A Computational Approach to Identification of Candidate Biomarkers in High-Dimensional Molecular Data

Complex high-dimensional datasets that are challenging to analyze are frequently produced through ‘-omics’ profiling. Typically, these datasets contain more genomic features than samples, limiting the use of multivariable statistical and machine learning-based approaches to analysis. Therefore, effective alternative approaches are urgently needed to identify features-of-interest in ‘-omics’ data. In this study, we present the molecular feature selection tool, a novel, ensemble-based, feature selection application for identifying candidate biomarkers in ‘-omics’ data. As proof-of-principle, we applied the molecular feature selection tool to identify a small set of immune-related genes as potential biomarkers of three prostate adenocarcinoma subtypes. Furthermore, we tested the selected genes in a model to classify the three subtypes and compared the results to models built using all genes and all differentially expressed genes. Genes identified with the molecular feature selection tool performed better than the other models in this study in all comparison metrics: accuracy, precision, recall, and F1-score using a significantly smaller set of genes. In addition, we developed a simple graphical user interface for the molecular feature selection tool, which is available for free download. This user-friendly interface is a valuable tool for the identification of potential biomarkers in gene expression datasets and is an asset for biomarker discovery studies.

Topology preserving stratification of tissue neoplasticity using Deep Neural Maps and microRNA signatures

Background

Accurate cancer classification is essential for correct treatment selection and better prognostication. microRNAs (miRNAs) are small RNA molecules that negatively regulate gene expression, and their dyresgulation is a common disease mechanism in many cancers. Through a clearer understanding of miRNA dysregulation in cancer, improved mechanistic knowledge and better treatments can be sought.

Results

We present a topology-preserving deep learning framework to study miRNA dysregulation in cancer. Our study comprises miRNA expression profiles from 3685 cancer and non-cancer tissue samples and hierarchical annotations on organ and neoplasticity status. Using unsupervised learning, a two-dimensional topological map is trained to cluster similar tissue samples. Labelled samples are used after training to identify clustering accuracy in terms of tissue-of-origin and neoplasticity status. In addition, an approach using activation gradients is developed to determine the attention of the networks to miRNAs that drive the clustering. Using this deep learning framework, we classify the neoplasticity status of held-out test samples with an accuracy of 91.07%, the tissue-of-origin with 86.36%, and combined neoplasticity status and tissue-of-origin with an accuracy of 84.28%. The topological maps display the ability of miRNAs to recognize tissue types and neoplasticity status. Importantly, when our approach identifies samples that do not cluster well with their respective classes, activation gradients provide further insight in cancer subtypes or grades.

Conclusions

An unsupervised deep learning approach is developed for cancer classification and interpretation. This work provides an intuitive approach for understanding molecular properties of cancer and has significant potential for cancer classification and treatment selection.

Multi-Omic Graph Transformers for Cancer Classification and Interpretation

Next-generation sequencing has provided rapid collection and quantification of 'big' biological data. In particular, multi-omics and integration of different molecular data such as miRNA and mRNA can provide important insights to disease classification and processes. There is a need for computational methods that can correctly model and interpret these relationships, and handle the difficulties of large-scale data. In this study, we develop a novel method of representing miRNA-mRNA interactions to classify cancer. Specifically, graphs are designed to account for the interactions and biological communication between miRNAs and mRNAs, using message-passing and attention mechanisms. Patient-matched miRNA and mRNA expression data is obtained from The Cancer Genome Atlas for 12 cancers, and targeting information is incorporated from TargetScan. A Graph Transformer Network (GTN) is selected to provide high interpretability of classification through self-attention mechanisms. The GTN is able to classify the 12 different cancers with an accuracy of 93.56% and is compared to a Graph Convolutional Network, Random Forest, Support Vector Machine, and Multilayer Perceptron. While the GTN does not outperform all of the other classifiers in terms of accuracy, it allows high interpretation of results. Multi-omics models are compared and generally outperform their respective single-omics performance. Extensive analysis of attention identifies important targeting pathways and molecular biomarkers based on integrated miRNA and mRNA expression.

Discriminating Neoplastic from Nonneoplastic Tissues Using an miRNA-Based Deep Cancer Classifier

Next-generation sequencing has enabled the collection of large biological data sets, allowing novel molecular-based classification methods to be developed for increased understanding of disease. miRNAs are small regulatory RNA molecules that can be quantified using next-generation sequencing and are excellent classificatory markers. Herein, we adapt a deep cancer classifier (DCC) to differentiate neoplastic from nonneoplastic samples using comprehensive miRNA expression profiles from 1031 human breast and skin tissue samples. The classifier was fine-tuned and evaluated using 750 neoplastic and 281 nonneoplastic breast and skin tissue samples. Performance of the DCC was compared with two machine-learning classifiers: support vector machine and random forests. In addition, performance of feature extraction through the DCC was also compared with a developed feature selection algorithm, cancer specificity. The DCC had the highest performance of area under the receiver operating curve and high performance in both sensitivity and specificity, unlike machine-learning and feature selection models, which often performed well in one metric compared with the other. In particular, deep learning was shown to have noticeable advantages with highly heterogeneous data sets. In addition, our cancer specificity algorithm identified candidate biomarkers for differentiating neoplastic and nonneoplastic tissue samples (eg, miR-144 and miR-375 in breast cancer and miR-375 and miR-451 in skin cancer).

A miR-375/YAP axis regulates neuroendocrine differentiation and tumorigenesis in lung carcinoid cells

Lung carcinoids are variably aggressive and mechanistically understudied neuroendocrine neoplasms (NENs). Here, we identified and elucidated the function of a miR-375/yes-associated protein (YAP) axis in lung carcinoid (H727) cells. miR-375 and YAP are respectively high and low expressed in wild-type H727 cells. Following lentiviral CRISPR/Cas9-mediated miR-375 depletion, we identified distinct transcriptomic changes including dramatic YAP upregulation. We also observed a significant decrease in neuroendocrine differentiation and substantial reductions in cell proliferation, transformation, and tumor growth in cell culture and xenograft mouse disease models. Similarly, YAP overexpression resulted in distinct and partially overlapping transcriptomic changes, phenocopying the effects of miR-375 depletion in the same models as above. Transient YAP knockdown in miR-375-depleted cells reversed the effects of miR-375 on neuroendocrine differentiation and cell proliferation. Pathways analysis and confirmatory real-time PCR studies of shared dysregulated target genes indicate that this axis controls neuroendocrine related functions such as neural differentiation, exocytosis, and secretion. Taken together, we provide compelling evidence that a miR-375/YAP axis is a critical mediator of neuroendocrine differentiation and tumorigenesis in lung carcinoid cells.

