NOHA as a key biomarker to link NOS2 activity to immune checkpoint and cytokine signaling in ER-negative breast cancer.

1
Introduction
Breast cancer is a biologically diverse disease and remains one of the most common and deadly cancers in women worldwide [[1], [2], [3], [4], [5]]. Classification into molecular subtypes based on hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) expression informs prognosis and guides treatment [[6], [7], [8], [9]]. Among HR expression subtypes, estrogen receptor-negative (ER−) breast cancers present a significant clinical challenge due to their aggressive nature, early metastatic potential, and lack of targeted therapies [10]. In contrast to estrogen receptor-positive (ER+) subtypes, where endocrine therapies are effective and receptor status is central to diagnosis and monitoring, ER− breast cancers are often diagnosed through imaging and tissue biopsy alone, with limited utility for longitudinal biomarker surveillance [[11], [12], [13]]. The urgent clinical need for diagnostic and disease monitoring tools in ER− breast cancer has led to initial investigations into immune-related biomarkers, particularly immune checkpoint molecules [14,15] and circulating cytokines [16].
ER− tumors frequently demonstrate an immunologically “hot” tumor microenvironment (TME), with elevated expression of immune checkpoint molecules like PD-1, PD-L1, and CTLA-4, which correlate with both disease aggressiveness and responsiveness to immune checkpoint inhibitor (ICI) therapies [17,18]. These markers not only influence response to immune checkpoint inhibitors (ICIs) but have been explored as potential diagnostic and prognostic tools [19,20]. For example, PD-L1 expression correlates with immune cell infiltration and clinical outcomes in subsets of ER− breast cancer [21].
Despite their biological relevance, the diagnostic use of immune checkpoint markers faces several challenges [22]. PD-L1 and related markers exhibit significant inter- and intra-tumoral heterogeneity, making consistent interpretation across biopsy sites or timepoints difficult [23]. Multiple commercial PD-L1 assays exist, each with distinct antibodies and scoring systems, leading to inconsistent results and poor cross-study comparability [24,25]. While PD-L1 may predict response to ICIs, its standalone value for early diagnosis or long-term prognosis remains unclear and often confounded by other tumor or host immune variables [22]. Additionally, immune checkpoint marker evaluation typically requires tumor tissue, limiting its role in real-time disease monitoring or longitudinal follow-up [22]. These limitations reduce the feasibility of incorporating immune checkpoint markers into routine diagnostics, although they remain valuable in selecting candidates for immunotherapy.
In parallel, pro-inflammatory cytokines such as interleukin-6 (IL-6), IL-1β, IL-8, and TNF-α are frequently upregulated in ER− breast cancers and play critical roles in tumor progression, immune modulation, and metastasis in ER− breast cancer [[26], [27], [28]]. Despite the mechanistic relevance of cytokines in ER− breast cancer biology for disease diagnostics or monitoring, their clinical use in ER− breast cancer remains experimental due to key limitations [[29], [30], [31]]. First, cytokine levels are inherently dynamic and influenced by systemic inflammation, infections, stress, and comorbid conditions, which complicates the specificity and reproducibility needed for reliable cancer diagnostics [30]. Second, inter-patient variability in cytokine expression and the absence of standardized reference ranges or cutoff thresholds hinder clinical interpretation [31]. Third, although certain cytokines correlate with disease stage or metastasis, they lack the sensitivity and specificity required to differentiate malignant from benign or inflammatory processes in a diagnostic setting [29].
Nevertheless, while immune checkpoint molecules and cytokines are central to disease biology, their clinical translation is limited by technical and biological challenges. Thus, there remains a significant unmet need for robust, less-invasive biomarkers in ER− breast cancer for early diagnosis, disease monitoring, and therapy outcome prediction [[32], [33], [34]].
Nitric oxide synthase-2 (NOS2) has been identified as a key feed-forward signaling regulator, promoting chronic inflammation and breast cancer progression [35]. NOS2 expression is commonly elevated in breast cancer and other cancer types (i.e. glioma, melanoma, cervical, ovarian, and pancreatic). In our prior studies, we have demonstrated that Nw-hydroxy-l-Arginine (NOHA), a stable NOS2 intermediate, is a highly sensitive marker that distinguishes ER– from ER + disease status in 3D spheroid cell culture medium and in breast cancer patient plasma samples [[36], [37], [38]]. Our earlier 3D spheroid cell culture studies, involving 12 distinctive 3D spheroid cell culture models representing CA and AA ethnic origins for healthy non-cancerous and ER breast cancer subtypes showed distinctive medium NOHA values of 68.3 ± 6.2 nM in healthy control spheroids versus 20.4 ± 2.1 nM of NOHA in ER– subtype [38]. Additionally, ER– 3D spheroid medium NOHA levels correlated with disease stage, grade, and molecular phenotype [37]. This distinction in medium NOHA level between healthy and ER– breast cancer subtype resonated in our pilot clinical trials among US and Tanzanian patient populations, with NOHA values of <4 nM in ER–, and > 8 nM among healthy patient samples [36,39]. Our patented [40,41] and validated NOHA measurement by enzyme-linked immunosorbent assay (ELISA) [42] in dry and fresh patient plasma samples further supports its utility potential for diagnosis, disease monitoring and therapy outcome measurement.
In ER− breast cancer, NOHA is particularly relevant due to its ability to: Modulate cytokine release, including IL-6, IL-8, and TNF-α, via NO-mediated activation of NF-κB and STAT3 pathways [43]; influence immune checkpoint expression, including PD-L1, through nitric oxide–dependent signaling cascades that affect T cell and macrophage function [44], and further serve as a metabolic indicator of immune activation within the TME [32,33,45,46], linking metabolic inflammation to disease progression.
Such an integrative role of NOHA in immune metabolism and signaling positions it uniquely as a potential biomarker at the intersection of immune checkpoint activity and cytokine-driven tumor progression. In this in vitro study, we utilize three-dimensional (3D) spheroid models representing ER–and ER + breast cancer subtypes, derived from Caucasian (CA) and African American (AA) patient origins, to offer a foundational determination of whether NOHA can function as a key biomarker capable of simultaneously reflecting the dynamic interplay between immune checkpoint signaling and pro-inflammatory cytokine activity. The 3D spheroid model offers a controlled and physiologically relevant system to elucidate the potential of NOHA response.

