CLINICAL MICROBIOLOGY REVIEWS, Apr. 1994, p. 265-275 Vol. 7, No.

2
0893-85 12/94/$04.00 + 0
Copyright ©C 1994, American Society for Microbiology

Biology of Parainfluenza Viruses

RAIJA VAINIONPAA* AND TIMO HYYPIA
Department of Virology, University of Turku, SF-20520 Turku, Finland

INTRODUCTION ........................................ 265

MOLECULAR BIOLOGY ........................................ 265
Structure .........................................265
Genome .........................................265
Proteins and Their Antigenic Structure ........................................ 267
Replication .........................................267
DIAGNOSIS .........................................268
CLINICAL AND EPIDEMIOLOGICAL ASPECTS ........................................ 268
PATHOGENESIS AND IMMUNE RESPONSE ........................................ 269
PREVENTION AND TREATMENT ........................................ 271
ACKNOWLEDGMENT ........................................ 272
REFERENCES .........................................272

INTRODUCTION cellular proteases to produce an active form of two disulfide-

linked subunits, F1 and F2. The fatty acid composition of the
Parainfluenza viruses 1 through 4 (PIVI through PIV4) are envelope is identical to that of the host cell. The matrix (M)
important human pathogens that cause upper and lower protein is highly hydrophobic in character, and its function in
respiratory tract infections, especially in infants and small the virion is to mediate interaction between the glycoproteins
children. Together with respiratory syncytial virus, PIVs are and the nucleocapsid. PIVs also contain cellular actin as a
the most frequent pathogens isolated in children with lower structural component; however, its function in virus structure
respiratory infections. In addition to the viruses infecting and replication is not fully understood (6, 39, 102, 111).
humans, PIVs include viruses that cause diseases in animals,
such as Sendai virus of mice, which has been used as a model Genome
for studying the course of PIV infection. Recent work has
revealed interesting aspects of the molecular biology of PIVs, The PIV genome is a linear, nonsegmented, negative-sense
and these findings can now be used in, for instance, develop- RNA molecule containing an average of 15,000 nucleotides
ment of new vaccine candidates. (nt) (15,463 nt in PIV3 [31]; 15,285 nt in Sendai virus [93];
15,156 nt in Newcastle disease virus (NDV) [116]) and consists
MOLECULAR BIOLOGY of six or seven genes that encode eight or nine proteins, some
of which are not detected in the virion but participate in the
Structure virus replication cycle in the cell. The genomic RNA is never
PIVs are typical members of the family Paramyxoviridae and found in naked form in the cells but is always tightly bound to
NP.
belong to the genus Paramyxovirus (Table 1). A number of The organization of the parainfluenza virus genome is
extensive reviews have recently been published concerning
their structure and replication (12, 30, 31, 60). These RNA 3'-NP-P/CNV-M-F-(SH)-HN-L-5'. The presence or absence of
the SH gene between the two glycoprotein genes divides
viruses are pleomorphic enveloped particles that are 150 to 300 parainfluenza viruses into two groups. The first group (includ-
nm in diameter. A structural model of the virion is shown in
ing PIV3, Sendai virus, and NDV) does not contain the SH
Fig. 1. The virion consists of a filamentous, herringbone-like gene, whereas the second group (e.g., the animal parainfluenza
nucleocapsid core surrounded by a lipid envelope with virus- virus, simian virus 5, and mumps virus, which is also a member
specific glycoprotein spikes. The nucleocapsid is composed of of the family Paramyxoviridae) does. This SH gene encodes a
the genome, a single-stranded RNA molecule of negative- small integral membrane protein, which is highly hydrophobic
sense polarity, which is tightly coated with the nucleocapsid
and is found only in infected cells. Its function is still unclear.
protein (NP). The nucleocapsid structure also contains two Figure 2 shows a schematic representation of the PIV genome
other proteins, the phosphoprotein (P protein) and the large and its replication cycle.
protein (L protein), which occur discontinuously as clusters. Extensive sequence analysis of PIV3, Sendai virus, and NDV
The NP is the most abundant protein in the virion, and it is genomes carried out during the past few years (10, 18, 26,
believed to be responsible, together with the P and L proteins, 32-35, 42, 55, 64, 65, 68, 70, 81, 90, 93, 96-99, 116) has shown
for RNA-dependent RNA polymerase activity (47). that the 3' and 5' ends of these genomes contain the short,
The viral envelope contains two virus-specific glycoproteins: extracistronic noncoding regions, approximately 50 and 40 nt
the hemagglutinin-neuraminidase (HN) and the fusion (F) long, respectively. The first 20 nt at the 3' end are highly
proteins. The F protein is synthesized as a biologically inactive conserved among paramyxoviruses (23, 35, 92, 93, 116), sug-
precursor form (Fo), which is cleaved posttranslationally with gesting their biological significance in initiation of mRNA
transcription and genome replication. The adjacent genes in
*
Corresponding author. Mailing address: Department of Virology, the genome are separated by intercistronic regions, which
University of Turku, Kiinamyllynkatu 13, SF-20520 Turku, Finland. contain a transcriptional control sequence found at the begin-
Phone: (358)-21-6337464. Fax: (358)-21-2513303. ning of every gene (gene start), a stop sequence (gene end) at
265
266 VAINIONPAA AND HYYPIA CLIN. MICROBIOL. REV.

