Abstract

   Protopanaxadiol (PPD) has potential immunomodulatory effects, but the
   underlying mechanism remains unclear. Here, we explored the potential
   roles of gut microbiota in the immunity regulation mechanisms of PPD
   using a cyclophosphamide (CTX)‐induced immunosuppression mouse model.
   Our results showed that a medium dose of PPD (PPD‐M, 50 mg/kg)
   effectively ameliorated the immunosuppression induced by CTX treatment
   by promoting bone marrow hematopoiesis, increasing the number of
   splenic T lymphocytes and regulating the secretion of serum
   immunoglobulins and cytokines. Meanwhile, PPD‐M protected against
   CTX‐induced gut microbiota dysbiosis by increasing the relative
   abundance of Lactobacillus, Oscillospirales, Turicibacter,
   Coldextribacter, Lachnospiraceae, Dubosiella, and Alloprevotella and
   reducing the relative abundance of Escherichia‐Shigella. Importantly,
   PPD‐M lost the ability to promote bone marrow hematopoiesis and enhance
   immunity when the gut microbiota was depleted by broad‐spectrum
   antibiotics. Moreover, PPD‐M promoted the production of
   microbiota‐derived immune‐enhancing metabolites including cucurbitacin
   C, l‐gulonolactone, ceramide, DG, prostaglandin E2 ethanolamide,
   palmitoyl glucuronide, 9R,10S‐epoxy‐stearic acid, and
   9′‐carboxy‐gamma‐chromanol. KEGG topology analysis showed that the
   PPD‐M treatment significantly enriched the sphingolipid metabolic
   pathway with ceramide as a main metabolite. Our findings reveal that
   PPD enhances immunity by manipulating gut microbiota and has the
   potential to be used as an immunomodulator in cancer chemotherapy.

   Keywords: chemotherapy, gut microbiota, immunity, immunomodulator,
   immunosuppression, protopanaxadiol
     __________________________________________________________________

   CTX treatment leads to immunosuppression and gut microbiota dysbiosis
   in mice. PPD treatment can protect against CTX‐induced gut microbiota
   dysbiosis by increasing the abundance of beneficial bacteria and
   inhibiting the proliferation of pathogenic bacteria. The beneficial
   bacteria metabolize to produce immune‐enhancing metabolites and
   anti‐inflammatory metabolites, which contribute to the immune‐enhancing
   and anti‐inflammatory functions of PPD‐M in CTX‐induced
   immunosuppressed mice, including promoting bone marrow hematopoiesis to
   increase the number of peripheral blood white blood cells,
   proliferating splenic lymphocytes, and eliminating inflammation.

   graphic file with name MCO2-4-e222-g006.jpg

1. INTRODUCTION

   Chemotherapy is the main method for the treatment of cancers,
   especially for patients with moderate and advanced metastatic tumors.
   However, chemotherapy agents, such as cyclophosphamide (CTX), tend to
   cause significant immunosuppression, including myelosuppression,
   leukopenia, anemia, and gastrointestinal mucosal barrier damage.[44] ^1
   , [45]^2 , [46]^3 The emergence of immunosuppression will lead to a
   relative reduction in the use of chemotherapy drugs, reduce the
   therapeutic effect, and result in infections and mortality in cancer
   patients. Therefore, discovering an effective immunomodulator to
   alleviate immunosuppression is of great significance to reduce the side
   effects and enhance the efficacy of chemotherapy drugs.

   Ginsenosides, the major active ingredients found in ginseng and other
   plants of the genus Panax, are very popular for their antioxidant,
   strengthening the immune system and many other activities.[47] ^4 ,
   [48]^5 Protopanaxadiol (PPD) is a major active metabolite from PPD‐type
   ginsenosides and possesses pleiotropic anticancer activities in many
   cancers.[49] ^6 , [50]^7 , [51]^8 Enhancing immunity is one of the main
   functions of ginsenosides. Whether PPD also regulates immunity has
   rarely been studied. Only one article reported that PPD significantly
   enhances the antilung cancer effect of CTX. In this process, PPD
   increases the spleen index, peripheral white blood cell (WBC) count,
   natural killer cell activity, and the levels of interleukin‑2 (IL‑2)
   and interferon‑γ (IFN‑γ) in CTX‑treated tumor‑bearing mice,[52] ^9
   suggesting that PPD may ameliorate the immunosuppression induced by
   CTX. However, the mechanism by which PPD enhances immunity remains
   unclear.

   The gut microbiota plays an important role in maintaining immune
   homeostasis.[53] ^10 , [54]^11 Under normal physiological conditions,
   the gut microbiota is in a dynamic balance with the immune system,
   which plays a critical role in maintaining human health.[55] ^12
   Disruption of gut microbiota will upset this balance, and the immune
   system will be compromised, leading to inflammation and related
   diseases.[56] ^10 , [57]^13 Chemotherapeutic agents often disrupt the
   ecological balance of gut microbiota, lead to the overexpression of
   certain pathogenic bacteria, and promote the occurrence of chronic
   inflammation and immunosuppression.[58] ^14 , [59]^15 Therefore, the
   concurrent use of gut microbiota modulators in cancer chemotherapy is
   considered to be a promising strategy. Many active ingredients in
   plants have been found to regulate gut microbiota and improve
   immunity.[60] ^16 , [61]^17 , [62]^18 , [63]^19 , [64]^20 Whether PPD
   can regulate gut microbiota has not been reported.

   Here, we investigated the protective effects of PPD on the immune
   system and gut microbiota in CTX‐induced immunosuppressed mice.
   16S‐rRNA sequencing and untargeted metabolomic analysis were employed
   to reveal the mechanism by which PPD enhances immunity. It was
   confirmed that PPD promoted bone marrow hematopoiesis and enhanced
   immunity by regulating gut microbiota to ameliorate the
   immunosuppression induced by CTX.

