More than a colour; how pigment influences colourblind microbes

Abstract

Many animals have pigments when they themselves cannot see colour. Perhaps those pigments enable the animal to avoid predators, or to attract mates. Maybe even those pigmented surfaces are hosts for microbes, even when the microbes do not see colour. Do some pigments then serve as a chemical signal for a good or bad microbial substrate? Maybe pigments attract or repel various microbe types? Echinoderms serve as an important model to test the mechanisms of pigment-based microbial interactions. Echinoderms are marine benthic organisms, ranging from intertidal habitats to depths of thousands of metres and are exposed to large varieties of microbes. They are also highly pigmented, with a diverse variety of colours between and even within species. Here we focus on one type of pigment (naphthoquinones) made by polyketide synthase, modified by flavin-dependent monoxygenases, and on one type of function, microbial interaction. Recent successes in targeted gene inactivation by CRISPR/Cas9 in sea urchins supports the contention that colour is more than it seems. Here we dissect the players, and their interactions to better understand how such host factors influence a microbial colonization.

This article is part of the theme issue ‘Sculpting the microbiome: how host factors determine and respond to microbial colonization’.

1. Introduction

Echinoderms come in all shapes and sizes, from the golf ball-shaped echinoid Lytechinus pictus, to the many-armed basket star Gorgonocephalus eucnemis, to the tube-shaped sea cucumber Apostichopus japonicus and the plant-like sea lily (the crinoid Metacrinus rotundus). Most echinoderms have bilaterally symmetric larvae that populate and disperse within the water column, and metamorphose into pentameric adults. These adults are usually shallow- to deep-water benthic organisms that consume sea grass and macroalgae (sea urchins), that consume shellfish (sea stars), that are suspension feeders (sea lilies), or that even process sediment (sea cucumbers). These animals are immersed in a rich variety of marine bacteria and survive through either acute defensive action or broad tolerance for beneficial microbes.

Echinoderms are also one of the more brightly coloured of marine organisms, advertizing white to reds, oranges, greens and even purples. The spines, the test (the internal skeleton making up the body of echininoids), the tube feet and the body wall each displays a plethora of colour options. These colourful echinoderms are also keystone species of the benthos and a debate over their ecological importance is currently raging. One needs look no further than the impact of the sea star wasting disease on the west coast of the United States. This disease began in 2014 and totally changed the ecosystem of the area (see current reviews [1–3]). A yet unclear pathogen decimated the population of the sunflower sea star (Pycnopodia helianthoides), a predator of the local sea urchins (largely Strongylocentrotus purpuratus). This die off of the sunflower sea star rapidly resulted in massively overpopulated prey (sea urchins), which ate the kelp beds (or liberated them by eating the base of the stalk) and anything else in the environment, such that a diverse and productive benthos ecosystem was transformed quickly into a marine desert.

Sea cucumbers are deposit-feeders, which reduce the organic load and redistribute surface sediments, making them bioremediators for coastal mariculture. Sea cucumbers also host more than 200 species of parasitic and commensal symbionts from seven phyla, thereby enhancing ecosystem biodiversity [4]. Clearly then, echinoderms are not just a pretty pigmented face, but a massively important taxon for the environment as well.

What is the function of pigments in an echinoderm? Echinoderms do contain visual capabilities (opsins) in the tips of their tube feet, although without lenses. The tissue morphology does suggest that the adults (mostly asteroids and echinoids have been tested) do actually perceive light [5,6]. Experimental conditions show that, in the species tested (e.g. the echinoid Diadema setosum and the sea star Asteria forbesi), the adults perceive light, with the greatest response sensitivities in the blue–green range [7]. Larvae also appear to perceive light and adjust their behaviour in a variety of ways, including swimming direction and gut constriction [8,9]. Larval pigment cells even change morphology in response to light [10]. There is no evidence, however, that echinoderms communicate with each other by the various pigmentation that members of the phylum contain.

Pigments in the echinoderm taxa are diverse and include carotenoids, porphyrins, melanins and quinones. Carotenoids and porphyrins are most likely plant products in the diet of the animal, which are used elsewhere in the animal as pigment. Currently, no evidence exists to argue that any echinoderm species is capable of biosynthesis of carotenoid and porphyrin pigments on their own. It is formally possible that microbes associated with the echinoderm biosynthesize carotenoid and porphyrins, though impact on the echinoderm is uncertain.

