高糖对大鼠脂肪细胞胰岛素信号蛋白磷酸化的影响

目的 探讨高浓度葡萄糖（高糖）对原代培养大鼠脂肪细胞的葡萄糖转运活动、胰岛素信号蛋白磷酸化及表达的影响。方法 分离的大鼠脂肪细胞在5，10，15和25mmol/L葡萄糖中孵育24h，然后测定：糖转运活动；胰岛素受体（IR）、胰岛素受体底物（IRS）1、2及蛋白激酶B（PKB）的磷酸化；IRS1，IRS2，肌醇磷脂－3－激酶85亚单位（p85)和PKB的蛋白表达。结果 高糖抑制了这些细胞的葡萄糖转运活动，削弱了IR、IRS1的酪氨酸磷酸化及PKB的丝氨酸磷酸化；下调IRS1而上调IRS2蛋白表达。结论 高糖能抑制脂肪细胞的糖转运活动，诱导胰岛素抵抗。其作用机制与影响胰岛素信号蛋白多部位的磷酸化及蛋白表达有关。     

慢性胰岛素刺激对胰岛素受体后不同信号转导途径的下降调节

目的　研究胰岛素慢性刺激对人肝癌细胞株 (HepG2 )胰岛素受体后不同信号转导途径的影响。方法　HepG2 细胞在无血清条件下与不同浓度的胰岛素 ( 0～ 10 0nmol/L)温育 16h ,然后用 10 0nmol/L胰岛素急性刺激 1min。这些细胞的溶解物中的胰岛素受体 β亚单位 (IRβ) ,胰岛素受体底物 (IRS) 1,IRS 2 ,磷酯酰肌醇 3激酶 (PI3K)的调节亚单位P85 ,有丝分裂原激活蛋白激酶 (MAPK)的蛋白表达水平和MAPK的磷酸化水平通过Western免疫印迹法测定 ,IRβ、IRS 1/ 2的蛋白磷酸化水平以及IRS 1/ 2与P85的结合反应用特异性抗体的免疫沉淀法。结果　胰岛素 1min急性刺激能迅速导致IRβ、IRS 1、IRS 2的酪氨酸磷酸化和MAPK的磷酸化 ,以及IRS 1( 2 )与P85的相互作用而激活PI3K。高浓度胰岛素慢性刺激显著降低了IRβ、IRS 1和IRS 2的酪氨酸磷酸化。细胞用 10 0nmol/L胰岛素预温育 16h后 ,IRβ、IRS 1和IRS 2的磷酸化水平降至最低值 ,分别为对照水平的 2 2 .2 % (P <0 .0 1)、10 .9% (P <0 .0 1)和 2 2 .0 % (P<0 .0 1) ,与IRS的磷酸化变化相平行 ,IRS 1和IRS 2与PI3K的相互作用分别降低至对照水平的 3 4.3 %(P <0 .0 1)和 3 0 .0 % (P <0 .0 1) ,MAPK的磷酸化水平降低至对照水平的 16.4% (P <0 .0 1)。IRβ的蛋白表达水平     

脂联素对高糖环境下胰岛细胞功能及腺苷酸活化激酶和胰岛素受体底物-1的影响

目的研究脂联素对高糖环境下体外胰岛细胞分泌胰岛素的影响、脂联素对胰岛素抵抗的作用,并观察脂联素对胰岛细胞腺苷酸活化蛋白激酶(AMPK)和胰岛素受体底物-1(IRS-1)酪氨酸磷酸化表达的影响。方法分为四组:对照组、对照组+脂联素、高糖组、高糖+脂联素组,分别干预24 h,48 h,72 h后用酶联免疫法(ELISA)测定上清液中胰岛素的含量;建立胰岛素抵抗(IR)细胞模型,脂联素作用下测定葡萄糖含量,细胞内糖原含量的变化以及Western印迹法检测AMPK,IRS-1酪氨酸磷酸化的表达。结果①经脂联素处理后的胰岛细胞,在高糖(25.6 mmol/L)培养48⁷2 h,其胰岛素分泌持续增加(P<0.05);②IR模型组培养液中葡萄糖含量明显高于正常对照组,细胞内糖原含量显著减少,加入脂联素组中培养液中葡萄糖含量较IR模型组显著降低,细胞内糖原含量升高(P<0.01);与对照组相比,IR模型组总AMPK蛋白水平较低(P<0.05),IRS-1酪氨酸磷酸化水平较低(P<0.05);脂联素组总AMPK升高(P<0.05),IRS-1酪氨酸磷酸化升高(P<0.05)。结论在高糖环境下,一定浓度的脂联素可以在体外促进胰岛细胞的分泌和释放;脂联素通过上调AMPK,IRS-1酪氨酸磷酸化的表达,改善胰岛素抵抗。     

