PubTator 3.0: an AI-powered literature resource for unlocking biomedical knowledge.

PubTator 3.0 (https://www.ncbi.nlm.nih.gov/research/pubtator3/) is a biomedical literature resource using state-of-the-art AI techniques to offer semantic and relation searches for key concepts like proteins, genetic variants, diseases and chemicals. It currently provides over one billion entity and relation annotations across approximately 36 million PubMed abstracts and 6 million full-text articles from the PMC open access subset, updated weekly. PubTator 3.0's online interface and API utilize these precomputed entity relations and synonyms to provide advanced search capabilities and enable large-scale analyses, streamlining many complex information needs. We showcase the retrieval quality of PubTator 3.0 using a series of entity pair queries, demonstrating that PubTator 3.0 retrieves a greater number of articles than either PubMed or Google Scholar, with higher precision in the top 20 results. We further show that integrating ChatGPT (GPT-4) with PubTator APIs dramatically improves the factuality and verifiability of its responses. In summary, PubTator 3.0 offers a comprehensive set of features and tools that allow researchers to navigate the ever-expanding wealth of biomedical literature, expediting research and unlocking valuable insights for scientific discovery.

Opportunities and challenges for ChatGPT and large language models in biomedicine and health.

ChatGPT has drawn considerable attention from both the general public and domain experts with its remarkable text generation capabilities. This has subsequently led to the emergence of diverse applications in the field of biomedicine and health. In this work, we examine the diverse applications of large language models (LLMs), such as ChatGPT, in biomedicine and health. Specifically, we explore the areas of biomedical information retrieval, question answering, medical text summarization, information extraction and medical education and investigate whether LLMs possess the transformative power to revolutionize these tasks or whether the distinct complexities of biomedical domain presents unique challenges. Following an extensive literature survey, we find that significant advances have been made in the field of text generation tasks, surpassing the previous state-of-the-art methods. For other applications, the advances have been modest. Overall, LLMs have not yet revolutionized biomedicine, but recent rapid progress indicates that such methods hold great potential to provide valuable means for accelerating discovery and improving health. We also find that the use of LLMs, like ChatGPT, in the fields of biomedicine and health entails various risks and challenges, including fabricated information in its generated responses, as well as legal and privacy concerns associated with sensitive patient data. We believe this survey can provide a comprehensive and timely overview to biomedical researchers and healthcare practitioners on the opportunities and challenges associated with using ChatGPT and other LLMs for transforming biomedicine and health.

Quality of Answers of Generative Large Language Models Versus Peer Users for Interpreting Laboratory Test Results for Lay Patients: Evaluation Study.

BACKGROUND: Although patients have easy access to their electronic health records and laboratory test result data through patient portals, laboratory test results are often confusing and hard to understand. Many patients turn to web-based forums or question-and-answer (Q&A) sites to seek advice from their peers. The quality of answers from social Q&A sites on health-related questions varies significantly, and not all responses are accurate or reliable. Large language models (LLMs) such as ChatGPT have opened a promising avenue for patients to have their questions answered. OBJECTIVE: We aimed to assess the feasibility of using LLMs to generate relevant, accurate, helpful, and unharmful responses to laboratory test-related questions asked by patients and identify potential issues that can be mitigated using augmentation approaches. METHODS: We collected laboratory test result-related Q&A data from Yahoo! Answers and selected 53 Q&A pairs for this study. Using the LangChain framework and ChatGPT web portal, we generated responses to the 53 questions from 5 LLMs: GPT-4, GPT-3.5, LLaMA 2, MedAlpaca, and ORCA_mini. We assessed the similarity of their answers using standard Q&A similarity-based evaluation metrics, including Recall-Oriented Understudy for Gisting Evaluation, Bilingual Evaluation Understudy, Metric for Evaluation of Translation With Explicit Ordering, and Bidirectional Encoder Representations from Transformers Score. We used an LLM-based evaluator to judge whether a target model had higher quality in terms of relevance, correctness, helpfulness, and safety than the baseline model. We performed a manual evaluation with medical experts for all the responses to 7 selected questions on the same 4 aspects. RESULTS: Regarding the similarity of the responses from 4 LLMs; the GPT-4 output was used as the reference answer, the responses from GPT-3.5 were the most similar, followed by those from LLaMA 2, ORCA_mini, and MedAlpaca. Human answers from Yahoo data were scored the lowest and, thus, as the least similar to GPT-4-generated answers. The results of the win rate and medical expert evaluation both showed that GPT-4's responses achieved better scores than all the other LLM responses and human responses on all 4 aspects (relevance, correctness, helpfulness, and safety). LLM responses occasionally also suffered from lack of interpretation in one's medical context, incorrect statements, and lack of references. CONCLUSIONS: By evaluating LLMs in generating responses to patients' laboratory test result-related questions, we found that, compared to other 4 LLMs and human answers from a Q&A website, GPT-4's responses were more accurate, helpful, relevant, and safer. There were cases in which GPT-4 responses were inaccurate and not individualized. We identified a number of ways to improve the quality of LLM responses, including prompt engineering, prompt augmentation, retrieval-augmented generation, and response evaluation.

