Iridescent Blue Plants

David W Lee. American Scientist. Volume 85, Issue 1. Jan/Feb 1997.

My first experience of the Malaysian tropical rain forest changed the direction of my scientific career. I had just taken up my first university position, as Lecturer in Experimental Taxonomy at the University of Malaya in Kuala Lumpur, when I was led on a walk of exploration in the forested mountains outside of the city. I was partly prepared for the bewildering diversity, the majesty of lofty trees festooned with epiphytes and lianes, the iridescent green-on-velvet-black wings of the Rajah Brooke birdwing butterfly-I had read about all that. But I was astonished by the leaf color of understory plants, especially the shimmering electric blue of Selaginella, a relative of the ferns.

We expect leaves to be green. The chlorophyll pigments in leaves absorb light at all visible wavelengths, but little in the green range; light scattered out of the leaf surfaces gives them their characteristic color. Leaf color other than green means that wavelengths normally captured by chlorophylls are absorbed by other pigments or reflected out of the leaf. This means that less energy is captured for photosynthesis and growth. Why did these spectacular blue plants, and other species I later found growing in such deeply shaded environments, reflect light that would otherwise add to their photosynthetic efficiency? How did they produce this spectacular color? When I left Southeast Asia in 1976 and eventually moved to Miami, I looked for iridescent blue plants in shady understory environments in the rain forests of Central and South America-and found them. The questions that arose about the nature and function of the color of these plants have stimulated two decades of research on iridescent blue plants in the rain forests of both the Asian and American tropics.

Structural Color

It was obvious that the basis of color production in the Selaginella plants was not pigmentation, because the leaves lost their color when immersed in water. Furthermore, only pigments found in the chloroplasts, the photosynthetic organelles found in plant cells, could be extracted. Blue color in flowers and fruits is almost always caused by anthocyanins modified from their association with metallic ions, or by other flavonoid pigments. Anthocyanins were not present in Selaginella. These leaves had to produce color by some physical means either by diffraction from a grating-like structure, by the selective scattering of small particles or by the constructive interference found in thin films. For Selaginella willdenow it was easy to eliminate the first two explanations. Color production was of a single hue, not the spectrum of colors that would result from diffraction. Reflectance measurements showed a clear peak, not a continued increase at shorter wavelengths as would be expected from small-particle scattering. Structural coloration commonly occurs in animals, particularly insects and birds, and is usually caused by thin-film interference. The iridescent green wing chevrons of the majestic Rajah Brooke birdwing (the largest wingspan of any butterfly) are produced in this way, by multiple layered structures in the wing scales that interfere with visible light.

The explanation for such color production was first provided by Thomas Young in 1801. Imagine a material made up of transparent layers with different optical densities, in other words where the index of refraction-the extent to which light waves are slowed down by the medium-is different for each layer. In this material light reflects at any boundary. Which wavelengths of light pass through the layer (destructive interference) and which are reflected from the layer (constructive interference) depends on the layer’s thickness, its refractive index and the angle at which the light enters. When light passing through and back is retarded by half a wavelength, the interference is destructive, so that these wavelengths pass through the layer. When light is reinforced at a full wavelength, the interference is constructive, producing an intense reflected metallic color. If one layer has a lower refractive index than the layers above and beneath it, a phase shift occurs. Because the balance between light reflectance and absorption depends on the thickness of individual layers as well as their optical density, a filter of different thickness has the same effect as a layer with a different refractive index. The layers in a filter can be extremely thin: At onequarter of a wavelength the interference effect is similar to that at a full wavelength of light.

By measuring the reflectance perpendicular to the leaf surface and assuming a refractive index of 1.45, typical of the moistened cellulose of cell walls or the membranes of organelles, it is possible to predict what thickness of layers in the leaves would produce an intense metallic coloration.

In Selaginella leaves, the task of locating the site of interference was simplified by the fact that we were sure it was located near the outer cell wall, where water could cancel its effect. Also, we could compare the cell structure in the iridescent blue leaves of some Selaginella plants with that in the green leaves of plants of the same species developing in more sunny locations. Charles Hebant, of the University of Montpellier, France, and I examined the epidermal cells of the leaves using transmission electron microscopy. We discovered two layers of the predicted thickness, approximately 80 nanometers, and opaque to the electron microscope, in the cell walls of Selaginella willdenow and S. uncinata. Such layers were absent from the green leaves of both species.