Classifying Lung Neuroendocrine Neoplasms through MicroRNA Sequence Data Mining

Lung neuroendocrine neoplasms (NENs) can be challenging to classify due to subtle histologic differences between pathological types. MicroRNAs (miRNAs) are small RNA molecules that are valuable markers in many neoplastic diseases. To evaluate miRNAs as classificatory markers for lung NENs, we generated comprehensive miRNA expression profiles from 14 typical carcinoid (TC), 15 atypical carcinoid (AC), 11 small cell lung carcinoma (SCLC), and 15 large cell neuroendocrine carcinoma (LCNEC) samples, through barcoded small RNA sequencing. Following sequence annotation and data preprocessing, we randomly assigned these profiles to discovery and validation sets. Through high expression analyses, we found that miR-21 and -375 are abundant in all lung NENs, and that miR-21/miR-375 expression ratios are significantly lower in carcinoids (TC and AC) than in neuroendocrine carcinomas (NECs; SCLC and LCNEC). Subsequently, we ranked and selected miRNAs for use in miRNA-based classification, to discriminate carcinoids from NECs. Using miR-18a and -155 expression, our classifier discriminated these groups in discovery and validation sets, with 93% and 100% accuracy. We also identified miR-17, -103, and -127, and miR-301a, -106b, and -25, as candidate markers for discriminating TC from AC, and SCLC from LCNEC, respectively. However, these promising findings require external validation due to sample size.

Characterizing and classifying neuroendocrine neoplasms through microRNA sequencing and data mining

Neuroendocrine neoplasms (NENs) are clinically diverse and incompletely characterized cancers that are challenging to classify. MicroRNAs (miRNAs) are small regulatory RNAs that can be used to classify cancers. Recently, a morphology-based classification framework for evaluating NENs from different anatomical sites was proposed by experts, with the requirement of improved molecular data integration. Here, we compiled 378 miRNA expression profiles to examine NEN classification through comprehensive miRNA profiling and data mining. Following data preprocessing, our final study cohort included 221 NEN and 114 non-NEN samples, representing 15 NEN pathological types and 5 site-matched non-NEN control groups. Unsupervised hierarchical clustering of miRNA expression profiles clearly separated NENs from non-NENs. Comparative analyses showed that miR-375 and miR-7 expression is substantially higher in NEN cases than non-NEN controls. Correlation analyses showed that NENs from diverse anatomical sites have convergent miRNA expression programs, likely reflecting morphological and functional similarities. Using machine learning approaches, we identified 17 miRNAs to discriminate 15 NEN pathological types and subsequently constructed a multilayer classifier, correctly identifying 217 (98%) of 221 samples and overturning one histological diagnosis. Through our research, we have identified common and type-specific miRNA tissue markers and constructed an accurate miRNA-based classifier, advancing our understanding of NEN diversity.

Exploring microRNA Regulation of Cancer with Context-Aware Deep Cancer Classifier

BACKGROUND

MicroRNAs (miRNAs) are small, non-coding RNA that regulate gene expression through post-transcriptional silencing. Differential expression observed in miRNAs, combined with advancements in deep learning (DL), have the potential to improve cancer classification by modelling non-linear miRNA-phenotype associations. We propose a novel miRNA-based deep cancer classifier (DCC) incorporating genomic and hierarchical tissue annotation, capable of accurately predicting the presence of cancer in wide range of human tissues.

METHODS

miRNA expression profiles were analyzed for 1746 neoplastic and 3871 normal samples, across 26 types of cancer involving six organ sub-structures and 68 cell types. miRNAs were ranked and filtered using a specificity score representing their information content in relation to neoplasticity, incorporating 3 levels of hierarchical biological annotation. A DL architecture composed of stacked autoencoders (AE) and a multi-layer perceptron (MLP) was trained to predict neoplasticity using 497 abundant and informative miRNAs. Additional DCCs were trained using expression of miRNA cistrons and sequence families, and combined as a diagnostic ensemble. Important miRNAs were identified using backpropagation, and analyzed in Cytoscape using iCTNet and BiNGO.

RESULTS

Nested four-fold cross-validation was used to assess the performance of the DL model. The model achieved an accuracy, AUC/ROC, sensitivity, and specificity of 94.73%, 98.6%, 95.1%, and 94.3%, respectively.

CONCLUSION

Deep autoencoder networks are a powerful tool for modelling complex miRNA-phenotype associations in cancer. The proposed DCC improves classification accuracy by learning from the biological context of both samples and miRNAs, using anatomical and genomic annotation. Analyzing the deep structure of DCCs with backpropagation can also facilitate biological discovery, by performing gene ontology searches on the most highly significant features.

Evaluating gastroenteropancreatic neuroendocrine tumors through microRNA sequencing

Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) can be challenging to evaluate histologically. MicroRNAs (miRNAs) are small RNA molecules that often are excellent biomarkers due to their abundance, cell-type and disease stage specificity and stability. To evaluate miRNAs as adjunct tissue markers for classifying and grading well-differentiated GEP-NETs, we generated and compared miRNA expression profiles from four pathological types of GEP-NETs. Using quantitative barcoded small RNA sequencing and state-of-the-art sequence annotation, we generated comprehensive miRNA expression profiles from archived pancreatic, ileal, appendiceal and rectal NETs. Following data preprocessing, we randomly assigned sample profiles to discovery (80%) and validation (20%) sets prior to data mining using machine-learning techniques. High expression analyses indicated that miR-375 was the most abundant individual miRNA and miRNA cistron in all samples. Leveraging prior knowledge that GEP-NET behavior is influenced by embryonic derivation, we developed a dual-layer hierarchical classifier for differentiating GEP-NET types. In the first layer, our classifier discriminated midgut (ileum, appendix) from non-midgut (rectum, pancreas) NETs based on miR-615 and -92b expression. In the second layer, our classifier discriminated ileal from appendiceal NETs based on miR-125b, -192 and -149 expression, and rectal from pancreatic NETs based on miR-429 and -487b expression. Our classifier achieved overall accuracies of 98.5% and 94.4% in discovery and validation sets, respectively. We also found provisional evidence that low- and intermediate-grade pancreatic NETs can be discriminated based on miR-328 expression. GEP-NETs can be reliably classified and potentially graded using a limited panel of miRNA markers, complementing morphological and immunohistochemistry-based approaches to histologic evaluation.