TABLE 1. Classification of the family Paramyxoviridae Intercistronic regions

/ \
Genus Human viruses Animal virusesa /
/
\
\
I \ \ \
Paramyxovirus PIVi Sendai virus (murine) I
PIV2 Simian virus 5 (canine) _NCR NP 11! P /CIV!II M ! F \ (SHV)\ HN \\ L NCR
PIV3 Bovine parainfluenza virus 3.' rI II -I 11 I5.
PIV4 PIV - RNA genome
Newcastle disease virus
(avian)
Mumps virus
Morbillivirus Measles virus 9
Canine distemper virus
Phocine distemper virus mRNAs b new viral antigenom e (+)RNA
Peste-des-petits-ruminats proteins
virus (caprine, ovine) we~~~~~~~~~~~~"Z
/o,
Rinderpest virus (bovine,
caprine, ovine, porcine) 0%Q
Pneumovirus Respiratory Bovine respiratory syncytial
syncytial virus virus new genomes
Pneumonia virus of mice
(rodent)
Turkey rhinotracheitis virus
(avian) ossembly of virions

"Animal viruses most closely related to a human pathogen are shown on the FIG. 2. Schematic representation of the PIV genome and the
same line. replicative cycle.

the end of every gene, and a short sequence between the genes. dependent G residues at a precise site in the mRNA molecule;
The control regions are conserved (42, 54, 63, 96, 116) and are the nonstructural V protein is encoded directly by the genomic
transcribed to mRNAs, whereas the short intergenic sequences sequence (61, 76, 78, 95, 108). A similar editing mechanism has
are not copied. been described for measles virus (9), which is also a member of
The first gene at the 3' end of the genome is the NP gene, the Paramyxoviridae, suggesting that this phenomenon has a
which is highly conserved among PIVs (32, 54, 55, 90, 92). The more general biological significance. The mechanism respon-
length of the monocistronic NP gene of PIV3 is in the range of sible for addition of the nontemplate residues is unknown, nor
1,641 to 1,850 nt, and it encodes a 515-amino-acid protein (32, is it known whether it occurs cotranscriptionally or posttran-
55, 90). scriptionally, but virus-encoded proteins are probably needed
The P gene is a multicistronic genome region with a unique for this process (36, 77, 108). The P and V proteins are amino
coding strategy. The P genes have continuous, discontinuous, coterminal and share 164 amino acids before the sequences
and overlapping cistrons (36), and, according to their coding diverge. The V proteins have a cysteine-rich region, which is
strategy, PIVs can be divided in two groups. In the first group, similar to the metal-binding domains present in proteins
PIV1 and PIV3 encode their 500- to 600-amino-acid P proteins known to bind to DNA or RNA. This cysteine-rich region has
from continuous P cistrons. The second start codon in the P been found in all paramyxovirus P genes, suggesting that it is
gene region initiates a shorter overlapping open reading frame biologically important (78).
coding for a smaller, nonstructural C protein (33, 65, 66, 97). The M gene is the most highly conserved gene among
The proportional quantities of the P and C proteins produced paramyxoviruses (34, 64, 99), and the homology is most
are evidently controlled at the level of translation. The C extensive at the carboxy-terminal third of the protein se-
protein (about 204 amino acids) is not found within the virion quences. Most conserved regions represent hydrophobic re-
but is detected in infected cells. gions, which may be structurally involved in membrane inter-
The second group, which includes PIV2, PIV4A, and actions.
PIV4B, produces more than one mRNA species by editing the The PIV3 F gene is 1,845 to 1,851 nt long and encodes a
P mRNA with insertion of two additional, non-template- 539-amino-acid protein (18, 98). The F gene sequences re-
ported for a number of clinical PIV3 isolates show a very high
degree of conservation at the nucleotide (>94%) and amino
acid (>97%) levels (82). The F protein contains two hydro-
phobic regions and a signal sequence, which is later removed
through a proteolytic cleavage. Comparison of the F genes of
Sendai virus and NDV indicates a close evolutionary relation-
ship between these human and animal pathogens (5, 10, 73,
93).
The HN genes of PIV3, Sendai virus, and simian virus 5
encode a 565- to 572-amino-acid protein containing one major
hydrophobic region and no signal sequence (4, 26, 51). The
highest homology, 62% at the amino acid level, occurs between
PIV3 and Sendai virus (73).
The L gene is located at the 5' end of the genome and is the
largest PIV gene (6,799 nt). It represents about 40% of the
FIG. 1. Structural model of PIV. coding capacity of the entire genome (72, 93, 116). Because of
VoL. 7, 1994 BIOLOGY OF PARAINFLUENZA VIRUSES 267

TABLE 2. Structural proteins of PIVs

Protein Designation Chemical nature Location Function

Hemagglutinin-neuraminidase HN Glycosylated Envelope Attachment to host cell receptors,

hemagglutinating and neuraminidase activities
Fusion protein F1F, Glycosylated Envelope Fusion, hemolysis, penetration
Nucleoprotein NP Phosphorylated Nucleocapsid Tightly complexed with genomic RNA, structural
component in nucleocapsid
Phosphoprotein P Phosphorylated Nucleocapsid Component of polymerase complex
Large protein L Phosphorylated Nucleocapsid Component of polymerase complex
Matrix protein M Phosphorylated Inside the envelope Assembly

