Patients and tissue samples
This study comprised patients who received diagnoses at the General Surgery Department of Zhongshan Hospital, Fudan University, during the period spanning from July 2019 to June 2020. The criteria for inclusion were as follows: (1) Confirmation of gastric or colorectal cancer through a pathological assessment; (2) Underwent radical surgery at the hospital; (3) Availability of complete clinical data. The exclusion criteria were as follows: (1) Previous receipt of preoperative anticancer strategy (chemotherapy, targeted therapy, immunotherapy, or radiotherapy); (2) Presence of severe organ dysfunction (renal failure, liver insufficient, heart failure or chronic obstructive pulmonary disease), immunity impairment (autoimmune disorders, lymphoma, leukemia or acquired immune deficiency syndrome) or uncontrolled metabolic syndrome (hyperthyroidism or diabetes). All enrolled patients were briefed about the research and granted their consent to take part.
Clinical data (age, gender, height, weight, weight loss, tumor site, and staging) for the enrolled patients were recorded. Nutrition-related laboratory parameters, including hemoglobin, albumin, pre-albumin, and total protein, were also recorded.
According to the 2011 international consensus on CAC [1], patients were defined as CAC if they satisfied one or more of these conditions: (1) A decrease in body weight exceeding 5% during the preceding 6 months; (2) A weight reduction of over 2% coupled with a body mass index (BMI) below 20 kg/m2.
Abdominal subcutaneous WAT was carefully separated and divided into approximately 500 mg pieces before being transferred to sample storage tubes. The specimens were then rapidly frozen using liquid nitrogen and preserved at -80℃ for subsequent examination.
CAC rodent model
Male C57/BL6 mice, weighing 18-20 grams and aged 6-7 weeks, were randomly assigned to two study groups: (1) healthy control group (CONT) and (2) tumor-bearing group (CAC). The mice were provided by the Shanghai Institutes for Biological Science. The CAC mice were subcutaneously injected with 5*106 Lewis lung carcinoma (LLC) cells dissolved in phosphate-buffered saline (PBS) near their ilium in the lower back. In contrast, the CONT group received PBS injections only. All mice were euthanized after 28 days following LLC cell or PBS injection. The samples of inguinal white adipose tissue (iWAT), epididymis white adipose tissue (eWAT), gastrocnemius, liver, and heart from each mouse were collected. The samples were promptly frozen in liquid nitrogen and preserved at -80℃ for subsequent examination.
RNA sequencing
Total RNA extraction followed the official guidelines provided by Invitrogen using the TRIzol method. Degradation and contamination of RNA were identified using agarose gel (1%) electrophoresis. The measurement of RNA concentration was determined by employing a NanoPhotometer spectrophotometer for OD260/280 and OD260/230. RNA integrity was assessed by an Agilent 2100 Bioanalyzer. Subsequently, mRNA was extracted and constructed into a cDNA library following official guidelines and technical documents from NEB (China). After assembly, the library underwent quantification with a Qubit 2.0 fluorometer and was then diluted to a concentration of 1.5 ng/μL using enzyme-free water. The insert fragments were detected by using an Agilent 2100 Bioanalyzer. The libraries were quantified using qRT-PCR, meanwhile ensuring an effective library concentration exceeding 2 nM. Sequencing of the library was performed by the Illumina Hiseq 4000 platform.
Bioinformatic data processing
The raw data in fastq format underwent initial processing and data cleaning with Perl scripts developed in-house. After that, the read number of each gene was counted and normalized. The reference genome and its corresponding index were assembled using Hisat2 v2.0.5 after being downloaded from the Ensembl database (Ensembl Mus musculus release-94). A differential expression analysis was performed between two sets of data utilizing the DESeq2 R package (version 1.16.1). Genes identified as differential expressions had an adjusted p-value (padj) below 0.05. The Gene Ontology (GO) database was used for enrichment analysis. Among the genes with differential expressions, GO terms that demonstrated a padj below 0.05 were deemed significantly enriched.
Real-time PCR
Total RNA from both animal and human tissue was extracted by using the TRIzol method. The reverse transcription of mRNAs was conducted by employing a Qiagen Quantinova kit. A real-time PCR reaction system was established with a Qiagen SYBR green PCR kit and was performed on a LightCycler 480 Platform. The list of PCR primers used was attached to supplemental materials.
Western blotting
RIPA lysis buffer was used for protein extraction from both tissue and cell samples, following which the proteins were quantified by the bicinchoninic acid method. Protein was separated by SDS-PAGE and transferred to a nitrocellulose membrane before blocking with Tris-buffered saline supplemented with 5% skimmed milk powder for 1 hour. Following this, the membranes underwent an incubation at 4℃ overnight with the primary antibody and then were subjected to a 1-hour incubation with goat anti-rabbit IgG marked with horse radish peroxidase (HRP). The Tanon 5200 imaging system was used for identifying and analyzing stained bands. In this study, β-tubulin (CST, USA; 1:1000) and Elovl6 (Novus Biologicals, USA; 1:1000) were the primary antibodies used.
