TY - JOUR
AU - Van Meeteren, U.
AB - Abstract A new method is presented that enables the induction of embolisms in a fraction of all xylem vessels, based on diameter, at one cut end of a stem segment. The method is based on the different capillary characteristic of xylem vessels of different cross‐sectional size. To verify the method, air embolisms were induced in cut xylem vessels of chrysanthemum (Dendranthema×grandiflorum Tzvelev cv. Cassa) stem segments at different xylem tensions and compared with the distribution of gas‐filled vessels as visualized by cryo‐scanning electron microscopy (Cryo‐SEM). At −6 kPa xylem pressure, air‐entrance was only induced in large diameter vessels (>30 μm), while at −24 kPa embolisms were induced in almost all xylem vessels (>10 μm). Although the principle of the embolization method worked well, smaller diameter vessels were observed to be embolized than was expected according to the calculations. The role of cross‐sectional shape and contact angle between xylem sap and vessel wall at the menisci are discussed. After correction for the observed (diameter independent) deviation from circularity of the cross‐sectional vessel shape the contact angle was calculated to be approximately 55°. Hydraulic resistance (Rh) measurements before and after embolization showed that the effect of embolizing only large diameter cut xylem vessels had only a small influence on overall Rh of a stem segment. Embolizing all cut xylem vessels at one cut end almost trebled overall Rh. The difference was discussed in the light of the networking capacity of the xylem system. Dendranthema grandiflorum, chrysanthemum, air, embolism, xylem, hydraulic conductance, hydraulic resistance, cryo scanning electron microscopy. ψxp, xylem pressure, Rh, hydraulic resistance, SEM, scanning electron microscopy, Ac, cross‐sectional surface area xylem vessel, Pc, perimeter xylem vessel, τ, surface tension, rc, radius xylem vessel, Θ, contact angle between water and the cell wall of xylem vessels. Introduction Air embolisms in xylem conduits can partially or completely block the water transport path between roots and sinks for water in plants. The consequent increase in hydraulic resistance (Rh) of the xylem may cause severe water stress and in extreme cases even plant death (Pickard, 1981). In intact plants, air embolisms can be the result of cavitation, due to freeze–thaw cycles or severe water stress (Sperry and Sullivan, 1992). Air embolisms may also occur as a result of mechanical injury of xylem conduits, for example, due to storm damage. Besides these natural reasons, well‐considered human action such as the harvest of cut flowers by cutting them in air, may also lead to air embolisms in xylem vessels that are later expected to conduct water again. Insufficient removal of these air embolisms may result in serious water stress which may lead to early leaf wilting as previously shown in chrysanthemum cut flowers (Van Meeteren, 1992). Until now, little is known about the physiological background of this phenomenon, despite its common occurrence. Vessel characteristics possibly play a role. Large diameter vessels are thought to be responsible for a disproportionate fraction of the overall flow through a stem because of the r4 relation of flow through circular pipes (Hagen‐Poiseuille relation; Dimond, 1966; Gibson et al., 1984; Pickard, 1981). Consequently, large diameter vessels are important in preventing water stress, although in chrysanthemum stems, bordered pits at vessel‐to‐vessel connections significantly add to the hydraulic resistance of the xylem system (van Ieperen et al., 2000). However, large diameter vessels are also thought to cavitate more easily than narrow vessels (Netting, 2000). The bordered pits, on the other hand, restrict the spread of air in the cut flower stem and therefore increase the hydraulic resistance due to air emboli, because they do not allow the passage of air–water interfaces (Zimmermann, 1983). There is a lot of evidence that embolized vessels may be repaired. Various studies on different plant species using Cryo Scanning Electron Microscopy (Cryo‐SEM) or measuring hydraulic resistance show that, after cavitation, xylem vessels can recover from air embolisms (Canny, 1997; McCully et al., 1998; Sperry et al., 1987; Utsumi et al., 1998; Yang and Tyree, 1992). The mechanism by which the plant manages to remove these air embolisms is still under debate (Holbrook and Zwieniecki, 1999; Tyree et al., 1999). An interesting mechanistic model has been developeded for air removal after cavitation (Yang and Tyree, 1992). In this model, capillary forces compress the air inside the trapped air bubble and thus generate the driving force for the removal of these air embolisms. A simple form of the capillary equation was used to calculate the driving force for air removal, assuming that the vessel is circular and the contact angle between xylem cell wall material and water at the meniscus is 0°. Model calculations showed that stems with large diameter vessels recover more slowly from air embolisms than stems with small diameter vessels (Yang and Tyree, 1992). Thus, vessel diameter probably not only influences the hydraulic resistance of the xylem system but also the recovering process from air embolism. Unfortunately, reality is more complex since plants usually contain vessels in a wide range of diameters (Zimmermann, 1983), which considerably complicates the process. Furthermore, the model calculations depend on the assumed cross‐sectional circularity of the embolized vessels and on the assumed 0° contact angle. These assumptions are both disputable (Pickard, 1981). Until now, experimental evidence for the role of vessel diameter in recovering from air embolisms is poor, due to the lack of a good method to embolize vessels of different diameters