Near-Infrared Fluorescent Highly Branched Poly(β-Amino Ester)s Nanoparticles for Gene Delivery: In vivo Biodistribution and Safety Evaluation Authors: Hongzhen Zhang, Jiahao Liu, Rui Huang, Ailin Hu, Rui Miao, Guang Chen, Haonan Li, Wenxin Wang, Zhonglei He Published: 2026 ABSTRACT AbstractBackgroundCationic highly branched poly(β-amino ester)s (HPAEs) represent a promising class of nonviral gene-delivery polymers; however, their in vivo distribution and biological fate remain challenging to monitor. Here, we explored covalent conjugation of indocyanine green (ICG) as a strategy to impart near-infrared-I (NIR-I) fluorescence to HPAEs while maintaining their DNA complexation capacity and gene-delivery performance.MethodsHPAE was modified with increasing feed amounts of ICG-N-hydroxysuccinimide (ICG-NHS), generating a series of fluorescent polymers designated HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9, where the numerical suffixes indicate ICG-NHS feed volumes rather than substitution ratios. The resulting conjugates and their DNA nanoparticles were characterized by spectroscopic and chromatographic analyses, DNA-binding assays, dynamic light scattering, zeta-potential measurements, transmission electron microscopy, and optical-stability evaluation. In vitro gene-delivery activity and cytocompatibility were assessed in multiple cell models, while systemic distribution, biocompatibility, and tissue responses were evaluated in healthy BALB/c mice following administration.ResultsIncreasing ICG-NHS feed resulted in tunable incorporation of fluorescent moieties into the HPAE backbone. Among the tested formulations, HPAE-3 exhibited an apparent amine substitution degree of 18.98% ± 0.54% and a fluorescence emission maximum at approximately 834 nm. HPAE-3-based nanoparticles displayed favorable physicochemical properties, including hydrodynamic diameters of approximately 170–290 nm, low-to-moderate dispersity (PDI, 0.18–0.40), positive surface potentials (+23 to +40 mV), and efficient DNA condensation at polymer/DNA ratios ≥20:1. Importantly, ICG incorporation at this level preserved reporter-gene expression in HEK293T, RAW264.7, and MLE-12 cells while maintaining acceptable cytocompatibility. Following systemic administration in mice, HPAE-3 mediated luciferase reporter-gene expression predominantly in the liver, spleen, and lungs. HPAE-3-associated NIR-I fluorescence was most evident in the liver and lungs at 6 h, with a weaker signal in the spleen, and declined thereafter; by 72 h, residual ex vivo fluorescence was detected predominantly in the liver. No apparent acute tissue damage, significant alterations in serum biochemical parameters, or deviations in body-weight profiles were observed compared with control groups.ConclusionHPAE-3 achieved a balanced integration of NIR-I fluorescence, DNA-delivery capability, nanoparticle stability, and preliminary in vivo tolerability. These results establish ICG-labeled HPAE as a potential platform for noninvasive visualization of polymer-mediated gene delivery and provide a foundation for further investigation of its biodistribution, intracellular fate, and therapeutic applications.Keywords: hyperbranched polymers, indocyanine green, nanoparticles, gene delivery, in vivo fluorescence imaging Introduction The development of nonviral gene delivery systems requires materials that can condense nucleic acids into nanoscale complexes, retain adequate stability in biological environments, support intracellular delivery, and limit carrier-associated toxicity. Among these materials, highly branched poly(β-amino ester)s (HPAEs) have attracted attention because their hydrolysable ester-containing backbones, high density of amino groups, and branched macromolecular architecture provide multiple opportunities for nucleic acid complexation and chemical functionalization. Previous studies have demonstrated that polymer topology, molecular composition, terminal groups, and charge distribution strongly influence the DNA- and RNA-delivery performance of HPAEs.1–6 These findings establish HPAEs as adaptable nonviral vectors, while also showing that their biological activity is highly dependent on polymer structure, formulation conditions, and the recipient cell type. Despite progress in optimizing HPAE-mediated gene delivery, the behavior of these polymeric vectors after systemic administration remains difficult to define. Reporter-gene expression provides a functional measure of delivery because it requires cellular uptake, intracellular processing, transcription, and translation; however, it does not directly reveal the physical location of the polymeric carrier. Fluorescent labeling offers a complementary approach for visualizing carrier-associated signals, but the resulting fluorescence must be interpreted cautiously. The chemical identity, conjugation density, and intraparticle location of a fluorophore can alter nanoparticle properties and biodistribution, while degradation or dissociation of a labeled carrier may generate fluorescent products that no longer represent the intact nanostructure.7,8 Fluorescence imaging therefore provides information on label-associated material rather than definitive evidence of intact-particle distribution or continued association between the carrier and its nucleic acid cargo. Indocyanine green (ICG) is a near-infrared fluorescent dye suitable for imaging within the NIR-I optical window. Although NIR-II imaging provides greater tissue penetration, ICG remains the only FDA-approved near-infrared fluorophore with broad clinical use.9–11 Therefore, NIR-I imaging based on ICG continues to represent the most practical strategy for translational fluorescent tracking of nanomedicines. Nevertheless, unformulated ICG is susceptible to concentration-dependent aggregation and quenching, chemical and photochemical degradation, nonspecific protein interactions, and rapid changes in its biological distribution.12–14 Incorporation of ICG into polymeric systems can modify its optical stability and in vivo behavior, but may also alter the charge, hydrophobicity, assembly, and biological interactions of the carrier.12,13,15 The NHS-activated derivative of ICG can react with primary amino groups, providing a practical route for covalent modification of amine-containing polymers. However, increasing the amount of ICG-NHS introduced during synthesis may consume cationic sites required for nucleic acid binding, promote hydrophobic association or fluorescence quenching, and affect nanoparticle size, dispersity, surface charge, colloidal behavior, transfection performance, and cytocompatibility. Consequently, fluorescent labeling must be optimized rather than maximized. Previous studies have incorporated ICG into polymeric micelles, nanocapsules, and highly branched polymer systems for fluorescence imaging, stimulus-responsive release, or phototherapy.12,13,16 However, those investigations did not systematically address whether covalent ICG labeling of an HPAE gene-delivery polymer could preserve DNA complexation and reporter-gene expression while providing a measurable NIR-I signal. A rigorous evaluation requires more than confirmation of fluorescence: the amount of ICG-NHS added during synthesis must be distinguished from its actual incorporation into the purified polymer, and the consequences of modification for molecular-weight distribution, particle size, DLS polydispersity index, zeta potential, morphology, stability, cytocompatibility, and cell-dependent gene expression must be considered together. Moreover, organ-associated fluorescence following systemic administration should be distinguished from active targeting, pharmacokinetic clearance, and the physical distribution of intact polymer/DNA nanoparticles. In the present study, HPAE was reacted with increasing feed volumes of ICG-NHS to prepare a formulation series designated HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9. The numerals identify the ICG-NHS stock-solution volumes used during synthesis and do not represent molar or substitution ratios. ICG incorporation was evaluated using Fourier-transform infrared spectroscopy, proton nuclear magnetic resonance spectroscopy, ultraviolet–visible absorption spectroscopy, fluorescence spectroscopy, and gel permeation chromatography. After complexation with plasmid DNA, the resulting HPAE-based NPs were assessed for DNA binding, hydrodynamic diameter, DLS polydispersity index, zeta potential, morphology, and short-term optical stability. Reporter-gene expression was examined in HEK293T, RAW264.7, MLE-12, and L-02 cells, while cytocompatibility was evaluated in HEK293T cells. In healthy BALB/c mice, bioluminescence imaging was used to assess functional luciferase expression, whereas time-resolved NIR-I imaging was used to examine ICG-associated fluorescence. Histopathology, serum biochemical measurements, and body-weight monitoring were included as preliminary tolerability assessments. The study was designed to identify a formulation that balanced fluorescent visibility with gene-delivery performance; it was not intended to demonstrate active organ targeting, intact-particle pharmacokinetics, synchronous polymer–DNA transport, or therapeutic efficacy. The overall formulation design and experimental workflow, including HPAE–ICG conjugation, DNA-polyplex formation, reporter-gene expression, and NIR-I fluorescence assessment, are summarized schematically in Figure 1. Schematic overview of HPAE–ICG conjugation, DNA-polyplex formation, and experimental evaluation. Primary amino groups of highly branched poly(β-amino ester)s (HPAEs) were reacted with ICG-NHS to generate HPAE–ICG conjugates, which subsequently complexed negatively charged plasmid DNA through electrostatic interactions to form cationic nanoparticles. Reporter-gene expression and ICG-associated NIR-I fluorescence were evaluated after systemic administration. Early HPAE-3-associated fluorescence was most evident in the liver and lungs, with a weaker splenic signal, whereas residual fluorescence at 72 h was predominantly hepatic. The schematic is conceptual and not drawn to scale. The fluorescence signal represents ICG-associated material and