OUP user menu

A role for carbon catabolite repression in the metabolism of phosphonoacetate by Agromyces fucosus Vs2

Sheryl Naomi O'Loughlin , Robert L.J. Graham , Geoff McMullan , Nigel G. Ternan
DOI: http://dx.doi.org/10.1111/j.1574-6968.2006.00344.x 133-140 First published online: 1 August 2006


A strain of Agromyces fucosus, designated Vs2, metabolized a range of organophosphonate compounds as sole phosphorus sources for growth and metabolized phosphonoacetate as a sole carbon, energy and phosphorus source for growth. With phosphonoacetate as the sole phosphorus source and a pyruvate carbon source, transient phosphate release to the medium was observed, in contrast to cultures grown with glucose and phosphonoacetate, where no phosphate release to the medium was observed. Carbon catabolite repression, specifically by means of inducer exclusion of phosphonoacetate, was proposed as the mechanism responsible, and phosphonoacetate hydrolase enzyme assays carried out on cell extracts confirmed that induced phosphonoacetate hydrolase activities were indeed higher in cells grown on pyruvate with phosphonoacetate as sole phosphorus source. This phenomenon has not previously been demonstrated in vivo, and must represent a significant metabolic control of organophosphonate metabolism. The catabolite repression phenomenon was also evident when A. fucosus grew on 2-aminoethylphosphonate as sole phosphorus source, allowing demonstration of a third mode of control for biodegradation of this compound. Excision of stained zymogram gel pieces, followed by tryptic digestion and mass spectrometric analysis, allowed the identification of phosphonoacetate hydrolase-derived peptides.

  • phosphonoacetate
  • 2-aminoethylphosphonate
  • catabolite repression
  • organophosphonate
  • regulation
  • biodegradation


Organophosphonate compounds, characterized by the presence of a stable, covalent, carbon to phosphorus (C–P) bond have found widespread commercial applications (Ternan, 1998a). As a result, the amount of these xenobiotic chemicals entering the environment every year is estimated to be in excess of 20000 tonnes per year (Jaworska, 2002). The vast majority of organophosphonate compounds are utilized, via the C–P lyase pathway, only as sole P sources for growth by microorganisms. This is because phosphate-limited conditions are required for induction of PHO regulon controlled genes encoding the organophosphonate uptake and C–P bond cleavage proteins that comprise the C–P lyase enzyme complex. However, a number of microorganisms have been isolated over the years that are capable of mineralizing certain, generally naturally occurring, organophosphonates to CO2 and H2O, thus allowing them to serve as sources of carbon and energy, or nitrogen, as well as phosphorus for growth (Ternan, 1998a,b; Klimek, 2001; Obojska, 2002; Obojska & Lejczak, 2003).

The complete mineralization of these compounds cannot be dependent on a requirement for phosphate-limited conditions as a C:P ratio in excess of 200:1 (Cook, 1988) is required for effective phosphorus limitation. When an organophosphonate serves as the sole source of carbon and phosphorus for growth, cleavage of the C–P bond of the organophosphonate molecules will result in non-phosphate-limiting conditions that in theory repress any further organophosphonate transport. Therefore, while the requirement for phosphate limitation before uptake and biodegradation of the majority of organophosphonates is an accepted consensus, little attention has been given thus far to the effect of carbon source availability on C–P bond cleavage activity within microorganisms metabolizing organophosphonates as sole phosphorus sources.

