The development of entirely new amino acids was the key opening the way to complex multicellularity. It permitted evolution of the structural proteins needed to provide mechanical support for increased size and greater morphogenetic experimentation.
If there is intelligent life elsewhere in the Universe with whom we might communicate, it is unlikely that such life is unicellular or primitively multicellular. It is more likely to be of the complex multicellular type-the so-called higher forms of life. The purpose of this paper is to emphasize the fundamental importance of two "rare" protein amino acids to the development of complex multicellular life on Earth.
There are only 20 amino acids found in proteins. A statement similar to this occurs in almost every biochemistry text. As a generalization to most proteins this statement is certainly true, but from the standpoint of the Structural proteins and of the evolutionary history of higher organisms it can be rather misleading. Despite the repetitive, degenerate nature of the genetic code, there are two important protein amino acids that are uncoded: hydroxyproline and hydroxylysine. These two amino acids play critical roles in the formation of structural glycoproteins in both plants and animals. In an evolutionary context, the absence of these hydroxyamino acids from the genetic code implies that they are relative "newcomers" on the biological scene, and it is fair to conclude that their functional importance must therefore outweigh the complexities required for their formation.
One or both of these hydroxyamino acids occur in all metazoans (Bairati, 1972; Adams, 1978), higher plants, most higher algae (Lamport, 1977), and certain types of fungi (LéJohn, 1971). They are absent (or very  rare) in "higher" fungi, procaryotes, protozoans, and red algae. In other words, these amino acids are widely distributed among the higher forms of life but are generally absent from lower organisms. The higher fungi may be the exception that proves the rule. The Ascomycetes and Basidomycetes like the red algae from which they were probably derived (Demoulin, 1974) lack both hydroxyamino acids.
The biosynthesis of both hydroxyproline and hydroxylysine follows a pathway whereby the parent amino acids proline and Iysine are hydroxylated only after incorporation into polypeptide linkage (see Miller and Matukas, 1974, for a review). The free amino acids cannot act as substrates The steps in the formation of these two uncoded, posttranslational hydroxy ammo acids are both specific and complicated. Hydroxylation is mediated by the enzymes proline and Iysine hydroxylase, respectively. Both enzymes meet the same four requirements: (1) molecular oxygen, (2) ferrous iron (3) ascorbic acid, and (4) o`-ketoglutarate. Moreover, the same basic pathway is found to exist in both plants and animals (Chrispeels, 1976), indicating a common evolutionary origin in some primitive, and probably photosynthetic, ancestor (Aaronson, 1970; Lamport, 1977).
The peptide-bound hydroxyproline and hydroxylysine in metazoans appear in the structural protein collagen. In the higher plants and green algae, hydroxyproline occurs ultimately in a family of glycoproteins termed extensins (Lamport and Miller, 1971). The ultimate location of peptide-bound hydroxyproline or hydroxylysine in the other organisms in which it occurs is not known for certain.
The requirement for -ketoglutarate by the hydroxylase enzymes is absolute and specific (Hutton et al., 1967). Since -ketoglutarate is a tricarboxylic acid (Krebs) cycle intermediate, it may not be coincidental that those organisms in which the TCA cycle may be limited (Klein and Cronquist, 1967, p. 189) -photosynthetic bacteria, blue-green algae, and red algae-are also among those organisms generally lacking in hydroxyproIine or hydroxylysme. -Ketoglutarate is also a requirement in the synthesis of Iysine via the aminoadipic acid pathway. Interestingly, the few organisms that synthesize Iysine by this pathway-red algae, euglenoids, and some higher fungi-also lack these two hydroxyamino acids
The enzyme requirement for molecular oxygen is also absolute. It has been demonstrated that the oxygen of the hydroxyl group is derived from molecular oxygen rather than from water in collagen hydroxyproline (Fujimoto and Tamiya, 1962; Prockop et al., 1962), in collagen hydroxylysme (Hausmann, 1967), and in extensin hydroxyproline (Lamport, 1963) Indeed, all collagens appear to require molecular oxygen, even those in the so-called anaerobic, parasitic worms (Fujimoto, 1967; Smith, 1969).
