Vallen E A, Eldridge K A, Culp L A
Department of Molecular Biology and Microbiology, School of Medicine, Case Western Reserve University, Cleveland, Ohio 44106.
J Cell Physiol. 1988 May;135(2):200-12. doi: 10.1002/jcp.1041350207.
Tissue culture substratum adhesion sites from EGTA-detached Platt human neuroblastoma cells were extracted with a buffer containing ocytlglucoside, NaCl, guanidine hydrochloride, and a variety of protease inhibitors, an extraction which resulted in quantitative solubilization of the 35SO4 = -radiolabeled proteoglycans and 3H-leucine-radiolabeled proteins. Of the sulfate-radiolabeled material, the vast majority was heparan sulfate proteoglycan (Kav = 0.15 on Sepharose C14B columns) and the remainder was chondroitin sulfate chains (no single chains of heparan sulfate were observed). This extract was then fractionated on DEAE-Sephadex columns under two different buffer elution conditions. Under DEAE-I conditions in low ionic strength acetate buffer, two major peaks of 35SO4 = -radiolabeled material (A,B) and a minor peak (C) could be resolved in the NaCl gradient; however, three-fourths of the material required 4 M guanidine hydrochloride to elute it from the column (peak D). Under DEAE-II conditions in acetate buffer supplemented with 8 M urea, the vast majority of the proteoglycan material could be eluted in the NaCl gradient as peak AB. Peak D material was shown to contain aggregated proteoglycan, along with nonproteoglycan protein, which high concentrations of urea or guanidine could dissociate, but not nonionic or zwitterionic detergents. Three different affinity chromatography systems were used to further characterize these components. Approximately 60% of peak A heparan sulfate proteoglycan from DEAE-I binds to the hydrophobic matrix, octyl-Sepharose, while 80% of the proteoglycan in DEAE-I peak D binds to this hydrophobic column. A sizable fraction of peak A proteoglycan fails to bind to plasma fibronectin but does bind to platelet factor-4 affinity columns. In contrast, peak AB proteoglycan from DEAE-II columns yields a much higher proportion of molecules which do bind to fibronectin. To examine the basis for these differences in affinity binding, nonproteoglycan protein from these adhesion sites was mixed with peak AB proteoglycan prior to affinity chromatography; proteoglycan binding to fibronectin decreased markedly while binding to platelet factor-4 was unaffected. This modulating activity involves the binding of nonproteoglycan protein in adhesion site extracts to both fibronectin on the column, as well as to heparan sulfate proteoglycan itself, and it could not be mimicked by a number of known proteins in adhesion site extracts or several other proteins. These results demonstrate selectivity and specificity in this modulation and indicate that a previously unidentified protein(s) is responsible.(ABSTRACT TRUNCATED AT 400 WORDS)
用含有辛基葡糖苷、氯化钠、盐酸胍和多种蛋白酶抑制剂的缓冲液，提取从用乙二醇双（2-氨基乙基醚）四乙酸（EGTA）分离的普拉特人神经母细胞瘤细胞中得到的组织培养基质粘附位点，该提取方法能定量溶解35SO4 = -放射性标记的蛋白聚糖和3H-亮氨酸放射性标记的蛋白质。在硫酸盐放射性标记的物质中，绝大多数是硫酸乙酰肝素蛋白聚糖（在琼脂糖C14B柱上的分配系数Kav = 0.15），其余是硫酸软骨素链（未观察到硫酸乙酰肝素的单链）。然后在两种不同的缓冲液洗脱条件下，将该提取物在二乙氨基乙基（DEAE）-葡聚糖柱上进行分级分离。在低离子强度乙酸盐缓冲液的DEAE-I条件下，在氯化钠梯度中可分辨出35SO4 = -放射性标记物质的两个主要峰（A、B）和一个次要峰（C）；然而，四分之三的物质需要4M盐酸胍才能从柱上洗脱下来（峰D）。在添加了8M尿素的乙酸盐缓冲液的DEAE-II条件下，绝大多数蛋白聚糖物质可在氯化钠梯度中作为峰AB洗脱下来。已证明峰D物质含有聚集的蛋白聚糖以及非蛋白聚糖蛋白，高浓度的尿素或盐酸胍可使其解离，但非离子或两性离子去污剂则不能。使用三种不同的亲和色谱系统进一步表征这些成分。DEAE-I的峰A硫酸乙酰肝素蛋白聚糖中约60%与疏水基质辛基-琼脂糖结合，而DEAE-I峰D中的蛋白聚糖80%与该疏水柱结合。峰A蛋白聚糖中有相当一部分不能与血浆纤连蛋白结合，但能与血小板因子-4亲和柱结合。相比之下，DEAE-II柱的峰AB蛋白聚糖产生的能与纤连蛋白结合的分子比例要高得多。为了研究这些亲和结合差异的基础，在亲和色谱之前，将这些粘附位点的非蛋白聚糖蛋白与峰AB蛋白聚糖混合；蛋白聚糖与纤连蛋白的结合显著降低，而与血小板因子-4的结合不受影响。这种调节活性涉及粘附位点提取物中的非蛋白聚糖蛋白与柱上的纤连蛋白以及硫酸乙酰肝素蛋白聚糖本身的结合，并且粘附位点提取物中的许多已知蛋白或其他几种蛋白都无法模拟这种活性。这些结果证明了这种调节的选择性和特异性，并表明一种以前未鉴定的蛋白质起了作用。（摘要截短至400字）
