Keywords

4.1 Introduction

Fungi are the major pathogens of plant , insects , and ectothermic vertebrates , but there are only a few of them that can cause diseases in mammals (Casadevall et al. 2019). Interaction between pathogens and their host can induce changes both in the host and in the pathogen, which might result in either the clearing of the pathogen from host or the establishing the pathogen within the host. These host-pathogen interactions can be analyzed with the help of systems biology approaches which generates novel hypothesis or decipher the effects of particular molecules or genes across the biological network (Peters et al. 2019).

Systems biology aims to understand the complex and dynamic biological information in the larger picture and requires the integration of different type of omics data such as proteomics, genomics, transcriptomics, and metabolomics (Pinu et al. 2019). This approach unravels the intricate network of interactions between host and fungal pathogen and helps to elucidate the complex pathogenesis processes of fungal interactions. Sayers et al. (2019) developed a web-based integrative resource for the analysis of virulence factor of different pathogens including fungus that causes infectious disease in human and animals. Virulence factors are small molecules that allow the microbial pathogens to survive and leads to diseased state in host.

To understand the infection mechanism , the whole host-fungal interaction system is more useful instead of investigating pathogen or host separately and is crucial to develop more effective solution. The combined investigation of host-fungal interaction might provide the understanding about the infectious disease in a way giving insight into the following questions.

  1. 1.

    How to identify direct interaction between fungus and host proteins during the infection and/or invasion process?

  2. 2.

    What signaling pathways and processes of the pathogen/host are implicated?

  3. 3.

    How to detect the crosstalk among the pathogen-host biochemical network?

  4. 4.

    What are the specific and unspecific responses of the host to pathogen invasion?

  5. 5.

    What specific protein or pathways could be targeted to control fungal infection and invasion in animals?

4.2 Systems Biology of Infection

Systems biology is an interdisciplinary research field which focuses on the study of nonlinear interactions among biological entities through combination of mathematical and computational approaches to study host-fungal interaction in biological system . This host-fungal interaction may be between proteins, nucleotides, metabolites, and small ligands. The traditional research typically focuses on single gene or fewer genes. The systems biology approach is insightful to understand physiology and infectious disease at cellular and molecular, network level. Systems biology approaches are categorized into bottom-up and top-down, where the bottom-up focus on network reconstruction through mathematical models while top-down approaches involves metabolic network reconstruction using “omics” data generated through high-throughput genomic techniques using appropriate statistical and bioinformatics methodologies (Shahzad and Loor 2012).

4.3 Data in Systems Biology and Analysis

Understanding the complex host fungal interaction requires high-throughput data as well as the annotations information available from public repositories (Table 4.1).

Table 4.1 Web-based bioinformatics resources for fungal systems biology
Full size table

4.3.1 Omics-Based Data

Fungal omics such as genomics , transcriptomics, proteomics, and metabolomics help to understand basic fungal biology and its associated functional implications. Due to recent advancement in sequencing techniques as well as a small size of the fungal genome, the analysis of the fungal genome and proteome data becomes much easier. The information obtained from omics data analysis may enhance our understanding about pathophysiology of infectious process and their underlying mechanisms of complex fungal biological process such as host-fungal interactions.

The free open source software “Bioconductor (Version 3.11) ” provides 1904 software package, 961 annotation, and 392 experimental datasets for bioinformatics analysis and comprehension of high-throughput genomic data.

4.3.1.1 Genomics

The new sequencing technology generates huge genomics datasets at a low cost and in a short time. This enables to explore more genomic information of fungal pathogen and would help to improve the diagnostic methods. SNP identification is one of the important methods for characterization of variants of different pathogens as well as to study the susceptibility of humans for different infections or pathogens. Saccharomyces cerevisiae is the first fungus whose complete genome sequence was available in 1996 (Goffeau et al. 1996); since then a number of human pathogen fungal genome have been published. The genome of C. albicans was published in 2004 (Jones et al. 2004), and the genome of C. neoformans and A. fumigatus were published in 2005 (Loftus et al. 2005; Nierman et al. 2005). There are also many freely sources available for different pathogenic fungal species (Table 4.1) such as Aspergillus Genome Database (AspGD) (Cerqueira et al. 2014) and Central Aspergillus Data Repository (CADRE) (Mabey et al. 2004) for Aspergillus and Candida Genome Database (CGD) (Skrzypek et al. 2017) and CandidaDB (d’Enfert et al. 2005) for Candida species. This freely available information allows researchers to identify and to investigate more about the host fungal interactions .