Carbon catabolite repression (CCR) in bacteria is generally regarded as a regulatory mechanism to ensure sequential utilization of carbohydrates (Bruckner & Titgemeyer, 2002) and is a widespread phenomenon in bacteria. The presence of a preferred carbon source such as glucose can repress genes and operons that may be involved in the metabolism of alternative carbon sources. Bacteria have developed global control mechanisms allowing them to use carbon sources in a strictly controlled hierarchical manner via CCR and the genetic and physiological control mechanisms are well characterized in both Gram-negative bacteria and low-GC Gram-positive bacteria (Titgemeyer & Hillen, 2002). In all bacteria studied so far, the phosphoenolpyruvate:sugar phosphotransferase system (PTS), which transports and phosphorylates a variety of sugars, is involved in CCR (Stülke, 1998). These mechanisms differ at the molecular level, with enteric bacteria transmitting CCR signals via the PTS protein IIAGlc, which triggers activity of the catabolite activator protein (CAP), required for activation of numerous catabolite controlled genes. In the presence of glucose unphosphorylated EIIAGlc binds to several sugar permeases and to glycerol kinase, thereby inhibiting them via a mechanism known as inducer exclusion (Saier, 1989; Stülke, 1998; Titgemeyer & Hillen, 2002). In low-GC Gram-positive bacteria, however, use of the PTS protein HPr exerts inducer exclusion and triggers the global regulator carbon catabolite protein A (CcpA), conferring CCR by global gene activation/repression (Titgemeyer & Hillen, 2002). In Gram-negative enteric bacteria, an external sugar is sensed by the sugar-recognition constituent of an Enzyme II complex of the PTS (IIC), and a dephosphorylating signal is transmitted via the Enzyme IIB/HPr proteins to the central regulatory protein, IIAGlc. Targets regulated include (1) permeases specific for lactose, maltose, melibiose and raffinose, (2) catabolic enzymes such as glycerol kinase that generate cytoplasmic inducers, and (3) the cAMP biosynthetic enzyme, adenylate cyclase that mediates catabolite repression (Saier, 1996). In low-GC Gram-positive bacteria, cytoplasmic phosphorylated sugar metabolites are sensed by the HPr kinase. HPr becomes phosphorylated on Ser-46, and this phosphorylated derivative regulates the activities of its target proteins. These targets include (1) the PTS, (2) non-PTS permeases (both of which are inhibited) and (3) a cytoplasmic sugar-P phosphatase which is activated to reduce cytoplasmic inducer levels (Saier, 1996). The general outcome is the same in both groups – preferential use of the carbon source allowing a maximal growth rate, mediated either by processes like inducer exclusion, inducer expulsion (Saier, 1996) and the control of regulator activity by phosphorylation (Stülke, 1998). While in these bacteria many molecular details are known, investigation of PTS function in high-GC Gram-positive members of the actinomycetes, such as Streptomyces coelicolor and Agromyces spp. is in its infancy (Bertram, 2004).

The genus Agromyces was established almost 40 years ago by Geldhill and Cassida for the single species Agromyces ramosus (Suzuki, 1996). This genus currently harbours at least 10 characterized species, including those exhibiting filamentous and nonfilamentous phenotypes, that may be microphilic to aerophilic and oxidase and catalase variable members of the Actinomycetales (Dorofeeva, 2003; Rivas, 2004). While the description of new Agromyces species continues apace (Jurado, 2005a,b), little is known of the metabolic capabilities of the genus, save for the utilization of certain hydrocarbons (Rivas, 2004) and their reported ability to lyse the cells of other microorganisms (Cherniakovskaia et alet al,, 2005).

In this paper, we describe for the first time a role for the involvement of catabolite repression by glucose in the metabolism of phosphonoacetate, within a strain of Agromyces fucosus, designated Vs2 which was recently isolated by Panas (2005) for its ability to mineralize phosphonoacetate.

Materials and methods


Chemicals, all of the highest purity available, were obtained from Sigma-Aldrich Chemical Co. (Poole, UK). Organophosphonate solutions were treated as described previously to remove contaminating inorganic phosphate (Ternan, 1998b). All glassware was stripped of contaminating phosphorus by soaking overnight in 2% Decon (Decon Laboratories, Hove, UK).


Complete M1 medium contained (L−1): KCl, 0.2g; MgSO4·7H2O, 0.2g; CaCl2·2H2O, 0.01g; NH4Cl, 5.0g; carbon source, 180g atoms CL−1; ferric ammonium citrate, 1.0mg; phosphate buffer, 1mM, and 1mL each of trace element solution and vitamin solution (Ternan, 1998b). Carbon-, nitrogen- and phosphorus-free M1 mineral salts medium was prepared by omitting the carbon sources and inorganic nitrogen and phosphorus sources from M1 medium.

Cell growth and culture conditions, cell harvest and cell extract preparation

Agromyces fucosus Vs2 was routinely maintained on nutrient agar plates at 30°C. Growth in batch cultures (50mL in 250mL Erlenmeyer flasks) was followed as attenuance at 650nm. Where phosphate-starved cells were required for experiments, cells from nutrient agar plates were initially inoculated into P-free medium and following 48h growth, harvested and washed aseptically in sterile saline. They were then transferred to P-free M1 medium, from whence they were used as inoculum. Cells were harvested in mid-late log phase by centrifugation at 9000g at 3–5°C for 10min and cell extracts prepared by the bead mill method of Graham (2006) before storage at −70°C.

