What Is a Recessive Allele?

WernerG. Heim

ONE of the commonlymisunderstood and misin- are termedthe dominant[dominirende], and those terpreted concepts in elementary genetics is that which becomelatent in the processrecessive [reces- sive]. The expression"recessive" has been chosen of dominance and recessiveness of alleles. Many becausethe charactersthereby designated withdraw students in introductory courses perceive the idea or entirelydisappear in the hybrids,but nevertheless that the dominant form of a gene is somehow stron- reappear unchanged in their progeny ... (Mendel ger than the recessive form and, when they are 1950,p. 8) together in a heterozygote, the dominant allele sup- Two important concepts are presented here: (1) presses the action of the recessive one. This belief is

Statements about dominance and recessiveness are Downloaded from http://online.ucpress.edu/abt/article-pdf/53/2/94/44793/4449229.pdf by guest on 27 September 2021 not only incorrectbut it can lead to a whole series of statements about the appearanceof characters,about further errors. Students, for example, often errone- what is seen or can be detected, not about the ously conclude that because the dominant allele is the underlying genetic situation; (2) The relationship stronger, it therefore ought to become more common between two alleles of a gene falls on a continuous in the course of evolution. scale from one of complete dominance and recessive- There are other common misconceptions, among ness to a complete lack thereof. In the latter case, the them that: expression of both alleles is seen in either of two ways 1) Dominance operates at the genotypic level. in the heterozygote. of domi- 2) Allelic families must have some order The actual situation is, however, somewhat more nance among the members. complicated than Mendel knew. Here, three propo- 3) Dominance is an all-or-none phenomenon. sitions will be briefly analyzed: (1) There exists a full 4) The dominance relationship depends on some spectrum of possible relationships between two al- mysterious, unknown and perhaps unknowable leles of a gene. (2) The nature of the relationship force rather than on molecular and physiological sometimes depends upon the level of analysis used. events subject to exploration and explanation. (3) The cause of the relationship can sometimes be Gregor Mendel recognized that "hybrids," i.e. the analyzed in great detail. For the last of these, we will F1 generation of a cross between two purebred par- return to one of Mendel's cases, that of the wrinkled ents differing in one or more characteristics, are pea. seldom exactly intermediate in appearance between the parents. In some cases, he says: The Spectrum of Relationships [w]ithsome of the morestriking characters, those, for instance,which relate to the form and size of the Generally, textbooks speak of four types of domi- leaves, the pubescenceof the severalparts, &c., the nance-recessiveness relationships:(1) complete dom- intermediate,indeed, is nearlyalways to be seen; in othercases, however,one of the two parentalcharac- inance, (2) codominance, (3) partial or incomplete ters is so preponderantthat it is difficult,or quite dominance, and (4) lack of dominance, although impossible,to detectthe otherin the hybrid.(Mendel some do not differentiateamong all these categories, 1950,p. 7) particularlythe last two. Complete dominance occurs where the of a between two dif- Turning to the results of his own work, he contin- offspring mating strains "resemblesthat of of ues: fering, purebred one the parentalforms so closely that the other either escapes In the case of each of the seven crossesthe hybrid- observation completely or cannot be detected with characterresembles that of one of the parentalforms certainty" (Mendel 1950, p. 7). In more modern so closely that the other either escapes observation terms, there is complete dominance when the phe- completely or cannot be detected with certainty .... [T]hosecharacters which are transmittedentire, or notype of the heterozygote cannot be distinguished almostunchanged in the hybridization,and therefore from that of the dominant parent. A classic example in themselvesconstitute the charactersof the hybrid, is the case of the round and wrinkled peas where the round heterozygotes are as round as the round homozygotes (Mendel 1950). Wemer G. Heim is a professorin the department of biology,The Codominance exists when the phenotypic expres- Colorado College, Colorado Springs,CO 80903. He was asso- ciate editor of ABTfrom 1970 to 1974. sion of both alleles is clearly and separately seen in the heterozygote. A good example here is the human