The simple explanation for iridescence in Selaginella left me unprepared for the more elaborate structures that turned up in other plants of the rain forest understory. In the New World tropical fern Danaea nodosa, repeated electron-opaque layers alternate with more transparent arcs of cellulose microfibrils, the long cylindrical fibers that make up plant cell walls. Similar patterns had been observed in the exocuticles of arthropods by A. C. Neville and his colleagues at the University of Bristol. Neville’s group showed that succeeding layers of fibrils are deposited at a regular angle in the formation of the cuticle, creating the patterns and the resulting iridescence.

The distance between the consecutive light and dark bands is a result of a step-by-step rotation, over a total of 180 degrees, in the orientation of the microfibrils. As a result, a slightly oblique cross-section of a beetle cuticle or, it turns out, a D. nodosa leaf, has a helicoidal appearance; it looks like a stack of coiled fibers. Neville and his co-workers demonstrated that this helicoidal structure is the cause of iridescent coloration in beetles.

Two mechanisms for color production may operate in these helicoidal layers. First, the periodicity of the layering provides the conditions for the reinforcing iridescence of multiple layers. Second, the helicoidal structures cause the circular polarization of white light, which can produce a narrow spectral region of reflected color. In D. nodosa the layers were of the predicted thickness (about 160 nanometers) to produce blue colors through interference. These layers were either absent or of the wrong thickness in the green leaves of adult plants. More recently, Kevin Gould, from the University of Auckland, and I elucidated the helicoidal basis for blue leaf iridescence in two Malaysian understory ferns, Diplazium tomentosum and Lindsaea lucida. We have not yet determined whether circular polarization also occurs.

Structures That Produce Color

Blue leaf iridescence is not limited to ferns and their allies, nor is blue iridescence limited to leaves. Some marine algae, particularly red algae, produce such color. Layers in the cuticle cause blue color by light interference in Iridaea cordata, but structures within cells may produce this color in other algae. A variety of flowering plants produce blue leaf iridescence, particularly in the Asian tropics. Gould and I studied two distantly related flowering herbs from the Malaysian rainforest understory. These studies helped us locate unusual structures that might produce the iridescence. In the peacock begonia, Begonia pavonina, and in Phyllagathis rotundifolia (of the Melastomataceae) the ultrastructural basis is a remarkably modified chloroplast, which we have termed an “iridoplast.”

Since chloroplasts are the sites where the hard work of photosynthesis takes place, they are the first place to look for interactions involving light. Normal chloroplasts contain discshaped membranes called thylakoids, collected to form stacks called grana, which in turn are connected by membranes in the stromal region (so called for the amorphous, enzyme-rich material called stroma that permeates the chloroplast). Chlorophyll-protein complexes on these membranes capture energy for the light reaction of photosynthesis.

Each iridoplast in the two plants we studied is shaped like a pancake and covers much of the bottom of the epidermal cell. Within, thylakoid stacks in close contact with one another (called “appressed” thylakoids) form the basis for the interference filter. Such structures are not found in the green leaves of these plants, which contain normal chloroplasts in other parts of the leaves.

In plants of Trichomanes elegans growing in extremely shady and wet sites in New World tropical rain forests, I have found similar structures in leaves. The filmy fern’s iridescent blue-green fronds contain modified chloroplasts in the epidermal cells. The grana stacks, which have five thylakoids each and are connected by extremely short stromal lamellae, form a repeating series of filters that produce the remarkable bluegreen interference color.

About 10 percent of fleshy fruits are blue in color. Such color is known to be produced by modified anthocyanin pigments in the outer layers of the fruit wall. But constructive interference could produce blue color in some fruits, and it may be that we have not looked carefully enough for this phenomenon.

When washing the outer pulp of fruits of the rudraksha tree (Elaeocarpus angustifolius) to collect the inner stones for germination, I noticed that no blue or red pigments were extracted in the water. These fruits are a persistent electric blue in color (see cover photograph), and I suspected that their brilliant color might be produced structurally. Subsequently I discovered that the interference color of the rudraksha fruits is indeed produced by a structure whose cellulose layers are of the predicted thickness to produce blue color. This structure, which I have called an “iridosome,” is different from those seen in leaves because it is secreted by the epidermal cells of the fruit and is located outside the cell membrane but inside the cell wall. A more careful search may reveal other fruits with structural color, particularly the close relatives of this tree.

Mysteries of Function

Earlier I asked what selective advantage a plant might gain by sacrificing photosynthetic efficiency for iridescence. The answer to this question, unfortunately, has proved elusive.

In a single habitat, the tropical rain forest understory, one finds structural coloration among distantly related species of shade-tolerant plants. This observation strongly suggests that there is selective advantage to this leaf character. Developing a hypothesis about the advantage is difficult: As I mentioned above, such blue reflectance means that these wavelengths are lost for photosynthesis by the leaves in a light-limited environment.

