Here we describe strong anion exchange high-performance liquid chromatography of [3H]-myo-inositol-labeled seedlings which is a highly sensitive method to detect and quantify inositol polyphosphates in plants.
The phosphate esters of myo-inositol, also termed inositol phosphates (InsPs), are a class of cellular regulators playing important roles in plant physiology. Due to their negative charge, low abundance and susceptibility to hydrolytic activities, the detection and quantification of these molecules is challenging. This is particularly the case for highly phosphorylated forms containing ‘high-energy’ diphospho bonds, also termed inositol pyrophosphates (PP-InsPs). Due to its high sensitivity, strong anion exchange high-performance liquid chromatography (SAX-HPLC) of plants labeled with [3H]-myo-inositol is currently the method of choice to analyze these molecules. By using [3H]-myo-inositol to radiolabel plant seedlings, various InsP species including several non-enantiomeric isomers can be detected and discriminated with high sensitivity. Here, the setup of a suitable SAX-HPLC system is described, as well as the complete workflow from plant cultivation, radiolabeling and InsP extraction to the SAX-HPLC run and subsequent data analysis. The protocol presented here allows the discrimination and quantification of various InsP species, including several non-enantiomeric isomers and of the PP-InsPs, InsP7 and InsP8, and can be easily adapted to other plant species. As examples, SAX-HPLC analyses of Arabidopsis thaliana and Lotus japonicus seedlings are performed and complete InsP profiles are presented and discussed. The method described here represents a promising tool to better understand the biological roles of InsPs in plants.
Almost four decades ago, inositol phosphates (InsPs) emerged as signaling molecules, after Ins(1,4,5)P3 (InsP3) was identified as a second messenger that activates the receptor-mediated release of Ca2+ in animal cells1,2. To date, no InsP3 receptor (IP3-R) has been identified in plants, which questions a direct signaling role for InsP3 in plant cells3. Regardless, InsP3 serves as a precursor for other InsPs involved in several plant developmental processes, including the regulation of specific signaling pathways3,4,5,6,7,8. For instance, InsP3 can be further phosphorylated to InsP6, also known as “phytic acid”, which represents a major source of phosphate, myo-inositol and cations, and was shown to play key roles in plant defense against pathogens, mRNA export and phosphate homeostasis5,9,10,11,12.
Inositol pyrophosphates (PP-InsPs) are a class of InsPs that contain at least one high-energy di-phospho bond, initially identified in animal cells, amoeba and yeast, where they play critical roles in various cellular processes13,14,15. Despite seminal work on PP-InsPs in plants16,17,18,19,20,21,22,23,24,25,26, the biological functions and isomer identity of these molecules still remain largely enigmatic. In the model plant Arabidopsis thaliana, cellular InsP8 was proposed to regulate defenses against insect herbivores and necrotrophic fungi via coincidence-detection of InsP8 and active jasmonate by the ASK1-COI1-JAZ receptor complex17. Furthermore, roles of InsP8 and other PP-InsPs in energy homeostasis and nutrient sensing, as well as phosphate homeostasis have been proposed17,23,24,25,26.
Regardless of the biological system employed, one major methodological challenge when studying InsPs has been the reliable detection and precise quantification of these molecules. Mass spectrometry-based methods have been used to detect InsPs, including PP-InsPs, from cell extracts. However, those studies failed to differentiate distinct isomers26,27. Another approach to analyze InsPs employs pull-down of InsPs from cell lysates using TiO2 beads, followed by polyacrylamide gel electrophoresis (PAGE) of the eluted InsPs. The InsPs can then be stained by either toluidine blue or DAPI24,28,29. However, it is so far not possible to reliably detect InsPs lower than InsP5 from plant extracts using this method. Recently, a method using [13C]-myo-inositol for nuclear magnetic resonance (NMR) analysis of InsPs was published as an alternative to strong anion exchange high-performance liquid chromatography (SAX-HPLC)30. This technique has been reported to achieve a similar sensitivity compared to SAX-HPLC and to allow the detection of 5-InsP7, as well as the discrimination of different non-enantiomeric InsP5 isomers from cell extracts. However, the implementation of the NMR-based method requires chemically synthesized and commercially not available [13C]-myo-inositol. Therefore, the method employed in most cases is radiolabeling samples with [3H]-myo-inositol, followed by SAX-HPLC31,32,33. This technique is based on the uptake of radioactive myo-inositol into the plant and its conversion into different InsPs by the combined activity of dedicated cellular kinases and phosphatases.
