Xin-Jun Liu

Induction of a non-encephalitogenic type 2 T helper-cell autoimmune response in multiple sclerosis after administration of an altered peptide ligand in a placebo- controlled, randomized phase II trial

In this ‘double-blind’, randomized, placebo-controlled phase II trial, we compared an altered peptide ligand of myelin basic protein with placebo, evaluating their safety and influence on magnetic resonance imaging in relapsing–remitting multiple sclerosis. A safety board suspended the trial because of hypersensitivity reactions in 9% of the patients. There were no increases in either clinical relapses or in new enhancing lesions in any patient, even those with hypersensitivity reactions. Secondary analysis of those patients completing the study showed that the volume and number of enhancing lesions were reduced at a dose of 5 mg. There was also a regulatory type 2 T helper-cell response to altered peptide ligand that cross-reacted with the native peptide.

CLONING OF A CDNA ENCODING A NOVEL INTERLEUKIN-1 RECEPTOR RELATED PROTEIN (IL1R-RP2)

We have identified and isolated both the rat and human cDNAs for a novel putative receptor related to the interleukin-1 type 1 receptor. We have named this protein interleukin 1 receptor related protein two (IL 1R-rp2). The rat cDNA for IL1R-rp2 was first identified using oligonucleotides of degenerate sequence in a polymerase chain reaction (PCR) paradigm with rat brain mRNA as the template. The protein encoded by both of these cDNAs are 561 amino acids long and exhibit 42% and 26% overall identity with the interleukin-1 type 1 and type 2 receptors, respectively. RNase protection assays from rat tissues revealed a predominant expression for IL 1R-rp2 in the lung and epididymis with lower levels detected in the testis and cerebral cortex. By in situ hybridization we were able to determine that the expression in rat brain appeared to be non-neuronal and associated with the cerebral vasculature. When expressed transiently in COS-7 cells the receptor was incapable of high affinity binding to either [125I]-recombinant human IL 1 alpha or [125I]-recombinant human IL 1 beta. Together, these data demonstrate the existence of a novel protein that is related to the interleukin-1 receptor but does not bind IL-1 by itself.

Cloning and Characterization of an Alternatively Processed Human Type II Interleukin-1 Receptor mRNA

Two types of interleukin (IL)-1 receptors with three extracellular immunoglobulin-like domains, limited homology (28%), and different pharmacological characteristics termed type I and type II have been cloned from mouse and human cell lines. Both receptors exist in transmembrane and soluble forms; the soluble IL-1 receptor is thought to be post-translationally derived from cleavage of the extracellular portion of the membrane receptors. In preliminary cross-linking studies with radiolabeled IL-1, we found that monkey kidney COS1 cells express a soluble receptor with molecular mass of ∼55-60 kDa, which is different from previously reported soluble IL-1 receptors. This soluble IL-1 receptor protein from COS1 cells was purified to homogeneity by affinity chromatography using recombinant IL-1β as the ligand and shown to have an affinity for human 125I-IL-1β (KD ∼2-3 nM) comparable to the human type II IL-1 receptor (IL-1RII). The purified protein was microsequenced, and the sequence information was used to design primers to clone the COS1 IL-1RII using reverse transcription-coupled polymerase chain reaction; the DNA comparison with monkey COS1 and human IL-1RII indicate that they are 95% identical at the nucleic acid and amino acid levels. In addition, another cDNA, which represents an alternatively processed mRNA of the IL-1RII gene, was also cloned both from monkey COS1 and human Raji cells and was shown to have ∼95% sequence identity between these species. While the cDNA of the novel alternatively processed gene has a 5′ end identical to the IL-1RII, the 200 base pairs at the 3′ end are different and the sequence predicts a soluble IL-1 receptor protein of 296 amino acids. Radioligand binding studies of the alternatively processed IL-1RII mRNA demonstrated kinetic and pharmacological characteristics similar to the known type II IL-1 receptor. COS7 cells (which lack IL-1 receptor) transfected with the transmembrane form of the human IL-1RII cDNA showed 125I-IL-1β binding in both the membrane fractions and supernatant. In contrast, COS7 cells transfected with the alternatively processed human IL-1RII cDNA showed high affinity 125I-IL-1β binding (Ki ∼ 1.2 nM) predominantly in the supernatant; a very small amount of detectable membrane IL-1 binding activity was also observed presumably due to association of the soluble IL-1 receptor and membrane-integrated proteins. In cross-linking and ligand blot studies, the alternatively processed human IL-1RII cDNA-transfected COS7 cells expressed a soluble IL-1 receptor with molecular masses ranging from 60 to 160 kDa, further indicating the association between this soluble IL-1 receptor and other soluble proteins. In summary, we report the purification and characterization of a soluble IL-1 receptor expressed by COS1 cells and the cloning of an alternatively processed type II IL-1 receptor mRNA from both human and COS1 cells. The alternative splicing of a primary transcript leading to a secreted protein provides a potentially important mechanism by which soluble IL-1RII can be produced.

Nephroblastoma overexpressed gene (NOV) codes for a growth factor that induces protein tyrosine phosphorylation.

