Systematic Evaluation of Physicochemical Properties of PEGylated Gold Nanorods Reveals Cell‐Specific Uptake Determinants in Human Immune Cell Subsets
Tista Roy Chaudhuri, Helene Giesler, Marija Kovacevic Sarmiento, Kim Lamers, Michelle Hechler, Michael Erkelenz, Marvin Haferkamp, Ronja Schirrmann, Milen Nachev, Rebeka Bosnjakovic, Bernd Sures, Sebastian Schlücker, Sven Brandau
TL;DR
This study explores how gold nanorods interact with human immune cells, finding that their size, PEG length, and surface charge determine selective targeting and uptake.
Contribution
The study identifies key physicochemical parameters of AuNR that enable selective targeting of immune cell subsets for nano-immunotherapy.
Findings
Antibody-targeted AuNR selectively bind and induce apoptosis in T-lymphocytes during photothermal therapy.
PMNs exhibit nonspecific uptake of AuNR, driven by their phagocytic activity.
Small AuNR size, large PEG size, and near-neutral surface charge reduce non-targeted uptake by PMNs.
Abstract
Gold nanorods (AuNR) are multifunctional transducers applied in heat‐ablating cancer therapy, bio‐imaging, and controlled drug release. Despite extensive studies of AuNR interactions with cancer cells, the AuNR characteristics that determine their interaction with primary human immune subsets remain poorly characterized. Here, we investigated the effect of AuNR physico‐chemical properties on the binding, uptake, and cell death responses by human T‐lymphocytes and polymorphonuclear neutrophils (PMNs). We demonstrate that antibody targeting with α‐CD3‐AuNR conjugates results in selective binding of AuNR to T‐lymphocytes. Photothermal therapy (PTT) using these AuNR triggered apoptotic pathways and evoked selective cell death in T‐lymphocytes. In contrast, PMN exhibited nonspecific target‐independent uptake, consistent with their inherent phagocytic activity. We evaluated the AuNR size, PEG…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 1
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5- —Else‐Kröner Fresenius Stiftung
- —Deutsche Forschungsgemeinschaft10.13039/501100001659
- —German Cancer Aid
- —Josepha and Charlotte von Siebold grant program of Medical Faculty, University of Duisburg‐Essen
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGold and Silver Nanoparticles Synthesis and Applications · Nanoplatforms for cancer theranostics · Nanocluster Synthesis and Applications
Introduction
1
Gold nanoparticles (AuNP) are chemically inert yet optically active colloids with a characteristic absorption and scattering spectrum in the visible region generated by localized surface plasmon resonance (LSPR) [1]. Both particle size and shape determine position and intensity of the plasmon peaks, and the large specific surface area promotes efficient photothermal conversion at resonant wavelengths upon irradiation [2]. The shape of the nanorods (AuNR) elicits electron oscillation in both longitudinal and transverse directions, generating a longitudinal plasmon band ranging from visible to the near‐infrared (NIR) region, which can be modulated by controlling the AuNR aspect ratio (length‐to‐diameter). NIR wavelengths enable deep tissue penetration [3], and therefore, AuNR are utilized in the field of biomedicine as heat‐ablating therapeutic agents for photothermal therapy (PTT), and as diagnostic agents in oncology.
If administered intravenously for cancer therapy, nanoparticles not only accumulate passively inside the tumor via the enhanced permeability and retention (EPR) effect [4], but also encounter multiple physiological barriers of delivery beyond those faced by small molecules [5, 6]. Such barriers include components of the mononuclear phagocytic system (MPS), and other phagocytic white blood cells of the innate immune system such as polymorphonuclear neutrophils (PMN) [4]. Although studied much less compared to macrophages, PMN are of special importance in this context because they are the most abundant leukocytes in the circulation of humans and have very high capacity to phagocytose particulate matter [7]. PMN are specialized for the clearance of both biological and nanoparticulate matter, including phagocytic uptake of micron‐scale pathogens (0.5–10 µm) [8] and receptor‐mediated endocytosis of nanoparticles (1–500 nm) [9]. Therefore, PMN substantially contribute to nanoparticle clearance and reduced circulation times, which ultimately hinder EPR‐driven delivery. AuNR size and surface chemistry drive internalization by PMN. Citrate‐ and cetyltrimethylammonium bromide (CTAB)‐capped AuNR are rapidly internalized by all phagocytic immune cells, including PMN, macrophages, and monocytes [10]. In addition, the metal surface of AuNR chelates blood proteins by forming a “protein corona”, which is recognized by PMN to internalize large cargoes of AuNR [11, 12]. Therefore, bypassing AuNR uptake by PMN is likely to improve the circulation half‐life of AuNR, enhances their EPR‐mediated delivery, and reduces AuNR‐mediated toxicity.
Steric stabilization of AuNR is a common approach to reduce non‐specific uptake by PMNs and phagocytes, but small molecular‐weight polyethylene glycol (PEG) provides insufficient coverage that results in poor tumor delivery and low temperature enhancements, which elicits only fever range hyperthermia and moderate activation of adaptive immunity [13, 14]. Alternately, antibody functionalization with surface ligands promotes higher internalization, enhances selective uptake of AuNR, and mediates endocytic internalization preventing receptor recycling to the cell surface [15]. Therefore, active antibody‐mediated AuNR targeting of tumor or immune cells may be a viable immunotherapeutic approach to trigger an antitumor immunity and is explored in addition to classical EPR‐mediated delivery [16, 17]. However, nonspecific uptake by PMN hinders targeted delivery, and our hypothesis was that reduction of such nonspecific uptake may promote targeted AuNR delivery. Therefore, in this study, we aimed to identify the key parameters that determine nonspecific uptake of AuNR by PMN to establish conditions that would facilitate the future design of AuNR‐based immunotherapeutics.
Results and Discussion
2
To delineate cell type‐specific uptake dynamics, we compared how AuNR of varying sizes, surface charges, and surface functionalization differentially interact with primary human immune cells, both phagocytic and non‐phagocytic, in well‐defined in vitro model systems. Experiments involved incubating functionalized and sterically stabilized AuNR with isolated primary human PMN, T‐lymphocytes, and whole blood samples at 37°C to assess AuNR uptake over time.
Targeted AuNR Bind to T Cells Specifically, but PMNs Show a Size‐Dependent Loss of Targeted Uptake
2.1
We tested whether surface functionalization with monoclonal antibodies (mAb) against phagocytic PMNs vs. non‐phagocytic T‐lymphocytes would lead to immune cell‐selective uptake. Because particle size is a key determinant that influences AuNP interaction with immune cells, we first investigated the effect of size on the specificity of antibody‐mediated binding in both immune cell types (Figure 1a,b). Multiple sizes of AuNR (15–60 nm) were functionalized with an anti‐CD3 (α‐CD3) antibody for targeting T‐lymphocytes and against the myeloid cell surface protein CD11b (α‐CD11b) for targeting PMN. All particles were sterically stabilized with 5 kDa NH_2_‐terminated PEG and incubated with rhodamine B isothiocyanate (RITC) for fluorescent labelling. Additionally, bovine serum albumin (BSA) was chelated onto the AuNR surface to minimize protein corona formation, and thereby, reduce non‐specific uptake by cells. As negative controls, particles were functionalized with IgG2a and IgG2b antibody isotype controls of α‐CD3 and α‐CD11b, respectively. Purified and isolated immune cells or whole blood samples were treated with the AuNR formulations for 30 min, and both surface bound and internalized AuNR intensity were quantified by flow cytometry.
Targeted AuNR bind to T cells specifically, but PMN show a size‐dependent loss of specific uptake. (a) T‐lymphocytes and (b) PMN isolated from blood of healthy donors via density gradients (left of dashed line) or analyzed in whole venous blood (right of dashed line) were treated with different sizes of RITC‐labeled AuNR (15–60 nm) conjugated with antibody‐based surface ligands—solid black circle: BSA; full blue circle: αCD3 mAb; blank blue circle: isotype control IgG2a for αCD3 mAb; full red circle: αCD11b mAb; empty red circle: rIgG2b isotype control for αCD11b mAb (10,000 AuNR /cell for 30 min). AuNR uptake was quantified by flow cytometry. n = 5 (c). The specificity of binding (ratio of αCD11‐AuNR/isotype control uptake) in isolated PMN and (d) in PMN from whole blood. (e) Representative confocal micrographs of isolated PMN (blue) and T‐lymphocytes (pink) mixed in a 1:1 ratio and treated with (from left to right): PBS, AuNR conjugates (yellow). All statistical analyses were done with unpaired t‐test with Welch's correction. The statistical significance is defined as p > 0.05 not significant (ns), * p < 0.05; ** p < 0.01; *** p < 0.001, **** p < 0.0001. n indicates the number of experimental repetitions, and each repetition performed with an independent donor. AuNR = gold nanorod; BSA = bovine serum albumin; PBS = phosphate‐buffered saline; PMN = polymorphonuclear neutrophil; RITC = rhodamine B isothiocyanate.
