A green, efficient and precise hydrogen therapy of cancer based on in vivo electrochemistry

Abstract By combined use of traditional Chinese acupuncture Fe needle electrode and in vivo electrochemistry, we achieved in vivo H2 generation in tumors in a controllable manner and exploited it for effective and green therapy of tumors for the first time. The cathodic acupuncture electrodes working under an applied voltage of ∼3 V (with minimal damage to the living body) undergo effective electrochemical reactions in the acidic tumor area that produce sufficient H2 locally to cause cancer cells to burst and die. Due to puncture positioning, the acidic tumor microenvironment and gas diffusion effect, the developed H2 generation electrochemotherapy (H2-ECT) strategy enables precise and large-scale tumor therapy, as demonstrated by in vivo treatment of diseased mice (glioma and breast cancers). Such green H2-ECT is simple, highly efficient and minimally invasive, requiring no expensive medical equipment or nano materials and medication, and is therefore very promising for potential clinical applications.


INTRODUCTION
Up to now, cancer is still one of the major diseases that threaten the survival of mankind, and it is difficult to cure clinically [1]. In addition to single or combined surgery, chemotherapy and radiotherapy, which are commonly used clinically [2], a number of promising therapeutic strategies have been recently put forward including immunotherapy and gene therapy, photothermal therapy (PTT), photodynamic therapy (PDT), and so on [3][4][5][6][7][8][9][10]. However, these techniques usually rely on chemical and genetic drugs or exotic nanomaterials to actualize treatments, making them quite difficult or debatable for practical clinical applications in the near future due to the uncertainties of potential biotoxicity and biohazards, and related genetic and ethical issues [11,12]. Also, immunotherapy and gene therapy are complex and expensive [13,14]. Therefore, the popularization of these techniques in clinical practice is restricted. Consequently, the development of simple, green, efficient and cheap treatment methods is urgently needed to combat cancer.
Hydrogen (H 2 ), owing to its small molecular size and physiological inertness, resistance to oxidation, and good gas diffusivity in vivo, is considered as a kind of green and endogenous gas [15]. H 2 can easily penetrate into the biological membrane to disperse into cytoplasm and other organelles such as the nucleus, mitochondria and so on due to its small size. It performs eminent physiological/pathological regulatory functions, and this has been exploited for the treatment of many diseases, such as Alzheimer's disease, arthritis and diabetes [16][17][18][19]. In 2007, Shigeo Ohta et al. demonstrated that the use of H 2 as a therapeutic antioxidant can selectively reduce cytotoxic oxygen radicals [16]. Several reports have also confirmed that a low concentration of H 2 has therapeutic effects against local inflammation such as eye, ear, nose, liver and systemic inflammation [20][21][22][23][24]. Dole et al. attempted to use the antioxidation ability of hyperbaric H 2 to treat skin squamous cell carcinoma in 1975 [25], but the need to provide hyperbaric H 2 with diving medical equipment limited the possible clinical application of hyperbaric H 2 in tumor therapy. Subsequently, to obtain a high concentration of H 2 , the conventional routes of H 2 administration are to inhale H 2 gas and intake of H 2rich water or saline [22,23,26]. These approaches,  The H 2 -production polarization curves of acupuncture electrodes recorded, respectively, in simulated body fluid (SBF, pH = 6.0) under three different voltages (C), in SBF but with different pH (5.5, 6.0, 6.5, 7.0 and 7.5) at 3 V (D), and in fresh pork (mimicking complex tumor tissue) at two pHs (6.0 and 7.4) (E). (F) Dissolved amount of iron from the anode during the EC process. (G) The dependence of H 2 production of the system on various pH levels in SBF recorded at 3 V for 10 min (n = 3). (H, I) The dependence of the amounts of dissolved iron and produced H 2 on voltages varying from 0 to 3 V, each for 10 min duration, at pH = 6.0 and 7.4, respectively (n = 3). All the data are represented as mean ± SEM (standard error of the mean). n denotes samples or number of experiments. n = 3, P < 0.0001 representing the statistical significance of different pH values (pH = 7.5 and pH = 6.0), which is evaluated by two-tailed Student's t-test with GraphPad Prism software.

