Lung adenocarcinoma (ADC) and squamous cell carcinoma (SCC) are two most common subtypes of lung cancer. Here, to identify new, targetable molecular properties of both subtypes, we monitored changes in the levels of heme- and oxidative phosphorylation (OXPHOS)-related proteins during lung tumorigenesis. Heme is a central molecule for oxidative metabolism and ATP generation via OXPHOS. Notably, both lung ADC and SCC tumors can be induced in the genetically engineered KLLuc mouse model harboring the G12D Kras mutation and a conditional Lkb1 knockout. We found that the levels of the rate-limiting heme synthesis enzyme ALAS1 and uptake protein SLC48A1, along with OXPHOS complex subunits, progressively increased as lung tumorigenesis advanced. Our data demonstrated that elevated levels of heme- and OXPHOS-related proteins were associated with both ADC and SCC. Importantly, treatment of KLLuc mice with a heme-sequestering protein, HeSP2, that inhibits heme uptake in tumor cells effectively arrested lung tumor progression, and both ADC and SCC tumors were strongly suppressed. Additionally, HeSP2 effectively suppressed the growth of both SCC and ADC tumor xenografts in NOD/SCID mice. Further analyses indicated that HeSP2 effectively diminished OXPHOS in both ADC and SCC, reduced angiogenesis, alleviated tumor hypoxia, and suppressed cell proliferation. These results show that the advancing of lung tumorigenesis requires progressive increase in cellular heme synthesis and uptake, leading to intensified OXPHOS activity and ATP generation and promoting aggressive tumorigenic functions.

Implications:

Heme sequestration is an effective strategy for the suppression of both ADC and SCC tumor initiation and development.

Lung cancer is the leading cause of cancer-related death in the United States and worldwide (1). Lung cancer is a very heterogeneous disease at both cellular and histologic levels. Lung cancer can be divided into two major types: non–small cell lung cancer (NSCLC), which accounts for ∼85% of cases, and small cell lung cancer, which accounts for the remaining 15% of lung cancer cases (2). NSCLC is further subdivided into lung adenocarcinoma (ADC), lung squamous cell carcinoma (SCC), and large cell carcinoma. Lung cancer exhibits widespread intertumoral and intratumoral heterogeneity as well as incidences of subtype transdifferentiation (3–8). Such molecular heterogeneity and transdifferentiation afford multiple mechanisms for the development of drug resistance and pose great challenges for the treatment of lung cancer. Indeed, despite the advent of targeted therapies and immunotherapies, an effective treatment or cure for lung cancer remains an unlikely outcome for most patients. The five-year survival rate remains 10% to 20%, lower than many other cancers, including breast (90%) and prostate (99%) cancers (1). Thus, novel strategies are still needed for dramatic improvement in the overall survival rate of lung cancer patients.

Accumulating evidence increasingly demonstrates that mitochondrial metabolism and oxidative phosphorylation (OXPHOS) are crucial for tumorigenesis (9, 10). Increased oxidative metabolism does not conflict with the fact that most cancer cells exhibit enhanced glycolytic rates, as Warburg observed (11). Studies of NSCLC tumors in human patients showed that pyruvate and lactate from elevated glycolysis enter and intensify the TCA cycle in all tumors, although NSCLC tumors are metabolically heterogeneous (12, 13). Additionally, components of OXPHOS complexes and markers of mitochondrial biogenesis are found to be highly predictive of reduced overall survival in NSCLC patients (14). Likewise, the expression of OXPHOS genes is negatively correlated with the prognosis of lung ADC (15). Heme is essential for OXPHOS/mitochondrial respiration and drug metabolism (16). Multiple subunits in OXPHOS complexes II–IV contain heme. Heme coordinates the expression and assembly of OXPHOS complexes (17). It is also worth noting that heme can have an array of adverse effect to cells (18). Importantly, recent studies in our laboratory have shown that enhanced heme biosynthesis and uptake lead to elevated levels of mitochondrial heme, which cause intensified OXPHOS and ATP generation, thereby promoting tumorigenic functions in NSCLC cells (19). Thus, it would be interesting to examine if and how heme metabolism and function change as lung tumorigenesis initiates and progresses and if such changes differ in different subtypes of NSCLC.

Genetically engineered mouse models (GEMM) have provided valuable tools for understanding the development of the various lung cancer subtypes (20). Notably, KL mice offer a powerful GEMM for NSCLC. They harbor a conditional activating mutation (G12D) at the endogenous Kras locus, a conditional Lkb1 knockout, and conditional luciferase expression for monitoring tumorigenesis in live mice (21, 22). Lung tumors developed in KL mice include lung ADC and SCC. They exhibit broad tumor heterogeneity (21–23). KL tumors also express low levels of immune markers, including PD-L1, and are not responsive to immunotherapy (24). Thus, KL tumors represent those lung tumors that are more drug resistant and difficult to treat. Thus, KL mice provide a powerful tool to define the molecular features of resistant NSCLC tumors and to identify effective therapeutic strategy to overcome their resistance. In this report, using KL mice, we monitored the levels of markers of heme synthesis and uptake, along with markers of OXPHOS, tumor vasculature and oxygenation, and glucose metabolism, as lung tumorigenesis initiates and advances. We found that the levels of proteins and enzymes involved in heme synthesis and uptake strongly increased along with markers of glucose metabolism and tumor vasculature, as lung tumors developed in KL mice. Elevated levels of markers of heme synthesis and uptake are associated with both ADC and SCC. Further, using a heme-sequestering protein (HeSP2), we showed that heme sequestration can effectively suppress the progression of both ADC and SCC.

Reagents

HeSP2 (previously known as HSP2) was purified using pET11a expression system. Dr. Mario Rivera (University of Kansas) kindly provided us the pET11a expression vector for Yersinia pestis HasA residues 1–193 (25). HeSP2 contains Q32H Y75M double mutations that were generated with the QuikChange II Site-Directed Mutagenesis Kit (Agilent Technologies; ref. 19). Ad5-CMV-Cre was purchased from Baylor Vector Development Laboratory. The Opal 4 color IHC Kit was purchased from Akoya Biosciences. ATP Determination Kit was purchased from Molecular Probes.

Cell culture

All NSCLC cell lines, H1299 (ATCC cat. no. CRL-5803, RRID: CVCL_0060), A549 (ATCC cat. no. CRM-CCL-185, RRID: CVCL_0023), H460 (ATCC cat. no. HTB-177, RRID: CVCL_0459), HCC15 (DSMZ cat. no. ACC-496, RRID: CVCL_2057), H226 (ATCC cat. no. CRL-5826, RRID: CVCL_1544), H520 (ATCC cat. no. HTB-182, RRID: CVCL_1566), and SW900 (ATCC cat. no. HTB-59, RRID: CVCL_1731), were purchased from ATCC and maintained in RPMI medium. H1299, A549, and H460 were supplemented with 5% heat-inactivated FBS, whereas HCC15, H226, H520, and SW900 were supplemented with 10% heat-inactivated FBS and switched to 5% FBS, 24 hours before experiments. All experiments were conducted using cells between passages 3 to 5 from revival of the frozen stocks. Cell lines were authenticated by Genetica, LabCorp and were found to be >95% identical to the standard. Mycoplasma testing for cell lines was done using a MycoFluor Mycoplasma Detection Kit (Molecular Probes), and the results were negative. Some known mutations in these cell lines are as follows: HCC15 (homozygous TP53 mutation, c.776A>T; STK11 deletion; NRAS mutation, c.181 C>A; CDKN2A deletion, c.1_471del471), NCI-H226 (homozygous CDKN2A deletion, c.1_150del150), NCI-H520 (homozygous TP53 mutation, c.438G>A; homozygous CDKN2A deletion, c.134delG), SW900 (heterozygous KRAS mutation, c.35G>T; homozygous TP53 mutation, c.499C>T), H1299 (heterozygous NRAS mutation, c.181C>A; homozygous TP53 deletion), A549 (homozygous KRAS mutation, c.34G>A; homozygous STK11 mutation, c.109C>T; homozygous CDKN2A deletion, c.1_471del471), and H460 (homozygous KRAS mutation c.183A>T; homozygous CDKN2A deletion; c.1_471del471, homozygous KRAS mutation c.183A>T; homozygous STK11mutation, c.109C>T).

