The work presented in this chapter is thus an attempt to narrow down the most likely mechanism that cytokinin uses to affect WUS transcription, translation, and protein distribution. Surprisingly, the results found that elevated levels of cytokinin did not directly affect WUS transcription, nuclear localization, or stability, nor did cytokinin have any significant effect on CLV3, eliminating a possible indirect mechanism. Instead, a novel absence of cytokinin response was identified in the CZ, and evidence suggests that this zone is maintained both by the lack of transcription, and by an unknown repressive mechanism that can affect B-type ARR proteins.Cytokinin responses were also correlated with WUS protein stability, starting roughly 12 hours after exogenous treatment. Auxin however, dramatically reduced WUS protein levels within just 4 hours, suggesting that this hormone has a more direct effect on protein stability. This suggests a model where auxin responses in the CZ and PZ cells stimulate protein degradation pathways that confine WUS proteins to the RM, where cytokinin responses may favor protein stability.In order to identify which part of the WUS-CLV3 feedback loop is affected by cytokinin responses, this work began by crossing the pTCSn1:mGFP5-ER reporter was crossed into wus-1 and clv3-2 mutant backgrounds. In untreated plants, both clv3-2 and WT meristems were found to have strong cytokinin responses in the RM, with a faint fluorescent signal extending deep into the pith and provascular tissue . The wus-1 mutant was similar, though its fluorescent signal did not become faint in the deeper tissue layers, presumably because this mutant meristem did not produce pith or provascular tissue. Treatment with exogenous 6- benzylaminopurine for 24 hours did not change the location of the cytokinin response in WT meristems, or in the meristems of either mutant. Instead, 25 liter plant pot the strength of the fluorescent signal was more than tripled in all three backgrounds, suggesting that endogenous cytokinin response mechanism is able to function over a wide range of concentrations.

The enhanced signal was most easily detected in the weakly fluorescent pith cells, but interestingly, the immature leaves, young anlagen, and the apex of the SAM all failed to produce any fluorescent response at all. In both WT and clv3-2 mutants, the absence of cytokinin responses occurred in a circular patch at the apex of the Central Zone and extended two cell layers deep. In wus-1 homozygousmutants, a similar response-free zone was found to be variable, but was detected in 84% of sectioned meristems, and extended only one cell layer deep . The near-complete lack of cytokinin responses in the CZ was unexpected, though presence of this function indicates a previously unrecognized feature of meristem organization involving two opposite and adjacent cytokinin response fields. As exogenous cytokinin applications were not able to induce pTCSn1:mGFP5-ER expression in the response-free zone to any significant degree, the two-component system pCLV3:GR-LhG4 x p6xOP:ARR1ΔDDK-GR was used to ectopically stimulate cytokinin responses specifically in the CZ tissue. As previously described, the ARR1ΔDDK-GR construct is a modified version of ARABIDOPSIS RESPONSE REGULATOR 1 , which activates cytokinin response genes following exposure to dexamethasone. Plants containing both constructs were then crossed to three different lines containing the fluorescent reporters: pTCSn1:mGFP5-ER, pCLV3:mGFP5-ER, and pWUS:eGFP-WUS. Surprisingly, when observed over a 48 hour time course of dexamethasone treatment, the pTCSn1:mGFP5-ER reporter did not immediately occur in the middle of the CZ as expected. For comparison, the pCLV3:mGFP5-ER reporter uses an identical promoter, indicating that the induced cytokinin response should occur in a pattern similar to the middle row in Figure 3.2. Instead, the pTCSn1:mGFP5-ER signal first appeared at the extreme edges of the peripheral zone, where it progressively appeared in adjacent cells in a centripetal manner, slowly constricting the cytokinin response-free zone until it disappeared between 24 and 48 hours later.