does not establish active targeting or the distribution of intact nanoparticles. Materials and Methods Synthesis and Purification of HPAE–ICG Conjugates The highly branched poly(β-amino ester)s (HPAEs) used in this study were proprietary materials supplied by Branca Bunús Ltd. (Dublin, Ireland; Batch No. 24052R05-41,223). N-Hydroxysuccinimide-activated indocyanine green (ICG-NHS) was purchased from Yunying Biotechnology Co., Ltd. (Nanjing, China). The gWiz-GFP and firefly luciferase reporter plasmids were obtained from General Biosystems Co., Ltd. (Anhui, China). Deuterated chloroform (CDCl3), lithium bromide (LiBr), dimethyl sulfoxide (DMSO), N,N-dimethylformamide (DMF), and diethyl ether were purchased from Merck (Darmstadt, Germany). Quant-iT PicoGreen double-stranded DNA reagent was obtained from Thermo Fisher Scientific (Waltham, MA, USA). Unless otherwise specified, all reagents were of analytical grade and used according to the manufacturers’ instructions. HPAE–ICG conjugates were synthesized through the reaction between the primary amino groups of HPAE and the activated NHS ester of ICG-NHS. HPAE and ICG-NHS stock solutions were prepared separately in DMSO. For each reaction, 375 μL of the HPAE stock solution was mixed with 0, 1, 3, 5, 7, or 9 μL of the ICG-NHS stock solution, as detailed in Table S1. The reaction mixtures were protected from light and stirred at 25°C for 3 h. The resulting formulations were designated HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9. HPAE-0 represents unmodified HPAE. For HPAE-1 to HPAE-9, the numeral denotes the volume, in microliters, of ICG-NHS stock solution added during synthesis while the HPAE feed volume remained constant. These designations are formulation codes and do not represent molar ratios, mass ratios, or experimentally determined substitution degrees. After the reaction, each mixture was added dropwise to excess cold diethyl ether to precipitate the polymer. The precipitated polymer was collected and washed three times with diethyl ether. Washing was continued until no detectable fluorescence was observed in the supernatant, thereby minimizing residual unreacted or non-covalently associated ICG-NHS. The purified products were dried, protected from light, and stored at −20°C until further use. Molecular Weight and Dispersity Measurements The number-average molecular weight (Mn), weight-average molecular weight (Mv), and molecular-weight dispersity (Đ) of HPAE-0 and the purified HPAE–ICG conjugates were determined by gel permeation chromatography equipped with a refractive-index detector. Polymer samples were dissolved in DMF at 5 mg/mL and filtered through a 0.20-μm membrane before analysis. DMF containing 0.1% LiBr was used as the mobile phase at a flow rate of 1.0 mL/min, and the column temperature was maintained at 60°C. Molecular-weight calibration was performed using linear poly(methyl methacrylate) standards. Molecular-weight dispersity was determined from the ratio of Mv to Mn. In this study, Đ refers specifically to the molecular-weight distribution of the polymer measured by GPC and is distinct from the nanoparticle polydispersity index determined by dynamic light scattering. The GPC-derived values reported in Table S2 were obtained from the original instrument output. Structural and Optical Characterization of HPAE–ICG Conjugates Fourier-Transform Infrared Spectroscopy Fourier-transform infrared spectroscopy was performed to provide supporting evidence of structural changes following ICG-NHS modification. Dried samples of HPAE-0 and the purified HPAE–ICG conjugates were analyzed over the wavenumber range of 4000–500 cm−1. Background subtraction and baseline correction were performed before comparison of the characteristic absorption bands associated with the HPAE backbone and ICG-containing structures. FTIR results were interpreted together with 1H nuclear magnetic resonance and optical spectroscopy and were not used as the sole evidence of covalent conjugation. Proton Nuclear Magnetic Resonance Spectroscopy The chemical structures of HPAE-0 and the purified HPAE–ICG conjugates were characterized by proton nuclear magnetic resonance spectroscopy. Dried polymer samples were dissolved directly in CDCl3 and transferred to NMR tubes. Spectra were acquired using a Bruker AVANCE NEO 400 MHz spectrometer (Bruker, Switzerland). For semi-quantitative analysis, the ICG-associated aromatic region at δ 8.3–8.0 ppm was designated A2, whereas the HPAE reference region at δ 4.2–4.0 ppm was designated A1. The A1 integral was normalized to 12, and the relative A2 integral was determined for each formulation. The resulting values were used to compare the relative incorporation of ICG among the formulations. Because of peak overlap and differences in spectral relaxation behavior, the NMR-derived values were not interpreted as exact molar substitution ratios. UV–Visible Absorption and Fluorescence Spectroscopy UV–visible absorption spectra of ICG-NHS and the purified HPAE–ICG conjugates were recorded in DMSO over a wavelength range of 300–900 nm. To examine concentration-dependent optical behavior, HPAE-5 was analyzed after 320-, 640-, 2560-, 5120-, and 10,240-fold dilution in DMSO. Fluorescence emission spectra were recorded following excitation at 780 nm, and emission was collected over the range of 780–900 nm. Samples within the same experiment were measured using identical acquisition settings, and the corresponding solvent background was subtracted before comparison. ICG-NHS Calibration and Estimation of ICG Incorporation A calibration curve was established using a series of ICG-NHS standards prepared in DMSO. Absorbance at 780 nm was measured in triplicate. Linear regression yielded the equation A = 3.2877C + 0.04510, with an R2 value of 0.99997, where A represents absorbance at 780 nm and C represents the ICG-NHS concentration in mg/mL. For estimation of ICG incorporation in HPAE-3, 5 μL of the HPAE-3 stock solution at 50 mg/mL was diluted to a final volume of 100 μL with DMSO. Absorbance at 780 nm was measured, and the apparent ICG-NHS-equivalent concentration was calculated using the calibration curve. The apparent ICG content per HPAE chain and the apparent proportion of substituted amino groups were estimated from the measured ICG-equivalent content, an average HPAE molecular weight of approximately 25 kDa, and approximately ten potentially reactive amino groups per HPAE chain, as estimated from the NMR analysis. Because conjugation to the polymer may alter the optical extinction coefficient of ICG, the calculated value was regarded as an apparent ICG-NHS-equivalent loading estimate rather than an absolute substitution measurement. Preparation of HPAE–ICG NPs HPAE-0 and the purified HPAE–ICG conjugates were prepared as stock solutions in DMSO at 50 mg/mL. Plasmid DNA was prepared in nuclease-free water at 1 mg/mL. A 25 mM sodium acetate buffer was prepared from a commercial 3 M stock solution and adjusted to pH 5.2. For nanoparticle formation, the polymer and DNA stock solutions were separately diluted to the required concentrations in sodium acetate buffer. Equal volumes of the diluted polymer and DNA solutions were combined, vortexed for 10s, and incubated for 2 min at room temperature to allow electrostatic complexation. Unless otherwise specified, DNA was diluted to 0.1 mg/mL before mixing, whereas the polymer was diluted to 1, 2, 3, or 4 mg/mL to prepare polymer-to-DNA mass ratios of 10:1, 20:1, 30:1, and 40:1, respectively. The resulting HPAE–ICG NPs were freshly prepared and subsequently diluted into the appropriate culture medium for cellular experiments. For in vivo experiments, HPAE–ICG NPs were prepared at a polymer-to-DNA mass ratio of 20:1 and a DNA concentration of 300 μg/mL. The formulations were prepared under sterile conditions immediately before administration. The final concentration of DMSO in the intravenously administered formulations did not exceed 2% (v/v). Cell Culture Human embryonic kidney HEK293T cells, murine macrophage-like RAW264.7 cells, murine lung epithelial MLE-12 cells, and human hepatocyte-lineage L-02 cells were obtained from the National Collection of Authenticated Cell Cultures, Chinese Academy of Sciences (Shanghai, China).HEK293T and RAW264.7 cells were maintained in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin. MLE-12 cells were maintained in DMEM/F12 medium at a 1:1 ratio supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin. L-02 cells were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin. All cells were cultured at 37°C in a humidified incubator containing 5% CO2 and were routinely passaged before reaching complete confluence. Cell Transfection and GFP Image Analysis Cells were seeded approximately 24 h before transfection at a density of 2 × 104 cells/cm2 and transfected at approximately 50–70% confluence. For HEK293T cells, gWiz-GFP plasmid was complexed with HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, or HPAE-9 at polymer-to-DNA mass ratios of 10:1, 20:1, 30:1, and 40:1. For RAW264.7, MLE-12, and L-02 cells, the polymer-to-DNA mass ratio was fixed at 20:1. The final plasmid concentration in the culture medium was 5 μg/mL, and unmodified BrB202 supplied by Branca Bunús Ltd. was used as the HPAE-0 comparator. GFP expression was evaluated 48 h after transfection using an EVOS M7000 imaging system (Thermo Fisher Scientific, Waltham, MA, USA). Images within each experiment were acquired using identical objective magnification, illumination intensity, exposure time, gain, and image-processing settings. GFP fluorescence was quantified using ImageJ software after applying the same background-subtraction and thresholding procedures to all groups. For HEK293T cells, reporter expression was quantified using background-corrected GFP integrated density, whereas for RAW264.7, MLE-12, and L-02 cells, it was quantified as the mean background-corrected fluorescence intensity. Three independent replicate measurements were included for each experimental condition. CCK-8 Cytotoxicity Assay Cell viability was evaluated in HEK293T cells 48 h after transfection using the CCK-8 assay. The culture medium was removed and replaced with 