Jump to content
Toggle menu
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Pleiotropy is the effect of one locus on two or more distinct traits.[1] The word is used both for a gene and for a single mutation, which are not always the same case: a mutation may alter only some of the functions of a gene that has several.[2] Pleiotropy and close linkage are the two main causes of genetic correlations between traits. Where such a correlation exists, breeding for one trait changes another with it.[3]

Mapping studies in Cannabis place several traits at one position on the genome. In a cross of two hemp cultivars, most quantitative trait loci (QTLs) fell into a few clusters. The authors ascribed much of the difference between the parents to a few pleiotropic loci of large effect.[4] In 145 samples of Iranian landraces, 15 of the 91 loci found were associated with several traits, among them flowering time, plant form, biomass and cannabinoid content.[5] Which gene acts on which trait at these positions remains largely open.

One locus or two close together

The third character in Gregor Mendel's work on peas combined three features. Plants had either a brown seed coat, violet flowers and spots in the leaf axils or a white seed coat, white flowers and no spots. Mendel put all three down to one factor.[1] That factor, the gene A, encodes a transcription factor that determines anthocyanin pigmentation throughout the plant. The white-flowered allele Mendel most likely used carries one changed base at a splice site.[6]

Traits that consistently segregate with one mutation are classed as pleiotropic, but genes lying close together on a chromosome also segregate together. The two cases can be hard to tell apart.[1] At the level of a QTL they are often confounded.[3] In 1995 Jiang and Zeng set out a formal test that compares one pleiotropic QTL at a position with several nearby QTLs that each affect one trait.[7] In the wild oat Avena barbata, where loci for separate traits mapped to similar positions, such tests mostly supported pleiotropy.[8]

A correlation caused by linkage is expected to break under recombination, while one caused by pleiotropy is expected to hold, and breeders want to know which they have. Annalise Paaby and Matthew Rockman point out that both can change. Modifier alleles, mutation and gene duplication can alter pleiotropy. Linkage can be tight enough to resist recombination. In their view the weight of the distinction for evolution varies from case to case.[2]

Direct effects and chains

In 1938 Hans Grüneberg separated "genuine" pleiotropy, two primary products from one locus, from "spurious" pleiotropy, in which a single product is used in different ways or sets off a cascade of consequences. The one gene, one enzyme hypothesis of the 1940s left no room for the first kind. Research turned to how one product could have several uses. Ernst Hadorn's terms of 1961 were mosaic pleiotropy for a locus acting directly on two traits and relational pleiotropy for a locus starting a cascade. Sequencing from the late 1970s then showed that one locus can make several products, through overlapping reading frames, alternative splicing and RNA editing.[1]

Later authors call the direct and the cascading forms horizontal and vertical pleiotropy.[2] Solovieff and colleagues distinguish biological pleiotropy, a direct influence on more than one trait, from mediated pleiotropy, where a variant reaches the second trait through the first. They keep "spurious" for an association that only appears to touch several traits, through biased sampling or misclassified traits, which is a different sense from Grüneberg's.[9] In a chain, environmental and genetic disturbances of the first trait both reach the second.[2] In plants, flowering time often varies together with height, seed yield and quality, leaf number, cell walls and secondary growth.[10]

What counts as a trait

A diagram with three circles labelled G1, G2 and G3 in a row above three circles labelled P1, P2 and P3. Arrows run from each upper circle down to two of the lower ones, and P2 receives three arrows.
A simple map from genes (G) to phenotypes (P), in which each gene acts on two traits and one trait depends on all three genes. Diagram by Alphillips6.

