Nanotherapeutics for hydrogen sulfide-involved treatment: An emerging approach for cancer therapy

Abbreviations

3-MST

3-mercaptopyruvate sulfurtransferase

ABMD

9,10-anthracenediyl-bis(methylene) dimalonic acid

ALs

ADT-encapsulated liposomes

AML

ADT-loaded magnetic nanoliposome

ATP

adenosine triphosphate

BAD-NEs

bovine serum albumin/α-linolenic acid (ALA)/diallyl disulfide (DADS) nanoemulsions

Bax

BCL2 associated X protein

Bcl-2

B-cell lymphoma 2

BODIPY

boron dipyrromethene

CaCO₃

calcium carbonate

CAT

catalase

CBS

cystathionine-β-synthase

CCH

CaCO₃-coated and HA-decorated Cu₂O nanoparticles

CDT

chemodynamic therapy

CIT

ion-interference therapy

COF

Cu(ii)-porphyrin-derived nanoscale covalent organic frameworks

COX IV

cytochrome c oxidase

CRC

colorectal cancer

CSE

cystathionine γ-lyase

(Cu₂(ZnTcpp)·H₂O) _n /(NP-1)

cupric ion and zinc organic ligands-containing metal-framework nanoparticles

Cu²⁺

cupric ion

Cu₂O

cuprous oxide

Cu-MOF/HKUST-1

copper metal-organic framework-constructed enzyme in nanoscale

CuS/Cu₉S₈

copper(ii) sulfide

CY

cyanine

EA-Fe@BSA NPs

ellagic acid-Fe-bovine serum albumin nanoparticles

EPR

enhanced permeability and retention effect

ESR

electron spin resonance

ETC

electron transport chain

Fe_1−x S-PVP NPs

polyvinyl pyrrolidone modified iron sulfide nanoparticles

Fe²⁺

iron(ii) ion

FeOOH NSs

iron oxide-hydroxide nanospindles

FeS@BSA

ferrous sulfide-embedded bovine serum albumin nanoclusters

GPX4

glutathione peroxidase 4

GSH

glutathione

H₂O₂

hydrogen peroxide

H₂S

hydrogen sulfide

HepG2 cells

human hepatocellular liver carcinoma cells

IC₅₀

half-maximal inhibitory concentration

ID

iron-dextrin

l-Cys

l-cysteine

LoVo

human colorectal adenocarcinoma cell

MRI

magnetic resonance imaging

MSNP-N₃-FA

folic acid-anchored azide functionalized mesoporous silica nanoparticles

N₃

azide

NaHS

sodium hydrosulfide

Nano-PT

phenyl tail small molecules

NEM

N-ethylmaleimide

NIR

near-infrared

NPs@BOD/CPT

nanoparticles loaded with boron-dipyrromethene dye (InTBOD-Cl) and Camptothecin-11 (CPT-11)

NSAIDs

nonsteroidal anti-inflammatory drugs

^·OH

hydroxyl radicals

¹O₂

singlet oxygen

PA

photoacoustic

PAs

photothermal agents

PBS

phosphate-buffered saline

PCA

PSBMA zwitterionic nanoparticles (zwitterionic sulfobetaine methacrylate (SBMA) and N,N′-bis-(acryloyl) cystamine (BAC)) loaded with l-Cys and α-CHCA

PCM

illumination sensitive phase change material

PDT

photodynamic therapy

PSD

polysulfide

PTT

photothermal therapy

ROS

reactive oxygen species

RT

radiotherapy

S^2-

sulfide ion

SAM

S-adenosyl-l-methionine

SeChry@PUREG4-FA

selenium chrysin encapsulated folate-targeted polyurea dendrimer generation four nanoparticles

SH-ASA/Cur-coloaded mPEG-PLGA NPs

methoxy poly(ethylene glycol)-poly(lactide-coglycolide) nanoparticles loaded with SH-aspirin (SH-ASA) and curcumin (Cur)

SPIOs-ADT-LPs

magneto-acoustic nanoliposomes loaded with superparamagnetic iron oxide nanoparticles (SPIOs) and anethole trithione (ADT)

SSS

self-assembly

T2

transverse relaxation time

TAMs

tumor-associated macrophages

TME

tumor microenvironment

US

ultrasound

VDAC 1

voltage-dependent anion-selective channel

VZnO

virus-like silica nanoparticles coated with Zinc oxide

ZnS@SiO₂

silica nanofiber coated zinc sulfide nanoparticles loaded with doxorubicin (DOX)

ZSZIT

zeolitic imidazolate framework-8 (ZIF-8) coated zinc sulfide nanoparticles loaded with tirapazamine (TPZ) and indocyanine green (ICG)

1 Introduction

Cancer, one of the most devastating diseases in human history, had led to approximately 19.3 million diagnosed cases and 10.0 million deaths in the year 2020 while the prevalence is still growing globally [1]. It is undoubtedly a serious impediment to increasing human life expectancy. The current modalities for treating cancers include surgery, chemotherapy, radiotherapy (RT), and immunotherapy. Among these, post-operative recurrence/metastasis and drug resistance to chemotherapy have been the two main causes of death in cancer patients so far [2]. Immunotherapy has been considered a potential anticancer treatment since the last decade [3,4,5], but the narrow therapeutic spectrum, autoimmunity-induced systematic responses, and high cost of the therapy are all obstacles hampering its clinical application [3,4]. Moreover, the hypoxic tumor microenvironment (TME) possesses nature resistance to RT [6], photodynamic therapy (PDT) [7], chemotherapy [8], and sonodynamic therapy [9]. Under such conditions, cancer still remains one of the major drivers of mortality worldwide, making the discovery and development of advanced therapies imperative [10].

It is well known that mammalian cells produce lots of gas molecules that can act as transmitters or signaling molecules in a variety of biological activities [11]. The endogenous gasotransmitters including nitric oxide (NO), carbon monoxide (CO), and hydrogen sulfide (H₂S) play important roles in promoting the growth, proliferation, and metastasis of cancer cells [12]. Recent research studies indicated that regulating the gasotransmitters can deplete the energy of tumor cells within a short period by binding to the heme iron in proteins, specifically the mitochondrial hemoglobin (known as the anti-Warburg effect), thereby inhibiting the proliferation and accelerating apoptosis [12]. Furthermore, some gas molecules like NO can augment the effects of chemotherapies in multidrug-resistant cancers by inhibiting the expression of P-glycoprotein [13,14]. The idea of introducing exogenous gas molecules into the tumor tissue opened up a new frontier in the field of cancer research, which is the so-called gas therapy. Gas therapy is a novel therapeutic strategy that can be used in numerous disease states such as inflammation, bacterial infections, and cardiovascular diseases [15,16,17]. Previously, different reviews have demonstrated the efficacy of CO- and NO-mediated nanotherapies [18,19]. When used at appropriate concentrations, these molecules have several advantages over chemotherapeutic agents, such as lower toxicity (except for H₂S) and less drug resistance. The gas can be given via different forms of donors to improve delivery efficiency and safety. It is therefore highly versatile and has become more popular over the past decade [20].

H₂S was first identified as a harmful gas before the discovery of its existence in mammalian cells [21], which rendered it a key endogenous molecule in both physiological and pathological conditions of the human body [22]. For example, it can increase the activity of vascular endothelial growth factor receptor 2 to promote angiogenesis, raise the level of glutathione (GSH) to protect the organs from oxidative damage, and participate in the sulfhydration of nuclear factor kappa-B (NF-κB) to regulate inflammatory responses [23,24,25]. Based on these special characteristics, H₂S was defined as the third gasotransmitter after NO and CO in 2002 [26]. The concentration of H₂S is abnormally high in certain types of cancer cells, and thus, H₂S can be used as a biomarker to guide the cancer diagnosis and treatment [27,28]. Indeed, the application of H₂S-involved treatment strategies has become an area of interest, possibly owing to the undesirable therapeutic outcomes and limitations of other anticancer modalities. High target specificity can be realized by H₂S-activated anticancer therapies, which decreases the risk of cytotoxicity on non-malignant cells [29,30,31]. In contrast to conventional oncotherapy, H₂S-involved therapy is highly plastic with different triggering mechanisms and extensive targets, such as mitochondria, catalase (CAT), cytochrome c, and the redox balance of tumor cells [32,33,34,35,36]. It can also be applied synergistically with other treatments like phototherapy, immunotherapy, and radiotherapy as an adjunct to potentiate the overall therapeutic effects.