94 THEAMERICAN BIOLOGY TEACHER, VOLUME 53, NO. 2, FEBRUARY1991 blood type AB, the phenotype jroduced by the designated as ts. Carriers (heterozygotes, genotype simultaneous action of the alleles I and IB. Partialor TS/ts) have usually 40-60 percent of the enzyme incomplete dominance exists if the heterozygote activity of normal, homozygous persons (genotype shows a phenotype more or less, but not exactly, TS/TS)(Bergsma 1979). In everyday life, health, lon- intermediate between the two parental phenotypes, gevity and fertility the heterozygotes cannot be dis- as where pink carnations result from the mating of tinguished from the normal homozygotes. The ho- red flowered with white flowered plants (Klug & mozygotes ts/ts die in infancy or early childhood Cummings 1986). Finally, if the heterozygote's char- (McKusick1988). acter state is precisely intermediate between that of Is, then, Tay-Sachs disease caused by a recessive the two parental characterstates, we may even speak allele? Since the gross phenotypes of the heterozy- of a lack of dominance. The differences among these gotes and homozygote normals cannot be distin- relationships-codominance, partial or incomplete guished, it is so by Mendel's criterion.Yet, the het- dominance, and lack of dominance-are small and erozygotes are easily distinguished from the may not be worth emphasizing in introductory homozygote normals (and from the affected persons) courses. by a relatively simple serum enzyme test (Thompson ideas follow from an understanding & Thompson 1979). Whether the allele yielding the

Two important Downloaded from http://online.ucpress.edu/abt/article-pdf/53/2/94/44793/4449229.pdf by guest on 27 September 2021 of these relationships between pairs of alleles. First, "normal"phenotype is dominant relative to the allele the dominant-recessiverelation may be considered as causing the disease depends upon the level of anal- on a scale from complete to nonexistent (Strickberger ysis. 1985). Second, it is sometimes possible to calculate When alleles of a gene are compared by DNA the degree of dominance exercised by each allele as in nucleotide sequencing, no two alleles are in a reces- the form: sive-dominant relationship. At this level of analysis every heterozygote must differ from either homozy- Degree of Dominance= gote since, by the very definition of "allele," the base (Heterozygous value) - (recessive value) sequences of the two homologous chromosomes must differ in some way. Again, the dominant- (Dominant value) - (recessive value) recessive relationship is not one fundamental to the (Stansfield 1977). As a simple example, consider a nature of genes but depends upon the level of anal- pair of purebred plant lines where one averages one ysis and the degree of detail to which we know the meter in height and the other three. Assume that the genetic situation. One might well argue that the heterozygotes resulting from a cross of these two concept of the relationship disappears entirely at the lines average 2.5 meters in height. The heterozygotes level of the DNA (or RNA) base sequences. are closer in height to the tall parents: The allele producing the greater height exerts 2.5-1/3-1or 3/4 of The Causes of Dominance a complete dominance effect on the height of the heterozygotes. Contrary to the belief among some A heterozygote may display in its phenotype the inexperienced students, the dominance-recessivere- action of one of its pair of alleles for any of several lationship is neither inevitable nor absolute. The reasons. Sometimes the recessive gene may not direct relationship neither has to exist (and often indeed the synthesis of an RNA or protein product; some- does not) nor must the dominant allele always com- times the gene product is defective or markedly low pletely prevent the expression of the recessive one. in activity. In other cases, especially those where the recessive allele, when homozygous, produces the normal phenotype, the dominant allele may be pro- The Level of Analysis ducing a structuralprotein incapable of proper func- Since dominance and recessiveness refer essen- tion by itself or with the product of the recessive tially to an organism's phenotype, we should not be allele. surprised that the relationship may be different at A case of the last type where, for structuralrea- different levels of analysis between phenotype and sons, the abnormalallele acts as a dominant relative genotype. This is well illustrated in the human auto- to the normal allele is illustrated by Ehlers-Danlos somal recessive condition usually called Tay-Sachs syndrome type IV (McKusick1988, entry no. 13005). disease or, more descriptively, GM2-gangliosidosisB The normal allele directs the synthesis of a normal variant (McKusick1988, entry no. 27280). The cause polypeptide. This polypeptide is combined with oth- of this condition is the virtual absence of an enzyme, ers to form the collagen molecule of connective tis- hexoseaminidase A (Hex A) in the affected homozy- sues. The abnormalallele causes the production of an gotes (Thompson & Thompson 1979). The normal abnormal polypeptide that prevents proper forma- allele is conventionally designated as TS; the one tion of the whole collagen protein from the polypep- which, in the homozygous state, causes the disease is tide subunits. Superti-Furga and Steinman (1988)