The α‐CD3‐AuNR binding or uptake by T‐lymphocytes was significantly higher compared to AuNR‐BSA and IgG2a‐AuNR controls in both isolated T‐lymphocytes and T‐lymphocytes present in whole blood, and this difference remained consistent over all sizes of AuNR (Figure 1a). Therefore, it was feasible to target T‐lymphocytes using α‐CD3‐AuNR ligand with high target specificity, irrespective of AuNR size. We also investigated the binding affinity of functionalized gold nanospheres (AuNSPH) with T‐lymphocytes to test whether nanoparticle shape influenced particle–cell interaction. In general, AuNR bound to T‐lymphocytes with greater affinity than AuNSPH (Figure S1a); like AuNR, all sizes of α‐CD3‐AuNSPH bound with greater avidity compared to BSA‐AuNSPH and IgG2a‐AuNSPH controls. Therefore, antibody functionalization facilitates antigen‐specific targeting of T‐lymphocytes with gold nanoparticles, irrespective of nanoparticle size and shape.
Next, we adapted the same test system to interrogate AuNR uptake by phagocytic PMN. The AuNR–antibody conjugates elicited higher fluorescence intensity in PMN compared to CD3+ T‐lymphocytes, suggesting that PMN exhibited an overall higher AuNR uptake. In contrast to T‐lymphocytes, PMN exhibited robust signal from internalizing isotype control‐AuNR. Only α‐CD11b‐AuNR measuring 15–30 nm were internalized with higher specificity than their isotype counterparts IgG2b‐AuNR and non‐functionalized BSA‐AuNR, suggesting that small‐sized AuNR could be adapted for antibody‐specific selective uptake by PMN (Figure 1b). It is interesting to note that the 24 and 30 nm α‐CD11b‐AuNR showed PMN‐selective uptake upon treatment in the whole blood system but were nonspecifically internalized by PMN isolated via density gradients. We attributed this effect to the process of PMN isolation over density gradients which may lead to cell biological changes and activation that enhances the phagocytic activity and uptake of AuNR [18].
The ratio of antigen‐specific uptake and isotype control signals of the AuNR conjugates confirmed that moderate levels of antigen‐specific targeting of PMN can only be achieved with small‐sized AuNR (Figure 1c,d). Confocal micrographs of mixed PMN and peripheral blood mononuclear cells (PBMCs) treated with functionalized 24 nm AuNR confirmed that PMN internalized all the AuNR formulations, whereas T‐lymphocytes only bound to α‐CD3‐AuNR (Figure 1e). AuNSPH were functionalized similarly, and both targeted AuNSPH and their isotype controls showed equal and higher nonselective internalization compared to BSA‐AuNSPH uptake by PMNs (Figure S1b). We conclude that surface receptor‐mediated binding provides a feasible strategy for antigen specific targeting of T‐lymphocytes. However, PMN internalize all large‐sized particles nonspecifically irrespective of ligands because of intrinsic phagocytosis, and moderate levels of antigen‐specific targeting of phagocytic primary human PMNs is achievable only with sub‐40 nm AuNR. Therefore, further physico‐chemical optimization of particles is necessary for enhancing selective antibody‐mediated internalization of AuNR in PMN.
Antibody‐Mediated Uptake of Functionalized AuNR Mediates Selective Cell Death in T Cells
2.2
PTT using AuNR leverages its NIR surface plasmon resonance for absorbing NIR light and converting this energy into heat [3]. Given that AuNR‐antibody conjugates target T‐lymphocytes specifically, we tested whether PTT induced selective cell death in T‐lymphocytes.
T‐lymphocytes were incubated with 24 nm α‐CD3‐AuNR and IgG2a‐controls, exposed to NIR laser radiation (808 nm) with 5 W power for 5 min, and rested for 1 and 24 h to assess cell death (Figure 2a). We observed cell death induction in the majority of the population after a 24‐hour rest period only, which suggested a time‐dependent cytotoxic mechanism. This 24‐hour time point was subsequently selected for comparative cell death assessments in T‐lymphocytes. The temperature increase during PTT‐irradiation with α‐CD3‐AuNR was significantly higher compared to all controls (Figure 2b), indicating T‐lymphocyte‐specific thermal effects. To establish the duration of irradiation optimal for cell death induction, AuNR‐bound cells were irradiated for 5, 10, and 15 min and rested for 24 h for cell death to manifest fully in the population (Figure 2c). PTT induced significant cell death within 5 min of irradiation in 75% of the population, and therefore, this duration of irradiation was adopted in subsequent experiments.
Antibody‐mediated targeting with AuNR and PTT mediates selective cell death in T‐lymphocytes. Isolated T cells were treated with RITC‐labeled 24 nm AuNR conjugated with different antibody‐based ligands for 30 min, unbound AuNR were removed by centrifugation, and T‐lymphocytes were irradiated with a 5 W NIR laser at 808 nm. The groups are blank black circle: PBS, no laser; solid black circle: PBS + laser irradiation; blank blue circle: mIgG2a isotype control for αCD3 mAb; full blue circle: αCD3 mAb. (a) The percentage of live T‐lymphocytes after 5 min of irradiation assessed by Annexin V‐7AAD staining using flow cytometry after 1 and 24 h of resting period. (b) The bulk temperature in the tissue culture wells was measured with an infrared camera (Optris Pix connect), and the temperature increase (Δ temperature) was plotted against time (in seconds). n = 3. (c) T‐lymphocytes were irradiated for 5 min and then rested for 24 h at 37°C postirradiation. Apoptosis‐related changes in cell membrane topology were assessed by Annexin V‐7AAD staining using flow cytometry. (d) Expression of active caspases 5–15 min after laser irradiation in T‐lymphocytes measured by flow cytometry. n = 4. All statistical analyses were done with unpaired t‐test with Welch's correction. The statistical significance is defined as p > 0.05 not significant (ns), * p < 0.05; ** p < 0.01; *** p < 0.001, **** p < 0.0001. n indicates the number of experimental repetitions, and each performed with independent donors. AuNR = gold nanorod; NIR = near‐infrared; PBS = phosphate‐buffered saline; PTT = photothermal therapy; RITC = rhodamine B isothiocyanate.
The delayed cytotoxic response 24 h after irradiation suggested that PTT mediated a regulated cell death mechanism rather than necrosis in T‐lymphocytes. Therefore, to test the hypothesis that PTT‐mediated heat ablation triggered apoptotic cell death in T‐lymphocytes, we quantified caspases activated at the beginning of death processes [19] during irradiation for 5, 10, and 15 min. T‐lymphocytes treated with α‐CD3‐AuNR showed a gradual but significant induction of caspases during PTT compared to control groups (p < 0.05, Figure 2d), suggesting that α‐CD3‐AuNR triggered an apoptotic mode of cell death, contrary to established consensus that heat ablation triggers necrotic cell death [20, 21]. Given T‐lymphocytes are non‐phagocytic, we surmise that the heat generated by membrane‐tethered ligand‐bound AuNR enhanced membrane porosity that triggered apoptosis rather than necrosis. However, further studies are needed to elucidate the mechanism of cell death observed in human primary T‐lymphocytes.
These findings suggest that PTT with α‐CD3‐AuNR mediates selective lymphocytic cell death, and therefore, receptor‐targeted AuNR represent a viable strategy for targeting human T‐lymphocytes. Such an approach would allow to therapeutically eliminate immunosuppressive subsets of T‐lymphocytes such as regulatory T cells (Treg) to induce antitumor immunity. Such an approach may be of special interest for target cells that are “undruggable” by small molecule therapeutics [22]. Thus, our findings serve as a basis for future investigations focused on targeting suppressive T‐lymphocytes in the tumor microenvironment. However, the intrinsic phagocytic ability of PMN drives nonspecific uptake of mAb‐functionalized AuNR and precludes PMN from straightforward targeting strategies. Therefore, to bypass nonspecific phagocytosis, and to resolve the contribution of key AuNR physico‐chemical properties to nonspecific phagocytosis, we systematically deconvoluted and evaluated how individual parameters such as AuNR‐size, surface properties and steric hindrance affected binding and PTT‐mediated cell death in primary human PMN.
