Cloning and expression of a divergent integrin subunit beta 8.

Rabbit and human cDNA clones have been identified that encode a novel integrin beta subunit. The sequences that encode this subunit, which has been designated as beta 8, were isolated initially from rabbit placental cDNA libraries using an oligonucleotide probe derived from a highly conserved region of integrin beta subunit sequences. The rabbit clone was used to isolate human beta 8 cDNA clones from human placental and MG-63 osteosarcoma cell libraries. The putative beta 8 polypeptides, which comprise 769 and 768 residues in human and rabbit, respectively, show a high degree of inter-species conservation (approximately 90% identity). In contrast, beta 8 is distinct from the other integrin beta subunits. At the amino acid level human beta 8 ranges from 31 to 37% identity with human beta 1-7. The domain structure of beta 8 is typical of the integrin beta subunits. Human beta 8 has a 42-residue N-terminal signal peptide, a large extracellular domain (approximately 639 residues) that contains four cysteine-rich repeats, a transmembrane domain (approximately 30 residues), and a C-terminal cytoplasmic domain (approximately 58 residues). There are several structural features that are unique to the beta 8 polypeptide, as compared with the other integrin beta subunits. Six of the 56 cysteine residues that are conserved within the extracellular domains of beta 1, beta 2, beta 3, beta 5, beta 6, and the beta subunit from Drosophila are absent in the beta 8 polypeptide. Also, the cytoplasmic domain of the beta 8 subunit shares no homology with the cytoplasmic regions of any of the other integrin beta subunits. Northern analysis demonstrated an approximately 8-kilobase beta 8 mRNA in rabbit placenta, kidney, brain, ovary, and uterus. PCR analysis revealed that beta 8 mRNA is also present in several transformed human cell lines. The beta 8 polypeptide has been transiently expressed in 293 human embryonic kidney cells. A polyclonal antipeptide antibody specific for beta 8 and a polyclonal antibody that recognizes alpha v epitopes were used to show that beta 8 can complex with the endogenous alpha v subunit in 293 cells and that the resulting integrin is expressed as a cell surface complex.

Characterization of an anti-p185HER2 monoclonal antibody that stimulates receptor function and inhibits tumor cell growth.

The HER2 protooncogene encodes a growth factor receptor-like transmembrane protein tyrosine kinase (p185HER2) whose ligand remains to be fully characterized. The overexpression of p185HER2 is implicated in aggressive forms of breast and ovarian cancers. The role of p185HER2 in aggressive malignancy, as well as its cell surface localization, makes it an attractive target for therapeutic monoclonal antibodies. In this report we have studied the modulation of p185HER2 function with 2 monoclonal antibodies, termed 4D5 and 6E9, which bind the extracellular domain of p185HER2. 4D5 inhibited proliferation of p185HER2 overexpressing SK-BR-3 human breast carcinoma cells (ED50 of approximately 1 nM) but did not inhibit proliferation of cultured human breast carcinoma MCF7 cells, low expressors of p185HER2. Monoclonal antibody 6E9 does not inhibit the growth of either cell line. Antibody binding studies revealed 2 populations of p185HER2 molecules on SK-BR-3 cells: one of high abundance (approximately 2 x 10(6) sites/cell) recognized by 4D5 (Kd approximately 6 nM) and the other of low abundance (2 x 10(4) sites/cell) recognized by 6E9 (Kd approximately 0.1 nM). 4D5, in an agonistic manner, downregulated SK-BR-3 cell surface p185HER2, was internalized, and stimulated p185HER2 phosphorylation in intact cells. Phosphoamino acid analysis of p185HER2 derived from SK-BR-3 cells incubated with the 4D5 monoclonal antibody demonstrated increased tyrosine, serine and threonine phosphorylation. 4D5, on short term (5 min) exposure to SK-BR-3 cells, stimulated inositol lipid hydrolysis as evidenced by increased intracellular levels of inositol polyphosphates (InsP) and sn-1,2-diacylglycerol (sn-1,2-DAG). On longer (24 h) exposure to the cells, the antibody appeared to downregulate this signalling pathway since the intracellular levels of InsP and sn-1,2-DAG decreased by 30 to 40%. 6E9 did not inhibit SK-BR-3 cell proliferation, downregulate surface p185HER2, stimulate receptor phosphorylation, or stimulate the second messenger pathway. Despite these agonistic properties, 4D5 was an inhibitor of SK-BR-3 cell proliferation at all concentrations tested (0.7 to 70 pM). The data suggest that 4D5 is a partial or weak agonist and thus may inhibit cell proliferation by mimicking ligand-like receptor downregulation.

