Capturing the dynamics of STAT6 macrophage polarization using bioluminescence temporal signatures

SUMMARY

Signal transducer and activator of transcription 6 (STAT6) signaling is activated by interleukin 4 (IL-4) and IL-13 and drives alternative macrophage polarization, which is pivotal in wound healing, immunosuppression, and tumor progression. STAT6 functions by forming a phosphorylated homodimer and binding to STAT6-responsive promoter elements to regulate anti-inflammatory genes. Measuring STAT6 activity can serve as a proxy for assessing macrophage polarization. We developed a STAT6-responsive-element (RE) THP-1 reporter to assess STAT6 activation in response to inflammatory stimuli. We quantitatively measured macrophage polarization by using bioluminescence temporal spectrometry (BTS). Human THP-1 monocytes were transduced with lentivirus to express STAT6-RE-firefly luciferase (FLuc)-green fluorescent protein (GFP). The STAT6-RE accurately reported endogenous STAT6 activity, as confirmed by ELISA, western blotting, bioluminescence, and imaging techniques. We developed a systems model to connect emergent bioluminescence to the kinetics of relevant STAT6 cellular signaling events. These results indicate a probe for assessing macrophage alternative polarization, enhancing our understanding of the relationship between molecular mechanisms and macrophage polarization.

Graphical abstract

In brief

Zheng et al. investigate macrophage polarization by linking transcription factor activity to intracellular signaling dynamics via the use of a synthetic responsive element bioluminescent reporter system. Using computational modeling, they infer signaling timescales that match experimentally observed dynamics. This integrated platform enables quantitative, longitudinal analysis to dissect mechanisms underlying macrophage polarization.

INTRODUCTION

Macrophages display extensive plasticity, i.e., polarization states, in response to micro-environmental stimuli, including biochemical signals, matrix and mechanical cues, and cell-cell interactions.^1–4 Macrophages probe environmental stimuli by expressing pattern recognition receptors, cytokines, and chemokine receptors. Agonist binding to these receptors initiates intracellular signaling, leading to dramatic changes in transcription factor activity and, ultimately, functional changes. There are key transcriptional factors involved linked to polarization states.^5–7 For example, nuclear factor κB (NF-κB) and interferon signaling are vital in pro-inflammatory macrophage polarization.^5,8 Interleukin (IL)-4-induced signal transducer and activator of transcription 6 (STAT6) signaling dominates M2a polarization, where membrane-bound Janus kinase 1 (JAK1) and Janus kinase 3 (JAK3) phosphorylate STAT6 and promote M2-associated genes.^9,10

The characterization of macrophage polarization is a broad challenge.^11,12 Differentiation and polarization are determined by clustering sets of differentially expressed genes or proteins. These are measured by flow cytometry, RNA sequencing, and proteomics.^13–15 Those phenotyping techniques have yielded a large amount of valuable omics information about macrophage functions; however, some challenges have not been addressed, including bulk population quantification, tissue homogenization, poorly defined macrophage subset markers, loss of spatial pattern, and lack of experimental continuum. Indeed, with advances in current techniques, spatial transcriptomics conserves the spatial aspect, and RNA sequencing and proteomics can be achieved at single-cell resolution. Still, they only offer one opportunity to measure cell behavior.^16–18 Moreover, these measurements do not reflect the temporal determinants of cell fate. Macrophages continuously process chemical and mechanical stimuli that vary over space and time. Given the macrophages’ dynamic sensory nature, there is a need to measure macrophage cell fate equally dynamically.

Understanding macrophage heterogeneity at a given time and space point can benefit predicting macrophage functions and deciphering the mechanism of disease progression. Here, we developed a bioluminescence temporal spectrometry (BTS) method to measure real-time changes in macrophage polarization states quantitatively. Our system includes the development of responsive-element (RE) luciferase-tagged macrophage reporter cells and computational approaches to assess the temporal signals generated. We created a STAT6-RE reporter to measure STAT6 activation in immortalized and primary macrophages. The dynamic curve can reflect the abundance of a specific genre of environmental stimuli and predict cell fate decisions.

BTS refers to repeatedly measuring the bulk luminescence intensity emitted from luciferase reporter cells at any given time. Bioluminescence is the light emitted when luciferase enzymes catalyze the oxidation of a substrate. Luciferase enzymes have a relatively short half-life, around 2 h, enabling the quantitative measurements of protein expression kinetics.¹⁹ The substrate has negligible toxicity, accommodating long-term measurements in vitro and in vivo.^20,21 Although the photon signal is lower than in fluorescent microscopy, bioluminescence is more sensitive due to the absence of endogenous auto-fluorescence.²² Moreover, bioluminescence does not require an external light source, eliminating the risk of photobleaching, sample damage from heating, and interference from light scattering and light source drift.²² Due to the short half-life of luciferase, low toxicity of substrate, and the high yield of luminescent emission, luciferases have been reported to monitor in vivo and in vitro physiological processes such as cellular cytotoxicity, gene expression, longitudinal cell migration, and survival.^19,23

RESULTS

Establishing an inducible luciferase platform for studying M2 macrophage polarization

The THP-1 cell line is a common immortalized human monocytic model to study macrophage polarization.^24–27 Pro-inflammatory features in macrophages have drawn extensive attention; however, anti-inflammatory phenotypes have not been well characterized but have many subsets. Of particular interest is the anti-inflammatory polarization associated with wound healing (i.e., M2a or alternative activation). M2a macrophage polarization is largely driven by IL-4 and the JAK-STAT6 signaling pathway.^9,10 To measure the dynamics of M2a macrophage polarization, we designed a lentivector with a promoter containing four repeats of STAT6-REs and mini-cytomegalovirus (CMV). The STAT6-RE sequence was amplified from the p4xSTAT6-Luc2P plasmid to construct the lentivector. Then, the sequence ends were tethered with XhoI and NheI restriction sites by PCR. Gel electrophoresis was performed to confirm that the molecular size of the amplified STAT6-RE sequences was 229 bp. Restriction enzymes XhoI and NheI were used to digest amplified STAT6-RE sequences and the lentiviral backbone pGreenFire 2.0 mCMV plasmid. The digested STAT6-RE inserts and the digested lentiviral backbone were mixed with a 3:1 mass ratio followed by the ligation process (Figure 1A). The sequence map indicates the promoter encodes four STAT6-CCAAT/enhancer binding protein (C/EBP) binding sites, allowing the recognition of phospho-STAT6 (p-STAT6) homodimers (Figure 1B). The reporter genes include a short half-life red firefly luciferase (rFluc) and a relatively long half-life green fluorescent protein (GFP) under the regulation of the STAT6-RE promoter. The puromycin gene is under the regulation of the mini phosphoglycerate kinase (PGK) promoter for transduced cell selection (Figure 1C). STAT6-RE specific primers were designed to confirm the successful insertion of the STAT6-RE sequence into the lentiviral backbone. Gel electrophoresis data supported the successful construction of the STAT6-RE lentivector (Figure 1C). The expression plasmid was sequenced, and BLASTN was performed to assess the sequence fidelity (Table S1). The measured STAT6-RE lentivector sequence displayed a 99.98% identity compared with the reference sequence. Wild-type (WT) THP-1 monocyte was incubated with the STAT6-RE lentivirus with the MOI = 3.02. The lentivirus-containing media were replaced by fresh RPMI complete media after a 3-day incubation. 1 μg/mL puromycin was supplemented to the RPMI complete media to remove non-transduced THP-1 cells. Provirus copy in STAT6-RE THP-1 cells was confirmed to be 15.62 copies/cell. This STAT6-RE THP-1 cell line was used in the following experiments.

Figure 1. Establishing an inducible luciferase reporter for monitoring macrophage M2 polarization.

(A) Construction of STAT6-RE lentivector with subcloning technique. Band (in the red box) in gel electrophoresis represents the PCR-amplified STAT6-RE sequence tethered XhoI and NheI restriction sites.

(B) Diagram of four repeats of STAT6-C/EBP binding sites within the promoter region.

