The chloroplast is a primary site for the production of immune signals in plants. In this issue ofDevelopmental Cell, Caplan et al. (2015) report that chloroplasts send out stromules as signal conduits for transmission of these immune signals to the nucleus during effector-triggered immunity.

Plastids are a family of plant-specific het-erogeneous double-membrane organ-elles that perform various vital functions.The best-known plastids are the chloro-plasts, which carry out photosynthesis.A remarkable feature of plant plastids istheirability to formhighly dynamictubularextensions called stromules. The functionof stromules is largely unresolved,although they are often associated withother organelles, raising the possibilitythat they may function as signal transmis-sion channels to mediate inter-organellarcommunications (Hanson and Sattarza-deh, 2011). In this issue of DevelopmentalCell, the team led by Dinesh-Kumar re-ports a function for stromules duringeffector-triggered immunity (ETI) and pro-grammed cell death (PCD) (Caplan et al.,2015). Their discovery advances our un-derstanding of communications betweenchloroplasts and nuclei in the context ofplant immunity.Both plants and animals employ struc-turally and functionally similar nucleo-tide-binding and leucine rich-repeat(NLR) proteins to respond to perturba-tions caused by intracellularly deliveredpathogen effectors. Upon activation,NLRs initiate ETI, which is often accom-panied by PCD. An earlier study byCaplan et al. (2008) showed that the chlo-roplast protein N receptor-interactingprotein 1 (NRIP1) is required for recogni-tion of the tobacco mosaic virus (TMV)effectorp50bytheNLRreceptorNduringETIagainstthisvirus.Intriguingly,uponin-duction, NRIP1 is relocated from chloro-plasts to the cytosol and the nucleus,where it forms a complex with p50. Thiscomplex activates N via interaction withitsToll-interleukin 1receptor(TIR)domain(Caplan et al., 2008). In that study, the au-thors first observed N-triggered stromuleformation.To confirm this observation, Caplanetal.,inthecurrentwork,usedaquantita-tive method to determine whether stro-mule induction is specific to N activationas a way to coordinate the release ofNRIP1 from the chloroplast. The results,however, suggest a different and moreinterestingconclusion.Caplanetal.foundthat stromules can be induced uponinfection by the bacterial pathogen Pseu-domonas syringae pv. tomato (Pst)DC3000 carrying different effectors,such as AvrRps4, AvrRpt2, or AvrRpm1.However, stromules are not induced byeither a Pst DC3000 strain lacking theseeffectors or the type III secretion systemmutant (Pst DC3000 hrcC), which isdefective in delivering effectors. Thissuggests that stromules are generallyinduced during ETI, and this induction isnot limited by the type of pathogen orthe effector-NLR pair.Further dissection revealed that stro-mule induction is not restricted to locallyinfectedcellswhereETI/PCDisactivated,but also occurs in intact neighboringcells, suggesting that stromules may beinduced by mobile signals produced dur-ing ETI/PCD. Activation of ETI results inproduction of an array of mobile immunesignals, including defense-related hor-mones, secondary metabolites, andsignaling proteins. Of those, ROS accu-mulation and production of the immunesignal salicylic acid (SA) are of key impor-tance in orchestrating cell death execu-tion and defense response (Fu andDong, 2013). Indeed, Caplan et al. foundthat exogenous application of eitherH