within samples of comparable xylem structure. The aim of this paper is to present and verify a method that enables controlled induction of air embolisms in cut xylem vessels at one cut end of a stem segment, which are larger than a certain pre‐defined cross‐sectional size. The embolization method is basically verified by visualization of the water and air‐filled cut xylem vessels at the cut surface and measuring their cross‐sectional size using Cryo Scanning Electron Microscopy (Cryo‐SEM) (Canny, 1997; Utsumi et al., 1998) in combination with image analysis techniques. Additionally, hydraulic resistance (Rh) measurements were done before and after embolization to investigate the effect of partial blockage of the xylem on overall Rh. Image analysis techniques were also used to measure the cross‐sectional shape of the vessels actually embolized in order to determine their deviation from circularity. Materials and methods Method basics and theoretical background The embolization method basically consists of the controlled suction of air into a pre‐defined fraction of all initially water‐filled cut xylem vessels at one cut end of an excised stem segment. This pre‐defined fraction consists of vessels that are larger than a certain cross‐sectional size. The controlling factor for water entry is the actual xylem pressure at the cut end of the stem segment where water usually enters, but which is temporarily exposed to air to enable air entrance. The capillary characteristics of the vessels, which strongly depend on their cross‐sectional size, are used as the distinguishing factor that determines whether a vessel embolizes or not. A common measurement system for Rh measurements is used to control the xylem pressure during embolization. The use of such a measurement system also makes it possibile to measure Rh before and after embolization to determine the effect of embolization on Rh of the stem segment. After cutting, air–water interfaces (menisci) appear at all the ends of cut xylem vessels. Upon exposure to air, such a meniscus only starts to move inwards if the inward force exceeds the outward force on the meniscus (Finw>Foutw). This could be the case if the xylem is under tension (negative pressure). However, xylem pressures are not always low enough to enforce the entrance of air in all diameter vessels at the cut end. Capillary characteristics of the xylem vessels tend to oppose the entrance of air. Finw depends on the pressure in the xylem vessel (ψxp [Pa]) and on the cross‐sectional surface area of the vessel (Ac, [m2], equation 1). Foutw, the capillary force that tends to keep the meniscus of the water column at the cut end of the vessel, is generated by the outward component of the surface force along the edge of the meniscus (Nobel, 1983). Foutw depends on the surface tension (τ, [N m−1]) of the xylem sap, the perimeter of the xylem conduit (Pc, [m]) and the contact angle (Θ, [−]) between xylem sap and cell wall at the meniscus (equation 2).   1  2 Combining equation 1 and 2 gives equation 3 where ψxp is the xylem pressure at the cut surface at which a xylem vessel with a perimeter–area ratio Pc/Ac starts to embolize.   3 Xylem sap is usually very dilute. Therefore τ is assumed to be equivalent to that of water (0.072 N m−1 at 20 °C). Xylem vessels are often assumed to be circular in cross‐section and the contact angle is often assumed to be 0° (Pickard, 1981). Following this simplification, substitution of the perimeter (2πrc) and the cross‐sectional area ( \({{\pi}r_{\mathrm{c}}^{2}}\) ) of circular shapes into equation 3 gives equation 4, in which rc [m] is the radius of the xylem vessel at its cut end.   4 Thus, according to the last equation, if ψxp is −6 kPa only cut vessels with a radius larger than 25 μm should embolize upon exposure to air. If ψxp is −24 kPa, all vessels with a radius larger than 6 μm should embolize. Deviation of the cross‐sectional shape of the meniscus from circularity tends to increase the calculated critical radius, while an increased contact angle tends to decrease the calculated critical radius. These pressures are certainly much smaller than what could normally be expected in xylem conduits of transpiring plants (Nobel, 1983). However, xylem pressures between 0 and −50 kPa may easily be established in water‐filled xylem vessels in excised stem segments using a simple laboratory system for hydraulic resistance (Rh) measurements. In fact, the use of small pressure differences to induce flow (<50 kPa) is common practice during hydraulic resistance determinations (Chiu and Ewers, 1993; Sperry et al., 1988; van Ieperen et al., 2000), although not always negative xylem pressures are induced. In contrast to the pressure gradient required to induce flow for Rh measurements, induction of air embolisms depends on the negative xylem pressure located at the cut end of the stem segment where water usually enters. During Rh‐measurements with negative xylem pressures actual xylem pressure at this point is usually close to atmospheric pressure. Fortunately, the pressure gradient during Rh measurements is clearly related to flow. Withdrawal of the water supply from the cut end of the stem segment stops the flow of water through the stem segment and forces the xylem pressure in the whole sample to equilibrate to a constant uniform negative xylem pressure (ψxp), caused by the pulling pressure at the other cut end of the stem segment. If ψxp is low enough to induce air entrance into a cut xylem vessel (equation 4), air enters only until the next vessel‐to‐vessel connection, because capillary forces in pit membranes of bordered pits do not allow the passage of air–water interfaces at these relatively small xylem pressures. Plant material and sample preparation Samples were collected from mature Chrysanthemum (Dendranthema×grandiflorum Tzvelev