A first glimpse of the molecular mechanism behind Ty-1, an atypical dominant resistance gene

Tomato yellow leaf curl virus (TYLCV) is a monopartite begomovirus within the family Geminiviridae with a single stranded (ss) circular DNA genome of 2.6 kb. TYLCV is one of the most devastating plant viruses worldwide and causes major yield losses in economically important crops like tomato. Its vector Bemisia tabaci is difficult to control and therefore breeding for resistance is the most effective way to combat TYLCV infections. In 2013, the first resistance gene against TYLCV, in casu Ty-1, was cloned and turned out to code for an RNA-dependent RNA polymerase (RDR) from the γ-class. While the members from the a-class (RDR1, 2 and 6, from Arabidopsis thaliana) have recognized functions in the amplification of the RNA interference (RNAi) response, either towards Post transcriptional gene silencing (PTGS) or Transcriptional gene silencing (TGS), the function of members of the γ-class (RDR3, 4 and 5) was completely unknown. Just prior to the start of this research first evidence was collected on the involvement of Ty-1 in the enhancement of antiviral RNAi. This thesis aimed to further unravel the molecular working mechanism of Ty-1 and its applicability in light of resistance spectrum and durability.

Ty-1 also turned out to be effective against Beet curly top virus (BCTV), a member of the Curtovirus genus (Chapter 2) and very distinct from TYLCV in the genus Begomovirus, supporting the idea that Ty-1 resistance is likely generic against all geminiviruses. Tomato Ty-1 introgression lines as well as stably transformed tomato and Nicotiana benthamiana (N. benthamiana) plants expressing the Ty-1 gene, showed resistance against TYLCV and BCTV. A betasatellite, often found co-replicating with monopartite begomoviruses and encoding a suppressor of TGS (βC1), was observed to compromise Ty-1 resistance. Not only co-replication of the betasatellite with TYLCV, but also co-expression of the encoded βC1 gene from a potato virus X (PVX) construct led to increased TYLCV titers in Ty-1 expressing plants (Chapter 2). This indicates that viruses/betasatellites encoding viral suppressors of TGS compromise Ty-1 resistance against the geminivirus during a co-infection.

In order to find out whether Ty-1 was involved in the PTGS and/or TGS response, small RNAs were purified from TYLCV-infected susceptible Moneymaker (MM) and Ty-1 bearing tomato plants, and sequenced to determine viral small interfering (vsi)RNA profiles. Compared to MM plants, profiles from Ty-1 tomato plants revealed a drastic reduction in the amount of 21-nt vsiRNAs, while the levels of those from the 22-nt and 24-nt size classes had increased. The genomic distribution of vsiRNAs also changed, showing less pronounced hotspots in Ty-1 plants (transitivity) and a relatively enhanced targeting of the Intergenic Region and the V1 gene. Whereas the increase in 24-nt vsiRNAs clearly suggests an enhanced TGS response, the elevated level of 22-nt vsiRNAs was unexpected and intriguing. Whether these 22-nt vsiRNAs play a role in a PTGS response or guide targeting of viral DNA for TGS via a non-canonical RdDM pathway, remains to be determined (Chapter 3).

Cytoplasmic Processing bodies (P-bodies) are responsible for RNA decay of defective RNA molecules and they compete with neighbouring siRNA-RDR6 (PTGS-amplification) bodies for these aberrant RNA molecules as substrate. Whereas impairing cytoplasmic P-bodies led to an enhanced PTGS response against a transgene (GFP), it did not change the susceptibility against TYLCV (Chapter 4). P-bodies thus do not seem to compete for RNA substrates with RDRs involved in amplifying the RNAi response against TYLCV and this finding suggests that the latter probably takes place somewhere else. 

In Chapter 5, Ty-1 was in situ localized in plant cells by transiently expressing GFP-tagged Ty-1 in N. benthamiana. The Ty-1 protein localized in nuclear bodies (NBs), in close proximity to COP1- containing photobodies. In the additional presence of TYLCV, the number of Ty-1 NBs increased and Ty-1 translocated to Cajal bodies. In addition, bimolecular fluorescence complementation assays showed interactions between Ty-1 and components of the sumoylation machinery, namely SIZ1, SCE1, SUMO1, and SUMO3. The interplay with the sumoylation machinery also appeared influenced by TYLCV infection. Since Cajal bodies were previously shown to host many components of the TGS pathway, these findings altogether point towards the involvement of Ty-1 in a TGS response.

In conclusion, evidence is presented of the applicability of Ty-1 to a wider range of geminiviruses, but at the same time an Achilles heel became visible, namely co-infections with other viruses that suppress TGS may break Ty-1 resistance.  Plants containing Ty-1 show increased levels of 24-nt vsiRNAs, needed for a TGS response, and of 22-nt vsiRNAs, of which the role in TGS or PTGS still needs to be determined. RDRs localized in cytoplasmic siRNA bodies and required for PTGS amplification, do not appear to be involved in an anti-geminiviral RNAi response. This fits with Ty-1 being localized in the nucleus, where the protein shows a dynamic distribution profile depending on the absence or presence of TYLCV infection, and interplays with the sumoylation machinery. In Chapter 6, the practical and scientific implications of the data presented in this thesis are discussed and summarized in a model to present a first glimpse of the mode of action of Ty-1 in a TGS response against geminiviruses.