Surface Charge of AuNR Drives Their Uptake by PMN
2.3
Surface properties including composition of the base material, presence of functional groups such as hydroxy and carboxy moieties, and surface charge influence nanoparticle interaction and their subsequent endocytosis in cells. Functional groups such as –OH or –COOH are biologically recognized and mediate cellular adhesion to enhance endocytic uptake, whereas positive charge drastically enhances endocytosis by membrane destabilization [23]. The cationic surface of metal nanoparticles chelates plasma proteins and forms a hard protein “corona” upon intravenous injection, which facilitates their recognition and uptake by phagocytes [24]. To overcome the above effects of metal surface, we synthesized sterically stabilized AuNR with multiple sizes of PEG that creates a protective shield around the nanoparticles and averts protein chelation [25, 26].
We investigated the influence of both positive and negative surface charge on uptake by altering the ratio of positive (NH_2_‐terminal; Figure 3a–c), negative (COOH‐terminal, Figure 3d,e), and neutral (methoxy‐terminal) PEGs (mPEG). Complete mPEG coverage produced AuNR with slight negative charge (zeta potential, ζ = ‐5 mV) because of the negatively charged glycol groups. PMN were incubated with AuNR (30 min, 2, and 4 h) and internalized gold was quantified indirectly by flow cytometry of RITC‐tagged particles, and directly by inductively coupled plasma mass spectrometry (ICP‐MS) or atomic absorption spectroscopy (AAS). All positively charged formulations manifested similar ζ potentials despite incubation with 15%, 50%, and 100% NH_2_‐PEG (Figure 3a) because the reaction required excess PEG, and therefore, the final proportion of NH_2_‐PEG on AuNR surface could not be controlled.
Surface charge drives the passive uptake of non‐targeted AuNR. AuNR with a size of 24 nm were coated with varying ratios of 5kDa NH2‐PEG, mPEG, and COOH‐PEG to produce a spectrum of surface charges, ranging from positive to negative. The AuNR with positive surface potential (NH2‐PEG) were labeled with RITC. Isolated PMN from healthy donors were incubated with PBS (black) and the AuNR formulations for 30 min, 2, and 4 h (25 000 AuNR/cell). The internalization of AuNRs by PMN was quantified using flow cytometry, ICP‐MS, or atomic spectroscopy. (a) Table representing the zeta potentials (ζ) of positively charged AuNR. (b) Representative flow cytometry histograms depicting the uptake of AuNR having 50% (yellow) and 15% (red) NH2‐PEG coverage after 2 and 4 h incubation with PMN. (c) The number of AuNR in each cell quantified using atomic spectroscopy (mPEG: NH2‐PEG ratios: blue‐100:0; green‐50:50; yellow‐85:15; red‐0:100) after 30 min, 2 and 4 h of incubation. n = 3. (d) Table with the ζ‐ potential of negatively‐charged AuNR formulations coated with multiple proportions of mPEG:COOH‐PEG. (e) The number of AuNR in each cell quantified using ICP‐MS (blue‐100:0; green‐75:25; yellow‐50:50; orange‐25:75; red‐0:100) after 30 min, 2 and 4 h of incubation. n = 3. Confocal micrographs of PMN (blue) internalizing Cy5‐labeled AuNR coated with (f) COOH‐PEG and (g) mPEG observed as red punctate intracellular vesicles. Statistical test: ANOVA. The statistical significance is defined as p > 0.05 not significant (ns), * p < 0.05; ** p < 0.01; *** p < 0.001, **** p < 0.0001. n indicates the number of independent donors as experimental replicates. AuNR = gold nanorod; ICP‐MS = ICP‐MS = inductively coupled plasma mass spectrometry; NIR = near‐infrared; PBS = phosphate‐buffered saline; PMN = polymorphonuclear neutrophil; PTT = photothermal therapy; RITC = rhodamine B isothiocyanate.
All RITC‐tagged NH_2_‐PEG covered AuNR exhibited near‐complete internalization by PMN, regardless of NH_2_‐PEG density (Figure 3b). AAS confirmed rapid saturation of gold accumulation by 2 h and minimal additional uptake at 4 h (Figure 3c). This demonstrates that even the moderate positive surface charge is a strong driver of phagocytosis, consistent with reports showing that positively charged particles interact with the negative charge of cellular membranes and destabilizes them and utilizes clathrin‐mediated endocytosis to traffic large cargoes into cells [9]. Additionally, antigen‐presenting and phagocytic immune cells including PMN are capable of constitutive macropinocytosis [27], providing additional transport routes for positive AuNRs into large vacuoles [28, 29].
AuNR formulations with a negative ζ potential (−5 to −20 mV) (Figure 3d) showed significantly reduced uptake compared to the cationic formulations and uptake increased with higher negative charge. This is consistent with published data showing that negatively charged nanoparticles are transported via caveolae‐mediated endocytosis, which achieves lower internalization than clathrin‐mediated endocytosis of positively charged particles [9]. The mPEG‐covered AuNR showed approx. 3‐fold lower uptake than AuNR with higher negative charges and COOH‐PEG coverages of 75% and 100% after 4‐h incubation; additionally, the lowest internalization only marginally exceeded vehicle treatment (Figure 3e). Of note COOH‐PEG‐covered AuNR showed slightly delayed kinetics as compared to NH_2_‐PEG‐AuNR. Confocal micrographs demonstrated that the negative AuNR were internalized via multiple punctate foci within the cell, whereas the “neutral” particles were internalized by a lower number of endocytic vesicles (Figure 3f,g). These observations suggest that the “bio‐active” COOH groups attract a larger protein “corona” that induces antibody‐mediated endocytosis and engages with the cellular membrane to a greater extent than the inert methoxy group.
Data show that both positive and negative surface charge enhanced AuNR uptake by PMN. However, the positive charge affected internalization more profoundly because it triggered high nonspecific uptake by direct engagement with the negatively charged membrane, whereas negative charges induce nonspecific caveolae‐mediated endocytosis [30]. This interpretation is supported by published reports showing that the negative charge of the ‐COOH group bound to cell membrane components for internalization via caveolae‐mediated endocytosis [31]. In contrast, the positively charged AuNR have a greater affinity for the negatively charged cellular membranes and mediated its entry via membrane permeabilization and rapid clathrin‐mediated endocytosis that deposits high cargo of gold into PMN [32]. Based on our findings that the surface charge of AuNR profoundly influence their uptake by PMN, we employed near‐neutral mPEG‐coated AuNR in the subsequent studies to mitigate charge‐driven nonspecific interactions. This approach enabled us to subsequently dissect the distinct individual contributions of particle size and PEG chain length to cellular uptake.
AuNR Size Affects the Uptake by PMN
2.4
Based on our previous findings that AuNR size hindered the specific uptake of antibody coupled AuNR in PMN (Ref. Figure 1), we tested the influence of AuNR size on the intrinsic, passive uptake of both neutral and positively charged particles. PMN were exposed to positively charged AuNR measuring 24 and 50 nm covered with 5kDa NH_2_‐PEG and RITC for multiple durations of time (ζ‐potential = 4.1 and 1.3 mV, respectively; Figure S2a–d). The 50 nm AuNR exhibited substantially stronger uptake signals compared with the smaller 24 nm AuNR and exhibited a further 7‐fold higher intracellular accumulation following a 4hr exposure (Figure S2b–d). These findings reflect the size‐dependent internalization of antibody‐conjugated AuNR in PMN (Ref. Figure 1) and show that the passive uptake of nonconjugated AuNR is also critically governed by particle size.
AuNR Size and PEG‐Size Jointly Drive the Passive Uptake of Non‐Targeted AuNR in Isolated PMN
2.5
Based on previous experiments, we next systematically combined PEG size and AuNR size to investigate potential interdependence of both parameters on the AuNR uptake by PMN. Because a minimum PEG size of 2 kDa is necessary for ensuring reduced protein adsorption and shielding by mononuclear immune cells [33], we selected Cy5‐labeled PEG sizes of 3, 5, and 10 kDa for sterically stabilizing 24, 40 and 60 nm AuNR. We sought to establish the optimal mPEG‐size required to effectively shield AuNR from phagocytic uptake by PMN. The ζ potentials of 24 nm particles were documented and are shown for 24 nm particles in Figure S3a.