RESEARCH ARTICLE
due to the lack of targeting for the tumor, along with poor bioavailability caused by poor water solubility of H 2 , limited their therapeutic efficacy. More recently, Zhao et al. tried using near-infrared (NIR) light-responsive PdH 0.2 nanocrystals as carriers to deliver H 2 and realize synergistic therapy of photothermal and H 2 therapy [27]. However, it is still difficult to achieve sufficient H 2 release in vivo with this method and it has potential biotoxicity due to the introduction of nanoparticles. To the best of our knowledge, no successful cancer therapy by the exploitation of H 2 alone has been reported. In fact, how to produce H 2 non-invasively and sufficiently without using nanomaterials and how to achieve the release of H 2 on demand in vivo are two huge challenges facing the H 2 therapy of cancers. Acupuncture is a traditional and unique minimally invasive method to treat diseases in China. It is quite effective for the treatment of systemic diseases, especially arthritis, cervical spondylopathy, psoatic strain and so on [28,29], but applying it to the treatment of major diseases, such as cancers, is still a great challenge. In this study, by innovative combined use of the traditional Chinese acupuncture Fe needle (electrode) and in vivo electrochemistry, we developed a simple and precise cancer therapy approach based on selective electrochemical

RESEARCH ARTICLE
generation of H 2 in the tumor region, termed in vivo H 2 generation electrochemotherapy (H 2 -ECT). Very impressively and excitingly, as depicted in Fig. 1A, the two acupuncture electrodes inserted into the tumor lump work well at ∼3 V direct current (DC) electric field (which causes minimal damage to healthy tissues), driving the electrochemical H 2 generation reaction effectively under the acidic tumor microenvironment and destroys solid tumors. Importantly, the H 2 -ECT method enables largescale tumor therapy by applying a gas diffusion effect, avoiding the shortcoming of limited effective area for classic electrochemical reactions. Moreover, due to the puncture positioning and acidic tumor microenvironment (compared to normal tissue), the method provides ideal selectivity and targeting to precisely kill tumors with minimal damage to normal tissue, which is very promising for potential clinical applications. The effectiveness of the method has been demonstrated by successful and fast treatment of glioma and breast cancers in mice in this study.

