Oleic acid differentially affects lipid droplet storage of de novo synthesized lipids in hepatocytes and adipocytes

Lipogenesis is a vital but often dysregulated metabolic pathway. Here we use optical photothermal infrared imaging to quantify lipogenesis rates of isotopically labelled oleic acid and glucose concomitantly in live cells. In hepatocytes, but not adipocytes, we find that oleic acid feeding at 60 μM increases the number and size of lipid droplets (LDs) while simultaneously inhibiting storage of de novo synthesized lipids in LDs. Our results demonstrate alternate regulation of lipogenesis between cell types.


Contents
Experimental Methods Materials Oleic acid conjugation and feeding protocol Huh-7 cell culture and 13 C and 2 H labeling Differentiated 3t3-L1 cell culture and 13 C and  S1: Average ratios of live and fixed Huh-7 cells fed with 2 H OA and 13 C glucose Table S2: Average ratios of live and fixed Huh-7 cells fed with 13 C glucose in BSA Table S5: Average ratios of live and fixed adipocytes fed with 2 H OA and 13 C glucose Table S6: Average ratios of live and fixed adipocytes fed with 2 H OA in BSA Note: Tables S3, S4, S7, S8, and S9 are found in the attached excel files.

Materials
Unless otherwise specified, all reagents were sourced from Sigma-Aldrich.

Oleic Acid Conjugation and Feeding Protocol
For feeding, 2 H oleic acid (oleic acid-d 33 , DLM-1891-PK, Cambridge Isotope Laboratories, Tewksbury, MA) was bound to fatty acid-free bovine serum albumin (BSA) using a protocol from Shi et al. 1 In a 2 mL clear glass container (VWR, Radnor, PA), Milli-Q purified water, 20 µL of 1 M NaOH, and 2.3 µL of 2 H oleic acid were combined and heated in a 70 C water bath until clear.
A solution of 150 mg/mL fatty acid-free bovine serum albumin (BSA) was then added to the 2 H oleic acid solution and heated again at 37 C in a water bath until clear.The solution was then filtered with a 0.2 m filter into another 2 mL clear glass container.The stock oleic acid concentration is approximately 3 mM.The ratio of oleic acid to BSA is 2:1, in line with physiological ratios. 2,3Oleic acid is then incubated with cell culture at a concentration of 60 µM, within a physiologically relevant range of oleic acid in human plasma 4 and less than the total fatty acid content in 10% FBS supplemented cell culture media. 5Free OA is likely much lower than 60 µM as most will remain bound to BSA.After a day of allowing cells to adhere to the coverslips, the media was replaced with glucose-free DMEM supplemented with 1% FBS, 1% penicillin/streptomycin, 13 C glucose (D-glucose-13 C 6 , CLM-1396-PK, Cambridge Isotope Laboratories) to 4.5 mg/mL, and either 2% BSA or 2% 2 H oleic acid conjugated with BSA (final concentration 60 µM OA and 30 µM BSA).

Cell fixing
For fixed cell samples, both cell lines were fixed at the desired time points with 4% paraformaldehyde (PFA) in PBS (Corning) for 20 minutes at room temperature and washed three times with PBS followed by three times with Milli-Q purified water and allowed to air-dry before analysis.

Live cell chamber construction
After feeding, live cell samples from both cell lines were washed once with PBS and mounted in PBS on a glass microscopy slide (VWR) with double-sided tape spacers (Nitto, San Diego, CA) with a 5 m thickness to keep the cells hydrated and sealed and minimize compression of the cell.

OPTIR data collection
All imaging was performed on a mIRage-LS IR microscope (Photothermal Spectroscopy Corporation, Santa Barbara, CA) integrated with a four-module-pulsed quantum cascade laser (QCL) system (Daylight Solutions, San Diego, CA) with a tunable range from 932 cm -1 to 2348 cm -1 .Brightfield optical images were collected using a low magnification 10X refractive objective with a working distance of 15 mm.Spectra and infrared images were collected in co-propagating mode using a 40X Cassegrain objective with a working distance of 8 mm.Fixed cell spectra and images were collected in standard (reflective) mode and live cell spectra and images were collected in transmission mode.Data was collected with an IR laser power of 20 % and a probe power in the range of 5-11 %.All spectra and images were collected using PTIR Studio 4.5.(Photothermal Spectroscopy Corporation).Image acquisition time was 2-5 minutes per frequency for a total acquisition time of 10-25 minutes per cell.

Data analysis
Images were processed in Fiji (NIH, Bethesda, MD). 7 Spectra were analyzed in IGOR Pro 9 (Wavemetrics, Portland, OR).Live and fixed cell ratio images were generated in Python 3.10 in Colab (Google, Mountain View, CA). 8tio images must be corrected to account for overlapping signals of other biomolecules (Figure S13).This is done through a correction protocol adapted 6 from Shi et al., 1 which removes contribution from the water band and protein amide I. Correction values are calculated for each cell line after analyzing single cell spectra in which no lipid signal is present.The intensity at the wavenumber at which lipid signal is collected (either 13 C=O, 12 C=O, or 12 C-2 H) is divided by the intensity at the amide I (1655 cm -1 ) or water band (2050 cm -1 ).These values are collected per cell across all time points and averaged to create a correction value with which to subtract off overlapping amide-I and water intensity for lipid intensities.
Fixed cell 13 C=O/ 12 C=O ratio images were corrected for the amide-I intensity using equation 1: S-4

Equation 1
Where A is the signal intensity of the frequency of the indicated biomolecule.The 12 C=O stretch is at 1744 cm -1 for Huh-7 cells and at 1747 cm -1 for adipocytes.For both cell lines, the 13 C=O stretch is at 1703 cm -1 .The variable b is the correction value for the amide-I intensity, with b being 0.34 for Huh-7 cells and 0.31 for adipocytes.There was no correction done for fixed 2 H-C/ 12 C=O ratio images.Fixed cell ratios were only performed on regions with sufficient 12 C=O signal intensity, defined as regions with 15% of the maximum value of 12 C=O signal for Huh-7 cells and 5% for adipocytes which visually corresponded to the lipid rich areas of the cells as determined by examining the single wavenumber images at 1744 cm -1 and the brightfield view of lipid droplets.
Live cell 13 C=O/ 12 C=O ratio images were corrected for the amide I and water band intensity using equation 2.