ALLELE95 have demonstrated a deletion of the DNA coding for fragments in the DNA from plants of either geno- the alpha subunit of one type of collagen in the type. fibroblastsof at least one patient with this condition. That the gene fragment derived from rr plants As might be expected in a heterozygote, both normal would bind to 30 or more sequences in the DNA of and abnormal forms of the corresponding mRNA plants of either genotype indicates that the extra were present. The collagen formed by the assembly piece of DNA in the r allele might be an inserted of correct and incorrect subunits had a reduced fragment, similar or identical to fragments inserted thermal stability, was secreted less efficientlyand was elsewhere in the genome. Indeed, the piece was processed abnormally in the cell. This seems to shown to have inverted tandem repeats and to be explain the unusually extensible and weak connective flanked by a tandem duplication eight base pairs tissue found in this form of the Ehlers-Danlos syn- long, two items commonly found in and around drome. inserted fragments. Sequence analysis of the terminal repeats showed them to be quite similar to those of transposable elements that are capable of becoming The Case of the WrinkledPeas inserted into genes of maize, snapdragon and pars- ley. Downloaded from http://online.ucpress.edu/abt/article-pdf/53/2/94/44793/4449229.pdf by guest on 27 September 2021 Perhaps the most common reason for the recessive However, the size of the inserted element in the r action of one allele relative to another is that an allele allele is too short to carrythe usual apparatusneeded yields either no product or a product with very little for excision and transposition to another gene. The or no activity. Even if a relatively or completely inserted sequence of DNA nucleotides does not ap- inactive protein is produced, it often does not inter- pear to be a complete movable element. Similar fere with the action of a normal protein. Thus the inserted sequences in maize can only move if a heterozygote exhibits only the character due to the full-sized version, constituting a complete transpos- allele directing the production of the functioning able element, is also present on the same chromo- protein. The allele yielding the functioning product is some. But the full-sized version does not appear to completely dominant because its product is the only exist in wrinkled peas, making the insertion essen- one of the two affecting the phenotype. tially a permanent one. (Were it to be excised, a part This relationship between an allele yielding a func- of the tandem repeats would likely be left behind, tional protein and one that does not is well illustrated probably causing a frameshift mutation and leaving by Mendel's (1950) famous case of the wrinkled the allele still nonfunctional). Quite possibly, a part of (genotype rr) and round (genotypes RR or Rr) peas. what was originally a transposable element was lost This case has now been analyzed to an exquisite level by mutation long ago. of detail. The wrinkling of the peas in the recessive We cannot now be certain that Mendel's wrinkled homozygous plants appears to result from a relatively peas had the same genetic arrangement as those high concentration of the sugar sucrose in the devel- studied by Bhattacharyyaet al. (1990), although it oping seed. This, in turn, causes an increased water seems very likely that they did. If so, we have finally content and stretched seed coat. When the usual learned, a century and a quarter later, just what drying occurs, wrinkling results. caused the F1 plants bred from Mendel's pure round The relatively high sugar concentrationis due to a and pure wrinkled stocks to look like their round defect in starch-branchingenzyme I (SBEI) whose parents. More importantly, we now have further normal function is to produce a branch-chainform of justificationfor thinking that the dominant-recessive starch (amylopectin). Some straight-chain starch relationship between alleles does not involve any (amylose) is also produced. When SBEIis deficient, direct interaction between them or, as one of my some sucrose is not converted to starchand an excess students put it, "One allele does not sit on the other accumulates (Fincham 1990). Bhattacharyya et al. and tell it to be quiet." (1990) have shown by cloning that the mRNA tran- scribed from the r allele is longer than that from the R allele and is also less abundant. The differenceis due to an extra 800 DNA base pairs inserted into the r References allele. The sequence with this insert produces a Bergsma, D. (1979). Birth defectscompendium (2nd ed.). New poorly functioning or nonfunctional SBEImolecule. York:Alan R. Liss, Inc. When tested against the genomic DNA from plants Bhattacharyya,M.K., Smith, A.M., Ellis, T.H.N., Hedley, of either homozygous genotype (RR or rr), a particu- C. & Martin, C. (1990). The wrinkled-seed characterof lar segment of the dominant, normal allele hybrid- pea described by Mendel is caused by a transposon-like ized with a single DNA segment. Somewhat insertion in a gene encoding starch-branchingenzyme. surpris- Cell, 60(1), 115-122. ingly, however, the corresponding segment from the Fincham, J.R.S. (1990). Mendel-now down to the molec- recessive allele hybridized with 30 or more short ular level. Nature,343 (6255):208-209.

