If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence Address correspondence to: Jonathan T. Busada, PhD, Department of Microbiology, Immunology and Cell Biology, West Virginia University School of Medicine, 64 Medical Center Drive, PO Box 9177, Morgantown, West Virginia 26506.
Molecular Endocrinology Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North CarolinaDepartment of Microbiology, Immunology and Cell Biology, West Virginia University School of Medicine, Morgantown, West Virginia
Molecular Endocrinology Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina
Post-Transcriptional Gene Expression Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina
Integrative Bioinformatics Support Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina
Molecular Endocrinology Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina
Molecular Endocrinology Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina
Post-Transcriptional Gene Expression Group, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina
Aberrant immune activation is associated with numerous inflammatory and autoimmune diseases and contributes to cancer development and progression. Within the stomach, inflammation drives a well-established sequence from gastritis to metaplasia, eventually resulting in adenocarcinoma. Unfortunately, the processes that regulate gastric inflammation and prevent carcinogenesis remain unknown. Tristetraprolin (TTP) is an RNA-binding protein that promotes the turnover of numerous proinflammatory and oncogenic messenger RNAs. Here, we assess the role of TTP in regulating gastric inflammation and spasmolytic polypeptide-expressing metaplasia (SPEM) development.
Methods
We used a TTP-overexpressing model, the TTPΔadenylate-uridylate rich element mouse, to examine whether TTP can protect the stomach from adrenalectomy (ADX)-induced gastric inflammation and SPEM.
Results
We found that TTPΔadenylate-uridylate rich element mice were completely protected from ADX-induced gastric inflammation and SPEM. RNA sequencing 5 days after ADX showed that TTP overexpression suppressed the expression of genes associated with the innate immune response. Importantly, TTP overexpression did not protect from high-dose-tamoxifen–induced SPEM development, suggesting that protection in the ADX model is achieved primarily by suppressing inflammation. Finally, we show that protection from gastric inflammation was only partially due to the suppression of Tnf, a well-known TTP target.
Conclusions
Our results show that TTP exerts broad anti-inflammatory effects in the stomach and suggest that therapies that increase TTP expression may be effective treatments of proneoplastic gastric inflammation. Transcript profiling: GSE164349.
Aberrant gastric inflammation damages the stomach and induces gastric metaplasia (spasmolytic polypeptide-expressing metaplasia). Increased expression of the RNA binding protein tristetraprolin suppresses adrenalectomy-induced gastric inflammation and spasmolytic polypeptide-expressing metaplasia development.
Gastric adenocarcinoma is the third leading cause of cancer deaths worldwide.
Chronic inflammation is strongly correlated with gastric cancer development and typically is initiated by Helicobacter pylori infection or autoimmune gastritis.
Chronic inflammation causes atrophic gastritis and loss of the acid-secreting parietal cells (oxyntic atrophy), leading to the development of spasmolytic polypeptide-expressing metaplasia (SPEM).
Excessive expression of proinflammatory cytokines such as interferon gamma, tumor necrosis factor alpha (TNF), and interleukin (IL)1B induce SPEM and dysplasia.
Hyperplastic gastric tumors with spasmolytic polypeptide-expressing metaplasia caused by tumor necrosis factor-alpha-dependent inflammation in cyclooxygenase-2/microsomal prostaglandin E synthase-1 transgenic mice.
Wild-type and interleukin-10-deficient regulatory T cells reduce effector T-cell-mediated gastroduodenitis in Rag2-/- mice, but only wild-type regulatory T cells suppress Helicobacter pylori gastritis.
Loss of these hormones leads to spontaneous activation of the innate immune response, driving SPEM development. However, the mechanisms that regulate gastric inflammation remain poorly defined.
Tristetraprolin (TTP) is a member of a small family of RNA binding proteins and is encoded by the gene Zfp36.
Proteins of the TTP family are characterized by highly conserved tandem zinc finger domains that bind to AU rich-elements (AREs) in the 3’ untranslated region (UTR) of target messenger RNAs (mRNAs).
The ideal binding sequence, UUAUUUAUU or its variants, recognized by TTP, is found in a host of transcripts that encode proinflammatory cytokines, chemokines, and oncogenes.
Germline Zfp36 knockout (KO) mice show numerous systemic inflammatory pathologies such as dermatitis, arthritis, autoimmunity, and myeloid hyperplasia, all of which are linked to aberrant expression of the proinflammatory cytokine TNF.
However, TTP expression usually is transient, in part owing to binding sites within the TTP transcript that allow the TTP protein to negatively regulate its own expression.
Mice with a germ-line deletion of a 136-base ARE within the TTP mRNA 3’ UTR (TTPΔARE) have enhanced TTP mRNA stability and moderately increased TTP protein expression in their tissues but are phenotypically normal during normal vivarium conditions.