Kistrin, a polypeptide platelet GPIIb/IIIa receptor antagonist, enhances and sustains coronary arterial thrombolysis with recombinant tissue-type plasminogen activator in a canine preparation.

BACKGROUND: Kistrin is a 68-amino acid polypeptide from the venom of the Malayan pit viper Agkistrodon rhodostoma, which inhibits the platelet GPIIb/IIIa receptor. Its effect on thrombolysis, reocclusion, and bleeding associated with administration of recombinant tissue-type plasminogen activator (rt-PA) was studied in a canine model of coronary artery thrombosis. METHODS AND RESULTS: Coronary patency was monitored for 2 hours by ultrasonic flow probe and repeated coronary angiography. The rt-PA was given as 0.45-mg/kg bolus injections at 15-minute intervals until recanalization or to a maximum of four boluses. Four groups of four or five dogs were studied: a control group that received intravenous heparin (4,000-unit bolus and 1,000 units each hour) and three groups that received heparin and 0.48, 0.24, or 0.12 mg/kg kistrin, administered as a 10% bolus injection and an infusion during a 60-minute period. In the control group, reflow occurred in four of five dogs within 37 +/- 47 minutes but was followed by cyclic reflow and reocclusion. Kistrin at a dose of 0.48 and 0.24 mg/kg reduced the time to reflow to 6 +/- 5 and 10 +/- 3 minutes, respectively, and abolished reocclusion. With 0.12 mg/kg kistrin, reflow occurred in all four animals, within 27 +/- 23 minutes, and reocclusion occurred in two animals. Kistrin induced a dose-related prolongation of the template bleeding time: with 0.48 mg/kg kistrin, the bleeding time was prolonged from 3.8 +/- 1.3 minutes before infusion to 29 +/- 2 minutes during infusion, but it was shortened to 8.3 +/- 2.6 minutes at 90 minutes after the end of infusion. Kistrin also caused a dose-related inhibition of platelet aggregation with ADP and collagen: with 0.48 mg/kg kistrin, platelet aggregation was abolished during the infusion but had partially recovered toward the end of the observation period. Pathological examination of recanalized coronary arterial segments of dogs given 0.48 or 0.24 mg/kg kistrin revealed widely patent arteries with some platelets layered on the damaged intimal surface. CONCLUSIONS: Kistrin increases the rate and extent of thrombolysis with a reduced dose of rt-PA, and it prevents reocclusion. At an effective dose, it is associated with a transient prolongation of the bleeding time and inhibition of platelet aggregation. Kistrin may offer promise as adjunctive treatment to thrombolytic agents in patients with acute myocardial infarction.

Growth hormone stimulates the formation of sn-1,2-diacylglycerol in rat hepatocytes.

The transmembrane signaling events of GH were investigated in the liver, a major target organ of GH action. Recombinant human GH when added to freshly isolated rat hepatocytes rapidly stimulated the production of sn-1,2-diacylglycerol (DAG). The generation of DAG was biphasic with the first transient peak observed at 2 min and the second peak at 15 min (1.2-fold and 1.4-fold over control, respectively). Levels of DAG continued to be elevated above those in control cells at 30 min. The response was dose-dependent with an EC50 of 0.15 nM. Both bovine GH and rat GH, which bind to the rat GH receptor but not to the PRL receptor, also stimulated DAG production. Similarly, human PRL, which binds to the PRL but not GH receptor, stimulated DAG formation to a comparable extent. These results suggest that production of DAG may be an early signaling event mediated by hormone stimulation of both the GH and PRL receptors.

Platelet glycoprotein IIb-IIIa protein antagonists from snake venoms: evidence for a family of platelet-aggregation inhibitors.