100 μL of fresh complete medium containing 10% CCK-8 reagent. After incubation at 37°C for 30 min, absorbance at 450 nm was measured using a SpectraMax M3 microplate reader (Molecular Devices, San Jose, CA, USA).Absorbance from cell-free blank wells was subtracted from all measurements. Cell viability was expressed as a percentage of the absorbance measured in untreated control cells. Three independently prepared replicate samples were analyzed for each condition, and the results are presented as the mean ± standard deviation. PicoGreen DNA-Binding Assay The DNA-binding capacity of HPAE-0 and the HPAE–ICG formulations was evaluated using the Quant-iT PicoGreen double-stranded DNA assay. Freshly prepared HPAE–ICG NPs were mixed with an equal volume of PicoGreen working solution in black 96-well plates and incubated for 5 min at room temperature in the dark. Fluorescence was measured using a microplate reader at the excitation and emission wavelengths recommended by the manufacturer. Naked plasmid DNA at the same concentration was used as the fully accessible DNA control, and reagent-only wells were included for background correction. DNA-binding efficiency was calculated from the reduction in PicoGreen-accessible DNA fluorescence relative to the naked-DNA control. Three independently prepared samples were analyzed for each formulation. Particle Size, Polydispersity Index, Zeta Potential, Morphology, and Short-Term Stability Hydrodynamic Diameter, DLS Polydispersity Index, and Zeta Potential The hydrodynamic diameter, DLS polydispersity index, and zeta potential of freshly prepared HPAE–ICG NPs were measured using a Zetasizer Pro instrument (Malvern Panalytical, Malvern, UK). For hydrodynamic diameter and polydispersity measurements, 100 μL of the freshly prepared nanoparticle suspension was transferred to a disposable low-volume cuvette and analyzed without further dilution. For zeta-potential measurements, the suspension was diluted tenfold in deionized water and transferred to a folded capillary cell. Measurements were performed at 25°C using the instrument’s automatic optimization settings. Three independently prepared samples were measured for each formulation, and the results are presented as the mean ± standard deviation. The term PDI in this subsection refers specifically to the cumulants-based DLS polydispersity index and is distinct from the GPC molecular-weight dispersity, Đ. Transmission Electron Microscopy The morphology and dry-state particle size of the selected HPAE-3 NPs were examined by transmission electron microscopy. A freshly prepared suspension was deposited onto a carbon-coated copper grid. After adsorption, excess liquid was removed using filter paper, and the grid was allowed to dry before imaging. Representative transmission electron microscopy images were obtained, and particle diameters were quantified using ImageJ software. At least 100 individual particles were measured to generate the number-based dry-state particle-size distribution. The dry-state diameter obtained by TEM was interpreted separately from the hydrodynamic diameter measured by DLS because the two methods assess particles under different physical conditions. Short-Term Optical Stability The short-term optical stability of HPAE-0 and the HPAE–ICG formulation series was evaluated in PBS and an FBS-containing medium. Each formulation was incubated in the corresponding medium at 37°C, and absorbance at 660 nm was measured at 0, 15, 30, 45, 60, 90, and 120 min. Values were normalized to the absorbance measured at 0 min. Three independently prepared samples were analyzed for each formulation, and the results are presented as the mean ± standard deviation. Because this experiment monitored absorbance rather than hydrodynamic diameter, it was considered a short-term optical-stability assessment rather than a definitive evaluation of long-term colloidal or plasma stability. Animal Experiments and in vivo Imaging Female BALB/c mice, 8–10 weeks of age and weighing 18–22 g, were obtained from Hangzhou Ziyuan Biotechnology Co., Ltd. (Hangzhou, China). Animals were maintained under specific pathogen-free conditions with controlled temperature and humidity, a 12-h light/dark cycle, and ad libitum access to standard food and water. Mice were acclimatized before experimentation.T he in vivo studies were exploratory, and three mice were included in each experimental group. For reporter-gene expression experiments, the firefly luciferase plasmid was administered either as naked DNA or after complexation with HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, or HPAE-9. HPAE–ICG NPs were prepared at a polymer-to-DNA mass ratio of 20:1 and a DNA concentration of 300 μg/mL. Each mouse received 200 μL of the corresponding formulation by tail-vein administration, equivalent to 60 μg of plasmid DNA per mouse. At 6 h after administration, mice received 150 μL of D-luciferin solution at 30 mg/mL in PBS by intraperitoneal administration. Bioluminescence images were acquired using a Tanon ABL-X5 in vivo imaging system (Tanon Science and Technology, Shanghai, China). Bioluminescence was interpreted as a readout of functional luciferase reporter-gene expression rather than the physical distribution of the polymer carrier. For NIR-I fluorescence imaging, mice received HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, HPAE-9, or ICG-NHS. Whole-animal fluorescence images were obtained at 0.5, 6, 24, 48, and 72 h after administration using the appropriate excitation and emission channels of the Tanon ABL-X5 system. Images included within the same comparison were acquired using identical imaging parameters and displayed using a common radiance scale. At 6 or 72 h after administration, mice assigned to the corresponding terminal groups were euthanized, and the heart, liver, spleen, lungs, and kidneys were collected for ex vivo bioluminescence or NIR-I fluorescence imaging. Organs were arranged in a consistent order, imaged using identical acquisition settings, and analyzed using Tanon Image Analysis Software. Regions of interest were applied consistently to the whole-animal and organ images. NIR-I fluorescence was interpreted as an ICG-associated signal that could originate from labeled polymer, polymer fragments, or ICG-containing degradation products. It was not considered definitive evidence of intact HPAE–ICG NPs or synchronous transport of the polymer and plasmid DNA. The ICG-NHS group was described as the ICG-NHS control rather than as free ICG because the administered compound was the NHS-activated derivative. Histopathological Analysis At 72 h after administration, the heart, liver, spleen, lungs, and kidneys were collected and fixed overnight in 4% paraformaldehyde. Tissues were dehydrated, embedded in paraffin, sectioned at a thickness of 5 μm, and stained with hematoxylin and eosin according to standard procedures. Histological sections from the control, HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9 groups were examined under a light microscope. Representative images were acquired using identical magnification settings, with scale bars of 100 μm. Histopathological evaluation was performed independently by two pathologists who were blinded to treatment allocation. Sections were examined for evidence of inflammatory-cell infiltration, edema, hemorrhage, tissue degeneration, necrosis, and other treatment-associated pathological alterations. Blood Biochemical Analysis and Body-Weight Monitoring At 72 h after administration, blood samples were collected under anesthesia by retro-orbital sampling. Samples were allowed to clot and were subsequently centrifuged to obtain serum. Serum alanine aminotransferase, aspartate aminotransferase, blood urea nitrogen, uric acid, and creatinine concentrations were measured using a Chemray 800 fully automated biochemical analyzer (Rayto Life and Analytical Sciences Co., Ltd., Shenzhen, China) according to the manufacturer’s instructions. These parameters were used as preliminary indicators of hepatic and renal function. For longer-term observation, body weights of control mice and mice receiving HPAE-3 NPs were recorded throughout the 30-day observation period using the same calibrated balance. Three mice were included in each group, and the results are presented as the mean ± standard deviation. Statistical Analysis Statistical analyses were performed using GraphPad Prism version 10.0 (GraphPad Software, Boston, MA, USA). Data are presented as the mean ± standard deviation (SD). The value of n represents independent biological replicates or individual animals, as specified in the corresponding figure legends; technical measurements were not treated as independent replicates. For comparisons involving one experimental factor, ordinary one-way analysis of variance (ANOVA) followed by Dunnett’s multiple-comparisons test was used when each formulation was compared with HPAE-0, whereas Tukey’s test was used for all pairwise comparisons. Data involving both formulation and organ type were analyzed using two-way ANOVA followed by Šídák’s multiple-comparisons test. Longitudinal body-weight data were analyzed using two-way repeated-measures ANOVA. When the assumptions for parametric analysis were not met, the Kruskal–Wallis test followed by Dunn’s multiple-comparisons test was applied. All tests were two-sided, and P<0.05 was considered statistically significant. Significance is denoted as follows: ns, not significant; *P<0.05; **P<0.01; ***P<0.001; and ****P<0.0001. Ethical Approval and Animal Welfare All animal experiments were reviewed and approved by the Biomedical Research Ethics Committee of Anhui University of Science and Technology (Approval No. SZ2024-004). All animal procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, Eighth Edition, and the institutional animal-care and use requirements of Anhui University of Science and Technology. For tail-vein administration and in vivo imaging, mice were anesthetized with 2% isoflurane in oxygen delivered at a flow rate of 1 L/min. At the designated experimental endpoint, mice were euthanized by carbon-dioxide inhalation. The carbon-dioxide displacement rate was maintained at approximately 20% of the chamber volume per minute. Death was confirmed by the absence of respiration and heartbeat, followed by cervical dislocation as a secondary physical method. Euthanasia procedures were performed in accordance with the American Veterinary Medical Association Guidelines for the Euthanasia of Animals. 