Notably, the impact of H₂S on oncogenesis is complex and thus cannot be simply depicted as positive or negative. A bell-shaped dose-response curve was proposed, which concluded that endogenous H₂S or low level of donated H₂S may have promoting effect on cancer cell proliferation, while relatively high level or long exposure of exogenous H₂S can be cytotoxic [37,38]. For instance, Wu et al. showed that NaSH (25–100 μM) facilitated vessel formation and tumor growth of hepatocellular carcinomas, both of which were significantly suppressed when the concentration of NaSH was increased to 800 μM [39]. Due to the fact that H₂S participates in maintaining the redox balance of tumor cells as a reducing agent, some research studies reported that inhibiting the synthesis of H₂S and depleting the existing H₂S can damage the redox system of tumor cells and thus induce death [40,41]. In this sense, inhibiting the endogenous H₂S production or donating exogenous H₂S at the cancer site are both feasible treatment directions that can severely impact the cell growth and proliferation by damaging the self-repairment of cancer cells [12,37]. The antitumor effect of regulating the endogenous H₂S concentration was also proved by previous studies [40,42].

Despite these merits of therapeutic gas molecules, the in vivo application of H₂S remains challenging due to its intrinsic defects of low diffusivity, poor solubility, and uncontrollable releasing profile, which collectively lead to low therapeutic efficacy and potential adverse reactions [43]. In terms of administration, the direct inhalation route is not available for H₂S due to its toxic nature [20]. To further explore the use of H₂S, several organic donors were designed and tested, such as hydrolysis-based donors (e.g., dithiolethione derivatives and GYY4137), light-triggered donors (e.g., ketoprofenate), natural compounds (e.g., diallyl trisulfide), enzyme-triggered donors [44], and thio-triggered donors (e.g., polysulfides and N-benzoylthiobenzamides). Based on the principles of photolysis, thiol/-enzyme activation, and hydrolysis, exogenous or endogenous triggers were coupled to optimize the H₂S-releasing property [20]. A variety of H₂S donors have been applied in clinical trials for treating different kinds of medical conditions, such as heart failure, visceral pain, and chronic kidney disease (Table 1). However, limitations like inherent functional group toxicity, premature or unintended gas release, and rapid systemic clearance due to low molecular weight still exist [45].

Delightfully, large progress has been made in the delivery of anticancer drugs with the development of nanotechnology [46]. A growing body of evidence suggests that the integration of nanotechnology is associated with a greater drug delivery efficiency and improved drug release profile compared to large-scale particles. For instance, certain nanoparticles can prolong the plasma circulation time, improve the water solubility, achieve the sustained release of long-acting drugs, and promote target specificity [47]. During the last decade, H₂S donor-loaded nanoplatforms capable of delivering H₂S to the target sites have been intensively studied and exhibited profound antitumor effects [32,34,48,49]. To minimize the toxic effects of H₂S in normal tissues, several metal sulfide nanomaterials (e.g., manganese(ii) sulfide, zinc sulfide, and iron(ii) sulfide) were designed with an area-restricted release in the TME. The decomposition or degradation of these metal sulfide nanomaterials elevates the H₂S concentration in tumor tissues and liberates metal ions to achieve other therapies at the same time, for example, chemodynamic therapy (CDT). On top of increasing the level of H₂S, a reduction in the concentration also generates an antitumor effect according to the Janus-faced pharmacological characteristic of H₂S. More interestingly, the photothermal effect or photodynamic effect can be achieved through metal sulfide nanomaterials (such as copper sulfide and iron sulfide) formed in the H₂S-rich environment in colon cancer [50,51]; the H₂S-releasing nanotherapeutics can inhibit tumor growth by damaging the cytoskeleton of tumor cells under ultrasound. In general, the therapeutic effects of H₂S become much more diverse with the support of nanotechnology.

Although the use of H₂S-involved nanotherapeutics has been gradually acknowledged, there have been few reviews classifying the mechanism of actions of those nanoparticles and emphasizing the role of H₂S in the therapies. Therefore, it is necessary to discuss and summarize the latest advances in this study area to guide and boost future research. This review outlines the recent developments of H₂S-involved anticancer nanodrugs and also provides a comprehensive view of their therapeutic activities, including H₂S-amplified gas therapy, H₂S-depleted chemotherapy, phototherapy, chemodynamic therapy, immunotherapy, and ultrasonic therapy (Figure 1 and Table 2).

Figure 1

Schematic illustration showing the classification of H₂S-involved nanotherapeutics for cancers.

2 Physicochemical properties of H₂S

Sulfur is an abundant element in nature, which accounts for 0.048% (mass fraction) of the Earth’s crust. It is a reactive element that can form compounds with other elements (except inert elements, iodide, and molecular nitrogen) at valence states of −2, +6, +4, +2, and +1. Among these, S²⁻ ion shows the strongest reducing ability [52]. H₂S is a small inorganic molecule formed by an S²⁻ ion and two protons, which has a boiling point of 0°C at ambient temperature and atmospheric pressure [44]. Therefore, it is in gaseous state at room temperature. In the saturated aqueous solution, H₂S is weakly acidic with a pKa of 0.98, which exists in the equilibrium of H₂S/HS⁻/S²⁻. Under the physiological environment (pH = 7.4, 37°C), 40% of the sulfides within unsaturated H₂S solution are in the form of H₂S, while the rest exist in the form of HS⁻. Due to the low molecular weight, H₂S can easily pass through the biological membranes without the aid of lipophilic paracrine signaling molecules that act as transporter proteins [53,54]. Thus, H₂S can interfere with signaling pathways and exert biological effects in the following ways: (a) reacting with reactive oxygen species (ROS) to maintain the redox balance of cells [55,56,57], (b) participating in protein persulfidation [58,59], (c) interacting with S-nitrosothiols [60,61,62], (d) inducing sulfhydration of electrophiles [63,64,65], (e) reacting with metal centers of the enzymes [66,67,68], and (f) engaging in the cellular metabolism of lactic acid and protons [69,70]. Apart from the aforementioned actions, ongoing research is now focusing on probing the functions of H₂S in biological activities, so that more precise treatment strategies can be made to target H₂S-associated diseases.

H₂S or sulfur in univalent or multivalent forms can react with metal ions to form sulfide ore to be stored in the Earth’s crust. Lately, researchers have found that nanosized metal sulfide compounds prepared by specific fabrication approaches (such as solvothermal approach, template-assisted approach, and biomineralization approach) exhibit special physical and chemical properties, such as Fenton catalysis, light conversion, radiation enhancement, and immune activation abilities [71,72,73,74,75]. These excellent features may well be adopted in tumor treatments [76]. In the past ten years, metal sulfide nanomaterials inaugurated a new era in the field of cancer therapy as illustrated in our previous review [77]. Moreover, H₂S donors or nanoplatforms can produce nanoscale H₂S bubbles with the catalysis of enzymes that exert powerful mechanical forces under ultrasound to eradicate tumors [78,79]. The H₂S-involved treatment strategies described later are all based on the above physicochemical properties of H₂S.

3 H₂S-amplified gas therapy

Gas therapy is the usage of gas molecules with special therapeutic effects in disease treatment. In the context of cancer therapy, hydrogen, NO, CO, and H₂S are the most representative gas signaling molecules. Different from other gases, the toxicity of H₂S renders the application of nanostructured H₂S donors essential as they can achieve target-specific H₂S accumulation. The significance of H₂S-amplified gas therapy in cancer treatment has also been underscored by previous research studies [80]. However, most of the previous studies were general without classifying the nanotherapeutics on the basis of fundamental mechanisms. In this section, we will discuss the action modes of nanotherapeutics based on four areas, including mitochondrial dysfunction, CAT inhibition, cytochrome c oxidase inhibition, and lactate metabolism disruption.

3.1 Mitochondrial dysfunction

It is well-known that mitochondrion is the “powerhouse” of the cell. The electron transport chain in the inner membrane syntheses adenosine triphosphate (ATP) via five protein complexes coupled with substrates from the Krebs cycle to function as a coordinated respiratory system. Although some types of tumor cells can generate ATP from anaerobic glycolysis even in the presence of sufficient oxygen (known as the Warburg effect), the energy provided by mitochondria is still essential for the survival of cancer cells [81]. Thus, the loss of ATP supply after treating with H₂S would benefit the decrease in cellular metabolism, leading to the starvation and death of cancer cells [82]. Apart from that, H₂S also interferes with mitochondrial apoptotic pathways where various functional proteins are involved.

