Ultrasound-targeted microbubble destruction enhances RSL3-induced ferroptosis in anaplastic thyroid carcinoma

Summary

Anaplastic thyroid carcinoma (ATC) is a highly aggressive malignancy with limited effective therapies. Ferroptosis, a form of programmed cell death, is closely associated with thyroid cancer prognosis and represents a potential therapeutic target. This study evaluated the antitumor effects of ultrasound-targeted microbubble (MB) destruction (UTMD)-mediated RAS-selective lethal 3 (RSL3) delivery to induce ferroptosis in ATC. MBs loaded with ferroptosis inducer RSL3 were synthesized and intravenously administered in a nude mouse ATC model, followed by localized ultrasound irradiation to enable targeted intratumoral drug release. RSL3-MBs exhibited uniform particle size, good biocompatibility, and excellent contrast performance. The combination of RSL3-MBs with UTMD significantly inhibited the growth of ATC without evident systemic toxicity. This strategy was shown to be more effective than the delivery methods involving either RSL3 solution or RSL3-MBs alone. These findings support UTMD-mediated RSL3 delivery as an effective approach to enhance ferroptosis-based therapy for ATC.

Graphical abstract

Highlights

• •RSL3-MBs enable contrast imaging and ferroptosis triggered by UTMD in ATC
• •UTMD enhances intratumoral accumulation of RSL3
• •RSL3-MBs combined with UTMD suppress ATC growth with minimal toxicity

Cancer; drug delivery system

Introduction

Anaplastic thyroid carcinoma (ATC) is an extremely aggressive malignant tumor characterized by rapid progression, a high rate of metastasis, and a very poor prognosis.¹ Despite accounting for less than 2% of thyroid cancers, ATC is responsible for 14%–39% of the mortality associated with thyroid carcinoma.² The median survival for patients with ATC is approximately 5 months, with a 1-year overall survival rate of 20%.³ With few exceptions, almost all patients die within 6 months of developing ATC.^2,4,5 Therefore, prolonging overall survival of ATC patients has become an urgent issue.

Surgical resection combined with radiotherapy and chemotherapy has traditionally been the mainstay of treatment for ATC.³ While surgery may offer benefit in selected patients with stage 4A or 4B disease, the overall prognosis has remained extremely poor. Over the past decade, however, the management of ATC has undergone a profound transformation from palliative intent to precision oncology. Comprehensive genomic profiling has revealed that BRAF^V600E and RAS mutations remain the main driver mutations in ATC.⁶ For patients with BRAF^V600E-mutated ATC, combined BRAF and MEK inhibition with dabrafenib and trametinib has emerged as a highly effective targeted therapy, demonstrating a high overall response rate, prolonged duration of response, and improved survival with manageable toxicity.^7,8 Importantly, immunotherapy has further expanded therapeutic possibilities, especially for patients without actionable driver mutations. Approximately, 22%–29% of ATC tumor express programmed death-ligand 1 (PD-L1), suggesting susceptibility to PD-1/PD-L1 blockade.^9,10 Spartalizumab, a humanized anti-PD-1 monoclonal antibody, has demonstrated promising antitumor activity and a good safety profile in patients with advanced or metastatic ATC.¹¹ Emerging evidence further indicates that integrating immunotherapy with targeted therapy may be synergistic. Recent data suggest that adding the immune checkpoint inhibitor pembrolizumab to dabrafenib/trametinib may significantly prolong survival in patients with BRAF^V600E-mutated ATC.¹² Despite these advances, many patients remain ineligible for targeted therapies or develop resistance to systemic treatments. Therefore, identifying additional therapeutic targets and developing more effective drug-delivery strategies remain critical priorities for improving outcomes and reducing mortality in ATC.

Ferroptosis, a type of programmed cell death, has been shown to play a crucial role in various cancers.^13,14 Recent accumulating evidence indicates that ferroptosis is strongly associated with the prognosis of thyroid cancer and holds promise as a potential therapeutic target.^15,16,17 However, there are currently few experimental studies on ferroptosis-based treatments for ATC, and the existing research primarily focuses on the cellular level.^18,19 Hence, investigating the efficacy of inducing ferroptosis in ATC treatment may offer potential strategies for managing ATC. RAS-selective lethal 3 (RSL3) acts as a ferroptosis inducer by binding to glutathione peroxidase 4 (GPX4), resulting in its irreversible inactivation. This prevents cells from converting lipid peroxides into lipid alcohols, leading to the accumulation of lipid peroxides, which causes fatal damage and rupture of the cell membrane, ultimately inducing cell death.²⁰ Previous studies have shown that RSL3 is capable of inducing ferroptosis in a range of cancers.^21,22,23,24 Unfortunately, due to the low water solubility, poor biocompatibility, and side effects associated with systemic administration, the application of RSL3 in vivo is limited.^20,25,26,27 In addition, ferroptosis also occurs in normal tissues. The systemic and nonspecific distribution of RSL3 may cause complications in cancer therapy. Therefore, it is important to study improved drug delivery systems that enhance biocompatibility and reduce severe side effects for the delivery of RSL3 to tumor cells.

Ultrasound-targeted microbubble destruction (UTMD) enables both acoustic imaging and targeted delivery of genes and drugs.^28,29 Drugs are encapsulated within acoustic microbubbles (MBs), which are administered via intravenous injection. Ultrasound irradiation is then applied to the tumor site, utilizing the cavitation, sonoporation, and mechanotransduction effects of ultrasound waves to selectively release drugs at the target site.³⁰ This technology enables targeted drug delivery and enhances cellular drug uptake, thereby maintaining therapeutic effect while reducing the required dosage and minimizing side effects.³¹

This study aims to explore the antitumor potential of UTMD-mediated RSL3 delivery in inducing ferroptosis for the treatment of ATC. We designed a drug delivery system in which the ferroptosis inducer RSL3 is encapsulated within acoustic MBs. By injecting RSL3-MBs into the mouse, the ultrasound imaging signal is enhanced, facilitating the identification of tumor location and size. Through UTMD technology, MBs were selectively disrupted in tumor regions, resulting in the localized release of RSL3. MB-mediated cavitation can improve the permeation of cell membranes, leading to a substantial accumulation of RSL3 within tumor tissue, and thereby inducing ferroptosis and treating ATC.