(C) The expression constructs were transduced to WT THP-1 monocytes through lentiviral transduction. The band (in the red box) in gel electrophoresis represents the PCR-amplified STAT6-RE sequence from the STAT6-RE lentivector.

(D) Workflow of STAT6-RE lentiviral transduction to WT THP-1 monocytes to construct the STAT6-RE THP-1 cell line.

Lentiviral transduction enables luciferase reporting during M2 polarization without impairing native STAT6 signaling

The addition of transgenes can impair the activity of the naive signaling dynamics.^28,29 To assess the intactness of polarization capacity and molecular machinery in the STAT6-RE THP-1 cell line, qPCR was performed to compare polarization-associated mRNA expression between WT and STAT6-RE THP-1 cells during IL-4 exposure. For the selected time points (12, 24, 48, and 72 h), M2a-associated genes (gata3, ccl22, ccl11, socs1, mrc1, cd86, il-24, harg1, and il-4)^30,31 and M1-associated genes (hnos2, ifnγ, il12b, cd80, hla-drα, cd64, il6, and tnfα)³² displayed a comparable expression profile in WT and STAT6-RE THP-1 cells (Figures 2A–2D). Some genes (gata3, ccl22, ccl11, socs1, mrc1, cd86, il-4, and hnos2) displayed a decrease in the STAT6-RE THP-1 cell line compared with WT THP-1 at 24-h post IL-4 activation (Figure 2B). However, all other gene expression patterns in STAT6-RE THP-1 matched the BTS-generated trajectory, suggesting the BTS curve effectively reported the functional transition of an M2 phenotype.

Figure 2. Assessing the functionalities and molecular machinery in the STAT6-RE THP-1 cell line.

(A–D) Expression profiles of polarization-associated genes in IL-4-induced phenotype M2a (IL-4) relative to unactivated phenotype for WT (gray) and STAT6-RE THP-1 (blue) macrophages. qPCR was performed at 12 (A), 24 (B), 48 (C), and 72-h (D) 40 ng/mL IL-4 activation. Each bar represents 3 biological replicates (n = 3). # represents undetectable genes. Statistical analysis was conducted with multiple unpaired t tests.

(E) THP1 WT and THP1 STAT6-RE cells were differentiated into macrophages. Following a 48-h period rest, cells were then incubated in 40 ng/mL IL-4 for 48 h then lysed for cell protein analysis. Control represents cells that received no IL-4 stimulation (Data S1).

(F) Quantitative measurement of p-STAT6 protein level from STAT6-RE THP-1 treated with 1, 20, and 40 ng/mL IL-4 for 48 h and the non-stimulated control.

(G) Reporter gene rFluc mRNA was measured with qPCR at 0, 12, 24, 48, and 72-h 40 ng/mL IL-4 incubation. rFluc expressions in MΦ and M2 phenotypes of STAT6-RE THP-1 macrophages were normalized to the corresponding rFluc expressions in WT THP-1 MΦ macrophages. Statistical analysis on rFluc gene expression was performed with one-way ANOVA followed by Tukey’s post hoc test.

(H) p-STAT6 protein from samples treated with 40 ng/mL IL-4 for 0, 12, 24, 48, or 72 h was quantitatively measured with ELISA. Each data bar represents three biological replicates (n = 3). Statistical analysis was performed with one-way ANOVA followed by Tukey’s post hoc test. Statistical analysis was performed with multiple unpaired t tests. ns, non-significant, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

To further assess the activation pattern in the STAT6 signaling pathway, western blotting and ELISA were used to quantify the active form of STAT6 transcription factor p-STAT6. p-STAT6 was barely detected in naive macrophages for WT and STAT6-RE THP-1 cell lines, while p-STAT6 was significantly upregulated in both THP-1 cell lines after 40 ng/mL IL-4 activation (Figure 2E; Data S1). ELISA was performed to quantify the p-STAT6 expression levels in WT and STAT6-RE THP-1 cell lines treated with IL-4 at different concentrations (1, 20, and 40 ng/mL). p-STAT6 expression patterns at different concentrations were similar in WT and STAT6-RE THP-1 cell lines (Figure 2F). Collectively, these data indicated that polarization functions and the STAT6 signaling pathway remained unimpaired in STAT6-RE THP-1 macrophages after lentiviral transduction.

To assess the correlation between the BTS curve and rFluc gene expression, rFluc mRNA expression was quantified at different time points (12, 24, 48, and 72 h) in unactivated control and 40 ng/mL IL-4-activated STAT6-RE macrophages (Figure 2G). rFluc gene displayed minimal expression in unactivated control, while rFluc in activated macrophages constantly increased at 24 h and after. Moreover, the BTS trajectory matched the rFluc expression trends, suggesting that the BTS curve accurately replicates STAT6-activated rFluc expression.

To assess the correlation between the BTS curve and p-STAT6 activity, a time course for p-STAT6 expression during single-dose 40 ng/mL IL-4 activation was measured. Compared with the unactivated control, p-STAT6 levels were significantly upregulated in 12- and 24-h activated groups. p-STAT6 expression peaked at 24 h, while the BTS-generated curve peaked at 72 h (Figure 2H). The 48–60 h difference in peak activity likely reflects the time for transcription and translation of rFluc, which is under the regulation of p-STAT6.

STAT6-RE rFluc THP-1 reporter monitors the M2 polarization in real time

To assess M2a polarization in STAT6-RE THP-1 macrophages, we generated BTS curves following protocols detailed in the STAR Methods (Figure 3A). We monitored STAT6 signaling through relative luminescence units (RLUs) over a 12-day period during which the macrophages were exposed to IL-4 (Figure 3B). While 1 ng/mL IL-4 produced minimal activation, higher concentrations showed substantial responses: 20 and 100 ng/mL IL-4 resulted in 5.8- and 10-fold increases in peak activity, respectively.

Figure 3. BTS measurement on the STAT6-RE THP-1 cell line shows specificity for M2a activation.

(A–D) Workflow of performing BTS on STAT6-RE THP-1 macrophages with the multimode microplate reader (A). Relative luminescence units (RLUs) were measured at a 12-h interval for up to 4 days and at a minimum 12-h interval for up to 12 days. BTS on STAT6-RE THP-1 macrophages activated with anti-inflammatory cytokines IL-4 (B), IL-6 (D), and IL-10 (E), at 20, 100 ng/mL, and the unactivated control. (C) Comparison of BTS on STAT6-RE-THP-1 macrophages activated by IL-4, IL-6, and IL-10 at 100 ng/mL. Each data point represents 3 biological replicates (n = 3). Statistical analysis was performed using the repeated measures ANOVA followed by Dunnett’s post hoc test. ns, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

(E and F) BTS on STAT6-RE THP-1 macrophages given anti-inflammatory stimuli IL-10 at 20 and 100 ng/mL (E) and pro-inflammatory stimuli IFN-α at 20 and 100 ng/mL (F) and the unactivated control. Each data point represents 3 biological replicates (n = 3). Statistical analysis was performed using the repeated measures ANOVA followed by Dunnett’s post hoc test. ns, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

To confirm the specificity of STAT6-RE cells to IL-4/STAT6 activity, we tested their response to other STAT-pathway-related inflammatory cytokines: IL-6/STAT3,³³ IL-10/STAT3,³⁴ and interferon α (IFN-α)/STAT1^35,36 (Figure 3C). Only IL-4 produced significant fold increases in activity across all tested concentrations. IL-6 (Figure 3D) and IL-10 (Figure 3E) induced only slightly higher STAT6 activity compared with the unactivated control. Western blot analysis confirmed that IL-6 exposure triggered STAT3 phosphorylation but not STAT6 (Figures S1 and S2; Data S1), while IL-10 showed minimal p-STAT3 activity and no p-STAT6 activation.

Pro-inflammatory stimuli produced contrasting effects. Exposure to IFN-α (Figure 3F) resulted in suppressed STAT6 signaling compared with the unactivated control. Similar inhibitory responses were observed with other pro-inflammatory factors, including IFN-γ, lipopolysaccharide (LPS), and tumor necrosis factor α (TNF-α) (Figure S3), all known STAT6 activity suppressors.^37,38 These results collectively demonstrate that the BTS curve from STAT6-RE cells specifically reports STAT6/M2a polarization and can effectively distinguish different response magnitudes.

