Corticotropin-releasing factor (CRF) receptors in intermediate lobe of the pituitary: biochemical characterization and autoradiographic localization

CRF receptors were characterized using radioligand binding and chemical affinity cross-linking techniques and localized using autoradiographic techniques in porcine, bovine and rat pituitaries. The binding of 125I-[Tyr0]-ovine CRF (125I-oCRF) to porcine anterior and neurointermediate lobe membranes was saturable and of high affinity with comparable KD values (200-600 pM) and receptor densities (100-200 fmoles/mg protein). The pharmacological rank order of potencies for various analogs and fragments of CRF in inhibiting 125I-oCRF binding in neurointermediate lobe was characteristic of the well-established CRF receptor in anterior pituitary. Furthermore, the binding of 125I-oCRF to both anterior and neurointermediate lobes of the pituitary was guanine nucleotide-sensitive. Affinity cross-linking studies revealed that the molecular weight of the CRF binding protein in rat intermediate lobe was identical to that in rat anterior lobe (Mr = 75,000). While the CRF binding protein in the anterior lobes of porcine and bovine pituitaries had identical molecular weights to CRF receptors in rat pituitary (Mr = 75,000), the molecular weight of the CRF binding protein in porcine and bovine intermediate lobe was slightly higher (Mr = 78,000). Pituitary autoradiograms from the three species showed specific binding sites for 125I-oCRF in anterior and intermediate lobes, with none being apparent in the posterior pituitary. The identification of CRF receptors in the intermediate lobe with comparable characteristics to those previously identified in the anterior pituitary substantiate further the physiological role of CRF in regulating intermediate lobe hormone secretion.     

The brain corticotropin-releasing factor (CRF) receptor is of lower apparent molecular weight than the CRF receptor in anterior pituitary. Evidence from chemical cross-linking studies

The ligand binding subunits of the corticotropin-releasing factor (CRF) receptors in brain and anterior pituitary of a number of species have been identified by chemical affinity cross-linking using the homobifunctional cross-linking agent disuccinimidyl suberate and 125I-Tyr0-oCRF (ovine CRF). In homogenates of rat, monkey, and human cerebral cortex, 125I-Tyr0-oCRF was covalently incorporated into a protein of Mr = 58,000. Under identical conditions in the anterior pituitary of rat, monkey, cow, and pig, 125I-Tyr0-oCRF was incorporated into a protein of apparent Mr = 75,000. The specificity of the labeling was typical of the CRF binding site since both the cerebral cortex- and pituitary-labeled proteins exhibited the appropriate pharmacological rank order profile characteristic of the CRF receptor (Nle21,Tyr32-oCRF approximately equal to rat/human CRF approximately equal to ovine CRF approximately equal to alpha-helical CRF(6-41) greater than alpha-helical oCRF(9-41) greater than or equal to oCRF(7-41) greater than rat/human CRF(1-20) approximately equal to vasoactive intestinal peptide). In addition to the major labeled proteins, 125I-Tyr0-oCRF was incorporated into higher molecular weight peptides which may represent precursors and into lower molecular weight components which may represent fragments of the major labeled proteins or altered forms of the CRF binding subunit. In summary, these data indicate a heterogeneity between brain and pituitary CRF receptors with the ligand binding subunit of the brain CRF receptor residing on a Mr = 58,000 protein, while in the anterior pituitary, the identical binding subunit resides on a protein of apparent Mr = 75,000.     

Molecular and Pharmacological Characterization of GABAA Receptors in the Rat Pituitary

Abstract: Levels of mRNA for the major subunits of the GABA_A receptor were assayed in the rat pituitary anterior and neurointermediate lobes by ribonuclease protection assay. α1, β1, β2, β3, and γ2s were found to be the predominant subunits in the anterior lobe, whereas α2, α3, β1, β3, γ2s, and γ1 were the predominant subunits expressed in the neurointermediate lobe. α5, α6, and δ subunits were not detectable. Hill and Scatchard analysis of [³H]muscimol binding to anterior and neurointermediate lobe membranes showed high-affinity binding sites with dissociation constants of 5.6 and 4.5 n M , respectively, and Hill coefficients near 1. Muscimol sites were present at a maximum of 126 fmol/mg in the anterior lobe and 138 fmol/mg in the neurointermediate lobe. The central-type benzodiazepine antagonist [³H]Ro 15-1788 bound to a high-affinity site with a dissociation constant of 1.5 n M in both tissues, at a maximum of 60 fmol/mg in anterior pituitary and 72 fmol/mg in neurointermediate lobe. A Hill coefficient of 1 was measured for this site in both tissues. Assays of CL 218 872 displacement of Ro 15-1788 were consistent with a pure type I benzodiazepine site in the anterior lobe and a pure type II site in the intermediate lobe. These results are consistent with both tissue-specific expression of particular GABA_A receptor subunits and receptor heterogeneity within individual cells in the pituitary.     