In addition, the functioning mechanism of the drug delivery system that inhibits ATC was studied (Figure 1). Our research may offer a promising strategy for ATC therapy and a basis for clinical drug investigations.

Figure 1.

An illustrative diagram showing the synthesis of RSL3-MBs and their delivery via UTMD for ATC treatment (by Figdraw)

Results

Characterizations of the RSL3-MBs

The microscopic images revealed that RSL3-MBs exhibit a spherical shape, uniform size, and uniform dispersion (Figure 2A). The frequency distribution chart showed that the particle sizes of MBs and RSL3-MBs were primarily distributed around 1 μm (Figure 2B). The quantitative results showed that the mean particle sizes of MBs and RSL3-MBs were 1.75 ± 1.41 and 1.89 ± 1.40 μm, respectively (p = 0.013). The above results suggested that adding RSL3 increased the size of RSL3-MBs compared to MBs. The results showed that the zeta potential of MBs was measured at −24.33 ± 0.95 mV. After adding RSL3, the surface charge changed from −24.33 ± 0.95 to −25.43 ± 0.12 mV (Figure 2C). The zeta potential shows no statistically significant difference between MBs and RSL3-MBs (p = 0.180). To evaluate the stability of RSL3-MBs, we monitored their mean diameter over a period of 1.5 h. The results showed a slight change in size within 1.5 h after preparation (Figure 2D) and suggested that RSL3-MBs could remain stable for longer than 1.5 h after preparation.

Figure 2.

Characterizations of the RSL3-MBs

(A) Microscopic image of RSL3-MBs. Scale bars, 5 μm.

(B) Size distribution of the MBs and RSL3-MBs.

(C) Zeta potential of MBs and RSL3-MBs (n = 3 per group; mean ± SEM; and independent samples t test).

(D) Size changes of RSL3-MBs within 1.5 h after preparation.

(E) Viability of Nthy-ori 3-1 cells incubated with different concentrations of MBs (n = 5 per group; mean ± SEM; and one-way ANOVA).

(F) Viability of Nthy-ori 3-1 cells incubated with different concentrations of RSL3-MBs (n = 5 per group; mean ± SEM; and one-way ANOVA).

(G) HE staining of heart, liver, spleen, lung, and kidney sections at day 12 after injection of MBs and RSL3-MBs. Scale bars, 50 μm.

High-performance liquid chromatography (HPLC) was used to measure the RSL3 content in RSL3-MBs. All samples were treated with methanol to disrupt the MB structure and subsequently subjected to sonication to ensure complete release of the encapsulated RSL3. The chromatogram showed a chromatography peak for the RSL3 standard at 3.75 min (Figure S1), and an RSL3 chromatography peak was also detected in the RSL3-MBs sample at the same time point (Figure S2), which confirms that RSL3 was successfully loaded into the MBs. In contrast, no RSL3 peak was observed at the corresponding time point in the chromatogram of the blank MBs sample (Figure S3). HPLC measured the RSL3 concentration in RSL3-MBs to be 1,020.72 ± 0.91 μg/mL. With 1.3 mg of RSL3 initially used for preparing RSL3-MBs, the drug encapsulation efficiency of RSL3-MBs was 78.46%.

Good biosafety is an essential prerequisite for the in vivo application of RSL3-MBs. Nthy-ori 3-1 cells were exposed to varying concentrations of MBs or RSL3-MBs for 24 h, and their cellular biosafety was evaluated using a Cell Counting Kit-8 (CCK-8) cytotoxicity assay. There was no significant effect on cell viability after exposure to various concentrations of MBs or RSL3-MBs compared to the control group (Figures 2E and 2F). The result demonstrated that MBs and RSL3-MBs exhibited no toxicity to Nthy-ori 3-1 cells, indicating their high biosafety. Next, histological analysis was performed to evaluate the in vivo biosafety of both MBs and RSL3-MBs. The findings indicated that the heart, liver, spleen, lung, and kidney did not exhibit significant histopathological damage (Figure 2G). Therefore, the results suggested that the MBs and RSL3-MBs exhibited good biosafety both in vitro and in vivo.

Ultrasound imaging performance and fluorescence molecular imaging

Contrast ultrasound images of RSL3-MBs solutions at different acoustic power were acquired with a high-frequency probe (4–12 MHz). As shown in Figure 3A and as the acoustic power increased, the amount of RSL3-MBs undergoing bursting gradually increased, indicating good acoustic response capabilities of RSL3-MBs in vitro. Then, we investigated the ultrasound imaging performance of RSL3-MBs on the tumor xenograft in vivo. As shown in Figure 3B, after RSL3-MBs were injected, an evident ultrasound contrast signal appeared at the tumor site. Subsequently, the application of UTMD at the tumor location significantly weakened the ultrasound contrast signal, suggesting that the MBs were disrupted. The dynamic images show that RSL3-MBs produce persistent contrast signals up to 4 min post-injection (Figure S4). This duration, which exceeds the 3-min UTMD treatment window, ensures sufficient availability of RSL3-MBs for ultrasound-triggered drug release. These results suggest that RSL3-MBs exhibit good imaging performance and a circulation time adequate for ultrasound imaging-guided ATC therapy.