Although the basically similar hydroxylation reactions are important in both extensin and collagen, additional strength is provided through subsequent  steps. It is the linkage between carbohydrate and protein that is crucial. In plants, the extensin hydroxyproline is covalently bonded to a sugar (arabinose) through an O-glycosidic linkage (Lamport, 1967). In metazoans, the situation is different. Collagen hydroxyproline, in association with glycine, forms and stabilizes the unique triple-stranded structure of the protein macromolecule (Berg and Prockop, 1973; Jimenez et al., 1973). It is the hydroxylysine in collagen that is linked to the carbohydrate (Butler and Cunningham, 1966). This material is further strengthened into true collagen fibrils by intermolecular crosslinks that involve Iysine and hydroxylysine mediated by another oxygen-dependent enzyme, Iysyloxidase (Traub and Piez, 1971). Hydroxylysine is essential to collagen crosslinks (Bailey et al., 1974). Among the numerous vertebrate and invertebrate collagens studied, only the nematode Ascaris is an exception (McBride and Harrington, 1967). The crosslinks are important to the high strength and chemical resistance necessary for the functioning collagen fibrils.
Is there significance in all of this for the origin and development of complex multicellular life? In plants, the precise function of extensin as a structural protein is not known, although its extracellular position and its association with periods of active growth in higher plants suggest an important role in the regulation of growth and development (Lamport, 1977). Similarly, collagen is also involved in developmental tissue interactions through extracellular matrices (see Lash and Burger, 1976). But in addition to this role, its primary role as the "tape and glue" of the metazoan world is undeniable. It has no substitute in this function as a connective-tissue protein, Collagen is the principal organic matrix component of vertebrate bone. It holds the calcitic plates of echinoderm skeletons together. Worm cuticles and body walls are collagenous, as is the mesoglea of jellyfish. It is a mayor component of sponges, and the muscles of bivalves are held to their shell with its help. These few examples serve to illustrate its importance.
The evolution of a functional substitute for collagen using only some combination of the 20 amino acids in the genetic code has not yet been achieved. And it is unlikely that it was ever achieved at any time in the past when one considers the lengths to which metazoans must go to manufacture this critical substance. First of all, a mixed-function oxygenase is required to synthesize hydroxyproline and hydroxylysine. Oxygenases are known to be comparatively inefficient and to compete with conventional respiratory pathways for free oxygen (Kaufman, 1962; Hayaishi, 1962). Second, the all important intermolecular crosslinks that provide the mechanical strength also need oxygen (Iysyloxidase) and in addition require more Iysine-an amino acid that animals are unable to synthesize and must obtain through food sources. Add to this the evolution of the posttranslational pathway with four enzyme cofactors and you have a protein that is complicated and expensive to manufacture. Thus, if any organism in the past had found a  simpler way to make this structural protein, or even a suitable substitute, I would have had a clear adaptive advantage. The importance of hydroxyproline and hydroxylysine is therefore indelibly expressed through their wide multicellular distribution in nature. Evolutionary increase in size and complexity through multicellularity was accomplished through oxidative metabolism, the eucaryotic condition, and structural proteins using new amine acids that placed a premium on the availability of free oxygen. Only the higher fungi have managed a small measure of success in increasing size and complexity without the new amino acids. But they lack mechanical strength and this is reflected in their poor fossil record.
If the first 2.5 billion years of Earth history were dominated by an environment in which free oxygen was generally unavailable to the biosphere because of primitive reduced inorganic acceptors or "sinks," then one of the principal strategies of life would have been to adapt to this environment Neither collagen nor extensin would have been possible in the absence of free oxygen, and the fossil record shows that during this period few significant increases in body size were made. After the exhaustion of most primordial inorganic oxygen sinks about 2 billion years ago (Cloud, 1976; Schopf, 1978) and the development of the eucaryotic cell and oxygenic photosynthesis, the major environments of the world began to turn from predominantly reducing to predominantly oxidizing conditions. Multicellularity would have been a priority development and collagen one of its certain prerequisites. The geologic evidence shows that the transition to an oxygenic world did not take place rapidly; if it had, there would be major deposits of reduced carbon in the sediments of the period, and none is found. The buildup of free oxygen is more likely to have been rather slow. With some limited free oxygen present in the environment, limited collagen (and extensin) could have evolved in some organisms where the wide-open ecological niches available and the adaptive advantages being conveyed outweighed the complexities and expense involved in its manufacture. I have already discussed how this may have affected the fossil record (Towe, 1970), but it is worth repeating that in this initially low-oxygen environment, the competition for the low levels of free oxygen by oxidative metabolisms would have restricted most collagen use to small, thin, diffusion-limited organisms unlikely to be found preserved as fossils. As Boaden (1977) has emphasized, the world at this time may have been like the modern thiobios-a sulfide-rich habitat dominated by mostly microscopic, interstitial meiofaunal elements.
