Home > News content

Well-known Nature Journal Published Tencent AI Lab and other medical cooperation research results

via:博客园     time:2017/11/27 19:01:29     readed:479

On November 23, a joint study by Jianmin Cui Lab of WUSTL, David Fedida Lab of UBC and the AI ​​Lab of Tencent has solved the problem of KCNQ1 channel inactivation mechanism over the years [1]. This is a heart rate control mechanism that is closely related to heart disease and controlled by an important potassium channel.

The latest research is published in Nature Communications, a subsidiary of Nature, a world-renowned journal whose impact factor 12.124, all SCI indexes in the JCR's Multidisciplinary ScienceRanked third in the journals, second only to the well-known Nature and Science.

orgsrc=https://images2018.cnblogs.com/news/66372/201711/66372-20171127184412675-716921229.jpg

JCR Integrated Science SCI index journals rankings

In the heart, KCNQ1 potassium channels control the repolarization of action potentials, and point mutations often lead to heart attacks. The KCNQ1 channel has a special switching mechanism and shows different deactivation characteristics than other ion channels. The elucidation of this inactivation characteristic is crucial for the channel's working mechanism, the pathology and treatment of heart disease. In this paper, the fluorescence inactivation mechanism of KCNQ1 channel is revealed by fluorescence electrophysiological, single-channel recording and data-driven Markov process modeling. This study provides a new idea for the further clinical study of diagnosis and treatment of abnormal heart rate.

Article link:https://www.nature.com/articles/s41467-017-01911-8

About cooperative research institutions

Jianmin Cui Lab from the University of Washington, St. Louis, USA, has been working on ion channel gating mechanisms and pharmacology for many years and his research team has been leading the way in the study of channels such as the BK channel and KCNQ1 / IKs Internationally enjoys a wide reputation. The research results are published in international top journals such as Nature, Nature Communications, Nature Genetics, Nature Structure and Molecular Biology, PNAS, Neuron, Journal of Neuroscience.

David Fedida Lab from UBC Previously studied the single-channel characteristics of IKs, which is the first successful direct experiment record of single channel of KCNQ1. The research results have been published on many international journals such as Nature Communications, Immunity, Circulation Research, PNAS, Trends in Pharmacological Sciences, Journal of Neuroscience.

Tencent AI Lab beach layout AI + medical

AI + health care industry increased its focus, may become the first AI technology boots. As an enterprise-class artificial intelligence laboratory in Tencent, Tencent AI Lab utilizes the dual advantages of basic research + application exploration and strikes the layout of the medical industry.

In August 2017, when AI Labs led the launch of the company by Tencent Internet + Partnerships, AI Lab was the first to enter the clinical pre-trial program, screening an endoscopy in less than 4 seconds and finding an accuracy rate of 90%.

In November 2017, "Tencent Mirage" combined AI + medical technology with exclusive technical support from AI Lab to combine natural language processing with machine learning and mass medical data to help doctors better understand medical records, reduce risks and improve diagnosis.

That month, the Ministry of Science and Technology released a new generation of workers in the intelligent development plan, and led the construction of medical imaging national artificial intelligence open and innovative platform.

Study Details

The American Heart Association measures cardiovascular health in adults using seven cardio-health indicators and is divided into four cardio-behavioral and four cardio-cardiac factors [2]. A recent survey of cardiovascular health in Chinese adults in the Journal of the American College of Cardiology found that in China, 13.5% of all four of the four cardiac health indicators met, The number of people who met all four of the four four-cardio heartbeat metrics was only 7 & permil; those who met all seven of their health indicators fell even further to 2 & permil; [3] Women usually only meet four to five of the seven indicators, while men only achieve three or four indicators in an ideal state. [3] In such a severe situation, an in-depth study of heart disease can not be delayed.

Heart rate is a measure of the number of beats per minute in a quiet, normal state and is closely related to heart disease. One of the most important heart rate regulators is controlled by an important potassium channel, the KCNQ1 channel. In the heart, KCNQ1 channels combine with its helper subunit KCNE1 to form IKs channels to pump out potassium ions to repolarize heart action potentials. Therefore, this channel is one of the most important factors determining the heart rate.

KCNQ1 channel about 700 amino acids, nearly 20 years, clinical studies have found that more than 300 point mutations in the channel can lead to heart disease, and a considerable part of the mutation will affect the channel inactivation characteristics, so KCNQ1 channel The study of inactivation mechanism is particularly important.

Common inactivation of channels is N-terminal inactivation (relatively fast) and C-terminal inactivation (relatively slow), whereas there is a unique inactivation in the KCNQ1 potassium channel, which is typical Of Type N or Type C inactivation characteristics, and only appear under certain voltage stimuli (as shown in Figure 1 below). So its mechanism has been elusive.

orgsrc=https://images2018.cnblogs.com/news/66372/201711/66372-20171127184412784-458383583.jpg

Figure 1 The KCNQ1 channel shows a "deactivation" behavior at certain voltage stimuli, as indicated by the hook current in the figure.

Previously, the Cui Lab at Washington University in St. Louis discovered a unique gating mechanism for the KCNQ1 channel in 2014 [4]. Unlike other ion channels, which have only one open state, the channel has two distinct open states: the intermediate open state (IO) and the activated open state (AO).