Figure 3.

Ultrasound imaging performance of RSL3-MBs

(A) Ultrasound contrast signal of RSL3-MBs at different acoustic power in vitro.

(B) Evaluation of UTMD in vivo.

(C) Fluorescence images of ATC tumors dissected from mice in the RSL3-MBs and the RSL3-MBs + UTMD groups.

(D) Quantitative analysis of fluorescence intensity in the RSL3-MBs and RSL3-MBs + UTMD groups (n = 3 per group; mean ± SEM; independent samples t test; and ^#p < 0.05).

At 24 h after injection of RSL3-MBs, tumors were excised from the mice, and targeting efficiency was assessed ex vivo by measuring the indocyanine green (ICG) fluorescence intensity at the tumors using an in vivo imaging system (IVIS). The RSL3-MBs + UTMD group exhibited significantly higher fluorescence intensity than the RSL3-MBs group (p = 0.025) (Figures 3C and 3D). This result indicated that UTMD significantly enhances local delivery efficiency.

The induction of ferroptosis in vivo with the combination of RSL3-MBs and UTMD treatment

Treatment begins on day 7 following subcutaneous tumor inoculation (Figure 4A). On the 12th day, we examined the changes in ferroptosis-related indicators across the different groups. The generation of reactive oxygen species (ROS) initiates lipid peroxidation and eventually results in ferroptosis.³² Therefore, the levels of ROS can serve as an indicator of the degree of ferroptosis. In this study, ROS levels were detected using dihydroethidium and were displayed as red fluorescence. As shown in Figure 4B, almost no red fluorescence was observed in the control group. In contrast, red fluorescence was observed in the RSL3, RSL3-MBs, and RSL3-MBs + UTMD. The RSL3-MBs + UTMD group exhibited a higher red fluorescence intensity compared to the RSL3 and RSL3-MBs groups, suggesting that the combination of RSL3-MBs with UTMD enhanced ROS generation in cells. 4-hydroxynonenal (4-HNE), an end product of lipid peroxidation, can evaluate the degree of ferroptosis.³³ Immunofluorescence results showed that no 4-HNE was detected in the control group. The 4-HNE content in the RSL3-MBs + UTMD group was higher compared to both the RSL3 and RSL3-MBs groups (Figure 4C). This result indicated that the combination of RSL3-MBs and UTMD significantly enhanced lipid peroxidation levels and boosted ferroptosis.

Figure 4.

Evaluation of ferroptosis-related indicators in tumor tissues after different treatments

(A) Schematic diagram of the treatment route (by Figdraw).

(B) The levels of ROS were detected by immunofluorescence staining in tumor tissues at day 12. Scale bars, 100 μm (upper) and 20 μm (lower).

(C) The levels of 4-HNE were detected by immunofluorescence staining in tumor tissues at day 12. Scale bars, 20 μm.

In addition, GPX4 and prostaglandin-endoperoxide synthase 2 (PTGS2) are considered key biomarkers of ferroptosis. GPX4 can directly reduce lipid hydroperoxides to nontoxic lipid alcohols, thereby suppressing ferroptosis.^34,35 PTGS2 upregulation is downstream of the lipid peroxidation that occurs during GPX4-regulated ferroptosis.²⁰ As shown in Figure 5A, the red fluorescence intensity representing GPX4 expression was observed in the following order: RSL3-MBs + UTMD < RSL3 and RSL3-MBs < control. In contrast, the green fluorescence intensity representing PTGS2 expression was observed in the following order: control <RSL3 and RSL3-MBs < RSL3-MBs + UTMD (Figure 5B). Consistent with the immunofluorescence results, western blot analysis showed that GPX4 expression was significantly decreased in the RSL3 group (p = 0.046), the RSL3-MBs group (p = 0.022), and the RSL3-MBs + UTMD group (p = 0.001), compared to the control group. There was no significant difference between the RSL3 and RSL3-MBs groups (p = 0.651). However, GPX4 expression in the RSL3-MBs + UTMD group was significantly lower than that in both the RSL3 group (p = 0.021) and the RSL3-MBs group (p = 0.044) (Figures 5C and 5D). Conversely, PTGS2 expression was significantly upregulated in the RSL3 group (p = 0.001), the RSL3-MBs group (p < 0.001), and the RSL3-MBs + UTMD group (p < 0.001), compared to the control group. There was no significant difference between the RSL3 and RSL3-MBs groups (p = 0.641), whereas PTGS2 expression in the RSL3-MBs + UTMD group was significantly higher than that in both the RSL3 group (p = 0.020) and the RSL3-MBs group (p = 0.043) (Figures 5C and 5E). Taken together, these data suggested that RSL3 treatment could promote ferroptosis in ATC xenografts. Moreover, the combined application of RSL3-MBs with UTMD markedly enhanced the degree of ferroptosis.

Figure 5.

Evaluation of PTGS2 and GPX4 expression levels in tumor tissues after different treatments

(A) The levels of PTGS2 were detected by immunofluorescence staining in tumor tissues at day 12. Scale bars, 20 μm.

(B) The levels of GPX4 were detected by immunofluorescence staining in tumor tissues at day 12. Scale bars, 20 μm.

(C) Western blot images showing PTGS2 and GPX4 expression in tumor tissues after different treatments.

(D and E) Quantification of GPX4 and PTGS2 protein levels based on the western blot bands (n = 3 per group; mean ± SEM; one-way ANOVA; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus control group; ^#p < 0.05 versus RSL3-MBs + UTMD group).