4.3.1.2 Transcriptomics

Pathogenic fungus during the course of infection process needs to effectively adapt to the host environment. These adaptation mechanisms are controlled by various transcriptional changes and it provides critical information regarding fungal pathogenesis (Amorim-Vaz and Sanglard 2015). The comparative analysis of the host and the fungal pathogen might help to design new antifungal drugs and to explore the clustered genes involved in the process of pathogenicity (Meijueiro et al. 2014).

Microarray , RNA-Seq , and nanoString are powerful tools to study the interaction between fungal pathogen and their host during the infection. Cryptococcus neoformans and Aspergillus fumigatus are the two important fungal species that cause high mortality in immune-compromised patients (Brown et al. 2012), and their transcriptomics study has already been performed (Chen et al. 2014; McDonagh et al. 2008). Bruno et al. in 2010 (Bruno et al. 2010) used RNA sequencing to generate a high-resolution transcriptome map of human pathogen Candida albicans under different environmental conditions and identified 602 new transcriptionally active regions (TARs) .

Deep sequencing also has been used to identify novel functional small RNAs, which has a great implication in the regulation of global gene expression in human pathogenic fungus (Nicolas and Ruiz-Vazquez 2013). Arthanari et al. (2014) used RNA sequencing data from ABI SOLiD platform to identify 939 novel long noncoding RNA (IncRNA) and 477 new natural antisense transcripts (NAT) from ascomycete fungus Neurospora crassa under different light and temperature conditions. Gene Expression Omnibus (GEO) (Barrett et al. 2013; Edgar et al. 2002) and ArrayExpress (Rustici et al. 2013) are the freely publicly available databases of fungal transcriptomics data.

Some reported virulence factors have been found to be related to transcriptional reprogramming associated with phase transition from avirulent mycelia to pathogenic yeast. For instance, in case of Paracoccidioides spp. , which causes paracoccidioidomycosis, the transition between these forms has been an important virulence factor (Tavares et al. 2015). Candida albicans , responsible for candidiasis, under comprised immunity and other opportunistic conditions results in superficial mucosal colonization and establishment of the systemic infection (Cheng et al. 2013; Wartenberg et al. 2014). According to RNA-seq studies in mice, during the vaginal infections caused by C. albicans, overexpression of aspartyl-proteinases 4, 5, and 6 (SAP4–6) has been reported which are hypha-associated secreted enzymes and function as inflammasome activators (Bruno et al. 2015). The virulence factor for Cryptococcus is a transcription factor Rim101 that is involved in the cell wall composition regulation (O’Meara et al. 2013). Cell wall forms a barrier at host-pathogen interface and is part of immune system. The structural integrity of the cell wall is maintained by the cell wall integrity signaling pathway that involves various crucial kinases. Dual RNA-seq is a high-throughput sequencing for simultaneous transcriptomic analysis of pathogen and host to explore the regulatory networks during the fungal infection (Das et al. 2015; Rienksma et al. 2015; Westermann et al. 2012).

4.3.1.3 Proteomics

Proteomics is another tool for the examination of expression patterns in the biological system. Proteins are the functional determinants and might have various roles such as to build the cellular structure; to mediate signal transduction, involved in the gene regulation; to influence different cellular processes; or to function as enzyme. Proteomic studies were initially performed using 2D-PAGE followed by mass spectrometry. Mass spectrometry is still the gold standard for proteomic analysis in the field of systems biology.

The release of the genome sequences of Candida albicans (Braun et al. 2005; Jones et al. 2004) and Aspergillus fumigatus (Nierman et al. 2005) has been of great benefit for a more detailed insight into the evolution and pathogenesis of these medically important fungus since for these pathogenic fungus, 2D gel proteome reference maps are available (Kniemeyer et al. 2011). Vodisch et al. (2009) identified 334 different mycelial proteins via 2D gel electrophoresis of Aspergillus fumigatus .

4.3.2 Image-Based Data

Image-based systems biology approaches are important tools for the investigation and elucidation of pathobiology of fungal infection. It can be divided into four basic steps: (i) extraction of image data through experimental techniques, (ii) preprocessing of the data, (iii) quantitative characterization of biological processes, and (iv) and image-based model development (Fig. 4.1).

Fig. 4.1
figure 1

Diagrammatic representation of image-based systems biology approaches

Full size image

Multiphoton microscopy (MPM) and fluorescence microscopy is one of the most valuable tools for live-cell imaging in infection research (Ettinger and Wittmann 2014; Sun et al. 2017). For the analysis of host-pathogen interaction , a large number of images are required that can be obtained by different experimental approaches such as fluorescence microscopy and positron emission tomography-computed tomography (PET/CT) . Ibrahim-Granet et al. (2003) described the detail analysis of interaction between A. fumigatus and phagocytes using fluorescence microscopy and manual image analysis.

