An integrated approach utilizing artificial neural networks and SELDI mass spectrometry for the classification of human tumours and rapid identification of potential biomarkers

MOTIVATION: MALDI mass spectrometry is able to elicit macromolecular expression data from cellular material and when used in conjunction with Ciphergen protein chip technology (also referred to as SELDI-Surface Enhanced Laser Desorption/Ionization), it permits a semi-high throughput approach to be taken with respect to sample processing and data acquisition. Due to the large array of data that is generated from a single analysis (8-10000 variables using a mass range of 2-15 kDa-this paper) it is essential to implement the use of algorithms that can detect expression patterns from such large volumes of data correlating to a given biological/pathological phenotype from multiple samples. If successful, the methodology could be extrapolated to larger data sets to enable the identification of validated biomarkers correlating strongly to disease progression. This would not only serve to enable tumours to be classified according to their molecular expression profile but could also focus attention upon a relatively small number of molecules that might warrant further biochemical/molecular characterization to assess their suitability as potential therapeutic targets. RESULTS: Using a multi-layer perceptron Artificial Neural Network (ANN) (Neuroshell 2) with a back propagation algorithm we have developed a prototype approach that uses a model system (comprising five low and seven high-grade human astrocytomas) to identify mass spectral peaks whose relative intensity values correlate strongly to tumour grade. Analyzing data derived from MALDI mass spectrometry in conjunction with Ciphergen protein chip technology we have used relative importance values, determined from the weights of trained ANNs (Balls et al., Water, Air Soil Pollut., 85, 1467-1472, 1996), to identify masses that accurately predict tumour grade. Implementing a three-stage procedure, we have screened a population of approximately 100000-120000 variables and identified two ions (m/z values of 13454 and 13457) whose relative intensity pattern was significantly reduced in high-grade astrocytoma. The data from this initial study suggests that application of ANN-based approaches can identify molecular ion patterns which strongly associate with disease grade and that its application to larger cohorts of patient material could potentially facilitate the rapid identification of validated biomarkers having significant clinical (i.e. diagnostic/prognostic) potential for the field of cancer biology. AVAILIBILITY: Neuroshell 2 is commercially available from ward systems.     

Expression and secretion of human alpha1(I) procollagen fragment by Hansenula polymorpha as compared to Pichia pastoris

Secretion of a human collagen alpha1(I) chain fragment was achieved in Hansenula polymorpha using the native alpha1(I) procollagen secretory signal sequence. The N-terminal propeptide in the fragment was cleaved off during secretion, yielding the N-terminus of mature alpha1(I) collagen. In Pichia pastoris transformants, the expression of the fragment could be detected on RNA-level, but the product could not be determined extracellularly. After fusion of the fragment with a myc-HIS6 epitope, the intact product was found intracellularly. The difference in the extracellular level of the protein between the two expression hosts is most likely caused by difference in secretion efficiency.     

Characterisation of<Emphasis Type="Italic">fusarium graminearum</Emphasis> and<Emphasis Type="Italic">F. culmorum</Emphasis> isolates by mycotoxin production and aggressiveness to wheat

A total of 27Fusarium culmorum isolates from Germany and 41F. graminearum isolates from Kenya were investigated for aggressiveness and mycotoxin production on wheat ears. In addition, ergosterol content of the kernels from ears inoculated withF. graminearum was determined and theF. culmorum isolates were tested for mycotoxin productionin vitro. For both pathogens, isolates markedly differed in aggressiveness. 59% and 37% of theF. culmorum isolates produced NIV and DON, respectively,in vivo andin vitro. The DON-producing isolates also produced 3-acDONin vitro. The more aggressive isolates produced mainly DON while the less aggressive isolates produced mainly NIV. 12% and 85% of theF. graminearum isolates produced NIV and DON, respectively. The highly aggressive isolates produced higher amounts of DON, aggressiveness being highly correlated to DON content in the kernels. NIV-producing isolates were less aggressive. Ergosterol content of kernels was moderately correlated to aggressiveness but highly correlated to DON content. Disease severity was associated with kernel weight reduction.     

Distribution of four recently discovered galaxiid species in the Taieri River, New Zealand: the role of macrohabitat

The physical features of 20 streams inhabited by four recently discovered species of allopatric, non-migratory galaxiid fishes in the Taieri River, South Island, New Zealand were examined. Significant differences in particle size frequencies, combined waterfall frequency, channel depth variation and stream width: flood channel width ratios, indicated that Galaxias anomalus lived in low gradient shallow streams whereas G. depressiceps and G. eldoni occupied low to high gradient streams with more variability in depth. Climbing ability differed significantly among the species. Their allopatric patterns of distribution were, in part, the probable consequence of the abilities of G. depressiceps and G. eldoni to colonize areas that G. anomalus could not penetrate. Historical factors and reproductive biology of the four species are hypothesized also to be causal determinants of their current distributions.     

Maternal prenatal depressive symptoms predict infant NR3C1 1F and BDNF IV DNA methylation

Prenatal maternal psychological distress increases risk for adverse infant outcomes. However, the biological mechanisms underlying this association remain unclear. Prenatal stress can impact fetal epigenetic regulation that could underlie changes in infant stress responses. It has been suggested that maternal glucocorticoids may mediate this epigenetic effect. We examined this hypothesis by determining the impact of maternal cortisol and depressive symptoms during pregnancy on infant NR3C1 and BDNF DNA methylation. Fifty-seven pregnant women were recruited during the second or third trimester. Participants self-reported depressive symptoms and salivary cortisol samples were collected diurnally and in response to a stressor. Buccal swabs for DNA extraction and DNA methylation analysis were collected from each infant at 2 months of age, and mothers were assessed for postnatal depressive symptoms. Prenatal depressive symptoms significantly predicted increased NR3C1 1F DNA methylation in male infants (β = 2.147, P = 0.044). Prenatal depressive symptoms also significantly predicted decreased BDNF IV DNA methylation in both male and female infants (β = −3.244, P = 0.013). No measure of maternal cortisol during pregnancy predicted infant NR3C1 1F or BDNF promoter IV DNA methylation. Our findings highlight the susceptibility of males to changes in NR3C1 DNA methylation and present novel evidence for altered BDNF IV DNA methylation in response to maternal depression during pregnancy. The lack of association between maternal cortisol and infant DNA methylation suggests that effects of maternal depression may not be mediated directly by glucocorticoids. Future studies should consider other potential mediating mechanisms in the link between maternal mood and infant outcomes.