An islet activating protein-sensitive G protein is involved in dopamine inhibition of angiotensin and thyrotropin-releasing hormone-stimulated inositol phosphate production in anterior pituitary cells

In primary culture of anterior pituitary cells, dopamine inhibited the angiotensin (AII)-stimulated inositol phosphate production by 28 +/- 2.5% (n = 14), with an EC50 of 660 +/- 228 nM (n = 8). This effect was blocked by (+)-butaclamol, a specific dopamine receptor antagonist. RU 24926, a D2 specific agonist, but not SKF 38393, a specific D1 agonist, inhibited AII-stimulated inositol phosphate production, suggesting that this dopamine effect is mediated through a dopamine receptor of the D2 subtype. Dopamine also partially inhibited (25%) inositol phosphate production stimulated by thyrotropin-releasing hormone (TRH). Our results suggest that the dopamine-mediated inhibition of hormonally stimulated inositol phosphate production is probably not mediated through the known inhibitory effects of dopamine on cAMP and Ca2+ intracellular concentrations. Although unknown, the mechanism by which dopamine inhibited the AII and TRH-stimulated inositol phosphate production implicates a GTP binding protein sensitive to the islet activating protein (IAP) since dopamine effects were blocked by this toxin. The alpha subunit of the GTP binding protein involved could be one of the three ADP-ribosylated proteins found in anterior pituitary cells in primary cultures, the alpha o (39 kDa), the alpha i (41 kDa), and an alpha subunit of 40 kDa. Indeed, we show here that this 40-kDa IAP substrate, already described in a few tissues, is present in anterior pituitary cells. The negative coupling between dopamine receptors and the AII or TRH inositol phosphate production systems, could be implicated in the dopamine inhibition of the AII- and TRH-stimulated prolactin release since such an inhibition is blocked by IAP. Our results suggest that the negative regulation of inositol phosphate production is one of the mechanisms by which dopamine controls hormonally stimulated prolactin release.     

Vasopressin and "CRF" in the regulation of ACTH secretion by the anterior and intermediate lobes of the pituitary gland (author's transl)

The aim of this review was to summarize the present state of knowledge concerning the mode of action of vasopressin (VP) and the putative corticotropin releasing factor (CRF) on ACTH secretion from the anterior and intermediate lobes of the pituitary gland. In vitro data show that although both CRF and VP enhanced release of anterior pituitary ACTH, the pattern of hormonal release, based on kinetical and dose-dependent studies, appeared to be different. Also, the effect of VP most probably was mediated by specific putative receptor sites. In contrast, VP was found not to alter ACTH secretion from the intermediate lobe; that secretion seems to be regulated by CRF-like material and neurotransmitters. The importance of VP as a corticotropin agent is discussed.     

Endocytosis of the vasopressin receptor by anterior pituitary cells is increased by corticotropin-releasing factor (CRF)

Endocytosis of certain receptors such as the transferrin receptor and the EGF-receptor appears to be influenced by second messengers. If second messengers are involved in modulation of endocytosis, not only endocytosis of the stimulated receptor itself but also of receptors for other ligands on the cell surface may be influenced by receptor occupancy. Corticotropin-releasing factor (CRF) and vasopressin act synergistically on secretion of ACTH from the anterior pituitary. The results presented here demonstrate that CRF increases retrieval of the vasopressin receptor in anterior pituitary cells in primary culture without influencing the surface binding of vasopressin. This is not a function of an increased membrane turnover since endocytosis of the transferrin receptor is not influenced by CRF.     

