Beyond Binary: Mapping the Evolution of Melanoma Across a Discrete Gradient of Acquired Chemoresistance

Beyond Binary: Mapping the Evolution of Melanoma Across a Discrete Gradient of Acquired Chemoresistance | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Beyond Binary: Mapping the Evolution of Melanoma Across a Discrete Gradient of Acquired Chemoresistance Emily Wang, Annabelle Hauss, Hiruni Lokuyaddehige, Amy Nielsen, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8929076/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Models of cancers with acquired MultiDrug Resistance (MDR) commonly employ a binary state that compares MDR to treatment naïve cohorts. While convenient, this paradigm also oversimplifies the dynamic process of acquiring MDR as cellular processes (and corresponding therapies targeting them) could perform differently at each intermediate stage of resistance. However, comparisons of discrete levels of chemoresistance, particularly with regards to carbohydrate transport or transiently expressed genes, remain limited. Here we characterize a B16 melanoma cell panel comprising a gradient of doxorubicin resistance (1 nM to 1 µM). Across the MDR gradient, we observed minimal changes in class I glucose transporter (GLUT) expression and reduced carbohydrate influx, despite increases in P-glycoprotein (P-gp) efflux. This suggests enhanced P-gp efflux is not compensated by GLUT-mediated influx, but rather, attenuated carbohydrate metabolism. Indeed, RNA-seq revealed decreases in glycolytic enzymes alongside additional interferon-stimulated genes (including Isg15 and Bst2 ) with transient differential expression only at intermediate levels of MDR. Overall, this study provides proof-of-principle that selection of a particular level of MDR is non-trivial, particularly for studies involving the transiently expressed genes or carbohydrate processing, and might explain why there remains significant debate over which in-vitro level of MDR is optimally prognostic of in-vivo performance. Biological sciences/Cancer Health sciences/Oncology P-gp chemoresistance doxorubicin melanoma GLUT carbohydrate transport Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction It is well-established that treatment of cancers with conventional cytotoxic chemotherapies results in recurrence of multidrug-resistant (MDR) phenotypes. 1 These MDR populations limit therapeutic efficacy through a range of genetic and epigenetic mechanisms, including elevated drug efflux, reduced drug influx, enzymatic drug inactivation, remodeled DNA repair systems, and irregular carbohydrate metabolism. 2 – 4 Elevated drug efflux via transport proteins, particularly the ATP-binding cassette (ABC) superfamily, has been well-characterized, 5 prompting interest in targeting these transporters through specific inhibition. 6 Of these, permeability glycoprotein (P-gp or ABCB1) is the most well-studied and is established to smoothly increase as a function of contact (both in terms of duration and concentration) with a range of cytotoxic chemotherapies, broadly, across many cancer types. 7 These observations have motivated extensive efforts to recapitulate MDR in controlled in-vitro settings. While numerous protocols now exist for inducing chemoresistance, 8 there remains significant debate over which in-vitro resistance level for a given cancer type is best to model in-vivo MDR cancers. Some studies demonstrate that typical MDR cell lines overstate what is found for MDR cancers in-vivo, whereas others contend that resistance presents across a spectrum that is typically 2–5 fold, but up to 16 fold, that inherent to the progenitor cancer type. 9 In practice, in-vitro models of MDR cancers typically study a small number of arbitrarily-defined resistance states, with many reducing the complex process of acquiring MDR to a binary resistant vs non-resistant comparison. 10 While such approaches have yielded valuable insights, this over-discretization can obscure early or transient resistance-associated phenotypes that may arise during intermediate stages of chemoresistance development in a typical in-vitro model with a high-level of resistance. Intentionally characterizing chemoresistance across a broader, stepwise range of resistance levels could help resolve these gradual or non-linear changes, while also capturing clinically relevant heterogeneity in drug biodistribution and intratumoral susceptibility to resistant phenotypes. This heterogeneity in therapeutic response is shaped by the plasticity of cellular processes that enable cancers to survive under sustained cytotoxic stress. Among these processes, metabolic regulation and membrane transport play central roles by governing energy availability, nutrient uptake, and drug efflux, and are themselves highly variable across tumor types and cellular contexts. For example, the Warburg effect, first characterized in the 1920s, describes cancer phenotypes with irregular, and often up-regulated, glucose transport and metabolism biased towards anerobic glycolysis. 11 These irregularities are now established to play a central role in acquired MDR as well as the persistence and progression of many cancers that manifest as solid tumors, ranging from glioblastoma 12 to melanoma. 13 , 14 This metabolic phenotype has inspired several emerging therapies which seek to exploit its distinctive overreliance on glucose. 15 , 16 New chemical entities targeting critical enzymes in glucose metabolism such as hexokinase, 17 phosphofructokinase, 18 and pyruvate kinase, 19 among others, have been developed in recent years and continue to improve in specificity and potency. Likewise, many new cancer drug delivery strategies target the overexpression of carbohydrate transporters, most commonly the glucose transporter (GLUT) family, such as in a glyco-conjugate form to enhance cell entry 20 – 23 or as a GLUT inhibitor to disrupt energy production and proliferation. 24 – 27 That said, to-date most GLUT inhibitors remain in preclinical development with only a handful in clinical trials and none that are currently FDA-approved cancer drugs. 28 – 30 Further, manipulating or otherwise targeting carbohydrate metabolism for therapeutic benefit presents a challenging moving target. Cancers are highly heterogeneous and their metabolomic profiles evolve as they develop resistance to therapeutic interventions. 31 , 32 Emerging glyco-targeted therapies will most often begin clinical investigations where there is a clear unmet need: MDR cancers that have failed multiple established therapies. 33 , 34 Thus, a more nuanced understanding of how glycometabolic shifts occur across the full spectrum of acquired MDR could inform the design of more effective therapies (Fig. 1 a). In particular, it could highlight nonlinear target expression across the MDR gradient which would improve the predictive accuracy of translating in-vitro results to in-vivo efficacy for cancers that persist across variable levels of drug resistance. However, to our knowledge, the acquisition of an MDR phenotype remains to be examined with such granularity that several intermediate resistance levels are characterized concurrently with treatment naïve and high-level MDR cells from the same line. As a proof-of-concept starting point to address this, we characterize a melanoma cell line as it acquires stable pharmacodynamic tolerance across incremental concentrations of a cytotoxic chemotherapeutic focusing on changes in efflux pumps, GLUTs, and functional carbohydrate processing (Fig. 1 b). This uniquely granular approach could be a useful system as it encompasses a wide range of clinically relevant resistance levels, while also enabling characterization of both transient and persistent transporter, and metabolic adaptations, that arise during the progression of chemoresistance. Results Establishing a discrete gradient of chemoresistance To create incremental levels of pharmacodynamic tolerance, we used a B16 melanoma cell line and the anthracycline doxorubicin (DOX). The B16 line was chosen to expand on, and give more granularity to, the classic studies done by the Parmiani lab, who similarly subjected B16 cells to a range of DOX concentrations (0–860 ng/mL) and characterized changes in drug sensitivity, DNA content, cross-resistance profiles, and in-vivo tumorigenicity. 35 , 36 We chose the Luc2 version of the B16 line for future in-vivo bioluminescence imaging as this could be used as a model of drug resistance that other drugs could be validated against in-vivo. DOX was chosen due to its established ability to stimulate P-gp expression and widespread affinity for multiple members of the common ABC efflux transporter family. 8 , 37 , 38 To establish an MDR gradient of the parent B16 line, we administered doxorubicin (DOX) at incrementally increasing concentrations, starting from 1 nM and doubling the dose at every stable passage, up to 1 µM (rounding at the 500 nM dose). Once we had created the gradient of MDR melanoma, we likewise observed a gradient of functional resistance, reflected by differences in cell viability following 48 h exposure to a high (5 µM) concentration of DOX (Fig. 2 a; Supplementary Fig. 1a ). As expected, DOX sensitivity decreased with increasing levels of pre-established resistance. Viability was comparable to PBS alone in the 500 nM and 1000 nM samples. We also evaluated DOX sensitivity in parent and 1000 nM-resistant B16 cells across a range (0–10 µM) of DOX concentrations (Fig. 2 b). After 48 h, the parental cells showed progressively lower viability with higher DOX concentrations, whereas the resistant line maintained stable viability. Furthermore, it has historically been demonstrated that enhanced P-gp-mediated drug efflux is at least one mechanism that lowers intracellular DOX accumulation in MDR B16 melanoma, 39 and this effect has been quantified with more modern flow cytometry analysis exploiting the intrinsic fluorescence of the anthracycline (Fig. 2 c; Supplementary Fig. 1b ). 40 , 41 Parental B16 cells showed higher DOX fluorescence than the 1000 nM-resistant B16 population following 30 min incubation with 10 µM DOX, indicating greater drug retention. Pre-treatment with verapamil (VPM), a competitive inhibitor of P-gp, 42 moderately increased DOX retention in the resistant cells. Assessing broader phenotypic differences across the MDR gradient by bright-field microscopy, we also observed increasing heterogeneity with higher DOX resistance, consistent with previous binary investigations of MDR B16 cells that also detailed the procedure of their generation (Fig. 2 d; Supplementary Fig. 1c ). 43 These morphological changes were further supported by increased granularity observed by flow cytometry as well ( Supplementary Fig. 1d ). GLUT expression is conserved across the MDR gradient With DOX sensitivity quantified across the MDR gradient, we then proceeded to characterize carbohydrate transport and drug efflux proteins as a function of acquired chemoresistance. Here, we primarily focused on class 1 GLUTs (family members 1–4) due to their established variability across many cancers, including melanomas. 15 , 44 We ran western blots to quantify relative P-gp and GLUT expression across the MDR gradient (Fig. 3 a). P-gp expression directly correlated with increasing DOX resistance, as expected. In contrast, GLUT1 and GLUT3 remained relatively stable across the entire gradient, while GLUT2 was not detectable. GLUT4 exhibited minimal overall expression albeit with a moderate increase at the 1 µM resistance level. Characterizing this in more detail, we also demonstrated GLUT4 insulin responsiveness which disappeared after longer DOX exposure, suggesting a temporary adaptive response ( Supplementary Fig. 2 ). Given the possibility that the functional efficiency of GLUT influx could involve intracellular redistribution rather than overall expression changes, we also used confocal microscopy to visualize GLUT localization (Fig. 3 b; Supplementary Fig. 3 ). Imaging confirmed upregulation of P-gp, though there were minimal differences in GLUTs 1, 3, and 4 between the parental and chemoresistant populations, with no considerable changes in localization. Complementing the proteomic analyses, we evaluated Slc2a (GLUT) transcript levels for GLUTs 1–4 by RT-qPCR to investigate whether regulatory changes at the mRNA level might occur independently of detectable changes in protein expression (Fig. 3 c; Supplementary Fig. 4 ). Changes in Slc2a1, Slc2a3 , and Slc2a4 (GLUTs 1, 3, and 4), remained modest, and correlated well with the microscopy and western blot characterization. As expected, Slc2a2 (GLUT2) was undetectable. Given that GLUT expression was conserved at both the transcriptomic and proteomic levels, we next asked whether functional glucose uptake was similarly maintained using the fluorescent glucose analog 2-NBDG. To isolate GLUT-mediated 2-NBDG uptake, we also examined pre-treatment with the pan-GLUT inhibitor DRB18 shown to inhibit GLUTs 1, 2, 3, and 4, 24 before adding 2-NBDG for 10 and 30 minutes. DRB18 significantly inhibited glucose uptake for both samples, though its effects were less pronounced by the 30 min timepoint (Fig. 3 d; Supplementary Fig. 5 ). Regardless, B16 cells resistant to 1000 nM DOX exhibited reduced glucose uptake compared to treatment-naïve cells at both time points, even with elevated concentrations of 2-NBDG ( Supplementary Fig. 6 ). To determine whether uptake was be influenced by nutrient availability, we subjected both parental and 1000 nM-resistant populations to varying glucose starvation periods (0–16 h) prior to 2-NBDG exposure. However, minimal differences in 2-NBDG uptake were detected within either group (Fig. 3 e; Supplementary Fig. 7 ). With clear alterations in P-gp efflux transporter expression and glucose uptake but limited changes in glucose transporter levels, we next sought to more broadly account for this difference in efflux relative to influx across the MDR gradient. To do this we used RNA-seq to determine transcriptional adaptations that might not have otherwise been captured by our targeted analyses. Here, in addition to the established gradient, we also included a 1000 nM-resistant sample that subsequently had DOX withheld from its culture media for two weeks (denoted as 1000X) to examine early regulatory changes following treatment withdrawal. Overall, our RNA-seq data was consistent with our previous findings: we observed minimal changes in the GLUT / Slc2a family and prominent increases in P-gp / Abcb1 isoforms that directly correlated to increasing chemoresistance (Fig. 3 f and 3 g). Of note, overexpression of Abcb1 isoforms persisted following DOX withdrawal, and none of the other ABC efflux pumps had expression correlating (either positively or negatively) with MDR level ( Supplementary Fig. 8 ). Mapping the transcriptional landscape across the MDR gradient Examining the RNA-seq data also provided a more comprehensive view into both correlative and transient transcriptomic changes that occur across the MDR gradient. Interestingly, distinct gene expression profiles emerged at each discrete resistance level (Fig. 4 a; Supplementary Fig. 9) . As expected, the early-stage resistant samples (16–128 nM) appeared to show relatively subtle transcriptional shifts, whereas more pronounced divergence became evident at higher resistance (256–1000 nM). Comparing each resistance level directly to the treatment-naive parent (0 nM) control revealed a general increase in the number of significantly upregulated and downregulated genes with increasing DOX resistance. However, the magnitude of these changes, and even the genes themselves, did not all directly correlate to DOX concentration. In contrast to P-gp expression, which remained consistently upregulated relative to parent through 1000 nM resistance (Fig. 4 b), many changes occurred transiently or irregularly across the gradient. The 0 vs 256 nM treatment comparison showed the highest total number of differentially expressed genes, while the 0 vs. 1000 nM comparison exhibited the greatest number of significantly upregulated genes (Fig. 4 c; Supplementary Fig. 10 ). Hierarchical clustering based on pairwise Euclidean distances demonstrated strong in-group similarity and clear segregation of samples by resistance level (Fig. 4 d). Summarizing these findings by Principal Component Analysis (PCA) illustrated that low to intermediate resistance populations (16–256 nM) exhibited greater transcriptional similarity to each other relative to highly resistant groups (500–1000X), with the 500 nM level of resistance appearing distinctively isolated, clustering tightly within themselves but not with neighboring resistance levels (Fig. 4 e). Notably, PC1 scores positively correlated with DOX resistance (r = 0.829), and loadings analysis identified Abcb1a as the strongest contributor to PC1, supporting resistance level as a major source of transcriptional variance lead by the well-established positive correlation with Abcb1 ( Supplementary Fig. 11 ). That said, noticeable separation emerged starting from the 64 to 128 nM transition, with 500 nM and higher resistance levels occupying distinct regions of the PCA space. Interestingly, the 1000X samples clustered distinctly from the 1000 nM samples and appeared marginally closer to the 0 nM group, possibly reflecting stabilization or partial reversion of the resistant phenotype following initial DOX withdrawal. Glycolytic suppression and transient viral response activation across the MDR gradient To further explore the transcriptional landscape associated with chemoresistance, we next investigated how gene ontologies, specifically those related to metabolism, are modulated across the MDR gradient. Building on our earlier observation that functional glucose uptake is reduced, despite relatively stable GLUT ( Slc2a ) expression and enhanced efflux, we examined the expression of glycolytic pathway components. Here, the rationale is that a slower pathway could account for the difference in influx and efflux. We detected significant downregulation in several key enzymes, especially in the 1000X sample (which had been taken off DOX treatment for two weeks), suggesting a suppression of glycolytic activity following prolonged drug exposure and/or withdrawal (Fig. 5 a; Supplementary Table 3 ). These glycolytic changes were further visualized in a combined heatmap with class I Slc2a genes, which confirmed notable reductions in glycolytic enzyme expression levels and similar behavior in Slc2a1 (Fig. 5 b). To contextualize these findings within a metabolic framework, we mapped gene expression onto the KEGG glycolysis pathway (Fig. 5 c). This revealed a broad decrease in glycolytic enzyme expression that was most pronounced in the 0 vs 1000X comparison, with other related pathways relatively unaffected. Complementing this, gene ontology (GO) analysis provided significance in the GO terms “response to virus” (GO:0009615) and “defense response to virus” (GO:0051607) in all comparisons to parent (0 nM), with the exception of 0 vs 1000X (Fig. 5 d). Combining significant genes in these two sets highlighted a noisy, but general trend of upregulation of viral response genes along the MDR gradient (Fig. 5 e and 5 f). Interestingly (and contributing to this noise), many of these increases were transient; some of the most pronounced differences occurred at intermediate resistance levels (128–500 nM) and became downregulated in the 1000 nM or 1000X samples. Their transient nature could imply that the response decreases as other chemoresistance mechanisms (like Abcb1 –mediated drug efflux) begin to dominate. DISCUSSION In our study, DOX resistance was conferred to parent B16 cells via incremental doubling of the DOX concentration for each successive passage ranging from 1 nM to 1 µM. Examining variable levels of resistance concurrently could better mimic clinical applicability, as chemotherapeutic biodistribution is unequal among tissue types, implying that various locations of a metastasized cancer are likely exposed to, and ultimately adapt to resist, equally variable concentrations of drug. 45 We selected the 0, 16, 64, 128, 256, 500, and 1000 nM resistance levels for our downstream analyses as considerable changes first occurred somewhere between the 64 and 128 nM resistance levels (Fig. 4 e). Cell viability assays confirmed incremental resistance, and we observed that cells resistant to higher DOX concentrations also had longer doubling times. This aligns with the observed changes in morphology; in particular, the observed cellular hypertrophy and flattening in the 1000 nM-resistant cells are suggestive of a senescent-like phenotype. 