Introduction

Our understanding of how Giardia causes disease is incomplete, particularly concerning the early stages of trophozoite pathogenesis1. Giardia trophozoites attach strongly to the intestinal epithelial cells via a ventral adhesive disc and cause significant damage and disruption to gastroepithelial cells in the absence of cell invasion, secreted toxins and overt inflammation2. The interplay between the host and the parasite on establishment is a gap in our knowledge. Recently, host-parasite interaction models with human intestinal epithelial cells (IEC) in vitro have provided a foundation for understanding disease induction by Giardia trophozoites. Results indicate that these interaction models are stimulatory, inducing expression of parasite factors which have limited or no expression in axenic culture alone3. Additional studies have addressed gene expression and transcriptional changes in Giardia trophozoites co-incubated with Caco-2 and HCT-8 cells4 and HT-29 cells5 and analysed the secreted proteomes3. There have also been complementary studies of transcripts from IECs exposed to Giardia trophozoites6,7. Together, these studies indicate the efficacy of in vitro models to explore the induction of Giardiasis.

Proteomics is one of the few exploratory tools available to understand parasite biology at a physiological levelAnalysis of differentially expressed proteins

The relative quantitation for host-cell interactions were derived by the ratio of TMT labels for each of the treatments over their respective replicate control (i.e. HSF R1 against Con R1 and CI IEC R1 against Con R1). A total of three expression ratios were derived for both HSF and CI IEC biological replicates and an average fold change was calculated for each protein identified. In addition to TMT ratios for differential expression, proteins were analysed statistically via one-sample t-test to evaluate significance of observed protein expression changes. Proteins were only considered differentially expressed if they met both fold change criteria as well as ≤0.05 p-value significance. Functional annotation of proteins was performed using Uniprot to assign GO function, subcellular localisation, Interpro protein domains and structure annotations where available.

Prediction of secreted proteins was analysed using a series of bioinformatics tools to assess subcellular localisation63, presence of signal peptides64, transmembrane helices65 and nuclear localisation66. Proteins were submitted to TargetP v1.1 using the default settings for the algorithm for non-plant sequences, with a reliability score ≤3 selected for cutoff (with 1 being the highest reliability score). For analysis of signal peptides, proteins were submitted to SignalP v4.01, with the default settings for eukaryotic sequences. The presence of transmembrane helices was predicted using THMHH Server v2.0. Finally, as an exclusionary tool, proteins were submitted to NucPred to assess nuclear localisation signals with proteins above a prediction reliability score of ≥0.90 considered significant.

Several additional statistical analyses of the TMT dataset were performed to evaluate the variability of the dataset and establish the presence of an underlying biological difference between HSF and CI treatments compared to control. Details of the additional statistical analyses can be found in Supplementary Methods. These include an assessment of sample variability based on control/control ratios, an unsupervised multivariate principal component analysis (PCA) and an analysis of the p-value distribution using paired t-tests between triplicates of HSF/Control and CI/Control ratios.

The mass spectrometry raw data files, database search results and TMT labelling protein quantitation results have all been deposited to the ProteomeXchange Consortium67 via the PRIDE partner repository with the dataset identifier PXD002398.

Additional Information

How to cite this article: Emery, S. J. et al. Induction of virulence factors in Giardia duodenalis independent of host attachment. Sci. Rep. 6, 20765; doi: 10.1038/srep20765 (2016).