3.Results Synthesis and Physicochemical Characterization of HPAE–ICG Conjugates and HPAE-Based NPs A series of HPAE–ICG conjugates was prepared by reacting the primary amino groups of HPAE with the activated ester of ICG-NHS. The formulations were designated HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9. HPAE-0 represents unmodified HPAE, whereas the numerals in HPAE-1 to HPAE-9 indicate the volume, in microliters, of ICG-NHS stock solution added to a fixed volume of HPAE stock solution. Accordingly, these labels are formulation codes and do not represent molar ratios, mass ratios, or experimentally determined substitution degrees (Table S1). The characteristic resonances of the HPAE backbone are labeled according to the corresponding proton environments, while the ICG-associated aromatic signals are observed at approximately δ 7.0–8.2 ppm. The simultaneous presence of HPAE and ICG resonances provides additional structural evidence for ICG incorporation into the HPAE conjugate (Figure S1). FTIR spectroscopy was first used to examine structural changes following ICG-NHS modification. HPAE-0 and the purified HPAE–ICG conjugates retained the broad O–H/N–H band at approximately 3522 cm−1 and the aliphatic C–H band near 2920 cm−1. Formulation-dependent changes were observed in the carbonyl and fingerprint regions, including bands around 1641, 1431, 1305, 1017, and 720 cm−1 (Figure S2). These spectral changes were consistent with the introduction of ICG-associated chemical groups while preserving the principal spectral features of the HPAE backbone. In the 1H NMR spectra, the characteristic signals of the HPAE backbone remained detectable after purification, whereas additional aromatic resonances appeared at approximately δ 7.3–8.3 ppm in the ICG-modified formulations (Figure 2A and Figure S3). These aromatic signals were negligible in HPAE-0 and became progressively more evident with increasing ICG-NHS feed volume. Semi-quantitative integration of the more selective ICG-associated region at δ 8.3–8.0 ppm was performed relative to the HPAE reference region at δ 4.2–4.0 ppm, which was normalized to 12. The corresponding ICG-associated integrals were 0, 0.16, 0.24, and 0.48 for HPAE-0, HPAE-1, HPAE-3, and HPAE-9, respectively (Figure S4 and Table S3). Thus, the selected formulations showed a feed-dependent increase in the relative ICG-associated signal. Structural, optical, DNA-binding, and physicochemical characterization of HPAE–ICG conjugates and HPAE/DNA nanoparticles. (A) Representative 1H NMR spectra of ICG-NHS, unmodified HPAE (HPAE-0), and the purified HPAE–ICG conjugates HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9 recorded in CDCl3. The red dashed box highlights the ICG-associated aromatic proton region at approximately δ 7.0–8.2 ppm. (B) Gel permeation chromatography elution profiles of HPAE-0 and the purified HPAE–ICG conjugates. The corresponding molecular-weight parameters and GPC-derived dispersity values are provided in Table S2. (C) Fluorescence emission spectra of the purified HPAE–ICG conjugates following excitation at 780 nm. The emission maximum was observed at approximately 834 nm. (D) PicoGreen-based DNA-entrapment efficiency of HPAE-0 and the HPAE–ICG formulations at polymer-to-DNA mass ratios of 10:1, 20:1, 30:1, and 40:1. DNA entrapment was calculated from the reduction in PicoGreen-accessible DNA fluorescence relative to naked plasmid DNA at the same concentration after background subtraction. The dashed horizontal line indicates 75% DNA entrapment. (E) Hydrodynamic diameter and DLS polydispersity index of HPAE/DNA and HPAE–ICG/DNA nanoparticles prepared at a polymer-to-DNA mass ratio of 20:1. Bars represent the mean hydrodynamic diameter on the left y-axis, and the line represents the DLS polydispersity index on the right y-axis. (F) Zeta potentials of the corresponding nanoparticles prepared at a polymer-to-DNA mass ratio of 20:1. Data in (D–F) are presented as mean ± SD from three independently prepared samples (n = 3). An independent UV–Vis calibration approach was used to estimate ICG incorporation in HPAE-3. The ICG-NHS calibration curve at 780 nm was linear over the tested concentration range, with the regression equation A = 3.2877C + 0.04510 and an R2 value of 0.99997 (Figure S5). Based on the measured ICG-NHS-equivalent content, an average HPAE molecular weight of approximately 25 kDa, and an estimated ten potentially reactive amino groups per polymer chain, the apparent amine substitution for HPAE-3 was calculated as 18.98% ± 0.54%. This value corresponds to approximately 1.9 ICG-NHS-equivalent molecules per HPAE chain under the stated assumptions. Because covalent attachment may alter the extinction coefficient of ICG, this calculation was regarded as an apparent ICG-NHS-equivalent loading estimate rather than an absolute grafting degree. GPC analysis showed broadly overlapping elution profiles for HPAE-0 and the purified HPAE–ICG conjugates (Figure 2B). The reported weight-average molecular weights ranged from 21.97 to 25.93 kDa, with values of 24.58, 23.86, 25.37, 21.97, 24.98, and 25.93 kDa for HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9, respectively. The corresponding GPC dispersity values ranged from 6.57 to 10.31 (Table S2). The absence of a pronounced displacement of the principal GPC peaks suggested that ICG-NHS modification did not produce a major change in the overall elution characteristics of the polymer. The GPC-derived dispersity describes the molecular-weight distribution of the polymer and is distinct from the DLS polydispersity index of the resulting nanoparticles. Optical characterization further demonstrated the introduction of ICG-associated near-infrared properties. The purified HPAE–ICG conjugates exhibited a broad absorption band centered at approximately 793 nm, whereas no corresponding near-infrared peak was expected for HPAE-0 (Figure S5 and S6). Serial dilution of HPAE-3 showed a concentration-dependent reduction in absorbance while preserving the position of the near-infrared absorption maximum (Figure S7). Following excitation at 780 nm, the conjugates displayed an emission maximum at approximately 834 nm. Among the formulations compared under identical measurement conditions, HPAE-3 showed the highest fluorescence emission intensity, whereas the signal decreased in the higher-feed formulations HPAE-5, HPAE-7, and HPAE-9 (Figure 2C). Serial dilution of HPAE-5 produced the expected concentration-dependent change in fluorescence intensity without an appreciable shift in the emission maximum (Figure S8). These findings indicate that the conjugates retained NIR-I fluorescence but also demonstrate that increasing the ICG-NHS feed volume did not lead to a proportional increase in fluorescence output, which may reflect formulation-dependent dye–dye interactions or fluorescence quenching. The DNA-binding capacity of the polymers was evaluated over polymer-to-DNA mass ratios of 10:1–40:1. At 10:1, DNA entrapment showed modest formulation-dependent variation, with the mean values remaining above approximately 90% for most formulations. Increasing the mass ratio to 20:1, 30:1, or 40:1 resulted in approximately 95–100% DNA entrapment for all formulations (Figure 2D). These results indicate that ICG modification did not abolish the ability of HPAE to electrostatically complex plasmid DNA. However, the data did not establish that ICG modification enhanced DNA binding relative to HPAE-0. At a polymer-to-DNA mass ratio of 20:1, the HPAE-based NPs had mean hydrodynamic diameters of approximately 170–290 nm (Figure 2E). HPAE-0 NPs were approximately 195 nm, whereas HPAE-1 and HPAE-3 produced the largest hydrodynamic diameters, at approximately 290 and 280 nm, respectively. HPAE-5, HPAE-7, and HPAE-9 showed approximate diameters of 240, 170, and 210 nm, respectively. The corresponding DLS polydispersity indices were approximately 0.18–0.40. HPAE-3 showed the lowest mean PDI, at approximately 0.18, whereas HPAE-5 and HPAE-7 showed broader distributions, with values of approximately 0.34 and 0.39, respectively. All formulations retained positive surface charge. The mean zeta potentials were approximately +25 mV for HPAE-0, +29 mV for HPAE-1, +23 mV for HPAE-3, +27 mV for HPAE-5, +31 mV for HPAE-7, and +40 mV for HPAE-9 (Figure 2F). Thus, although the extent of ICG-NHS modification affected particle size, dispersity, and surface charge, all formulations remained cationic under the measurement conditions. TEM analysis of the selected HPAE-3 formulation showed nanoscale particulate structures in the dry state (Figure S11A). Image analysis of 100 particles yielded a mean dry-state diameter of 33.39 ± 8.01 nm (Figure S11B). The substantially smaller TEM diameter relative to the DLS hydrodynamic diameter was consistent with differences between dry-state electron microscopy and solution-phase DLS, including dehydration and the absence of the hydrated polymer layer during TEM analysis. Short-term optical stability was assessed by monitoring normalized absorbance at 660 nm. In PBS, the normalized signals of all formulations remained close to their initial values throughout the 120-min observation period, with only small fluctuations (Figure S9). In the FBS-containing medium, HPAE-0, HPAE-1, HPAE-3, and HPAE-5 showed relatively limited changes, whereas HPAE-7 displayed greater variability and HPAE-9 exhibited a pronounced transient increase in normalized absorbance, reaching approximately fourfold above baseline at 45–60 min before decreasing toward the initial level at 90–120 min (Figure S10). These results indicate greater optical instability or protein-associated changes for the higher-feed formulations, particularly HPAE-9. Because absorbance rather than particle size was monitored, these data were interpreted as short-term optical stability results and not as definitive evidence of serum or