In Vivo Entombment of Bacteria and Fungi during Calcium Oxalate, Brushite, and Struvite Urolithiasis

Abstract

Background: Human kidney stones form via repeated events of mineral precipitation, partial dissolution, and reprecipitation, which are directly analogous to similar processes in other natural and manmade environments, where resident microbiomes strongly influence biomineralization. High-resolution microscopy and high-fidelity metagenomic (microscopy-to-omics) analyses, applicable to all forms of biomineralization, have been applied to assemble definitive evidence of in vivo microbiome entombment during urolithiasis.

Methods: Stone fragments were collected from a randomly chosen cohort of 20 patients using standard percutaneous nephrolithotomy (PCNL). Fourier transform infrared (FTIR) spectroscopy indicated that 18 of these patients were calcium oxalate (CaOx) stone formers, whereas one patient formed each formed brushite and struvite stones. This apportionment is consistent with global stone mineralogy distributions. Stone fragments from seven of these 20 patients (five CaOx, one brushite, and one struvite) were thin sectioned and analyzed using brightfield (BF), polarization (POL), confocal, super-resolution autofluorescence (SRAF), and Raman techniques. DNA from remaining fragments, grouped according to each of the 20 patients, were analyzed with amplicon sequencing of 16S rRNA gene sequences (V1-V3, V3-V5) and internal transcribed spacer (ITS1, ITS2) regions.

Results: Bulk-entombed DNA was sequenced from stone fragments in 11 of the 18 patients who formed CaOx stones, and the patients who formed brushite and struvite stones. These analyses confirmed the presence of an entombed low-diversity community of bacteria and fungi, including Actinobacteria, Bacteroidetes, Firmicutes, Proteobacteria, and Aspergillus niger. Bacterial cells approximately 1 μm in diameter were also optically observed to be entombed and well preserved in amorphous hydroxyapatite spherules and fans of needle-like crystals of brushite and struvite.

Conclusions: These results indicate a microbiome is entombed during in vivo CaOx stone formation. Similar processes are implied for brushite and struvite stones. This evidence lays the groundwork for future in vitro and in vivo experimentation to determine how the microbiome may actively and/or passively influence kidney stone biomineralization.

Keywords: Raman spectroscopy; bacteria; basic science; fungi entombment; geomicrobiology; kidney stone mineralogy; microbiome; nephrolithiasis; super-resolution autofluorescence (SRAF) microscopy; urolithiasis.

Graphical abstract

Figure 1.

Study flow. A total of 20 patients consented and enrolled in the study. Mineralogies analyzed in this study include calcium oxalate (CaOx) (n=18 patients), struvite (n=1 patient), and brushite (n=1 patient). After filtering reads through pipelines for quality, preprocessing, and contamination, *16S rRNA gene sequences (V3–V5 hypervariable region) were detected in nine out of 20 patients (45%), whereas internal transcribed spacer ^†internal transcribed spacer 2 (ITS2) sequences were detected in 11 out of 20 patients (55%). PCNL, percutaneous nephrolithotomy; IR, infrared.

Figure 2.

Total number of reads and taxa grouped by kidney stone mineralogical type. Number of amplicon sequence variants (ASVs) detected within calcium oxalate (CaOx), brushite, and struvite kidney stones, respectively, targeting the (A) V3–V5 hypervariable region of the 16S rDNA gene and (B) internal transcribed spacer 2 (ITS2) hypervariable region of the fungal rDNA gene. The number of reads for each individual patient stone fragment group are presented in Supplemental Figure 1. The asterisk (*) represents six of the stone fragment groups presented in (A), and an additional three stone fragment groups included in calculation. The dagger (^†) represents the same stone fragment groups presented in (A). (C) Binary heatmap indicating the presence or absence of unique taxonomical divisions identified from ASVs (blue = bacteria; gray = nonribosomal host human DNA; green = fungi). The n values correspond to individual patient stone fragment groups.

Figure 3.

Phylogenetic diversity of three mineralogical types of human kidney stones. Pie charts represent the community diversity of bacterial 16S rRNA gene sequences, nonribosomal human host (A, C, and E), and internal transcribed spacer (ITS) fungal sequences (B, D, and F) from calcium oxalate (CaOx), brushite, and struvite patient-specific kidney stone fragment groups, respectively. The community diversity of bacterial and fungal sequences for each individual patient kidney stone fragment group is presented in Supplemental Figure 1.

Figure 4.

Patient metadata analysis. Wilcoxon signed-rank tests that the detection of fungal sequences is not significantly correlated with elevated urine calcium excretion at the α=0.05 level (P=0.07). No statistically significant correlation was observed between the presence or absence of fungal sequences and urine oxalate excretion. Mean and SD for each sample group are depicted in the text below the graphs. Purple and blue boxes are box plots for urine calcium and oxalate levels, respectively. The box plot summary statistics are as follows: lower boundary of box = first quartile, line in center of box = second quartile (median), upper boundary of box = third quartile. Dots represents individual patient data points. A dot unattached to the vertical lines represents outlier points, which were included when calculating the interquartile range.

