Abstract

   Inflammatory bowel disease (IBD) affects millions of people each year.
   The overproduction of reactive oxygen species (ROS) plays a critical
   role in the progress of IBD and will be a potential therapeutic target.
   Here, we synthesize a kind of oral zero-valent-molybdenum nanodots
   (ZVMNs) for the treatment of IBD by scavenging ROS. These ultrasmall
   ZVMNs can successfully pass through the gastric acid and then be
   absorbed by the intestine. It has been verified that ZVMNs can
   down-regulate the quantity of ROS and reduce colitis in a mouse IBD
   model without distinct side effects. In addition, RNA sequencing
   reveals a further mechanism that the ZVMNs can protect colon tissues
   from oxidative stress by inhibiting the nuclear factor κB signaling
   pathway and reducing the production of excessive pro-inflammatory
   factors. Together, the ZVMNs will offer a promising alternative
   treatment option for patients suffering from IBD.
     __________________________________________________________________

   Oral zero-valent-molybdenum nanodots were developed for the treatment
   of inflammatory bowel disease by scavenging reactive oxygen.

INTRODUCTION

   Inflammatory bowel disease (IBD) is a nonspecific chronic inflammatory
   disease that affects certain segments of the gastrointestinal tract and
   often causes extraintestinal complications ([44]1). Approximately 6.8
   million cases of IBD have been reported in 2017, and the incidence and
   prevalence of IBD are increasing worldwide, which brings a huge
   challenge to the health care system ([45]2–[46]4). Although some
   anti-inflammatory agents, immunosuppressants, and biologic agents such
   as infliximab and adalimumab have been used for IBD treatment in
   clinic, long-term use of these drugs will lead to a series of side
   effects including autoimmune responses, viral infection, and
   tumorigenesis ([47]5–[48]7). Therefore, it is urgent to explore more
   effective and suitable treatment for the therapy of IBD.

   Despite the complicated and poorly understood pathogenesis of IBD, many
   studies have pointed out that reactive oxygen species (ROS) play an
   important role in the progress of IBD ([49]1, [50]8). ROS up-regulation
   has been observed in colitis tissues of patients with IBD and murine
   colitis models, which eventually contributes to the intestinal mucosa
   injury ([51]9–[52]11). First, pro-inflammatory cytokines can activate
   immune cells and amplify both inflammation and ROS generation in the
   occurrence of IBD ([53]12, [54]13). Meanwhile, excessive ROS production
   will further activate the inflammatory/immune response through nuclear
   factor κB (NF-κB) signal pathway, leading to increased expression and
   secretion of pro-inflammatory cytokines ([55]14–[56]16). As a result,
   high ROS exposure gives rise to oxidative damage to mitochondria and
   consequently induces apoptosis of intestinal epithelial cells (IECs)
   ([57]16–[58]20). Therefore, scavenging the ROS in the gut is necessary
   to halt IBD progression. Unfortunately, several nonenzymatic and
   enzymatic antioxidants currently are not efficient, and many of them
   will cause undesirable immunological response ([59]21, [60]22). A few
   trials have shown that the curative effect of natural antioxidants such
   as vitamin E, vitamin C, and coenzyme Q was not impressive ([61]23). In
   addition, it is reported that the ROS scavenger N-acetyl cysteine
   potentiated the generation of T helper 17 (T[H]17) cells (T[H] cells)
   in vivo in a Misshapen/NIK-related kinase 1-dependent manner and
   increased the risk of promoting autoimmune inflammatory diseases
   ([62]24). Hence, it is of great significance to develop novel
   antioxidants to overcome these limitations.

   Recently, Mo-based nanomaterials such as MoS[2] and phosphomolybdate
   have achieved much attention in the area of antioxidants because of
   their favorable biocompatibility and effective antioxidant properties
   ([63]25–[64]27). For example, our group has reported that the
   molybdenum-based polyoxometalate nanoclusters could act as antioxidants
   for acute kidney injury repair with favorable ROS scavenging capability
   and biocompatibility ([65]26). Here, we synthesize a new kind of oral
   zero-valent-molybdenum nanodots (ZVMNs) through mechanical stripping
   for IBD treatment. Because of the high stability in an acidic
   environment, ZVMNs can be taken orally and accumulate in the gut.
   Because of the strong reducing action of zero-valent-molybdenum atoms,
   ZVMNs can effectively scavenge ROS and remodel the inflammatory
   microenvironment in dextran sulfate sodium (DSS)–induced IBD murine
   models. Both in vitro and in vivo experiments confirm the therapeutic
   effect of the ZVMNs. Moreover, the related molecular mechanisms are
   observed by RNA sequencing (RNA-seq) and 16S RNA-seq. We believe that
   this research will open up new perspectives for the treatments of IBD.

