From code to art: how does a linear set of instructions result in a beautifully crafted pattern? Diatoms do it, and scientists are struggling to figure out how. So far, they can see where spots of paint are appearing on the canvas, but the system that directs the finished masterpiece eludes them. 

Morphogenesis is the construction of a functional shape from component parts. It’s a huge mystery in biology. It isn’t enough to accumulate the parts; they must be fit together according to an overall plan—in the right places, in the right order, and at the right time. Laufmann and Glicksman pointed out the mystery of bone morphogenesis in a sidebar on page 78 of Your Designed Body, comparing it to the construction of a house. “What does it take to build a house?” they ask; “Where are the shapes for these bones specified?”

Since bones are made by many individual (and independent) bone cells, building a bone is an inherently distributed problem. How do the individual bone cells know where to be, and where and how much calcium to deposit? How is this managed over the body’s development cycle, as the sizes and shapes of many of the bones grow and change? Surely the specifications for the shapes, their manufacturing and assembly instructions, and their growth patterns must be encoded somewhere. There must also be a three-dimensional coordinate systemfor the instructions to make sense. Is the information located in each bone cell, or centrally located and each individual bone cell receives instructions? If each bone cell contains the instructions for the whole, how does it know where it is in the overall scheme? How do all those bone cells coordinate their actions to work together rather than at odds with each other? [Emphasis added.]

On a smaller scale, diatoms face (and solve) this same challenge. Instead of organizing cells together, they fit proteins and inorganic silica together. Diatoms have a blueprint (DNA) and a parts list (proteins) for construction of their glass houses. What drives them to create geometrical shapes from the parts, and place them in accurate positions in 3D space? This seems beyond the capabilities of parts and blueprints. It would be like placing a blueprint for a house on a pile of lumber, pipe, wire, and glass and expecting the house to self-organize. Even if a complete set of tools were available nearby, nothing would happen without foremen and skilled workers. 

Diatom Houses

There are tens of thousands of species of diatoms exhibiting a multitude of shapes. A few are five-pointed stars; others form triangles, squares, or rods. Each consists of two halves, called frustules or valves, that are made of silica, fitting together like halves of a pill box. The edges of the valves contain girdle bands that hold the top and bottom together. The faces of the valves are often adorned with pores organized into wonderfully detailed arrays that display intricate geometry under a microscope. Diatoms build their houses from the inside out. 

A species named Thalassiosira pseudonana is shown at the top of this article. Its circular lid displays “hierarchical patterns of meso- and macropores, ribs, tubes, and spines,” arranged with geometrical precision. These valves are as functional as they are beautiful. Possessing enormous strength, they protect the organism from predators and possibly from UV light. The pores may act like lenses, too, channeling the proper wavelengths of light to the photosynthetic machinery inside (see my article from April 2021).

Valiant Effort

In a determined effort to understand how the diatom builds its glass house, a team from the Center for Molecular and Cellular Bioengineering in Dresden, Germany, attempted to identify the parts of the “silica deposition vesicle” (SDV) for T. pseudonana and determine how the protein machinery works to fit the “silica precursors” into their positions on the valve. Their results were published in PNAS by Christoph Heintze et al., “The molecular basis for pore pattern morphogenesis in diatom silica.” 

The magnitude of the problem can be appreciated in their introduction. The mystery of morphogenesis extends beyond diatoms, and understanding it could lead to revolutionary technologies.

Numerous organisms produce inorganic materials with amazingly complex morphologies and extraordinary properties in a process termed biomineralization. Prominent examples include the single-domain magnetite nanocrystals of bacteria that act as sensitive magnetic field sensors, the nacreous calcium carbonate layers of mollusks with exceptionally high fracture resistance, and the hierarchically porous, silica cell walls of diatomswith intriguing photonic properties. A fundamental understanding how genetically encoded machineries are capable of establishing physical and chemical forces that drive morphogenesis of such intricate mineral structures is currently lacking. Therefore, unveiling the mechanisms of biomineralization holds the promise of gaining advanced capabilities to synthesize minerals with tailored properties using environmentally benign processes.

This is all highly appetizing to consider, but reading their paper is like watching a Sherlock Holmes movie that never resolves: a plethora of clues, but no answer to the big question: who dunnit? “Come back later for the next exciting episode” is hardly satisfying. 