2. RESULTS

2.1. PPD improves immune function in immunosuppressed mice

   ICR mice were intraperitoneally injected with CTX to construct an
   immunosuppressive model (Figure [65]S1). The effects of PPD on immune
   function in the immunosuppressed mice were evaluated by splenic
   lymphocyte subsets, inflammatory factors, immunoglobulins, spleen, and
   thymus. As shown in Figure [66]1A, mice were prophylactically treated
   with different concentrations of PPD for 14 days before CTX injection.
   On the 17th day, the blood of the fundus venous plexus, right femur,
   thymus, and spleen of mice was collected for the following experiments.
   The obtained results showed that, compared with the control (CTR)
   group, peripheral blood WBCs were significantly reduced in the CTX
   group, PPD significantly ameliorated this reduction, and the medium
   dose of PPD (50 mg/kg, PPD‐M) had the best effect (Figure [67]1B).
   Similarly, the spleen index and thymus index of mice in the CTX group
   significantly decreased, PPD significantly reversed the spleen index
   and thymus index depression caused by CTX, and PPD‐M had the best
   effect (Figures [68]1C and D, Figure [69]S2). Here, PPD‐H (100 mg/kg)
   is less effective than PPD‐M in enhancing immunity, which may be due to
   certain cytotoxicity of higher concentration of PPD‐H. Our previous
   studies have found that PPD‐H, but not PPD‐M, has significant antitumor
   effects.[70] ^21 Therefore, PPD‐M was used as an immunomodulator to
   treat immunosuppressed mice in the following experiments.

FIGURE 1.

   FIGURE 1
   [71]Open in a new tab

   Effects of PPD on immunity in immunosuppressed mice. (A) Mice were
   intragastrically treated with different concentrations of PPD (PPD‐L:
   25 mg/kg, PPD‐M: 50 mg/kg, PPD‐H: 100 mg/kg) for 16 days, and the CTR
   group and CTX group mice were intragastrically administered PPD
   vehicle. Intraperitoneal injection of CTX (50 mg/kg) was performed on
   day 14 (CTR group mice with CTX vehicle). Blood was collected from the
   retro‐orbital venous plexus and mice were sacrificed on day 17. (B)
   Changes of WBCs in the peripheral blood of mice in each group. (C)
   Spleen index of mice in each group. (D) Thymus index of mice in each
   group. (E) The expression of CD3^+ T lymphocytes in the mouse spleen.
   (F) The expression of CD4^+ and CD8^+ T lymphocytes in the mouse
   spleen. (G) The levels of IgG, IgM, IgA, IL‐4, IFN‐γ, and TNF‐α in
   serum. (n = 6, *p < 0.05, **p < 0.01, ***p < 0.001, “ns” means no
   significant difference.)

   Flow cytometry results showed that the number of CD3^+ T lymphocytes
   and the ratio of CD4^+/CD8^+ T lymphocytes in the spleen were
   significantly decreased in the CTX group, while PPD‐M treatment
   significantly reversed the decrease (Figures [72]1E and F). Serum
   immunoglobulin and cytokine levels were detected by ELISA. As shown in
   Figure [73]1G, the serum levels of IL‐4 and IgG were significantly
   decreased after CTX treatment, while PPD‐M treatment significantly
   increased their levels. Conversely, the contents of IFN‐γ and TNF‐α
   were clearly increased after CTX treatment, whereas PPD‐M reduced them
   to the control level. In addition, the IgM level in the PPD‐M group was
   significantly higher than that in the CTX group.

   The HE staining analyses of the spleen and thymus of mice are shown in
   Figure [74]S3. Compared with the CTR group, the splenic capsule
   thickness of the CTX group was uneven, the number and volume of white
   pulp were decreased, and the boundary between white and red pulp was
   unclear. However, in the PPD‐M group, these symptoms were improved, and
   a diffuse distribution of extramedullary hematopoietic cells was
   observed in the red pulp. The thymus volume of mice in the CTX group
   was reduced and the medulla structure was unclear. Although the
   structure of the thymus was also unclear in the PPD‐M group, the thymus
   volume was normal. The abovementioned results demonstrate that PPD can
   restore the impaired immune function induced by CTX in immunosuppressed
   mice.

2.2. PPD ameliorates leukopenia by promoting bone marrow hematopoiesis

   Chemotherapy is known to target bone marrow cells to induce leukopenia,
   and its recovery may also occur through bone marrow hematopoiesis.[75]
   ^22 The number of bone marrow nucleated cells can reflect the strength
   of bone marrow hematopoietic function. Therefore, bone marrow
   histopathology and flow cytometry were used to analyze the effect of
   PPD on bone marrow hematopoiesis in immunosuppressed mice. A flow
   cytometry gating strategy was used to identify hematopoietic stem and
   progenitor cells (HSPCs), common lymphoid progenitors (CLPs), common
   myeloid progenitors (CMPs), granulocyte‐macrophage progenitors (GMPs),
   and megakaryocyte‐erythrocyte progenitors (MEPs)[76] ^23 (Figure
   [77]S4).

   Flow cytometry results showed that PPD‐M treatment significantly
   increased the number of HSPCs compared with those in the CTR and CTX
   groups (Figure [78]2A). Moreover, CTX could significantly reduce CMP,
   MEP, and CLP levels, among which PPD‐M clearly reversed the CLP
   reduction caused by CTX (Figures [79]2B and C). However, the GMP level
   in the CTX group was significantly higher than that in the CTR group,
   which may be a temporary increase from hematopoietic stress,[80] ^24
   and PPD‐M treatment further enhanced the level of GMPs. HE staining
   analysis of mouse bone marrow is shown in Figure [81]2D. Compared with
   the CTR group, nucleated cells in the bone marrow of mice in the CTX
   group were significantly reduced, and a large area of hemorrhage was
   observed, while these phenomena were improved in the PPD‐M group. These
   results suggest that PPD promotes HSPC self‐renewal and differentiation
   in immunosuppressed mice.