RESULTS AND DISCUSSIONS

Synthesis and characterization of ZVMNs

   ZVMNs were synthesized by mechanical exfoliation of the mixture of Mo
   powder and isopropanol. Transmission electron microscopy (TEM) image
   showed that the size of the as-prepared ZVMNs was about 3 nm ([66]Fig.
   1A). The x-ray diffraction (XRD) pattern of Mo powder was the same as
   the standard powder diffraction file ([67]Fig. 1B). However, after
   mechanical exfoliation, there were no obvious peaks in the ZVMNs,
   indicating the super small particle size of ZVMNs, which may promote
   the reaction rate ([68]Fig. 1B). Next, the reaction between ZVMNs and
   ROS was characterized by using H[2]O[2] on behalf of ROS. The solution
   of ZVMNs was bluish violet, while after oxidation, the products were
   colorless (fig. S1). The XRD pattern showed that the products were
   between MoO[3] and MoO[3]·H[2]O ([69]Fig. 1C). The peak spread of XRD
   and TEM image showed that the size of products was rather small (fig.
   S2). The increased binding energy of Mo 3d orbital of products in x-ray
   photoelectron spectroscopy (XPS) proved the raised valence state of Mo
   ([70]Fig. 1D). Hence, these products were called hexavalent-molybdenum
   nanoparticles (HVMNs).

Fig. 1. Characterization of ZVMNs.

   [71]Fig. 1.
   [72]Open in a new tab

   (A) TEM images of ZVMNs. (B) XRD pattern of ZVMNs and Mo powder. (C)
   XRD pattern of hexavalent-molybdenum nanoparticles (HVMNs). (D) XPS
   spectra of Mo 3d of ZVMNs and HVMNs. R space of Mo K-edge x-ray
   absorption fine structure (EXAFS) of (E) ZVMNs and (F) HVMNs. Wavelet
   transform of EXAFS of (G) ZVMNs and (H) HVMNs. (I) Ability of ZVMNs to
   scavenge H[2]O[2] at different concentrations (n = 5, means ± SD). (J)
   Zeta potential of ZVMNs in water and HCl solutions. (K) Hydrated radius
   of ZVMNs in water and HCl solutions. (L) Photographs of ZVMNs in water
   and HCl solutions.

   Mo K-edge–extended x-ray absorption fine structure (EXAFS) was further
   carried out to analyze the structure of ZVMNs and HVMNs. Through
   comparison with standard sample, ZVMNs had major Mo─O bonds but few
   Mo─Mo bonds ([73]Fig. 1E and fig. S3, A and B). HVMNs also had major
   Mo─O bonds ([74]Fig. 1F and fig. S3B), and the fitted coordination
   number (table S1) was similar to ZVMNs. Wavelet transform of EXAFS
   provided more information about the difference between ZVMNs and HVMNs.
   As shown in [75]Fig. 1 (G and H), most coordinate atoms in ZVMNs were
   located farther than that in HVMNs, indicating ZVMNs had a looser
   structure. On the basis of the above data, we supposed that ZVMNs
   consisted of loosely assembled monatomic Mo with O-containing ligands
   (such as hydroxyl and H[2]O), and HVMNs consisted of closely assembled
   MoO[3] clusters with bound water. Then, the ability of ZVMNs to
   scavenge H[2]O[2] was detected. As shown in [76]Fig. 1I, ZVMNs could
   effectively reduce the content of H[2]O[2] in solution.
   Quantificationally, each Mo atom could scavenge about 0.27 H[2]O[2]
   molecules. We further measured the ability of ZVMNs to scavenge other
   kinds of ROS. As shown in fig. S4 (A and B), the ZVMNs exhibited
   favorable superoxide anions and hydroxyl radicals eliminating
   efficiencies, further indicating the high ROS scavenging activities of
   ZVMNs.

   Last, to clarify whether ZVMNs could be taken orally, the stability of
   ZVMNs in saline and gastric acid was evaluated. The hydrated radius of
   ZVMNs in saline showed few differences at 1-, 7-, and 14-day periods
   (fig. S6A), indicating the stability of ZVMNs in saline. Diluted HCl
   solutions (pH 1.5) were applied to simulated gastric acid. Because of
   the abundant H^+, zeta potential of ZVMNs in HCl solutions decreased in
   half compared with that in water ([77]Fig. 1J), leading to increased
   hydrated radius ([78]Fig. 1K). Nevertheless, the TEM image showed the
   aggregation of ZVMNs rather than dissolution in HCl solutions (fig.
   S5). In addition, the hydrated radius of ZVMNs showed few differences
   at 1-, 7-, and 14-day periods (fig. S6B), and the color did not change
   ([79]Fig. 1L), further indicating that ZVMNs could stay stable in HCl
   solutions. In summary, all these data present the favorable performance
   of scavenging ROS and the feasibility to be taken orally of ZVMNs.

Biocompatibility assessment of ZVMNs

   Cytotoxicity assay kit and cell counting kit–8 (CCK-8) assessment and
   the staining of living and dead cells were performed to detect the
   biocompatibility of the ZVMNs in vitro. HCT116 cells (human colon
   cancer cells) and immortalized bone marrow–derived macrophages (iBMDMs)
   exposed to different concentrations of the ZVMNs for 24 hours exhibited
   normal cell morphology. Both CCK-8 assay results (figs. S7 and S9A) and
   calcein/propidium iodide (PI) staining results (fig. S8) indicated that
   the ZVMNs at test concentrations did not exhibit noticeable
   cytotoxicity. Moreover, blood routine and biochemistry indexes (fig.
   S9, B to D) and histopathology assays of the main organs (fig. S10)
   after ZVMNs administration demonstrated no observable impairments.
   These results are in accordance with previous findings in literature
   that the Mo-based nanomaterials are nontoxic at relatively low
   concentrations ([80]26, [81]28, [82]29).