However, TTPΔARE mice are resistant to experimental models of imiquimod-induced dermatitis, collagen antibody–induced arthritis, experimental autoimmune encephalomyelitis, bacterial gingivitis and dental bone erosion, inflammatory lung damage, experimental autoimmune uveitis, and chemically induced skin carcinogenesis.
We previously showed that glucocorticoids are master regulators of gastric inflammation. Systemic removal of endogenous glucocorticoids by ADX triggers massive, spontaneous gastric inflammation and SPEM.
Adrenalectomy is a useful model to study factors that participate in gastric inflammation and SPEM development. In this study, we used the ADX model to investigate the effects of enhanced TTP expression on gastric inflammation and metaplasia. We found that increased, regulated, whole-body TTP expression completely blocked the development of gastric inflammation and metaplasia after bilateral ADX. Surprisingly, protection from ADX-induced gastric inflammation was not recapitulated in Tnf KO mice, suggesting that TTP regulation of gastric inflammation is more complex than suppressing a single proinflammatory cytokine. Our results suggest that treatments that increase TTP protein expression may effectively treat gastric inflammation and potentially protect against neoplasia development.
Gastric inflammation is associated with the development of gastritis, oxyntic atrophy, and metaplasia. TTP enhances the turnover of numerous proinflammatory mRNAs such as those encoding TNF.
We hypothesized that enhanced systemic TTP expression could protect mice from gastric inflammation and metaplasia. To test this hypothesis, we used TTPΔARE mice in which a 136-base AU-rich instability region was deleted from the 3’ UTR of the gene encoding TTP, Zfp36.
we confirmed that germline deletion of the ARE region results in the accumulation of TTP mRNA in the mouse gastric fundus at 2 months of age (Figure 1A). We previously showed that adrenalectomy (ADX) rapidly induces spontaneous gastric inflammation and SPEM.
We used bilateral ADX to assess gastric inflammation and SPEM development in TTPΔARE mice (Figure 1B). As expected, wild-type (WT) control mice showed prominent inflammation within the gastric corpus 2 months after ADX (Figure 1C). In contrast, both TTPΔARE heterozygous and homozygous mice were protected from increased inflammation. We previously have shown that ADX-induced gastric inflammation is composed predominately of macrophages and eosinophils.
Analysis of the WT mice showed 4.7-fold and 28-fold increases in gastric macrophages and eosinophils 2 months after ADX, respectively (Figure 1D). In contrast, neither TTP heterozygous mice nor homozygous mice showed a significant increase in inflammatory cells. These data indicate that increased systemic TTP expression from normally regulated Zfp36 can protect the stomach from ADX-induced chronic inflammation.
Figure 1Increased levels of TTP prevent ADX-induced gastric inflammation. (A) qRT-PCR for TTP mRNA within the gastric corpus. (B) Experimental model. (C) Representative immunostaining of the gastric corpus from WT, TTPΔARE heterozygous, and TTPΔARE homozygous mice 2 months after sham surgery or ADX. Gastric sections were stained with CD45 antibodies (green) and nuclei were stained with 4′,6-diamidino-2-phenylindole. Scale bars: 100 μm. n ≥ 7 mice/group. (D) Quantitation of macrophages (CD68 and CD45 double-positive) and eosinophils (Siglec F and CD45 double-positive). n ≥ 4 mice/group. (A and D) Data are means ± SD. P values were determined by (A) an unpaired Student t test or (D) 1-way analysis of variance with a post hoc Tukey t test. ∗∗∗P ≤ .001 and ∗∗∗∗P ≤ .0001.
Because TTPΔARE mice were resistant to ADX-induced inflammation, we asked whether increased TTP expression could prevent the development of oxyntic atrophy and metaplasia. The gross morphology of sham-operated TTPΔARE heterozygous and homozygous mice was indistinguishable from sham-operated WT mice (Figure 2A), and there were no significant differences in the number of parietal cells or chief cells (Figure 2B). Two months after ADX, WT mice had lost 82% of their parietal cell population and 99% of their mature chief cells (Figure 2). Moreover, WT mice showed prominent mucous cell hyperplasia within the gastric corpus, identified by an increase in Griffonia simplicifolia (GSII) lectin staining, which binds to mucin 6. In contrast to ADX-WT mice, neither TTPΔARE heterozygous nor homozygous mice showed a significant change in their parietal and chief cell populations, and both genotypes had normal gastric morphology 2 months after ADX (Figure 2).