2
Materials and methods
2.1
Supplies, reagents and kits
All cell culture reagents and supplies were obtained from Invitrogen (Carlsbad, CA), and analytical-grade chemicals (≥95 % purity) were sourced from Sigma-Aldrich (St. Louis, MO). General laboratory consumables were procured from VWR International (Bridgeport, NJ) and CellTreat (Pepperell, MA). Commercial assay kits were used for the following: NOS2 expression (Antibodies-online, Atlanta, GA); NOS2 enzymatic activity via total nitrite detection by fluorometric assay (Cayman Chemical, Ann Arbor, MI); total PD-L1 expression (Cell Signaling Technology, Danvers, MA); and IL-8 expression (Abbexa, Sugar Land, TX). All experimental procedures utilized 18.2 MΩ deionized water.

2.2
Cell culture
Human breast cancer cell lines of CA and AA origin were obtained from the American Type Culture Collection (ATCC, Manassas, VA), that involved – MDA-MB-436 (ER–, CA), MDA-MB-157 (ER–, AA), MCF7 (ER+, CA), and HCC1007 (ER+, AA). Non-tumorigenic control cell lines (i.e., Human Mammary Epithelial Cells, HMECs) were also obtained from ATCC for this study. All cells were cultured in T75 flasks to about 80 % confluence in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 12 % fetal bovine serum (FBS), 2 mM l-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin, at 37 °C in a humidified atmosphere with 5 % CO2 for 7 days.
Cells were harvested by washing twice with 1X phosphate-buffered saline (PBS, 5 mL/wash), followed by incubation with 0.25 % trypsin-EDTA (1 mL) at 37 °C for 2–3 min to detach adherent cells. Trypsin treated cells were neutralized with 1 mL of complete DMEM, centrifuged at 300×g for 5 min, washed twice with PBS (1 mL/wash), and resuspended in 0.5 mL DMEM for subsequent 3D spheroid culture.

