We recently evaluated several tissue culture model systems for the study of invasion and intracellular multiplication of Mycobacterium tuberculosis. These model systems include a human alveolar pneumocyte epithelial cell line, a murine macrophage cell line (J774), and fresh human peripheral blood-derived macrophages. Our data indicated that the initial level of association of M. tuberculosis with human alveolar pneumocyte cells (2%) was less than that observed with fresh human peripheral blood macrophages (9%) or J774 murine macrophages (13%) within 6 h of the addition of the bacteria. M. tuberculosis replicated in association with the pneumocyte cells by more than 55-fold by day 7 postinfection. In contrast, total bacterial growth in the J774 cells and human macrophages was considerably less, with increases of only fourfold and threefold, respectively, over the same 7-day period. Amikacin, an aminoglycoside antimicrobial agent, was added to inhibit the growth of extracellular bacteria after the initial 6-h infection period. Decreases in viable counts were observed in all three cell cultures within the first 3 days after infection. However, unlike the case with either macrophage culture, intracellular bacterial CFU obtained from the infected pneumocytes increased by fourfold by day 7 after the addition of amikacin. These data indicate that M. tuberculosis infects and multiplies intracellularly in human lung epithelial cells and that these cells may be an alternative in vitro model for the study of intracellular multiplication of M. tuberculosis in the human lung.

No abstract

The predicted protein domains coded by exons 9–12 and 19–23 of the 27 exon cystic fibrosis transmembrane conductance regulator (CFTR) gene contain two putative nucleotide‐binding fold regions. Analysis of CFTR mRNA transcripts in freshly isolated bronchial epithelium from 12 normal adult individuals demonstrated that all had some CFTR mRNA transcripts with exon 9 completely deleted (exon 9‐ mRNA transcripts). In most (9 of 12), the exon 9‐ transcripts represented less than or equal to 25% of the total CFTR transcripts. However, in three individuals, the exon 9‐ transcripts were more abundant, comprising 39, 62 and 66% of all CFTR transcripts. Re‐evaluation of the same individuals 2–4 months later showed the same proportions of exon 9‐ transcripts. Of the 24 CFTR alleles in the 12 individuals, the sequences of the exon‐intron junctions relevant to exon 9 deletion (exon 8‐intron 8, intron 8‐exon 9, exon 9‐intron 9, and intron 9‐exon 10) were identical except for the intron 8‐exon 9 region sequences. Several individuals had varying lengths of a TG repeat in the region between splice branch and splice acceptor consensus sites. Interestingly, one allele in each of the two individuals with 62 and 66% exon 9‐ transcripts had a TT deletion in the splice acceptor site for exon 9. These observations suggest either the unlikely possibility that sequences in exon 9 are not critical for the functioning of the CFTR or that only a minority of the CFTR mRNA transcripts need to contain exon 9 sequences to produce sufficient amounts of a normal CFTR to maintain a normal clinical phenotype.

ADP-ribosylation factors (ARFs) are small guanine nucleotide-binding proteins that enhance the enzymatic activities of cholera toxin. Two ARF cDNAs, ARF1 and ARF3, were cloned from a human cerebellum library. Based on deduced amino acid sequences and patterns of hybridization of cDNA and oligonucleotide probes with mammalian brain poly(A)+ RNA, human ARF1 is the homologue of bovine ARF1. Human ARF3, which differs from bovine ARF1 and bovine ARF2, appears to represent a newly identified third type of ARF. Hybridization patterns of human ARF cDNA and clone-specific oligonucleotides with poly(A)+ RNA are consistent with the presence of at least two, and perhaps four, separate ARF messages in human brain. In vitro translation of ARF1, ARF2, and ARF3 produced proteins that behaved, by SDS/PAGE, similar to a purified soluble brain ARF. Deduced amino acid sequences of human ARF1 and ARF3 contain regions, similar to those in other G proteins, that are believed to be involved in GTP binding and hydrolysis. ARFs also exhibit a modest degree of homology with a bovine phospholipase C. The observations reported here support the conclusion that the ARFs are members of a multigene family of small guanine nucleotide-binding proteins. Definition of the regulation of ARF mRNAs and of function(s) of recombinant ARF proteins will aid in the elucidation of the physiologic role(s) of ARFs.