What characteristic is common of both a genetic bottleneck and a founder effect?

Founder Effect

One extreme example of a difference in the incidence of genetic disease among different ethnic groups is the high incidence ofHuntington disease (Case 24) among the indigenous inhabitants around Lake Maracaibo, Venezuela, that resulted from the introduction of a Huntington disease mutation into this genetic isolate. There are numerous other examples of founder effect involving other disease alleles in genetic isolates throughout the world, such as the French-Canadian population of Canada, which has high frequencies of certain disorders that are rare elsewhere. For example, hereditarytype I tyrosinemia is an autosomal recessive condition that causes hepatic failure and renal tubular dysfunction due to deficiency of fumarylacetoacetase, an enzyme in the degradative pathway of tyrosine. The disease frequency is 1 in 685 in the Saguenay–Lac-Saint-Jean region of Quebec, but only 1 in 100,000 in other populations. As predicted for a founder effect, 100% of the mutant alleles in the Saguenay–Lac-Saint-Jean patients are due to the same mutation.

Thus one of the outcomes of the founder effect and genetic drift is that each population may be characterized by its own particular mutant alleles, as well as by an increase or decrease in specific diseases. The relative mobility of most present-day populations, in comparison with their ancestors of only a few generations ago, may reduce the effect of genetic drift in the future while increasing the effect of gene flow.

Abstract

Founder effect is the common outcome of the establishment of new populations from a small number of founding individuals. These founding individuals carry with them only a fraction of the genetic diversity of the parental population, and therefore, the founder effect results with a decreased genetic diversity and distinctive allele frequency patterns in the newly established population. The founder effect can increase the frequency of certain rare disorders, while other disease alleles characteristic of the parental population may disappear. Disease alleles that have negative effect on fitness will be eliminated over time, and eventually, the signature of founder effect can be erased.

Founder effects

Thefounder effect was defined by Ernst Mayr as ‘The establishment of a new population by a few original founders (in an extreme case, by a single fertilised female) which carry only a small fraction of the total genetic variation of the parental population’, and is recognised when a particular polymorphism can be traced back to a single individual.

The reasons for this phenomenon are twofold. First, a particular area may become populated with a small number of individuals, with all subsequent generations originating from these people while the particular population remains isolated(Fig. 5.17). For example, many individuals living in Tristan da Cunha originate from the original British settlement of 15 in 1816, one of whom was a carrier for retinitis pigmentosa, a disease which leads to premature blindness in affected homozygotes and has led to this disease remaining more prevalent in this small island population than elsewhere.

The second reason concerns the origin of a particular set of Y chromosome polymorphisms. The male Y chromosome is passed without change (other than rare mutations) through the generations. Thus, males with the same paternal ancestors are very likely to share identical Y chromosome polymorphism (known as ahaplotype, as there is only one Y chromosome).

Founder Effect

The founder effect is a paucity of genetic variation attributable to small effective population size in a founding population. A small number of individuals are not likely to carry a random sample of the genetic variation in a large, genetically diverse population, so a small group of individuals founding a population can be a unique subset of the variation in the large source population. The founder effect could conceivably be a fortuitous sample, perhaps lacking deleterious, recessive alleles. But a common consequence is for founding populations to have higher levels of one or a few deleterious alleles. For example, a founder effect during the establishment of the Amish community in Pennsylvania, USA, is credited for the elevated frequency of a syndrome characterized by dwarfism and polydactyly.

Regional and Ethnic Distributions of Spinocerebellar Ataxia

SCAs 1, 2, 3, 6, 7, and 8 are most common in the United States and Europe, while geographic predilection of specific SCAs and distinctive founder effects exist in various parts of the world (Fig. 23.6). For example, a high prevalence has been found for SCA1 in Poland; SCA2 in Cuba, Mexico, and Italy; SCA6 in UK, Germany, and Japan; SCA7 in South Africa, Mexico, and Venezuela; SCA10 in Latin America; SCA12 in India and Italy; while SCA3 is the most common SCA worldwide. However, only limited population-based data (Coutinho et al., 2013) exist for incidence and prevalence of SCAs, and estimated frequency of SCAs in a given region is often reflecting founder effects.