Inhibition of STAT6 signaling pathways captured by STAT6-RE THP-1 reporter cells

Having established the BTS curve’s effectiveness for monitoring M2 polarization, we evaluated how STAT6-RE THP-1 cells responded to inhibitors of the STAT6 signaling pathway. We tested inhibitors with different mechanisms of action: leptomycin B (LMB), tofacitinib, and AS1517499 (Figures 4A and S4). LMB, a nuclear export inhibitor, blocks protein nuclear export^39,40 and disrupts the translocation of both STAT6 and luciferase to the cytoplasm. When we treated IL-4-activated macrophages with LMB (1 or 10 ng/mL), we observed significantly reduced BTS readings, indicating that STAT6 signaling activation requires an uninterrupted STAT6 cycle (Figure 4B). Tofacitinib inhibits JAK1/JAK3 phosphorylation^41,42 and consequently reduces p-STAT6 formation (Figure 4A). Treatment with tofacitinib (1 or 5 μM) during IL-4 activation significantly decreased relative luminescence units (RLUs), demonstrating that intact JAK function is necessary for STAT6 signaling activation (Figure 4C). We also tested AS1517499, a specific STAT6 phosphorylation inhibitor.^43–45 24 h after incubation, there was a 35.7% decrease in STAT6-RE activation between the DMSO control and the 500 nM group (Figure S4).

Figure 4. Tofacitinib and LMB inhibited the STAT6 signaling pathways.

(A) Diagram of tofacitinib inhibiting STAT6 phosphorylation by interfering with JAK1/JAK3 activities and LMB blocking STAT6 nuclear export by interfering exportin 1 function.

(B) BTS on STAT6-RE THP1 macrophages treated with 1, 10 ng/mL LMB or methanol after the 24-h 40 ng/mL IL-4 activation. Each data point represents 3 biological replicates (n = 3). RM-ANOVA with Dunnett’s post hoc test was performed on BTS curves. *p < 0.05, ***p < 0.001.

(C) BTS on STAT6-RE THP1 macrophages treated with 1 μM, 5 μM tofacitinib, or DMSO after the 24-h 40 ng/mL IL-4 activation. Each data point represents 3 biological replicates (n = 3). RM-ANOVA Dunnett’s post hoc test was performed on the BTS curves. *p < 0.05, ***p < 0.001.

(D) Immunostaining on 24-h IL-4-activated STAT6-RE THP1 macrophages treated with 10 ng/mL LMB, methanol, 5 μM tofacitinib, and DMSO for 48 h, respectively. The cell nucleus was stained by Hoechst 33342 (blue), filamentous actin (F-actin) was stained by Alexa Fluor 488 phalloidin (green), and p-STAT6 was stained with rabbit-anti-p-STAT6 (Y641) primary antibody labeled by Alexa Fluor 648-conjugated secondary antibody (red). The scale bar is 20 μm.

(E) Mean fluorescence intensity (MFI) of p-STAT6 was quantified within the nucleus, cytoplasm, and whole cell with Fiji ImageJ based on immunostaining images. Each data point represents 8 biological replicates (n = 8). The unpaired t test was used for statistical analysis.ns, p > 0.05, ***p < 0.001.

(F) Quantitative measurement on p-STAT6 protein level in STAT6-RE THP1 macrophages treated with 10 ng/mL LMB, methanol, 5 μM tofacitinib, and DMSO for 48 h, respectively, by using ELISA. Each data point represents 3 biological replicates (n = 3). The unpaired t test was used for statistical analysis. ns, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

To further investigate these effects, we examined p-STAT6 localization and distribution in STAT6-RE macrophages treated with either inhibitor, comparing them to appropriate vehicle controls (methanol or DMSO) through immunostaining (Figure 4D). LMB-treated macrophages showed decreased mean fluorescence intensity (MFI) compared with methanol controls, regardless of IL-4 treatment timing (Figures 4E and S5). The reduction in p-STAT6 was more pronounced in the cytoplasm than in the nucleus, consistent with impaired STAT6 transport. Similarly, tofacitinib-treated macrophages exhibited reduced fluorescent intensity compared with DMSO controls, an effect that was independent of IL-4 treatment timing (Figures 4E and S5).

Contrary to LMB treatment, p-STAT6 was significantly decreased in the cell nucleus but not in the cytoplasm, mainly contributing to total p-STAT6 downregulation (Figures 4E and S6). This heterogeneous p-STAT6 distribution may result from less STAT6 phosphorylation and nucleus translocation due to JAK1/JAK3 kinase knockdown. ELISA was performed to confirm that p-STAT6 levels were significantly decreased in LMB- or tofacitinib-treated groups compared with the corresponding controls (Figure 4F). In conclusion, those data are consistent with each other, suggesting the feasibility of utilizing BTS and STAT6-RE THP-1 cells to assess the efficacy of STAT6 signaling pathway inhibitors.

The reporter cell is used for measuring TAM M2 polarization in A549 lung cancer models, and the interruption of STAT6 signaling attenuates the M2 polarization

Research has shown that IL-4 abundantly exists in various cancers^46–48 and that the STAT6 signaling pathway plays a critical role in the development of tumor-associated macrophages (TAMs).^49,50 To investigate this relationship further, we examined how A549 cancer cells, a human non-small cell lung cancer cell line, influence STAT6 signaling in our reporter cells. STAT6-RE THP-1 monocytes were treated with three independent conditions: A549 conditioned media, 2D co-culture with A549 cells, and 3D spheroid culture with A549 cells (Figure 5A).

Figure 5. M2 polarization of STAT6-RE THP-1 monocytes in A549 lung cancer models.

(A) Diagram of BTS measurements on STAT6-RE THP-1 monocytes with different experimental settings: A549-CM, 2D co-culture, and 3D spheroid.

(B and C) BTS on STAT6-RE THP-1 monocyte control (gray; no cytokines added), STAT6-RE THP-1 monocytes cultured in A549-CM, 2D co-culture, or 3D spheroid for 10 days. Each point represents 3 biological replicates (n = 3). Paired t test was performed on A549 CM and THP-1 monocyte control. RM-ANOVA with Dunnett’s post hoc test was performed on 2D co-culture and 3D spheroid and THP-1 monocyte control. ns, p > 0.05, ****p < 0.001.

(D and E) BTS on A549-STAT6-RE THP-1 monocyte 3D spheroid with either methanol or 10 ng/mL LMB treatment. BTS on A549-STAT6-RE THP-1 3D spheroid with either DMSO or 10 μM IL-4 inhibitor.*p < 0.05.

(F and G) BTS on A549-STAT6-RE THP-1 3D spheroid with 50 ng/mL pro-inflammatory cytokines IFN-γ or IFN-α. Each point represents 3 biological replicates (n = 3). A paired t test was performed on BTS curves. *p < 0.05.

The results revealed interesting patterns in STAT6 activation. While A549 conditioned media failed to activate the STAT6 signaling pathway, the 2D co-culture showed slightly elevated STAT6 activity compared with control conditions (Figure 5B). The 3D spheroid co-culture demonstrated the most robust STAT6 activation, showing a 4-fold increase compared with the 2D model (Figure 5C). In addition to a luciferase reporter, the STAT-RE cells also have a GFP reporter (Figure 1). Fluorescence imaging of the GFP confirmed the increased STAT6-activated macrophages in the spheroid (Figure S7). These findings suggest that direct cell-to-cell interaction, particularly in a 3D environment, may be crucial for STAT6 signaling activation in monocytes.

Then we evaluated various inhibitors to better understand the STAT6 signaling with 2D and 3D cocultures. The nuclear export inhibitor LMB inhibitor (Figure 5D) and IL-4 inhibitor (Figure 5E) both effectively reversed STAT6 signaling activation in both 2D and 3D co-culture systems. Similarly, IFN-α (Figure 5F) and IFN-γ (Figure 5G), known for their ability to repolarize TAM macrophages into an M1 phenotype, successfully suppressed the STAT6 signaling pathway. Even high doses of tofacitinib (50 μM) (Figure S7) showed minimal effect on inhibiting STAT6 signaling in either culture system. In summary, this result supported the use of the STAT6-RE THP-1 cell line to monitor the TAM development in the lung cancer scenario and assess the efficacy of strategies targeting TAMs.