Manual image data analysis with regard to host-pathogen interaction is time consuming, and also error-prone ; on the other hand, an automated image data analysis can analyze a large amount of data in a shorter period of time and is not labor extensive. Mech et al. (2011) developed an automated image analysis for the host-pathogen interaction between Aspergillus fumigatus conidia with immune cells. They used confocal laser scanning microscopy images from macrophages with different A. fumigatus strains. In this study, phagocytosed cells were differentially stained with fluorescent dyes and visualized with confocal laser scanning microscope, leading to discrimination between macrophages , internalized conidia , and non-internalized conidia . Currently there are a number of image analysis tools available for fungal analysis (Brunk et al. 2018; Cairns et al. 2019; Mader et al. 2015; Wurster et al. 2019). All image analysis methods comprise mainly three main parts: (i) pre-processing, (ii) segmentation, and (iii) classification.

Image-based systems biology is a relatively new scientific approach that comprises imaging, quantitative characterization, and modeling. In host-fungal interactions, image-derived models have been used to investigate the phagocytosis of fungal spores and the growth of fungal hyphae (Mech et al. 2011). It enhances our understanding of many aspects of cell migration and interaction behavior of the immune cells. Image-based systems biology focuses on the spatial factors and on cellular morphology (Horn et al. 2012).

4.4 Modeling

The computational modeling of networks of the genes, transcripts, proteins, and metabolites is of great importance in biomedical research to understand the molecular mechanism of the host-fungal interactions and helps to understand the complex regulatory mechanism. The aim of the modeling is to support the standardized design of experiments, to generate the hypothesis, and to validate the hypothesis. The modeling also allows the prediction of novel regulatory biological strategies (Horn et al. 2012). May and Anderson (1979) described the host-pathogen interaction through the evolutionary model. Due to the challenges of modeling of a complex system, only a few models of fungal-human interactome network have been studied (Horn et al. 2012). Rodrigues et al. (2018) developed transkingdom network (TransNet) analysis pipeline that allows to make biological inferences. They constructed a network using correlation between differentially expressed elements and integration of high-throughput data from different taxonomic kingdoms.

4.4.1 Network Modeling

Biological network modeling enables to study how the system can respond to the ever-changing external environment. Predictive power of such model enables the diagnosis by the prediction of biomarkers and drug targets. Cellular behavior of organism can be represented by gene regulatory network, protein-protein network, signaling network, and metabolic network. In the network modeling, the nodes represent molecular entities such as genes, proteins, and metabolites, and the edges represent the relationship between the nodes and can be modeled in different ways like directed or undirected edges (Horn et al. 2012). Aho et al. (2010) made an effort to reconstruct a model, which integrated the genomics, transcriptomics, proteomics, and metabolic data. Some of the biological networks of fungus that have been extensively studied in recent years have been discussed below.

4.4.1.1 Gene Regulatory Network

Gene regulatory network (GRN) is the network of genes that either interact physically or have genetic interaction to regulate a pathway or to carry out a specific role (Karlebach and Shamir 2008). Gene regulatory network helps to predict the gene that might function as central and key player in the network. It also describes the relationship between the regulatory components and can help to predict their target gene (Guo et al. 2016). Among the fungal species, GRN studies has been extensively focused on Saccharomyces cerevisiae (Darabos et al. 2011; Guelzim et al. 2002; Hu et al. 2007; Kim et al. 2006; Lee et al. 2002; Pe’er et al. 2002; Segal et al. 2003), Candida albicans (Homann et al. 2009; Ramachandra et al. 2014), and Fusarium graminearum (Guo et al. 2016). Many computational methods such as Gaussian graphical model (Kishino and Waddell 2000; Schafer and Strimmer 2005), probabilistic Boolean network (Glass and Kauffman 1973; Kauffman 1969; Shmulevich et al. 2002), linear model, and regression and Bayesian network (Friedman et al. 2000; Pe’er et al. 2002; Segal et al. 2003) are available for GRN reconstruction. Pe’er et al. (2002) used Bayesian network to reconstruct yeast GRN. Tierney et al. (2012) generated the first interspecies computational model of host-pathogen interactions and used it to decipher the mechanism of microbial pathogenesis. They used RNA-Seq expression data from Candida albicans and bone marrow-derived dendritic cells from M. musculus.

4.4.1.2 Protein-Protein Interaction (PPI) Network

In the post-genomic era, genes and its corresponding proteins are very useful for the identification of intra- and interspecies protein interaction networks (Durmus et al. 2015). These protein-protein interaction (PPI) networks have been used to identify the host immune-associated genes and the pathogenic effector proteins associated with host infection. Wang et al. (2013) integrated multiple omics data and based on the inference of ortholog-based PPI and dynamic modeling of regulatory responses have constructed an interspecies PPI network for Candida albicans and zebrafish. Development of experimental techniques to produce large-scale molecular interaction data and further an increase in the amount of experimentally validated host-fungal PPI data provides opportunity to perform a number of computational studies to investigate infection mechanism for different pathogen types (Durmus et al. 2015).