Antitumor efficacy in vivo with the combination of RSL3-MBs and UTMD treatment

Tumor volume was measured from day 0 to day 12. There was no significant difference in the average tumor volume among the groups at the beginning of treatment (p = 0.945). As shown in Figures 6A–6D, the results indicated that the control group exhibited rapid tumor growth, whereas both the free RSL3 and RSL3-MBs groups showed decreased tumor growth. Notably, the RSL3-MBs + UTMD group showed a marked inhibition of tumor growth. Statistical analyses indicated that the tumor volumes in the free RSL3, RSL3-MBs, and RSL3-MBs + UTMD groups were significantly reduced compared to those in the control group (all p < 0.001). The RSL3-MBs + UTMD group showed the best antitumor effect, with tumor volumes significantly smaller than those in the free RSL3 (p = 0.005) and RSL3-MBs (p = 0.015) groups (Figure 6E). The body weight of mice did not show obvious variation among the different groups during the treatment duration, suggesting that the side effects and toxicity of the different treatments were negligible (Figure 6F).

Figure 6.

Assessment of the anti-tumor effect in vivo

(A–D) The tumor volume changes over time after different treatments (n = 5 per group).

(E) Statistical comparison of tumor volumes among different groups (n = 5 per group; mean ± SEM; one-way ANOVA; ^∗∗∗p < 0.001 versus control group; ^#p < 0.05, and ^##p < 0.01 versus RSL3-MBs + UTMD group).

(F) The body weight changes of mice after different treatments (n = 5 per group; and mean ± SEM).

(G) The bioluminescence imaging of tumor growth after different treatments.

(H) Quantification of tumor bioluminescence images after different treatments. Data were plotted as log10 transformed for ease of comparison and visualization (n = 3 per group; mean ± SEM; one-way ANOVA; ^∗∗∗p < 0.001 versus control group; ^#p < 0.05, and ^##p < 0.01 versus RSL3-MBs + UTMD group).

(I) Immunohistochemical staining for Ki-67 expression in tumor tissues after different treatments. Scale bars,100 μm.

(J) Quantification of Ki-67 immunohistochemical staining in tumor tissues after different treatments (n = 3 per group; mean ± SEM; one-way ANOVA; ^∗∗∗p < 0.001 versus control group; and ^##p < 0.01 versus RSL3-MBs + UTMD group).

Furthermore, the in vivo bioluminescence imaging of tumor growth was assessed using the IVIS imaging system on days 0, 6, and 12 (Figure 6G), exhibiting great concordance with the tumor growth curves (Figure 6H). On day 12, the mice were euthanized, and the tumors were excised for Ki67 staining. As shown in Figure 6I, the free RSL3, RSL3-MBs, and RSL3-MBs + UTMD all significantly inhibited tumor cell proliferation compared to the control group (all p < 0.001). Furthermore, the inhibitory effect in the RSL3-MBs + UTMD group was significantly greater than that of the free RSL3 group (p = 0.002) and the RSL3-MBs group (p = 0.007) (Figure 6J). To further assess the safety of different treatments, the major organs of mice were excised for histopathological examination. The hematoxylin-eosin (HE) staining results indicated that the structures of the major organs in all groups remained intact, with no significant inflammatory infiltration, edema, hemorrhage, or other pathological changes, further demonstrating the good biocompatibility of the delivery system (Figure 7A). To further assess the safety of our treatment, serum biochemical parameters, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), and creatinine (CREA), were measured. The liver and kidney function indices in RSLS-MBs + UTMD group were similar to those of the control group (Figures 7B–7E). While slight elevations were observed in the RSL3 and RSL3-MBs groups, these differences were not statistically significant. Overall, although no significant differences were found among the four groups, these results suggest that RSL3-MBs combined with UTMD exert minimal impact on liver and kidney function in ATC-bearing mice.

Figure 7.

Evaluation of the safety of different treatments

(A) H&E staining of major organs of mice in each group at day 12 following treatments. Scale bars, 50 μm.

(B–E) Serum levels of ALT (B), AST (C), BUN (D), and CREA (E) of mice in each group at day 12 following treatments (n = 3 per group; mean ± SEM; and one-way ANOVA).

Discussion

ATC is an extremely aggressive cancer, and there are currently no efficacious treatments available for patients with ATC.¹ Ferroptosis is a recently identified type of programmed cell death, which has a strong association with thyroid cancer prognosis and holds promise as a potential therapeutic target.^15,16,17 Therefore, in this study, we synthesized RSL3-loaded MBs combined with UTMD for the targeted release of RSL3 in tumor tissues, thereby inducing ferroptosis to treat ATC. Our findings demonstrated that the combination of RSL3-MBs and UTMD was a safe and effective therapeutic approach for ATC. This strategy can increase drug concentration in tumor cells, minimize side effects, and has proven to be more effective than the delivery methods using either RSL3 solution or RSL3-MBs alone.

In this study, we used self-prepared MBs, primarily composed of an C₃F₈ gas core and a phospholipid shell made of 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) and distearoylphosphatidylethanolamine with covalently linked polyethylene glycol of molecular weight 2,000 (DSPE-PEG2000). Studies have shown that phospholipid shell materials, DSPC and DSPE-PEG2000, are safe.^36,37 Our experimental results also demonstrated that the injection of MBs does not cause any toxic side effects. In vitro results showed that cell viability was not significantly affected by MBs. In vivo studies indicated that there was no obvious histopathological damage in the kidney, lung, spleen, liver, and heart after MBs’ administration. These findings suggested that MBs and RSL3-MBs exhibited good biosafety both in vitro and in vivo, which is the primary prerequisite for their application. When RSL3-MBs were injected in vivo, an evident ultrasound contrast signal appeared at the tumor site, indicating that synthesized RSL3-MBs have excellent imaging capabilities. The application of UTMD at the tumor location significantly weakened the ultrasound contrast signal, indicating the burst of RSL3-MBs. In summary, the RSL3-MBs we synthesized can be safely utilized for both imaging and ultrasound-guided tumor therapy.

