The interest in a fast, high specific and reliable detection method for bacteria identification is increasing. We will show that the application of vibrational spectroscopy is feasible for the validation of bacteria in microfluidic devices. For this purpose, reproducible and specific spectral pattern as well as the establishment of large databases are essential for statistical analysis. Therefore, short recording times are beneficial concerning the time aspect of fast identification. We will demonstrate that the requirements can be fulfilled by measuring ultrasonic busted bacteria by means of microfluidic lab-on-a-chip based SERS. With the applied sample preparation, high specificity and reproducibility of the spectra are achieved. Taking advantage of the SERS enhancement, the spectral recording time is reduced to 1 s and a database of 11,200 spectra is established for a model system E. coli including nine different strains. The validation of the bacteria on strain level is achieved accomplishing SVM accuracies of 92%. Within this contribution the potential of our approach of bacterial identification for future application is discussed, focusing on the time-benefit and the combination with other microfluidic applications.

Bacterial resistances against antibiotics are increasingly problematic for medical treatment of pathogenic bacteria, e.g., in hospitals. Resistances are, among other genes, often encoded on plasmids which can be transmitted between bacteria not only within one species, but also between different species, genera, and families. The plasmid pDrive is transformed into bacteria of the model strain Escherichia coli DH5α. Within this investigation, we applied micro-Raman spectroscopy with two different excitation wavelengths in combination with support vector machine (SVM) and linear discriminant analysis (LDA) to differentiate between bacterial cultures according to their cultural plasmid content. Recognition rates of about 92% and 90% are achieved by Raman excitation at 532 and 244 nm, respectively. The SVM loadings reveal that the pDrive transformed bacterial cultures exhibit a higher DNA content compared to the untransformed cultures. To elucidate the influence of the antibiotic, ampicillin-treated cultures are also comprised within this study and are classified with rates of about 97% and 100% for 532 and 244 nm Raman excitation, respectively. The Raman spectra recorded with 532 nm excitation wavelength show differences of the secondary protein structure and enhanced stress-related respiration rates for the ampicillin-treated cultures. Independent cultural replicates of either ampicillin-challenged or non-challenged cultures are successfully identified with identification rates of over 90%.

Micro-Raman spectroscopy is a fast and sensitive tool for the detection, classification, and identification of biological organisms. The vibrational spectrum inherently serves as a fingerprint of the biochemical composition of each bacterium and thus makes identification at the species level, or even the subspecies level, possible. Therefore, microorganisms in areas susceptible to bacterial contamination, e.g., clinical environments or food-processing technology, can be sensed. Within the scope of point-of-care-testing also, detection of intentionally released biosafety level 3 (BSL-3) agents, such as Bacillus anthracis endospores, or their products is attainable. However, no Raman spectroscopy-compatible inactivation method for the notoriously resistant Bacillus endospores has been elaborated so far. In this work we present an inactivation protocol for endospores that permits, on the one hand, sufficient microbial inactivation and, on the other hand, the recording of Raman spectroscopic signatures of single endospores, making species-specific identification by means of highly sophisticated chemometrical methods possible. Several physical and chemical inactivation methods were assessed, and eventually treatment with 20% formaldehyde proved to be superior to the other methods in terms of sporicidal capacity and information conservation in the Raman spectra. The latter fact has been verified by successfully using self-learning machines (such as support vector machines or artificial neural networks) to identify inactivated B. anthracis-related endospores with adequate accuracies within the range of the limited model database employed.The detection of biological warfare agents requires methods for detecting and rapidly identifying bacterial endosporessuch as Bacillus anthracis, the etiological agent of the acute fatal zoonosis anthrax in mammals-that are released in buildings or distributed in the environment. A great number of different technologies and combinations, such as DNA detection by PCR or DNA sequencing (42), are applied for genetic analysis. In addition, different microscopic approaches, such as atomic force microscopy (64) or fluorescence microscopy (26, 27), mass spectroscopy, and infrared (23,34,50) and Raman spectroscopy (15), have been used. With these optical detection methods, preparation and analysis time can be considerably shortened relative to that required for currently established methods based on, e.g., microbiology, immunoassays, and genetic and molecularly based approaches for identification.

