Describing Ordinal Odds Ratios for Stratified r × c Tables

For an r × ctable with ordinal responses, odds ratios are commonly used to describe the relationship between the row and column variables. This article shows two types of ordinal odds ratios where local‐global odds ratios are used to compare several groups on a c‐category ordinal response and a global odds ratio is used to measure the global association between a pair of ordinal responses. When there is a stratification factor, we consider Mantel‐Haenszel (MH) type estimators of these odds ratios to summarize the association from several strata. Like the ordinary MH estimator of the common odds ratio for several 2 × 2 contingency tables, the estimators are used when the association is not expected to vary drastically among the strata. Also, the estimators are consistent under the ordinary asymptotic framework in which the number of strata is fixed and also under sparse asymptotics in which the number of strata grows with the sample size. Compared to the maximum likelihood estimators, simulations find that the MH type estimators perform better especially when each stratum has few observations. This article provides variances and covariances formulae for the local‐global odds ratios estimators and applies the bootstrap method to obtain a standard error for the global odds ratio estimator. At the end, we discuss possible ways of testing the homogeneity assumption.     

Pre‐copulation intervals,copulation frequencies,and initial progeny sex ratios in two biotypes of whitefly,Bemisia tabaci

The whitefly Bemisia tabaci (Gennadius) (Homoptera: Aleyrodidae) is a species complex, and its systematic classification requires controlled crossing experiments among its genetic groups. Accurate information on pre‐copulation intervals, copulation frequencies, and initial frequency of egg fertilization of newly emerged adults is critical for designing procedures for collecting the virgin adults necessary for these experiments. In the literature, considerable variation is reported between B. tabaci populations, with respect to the length of the pre‐copulation interval and the initial frequency of egg fertilization. Here, we used a video‐recording method to observe continuously the copulation behaviour of the Mediterranean/Asia Minor/Africa (B biotype) and the Asia II (ZHJ1 biotype) groups of B. tabaci. We also recorded the initial frequency of egg fertilization, as determined by the sex of the progeny. When adults were caged in female–male pairs on leaves of cotton plants, the earliest copulation events occurred 2–6 h after emergence; at 12 h after emergence 56–84% of the females had copulated at least once, and nearly all (92–100%) had copulated at least once by 36 h after emergence. Both females and males copulated repeatedly. Approximately 80 and 20% of copulation events occurred during the photophase and scotophase, respectively. By 72 h post‐emergence, the females of the B and ZHJ1 biotypes had copulated on average 6.1 and 3.9 times, respectively. When adults were caged in groups on plants 1–13 h after emergence, 30–35% of the eggs deposited during this period were fertilized, and approximately 90% of females were fertilized by the end of the 13 h. Although timing of copulation differed in detail between the two genetic groups, the results demonstrate that B. tabaci adults can start to copulate as early as 2–6 h post‐emergence and the majority of females can become fertilized on the day that they emerge.     

Structural analysis of the locus containing the human C-reactive protein gene and its related pseudogene

The gene for human C-reactive protein (CRP) is mapped within a 34-kilobase pair genomic DNA segment identified by chromosome walking through overlapping DNA fragments cloned into a lambda phage library. Within 16 kilobase pairs upstream and downstream of the locus for the authentic CRP gene, only one other sequence homologous to that for CRP could be found. Sequencing analysis indicates this sequence to be a pseudogene with 50-80% region-specific homology. Comparison of the authentic CRP gene cloned from genomic DNA libraries independently prepared from three patients indicates no difference in the 5' and 3' flanking region, promoter region, or coding sequence. Only a polymorphism in the length of the poly(GT) stretch located in the intron is observed. There appears to be only one gene locus and copy per haploid chromosome for the authentic CRP gene and its pseudogene.     