Melanins, however, do appear to be biosynthesized by many groups of echinoderms [11,12]. It is unclear, though, what genes are responsible for this activity as a perusal of the genomes of several species (Strongylocentrotus purpuratus, Lytechinus variegatus, Patiria miniata, Acanthaster planci, Apostichopus japonicas, Asterias rubens, Lytechinus pictus) listed in Echinobase.org by a general NCBI BLAST of tyrosinase (essential enzyme in melanin production) resulted in no significant orthology with melanin (tyrosinase)-positive species. A possible tyrosinase orthologue was found in the recently released Holothuria leucospilota (sea cucumber) genome, although it was fused with the haemocyanin (HCY) gene and its expression was unchanged during pigmentation. Its role in melanization in this animal is yet unclear.

2. Naphthoquinones—terminology and biochemistry

Most work on echinoderm pigmentation has been directed to understanding the biochemistry and function of the naphthoquinones present in echinoderms. The biosynthesis of naphthoquinones is widespread in plants, bacteria and fungi, including a diverse cohort of important chemicals such as jugalone, lawsone, chimaphilin and menadione (vitamin K3). These chemicals have antibacterial, anti-inflammatory, neuroprotective, dye and anti-tumour properties and these structures and activities currently are being repurposed for unique applications by tweaking their biosynthetic pathways [13,14].

The first identification of naphthoquinones in animals was made in sea urchins with the discovery of echinochromes in the coelomocytes of Echinus esculentus and Paracentrotus lividus [15,16]. On the grounds of an observed change in the absorbance spectrum under the action of strong reducing agents, MacMunn [16] concluded that the pigment was an oxygen carrier. Griffiths [17] confirmed this opinion but stated that the oxygen was much more firmly held than in oxyhaemoglobin, being removable only by the reducing action of the living cells. McLendon [18] separated a pigment in a partially purified state that he found to be spectroscopically identical with the pigment described by MacMunn, and the name echinochrome was, therefore, extended to it. It should be noted that the pigmented immune cells of the coelomic fluid (red spherule cells) have several similarities to the larval pigmented cells. Several related quinones were subsequently identified in the spines and tests of echinoids, now referred to as spinochromes [19] (of interest, Chika Kuroda was the first woman to be awarded a chemistry doctorate in Japan—and made major contributions to our understanding of pigments and in several other fields of chemistry). We now know that echinoderms have three basic quinone structures: (1) ubiquinone, a derivative of p-benzoquinone; (2) naphthoquinone in which a benzene ring is condensed with the p-benzoquinone ring; and (3) anthraquinones, wherein a benzene ring is condensed on both sides of the p-benzoquinone ring. Detection of these quinones was historically by spectroscopic analysis, then by paper chromatography (see [19]) and more recently by mass spectroscopy (LC-MS/MS). Six known polyhydroxylated 1,4-naphthoquinones (PHNQs), including spinochrome E, 2,7-dihydroxynaphthazarin, spinochrome B, spinochrome C, spinochrome A and echinochrome A, and two new PHNQs were identified tentatively from purple sea urchin (Strongylocentrotus nudus) spines by mass spectroscopy (figure 1) [20]. The pigment extract was evaluated in multiple ways, including an antioxidant activity by using 1,1-diphenyl-2-picrylhydrazyl (DPPH) scavenging capacity, an Fe²⁺ chelating assay, a reducing power assay, lipid peroxidation inhibition assay and a tertiary-butyl hydroperoxide (t-BOOH)-induced macrophages protection assay. The antioxidant ability of PHNQ is thought to be the result of a combination of iron chelation, reducing power and free-radical scavenging activity [20].

Figure 1.

Structure of major naphthoquinones in sea urchin spines.

Naphthoquinones have also been identified biochemically in various body parts of sea urchins [21]) and in other echinoderms, e.g. holothurians (sea cucumbers; [22,23]) and crinoids (sea lilies; [24]). That crinoids contain naphthoquinones is particularly important since this taxon is believed to be a member of the most ancestral echinoderm branch and one reflective of the deuterostome clade [25]. Naphthoquinones in this animal would imply that this branch of phylogeny (deuterostomes), which includes chordates, contained the machinery necessary to make these compound chemicals, and perhaps imply a function in withstanding microbial colonization and maybe in survival in an early evolutionary selection process.