4.4.1.3 Signaling Network

Signaling network is a cell-to-cell communication network that allows the cell to respond to external signals through the change in transcription. When the cell receives the external signal through its membrane, it activates the cascade of events and ultimately affects the transcriptional. These signals or stimulus can be of different type such as chemical, physical, radiation, pathogen, etc. An excellent example of a highly conserved signaling network within all eukaryotes is the Regulation of Ace2 and Morphogenesis (RAM) network (Kurischko et al. 2005; Nelson et al. 2003; Saputo et al. 2012). This network is most extensively studied in Saccharomyces cerevisiae and Candida albicans and is less characterized for pathogenic fungus.

4.4.1.4 Metabolic Network

During infection, the pathogen interacts with the host cell and it causes alterations in the level of metabolites by affecting the host metabolic pathways (Cakir et al. 2020). Metabolic network model describes the metabolic state of the cell, and it allows for an in-depth insight into the molecular mechanisms and systems level predictions of metabolism in a variety of organisms. Recently, Wang et al. (2018) used a systems biology approach to predict and characterize human gut microbial metabolite in colorectal cancer.

The first genome-scale metabolic model for Saccharomyces cerevisiae was presented in 2003 (Forster et al. 2003), and it was the first and the most studied genome-scale network reconstruction for the eukaryotic organism. Since eukaryotic organisms are very complex in nature, researchers are continuing applying their efforts to improve and upgrade information into the metabolic network. Till date a number of genome-scale reconstruction of metabolic networks have been released (Dobson et al. 2010; Duarte et al. 2004; Heavner et al. 2012; Herrgard et al. 2008; Mo et al. 2009; Nookaew et al. 2008; Osterlund et al. 2013; Zomorrodi and Maranas 2010). Saccharomyces cerevisiae metabolic network has also been used as the basis for construction of metabolic models for other yeast, such as Saccharomyces pombe (Sohn et al. 2012), Yarrowia lipolytica (Loira et al. 2012), Pichia pastoris, and Pichia stipites (Caspeta et al. 2012). There are many researchers who are working on the tool development for reconstruction of genome-scale metabolic network (O’Brien et al. 2015; Pusa et al. 2020; Tefagh and Boyd 2020). Recently, Tefagh and Boyd in 2020 (Tefagh and Boyd 2020) proposed SWIFTCC and SWIFTCORE as effective methods for flux consistency checking and for the context specific reconstruction of genome-scale metabolic networks, while Pusa et al. (2020) developed a mathematical tool “MOOMIN” which uses genome-scale metabolic reconstruction to infer metabolic changes from differential expression data.

4.5 Conclusion and Future Implications

Unlike bacterial and viral infections, the fungal infections are uncommon and have been mainly associated with the immunocompromised patients due to conditions such as transplantation, tuberculosis, and HIV due to which they lead to a high mortality rate (Kumar and Ruhel 2019; Romani 2011). Understanding the intrinsic complexity of the invasive infection and exploring the transcriptional reprogramming in the host cells upon fungal infection might help to design more efficient and broad-range therapeutic strategies and drug targets. This would also expand the knowledge about the biomarkers for progression of the infection. Noncoding RNAs (ncRNAs) are emerging as important players in various stages of infection process such as colonization of cells and signaling and in other pathologically related functions. It would be interesting to explore the involvement of more players such as the noncoding RNAs along with the epigenetic factors during the host-fungal interaction and its implications on the infection process. The study of ncRNAs by small RNA (sRNA) sequencing during the fungal infection in the host cells combined with transcriptomics data could be another approach.

The recent advanced high-throughput techniques such as RNA-seq, sRNA-seq, and dual-seq analysis have been of a great importance to understand the crosstalk between various pathways triggered by the fungus pathogen in the infected host cells. Expansion in the data repositories related to host-fungal interaction along with systems biology modeling could be helpful for personalized medicine which is an emerging research (Dix et al. 2016). Combining various approaches will be an important route for infection-related research over the next decade (Yeung et al. 2019). Integration of the experimental data, computational and mathematical model, and information from the database provide a better model (Fig. 4.2) which can be used to validate a working hypothesis as well as to test new computational models. In future, the modeling of host-fungal interaction networks will be important to complement the experimental work to enhance the understanding of the system, to reduce animal experiments, and to generate and test hypothesis faster.

Fig. 4.2
figure 2

An illustration of the various system approaches that can be applied to test the hypothesis to study host-fungal interactions

Full size image