Most documented interference has been positive, but some reports of negative interference exist in experimental organisms. The investigation of interference is important because accurate modeling of interference will provide better estimates of true genetic map length and intermarker distances, and more accurate mapping of trait loci. Interference is very difficult to measure in humans, because exceedingly large sample sizes, usually on the order of three hundred to one thousand fully informative meiotic events, are required to detect it.
Bibliography Strachan, Tom, and Andrew P. Human Molecular Genetics. Subhashree Subhi. ZaRminaa Khan. Shankar G. Refat Khan. Show More. Views Total views. Actions Shares. No notes for slide. Interference 1. Interference shown in this diagram using the model of meiotic recombination.
One crossover reduces the chance of another crossover in the adjacent region. One crossover enhances the chance of another crossover in the adjacent region. Mostly in eukaryotes 2. Found in some lower organisms 3. In this case coefficient of coincidence is less than one.
In this case coefficient of coincidence is always more than one. Genotype No. Total views 10, On Slideshare 0. From embeds 0. Number of embeds 1. Downloads Shares 0. Comments 0. Closely spaced COs are observed less frequently than would be expected from a random distribution.
This phenomenon is known as crossover interference — though the more general term genetic interference may be more useful since there is growing evidence described below that other events can also interfere with one another. Alfred H. Sturtevant and Hermann J. Muller are typically given equal billing for the discovery of interference. Sturtevant clearly describes the phenomenon as early as [ 2 ].
Although interference was originally observed almost a century ago and has subsequently been validated in numerous studies, fundamental questions regarding its underlying mechanisms still exist. The goal of this article is to outline unanswered questions about interference and also to review existing models. The genomic distribution of COs is regulated in multiple ways.
A random distribution of COs among chromosomes predicts a class of chromosomes that have no COs, yet the observed number of chromosomes without a CO is quite small in most organisms [ 4 , 5 ]. Growing evidence suggests that interference is the result of multiple levels of recombination regulation and CO assurance could be a result of the interference mechanism.
Meiotic DSBs are enzymatically catalyzed by a topoisomerase I-like protein called Spo11 that remains covalently attached to 5' ends of the break Fig. Following Spo11 removal and further end processing resection , the breaks are left with single-stranded 3' tails. One of these tails can then invade a non-sister chromatid, which is known as strand invasion. Stabilized strand invasion intermediates are known as single-end invasion SEI intermediates.
The DNA synthesis that occurs at this stage is primed by one chromatid but uses a non-sister chromatid as a template — therefore any polymorphisms that exist at this locus will be copied from the template chromatid to the invading chromatid.
This transfer of parental information is called gene conversion GC [ 6 , 7 ]. At this point the invading end can dissociate from the non-sister chromatid and re-associate with the other end of the break in a process called synthesis dependant strand annealing SDSA [ 8 ].
After additional DNA synthesis and ligation the break is repaired resulting in a non-crossover NCO , potentially with associated GC if a polymorphism existed at the locus. Alternatively, the mismatch repair system can fail to recognize the mismatch. In this case, both genotypes in the heteroduplex will be propagated during the next mitotic division in a process called post-meiotic segregation PMS. PMS results in two populations of cells, one that has experienced GC at the polymorphic site and the other that has not.
Single strands of DNA are shown as either green parent 1 or yellow parent 2 rods. The Spo11 complex initiates programmed DSBs. Also shown is a pathway center describing aberrant JMs, that are hypothesized to either be resolved back to the strand invasion stage by Sgs1 or resolved as COs by the MusEme1 heterodimer [ 52 ].
Prior to stabilization, Sgs1 could wire SEIs back to the strand invasion stage [ 52 , 89 ]. The displaced strand, called a D-loop is captured by the resected break of opposite homolog and subsequent DNA synthesis results in a dHJ intermediate. This intermediate is resolved as a CO upon appropriate resolution of the two HJs. If SDSA does not occur, then as the D-loop is extended it can hybridize to the single-stranded 3' tail on the other side of the break in a process called second end capture Fig.
Again, this structure can be acted on by DNA polymerase, which extends the second single-stranded 3' tail.
As before, the priming and template DNA are from non-sister chromatids, therefore GC can occur during this stage. In principle, this structure can be resolved to produce either a crossover CO or NCO depending on how the individual junctions are cut to release the chromatids. However, evidence in Saccharomyces cerevisiae suggests that these intermediates are resolved predominantly as COs [ 4 , 9 , 10 ].
This marked the beginning of the debate as to the best way to measure interference, which is still being argued to this day [ 11 ]. Data generated in Drosophila dominated advancements in interference research for the next 40 years [ 12 - 15 ].
Nonetheless, several key observations were made. Muller first proposed that interference does not act between different chromosomes, i. Subsequent research by Weinstein established the reach of interference on the X chromosome. In this case, interference does not manifest when the distance between the two intervals being studied is greater than 46 cM [ 14 ]. Additionally, COs in intervals on the same chromosomal arm interfere more strongly than intervals on opposite arms that have approximately equivalent genetic distances [ 14 ].