ZVMNs protect iBMDM cells from ROS-induced damage

   Given the fact that the ZVMNs have strong antioxidative property, their
   potential capability in scavenging ROS in vitro was measured. A
   cellular inflammation model was established by treating iBMDMs with
   lipopolysaccharide (LPS). After LPS stimulation, an ROS-sensitive
   fluorescent dye, 2′,7′-dichlorofluorescin diacetate (DCFH-DA), was
   applied to stain the cells. As shown in [83]Fig. 2A, the intracellular
   green fluorescent signal of iBMDMs increased markedly after treatment
   with LPS. In contrast, the intracellular ROS level obviously decreased
   when the iBMDM cells were pretreated with ZVMNs. We further performed
   the quantitative analysis of intracellular ROS levels via flow
   cytometry. The results ([84]Fig. 2, B and C) showed that ZVMNs
   eliminated about 66.7% of intracellular ROS from LPS-treated iBMDM
   cells, demonstrating the favorable ROS scavenging capability of ZVMNs.
   To further confirm the anti-inflammatory activity of ZVMNs, the
   messenger RNA (mRNA) expression levels of inflammatory cytokines
   stimulated by LPS in the iBMDM cell line were detected. As shown in
   [85]Fig. 2 (D to F), the levels of interleukin-1β (IL-1β), IL-10, and
   tumor necrosis factor–α (TNF-α) in the ZVMN-treated group were declined
   significantly compared to the levels in the LPS group (P < 0.05),
   indicating that the ZVMNs could relieve inflammation responses in
   vitro.

Fig. 2. ZVMNs protect iBMDM cells from ROS-induced damage.

   [86]Fig. 2.
   [87]Open in a new tab

   (A) Representative ROS staining (green fluorescence) of iBMDM cell
   lines under different treatment conditions. Scale bar, 10 μm. (B) ROS
   levels in untreated and ZVMN-treated iBMDM cell lines incubated with
   LPS. (C) Statistical analysis of ROS levels in the iBMDM cell lines
   under different treatment conditions (n = 3, means ± SD). Messenger RNA
   (mRNA) levels of pro-inflammatory cytokines including interleukin-1β
   (IL-1β) (D), IL-10 (E), and tumor necrosis factor–α (TNF-α) (F) (n = 3,
   means ± SD). (G) ECAR reflects the glycolytic flux. (H) OCR indicates
   mitochondrial respiration (n = 12, means ± SD). Significance between
   every two groups was calculated using Mann-Whitney U test. *P < 0.1,
   **P < 0.01, ***P < 0.001, and ****P < 0.0001. 2-DG, 2-Deoxy-Glucose.

   Furthermore, metabolic flux analysis using Seahorse technology was
   performed. The extracellular acidification rate (ECAR) showed
   significant changes between LPS-stimulated cells and ZVMN-treated
   cells, indicating that the ZVMNs affected glycolytic function ([88]Fig.
   2G). The basal oxygen consumption rate (OCR) showed no changes in basal
   mitochondrial respiration, but maximal mitochondrial respiration in
   LPS-stimulated iBMDM cells was found in ZVMN-treated cells ([89]Fig.
   2H).

Therapeutic efficacy of ZVMNs in DSS-induced IBD mice model

   On the basis of the good ROS scavenging activity of the ZVMNs in vitro,
   its therapeutic efficacy in vivo was then investigated for the
   treatment in the DSS-induced C57BL/6 mice IBD model. Mice were orally
   treated with 2.5% DSS in the drinking water to induce colitis ([90]Fig.
   3A). All mice were randomly divided into four groups, including the
   control group, model group (DSS), 5-aminosalicylic acid (5-ASA) group
   (DSS + 5-ASA), and ZVMNs treatment group (DSS + ZVMNs). Among them, the
   5-ASA group was set as the positive control, which was the first-line
   drug in IBD treatment but with a risk of relapse and multiple adverse
   effects. The 5-ASA and ZVMNs were given separately by oral gavage every
   day. The therapeutic efficacy of the ZVMNs was assessed by measuring
   the body weight changes, the disease activity index (DAI), colon
   lengths, and levels of pro-inflammatory cytokines of IL-1, TNF-α, IL-6,
   and interferon-γ (IFN-γ) in colon tissues and histological analysis of
   colon sections.

Fig. 3. Favorable therapeutic Efficacy of ZVMNs in the mice model with
DSS-induced colitis.