Measurement of heme synthesis and uptake

Measurement of heme synthesis and uptake in cells was carried out exactly as described previously (19, 26). Experiments were conducted in triplicates and were normalized with total cellular proteins.

Animals

LSL-KrasG12D; Lkb1flox/flox; LSL-Luciferase (KLLuc) mice were generated, bred, and confirmed as described (27, 28). NOD/SCID mice (IMSR cat. no. CRL:394, RRID: IMSR_CRL:394) were purchased from Charles River Laboratories. In accordance with NIH guidelines, mice were bred and cared for in a University of Texas at Dallas (Richardson, TX)–specific pathogen-free animal facility. All animal procedures were performed under protocols approved by Institutional Animal Care and Use Committee at the University of Texas at Dallas. Animals were regularly examined for any signs of stress and euthanized according to preset criteria.

In vivo experiments with KLLuc mice

Lung tumors were induced by intranasal inhalation of Ad5-CMV-Cre virus into 6- to 11-week-old LSL-KrasG12D; Lkb1flox/flox; LSL-Luciferase (KLLuc) mice at 3 × 107 PFU per mouse, as described (28). Mice were anesthetized using intraperitoneal administration of Ketamine-Xylazine cocktail (ketamine, 100 mg/kg; xylazine, 12 mg/kg). Tumor growth was monitored using bioluminescence imaging (BLI) and histology. For studies on tumor progression, mice were sacrificed at 4, 6, 8, and 10 weeks after infection. For treatment experiments, mice were randomized into two groups (n ≥ 6 per group) that received saline (for control) or HeSP2 (25 mg/kg, i.v., every three days). Treatment was started postinfection when BLI signal from lung tumors reached above 5 × 106 photons/sec, after 6 to 7 weeks postinfection. Body masses were recorded once every week. Mice were sacrificed after three weeks of treatment.

Treatment of human NSCLC xenografts in NOD/SCID mice

To generate subcutaneous xenograft models, 2 × 106 HCC15 cells containing 50% Matrigel in serum-free medium were injected subcutaneously into the left flank region of 4- to 6-week-old female NOD/SCID mice (n  = 6 per group). Mice were randomized into treatment groups that received saline (for control) and HeSP2 (i.v. 25 mg/kg every three days), respectively. Body masses were recorded once every week. Treatments were started only after tumors were palpable to ensure successful implantation. Tumor growth was monitored with caliper measurements.

Measurement of oxygen consumption and ATP levels

Oxygen consumption was measured, as described previously (26). Oxygen consumption rates (OCR) and ATP levels from freshly isolated tumors were measured as described (19). Both OCRs and ATP levels were normalized with protein amounts. For treatment with the inhibitor of heme synthesis, succinyl acetone, and heme-add back, cells were cultured in medium containing heme-depleted serum. Heme-depleted serum was prepared as described (29). Cells were grown in media containing 0.5 mmol/L succinyl acetone, then supplemented with 0, 5, 10, and 20 μmol/L heme, respectively, for six days.

Histology and tumor lesion area measurement

Mice were sacrificed, and lungs were excised and prepared for histology. Lung tissues were fixed in 4% formalin. Paraffin embedding was performed at Histology core at University of Texas Southwestern Medical Center (Dallas, TX). The paraffin blocks were sectioned into 5 μm sections and stained with hematoxylin and eosin (H&E). Sections were scanned at 40× resolution with an Olympus VS120 slide scanner. Tumor lesion areas were measured using cellSens software from Olympus.

Immunohistochemistry (IHC)

IHC was carried out exactly as described previously (30). Primary antibodies were diluted in 1× TBS/1% BSA/5% goat serum (16210-072, Gibco). The dilutions were 1:100 DeltaNp63 (Abcam cat. #ab172731, RRID:AB_2891015), 1:100 TTF-1 (Agilent cat. #M3575, RRID:AB_2877699), 1:200 Ki-67 (Cell Signaling Technology cat. #12202, RRID:AB_2620142), 1:200 cleaved caspase-3 (Cell Signaling Technology cat. #9661, RRID:AB_2341188), 1:100 Alas1 (Novus cat. #NB100-56415, RRID:AB_837550), 1:200 SLC48A1 (Novus cat. #NBP1-91563, RRID:AB_11009875), 1:50 HCCS (Abcam cat. #ab224321, RRID:AB_2891018), 1:200 cytochrome c (Novus cat. #NBP2-21569, RRID:AB_2891022), 1:100 COX5A (Abcam cat. #ab110262, RRID:AB_10861723), 1:500 COX5B (Abcam cat. #ab180136, RRID:AB_2891023), 1:500 COX7B (Abcam cat. #ab137094, RRID:AB_2891024), 1:1,000 ATP5A (Abcam cat. #ab14748, RRID:AB_301447), 1:200 SLC2A1 (Abcam cat. #ab40084, RRID:AB_2190927), 1:500 HK2 (Thermo Fisher Scientific cat. #PA5-29326, RRID:AB_2546802), 1:200 SLC1A5 (Abcam cat. #ab58690, RRID:AB_943478), 1:200 GLS (Abcam cat. #ab156876, RRID:AB_2721038), 1:35 VEGFA (Abcam cat. #9570, RRID: AB_308723), 1:50 VEGFR1 (Abcam cat. #ab2350, RRID:AB_303000), 1:200 HIF1α (Novus Biologicals cat. #NB100-449, RRID: AB_10001045), 1:200 CA9 (Novus biologicals cat. #NB100-417, RRID:AB_ 10003398), 1:200 CD34 (Abcam cat. #ab185732, RRID:AB_2811308), and CD31 (PECAM1; Cell Signaling Technology cat. #77699, RRID:AB_2722705).

Sections were incubated with primary antibodies overnight at 4°C and then with horseradish peroxidase–conjugated goat anti-rabbit IgG (Thermo Fisher Scientific cat. # 31460, RRID: AB_228341) or horseradish peroxidase–conjugated goat anti-mouse IgG (Thermo Fisher Scientific cat. #31430, RRID: AB_228307) at room temperature for 45 minutes. Sections were stained with tyramide signal amplification-conjugated fluorophores in 1× Plus Amplification Diluent (Akoya Biosciences cat. #FP1498).

IHC sections were scanned at a 10× resolution with an Olympus VS120 slide scanner and quantified using cellSens software from Olympus. For each lung tissue, protein expression in the tumor region is normalized to expression in the nontumor region. Microvessel density was calculated based on PECAM1 and CD34-positive objects as described previously (31). Experiments with all antibodies were repeated at least twice. Intensity or object counts from 10 ROIs were averaged for each tumor tissue. At least three sets of lung tissues per treatment were quantified, and the averages were calculated and plotted.

Statistical analyses of data

Data from different treatment groups of cells, mice, and tissues were compared, and statistical analysis was performed with a Welch two-sample t test.

Data availability statement

The data generated in this study are available within the article and its supplementary data files.

The levels of the heme uptake protein and rate-limiting heme synthetic enzyme increase progressively as lung tumorigenesis advances

Our previous experimental data have shown that the levels of heme synthesis and uptake, along with the levels of heme uptake protein and heme synthetic enzyme ALAS1, are elevated in NSCLC cell lines relative to nontumorigenic lung cell lines, leading to increased mitochondrial heme levels (19). Here, to monitor changes during the process of lung tumor development, we took advantage of the KLLuc (LSL-KrasG12D; Lkb1flox/flox; LSL-Luciferase) mouse model (32). Previous studies have shown that these mice develop a full spectrum of NSCLC tumor types, including SCC, ADC, and mixed adenosquamous tumors (21–23). Consistent with previous studies, we detected small lung tumors starting at four weeks after tumor initiation by infection with the Ad-Cre virus (see Fig. 1A). Tumor burden/size progressively increased as lung tumorigenesis advanced to 10 weeks after tumor initiation (see Fig. 1A and E). We found that ADC appeared before SCC in KLLuc mice, which is consistent with previous observations (21, 23). As shown in Fig. 1B, F, and G, ADC, revealed by the NKX2-1 (TTF-1) marker, was clearly detected starting at four weeks after tumor initiation, while SCC, revealed by the ΔNp63 (p40) marker, was detected at eight weeks after tumor initiation.

