Contamination source modeling with SCRuB improves cancer phenotype prediction from microbiome data

Association between semen microbiome disorder and sperm DNA damage

DNA fragmentation index (DFI), a new biomarker to diagnose male infertility, is closely associated with poor reproductive outcomes. Previous research reported that seminal microbiome correlated with sperm DNA integrity, suggesting that the microbiome may be one of the causes of DNA damage in sperm. However, it has not been elucidated how the microbiota exerts their effects. Here, we used a combination of 16S rRNA sequencing and untargeted metabolomics techniques to investigate the role of microbiota in high sperm DNA fragmentation index (HDFI). We report that increased specific microbial profiles contribute to high sperm DNA fragmentation, thus implicating the seminal microbiome as a new therapeutic target for HDFI patients. Additionally, we found that the amount of Lactobacillus species was altered: Lactobacillus iners was enriched in HDFI patients, shedding light on the potential influence of L. iners on male reproductive health. Finally, we also identified enrichment of the acetyl-CoA fermentation to butanoate II and purine nucleobase degradation I in the high sperm DNA fragmentation samples, suggesting that butanoate may be the target metabolite of sperm DNA damage. These findings provide valuable insights into the complex interplay between microbiota and sperm quality in HDFI patients, laying the foundation for further research and potential clinical interventions.IMPORTANCEThe DNA fragmentation index (DFI) is a measure of sperm DNA fragmentation. Because high sperm DNA fragmentation index (HDFI) has been strongly associated with adverse reproductive outcomes, this has been linked to the seminal microbiome. Because the number of current treatments for HDFI is limited and most of them have no clear efficacy, it is critical to understand how semen microbiome exerts their effects on sperm DNA. Here, we evaluated the semen microbiome and its metabolites in patients with high and low sperm DNA fragmentation. We found that increased specific microbial profiles contribute to high sperm DNA fragmentation. In particular, Lactobacillus iners was uniquely correlated with high sperm DNA fragmentation. Additionally, butanoate may be the target metabolite produced by the microbiome to damage sperm DNA. Our findings support the interaction between semen microbiome and sperm DNA damage and suggest that seminal microbiome should be a new therapeutic target for HDFI patients.

The impact of kit, environment, and sampling contamination on the observed microbiome of bovine milk

In low-microbial biomass samples such as bovine milk, contaminants can outnumber endogenous bacteria. Because of this, milk microbiome research suffers from a critical knowledge gap, namely, does non-mastitis bovine milk contain a native microbiome? In this study, we sampled external and internal mammary epithelia and stripped and cisternal milk and used numerous negative controls, including air and sampling controls and extraction and library preparation blanks, to identify the potential sources of contamination. Two algorithms were used to mathematically remove contaminants and track the potential movement of microbes among samples. Results suggest that the majority (i.e., >75%) of sequence data generated from bovine milk and mammary epithelium samples represents contaminating DNA. Contaminants in milk samples were primarily sourced from DNA extraction kits and the internal and external skin of the teat, while teat canal and apex samples were mainly contaminated during the sampling process. After decontamination, the milk microbiome displayed a more dispersed, less diverse, and compositionally distinct bacterial profile compared with epithelial samples. Similar microbial compositions were observed between cisternal and stripped milk samples, as well as between teat apex and canal samples. Staphylococcus and Acinetobacter were the predominant genera detected in milk sample sequences, and bacterial culture showed growth of Staphylococcus and Corynebacterium spp. in 50% (7/14) of stripped milk samples and growth of Staphylococcus spp. in 7% (1/14) of cisternal milk samples. Our study suggests that microbiome data generated from milk samples obtained from clinically healthy bovine udders may be heavily biased by contaminants that enter the sample during sample collection and processing workflows.IMPORTANCEObtaining a non-contaminated sample of bovine milk is challenging due to the nature of the sampling environment and the route by which milk is typically extracted from the mammary gland. Furthermore, the very low bacterial biomass of bovine milk exacerbates the impacts of contaminant sequences in downstream analyses, which can lead to severe biases. Our finding showed that bovine milk contains very low bacterial biomass and each contamination event (including sampling procedure and DNA extraction process) introduces bacteria and/or DNA fragments that easily outnumber the native bacterial cells. This finding has important implications for our ability to draw robust conclusions from milk microbiome data, especially if the data have not been subjected to rigorous decontamination procedures. Based on these findings, we strongly urge researchers to include numerous negative controls into their sampling and sample processing workflows and to utilize several complementary methods for identifying potential contaminants within the resulting sequence data. These measures will improve the accuracy, reliability, reproducibility, and interpretability of milk microbiome data and research.