   [91]Fig. 3.
   [92]Open in a new tab

   (A) Experimental design of DSS-induced IBD model mice. The mice were
   provided with sterile water or water containing 2.5% DSS for 7 days.
   Oral administration of treatments was given from the second day. All
   mice were euthanized 1 day after DSS drinking stopped. (B) Macroscopic
   colon appearance of each group was shown. (C) Colon lengths of mice
   with indicated treatments on day 8 (n = 6, means ± SD). (D)
   Representative macroscopic spleen appearance of each group. (E) Spleen
   weight of mice with indicated treatments was measured and analyzed (n =
   6, means ± SD). (F) Daily changes of body weight were recorded in
   detail every day (n = 6, means ± SD). (G) Everyday DAI scores were
   recorded and analyzed (n = 6, means ± SD). (H) Representative
   hematoxylin and eosin (H&E) and MPO histology images of colon tissue of
   each group. Scale bar, 200 μm. Significance between every two groups
   was calculated using Mann-Whitney U test. *P < 0.1, ***P < 0.001, and
   ****P < 0.0001. ns, not significant.

   Compared with the DSS group, the DSS + ZVMN–treated group showed a
   longer colon length (P < 0.0001; [93]Fig. 3, B and C), lighter spleen
   weight (P < 0.001; [94]Fig. 3, D and E), heavier body weight (P <
   0.001; [95]Fig. 3F), and lower DAI (P < 0.0001; [96]Fig. 3G). These
   results indicate the significant therapeutic effect of the ZVMNs on
   mice with DSS-induced colitis. In the histological analysis, mice with
   colitis showed severe destruction of crypts, massive infiltrations of
   immune cells, and severe colonic epithelial damage in the inflamed
   colon, while nearly normal histological microstructure and less
   inflammatory cell infiltration were observed in DSS + ZVMNs group. The
   myeloperoxidase (MPO)–positive neutrophils were significantly augmented
   in the colon tissue of the DSS group ([97]Fig. 3H), and the ZVMNs
   markedly reduced its congregation, further verifying that the ZVMNs
   successfully could relieve the disease states of IBD.

Therapeutic mechanisms of ZVMNs on IBD

   To further elucidate the therapeutic mechanism of the ZVMNs to IBD,
   RNA-seq of colonic tissues of mice was performed. An unguided principal
   components analysis (PCA) of the data revealed substantially different
   transcriptomic profiles among the control group, DSS group, and DSS +
   ZVMNs group (fig. S11). Volcano plots showed that 3229 genes are
   up-regulated, and 2818 genes were down-regulated in DSS-induced colitis
   mice compared to the control group. While 2550 genes were up-regulated,
   and 1863 genes were down-regulated after the ZVMNs treatment ([98]Fig.
   4A). Among the up-regulated and down-regulated genes, there were 1731
   genes and 1234 genes overlapped between those two sets of comparison,
   respectively, which could be attributed to the presence of inflammation
   ([99]Fig. 4B).

Fig. 4. RNA-seq analysis of DSS-induced colitis regulated by ZVMNs.

   [100]Fig. 4.
   [101]Open in a new tab

   (A) Volcano plot exhibiting the differentially expressed genes (fold
   change > 1.5; P adjust < 0.05; up-regulated genes: red; down-regulated
   genes: blue) in the DSS group compared to the control group (left) and
   in the DSS group versus the ZVMN-treated group (right) from RNA-seq
   data. (B) Venn diagram of RNA-seq analysis showing significantly
   up-regulated (bottom) or down-regulated (top) genes in the DSS group
   compared to the control group and in the DSS group versus the
   ZVMN-treated group. (C) Gene set enrichment analysis of response to ROS
   genes between different conditions. NES, normalized enrichment score.
   (D) Heatmap from RNA-seq analysis indicating the response to
   ROS-related genes in the control group or DSS group and ZVMN-treated
   group (fold change in the right). (E) KEGG pathways enriched the
   up-regulated genes in the DSS group compared to the control group
   (left) and in the DSS group versus the ZVMN-treated group (right).
   JAK-STAT, Janus kinase–signal transducers and activators of
   transcription; PD-L1, programmed cell death ligand 1; PD-1, programmed
   cell death protein 1; NOD, nonobese diabetic; MAPK, mitogen-activated
   protein kinase. AGE-RAGE, Advanced Glycation End products-the Receptor
   of Advanced Glycation End products.

   Similarly, among the top ranking up-regulated signaling pathways
   (hypergeometric tests, P < 0.05), 15 overlapped signaling pathways were
   revealed in the comparison of colonic sections after the ZVMNs
   treatment in the colitis mice, according to Kyoto Encyclopedia of Genes
   and Genomes (KEGG) pathway enrichment analysis (fig. S12). Gene set
   enrichment analysis (GSEA) revealed that the up-regulated genes of
   response to ROS are significantly enriched in the DSS group ([102]Fig.
   4C), which were decreased to normal level under ZVMNs treatment. As a
   further analysis, the normalized heatmap showed that the gene signature
   of response to ROS was predominantly decreased in IBD mice after the
   ZVMNs treatment ([103]Fig. 4D). Functional pathway enrichment analysis
   displayed that the TNF signaling pathway, IL-17 signaling pathway,
   chemokine signaling pathway, and NF-κB signaling pathway are highly
   associated with the therapeutic mechanisms of the ZVMNs ([104]Fig. 4E).
   Literature has reported that ROS could promote the production of
   pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6. TNF-α could
   further trigger a strong cascade inflammatory response throughout the
   TNF-α/NF-κB signaling pathways, resulting in an excessive inflammation
   response in IBD ([105]30). Consistently, the above results indicated
   that the ZVMNs could reduce colitis in mice via inhibiting the NF-κB
   signaling pathway through decreasing the ROS level.