Enzyme assays

Phosphonoacetate hydrolase activity was assayed by the method of McMullan & Quinn (1992) and activity expressed as nanomoles of inorganic phosphate (Fiske & Subbarow, 1925) released min−1 mg−1 of cell-extract protein. Bacterial alkaline phosphatase activity in cell extracts was assayed by the method of Ternan & Quinn (1998) and activity expressed as nanomoles of p-nitrophenol released min−1 mg−1 of cell-extract protein. Enzyme assays were carried out in at least duplicate on cell extracts prepared from two separate cultures.

Biochemical determinations

Glucose in culture supernatant samples was determined using a glucose assay kit (Sigma) and glucose concentrations were expressed as mg mL−1 glucose, read off a standard calibration graph. Pyruvate in culture supernatants was assayed by the method of Anthon & Barrett (2003) and pyruvate concentrations were expressed as mg mL−1 pyruvate, determined from a standard calibration graph. All standards and samples were assayed in duplicate.

In-gel digestion of proteins and liquid chromatography-mass spectrometry of peptide mixtures

Proteins in excised native polyacrylamide gel electrophoresis (PAGE) zymogram gel pieces were subject to trypsinization by the method of Chong & Wright (2005) before mass spectrometric analysis as described previously (Graham, 2006).

Results and discussion

Agromyces fucosus Vs2 was initially grown on a range of organophosphonates (final concentration, 1.0mM) as sole P source with either glucose or pyruvate as a carbon source. Pi release into the culture medium during growth on both phosphonoacetate and 2-aminoethylphosphonate (2AEP), when cultured with pyruvate as the C source was observed; no Pi was released under any other growth condition (Table 1). Release of Pi during growth on phosphonoacetate/pyruvate was transient (Fig. 2), occurring at the early log phase of growth, similar to that previously described by Ternan & Quinn (1998). Growth yields were similar for both phosphonoacetate and Pi at concentrations of 0.1, 0.25, 0.5, 0.75 and 1.0mM, where biomass protein yields increased proportionately in relation to the amount of phosphorus source supplied. In all cases, over 90% of the carbon substrate had been removed from the cultures by the end of logarithmic growth and biomass production (results not shown). In A. fucosus Vs2, the supplied carbon source therefore influences the regulation not only of phosphonoacetate but also 2AEP metabolism, allowing luxuriant substrate uptake and C–P bond cleavage under P-limited conditions.

View this table:

Range of organophosphonate substrates utilized by Agromyces fucosus Vs2 as sole phosphorus source (final concentration 1.0mM) with either glucose (final concentration 1%, 55mM) or sodium pyruvate (final concentration 1%, 110mM) as carbon source

SubstrateProtein (μgmL−1)Pi release (mM)Protein (μgmL−1)Pi release (mM)
Positive control (1mM Pi)90ND380ND
Negative control100.0300.0
Methyl phosphonate110.0200.0
Ethyl phosphonate350.0300.0
Phenyl phosphonate800.0200.0
2-aminoethyl phosphonate690.01900.15
3-aminopropyl phosphonate150.0400.0
4-aminobutyl phosphonate150.0600.0
2-amino-3-phosphono propionate350.0300.0
2-amino-4-phosphono butyrate50.0100.0
  • * Growth is expressed as μg protein produced per mL of culture; the values presented are the highest values obtained from cultures entering the stationary phase of growth. Results are the mean of duplicate culture determinations that on no occasion varied by more than 5%.

  • Limit of detection for Pi was less than 5.0μM–0.0=below limit of detection.

  • ND, not determined.


Growth of Agromyces fucosus Vs2 on phosphonoacetate (1mM) as sole phosphorus source, with 1% pyruvate as sole carbon source. •, culture attenuance (D650nm); □, pyruvate in culture supernatant (%); ○, phosphate in culture supernatant (mM).


Characterized biochemical pathways for the biodegradation of phosphonoacetate and 2AEP by microorganisms.

