Search

Yan Ji,1 Haorong Li,1 Jiguo Li,2 Guangqiang Yang,1 Wenli Zhang,1 Yan Shen,1 Bohui Xu,3 Jianping Liu,1 Jingyuan Wen,4 Wenting Song1 1Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, People’s Republic of China; 2Nanjing Miaobang Meiye Enterprise Management Co, LTD, Nanjing, People’s Republic of China; 3School of Pharmacy, Nantong University, Nantong, 226001, People’s Republic of China; 4School of Pharmacy, Faculty of Medical and Health Sciences, the University of Auckland, Auckland, New ZealandCorrespondence: Wenting Song, Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, 639 Longmian Avenue, Jiangning District, Nanjing, 211198, People’s Republic of China, Email songwenting908@163.com Jingyuan Wen, School of Pharmacy, Faculty of Medical and Health Sciences, the University of Auckland, Auckland, New Zealand, Email j.wen@auckland.ac.nzPurpose: Acne vulgaris is a chronic inflammatory skin disorder centered on hair follicles, making hair follicle-targeted delivery of anti-acne drugs a promising option for acne treatment. However, current researches have only focused on the delivering to healthy hair follicles, which are intrinsically different from pathologically clogged hair follicles in acne vulgaris.Patients and Methods: Azelaic acid (AZA) micro/nanocrystals with different particle sizes were prepared by wet media milling or high-pressure homogenization. An experiment on AZA micro/nanocrystals delivering to healthy hair follicles was carried out, with and without the use of physical enhancement techniques. More importantly, it innovatively designed an experiment, which could reveal the ability of AZA micro/nanocrystals to penetrate the constructed clogged hair follicles. The anti-inflammatory and antibacterial effects of AZA micro/nanocrystals were evaluated in vitro using a RAW264.7 cell model stimulated by lipopolysaccharide and a Cutibacterium acnes model. Finally, both the anti-acne e

To facilitate the research on intelligent and human-like chatbots with multi-modal context, we introduce a new video-based multi-modal dialogue dataset, called TikTalk. We collect 38K videos from a popular video-sharing platform, along with 367K conversations posted by users beneath them. Users engage in spontaneous conversations based on their multi-modal experiences from watching videos, which helps recreate real-world chitchat context. Compared to previous multi-modal dialogue datasets, the richer context types in TikTalk lead to more diverse conversations, but also increase the difficulty in capturing human interests from intricate multi-modal information to generate personalized responses. Moreover, external knowledge is more frequently evoked in our dataset. These facts reveal new challenges for multi-modal dialogue models. We quantitatively demonstrate the characteristics of TikTalk, propose a video-based multi-modal chitchat task, and evaluate several dialogue baselines. Experimental results indicate that the models incorporating large language models (LLM) can generate more diverse responses, while the model utilizing knowledge graphs to introduce external knowledge performs the best overall. Furthermore, no existing model can solve all the above challenges well. There is still a large room for future improvements, even for LLM with visual extensions. Our dataset is available at \url{https://ruc-aimind.github.io/projects/TikTalk/}., Comment: Accepted to ACM Multimedia 2023

Multimodal learning, especially large-scale multimodal pre-training, has developed rapidly over the past few years and led to the greatest advances in artificial intelligence (AI). Despite its effectiveness, understanding the underlying mechanism of multimodal pre-training models still remains a grand challenge. Revealing the explainability of such models is likely to enable breakthroughs of novel learning paradigms in the AI field. To this end, given the multimodal nature of the human brain, we propose to explore the explainability of multimodal learning models with the aid of non-invasive brain imaging technologies such as functional magnetic resonance imaging (fMRI). Concretely, we first present a newly-designed multimodal foundation model pre-trained on 15 million image-text pairs, which has shown strong multimodal understanding and generalization abilities in a variety of cognitive downstream tasks. Further, from the perspective of neural encoding (based on our foundation model), we find that both visual and lingual encoders trained multimodally are more brain-like compared with unimodal ones. Particularly, we identify a number of brain regions where multimodally-trained encoders demonstrate better neural encoding performance. This is consistent with the findings in existing studies on exploring brain multi-sensory integration. Therefore, we believe that multimodal foundation models are more suitable tools for neuroscientists to study the multimodal signal processing mechanisms in the human brain. Our findings also demonstrate the potential of multimodal foundation models as ideal computational simulators to promote both AI-for-brain and brain-for-AI research.