Robustness of cancer microbiome signals over a broad range of methodological variation

In 2020, we identified cancer-specific microbial signals in The Cancer Genome Atlas (TCGA) [1]. Multiple peer-reviewed papers independently verified or extended our findings [2-12]. Given this impact, we carefully considered concerns by Gihawi et al. [13] that batch correction and database contamination with host sequences artificially created the appearance of cancer type-specific microbiomes. (1) We tested batch correction by comparing raw and Voom-SNM-corrected data per-batch, finding predictive equivalence and significantly similar features. We found consistent results with a modern microbiome-specific method (ConQuR [14]), and when restricting to taxa found in an independent, highly-decontaminated cohort. (2) Using Conterminator [15], we found low levels of human contamination in our original databases (~1% of genomes). We demonstrated that the increased detection of human reads in Gihawi et al. [13] was due to using a newer human genome reference. (3) We developed Exhaustive, a method twice as sensitive as Conterminator, to clean RefSeq. We comprehensively host-deplete TCGA with many human (pan)genome references. We repeated all analyses with this and the Gihawi et al. [13] pipeline, and found cancer type-specific microbiomes. These extensive re-analyses and updated methods validate our original conclusion that cancer type-specific microbial signatures exist in TCGA, and show they are robust to methodology.

The prevalence of controls in phyllosphere microbiome research: a methodological review

DNA contamination can critically confound microbiome studies. Here, we take a systematic approach to review the current literature and investigate the prevalence of contamination controls in phyllosphere microbiome research over the past decade. By utilising systematic review principles for this review, we were able to conduct a thorough investigation, screening 450 articles from three databases for eligibility and extracting data in a controlled and methodical manner. Worryingly, we observed a surprisingly low usage of both positive and negative contamination controls in phyllosphere research. As a result, we propose a set of minimum standards to combat the effects of contamination in future phyllosphere research.

Processing-bias correction with DEBIAS-M improves cross-study generalization of microbiome-based prediction models

Every step in common microbiome profiling protocols has variable efficiency for each microbe. For example, different DNA extraction kits may have different efficiency for Gram-positive and -negative bacteria. These variable efficiencies, combined with technical variation, create strong processing biases, which impede the identification of signals that are reproducible across studies and the development of generalizable and biologically interpretable prediction models. "Batch-correction" methods have been used to alleviate these issues computationally with some success. However, many make strong parametric assumptions which do not necessarily apply to microbiome data or processing biases, or require the use of an outcome variable, which risks overfitting. Lastly and importantly, existing transformations used to correct microbiome data are largely non-interpretable, and could, for example, introduce values to features that were initially mostly zeros. Altogether, processing bias currently compromises our ability to glean robust and generalizable biological insights from microbiome data. Here, we present DEBIAS-M (Domain adaptation with phenotype Estimation and Batch Integration Across Studies of the Microbiome), an interpretable framework for inference and correction of processing bias, which facilitates domain adaptation in microbiome studies. DEBIAS-M learns bias-correction factors for each microbe in each batch that simultaneously minimize batch effects and maximize cross-study associations with phenotypes. Using benchmarks of HIV and colorectal cancer classification from gut microbiome data, and cervical neoplasia prediction from cervical microbiome data, we demonstrate that DEBIAS-M outperforms batch-correction methods commonly used in the field. Notably, we show that the inferred bias-correction factors are stable, interpretable, and strongly associated with specific experimental protocols. Overall, we show that DEBIAS-M allows for better modeling of microbiome data and identification of interpretable signals that are reproducible across studies.

Broadening oncological boundaries: the intratumoral microbiota

The microbiota of solid tumors was identified >100 years ago; however, heterogeneous composition and diversity have been revealed only recently. Growing evidence has suggested that several functional mechanisms of the intratumoral microbiota affect tumorigenesis and progression, suggesting that the intratumoral microbiota is a promising biomarker for multiple cancers. The low biomass of the intratumoral microbiota poses a major challenge to related research, thus necessitating the use of a multiple-modality integrated framework to resolve this dilemma. Advanced techniques such as single-cell sequencing provide significant clues, and the gradual optimization of functional experiments and culture-based methods enables deeper investigation of the underlying mechanisms involved. In this review, we outline the current state of research on the intratumoral microbiota and describe the challenges and comprehensive strategies for future research.