Naphthoquinones are also known to act as algistats and have been shown to restrict growth and viability of especially blue–green algae (Coccomyxa astericola; [26,27]). Recent work has explored the potential activity of naphthoquinones in much greater depth. A naphthoquinone equivalent, NQ 2-0, was bioinspired, synthesized in quantity, and tested against a variety of algae species. The species most sensitive to this compound were all blue–green algae, and non-blue–greens were not at all affected, or sensitive to only high concentrations of the compound [13]. The mechanism of action of NQ 2-0 was also tested, showing that it acted by blocking electron transfer within the photosynthetic electron transport system, and it caused increased levels of reactive oxygen species (ROS), resulting in membrane damage through lipid peroxidation. Through ultrastructural analysis it was seen that the thylakoid membranes disintegrated within 12 h following NQ 2-0 treatment, and subsequent vacuolation within the cytoplasm occurred, each likely contributing to the algicidal activity observed within 24 h. These effects were highly selective for the multiple blue–green algae species tested. Further, NQ 2-0 quickly dissipated from the field sites tested, with a half-life of approximately 3.2 days. It has been reported that a broad spectrum of free-living phytoplankton can be controlled by low concentrations of naphthoquinones (as low as mg l⁻¹) [14]. These results suggest that the naphthoquinone derivative might be useful in ecological remediation projects that include blue–green algae blooms. In addition, naphthoquinones (e.g. lawsone) are widely used in pharmaceutical, chemical and other industries owing to their demonstrable bactericidal, fungicidal, anti-malarial and cytostatic properties [28]. Thus, microbial interactions with pigment chemistry play important and diverse roles both naturally in the environment and in biotechnological approaches.

3. Microbial interactions with distinct color morphs

An immune function for naphthoquinones in echinoderms in vivo has also been proposed and tested [29–33]. In fact, echinochrome showed antibiotic properties against six Gram-positive and Gram-negative marine bacteria [34]. In eggs, echinochrome A is oxidized and released, especially after fertilization, producing H₂O₂ potentially at sufficient levels for antimicrobial function [33]. Kiselev et al. [32] tested the effects of sea urchin larvae exposure to several pathogenic marine bacteria and found that pigment cell number and Sp-Pks1 expression were both increased as compared with untreated larvae. More recently Ho et al. [31] showed that the pigmented cells within a larva will track, capture and kill bacteria experimentally introduced into the gut, or the blastocoel, and that the pigment (echinochrome A) in these pigmented cells is released during these events. George [10] confirmed that pigment cells released echinochrome in response to infection. George and co-workers also found that pigment cells respond to wounding and function to facilitate repair [10,35].

Possible immune function of pigment cells in larvae inspired experiments formally testing pigment function in adults. It should be noted that the pigment cells in the larvae, as well as the coelomocytes, and tube feet all have the same basic pigment property of a deep red colour. We refer to the highly migratory, individual cells harbouring this type of pigment as immune-functioning pigment cells. These cells are present in larvae and spines, and are enriched in tube feet and coelomocytes. This pigment and these cells, though, are distinct from the highly variable pigment in the static cells of the spine and test, for example. It is these cells that give the great variation in colour within the phylum.

Spines from several different species were tested for colonizing microbes using 16S rDNA sequences to identify prokaryotic colonizers. Each spine type showed markedly distinct, and species-specific microbial populations, even allowing for the environmental contribution. Perhaps this species-specific colonization difference is not surprising since the spines from different species might have different compositions. Surprisingly, much of this community was lost and replaced by opportunistic microbes when the animals were transferred to a laboratory environment. Lytechinus variegatus was then used as a natural experiment to test microbial interactions with host pigment since the spine pigment is highly variable between individuals of this species (figure 2). Pigment was analysed by mass spectroscopy, and microbes associated with the spine were screened by 16S rDNA sequences. Although no major difference was seen in microbial load between the colour morphs, a distinct trend was seen in which the pigment compilation of echinochromes and spinochromes resulted in distinct microbial colonization [36] (figure 2). We also noted that fresh field-caught adults harboured a statistically distinct population of microbes compared with those housed within a lab/aquarium habitat. These results suggest perhaps that the pigments may only fine-tune the types of microbes, and not function in a microbial-static function. They also do not distinguish between pigment forming a repressive habitat versus a conducive habitat, and they ignore the potential microbial interactions that likely occur on such structures, whose results may not be binary. Plus, we must keep in mind that although the L. variegatus colour-morph animals were collected from the same site, they may have been only distantly related and therefore offered distinct microbial environments and transcriptional profiles.

