创「基因剔除鼠」 3杰夺诺贝尔医学奖

Knowing · Post by **Knowing** » 2007-10-08 12:45

小白，这不就是你们用的LV 老鼠么？

今年诺贝尔医学奖获得者八日在斯德哥尔摩揭晓，由美国的卡沛奇、史密西斯与英国的伊凡斯共同获得。三位科学家因为发明「基因标的」技术，并利用这种基因改造技术创造出「基因剔除小鼠」（knockout mice），共同赢得此一殊荣，将平分奖金一千万瑞典克朗（约台币五千万元）。

诺贝尔委员会在颂词中表示，三位科学家「发现如何对老鼠的胚胎干细胞进行基因改造」，让「心血管与神经退化疾病、糖尿病和癌症等五百多种人类疾病，得以复制在实验室老鼠身上」。

三位科学家发明的技术称为「基因标的」（gene targeting），又称「基因剔除」（gene knockout）。科学家利用基因已被改造的「基因剔除小鼠」，来开关老鼠和人类共有的特定基因。诺贝尔委员会指出，基因改造的老鼠提供研究人员一个实验模型，藉以对阿兹海默症、癌症等疾病产生根本的认识，进而研发新的药物。

委员会在颂词中说：「对老鼠基因标的技术如今已广泛应用在生物制药的各个领域，拓展了人类对胚胎发展、人体生理学、老化和疾病方面许多基因的认识。这种技术在增进对基因功能的了解和造福人类方面的贡献，在未来许多年仍将持续扩大。」

自从第一只基因剔除小鼠在一九八九年宣布制成，迄今已有一万多组老鼠基因被标定剔除，差不多是老鼠基因数量的一半。

三位得奖者当中，八二岁的史密西斯是英国出生的美国公民，现在北卡大学研究。他告诉瑞典广播电台，自己「很高兴能与旁人共同得奖」，还说多次获提名皆未得奖，差点就放弃了希望。

义大利出生、现年七岁的卡沛奇则表示，八日一早获得诺贝尔委员会通知时，他还在睡觉，感觉是「神奇的惊喜」，对他的实验室和任教的犹他大学来说，得奖是「莫大的殊荣」。

卡沛奇表示：「希望这些发现未来能够带给人类更好的药物和更健康的生活。」

在英国加地夫大学任教的伊凡斯形容，获奖是对他研究生涯的「最高表扬与肯定」。

六六岁的他打趣道，八日原先计画帮女儿清空房子，现在得爽约了。

诺贝尔医学奖揭开诺贝尔奖的序幕，物理奖获得者将于九日宣布，日是化学奖，一日是文学奖，二日是和平奖，压轴的经济奖将于五日宣布。颁奖典礼将于二月日在，斯德哥尔摩举行。

洛洛 · Post by 洛洛 » 2007-10-08 12:47

义大利出生、现年七岁的卡沛奇则表示

？？这里的数字看得我很糊涂呢。

森林的火焰 · Post by **森林的火焰** » 2007-10-09 9:13

我记得最早有个八卦说：当初业界普遍不认为能用gene targeting技术做胚胎干细胞，虽然在别的生物里已经有很多了。原因是认为效率太低，不可能成功。不过这三人里的某一个觉得，试试又何妨？于是就成功了。
现在几乎每个训练有素的小鼠遗传实验室都可以做了，我自己就做了俩，深受其惠啊。但坏处就是：这年头儿要出cell paper，似乎逃不过要做一两个，这一下子就一两年的时间。

tiffany · Post by **tiffany** » 2007-10-09 10:48

物理奖不是那几个发明磁盘储存技术的？怎么才发给他们呢？我很迷惑的问。

Knowing · Post by **Knowing** » 2007-10-09 12:00

你还没回答呢。你们用的就是这种敲掉的老鼠么？

tiffany · Post by **tiffany** » 2007-10-09 12:16

准确的说，是用他们的技术发展出来的很多支老鼠中的一支。
一个lv包包要两万啊？够贵。

森林的火焰 · Post by **森林的火焰** » 2007-10-09 13:45

如果自己做的话，六七千应该也够了，特别是自己买血清，只花钱让老鼠技术员做嵌合体。我们的老鼠技术员做一次嵌合体一千五百块。

Jun · Post by **Jun** » 2007-10-09 14:09

发明microarray gene chips的人有没有得到炸弹奖呢？照我看是早该得了。

tiffany · Post by **tiffany** » 2007-10-09 14:11

那个技术太新且不够成熟－－－听说，等等吧，还得。

森林的火焰 · Post by **森林的火焰** » 2007-10-09 14:16

Microarray设想很好，但是实验必须很小心地设计对照组才行。因为microarray会产生海量数据，而且根据实验的具体情况具体控制，产生数据的差别会十分大，不太consistent。而且海量数据的分析，又是另一门学问了。
其实flow cytometry也是很有用的技术，不知道会不会最终得。
白博用的大鼠转基因的多么？似乎比小鼠要难得多。