Quantification of the fluorescence signal represented by flow cytometry histograms (Figure 4a) revealed significant stepwise size‐dependent increase in AuNR accumulation when comparing rods with similar PEG coverage (p < 0.001; Figure 4b). Quantification by ICP‐MS measuring intracellular gold from AuNR revealed comparable uptake levels between 24 and 40 nm AuNR, whereas 60 nm AuNR exhibited a significant 2.5‐fold increase in cellular internalization compared to the smaller AuNR (Figure 4c–e; p < 0.05). Such results suggest an AuNR size threshold of ~40 nm, above which the AuNR‐PMN interactions shifted markedly towards enhanced phagocytosis. Steric stabilization with both 5 kDa‐ and 10 kDa‐AuNR significantly lowered internalization, and the uptake of 3 kDa‐PEG‐AuNR remained consistently elevated across all AuNR sizes compared to both 5 and 10 kDa‐AuNR (p < 0.001; Figure 4b–e). The 10 kDa PEG AuNR consistently exhibited a ~3–4 fold reduced phagocytic uptake than 3 kDa‐AuNR. Notably, the 24 nm 10 kDa PEG‐AuNR only showed increased uptake over controls as late as 4 h after PMN challenge (p < 0.05; Figure 4c). This delayed onset of measurable uptake suggested that high steric stabilization with 10 kDa PEG substantially reduced the kinetics of uptake, and may serve as the critical mPEG size needed to impede phagocytosis by PMN. When translated to the in vivo setting, these data may be interpreted in a way that 10kDa‐mPEG‐24 nm‐AuNR will be largely invisible to the human immune system, a hypothesis that should be tested in future translational research.
*AuNR size and PEG‐size drive the passive uptake of non‐targeted AuNR in isolated PMN. AuNR measuring 20, 40 and 60 nm were sterically stabilized with Cy5‐mPEG having molecular weights of 3, 5, and 10 kDa (red, yellow and blue, respectively). Isolated PMN (n = 3) were incubated with PBS (black) and the AuNR formulations for 30 min, 2 and 4 h at saturating doses (25,000 AuNR/cell). The uptake was quantified using flow cytometry and ICP‐MS. (a) Representative fluorescence histograms showing the cellular uptake of the AuNR formulations by PMN from one representative donor. (b) Quantification of Cy5 fluorescence in PMN following a 2‐h incubation period of 24, 40 and 60 nm Cy5‐AuNR. Inset: Quantification of 24 nm Cy5‐AuNR. n= 5. The internalized number of (c) 24 nm, (d) 40 nm, and (e) 60 nm AuNR particles with different PEG coverage (3 kDa ‐ red, 5 kDa ‐yellow and 10 kDa ‐blue; PBS control ‐black) quantified by ICP‐MS. (f) Confocal micrographs revealed that the smaller 24 and 40 nm AuNR were internalized via fewer number of vesicles compared to the larger 60 nm AuNR (all AuNR covered with 5 kDa mPEG). Statistical test: ANOVA and unpaired t‐test with Welch's correction (for panel d only, 2 and 4 h incubation time). The statistical significance is defined as p > 0.05 not significant (ns), * p < 0.05; *, p < 0.01; *** p < 0.001, **** p < 0.0001. N indicates the number of independent donors as experimental replicates. AuNR = gold nanorod; ICP‐MS = inductively coupled plasma mass spectrometry; PBS = phosphate buffered saline; PMN = polymorphonuclear neutrophil.
High‐resolution confocal microscopy provided further cell biological insight on size‐dependent uptake mechanisms, whereas 24 and 40 nm AuNR localized into sparse, punctate vesicles, 60 nm AuNR accumulated as abundant intracellular foci within enlarged PMNs with circular morphologies and distended nuclei (Figure 4f), indicative of transcriptional and cytoplasmic activation [34]. Confocal micrographs and dark field microscopy showed that 3kDa‐AuNR were internalized via multifocal punctate vesicles, whereas 5kDa‐ and 10kDa‐AuNR were associated with singular vesicles (Figure S3 a–c). These data indicate that both 5 and 10kDa PEGs mitigated cell membrane engagement, lowered internalization, and rendered “stealth” properties to 24 nm AuNR.
Overall, our findings show that both AuNR size and PEG size substantially influence the intrinsic uptake by PMN. Collectively, the results allude to an AuNR size‐threshold of approx. 40 nm, above which the endocytic efficacy of PMN appeared to be dramatically enhanced. We conclude that high molecular weight 10 kDa PEG coverage on 24–40 nm AuNR is essential to delay uptake kinetics and alleviate intrinsic phagocytosis of AuNR by PMN.
AuNR Size and PEG Molecular Weight Influence PTT‐Mediated Cell Death in PMN
2.6
In cancer therapy, the most frequent application of AuNR‐mediated PTT is through direct ablation of tumor cells. In order to specifically target cancer cells, AuNR should be as ‘invisible’ as possible to PMN and other phagocytic cells that would unspecifically and intrinsically take up those particles and limit tumor target cell specificity. In the experiments presented above, we have established such parameters. An alternative approach in modern cancer immunotherapy is the targeted elimination of pro‐tumorigenic immune cells [35, 36]. One such immune cell subset are tumor‐associated neutrophils (TAN), and we and others have identified specific surface markers that mark tumor‐promoting and immunosuppressive TAN subsets [36, 37].In order to specifically target such subsets and leave remaining PMN unaffected, a low intrinsic uptake of targeted AuNR by PMN is likewise essential. Against this background, we explored how the AuNR physico‐chemical parameters identified in the previous sections would affect PTT‐mediated cell death of PMN. Based on the rationale that cell death may be induced by both apoptotic and necrotic routes [38, 39], we investigated whether PTT with different sizes of AuNR‐induced different cytotoxic routes by Annexin V (apoptosis) and 7‐AAD‐staining (necrosis) staining in PMN.
PMN exposed to multiple AuNR formulations (Ref. Figure 4) for 1 and 3 h were irradiated with NIR laser, and the temperature rise as well as cell death was assessed. The rise in temperature reflected AuNR uptake (Figure 5a,b) and showed significantly higher temperature enhancement in PMN cultures treated with 60 nm AuNR (55°C–65°C) for 1 h compared to smaller 24 and 40 nm AuNR (9°C–15°C and 30°C–50°C, respectively) and vehicle‐treated PMN (7°C). In each size category, the 3 kDa PEG‐covered AuNR‐induced higher temperatures than 10 kDa PEG‐covered AuNR. Higher AuNR incubation time of 3 h (Figure S4) generated the same rise in temperature as 1 h, suggesting that AuNR accumulation in PMN over time did not influence PTT efficiency. PTT with 24 nm AuNR did not induce cell death under our experimental conditions, where laser irradiation experiments were performed at room temperature and not at 37°C so that a temperature rise of 7°C was not sufficient to induce detectable cell damage (Figure 5c). Consistent with the higher increase in temperature, only AuNR sizes 40 and 60 nm induced substantial cell death in PMN cultures (Figure 5d,e), indicating that nanoparticle size, and not PEG molecular weight, is the primary determinant of cell death induction in PMN. Notably, with 40 nm AuNR (Figure 5d), but not 60 nm AuNR (Figure 5e), molecular weight of the PEG coverage influenced efficacy of cell death induction, with small PEG (3 kDa) inducing higher death count of PMN. In the majority of PMN, PTT‐induced positivity for 7‐AAD staining, which is indicative of cell membrane disintegration and subsequent nuclear staining of dying cells. Collectively, the data demonstrated that phagocytic internalization of AuNR by PMN and subsequent heat generation elicited rapid cytotoxicity, consistent with features of necrotic pathways. This finding contrasted with cell death induction in T‐lymphocytes, which exhibited a delayed caspase‐dependent apoptotic cell death 24 h after photothermal irradiation (Ref. Figure 2), likely mediated via membrane destabilization by receptor‐tethered α‐CD3‐AuNR triggering programmed death pathways.
PMN death by photothermal therapy is dependent on AuNR size. Isolated PMN (n = 3) were incubated for 1 h (25 000 AuNR/PMN) with PBS (black) and AuNR (24, 40, and 60 nm) sterically stabilized with 3, 5, and 10 kDa mPEG (red, yellow, and blue, respectively). Purified PMNs were irradiated for PTT using an 808 nm laser (5 W for 5 min). (a) The temperature was monitored with an infrared camera (initial temperature 28°C), and temperature‐increase relative to initial (ΔT) was plotted (n = 4 for 24 nm and 3 for 40 and 60 nm). The inset graph represents scaled ΔT induced by PTT with 24 nm AuNR. (b) Representative thermograms illustrating tissue culture well temperatures in treatment groups of PBS and different sizes of 5 kDa PEG‐AuNR (25‐100°C). (c–e) Cell death was assessed after 1 h rest at 37°C by Annexin V‐7AAD staining and flow cytometry. Left panels: The proportion of live PMN cells (Annexin‐V negative, 7‐AAD negative) remaining after PTT with (c) 24 nm AuNR, (d) 40 nm, and (e) 60 nm AuNR. Right panels: representative Annexin V‐7AAD 2D‐scatter plot from a single PMN donor (n = 3). Statistical test: ANOVA (for a) and unpaired t‐test with Welch's correction (for panels c–e). The statistical significance is defined as p > 0.05 not significant (ns), * p < 0.05; ** p < 0.01; *** p < 0.001, **** p < 0.0001. n indicates the number of experimental repetitions, and each performed with independent donors. AuNR = gold nanorod; NIR = near‐infrared; PBS = phosphate buffered saline; PMN = polymorphonuclear neutrophil; PTT = photothermal therapy.