The [3H]-labeled InsPs are then acid-extracted and fractionated using SAX-HPLC. Because of their negative charge, the InsPs strongly interact with the positively charged stationary phase of the SAX-HPLC column and can be eluted with a buffer gradient containing increasing phosphate concentrations to outcompete InsPs from the column. Elution times thus depend on charge and geometry of the InsP species to be separated. In the absence of chiral columns, only non-enantiomeric isomers can by separated by this protocol. However, radiolabeled standards can be used to assign the isomeric nature of a specific InsP peak. Multiple efforts in the past by various laboratories to generate labeled and unlabeled standards with (bio)chemical methods or to purify them from various cells and organisms have helped assigning peaks to certain InsP species, and also to narrow down the isomeric identity of individual InsP species5,7,21,34,35,36,37,38,39,40,41,42,43. Also, the recent elucidation of enzymatic pathways leading to the formation of PP-InsPs in plants, as well as the discovery of a bacterial type III effector with a specific 1-phytase activity, provide information on how to generate useful standards for these analyses10,17,18,22,23.
The resulting fractions can be measured in a liquid scintillation counter due to the β-decay of tritium (3H). With increasing labeling time, a steady-state isotopic equilibrium is reached, after which the obtained InsP profiles should represent the InsP status of the plant31. The major advantage of this protocol in comparison to other available techniques is the high sensitivity achieved by the use of the direct precursor for InsPs and the measurement of a radioactive signal.
SAX-HPLC of samples extracted from [3H]-myo-inositol-labeled plants or other organisms is commonly used for the detection and quantification of InsPs ranging from lower InsP species to PP-InsPs, representing a valuable tool to better understand the metabolism, function and modes of action of InsPs. So far, this method is also the most appropriate choice for researchers with special interest in lower InsP species. While the basics of this procedure, on which the protocol described here builds on, have been previously described7,21,31,34, a detailed protocol tailored to the analysis of plant-derived InsPs and especially of PP-InsPs is still missing. Previous publications reported difficulties to reliably detect the low abundant PP-InsPs, especially InsP8, due to one or more of the following factors: relatively low amounts of plant material, [3H]-myo-inositol with low specific activity (> 20 Ci/mmol), usage of extraction buffers that are either not based on perchloric acid or are less concentrated than 1 M, different neutralizing buffers, as well as sub-optimal gradients or detection of [3H] with an on-line detector. In comparison to those studies, the protocol presented here is designed for the reliable detection of PP-InsPs7,21,34.
Here we present a detailed workflow, starting from the setup of the equipment to plant cultivation and labeling, InsP extraction and the SAX-HPLC run itself. Although the method was optimized to the model plant A. thaliana, it can be easily modified to study other plant species, as shown here with the first reported InsP profile of the model legume Lotus japonicus. Although the use of a different plant species might require some optimization, we envisage that those will be minor, making this protocol a good starting point for further research in plant InsPs. In order to facilitate possible optimizations, we indicate every step within the protocol in which modifications are possible, as well as all critical steps that may be challenging when establishing the method for the first time. Additionally, we report how data obtained by this method can be used for the quantification of specific InsPs and how different samples can be analyzed and compared.
Here we present a versatile and sensitive method to quantify InsPs including PP-InsPs in plant extracts and provide practical tips on how to get this method established. Even though the protocol is generally robust, suboptimal runs and analyses can occur. In most cases, those runs can be identified by a strong reduction or even complete loss of highly phosphorylated InsPs, especially the PP-InsP species InsP7 and InsP8. Possible reasons can be microbial contaminations of the plant material and insufficient deactivation of endogenous plant PP-InsP hydrolases during extraction due to insufficient grinding and thawing of plant material that will not be in immediate contact with extraction buffer. Further reasons include inaccurate pH adjustment by insufficient or excess addition of neutralization buffer, or simply insufficient sample material. The latter can make it difficult to detect PP-InsPs, since those are often present in very low amounts in the cells. An excess of sample material or inefficient drying during step 3.5 may cause dilution of the perchloric acid, therefore also leading to insufficient enzyme deactivation and a specific loss of InsP6 and PP-InsPs. The amount of plant material, as well as radiolabel used in this protocol were optimized based on costs and performance, and is therefore close to the lowest amount that is still sufficient for providing optimal results. In addition, the column resin will gradually loose its resolution capacity. The first sign of this process is (for reasons not entirely clear to the authors) a specific loss of higher phosphorylated InsP species like the PP-InsPs in the HPLC spectrum. With further aging, even InsP6 will not be resolved properly by the column (Figure 1D). Therefore, the use of an adequate column, as well as meticulous handling of the sample and proper maintenance of the HPLC components is crucial for ensuring accurate results.