FAME analysis
A mixture of ether and petroleum ether was used to extract fatty acids. Fatty acid was esterified to form fatty acid methyl esters (FAMEs) before being injected into the gas chromatograph for flame ionization detection. The percentage of a particular fatty acid is determined by the ratio of the peak area of that fatty acid to the sum of the peak areas of all fatty acid components.
Cell culture
The 3T3-L1 mouse preadipocytes were cultivated in Dulbecco's modified eagle medium with high-glucose (HG-DMEM, Thermo Fisher), into which 10% neonatal calf serum (NCS) from AusgeneX and 1% penicillin-streptomycin solution (PSS) from ThermoFisher were added. Similarly, LLC and the C26 mouse colon cancer cell lines were cultured in HG-DMEM. These cultures received additional supplementation, including 10% fetal bovine serum (FBS) from AusgeneX and 1% PSS. The cells were kept in a microbiological incubator at 37°C with a 5% CO2 environment. The Cell Bank at the Chinese Academy of Sciences generously provided all the cells.
Differentiation of 3T3-L1 cells
For the initial differentiation of 3T3-L1 cells, a differentiation-induced medium (DIM-I) was employed, consisting of DMEM with 1 μM dexamethasone, 0.5 mM 3-isobutyl-1-methylxanthine (IBMX), 1 μg/mL insulin, 1% PSS, and 10% FBS. Subsequently, another differentiation-induced medium (DIM-II) was used to maintain the differentiation environment. DIM-II was composed of DMEM with 1 μg/mL insulin, 1% PSS, and 10% FBS. The 3T3-L1 cell differentiation process commenced two days after achieving 100% confluency (D0) by switching to the DIM-I medium. On day 2 (D2), the medium was switched to DIM-II, which was renewed every 2 days until further intervention.
Preparation and use of CAC-mimic medium
The preparation of a CAC-mimic medium has been described in our previous work [33]. Two days after passaging C26 cells (or 3T3-L1 cells), the medium supernatant was gathered and filtered. The CAC-mimic medium consisted of 66% C26 medium supernatant and 33% complete medium, replacing D4 in the CAC group to form a cachexic environment. A mixture of 66% 3T3-L1 medium supernatant and 33% complete medium served as the corresponding intervention for the negative control (NC) group. Interleukin-6 (IL-6) was added to the complete medium as a positive control at a final concentration of 100 ng/ml [34].
Cell transfection
In 6-well plates, 3T3-L1 cells were cultured until they reached a confluency of 70-90% for transfection. Following a PBS rinse, 700 µl of reduced-serum medium (Opti-MEM) was added to each well. The transfection solution consisted of two mixtures. Liquid A contained 9 µl of ThermoFisher Lipofectamine 2000, while liquid B contained 75 pmol per well of siRNA for the Elovl6 KD group. In both cases, 150 µl of Opti-MEM was added. The negative control group also used the same volume but with only Opti-MEM. The procedure involved combining liquid A and B in equal volumes, incubating for five minutes, and then adding the mixture to the 6-well plate. After 8 hours of transfection, the medium was switched back to a complete medium. Following that, the cells were allowed to grow until they reached 100% confluency for differentiation, as previously described. Sangon Biotech in Shanghai (China) custom-designed and produced the siRNA used for Elovl6 knockdown. It featured a sense strand sequence of GCUCUUCGAACUGGUGCUUTT, and an antisense strand sequence of AAGCACCAGUUCGAAGAGCTT.
Oil red O (ORO) staining
ORO solid powder was dissolved in isopropanol to reach a concentration of 0.5% and then filtered with filter paper to obtain the ORO staining storage solution. It was stored in the dark at a temperature of 4°C. Upon being prepared for utilization, the storage solution was diluted 60% in water and filtered with filter paper to obtain the ORO staining working solution. The cells underwent two rounds of PBS rinsing, followed by fixation with 4% paraformaldehyde at room temperature (RT) for half an hour. Subsequently, they received another round of washing with diluted water. Following that, the cells were subjected to an ORO staining solution, allowing them to stain for 30 min at RT. Afterward, destaining was carried out by a single wash with absolute ethanol. After staining each well, 1 ml of isopropanol was introduced, and the absorbance was measured at 510 nanometers utilizing the TECAN spectrophotometer.
Statistical analysis
Statistical hypothesis testing and plotting were performed using Prism GraphPad 9.0.0. Normality testing was conducted using both the Anderson-Darling and the Shapiro-Wilk tests. Based on the normality of data, a Student's t-test or Kruskal-Wallis’s rank-sum test was used to evaluate overall mean differences between groups. The Chi-squared test was used to compare the proportion of categorical variables between groups.