Creativity, Confusion For Genes

Mouse has a kinky-waltzer. Drosophila has a bride of sevenless. Yeast has a Wee-1. Gene and protein names often are based on the flamboyant, the descriptive, and the intentionally obscure. For many researchers, naming their discovery may be a rare opportunity to imbue their science with creativity. (See list on page 6 for origins of these names.) NOMEN-CLUTTER: Multiple names for genes and gene products causes confusion, says University of Alberta's Lawrence Puente. But creativity plus competition can equal confusion, says Lawrence Puente, a molecular biologist at the University of Alberta in Canada. "There are multiple names for the same things, the same names for different things, names that are misleading, and names that are just plain hard to remember. There's no simple answer to this problem, and consequently many different opinions on what the best nomenclature method would be." Competition and a quicker pace of discovery contribute to the confusion. As multiple groups of scientists rush to discover, report, and name the same gene or protein, they unintentionally erect Towers of Babel. For example, one catalytically inactive caspase has been published as MRIT, CLARP, Usurpin, FLIP, Casper, FLAME, CASH, and I-FLICE. Such instances present obstacles to scientists, agrees Julia A. White, a member of the Human Genome Organization (HUGO) nomenclature committee, based at University College, London. "When trying to find information in a database or in literature searches, it is very frustrating to find that you need to use three or four different names for the gene--with maybe a couple of variations in punctuation for each--in order to retrieve all the information you require." The committee strives to sort out this linguistic chaos. However, they inevitably remain behind as a result of both the speed and the scope of the Human Genome Project. They--and nomenclature committees for other species--also must weigh competing rationales like basing a name on a phenotype, on a function, or on a genotype. Some of the earliest--and most colorful--names are artifacts of the phenotype methodology. Geneticists would cause a mutation, then assign a name to that gene based on the mutation's manifestation. Drosophila gene names emerging from this method include fairly obvious descriptions like wingless, slowpoke, and hyperkinetic. The results of other mutations were harder to summarize and resulted in names like hedgehog and ether a go-go. "When the phenotype is the only relevant information you have, this is perfectly reasonable," White says. However, she favors changing names once scientists unearth more information about a gene. "There is a strong argument that the 'official' name for a gene should reflect its normal function and not the results of abnormal or nonexistent activity," she reasons. Her committee has changed many human gene names after scientists discerned their function. For example, one gene was assigned the symbol ACH for achondroplasia--a condition also known as thanatophoric dwarfism. However, it was reassigned FGFR3 for "fibroblast growth factor receptor 3." In rare instances, such as JAK, the scientific community keeps the letters of the acronym but changes the words that compose it to reflect a new knowledge of a protein's function. MISNOMER: Brandeis University's Jeffrey Hall points out that some researchers muddy the semantic waters by assigning names for phenotypes that appear to pinpoint functions that are actually unknown. Jeffrey C. Hall, a professor of biology at Brandeis University, points out that some researchers muddy the semantic waters by assigning names for phenotypes that appear to pinpoint functions that in actuality aren't known yet. He recalls good-naturedly chiding a colleague who named a mouse gene clock because, while it appeared that the mutant mouse had its circadian rhythm interrupted, the biological basis for the behavioral manifestation hadn't yet been established. "It's better to be a bit cryptic than [to be] possibly misleading," Hall says." For example, one gene he refers to as fruitless might sound vague to the uninitiated, but to people who study fly mating habits, the name makes perfect sense. It also opens the door to puns: "Our fruitless work has been going on for 20 years," Hall deadpans. He also points out that genes are often first found through phenotypes. Functions often aren't learned until years after a gene's initial discovery. Another system based on technology is emerging--which may lead to yet more name changes. "Now that more sequences are becoming available and the computer software to compare them is fast and efficient, it is possible to base systematic, hierarchical naming systems on the sequence relationships between genes," White says. "Names derived from this system could be applied