plasma colloidal stability. Overall, HPAE-3 combined detectable NIR-I fluorescence with high DNA entrapment, a comparatively low DLS PDI, positive zeta potential, and retention of nanoscale morphology. On this basis, HPAE-3 was selected as the principal modified formulation for subsequent in vitro and in vivo comparisons. Formulation- and Cell Type-Dependent Reporter-Gene Expression in vitro The ability of the HPAE-based NPs to deliver the gWiz-GFP reporter plasmid was initially evaluated in HEK293T cells at polymer-to-DNA mass ratios of 10:1, 20:1, 30:1, and 40:1. GFP fluorescence was observed in all formulation groups, but the intensity varied according to both the formulation and the polymer-to-DNA ratio (Figure 3A).Image-based integrated-density analysis showed that HPAE-0, HPAE-1, and HPAE-3 produced their highest or near-highest reporter signals at a mass ratio of 20:1 (Figure 3B). At this ratio, the integrated-density values were 207.26 for HPAE-0, 201.96 for HPAE-1, and 202.62 for HPAE-3. The corresponding values for HPAE-5, HPAE-7, and HPAE-9 were 195.56, 167.83, and 180.93, respectively. HPAE-3 therefore retained reporter-expression intensity close to that of HPAE-0 and HPAE-1 at 20:1. At the 10:1 ratio, the integrated-density value for HPAE-3 was 150.42, compared with 161.04 for HPAE-0 and 161.23 for HPAE-1. Increasing the ratio from 10:1 to 20:1 enhanced the reporter signal for these three formulations. By contrast, the higher-feed formulations showed less consistent ratio-dependent improvement. HPAE-9 produced the lowest signal at 10:1, with an integrated-density value of 108.13, and remained below HPAE-0 and HPAE-3 at the other ratios. These data indicate that increasing ICG-NHS modification beyond the HPAE-3 level did not improve GFP reporter expression.CCK-8 analysis demonstrated that cell viability was influenced by both formulation and polymer-to-DNA ratio (Figure 3C). At lower polymer ratios, most formulations retained comparatively high viability. As the ratio increased, viability generally declined, with the most marked reductions observed at 40:1. At this ratio, the mean viability of HPAE-5, HPAE-7, and HPAE-9 decreased to approximately 30–45%, whereas HPAE-0 retained a higher mean viability. HPAE-3 showed an intermediate profile, maintaining comparatively strong GFP expression at 20:1 while avoiding the pronounced loss of viability observed with the higher-feed formulations at elevated polymer ratios. The 20:1 mass ratio was therefore selected for subsequent experiments because it provided a practical balance between GFP reporter expression and cell viability. This selection was based on the combined transfection and CCK-8 results and was not intended to indicate that 20:1 was optimal for every cell type or experimental setting. In Vitro Toxicity of HPAE-ICG NPs Fluorescent Nanoparticles (A) Observation of the cellular transfection of HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, and HPAE-9 fluorescent nanomaterials in 293T cells was conducted using fluorescence microscopy. (B) Heatmap showing green fluorescent protein (GFP) expression in 293T cells 48 hours post-transfection with complexes based on HPAE-0 to HPAE-9 at different polymer/DNA weight ratios (w/w). (C) Cell viability of 293T cells 48 hours post-transfection with complexes based on HPAE-0 to HPAE-9 at varying polymer/DNA ratios. To address the limited representativeness of HEK293T cells, reporter-gene expression was additionally evaluated in RAW264.7 macrophage-like cells, MLE-12 lung epithelial cells, and L-02 cells. In RAW264.7 cells, HPAE-1 and HPAE-3 generated mean GFP fluorescence intensities that were not significantly different from that of HPAE-0, whereas HPAE-5, HPAE-7, and HPAE-9 showed substantially lower signals (P<0.0001 versus HPAE-0; Figure S12). A similar pattern was observed in MLE-12 cells: HPAE-1 and HPAE-3 were not significantly different from HPAE-0, while the three higher-feed formulations showed marked reductions in GFP intensity (P<0.0001; Figure S13). In L-02 cells, HPAE-0 produced the highest GFP fluorescence signal. All ICG-modified formulations showed lower reporter-expression intensity than HPAE-0, with the comparisons reaching P<0.0001 (Figure S14). Accordingly, the effect of ICG modification was cell-type dependent: HPAE-3 retained reporter expression comparable to HPAE-0 in RAW264.7 and MLE-12 cells but not in L-02 cells. The fluorescence measurements in these experiments represent image-based reporter-expression intensity and not the percentage of GFP-positive cells or a flow-cytometry-derived transfection efficiency. Moreover, cytocompatibility in RAW264.7, MLE-12, and L-02 cells was not directly evaluated by CCK-8 in the present experiments. The additional cell-line results therefore extend the reporter-expression assessment beyond HEK293T cells but do not establish comparable viability across all tested cell types. HPAE-3 Retained Functional Luciferase Reporter Expression in vivo The experimental workflow for the in vivo reporter-gene study is shown in Figure 4A. Female BALB/c mice received naked firefly luciferase plasmid or plasmid complexed with HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, or HPAE-9. Whole-animal bioluminescence was assessed 6 h after administration. Naked plasmid generated only weak and spatially limited bioluminescence, whereas all polymer-complexed groups produced detectable signals (Figure 4B). HPAE-0 and HPAE-3 generated the strongest whole-animal signals, with mean intensities of approximately 2.0×104 p/s/cm2/sr (Figure 4C). The HPAE-3 signal was not significantly different from that of HPAE-0. HPAE-1 showed a moderate reduction relative to HPAE-0, while HPAE-5 produced a further reduction. HPAE-7, HPAE-9, and naked DNA generated the lowest signals, with the differences from HPAE-0 reaching the higher levels of statistical significance indicated in Figure 4C. These data demonstrate that moderate ICG-NHS modification, represented by HPAE-3, retained whole-animal functional reporter expression at 6 h. In contrast, the progressive decline observed with HPAE-5, HPAE-7, and HPAE-9 was consistent with the in vitro finding that higher ICG-NHS feed levels were associated with reduced GFP reporter expression. In vivo luciferase reporter-gene expression mediated by HPAE/DNA polyplexes. BALB/c mice were intravenously administered with HPAE/DNA polyplexes prepared at a polymer-to-DNA mass ratio of 20:1, followed by bioluminescence imaging and biosafety evaluation. (A) Experimental timeline for in vivo reporter-gene expression and preliminary safety assessment. (B) Representative whole-body bioluminescence images at 6 h after administration. (C) Quantification of whole-body bioluminescence intensity at 6 h. (D) Representative ex vivo bioluminescence images of major organs (heart, liver, spleen, lung, and kidney) collected after administration. (E) Quantitative analysis of organ-associated bioluminescence signals. The detected bioluminescence represents functional luciferase reporter-gene expression following plasmid delivery and expression, rather than direct measurement of nanoparticle distribution. Data are presented as mean ± SD (n = 3). Statistical analysis was performed using one-way ANOVA followed by Dunnett’s multiple-comparisons test. ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Ex vivo organ imaging showed that luciferase activity was not uniformly distributed among the collected organs (Figure 4D). Signals were generally low in the heart and kidneys and were concentrated primarily in the liver, spleen, and lungs. Quantitative analysis confirmed substantial formulation-dependent differences within these organs (Figure 4E). HPAE-3 produced the strongest hepatic reporter signal among the tested formulations, with a mean value of approximately 3.7×104 p/s/cm2/sr. Detectable hepatic expression was also observed in the HPAE-0, HPAE-5, and HPAE-7 groups, but at lower intensities. In the spleen, HPAE-3 again produced a comparatively strong signal, whereas HPAE-5, HPAE-7, HPAE-9, and naked DNA showed limited expression. Pulmonary luciferase expression was prominent in the HPAE-1, HPAE-3, and HPAE-5 groups, with HPAE-3 and HPAE-5 producing mean signals of approximately 2.8–3.0 × 104 p/s/cm2/sr. By comparison, pulmonary signals in HPAE-7, HPAE-9, and naked-DNA groups were markedly lower. The organ-level data indicate that HPAE-3 supported functional reporter expression in the liver, spleen, and lungs after systemic administration. However, the pattern was not identical across all formulations, and the reporter signal should not be interpreted as the physical biodistribution of the polymer. Bioluminescence requires successful cellular uptake, intracellular plasmid processing, transcription, and translation; therefore, it identifies sites of functional reporter expression rather than the location of intact nanoparticles. The animal experiments included three mice per group and evaluated reporter expression at a single 6-h time point. Consequently, the results were treated as exploratory. They demonstrate retained early reporter expression with HPAE-3 but do not establish statistical equivalence with HPAE-0, the duration of expression, transgene kinetics, or therapeutic efficacy. Time-Resolved NIR-I Imaging of ICG-Associated Signals NIR-I fluorescence imaging was performed independently of the luciferase reporter assay to characterize the organ-associated distribution and temporal persistence of ICG-containing species. At 6 h, ex vivo imaging showed near-background signals in the HPAE-0 group, whereas HPAE-3-associated fluorescence was detected predominantly in the liver, with lower signals in the lungs and spleen and minimal fluorescence in the heart and kidneys (Figure 5A and Figure S15). Quantitative analysis confirmed that the hepatic fluorescence intensity of HPAE-3 was approximately 2×1011 p/s/cm2/sr and exceeded the corresponding splenic and pulmonary signals (Figure 5B). The ICG-NHS control produced substantially stronger fluorescence and a broader organ-associated pattern, including pronounced signals in the liver, lungs, and kidneys and detectable fluorescence in the heart and spleen. Because