Hexa-atom Pt Catalyst Fabricated by a Ligand Engineering Strategy for Efficient Hydrogen Oxidation Reaction.

Atomically precise supported nanocluster catalysts (APSNCs), which feature exact atomic composition, well-defined structures, and unique catalytic properties, offer an exceptional platform for understanding the structure-performance relationship at the atomic level. However, fabricating APSNCs with precisely controlled and uniform metal atom numbers, as well as maintaining a stable structure, remains a significant challenge due to uncontrollable dispersion and easy aggregation during synthetic and catalytic processes. Herein, we developed an effective ligand engineering strategy to construct a Pt6 nanocluster catalyst stabilized on oxidized carbon nanotubes (Pt6/OCNT). The structural analysis revealed that Pt6 nanoclusters in Pt6/OCNT were fully exposed and exhibited a planar structure. Furthermore, the obtained Pt6/OCNT exhibited outstanding acidic HOR performances with a high mass activity of 18.37 A·mgpt-1 along with excellent stability during a 24 h constant operation and good CO tolerance, surpassing those of the commercial Pt/C. Density functional theory (DFT) calculations demonstrated that the unique geometric and electronic structures of Pt6 nanoclusters on OCNT altered the hydrogen adsorption energies on catalytic sites and thus lowered the HOR theoretical overpotential. This work presents a new prospect for designing and synthesizing advanced APSNCs for efficient energy electrocatalysis.

Al3+ Dopants Induced Mg2+ Vacancies Stabilizing Single-Atom Cu Catalyst for Efficient Free-Radical Hydrophosphinylation of Alkenes.

Utilizing heterogeneous catalysts to overcome obstacles for homogeneous reactions is fascinating but very challenging owing to the difficult fabrication of such catalysts based on the character of target reactions. Herein, we report a Al3+ doping strategy to construct single-atom Cu on MgO nanosheets (Cu1/MgO(Al)) for boosting the free-radical hydrophosphinylation of alkenes. Al3+ dopants in MgO bring about abundant Mg2+ vacancies for stabilizing dense independent Cu atoms (6.3 wt %), while aggregated Cu nanoparticles are formed without Al3+ dopants (Cu/MgO). Cu1/MgO(Al) exhibits preeminent activity and durability in the hydrophosphinylation of various alkenes with great anti-Markovnikov selectivity (99%). The turnover frequency (TOF) value reaches up to 1272 h-1, exceeding those of Cu/MgO by ∼6-fold and of traditional homogeneous catalysts drastically. Further experimental and theoretical studies disclose that the prominent performance of Cu1/MgO(Al) derives from the accelerated initiating step of phosphinoyl radical triggered by individual Cu atoms.

Iridium Doped Pyrochlore Ruthenates for Efficient and Durable Electrocatalytic Oxygen Evolution in Acidic Media.

Developing highly active, durable, and cost-effective electrocatalysts for the oxygen evolution reaction (OER) is of prime importance in proton exchange membrane (PEM) water electrolysis techniques. Ru-based catalysts have high activities but always suffer from severe fading and dissolution issues, which cannot satisfy the stability demand of PEM. Herein, a series of iridium-doped yttrium ruthenates pyrochlore catalysts is developed, which exhibit better activity and much higher durability than commercial RuO2 , IrO2 , and most of the reported Ru or Ir-based OER electrocatalysts. Typically, the representative Y2 Ru1.2 Ir0.8 O7 OER catalyst demands a low overpotential of 220 mV to achieve 10 mA cm-2 , which is much lower than that of RuO2 (300 mV) and IrO2 (350 mV). In addition, the catalyst does not show obvious performance decay or structural degradation over a 2000 h stability test. EXAFS and XPS co-prove the reduced valence state of ruthenium and iridium in pyrochlore contributes to the improved activity and stability. Density functional theory reveals that the potential-determining steps barrier of OOH* formation is greatly depressed through the synergy effect of Ir and Ru sites by balancing the d band center and oxygen intermediates binding ability.