The centripetal pattern was visible in both L1 and L2 cells, though the L2 signal was weaker and lagged behind the L1 by 1-3 cell diameters. By 48 hours, the response-free zone was completely lost, and pTCSn1:mGFP5-ER expression become nearly homogenous throughout the SAM.The pCLV3:mGFP5-ER reporter in contrast, was expressed in the apical pattern as expected for the CLV3 promoter. The fluorescent pattern occurred in conical patch of cells at the apex of the SAM, and extended up to four cells deep. The fluorescence levels were mostly uniform, though the L2 frequently had significantly less expression than the other layers. The expression pattern was already fully formed in the absence of dexamethasone treatment, and remained unchanged through at least 24 hours. The fluorescent pattern became broader in proportion to the size of the meristem at 48 hours, but the longitudinal pattern did not significantly change. Deep inside the SAM tissues however, a faint pCLV3:mGFP5-ER signal could be detected, which produced a central hourglass-shaped column more than 20 cell layers deep. At the beginning of the time course, the pWUS:eGFP-WUS reporter produced a nuclear-localized pattern centered on the RM, with a radial concentration gradient spreading into all adjacent cells as expected. This pattern did not change after 6 hours of dexamethasone treatment, but by 12 hours a subtle increase in the number of cells displaying the pWUS:eGFP-WUS fluorescent reporter was apparent. The number of small meristem-like cells also began to increase over time, accumulating in a rootward direction at a rate directly proportional to the loss of the underlying large pith cells. The pWUS:eGFP-WUS expression pattern followed the downward appearance of the new cells, eventually producing a brightly visible fluorescent column extending more than 20 cells deep.

Elongate voids with no fluorescence were visible on either side, similar to the faint column produced by pCLV3:mGFP5-ER. Interestingly, long term ectopic induction of the pCLV3:GR-LhG4 x p6xOP:ARR1ΔDDK-GR system did not significantly change the volume of the SAM over the first 24 hours, but by 48 hours the SAM volume had quadrupled. This exponential growth pattern continued in plants subjected to prolonged 120 hour treatments, eventually producing a spherically swollen SAMs 1- 2mm in diameter, with frequent superficial cracks . Curiously, while the change in cell proliferation appeared to be abrupt in the pTCSn1:mGFP5-ER and pCLV3:mGFP5-ER reporter backgrounds, the proliferation rate in the pWUS:eGFP-WUS background was more gradual, beginning at least 12 hours earlier than in the other two lines. This precocious behavior may be related to the concentration of WUS proteins in this line, as the presence of the pWUS:eGFP-WUS construct can complement wus-1 mutants, and likely double doubles the concentration of WUS proteins in the presence of the WT copy of WUS.While CLV3 does not appear to induce or respond to cytokinin to any significant extent, WUS proteins displayed a more complicated pattern as shown by pWUS:eGFP-WUS reporter in Figure 3.4: Part of the WUS pattern overlaps with the cytokinin-response-free zone, and typically no WUS was found in either the deep RM or the PZ, where cytokinin responses were clearly present at comparable time points. The failure of WUS to activate cytokinin responses in the CZ is somewhat surprising, as 24 hours of exogenous 6-bap treatment moderately increased pWUS:eGFP-WUS fluorescent levels in both WT and clv3-2 mutant backgrounds . On the other hand, if cytokinin is required to activate WUS transcription, the presence of WUSexpression in tissues that lack a clear cytokinin response is equally difficult to explain. When cytokinin responses are eliminated with the cytokinin receptor triple mutant ahk2/3/4, only trace amounts of pWUS:eGFP-WUS fluorescent signal could be detected in seven day old plants. The lack of cytokinin responses was further confirmed by treating pWUS:eGFP-WUS x ahk2/3/4 plants with exogenous 6-bap, which did not significantly change the fluorescent pattern . However, the ahk2/3/4 mutant was quite variable, as 74% of examined SAM tissues displayed no fluorescence, while the remaining 26% ranged from faint GFP patterns to nearly full WT-like patterns . To deplete native cytokinin in WT meristems without the physical defects of the ahk2/3/4 mutant, the dex-inducible construct p35S:GR-LhG4::p6xOP:CKX3 was used to over-expresses CYTOKININ OXIDASE 3 , which degrades native cytokinin molecules. Following 24 hours of dexamethasone treatment in this background, the pCLV3:mGFP5-ER reporter showed no significant change in expression . Parallel attempts to study pWUS:eGFPWUS in the p35S:GR-LhG4::p6xOP:CKX3 background produced extremely variable results during the first 24 hours, black plastic plant pots ranging from the complete absence of fluorescent signal, to near-WT patterns, but became consistent by 48 hours of dex treatment.When WUS transcription was checked with RT-PCR however, both WT and ahk2/3/4 mutants background were found to have detectable WUS transcription localized to the RM .

