PCNA in extracted chromatin. Chromatin from crosslinked embryos for different times was isolated and then the bulk was im- munoprecipitated using anti-PCNA antibodies. After eluting the com- plexes, crosslinking was reversed and PCNA was recognized by using the anti-PCNA antibodies. IgG, immunoglobulin heavy band; α, anti - body without chromatin added; CE, crude extract of non-crosslinked embryos. 

Context in source publication

Context 1

... Connector Loop (IDCL) which is bound by different proteins possessing a domain known as of Interaction to PCNA or PIP. PCNA is essential for DNA metabolism, playing funda- mental roles in replication, repair, recombination and also in cell cycle control [5-7]. It was initially described as a protein factor that stimulated activity and increased processivity of δ and ε -type DNA polymerases [8-10], acting as a “sliding clamp” on DNA in a Replication Factor C (RF-C)-dependent way to allow the entry of the replicative DNA polymerase to the replication fork, stimulating it. Accumulating evidence indicates that, in mammal cells, PCNA associates to an increasing number of proteins that participate in DNA metabolism, as DNA polymerases δ and ε , RFC, Fen1, DNA ligase 1, DNA topoisomerase, MLHI, MSH2, XP-G, DNA methyl-transferases and many others [10- 16]. Thus, it has been proposed that PCNA, besides being a processivity factor for DNA polymerases, it works as a recruit- ing platform for these and other proteins to the replication fork, or else to DNA with single stranded fragments, indicative of DNA damage [17]. PCNA also associates to proteins functioning in cell cycle control, as D-cyclins, cyclin-dependent kinases (Cdks) and p21 [7, 18, 19]. Cyclin D is a protein that binds and activates a Cdk [20]. Cdks are a family of protein kinases whose function is to regulate the cell cycle and p21 is a regulator of cyclin-Cdk kinases. PCNA has been found forming a quaternary complex with cyclin D, Cdk and p21. The physiological meaning of this complex has not been determined, but it is suggested that it may contribute to coordination between cell cycle progression and DNA replication [21, 22]. Research in our group has been focused in the study of the G1 phase and the G1-S transition of the cell cycle and its relationship with a developmental process such as maize seed germination and, for this purpose, it has been focused to the study of some of the proteins that determine entry into the G1 phase such as the D-type cyclins and the A-type Cdks [23, 24], and also in a fundamental protein for S phase, PCNA. Indepen- dently of its S-phase function, and similarly to what it has been found in mammal cells, during maize germination, PCNA also associates to D-type cyclins and to A-type Cdks [25]. These ternary complexes show CdkA-type kinase activity and this activity seems to be more relevant in the early stages of germi- nation, precisely when the G1 phase is in progress. Further, the composition of these ternary complexes and their activity seem to be regulated by phytohormones during germination [26]. Since D-type cyclins participate in G1 phase, associated to Cdks and these proteins bind to PCNA, the question is if these ternary complexes are formed when PCNA is anchored to DNA. This could mean that PCNA functions as a sliding clamp so that the bound cyclin-cdk complex localizes its target proteins and phosphorylates them, which could constitute a new function of PCNA. The purpose of the present research is to determine, using the Chromatin ImmunoPrecipitation method (ChIP), if PCNA joins the cyclin-cdk complex when PCNA is associated to the chromatin. As the isolation of a high number of maize embryo axes in a short time is a very heavy burden, and our initial purpose was to establish the ChIP methodology, we started using maize embryos from 21 days post anthesis kernels. Searching for the optimal conditions for crosslinking, dif- ferent times of treatment of maize embryos with 1% formalde- hyde were tested (0, 15 and 30 min, 2 and 4 h). It was interest- ing to observe that western blots of total protein extracts, using antibodies against PCNA, showed bands of around 29, 87 and 174 kDa (Fig. 1), which appeared to correspond to the sizes of a monomer, a trimer and a dimer of a trimer of PCNA; the trimer and dimer of trimer bands increased proportionately to the increase in crosslinking time, that correlated with the con- sequent decrease of the monomer, 29 kDa PCNA. The appear- ance of the dimer of trimer band would imply the existence in plants of a PCNA novel structure, which could help explaining the great capacity of PCNA to associate to several different proteins at the same time. The nature of the bands at around 50 kDa is not known. Assays in search for the optimal reversion of crosslinking time, using 2 h-crosslinked samples as starting material, were performed. Reversion times were 0, 6, 12, 18 and 24 h, at 65 o C. Figure 2 shows that the band associated to PCNA mono- mer increased proportionately to the incubation time whereas the intensity of the bands associated to the trimer and dimer of trimer decreased with time. The intensity of PCNA bands in non-crosslinked samples incubated for 