Our data provide a molecular mechanism of elevated OXPHOS in TNBC via CDCP1/mitochondrial Src pathway. Importantly, our data specifically show that this pathway activates Complex I. Moreover, we show that NAD+ produced by Complex I is important for TNBC migration. Thus, our data position CDCP1/mitochondrial Src pathway as therapeutic targets to reduce Complex I activity and OXPHOS in TNBC. Since OXPHOS/metastasis pathway has been reported in other types of cancer besides TNBC, including pancreatic, ovarian, melanoma and prostate cancers [38,39,40,41,42], the CDCP1/mitochondrial Src pathway needs to be further investigated in these cancer cell contexts.
MDA-MB-231, SUM159, HEK293T, MDA-MB-468 and UCI-082014 [11] cell lines were grown in Dulbecco's Modified Eagle Medium (DMEM) (Genesee Scientific #25-500) with 10% Fetal Bovine Serum (FBS) (Genesee Scientific #25-514), 100 u/mL penicillin and 100 μg/mL streptomycin (1% P/S). Cells were grown in mixed gas CO water-jacketed incubators (21% O, 5% CO) and regularly confirmed to be negative for mycoplasma contamination. TNBC cells were treated with 1 μM Saracatinib (Cayman Chemical Company, #11497), 10 μM Dasatinib (Enzo Life Sciences, #76003-104), 0.5 mM nicotinamide riboside (Cayman Chemical Company, #23132) or 0.5 mM nicotinamide mononucleotide (Cayman Chemical Company, #16411) for 16 hours unless noted otherwise.
Detailed protocol is described in Razorenova et al. [43]. Lenti-X 293T cells were transfected with lentiviral plasmids for shRNA [11] or gene expression and packaging plasmids, pVSVG and ΔR8.2. TNBC cells were treated with virus-containing media supplemented with 6 µg/mL of polybrene, followed by selection in 1 µg/mL puromycin- or 2 µg/mL blasticidin-containing media for a minimum of 1 week. MTS-Kd-Src-Flag-Puro lentiviral expression vectors were packaged into lentiviral particles purchased from VectorBuilder Inc, Chicago, IL USA (VB240904-1416wze).
LentiGuide-puro and LentiCas9-blast viruses were produced in Lenti-X cells as described above. Three different LentiGuide-puro viruses were produced to target CDCP1 (constructs were generated by cloning using the following self-annealing oligo pairs: gCDCP1#1 5'caccggggtctctatcgcactgcta + 5'aaactagcagtgcgatagagaccc; gCDCP1#2 5'caccgcgatagagaccccgcagttc + 5'aaacgaactgcggggtctctatcg; gCDCP1#3 5'caccggtaggcaacaacgatgtcga + 5'aaactcgacatcgttgttgcctac, sequences complementary to CDCP1 gene are underlined). The LentiGuide-puro virus was produced to target GFP (and served as a non-targeting control), the sequence complementary to GFP is as follows: 5' ggcgaggagctgttcaccg. First, cells were transduced with LentiCas9-blast and selected with blasticidin at 2 µg/mL. Second, cells were transduced with LentiGuides and selected with puromycin at 1 µg/mL. Third, CDCP1 knockout cells were enriched by fluorescence-activated cell sorting (FACS) of cells with no CDCP1 expressed on the cell surface. Antibodies are listed in Supplemental Table 1. Stable CDCP1 knockdown (shCDCP1) and control (shGFP) cell lines were created as previously described [11], and pLKO.1shGFP was a kind gift from Silvestre Vicent (Stanford University).
The western blot protocol was adapted from Razorenova et al. [44]. Cells were lysed in lysis buffer (20 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X100, 2.5 mM NaPO, 1 mM β-glycerophosphate, 1 mM NaVO) with protease inhibitors (Fisher Scientific, Pittsburgh, PA #P1-88266) and phosphatase inhibitors (Roche #04906845001). Fifty micrograms of protein were used per lane. Antibodies are listed in Supplemental Table 1.
The Oxygen Consumption Rate (OCR) was measured using a Mito Stress Test Kit and XF24 Extracellular Flux Analyzer (Agilent) according to the manufacturer's protocol. In brief, 80,000 MDA-MB-231/UCI-082014 cells or 50,000 SUM159/HEK293T cells were plated in 100 μL of their standard growth media and cultured overnight. HEK293T cells were plated on poly-L-lysine (0.1 mg/ml) to improve attachment. Next day, cells were washed with XF base medium supplemented with 17.5 mM glucose, 2 mM L-Glutamine and 10 mM sodium pyruvate and incubated in a CO-free incubator at 37 °C for 1 hour to equilibrate prior to loading. OCR measurements were taken before and after the addition of oligomycin (ATP synthase inhibitor, 1 μM), FCCP (mitochondrial oxidative phosphorylation uncoupler, MDA-MB-231/UCI-082014 = 1 μM, SUM159 = 0.5 μM, HEK293T = 0.25 μM), Rotenone/Antimycin A (Complex I and III inhibitors, respectively, 0.5 μM each), and used to calculate basal respiration, maximal respiration and ATP production. OCR rates were normalized to cell numbers. For Complex I activity, cells were prepared as described above, and OCR was measured before and after the addition of 0.5 μM rotenone [45,46,47,48]. Complex I activity was determined by subtracting the lowest OCR measurement post-rotenone from the last basal OCR measurement pre-rotenone and normalized to cell number.