FIGURE 2.

   FIGURE 2
   [82]Open in a new tab

   PPD improves bone marrow suppression induced by CTX in mice. (A)
   Changes in bone marrow hematopoietic stem and progenitor cells (HSPCs,
   KLS^+) in the CTR, CTX, and PPD‐M groups. (B) and (C) Changes in bone
   marrow hematopoietic progenitor cells (CMPs, GMPs, MEPs, and CLPs) in
   the CTR group, CTX group and PPD‐M group. (D) HE staining of mouse
   femur tissue. (Data are expressed as the mean ± SD. n = 4, *p < 0.05,
   **p < 0.01, ****p < 0.0001, “ns” means no significant difference.)

2.3. PPD improves the composition and structure of gut microbiota in
immunosuppressed mice

   Chemotherapy is typically associated with dysbiosis of the gut
   microbiota.[83] ^15 To study whether PPD could prevent CTX‐induced gut
   microbiota dysbiosis, we performed high‐throughput gene sequencing of
   16S rRNA from the fecal bacteria of mice. The results of the dilution
   curves of the Chao diversity index and Shannon diversity index
   indicated that most bacterial diversity and many new phylotypes could
   be captured (Figure [84]S5). To determine the gut microbiota richness
   and diversity, the ACE, Chao, Shannon, and Simpson indices were used to
   evaluate the α‐diversity. As shown in Table [85]1, CTX reduced the
   richness and diversity of gut microbiota with lower ACE, Chao, and
   Shannon indices and a higher Simpson index. Of note, PPD reversed this
   phenomenon. These results indicate that PPD has a regulatory effect on
   the gut microbiota and increases the richness and diversity of
   intestinal bacteria.

TABLE 1.

   α‐Diversity indices of gut microbiota in each group
   Groups ACE            Chao            Shannon     Simpson
   CTR    425.89 ± 51.48 434.47 ± 59.55  3.98 ± 0.31 0.05 ± 0.02
   CTX    390.95 ± 94.30 395.90 ± 102.48 3.57 ± 0.61 0.10 ± 0.08
   PPD‐M  424.39 ± 69.32 430.25 ± 73.75  3.95 ± 0.20 0.05 ± 0.01
   [86]Open in a new tab

   ACE (Abundance‐based Coverage Estimator) and Chao indices reflect
   community richness, and the larger the value, the higher the community
   richness. Shannon (Shannon‐Wiener index) and Simpson (proposed by
   Edward Hugh Simpson) indices represent community diversity. The higher
   the Shannon value, the higher the community diversity. Conversely, the
   lower the Simpson value, the higher the community diversity.

   Partial least squares discriminant analysis (PLS‐DA) was used to
   compare the similarity between groups and showed that the mouse samples
   of the CTR group, CTX group and PPD‐M group could be clearly
   distinguished and clustered into three groups (Figure [87]3A),
   indicating that the gut microbiota composition of the three groups of
   mice had significant differences. The proportions for shared and unique
   OTUs are shown in a Venn diagram (Figure [88]3B). There were 505 shared
   OTUs among the three groups. The number of unique OTUs in the CTR, CTX,
   and PPD‐M groups was 22, 11, and 14, respectively. Multilevel sunburst
   maps at the species level showed that the dominant bacteria in the CTR
   group were Firmicutes and Bacteroidota. CTX treatment decreased the
   abundance of Bacteroidota and increased that of Firmicutes,
   Proteobacteria and Campilobacterota. Interestingly, PPD‐M treatment
   reversed the increase in Proteobacteria and Campilobacterota caused by
   CTX and restored the proportion of dominant bacteria to CTR group
   levels (Figure [89]3C). The community bar diagram showed that CTX and
   PPD had different influences on the relative abundance of microbial
   community composition, and the PPD group showed the advantage of having
   a closer evolutionary relationship with the CTR group (Figure [90]3D).
   These results indicate that PPD regulates the structure and composition
   of gut microbiota and reverses CTX‐induced gut microbiota dysbiosis.

FIGURE 3.

   FIGURE 3
   [91]Open in a new tab

   Effects of PPD on gut microbiota composition and structure in
   immunosuppressed mice. (A) PLS‐DA analysis (OTU level). (B) Venn
   diagram (similarity and difference in bacterial compositions in
   different experimental groups at the OTU level). (C) Community analysis
   sunburst plot at the species level (the species composition at the
   domain, kingdom, phylum, class, order, family, genus, and species
   levels is displayed from the inner circle to the outer circle). (D)
   Community bar plot analysis at the species level (the ordinate is the
   percentage of species in the sample, the color of the column represents
   different species, and the length of the column represents the size of
   the percentage of the species). (E) LEfSe (LDA effect size, LDA > 3)
   analysis of species with significant differences between CTR, CTX, and
   PPD‐M (the greater the LDA score is, the greater the impact of species
   abundance on the differential effect).