Jun · Post by **Jun** » 2007-10-09 14:18

可是我不认为要技术成熟才能给奖，关键不在于成熟而是创新，尤其是概念上的创新。Microarray绝对起到革命性的作用，历史地位不比PCR差（我的个人意见）。

tiffany · Post by **tiffany** » 2007-10-09 14:25

创新idea比较容易，工作比较难

silkworm · Post by **silkworm** » 2007-10-09 14:46

炸弹奖还是很重视impact吧。microarray的技术目前产生了海量的数据，但是有重大意义的结果还不算显著。

森林的火焰 · Post by **森林的火焰** » 2007-10-09 14:52

就impact来讲，microarray比PCR要差远了。现在普通的分子生物学实验室离了ＰＣＲ就不用转了，不碰microarray的大有人在。而且microarray的idea也是一点点积累来的，不是横空出世的。如果microarray能够consistent一些，未始不会得到更广泛的应用。

Jun · Post by **Jun** » 2007-10-09 14:53

Hmm. Maybe I'm totally wrong about this, but I think the impact of microarray technology is transformative and revolutionary in genetic medical research. It has made gene-mining exponentially cheaper and faster. Without gene chips there is NO WAY we could have identified so many genes linked to this disease and that disease. For neurological disorders alone the impact seems ... massive.
在动物身上那是不需要用，但是要在一群有同样疾病的人身上找出他们的共同基因，就那么几个，轰隆一下子大海捞针，没有gene chips是绝对不行的。拿特定基因的probe去一个一个试那得猴年马月？

森林的火焰 · Post by **森林的火焰** » 2007-10-09 15:19

所有人的基因基本构成都是一样的，但是基因和基因之间的差别有很多，这叫基因的多样性。还有使基因失去功能所以产生病变的，几个碱基甚至一个碱基的不同。目前来说，对于人类主要基因多样性的了解和哪些突变是致病的，哪些是普通的多样性，仍然缺乏基本的大图。我不知道Jun说的”找出病人的共同基因“是要找这个致病的，还是致病的引起的一连串异常表达的基因。如果是要找致病的，那它必须是凑巧因为表达量高低而不是编码对错而导致疾病；如果是要分辨对错的，它找不出来。遗憾的是，大部分遗传病是因为编码出错导致的。
Microarray只能检测mRNA的丰度，不能检测mRNA是否有正确的编码，也不能检测最后的蛋白产物是否正确。就是说，在整个基因编译的过程中，microarray只提供其中一步的片面信息。而这些还只是关于能表达的基因。还有很多基因是不编码蛋白的，或者只是调节序列。它们出错了，后果一样很严重。但是到底应该怎样检测，没有定规。
就先天性心脏病来说，目前的手段是：收集一些先心病人的病例（最好能上千，起码也要上百），取得他们的组织，猜几个目标基因，挨个儿测序（测到什么程度就看有多少钱和人手了），最后取得的一些信息，拿去做报告基因，报告基因如果有情况，就做老鼠模型；老鼠模型如果能重现先天性心脏病，那简直太好了，cell paper。到这个地步，才能有结论说这个位置的这个点突变或者缺失能导致心脏病。如果来个有家族病史的孕妇要求作检查，就把几个可能的位点都ＰＣＲ出来，测序。靠gene chip，是查不出来的。
Genechip的海量数据分析综合，简直是另外一个新领域。很多人在搞这些方面，试图从基因的global regulation上得到一些结论。但是迄今为止没有什么定论。

Jun · Post by **Jun** » 2007-10-09 15:28

我见到最多的研究是：取一个致病可能性高的基因（例如serotonin transporter），从图书馆里拿一大把SNP变异，用来对照一群得病（例如忧郁症）的人的这个基因，map out statistical association, 就是说，得XX病的人有超乎寻常的高发某种特定的基因组合，而健康人里这种组合就特别少，其他组合多。。这个现在很多人都在搞这个。最近NIH砸进去大笔钱要挖出一系列病的相关基因。I don't know exactly how it's conducted but it's based on DNA microarray assays like Bead Array.