the length of the gene and its location, the number of L exhibits extensive homology (70%) with that of Sendai virus in
mRNAs is the smallest among the transcripts. the first 424 amino-terminal amino acids, but no homology is
observed between the 95 carboxy-terminal amino acids (90).
Proteins and Their Antigenic Structure Moreover, NP contains two other identical, highly conserved
regions, suggesting their structural or functional importance.
The PIV genomes encode six structural proteins, which are Like other internal components of the virions, M, P, and L
found in virions, and two (or more) nonstructural proteins, proteins, NP would not be expected to induce a good humoral
which are detected only in infected cells. Table 2 summarizes responses. On the other hand, these proteins may act in the
the chemical nature, location in the virion, and functions of the induction of T-cell immunity, as has been shown with Sendai
structural proteins. virus and measles virus NPs (22, 89).
The HN glycopolypeptide is a receptor-binding protein that The P and L proteins are crucial components in RNA-
mediates virus-cell attachment. It is anchored to the envelope dependent RNA polymerase activity. P protein is the most
by the hydrophobic region near its amino terminus (4, 51, 70), extensively phosphorylated protein in the virion structure,
whereas the F glycoprotein is anchored near its carboxy although NP and L protein are also phosphorylated to some
terminus. The HN protein has hemagglutinating and neur- extent. L protein is the largest virus-specific protein and is
aminidase activities. Its putative three-dimensional structure is believed to be responsible, together with P protein, for tran-
thought to be broadly similar to that of influenza virus neur- scriptase activity as well as for capping, methylation, and
aminidase, but the framework residues of the active site are polyadenylation of virus-specific mRNAs.
less strictly conserved (17). By using recombinant vaccinia The hydrophobic M protein is found in close association
viruses expressing the HN and F glycoproteins of PIV3, it has with the envelope. M protein plays a critical role in the
been shown that the HN protein is also required for complete structure of the virion by mediating interaction between the
fusion activity (25). nucleocapsid core and the surface glycoprotein spikes. The
The PIV3 HN protein contains six antigenic sites (A to F), attachment of the nucleocapsid to glycoproteins through the M
most of which are conformational (15, 16, 106). Three of these protein is thought to start the budding process of the virus
sites (A, B, and C, together containing 11 epitopes) are particle through the plasma membrane; this is the final step in
involved in neutralization and hemagglutination activities; in virion maturation. Tight association of M protein with the
PIV3, these regions are highly conserved among clinical iso- nucleocapsid structure in infected cells suggests that it may
lates (14, 88, 104). However, in PIVI, HN protein genetic also have a regulatory role in transcription and/or replication.
variation and evolution occur (49, 50). On the other hand, the The fact that the M protein is highly conserved among PIV
HN proteins of Sendai virus and NDV contain only four proteins also indicates that it plays an important role in virus
antigenic sites, but antibodies against all of them are needed replication (34, 99). Furthermore, it has been speculated that
for complete neutralization of the virus (52, 53, 75, 79). In decreased synthesis or stability of M protein plays a role in the
addition, glycosylation of the protein is important for antige- establishment and maintenance of persistent infections.
nicity. Monoclonal antibodies to site A epitopes also inhibit
neuraminidase activity, suggesting that site A is located on the Replication
PIV3 HN molecule in a region topographically close to the site
of neuraminidase activity (14, 15). Replication of RNA viruses takes place in the cytoplasm of
The F glycoprotein is involved in hemolysis, cell fusion, and the host cell, and the mature virus particles are released by
virus penetration into the cell. It is a disulfide-linked protein budding through the plasma membrane. The first step in the
composed of a distal polypeptide, F2, and a larger, proximal infection cycle is the attachment of the virus particle to its
polypeptide, Fl, anchored to the envelope near its C-terminal cellular receptor, a process mediated by the HN glycoprotein
end (10). It induces neutralizing antibodies, and the PIV3 F in PIVs. The F protein catalyzes subsequent fusion of the virus
protein contains eight antigenic sites. The existence of identi- envelope and cell membrane. After the uncoating event, the
cal changes in F epitopes in naturally occurring PIV3 strains nucleocapsid structure containing the genome is released into
raises the possibility that mutations arise easily (104, 105). the cytoplasm.
NP is the most abundant component in the nucleocapsid The genome of negative-sense RNA viruses cannot function
structure. It is tightly bound to genomic RNA, having an directly as mRNA but must first be transcribed to virus-specific
essential role in the stability of the nucleocapsid. NP also mRNA species by RNA-dependent RNA polymerase. Because
interacts with the P and L proteins in the nucleocapsid eukaryotic cells do not contain this enzyme, the virus nucleo-
structure, but the precise role of NP in transcription and capsid structure has to transport the activity into cells. The
replication is not completely understood. It has been proposed viral P and L proteins, and possibly also NP, are needed for this
that this polypeptide might function as a switching factor in the activity. During primary transcription, the genomic RNA is
change from transcription to replication. The NP of PIV3 sequentially transcribed starting from the 3' end of the genome
268 VAINIONPAA AND HYYPIA CLIN. MICROBIOL. REV.

