Male-biased hermaphrodites in a gynodioecious shrub, Daphne jezoensis.

Gynodioecy, a state where female and hermaphrodite plants coexist in populations, has been widely proposed an intermediate stage in the evolutionary pathway from hermaphroditism to dioecy. In the gynodioecy-dioecy pathway, hermaphrodites may gain most of their fitness through male function once females invade populations. To test this prediction, comprehensive studies on sex ratio variation across populations and reproductive characteristics of hermaphrodite and female phenotypes are necessary. This study examined the variation in sex ratio, sex expression, flower and fruit production and sexual dimorphism of morphological traits in a gynodioecious shrub, Daphne jezoensis, over multiple populations and years. Population sex ratio (hermaphrodite:female) was close to 1:1 or slightly hermaphrodite-biased. Sex type of individual plants was largely fixed, but 15% of plants changed their sex during a 6-year census. Hermaphrodite plants produced larger flowers and invested 2.5 times more resources in flower production than female plants, but they exhibited remarkably low fruit set (proportion of flowers setting fruits). Female plants produced six times more fruits than hermaphrodite plants. Low fruiting ability of hermaphrodite plants was retained even when hand-pollination was performed. Fruit production of female plants was restricted by pollen limitation under natural conditions, irrespective of high potential fecundity, and this minimised the difference in resources allocated to reproduction between the sexes. Negative effects of previous flower and fruit production on current reproduction were not apparent in both sexes. This study suggests that gynodioecy in this species is functionally close to a dioecious mating system: smaller flower production with larger fruiting ability in female plants, and larger flower production with little fruiting ability in hermaphrodite plants.

Current Status of CD4-Based Therapies for Prophylaxis and Treatment of HIV Infection.

CD4 is a member of the immunoglobulin superfamily. It is usually found as a surface glycoprotein on mature circulating T lymphocytes with helper/inducer function. The first step of HIV infection involves binding of the CD4 of host cells to the gp120 portion of viral envelope glycoprotein. In addition, uninfected CD4+ cells with free gp120 bound to their surfaces are the target of antibody-dependent cellular cytotoxicity, and a CD4-gp120 interaction is necessary for the cell-to-cell spread of HIV. Early enthusiasm for therapy with recombinant soluble CD4 (rsCD4) was supported by evidence of potent antiviral effects against acquisition and spread of HIV-1 infection in vitro. Subsequent in vivo results were not as encouraging; the pharmacokinetics of rsCD4 were unfavourable, and there was little apparent effect on clinical markers of infection. With the recognition that laboratory strains and clinical strains of HIV-1 differ enormously in their response to rsCD4, it was realised that early trials had used subtherapeutic doses. High dose intravenous rsCD4 abrogates HIV-1 viraemia in a dose-dependent fashion, with the duration of effect limited by the pharmacokinetics of rsCD4. With the intention of improving pharmacokinetics, and of broadening the mechanism of action, further molecules combining rsCD4 and immunoglobulin molecules were developed. Their pharmacokinetics had improved as anticipated, with a marked increase in half-life, and the molecules retained the Fc-mediated effector functions of the parent immunoglobulin molecule. However, therapeutic benefit in the treatment of established HIV-1 infection has been limited, with some animal evidence of prevention of infection after known exposure to HIV, and possible implications for the prevention of perinatal and needle-stick infection. Synergism between different retroviral therapies, using rsCD4 and CD4-IgG molecules in combination with various other agents, has been demonstrated in vitro, but the clinical effectiveness of this strategy has not been demonstrated. Synergism between CD4-IgG and naturally occurring or monoclonal HIV-1 antibodies is being evaluated. Attempts to use the CD4 molecule to give Pseudomonas toxin access to HIV-1 for specific killing have demonstrated potent in vitro effectiveness, but dose-limiting toxicity in phase I human studies made this strategy ineffective. Attempts using other toxins are continuing.