Chemical synthesis of a gene coding for human angiogenin, its expression in Escherichia coli and conversion of the product into its active form.

A synthetic gene coding for human angiogenin was synthesized by solid support phosphoramidite chemistry as eight long oligodeoxynucleotides which were subsequently assembled and cloned in Escherichia coli. The gene was designed to use codons found in highly expressed E. coli proteins. A pBR322-derived expression vector was constructed containing the E. coli trp promoter, the ribosome-binding site of the bacteriophage lambda cII gene, the angiogenin coding sequence, and the transcription terminator region of the E. coli rrnB operon. Under tryptophan deprivation, angiogenin was strongly expressed in E. coli cells at a yield of 5-10% of total protein. The eukaryotic protein was found to be insoluble but could be easily renatured and purified. The purified angiogenin was demonstrated to be active as an angiogenic factor and exhibited a characteristic RNase activity.

Heterologous protein export in Escherichia coli: influence of bacterial signal peptides on the export of human interleukin 1 beta.

Expression plasmids carrying the coding sequence of mature human interleukin 1 beta (IL 1 beta) linked either to a Met start codon, or fused to different efficient Escherichia coli secretion signal sequences, have been constructed. In the latter case, we used signal peptides derived either from an outer membrane protein (OmpA) or from a periplasmic protein (PhoA). The synthesis of IL1 beta from these fusions was investigated in an otherwise strictly isogenic context using identical conditions of derepression and culture media. The Met-IL1 beta fusion produced a soluble cytoplasmic protein which could be released from the cells by osmotic shock whereas the OmpA and PhoA fusions were always insoluble. The extent of sOmpA-IL1 beta maturation was found to vary from 50 to 100%, mainly depending on the medium used, whereas no significant maturation of the signal peptide could be detected in the case of the sPhoA-IL1 beta fusion. Immuno-electron microscopy revealed that the sOmpA-IL1 beta fusion was targeted to the inner membrane, whereas the sPhoA-IL1 beta fusion remained within the cytoplasm and thus did not appear to enter the secretion pathway. Amplifying the E. coli signal peptidase lep gene on a multicopy plasmid did not improve signal peptide removal from sOmpA-IL1 beta. Moreover, these E. coli secretion vectors allowed us to produce, in high levels, IL1 beta fragments which otherwise could not be stably accumulated within the cytoplasmic compartment.

Mucocutaneous and invasive candidiasis among very low birth weight (less than 1,500 grams) infants in intensive care nurseries: a prospective study.

To determine whether mucocutaneous candidiasis presages the development of invasive candidiasis and to assess factors influencing the development of mucocutaneous candidiasis and invasive candidiasis among infants requiring neonatal intensive care, all infants admitted to our neonatal intensive care unit during a 47-month period were prospectively examined twice weekly for mucocutaneous candidiasis. Because 16 of 18 (89%) infants in whom invasive candidiasis (defined by positive cultures of blood, CSF, deep tissue or greater than or equal to 2 supra-pubic urine aspirates) developed had birth weights less than 1,500 g, further analysis was focused toward the very low birth weight group. Of 358 very low birth weight infants hospitalized for less than three days and serially studied until discharge from the neonatal intensive care unit, mucocutaneous candidiasis developed in 28 (7.8%), invasive candidiasis developed in 16 (4.5%), and in 323 there was no evidence of mucocutaneous candidiasis or invasive candidiasis. Although many risk factors were shown by univariate analysis to be significantly more common among those with invasive candidiasis and mucocutaneous candidiasis, adjustment for the covariant effects of duration of hospitalization and gestational age revealed that only prolonged duration of antibiotic therapy and duration of endotracheal intubation were significantly associated with invasive candidiasis. Invasive candidiasis developed later in nine of 28 (32%) infants with mucocutaneous candidiasis despite nystatin therapy of mucocutaneous candidiasis in all nine (median duration of therapy before invasive candidiasis, nine days). Very low birth weight infants in whom mucocutaneous candidiasis develops are at significantly greater risk of invasive candidiasis developing later than those in whom mucocutaneous candidiasis did not develop (9/28 v 7/330, P less than .001).

