Plant cell walls are amazingly complex amalgams of carbohy? drates, proteins, lignin, water, and incrusting substances such as cutin, suberin, and certain inorganic compounds that vary among plant species, cell types, and even neighboring cells Developmental events and exposure to any of a number of abiotic and biotic stresses further increase this compositional and structural variation Moreover, the dynamic nature and func? tions of plant cell walls in terms of growth and development, environmental sensing and signaling, plant defense, intercel? lular communication, and selective exchange interfaces are reflected in these variations Much is currently known about the structure and metabolic regulation of the various cell wall components, but relatively little is known about their precise functions and intermolecular interactions In this review, I will discuss the accumulated structural and regulatory data and the much more limited functional and in? termolecular interaction information on five plant cell wall protein classes These five protein classes, listed in Table 1, include the extensins, the glycine-rich proteins (GRPs), the proline-rich proteins (PRPs), the solanaceous lectins, and the arabinogalactan proteins (AGPs) These five proteins may be evolutionarily related to one another, most obviously because each of them, with the exception of the GRPs, contains hydroxyproline, and less obviously in the case of the GRPs because this class has nucleotide sequence similarity to the extensins For completeness, I should mention that these are not the only cell wall proteins that are known Others exist, such as cysteine-rich thionins, 28and 70-kD water-regulated proteins, a histidine-tryptophan-rich protein, and many cell wall enzymes such as peroxidases, phosphatases, invertases, a-mannosidases, P-mannosidases, p-1,3-glucanases, (3-1,4glucanases, polygalacturonase, pectin methylesterases, malate dehydrogenase, arabinosidases, a-galactosidases, (3-galactosidases, |3-glucuronosidases, p-xylosidases, proteases, and ascorbic acid oxidase (Varner and Lin, 1989) However, the above five classes generally represent the most abundant, and to date, the most well-studied and widely documented, plant cell wall proteins Before describing these five wall protein classes, I should point out that research on these individual proteins has oc? curred in several plant species, but relatively few examples exist where these cell wall proteins have been studied together in one plant, let alone in one particular plant organ or type of cell Thus, data from one plant species are often extrapolated to represent the situation in other plant species Although such extrapolations are usually valid, enough variations are now known that caution should be exercised in making or believing such claims

Structure and function of plant cell wall proteins.

Surface infrared spectroscopy

Ultrafast vibrational dynamics of hydrogen bonds in the condensed phase.

TRANSPORT of the essential nutrient phosphorus—primarily in the form of orthophosphate—into cells and organelles is highly specific. This is exemplified by the uptake of phosphate or its close analogue arsenate by bacterial cells by way of a high affinity active transport system dependent on a phosphate-binding protein; this system is unable to recognize other inorganic oxyanions and is, moreover, distinct from the one for sulphate transport1,2. The phosphate-binding protein is a member of a family of periplasmic proteins acting as initial high-affinity receptors for the osmotic shock-sensitive active transport systems or permeases for various sugars, amino acids, oligopeptides, and oxyanions2,3. We report here the highly refined 1.7 A resolution X-ray structure of the liganded form of the phosphate-binding protein. The structure reveals the atomic features responsible for phosphate selectivity, either in monobasic or dibasic form, and the exclusion of sulphate. These features are fundamental to understanding phosphate transport systems and molecular recognition of charged substrates or ions in other biological processes.

High specificity of a phosphate transport protein determined by hydrogen bonds

Regulation of gene expression during plant embryogenesis.

We present a simple model to describe the phase relaxation of the ν s (XH) mode of a hydrogen-bonded species XH⋯Y in solution in an inert solvent. The major assumptions made are that the ν σ (XH⋯Y) stretching mode of the hydrogen bond can be represented by the Ornstein—Uhlenbeck stochastic process, and that the ν s (XH) phase coherence is lost because of the coupling between the two vibrational modes. The model can be analysed in terms of Kubo's theory of a randomly modulated oscillator. An expression is derived for the transition dipole moment autocorrelation function which characterises the line shape of the mid infrared ν s (XH) absorption band, and various limiting cases of the formula are discussed. It is further shown that vibrational relaxation may be expected to influence the far infrared ν σ (XH⋯Y) bandshape and the one-phonon neutron inelastic scattering cross section, and that the parameters required to characterise all three types of spectra are closely related. Experimental tests of the theory are reported in the following paper.