The expression pattern of WUS also largely unchanged in ahk2/3/4 mutant RNA in-situ’s, suggesting that cyokinin responses primarily affect WUS protein. Further RNA in-situ’s following the time course treatment of the pCLV3:GR-LhG4 x p6xOP:ARR1ΔDDK-GR system found that cytokinin did not significantly increase WUS transcription in the CZ cells .This indicates that the pWUS:eGFP-WUS fluorescence observed in CZ cells is a product of protein movement, not local transcription. The RNA in-situ’s further revealed that WUS transcription patterns also expanded in a rootward direction, similar to the pWUS:EGFP-WUS pattern shown in Figure 3.2. By 48 hours, WUS expression was clearly found throughout the entire volume of the enlarged RM, with the exception of L1 and L2, which had little or no WUS transcripts. In many cases, large elliptical voids appeared in the peripheral zone, which corresponded to the presence of lateral anlagen. When two voids were present simultaneously , the central RNA expression pattern is reminiscent of the central column displayed by the pWUS:eGFP-WUS reporter in Figure 3.2. Though WUS is known to be non-cell autonomous, the close correlation between RNA and GFP patterns suggests that WUS proteins has a short mobile range, here estimated at 3 cell diameters.Previous research has shown that the non-cell autonomous movement of WUS protein does not have tissue-specific patterns , suggesting that the plasmodesmata are unlikely targets of cytokinin regulation. In order to explore other possible means of protein movement regulation, this study began by performing hand-cut longitudinal sections of pWUS:eGFP-WUS plants, providing an un-biased view of the WUS concentration profile in the deeper layers of the SAM, thereby avoiding the loss of signal associated with tissue depth. Special care was taken to avoid saturating the pWUS:eGFP-WUS reporter during imaging, so that semi-quantitative analysis might reveal subtle patterns . In untreated plants, the pWUS:eGFP-WUS reporter revealed a nearly symmetrical concentration profile, with a triangular peak centered on the RM, tapering off over 3-4 cell diameters . The location of the peak varied between L3-L5 in different sections, which likely reflects error introduced by tangential or oblique cuts. Above the peak, the fluorescent gradient was strongly linear, tapering to near undetectable levels in L1 cells. In the deep meristem tissues, the rootward gradient was equally linear and symmetric for the first 2-3 cell diameters, but then began to flatten out into a low but relatively constant background signal. It is not clear how much of the deep-layer signal reflects the presence of WUS, as the pWUS-eGFP-WUS reporter did not display its characteristic nuclear-localized pattern in these cells. Instead, the fluorescent signal largely co-localized with the developing chloroplasts in deepest cells layers, suggesting that this background signal is at least partially derived from chlorophyll auto-fluorescent noise. However, no such noise can be detected in the absence of pWUS-eGFP-WUS, or when histone or ERtagged fluorescent proteins are used , strongly implying that this background signal reflects the actual WUS protein distribution. When pWUS:eGFP-WUS plants were treated with exogenous 6-bap for 48 hours, no significant changes were observed in the upper gradient , either in the slope or in the total fluorescent concentration. The signal started to diverge by L4 however, where fluorescent signal became as much as 2x brighter down through at least L10 .When cytokinin responses were ectopically induced in the pCLV3:GR-LhG4 x p6xOP: ARR1ΔDDK-GR background, the pWUS:eGFP-WUS reporter produced patterns very similar to the exogenous cytokinin treatments. The upper gradient remained unchanged, while the deeper cell layers starting at roughly L5 doubled their fluorescent signal. A time-course analysis further revealed that the deep-meristem signal began to appear after 12 hours, and was fully formed by 24 hours. Interestingly, by 48 hours the gradient in L1-L3 cells suddenly increased their fluorescent amplitude by 140%, yet the slope of the gradient in these cells remained unchanged.When cytokinin responses were reduced with the p35S:GR-LhG4::p6xOP:CKX3 construct, a slightly different pattern emerged.