Unveiling the role of DXS-Interacting (DXI) proteins in the regulation of plastidial isoprenoid biosynthesis

Abstract

Chloroplasts provide plants with metabolic pathways that are unique among eu- karoytes, including the methylerythritol 4-phosphate pathway (MEP) for the pro- duction of isoprenoids essential for photosynthesis and plant growth. The first re- action of the MEP pathway involves the synthesis of deoxyxylulose 5-phosphate (DXP) from the central metabolic intermediates glyceraldehyde 3-phosphate (GAP) and pyruvate catalyzed by DXP synthase (DXS). DXS has a major role in regulating the MEP pathway flux, but little is known about how its levels and activity are regu- lated. It has been shown that DXS stability and enzymatic activity can be modulated by interaction with other plastidial DXS-interacting (DXI) proteins. The goal of this thesis work has been to characterize the physiological role of DXS-DXI interactions and the molecular pathways leading from the interactions to the eventual biological effects in the model plant Arabidopsis thaliana. To investigate whether loss of DXI function in mutants impacted DXS activity, we analyzed their resistance to clomazone (CLM), a DXS-specific inhibitor. Most of the loss-of-function mutants tested did not show resistance or sensitivity to this in- hibitor. However, two mutant alleles for the gene SBP, which encodes the Calvin cycle enzyme sedoheptulose 1,7-bisphosphatase (SBPase), showed an increased re- sistance to CLM. In contrast, overproduction of SBPase in transgenic Arabidopsis plants resulted in reduced CLM resistance. Strikingly, we found that DXS protein levels or activity did not change when SBPase levels are altered in plants. Although co-immunoprecipitation assays were unable to confirm the interaction DXS-SBPase, our results do show a functional relationship between the Calvin cycle and the MEP pathway. We propose that excess GAP in the sbp mutant is diverted into the MEP pathway, preventing CLM binding to DXS. The second part of the thesis continued previous work with DXI1, a DXS-interacting J-protein that facilitates the recognition of inactive DXS forms to deliver them to eventual reactivation or degradation pathways. In particular, we focused on inves- tigating the molecular components involved in these two opposite pathways. By bioinformatic and experimental approaches, we confirmed that DXS is prone to ag- gregate in the chloroplast and associate to insoluble (membrane) fractions in an in- active form. These inactive forms of the enzyme were found to overaccumulate in plants defective in DXI1 (renamed J20). J20 is an adaptor of the Hsp70 chaperone. We demonstrated that the DXS-Hsp70 complex interacts with the Hsp100/ClpC1

vi

chaperone to unfold DXS for delivery into the proteolytic chamber of the Clp pro- teolytic complex. On the other hand, correct folding of DXS is achieved with the contribution of Hsp100/ClpB3. Our work suggests that degradation or activation of DXS might depend mostly on changes in ClpB3 levels. This disaggregase accumulates when the MEP pathway flux is decreased and in situations causing protein folding stress. Through molecular, genetic and pharmacological approaches, we demonstrated that this accumulation depends on a mechanism called chloroplast Unfolded Protein Response (cpUPR). Elicitation of this cpUPR by inhibition of protein synthesis in the chloroplast led to increased expression of nuclear genes encoding ClpB3 and other chloroplast chap- erones, eventually causing a stress acclimation response. We further demonstrated that cpUPR is independent of GUN1, an integrator of retrograde signaling, since we observed that chaperones accumulate in both wild-type and gun1 mutant plants. However, GUN1-defective plants were unable to develop the acclimation response. Our data therefore confirm that GUN1 is a central integrator of different pathways controlling chloroplast protein homeostasis beyond the control of nuclear gene ex- pression. Our results will contribute to taking more informed decisions on future approaches to manipulate levels of chloroplast isoprenoids of interest (such as vitamins, biofuels or drugs against cancer and malaria) in crop plants.

Los cloroplastos tienen vías metabólicas únicas entre los eucariontes, incluyendo la vía del metileritritol 4-fosfato (vía MEP) para la producción de isoprenoides. La primera reacción de la vía MEP es la síntesis de desoxixilulosa 5-fosfato (DXP) a partir de gliceraldehído 3-fostato (GAP) y piruvato, catalizada por la enzima DXP sintasa (DXS). DXS tiene un papel central en regular el flujo de la vía MEP, pero aún se sabe relativamente poco acerca de cómo se regulan sus niveles y actividad enz- imática. En esta tesis se ha estudiado cómo la interacción de DXS con otras proteínas regula su función en Arabidopsis thaliana. De entre estas proteínas interactoras, solo la pérdida de función de la enzima sedoheptulosa 1,7-bifosfatasa generó un fenotipo de resistencia a la inhibición de DXS. Los resultados de la tesis permiten concluir que existe un vínculo funcional entre el ciclo de Calvin y la vía MEP, seguramente medi- ado por la disponibilidad de GAP. Por otro lado, se ha observado que DXS es un en- zima propenso a agregarse. Estas formas inactivas interaccionan con la proteína J20, un adaptador de la chaperona Hsp70. El complejo DXS-Hsp70 a su vez interactúa con ClpC1 para degradar DXS mediante la proteasa Clp. Por otro lado, la interacción del complejo con ClpB3 pliega correctamente y activa DXS. El destino de DXS podría depender mayoritariamente de los cambios en los niveles de ClpB3. Un menor flujo de vía MEP, la pérdida de la homeostasis proteica en el cloroplasto, o la expresión defectuosa del plastoma causan la acumulación de ClpB3 y de otras chaperonas me- diante la activación de la expresión de los correspondientes genes nucleares. Esta respuesta no requiere de la actividad de GUN1, un nodo central en la comunicación cloroplasto-núcleo. Sin embargo, GUN1 es esencial para la posterior respuesta de aclimatación que permite a las plantas soportar mejor otros tipos de estrés como los causados por la inhibición de la síntesis de isoprenoides. Nuestros resultados pueden contribuir a tomar decisiones más informadas para manipular los niveles de isoprenoides cloroplastídicos de interés (como vitaminas, biocombustibles o fárma- cos contra el cáncer y la malaria) en plantas de cultivo.