Figure 1.

Markers of ADC and SCC, as well as heme pathway proteins, increase with lung tumorigenesis. A, Representative H&E images of lung tissue sections from KLLuc mice sacrificed at 4, 6, 8, and 10 weeks after the induction of lung tumorigenesis via viral infection. B, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the ADC and SCC markers, NKX2-1 and ΔNp63, respectively. White and blue rectangles in DAPI montage images denote ADC and SCC regions shown in 10×, respectively. C, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the heme biosynthetic enzyme ALAS1. D, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the heme transporter SLC48A1. Shown in C and D are montage images of lung tissue sections stained with H&E (row 1), montage images stained with DAPI (row 2), or antibodies against the indicated protein (row 3), and 10× images of lung tissue sections stained with DAPI (row 4) or antibodies against the indicated protein (row 5). The light blue lines in H&E montage images outline the tumors in the lung. The blue rectangles in H&E montage images denote the regions shown in 10×.The light green lines in DAPI montage images outline the tumors in the lung. The blue rectangles in H&E and white rectangles in DAPI montage images denote the regions shown in 10×. Scale bar, montage, 1 mm; 10×, 100 μm. E, Graph showing quantified tumor lesion areas correlating with A. F, Graph showing quantified levels of NKX2-1 correlating with B. G, Graph showing quantified levels of ΔNp63 correlating with B. H, Graph showing quantified levels of ALAS1 correlating with C. I, Graph showing quantified levels of SLC48A1 correlating with D. Protein levels were quantified, and data are plotted as mean ± SEM.

Figure 1.

Markers of ADC and SCC, as well as heme pathway proteins, increase with lung tumorigenesis. A, Representative H&E images of lung tissue sections from KLLuc mice sacrificed at 4, 6, 8, and 10 weeks after the induction of lung tumorigenesis via viral infection. B, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the ADC and SCC markers, NKX2-1 and ΔNp63, respectively. White and blue rectangles in DAPI montage images denote ADC and SCC regions shown in 10×, respectively. C, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the heme biosynthetic enzyme ALAS1. D, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the heme transporter SLC48A1. Shown in C and D are montage images of lung tissue sections stained with H&E (row 1), montage images stained with DAPI (row 2), or antibodies against the indicated protein (row 3), and 10× images of lung tissue sections stained with DAPI (row 4) or antibodies against the indicated protein (row 5). The light blue lines in H&E montage images outline the tumors in the lung. The blue rectangles in H&E montage images denote the regions shown in 10×.The light green lines in DAPI montage images outline the tumors in the lung. The blue rectangles in H&E and white rectangles in DAPI montage images denote the regions shown in 10×. Scale bar, montage, 1 mm; 10×, 100 μm. E, Graph showing quantified tumor lesion areas correlating with A. F, Graph showing quantified levels of NKX2-1 correlating with B. G, Graph showing quantified levels of ΔNp63 correlating with B. H, Graph showing quantified levels of ALAS1 correlating with C. I, Graph showing quantified levels of SLC48A1 correlating with D. Protein levels were quantified, and data are plotted as mean ± SEM.

Close modal

Next, we monitored the levels of the rate-limiting heme synthetic enzyme ALAS1 and the heme uptake protein SLC48A1. Increased levels of these proteins have been shown to strongly increase mitochondrial heme levels and OCRs, promoting ATP generation and tumorigenic functions in tumor cells (19). The levels of both ALAS1 (Fig. 1C and H) and SLC48A1 (Fig. 1D and I) were strongly and progressively increased as lung tumorigenesis advanced from 4 to 10 weeks after tumor initiation. These results are consistent with previous observation showing that elevated heme synthesis and uptake underpin tumorigenic functions in NSCLC cells (19). Further, these results indicate that heme synthesis and uptake are progressively increased to support tumorigenic functions in lung cells, as tumorigenesis advances.

The levels of important proteins and enzymes relating to cellular bioenergetics increase concomitantly as lung tumorigenesis advances

Heme serves as a prosthetic or cofactor for multiple subunits of mitochondrial OXPHOS complexes and coordinates the generation of OXPHOS complexes. Elevated heme synthesis and uptake should promote OXPHOS complex synthesis. We therefore detected the levels of OXPHOS complex subunits at various time points of lung tumorigenesis. We found that the levels of OXPHOS complex subunits, like heme synthetic enzyme and transporter, progressively increased as lung tumorigenesis advanced from 4 to 10 weeks after tumor initiation. These include cytochrome c (CYCS; Fig. 2A and F), cytochrome c oxidase subunit COX7B (Fig. 2B and G), and ATP synthase subunit alpha ATP5F1A (Fig. 2C and H). Likewise, the levels of cytochrome c-type heme lyase HCCS also increased strongly as lung tumorigenesis advanced (Supplementary Fig. S1A and S1D). We have also confirmed the specificity of the antibodies by showing that ATP5F1A, COX7B, and CYCS all colocalized with the mitochondrial localization marker TOMM20 (see Supplementary Fig. S1G). Furthermore, Supplementary Fig. S1H shows that NSCLC cells supplemented with higher levels of exogenous heme exhibited significantly higher OCRs. This result indicates that increased heme availability is linked to elevated OXPHOS activity.

Figure 2.

The levels of OXPHOS and glycolytic proteins increase with lung tumor progression. A, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the OXPHOS protein CYCS. B, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the OXPHOS protein COX7B. C, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the OXPHOS protein ATP5F1A. D, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the glucose transporter SLC2A1. E, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the glycolytic enzyme HK2. Shown are montage images of lung tissue sections stained with H&E (row 1), montage images stained with DAPI (row 2), or antibodies against the indicated protein (row 3), and 10× images of lung tissue sections stained with DAPI (row 4) or antibodies against the indicated protein (row 5). The light green lines in DAPI montage images outline the tumors in the lung. The blue rectangles in H&E and white rectangles in DAPI montage images denote the regions shown in 10×. Scale bar, montage, 1 mm; 10×, 100 μm. F, Graph showing quantified levels of CYCS correlating with A. G, Graph showing quantified levels of COX7B correlating with B. H, Graph showing quantified levels of ATP5F1A correlating with C. I, Graph showing quantified levels of SLC2A1 correlating with D. J, Graph showing quantified levels of HK2 correlating with E. Protein levels were quantified, and data are plotted as mean ± SEM.

Figure 2.

The levels of OXPHOS and glycolytic proteins increase with lung tumor progression. A, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the OXPHOS protein CYCS. B, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the OXPHOS protein COX7B. C, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the OXPHOS protein ATP5F1A. D, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the glucose transporter SLC2A1. E, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of the glycolytic enzyme HK2. Shown are montage images of lung tissue sections stained with H&E (row 1), montage images stained with DAPI (row 2), or antibodies against the indicated protein (row 3), and 10× images of lung tissue sections stained with DAPI (row 4) or antibodies against the indicated protein (row 5). The light green lines in DAPI montage images outline the tumors in the lung. The blue rectangles in H&E and white rectangles in DAPI montage images denote the regions shown in 10×. Scale bar, montage, 1 mm; 10×, 100 μm. F, Graph showing quantified levels of CYCS correlating with A. G, Graph showing quantified levels of COX7B correlating with B. H, Graph showing quantified levels of ATP5F1A correlating with C. I, Graph showing quantified levels of SLC2A1 correlating with D. J, Graph showing quantified levels of HK2 correlating with E. Protein levels were quantified, and data are plotted as mean ± SEM.