Li et al. reported a nanotheranostic approach for triple-negative breast cancer [32]. The CY-PSD nanoparticles comprised of polysulfide (PSD) and cyanine (CY) achieved tumor-targeted drug delivery concurrently with real-time gas monitoring. TME contains a higher level of thiols than the physiological environment, which can boost the H₂S production via specific thiol-polysulfide bond cleavage at the lesion site (Figure 2a). With the gradual addition of GSH, H₂S was then produced from PSD nanoparticles (50 × 10⁻⁶ M) (Figure 2b). The formed H₂S accounted for the changes in several cellular activities in the mitochondria. For example, the alternations in the level of functional proteins like abnormal regulation of membrane protein (e.g., long-chain specific acyl-CoA dehydrogenase) that is involved in fatty acid oxidation and overexpression of enzymes responsible for cell metabolism (e.g., voltage-dependent anion-selective channel 1 (VDAC 1)). The highly expressed VDAC 1 in the mitochondrial outer membrane can result in oligomerization, which in turn releases the apoptosis-inducing cytochrome c. These effects simultaneously contributed to the abrogation of mitochondrial functions, thus inhibiting the energy generation of cells and promoting mitochondrial apoptosis. The viability of MDA-MB 231 cells was decreased to 30% with an increased concentration of CY-PSD nanoparticles, resulting in half-maximal inhibitory concentration (IC₅₀) of 12 × 10⁻⁶ M (Figure 2c). The final viability (87%) of thiol-inhibited cells after CY-PSD nanoparticles administration (0–100 × 10⁻⁶ M) was similar to that of the normal 3T3 cells (81%) under N-ethylmaleimide (NEM)-mediated cytotoxicity test, as the production of H₂S from thiol-polysulfide cleavage of PSD was depleted due to low thiol concentration, which further confirmed the proapoptotic effect of H₂S. Compared to the minimal volume changes from the control (PBS) group and CY group, the tumor volumes of PSD and CY-PSD groups after 21 days of treatment were decreased to 37 and 35.6%, respectively, conferring the antitumor effects of H₂S-producing treatments (Figure 2d and e). This was in line with the lowered tumor weights after the treatment of CY-PSD and PSD nanoparticles (Figure 2e). It was also suggested that CY-PSD nanoparticles alleviated the oxidative stress in TME, which eventually resulted in toxicity against tumor cells through sulfide-sulfite conversion. H&E staining of the extracted organs (heart, liver, spleen, lung, and kidneys) showed negligible damage, which suggested the biosafety of CY-PSD nanoparticles. Apart from the therapeutic effects, these nanoparticles also benefited the bioimaging with CY being an H₂S-sensitized probe and therefore can be used in cancer diagnosis.

Figure 2

(a) Schematic illustration of PA monitored thiol-induced H₂S therapy for triple-negative breast cancer. (b) Yielded H₂S concentration from the interaction between PSD (50 × 10⁻⁶ M) and GSH (0–11 × 10⁻³ M). (c) Concentration–response curves of cell viability of CY-PSD nanoparticles-incubated 3T3 (square), MDA-MB 231 (dot), and NEM-pretreated MDA-MB 231 (triangle) cells. (d) Tumor volume change curves of MDA-MB 231 tumor-bearing mice after injecting various materials (CY nanoparticles, CY-PSD nanoparticles, and PSD nanoparticles) and with phosphate-buffered saline (PBS) controlled. (e) Tumor sizes and weights from each treatment group. Reprinted with permission from ref. [32]. Copyright © 2020 Wiley-VCH.

Furthermore, the intrinsic pathway of mitochondria is also considered a drug target since the alteration of this pathway is closely related to cell apoptosis. Zhou et al. reported a methoxy poly(ethylene glycol)-poly (lactide-coglycolide) nanoparticles loaded with SH-aspirin (SH-ASA) and curcumin (Cur) (SH-ASA/Cur-coloaded mPEG-PLGA NPs) that could achieve co-distribution of both functional agents to obtain synergistic anticancer effects against ovarian cancer cells [33]. SH-ASA is a type of SH-NSAIDs (nonsteroidal anti-inflammatory drugs) that generated H₂S with antitumor activity through hydrolyzing ester bonds within its molecular structure so that the covalently bonded H₂S-containing moiety could be detached from aspirin. Different from the previous designs of H₂S-releasing drugs, polymeric nanoparticles were used in this study as the delivery vehicle that increased the dispersity, solubility, and circulation time of the encapsulated hydrophobic drugs. The H₂S might trigger a redox shift in TME by promoting oxidative stress, which inhibited the cancer cell proliferation and further sensitized the ovarian carcinoma to the combined treatment; hence, the overall therapeutic outcome was improved. The results illustrated that SH-ASA might further induce apoptosis via elevating the level of cytochrome c, therefore activating the downstream mitochondrial pathway and evolving an increase in caspase-3 and caspase-9 levels. In addition, a decreased level of B-cell lymphoma 2 (Bcl-2) and an increased level of Bcl-2 associated X protein (Bax) were observed upon the treatment. Bcl-2 protein exerted an anti-apoptotic effect while Bax protein tended to be pro-apoptotic and mediated the release of cytochrome c in the cytosol followed by the initiation of an apoptotic pathway [83]. In addition, Cur acted as another anticancer agent, synergistically attaining potent inhibitory effects against tumor growth. According to the results, the early and late apoptosis rates were 1.7 and 3.5% in the control group, and 21 and 64.7% in the SH-ASA/Cur NPs group, respectively, which made the effectiveness evident. Furthermore, the greatest synergistic effect (50% reduction in viability) was achieved with a combination ratio of 1:6 for SH-ASA and Cur. However, the action profile of the nanodrugs (e.g., release, uptake and cytotoxicity) was established on the basis of in vitro study only, which might require further investigation via in vivo analysis.

In another research, Ciocci et al. introduced novel H₂S-releasing nanoemulsions (BAD-NEs) consisting of diallyl disulfide (DADS) as a type of organo-sulfur compounds (OSCs) and α-linolenic acid (ALA) as a dietary polyunsaturated fatty acid [48]. BAD-NEs, as promising H₂S donors, demonstrated anti-proliferative and pro-apoptotic activities against breast cancer cells and T-cell lymphoma cells through caspase-3-involved apoptotic pathways in mitochondria together with the translocation of cytochrome c into the cytosol and an alteration of mitochondrial membrane potential. The releasing mechanism might involve a low-speed and low-productive reaction between DADS and GSH, which was achieved via α-carbon nucleophilic substitution [84]. In addition to the H₂S donated from DADS, the sulfane sulfur species on the amino acid outer layer of BAD-NEs could further induce the release of antioxidant H₂S. Therefore, BAD-NEs exerted an antitumor effect via both promoting apoptosis and inhibiting cancer cell proliferation. In this section, the anti-proliferative H₂S was released from organic polysulfides, mPEG-PLGA polymers and nature-derived organo-sulfur compounds via interacting with thiols, water molecules, or GSH in the TME. H₂S-mediated mitochondria damage was achieved by disrupting the mitochondrial pathway at the protein level. From the current research results, it is believed that the H₂S-mediated mitochondrial interference is a promising strategy for cancer therapy.

3.2 CAT inhibition

Cells living in oxygen-containing environments carry a risk of being damaged by oxidative stress. Catalase (CAT) is the enzyme that resists oxidative damage by catalyzing the metabolism of reactive hydrogen peroxide (H₂O₂) to produce more inert oxygen [85]. Since the protective role of CAT is indispensable for cell survival, elevated CAT is a distinct feature in a variety of cancer cells and has been widely acknowledged as a protein target [86]. Some previous proteomic studies on animal models showed that the cysteine residues of CAT can be covalently modified by H₂S via S-sulfhydration to produce persulfides, so that the ROS produced by concurrent treatments will remain stable and cytotoxic [87,88]. This post-translational modification alters the functionality of CAT and may well be the underlining mechanism of H₂S-triggered CAT inhibition.

Based on the findings above, Xie et al. synthesized ferrous sulfide (FeS)-embedded bovine serum albumin (BSA) (FeS@BSA) nanoclusters that demonstrated favorable anticancer effect in Huh7 cell lines by targeting the hyperreactive CAT activity [34]. In the weakly acidic TME, the nanoclusters were decomposed/ionized, releasing ferrous ions (Fe²⁺) and sulfide ions (S²⁻) (precursor of H₂S gas) for Fenton reaction catalysis and CAT inhibition, respectively (Figure 3a). As a result, the formation of toxic hydroxyl radicals (^·OH) was further increased due to H₂S-induced H₂O₂ accumulation as it synergistically enhanced the efficiency of the Fenton reaction. The pathological features of tumor cells result in a phenomenon of abnormal vasculature and poor fluid drainage, named the enhanced permeability and retention (EPR) effect. These nanoclusters, however, took advantage of the EPR effect to achieve tumor site accumulation, which enabled transverse relaxation time (T2)-weighted magnetic resonance imaging (MRI) to locate the tumor site. The release profile of H₂S was associated with the pH of the solution, in which the H₂S concentration increases with decreased pH. The highest releasing concentration (20 × 10⁻⁶ M) was achieved at pH 5.5, followed by a concentration of 10 × 10⁻⁶ M at pH 6.5, while no H₂S was released at pH 7.4 (Figure 3b). The cytotoxic effect of FeS@BSA nanoclusters was contributed by ^·OH induction. The control group showed an absence of fluorescence, while a green color was observed in the treated groups that were further enhanced at pH 6, suggesting a higher level of ^·OH (Figure 3c). The inhibition of CAT activity was also investigated by comparing the results of control and FeS@BSA-treated groups after a 30-min incubation (Figure 3d). After 14 days of treatment with a dosage frequency of once a week, moderate inhibition rates were found following the use of Na₂S (≈27 wt%) and Fe²⁺ @BSA (≈50 wt%), while the most significant inhibition rate (≈71 wt%) was shown in the FeS@BSA group (Figure 3e and f). This finding was also in concert with the size difference observed in the photographs of extracted tumors (Figure 3g). The biosafety of FeS@BSA nanoclusters was confirmed by the serum biochemical data and H&E staining analysis. Therefore, FeS@BSA nanoclusters showed an excellent therapeutic effect against cancer cells via a combination of H₂S-mediated Fenton reaction enhancement and CAT inhibition.