In sum, our study shows that AuNR size, surface charge, and PEG coverage collectively influence uptake of AuNR in human professional PMN phagocytes. We also demonstrate that the AuNR size is dominant over PEG‐size in driving PTT‐mediated cell death of primary human PMN, subsequent to internalization.
Conclusion
3
This article establishes a framework and AuNR design strategy for PTT applications. We have identified crucial parameters that influence intrinsic uptake of AuNP by human phagocytes and T‐lymphocytes, and the resulting cell death responses. The principal function of phagocytic cells is the uptake and elimination of foreign particles [40], thereby PMN rapidly internalize particulate matter. The surface of AuNR triggers rapid macropinocytic uptake into large vacuoles, which activates stress‐responses characterized by elevated levels of cytokines such as IL‐8 and MMP9, reactive oxygen species (ROS) generation [41, 42], LOX‐1 upregulation‐mediated autocrine loop [10], and immunosuppression [36]. However, targeting phenotypic subpopulations of phagocytic cells, notably the immunosuppressive tumor‐associated‐PMN, requires circumventing intrinsic phagocytosis. In this study, we identify critical AuNR physico‐chemical parameters that may be utilized in several ways: (a) limit unwanted nonspecific phagocytosis of AuNR by PMN, (b) promote targeting of non‐PMN human cell types and (c) enable selective targeting of functional PMN subpopulations. AuNR size emerged as the most critical determinant of AuNR–PMN interactions, with smaller particles exhibiting a significantly diminished nonspecific uptake. Therefore, ultra‐small AuNR with PTT‐compatible aspect ratio may be essential for selectively targeting PMN subpopulations. In contrast, non‐phagocytic T‐lymphocytes demonstrated minimal non‐specific uptake, and therefore, can be targeted selectively using antibody functionalized AuNR. Our strategy also integrated AuNR physicochemical parameters with cell death outcomes in both immune cell types. The cell death mechanisms diverged by cell type, in which T‐lymphocytes underwent caspase‐mediated apoptosis, whereas PMN adopted a necrotic route that was dependent upon AuNR size. For these reasons, our data highlight the need to design AuNR formulations tailored for distinct cell types in order to achieve high cell type selectivity and low unwanted nontargeted delivery to bystander cells.
Materials and Methods
4
Materials
4.1
All consumables (pipette tips, culture plates, Falcon tubes) were purchased from Sarstedt, reaction tubes from Eppendorf, blood monovettes from Carl Roth, chemicals (Cytofix/cytoperm, FACS solutions) from BD Bioscience, cell culture medium, phosphate buffered saline (PBS), and Fluoromount G from Thermo Fischer Scientific. Tetrachloroauric(III)acid monohydrate (≥ 99.9%), silver nitrate (AgNO3, ≥ 99%), L‐ascorbic acid (≥ 99%), sodium borohydride (NaBH_4_, 99%), hydrochlorid acid (HCl, 1 M), hexadecyltrimethylammonium bromide (CTAB, ≥ 96%), 1‐decanol (≥ 99%), 2‐(4‐(2‐hydroxyethyl)piperazine‐1‐ethansulfonic acid (HEPES, ≥ 99.5%), sulfo‐N‐hydroxysuccinimide (sNHS), 1‐ethyl‐3‐(3‐dimethylaminopropyl) carbodiimide (EDC, ≥ 97%), Rhodamin B isothiocyanate (RITC, ≥ 90.0%), and glycerol (≥ 99.0%) were obtained from Sigma–Aldrich. α‐carboxy‐ω‐thiol PEG (α‐butyric acid‐ω‐mercaptopropanamido PEG, HS‐PEG‐COOH, 3000 Da and 5000 Da, ≥ 95%) and α‐mercapto‐ω‐amino PEG hydrochloride (HS‐PEG‐NH_2_, 5000 Da, ≥ 95%) were purchased from Rapp Polymere. Methoxy PEG thiol (mPEG‐SH, 3000, 5000, and 10 000 Da, ≥ 95%) and Cy5‐labelled PEG thiol (Cy5‐PEG‐SH, 3000, 5000, and 10 000 Da, ≥ 95%) were purchased from Biopharma PEG. For all syntheses deionized (DI) water was used (18.2 mΩ, Merck Millipore). Annexin V: PE Apoptosis Detection Kit was purchased from BD Bioscience, and ApoStat Intracellular Caspase Detection Kit was purchased from Biotechne.
Antibodies used for AuNR or AuNSPH functionalization were CD3, isotype mIgG2a, clone: OKT‐3 and its respective isotype antibody mIgG2a, clone: Cl.18.4, CD11b, isotype rIgG2a, clone: M1/70, and its respective isotype antibody rIgG2a, clone: L TF‐2, all purchased from BioXcell.
Instrumentation used in this publication is the following: The physicochemical properties of the synthesized gold nanorods (AuNR) were determined by UV/Vis absorption spectroscopy (JASCO V‐730), transmission electron microscopy (TEM, ZEISS EM910, JEOL JEM 1400+), and dynamic light scattering with ζ‐potential detection (WYATT, Moebiuζ). Inductively coupled plasma mass spectroscopy (ICP‐MS) (AGILENT, 7500a) and AAS was used to determine AuNR concentration. Cell counting was done with CASY Cell Counter + Analyzer System version 2 (Omni life science). Lymphocytes were isolated by autoMACS, using CD3 µ beads (Milteny Biotech). For assessing uptake, caspase activity and cell death, we used flow cytometry (BD FACS Canto II). Confocal imaging was performed by ZEISS Elyra PS.1 LSM710 with a ZEN system 2012 black edition of the Imaging Center Essen (IMCES). The irradiation was performed in a 96‐well plate format, with a single laser beam diameter of 0.64 cm (15.5 W/cm^2^) at 808 nm wavelength for 5 min, with an in‐house made setup, equipped with an infrared camera (Optris Pix connect). Darkfield imaging of a single PMN cell was performed at a self‐built darkfield microscopy set‐up, consisting of an inverted microscope (Nikon Eclipse Ti‐S) equipped with a white light source (Zeiss HAL 100 quartz collector), an automated piezo sample stage from Physik Instrumente (PI), a 20 x (Nikon, NA 0.4, WD 3.9) and 100x (Nikon, NA 0.9, WD 0.26) microscope objective, a camera (IDS, UI‐3240 CP‐M‐GL R2) and an OceanView QEPro spectrometer.
Methods
4.2
Antibody‐functionalized AuNR production
4.2.1
Gold nanorods for antibody functionalization were synthesized by a modified seed‐mediated growth method [43]. The nucleating solution was prepared by mixing 3.0 mL of a freshly prepared ice‐cold NaBH_4_ solution (0.1 M) with a solution of 1.25 mL HAuCl_4_ (0.01 M) and 48.75 mL CTAB (0.1 M) under rapid stirring. The resulting seed solution was stirred for 2 minutes and incubated at 30°C for 30 min. The growth solution was prepared by continuously mixing of CTAB (46 mL, 0.1 M), HAuCl_4_ (2.5 mL, 0.01 M), AgNO_3_ (0.8 mL, 0.01 M), HCl (1.0 mL, 1 M),and ascorbic acid (0.4 mL, 0.1 M). Upon the transformation of the growth solution to colorless, 50 mL of the growth solution was combined with 50 mL of the incubated seed solution under vigorous agitation (1 000 rpm) for 2 minutes. The resulting solution was incubated for approximately 24 h at 30°C. To remove cytotoxic CTAB, resulting final AuNRs were purified by centrifugation and resuspension in water with 0.2% sodium dodecyl sulfate (SDS). Subsequently, PEG‐ylation was achieved by incubating 1 mL of AuNR (OD = 1) with 100 µL of a freshly prepared aqueous PEG solution (2 mM, HS‐PEG‐COOH, 3 kDa) overnight, followed by centrifugation and resuspension in HEPES. For antibody conjugation, 500 µL of the PEG‐ylated AuNR (OD = 1) was incubated with sNHS (20 µL, 15 mM) and EDC (20 µL, 6 mM) for 25 min (500 rpm, RT). The reaction mixture was purified by centrifugation and resuspended in HEPES. Subsequently, 1 µL of the antibody solution (0.5 mg/mL in glycerol) was added. The sample was agitated for 3.5 h (500 rpm, RT) and centrifuged at 4°C, followed by resuspension in 1x PBS buffer. Depending on the readout strategy in different experiments, antibody‐coupled AuNPs were additionally labelled with fluorophores (Alexa 647) for detection in flow cytometry. Quality control was performed for all antibody‐functionalized particles used in this paper by UV/Vis absorption spectroscopy and TEM imaging (Figures S5 and S6). The employed amount of AuNR for each experiment equaled a final concentration of 0.4875 µg/mL +/‐ 15.7 % (0.078 nM, 0.4875 pg / cell) for isolated cell samples, and a final concentration of 0.929 µg/mL +/‐ 15.7 % (0.149 nM, maximal concentration of 0.4875 pg / PMN) AuNPs for multicellular culture (whole blood) samples.