Glossary

Founder effect

the random change in genetic composition of a population due to a extreme reduction in its size during a colonization or infection episode.

the random change in the genetic composition of a population due to its finite size. Every population experiences genetic drift but its effects, a reduction in genetic variation eventually leading to fixation of a variant, are more intense, both in magnitude and speed, the smaller its population size.

bacterial strains with an increased mutation rate usually due to a defective mismatch-repair system.

the effect of a gene on two different traits with opposite consequences on fitness.

the set of antibiotic-resistance genes or proteins found in a given environment.

5.3.3 Genomics of founder populations

Due to the founder effect, many variants have drifted to allele frequencies that are significantly different from available population controls. Therefore it is important to genetically study and sample individuals from these populations to create population-specific genomic databases that are informative for these populations. These databases are necessary to provide accurate population-specific allele frequencies for variants to enable appropriate filtering during analyses of genomic data for patients with rare diseases from these populations. Variants with frequencies that are too high to be reasonably considered pathogenic can be more confidently excluded. Simultaneously, these databases enable the identification of healthy control individuals who are homozygous for rare alleles in the population, deeming these variants unlikely to be pathogenic. This is especially true for founder populations where underlying levels of genome-wide homozygosity are high due to shared common ancestry, but also for consanguineous populations that will have large genome-wide homozygous regions due to inbreeding. Furthermore, in cases of pathogenic variants, having a catalog of disease-associated variation in these populations enables rapid, early, and accurate diagnoses that may improve patient outcomes due to informed clinical management and early interventions.

The explosion of population-level genomic data for founder and consanguineous populations has also highlighted new avenues for research beyond disease gene identification. Owing to the presence of drifted alleles, founder populations have provided new insights into variant interpretation and classification which could not have been accomplished previously. The presence of large numbers of heterozygotes for a globally rare allele affords the unusual ability to study the phenotypic impacts of these alleles at unprecedented resolution. For example, an allele in the KCNQ1 gene (c.671C>T, p.Thr224Met) was previously identified in two patients with long QT syndrome, but was still classified as a “variant of unknown significance (VUS)”. The variant is extremely rare in the general population (1 heterozygote in 112,482 European control individuals), but relatively common in the Amish (1 in 45) [49]. The high prevalence of this allele allowed researchers to identify 124 Amish heterozygotes and to perform detailed EKG studies on 88 carriers. The variant was shown to be associated with a 20 ms longer QT interval than the normal allele. This information allowed the variant to be reclassified as a “pathogenic” variant and led to culturally appropriate return of results, including recommendations for treatment and cascade testing.

The pursuit to understand the function of each human gene has been most strongly advanced by the study of Mendelian diseases. The LoF of a gene in a diseased individual informs researchers about the normal function of that gene. A related and complementary method is to identify healthy individuals who are homozygous for rare LoF alleles (so-called “human knockouts”) and determine whether there is an associated phenotype with the natural absence of that gene [44]. In these studies, founder populations, and especially populations with a high level of consanguinity, have helped to elucidate the function of genes that contribute more subtly to human phenotypic variation [50]. It is particularly important to acknowledge that many of these phenotypic effects are actually medically beneficial, such as the finding of LoF variants in LPA in the Finnish population which confer protection against cardiovascular disease or the identification of a CCR5 homozygous LoF variant conferring resistance to HIV infection [51,52]. These studies enable a richer understanding of the genetic contribution to phenotypic variation in humans without relying solely on a classical disease model.