Agromyces fucosus Vs2 was subsequently grown on a range of carbon sources with either 1mM phosphonoacetate or 1mM Pi as the P source (Table 2). Agromyces fucosus Vs2 could utilize glucose, pyruvate, fructose, galactose, sucrose, glycerol, arabinose, trehalose and lactose as sources of carbon and energy, but could not utilize any of succinate, acetate, citrate, tartrate, or l-sorbose when exogenously supplied. Pi release during growth with phosphonoacetate was observed for a number of carbon sources. It appears therefore that while the presence of free exogenous glucose molecules in the medium induces a regulatory effect, a glucose molecule as a constituent of a disaccharide sugar does not lead to catabolite repression. Where A. fucosus Vs2 was incapable of utilizing a supplied carbon source, Pi release to the medium, commensurate with the metabolism of the carbon contained within the supplied phosphonoacetate (1mM) was observed. With galactose, sucrose and glycerol carbon sources, however, lack of Pi release to culture supernatants suggested that under these growth conditions, C–P bond cleavage was more tightly regulated – as in the case of glucose – such that no Pi release was observed. As with pyruvate-grown cells, cultures grown on trehalose, lactose or fructose demonstrated transient release of Pi during growth on phosphonoacetate as sole P source, making it evident that uptake of both sugar-carbon source and phosphonoacetate P-source, with concomitant C–P bond cleavage, was occurring.

View this table:

Range of carbon substrates utilized by Agromyces fucosus Vs2 with either phosphonoacetate (final concentration, 1.0mM) or inorganic phosphate (final concentration 1.0mM) as sole phosphorus source

Phosphorus source (1mM)
Carbon sourcePhosphono
protein yield
(μgmL−1)Pi release
acetate) (mM)Inorganic
yield (μgmL−1)
Carbon-free control180.9819
  • * Growth is expressed as μg protein produced per mL of culture; the values presented are the highest values obtained from cultures entering the stationary phase of growth. Results are the means of duplicate culture determinations that on no occasion varied by more than 5%.

  • All carbon sources were supplied at concentrations such that the total concentration of C atoms in each culture was the same (180g atoms CL−1, equivalent to 1% pyruvate).

  • Limit of detection for Pi was less than 5.0μM–0.0=below limit of detection.

  • § This culture was effectively carbon and phosphorus-limited: the 1mM phosphonoacetate added was then metabolized as a carbon source with the excess phosphate being released as Pi. Cultures containing up to 10mM phosphonoacetate as sole C and P source yielded proportionally increasing protein yields, with phosphate release that was effectively equimolar with the concentration of phosphonoacetate supplied (data not shown).

Initial screening on organophosphonates as sole P source showed that 2AEP served as a source of P for growth of A. fucosus Vs2. Release of Pi from 2AEP to the culture supernatant when A. fucosus Vs2 was grown with pyruvate carbon source suggested that the same catabolite repression phenomenon as that observed for phosphonoacetate was in effect. To test this hypothesis, A. fucosus was grown on 2AEP supplied as sole P (1mM), sole N (5mM), sole N and P (5mM), or as sole C, N, and P (10mM) source, with either glucose or pyruvate as carbon source. While Pi release was observed only with pyruvate carbon source when 2AEP served as sole P source, an unexpected release of Pi was observed when 2AEP served as either sole N or sole N and P source for growth, with either glucose or pyruvate as the carbon source (results not shown). It was of note that 2AEP was unable to serve as a C and energy source for growth of A. fucosus Vs2 and under this condition, no phosphate release was observed. This suggests that while catabolite repression control is exerted over metabolism of 2AEP as a sole P source, this cannot be the case when 2AEP serves as either sole N source of as sole N & P source. It is evident that luxuriant uptake of 2AEP, and the necessary C–P bond cleavage to result in excess Pi release over and above that required for growth, is permissible under certain culture conditions. However, A. fucosus Vs2 does not appear to be capable of utilizing the carbon skeleton of 2AEP and was incapable of utilizing exogenously supplied acetaldehyde (10mM), the breakdown product of 2AEP, as a sole source of C and energy for growth. Acetaldehyde did not, however, affect either growth rate or yield in full medium, suggesting that it was not toxic to the cells at this concentration.