Figure 2.

Colour morphs have different echinochrome/spinochrome profiles. Colour morphs of Lytechinus variegatus and their pigmentation chemistry. Aboral photographs (left) and high-performance liquid chromatography profiles of spine pigments from green (first row), red (second row) and white (third row) colour morphs of Lytechinus variegatus, as well as from Strongylocentrotus purpuratus (fourth row) (Sp, spinochrome; Ech, echinochromes). A natural pigment experiment using Lytechinus variegatus (Lv): spines from Lv colour morphs have distinct microbial profiles as seen by 16S ribosomal DNA sequencing.

To test more specifically the effect that pigment may have on microbial composition, we manipulated pigment gene expression in sibling animals of Hemicentrotus pulcherrimus, a sea urchin species that has a long history in developmental biology, and where genome editing using CRISPR/Cas9 has been established [37,38]. In sibling animals, the genes encoding polyketide synthase (PKS), which synthesizes the base polyketide, and flavin-dependent monooxygenase (Fmo3), which appears to modify the base polyketide, were inactivated by Cas9 targeting. The spines resulting from PKS inactivation were albino, and Fmo3 knockout produced a shift in pigmentation from brownish-green to pastel purple. No other phenotypic differences were observed. 16S rDNA sequencing of microbes from the spines of these colour-morph siblings showed that the membership of the spine microbiome differed significantly between each genotype. Based on pairwise comparisons, the memberships of the PKS and Fmo3 knockouts were significantly different from the wild-type while being more similar to each other. These results argue that pigment is a key determinant of microbial colonization on sea urchin spines, and the goal is now to understand how.

4. Genetic machinery for pigment biosynthesis in echinoderms

A differential screen was devised in the sea urchin embryo to reveal genes involved in endomesoderm development [39]. Hatched blastula stage embryos lacking Delta/Notch signalling (by expression of a dominant-negative Notch construct) were compared with sibling embryos whose Wnt signalling pathway was activated by repression of GSK3β (by use of lithium chloride). Dominant-negative Notch signalling eliminates almost completely the formation of pigment cells and severely reduces the formation of blastocoelar cells, circumoesophageal muscles and coelomic pouch cells. Wnt signalling (LiCl treatment), on the other hand, results in significant expansion of the endomesoderm territory. Subtractive hybridization to enrich for mRNAs present selectively in the altered embryos, and hybridization of the non-subtracted mRNAs on macroarrays, resulted in an excess of 400 enriched mRNAs in the expanded endomesoderm/LiCl sample, which were identified by sequencing. Of those, sequences encoding metabolic enzymes, transcription factors, extracellular matrix proteins, and others, as well as orthologues of PKS, Fmo3 and a sulfotransferase, were found repeatedly [39]. This category of factors was tested by in situ hybridization, which showed selective expression in pigment cells. Further, use of morpholino antisense oligonucleotides (not shown) or Cas9/guide RNA (gRNA) to reduce expression of these target mRNAs resulted in albino animals with no pigment expression in the larvae (figure 3). This was a monumental discovery and the first time to our knowledge that PKS was identified and functionally proven in a metazoan.

Figure 3.

Polyketide synthase (PKS) is essential for making pigment in the pigmented immune cells (red spherule cells, (a), with inset magnified). Guide RNAs (gRNA) and Cas9 were injected into early embryos and the impact on pigment was assessed in early larvae. The pigment cells remained, but they made no pigment (b), with inset magnified. The transcription factor 'glial cells missing' (GCM) transcribes polyketide synthase (PKS) (c). (d–f) An embryo injected with Cas9/gRNA by brightfield (DIC), Cas9 immunolabelling in the nucleus, and DNA stain. The pigmented immune cells have an intense red colour, distinct from the spine pigmentation, and they are thought to be an immune type of pigmented cell. Similar cells bearing the same pigment are also seen patrolling the adult tube feet (g), and in the spines (h).