to produce individual mRNA species. According to the Kings- incubated with a europium-labeled detector antibody for 1 h in
bury-Kolakofsky model of replication, the polymerase has a microtiter strips previously coated with the capture antibody.
single site of attachment to the genomic RNA, and it copies After the strips are washed, the result is measured with a
the viral genes one by one (60). At the gene boundaries the time-resolved fluorometer.
transcriptase moves to the next gene without copying the short A more recent development for detection of viruses in, e.g.,
intergenic sequences, and at each of the gene junctions the nasopharyngeal samples is the use of PCR. In this assay as
transcription efficiency decreases. Viral mRNA molecules are applied to the detection of PIVs, DNA complementary to viral
modified to contain a 5'-end methylated cap structure and a RNA is first generated by using reverse transcriptase and an
polyadenylated 3' end resembling eukaryotic messengers. The oligonucleotide primer. This product is further amplified in
mRNAs are translated subsequently to full-size proteins on repeated cycles of DNA synthesis, initiated with a specific
host cell ribosomes, except for the F protein, which is synthe- primer pair and catalyzed by a heat-stable DNA polymerase.
sized as a precursor (FO) and later cleaved to its active form The PCR product can then be detected by agarose gel elec-
(F1F2) by host cell proteolytic enzymes. Proteolytic cleavage of trophoresis or nucleic acid hybridization methods or used for
the inactive precursor protein F0 is important for virus infec- direct sequencing, thus allowing exact comparisons of isolated
tivity, because cells lacking this activity are not able to support strains for epidemiological purposes.
virus replication. Thus proteolytic activation is a determinant In a PCR approach, Karron et al. (58) used primers that
for PIV tissue tropism and pathogenesis. The glycoproteins are detected a sequence at the 5' noncoding region of the PIV3 F
processed (e.g., glycosylation, attachment of fatty acids) during protein gene. Specimens obtained from 10 children during a
their complex transportation process through the Golgi com- hospital outbreak were studied by sequence analysis after the
plex to the plasma membrane. PCR. The results showed that four different strains were
Genome replication takes place in two steps. First the present among the 10 isolates. Six isolates which represented
negative-strand RNA is copied to a complementary positive- one strain were obtained from a cluster of nosocomial cases in
sense RNA, and then this molecule functions as a template for pediatric intermediate care unit. The presence of the other
the synthesis of genomic RNA. In genome replication, the three strains in the community indicated that multiple strains
polymerase must ignore the transcription stop signals at the can be found during a single epidemic.
gene boundaries in order to make full-length genomic RNA. Serological assays based on hemagglutination inhibition
On the basis of studies with Sendai virus, it has been argued have been widely used for the detection of antibody responses
that the concentration of NP regulates the activities of poly- against PIVs (11). In these assays, removal of nonspecific
merase, with an excess favoring production of full-length inhibitors is needed for optimal results. More recently, enzyme
genome copies by a read-through mechanism of the replicase. immunoassays have been developed for PIV serological assays.
On the other hand, NP scarcity favors production of monocis- These tests are based either on the demonstration of an
tronic mRNAs (109). Figure 2 represents a schematic model of increase in the levels of specific immunoglobulin G (IgG)-class