Existence of a common precursor to ACTH and endorphin in the anterior and intermediate lobes of the rat pituaitary

Extracts of rat anterior and intermediate-posterior pituitary were fractionated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and assayed for immunoactive ACTH and endorphin. In both lobes the major forms of immunoactive ACTH have apparent molecular weights of 31,000 (31K), 20–21K, 14K, and 4.5K, and the major forms of immunoactive endorphin have apparent molecular weights of 31K (coincident with the peak of immunoactive ACTH), 13K (a βLPH-like peptide), and 3.5K (a β-endorphin-like peptide). However, the quantitative distribution of immunoactivity among the various forms differs greatly between the lobes. Assays using an extreme COOH-terminal ACTH antiserum indicate that the 31K ACTH/endorphin molecule in rat antierior and intermediate pituitary is similar to the pro-ACTH/endorphin molecule from mouse pituitary tumor cells. A radioimmunoassay that is specific for the NH₂-terminal non-ACTH, nonendorphin segment (referred to as 16K fragment) of the mouse pro-ACTH/endorphin molecule was used to assay extracts of rat pituitary. In addition to detecting material at 31K and 20–21K, the 16K fragment radioimmunoassay detects significant amounts of cross-reactive material with an apparent molecular weight of 16K in extracts of both lobes. This result also suggests that the structure and processing of the rat 31K ACTH/endorphin molecule is similar to that of mouse tumor cell pro-ACTH/endorphin. Cell suspensions were prepared from the anterior and intermediate lobes of the rat pituitary and maintained in culture for a 24-h period. The isolated cells from both lobes incorporate [³H] phenylalanine into immunoprecipitable ACTH- and endorphin-containing molecules. By sequential immunoprecipitation with ACTH and endorphin antisera, it is possible to demonstrate directly that a single molecule (31K ACTH/endorphin) has antigenic determinants for both ACTH and endorphin. Significant amounts of 31K ACTH/endorphin are released into the culture medium by isolated anterior lobe and intermediate lobe cells. The isolated intermediate lobe cells synthesize and secrete relatively large amounts of a β-endorphin-like molecule; the isolated anterior lobe cells secrete significant amounts of both a βLPH-like molecule and a β-endorphin like molecule. These same quantitative differences between anterior and intermediate lobe tissue were observed in immunoassays of extracts of the separated lobes and probably reflect differences in the processing of the common precursor. The isolated anterior lobe cells can be stimulated to release increased amounts of immunoprecipitable ACTH and endorphin by incubation with a cyclic AMP analog and a phosphodiesterase inhibitor.     

Comparison of the Molecular Forms of the Kex2/Subtilisin-Like Serine Proteases SPC2, SPC3, and Furin in Neuroendocrine Secretory Vesicles Reveals Differences in Carboxyl-Terminus Truncation and Membrane Association

Abstract: The molecular forms and membrane association of SPC2, SPC3, and furin were investigated in neuroendocrine secretory vesicles from the anterior, intermediate, and neural lobes of bovine pituitary and bovine adrenal medulla. The major immunoreactive form of SPC2 was the full-length enzyme with a molecular mass of 64 kDa. The major immunoreactive form of SPC3 was truncated at the carboxyl terminus and had a molecular mass of 64 kDa. Full-length 86-kDa SPC3 with an intact carboxyl terminus was found only in bovine chromaffin granules. Immunoreactive furin was also detected in secretory vesicles. The molecular masses of 80 and 76 kDa were consistent with carboxyl-terminal truncation of furin to remove the transmembrane domain. All three enzymes were distributed between the soluble and membrane fractions of secretory vesicles although the degree of membrane association was tissue specific and, in the case of SPC3, dependent on the molecular form of the enzyme. Significant amounts of membrane-associated and soluble forms of SPC2, SPC3, and furin were found in pituitary secretory vesicles, whereas the majority of the immunoreactivity in chromaffin granules was membrane associated. More detailed analyses of chromaffin granule membranes revealed that 86-kDa SPC3 was more tightly associated with the membrane fraction than the carboxyl terminus-truncated 64-kDa form.     

