Volume 11 Supplement 11
Revealing parasite influence in metabolic pathways in Apicomplexa infected patients
- Tao Xu†1,
- Jie Ping†1,
- Yao Yu2,
- Fudong Yu3,
- Yongtao Yu3,
- Pei Hao2, 3Email author and
- Xuan Li2Email author
© Hao et al; licensee BioMed Central Ltd. 2010
Published: 14 December 2010
As an obligate intracellular parasite, Apicomplexa interacts with the host in the special living environment, competing for energy and nutrients from the host cells by manipulating the host metabolism. Previous studies of host-parasite interaction mainly focused on using cellular and biochemical methods to investigate molecular functions in metabolic pathways of parasite infected hosts. Computational approaches taking advantage of high-throughput biological data and topology of metabolic pathways have a great potential in revealing the details and mechanism of parasites-to-host interactions. A new analytical method was designed in this work to study host-parasite interactions in human cells infected with Plasmodium falciparum and Cryptosporidium parvum.
We introduced a new method that analyzes the host metabolic pathways in divided parts: host specific subpathways and host-parasite common subpathways. Upon analysis on gene expression data from cells infected by Plasmodium falciparum or Cryptosporidium parvum, we found: (i) six host-parasite common subpathways and four host specific subpathways were significantly altered in plasmodium infected human cells; (ii) plasmodium utilized fatty acid biosynthesis and elongation, and Pantothenate and CoA biosynthesis to obtain nutrients from host environment; (iii) in Cryptosporidium parvum infected cells, most of the host-parasite common enzymes were down-regulated, whereas the host specific enzymes up-regulated; (iv) the down-regulation of common subpathways in host cells might be caused by competition for the substrates and up-regulation of host specific subpathways may be stimulated by parasite infection.
Results demonstrated a significantly coordinated expression pattern between the two groups of subpathways. The method helped expose the impact of parasite infection on host cell metabolism, which was previously concealed in the pathway enrichment analysis. Our approach revealed detailed subpathways and metabolic information are important to the symbiosis in two kinds of the apicomplex parasites, and highlighted its significance in research and understanding of parasite-host interactions.
Living organisms gain energy from and discharge wastes to their environment. In the case of obligate intracellular parasites, living environment changes with the internal biochemical state of hosts. The need of intracellular parasites to retrieve nutrients and fulfill their energy requirements is achieved by manipulating the host metabolism. Such parasitic relationships, as well as other forms of species-to-species interactions, set the selective pressures to shape the evolution of each species . Therefore the metabolic networks are markedly impacted, e.g. many intracellular parasites have lost a substantial number of genes related to biosynthetic functions, and depend on the hosts for nutrients that they can not produce themselves[2, 3].
Many different approaches have been developed to study the possible interactions of host and parasite in metabolic pathways. Some traditional cellular and biochemical methods investigating protein or molecular functions in metabolic pathways were used in previous studies [4, 5]. Computational approaches using high throughput biological data would have great potential in revealing parasites-to-host interactions. Topology of metabolic pathways can be used to evaluate the biosynthetic support for a given pair of species by identifying the set of exogenously acquired compounds in the parasite metabolic network. Metabolites that parasites gained from the host can be predicted on the basis of topological structure of the metabolic pathway. It was shown previously that parasite tends to reduce their own metabolic pathways to make the metabolism most efficient, and utilize the metabolites from the host cells. The adaptation to parasitic environment, i.e. obtaining metabolites from their hosts, represents the life habit of these microorganisms [6, 7]. In such cases, it was hypothesized that there would be an inhibitive effect on the host genes which are downstream of metabolites that parasites need, but an increase on the genes that produce such metabolites. Thus, the infection of parasite would cause a biochemical imbalance in host cells . Our study is focused on a type of special protozoan, apicomplexa, which interacts with its host via its unique apical organelles, apicolast. This kind of parasite has a complex life cycle. After attachment and orientation on the erythrocyte surface, apicomplexa produce the small endocytic vacuole which accommodates the parasite in the feeding stage. This organelle is vital for the nutrient obtaining, reproducing, and other aspects of the parasite. It facilitates the direct protein-protein interactions and nutrient transportation between host and parasite [10, 11]. In the current study, the expressions of genes that encoded the related enzymes are examined. We hoped to evaluate the impact of infection on the expression of related genes based on the topological information of metabolic pathways, and the host-parasite ortholog gene pairs. The biological pathways that were involved in the nourishing relationship between host and parasite were analyzed to reveal the parasite-host interactions in greater details.