The STAT6-RE reporter system can be applied to primary macrophages

We first differentiated human peripheral blood monocytic cells (hPBMCs) into human monocyte-derived macrophages (hMDMs) to establish primary macrophage luciferase reporters. The transduction process was performed as provided in the STAR Methods (Figure 6A). STAT6-RE hMDMs were administered with IL-4 at 1, 20, and 100 ng/mL for BTS measurements (Figure 6B). 1 ng/mL dose resulted in a significant activation in the STAT6 signaling pathway relative to unactivated control, and an increased dose (20 ng/mL) led to a more robust activation. However, further increasing the dose to 100 ng/mL decreased activation in the STAT6 signaling pathway (Figure 6B). The signal intensity was an order of magnitude lower than in the THP-1 cells, likely due to differences in transduction efficiency and experimental conditions (i.e., cell density). Nonetheless, these results suggested that the STAT6-RE reporter system enables real-time M2 polarization measurements in primary macrophages.

Figure 6. STAT6-RE Fluc reporter system reports M2 polarization in primary human macrophages.

(A) The workflow of introducing STAT6-RE reporter genes in hPBMC-derived hMDMs with lentiviral transduction.

(B) BTS on the STAT6-RE Fluc hMDM reporter with IL-4 administration at 1, 20, and 100 ng/mL and the unactivated control. Plate reading was performed at a 24-h interval for up to 10 days. Each data point represents 3 biological replicates (n = 3). Statistical analysis was performed with RM-ANOVA followed by Dunnett’s post hoc test. ***p < 0.001; ****p < 0.0001.

The STAT6-RE reporter system provides an emergent read-out that supports inference of STAT signaling dynamics in primary macrophages

One promise of the STAT6-RE reporter system is the potential to use bioluminescence as a read-out that is indicative of intracellular signaling phenomena. Reporter activity lies downstream of JAK phosphorylation due to IL-4 engagement with the JAK receptor. However, the transduction of JAK activation to STAT6-RE activity involves several intermediate events, and RE activation induces feedback mechanisms that alter JAK availability at the membrane. Thus, BTS curves are an emergent outcome dependent on the non-linear coupling of membrane, cytoplasmic, and nuclear events. To assess the informative value of BTS signal dynamics, we developed a preliminary model of JAK/STAT6 signaling (Figure 7A) and estimated the plausible kinetics of intracellular events. Despite the practical unidentifiability of the large number of model parameters, the BTS readouts over four different levels of IL-4 stimulation were sufficient to constrain a search for credible parameters that reproduced the observed BTS signal (Figure 7B). Furthermore, the considerable uncertainty in estimates for most parameters mapped to a comparatively narrow range of emergent dynamics (Figure 7C, top).

Figure 7. Model of the STAT6 signaling pathway with kinetics inferred from BTS luminescence.

(A) Schematic indicating molecular events considered in the ordinary differential equation (ODE) model. Circles indicate molecular species. Squares indicate events. Arrows indicate the logical requirements for an event to occur and the consequences of that event (input and output arrows, respectively). Event connections with a round head indicate a logical requirement for a species to be present without the model accounting for changes in the abundance of that species. In these cases, the required species is assumed to be available in excess.

(B) Predicted bioluminescence time courses for best-fit simulations identified from N = 150,000 examined parameter sets over the course of 150 sequential Monte Carlo (SMC) populations. Each corresponds to a single-dose IL-4-activated scenario. Symbols (circle, triangle, cross, and asterisk) represent experimental BTS data. Lines (solid and dashed) represent simulated curves.

(C) Simulated BTS dynamics (top) and the predicted correlation between emergent luminescence and STAT6-RE activation at the 40 ng/mL IL-4 condition (bottom). Each trace represents the simulation outcome for one of N = 1,000 inferred parameter sets.

(D) Range of predicted BTS and total p-STAT6 time courses. Plots indicate median and 95% CIs (central line and shaded region, respectively) from the ensemble of N = 1,000 simulations. Results are shown for simulated stimulation with 40 ng/mL IL-4.

(E) Distributions reflecting timing of BTS and p-STAT6 maxima.

(F) Hysteresis of predicted BTS dynamics with respect to total p-STAT6 dynamics.

In all cases, the predicted time course of STAT6-RE activity exhibited a linear correlation with the BTS signal (Figure 7C, bottom). In addition, the ensemble of simulations recapitulated known qualitative features of the signaling pathway. This included the order-of-magnitude difference between the luciferase half-life and the decay of luminescence (~1 h vs. ~1 min, respectively). The model also captured the timing of the p-STAT6 maximum after IL-4 stimulation—despite not being trained on direct p-STAT6 measurements. Specifically, the model identified a time-to-peak of 0.87 days (95% confidence interval [CI]: [0.07, 1.7]) for total p-STAT6, which precedes the maximum of the BTS curve at 2.6 days (95% CI: [2.5, 2.8]) a p-STAT6 with the p-STAT6 protein dynamics measured by ELISA (Figure 7D). Notably, p-STAT6 is a state variable whose simulated time course did not correlate with the dynamics of the BTS curves. This variable instead exhibited considerable hysteresis when plotted against STAT6-RE activation and corresponding luminescence (Figure 7F).

DISCUSSION

Macrophages display extensive plasticity in response to microenvironmental stimuli, resulting in heterogeneous polarization dynamics. This study established a stable STAT6-RE THP-1 cell line to characterize the temporal profiles during IL-4-induced M2a polarization. After lentiviral transduction, we first confirmed the intactness of polarization capacities and molecular machinery in the STAT6-RE THP-1 cell line. Combined with the models generated from the BTS curves, the STAT6-RE THP-1 cell line reports M2a polarization kinetics in different biological contexts.

In the single-dose cytokine activation model, the STAT6-RE THP-1 macrophages revealed dose-dependent activation upon IL-4 exposure but inhibition for M1 stimuli. By analyzing the parameters (peak value, peak time point, and activation period) extracted from the BTS curves, we can interpret the macrophage responses to the environmental stimuli more quantitatively. In our study, the p-STAT6 activation via western blot and M2a genes (MRC1, SOCS1, and CD86) via gene expression displayed similar activation patterns as the BTS curve in M2a polarization. We also demonstrated the applicability of the STAT6-RE reporter system on primary macrophages, which indicated the potential for studying macrophage behaviors using in vivo models. Previous studies have shown that M2 genes (mrc1, cd163, ccl17, arg1, and tgfb1) and an M2-polarizing cytokine (IL-4) exhibited the activation curve in the wound healing process⁵¹ and spinal cord ischemia/reperfusion injury (SCIRI),⁵² respectively. Each factor can be measured individually; however, using the STAT6-RE system simplifies the complexity and presents a more easily quantifiable model. Therefore, the BTS curve generated from STAT6-RE reporter cells accurately represents M2a polarization. In addition, our system can be used to validate gene expression results.

As all the cell populations are stably transfected, the STAT6-RE THP-1 cell line has high sensitivity and reproducibility and can assess chemical compounds targeting STAT6-dependent M2a polarization. Our results are consistent with previous studies that STAT6 signaling pathway inhibitors LBM,¹⁰ tofacitinib,⁵³ and AS1517499⁵⁰ significantly knocked down the M2 activation (Figures 3B, 3C, and S6–S8). However, leflunomide treatment failed to inhibit STAT6 signaling pathway activation in our studies, contrary to previous studies (data not shown).⁵⁴

We investigated how the STAT6-RE THP-1 cell line could be used to monitor complex tumor environments such as a tumor. Extensive studies have shown the association between M2-like TAM and cancer progression.⁵⁵ STAT6 was shown to express abundantly in human lung carcinoma.⁴⁷ Studies have shown that inhibiting STAT6 expression suppresses lung cancer progression both in vivo and in vitro.^30,39 Our findings aligned with this research, demonstrating that STAT6-dependent M2a polarization continues steadily in macrophages within lung cancer environments. Using chemical compounds (LMB or IL-4 inhibitor) or IFN cytokines to suppress the STAT6 signaling pathway reduced M2a polarization in lung cancer models. This method allowed us to repeatedly measure polarization from the same samples, enabling us to track TAM development over time while reducing sample-to-sample variability.