Polymicrobial sepsis among intensive care nursery infants.

To determine the incidence, characteristics, and course of polymicrobial sepsis among infants in intensive care nurseries, we reviewed all such episodes in our neonatal unit from September 1971 through June 1986. We identified 15 episodes (3.9% of all cases of culture-proven sepsis during the survey period) in which blood or cerebrospinal fluid (CSF) culture yielded multiple organisms felt to represent true pathogens. Mortality associated with late-onset polymicrobial sepsis (7 of 10; 70%) was significantly higher (P less than .001) than in late-onset monomicrobial sepsis (86 of 370; 23%). Six patients were 37 weeks' gestation or greater at birth, and five were younger than 4 days of age when the polymicrobial culture was obtained. Group D streptococci were recovered in eight cases (53%). Gastrointestinal foci appeared to be common among infants with late-onset polymicrobial infection (5 of 10), while prolonged rupture of membranes was frequently associated with early-onset infection (4 of 5). Though recovery of multiple organisms from blood or CSF may not always be significant, one should not immediately assume contamination. A report of more than one organism growing from a normally sterile body fluid in an intensive care nursery infant should be considered significant, and therapy should be adjusted to provide appropriate antimicrobial agents for all reported organisms if the infant has not substantially improved in the interval since the culture was actually obtained.

The impact of gestational age on dysmaturity/postmaturity.

The medical records of 403 infants admitted to the neonatal intensive care unit were reviewed. All were postterm (greater than or equal to 42 weeks' gestation) infants or infants who were full term (greater than or equal to 38 weeks' gestation) and had clinical diagnoses associated with the neonatal postmaturity/dysmaturity syndrome. Data collected from these 403 records were used to generate frequency distribution tables for a variety of obstetric and neonatal outcome variables. Regression analyses were used to assess associations among these variables and the presence or absence of fetal malnutrition (dysmaturity) or postdatism. Fetal distress and neonatal acidosis were associated with both dysmaturity and postdatism. Primigravidas, meconium-stained amniotic fluid, cesarean section, birth trauma, and neonatal death were associated with postdatism but not with dysmaturity. Preeclampsia, maternal smoking, oligohydramnios, low Apgar score, neonatal pulmonary hypertension, neurologic abnormalities, and a need for extracorporeal membrane oxygenation were associated with dysmaturity. No interaction between postdatism and dysmaturity was seen for any outcome variable. Postdatism and dysmaturity appear to contribute risk factors independently to infants admitted to the intensive care unit.

Gag-specific cytotoxic T lymphocytes from human immunodeficiency virus type 1-infected individuals: Gag epitopes are clustered in three regions of the p24gag protein.

Virus-specific cytotoxic T lymphocytes (CTL) may be an important host defense mechanism in the control of virus replication in persons infected with human immunodeficiency virus type 1 (HIV-1). Cytotoxic T-cell lines generated by nonspecific stimulation with anti-CD3 monoclonal antibodies and interleukin 2 were used to identify regions within the HIV-1 Gag protein that are the most frequently recognized. Using autologous Epstein-Barr virus-transformed target cells infected with recombinant vaccinia viruses encoding p18gag, p24gag, and p55gag proteins of HIV-1/Lai or selected truncations of p24gag, we show that within a group of 29 infected subjects, the p24gag protein is the target of Gag-specific CTL in most donors. Using autologous Epstein-Barr virus-transformed target cells coated with different synthetic peptides spanning the Gag amino acid sequence, we found clusters of partially overlapping peptides in three conserved regions of the p24 protein (amino acids [aa] 169 to 192, aa 219 to 304, and aa 335 to 372) that are frequently recognized by CTL and presented by a variety of human leukocyte antigen class I molecules. Since there are experiments both in vitro and in vivo showing the role of CTL in the control of virus replication in HIV and simian immunodeficiency virus infections, these results may be particularly important for vaccine development.