the PIV genome and its replicative cycle. The relative gene antibodies in serum samples collected during the convalescent
sizes have been drawn to scale. phase of the disease over those in samples collected during the
acute phase or on the detection of specific IgM present during
DLAGNOSIS the acute phase of infection. The former approach is in routine
use in our laboratory for PIV1, PIV2, and PIV3 (110). Serum
Although some clinical signs and symptoms (e.g., croup [see specimens are tested at a single dilution, and the results are
below]) are more frequent in illnesses caused by PIVs than in expressed by comparison with a standard serum. In paired sera
other respiratory infections, specific etiological diagnosis al- from patients with virologically confirmed PIV infection, in-
ways requires detection of infectious virus or virus components creases in IgG antibody levels were detected in 69 to 87% of
in clinical samples or a serological response. PIVs grow and the patients while specific IgM antibodies were present in 42%.
produce a syncytial cytopathic effect in certain cell lines,
enabling isolation of viruses from specimens taken from the CLINICAL AND EPIDEMIOLOGICAL ASPECTS
respiratory tract of patients. Primary rhesus monkey kidney
(MK) cells were once the only choice for growing PIVs. The The clinical diseases caused by PIVs include rhinorrhea,
LLC-MK2 cell line is now also widely used for PIV isolation cough, croup (laryngotracheobronchitis), bronchiolitis, and
(28), and a continuous line of mucoepidermoid human lung pneumonia. When 99 Finnish children with virologically con-
carcinoma cells, NCI-H292 (8), has recently been shown to be firmed PIV infection were studied, laryngitis was the clinical
equivalent to MK cells. In culture, the presence of the virus diagnosis in 64, 86, and 21% of PIV1, PIV2, and PIV3
may be detected by hemadsorption with guinea pig erythro- infections, respectively (83). Upper respiratory tract infection
cytes (11) or by using specific immunological reagents. The was clinically diagnosed in 58% of the children with PIV3
isolation procedure can be accelerated by using the immuno- infection. Otitis was detected in 10 to 34% of children with PIV
peroxidase method to stain cultures with monoclonal antibod- infection while pneumonia was found in 12% of the children
ies 2 days after inoculation of the sample, when the cytopathic with PIV2 and 11% with PIV3 infections but not in those
effect is not always visible. This method has been used success- infected with PIV2. Bronchiolitis was not found in association
fully at the Department of Virology, University of Turku, for with PIV infections in this study. The maximum temperature of
other respiratory viruses (113) and recently also for PIV1, these patients was approximately 40°C. Slightly elevated or
PIV2, and PIV3 grown in NCI-H292 cells (112). normal C-reactive protein, leukocyte count, and erythrocyte
When more rapid demonstration of the virus is needed, sedimentation rate values were found in children with PIV
detection of viral antigens by immunoassays is recommended. infections (85).
Radioimmunoassays and enzyme immunoassays have been Murphy et al. (74) have summarized the virological findings
developed for the detection of PIV1, PIV2, and PIV3, using in pediatric inpatients with respiratory infections. In this
either polyclonal sera (91) or, more recently, monoclonal report, PIV3 and respiratory syncytial virus were identified as
antibodies in a highly sensitive one-step time-resolved fluoro- the two leading causes of serious lower respiratory tract
immunoassay (46). In the assay, the nasopharyngeal sample is disease. PIV1, PIV2, and PIV3 were primarily associated with
VOL. 7, 1994 BIOLOGY OF PARAINFLUENZA VIRUSES 269

