Cryptic coloration of Macaranga bancana seedlings: A unique strategy for a pioneer species

ABSTRACT

Macaranga bancana is considered as a successful pioneer plant species. Usually found in disturbed and open areas, most of the current research focused on its relations with ants. One of the unique feature of the plants is that the seedling leaves are red, resembling and almost matching the background. Using a portable spectrometer, we measured the color reflectance of M. bancana seedlings (less than 20 cm in height). We also measured the leaf litter reflectance, adult M. bancana leaves and also seedlings of several other species found in the vicinity of M. bancana seedlings. The reflectances of M. bancana seedlings are very similar to that of the leaf litter background. We suggest that this cryptic coloration is crucial during the early stages of the plant when it still cannot rely on the protection of ants.

KEYWORDS:

• Anthocyanins
• cryptic colors
• leaf color
• Macaranga bancana
• myrmecophyte
• red leaf color
• reflectance analysis

Plants are sessile organisms that rely on multiple signaling systems. Plants rely on both physical and chemical signals either to attract pollinators and seed dispersers, or to defend from grazing herbivores. Plants also use cryptic coloration as camouflage from grazing herbivores.^1-3 There are also plants that use aposematic coloration as a warning signal.^4,5

There are a few species of plants that uses different color strategies throughout their ontogeny. For example, Pseudopanax crassifolius adapts cryptic colorations during the seedling stage and changes to aposematic colors during the juvenile stage.⁴ Another unique example is the Elaeocarpus hookerianus, a heteroblastic plant that has varied leaf shapes and brown mottled colors mimicking the leaf litter.¹ La Rocca et al.⁶ reported the dual functioned colors of Erythronium dens-canis. This amazing understory Lilly has complex leaf pattern to remain cryptic to herbivores, while certain portions of the reflective colors function as attractants for pollinators. Corydalis benecincta, an alpine plant with dimorphic colors (green and gray), in which the gray colored variety had fewer predation effect while still retaining a similar photosynthetic performance and visual cues to attract pollinators with the green colored variety.³

These visual defenses (either to remain cryptic or aposematic) are plant and environment specific. Most of the recorded plants originate from the temperate climate and in some cases, are evolutionary linked with the environment or to a certain animal taxon. There is a scarcity of examples of such visual defenses from the tropics. They include being red/brown to look dead,⁷ delayed greening,⁸ red leaf undersides that defend from fungal attacks,⁹ different adaxial and abaxial leaf colors that undermine herbivorous insect camouflage¹⁰ and aposematism.¹¹

Macaranga bancana (Miq.) Müll.Arg. of the Euphorbiaceae is a common tree species in the South East Asia region, especially in Malaysia. Previously, the species was commonly mistaken as M. triloba, which has an almost similar appearance except for lacking hollow twigs. The seedlings are often described as reddish purple in color with prominent glands lining leaf margin.

Considered as a pioneer species, M. bancana are usually found along roads and recently disturbed areas.¹² The species is also well known as an example of myrmecophytic relation with ants. The hollow space within the twigs provides excellent shelter for ants and in return, the ants provide protection to the plants. Although there had been numerous studies on the myrmecophyte relation with ants,^13-15 there is almost no information on other types of adaptive strategies employed by M. bancana.

Here we suggest that M. bancana utilizes different adaptive tactics throughout its ontogeny. A shade intolerant species, with a fast sprouting and growth, M. bancana has been known to thrive and occupy disturbed habitats and open areas with access to greater sunlight. Using spectral analysis, we measured the color reflectance of adult and seedlings of M. bancana. We also measured the spectral reflectance of the background, and of other plant seedlings growing in the vicinity of the M. bancana seedlings. We hypothesize that the M. bancana seedlings would have a similar color reflectance to the background and to be distinctive compared to adult M. bancana plants and to seedlings of other plant taxa.

Methods

The common mahang (Macaranga bancana) is a small to medium sized tree that grows up to 20 to 23 m high (Fig. 1a). The bark is pale-brown to whitish, mottled. The large leaves are characteristically 3-lobed (Fig. 1b). At the base of each leaf stalk is a pair of dark purple stipules, and many white starch grains are found on their abaxial surface. Young leaves are reddish-brown, densely covered by trichomes (Fig. 2). It is common in Peninsular Thailand, Peninsular Malaysia, Sumatra and Borneo. Macaranga bancana can be found in a broad range of habitats. The species can thrive on both clay and sandy soils, although the preference is toward the latter due to the water retention in clay. As such, M. bancana can be found in places where non-pioneer species could not thrive, such as wet gullies along roadsides, alluvial forest areas, and the margins of degraded peat swamp forests.

Figure 1 (a) Adult M. bancana.(b) Close up of a normal adult Macaranga Bancana. It is easily identified based on its trilobe shape.

Figure 2. M. bancana seedling. There is evidence of insect herbivory at some of the seedlings based on the damages on the young leaves. The seedlings, however are surrounded by other syntopic green seedlings that stands out among the background.