Well-Defined Materials for Heterogeneous Catalysis: From Nanoparticles to Isolated Single-Atom Sites.

The use of well-defined materials in heterogeneous catalysis will open up numerous new opportunities for the development of advanced catalysts to address the global challenges in energy and the environment. This review surveys the roles of nanoparticles and isolated single atom sites in catalytic reactions. In the second section, the effects of size, shape, and metal-support interactions are discussed for nanostructured catalysts. Case studies are summarized to illustrate the dynamics of structure evolution of well-defined nanoparticles under certain reaction conditions. In the third section, we review the syntheses and catalytic applications of isolated single atomic sites anchored on different types of supports. In the final part, we conclude by highlighting the challenges and opportunities of well-defined materials for catalyst development and gaining a fundamental understanding of their active sites.

A Machine-Learning Based Approach for Predicting Older Adults' Adherence to Technology-Based Cognitive Training.

Adequate adherence is a necessary condition for success with any intervention, including for computerized cognitive training designed to mitigate age-related cognitive decline. Tailored prompting systems offer promise for promoting adherence and facilitating intervention success. However, developing adherence support systems capable of just-in-time adaptive reminders requires understanding the factors that predict adherence, particularly an imminent adherence lapse. In this study we built machine learning models to predict participants' adherence at different levels (overall and weekly) using data collected from a previous cognitive training intervention. We then built machine learning models to predict adherence using a variety of baseline measures (demographic, attitudinal, and cognitive ability variables), as well as deep learning models to predict the next week's adherence using variables derived from training interactions in the previous week. Logistic regression models with selected baseline variables were able to predict overall adherence with moderate accuracy (AUROC: 0.71), while some recurrent neural network models were able to predict weekly adherence with high accuracy (AUROC: 0.84-0.86) based on daily interactions. Analysis of the post hoc explanation of machine learning models revealed that general self-efficacy, objective memory measures, and technology self-efficacy were most predictive of participants' overall adherence, while time of training, sessions played, and game outcomes were predictive of the next week's adherence. Machine-learning based approaches revealed that both individual difference characteristics and previous intervention interactions provide useful information for predicting adherence, and these insights can provide initial clues as to who to target with adherence support strategies and when to provide support. This information will inform the development of a technology-based, just-in-time adherence support systems.

Long noncoding RNA LBX2-AS1-modulated miR-4766-5p regulates gastric cancer development through targeting CXCL5.

BACKGROUND: Long noncoding RNAs (LncRNAs) have been reported to critically regulate gastric cancer (GC). Recently, it was reported that LBX2 antisense RNA 1 (LBX2-AS1) is abnormally expressed in GC. However, the role of LBX2-AS1 in the malignancy of GC is worth further discussion. METHODS: Quantitative real-time polymerase chain reaction (qRT-PCR) was used to determine the LBX2-AS1, miR-4766-5p and C-X-C motif chemokine (CXCL5) expression in GC tissues and cells. Dual-luciferase reporter assay was applied to examine the target relationship between LBX2-AS1 and miR-4766-5p or miR-4766-5p and CXCL5. Cell counting kit-8 (CCK-8) and Transwell assays were used to detect cell proliferation, migration and invasion rates. The protein expression of CXCL5 was confirmed using western blot. The RNA pull down experiment was used to verify the specificity of LBX2-AS1 and miR-4766-5p on BGC-823 and SGC-7901 cells. RESULTS: LBX2-AS1 was up-regulated in GC tissues and cells, and its knockdown suppressed proliferation, migration and invasion of GC cells. While, overexpression of LBX2-AS1 increased proliferation and increased CXCL5 mRNA level. CXCL5 improved cell proliferation, migration and invasion of GC cells. LBX2-AS1 could bind to miR-4766-5p to regulate CXCL5 expression. Overexpression of CXCL5 overturned those effects of miR-4766-5p in GC cells. RNA Pull down shown that BGC-823 and SGC-7901 cells, miR-4766-5p specifically binds to LBX2-AS1. CONCLUSIONS: In short, this study demonstrated that LBX2-AS1 promoted proliferation, migration and invasion through up-regulation CXCL5 mediated by miR-4766-5p in GC. The LBX2-AS1/miR-4766-5p/CXCL5 regulatory axis provides a theoretical basis for the research on lncRNA-directed therapeutics in GC.