MicroRNA-206 suppresses TGF-β signalling to limit tumor growth and metastasis in lung adenocarcinoma

MicroRNA-206 (miR-206) has demonstrated tumor suppressive effects in a variety of cancers. Numerous studies have identified aberrantly expressed targets of miR-206 that contribute to tumor progression and metastasis, however, the broader gene-networks and pathways regulated by miR-206 remain poorly defined. Here, we have ectopically expressed miR-206 in lung adenocarcinoma cell lines and tumors to identify differentially expressed genes, and study the effects on tumor growth and metastasis. In H1299 tumor xenograft assays, stable expression of miR-206 suppressed both tumor growth and metastasis in mice. Profiling of xenograft tumors using small RNA sequencing and a targeted panel of tumor progression and metastasis-related genes revealed a network of genes involved in TGF-β signalling that were regulated by miR-206. Among these were the TGFB1 ligand, as well as direct transcriptional targets of Smad3. Other differentially expressed genes included components of the extracellular matrix involved in TGF-β activation and signalling, including Thrombospondin-1, which is responsible for the activation of latent TGF-β in the stroma. In cultured lung adenocarcinoma cells treated with recombinant TGF-β, ectopic expression of miR-206 impaired canonical signalling, and expression of TGF-β target genes linked to epithelial-mesenchymal transition. This was due at least in part to the suppression of Smad3 protein levels in lung adenocarcinoma cells with ectopic miR-206 expression. Together, these findings indicate that miR-206 can suppress tumor progression and metastasis by limiting autocrine production of TGF-β, and highlight the potential utility of TGF-β inhibitors for the treatment of lung adenocarcinomas.

microRNA-guided diagnostics in clinical samples

miRNA-guided diagnostics is a powerful molecular approach for evaluating clinical samples through miRNA detection and/or visualization. To date, this approach has been successfully used to diagnose, manage, and/or monitor a wide range of neoplastic and non-neoplastic diseases. Despite the promise of miRNA-guided diagnostics, particularly in the field of minimally invasive biomarkers, several knowledge and practical issues confound or hinder translation into routine clinical practice including: miRNA sequence database errors, suboptimal RNA extraction methods, detection assay variability, a vast array of online resources for bioinformatic analyses, and non-standardized statistical analyses for miRNA clinical testing. In this review, we raise awareness of these issues and recommend research directions to help specialists in endocrinology and metabolism integrate miRNA testing into clinical decision-making.

Unique microRNAs appear at different times during the course of a delayed-type hypersensitivity reaction in human skin

Diphencyprone (DPCP) is a hapten that induces delayed-type hypersensitivity (DTH) reactions. MicroRNAs (miRNAs) are short non-coding RNAs that negatively regulate gene expression and have been implicated in various inflammatory skin diseases, but their role in DTH reactions is not well understood. We generated global miRNA expression profiles (using next-generation sequencing) of DPCP reactions in skin of seven healthy volunteers at 3, 14 and 120 days after challenge. Compared to placebo-treated sites, DPCP-challenged skin at 3 days (peak inflammation) had 127 miRNAs significantly deregulated. At 14 days (during resolution of inflammation), 43 miRNAs were deregulated and, at 120 days (when inflammation had completely resolved), six miRNAs were upregulated. While some miRNAs have been observed in psoriasis or atopic dermatitis, most of the deregulated miRNAs have not yet been studied in the context of skin biology or immunology. Across the three time points studied, many but not all miRNAs were uniquely expressed. As various miRNAs may influence T cell activation, this may indicate that the miRNAs exclusively expressed at different time points function to promote or resolve skin inflammation, and therefore, may inform on the paradoxical ability of DPCP to treat both autoimmune conditions (alopecia areata) and conditions of ineffective immunity (melanoma).

miR-375 gene dosage in pancreatic β-cells: implications for regulation of β-cell mass and biomarker development

Abstract

MicroRNAs play a crucial role in the regulation of cell growth and differentiation. Mice with genetic deletion of miR-375 exhibit impaired glycemic control due to decreased β-cell and increased α-cell mass and function. The relative importance of these processes for the overall phenotype of miR-375KO mice is unknown. Here, we show that mice overexpressing miR-375 exhibit normal β-cell mass and function. Selective re-expression of miR-375 in β-cells of miR-375KO mice normalizes both, α- and β-cell phenotypes as well as glucose metabolism. Using this model, we also analyzed the contribution of β-cells to the total plasma miR-375 levels. Only a small proportion (≈1 %) of circulating miR-375 originates from β-cells. Furthermore, acute and profound β-cell destruction is sufficient to detect elevations of miR-375 levels in the blood. These findings are supported by higher miR-375 levels in the circulation of type 1 diabetes (T1D) subjects but not mature onset diabetes of the young (MODY) and type 2 diabetes (T2D) patients. Together, our data support an essential role for miR-375 in the maintenance of β-cell mass and provide in vivo evidence for release of miRNAs from pancreatic β-cells. The small contribution of β-cells to total plasma miR-375 levels make this miRNA an unlikely biomarker for β-cell function but suggests a utility for the detection of acute β-cell death for autoimmune diabetes.

Key messages

Overexpression of miR-375 in β-cells does not influence β-cell mass and function.

Increased α-cell mass in miR-375KO arises secondarily to loss of miR-375 in β-cells.

Only a small proportion of circulating miR-375 levels originates from β-cells.

Acute β-cell destruction results in measurable increases of miR-375 in the blood.

Circulating miR-375 levels are not a biomarker for pancreatic β-cell function.

Electronic supplementary material

The online version of this article (doi:10.1007/s00109-015-1296-9) contains supplementary material, which is available to authorized users.

Dysregulation of microRNA-219 promotes neurodegeneration through post-transcriptional regulation of tau

Tau is a highly abundant and multifunctional brain protein that accumulates in neurofibrillary tangles (NFTs), most commonly in Alzheimer's disease (AD) and primary age-related tauopathy. Recently, microRNAs (miRNAs) have been linked to neurodegeneration; however, it is not clear whether miRNA dysregulation contributes to tau neurotoxicity. Here, we determined that the highly conserved brain miRNA miR-219 is downregulated in brain tissue taken at autopsy from patients with AD and from those with severe primary age-related tauopathy. In a Drosophila model that produces human tau, reduction of miR-219 exacerbated tau toxicity, while overexpression of miR-219 partially abrogated toxic effects. Moreover, we observed a bidirectional modulation of tau levels in the Drosophila model that was dependent on miR-219 expression or neutralization, demonstrating that miR-219 regulates tau in vivo. In mammalian cellular models, we found that miR-219 binds directly to the 3'-UTR of the tau mRNA and represses tau synthesis at the post-transcriptional level. Together, our data indicate that silencing of tau by miR-219 is an ancient regulatory mechanism that may become perturbed during neurofibrillary degeneration and suggest that this regulatory pathway may be useful for developing therapeutics for tauopathies.