across species, revealing evolutionary relationships, and usually sequence similarities are reflected by similarities in function." Lois J. Maltais, the nomenclature coordinator for the mouse genome database at the Jackson Laboratory in Bar Harbor, Maine, thinks such a system could be useful--especially for categorizing related proteins. "This type of naming schema is becoming more and more prevalent today with the advances in sequence-analysis tools and the ever-increasing identity of genes belonging to gene families," Maltais explains. Mouse and human sequences with a high degree of similarity are considered orthologous and use the same naming scheme. Examples include the mammalian ALDH (aldehyde dehydrogenase) family, the cytochrome P450 family, and the AAA superfamily. However, Maltais acknowledges that changing existing names to new ones proves troublesome to some scientists who coined them years ago. "They prefer to keep the old names and symbols." White concedes that each system has its advantages and disadvantages: "The Drosophila system has advantages of memorability and constancy; the system in humans has advantages of currency and meaning. The functional approach tends to lead to the grouping of genes into families with similar function, but at the stage where knowledge is limited, it can result in names which cover too broad a category or in problems with genes that could belong in more than one functional category." NOMEN-CLUTTER: Multiple names for genes and gene products causes confusion, says University of Alberta's Lawrence Puente. MISNOMER: Brandeis University's Jeffrey Hall points out that some researchers confuse matters by assigning names that appear to pinpoint functions that are actually unknown. bride of sevenless (boss)--Drosophila. Root "sevenless" refers to phenotype mutation in which the fly lacks "cell seven" function in the eye--resulting in blindness. Other names with the sevenless root include daughter of sevenless (dos) and son of sevenless (sos). circadian locomoter output cycles kaput (Clock)--Mouse. Mutant phenotype appears to disrupt sleeping and waking cycle. ether a go-go (eag)--Drosophila. One of a number of genes isolated on the basis of a twitching abnormality shown by mutant flies while under ether anesthesia. Other genes of this phenotype include Shaker (Sh) and Hyperkinetic (Hk). Most genes of this type were isolated by a group of Drosophilists who were interested in isolating genes that affect neural function. fruitless (fru)--Drosophila. The name emerged because mutant males would court females but were unable to breed with them. Synonyms include fru-satori and fruity (fty). The latter is considered offensive by some, because this earlier, pre-political correctness-era name referred to the mutant male's tendency to court other hedgehog (hh)--Drosophila. Refers primarily to absence of segmental boundaries and the lawn of denticles replacing lack of cuticles in fatal phenotypes. Synonyms include Moonrat (Mrt) and Mirabile (Mir). Janus Kinase (JAK): Originally named Just Another Kinase because this family of tyrosine kinases was isolated in a random screen and the function remained unknown. Later, it was revealed that these kinases are critical messengers in cytokine responses. Because cytokines are like hormones of the immune system, binding to cell-surface receptors, the JAKs were recast as Janus Kinases, as a reference to the Greek god of gates and doorways. kinky-waltzer (Kw)--Mouse. The name waltzer (v) derived from phenotype description of a mutant mouse dancing about its cage. The "kinky" refers to kinks in the tale of some mutant phenotypes. Other "waltzers" include Ames waltzer (av), Bronx Waltzer (bv), Jackson waltzer (jv), Nijmegen waltzer (nv). Waltzer types are not necessarily related by function or by chromosomal location. Wee-1--Yeast. Cells with a mutation resulted in an abnormally small protein. The name may be derived from a Scottish group that first observed this mutant, but the story is unverified. Function of the normal form of Wee-1 was later discerned; it helps regulate cellular division. Yama: Death protease name refers to the Hindu goddess of death; named thus for its death effector function. Also published as apopain and CPP32. Renamed caspase-3. --P.S. Whatever the rationale behind full names, even the most clever gene and molecule names invariably get shortened. For example, fruitless is shortened to fru. "Acronyms are an important tool in molecule naming, because they are a means to [shorten] a long, descriptive name," says Avi Ashkenazi, senior scientist at Genentech Inc. in South San Francisco, Calif. Good acronyms retain some form of meaning, and the best also allow for "nice-sounding, easy-to-pronounce" phonetics. Ashkenazi cites CED, short for "C. elegans death" as a good example. However, Christopher