Targets Of The Mos4 Associated Complex Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Plants are able to defend themselves against pathogens in a variety of ways. In non-host resistance, pathogens are unable to infect the plant, usually due to pre-formed physical barriers in the cell wall or plasma membrane (Weirmer et al., 2005). Host resistance is though conceptually as a multilayered system (Dangl and Jones, 2006). Plant pathogen recognition receptors are able to detect conserved features of pathogens and mount a relatively weak defence response. To mount a stronger defence response plants use Resistance (R) proteins which can detect secreted pathogen-derived compounds (effectors). R-protein activation is characterized by two major events; a localized hypersentive response which kills plant cells at the site of infection to prevent further pathogen colonization and systemic acquired resistance which increases the plant's ability to resist future infection (Durrant and Dong, 2004). Between R-protein activation and the end defence responses, many complex intracellular signalling events take place. Our laboratory uses a unique gain-of-function R-gene mutant to try to unravel these signalling events.

The snc1 mutant has been a useful tool for the discovery of defence related genes. It was initially identified in a mutant screen designed to search for mutants with enhanced defence signalling (Li et al., 2001). Using an npr1 mutant in the Columbia (Col) ecotype with a PR2::GUS reporter construct in the background, mutagenized plants were screened for the upregulation of PR2. snc1 was found to be a gain-of-function mutation which results in constitutive activation of the R-protein in the absence of pathogen effectors (Zhang et al., 2003). Since snc1 functions at the initiation of defence signalling and it is always actively signalling, it became apparent that this mutant could be an ideal tool to identify downstream defence-related genes. To identify genes downstream of snc1, a snc1 suppressor screen was initiated. This screen has identified fifteen modifier of snc1 (mos) genes. The nine fully characterized MOS proteins function in RNA processing, nuclecytoplasmic transport, protein modification and epigenetic control of gene expression, which revealed the complexity of downstream R-protein activation (Li et al., 2010; Zhang et al., 2005A; Zhang et al., 2005B; Palma et al., 2005; Goritschnig et al., 2007; Palma et al., 2007; Wiermer, M. al., 2007; Cheng et al., 2009; Goritschnig et al., 2008; Germain, H et al., in preparation).

The mos4 mutation is recessive and it completely suppresses all snc1 phenotypes (Palma et al., 2007). In the snc1 mos4 mutant, snc1 mediated dwarfism, constitutive PR gene expression, heightened salicylic acid (SA) levels and enhanced resistance to both Hyloperonospora parasitica Noco2 and Pseudomonas syringae pv maculicola ES4326 are all reverted to wild type (WT). Epistasis analysis revealed that MOS4 functions in an SA independent pathway. The MOS4 protein is a small (253 amino acids) evolutionarily conserved coiled coil protein predicted to be involved in protein-protein interactions. Yeast hybrid and co-immunoprecipitation experiments have found that MOS4 associates with at least twenty other proteins in a complex, forming the MOS4-associated complex (MAC) (Palma et al., 2007; Monaghan et al., 2009). Furthermore, many of these proteins have homologs in homologous complexes in yeast (nineteen complex; NTC) and human (Prp19/CDC5L complex), suggesting that the complex is highly conserved. In yeast, the NTC plays an essential role in general splicing (Wahl et al., 2009). Therefore, it was initially hypothesized that the mos4 phenotype is caused by impaired splicing (Palma et al., 2007). However, splicing defects have not been observed in mos4 or other known MAC mutants. However, combined MAC double mutants are lethal, suggesting that the MAC could be essential, as a complex, for splicing. These results suggest MOS4 is performs a function in defence, independent of splicing, which is disabled in mos4. Recently the molecular structure of the human Prp19/CDC5L complex has been characterized (Grote et al., 2010). The core component of the human Prp19/CDC5L complex (Arabidopsis MAC) are PRP19 (MAC3A/3B), SPF27 (MOS4), PRL1 (PRL1) and CDC5L (AtCDC5). Grote et al., (2010) found that SPF27 forms a hub to which four units of PRP19 and one unit each CDC5L and PRL1 bind. From these data we can infer that although mos4 does not effect splicing, it could be a fundamental protein holding parts the MAC together, and without it other proteins may not be able to bind to the MAC.