我一直想找几篇深入浅出的解释这方面统计原理的review article。有人能帮帮忙吗？猫咪头？我的统计实在太烂了。。。

森林的火焰 · Post by **森林的火焰** » 2007-10-09 15:40

写grant的时候肯定这么写啦，ＮＩＨ批grant的时候也是这么批。但是从相关到证实，特别是这种群体性的相关，路还长着呢。

照我看，得到的数据不能说没有有用的，但恐怕会被淹没在polymorphism的海洋里。

Jun · Post by **Jun** » 2007-10-09 16:50

I completely disagree.

gigi · Post by **gigi** » 2007-10-09 17:03

Jun, review À´ÁË.

1: Am J Hum Genet. 2007 Aug;81(2):199-207. Epub 2007 Jun 29.

New perspectives for the elucidation of genetic disorders.

Ropers HH.

Max Planck Institute for Molecular Genetics, Berlin, Germany.
ropers@molgen.mpg.de

For almost 15 years, genome research has focused on the search for major risk factors in common diseases, with disappointing results. Only recently, whole-genome association studies have begun to deliver because of the introduction of high-density single-nucleotide-polymorphism arrays and massive enlargement of cohort sizes, but most of the risk factors detected account for
only a small proportion of the total genetic risk, and their diagnostic value is negligible. There is reason to believe that the complexity of many "multifactorial" disorders is primarily due to genetic heterogeneity, with defects of different genes causing the same disease. Moreover, de novo copy-number variation has been identified as a major cause of mental retardation and other complex disorders, suggesting that new mutations are an important, previously overlooked factor in the etiology of complex diseases. These observations support the notion that research into the previously neglected monogenic disorders should become a priority of genome research. Because of the introduction of novel high-throughput, low-cost sequencing methods, sequencing
and genotyping will soon converge, with far-reaching implications for the elucidation of genetic disease and health care.

2: PLoS Genet. 2006 Oct 6;2(10):e150.

Application of genome-wide single nucleotide polymorphism typing: simple
association and beyond.

Gibbs JR, Singleton A.

Computational Biology Core, National Institute on Aging, National Institutes of Health, Porter Neuroscience Research Center, Bethesda, Maryland, United States of America.

The International HapMap Project and the arrival of technologies that type more than 100,000 SNPs in a single experiment have made genome-wide single nucleotide polymorphism (GW-SNP) assay a realistic endeavor. This has sparked considerable
debate regarding the promise of GW-SNP typing to identify genetic association in disease. As has already been shown, this approach has the potential to localize common genetic variation underlying disease risk. The data provided from this technology also lends itself to several other lines of investigation; autozygosity mapping in consanguineous families and outbred populations, direct detection of structural variation, admixture analysis, and other population genetic approaches. In this review we will discuss the potential uses and practical application of GW-SNP typing including those above and beyond simple association testing.

Jun · Post by **Jun** » 2007-10-09 17:05

多谢多谢！

orangetabby · Post by **orangetabby** » 2007-10-09 19:28

Without a disease family, it is extremely hard to map out a disease gene since there could be multiple genes at play. It's doable but very very hard.

I am not a believer of micro array. One big disadvantage of miroarray is that it only takes account of mRNA expression.

However, mRNA stability varies a lot among different genes. Therefore, high mRNA expression doesn't mean high protein expression. It's not unusual to see high mRNA expression but no protein expression.

orangetabby · Post by **orangetabby** » 2007-10-09 19:36

"map out statistical association of a region" is realtively easy. However, I wonder if there is a single disease gene discovered using this technique at all.

gigi · Post by **gigi** » 2007-10-09 20:20

Nowadays SNP arrays which detect single nucleotide polymorphism at the DNA level on genome (instead of the old fashioned mRNA expression arrarys) are used widely in disease association studies. And there are quite a few successful studies using SNP arrays.

