary enzymatic machinery for anthocyanin synthesis. It remains a remarkable and largely unexplained biological conundrum that has been reinforced by repeated observations for over fifty years [6,7,8].

With this as context, Fan et al. [9] recently reported anthocyanins within the betalain-pigmented genus Hylocereus (Cactaceae), also commonly called Pitaya. Fan et al. [9] analysed the fruits of three closely related species – a red-fleshed Hylocereus polyrhizus, a white-fleshed Hylocereus undatus, and an intermediate pink-fleshed hybrid (H. polyrhizus x H. undatus). Based on the analysis, they reported to correlate the accumulation of anthocyanins with the colour of red and pink fruit pulps, and the expression levels of anthocyanin biosynthesis genes. Fan et al. [9] suggest that their findings “refute the paradigm of mutual exclusion of anthocyanins and betalains within the same species/tissue”.

However, we have doubts about the findings of Fan et al. [9] and below, we outline our concerns.

Main text

No detection of betalains

Fan et al. [9] did not report the detection of betalains in the fruits of Hylocereus cultivars in their analyses. But as cited by Fan et al. [9], a range of betalain pigments have previously been detected by numerous studies in the same species [10,11,12,13,14]. Using the same two species as Fan et al. [9], earlier studies convincingly report betalain accumulation in red-fleshed H. polyrhizus and white-fleshed H. undatus [13, 14], and, also, correlated betalain accumulation with colour development at different stages of maturity of the red-fleshed species H. polyrhizus [12]. The first step in the analysis by Fan et al. [9] was an untargeted metabolite analysis that identified 443 different metabolites, including tyrosine, L-DOPA and cyclo-DOPA-5-O-glucoside which are intermediate metabolites in the betalain pathway - but no betalains. We do not understand why betalains were not detected in an untargeted metabolite analysis, when they have been previously shown to abundantly occur in these species [11,12,13,14].

Profiling of anthocyanins does not meet widely held standards

Fan et al. [9] reported the detection of five distinct anthocyanins. However, they provided very little methodology with respect to the initial metabolic profiling, with only the following brief statement: “Extract preparation, metabolite extraction, identification and quantification were performed following standard procedures of Suzhou BioNovoGene Metabolomics Platform, Suzhou, China”. We find this to be insufficient evidence and would expect at least to see the LC-MS/MS spectra and co-elution/fragmentation of the pigments versus authentic reference compounds. This standard practice is particularly important to uphold in betalain-pigmented species where there is no prior expectation to detect anthocyanins. We believe it is also important to highlight the need for standards in metabolite analyses more widely, because studies cited by Fan et al., [9] as evidence for anthocyanins in Hylocereus have similar methodological limitations and are likely also not solid evidence for the presence of anthocyanins.

Reported anthocyanins do not clearly correlate with flesh colouration

Fan et al. [9] reported that the accumulation of anthocyanins positively correlated with pink- and red-pigmented flesh indicating their “probable contribution to flesh coloration”. However, the reported anthocyanin Delphinidin 3-rutinoside, which is blue or pink coloured, accumulates to higher levels in the white-fleshed Hylocereus undatus. Indeed, based on their ion-intensity analyses (Fig. 3) the accumulation of Delphinidin 3-rutinoside is at a higher level in the white-fleshed Hylocereus undatus than the combined accumulation of the other 4 anthocyanins in the red-fleshed Hylocereus polyrhizus. Further, if all the detected anthocyanins were combined, the white pulp cultivar (H. undatus) would have the maximum anthocyanin content in its pulp. We therefore cannot understand why the flesh of H. undatus is white, given that the authors claim anthocyanins significantly contribute to flesh colouration.

qPCR quantification is missing for key anthocyanin biosynthesis genes

Fan et al. [9] reported a significant increase in transcript abundance of genes associated with flavonoid synthesis correlated with pink- and red-coloured flesh. Genes they reported as showing this pattern include C4H (Cinnamate 4-hydroxylase, F3H (flavonoid-3’–hydroxylase), F3’5’H (flavonoid-3′,5′-hydroxylase), DFR (dihydroflavonol-4-reductase), and ANS (anthocyanidin synthase). Fan et al. [9] provide a qPCR analysis of selected genes to support their RNAseq experiment, and which is largely concordant. However, both ANS and DFR are missing from the qPCR data. This omission is difficult to understand, as these are the two most important genes for their data interpretation, as they encode late stage enzymes in anthocyanin biosynthesis. Nonetheless, their RNAseq data reports very low transcript abundance of ANS in the white-fleshed H. undatus, which is difficult to reconcile with their report of high levels of anthocyanins in the same species. Especially as anthocyanin biosynthesis is considered a model system for regulation through transcriptional control thus a good correlation between the abundance of enzyme encoding transcripts and anthocyanins is common [15,16,17].