Multiplexed miRNA fluorescence in situ hybridization for formalin-fixed paraffin-embedded tissues

Multiplexed miRNA fluorescence in situ hybridization (miRNA FISH) is an advanced method for visualizing differentially expressed miRNAs, together with other reference RNAs, in archival tissues. Some miRNAs are excellent disease biomarkers due to their abundance and cell-type specificity. However, these short RNA molecules are difficult to visualize due to loss by diffusion, probe mishybridization, and signal detection and signal amplification issues. Here, we describe a reliable and adjustable method for visualizing and normalizing miRNA signals in formalin-fixed paraffin-embedded (FFPE) tissue sections.

Multicolor microRNA FISH effectively differentiates tumor types

MicroRNAs (miRNAs) are excellent tumor biomarkers because of their cell-type specificity and abundance. However, many miRNA detection methods, such as real-time PCR, obliterate valuable visuospatial information in tissue samples. To enable miRNA visualization in formalin-fixed paraffin-embedded (FFPE) tissues, we developed multicolor miRNA FISH. As a proof of concept, we used this method to differentiate two skin tumors, basal cell carcinoma (BCC) and Merkel cell carcinoma (MCC), with overlapping histologic features but distinct cellular origins. Using sequencing-based miRNA profiling and discriminant analysis, we identified the tumor-specific miRNAs miR-205 and miR-375 in BCC and MCC, respectively. We addressed three major shortcomings in miRNA FISH, identifying optimal conditions for miRNA fixation and ribosomal RNA (rRNA) retention using model compounds and high-pressure liquid chromatography (HPLC) analyses, enhancing signal amplification and detection by increasing probe-hapten linker lengths, and improving probe specificity using shortened probes with minimal rRNA sequence complementarity. We validated our method on 4 BCC and 12 MCC tumors. Amplified miR-205 and miR-375 signals were normalized against directly detectable reference rRNA signals. Tumors were classified using predefined cutoff values, and all were correctly identified in blinded analysis. Our study establishes a reliable miRNA FISH technique for parallel visualization of differentially expressed miRNAs in FFPE tumor tissues.

MicroRNAs miR-26a, miR-26b, and miR-29b accelerate osteogenic differentiation of unrestricted somatic stem cells from human cord blood

BACKGROUND

MicroRNAs are a population of short non-coding RNAs with widespread negative regulatory impact on mRNA translation. Unrestricted somatic stem cells (USSC) are a rare population in human cord blood that can be induced into cells representative of all three germinal layers. Here we analyzed the functional impact of miRNAs on the osteogenic differentiation in USSC.

RESULTS

Gene expression profiling identified 20 microRNAs that were consistently upregulated during osteogenic differentiation of two different USSC cell lines (SA5/73 and SA8/25). Bioinformatic target gene prediction indicated that among these microRNAs, miR-10a, -22, -26a, -26b, and -29b recognize transcripts that encode a set of proteins inhibiting osteogenesis. We subsequently verified osteo-inhibitory CDK6, CTNNBIP1, HDAC4, and TOB1 and osteo-promoting SMAD1 as targets of these microRNAs. In Western blot analyses demonstrated that endogenous levels of CDK6 and HDAC4 were downregulated during osteogenic differentiation of USSC and reduced following ectopic expression of miR-26a/b and miR-29b. In contrast, endogenous expression of SMAD1, targeted by miR-26a/b, was unaltered during osteogenic differentiation of USSC or following ectopic expression of miR-26a/b. Functional overexpression analyses using microRNA mimics revealed that miR-26a/b, as well as miR-29b strongly accelerated osteogenic differentiation of USSC as assessed by Alizarin-Red staining and calcium-release assays.

CONCLUSIONS

miR-26a/b and miR-29b are upregulated during osteogenic differentiation of USSC and share target genes inhibiting osteogenesis. Furthermore, these microRNAs accelerate osteogenic differentiation, likely mediated by osteo-inhibitory proteins such as CDK6 and HDAC4.

Barcoded cDNA library preparation for small RNA profiling by next-generation sequencing

The characterization of post-transcriptional gene regulation by small regulatory (20-30 nt) RNAs, particularly miRNAs and piRNAs, has become a major focus of research in recent years. A prerequisite for characterizing small RNAs is their identification and quantification across different developmental stages, and in normal and disease tissues, as well as model cell lines. Here we present a step-by-step protocol for generating barcoded small RNA cDNA libraries compatible with Illumina HiSeq sequencing, thereby facilitating miRNA and other small RNA profiling of large sample collections.

Bioinformatic analysis of barcoded cDNA libraries for small RNA profiling by next-generation sequencing

The characterization of post-transcriptional gene regulation by small regulatory RNAs of 20-30 nt length, particularly miRNAs and piRNAs, has become a major focus of research in recent years. A prerequisite for the characterization of small RNAs is their identification and quantification across different developmental stages, normal and diseased tissues, as well as model cell lines. Here we present a step-by-step protocol for the bioinformatic analysis of barcoded cDNA libraries for small RNA profiling generated by Illumina sequencing, thereby facilitating miRNA and other small RNA profiling of large sample collections.