Figure 3

(a) Schematic illustration of the synthesis of FeS@BSA nanoparticles. (b) H₂S releasing profile at pH 7.4, 6.5, and 5.5. (c) Fluorescent images of DCFH-DA-stained Huh7 cells incubated with 20 µg mL⁻¹ of FeS@BSA in neutral and acidic (pH = 6) medium for 6 h, showing intracellular ROS. The untreated cells were used as the control group. (d) The effect of H₂S from FeS@BSA (0.1 mg mL⁻¹) on catalase activity of WRL-68 and Huh7 cells. (e and f) Relative tumor volumes and tumor weights after different treatments. (g) Photograph of the tumors from each treatment group. Reprinted with permission from ref. [34]; Copyright © 2020 Wiley-VCH.

Hypoxia is a hallmark of tumors that can weaken the efficacy of oxygen-required therapies. To bypass this restriction, several hypoxia-activated chemotherapeutic agents were designed and oxygen-consuming modalities like PDT can be synergistically used to potentiate the overall effect [89]. However, CAT-mediated conversion of H₂O₂ to oxygen can reversibly oxidize the free radicals produced by PDT, thereby masking the cytotoxicity. Taking that into consideration, Fang et al. prepared zinc sulfide (ZnS) nanoparticles with a shell of zeolitic imidazolate framework-8 (ZIF-8), which was then loaded with tirapazamine (TPZ) and indocyanine green (ICG) to form a co-delivery system (ZSZIT) that realized both PDT and chemotherapy [49]. In the weakly acidic TME, H₂S gas was released from ZnS followed by the dissociation of the ZIF-8 outer layer, which then played a dual role of exhibiting toxic effects and blocking the production of oxygen from H₂O₂ via inhibiting the activity of CAT. At the same time, ICG acted as a photosensitizer in PDT to assist the consumption of oxygen, facilitating the formation of singlet oxygen (¹O₂) under 808 nm NIR and exacerbating hypoxia. Therefore, the antitumor effect of hypoxia-stimulated TPZ was enhanced. The slices of ZSZIT + NIR-treated tumors demonstrated the most acute tissue damage under H&E stain, which was in line with the tumor weight when compared to other groups at day 14 after the treatment. Overall, the pH-triggered H₂S release from ZSZIT had an amplification effect that increased the efficiency of both PDT and gas therapy.

Wang et al. constructed H₂S-releasing ZnS nanoparticles with a silica nanofiber coating loaded with doxorubicin (DOX-ZnS@SiO₂), which enhanced the effect of chemotherapy in a similar way to the above study [35]. Nevertheless, in order to achieve a stable plasma concentration and avoid systemic clearance, the nanoparticles were implanted locally at the lesion site. After administration, the nanoplatforms were endocytosed into the hepatic cancer cells in which ZnS could react with hydrogen ions to synthesize H₂S gas under the acidic TME, which rose the ROS level by suppressing the CAT activity. In this way, DNA was damaged together with the pH-triggered DOX liberation. The effect of H₂S was shown by the DOX-ZnS@SiO₂ fiber-induced reduction in cell viability from ≈68.6% (at neutral pH) to ≈24.8% (at pH 6) after 48 h of incubation. The tumor sizes were also decreased dramatically after a 14-day treatment of DOX-ZnS@SiO₂ compared with the free DOX group. The effect of chemotherapy was successfully augmented by H₂S. To summarize, the exogenously delivered metal sulfide nanoparticles (e.g., ZnS and FeS) formed H₂S in the acidic TME to interfere with CAT activity and stimulate ROS formation, which potentiated the therapeutic effects of concurrent anticancer treatment.

3.3 Cytochrome c oxidase inhibition

The cytotoxic effect of H₂S gas may also be the consequence of cytochrome c oxidase (COX IV) inhibition. COX IV is a transmembrane enzyme in mitochondria that plays an intermediate role in the energy-generating process of electron transport chain (ETC) and is responsible for cellular respiration at the same time. Yang et al. designed polyvinyl pyrrolidone (PVP)-modified iron sulfide nanoparticles (Fe_1−x S-PVP NPs) that produced H₂S gas in the acidic pancreatic TME via a combination reaction of two protons and a sulfide ion [90]. The resultant H₂S gas was toxic against COX IV, thus causing respiratory chain dysfunction. It deprived the energy source of the cell, leading to decreased cell proliferation and ultimate tumor growth cessation. In addition to the aforementioned gas-formation ability, these nanoparticles also acted as photothermal agents (PAs) to generate heat energy under NIR irradiation (808 nm), which further accelerated toxic ROS (^·OH) production in the Fenton reaction (Figure 4a). The group treated with Fe_1−x S-PVP NPs under pH of 6.5 had a stronger Hsip-1DA-mediated fluorescence signal in contrast to the control group, indicating a higher H₂S concentration in TME (Figure 4b), where the mitochondrial COX IV was greatly inhibited. This was in line with the lowest intensity of COX IV found in Fe_1−x S-PVP + pH 6.5 group (Figure 4c). Fe_1−x S-PVP NPs + NIR group led to the largest tumor growth reduction among the four groups (Figure 4d). Moreover, the average weight of tumors removed from the mice was mostly reduced in Fe_1−x S-PVP NPs + NIR group, conclusively showing a greater tumor inhibitory effect compared to other groups (Figure 4e). Furthermore, the effect of H₂S was confirmed using photoacoustic (PA) imaging, which showed a decrease in mean tissue oxygenated hemoglobin at the tumor site from just before intravenous injection (12.5%) to 6 h post-treatment (5.76%). Also, the tumor volume was more significantly reduced in Fe_1−x S-PVP NPs + NIR group (78.3%) compared to that of the group using NIR alone (1.6%) after 14 days. In conclusion, the cytotoxic H₂S sourced from metal sulfide nanoparticles realized gas therapy by inhibiting COX IV and triggering ETC dysfunction under acidic TME.

Figure 4

(a) Schematic illustration of the Fe_1−x S-PVP-mediated synergistic gas therapy/PTT/CDT treatment. (b) CLSM images of intracellular H₂S. Scale bar = 50 µm. (c) Western blot analysis and intensity of COX IV in PAN-02 cells. (d) Tumor volume change of PAN-02 tumor-bearing nude mice after various treatments (PBS, NIR, Fe_1−x S-PVP, and Fe_1−x S-PVP + NIR). (e) Average tumor weight at the end of each treatment. Reprinted with permission from Ref. [90]; Copyright © 2020 Wiley-VCH.

3.4 Lactate metabolism disruption

With a greater understanding of the metabolic pathways specific to cancer cells, the potential to increase the target selectivity of nanotherapeutics was brought one step closer. Warburg effect is of particular interest in many studies as a unique metabolic strategy of cancer cells, which includes a fermentation process that forms lactic acid as a product with a risk of acidosis when accumulates at the tumor site. However, some types of cancer cells can degrade the lactic acid to form ATP via the Krebs cycle to prevent cell damage. Wan et al. designed PSBMA zwitterionic nanoparticles (P NPs) comprising zwitterionic sulfobetaine methacrylate (SBMA) as a polymerization monomer and N,N′-bis-(acryloyl) cystamine (BAC) as a crosslinking agent [91]. The synthesis was achieved by free radical polymerization that enabled the nanoparticles to be decomposed in TME with a high GSH level. On top of that, l-cysteine (l-Cys) and α-CHCA were co-loaded into the nanoparticles to assemble a H₂S-activated nanomotor PSBMA/l-Cys/α-CHCA (PCA) (Figure 5a). l-Cys is capable of producing H₂S with the catalysis of overexpressed cystathionine β synthase (CBS) in TME. The formed H₂S played a dual role. One was acting as the “fuel” of nanomotor after being released to improve the motility and the deep tissue penetration ability of the nanoparticles. The other was functioning as a therapeutic agent to enhance the cellular uptake of glucose to produce a large quantity of lactic acid together with blocking the export of protons out of the cell, both contributing to acidosis. On the other hand, α-CHCA performed as a monocarboxylic acid transporter inhibitor to further exacerbate the acidosis by disturbing the intratumoral lactic acid delivery and downregulating the lactate transporters (MCT-1/4) to synergistically accumulate intracellular lactic acid with H₂S. The lactic acid content ratio after treatment of PCA nanomotor was approximately 5.7, which was greater compared to the PA (P NPs loaded with α-CHCA) group (≈2.9) due to an additional H₂S-releasing capacity (Figure 5b). In order to evaluate the target specificity, HUVECs (vascular endothelial cells) were used to represent non-cancerous tissue, which did not show significant responses in terms of intracellular pH reduction. While the PCA group resulted in the lowest intracellular pH due to its strong effect of triggering acidosis (Figure 5c). A decrease in ATP concentration in the PA group was observed in 24 h, possibly attributed to the transporter inhibition of α-CHCA. The inability of excreting lactic acid generated negative feedback to impede the glycolysis process. Meanwhile, the oxidative tumor cells were not able to adsorb lactic acid into the Krebs cycle, so that the ATP production was depleted. PCA group demonstrated the greatest reduction of ATP production after 24 h due to the synergism of α-CHCA and H₂S (≈2.5 μM) in contrast to that of the control group (>7 μM) (Figure 5d). Accordingly, the lowest cell viability (≈34.3%) was obtained in the PCA group in 24 h. The powering ability of H₂S was assessed by setting H₂S-involved groups (PC and PCA) and H₂S-absent groups (P + C and PA + C). The latter groups did not exert sufficient tumor inhibitory effects after 14 days by showing relative tumor volumes of 3.51 and 2.27, respectively. In comparison, the deep tissue penetration of PCA led to large tumor growth reduction and a relative tumor volume of 1.42 (Figure 5e). Owing to the high biocompatibility and antitumor effects, all subjects receiving the PCA nanomotors survived 60 days (Figure 5f). Overall, the exogenously donated H₂S was used as the “motive power” of PCA nanomotors to realize deep tissue penetration and also as a therapeutic agent in conjunction with α-CHCA to trigger acidosis and ultimately apoptosis of tumor cells.

