CYN determination in tissues from freshwater fish by LC-MS/MS: validation and application in tissues from subchronically exposed tilapia (Oreochromis niloticus).

Harmful cyanobacterial blooms are occurring in eutrophic freshwater lakes and reservoirs throughout the world and, because of the production of toxins such as cylindrospermopsin (CYN), they can present a public safety hazard through contamination of seafood and fish for human consumption. Therefore it is important to develop methods to determine CYN at trace levels in those organisms. A new method for unconjugated CYN determination in tissues (liver and muscle) of tilapia (Oreochromis niloticus) is herein described and discussed; it is based on solvent extraction and purification with C18 and graphitized carbon cartridges, and quantification by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). The method was optimized and suitably validated, with a linear range from 0.125-12.5 µg CYN/g dry weight (dw) in the case of the liver, and 0.02-1 µg CYN/g dw for the muscle. Limits of detection and quantitation were 0.07 and 0.12 µg/g dw for the liver, and 0.002 and 0.007 µg/g dw for the muscle, respectively. Mean recoveries ranged 80-110% in liver, and 94-104% in muscle, and intermediate precision values from 6 to 11%. The method is robust against the three factors considered for purification (batch of the graphitized carbon cartridges, time for the sample to pass through the cartridge, and final dissolving water volume). Furthermore, it has been successfully applied to the extraction and quantification of CYN in tissue samples from tilapia subchronically exposed to CYN in the laboratory. This represents a sensitive, reproducible, accurate, and robust method for extraction and determination of unconjugated CYN in tissues of fish exposed to the toxin. This procedure can be used for confirmatory routine monitoring of CYN in fish samples in environmental studies.

Development and optimization of a method for the determination of Cylindrospermopsin from strains of Aphanizomenon cultures: intra-laboratory assessment of its accuracy by using validation standards.

The occurrence of cyanobacterial blooms in aquatic environments is increasing in many regions of the world due to progressive eutrophication of water bodies. Because of the production of toxins such as Cylindrospermopsin (CYN), contamination of water with cyanobacteria is a serious health problem around the world. Therefore it is necessary to develop and validate analytical methods that allow us to quantify CYN in real samples in order to alert the public of this toxin. In this work, an analytical method has been developed an optimized for the determination of CYN from Aphanizomenon ovalisporum cultures. The analytical procedure is based on solvent extraction followed by a purification step with graphitized cartridges and CYN quantification by LC-MS/MS. The extraction and purification steps were optimized using a two-level full factorial design with replications. A suitable and practical procedure for assessing the trueness and precision of the proposed method has been applied by using validation standards. The method has been suitably validated: the regression equation was calculated from standards prepared in extracts from lyophilized M. aeruginosa PCC7820 (r(2)≥0.9999) and the linear range covered is from 5 to 500 µg CYN/L, equivalent to 0.18-18.00 µg CYN/g dry weight lyophilized cells. Limits of detection and quantification were 0.04 and 0.15 µg CYN/g, respectively, the recovery range (%) oscillated between 83 and 94% and intermediate precision (RSD %) values from 5.6 to 11.0%. Moreover, the present method showed to be robust for the three factors considered: the batch of the graphitized carbon cartridges, the flow rate of the sample through the cartridge, and the final re-dissolved water volume after SPE treatment, which permits its validation. The validated method has been applied to different lyophilized cultures of A. ovalisporum (LEGE X-001) to evaluate CYN content. This procedure can be used for determining CYN in lyophilized natural blooms samples in environmental studies.

Cylindrospermopsin determination in water by LC-MS/MS: optimization and validation of the method and application to real samples.

A new method for determining dissolved cylindrospermopsin (CYN) in waters using solid-phase extraction (SPE) with graphitized carbon cartridges and quantification by liquid chromatography coupled with tandem mass spectrometry is described and discussed. The method has been suitably validated: the linear range covered is from 0.900 to 125 µg CYN/L. Limits of detection and quantification were 0.5 and 0.9 µg CYN/L, respectively, and allow CYN determination at concentrations below the guideline proposed of 1 µg CYN/L in natural waters. The method exhibits mean recoveries from 83 to 95%, and intermediate precision (relative standard deviation (%)) values from 5 to 12%, ensuring adequacy against the Association of Official Analytical Chemists guidelines. The method is robust against the following three influential factors considered in the cleanup stage: the batch of the graphitized carbon cartridges, the flow rate of the water sample through the cartridge, and the final redissolved water volume after SPE treatment. The method has been successfully applied to detection and quantification of CYN in water samples from aquaria of a toxicological in vivo laboratory experiment.