46 , 47 Further, the increased granularity may be associated with increased lysosomal and autophagic vesicle content, 48 and is enhanced by increased pigmentation due to melanin, which has been implicated in helping to attenuate the effect of DOX. 49 , 50 With the chemoresistance gradient established, we next characterized the relationship between DOX resistance and glucose transporter (GLUT) expression. Elevated expression of permeability glycoprotein (P-gp), a key mediator of MDR, imposes considerable metabolic stress due to its ATP-dependent efflux activity. 51 In turn, cells may require enhanced energy-producing pathways, such as glycolysis, to sustain chemoresistance. 51 – 54 Although many studies have examined altered glucose transport and metabolism in cancers, 55 few have explored its relationship with discrete variable levels of chemoresistance, and none have investigated how GLUT activity and expression changes across incremental levels of chemoresistance. Given P-gp’s increased demand for ATP, we were surprised to find that the expression of any given GLUT family member remained largely unchanged across the MDR gradient, despite confirmed increases in P-gp expression and activity. These results were consistent across imaging, proteomic, and transcriptomic analyses. Among the class I GLUTs, GLUT1 and GLUT3 remained abundant in expression, while GLUT2 was undetectable, aligning with prior characterizations across multiple melanocytic lesions. 44 While GLUT4 generally exhibited lower expression relative to GLUTs 1 or 3, we did note a transient upregulation of GLUT4 at the 1 µM-resistance level, with the previously well-established link between GLUT4 expression and insulin exposure retained. 56 This transient increase in GLUT4 may reflect a brief supplementation in cellular energy via insulin-dependent glucose uptake, though its subsequent downregulation suggests that this pathway is unsustainable or inefficient under prolonged chemotherapeutic pressure. This variability may also reflect the inherent heterogeneity of melanoma. Previous studies have shown that while most melanomas lack strong GLUT4 expression, with near complete absence for benign nevi, a small subset of melanomas (7.8%) exhibit moderate to strong GLUT4 expression. 57 We consider this likely heterogeneity a fundamental limitation of our current proof-of-concept study and something that should be examined in more detail. Regardless, to supplement our expression-based analyses, we evaluated functional GLUT activity through glucose uptake, which revealed decreased uptake in 1000 nM-resistant cells relative to parent. Pre-incubation with the pan-GLUT inhibitor DRB18 24 competitively inhibited uptake in both MDR and parent lines, though its effects became less pronounced with time. This may reflect that some 2-NBDG enters through GLUT-independent pathways, 58 or that the inhibitory effect is eventually overcome. Further, while prior studies have shown enhanced GLUT expression with glucose deprivation, 23,59 we did not find glucose deprivation to appreciably affect glucose uptake in either the parental or resistant line, although uptake remained reduced in the 1000 nM-resistant cells relative to parent across all conditions tested. Taken together, this suggests that MDR B16 cells acquire altered phenotypes in which glucose accumulation and general influx are repressed. This is also consistent with the reduced proliferation rates we previously observed and decreased membrane permeability reported in MDR contexts. 60 , 61 Broadening our analysis with RNA-seq across the MDR gradient provided a comprehensive view of DOX-induced changes throughout the entire transcriptome. These results confirmed minimal changes in Slc2a (GLUT) expression and significant increases in Abcb1 (P-gp) expression towards higher levels of chemoresistance. Notably, the trends in expression of the two murine P-gp isoforms, Abcb1a and Abcb1b , seemed to switch past the 128 nM resistance level with the transcripts per million (TPM) abundance of the Abcb1a isoform markedly surpassing Abcb1b at the 1000 nM-resistance level ( Supplementary Fig. 12 ). This is consistent with a similar phenomenon reported for J774.2 cells by Lothstein et al, who observed that while Abcb1b expression dominates during early levels of vinblastine resistance, it declines as drug concentration increases, coinciding with a rise in Abcb1a expression. 62 It was also found that these isoforms are independently regulated: Abcb1b becomes overexpressed primarily through gene amplification, whereas Abcb1a is induced via transcriptional activation. 63 Together, this suggests that for mouse model systems Abcb1a may act as an inducible backup, becoming engaged only under extreme drug pressure when Abcb1b -mediated efflux is no longer sufficient. Investigating broader transcriptional adaptations beyond Slc2a and Abc transporters via pathway enrichment and GO analyses revealed notable decreases in a group of glycolytic enzymes with increasing levels of chemoresistance or DOX withdrawal from 1000 nM-resistant MDR cells. This reduction, coupled with the slowed growth and reduced uptake observed in MDR cells, may reflect the ability of DOX and other genotoxic agents to suppress glycolytic flux and promote cellular senescence, both of which can contribute to long-term metabolic reprogramming. 46 , 64 , 65 As DOX targets DNA replication, cancers that slow cell cycle progression likely allow more time for MDR mechanisms (like drug efflux) to take place, thereby enhancing survival. In addition to these metabolic changes, we also detected transient increases in antiviral response genes, some of which diminished upon DOX withdrawal, likely reflecting an adaptive defense against DOX’s DNA-intercalating mechanism of action. 66 Among these were many interferon-stimulated genes, which have been implicated in resistance to DNA damage and genotoxic therapy-induced senescence. 67 , 68 For example, the progressive upregulation of Ifit1 and Ifit3 are implicated in mechanisms that block apoptosis, and disrupting them can restore chemosensitivity. 69 , 70 However, they only exhibited pronounced increases at the higher (500 and 1000 nM) resistance levels. While this would indicate an attractive target in lab cell lines with corresponding higher resistance levels, the increase is less pronounced at lower (and perhaps more clinically relevant) levels of drug resistance. More importantly, the most transient induction of the genes included Isg15 and Bst2 , which provided maximal expression centered on the 256 nM resistance level, smoothly decreasing for both lower and higher levels of resistance. This is important because BST2 has been proposed as a druggable target 71 and our experiments demonstrate that the potential increased therapeutic value of a BST2-targeted drug in MDR melanoma would be missed if a simple binary comparison of treatment naïve and MDR melanoma was used as a preclinical model. We also detected significant transient increases in Samhd1 (associated with DOX resistance), as well as Tap1/2 and Psmb8/9 reflecting transient activation of antigen-processing pathways and immunoproteasome-mediated proteostasis under extreme genotoxic and oxidative stress ( Supplementary Fig. 13 ). 72 – 74 Overall, our analyses offer new insights into the stress-adaptive mechanisms melanoma cells may utilize to enhance survival under chemotherapeutic stress and highlight several adaptations that are non-linear with respect to the level of chemoresistance. This is important given the discrepancies in clinically relevant MDR levels and those typically used in-vitro for drug discovery research. While long-term functional resistance appears largely driven by P-gp expression, which directly scales with DOX concentration, it is not accompanied by a compensatory rise in glucose influx via GLUTs. Rather, MDR phenotypes are accompanied by reduced glucose uptake and glycolytic enzyme expression, suggesting a shift away from canonical Warburg metabolism and introducing a more nuanced facet of chemoresistance-associated metabolic reprogramming. Conclusion In summary, we have established that a panel of B16 melanoma, with incremental discrete concentrations of pharmacodynamic tolerance to doxorubicin, has both correlative and transient changes in phenotype yet minimal changes in GLUT expression. After confirming variable levels of resistance through viability assays, we examined functional glucose uptake along with transport protein expression, specifically class I GLUTs and P-gp. Here, we observed minimal changes in the expression of GLUTs despite reduced glucose uptake, along with the expected increase in P-gp expression across the MDR gradient. To expand beyond this group of transporters, we ran RNA-seq analysis on the MDR gradient, where only Abcb1 (P-gp) isoforms demonstrated noticeable change among the ABC transporters. Other notable changes included decreases in glycolytic enzymes and pronounced, yet transient, increases in viral response-related genes. Overall, our study contributes insight into the evolution of cancer cells as they progress towards MDR phenotypes. That said, this particular study should be understood to have two primary limitations. First, DOX is not the most clinically relevant chemotherapeutic for melanoma; rather, we chose DOX due to its well-documented ability to induce an MDR phenotype, enabling direct comparison with a broad body of prior research. Second, our experiments were performed in-vitro at the bulk cell-culture level, which cannot mimic the full complexity or heterogeneity of the tumor microenvironment. Indeed, many factors such as hypoxia, nutrient deprivation, varying pH, and immune infiltration can affect drug response, metabolism, and transporter activity, 75,76 and this heterogeneity could be a major driving force in melanoma. 77 To address this, in the future this type of study could be expanded to include single-cell analysis of tumor isolates generated in-vivo. Regardless, we expect our initial work here will aid in the development of anti-cancer drugs that target carbohydrate influx (or efflux) mechanisms and inform the rational for in-vitro testing of new drugs aimed at cancers that have failed multiple rounds of chemotherapies. Methods Materials The following biologics, chemicals, and instruments were used. For cell culture: B16-F10-Luc2 (ATCC), DMEM (VWR), HI-FBS (VWR), Blasticidin (Invivogen), Normocin (Invivogen), Trypsin-EDTA (ThermoFisher), doxorubicin (BroadPharm). For viability assays: Resazurin (Sigma-Aldrich), DMEM without phenol red (VWR), L-glutamine (ThermoFisher), BioTek Synergy LX microplate reader (BioTek). For flow cytometry: Verapamil (Ambeed), 2-NBDG (ThermoFisher), DRB18 (Sigma-Aldrich), DMEM without glucose (ThermoFisher), Glucose solution (ThermoFisher), Attune NxT (ThermoFisher). For western blots: 1% Triton-X Lysis buffer (ThermoFisher), protease inhibitor cocktail (Thermofisher), BCA assay (ThermoFisher), electrophoresis equipment (BioRad), protein ladder (ThermoFisher), PVDF membranes (BioRad), 4X Laemmli buffer (BioRad), nonfat dried milk (RPI), insulin solution (ThermoFisher), primary and secondary antibodies (abcam), HRP substrate (ThermoFisher), BioRad ChemiDoc MP (BioRad). For immunofluorescence: 22 x 22 mm coverslips (Fisher Scientific), Cytofix / Cytoperm kit (BD Biosciences), BSA (Glentham Life Sciences), primary and secondary antibodies (abcam), Phalloidin-iFluor 594 (AAT Bioquest), Fluoromount-G with DAPI (SouthernBiotech), ZEISS LSM 710 confocal microscope (ZEISS), BioTek Lionheart FX (BioTek). For RT-qPCR: Cells-to-Ct kit (ThermoFisher), primers (ThermoFisher), nuclease-free water (ThermoFisher), RNAseZap (ThermoFisher), DNAZap (ThermoFisher), QuantStudio 7 (ThermoFisher). Cell culture The B16-F10-Luc2 (B16) cell line was purchased through ATCC and cultured based on manufacturer instructions. The B16 cells were grown in complete culture media containing DMEM with 4.5 g/L glucose, 4 mM L-glutamine, and 110 mg/L sodium pyruvate, and supplemented with 10% (v/v) FBS, 10 µg/mL Blasticidin, and 100 µg/mL Normocin. Media was changed every 2–4 days. Cells were passaged using 0.25% Trypsin-EDTA solution upon reaching 80% confluence and 1–2 million cells were seeded into a new T150 flask. The chemotherapeutic doxorubicin (DOX) was used to create the chemoresistance gradient consisting of 1, 2, 4, 8, 16, 32, 64, 128, 256, 500, and 1000 nM DOX concentrations. In brief, 1 nM DOX was added to the growth media of parent cells, and the concentration was doubled at each stable passage until the 1000 nM resistance level was reached. For all downstream experiments, cultures were incubated in their respective DOX-resistance concentrations for at least two passages. All cultures were maintained in a 37 ℃, 5% CO 2 incubator. Tolerance to DOX exposure Cell viability following DOX exposure was assessed using a resazurin assay. To confirm incremental resistance to DOX, B16 cells resistant to 0 (parent), 16, 64, 128, 256, 500, and 1000 nM DOX were seeded into a 96-well plate at 20,000 cells/well in complete media and allowed to adhere for 24 h. DOX was then added to each sample, reaching a final concentration of 5 µM, and these experiments were run alongside corresponding control wells that received PBS only. After 48 h, the media was changed to phenol red-free DMEM, and resazurin solution (0.1 mg/mL) was added to 10% of the final well volume. To test resistance capacity across varying doses of DOX, parent B16 and 1000 nM-resistant B16 were seeded into a 96-well plate at 20,000 cells/well. After 24 h, both cell types were given a range of DOX concentrations (0.05, 0.1, 0.5, 1, 5, and 10 µM), along with PBS as vehicle control. Cells were similarly incubated for 48 h before the media change and addition of resazurin. All plates were incubated with resazurin for 4 h before taking absorbance readings at 570 and 600 nm, which correspond to the absorbance of resazurin and resazurin converted to resorufin by live cells respectively. Cell viability was determined by calculating the difference between the 570 nm and 600 nm reads, then subtracting out the background absorbance of wells with resazurin alone (no cells). Data was normalized relative to cells receiving vehicle control to yield % viability. Western blot B16 cells resistant to 0 (parent), 16, 64, 128, 256, 500, and 1000 nM DOX were grown in separate wells in 6-well plates. Upon reaching 80% confluency, the cells were scraped in 1% Triton-X lysis buffer with protease inhibitors, and a BCA assay was used to standardize protein concentration used from each lysate. Samples were then loaded on a 10% cross-linked gel at 25 µg/well to be separated by SDS-PAGE. Then, the gel was electrically transferred to a PVDF membrane, and the membrane was blocked in 5% nonfat milk overnight at 4°C. Primary rabbit anti-mouse antibodies against GLUT1, GLUT2, GLUT3, GLUT4, P-gp, and β-actin were incubated for 2 h at RT and HRP-conjugated goat anti-rabbit secondary antibodies were incubated afterwards for 1 h at RT ( Supplementary Table 1 ). Three 10 min TBST washes at RT were performed following each antibody incubation. HRP substrate was added for 3–5 min, and the signal was detected by chemiluminescence using a BioRad ChemiDoc MP imaging system. Immunofluorescence Parent B16 and 1 µM-resistant B16 were seeded onto 22 x 22 mm coverslips placed inside 6-well plates. Once cells reached 50% confluency, cells were fixed and permeabilized using the Cytofix/Cytoperm kit per manufacturer instructions. The coverslips were then blocked with 3% BSA in PBS for 1 h at RT, followed by overnight incubation at 4°C with primary antibody solutions for GLUT1, GLUT3, GLUT4, or P-gp ( Supplementary Table 1 ). The next day, secondary antibodies conjugated to Alexa Fluor 488 were added for 1 h at RT in the dark, with 3 x 5 min PBS washes performed at RT after each antibody incubation. Secondary antibody-only controls were included for both samples. Phalloidin-iFluor 