Experimental methods
Human sample preparation
Human samples and clinical data were obtained from patients undergoing either hepatectomy surgery due to a hepatic pathology or a donor hepatectomy for a living donor liver donation (Fig. 1b). In the hepatectomies that were performed due to liver pathologies, a sample was taken at least 5 cm from the closest margin of the pathology. In the hepatectomies that were performed as donor hepatectomy for a living donor liver donation, a liver biopsy of 1 cm³ was obtained immediately upon abdominal opening, prior to mobilization of the liver and initiation of resection. All hepatectomies due to liver pathologies were performed at Sheba Medical Center, Israel with approval from the Sheba Medical Center Helsinki committee. All donor hepatectomies for a living donor liver donation were performed at Mayo Clinic, Rochester, MN, with approval from the Mayo Clinic IRB committee. Informed consent was obtained from all patients, and all experiments followed all the Helsinki committee guidelines and regulations. Samples were taken only in cases where no pathology was observed in the sampling area during pre-operative and intraoperative evaluations. In all cases ~1 cm of hepatic parenchyma distant from the liver capsule area were resected. Tissues were gently washed in PBS and were embedded in OCT for Visium and MERFISH or fresh-frozen for Visium HD.
Non-human sample preparation
Whole washed pig (Sus scrofa domesticus) livers were bought from a butcher at Kibbutz Lahav C.R.O., Israel, which sells meat products from surplus animals. Specimens were adult males (5.5 months, ~85 kg) and an adult female (5.5 months, ~75 kg) raised under farming conditions. Washed cattle (Bos taurus) liver sections were donated by a butcher in Haifa, Israel. Specimens were adult males (17 months, ~670 kg) raised under farming conditions. Wild Boar (Sus scrofa) liver samples were taken from the cadavers of animals euthanized as part of population control regulated hunting done by the Israel Nature and Parks Authority (INPA). The sampling permit was acquired beforehand under INPA regulation. The researchers did not influence the time, location or manner of death; they were solely permitted to take samples from the cadavers. Age and weight estimation of the specimens were evaluated by trained INPA rangers and a veterinarian who participated in the hunting expedition. The boars’ ages were approximately 1–2 years, and they weighed approximately 60 kg. Liver tissues were gently washed in PBS and were embedded in OCT for spatial transcriptomics (Fig. 3b–d and Extended Data Fig. 6).
10X Visium spatial transcriptomics
Fresh-frozen OCT embedded blocks were cryo-sectioned into 10-µm-thick slices and carefully placed on a Visium Spatial Gene Expression slide. Each slide contained 4 slices from 4 different patients, total of 4 Visium slides were used for the 16 human samples. Slides were fixed with pre-chilled methanol at −20 °C, stained for H&E as described in the Visium user guide and were captured using a Leica DMi8 widefield inverted microscope equipped with a Leica DFC7000T colour camera and an HC PL APO 20×/0.80 DRY objective (506530, Leica Microsystems). Permeabilization time was set according to the 10X Tissue optimization protocol, resulting in 24 min for all 16 human and 7 non-human mammalian samples. After tissue permeabilization and transcript capture, libraries were prepared according to the Visium Spatial Gene Expression User guide with 14–17 cycles of PCR for cDNA amplification. Four Libraries for each slide were pooled and loaded at a concentration of 400 pM to a SP100 flow cell and sequenced on Illumina NovaSeq 6000. The Fastq reads generated from the sequencer were preprocessed by 10X Genomics space-ranger software (version 2.0.1) which included spatial de-barcoding, read-alignment to hg38 and UMI-generation. Further preprocessing was done in MATLAB R2022a.
10X Visium HD spatial transcriptomics
Fresh-frozen tissue samples were processed into FFPE blocks. Sections of 5 μm thickness were mounted onto Visium HD slides and deparaffinized according to the Visium HD Spatial Gene Expression User Guide. H&E staining was performed to visualize tissue morphology, and high-resolution brightfield images were captured using a Leica DMi8 widefield inverted microscope equipped with a Leica DFC7000T colour camera and an HC PL APO 20×/0.80 DRY objective (506530, Leica Microsystems). Following imaging, sections were destained and underwent de-crosslinking as described in the Visium HD FFPE Tissue Preparation Handbook. Three specific probes for each targeted gene were added to the deparaffinized, destained, and decrosslinked tissues.
MERFISH spatial transcriptomics
... continue reading