   To further confirm the therapeutic mechanisms of ZVMNs in vivo, we
   first detected the ROS levels in the colonic mucosa by using
   dihydroethidium (DHE) staining. The green staining was positive for ROS
   content. As shown in [106]Fig. 5A, the ROS level in ZVMN-treated IBD
   mice was similar to that of healthy mice, indicating that the ZVMNs
   could function as antioxidants to scavenge ROS in vivo. Besides, as we
   know, the mitochondria are the central organelles in energy metabolism
   and are exposed to ROS. Excess ROS generation leads to lower ATP
   production, suppression of the intracellular electron transport chain,
   and DNA damage in mitochondria ([107]21, [108]31, [109]32). However, we
   found that there was no significant difference in mitochondrial DNA
   (mtDNA) between the three groups (control group, DSS group, and DSS +
   ZVMNs group) (fig. S13), which was probably due to the short modeling
   period.

Fig. 5. Therapeutic mechanisms of the ZVMNs on IBD.

   [110]Fig. 5.
   [111]Open in a new tab

   (A) Representative ROS staining (green fluorescence) of colon tissues
   of each group by DHE. Scale bar, 500 μm. (B) Western blot analysis of
   the expression of phospho–NF-κB p65 (P-NF-κB p65), total NF-κB p65,
   phospho–IκB-α (P-IκB p65), and total IκB-α in colon tissues of
   DSS-induced colitis mice. Colonic mRNA levels of IFN-γ (C), IL-6 (D),
   IL-1β (E), TNF-α (F), and IL-10 (G) (n = 6, means ± SD). Significance
   between every two groups was calculated using Mann-Whitney U test. **P
   < 0.01, ***P < 0.001, and ****P < 0.0001. IκB, inhibitor of NF-κB.

   Next, we validated the effects of the ZVMNs on the key proteins in the
   NF-κB signaling pathway, including NF-κB, phosphorylation of NF-κB,
   inhibitor of NF-κB (IκB), and phosphorylation of IκB. Western blotting
   of colon tissues was performed. As shown in [112]Fig. 5B, the
   phosphorylation of NF-κB and IκB were significantly enhanced in the
   DSS-induced colitis mice, indicating the activation of NF-κB signaling
   pathway in IBD. While the phosphorylation of NF-κB and IκB was
   significantly decreased in the ZVMN-treated group, indicating that the
   ZVMNs could inhibit NF-κB signaling pathway, in accordance with the
   RNA-seq analysis result. The downstream inflammatory factors of the
   NF-κB signaling pathway were also detected. The mRNA expression of
   pro-inflammatory cytokines in the colonic mucosa was measured,
   including IFN-γ, IL-6, IL-1β, TNF-α, and IL-10, which played important
   roles in the pathologies of IBD ([113]Fig. 5, C to G). Compared to the
   control group, these cytokines are highly increased in IBD group but
   significantly reduced in the ZVMN treatment group, which was consistent
   with our findings in vitro. According to the previous research, IL-1β
   is oversecreted because of the increased activity of the NOD-like
   receptor thermal protein domain associated protein 3, NLRP3
   inflammasome on the development of IBD. Meanwhile, IL-1β can also
   stimulate the release of ROS by neutrophils, which will further
   increase TNF-α synthesis ([114]31, [115]32). IL-6 and IFN-γ are also
   main pro-inflammatory cytokines activating immune cells. In patients
   with IBD, levels of IL-6 and IFN-γ will increase and mediate the
   activation of T cells, which eventually aggravate the intestinal
   inflammation ([116]32). In summary, the production of IL-1β, IL-6,
   IFN-γ and TNF-α by immune cells and IECs presented the key
   pro-inflammatory events in the initiation of IBD, while a markedly
   reduction in the inflammatory cytokines was observed, proving the
   efficacy of the ZVMNs.

   Besides, it is widely known that the perturbation of intestinal
   microbiota is densely associated with IBD onset and exacerbation
   ([117]33). The intestinal microbiota has been identified as a large
   source of pro-oxidants ([118]34). Excessive production of ROS could
   also result in the gut barrier dysfunction and the expansion of
   potentially harmful bacteria such as facultative anaerobic bacteria
   ([119]35). Now, targeting gut microbiota dysbiosis and overly
   intestinal inflammation with dietary or microbial interventions—such as
   the applications of probiotics, antibiotics, defined enteral
   nutritional therapy (ENT), and fecal microbiota transplantation—have
   become attractive strategies for IBD treatment ([120]33). To date,
   several studies have shown that microbial interventions could improve
   the disease severity. Nevertheless, part of these studies are not
   always repeatable. In addition, the gut microbiota varies greatly among
   individuals, targeting gut microbiota does not always bring benefits.
   For example, ENT has been reported to be associated with rapid changes
   in gut microbiota composition that could lead to even greater dysbiosis
   relative to healthy individuals ([121]33). Thus, to explore whether the
   ZVMNs affect the composition or abundance of gut microbiota, 16S
   ribosomal RNA-seq (rRNA-seq) technology was carried out to analyze the
   changes of mice enterobacterial composition. The results showed no
   significant difference in alpha diversity between the three groups
   (control group, DSS group, and DSS + ZVMNs group) (figs. S14 and S15).
   Principal Co-ordinates analysis (PCOA) indicated that the treatment of
   ZVMNs partly altered the composition of microflora in mice with colitis
   (fig. S16). Further investigation should be carried out in the future
   to clarify the effect of ZVMNs on the gut microbiota.