When comparing samples and runs, especially when generated with different equipment (e.g., HPLC systems and columns) or on different days, it is crucial to normalize the samples to each other (as described in step 7.3) and to analyze them in the same way. Only through normalization it is possible and accurate to show multiple samples in the same graph (Figure 3). For quantification of individual InsPs relative to total InsPs, or to another specific InsP species, it is not necessary to normalize, as long as only relative values and not absolute values are shown. Ideally, both the InsP profiles and the quantifications are shown. However, in some cases it is not possible to adequately show two or more runs in the same graph. Different retention times or different levels of background activity can make it difficult to compare unquantified SAX-HPLC profiles alone. The same is the true when many samples need to be compared. In such cases, a further evaluation using an additional software (e.g., Origin) for individual peak quantification is necessary.
The authors are aware that the protocol described here can be optimized and needs to be adapted to each individual research question. Although being optimized for Arabidopsis extracts7,17 in this protocol, this method is versatile and can help determining InsP profiles of other plant species as well. Here we exemplify this possibility by presenting for the first time a InsP profile for L. japonicus, which required no modifications of the labeling conditions, InsP extraction or SAX-HPLC run (Figure 2). Notably, while overall similar, differences are observed between L. japonicus and Arabidopsis InsP profiles. For instance, in L. japonicus InsP5 [4-OH] or its enantiomeric form InsP5 [6-OH] are more abundant than InsP5 [1-OH] or its enantiomeric form InsP5 [3-OH] in comparison to Arabidopsis, where InsP5 [1-OH] or its enantiomeric form InsP5 [3-OH] are the dominant InsP5 species. Likewise, we anticipate that alterations in the media composition, [3H]-myo-inositol concentration, plant age, environmental conditions (e.g., light and temperature), addition of chemical compounds or analyses of plant-microbial interactions among other factors, might need to be tested and adapted.
One important drawback of this method that needs to be considered is that the labeling is done in a (sterile) liquid culture, which does not represent a physiological environment for most land plants. In addition, due to the high costs of [³H]-myo-inositol, the volume of the labeling solution and the size of the culture vessel is generally limited, which also restricts the size of the plants that can be used. Cultivation in liquid culture can be avoided by directly infiltrating for instance leaves of soil-grown plants with [³H]-myo-inositol and subsequently following the protocol described here, as previously reported10.
There are several drawbacks of this protocol in comparison to alternative methods, such as TiO2 pull-down followed by PAGE or mass spectrometry based techniques. Due to the [3H]-myo-inositol labeling, only InsP species that directly originate from radiolabeled myo-inositol will be detected in the end. The method described here is blind to other Ins isomers such as scyllo-inositol and other isomers some of which have been identified in certain plants44. Furthermore, myo-InsPs derived from other pathways will be excluded, including those synthesized by de novo synthesis of myo-inositol and myo-inositol-3-phosphate via isomerization of glucose-6-phosphate, catalyzed by myo-inositol-3-phosphate synthase (MIPS) proteins45. Although [32P] or [33P]-ortho-phosphate can be used as alternative labels, their use poses a major disadvantage, since every phosphate-containing molecule, including the abundant nucleotides and its derivatives, will be labeled. Those molecules can also be extracted with this protocol and bind to the SAX column, which will result in a high level of background activity that will interfere with the identification of individual InsP peaks5. In addition, quantification of [32P]- or [33P] -labeled InsPs and PP-InsPs can be strongly influenced by phosphate and pyrophosphate moiety turnover and might not report a mass readout for inositol species.
On the other hand, [3H]-myo-inositol specifically labels myo-inositol-containing molecules. InsPs, inositol-containing lipids, such as phosphoinositides, and galactinol are in this case labeled. However, only InsPs will be analyzed with this protocol, since lipids are insoluble in the extraction buffer and galactinol does not bind to the SAX column.
So far, the differences from a plant InsP profile generated by [3H]-myo-inositol labeling compared with one determined by TiO2 pulldown/PAGE remains unknown, since such comparisons have not been performed in plants. A recent study in animal cells addressed this question46. In that work, a pool of InsP6 that is invisible by [3H]-myo-inositol labeling, which should thereby be directly derived from glucose-6-phosphate, was identified by comparing SAX-HPLC profiles with PAGE gels of mammalian cell lines. 24 h of phosphate starvation resulted in a 150% increase of InsP6 when quantifying PAGE gels of InsPs purified using TiO2 pulldown. SAX-HPLC analyses of [3H]-myo-inositol-labeled cells that were treated identically only showed an increase by 15% of [3H]-InsP6. As previously mentioned, InsPs lower than InsP5 are undetectable with PAGE analysis in most cases. Radiolabeling followed by SAX-HPLC appears to be the method of choice, as long as mass spectrometric protocols are not optimized to detect this group of highly negatively charged molecules.