Close modal

Next, we examined the levels of proteins involved in the utilization of two key fuels promoting tumorigenic functions, glucose, and glutamine. We found that the levels of the glucose transporter SLC2A1 (Fig. 2D and I) and the key glycolysis enzyme hexokinase HK2 (Fig. 2E and J) progressively and strongly increased as lung tumorigenesis advanced from 4 to 10 weeks after tumor initiation. Note that the increase of SLC2A1 was somewhat delayed compared with HK2, likely due to its low expression associated with ADC (see below). The levels of the glutamine transporter SLC1A5 (Supplementary Fig. S1B and S1E) and the key enzyme of glutamine consumption, glutaminase GLS1 (Supplementary Fig. S1C and S1F), also strongly increased in lung tumors compared with normal tissues. These results show that enzymes of key bioenergetic pathways, including OXPHOS, glucose, and glutamine consumptions, along with key heme synthetic enzyme and transporter, are upregulated progressively as lung tumorigenesis advances, to support increased needs for cellular energy.

Markers of angiogenesis and hypoxia increase as lung tumorigenesis advances

As tumorigenesis advances, it is expected that angiogenesis will be induced. Indeed, we found that the levels of the angiogenic factor VEGFA (Fig. 3A and F) and its receptor VEGFR1 (Fig. 3B and G) increased progressively as lung tumorigenesis advanced from 4 to 10 weeks after tumor initiation. Likewise, the levels of the vascular marker CD34 (Fig. 3C and H), as well as microvessel density (Fig. 3I), progressively increased along with lung tumorigenesis. Further, we examined tumor hypoxia by detecting the levels of hypoxia-inducible factor HIF1A and the long-term hypoxia marker carbonic anhydrase 9 (CA9). The levels of HIF1A increased in all 4- to 10-week tumors (Fig. 3D and J), and the levels of CA9 increased progressively (Fig. 3E and K), indicating tumor hypoxia progressively intensified in 4- to 10-week tumors.

Figure 3.

Markers of tumor vasculature and hypoxia increase with lung tumorigenesis. A, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of VEGFA. B, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of VEGFR1. C, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of CD34. D, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of HIF1A. E, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of CA9. Shown are montage images of lung tissue sections stained with H&E (row 1), montage images stained with DAPI (row 2), or antibodies against the indicated protein (row 3), and 10× images of lung tissue sections stained with DAPI (row 4) or antibodies against the indicated protein (row 5). The light green lines in DAPI montage images outline the tumors in the lung. The blue rectangles in H&E and white rectangles in DAPI montage images denote the regions shown in 10×. Scale bar, montage, 1 mm; 10×, 100 μm. F, Graph showing quantified levels of VEGFA correlating with A. G, Graph showing quantified levels of VEGFR1 correlating with B. H, Graph showing quantified levels of CD34 correlating with C. I, Graph showing quantified levels of microvessel density based on IHC analysis with antibodies to CD34 shown in C. J, Graph showing quantified levels of HIF1A correlating with D. K, Graph showing quantified levels of CA9 correlating with E. Protein levels were quantified, and data are plotted as mean ± SEM.

Figure 3.

Markers of tumor vasculature and hypoxia increase with lung tumorigenesis. A, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of VEGFA. B, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of VEGFR1. C, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of CD34. D, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of HIF1A. E, Representative H&E and IHC images of lung tissue sections from KLLuc mice showing levels of CA9. Shown are montage images of lung tissue sections stained with H&E (row 1), montage images stained with DAPI (row 2), or antibodies against the indicated protein (row 3), and 10× images of lung tissue sections stained with DAPI (row 4) or antibodies against the indicated protein (row 5). The light green lines in DAPI montage images outline the tumors in the lung. The blue rectangles in H&E and white rectangles in DAPI montage images denote the regions shown in 10×. Scale bar, montage, 1 mm; 10×, 100 μm. F, Graph showing quantified levels of VEGFA correlating with A. G, Graph showing quantified levels of VEGFR1 correlating with B. H, Graph showing quantified levels of CD34 correlating with C. I, Graph showing quantified levels of microvessel density based on IHC analysis with antibodies to CD34 shown in C. J, Graph showing quantified levels of HIF1A correlating with D. K, Graph showing quantified levels of CA9 correlating with E. Protein levels were quantified, and data are plotted as mean ± SEM.

Close modal

Elevated levels of heme synthesis, uptake, and OXPHOS activity are associated with both ADC and SCC

ADC and SCC have previously been shown to exhibit different metabolic properties (27, 33). One particularly notable example is the selectively high expression of GLUT1/SLC2A1 in SCC. Indeed, we detected high levels of GLUT1/SLC2A1 expression associated with SCC, whereas low levels of SLC2A1 expression were associated with ADC in the KLLuc tumors (Fig. 4A and F). Notably, elevated levels of the glutamine transporter SLC1A5 were associated with both ADC and SCC (Fig. 4B and G). Next, we examined the levels of the rate-limiting heme synthetic enzyme ALAS1 and heme transporter SLC48A1 in ADC and SCC. We found that high levels of ALAS1 (Fig. 4C and H) and SLC48A1 (Fig. 4D and I) were associated with both ADC and SCC. Furthermore, elevated levels of OXPHOS complex subunits COX5A (Fig. 4E and J), COX5B (Supplementary Fig. S2A and S2D), COX7B (Supplementary Fig. S2B and S2E), and CYCS (Supplementary Fig. S2C and S2F) were associated with both ADC and SCC. These results suggest that OXPHOS activities are elevated in both ADC and SCC to support tumorigenic functions of these cancer cells.

Figure 4.

The levels of representative metabolic proteins and enzymes in ADC and SCC tumors developed in KLLuc mice. A, Representative IHC images of lung tissue sections showing levels of glucose transporter SLC2A1, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. B, Representative IHC images of lung tissue sections showing levels of glutamine transporter SLC1A5, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. C, Representative IHC images of lung tissue sections showing levels of ALAS1, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. D, Representative IHC images of lung tissue sections showing levels of SLC48A1, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. E, Representative IHC images of lung tissue sections showing levels of COX5A, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. Shown are montage and 10× images of lung tissue sections stained with DAPI (row 1), antibodies to NKX2-1 or ΔNp63 (row 2), antibodies to the indicated protein (row 3), and merged images of NKX2-1 or ΔNp63 with the indicated protein (row 4). The light green lines in DAPI montage images outline the tumors in the lung. The white arrows denote the regions shown in 10× images. Scale bar, montage, 1 mm; 10×, 100 μm. F, Graph showing quantified levels of SLC2A1 in ADC and SCC tumors correlating with A. G, Graph showing quantified levels of SLC1A5 in ADC and SCC tumors correlating with B. H, Graph showing quantified levels of ALAS1 in ADC and SCC tumors correlating with C. I, Graph showing quantified levels of SLC48A1 in ADC and SCC tumors correlating with D. J, Graph showing quantified levels of COX5A in ADC and SCC tumors correlating with E. Protein levels were quantified, and data are plotted as mean ± SEM.

Figure 4.

The levels of representative metabolic proteins and enzymes in ADC and SCC tumors developed in KLLuc mice. A, Representative IHC images of lung tissue sections showing levels of glucose transporter SLC2A1, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. B, Representative IHC images of lung tissue sections showing levels of glutamine transporter SLC1A5, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. C, Representative IHC images of lung tissue sections showing levels of ALAS1, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. D, Representative IHC images of lung tissue sections showing levels of SLC48A1, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. E, Representative IHC images of lung tissue sections showing levels of COX5A, along with the ADC marker NKX2-1 and the SCC marker ΔNp63, from KLLuc mice at 10 weeks after the initiation of tumorigenesis via viral infection. Shown are montage and 10× images of lung tissue sections stained with DAPI (row 1), antibodies to NKX2-1 or ΔNp63 (row 2), antibodies to the indicated protein (row 3), and merged images of NKX2-1 or ΔNp63 with the indicated protein (row 4). The light green lines in DAPI montage images outline the tumors in the lung. The white arrows denote the regions shown in 10× images. Scale bar, montage, 1 mm; 10×, 100 μm. F, Graph showing quantified levels of SLC2A1 in ADC and SCC tumors correlating with A. G, Graph showing quantified levels of SLC1A5 in ADC and SCC tumors correlating with B. H, Graph showing quantified levels of ALAS1 in ADC and SCC tumors correlating with C. I, Graph showing quantified levels of SLC48A1 in ADC and SCC tumors correlating with D. J, Graph showing quantified levels of COX5A in ADC and SCC tumors correlating with E. Protein levels were quantified, and data are plotted as mean ± SEM.