Folate Receptors in Malignant and Benign Tissues of Human Female Genital Tract

We have characterized the folate receptor in malignant and benign tissues of human female genital tract (Fallopian tube and benign and malignant tissues of uterus). Radioligand binding displayed characteristics similar to those of other folate binding proteins. Those include a high-affinity type of binding (K=10¹⁰M^–1), apparent positive cooperativity, a slow dissociation at pH 7.4 becoming rapid at pH 3.5, and inhibition of binding by folate analogues. The gel filtration profile of Triton X-100 solubilized tissue contained two large peaks of ³H-folate labelled protein (>=130 and 100kDa) as well as a 25 kDa peak. Only a single band of 70 kDa was seen on SDS-PAGE immunoblotting. The large molecular size forms on gel filtration appear to represent folate receptors having a hydrophobic membrane anchor inserted into Triton X-100 micelles. The folate receptor of female genital tract showed cross-reactivity in ELISA and positive immunostaining with rabbit antibodies against human milk folate binding protein. Variations in the ratio of immunoresponse to total high affinity folic acid binding suggests the presence of multiple isoforms of the receptor in different types of malignant and benign tissues.     

Brain and pituitary receptors for corticotropin releasing factor: Localization and differential regulation after adrenalectomy

Specific receptors for corticotropin releasing factor (CRF) were identified in two functionally distinct systems within the brain, the cortex and the limbic system. Autoradiographic mapping of the CRF receptors in the brain revealed high binding density throughout the neocortex and cerebellar cortex, subiculum, lateral septum, olfactory tract, bed nucleus of the stria terminalis, interpeduncular nucleus and superior colliculus. Moderate to low binding was found in the hippocampus, nucleus accumbens, claustrum, nucleus periventricularis thalamus, mammillary bodies, subthalamic nucleus, periaqueductal grey, locus coeruleus and nucleus of the spinal trigeminal tract. As in the anterior pituitary gland, CRF receptors in the brain were shown to be coupled to adenylate cyclase. However, in contrast to the marked decrease in CRF receptors observed after adrenalectomy in the anterior pituitary gland, CRF receptor concentration in the brain and pars intermedia of the pituitary was unchanged. The presence of CRF receptors in areas involved in the control of hypothalamic and autonomic nervous system functions is consistent with the major role of CRF in the integrated response to stress.     

[3H]Nitrendipine Binding to Calcium Channels in Bovine and Rat Pituitary

[3H]Nitrendipine was used to label sites in homogenates of bovine anterior and neurointermediate lobes of the pituitary gland. The amount of specific binding in the anterior lobe was 1.82 +/- 0.30 pmol/g wet weight of tissue and the KD was 1.44 +/- 0.02 X 10(-10) M. Preliminary experiments indicated a similar amount of binding in bovine neurointermediate lobe. In competition studies nimodipine and nisoldipine (two potent voltage-sensitive calcium channel blockers) displayed IC50 values of 1.6 and 6.8 X 10(-10) M, respectively. Verapamil and the verapamil-like calcium channel blockers D-600 and tiapamil competed in a complex manner for the [3H]nitrendipine specific binding to bovine anterior pituitary homogenates. Autoradiographical studies demonstrated specific [3H]nitrendipine binding sites distributed approximately equally in the anterior and posterior lobes, but not in the intermediate lobe of the rat pituitary. In general the properties of [3H]nitrendipine binding in the pituitary tissue resemble strongly the properties of [3H]nitrendipine binding in the brain which is believed to be to voltage-sensitive calcium channels. These results provide support for the hypothesis that calcium channels are involved in pituitary hormone secretion and that drugs that interact with calcium channels may modulate the secretory process directly at the level of the pituitary.     

Identification and characterization of a pituitary corticotropin-releasing factor binding protein by chemical cross-linking

A corticotropin-releasing factor (CRF) binding protein has been identified based on the chemical cross-linking of ovine [Nle21,m-125I-Tyr32]CRF (125I-oCRF) to bovine anterior pituitary membranes using disuccinimidyl suberate (DSS). The apparent molecular weight of the cross-linked complex determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by autoradiography was approximately 75,000 and was slightly decreased in its nonreduced state, suggesting the presence of intramolecular disulfide bonds. Subtracting the molecular weight of 125I-oCRF, the binding protein appeared to have a molecular weight of approximately 70,000. The cross-linking was specific since an excess (1 microM) of an unrelated peptide (insulin) did not affect the appearance of the Mr 75,000 band. The concentration of CRF required to inhibit cross-linking by 50% was found to be similar to that determined for bovine pituitary CRF receptors by radioreceptor assay. The nonhydrolyzable GTP analogue 5'-guanylylimidodiphosphate dose dependently inhibited the cross-linking of 125I-oCRF to the Mr 70,000 protein. 50 nM of the inactive CRF analogue, [Ala14]oCRF, had no effect on the cross-linking, an observation which is consistent with this compound's low potencies in bioassays and radioreceptor assays. These results strongly suggest that this Mr 70,000 protein is the biological bovine anterior pituitary CRF receptor.     