594 and DAPI stains were used per manufacturer instructions to visualize the cytoskeleton and nucleus respectively. Coverslips were mounted onto glass slides and images were captured using ZEISS confocal and BioTek Lionheart FX microscopes at 40X and 10X air objectives respectively. RT-qPCR B16 cells resistant to 0, 16, 64, 128, 256, 500, and 1000 nM DOX were seeded into a 96-well plate at 10,000 cells/well. After 48 h, the cells were subjected to lysis and RNA extraction using a Cells-to-Ct kit. In the reverse transcription step, 10 µL lysate was mixed with 40 µL of RT master mix, and the plate was incubated at 37°C for 60 min, followed by 95°C for 5 min to inactivate the RT enzyme. For subsequent qPCR, 4 µL of the synthesized cDNA was combined with TaqMan Gene Expression Master Mix containing ROX passive reference dye, along with primers for Slc2a1 , Slc2a2 , Slc2a3 , Slc2a4 , or Actb , resulting in a 20 µL reaction ( Supplementary Table 2 ). PCR cycles started with denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. A consistent control sample with Actb was included across all plates and used to correct for inter-plate variation. All reactions were run in triplicate and the Slc2a expression for each sample was normalized to its respective Actb expression. Relative expression was calculated using the ΔΔCt method. Functional glucose uptake Parent and 1000 nM DOX-resistant B16 were grown in 6-well plates and allowed to grow to 80% confluency. Media was then changed to glucose and pyruvate-free DMEM for 30 min with and without 25 µM of pan-GLUT inhibitor (DRB18). Then, 100 µM 2-NBDG was added, and plates were allowed to incubate for 10 and 30 min at 37°C before two PBS washes and trypsinization. Cell pellets were collected and resuspended in ice-cold PBS, and all samples were maintained on ice throughout the experiment. To assess impact of nutrient deprivation on uptake, parent and 1000 nM-resistant B16 were similarly grown in 6-well plates. At 80% confluence, cells were then subjected to varying periods of glucose starvation (0, 4, 8, and 16 h) by changing the media to a glucose-free version for the indicated timepoints. Then, 100 µM 2-NBDG was added, and plates were incubated for 30 min at 37°C before two PBS washes, trypsinization, and resuspension in ice-cold PBS. For all runs, parent and 1000 nM-resistant B16 that did not receive any treatment were used to calibrate and set gates on the single-cell population. For each sample, 30,000 events were collected within the gated population with single-cell fluorescence measured using a 488 nm blue laser on an Attune NxT flow cytometer. RNA-seq sample preparation B16 cells resistant to 0, 16, 64, 128, 256, 500, and 1000 nM DOX were cultured in T25 flasks. Additionally, a 1000 nM DOX-resistant sample was incubated in DOX-free media for two passages/two weeks (1000X) to explore the initial regulatory effects of treatment cessation. For each discrete resistance level, three biological replicates were prepared. Upon reaching 80% confluency, cells were trypsinized, pelleted, and washed in ice-cold PBS. Supernatant was removed and pellets were snap-frozen before storing in -80°C. The pellets were shipped on dry ice to GENEWIZ (Azenta Life Sciences) for downstream RNA extraction, QC testing, library preparation, sequencing, and bioinformatic analyses as part of their standard RNA-seq package. Libraries were prepared using poly(A) selection and sequenced on an Illumina platform with a paired-end 2 × 150 bp configuration, generating approximately 20 million reads per sample. RNA-seq data processing Raw RNA-seq reads were quality-trimmed and adapter-filtered using fastp (v0.23.4). 78 Transcript-level quantification of expression was performed using Salmon (v1.10.0) 79 in pseudoalignment mode with a decoy-aware index constructed from the Mus musculus reference genome GRCm39 (ENSEMBL release 113). The index incorporated (1) combined cDNA, ncRNA, and genomic sequences and (2) chromosome/scaffold-derived decoys to minimize spurious mappings. Transcript abundances were aggregated to gene-level counts using tximport (v1.28.0) 80 with ENSEMBL transcript-to-gene mappings. Differential expression analysis was performed with PyDESeq2 (v0.5.0) 81 using standard thresholds (FDR 1). The variance-stabilizing transformation (VST) was applied to DESeq2-normalized counts for heatmaps, PCA plots, and statistical testing. 82 Significant differentially expressed genes were grouped according to their gene ontology annotations, and enrichment of GO terms was assessed by GENEWIZ using Fisher’s exact test (GeneSCF v1.1-p2). 83 Pathways were visualized using pathview (v1.48.0). 84 Statistical Analysis All statistical analyses were performed using Microsoft Excel and/or Python (v3.11.13). Data are presented as mean ± standard deviation (SD), and statistical significance was assessed using Welch’s t-test for independent comparisons unless otherwise specified. A p-value < 0.05 was considered statistically significant and noted as: * p < 0.05; ** p < 0.01; *** p < 0.001 Data Availability Data reported in this study are available within the article with the original files available from the authors upon request. The RNA-seq data from this study is available from the Gene Expression Omnibus (GEO). Abbreviations ABC (ATP-binding cassette) B16 (B16-F10-Luc2) Ct (cycle threshold) DEG (differentially expressed gene) DOX (doxorubicin) FDR (false discovery rate) GLUT (glucose transporter) GO (gene ontology) Log 2 FC (log 2 (fold change)) MDR (multidrug-resistance) 2-NBDG (2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxy-D-glucose) PCA (principal component analysis) P-gp (permeability glycoprotein) Slc2a (solute carrier family 2A) TPM (transcripts per million) VPM (verapamil) VST (variance-stabilizing transformation) Declarations Acknowledgements The authors greatly thank GENEWIZ for performing the RNA extractions and subsequent RNA-seq workflow. The authors also thank Abbey Michaelson, Emma Johnson, and Colin O’Malley for their valuable technical assistance. Authors’ Contributions ETW created the gradient of doxorubicin-resistant B16 cells and performed the viability assays, flow cytometry assays, western blots, imaging, and RT-qPCRs. CL and HL processed the RNA-seq raw reads. ETW and AH performed the downstream analyses. ETW wrote the original first draft of the manuscript. AEN and RJM conceived of the project, supervised, and directed the research. All authors contributed to the review, editing, and approval of the final manuscript. Funding This work was supported by the National Cancer Institute of the National Institutes of Health R01CA234115. ETW acknowledges the Beckman and Goldwater Foundations as well as multiple research scholarships from Miami University for substantial funding. Conflict of interest All authors declared that there are no conflicts of interest. References Bukowski, K., Kciuk, M. & Kontek, R. Mechanisms of Multidrug Resistance in Cancer Chemotherapy. Int. J. Mol. Sci. 21 , 3233 (2020). Scheer, A. et al. Twelve weeks of water-based circuit training exercise improves fitness, body fat and leg strength in people with stable coronary heart disease: a randomised trial. J. Physiother . 67 , 284–290 (2021). Lin, J. et al. The roles of glucose metabolic reprogramming in chemo- and radio-resistance. J. Exp. Clin. Cancer Res. 38 , 218 (2019). Alexa-Stratulat, T., Pešić, M., Gašparović, A. Č., Trougakos, I. P. & Riganti, C. What sustains the multidrug resistance phenotype beyond ABC efflux transporters? Looking beyond the tip of the iceberg. Drug Resist. Updat . 46 , 100643 (2019). Alam, A. & Locher, K. P. Structure and Mechanism of Human ABC Transporters. Annu. Rev. Biophys. 52 , 275–300 (2023). Xiao, H., Zheng, Y., Ma, L., Tian, L. & Sun, Q. Clinically-Relevant ABC Transporter for Anti-Cancer Drug Resistance. Front Pharmacol 12 , (2021). Guo, Y. et al. P-glycoprotein (P-gp)-driven cancer drug resistance: biological profile, non-coding RNAs, drugs and nanomodulators. Drug Discov Today . 29 , 104161 (2024). Amaral, M. V. S. et al. Establishment of Drug-resistant Cell Lines as a Model in Experimental Oncology: A Review. Anticancer Res. 39 , 6443–6455 (2019). McDermott, M. et al. In vitro Development of Chemotherapy and Targeted Therapy Drug-Resistant Cancer Cell Lines: A Practical Guide with Case Studies. Front Oncol 4 , (2014). Moralev, A. D., Markov, O. V., Zenkova, M. A. & Markov, A. V. Novel Cross-Cancer Hub Genes in Doxorubicin Resistance Identified by Transcriptional Mapping. Biomedicines 13 , 2527 (2025). Warburg, O., Wind, F. & Negelein, E. The metabolism of tumors in the body. J. Gen. Physiol. 8 , 519–530 (1927). Stanke, K. M., Wilson, C. & Kidambi, S. High Expression of Glycolytic Genes in Clinical Glioblastoma Patients Correlates With Lower Survival. Front. Mol. Biosci. 8 , 752404 (2021). Bettum, I. J. et al. Metabolic reprogramming supports the invasive phenotype in malignant melanoma. Cancer Lett. 366 , 71–83 (2015). Koch, A. et al. Characterization of glycolysis-related gene expression in malignant melanoma. Pathol. - Res. Pract. 216 , 152752 (2020). Shafi, S. et al. Envisioning Glucose Transporters (GLUTs and SGLTs) as Novel Intervention against Cancer: Drug Discovery Perspective and Targeting Approach. Curr. Drug Targets . 26 , 109–131 (2025). Sun, X. et al. Discovery and development of tumor glycolysis rate-limiting enzyme inhibitors. Bioorg. Chem. 112 , 104891 (2021). Garcia, S. N., Guedes, R. C. & Marques, M. M. Unlocking the Potential of HK2 in Cancer Metabolism and Therapeutics. Curr. Med. Chem. 26 , 7285–7322 (2020). Lešnik, S. et al. Small-molecule inhibitors of 6-phosphofructo-1-kinase simultaneously suppress lactate and superoxide generation in cancer cells. PLOS One . 20 , e0321998 (2025). Li, R. et al. Identification of a new pyruvate kinase M2 isoform (PKM 2) activator for the treatment of non-small‐cell lung cancer (NSCLC). Chem. Biol. Drug Des. 92 , 1851–1858 (2018). Pohl, J. et al. D-19575—a sugar-linked isophosphoramide mustard derivative exploiting transmembrane glucose transport. Cancer Chemother. Pharmacol. 35 , 364–370 (1995). Liu, Z. et al. Enhanced Sampling Molecular Dynamics Simulations Reveal Transport Mechanism of Glycoconjugate Drugs through GLUT1. Int. J. Mol. Sci. 25 , 5486 (2024). Porter, J., Noble, A. R., Signoret, N., Fascione, M. A. & Miller, G. J. Exploring a Gemcitabine-Glucose Hybrid as a Glycoconjugate Prodrug. ACS Omega . 9 , 31703–31713 (2024). Iorio, A. L. et al. Oxime Linked Doxorubicin Glycoconjugates Improve the Specific Targeting of Glioblastoma in High-Grade Glioma Therapy. ACS Med. Chem. Lett. 15 , 1953–1960 (2024). Shriwas, P. et al. A small-molecule pan-class I glucose transporter inhibitor reduces cancer cell proliferation in vitro and tumor growth in vivo by targeting glucose-based metabolism. Cancer Metab. 9 , 14 (2021). Wang, Y., Sun, Z., Zhao, Z., Pang, J. & Chen, J. Recent Progress in the Development of Glucose Transporter (GLUT) Inhibitors. J. Med. Chem. 68 , 1033–1050 (2025). Hayashi, M. et al. GLUT1 inhibition by BAY-876 induces metabolic changes and cell death in human colorectal cancer cells. BMC Cancer . 25 , 716 (2025). Zhang, R. S., Li, Z. K., Liu, J., Deng, Y. T. & Jiang, Y. WZB117 enhanced the anti-tumor effect of apatinib against melanoma via blocking STAT3/PKM2 axis. Front. Pharmacol. 13 , 976117 (2022). Reckzeh, E. S. & Waldmann, H. Small-Molecule Inhibition of Glucose Transporters GLUT‐1–4. ChemBioChem 21, 45–52 (2020). Fu, J., Yang, J., Seeberger, P. H. & Yin, J. Glycoconjugates for glucose transporter-mediated cancer-specific targeting and treatment. Carbohydr. Res. 498 , 108195 (2020). Cao, X. et al. Carbohydrate-based drugs launched during 2000 – 2021. Acta Pharm. Sin B . 12 , 3783–3821 (2022). Giuliani, S. et al. Metabolic Reprogramming in Melanoma: An Epigenetic Point of View. Pharmaceuticals 18 , 853 (2025). Kim, J. & DeBerardinis, R. J. Mechanisms and Implications of Metabolic Heterogeneity in Cancer. Cell. Metab. 30 , 434–446 (2019). Seruga, B., Ocana, A., Amir, E. & Tannock, I. F. Failures in Phase III: Causes and Consequences. Clin. Cancer Res. 21 , 4552–4560 (2015). Tilekar, K. et al. Power of two: combination of therapeutic approaches involving glucose transporter (GLUT) inhibitors to combat cancer. Biochim. Biophys. Acta BBA - Rev. Cancer . 1874 , 188457 (2020). Supino, R., Prosperi, E., Formelli, F., Mariani, M. & Parmiani, G. Characterization of a doxorubicin-resistant murine melanoma line: Studies on cross-resistance and its circumvention. Br. J. Cancer . 54 , 33–42 (1986). Formelli, F., Rossi, C., Supino, R. & Parmiani, G. In vivo characterization of a doxorubicin resistant B16 melanoma cell line. Br. J. Cancer . 54 , 223–233 (1986). Gottesman, M. M., Fojo, T. & Bates, S. E. Multidrug resistance in cancer: role of ATP–dependent transporters. Nat. Rev. Cancer . 2 , 48–58 (2002). Molinari, A. et al. Induction of P-glycoprotein expression on the plasma membrane of human melanoma cells. Anticancer Res. 20 , 2691–2696 (2000). Supino, R., Mariani, M., Capranico, G., Colombo, A. & Parmiani, G. Doxorubicin cellular pharmacokinetics and DNA breakage in a multi-drug resistant B16 melanoma cell line. Br. J. Cancer . 57 , 142–146 (1988). Haroon, M. et al. Efflux-Enhanced Imidazoquinolines To Exploit Chemoresistance. ACS Omega . 10 , 12319–12333 (2025). Kauffman, M., Kauffman, M., Zhu, H., Jia, Z. & Li, Y. Fluorescence-Based Assays for Measuring Doxorubicin in Biological Systems. React. Oxyg Species . https://doi.org/10.20455/ros.2016.873 (2016). Wang, L. & Sun, Y. Efflux mechanism and pathway of verapamil pumping by human P-glycoprotein. Arch. Biochem. Biophys. 696 , 108675 (2020). Pulukuri, A. J. et al. Acquired drug resistance enhances imidazoquinoline efflux by P-glycoprotein. Pharmaceuticals 14 , 1292 (2021). Ruby, K. N. et al. Diagnostic and prognostic value of glucose transporters in melanocytic lesions. Melanoma Res. 29 , 603–611 (2019). Seo, J. W. et al. Real-time monitoring of drug pharmacokinetics within tumor tissue in live animals. Sci. Adv. 8 , eabk2901 (2022). Jolles, B. & Stierlé, V. Doxorubicin-induced senescence is modulated by the eukaryotic release factor 3a and its polyglycine expansion in HCT116 cells. Biochimie 233 , 81–87 (2025). Bojko, A., Czarnecka-Herok, J., Charzynska, A., Dabrowski, M. & Sikora, E. Diversity of the Senescence Phenotype of Cancer Cells Treated with Chemotherapeutic Agents. Cells 8 , 1501 (2019). González-Gualda, E., Baker, A. G., Fruk, L. & Muñoz‐Espín D. A guide to assessing cellular senescence in vitro and in vivo . FEBS J. 288 , 56–80 (2021). Svensson, S. P. S., Lindgren, S., Powell, W. & Green, H. Melanin Inhibits Cytotoxic Effects of Doxorubicin and Daunorubicin in MOLT 4 Cells. Pigment Cell. Res. 16 , 351–354 (2003). Fedele, C. et al. Prospective Isolation According to Melanin Pigment Content of Melanoma Cells With Heterogeneous Potentials for Disease Propagation. Pigment Cell. Melanoma Res 38 , (2025). Sauna, Z. E. & Ambudkar, S. V. Evidence for a requirement for ATP hydrolysis at two distinct steps during a single turnover of the catalytic cycle of human P-glycoprotein. Proc. Natl. Acad. Sci. 97, 2515–2520 (2000). Gao, X., Aguanno, D., Board, M. & Callaghan, R. Exploiting the metabolic energy demands of drug efflux pumps provides a strategy to overcome multidrug resistance in cancer. Biochim. Biophys. Acta BBA - Gen. Subj. 1865 , 129915 (2021). Broxterman, H. J., Pinedo, H. M., Kuiper, C. M., Schuurhuis, G. J. & Lankelma, J. Glycolysis in P-glycoprotein‐overexpressing human tumor cell lines Effects of resistance‐modifying agents. FEBS Lett. 247 , 405–410 (1989). Wartenberg, M. et al. Glycolytic pyruvate regulates P-Glycoprotein expression in multicellular tumor spheroids via modulation of the intracellular redox state. J. Cell. Biochem. 109 , 434–446 (2010). Chen, J. Q. & Russo, J. Dysregulation of glucose transport, glycolysis, TCA cycle and glutaminolysis by oncogenes and tumor suppressors in cancer cells. Biochim. Biophys. Acta BBA - Rev. Cancer . 1826 , 370–384 (2012). Chang, Y. C., Chan, M. H., Yang, Y. F., Li, C. H. & Hsiao, M. Glucose transporter 4: Insulin response mastermind, glycolysis catalyst and treatment direction for cancer progression. Cancer Lett. 563 , 216179 (2023). Ma, W., Li, Z., Wen, J. & Yan, S. The expression and prognostic value of multiple glucose transporters in cutaneous melanoma. Anticancer Res. 44 , 3747–3756 (2024). Hamilton, K. E., Bouwer, M. F., Louters, L. L. & Looyenga, B. D. Cellular binding and uptake of fluorescent glucose analogs 2-NBDG and 6-NBDG occurs independent of membrane glucose transporters. Biochimie 190 , 1–11 (2021). Gonzalez-Menendez, P. et al. GLUT1 protects prostate cancer cells from glucose deprivation-induced oxidative stress. Redox Biol. 17 , 112–127 (2018). Pisco, A. O., Jackson, D. A. & Huang, S. Reduced Intracellular Drug Accumulation in Drug-Resistant Leukemia Cells is Not Only Solely Due to MDR-Mediated Efflux but also to Decreased Uptake. Front Oncol 4 , (2014). Peetla, C. et al. Drug Resistance in Breast Cancer Cells: Biophysical Characterization of and Doxorubicin Interactions with Membrane Lipids. Mol. Pharm. 7 , 2334–2348 (2010). Lothstein, L., Hsu, S. I., Horwitz, S. B. & Greenberger, L. M. Alternate Overexpression of Two Phosphoglycoprotein Genes Is Associated with Changes in Multidrug Resistance in a J774.2 Cell Line. J. Biol. Chem. 264 , 16054–16058 (1989). Hsu, S. I., Lothstein, L. & Horwitz, S. B. Differential overexpression of three mdr gene family members in multidrug-resistant J774.2 mouse cells. J. Biol. Chem. 264 , 12053–12062 (1989). Achkar, I. W. et al. Metabolic Signatures of Tumor Responses to Doxorubicin Elucidated by Metabolic Profiling in Ovo. Metabolites 10 , 268 (2020). Zhou, R., Vander Heiden, M. G. & Rudin, C. M. Genotoxic exposure is associated with alterations in glucose uptake and metabolism. Cancer Res. 62 , 3515–3520 (2002). Kciuk, M. et al. Doxorubicin—An Agent with Multiple Mechanisms of Anticancer Activity. Cells 12 , 659 (2023). Yu, Q. et al. DNA-Damage-Induced Type I Interferon Promotes Senescence and Inhibits Stem Cell Function. Cell. Rep. 11 , 785–797 (2015). Cheon, H., Wang, Y., Wightman, S. M., Jackson, M. W. & Stark, G. R. How cancer cells make and respond to interferon-I. Trends Cancer . 