Variations in the composition of gelling agents affect morphophysiological and molecular responses to deficiencies of phosphate and other nutrients

Abstract

Low inorganic phosphate (Pi) availability triggers an array of spatiotemporal adaptive responses in Arabidopsis (Arabidopsis thaliana). There are several reports on the effects of Pi deprivation on the root system that have been attributed to different growth conditions and/or inherent genetic variability. Here we show that the gelling agents, largely treated as inert components, significantly affect morphophysiological and molecular responses of the seedlings to deficiencies of Pi and other nutrients. Inductively coupled plasma-mass spectroscopy analysis revealed variable levels of elemental contaminants not only in different types of agar but also in different batches of the same agar. Fluctuating levels of phosphorus (P) in different agar types affected the growth of the seedlings under Pi-deprivation condition. Since P interacts with other elements such as iron, potassium, and sulfur, contaminating effects of these elements in different agars were also evident in the Pi-deficiency-induced morphological and molecular responses. P by itself acted as a contaminant when studying the responses of Arabidopsis to micronutrient (iron and zinc) deficiencies. Together, these results highlighted the likelihood of erroneous interpretations that could be easily drawn from nutrition studies when different agars have been used. As an alternative, we demonstrate the efficacy of a sterile and contamination-free hydroponic system for dissecting morphophysiological and molecular responses of Arabidopsis to different nutrient deficiencies.

Figure 1.

Pi-deficiency responses on primary root growth and root hair development of Arabidopsis grown on different types of agar. Wild-type (Col-0) seeds were initially grown on one-half-strength MS with different agar types for 5 d and then subsequently transferred to the respective types containing medium deprived of Pi (0 μm) on vertically oriented petri plates for 2 d. A, Mark on the primary root indicates the length achieved before transferring to Pi-deprived medium. C, Development of root hairs in 5-mm sections from the primary root tip. Photographs in A and C are representative of 15 and 10 seedlings each, respectively, grown on different agar types. B and D, Increase in primary root length after transfer to Pi-deprived medium (n = 15; B) and number of root hairs (n = 10; D). Values are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05).

Figure 2.

Long-term Pi-deficiency-mediated morphophysiological responses of Arabidopsis grown on different types of agar. Wild-type (Col-0) seeds were grown on different agar types, as described in the legend to Figure 1, on P− medium for 7 d. A, Lateral roots were spread to reveal architectural details. RSA traits are representative of 15 seedlings each for different agar types. Mark on the primary root indicates the length achieved before transferring to P− medium. B, Shoots were stained with iodine solution for detection of starch accumulation. C to G, Dry weight (n = 6 replicates of a pool of 10 seedlings each; C), shoot area (n = 6; D), increase in primary root length after transfer from one-half-strength MS to P− (n = 15; E), and number (F) and length (G) of first-order and higher order lateral roots (n = 15). Values in C to G are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05). H, Histochemical GUS staining of primary root tips of CycB1;1∷uidA seedlings grown for the indicated time intervals on different agar types. GUS-stained seedlings were observed with a compound microscope. Photographs are representative of 12 to 15 seedlings. I, Primary roots of CycB1;1∷uidA seedlings grown on Sigma A-1 (A-1), Sigma A-2 (A-2), Sigma E (E), and Caisson (C) agar types showing meristematic activity (%). [See online article for color version of this figure.]

Figure 3.

ICP-MS analysis of the Pi-deprived Arabidopsis grown on different agar types. Wild-type (Col-0) seedlings were grown under Pi-deficient condition, as described in the legend to Figure 2, for 7 d. Data are presented for the ICP-MS analysis of the whole seedling (n = 6 replicates of a pool of 10 seedlings each). Values are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05).

Figure 4.

Effects of agar types on the molecular responses of Pi-deprived Arabidopsis. Real-time PCR analysis of the relative expression levels of PSI genes in the roots of wild-type (Col-0) seedlings grown under Pi-deficient condition, as described in the legend to Figure 2, for 7 d. ACT2 was used as an internal control. Data presented are means of six technical replicates ± se.