Figure 5

(a) Schematic illustration of the synthetic process and action of PCA nanomotor. (b) Increased intracellular and extracellular lactate content ratio in different treatment groups. (c) The pH of MCF-7 and HUVECs cells after a 24 h treatment with different groups. (d) The ATP level in MCF-7 cells after a 24 h incubation with different treatment groups. (e) The change in tumor volumes of mice during different treatments. (f) The survival rate of mice in 62 days. Reprinted with permission from Ref. [91]; Copyright © 2021 Wiley-VCH.

4 H₂S-depleted therapy

The overproduced endogenous H₂S itself is involved in a diverse set of biological activities to promote cancer invasion and progression, for example, DNA repairment, tumor cell metabolism, and angiogenesis [92, 93, 94]. Therefore, aside from using H₂S as a “switch” for drug release, disturbing the balance of H₂S levels can have an inhibitory effect on tumor growth. Regulating the endogenous production of H₂S has become a strategy of choice, but fewer strategies are available to deplete endogenous H₂S production in contrast to donating exogenous H₂S, possibly due to a research gap in the role of H₂S in tumor pathogenesis. CBS, cystathionine-γ-lyase (CSE), and 3-mercaptopyruvate sulfurtransferase (3-MST) are the three enzymes involved in the synthesis of endogenous H₂S. CBS is usually the most targeted enzyme since the levels of CSE and 3-MST in cancer cells are comparable to that of the non-cancerous cells [95]. The augmentation of intratumoral H₂S can be a result of CBS overproduction in certain types of cancers such as colon cancer, which specifically converts l-Cys to H₂S and leads to an induction of angiogenesis and tumor cell proliferation [27,96]. This finding was further proved by the successful application of CBS inhibitors like aminooxyacetic acid against overproduced H₂S [97]. Moreover, the distinct EPR effect results from the pathological features of tumor cells can accumulate nanodrugs at the lesion site and enhance the target specificity, so that the treatment efficiency and safety are improved. Based on these features, a series of endogenous H₂S-depleted and stimulated drug delivery systems were constructed for cancer therapy.

4.2 Glutathione peroxidase 4 inhibition

Given the importance of GSH as a reducing agent against oxidative damage, its depletion results in GSH-dependent enzyme inhibition and ROS accumulation, which eventually leads to cellular damage [99,100]. Glutathione peroxidase 4 (GPX4), an intracellular GSH-dependent lipid repairing enzyme, catalyzes the reduction of membrane-bound phospholipid hydroperoxides to prevent associated cell damage and ferroptotic death [101,102,103,104]. GPX4 can also mediate the degradation of H₂O₂, lipid peroxides, and other peroxides to yield water and alcohol molecules [105]. GSH-GPX4 redox axis plays an important role in maintaining tumor cell redox balance and normal life activities. H₂S could increase the production of GSH by enhancing cystine/cysteine transporters and redistributing GSH to mitochondria, and thus promote the effect of GSH-GPX4 redox axis [24].

Based on the above mechanisms, Pan et al. constructed zinc oxide-coated virus-like silica nanoparticles (VZnO) against colorectal cancer (CRC) via a reaction between H₂S and ZnO, which produced ZnS and water (Figure 6a) [41]. This desulfurization process removed the S²⁻ ion from H₂S to reduce the level of endogenous H₂S, resulting in a lower concentration of GSH at the tumor site. As a co-substrate of GPX4, a decreased level of GSH could inactivate GPX4 and thus deprive its role as an antioxidant against lipid peroxidation. Subsequently, Fe²⁺ ions oxidized the lipids via a Fenton-like process that induced the cancer cells to undergo ferroptosis as a result of ROS overexpression [106]. The H₂S concentration was measured to evaluate the effect of VZnO and iron-dextrin (ID) group (50 mg kg⁻¹), which revealed a decreasing trend with a maximum reduction being observed after the fourth treatment (Figure 6b and c). As shown by the photograph and measured tumor volumes, the combined therapy of VZnO and ID caused the most tumor growth inhibition among all the groups on day 14 (Figure 6d and e). Moreover, the biosafety was ensured by the blood tests and body weight measurements of mice treated with VZnO. Overall, VZnO was considered a potent treatment for CRC that displayed a relatively high selectivity toward H₂S-enriched TME. To sum up, the antitumor effect can be achieved via depleting H₂S molecules that have already been evolved at the tumor site as well as inhibiting its formation process. The redox ability of H₂S enables it to participate in multiple facets of cellular metabolism, for instance, the aforementioned ability to disrupt the normal function of the GSH-GPX4 redox axis. The imbalance of redox reactions can lead to ferroptosis of tumor cells, which provides a breakthrough point for us to further investigate the treatment mechanisms of H₂S from the aspect of ferroptosis.

Figure 6

(a) Schematic illustration of VZnO synthesis and H₂S depletion for CRC treatment. (b) Measurement of H₂S content in the mouse model. (c) H₂S content measured on day 14. (d–e) Photograph and volume of extracted tumors on day 14 in different treatment groups. Reprinted with permission from Ref. [41]; Copyright © 2021 Springer Nature.

4.3 H₂S-triggered drug release

It was reported by Thirumalaivasan et al. that folic acid-anchored azide functionalized mesoporous silica nanoparticles (MSNPs-N₃-FA) loaded with DOX were effective against ovarian and colon cancer cells [107]. MSNPs were applied as carriers of DOX that kept it stored in the mesopores for subsequent H₂S-triggered release (Figure 7a). A high concentration of H₂S in TME stimulated the reduction of azide (N₃) into amine, breaking the ester bond of carbamate that is used to link folic acid and the drug-loading mesopores covered by an α-cyclodextrin valve. The opening of α-cyclodextrin valves then liberated the DOX from mesopores to exert its cytotoxic effect (Figure 7b). The fluorescein stored in the mesopores of nanovalves was also restricted by the α-cyclodextrin cap and was released later in H₂S-riched TME, which demonstrated a positive correlation between H₂S level and the amount of released fluorescein (Figure 7c). H₂S-triggered drug release feature of MSNPs-N₃-FA will improve the biosafety of therapeutic molecules. In the in vivo study, the tumor volumes were 47.56 and 34.12% decreased in the free DOX group and MSNP-N₃-FA group, respectively, while DOX-loaded MSNP-N₃-FA exerted the most remarkable inhibition rate of 68.77% with the smallest tumor weight after 30 days (Figure 7d and e). To conclude, MSNPs-N₃-FA acted as a nanocarrier in the treatment of colon and ovarian cancers by using the tumor-site H₂S as a triggering molecule to assist chemotherapeutic drug release.

Figure 7

(a) Schematic illustration of the integration and release of DOX-loaded MSNP-N₃-FA into HT-29 cells. (b) H₂S-triggered drug release. (c) Fluorescein release from MSNP-N3-FA in different H₂S concentrations. (d) Tumor growth curves after intra-arterial injection of PBS, free DOX, MSNP-N₃-FA, and DOX-loaded MANP-N₃-FA. (e) Changes in tumor weight in different treatment groups. Reprinted with permission from Ref. [107]; Copyright © 2019 American Chemical Society.