2 Founder effects and linkage disequilibrium

In isolated populations, a founder effect can sometimes be involved in the high prevalence of certain disorders. The term founder effect, proposed by Mayr,18 describes the establishment of a new population by a few original founders, which carries only a small fraction of the total genetic variation of the original population. For example, an allele for a Mendelian disorder can be very rare in the original population, but (by chance) be present at a much higher frequency in the founder population. This can result in abnormally high frequencies of specific disorders in different population isolates after a few generations, often with additional action of genetic drift and inbreeding. This is the case of the Amish population of Lancaster County in Pennsylvania, which presents a high frequency of the autosomal recessive Ellis–van Creveld syndrome.8

A founder effect in a population is suspected when there is an unusually high prevalence of some genetic disorder and/or a very low prevalence of others. When the subjects in the population are genotyped for one of these high prevalence disorders, if they share the same pathogenic variant as well as surrounding genetic variants in a common haplotype, a founder mutation transmitted by a common ancestor is strongly suspected.1,19 The size of the shared haplotype identical by descent from the common ancestor will be inversely proportional to the number of generations to the founder (Fig. 6.1).

The four colored horizontal bars represent four ancestral chromosomes. The star in the first bar represents a founder genetic variant. Two hypothetical populations derived from a small number of founders are represented; a young isolate on the left and an old isolate on the right. The colored bars below represent the different haplotypes present in each kind of isolate, with a reduction in the area in linkage disequilibrium evident for the older isolate.

The contribution of a founder effect for the detection of disease-associated genetic variants varies between Mendelian and complex disorders.5,20 This is in part due to the different population frequencies of both kinds of disorders, and also to the different allele frequencies and effect sizes of risk variants involved. In Mendelian conditions, disease alleles are usually rare in the original population and it is probable that in a small number of founders a single copy of an allele causing monogenic disease is present. The consequence some generations later is that almost all affected individuals will share the same disease-causing allele. For common complex disorders, both common and rare alleles play a role in the genetic architecture of disease.21 It is conceivable that a single copy of a rare risk allele for a common disorder was present in a founding population and it can become common in the isolate because of genetic drift and/or inbreeding. However, detection of the risk variant is affected by the effect size of the variant. Variants that increase risk by a very small amount are difficult to detect. For common risk variants for common disorders, on the other hand, the founding population can have several different risk alleles even if the size is small, and effect sizes of these alleles can potentially be very small—two factors that complicate detection of risk variants.22 However, the argument can be made that even in such a case, the use of isolated populations can be beneficial for the study of complex disorders, because of a reduction of the background genetic diversity in the population.1,23

When linkage disequilibrium (LD) is present in a genomic region, some haplotypes (combinations of alleles along one chromosome) are found more frequently than expected. In isolated populations, this reflects the allele combinations from the founders. As will be further discussed, the occurrence of high LD facilitates identification of genomic regions related to disease. Particularly in young isolates, the extension of LD can be much greater than in outbred populations. Older isolates have been found to show comparable LD to outbred populations.24

In genetic isolates, largely due to the existence of founder effects and high LD, gene mapping can be performed at the population level instead of the family level. Ideal isolates for this purpose are those with a relatively low number of founders, followed by important growth for 10–20 generations, and a current population large enough to find several hundred individuals affected with the disorder of interest.2

INTERPRETATION OF EVIDENCE FROM DIVERSE HUMAN POPULATIONS

It is tempting to try to understand the origins of the instances of paralanguage reported above, and to determine how much is conventional. Others have. Investigators like Birdwhistell and LaBarre used cultural differences in facial paralanguage to assert a cultural relativism and to mitigate a phyletic contribution. I consider this dispositiveness premature. As I have tried to make clear for nonparalinguistic facial displays, regional differences in facial paralanguage do not imply that it is conventional (i.e., epigenetic). Determination of what is “nature” and what is “nurture” in facial paralanguage is just as labyrinthine, and any attempt must consider the following:

1.

Genetic drift or founder effects may have produced variations in facial paralanguage among diverse cultures. As I indicated in Chapter 11, these are strictly genetic mechanisms, but they can produce quite pronounced variations in local populations that masquerade as “cultural.“

2.