软脂酸诱导HepG2细胞胰岛素抵抗及花生四烯酸防治作用的机制研究

目的 探讨高浓度软脂酸(PA)诱导HepG2细胞胰岛素抵抗(IR)的机制及花生四烯酸(AA)对IR的防治作用。方法 (1)用高浓度软脂酸(PA)或10^-7mol／L高胰岛素(HI)培养HepG2细胞建立具有IR的细胞模型，测定培养液中葡萄糖含量及细胞内糖原含量作为鉴定指标；(2)用Western blot检测胞内糖原合酶(GS)和蛋白激酶B(PKB)蛋白水平；(3)用磷脂酰肌醇3激酶(P13K)抑制剂Wortmannin(WT)探讨其对胰岛素信号通路的影响；(4)观察AA是否对PA引起的IR有防治作用。结果 (1)0．20mmol／L PA或川培养HepG2细胞36h后，培养液中葡萄糖含量极显著增高，细胞内糖原含量极显著减少；(2)高浓度PA使磷酸化的PKB(P-Ser473)蛋白水平显著减少，磷酸化的糖原合酶(P-Ser641 GS)蛋白水平极显著增加；(3)WT使对照组GS活性及胞内糖原含量极显著减少，HI组和PA组胞内糖原含量均无统计学差异，但各实验组PKB活性都极显著减少；(4)PA AA组培养液中葡萄糖含量显著低于PA组，GS和PKB活性及胞内糖原含量显著增加。结论 高浓度PA或HI培养HepG2细胞能够诱导IR，其机制可能是其引起胰岛素信号传递途径中自PKB下游到GS之间的信号通路受阻所致。AA能改善PA引起的IR。     

复方石斛合剂调节胰岛素受体表达促进HepG2细胞糖代谢

目的 探讨复方石斛合剂增加HepG2细胞胰岛素信号促进葡萄糖代谢的机制.方法 以高胰岛素培养HepG2细胞24 h,诱导胰岛素抵抗(IR)状态,继以10％复方石斛合剂的含药血清干预48 h,测定细胞6-磷酸果糖激酶、异柠檬酸脱氢酶的活性,RT-PCR与免疫印迹胰检测岛素受体mRNA与蛋白水平的表达.结果 高浓度胰岛素能降低6-磷酸果糖激酶、异柠檬酸脱氢酶的活性,降低胰岛素受体的表达;复方石斛合剂的治疗能增加6-磷酸果糖激酶、异柠檬酸脱氢酶的活性(P＜0.05),促进胰岛素受体的转录与翻译水平表达(P＜0.05),逆转高胰岛素的下调影响.结论 高浓度胰岛素可诱发IR,复方石斛合剂能增加HepG2细胞胰岛素受体的表达,提高葡萄糖分解代谢关键酶活性,缓解IR.     

多囊卵巢综合征患者卵巢黄素化颗粒细胞胰岛素受体底物1、2表达及其酪氨酸磷酸化

目的检测多囊卵巢综合征（PCOS）患者卵巢黄素化颗粒细胞胰岛素受体底物（IRS）-1、IRS-2蛋白表达及其酪氨酸磷酸化水平，探讨卵巢局部胰岛素抵抗的分子机制。方法收集行体外受精-胚胎移植治疗的11例PCOS患者（PCOS组）和15例排卵正常的输卵管性不孕患者（对照组）促排卵后卵巢黄素化颗粒细胞，采用放射免疫法检测血清LH、FSH、睾酮及空腹胰岛素（FINS）水平；采用葡萄糖氧化酶法测定空腹血糖（FPG）水平；利用稳态模型（HOMA）计算胰岛素抵抗指数（HOMA—IR）；采用RT—PCR、Western印迹及免疫沉淀法分别检测两组卵巢黄素化颗粒细胞IRS-1和IRS-2 mRNA、蛋白的表达及其酪氨酸磷酸化程度。结果（1）PCOS患者血清LH、LH／FSH、睾酮、FINS及HOMA—IR均明显高于对照组（均P〈0．05）；（2）与对照组比较，PCOS患者卵巢黄素化颗粒细胞IRS-1 mRNA、蛋白的表达显著增加（均P〈0．05），IRS-2 mRNA、蛋白的表达显著降低（均P〈0．05）；（3）PCOS患者卵巢黄素化颗粒细胞IRS-1和IRS-2酪氨酸磷酸化程度均较对照组明显降低（均P〈0．05）。结论PCOS患者卵巢局部存在胰岛素抵抗，其原因可能与IRS-1和IRS-2蛋白表达及其酪氨酸磷酸化异常有关。     