cv. Cassa) plants. Plants were propagated via stem cuttings, transplanted into pots and grown until commercial maturity (Van Meeteren et al., 2000). The length of the growth period from transplanting to sample collection was approximately 12 weeks. One night before cutting the plants were well irrigated and put in the dark for 12 h to ensure maximal turgidity and minimal presence of natural air embolisms. The plants were approximately 90–100 cm long and fully flowered. Each sample (stem segment) originated from an individual plant and was collected from the same position on the plant. All manipulations with plants and stem segments in the laboratory were done under water to prevent the entrance of air into the vessels at their cut ends. Stem segments of 40 cm length were cut from the plants at 5 cm above the root/shoot junction with sharp shears and re‐cut with a razor blade to approximately 30 cm length by cutting off 5 cm from both ends. A new razor blade was used for each cut. The length of the stem segments was exactly measured. The stem segment was always longer than the maximal length of cut open vessels (approximately 25 cm), which was determined using the latex particle method previously described (Zimmermann and Jeje, 1981) on three similarly grown chrysanthemum plants (Van Ieperen et al., 2000). Leaves were removed from the stem segment with a razor blade, leaving 1 cm of the petioles on the stem. Each stem segment had 5–6 petiole stubs and was approximately 5–6 mm in diameter. A silicone tube was pushed over the upper cut end of the stem segment (cut end at largest distance from the roots) and attached to the Rh‐measurement system (see below and Fig. 1). During preparation a small positive pressure was applied to the upper side of the stem segment to avoid penetration of air into open vessels at the lower cut‐end. The time between collection from the plant and the start of the Rh measurement was approximately 15 min. Fig. 1. View largeDownload slide Apparatus for (A) measuring hydraulic resistance (Rh) and (B) controlling the xylem pressure at the basal cut end of excised stem segments during exposure to air (B is A but without water supply to the basal cut end). To induce a pressure gradient (of below atmospheric xylem pressures), the upper part of a stem segment was connected to a flexible silicon tube holding a hanging water column of de‐gassed water, which could be varied in length. During Rh measurements the basal cut surface of the stem segment was positioned approximately 0.5 cm below the surface of the water in the container on the balance. The pressure gradient to induce flow was determined by the length of the water column (h1−h0) and the length of the stem segment. The length of the water column which induced the pressure at the basal cut end during embolization was slightly smaller ( \({h_{1}^{{\ast}}}\) −h0). Fig. 1. View largeDownload slide Apparatus for (A) measuring hydraulic resistance (Rh) and (B) controlling the xylem pressure at the basal cut end of excised stem segments during exposure to air (B is A but without water supply to the basal cut end). To induce a pressure gradient (of below atmospheric xylem pressures), the upper part of a stem segment was connected to a flexible silicon tube holding a hanging water column of de‐gassed water, which could be varied in length. During Rh measurements the basal cut surface of the stem segment was positioned approximately 0.5 cm below the surface of the water in the container on the balance. The pressure gradient to induce flow was determined by the length of the water column (h1−h0) and the length of the stem segment. The length of the water column which induced the pressure at the basal cut end during embolization was slightly smaller ( \({h_{1}^{{\ast}}}\) −h0). Rh measurements To determine Rh of a stem segment, the water flow rate is measured by pulling water through the stem segment at a known pressure gradient. To standardize the Rh measurement an hydraulic pressure difference of −24 kPa is used. During measurements, flow was always in natural direction, for example, from the lower cut end (closest to the roots) to the upper cut end of the stem segment. A hanging water column of degassed water in a flexible tubing system provided the pulling pressure at the upper cut end of the stem segment (Fig. 1A). The pressure gradient was calculated from the length of the stem segment and the difference in height between the water level in the container on the balance and the lower open end of the tubing system. Changes in the pressure gradient during Rh measurements were negligible because the water level in the container changed less than 3 mm during each individual measurement. The rate of water flow through the stem segment was calculated from weight changes measured with a Sartorius LC3200 d balance interfaced to a personal computer. Weight was measured at 1 s−1 sample rate and averaged over 30 s. Two subsequent averaged weights were used to calculate the rate of water flow. Calculated flow rates were corrected for direct evaporation from the container, which was measured before and after each individual measurement and during the embolization treatment. No correction was done for the change in submerged volume of the stem segment in the solution because the calculated error on the rate of water flow (of a 5 mm diameter stem segment submerged in a solution in a 9.5 cm diameter container) is lower than 0.5%. This error could even be lower due to the porosity of the stem segment. Flow measurements were done using an aqueous solution containing 1.49 mol m−3 NaHCO3, 0.67 mol m−3 CaCl2, 4.8 mmol m−3 CuSO4, (pH 6.6; 20±2 °C). This solution sufficiently delays the long‐term increase in Rh, which, as in other species (Sperry et al., 1988), also occurs in excised Chrysanthemum stem segments when using