Metabolic pathway information and expression data
A study on malaria previously reported by Ockenhouse, C. F. et al. provided the expression data of epithelial blood cell from acute infected patients and uninfected persons. The data was obtained from the GEO (GDS2362). Time series data were as retrieved from a work on the expression of Cryptosporidium parvum infected HCT-8 ileocecal epithelial cells at various time points, including 6 time points, up to 72 hours after infection with Cryptosporidium parvum (GDS2240). KEGG orthology (KO) information and pathway information of both host’s and parasite’s genes were downloaded from the Kyoto Encyclopedia of Genes and Genomes (KEGG) . The genes with ortholog in both parasite and the host were defined as host-parasite ortholog genes. Genes owned by host alone (without ortholog in parasite) were defined as host-specific genes.
Identification of differentially expressed genes (DEGs)
The dataset of GDS2362 was obtained from Affymetrix Human Genome U95 Version 2 Array (GPL8300), and that of GDS2362 from Affymetrix Human Genome U133A Array (GPL96). We used the BioConductor package ‘affy’ to perform the data preprocessing and normalization. Both datasets were background corrected and normalized using Robust Multichip Average (RMA) algorithm in ‘affy’ package. Genes with absolute log2-fold changes greater than 1.5 and a P value less than 0.05 were considered significantly differentially expressed genes. A list of differentially expressed genes, their fold changes, and P values of t-test at different time points is provided in the additional files 1,2,3.
Definition of host-parasite ortholog genes
Original KEGG-defined ortholog genes were used in our analysis without any changes. KEGG provides a list of ortholog genes, which were downloaded from KEGG web site (http://www.genome.jp/kegg/). We identified the host-parasite ortholog gene pairs as: genes from the genomes of both host and parasite that belong to a group of ortholog genes in KEGG (one KO group).
Enrichment analysis of DEGs on metabolic pathways
The pathway regulation pattern in different gene sets
The host-parasite common subpathways significantly up- or down- regulated in plasmodium infected human blood cells.
Synthesis and degradation of ketone bodies
Linoleic acid metabolism
Pantothenate and CoA biosynthesis
Terpenoid backbone biosynthesis
Fatty acid biosynthesis
Fatty acid elongation in mitochondria
Arginine and proline metabolism
alpha-Linolenic acid metabolism
One carbon pool by folate
Limonene and pinene degradation
Biosynthesis of unsaturated fatty acids
Valine, leucine and isoleucine biosynthesis
Analysis of plasmodium infected human cell
Firstly, we tried to find which genes are differentially expressed after infection. Using false discovery rate (FDR) less than 0.01, we found a group of genes whose expressions were significantly altered when comparing the epithelial blood expression profile from acutely infected patient with from uninfected people. After that, we performed enrichment analysis to detect significantly altered expression pathways using the hyper geometric method (see Method), which may reveal significance of gene-set wise alternation. The result indicates two pathways (KEGG ID hsa00062: Fatty acid elongation in mitochondria, hsa00410: beta-Alanine metabolism) were significantly altered (FDR<=0.01) in the samples obtained from patients having clinically apparent malaria.
Then we separated pathways into host specific pathways and host-parasite common pathways. According to the KEGG orthology (KO) information, we first identified 1357 genes for which host-parasite ortholog pairs exist, and 8092 genes which were found only in host. We defined a subpathway consisting of ortholog pairs as host-parasite common subpathway, and a subpathway consisting of host only genes as a host specific subpathway. We subsequently performed enrichment test on each subpathway. Filtering out the subpathway s that contain only one gene, the result showed 6 host-parasite common subpathways and 4 host specific subpathways were significantly altered (Additional File 1).
Previously differentially expressed genes were classified to up-regulated or down-regulated gene sets in order to analyze the response of host cells after infection of pathogen. In this study, we further divided them into four different sub-sets according to their up- or down-regulation. We did enrichment test on each of these four sets, and found several pathways that had either host-parasite common or host specific subpathways showing an up- or down-regulation pattern (Table 1). All the test p-values were adjusted using Bonferroni correction. To analyze data more comprehensively we provided the results with multiple thresholds: p-value≤0.05, and p-value≤0.1. Furthermore, we found there exist correlated changes among these subpathways, and more importantly we identified subpathways that plasmodium utilized to obtain nutrients for its own development (Figure 1). They include Fatty acid biosynthesis and elongation (KEGG ID hsa00061 and hsa00062), and Pantothenate and CoA biosynthesis (KEGG ID hsa00770).