The question remains, how does mos4 suppress snc1? Two of the core components of the MAC, AtCDC5 and PRL1, offer some clues. AtCDC5 is an atypical R2R3-Myb transcription factor and MAC3A/3B is a U-box E3 ubiquitin ligase (Monaghan et al., 2009). Given the aforementioned data, it is possible that MOS4 tethers AtCDC5 and MAC3A/3B to the MAC. AtCDC5 and MAC3A/3B could effect the transcription of defence related genes; while MAC3A/3B could tag negative regulators of defence signalling for degradation, AtCDC5 could promote the transcription of defence related transcripts. Without MOS4, AtCDC5 and MAC3A/3B would not be able to bind to the MAC and their functions could be lost. In support of this hypothesis, we found that Atcdc5 mutants partially suppress snc1(Palma et al., 2007) and mac3A/3B double mutants have exactly the same phenotypes as mos4 (Monaghan et al., 2009).

To search for the targets of the MAC, a suppressor screen in the snc1 mos4 background was conducted. If our hypothesis that the MAC targets negative regulators of defence for degradation (through MAC3A/3B) is correct, then this screen should allow us to find those targets. A loss of function to one of these negative regulators would lead to upregulated defence. However, regardless of the mechanism of MAC function, we know that mos4 fully suppresses snc1. Therefore, even if our hypothesis is incorrect, a suppressor screen in this genetic background could enable the discovery of other negative regulators that are unrelated to the MAC and unique gain-of-function mutants to defence genes.

Thesis Objective

Identify at least one novel defence gene, preferably a target of the MAC, by studying suppressors of snc1 mos4.

Proposed Research

Isolation of putative suppressors of snc1 mos4

snc1 mos4 seeds were mutagenized with ethyl methanesulfonate (EMS) treatment. The 2500 M1 plants were allowed to self-fertilize. In the M2 population 50 000 plants were screened and mutants with dwarf morphology were selected (work done by K. Palma, J. Monaghan and L. Li). We suspected that these mutants were suppressing the snc1 mos4 phenotype. Six seeds from each dwarf M2 plant were grown to check for heritability. At this point, one of the six plants were genotyped at the MOS4 locus. This was done because a mutation reverting mos4 to WT MOS4 could have been responsible for the dwarf phenotype. Lines with confirmed heritability of the dwarf phenotype and homozygosity for mos4 were collected for a secondary screen.

Secondary Screen

To further characterize and identify the mutants most likely contributing to innate immunity, a secondary screen was performed. Morphology, PR gene expression, resistance to a virulent pathogen, snc1 gene expression were assessed in each mutant.

From studying the snc1 mutant and other mutants with constitutively active defence, we know that these mutants typically have a specific type of dwarf morphology, which mainly includes twisted leaves, bushy siliques and an enhanced ability to grow at high temperatures. However, other morphological features can be present. We took note of leaf shape (e.g. round, arrow-like, serrated), leaf colour relative to WT, intra-plant variations in colour and silique features (e.g. length, bushy or evenly separated). Photographs were taken of each mutant to document the morphology.

The original mutant line, snc1 mos4, has a PR2 promoter GUS construct in its genetic background. This has allowed us to determine if this defence marker gene is unregulated in the mutants. The mutant seedlings (10 days old, grown on MS plates or soil) were stained for GUS. This assay gave us a clear idea of whether or not defence signalling is effected in the mutants.

To more directly assess defence signalling, we measured the resistance of the mutants to a virulent oomycete pathogen, Hyaloperonospora Arabidopsis(H.a.) Noco2. Plants were inoculated at the two leaf stage and after a one week incubation, levels of infection were measured relative to a resistant (snc1) and a susceptible (Col) control.

In addition to a revertant mutation to mos4, a mutation causing the upregulation of snc1 would also lead to the suppression of snc1 mos4. In an attempt to identify these mutants, snc1 gene expression was measured. One potential limitation of this approach is that it could only detect mutations that affect the snc1 transcription or protein accumulation. Mutations leading to a change in SNC1 structure and a further gain of SNC1 protein function will not be identified using this method. Mutants that effect the expression of snc1, such as MOS1 (Li et al., 2010), are of interest to the field, however, since they would not be targets of the MAC, my project will not focus on these mutants.