The purification, complete amino acid sequence, and biological activity are described for several homologous snake venom proteins that are platelet glycoprotein (GP) IIb-IIIa antagonists and potent inhibitors of platelet aggregation. The primary structures of kistrin (from Agkistrodon rhodostoma), bitan (from Bitis arietans), three isoforms of trigramin (from Trimeresusus gramineus), and an isoform of echistatin (from Echis carinatus) were determined by automated sequence analysis and fast atom bombardment mass spectrometry analysis. Each of the protein in this family, which range from 47 to 83 residues, contains an Arg-Gly-Asp amino acid sequence found in protein ligands that binds to GPIIb-IIIa, a high (17 +/- 1%) cysteine content conserved in the primary sequence, and a homologous N-terminal region absent only in the echistatin isoforms. Each protein directly inhibits the interaction of purified platelet GPIIb-IIIa to immobilized fibrinogen about 100 times more effectively than does the pentapeptide Gly-Arg-Gly-Asp-Ser; IC50 values range from 1.1 to 3.0 nM. The IC50 value for the inhibition of platelet aggregation, using human platelet-rich plasma stimulated with ADP, ranges from 110 to 550 nM for the various proteins, about 1000-fold more potent than Gly-Arg-Gly-Asp-Ser. Kistrin binds reversibly to both resting and ADP-activated human platelets with high affinity (Kd = 10.8 nM and 1.7 nM, respectively) and to purified GPIIb-IIIa with a lower affinity (Kd = approximately 100 nM). Finally, kistrin injected at 1.0 mg/kg into rabbits reversibly inhibits platelet aggregation ex vivo over 30 min without induction of thrombocytopenia. We propose that these proteins are members of a general class of proteins found in the venom of pit vipers that inhibit platelet aggregation by antagonism of the GPIIb-IIIa-fibrinogen interaction and as such serve as potential antithrombotic agents.

Characterization of murine monoclonal antibodies reactive to either the human epidermal growth factor receptor or HER2/neu gene product.

High levels of expression of either the epidermal growth factor receptor or the receptor-like HER2/neu gene product p185HER2 have been observed in a variety of human malignancies. Because of the association of this high level expression with certain human tumors, we have generated a panel of monoclonal antibodies specific for either the epidermal growth factor receptor or p185HER2 to study their structure, function, and antigenic domains in the normal and neoplastic states. We used the epidermoid carcinoma line A431 to generate five monoclonal antibodies which immunoprecipitate the epidermal growth factor receptor. These monoclonal antibodies bind to the extracellular domain of the epidermal growth factor receptor and demonstrate variable effects on epidermal growth factor binding. We used a stably transfected NIH 3T3 cell line expressing the HER2/neu gene to produce and characterize 10 monoclonal antibodies which immunoprecipitate p185HER2. These monoclonal antibodies bind to the extracellular domain of p185HER2 and do not cross-react with the epidermal growth factor receptor. The characteristics and potential applications of these monoclonal antibodies will be discussed.

Expression of recombinant platelet glycoprotein IIbIIIa results in a functional fibrinogen-binding complex.

The ability of cDNAs encoding the human platelet glycoprotein IIbIIIa to be expressed and assembled into a functional integrin receptor was assessed by transient transfection into a human cell line. Transfection of full length cDNAs resulted in synthesis of high levels of integrin subunits which appear to be stable within the cell for several days. Coexpression of both subunits resulted in a proteolytically processed form of GPIIb that associated with GPIIIa as a heterodimeric complex as the cell surface. Transport to the cell surface required association of these subunits with each other or with endogenous integrin subunits. When expressed alone, the GPIIb subunit remained intracellular, while the GPIIIa subunit was found to complex with endogenous proteins and was mobilized to the cell surface. The GPIIbIIIa receptor complex facilitated attachment of cells to known ligands for GPIIbIIIa: fibrinogen, vitronectin, and von Willebrand factor. This adhesion was sensitive to inhibition by the peptide GRGDV and the monoclonal antibody AP2, known inhibitors of platelet aggregation

Epidermal growth factor receptor tyrosine kinase phosphorylation of glucose-6-phosphate dehydrogenase in vitro.

The specific tyrosine phosphorylation of glucose-6-phosphate dehydrogenase (G6PDH) by the epidermal growth factor (EGF) receptor in vitro is demonstrated. The Km values of the substrate G6PDH and of ATP for the receptor tyrosine kinase were ca. 1 and 10 microM, respectively. The rate of phosphorylation was EGF dependent, with a four-fold increase in Vmax in the presence of EGF. The phosphorylation was stimulated maximally by 0.2 microM or greater EGF, with an ED50 of ca. 20 nM which is consistent with the affinity of the solubilized receptor for EGF. Using conditions of 5 microM G6PDH, 100 microM ATP, 5 mM Mg2+, and 1 mM Mn2+, up to 0.3 mol phosphate was incorporated into 1 mol of the 55-kDa subunit of Baker's yeast G6PDH. Tryptic peptide mapping revealed several unique phosphopeptides for both Baker's yeast and bovine adrenal G6PDH. The patterns of phosphopeptides for a given enzyme were identical for basal and EGF-stimulated phosphorylation.