   IBD is a refractory chronic inflammatory disease, which has a close
   connection with excess ROS production. In this research, we have
   developed ZVMNs with the capability of antioxidant to treat IBD via
   oral administration. With IECs markedly impaired by IBD, oral
   administration could provide fast access to the inflamed region and
   avoid the inconvenience of injections and the associated pain. Then,
   ZVMNs can accumulate at the inflammatory site, scavenge the
   overproduced ROS, and repair the enlarging tight junctions and the
   increasing permeability. Combined with the good biocompatibility of
   ZVMNs, these features will facilitate the application of ZVMNs in
   clinic.

   The as-prepared ZVMNs with ultrasmall particle size exhibited good
   physiological stability especially in gastric acid, favorable
   biosafety, and significant ROS scavenging abilities. The structure of
   ZVMNs and their interaction with ROS has been clarified by XRD, XPS,
   and EXAFS. Both in vitro and in vivo experiments demonstrated the
   favorable therapeutic effects of the ZVMNs for IBD. RNA-seq revealed a
   potential mechanism that the ZVMNs could protect colon tissues from
   oxidative stress by inhibiting the NF-κB signaling pathway and reducing
   the production of excessive pro-inflammatory factors ([122]Fig. 6). In
   addition, under laboratory conditions, one ultrasonic disrupter can
   produce about 300 mg of ZVMNs [or 300 ml of 1000–parts per million
   (ppm) ZVMNs solution] each day. Higher ultrasonic power and more
   ultrasonic disrupters will bring higher yield of ZVMNs. The cost of
   ZVMNs is about $5/g in consideration of the recycled raw materials (Mo
   powder and isopropanol), energy charge, and the abrasion of the
   ultrasonic amplitude transformer. Hence, ZVMNs have much potential for
   industrial production. We believe that our findings will provide a
   potential alternative treatment option for patients with IBD.

Fig. 6. Schematic illustration of ZVMNs in the treatment of IBD.

   [123]Fig. 6.
   [124]Open in a new tab

   The ZVMNs protect colon tissues from oxidative stress by inhibiting the
   NF-κB signaling pathway and reducing the production of excessive
   pro-inflammatory factors.

MATERIALS AND METHODS

Experimental apparatus

   TEM image was carried out by FEI Talos F200X. XRD was measured by
   Rigaku D/Max 2250 V. Dynamic light scattering was measured by Malvern
   Zetasizer Nano S. XPS spectra were measured by Thermo Fisher Scientific
   ESCALAB 250Xi. Confocal laser scanning microscopy was carried out on
   Leica TCS SP8 STED 3X. The content of Mo was measured by inductively
   coupled plasma optical emission spectrometry (Agilent 725). The
   fluorescence microplate system was TECAN Spark. Mechanical exfoliation
   was carried out by an ultrasonic disrupter (Scientz JY92-IIN). Hydroxyl
   radicals were generated by Varian Clinac 21EX (Trilogy). The data of
   EXAFS were collected at the beamline of TPS44A1 in the National
   Synchrotron Radiation Research Center. For wavelet transform analysis,
   χ(k) exported from Athena was imported into the HAMA FORTRAN code. The
   parameters were listed as follow: R range, 0 to 6 Å; k range, 0 to 13.9
   Å^−1; k weight, 3; and Morlet function with κ = 10, σ = 1 was used as
   the mother wavelet to provide the overall distribution.

Reagents

   Mo powder, isopropanol, 5-ASA, salicylic acid, and sodium chloride were
   purchased from Macklin. Phosphate-buffered saline (PBS), Dulbecco’s
   modified Eagle’s medium (DMEM), and fetal bovine serum (FBS) were
   obtained from Gibco. Streptomycin and penicillin were obtained from
   Adamas.

Synthesis of ZVMNs

   ZVMNs were synthesized by mechanical exfoliation (450 W, 75% duty
   cycle, 20 hours) of the mixture of 3 g of Mo powder and 50 ml of
   isopropanol. The supernatant was taken out, and the isopropanol was
   removed by a rotary evaporator. At last, the as-prepared ZVMNs were
   dispersed in deionized water.

Cell culture

   HCT116 cell line and mice iBMDMs were cultured in DMEM supplemented
   with 10% FBS, streptomycin (100 μg/ml), and penicillin (100 U/ml) at
   37°C in an incubator supplied with a humidified atmosphere of 5% CO[2].

Cell viability

   HCT116 cells and iBMDMs were seeded at 5 × 10^3 cells per well in
   96-well plates. After 1 day, cells were cocultured with different
   concentrations of nanomaterials. The cell viability was tested by using
   a CCK-8 kit (Dojindo Laboratories) or calcein/PI cell
   viability/cytotoxicity assay kit (Beyotime). By incubating the CCK-8
   reagent with HCT116 cells and iBMDMs for 2 hours, the live cells
   reacting with the reagent can be detected by a microplate reader. After
   incubating with the calcein/PI buffer for 30 min, the live or dead
   cells can be observed under a fluorescent microscope.