RNA-ligase-dependent biases in miRNA representation in deep-sequenced small RNA cDNA libraries

Sequencing of small RNA cDNA libraries is an important tool for the discovery of new RNAs and the analysis of their mutational status as well as expression changes across samples. It requires multiple enzyme-catalyzed steps, including sequential oligonucleotide adapter ligations to the 3' and 5' ends of the small RNAs, reverse transcription (RT), and PCR. We assessed biases in representation of miRNAs relative to their input concentration, using a pool of 770 synthetic miRNAs and 45 calibrator oligoribonucleotides, and tested the influence of Rnl1 and two variants of Rnl2, Rnl2(1-249) and Rnl2(1-249)K227Q, for 3'-adapter ligation. The use of the Rnl2 variants for adapter ligations yielded substantially fewer side products compared with Rnl1; however, the benefits of using Rnl2 remained largely obscured by additional biases in the 5'-adapter ligation step; RT and PCR steps did not have a significant impact on read frequencies. Intramolecular secondary structures of miRNA and/or miRNA/3'-adapter products contributed to these biases, which were highly reproducible under defined experimental conditions. We used the synthetic miRNA cocktail to derive correction factors for approximation of the absolute levels of individual miRNAs in biological samples. Finally, we evaluated the influence of 5'-terminal 5-nt barcode extensions for a set of 20 barcoded 3' adapters and observed similar biases in miRNA read distribution, thereby enabling cost-saving multiplex analysis for large-scale miRNA profiling.

MicroRNAs MiR-17, MiR-20a, and MiR-106b act in concert to modulate E2F activity on cell cycle arrest during neuronal lineage differentiation of USSC

Background

MicroRNAs are short (∼22 nt) non-coding regulatory RNAs that control gene expression at the post-transcriptional level. Here the functional impact of microRNAs on cell cycle arrest during neuronal lineage differentiation of unrestricted somatic stem cells from human cord blood (USSC) was analyzed.

Methodology/Principal Findings

Expression profiling revealed downregulation of microRNAs miR-17, -20a, and -106b in USSC differentiated into neuronal lineage but not in USSC differentiated into osteogenic lineage. Transfection experiments followed by Ki67 immunostainings demonstrated that each of these microRNAs was able to promote proliferation of native USSC and to prevent in part cell cycle arrest during neuronal lineage differentiation of USSC. Bioinformatic target gene predictions followed by experimental target gene validations revealed that miR-17, -20a, and -106b act in a common manner by downregulating an overlapping set of target genes mostly involved in regulation and execution of G₁/S transition. Pro-proliferative target genes cyclinD1 (CCND1) and E2F1 as well as anti-proliferative targets CDKN1A (p21), PTEN, RB1, RBL1 (p107), RBL2 (p130) were shown as common targets for miR-17, -20a, and -106b. Furthermore, these microRNAs also downregulate WEE1 which is involved in G₂/M transition. Most strikingly, miR-17, -20a, and -106b were found to promote cell proliferation by increasing the intracellular activity of E2F transcription factors, despite the fact that miR-17, -20a, and -106b directly target the transcripts that encode for this protein family.

Conclusions/Significance

Mir-17, -20a, and -106b downregulate a common set of pro- and anti-proliferative target genes to impact cell cycle progression of USSC and increase intracellular activity of E2F transcription factors to govern G₁/S transition.

Millennium Development Goal 1: poverty, hunger and decent work in Southeast Asia

This article considers three questions: 1) what progress has been made in achieving MDG1 targets?; 2) what challenges remain?; and 3) what more could and should be done? To examine these questions, the article assesses the progress of Southeast Asia in seeking to achieve MDG1. It argues that the region is 'on track' to achieve MDG 1 targets, although significant challenges such as inequality remain. Economic growth, significant structural change and incorporation into global value chains have contributed to MDG progress. However, this is a double-edged sword as exposure to global economic turbulence can increase. The longer-term reduction of poverty, inequality and social exclusion is a question of empowerment of local producers within value chains-a shift in economic power and control through pro-poor strategies strong enough to effect substantive structural change. The article outlines key concepts; identifies the main characteristics of Southeast Asian poverty; outlines what more needs to be done; and concludes by reprising the article's findings and weighing the prospects for 2010-15 and beyond.

Global distribution of novel rhinovirus genotype

Global surveillance for a novel rhinovirus genotype indicated its association with community outbreaks and pediatric respiratory disease in Africa, Asia, Australia, Europe, and North America. Molecular dating indicates that these viruses have been circulating for at least 250 years.

Family clustering of Viliuisk encephalomyelitis in traditional and new geographic regions

Transmission occurs through patient contact; human migration from disease-endemic villages leads to disease emergence in new communities.

Viliuisk encephalomyelitis is an acute, often fatal, meningoencephalitis that tends to develop into a prolonged chronically progressive panencephalitis. Clinical, neuropathologic, and epidemiologic data argue for an infectious cause, although multiple attempts at pathogen isolation have been unsuccessful. To assess mechanisms of disease transmission and spread, we studied 6 multiplex families. Secondary cases occurred among genetically related and unrelated persons in a setting of prolonged intrahousehold contact with a patient manifesting the disease. Transmission to unrelated persons was documented in a densely populated region around the city of Yakutsk in which Viliuisk encephalomyelitis had not been previously known. Initially identified in a small Yakut-Evenk population on the Viliui River of eastern Siberia, the disease subsequently spread through human contacts to new geographic areas, thus characterizing Viliuisk encephalomyelitis as an emerging infectious disease.

A recently identified rhinovirus genotype is associated with severe respiratory-tract infection in children in Germany

Acute respiratory infection is a significant cause of morbidity and mortality in children worldwide. Accurate identification of causative agents is critical to case management and to prioritization in vaccine development. Sensitive multiplex diagnostics provide us with an opportunity to investigate the relative contributions of individual agents and may also facilitate the discovery of new pathogens. Recently, application of MassTag polymerase chain reaction (PCR) to undiagnosed influenza-like illness in New York State led to the discovery of a novel rhinovirus genotype. Here we report the investigation, by MassTag PCR, of pediatric respiratory-tract infections in Germany, studying 97 cases for which no pathogen was identified through routine laboratory evaluation. Respiratory viruses were identified in 49 cases (51%); of the 55 identified viruses, 41 (75%) were rhinoviruses. The novel genotype represented 73% of rhinoviruses and 55% of all identified viruses. Infections with the novel genotype were associated with upper-respiratory-tract symptoms but, more frequently, with bronchitis, bronchiolitis, and pneumonia.

Detection of respiratory viruses and subtype identification of influenza A viruses by GreeneChipResp oligonucleotide microarray

Acute respiratory infections are significant causes of morbidity, mortality, and economic burden worldwide. An accurate, early differential diagnosis may alter individual clinical management as well as facilitate the recognition of outbreaks that have implications for public health. Here we report on the establishment and validation of a comprehensive and sensitive microarray system for detection of respiratory viruses and subtyping of influenza viruses in clinical materials. Implementation of a set of influenza virus enrichment primers facilitated subtyping of influenza A viruses through the differential recognition of hemagglutinins 1 through 16 and neuraminidases 1 through 9. Twenty-one different respiratory virus species were accurately characterized, including a recently identified novel genetic clade of rhinovirus.