deionized water as the flowing solution (Van Meeteren et al., 2000). Due to the almost neutral pH and low concentration of solutes, this solution is likely to have a surface tension and contact angle similar to water or xylem sap. To justify the use of the standard pressure of −24 kPa during Rh measurements and the use of defoliated stem segments, tests were done to check whether measured Rh was affected by the magnitude of the pulling pressure and whether leaf removal influenced Rh. Two sets of four different pressure gradients were subsequently applied on a single stem segment (and repeated on four different stem segments): one set of pressures was applied before and one after leaf removal. To check if all vessels near the lower cut surface were involved in water transport, tests were done with a 1% acid Fuchsin solution (Sigma Chemical Co., St Louis, USA). Stem segments were harvested, prepared and attached to the measurement system as described above. Flow was permitted during 2 min after dye appeared at the upper cut end of the stem segments. Fresh sections were cut at several distances from the bottom cut surface with a razor blade and examined under a light microscope. Air embolization treatment After the first Rh measurement on a stem segment without initial air embolisms (at −24 kPa) one of the embolization pressures (−6.0, −10.4 or −24.0 kPa) was established. Then the water level in the container was lowered just below the lower cut end of the stem segment in order to (1) stop the flow of water, (2) induce equilibration of ψxp along the stem segment and (3) allow air entrance into cut vessels at the lower cut end. After a period of approximately 6 min of exposure to air the water level was raised again and a second Rh measurement (at −24 kPa) was made. At −6 and −24 kPa four replicate cycles (Rh measurements–embolization–Rh measurement) were done, each cycle on a new stem segment. The whole procedure was repeated after approximately 3 weeks with new stem segments collected from a new batch of plants. During the second batch of measurements an extra embolization pressure (−10.4 kPa) was added. Cryo Scanning Electron Microscopy Stem segments were prepared for visual analysis of embolized vessel ends using Cryo‐SEM. The basal 10–15 cm of these stem segments were carefully frozen in liquid nitrogen at the following stages during a cycle: (1) immediately after harvest to check for the presence of initial air‐embolisms; (2) after 6 min of exposure to air to verify the effect of ψxp on the embolization of xylem vessels. The frozen stem segments were stored in liquid nitrogen (−196 °C) prior to final preparation for Cryo‐SEM. To visualize the contents of the xylem vessels in the stem segments, the lower 1 cm portions of the frozen stem segments were mounted on hollow brass stubs using Tissue Tek adhesive (No. 4583 Miles Inc., Elkhart, Indiana‐USA) with the lower cut ends positioned upwards. Cut surfaces of stem segments that had been frozen immediately after harvest were not suitable for direct examination of vessel content due to adhering, frozen water. Therefore, these samples were sawn at 2 mm and 3 cm from the basal cut surface and planed with steel blades or a glass knife in a cryo‐microtome at a maximum temperature of −30 °C. The cut surfaces of samples taken immediately after air entrance were examined without any re‐cutting. The surface was etched for 5 min at −89 °C and 10−4 Pa in an Oxford CT 1500 HF cryo‐transfer unit and sputter‐coated with platinum. Vessel contents were investigated at the cut surface in a JEOL 6300F field emission scanning electron microscope at −170 °C and 2.5 kV. Digitized images of the xylem were recorded at several magnifications. Cryo‐SEM images from the stem segments that were frozen immediately after embolization at −6 and −24 kPa, were analysed using image analysis software (Scion Image, a free PC version of NIH‐Image, Scion Corporation, Maryland, USA). Each of the selected images contained at least four and maximal six vascular bundles with embolized vessels. The cross‐sectional surface area (Ac), the perimeter (Pc) and the diameters of the smallest and largest inscribed ellipses of all embolized vessels were analysed. The critical diameter for air‐entrance (which equals 2rc) was defined as the average of the equivalent circle diameters of the largest and smallest inscribed ellipses (Pickard, 1981). To check the circularity of the embolized vessels, (2/rc) ratios (see equation 4) of embolized vessels were calculated, and compared with the actual measured Pc/Ac ratios. Results Stem segment characteristics After applying dye at the basal cut end of the stem segments, stain was found in walls of xylem vessels of all sizes. Most coloured vessels (small and large diameter) were located in the approximately 20–25 vascular bundles in a cross‐section. However, not all vascular bundles were coloured at all heights in the stem segments. At 5 mm above the basal cut surface 4–5 vascular bundles were unstained, probably because they were previously connected to leaves that were removed from the stem segment before the measurement started. These bundles are disconnected from the negative pressure applied at the top of the stem segment. This explanation was confirmed by observations of colouring above the highest, second highest and third highest petiole stubs on the stem segments, where respectively, all, all minus one and all minus two vascular bundles were stained. No dye was found in the central pith area of the stem segment except close (<2 mm) to the basal cut surface. The length distribution of the xylem vessels, determined by the latex particle method (Zimmermann and Jeje, 1981), and measured up from the cut surface, was found to be most accurately described by the exponential curve N=100 e(−αL), where N is the percentage of vessels