96 THEAMERICAN BIOLOGY TEACHER, VOLUME 53, NO. 2, FEBRUARY1991 Klug, W.S. & Cummings, M.R. (1986). Conceptsof genetics Strickberger, M.W. (1985). Genetics (3rd ed.). New York: (2nd ed.). Columbus: MerrillPublishing Company, Inc. Macmillian Publishing Company. McKusick, V.A. (1988). Mendelianinheritance in man (8th Superti-Furga, A. & Steinman, B. (1988). Impaired secre- ed.). Baltimore:The Johns Hopkins Press. tion of type III procollagen in Ehlers-Danlos syndrome Mendel, G. (1950). Experimentsin plant hybridization.Cam- type IV fibroblasts: correction of the defect by incubation bridge: Harvard University Press. Translation used by at reduced temperature and demonstration of subtle the Royal HorticulturalSociety of London, published as alterations in the triple-helical region of the molecule. Mendel, G. (1865). Versuche uber Pflanzenhybriden. Biochemicaland BiophysicalResearch Communications, 45, Verhandlungendes naturforschendenVereines in Brunn, 4, 140-147. 1-47. Thompson, J.S. & Thompson, M.W. (1979). Genetics in Stansfield, W.D. (1977). Thescience of evolution.New York: medicine (4th ed.). Philadelphia: W.B. Saunders Com- MacmillanPublishing Company, Inc. pany. Downloaded from http://online.ucpress.edu/abt/article-pdf/53/2/94/44793/4449229.pdf by guest on 27 September 2021

I ! I ! l ! ! !1SoI Ap, V

surces...... th fife scienceBIS...S information always cangng, ture a easily searchabe on the educating high school and college students in this Connection, the online end-user service from| subject area can be a challenging responsibility. BIOSIS.| Updated monthly throughout the academic year,| Luckily,there is a solution: Biology Digest Online! Biology Digest Online providesyour studentsl with the information they need in a format| Biology Digest Online, the online version of the geared to their comprehension level. Learningl printed publication, Biology Digest, from Plexus about the life sciences has never been so interest-l Publishing, enables students to build an aware- ing...... 50.f... ..soeasy! l ness of current developments in all areas of the Call Today! biological sciences through access to informative 1-800-523-4806 (USA except PA)l references and abstracts from worldwide litera- (215) 587-4800 (worldwide)l

I, BIOSIS | Information for Today'sDecisions and Discoveries|

BIOSIS,Marketing Section ABT291BD, 2100 Arch Street, Philadelphia, PA 19103-1399 USA; Fax (215) 587-2016.l BI0515 is a registered trademark of Biological Abstracts.|

ALLELE97