Philip W. Lowry

Mycobacterium chelonae causing otitis media in an ear-nose-and throat practice

Seventeen cases of otitis media caused by Mycobacterium chelonae were detected among patients seen at a single ear-nose-and-throat (ENT) office (Office A) in Louisiana between May 5 and September 15, 1987. All the patients had a tympanotomy tube or tubes in place or had one or more tympanic-membrane perforations, with chronic otorrhea that was unresponsive to standard therapy with antimicrobial agents. Middle-ear exploration in six patients revealed abundant granulation tissue; multiple granulomas and acid-fast bacilli were demonstrated on a section of tissue from one patient with a nonhealing mastoidectomy incision. Thirteen of the 14 ear isolates obtained from patients seen in Office A had the same unusual pattern of high-level resistance to aminoglycosides. M. chelonae and other nontuberculous mycobacteria were recovered from several sources of water in Office A, as well as in another ENT office (Office B) in a neighboring city that was visited by the index patient. Only one additional case was detected in Office B during the same period. Otologic instruments in Office A were cleaned in an ultrasonic bath with tap water and a liquid detergent; the contents of the bath were changed only once weekly. Instruments in Office B were placed in boiling water between patient examinations. This outbreak establishes M. chelonae as an agent of otitis media and underscores the need for high-level disinfection or sterilization of ENT instruments between examinations to prevent the transmission of this organism to patients in the office setting.

Nosocomial legionellosis: A review of pulmonary and extrapulmonary syndromes

Surgical patients appear to be at highest risk for acquisition of nosocomial Legionella pneumonia; most appear to become infected during respiratory tract manipulation and mechanical ventilation. Although the lungs are the most common site of nosocomial Legionella infection, an important subset of patients have infection at extrapulmonary sites. We describe 22 cases of extrapulmonary legionellosis reported in the literature. Most of these patients were surgical patients; more than half did not have serious underlying illnesses, and only five (23%) were receiving immunosuppressive agents. A total of 13 extrapulmonary sites of infection were reported, many in the absence of clinical pneumonia; these infections included sinusitis, hip wound infection, and prosthetic valve endocarditis. Five patients (23%) had fatal infections; in four of these cases diagnosis of Legionella infection was made after death, underscoring the need for a high index of clinical suspicion. A large percentage of extrapulmonary Legionella infections may result from direct topical exposure of susceptible tissue to contaminated tap water. Use of tap water must be carefully monitored, particularly in dressing changes and bathing of surgical patients.

Immunity in strain 2 guinea-pigs inoculated with vaccinia virus recombinants expressing varicella-zoster virus glycoproteins I, IV, V or the protein product of the immediate early gene 62

The immunogenicity of specific varicella-zoster virus (VZV) proteins, with emphasis upon cell-mediated immune responses, was evaluated by immunizing strain 2 guinea-pigs with vaccinia virus recombinants that express gpI (vac-gpI), gpIV (vac-gpIV) and gpV (vac-gpV) or the IE-62 protein (vac-IE-62). Vac-gpI elicited the highest initial mean T cell proliferation response [stimulation index (S.I.) 3.8 +/- 0.9 S.E.M.] whereas inoculation with vac-gpV produced the lowest primary T cell response (S.I. 2.5 +/- 1.1 S.E.M.). T cell proliferation was detected for a shorter period after immunization with vac-gpV compared to vac-gpI, vac-gpIV or vac-IE-62. A comparison of the immunogenicity of vac-gpI and vac-IE-62 with the same proteins prepared by immunoaffinity purification showed that immunization with these proteins in either form elicited virus-specific IgG antibodies and T cell recognition. The presence or absence of IgG antibodies to the IE-62 protein was used to assess protection against challenge with guinea-pig cell-adapted infectious VZV in animals that had been inoculated with vac-gpI, vac-gpIV or vac-gpV. Immunization with vac-gpI and vac-gpIV restricted VZV replication but all animals given vac-gpV developed antibodies to IE-62 after challenge with infectious VZV. Priming of the T lymphocyte response was observed in all animals immunized with VZV-vaccinia virus recombinants after subsequent exposure to infectious VZV. These experiments with VZV vac-gpI, vac-gpIV and vac-gpV in guinea-pigs suggest variability in the capacity of herpesviral glycoproteins to elicit cell-mediated immunity in vivo. Induction of virus-specific immunity using IE-62 means that this major tegument protein of VZV could be a useful component for vaccine development.

The synthesis and immunogenicity of varicella-zoster virus glycoprotein E and immediate-early protein (IE62) expressed in recombinant herpes simplex virus-1.

In order to evaluate the conditions for optimal expression and immunogenicity of varicella-zoster virus (VZV) proteins in a herpes simplex virus-1 (HSV-1) vector, we selected the VZV glycoprotein E (gE), encoded by ORF 68 and the VZV product of ORF 62, an immediate-early major tegument protein (IE62). Three HSV/VZV recombinants were generated: (1) VZV gE protein coding sequences along with the promoter region were inserted into the thymidine kinase (TK) gene of HSV-1 strain KOS; (2) VZV gE expressed from the HSV-1 ICP4 promoter was inserted into the glycoprotein C (gC) gene of HSV-1 strain F; and (3) VZV IE62 protein coding sequences under the control of the HSV-1 ICP4 promoter were inserted into the gC gene of HSV-1 strain F. Immunoblot analysis and immunoperoxidase staining of infected cell monolayers demonstrated vector expression of VZV proteins. Following intracranial inoculation in mice, both VZV gE-HSV (TK) and VZV IE62-HSV (gC) induced an IgG response against VZV gE or VZV IE62. When tested in cytotoxicity assays using T-lymphocytes from VZV immune human donors, the range of precursor frequencies for T-lymphocytes that recognized VZV gE or VZV IE62 was similar whether these proteins were expressed by HSV-1 or a vaccinia vector. These experiments demonstrate that HSV-1 is a competent vector for expression of these VZV proteins and support the feasibility of engineering a combined vaccine for these closely related alpha-herpesviruses.