Annotation and orthology assignment appear erroneous

The authors have deposited their raw RNAseq datasets but not their transcriptome assembly and it was not available on request. It is therefore not possible to assess their annotation directly. Nonetheless, we re-assembled their RNAseq datasets for their three taxa, with our own protocols [18]. We attempted to identify an equivalent set of anthocyanin and flavonoid biosynthesis candidate genes based on homology to previously described sequences involved in the flavonoid biosynthesis (see methods for details). The results of our annotation differ markedly from Fan et al. [9]. Most striking is that our phylogenetic analysis did not reveal a F3’5’H candidate, but strongly suggested that the only candidate is actually a F3’H. These candidates show all conserved amino acid residues expected of F3’Hs, while they lacked at least one conserved residue of F3’5’Hs. F3’5’H is a key enzyme in the biosynthetic pathway of some of their reported anthocyanins, including delphinidin. Equally striking is the absence of a true DFR sequence in our H. polyrhizus transcriptome assembly (Additional file 1). There are several putative DFR-like candidates, but DFR belongs to a large multi-gene family, and none of the putative DFR sequences is confirmed to be a DFR ortholog. When analyzing all DFR candidate sequences in a phylogenetic tree with previously described DFR sequences of other species, no sequence of the H. polyrhizus assembly falls into the Caryophyllales DFR clade. This last finding is important as DFR is a late-stage enzyme in the pathway to anthocyanin synthesis and DFR has previously been shown to have reduced and/or tissue specific expression in betalain-pigmented species [19].

Reported transcript abundance for anthocyanin genes is not reproducible

We quantified transcript abundance of each homolog separately to examine their correlation across the three differently pigmented species (Fig. 1). We did not recover the same patterns of transcript abundance for anthocyanin synthesis genes as Fan et al. [9]. We find the depiction of transcript abundance in Fan et al. [9] to be slightly visually misleading, as each gene homolog is plotted individually with the Y-axis length normalised, which has the effect of under-emphasizing when genes have relatively low abundance. We therefore re-plotted all gene homologs on the same axis, to highlight that DFR cannot be detected in transcriptome assemblies of two of three species, and ANS expression in all three Hylocereus cultivars was negligible (RPKM < 2). In summary, from re-analysis of the transcript abundances of flavonoid and anthocyanin genes, we find no evidence to support the presence of a functional anthocyanin synthesis pathway in the fruits of Hylocereus, and no evidence of correlation with pigmentation in the fruit flesh (Table 1).

Fig. 1
figure 1

Transcript abundance on a gene set that includes all genes reported by Fan et al., [9] (and with the addition of PAL, 4CL and CHI) presented in Fig. 5 of Fan et al. [9] and additional genes of the flavonoid biosynthesis. F3’5’H and DFR (marked with an *) were not detected in the transcriptome assembly and are therefore considered as no expression detectable. Transcript abundances of multiple isoforms or homologs were summarized per step in the pathway

Table 1 Comparison of the de novo transcriptome assemblies. BR Hylocereus undatus Bai Rou, FR Hylocereus polyrhizus x undatus Fen Rou, DH Hylocereus polyrhizus Da Hong

Putative MYB regulators are not homologs of known activators of the anthocyanin pathway

