Page 88 - Read Online

Longhi et al. Microbiome Res Rep 2024;3:4 https://dx.doi.org/10.20517/mrr.2023.02 Page 3 of 15

Several approaches have been proposed to optimize DNA obtained from samples of different sources low in
[18]
microbial counts. Many commercially available microbiome-specific DNA extraction kits and the use of a
[16]
variety of additional steps employing detergents including saponin, Tween 20 and Triton X-100 , or
[19]
Benzonase are amongst the approaches that have been developed to deplete host DNA without apparently
influencing the prokaryotic DNA. Nevertheless, while many studies have demonstrated that these
approaches could reduce human DNA contamination, few have pointed out their significant impact on the
final microbial community structure delineated by shotgun metagenomics attempts. Among these
approaches, different saponin percentages have been extensively tested and proposed as the golden standard
for the depletion of eukaryotic DNA in highly contaminated samples with high relative concentrations of
human DNA compared to bacterial DNA [20-22] . However, an important limitation of the use of saponin is
linked to the differential impact that this reagent produces on the DNA of various bacteria and, thus, on the
generation of artifacts in the relative abundance of the different microbial groups. In this regard, saponin
has been previously reported to possess strong antimicrobial activity towards specific bacterial taxa through
in vitro and in vivo investigations .
[23]
For this reason, in this study, we have carefully evaluated the saponin-based protocol for extracting DNA
from different human sample specimens to overcome high eukaryotic turnover and highlighted its main
limitations in terms of microbial profiling.

METHODS
Human samples collection
Regarding the biological specimens included in this study, one sample was collected for each human matrix.
Specifically, the vaginal swab sample was gathered by a healthy woman. The oral sample consisted of a
lingual swab. Sputum samples represent the thick mucus expelled from the lower airways (bronchi and
lungs) through a deep cough. Saliva samples were the early morning sampling before teeth washing. The
skin sample represented the forehead microbiota sampled with a film dressing. Biopsies were different
sections of the stomach (gastric antrum and body), and the nursing staff performed the nasopharyngeal
swab collection at Parma Day Hospital. All the biological specimens included in this study were collected
from different healthy donors and stored at -80 °C until they were processed.

Signed informed consent was obtained from the individuals enrolled in this study.

Human DNA depletion and bacterial enrichment
Samples (1 mL) were centrifuged at 6,000 g for 3 min. For biopsies and skin samples, a fraction of the
specimen was homogenized with 1 mL of phosphate buffer saline (PBS), and the suspension was then
treated like other samples. After centrifugation, the supernatant was carefully removed, and cell pellets were
incubated with PBS and saponin at the required concentration (see in the main text the different
concentrations used) at room temperature for 10 min. The depletion protocol involved a commercially
available product, known as saponin, obtained from the internal bark of the Quillaja Saponaria (Sigma-
Aldrich, MO, USA).

Following the incubation, 350 µL of sterile water was added, and incubation was continued at room
temperature for 30 s, after which 12 µL of 5 M NaCl was added to deliver an osmotic shock, lysing the
damaged host cells. Samples were centrifuged at 6,000 g for 5 min, with the supernatant removed and the
pellet resuspended in 100 µL of PBS. Turbo DNase buffered (Thermo Fisher Scientific, USA) with Turbo
DNase enzyme (Thermo Fisher Scientific, USA) was added at 37 °C for 30 min to promote host cell lysis.
Finally, the host DNA-depleted samples were washed two times with decreasing volumes of PBS (1 mL and