longer than length L (measured from the cut surface). (R2>0.95 in all individual determinations; α=0.0275±0.0021 (mean and standard deviation); n=3). From this relationship it can be calculated that 75% of the vessels were longer than 10 mm and 50% longer than 25 mm. The maximal vessel length was between 24 and 28 cm. Observations during the embolization treatment confirmed that in none of the stem segments xylem vessel length exceeded stem segment length because no air appeared at the upper cut end during and after exposure of the basal cut end to air. Verification of the method by Cryo‐SEM In the Cryo‐SEM images water‐ and air‐filled vessels could easily be distinguished by their structure and contents. No embolized vessels were observed in stem segments, which were frozen immediately after harvest (Fig. 2A). This strongly indicates the absence of natural air embolisms before embolization. Xylem pressure during embolization clearly influenced the cross‐sectional size fraction of embolized vessels in a stem segment. After embolization at −6 kPa, emboli were almost exclusively observed in large diameter vessels (Fig. 2B, C), while after embolization at −24 kPa many of the small diameter vessels also became embolized (Fig. 2D, E). Not all 20–25 vascular bundles in a cross‐section contained embolized vessels: the vessels in four or five vascular bundles remained completely water‐filled. This is in agreement with the results of the staining experiment, which made clear that probably a few bundles at the lower cut surface are disconnected from the pulling pressure at the upper cut end of the stem segment. Consequently, these disconnected bundles do not add to the water flow through the stem segment in this experimental set up. The number of embolized vessels at −24 kPa was much larger than at −6 kPa. This was almost completely due to the large number of embolized vessels with diameters lower than 20 μm. Frequency distributions of measured diameters of embolized vessels after air‐entrance at −6 and −24 kPa are shown in Fig. 3. The distribution at −24 kPa reflects more or less the diameter distribution of all vessels since almost all vessels were embolized. The distribution at −6 kPa shows a break between the 25–30 and 30–35 μm diameter classes: instead of an increasing percentage of embolized vessels in the lower diameter classes (below 25–30 μm) a decreasing percentage was observed. This shows that most vessels with a diameter smaller than approximately 30 μm did not embolize at −6 kPa while almost all larger diameter vessels did. This critical diameter for embolization at −6 kPa (30 μm) is smaller than was calculated by equation 4 (50 μm). The cross‐sectional shape of embolized conduits was clearly not circular (Fig. 4) as was assumed in equation 4. The actual measured Pc/Ac ratio of embolized vessels after embolization at −6 and −24 kPa was on average 1.13 times the calculated Pc/Ac ratio (=2/rc). The difference between calculated and measured Pc/Ac was independent of vessel diameter and critical cross‐sectional vessel size (Fig. 4). Rh test measurements The test measurements revealed that measured Rh was independent of the applied negative pressure and of leaf removal (Fig. 5). Linear relationships were found between the applied pressure gradients and observed rates of water flow, while the absence of leaves did not affect the slope of the pressure‐flow rate relationships (Fig. 5). The positive intercept at the flow‐axis of the curve of the stem segment with leaves agreed with the measured transpiration of the stem segment with leaves (which was measured when no pressure was applied). After starting an Rh measurement with a new stem segment, the measured flow rate of water sometimes started at approximately 85–90% of its final flow rate. The final flow rate was reached within 10–40 min and afterwards remained constant for a period of at least 3–4 h. During this period sudden changes in pressure gradient usually caused transient responses in flow rate, which, however, never lasted longer than 2 min (results not shown). Fig. 2. View largeDownload slide Cross‐sectional faces of Dendranthema morifolium Ramat cv. Cassa stem segments examined by Cryo‐SEM. (B, D) A sector of a whole stem cross‐section containing several vascular bundles with water‐ and air‐filled xylem vessels. (A, C, E) Details of the xylem tissue. (A) 5 mm above basal cut surface after harvest under water; (B) basal cut surface after embolization at −6 kPa; (C) detail of (B); (D) basal cut surface after embolization at −24 kPa; (E) detail of (D); c, cortex; e, empty xylem vessel; f, xylem fibre cells; p, pith; ph, phloem; w, water‐filled xylem vessel; x, xylem. All bars represent 100 μm. Fig. 2. View largeDownload slide Cross‐sectional faces of Dendranthema morifolium Ramat cv. Cassa stem segments examined by Cryo‐SEM. (B, D) A sector of a whole stem cross‐section containing several vascular bundles with water‐ and air‐filled xylem vessels. (A, C, E) Details of the xylem tissue. (A) 5 mm above basal cut surface after harvest under water; (B) basal cut surface after embolization at −6 kPa; (C) detail of (B); (D) basal cut surface after embolization at −24 kPa; (E) detail of (D); c, cortex; e, empty xylem vessel; f, xylem fibre cells; p, pith; ph, phloem; w, water‐filled xylem vessel; x, xylem. All bars represent 100 μm. Fig. 3. View largeDownload slide Percentage of all embolized vessels per diameter class (diameter is defined as the average of the circle diameters of the largest and smallest inscribed ellipse in an embolized vessel) after air‐entrance at −6 kP and −24 kP xylem pressure, obtained from three images per pressure containing each at least four vascular bundles. Fig. 3. View largeDownload slide Percentage of all embolized vessels per diameter class (diameter is defined as the average of