Metabolic products of the intestinal microbiome and extremes of atherosclerosis.

BACKGROUND AND AIMS: There is increasing awareness that the intestinal microbiome plays an important role in human health. We investigated its role in the burden of carotid atherosclerosis, measured by ultrasound as total plaque area. METHODS: Multiple regression with traditional risk factors was used to identify three phenotypes among 316/3056 patients attending vascular prevention clinics. Residual score (RES; i.e. the distance off the regression line, similar to standard deviation) was used to identify the 5% of patients with much less plaque than predicted by their risk factors (Protected, RES <-2), the 90% with about as much plaque as predicted (Explained, RES -2 to 2), and the 5% with much more plaque than predicted (Unexplained RES >2). Metabolic products of the intestinal microbiome that accumulate in renal failure - gut-derived uremic toxins (GDUT) - were assayed in plasma by ultra-performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry. RESULTS: Plasma levels of trimethylamine n-oxide (TMAO), p-cresyl sulfate, p-cresyl glucuronide, and phenylacetylglutamine were significantly lower among patients with the Protected phenotype, and higher in those with the Unexplained phenotype, despite no significant differences in renal function or in dietary intake of nutrient precursors of GDUT. In linear multiple regression with a broad panel of risk factors, TMAO (p = 0.011) and p-cresyl sulfate (p = 0.011) were significant independent predictors of carotid plaque burden. CONCLUSIONS: The intestinal microbiome appears to play an important role in atherosclerosis. These findings raise the possibility of novel approaches to treatment of atherosclerosis such as fecal transplantation and probiotics.

Expanding the UniFrac Toolbox.

The UniFrac distance metric is often used to separate groups in microbiome analysis, but requires a constant sequencing depth to work properly. Here we demonstrate that unweighted UniFrac is highly sensitive to rarefaction instance and to sequencing depth in uniform data sets with no clear structure or separation between groups. We show that this arises because of subcompositional effects. We introduce information UniFrac and ratio UniFrac, two new weightings that are not as sensitive to rarefaction and allow greater separation of outliers than classic unweighted and weighted UniFrac. With this expansion of the UniFrac toolbox, we hope to empower researchers to extract more varied information from their data.

Ras Signaling Is a Key Determinant for Metastatic Dissemination and Poor Survival of Luminal Breast Cancer Patients.

Breast cancer is associated with alterations in a number of growth factor and hormone-regulated signaling pathways. Mouse models of metastatic breast cancer typically feature mutated oncoproteins that activate PI3K, Stat3, and Ras signaling, but the individual and combined roles of these pathways in breast cancer progression are poorly understood. In this study, we examined the relationship between oncogenic pathway activation and breast cancer subtype by analyzing mouse mammary tumor formation in which each pathway was activated singly or pairwise. All three oncogenes showed cooperation during primary tumor formation, but efficient dissemination was only dependent on Ras. In addition, transcriptional profiling demonstrated that Ras induced adenocarcinomas with molecular characteristics related to human basal-like and HER2(+) tumors. In contrast, Ras combined with PIK3CA(H1047R), an oncogenic mutant linked to ERα(+)/luminal breast cancer in humans, induced metastatic luminal B-like tumors. Consistent with these data, elevated Ras signaling was associated with basal-like and HER2(+) subtype tumors in humans and showed a statistically significant negative association with estrogen receptor (ER) signaling across all breast cancer. Despite this, there are luminal tumors with elevated Ras signaling. Importantly, when considered as a continuous variable, Ras pathway activation was strongly linked to reduced survival of patients with ERα(+) disease independent of PI3K or Stat3 activation. Therefore, our studies suggest that Ras activation is a key determinant for dissemination and poor prognosis of ERα(+)/luminal breast cancer in humans, and hormone therapy supplemented with Ras-targeting agents may be beneficial for treating this aggressive subtype.

Cooperation between Pik3ca and p53 mutations in mouse mammary tumor formation.

PIK3CA, which codes for the p110α catalytic subunit of phosphatidylinositol 3-kinase, is one of the most frequently mutated genes in human breast cancer. Here, we describe a mouse model for PIK3CA-induced breast cancer by using the ROSA26 (R26) knock-in system, in which targeted Pik3ca alleles can be activated through transgenic expression of Cre recombinase. We mated Pik3ca(H1047R) and Pik3ca(wt) knock-in lines with MMTV-Cre transgenics, which express Cre in mammary epithelium. Starting at approximately 5 months of age, female R26-Pik3ca(H1047R);MMTV-Cre mice, but not control R26-Pik3ca(wt);MMTV-Cre mice, developed mammary tumors, as well as lymphoid and skin malignancies. R26-Pik3ca(H1047R);MMTV-Cre mammary tumors were typically either adenosquamous carcinoma or adenomyoepithelioma. As p53 is the most commonly mutated gene in breast cancer, we tested for genetic interaction between Pik3ca(H1047R) and p53 loss-of-function mutations in R26-Pik3ca(H1047R);p53(loxP/+);MMTV-Cre mice. This led to decreased survival of double-mutant animals, which developed lymphoma and mammary tumors with rapid kinetics. Mammary tumors that formed in p53(loxP/+);MMTV-Cre conditional mutants were either poorly differentiated adenocarcinoma or spindle cell/EMT, whereas R26-Pik3ca(H1047R);p53(loxP/+);MMTV-Cre mammary tumors were mostly adenosquamous carcinoma or spindle cell/EMT indicating that double-mutant mice develop a distinct spectrum of mammary tumors. Thus, an oncogenic variant of PIK3CA implicated in multiple human breast cancer subtypes can induce a very diverse spectrum of mammary tumors in mice. Furthermore, Pik3ca(H1047R) shows cooperation with p53, which altered the specific tumors that formed. Thus, the two most frequently mutated genes in human breast cancer show cooperation in mammary tumor formation.