After gaining more knowledge on H₂S-triggered drug delivery and release, the target specificity is largely improved. The new frontier is now established in the exploration of potential H₂S applications in combination with other therapeutic modalities. Shi et al. designed nanoparticles (NPs@BOD/CPT) with co-loaded boron-dipyrromethene dye (InTBOD-Cl) and Camptothecin-11 (CPT-11) inside an illumination sensitive phase change material (PCM). In the H₂S-rich environment of CRC, the chlorine on the aromatic ring of InTBOD-Cl molecule was substituted with nucleophilic HS⁻, achieving in situ formation of NIR-activable InTBOD-SH [108]. The photothermal converting capacity of InTBOD-SH enabled a temperature elevation (to 44.9°C) at the tumor site that exceeded the eutectic melting point (39°C) after a 10 min NIR irradiation (785 nm, 1.00 W cm⁻²). This resulted in a state change of PCM matrix from solid to liquid and therefore triggered the release of encapsulated cytotoxic CPT-11. Complete tumor eradication was then observed after 14 days of treatment. Hence, in this study, the overexpressed H₂S was used as a formation moiety of the photothermal mediator to be used for consequent light-controlled CPT-11 release. In brief, the H₂S-triggered drug release is often realized by inducing chemical reactions between the drug delivery vehicles and endogenous H₂S. For example, a bond cleavage or functional group substitution ultimately leads to the liberation of cytotoxic drugs for tumor cell elimination. This conditional release of drugs from the nanoplatforms possesses the advantages of less normal tissue damage and high biocompatibility.

5 H₂S-involved phototherapy

Phototherapy refers to the therapeutic strategy based on the conversion of light energy to heat or ROS, which can be categorized into photothermal therapy (PTT) or PDT according to the substrate and end product. In PDT, the oxygen in TME was utilized by photosensitizers to be transformed into highly toxic ¹O₂. In PTT, however, the heat is generated from oxygen by NIR-activable PAs that raises the temperature of TME and thus impedes cancer cell survival. Notably, nanoscale biomaterials are gaining popularity as a new generation of photoactive drugs with remarkable improvements like modified biodistribution and stimuli-controlled release. In our previous review, lots of metal sulfides nanomaterials were shown to have excellent photosensitivity, absorbability, and transduction capacity, and therefore can be applied as novel phototherapy agents [77]. For instance, silver sulfide nanoparticles (Ag₂S NPs) can be used for PTT, and cobalt sulfide nanodots (Co₉S₈ NDs) can be used for PDT. In other cases, some metal oxides are able to convert to metal sulfides in H₂S-enriched environment, where the products can participate in PTT or PDT. Based on these features, researchers have developed various H₂S-stimulated nano delivery systems for the phototherapy of cancers.

5.1 Endogenous H₂S-involved PDT

The indirect H₂S-responsive release of ¹O₂ was first achieved by Ma et al. using cupric (Cu²⁺⁾ ion and zinc organic ligands-containing metal-framework nanoparticles (Cu₂(ZnTcpp)·H₂O _n )/(NP-1) as photosensitizers for PDT [50]. In the colon adenocarcinoma cells, H₂S reacted with the active site of the framework containing Cu²⁺ ions that were appended as a “stopper” to allow the generation of photosensitive ligand after its dissociation. After the attack of H₂S, the bond between Cu²⁺ and carboxylate groups was cleaved, empowering the sequential cytotoxic ¹O₂ production under NIR exposure (Figure 8a). The inhibition of tumor growth was successfully detected within 2 days in Cu₂ (ZnTcpp) + NIR group (Figure 8b). The perfect antitumor effect of Cu₂ (ZnTcpp) + NIR was also in good agreement with the MTT assay on HTC116 cells activated by the intrinsic H₂S (Figure 8c). To classify the impact of H₂S concentration in the activation of NP-1 under irradiation, the tumor inhibition curves were depicted using three model cell lines. The results from the HepG2 (human hepatocellular liver carcinoma cells) group demonstrated a minimal tumor inhibitory effect in low H₂S TME, which was similar to that of the LoVo (human colorectal adenocarcinoma cell) control group. On the contrary, the tumor growth in the HCT116 group (high H₂S level) was effectively inhibited. Thus, the effect of PDT treatment was enhanced with increased H₂S concentration in the TME (Figure 8d–f).

Figure 8

(a) Schematic illustration of MOF NP-1 structure and mechanism of action. (b) Tumor growth inhibition in different treatment groups. (c) MTT assay of the HCT116 cells in the presence of different concentrations of NP-1 activated by intrinsic H₂S. (d) Tumor growth inhibition curves for three cell models after the treatment of 1.0 mg mL⁻¹ NP-1 dispersed in PBS buffer. (e–f) Photographs of HepG2 and LoVo tumor-bearing mice in PBS group and NP-1 group, respectively. Reprinted with permission from Ref. [50]; Copyright © 2017 Wiley-VCH.

5.2 Endogenous H₂S-involved PTT

Photothermal therapy, an emerging anticancer therapy that consists of non-invasive light absorption and heat production, is characterized by deep tissue penetration capacity and good biocompatibility. The NIR irradiation (700–2,500 nm) and a photosensitizer are required to convert light energy into heat, which is manifested as a temperature rise in local tissue. When the temperature reaches above 41°C, the damage to tumor tissues will then be present [109]. By the precise irradiation of NIR and efficient photothermal conversion of the nanomaterials (including but not limit to nano-sized metal sulfide like copper(ii) sulfide (CuS), FeS and molybdenum disulfide [77]), PTT is a relatively safe and non-invasive treatment. In order to prevent heat damage in non-tumor tissues, metal oxides or metal hydroxides were designed in previous studies to be converted into metal sulfides with photothermal activity in TME enriched with H₂S.

An et al. proposed a theranostic approach using cuprous oxide (Cu₂O) nanoparticles as substrates, which underwent sulfidation in the H₂S-rich TME and were converted into copper sulfide (Cu₉S₈) [110]. The thermal imaging demonstrated a change in wavelength in the NIR region, which recognized Cu₂O as a potent antitumor precursor of the photosensitive Cu₉S₈. As expected, the relative tumor volume nearly approached zero after a 16-day treatment. In another research, Li et al. proposed a possible defect of the above study, which was the toxicity of the Cu element. Therefore, another H₂S-sensitive iron oxide-hydroxide nanospindles (FeOOH NSs) were constructed for the treatment of colon cancer [51]. The nanospindles exerted a dual effect in the presence of H₂S by effectively decreasing the intracellular H₂S concentration by 85% to cease the tumor progression and providing a target-specific PS (FeS) via interacting with the endogenous H₂S. The tumor size and weight after the administration of FeOOH NSs and NIR (≈408 mm³ and 0.25 ± 0.15 g) on day 15 post-treatment were around six times smaller than that of the control group (≈1,940 mm³ and 1.5 ± 0.51 g). Moreover, the ferric ions in the nanostructure also contributed to MRI imaging and ferroptosis, which respectively assisted in cancer diagnosis and overall inhibitory effect on carcinoma.

In addition to forming metal sulfides, polymer micelles could also exhibit a photothermal effect after H₂S stimulation at the tumor site. For instance, Shi et al. developed nanocomplexes made up of self-assembled phenyl tail small molecules (nano-PT) that used NIR as a trigger to elicit photothermal effect in colorectal cancer cells after H₂S-mediated aromatic substitution of the chloride group [111]. Thus, this substitution reaction was key to the action of nano-PT. The nanoparticles, on the other hand, were also named H₂S-responsive small molecules capable of self-assembly (SSS). SSS were synthesized via alkylating 2,3,3-trimethylindolenine with compound 1 and then condensed with aldehyde boron dipyrromethene (BODIPY) (Figure 9a). The structure therefore included a three triethylene glycol monomethyl ether chain-functionalized phenyl ring tail and a BODIPY core. On top of the light absorption, nano-PT also ejected NIR-II signals deeply into cancer regions to trail the intracellular activation of the nanoparticles (Figure 9b). The obtained high-quality NIR-II imaging with an infinitesimal amount of signals in normal tissues made nano-PT a potential nanoprobe for CRC detection and PTT guidance. In order to evaluate the photothermal activity of nano-PT, the temperature change was measured along with time. NaHS was introduced to reveal that both H₂S and SSS concentrations contributed to the temperature changes, and the photothermal effect would only exist with the presence of H₂S (Figure 9c). The specificity test was then performed to assess the targeting ability of nano-PT in tumor tissues by irradiating with a 785 nm laser (1.66 W cm⁻²) at 2 h after injection. The injected Nano-PT was switched off in healthy tissues, while the nano-PT injection into tumoral tissues resulted in a rapid rise in temperature to around 60.9°C at 10 min post-injection. In contrast, there were only minimal changes in the temperature in mice treated with irradiation alone (Figure 9d and e). As a result, the in vivo effect of nano-PT as an H₂S-activable thermal agent was confirmed. The treatment course was continuously monitored for a total of 15 days. The mice exposed to both nano-PT and laser irradiation had complete tumor weight and volume reductions. This finding was different from the other groups, in which neither nano-PT nor NIR irradiation was shown to be effective as standalone therapy (Figure 9f–h). The blood biochemistry analysis showed great biosafety of Nano-PT, in which no abnormal results were detected. Overall, nano-PT was considered the first H₂S-stimulative photothermal agent capable of self-assembling at the CRC site.