Figure 2Increased levels of TTP protect the stomach from ADX-induced pathogenic inflammation. (A) Representative immunostaining of the gastric corpus from WT, TTPΔARE heterozygous, and TTPΔARE homozygous mice killed 2 months after sham surgery or ADX. Gastric sections were probed for ATP4B (parietal cells, red), MIST1 (chief cells, green), and GSII lectin (mucous neck cells, grey). Nuclei were stained with 4′,6-diamidino-2-phenylindole. Scale bars: 100 μm. (B) Quantitation of the number of parietal cells and chief cells observed per 20× field (n ≥ 6 mice/group). Data are means ± SD, P values were determined by 1-way analysis of variance with a post hoc Tukey t test. ∗∗∗∗P ≤ .0001.
Oxyntic atrophy, loss of the mature chief cell marker BHLHA15 (also known as MIST1), and expansion of GSII+ cells are among the defining characteristics of SPEM. We confirmed SPEM development by immunostaining for the de novo SPEM marker CD44v9, a splice variant of CD44.
Although there was widespread staining of CD44v9 in ADX WT mice, CD44v9 was not detected within the gastric glands of ADX TTPΔARE mice (Figure 3A). Re-entry into the cell cycle accompanies chief cell transdifferentiation.
We performed co-immunofluorescence for Ki67 and β-catenin (CTNNB1) to identify proliferative epithelial cells. In sham mice, proliferation was restricted to the gland isthmus, which is widely regarded as the stem cell compartment within the gastric corpus (Figure 3B). In contrast, 2 months after ADX, WT mice showed numerous Ki67+ cells throughout the neck and base. However, proliferation remained unchanged 2 months after ADX in TTPΔARE heterozygous and homozygous mice. In addition, we performed quantitative reverse-transcription polymerase chain reaction (qRT-PCR) on a panel of transcripts from the advanced SPEM-associated genes Cftr, Wfdc2, and Olfm4.
Consistent with the increase in CD44v9 staining, there was significant induction of all 3 SPEM markers in ADX WT mice (Figure 3C). However, these transcripts did not significantly increase in TTPΔARE homozygous mice. These results show that increased TTP expression protected the mice from oxyntic atrophy and SPEM development.
Figure 3Increased levels of TTP prevent SPEM development. (A and B) Representative immunostaining of the gastric corpus from WT, TTPΔARE heterozygous, and TTPΔARE homozygous mice 2 months after sham surgery or adrenalectomy. Gastric sections were probed for (A) the SPEM marker CD44v9 (green) and the lectin GSII (mucous neck cells, grey) or (B) with Ki67 (green) and CTNNB1 (red, epithelial cells). Nuclei were stained with 4′,6-diamidino-2-phenylindole. Scale bars: 100 μm. n ≥ 6 mice/group. (C) qRT-PCR of the indicated SPEM marker genes using RNA isolated from the gastric corpus (n ≥ 4 mice/group). Data are means ± SD. P values were determined by 1-way analysis of variance with a post hoc Tukey t test. ∗∗∗P ≤ .001 and ∗∗∗∗P ≤ .001.
TTP Suppresses the Induction of Proinflammatory Gene mRNAs After ADX
Because TTPΔARE mice were protected from ADX-induced gastric inflammation and SPEM, we next used RNA sequencing (RNAseq) to examine their gastric transcriptomes 5 days after ADX (Figure 1B). We used this early time after ADX to avoid secondary changes caused by the anatomic alterations seen in long-term ADX mice. Moreover, there was limited gastric inflammation 5 days after ADX, as shown by a modest increase in the pan-immune cell marker Ptprc (CD45) and the pan macrophage marker Cd68 (Figure 4A). RNAseq showed significant increases in inflammatory gene expression 5 days after ADX in WT mice. Gene set enrichment analysis (GSEA) comparing sham WT vs ADX WT groups showed significant enrichment of mRNAs associated with the Gene Ontology (GO) inflammatory response pathway (Figure 4B). Surprisingly, there also was significant enrichment of inflammatory genes 5 days after ADX in TTPΔARE homozygous mice. However, the normalized enrichment score was 6.36 in the WT group compared with 5.02 in the TTPΔARE group, suggesting moderately increased inflammation within the WT group. Moreover, a comparison of the ADX WT group with the ADX TTPΔARE group showed greater activation of inflammatory response pathways in ADX WT mice (Figure 4B). Next, we ranked the GSEA data and found that the GO innate immune response pathway was the seventh highest activated pathway in the WT group (normalized enrichment score, 5.32) (Figure 4C). In contrast, this pathway was ranked 46th in the TTPΔARE group (normalized enrichment score, 3.97). Comparison of the WT ADX group with the TTPΔARE ADX group showed significant positive enrichment (Figure 4C), suggesting increased innate immune system activation in WT ADX mice.