The biochemistry of collagens from modern near-anaerobic nematodes is instructive in making further comparisons with the Late Precambrian fossil record. Both the cuticle and body wall of Ascaris lumbricoides contain collagen. The body-wall collagen has large amounts of hydroxyproline and hydroxylysine. Adapted to low oxygen tensions, their formation through the  hydroxylating enzymes is actually inhibited by too much oxygen (>1%) in the environment (Fujimoto and Prockop, 1969). The cuticle collagen lacks hydroxylysine, contains little hydroxyproline, and appears to be strengthened by disulfide crosslinks (McBride and Harrington, 1967), another adaptation to low-oxygen tensions.
Given this information, one can speculate that the early use of collagen in Late Precambrian low-oxygen environments may have eventually produced "worms" with similarly adapted collagen metabolism, which permitted some of them to attain much larger sizes than the remaining interstitial faunal elements. Perhaps the enigmatic coiled fossils from the billion-year-old Greyson Shale (Walter et al., 1976) or from the Little Dal Group (Hofmann and Aitken, 1979) are rare body fossils of such worms. Or perhaps some of the Late Precambrian burrows were produced by similar worms who had become adapted to the low-oxygen environment and were, like their modern thiobiotic descendants, burrowing to avoid the increasing oxygen tensions that must inevitably have taken place. Burrowing to avoid oxygen at the sediment-water interface in this very early stage of metazoan history seems more likely than burrowing to avoid predators, the types of which are unknown and the fossil evidence for which is otherwise nonexistent.
Ultimately, the further increase in availability of free oxygen as the result of increasing oxygenic photosynthesis would have brought an end to such adaptations, and therefore any experiments toward developing collagens completely free of an oxygen requirement were terminated. At the same time, the high competitive priority of respiratory events for oxygen would have been moderated, allowing many more morphological experiments with collagen to take place in other previously limited metazoan phyla. Even the sclerotization of the arthropod cuticle, which is also inhibited by lack of atmospheric oxygen (Richards, 1951), would have been improved All this would have caused a dramatic worldwide increase in the size and hence ready preservability of numerous organisms. The Late Precambrian-Early Cambrian fossil record can then be interpreted as an explosion of fossils rather than as a sudden eruption of metazoan phylogenesis with highly evolved, diverse, and morphogenetically advanced forms appearing suddenly side by side around the world, few of which have any plausible immediate ancestors as fossils.
What does the hydroxyproline-hydroxylysine connection with multicellular evolution have to tell us about phylogeny? Although the data are very incomplete and hence subject to change, there are some interesting Observations that presently invite speculation and comment. These speculations fall into the category of the "wild and wooly," but they may serve to provoke thought and further data collection.
 It is conventional to consider that the origin of the Metazoa lies somewhere among the protozoans (animal-like protists) and probably among those that are nonpseudopodial (Hanson,1976). There is a problem with this viewpoint. It concerns the origin of collagen, which, although it occurs in all metazoans, including the Porifera (Garrone, 1978), has not been found in any protist. As pointed out by Delmer and Lamport (1977), "the origin of collagen predicates the origin of the Metazoa." And collagen requires hydroxyproline and hydroxylysine biosynthesis.
Hydroxyproline has been found in only one protozoan to date: the shell of an amoeboid foraminifera (Hedley and Wakefield, 1969). Hydroxylysine has never been reported in any protozoan. While it is true that many more protozoans need to be examined carefully (especially the choanoflagellates), it is still fair to ask those who will derive the Metazoa from animal-like protists to explain how collagen could have been evolved from organisms not equipped with the complex pathways required to synthesize either of the basic building blocks. To argue for the independent discovery in some protozoan of the complicated biosynthetic pathways involved is to suggest a highly unlikely convergence. Lamport (1977) recognized this problem and resurrected the alga Volvox as a branchpoint. It contains abundant hydroxyproline and has the appearance of a blastula. But the close derivation of the Metazoa from any alga requires, in addition to the origination of collagen, the loss of Iysine synthesis and the replacement of phototrophy by ingestion. The ability to synthesize chitin and ferritin might also be a problem since the former occurs in few algae and the latter in none.
Implausible as it may sound, it is instructive to examine the potential of the fungi (fungus-like protists) in this regard. Considered as a broad group, fungi have the ability to synthesize hydroxyproline, chitin, cellulose, and even ferritin. They have already replaced phototrophy with heterotrophic absorption. And fungi were surely not derived from protozoa, simply because to do so would require that the protozoan ancestor, already having lost the ability to synthesize Iysine, rederived it later - a peculiar step indeed. Whittaker (1977) has already noted that "metazoans with digestive tracts have probably evolved from absorptive flagellates and, in this evolution, internalized the process of food absorption and added it to the process of ingestion." Could some lower fungal type have given rise to a protometazoan by some neotenous or paedomorphic transformation during a flagellate stage? And if it were one of the fungi that had utilized the aminoadipic acid pathway for Iysine synthesis, the loss of this capability might have conserved o`-ketoglutarate for proline hydroxylation. There would have been nothing to lose but the ability to synthesize Iysine and everything to gain in evolutionary potential.