Characterization of H 2 generation in vitro by the H 2 -ECT method
To verify the production of H 2 during the in vivo electrochemical (EC) process, we first carried out in vitro EC studies by using two medical acupuncture needles, whose main component is stainless steel, as cathode and anode electrodes. Figure 1B depicts the two-electrode system used in the experiments. The whole EC reactions are carried out at the electrode/electrolyte interfaces under an applied DC electric field and the simulated body fluid (SBF, pH = 6.0), in which the H 2 evolution reaction and the acupuncture electrode corrosion process were carried out at the cathode and anode, respectively. The reaction equations are as follows: Cathode reaction : 2H + + 2e − → H 2 ; Anode reaction : Fe − 2e − → Fe 2 + . Figure 1C shows typical polarization curves of the cathode in the SBF under three different voltages. H 2 is gradually generated during the EC process when the cathode voltage is greater than ∼−1.64 V. The known half-reaction standard electrode potentials (E 0 ) for Fe/Fe 2+ , Fe/Fe 3+ and Fe 2+ /Fe 3+ are 0.44, 0.036 and −0.77 V, respectively, relative to the standard electrode potential of H 2 [30][31][32]. In this study, the overpotential of acupuncture electrode to produce H 2 was about 1.2 V, which is much higher than the E 0 of Fe/Fe 2+ for iron dissolution. The overpotential is caused by the surface passivation of iron in a weakly acidic solution [33,34]. When cathode potential exceeded the passivation potential, the anode electrolysis was increased ( Fig. S1 in the online supplementary material). Under the EC conditions, the anode and cathode produced ferrous iron (Fe 2+ ) and H 2 , respectively, affirmed by X-ray photoelectron spectroscopy (XPS) and methyl blue (MB) probe UV-Vis spectrum characterizations (Figs S2 and S3). The pH effect of media on the H 2 production of the system was also checked by recording the polarization curves of the electrodes in SBF with various pH values. The results showed that the more acidic the solution, the more H 2 was produced (Fig. 1D). The result is the same when using fresh pork (to simulate complex tumor tissue) of two pH values ( Fig. 1E and Fig. S4B), which is favorable for potential H 2 -ECT applications in vivo since tumor tissue is weakly acidic compared to neutral healthy tissue. To facilitate the evaluation of the therapeutic efficacy of the method, we measured the iron consumption of the acupuncture electrode semiquantitatively using inductively coupled plasma mass spectrometry (ICP-MS), and calculated the volume of H 2 generation under the various conditions (see online supplementary material for detailed calculation).
As shown in Fig. 1F and G, the iron consumption of the acupuncture electrode and the volume of H 2 generated were gradually decreased as the pH of SBF increased from 5.5 to 7.5 during the 3 V electrolytic process. We then examined the amount of iron consumption and H 2 generation volume at pH 7.4 and 6.0, respectively, under three different applied voltages (0, 1.2, 2.1 and 3.0 V) each for 10 min duration. The results indicated that at the two pH levels the volume of H 2 generated was rapidly increased as voltage increased, especially at pH 6.0 ( Fig. 1H and I). This dependency and flexibility of H 2 generation on pH and voltage make the method promising and operational in a controlled manner for precise H 2 therapy of cancer in vivo.
Before applying the method to in vivo cancer therapy, we assessed the potential cytotoxicity of the dissolved Fe 2+ in vitro by live/dead staining and MTT assay, using MCF-7 and C6 cells as models. The cells were treated with different concentrations of Fe 2+ (from 5 μM to 50 mM). As shown in Fig. S5, the cytotoxicity of Fe 2+ with concentrations less than 5 mM was negligible as iron is an essential nutrient and the most abundant transition metal in the human body [35]. In our system, since the maximum concentration of Fe 2+ produced by 10 min electrolysis of the electrode at the applied voltage of 3 V (at pH 6.0) was estimated to be about 2.5 μM, the biotoxicity of Fe 2+ is not a concern. For safety, before cancer treatment in vivo, we also checked the injury degree of the H 2 -ECT method for healthy tissue versus can-cer tissue, by using the fresh pork immersed in SBF with pH 7.4 and 6.0, respectively, as mimics. As expected, the injury degree of tissue at pH 6.0 was more serious than that on the neutral tissue (Fig. S6A), implying the effectiveness of the method for cancer therapy in vivo. We also performed an additional hematoxylin and eosin (H&E) staining assay to further accurately discriminate the injury degree of tissue at different pH levels (pH = 6.0 and pH = 7.4). As shown in Fig. S6B, degeneration and necrosis of the muscle fibers of cytoplasm were observed in the tissue at pH = 6.0 after H 2 -ECT, indicating that cells were seriously damaged. However, after the H 2 -ECT, the muscle fibers were still intact for the tissue at pH = 7.4, which indicated that this method is basically harmless to normal tissue. Consequently, the H 2 -ECT has great medical value in the treatment of cancer.

Characterizations of H 2 generation in vivo during the H 2 -ECT process
Firstly, for better observing the therapeutic effect, the tumor-bearing BALB/c mice with similar average tumor size were divided into different groups before H 2 -ECT (Fig. S7). To assess H 2 production efficiency in vivo during the H 2 -ECT process, two closely spaced acupuncture electrodes were inserted carefully and precisely with the naked eyes into the tumor lump of a C6 tumor-bearing mouse and the depth of insertion of the electrodes was decided by the depth and thickness of the tumor. The part of the electrode outside the tumor was insulated by encasing with plastic hose to avoid short circuits and unnecessary damage to normal tissue (Fig. S8). As shown in Fig. 2A, the acupuncture electrodes were gradually electrolyzed with increasing voltages applied from 0 to 3 V, with some surface corrosion observed by microscope after the in vivo treatment, indicating the occurrence of the EC reactions. The voltage-related H 2 production in nude mouse solid tumors was further evaluated by in vivo T1-weighted magnetic resonance imaging (MRI) and computed tomography (CT) imaging. As clearly shown in Fig.  2B and C, as the treatment voltage became higher than 2.1 V, cavities emerged in the tumors, indicating effective gas (H 2 ) production within the tumor after the H 2 -ECT treatment. Moreover, with the increase of voltage, the cavity area increased gradually, confirming the effectiveness and flexibility of the method for cancer therapy in vivo. Figure 2D shows the in vivo H 2 -generation polarization curves of the cathode Fe electrodes recorded inside the tumor in a three-electrode system (with the AgCl-treated Ag puncture electrode as the reference electrode), under three different applied volt-ages. The result is quite similar to the in vitro curves in SBF (pH = 6, cf. Fig. 1C), indicating effective occurrence of the EC reactions in the acidic tumor microenvironment (which provides sufficient protons for the reaction). We further checked electrogravimetric loss (Fe dissolution) of the anode after the H 2 -ECT treatment (Fig. 9). The electrogravimetric loss of the electrode increases with increasing voltage, illustrating increased iron consumption (Fig. 2E). Then, the average amount of H 2 produced by a single cathode electrode during each 10-min treatment session can be calculated electrochemically according to the above-mentioned equations, to be ∼0.35, 2.95 and 7.69 mL under the applied voltages of 1.2, 2.1 and 3.0 V, respectively. Therefore, in the whole H 2 -ECT process (10 min twice a day for 3 days), the total amount of H 2 produced in a diseased mouse at a corresponding voltage was ∼2.1, 17.7 and 46.1 mL, respectively. The calculated results are consistent with that obtained by Faraday's law of electrolysis (Fig. S10). These results also agreed well with the electrogravimetric analysis. For further safe tumor therapy in vivo, dynamic processes of H 2 generation and Fe consumption of the electrodes during the in vivo H 2 -ECT treatment were monitored (Fig. S11) and the results showed that the amounts of H 2 generation and Fe consumption tended to stop increasing after the electrolysis time reached more than 10 min during one treatment (Fig. S11B). The reason is that limited tumor local electrolytes near the electrodes will be depleted during one rushing EC run due to the diffusion limitation. In addition, locally growing H 2 bubbles will increase the EC resistance and slow down the reaction. Fortunately, such self-protection mechanism further reduces the damage to healthy tissue, and these can be easily addressed by intermittent H 2 -ECT or multi-needle treatment (especially for large tumors) to improve therapeutic effects. Therefore, 10 min per treatment is safe and effective and will be used as an optimal time in the following studies.