0 and 24 h at 65 o C did not vary (Fig. 2, lanes 7 and 8), suggesting that there is no loss of protein during the processing of the samples. The next step was to demonstrate that PCNA was present in the chromatin pool. Figure 3 clearly shows that, after extract- ing chromatin, PCNA was still bound in samples crosslinked for up to two hours. That chromatin had been isolated was demonstrated by co-localizing histone H3 in chromatin extracts (data not shown). For an optimal immunoprecipitation, it is necessary to fragment the chromatin to sizes in the range of 200-600 bp DNA since longer chromatin fragments may be difficult to bring down and shorter fragments could cause the loss of both the DNA and the protein. For this purpose, micrococcal nucle- ase was used for different times until the desired chromatin size was obtained (Fig. 4A). Finally, optimal immunoprecipitation conditions using the anti-PCNA antibody were: optimal time of crosslinking, 1-2 h, optimal time of crosslinking reversion, 24 h, antibody dilution, 1:100 and incubation with protein A- agarose, 4 h. Figure 4B shows that under these conditions, PCNA was immunoprecipitated with crosslinking times of up to two hours; longer times appeared to make immunoprecipi- tation more difficult, probably due to the formation of more intricate or stronger nets of DNA-protein. Having developed the optimal conditions for the ChIP assay, the method was now used to determine the time at which PCNA is found bound to chromatin during the germination process, and its possible association to cell cycle control proteins, name- ly, cyclin D4;2 and the CdkA type protein. Figure 5 shows that PCNA was bound to chromatin in maize embryo axes of dry seeds and that its amount increased as the germination process advanced. In the absence of added chromatin, or adding pro- tein A-agarose without anti-PCNA antibody, no PCNA signal was detected. Under these conditions, it was found that PCNA bound to chromatin also associated to the CdkA-type protein (as determined by using an antibody against the PSTAIRE sequence present in all CdkA-type proteins), since this protein was present in the anti-PCNA immunoprecipitates, particularly in the early germination times and then its amount decreased by 24 h (Fig. 6). Of the many possible existing maize D-type cyclins, Cy- clin D4;2 was selected to search for its association to PCNA because there was previous evidence of this association [27] and because the antibody we have developed gives a good, clear signal by western blot. Figure 7 shows that similarly to the CdkA-type protein, cyclin D4;2 associated to PCNA when this was bound to chromatin during maize germination, although with some variation with respect to time as the cyclin, being present at time 0, diminished during the early hours of germina- tion, to recover importantly by 24 h. The establishment of the methodology of chromatin immu- noprecipitation has allowed us to determine the existence of PCNA complexes with fundamental proteins for the regulation of the cell cycle in plants, a research area that has been largely neglected in the world. While this methodology was being optimized, it was found that PCNA can form more complex structures than a trimer, since the presence of a dimer of a tri- mer was evidenced, this is to say, two trimers that, apparently, would be sliding one very close to the other, and that therefore would be crosslinked by the formaldehyde treatment so that they would appear together in a denaturing gel. Evidence of the existence of such dimer of trimer has been found in hu- man cells [28], although new reports corroborating these early findings have not been published. A dimer of trimer of PCNA would help explaining the capability of this protein to associate to different proteins at the same time. PCNA has been considered as a sliding platform that is localized and bound by different proteins related to different aspects of DNA metabolism, like replication, repair or recom- bination, thus allowing their interaction with DNA and with other proteins to fulfill their role. One example would be the interaction of a replicative DNA polymerase with a DNA li- gase during Okazaki fragment maturation, actions mediated by their binding to PCNA. However, it is not so simple to under- stand the association of PCNA with proteins that regulate cell cycle advancement in G1 phase because: a) there is no DNA replication going on, i.e., the S phase has not started and, b) there is no evidence that the cyclin-Cdk complex binds to the chromatin. This is the reason why it is important to determine if the cyclin-Cdk complex binds PCNA when this is associated to the chromatin. Results presented here strongly suggest that PCNA, bound to the chromatin, carries on Cyclin D4;2 and a CdkA-type protein. We do not demonstrate the existence of a ternary complex but the association of PCNA with each one of the two proteins; nonetheless, since a cyclin, or a Cdk, by itself has no apparent biological activity, it is to be expected that in fact PCNA binds to both of them, or perhaps binds one of them, lets say the cyclin and it is this cyclin that carries on the Cdk. It will be most convenient to ...