Figure 4Increased TTP expression elicits broad suppression of inflammatory genes within the stomach of adrenalectomized mice. Shown are RNAseq data from total cellular RNA isolated from the gastric corpus of WT and TTPΔARE homozygous mice killed 5 days after sham surgery or ADX. (A) FPKM values for the indicated genes. P values were determined by 1-way analysis of variance with a post hoc Tukey t test. ∗P ≤ .05 and ∗∗∗P ≤ .001. (B, C, E, and F) GSEA of the significantly regulated genes comparing adrenalectomized WT mice with adrenalectomized TTP mice. (D) IPA of the total number of significantly regulated genes in sham vs adrenalectomized WT mice and TTP mice, respectively. n = 4 mice/group. NES, normalized enrichment score.
Therefore, we next analyzed the differentially expressed gene (DEG) lists using Ingenuity Pathway Analysis (IPA) to assess transcripts associated with macrophage activation. IPA predicted significant activation of the "Activation of Macrophages" pathway in ADX WT mice (activation z-score, 2.43) (Figure 4D). However, this pathway was not significantly activated in ADX TTPΔARE mice. Importantly, GSEA showed that pathways associated with adaptive immunity, such as the GO adaptive immune response (Figure 4E) and gene GO lymphocyte activation (Figure 4F), were activated equivalently in both WT and TTPΔARE mice. These results are consistent with published reports that mature lymphocytes are dispensable for inducing SPEM development.
Transcripts Containing AREs Are Only a Small Portion of the ADX-Induced Genes
TTP is an RNA binding protein that binds to adenylate-uridylate–rich target sequences in mRNAs before promoting the turnover of those mRNAs. RNAseq showed 760 DEGs between the sham WT and ADX WT groups. In contrast, there were only 490 DEGs between the sham TTPΔARE mice and ADX TTPΔARE groups (Figure 5A). Of the DEGs, 189 genes were regulated in both groups. We screened the transcripts that were up-regulated by ADX in the WT group for the presence of ideal TTP binding sequences (UAUUUAU and UAUUUUAU). We identified 94 mRNAs that contained a potential TTP binding motif (Figure 5B). Up-regulation of 93 of these transcripts was blunted significantly in ADX TTPΔARE mice, indicating that TTP may enhance the degradation of these transcripts. Importantly, there were established TTP targets among the 94 ARE-containing transcripts, such as the mRNA encoding Tnf, and inflammatory genes associated with SPEM development, including Il13. Il13 is potently induced by the alarmin IL33.
Interestingly, we found that Il33 expression was increased significantly only in TTPΔARE mice 5 days after ADX (Figure 5C). Consistent with this increase, we did not identify an ARE within the Il33 transcript, suggesting it may not be a direct TTP target. In contrast, Il13, which does contain potential TTP binding sites, was blunted significantly in ADX TTPΔARE mice; thus, TTP suppression of Il13 may disrupt macrophage activation. Together, these data show that TTP directly regulates numerous proinflammatory genes within the stomach.
Figure 5ARE-containing mRNAs are a small proportion of the DEGs in the stomach. (A) Venn diagram of the total number of DEGs in the gastric corpus 5 days after ADX. (B) Heatmap visualizing ARE-containing mRNAs in the indicated data sets. Transcripts were up-regulated significantly in ADX WT mice compared with sham WT mice. Scale is z-score minimum/maximum within the data set. (C) qRT-PCR of RNA isolated from the gastric corpus 5 days after sham or ADX surgery. n ≥ 6 mice/group. Data are means ± SD. P values were determined by 1-way analysis of variance with a post hoc Tukey t test. ∗∗P ≤ .01 and ∗∗∗∗P ≤ .0001.
Tnf Knockout Mice Are Partially Protected From ADX-Induced SPEM
TNF-α is a prominent proinflammatory cytokine produced by macrophages and other leukocytes. Aberrant TNF production is associated with inflammatory disease within the gastrointestinal tract, and may increase the risk of gastric cancer.
Hyperplastic gastric tumors with spasmolytic polypeptide-expressing metaplasia caused by tumor necrosis factor-alpha-dependent inflammation in cyclooxygenase-2/microsomal prostaglandin E synthase-1 transgenic mice.
We hypothesized that suppression of Tnf in TTPΔARE mice may protect against ADX-induced inflammation and metaplasia. Therefore, we adrenalectomized Tnf KO mice and assessed their stomachs 2 months after surgery. Interestingly, Tnf KO mice showed only intermediate protection from SPEM (Figure 6A and B). In ADX Tnf KO mice, there were regions of the gastric corpus that appeared identical to sham controls, with the normal complement of parietal and chief cells, and that were negative for the SPEM marker CD44v9 (Figure 6A). In contrast, other regions of the lesser curvature appeared identical to sections from the ADX WT mice (Figure 6A, far right panel). We quantitated the number of parietal and chief cells present in both normal and SPEM regions. Quantitation showed that although ADX Tnf KO mice showed a significant loss of parietal and chief cells relative to sham controls, these effects were diminished significantly compared with ADX WT mice (Figure 6B). In addition to stomach inflammation, ADX WT mice developed splenomegaly (Figure 6C and D), a classic feature of ADX in rodents.