Close modal

To further evaluate and compare heme metabolism and OXPHOS activities in ADC versus SCC cells, we examined four representative SCC cell lines: HCC15, NCI-H226, NCI-H520, and SW 900. In comparison, we examined three previously characterized ADC cell lines—H1299, A549, and H460—that have been shown to exhibit elevated heme synthesis, uptake, and OXPHOS activities (19). We found that the levels of heme synthesis (Supplementary Fig. S3A), heme uptake (Supplementary Fig. S3B), and OXPHOS activity (Supplementary Fig. S3C) in SCC cell lines were generally comparable with the levels in ADC cell lines, although the levels are heterogeneous in both SCC and ADC cell lines. Together, these results indicate that heme synthesis, uptake, and OXPHOS are intensified in both SDC and SCC cells to support their tumorigenic functions.

Heme sequestration effectively suppresses the growth and progression of both ADC and SCC tumors

To test the possibility of targeting heme availability for the suppression of lung tumorigenesis and progression, we took advantage of a heme-sequestering protein 2 (HeSP2, previously named HSP2), which has been shown to inhibit heme uptake, OXPHOS activity, ATP generation, and tumorigenic functions in NSCLC cell lines (19). We carried out the treatment of endogenously developed lung tumors in KLLuc mice with HeSP2. We found that HeSP2 strongly suppressed the growth and progression of endogenously developed lung tumors (Fig. 5A). Using markers for ADC (NKX2-1) and SCC (ΔNp63), we showed that the growth and progression of both ADC and SCC were suppressed by HeSP2 (Fig. 5B). Quantification of tumor lesion areas showed that the effects of HSP2 on both ADC and SCC progression were strong and statistically significant (Fig. 5C).

Figure 5.

HeSP2 effectively suppress the growth of both ADC and SCC lesions. A, Representative H&E images of lung tissue sections from pretreatment KLLuc mice, control KLLuc mice treated with saline, and HeSP2-treated KLLuc mice. The light blue lines in H&E montage images outline the tumors in the lung. The black rectangles in H&E denote the regions shown in 10×. B, Representative DAPI and IHC images of lung tissue sections from pretreatment, control and HeSP2-treated KLLuc mice showing levels of NKX2-1 and ΔNp63. Shown are montage images of lung tissue sections stained with DAPI (row 1), antibodies to NKX2-1 (row 2), and antibodies to ΔNp63 (row 3). The light green lines in DAPI montage images outline the tumors in the lung. Scale bar, montage, 1 mm; 10×, 100 μm. C, Graph showing quantified ADC and SCC lesion areas in pretreatment, control, and HeSP2-treated KLLuc mice (n = 7/group). For statistical analysis, tumor lesion areas in control were compared with HeSP2-treated lung tissues, with a Welch two-sample t test. *, P < 0.05; **, P < 0.005.

Figure 5.

HeSP2 effectively suppress the growth of both ADC and SCC lesions. A, Representative H&E images of lung tissue sections from pretreatment KLLuc mice, control KLLuc mice treated with saline, and HeSP2-treated KLLuc mice. The light blue lines in H&E montage images outline the tumors in the lung. The black rectangles in H&E denote the regions shown in 10×. B, Representative DAPI and IHC images of lung tissue sections from pretreatment, control and HeSP2-treated KLLuc mice showing levels of NKX2-1 and ΔNp63. Shown are montage images of lung tissue sections stained with DAPI (row 1), antibodies to NKX2-1 (row 2), and antibodies to ΔNp63 (row 3). The light green lines in DAPI montage images outline the tumors in the lung. Scale bar, montage, 1 mm; 10×, 100 μm. C, Graph showing quantified ADC and SCC lesion areas in pretreatment, control, and HeSP2-treated KLLuc mice (n = 7/group). For statistical analysis, tumor lesion areas in control were compared with HeSP2-treated lung tissues, with a Welch two-sample t test. *, P < 0.05; **, P < 0.005.

Close modal

Our previous studies have shown that HeSP2 strongly suppresses the growth of ADC tumor xenografts, such as H1299 and A549 tumors (19). Here, to further confirm the effect of HeSP2 on suppressing SCC tumors, we examined the effect of HeSP2 on the growth of HCC15 SCC tumor xenografts in NOD/SCID mice. Figure 6AC shows that HeSP2 strongly suppressed the growth of HCC15 tumors. The treatments of HeSP2 did not cause significant changes in body masses of mice (Fig. 6D). The measurements of OCRs (Fig. 6E) and ATP levels (Fig. 6F) in isolated tumor cells from mice showed that HeSP2 substantially reduced OCR and ATP generation in tumors.

Figure 6.

HeSP2 effectively suppresses the growth, OCR, and ATP generation in subcutaneous squamous cell carcinoma tumor xenografts. A, Average tumor volumes from mice bearing subcutaneous HCC15 tumor xenografts treated with saline (control) or HeSP2-treated tumors (n = 6/group). HeSP2 treatment (25 mg/kg i.v. every 3 days) started at 1.5 weeks after tumor implantation when tumors were palpable. B, Images showing dissected tumors from mice implanted with subcutaneous xenografts, treated with saline (control) or HeSP2. C, Average masses for control and HeSP2-treated tumors. D, Average body masses of control and HeSP2-treated mice. E, OCRs measured in cells from control and HeSP2-treated tumors. F, ATP levels measured in cells from control and HeSP2-treated tumors. Data are plotted as mean ± SEM. For statistical analysis, the levels in HeSP2-treated tumors were compared with the levels in control tumors with a Welch two-sample t test. **, P < 0.005.

Figure 6.

HeSP2 effectively suppresses the growth, OCR, and ATP generation in subcutaneous squamous cell carcinoma tumor xenografts. A, Average tumor volumes from mice bearing subcutaneous HCC15 tumor xenografts treated with saline (control) or HeSP2-treated tumors (n = 6/group). HeSP2 treatment (25 mg/kg i.v. every 3 days) started at 1.5 weeks after tumor implantation when tumors were palpable. B, Images showing dissected tumors from mice implanted with subcutaneous xenografts, treated with saline (control) or HeSP2. C, Average masses for control and HeSP2-treated tumors. D, Average body masses of control and HeSP2-treated mice. E, OCRs measured in cells from control and HeSP2-treated tumors. F, ATP levels measured in cells from control and HeSP2-treated tumors. Data are plotted as mean ± SEM. For statistical analysis, the levels in HeSP2-treated tumors were compared with the levels in control tumors with a Welch two-sample t test. **, P < 0.005.

Close modal

Additionally, we assessed the effect of HeSP2 treatment on OXPHOS in endogenously developed lung tumors in KLLuc mice by detecting the levels of OXPHOS complex subunits. We found that HeSP2 reduced the levels of OXPHOS complex subunits, including COX5A (Fig. 7A and G), COX5B (Supplementary Fig. S4A and S4E), COX7B (Supplementary Fig. S4B and S4F), ATP5F1A (Supplementary Fig. S4C and S4G), and CYCS (Supplementary Fig. S4D and S4H) in KLLuc tumors. These results indicate that HeSP2 treatment should reduce OXPHOS activity and are consistent with the effect of HeSP2 on OCR in SCC tumor xenografts.

Figure 7.