Subsequently, pigment biosynthesis was used as a simple, observational metric for optimizing CRISPR/Cas9 gene targeting in sea urchins. When an N-terminal exon of the PKS gene was targeted, the larvae developed perfectly well, but they were albino. This replicated the previous result of albinism in the larvae using morpholinos, and in both approaches, the larvae still made pigment cells based on other markers of this cell type, but those cells were no longer pigmented. The albino animals also still develop, swim and feed as normal and, aside from colour, the developing albino larvae are otherwise identical to the wild-type larvae. Thus, PKS and the polyketides resulting from its activity appear not to be involved in developmental signalling.

Since that discovery, efforts have been made to understand how PKS functions in pigment formation. The major products of pigment from a sea urchin are the echinochromes, spinochromes and related naphthoquinones. Radioactive labelling experiments indicated that echinochrome is a polyketide derived in part, at least, from acetate [40]. Biochemical evidence of PKS function, though, was lacking until very recent direct testing of this biosynthetic step.

Li et al. [41] undertook a direct approach to determine if PKS from a sea urchin was capable of synthesizing naphthoquinones and other aromatic polyketides in vitro. These investigators expressed the sea urchin PKS sequence in yeast, isolated the protein, and performed biosynthetic assays in vitro to test for mechanisms of product synthesis. Remarkably, sea urchin PKS was capable of forming a naphthalene precursor product to pigment (2-acetyl-1,3,6,8-tetrahydroxynaphthalene) when the enzyme was incubated with malonyl-CoA. Through analysis of the domain structure of PKS in the sea urchin, and by testing of PKS domains selectively, it was then found that the sea urchin SpPks1 is distinct from the bacterial, fungal and plant PKS gene functions (figure 4). The many PKSs in each kingdoms have mechanisms for polyketide synthesis, with a multitude of products (figure 5), but such function is distinct in the sea urchin based on domain membership. Indeed, instead of a giant seven-domain protein of approximately 3300 residues, truncation studies showed that only three of the SpPKS domains (ketoacylsynthase (KS), acyltransferase and the acyl carrier domains) were sufficient to make the pigment precursor (figure 4). We now know that SpPKS and its close relatives are a new class of animal carbocyclic aromatic PKS (aPKS) [41].

Figure 4.

Domain architecture of SpPks1 in comparison with other iterative polyketide synthases (PKSs) and fatty acid synthase (FAS) enzymes. HRPKS = highly reducing polyketide synthase; NRPKS, non-reducing PKS; KS = ketosynthase; AT = acyltransferase; DH = dehydratase; cMT = C-methyl-transferase; ER = enoylreductase; KR = ketoreductase; ACP = acyl carrier protein; TE = thioesterase; SAT = starter unit : ACP transacylase; PT = product template domain. The KS–AT–ACP domains are sufficient to yield naphthalene in vitro. (From [42]) (after [41]).

Figure 5.

Sampling of diverse metabolites made by polyketide synthase in various organisms.

By use of cryo-electron microscopy and crystal structure analysis we now know that the modular PKS architecture in animals is a functional homodimer, in which the two PKS polypeptides are aligned head-to-head and tail-to-tail, twisted upon each other to form a double-helix-like structure (figure 6). These studies confirmed the dimeric nature of the PKS, and showed that, at least in isolation, the structures of many of the PKS domains or didomains closely resemble their counterparts in animal fatty acid synthase (FAS), with the extra helical feature of dimerization in PKS [42–45]. The family of PFS/FAS enzymes and their domain-swapping evolution are powerful targets for repurposing biosynthetic activities. The identification of products from the enzymes, and the structures used to accomplish these tasks, will undoubtably be fodder for future creativity.

Figure 6.

Atomic structure of polyketide synthase (PKS) (a) and fatty acid synthase (FAS) (b) by cryo-electron microscopy (cryo-EM) and X-ray crystallography respectively. Domain labels as documented for figure 4.

PKS appears to be essential for making the core polyketide, whereas FMOs, also found in the original sea urchin larval screen, appear to be essential for modifications to the core polyketide. Flavin-dependent monooxygenases (FMOs) compose a large family of monoamine oxidases. FMO is an excellent multifunctional oxidoreductase, which can catalyse the formation of different types of C–O bonds, including hydroxylation, Baeyer–Villiger oxidation, sulfur oxidation, epoxidation, and other reactions with high chemo-, regio- and stereo-selectivity. Compared with the CYP450 family, FMO has unique oxidation advantages such as producing fewer toxic substances and being self-sufficient in the process of catalysis with higher efficiency [46].