Evaluation of the therapeutic effect of H 2 -ECT in vivo
The developed H 2 -ECT approach was then exploited for in vivo treatment of glioma in diseased mice as it is one of the deadliest cancers and difficult to cure, with a 5-year survival rate of patients less than 5% [36] due to the strongest malignant degree. Consequently, the C6 tumor-bearing BALB/c nude mice were used as a tumor xenograft model to evaluate the in vivo tumor therapy capability of the H 2 -ECT method. The control groups and therapeutic groups were administrated to assess the therapeutic performances, respectively (Fig. 3A). Tumor-bearing mice with similar average tumor size were randomly divided into five groups: the blank, needle stimulation (NS + 0 V), NS + 1.2 V, NS + 2.1 V and NS + 3.0 V groups. Although the establishment of the tumor model and feeding of the mice were exactly under the same conditions, there was some obvious volume variation of tumor lumps in the same group due to the individual differences between mice. When the tumor size reached ∼200 mm 3 , each tumor-bearing mouse was treated for 10 min a time twice a day for 3 days using H 2 -ECT and then observed for 12 days, as shown in Fig. S8. The tumor sizes and body weights of the mice were recorded every day. As shown in Fig. 3B (Fig. 3F). During the 3-day treatment and subsequent 12-day physical recovery, the body weights of mice in control and all therapeutic groups showed no significant changes (Fig. 3C), manifesting the low systemic toxicity of the H 2 -ECT method. Although scab of the treatment area was observed after the H 2 -ECT treatment at 3.0 V, it recovered very well during the recovery process, as shown in Fig. S12. Impressively, the NS + 3.0 V group had the H 2 -ECT treatment conducted  only for the first 3 days, but maintained remarkable tumor inhibition effect during the following 12-day observation, indicating that H 2 -ECT is a simple and effective method for tumor therapy. Simultaneously, we also compared the therapeutic process under the 1.2 and 2.1 V treatments and found that the volumes of tumor showed no significant change in the first 4 days after the treatment, but the volumes of the tumor increased gradually during the following recovery process. It implied that insufficient dosage of H 2 generated during the whole H 2 -ECT process may lead to the rebound of suppressed tumor growth (Fig. S12) To gain insights into the therapy efficacy, tumor tissues of mice in the different groups were collected and analyzed by hematoxylin and eosin (H&E) staining assays. As shown in Fig. 3G, clear deformation and shrinking of the nuclei and destruction of the membrane integrity were observed in the NS + 3.0 V and NS + 2.1 V groups, indicating that tumor cells were seriously damaged. The tumors of the other groups had no distinct injury. To further confirm the destruction of the tumor by this method, the tumors were sliced after H 2 -ECT and stained using the immunofluorescence terminal deoxynucleotidyl transferase-mediated dUTPbiotin nick end labeling (TUNEL). As clearly seen from Fig. 3H, an intensive green fluorescence, a characteristic of apoptotic signal caused by DNA fragmentation, was observed in the NS + 3.0 V group as compared to the other groups, confirming that the H 2 -ECT is capable of activating cell apoptosis in tumor tissues. However, the underlying molecular and cellular mechanisms are unknown and require further study.