饮酒对大鼠骨骼肌胰岛素受体及其底物表达和磷酸化的影响

观察摄入不同剂量的酒精5个月后,大鼠骨骼肌胰岛素刺激后葡萄糖摄取能力和胰岛素受体(IR)、IR底物(IRS)1及IRS-2的表达及胰岛素刺激后酪氨酸磷酸化水平的变化,发现饮酒可降低骨骼肌胰岛素刺激后糖摄取,同时伴有IR、IRS-1和IRS-2表达及酪氨酸磷酸化水平代偿性上调。     

肿瘤坏死因子α和白细胞介素6致脂肪细胞胰岛素抵抗的不同机制

细胞介素6(IL-6)和肿瘤坏死因子α(TNF-α)分别作用于3T3-L1细胞6h和72h,观察对葡萄糖摄取和胰岛素信号转导的影响。两者作用72h抑制胰岛素刺激的葡萄糖摄取、胰岛素受体底物1 (IRS-1)酪氨酸及蛋白激酶B(PKB)磷酸化,并使IRS-1表达下降。TNF-α作用6h增加IRS-1丝氨酸307磷酸化。提示IL-6和TNF-α致胰岛素抵抗的作用机制不尽相同。     

胰岛素信号传导及其调节机制

1 胰岛素信号传导通路 胰岛素信号传导可大致分为几个步骤：(1)胰岛素首先与细胞表面胰岛素受体(IR)结合，激活其β亚基的酪氨酸蛋白激酶(protein Tyrosine kinase，PTK)。(2)PTK磷酸化IRS而使之激活，导致受体本身磷酸化和几种底物蛋白磷酸化。(3)IR底物主要包括不同类型的胰岛素受体底物(IRS)，IRS作为一种船坞蛋白(docking protein)，     

吡格列酮减轻肿瘤坏死因子α所致3T3-L1脂肪细胞胰岛素抵抗

目的观察吡格列酮对TNF-α介导的胰岛素抵抗和胰岛素信号通路的影响方法经或未经吡格列酮预处理的3T3-L1细胞与TNF-α作用24h后,分别与对照组比较细胞对胰岛素刺激的葡萄糖摄取,IRS-1及其酪氨酸磷酸化以及PKB,PKCλ及其磷酸化的变化。结果TNF-α抑制胰岛素刺激的葡萄糖摄取以及IRS-1酪氨酸及PKB磷酸化,并使IRS-1蛋白水平下降,对PKC-λ磷酸化无影响。吡格列酮预处理可以逆转TNF-α导致的胰岛素抵抗,部分恢复IRS-1蛋白水平,增强胰岛素刺激的IRS-1酪氨酸、PKB磷酸化及PKC-λ磷酸化。结论TNF-α导致胰岛素抵抗与IRS-1酪氨酸磷酸化水平下降密切相关,吡格列酮可以逆转TNF-α的上述作用。     

Insulin resistance induced by high glucose and high insulin precedes insulin receptor substrate 1 protein depletion in human adipocytes