2.3
3D-spheroid generation and lysis
Spheroids were formed by mixing 200 μL of cell suspension (from ER–, ER+, or control lines) with 30 μL of Cultrex® Bio-Techne spheroid formation medium containing extracellular matrix proteins from R&D systems (Gaithersburg, MD). Mixtures were plated in triplicate in 96-well InSphero GravityPLUS™ hanging drop plates (PerkinElmer, Hopkinton, MA) at 60 μL/well (∼1 × 104 cells/well). Cultures were maintained at 37 °C and 5 % CO2, with medium changed daily. Culture media were collected weekly for NOHA quantification.
Spheroid growth was monitored every other day in any given week using a high-content imaging system (PerkinElmer, Waltham, MA) with a 2× objective. Spheroid viability and cell cycle status were assessed weekly, and before cell extraction, using Guava® 8HT flow cytometry assays (Millipore, MA). Spheroids that maintained close histological similarity to control/healthy cells (i.e., well differentiated) and with an overall size measurement in volume of ≤250 μm were characterized as “low-grade. 3D-spheroids with tumor size >400 μm, and with less characteristic similarity to control spheroids were categorized as “intermediate-grade”. Longitudinal spheroid growth were monitored, and a lack of measurable increase in tumor size across consecutive weeks were defined as growth plateauing. This plateau in spheroid growth were used to determine the effective study duration for each experimental group. A final consensus on a standardized 3D spheroid study duration of 7-weeks got established based on overall 3D-spheroid growth assessment in all tested groups.
At the end of each week, spheroids were washed twice with 500 μL of PBS and incubated with 250 μL of 0.25 % trypsin-EDTA at 37 °C for 1–3 min with gentle agitation to dissociate the spheroids. A 20 μL aliquot of the dissociated cell suspension was used to assess viability using the ViaCount® kit (Millipore) on the Guava 8HT cytometer. The remaining suspension was centrifuged at 300×g for 5 min, and the resulting cell pellet was lysed in 500 μL of ice-cold lysis buffer (20 mM Tris, pH 7.5; 150 mM NaCl; 1 % Triton X-100; 2.5 mM sodium pyrophosphate; 1 mM β-glycerophosphate; 1 mM PMSF) at 2–8 °C for 5 min. Lysates were vortexed for 2 min, centrifuged at 14,000×g for 5 min at 2–4 °C, and 400 μL of the supernatant was collected for downstream quantification of cellular NOHA, PD-L1, IL-8, NOS2, and nitric oxide.

2.4
NOHA quantification
NOHA levels were quantified using our patented, and previously validated competitive ELISA developed in-house, which utilizes a proprietary monoclonal antibody (mAb) and serves as a sensitive alternative to LC–MS-based detection [42]. Our prior validation studies have shown the NOHA assay to maintain sensitivity at ≥60 pg/ml to NOHA antigen with 1 ng/ml NOHA mAb use, and to maintain NOHA antigen specificity even in the presence of other closely related cationic amino acids (i.e. LArginine, d-Arginine, l-Lysine, d-Lysine, l-Ornithine, and l-Citrulline) [42]. The reliability of the ELISA protocol were confirmed with the low percent-covariance, for all tested parameters of sensitivity (≤8.2 %), selectivity (≤8.6 %), precision (≤12.6 %), dilution linearity (≤11.2 %) and recovery (≤6.7 %) [42].
BSA–NOHA–coated strips were washed three times with 200 μL of 1X PBS. Each well received 100 μL of a complex mixture (80 μL of either sample lysate or standard, combined with 20 μL of NOHA–mAb at 5 ng/mL) and was incubated for 1 h at 25 °C. After discarding the contents, wells were washed eight times with 200 μL of PBS. A horseradish peroxidase-conjugated secondary antibody (Abcam, Cambridge, MA), diluted 1:20,000 in PBS, was added (100 μL/well) and incubated for 1 h at 25 °C. Following another set of eight washes, 100 μL of TMB substrate (MossBio, Pasadena, MD) was added and incubated in the dark for 10 min. The reaction was terminated with 100 μL of 0.1 N HCl. Absorbance was measured at 450 nm using a VersaMax® spectrophotometer (Molecular Devices, NH). A second-order polynomial standard curve was generated, and R2 values were used to assess curve fitting and confidence in NOHA quantification.