5. Does pigment impact microbial load?

To test the hypothesis directly that pigment influences the microbial load on sea urchin spines, the genomic loci of PKS and Fmo3 were inactivated by single guide RNA (sgRNA)-directed Cas9 activity. Mutations of these genes resulted in no gross morphological difference in development, metamorphosis or adult development—except for their colour, all other features of the animals were normal (figure 7). When microbial diversity was assessed, a clear distinction was seen between siblings that differed only in their pigment (figure 7). Thus, we see a direct correlation between microbial populations and the naphthoquinone pigment in the adults. Editing the host genome resulted in taxonomically convergent shifts in the spine microbiome, and, by this means, we identified two host factors in vivo that influence which microbes associate with sea urchin spines [36]. This is the only instance that we are aware of where host genome editing via CRISPR/Cas9 resulted in a change in the microbiome.

Figure 7.

Polyketide synthase (PKS) and flavin-dependent monooxygenase (Fmo3) are necessary for pigment formation in the adult sea urchin Hemicentrotus pulcherrimus. Spine pigment alters microbial colonization in the spines of adult H. pulcherrimus. WT, wild-type. (After [36].)

Remarkably, the albino adults that formed as a result of PKS inactivation succumbed early. Their longevity was compromised either by an ageing issue without pigment or by a vulnerability to pathogens. The full-grown adults (1 year 8 months) of PKS knockout all died within two weeks of each other, whereas their sibling controls, and the Fmo3 knockout animals, lived well beyond 3 years of age. This same phenomenon of early demise was seen in two different species in two different labs under controlled conditions [38,47]. It should be noted that the Fmo3 knockout animals still had echinochrome in the pigmented coelomocytes, such that control and Fmo3 knockout had similar immune cells, but different colours of spines. This change in only the spine pigment also correlated with a change in microbial populations. Currently no technology allows us to inactivate the PKS only in the spines or test, and instead, it inactivates pigment synthesis in all cell types. Thus, we cannot determine if loss of the spine pigment is responsible for this demise, or if it is the immune-type pigment cells with their deep red echinochrome A pigment that are responsible for this compromised health. One approach to test this may be to transplant coelomocytes from a wild-type animal to a PKS-albino sibling coelom that contains all the usual cell types, except in a non-pigmented form. We might hypothesize that such transplantations, were they to contain mitotic cells, could rescue the albino early death phenotype, whatever its cause.

Since polyketides are generally cytotoxic, the echinochrome in the sea urchin larval pigment cells and red spherule cells of the coelom must be either conjugated with some protein or other compound(s) to neutralize the toxicity, or compartmentalized promptly upon biosynthesis; otherwise the toxicity would presumably compromise the maker. Alternatively, the PKS may be sequestered into vesicles that protect the cell synthesizing the toxic compound. A subcellular localization of PKS (anti-PKS antibodies) to identify the location of the enzyme by e.g. immune electron microscopy may help answer this question. Perhaps PKS is in the cytoplasm and the product is targeted for vesiculated sequestration, or the PKS is in such vesicles, making the naphthoquinone in a protected environment. This strategy may place exceptional localization and/or regulatory constraints on PKS in its synthesis of a toxic agent. Are PKS and FMO always found together? Is ABCg11, another gene product enriched in the immune-type pigment cells [39,48], in the membrane of an echinochrome vesicle pumping precursor into the vesicle? Or pumping the completed product (echinochrome) into the vesicle to protect the cell? If these products are in vesicles, what do we call them—colour vesicle, chrome vesicle, chrome body, chromosome (!)? We preferred the last term though hesitated to use it since dibs have been called on it already. Instead, since we know the products to contain naphthoquinones, and the cells that make them as quinocytes, we will refer to these vesicles as quinosomes. This premise is analogous to how melanin is regulated. Melanosomes are lysosome-related organelles in which melanins are synthesized and stored and the melanosomes function in secretion and phagocytosis. Reductions in the number, structure, and/or function of melanosomes in melanocytes leads to albinism.

