Drug Development of GLP-1 receptor agonists

当出现2型糖尿病时，胰腺中制造和分泌胰岛素的β细胞不会像非糖尿病受试者那样做出反应。具体地说，患有2型糖尿病的受试者对葡萄糖的反应是第一时相迟钝甚至完全丧失，第二时相胰岛素释放严重迟钝。与此同时，尽管目前有所有治疗糖尿病的方法，但随着时间的推移，β细胞功能继续恶化。使用英国前瞻性糖尿病研究(9月1日)的数据。1998)这一点被更有力地带回了家。尽管对参与研究的患者进行了持续的监测，但由于β细胞功能下降，即使进行强化治疗，正常血糖也无法维持。我们研究GLP-1已经有一段时间了，GLP-1是一种自然产生的多肽，通过肠道产生并释放出来，对食物做出反应。释放的量取决于摄入的葡萄糖和脂肪的量。血浆GLP-1水平升高后，GLP-1与β细胞上的GLP-1受体(GLP-1R)结合，并因cAMP的产生而增加PKA活性。在PKA活性增加的下游，胰岛素的释放也以葡萄糖依赖的方式发生，血糖因膳食而上升。最终结果是血糖恢复到基线水平。因此，GLP-1类似物和GLP-1R激动剂作为治疗2型糖尿病的药物正在紧张的研究中。一种自然产生的GLP-1R激动剂exendin-4现在可以用于治疗。它与GLP-1R结合的亲和力是天然GLP-1的十倍。它含有许多与GLP-1同源的氨基酸，有趣的是，它还含有一个独特的9个氨基酸的C末端。我们通过将C末端加入到GLP-1的C末端时对GLP-1的影响以及从exendin-4中依次删除三个三联体中的9个氨基酸的方法，广泛地研究了C末端的功能意义。它与GLP-1的结合提高了它与自身受体的结合亲和力，而从exendin-4中删除它则取消了它与GLP-1R亲和力的增加。因此，对于exendin-4到GLP-1R的超级激动剂性质来说，9个氨基酸的C末端是必需的。最近，我们一直在研究一种GLP-1融合蛋白，将GLP-1与人转铁蛋白融合，从而将其体外半衰期延长几天。融合蛋白是在酵母中通过重组技术合成的，并分泌到从中提取它的酵母发酵液中。使用GLP-1本身的一个主要障碍是它的半衰期只有几分钟，因为它的N端的前两个氨基酸被二肽基肽酶1V迅速灭活。我们对融合蛋白中的GLP-1进行了修饰，使其不再是二肽基肽酶1V的底物。我们已经在小动物和大动物身上进行了体内工作，以描绘融合蛋白的动力学和给药时间表，FDA已经批准了在人类身上的给药时间表。此外，我们现在已经将exendin-4融合到转铁蛋白中，并发现与单独使用exendin-4相比，啮齿动物中没有失去exendin-4的活性。另一种促胰岛素多肽GIP也会在进食后从肠道释放，但根据BLSA的数据，一旦血糖超过126 mg/dl，胰腺的β细胞似乎就会抵抗它的影响。此外，使用外源性GIP的人体体内实验方案表明，β细胞不会因外源性人GIP而增加胰岛素分泌。GIP也容易被二肽基肽酶1V迅速分解。因此，我们目前正在2型糖尿病患者身上测试抗二肽基肽酶1V的GIP多肽是否能成功地增加胰岛β细胞的胰岛素分泌。这项研究涉及24名受试者，他们被给予的多肽浓度最高为20 ng/kg/min。该协议现已完成，数据正在分析中。

Beta cells of the pancreas, which make and secrete insulin, do not respond like those of non-diabetic subjects when type 2 diabetes is present. Specifically, subjects suffering from type 2 diabetes have a blunted or even absolute loss of first phase and a severely blunted second phase insulin release in response to glucose. In conjunction with this, and despite all treatments currently available to treat diabetes, beta cell function continues to deteriorate over time. With the data now available from the United Kingdom Prospective Diabetes Study (Sept. 1998) this point was brought home even more forcefully. Despite continual monitoring of patients enrolled in the study, euglycemia could not be maintained even with intensive therapy, because of declining beta cell function. We have been working for some time with GLP-1, a naturally occurring peptide produced and released from the gut in response to food. The amount released depends on the amount of glucose and fat that has been ingested. After its plasma levels increase, GLP-1 binds to the GLP-1 receptor (GLP-1R) on beta cells, and increases PKA activity because of cAMP generation. Downstream of the increased PKA activity, insulin release occurs also in a glucose-dependent manner, plasma glucose having risen because of the meal. The end result is a restoration of plasma glucose back to baseline. Consequently, GLP-1 analogs and GLP-1R agonists are under intense study as treatments for type 2 diabetes. A naturally occurring GLP-1R agonist, exendin-4, is now available for treatment. It binds to GLP-1R with ten times more avidity than does native GLP-1. It contains many amino acids homologous to GLP-1 and, interestingly, contains a unique 9 amino acid C-terminus. We have extensively investigated the functional significance of the C-terminus by studying the effects on GLP-1 when it is added to the C-terminus of GLP-1 and by deleting the 9 amino acids sequentially and in triads from exendin-4. Its addition to GLP-1 improved its binding affinity to its own receptor and its deletion from exendin-4 abrogated its increased affinity for GLP-1R. Therefore, for superagonist properties of exendin-4 to GLP-1R the 9 amino acid C-terminus is required. More recently, we have been working on a GLP-1 fusion protein whereby GLP-1 is fused to human transferrin, thus extending its half-life, in vitro, by several days. The fusion protein is synthesised in yeast by recombinant technology and secreted into the yeast broth from which it is extracted. A major impediment to using GLP-1 itself is that its half-life is only a few minutes because it is rapidly inactivated by removel of its first two amino acids at its N-terminus by dipeptidyl peptidase 1V. We have modified the GLP-1 in the fusion protein so that it is not a substrate for dipeptidyl peptidase 1V. We have carried out in vivo work in small and large animals to delineate kinetics and dosing schedules of the fusion protein and FDA approval has been granted for a dosing schedule in humans. Additionally, we have now fused exendin-4 to transferrin and found there is no loss of exendin-4 activity in rodents, when compared to exendin-4 alone. Another insulinotropic peptide, GIP, is also released from the gut after eating, but, based on BLSA data, beta cells of the pancreas appear resistant to its effects once blood glucose exceed 126 mg/dl. Additionally, in vivo experimental protocols in humans using exogenous GIP have shown that beta cells do not increase insulin secretion in response to exogenous human GIP. GIP is also subject to rapid breakdown by dipeptidyl peptidase 1V. Therefore, we are currently testing in type 2 diabetic humans if a dipeptidyl peptidase 1V-resistant GIP peptide might be successful at increasing insulin secretion from beta cells. This study involves 24 subjects to which increasing concentrations of the peptide have been given up to a maximum of 20 ng/kg/min. This protocol is now finished and the data is being analyzed.