The ability to continuously measure macrophage responses provides a tool for assessing organ functions. There are existing models that use RE reporters. Reporter systems based on inflammatory signaling pathways, such as protein kinase R-like endoplasmic reticulum kinase (PERK),⁵⁶ NF-κB,^57,58 arginase-1 (ARG1),⁵⁹ and inducible nitric oxide sythase (iNOS),⁶⁰ have previously been used to study neuroinflammation, tumor immunosuppression, and assess anti-inflammatory compounds. However, a key advantage of our approach is its ability to capture temporal dynamics rather than single time points. The BTS curve generated by time course kinetics from the STAT6-RE reveals comprehensive information about macrophage polarization dynamics. We leveraged this temporal BTS signal data to inform a mathematical model of JAK/STAT6 signaling. Our model suggests a linear correlation between BTS signal dynamics and STAT6-RE activation and successfully predicts the timing of the p-STAT6 peak—even though this state variable does not directly correlate with STAT6-RE luminescence. These results support the BTS curves as an informative tool for real-time measurement of STAT6 activation in macrophages, offering a more complete picture of the signaling dynamics than traditional single-time point approaches.

In conclusion, the STAT6-RE THP-1 cell line provides a platform for characterizing the kinetics of M2a polarization under various biological contexts. BTS curves are robust and rapid to measure the global polarization kinetics in M2a polarization. Further, we introduce STAT6-RE primary macrophages, confirming that this model can be used in primary as well as immortalized cells. In future studies, we hope to use these reporter cells within in vivo model contexts.

RESOURCE AVAILABILITY

Lead contact

Requests for further information and resources and reagents should be directed to and will be fulfilled by the lead contact, Elizabeth Wayne (lizwayne@uw.edu).

Materials availability

Plasmids generated in this study are available upon request to the lead contact.

STAR★METHODS

EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS

Human monocyte wildtype (WT) THP-1 cells (ATCC TIB-202) and STAT6-RE THP-1 cells were used in all the experiments and cultured in RPMI high glucose medium (ATCC) supplemented 10% fetal bovine serum (Avantor) at 5% CO2 in 37°C. Non-small cell lung cancer (NSCLC) A549 cell line (ATCC CCL-185) was cultured in F-12K medium (ATCC) supplemented with 10% fetal bovine serum (Avantor). HEK293T (ATCC CRL-3216) was cultured in DMEM high glucose medium (ATCC) supplemented with 10% fetal bovine serum (Avantor) at 5% CO2 at 37°C.

Human THP-1 monocytes were differentiated into macrophages using phorbol 12-myristate-13-acetate (PMA). Cells were seeded at the density of 2x10⁵ cells/ml in a white transparent bottom 96-well plate with 10ng/ml PMA. After 24-hour incubation, the media was changed following a 48-hour resting period.

A549 cells and HEK293T cells were trypsinized with trypsin-EDTA (0.25%) (ThermoFisher) when cells reached 80-90% confluency and passed to new cell culture flasks. Periodically, cells were screened for mycoplasma contamination with MycoAlert^™ PLUS Mycoplasma Detection Kit (Lonza).

Human peripheral blood monocytes (STEMCELL 70034) were seeded at 1x10⁶ cells/ml in a 6-well tissue-treated plate (Avantor). They were differentiated in ImmunoCult^™-SF macrophage media (STEMCELL) supplemented with 50 ng/ml M-CSF for 7 days. The medium was changed every 3 days. Human monocyte-derived macrophages (hMDMs) were used for lentiviral transduction.

METHOD DETAILS

Vector construct and molecular cloning

To amplify the STAT6-RE sequence from the donor plasmid (p4xSTAT6-Luc2P, addgene), PCR cloning was performed with forward primer (5’-CATGTACTCGAGTGCAGGTGCCAGAACATT-3’) and reverse primer (5’-GATGATGCTAGCCCAGATCCCTGCTGTTCTT-3’). Agarose gel electrophoresis was performed to determine the molecular size of the amplified sequence. The amplified STAT6-RE sequence was cloned into pGF2-mCMV-rFluc-T2A-GFP-mPGK-Puro (System Biosciences) using XhoI and NheI restriction enzymes. To confirm the successful insertion of STAT6-RE into the recipient lentivector, PCR was performed with forward primer (5’-ACATTTCTCTGGCCTAACTGG-3’) and reverse primer (5’-GATCCAACGAATGTCGAGAGG-3’). Agarose gel electrophoresis was performed to determine the molecular size of the amplified sequence.

Lentivirus particle production and cell transduction

HEK293T cells were seeded in a 6-well plate at 0.5x10⁶ cells/ml one day before the transfection. 2^nd generation lentiviral packaging plasmid pCMV-p8.9, envelope plasmid pMDG-VSVG and the shuttle plasmid pLV-STAT6-RE-Ffluc-GFP were transfected into HEK293T cells via TransIT-293 transfection reagent (Mirus Bio). Lentivirus was harvested after 72 hours, and virus titer was measured with a virus titration kit (Applied Biological Materials). Lentivirus was concentrated with Lenti-X concentrator (Takara Bio) and then drop-wise onto WT THP-1 cells. To achieve better transduction efficiency, spinoculation was performed at 200xg for 1 hour. After 3-day viral incubation, media was changed, and then 1μg/ml puromycin (ThermoFisher) was added into RPMI complete media for selection. Provirus copy in transduced THP-1 cell line was quantified with Lenti-X provirus quantification kit (Takara Bio). hMDMs were incubated with lentivirus a multiplicity of infection (MOI = 30). Transduction enhancer lentiBOOST-P and polybrene were used at 1 mg/ml and 1 μg/ml, respectively. Spinoculation was performed at 200xg for 1 hour to enhance the lentiviral transduction. The media was changed after overnight incubation. A second dose of lentivirus was given to hMDMs two days after the first dose with the same transduction procedures. After a 10-day resting period, transduced STAT6-RE hMDMs were used for bioluminescent measurements.

Bioluminescence temporal spectroscopy (BTS)

After differentiation into macrophages, STAT6-RE cells were incubated with cytokines and 1mM D-luciferin-supplemented RPMI complete media to detect luciferase activity. The relative luminescence units (RLU) signal from THP-1 STAT6-RE macrophages were measured in the multimode microplate reader at 12-hour interval time points. RLU represents the numerical value generated by the luminometer (Tecan Spark multimode microplate reader) and is proportional to the luminescence photons cells emit. Exposure time: 10000 ms.30 μL of media was replaced with fresh 1mM D-luciferin-supplemented RPMI complete media every 24 hours.

Immunocytochemistry (ICC)

Cells were plated in 4 well glass chamber slides (Lab-Tek) with 10ng/ml PMA. After 24-hour incubation, the media was changed following a 48-hour resting period. Cells were treated with 40 ng/ml IL-4 for 24 hours. After that, cells were added with 70% Methanol,10ng/ml Leptomycin, DMSO, and 5μM Tofacitinib, respectively, for an additional 48 hours. Cells were fixed with 4% paraformaldehyde (PMA), then permeabilized with 0.1% Tween-20 for 30 minutes. phospho-STAT6 (p-STAT6) were stained with rabbit anti-p-STAT6 (Cell Signaling Technology) at 1:1000 dilution in 0.1% BSA overnight. Goat anti-rabbit antibodies conjugated to Alexa fluor 647 were incubated with p-STAT6 antibody-labeled samples at 1: 200 dilutions in 0.1% BSA for an hour. Lastly, cells were stained with DAPI (Biolegend) at 300nM and Alexa Fluor 488 phalloidin (Invitrogen) at 1:400 For 30 minutes. Cells were washed with 1xPBS three times in between every step.