We gathered the M. bancana samples from the Gunung Ledang National Park, Johore, Malaysia (N2 20.802 E102 38.181). Spectral measurements follow protocols suggested by Fadzly et al.⁴. We took spectral readings from 30 individual M. bancana seedlings (less than 20 cm in height). We also took spectral readings from 30 adult M. bancana (more than 20 cm in height). These include both saplings and full grown adult trees. The spectral measurements for both groups were similar, therefore we grouped them together. The background reflectance was also measured. We took the average of 15 spectral measurements of the background which consists mostly of dead leaves, woody debris, soil and rocks. We sorted the spectral results for the background into 2 categories, which were background soil and background dry leaves. The categories infer to the main component. For comparison purposes, we took the spectral measurements of other types of seedlings growing nearby the M. bancana seedlings. We took 30 individual measurements of the other species seedlings (less than 20 cm in height). However, most of the seedlings were unidentifiable. Although we could not clearly identify each of the seedlings, their general appearance and coloration suggests that they are very different from the M. bancana seedlings.

Spectral measurements were conducted using OceanOptics Jazz Portable Spectrophotometer with Tungsten lamp light source. Reflectance was measured as the proportion if a diffuse reflectance standard (Teflon coated white standard, OceanOptics). The distance between each object and the probe was fixed at 1 cm, with the angle of illumination and reflection fixed at 45° to reduce glare. Irradiance was measured using cosine corrected sensor and D65 (normal daylight) bulb as reference. Spectra were calculated at 5-nm intervals from 300 to 700 nm with SpectraSuite software (OceanOptics).

There are multiple options in analyzing the reflectance properties. Endler and Mielke¹⁶ suggested using the simple color pattern measures related to photon capture of a known, specific receptor spectral sensitivity of the observer. Fadzly et al.⁴ based their calculations on the eye receptor sensitivities of ostriches, as the closest proxy of the extinct Moa birds (the targeted herbivore in the study). Niu et. al³ uses butterflies (as potential pollinators) to calculate discriminability among the 2 variants of Corydalis benecincta. However for this study, we analyzed the reflectance based on a slightly modified protocol described by Klooster et al.² The reason we did not choose a specific targeted observer, is because M. bancana is a pioneer species found in an open/disturbed areas in a tropical rainforest. In a tropical rain forest, there are various types of herbivores ranging from ungulates to small insects. There are no specific (monophagus) herbivores that exclusively target M. bancana. Principal Component Analysis has an advantage in this study because the method considers both hue and chroma in a simple graphical summary (see Klooster et al.²). There might be differences based on different types of animal eye receptor sensitivities, but for this analysis, PCA transformation is sufficient to validate our results. A PCA analysis was performed with JMP 10 software package using raw spectral values sampled at 5 nm increments from the M. bancana seedlings, other species seedlings, adult M. bancana and the background (both soil and leaves).

Results

Fig. 3 shows the reflectance analysis results. Based on the curves, Adult M. bancana and seedlings of other species have an almost similar reflected spectrum. Both showed a peak at around 550 nm, which is green in the visible light. The adult leaves give a higher reflected light intensity compared to that of the seedlings. The light reflected from M. bancana seedlings does not show a specific peak, and shows an increase from 500 nm to 650 nm, a combination between green-red in the visible color spectrum. For the background spectrum, soil is much darker/ lower intensity and shows a range of wavelengths ranging from 400 nm to 700 nm. Dry leaves also show similar colors, with a much higher intensity compared to the other variables. Although the spectrum of the background does suggest the appearance of blue, this is attributed to small stones and clay soil that are mixed in the background. The spectral analysis alone does show how the M. bancana seedlings are remarkably visually different from the adult stage and also from the other species seedlings coloration. The seedling's spectral reflectance closely resembles the soil background.

Figure 3. Spectral reflectance results. The adult Macaranga bancana and other seedlings shows similar pattern, whereas the seedlings Macaranga bancana are almost similar to the background.

Figure 4. Principal component analysis results. PC1 and PC2 accounts for 93.8 % of the information.

As an additional analysis, our PCA analysis created 5 eigenvalues, with the first 2 factors account for 93.8 % of the variation. PC1 independently accounted for the majority of the variation at 67 % with PC2 representing the next highest value at 26.8 %. Table 1 shows the eigenvector values for PC1 and PC2. Other seedlings and M. bancana adults are clustered independently of each other in the ordination space. In contrast M. bancana seedlings, background soil and background dry leaves are clustered together within the same region (Fig. 4).

Table 1. Eigenvectors For PCA analysis.

	Principal component 1	Principal component 2
M. bancana Adult	0.46499	0.42675
M. bancana Seedling	0.51955	−0.07344
Other Seedlings	0.08106	0.84118
Background soil	0.50250	−0.19450
Background dry leaves	0.50475	−0.25900

Discussion

Despite the research attention that had been given M. bancana adults, very little is known about the seedlings. Experienced park rangers at the sampling site had acknowledged, that even though they could recognize an adult M. bancana, they have never bothered to inspect and examine the seedlings. When asked to identify the seedlings, the rangers failed to locate them. After further information and an example picture were given, only then the rangers managed to spot the seedlings.