Pantothenate is another important nutrient for plasmodium. Here, we also detected significant alternation in the pathway of Pantothenate and CoA biosynthesis (KEGG ID hsa00770). P. falciparum requires an extracellular supply of pantothenate to support its proliferation during the erythrocytic stage of its development in humans . We found host-specific part of its upstream pathways, such as Propanoate metabolism (KEGG ID hsa00640), beta-Alanine metabolism (KEGG ID hsa00410), and Arginine and proline metabolism (KEGG ID hsa00330) all significantly up-regulated. But the host-parasite common part of Pantothenate and CoA biosynthesis (KEGG ID hsa00770) and its downstream pathway Terpenoid backbone biosynthesis (KEGG ID hsa00900) both significantly down-regulated. Based on this result, we hypothesized that some compound in these pathway being scavenged by plasmodium from host cell. One recent finding by Kellen L. Olszewski et.al. provided support to our hypothesis. Their study showed that the extracellular arginine in infected host cells differed from uninfected cells irrespective of the developmental stage. The Plasmodium converts arginine to ornithine and causes depletion in the host cell which is related to the human malarial hypoargininemia associated with cerebral malaria pathogenesis.
Other detected pathways that have their subpathways up or down-regulated also show important aspects of plasmodium during its development. The host-parasite common subpathway of Linoleic acid metabolism (KEGG ID hsa00591) was gene enriched and showed a down-regulation pattern. Linoleic acid is well known for its promotion of beta-hematin formation in vitro. Previous study suggests that linoleic acid or diglyceride containing dilinoleolyglycerol plays the critical role in promoting FP polymerization in malaria parasites.
Analysis of Cryptosporidium parvum and its infection in different time points
Metabolic pathways which were categorized as lipid metabolism were also included in DEGs significantly enriched host-parasite common subpathways. According to the study of Zhu and Fritzler, de novo biosynthesis of fatty acids probably did not occur in C. parvum [23, 24]. But C. parvum possessed three fatty acyl synthases (ACS). These enzymes activated fatty acids by converting them into fatty acyl-CoAs, and in some systems these same proteins can function in lipid trafficking and import .
Purine biosynthesis and amino acid metabolic pathways, including Alanine, aspartate and glutamate metabolism, and Glycine, serine and threonine metabolism did not exist in C. parvum either. The metabolites were imported from hosts instead, which is a common feature of apicomplexans. It is worth noting that these pathways were all detected as significant according to our results.
Significantly altered sub-pathways at different time points after infection of Cryptosporidium
Time post infection
From the analysis of gene expression of Plasmodium infected cells, we found the host-parasite common subpathway are more likely to present a down-regulation pattern, as there is only one subpathway that is enriched with up-regulated genes. Moreover, subsequent analysis indicates this phenomenon also existed in Cryptosporidium parvum infected human cells. We performed Wilcoxon signed rank test between the number of up- and down-regulation genes enriched host parasite common subpathways, the result was significant (p-value= 0.02726). Here we did not include the dataset of infection after 6 hours which is still in the invasion stage, and different from others.
Something interesting in the cells infected by the Plasmodium is that the number of pathways having a down-regulated pattern on ortholog genes grew with time, while the same manner appeared in the pathways with a significant alteration in ortholog genes. We excluded the result of 6 hours after infection which was still in the invasion stage, and Spearman correlation test showed positive correlation (correlation coefficient= 0.9486833, p-value= 0.02566, one-sided test), indicating that as infection progressed, more host-parasite common subpathways were impacted, which displayed down-regulation pattern on enzymes or compounds in the corresponding subpathways. However, more samples at different times after infection are needed to further test and verify this trend.
One factor that caused the down-regulation of host-parasite common subpathways in the metabolic pathways was competition of metabolites by the parasite. The gene in fatty acid metabolism pathway (KEGG ID hsa00071) which encodes the acyl-CoA synthetase long-chain family member 1 (EC: 184.108.40.206) was significantly down-regulated. Previous study suggested that the enzymes encoded by ortholog of plasmodium, long chain fatty acid ligase (EC: 220.127.116.11), would enter the host cytoplasm through the apicomplex and was involved in the formation of Hexadecanoyl-CoA [4, 5]. The activity of parasite enzyme also activated the downstream formation of CoA (see Figure 2), which would cause downstream genes in the host-specific subpathway significantly up-regulated.