ICG-NHS is a chemically reactive activated ester, its behavior should not be regarded as equivalent to that of conventional non-activated free ICG. Whole-animal imaging at 0.5, 6, 24, 48, and 72 h further demonstrated clear formulation- and time-dependent differences in fluorescence magnitude (Figure 5C and Figure S16). Quantification showed that ICG-NHS generated the highest signal throughout the observation period, whereas the HPAE–ICG formulations produced lower signals, generally in the 109–1010 p/s/cm2/sr range (Figure 5D). Among the conjugated formulations, HPAE-7 generally showed the strongest whole-animal fluorescence, followed by HPAE-5 and HPAE-9; HPAE-3 produced a lower but readily detectable signal, HPAE-1 was weaker, and HPAE-0 remained close to background. All HPAE–ICG formulations exhibited a pronounced decrease between 0.5 and 6 h, followed by a more gradual decline from 6 to 72 h. These curves represent time-dependent changes in ICG-associated fluorescence and should not be interpreted as blood pharmacokinetic profiles or clearance kinetics of intact nanoparticles. Time-resolved in vivo NIR-I fluorescence and ex vivo organ-associated signals. (A) Representative ex vivo NIR-I fluorescence images of the heart, liver, spleen, lungs, and kidneys collected 6 h after administration of HPAE-0, ICG-NHS, or HPAE-3. (B) Quantification of organ fluorescence intensity at 6 h. (C) Representative whole-animal NIR-I fluorescence images obtained at 0.5, 6, 24, 48, and 72 h after administration of HPAE-0, ICG-NHS, or HPAE-3. (D) Quantification of whole-body fluorescence intensity over 72 h for HPAE-0, HPAE-1, HPAE-3, HPAE-5, HPAE-7, HPAE-9, and ICG-NHS. The curves indicate time-dependent changes in ICG-associated fluorescence and do not represent blood pharmacokinetic profiles or intact-nanoparticle clearance. (E) Representative ex vivo NIR-I fluorescence images of the major organs collected 72 h after administration of HPAE-0, ICG-NHS, or HPAE-3. (F) Quantification of organ fluorescence intensity at 72 h. Data are presented as mean ± SD (n = 3 mice per group). Statistical comparisons in panels B and F were performed using two-way ANOVA followed by Šídák’s multiple-comparisons test. ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001. At 72 h, ex vivo imaging demonstrated that residual fluorescence from the HPAE–ICG formulations was detected predominantly in the liver, with comparatively weak signals in the heart, spleen, lungs, and kidneys (Figure 5E and Figure S17). Quantitative analysis showed that HPAE-7 retained the strongest hepatic signal among the conjugates, followed by HPAE-5, HPAE-9, and HPAE-3, whereas the ICG-NHS control maintained pronounced fluorescence in the liver and lungs together with substantial renal fluorescence (Figure 5F). The liver-, spleen-, and lung-associated HPAE-3 signals observed at 6 h showed partial spatial correspondence with the organs exhibiting luciferase reporter expression; however, the two imaging modalities represent distinct biological endpoints. Bioluminescence reflects functional reporter-gene expression, whereas NIR-I fluorescence indicates the presence of ICG-containing species that may include intact HPAE–ICG conjugates, polymer fragments, protein-associated products, or other degradation products. The present data therefore do not demonstrate synchronous transport of the polymer and plasmid DNA or direct tracking of intact HPAE–ICG nanoparticles. Moreover, because the formulations contained different amounts of incorporated ICG and were not administered at matched ICG-equivalent doses, differences in absolute fluorescence intensity reflect both fluorophore content and formulation-dependent in vivo behavior and cannot be attributed exclusively to organ accumulation or clearance. Overall, although HPAE-3 did not generate the strongest fluorescence among the conjugates, it provided the most balanced profile when fluorescence visibility, reporter-gene expression, particle dispersity, and cytocompatibility were considered together. Preliminary in vivo Tolerability Acute tissue tolerability was evaluated by H&E staining of the heart, liver, spleen, lungs, and kidneys at 72 h. Compared with the control group, tissue sections from mice receiving HPAE-0 or HPAE-3 retained the expected overall architecture (Figure 6A). The myocardium showed no overt necrotic lesions or extensive inflammatory-cell infiltration. Hepatic sections retained recognizable lobular organization without widespread hepatocellular degeneration or confluent necrosis. Splenic white- and red-pulp structures remained identifiable, and the renal cortex showed preserved glomerular and tubular morphology. Lung sections showed open alveolar spaces without widespread hemorrhage or dense inflammatory consolidation. Preliminary in vivo tolerability evaluation of HPAE-ICG nanoparticles. Representative H&E staining images of major organs (heart, liver, spleen, lung, and kidney) collected from control, HPAE-0, and HPAE-3 groups at 72 h after intravenous administration showed no obvious histopathological abnormalities (A). Serum biochemical parameters, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), uric acid (UA), and creatinine (CREA), were measured after administration of different HPAE formulations and ICG-NHS control (B–F). The red dashed lines indicate the upper limits of the corresponding reference ranges for each biochemical parameter. All measured values remained within the indicated reference ranges under the tested conditions. Data are presented as mean ± SD (n = 3 mice per group). The supplementary histological evaluation of HPAE-1, HPAE-5, HPAE-7, and HPAE-9 similarly showed no obvious acute treatment-associated lesions in the examined organs (Figure S18). No marked formulation-dependent increase in inflammatory infiltration, edema, hemorrhage, tissue degeneration, or necrosis was identified by qualitative examination. These findings support the absence of overt acute histopathological injury at the tested dose but do not exclude subtle cellular, immunological, or molecular alterations. Serum biochemical parameters were evaluated as preliminary indicators of hepatic and renal function (Figure 6B–F). Mean ALT concentrations were approximately 37–65 U/L across the polymer groups and remained below the upper reference value of 96.47 U/L. Mean AST concentrations were approximately 54–113 U/L, within the displayed reference interval of 36.31–235.48 U/L. BUN values were approximately 22–28 mg/dL and remained within the stated reference interval of 10.81–34.74 mg/dL. Mean uric acid values showed greater numerical variability, ranging from approximately 60 to 210 μmol/L, but remained within the displayed interval of 44.42–224.77 μmol/L. Mean creatinine concentrations were approximately 20–31 μmol/L, substantially below the displayed upper reference value of 85.09 μmol/L. No treatment group showed a coordinated elevation across multiple hepatic or renal markers. Because the animal group size was small and no statistically significant comparisons were indicated in Figure 6B–F, the biochemical findings were interpreted descriptively rather than as definitive evidence of equivalence between groups. Body weight was monitored for 30 days in control mice and mice receiving HPAE-3 NPs (Figure S19). Both groups showed progressive weight gain during the observation period. The control group increased from approximately 19 g to 23.5 g, whereas the HPAE-3 group increased from approximately 20 g to 23.8 g. Their trajectories remained broadly parallel, and no persistent loss of body weight was observed in the HPAE-3 group. Collectively, the preserved histological appearance, biochemical values within the displayed reference ranges, and continued body-weight gain support preliminary tolerability of the tested HPAE-based formulations after a single administration. However, these findings do not establish comprehensive systemic biosafety. The small sample size, single-dose design, limited biochemical panel, use of healthy animals, and absence of dedicated hemolysis, coagulation, complement-activation, immunogenicity, repeated-dose, and long-term organ-toxicity assessments require the results to be interpreted cautiously. Discussion In this study, covalent ICG modification was used to introduce NIR-I fluorescence into an HPAE-based gene-delivery system. Rather than establishing a therapeutic or actively targeted nanoplatform, the present work examined whether fluorescent labeling could be introduced while preserving the principal physicochemical and gene-delivery properties of HPAE. Previous HPAE studies have primarily optimized polymer branching, end-group chemistry, backbone composition, and charge distribution to improve nucleic-acid condensation and transfection efficiency. By contrast, the present study introduced an additional design constraint: the fluorophore density had to be sufficient for NIR-I visualization without excessively compromising the cationic and assembly-related properties required for DNA delivery. Previous ICG-containing polymeric systems mainly focused on improving dye stability, fluorescence imaging, or photothermal performance, whereas the current study systematically evaluated how increasing ICG-NHS feed volumes affected polymer structure, DNA association, nanoparticle properties, reporter-gene expression, and preliminary in vivo tolerability. Therefore, the principal contribution of this work is not the first use of ICG in a polymeric carrier, but the identification of an intermediate modification level that balanced fluorescence labeling with preservation of HPAE-mediated gene delivery. The formulation codes HPAE-1 to HPAE-9 were defined by the volume of ICG-NHS stock solution added during synthesis and should not be interpreted as molar ratios or experimentally determined substitution degrees. The progressive increase in ICG-associated aromatic signals in the 1H NMR spectra, together with FTIR and optical spectroscopy, supported ICG incorporation into the purified HPAE conjugates. Nevertheless, the NMR analysis was semi-quantitative, and the UV–Vis-derived loading estimate for HPAE-3 depended on assumptions regarding polymer