Panmicrobial oligonucleotide array for diagnosis of infectious diseases

To facilitate rapid, unbiased, differential diagnosis of infectious diseases, we designed GreeneChipPm, a panmicrobial microarray comprising 29,455 sixty-mer oligonucleotide probes for vertebrate viruses, bacteria, fungi, and parasites. Methods for nucleic acid preparation, random primed PCR amplification, and labeling were optimized to allow the sensitivity required for application with nucleic acid extracted from clinical materials and cultured isolates. Analysis of nasopharyngeal aspirates, blood, urine, and tissue from persons with various infectious diseases confirmed the presence of viruses and bacteria identified by other methods, and implicated Plasmodium falciparum in an unexplained fatal case of hemorrhagic feverlike disease during the Marburg hemorrhagic fever outbreak in Angola in 2004-2005.

Greene SCPrimer: a rapid comprehensive tool for designing degenerate primers from multiple sequence alignments

Polymerase chain reaction (PCR) is widely applied in clinical and environmental microbiology. Primer design is key to the development of successful assays and is often performed manually by using multiple nucleic acid alignments. Few public software tools exist that allow comprehensive design of degenerate primers for large groups of related targets based on complex multiple sequence alignments. Here we present a method for designing such primers based on tree building followed by application of a set covering algorithm, and demonstrate its utility in compiling Multiplex PCR primer panels for detection and differentiation of viral pathogens.

MassTag polymerase-chain-reaction detection of respiratory pathogens, including a new rhinovirus genotype, that caused influenza-like illness in New York State during 2004-2005

In New York State during winter 2004, there was a high incidence of influenza-like illness that tested negative both for influenza virus, by molecular methods, and for other respiratory viruses, by virus culture. Concern that a novel pathogen might be implicated led us to implement a new multiplex diagnostic tool. MassTag polymerase chain reaction resolved 26 of 79 previously negative samples, revealing the presence of rhinoviruses in a large proportion of samples, half of which belonged to a previously uncharacterized genetic clade. In some instances, knowledge of the detected viral and/or bacterial (co)infection could have altered clinical management

MassTag polymerase chain reaction for differential diagnosis of viral hemorrhagic fever

Viral hemorrhagic fevers are associated with high rates of illness and death. Although therapeutic options are limited, early differential diagnosis has implications for containment and may aid in clinical management. We describe a diagnostic system for rapid, multiplex polymerase chain reaction identification of 10 different causes of viral hemorrhagic fevers.

Molecular identification of mumps virus genotypes from clinical samples: standardized method of analysis

A sensitive nested reverse transcription-PCR assay, targeting a short fragment of the gene encoding the small hydrophobic protein (SH gene), was developed to allow rapid characterization of mumps virus in clinical samples. The sensitivity and specificity of the assay were established using representative genotypes A, B, C, D, E, and F. Mumps virus RNA was characterized directly from cerebrospinal fluid (CSF) samples and in extracts of mumps virus isolates from patients with various clinical syndromes. Direct sequencing of products and subsequent phylogenetic analysis enabled genetic classification. A simple web-based system of sequence analysis was established. The study also allowed characterization of mumps virus strains from Argentina as part of a new subgenotype. This PCR assay for characterization of mumps infections coupled to a web-based analytical program provides a rapid method for identification of known and novel strains.

Diagnostic system for rapid and sensitive differential detection of pathogens

Naturally emerging and deliberately released pathogens demand new detection strategies to allow early recognition and containment. We describe a diagnostic system for rapid, sensitive, multiplex discrimination of microbial gene sequences and report its application for detecting 22 respiratory pathogens in clinical samples.

Severe acute respiratory syndrome coronavirus persistence in Vero cells

BACKGROUND

Several coronaviruses establish persistent infections in vitro and in vivo, however it is unknown whether persistence is a feature of the severe acute respiratory syndrome coronavirus (SARS-CoV) life cycle. This study was conducted to investigate viral persistence.

METHODS

We inoculated confluent monolayers of Vero cells with SARS-CoV at a multiplicity of infection of 0.1 TCID50 and passaged the remaining cells every 4 to 8 days for a total of 11 passages. Virus was titrated at each passage by limited dilution assay and nucleocapsid antigen was detected by Western blot and immunofluoresence assays. The presence of viral particles in passage 11 cells was assessed by electron microscopy. Changes in viral genomic sequences during persistent infection were examined by DNA sequencing.

RESULTS

Cytopathic effect was extensive after initial inoculation but diminished with serial passages. Infectious virus was detected after each passage and viral growth curves were identical for parental virus stock and virus obtained from passage 11 cells. Nucleocapsid antigen was detected in the majority of cells after initial inoculation but in only 10%-40% of cells at passages 2-11. Electron microscopy confirmed the presence of viral particles in passage 11 cells. Sequence analysis at passage 11 revealed fixed mutations in the spike (S) gene and ORFs 7a-8b but not in the nucleocapsid (N) gene.

CONCLUSIONS

SARS-CoV can establish a persistent infection in vitro. The mechanism for viral persistence is consistent with the formation of a carrier culture whereby a limited number of cells are infected with each round of virus replication and release. Persistence is associated with selected mutations in the SARS-CoV genome. This model may provide insight into SARS-related lung pathology and mechanisms by which humans and animals can serve as reservoirs for infection.

Genetic similarities between Spitz nevus and Spitzoid melanoma in children

BACKGROUND

Melanoma in children is rare. Diagnosis of the subtype of melanoma known as Spitzoid melanoma can be extremely challenging in this age group. Spitzoid melanoma clinically and histopathologically resembles a benign melanocytic proliferation referred to as Spitz nevus. In some cases, distinction between the two is impossible. Initial misdiagnoses of Spitzoid melanomas as Spitz nevi, thus leading to fatal outcomes, have occurred. The genetic basis and biologic behavior of Spitzoid melanoma is unknown. Although melanoma specimens exhibit high rates of mutation in the B-RAF and N-RAS genes, the Spitzoid melanoma subtype has not been evaluated. Spitz nevi have been found to be associated with a low percentage of mutations in the H-RAS gene; however, the mutational profile of H-RAS in Spitzoid melanoma is unknown.