However, TTPΔARE homozygous spleen weights did not differ significantly from WT mice 2 months after ADX. Surprisingly, Tnf KO completely rescued the splenomegaly observed in ADX WT mice. Together, these data indicate that although Tnf contributes to SPEM development, there likely are redundant mechanisms that control pathogenic gastric inflammation. Moreover, these results show that TTP’s protective effects in the stomach are the result of broad anti-inflammatory effects beyond the suppression of Tnf.
Figure 6Tnf KO mice are partially protected from ADX-induced gastric inflammation (A) Representative images of stomachs taken from WT and Tnf KO mice killed 2 months after sham surgery or ADX.Scale bars: 100 μm. (B) Quantitation of the number of parietal cells and chief cells observed per 20× field (n ≥ 4 mice/group). (C) Representative images of spleens from mice killed 2 months after sham surgery or ADX. (D) Ratio of spleen weight normalized to total body weight. Data are means ± SD. P values were determined by 1-way analysis of variance with a post hoc Tukey t test. ∗P ≤ .05 and ∗∗∗∗P ≤ .0001.
TTP Does Not Prevent High-Dose-Tamoxifen–Induced SPEM Development
SPEM development occurs in response to glandular damage within the gastric corpus. Adrenalectomy induces SPEM development by triggering massive gastric inflammation.
Our results show that TTP overexpression suppresses ADX-induced gastric inflammation. We hypothesized that TTP protected from SPEM by regulating gastric inflammation. To test this hypothesis, we used the high-dose tamoxifen (HDT) model. HDT induces chief cell transdifferentiation toward the SPEM lineage by killing parietal cells and is largely noninflammatory.
WT and TTPΔARE homozygous mice were treated with HDT 3 times over 72 hours, and stomachs were collected 24 hours after the final dose. There were no morphologic differences between the stomachs of vehicle-treated WT and TTPΔARE mice (Figure 7A). As expected, HDT treatment induced nearly complete oxyntic atrophy in both genotypes. Importantly, loss of mature chief cells, denoted by loss of MIST1 staining (Figure 7A) and Gif mRNAs (Figure 7B) was equivalent in both HDT-treated WT and TTPΔARE mice. Moreover, there was concurrent induction of the SPEM markers CD44v9 as well as Cftr mRNAs. These results show that TTP overexpression does not directly inhibit SPEM development and suggests that SPEM protection occurs through inhibition of the intensity and type of inflammation.
Figure 7TTP overexpression does not affect high-dose-tamoxifen–induced SPEM development. (A) Representative immunostaining of the gastric corpus from WT and TTPΔARE homozygous mice treated with tamoxifen for 3 consecutive days. Gastric sections were probed for ATP4B (parietal cells, red), MIST1 (chief cells, green), and GSII lectin (mucous neck cells, grey); or with CD44v9 (SPEM, green) and GSII. Nuclei were stained with 4′,6-diamidino-2-phenylindole. Scale bars: 100 μm. n ≥ 5 mice/group. (B) Quantitation of the number of parietal cells and chief cells observed per 20× field (n ≥ 6 mice/group). Data are means ± SD. P values were determined by 1-way analysis of variance with a post hoc Tukey t test. ∗∗P ≤ .01, ∗∗∗P ≤ .001, and ∗∗∗∗P ≤ .0001. Tam, tamoxifen; Veh, vehicle.
Post-transcriptional regulation of gene expression by RNA binding proteins is critical for maintaining cellular and tissue homeostasis. Dysregulation of RNA binding proteins is associated with a host of diseases including cancer.
Zfp36 encodes a zinc finger RNA binding protein, TTP, that binds to ARE-containing mRNAs and destabilizes them by recruiting deadenylases, thus promoting mRNA decay.
RNA sequence elements required for high affinity binding by the zinc finger domain of tristetraprolin: conformational changes coupled to the bipartite nature of Au-rich MRNA-destabilizing motifs.
TTP is a critical regulator of numerous proinflammatory cytokines. TTP KO mice develop multisystem inflammatory disease that is largely caused by excessive TNF expression.
Here, we report that knockin mice that have regulated increases in TTP levels throughout the body are protected from ADX-induced gastric inflammation and SPEM. Our results suggest that TTP could be a master regulator of gastric inflammation, and therapies that lead to increased TTP protein levels may be effective at treating gastric inflammation.
Chronic inflammation is strongly associated with gastric cancer development. Within the stomach, inflammation induces a well-defined histopathologic progression in which stomach damage leads to gastric atrophy, metaplasia, dysplasia, and adenocarcinoma.
SPEM is a potentially preneoplastic form of metaplasia that develops in response to damage within the gastric corpus that also may serve as a healing mechanism.