Strain Engineering to Enhance the Electrooxidation Performance of Atomic-Layer Pt on Intermetallic Pt3Ga.

Strain engineering has been a powerful strategy to finely tune the catalytic properties of materials. We report a tensile-strained two-to-three atomic-layer Pt on intermetallic Pt3Ga (AL-Pt/Pt3Ga) as an active electrocatalyst for the methanol oxidation reaction (MOR). Atomic-resolution high-angle annular dark-field scanning transmission electron microscopy characterization showed that the AL-Pt possessed a 3.2% tensile strain along the [001] direction while having a negligible strain along the [100]/[010] direction. For MOR, this tensile-strained AL-Pt electrocatalyst showed obviously higher specific activity (7.195 mA cm-2) and mass activity (1.094 mA/µgPt) than those of its unstrained counterpart and commercial Pt/C catalysts. Density functional theory calculations demonstrated that the tensile-strained surface was more energetically favorable for MOR than the unstrained one, and the stronger binding of OH* on stretched AL-Pt enabled the easier removal of CO*.

Temperature-Controlled Selectivity of Hydrogenation and Hydrodeoxygenation in the Conversion of Biomass Molecule by the Ru1/mpg-C3N4 Catalyst.

Hydrogenation and hydrodeoxygenation are significant and distinct approaches for the conversion of biomass and biomass-derived oxygenated chemicals into high value-added chemicals and fuels. However, it remains a great challenge to synthesize catalysts that simultaneously possess excellent hydrogenation and hydrodeoxygenation performance. Herein, we report a catalyst made of isolated single-atom Ru supported on mesoporous graphitic carbon nitride (Ru1/mpg-C3N4), fabricated by a wet impregnation method. The as-prepared Ru1/mpg-C3N4 catalyst shows excellent hydrogenation and hydrodeoxygenation performance. First-principles calculations reveal that the Ru atom is mobilized, and the active site is induced by adsorption of the reactants. A systematic reaction mechanism is proposed, suggesting that vanillyl alcohol is the deoxygenation prohibited product, while 2-methoxy- p-cresol is the deoxygenation allowed product. Thus, the excellent selectivity for the hydrogenation or hydrodeoxygenation of vanillin at different temperatures results from switching between the two types of products.

Metal (Hydr)oxides@Polymer Core-Shell Strategy to Metal Single-Atom Materials.

Preparing metal single-atom materials is currently attracting tremendous attention and remains a significant challenge. Herein, we report a novel core-shell strategy to synthesize single-atom materials. In this strategy, metal hydroxides or oxides are coated with polymers, followed by high-temperature pyrolysis and acid leaching, metal single atoms are anchored on the inner wall of hollow nitrogen-doped carbon (CN) materials. By changing metal precursors or polymers, we demonstrate the successful synthesis of different metal single atoms dispersed on CN materials (SA-M/CN, M = Fe, Co, Ni, Mn, FeCo, FeNi, etc.). Interestingly, the obtained SA-Fe/CN exhibits much higher catalytic activity for hydroxylation of benzene to phenol than Fe nanoparticles/CN (45% vs 5% benzene conversion). First-principle calculations further reveal that the high reactivity originates from the easier formation of activated oxygen species at the single Fe site. Our methodology provides a convenient route to prepare a variety of metal single-atom materials representing a new class of catalysts.

Rational Design of Single Molybdenum Atoms Anchored on N-Doped Carbon for Effective Hydrogen Evolution Reaction.

The highly efficient electrochemical hydrogen evolution reaction (HER) provides a promising pathway to resolve energy and environment problems. An electrocatalyst was designed with single Mo atoms (Mo-SAs) supported on N-doped carbon having outstanding HER performance. The structure of the catalyst was probed by aberration-corrected scanning transmission electron microscopy (AC-STEM) and X-ray absorption fine structure (XAFS) spectroscopy, indicating the formation of Mo-SAs anchored with one nitrogen atom and two carbon atoms (Mo1 N1 C2 ). Importantly, the Mo1 N1 C2 catalyst displayed much more excellent activity compared with Mo2 C and MoN, and better stability than commercial Pt/C. Density functional theory (DFT) calculation revealed that the unique structure of Mo1 N1 C2 moiety played a crucial effect to improve the HER performance. This work opens up new opportunities for the preparation and application of highly active and stable Mo-based HER catalysts.