It took over 2 billion years for life on Earth to evolve the capacity for complex multicellular development. This length of time alone would seem to  be good evidence that, with or without free oxygen, no combination of the 20 coded amino acids could be found that could produce a structural protein comparable to the extensin-collagen family. And while free oxygen was pivotal for efficient respiration and the eucaryotic condition, these advances were also insufficient to allow for the increase in size and complexity we find in metazoans and higher plants. The development of entirely new ammo acids was the real key to opening the way for complex multicellularity. This permitted the evolution of the structural proteins necessary to provide the mechanical support for increased size and morphogenetic experimentation. Only then could the long-vacant niches begin to be filled by larger, rapidly diversifying organisms better suited to fossilization. If there had remained only 20 amino acids in all proteins, Earth would probably still be dominated, as it was for much of the Precambrian, by procaryotes and smaller eucaryotes, and fossils would be rare everywhere.
I thank Dr. James W. Valentine for agreeing to read a draft of the manuscript and for making helpful suggestions for its improvement.
- Aaronson, S.: Molecular Evidence for Evolution in the Algae: A Possible Affinity Between Plant Cell Walls and Animal Skeletons. Ann. N.Y. Acad. Sci., vol.175, art. 2,1970, pp. 531-540.
- Adams, Elijah: Invertebrate Collagens. Science, vol. 202,1978, pp. 591-598.
- Bailey, A. J.; Robins, S. P.; and Balian, G.: Biological Significance of the Intermolecular Crosslinks of Collagen. Nature, vol. 251, 1974, pp.105-109.
- Bairati, A.: Collagen: An Analysis of Phylogenetic Aspects. Bull. Zool., vol. 39,1972, p. 205.
- Berg, R. A; and Prockop, D. J.: The Thermal Transition of a Non-Hydroxylated Form of Collagen. Evidence for a Role for Hydroxyproline in Stabilizing the Triple-Helix of Collagen. Biochem. Biophys. Res. Comm., vol. 52,1973, pp.115-129.
 - Boaden, P. J. S.: Thiobiotic Facts and Fancies. Mikrofauna Meersboden vol. 61,1977, p. 45.
- Butler, William T.; and Cunningham, Leon W.: Evidence for the Linkage of a Disaccharide to Hydroxylysine in Tropocollagen. J. Biol. Chem., vol. 241,1966, pp. 3882-3888.
- Chrispeels, Maarten J.: Biosynthesis, Intracellular Transport, and Secretion of Extracellular Macromolecules. Ann. Rev. Plant Physiol., vol. 27 1976, pp.19-38.
- Cloud, Preston: Beginnings of Biospheric Evolution and Their Biogeochemical Consequences. Paleobiol., vol. 2,1976, p. 351.
- Delmer, D. P.; and Lamport, D. T. A.: Cell Wall Biochemistry Related to Specificity in Host-Plant Pathogen Interactions. In: Cell Wall Biochemistry, B. Solheim and J. Raa, eds. Proceedings of a symposium held at the University of Trowis, Trowis, Norway, Aug. 1976. Columbia Univ. Press, N.Y., 1977, pp. 85-104.
- Demoulin, V.: Origin of Ascomycetes and Basidiomycetes-Case for a Red Algal Ancestry. Botanical Rev., vol. 40, 1974, pp. 315-345.
- Fujimoto, Daisaburo: Biosynthesis of Collagen Hydroxyproline in Ascaris. Biochem. Biophys. Acta, vol. 140,1967, pp. 148-154.
- Fujimoto, D.; and Tamiya, N.: Incorporation of 18O from Air into Hydroxyproline by Chick Embryo. Biochem. J., vol. 84, 1962, pp. 333-335.
- Fujimoto, Daisaburo; and Prockop, Darwin J.: Protocollagen Proline Hydroxylase from Ascaris lumbricoides. J. Biol. Chem., vol. 244, 1969,
- Garrone, Robert: Phylogenesis of Connective Tissue. Morphological Aspects and Biosynthesis of Sponge Intercellular Matrix. Frontiers of Matrix Biology, vol. 5,1978, p. 1.
- Hanson, E. D.: Major Evolutionary Trends in Animal Protists. J. Protozool., vol. 23,1976, pp. 4-12.
 - Hausmann, E: Cofactor Requirements for the Enzymatic Hydroxylation of Lysine in a Polypeptide Precursor of Collagen. Biochem. Biophys. Acta, vol. 133,1967, pp. 591-593.