Fan et al. [9] discussed two MYBs and one bHLH and suggested a role for these transcription factors in the pigmentation patterns of interest. However, we found no evidence of any PAP1 R2R3-MYB homologs which typically up-regulate ANS [20] in our assemblies (Additional file 1). Moreover, we used the sequences of qPCR primers to recover the corresponding full-length sequences from our assemblies and found 3 corresponding MYB sequences compared to the 2 discussed by Fan et al. [9]. None of their sequences fall into the clade of R2R3-MYBs, but rather are similar to MYBS3 which have only a single MYB repeat (MYB1Rs). Single repeat MYBs have previously been reported as repressors of anthocyanin and flavonoid biosynthesis [21] rather than activation. Single repeat MYBs also do not interact with bHLH transcription factors, as do the R2R3-MYBs, so it is not clear what significance the authors are drawing from the expression of these MYBs or the bHLH gene. Finally, we quantified transcript abundance in their datasets using our assemblies and did not recover patterns commensurate with their qPCR data (Fig. 2).

Fig. 2
figure 2

Transcript abundance of MYB and bHLH transcription factors. 1R-MYBa, 1R-MYBb, 1R-MYBc, and bHLH were identified based on qPCR primer sequences provided by Fan et al., [9]

Materials and methods

Transcriptome assembly

RNAseq datasets of different cultivars were retrieved from the Sequence Read Archive via fastq-dump [22]. Trimming and adapter removal based on a set of all available Illumina adapters were performed via Trimmomatic v0.39 [23] using SLIDINGWINDOW:4:15 LEADING:5 TRAILING:5 MINLEN:50 TOPHRED33. We decided to use separate transcriptome assemblies for the three species, because the assembly quality appears to be superior to the quality of a combined assembly. If transcripts are not recovered through this approach, it is unlikely that they have a substantial contribution to the fruit colour. Clean read pairs were subjected to Trinity v2.4.0 [24] for de novo transcriptome assembly using a k-mer size of 25. Short contigs below 200 bp were discarded. Previously described Python scripts [25] and BUSCO v3 [26] were applied for the calculation of assembly statistics for evaluation. Assembly quality was assessed based on continuity and completeness. Although assemblies were generated for all three species, the assembly generated on the basis of the data sets of Hylocereus undatus (SRR11190792-SRR11190794) was used for all down-stream analyses.

Transcriptome annotation

Prediction of encoded peptides was performed using a previously described approach to identify and retain the longest predicted peptide per contig [25]. Functional annotation was performed by combining InterProScan5 [27] with annotation transfer from Arabidopsis thaliana and Beta vulgaris based on reciprocal best BLASTp hits [25]. Genes involved in the flavonoid biosynthesis were identified via KIPEs [28] using the peptide mode (Additional files 2 and 3). An additional tBLASTn [29] search with DFR peptide sequences was performed to screen for a putative degenerated DFR transcript which could have been missed in the BLASTp search. Predicted peptide sequences were also screened via KIPEs to identify MYBs for the transcript abundance analysis. Phylogenetic trees with pitaya candidate sequences and previously characterized sequences [30, 31] were constructed with FastTree v2 [32] (WAG + CAT model) based on alignments constructed via MAFFT v7 [33] and cleaned with pxclsq [34] to achieve a minimal occupancy of 0.1 for all alignment columns.

Transcript abundance quantification

Quantification of transcript abundance was performed with kallisto v0.44.0 [35] using the RNAseq reads and our Hylocereus undatus transcriptome assembly [18]. Customized Python scripts were applied to summarize individual count tables and to compare expression values [36].

Availability of data and materials

The datasets generated and/or analysed during the current study are available in the Bieldefeld University repository:


  1. Brockington SF, Walker RH, Glover BJ, Soltis PS, Soltis DE. Complex pigment evolution in the Caryophyllales. New Phytol. 2011;190:854–64.

    Article  CAS  Google Scholar 

  2. Brockington SF, Yang Y, Gandia-Herrero F, Covshoff S, Hibberd JM, Sage RF, et al. Lineage-specific gene radiations underlie the evolution of novel betalain pigmentation in Caryophyllales. New Phytol. 2015;207:1170–80.

    Article  CAS  Google Scholar 

  3. Kimler L, Mears J, Mabry TJ, Rösler H. On the Question of the Mutual Exclusiveness of Betalains and Anthocyanins. Taxon. 1970;19:875–8.