croup, and PIV3, together with respiratory syncytial virus, was Patients with immunodeficiencies have a tendency to de-
a major cause of bronchiolitis and pneumonia. velop persistent and severe PIV and other respiratory virus
Sites of PIV infection other than the respiratory tract have infections (20, 56). Serious lower respiratory tract disease
been described. In rare cases, parotitis may occur. Meningitis, caused by PIVs has been reported in children and adults who
also a rare complication, has been reported in individual cases, undergo bone marrow transplantation (114). Of 1,253 trans-
and the etiology has also been confirmed by isolation of these plant recipients, 27 had PIV infection as demonstrated by
viruses from the cerebrospinal fluid (2). culture of the pathogen from nasopharyngeal or bronchoalveo-
In industrialized countries, the epidemiology of PIV infec- lar samples. Nineteen of the PIV-infected patients had lower
tions has been analyzed by several investigators. Surveillance respiratory tract involvement. Fatal respiratory failure devel-
of PIV3 infections in Houston, Tex., from 1975 to 1980 showed oped in six of these patients.
that most of the cases occurred during the late winter or spring It is known that infections caused by respiratory viruses are
(41). At least two-thirds of children were infected with PIV3 in often complicated by bacterial infections, such as otitis and
each of the first 2 years of life. The risk of illness was pneumonia. Korppi et al. (62) studied specimens collected
approximately 30 per 100 children per year and decreased after from 37 children with serologically confirmed PIV infection for
2 years of age. It is of interest that in the same region, PIV3 evidence of concomitant bacterial infection. They found no
infections had previously followed an endemic rather than an evidence of bacterial involvement in 24 children with croup,
epidemic pattern. PIV1 and PIV2 infections occurred in 2-year whereas in 3 of the 13 children with lower respiratory tract
cycles. In another study, carried out in a small community in infection, serological findings supported Streptococcus pneu-
Tecumseh, Mich., PIV1 and PIV2 infections occurred together moniae infection, and in 1 case, the findings supported involve-
in alternate years from 1976 to 1981 (71). The peak of PIV1 ment of Haemophilus influenzae.
infections was in October, and that of PIV2 infections was in
December; the monthly distribution of PIV3 infections was
more consistent. The viruses were isolated predominantly from PATHOGENESIS AND IMMUNE RESPONSE
children less than 2 years old. Analysis of virological data
concerning PIV3 infections over a 12-year period (1978 Primary inoculation of PIVs occurs through the nasal mu-
through 1989) in Sydney, Australia, showed that the peak cosal surface, and the first symptoms, including rhinorrhea, are
incidence was in the spring (21). observed after an incubation period of 2 to 4 days. The usual
Easton and Eglin (24) have reported a different seasonal course involves slow recovery after upper respiratory tract