However, in the setting of prolonged damage, such as during chronic inflammation, SPEM becomes increasingly proliferative and eventually may progress toward carcinogenesis.
We found that TTP overexpression protected mice from gastric inflammation and SPEM development. We used ADX as a model to challenge the TTPΔARE mice. In WT mice, ADX triggered massive spontaneous inflammation of the gastric corpus followed by SPEM development. Both homozygous and heterozygous TTPΔARE mice were completely protected from ADX-induced gross inflammation and SPEM development. We previously reported that suppressing gastric inflammation by depleting macrophages in ADX WT mice protects them from SPEM development.
Thus, it is likely that TTP prevents SPEM development by suppressing inflammation. Importantly, we found that TTP overexpression did not affect HDT-induced SPEM development. These results suggest that TTP does not directly inhibit SPEM development and that protection from SPEM in the ADX model likely occurs by inhibiting inflammation. Our results suggest that therapies that elicit even a modest increase in TTP expression may effectively control gastric inflammation.
TTP primarily functions by binding to specific AREs within the 3’ UTR of target mRNAs, eventually promoting the degradation of the mRNA.
Our RNAseq studies showed that TTP potently suppressed genes associated with macrophage activation in ADX mice. Importantly, TTP regulates the expression of IL13 and TNF⍺, cytokines that have been implicated in inducing SPEM development.
Our RNAseq data showed that 33% of DEG transcripts in ADX WT mice contained potential TTP binding sites, including Tnf and Il13. TTP regulation of Il13 may be an important mechanism protecting from SPEM. Within the stomach, Il13 is potently expressed by type 2 innate lymphoid cells.
In response to gastric epithelial damage, Il13 is induced by IL33, which is released from the surface epithelial cells. IL13 drives alternative macrophage activation, which in turn drives SPEM development.
Interestingly, Il33 induction was greater in ADX TTPΔARE mice than in WT mice, and our analysis did not identify any TTP binding sites within the Il33 gene, suggesting that TPP may not directly regulate Il33 expression. Thus, TTP suppression of Il13 may be important for disrupting macrophage activation and protecting from SPEM development. However, given that TTP can regulate other cellular pathways, including those involving nuclear factor-κB,
it is likely that TTP can indirectly regulate the expression of additional inflammatory genes within the stomach.
Surprisingly, despite the almost complete suppression of inflammatory infiltrates into the stomachs of ADX TTPΔARE mice, we found striking up-regulation of numerous inflammatory transcripts and pathways. Increased TTP specifically suppressed the innate immune response, while pathways associated with the adaptive immune response were not affected significantly. It has been postulated previously that TTP preferentially regulates the innate immune response. However, although myeloid-specific TTP KO mice have an abnormal inflammatory response when challenged with lipopolysaccharide, they do not phenocopy the spontaneous inflammatory pathologies that develop in the whole-body TTP KO.
Hyperplastic gastric tumors with spasmolytic polypeptide-expressing metaplasia caused by tumor necrosis factor-alpha-dependent inflammation in cyclooxygenase-2/microsomal prostaglandin E synthase-1 transgenic mice.
Hyperplastic gastric tumors with spasmolytic polypeptide-expressing metaplasia caused by tumor necrosis factor-alpha-dependent inflammation in cyclooxygenase-2/microsomal prostaglandin E synthase-1 transgenic mice.
Thus, TNF may contribute to gastric carcinogenesis. Tnf mRNA is a well-known TTP target, and the numerous inflammatory pathologies that develop in TTP KO mice were rescued by treatment with TNF neutralizing antibodies
we hypothesized that TTP suppression of Tnf was the underlying mechanism by which TTPΔARE mice were protected from ADX-induced gastric inflammation. Surprisingly, we found that Tnf KO mice were at least partially susceptible to ADX-induced gastric inflammation and metaplasia. Interestingly, Tnf KO mice did not develop ADX-induced splenomegaly. These results show tissue-specific roles for TTP in regulating inflammation, and suggest that TTP’s anti-inflammatory role in the stomach is more complex than the suppression of a single proinflammatory cytokine.
Regulation of inflammation is multifaceted, occurring at the transcriptional level, post-transcriptional level, and beyond. We previously have shown that glucocorticoids are critical transcriptional regulators of gastric inflammation.
Here, we report that increased expression of the RNA binding protein TTP protects mice from gastric inflammation and metaplasia. Importantly, TTP transcription is induced by glucocorticoids.
TTP may be a key effector molecule by which glucocorticoids regulate the gastric inflammatory response and may be a useful therapeutic target for treating gastric inflammatory disease. Recent reports have found that TTP expression is decreased in gastric cancer samples.
Thus, there is a need for continued study into the role of TTP in suppressing gastric inflammation and carcinogenesis.