The nodulation of alfalfa by the acid-tolerant Rhizobium sp. strain LPU83 does not require sulfated forms of lipochitooligosaccharide nodulation signals.

The induction of root nodules by the majority of rhizobia has a strict requirement for the secretion of symbiosis-specific lipochitooligosaccharides (nodulation factors [NFs]). The nature of the chemical substitution on the NFs depends on the particular rhizobium and contributes to the host specificity imparted by the NFs. We present here a description of the genetic organization of the nod gene cluster and the characterization of the chemical structure of the NFs associated with the broad-host-range Rhizobium sp. strain LPU83, a bacterium capable of nodulating at least alfalfa, bean, and Leucena leucocephala. The nod gene cluster was located on the plasmid pLPU83b. The organization of the cluster showed synteny with those of the alfalfa-nodulating rhizobia, Sinorhizobium meliloti and Sinorhizobium medicae. Interestingly, the strongest sequence similarity observed was between the partial nod sequences of Rhizobium mongolense USDA 1844 and the corresponding LPU83 nod genes sequences. The phylogenetic analysis of the intergenic region nodEG positions strain LPU83 and the type strain R. mongolense 1844 in the same branch, which indicates that Rhizobium sp. strain LPU83 might represent an early alfalfa-nodulating genotype. The NF chemical structures obtained for the wild-type strain consist of a trimeric, tetrameric, and pentameric chitin backbone that shares some substitutions with both alfalfa- and bean-nodulating rhizobia. Remarkably, while in strain LPU83 most of the NFs were sulfated in their reducing terminal residue, none of the NFs isolated from the nodH mutant LPU83-H were sulfated. The evidence obtained supports the notion that the sulfate decoration of NFs in LPU83 is not necessary for alfalfa nodulation.

Sinorhizobium fredii HH103 cgs mutants are unable to nodulate determinate- and indeterminate nodule-forming legumes and overproduce an altered EPS.

Sinorhizobium fredii HH103 produces cyclic beta glucans (CG) composed of 18 to 24 glucose residues without or with 1-phosphoglycerol as the only substituent. The S. fredii HH103-Rifr cgs gene (formerly known as ndvB) was sequenced and mutated with the lacZ-gentamicin resistance cassette. Mutant SVQ562 did not produce CG, was immobile, and grew more slowly in the hypoosmotic GYM medium, but its survival in distilled water was equal to that of HH103-Rifr. Lipopolysaccharides and K-antigen polysaccharides produced by SVQ562 were not apparently altered. SVQ562 overproduced exopolysaccharides (EPS) and its exoA gene was transcribed at higher levels than in HH103-Rifr. In GYM medium, the EPS produced by SVQ562 was of higher molecular weight and carried higher levels of substituents than that produced by HH103-Rifr. The expression of the SVQ562 cgsColon, two colonslacZ fusion was influenced by the pH and the osmolarity of the growth medium. The S. fredii cgs mutants SVQ561 (carrying cgs::Omega) and SVQ562 only formed pseudonodules on Glycine max (determinate nodules) and on Glycyrrhiza uralensis (indeterminate nodules). Although nodulation factors were detected in SVQ561 cultures, none of the cgs mutants induced any macroscopic response in Vigna unguiculata roots. Thus, the nodulation process induced by S. fredii cgs mutants is aborted at earlier stages in V. unguiculata than in Glycine max.

Structural determination of the Nod factors produced by Rhizobium gallicum bv. gallicum R602.

Rhizobium gallicum is a fast-growing bacterium found in European, Australian and African soils; it was first isolated in France. It is a microsymbiont which is able to nodulate plants of the genus Phaseolus. Rhizobium gallicum bv. gallicum R602 produces four extracellular signal molecules consisting of a linear backbone of N-acetyl glucosamine, bearing on the nonreducing terminal residue an N-methyl group and different N-acyl substituents. The four acyloligosaccharides terminate with a sulfated N-acetylglucosaminitol. This unit may be also acetylated. These structures were determined using carbohydrate and methylation analysis, mass spectrometric analysis and one-dimensional- and two-dimensional-nuclear magnetic resonance experiments. This work establishes the common structure that a lipochito-oligosaccharide must have so that the Rhizobium that produces and excretes it is able to nodulate plants of Phaseolus vulgaris. The substituents common to all the molecules are an N-methyl group and a C(18:1) fatty acid on the nonreducing terminal residue.