Melanoma and microbiota: Current understanding and future directions

Over the last decade, the composition of the gut microbiota has been found to correlate with the outcomes of cancer patients treated with immunotherapy. Accumulating evidence points to the various mechanisms by which intestinal bacteria act on distal tumors and how to harness this complex ecosystem to circumvent primary resistance to immune checkpoint inhibitors. Here, we review the state of the microbiota field in the context of melanoma, the recent breakthroughs in defining microbial modes of action, and how to modulate the microbiota to enhance response to cancer immunotherapy. The host-microbe interaction may be deciphered by the use of "omics" technologies, and will guide patient stratification and the development of microbiota-centered interventions. Efforts needed to advance the field and current gaps of knowledge are also discussed.

The fungal intestinal microbiota predict the development of bronchopulmonary dysplasia in very low birthweight newborns

RATIONALE

Bronchopulmonary dysplasia (BPD) is the most common morbidity affecting very preterm infants. Gut fungal and bacterial microbial communities contribute to multiple lung diseases and may influence BPD pathogenesis.

METHODS

We performed a prospective, observational cohort study comparing the multikingdom fecal microbiota of 144 preterm infants with or without moderate to severe BPD by sequencing the bacterial 16S and fungal ITS2 ribosomal RNA gene. To address the potential causative relationship between gut dysbiosis and BPD, we used fecal microbiota transplant in an antibiotic-pseudohumanized mouse model. Comparisons were made using RNA sequencing, confocal microscopy, lung morphometry, and oscillometry.

RESULTS

We analyzed 102 fecal microbiome samples collected during the second week of life. Infants who later developed BPD showed an obvious fungal dysbiosis as compared to infants without BPD (NoBPD, p = 0.0398, permutational multivariate ANOVA). Instead of fungal communities dominated by Candida and Saccharomyces, the microbiota of infants who developed BPD were characterized by a greater diversity of rarer fungi in less interconnected community architectures. On successful colonization, the gut microbiota from infants with BPD augmented lung injury in the offspring of recipient animals. We identified alterations in the murine intestinal microbiome and transcriptome associated with augmented lung injury.

CONCLUSIONS

The gut fungal microbiome of infants who will develop BPD is dysbiotic and may contribute to disease pathogenesis.

Leveraging circulating microbial DNA for early cancer detection

Microbial cell-free DNA (microbial cfDNA) offers a minimally invasive approach for profiling the microbiome. Despite technical and biological challenges, the potential of microbial cfDNA for cancer detection is increasingly being evaluated using deep sequencing, novel laboratory approaches, and computational methods. Targeting the microbiome using liquid biopsies could improve early cancer detection.

Intracellular bacteria in cancer-prospects and debates

Recent evidence suggests that some human cancers may harbor low-biomass microbial ecosystems, spanning bacteria, viruses, and fungi. Bacteria, the most-studied kingdom in this context, are suggested by these studies to localize within cancer cells, immune cells and other tumor microenvironment cell types, where they are postulated to impact multiple cancer-related functions. Herein, we provide an overview of intratumoral bacteria, while focusing on intracellular bacteria, their suggested molecular activities, communication networks, host invasion and evasion strategies, and long-term colonization capacity. We highlight how the integration of sequencing-based and spatial techniques may enable the recognition of bacterial tumor niches. We discuss pitfalls, debates and challenges in decisively proving the existence and function of intratumoral microbes, while reaching a mechanistic elucidation of their impacts on tumor behavior and treatment responses. Together, a causative understanding of possible roles played by intracellular bacteria in cancer may enable their future utilization in diagnosis, patient stratification, and treatment.

Microbial signatures of neonatal bacterial meningitis from multiple body sites

As a common central nervous system infection in newborns, neonatal bacterial meningitis (NBM) can seriously affect their health and growth. However, although metagenomic approaches are being applied in clinical diagnostic practice, there are some limitations for whole metagenome sequencing and amplicon sequencing in handling low microbial biomass samples. Through a newly developed ultra-sensitive metagenomic sequencing method named 2bRAD-M, we investigated the microbial signatures of central nervous system infections in neonates admitted to the neonatal intensive care unit. Particularly, we recruited a total of 23 neonates suspected of having NBM and collected their blood, cerebrospinal fluid, and skin samples for 2bRAD-M sequencing. Then we developed a novel decontamination method (Reads Level Decontamination, RLD) for 2bRAD-M by which we efficiently denoised the sequencing data and found some potential biomarkers that have significantly different relative abundance between 12 patients that were diagnosed as NBM and 11 Non-NBM based on their cerebrospinal fluid (CSF) examination results. Specifically, we discovered 11 and 8 potential biomarkers for NBM in blood and CSF separately and further identified 16 and 35 microbial species that highly correlated with the physiological indicators in blood and CSF. Our study not only provide microbiological evidence to aid in the diagnosis of NBM but also demonstrated the application of an ultra-sensitive metagenomic sequencing method in pathogenesis study.