Based on our spectral results, the reflectance curve shows that the adults and other seedling colors are similar, whereas the M. bancana seedlings are more similar to the background color. The average spectral properties of both adults M. bancana and other seedlings are also ordinary, that is mostly within the standard error of the leaves of syntopic species. Their colors are consistent with the reflectance properties of the primary pigments involved in photosynthesis (chlorophyll a and b). These two pigments have peak absorption values above and below the peak in reflectance of adult leaves at 540–550 nm. Further PCA results also support our hypothesis that the M. bancana seedlings are more similar to the background color (both soil and fallen leaves).

There are multiple hypotheses of why leaves are red colored. The red coloration could be attributed to the presence of anthocyanin pigments. These pigments produce colors ranging from brown to red or in some cases as a trade-off between the amount of chlorophyll in the leaves. The most common hypothesis is protection against photoinhibition. Excessive bright light condition could reduce photosynthesis efficiency and could possibly lead permanent damage to the chloroplasts, cells and tissues.¹⁷ However, Krause et al.¹⁸ argued that for tropical plants (young leaves), the main component against photoinhibition are xanthophylls, α-carotene, and zeaxanthin. This argument is supported by Dominy et al.,¹⁹ in stating that if anthocyanins are involved in photoprotection, there should be more tropical trees with red leaves at the canopy level, compared to the understory.

Another hypothesis that might explain red colorations of young leaves in the tropics is that red color makes the leaves cryptic to insect herbivores.¹⁹ Most of the insect herbivores lack the spectral sensitivity in the red region (600– 700 nm wavelength).^19,20 Tellez et al.²¹ suggests that the anthocyanins protect the young leaves from fungal damage. The amount of anthocyanins in the leaves could also produce a bitter, inedible taste that could possibly repel grazing herbivores.²²

The cryptic red colors of M. bancana seedlings and its subsequent change during the adult stage resembles the Pseudopanax crassifolius and Elaocarpus hookerianus that changes throughout its ontogeny.^1,4 It can be speculated that M. bancana changes its cryptic coloration after development of the hollow stem and food bodies that will attract ants. The ants in turn will provide defense for the plants from browsing herbivores. This cryptic coloration is quite peculiar and undocumented strategy for M. bancana. As a pioneer species, fast resprouting and growth would be an advantage.

There is a caveat of having red coloration, Archetti et al.²³ argued that red colored leaves is counterproductive as it needs extra costs of pigment synthesis, resource loss and decreases the photosynthesis rate. Therefore, the red leaves are thought of as a handicap signal to deter colonizing insect herbivores. It must be noted that this hypothesis is a based on temperate trees, rather than tropical. A more recent study by Chen and Huang,²⁴ suggests that through phylogenetic analysis, young red leaves have less mechanical protection (in the form of enhanced cuticle, epidermis and trichomes) compared to the green young leaves. Menzies et al²⁵ demonstrated that the red leaves of the Pseudowintera colorata, has a higher concentration of antifeedant chemical components, incurring less herbivory damage and reduced leaf eating infestation from insects. Hughes and Lev-Yadun ²⁶ discusses the numerous potential ecological and physiological of having red colored leaf margin. Various existing hypothesis supports one another, depending on the environment or the locality of the species. For further example, delayed leaf greening is a useful trait for tropical trees. A delayed greening or red-colored young leaf strategy in seedlings have been linked to higher survival and growth rate further influencing the species composition in later stages.²⁷

For the M. bancana, the cryptic coloration serves it's purpose based on the species ecological niche. As a pioneer species, the seedlings are quite unprotected in an open area. Being cryptic definitely increases the survival rate, as browsing herbivores would be more inclined toward a conspicuous green foliage. In a tropical forest and disturbed areas, there are numerous herbivores, from deer, cows and to the smallest leaf cutter insects. We do note that there was evidence of insect herbivory for some of the M. bancana seedlings sample, however, it was in a smaller amount and limited to a certain ant infested treefall gap area. Whether the red colors work to deter insect herbivory for M. bancana seedlings still remain open to experimentation and discussion. As far as for humans' eye vision, without prior information and a trained eye, the seedlings remain quite inconspicuous. We would like to suggest that the cryptic coloration of M. bancana seedlings as an additional useful defensive strategy for a pioneering plant species.

Disclosure of potential conflict of interest

The authors would like to declare that we have no conflict of interest and financial obligations concerning the publication of this data and manuscript. The authors also adhere to the code and conduct of scientific publication.

Acknowledgments

The authors would like to thank the Perbadanan Taman Negara Johor and Gunung Ledang National Park rangers for their help and guidance. We would like to thank Prof Simcha Lev-yadun and another anonymous reviewer for their comments in improving the manuscripts. We would also like to thank Nor Amira Rahman for helping out with the data analysis.

Funding

This study is supported by RU USM Grant 1001/PBIO/811329.