Synthesis and Characterization of Quasi‐Spherical Gold Nanoparticles (AuNSPH)
4.2.2
AuNSPH were prepared in a two‐step synthesis. Firstly, CTAB (10 mL, 100 mM) was mixed with HAuCl_4_ (500 µL, 10 mM) and NABH4 (600 µL, 10 mM). The resulting seed solution was stirred for 2 h at 30°C. Subsequently, 15 µL of the seed solution was mixed with CTAC (2 mL, 200 mM) and L‐ascorbic acid (1.5 mL, 100 mM). Furthermore, HAuCl_4_ (2 mL, 0.5 mM) was added, and the solution was stirred for another 30 min. The resulting AuNSPH were characterized by UV/Vis absorption spectroscopy and TEM imaging. ζ‐potential of AuNSPH during different steps of synthesis and functionalization (CTAB and PEG‐ylation) was analyzed in the absence and presence of human serum.
Dynamic Light Scattering
4.2.3
Dynamic light scattering (DLS) was used to determine the particles hydrodynamic diameter, ζ potential and monodispersity. Thus, AuNR (OD = 0.1) were analyzed as triple replicates with ten DLS acquisitions and an acquisition time of 5 s. For ζ potential measurements, the samples were analyzed in a ζ cell with the voltage amplitude being set to 5 V, an electric field frequency of 20 Hz, and a PALS collection period of 15 s. The temperature was set to 20°C for all measurements. ζ values correspond to the mean value of three runs with ten DLS acquisitions each.
Unfunctionalized AuNR Production
4.2.4
The production of AuNR is based on a seed‐mediated method [43]. The synthesis was performed in 25 mL vials. Initially, a cluster solution containing 1–2 nm gold clusters was prepared. Thus, a CTAB/decanol mixture (10 mL, 50 mM CTAB, and 13.5 mM decanol) was combined with tetrachloroauric acid (500 µL, 10 mM) and ascorbic acid (50 µL, 100 mM) under stirring at 400 rpm. Upon decoloration, a freshly prepared sodium borohydride solution (400 µL, 20 mM) was added rapidly while stirring at 1000 rpm. The cluster solution was stirred for an additional 2 minutes at 400 rpm and then incubated at 30°C for 1 h. To synthesize AuNR approximately 24 nm in length, a growth solution was prepared by mixing the same CTAB/decanol solution, tetrachloroauric acid (500 µL, 10 mM), silver nitrate (105 µL, 10 mM), and hydrochloric acid (700 µL, 1 M). Subsequently, ascorbic acid (130 µL, 100 mM) was added, followed by addition of the pre‐synthesized cluster solution (850 µL) upon decoloration. All components were combined under stirring at 400 rpm. The mixture was incubated overnight at 30°C, followed by two centrifugation steps and resuspension in DI water. To synthesize AuNR approximately 40/60 nm in length, the synthesis regarding volumes in the growth solution was adjusted as following: 150/120 µL silver nitrate, 600/550 µL HCl, 80/130 µL ascorbic acid and 950/1100 µL of the cluster solution, respectively. After purification, the concentration of AuNR samples were determined using UV/Vis absorption spectroscopy [44]. Given the concentration and size dimensions of the particles from TEM analysis, the needed volume of PEG derivates (10 mg/mL) yielding in a 100x excess was calculated and incubated with AuNR overnight, followed by two centrifugation steps. For flow cytometry experiments, 30% of Cy5‐labelled mPEG was included in this excess of PEG. For RITC functionalization, the colloid was incubated with amino PEG (HS‐PEG‐NH_2_) overnight and purified by centrifugation. An ethanolic solution of RITC (1 µL, 10 mM) was added to 1 mL AuNR (OD = 1) and incubated overnight at 4°C, followed by two centrifugation steps and resuspension in DI water. Quality control was performed for all unfunctionalized particles used in this paper by UV/Vis absorption spectroscopy and TEM imaging (Figures S7 and S8).
Isolation of immune cells from the peripheral blood (PMN and lymphocytes)
4.2.5
Fresh 3.8% sodium citrate anticoagulated blood of healthy donors was diluted 1:1 with PBS and applied to density gradient centrifugation using Biocoll separating solution (Biochrom/Merck, Berlin, Germany). PBMCs were collected from the upper gradient (above the Biocoll). After washing with PBS, T cells were isolated using magnetically labeled CD3 micro beads (Miltenyi Biotec, Bergisch‐Gladbach, Germany) according to the manufacturer's instructions. The neutrophil fraction on top of the erythrocytes pellet was collected in a fresh test tube. Contaminating erythrocytes were removed by sedimentation for 30 min with a solution containing 1% poly vinyl alcohol (Sigma–Aldrich, Taufkirchen, Germany), and residual erythrocytes were removed by hypotonic shock. The isolated PMN were rested for 1 h at 37°C prior to treatment in RPMI‐1640 (Thermo Fisher Scientific, Karlsruhe, Germany) supplemented with 10% autologous serum (10% AS medium).
Whole Blood Flow Cytometry Assay
4.2.6
3.8% sodium citrate anticoagulated blood of healthy donors was always freshly used. All experiments were performed in protein low‐binding 1.5 mL reaction tubes (Eppendorf AG). Whole blood was treated with water (NC) or 5000 AuNR or AuNSPH/PMN. Afterward, erythrocytes were lysed using 1× BD Pharm lysing buffer (BD Bioscience), followed by centrifugation and washing with PBS. The samples were resuspended in PBS suppelemented with 3% human serum and processed by flow cytometry.
Treatment of Immune Cells with Gold Nanorods and Nanospheres
4.2.7
Cells were treated with varying formulations of AuNR or AuNSPH (25,000 NP/cell) (15 nm AuNR = 0.449 µg/mL, 0.449 pg/cell, AuNSPH = 1.67 µg/mL, 1.67 pg/cell, 30 nm AuNR = 4.08 µg/mL, 4.08 pg/cell, AuNSPH = 13.2 µg/mL, 13.2 pg/cell, 60 nm AuNR = 33.31 µg/mL, 33.31 pg/cell, AuNSPH = 106.23 µg/mL, 106.23 pg/cell) for varying amounts of time (ranging from 30 min to 4 h) at 37°C. The excess of the NPs was removed by washing with PBS and subsequent centrifugation. Antibody and isotype functionalized AuNR that were not fluorescent were stained with a secondary antibody for 30 min at 37°C. Cells were resuspended in PBS supplemented with 3% of human serum, and the uptake was measured by flow cytometry.
Laser Irradiation and Temperature Measurement
4.2.8
Cell suspensions pre‐treated with AuNP for laser irradiation were washed with PBS and resuspended with RPMI + 10% AS. The samples were irradiated with laser beam of 5 W power, at 808 nm wavelength for varying periods of time (5, 10 or 15 min). After removing the excess of AuNR washing with PBS, cells were irradiated for 5 min using a 5 W laser at 808 nm, and temperature was measured with an infrared camera (Optris Pix connect). The temperature increase was determined by subtracting the temperature before the irradiation (n = 4). Afterwards, plates were rested at 37°C for one or 24 h, before being processed for viability analysis by caspase or Annexin‐V/7‐AAD (BD Bioscience) staining according to manufacturers’ instruction by flow cytometry.