molecular weight, the number of potentially reactive amino groups, and the extinction coefficient of polymer-associated ICG. Because the local polymer environment can alter the optical properties of cyanine dyes, the calculated value should be regarded as an apparent ICG-NHS-equivalent loading estimate rather than an absolute grafting degree. The broadly overlapping GPC profiles suggested that ICG modification did not produce a pronounced shift in the overall chromatographic behavior of HPAE, although the broad molecular-weight distributions indicated that the polymers were not molecularly uniform. This observation is consistent with previous HPAE studies showing that average molecular weight alone does not fully describe gene-delivery performance, because branching topology, end-group chemistry, charge distribution, and degradation behavior also contribute substantially. A recent direct comparison of highly branched and linear PBAEs similarly demonstrated that polymer topology and end-group hydrophilicity jointly govern DNA retention, degradation kinetics, transfection, and biocompatibility.17 However, the current GPC data do not resolve subtle differences in branch distribution or polymer conformation and should not be interpreted as evidence that all HPAE–ICG conjugates were structurally identical. The purified conjugates retained an absorption maximum near 793 nm and an emission maximum near 834 nm, confirming that the incorporated ICG remained fluorescent within the NIR-I region. HPAE-3 exhibited the highest emission intensity under identical measurement conditions, whereas further increases in ICG-NHS feed volume did not proportionally enhance fluorescence. Similar nonlinear optical behavior has been reported for ICG incorporated into polymeric micelles and other nanostructures, in which the local dye concentration, molecular packing, polarity, and dye–dye interactions can strongly influence fluorescence output. Rodriguez et al showed that polymeric encapsulation can stabilize ICG, whereas Bahmani et al demonstrated that nanoencapsulation and PEGylation modify both its optical behavior and in vivo distribution.12,13 In the present system, increasing the local ICG density may have promoted self-association or aggregation-associated quenching, while changes in polymer conformation and hydrophobic association may also have contributed. These mechanisms were not measured directly and therefore remain possible explanations rather than established causes. Nevertheless, the findings clearly indicate that fluorophore incorporation should be optimized rather than maximized. All formulations retained strong plasmid-DNA association, with approximately 95–100% entrapment at polymer-to-DNA mass ratios of 20:1 or higher. This finding is consistent with previous reports that highly branched PBAEs provide multiple cationic interaction sites and can condense nucleic acids efficiently. Liu et al similarly reported that branching and hydrophobic modification acted together to promote tight DNA condensation, polyplex stability, cellular uptake, and transfection.18 However, the present findings also reveal an important difference: whereas moderate hydrophobic modification can benefit some branched PBAE systems, increasing ICG modification beyond HPAE-3 did not improve reporter expression and was associated with less favorable particle dispersity or cytocompatibility in several higher-feed formulations. Thus, hydrophobic modification does not appear to provide a universally monotonic benefit and must be balanced against preservation of cationic functionality and appropriate nanoparticle assembly. In addition, PicoGreen inaccessibility only indicates reduced accessibility of the fluorescent reagent to DNA; it does not independently demonstrate complete condensation, nuclease resistance, intracellular release, or improved binding relative to HPAE-0. At a polymer-to-DNA mass ratio of 20:1, the HPAE-based NPs exhibited hydrodynamic diameters of approximately 170–290 nm, positive zeta potentials, and formulation-dependent DLS PDI values. HPAE-3 retained a positive surface charge and showed a comparatively low PDI, whereas HPAE-5 and HPAE-7 showed broader particle-size distributions and HPAE-9 displayed the greatest optical variation in the FBS-containing medium. These findings suggest that extensive ICG modification changed several interdependent properties rather than simply reducing the polymer charge. Walkey et al demonstrated that nanoparticle size and surface chemistry affect the composition of adsorbed serum proteins and subsequent macrophage uptake.19 Similarly, Monopoli et al described the protein corona as a biologically acquired interface that can modify the effective identity of nanoparticles after exposure to biological fluids.20 The greater variation observed for the higher-feed formulations in FBS is therefore consistent with formulation-dependent protein association or aggregation, although protein-corona composition was not directly characterized in the present study. The dry-state diameter of HPAE-3 NPs determined by TEM was substantially smaller than the hydrodynamic diameter measured by DLS. This difference is methodologically expected. TEM visualizes particles after deposition and dehydration on a solid support, whereas DLS estimates an apparent hydrodynamic diameter from Brownian diffusion in suspension and includes contributions from the solvated polymer layer and any larger particle populations. Maguire et al,21 similarly emphasized that light-scattering and electron-microscopy methods characterize particles under different physical conditions and should be interpreted as complementary rather than directly interchangeable. Therefore, the approximately 33-nm TEM diameter does not contradict the larger DLS hydrodynamic size. The relatively stable absorbance in PBS but greater variation in the FBS-containing medium, particularly for HPAE-9, may reflect protein adsorption, reversible particle association, changes in light scattering, or alterations in the optical environment of ICG. Because time-resolved particle size, PDI, zeta potential, and protein-corona composition were not measured, these experiments support only short-term optical stability and do not establish long-term serum, plasma, or colloidal stability. The in vitro results demonstrated that reporter-gene expression depended on both the polymer-to-DNA mass ratio and the extent of ICG modification. HPAE-3 retained GFP reporter expression comparable to HPAE-0 at a mass ratio of 20:1 in HEK293T, RAW264.7, and MLE-12 cells, whereas the higher-feed formulations generally produced weaker signals. Similar cell- and formulation-dependent behavior has been reported in previous HPAE studies. Zhou et al,2 showed that branched PBAEs can transfect diverse cell types, but also demonstrated that polymer topology and terminal functionalization substantially affect performance. Liu et al,3 achieved high transfection in adipose-derived stem cells and primary astrocytes only after coordinated optimization of branching, degradability, and guanidinylation. More recently, Qiu et al,5 evaluated HPAE-mediated CFTR plasmid delivery in disease-relevant lung models, further illustrating that performance in therapeutically relevant cells cannot be inferred from a single highly transfectable cell line. Compared with these studies, the present work provides evidence that moderate ICG modification can preserve reporter expression across several cell lines, but it does not establish therapeutic-gene delivery or equivalent cytocompatibility in all tested cells. The decline in reporter expression and viability at higher polymer ratios, particularly for HPAE-5, HPAE-7, and HPAE-9, may reflect increased exposure to cationic polymer, altered effective charge density, increased hydrophobicity, or changes in particle–membrane interactions. Similar structure–activity relationships have been reported for PBAEs, in which relatively small changes in topology, hydrophobicity, end-group chemistry, or ionization behavior can produce large differences in intracellular delivery.17,18 However, endosomal escape, intracellular trafficking, buffering capacity, and polymer degradation were not directly measured in the present study. The reduced reporter expression associated with the higher-feed formulations therefore cannot be attributed to a single mechanism or specifically to an impaired proton-sponge effect. In L-02 cells, all ICG-modified formulations showed lower reporter intensity than HPAE-0. This result reinforces the conclusion that the effect of ICG modification was cell-type dependent. Importantly, L-02 has been reported as a misidentified cell line and should not be considered an authenticated normal human hepatocyte model.22 Accordingly, the present result should be reported only as reporter expression in L-02 cells and should not be extrapolated to primary human hepatocytes. Future studies should include authenticated hepatocyte models and directly measure cellular viability, uptake, intracellular trafficking, and reporter-positive cell percentages in RAW264.7, MLE-12, and liver-relevant cells. Following systemic administration, HPAE-3 retained the ability to mediate functional luciferase expression, with organ-level signals observed mainly in the liver, spleen, and lungs. Previous HPAE studies have demonstrated that structural optimization can influence organ-selective nucleic-acid expression. Yong et al,6 reported organ-selective mRNA delivery after tailoring highly branched PBAE structures, while Kavanagh et al,23 demonstrated ligand-free systemic PBAE-mediated mRNA delivery to pulmonary epithelial and endothelial cell populations using optimized end-capped polymers and detailed cell-type analyses. Compared with those studies, the present work measured luciferase expression at only one early time point and did not identify the transfected cell populations. Therefore, the liver-, spleen-, and lung-associated signals should be interpreted as sites of functional reporter expression rather than evidence of equivalent organ selectivity or active targeting. Bioluminescence and NIR-I fluorescence also represent distinct biological