Dominance/Recessiveness of the Mutation

To determine whether the mutation is dominant or recessive, a backcross to BGL2 (Col with PR2::GUS in the background) was performed. Typically this backcross would be done with the original line (snc1 mos4). We have decided on the above approach because this cross will allow us to get the single suppressor mutants. We can use this approach because in snc1, the morphology is recessive (i.e., snc1/SNC1 has WT morphology) (Li et al., 2001). mos4 is recessive in all its phenotypes (Palma et al., 2007). Therefore, the F1 will have WT morphology if the third mutation is recessive and dwarf morphology if the mutation is dominant. GUS staining could be used to confirm dominance/recessiveness of the defence signalling phenotype. If the mutation is dominant, we would expect to see more GUS staining relative to a snc1/SNC1 mos4/MOS4 mutant. If the mutation is recessive, we would expect to see the same level of GUS staining relative to a snc1/SNC1 mos4/MOS4 mutant.

The minor disadvantage of this approach is that it will be very difficult to confirm that the mutant phenotype is caused by a single gene mutation. Typically, after determining dominance or recesiveness in the F1, we would allow the plants to self and if the mutant phenotype is caused by a single gene mutation, a 3:1 ratio would be anticipated in the F2. Since in my proposed cross all three genes will be segregating in the F2 and the single mutant phenotype of the third mutation is unknown, this approach is not likely to work. However, the vast majority of mutant phenotypes are caused by a single gene mutation. In addition, crude mapping will be able to verify the number of genes contributing to the phenotype (i.e., linkage with more than one loci, would indicate that more than one mutation contributes to the mutant phenotype).

Map Based Cloning

To clone the genes of interest we first crossed the original triple mutants with Landsberg erecta (Ler). The F1 was allowed to self and approximately 200 F2 seeds from each cross were planted. Approximately four independent crosses were done for each mutant line. DNA was extracted from twenty-four F2 plants having the same morphology as the original mutant. Insertion/deletion markers throughout the genome were used to search for a bias towards the Col genotype; this indicates linkage of the mutant phenotype to the particular marker being used. Once linkage was identified, other markers were used to flank the gene. After the mutation has been flanked, chromosome walking was done to map the gene to the smallest possible region.

Fine mapping will allow us to narrow down the gene to approximately a 100kb region of the genome. To generate a population of 1000 recombinants for fine mapping, it is necessary to identify plants that are heterozygous for our mutation of interest and homozygous at the SNC1 locus (either mutant WT SNC1 or snc1) to avoid interference of the SNC1 locus. In the F3, such plants will produce a 3:1 ratio of WT to dwarf (or snc-1 like to very snc-1 like) plants. This F3 population will be used to fine map the gene of interest using recombination mapping and chromosome walking. Once the gene has been mapped to the smallest possible region, the genes within that region will be directly sequenced in the mutant to identify the molecular lesion.

Verifying that the Correct Gene has been Cloned

Verifying that the correct gene was cloned can be accomplished using transgene complementation or by using available T-DNA insertion mutants. Transgene complementation could be performed by cloning the WT gene into a vector and using Agrobacterium mediated transformation to transfer the construct into the triple mutant. If the correct gene was cloned, this would rescue the triple mutant phenotype and we would observe a snc1 mos4-like phenotype. If the mutation of interest is a loss of function mutation, T-DNA mutants of the gene of interest can be used to confirm the correct gene was cloned. For examples, a full knock-out, homozygous T-DNA mutant of the gene of interest could be crossed with snc1 mos4. If the correct gene was cloned, a 3:1 ration of WT to triple mutant-like plants would be observed in the F2. Theses strategies assume that the single mutant does not have a phenotype, if it does, there are other ways these experiments could be conducted.

Progress of the Proposed Research

Fifty-two dwarf lines were identified in the M2. Six of these lines were shelved after the secondary screen, mostly because of a combination of low GUS staining (indicates low PR2 expression) and H.a. Noco2 susceptibility. Two lines have been lost because none of the seeds would germinate even on agar plates. Ten lines had very low fecundity and I am attempting to bulk up seeds. Three lines of these ten lines will not grow enough to set seed, even when grown at high temperature (30°C) and high humidity (60%).