This introduced the idea that interference does not cross the centromere. In , Graubard observed that chromosome 2 carrying an inversion max size 25 cM did not affect interference values of intervals on that chromosome [ 15 ]. This result was the first to suggest that pre-existing chromosomal features are not the primary determinant of interference, which is widely believed to this day. Model fungal organisms that form tetrads fused meiotic products , allow for both the recovery of non-reciprocal recombination products and novel statistical approaches for interference analysis [ 16 - 18 ].
Non-Mendelian segregation the result of post-meiotic mitotic division of a tetrad at a single locus, indicative of a GC, was first observed in spore pigmentation mutants of Bombardia lunata [ 18 ]. These observations led researchers to question if GCs were born from the same mechanism that produces COs.
If true, this would predict that they should exhibit interference. Using Neurospora crassa as a model system, Stadler tested this idea by seeing if GC events interfered with the probability of COs in an adjacent interval. These data suggested that all recombination events have a common molecular basis which we now know to be DSBs and strand invasion [ 6 , 7 ] and that initial events are distributed independently of one another, but that the occurrence of a CO at one site will subsequently influence nearby events to be resolved as NCOs, thus resulting in a CO distribution that displays interference [ 20 ].
Determining what combinations of recombination events are subject to interference is key to understanding both when interference is imposed as well as the mechanism that mediates it. Interestingly, in many organisms, not even all COs are subject to interference discussed below. However, those COs that do interfere could theoretically be a reflection of an inhibition of closely spaced DSBs. Whether or not, and to what degree DSBs interfere with one another remains an open question.
Meiotic DSB mapping studies in S. However, in S. Ohta et al. This led to the proposal that strong DSBs sites could outcompete nearby sites for limiting factors essential for DSB formation.
Another proposal is that structural features that influence DSB formation create domain boundaries that isolate regions from one another [ 25 ]. DSB distribution has also been addressed via mapping the physical position of structures called recombination nodules RNs. RNs are proteinaceous structures that are associated with the axial elements of the synaptonemal complex SC from leptotene to pachytene and are thought to be locations where meiotic recombination reactions are occurring [ 27 ].
RNs are divided into two sub-classes: early nodules ENs and late nodules LNs , which differ in respect to timing, size, shape and number. In tomato, both ENs and LNs exhibit a distribution indicative of interference, but the strength of interference is much stronger among LNs [ 30 , 31 ].
Analysis of the distribution of MSH4 foci in mouse meiocytes strongly supports the idea that DSBs interfere with one another. Early MSH4 foci exhibit a distribution indicative of interference, but do not exhibit as strong interference as MLH1 foci [ 31 , 35 , 36 ].
Taken together the above results argue that interference is the result of multiple layers of control [ 31 , 32 , 36 ]. Another observation that suggests interference is not merely a reflection of an underrepresentation of closely spaced DSBs is that in S. More recently, analysis of genome-wide recombination maps based on DNA tiling arrays reaffirmed that the distance between GCs not associated with COs does not differ significantly between experimental samples and randomized control data [ 37 ].
The first results to address this issue came from S. However, they did not observe statistically significant negative interference between GCs not associated with CO at met13 and COs.
Recently, Getz et al. However, most recently, a contradictory result showing that NCOs and COs interfere with one another was recently presented by Mancera et al. While locus-by-locus studies have consistently shown negative or no interference between NCOs and COs, this genome-wide approach showed the opposite. It is difficult to reconcile these contradictory results, but because the Mancera data were generated using a genome-wide analysis the advantage appears to be with the idea that COs and NCOs exhibit positive interference.
One reconciliatory possibility is that there exist both interfering NCOs and non-interfering NCOs as in COs, see below and the single locus studies mentioned above happened to measure only the latter class.
In many organisms, there are at least two pathways for producing COs. Arabidopsis thaliana , humans, mouse and S. In these organisms, primary pathway COs are characterized by the Msh4-Msh5 heterodimer while secondary pathway COs are dependent on the MusEme1 heterodimer [ 40 ].
Not all organisms produce both interfering and non-interfering COs. Haploid chromosome number n and presence or absence of CO interference is noted.
Also shown are presence or absence of Msh4-Msh5 interference-sensitive and MusEme1 interference-insensitive mediated CO pathways.
While it is known that both interfering and non-interfering COs can be produced in the same cell, the mechanistic difference between the two and specifically why one class exhibits interference and the other does not is unknown.
One possibility is based on the work in Arabidopsis by Franklin et al. This could result in interfering and non-interfering COs and perhaps non-interfering NCOs if the Msh4-Msh5 pathway is subject to interference, but the smaller and randomly distributed population of aberrant JMs resolved as COs by MusEme1 acts later, after interference has already been established. In congruence with this idea, recent biochemical and genetic analysis of mus81 mutants in S. In this model, secondary COs do not produce an interference signal, and are resolved after the interference signal has been imposed.