   To identify the key phylotypes and biomarkers of gut microbiota among
   the three groups, LEfSe (LDA > 3, p < 0.05) analysis was performed
   (Figure [92]3E). According to the analysis results, p_Bacteroidota,
   c_Bacteroidia, and o_Bacteroidales displayed high LDA scores, showing
   that they were abundant OTUs in the CTR group. In the CTX group,
   g_Escherichia‐Shigella, f_Enterobacteriaceae, and o_Enterobacterales
   exhibited higher scores, which indicated that they were significantly
   affected by CTX. In the PPD‐M group, p_Firmicutes, c_Bacilli,
   o_Lactobacillales, f_Lactobacillaceae, and g_Lactobacillus were
   significantly enriched, and o_Oscillospirales, g_Turicibacter, and
   g_Parabacteroides were also abundant. Thus, CTX leads to an increase in
   pathogenic bacteria, while PPD can enhance the abundance of beneficial
   bacteria, especially Lactobacillus. In general, these results indicate
   that PPD intervention ameliorates the imbalance of gut microbiota
   induced by CTX and promotes the proliferation of beneficial bacteria.

2.4. Gut microbiota depletion impairs HSPC self‐renewal and immune function
recovery

   To determine whether the gut microbiota plays a key role in PPD
   promoting bone marrow hematopoiesis and restoring immune functions,
   broad‐spectrum antibiotics (ABX) were used to deplete the microbiota in
   mice (Figure [93]4A). In this model, the fecal bacteria of mice
   significantly decreased compared with the normal group (Figure [94]S6).
   Interestingly, in ABX‐treated mice, PPD lost the ability to increase
   the number of WBCs and HSPCs (Figures [95]4B and C). Similarly, PPD
   could not reverse the decrease in CD3^+ T cells, IgA, IgG, and IL‐4
   caused by CTX (Figures [96]4D–F), and IFN‐γ did not return to the
   control level after PPD treatment (Figure [97]4F). These results
   indicate that PPD promotes bone marrow hematopoiesis and enhances
   immunity by regulating the gut microbiota, and the gut microbiota plays
   an indispensable role in the immune regulation of PPD.

FIGURE 4.

   FIGURE 4
   [98]Open in a new tab

   Correlation between gut microbiota and PPD enhancing immunity in
   immunosuppressed mice. (A) Antibiotic‐treated mouse protocol. In the
   first 7 days of the experiment and throughout the experiment,
   broad‐spectrum antibiotics were freely drunk by mice (CTR group was
   free to drink sterile water). The mice in the ABX+CTX+PPD group were
   intragastrically treated with PPD (50 mg/kg) for 16 days, and the CTR
   group, ABX group and ABX+CTX group mice were intragastrically
   administered PPD vehicle. Intraperitoneal injection of CTX was
   performed on day 14 (CTR group and ABX group mice with CTX vehicle).
   Blood was collected from the retro‐orbital venous plexus and mice were
   sacrificed on day 17. (B–D) Effects of PPD on WBCs (B), HSPCs (C),
   spleen CD3^+ T cells (D), IgG, IgA and IgM (E), IL‐4, TNF‐α, and IFN‐γ
   (F) in antibiotic‐treated mice (data are expressed as the mean ± SD.
   n = 4–6, *p < 0.05, **p < 0.01, ***p < 0.001, “ns” means no significant
   difference). (G) Spearman correlation heatmap at the genus level. (H)
   Spearman correlation heatmap at the species level. Squares in red
   (positive correlation) or blue (negative correlation) with a white
   asterisk (*) indicate significant correlations with p < 0.05 measured
   by the Spearman parametric correlation test.

   Furthermore, Spearman correlation analysis was performed to evaluate
   the potential relationship between immune parameters and the major
   microbiota. As shown in Figure [99]4G, many bacteria were positively
   correlated with immunoglobulins and negatively correlated with
   inflammatory factors. Specifically, Colidextribacter,
   f_Oscillospiraceae, and Oscillibacter were significantly positively
   correlated with IgG, IgM, and IL‐4 and negatively correlated with IFN‐γ
   and TNF‐α. In addition, f_Peptococcaceae, Blautia, and
   f_Lachnospiraceae were positively related to IL‐4 and negatively
   correlated with IFN‐γ and TNF‐α. Moreover, f_Lachnospiraceae,
   g_Colidextribacter, g_Dubosiella, and g_Alloprevotella were
   significantly positively correlated with HSPCs, WBCs, and CD3^+ T
   cells, while g_Escherichia‐Shigella was negatively correlated with WBCs
   and CD3^+ T cells (Figure [100]4H). These results further confirm that
   the gut microbiota plays an important role in PPD regulation of
   immunity.

2.5. PPD promotes the production of immune‐related fecal metabolites

   Intestinal metabolites are produced through the interaction between gut
   microbiota and host, which are essential for the maintenance of immune
   homeostasis.[101] ^25 We hypothesized that the potential mechanisms by
   which PPD improves immune functions are via the secondary metabolites
   generated by microbial metabolism in the gut. Then, untargeted
   metabolomic analysis of fecal samples was performed using liquid
   chromatography‒mass spectrometry (LC‒MS) in the CTR, CTX, and PPD‐M
   groups. The PLS‐DA model exhibited a significant separation of clusters
   among the three mouse groups in both cationic mode and anionic mode
   (Figures [102]S7A and B). In the PLS‐DA model verification (the number
   of random replacement tests was 200 times), the replacement tests
   passed, and there was no overfitting phenomenon in the model (Figures
   [103]S7C and D). A Venn diagram and volcano map showed that both CTX
   treatment and PPD‐M treatment resulted in dramatic alterations of
   metabolites, with a total of 101 metabolites (43 upregulated and 58
   downregulated) in the CTX versus. CTR group and 117 metabolites (78
   upregulated and 39 downregulated) in the PPD‐M versus CTX group
   (Figures [104]S7E–G). These results suggest that gut microbiota
   regulated by PPD alters the composition of fecal metabolites.

