Background:

Butterflies are well known for the variation of their wing patterns which can be used for a variety of ecological functions, such as aposematic warning signals (think bright colors and high-contrast patterns) or conspecific mating cues [1, 2]. Decades of research on these wing patterns has found that a few key genetic loci control the variations that are sometimes observed within populations. In most cases, these studies found genes encoding developmental regulatory factors such as the WntA signaling ligand, optix, doublesex, aristaless1, and bar-h1 transcription factors [3, 4]. Transcriptional profiling followed by functional validation using CRISPR knock-outs confirmed that many of these genes are involved in the development of spatially restricted patterns and colors.

The cortex locus, in contrast, has been more intriguing. It has been repeatedly associated with cases of phenotypic adaptation in butterfly color patterns, even in the illustrious peppered moth [3, 5]. But, despite the contributions of several studies that have looked further into cortex function and evolution, a few puzzling issues remain. For example, previous work has shown that cortex is expressed in all epithelial and scale cell precursors during pupal development, with seemingly no correlation to adult wing color patterns [6]. Furthermore, the cortex protein is known to play a role in meiosis in flies, a function that is hard to reconcile with a cell identity determinant role that would explain a color switching effect in butterflies. In the series of preprints highlighted here, the authors show that it is actually an adjacent and previously undetected set of non-coding RNAs that is responsible for wing color patterning–not cortex. These transcripts, the ivory long non-coding RNA (lncRNA) and a micro-RNA dubbed mir-193, are probably co-transcribed and are both required for the activation of melanic color scales. Together, these researchers show just how much we still have to learn from what has thus far been an overlooked genetic modulator.

Main Findings of Livraghi et al., 2024:

In this study, the authors began by re-analyzing sequencing data and revealing the presence of the lncRNA ivory. They also identified its first exon and its location immediately 3’ of a differentially accessible region associated with pupal wing tissue, supporting its role as a potential promoter. To probe possible functional roles for ivory, the authors targeted this first exon using CRISPR-Cas9 mutagenesis in three different butterflies: two species of Heliconius (H. erato and H. charithonia) as well as Vanessa cardui, colloquially known as the Painted Lady. These mosaic knockouts resulted in scale transformation from melanic, or dark, states to yellow-white states. The mutant phenotype was highly penetrant and did not seem to affect epithelial scales in other parts of the adult wing or body, suggesting that ivory has low pleiotropic effects. In situ hybridization experiments against ivory’s first exon also showed that ivory lncRNA is expressed in melanic wing scales during pupal wing development, further supporting ivory as the causative agent in wing pattern divergence.

While these experiments showed that expression of ivory positively regulates melanic scale differentiation during pupal wing development, the authors also wanted to test the function of cortex, especially given the fact that previous mosaic knockouts generated rare phenotypes. They generated V. cardui null mutants targeting the second exon of cortex–an obligatory exon with conserved elements–and found that these mutants were indistinguishable from wild-type controls. Two confirmed G0 mutants were crossed to produce F1 individuals, which also showed no pigmentation changes. These experiments suggest that ivory does not have a trans-regulatory effect on neighboring cortex and that its promoter does not act in cis to regulate cortex, either. However, analysis of previous Hi-C and histone ChIP-seq data from H. erato pupal wings, as well as histone modification profiling done in pupal wings, did reveal that ivory is likely acting in trans and is not involved in cis-regulation of nearby genes.

The authors next assessed the conservation of the ivory promoter and first exon, finding that they are both conserved in ditrysian Lepidoptera. Subsequent CRISPR-Cas9 targeting of ivory in two nymphalid butterflies, Agraulis incarnata and Danaus plexippus, showed the transformation of melanic scales to yellow-white scales. Interestingly, in A. incarnata, these transformed scales also underwent a change in their ultrastructure and some melanic spots showed a complete conversion to white as opposed to a lighter intensity. Similarly, melanic scales in D. plexippus turned white and orange scales became lighter in appearance. However, some distal white spots became orange and more distal orange areas became white, hinting at local differences in ivory function as well.

Main Findings of Fandino et al., 2024:

Following up on van der Burg et al. 2020 [7]—a study that identified the genes most likely to be associated with differences in color plasticity in the buckeye butterfly, Junonia coenia—the authors of this study set out to further explore the genetic locus underlying seasonal polymorphism. They began by identifying ivory’s first exon, specifically its 5’ end. They set out to identify cis-regulatory elements (CREs) of interest by both generating new ChIP-seq data and analyzing previous datasets of mutants associated with wing depigmentation, identifying three transcription factors of interest: Spineless, Fushi-tarazu-F1 (Fitz-F1), and Bric-a-brac (Bab). All three showed strong binding signals at the ivory promoter, with Spineless showing a strong cluster of binding sites in ivory’s intron, a site that has been previously associated with seasonal polyphenism in J. coenia. Further analysis using ATAC-seq datasets of other nymphalid butterflies also showed that ivory’s first exon and multiple CREs are conserved, suggesting an established role for ivoryin pupal wing development.