The fundamental goal of artificial intelligence (AI) is to mimic the core cognitive activities of human. Despite tremendous success in the AI research, most of existing methods have only single-cognitive ability. To overcome this limitation and take a solid step towards artificial general intelligence (AGI), we develop a foundation model pre-trained with huge multimodal data, which can be quickly adapted for various downstream cognitive tasks. To achieve this goal, we propose to pre-train our foundation model by self-supervised learning with weak semantic correlation data crawled from the Internet and show that promising results can be obtained on a wide range of downstream tasks. Particularly, with the developed model-interpretability tools, we demonstrate that strong imagination ability is now possessed by our foundation model. We believe that our work makes a transformative stride towards AGI, from our common practice of "weak or narrow AI" to that of "strong or generalized AI"., Comment: Published by Nature Communications, see https://www.nature.com/articles/s41467-022-30761-2

Multi-modal pre-training models have been intensively explored to bridge vision and language in recent years. However, most of them explicitly model the cross-modal interaction between image-text pairs, by assuming that there exists strong semantic correlation between the text and image modalities. Since this strong assumption is often invalid in real-world scenarios, we choose to implicitly model the cross-modal correlation for large-scale multi-modal pre-training, which is the focus of the Chinese project `WenLan' led by our team. Specifically, with the weak correlation assumption over image-text pairs, we propose a two-tower pre-training model called BriVL within the cross-modal contrastive learning framework. Unlike OpenAI CLIP that adopts a simple contrastive learning method, we devise a more advanced algorithm by adapting the latest method MoCo into the cross-modal scenario. By building a large queue-based dictionary, our BriVL can incorporate more negative samples in limited GPU resources. We further construct a large Chinese multi-source image-text dataset called RUC-CAS-WenLan for pre-training our BriVL model. Extensive experiments demonstrate that the pre-trained BriVL model outperforms both UNITER and OpenAI CLIP on various downstream tasks., Comment: This paper is the outcome of the Chinese multi-modal pre-training project called 'WenLan'

Qinxin Song,1–3 Danhui Li,3 Yongzhi Zhou,3 Jie Yang,1 Wanqi Yang,1 Guohua Zhou,2 Jingyuan Wen31Key Laboratory of Drug Quality Control and Pharmacovigilance, Ministry of Education, School of Pharmacy, China Pharmaceutical University, 2Department of Pharmacology, Jinling Hospital, Nanjing University School of Medicine, Nanjing, People’s Republic of China; 3School of Pharmacy, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New ZealandAbstract: The aim of this study was to evaluate (+)-catechin and (−)-epigallocatechin gallate (EGCG) cellular uptake and transport across human intestinal Caco-2 cell monolayer in both the absence and presence of niosomal carrier in variable conditions. The effect of free drugs and drug-loaded niosomes on the growth of Caco-2 cells was studied. The effects of time, temperature, and concentration on drug cellular uptake in the absence or presence of its niosomal delivery systems were investigated. The intestinal epithelial membrane transport of the drug-loaded niosomes was examined using the monolayer of the human Caco-2 cells. The kinetics of transport, and the effect of temperature, adenosine triphosphate inhibitor, permeability glycoprotein inhibitor, multidrug resistance-associated protein 2 inhibitor, and the absorption enhancer on transport mechanism were investigated. It was found that the uptake of catechin, EGCG, and their niosomes by Caco-2 cells was 1.22±0.16, 0.90±0.14, 3.25±0.37, and 1.92±0.22 µg/mg protein, respectively (n=3). The apparent permeability coefficient values of catechin, EGCG, and their niosomes were 1.68±0.16, 0.88±0.09, 2.39±0.31, and 1.42±0.24 cm/second (n=3) at 37°C, respectively. The transport was temperature- and energy-dependent. The inhibitors of permeability glycoprotein and multidrug resistance-associated protein 2 and the absorption enhancer s