The effect of HeSP2 treatment on the levels of OXPHOS complex subunits and markers of tumor angiogenesis, vasculature, and tumor cell proliferation. A, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of COX5A. B, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of VEGFA. C, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of VEGFR1. D, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of PECAM1. E, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of CD34. F, Representative IHC images of lung tissue sections from KLLuc mice showing the levels MKI67. Shown are montage and 10× images of lung tissue sections stained with DAPI (row 1) and antibodies against the indicated protein (row 2). The light green lines in DAPI montages outline the tumors in the lung. The white rectangles denote the regions shown in 10× images. Scale bar, montage, 1 mm; 10×, 100 μm. G, Graph showing quantified levels of COX5A correlating with A. H, Graph showing quantified levels of VEGFA correlating with B. I, Graph showing quantified levels of VEGFR1 correlating with C. J, Graph showing quantified levels of PECAM1 correlating with D. K, Graph showing microvessel density quantified from IHC images of PECAM1. L, Graph showing quantified levels of CD34 correlating with E. M, Graph showing microvessel density quantified from IHC images of CD34. N, Graph showing quantified levels of MKI67 correlating with F. For statistical analysis, the levels in HeSP2-treated tumors were compared with the levels in control tumors with a Welch two-sample t test. **, P < 0.005.

Figure 7.

The effect of HeSP2 treatment on the levels of OXPHOS complex subunits and markers of tumor angiogenesis, vasculature, and tumor cell proliferation. A, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of COX5A. B, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of VEGFA. C, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of VEGFR1. D, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of PECAM1. E, Representative IHC images of lung tissue sections from KLLuc mice showing the levels of CD34. F, Representative IHC images of lung tissue sections from KLLuc mice showing the levels MKI67. Shown are montage and 10× images of lung tissue sections stained with DAPI (row 1) and antibodies against the indicated protein (row 2). The light green lines in DAPI montages outline the tumors in the lung. The white rectangles denote the regions shown in 10× images. Scale bar, montage, 1 mm; 10×, 100 μm. G, Graph showing quantified levels of COX5A correlating with A. H, Graph showing quantified levels of VEGFA correlating with B. I, Graph showing quantified levels of VEGFR1 correlating with C. J, Graph showing quantified levels of PECAM1 correlating with D. K, Graph showing microvessel density quantified from IHC images of PECAM1. L, Graph showing quantified levels of CD34 correlating with E. M, Graph showing microvessel density quantified from IHC images of CD34. N, Graph showing quantified levels of MKI67 correlating with F. For statistical analysis, the levels in HeSP2-treated tumors were compared with the levels in control tumors with a Welch two-sample t test. **, P < 0.005.

Close modal

To further ascertain the molecular mechanisms underlying the suppression of tumor growth, we examined the effects of HeSP2 on markers of tumor angiogenesis and vasculature in KLLuc mice. HeSP2 significantly reduced the levels of angiogenic factor VEGFA (Fig. 7B and H) and receptor VEGFR1 (Fig. 7C and I), as well as the endothelial cell markers PECAM1 (Fig. 7D and J) and CD34 (Fig. 7E and L). Consequently, HeSP2 strongly decreased microvessel density in treated tumors (Fig. 7K and M) calculated based on both IHC images of PECAM1 and CD34. HeSP2 also diminished tumor hypoxia, as shown by reduced levels of hypoxia markers HIF1A (Supplementary Fig. S5A and S5D) and CA9 (Supplementary Fig. S5B and S5E). These results show that HeSP2 treatment normalized tumor vasculature and alleviated tumor vasculature in KLLuc mice.

Furthermore, we detected the levels of proliferation marker MKI67 and apoptosis marker cleaved caspase-3 (CASP3). Figure 7F and N shows that HeSP2 strongly reduced the levels of Ki67. HeSP2 increased the levels of cleaved CASP3 in treated tumors (Supplementary Fig. S5C and S5F). These results show that decreased proliferation and increased apoptosis contribute to tumor suppression by HeSP2. Together, these results show that HeSP2 effectively decreased OXPHOS activity, normalized tumor vasculature, alleviated tumor hypoxia, inhibited tumor cell proliferation, and induced apoptosis, thereby leading to the suppression of ADC and SCC tumor growth and progression.

Despite improved early detection and the use of targeted therapies and immunotherapies, the lung cancer five-year survival rate remains 10% to 20% in most countries. Targeted therapies are limited by the lack of targetable genetic mutations in many patients and the development of drug resistance (34, 35). Recently, immunotherapies, including nivolumab, pembrolizumab, and atezolizumab, have been used for treating NSCLC (36). Immunotherapies have extended the median overall survival by several months in both the first-line and second-line settings in most clinical trials (35, 37, 38). Furthermore, only a fraction of patients can benefit from immunotherapy (39). For example, patients with LKB1 mutations often do not respond to immune therapy (24, 40). Clearly, identifying new molecular features of NSCLC that enable novel therapeutic strategies is of great biological and clinical importance.

Heme is a central metabolic and signaling molecule that regulates diverse processes, including transcription and cell signaling (41–43). Heme also serves as a prosthetic group in proteins and enzymes involved in oxygen utilization and metabolism (16). One crucial function of mitochondria is to carry out mitochondrial respiration/OXPHOS for ATP generation. Heme function and mitochondrial respiration are tightly linked. Multiple subunits in OXPHOS complexes II–IV contain heme. Heme also coordinates the expression and assembly of OXPHOS complexes (17). Clearly, heme possesses unique signaling and structural properties that enable it to coordinate elevated OXPHOS in cancer cells (9). Here, using the KLLuc mouse model harboring a conditional activating mutation (G12D) at the endogenous Kras locus and a conditional Lkb1 knockout, we are able to monitor the changes in the levels of enzymes and transporters involved in heme synthesis and uptake, OXPHOS, as well as those involved in the uptake and utilization of glucose and glutamine (Figs. 1 and 2; Supplementary Fig. S1). Our data showed that the levels of transporters and enzymes involved in heme acquisition and ATP generation via oxidative metabolism of glucose and glutamine all progressively increased along with lung tumorigenesis. Previously, work in the authors' lab has shown that elevated heme flux and function underlie enhanced OXPHOS and tumorigenicity of NSCLC cells (19). The data shown here indicate that elevated levels of heme and OXPHOS complex enzymes correlate with the stages of lung tumor development. Evidently, as lung tumorigenesis advances, more heme is synthesized and taken up to support OXPHOS, which leads to increased ATP generation, thereby further promoting tumor progression. Additionally, our data showed that as with intensified heme and OXPHOS function, tumor angiogenesis and hypoxia increased with lung tumorigenesis (Fig. 3).

Notably, consistent with previous studies (27), IHC showed that elevated levels of the glucose transporter GLUT1/SLC2A1 are preferentially associated with only SCC (Fig. 4A and F). However, elevated levels of heme and OXPHOS complex enzymes are associated with both ADC and SCC (Fig. 4CE and HJ; Supplementary Fig. S2), indicating that elevated heme function and OXPHOS are important for the development of both ADC and SCC. This is further supported by data from ADC and SCC cell lines. Measurements in cell lines showed that the levels of heme synthesis, uptake, and oxygen consumption fall in the comparable range in both ADC and SCC cell lines (Supplementary Fig. S3). These results together strongly support the conclusion that elevated heme and OXPHOS function are a common feature of ADC and SCC.

Importantly, we showed that heme sequestration with HeSP2 is effective at suppressing the progression of both ADC and SCC in KLLuc mice (Fig. 5). This is in complete agreement with previous studies showing that HeSP2 (HSP2) effectively suppresses the growth of NSCLC H1299 and A549 tumor xenografts in mice (19, 44). Additionally, we demonstrated that HeSP2 effectively suppresses HCC15 SCC tumor xenografts in mice (Fig. 6). Together, these results show that HeSP2 strongly suppresses the growth and progression of endogenously developed lung ADC and SCC tumors, as well as ADC and SCC tumor xenografts. Thus, targeting heme and OXPHOS via heme sequestration is a novel, effective strategy for the suppression of both ADC and SCC tumors.