Quantitatively Monitoring the Size-Focusing of Au Nanoclusters and Revealing What Promotes the Size Transformation from Au44(TBBT)28 to Au36(TBBT)24.

"Size-focusing" is a well-recognized process and widely employed for the synthesis of atomically monodisperse metal nanoclusters. However, quantitatively monitoring the size-focusing of Au nanoclusters has not been achieved yet, and the in-depth understanding of the size focusing is far from completed. Herein, we introduce a facile, cheap, and powerful tool, preparative thin-layer chromatography (PTLC), to quantitatively track the size-focusing process, to reveal that mainly ∼3 nm nanoparticles promote the transformation from Au44(TBBT)28 to Au36(TBBT)24 (where TBBT is 4-tert-butylbenzenethiolate) and to improve the syntheses of Au44(TBBT)28 and Au36(TBBT)24. Our work further demonstrates the usefulness of PTLC in nanocluster research and advances one step toward understanding the "size-focusing" process of nanoclusters.

Effect of Neoadjuvant Chemotherapy Treatment on Prognosis of Patients with Advanced Gastric Cancer: a Retrospective Study.

OBJECTIVE: To evaluate the prognostic effects of neoadjuvant chemotherapy (NAC) in patients with local advanced gastric cancer. METHODS: We retrospectively analyzed prognosis in 191 patients with advanced gastric cancer, of whom 71 were treated with NAC and 120 received surgery only between February 2007 and July 2013. Postoperative complication rate was recorded. Survival by clinicopathological features, pathological T and N stages, and histopathological tumor regression was retrospectively compared between the two groups. RESULTS: According to Response Evaluation Criteria in Solid Tumors, none of the 71 patients in the NAC followed by surgery group showed complete response, 36 showed partial response, 25 had stable disease, and 10 had progressive disease. The chemotherapy response rate was 50.7%; the disease control rate was 85.9%. Grade 3/4 adverse events were seen in less than 20% patients, with acceptable toxicities. No difference was found in the overall postoperative complication rates between the two groups (7 versus 22 cases, P=0.18). Median survival time was significantly different, at 54 months in the NAC combined with surgery group and 25 months in the surgery-only group (P=0.025). CONCLUSION: In patients with operable gastric adenocarcinomas, NAC can significantly improve overall survival without increasing surgical complications.

[Clinical pathological features and prognosis analysis of gastrointestinal stromal tumor: a series of 558 cases].

OBJECTIVE: To investigate and analysis the clinical and pathological characteristic of gastrointestinal stromal tumor (GIST) patients, and to clarify the factors that effect on prognosis. METHODS: The clinical and pathological features and follow-up of GIST patients who received surgery in Peking Union Medical College Hospital from May 2002 to December 2013 were analyzed retrospectively. The prognosis was evaluated by univariate and multivariate analysis. Kaplan-Meier unvariate analysis and Log-rank test were used to compare the survival rates. Multivariate factors for survival were analyzed by Cox proportional hazards regression model. RESULTS: A total of 558 GIST patients were collected, including 284 males and 272 females. The high incidence was in the elderly and age of 50 to 70 years. Most of the primary tumors are located in stomach (303 cases), followed by the small intestine (118 cases). Surgical procedures included R0 resection in 517 cases, R1 resection in 4 cases, R2 or palliative resection in 37 cases. The recurrence risk was very low in 102 cases, low in 156 cases, moderate in 67 cases and high in 233 cases. Of all the patients, 495 cases completed the follow-up, the follow-up rate was 88.7%. Five year survival rate was 87.4%. Patients who took targeted therapy with moderate and high risk of recurrence had a better prognosis compared with not taking the drug. Univariate analysis revealed that the factors impacting the prognosis were age, tumor size, tumor site and mitotic count. Multivariate analysis showed that tumor size (P=0.01, RR=1.562, 95% CI: 1.452 to 15.664), location (P=0.01, RR=1.552, 95% CI:1.324 to 12.225), mitotic figures (P<0.01, RR=1.415, 95% CI: 2.126 to 7.968) and tumor rupture (P=0.01, RR=1.578, 95% CI: 1.543 to 15.892) were independent prognostic factors. CONCLUSION: R0 resection combined with targeted therapy is the best treatment of GIST. Tumor size, location, mitosis count and tumor rupture are independent prognostic factors of GIST patients.

[Effect of n-3 polyunsaturated fatty acids on gut microbiota and endotoxin levels in portal vein of rats fed with high-fat diet].