The aim of this study was to investigate whether high glucose and/or high insulin produces cellular insulin resistance in human adipocytes and, if so, to evaluate the time course and content of key proteins in the insulin signaling pathway. Subcutaneous fat biopsies were taken from 27 nondiabetic subjects. Insulin action in vitro was studied by measurement of glucose uptake after incubation at a physiologic glucose level (6 mmol/L) for 24 hours or with the last 2, 6, or 24 hours at a high glucose level (20 mmol/L) with or without high insulin (10(4)microU/mL). High glucose alone for 24 hours produced a small but significant impairment (by approximately 20%, P < .05) of insulin's effect to stimulate glucose transport, whereas nonstimulated glucose uptake was left intact. In contrast, the combination of high glucose and high insulin for 6 hours or more reduced basal glucose uptake by approximately 40% (P < .05). In addition, insulin-stimulated glucose uptake capacity was reduced by approximately 40% already after 2 hours (P < .05) and reached a maximal decline (by approximately 50%, P < .05) after a 6-hour culture in high glucose and high insulin. Treatment with high glucose and high insulin in combination for at least 6 hours reduced cellular insulin receptor substrate (IRS)-1, but not IRS-2, protein content by approximately 45% or more (P < .05). Moreover, after 24 hours, the ability of insulin to activate protein kinase B (ie, the phosphorylated protein kinase B [pPKB]-protein kinase B ratio) was decreased by approximately 50% (P < .05). No significant effects were seen on insulin signaling proteins or glucose transporter 4 after a long-term high-glucose culture. Culture with high insulin alone (and low glucose, 6 mmol/L) decreased basal and insulin-stimulated glucose uptake in conformity with the high-glucose/high-insulin setting. However, IRS-1 protein content remained unchanged. We conclude that, in adipocytes from healthy humans, high insulin alone for 2 hours or more decrease glucose uptake capacity. Likewise, high glucose and high insulin in combination for 2 hours or more decrease glucose uptake to the same extent as when cells were cultured with high insulin alone but, in addition, with a diminishment in IRS-1 protein lagging behind. Thus, IRS-1 depletion appears to be a secondary phenomenon in this model of insulin resistance. High glucose alone induces only a minor insulin resistance in human fat cells.     

Effect of a high glucose diet on insulin binding and insulin action in rat adipocytes

Summary To elucidate the mechanisms whereby changes in dietary composition affect the action of insulin on glucose metabolism, insulin binding and glucose uptake and oxidation have been studied in epididymal fat pad adipocytes from rats fed high glucose diets for 5 and 10 days. After 5 days, insulin binding was increased, due mainly to an increased number of receptors (3.4×10⁵ vs. 2.4×10⁵ sites per cell) in spite of increased plasma insulin levels (3.0±0.2 vs. 2.1±0.1 g/l; p<0.05). The maximal response of glucose oxidation to insulin was increased (925±55 vs. 510±58 n moles/2×10⁵ cells/2h; p<0.01) and the dose-response curve of glucose uptake was shifted to the left. After 10 days, receptor number decreased to the control level and the effect of insulin on glucose uptake and oxidation (% basal) were similar to controls. Thus, in the early stage of high glucose feeding, insulin receptor number, insulin sensitivity of glucose uptake, and insulin responsiveness of glucose oxidation were increased.     

Obesity and insulin secretion in fasting high school students

Summary The relationship between plasma glucose, serum insulin, serum C-peptide and obesity was studied in 320 fasting high school students (13–18 years old), as part of a Busselton population study. For males and females respectively plasma glucose was 4.5±0.4 and 4.4±0.5 mmol/1 (mean±SD), serum insulin 0.51±0.35 and 0.69±0.39 log₁₀ (nmol/lx100), and serum C-peptide 0.48±0.15 and 0.55±0.14 nmol/1. These sex differences were not statistically significant. Plasma glucose correlated with C-peptide (r=0.21, p<0.001) and insulin (r= 0.32, p<0.001), indicating greater secretion where fasting glucose was higher. Obesity, measured as skin fold thickness, was also associated with serum C-peptide (r=0.32, p<0.001) and insulin (r=0.37, p< 0.001).     

Effects of dexamethasone on glucose-induced insulin and proinsulin release in low and high insulin responders

We compared the effects of dexamethasone-induced insulin resistance on B-cell secretory performance in 12 low insulin responders (LIR) and in eight high insulin responders (HIR). A hyperglycemic clamp (120 minutes) was performed before and after the subjects had ingested dexamethasone 3 mg x 2 for 2 1/2 days. Fasting levels of blood glucose increased from 4.60 +/- 0.13 to 5.74 +/- 0.23 mmol/L after dexamethasone in LIR and from 4.37 +/- 0.18 to 5.26 +/- 0.13 mmol/L in HIR. Dexamethasone treatment increased fasting levels of total immunoreactive insulin (IRI), C-peptide, and proinsulin, as well as the proinsulin to IRI ratio to a similar degree in LIR and HIR. The amount of glucose infused to uphold hyperglycemia during the clamp decreased by 54% after dexamethasone in LIR and by 46% in HIR. Mean level of stimulated IRI during the clamp increased after dexamethasone by 43% in LIR and by 53% in HIR. Mean level of stimulated C-peptide increased by 11% (not significant) in LIR and by 24% in HIR. Mean level of stimulated proinsulin increased by 86% in LIR and by 93% in HIR. The effects of dexamethasone on insulin secretion varied among individuals, since steroid treatment failed to affect IRI responses to glucose in two LIR and two HIR. The magnitude of dexamethasone effects on secretion was not correlated to pre-dexamethasone insulin sensitivity as assessed by a somatostatin-insulin-glucose infusion test (SIGIT) or by M/I (glucose infused/insulin level) ratios of the control clamp.(ABSTRACT TRUNCATED AT 250 WORDS)     