NADH fluorescence lifetime is used to determine its enzyme-bound (active) or free (inactive) state. A higher ratio of bound/free NADH corresponds to higher OXPHOS. The NADH FLIM imaging experiments were carried out as previously described [10]. In short, the FLIM images were obtained using an inverted laser scanning confocal microscope (LSM 710) with a ×40/1.2 numerical aperture C-Apochromat water-immersion objective, with the cells maintained in a stage-top incubator at 37 °C and 5% CO throughout imaging. NADH in MDA-MB-231 cells was excited using a two-photon Ti:Sapphire laser (Spectra-Physics, MaiTai) with an 80 MHz repetition rate at 740 nm and at ~2 mW power at the sample. This signal was passed through a 690 nm dichroic filter to separate excitation and emission signals. The fluorescence emission was separated by a bandpass filter (442/46 nm) to capture the cell auto-fluorescence and exclude any mApple emission and was then detected using a photomultiplier tube (Hamamatsu, #H7422P-40). The fluorescence lifetime decays were captured in the frequency domain using an A320 FastFLIM box (ISS) which were then mapped onto the phasor plot using the SimFCS 4 software -- developed at the Laboratory for Fluorescence Dynamics at the University of California, Irvine -- for quantitative NADH lifetime analysis. The lifetime free and protein bound form of NADH were used to determine the metabolic trajectory (the "M-Trajectory") on phasor FLIM plot, as demonstrated by Stringari et al. [49]. The instrument response time was calibrated before each experiment by imaging Coumarin-6 in ethanol (~10 μM), which has a known single exponential fluorescence lifetime of 2.4 ns.
The NAD/NADH ratio was determined following the manufacturer's protocol (Millipore Sigma MAK460).
The Luminescent ATP Detection Assay was performed following the manufacturer's protocol (Abcam #ab113849). Luminescence read with Bio Tek Cytation5 microplate reader (Agilent) was normalized to cell number.
Cells were fixed with 4% paraformaldehyde for 10 min, then treated with 0.25% Triton X-100 in 1xPBS for 5 min and blocked for 1 h in 3% bovine serum albumin (BSA) in 1xPBS. Slides were then incubated overnight in primary antibody at 4 °C, followed by incubation for 2 h in secondary antibody at room temperature in the dark. Antibodies are listed in Supplementary Table 1. Both primary and secondary antibodies were diluted in 3% BSA in 1xPBS. Coverslips were mounted with 1xPBS. Cells were imaged using a ZEISS Elyra 7 super-resolution microscope with Lattice SIM². Colocalization analysis and quantitation were performed using the surface-to-surface algorithm in Imaris Microscopy Image Analysis Software. The total Src volume and total mitochondrial volume were masked as "surfaces." The shortest distance between surfaces was quantified, and only surfaces that were within 0.3 microns were chosen (as tomm20 is an outer mitochondrial matrix protein and MTS-Src-Flag constructs are localized to the mitochondrial matrix, we don't expect a perfect overlap. The same logic applies to endogenous Src localization. The diameter of mitochondria is anywhere from 0.5 to 1.5 micron [50, 51], so 0.3 micron is a conservative constraint). The total volume of colocalizing surfaces as well as the total Src volume was quantified and the percentage of total Src in the mitochondria was then calculated using 5-8 fields of view. To assess CDCP1 localization, cells were prepared as described above and the Zeiss LSM 780 confocal microscope was used to image CDCP1 and mitochondria (using Tomm20 as a mitochondrial marker).
To prepare mitochondrial and cytosolic fractions, cells were harvested, washed in ice-cold PBS, minced and then resuspended in homogenizing buffer (20 mM HEPES-KOH, pH 7.5, 10 mM KCl, 1.5 mM MgCl, 1 mM sodium EDTA, 1 mM sodium EGTA, and 1 mM dithiothreitol) containing 250 mM sucrose and a mixture of protease inhibitors (1 mM PMSF, 1% aprotinin, 1 mM leupeptin, 1 mg/ml pepstatinA, and 1 mg/ml chymostatin). After 30 min incubation on ice, tissues were homogenized using a glass Pyrex homogenizer (type B pestle, 50-60 strokes) and centrifuged at 1000 × g for 5 min at 4 °C to remove debris and unbroken cells. The resulting supernatant was centrifuged at 12,000 × g for 12 min to obtain mitochondria-enriched preparation as a pellet. The supernatant is the cytosolic fraction. The mitochondrial pellet was washed 3 times in homogenizing buffer. 5-10 μg of total protein, cytosolic and mitochondrial fractions were analyzed by western blot as described above.
