Dynamics of a mobile RNA of potato involved in a long-distance signaling pathway

Abstract

BEL1-like transcription factors interact with Knotted1 types to regulate numerous developmental processes. In potato (Solanum tuberosum), the BEL1 transcription factor St BEL5 and its protein partner POTH1 regulate tuber formation by mediating hormone levels in the stolon tip. The accumulation of St BEL5 RNA increases in response to short-day photoperiods, inductive for tuber formation. RNA detection methods and heterografting experiments demonstrate that BEL5 transcripts are present in phloem cells and move across a graft union to localize in stolon tips, the site of tuber induction. This movement of RNA to stolon tips is correlated with enhanced tuber production. Overexpression of BEL5 transcripts that include the untranslated sequences of the BEL5 transcript endows transgenic lines with the capacity to overcome the inhibitory effects of long days on tuber formation. Addition of the untranslated regions leads to preferential accumulation of the BEL5 RNA in stolon tips under short-day conditions. Using a leaf-specific promoter, the movement of BEL5 RNA to stolon tips was facilitated by a short-day photoperiod, and this movement was correlated with enhanced tuber production. These results implicate the transcripts of St BEL5 in a long-distance signaling pathway that are delivered to the target organ via the phloem stream.

Figure 1.

Accumulation of St BEL5 RNA in Stems and Stolons of Wild-Type Plants under LD and SD Photoperiod Conditions. Plants of the photoperiodic responsive line S. tuberosum ssp andigena were grown in the greenhouse under LDs until the 12-leaf stage and then moved to a growth chamber under LD (16 h light/8 h dark) or SD (8 h light/16 h dark) conditions. After 12 d, tissue was harvested from shoot tips, stem sections, and stolon tips, and RNA was extracted and subjected to electrophoresis on a denaturing gel. Hybridization was performed on RNA gel blots with a ³²P-labeled DNA probe specific for St BEL5. Ten micrograms of total RNA was loaded per lane. Equal loading was verified by visualizing rRNA loading under UV light.

Figure 2.

In Situ Hybridizations of Sections of the Potato Stem and Stolon Tip of Wild-Type Plants during Early Tuber Formation. Sections were hybridized with a digoxygenin-labeled 0.2-kb RNA copy of sequence from the 3′ untranslated region of St BEL5 ([A] and [C], sense riboprobe; [B] and [D], antisense riboprobe). The presence of St BEL5 mRNA is indicated by the dark stain under bright-field microscopy. Bars = 100 μm in (B) and 0.5 mm in (C) and (D). The stolon sections ([C] and [D]) are from an SD-grown plant in the early stages of tuber formation. The stem sections ([A] and [B]) are from LD-grown plants. Similar results were obtained in stem sections from SD-grown plants. (A) Transverse section of an LD stem; a negative control. p, phloem cells; co, cortex; x, xylem. (B) Transverse section of an LD stem with antisense probe. E, external phloem cells; I, internal phloem cells; x, xylem; vc, vascular cambium. (C) and (D) Longitudinal sections of newly tuberizing stolon tips. Sieve elements in (D) were identified by a comparison to longitudinal sections of stolon tips in Cutter (1978). The arrows in (D) indicate positive St BEL5 signal.

Figure 3.

Laser Microdissection and LPC to Identify Specific RNAs in Phloem Cells of Potato Stems. Microdissection was performed on paraffin-imbedded transverse sections of potato stems of wild-type plants (S. tuberosum ssp andigena) using the PALM Microlaser system (Plant Sciences Institute, Iowa State University). For microdissection, focused laser light was used to excise selected cells from regions of the xylem, phloem, or epidermis (X, P, and E in [A], respectively). After microdissection, the sample cells were directly catapulted into an appropriate collection device. RNA was extracted from these cells and used as a template for RT-PCR using gene-specific primers for each RNA sample (B). NT is nitrate transporter (specific for root cells). G2 is the G2-like transcription factor (specific for phloem cells). Identity of PCR products was confirmed by sequencing isolated bands. Data shown are from SD-grown plants. Identical results were obtained when using material from LD-grown plants.

Figure 4.