2.5
Statistical analysis
All data are expressed as mean ± standard deviation from independent spheroids harvested weekly for 7-weeks, with six individual biological sample replicates per treatment time point each week (n = 6) and averaged, unless stated otherwise. Ratio analysis for NOHA/PD-L1 and NOHA/IL8 were calculated from six independent biological sample replicates per treatment time point each week, and averaged for interpretation. All statistical comparisons over 7-weeks of dataset derived from each week independently harvested spheroid samples were performed using one-way analysis of variance (ANOVA), followed by Fisher's Least Significant Difference (LSD) and Tukey's post hoc tests, using GraphPad Prism (v10.2, La Jolla, CA). A p-value of <0.01 was considered statistically significant.

3
Results
3.1
3D spheroid model analyte measurements
Our in vitro studies showed a progressive ≥1-fold increase in cellular NOS2, total nitrites, PD-L1, and IL8, with a ≥1-fold decrease in both cellular and extracellular NOHA over a 7-week culture period in ER– AA and ER– CA 3D spheroid groups (p < 0.01; Fig. 1). In contrast, the ER+ and control 3D spheroid groups showed no significant changes in these parameters throughout the study period (Fig. 1).
In ER– AA spheroids, cellular and medium NOHA levels declined weekly by ≥ 8.2 % and ≥11.3 %, respectively, while PD-L1 and IL8 expression increased by ≥ 20.4 % and ≥23.9 %, respectively. These cultures also exhibited weekly increases in cellular NOS2 and total nitrites by ≥ 11.1 % and ≥19.8 %, respectively. Similarly, ER– CA spheroids showed weekly reductions in cellular and medium NOHA by ≥ 5.2 % and ≥7.3 %, respectively, along with increases in PD-L1 and IL8 expression by ≥ 16.4 % and ≥18.1 %, respectively. NOS2 and total nitrite levels also rose in ER– CA cultures, with weekly gains of ≥6.1 % and ≥16.7 %, respectively.

3.2
NOHA to PD-L1 ratio comparison in ER– groups
To evaluate the relationship between NOHA depletion and PD-L1 upregulation, we conducted a longitudinal ratio analysis comparing cellular PD-L1 expression to medium NOHA levels across ER− 3D spheroid groups. This analysis demonstrated a progressive increase in the PD-L1/NOHA ratio over the 7-week culture period, with week-to-week change of ≥33.8 % in the ER− AA group and ≥36.4 % in the ER− CA group. Importantly, at each week time point, the PD-L1/NOHA ratio mean was consistently ≥43.8 % higher in ER− AA spheroids compared to ER− CA spheroids (Fig. 2A, p < 0.01).
A corresponding analysis using cellular PD-L1 and cellular NOHA levels yielded a similar trend with week-to-week change of ≥31.5 % in the ER− AA group and ≥28.9 % in the ER− CA group. Over the same 7-week time frame, the PD-L1/NOHA ratio mean was ≥52.4 % higher in ER− AA spheroids relative to ER− CA spheroids, further supporting a more pronounced immune checkpoint activation associated with NOHA depletion in the AA-derived ER− models (Fig. 2B, p < 0.01).

3.3
NOHA to IL8 ratio comparison in ER– groups
To determine the connection between NOHA reduction and IL8 upregulation, we conducted a week-to-week ratio analysis of IL8 expression to NOHA for each ER– 3D spheroid group. This analysis demonstrated a progressive increase in the % ratio variation, with week-to-week change of ≥21.2 % in the ER– AA group and ≥29.9 % in the ER– CA group. Of interest, at any given week over the 7-week culture period, the IL8/NOHA ratio mean was ≥48.4 % higher in ER– AA 3D spheroids from ER– CA spheroids (Fig. 3A, p < 0.01).
A parallel analysis using cellular IL-8 and cellular NOHA levels yielded a comparable trend, with week-to-week changes of ≥32.4 % in the ER− AA group and ≥30.7 % in the ER− CA group. Over the full 7-week culture period, the IL-8/NOHA ratio mean was ≥55.9 % higher in ER− AA spheroids than in ER− CA spheroids, reinforcing the association between NOHA depletion and enhanced pro-inflammatory signaling in AA-derived ER− models (Fig. 3B, p < 0.01).