9 , 83–92 (2023). Pidugu, V. K., Pidugu, H. B., Wu, M. M., Liu, C. J. & Lee, T. C. Emerging Functions of Human IFIT Proteins in Cancer. Front. Mol. Biosci. 6 , 148 (2019). Weichselbaum, R. R. et al. An interferon-related gene signature for DNA damage resistance is a predictive marker for chemotherapy and radiation for breast cancer. Proc. Natl. Acad. Sci. 105, 18490–18495 (2008). Yu, H. et al. Bone marrow stromal cell antigen 2: Tumor biology, signaling pathway and therapeutic targeting (Review). Oncol. Rep. 51 , 1–12 (2024). Grabosch, S. et al. Cisplatin-induced immune modulation in ovarian cancer mouse models with distinct inflammation profiles. Oncogene 38 , 2380–2393 (2019). Uchihara, Y. et al. DNA damage promotes HLA class I presentation by stimulating a pioneer round of translation-associated antigen production. Mol. Cell. 82 , 2557–2570e7 (2022). Seifert, U. et al. Immunoproteasomes Preserve Protein Homeostasis upon Interferon-Induced Oxidative Stress. Cell 142 , 613–624 (2010). Bayar, I., Ekren Aşıcı, G. S., Bildik, A. & Kıral, F. Gene Expression of Glycolysis Enzymes in MCF-7 Breast Cancer Cells Exposed to Warburg Effect and Hypoxia. Int J. Mol. Cell. Med 13 , (2024). Mengistu, B. A. et al. Comprehensive review of drug resistance in mammalian cancer stem cells: implications for cancer therapy. Cancer Cell. Int. 24 , 406 (2024). Yang, C., Tian, C., Hoffman, T. E., Jacobsen, N. K. & Spencer, S. L. Melanoma subpopulations that rapidly escape MAPK pathway inhibition incur DNA damage and rely on stress signalling. Nat. Commun. 12 , 1747 (2021). Chen, S., Zhou, Y., Chen, Y. & Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34 , i884–i890 (2018). Patro, R., Duggal, G., Love, M. I., Irizarry, R. A. & Kingsford, C. Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods . 14 , 417–419 (2017). Soneson, C., Love, M. I. & Robinson, M. D. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research 4, 1521 (2015). Muzellec, B., Teleńczuk, M., Cabeli, V. & Andreux, M. PyDESeq2: a python package for bulk RNA-seq differential expression analysis. Bioinformatics 39 , btad547 (2023). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15 , (2014). Subhash, S. & Kanduri, C. GeneSCF: a real-time based functional enrichment tool with support for multiple organisms. BMC Bioinform. 17 , 365 (2016). Luo, W. & Brouwer, C. Pathview: an R/Bioconductor package for pathway-based data integration and visualization. Bioinformatics 29 , 1830–1831 (2013). Additional Declarations No competing interests reported. Supplementary Files WangSupportingInformation2026.pdf Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 24 Feb, 2026 Editor assigned by journal 24 Feb, 2026 Submission checks completed at journal 24 Feb, 2026 First submitted to journal 20 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8929076","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":596339955,"identity":"31095d02-9f4b-4cc4-9807-47c497e7e280","order_by":0,"name":"Emily Wang","email":"","orcid":"","institution":"Miami University","correspondingAuthor":false,"prefix":"","firstName":"Emily","middleName":"","lastName":"Wang","suffix":""},{"id":596339956,"identity":"5a6829cf-3563-4cca-9939-1ba1955bace0","order_by":1,"name":"Annabelle Hauss","email":"","orcid":"","institution":"Miami University","correspondingAuthor":false,"prefix":"","firstName":"Annabelle","middleName":"","lastName":"Hauss","suffix":""},{"id":596339957,"identity":"b1b44598-c2b4-45e5-97ef-769e4b3a5c78","order_by":2,"name":"Hiruni Lokuyaddehige","email":"","orcid":"","institution":"Miami University","correspondingAuthor":false,"prefix":"","firstName":"Hiruni","middleName":"","lastName":"Lokuyaddehige","suffix":""},{"id":596339958,"identity":"769a606a-da66-463b-ad8c-fd43d1fd6c74","order_by":3,"name":"Amy Nielsen","email":"","orcid":"","institution":"Astante Therapeutics Inc.","correspondingAuthor":false,"prefix":"","firstName":"Amy","middleName":"","lastName":"Nielsen","suffix":""},{"id":596339959,"identity":"64ed6fe0-c776-4b34-9f23-2d6a448ca6ac","order_by":4,"name":"Chun Liang","email":"","orcid":"","institution":"Miami University","correspondingAuthor":false,"prefix":"","firstName":"Chun","middleName":"","lastName":"Liang","suffix":""},{"id":596339960,"identity":"b86f0f65-bf65-4b00-ac98-24030a78a499","order_by":5,"name":"Rock Mancini","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYLCCBAYGOQZmMOMAhEuMFmMStQBBYgOEJkKLbvvZgx8e7riXvrad+emGBxV3GPjZcwzwajE7k5cskXimOHfbYTazGwlnnjFI9rwhoOVAjhlDYlsCUAuD2Y3EtsMMBjcI2XL+DVhLutlh9m9gLfYEtdyA2JJgdpgHaosEQS1vjCWAWgy3HeYpA/rlMI/EmWcFBByWY/jxZ1uCvNn549tu/qg4LMffnrwBrxYMwEOa8lEwCkbBKBgFWAEAGRVOA5y7gKcAAAAASUVORK5CYII=","orcid":"","institution":"Miami University","correspondingAuthor":true,"prefix":"","firstName":"Rock","middleName":"","lastName":"Mancini","suffix":""}],"badges":[],"createdAt":"2026-02-20 21:08:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8929076/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8929076/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103472735,"identity":"87d925c3-916e-433f-9d8d-7eb4caacf5db","added_by":"auto","created_at":"2026-02-26 06:13:55","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":18789101,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a)\u003c/strong\u003e Targets of carbohydrate transport, metabolism, and drug efflux in cancer cells. GLUT inhibitors restrict carbohydrate availability to cancer cells, while glyco-targeted drugs incorporate sugar moieties to enhance cellular uptake. Metabolic enzyme inhibitors have also been explored, such as those targeting the glycolytic enzymes HK, PFK, PK, PGAM, ALDO, ENO, and LDH. Efflux inhibitors have been developed to block drug efflux through common transporters like P-gp that otherwise allow MDR phenotypes to persist following treatment. \u003cstrong\u003e(b)\u003c/strong\u003e MDR cancers resist the action of cytotoxic chemotherapeutics like the anthracycline doxorubicin (DOX) by overexpressing ABC transport proteins like P-gp which actively efflux xenobiotics and cellular components alike to the extracellular space. Based on this established effect of enhanced efflux, we generated cell lines with stable resistance to discrete concentrations of DOX across a gradient of chemoresistance. We then examined associated genome-wide compensatory changes as a function of chemoresistance level with a specific focus on in carbohydrate metabolism and GLUT pathways.\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-8929076/v1/35cbe30f29612935e8bbe2d8.png"},{"id":103508093,"identity":"fb54b1a4-e210-4160-85b8-f4a4ef9d4421","added_by":"auto","created_at":"2026-02-26 13:47:10","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":32856367,"visible":true,"origin":"","legend":"\u003cp\u003eCharacterization of DOX tolerance across the B16 chemoresistance gradient. (\u003cstrong\u003ea\u003c/strong\u003e) Cell viability of the B16 gradient ± 5 μM DOX for 48 h. (\u003cstrong\u003eb\u003c/strong\u003e) Cell viability of parent and 1 µM-resistant B16 with varying amounts of DOX added for 48 h. For both (A) and (B), cell viability was measured using resazurin assay and data are represented as mean ± SD normalized to the no-DOX control for each unique sample. * \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; ** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; *** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001 (Welch’s t-test, n = 3). (\u003cstrong\u003ec\u003c/strong\u003e) Flow cytometry histograms of DOX fluorescence in B16 cells examined previously. Parent and 1 µM-resistant B16 were incubated with 10 μM DOX for 30 min ± pre-treatment with 10 μM verapamil (VPM) for 16 h. Adapted from Haroon et al., ACS Omega 2025, 10, 12319–12333, under CC BY 4.0.\u003csup\u003e40\u003c/sup\u003e (\u003cstrong\u003ed\u003c/strong\u003e) Brightfield images revealing morphological changes with induced chemoresistance. Images taken on 10X objective. The scale bar represents 200 μm.\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-8929076/v1/dd127555f501cc49f3b23f5a.png"},{"id":103472740,"identity":"acb5aa80-ed1d-42ed-be5e-54ff6b072d47","added_by":"auto","created_at":"2026-02-26 06:13:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":61833190,"visible":true,"origin":"","legend":"\u003cp\u003eP-gp upregulation in response to DOX is linked to impaired glucose uptake despite conserved GLUT expression. \u003cstrong\u003e(a) \u003c/strong\u003eWestern blot detection of P-gp and GLUTs 1–4 in B16 melanoma cell lysates in order of increasing DOX resistance (nM). β-actin was used as the internal loading control.\u003cstrong\u003e (b)\u003c/strong\u003e Immunofluorescence staining on P-gp and GLUTs in parent B16 and 1000 nM DOX-resistant B16. Confocal images obtained using a 40X air objective. Scale bar represents 50 μm. \u003cstrong\u003e(c) \u003c/strong\u003eRT-qPCR fold change in expression (mean ± SD) of \u003cem\u003eSlc2a1 \u003c/em\u003e(GLUT1), \u003cem\u003eSlc2a3 \u003c/em\u003e(GLUT3), and \u003cem\u003eSlc2a4 \u003c/em\u003e(GLUT4) across different DOX resistance levels. \u003cem\u003eSlc2a2\u003c/em\u003e (GLUT2) showed minimal amplification (Ct \u0026gt; 35). All fold changes (2\u003csup\u003e–ΔΔCt\u003c/sup\u003e) are relative to parent B16 (n=3). (\u003cstrong\u003ed\u003c/strong\u003e) Parent and 1000 nM-resistant B16 were glucose starved with and without 25 μM DRB18 for 30 min prior to adding 100 μM 2-NBDG for 10 and 30 min. (\u003cstrong\u003ee\u003c/strong\u003e) Parent and 1000 nM-resistant B16 were glucose starved for 0, 4, 8, and 16 h before adding 2-NBDG. PBS represents the negative control for background autofluorescence. \u003cstrong\u003e(f) \u003c/strong\u003eDESeq2-normalized mean ± SD gene expression of all murine \u003cem\u003eSlc2a \u003c/em\u003emembers with mean expression \u0026gt; 10 in at least one sample across the chemoresistance gradient measured by RNA-seq. \u003cstrong\u003e(g) \u003c/strong\u003eDESeq2-normalized mean ± SD gene expression of four ABC efflux transporter family members most implicated in MDR cancers\u003csup\u003e6\u003c/sup\u003e across the chemoresistance gradient measured by RNA-seq. For (c), (f), and (g), statistical significance was compared between each resistant sample and parent using Welch’s t-test (n = 3), and VST-normalized counts were used in statistical tests for (f) and (g). * \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; ** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; *** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-8929076/v1/7cbd6a1c375343c595c76890.png"},{"id":103472737,"identity":"13c80d2a-6f6f-48e7-8377-a96701186007","added_by":"auto","created_at":"2026-02-26 06:13:55","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":30922806,"visible":true,"origin":"","legend":"\u003cp\u003eDistinct transcriptional profiles emerge with chemoresistance. \u003cstrong\u003e(a) \u003c/strong\u003eHeatmap showing top differentially expressed genes (n = 6598) between at least one pairwise comparison across the chemoresistance gradient (|log\u003csub\u003e2\u003c/sub\u003eFC| \u0026gt; 1, adjusted p \u0026lt; 0.05, Wald test). Expression values represent the average normalized log expression per condition (n = 3), and heatmap colors reflect row-wise z-scores. (\u003cstrong\u003eb\u003c/strong\u003e) Volcano plot comparing parent (0) and 1000 nM B16 expression profiles. Dashed lines denote thresholds for significance: adjusted \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05 and |log\u003csub\u003e2\u003c/sub\u003eFC| \u0026gt; 1. \u003cstrong\u003e(c) \u003c/strong\u003eDiverging bar chart representing the number of significantly up- and down-regulated genes for each resistance condition relative to 0 nM. \u003cstrong\u003e(d)\u003c/strong\u003e Sample-to-sample distance heatmap illustrating hierarchical clustering of chemoresistant samples based on Euclidean distances. Darker colors indicate greater similarity between samples. \u003cstrong\u003e(e)\u003c/strong\u003e Principal component analysis (PCA) projecting each sample onto the PC1 and PC2 axes representing the major sources of transcriptomic variance. Counts are VST-normalized (from DESeq2 analysis) in panels (a), (d), and (e).\u0026nbsp;\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-8929076/v1/80455c2930caeeec4c05aa75.png"},{"id":103472739,"identity":"e6276330-767d-40f1-b2d8-22c598cd62e0","added_by":"auto","created_at":"2026-02-26 06:13:56","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":42870203,"visible":true,"origin":"","legend":"\u003cp\u003eChemoresistance is associated with reduced glycolytic enzyme expression and induction of antiviral response genes. \u003cstrong\u003e(a)\u003c/strong\u003e Lineplot including genes encoding the 10 glycolytic enzymes and lactate dehydrogenase expressed across the MDR gradient, filtered by expression \u0026gt; 10,000 for at least one sample. \u003cstrong\u003e(b)\u003c/strong\u003e Heatmap of \u003cem\u003eSlc2a1–4\u003c/em\u003e and glycolytic enzymes across the chemoresistance spectrum. \u003cstrong\u003e(c)\u003c/strong\u003e KEGG pathway map of glycolysis showing log₂FC relative to the untreated (0 nM) control, comparing each DOX-treated sample in increasing resistance from left to right in each box (16 to 1000X nM). \u003cstrong\u003e(d) \u003c/strong\u003eBubble chart showing all significant GO terms. Each GO term shown is significantly enriched (adjusted p \u0026lt; 0.05, Fisher’s exact test) in at least one treatment condition relative to the 0 nM control. Bubble size correlates with the number of genes per GO term, and color represents enrichment significance (-log10(adjusted p-value)). \u003cstrong\u003e(e) \u003c/strong\u003eLineplot of genes significantly enriched in “response to virus” and “defense response to virus” gene ontologies (relative to 0 nM), filtered for expression \u0026gt; 500 in at least one sample. \u003cstrong\u003e(f)\u003c/strong\u003e Heatmap including all significantly enriched genes (relative to 0 nM) in “response to virus” and “defense response to virus” gene ontologies. Lineplot data in (a) and (e) are represented as mean ± SD (n=3); all genes shown exhibit |log\u003csub\u003e2\u003c/sub\u003eFC| \u0026gt; 1 and \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05 for at least one pairwise comparison. Heatmap data in (b) and (f) represent VST-normalized counts, with the scalebar representing z-scored expression. All genes shown have expression \u0026gt; 10 in at least one sample.\u003c/p\u003e","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-8929076/v1/5608c28607f95199621d85ab.png"},{"id":103472734,"identity":"dd0d210e-d72e-4ed6-9e68-c2efa1f9d41a","added_by":"auto","created_at":"2026-02-26 06:13:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":2364679,"visible":true,"origin":"","legend":"","description":"","filename":"WangSupportingInformation2026.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8929076/v1/54c44fdb07afcb0abc3c6860.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Beyond Binary: Mapping the Evolution of Melanoma Across a Discrete Gradient of Acquired Chemoresistance","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIt is well-established that treatment of cancers with conventional cytotoxic chemotherapies results in recurrence of multidrug-resistant (MDR) phenotypes.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e These MDR populations limit therapeutic efficacy through a range of genetic and epigenetic mechanisms, including elevated drug efflux, reduced drug influx, enzymatic drug inactivation, remodeled DNA repair systems, and irregular carbohydrate metabolism.\u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e Elevated drug efflux via transport proteins, particularly the ATP-binding cassette (ABC) superfamily, has been well-characterized,\u003csup\u003e5\u003c/sup\u003e prompting interest in targeting these transporters through specific inhibition.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e Of these, permeability glycoprotein (P-gp or ABCB1) is the most well-studied and is established to smoothly increase as a function of contact (both in terms of duration and concentration) with a range of cytotoxic chemotherapies, broadly, across many cancer types.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThese observations have motivated extensive efforts to recapitulate MDR in controlled in-vitro settings. While numerous protocols now exist for inducing chemoresistance,\u003csup\u003e8\u003c/sup\u003e there remains significant debate over which in-vitro resistance level for a given cancer type is best to model in-vivo MDR cancers. Some studies demonstrate that typical MDR cell lines overstate what is found for MDR cancers in-vivo, whereas others contend that resistance presents across a spectrum that is typically 2\u0026ndash;5 fold, but up to 16 fold, that inherent to the progenitor cancer type.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eIn practice, in-vitro models of MDR cancers typically study a small number of arbitrarily-defined resistance states, with many reducing the complex process of acquiring MDR to a binary resistant vs non-resistant comparison.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e While such approaches have yielded valuable insights, this over-discretization can obscure early or transient resistance-associated phenotypes that may arise during intermediate stages of chemoresistance development in a typical in-vitro model with a high-level of resistance. Intentionally characterizing chemoresistance across a broader, stepwise range of resistance levels could help resolve these gradual or non-linear changes, while also capturing clinically relevant heterogeneity in drug biodistribution and intratumoral susceptibility to resistant phenotypes.\u003c/p\u003e \u003cp\u003eThis heterogeneity in therapeutic response is shaped by the plasticity of cellular processes that enable cancers to survive under sustained cytotoxic stress. Among these processes, metabolic regulation and membrane transport play central roles by governing energy availability, nutrient uptake, and drug efflux, and are themselves highly variable across tumor types and cellular contexts. For example, the Warburg effect, first characterized in the 1920s, describes cancer phenotypes with irregular, and often up-regulated, glucose transport and metabolism biased towards anerobic glycolysis.