Most of the lines have been through the morphology (52 of 52 lines), mos4 genotyping (44 of 52 lines), GUS staining (41 of 52 lines), H.a. Noco2 assay (29 of 52 lines) and snc1 gene expression (20 of 52 lines) aspects of the secondary screen. Based on these results thirteen lines with very promising phenotypes were identified (i.e., lines having snc1-like morphology, mos4 homozygous, strong GUS staining, resistance to H.a. Noco2 and unchanged snc1 expression). Forty-five of these lines have been crossed with Ler to generate a population for crude mapping in the F2. Thirty-one lines have been backcrossed with BGL2 to determine dominance or recessiveness. We are still awaiting the data.

Line 60B-1, one of the promising mutants, has been crude mapped. Twenty-four plants with the mutant phenotype were used for crude mapping. Linkage was found on two chromosomes. Initially, linkage was found on chromosome 4 near the snc1 locus. This suggested that either the mutation was closely linked to snc1 or the phenotype of the triple mutant was dependent on snc1. However, linkage was also discovered on chromosome 5, which suggests that the triple mutant phenotype is dependent on snc1. The mutation on chromosome 5 was flanked using markers MJG14 (at 14.7 Mb) and KIL20 (at 26.5 Mb). Using chromosome walking, the mutation was narrowed down to a ~6 Mb region between K19E20 (at 20 Mb) and MU133 (at 28.82 Mb). Unfortunately, three known genes involved in defence are also in this region of the genome, SIZ1, BON1 and CIM3. To determine whether 60B-1 is allelic to these mutants, a complementation test will be performed. We will have to wait for these results before moving on to fine mapping this mutation.

Another promising mutant, 14-2 has also been crude mapped. F2 plant from the mapping cross that had the 14-2 triple mutant phenotype were selected and DNA was extracted for crude mapping. The phenotype of this mutant is also dependant on snc1. Besides snc1, the only linkage found was on chromosome 2 at three different widely separated markers. The linkage at these loci was weak (~40 cM). Since it is not yet known whether 14-2 is dominant or recessive, we decided to look at the data for chromosome 2 as though the mutation were dominant. This analysis showed that a dominant mutation on chromosome 2 is a possibility for this mutant. However, the mutation could still be a recessive mutation that is closely linked to snc1. snc1 is on chromosome 4 at 9.5 Mb. In 14-2, if the marker F16G20-2 (4 at 11 Mb) is used, 2 recombinants are found (~4 cM from the mutation of interest). When a 60B-1 DNA was tested for the same marker, 6 recombinants were found (~12 cM from the nearest mutation). This data suggests that the mutation of interest in 14-2 could be closely linked to snc1. We will wait for the data from the backcross to determine dominance/recessiveness before proceeding with this line.

The next step in the project will be completing the secondary screen. Recently, we have obtained enough seeds from some of the lines that were being bulked up. However, with some very low yielding lines, it may not be possible. The secondary screen will be mostly completed by the end of May.

Many of the F1 from the backcrosses to BGL2 are currently on soil. However, due to low seed numbers or failed crosses, some crosses still need to be done. It is anticipated that dominance/recessiveness of most of the lines will be determined by the end of May or early June.

In May, I will crude map three more lines: 83-2, 84-1 and 39-1. The F2 seeds from the mapping cross are currently on soil. Once these lines have been crude mapped, two lines will be selected for fine mapping. Besides the results from the secondary screen, there are a few features that will be looked for in a line to fine map. If a recessive mutant is selected, then a line with a phenotype that is dependant on snc1 is preferred. If the phenotype is dependant on snc1 and the single mutant does not have phenotype, than it is more likely that this mutant will be novel; the chances that it has previously been identified in a mutant screen are low. However, if the mutation is crude mapped to a locus with no known defence related genes, this will also be a good sign that the gene is novel. If a dominant mutant is selected, dependency on snc1 is not desirable because this could a mutation that simply causes a further gain of SNC1 function. However, this can also be tested by observing the segregation of the F2 of the mapping cross and crude mapping the mutation. A 3:1 ratio of WT to snc1-like plants will suggest the mutation could be a further gain of function to SNC1.

Once our lab has crude mapped most of the mutants, I will select two to fine map, to guarantee that at least one mutation will be cloned. Most of the crude mapping will probably be finished by the end of the summer. If everything goes as planned, I should be able to have my first gene cloned by this time next year.