N-cadherin loss in POMC-expressing cells leads to pituitary disorganization

Pituitary tumors are the third most common intracranial tumor in humans and can cause altered hormone secretions leading to hypercortisolism, acromegaly, and infertility. Reduced expression of the cell adhesion molecule N-cadherin has been linked with the formation of pituitary tumors, but its role in normal pituitary gland physiology or tumor initiation is unknown. In the murine pituitary, N-cadherin expression is detected in virtually all cells of the posterior, intermediate, and anterior lobes. N-cadherin may function to initiate important cues such as controlling proliferation, directing cell placement, and promoting formation of cell networks that coordinately release hormones into the bloodstream. To address this, we generated mice lacking N-cadherin in proopiomelanocortin-expressing melanotrope and corticotrope cells of the intermediate and anterior lobes of the pituitary. We observed that intermediate lobe cells can aberrantly displace SOX2-containing progenitor cells in the N-cadherin conditional knockout mice at postnatal d 1. By postnatal d 30, although a reduction in α- and β-catenin membrane staining occurs, there is little effect on intermediate lobe architecture with N-cadherin loss. Also, despite these changes in adherens junction molecules, no alterations in cell proliferation occur. In contrast, loss of N-cadherin in the corticotropes leads to aberrant cell clustering and a reduction in Pomc mRNA. Taken together, our data reveal important roles of N-cadherin in pituitary cell placement and that loss of N-cadherin alone does not lead to pituitary tumor formation.     

Characterization of Leydig cell gonadotropin-releasing hormone binding sites utilizing radiolabeled agonist and antagonist

Gonadotropin-releasing hormone (GnRH) binding sites in intact Leydig cells and in membrane preparations were investigated using 125I-labeled GnRH agonist and antagonist. Binding was saturable and involved a single class of high affinity sites. Intact Leydig cells and a membrane preparation had a higher affinity for GnRH agonist (Kd 3.0 +/- 1.7 X 10(-10) M) than for GnRH antagonist (Kd 10.0 +/- 1.8 X 10(-10) M). With anterior pituitary membranes the Kd was 2.8 +/- 0.7 X 10(-10) M for the agonist and 2.4 +/- 1.4 X 10(-10) M for the antagonist. The Kd for GnRH was similar for Leydig cells and the anterior pituitary. Chymotrypsin and trypsin digestion decreased receptor binding, but neuraminidase increased Leydig cell binding in contrast to the decrease in binding observed with pituitary receptors. The results suggest that the Leydig cell GnRH binding sites may differ from the pituitary receptor which may be related to structural differences in GnRH-like peptides recently described in extracts of rat testis.     

Lectin-labeled membrane is transferred to the Golgi complex in mouse pituitary cells in vivo

Labeling of the Golgi complex with the lectin conjugate wheat germ agglutinin-horseradish peroxidase (WGA-HRP), which binds to cell surface membrane and enters cells by adsorptive endocytosis, was analyzed in secretory cells of the anterior, intermediate, and posterior lobes of mouse pituitary gland in vivo. WGA-HRP was administered intravenously or by ventriculo-cisternal perfusion to control and salt-stressed mice; post-injection survival times were 30 min-24 hr. Peroxidase reaction product was identified within the extracellular clefts of anterior and posterior pituitary lobes through 24 hr but was absent in intermediate lobe. Endocytic vesicles, spherical endosomes, tubules, dense and multivesicular bodies, the trans-most saccule of the Golgi complex, and dense-core secretory granules attached or unattached to the trans Golgi saccule were peroxidase-positive in the different types of anterior pituitary cells and in perikarya of supraoptico-neurohypophyseal neurons; endoplasmic reticulum and the cis and intermediate Golgi saccules in the same cell types were consistently devoid of peroxidase reaction product. Dense-core secretory granules derived from cis and intermediate Golgi saccules in salt-stressed supraoptic perikarya likewise failed to exhibit peroxidase reaction product. The results suggest that in secretory cells of anterior and posterior pituitary lobes, WGA-HRP, initially internalized with cell surface membrane, is eventually conveyed to the trans-most Golgi saccule, in which the lectin conjugate and associated membrane are packaged in dense-core secretory granules for export and potential exocytosis of the tracer. Endoplasmic reticulum and the cis and intermediate Golgi saccules appear not to be involved in the endocytic/exocytic pathways of pituitary cells exposed to WGA-HRP.     