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e These irregularities are now established to play a central role in acquired MDR as well as the persistence and progression of many cancers that manifest as solid tumors, ranging from glioblastoma\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e to melanoma.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThis metabolic phenotype has inspired several emerging therapies which seek to exploit its distinctive overreliance on glucose.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e New chemical entities targeting critical enzymes in glucose metabolism such as hexokinase,\u003csup\u003e17\u003c/sup\u003e phosphofructokinase,\u003csup\u003e18\u003c/sup\u003e and pyruvate kinase,\u003csup\u003e19\u003c/sup\u003e among others, have been developed in recent years and continue to improve in specificity and potency. Likewise, many new cancer drug delivery strategies target the overexpression of carbohydrate transporters, most commonly the glucose transporter (GLUT) family, such as in a glyco-conjugate form to enhance cell entry\u003csup\u003e\u003cspan additionalcitationids=\"CR21 CR22\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e or as a GLUT inhibitor to disrupt energy production and proliferation.\u003csup\u003e\u003cspan additionalcitationids=\"CR25 CR26\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThat said, to-date most GLUT inhibitors remain in preclinical development with only a handful in clinical trials and none that are currently FDA-approved cancer drugs.\u003csup\u003e\u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e Further, manipulating or otherwise targeting carbohydrate metabolism for therapeutic benefit presents a challenging moving target. Cancers are highly heterogeneous and their metabolomic profiles evolve as they develop resistance to therapeutic interventions.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e Emerging glyco-targeted therapies will most often begin clinical investigations where there is a clear unmet need: MDR cancers that have failed multiple established therapies.\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e Thus, a more nuanced understanding of how glycometabolic shifts occur across the full spectrum of acquired MDR could inform the design of more effective therapies (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). In particular, it could highlight nonlinear target expression across the MDR gradient which would improve the predictive accuracy of translating in-vitro results to in-vivo efficacy for cancers that persist across variable levels of drug resistance. However, to our knowledge, the acquisition of an MDR phenotype remains to be examined with such granularity that several intermediate resistance levels are characterized concurrently with treatment na\u0026iuml;ve and high-level MDR cells from the same line.\u003c/p\u003e \u003cp\u003eAs a proof-of-concept starting point to address this, we characterize a melanoma cell line as it acquires stable pharmacodynamic tolerance across incremental concentrations of a cytotoxic chemotherapeutic focusing on changes in efflux pumps, GLUTs, and functional carbohydrate processing (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). This uniquely granular approach could be a useful system as it encompasses a wide range of clinically relevant resistance levels, while also enabling characterization of both transient and persistent transporter, and metabolic adaptations, that arise during the progression of chemoresistance.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEstablishing a discrete gradient of chemoresistance\u003c/h2\u003e \u003cp\u003eTo create incremental levels of pharmacodynamic tolerance, we used a B16 melanoma cell line and the anthracycline doxorubicin (DOX). The B16 line was chosen to expand on, and give more granularity to, the classic studies done by the Parmiani lab, who similarly subjected B16 cells to a range of DOX concentrations (0\u0026ndash;860 ng/mL) and characterized changes in drug sensitivity, DNA content, cross-resistance profiles, and in-vivo tumorigenicity.\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e We chose the \u003cem\u003eLuc2\u003c/em\u003e version of the B16 line for future in-vivo bioluminescence imaging as this could be used as a model of drug resistance that other drugs could be validated against in-vivo. DOX was chosen due to its established ability to stimulate P-gp expression and widespread affinity for multiple members of the common ABC efflux transporter family.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eTo establish an MDR gradient of the parent B16 line, we administered doxorubicin (DOX) at incrementally increasing concentrations, starting from 1 nM and doubling the dose at every stable passage, up to 1 \u0026micro;M (rounding at the 500 nM dose). Once we had created the gradient of MDR melanoma, we likewise observed a gradient of functional resistance, reflected by differences in cell viability following 48 h exposure to a high (5 \u0026micro;M) concentration of DOX (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea; \u003cb\u003eSupplementary Fig.\u0026nbsp;1a\u003c/b\u003e). As expected, DOX sensitivity decreased with increasing levels of pre-established resistance. Viability was comparable to PBS alone in the 500 nM and 1000 nM samples. We also evaluated DOX sensitivity in parent and 1000 nM-resistant B16 cells across a range (0\u0026ndash;10 \u0026micro;M) of DOX concentrations (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). After 48 h, the parental cells showed progressively lower viability with higher DOX concentrations, whereas the resistant line maintained stable viability. Furthermore, it has historically been demonstrated that enhanced P-gp-mediated drug efflux is at least one mechanism that lowers intracellular DOX accumulation in MDR B16 melanoma,\u003csup\u003e39\u003c/sup\u003e and this effect has been quantified with more modern flow cytometry analysis exploiting the intrinsic fluorescence of the anthracycline (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec; \u003cb\u003eSupplementary Fig.\u0026nbsp;1b\u003c/b\u003e).\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e Parental B16 cells showed higher DOX fluorescence than the 1000 nM-resistant B16 population following 30 min incubation with 10 \u0026micro;M DOX, indicating greater drug retention. Pre-treatment with verapamil (VPM), a competitive inhibitor of P-gp,\u003csup\u003e42\u003c/sup\u003e moderately increased DOX retention in the resistant cells. Assessing broader phenotypic differences across the MDR gradient by bright-field microscopy, we also observed increasing heterogeneity with higher DOX resistance, consistent with previous binary investigations of MDR B16 cells that also detailed the procedure of their generation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed; \u003cb\u003eSupplementary Fig.\u0026nbsp;1c\u003c/b\u003e).\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e These morphological changes were further supported by increased granularity observed by flow cytometry as well (\u003cb\u003eSupplementary Fig.\u0026nbsp;1d\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eGLUT expression is conserved across the MDR gradient\u003c/h3\u003e\n\u003cp\u003eWith DOX sensitivity quantified across the MDR gradient, we then proceeded to characterize carbohydrate transport and drug efflux proteins as a function of acquired chemoresistance. Here, we primarily focused on class 1 GLUTs (family members 1\u0026ndash;4) due to their established variability across many cancers, including melanomas.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e We ran western blots to quantify relative P-gp and GLUT expression across the MDR gradient (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). P-gp expression directly correlated with increasing DOX resistance, as expected. In contrast, GLUT1 and GLUT3 remained relatively stable across the entire gradient, while GLUT2 was not detectable. GLUT4 exhibited minimal overall expression albeit with a moderate increase at the 1 \u0026micro;M resistance level. Characterizing this in more detail, we also demonstrated GLUT4 insulin responsiveness which disappeared after longer DOX exposure, suggesting a temporary adaptive response (\u003cb\u003eSupplementary Fig.\u0026nbsp;2\u003c/b\u003e). Given the possibility that the functional efficiency of GLUT influx could involve intracellular redistribution rather than overall expression changes, we also used confocal microscopy to visualize GLUT localization (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb; \u003cb\u003eSupplementary Fig.\u0026nbsp;3\u003c/b\u003e). Imaging confirmed upregulation of P-gp, though there were minimal differences in GLUTs 1, 3, and 4 between the parental and chemoresistant populations, with no considerable changes in localization.\u003c/p\u003e \u003cp\u003eComplementing the proteomic analyses, we evaluated \u003cem\u003eSlc2a\u003c/em\u003e (GLUT) transcript levels for GLUTs 1\u0026ndash;4 by RT-qPCR to investigate whether regulatory changes at the mRNA level might occur independently of detectable changes in protein expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec; \u003cb\u003eSupplementary Fig.\u0026nbsp;4\u003c/b\u003e). Changes in \u003cem\u003eSlc2a1, Slc2a3\u003c/em\u003e, and \u003cem\u003eSlc2a4\u003c/em\u003e (GLUTs 1, 3, and 4), remained modest, and correlated well with the microscopy and western blot characterization. As expected, \u003cem\u003eSlc2a2\u003c/em\u003e (GLUT2) was undetectable.\u003c/p\u003e \u003cp\u003eGiven that GLUT expression was conserved at both the transcriptomic and proteomic levels, we next asked whether functional glucose uptake was similarly maintained using the fluorescent glucose analog 2-NBDG. To isolate GLUT-mediated 2-NBDG uptake, we also examined pre-treatment with the pan-GLUT inhibitor DRB18 shown to inhibit GLUTs 1, 2, 3, and 4,\u003csup\u003e24\u003c/sup\u003e before adding 2-NBDG for 10 and 30 minutes. DRB18 significantly inhibited glucose uptake for both samples, though its effects were less pronounced by the 30 min timepoint (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed; \u003cb\u003eSupplementary Fig.\u0026nbsp;5\u003c/b\u003e). Regardless, B16 cells resistant to 1000 nM DOX exhibited reduced glucose uptake compared to treatment-na\u0026iuml;ve cells at both time points, even with elevated concentrations of 2-NBDG (\u003cb\u003eSupplementary Fig.\u0026nbsp;6\u003c/b\u003e). To determine whether uptake was be influenced by nutrient availability, we subjected both parental and 1000 nM-resistant populations to varying glucose starvation periods (0\u0026ndash;16 h) prior to 2-NBDG exposure. However, minimal differences in 2-NBDG uptake were detected within either group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee; \u003cb\u003eSupplementary Fig.\u0026nbsp;7\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eWith clear alterations in P-gp efflux transporter expression and glucose uptake but limited changes in glucose transporter levels, we next sought to more broadly account for this difference in efflux relative to influx across the MDR gradient. To do this we used RNA-seq to determine transcriptional adaptations that might not have otherwise been captured by our targeted analyses. Here, in addition to the established gradient, we also included a 1000 nM-resistant sample that subsequently had DOX withheld from its culture media for two weeks (denoted as 1000X) to examine early regulatory changes following treatment withdrawal. Overall, our RNA-seq data was consistent with our previous findings: we observed minimal changes in the GLUT / \u003cem\u003eSlc2a\u003c/em\u003e family and prominent increases in P-gp / \u003cem\u003eAbcb1\u003c/em\u003e isoforms that directly correlated to increasing chemoresistance (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eg). Of note, overexpression of \u003cem\u003eAbcb1\u003c/em\u003e isoforms persisted following DOX withdrawal, and none of the other ABC efflux pumps had expression correlating (either positively or negatively) with MDR level (\u003cb\u003eSupplementary Fig.\u0026nbsp;8\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eMapping the transcriptional landscape across the MDR gradient\u003c/h3\u003e\n\u003cp\u003eExamining the RNA-seq data also provided a more comprehensive view into both correlative and transient transcriptomic changes that occur across the MDR gradient. Interestingly, distinct gene expression profiles emerged at each discrete resistance level (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea; \u003cb\u003eSupplementary Fig.\u0026nbsp;9)\u003c/b\u003e. As expected, the early-stage resistant samples (16\u0026ndash;128 nM) appeared to show relatively subtle transcriptional shifts, whereas more pronounced divergence became evident at higher resistance (256\u0026ndash;1000 nM). Comparing each resistance level directly to the treatment-naive parent (0 nM) control revealed a general increase in the number of significantly upregulated and downregulated genes with increasing DOX resistance. However, the magnitude of these changes, and even the genes themselves, did not all directly correlate to DOX concentration. In contrast to P-gp expression, which remained consistently upregulated relative to parent through 1000 nM resistance (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb), many changes occurred transiently or irregularly across the gradient. The 0 vs 256 nM treatment comparison showed the highest total number of differentially expressed genes, while the 0 vs. 1000 nM comparison exhibited the greatest number of significantly upregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec; \u003cb\u003eSupplementary Fig.\u0026nbsp;10\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eHierarchical clustering based on pairwise Euclidean distances demonstrated strong in-group similarity and clear segregation of samples by resistance level (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). Summarizing these findings by Principal Component Analysis (PCA) illustrated that low to intermediate resistance populations (16\u0026ndash;256 nM) exhibited greater transcriptional similarity to each other relative to highly resistant groups (500\u0026ndash;1000X), with the 500 nM level of resistance appearing distinctively isolated, clustering tightly within themselves but not with neighboring resistance levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee). Notably, PC1 scores positively correlated with DOX resistance (r\u0026thinsp;=\u0026thinsp;0.829), and loadings analysis identified \u003cem\u003eAbcb1a\u003c/em\u003e as the strongest contributor to PC1, supporting resistance level as a major source of transcriptional variance lead by the well-established positive correlation with \u003cem\u003eAbcb1\u003c/em\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;11\u003c/b\u003e). That said, noticeable separation emerged starting from the 64 to 128 nM transition, with 500 nM and higher resistance levels occupying distinct regions of the PCA space. Interestingly, the 1000X samples clustered distinctly from the 1000 nM samples and appeared marginally closer to the 0 nM group, possibly reflecting stabilization or partial reversion of the resistant phenotype following initial DOX withdrawal.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eGlycolytic suppression and transient viral response activation across the MDR gradient\u003c/h3\u003e\n\u003cp\u003eTo further explore the transcriptional landscape associated with chemoresistance, we next investigated how gene ontologies, specifically those related to metabolism, are modulated across the MDR gradient. Building on our earlier observation that functional glucose uptake is reduced, despite relatively stable GLUT (\u003cem\u003eSlc2a\u003c/em\u003e) expression and enhanced efflux, we examined the expression of glycolytic pathway components. Here, the rationale is that a slower pathway could account for the difference in influx and efflux. We detected significant downregulation in several key enzymes, especially in the 1000X sample (which had been taken off DOX treatment for two weeks), suggesting a suppression of glycolytic activity following prolonged drug exposure and/or withdrawal (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea; \u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e). These glycolytic changes were further visualized in a combined heatmap with class I \u003cem\u003eSlc2a\u003c/em\u003e genes, which confirmed notable reductions in glycolytic enzyme expression levels and similar behavior in \u003cem\u003eSlc2a1\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). To contextualize these findings within a metabolic framework, we mapped gene expression onto the KEGG glycolysis pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec). This revealed a broad decrease in glycolytic enzyme expression that was most pronounced in the 0 vs 1000X comparison, with other related pathways relatively unaffected.\u003c/p\u003e \u003cp\u003eComplementing this, gene ontology (GO) analysis provided significance in the GO terms \u0026ldquo;response to virus\u0026rdquo; (GO:0009615) and \u0026ldquo;defense response to virus\u0026rdquo; (GO:0051607) in all comparisons to parent (0 nM), with the exception of 0 vs 1000X (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ed). Combining significant genes in these two sets highlighted a noisy, but general trend of upregulation of viral response genes along the MDR gradient (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ee and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ef). Interestingly (and contributing to this noise), many of these increases were transient; some of the most pronounced differences occurred at intermediate resistance levels (128\u0026ndash;500 nM) and became downregulated in the 1000 nM or 1000X samples. Their transient nature could imply that the response decreases as other chemoresistance mechanisms (like \u003cem\u003eAbcb1\u003c/em\u003e\u0026ndash;mediated drug efflux) begin to dominate.