RESEARCH ARTICLE
To confirm the vital role of H 2 generation in the H 2 -ECT, we further investigated the effects of temperature and applied DC electric field on tumor growth. As shown in Fig. S13, since the temperature of the tumor remained basically unchanged before and after the H 2 -ECT treatment, the thermal effect in the H 2 -ECT process can be ignored. To check possible electrical field effect on tumor treatment, we also performed experiments under the same EC conditions (3 V, 10 min) but using a silver needle with poorer EC H 2 production capability instead of a stainless steel needle. As seen from Fig. S14, since the EC activity of silver (compared to Fe) is too low to effectively generate enough H 2 , the tumor suppression effect was obviously weakened. All these results fully illustrated the important role of sufficient H 2 generation in the tumor therapy.
Next, to verify the generality of the method, we also applied the H 2 -ECT approach for the therapeutic treatment of another tumor model (MCF-7 breast cancer tumor), which was established by subcutaneously injecting MCF-7 breast cancer cells into the hind limb space of BALB/c nude mice (5 weeks old). Tumor-bearing mice were randomly divided into three groups: the blank, NS + 0 V and NS + 3.0 V groups. Similarly, the H 2 -ECT treatment under an applied voltage of 3 V was also applied for the MCF-7 tumor-bearing mice for 10 min twice a day for 3 days. After successive 3-day therapeutic periods and subsequent 12 days of observation, similar high-efficiency tumor suppression on the breast cancer was also achieved in the NS + 3.0 V group (Fig. S15). Our preliminary results implied the generality and rosy prospect of the H 2 -ECT for future minimally invasive treatments of cancers.

In vivo biosafety of the H 2 -ECT
To evaluate the biosafety of the H 2 -ECT method, bio-distribution of iron element in the main organs for two kinds of tumor was determined using ICP-MS. As shown in Fig. 4A and B, iron element was found mainly accumulated into the tumor, without serious effects on other organs (heart, liver, spleen, lung, kidney and muscle). Moreover, the histopathological evaluation of major organs and hematology-related assays were carried out after the in vivo H 2 -ECT experiments. As shown in Fig. 4C and Fig. S16, major organs stained by H&E showed no significant pathological changes in the main organs between the control groups and the treatment groups. Meanwhile, the blood chemistry analyses were performed via checking the standard nine hematological biomarkers (see Figs S17 and S18), the results of which showed no distinct abnormality before and after the treatment for 15 days, indicating negligible side effects of the H 2 -ECT method. In addition, the microelements analyses ( Fig. 4D and E, and Figs S19 and S20) showed no distinct microelement abnormality before and after the H 2 -ECT treatment, and the iron element in blood remained at a healthy level even after the treatment of H 2 -ECT at 3 V. All the above results showed the excellent biosafety of the H 2 -ECT method.

DISCUSSION
Although H 2 therapy with several successful clinical trials has been developed in the past decade, there remain huge challenges regarding its efficacy and methods. Compared to other biomedical applications, H 2 therapy of cancer is rarely reported due to the lack of effective methods to control the production and release of enough H 2 in the body. Due to the disadvantages of insufficient and uncontrollable release of H 2 along with potential metabolic toxicity, current attempts to use nanomaterials as H 2 carriers are still far from clinical oncotherapy.
In our study, we report a green and conceptually new in vivo H 2 generation electrochemotherapy (H 2 -ECT) of tumors, by combined use of Chinese acupuncture Fe needle electrode and in vivo electrochemistry. The method solves the above problems by virtue of the acidic microenvironment of the tumor and in vivo electrochemical technology. The acupuncture electrodes inserted in the acidic tumor area, and working under the applied voltage of 3 V, produce effectively sufficient H 2 to damage and kill cancer. The method is highly biosafe without an obvious poisoning effect, solving the difficult problems of material metabolism and biotoxicity in conventional therapeutic methods for clinical applications. By taking advantage of the puncture positioning, gas diffusion effect and self-protection mechanism of H 2 generation, the H 2 -ECT method provides strong selectivity and targeting to effectively and precisely kill tumors, with minimal damage to healthy tissues. The effectiveness of the method has been demonstrated by in vivo treatment of glioma and breast cancers of diseased mice. More importantly, as the volume of the tumor tissue determines the total amount of hydrogen ions within the tumor, the larger tumors provide more hydrogen ions to produce enough hydrogen for effective treatment. Consequently, in