Paaby and Rockman separate three questions that share the word. Molecular-gene pleiotropy concerns how many functions a gene has, whether counted by the traits a knockout alters or by the proteins it interacts with. Developmental pleiotropy concerns how many parts of the phenotype a mutation changes, whatever the effect on fitness. Selectional pleiotropy concerns how many components of fitness a mutation affects. The three can come apart. A mutation in a multifunctional gene may alter one function only. A mutation with one phenotypic effect may help in one environment and harm in another.[2]

Across five data sets from yeast, a nematode and the mouse, Wang and colleagues found a median of one to seven traits per gene (1 to 9 per cent of the traits considered), with genes and traits falling into modules.[11] Mapping of 70 skeletal traits in mice gave 102 QTLs, each affecting 7.8 traits on average and at most 30. In sticklebacks, 26 QTLs for 54 body-shape traits averaged 3.5.[2] Such figures favour modular pleiotropy over the "universal" pleiotropy that Sewall Wright and others assumed.[1] William Hill and Xu-Sheng Zhang showed that models with effects on every trait at every locus can give similar counts. The conclusion depends on statistical assumptions such as the significance level.[12] When lethal deletions are counted as affecting every trait, the distribution in yeast has a second peak at the maximum. Paaby and Rockman rank limited statistical power above any upward bias from reading linked loci as one.[2]

Selection on correlated traits

In Lande's model of 1980, for large randomly mating populations, pleiotropy is the main cause of genetic correlation between characters when selection on polygenic variation is weak in relation to recombination.[13] A few pleiotropic loci of large effect typically induce strong correlations, while the effects of many small ones may cancel out. Heritability predicts the expected response to selection in a trait. If a locus under selection has other effects that lower fitness, the amount by which the response falls short measures the extent of pleiotropy. In some cases there is no response at all.[2]

Ghd7 in rice has major effects on heading date, plant height and grains per panicle. Natural alleles of reduced function allowed rice to be grown in cooler regions.[14] Engineered mutations in tomato SP5G cause rapid flowering and a more compact habit together, giving an early yield.[15] Isogenic soybean lines with early-flowering alleles were often shorter, lodged less and matured early, but gave less seed. Earlier varieties accumulate less biomass in stem and leaf, so less carbon is available for seed. Overexpression of another gene, GmAGL1, brought early flowering without that loss.[10]

Mutations in cis-regulatory sequence often affect only part of a gene's expression, while a change in the protein applies wherever it is expressed.[2] Duplicate genes may divide the effects: in maize, zfl1 is more strongly associated with flowering time and zfl2 with branching and inflorescence structure.[16]

Effects on fitness

Under antagonistic pleiotropy, one allele raises one part of fitness and lowers another. G. C. Williams proposed in 1957 that genes favourable before reproduction and harmful after it could explain senescence.[1] In Avena barbata, alleles that hastened first flowering generally reduced vegetative mass, increased reproductive mass and were associated with high fitness.[8] The trade-off may also fall between environments, as in local adaptation. It may fall between the sexes where one mutation has different consequences for the fitness of each.[2] Up to 2011, conditional neutrality, an allele that matters in only one environment, had more documented plant examples than antagonistic pleiotropy between environments.[17] Later field work in Arabidopsis found trade-offs at a third of 15 fitness QTLs.[18]

In Fisher's model of adaptation, complex organisms pay a cost. Orr calculated that the rate of adaptation falls at least as fast as the inverse of the number of independent characters, for mutations of the same phenotypic size.[19] Wang and colleagues found that genes affecting more traits have larger effects on each, which substantially lessens that cost.[11]

Measured in cannabis

Two mapped crosses

Ten vertical bars for the cannabis chromosomes, with coloured blocks beside them marking mapped loci. Solid blocks for plant traits cluster on chromosomes 2 and 9. Striped blocks for seed traits sit on chromosomes 1, 3, 4 and 5.
Major QTLs in an F2 of SI-1 and a Syrian landrace accession. Plant traits (solid) cluster on chromosomes 2 and 9, seed traits (striped) lie elsewhere. From Manansala-Siazon and colleagues (2025).