Not that the authors didn’t try; they isolated the SDV, a first. They identified thousands of protein molecules in the SDV and searched for matched sequences in the UniProt database to sift the data down to the most likely candidates involved in pattern formation. After pinpointing some, they ran gene knockout experiments to see what resulted. They also identified receptors in the cell membrane for these proteins. Transporter proteins, ion pumps, silica transporters and other parts were labeled in a model diagram (Figure 5). They followed up on hypotheses that silica deposition involves liquid-liquid phase separation (LLPS), causing unassembled silica in the organic matrix to organize into “droplets” ready for deposition.

Proteomics analysis of the intracellular organelle for silica biosynthesis led to the identification of new biomineralization proteins. Three of these, coined dAnk1-3, contain a common protein–protein interaction domain (ankyrin repeats), indicating a role in coordinating assembly of the silica biomineralization machinery. Knocking out individual dank genes led to aberrations in silica biogenesis that are consistent with liquid–liquid phase separation as underlying mechanism for pore pattern morphogenesis.

Their model shows that dAnk1 appears to assemble the droplets, and dAnk2 and dAnk3 cooperate in stabilizing and disassembling them. But they never found the artist. The morpho genius behind morphogenesis remains unknown. 

Origin of Morpho: First recorded in 1850–55; from New Latin Morphō, genus name, from Greek Morphṓ “the Shapely, the Beautiful” (an epithet of Aphrodite in Sparta), akin to morphḗ “form, shape, figure, beauty.”

Something Missing

The authors considered alternative hypotheses regarding the mechanism of pore pattern formation: is it template-driven or self-organizational? While finding that the dAnk1-3 genes “influence the morphogenesis of pore patterns,” no “morphogene” was found. This leads them to believe that “additional proteins and possibly other components are involved in the morphogenesis process.” But is this like looking for more tools lying around on a house construction site? Where is an entity that knows the master plan and understands how to carry it out?

Considerations such as this over many years led Michael Denton to reject genetic determinism and embrace structuralism: the philosophy that form precedes function, not vice versa. Biological structures arise from internal “laws of biological form,” he said in 2016, that are universal and built into the properties of matter.

On a personal note, it was my own increasing recognition that the gene-centric paradigm was failing at the cellular level and that the architecture of cells is an “epigenetic affair,” the result of the self-organization of cellular matter, which was one of the major factors influencing my own move to structuralism.

Denton, Evolution: Still a Theory in Crisis, p. 259.

Yet even structuralism seems to be missing something. There is no law of biological form that necessarily makes it self-organize into a five-pointed star or a triangle, else all diatoms would look the same. Thousands of other species within the same environment inherit very different shapes. There is no material law that could take a form like a five-pointed star and encode it into a genome that leads to consistent inheritance of that form in its progeny. Structuralism might explain snowflakes, which though profoundly unique, are nevertheless built on the same structural pattern inherent in ice crystals. Snowflakes, are also not inherited from a linear code, as in biology. The organizing principles in biological structures seem very different from any other examples of self-organization in nature.

Software in the Hardware

The only mechanism we know for sure can faithfully reproduce a form from a linear code is software. A designing intelligence that writes software does not have to be present when it is operating. A 3D printer can run automatically, producing unlimited copies of a shape, given sufficient supply of resin. If a toy shop were running software-directed morphogenesis, no amount of inspection of the properties of the resin and machinery would predict the shape of a statuette coming out. The shape was in the mind of the programmer, not in the properties of the ingredients, the laws of nature, or the environment.

More Design in Diatoms

Two other facts about diatoms should arouse our appreciation of their intelligent design. One is that another atomic element — silicon — has found its way into the periodic table of biology. See our other articles about elements in life: phosphorus, boron, potassium, and about 20 others. The finely-tuned properties of elements have been detailed by Michael Denton in The Miracle of Man and his other “Privileged Species” books and videos. Diatoms could have taken on plain shapes without silica shells and still survive, as do other plankton. But the world is enriched by their crystalline architectures.

Another awesome fact about diatoms is their contribution to animal life. Diatoms produce about 25 percent of the air we breathe. This raises a philosophical puzzle of how necessary things can also be beautiful. Would we expect a statue in a museum to sweep its own floor? A healthy atmosphere might come from amorphous blobs of photosynthetic organisms, but the beauty of diatoms adds artistry to function.