The authors next set out to assess ivory expression in pupal wings by using hybridization chain reaction (HCR) in situ hybridization. The authors found high expression of ivory lncRNAs in specific elements, particularly the eyespot and in the outer boundaries of the discal bands. These patterns were consistent with the in situs shown by Livraghi et al., once again suggesting a role for ivory in wing color patterning. Through these experiments, the authors found ivory to be largely represented in melanic patterns. In order to better understand ivory’s functional role in color pattern development, the authors used CRISPR/Cas9 to generate two mutant deletion lines: one targeting the ivory promoter and another targeting the promoter and first exon. The mutants with the promoter-only deletion showed an overall faded phenotype but appeared to retain most of the pigment types. The mutants with the combined promoter and exon deletion, however, showed an almost completely pigmentless phenotype resembling the original H. melpomene ivory mutant. This finding emphasizes the importance of the ivory promoter as an important hub for nuanced effects on coloration.

To understand how ivory cis-regulatory regions may be involved in J. coenia’s polyphenic variation, the authors decided to generate “shotgun deletions” in five presumptive ivory regulatory regions using CRISPR/Cas9. This approach generated a spectrum of mutations of different lengths, allowing the authors to assess the variety of effects that regulatory mutations can cause. Three main phenotypes resulted from this experiment: a phenotype replicating different seasonal color patterns, strongly hypomorphic mutants resembling the promoter and first exon deletion phenotype, and pigmentation hypermorphs that showed a darkening of pattern elements. Taking these results together, the authors determined that enhancers within the ivory locus govern the red-tan polyphenism of J. coenia by regulating ivory’s transcription or dosage.

Because of ivory’s effects on the regulation of multiple pigment types, the authors assumed that lncRNAs are likely involved in the regulation of multiple biological processes. To look at this, they returned to their ivory promoter and first exon knock-out mutants and characterized the wing transcriptome at three different stages of development. The authors found that a number of pigmentation genes were downregulated in knockouts. Interestingly, some pigment-related genes were actually upregulated in mutants, suggesting a more nuanced relationship with their functional roles in wing pigmentation. The authors also identified targets with no known color patterning role—namely, the Osiris family of genes—which present intriguing candidates for future functional analysis. Overall, then, loss of ivory led to broad transcriptional effects that extend to genes in other pigment pathways, showing a more global effect of the mutation. Taken together, these data begin to piece together a mechanism by which multiple transcriptional pathways are integrated to provide the genetic cues that govern an environmentally plastic response.

Main Findings of Tian et al., 2024:

The authors of this study began by screening the cortex locus and subsequently honed in on two deeply conserved micro-RNAs (mi-RNAs) located in the intergenic region between cortex and adjacent parn, mir-193 and mir-2788. To assess the function of these mi-RNAs, the authors performed CRISPR-Cas9 knockouts in Bicyclus anynana embryos. For mir-193, 74-83% of the mosaic knockout individuals showed reduced melanin levels across the body of the butterfly. Variation in wing pigmentation was also observed between both individuals and sexes. Additionally, many of these mutants could not fly. Conversely, only 0-8% of the mir-2788 knockouts showed a color phenotype and no flight defects were observed.

The authors determined that the various phenotypes of the mir-193 mutants were likely caused by the mutant alleles around the miRNA locus. Sequencing of mir-193 mutants revealed 1-3 major mutant alleles and a near 100% mutation rate. In the case of the mir-2788 mutants, however, they determined that the color phenotypes were not caused by disruption of mir-2788 but of a flanking locus. To further confirm this, the authors generated homozygous mutant lines with short deletions for both miRNAs. Four mutant lines were generated for mir-193, three of which showed a light brown wing color, white eyespots, and flight defects. Heterozygotes of all four lines showed wild-type patterning and flight phenotypes, indicating recessive alleles. One mutant line was generated for mir-2788. Neither homozygotes nor heterozygotes showed a color or patterning phenotype despite depletion of the mature strand as confirmed by qPCR. This led the authors to conclude that mir-193, not mir-2788, promotes wing color in B. anynana.