Patrick Woods and colleagues mapped 372 second-generation lines from a cross of the tall, late fibre cultivar Carmagnola with the short, early USO31, grown as clones in one field in Colorado in 2017. Most of the 34 QTLs for agronomic traits fell into four clusters. Each model for the eight agronomic traits held a QTL on linkage group 3 and one on linkage group 5, explaining up to 22 and 34 per cent of the variance in a trait. The authors read this as a few pleiotropic loci of large effect. They note that the QTLs of most traits sum to about half of the variance. In the cluster on group 5, a narrow interval held a homologue of the gene TINY with one amino-acid change. Genotypes at that site differed for all eight traits, which the authors offer as a hypothesis.[4]

Two clusters on groups 6 and 9 held QTLs for both cannabinoids and terpenes. At the group 9 cluster the USO31 allele of olivetol synthase was less active when expressed in yeast, which fits the lower cannabinoid content of plants carrying it. It does not explain why those plants generally made less of each terpene too. Woods and colleagues suggest either an unknown interaction between the two pathways or a pleiotropic regulator of both. Agronomic and biochemical QTLs did not share positions in this cross. The traits of the two classes were not significantly correlated.[4]

In an F2 of the hemp accession SI-1 and a Syrian landrace accession, seed size and plant size were not correlated and did not share QTLs, the opposite of the seed mass result in Carmagnola × USO31. Manansala-Siazon and colleagues point to genetic background or growing conditions. Trichome density in this cross shared a cluster on chromosome 2 with agronomic traits while showing little or no correlation with plant height.[20]

Iranian landrace panel

Mehdi Babaei and Davoud Torkamaneh classed 15 of 91 loci, marked by 17 markers, as pleiotropic.[5]

Percentage of phenotypic variance explained by markers at three loci in 145 pooled samples of Iranian landraces
Locus (chromosome) Flowering stages Plant form Biomass Cannabinoid
CsFTL3 (8), three markers in a block of 3 kilobases 19.9 to 39.3, six stages Main-stem nodes 35.0; nodes to the main inflorescence 34.3; height 28.5; stem diameter 19.6; number of lateral shoots 7.8 Stem dry weight 46.9; stem fresh weight 24.1; total fresh weight 13.9; total dry weight 11.4 Cannabidivarin 82.1
CsCFL1 (9), one marker 24.3 to 41.4, four stages Longest shoot 52.7; shortest shoot 54.4; height 47.9; stem diameter 57.8; number of lateral shoots 47 Total fresh weight 30.5; stem fresh weight 20.4; total dry weight 13.1 none listed
CsFTL4 (7), one marker 13.3 and 30.6, two stages none listed Total dry weight 36.0; stem dry weight 32.4; total fresh weight 24.3; stem fresh weight 13 none listed[5]

At CsCFL1, plants of genotype CC reached 50 per cent flowering at 121 days against 79 for AA and stood 161 cm tall at harvest against 85 cm. Babaei and Torkamaneh take flowering loci of this kind to act high in a regulatory chain, with effects that cascade to plant form, growth and biomass. Separate markers in the CsFTL3 block carry different traits. One of them, Cs1, also reaches cannabidivarin. Its AA genotype had six times the stem dry weight of CC (51.73 against 8.41 g), while CC had eight times the cannabidivarin of AA (0.0097 against 0.0012 per cent by weight).[5] They also call CsFTL3 Autoflower3. In a later paper from the same group the block occurs only in auto-flowering landraces.[21]

Some of the counted traits are sums of others. Total weight adds stem, leaf and flower. Nodes on the main stem add those on and below the main inflorescence.[5] An effect on a part can therefore appear again in the total.