Qinxin Song,1–3 Danhui Li,3 Yongzhi Zhou,3 Jie Yang,1 Wanqi Yang,1 Guohua Zhou,2 Jingyuan Wen31Key Laboratory of Drug Quality Control and Pharmacovigilance, Ministry of Education, School of Pharmacy, China Pharmaceutical University, 2Department of Pharmacology, Jinling Hospital, Nanjing University School of Medicine, Nanjing, People’s Republic of China; 3School of Pharmacy, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New ZealandAbstract: The aim of this study was to evaluate (+)-catechin and (−)-epigallocatechin gallate (EGCG) cellular uptake and transport across human intestinal Caco-2 cell monolayer in both the absence and presence of niosomal carrier in variable conditions. The effect of free drugs and drug-loaded niosomes on the growth of Caco-2 cells was studied. The effects of time, temperature, and concentration on drug cellular uptake in the absence or presence of its niosomal delivery systems were investigated. The intestinal epithelial membrane transport of the drug-loaded niosomes was examined using the monolayer of the human Caco-2 cells. The kinetics of transport, and the effect of temperature, adenosine triphosphate inhibitor, permeability glycoprotein inhibitor, multidrug resistance-associated protein 2 inhibitor, and the absorption enhancer on transport mechanism were investigated. It was found that the uptake of catechin, EGCG, and their niosomes by Caco-2 cells was 1.22±0.16, 0.90±0.14, 3.25±0.37, and 1.92±0.22 µg/mg protein, respectively (n=3). The apparent permeability coefficient values of catechin, EGCG, and their niosomes were 1.68±0.16, 0.88±0.09, 2.39±0.31, and 1.42±0.24 cm/second (n=3) at 37°C, respectively. The transport was temperature- and energy-dependent. The inhibitors of permeability glycoprotein and multidrug resistance-associated protein 2 and the absorption enhancer s

Ye Jin,1,4 Jingyuan Wen,2 Sanjay Garg,3 Da Liu,1 Yulin Zhou,1 Lirong Teng,1,* Weiyu Zhang,4,*1College of Life Science, Jilin University, Jilin, People’s Republic of China; 2School of Pharmacy, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; 3School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, Australia; 4School of Pharmacy, Changchun University of Chinese Medicine, Jilin, People’s Republic of China *These authors contributed equally to this workBackground: The aim of this study was to develop an optimal niosomal system to deliver Ginkgo biloba extract (GbE) with improved oral bioavailability and to replace the conventional GbE tablets.Methods: In this study, the film dispersion-homogenization method was used to prepare GbE niosomes. The resulting GbE niosome suspension was freeze-dried or spray-dried to improve the stability of the niosomes. GbE-loaded niosomes were formulated and characterized in terms of their morphology, particle size, zeta potential, entrapment efficiency, and angle of repose, and differential scanning calorimetry analysis was performed. In vitro release and in vivo distribution studies were also carried out.Results: The particle size of the optimal delivery system prepared with Tween 80, Span 80, and cholesterol was about 141 nm. There was a significant difference (P < 0.05) in drug entrapment efficiency between the spray-drying method (about 77.5%) and the freeze-drying method (about 50.1%). The stability study revealed no significant change in drug entrapment efficiency for the GbE niosomes at 4°C and 25°C after 3 months. The in vitro release study suggested that GbE niosomes can prolong the release of flavonoid glycosides in phosphate-buffered solution (pH 6.8) for up to 48 hours. The in vivo distribution study showed that the flavonoid glycoside content in the heart, lung, kidney, brain, and blood of rats treated with the GbE nios