The Wellcome Trust Case Control Consortium (WTCCC) is the leading institute in this field. You can go to their website to explore
http://www.wtccc.org.uk/info/070606.shtml

And here is the link to their publications
http://www.wtccc.org.uk/info/publicatio ... ions.shtml

Jun · Post by **Jun** » 2007-10-09 21:02

You don't have to totally shoot in the dark. In the few neuropsychiatric genetic studies that I have seen, the researchers started with a limited number of candidate genes strongly suspected to be involved in a particular disease phenotype. For example, the search for depression-linked genetic defects focused largely on serotonin receptor genes and transporter genes. Many of the serotonin receptor genes have not panned out, but a couple transporter genes actually did, which proves this approach works. A lot of research has followed the same path. If I remember correctly, the particular polymorphism associated with a dramatic increase in major depression episode --- coupled with traumatic life event --- that was discovered a few years ago, was exactly done in this fashion, because serotonin reuptake proteins and receptors were tested as candidate/suspect genes. The two recent examples I cited in the suicide thread, on the genetic polymorphisms linked to SSRI-induced suicidal ideation, were discovered by the same approach. The study published in American Journal of Psychiatry is especially significant, because one of the two SNPs has an odds ratio of 8 (!) in its association with the phenotype of SSRI-induced suicidal thoughts. Moreover, a company called Neuromark has already signed contracts with NIMH to market, ie, sell, the two genetic markers associated with SSRI-induced suicidal risks. This is no longer bench/lab science but clinical applications we are talking about.

There have been no less than two dozens of schizophrenia-associated genes identified so far. The difficulty is that they all seem to be NOT the overwhelming determining factor, but rather interact with environmental or other triggers or factors. That does NOT mean, however, that their discovery is too preliminary to be relevant. Because the mere identification of these genes confirm certain receptors' involvement in the disease pathology, while excluding other receptors suspected. Then researchers can further look into these particular receptors and try to tinker with their activation/inactivation etc. A lot of pathways and target receptors for new drugs are being tested because the SNP studies have zeroed in on these targets. For example, schizophrenia has always been blamed on dopamine neurotransmission, but now we know glutamate-activated pathways are more intimately involved or downstream to the old dopamine/norepinephrine cascade. And the identification of association with glutamatergic receptor genes in these patients confirm that we are on the right track to continue to beat the crap out of certain glutamate receptors.

A lot of research money, private and public, and manpower have been poured into the massive search for the mutations associated with obesity using similar DNA microarray method. The problem is the vast variation in phenotypes and physiological pathways. Once the phenotypes and metabolic pathways are better carved out, it is only a matter of time the many genetic variations will be implicated in specific clinical manifestation.

I'm not particularly up to date on genetic clinical research outside of neuropsychiatry, but I'm sure there is a lot of candidate genes being washed and sifted through the high density gene chips every day.

Even mRNA microarrays are vastly efficient to help clarify how things work in a physiological sense. I was once peripherally involved in a clinical study on an experimental drug to treat lupus and myositis. The basic research people convinced the medical lead to take tissue samples from patients' skin or muscle lesions and test mRNA expression. They found association between the immunosuppressive drug and the suppression of inflammatory chemicals in these tissues (but not in blood because they are too diluted). So even though the clinical findings were inconclusive because the sample size was too small and the effect size was also modest, the expression results convinced the company to continue invest in the drug's development, because we saw it did something. And the mRNA expression suppression helped guide more precise study design in the next, and the next, clinical studies. This is real clinical impact, not theoretical stuff.

橙 · Post by 橙 » 2007-10-09 22:30

Array太不准确啦太不准确啦！

single gene做PCR，能出来十几倍的变化，用array连两三倍都不一定能保证。我还敢相信那些变了一倍半的么？

还有，array做那种模型，就是找什么基因，不管因果关系，跟什么什么有关，是可行的，但是只有那个是绝对不够的，一定要用传统方法在单个基因上检测，要做动物模型，当然microRNA也行。发表的文章当然都是说传统方法跟array结果一致，做的时候有多一致可难说的很。一般来说，就mRNA 来说，反转录然后cDNA跑PCR，结果几乎永远可重复，可是如果用array，1千个gene里头能有一半能 consistant 就不错了。当然这也跟你看的什么变化有关，要是把RNA聚合酶2给抑制了估计出来的结果能每一次都象另一次的孪生姐妹，可是现在那些明显的早都看的差不多了，谁还会有那个运气呢。