Other results

Plants of the family GVA-H-20-1080 differed by genotype at Autoflower1 in flowering date, height and wet biomass, with heterozygotes between the homozygotes.[22] Early flowering in hemp shortens the vegetative phase in which biomass accumulates.[22]

Association mapping of bast fibre composition placed QTLs for glucose and lignin on one genomic scaffold and those for xylose and glucuronic acid on another. The last two are both components of xylans.[23] Pancaldi and colleagues regard fibre yield and quality as complex traits under loci and genes with intricate pleiotropic effects.[24]

Aardema and DeSalle took phenotypes for 845 strains from public online databases and found 29 of 94 variants significant in more than one category of trait, which they call possible pleiotropic effects.[25]

The word

The German Pleiotropie was coined by Ludwig Plate in 1910, in a volume in honour of Richard Hertwig. He called a unit of inheritance pleiotropic when several characters depend on it, so that they always appear together and may seem correlated. Haecker described the same mechanism in 1925 under another name, but Plate's term was by then established.[1]

See also

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 Stearns, F.W. (2010). "One hundred years of pleiotropy: a retrospective". Genetics. 186 (3): 767–773. doi:10.1534/genetics.110.122549.
  2. ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 Paaby, A.B.; Rockman, M.V. (2013). "The many faces of pleiotropy". Trends in Genetics. 29 (2): 66–73. doi:10.1016/j.tig.2012.10.010. PMC 3558540.
  3. ↑ 3.0 3.1 Chen, Y.; Lübberstedt, T. (2010). "Molecular basis of trait correlations". Trends in Plant Science. 15 (8): 454–461. doi:10.1016/j.tplants.2010.05.004.
  4. ↑ 4.0 4.1 4.2 Woods, P.; Campbell, B.J.; Nicodemus, T.J.; Cahoon, E.B.; Mullen, J.L.; McKay, J.K. (2021). "Quantitative trait loci controlling agronomic and biochemical traits in Cannabis sativa". Genetics. 219 (2) iyab099. doi:10.1093/genetics/iyab099.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 Babaei, M.; Torkamaneh, D. (2026). "Genetic architecture of phenological, morphological, and phytochemical traits in Cannabis landraces". The Plant Genome. 19 (2) e70243. doi:10.1002/tpg2.70243.
  6. ↑ Hellens, R.P.; Moreau, C.; Lin-Wang, K.; Schwinn, K.E.; Thomson, S.J.; et al. (2010). "Identification of Mendel's white flower character". PLoS ONE. 5 (10) e13230. doi:10.1371/journal.pone.0013230.
  7. ↑ Jiang, C.; Zeng, Z.B. (1995). "Multiple trait analysis of genetic mapping for quantitative trait loci". Genetics. 140 (3): 1111–1127. doi:10.1093/genetics/140.3.1111.
  8. ↑ 8.0 8.1 Latta, R.G.; Gardner, K.M. (2009). "Natural selection on pleiotropic quantitative trait loci affecting a life-history trade-off in Avena barbata". Evolution. 63 (8): 2153–2163. doi:10.1111/j.1558-5646.2009.00701.x.
  9. ↑ Solovieff, N.; Cotsapas, C.; Lee, P.H.; Purcell, S.M.; Smoller, J.W. (2013). "Pleiotropy in complex traits: challenges and strategies". Nature Reviews Genetics. 14 (7): 483–495. doi:10.1038/nrg3461. PMC 4104202.
  10. ↑ 10.0 10.1 Salentijn, E.M.J.; Petit, J.; Trindade, L.M. (2019). "The complex interactions between flowering behavior and fiber quality in hemp". Frontiers in Plant Science. 10 614. doi:10.3389/fpls.2019.00614.