Vibrational relaxation of hydrogen-bonded species in solution. I. Theory

The major storage protein in many legume seeds is legumin, a hexameric molecule of molecular weight (Mr) 360,000–400,000 consisting of six subunit pairs, each of which comprises a 40,000-Mr, acidic polypeptide linked by disulphide bonds to a 20,000-Mr, basic polypeptide1. We have recently shown that these subunit pairs are synthesized, in vivo and in vitro in Pisumand Vicia2,3, as a 60,000-Mr precursor polypeptide, putatively containing one copy each of the acidic and basic summits covalently linked together. We describe here the cloning and identification of cDNA sequences coding for the legumin precursor. Additionally, we have used these cDNAs to verify that the legumin mRNAs are of sufficient size to code for the precursor and to demonstrate the presence of four single-copy legumin genes in genomic DNA. Possible mechanisms for the processing of the legumin precursor are discussed. Identification of cloned cDNA sequences coding for vicilin, the secondary storage protein of Pisum sativum4, is also reported.

Cloning and analysis of cDNAs encoding plant storage protein precursors

Studies of molecular motions and vibrational relaxation in acetonitrile. I. Raman spectral studies of the ν1 ν3 and 2ν3 bands in the

A 3.4-kilobase genomic DNA fragment from Pisum sativum L. containing the LegA gene, which encodes a major legumin storage protein, was transferred to Nicotiana plumbaginifolia using an Agrobacterium tumefaciens strain containing the Bin 19 binary vector system. Northern hybridisation analysis of legA-transformed plants demonstrated that legumin-specific RNA was present in developing seeds but not in developing leaves. Legumin protein was immunologically detected in the mature seeds of legA-transformed plants, and was present as the correct-size protein composed of disulphide-bonded polypeptides. It is concluded that the transferred pea genomic fragment contains all the information necessary for seed-specific expression of the legA gene, and for correct processing of the primary transcript and the precursor legumin protein.

Tissue-specific expression of a pea legumin gene in seeds of Nicotiana plumbaginifolia.

The v s (XH) absorption bands of the following complexes have been recorded in digitised form: phenol (OH)- and phenol (OD)- dioxan in CCl 4 ; phenol (OH)- and phenol (OD)-acetonitrile in CCl 4 ; and phenol (OH)-acetonitrile in CDCl 3 . All these bands are broad and nearly featureless, and have widths at half height ranging from 75 to 148 cm −1 . With one possible exception they appear to be free from Fermi resonances. By least-squares fitting of the theoretical bandshape function derived in part I to the experimental spectra we have been able to estimate the parameters Δ, ω 2 , and τ C which characterise the theoretical autocorrelation function, and also to determine the v s (XH) stretching frequency very accurately. In the case of phenol (OD)-dioxan in CCl 4 the discrepancies between the observed and fitted spectra can be entirely attributed to errors of observation. It is found that the Kubo parameter τ C Δ is close to unity in all cases, showing that the v s (XH) bands are subject to partial (but not extreme) motional narrowing. The v σ (XH⋯Y) vibration is always heavily damped and has a broad spectral density. The changes in the spectral profiles produced by deuteration and change of solvent are easily understood in terms of the theory. Deuteration affects the amplitude of modulation, while changing the solvent affects the value of τ C . In either case τ C Δ changes, and the speed of modulation is altered. Since the system is far from the gaussian (slow modulation) limit, the bandshape changes markedly in consequence.

J. Yarwood

Papers

Vibrational relaxation of hydrogen-bonded species in solution. I. Theory

Cloning and analysis of cDNAs encoding plant storage protein precursors

Studies of molecular motions and vibrational relaxation in acetonitrile. I. Raman spectral studies of the ν1 ν3 and 2ν3 bands in the

Tissue-specific expression of a pea legumin gene in seeds of Nicotiana plumbaginifolia.

Vibrational relaxation of hydrogen-bonded species in solution. Il. Analysis of vs(XH) absorption bands