Molecular characteristics of receptors for atrial natriuretic factor.

Specific, high-affinity receptors for atrial natriuretic factor (ANF) have been identified on membranes from a variety of tissues and cultured cells. By affinity labeling procedures, radioactivity from 125I-labeled ANF was specifically incorporated into three different polypeptides of ca. 120,000, 70,000, and 60,000 daltons, which may represent the binding subunits of ANF receptors. These polypeptides were present in varying amounts in different target tissues. In rat adrenal membranes, the 120,000- and 70,000-dalton peptides were specifically labeled whereas in A10 rat smooth muscle cells, only the 60,000-dalton peptide was labeled. Membranes from rat kidney and rabbit aorta contain all three peptides. Gel filtration chromatography of solubilized receptors suggested that intact ANF receptors are large molecular complexes with apparent molecular masses in the range of 250,000-350,000 daltons. The differential labeling pattern observed with the various tissues suggested that there might be at least two different receptors composed of unique ANF-binding polypeptides.

Binding and internalization of atrial natriuretic factor by high-affinity receptors in A10 smooth muscle cells.

A10 smooth muscle cells, derived from embryonic rat thoracic aorta, responded to the atrial natriuretic factor (ANF) with increased levels of cyclic GMP. These cells possess high-affinity (apparent Kd = 50 pM) plasma membrane receptors for ANF. Internalization of ANF at 37 degrees C was indicated by the following: approximately 25% of the 125I-ANF associated with the cells at elevated temperatures could not be dissociated from the surface of the cells, but could be released by permeabilization with saponin, and the amount of nondissociable ANF increased in the presence of chloroquine. In whole cells and in membranes, a single polypeptide of 60,000 Da was specifically labeled by a photoaffinity analog of 125I-ANF, as well as by crosslinking, and an IC50 of 80 pM for inhibition of the labeling by ANF was observed. The ANF receptor in A10 cells was distinguished from that in rabbit aorta by its high affinity for shorter and linear analogs of ANF, as well as by a different photolabeling pattern.

Effects of atrial natriuretic factor on renal function and cyclic GMP production.

Anesthetized beagle dogs received increasing doses of continuous infusions of a 26-amino-acid synthetic atrial natriuretic factor (ANF). Urinary sodium excretion rose in a dose-dependent manner to a maximum level similar to that seen after hydrochlorothiazide administration. Mean arterial blood pressure decreased, but only modestly, and not in a dose-dependent fashion. Dogs chronically retaining NaCl secondary to constriction of the thoracic inferior vena cava showed only modestly enhanced natriuresis when infused with similar levels of ANF. When ANF was infused directly into the renal artery of anesthetized beagles, a dose-dependent natriuresis and calciuresis were observed with maximal fractional sodium excretion averaging approximately 8%. Although glomerular filtration tended to increase, the average dose-related changes were not significant. Cyclic GMP excretion was increased during intra-renal-arterial infusion of ANF. Excretion of cyclic GMP by both the infused and noninfused kidneys was equal, which suggests that urinary cyclic GMP was not nephrogenous but derived from the elevated circulating levels. These and other data from rats dissociate changes in urinary cyclic GMP excretion and sodium excretion.

Identification of a receptor for atrial natriuretic factor in rabbit aorta membranes by affinity cross-linking.

Binding sites in rabbit aorta membranes for atrial natriuretic factor (ANF) have been specifically and covalently labeled by two methods. In the first, the photoreactive analog of ANF, 125I-azidobenzoyl-ANF, was synthesized and used to photoaffinity label ANF receptors. In the second, 125I-ANF was covalently attached to its binding site by treatment of the 125I-ANF-receptor complex with bifunctional cross-linking agents. Analysis of the labeled proteins by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography revealed that by both methods the same three protein bands were labeled. These bands had apparent molecular masses of 60,000, 70,000, and 120,000 daltons. With the photoaffinity label, half-maximal inhibition of labeling of each of these bands was achieved when approximately 200 pM of unlabeled ANF was included in the binding assay. These results suggest that these three different polypeptides are specific components of ANF receptors in rabbit aorta membranes.

Renal and blood pressure responses to synthetic atrial natriuretic factor in spontaneously hypertensive rats.