Figure 5.

Variable elemental contaminants in different batches and types of agar. Data are presented for the ICP-MS analysis of different agar types (Sigma A-1, Sigma A-2, Sigma E, and Caisson). Values are means ± se (n = 6). Different letters on the histograms indicate that the means differ significantly (P < 0.05). Values in the histograms indicate concentrations (μm) in 1.2% agar.

Figure 6.

Low-percentage (0.6%) agar altered Pi-starvation responses of Arabidopsis. Transgenic CycB1;1∷uidA Arabidopsis seedlings were initially grown on one-half-strength MS with 0.6% and 1.2% each of Sigma A-2 and Sigma E and then transferred to the respective types containing P− medium on vertically oriented petri plates for 7 d. A, RSA traits are representative of 15 seedlings each for different treatments. Mark on the primary root indicates the length achieved before transferring to Pi-deprived medium. B, Representative photographs of 12 to 15 histochemical GUS-stained primary root tips of CycB1;1∷uidA seedlings. C, Increase in primary root length after transfer from one-half-strength MS to P− (n = 15). Values are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05). [See online article for color version of this figure.]

Figure 7.

Responses of Arabidopsis to P−Fe− on different agar types. Wild-type (Col-0) seeds, grown on one-half-strength MS with different agar types for 5 d, were transferred to the respective types containing P−Fe− medium on vertically oriented petri plates for 7 d. A, RSA traits are representative of 15 seedlings each for different treatments. Mark on the primary root indicates the length achieved before transferring to P−Fe− medium. B to D, Increase in primary root length after transfer from one-half-strength MS to P−Fe− medium (n = 15; B) and number (C) and length (D) of first-order and higher order lateral roots (n = 10). Values in B to D are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05). E and F, Real-time PCR analysis of the relative expression levels of Pht1;4 (E) and IRT1 (F) in the roots of Pi−Fe− seedlings. ACT2 was used as an internal control. Data presented are means of six technical replicates ± se. [See online article for color version of this figure.]

Figure 8.

Responses of Arabidopsis starved of Fe and Zn on different agar types. Arabidopsis seedlings were grown on one-half-strength MS with different agar types and then transferred to the respective types containing Fe- and Zn-deficient media on vertically oriented petri plates for 7 d. A, RSA traits are representative of 15 seedlings each for different treatments. Mark on the primary root indicates the length achieved before transferring to Zn− and Fe− media. B to D, Increase in primary root length after transfer from one-half-strength MS to Zn− and Fe− media (n = 15; B) and number (C) and length (D) of first-order and higher order lateral roots (n = 5). Values in B to D are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05). E and F, Real-time PCR analysis of the relative expression levels of ZIP3 and ZIP9 in Zn− roots (E) and IRT1 and FER1 in Fe− roots (F). ACT2 was used as an internal control. Data presented are means of six technical replicates ± se. [See online article for color version of this figure.]

Figure 9.

Hydroponic system for an evaluation of the Pi-starvation responses of Arabidopsis. Transgenic CycB1;1∷uidA Arabidopsis seedlings were initially grown hydroponically on one-half-strength MS for 5 d and then subsequently transferred to P+ (1.25 mm Pi) and P− (0 μm Pi) for 7 d. A, Seedlings were removed from the hydroponic system and spread on agar plates for documenting the RSA traits. B, D, and E, Primary root length (B) and number (D) and length (E) of first-order and higher order lateral roots (n = 10). Values in B, D, and E are means ± se. Different letters on the histograms indicate that the means differ significantly (P < 0.05). C, Histochemical GUS staining of the primary root tip of CycB1;1∷uidA seedlings was documented as described in the legend to Figure 2. F, Real-time PCR analysis of the relative expression levels of genes involved in Pi, Fe, and Zn homeostasis in the roots. ACT2 was used as an internal control. Data presented are means of six technical replicates ± se. [See online article for color version of this figure.]