Materials and Methods
Animal Care and Treatment
All mouse studies were performed with approval by the National Institute of Environmental Health Sciences Animal Care and Use Committee. C57BL/6J mice were purchased from the Jackson Laboratories (000664; Bar Harbor, MA). TTPΔARE mice were generated as previously described and were maintained on a congenic C57Bl/6 genetic background.
Mice were administered standard chow and water ad libitum and maintained in a temperature and humidity-controlled room with standard 12-hour light/dark cycles. Sham, adrenalectomy, and castration surgeries were performed at 8 weeks of age by the National Institute of Environmental Health Sciences Comparative Medicine Branch. After ADX, mice were maintained on 0.85% saline drinking water to maintain ionic homeostasis. HDT treatment was performed as previously described by Saenz et al.
Briefly, mice received 3 consecutive intraperitoneal injections of 0.25 mg/g body weight tamoxifen (MilliporeSigma, Burlington, MA) every 24 hours. Stomach tissue was collected 24 hours after the final dose.
Histology
Mice were euthanized by cervical dislocation at the indicated time points. Stomachs were removed and opened along the greater curvature and washed in phosphate-buffered saline to remove gastric contents. Stomachs were fixed overnight in 4% paraformaldehyde at 4°C and then cryopreserved in 30% sucrose and embedded in optimal cutting temperature media. Histology and cell quantitation were performed as previously described.
Briefly, 5-μm stomach cryosections were incubated with antibodies against the H+/K+ adenosine triphosphatase α subunit (clone 1H9; MBL International Corporation, Woburn, MA), MIST1 (clone D7N4B; Cell Signaling Technologies, Danvers, MA), CD45 (clone 104; BioLegend, San Diego, CA), CD44v9 (Cosmo Bio, Tokyo, Japan), CD68 (clone E307V; Cell Signaling Technologies), Siglec F (clone 1RNM44NN; eBiosciences, San Diego, CA), Ki67 (clone D3B5; Cell Signaling Technologies), or CTNNB1 (clone 14; BD BioSciences, Franklin Lakes, NJ) for 1 hour at room temperature or overnight at 4°C. After washing in phosphate-buffered saline with 0.1% Triton X-100 (Thermo Fisher Scientific, Waltham, MA), sections were incubated in secondary antibodies for 1 hour at room temperature. Fluorescent-conjugated GSII lectin (Thermo Fisher Scientific, Waltham, MA) was added with secondary antibodies. Sections were mounted with Vectastain mounting media containing 4′,6-diamidino-2-phenylindole to visualize nuclei (Vector Laboratories, Burlingame, CA). Images were obtained using a Zeiss 710 confocal laser-scanning microscope equipped with Airyscan (Carl-Zeiss GmbH, Jena, Germany) and running Zen Black (Carl-Zeiss GmbH) imaging software.
Image Quantitation
Parietal cells and chief cells were quantitated as previously described
using confocal micrographs captured using a 20× microscope objective and 1-μm–thick optical sections. Cells were counted using the ImageJ (National Institutes of Health, Bethesda, MD) count tool. Cells that stained positive with anti-H+/K+ antibodies were identified as parietal cells, while cells that stained positive with anti-MIST1 antibodies and were GSII negative were identified as mature chief cells. Counts were reported as the number of cells observed per 20× field. Images were chosen that contained gastric glands cut longitudinally. Leukocytes were quantitated using Nikon Elements General Analysis (Nikon, Tokyo, Japan). Six tile-scanned images were captured using the 20× objective and stitched on Zen Black. Eosinophils were identified as CD45/Siglec F double-positive, while macrophages were CD45/CD68 double-positive.
RNA Isolation and qRT-PCR
RNA used for qRT-PCR and RNAseq was isolated from a 4-mm biopsy specimen of the gastric corpus lesser curvature. RNA was extracted in TRIzol (Thermo Fisher Scientific) and precipitated from the aqueous phase using 1.5 volumes of 100% ethanol. The mixture was transferred to a RNeasy column (Qiagen, Hilden, Germany), and the remaining steps were followed according to the RNeasy kit manufacturer’s recommendations. RNA was treated with RNase-free DNase I (Qiagen) as part of the isolation procedure. Reverse-transcription followed by qPCR was performed in the same reaction using the Universal Probes One-Step PCR kit (Bio-Rad Laboratories, Hercules, CA) and the TaqMan primers (Thermo Fisher Scientific) Cftr (Mm00445197_m1), Olfm4 (Mm01320260_m1), Wfdc2 (Mm00509434_m1), Zfp36 (Mm00457144_m1), Il33 (Mm00505403_m1), and Il13 (Mm00434204_m1) (Thermo Fisher Scientific) on a Quantstudio 6 (Thermo Fisher Scientific). mRNA levels were normalized to the reference gene Ppib (Mm00478295_m1).