Microscopy and image analysis

Fixed cell samples were imaged on a confocal microscope (LSM 880; Zeiss) at 63x with oil immersion. Imaging settings, including power, pinhole, gain, and other parameters, remained identical throughout the imaging process. All imaging analyses were performed on Fiji ImageJ. To identify boundaries of cytoplasm and nucleus, cytoskeleton-Alexa fluor-488 channel and DAPI-nucleus channel were used to extract corresponding regions of interest (ROIs) from identical cells. After applying ROIs drawn from 488 and DAPI channels onto p-stat6-Alexa fluor 647 channels, fluorescence intensity was quantified within the cytoplasm, the nucleus, and the whole cell, respectively.

Enzyme-linked immunosorbent assay (ELISA)

P-STAT6 was measured with Phospho-STAT6 (Tyr641) Sandwich ELISA Kit (Cell Signaling Technology) according to manufacturing protocol. Cell lysates from WT and STAT6-RE cells were collected after experimental treatments, followed by a BCA assay to determine protein concentration. Cell lysates were incubated in Phospho-STAT6 antibody-coated microwells for 2 hours at 37°C. After removing the samples, the detection antibody and HRP-linked secondary antibody were incubated in the microwells for 1 hour and 30 minutes, respectively. By adding the TMB substrate, the absorbance was measured with a spectrophotometer at 450 nm. Three washing steps were performed between each step.

Western blotting

WT and STAT6-RE THP-1 cells were plated in T25 flasks and then were differentiated into macrophages with 24-hour 10ng/ml PMA incubation. After a 48-hour resting period, cells either were cultured in RPMI complete media or were treated with 40 ng/ml IL-4 for 48 hours. Cell lysates were extracted from WT and STAT6-RE-fluc-GFP THP-1 macrophages with 0.1% sodium dodecyl sulfate (SDS) (Supelco). Protein concentration for each sample was quantified with a BCA protein assay kit (ThermoFisher). Protein samples were normalized to equal concentrations and then loaded onto a 8% polyacrylamide gel and transferred to PVDF membrane (Bio-Rad 1620177) for immunodetection. Membranes were blocked for two hours in tris-buffered saline (20 mM Tris, 150 mM NaCl, pH 7.6) containing 0.1% Tween 20 and 5% nonfat powdered milk. Primary antibodies for pSTAT6/pSTAT3 and GAPDH (Fitzgerald 10R-2932, 1:10,000, mouse) were incubated overnight at 4°C under gentle agitation. Rabbit or mouse primary antibodies were detected using a horseradish peroxidase-conjugated anti-rabbit IgG (Sigma AP307P, 1:4000) or anti-mouse IgG (Sigma AP308P, 1:4000) secondary antibody for 60 min at room temperature, then developed using SuperSignal West Pico PLUS (Thermo Fisher) chemiluminescence substrate. Then the blots were imaged in the ChemiDoc imaging system (BioRad). After imaging, blots were rinsed with 1xTBST, then incubated with 10mL of Restore Western Stripping Buffer in a rocker for 10 minutes. Then blots were washed with 1xTBST (5 minutes and 3 washes). After stripping, blots were incubated with primary antibodies against STAT6 or STAT3 and detected as described above. Uncropped Western Blots are included in (Data S1).

RT-qPCR

mRNA was extracted from both WT and STAT6-RE THP-1 cells with RNeasy Maxi Kit (Qiagen). The mRNA concentration from each sample was measured with Nanodrop spectrophotometers and normalized to the same concentration for reverse transcription. cDNA was generated by using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer’s instructions. 1000 ng cDNA were amplified with 500nM forward and reverse primers for each gene of interest and PowerTrack SYBR Green Master Mix (Applied Biosystems) for each sample. The experiments were performed on an Applied Biosystems ViiA 7 system. The primer sequences for SYBR green amplification were listed in the supplementary form. Relative mRNA expressions were normalized with the mean of the housekeeping gene ribosomal RNA (18S). For the comparison of polarization associated gene expression in M2 phenotypes between WT and STAT6-RE THP-1 cell lines, relative gene expression levels were normalized to the corresponding naïve control using the 2^-ddCt method.

Statistical analysis

Statistical analyses are described in detail in each figure legend. Unless otherwise specified, analysis was performed using one-way-ANOVA followed by Tukey’s post-hoc testing. Data points represent a minimum of 3 biological replicates. P>0.05, non-significant (ns); P<0.05, *; P<0.01, **; P<0.001, ***. P<0.0001, ****. All the statistical analyses were performed on Prism Graphpad software.

Systems modeling and parameter inference

Based on current knowledge about JAK/STAT signaling involving JAK1/3 and STAT6 – as stimulated by IL-4 – we implemented a mathematical model of STAT6-RE activation and expected BTS luminescence. The model abstraction reflects: (1) IL-4-driven augmentation of basal JAK phosphorylation; (2) resulting STAT6 phosphorylation and consequent dimerization; (3) disassociation of the dimer, competing with dimer translocation to the nucleus; (4) activation of the response element by STAT6 dimers; (5) consequent luciferase synthesis and generation of luminescence; and (6) synthesis of SOCS and its mediation of JAK degradation, acting as a negative feedback mechanism. The resulting model, implemented as a system of twelve ordinary differential equations (ODEs), consisted of twenty-six rate expressions. All expressions assumed mass-action kinetics. After non-dimensionalizing, the system of ODEs included 23 rate constants. We captured the impact of IL-4 stimulation on JAK phosphorylation using a saturable effect model that contributed two additional parameters: a maximal effect constant and a saturation constant. We then used the BTS data recorded for IL-4 stimulation at 20, 40, 80, and 100 ng/ml to perform parameter estimation. In each case, we normalized the IL-4-stimulated data against the control and likewise quantified the ratio between predicted BTS signal for IL-4stimulated simulations and a control simulation. The simulation also accounts for periodic dilution of the IL-4 concentration, corresponding to daily replacement of 1/15 of the cell media volume.

With such complexity, the model is underdetermined and the parameters not practically identifiable. We endeavored instead to assess the extent to which the emergent read-out of bioluminescence could constrain the model’s kinetic parameters. To this end, we performed parameter inference using the simulation-based inference approach of approximate Bayesian computation with sequential Monte Carlo (ABC-SMC).^62,63 To identify a distribution of plausible parameter values, we began with N=1000 randomly sampled parameter sets, distributed by uniform Sobol sampling in a 25-dimensional log-space, with values in the range [−4,4]. Our algorithm then iterated with a uniform perturbation kernel and an increasingly strict rejection constant until the summary statistic (least-squared cost function summed for all data series) across the population of N parameter sets fell below a selected threshold (i.e., a mean absolute difference per datapoint < average standard deviation of the BTS measurements).

Simulations and analyses were performed using MATLAB (R2023A; The MathWorks, Inc., Natick, Massachusetts, USA). All original code has been deposited at Zenodo and is publicly available at Zenodo: https://doi.org/10.5281/zenodo.14545524 as of the date of publication.