To sum up, our data presented in this report provides the following new insights about NSCLC: (i) The levels of enzymes and proteins relating to heme synthesis and uptake increase progressively as lung tumorigenesis advances. (ii) The levels of enzymes relating to OXPHOS, glucose metabolism, and glutamine metabolism increase progressively during lung tumorigenesis. (iii) Angiogenesis and tumor hypoxia increase progressively during lung tumorigenesis. (iv) The increase in heme- and OXPHOS-related proteins and enzymes accompanies both ADC and SCC, and heme sequestration effectively suppresses the growth and progression of both ADC and SCC tumors. Notably, our results indicate that HeSP2 alleviates tumor hypoxia and normalizes vasculature (Fig. 7BE and HM, Supplementary Fig. S5A and S5B, S5D and S5E). Tumor hypoxia strongly affects cancer progression. It promotes several processes that are crucial for migration, invasion, metastasis, immune surveillance, and drug resistance (45–47). Tumor hypoxia generally indicates poor prognosis in human cancer. Even early-stage NSCLC exhibits substantial tumor hypoxia (48). Thus, alleviating tumor hypoxia presents a high potential to improve the outcome of NSCLC treatment (49). Given the effectiveness of HeSP2 in alleviating tumor hypoxia, it is highly likely that HeSP2 or agents with similar properties can be used in combination with chemotherapy, radiotherapy, and immunotherapy, to dramatically improve the treatment of lung ADC and SCC. The application of the strategy of heme sequestration can fundamentally advance cancer therapy.

L. Zhang reports a patent for US 2020/0361999 A1 pending; and L. Zhang is the president of HemePro Therapeutics, LLC. UT Dallas is actively managing the potential conflict of interest. No disclosures were reported by the other authors.

S. Dey: Conceptualization, data curation, formal analysis, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. A. Ashrafi: Conceptualization, data curation, formal analysis, validation, investigation, methodology, writing–original draft, writing–review and editing. C. Vidal: Data curation, formal analysis, investigation, methodology, writing–review and editing. N. Jain: Data curation, investigation, methodology, writing–review and editing. S.P. Kalainayakan: Conceptualization, data curation, validation, investigation, methodology, writing–review and editing. P. Ghosh: Conceptualization, data curation, validation, investigation, methodology, writing–review and editing. P.S. Alemi: Data curation, investigation, visualization, writing–review and editing. N. Salamat: Data curation, investigation, methodology, writing–review and editing. P.C. Konduri: Data curation, investigation, methodology, writing–review and editing. J.-w. Kim: Conceptualization, supervision, methodology, writing–review and editing. L. Zhang: Conceptualization, resources, formal analysis, supervision, funding acquisition, validation, methodology, writing–original draft, project administration, writing–review and editing.