Movement of St BEL5-cds Transcripts across a Graft Union in Vitro and in Soil-Grown Plants. Using heterografts with tissue culture plants (A) and RT-PCR with gene-specific primers, RNA for St BEL5 moves across a graft union toward the base of the plant ([B], WT stock lanes). All PCR products detected in scion (positive control) and stock (test for movement) RNA samples represent tagged transgenic RNA. Arrows in (A) indicate where samples for RNA extraction were taken. Lines 19 and 11 are transgenic potato lines that overexpresses St BEL5 mRNA coding sequence only. RNA from line 19 and line 11 scion samples ([A], top arrow) was used as a positive control. Wild-type RNA from stock material ([A], bottom arrow) was sampled for both heterografts. Heterografts were cultured in vitro for 10 d under SD conditions (8 h light/16 h dark). PCR was performed twice off cDNA template made from RNA and reverse transcriptase. Two different gene-specific primers were used with a nonplant DNA tag specific for the transgenic RNA (designated NT-2 in Methods). RNA from wild-type/wild-type autografts was used as a negative control (WT lane). Ethidium bromide–stained PCR product of expected size (360 bp) is shown in (B). The identity of these bands was verified with a DNA gel blot using a St BEL5–specific probe. RT-PCR results from whole plant heterografts of St BEL5-cds/wild type are shown in (C). Scion leaf samples were from overexpression lines of St BEL5-cds from three separate heterografts, whereas wild-type stock samples were harvested from 5.0 mm of the stolon tip after 14 d of SD conditions. Transcript-specific primers were used for St BEL5 and for the rRNA reactions. No PCR product was detected from reactions without template or from RNA extracted from wild-type/wild-type autografts.

Figure 5.

Rate of Tuberization of Full-Length St BEL5 Transgenic Lines. Previous work on the analysis of transgenic lines that overexpressed St BEL5 was done with the coding sequence only (A) (Chen et al., 2003). In this experiment, transgenic lines overexpressing the full-length St BEL5 transcript (A), including ∼650 nucleotides of the UTRs (accession number AF406697) were tested. Line 7540 is nontransformed S. tuberosum ssp andigena. Evaluations were performed on 10 plants per line cultured on Murashige and Skoog (MS) medium with 6% sucrose (B). Plants were scored for time to tuber formation under LDs (earliness), number of tubers after 14 LDs, and total tuber yield after 56 d under SD conditions in vitro. The greenhouse LD experiment was with soil-grown plants. The top five lines (asterisks) tuberized as soil-grown plants in 10-cm pots after 8 weeks under LD conditions in the greenhouse. Most of these tubers ranged from 1.5 to 2.5 cm in diameter (C). Overexpression lines of the coding sequence only (Chen et al., 2003) did not form tubers under LDs even after 6 months in the greenhouse. The means of the St BEL5-FL lines for earliness, tuber number, and yield were significantly greater than the 7540 control plants at P = 0.05 (Fisher's test), and this difference is designated by the letters a or b. ND, not detected.

Figure 6.

Movement of Full-Length Transcripts of St BEL5 across a Graft Union and Its Effect on Tuber Yield. Soil-grown wild-type stock plants were grafted with scions from an St BEL5 overexpression line (5FL1-3) in the greenhouse. Both stock and scion material were grown initially under LDs. Grafts were sealed with plastic soda straw sections. Plants were placed in plastic bags, and graft unions were allowed to form. After 4 weeks of LDs in the greenhouse, grafted plants were transferred to a growth chamber and acclimated under LD conditions for 1 week before transfer to SD conditions. Leaf and stolon tip samples (arrow in [A]) were harvested after 12 d of SD conditions and the RNA extracted. RT-PCR with gene-specific primers was performed for both a negative control (a potato MADS box gene, 102-23) and test samples (St BEL5-FL). Grafts made from an overexpression line for an antisense sequence of a potato MADS box gene (line 102-23) were used as a nonmobile control. RNA from scion leaf samples was used as a positive control (scion leaf). Wild-type RNA from stolon tips, 0.5 cm in length, was sampled for both heterografts and used in the RT-PCR reactions. PCR was performed twice off template made from RNA and reverse transcriptase. Two different gene-specific primers were used with a nonplant DNA tag specific for the transgenic RNA to discriminate from the native RNA. Three plants were assayed for both heterografts and are designated 1, 2, and 3. RNA from leaves of a wild-type/wild-type autograft was used as a negative PCR control (WT leaf lane). Similar negative results were obtained with RNA from autograft stolons. For tuber yields, plants were harvested after 28 d, and the mean of three plants was calculated for wild-type and St BEL5 grafted plants. Wild-type scions grafted onto wild-type stocks were used as the yield controls. Identity of PCR products was confirmed by blot hybridization with a gene-specific probe.

Figure 7.