   A heatmap was used to visualize the importance and expression trend of
   the differential metabolites in the CTX versus CTR group and PPD‐M
   versus CTX group with variable importance in the projection (VIP) value
   of multivariate statistical analysis and P values of one‐dimensional
   statistics (Figures [105]5A and B). Of note, the metabolites of
   9R,10S‐epoxy‐stearic acid (ESA), ceramide (d18:1/9Z‐18:1), DG
   (18:1(11Z)/16:0/0:0), palmitoyl glucuronide (PG), and prostaglandin E2
   ethanolamide (PEE) were significantly decreased in CTX‐treated mice,
   while PPD treatment dramatically reversed the CTX‐induced decrease in
   these five metabolites (Figure [106]5C). In addition, PPD treatment
   also significantly enhanced the levels of Cucurbitacin C,
   l‐gulonolactone and 9′ ‐carboxy‐gamma‐chromanol (CGC) (Figure [107]5C).

FIGURE 5.

   FIGURE 5
   [108]Open in a new tab

   PPD induces changes in fecal metabolites. (A) and (B) Heat map of
   significantly altered metabolites in each sample. Metabolites with a
   variable importance in the projection (VIP) value of OPLS‐DA model > 1
   and p values of Student's t‐test < 0.05 were considered to be
   significantly different. (n = 6, *p < 0.05, **p < 0.01, ***p < 0.001).
   (C) Metabolites with significantly different changes in the PPD‐M group
   and the CTX group (data expressed as the mean ± SD. n = 6, *p < 0.05,
   **p < 0.01, ***p < 0.001, ****p < 0.0001, “ns” means no significant
   difference.)

   Spearman correlation analysis showed that the abovementioned eight
   differential metabolites (ESA, ceramide, DG, PG, PEE, cucurbitacin C,
   l‐gulonolactone, and CGC) were all positively correlated with IgG, IgA,
   IgM, and IL‐4 but negatively correlated with IFN‐γ and TNF‐α
   (Figure [109]6A). Similarly, HSPCs, CLPs, WBCs, and CD3^+ T cells were
   positively correlated with these 8 differential metabolites
   (Figure [110]6B). In addition, these metabolites were well correlated
   with many gut bacteria identified in the PPD‐M group (Figure [111]6C).
   Remarkably, cucurbitacin C was positively correlated with
   uncultured_bacterium_g_Lactobacillus, Lactobacillus_reuteri, and
   Lactobacillus_johnsonii, which was consistent with the fact that PPD
   treatment could significantly increase the abundance of Lactobacillus
   (Figure [112]3E), suggesting that cucurbitacin C may be a metabolite of
   Lactobacillus. Similarly, l‐gulonolactone was positively correlated
   with uncultured_bacterium_g_Colidextribacter,
   uncultured_bacterium_g_Alistipes, and
   uncultured_bacterium_g_Lachnospiraceae_NK4A136_group (Figure [113]6C).
   Figures [114]4G and H show that Colidextribacter was significantly
   positively correlated with IgG, IgM, IL‐4, WBCs and CD3^+ T cells and
   negatively correlated with IFN‐γ and TNF‐α, and
   g_Lachnospiraceae_NK4A136_group was negatively correlated with IFN‐γ,
   suggesting that l‐gulonolactone may be a metabolite of
   g_Colidextribacter and g_Lachnospiraceae_NK4A136_group. The
   abovementioned results indicate that PPD promotes the production of
   immune‐related metabolites by remodeling the gut microbiota.

FIGURE 6.

   FIGURE 6
   [115]Open in a new tab

   Functional analysis of differential metabolites. (A) Heat map of linear
   correlation analysis of differential metabolites with immunoglobulin
   and cytokines. (B) Heat map of linear correlation analysis of
   differential metabolites with immune indices (HSPCs, CLPs, WBCs, CD3^+
   T cells). (C) Heat map of linear correlation analysis between
   differential metabolites and gut microbiota. (D) KEGG Topology Analysis
   (CTX vs CTR); (E) KEGG Topology Analysis (PPD‐M vs. CTX). (Each bubble
   represents a KEGG pathway.)

   To further clarify the influence of PPD‐M on metabolic function in
   CTX‐induced immunosuppressed mice, the Kyoto Encyclopedia of Genes and
   Genomes (KEGG) pathway enrichment analysis was performed. The obtained
   results showed that two biological pathways (impact values>0.1, p<0.05)
   were significantly affected by CTX compared with the CTR group, and
   these changes were primarily linked to sphingolipid metabolism and
   alanine, aspartate and glutamate metabolism (Figure [116]6D). The
   comparison between the PPD‐M group and the CTX group showed that the
   sphingolipid metabolism pathway was also significantly affected by
   PPD‐M (impact values>0.1, p<0.05) (Figure [117]6E), indicating the
   importance of the sphingolipid metabolism pathway. According to the
   KEGG topology analysis, the main metabolite enriched in the
   sphingolipid metabolism pathway was ceramide (D18:1/9Z‐18:1).
   Figure [118]5 shows that CTX treatment significantly reduced the
   ceramide level, while PPD treatment perfectly restored the CTX‐induced
   ceramide reduction. These results illustrate that the sphingolipid
   metabolism pathway plays an important role in PPD‐M improvement of
   CTX‐induced immunosuppression.

