Alterations of gut microbiome accelerate multiple myeloma progression by increasing the relative abundances of nitrogen-recycling bacteria

Abstract

Background: Gut microbiome alterations are closely related to human health and linked to a variety of diseases. Although great efforts have been made to understand the risk factors for multiple myeloma (MM), little is known about the role of the gut microbiome and alterations of its metabolic functions in the development of MM.

Results: Here, in a cohort of newly diagnosed patients with MM and healthy controls (HCs), significant differences in metagenomic composition were discovered, for the first time, with higher bacterial diversity in MM. Specifically, nitrogen-recycling bacteria such as Klebsiella and Streptococcus were significantly enriched in MM. Also, the bacteria enriched in MM were significantly correlated with the host metabolome, suggesting strong metabolic interactions between microbes and the host. In addition, the MM-enriched bacteria likely result from the regulation of urea nitrogen accumulated during MM progression. Furthermore, by performing fecal microbiota transplantation (FMT) into 5TGM1 mice, we proposed a mechanistic explanation for the interaction between MM-enriched bacteria and MM progression via recycling urea nitrogen. Further experiments validated that Klebsiella pneumoniae promoted MM progression via de novo synthesis of glutamine in mice and that the mice fed with glutamine-deficient diet exhibited slower MM progression.

Conclusions: Overall, our findings unveil a novel function of the altered gut microbiome in accelerating the malignant progression of MM and open new avenues for novel treatment strategies via manipulation of the intestinal microbiota of MM patients. Video abstract.

Keywords: Fecal microbiota transplantation; Gut microbiome; Multiple myeloma; Nitrogen-recycling bacteria.

Fig. 1

Gut microbiome characterization and bacterial diversity. a PCoA based on the Bray-Curtis dissimilarity index shows the between-subjects (β) diversity across groups, in which the blue circles and red triangles represent HC and MM subjects, respectively (PERMANOVA, P = 0.001). b Species rarefaction curves in blue or red indicate HC and MM subjects, respectively. c, d Violin graphs in blue or red show the Shannon-Wiener index of HC and MM subjects, respectively, at the species level (c) and at the genus level (d) (one-tailed Wilcoxon rank-sum test, P = 0.045 at the species level and P = 0.043 at the genus level)

Fig. 2

Species with differential abundance in HC and MM subjects of the cohort. a Left panel: the fold change of abundance of 36 different species in MM in relation to HC. Right panel: heatmap shows the scaled logarithm base-10 of the abundance of 20 MM-enriched species and 16 HC-enriched species. b Graphs indicate the species with significant differential abundance in an expanded cohort as verified using qPCR. HC-enriched species are highlighted in blue frame, while MM-enriched species in red frame. In HC, circles in red and white represent the subjects from a new collection of controls and metagenomic sequenced group, respectively. In MM, squares in red and white represent the subjects from a new collection of MM patients and metagenomic sequenced group, respectively. P value was determined by using two-tailed Mann-Whitney test. Note that some species were undetected in some samples in Fig. 2b. There are 6 and 4 undetected samples for Anaerostipes hadrus and Clostridium saccharobutylicum, respectively. And, there are 9, 3, 2, and 3 undetected samples for Raoultella ornithinolytica, Citrobacter freundii, Klebsiella variicola, and Klebsiella pneumoniae, respectively

Fig. 3

Functional alterations of microbes (KOs) in MM. a PCoA based on Bray-Curtis dissimilarity index shows the between-samples (β) diversity (PERMANOVA, P = 0.004). b Eight modules mapped by at least 2 identified KOs. c Metabolic pathways of NH₄⁺ in the intestinal lumen. d–f Cell proliferation curves with different glutamine concentrations, including 5TGM1 (d), ARP1 (e), and 8226 (f). g, h The activities of urease (g) and glutamine synthase (h) in HC and MM subjects, respectively. i Linear fitting of urease and glutamine synthase activity in all subjects, in which blue dots and red triangles represent HC and MM samples, respectively. j, k The concentrations of serum urea (j) and NH₄⁺ (k) in HC, MGUS, and MM samples, respectively. P value was determined by using two-tailed unpaired t test. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 4

Metabolic profiling of host serum. a OPLS-DA score plot based on the metabolic profiling of serum samples, in which blue dots and red triangles represent HC and MM subjects, respectively (R2Y = 0.853; Q2 = 0.679). b Heatmap shows the scaled abundance of 26 differential metabolites. c Heatmap illustrates the Spearman’s correlation between 14 differential species and 26 differential metabolites, in which the red and blue species and metabolites denote enrichment in MM and HC samples, respectively. *, +, and # all suggest the significance, *P < 0.05, +P < 0.01, #P < 0.001