Next, the authors wanted to study the primary transcripts of these miRNAs, which would in turn allow them to elucidate their transcriptional control and expression patterns. Because miRNAs in proximity are usually co-transcribed, the authors assumed that mir-193 and mir-2788 were likely being processed from the same primary transcript. Using RNA-seq data from both wild-type and mutant homozygotes, they determined that this transcript was a lncRNA and subsequently found that the core promoter region displayed clear sequence homology, suggesting conservation across Lepidoptera.

Having isolated mir-193 as a promoter of wing color, the authors next set out to find the direct and indirect targets of this miRNA. Examination of the mir-193 mutant and wild-type transcriptomes at different stages revealed varying numbers of differentially expressed genes. None of the genes within the cortex locus were differentially expressed, suggesting that mir-193 acts in trans. Because direct targets of the miRNA should be overexpressed in the mutants, all significantly upregulated genes were pooled and assessed for complementary binding sites to the guide strand of mir-193. One of these targets, ebony, is a well-known insect melanin pathway gene responsible for converting dopamine to a light yellow pigment known as NBAD. Their analysis indicated that ebony had two mir-193 binding sites within it. This led the authors to hypothesize that disruption of mir-193 in the dark wing regions of their mutants allows for de-repression of ebony, which shunts dopamine away from melanin pigment production and results in lighter wing phenotypes.

Why I Chose These Preprints:

While I originally intended to cover the Livraghi et al. preprint by itself, it quickly became clear to me that this story would be incomplete if I didn’t include the work of Fandino et al. and Tian et al. in this preLight as well. Together, these three studies present an incredibly thorough and intriguing case study on the role of ivory and its miRNA mir-193 and what we might be missing by overlooking genetic modulators such as lncRNAs. While there have been a few examples of non-coding RNAs playing a role in color patterning and adaptation–such as Liang et al. 2023 [8], which showed how the YUPgenetic locus produces small interfering RNAs (siRNAS) that regulate floral carotenoid pigmentation–it is still an interesting phenomenon. Given that there are many non-coding RNAs expressed in genomes that we currently know nothing about, these studies make the case that we can ignore them no longer.

However, while the research more than speaks for itself, what I really enjoyed about these studies is their use of collaboration. Science can sometimes be a hypercompetitive environment that encourages individual achievement, but these researchers instead chose to work together to uncover what was really going on with ivory and cortex. The work they have produced is admirable, and I hope it encourages us all to continually challenge our beliefs and our data–and when we can’t figure out what’s going on, to put our heads together to figure it out.

References:

[1] Merrill, R. M., Rastas, P., Martin, S. H., Melo, M. C., Barker, S., Davey, J., et al. Genetic dissection of assortative mating behavior. PLoS Biol. 17:e2005902 (2019). doi: 10.1371/journal.pbio.2005902

[2] Finkbeiner, S. D., Briscoe, A. D., Reed, R.D. Warning signals are seductive: Relative contributions of color and pattern to predator avoidance and mate attraction in Heliconius butterflies, Evolution, 68, 12:3410–3420 (2014). https://doi.org/10.1111/evo.12524

[3] McMillan, W.O., Livraghi, L., Concha, C., Hanly J.J. From Patterning Genes to Process: Unraveling the Gene Regulatory Networks That Pattern Heliconius Wings. Front. Ecol. Evol. 8:221 (2020). doi: 10.3389/fevo.2020.00221

[4] Bayala, E.X., et al. aristaless1 has a dual role in appendage formation and wing color specification during butterfly development. BMC Biol 21, 100 (2023). https://doi.org/10.1186/s12915-023-01601-6

[5] Nadeau, N. J., Pardo-Diaz, C., Whibley, A., Supple, M. A., Saenko, S. V., Wallbank, R. W. R., et al. The gene cortex controls mimicry and crypsis in butterflies and moths. Nature, 534, 106–110 (2016). doi: 10.1038/nature17961

[6] Livraghi, L., et al., Cortex cis-regulatory switches establish scale colour identity and pattern diversity in Heliconius. eLife 10, e68549 (2021). https://doi.org/10.7554/eLife.68549

[7] Karin R. L. van der Burg et al., Genomic architecture of a genetically assimilated seasonal color pattern. Science 370, 721-725 (2020). doi:10.1126/science.aaz3017

[8] Mei Liang et al., Taxon-specific, phased siRNAs underlie a speciation locus in monkeyflowers. Science 379, 576-582 (2023). doi:10.1126/science.adf1323

Tags: butterfly, color patterning, lncrna, mirna, non coding rna

doi: https://doi.org/10.1242/prelights.36968

(No Ratings Yet)