Supplementary Material

Supplemental information can be found online at https://doi.org/10.1016/j.cels.2026.101643.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Cell lines
NEB® 5-alpha Competent E. coli (High Efficiency)	NEB	antib
NEB® Stable Competent E. coli (High Efficiency)	NEB	C3040I
HEK293T	ATCC	Cat# CRL-11268; RRID: CVCL_1926
Wildtype (WT) THP-1	ATCC	Cat# TIB-202; RRID: CVCL_0006
Human Peripheral Blood Monocyte	STEMCELL Technologies	70034
A549	ATCC	CCL-185
Oligonucleotides
STAT6-RE-Clone-Forward Primer: 5’-CATGTACTCGAGTGCAGGTGCCAGAACATT-3’	This paper	N/A
STAT6-RE-Clone-Reverse Primer: 5’-GATGATGCTAGCCCAGATCCCTGCTGTTCTT-3’	This paper	N/A
STAT6-TEST-Forward Primer: 5’-ACATTTCTCTGGCCTAACTGG-3’	This paper	N/A
STAT6-TEST-Reverse Primer: 5’-GCGATCTGACGGTTCACTAA-3’	This paper	N/A
GATA3-Forward Primer: 5’-GCCCCTCATTAAGCCCAAG-3’	This paper	N/A
GATA3-Reverse Primer: 5’-TTGTGGTGGTCTGACAGTTCG-3’	This paper	N/A
CCL22-Forward Primer: 5’-ATCGCCTACAGACTGCACTC-3’	This paper	N/A
CCL22-Reverse Primer: 5’-GACGGTAACGGACGTAATCAC-3’	This paper	N/A
CCL11-Forward Primer: 5’-CCCCTTCAGCGACTAGAGAG-3’	This paper	N/A
CCL11-Reverse Primer: 5’-TCTTGGGGTCGGCACAGAT-3’	This paper	N/A
SOCS1-Forward Primer: 5’-CACGCACTTCCGCACATTC-3’	This paper	N/A
SOCS1-Reverse Primer: 5’-TAAGGGCGAAAAAGCAGTTCC-3’	This paper	N/A
MRC1-Forward Primer: 5’-TCCGGGTGCTGTTCTCCTA-3’	This paper	N/A
MRC1-Reverse Primer: 5’-CCAGTCTGTTTTTGATGGCACT-3’	This paper	N/A
CD86-Forward Primer: 5’-CTGCTCATCTATACACGGTTACC-3’	This paper	N/A
CD86-Reverse Primer: 5’-GGAAACGTCGTACAGTTCTGTG-3’	This paper	N/A
IL24-Forward Primer: 5’-CACACAGGCGGTTTCTGCTAT-3’	This paper	N/A
IL24-Reverse Primer: 5’-TCCAACTGTTTGAATGCTCTCC-3’	This paper	N/A
hARG1-Forward Primer: 5’-TCATCTGGGTGGATGCTCACAC-3’	This paper	N/A
hARG1-Reverse Primer: 5’-GAGAATCCTGGCACATCGGGAA-3’	This paper	N/A
IL4-Forward Primer: 5’-GCCAAGACCCCTTCGAGAAAT-3’	This paper	N/A
IL4-Reverse Primer: 5’-CCGATCCTGTTATCTGCCTCC-3’	This paper	N/A
hNOS2-Forward Primer: 5’-GCTCTACACCTCCAATGTGACC-3’	This paper	N/A
hNOS2-Reverse Primer: 5’-CTGCCGAGATTTGAGCCTCATG-3’	This paper	N/A
IFNγ-Forward Primer: 5’-TCGGTAACTGACTTGAATGTCCA-3’	This paper	N/A
IFNγ-Reverse Primer: 5’-TCGCTTCCCTGTTTTAGCTGC-3’	This paper	N/A
IL12B-Forward Primer: 5’-ACCCTGACCATCCAAGTCAAA-3’	This paper	N/A
IL12B-Reverse Primer: 5’-TTGGCCTCGCATCTTAGAAAG-3’	This paper	N/A
CD80-Forward Primer: 5’-AAACTCGCATCTACTGGCAAA-3’	This paper	N/A
CD80-Reverse Primer: 5’-GGTTCTTGTACTCGGGCCATA-3’	This paper	N/A
HLADRα-Forward Primer: 5’-AGTCCCTGTGCTAGGATTTTTCA-3’	This paper	N/A
HLADRα-Reverse Primer: 5’-ACATAAACTCGCCTGATTGGTC-3’	This paper	N/A
CD64-Forward Primer: 5’-TGGGTCAGCGTGTTCCAAG-3’	This paper	N/A
CD64-Reverse Primer: 5’-CACCTGTATTCACCACTGTCATT-3’	This paper	N/A
IL6-Forward Primer: 5’-ACTCACCTCTTCAGAACGAATTG-3’	This paper	N/A
IL6-Reverse Primer: 5’-CCATCTTTGGAAGGTTCAGGTTG-3’	This paper	N/A
TNFα-Forward Primer: 5’-CCTCTCTCTAATCAGCCCTCTG-3’	This paper	N/A
TNFα-Reverse Primer: 5’-GAGGACCTGGGAGTAGATGAG-3’	This paper	N/A
18S-Forward Primer: 5’-TGTGCCGCTAGAGGTGAAATT-3’	This paper	N/A
18S-Reverse Primer: 5’-TGGCAAATGCTTTCGCTTT-3’	This paper	N/A
rFluc-Forward Primer: 5’-TGACCGGCGTGGACTACAGCTA-3’	This paper	N/A
rFluc-Reverse Primer: 5’-TTCCTCGCAGTTCTCGCTGCAC-3’	This paper	N/A
Recombinant DNA
p4xSTAT6-Luc2P	Addgene	RRID: Addgene_35554
pGL4.45[luc2P/ISRE/Hygro]	Promega	E4141
pGreenFire mCMV 2.0	System Biosciences	TR410PA
Packaging Plasmids pCMV-R8.9	Gift from Dr. Alok Joglekar	Joglekar et al.61
Envelope Plasmid pMDG-VSVG	Gift from Dr. Alok Joglekar	Joglekar et al.61
Chemicals, peptides, and recombinant proteins
Xhol Restriction Enzyme	NEB	R0146S
Nhel-HF Restriction Enzyme	NEB	R3131S
DpnI Restriction Enzyme	NEB	R0176S
UltraPure™ DNase/RNase-Free Distilled Water	Thermo Fisher Scientific	10977015
5x Phusion™ HF Buffer	NEB	B0518S
Phusion™ High-Fidelity DNA Polymerase (2U/ul)	NEB	F530L
Deoxynucleotide (dNTP) Solution Set	NEB	N0446S
Qiagen QIAquick PCR Purification Kit	Qiagen	28104
GeneRuler 1 kb DNA Ladder, Ready-to-Use	Thermo Fisher Scientific	SM0313
SYBR™ Safe DNA Gel Stain	Thermo Fisher Scientific	S33103
UltraPure™ Ethidium Bromide, 10 mg/ml	Thermo Fisher Scientific	15585011
TopVision Agarose Tablets	Thermo Fisher Scientific	R2801
TBE Buffer (10X)	Thermo Fisher Scientific	B52
10x rCutSmart™ Buffer	NEB	B6004S
Antarctic Phosphatase Reaction Buffer (10x)	NEB	B0289S
Antarctic Phosphatase (5U/ 1pmol DNA ends)	NEB	M0289S
T4 DNA Ligase Buffer (10X)	NEB	B0202S
T4 DNA Ligase	NEB	M0202S
Tryptone	RPI	T60060-5000.0
Sodium Chloride for Biotechnology	VWR	97061-278
Sodium Hydroxide (NaOH) Pellets	VWR	M137-1KG
Selected Yeast Extract	Sigma-Aldrich	Y0375-500G
Methanol, 99%	Thermo Fisher Scientific	L13255.AP
Sodium dodecyl sulfate	Sigma-Aldrich	74255-250G
Tween-20	Sigma-Aldrich	P7949-100ML
Triton™ X-100	Millipore Sigma	T8787
Bovine Serum Albumin (BSA) DNase- and Protease-free Powder	Fisher Scientific	BP9706100
Tofacitinib citrate	Sigma-Aldrich	PZ0017-5MG
Leptomycin B	Sigma-Aldrich	L2913-.5UG
AS1514197	Sigma-Aldrich	PZ0017-5MG
IL-4 Inhibitor	Cayman Chemical	32548
Nonfat dry milk	Cell Signaling Technologies	9999S
Tris	Thermo Fisher Scientific	17926
PVDF Membranes	Bio-Rad	1620177
SYPRO™ Protein Gel Stains	Thermo Fisher Scientific	S12000
Ampicillin Sodium Salt	VWR	0339-EU-25G
Glycerine ≥99.7%	VWR	BDH1172-1LP
Taq DNA Polymerase with Standard Taq Buffer	NEB	M0273
Wizard® Plus SV Minipreps DNA Purification Systems	Promega	A1330
Dulbecco’s Modified Eagle Medium (DMEM)	ATCC	30-2002
RPMI-1640 Medium	ATCC	30-2001
Kaighn’s Modification of Ham’s F-12 Medium (F-12K)	ATCC	30-2004
Fetal Bovine Serum (FBS)	Avantor	89510-186
Poly-D-Lysine, 0.1 mg/ml	Thermo Fisher Scientific	A3890401
0.25% Trypsin-EDTA, Phenol Red	Thermo Fisher Scientific	25200114
Dulbecco’s Phosphate-Buffered Saline (DPBS)	Corning	20-030-CV
Phosphate-Buffered Saline (PBS)	Thermo Fisher Scientific	10010049