3.4
Regression assessment of medium NOHA to PD-L1 and IL8 in ER– AA and ER–CA groups
Linear regression analysis of ER– 3D spheroid cultures revealed strong inverse associations between medium NOHA concentrations and both PD-L1 and IL-8 protein abundance in ER– CA (i.e., MDB-436) and ER– AA (i.e., MDA-MB157) models (Fig. 4). In ER– CA spheroids, NOHA levels were tightly correlated with reduced PD-L1 (R2 = 0.963) and IL-8 (R2 = 0.972), and similarly strong correlations were observed in ER– AA spheroids for both PD-L1 (R2 = 0.973) and IL-8 (R2 = 0.988). Across all panels (Fig. 4. A–D), increase in NOHA corresponded to a steep, monotonic decline in PD-L1 and IL-8 expression, with nearly overlapping regression slopes between CA and AA groups. These findings demonstrate that medium NOHA is a highly predictive metabolic indicator of PD-L1 and IL-8 levels in ER– breast cancer 3D spheroids.

4
Discussion
This study provides the first evidence that NOHA acts as an key biomarker linking immune checkpoint activation and inflammatory cytokine signaling in ER− breast cancer. Using physiologically relevant 3D spheroid models derived from both AA and CA patient derived breast cancer cell lines, we observed a consistent inverse relationship between NOHA levels and both PD-L1 and IL-8 expression, specifically under ER− conditions. In contrast, no such trends were observed in ER + or control spheroids, underscoring the subtype specificity and potential clinical utility of NOHA in the ER− breast cancer landscape.
Our findings complement prior work demonstrating that ER− breast tumors are immunologically active and often characterized by high PD-L1 expression and pro-inflammatory cytokine release [47,48]. These immune features, while mechanistically important, have shown limited reliability as diagnostic or prognostic markers due to their temporal variability, contextual expression, and dependence on tumor tissue [22,[49], [50], [51]]. Our data suggest that NOHA detectable in extracellular medium may overcome such limitations by offering a non-invasive, and biologically meaningful surrogate for immune activation in ER− breast cancer.
Using cell lines such as MDA-MB-436, MDA-MB-157, MCF7 and HCC1007 in an in vitro study such as this is helpful to approximate some aspects of the heterogeneity observed in human breast cancer at the population level [52]. By including cell lines from different molecular subtypes (e.g., luminal/ER-positive such as MCF7, and triple-negative/basal-like such as MDA-MB-436 or MDA-MB-157), the experiment captures variability in hormone-receptor status, differentiation state, and aggressiveness, which mirrors the diversity of breast cancer clinical presentations [53,54]. Studies have shown that cell lines differ markedly from one another in biomarker expression, signaling pathways, mechanical properties, and metabolic phenotypes; even among those classified under the same broad subtype [[52], [53], [54]].
At the same time, it is important to recognize the limitations of this approach. Even a diverse panel of cell lines does not replicate the full intra-tumor heterogeneity of human cancers as individual cell lines tend to adopt lineage-restricted phenotypes in culture and lose many of the rare cell subpopulations that may be present in actual tumors [55,56]. Moreover, long-term culture induces genetic drift, copy-number alterations, and selection for clones that adapt well to plastic-dish conditions; changes that may diverge substantially from the original tumor biology [57]. As a result, while such a cell-line panel improves the external validity of in vitro findings compared with using a single line, the model remains an imperfect and reduced representation of the patient population.
Compared to traditional two-dimensional cultures use, our study involved 3D spheroids better replicate tumor architecture, nutrient gradients, and cellular interactions, which are essential for modeling immune-metabolic shifts over extended culture duration and enabling tracking of tumor evolution [[58], [59], [60]]. The week-to-week spheroid characterization that was involved for this study helped us to determine the overall longitudinal growth time frame as 7-weeks. This coincides with the plateau in spheroid growth observed after 7–8 weeks in our [38] and other previous studies [61,62] further confirms the physiological relevance for adopting 7-week timeframe for this in vitro study design.