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn our study, DOX resistance was conferred to parent B16 cells via incremental doubling of the DOX concentration for each successive passage ranging from 1 nM to 1 \u0026micro;M. Examining variable levels of resistance concurrently could better mimic clinical applicability, as chemotherapeutic biodistribution is unequal among tissue types, implying that various locations of a metastasized cancer are likely exposed to, and ultimately adapt to resist, equally variable concentrations of drug.\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e We selected the 0, 16, 64, 128, 256, 500, and 1000 nM resistance levels for our downstream analyses as considerable changes first occurred somewhere between the 64 and 128 nM resistance levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee). Cell viability assays confirmed incremental resistance, and we observed that cells resistant to higher DOX concentrations also had longer doubling times. This aligns with the observed changes in morphology; in particular, the observed cellular hypertrophy and flattening in the 1000 nM-resistant cells are suggestive of a senescent-like phenotype.\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e Further, the increased granularity may be associated with increased lysosomal and autophagic vesicle content,\u003csup\u003e48\u003c/sup\u003e and is enhanced by increased pigmentation due to melanin, which has been implicated in helping to attenuate the effect of DOX.\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eWith the chemoresistance gradient established, we next characterized the relationship between DOX resistance and glucose transporter (GLUT) expression. Elevated expression of permeability glycoprotein (P-gp), a key mediator of MDR, imposes considerable metabolic stress due to its ATP-dependent efflux activity.\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e In turn, cells may require enhanced energy-producing pathways, such as glycolysis, to sustain chemoresistance.\u003csup\u003e\u003cspan additionalcitationids=\"CR52 CR53\" citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e Although many studies have examined altered glucose transport and metabolism in cancers,\u003csup\u003e55\u003c/sup\u003e few have explored its relationship with discrete variable levels of chemoresistance, and none have investigated how GLUT activity and expression changes across incremental levels of chemoresistance.\u003c/p\u003e \u003cp\u003eGiven P-gp\u0026rsquo;s increased demand for ATP, we were surprised to find that the expression of any given GLUT family member remained largely unchanged across the MDR gradient, despite confirmed increases in P-gp expression and activity. These results were consistent across imaging, proteomic, and transcriptomic analyses. Among the class I GLUTs, GLUT1 and GLUT3 remained abundant in expression, while GLUT2 was undetectable, aligning with prior characterizations across multiple melanocytic lesions.\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e While GLUT4 generally exhibited lower expression relative to GLUTs 1 or 3, we did note a transient upregulation of GLUT4 at the 1 \u0026micro;M-resistance level, with the previously well-established link between GLUT4 expression and insulin exposure retained.\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e This transient increase in GLUT4 may reflect a brief supplementation in cellular energy via insulin-dependent glucose uptake, though its subsequent downregulation suggests that this pathway is unsustainable or inefficient under prolonged chemotherapeutic pressure. This variability may also reflect the inherent heterogeneity of melanoma. Previous studies have shown that while most melanomas lack strong GLUT4 expression, with near complete absence for benign nevi, a small subset of melanomas (7.8%) exhibit moderate to strong GLUT4 expression.\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e We consider this likely heterogeneity a fundamental limitation of our current proof-of-concept study and something that should be examined in more detail.\u003c/p\u003e \u003cp\u003eRegardless, to supplement our expression-based analyses, we evaluated functional GLUT activity through glucose uptake, which revealed decreased uptake in 1000 nM-resistant cells relative to parent. Pre-incubation with the pan-GLUT inhibitor DRB18\u003csup\u003e24\u003c/sup\u003e competitively inhibited uptake in both MDR and parent lines, though its effects became less pronounced with time. This may reflect that some 2-NBDG enters through GLUT-independent pathways,\u003csup\u003e58\u003c/sup\u003e or that the inhibitory effect is eventually overcome. Further, while prior studies have shown enhanced GLUT expression with glucose deprivation,\u003csup\u003e23,59\u003c/sup\u003e we did not find glucose deprivation to appreciably affect glucose uptake in either the parental or resistant line, although uptake remained reduced in the 1000 nM-resistant cells relative to parent across all conditions tested. Taken together, this suggests that MDR B16 cells acquire altered phenotypes in which glucose accumulation and general influx are repressed. This is also consistent with the reduced proliferation rates we previously observed and decreased membrane permeability reported in MDR contexts.\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eBroadening our analysis with RNA-seq across the MDR gradient provided a comprehensive view of DOX-induced changes throughout the entire transcriptome. These results confirmed minimal changes in \u003cem\u003eSlc2a\u003c/em\u003e (GLUT) expression and significant increases in \u003cem\u003eAbcb1\u003c/em\u003e (P-gp) expression towards higher levels of chemoresistance. Notably, the trends in expression of the two murine P-gp isoforms, \u003cem\u003eAbcb1a\u003c/em\u003e and \u003cem\u003eAbcb1b\u003c/em\u003e, seemed to switch past the 128 nM resistance level with the transcripts per million (TPM) abundance of the \u003cem\u003eAbcb1a\u003c/em\u003e isoform markedly surpassing \u003cem\u003eAbcb1b\u003c/em\u003e at the 1000 nM-resistance level (\u003cb\u003eSupplementary Fig.\u0026nbsp;12\u003c/b\u003e). This is consistent with a similar phenomenon reported for J774.2 cells by Lothstein et al, who observed that while \u003cem\u003eAbcb1b\u003c/em\u003e expression dominates during early levels of vinblastine resistance, it declines as drug concentration increases, coinciding with a rise in \u003cem\u003eAbcb1a\u003c/em\u003e expression.\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e It was also found that these isoforms are independently regulated: \u003cem\u003eAbcb1b\u003c/em\u003e becomes overexpressed primarily through gene amplification, whereas \u003cem\u003eAbcb1a\u003c/em\u003e is induced via transcriptional activation.\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e Together, this suggests that for mouse model systems \u003cem\u003eAbcb1a\u003c/em\u003e may act as an inducible backup, becoming engaged only under extreme drug pressure when \u003cem\u003eAbcb1b\u003c/em\u003e-mediated efflux is no longer sufficient.\u003c/p\u003e \u003cp\u003eInvestigating broader transcriptional adaptations beyond \u003cem\u003eSlc2a\u003c/em\u003e and \u003cem\u003eAbc\u003c/em\u003e transporters via pathway enrichment and GO analyses revealed notable decreases in a group of glycolytic enzymes with increasing levels of chemoresistance or DOX withdrawal from 1000 nM-resistant MDR cells. This reduction, coupled with the slowed growth and reduced uptake observed in MDR cells, may reflect the ability of DOX and other genotoxic agents to suppress glycolytic flux and promote cellular senescence, both of which can contribute to long-term metabolic reprogramming.\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e,\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e As DOX targets DNA replication, cancers that slow cell cycle progression likely allow more time for MDR mechanisms (like drug efflux) to take place, thereby enhancing survival.\u003c/p\u003e \u003cp\u003eIn addition to these metabolic changes, we also detected transient increases in antiviral response genes, some of which diminished upon DOX withdrawal, likely reflecting an adaptive defense against DOX\u0026rsquo;s DNA-intercalating mechanism of action.\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e Among these were many interferon-stimulated genes, which have been implicated in resistance to DNA damage and genotoxic therapy-induced senescence.\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e,\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e For example, the progressive upregulation of \u003cem\u003eIfit1\u003c/em\u003e and \u003cem\u003eIfit3\u003c/em\u003e are implicated in mechanisms that block apoptosis, and disrupting them can restore chemosensitivity.\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e,\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e However, they only exhibited pronounced increases at the higher (500 and 1000 nM) resistance levels. While this would indicate an attractive target in lab cell lines with corresponding higher resistance levels, the increase is less pronounced at lower (and perhaps more clinically relevant) levels of drug resistance. More importantly, the most transient induction of the genes included \u003cem\u003eIsg15\u003c/em\u003e and \u003cem\u003eBst2\u003c/em\u003e, which provided maximal expression centered on the 256 nM resistance level, smoothly decreasing for both lower \u003cem\u003eand higher\u003c/em\u003e levels of resistance. This is important because BST2 has been proposed as a druggable target\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e and our experiments demonstrate that the potential increased therapeutic value of a BST2-targeted drug in MDR melanoma would be missed if a simple binary comparison of treatment na\u0026iuml;ve and MDR melanoma was used as a preclinical model. We also detected significant transient increases in \u003cem\u003eSamhd1\u003c/em\u003e (associated with DOX resistance), as well as \u003cem\u003eTap1/2\u003c/em\u003e and \u003cem\u003ePsmb8/9\u003c/em\u003e reflecting transient activation of antigen-processing pathways and immunoproteasome-mediated proteostasis under extreme genotoxic and oxidative stress (\u003cb\u003eSupplementary Fig.\u0026nbsp;13\u003c/b\u003e).\u003csup\u003e\u003cspan additionalcitationids=\"CR73\" citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eOverall, our analyses offer new insights into the stress-adaptive mechanisms melanoma cells may utilize to enhance survival under chemotherapeutic stress and highlight several adaptations that are non-linear with respect to the level of chemoresistance. This is important given the discrepancies in clinically relevant MDR levels and those typically used in-vitro for drug discovery research. While long-term functional resistance appears largely driven by P-gp expression, which directly scales with DOX concentration, it is not accompanied by a compensatory rise in glucose influx via GLUTs. Rather, MDR phenotypes are accompanied by reduced glucose uptake and glycolytic enzyme expression, suggesting a shift away from canonical Warburg metabolism and introducing a more nuanced facet of chemoresistance-associated metabolic reprogramming.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, we have established that a panel of B16 melanoma, with incremental discrete concentrations of pharmacodynamic tolerance to doxorubicin, has both correlative and transient changes in phenotype yet minimal changes in GLUT expression. After confirming variable levels of resistance through viability assays, we examined functional glucose uptake along with transport protein expression, specifically class I GLUTs and P-gp. Here, we observed minimal changes in the expression of GLUTs despite reduced glucose uptake, along with the expected increase in P-gp expression across the MDR gradient. To expand beyond this group of transporters, we ran RNA-seq analysis on the MDR gradient, where only \u003cem\u003eAbcb1\u003c/em\u003e (P-gp) isoforms demonstrated noticeable change among the ABC transporters. Other notable changes included decreases in glycolytic enzymes and pronounced, yet transient, increases in viral response-related genes. Overall, our study contributes insight into the evolution of cancer cells as they progress towards MDR phenotypes. That said, this particular study should be understood to have two primary limitations. First, DOX is not the most clinically relevant chemotherapeutic for melanoma; rather, we chose DOX due to its well-documented ability to induce an MDR phenotype, enabling direct comparison with a broad body of prior research. Second, our experiments were performed in-vitro at the bulk cell-culture level, which cannot mimic the full complexity or heterogeneity of the tumor microenvironment. Indeed, many factors such as hypoxia, nutrient deprivation, varying pH, and immune infiltration can affect drug response, metabolism, and transporter activity,\u003csup\u003e75,76\u003c/sup\u003e and this heterogeneity could be a major driving force in melanoma.\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e To address this, in the future this type of study could be expanded to include single-cell analysis of tumor isolates generated in-vivo. Regardless, we expect our initial work here will aid in the development of anti-cancer drugs that target carbohydrate influx (or efflux) mechanisms and inform the rational for in-vitro testing of new drugs aimed at cancers that have failed multiple rounds of chemotherapies.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eMaterials\u003c/h2\u003e \u003cp\u003eThe following biologics, chemicals, and instruments were used. For cell culture: B16-F10-Luc2 (ATCC), DMEM (VWR), HI-FBS (VWR), Blasticidin (Invivogen), Normocin (Invivogen), Trypsin-EDTA (ThermoFisher), doxorubicin (BroadPharm). For viability assays: Resazurin (Sigma-Aldrich), DMEM without phenol red (VWR), L-glutamine (ThermoFisher), BioTek Synergy LX microplate reader (BioTek). For flow cytometry: Verapamil (Ambeed), 2-NBDG (ThermoFisher), DRB18 (Sigma-Aldrich), DMEM without glucose (ThermoFisher), Glucose solution (ThermoFisher), Attune NxT (ThermoFisher). For western blots: 1% Triton-X Lysis buffer (ThermoFisher), protease inhibitor cocktail (Thermofisher), BCA assay (ThermoFisher), electrophoresis equipment (BioRad), protein ladder (ThermoFisher), PVDF membranes (BioRad), 4X Laemmli buffer (BioRad), nonfat dried milk (RPI), insulin solution (ThermoFisher), primary and secondary antibodies (abcam), HRP substrate (ThermoFisher), BioRad ChemiDoc MP (BioRad). For immunofluorescence: 22 x 22 mm coverslips (Fisher Scientific), Cytofix / Cytoperm kit (BD Biosciences), BSA (Glentham Life Sciences), primary and secondary antibodies (abcam), Phalloidin-iFluor 594 (AAT Bioquest), Fluoromount-G with DAPI (SouthernBiotech), ZEISS LSM 710 confocal microscope (ZEISS), BioTek Lionheart FX (BioTek). For RT-qPCR: Cells-to-Ct kit (ThermoFisher), primers (ThermoFisher), nuclease-free water (ThermoFisher), RNAseZap (ThermoFisher), DNAZap (ThermoFisher), QuantStudio 7 (ThermoFisher).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCell culture\u003c/h2\u003e \u003cp\u003eThe B16-F10-Luc2 (B16) cell line was purchased through ATCC and cultured based on manufacturer instructions. The B16 cells were grown in complete culture media containing DMEM with 4.5 g/L glucose, 4 mM L-glutamine, and 110 mg/L sodium pyruvate, and supplemented with 10% (v/v) FBS, 10 \u0026micro;g/mL Blasticidin, and 100 \u0026micro;g/mL Normocin. Media was changed every 2\u0026ndash;4 days. Cells were passaged using 0.25% Trypsin-EDTA solution upon reaching 80% confluence and 1\u0026ndash;2\u0026nbsp;million cells were seeded into a new T150 flask. The chemotherapeutic doxorubicin (DOX) was used to create the chemoresistance gradient consisting of 1, 2, 4, 8, 16, 32, 64, 128, 256, 500, and 1000 nM DOX concentrations. In brief, 1 nM DOX was added to the growth media of parent cells, and the concentration was doubled at each stable passage until the 1000 nM resistance level was reached. For all downstream experiments, cultures were incubated in their respective DOX-resistance concentrations for at least two passages. All cultures were maintained in a 37 ℃, 5% CO\u003csub\u003e2\u003c/sub\u003e incubator.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eTolerance to DOX exposure\u003c/h2\u003e \u003cp\u003eCell viability following DOX exposure was assessed using a resazurin assay. To confirm incremental resistance to DOX, B16 cells resistant to 0 (parent), 16, 64, 128, 256, 500, and 1000 nM DOX were seeded into a 96-well plate at 20,000 cells/well in complete media and allowed to adhere for 24 h. DOX was then added to each sample, reaching a final concentration of 5 \u0026micro;M, and these experiments were run alongside corresponding control wells that received PBS only. After 48 h, the media was changed to phenol red-free DMEM, and resazurin solution (0.1 mg/mL) was added to 10% of the final well volume.\u003c/p\u003e \u003cp\u003eTo test resistance capacity across varying doses of DOX, parent B16 and 1000 nM-resistant B16 were seeded into a 96-well plate at 20,000 cells/well. After 24 h, both cell types were given a range of DOX concentrations (0.05, 0.1, 0.5, 1, 5, and 10 \u0026micro;M), along with PBS as vehicle control. Cells were similarly incubated for 48 h before the media change and addition of resazurin.\u003c/p\u003e \u003cp\u003eAll plates were incubated with resazurin for 4 h before taking absorbance readings at 570 and 600 nm, which correspond to the absorbance of resazurin and resazurin converted to resorufin by live cells respectively. Cell viability was determined by calculating the difference between the 570 nm and 600 nm reads, then subtracting out the background absorbance of wells with resazurin alone (no cells). Data was normalized relative to cells receiving vehicle control to yield % viability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eWestern blot\u003c/h2\u003e \u003cp\u003eB16 cells resistant to 0 (parent), 16, 64, 128, 256, 500, and 1000 nM DOX were grown in separate wells in 6-well plates. Upon reaching 80% confluency, the cells were scraped in 1% Triton-X lysis buffer with protease inhibitors, and a BCA assay was used to standardize protein concentration used from each lysate. Samples were then loaded on a 10% cross-linked gel at 25 \u0026micro;g/well to be separated by SDS-PAGE. Then, the gel was electrically transferred to a PVDF membrane, and the membrane was blocked in 5% nonfat milk overnight at 4\u0026deg;C. Primary rabbit anti-mouse antibodies against GLUT1, GLUT2, GLUT3, GLUT4, P-gp, and β-actin were incubated for 2 h at RT and HRP-conjugated goat anti-rabbit secondary antibodies were incubated afterwards for 1 h at RT (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e). Three 10 min TBST washes at RT were performed following each antibody incubation. HRP substrate was added for 3\u0026ndash;5 min, and the signal was detected by chemiluminescence using a BioRad ChemiDoc MP imaging system.