  11. ↑ 11.0 11.1 Wang, Z.; Liao, B.Y.; Zhang, J. (2010). "Genomic patterns of pleiotropy and the evolution of complexity". Proceedings of the National Academy of Sciences. 107 (42): 18034–18039. doi:10.1073/pnas.1004666107.
  12. ↑ Hill, W.G.; Zhang, X.S. (2012). "On the pleiotropic structure of the genotype–phenotype map and the evolvability of complex organisms". Genetics. 190 (3): 1131–1137. doi:10.1534/genetics.111.135681.
  13. ↑ Lande, R. (1980). "The genetic covariance between characters maintained by pleiotropic mutations". Genetics. 94 (1): 203–215. doi:10.1093/genetics/94.1.203.
  14. ↑ Xue, W.; Xing, Y.; Weng, X.; Zhao, Y.; Tang, W.; et al. (2008). "Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice". Nature Genetics. 40 (6): 761–767. doi:10.1038/ng.143.
  15. ↑ Soyk, S.; Müller, N.A.; Park, S.J.; Schmalenbach, I.; Jiang, K.; et al. (2017). "Variation in the flowering gene SELF PRUNING 5G promotes day-neutrality and early yield in tomato". Nature Genetics. 49 (1): 162–168. doi:10.1038/ng.3733.
  16. ↑ Bomblies, K.; Doebley, J.F. (2006). "Pleiotropic effects of the duplicate maize FLORICAULA/LEAFY genes zfl1 and zfl2 on traits under selection during maize domestication". Genetics. 172 (1): 519–531. doi:10.1534/genetics.105.048595.
  17. ↑ Anderson, J.T.; Willis, J.H.; Mitchell-Olds, T. (2011). "Evolutionary genetics of plant adaptation". Trends in Genetics. 27 (7): 258–266. doi:10.1016/j.tig.2011.04.001. PMC 3123387.
  18. ↑ Ågren, J.; Oakley, C.G.; McKay, J.K.; Lovell, J.T.; Schemske, D.W. (2013). "Genetic mapping of adaptation reveals fitness tradeoffs in Arabidopsis thaliana". Proceedings of the National Academy of Sciences. 110 (52): 21077–21082. doi:10.1073/pnas.1316773110. PMC 3876199.
  19. ↑ Orr, H.A. (2000). "Adaptation and the cost of complexity". Evolution. 54 (1): 13–20. doi:10.1111/j.0014-3820.2000.tb00002.x.
  20. ↑ Manansala-Siazon, S.E.; Siazon, P.M.; Tandayu, E.; Garcia-de Heer, L.; Burn, A.; et al. (2025). "Seed the difference: QTL mapping reveals several major loci for seed size in Cannabis sativa L." Plants. 14 (24) 3853. doi:10.3390/plants14243853. PMC 12737016.
  21. ↑ Babaei, M.; Nemati, H.; Arouiee, H.; Torkamaneh, D. (2026). "Integrating temporal morphophysiological and genomic markers for precise classification of flowering time in cannabis". Scientific Reports. 16 (1) 23150. doi:10.1038/s41598-026-53686-y.
  22. ↑ 22.0 22.1 Toth, J.A.; Stack, G.M.; Carlson, C.H.; Smart, L.B. (2022). "Identification and mapping of major-effect flowering time loci Autoflower1 and Early1 in Cannabis sativa L." Frontiers in Plant Science. 13 991680. doi:10.3389/fpls.2022.991680.
  23. ↑ Petit, J.; Salentijn, E.M.J.; Paulo, M.-J.; Denneboom, C.; van Loo, E.N.; et al. (2020). "Elucidating the genetic architecture of fiber quality in hemp (Cannabis sativa L.) using a genome-wide association study". Frontiers in Genetics. 11 566314. doi:10.3389/fgene.2020.566314. PMC 7527631.
  24. ↑ Pancaldi, F.; Salentijn, E.M.J.; Trindade, L.M. (2025). "From fibers to flowering to metabolites: unlocking hemp (Cannabis sativa) potential with the guidance of novel discoveries and tools". Journal of Experimental Botany. 76 (1): 109–123. doi:10.1093/jxb/erae405.
  25. ↑ Aardema, Matthew L.; DeSalle, Rob (23 February 2021). "Can public online databases serve as a source of phenotypic information for Cannabis genetic association studies?". PLOS ONE. 16 (2) e0247607. doi:10.1371/journal.pone.0247607.