One day when I was first contemplating the Selaginella of the Malaysian rain forests, I looked at the compound lens of my camera and had the intuition that perhaps the interference of these leaf cell layers could also function as an anti-reflection coating, capturing certain wavelengths for use by the plant. A careful analysis of the optical properties of the blue leaves, compared to the green leaves of the same species, showed that the blue leaves were in fact able to absorb more radiation in the longer wavelengths of the visual range. In a sunny setting chlorophylls have access to a wide spectrum of visible light. But the longer wavelengths are more available for photosynthesis in rain forest shade.

In Selaginella I also learned that the blue iridescence develops under experimental conditions that replicate the shift in spectral quality characteristic of forest shade conditions. However, in the labratory these conditions changed the leaves in other ways, so it has not been possible to obtain direct physiological evidence for the advantage of iridescence in low light conditions.

The selective advantage that might be conveyed by iridescent blue leaves, then, particularly where the structures are beneath the leaf surface (suggesting they are not altering surface reflectance), requires further study. What may be important for a plant is the spectral environment within the iridescent leaves. Interference significantly affects the spectral quality of the light environment within iridescent leaves. The leaves of most species produce a reflectance peak at wavelengths of 460 to 480 nanometers, which means a reduction of these wavelengths within the leaf.

Might there be an advantage in keeping these wavelengths out of the leaf interior? Kevin Gould and I, along with our colleagues David Kuhn and Steve Oberbauer at Florida International University, have argued that the anthocyanins in red-undersurfaced leavesalso common in rain forest understory plants-may protect leaf cells against damage to chloroplasts. Anthocyanins absorb light down into the blue region, to about 470 nanometers. Bright light at this wavelength would otherwise be absorbed by chlorophyll b, and the anthocyanins may protect the photosynthetic apparatus from brief flecks of intense sunlight. (In particular, the anthocyanins may protect the action centers of the group of pigments and proteins called photosystem II that are located on the grana lamellae. These contain most of the chlorrophyll b and are the principal sites of photoinhibition and photodamage in the chloroplast.)

Interference may offer a similar protection, reflecting wavelengths otherwise absorbed by chlorophyll b. Two of the taxa produce a reflectance peak of 530-538 nanometers. This spectral region coincides with a wavelength considered by John Gamon, of California State University, Los Angeles, to be a critical indicator of photosynthetic function in plants.

In the iridescent blue fruits of rudraksha, the search for the possible selective advantage of structural coloration began with examining how an interference filter, compared to pigmentation, would produce color. In typical blue fruits, such as those of Heliconia psittacorum from Central America, color is produced when light passes through epidermal cells that have pigments concentrated in their large central vacuoles. The colorless, spongy tissue directly beneath scatters light back through the pigmented cells. The blue color is the small portion of the spectrum not absorbed by the pigments. In the rudraksha fruits, the interference filter constructively interferes in the blue range (best at wavelengths of about 430 nanometers). The skin is transparent at longer wavelengths.

Fruit pigeons and flightless cassowaries in Queensland and New Guinea find the fruits of these trees on the ground and avidly consume them, exposing and dispersing the tough rudraksha seeds within. The fruit color may be selectively advantageous because it is durable and lasts even during the breakdown of the pulp. In addition, this structural color may provide a physiological advantage. The ripe rudraksha fruits contain tissues rich in chlorophyll beneath their blue skin. Since the fruit skin is transparent to much of the light that plants use in photosynthesis, structural color may enable the fruits to reduce their costs of production by permitting light for photosynthesis to penetrate when the fruits are ripe, even after they have fallen from the parent tree. Since photosynthetic plants release carbon dioxide by respiration when they are in the dark, I looked at the amount of carbon dioxide lost by these fruits under shady compared to dark conditions. I detected a reduction in the COZ lossthat is, an enhancement of the fruits’ carbon budgets-a finding consistent with this hypothesis. The leaves of plants found in the deep shade of tropical rain forests have evolved especially sophisticated mechanisms for modifying light environments in their interiors. Their internal anatomy displays chloroplasts, or allows them to move, to optimize interception. Lens-like epidermal cells direct light towards these chloroplasts in shady conditions. Accessory pigments absorb particular wavelengths, providing protection from intense sunlight. Here we describe structures that physically modify the interior light environments in another way, through constructive interference. We see this as a remarkable color display How it aids in the survival of these shade-tolerant plants, if at all, will require much more study One thing is clear: The more we look, the more treasures we find in the world’s tropical rain forests.