Structural determination of the lipo-chitin oligosaccharide nodulation signals produced by Rhizobium giardinii bv. giardinii H152.

Rhizobium giardinii bv. giardinii is a microsymbiont of plants of the genus Phaseolus and produces extracellular signal molecules that are able to induce deformation of root hairs and nodule organogenesis. We report here the structures of seven lipochitooligosaccharide (LCO) signal molecules secreted by R. giardinii bv. giardinii H152. Six of them are pentamers of GlcNAc carrying C 16:0, C 18:0, C 20:0 and C 18:1 fatty acyl chains on the non-reducing terminal residue. Four are sulfated at C-6 of the reducing terminal residue and one is acetylated in the same position. Six of them are N-methylated on the non-reducing GlcN residue and all the nodulation factors are carbamoylated on C-6 of the non-reducing terminal residue. The structures were determined using monosaccharide composition and methylation analyses, 1D- and 2D-NMR experiments and a range of mass spectrometric techniques. The position of the carbamoyl substituent on the non-reducing glucosamine residue was determined using a CID-MSMS experiment and an HMBC experiment.

Alfalfa nodulation by Sinorhizobium fredii does not require sulfated Nod-factors.

Rhizobium strain 042B(s) is able to nodulate both soybean and alfalfa cultivars. We have demonstrated, by mass spectrometry, that the nodulation (Nod) factors produced by this strain are characteristic of those produced by Sinorhizobium fredii, which typically nodulates soybean; they have 3-5 N-acetylglucosamine (GlcNAc) residues, a mono-unsaturated or saturated C16, C18 or C20 fatty-acyl chain, and a (methyl)fucosyl residue on C6 of the reducing-terminal GlcNAc. In order to study Rhizobium strain 042B(s) and its nodulation behaviour further, we introduced an insertion mutation in the noeL gene, which is responsible for the presence of the (methyl)fucose residue on the reducing terminal GlcNAc of the Nod-factors, yielding mutant strain SVQ523. A plasmid (pHM500) carrying nodH, nodP and nodQ, the genes involved in sulfation of Nod-factors on C6 of the reducing-terminal GlcNAc, was introduced into SVQ523, generating SVQ523.pHM500. As expected, strain SVQ523 produces unfucosylated Nod-factors, while SVQ523.pHM500 produces both unfucosylated and unfucosylated sulfated Nod-factors. Plant tests showed that soybean nodulation was reduced if the inoculant (SVQ523.pHM500) produced sulfated Nod-factors. In the Asiatic alfalfa cultivar Baoding, SVQ523 (absence of a substitution at C6) failed to nodulate, but both 042B(s) (fucosyl at C6) and SVQ523.pHM500 (sulfate at C6) formed nodules. In contrast, SVQ523 showed enhanced nodulation capacity with the western alfalfa cultivars ORCA and ARC. These results indicate that Nod-factor sulfation is not a requisite for S. fredii to nodulate alfalfa.

Sinorhizobium fredii HH103 has a truncated nolO gene due to a -1 frameshift mutation that is conserved among other geographically distant S. fredii strains.

Strain SVQ121 is a mutant derivative of Sinorhizobium fredii HH103 carrying a transposon Tn5-lacZ insertion into the nolO-coding region. Sequence analysis of the wild-type gene revealed that it is homologous to that of Rhizobium sp. NGR234, which is involved in the 3 (or 4)-O-carbamoylation of the nonreducing terminus of Nod factors. Downstream of nolO, as in Rhizobium sp. NGR234, the noeI gene responsible for methylation of the fucose moiety of Nod factors was found. SVQ121 Nod factors showed lower levels of methylation into the fucosyl residue than those of HH103-suggesting a polar effect of the transposon insertion into nolO over the noel gene. A noeI HH103 mutant was constructed. This mutant, SVQ503, produced Nod factors devoid of methyl groups, confirming that the S. fredii noeI gene is functional. Neither the nolO nor the noeI mutation affected the ability of HH103 to nodulate several host plants, but both mutations reduced competitiveness to nodulate soybean. The Nod factors produced by strain HH103, like those of other S. fredii isolates, lack carbamoyl residues. By using specific polymerase chain reaction primers, we sequenced the nolO gene of S. fredii strains USDA192, USDA193, USDA257, and 042B(s). All the analyzed strains showed the same -1 frameshift mutation that is present in the HH103 nolO-coding region. From these results, it is concluded that, regardless of their geographical origin, S. fredii strains carry the nolO-coding region but that it is truncated by the same base-pair deletion.