the circle diameters of the largest and smallest inscribed ellipse in an embolized vessel) after air‐entrance at −6 kP and −24 kP xylem pressure, obtained from three images per pressure containing each at least four vascular bundles. Fig. 4. View largeDownload slide Relationship between vessel diameter (diameter defined as the average of the circle diameters of the largest and smallest inscribed ellipse in an embolized vessel) and the quotient of the actual measured and the calculated perimeter/cross‐sectional area ratio (Pc/Ac) of the same conduits. Calculated Pc/Ac is based on vessel diameter and assumed circularity. Circles: embolized conduits after air‐entrance at −6 kPa (n=273). Squares: embolized conduits after air‐entrance at −24 kPa (n=610). Fig. 4. View largeDownload slide Relationship between vessel diameter (diameter defined as the average of the circle diameters of the largest and smallest inscribed ellipse in an embolized vessel) and the quotient of the actual measured and the calculated perimeter/cross‐sectional area ratio (Pc/Ac) of the same conduits. Calculated Pc/Ac is based on vessel diameter and assumed circularity. Circles: embolized conduits after air‐entrance at −6 kPa (n=273). Squares: embolized conduits after air‐entrance at −24 kPa (n=610). Fig. 5. View largeDownload slide Pressure‐flow relationships measured on individual Dendranthema×grandiflorum Tvelev cv. Cassa stem segments before (circles) and after (squares) leaf excision (averages of four stem segments). Fig. 5. View largeDownload slide Pressure‐flow relationships measured on individual Dendranthema×grandiflorum Tvelev cv. Cassa stem segments before (circles) and after (squares) leaf excision (averages of four stem segments). Rh before and after embolizing the xylem at different xylem pressures Typical courses of measured water flow rates before and after air‐entrance at different xylem pressures are shown in Fig. 6. Calculated hydraulic resistances before and after air‐entrance are presented in Table 1. Rh before embolization was approximately similar in all treatments and replicates, although the variability between individual stem segments was substantial. Surprisingly, measured water flow rate and Rh were not decreased after embolization at −6 kPa. After embolization at −24 kPa in contrast, Rh almost trebled. Based on means and standard deviations presented in Table 1, Rh seem to be hardly changed after embolization at −10.4 kPa. However, comparisons of paired measurements of Rh (before and after air‐entrance) on individual stem segments showed consistent increases in Rh (Fig. 6). The increase in Rh after embolization at −10.4 kPa appeared in all stem segments and ranged between 8% and 23%. Fig. 6. View largeDownload slide Comparison of typical water flow rate patterns through stem segments before and after embolizing different parts of the xylem. Embolizing xylem pressures: −6 kPa (only large vessels), −10.4 kPa (large and intermediate diameter vessels) and −24 kPa (large, intermediate and small diameter vessels). The pressure gradient to induce flow during the flow rate measurement was 80 kPa m−1. Fig. 6. View largeDownload slide Comparison of typical water flow rate patterns through stem segments before and after embolizing different parts of the xylem. Embolizing xylem pressures: −6 kPa (only large vessels), −10.4 kPa (large and intermediate diameter vessels) and −24 kPa (large, intermediate and small diameter vessels). The pressure gradient to induce flow during the flow rate measurement was 80 kPa m−1. Table 1. The effect of xylem pressure during air‐entrance on changes in hydraulic resistance (Rh) due to air‐blockages Stem segments were collected from two groups of plants (A and B) which were grown separate in time. Rh was determined before and after air entrance on each individual stem segments. Mean and standard deviation of Rh before and after air‐entrance at each pressure and group combination are indicated Xylem pressure during air‐entrance [kPa]   Rh before (kPa m−1 mol−1 s)   Rh after (kPa m−1 mol−1 s)   Group   n   −6.0  2.1±0.32  2.0±0.31  A  4    2.2±0.49  2.2±0.52  B  3  −10.4  –  –  A  –    2.1±0.49  2.4±0.64  B  8  −24.0  2.1±0.28  5.7±1.41  A  4    2.3±0.11  7.2±1.10  B  4  Xylem pressure during air‐entrance [kPa]   Rh before (kPa m−1 mol−1 s)   Rh after (kPa m−1 mol−1 s)   Group   n   −6.0  2.1±0.32  2.0±0.31  A  4    2.2±0.49  2.2±0.52  B  3  −10.4  –  –  A  –    2.1±0.49  2.4±0.64  B  8  −24.0  2.1±0.28  5.7±1.41  A  4    2.3±0.11  7.2±1.10  B  4  View Large Discussion The technique of embolizing cut xylem vessels strongly depends on the actual established hydraulic pressure at the lower cut surface where air could possibly enter the cut xylem vessel ends. It is therefore important to be sure about the magnitude of hydraulic pressure localized at that specific position. The absence of flow and the continuity of the water column between the basal cut end of the stem segment and the lower open end of the flexible tube (Fig. 1) ensure this well‐defined xylem pressure at the cut end of the stem segment. Immediately after the start of air entrance in a vessel the pressure on the moving meniscus of the water column further decreases, due to the vertical position of the stem segment in the system (Fig. 1). This facilitates further movement of the meniscus inward to the embolizing vessel. The results show that the proposed method enables the controlled embolization of xylem conduits larger than a certain diameter within samples of comparable xylem structure (Fig. 2). The diameters of embolized vessels in Cryo‐SEM images only roughly agreed with calculated critical vessel diameters (Fig. 3). This indicates that equation 4, which reflects the simple capillary equation, could not be used as such. The simplification of equation 3 to equation 4 is based on two