Another remaining challenge is to distinguish enantiomers in SAX-HPLC analyses (or in any other method for InsP analysis)10,17. This challenge can be tackled by the addition of chiral selectors, i.e., enantiopure compounds like L-arginine amide that interact with the respective enantiomeric molecules to form diastereomeric complexes that can be separated10. To our knowledge, this approach has only been implemented to discriminate the enantiomeric InsP5 isomers InsP5 [1-OH] and InsP5 [3-OH] by NMR analyses10. Discrimination of other enantiomeric pairs or successful discrimination of enantiomers by chiral SAX-HPLC analysis or chiral PAGE-based methods have not yet been reported and should be further developed. Considering the conserved synthesis and the conserved regulation of PP-InsPs by phosphorous availability, we envision that especially non-radioactive methods such as PAGE or MS-based methods, together with nutrient analyses, will help ground truthing efforts to calibrate remote sensing data designed to diagnose nutrient deficiencies in crops17,18,24,25. However, the method presented here can currently still be considered the gold standard for InsP analyses and will be instrumental to discover new functions of these intriguing messengers in plants.
The authors have nothing to disclose.
This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC-2070 – 390732324 (PhenoRob), the Research Training Group GRK2064 and individual research grants SCHA1274/4-1 and SCHA1274/5-1 to G.S.. We also thank Li Schlüter and Brigitte Ueberbach for technical assistance.
Centrifuge | Eppendorf AG | model: 5430 R | |
Diammonium hydrogen phosphate, ≥97 % | Carl Roth GmbH + Co. KG | 0268.1 | |
Ethylenediamine tetraacetic acid disodium salt dihydrate, ≥99 %, p.a., ACS | Carl Roth GmbH + Co. KG | 8043.2 | |
Falcon 12-well clear flat bottom TC-treated multiwell cell culture plate, with lid, individually wrapped, sterile | Corning Inc. | 353043 | |
Fraction collector | LAMBDA Instruments GmbH | model: OMNICOLL single channel collector | |
Growth incubator | poly klima GmbH | model: PK 520-LED | |
HPLC pumps | Kontron Instruments | model: 420 | |
HPLC syringe for Rheodyne valves, 1 mL | Hamilton Company | 81365 | |
Injector for HPLC | Supelco | model: Rheodyne 9725 | |
Inositol, myo-[1,2-3H(N)] | American Radiolabeled Chemicals Inc. | ART 0261 | |
Liquid nitrogen | University, Chemistry Department | ||
Liquid scintillation counter | PerkinElmer Inc | model: TRI-CARB 2900TR | |
Micro pestle | Carl Roth GmbH + Co. KG | CXH7.1 | |
Mixed cellulose Eester filter, ME range (ME 24), plain, 0.2 µm pore size, 47 mm circle | GE Healthcare Life Sciences | 10401770 | |
Mixer for HPLC | Kontron Instruments | model: M 800 | |
Murashige & Skoog medium, salt mixture | Duchefa Biochemie | M0221 | |
OriginPro software | OriginLab Corp. | ||
Orthophosphoric acid, ≥85 %, p.a., ISO | Carl Roth GmbH + Co. KG | 6366.1 | |
Partisphere 5 µm SAX cartridge column, 125 x 4.6 mm | Hichrom Limited | 4621-0505 | |
Perchloric acid, 70 %, 99.999 % trace metals basis | Sigma-Aldrich | 311421 | |
Petri dish, square, PS, clear, 120/120/17 mm, sterile | Greiner Bio One International GmbH | 688161 | |
pH-indicator paper pH 5.5 – 9.0, Neutralit | Merck KGaA | 109564 | |
Phytagel | Sigma-Aldrich | P8169 | |
Potassium carbonate | Carl Roth GmbH + Co. KG | P743.2 | |
Safe-Lock tubes, 1.5 mL | Eppendorf AG | 30120086 | |
Sample loop for 9725 injectors, volume 2 mL, PEEK | Supelco | 57648 | |
SNAPTWIST scintillation vial, 6.5 mL | Simport Scientific Inc. | S207-5 | |
Sterile bench | LaboGene | model: ScanLaf MARS 900 | |
Sucrose, ≥99,5 %, p.a. | Carl Roth GmbH + Co. KG | 4621.1 | |
Ultima-Flo AP liquid scintillation cocktail | PerkinElmer Inc | 6013599 | |
Ultra-pure deionized water | Milli-Q | ||
Wrenchless WVS End Fitting Kit | Hichrom Limited | 4631-1001 |