OBJECTIVE: To investigate the effect of n-3 polyunsaturated fatty acids (n-3PUFAs) on gut microbiota and endotoxin levels in portal vein of rats fed with a high-fat diet (HFD). METHODS: Thirty-six male Sprague-Dawley rats were randomly divided into four groups and fed with normal control diet (CD), HFD, CD supplemented with n-3PUFAs, and HFD supplemented with n-3PUFAs, respectively. Fresh fecal samples were collected to analyze the gut microbiota 10 weeks after feeding. DNA was exacted from the fresh fecal samples. Quantitative PCR was used to detect the composition of the gut microbiota. The endotoxin levels were detected through modified azo chromogenic substrate limulus amebocyte lysate assay. RESULTS: The differences in body weight before breeding in each group were not statistically significant among these four groups (P=0.613). The increase in the body weight was significantly larger in the HFD group than in the CD group (P=0.0002), CD+n-3PUFAs group (P=0.0001), and HFD+n-3PUFAs group (P=0.022). There were significantly more firmicutes (P=0.002) and enterobacteriales (P=0.022) and significantly less bacteroidetes (P=0.026) and bifidobactera (P=0.034) in the gut of rats from HFD group than those from the CD group. There were significantly more bacteroidetes in the fecal samples of the rats from the CD+n-3PUFAs group compared to those from the CD group (P=0.043). There were significantly more firmicutes (P=0.044)and enterobacteriales (P=0.012) and less bacteroidetes (P=0.042) in the fecal samples of the rats from HFD group compared to those from the HFD+n-3PUFAs group. The endotoxin in plasma form portal vein of rats in HFD group were significantly higher than in CD group (P=0.007) and HFD+n-3PUFAs group (P=0.042) but showed no significant difference between CD+n-3PUFAs and CD group (P=0.210). CONCLUSIONS: HFD can increase body weight and change gut microbiota. Supplementation of n-3PUFAs can partially counteract such gut dysbiosis, lower endotoxin level in portal vein blood, and improve the body weight.

[MicroRNA and gastric cancer].

Gastric cancer is caused by the interaction of genetic and environmental factors. MicroRNA (miRNA) is involved in many cellular processes including proliferation, differentiation, and apoptosis and plays an important role in pathogenesis of gastric cancer, as demonstrated in many recent studies from perspectives including miRNA profiling, reciprocal modulation between epigenetic and miRNA, and Helicobacter pylori infection. MiRNA is highly stabe in blood, and therefore non-invasive diagnosis of gastric cancer using circulating miRNA may be promising.

[Effects of eukaryotic translation initiation factor 5A2 down-regulation by small interfering RNA on aggressiveness of MKN28 human].

OBJECTIVE: To investigate the effects of eukaryotic translation initiation factor 5A2 (EIF5A2) down-regulation by small interfering RNA (siRNA) on aggressiveness of human gastric cancer cell and its potential mechanisms. METHODS: The expressions of EIF5A2 in human gastric cancer cell lines (MKN28 and HGC27) and immortalized gastric mucosal epithelial cells (GES-1) were measured by real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR) and Western blotting. EIF5A2 gene in MKN28 cells was silenced by RNA interference and the inhibitory effect was evaluated by both qRT-PCR and Western blotting. Cell proliferation was assessed by CCK-8 assay. Cell migration and invasion were assessed by Transwell assay. The possible downstream targets of EIF5A2, such as CyclinD1, CyclinD3, matrix metallopeptidase-9 (MMP-9), E-cadherin, vimintin, C-myc, and metastasis-associated protein 1 (MTA1) expression levels, were examined by Western blotting. RESULTS: High expressions of EIF5A2 were found in MKN28 cells and human gastric adenocarcinoma tissues. Both EIF5A2 mRNA and protein expression in MKN28 cells were significantly down-regulated by siRNA#1 and siRNA#2, especially siRNA#1. Knockdown of EIF5A2 caused an apparent suppression of MKN28 cell proliferation (all P<0.01), migration (P<0.001), and invasion (P<0.001). After the knockdown of EIF5A2 in MKN28 cells, E-cadherin levels were upregulated, whereas vimentin, Cyclin D1, Cyclin D3, C-myc and MTA1 levels were downregulated. CONCLUSION: Knockdown of EIF5A2 may inhibit MKN28 cell proliferation by downregulating the CyclinD1 and CyclinD3 and suppressing the cell migration and invasion by inhibiting MTA1, C-myc and epithelial-mesenchymal transition.