Quantitative reverse transcription PCR

   To assess the anti-inflammatory effect, iBMDM cells were seeded into
   six-well plates, which were stimulated by LPS (1 μg/ml) and then
   incubated with 50-ppm ZVMNs. The total RNA of cell and mice colon
   tissue was extracted by TRIzol reagent (Takara) and then transcribed
   into cDNA by the Hifair II First Strand cDNA Synthesis Kit (Yeasen).
   Next, quantitative polymerase chain reaction (qPCR) was performed using
   Hieff qPCR SYBR Green Master Mix (Yeasen) on LightCycler 480 Instrument
   II (Roche). The following qPCR conditions were used: 40 cycles of
   denaturation at 95°C for 10 s and annealing at 60°C for 30 s. A
   comparative threshold cycle method was used to analyze the qPCR data,
   where the amount of target was normalized to the endogenous reference
   of β-actin in each sample. The primer sequences used were as follows:
   β-actin (forward: CGTTGACATCCGTAAAGACC; reverse: TAGGAGCCAGAGCAGTAATC),
   IL-10 (forward: CAGGGATCTTAGCTAACGGAAA; reverse:
   GCTCAGTGAATAAATAGAATGGGAAC), TNF-α1 (forward: CAGGCGGTGCCTATGTCTC;
   reverse: CGATCACCCCGAAGTTCAGTAG), IL-1β (forward:
   TTCAGGCAGGCAGTATCACTC; reverse: GAAGGTCCACGGGAAAGACAC), IL-6 (forward:
   CTGCAAGAGACTTCCATCCAG; reverse: AGTGGTATAGACAGGTCTGTTGG), and IFN-γ
   (forward: GCCACGGCACAGTCATTGA; reverse: TGCTGATGGCCTGATTGTCTT).

DSS-induced colitis mouse model

   All the experiments in vivo were approved by the Animal Ethics
   Committee of Shanghai Jiao Tong University, and the ethics number was
   B-2021-009. Six-week-old male C57BL/6 mice were housed in groups of six
   mice per cage and acclimatized for 1 week before inclusion in the
   study. Mice were randomly divided into four groups, including the
   normal control group, DSS-induced colitis group, DSS-induced colitis
   treated with the 5-ASA group, and the DSS-induced colitis treated with
   the ZVMNs group. Mice with colitis were induced by DSS (2.5 weight %,
   molecular weight: 36,000 to 50,000; MP Biomedicals) added in the
   drinking water for 7 days. The 5-ASA (50 mg/kg) and ZVMNs (4 mg/kg)
   were given daily by oral gavage. During the period, weight loss, stool
   consistency, and fecal blood of mice were evaluated daily to determine
   the DAI. On day 8, all mice were euthanized under isoflurane anesthesia
   and the colon, spleen, liver, lung, kidney, and heart were collected
   for further analysis.

Biocompatibility in vivo

   The healthy mice were divided into three groups randomly. ZVMNs (4
   mg/kg) were taken via oral administration into healthy mice of two
   groups. The other one group served as the control group. After ZVMNs
   oral administration of different times (7 and 30 days), the mice were
   euthanized. The blood of mice was collected for serum biochemistry
   tests and blood routine index tests.

Histological analysis

   For histological analyses, the colon, spleen, liver, lung, kidney, and
   heart were fixed with paraformaldehyde (4% in PBS), embedded in
   paraffin wax, stained with hematoxylin and eosin, and analyzed by
   microscopy. For immunohistochemistry, tissue sections were
   deparaffinized, rehydrated, and rinsed, followed by antigen retrieval
   and blocking. Next, the tissue sections were incubated with anti-MPO
   (Abcam) prepared in 5% goat serum overnight in a moist chamber and then
   the Alexa Fluor–conjugated secondary antibody. After that, the tissue
   sections were observed under the fluorescence microscope. ROS staining
   was performed using a ROS staining kit (Beyotime) following the
   manufacturer’s protocol. A 0.5-cm-long colon section taken 1 cm from
   the anus was used to make frozen sections. Prepared colon sections were
   then incubated with DHE for 30 min at 37°C. After washing, colon
   sections were then incubated with 4′,6-diamidino-2-phenylindole (DAPI)
   solution at room temperature for 10 min to accomplish DAPI
   counterstain. Last, ROS staining was observed under a fluorescence
   microscope.

ROS scavenging in solution

   The content of H[2]O[2], superoxide anions, and hydroxyl radicals were
   measured by a hydrogen peroxide assay kit (Beyotime), total superoxide
   dismutase assay kit with nitroblue tetrazolium (Beyotime), and
   salicylic acid, respectively. Different concentrations of ZVMNs (0, 10,
   20, 30, and 40 ppm) were applied to scavenge H[2]O[2] (about 100 μM),
   superoxide anions (xanthine/xanthine oxidase), and hydroxyl radicals
   (10-gray x-rays). The content of H[2]O[2], superoxide anions, and
   hydroxyl radicals was measured by the absorption at 560, 560, and 510
   nm, respectively.