- Hayaishi, Osamu: History and Scope. In: Oxygenases, O. Hayaishi, ed., Academic Press, N.Y., 1962, pp.1-29.
- Hedley, Ronald H.; and Wakefield, James: Fine Structure of Gromia oviformis (Rhizopodea: Protozoa). Bull. Brit. Mus. Nat. Hist., vol.18,1969, p. 67.
- Hofmann, H. J.; and Aitken, J. D.: Precambrian Biota from the Little Dal Group, Mackenzie Mountains, Northwestern Canada. Canadian J. Earth Sci., vol.16,1979, p. 150.
- Hutton, John J., Jr.; Tappel, A. L.; and Udenfriend, Sidney: Cofactor and Substrate Requirements of Collagen Proline Hydroxylase. Arch. Biochem. Biophys., vol.118,1967, p. 231.
- Jimenez, Sergio; Harsch, Margaret; and Rosenbloom, Joel: Hydroxyproline Stabilizes the Triple Helix of Chick Tendon Collagen. Biochem. Biophys. Res. Comm., vol. 52,1973, pp. 106-114.
- Kaufman, Seymour: Aromatic Hydroxylations. In: Oxygenases, O. Hayaishi, ea., Academic Press, N.Y., 1962, pp.129-180.
- Klein, Richard M.; and Cronquist, Arthur: A Consideration of the Evolutionary and Taxonomic Significance of Some Biochemical, Micromorphological, and Physiological Characters in the Thallophytes. Quart. Rev. Biol., vol. 42, no. 2, June 1967, pp.105-296.
- Lamport, Derek T. A.: Oxygen Fixation into Hydroxyproline of Plant Cell Wall Protein. J. Biol. Chem., vol. 238, no. 4,1963, pp. 1438-1440.
- Lamport, Derek T. A.: Hydroxyproline-O-glycosidic Linkage of the Plant Cell Wall Glycoprotein Extensin. Nature, vol. 216, 1967, pp.1322-1324.
- Lamport, Derek T. A.: Structure, Biosynthesis and Significance of Cell Wall Glycoproteins. Recent Adv. Phytochem., vol. 11, 1977, pp. 79-115.
- Lamport, Derek T. A.; and Miller, David H.: Hydroxyproline Arabinosides in the Plant Kingdom. Plant Physiol., vol. 48,1971, pp. 454-456.
- Lash, J. W.; and Burger, M. M., eds.: Cell and Tissue Interactions. Soc. Gen Physiol. Ser., vol. 32,1976.
- LeJohn, H. B.: Enzyme Regulation, Lysine Pathways and Cell Wall Structures as Indicators of Major Lines of Evolution in Fungi. Nature vol. 231,1971, pp. 164-168.
- McBride, O. Wesley; and Harrington, William F.: Ascaris Cuticle Collagen on the Disulfide Cross-Linkages and the Molecular Properties of the Subunits. Biochemistry, vol. 6,1967, pp.1484-1498.
- Miller, Edward J.; and Matukas, Victor J.: Biosynthesis of Collagen. Fed Proc., vol. 33, May 1974, pp.1197-1204.
- Prockop, D.; Kaplan, A.; and Udenfriend, S.: Oxygen-18 Studies on the Conversion of Proline to Hydroxyproline. Biochem. Biophys. Res. Comm. vol. 9,1962, pp. 162-166.
- Richards, A. Glenn: The Integument of Arthropods. The Chemical Components and Their Properties, the Anatomy and Development, and the Permeability. Univ. of Minnesota Press,1951.
- Schopf, J. William: The Evolution of the Earliest Cells. Sci. American, vol. 239,1978, pp.110-138.
- Smith, Malcolm H.: Do Intestinal Parasites Require Oxygen? Nature, vol. 223,1969, pp.1129-1132.
- Towe, Kenneth M.: Oxygen-Collagen Priority and the Early Metazoan Fossil Record. Proc. Natl. Acad. Sci. USA, vol. 65,1970, pp. 781 -788.
- Traub, Wolfie; and Piez, Karl A.: The Chemistry and Structure of Collagen In: Advances in Protein Chemistry, vol. 25,1971, pp. 243-352.
- Walter, M. R.; Oehler, John H.; and Oehler, Dorothy Z.: Megascopic Algae 1300 Million Years Old from the Belt Supergroup, Montana: A Reinterpretation of Walcott's Helminthoidichnites. J. Paleontol., vol. 50,
- Whittaker, R. H.: In: Protozoa of Medical and Veterinary Interest, J. P. Kreier. ed.. Academic Press. N.Y. 1977