Benzodiazepine suppression of corticotropin-releasing factor (CRF)-induced beta-endorphin release from rat neurointermediate pituitary.

Dopamine and gamma-aminobutyric acid (GABA) inhibit POMC peptide release from the pituitary intermediate lobe, via interaction with D2 or GABA-A/benzodiazepine receptors. Here, we examined the effects of an antianxiety triazolobenzodiazepine, adinazolam, on corticotropin-releasing factor (CRF)-stimulated POMC peptide secretion from the rat neurointermediate pituitary. Neurointermediate lobes (NILS) were incubated with CRF (10(-7) M), then adinazolam (10(-8) or (10(-9) M) was added, with CRF remaining in the medium. Aliquots were removed at 15-min intervals and frozen for radioimmunoassay of beta-endorphin. Adinazolam alone did not significantly affect secretion as compared to controls or CRF alone. Adinazolam incubated with CRF led to significant inhibition of beta-endorphin secretion, as compared to CRF alone. In addition, adinazolam was as effective as dopamine or the CRF antagonist, alpha-helical CRF, in preventing CRF-induced beta-endorphin release. Adinazolam appears to act directly on the pituitary to suppress hormone release induced by a stress-related hypothalamic peptide.     

Activation of protein kinase C differentially regulates corticotropin-releasing factor-stimulated peptide secretion and cyclic AMP formation of intermediate and anterior pituitary cells in culture

The present study was aimed at investigating the effect of protein kinase C (PKC) activation on CRF receptor function of proopiomelanocortin (POMC) cells in culture. Incubation of tissues with the phorbol ester PMA selectively potentiated corticotropin-releasing factor (CRF)-stimulated ACTH secretion and cyclic AMP formation of anterior pituitary (AP) cells, while, in sharp contrast, it failed to similarly affect intermediate pituitary (IP) cells and AtT-20 corticotrophs exposed to CRF. Unexpectedly, however, long-term treatment of cultures with PMA, which depletes cell stores of PKC, resulted in a similar dramatic attenuation of stimulated peptide release from both corticotrophs and melanotrophs, while being without significant effect on cyclic AMP production. Exposure of cells to PMA did not change either basal or CRF-enhanced levels of POMC mRNA. We conclude that activation of PKC fails to synergize with CRF-mediated signalling in IP and AtT-20 cells, although optimal CRF receptor expression requires the presence of a functional kinase C pathway, thus suggesting cross-talks between both messenger systems.     

Affinity purification of human granulocyte macrophage colony-stimulating factor receptor alpha-chain. Demonstration of binding by photoaffinity labeling

The human granulocyte macrophage colony-stimulating factor (GM-CSF) receptor alpha-chain, a low affinity component of the receptor, was solubilized and affinity-purified from human placenta using biotinylated GM-CSF. Scatchard analysis of 125I-GM-CSF binding to the placental membrane extract disclosed that the GM-CSF receptor had a dissociation constant (Kd) of 0.5-0.8 nM, corresponding to the Kd value of the GM-CSF receptor alpha-chain on the intact placental membrane. Affinity labeling of the solubilized protein using a photoreactive cross-linking agent, N-hydroxysuccinimidyl-4-azidobenzoate (HSAB), demonstrated a single specific band of 70-95 kDa representing a ligand-receptor complex. Approximately 2 g of the placental membrane extract was subjected to a biotinylated GM-CSF-fixed streptavidin-agarose column, resulting in a single major band at 70 kDa on a silver-stained sodium dodecyl sulfate gel. The radioiodination for the purified material disclosed that the purified protein had an approximate molecular mass of 70 kDa and a pI of 6.6. Binding activity of the purified material was demonstrated by photoaffinity labeling using HSAB-125I-GM-CSF, producing a similar specific band at 70-95 kDa as was demonstrated for the crude protein.