This work was supported by Cancer Prevention and Research Institute of Texas grant RP200021.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Siegel
RL
,
Miller
KD
,
Jemal
A
. 
Cancer statistics, 2020
.
CA Cancer J Clin
2020
;
70
:
7
30
.
2.
Araujo
LH
,
Horn
L
,
Merritt
RE
,
Shilo
K
,
Xu-Welliver
M
,
Carbone
DP
. 
Cancer of the lung: non-small cell lung cancer and small cell lung cancer
. In:
Niederhuber
JE
,
Armitage
JO
,
Doroshow
JH
,
Kastan
MB
,
Tepper
JE
, editors.
Abeloff's clinical oncology
.
Philadelphia, PA
:
Elsevier
; 
2020
. p.
1108
58
.
3.
Jamal-Hanjani
M
,
Wilson
GA
,
McGranahan
N
,
Birkbak
NJ
,
Watkins
TBK
,
Veeriah
S
, et al
Tracking the evolution of non-small-cell lung cancer
.
N Engl J Med
2017
;
376
:
2109
21
.
4.
Hua
X
,
Zhao
W
,
Pesatori
AC
,
Consonni
D
,
Caporaso
NE
,
Zhang
T
, et al
Genetic and epigenetic intratumor heterogeneity impacts prognosis of lung adenocarcinoma
.
Nat Commun
2020
;
11
:
2459
.
5.
de Sousa
VML
,
Carvalho
L
. 
Heterogeneity in lung cancer
.
Pathobiology
2018
;
85
:
96
107
.
6.
Marino
FZ
,
Bianco
R
,
Accardo
M
,
Ronchi
A
,
Cozzolino
I
,
Morgillo
F
, et al
Molecular heterogeneity in lung cancer: from mechanisms of origin to clinical implications
.
Int J Med Sci
2019
;
16
:
981
9
.
7.
Han
X
,
Li
F
,
Fang
Z
,
Gao
Y
,
Li
F
,
Fang
R
, et al
Transdifferentiation of lung adenocarcinoma in mice with Lkb1 deficiency to squamous cell carcinoma
.
Nat Commun
2014
;
5
:
3261
.
8.
Ahmed
T
,
Vial
MR
,
Ost
D
,
Stewart
J
,
Hasan
MA
,
Grosu
HB
. 
Non-small cell lung cancer transdifferentiation into small cell lung cancer: a case series
.
Lung Cancer
2018
;
122
:
220
3
.
9.
Alam
MM
,
Lal
S
,
FitzGerald
KE
,
Zhang
L
. 
A holistic view of cancer bioenergetics: mitochondrial function and respiration play fundamental roles in the development and progression of diverse tumors
.
Clin Transl Med
2016
;
5
:
3
.
10.
DeBerardinis
RJ
,
Chandel
NS
. 
Fundamentals of cancer metabolism
.
Sci Adv
2016
;
2
:
e1600200
.
11.
Warburg
O
,
Wind
F
,
Negelein
E
. 
The metabolism of tumors in the body
.
J Gen Physiol
1927
;
8
:
519
30
.
12.
Hensley
CT
,
Faubert
B
,
Yuan
Q
,
Lev-Cohain
N
,
Jin
E
,
Kim
J
, et al
Metabolic heterogeneity in human lung tumors
.
Cell
2016
;
164
:
681
94
.
13.
Faubert
B
,
Li
KY
,
Cai
L
,
Hensley
CT
,
Kim
J
,
Zacharias
LG
, et al
Lactate metabolism in human lung tumors
.
Cell
2017
;
171
:
358
71
.
14.
Sotgia
F
,
Lisanti
MP
. 
Mitochondrial markers predict survival and progression in non-small cell lung cancer (NSCLC) patients: use as companion diagnostics
.
Oncotarget
2017
;
8
:
68095
107
.
15.
Jeon
JH
,
Kim
DK
,
Shin
Y
,
Kim
HY
,
Song
B
,
Lee
EY
, et al
Migration and invasion of drug-resistant lung adenocarcinoma cells are dependent on mitochondrial activity
.
Exp Mol Med
2016
;
48
:
e277
.
16.
de Montellano
PRQ
. 
Hemes in biology
. In:
Begley
TP
, editor.
Wiley encyclopedia of chemical biology
.
West Sussex
:
John Wiley & Sons, Ltd
; 
2009
. p.
240
9
.
17.
Kim
HJ
,
Khalimonchuk
O
,
Smith
PM
,
Winge
DR
. 
Structure, function, and assembly of heme centers in mitochondrial respiratory complexes
.
Biochim Biophys Acta
2012
;
1823
:
1604
16
.
18.
Smith
A
,
McCulloh
RJ
. 
Hemopexin and haptoglobin: allies against heme toxicity from hemoglobin not contenders
.
Front Physiol
2015
;
6
:
187
.
19.
Sohoni
S
,
Ghosh
P
,
Wang
T
,
Kalainayakan
SP
,
Vidal
C
,
Dey
S
, et al
Elevated heme synthesis and uptake underpin intensified oxidative metabolism and tumorigenic functions in non-small cell lung cancer cells
.
Cancer Res
2019
;
79
:
2511
25
.
20.
Ferone
G
,
Lee
MC
,
Sage
J
,
Berns
A
. 
Cells of origin of lung cancers: lessons from mouse studies
.
Genes Dev
2020
;
34
:
1017
32
.
21.
Shackelford
DB
,
Abt
E
,
Gerken
L
,
Vasquez
DS
,
Seki
A
,
Leblanc
M
, et al
LKB1 inactivation dictates therapeutic response of non-small cell lung cancer to the metabolism drug phenformin
.
Cancer Cell
2013
;
23
:
143
58
.
22.
Ji
H
,
Ramsey
MR
,
Hayes
DN
,
Fan
C
,
McNamara
K
,
Kozlowski
P
, et al
LKB1 modulates lung cancer differentiation and metastasis
.
Nature
2007
;
448
:
807
10
.
23.
Li
F
,
Han
X
,
Li
F
,
Wang
R
,
Wang
H
,
Gao
Y
, et al
LKB1 inactivation elicits a redox imbalance to modulate non-small cell lung cancer plasticity and therapeutic response
.
Cancer Cell
2015
;
27
:
698
711
.
24.
Skoulidis
F
,
Goldberg
ME
,
Greenawalt
DM
,
Hellmann
MD
,
Awad
MM
,
Gainor
JF
, et al
STK11/LKB1 mutations and PD-1 inhibitor resistance in KRAS-mutant lung adenocarcinoma
.
Cancer Discov
2018
;
8
:
822
35
.
25.
Kumar
R
,
Lovell
S
,
Matsumura
H
,
Battaile
KP
,
Moenne-Loccoz
P
,
Rivera
M
. 
The hemophore HasA from Yersinia pestis (HasAyp) coordinates hemin with a single residue, Tyr75, and with minimal conformational change
.
Biochemistry
2013
;
52
:
2705
7
.
26.
Hooda
J
,
Cadinu
D
,
Alam
MM
,
Shah
A
,
Cao
TM
,
Sullivan
LA
, et al
Enhanced heme function and mitochondrial respiration promote the progression of lung cancer cells
.
PLoS One
2013
;
8
:
e63402
.
27.
Goodwin
J
,
Neugent
ML
,
Lee
SY
,
Choe
JH
,
Choi
H
,
Jenkins
DMR
, et al
The distinct metabolic phenotype of lung squamous cell carcinoma defines selective vulnerability to glycolytic inhibition
.
Nat Commun
2017
;
8
:
15503
.
28.
DuPage
M
,
Dooley
AL
,
Jacks
T
. 
Conditional mouse lung cancer models using adenoviral or lentiviral delivery of Cre recombinase
.
Nat Protoc
2009
;
4
:
1064
72
.
29.
Zhu
Y
,
Hon
T
,
Ye
WZ
,
Zhang
L
. 
Heme deficiency selectively interferes with the Ras-MAPK signaling pathway and the expression of a subset of neuronal genes
.
Cell Grow & Diff
2002
;
13
:
431
9
.
30.
Dey
S
,
Kumari
S
,
Kalainayakan
SP
,
Campbell
J
 3rd
,
Ghosh
P
,
Zhou
H
, et al
The vascular disrupting agent combretastatin A-4 phosphate causes prolonged elevation of proteins involved in heme flux and function in resistant tumor cells
.
Oncotarget
2018
;
9
:
4090
101
.
31.
Liu
L
,
Cao
Y
,
Chen
C
,
Zhang
X
,
McNabola
A
,
Wilkie
D
, et al
Sorafenib blocks the RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell apoptosis in hepatocellular carcinoma model PLC/PRF/5
.
Cancer Res
2006
;
66
:
11851
8
.
32.
Hsieh
MH
,
Choe
JH
,
Gadhvi
J
,
Kim
YJ
,
Arguez
MA
,
Palmer
M
, et al
p63 and SOX2 dictate glucose reliance and metabolic vulnerabilities in squamous cell carcinomas
.
Cell Rep
2019
;
28
:
1860
78
.e9.
33.
Momcilovic
M
,
Jones
A
,
Bailey
ST
,
Waldmann
CM
,
Li
R
,
Lee
JT
, et al
In vivo imaging of mitochondrial membrane potential in non-small-cell lung cancer
.
Nature
2019
;
575
:
380
4
.
34.
Santarpia
M
,
Karachaliou
N
,
Rosell
R
. 
Beyond platinum treatment for NSCLC: what does the future hold?
Expert Rev Anticancer Ther
2017
;
17
:
293
5
.
35.
Barlesi
F
,
Mazieres
J
,
Merlio
JP
,
Debieuvre
D
,
Mosser
J
,
Lena
H
, et al
Routine molecular profiling of patients with advanced non-small-cell lung cancer: results of a 1-year nationwide programme of the French Cooperative Thoracic Intergroup (IFCT)
.
Lancet
2016
;
387
:
1415
26
.
36.
Ettinger
DS
,
Wood
DE
,
Aisner
DL
,
Akerley
W
,
Bauman
J
,
Chirieac
LR
, et al
Non-small cell lung cancer
,
Version 4.2018, NCCN Clinical Practice Guidelines in Oncology
. https://www.nccn.org/patients/guidelines/content/PDF/lung-early-stage-patient.pdf; 
2018
.
37.
Reck
M
,
Rodriguez-Abreu
D
,
Robinson
AG
,
Hui
R
,
Csoszi
T
,
Fulop
A
, et al
Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer
.
N Engl J Med
2016
;
375
:
1823
33
.
38.
Martinez
P
,
Peters
S
,
Stammers
T
,
Soria
JC
. 
Immunotherapy for the first-line treatment of patients with metastatic non-small cell lung cancer
.
Clin Cancer Res
2019
;
25
:
2691
8
.
39.
Haslam
A
,
Prasad
V
. 
Estimation of the percentage of US patients with cancer who are eligible for and respond to checkpoint inhibitor immunotherapy drugs
.
JAMA Netw Open
2019
;
2
:
e192535
.
40.
Mograbi
B
,
Heeke
S
,
Hofman
P
. 
The importance of STK11/LKB1 assessment in non-small cell lung carcinomas
.
Diagnostics (Basel)
2021
;
11
:
196
.
41.
Barr
I
,
Smith
AT
,
Chen
Y
,
Senturia
R
,
Burstyn
JN
,
Guo
F
. 
Ferric, not ferrous, heme activates RNA-binding protein DGCR8 for primary microRNA processing
.
Proc Natl Acad Sci U S A
2012
;
109
:
1919
24
.
42.
Kuhl
T
,
Imhof
D
. 
Regulatory Fe(II/III) heme: the reconstruction of a molecule's biography
.
ChemBioChem
2014
;
15
:
2024
35
.
43.
Mense
SM
,
Zhang
L
. 
Heme: a versatile signaling molecule controlling the activities of diverse regulators ranging from transcription factors to MAP kinases
.
Cell Res
2006
;
16
:
681
92
.
44.
Ghosh
P
,
Guo
Y
,
Ashrafi
A
,
Chen
J
,
Dey
S
,
Zhong
S
, et al
Oxygen-enhanced optoacoustic tomography reveals the effectiveness of targeting heme and oxidative phosphorylation at normalizing tumor vascular oxygenation
.
Cancer Res
2020
;
80
:
3542
55
.
45.
Lee
CT
,
Boss
MK
,
Dewhirst
MW
. 
Imaging tumor hypoxia to advance radiation oncology
.
Antioxid Redox Signal
2014
;
21
:
313
37
.
46.
Gillies
RJ
,
Verduzco
D
,
Gatenby
RA
. 
Evolutionary dynamics of carcinogenesis and why targeted therapy does not work
.
Nat Rev Cancer
2012
;
12
:
487
93
.
47.
McDonald
PC
,
Chafe
SC
,
Dedhar
S
. 
Overcoming hypoxia-mediated tumor progression: combinatorial approaches targeting pH regulation, Angiogenesis and Immune Dysfunction
.
Front Cell Dev Biol
2016
;
4
:
27
.
48.
Le
QT
,
Chen
E
,
Salim
A
,
Cao
H
,
Kong
CS
,
Whyte
R
, et al
An evaluation of tumor oxygenation and gene expression in patients with early stage non-small cell lung cancers
.
Clin Cancer Res
2006
;
12
:
1507
14
.
49.
Salem
A
,
Asselin
MC
,
Reymen
B
,
Jackson
A
,
Lambin
P
,
West
CML
, et al
Targeting hypoxia to improve non-small cell lung cancer outcome
.
J Natl Cancer Inst
2018
;
110
.

Supplementary data