RNAseq
RNA was isolated 5 days after sham surgery or adrenalectomy as described earlier. Four mice were used for each experimental group. Indexed samples were sequenced using the 75-bp paired-end protocol via the NextSeq500 (Illumina) per the manufacturer’s protocol. Raw reads (27–41 million pairs of reads per sample) were filtered using a custom perl script and the cutadapt program (v2.8) to remove low-quality reads and adapter sequences. Preprocessed reads were aligned to the University of California, Santa Cruz mm10 reference genome using STAR (v2.7.0f) with default parameters.
The quantification results from featureCounts (available in Subread software, v1.6.4) then were analyzed with the Bioconductor package DESeq2, which fits a negative binomial distribution to estimate technical and biological variability.
Comparisons were made between sham WT vs ADX WT, sham TTPΔARE vs ADX TTPΔARE, and ADX WT vs TTPΔARE. An abundance cut-off value was used so that only transcripts were evaluated whose average expression in the WT samples was greater than 0.1 fragments per kilobase of transcript per million mapped reads (FPKM). A transcript was considered differentially expressed if the adjusted P value was less than .05 and its expression changed -1.5-fold or less or 1.5-fold or more. Lists of significant transcripts were analyzed further using IPA (version 01-18-05; Qiagen). Enrichment or overlap was determined by IPA using the Fisher exact test (P < .05). GSEA was performed using GSEA v4.0.3 software (Broad Institute, San Diego, CA) and Molecular Signatures Database v7.0.
Transcripts were preranked based on their P value and their fold change of gene expression. This application scores a sorted list of transcripts with respect to their enrichment in selected functional categories (KEGG, Biocarta, Reactome and GO). The significance of the enrichment score was assessed using 1000 permutations. Benjamini and Hochberg’s false-discovery rate was calculated for multiple testing adjustments. A q value of 0.05 or less was considered significant. The heatmap was generated with the mean expression values of the 94 selected genes. The expression values were log2-transformed before subjecting to heatmap generation with scale by row in the pheatmap function available in R package pheatmap. The RNAseq data are available in the Gene Expression Omnibus repository at the National Center for Biotechnology Information (accession number: GSE164349; available at https://www.ncbi.nlm.nih.gov/geo).
Statistical Analysis
All error bars are ± SD of the mean. The sample size for each experiment is indicated in the figure legends. Experiments were repeated a minimum of 2 times. Statistical analyses were performed using 1-way analysis of variance with the post hoc Tukey t test when comparing 3 or more groups or by an unpaired t test when comparing 2 groups. Statistical analysis was performed by GraphPad Prism 8 software (GraphPad Software, San Diego, CA). Statistical significance was set at P ≤ .05. Specific P values are listed in the figure legends.
Acknowledgments
The authors thank the National Institute of Environmental Health Sciences Comparative Medicine Branch, Epigenomes Core Laboratory, and Fluorescence Microscopy and Imaging Center for their assistance. The authors also thank Michael Fessler and Donald Cook for critical reading of the manuscript.
CRediT Authorship Contributions
Jonathan T Busada, PhD (Conceptualization: Lead; Data curation: Lead; Formal analysis: Lead; Funding acquisition: Equal; Project administration: Lead; Writing – original draft: Lead)
Hyperplastic gastric tumors with spasmolytic polypeptide-expressing metaplasia caused by tumor necrosis factor-alpha-dependent inflammation in cyclooxygenase-2/microsomal prostaglandin E synthase-1 transgenic mice.
Wild-type and interleukin-10-deficient regulatory T cells reduce effector T-cell-mediated gastroduodenitis in Rag2-/- mice, but only wild-type regulatory T cells suppress Helicobacter pylori gastritis.
RNA sequence elements required for high affinity binding by the zinc finger domain of tristetraprolin: conformational changes coupled to the bipartite nature of Au-rich MRNA-destabilizing motifs.
Conflicts of interest The authors disclose no conflicts.
Funding This research was supported by the National Institute of General Medical Sciences 1Fi2GM123974 (J.T.B.) and P20GM121322 (J.T.B.), West Virginia University start-up funds (J.T.B), and by the Intramural Research Program of the National Institutes of Health/National Institute of Environmental Health Sciences 1ZIAES090057 (J.A.C) and 1ZIAES090080 (P.J.B.). Confocal microscopy was performed at the West Virginia University Microscope Imaging Facility, which is supported by National Institutes of Health grants P20RR016440, P30GM103488, U54GM104942, P30GM103503, and P20GM103434.
00161-2&id=gr1.jpg){kind=link}
00161-2&id=gr2.jpg){kind=link}
00161-2&id=gr3.jpg){kind=link}
00161-2&id=gr4.jpg){kind=link}
00161-2&id=gr5.jpg){kind=link}
00161-2&id=gr6.jpg){kind=link}
00161-2&id=gr7.jpg){kind=link}