Atrial natriuretic factor (ANF) is a potent natriuretic and vasorelaxant agent that also stimulates guanosine 3',5'-cyclic monophosphate (cGMP) excretion in normotensive animals. These properties suggest that ANF may be involved in the regulation of blood pressure. To test a pure preparation of ANF in both normotensive and hypertensive animals, a synthetic 26 amino acid peptide (sANF) contained within endogenous rat ANF was infused intravenously into conscious Wistar-Kyoto rats (WKY) and spontaneously hypertensive rats (SHR) at doses from 12 to 190 pmol/minute. Mean arterial pressure fell progressively as doses of sANF were increased until maximum responses of -41 +/- 5 mm Hg and -29 +/- 5 mm Hg were obtained during infusion of 95 pmol/minute sANF in SHR and WKY, respectively. Heart rate was not significantly affected in either group. At sANF doses of 12 to 50 pmol/minute, urinary electrolyte excretion rose in a dose-related fashion and was similar in WKY and SHR. At infusions of 95 to 190 pmol/minute, the diuretic and saluretic responses were diminished in the hypertensive animals. Only the 190 pmol/minute sANF dose significantly enhanced cGMP excretion in SHR (p less than 0.05); however, in WKY urinary cGMP excretion was elevated in a dose-related fashion. At the highest sANF dose, cGMP excretion was approximately 15 times that observed in the pretreatment urine. The differences in the renal and blood pressure responses to sANF in SHR and WKY suggest that the actions of endogenous ANF may be altered in hypertension.

Renal and systemic effects of synthetic atrial natriuretic factor.

A synthetic peptide corresponding to a sequence of 26 amino acids contained in endogenous rat atrial natriuretic factor (ANF), was infused into one renal artery of anesthetized dogs for a comprehensive in vivo evaluation of the renal and systemic effects of pure ANF. The results proved conclusively that ANF acted directly on the kidney since urine volume and fractional excretion of sodium, potassium, chloride and calcium were elevated in a dose-related manner in the ANF-treated kidney, but were not significantly affected in the contralateral saline-infused organ. The maximum effects achieved with the synthetic ANF were higher than any reported following intravenous administration of crude extracts of rat atria and were similar to those produced by thiazide diuretics. In four of the five dogs studied, renal vascular resistance fell progressively as doses of ANF were increased. Glomerular filtration rate was not significantly elevated during ANF infusion, but was correlated with sodium excretion rates. Even though mean arterial pressure was progressively reduced, there was no significant change in heart rate and no stimulation of renin secretion. Arterial cyclic GMP concentration was higher in the basal state and rose more rapidly than did renal venous levels, indicating that increases in circulating concentrations of arterial cyclic GMP originated from an extrarenal source. Dose-related elevations in urinary cyclic GMP excretion could be explained by increased cyclic GMP filtration, by enhanced production in tubular cells, or by renal tubular secretion. Especially in the saline-infused kidney, there was a clear dissociation between excretion of cyclic GMP and fractional sodium excretion. We conclude that the synthetic ANF increased electrolyte excretion via a direct renal action which was not solely dependent upon changes in renal vasculature, renin secretion or cyclic GMP levels.

Pharmacology and receptor binding of atrial natriuretic factor in vascular smooth muscle.

Characterization of the vascular pharmacology and receptor binding of atrial natriuretic factor (ANF) has been achieved utilizing a synthetic peptide which contains the sequence and biological activity of ANF. The synthetic ANF (sANF) relaxes aortic segments contracted by agonists or by low (15 to 20 mM) but not high (greater than or equal to 40 mM) concentrations of K+. The relaxation to sANF is well maintained, reversible, independent of the vascular endothelium and correlated with increases in cyclic GMP (with no change in cyclic AMP). Plasma membranes prepared from rabbit aorta and kidney possess high affinity (Kd = 100 pM) specific binding sites for sANF. An excellent correlation exists between the receptor binding and pharmacology for several synthetic analogs of ANF. The presence of these receptors appears consistent with the activation of particulate (but not soluble) guanylate cyclase by sANF. sANF does exhibit a profound regional vasodilator selectivity which can be explained, in part, by changes in receptor density.

Specific membrane receptors for atrial natriuretic factor in renal and vascular tissues.