3. DISCUSSION

   The gut microbiota is strongly implicated in host physiology and
   pathophysiology.[119] ^26 Gastrointestinal mucositis is one of the
   serious complications of chemotherapy and radiotherapy and affects
   approximately 50% of cancer patients.[120] ^27 CTX has the side effect
   of damaging the intestinal mucosal barrier and causing dysregulation of
   gut microbiota.[121] ^14 In this study, 16S rRNA gene sequencing was
   used to detect the gut microbiota of mice. LEfSe analysis showed that
   g_Escherichia‐Shigella was enriched in the CTX group, which has been
   identified as an opportunistic pathogen that can cause
   inflammation.[122] ^28 , [123]^29 After PPD treatment, Lactobacillus
   was significantly enriched, and o_Oscillospirales, g_Turicibacter, and
   g_Parabacteroides were also abundant (Figure [124]3E). Lactobacillus is
   defined as a probiotic that can promote digestion and absorption,
   improve the function of the human gastrointestinal tract, increase
   beneficial intestinal bacteria, restore the balance of intestinal
   bacteria, and enhance human immunity and resistance.[125] ^30 Butyrate
   is an important reference index for screening “the next generation of
   probiotics.” Oscillospirales can produce butyrate and other short‐chain
   fatty acids (SCFAs); thus, it is listed as a candidate for the next
   generation of probiotics.[126] ^31 Turicibacter was positively
   correlated with enhanced immune functions,[127] ^32 which was confirmed
   by the fact that the populations of Turicibacter in the gastrointestine
   were virtually nonexistent in both innate and adaptive immunodeficiency
   mice.[128] ^33 , [129]^34 Therefore, in our study, CTX treatment led to
   the proliferation of pathogenic bacteria, while PPD‐M treatment
   increased the abundance of beneficial bacteria.

   Broad‐spectrum ABX experiments showed that the gut microbiota plays an
   important role in PPD‐M enhancing immunity (Figures [130]4A–F).
   Spearman correlation analysis indicated that Colidextribacter,
   f_Oscillospiraceae, and Oscillibacter were significantly positively
   correlated with IgG, IgM and IL‐4 but negatively correlated with IFN‐γ
   and TNF‐α. Moreover, HSPC, WBC and CD3^+ T cells were significantly
   positively correlated with f_Lachnospiraceae, g_Colidextribacter,
   g_Dubosiella, and g_Alloprevotella but negatively correlated with
   g_Escherichia‐Shigella (Figures [131]4G and H). Recent studies have
   reported that Coldextribacter metabolizes to produce inosine, which can
   enhance the efficacy of checkpoint blockade immunotherapy and has
   anti‐inflammatory effects.[132] ^35 , [133]^36 Lachnospiraceae can
   protect mice against damage to the hematopoietic system and intestinal
   system caused by radiation and chemotherapy.[134] ^37 Dubosiella also
   regulates metabolism in vivo, improving intestinal immunity and
   promoting the body's resistance to inflammatory diseases.[135] ^38 ,
   [136]^39 Alloprevotella has been shown to benefit the production of
   SCFAs.[137] ^40

   How does the gut microbiota influence the immunomodulatory function of
   PPD‐M? Untargeted metabolomic analysis of mouse fecal contents was used
   to explain this mechanism. CTX treatment decreased the levels of
   9R,10S‐epoxy‐stearic acid, ceramide, DG, palmitoyl glucuronide, and
   prostaglandin E2 ethanolamide, while PPD‐M treatment dramatically
   reversed these decreases. In addition, PPD treatment also significantly
   enhanced the levels of cucurbitacin C, l‐gulonolactone and 9′
   ‐carboxy‐gamma‐chromanol (Figure [138]5C). 9R, 10S‐epoxy‐stearic acid
   has anti‐inflammatory and preventive effects, as well as anticancer
   activity.[139] ^41 Palmitoyl glucuronide is negatively correlated with
   ulcerative colitis.[140] ^42 Prostaglandin E2 ethanolamide is a
   prostaglandin derivative that can reduce cytokine‐induced intestinal
   epithelial injury.[141] ^43 l‐gulonolactone does not enhance immunity
   by itself, but it is an essential component of l‐gulono‐lactone oxidase
   (GLO), a critical enzyme in the vitamin C synthesis pathway, promoting
   the synthesis of vitamin C. Vitamin C has a variety of beneficial
   effects on the cellular functions of both the innate and adaptive
   immune systems.[142] ^44 Cucurbitacin C itself has immune‐enhancing
   effects, and a recent study found that cucurbitacin C also has a
   variety of antitumor effects.[143] ^45 In this study, Spearman
   correlation analysis showed that these 8 differential metabolites were
   positively correlated with IgG, IgA, IgM, IL‐4, HSPCs, CLPs, WBCs, and
   CD3^+ T cells and negatively correlated with IFN‐γ and TNF‐α
   (Figures [144]6A and B), illustrating that these metabolites have
   immune‐enhancing and/or anti‐inflammatory effects. More importantly,
   these differential metabolites are closely related to the gut
   microbiota (Figure [145]6C). In particular, cucurbitacin C was
   positively correlated with Lactobacillus_reuteri and
   Lactobacillus_johnsonii, suggesting that cucurbitacin C may be a
   metabolite of Lactobacillus. l‐gulonolactone was positively correlated
   with g_Colidextribacter, g_Alistipes, and
   g_Lachnospiraceae_NK4A136_group, suggesting that l‐gulonolactone may be
   their metabolite. Therefore, we conclude that PPD‐M exerts its
   immunomodulatory function by promoting beneficial bacteria to produce
   immunity‐enhancing metabolites.