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eImmunofluorescence\u003c/h2\u003e \u003cp\u003eParent B16 and 1 \u0026micro;M-resistant B16 were seeded onto 22 x 22 mm coverslips placed inside 6-well plates. Once cells reached 50% confluency, cells were fixed and permeabilized using the Cytofix/Cytoperm kit per manufacturer instructions. The coverslips were then blocked with 3% BSA in PBS for 1 h at RT, followed by overnight incubation at 4\u0026deg;C with primary antibody solutions for GLUT1, GLUT3, GLUT4, or P-gp (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e). The next day, secondary antibodies conjugated to Alexa Fluor 488 were added for 1 h at RT in the dark, with 3 x 5 min PBS washes performed at RT after each antibody incubation. Secondary antibody-only controls were included for both samples. Phalloidin-iFluor 594 and DAPI stains were used per manufacturer instructions to visualize the cytoskeleton and nucleus respectively. Coverslips were mounted onto glass slides and images were captured using ZEISS confocal and BioTek Lionheart FX microscopes at 40X and 10X air objectives respectively.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eRT-qPCR\u003c/h2\u003e \u003cp\u003eB16 cells resistant to 0, 16, 64, 128, 256, 500, and 1000 nM DOX were seeded into a 96-well plate at 10,000 cells/well. After 48 h, the cells were subjected to lysis and RNA extraction using a Cells-to-Ct kit. In the reverse transcription step, 10 \u0026micro;L lysate was mixed with 40 \u0026micro;L of RT master mix, and the plate was incubated at 37\u0026deg;C for 60 min, followed by 95\u0026deg;C for 5 min to inactivate the RT enzyme. For subsequent qPCR, 4 \u0026micro;L of the synthesized cDNA was combined with TaqMan Gene Expression Master Mix containing ROX passive reference dye, along with primers for \u003cem\u003eSlc2a1\u003c/em\u003e, \u003cem\u003eSlc2a2\u003c/em\u003e, \u003cem\u003eSlc2a3\u003c/em\u003e, \u003cem\u003eSlc2a4\u003c/em\u003e, or \u003cem\u003eActb\u003c/em\u003e, resulting in a 20 \u0026micro;L reaction (\u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e). PCR cycles started with denaturation at 95\u0026deg;C for 10 min, followed by 40 cycles of 95\u0026deg;C for 15 s and 60\u0026deg;C for 1 min. A consistent control sample with \u003cem\u003eActb\u003c/em\u003e was included across all plates and used to correct for inter-plate variation. All reactions were run in triplicate and the \u003cem\u003eSlc2a\u003c/em\u003e expression for each sample was normalized to its respective \u003cem\u003eActb\u003c/em\u003e expression. Relative expression was calculated using the ΔΔCt method.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eFunctional glucose uptake\u003c/h2\u003e \u003cp\u003eParent and 1000 nM DOX-resistant B16 were grown in 6-well plates and allowed to grow to 80% confluency. Media was then changed to glucose and pyruvate-free DMEM for 30 min with and without 25 \u0026micro;M of pan-GLUT inhibitor (DRB18). Then, 100 \u0026micro;M 2-NBDG was added, and plates were allowed to incubate for 10 and 30 min at 37\u0026deg;C before two PBS washes and trypsinization. Cell pellets were collected and resuspended in ice-cold PBS, and all samples were maintained on ice throughout the experiment.\u003c/p\u003e \u003cp\u003eTo assess impact of nutrient deprivation on uptake, parent and 1000 nM-resistant B16 were similarly grown in 6-well plates. At 80% confluence, cells were then subjected to varying periods of glucose starvation (0, 4, 8, and 16 h) by changing the media to a glucose-free version for the indicated timepoints. Then, 100 \u0026micro;M 2-NBDG was added, and plates were incubated for 30 min at 37\u0026deg;C before two PBS washes, trypsinization, and resuspension in ice-cold PBS.\u003c/p\u003e \u003cp\u003eFor all runs, parent and 1000 nM-resistant B16 that did not receive any treatment were used to calibrate and set gates on the single-cell population. For each sample, 30,000 events were collected within the gated population with single-cell fluorescence measured using a 488 nm blue laser on an Attune NxT flow cytometer.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eRNA-seq sample preparation\u003c/h2\u003e \u003cp\u003eB16 cells resistant to 0, 16, 64, 128, 256, 500, and 1000 nM DOX were cultured in T25 flasks. Additionally, a 1000 nM DOX-resistant sample was incubated in DOX-free media for two passages/two weeks (1000X) to explore the initial regulatory effects of treatment cessation. For each discrete resistance level, three biological replicates were prepared. Upon reaching 80% confluency, cells were trypsinized, pelleted, and washed in ice-cold PBS. Supernatant was removed and pellets were snap-frozen before storing in -80\u0026deg;C. The pellets were shipped on dry ice to GENEWIZ (Azenta Life Sciences) for downstream RNA extraction, QC testing, library preparation, sequencing, and bioinformatic analyses as part of their standard RNA-seq package. Libraries were prepared using poly(A) selection and sequenced on an Illumina platform with a paired-end 2 \u0026times; 150 bp configuration, generating approximately 20\u0026nbsp;million reads per sample.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eRNA-seq data processing\u003c/h2\u003e \u003cp\u003eRaw RNA-seq reads were quality-trimmed and adapter-filtered using fastp (v0.23.4).\u003csup\u003e78\u003c/sup\u003e Transcript-level quantification of expression was performed using Salmon (v1.10.0)\u003csup\u003e79\u003c/sup\u003e in pseudoalignment mode with a decoy-aware index constructed from the \u003cem\u003eMus musculus\u003c/em\u003e reference genome GRCm39 (ENSEMBL release 113). The index incorporated (1) combined cDNA, ncRNA, and genomic sequences and (2) chromosome/scaffold-derived decoys to minimize spurious mappings.\u003c/p\u003e \u003cp\u003eTranscript abundances were aggregated to gene-level counts using tximport (v1.28.0)\u003csup\u003e80\u003c/sup\u003e with ENSEMBL transcript-to-gene mappings. Differential expression analysis was performed with PyDESeq2 (v0.5.0)\u003csup\u003e81\u003c/sup\u003e using standard thresholds (FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.05, |log\u003csub\u003e2\u003c/sub\u003eFC| \u0026gt; 1). The variance-stabilizing transformation (VST) was applied to DESeq2-normalized counts for heatmaps, PCA plots, and statistical testing.\u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e Significant differentially expressed genes were grouped according to their gene ontology annotations, and enrichment of GO terms was assessed by GENEWIZ using Fisher\u0026rsquo;s exact test (GeneSCF v1.1-p2).\u003csup\u003e\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e Pathways were visualized using pathview (v1.48.0).\u003csup\u003e84\u003c/sup\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses were performed using Microsoft Excel and/or Python (v3.11.13). Data are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (SD), and statistical significance was assessed using Welch\u0026rsquo;s t-test for independent comparisons unless otherwise specified. A p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant and noted as: * \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05; ** \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01; *** \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eData Availability\u003c/h2\u003e \u003cp\u003eData reported in this study are available within the article with the original files available from the authors upon request. The RNA-seq data from this study is available from the Gene Expression Omnibus (GEO).\u003c/p\u003e \u003c/div\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eABC (ATP-binding cassette)\u003c/p\u003e\n\u003cp\u003eB16 (B16-F10-Luc2)\u003c/p\u003e\n\u003cp\u003eCt (cycle threshold)\u003c/p\u003e\n\u003cp\u003eDEG (differentially expressed gene)\u003c/p\u003e\n\u003cp\u003eDOX (doxorubicin)\u003c/p\u003e\n\u003cp\u003eFDR (false discovery rate)\u003c/p\u003e\n\u003cp\u003eGLUT (glucose transporter)\u003c/p\u003e\n\u003cp\u003eGO (gene ontology)\u003c/p\u003e\n\u003cp\u003eLog\u003csub\u003e2\u003c/sub\u003eFC (log\u003csub\u003e2\u003c/sub\u003e(fold change))\u003c/p\u003e\n\u003cp\u003eMDR (multidrug-resistance)\u003c/p\u003e\n\u003cp\u003e2-NBDG (2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxy-D-glucose)\u003c/p\u003e\n\u003cp\u003ePCA (principal component analysis)\u003c/p\u003e\n\u003cp\u003eP-gp (permeability glycoprotein)\u003c/p\u003e\n\u003cp\u003eSlc2a (solute carrier family 2A)\u003c/p\u003e\n\u003cp\u003eTPM (transcripts per million)\u003c/p\u003e\n\u003cp\u003eVPM (verapamil)\u003c/p\u003e\n\u003cp\u003eVST (variance-stabilizing transformation)\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eThe authors greatly thank GENEWIZ for performing the RNA extractions and subsequent RNA-seq workflow. The authors also thank Abbey Michaelson, Emma Johnson, and Colin O\u0026rsquo;Malley for their valuable technical assistance.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Authors\u0026rsquo; Contributions\u003c/p\u003e\n\u003cp\u003eETW created the gradient of doxorubicin-resistant B16 cells and performed the viability assays, flow cytometry assays, western blots, imaging, and RT-qPCRs. CL and HL processed the RNA-seq raw reads. ETW and AH performed the downstream analyses. ETW wrote the original first draft of the manuscript. AEN and RJM conceived of the project, supervised, and directed the research. All authors contributed to the review, editing, and approval of the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Cancer Institute of the National Institutes of Health R01CA234115. ETW acknowledges the Beckman and Goldwater Foundations as well as multiple research scholarships from Miami University for substantial funding.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors declared that there are no conflicts of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBukowski, K., Kciuk, M. \u0026amp; Kontek, R. Mechanisms of Multidrug Resistance in Cancer Chemotherapy. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 3233 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheer, A. et al. Twelve weeks of water-based circuit training exercise improves fitness, body fat and leg strength in people with stable coronary heart disease: a randomised trial. \u003cem\u003eJ. Physiother\u003c/em\u003e. \u003cb\u003e67\u003c/b\u003e, 284\u0026ndash;290 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, J. et al. The roles of glucose metabolic reprogramming in chemo- and radio-resistance. \u003cem\u003eJ. Exp. Clin. Cancer Res.\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e, 218 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlexa-Stratulat, T., Pešić, M., Gašparović, A. Č., Trougakos, I. P. \u0026amp; Riganti, C. What sustains the multidrug resistance phenotype beyond ABC efflux transporters? Looking beyond the tip of the iceberg. \u003cem\u003eDrug Resist. Updat\u003c/em\u003e. \u003cb\u003e46\u003c/b\u003e, 100643 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlam, A. \u0026amp; Locher, K. P. Structure and Mechanism of Human ABC Transporters. \u003cem\u003eAnnu. Rev. Biophys.\u003c/em\u003e \u003cb\u003e52\u003c/b\u003e, 275\u0026ndash;300 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao, H., Zheng, Y., Ma, L., Tian, L. \u0026amp; Sun, Q. Clinically-Relevant ABC Transporter for Anti-Cancer Drug Resistance. \u003cem\u003eFront Pharmacol\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo, Y. et al. P-glycoprotein (P-gp)-driven cancer drug resistance: biological profile, non-coding RNAs, drugs and nanomodulators. \u003cem\u003eDrug Discov Today\u003c/em\u003e. \u003cb\u003e29\u003c/b\u003e, 104161 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmaral, M. V. S. et al. Establishment of Drug-resistant Cell Lines as a Model in Experimental Oncology: A Review. \u003cem\u003eAnticancer Res.\u003c/em\u003e \u003cb\u003e39\u003c/b\u003e, 6443\u0026ndash;6455 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcDermott, M. et al. In vitro Development of Chemotherapy and Targeted Therapy Drug-Resistant Cancer Cell Lines: A Practical Guide with Case Studies. \u003cem\u003eFront Oncol\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoralev, A. D., Markov, O. V., Zenkova, M. A. \u0026amp; Markov, A. V. Novel Cross-Cancer Hub Genes in Doxorubicin Resistance Identified by Transcriptional Mapping. \u003cem\u003eBiomedicines\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 2527 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarburg, O., Wind, F. \u0026amp; Negelein, E. The metabolism of tumors in the body. \u003cem\u003eJ. Gen. Physiol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 519\u0026ndash;530 (1927).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStanke, K. M., Wilson, C. \u0026amp; Kidambi, S. High Expression of Glycolytic Genes in Clinical Glioblastoma Patients Correlates With Lower Survival. \u003cem\u003eFront. Mol. Biosci.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 752404 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBettum, I. J. et al. Metabolic reprogramming supports the invasive phenotype in malignant melanoma. \u003cem\u003eCancer Lett.\u003c/em\u003e \u003cb\u003e366\u003c/b\u003e, 71\u0026ndash;83 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoch, A. et al. Characterization of glycolysis-related gene expression in malignant melanoma. \u003cem\u003ePathol. - Res. Pract.\u003c/em\u003e \u003cb\u003e216\u003c/b\u003e, 152752 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShafi, S. et al. Envisioning Glucose Transporters (GLUTs and SGLTs) as Novel Intervention against Cancer: Drug Discovery Perspective and Targeting Approach. \u003cem\u003eCurr. Drug Targets\u003c/em\u003e. \u003cb\u003e26\u003c/b\u003e, 109\u0026ndash;131 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun, X. et al. Discovery and development of tumor glycolysis rate-limiting enzyme inhibitors. \u003cem\u003eBioorg. Chem.\u003c/em\u003e \u003cb\u003e112\u003c/b\u003e, 104891 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia, S. N., Guedes, R. C. \u0026amp; Marques, M. M. Unlocking the Potential of HK2 in Cancer Metabolism and Therapeutics. \u003cem\u003eCurr. Med. Chem.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 7285\u0026ndash;7322 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLešnik, S. et al. Small-molecule inhibitors of 6-phosphofructo-1-kinase simultaneously suppress lactate and superoxide generation in cancer cells. \u003cem\u003ePLOS One\u003c/em\u003e. \u003cb\u003e20\u003c/b\u003e, e0321998 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, R. et al. Identification of a new pyruvate kinase M2 isoform (PKM 2) activator for the treatment of non-small‐cell lung cancer (NSCLC). \u003cem\u003eChem. Biol. Drug Des.\u003c/em\u003e \u003cb\u003e92\u003c/b\u003e, 1851\u0026ndash;1858 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePohl, J. et al. D-19575\u0026mdash;a sugar-linked isophosphoramide mustard derivative exploiting transmembrane glucose transport. \u003cem\u003eCancer Chemother. Pharmacol.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 364\u0026ndash;370 (1995).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Z. et al. Enhanced Sampling Molecular Dynamics Simulations Reveal Transport Mechanism of Glycoconjugate Drugs through GLUT1. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 5486 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePorter, J., Noble, A. R., Signoret, N., Fascione, M. A. \u0026amp; Miller, G. J. Exploring a Gemcitabine-Glucose Hybrid as a Glycoconjugate Prodrug. \u003cem\u003eACS Omega\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e, 31703\u0026ndash;31713 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIorio, A. L. et al. Oxime Linked Doxorubicin Glycoconjugates Improve the Specific Targeting of Glioblastoma in High-Grade Glioma Therapy. \u003cem\u003eACS Med. Chem. Lett.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 1953\u0026ndash;1960 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShriwas, P. et al. A small-molecule pan-class I glucose transporter inhibitor reduces cancer cell proliferation in vitro and tumor growth in vivo by targeting glucose-based metabolism. \u003cem\u003eCancer Metab.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 14 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Y., Sun, Z., Zhao, Z., Pang, J. \u0026amp; Chen, J. Recent Progress in the Development of Glucose Transporter (GLUT) Inhibitors. \u003cem\u003eJ. Med. Chem.\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e, 1033\u0026ndash;1050 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHayashi, M. et al. GLUT1 inhibition by BAY-876 induces metabolic changes and cell death in human colorectal cancer cells. \u003cem\u003eBMC Cancer\u003c/em\u003e. \u003cb\u003e25\u003c/b\u003e, 716 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, R. S., Li, Z. K., Liu, J., Deng, Y. T. \u0026amp; Jiang, Y. WZB117 enhanced the anti-tumor effect of apatinib against melanoma via blocking STAT3/PKM2 axis. \u003cem\u003eFront. Pharmacol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 976117 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReckzeh, E. S. \u0026amp; Waldmann, H. Small-Molecule Inhibition of Glucose Transporters GLUT‐1\u0026ndash;4. \u003cem\u003eChemBioChem\u003c/em\u003e 21, 45\u0026ndash;52 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu, J., Yang, J., Seeberger, P. H. \u0026amp; Yin, J. Glycoconjugates for glucose transporter-mediated cancer-specific targeting and treatment. \u003cem\u003eCarbohydr. Res.