endpoints. Bioluminescence requires successful plasmid delivery into viable cells followed by intracellular processing, transcription, translation, and enzymatic light generation. Fluorescence, by contrast, reports the presence of ICG-containing species. Di et al,24 similarly demonstrated in mRNA lipid nanoparticles that nanoparticle exposure and transgene expression are not linearly correlated because organ-specific transfection efficiency, particle size, administration route, and pharmacokinetics all affect the relationship between carrier distribution and reporter expression. Consequently, the partial spatial correspondence between luciferase expression and HPAE-3-associated fluorescence does not establish that the polymer and plasmid remained physically associated in vivo. The detected NIR-I fluorescence cannot be assumed to originate exclusively from intact HPAE–ICG NPs. Álamo et al,7 demonstrated that fluorescent-dye labeling can alter the apparent biodistribution of nanoparticles, whereas Hinz et al,8 showed that the location of the fluorophore within a nanocarrier can substantially affect the resulting biodistribution profile. These findings support cautious interpretation of the present fluorescence data. Hydrolysis of the HPAE backbone, fragmentation of the conjugate, binding of ICG-containing products to plasma proteins, and intracellular processing could all contribute to the signal. The reduction in fluorescence over 72 h therefore indicates a decrease, redistribution, or elimination of ICG-associated material but does not establish the clearance rate, blood half-life, or pharmacokinetic behavior of intact HPAE–ICG NPs. Dual labeling of the polymer and DNA, together with chemical quantification in plasma and tissues, would be required to determine whether the carrier and nucleic acid remain associated. The substantially stronger and more broadly distributed signal of the ICG-NHS control should also be interpreted cautiously. ICG-NHS is a chemically reactive activated ester rather than conventional free ICG, and the HPAE–ICG formulations were not administered at matched ICG-equivalent doses. Differences in absolute fluorescence therefore reflect a combination of fluorophore content, optical quenching, protein association, tissue distribution, degradation, and elimination. The current imaging data are consequently suitable for comparing formulation-dependent fluorescence patterns but not for ranking nanoparticle accumulation quantitatively. The observed liver-, spleen-, and lung-associated pattern does not demonstrate active targeting. No organ-specific ligand was incorporated, and no direct comparison with an otherwise identical ligand-targeted or non-targeted formulation was performed. Walkey et al,19 and Monopoli et al,20 showed that surface properties and protein adsorption can modify macrophage recognition, while Gustafson et al,25 emphasized that phagocytic uptake is a major determinant of nanoparticle clearance and organ distribution. Thus, the present distribution is more appropriately described as passive mononuclear-phagocyte-system- or reticuloendothelial-system-associated distribution. Although some ligand-free PBAE formulations can achieve reproducible organ-selective expression through carefully optimized polymer chemistry6,23 the current data do not establish the cellular mechanism or selectivity required to make an equivalent claim. The preliminary tolerability findings were generally favorable. No overt acute histopathological injury or coordinated hepatic or renal biochemical abnormalities were observed after a single administration, and HPAE-3-treated mice continued to gain weight during the 30-day observation period. Previous optimized HPAE systems have also shown favorable in vitro viability and in vivo tolerability.17,23 However, several of those investigations included primary cells, disease-relevant models, longer observation periods, or repeated administration. For example, Kavanagh et al evaluated repeated systemic administration of lung-directed PBAE nanoparticles, whereas the present study included only three female mice per group and a single dose.23 Therefore, the current findings support preliminary tolerability under the tested conditions rather than comprehensive systemic biosafety or clinical readiness. Collectively, these findings suggest that HPAE-3 represented the most balanced formulation among those tested. Its performance did not result from maximal ICG incorporation, fluorescence intensity, or surface charge, but from an intermediate modification level that retained detectable NIR-I fluorescence, strong DNA association, comparatively favorable particle dispersity, and reporter-gene expression. Compared with previous HPAE studies focused solely on maximizing nucleic-acid delivery, the present work demonstrates the additional trade-offs introduced by covalent fluorophore labeling. Compared with earlier ICG-containing polymeric systems developed mainly for imaging or phototherapy, the current study evaluates whether fluorescent functionality can coexist with plasmid delivery. Nevertheless, ICG incorporation was not absolutely quantified for every formulation, fluorescence comparisons were not performed at matched ICG-equivalent doses, serum stability was not assessed by time-resolved DLS, and the animal experiments were exploratory. Future work should therefore quantify the polymer, dye, and DNA independently; characterize protein-corona formation and intracellular trafficking; evaluate reporter-expression kinetics; and investigate hemocompatibility, immune activation, repeated-dose toxicity, and therapeutic efficacy in disease-relevant models. In conclusion, covalent ICG modification generated NIR-I-fluorescent HPAE formulations while preserving DNA association and, at an intermediate modification level, functional reporter-gene delivery. HPAE-3 retained reporter expression comparable to unmodified HPAE in HEK293T, RAW264.7, and MLE-12 cells and supported early luciferase expression following systemic administration. However, the NIR-I signal represented ICG-associated material and did not establish active targeting, intact-particle tracking, or pharmacokinetic clearance. These results support further investigation of HPAE-3 as a fluorescently labeled polymeric gene-delivery formulation while defining the additional mechanistic, pharmacokinetic, hemocompatibility, therapeutic, and long-term safety studies required for translation. Conclusion This study developed a series of highly branched poly(β-amino ester)–indocyanine green nanoparticles (HPAE–ICG NPs) and evaluated their physicochemical properties, reporter-gene delivery, in vivo fluorescence, and preliminary tolerability. Among the formulations, HPAE-3 achieved the most favorable balance of NIR-I fluorescence, DNA complexation, particle dispersity, positive surface charge, and reporter-gene expression. HPAE-3 retained reporter-gene expression comparable to unmodified HPAE in HEK293T, RAW264.7, and MLE-12 cells and mediated luciferase expression in the liver, spleen, and lungs after systemic administration. Time-resolved NIR-I imaging showed early liver- and lung-predominant ICG-associated fluorescence, with a weaker splenic signal, followed by an overall decline and predominantly hepatic residual fluorescence at 72 h. Histopathological, biochemical, and body-weight assessments indicated preliminary tolerability after a single administration. Collectively, these findings establish HPAE-3 as a rationally optimized fluorescent HPAE formulation for future development of image-guided polymeric gene-delivery systems and provide a practical framework for integrating molecular imaging with nonviral nucleic acid delivery. Acknowledgment This work was supported by Scientific Research Foundation for High-level Talents of Anhui University of Science and Technology (2023yjrc52 to Zhonglei He); Research Funds of Joint Research Center for Occupational Medicine and Health of IHM (OMH-2023-02 to Wenxin Wang; OMH-2023-15 to Zhonglei He); Key Science & Technology Project of Anhui Province (202423110050053 to Wenxin Wang); National Natural Science Foundation of China (82471863 to Wenxin Wang); the Fundamental Research Funds of the AUST (2024JBQN0032 to Zhonglei He).And this work is supported by Anhui University of Science and Technology Graduate Innovation Fund led by Hongzhen Zhang[Graduate Student Document No. 24 (2024)]. Declaration of Generative AI in Scientific Writing During the preparation of this manuscript, the authors used ChatGPT (GPT-5.5-mini, OpenAI) as an AI-assisted language tool to improve English language expression, grammar, readability, and overall manuscript organization. The tool was used only for language refinement and editorial assistance; it was not used to generate experimental data, analyze results, interpret scientific findings, or create references.The authors have carefully reviewed and revised all AI-assisted content and confirm that the manuscript represents original work by the authors. The authors take full responsibility for the originality, accuracy, integrity, and scientific validity of the entire manuscript, including the accuracy and completeness of all references and citations. The authors have reviewed the terms of use and policies of the AI tool and confirm that its use is appropriate and compliant for manuscript preparation and publication. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure W.W. is the scientific founder and holds equity in Branca Bunús Ltd and Beijing PolyGene Biotechnology Co., Ltd. W.W. is the inventor on a patent for highly branched cationic polymers useful as nucleic acid delivery vectors for transfecting cells (WIPO Patent Application WO 2021/058491 A1), which has been developed into BrPERfect™ and TransPi™ series transfection reagents. W.W. is the inventor on a patent for nanoparticle compositions for gene therapy (WIPO Patent Application WO 2021/058492 A1). These interests have been fully disclosed to the relevant affiliations, and the approved plans are in place for managing any potential conflicts arising from licensing these patents. The authors report no other conflicts of interest in this work. References