   In addition, KEGG topology analysis showed that the sphingolipid
   metabolic pathway was significantly enriched in both the CTX versus CTR
   and PPD‐M versus CTX groups, and ceramide was the main metabolite of
   this pathway (Figures [146]6D and E). In CTX‐induced immunosuppressed
   mice, ceramide levels were significantly reduced, while PPD treatment
   perfectly restored this reduction (Figure [147]5). The sphingolipid
   metabolic pathway has become an important target for the treatment of
   many diseases. Currently, many bioactive sphingolipid analogs have been
   proven to be used for the treatment of immunity, metabolic disorders,
   and cancer, such as FTY720.[148] ^46 Ceramide is a central hub of the
   sphingolipid metabolic pathway. Ceramide accumulation is considered to
   be a key step in mediating and regulating various immune cell functions
   such as regulation of the immune cell response to bacteria, viruses,
   and other foreign pathogens, production of cytokines, optimization of
   hematopoietic stem cell mobilization and homing, formation of NETs in
   neutrophils, and mediation of T cell responses during
   inflammation.[149] ^47 , [150]^48 In this study, ceramide was
   positively correlated with CLP and IgA but negatively correlated with
   TNF‐α (Figures [151]6A and B), suggesting that ceramide was involved in
   the processes of promoting bone marrow hematopoiesis, enhancing
   immunity and inhibiting inflammation. In industry, microbial
   fermentation technology is a commonly used production method of
   ceramide,[152] ^49 and a previous study has shown that gut‐derived
   sphingolipids produced by gut Bacteroides affect ceramide levels,[153]
   ^50 suggesting that ceramide production is closely related to gut
   bacteria. In summary, these results demonstrate that the sphingolipid
   metabolic pathway may be a potential target of PPD for enhancing
   immunity in immunosuppressed mice.

   In conclusions, this work provides the first evidence that PPD promotes
   bone marrow hematopoiesis and improves immunity by regulating gut
   microbiota in an immunosuppressed mouse model. PPD‐M effectively
   ameliorated CTX‐induced immunosuppression in mice by promoting bone
   marrow hematopoiesis, increasing the number of peripheral blood WBCs,
   protecting peripheral lymphoid tissue and increasing the levels of
   immunoglobulin IgG and IgM in blood. More importantly, PPD‐M protected
   against CTX‐induced gut microbiota dysbiosis by increasing the
   abundance of beneficial bacteria and inhibiting the proliferation of
   pathogenic bacteria. When the gut microbiota was depleted, PPD‐M lost
   its ability to promote bone marrow hematopoiesis and enhance immunity.
   In addition, PPD‐M promoted the production of microbiota‐derived
   anti‐inflammatory and immune‐enhancing metabolites and the enrichment
   of the sphingolipid metabolic pathway, which contributed to the
   immune‐enhancing function of PPD‐M in immunosuppressed mice
   (Figure [154]7). Our study revealed a new mechanism by which PPD
   enhances immunity and provided a scientific basis for PPD as an
   immunomodulator. PPD has potential clinical applications in reducing
   the side effects of cancer chemotherapy.

FIGURE 7.

   FIGURE 7
   [155]Open in a new tab

   A schematic model showing the immunomodulatory mechanism of PPD in
   CTX‐induced immunosuppressed mice. CTX treatment leads to
   immunosuppression and gut microbiota dysbiosis in mice. PPD treatment
   can protect against CTX‐induced gut microbiota dysbiosis by increasing
   the abundance of beneficial bacteria and inhibiting the proliferation
   of pathogenic bacteria. The beneficial bacteria metabolize to produce
   immune‐enhancing metabolites and anti‐inflammatory metabolites, which
   contribute to the immune‐enhancing and anti‐inflammatory functions of
   PPD‐M in CTX‐induced immunosuppressed mice, including promoting bone
   marrow hematopoiesis to increase the number of peripheral blood white
   blood cells, proliferating splenic lymphocytes and eliminating
   inflammation. (The schematic was drawn with the help of Biorender.)

4. MATERIALS AND METHODS

4.1. Animal and experimental design

   ICR mice (5‐6 weeks old, 22 ± 2 g, male) used in this study were
   obtained from the Pengyue Experimental Animal Center (No. SCXK20190003,
   Jinan, China). For immunosuppressed mouse model preparation, 56 mice
   were randomly divided into two groups: the control (CTR) group and the
   CTX group. Mice in the CTX group were intraperitoneally injected with
   CTX (50 mg/kg) solution, and mice in the CTR group were
   intraperitoneally injected with the same volume of physiological
   saline. On the day before treatment with CTX and on the first, third,
   and sixth days after treatment with CTX, seven mice in each group were
   randomly removed for experimental testing at each time point.

   To explore the immune improvement of PPD, 30 mice were randomly divided
   into five groups: the control group (CTR), CTX group, CTX+PPD low‐dose
   group (PPD‐L), CTX+PPD medium‐dose group (PPD‐M), and CTX+PPD high‐dose
   group (PPD‐H). PPD (HPLC ≥98%) was dissolved in vehicle containing 5%
   anhydrous ethanol, 0.5% Tween‐80 and 94.5% deionized water. The mice in
   the PPD groups were administered PPD by oral gavage at doses of
   25 mg/kg/day (PPD‐L), 50 mg/kg/day (PPD‐M), and 100 mg/kg/day (PPD‐H)
   for 16 days. The mice in the CTR and CTX groups were given PPD vehicle
   for 16 days. On the 14th day, the mice in the CTX, PPD‐L, PPD‐M, and
   PPD‐H groups were given CTX (50 mg/kg) intraperitoneally, and the CTR
   group was administered an equivalent dose of saline. On the 17th day,
   the mice were sacrificed for fecal, blood, and organ samples for
   further testing.

4.2. Depletion of gut microbiota

   Mice freely drank water with broad‐spectrum ABX, containing ampicillin
   (1 mg/ml), kanamycin (1 mg/ml), streptomycin (1 mg/ml), and vancomycin
   (0.5 mg/ml) (Meilunbio, China). The control group mice were also
   maintained with sterile water and sterile food during the experiment.
   Rectal feces were collected at 1 week after antibiotic administration
   and at the end of the experiment and then cultured in LB liquid medium
   or LB solid medium for 12 h. The absorbance and photographs were
   recorded.