\u003c/em\u003e \u003cb\u003e498\u003c/b\u003e, 108195 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao, X. et al. Carbohydrate-based drugs launched during 2000\u0026thinsp;\u0026ndash;\u0026thinsp;2021. \u003cem\u003eActa Pharm. Sin B\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e, 3783\u0026ndash;3821 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiuliani, S. et al. Metabolic Reprogramming in Melanoma: An Epigenetic Point of View. \u003cem\u003ePharmaceuticals\u003c/em\u003e \u003cb\u003e18\u003c/b\u003e, 853 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, J. \u0026amp; DeBerardinis, R. J. Mechanisms and Implications of Metabolic Heterogeneity in Cancer. \u003cem\u003eCell. Metab.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 434\u0026ndash;446 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeruga, B., Ocana, A., Amir, E. \u0026amp; Tannock, I. F. Failures in Phase III: Causes and Consequences. \u003cem\u003eClin. Cancer Res.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 4552\u0026ndash;4560 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTilekar, K. et al. Power of two: combination of therapeutic approaches involving glucose transporter (GLUT) inhibitors to combat cancer. \u003cem\u003eBiochim. Biophys. Acta BBA - Rev. Cancer\u003c/em\u003e. \u003cb\u003e1874\u003c/b\u003e, 188457 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSupino, R., Prosperi, E., Formelli, F., Mariani, M. \u0026amp; Parmiani, G. Characterization of a doxorubicin-resistant murine melanoma line: Studies on cross-resistance and its circumvention. \u003cem\u003eBr. J. Cancer\u003c/em\u003e. \u003cb\u003e54\u003c/b\u003e, 33\u0026ndash;42 (1986).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFormelli, F., Rossi, C., Supino, R. \u0026amp; Parmiani, G. In vivo characterization of a doxorubicin resistant B16 melanoma cell line. \u003cem\u003eBr. J. Cancer\u003c/em\u003e. \u003cb\u003e54\u003c/b\u003e, 223\u0026ndash;233 (1986).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGottesman, M. M., Fojo, T. \u0026amp; Bates, S. E. Multidrug resistance in cancer: role of ATP\u0026ndash;dependent transporters. \u003cem\u003eNat. Rev. Cancer\u003c/em\u003e. \u003cb\u003e2\u003c/b\u003e, 48\u0026ndash;58 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMolinari, A. et al. Induction of P-glycoprotein expression on the plasma membrane of human melanoma cells. \u003cem\u003eAnticancer Res.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 2691\u0026ndash;2696 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSupino, R., Mariani, M., Capranico, G., Colombo, A. \u0026amp; Parmiani, G. Doxorubicin cellular pharmacokinetics and DNA breakage in a multi-drug resistant B16 melanoma cell line. \u003cem\u003eBr. J. Cancer\u003c/em\u003e. \u003cb\u003e57\u003c/b\u003e, 142\u0026ndash;146 (1988).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaroon, M. et al. Efflux-Enhanced Imidazoquinolines To Exploit Chemoresistance. \u003cem\u003eACS Omega\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e, 12319\u0026ndash;12333 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKauffman, M., Kauffman, M., Zhu, H., Jia, Z. \u0026amp; Li, Y. Fluorescence-Based Assays for Measuring Doxorubicin in Biological Systems. \u003cem\u003eReact. Oxyg Species\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.20455/ros.2016.873\u003c/span\u003e\u003cspan address=\"10.20455/ros.2016.873\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, L. \u0026amp; Sun, Y. Efflux mechanism and pathway of verapamil pumping by human P-glycoprotein. \u003cem\u003eArch. Biochem. Biophys.\u003c/em\u003e \u003cb\u003e696\u003c/b\u003e, 108675 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePulukuri, A. J. et al. Acquired drug resistance enhances imidazoquinoline efflux by P-glycoprotein. \u003cem\u003ePharmaceuticals\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1292 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuby, K. N. et al. Diagnostic and prognostic value of glucose transporters in melanocytic lesions. \u003cem\u003eMelanoma Res.\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e, 603\u0026ndash;611 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeo, J. W. et al. Real-time monitoring of drug pharmacokinetics within tumor tissue in live animals. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, eabk2901 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJolles, B. \u0026amp; Stierl\u0026eacute;, V. Doxorubicin-induced senescence is modulated by the eukaryotic release factor 3a and its polyglycine expansion in HCT116 cells. \u003cem\u003eBiochimie\u003c/em\u003e \u003cb\u003e233\u003c/b\u003e, 81\u0026ndash;87 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBojko, A., Czarnecka-Herok, J., Charzynska, A., Dabrowski, M. \u0026amp; Sikora, E. Diversity of the Senescence Phenotype of Cancer Cells Treated with Chemotherapeutic Agents. \u003cem\u003eCells\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 1501 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez-Gualda, E., Baker, A. G., Fruk, L. \u0026amp; Mu\u0026ntilde;oz‐Esp\u0026iacute;n D. A guide to assessing cellular senescence \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e. \u003cem\u003eFEBS J.\u003c/em\u003e \u003cb\u003e288\u003c/b\u003e, 56\u0026ndash;80 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSvensson, S. P. S., Lindgren, S., Powell, W. \u0026amp; Green, H. Melanin Inhibits Cytotoxic Effects of Doxorubicin and Daunorubicin in MOLT 4 Cells. \u003cem\u003ePigment Cell. Res.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, 351\u0026ndash;354 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFedele, C. et al. Prospective Isolation According to Melanin Pigment Content of Melanoma Cells With Heterogeneous Potentials for Disease Propagation. \u003cem\u003ePigment Cell. Melanoma Res\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e, (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSauna, Z. E. \u0026amp; Ambudkar, S. V. Evidence for a requirement for ATP hydrolysis at two distinct steps during a single turnover of the catalytic cycle of human P-glycoprotein. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 97, 2515\u0026ndash;2520 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, X., Aguanno, D., Board, M. \u0026amp; Callaghan, R. Exploiting the metabolic energy demands of drug efflux pumps provides a strategy to overcome multidrug resistance in cancer. \u003cem\u003eBiochim. Biophys. Acta BBA - Gen. Subj.\u003c/em\u003e \u003cb\u003e1865\u003c/b\u003e, 129915 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBroxterman, H. J., Pinedo, H. M., Kuiper, C. M., Schuurhuis, G. J. \u0026amp; Lankelma, J. Glycolysis in P-glycoprotein‐overexpressing human tumor cell lines Effects of resistance‐modifying agents. \u003cem\u003eFEBS Lett.\u003c/em\u003e \u003cb\u003e247\u003c/b\u003e, 405\u0026ndash;410 (1989).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWartenberg, M. et al. Glycolytic pyruvate regulates P-Glycoprotein expression in multicellular tumor spheroids via modulation of the intracellular redox state. \u003cem\u003eJ. Cell. Biochem.\u003c/em\u003e \u003cb\u003e109\u003c/b\u003e, 434\u0026ndash;446 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, J. Q. \u0026amp; Russo, J. Dysregulation of glucose transport, glycolysis, TCA cycle and glutaminolysis by oncogenes and tumor suppressors in cancer cells. \u003cem\u003eBiochim. Biophys. Acta BBA - Rev. Cancer\u003c/em\u003e. \u003cb\u003e1826\u003c/b\u003e, 370\u0026ndash;384 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang, Y. C., Chan, M. H., Yang, Y. F., Li, C. H. \u0026amp; Hsiao, M. Glucose transporter 4: Insulin response mastermind, glycolysis catalyst and treatment direction for cancer progression. \u003cem\u003eCancer Lett.\u003c/em\u003e \u003cb\u003e563\u003c/b\u003e, 216179 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa, W., Li, Z., Wen, J. \u0026amp; Yan, S. The expression and prognostic value of multiple glucose transporters in cutaneous melanoma. \u003cem\u003eAnticancer Res.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 3747\u0026ndash;3756 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamilton, K. E., Bouwer, M. F., Louters, L. L. \u0026amp; Looyenga, B. D. Cellular binding and uptake of fluorescent glucose analogs 2-NBDG and 6-NBDG occurs independent of membrane glucose transporters. \u003cem\u003eBiochimie\u003c/em\u003e \u003cb\u003e190\u003c/b\u003e, 1\u0026ndash;11 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonzalez-Menendez, P. et al. GLUT1 protects prostate cancer cells from glucose deprivation-induced oxidative stress. \u003cem\u003eRedox Biol.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 112\u0026ndash;127 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePisco, A. O., Jackson, D. A. \u0026amp; Huang, S. Reduced Intracellular Drug Accumulation in Drug-Resistant Leukemia Cells is Not Only Solely Due to MDR-Mediated Efflux but also to Decreased Uptake. \u003cem\u003eFront Oncol\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeetla, C. et al. Drug Resistance in Breast Cancer Cells: Biophysical Characterization of and Doxorubicin Interactions with Membrane Lipids. \u003cem\u003eMol. Pharm.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 2334\u0026ndash;2348 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLothstein, L., Hsu, S. I., Horwitz, S. B. \u0026amp; Greenberger, L. M. Alternate Overexpression of Two Phosphoglycoprotein Genes Is Associated with Changes in Multidrug Resistance in a J774.2 Cell Line. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cb\u003e264\u003c/b\u003e, 16054\u0026ndash;16058 (1989).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsu, S. I., Lothstein, L. \u0026amp; Horwitz, S. B. Differential overexpression of three mdr gene family members in multidrug-resistant J774.2 mouse cells. \u003cem\u003eJ. Biol. Chem.\u003c/em\u003e \u003cb\u003e264\u003c/b\u003e, 12053\u0026ndash;12062 (1989).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAchkar, I. W. et al. Metabolic Signatures of Tumor Responses to Doxorubicin Elucidated by Metabolic Profiling in Ovo. \u003cem\u003eMetabolites\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 268 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, R., Vander Heiden, M. G. \u0026amp; Rudin, C. M. Genotoxic exposure is associated with alterations in glucose uptake and metabolism. \u003cem\u003eCancer Res.\u003c/em\u003e \u003cb\u003e62\u003c/b\u003e, 3515\u0026ndash;3520 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKciuk, M. et al. Doxorubicin\u0026mdash;An Agent with Multiple Mechanisms of Anticancer Activity. \u003cem\u003eCells\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 659 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, Q. et al. DNA-Damage-Induced Type I Interferon Promotes Senescence and Inhibits Stem Cell Function. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 785\u0026ndash;797 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheon, H., Wang, Y., Wightman, S. M., Jackson, M. W. \u0026amp; Stark, G. R. How cancer cells make and respond to interferon-I. \u003cem\u003eTrends Cancer\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e, 83\u0026ndash;92 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePidugu, V. K., Pidugu, H. B., Wu, M. M., Liu, C. J. \u0026amp; Lee, T. C. Emerging Functions of Human IFIT Proteins in Cancer. \u003cem\u003eFront. Mol. Biosci.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 148 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeichselbaum, R. R. et al. An interferon-related gene signature for DNA damage resistance is a predictive marker for chemotherapy and radiation for breast cancer. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 105, 18490\u0026ndash;18495 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, H. et al. Bone marrow stromal cell antigen 2: Tumor biology, signaling pathway and therapeutic targeting (Review). \u003cem\u003eOncol. Rep.\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e, 1\u0026ndash;12 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrabosch, S. et al. Cisplatin-induced immune modulation in ovarian cancer mouse models with distinct inflammation profiles. \u003cem\u003eOncogene\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e, 2380\u0026ndash;2393 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUchihara, Y. et al. DNA damage promotes HLA class I presentation by stimulating a pioneer round of translation-associated antigen production. \u003cem\u003eMol. Cell.\u003c/em\u003e \u003cb\u003e82\u003c/b\u003e, 2557\u0026ndash;2570e7 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeifert, U. et al. Immunoproteasomes Preserve Protein Homeostasis upon Interferon-Induced Oxidative Stress. \u003cem\u003eCell\u003c/em\u003e \u003cb\u003e142\u003c/b\u003e, 613\u0026ndash;624 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBayar, I., Ekren Aşıcı, G. S., Bildik, A. \u0026amp; Kıral, F. Gene Expression of Glycolysis Enzymes in MCF-7 Breast Cancer Cells Exposed to Warburg Effect and Hypoxia. \u003cem\u003eInt J. Mol. Cell. Med\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMengistu, B. A. et al. Comprehensive review of drug resistance in mammalian cancer stem cells: implications for cancer therapy. \u003cem\u003eCancer Cell. Int.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 406 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, C., Tian, C., Hoffman, T. E., Jacobsen, N. K. \u0026amp; Spencer, S. L. Melanoma subpopulations that rapidly escape MAPK pathway inhibition incur DNA damage and rely on stress signalling. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 1747 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, S., Zhou, Y., Chen, Y. \u0026amp; Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, i884\u0026ndash;i890 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatro, R., Duggal, G., Love, M. I., Irizarry, R. A. \u0026amp; Kingsford, C. Salmon provides fast and bias-aware quantification of transcript expression. \u003cem\u003eNat. Methods\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e, 417\u0026ndash;419 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoneson, C., Love, M. I. \u0026amp; Robinson, M. D. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. \u003cem\u003eF1000Research\u003c/em\u003e 4, 1521 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuzellec, B., Teleńczuk, M., Cabeli, V. \u0026amp; Andreux, M. PyDESeq2: a python package for bulk RNA-seq differential expression analysis. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e39\u003c/b\u003e, btad547 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubhash, S. \u0026amp; Kanduri, C. GeneSCF: a real-time based functional enrichment tool with support for multiple organisms. \u003cem\u003eBMC Bioinform.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 365 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo, W. \u0026amp; Brouwer, C. Pathview: an R/Bioconductor package for pathway-based data integration and visualization. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e, 1830\u0026ndash;1831 (2013).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":false,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"P-gp, chemoresistance, doxorubicin, melanoma, GLUT, carbohydrate transport","lastPublishedDoi":"10.21203/rs.3.rs-8929076/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8929076/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eModels of cancers with acquired MultiDrug Resistance (MDR) commonly employ a binary state that compares MDR to treatment na\u0026iuml;ve cohorts. While convenient, this paradigm also oversimplifies the dynamic process of acquiring MDR as cellular processes (and corresponding therapies targeting them) could perform differently at each intermediate stage of resistance. However, comparisons of discrete levels of chemoresistance, particularly with regards to carbohydrate transport or transiently expressed genes, remain limited. Here we characterize a B16 melanoma cell panel comprising a gradient of doxorubicin resistance (1 nM to 1 \u0026micro;M). Across the MDR gradient, we observed minimal changes in class I glucose transporter (GLUT) expression and reduced carbohydrate influx, despite increases in P-glycoprotein (P-gp) efflux. This suggests enhanced P-gp efflux is not compensated by GLUT-mediated influx, but rather, attenuated carbohydrate metabolism. Indeed, RNA-seq revealed decreases in glycolytic enzymes alongside additional interferon-stimulated genes (including \u003cem\u003eIsg15\u003c/em\u003e and \u003cem\u003eBst2\u003c/em\u003e) with transient differential expression only at intermediate levels of MDR. Overall, this study provides proof-of-principle that selection of a particular level of MDR is non-trivial, particularly for studies involving the transiently expressed genes or carbohydrate processing, and might explain why there remains significant debate over which in-vitro level of MDR is optimally prognostic of in-vivo performance.\u003c/p\u003e","manuscriptTitle":"Beyond Binary: Mapping the Evolution of Melanoma Across a Discrete Gradient of Acquired Chemoresistance","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-26 06:13:50","doi":"10.21203/rs.3.rs-8929076/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-24T13:32:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-24T08:51:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-24T08:49:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-02-20T20:51:49+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1630ac6b-712e-434e-aaae-ce12fb82a53a","owner":[],"postedDate":"February 26th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":63453861,"name":"Biological sciences/Cancer"},{"id":63453862,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2026-03-25T16:38:18+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-26 06:13:50","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8929076","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8929076","identity":"rs-8929076","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}