METHODS

The authors evaluated a unique series of melanomas occurring in prepubescent children that showed Spitz nevus-like histopathology (Spitzoid melanoma). All of the melanomas in the current series have metastasized to lymph nodes, confirming the diagnosis of melanoma. The authors examined these tumors, as well as age-matched Spitz nevi, for mutations in the B-RAF, N-RAS, and H-RAS genes.

RESULTS

Activating hotspot mutations in the B-RAF, N-RAS, and H-RAS genes were not identified in Spitzoid melanoma or Spitz nevus specimens.

CONCLUSIONS

There are genetic similarities with respect to the B-RAF, N-RAS, and H-RAS genes between Spitzoid melanoma and Spitz nevi. Such similarities further differentiate these two tumor types from other melanoma subtypes and from melanocytic nevi, respectively. However, mutation analysis of B-RAF, N-RAS, and H-RAS was not useful in differentiating between Spitzoid melanoma and Spitz nevus in children.

Vascular endothelial growth factor levels in serum do not increase following HIV type 1 and HHV8 seroconversion and lack correlation with AIDS-related Kaposi's sarcoma

The utility of vascular endothelial growth factor (VEGF) concentrations in serum as a predictive marker for AIDS-KS was studied. Sera were obtained from 40 homosexual men who seroconverted for HIV-1 and HHV8 prior to or during their participation in the Amsterdam Cohort Studies (1984-2000). We designed an ELISA to detect VEGF and measured VEGF prior to either infection, at HIV-1 and HHV8 seroconversion, after both infections, at AIDS-KS diagnosis (n = 11), and in the most recently available serum sample. The geometric mean serum VEGF concentration was 81.5 pg/ml in those with AIDS-KS and 80.4 pg/ml in those with AIDS but without KS. Median serum VEGF concentrations did not differ between the time points described above. Higher VEGF concentrations in serum were observed at higher CD4(+) cell counts. Serum concentrations of VEGF were not influenced by HIV-1 or HHV8 infection or by the conditions leading to AIDS-KS. Sequential measurement of VEGF in serum is not expected to contribute to the prediction or therapeutic monitoring of AIDS-KS.

Risk factors for human herpesvirus 8 infection in a cohort of drug users in the Netherlands, 1985-1996

To elucidate the mode of human herpesvirus 8 (HHV-8) transmission in a population of Amsterdam drug users, HHV-8 seroprevalence and seroincidence were determined in 1179 drug users in the Amsterdam Cohort Studies (1985-1996). Risk factors for HHV-8 infection were examined. Serum samples were screened with an enzyme immunoassay by using HHV-8 lytic capsid (open-reading frame [ORF] 65) and latent nuclear (ORF73) antigens; positive results were confirmed by Western blot and immunofluorescence assay. Seroprevalence (men, 3.4%; women, 1.4%) and seroincidence (men, 0.08; women, 0.05/100 person-years) were low in this study. Infections with human immunodeficiency virus (HIV) type 1, hepatitis B virus (HBV), and hepatitis C virus (HCV), but not HHV-8, were associated with injection drug use (IDU). Independent risk factors for HHV-8 seropositivity were homosexual contacts and Mediterranean nationality for men and sexual contact with bisexual men, absence of a steady partner, and unprotected commercial sex for women. Unlike HIV-1, HBV, or HCV infection, HHV-8 infection is uncommon in Amsterdam drug users, as is HHV-8 transmission through IDU.

Kaposi's sarcoma and human herpesvirus 8 infection do not protect HIV-1 infected homosexual men from AIDS dementia complex

OBJECTIVE

To examine the association between Kaposi's sarcoma (KS), human herpes virus 8 (HHV8) and AIDS dementia complex (ADC).

DESIGN

A total of 599 HIV-1 infected homosexual men participated in a prospective cohort study (Amsterdam, 1984-1996).

METHODS

The risk for ADC in patients with prior KS or HHV8 infection was estimated using the Cox proportional hazards method with adjustments for antiretroviral medication and low CD4 cell counts.

RESULTS

Of the 599 participants, 290 (48.4%) had HHV8 antibodies, 99 (16.5%) had KS and 30 (5.0%) had ADC. ADC was diagnosed in 5.2% of participants with KS and 5.0% of those without KS, and in 4.8% of HHV8 seropositive compared to 5.2% seronegative individuals and thus was not associated with KS or HHV8 infection. Using a time-dependent Cox proportional hazards analysis with the date of KS as risk factor, the risk for ADC was 2.7 [95% confidence interval (CI), 0.92-7.96; P = 0.07) and when only definite ADC was considered it was 3.5 (95% CI, 1.00-12.26;P = 0.05). After adjusting for decreases in CD4 cell count and use of medication, the hazards ratio for participants with KS to develop ADC was 2.0 (95% CI, 0.66-5.77; P = 0.23) and 2.6 (95% CI, 0.73-9.12; P = 0.14), respectively. HHV8 seropositivity, adjusted for the same variables, showed a risk for ADC of 0.85 (95% CI, 0.41-1.77;P = 0.66) and for definite ADC 0.69 (95% CI, 0.27-1.73; P = 0.42). The expected neuroprotective effects of antiretroviral medication were observed.

CONCLUSIONS

KS or HHV8 does not significantly influence the risk for developing ADC in a group with a uniform risk for developing KS therefore we recommend caution in searching for a KS-associated or HHV8-derived therapy for ADC.

High prevalence of Epstein-Barr virus type 2 among homosexual men is caused by sexual transmission

To investigate whether Epstein-Barr virus (EBV) type 2 infection is highly prevalent among homosexual men, the prevalence of EBV type 2 was studied among homosexual and heterosexual white men who were at high and low risk for sexually transmitted diseases; these data were correlated with sexual behavior. The prevalence of EBV type 2 among homosexual men was significantly higher than it was among heterosexual men (39% vs. 6%). Among high-risk heterosexual men, prevalence was significantly higher than it was among low-risk heterosexual men (15% vs. 0). In univariate analyses, EBV type 2 infection in homosexual men was significantly associated with human immunodeficiency virus (HIV) seropositivity, increased numbers of intercourse partners, non-Dutch nationality, and human herpesvirus 8 seropositivity. In multivariate analyses, an independent association with EBV type 2 was observed only for HIV seropositivity and number of sex partners. These data support the conclusion that EBV type 2 infection is more prevalent among white homosexual men and is caused by sexual transmission.