    Article  CAS  Google Scholar 

  4. Sheehan H, Feng T, Walker-Hale N, Lopez‐Nieves S, Pucker B, Guo R, et al. Evolution of l-DOPA 4,5-dioxygenase activity allows for recurrent specialisation to betalain pigmentation in Caryophyllales. New Phytol. 2020;227:914–29.

    Article  CAS  Google Scholar 

  5. Polturak G, Grossman N, Vela-Corcia D, Dong Y, Nudel A, Pliner M, et al. Engineered gray mold resistance, antioxidant capacity, and pigmentation in betalain-producing crops and ornamentals. PNAS. 2017;114:9062–7.

    Article  CAS  Google Scholar 

  6. Stafford HA. Anthocyanins and betalains: evolution of the mutually exclusive pathways. Plant Sci. 1994;101:91–8.

    Article  CAS  Google Scholar 

  7. Clement JS, Mabry TJ. Pigment Evolution in the Caryophyllales: a Systematic Overview*. Botan Acta. 1996;109:360–7.

    Article  CAS  Google Scholar 

  8. Timoneda A, Feng T, Sheehan H, Walker-Hale N, Pucker B, Lopez‐Nieves S, et al. The evolution of betalain biosynthesis in Caryophyllales. New Phytol. 2019;224:71–85.

    Article  Google Scholar 

  9. Fan R, Sun Q, Zeng J, Zhang X. Contribution of anthocyanin pathways to fruit flesh coloration in pitayas. BMC Plant Biol. 2020;20:361.

    Article  Google Scholar 

  10. Khan MI, Giridhar P. Plant betalains: Chemistry and biochemistry. Phytochemistry. 2015;117:267–95.

    Article  CAS  Google Scholar 

  11. Wu L, Hsu H-W, Chen Y-C, Chiu C-C, Lin Y-I, Ho JA. Antioxidant and antiproliferative activities of red pitaya. Food Chem. 2006;95:319–27.

    Article  CAS  Google Scholar 

  12. Wu Y, Xu J, He Y, Shi M, Han X, Li W, et al. Metabolic Profiling of Pitaya (Hylocereus polyrhizus) during Fruit Development and Maturation. Molecules. 2019;24.

  13. Suh DH, Lee S, Heo DY, Kim Y-S, Cho SK, Lee S, et al. Metabolite profiling of red and white pitayas (Hylocereus polyrhizus and Hylocereus undatus) for comparing betalain biosynthesis and antioxidant activity. J Agric Food Chem. 2014;62:8764–71.

    Article  CAS  Google Scholar 

  14. Wybraniec S, Mizrahi Y. Fruit Flesh Betacyanin Pigments in Hylocereus Cacti. J Agric Food Chem. 2002;50:6086–9.

    Article  CAS  Google Scholar 

  15. Castellarin SD, Pfeiffer A, Sivilotti P, Degan M, Peterlunger E, Gaspero DI. G. Transcriptional regulation of anthocyanin biosynthesis in ripening fruits of grapevine under seasonal water deficit. Plant Cell Environ. 2007;30:1381–99.

  16. Yuan Y, Chiu L-W, Li L. Transcriptional regulation of anthocyanin biosynthesis in red cabbage. Planta. 2009;230:1141–53.

    Article  CAS  Google Scholar 

  17. Outchkourov NS, Karlova R, Hölscher M, Schrama X, Blilou I, Jongedijk E, et al. Transcription Factor-Mediated Control of Anthocyanin Biosynthesis in Vegetative Tissues. Plant Physiol. 2018;176:1862–78.

    Article  CAS  Google Scholar 

  18. Pucker B, Brockington S. Pitaya transcriptome assemblies and investigation of transcript abundances. 2020.

  19. Shimada S, Otsuki H, Sakuta M. Transcriptional control of anthocyanin biosynthetic genes in the Caryophyllales. J Exp Bot. 2007;58:957–67.

    Article  CAS  Google Scholar 

  20. Borevitz JO, Xia Y, Blount J, Dixon RA, Lamb C. Activation Tagging Identifies a Conserved MYB Regulator of Phenylpropanoid Biosynthesis. Plant Cell. 2000;12:2383–93.

    Article  CAS  Google Scholar 

  21. Nakatsuka T, Yamada E, Saito M, Fujita K, Nishihara M. Heterologous expression of gentian MYB1R transcription factors suppresses anthocyanin pigmentation in tobacco flowers. Plant Cell Rep. 2013;32:1925–37.