Reactivation of inhibited bone acid phosphatase and its significance in bone histomorphometry

Despite biochemical demonstration of acid phosphatase (AcP) activation or reactivation in bone, few attempts have been made to show similar effects histochemically. Bones from growing rats, when fixed in 4% buffered formaldehyde at room temperature and demineralized in 5% formic acid, exhibited expected inactivation of AcP. The inhibited AcP, however, was reactivated by pre-incubation of sections for 1 hr at 37 degrees C in the following buffers: 0.2 M Tris, 0.2 M glycine, 0.2 M NaHCO3, or 0.1 M borax, as well as in alkaline water, but not in 0.2 M Na2HPO4 (all at pH 9). The reactivation was (a) site-specific (e.g., osteoclasts, osteoblasts, osteocytes, and cement lines), (b) temperature- and pH-dependent, (c) unaffected by OH- or SH--binding agents or by an alkaline phosphatase inhibitor, and (d) inhibited completely by 10 mM Na2HPO4. The reactivation process, much simplified and/or more effective than with the methods previously reported, was observed in all 83 human biopsy bones embedded in methyl methacrylate and in human bones stored in cold buffered formaldehyde for 7 months. This study demonstrates a unique method for reactivating and thus localizing the inhibited AcP in bones, and suggests possible applications in bone histomorphometry.     

Immune regulation of the L5178Y murine tumor-dormant state. II. Interferon-gamma requires tumor necrosis factor to restrain tumor cell growth in peritoneal cell cultures from tumor-dormant mice

L5178Y lymphoma cells are restrained from progressive growth in peritoneal cell ("in vitro tumor-regressor" PC) cultures prepared from many DBA/2 mice which harbor the tumor cells in the peritoneal cavity in a tumor-dormant state. Treatment of these PC cultures with 'antibodies to murine interferon-gamma (MuIFN-gamma) and murine tumor necrosis factor (MuTNF) but not with antibody to interleukin 2 (IL-2) receptors eliminated the restraint on tumor cell growth and permitted their progressive proliferation. L5178Y cells were found to be resistant to the direct toxic effects of large concentrations (3,000 U/ml) of MuIFN-gamma and of MuTNF, either alone or in combination. Treatment of PC cultures from tumor-dormant mice, in which tumor cells grew progressively ("in vitro tumor-progressor"), but not PC cultures from normal mice, with exogenous MuIFN-gamma resulted in a marked inhibition of tumor cell growth. The MuIFN-gamma-induced cytotoxic activity was cell-mediated since no soluble tumor-cytotoxic factors could be detected in the cultures. MuIFN-gamma induced cytotoxic activity in plastic-adherent peritoneal cell (AD-PC) cultures, but induced no cytotoxic activity in nonadherent-PC cultures unless small numbers (2%) of AD-PC were present, and inclusion of antibody to MuTNF in these mixed PC cultures blocked the development of cytotoxic activity. Antibody to MuTNF also blocked the development of cytotoxic activity in cultures of MuIFN-gamma-treated whole PC and AD-PC from tumor-dormant mice. These results indicate that MuIFN-gamma and MuTNF are both important in restraining tumor cell growth in PC cultures from tumor-dormant mice, and that MuIFN-gamma requires the presence of MuTNF to induce cytotoxic activity in these cultures.     

Expression of a recombinant murine IgE in transfected myeloma cells

We constructed a recombinant gene encoding an immunoglobulin (Ig) heavy chain consisting of the variable region from the phosphorylcholine (PC)-specific secreting myeloma MOPC167 and the epsilon constant region from SJL mice. This gene, cloned into the shuttle vector pSV2gpt, was transfected into J558L myeloma cells, and stable transformants that expressed the epsilon gene were cloned. The IgE heavy chain in these transformants is associated with the endogenous lambda light chain and is secreted as an intact IgE molecule. However, the secreted IgE does not bind to PC conjugated to bovine serum albumin (PC-BSA). The MOPC167 kappa chain gene was cloned into the shuttle vector pSV2neo and was transfected into the epsilon heavy-chain transformant. Stable transformants were cloned that expressed both the epsilon heavy chain and the kappa light chain. IgE secreted from such a transformant was shown to bind to PC-BSA. Both types of secreted recombinant IgE bound to rat basophilic leukemia (RBL) cells, but only the IgE produced by the cell line transformed with the MOPC167 kappa gene could be cross-linked with PC-BSA to cause serotonin release.     