The host-specific subpathways might tend to have a up-regulate pattern because of the interference of parasite enzyme in the host-parasite common subpathways, as was shown in the numbers of host-specific subpathways in Plasmodium infected cells: 12 subpathways were enriched with up-regulated genes and 4 were enriched with down-regulated genes. This assumption could not be fully proved from the analysis of C. parvum (p-value = 0.5724). The expression pattern of host specific subpathways was rather complex as genes in these pathways were under complicated regulations. The topology of these subpathways may also contribute to the complexity: some of them are up-stream of the host-parasite common subpathways and some have parallel host-specific subpathways. Another reason for the variation to predict nutrition relationship between host and parasite might be due to the incompleteness of the pathways from which we retrieve the topological information. Because the KEGG pathway was structured mainly on the genome annotation from the sequence data of a few species, unavailablely there might be inaccuracy associated with some particular pathway. There would be gene merely similar at the sequence level, but it may not necessarily have the same function as the orthologs in other species. According to the gene annotation of KEGG database, a large number of putative genes were found in the KEGG metabolic pathway of these two species: Plasmodium falciparum and Cryptosporidium parvum. Care must be taken in metabolic pathways analysis, and more study needs to be done with updated pathway information.
In this study, we introduced a new method for exposing the interactions between infected host cells and parasites: Plasmodium falciparum and Cryptosporidium parvum. We divided the host metabolic pathway into two parts: host specific subpathways and host-parasite common subpathways according to ortholog gene information of host and parasite. We performed analysis on the subpathways separately, and the results showed different character between these two groups subpathways. With this method, we revealed the impact of parasite infection on host cell gene expression, which was previously concealed in the pathway enrichment analysis. We provided the results from statistical analyses and visualized them by mapping the data to the KEGG pathway. Our approach revealed detailed subpathways and metabolic information that are important to the symbiosis in two kinds of the apicomplex parasites. The results highlight the significance of our approach in research and understanding of the interactions between parasites and host cells.
This work was supported by the grants from the National High Technology Research and Development Program of China (Contract No. 2007AA02Z330, 2009AA02Z308 and 2007AA02Z332), the National Natural Science Foundation of China (90913009, 30900834), National Key Basic Research Program (Grant No. 30800210) and Shanghai Committee of Science and Technology (grant 07JC14048, 08JC1416600) , Shanghai Postdoctoral Scientific Program (09R21411100) and Major National Science and Technology Projects (2009ZX10004-302).
This article has been published as part of BMC Bioinformatics Volume 11 Supplement 11, 2010: Proceedings of the 21st International Conference on Genome Informatics (GIW2010). The full contents of the supplement are available online at http://www.biomedcentral.com/1471-2105/11?issue=S11.
- Gandon S, Buckling A, Decaestecker E, Day T: Host-parasite coevolution and patterns of adaptation across time and space. J Evol Biol 2008, 21(6):1861–1866. 10.1111/j.1420-9101.2008.01598.xView ArticlePubMedGoogle Scholar
- McCutcheon JP, Moran NA: Parallel genomic evolution and metabolic interdependence in an ancient symbiosis. Proc Natl Acad Sci U S A 2007, 104(49):19392–19397. 10.1073/pnas.0708855104PubMed CentralView ArticlePubMedGoogle Scholar
- Edwards JS, Palsson BO: Robustness analysis of the Escherichia coli metabolic network. Biotechnol Prog 2000, 16(6):927–939. 10.1021/bp0000712View ArticlePubMedGoogle Scholar
- Tellez M, Matesanz F, Alcina A: The C-terminal domain of the Plasmodium falciparum acyl-CoA synthetases PfACS1 and PfACS3 functions as ligand for ankyrin. Mol Biochem Parasitol 2003, 129(2):191–198. 10.1016/S0166-6851(03)00123-3View ArticlePubMedGoogle Scholar