Prior studies have shown NOS2-derived NOHA can modulate NF-κB and STAT3 signaling pathways [43,63], both of which drive pro-inflammatory cytokine production and PD-L1 upregulation [64,65]. In this context, the progressive decline in NOHA concurrent with rising PD-L1 and IL-8 levels over 7-weeks suggests a plausible mechanistic link that warrents further investigation. The close association between dcreasing NOHA levels with increased NOS2 activity and elevated nitrite production [37] further supports NOHA role as a dynamic metabolic analyte in the tumor immune microenvironment. These findings align with studies identifying NOS2 as a mechanistic indicator of immune evasion and tumor aggressiveness, particularly in ER− and triple-negative breast cancers. The progressively rising PD-L1/NOHA and IL8/NOHA ratios in ER− spheroids, especially among AA spheroid group, highlight potential for anticipating biological differences by ethnicity in future translational studies involving NOHA clinical utility testing among high risk subgroups.
The consistently high R2 values seen across all four regression analyses in Fig. 4, underscore a conserved and robust relationship between NOHA abundance and the suppression of key immunomodulatory (PD-L1) and pro-inflammatory (IL-8) proteins in ER– breast cancer, regardless of ethnic background. The nearly identical dose–response behavior between ER– AA and ER– CA spheroids suggests that NOHA regulates these pathways via shared biological mechanisms rather than ancestry-specific factors. Future mechanistic studies will be essential to determine whether NOHA directly modulates transcriptional or signaling pathways governing PD-L1 and IL-8 expression, and whether manipulating NOHA in vivo could shift the immune microenvironment toward a more therapeutically favorable state.
While this study findings support the hypothesis that NOHA reduction reflects a metabolic shift associated with tumor progression, NOS2 upregulation, and immune checkpoint activation, making it a promising candidate in future delineating utility this study has its limitations. First, while the in vitro 3D spheroid models provide important mechanistic insights, in vivo validation in patient-derived xenografts or clinical cohorts is necessary to confirm translatability. Second, while IL-8 and PD-L1 were assessed here, future studies should expand to other cytokines (e.g., IL-6, TNF-α) and immune checkpoints (e.g., CTLA-4) to fully map NOHA's regulatory network. Third, despite accounting for ethnicity by including AA and CA cellular models, more comprehensive genomic profiling could clarify whether observed differences reflect intrinsic tumor biology or broader ancestral traits. Additional limitations include the absence of tumor–stromal and immune cell interactions inherent to spheroid-only systems, lack of in vivo confirmation, and the absence of direct comparisons with clinically used circulating biomarkers.

5
Conclusion
This study identifies NOHA as a key immune-metabolic marker in ER− breast cancer, capable of reflecting dynamic changes in NOS2 activity, PD-L1 expression, and IL-8 production. These findings provide a foundational framework for evaluating NOHA in future clinical studies as a potentially actionable biomarker for disease diagnosis, monitoring, and possibly guiding immunotherapy decisions in ER− breast cancer. Moreover, its utility may have broader implications in the future for addressing racial disparities in breast cancer outcomes, given the disproportionate burden of aggressive ER− and TNBC subtypes in patient populations.

Declaration of generative artificial intelligence (AI) and AI-assisted technologies in the writing process
During the preparation of this work the author(s) used ChatGPT free online version service for grammatical corrections. By using this AI tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

CRediT authorship contribution statement
Olivia Chase: Formal analysis, Methodology, Resources. Emmeline Graham: Formal analysis, Methodology. Mack Buczek: Formal analysis, Methodology, Resources. William Speltz: Data curation, Formal analysis, Methodology, Resources. Megan Steele: Data curation, Formal analysis, Methodology, Resources. Connor LaBonte: Data curation, Formal analysis, Methodology, Resources. Srinidi Mohan: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Validation, Writing – original draft, Writing – review & editing.

Declaration of competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.