Our research results revealed that RSL3, RSL3-MBs, and RSL3-MBs + UTMD could all induce ferroptosis in ATC xenograft model. Among the three groups, the RSL3-MBs + UTMD group exhibited the highest levels of ROS, 4-HNE, and PTGS2, while the level of GPX4 was the lowest. This indicates that the combination of RSL3-MBs and UTMD triggers ferroptosis in ATC cells more effectively than either RSL3 or RSL3-MBs alone. This may be attributed to the synergistic effect of RSL3-MBs and UTMD, which facilitates enhanced accumulation of RSL3 in the tumor tissue. After the intravenous injection of RSL3-MBs, they flow through the bloodstream to the tumor site. Upon ultrasound irradiation of the tumor site, the interaction of ultrasound with intravascular RLS3-MBs generates inertial cavitation, causing RSL3-MBs to burst and simultaneously perforate the nearby cell membrane.^38,39 This process enables the targeted release of RSL3 and significantly enhances cell membrane permeability, facilitating the intake of RSL3 into the tumor tissue. Therefore, the combination of RSL3-MBs and UTMD increases the concentration of RSL3 in the tumor tissue, inducing more extensive ferroptosis. It is worth noting that the therapeutic benefit of the RSL3-MBs + UTMD group may not derive solely from enhanced intracellular drug delivery via sonoporation. Increasing evidence indicates that UTMD-induced cavitation can generate ROS.^40,41 Prior mechanistic studies have demonstrated that the micro-streaming and shear stress generated by ultrasound can stimulate the formation of superoxide anions and other oxidative species.⁴⁰ Given that RSL3 inhibits GPX4, the key enzyme responsible for detoxifying lipid peroxides, tumor cells become highly susceptible to oxidative damage. Thus, the additional ROS generated by UTMD cavitation may act as an exogenous oxidative stressor that synergizes with RSL3-induced endogenous lipid peroxidation. This cooperative effect may accelerate the accumulation of lipid peroxides beyond the lethal threshold required to trigger ferroptosis. Therefore, UTMD-induced ROS production may further potentiate lipid peroxidation in tumor cells and thereby amplify the ferroptotic response initiated by RSL3.

Subsequently, we further investigated whether inducing ferroptosis could inhibit the growth of ATC tumors. Compared with the control group, RSL3, RSL3-MBs, and RSL3-MBs + UTMD groups all could inhibit tumor growth. The RSL3-MBs + UTMD group exhibited the most effective outcome, with tumor volume remaining nearly unchanged over 12 days. This may be attributed to both the RSL3-induced and UTMD-enhanced ferroptosis-based targeted therapy. Previous studies have only demonstrated that RSL3 can inhibit ATC cell viability at the cellular level.¹⁹ Currently, there are no studies for systematically evaluating the antitumor potential of RSL3 on ATC in animal models. This study investigated the antitumor potential of RSL3 for ATC in vivo. Furthermore, we designed an RSL3-MBs + UTMD delivery system, which significantly increased the concentration of RSL3 in the tumor tissue, expanded the extent of ferroptosis, and markedly improved the antitumor effect of ATC. Our results showed no obvious changes in the body weight of mice in the RSL3, RSL3-MBs, and RSL3-MBs + UTMD groups during the treatment period. Additionally, histopathological analysis revealed no evident damage in the main organs of mice across different groups. These results suggest that the various treatment methods do not exhibit obvious toxicity or adverse effects, and the RSL3-MBs + UTMD delivery system shows high anti-cancer efficiency against ATC and good safety in vivo.

Ferroptosis has rapidly emerged as a promising therapeutic vulnerability across multiple malignancies. To overcome the poor solubility and systemic toxicity of ferroptosis inducers such as RSL3, numerous stimuli-responsive nanoplatforms have been developed. For example, Chen et al. designed a glutathione (GSH)-responsive self-assembled nano-prodrug for the co-delivery of chemotherapeutics, ferrocene, and RSL3, which utilizes high intratumoral GSH levels to trigger drug release.²⁶ Similarly, Gao et al. developed arachidonic acid-conjugated polymer micelles that release RSL3 upon free radical-triggering in the tumor microenvironment, effectively eliminating drug-tolerant persister cells.⁴² These systems rely on endogenous biochemical triggers and passive processes for tumor accumulation. In contrast, our UTMD-based strategy provides active, externally controlled, spatiotemporally precise delivery. By encapsulating RSL3 directly within MBs and acoustically rupturing them only within the insonated tumor region, we achieve site-specific drug release. The UTMD-induced sonoporation further facilitates transvascular and transmembrane transport, partially overcoming physical barriers such as high interstitial fluid pressure that impede conventional nanoparticles. Moreover, the additional ROS generated by UTMD cavitation may act as an exogenous oxidative stressor that synergizes with RSL3-induced endogenous lipid peroxidation, thereby amplifying ferroptosis. UTMD has been widely applied to enhance intratumoral drug delivery of chemotherapeutics and biological agents across diverse tumor types.^43,44,45,46 However, in conventional co-administration approaches, the therapeutic effects remain spatially limited to the close proximity of the insonated MBs, and the perturbations in cell membranes caused by ultrasound are very short-lived.⁴⁷ Our strategy encapsulates RSL3 directly within MBs and overcomes these limitations by enabling ultrasound-triggered, site-specific drug release at the tumor site, thereby improving delivery precision and reducing off-target exposure. Recent work using immunomodulatory MBs targeting integrin αvβ3, combined with immune checkpoint blockade, has demonstrated potent theranostic potential in mouse models of ATC.⁴⁶ This biologically targeted MB platform achieves selective vascular adhesion and immune microenvironment modulation, whereas UTMD provides deep tissue penetration and acoustic control of drug release. Consequently, future research could aim to converge these strategies, developing ligand-functionalized, RSL3-loaded MBs that combine biological targeting with UTMD, thereby maximizing therapeutic outcomes.

