The Makgadikgadi Salt Pans are the remnants of a mega paleo-lake system in the central Kalahari, Botswana. Today, the Makgadikgadi Basin is an arid to semi-arid area receiving water of meteoric origin during the short, wet season. Large microbial mats, which support primary production, are formed due to desiccation during the dry season. This study aimed to characterise the microbial diversity of the microbial mats and the underlying sediment. The focus was the Ntwetwe Pan, located west of the Makgadikgadi Basin. Metagenomic analyses demonstrated that the mats consisted of a high relative abundance of Cyanobacteriota (synonym Cyanobacteria) (20.50–41.47%), Pseudomonadota (synonym Proteobacteria) (15.71 to 32.18%), and Actinomycetota (synonym Actinobacteria) (8.53–32.56%). In the underlying sediments, Pseudomonadota, Actinomycetota, and Euryarchaeota represented over 70% of the community. Localised fluctuations in water content and pH did not significantly affect the microbial diversity of the sediment or the mats.
An amperometric immunoanalysis system based on monoclonal antibodies immobilized on Sepharose beads and packed into a micro-immunocolumn was developed for the quantification of microcystin-LR. Microcystin-LR (MCLR) was used as a reference microcystin variant. Inside the immunocolumn, free microcystins and microcystin-horseradish peroxidase (tracer) were sequentially captured by the immobilized antibodies, and the detection was performed electrochemically using Super AquaBlue ELISA substrate 2,2′-azinobis(3-ethylbenzothiazoline-sulfonic acid) (ABTS). The ABTS●+ generated by enzymatic oxidation of ABTS was electrochemically determined at a carbon working electrode by applying a reduction potential set at 0.4 V versus Ag/AgCl reference electrode. The peak current intensity was inversely proportional to the amount of analyte bound to the immunocolumn. The amperometric flow-ELISA system, which was automatically controlled through the CapSenzeTM (Lund, Sweden) computer software, enabled determination of MCLR as low as 0.01 µg/L. The assay time was very short (20 min for one assay cycle). In addition, the electrochemical signals were not significantly affected by possible interferences which could be present in the real samples. Along with the simplicity of automation, this makes the developed method a promising tool for use in water quality assessment.