Caspase Staining
4.2.9
Cells were stained for caspase activity during the last 30 min of incubation/resting time after apoptosis induction, using an intracellular Caspase Detection Kit (BioTechne), containing fluorescein‐conjugated V‐D‐FMK. The cell‐permeable caspase inhibitor was added at 37°C in a concentration of 10 µL / mL cell suspension, according to manufacturer's instruction. To remove unbound reagent, cell suspensions were washed with PBS (300 g, 5 min) and subsequently measured by flow cytometry using FlowJo 10 software.
Confocal Microscopy
4.2.10
Isolated PMN were incubated with 5000 RITC‐labeled AuNR/PMN for 30 min at 37°C in protein low‐binding reaction tubes (Eppendorf AG, Germany) in rotation. Afterward, samples were washed with PBS, resuspended in fetal calf serum, and cytospins onto 24 × 48 mm thick coverslips were prepared. Cells were fixed using BD Cytofix/Cytoperm (BD Bioscience, Heidelberg, Germany), and nuclear counter staining with 4,6‐Diamidino‐‐2‐phenylindol (BioLegend, Amsterdamm, Netherlands) was performed. Coverslips were mounted on slides with Fluoromount G (Thermo Fisher scientific). Confocal microscopy was performed using a 63× magnification oil objective and ZEISS Elyra PS.1 LSM710 with a ZEN system 2012 black edition of the Imaging Center Essen (IMCES). Single cell fluorescence intensity of AuNR within PMN was analyzed using Fiji ImageJ 1.53c
Inductively Coupled Plasma‐Mass Spectrometry (ICP‐MS)
4.2.11
Isolated PMN were incubated with 10 000, 5000, 2500, 1250, 625, 312, or 156 AuNR, for varying time points (30 min, 2 and 4 h). Subsequently, excess AuNR were removed, PMN were counted, and PMN‐associated AuNR were analyzed in ICP‐MS. For this, 500 μL of the PMN solution (between 7 × 10^5^ and 1.5 × 10^6^ PMN) was placed into 20 mL TFM vessels (MarsXpress, CEM Corporation, Kamp‐Lintfort, Germany). Samples were digested in a mixture of 65% nitric acid (sub boiled quality, Carl Roth GmbH + Co. KG, Karlsruhe, Germany) and 35% hydrochloric acid (suprapur quality, Carl Roth GmbH + Co. KG, Karlsruhe, Germany) in a 3:1 ratio (reverse aqua regia). Subsequently, sample solution was ramp heated up to 170°C and held at this temperature for 40 min using a Mars 6 microwave digestion system (CEM Corporation, Kamp‐Lintfort, Germany). After digestion, the solution was transferred in a volumetric flask and brought to a volume of 5 mL with Milli‐Q water. The concentration of gold within the sample solution was analyzed using ICP‐MS. The analyses were carried out with a quadrupole ICP‐MS system (PerkinElmer Sciex Elan DRC‐e, PerkinElmer Corporation, Waltham, USA) operating at 1100 W plasma power, 14 L/min plasma gas flow and 0.95 L/min nebulizer gas flow and an auto sampler system (PerkinElmer AS‐90) connected with a peristaltic pump with a sample flow of 1 mL/min. To avoid contamination and memory effects, the wash time between measurements was set at 30 s (1% reverse aqua regia, supra pure quality). Before analyses, the samples were diluted 1:10 using a solution of 1% HNO_3_ (sub boiled) with a concentration of 10 ng/L of thulium (CetriPUR, Merck, Darmstadt, Germany) as an internal standard. To control the accuracy and stability during measurements, a standard solution of gold with a concentration of 10 μg/L was analyzed after every 10 samples. The calibration was carried out with a series of 11 dilutions of a gold standard solution (Gold ICP‐Standard, Bernd Kraft GmbH, Duisburg, Germany). Element concentrations were calculated as mg/L using corresponding regression lines (correlation factor ≥ 0.999). All samples were normalized to respective cell count initially added (ng/10^5^ PMN).
Atomic Absorption Spectroscopy (AAS)
4.2.12
Gold was also analyzed using a graphite furnace AAS (GF‐AAS; Perkin Elmer AAnalyst 600, PerkinElmer Corporation, Waltham, USA) with Zeeman‐effect background correction. The instrument was calibrated with a series of six standard solutions ranging from 5 to 100 µg/L, to which the sample matrix was added (matrix‐adapted calibration). Prior to the measurements, the samples were diluted two times using 1% reversed aqua regia. Concentrations were determined based on the respective calibration curves, following the same regression‐based approach applied in the ICP‐MS analysis.
Statistical Analysis
4.3
Statistical analysis was performed with GraphPad Prism 8 software. Unless otherwise indicated we used unpaired t‐test with Welch's correction and ANOVA. Data are depicted as the mean value with standard error of the mean (±SEM). N indicates the number of experimental repetitions, and each performed with individual, independent blood donors.
Supporting Information
Additional supporting information can be found online in the Supporting Information section. Supporting Fig. S1: AuNSPH can specifically target T‐Lymphocytes, while PMN exhibit unspecific uptake. (a) Isolated T‐cells (left) and whole venous blood (right) were treated with differentially functionalized (full back circle: BSA‐functionalized, empty blue circle: mIgG2a (isotype)‐functionalized, full blue circle: CD3‐functionalized, empty red circle: rIgG2b (isotype)‐functionalized, full red circle: CD11b‐functionalized) and sized (15, 30 and 60 nm) AuNSPH. All cells were treated with 5000 AuNSPH‐RITC/cell for 30 min. Uptake was analyzed by flow cytometry (n = 5). (b) Isolated PMN (left) and whole venous blood (right) was treated with differentially functionalized (full back circle: BSA‐functionalized, empty blue circle: mIgG2a (isotype)‐functionalized, full blue circle: CD3‐functionalized, empty red circle: rIgG2b (isotype)‐functionalized, full red circle: CD11b‐functionalized) and sized (15, 30 and 60 nm) AuNSPH. All cells were treated with 5000 AuNSPH‐RITC/cell for 30 min. Uptake was analyzed by flow cytometry (n = 5). Statistics performed with an unpaired t‐test with Welch's correction. The statistical significance is defined as: p>0.05 not significant (ns), * p<0.05; ** p<0.01; *** p<0.001.). n indicates the number of experimental repetitions, each performed with independent donors. Supporting Fig. S2: Size‐dependent non‐specific uptake of positively charged AuNR. (a) Table representing measured ζ potential of two different‐sized AuNR (24 and 50 nm). (b) Representative flow cytometry graphs showing the uptake of PBS (black), 24 nm (red) and 50 nm (yellow) RITC‐AuNR in isolated PMN (5000 AUNR/cell) after 30 min, 2 hr and 4 hr. (c) Quantification of the RITC‐AuNR taken up by isolated PMN after 30 min, 2 hr and 4 hr, as measured by flow cytometry. Red: 24 nm AuNR; Yellow: 50 nm AuNR. X‐axis: AuNR size, y‐axis: Delta median RITC intensity (median PBS intensity subtracted from measured median RITC intensity), n=3 (d) Quantification of gold present in the PMN via AAS. PMNs were isolated from whole venous blood and treated with 24 nm or 50 nm AuNR (5000 AuNR/cell) for 30 min, 2 hr and 4 hr. n=3. n indicates the number of experimental repetitions, each performed with independent donors. Supporting Fig. S3: Size of PEG shell regulates non‐specific uptake of small (24 nm) AuNR. (a) Table representing measured ζ potential of 24 nm‐sized AuNR, with 3 kDa, 5 kDa and 10 kDa mPEG. (b) Confocal micrographs representing PMNs (blue) taking up 24 nm‐sized Cy5‐AuNR (red), covered (from left to right) with 3 kDa, 5 kDa and 10 kDa mPEG, localization associated with punctate intracellular vesicle‐like structures. (c) Representative dark‐field micrographs showing 24 nm AuNR, covered with 3 kDa (up), 5 kDa (middle) or 10 kDa mPEG (lowest panel), taken up by a single PMN cell, displaying PEG size‐dependent intracellular staining pattern. Each line represents a z‐stack through a single cell (step‐size: 2 μM). Supporting Fig. S4: AuNR size‐based temperature increase by photothermal therapy. Isolated PMN were treated with PBS (black) or 24 nm, 40 nm or 60 nm AuNR covered with varying mPEG size (3 kDa (red), 5 kDa (yellow) or 10 kDa (blue)) for 3 hr. After removing the excess of AuNR by washing with PBS, cells were irradiated for 5 min using a 5 W laser at 808 