Cells were lysed as described in western blotting. 15 μL of Flag magnetic beads (Millipore Sigma #M8823) were added to 250 μg of protein in 250 μL of cell lysis buffer and incubated at room temperature with gentle agitation for 3 h. Immunoprecipitated proteins were collected using a magnetic stand, washed 5 times for 5 min with cell lysis buffer without protease or phosphatase inhibitors, boiled in 2× sample loading buffer (125 mM Tris-HCl pH 6.8, 4% SDS, 0.02% bromophenol blue, 20% glycerol, 4% mercaptoethanol) for 6 min and analyzed by western blot.
MDA-MB-231 cells were transfected using Lipofectamine 3000 reagent (Thermo Fisher Scientific #L3000008) following the manufacturer's protocol. SUM159 and HEK293T cells were transfected using Lipofectamine and Plus reagents (Thermo Fisher Scientific #18324012 and #11514015) following the manufacturer's protocol. ON-TARGETplus smartpool of 4 siRNAs targeting Src and CDCP1 as well as a non-targeting pool (#L-003175-00-0005, #L-010732-00-0005, and #D-001810-10-05) were purchased through Horizon Discovery. Mitochondrially-localized FLAG-tagged WT and kinase dead Src (pcDNA3-MTS-WT-c-Src-FLAG and pcDNA3-MTS-KD-c-Src-FLAG, respectively), and constitutively active Src (pEF6-Src-Y530F) were purchased from Addgene (# 44652, # 44653, and #124659) [33, 52]. Fl-CDCP1 and cCDCP1 (pLM-CMV-flCDCP1-puro and pLM-CMV-cCDCP1-puro, respectively) were described in Wright et al. [11]. pWP1-EV-GFP and pWPI-NDI1-GFP were provided by Dr. Chandel [53].
Mitochondrial-specific ROS was detected using MitoSOX Red reagent and was performed following the manufacturer's protocol (Thermo Fisher, #M36008). Mitochondrial ROS was quantified in individual cells on the BD Fortessa X20.
Cells were treated with 100 μM mitomycin C for 2.5 h before being detached with 0.05% trypsin in 1×PBS. Thirty thousand SUM159 cells in serum-free DMEM were added to the top chamber of the transwell inserts (Corning, NY, USA; #3464) and immersed in 10% FBS-containing media in the bottom chamber. Cells were allowed to migrate for 24 h. Cells from the top of the transwell inserts were removed with a cotton swab and cells that migrated through the transwell inserts were fixed, and fluorescent microscopy was used to count the GFP-positive cells. Cells were counted in 6-7 fields of view per transwell at 20× and normalized to control to quantify average fold change (FC) in migrated cells.
Cells were plated at 5000 cells per well in 1:1 mix of matrigel (growth factor reduced) (Corning, Cat. No. CB40230) and EpiCult -B Media supplemented with 5% FBS, 1% P/S, 10 ng/mL human epidermal growth factor, and 10 ng/mL basic fibroblast growth factor. Cells were allowed to grow and form spheres for 7 days.
1.0 ×10 gGFP, gCDCP1-1, gCDCP1-3, EV-puro or MTS-Kd-Src-flag-puro cells were injected into the 4 mammary fat pad of 8-week-old female NSG mice (N = 6, 5, 6, 7, 7 per group, respectively). Mouse sample size for metastasis experiments was consistent with our previous published work [12]. No randomization was used. For gGFP and gCDCP1 cells, cells were injected bilaterally to produce two tumors per mouse and combined tumor volume was monitored. Caliper measurements of tumors were taken twice a week and tumor volume was calculated as Volume = Length × Width × 0.51. Tumors were resected when volume reached 1.0 cm (30 days post-injection). Analysis of metastasis in lung tissue was performed as previously described [54]. Briefly, the lungs of NSG mice were harvested, dissociated into single cells, and stained with fluorescently conjugated antibodies for CD298 (Biolegend, Cat. no. 341704) and MHC-I (eBioscience, Cat. no. 17-5957-82), and flow cytometry was used to analyze and quantify disseminated CD298/MHC-I MDA-MB-231 cells using the BD Fortessa flow cytometer. All mouse experiments were approved by the Institutional Animal Care and Use Committee at UC Irvine (#AUP-22-012).
Statistical analysis was conducted in GraphPad Prism version 10.0 using a Student's t test or ANOVA between groups with *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 unless otherwise noted. Error bars (StDev or SEM) and the number of biological replicates are indicated in each figure legend.