Despite these promising preclinical results, UTMD-mediated therapy still faces several obstacles to clinical translation. Given the high metastatic potential of ATC, it remains an important question how ultrasound can accurately locate and target occult metastatic lesions. Subsequent studies should improve and fine-tune MB design, including surface modification with tumor-targeting ligands, to enhance recognition and accumulation in hidden or metastatic lesions. For deep-seated tumors, ultrasound waves undergo severe attenuation during penetration, which may limit the energy available to trigger MB rupture. Therefore, ultrasound parameters should be carefully optimized, and advanced imaging modalities, such as computed tomography (CT), magnetic resonance imaging (MRI), or photoacoustic imaging, should be integrated to improve localization accuracy and ensure adequate acoustic energy delivery to deep or anatomically complex tumor sites.

In conclusion, we successfully constructed RSL3-loaded MBs with good biocompatibility and excellent contrast performance for contrast-enhanced ultrasound imaging of ATC tumor. Our results indicate that the combination of RSL3-MBs and UTMD significantly enhances the delivery efficiency of RSL3 to tumor tissue and markedly suppresses ATC growth, with no obvious toxicity. The underlying mechanism involves RSL3-induced inactivation of GPX4, thereby triggering ferroptosis in ATC cells. Meanwhile, the acoustic pore effect generated by ultrasound irradiation enhances vascular permeability and creates transient pores in the cell membrane, facilitating the efficient transmembrane delivery of RSL3 into tumor cells. This study provides an effective treatment approach for ATC and establishes an experimental foundation for the application of RSL3 in ATC treatment. Future studies could further optimize the design of MBs to enhance their targeting capability to ATC. In addition, further experimental studies are needed to explore the combination of RSL3-MBs + UTMD with other therapeutic strategies, such as immunotherapy or chemotherapy, to enhance the therapeutic effect.

Limitations of the study

One limitation of the study is insufficient observation period for the mice after treatment, which restricted the investigation of the effects of various treatments on mouse survival rates and tumor metastasis. Future studies can further explore the effects of RSL3-MBs combined with UTMD on survival rates and tumor metastasis in ATC model mice.

Resource availability

Lead contact

Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Bo Zhang (thyroidus@163.com).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •Data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (grant 22278449, U24A20731, 82272111, and 92159303), Beijing Natural Science Foundation (grant 7252292), and the National High Level Hospital Clinical Research Funding (grant 2023-NHLHCRF-BQ-25).

Author contributions

Methodology, software, investigation, data curation, writing – original draft, and visualization, Xinyao Liu; investigation, data curation, and visualization, L.W.; formal analysis, software, and data curation, Xinyi Liu; conceptualization, resources, writing – review and editing, supervision, project administration, and funding acquisition, Y.D.; conceptualization, writing – review and editing, supervision, project administration, funding acquisition, and validation, B.Z.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
rabbit anti-GPX4	Abcam	Cat#ab125066; RRID:AB_10973901
rabbit anti-PTGS2	Servicebio	Cat#GB11077-1; RRID:AB_3719447
rabbit anti-Ki67	Servicebio	Cat# GB111499; RRID:AB_2927572
mouse anti-PTGS2	Proteintech	Cat# 66351-1-Ig; RRID:AB_2881731
mouse anti-4-HNE	Thermo Fisher Scientific	Cat# MA5-27570;RRID:AB_2735095
Biological samples
Mouse blood and tissue samples	BALB/c nude mice	Not applicable
Chemicals, peptides, and recombinant proteins
RSL3	Selleck Chemicals	Cat# S8155
DSPE-PEG2000	Ruixi Biotechnology Co., Ltd.	Cat# R-1028-2K
DSPC	Ruixi Biotechnology Co., Ltd.	Cat# LP-R4-076
DHE	Sigma-Aldrich	Cat# D7008
DSPE-ICG	Ruixi Biotechnology Co., Ltd.	Cat# R-DSPE-ICG
Critical commercial assays
Cell Counting Kit-8	DoJindo	Cat#AQ308
Experimental models: Cell lines
CAL-62 cells	Procell Life Science &Technology Co. Ltd.	Cat#CL-0618
Nthy-ori 3-1 cells	Procell Life Science &Technology Co. Ltd.	Cat#CL-0817
Experimental models: Organisms/strains
Mouse:BALB/c nude mice	Beijing Vital River Laboratory Animal Technology Co., Ltd.	RRID:IMSR_CRL:194
Software and algorithms
ImageJ	NIH	RRID:SCR_003070
GraphPad Prism	GraphPad Software	RRID: SCR_002798
SPSS	IBM	RRID:SCR_002865
Adobe Photoshop	Adobe Systems	RRID:SCR_014199

Experimental model and study participant details

Animals

Female BALB/c nude mice (4–5 weeks old) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). The animals were housed in specific pathogen-free conditions with food and water freely available, and maintained on a 12-h light/dark cycle under standard humidity and temperature. All animal experiments were approved by the Animal Ethics Committee of China-Japan Friendship Hospital (Approval Number: ZRDWLL230107) and were conducted in accordance with relevant institutional and national guidelines. Only female mice were used in this study. Therefore, sex-based comparisons were not performed, and potential sex-specific effects could not be assessed, which represents a limitation of the study.