4.3. Cytokine and immunoglobulin assays

   Serum was collected from blood samples by centrifugation at 4000 rpm
   and 4°C for 15 min. The contents of TNF‐α, IFN‐γ, IL‐4, IgG, IgM, and
   IgA were measured by ELISA kits (Shanghai MLBIO Biotechnology, China).

4.4. Flow cytometry assay

   The spleen was ground and filtered through a 70 μm nylon mesh to obtain
   a single‐cell suspension. Lymphocytes were isolated using a Mouse
   Spleen Lymphocyte Isolation Kit (Solarbio, China). The lymphocytes were
   washed twice with phosphate‐buffered saline (PBS) and resuspended.
   Then, antibodies against CD3‐PE, CD4‐APC, and CD8a‐FITC (BioLegend,
   USA) were added to the cell suspension in sequence. After gentle
   mixing, the mixture was incubated for 30 min in a dark room at 4°C. The
   percentages of CD3^+ T cells, CD4^+ T cells and CD8a^+ T cell subsets
   were analyzed by flow cytometry (BD Canto II, USA).

   Bone marrow was collected by flushing femurs with RPMI 1640 medium. Red
   blood cells were removed from single‐cell suspensions with RBC lysis
   buffer (Solarbio, China). One milliliter of cell suspension was added
   to Fixable Dye eFluor 780 (Thermo Fisher Scientific, USA) to isolate
   dead cells. After gentle swirling, the cells were incubated for 30 min
   in the dark at 4°C. After washing with PBS, the cells were resuspended.
   The treated cells were analyzed for hematopoietic stem/progenitor cells
   (HSPCs). In short, lineage marker‐negative cells (Lin^−) were first
   separated using Hematopoietic Lineage Antibody Cocktail FITC (Thermo
   Fisher Scientific, USA) (containing antibodies against CD3, CD45R,
   CD11b, Gr‐1, and Ter‐119). Lin^− cells were then further stained with
   Sca‐1‐PerCP/Cyanine5.5, c‐Kit‐APC, IL‐7Rα‐PE/Cyanine7, CD34‐Brilliant
   Violet 421™, and CD16/32‐PE (BioLegend, USA). Populations of KLS^+
   cells (Lin^− c‐Kit^+ Sca‐1^+, HSPCs), KLS^− cells (Lin^− c‐Kit^+
   Sca‐1^−), CLPs (KLS^+ IL‐7Rα^+), CMPs (KLS^− CD34^+ IL‐7Rα^−
   CD16/32^−), GMPs (KLS^− CD34^+ IL‐7Rα^− CD16/32^+), and MEPs (KLS^−
   IL‐7Rα^− CD34^− CD16/32^−) were analyzed by flow cytometry. Data were
   analyzed with FlowJo software (Tree Star, USA), and the loop‐gate
   strategy is shown in Supplementary information Figure [156]S2.

4.5. Histological analysis

   The thymus, spleen, and thighbones were prepared for histological
   analysis according to previous methods.[157] ^12 Briefly, the spleens
   and thymus were fixed in 4% paraformaldehyde, embedded in paraffin, and
   sliced into 5‐μm‐thick sections. Thighbones also need to be decalcified
   after fixation with 4% paraformaldehyde. To observe the changes within
   the organization, the tissues were stained with hematoxylin and eosin
   (HE) after deparaffinization. Images were obtained with a light
   microscope (OLYMPUS BX53, Japan).

4.6. Gut microbiota analysis

   Fecal samples from mice in the CTR group, CTX group and PPD‐M group
   were collected aseptically, quickly frozen with liquid nitrogen, and
   then stored in the refrigerator at −80°C. DNA was extracted for gut
   microbiota diversity analysis. Fecal sample DNA extraction, PCR
   amplification and sequencing, and processing of sequencing data were
   performed at Majorbio Biotech (Shanghai, China). The data were analyzed
   using the Majorbio Cloud Platform ([158]www.majorbio.com).

4.7. Fecal metabolite assessment

   Fecal samples of mice in the CTR group, CTX group, and PPD‐M group were
   frozen at −80°C prior to LC‒MS analysis. Metabolite sample preparation
   and LC‒MS analysis parameters were conducted by Majorbio Biotech
   (Shanghai, China). Detailed steps are available in the Supplementary
   information. The data were analyzed using the Majorbio Cloud Platform
   ([159]www.majorbio.com).

4.8. Statistical analysis

   For two‐group comparisons, an unpaired two‐tailed Student's t‐test was
   applied. For significantly changing bacterial lists, we used the online
   Majorbio Cloud Platform to perform the Wilcoxon rank‐sum test, and the
   p value was based on a two‐tailed FDR‐corrected test. The significance
   level was set to 0.05, and the 0.95 confidence intervals were
   calculated through the bootstrap algorithm. For comparisons of more
   than two groups, one‐way ANOVA or two‐way ANOVA followed by Dunnett's
   test was performed. All data with error bars are represented as the
   mean ± SD, and p < 0.05 was considered statistically significant. Most
   of the data were analyzed in GraphPad Prism (Graph Pad Software Inc.,
   USA).

AUTHOR CONTRIBUTION

   Y. C. and X. P. designed the study and wrote the article. Y. C., B. L.,
   and W. L. performed the experiments and analyzed the data. L. Y., H.
   L., and C. L. assisted with the animal work. F. G. and X. G. provided
   useful suggestions. Z. S. helped with the data analysis. J. L.
   interpreted the results and revised the manuscript. All authors have
   read and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT

   The authors declare no conflicts of interest.

ETHICS STATEMENT

   All animal experiments were approved by the Laboratory Animal Ethics
   Committee of Binzhou Medical University and complied with the national
   and international guidelines for the Care and Use of Laboratory
   Animals.

Supporting information

   Supporting Information
   [160]Click here for additional data file.^ (1.2MB, pdf)

ACKNOWLEDGMENTS