Twenty-six Pseudomonas aeruginosa strains from patients with cystic fibrosis were typed by the Fisher immunotyping scheme. Only 6 strains were agglutinated by a single typing serum, whereas 15 strains were agglutinated with more than one serum and 5 were not agglutinated by any serum. Neither the polyagglutinable nor the nonagglutinable strains were typable by hemagglutination inhibition or immunodiffusion, suggesting that these polyagglutinable strains did not express multiple serotype antigens, but were instead being agglutinated by antibody to nonserotype determinants. Four typable isolates were resistant to pooled normal human serum, whereas the 12 polyagglutinable and nonagglutinable isolates studied were very sensitive to normal human serum. The outer membranes of 16 strains were isolated and characterized. The data suggested, in general, strong conservation of outer membrane protein patterns. Lipopolysaccharides (LPS) were purified by a new technique which allowed isolation of both rough and smooth LPS in high yields. Three of four typable, serum-resistant strains examined had amounts of smooth, O-antigen-containing LPS equivalent to our laboratory wild type, P. aeruginosa PAO1 strain H103. In contrast, 10 of 12 polyagglutinable or nonagglutinable, serum-sensitive strains had very little or no smooth, O-antigen-containing LPS, and the other two contained less smooth LPS than our wild-type strain H103. In agreement with this data, five independent, rough, LPS O-antigen-deficient mutants of strain H103 were nontypable and serum sensitive. We suggest that the LPS defects described here represent a significant new property of many P. aeruginosa strains associated with cystic fibrosis.

Three spontaneously arising rough mutants of Pseudomonas aeruginosa have been isolated by selection for resistance to virulent lipopolysaccharide (LPS) specific bacteriophages. In addition, the first phages specific for rough mutants of P. aeruginosa were isolated. Using these phage and autoagglutination patterns in 4% NaCl and acriflavine, these mutants could be clearly distinguished from the wild-type strain and each other. Chemical analysis of the LPS together with chromatographic resolution of the polysaccharide moieties showed alterations in both O-specific side chains and core regions.

Alpha(1)-aderenoceptor-mediated constriction of rabbit inferior vena cava (IVC) is signaled by asynchronous wavelike Ca(2+) oscillations in the in situ smooth muscle. We have shown previously that a putative nonselective cationic channel (NSCC) is required for these oscillations. In this report, we show that the application of 2-aminoethoxyphenyl borate (2-APB) to antagonize inositol 1,4,5-trisphosphate (InsP(3))-sensitive Ca(2+) release channels (IP(3)R channels) can prevent the initiation and abolish ongoing alpha(1)-aderenoceptor-mediated tonic constriction of the venous smooth muscle by inhibiting the generation of these intracellular Ca(2+) concentration ([Ca(2+)](i)) oscillations. The observed effects of 2-APB can only be attributed to its selective inhibition on the IP(3)R channels, not to its slight inhibition of the L-type voltage-gated Ca(2+) channel and the sarco(endo)plasmic reticulum Ca(2+) ATPase. Furthermore, 2-APB had no effect on the ryanodine-sensitive Ca(2+) release channel and the store-operated channel (SOC) in the IVC. These results indicate that the putative NSCC involved in refilling the sarcoplasmic reticulum (SR) and maintaining the tonic contraction is most likely an SOC-type channel because it appears to be activated by IP(3)R-channel-mediated SR Ca(2+) release or store depletion. This is in accordance with its sensitivity to Ni(2+) and La(3+) (SOC blockers). More interestingly, RT-PCR analysis indicates that transient receptor potential (Trp1) mRNA is strongly expressed in the rabbit IVC. The Trp1 gene is known to encode a component of the store-operated NSCC. These new data suggest that the activation of both the IP(3)R channels and the SOC are required for PE-mediated [Ca(2+)](i) oscillations and constriction of the rabbit IVC.

Receptors for phages specific to Pseudomonas aeruginosa strain PAO were studied. Phages 16, 44, 109, F8, and PBI are lipopolysaccharide (LPS) specific as shown by neutralization tests. The PhI50's of the LPS, adsorption rate constants with strain PAO and the plaque morphologies of these five phages were quite similar. Phages 1214 and 7 also appear to be LPS-specific on the basis of host-range studies. Phage 73 is pilus-specific, while phages 21 and 68 fall into a group which does not attach to pili, flagella, or LPS. A theoretical approach to the interpretation of phage-cell interactions is presented.