    Article  CAS  Google Scholar 

  22. NCBI. sra-tools C NCBI - National Center for Biotechnology Information/NLM/NIH; 2020. Accessed 8 Oct 2020.

  23. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30:2114–20.

    Article  CAS  Google Scholar 

  24. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Trinity: reconstructing a full-length transcriptome without a genome from RNA-Seq data. Nat Biotechnol. 2011;29:644–52.

    Article  CAS  Google Scholar 

  25. Haak M, Vinke S, Keller W, Droste J, Rückert C, Kalinowski J, et al. High Quality de Novo Transcriptome Assembly of Croton tiglium. Front Mol Biosci. 2018;5.

  26. Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. 2015;31:3210–2.

    Article  Google Scholar 

  27. Finn RD, Attwood TK, Babbitt PC, Bateman A, Bork P, Bridge AJ, et al. InterPro in 2017-beyond protein family and domain annotations. Nucleic Acids Res. 2017;45:D190–9.

    Article  CAS  Google Scholar 

  28. Pucker B, Reiher F, Schilbert HM. Automatic Identification of Players in the Flavonoid Biosynthesis with Application on the Biomedicinal Plant Croton tiglium. Plants. 2020;9:1103.

    Article  CAS  Google Scholar 

  29. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–10.

    Article  CAS  Google Scholar 

  30. Stracke R, Werber M, Weisshaar B. The R2R3-MYB gene family in Arabidopsis thaliana. Curr Opin Plant Biol. 2001;4:447–56.

    Article  CAS  Google Scholar 

  31. Stracke R, Holtgräwe D, Schneider J, Pucker B, Rosleff Sörensen T, Weisshaar B. Genome-wide identification and characterisation of R2R3-MYB genes in sugar beet (Beta vulgaris). BMC Plant Biol. 2014;14:249.

    Article  Google Scholar 

  32. Price MN, Dehal PS, Arkin AP. FastTree 2 – Approximately Maximum-Likelihood Trees for Large Alignments. PLOS ONE. 2010;5:e9490.

    Article  Google Scholar 

  33. Katoh K, Standley DM. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Mol Biol Evol. 2013;30:772–80.

    Article  CAS  Google Scholar 

  34. Brown JW, Walker JF, Smith SA. Phyx: phylogenetic tools for unix. Bioinformatics. 2017;33:1886–8.

    Article  CAS  Google Scholar 

  35. Bray NL, Pimentel H, Melsted P, Pachter L. Near-optimal probabilistic RNA-seq quantification. Nat Biotechnol. 2016;34:525–7.

    Article  CAS  Google Scholar 

  36. Pucker B. pitaya scripts. Python. 2020. Accessed 8 Oct 2020.

Download references


We thank the Center for Biotechnology (CeBiTec) at Bielefeld University for providing an environment to perform the computational analyses. We thank Nathanael Walker-Hale for useful discussion.


BP is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – 436841671. HBS, MK and MIK are funded by Science and Engineering Research Board (ECR/2016/000952) & Department of Biotechnology (BT/PR16902/NER/95/422/2015), Government of India. SFB is funded by BBSRC High Value Chemicals from Plants Network & NERC-NSF-DEB RG88096.

Author information

Authors and Affiliations



BP performed all analyses, with contributions from HBS and MK. BP, SFB, HBS and MK prepared figures. BP, SFB and MIK wrote the manuscript. The authors read and approved the final manuscript.

Corresponding authors

Correspondence to Mohammad Imtiyaj Khan or Samuel F. Brockington.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Phylogenetic trees of anthocyanin biosynthesis sequences and putative MYB sequences detected in our pitaya transcriptome assemblies.

Additional file 2.

Peptide sequences of enzymes associated with the flavonoid biosynthesis of pitaya.

Additional file 3.

Peptide sequences of putative pitaya MYBs.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Pucker, B., Singh, H.B., Kumari, M. et al. The report of anthocyanins in the betalain-pigmented genus Hylocereus is not well evidenced and is not a strong basis to refute the mutual exclusion paradigm. BMC Plant Biol 21, 297 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: