SEARCH

SEARCH BY CITATION

Keywords:

  • Cockfoot;
  • birch;
  • isotopic percentage;
  • 15N labelling;
  • nitrogen oxides;
  • NOx;
  • pollen;
  • SIMS

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References

We used secondary ion mass spectrometry to image cellular targets of nitrogen oxides (widespread air pollutants) in pollen grains of birch (Betula verrucosa Ehrh.) and cockfoot (Dactylis glomerata L.). The pollen samples were exposed to air supplemented with high doses of 15NO. The pollen grains were then fixed, dehydrated using a newly developed ‘vapour phase’ preparation method and embedded in LRW resin. Semithin sections were then analysed. Imaging was performed in scanning mode. As usual, the two isotopes 14N and 15N were imaged as 12C14N and 12C15N, respectively. The isotopic percentages of 15N were quantitatively determined either by image processing or by direct analysis. We show that the preferential areas of NO fixation in the pollen cell are the sporoderm and discrete intracytoplasmic structures that we tentatively describe as globoid-like structures similar to those encountered in seeds.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References

In industrialized countries, air pollution has become an environmental concern requiring extensive study. Nitrogen oxides are mainly produced by petrol combustion and are among the most common and hazardous of air pollutants. As air pollutants are known to increase sensitivity to pollen allergens (Charpin, 1996), it is important to determine whether nitrogen oxides are fixed by pollen and, if this is the case, to localize the sites of fixation within the pollen grains. In the absence of any radioactive isotope of nitrogen and oxygen with an appropriate half-life, we used the stable isotope 15N for the labelling and imaging of nitrogen oxides. Secondary ion mass spectrometry (SIMS) is especially well suited to the isotopic discrimination and imaging of nitrogen isotopes with remarkable sensitivity and mass resolution (Grignon et al., 1999). However, SIMS imaging and quantification (Ramseyer & Morrison, 1983; Burns & File, 1986; Thellier et al., 1993) have some difficulties that are inherent to the method and needed to be circumvented in the experimental design. In this present work we have studied the applicability of SIMS imaging to the determination of the targets of nitrogen oxides in birch and cockfoot pollen. The reason for choosing these two species is that birch and cockfoot are responsible for more than 95% of the cases of allergy to tree and grass pollens (Stewart & Thompson, 1996).

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References

Pollen

Birch (Betula verrucosa Ehrh.) pollen was harvested in April 1996 at Forges-les-Eaux, north-east of Rouen, France. The catkins were collected shortly before anther dehiscence and were stored at 25 °C on a glossy paper sheet until opening; their pollen was then sampled, sieved and stored in cryogenic vials (Nunc, PolyLabo, Strasbourg, France) at −80 °C until their experimental use. Cockfoot (Dactylis glomerata L.) pollen was purchased from the Allergon Company (batch no. 010994101; Angelholm, Germany); it was also stored at −80 °C until used. The pollen samples (150 mg in mass) were exposed for 48 h in an exposure chamber built in our laboratory (Fig. 1a). The chamber had an internal volume of 1 L and included a sample holder (Fig. 1b) delimited by two filters (0.45 μm pore size, Millipore, St Quentin en Yvelines, France). An appropriate volume of 15NO (98.30% 15N, Euriso-top, CEA St-Aubin, France) was added to the air within the chamber with a syringe, giving final concentrations of 1, 3 or 5% (v/v). It is well known that in air NO is partly converted into other nitrogen oxides. In the following, we shall use NOx as a general name for all nitrogen oxides. An electric fan was used to homogenize the gases within the exposure chamber. After exposure, the pollen was harvested and stored at − 80 °C in cryogenic vials until used. The chamber was aired for half a day after use.

image

Figure 1. 5N-labelled NO; the whole device is made of polymethylmethacrylate. A small fan, as commonly used in electronic devices, is introduced into the chamber for gentle homogenization of the chamber atmosphere. (b) Scheme of the sample holder; the pollen is sandwiched between two Millipore filters in order to allow gas exchange while preventing pollen dissemination within the chamber.

Download figure to PowerPoint

Preparation of pollen samples and magnesium standards

Pollen fixation and dehydration were carried out at room temperature using a ‘vapour phase’ technique recently developed in our group for studying the distribution of inorganic, mobile ions (Lhuissier et al., 1997). The technique was used here for the sake of homogeneity of sample preparation (the same specimens were used for imaging the distributions of inorganic ions) and to decrease the risk of losing the NOx molecules possibly attached to diffusible substances. Briefly, pollen was fixed at room temperature in acrolein vapour (10 days), dehydrated in dimethoxypropane vapour containing gaseous HCl (2 h), rinsed in liquid propylenoxide (1 h) and progressively embedded in London Resin White (15 days). Polymerization was carried out at 60 °C (24 h) in gelatine capsules. Then dry sections (2.5 μm thick) were cut using a diamond knife (Diatome, Biel, Switzerland) mounted on an ultramicrotome (Leica, Ultracut S, Rueil-Malmaison, France). The sections were collected using an eyelash and attached to polished stainless steel rods with a double-sided adhesive carbon-based disk. The stainless steel rods with the associated sections were stored under vacuum before use.

To investigate possible mass interference with magnesium (see the section on ‘mass interference’ in Results), standards of pure Mg-phthalocyanine were also prepared: a small drop of a saturated solution of Mg-phthalocyanine in benzene was deposited at the surface of a finely polished and cleaned stainless steel stub. After evaporation at room temperature in a dust protected vessel, the solution left a thin adherent film on the stub surface. The deposit was further degassed overnight under primary vacuum.

SIMS analyses

For the principle of SIMS and recent application to pollen and other plant samples, see for example Thellier et al. (1993), Dérue et al. (1999), Grignon et al. (1999) and Lhuissier et al. (1997, 1999). Briefly, when a finely focused ion beam (primary ions) a few keV in energy impinges on the surface of a solid, the most superficial atomic layers of the solid are sputtered out. Some is sputtered as monoatomic or polyatomic ions, which are termed the secondary ions. These ions are collected and dispersed using a high performance mass spectrometer, at the exit of which the ion to be analysed is selected. The selected secondary ions are counted using an appropriate device (electron multiplier or Faraday cup) and the counting value is stored in a computer. The beam is then rastered over the surface of the sample to produce a 256 × 256 pixel scanning digital image. The images are processed and edited using grey level or false colour scales. The method is very sensitive. Typically it allows the detection of a few parts per million of any isotope of most elements. The sample sections prepared as described above were introduced in the sample chamber of an IMS 4f (Cameca, Courbevoie, France) SIMS instrument. Caesium was used for primary ions in order to enhance the negative secondary ion emission. The accelerating voltage was set to 10 kV and the primary current was in the range 3–30 pA. The time for image acquisition at each mass under study was 300 s and the mass resolution was 6000 (which means that the instrument can discriminate between two ions whose masses, m′ and m, differ by only m/6000).

Processing of the SIMS images and isotopic percentage determination

The SIMS images were processed using the new Vizir software developed in our group. This is a 32-bit image processing program which can be used under Windows 95 or Windows NT 4.0 environment. The program can perform standard functions such as filtering or segmentation and new dedicated functions for SIMS image processing. We used such dedicated functions to sum up the count values associated with the pixels contained in regions of interest (ROI) corresponding to the precise contours of whole or parts of pollen grains. These contours (polygonal lines) are drawn by the user on the SIMS images using the computer mouse. The program can automatically draw the same polygons in the corresponding and simultaneously acquired SIMS images and then perform either pixel-to-pixel calculations in the drawn ROI or calculations using the mean of the count values of the pixels enclosed within the ROI. For instance, using the images of the distributions of 12C15N and 12C14N, the software computes a pixel-to-pixel image of the isotopic percentage {12C15N/(12C14N− 12C15N)} × 100. This allowed us to obtain the mean values of the isotopic percentages in different areas of the pollen grains for each type of labelling used (i.e. unlabelled controls and samples exposed to 1, 3 or 5% 15N-labelled NO).

Mean isotopic percentages over whole sections of pollen grain or Mg standards were also determined directly (non-imaging method) by using the isotopic-ratio function of the CAMECA software package with the following parameters: 10 blocks of 10 cycles for each sample; waiting time 1.999 s for each mass; counting time 1.006, 5.003, 1.006 and 5.003 s for 12C, 13C, 12C14N and 12C15N, respectively; for pollen samples, the secondary ions were collected from sample areas with diameters in the range 25–50 μm and the primary currents were in the range 30–150 pA; for Mg standards these values were 100–500 μm and 0.5–3 nA, respectively.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References

Mass interference

In organic matrices nitrogen is sputtered mainly in the form of CN secondary ions, hence this ionic species was used to image nitrogen distribution. Moreover, the isotopic percentage, {15N/(14N + 15N)} × 100, was determined from the ratio {12C15N/(12C14N– 12C15N)} × 100. However, mass interference may occur. For example, at mass 27 the ions 13C14N and, to a lesser extent, 12C2H3 may interfere with 12C15N. A mass resolution of 4300 is necessary to discriminate 13C14N from 12C15N. The high resolution spectra, acquired with a mass resolution of 12 000, show (Fig. 2a) that no other interference was detected at this value of mass resolution. However, we had to check for possible interference by 25MgH2 at mass 27, because the mass resolution required to separate it from 12C15N would be equal to 20 000, a resolution which cannot be attained with an IMS 4f SIMS instrument without a drastic loss of transmittance of the spectrometer. The reason for checking for a possible interference by 25MgH2 is that cytoplasmic structures containing high Mg concentrations are known to exist in pollen grains (Lhuissier et al., 1997). Using the isotopic-ratio function of the CAMECA software package, we measured a 15N percentage value of 0.357 ± 0.009 and (in order to check the reliability of the measurements) a 13C percentage value of 1.03 ± 0.06 (mean ± standard error of 40 measurements on total, made on four different areas) on the Mg-phthalocyanine standards. Neither of the isotopic percentages thus measured are significantly different from the expected natural values. This proves that there is no appreciable interference of 12C15N by 25MgH2. At mass 26, the high resolution spectrum (Fig. 2b) shows that the secondary current of the ion 12C14N is not significantly altered by the interfering ions 13C2 and 12C2H2. The mass resolution of 6000 used in our experiments was therefore appropriate for acquiring the images.

Control samples

Two different types of control pollen sample were used. Pollen samples which have been subjected to no treatment at all were used as a first sort of control (termed the ‘non-treated controls’). The pollen for the second type of control (termed the ‘air-treated controls’) was prepared as follows: after using the exposure chamber for exposure of experimental pollen samples to NO contamination, the chamber was aired as usual (see Materials and Methods), then the pollen used as control was subjected to a 48-h exposure in normal air. The mean values of the isotopic percentages, {13C/(12C− 13C)} × 100 and {12C15N/(12C14N− +12C15N)} × 100, measured with the non-treated controls were equal to 1.04 ± 0.02 and 0.36 ± 0.01, respectively (120 measurements in total, made on 10 different pollen areas), i.e. they were not significantly different from the natural values. By contrast, with the air-treated controls, the isotopic percentages of 13C and 15N were equal to 1.07 ± 0.01 and 0.62 ± 0.02 (70 measurements in total, made on seven different pollen areas), i.e. the 15N percentage remained low but significantly larger than the natural value, while the isotopic percentage of 13C was not significantly different from the natural value. Hence, it may be inferred that (i) some 15NO was absorbed on the chamber walls during the 15NO-treatment of the experimental samples and that (ii) this absorbed 15NO was slowly desorbed when the chamber was used to prepare the air-treated controls, with the consequence that these controls were slightly contaminated with 15N. This slight contamination was, however, insufficient to alter significantly the validity of the experimental results described below.

Sites of 15N fixation in the pollen grains

The 15N isotopic percentages computed from the SIMS images obtained with the cockfoot pollen samples after exposure to the different NO concentrations are shown in Fig. 3(a). Using the Vizir software, three different areas were studied on the pollen grain: the sporoderm, the cytoplasm and the whole grain. The results obtained with 1% 15NO-treated cockfoot pollen samples were very close to one another (isotopic percentages close to 5.3) in the different parts of the pollen grain. In 3% 15NO-treated cockfoot pollen samples, the isotopic percentage was approximately 1.5-fold higher in the sporoderm (12 ± 1) than in the cytoplasm (8 ± 1). A similar result was obtained with 5% 15NO-treated cockfoot pollen samples, in which the sporoderm was the major site of accumulation of 15NO (isotopic percentage = 14.5 ± 0.9 instead of 8.5 ± 0.4 in the cytoplasm).

image

Figure 3. . Computed isotopic percentages in three different areas of (a) cockfoot and (b) birch pollen grains as a function of the dose of 15NO used for pollen exposure. ▴: sporoderm, ▪: cytoplasm, ●: whole pollen grain.

Download figure to PowerPoint

The isotopic percentages computed from the SIMS images using birch pollen are given in Fig. 3(b). Treatment of pollen by 15NO caused a significant increase of this signal in sporoderm and in cytoplasm inclusions. The values of the isotopic percentages obtained after treatment by 1% and 3% 15NO were very similar to those obtained with cockfoot pollen, whereas they were twice as high after treatment with 5% 15NO (Fig. 3b).

Typical SIMS images of a 5% 15NO-treated cockfoot pollen grain are shown in Fig. 4 and corresponding processed images in Fig. 5. The distribution of the 12C14N signal (Fig. 4a) appears to be almost identical to that of 32S (Fig. 4c). This is clearly shown in Fig. 5(c), where the areas of low 12C14N intensity (mask computed with Fig. 4a and shown in Fig. 5a) match perfectly those of low 32S intensity which are visible in Fig. 4(c). In these low-intensity areas, the signal for 12C14N (Fig. 4a), 216 ± 6 cps, is about half as high as in the rest of the cytoplasm, 440 ± 32 cps, and the 32S signal (Fig. 4c), 5.91 ± 0.15 cps, is approximately 2.6-fold lower than it is in the surrounding cytoplasm, 16 ± 1 cps. Similarly, it is seen in Figs 4(b) and (d) and 5(b) and (d) that the high intensity areas of 12C15N and 16O231P match each other perfectly. By contrast, although the areas of low intensity of 12C14N (Figs 4a and 5a) and those of high intensity of 12C15N (Figs 4b and 5b) match reasonably well (Fig. 5e), a small shift is sometimes observed between them. Figure 5(f), which shows the pixel-by-pixel computed 15N isotopic percentage in the pollen grain, exemplifies the discrete distribution of intracytoplasmic sites of 15N fixation and the corresponding sporoderm 15N enrichment. The distributions of 12C14N, 12C15N, 32S and 16O231P observed in SIMS images of different samples of cockfoot pollen grains (not shown) were quite similar to those which we have just described, that is, the results were reasonably reproducible.

image

Figure 4. .9 pA. Field 35 × 35 μm.

Download figure to PowerPoint

image

Figure 5. corresponds to 100% (isotopic percentage).

Download figure to PowerPoint

The images obtained with birch pollen grains (not shown) were quite similar to those with the cockfoot pollen grains. The range of secondary current intensity differed from one image to another, but the isotopic percentages calculated from these data were highly reproducible (see the relatively small standard errors in Figs 3a and b).

Discussion and conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References

The NOx fixation occurring in the sporoderm of the pollen grains is likely to be a consequence of the nitrating activity of NO (Halfpenny & Robinson, 1952; Henry, 1994; Van der Vliet et al., 1994; Niester-Nyveld et al., 1997), which acts for example on the aromatic cycles of the phenolic compounds that exist in the sporopollenin (Niester-Nyveld et al., 1997) present in the sporoderm. Moreover, peroxidation reactions by NOx (Halliwell et al., 1992; Geletii & Balovoine, 1994) may occur on the long unsaturated aliphatic chains of the carotenoids (Guilford et al., 1988; Meuter-Gerhards et al., 1995), carotenoid esters (Niester-Nyveld et al., 1997) and other compounds (Wiermann & Vieth, 1983) that are present in the sporoderm.

It is more difficult to propose the likely molecular targets for the fixation of 15NO in the pollen cytoplasm. However, it is known that the cytoplasm of the pollen of both Cucurbita maxima and Cucurbita andreana contains electron-dense particles that resemble the globoid crystals which exist in protein bodies present in seeds (Skilnyk & Lott, 1992). Seed globoid crystals have been shown to contain large amounts of phytates, hence large amounts of phosphorus (Lott et al., 1984; Jauneau et al., 1994a, b). Because, in the SIMS images of our present 15NO-treated pollen samples, the phosphorus spots match the 12C15N spots perfectly, it cannot be excluded that the 15N-labelling of the pollen cytoplasm occurs in structures somewhat comparable to seed globoid crystals. In addition, the fact that the sites of preferential fixation of NO do not perfectly superimpose the areas of low 12C14N intensity might be a consequence of the heterogeneous structure of the protein bodies. The low intensity of these areas might be due to matrix effects arising from a reduced sputtering rate; this might originate from strong chemical heterogeneity (e.g. tight packing of proteins) causing either differential sputtering rate or non-perfect surface flatness. An alternative explanation may be that the low intensity areas correspond to amyloplasts (Lhuissier, 1998) rather than protein bodies.

The concentrations of nitrogen oxides in the atmosphere of polluted environments are three or four orders of magnitude lower than those used in our present experiments. Therefore, the cell targets which we have revealed here have certainly fixed NO until saturation, a state which may be not directly relevant to the effects of atmospheric NOx on pollen under natural conditions. However, as our present approach has shown that the slight desorption of 15NO from the walls of the exposure chamber was detectable under our experimental conditions, this means that the SIMS method may be applied to the study of the fixation of considerably lower doses of NO than those which have been investigated here to accurately localize the cell targets of this molecule.

It may also be argued that the site(s) of accumulation of the 15N label are not the real target(s) of NOx but are the site(s) of accumulation of some secondary product. Even if that possibility is rather unlikely, it can be checked in future work by using doubly labelled nitrogen oxides, 15N17O or 15N18O, for pollen treatment and then determining whether or not the two labels remain closely associated with each other in the SIMS images. This is yet another advantage of the SIMS technique, which has the potential to easily image two different isotopes simultaneously in a single specimen.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References

F.L. was supported by a doctoral fellowship from ADEME and Région de Haute-Normandie. Our SIMS instrument was purchased with grants from Ministère de la Recherche, Région de Haute-Normandie and CAMECA Company. The authors wish to thank Dr Marie-Claire Verdus for her help in preparing the Mg-phthalocyanine samples and Professor Vic Norris and Dr Ross Jeffree for discussion and help with the final preparation of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion and conclusion
  7. Acknowledgements
  8. References
  • 1
    Burns, M.S. & File, D.M. (1986) Quantitative microlocalization of diffusible ions in normal and galactose cataractous rat lens by secondary ion mass spectrometry. J. Microsc. 144, 157 182.
  • 2
    Charpin, D. (1996) Pollution atmosphérique et atopie. Revue Française d'Allergologie, 3, 327 335.
  • 3
    Dérue, C., Gibouin, D., Lefebvre, F., RasSeries, B., Robin, A., Le Sceller, L., Verdus, M.C., Demarty, M., Thellier, M., Ripoll, C. (1999) A new cold stage for SIMS analysis and imaging of frozen-hydrated biological samples. J. Trace Microprobe Techn. 17, 451 460.
  • 4
    Geletii, Y.V. & Balavoine, G. (1994) Chimie et aspects biologiques du peroxynitrite. Cahier IMABIO, Chimie et Biochimie du Monoxyde d'Azote (ed. by P. Auger and J. P. Frénoy), pp. 27 36. CNRS Publications, Paris.
  • 5
    Grignon, N., Jeusset, J., Lebeau, E., Moro, C., Gojon, A., Fragu, P. (1999) SIMS localization of nitrogen in the leaf of soybean: basis of quantitative procedures by localized measurements of isotopic ratios. J. Trace Microprobe Techn. 17, 477 490.
  • 6
    Guilford, W.J., Schneider, D.M., Labovitz, J., Opella, S.J. (1988) High resolution solid state 13C NMR spectroscopy of sporopollenins from different plant taxa. Plant Physiol. 86, 134 136.
  • 7
    Halfpenny, E. & Robinson, P.L. (1952) The nitration and hydroxylation of aromatic compounds by pernitrous acid. J. Chem. Soc. Part I, 939 946.
  • 8
    Halliwell, B., Hu, M.L., Louie, S., Duvall, T.R., Tarkington, K., Motchnik, P., Cross, C.E. (1992) Interaction of nitrogen dioxide with human plasma. Antioxidant depletion and oxidative damage. FEBS Lett. 313, 62 66.
  • 9
    Henry, Y. (1994) Cibles moléculaires du monoxyde d'azote. Détection par résonnance paramagnétique électronique Cahier IMABIO, Chimie et Biochimie du Monoxyde d'Azote (ed. by P. Auger and J. P. Frénoy), pp. 55 61. CNRS Publications, Paris.
  • 10
    Jauneau, A., Pariot, C., Verdus, M.C., Ripoll, C. (1994a) SIMS imaging of alkaline and alkaline-earth cations in flax seeds. J. Trace Microprobe Techn. 12, 51 60.
  • 11
    Jauneau, A., Ripoll, C., Verdus, M.C., Lefebvre, F., Demarty, M., Thellier, M. (1994b) Imaging the K, Mg, Na and Ca distributions in flax seeds using SIMS microscopy. Bot. Acta, 107, 81 89.
  • 12
    Lhuissier, F. (1998) Les pollens biomarqueurs de la pollution atmosphérique? Approche microscopique et biochimique. Thesis, University of Rouen, France.
  • 13
    Lhuissier, F., Verdus, M.C., Labulle, B., Lefebvre, F., Bocquel, C., Ripoll, C. (1997) SIMS investigation of the distribution of K+, Na+, Mg2+ and Ca2+ cations in birch pollen. Bot. Acta, 110, 378 387.
  • 14
    Lhuissier, F., Lefebvre, F., Gibouin, D., Sénéchal, H., Mayer, C., Desvaux, F.X., Thellier, M., Peltre, G., Ripoll, C. (1999) Structural and biochemical study of the effects of carbon monoxide on tree and grass pollen: labelling CO with 18O. J. Trace Microprobe Techn. 17, 461 476.
  • 15
    Lott, J.N.A., Goodchild, D.J., Craig, S. (1984) Studies of mineral reserves in pea (Pisum sativum) cotyledons using low-water content procedures. Aust. J. Plant Physiol. 11, 459 469.
  • 16
    Meuter-Gerhards, A., Schwerdtfeger, C., Steuernagel, S., Wilmesmeier, S., Wiermann, R. (1995) Studies on sporopollenin structure during pollen development. Z. Naturforsch. 50c, 487 492.
  • 17
    Niester-Nyveld, C., Haubrich, A., Kampendonk, H., Gubatz, S., Tenberge, K.B., Rittscher, M., Wilmesmeier, S., Wiermann, R. (1997) Immunocytochemical localization of phenolic compounds in pollen walls using antibodies against p-coumaric acid coupled to bovine serum albumin. Protoplasma, 197, 148 159.
  • 18
    Ramseyer, G.O. & Morrison, G.H. (1983) Relative sensitivity factors of elements in quantitative secondary ion mass spectrometric analysis of biological reference elements. Anal. Chem. 55, 1963 1970.
  • 19
    Skilnyk, H.R. & Lott, J.N.A. (1992) Mineral analysis of storage reserves of Cucurbita maxima and Cucurbita andreana pollen. Can. J. Bot. 70, 491 495.
  • 20
    Stewart, G.A. & Thompson, P.J. (1996) The biochemistry of common aeroallergens. Clin. Exp. Allergy, 26, 1020 1044.
  • 21
    Thellier, M., Ripoll, C., Quintana, C., Sommer, F., Chevallier, P., Dainty, J. (1993) Physical methods to locate metal atoms in biological systems. Meth. Enzymol. 227, 535 586.
  • 22
    Van der Vliet, A., O'Neill, C.A., Halliwell, B., Cross, C.E., Kaur, H. (1994) Aromatic hydroxylation and nitration of phenylalanine and tyrosine by peroxynitrite. Evidence for hydroxyl radical production. FEBS Lett. 339, 89 92.
  • 23
    Wiermann, R. & Vieth, K. (1983) Outer pollen wall, an important accumulation site for flavonoids. Protoplasma, 118, 230 233.