Dimethyl Sulfoxide (DMSO)	Thermo Fisher Scientific	20688
Cell Culture Water	Corning	25-055-CV
TransIT®-293 Transfection Reagent	Mirus Bio	MIR 2704
Opti-MEM™ I Reduced Serum Medium	Thermo Fisher Scientific	31985070
Lenti-X Concentrator	TAKARA Bio	631231
qPCR Lentivirus Titer Kit	abm	LV900 (100rxn)
Puromycin Dihydrochloride	Thermo Fisher Scientific	A1113803
Lenti-X Provirus Quantification Kit	TAKARA Bio	631239
ImmunoCult™-SF Macrophage Medium	STEMCELL Technologies	10961
Trypan Blue Solution, 0.4%	Thermo Fisher Scientific	15250061
Recombinant Human Macrophage Colony-Stimulating Factor (M-CSF)	STEMCELL	78057.1
Human IL-4 Recombinant Protein	Thermo Fisher Scientific	200-04-20UG
Human IL-6 Recombinant Protein	Thermo Fisher Scientific	200-06-100UG
Human IL-10 Recombinant Protein	Thermo Fisher Scientific	200-10-10UG
Human IFN-gamma Recombinant Protein	Thermo Fisher Scientific	300-02-100UG
Human IFN-alpha 2 (alpha2a) Recombinant Protein	Thermo Fisher Scientific	300-02AA-500UG
Human TNF-alpha Recombinant Protein	Thermo Fisher Scientific	300-01A-50UG
Lipopolysaccharide (LPS) Solution (500X)	Invitrogen	00-4976-03
Phorbol 12-Myristate 13-Acetate (PMA)	biogems	1652981
Pierce™ D-Luciferin, Monosodium Salt	Thermo Fisher Scientific	88294
Select Agar	Sigma-Aldrich	9002-18-0
Polybrene, 10 mg/ml	americanBIO	AB01643-00001
LentiBOOST-P	Mayflower Bioscience	SB-P-LV-101-10
PathScan® RP Phospho-Stat6 (Tyr641) Sandwich ELISA Kit	Cell Signaling Technology	7275
Pierce™ BCA Protein Assay Kits	Thermo Fisher Scientific	23227
RNeasy Mini Kit	Qiagen	74106
High-Capacity cDNA Reverse Transcription Kit	Applied Biosystems	4374966
PowerTrack™ SYBR Green Master Mix for qPCR	Applied Biosystems	A46112
Phospho-Stat6 (Tyr641) Rabbit mAb	Cell Signaling Technology	Cat#Cat# 56554S; RRID: AB_2799514; RRID: AB_2799514
STAT6 Monoclonal Antibody (7D3)	Thermo Fisher Scientific	Cat# MA5-15659; RRID: AB_10987377
Phospho-Stat3 (Tyr705) (D3A7) XP® Rabbit mAb	Cell Signaling Technology	9145SCat# 9145; RRID: AB_2491009 ; RRID: AB_2491009
Stat3 (124H6) Mouse mAb	Cell Signaling Technology	9139SCat# 9139; RRID: AB_331757; RRID: AB_331757
GAPDH Mouse Antibody	Fitzgerald	−2932Cat# 10R-2932; RRID: AB_11199818 ; RRID: AB_11199818
Goat anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 647	Thermo Fisher Scientific	21245Cat# A-21245; RRID: AB_2535813; RRID: AB_2535813
Alexa Fluor™ 488 Phalloidin	Thermo Fisher Scientific	A12379
Goat anti rabbit IgG (H+L) HRP Conjugated	Millipore Sigma	P307PCat# AP307P; RRID: AB_92641; RRID: AB_11212848
Goat anti mouse IgG (H+L) HRP Conjugated	Millipore Sigma	P308PCat# AP308P; RRID: AB_92635; RRID: AB_92635
SuperSignal™ West Pico PLUS Chemiluminescent Substrate	Thermo Fisher Scientific	34580
Hoechst 33342 solution, 20 mM	Thermo Fisher Scientific	62249
1.5M Tris-HCL PH 8.8	BioRad	1610798
30% Acrylamide/Bis solution 29:1	BioRad	1610156
TEMED	BioRad	1610800
Ammonium Persulfate (APS)	BioRad	1610700
Halt™ Protease and Phosphatase Inhibitor Cocktail (100X)	Thermo Fisher Scientific	78440
RIPA Lysis Buffer	Thermo Fisher Scientific	89900
Precision Plus Protein Dual Color Standards	BioRad	1610374
MycoAlert® PLUS Mycoplasma Detection Kit	Lonza	LT07-710
Deposited Data
Ctrl and IL4 replicates data and Figure 7B parameter values. The code uses the best parameter set identified in the ABC-SMC search to generate response curves for IL4 stimulation. These curves are then plotted against the experimental data.	Zenodo	https://doi.org/10.5281/zenodo.14545524
Software and Algorithms
Thermo Fisher Tm Calculator	Thermo Fisher Scientific	N/A
NEBiocalculator®	NEB	N/A
Graphpad Prism	Graphpad	10.0.3
Benchling	Benchling	N/A
Basic Local Alignment Search Tool (BLAST)	National Institutes of Health	N/A
Code for JAKSTAT systems model	This paper	10.5281/zenodo.14545524
Others
NanoDrop Spectrophotometer	Thermo Fisher Scientific	2000c
Multimode microplate reader	Tecan	Spark
DeNovix Cell Counter (or equivalent cell counter)	DeNovix	CellDrop FL
Tissue Culture Flask (T-25, T-75)	VWR	10062-872, 10062-860
Petri Dishes (90 mm diameter)	Thermo Fisher Scientific	263991
Centrifuge Tubes (5 ml, 15 ml, 50 ml)	VWR	10002-738, 10025-686, 21008-242
Microcentrifuge Tubes (0.5 ml, 1.7 ml)	VWR	89000-010, 870030296
Cryogenic Tubes	Thermo Fisher Scientific	374510
Nuclease-Free 0.2 ml PCR Tube	Thermo Fisher Scientific	AB0620
Disposable Culture Tube (18x150 mm)	VWR	47729-583
Culture Tube Cap (18 mm)	VWR	76184-780
Tissue Culture Plate (6-wells)	VWR	10062-892
Nunc™ Lab-Tek™ II Chamber Slide™ System	Thermo Fisher Scientific	154526PK
96-Well Assay Plate (white)	Corning	3610
Cell Culture Microplate, 96 Well, PS, U-Bottom	Greiner Bio-One	650979
Inoculating Loop	VWR	89126-872
Thermocycler	Applied Biosystems	2720
Centrifuge	Eppendorf	5810R, 5418
Dry Block Heater	VWR	75838-282, 13259-286
Mr. Frosty™ Freezing Container	Thermo Fisher Scientific	5100-0001
CO2 Incubator	Thermo Fisher Scientific	51033557
Incubator Shaker	Eppendorf	Innova 42
Gel Electrophoresis Equipment	Bio-Rad	PowerPac™ basic power supply, sub-cell model 96
ChemiDoc™ Imaging System	Bio-Rad	MP
Real-Time PCR Instrument	Applied Biosystems	ViiA 7
Laser Scanning Microscope	Zeiss	Zeiss LSM 880

Highlights.

• Bioluminescence enables non-invasive, longitudinal macrophage polarization measurement

• Real-time reporting of TF activity reveals stimulus-specific signaling dynamics

• Computational model links time-resolved TF activity with underlying signaling kinetics

• TF measurements of macrophage polarization performed in multiarray 2D and 3D culture

Data and code availability

• Source data statement: data reported in this paper will be shared by the lead contact upon request.

• Code statement: all original code has been deposited at Zenodo and is publicly available as of the date of publication at Zenodo: https://doi.org/10.5281/zenodo.14545524.

• Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.