RNA Accumulation in Transgenic Lines Expressing St BEL5 Transcripts with (St BEL5-FL) and without (St BEL5-cds) the UTRs of the RNA and the Effect of Photoperiod. Transgenic plants were grown under SDs for 14 d, and the RNA was extracted from 0.5-cm stolon tips (S) and new leaves (L) from three separate plants for each construct (A). For the photoperiod experiment, transgenic plants with the St BEL5-FL construct were grown under LDs for 14 d ([B], St BEL5-FL-LD), and the RNA was extracted from 0.5-cm stolon tips and new leaves from three separate plants. The significant accumulation of St BEL5 RNA in stolon tips shows that signal in the stolon is dependent on photoperiod. One-step RT-PCR was performed using 20 ng of total RNA, a nonplant sequence tag fused to all transgenic RNAs (designated NT-2), and a gene-specific primer from either the open reading frame of St BEL5 (for St BEL-cds) or from the 3′ UTR (for St BEL5-FL). Use of the NT-2 primer makes it possible to discriminate transgenic RNA from native BEL5 RNAs. No PCR product was detected from native St BEL5 RNA extracted from leaves of wild-type plants ([A], WT lanes). The PCR reactions were normalized using rRNA primers. All PCR reactions were standardized and optimized to yield product in the linear range (31 cycles for both BEL5 RNAs and 17 cycles for the rRNA). Expected size of the full-length product was 375 and 550 nucleotides for the cds product. Both constructs were driven by the CaMV 35S promoter in the binary vector pCB201. The 5′ UTR of St BEL5 is 146 nucleotides, and the 3′ UTR is 505 nucleotides. Homogenous PCR products were quantified (B) using ImageJ software (Abramoff et al., 2004) and normalized using the rRNA values. The results of (A) are a representative sample of one of the biological replicates for each treatment. Standard errors of the means of the three biological replicates are shown in (B). Open bars, leaf; closed bars, stolon.

Figure 8.

The Effect of Photoperiod on RNA Accumulation in Stolon Tips of Transgenic Lines Expressing Full-Length St BEL5 Transcripts Driven by the Leaf-Specific Cm GAS Promoter (Ayre et al., 2003). Transgenic plants were grown under greenhouse conditions until the 12- to 14-leaf stage and then grown under SDs or LDs for 10 d before harvest. The RNA was extracted from 0.5-cm stolon tips (S) and new leaves (L) from three separate plants for each construct. Harvested plants were scored for tuber numbers after 10 d and tuber yields after 28 d under SD conditions (C). One-step RT-PCR (A) was performed using 500 ng of total RNA, a nonplant sequence tag fused to all transgenic RNAs (designated NT-2), and a gene-specific primer from the 3′ UTR of St BEL5 (GSP1). Use of the NT-2 primer makes it possible to discriminate transgenic RNA from native BEL5 RNAs. No PCR product was detected from native St BEL5 RNA extracted from leaves of wild-type plants ([A], WT lane). All PCR reactions were standardized and optimized to yield product in the linear range (39 cycles for the BEL5 RNA and 17 cycles for the rRNA). Expected size of the BEL5 product was 550 nucleotides. The full-length BEL5 construct was driven by the Cm GAS promoter in the binary vector pBI101.2. Homogenous PCR products were quantified (B) using ImageJ software (Abramoff et al., 2004) and normalized using the rRNA values. The results of (A) are a representative sample of one of the biological replicates for each treatment. Standard errors of the means of the three biological replicates are shown ([B] and [C]). Open bars, leaf; closed bars, stolon. GUS expression was detected in leaves but not stolons of GAS:GUS transgenic lines of S. tuberosum ssp andigena (data not shown).

Figure 9.

GUS Activity of the Upstream Region of St BEL5. The promoter sequence used in this construct to drive GUS expression contained a 202-nucleotide intron located within 103 nucleotides of the 5′ UTR and 1971 nucleotides of genomic sequence upstream from the 5′ UTR (A). GUS activity was detected in stolons and newly formed tubers ([B], arrow) and in leaf veins and petioles ([C] to [E]). The plant in (D) was cultured in vitro under LD conditions. All other material was from soil-grown plants. The leaf in (F) is 30 d older than the leaf in (E). Transverse section of a petiole exhibiting staining throughout the tissue, with the greatest level of activity around vascular cells is shown in (G) (arrows). Under higher magnification, blue staining was detected in companion cells (cc), in parenchyma-associated phloem cells (pp) ([H] and [I]), and in xylem cell walls (x) (H). No staining was observed in transverse sections of internodal regions of the stem (J). Bars = 100 μm in (G) and (J), 20 μm in (H), and 5.0 μm in (I).