Reverse phase high performance liquid chromatographic analysis of circulating insulin in the insulin autoimmune syndrome

Some patients with the insulin autoimmune syndrome have circulating insulin that is heterogeneous. We used reverse phase high performance liquid chromatographic analysis to identify the forms of plasma insulin in patients with this syndrome and compared the results with those in patients with insulin-treated diabetes and patients with hyperinsulinism. Under acidic conditions, free insulin dissociated from insulin antibodies eluted from Bio-Gel P-30 columns as a single peak. When such insulin fractions were applied to reverse phase high performance liquid chromatography, a major insulin peak emerged with the same retention time as standard human insulin in all six patients with the syndrome. In addition, a minor insulin peak was consistently found at relatively high acetonitrile concentrations. However, this hydrophobic insulin also was found in two of four insulin-treated diabetic patients and in one of two hyperinsulinemic patients who did not have insulin antibodies. Preliminary characterization of the variant insulin revealed that it has a molecular size between those of proinsulin and insulin and retains the immunoreactivity of insulin, but not C-peptide. It may be an aggregate of insulin molecule or proinsulin intermediates. Since the variant insulin was not found only in patients with the insulin autoimmune syndrome, it seems unlikely that an altered endogenously produced insulin induces the generation of autoantibodies to insulin in this syndrome.     

Associations between plasma insulin and high blood pressure]

We studied the relationship between plasma insulin level and hypertension in 510 cases with normal glucose tolerance and impaired glucose tolerance. In nonobese group (BMI < 25kg/m2), plasma insulin was higher in those with hypertension than those with normal blood pressure (P < 0.0001). There was no correlation between diastole blood pressure and plasma insulin; multiple regression analysis showed that fasting plasma insulin was significantly associated with systolic blood pressure after controlling age, BMI and plasma glucose level (beta = 0.27, P = 0.0078). The result suggested that age, BMI and plasma insulin level were independent risk factors of hypertension. In obese group (BMI > 25kg/m2), blood pressure was significantly associated with age and BMI, there was no association between blood pressure and plasma insulin level.     

An atypical insulin receptor with high affinity for insulin-like growth factors copurified with placental insulin receptors

Insulin receptors purified from human placenta by sequential affinity chromatography on wheat germ lectin-agarose and insulin coupled to 1,1'-carbonyldiimidazole-activated agarose (CDI-agarose) retained full binding activity but bound a greater than predicted amount of 125I-labeled insulin-like growth factor I (IGF-I). IGF-I and multiplication-stimulating activity (MSA; the rat homologue of IGF-II) were equipotent in displacing either 125I-labeled IGF-I or 125I-labeled MSA from the purified receptors; insulin was 5-15 times more potent. Competitive binding studies indicated that this IGF binding activity could not be explained by cross-reaction with classical insulin receptors or by coelution of IGF-I or IGF-II receptors. Instead, it was due to a minor population of discrete atypical insulin receptors (6-18% total insulin receptors) with moderately high affinity (Kd = 2-4 X 10(-9) M) for IGF-I and MSA. These receptors were not an artifact of insulin-CDI-agarose chromatography, since they were present in wheat germ lectin-agarose-purified preparations and could also be purified from insulin-succinyldiaminodipropylamino-agarose. Affinity labeling with 125I-labeled MSA revealed that these atypical receptors had the same binding subunit (Mr 140,000) as classical insulin and IGF-I receptors. They displayed intermediate reactivity with polyclonal and monoclonal antibodies to the insulin and IGF-I receptors. It is therefore likely that insulin receptors purified by immunoadsorption would also contain atypical insulin receptors. The finding of more than one type of insulin receptor might relate to the slight variations in the cDNA nucleotide sequences and the multiple mRNA species reported for the insulin receptor [Ebina, Y., Ellis, L., Jarnagin, K., Edery, M., Graf, L., Clauser, E., Ou, J.-H., Masiarz, F., Kan, Y. W., Goldfine, I. D., Roth, R. A. & Rutter, W. J. (1985) Cell 40, 747-758].     