Figure 5.

Microscopy evidence of microorganism entombment within human kidney stones. Mineralogical identifications determined with a combination of bulk stone Fourier Transform Infrared (FTIR) analyses and determination of individual crystal morphologies (2) and Raman spectroscopy (Figure 7) in thin sections. (A) transmitted light photomultiplier (TPMT), (B) super-resolution autofluorescence (SRAF), and (C) TPMT overlaid on SRAF image from a struvite kidney stone documenting entombed bright orange autofluorescent coccoidal and rod-shaped bacterial cells. See also Supplemental Figure 5 for contextualization of the occurrence of both coccoidal and rod-shaped bacteria in surrounding regions of the thin section.

Figure 6.

Evidence for bacteria entombed within struvite and brushite kidney stones. Mineralogical identifications determined with a combination of bulk stone Fourier Transform Infrared (FTIR) analyses and determination of individual crystal morphologies (2) and Raman spectroscopy (Figure 7) in thin sections. (A) Color brightfield (BF) image of a 25 µm thin section prepared from a struvite stone. (B) The same field of view as in a indicating that concentrically layered spherulitic hydroxyapatite exhibits extinction under polarized light (POL) and are therefore amorphous (non-crystalline). Conversely, the radiating needle-like (acicular) crystals of struvite are strongly birefringent. (C) Color BF image of enlargement box shown in (A). White arrows in (C and D) indicate cross-sections at various oblique angles of entombed coccoidal and rod-shaped bacteria. (D) Super-resolution autofluorescence (SRAF) image of the same field of view shown in (C). (E) BF image of polymorphic twinning of radiating acicular brushite crystals. This is an approximately 600 nm thick slice in reflection mode, therefore not all BF objects are not in focus. White arrows in (E and F) indicate cross-sections at various oblique angles of entombed coccoidal and rod-shaped bacteria. (F) SRAF image of the same field of view shown in (E). Regions of blurred (fuzzy) concentric zonations represent regions of mimetic dissolution and replacement (2,4).

Figure 7.

Raman spectroscopy evidence for entombed bacteria within a struvite (NH₄MgPO₄0.6H₂O) human kidney stone. (A–C) Mineral component 1 (A, pseudo-colored red), mineral component 2 (B, pseudo-colored blue), and the corresponding merged images extracted from Raman spectra (C). (D) Transparent overlay of image (C) on a lower magnification color brightfield (BF) image illustrating optical microscopy correlated with Raman spectroscopy. (E) Raman spectra for mineral components 1 (A) and 2 (B), with legends highlighting chemical components for identified peaks on the basis of Takasaki (Supplemental Reference 11) and Balan et al. (Supplemental Reference 12). (F–J) Enlargement of box in (D) similar to Figure 5C but with high magnification Raman scan pseudo-colored red for hydroxyapatite Ca₁₀(PO₄)₆(OH₂) and green for struvite. Note the 959 peak (italicized in J) is the “high” peak for hydroxyapatite among the other peaks. Also note the similarity between struvite peaks in (E) (larger field of view) and (J) (smaller field of view).

Figure 8.

Microscopy evidence for entombed bacteria in brushite human kidney stones. (A) Color brightfield (BF) image of a 25 µm thin section prepared from a brushite stone. (B) Same field of view as in a indicating that concentrically layered spherules exhibit extinction under polarized light (POL) and are therefore amorphous (noncrystalline; white arrows). Conversely, the radiating acicular crystals of brushite (white box in B, see also Figure 5, E and F) are strongly birefringent. Inset in lower right in (B) is a transmitted light photomultiplier (TPMT) image of spherules with entombed coccoidal and rod-shaped bacteria throughout each crystal (white arrows). White arrows in (B) indicate cross-sections at various oblique angles of entombed coccoidal and rod-shaped bacteria.

Figure 9.

Microscopy-to-omics evidence for microbiome entombment during biomineralization in natural environments. Images and figures modified from previous publications (6,7). (A) A skeleton of the scleractinian coral Orbicella annularis from the leeward reef tract of Curaçao exhibits extensive fungal hyphae and borings (white arrow) in brightfield (BF). (B) Phase contrast (PC) image of the same field of view shown in (A). (C) Rapid growth and accretion of CaCO₃ travertine at Mammoth Hot Springs in Yellowstone National Park is coated by and entombs coccoidal, rod, and filaments of the bacterium Sulfurihydrogenibium yellowstonense. A merged blue and red super-resolution autofluorescence (SRAF) image (violet to red color) overlaid on a BF image. (D) Environmental scanning electron microscope (ESEM) of the same sample shown in (C). (E) Phylogenetic diversity pie chart of the microbiome associated the deposition of hot spring travertine at Mammoth Hot Springs. (F) Phylogenetic diversity pie chart of the microbiome entombed in all calcium oxalate (CaOx) kidney stones analyzed in this study.