Figure 9

(a) Synthesis of SSS and its conversion into a NIR-triggered photothermal agent. (b) An illustration of the action of photothermally active Nano-PT for NIR-II fluorescence-guided colorectal cancer therapy. (c) Temperature changes of PBS and Nano-PT (20 μM) with or without NaHS (100 μM) under 785 nm laser irradiation (5.37 W cm⁻²). (d and e) Infrared thermal images of HCT116 tumor-bearing mice under NIR laser irradiation: (i) no probe administered; (ii) probe-administered in normal tissue; (iii) probe-administered in tumor tissue. (f) Photographs of grouped mice during PTT. The tumor location was represented by red circles. (g) The curves of tumor growth in different treatment groups. (h) Ratio between tumor weight and photographs of extracted tumors on day 15. Reprinted with permission from Ref. [111]; Copyright © 2018 American Chemical Society.

6 Endogenous H₂S-involved chemodynamic therapy

The application of anticancer nanomaterials in CDT is arising, which uses Fenton or Fenton-like reaction to produce highly toxic ^·OH. Metal ions such as Fe(iii) and Fe(ii) are often considered as Fenton catalysts, whereas the catalytic ability of Fe(iii) is weaker than that of Fe(ii) with the conversion between Fe(iii)/Fe(ii) being a rate-limiting step. Therefore, a H₂S-mediated reduction reaction was embedded in the study of Tian et al. by fabricating ultra-small ellagic acid-Fe-bovine serum albumin (EA-Fe@BSA) NPs as a CDT agent through compounding FeCl₃, BSA, and ellagic acid (polyphenol) [113]. The Fe(ii) supplied from EA-Fe@BSA NPs successfully met the requirement for highly efficient catalysis and thus enhanced the efficiency of CDT against colon cancer cells (Figure 10a). The reducing activity of endogenous H₂S boosted the conversion of Fe(iii) into Fe(ii) and therefore offered a greater therapeutic advantage by producing more cytotoxic ^·OH. Apart from the oxidative stress exerted by the end product of Fenton reaction in CDT, the strength of EA-Fe@BSA NPs was further amplified by having photothermal effect. It is worth noting that the report of Cu₂O nanoparticles discussed in the PTT section was from previous research of the same group, in which the effective concentrations of endogenous H₂S and Cu₂O were exceedingly high, thus making that approach less practical in the clinical settings. Hence, CDT was then proposed as an advanced method with lowered demand for therapeutic reagents and H₂S levels. The electron spin resonance (ESR) was used to evaluate the ^·OH-generating ability of the nanotherapeutics. As shown by the results, the classic four-line pattern disappeared in the absence of H₂O₂ but was reinforced in the presence of NaHS and an elevated temperature, indicating an increased production of ^·OH due to both H₂S- and heat-accelerated Fenton reactions (Figure 10b). Besides, the relative tumor volume was dramatically decreased at 16 days post-treatment in the EA-Fe@BSA NPs + laser group. However, recurrence was observed around the eradicated tumor tissue in the same group at day 10, which diminished its long-term efficacy. In comparison, complete tumor elimination was achieved in mice given EA-Fe@BSA NPs, S-adenosyl-l-methionine (SAM), and laser irradiation. No tumor suppression was reported in other treatment groups (Figure 10c and d). Generally, the potential role of endogenous H₂S as a reducing agent in augmenting CDT efficiency was corroborated by EA-Fe@BSA NPs.

Figure 10

(a) An illustration of the synthesis of the EA-Fe@BSA NPs and H₂S-triggered Fe(iii)/Fe(ii) conversion and PTT synergistically enhanced CDT. (b) ESR spectra for ^·OH detection. (c) Relative tumor volume. (d) Photographs of mice on days 0, 8, and 16 of treatments. Reprinted with permission from Ref. [113]; Copyright © 2020 IVY Publisher.

7 H₂S-involved immunotherapy

Although PTT is a promising anticancer therapy with unique advantages, there remain issues in preventing the recurrence and metastasis of tumors. To this end, a combination use of PTT and immunotherapy has been attempted by researchers to improve the tumor eradication efficacy and long-term prognosis [114,115]. The hyperthermia resulting from PTT can activate immune cells and enhance the antigen presentation, collaboratively achieving immunogenic cell death to further potentiate the effect of immunotherapy while ablating the tumor physically.

Chang et al. developed calcium carbonate (CaCO₃)-coated and hyaluronic acid (HA)-decorated Cu₂O nanoparticles (CCH) on the basis of Cu_2−x S via in situ mineralization and HA coating. CCH could accumulate at the colorectal cancer site and bind firmly to the highly expressed CD44 receptors [116]. The nanoplatforms successfully increased the target specificity through acid-triggered CaCO₃ decomposition in the TME. Followed by a reaction between tumor-site H₂S and CCH, smaller-sized Cu₃₁S₁₆ NPs with improved NIR absorbance were formed from the Cu₂O core to realize PTT and PDT. The breakdown of CaCO₃ layer not only liberated inner particles but also accumulated Ca²⁺ ions, which led to calcified tumors with inhibited growing activity. It was therefore named ion-interference therapy (CIT). Meanwhile, the Cu²⁺ ions could also display Fenton catalytic property to achieve CDT. As a result, CCH achieved a range of different anticancer therapies including CDT, PTT, PDT, and CIT. On top of that, the ROS and thermal energy produced from PTT facilitated the immune cells to bypass the immunosuppression by transforming the pro-tumoral M2 phenotype tumor-associated macrophages (TAMs) into antitumoral M1 phenotype TAMs. This conversion could stimulate the Fenton reaction, which led to the programmed cell death. The resultant Cu₃₁S₁₆ NPs were renally excreted from the body after fully exerting the therapeutic effect (Figure 11a). The photothermal conversion efficiency of Cu₂O was increased from 14.48 to 65.89% in the presence of 200 µg mL⁻¹ NaHS. In the CCH group exposed to 1,064 nm laser irradiation, the ratio of M1 macrophages increased from 34.2 to 64.8%, while the ratio of M2 macrophages decreased from 68.8 to 41.9% (Figure 11b). This result demonstrated that the CCH could reverse the immunosuppressive TME under 1,064 nm laser irradiation by re-converting M2 macrophages to M1 macrophages. This conclusion was further confirmed by the upregulation of IL-12 (secreted by M1 macrophages) and the downregulation of IL-10 (secreted by M2 macrophages) (Figure 11c and d). Under the same irradiation condition, the tumoral temperature after CCH treatment was increased by 14.5°C compared to only a 3.0°C increase in the PBS-treated group. Altogether, the CCH + 1,064 nm + anti-CD47 group had the strongest tumor inhibitory effect shown by the smallest tumor size, highest survival rate, and the greatest reversed M1 TAMs (42.2%) and M2 TAMs (8.7%) percentages after the treatment (Figure 11e–h).

Figure 11

(a) Systemic illustration of the synthesis, bio-decomposition, antitumor immune responses, and renal clearance of CCH. (b) The flow cytometry of CD206 (M2 macrophage marker) and CD86 (M1 macrophage marker) expression in different treatments. (c and d) Secretion levels of IL-10 and IL-12 in the supernatant after different treatments. (e and f) The tumor sizes and survival rate after different treatments. (g and h) M1 and M2 macrophage percentages in the tumors. Groups: (1) control, (2) 1,064 nm, (3) anti-CD47, (4) CH, (5) CCH, (6) CCH + 1,064 nm, and (7) CCH + 1,064 nm + anti-CD47. Reprinted with permission from Ref. [116]; Copyright © 2020 Wiley-VCH.

8 H₂S-involved ultrasonic therapy

Ultrasound (US) is commonly embedded in multiple therapeutic modalities such as kidney stone elimination and fracture recovery acceleration [117,118]. Over the past few decades, the US was also found to play roles in different facets of cancer treatment depending on the cancer type and concurrently used therapies, for instance, sonodynamic therapy, which can be realized by using sonosensitizers to form highly toxic ROS, and US-mediated chemotherapy, in which the drug uptake can be enhanced by reversibly increasing the cell membrane permeability [119,120]. Although the specific target can differ, the main action of the US is to induce the size change and burst of microbubbles within TME, which also concluded as the cavitation effect [121]. Under the US, the microbubbles oscillate and collapse, resulting in jet formation on the cell membrane and mechanical stress toward cellular structures. The resulted increase in drug internalization and cell damage then triggers apoptosis. While some of the studies utilized the co-delivery technology to transfer microbubbles together with nanoparticles via the bloodstream, a new concept of H₂S-mediated microbubble generation was envisaged and achieved via enzymatic catalysis of nanoscale H₂S-donors (e.g., anethole trithione).