NOV (nephroblastoma overexpressed gene) is a member of the CCN (connective tissue growth factor [CTGF], Cyr61/Cef10, NOV) family of proteins. These proteins are cysteine-rich and are noted for having growth-regulatory functions. We have isolated the rat NOV gene, and the DNA sequence shares 90% identity with the mouse and 80% identity with the human sequences. The rat NOV gene was expressed in all rat tissues examined, including brain, lung, heart, kidney, liver, spleen, thymus and skeletal muscle. Higher levels of rat NOV mRNA were seen in the brain, lung and skeletal muscle compared to the other tissues. Examination of NOV expression in various human cell lines revealed that NOV was expressed in U87, 293, T98G, SK-N-MC and Hs683 but not in HepG2, HL60, THP1 and Jurkat. The human NOV gene was transfected into 293 cells and the expressed protein purified. When 3T3 fibroblasts were treated with this recombinant NOV protein, a dose-dependent increase in proliferation was observed. Analysis of tyrosine-phosphorylated proteins revealed that when 3T3 cells were treated with NOV, a 221 kDa protein was phosphorylated. These data suggest that NOV can act as a growth factor for some cells and binds to a specific receptor that leads to the phosphorylation of a 221 kDa protein.

Selective Impairment of Corticotropin-Releasing Factor1 (CRF1) Receptor-Mediated Function Using CRF Coupled to Saporin1

CRF is the main component in the brain neuropeptide effector system responsible for the behavioral, endocrine, and physiological activation that accompanies stress activation. Reduced CRF system activation plays a role in the etiology of a variety of psychiatric and metabolic disease states. We have developed a novel protein conjugate that joins native rat/human CRF to a ribosome-inactivating protein, saporin (CRF-SAP), for the purpose of targeted inactivation of CRF receptor-expressing cells. Cytotoxicity measurements revealed that CRF-SAP (1-100 nM) produced concentration-dependent and progressive cell death over time in CRF1 receptor-transfected L cells, but at similar concentrations had no effect on CRF2alpha receptor-transfected cells. The CRF-SAP-induced toxicity in CRF1-transfected cells was prevented by coincubation with the competitive CRF1/CRF2 receptor peptide antagonist, [D-Phe12]CRF-(12-41), or the selective nonpeptide CRF1 receptor antagonist, NBI 27914. Finally, in cultured rat pituitary cells that express native CRF1 receptors, CRF-SAP suppressed CRF-induced (1 nM) ACTH release. GnRH (1-10 nM) stimulated LH release was also assessed in the same pituitary cultures. Although there was a slight decrease in LH release from these cultures, this decrease was observed with CRF-SAP or SAP alone, suggesting that the response was nonspecific. Taken together, these results suggest the utility of CRF-SAP as a specific and subtype-selective tool for long term impairment of CRF1 receptor-expressing cells.

Characterization of [125I-Tyr0]-corticotropin releasing factor (CRF) binding to the CRF binding protein using a scintillation proximity assay

We describe the characterization of high affinity [125I-Tyr0]-human CRF binding to purified recombinant human CRF-binding protein (CRF-BP) using a scintillation proximity assay (SPA). For this stable nonseparation technique developed in 96 well microtiter plates, biotinylated CRF-BP is captured by streptavidin-coated SPA beads for the detection of bound [125I-Tyr0]-CRF. Unbound [125I-Tyr0]-CRF represented little or no signal in the assay. Total binding observed was greater than 5000 cpm with a nonspecific signal of < 100 cpm determined in the presence of excess unlabeled human CRF. A comparison of the SPA method with a charcoal precipitation method confirmed that the biotinylation procedure did not adversely affect affinity of the CRF-BP for [125I-Tyr0]-CRF. Saturation binding analysis yielded an apparent equilibrium dissociation constant (Kd) of 208 +/- 5.0 pM (+/- S.D., n = 3). An inhibition constant (Ki) for unlabeled CRF was calculated to be 0.22 +/- 0.03 nM (+/- S.D., n = 8) and a pharmacological profile for eight CRF-related neuropeptides gave a rank potency similar to previously reported results. Finally, the assay variability was assessed with intra- and inter-plate coefficients of variation which were less than 5% each.

Regulation of IGFBP-4 and -5 Expression in Rat Granulosa Cells

One of the central questions in reproductive physiology concerns how follicles in the ovary are selected during folliculogenesis. Although it has been well established that folliculogenesis is initiated and maintained by the gonadotropin, follicle-stimulating hormone (FSH), how follicle selection takes place in the ovary is still an enigma.1 Based on the notion that selection of the dominant follicle might be due to the action of a locally produced inhibitor in the follicular fluid which could block the effects of FSH on the non-selected follicles, our laboratory initiated a project in 1988 to isolate and identify the FSH inhibitor present in porcine ovarian follicular fluid. Using a well-defined rat granulosa cell culture assay to monitor the purification, a polypeptide was isolated which exhibited a potent inhibition of FSH-stimulated estradiol production in those cells.2 Chemical characterization of the inhibitor, however, revealed that its structure corresponded to insulin-like growth factor binding protein-3 (IGFBP-3).3 Subsequent studies carried out by our laboratory identified that the rat ovary produced five IGFBPs in a tissue-specific manner, with the mRNA for IGFBP-2 mainly expressed by theca interstitial cells, IGFBP-3 by luteinized granulosa cells, IGFBP-4 by the granulosa of atretic antral follicles, IGFBP-5 by the granulosa of atretic preantral follicles and IGFBP-6 by the theca externa, stroma and smooth muscle cells4–6. IGFBP-1 was not detected in the rat ovary. These findings indicate that the transcription of each IGFBP gene is regulated differently in the rat ovary and that each IGFBP may have a different physiological function. This hypothesis is strengthened by our finding that exogenously added IGFBP-2, -3, -4 and -5 were able to attenuate the FSH-stimulated production of estradiol and progesterone in cultured rat granulosa cells7,8 and ovarian intrabursal injection of IGFBP-3 was able to decrease the ovulation rate of pregnant mare’s serum gonadotropin/human choriogonadotropin-primed animals.9