Cell lines

The human thyroid follicular epithelial cell line Nthy-ori 3-1 (CL-0817) and the human anaplastic thyroid cancer cell line CAL-62 (CL-0618) were obtained from Procell Life Science & Technology Co., Ltd. The CAL-62 cell line was established from a 70-year-old woman and was cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). The Nthy-ori 3-1 cell line was established from a 35-year-old woman and was maintained in Roswell Park Memorial Institute 1640 (RPMI-1640) medium supplemented with 10% FBS. Both cell lines were incubated at 37 °C in a humidified atmosphere containing 5% CO₂. All cell lines used in this study were authenticated by short tandem repeat (STR) profiling and were routinely tested for mycoplasma contamination. All tests confirmed that the cell lines were mycoplasma-negative prior to use.

Construction of xenograft tumor model and groups

The subcutaneous ATC model was established using female BALB/c nude mice. Luciferase-expressing CAL-62 cells during the logarithmic phase were digested and resuspended in PBS at a concentration of 1 ×10⁸ cells/mL. A subcutaneous injection of 100 μL of cell suspension was performed in the inguinal area of mouse. The mice were randomly assigned to four groups: control, RSL3, RSL3-MBs, RSL3-MBs + UTMD (n = 9 in each group). Mice were treated 7 days after tumor implantation, following confirmation of tumor formation by bioluminescence imaging. 100 μL of RSL3 or RSL3-MBs (equivalent RSL3 concentrations: 1.0 mg/mL) was injected via the tail vein every two days for a total of four doses. The tumors in the fourth group were exposed to ultrasound irradiation immediately (frequency: 3.0 MHz; pulse repetition frequency: 20 HZ; pulse length: 4.0 cycle; pulse time: 1 s; time interval: 5 s; acoustic power: 60%; duration: 3 min).

Method details

Synthesis of RSL3-MBs

The MBs synthesis method was based on previously literature.^36,48 First, ethanol was used as a solvent to dissolve 0.7 mg of 1,2-Distearoyl-sn-glycero-3-phosphocholine (DSPC) and 0.3 mg of 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(poly-ethylene glycol)-2000] (DSPE-PEG2000). The resulting solution was then gradually added to 800 μL of deionized water in a sonication water bath. Second, 1.3 mg of RSL3 (S8155, Selleck Chemicals, USA) was dissolved in dimethyl sulfoxide (DMSO) and injected into the above solution in a sonication water bath. Third, dialysis was used to remove the organic solvent. Finally, the dialysate was transferred into a 3 mL glass vial and diluted to 1 mL with deionized water, after which the vial was filled with C₃F₈. The RSL3-MBs were acquired following 30 s of mechanical shaking. The same process was used to prepare MBs without including RSL3.

Characterizations

The morphology of MBs and RSL3-MBs were evaluated by a fluorescence microscope (Leica, Germany). The hydrodynamic diameter of MBs and RSL3-MBs were determined by an Accusizer particle size analyzer (Particle Sizing Systems, USA). The zeta potential of MBs and RSL3-MBs were assessed using a Zetasizer particle analyzer (Malvern Panalytical, UK). The amounts of RSL3 in RSL3-MBs was measured by high-performance liquid chromatography (HPLC, Agilent, USA). The ultrasound imaging was performed using the VINNO 70 diagnostic US scanner (Vinno Technology Co. Ltd., China).

Biosafety assay

The human thyroid follicular epithelial cell line Nthy-ori 3-1 cells (CL-0817, Procell Life Science &Technology Co. Ltd., China) were plated in 96-well plates and cultured overnight. MBs and RSL3-MBs were added at different concentrations, followed by incubated for another 24 h. Cell viability was assessed using the Cell Counting Kit-8 (CCK-8, DoJindo, Japan) in accordance with the manufacturer’s protocol. Fresh medium was added to each well, followed by the addition of CCK-8 solution and incubation at 37 °C for 2 h. Absorbance was recorded at 450 nm to determine cell viability.

Next, histological analysis was performed to evaluate the in vivo biosafety of bothMBs and RSL3-MBs. 100 μL of MBs or RSL3-MBs was administered intravenously to healthy female BALB/c nude mice at the dosage used in the following experiment. On day 12, the mice were euthanized to remove the main organs, followed by HE staining for histological examination.

Ultrasound imaging performance

We first evaluated the ultrasound imaging performance of RSL3-MBs in vitro using a clinical ultrasound system (VINNO70, China). A total of 100 μL of RSL3-MBs was added into a water tank filled with degassed water, mixed thoroughly, and briefly left to stand. The ultrasound probe was then immersed in the microbubble-containing water, and the region of interest was adjusted to an appropriate range. The UTMD mode was activated (frequency: 3.0 MHz; pulse repetition frequency: 20 HZ; pulse length: 4.0 cycle; pulse time: 1 s; time interval: 5 s; duration: 3 min), and the MBs were exposed to varying levels of acoustic power (0%, 20%, 40%, and 60%) to observe the extent of MB disruption within the acoustic field.

Subsequently, the in vivo ultrasound performance of RSL3-MBs was examined. 100 μL of RSL3-MBs was injected via the tail vein into xenograft tumor-bearing mice.⁴⁸ Ultrasound irradiation was immediately applied (frequency: 3.0 MHz; pulse repetition frequency: 20 HZ; pulse length: 4.0 cycle; pulse time: 1 s; time interval: 5 s; acoustic power: 60%; duration: 3 min). The in vivo imaging performance of RSL3-MBs and the efficiency of RSL3-MBs destruction by UTMD were assessed.