DNA array就更惨了。DNA array一般是chip on ChIP。是看的某种蛋白对很多基因的影响，用的方法是先用这个蛋白的抗体把跟她有联系的DNA沾上，沉淀下来，然后array这些DNA。这个沾下来的DNA的量跟蛋白在细胞里的多少有很大关系，跟抗体的质量和针对的位点也有很大关系。现在做出来的chip on ChIP还是以组蛋白Pol II之类为主，组蛋白在细胞里那是什么量啊！而且就算组蛋白沾下来的，还要经过一步扩增。我的妈呀！扩增完了就全变了！

反正，就眼下来看，除非histone 和polII这样的，不然只有array的结果是不行的。虽然人人都说要走genome-wide道路，但是是不是一定要array也难说的很。新技术也在出呢，array的缺点太明显，过几年也许就淘汰了或者边缘化了。

gigi · Post by **gigi** » 2007-10-10 4:44

¿´À´³È²»Ì«ÁË½âSNP array¹þ. Microarray µÄ¼¼Êõ(ºÍºÜ¶àÆäËüµÄÉúÎï¼¼Êõ)ÈÕÐÂÔÂÒì, Ô¶·ÇÎÒÃÇËÄÎåÄêÇ°ËùÓÃµÄÁË. SNP array Ì½²âµÄÊÇDNA Æ¬¶ÎÉÏµ¥¸öÎ»µãµÄ¼î»ùÍ»±ä,½á¹ûÊÇ0 »ò1, ¶ø²»ÊÇÁ¬ÐøÊýÖµ. ÕâÀïÓÐÒ»¸ö³§¼ÒµÄÁ´½Ó, Äã¿ÉÒÔÈ¥¿´¿´µ½µ×ÊÇÔõÃ´»ØÊÂ.
http://www.affymetrix.com/technology/index.affx

²»¹ýÏÖÔÚ DNA sequencing µÄ¼¼ÊõÓÖÔÚ´ó·¢Õ¹ÖÐ, ¹ý¼¸Äêmicroarray ÓÖ¸Ã±»ÌÔÌÁË, Ö±½Ó²âÐòÁË.

橙 · Post by 橙 » 2007-10-10 5:57

qiqi同学难道就是在那个传说中的地方，从8个也不是10个病人身上发现的某某位点上的某个SNP决定某药对某肺癌是不是有用？

gigi · Post by **gigi** » 2007-10-10 6:37

ÄÇÊÇÊ²Ã´µØ·½? ÔÙËµÊ®¸ö°Ë¸ö²¡ÈËÑù±¾Á¿Ò²Ì«Ð¡ÁË,ÕâÑù×÷³öµÄÍ³¼ÆÊý¾Ýsignificant Âð?

橙 · Post by 橙 » 2007-10-10 6:45

affymetrix的rep讲的故事末！

Jun · Post by **Jun** » 2007-10-10 7:04

From: New England Journal of Medicine, August 2, 2007

[quote]Genomewide Association Analysis of Coronary Artery Disease

Nilesh J. Samani, F.Med.Sci., Jeanette Erdmann, Ph.D., Alistair S. Hall, F.R.C.P., Christian Hengstenberg, M.D., et al.

ABSTRACT

Background: Modern genotyping platforms permit a systematic search for inherited components of complex diseases. We performed a joint analysis of two genomewide association studies of coronary artery disease.

Methods: We first identified chromosomal loci that were strongly associated with coronary artery disease in the Wellcome Trust Case Control Consortium (WTCCC) study (which involved 1926 case subjects with coronary artery disease and 2938 controls) and looked for replication in the German MI [Myocardial Infarction] Family Study (which involved 875 case subjects with myocardial infarction and 1644 controls). Data on other single-nucleotide polymorphisms (SNPs) that were significantly associated with coronary artery disease in either study (P<0.001) were then combined to identify additional loci with a high probability of true association. Genotyping in both studies was performed with the use of the GeneChip Human Mapping 500K Array Set (Affymetrix).

Results: Of thousands of chromosomal loci studied, the same locus had the strongest association with coronary artery disease in both the WTCCC and the German studies: chromosome 9p21.3 (SNP, rs1333049) (P=1.80x10

Knowing · Post by **Knowing** » 2007-10-13 9:16

昏倒，戈尔居然得了和平奖！

Jun · Post by **Jun** » 2007-10-13 14:54

文学奖给了Doris Lessing，她一点也不意外，记者问她感想如何，老太太恶狠狠地答：我都八十八岁了，大概他们怕再不给我就没机会了。