Human herpesvirus 8 infections in the Amsterdam Cohort Studies (1984-1997): analysis of seroconversions to ORF65 and ORF73

We have shown previously that human herpesvirus 8 (HHV8) seroconversion for antibodies to the latency-associated nuclear antigen encoded by ORF73 and/or the lytic capsid antigen (vp19) encoded by ORF65 is associated with orogenital contact and is strongly linked to the development of Kaposi's sarcoma among HIV-infected individuals in the Amsterdam Cohort Studies. Here, we investigate the relationship between seroconversion to these antigens and primary HHV8 infection. Between 1984 and 1997, 215 HHV8 seroconversions to ORF73 (106 cases or 49%) and/or to ORF65 (159 cases or 74%) were recorded in the cohort of homosexual men. The HHV8 seroconversion rate among HIV-infected homosexual men (6.2 per 100 person years) was consistently higher than among HIV-uninfected men (2.6 per 100 person years). In HIV-infected but not in uninfected individuals, seroconversion to ORF73/latency-associated nuclear antigen precedes that to ORF65/vp19. Antibody levels to both ORF65- and ORF73-encoded antigens were higher in HIV-infected than in HIV-uninfected men, and among HIV-seropositives, antibody levels to ORF65/vp19 rise even higher with declining CD4 cell counts and peak with Kaposi's sarcoma development, suggesting continuing and increasing viral replication. In 10.3% of HHV8 seroconversions, transient serum viremia could be demonstrated before or at seroconversion. Together with the previously reported link between unprotected orogenital sex and HHV8 seroconversion, our observations suggest that HHV8 seroconversions result from primary infections.

Risk factors for human herpesvirus 8 seropositivity and seroconversion in a cohort of homosexual men

Sexual and nonsexual modes of transmission of human herpesvirus 8 (HHV8) have been suggested, but specific routes remain unclear. Therefore, the objective of this study was to assess risk factors for HHV8 seropositivity and determine specific sexual practices associated with HHV8 seroconversion. Sera from 1,458 homosexual men (Amsterdam Cohort Study, 1984-1996) were tested for antibodies to HHV8 with a modified version of an enzyme immunoassay, using recombinant HHV8 lytic phase capsid (ORF65) and latent phase nuclear (ORF73) proteins. HHV8 seroprevalence at study entry was 20.9% (305/1,458); was highest among those with positive human immunodeficiency virus (HIV) status, no steady partner, and southern European or Latin American nationality; and increased with older age and higher number of sexual partners. During follow-up, 215 men seroconverted for HHV8 (incidence: 3.6/100 person-years). Both prevalence and incidence rates remained more or less stable during the study period. Orogenital insertive sex (odds ratio (OR) = 5.95; 95% confidence interval (CI): 2.88, 12.29) or orogenital receptive sex (OR = 4.29; 95% CI: 2.11, 8.71) with more than five partners in the past 6 months, older age (OR = 2.89; 95% CI: 1.13, 7.34, when older than 45 years), and preceding HIV infection (OR = 2.47; 95% CI: 1.53, 3.99) were independent predictors for HHV8 seroconversion. The authors found strong evidence for orogenital transmission of HHV8 among homosexual men.

Favourable effect of chemotherapy on clinical symptoms and human herpesvirus-8 DNA load in a patient with Kaposi's sarcoma presenting with fever and anemia

The case of a patient infected with human immunodeficiency virus type 1 (HIV-1) with Kaposi's sarcoma who presented with fever of unknown origin, severe anemia, thrombocytopenia and hypoalbuminemia but only limited involvement of the skin is presented. Chemotherapy directed at Kaposi's sarcoma resulted in resolution of these clinical signs and symptoms and was associated with a significant reduction in human herpesvirus-8 DNA load in serum, despite continued HIV-1 replication. Such a decreasing human herpesvirus-8 load following Kaposi's sarcoma-directed chemotherapy has not been reported previously. These findings suggest that Kaposi's sarcoma was indeed responsible for the clinical syndrome and that this neoplasm is a source of human herpesvirus-8 virus particle production, which can be inhibited by chemotherapy-induced reduction in tumor burden.

Seroconversion for human herpesvirus 8 during HIV infection is highly predictive of Kaposi's sarcoma

BACKGROUND

The finding of antibodies against human herpesvirus 8 (HHV-8) is associated with the occurrence of Kaposi's sarcoma in persons infected with HIV. However, the predictive value of HHV-8 antibodies for Kaposi's sarcoma in HIV infection is unknown.

METHODS

The Amsterdam Cohort Studies on HIV infection and AIDS started in 1984 for homosexual men and in 1985 for injecting drug users. Serum samples from 1459 homosexual men and 1167 drug users were tested for antibodies to recombinant HHV-8 lytic-phase capsid (ORF65) antigen and latent-phase nuclear (ORF73) antigen. Individuals were retrospectively identified as HHV-8-positive or HHV-8-negative at enrolment or HHV-8 seroconverter during the study. Kaposi's sarcoma-free survival time was compared between HIV-infected men who were positive for HHV-8 at enrolment and those who later seroconverted for HHV-8. Hazard ratios were estimated for Kaposi's sarcoma, lymphoma, and opportunistic infection according to the HHV-8 serostatus.

RESULTS

The incidence of HHV-8 seroconversion among drugs users was 0.7 per 100 person-years based on 31 seroconversions, whereas an incidence of 3.6 was found among homosexual men based on 215 seroconversions. The hazard ratio for Kaposi's sarcoma was 3.15 (95% CI: 1.89-5.25) in HIV-infected individuals if HHV-8 antibodies were present either at enrolment or at HIV seroconversion. In HIV-infected persons who later seroconverted to HHV-8, Kaposi's sarcoma developed more rapidly: hazard ratio of 5.04 (95% CI: 2.94-8.64), an additional risk of 1.60 (95% CI: 1.01-2.53; P = 0.04). Time-dependent adjustment for CD4+ cell count and HIV RNA had no impact on the additional risk, although the CD4+ cell count was an independent risk factor for Kaposi's sarcoma. HHV-8 infection did not increase the risk of AIDS-related lymphoma or opportunistic infections.

CONCLUSIONS

The incidence of HHV-8 infection is higher in homosexual men than in drug users. The presence of HHV-8 antibodies in HIV-infected persons increases the risk of Kaposi's sarcoma. Among HIV-infected persons, those who subsequently seroconvert for HHV-8 are at highest risk. These results strongly confirm the causal role of HHV-8 in Kaposi's sarcoma and emphasize the clinical relevance of HHV-8 seroconversion before and after the HIV infection.