Equation 2
Where c is the correction value for and is 0.61 for Huh-7 cells and 0.66 for adipocytes and d is the correction value for and is 0.31 for Huh-7 cells and 0.28 for adipocytes.Live cell 2 H/ 12 C ratio images were corrected for the water band using equation 3.

Equation 3
The 12 C -2 H is measure from a shoulder at 2214 cm -1 for Huh-7 cells and at 2212 cm -1 for adipocytes.The water band is measured from 2050 cm -1 for both cell lines as this does not overlap with other 12 C -2 H stretches.The variable f is the correction value for the water band and is 0.87 for Huh-7 cells and 0.85 for adipocytes and d is the correction value discussed above.Live cell ratios were also only performed on regions with sufficient 12 C lipid ester carbonyl signal intensity, defined as regions with 15% of the maximum value of 12 C lipid ester carbonyl signal for Huh-7 cells and 5 % for adipocytes.

Lipid droplet size analysis
Lipid droplet size and diameter were determined using Fiji on cells at 48 hours after feeding with 13 C glucose and either 2 H oleic acid or BSA.The number of lipid droplets per cell were counted and the size of each lipid droplet determined using the line tool and measuring the diameter.These values were averaged over a total of 30 cells for both 2 H oleic acid fed and BSA fed trials.][11] Figure S3: Full assignments of (A) 2 H oleic acid IR spectrum 9,11 and (B) a live differentiated 3T3-L1 cell 72 hours after feeding with 2 H oleic acid and 13 C glucose.
Spatial resolution of OPTIR FigureS2.Full assignments of a 12 C glucose fed cell and 13 C glucose and 2 H OA fed cells FigureS3.Full assignments of 2 H oleic acid spectrum and cell fed with probes FigureS4.Sensitivity limit of OPTIR FigureS5.Live Huh-7 cell after 72 hours of feeding13  C glucose Figure S6.Line scan across a lipid droplet Figure S7.Visualization of rates of de novo lipid storage in live differentiated 3T3-L1 Figure S8.Visualization of rates of de novo lipid storage in fixed Huh-7 cells Figure S9.Visualization of rates of de novo lipid storage in fixed differentiated 3T3-L1 Figure S10.Fixed cell rates of 13 C incorporation in the vehicle and 2 H oleic acid Figure S11.Lipid droplet comparison brightfield images Figure S12.Slopes of 13 C incorporation rates Figure S13.Spectra based image correction Supplemental TablesTable

Figure S1 .
Figure S1.Instrument resolution.Profile line (left) and full image (right) of 1720 cm-1 intensity of section of PMMA beads embedded in epoxy.Beads are resolved from the background at a resolution of 500 nm or less (in-plane distance).The authors acknowledge Photothermal Spectroscopy Inc as this data was collected during instrument install.

Figure S5 :
Figure S5: Rates of de novo lipid storage in live Huh-7 cells can be observed by OPTIR imaging approach previously used on adipocytes.(A) Brightfield image of a 13 C glucose-fed Huh-7 cell at 72 hours.(B) Single wavenumber image collected at the 12 C=O lipid band (1744 cm -1 ) and (C) the corresponding ratio image of 13 C=O lipid ester carbonyl of triglycerides to 12 C=O lipid band, after correction.Scale bars are 10 m.

Figure S6 . 12 Figure S7 :
Figure S6.Spatial resolution of OPTIR.(A) Brightfield image of live 3T3-L1 cell with points indicating where spectra (B) were collected.500 nm separate each point except for the purple spectrum at the center of the droplet.

Figure S8 :
Figure S8: Visualization of rates of de novo lipid storage in fixed Huh-7 cells after feeding of 2 H oleic acid and 13 C glucose for 24-72 h.(A) Brightfield images of fixed Huh-7 cells (B) 2 H OA ratio images of 2 H lipid to the 12 C lipid band, after correction.(C) 13 C glucose ratio images of the 13 C lipid band to the 12 C lipid band, after correction Scale bars are 10 µm.(D) Average ratios of 2 H lipid band (grey) and 13 C lipid band (red) to 12 C lipid band across many cells.N=12-16 cells per time point.

Figure S9 :
Figure S9: Visualization of rates of de novo lipid storage in fixed differentiated 3T3-L1 cells after feeding of 2 H oleic acid and 13 C glucose for 24-72 h.(A) Brightfields of fixed differentiated 3T3-L1 cells.B) 2 H OA ratio images of 2 H lipid to the 12 C lipid band, after correction.(C) 13 C glucose ratio images of the 13 C lipid band to the 12 C lipid band, after correction Scale bars are 10 µm.(D) Average ratios of 2 H lipid band (grey) and 13 C lipid band (red) to 12 C lipid band across many cells N=11-13 cells per time point

Figure S11 :Figure S12 :
Figure S11: Effect of OA feeding on de novo lipid storage in fixed cells.Rates of uptake of 13 C glucose in 2 H OA trials (red) compared to in the BSA vehicle (purple) show that oleic acid has different effects between (A) fixed Huh-7 cells and (B) fixed differentiated 3T3-L1 cells.N= 7-16 cells per time point.* p-value< 0.05, ** p-value<0.005,***p-value<0.0005