Membranes from rabbit aorta and from rabbit and rat kidney cortex possess high-affinity (Kd = 10(-10) M) specific binding sites for atrial natriuretic factor (ANF). Similar high-affinity sites are present in an established cell line from pig kidney, LLC-PK1. Results of fractionation studies indicate that the receptors are localized in the plasma membrane of these tissues. The binding is time-dependent and saturable. An excellent quantitative correlation was found between the affinity of synthetic ANF and analogs of intermediate activity to aorta membranes and the half-maximal concentration needed for relaxation of rabbit aorta rings contracted by addition of serotonin. Furthermore, the binding affinity of the receptor in kidney membranes is consistent with the concentration required for in vivo natriuresis in the rat. Biologically inactive synthetic ANF fragments and other peptide hormones such as angiotensin II and vasopressin do not significantly inhibit binding. These data suggest that the receptors for ANF in vascular and renal tissues are responsible for mediating the physiological actions of this peptide in these target tissues.

Isolation and sequence determination of peptide components of atrial natriuretic factor.

The independent isolation and sequence determination in our laboratories of three closely related Atrial Natriuretic Factor peptides from rat atria confirm the sequences of ANF peptides reported by Seidah et al and synthesized by Nutt et al [Proc. Natl. Acad. Sci., (1984) in press] and contain the sequences reported by Flynn et al [Biochem. Biophys. Res. Commun. (1983) 117: 859-865] and by Currie et al [Science (1984) 223: 67-69]. In addition, we provide proof for a C-terminal tyrosine rather than tyrosine amide in our isolated peptides.

Neocarzinostatin chromophore: presence of a highly strained ether ring and its reaction with mercaptan and sodium borohydride.

Spectroscopic evidence suggests the presence of a highly strained ether ring (Fig. 1) (possibly an epoxide) in the C12-subunit of the previously determined partial structure 2a (Fig. 2) of the major neocarzinostatin chromophore (NCS-Chrom A) which completes assignment of all the oxygens in the molecule. The main product from mercaptan treatment suggests opening of the ether ring involving the addition of one molecule of mercaptan as well as reduction of the C12-substructure, whereas a parallel two-step reduction occurs on NaBH4 treatment. Both reactions occur with rearrangement of the C12-substructure and the implication for the mechanism of action of NCS-Chrom A in DNA strand scission activity is discussed. The evidence suggests a downward revision of the molecular formula for NCS-Chrom A as well as minor components B and C by two protons.

Neocarzinostatin chromophore. Assignment of spectral properties and structural requirements for binding to DNA.

The site responsible for the mercaptan (or borohydride)-stimulated DNA scission activity of neocarzinostatin chromophore (NCS-Chrom) is located in the central C12-subunit of the molecule. This has been determined by studies of the characteristic spectral properties of the chromophore and its reduction products and of the spectral changes induced by their interaction with its apoprotein (apo-NCS) and DNA. The UV-visible absorption, fluorescence, CD, and MCD spectral properties of the major nonprotein chromophoric component of neocarzinostatin (NCS-Chrom A) are assigned to its component substructures, the 2-hydroxy-5-methoxy-7-methyl-1-naphthoate, the five-membered cyclic carbonate ring (1,3-dioxolan-2-one), the 2,6-dideoxy-2-methylaminogalactose, and the incompletely defined C12-subunit which links the other three residues. Although the major source of its UV-visible absorption is the naphthoic acid residue (HNA-NCS), a significant absorption from approximately 260 to 330 nm is due to the presence of the highly unsaturated C12-subunit. The presence of the C12-subunit and, to a lesser extent, the cyclic carbonate reduces the intensity of the fluorescence emission of the fluorophore of NCS-Chrom A, the HAN-NCS subunit. The CD activity of NCS-Chrom A is also due to the presence of the C12-subunit. The MCD activity of NCS-Chrom A, however, is completely accounted accounted for by the naphthoic acid residue. A role for the C12-subunit in the binding of NCS-Chrom to DNA or apo-NCS is indicated by the resultant absorption hypochromicity in the region assigned to the C12-moiety. Limited modification of the C12-subunit (by mercaptan or borohydride) inactivates the chromophore for DNA strand scission, although both products still bind DNA. The naphthoic acid residue alone is not sufficient for binding to DNA. Therefore, the intercalation of the naphthoate residue between DNA base pairs requires the binding of NCS-Chrom to DNA probably via electrostatic interaction between the positively charged 2-methylamino group of the galactose residue and the negatively charged oxygens of the phosphate in the DNA backbone. The C12-unit is viewed as forming a short-lived reduction-activated species that, in the presence of oxygen, causes single-strand breaks in DNA. The cyclic carbonate residue is not required for in vitro DNA strand scission activity but affects its stability with respect to hydrolysis and reactivity with mercaptan. If DNA is absent the activated species decompose to inactive products, including a mercaptan (or hydrogen) addition product of the C12-subunit.