We wished to test the hypothesis that inducer exclusion of phosphonoacetate in the presence of glucose was responsible for the observed lack of phosphate release. Under such a regime, phosphonoacetate would not be transported into the cells in amounts in excess of those necessary to supply P for growth. As a result, we proposed that the level of phosphonoacetate hydrolase activity in cell extracts would be lower in glucose-grown cells than those grown on pyruvate. Indeed, we have previously shown that increasing medium phosphonoacetate levels resulted in increased levels of phosphonoacetate hydrolase enzyme activity in cell extracts of A. fucosus Vs2.

Agromyces fucosus Vs2 was grown on phosphonoacetate as sole P source with either glucose or pyruvate, on Pi with either glucose or pyruvate, and on phosphonoacetate (10mM) as sole C, energy and P source. Cell extracts were assayed for alkaline phosphatase activity and for phosphonoacetate hydrolase activity (Table 3). Glucose/phosphonoacetate-grown cells contained lower phosphonoacetate hydrolase activity levels than pyruvate/phosphonoacetate-grown cells. Both extracts had similar levels of alkaline phosphatase activity. These trends were also observed when 0.1% (w/v) carbon source was supplied (results not shown), suggesting that the presence of glucose, with its effect on global substrate transport, rather than the carbon to phosphorus (as Pi) ratio of the medium was responsible. Of note was the observation that glucose-grown cells were P-limited, as evidenced by the high levels of alkaline phosphatase activity in cell extracts, which were similar to those in extracts prepared from P-starved cells (results not shown). The presence of glucose appears to exert strong negative control on global substrate uptake, reducing transport of both phosphonoacetate and inorganic phosphate in this microorganism.

View this table:

Phosphonoacetate hydrolase and alkaline phosphatase activities induced in cell extracts of Agromyces fucosus Vs2 grown under a range of nutrient limitations

Culture conditionPhosphonoacetate
hydrolase activityAlkaline
Phosphonoacetate as sole P, glucose1.1 ± 0.125.2 ± 6.4
Phosphonoacetate as sole P, pyruvate4.5 ± 0.235.1 ± 7.6
Phosphate-grown cells, glucose0.0 ± 0.0133.6 ± 3.7
Phosphate-grown cells, pyruvate0.0 ± 0.06.1 ± 2.7
Phosphonoacetate as sole C, energy & P115.0 ± 6.02.9 ± 0.6
  • * Phosphonoacetate hydrolase activity is expressed as nanomoles of inorganic phosphate produced min−1 mg−1 of cell-extract protein. Assays were carried out twice in duplicate and means presented, ± standard deviation of the mean.

  • Alkaline phosphatase activity is expressed as nanomoles of p-nitrophenol produced min−1 mg−1 of cell-extract protein. Assays were carried out twice in duplicate and means presented, ± standard deviation of the mean.

Cell extracts from pyruvate/Pi-grown cells contained only low levels of alkaline phosphatase activity, showing that they were not phosphate-limited. No phosphonoacetate hydrolase or alkaline phosphatase activity was detectable in cell extracts prepared from nutrient broth-grown cells. Cell extracts prepared from cells grown on phosphonoacetate as sole carbon, energy and phosphorus source contained much higher levels of phosphonoacetate hydrolase activity and had only low levels of alkaline phosphatase activity. Only when phosphonoacetate was present in the culture medium was any phosphonoacetate hydrolase activity detectable in cell extracts, an observation consistent with the fact that to date, expression of the phosphonoacetate hydrolase gene in microorganisms has only been induced by the presence of phosphonoacetate (McMullan & Quinn, 1994).

Cell extracts were electrophoresed and zymogram staining carried out as described previously (Ternan, 1999), using phosphonoacetate as a substrate. Phosphonoacetate hydrolase activity was identified as a discrete blue band on a clear background and the A. fucosus Vs2 phosphonoacetate hydrolase moved further on the gel than the phosphonoacetate hydrolase of Pseudomonas fluorescens 23F, which was included as a reference. The blue-stained portions of the zymogram gel were excised (0.5cm × 0.2cm) and subject to in-gel digestion of proteins before mass spectrometric analysis. Analysis of MASCOT data produced from tryptic peptides of the P. fluorescens 23F zymogram allowed the identification of peptides matching the published sequence of that phosphonoacetate hydrolase. This proved that despite the relatively harsh treatment during development of the zymogram, intact tryptic peptides could be recovered from gels. No peptides matching phosphonoacetate hydrolase were detected from of the A. fucosus Vs2 zymogram stain, reinforcing the hypothesis that the A. fucosus Vs2 phosphonoacetate hydrolase must be significantly different to the canonical P. fluorescens 23F enzyme. Furthermore, we were unable to generate PCR amplicons from A. fucosus Vs2 genomic DNA despite using nine different primers combinations designed from the 23F nucleotide sequence (Panas, 2006) that did allow amplification of phosphonoacetate hydrolase nucleotide sequences with a number of Pseudomonas isolates described by Panas (2005) and Panas (2006). The phosphonoacetate hydrolase enzymes within these organisms, from which tryptic peptides could be recovered and identified from zymograms, all migrated the same distance on a native PAGE gel as the P. fluorescens 23F enzyme (N.G. Ternan, unpublished results), showing that their biochemical properties, in addition to their coding nucleotide sequences (Panas, 2006) were similar. The phosphonoacetate hydrolase of A. fucosus Vs2 must be significantly different to the well-characterized P. fluorescens 23F enzyme, both at the nucleotide and amino acid sequence levels.