RESEARCH ARTICLE
principle, irrespective of the size of the tumor, the inherent volume effect of the method can solve the defect of insufficient hydrogen generation during the therapy process.
Although currently the method may not be valid for the treatment of non-solid tumors like lung cancer and liver cancer, our method is promising and reliable for the treatment of solid tumors. With the help of various advanced imaging techniques and cytopathology knowledge, we can accurately locate the tumor and carry out accurate treatment in future clinic applications, avoiding damage to normal tissue to the greatest extent. The developed H 2 -ECT is simple, safe, efficient and minimally invasive and therefore very promising for potential clinic applications to cure solid tumors. We believe that the H 2 -ECT method will also be used for the treatment of deep tumors with the help of accurate medical imaging technology in the future.

Animal experiments
BALB/c nude mice (4 weeks old, 15 ± 0.2 g, male) were purchased from Beijing HFK Biotechnology Ltd (Beijing, China). Animal care and handing procedures were in agreement with the guidelines of the Regional Ethics Committee for Animal Experiments.

Tumor models and in vivo H 2 -ECT treatments
The MCF-7 and C6 cells were cultured in DMEM supplemented with 10% FBS at 37 • C in an atmosphere of 5% CO 2 . The MCF-7 and C6 cells were then collected and washed in cold PBS three times, respectively. After that 5 × 10 6 cells/mL was dispersed into PBS and 50 μL of cell solution was injected into each mouse. The mice were used for further experiments when the tumor had grown to 8-10 mm in diameter. The mice were then divided into five groups at random: blank, needle stimulation (NS + 0 V) and NS + (1.2 V, 2.1 V and 3.0 V) and three mice were set as one group. For observation of the therapy effect of H 2 -ECT, C6 tumor-bearing mice were firstly anesthetized using intraperitoneal injection of chloral hydrate solution (10 wt%). The purpose of anesthesia was to make the mice unconscious and eliminate pain, and keep them still during the H 2 -ECT therapy. Then the tumor sites were disinfected using 75% alcohol and subsequently two acupuncture electrodes were inserted into the tumor lump. The acupuncture electrode was inserted into the central part of the tumor tissue, which can avoid damage to the normal tissue. From an electrochemical point of view, the distance between the two electrodes will affect the efficiency of H 2 generation in vivo to some extent. Therefore, in order to avoid short circuits and excessive electrochemical resistance in tumor tissues, and thus achieve the effectiveness and repeatability of the treatment, we chose the optimal distance (between two electrodes) of ∼3-5 mm for the treatment in this study. Finally, the two electrodes were connected to the positive and negative poles of the voltage regulator, respectively, and single H 2 -ECT treatment carried out under the various applied voltages from 0 to 3.0 V for 10 min. Each tumor-bearing mouse was treated with H 2 -ECT twice per day for 3 days and then observed for another 12 days. To avoid excessive injury to normal tissue, the negative pole should be inserted into the center of the tumor bulk. The insertion depth of the acupuncture electrode depends on the depth and thickness of the tumor and the outside part of the acupuncture electrodes was insulated to avoid unnecessary injury to normal tissue. Although acupuncture Fe electrodes can be reused for H 2 -ECT, since they are very cheap (less than $0.1) and readily available, we used them only once per treatment to avoid infection and contamination between mice.
The tumor size was measured using the digital caliper in two dimensions and the tumor volume was calculated by the following equation: tumor volume = (length × width 2 )/2. Meanwhile, the body weight changes of mice were recorded every day to estimate the efficacy of physical recovery. On the 16th day, all the mice were dissected to collect their organs (heart, liver, lung, kidney and spleen) fixed by paraformaldehyde (4 wt%) for pathological analysis. The tumors were weighed to assess therapy efficacy of different experimental groups.