A Ca2+-Dependent Protein Kinase Activity Associated with Serotonin Binding Protein

The endogenous phosphorylation of serotonin binding protein (SBP), a soluble protein found in central and peripheral serotonergic neurons, inhibits the binding of 5-hydroxytryptamine (5-HT, serotonin). A protein kinase activity that copurifies with SBP (SBP-kinase) was partially characterized and compared with calcium/calmodulin-dependent protein kinase II (CAM-PK II). SBP itself is not the enzyme since heating destroyed the protein kinase activity without affecting the capacity of the protein to bind [3H]5-HT. SBP-kinase and CAM-PK II kinase shared the following characteristics: (1) size of the subunits; (2) autophosphorylation in a Ca2+-dependent manner; and (3) affinity for Ca2+. In addition, both forms of protein kinase phosphorylated microtubule-associated proteins well and did not phosphorylate myosin, phosphorylase b, and casein. Phorbol esters or diacylglycerol had no effect on either of the protein kinases. However, substantial differences between SBP-kinase and CAM-PK II were observed: (1) CAM enhanced CAM-PK II activity, but had no effect on SBP-kinase; (2) synapsin I was an excellent substrate for CAM-PK II, but not for SBP-kinase; (3) 5-HT inhibited both the autophosphorylation of SBP-kinase and the phosphorylation of SBP, but had no effect on CAM-PK II. These data indicate that SBP-kinase is different from CAM-PK II. Phosphopeptide maps of SBP and SBP-kinase generated by digestion with S. aureus V8 protease are consistent with the conclusion that these proteins are distinct molecular entities. It is suggested that phosphorylation of SBP may regulate the transport of 5-HT within neurons.     

Evidence for active half-molecules of alpha 2-macroglobulin formed by dissociation in urea

Urea caused dissociation of alpha 2-macroglobulin (alpha 2M) into half-molecules (two disulfide-bonded subunits) as revealed by gel electrophoresis. The fraction of whole molecules remaining decreased with increasing urea concentration. Half-dissociation occurred at about 2.2 M. The ability of alpha 2M to inhibit trypsin also decreased with increasing urea concentration, but the activity-urea curve was shifted to the right as compared to the dissociation-urea curve. Thus, at 3 M urea, gel electrophoresis showed only 6.6% whole molecules, whereas the trypsin inhibitory activity was 95% of that in buffer with no urea, suggesting that half-molecules retain activity. In addition, complexes formed in urea with 125I-labeled trypsin were observed to migrate as half-molecules even though only 50% of such complexes were covalent. These results are surprising in light of the report by Gonias and Pizzo [Gonias, S., & Pizzo, S. (1983) Biochemistry 22, 536-546] that half-molecules formed by mild reduction are active; reduction is assumed to divide the molecule along an axis orthogonal to the break caused by urea. This suggests that active half-molecules can be formed by splitting either the covalent or noncovalent bonds that hold the subunits together. A model is proposed that can account for this possibility. It has the same dimensions and symmetry as a previous model of Feldman et al. [Feldman, S.R., Gonias, S.L., & Pizzo, S.V. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 5700-5704] and accounts in a similar way for previous functional studies of the protein.(ABSTRACT TRUNCATED AT 250 WORDS)     

The ovulated ovum of the horse: cytology of nonfertilized ova to pronuclear stage ova

Fertilization and early development in the horse were studied by recovering oviductal ova at various times after postovulatory mating. Ova collected between 7 and 22 h post coitum (pc) were examined for evidence of fertilizing sperm, cellular changes accompanying fertilization, and pronuclear development. Five ova collected between 7 and 9 h pc contained a marginal metaphase plate, but had no indication of sperm components; three of these, however, showed reduced numbers of cortical granules. Two activated ova (10 and 14 h pc) were in telophase of the second meiotic division, following incorporation of the fertilizing sperm. The fertilizing sperm was situated in a slight elevation; the nucleus was expanding but lacked a nuclear envelope. The pronuclear stage in the horse began as early as 12 h pc, and lasted at least until 21 h pc. Sperm tail remnants were seen in 5 of 7 pronuclear-stage ova, although the crowding of the cytoplasm with clusters of lipid and vacuoles made discerning sperm tail remnants difficult. The spindles of the metaphase stage of the second meiotic division were oriented radially, that is, at right angles to the cell surface, in all but one ovum, so this orientation is not a response to fertilization.     

N-Deacetylation of 2-acetamido-2-deoxy-hexosecontaining oligosaccharides and polysaccharides by calcium in liquid ammonia

N-Deacetylation of 2-acetamido-2-deoxy-hexose residues is accomplished in liquid ammonia containing calcium. Oligosaccharides, lacto-N-fucopentaose II and lacto-N-difucohexaose I, containing 3,4-disubstitutedN-acetylhexosamine residues are quantitativelyN-deacetylated. When applied to polysaccharides, however, only partialN-deacetylation was achieved.Author for correspondence. AXRD