6. Future work

• 1. Microbial colonization is distinct in different colour morphs of two species of sea urchins. Recent single-cell RNA-seq work in sea urchins and sea cucumbers now enables a comparison of both what pigment cell transcriptomes look like and what microbial dynamics might be seen between taxa of echinoderms. Both sea urchins and sea cucumbers have great diversity in body colours and are widely distributed from tropical to frigid regions. Compared with sea urchins, preliminary results suggest that sea cucumbers have a richer variety of microbes attached to their body surface. We assume, reflected by the article title, that the microbes may sense the chemistry of the pigment, but not the colour of the pigment. Some marine microbes, however, such as certain halophiles, can recognize different colours of light [49]. Perhaps these microbes can select hosts based on the different colours of light absorbed and reflected by the pigmented surfaces of marine animals and thereby use that energy for survival or growth.
• 2. PKS appears to synthesize the core polyketide, but different FMOs appear to modify them for colour variation. FMOs share special protein structures and unique mechanisms for catalysing substrate oxidation. FMOs stabilize the internal C4a-hydroperoxide flavin intermediate at special sites, and they catalyse the synthesis of a variety of substances, including natural products, through a unique catalytic mechanism. The reactions involved in the synthesis of natural products catalysed by FMOs include hydroxylation, epoxidation, Baeyer–Villiger oxidation, oxidative decarboxylation, halogenation, and sulfur oxidation. Currently, only one FMO gene (Fmo3) in sea urchins has been knocked out for verification testing this premise, so other FMOs (the sea urchin has about a dozen) are important to test for comparison. We can follow up with research work such as: construction of FMO expression vectors; isolation and full characterization for extracted compounds; absolute stereochemistry analysis.
• 3. We have focused here on naphthoquinones, but echinoderms have other pigments defined by classic biochemistry [11,12]. Now with many genomes sequenced in echinoderms it is important to identify their cognate biosynthetic machinery, and test functions of those diverse pigments with a focus on microbial interactions. We note even a report of fluorescent pigments [50] present in echinoderms, which would be of particular interest in engineering reporters for research purposes.
• 4. Sea urchin spine cultures—echinoderms are slow in adult growth but by using spines in culture perhaps one can begin testing function without having to go through a life cycle of disrupting gene function in embryos, and then culture to adulthood. Perhaps spine cultures in vitro could be used as a proxy for function in adults and one may have more direct control over microbial interactions in dynamic ranges.
• 5. What is the chemistry of pigments that cause anti-microbial activity? We should start from the specific regulatory mechanisms by which pigments affect microbes. For example, pigments with flavonoid structure have more phenolic hydroxyl groups in their molecules, and these functional groups bind to proteins or enzymes by hydrogen bonding, destroying the proteins' molecular structure and denaturing or inactivating them, leading to cytoplasmic solidification, and thus disintegration and death of microbes.
• 6. Since bacteria have PKS activities also—what features of the animal PKS contribute uniquely? How do bacteria retain the naphthoquinones and not die? (A) Microbes coordinate through the overlap of PKS genes with their resistance genes. For example, this mutual regulation exists between the biosynthesis of antibiotics such as erythromycin, neomycin and streptomycin and the expression of resistance genes. (B) The target of naphthoquinone in bacteria is modified. For example, erythromycin-producing and streptozotocin–sulfur-producing bacteria whose ribosomal RNAs were modified by methylation exhibited insensitivity to these two antibiotics that act on rRNA. (C) The synthesized naphthoquinone exists in an inactive condition in the cell after being subjected to its own modifications (e.g. acetylation) or being an inactive structure. When naphthoquinone is secreted outside the cell, it is de-modified, catalysed by certain membrane proteins, making it an active structure outside the cell.
• 7. Recent reviews give a comprehensive summary of how pigment might be used clinically [51,52]. Combining the algalstatic repurposing by N-20, and the recent in vitro synthesis of polyketides by PKS, the future looks colourful, especially for various medical applications which may be coupled, or not, to the anti-microbial theme.

Data accessibility

This article has no additional data.

Declaration of AI use

We have not used AI-assisted technologies in creating this article.

Authors' contributions

G.M.W.: conceptualization, data curation, funding acquisition, project administration, supervision, writing—original draft, writing—review and editing; L.X.: conceptualization, formal analysis, resources, validation, writing—original draft, writing—review and editing; N.O.: formal analysis, software, writing—review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed herein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

The authors thank the National Institutes of Health (grant no. 1R35GM140897) and the National Science Foundation (grant no. IOS-1923445) for support of this work.