Fig. 5

FMT experiment with MM and HC feces. a Schematic representation of FMT experiments. b Based on the Euclidean distance calculated from the relative abundances of several characteristic bacteria, experiments with PCoA show that these bacteria changed in FMT_HC and FMT_MM mice between week -4 and week 0. c Lines in blue, in red, and in black denote the concentration of serum lgG2b in FMT_HC, FMT_MM, and PBS mice, respectively. d Lines in blue, in red, and in black indicate the fluorescence intensity of live imaging in FMT_HC, FMT_MM, and PBS mice, respectively. e Live imaging of all recipient mice. f Heatmap reflects the changes of the scaled relative abundance of bacteria over time in FMT_HC, FMT_MM, and PBS mice. g Biplot of RDA of the bacteria in FMT_HC, FMT_MM, and PBS mice. P value was determined by two-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 6

Differential metabolic profiling in FMT_HC, FMT_MM, and PBS mice. a, b The concentrations of BM l-glutamic acid (a) and l-glutamine (b) in FMT_HC, FMT_MM, PBS, and normal mice. P value was calculated by two-tailed unpaired t test. c, d The concentrations of serum l-glutamic acid (c) and l-glutamine (d) in FMT_HC, FMT_MM, PBS, and normal mice. P value was calculated by two-tailed unpaired t test. e, f The concentrations of cecal l-glutamic acid (e) and l-glutamine (f) in FMT_HC, FMT_MM, PBS, and normal mice. P value was calculated by two-tailed unpaired t test. g, h The concentrations of serum urea (g) and NH₄⁺ (h) in FMT_HC, FMT_MM, and PBS mice, respectively. P value was determined by using two-way ANOVA test. i, j The deposition of monoclonal protein (lgG2b kappa) in the kidney surface of FMT_HC, FMT_MM, PBS, and normal mice. P value was determined by using one-tailed t test. k, l The concentrations of cecal urea after microbial cells being unbroken (k) and broken (l), respectively. P value was calculated using two-tailed unpaired t test. m, n The concentrations of cecal NH₄⁺ after microbial cells being unbroken (m) and broken (n), respectively. P value was calculated using two-tailed unpaired t test. o, p The activities of cecal urease (o) and glutamine synthase (p) in FMT_HC, FMT_MM, and PBS mice, respectively. P value was calculated by using two-tailed unpaired t test. q Scatter plot of enzyme activities in cecal contents of FMT_HC, FMT_MM, and PBS mice. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 7

Mice experiment with Klebsiella pneumoniae and Clostridium butyricum. a Schematic representation of the characteristic bacteria transplantation experiment. b, c Lines in blue, in black, and in red denote the concentration of serum lgG2b (b) and the fluorescence intensity of live imaging (c) in FMT_CBu, PBS, and FMT_KPn mice, respectively. P value was determined by using two-way ANOVA test. d Live imaging of all recipient mice. e, f At week 6, the IgG2b concentration (e) and the tumor fluorescence intensity (f) in PBS, KPn_WT, and KPn_Mut mice, respectively. g, h At week 6, the concentrations of serum glutamine (g) and glutamic acid (h) in PBS, KPn_WT, and KPn_Mut mice, respectively. i, j At week 6, the concentrations of cecal glutamine (i) and glutamic acid (j) in PBS, KPn_WT, and KPn_Mut mice, respectively. k, l The activities of cecal urease (k) and glutamine synthase (l) in PBS, KPn_WT, and KPn_Mut mice, respectively. m, n The concentrations of cecal urea (m) and NH₄⁺ (n) in PBS, KPn_WT, and KPn_Mut mice, respectively. P value was determined by using two-tailed unpaired t test. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 8

Mice experiment by gavage with ammonium and urea. a Schematic representation of the mice experiments. b Based on the Euclidean distance calculated from the relative abundances of several MM-enriched bacteria, experiments with PCoA show that these bacteria changed in NH₄Cl and urea mice between week -1 and week 0. c, d Lines in gray, in pink, and in orange denote the serum lgG2b concentrations (c) and the tumor fluorescence intensity (d) in NaCl, NH₄Cl, and urea mice, respectively. e Live imaging of all recipient mice. f, g At week 6, the concentrations of serum urea (f) and NH₄⁺ (g) in NaCl, NH₄Cl, and urea mice, respectively. h, i At week 6, the activities of cecal urease (h) and glutamine synthase (i) in NaCl, NH₄Cl, and Urea mice, respectively. j, k At week 6, the concentrations of cecal glutamine (j) and glutamic acid (k) in NaCl, NH₄Cl, and urea mice, respectively. l, m At week 6, the concentrations of serum glutamine (l) and glutamic acid (m) in NaCl, NH₄Cl, and Urea mice, respectively. P value was determined by using two-tailed unpaired t test. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 9

Mice experiment fed with glutamine- and cysteine-deficient diet. a Schematic representation of the mice experiment. b At week 7, the tumor fluorescence intensity in Ctr, Gln-, and Plus- mice, respectively. c Live imaging of all recipient mice at week 7. d–f The concentrations of serum lgG2b (d), glutamine (e), and glutamic acid (f) at week 7. P value was determined by using two-tailed unpaired t test. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 10

Schematic representation of the host-microbes interaction centered on nitrogen recycling