assumptions: cross‐sectional circularity of the vessels and a 0° contact angle between xylem cell wall and meniscus at the air–water interface. The analysis of the cross‐sectional shape of a large number of embolized vessels (Fig. 4) revealed that the assumption of circularity in equation 4 was not justified. However, the diameter independent (Fig. 4) deviation from circularity could not explain the difference between the observed and the calculated critical diameter for embolization (approximately 30 μm versus 50 μm at −6 kPa). Instead, the observed deviation from circularity even increased the difference between the observed and the calculated critical diameter. A larger than 0° contact angle on the other hand decreases the difference between the observed and the calculated critical diameter. Taking into account the observed deviation from circularity and the observed critical diameter of approximately 30 μm, the contact angle must have been approximately 55°. Earlier remarks (Pickard, 1981) concerning the contact angle at xylem walls strengthen this explanation; he mentioned substantially higher contact angles than 0° between water and wall material of xylem conduits (up to 45°). Recent measurements of the contact angle in six species using a completely different approach, showed that it varied between 42° and 55° (Zwieniecki and Holbrook, 2000). The deviation of the contact angle from 0° could be of importance for the removal of air embolisms. According to the model of Yang and Tyree this larger contact angle could significantly decrease the rate of recovering from air embolisms (Yang and Tyree, 1992). The general response of flow during the Rh determination, including the decline 3–4 h after applying the pressure gradient on the stem segment, was about the same as that reported in the literature when using de‐ionized water or a low concentration sodium chloride solution (Sperry et al., 1988). They explained the long‐term decline by microbial clogging. A simple explanation for the rise of the flow rate during the first 30–40 min of the measurement period could not be given. A non‐degassed solution was used to prevent the undesired rapid removal of the air embolisms in the xylem sap after reapplying the solution to the basal cut end of the stem segment. The possibility of artefacts due to the use of non‐degassed water needs to be considered. It seems, however, that the use of non‐degassed solutions did not result in extra embolisms, as the flow rate initially increased instead of decreased and remained constant during the following 3–4 h. Solutions with low pH (Sperry et al., 1988) were not used in order to prevent unexpected effects on forces between the cell wall and the water: low pH could possibly change the contact angle between wall and water, which affects the embolization treatment. It is possible that some kind of adaptation appeared at the start of the Rh determinations in response to the change in content of the xylem vessels. It is known that the Rh of chrysanthemum stems could be influenced by the ionic composition of the xylem fluid (Van Ieperen et al., 2000). This explanation is strengthened by the fact that no transient responses were observed after changes in pressure gradients applied after the initial period of increase of flow rate during the test measurements on stem segments with and without leaves. The Rh of an non‐embolized stem segment is the collective resistance of many xylem conduits in parallel, which are interconnected and which have different lengths. In chrysanthemum stem segments 50% of the cut vessels at the basal cut surface were shorter than 25 mm. Vessel‐to‐vessel connections significantly add to Rh. Prediction of the overall Rh of whole stem segments is difficult even without including air‐embolisms, due to the non‐normal distribution of vessel lengths and vessel diameters, and the unknown correlation between vessel length and diameter (Chiu and Ewers, 1993; Van Ieperen et al., 2000). The effect of air embolisms in the cut vessels at the basal end of the stem segment on overall Rh is therefore difficult to assess. Usually, only a small portion of the path for water flow is blocked. Even if all the severed vessels are blocked by air, water can move via cell walls along the shortest distance to the intact vessels whose ends are near the cut surface. Once water gets into these vessels it may flow via the normal network of functioning vessels. The following simple calculation, ignoring the actual vessel lengths, diameters and resistances of pit membranes, shows that the impact of embolisms on overall hydraulic conductance (the reciprocal of Rh) is limited. Suppose that air embolisms drop the hydraulic conductance of the lowest 2.5 cm of a 30 cm long stem segment to 5% of its value before air entrance, and that the hydraulic conductance in the remaining part of the stem segment is unchanged. In that case, overall hydraulic conductance of the 30 cm long stem segment decreases by 61%. If hydraulic conductance in the lower 2.5 cm of the stem segment drops to 90%, overall hydraulic conductance decreases by 1%. These calculations partially explain why the effect of embolization at −6 kPa (Table 1) on Rh was very small, although the Cryo‐SEM images obtained from samples frozen immediately after the embolization treatment (Fig. 2B, C) clearly show that most large diameter vessels were embolized at the cut surface. On the other hand, embolization of small and large diameter vessels at −24 kPa xylem pressure (Fig. 2C, D) was accompanied with a considerable increase in Rh (Table 1). Theoretically there are several explanations for the observed discrepancy between the visually observed embolization and measured changes in Rh in the −6 kPa treatment: an unlikely explanation is that large diameter vessels are not important for water flow. This