nm and temperature was measured with an infrared camera (Optris Pix connect) for 5 min. ΔT was calculated by subtracting the temperature before starting the irradiation (n = 6 for 24 nm and n=3 for 40 and 60 nm). Statistical test: ANOVA. The statistical significance is defined as: p>0.05 not significant (ns), * p<0.05; ** p<0.01; *** p<0.001.). n indicates the number of experimental repetitions, each performed with independent donors. Supporting Fig. S5: Quality control data of antibody‐functionalized AuNR. (a) UV/Vis absorption spectra and representative TEM images of 15 nm (black), 24 nm (green), 30 nm (red) and 60 nm (blue) AuNR with absorption maxima in the range of 630‐650 nm. Scale bar: 100 nm. (b) UV/Vis absorption spectra of unfunctionalized AuNR (black), PEGylated AuNR (red) and mIgG2a‐functionalized AuNR (blue) shown for a particle length of 30 nm. The spectra show a stable colloid upon PEGylation antibody functionalization. Supporting Fig. S6: Quality control data of antibody‐functionalized AuNSPHs. (a) UV/Vis absorption spectra and representative TEM images of 15 nm (black), 30 nm (red) and 60 nm (blue) spherical goldnanoparticles (AuNSPH). Scale bar: 100 nm. (b) UV/Vis absorption spectra of unfunctionalized AuNSPH (black), PEGylated AuNSPH (red) and mIgG2a‐functionalized AuNSPH (blue) shown for a particle size of 30 nm. The spectra show a stable colloid upon PEGylation and antibody functionalization. Supporting Fig. S7: Quality control data of different sized AuNR. (a) UV/Vis absorption spectra of 24 nm (red), 40 nm (pink), 50 nm (green) and 60 nm (blue) AuNR with absorption maxima in the range of 800‐815 nm, which fits the excitation wavelength of the near‐infrared (NIR) laser at 808 nm used for photothermal therapy experiments. (b) Representative TEM images of 24 nm (red), 40 nm (pink), 50 nm (green) and 60 nm (blue) AuNR. Scale bar: 100 nm. (c) ζ potential (left y axis, grey) and hydrodynamic radius (right axis, orange) measured by dynamic light scattering (DLS) of 24 nm and 50 nm sized AuNR functionalized with 5 kDa NH_2_‐PEG and RITC. Corresponding to the positively charged PEG bound to the particle surface, a positive ζ potential is obtained. It must be noted that the hydrodynamic radius determined by DLS refers to spherical particles. As a consequence, the determined radius values do not reflect the actual particle size of nanorods. Supporting Fig. S8: Quality control data of AuNR with varying surface modification. (a) UV/Vis absorption spectra of 24 nm AuNR before and after PEGylation, while the ratio of methoxy/carboxy PEG was varied. A red‐shift of the longitudinal plasmon band upon PEGylation and with increase of methoxy PEG is obtained. (b) ζ potential (left y axis, grey) and hydrodynamic radius (right axis, orange) measured by DLS of 24 nm AuNR with different sizes of methoxy PEG (in kDa). 0 kDa mPEG corresponds to AuNR before PEGylation. A large positive ζ potential is detected corresponding to the positively charged ligand CTAB being bound onto the particle surface. The ζ potential turns negative upon PEGylation with methoxy PEG, showing the surface potential change after washing and PEGylating the particles. The hydrodynamic radius increases with size of PEG, from 3 kDa to 10 kDa. (c) ζ potential (left y axis, grey) and hydrodynamic radius (right axis, orange) measured by DLS of 24 nm AuNR with a varying ratio of methoxy PEG and carboxy PEG. 0% mPEG corresponds to 100% carboxy PEG, etc. The ζ potential decreases with higher percentages of negatively charged carboxy PEG.
Author Contributions
Tista Roy Chaudhuri: conceptualization (equal), formal analysis (lead), methodology (equal), visualization (supporting), writing – original draft (lead). Helene Giesler: data curation (supporting), formal analysis (supporting), investigation (supporting), methodology (supporting), writing – review & editing (supporting). Marija Kovacevic Sarmiento: conceptualization (supporting), data curation (lead), formal analysis (supporting), writing – original draft (lead), writing – review & editing (lead). Kim Lamers: data curation (supporting), formal analysis (supporting), investigation (equal). Michelle Hechler: data curation (supporting), formal analysis (supporting), investigation (supporting), methodology (supporting), writing – review & editing (supporting). Michael Erkelenz: data curation (supporting). Marvin Haferkamp: investigation (supporting). Ronja Schirrmann: data curation (supporting), formal analysis (supporting), supervision (supporting), writing – review & editing (equal). Milen Nachev: data curation (supporting), methodology (supporting), writing – review & editing (supporting). Rebeka Bosnjakovic: data curation (supporting), formal analysis (supporting), investigation (supporting). Bernd Sures: data curation (supporting), formal analysis (supporting), methodology (equal), resources (equal), writing – review & editing (supporting). Sebastian Schlücker: conceptualization (equal), funding acquisition (equal), project administration (equal), resources (equal), supervision (equal), writing – original draft (equal), writing – review & editing (equal). Sven Brandau: conceptualization (lead), funding acquisition (lead), project administration (equal), resources (equal), supervision (equal), writing – original draft (lead), writing – review & editing (equal).
Funding
T.R.C. was supported by the Else‐Kröner Fresenius Stiftung through funding of UMESCIA (University Medicine Essen Medical Scientist Academy, spokesperson S.B., project to S.S.). Research in this manuscript was also supported by Deutsche Forschungsgemeinschaft through funding of TRR 332 (project A4 to S.B.), by German Cancer Aid through the ‘visionary novel concepts’ call with grant number 70113769 and by the Josepha and Charlotte von Siebold grant program of Medical Faculty, University of Duisburg‐Essen.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supplementary Material
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1H. S. Han and K. Y. Choi , “Advances in Nanomaterial‐Mediated Photothermal Cancer Therapies: Toward Clinical Applications,” Biomedicines 9, no. 3 (2021): 305, 10.3390/biomedicines 9030305.33809691 PMC 8002224 · doi ↗ · pubmed ↗
- 2J. B. Vines , J. H. Yoon , N. E. Ryu , D. J. Lim , and H. Park , ”Gold Nanoparticles for Photothermal Cancer Therapy.,” Frontiers in Chemistry 7 (2019): 167, 10.3389/fchem.2019.00167.31024882 PMC 6460051 · doi ↗ · pubmed ↗
- 3S. He , J. Song , J. Qu , and Z. Cheng , “Crucial Breakthrough of Second Near‐Infrared Biological Window Fluorophores: Design and Synthesis toward Multimodal Imaging and Theranostics,” Chemical Society Reviews 47, no. 12 (2018): 4258–4278, 10.1039/c 8cs 00234 g.29725670 · doi ↗ · pubmed ↗
- 4Y. Nakamura , A. Mochida , P. L. Choyke , and H. Kobayashi , “Nanodrug Delivery: Is the Enhanced Permeability and Retention Effect Sufficient for Curing Cancer?,” Bioconjugate Chemistry 27, no. 10 (2016): 2225–2238, 10.1021/acs.bioconjchem.6b 00437.27547843 PMC 7397928 · doi ↗ · pubmed ↗
- 5T. Roy Chaudhuri , N. L. Straubinger , R. F. Pitoniak , et al., “Tumor‐Priming Smoothened Inhibitor Enhances Deposition and Efficacy of Cytotoxic Nanoparticles in a Pancreatic Cancer Model,“ Molecular Cancer Therapeutics 15, no. 1 (2016): 84–93, 10.1158/1535-7163.MCT-15-0602.26516158 PMC 4707122 · doi ↗ · pubmed ↗
- 6R. M. S. Tista Roy Chaudhuri , “Nanoparticles for Brain Tumor Delivery,” in Nervous System Drug Delivery ed. Russell R. Lonser and M. S. Krystof Bankiewicz (Academic Press, 2019), 229–250.
- 7C. Summers , S. M. Rankin , A. M. Condliffe , N. Singh , A. M. Peters , and E. R. Chilvers , “Neutrophil Kinetics in Health and Disease,” Trends in Immunology 31, no. 8 (2010): 318–324, 10.1016/j.it.2010.05.006.20620114 PMC 2930213 · doi ↗ · pubmed ↗
- 8P. Makvandi , M. Chen , R. Sartorius , et al., “Endocytosis of Abiotic Nanomaterials and Nanobiovectors: Inhibition of Membrane Trafficking,” Nano Today 40 (2021): 101279, 10.1016/j.nantod.2021.101279.34518771 PMC 8425779 · doi ↗ · pubmed ↗