In this article, researchers found that the AO open state is actually the so-called "inactivation state" of the KCNQ1 channel. Since the deactivation properties of the channel are in the same time-dependent and voltage dependent manner as their AO open state (as shown in Figure 2a below), and the two point mutations S338F and F351A located on the sixth transmembrane structure of the channel can be selectively Both IO and AO states are open, and because they all have only one open state, both mutant channels no longer have inactivation properties (as shown in Figures 2b and c below). Researchers also found the "inactivation" of the KCNQ1 channel from the single-channel current level: the odds ratio for the AO state is lower than for the IO state (as shown in Figure 2d below). This is the first time that KCNQ1 single-channel current is recorded directly and experiments have shown that its electrical channel conductance is as low as ~ 0.18 pS, which is the smallest known single-channel conductance (single-channel conductance for large-conductance calcium-activated potassium channels can be large To 300 pS).

orgsrc=https://images2018.cnblogs.com/news/66372/201711/66372-20171127184412878-1161899041.jpg

Figure 2 (a) KCNQ1 channel inactivation properties (A2) have the same time-dependent and voltage-dependent properties as their AO open states (Islow). (b-c) Both S338F and F351A are selectively open only to the IO and AO states, respectively, and therefore no longer have inactivation properties. (d) Single-channel current of KCNQ1 channel.

The researchers then further studied the "inactivation" of the KCNQ1 channel through a data-driven approach. In particular, a Six-State Markov Model is established in the paper, which includes IO state and AO state. In this six-state model, the researchers assume that the one-step transfer probability is independent of time, ie, a homogeneous Markov model. Since then, the study models the one-step transition probability matrix of the homogeneous Markov model and fits the optimal parameters (as shown in Figure 3a below).

KRI, KIR, KIA and KAI are all voltage functions in this one-step transfer probability matrix, which allows researchers to establish different homogeneous Markov models of the "inactivation" of KCNQ1 channels at different voltages . Finally, the study passes Kolmogorov forward and backward equations

(Kolmogorov Forward / Backward Equation) and the smooth distribution of the Markov process, the KCNQ1 channel's dynamic and steady-state outputs at different voltages, namely the different "inactivation" characteristics, are successfully simulated (as shown in Figure 3b below) . So far, this study further demonstrates the correctness of the model presented in the paper while further understanding the "inactivation" of the KCNQ1 channel.

orgsrc=https://images2018.cnblogs.com/news/66372/201711/66372-20171127184412737-1889684353.jpg

Figure 3 KCNQ1 channel modeling and its "deactivation" simulation.

This article advances the study of KCNQ1 channel gating mechanism and related pathology. In recent years, with the rapid development of structural biology, more and more ion channel structures have been analyzed. However, how to switch ion channels remains a hot and difficult topic in the field. This study shows that the AO state of KCNQ1 channel may have a different mechanism from the classical voltage-sensing region to pore region, and a breakthrough is found in the study of non-canonical VSD-pore coupling mechanism.

This research is a joint research result of Tencent's AI Lab and internationally renowned universities. The partners include Washington University in St. Louis and University of British Columbia in Canada. This study provides a new idea for the further clinical study of diagnosis and treatment of abnormal heart rate.

In the future, Tencent AI Lab will continue to promote AI + medical deployment, deepen cooperation with world-renowned universities and medical schools, and overcome more medical challenges with AI technology.

[1] Panpan Hou, Jodene Eldstrom, Jingyi Shi, Ling Zhong, Kelli McFarland, Yuan Gao, David Fedida, Jianmin Cui. Inactivation of KCNQ1 potassium channels reveals dynamic coupling between voltage sensing and pore opening. Nature Communications 8 (1): 1730 2017

[2] The American Heart Association is the American Heart Association. The seven Cardiovascular Health Metrics are: smoking, body mass index, physical activity, diet health, lipids, blood pressure, and blood sugar. The four Cardiovascular Herself Behaviors include smoking, BMI, physical activity, and diet. Four Cardiovascular Health Factors include smoking, blood fat, blood pressure, blood sugar. Whether smoking is both a heart health behavior index, but also be counted as cardiac health indicators.

[3] Yufang Bi, Yong Jiang, Jiang He, Yu Xu, Limin Wang, Min Xu, Mei Zhang, Yichong Li, Tiange Wang, Meng Dai, Jieli Lu, Mian Li, Chung-Shiuan Chen, Shenghan Lai, Weiqing Wang, Status of Cardiovascular Health in Chinese Adults. Journal of the American College of Cardiology, 65 (10): 1013-1025, 2015.

[4] Mark Zaydman, Marina Kasimova, Kelli McFarland, Zachary Beller, Panpan Hou, Holly E Kinser, Hongwu Liang, Guohui Zhang, Jingyi Shi, Mounir Tarek, Jianmin Cui. Domain & ndash; domain interactions determine the gating, permeation, pharmacology, and subunit modulation of the IKs ion channel. eLife, 3: e03606, 2014.

China IT News APP

Download China IT News APP

Please rate this news

The average score will be displayed after you score.

Post comment

Do not see clearly? Click for a new code.

User comments