In vivo dose response curves of insulin action in heart: anomalous effects at high insulin doses

The euglycaemic hyperinsulaemic clamp technique in conscious unrestrained rats was used to compile insulin dose response curves of glucose metabolism in the heart in vivo. An estimate of heart glucose uptake (Rg') was obtained using [3H]-2-deoxyglucose and glucose disposal was examined by measuring cardiac glycogen content. Elevation of insulin from 29 to 54 mU/l resulted in a significant increase in Rg' in heart from 41 +/- 6 to 77 +/- 4 mumol/100 g/min (P less than 0.01) with no effect on glycogen content. This is consistent with increased glucose oxidation. At 150 mU/l of insulin both Rg' and glycogen synthesis were increased. Glycogen content increased from 18.5 +/- 1.7 mumol/g under basal conditions to 27.9 +/- 1.6 mumol/g with insulin. However, at subsequent insulin doses producing plasma levels exceeding 600 mU/l there was an anomalous reversal of Rg' back to basal levels while glycogen content was significantly elevated (2.4-fold, P less than 0.01). This effect may be related to feedback inhibition of tissue glycogen on glucose transport or to accumulation of tissue metabolites such as glucose-6-phosphate. The dose response curve for insulin stimulated Rg' in heart does not resemble either the whole body glucose utilization curve or that in individual skeletal muscles.     

Tolerance to high and low doses of natural crystalline insulin

Summary In order to examine the induction of an immunological tolerance to insulin, six different doses of once-crystallized insulin were injected i.p. into NMRI mice 3 times a week for 3 months. The doses were: 100 g, 10 g, 1 g, 100 ng, 10 ng and 1 ng per injection. A test immunization of 100 g insulin in complete Freunds adjuvant was given at the end of the immunization period to test the immunologic reactivity. It was found that one dose (1 g) had stimulated antibody production, and that two doses had induced tolerance: a low dose of 100 ng and a high dose of 100 g. In the tolerant animals at 3 and 5 weeks after the test immunization the antibody level was less than 10% of that present in the controls.

---

Toleranz gegen hohe und niedrige Dosen von natürlichem krystallinem Insulin

Zusammenfassung Um die Induktion einer Immun Toleranz gegen Insulin zu untersuchen, wurden 6 verschiedene Dosen von einmal kristallisiertem Insulin NMRJ Mäusen dreimal wöchentlich über 3 Monate injiziert. Die Dosen betrugen 100 , 10 , 1 , 100 ng, 10 ng, 1 ng pro Injektion. Eine Test-Immunisierung von 100 ng Insulin in komplettem Freund'schen Adjuvans wurde am Ende der Immunisierungsperiode gegeben, um die immunologische Reaktivität zu prüfen. Es wurde gefunden, daß eine Dosis (1 ng) die Antikörperproduktion stimuliert hatte, und daß zwei Dosen eine Toleranz induziert hatten: Eine niedrige Dosis von 100 ng und eine hohe Dosis von 100ng. Im toleranten Tier betrug 3 und 5 Wochen nach der Testimmunisierung der Antikörperspiegel weniger als 10% derjenigen bei den Kontrollen key-words Immunological Tolerance to Insulin.

---

Tolérance à des doses fortes et faibles d'insuline cristalline naturelle

Résumé Afin d'étudier l'induction d'une tolérance immunologique à l'insuline, six doses différentes d'insuline recristallisée une fois ont été injectées i. p. à des souris NMRI 3 fois par semaine pendant 3 mois. Les doses étaient de 100 g, 10 g, 1 g, 100ng, 10ng et 1 ng par injection. Une immunisation-test de 100 g d'insuline dans de l'adjuvant de Freund complet était administrée à la fin de la période d'immunisation, afin de tester la réactivité immunologique. Il a été trouvé qu'une dose (1 g) avait stimulé la production d'anticorps et que deux doses avaient provoqué la tolérance: une faible dose de 100ng et une forte dose de 100 g. Chez les animaux tolérants au bout de 3 et 5 semaines après l'immunisation-test, le taux d'anticorps représentait moins de 10% de celui des animaux témoins.

---

---

I am very grateful to Miss H. Dorsel and Miss G. Nahler for their good technical assistance.