[Frontiers in Bioscience 2, d588-591, December 1, 1997]
DETERMINANTS OF ORGAN TROPISM OF SENDAI VIRUS
M. Tashiro1 and J.T. Seto2
1Department of Virology 1, National Institute of Infectious Diseases, Tokyo, Japan, 162 and 2 Department of Microbiology, California State University, Los Angeles, Los Angeles, California, 90032-8201
Received 11/10/97 Accepted 11/14/97
3. SENDAI VIRUS
3.1 Genome and proteins
The genome of Sendai virus, a negative strand RNA virus, consists of 15,384 nucleotides with the gene order of 3'NP-P/C/V-M-F-HN-L-5' (13,14,15). The genome codes for the nucleocapsid (NP), phosphoprotein (P), matrix (M), fusion (F), hemagglutinin-neuraminidase (HN), and large (L) proteins, respectively. The entire genome of Sendai virus (Z strain), of the two host range mutants (ts-f1 and F1-R), and a revertant (T-5) derived from the pantropic mutant (F1-R) has been sequenced (13,14,15,16).
3.2 Molecular biology
In search of the molecular basis of the virulence of paramyxoviruses, the focus has been on the role of the F glycoprotein. In infected tissue cultures multiple cycles of viral replication are the consequence of proteolytic cleavage of the F protein (1). Activation of the viral infectivity is dependent upon post-translational cleavage of the glycoprotein precursor Fo to F 1 and F 2 (4,5). For wild-type Sendai virus cleavage in vivo is only by tryptase Clara, a trypsin-like host protease secreted by Clara cells that are restricted to the respiratory epithelium (8,9). Arginine is the residue at the cleavage site of wild-type virus F protein. The F protein of the pantropic mutant, F1-R, has a proline residue next to the cleavage site (17) which renders the protein cleavable by ubiquitous host protease(s). Therefore, cleavage activation and mutiple cycles of replication of the wild-type virus occur exclusively in the lungs, whereas, for F1-R, they occur in many organs. Thus, the primary determinant of pantropism is proteolytic cleavability and the organ distribution of the proteases (12,18,19).
3.3 Budding domain
In addition to proteolytic cleavage of the F glycoprotein, the budding polarity of F1-R has been proposed to be a determinant of pantropism. Another important difference between F1-R and wild-type Sendai virus has been identified. The mutant was shown to bud bipolarly, at the apical and basolateral domains of the plasma membrane, in the bronchial epithelial cells of mice and in polarized MDCK cells, whereas budding of wild-type virus was at the apical domain (18,19). Specific viruses have been shown to be targeted to the apical domain and other viruses to the basolateral domain (20). For example, paramyxoviruses and influenza viruses bud at the apical domain whereas HIV and vesicular stomatitis viruses bud at the basolateral surfaces. The latter viruses are systemic agents. Systemic mutants of influenza virus bud at the apical domain. It has been postulated that budding at the basolateral domain facilitates the spread of virus into subepithelial tissues and gaining access to the peripheral blood (18, 19, 21). The differential budding behavior of wild-type and F1-R viruses may explain, in part, why infection by wild-type virus is localized in the respiratory tract whereas infection by the F1-R mutant virus becomes systemic, presumably by spreading to distant organs via the blood.
3.4 Nucleotide sequence
By comparative nucleotide sequence analysis of the F gene of the host range mutants (ts-f1 and F1-R), six mutations were revealed in the F gene, two in the M gene, none in the HN gene, one in the NP gene of ts-f1, one in the P gene of F1-R, and one in the L gene of F1-R. Of significance is the predicted amino acid substitutions in the F protein at residues 115 (Ser to Pro) and 116 (Arg to Lys) at the site of cleavage of the F protein. One mutation was found in a revertant, T-5, derived from F1-R that was no longer pantropic. It was at residue 115 (Pro to Ser), which represents the mutation in the protein involved in the cleavability of the protein. This also rules out the possible role of the other mutations found in the F, P, and L genes of F1-R for pantropism.
3.5 Determinants of pantropism
Two determinants of pantropism have been suggested, the primary determinant of proteolytic cleavage and the secondary one as bipolar budding. One of the distinguishing phenotypes of F1-R was the abnormal migration of the M protein in SDS-PAGE gels. The protein migrated faster indicating a structural change in the M protein of F1-R. Two mutations in the M protein were at residues 128 (Asp to Gly) and 210 (Ile to Thr). These mutations have been postulated to cause bipolar budding of F1-R. Additionally, polarized transport of proteins in various epithelial cells depend upon the cytoskeletal system (22). It was shown that wild-type Sendai virus in polarized MDCK cells depend on intact microtubules, whereas in F1-R infected cells, the microtubules were disrupted (23).
To clarify which gene or mutations in F1-R was responsible for the disruption of microtubules leading to the altered budding phenotype, MDCK cells containing the F and M genes of wild-type and F1-R were established. Expression of the mutated F1-R M protein resulted in the formation of giant cells, bipolar transport of the F protein, and in the disruption of the microtubular network. This leads to the impairment of cellular polarity, bipolar transport of the F glycoprotein, and bipolar budding of the virus (24). The M protein plays an important role in the assembly process by binding the cytoplasmic tails of the glycoproteins with the nucleocapsid which is composed of the genomic RNA and NP, P, and L proteins (25).
Additional mutants with phenotypes similar to F1-R were derived from wild-type virus. This was to provide further evidence whether specific mutations in the F protein are solely responsible for the drastic change in infectivity, whether specific mutations in the M protein are responsible for bipolar budding, and whether bipolar budding of Sendai virus is a cofactor for pantropism.
Protease activation mutants that cleaved the F protein and mutants that bud bipolarly were isolated. Mutants that bud bipolarly, mutants with protease activation activity, and mutants with both phenotypes were characterized. Nucleotide sequence analysis of the F and M genes revealed deduced amino acid substitutions in the F and M proteins that were different from those of the respective proteins of F1-R, T-5 revertant of F1-R, and wild-type virus. The mutants did not cause systemic infections in mice and they did not disrupt the microtubules in MDCK cells (unpublished data).
These findings support the premise that one mutation (residue 115) in the F protein and two mutations (residues 128 and 210) in the M protein are the specific mutations required for the systemic infection of F1-R, and proteolytic cleavage of the F protein and bipolar budding must function in concert for the systemic infection.