contradicts the generally assumed analogue with Poiseuilles law, and with the observed colouring of the large diameter vessels in the present experiments. Moreover, the fact that the vessels were embolized implies that they were connected to the main longitudinal water transport path and therefore did add to longitudinal water transport. A second explanation could be that all large diameter vessels were short or only embolized for a small percentage of their length. In both these cases, water could have followed relative short side‐paths via adjacent non‐embolized conduits before entering large diameter vessels, thus not significantly changing the overall hydraulic resistance. It is, indeed, possible that in the −6 kPa treatment the large diameter vessels were not completely embolized due to longitudinal tapering of the vessels. A third explanation could be that small diameter vessels, which were still water filled after embolization at −6 kPa, provided the short side‐paths to large diameter vessels that start just above the cut surface after air‐penetration at −6 kPa. These short‐side paths were not available after embolization at −24 kPa, which might have been the reason for the much larger increase in Rh after embolization at −24 kPa. Verification of the presented embolizing technique by Cryo‐SEM showed that embolizing a part of the cut xylem vessels, discriminating on diameter, is quite possible. Based on the present experiments it can be concluded that the effect of embolizing only all large diameter vessels at one cut end of a chrysanthemum stem segment hardly changes overall Rh. 3 To whom correspondence should be addressed. Fax: +31317484709. E‐mail: Wim.vanIeperen@hpc.dpw.wau.nl This work was granted by the Dutch Technology Foundation STW. We thank Adriaan van Aelst for the stimulating discussions and his helpful advice in operating the Cryo‐SEM equipment. References Canny MJ. 1997. Vessel contents of leaves after excision—a test of Scholander's assumption. American Journal of Botany  84, 1217–1722. Google Scholar Chiu ST, Ewers FW. 1993. The effect of segment length on conductance measurements in Lonicera fragrantissima. Journal of Experimental Botany  44, 175–181. Google Scholar Dimond AE. 1966. Pressure and flow relations in vascular bundles of the tomato plant. Plant Physiology  41, 119–131. Google Scholar Gibson AC, Calkin HW, Nobel PS. 1984. Xylem anatomy, water flow and hydraulic conductance in the fern Cyrtomium falcatum [Pteridophyta]. American Journal of Botany  84, 564–574. Google Scholar Holbrook NM, Zwieniecki MA. 1999. Embolism repair and xylem tension: do we need a miracle? Plant Physiology  120, 7–10. Google Scholar McCully ME, Huang CX, Ling LEC. 1998. Daily embolism and refilling of xylem vessels in the roots of field‐grown maize. New Phytologist  138, 327–342. Google Scholar Netting AG. 2000. PH, abscisic acid and the integration of metabolism in plants under stressed and non‐stressed conditions: cellular responses to stress and their implication for plant water relations. Journal of Experimental Botany  51, 147–158. Google Scholar Nobel PS. 1983. Biophysical plant phyiology and ecology . New York: WH Freeman and Company. Google Scholar Pickard WF. 1981. The ascent of sap in plants. Progress in Biophysics and Molecular Biology  37, 181–229. Google Scholar Sperry JS, Donnelly JR, Tyree MT. 1988. A method for measuring hydraulic conductivity and embolism in xylem. Plant, Cell and Environment  11, 35–40. Google Scholar Sperry JS, Holbrook NM, Zimmermann MH, Tyree MT. 1987. Spring filling xylem vessels in wild grapevine. Plant Physiology  83, 414–417. Google Scholar Sperry JS, Sullivan JEM. 1992. Xylem embolism in response to freeze‐thaw cycles and water stress in ring‐porous, diffuse‐porous, and conifer species. Plant Physiology  100, 605–613. Google Scholar Tyree MT, Salleo S, Nardini A, Lo Gullo MA, Mosca R. 1999. Refilling of embolized vessels in young stems of laurel. Do we need a new paradigm? Plant Physiology  120, 11–21. Google Scholar Utsumi Y, Sano Y, Fujikawa S, Funada R, Ohtani J. 1998. Visualization of cavitated vessels inwinter and refilled vessels in spring in diffuse‐porous trees by cryo‐scanning electronmicroscopy. Plant Physiology  117, 1463–1471. Google Scholar Van Ieperen W, van Meeteren U, van Gelder H. 2000. Fluid ionic composition influences hydraulic conductance of xylem conduits. Journal of Experimental Botany  51, 769–776. Google Scholar Van Meeteren U. 1992. Role of air embolism and low water temperature in water balance of cut chrysanthemum flowers. Scientia Horticulturae  51, 275–284. Google Scholar Van Meeteren U, Van Gelder H, Van Ieperen W. 2000. Reconsideration of the use of deionized water as vase water in post‐harvest experiments on cut flowers. Postharvest Biology and Technology  18, 169–181. Google Scholar Yang S, Tyree MT. 1992. A theoretical model of hydraulic conductivity recovery from embolism with comparison to experimental data on Acer saccharum. Plant, Cell and Environment  15, 633–643. Google Scholar Zimmermann MH. 1983. Xylem structure and the ascent of sap . Berlin: Springer‐Verlag. Google Scholar Zimmermann MH, Jeje AA. 1981. Vessel length distribution in stems of some American woody plants. Canadian Journal of Botany  59, 1882–1892. Google Scholar Zwieniecki MA, Holbrook NM. 2000. Bordered pit structure and vessel wall surface properties. Implications for embolism repair. Plant Physiology  123, 1015–1020. Google Scholar © Society for Experimental Biology
TI - Induction of air embolism in xylem conduits of pre‐defined diameter
JF - Journal of Experimental Botany
DO - 10.1093/jexbot/52.358.981
DA - 2001-05-01
UR - https://www.deepdyve.com/lp/oxford-university-press/induction-of-air-embolism-in-xylem-conduits-of-pre-defined-diameter-jbky2CM1oQ
SP - 981
EP - 991
VL - 52
IS - 358
DP - DeepDyve
ER -