An alternate hypothesis regarding the difference between interfering and non-interfering COs was introduced by Getz et al. In addition to being independent of MSH4 , pairing phase COs are hypothesized to be less proficient at repairing mismatches.
This model is also compatible with the phenotype of ndj1 mutants, which have decreased interference and increased rates of nondisjunction [ 57 , 58 ], which can be explained by increased pairing COs at the expense of disjunction phase COs. Since pairing phase COs are proposed to be interference insensitive it follows that they should be MUS81 -dependent, which predicts that mus81 and eme1 mutants will be pairing defective.
In yet another layer of complexity the toolbox and two-phase hypotheses are not mutually exclusive as there could be pairing phase non-interfering COs that have nothing to do with MUS81 as well as non-interfering disjunction phase COs produced by MUS81 that are reflective of MSH4 -independent aberrant JM resolution. Determining the timing of events that lead to interference is extremely challenging since diverse cellular processes likely play a role.
Chromatin structure e. These features are not constant along chromosomes and are dynamic in both the mitotic and meiotic cell cycles. Meiotic chromosome condensation, which begins at the start of meiotic prophase and does not end until after recombination is complete, also likely influences interference [ 1 , 29 , 59 ].
However, pre-recombination chromosomal features are not the only important determinants. This strongly suggests that the assembly and distribution of recombination complexes is critical for the timing of the imposition of interference. One attractive proposal is that interference is imposed during strand invasion when Msh4-Msh5 complexes stabilize CO-specific SEI recombination intermediates [ 60 ]. This idea is based on the observations that msh5 ndt80 mutants result in very low levels of JM accumulation along with absence of interference in S.
The ndt80 mutation was used in this case because it removes the late pachytene checkpoint and results in accumulation of recombination intermediates [ 60 ]. The spo16 mutant also offers important insight as to the latest interference could be acting.
Supporting this idea is the observation that SC initiation complexes exhibit a distribution indicative of interference [ 62 ]. Additional support for the idea that interference involves regulation of the strand invasion step comes from analysis of the tid1 mutant in S. The timing of events leading to interference is different in Drosophila and C. Thus, interference in these organisms is likely implemented after though not dependent on SC formation.
How this decision is enforced is unknown. ZMM mutants are strongly CO defective, have abolished or greatly reduced interference, yet are not defective for NCO formation [ 42 , 65 , 66 ]. It is important to determine if designation of a CO at one site subsequently results in NCO designation of nearby DSBs, which is an attractive, but as yet unproven, proposal. Two independent studies in S. However, the two studies presented incongruent results regarding CO frequency on other chromosomes in that Joshi et al.
In light of these findings, the researchers propose a short-range interference model discussed further below in which Pch2 aids in the establishment of chromosomal domains in which only one CO can occur [ 67 ]. In addition to its role in sister chromatid cohesion, Rec8 has been implicated in a diverse set of meiotic processes including pairing, SC polymerization, recombination, and disjunction [ 70 , 71 ].
Many of these functions have been shown to be separable from its role in sister chromatid cohesion [ 71 ]. ChIP-chip reveals an interesting correlation between binding of Spo11 and cohesins. Spo11 has been shown to co-localize with Rec8 in early meiosis, and the frequency of co-localization decreases as meiosis progresses [ 73 ].
It has been proposed that Rec8 provides structural landmarks that dictate the proper distribution of Spo11 [ 73 ]. Spo11 could transfer from Rec8 binding sites to chromatin loops before initiating DSBs [ 73 ]. The Rec8 binding landscape could serve to bias Spo11 distribution toward uniformity. However, a role for Rec8 in the mediation of interference has not yet been demonstrated, so additional work to elucidate this relationship will have to be conducted. The central unanswered question regarding interference is how it is achieved within the cell.
The answer to this question has been elusive partly because traditional genetic screens designed to discover interference mutants are labor-intensive and problematic. Isolating the interference machinery is difficult because the relevant players likely have overlapping roles with other critical meiotic processes. Most mutations that affect interference also affect CO frequency and only a handful of mutants has been identified all in S.
Further complicating matters is that many of these mutants behave differently with regards to CO frequency in different studies and within studies at different loci [ 39 , 57 , 63 , 67 , 68 , 74 ]. Additional complications arise from the fact that interference is not absolute in most species but instead reflects the reduced probability of an event in a population of events — making the implementation of efficient screens difficult. Models aimed at explaining the interference mechanism have been proposed, but there is no dominant paradigm as no empirical observations vastly favor one particular explanation.
Several commonly referenced models are discussed below. Because interference is maximized at close distances and decreases with increasing distance [ 3 ], interference models that rely on a mechanical explanation invoke a signal that spreads from CO sites. The modern mechanical stress model proposed by Kleckner et al.
Stress is generated as meiotic chromosomes compress and expand. COs result in a localized relief of this stress that spreads in both directions down the axis of the SC. This model is attractive because it posits a simple explanation that predicts many properties of interference including CO assurance and CO homeostasis discussed below.
As each chromosome will be under stress, the occurrence of a first event, CO assurance, is easily obtained.
0コメント