ROS scavenging in vitro

   iBMDM cells were seeded into six-well plates, and the cells were
   stimulated by LPS (1 μg/ml) and then incubated with 50-ppm ZVMNs.
   DCFH-DA (Sigma-Aldrich), an oxidation-sensitive fluorescent dye, was
   used to detect the intracellular ROS level according to the previous
   literature. After the aforementioned incubation with LPS for 6 hours,
   cells were gently rinsed with serum-free medium, and then 10 μM DCFH-DA
   was added to the cells and incubated in dark at 37°C for 30 min.
   Afterward, the cells were washed with serum-free medium thrice to
   remove unloaded DCFH-DA probe, then were imaged using a laser confocal
   microscope, and were subjected to a flow cytometry analysis to quantify
   the intracellular ROS levels, respectively.

Western blot

   The total proteins from colon tissue were extracted using
   radioimmunoprecipitation assay lysis buffer containing phosphatase and
   protease inhibitor. Bicinchoninic acid protein assay kit was used to
   determine the concentrations of extracted proteins. Eight percent of
   SDS-PAGE gel was prepared for an equal amount of protein from each
   sample to run and then transferred to polyvinylidene difluoride (PVDF)
   membranes. PVDF membranes were blocked with 5% skim milk and incubated
   with primary antibodies (including NF-κB p65, phospho–NF-κB p65, IκB-α,
   phospho-IκB-α, and β-actin) at 4°C overnight, followed by incubation
   with secondary horseradish peroxidase–conjugated antibodies for 1 hour
   at room temperature.

Seahorse analysis

   To perform real-time ECAR and OCR analyses, iBMDMs were analyzed using
   the Seahorse XF24 Analyzer from Agilent Technologies by Mito Stress
   Test Kit or Glycolysis Stress Test Kit according to the manufacturer’s
   instructions. A total of 40,000 cells were used in each assay with n =
   12 replicates in three independent experiments.

RNA-seq analysis

   C57BL/6 mice were euthanized to collect the frozen colon sections. The
   total RNA of samples was extracted using TRIzol reagent (Invitrogen)
   according to the manufacturer’s instructions, and genomic DNA was
   removed using deoxyribonuclease I. Then, RNA quality was analyzed by
   2100 Bioanalyzer (Agilent) and quantified by the NanoDrop Technologies.
   Only a high-quality RNA sample was used to construct the sequencing
   library. The isolation of the desired RNA molecules, reverse
   transcription of RNA to cDNA, amplification of primed cDNA molecules,
   and sequencing were performed at Majorbio Bio-Pharm Biotechnology Co.
   Ltd. using Illumina HiSeq X10 (Illumina) according to the
   manufacturer’s instructions. For bioinformatics analysis, the
   expression level of each transcript was calculated according to the
   fragments per kilobase of exon per million mapped reads method. HTSeq
   was used to count the read numbers mapped to each gene. Data analysis
   was performed using R software (R Foundation for Statistical
   Computing). The DESeq2 was used to normalize the raw counts and
   identify differentially expressed genes (|fold change| ≥ 1.5; false
   discovery rate < 0.05) and generate the principal component plot. Gene
   Ontology enrichment analysis was performed by R package cluster
   Profiler, where the differentially expressed genes identified as
   described above were supplied as the input for genes by function
   enriched KEGG. Hypergeometric tests were used to calculate P values,
   which were then subjected to multiple testing adjustments by
   Benjamini-Hochberg correction. GSEA was performed to test whether
   interesting gene sets are substantially enriched in corresponding
   conditions. Besides, protein-protein interactions of genes were
   analyzed by the Search Tool for the Retrieval of Interacting
   Genes/Proteins algorithm ([125]www.string-db.org/). The Cytoscape
   software was used to screen for hub genes according to degrees.

Microbiome analysis

   Feces of each mouse were collected separately on day 8 and stored at
   −20°C. Total genomic DNA was extracted by using the E.Z.N.A. soil DNA
   Kit (Omega Bio-Tek, Norcross, GA, USA). The bacterial 16S rRNA gene
   (V3-V4 region) was amplified using the primer pairs 338F
   (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) in
   an ABI GeneAmp 9700 PCR thermocycler (ABI). Microbiota composition was
   determined on Illumina MiSeq platform (Illumina) according to the
   standard protocols from Majorbio Bio-Pharm Technology Co. Ltd.
   Operational taxonomic units (OTUs) were clustered in UPARSE (version
   7.1) and the taxonomy of each OTU representative sequence was analyzed
   by RDP Classifier. All data were lastly analyzed on the Majorbio
   Biocloud platform.

Statistical analysis

   Statistical analyses were conducted using SPSS software version 19.0
   (IBM) and GraphPad Prism 9.0 (GraphPad). Quantitative variables are
   expressed as means ± SD or median, whereas categorical variables are
   expressed as numbers and percentages. The Mann-Whitney U test and
   analysis of variance (ANOVA) were used to compare the differences in
   two or more groups when appropriate. P < 0.05 was considered
   statistically significant.

Acknowledgments