- Beaumelle BD, Vial HJ: Acyl-CoA synthetase activity in Plasmodium knowlesi-infected erythrocytes displays peculiar substrate specificities. Biochim Biophys Acta 1988, 958(1):1–9.View ArticlePubMedGoogle Scholar
- Borenstein E, Feldman MW: Topological signatures of species interactions in metabolic networks. J Comput Biol 2009, 16(2):191–200. 10.1089/cmb.2008.06TTPubMed CentralView ArticlePubMedGoogle Scholar
- Borenstein E, Kupiec M, Feldman MW, Ruppin E: Large-scale reconstruction and phylogenetic analysis of metabolic environments. Proc Natl Acad Sci U S A 2008, 105(38):14482–14487. 10.1073/pnas.0806162105PubMed CentralView ArticlePubMedGoogle Scholar
- Wood PA, Eaton JW: Hemoglobin catabolism and host-parasite heme balance in chloroquine-sensitive and chloroquine-resistant Plasmodium berghei infections. Am J Trop Med Hyg 1993, 48(4):465–472.PubMedGoogle Scholar
- Elford BC, Cowan GM, Ferguson DJ: Parasite-regulated membrane transport processes and metabolic control in malaria-infected erythrocytes. Biochem J 1995, 308(Pt 2):361–374.PubMed CentralView ArticlePubMedGoogle Scholar
- van Dooren GG, Tomova C, Agrawal S, Humbel BM, Striepen B: Toxoplasma gondii Tic20 is essential for apicoplast protein import. Proc Natl Acad Sci U S A 2008, 105(36):13574–13579. 10.1073/pnas.0803862105PubMed CentralView ArticlePubMedGoogle Scholar
- Parsons M, Karnataki A, Feagin JE, DeRocher A: Protein trafficking to the apicoplast: deciphering the apicomplexan solution to secondary endosymbiosis. Eukaryot Cell 2007, 6(7):1081–1088. 10.1128/EC.00102-07PubMed CentralView ArticlePubMedGoogle Scholar
- Kanehisa M, Goto S, Hattori M, Aoki-Kinoshita KF, Itoh M, Kawashima S, Katayama T, Araki M, Hirakawa M: From genomics to chemical genomics: new developments in KEGG. Nucleic Acids Res 2006, 34(Database issue):D354–357. 10.1093/nar/gkj102PubMed CentralView ArticlePubMedGoogle Scholar
- Gautier L, Cope L, Bolstad BM, Irizarry RA: affy--analysis of Affymetrix GeneChip data at the probe level. Bioinformatics 2004, 20(3):307–315. 10.1093/bioinformatics/btg405View ArticlePubMedGoogle Scholar
- Sartor MA, Leikauf GD, Medvedovic M: LRpath: a logistic regression approach for identifying enriched biological groups in gene expression data. Bioinformatics 2009, 25(2):211–217. 10.1093/bioinformatics/btn592PubMed CentralView ArticlePubMedGoogle Scholar
- Falcon S, Gentleman R: Using GOstats to test gene lists for GO term association. Bioinformatics 2007, 23(2):257–258. 10.1093/bioinformatics/btl567View ArticlePubMedGoogle Scholar
- Tarun AS, Vaughan AM, Kappe SH: Redefining the role of de novo fatty acid synthesis in Plasmodium parasites. Trends Parasitol 2009, 25(12):545–550. 10.1016/j.pt.2009.09.002View ArticlePubMedGoogle Scholar
- Spry C, van Schalkwyk DA, Strauss E, Saliba KJ: Pantothenate Utilization by Plasmodium as a Target for Antimalarial Chemotherapy. Infect Disord Drug Targets 9(5):Google Scholar
- Olszewski KL, Morrisey JM, Wilinski D, Burns JM, Vaidya AB, Rabinowitz JD, Llinas M: Host-parasite interactions revealed by Plasmodium falciparum metabolomics. Cell Host Microbe 2009, 5(2):191–199. 10.1016/j.chom.2009.01.004PubMed CentralView ArticlePubMedGoogle Scholar
- Fitch CD, Cai GZ, Shoemaker JD: A role for linoleic acid in erythrocytes infected with Plasmodium berghei. Biochim Biophys Acta 2000, 1535(1):45–49.View ArticlePubMedGoogle Scholar
- Deng M, Lancto CA, Abrahamsen MS: Cryptosporidium parvum regulation of human epithelial cell gene expression. Int J Parasitol 2004, 34(1):73–82. 10.1016/j.ijpara.2003.10.001View ArticlePubMedGoogle Scholar
- Abrahamsen MS, Templeton TJ, Enomoto S, Abrahante JE, Zhu G, Lancto CA, Deng M, Liu C, Widmer G, Tzipori S, et al.: Complete genome sequence of the apicomplexan, Cryptosporidium parvum. Science 2004, 304(5669):441–445. 10.1126/science.1094786View ArticlePubMedGoogle Scholar
- Zapata F, Perkins ME, Riojas YA, Wu TW, Le Blancq SM: The Cryptosporidium parvum ABC protein family. Mol Biochem Parasitol 2002, 120(1):157–161. 10.1016/S0166-6851(01)00445-5View ArticlePubMedGoogle Scholar
- Fritzler JM, Millership JJ, Zhu G: Cryptosporidium parvum long-chain fatty acid elongase. Eukaryot Cell 2007, 6(11):2018–2028. 10.1128/EC.00210-07PubMed CentralView ArticlePubMedGoogle Scholar
- Zhu G: Current progress in the fatty acid metabolism in Cryptosporidium parvum. J Eukaryot Microbiol 2004, 51(4):381–388. 10.1111/j.1550-7408.2004.tb00384.xView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.