W. Arendt, a mo- lecular biologist at the University of Alberta in Canada, argues that many acronyms confuse rather than clarify, especially in verbal communication. "Even a patent lawyer would shiver to try to sort out the differences between Csk, Syk, Shc, Shp, Ship, Sirp, Syp, and Src--all of which are intracellular signal transducing proteins." Bruce J. Mayer, an assistant professor of microbiology and molecular genetics at Harvard University, still recalls the phonetic difficulties posed by his decision to assign an oncogene name. "When I announced to my [principal investigator] . . . that I would like to name my new oncogene Crk, he merely raised his eyebrows. But the next morning I found on my bench an extensive list of possible pronunciations: cork, crack, crank, creak, crick, crock, croak, crook." Mayer acknowledges that it's easy to complain about the system, but difficult to improve it. "The problem in the signaling field is that we have gotten sloppy and ended up with multiple proteins or genes with the same name." G. Steven Martin, a professor of molecular and cellular biology at the University of California, Berkeley, blames history for the proliferation of three-letter acronyms. At some point, he thinks, it became impolite to name a virus or disease after oneself. Thus, the three-letter system was born; it was kept afloat through inertia, he maintains. "[Three-letter acronyms] started out with bacterial nomenclature and [were] adopted by the oncogene people." Martin agrees that the problem of confusing three-letter acronyms seems most legion in the cellular signaling community. He says he's unaware of any formal, permanent group to tackle nomenclature in the signal transduction field. "It happens on an ad hoc basis in a certain field occasionally." Nomenclature committees try to sort out multiple names, settle disputes, assign symbols, and, frequently establish databases about names and their synonyms. Many, like the HUGO Nomenclature Committee, are permanent. But others form on an ad hoc basis after research in one field explodes. For example, about three years ago, the then-unnamed caspase field boomed, resulting in multiple names for the same proteins popping up in different journals. One caspase called Yama, for example, had two other published names. Confusion ensued. "It was clear that we had to do something," recalls Nancy A. Thornberry, senior research fellow at Merck and Co. Inc. in Rahway, N.J. Thornberry and colleagues formed an ad hoc committee, which coined the term "caspase" ("c" for "cysteine protease mechanism," "asp" for "artate specific protease," and "ase," signifying an enzyme). The members then published their recommendations, which Thornberry says have become widely accepted (E.S. Alnemri et al., Cell, 87:171, 1996). There are now 10 human caspases. Not everyone is comfortable with the role of such committees. "I feel far too much time is wasted by scientists participating in nomenclature committees, renaming things they didn't even discover," comments David V. Goeddel, chief executive officer of Tularik Inc. in South San Francisco, Calif. He feels proteins should be renamed only if the original name was clearly misleading regarding function or if a family of related proteins is found. White disagrees, pointing out that nomenclature committees can manufacture consensus more effectively than can the groups competing for the claim to a gene. "Perhaps surprisingly, when a gene has been called a different name by virtually every group working on it, it is often easier to get all groups to agree on an independent name chosen by us than to get them to agree to use one of the names already used," White says. "There is a saying in nomenclature circles that scientists would rather use someone s toothbrush than someone s nomenclature." White and Maltais both remark that databases help clarify the issue and help researchers keep track of multiple names. But the databases need to be updated and maintained if that is to happen. While a search conducted by The Scientist for ACH on the HUGO Genome Database resulted in the message "symbol withdrawn, see FGFR3," other searches weren't as fruitful. For example, a search for Yama, a synonym for caspase-3, produced no results. A search for "caspase" yielded a list of all 10 of the human proteins, but none of the entries contained earlier synonyms. And the searches failed to retrieve a single one of the problematic synonyms MRIT, CLARP, Usurpin, FLIP, Casper, FLAME, CASH, and I-FLICE. The database, which contains 7,249 active gene symbols, is available at www.gene.ucl.ac.uk/cgi-bin/nomenclature/searchgenes.pl The site contains explanations of nomenclature rules as well as links to numerous other nomenclature databases.

Creativity, Confusion For Genes

Interested in reading more?

Become a Member of