This paper brings to light a number of novel findings in the area of organophosphonate metabolism by environmental microorganisms. Firstly we have demonstrated the significant and hitherto unrecognized role played by CCR in both the metabolism of phosphonoacetate and 2AEP, and in the induction of both phosphonoacetate hydrolase and alkaline phosphatase enzymes in A. fucosus Vs2. The phosphonoacetate hydrolase enzyme activity levels measured by us in crude cell extracts appear to support our hypothesis that inducer exclusion may be the means by which catabolite repression control is exerted in A. fucosus Vs2. This is the first time that the CCR phenomenon has been investigated with regard to organophosphonate metabolism and it is noteworthy that the same control appears to exist in this organism for both the natural organophosphonate, 2AEP, as well as for phosphonoacetate – a compound that we have recently suggested may be a biogenic molecule (Panas, 2006). However, we believe that the metabolism of 2AEP as either P and/or N source concomitant with luxuriant Pi release, while it is not mineralized as a C source, represents a third mode of 2AEP metabolism in addition to (i) organisms that metabolize 2AEP only as sole P source, under P-limited conditions and (ii) organisms that can mineralize the compound as a carbon and energy source and are not subject to phosphate-starvation control of the enzymes involved.

Upon reviewing the literature on organophosphonate metabolism, we discovered that in those cases where microorganisms capable of carrying out C–P bond cleavage in the presence of inorganic phosphate were isolated, the carbon source(s) supplied did not include glucose. Those microorganisms where organophosphonate metabolism was under strict PHO regulon control, however, were most often isolated with a glucose carbon source. It is intriguing that researchers eschewing glucose as a readily assimilated carbon source for microbial growth have had most success in isolating novel microorganisms capable of the phosphate-starvation independent cleavage of the C–P bond. We are led to propose therefore, that in certain cases catabolite repression has a role to play in organophosphonate metabolism and that the significance of this additional level of control has only just begun to be recognized. Indeed, the results of our study add to the conclusions drawn by the recent work of Puri-Taneja (2006) and Choi & Saier (2005) who demonstrated the importance of the ccpA in negative regulation of pho regulon controlled genes in Bacillus subtilis. They showed that the presence of ccpA (which is induced under glucose replete conditions) in fact switched off transcription of phoBR controlled PHO regulon genes. Based on direct in vitro measurements of alkaline phosphatase activity in cell extracts, we have shown that the opposite situation exists in A. fucosus with alkaline phosphatase levels being higher in glucose/Pi-grown cells. Little is known of the means by which CCR is effected in Streptomycetes, and it would appear that different control mechanisms to those of either Gram-negative or low-GC Gram-positive microorganisms must be present in A. fucosus Vs2. Future investigations will undoubtedly yield new insights into the biochemistry of C–P bond cleavage, its genetic control, and carbon metabolism in this microorganism.


N.O'L. gratefully acknowledges receipt of a Nuffield Foundation Undergraduate Research Bursary in Science (Ref URB/01988/A 29999). R.L.J.G. was supported by the Northern Ireland Centre of Excellence in Functional Genomics, which was established at the University of Ulster at Coleraine in 2003, with funding from the European Union (EU) Programme for Peace and Reconciliation, under Technology Support for the Knowledge-Based Economy. N.G.T. acknowledges research funding for this work from the Biomedical Sciences Research Institute, University of Ulster, Coleraine.


View Abstract