Cytoskeleton, the scaffold of a cell, is responsible for structural support and is included in activities such as cell movement and transportation of intracellular components. Based on its significant role, Liu et al. constructed magneto-acoustic nanoliposomes loaded with superparamagnetic iron oxide nanoparticles (SPIOs) and anethole trithione (ADT) (SPIOs-ADT-LPs) to disrupt the cytoskeletal structure in HepG2 cells [122]. ADT is a kind of H₂S-releasing drug that caused imbalanced redox homeostasis in TME via magnetic field-aided membrane adherence and endocytosis. The donation of H₂S bubbles activated the membrane calcium channels to allow a higher entrance rate of Ca²⁺ ions that were capable of aggravating cytoskeleton damage via disrupting the actin structure. The raised level of intracellular calcium also enhanced the activity of cell myosin V ATPase, thus accelerating the dispersion of SPIOs-ADT-LPs-containing phagosomes inside the cell. After the gradual expansion of H₂S gas, US acted as an acoustic stimulus to foster the eruption of sensitive H₂S bubbles through increasing hydrostatic pressure. The intracellular “explosive” event ultimately gave rise to cell apoptosis and necrosis due to a distorted cytoskeletal structure. This was verified by fluorescence images that manifested F-actin cytoskeleton network destruction after 12 h of incubation with SPIOs-ADT-LPs, in which H₂S bubbles were gradually generated. The cell structure was creased and an unclear membrane edge appeared in the transmission electron microscopy images followed by a complete distortion after the integration of acoustic waves. In summary, the extracellular H₂S donor (ADT) was delivered to form US-triggerable H₂S bubbles, which destructed the cytoskeleton structure of tumor cells by applying mechanical stresses.

More recently, another magnetic nanoliposome bomber (AML) containing ADT was synthesized in a separate study by Liu et al. [79]. Likewise, the antitumor effect implicated both magnetism-guided nanoparticle distribution and H₂S-induced cell explosion, but a new concept of intratumoral particle size conversion from the nanoscale (∼200 nm diameter) to microscale was introduced to conquer the limitation of nanomaterials (Figure 12a). For example, the small particle-sized nanoparticles were likely to be cleared out of cells. Owing to the ability of AML to act as a contrast media, H₂S production was achieved and monitored by US. The T2 MR images did not show a decrease in the T2 signal of blank liposomes and ADT-encapsulated liposomes (ALs) groups after injection. While a 22% of reduction in the mean gray value of the tumor area was observed in the AML group 6 h post-injection, illustrating the effect of AML in decreasing the tumor size. The H₂S concentration was similar in both AMLs (6.56 × 10⁻³ μmol) and ALs (9.29 × 10⁻³ μmol), while no H₂S was detected in liposomes, which demonstrated the H₂S-generating behavior of ADT (Figure 12b). The tumor cell viability in both ALs and AMLs groups was decreased after 4 h post-treatment, and approximately 50% of cell deaths were observed at 12 h post-treatment in both groups, plausibly attributed to the cytotoxic effect of destructive H₂S bombers (Figure 12c). In the AML-incubated HepG2 cell culture, the cell morphology started to change at 2.5 h, and the membrane rupture was observed after 6 h in conjunction with the detachment of cells from the culture dish, indicating cell damage and death. However, no significant changes were found in liposome-treated cells (Figure 12d). The intensity of US in the AMLs group was maximized at 4 h following the injection, which was about 3.4 times greater than that of the ALs group. The differences in results between the two treatment groups were contributed by the additional superparamagnetic nanoparticles in AMLs, which generated a greater response to the magnetic field and therefore enhanced the accumulation of AMLs at the tumor site for later H₂S production. Among all treatment groups, the tumor-bearing mice given AMLs under a magnetic field achieved the greatest tumor growth inhibition, showing that the magnetically stimulated AML accumulation could promote antitumor effects by generating a high intracellular H₂S concentration (Figure 12e). Accordingly, the cavitation areas and extent of cell apoptosis in AMLs group and magnetic field group were clearly demonstrated by the TUNEL assay (Figure 12f). Generally, AMLs displayed a prominent antitumor effect. There was negligible mortality and weight reduction in the AMLs-treated group, confirming the biosafety of these nanodrugs. To conclude, ADT acted as an H₂S donor in both studies of Liu et al. by incorporating it into the lipophilic membrane of magnetic liposomes. Following the spatiotemporally controlled drug delivery under external magnetic fields, H₂S bubbles were progressively produced and accumulated to an extent in which addition of acoustic energy could trigger the explosion and cell rupture.

Figure 12

(a) Schematic illustration of the composition and mechanism of actions of AMLs. (b) Detection of H₂S production by different treatment groups. (c) Cell viability in different treatment groups at different time points. (d) Changes in cell morphology and bubble generation at 0, 2, 4, 6, 8, 10, 12, and 24 h in AMLs, ALs, and liposome groups. The presence of bubble and cell membrane damage was indicated by red arrows. (e) The changes of relative tumor volume throughout the 7 day follow-up treatment of different groups. (f) TUNEL assay for tumors. Reprinted with permission from Ref. [79]; Copyright © 2017 American Chemical Society.

9 Summary and prospects

The use of H₂S in cancer treatment was not favored at first due to its toxic nature. Therefore, studies on H₂S in the field of life science are delayed compared to NO and CO. Similar to other gas molecules, increased cellular H₂S level via gas therapy is one of the most prevalent topics as the overexpression of H₂S can inhibit the cancer cell proliferation through damaging mitochondria and other key enzymes like CAT and COX IV. In addition, the integration of nanotechnology allows tumor-site specific release of H₂S gas to optimize the safety of gas therapy. H₂S demonstrated different characteristics compared to other gas molecules, such as the powerful reducing ability. As a result, lowering the H₂S concentration at the tumor site can alter the redox balance to prevent the proliferation and metastasis of tumor cells. It should be emphasized that H₂S depletion has been less studied as a treatment option in comparison with H₂S donation. Thus, more attention should be drawn to further explore the role of H₂S depletion in the occurrence and development of tumors. Nanotechnology enhances the diversity of H₂S-involved cancer treatments. As a result, phototherapy, CTD, immunotherapy, and ultrasonic therapy have all been realized by designing nanotherapeutics based on the physicochemical properties of H₂S. In addition, the active targeting, passive targeting, and TME-specific H₂S-releasing features of nanotherapeutics can improve the biosafety of H₂S-involved nanodrugs. In view of the versatility of metal sulfide nanomaterials, more antitumor therapeutic strategies are expected to be developed hereafter. For instance, the synergistic combinations of H₂S-involved nanotherapeutics and other treatment strategies may be the focus of future research.

Despite the significant progress on H₂S-involved nanotherapeutics in the past decade, several challenges are to be resolved before clinical translation: (1) Since the effects of H₂S are dose-dependent, its mode of action at different concentrations needs to be explicitly explored in order to optimize the therapeutic efficacy. At this stage, however, the therapeutic window of H₂S-mediated nanodrugs cannot be precisely defined, in which the adverse effects and therapeutic effects are balanced to obtain the greatest outcome without damaging healthy tissues. (2) It is also unclear whether a certain dose of the drug has similar efficacy in all kinds of TME considering the large variability between cancer types and/or stages. Of all the studies reviewed in this paper, colon cancer and ovarian cancer are the most reported cancer types due to their typical features of high H₂S levels. Hence, the treatments can have greater effects on certain cancers than the others. For this reason, it is worthy to investigate different types of malignancies to acquire a more holistic view of the efficacy spectrum of the H₂S-mediated nanodrugs. (3) To further guide and track the release of H₂S, the continuous development of real-time gas monitoring techniques should be encouraged considering the fact that a large portion of the aforementioned nanoplatforms lacks effective monitoring of H₂S activity, which increases the risk of gas poisoning. (4) Most of the nanovehicles have demonstrated good biocompatibility but tend to be unstable in harsh microenvironments, leading to uncontrolled gas release toward normal tissues. Thus, the spatiotemporal control over H₂S-releasing nanoplatforms is continuously desired, although it has already been achieved by a few studies. The inorganic metal sulfide nanomaterials, on the other hand, are highly stable but have a problem of potential endogenous toxicity. (5) On top of that, some of the synthesis procedures are complicated with low reproducibility and therefore dwindle the opportunity for future clinical application. (6) Although the discovery of H₂S-regulated nanoplatforms has been advanced, only a small number of studies were mature enough to be translated into clinical trials. The data presented in this review mainly came from in vitro studies and animal models, hence were only discussed in the context of pharmacokinetics and pharmacogenetics. Therefore, the biosafety evaluations such as biodistribution, metabolism, excretion, biocompatibility and degradation are all prerequisites for future clinical translation. To summarize, developing translational anticancer treatments is a prolonged and complicated process that requires long-term multidisciplinary research to gain a deeper understanding of the underlining mechanisms of H₂S in cancers. It is believed that H₂S-involved nanotherapeutics can be good candidates for future cancer treatments and more significant roles of H₂S will be revealed as this research area matures.