The aim of this work was to prepare L-DOPA loaded poly(d,l-lactide-co-glycolide) (PLGA) nanoparticles by a modified water-in-oil-in-water (W1/O/W2) emulsification solvent evaporation method. A central composite design was applied for optimization of the formulation parameters and for studying the effects of three independent variables: PLGA concentration, polyvinyl alcohol (PVA) concentration and organic solvent removal rate on the particle size and the entrapment efficiency (response variables). Second-order models were obtained to adequately describe the influence of the independent variables on the selected responses. The analysis of variance showed that the three independent variables had significant effects (p < 0.05) on the responses. The experimental results were in perfect accordance with the predictions estimated by the models. Using the desirability approach and overlay contour plots, the optimal preparation area can be highlighted. It was found that the optimum values of the responses could be obtained at higher concentration of PLGA (5%, w/v) and PVA (6%, w/v); and faster organic solvent removal rate (700 rpm). The corresponding particle size was 256.2 nm and the entrapment efficiency was 62.19%. FTIR investigation confirmed that the L-DOPA and PLGA polymer maintained its backbone structure in the fabrication of nanoparticles. The scanning electron microscopic images of nanoparticles showed that all particles had spherical shape with porous outer skin. The results suggested that PLGA nanoparticles might represent a promising formulation for brain delivery of L-DOPA. The preparation of L-DOPA loaded PLGA nanoparticles can be optimized by the central composite design.

The aim of this work was to prepare L-DOPA loaded poly(d,l-lactide-co-glycolide) (PLGA) nanoparticles by a modified water-in-oil-in-water (W1/O/W2) emulsification solvent evaporation method. A central composite design was applied for optimization of the formulation parameters and for studying the effects of three independent variables: PLGA concentration, polyvinyl alcohol (PVA) concentration and organic solvent removal rate on the particle size and the entrapment efficiency (response variables). Second-order models were obtained to adequately describe the influence of the independent variables on the selected responses. The analysis of variance showed that the three independent variables had significant effects (p < 0.05) on the responses. The experimental results were in perfect accordance with the predictions estimated by the models. Using the desirability approach and overlay contour plots, the optimal preparation area can be highlighted. It was found that the optimum values of the responses could be obtained at higher concentration of PLGA (5%, w/v) and PVA (6%, w/v); and faster organic solvent removal rate (700 rpm). The corresponding particle size was 256.2 nm and the entrapment efficiency was 62.19%. FTIR investigation confirmed that the L-DOPA and PLGA polymer maintained its backbone structure in the fabrication of nanoparticles. The scanning electron microscopic images of nanoparticles showed that all particles had spherical shape with porous outer skin. The results suggested that PLGA nanoparticles might represent a promising formulation for brain delivery of L-DOPA. The preparation of L-DOPA loaded PLGA nanoparticles can be optimized by the central composite design.

Ye Jin,1,4 Jingyuan Wen,2 Sanjay Garg,3 Da Liu,1 Yulin Zhou,1 Lirong Teng,1,* Weiyu Zhang,4,*1College of Life Science, Jilin University, Jilin, People’s Republic of China; 2School of Pharmacy, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; 3School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, Australia; 4School of Pharmacy, Changchun University of Chinese Medicine, Jilin, People’s Republic of China *These authors contributed equally to this workBackground: The aim of this study was to develop an optimal niosomal system to deliver Ginkgo biloba extract (GbE) with improved oral bioavailability and to replace the conventional GbE tablets.Methods: In this study, the film dispersion-homogenization method was used to prepare GbE niosomes. The resulting GbE niosome suspension was freeze-dried or spray-dried to improve the stability of the niosomes. GbE-loaded niosomes were formulated and characterized in terms of their morphology, particle size, zeta potential, entrapment efficiency, and angle of repose, and differential scanning calorimetry analysis was performed. In vitro release and in vivo distribution studies were also carried out.Results: The particle size of the optimal delivery system prepared with Tween 80, Span 80, and cholesterol was about 141 nm. There was a significant difference (P < 0.05) in drug entrapment efficiency between the spray-drying method (about 77.5%) and the freeze-drying method (about 50.1%). The stability study revealed no significant change in drug entrapment efficiency for the GbE niosomes at 4°C and 25°C after 3 months. The in vitro release study suggested that GbE niosomes can prolong the release of flavonoid glycosides in phosphate-buffered solution (pH 6.8) for up to 48 hours. The in vivo distribution study showed that the flavonoid glycoside content in the heart, lung, kidney, brain, and blood of rats treated with the GbE nios