Fluorescence molecular imaging

To evaluate the targeting of RSL3-MBs combined with UTMD, ICG was conjugated to the MBs as a surrogate tracer. Briefly, 0.01 mg of DSPE-ICG was thoroughly mixed under an ultrasonic water bath and added to a solution containing DSPC, DSPE-PEG2000, and RSL3. Mice were randomly assigned to the RSL3-MBs group or the RSL3-MBs + UTMD group (n = 3 per group). After tail vein injection of RSL3-MBs, UTMD was immediately applied to the subcutaneous tumors in the RSL3-MBs + UTMD group. At 24 h after injection, tumors were excised from the mice, and targeting efficiency was assessed ex vivo by measuring the ICG fluorescence intensity at the tumors using an IVIS imaging system (PerkinElmer, USA).

Antitumor effect

To investigate antitumor efficacy of UTMD drug delivery system, the mice were randomly assigned to four groups: control, RSL3, RSL3-MBs, RSL3-MBs + UTMD (n = 9 in each group). Treatment begins on day 7 following subcutaneous tumor inoculation. 100 μL of RSL3 or RSL3-MBs (equivalent concentrations of RSL3: 1.0 mg/mL) was injected via the tail vein every two days for a total of four doses. The tumors in the fourth group were exposed to ultrasound irradiation immediately (frequency: 3.0 MHz; pulse repetition frequency: 20 HZ; pulse length: 4.0 cycle; pulse time: 1 s; time interval: 5 s; acoustic power: 60%; duration: 3 min).

Following different treatments, the tumor volumes and weights of the mice were assessed every 2 days. Tumor volume was calculated as 0.5 × width² × length.⁴⁹ Tumor growth was monitored using IVIS imaging system for bioluminescence imaging. After a period of 12 days, the mice were euthanized by cervical dislocation, and both the tumors and major organs were excised for subsequent histological examination. Blood was collected from each group of mice to assess liver and kidney function, including ALT, AST, BUN, and CREA. The dihydroethidium (DHE) staining was used to measure ROS level. Immunofluorescence, immunohistochemistry and western blot were used to detect the protein expression.

HE staining

Tissue sections were immersed in hematoxylin solution for 3–5 min, followed by rinsing with tap water. Subsequently, the sections were treated with a differentiation solution and rinsed again with tap water. Afterward, the sections were exposed to a hematoxylin bluing solution and rinsed again. The sections were then immersed in 95% ethanol for 1 min, stained with eosin for 15 s, and dehydrated through graded ethanol. Finally, the sections were cleared in xylene and sealed with neutral gum.

ROS staining

ROS levels were detected using DHE (D7008, Sigma-Aldrich, USA). The frozen sections were rewarmed at room temperature, and excess moisture was removed. The sections were then treated with an autofluorescence quencher for 5 min. Subsequently, the sections were incubated with DHE dye at 37 °C in a dark incubator for 30min. Afterward, the sections were incubated with 4′,6-Diamidino-2-phenylindole (DAPI) staining solution in the dark at room temperature for 10 min. Finally, the sections were sealed with an anti-fade mounting medium.

Immunofluorescence

After antigen retrieval, the tissue sections were blocked with 3% bovine serum albumin (BSA) at room temperature for 30 min to prevent non-specific binding. The primary antibodies, including 4-HNE (MA5-27570, Thermo Fisher Scientific, USA), rabbit anti-GPX4 (ab125066, Abcam, USA) and rabbit anti-PTGS2 (GB11077-1, Servicebio, China), were applied to the sections, which were then placed flat in a humidified chamber and incubated overnight at 4°C. The next day, corresponding secondary antibodies were added and incubated for 50 min at room temperature in the dark. DAPI solution was then applied for nuclear staining and incubated for 10 min at room temperature in the dark. After that, autofluorescence quencher solution was added for 5 min, followed by rinsing under running water for 10 min. Finally, the sections were coverslipped with anti-fade mounting medium.

Immunohistochemistry

After antigen repair, endogenous peroxidase activity was blocked by incubating the tissue sections in 3% hydrogen peroxide solution at room temperature for 25 min in the dark. The sections were then blocked with 3% BSA for 30 min at room temperature. After gently removing the blocking solution, the sections were incubated with a rabbit anti-Ki67 primary antibody (GB111499, Servicebio, China), and placed in a humidified chamber overnight at 4°C. The next day, the sections were incubated with a secondary antibody at room temperature for 50 min. After thorough washing, 3,3′-diaminobenzidine (DAB) chromogenic detection and hematoxylin counterstaining were performed sequentially. Finally, the sections were dehydrated, cleared in xylene, and mounted.

Western blotting

Briefly, tissues were lysed in nondenaturing lysis buffer (C1050, Applygen, China) at 4 °C for 30 min. The extracted proteins were then mixed with loading buffer and denatured by boiling at 100 °C for 5 min. Subsequently, proteins were separated by electrophoresis on a 10% Tris-Glycine gel and transferred onto polyvinylidene fluoride membranes (ISEQ00010, Millipore, USA). After blocking with 5% skim milk for 1 h, the membranes were incubated with the primary antibodies overnight at 4 °C. Primary antibodies included anti-GPX4 (ab125066, Abcam, USA) and anti-PTGS2 (66351-1-Ig, Proteintech, China). On the following day, the membranes were incubated with corresponding secondary antibodies at room temperature for 1 h. The protein signals were visualized using an enhanced chemiluminescence substrate and ultrasensitive multifunctional imager.

Quantification and statistical analysis

Statistical analyses were performed using SPSS 25.0 software (SPSS Inc., IBM, NY, USA). All data were presented as mean ± SEM. An independent samples t test was conducted to compare two independent groups. One-way analysis of variance (ANOVA) was used for multiple comparisons. A p value <0.05 was considered statistically significant. Asterisks denote statistical significance compared with the control group (^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001), while hash symbols indicate comparisons with the RSL3-MBs + UTMD group (^#p < 0.05, ^##p < 0.01, ^###p < 0.001).

Published: February 10, 2026