Population comparisons of facial paralanguage that find regional variation must still control for language. Some paralanguage may be universally, innately coupled to certain aspects of language, but its manifestation would depend entirely upon the existence or nonexistence of those aspects of the language in the given population. This point can be clarified with an example. Conceivably, humans might be “prewired” to display a head tilt with eyebrow raise whenever they use the subjunctive mood; the action might occur as an illustrator during speech, or as an emblematic conversational response (symbolizing the sentence, “It could be true.“). What if we found wide variations in usage of this display across geographic regions or cultures? It would be presumptuous to consider this a variation in display, because it might as well represent a difference in language. Populations that did not show the display might, for example, lack the subjunctive mood for demarcating counterfactuals (e.g., putatively the Chinese; see Au, 1983; Bloom, 1981). Conversely, a population may lack the language for counterfactuals and depend upon paralanguage to denote them. And we simply do not know what constitutes the innate, “deep structure” of facial paralanguage. It may be coupled with phonological, syntactic, and/or semantic features of language, and it may be “released” with the emphatic or prosodic features of particular dialects.

3.

Even if facial actions like emblems appear wholly symbolic, it cannot be assumed that they are simply conventions. As I illustrated in the case of the hypothetical subjunctive-illustrator, and in my earlier discussion of display rules, rapid cultural evolution does not exclude phyletic contribution. To take a sample emblem, Elvis Presley's cinematic sneer while speaking to female costars became a teenage American male flirtation display nearly instantaneously in the 1960s, although its use in contempt (and derision, scorn, etc.) was an excellent preadaptation. Are we to conclude that this emblem was entirely conventional (i.e., a learned token)? This position is defensible only if it is demonstrated that attitudes females found attractive in males did not change. Instead, the sneer may be entirely canalized in displaying contempt, derision, and so on, and Elvis Presley induced males totake this attitude toward females. Ultimately, had the sneer persisted culturally, males genetically predisposed to sneering should have greater reproductive advantage. Furthermore, morphological change (a lip structure conducive to sneers) might become weakly selected for, but detecting such selection would take many generations.

4.

The above three dicta urge against the disqualification of a genetic basis for facial paralanguage even amid population variations in it. Conversely, communalities in facial paralanguage among human populations—or resemblances among human paralinguistic and nonhuman paravocal actions—do not guarantee a phyletic contribution. This is so for the same arguments cited in my discussion of “facial expressions of emotion.” As before, it is necessary to exclude cultural transmission, convergent evolution, and common learning. As in those cases, both nonhuman and human ontogenetic data are crucial, even though the ideal human experiments (i.e., blind children reared and taught to speak solely by blind caretakers) are wildly improbable and perhaps ill advised.

These are just cautions. It is too early to determine just which elements of facial paralanguage are under genetic control and which are conventional. We know far too little about human facial paralanguage in even one population, much less across humanity in toto. And as I indicated with respect to putative “display rules,” ethnographic and linguistic perspectives must come to bear before the existence or nonexistence of any facial paralanguage is taken to implicate genetic versus epigenetic contributions.

Genetic Drift

Although there is debate about whether drastic founder effects and population bottlenecks enable dramatic adaptive peak shifts triggering speciation (Templeton, 2008), this does not mean that genetic drift does not sometimes play an important role in population divergence. As neutral or near neutral differences accumulate between populations in allopatry they may create large and numerous depressions of decreased fitness in the adaptive landscape in hybrids due to B–D–M incompatibilities. This is because mutations that are neutral or nearly neutral within demes are not necessarily inconsequential and may cause significant postzygotic isolation between populations when mixed in hybrids. As a result, as many slight differences accumulate between allopatric populations, a highly convoluted and holey adaptive landscape can exist for individuals of mixed ancestry, generating a high degree of RI and severely restricting gene flow genome wide between taxa (Gavrilets, 2004). In addition, during early stages of speciation with gene flow, the initial build up genetic divergence prior to a tipping point being reached and genomic hitchhiking being enabled may have dynamics that are equivalent to genetic drift, as many selected sites may be only slightly favored and behave as if they are neutral until enough differences have accumulate to reduce the effective migration rate between populations genome wide (Flaxman et al., 2013; Feder et al., 2014).