Previous studies dealing with bacterial identification by means of Raman spectroscopy have demonstrated that micro-Raman is a suitable technique for single-cell microbial identification. Raman spectra yield fingerprint-like information about all chemical components within one cell, and combined with multivariate methods, differentiation down to species or even strain level is possible. Many microorganisms may accumulate high amounts of polyhydroxyalkanoates (PHA) as carbon and energy storage materials within the cell and the Raman bands of PHA might impede the identification and differentiation of cells. To date, the identification by means of Raman spectroscopy have never been tested on bacteria which had accumulated PHA. Therefore, the aim of this study is to investigate the effect of intracellular polymer accumulation on the bacterial identification rate. Combining fluorescence imaging and Raman spectroscopy, we identified polyhydroxybutyrate (PHB) as a storage polymer accumulating in the investigated cells. The amount of energy storage material present within the cells was dependent on the physiological status of the microorganisms and strongly influenced the identification results. Bacteria in the stationary phase formed granules of crystalline PHB, which obstructed the Raman spectroscopic identification of bacterial species. The Raman spectra of bacteria in the exponential phase were dominated by signals from the storage material. However, the bands from proteins, lipids, and nucleic acids were not completely obscured by signals from PHB. Cells growing under either oxic or anoxic conditions could also be differentiated, suggesting that changes in Raman spectra can be interpreted as an indicator of different metabolic pathways. Although the presence of PHB induced severe changes in the Raman spectra, our results suggest that Raman spectroscopy can be successfully used for identification as long as the bacteria are not in the stationary phase.

In this paper, a novel method for developing a tree-like classifier which differentiates between organic and inorganic particulate matter by means of Raman spectroscopy is introduced. The algorithm is fully automatic and optimises itself without any human interaction. This method uses a tree-like structure to classify Raman spectra as a decision tree. On every knot of this tree, the optimal classifier is automatically obtained, tested and trained. The optimal classifier is an artificial neural network, linear discriminant analysis or a support vector machine, where different kernels are possible. The support vector machine is optimised by the simulated annealing method to achieve the best possible classifier. After the training, a hold-out experiment with two completely independent sets of Raman spectra was tried to show the abilities of this method for real-world application.

Classification of Raman spectra recorded from single cells is commonly applied to bacteria that exhibit small sizes of approximately 1 to 2 μm. Here, we study the possibility to adopt this classification approach to filamentous bacteria of the genus Streptomyces. The hyphae can reach extensive lengths of up to 35 μm, which can correspond to a single cell identified in light microscopy. The classification of Raman bulk spectra will be demonstrated. Here, ultraviolet resonance Raman (UV RR) spectroscopy is chosen to classify six Streptomyces species by the application of a tree-like classifier. For each knot of the hierarchical classifier, estimated classification accuracies of over 94% are accomplished. In contrast to the classification of bulk spectra, the classification of single-cell spectra requires a homogenous substance distribution within the cell. Consequently, the bacterial cell chemistry can be represented by one individual spectrum. This requirement is not fulfilled when different spectra are processed from different locations within the cell. Bacteria of the investigated genus Streptomyces exhibit, besides the normal bacterial spectra, lipid-rich spectra. The occurrence of lipid enrichment depends on culture age and nutrition availability. With this study, we investigate the cell substance distribution, especially of lipid-rich fractions. The classification utilizing a tree-like classifier is also applied to the Streptomyces single-cell spectra, resulting in classification accuracies between 80 and 93% for the investigated Streptomyces species.

This paper presents four different ways to generate quantitative and qualitative mineral information out of a Raman mapping experiment. The results of the four methods are compared against each other, and the advantages and disadvantages are pointed out. The combination of the point-counting method with chemometric approaches is presented for the first time. The method with its different possible techniques is explained in detail. It will be shown that this technique works best for minerals.

No abstract