occurrence of PIV3 infections in England and Wales over a involvement, but the illness is occasionally complicated by
10-year period from 1978 through 1987. They showed that otitis. In more severe cases, the infection spreads to the lower
PIV3 causes epidemics in the summer, which were detected respiratory tract, presumably through aspiration of secretions,
yearly, whereas respiratory syncytial virus infections occur in and bronchiolitis or pneumonia may follow (69). PIV3 has also
the winter. been isolated from the blood of children with acute respiratory
In Turku, Finland, 249 PIVI, 155 PIV2, and 812 PIV3 infection (84).
infections were diagnosed from 1981 to 1992 by using viral Ciliated epithelial cells of the respiratory tract are infected
antigen detection (Fig. 3). An epidemic of PIV1 infection (27), and a peribronchiolar infiltrate containing lymphocytes,
occurred in the winter of 1982; otherwise, only individual cases plasma cells, and macrophages appears, together with edema
(a maximum of nine cases per month, usually in the early and excess secretion of mucus. When pneumonia is predomi-
autumn) were observed. The findings concerning PIV2 infec- nant, a moderate hyperplasia of alveolar epithelia is observed
tions were similar. PIV3 caused small epidemics every year in and fluid containing some macrophages, erythrocytes, and
the spring. When compared with other respiratory viruses leukocytes accumulates (117). The mechanisms of cell damage
detected during the same period, the frequency of PIV infec- include both direct destruction by the virus and effects of the
tions is similar or lower (3,945 respiratory syncytial virus, 1,027 immune response (70). The latter may consist of formation of
influenza A virus, 400 influenza B virus, and 1,805 adenovirus antigen-antibody complexes, allergic injury due to IgE, cyto-
infections were diagnosed). toxic T cells, or delayed-type hypersensitivity. The release of
The epidemiology of PIV and other respiratory virus infec- soluble mediators may also play a role.
tions in the developing world has recently been investigated. Several animal models closely mimic the course of PIV
Ruutu et al. (86) studied specimens from 312 Filipino children infection in humans. The most widely used model is Sendai
(less than 5 years old), living in periurban slums and middle- virus infection in rodents, and a rat model is described here as
class housing, who fulfilled the criteria for acute lower respi- an example. Garlinghouse et al. (37) and Giddens et al. (38)
ratory tract infection. The etiological diagnosis was made on studied different parameters during infection of Sprague-
the basis of viral antigen detection, virus isolation, and sero- Dawley rats. When 5- to 8-week-old animals were infected
logical assays. A total of 198 virus infections were confirmed in intranasally with the virus, severe rhinitis and pneumonia
162 patients, and PIVs were found in 8.8% of the patients. In developed over 4 days. Signs of the respiratory tract involve-
the remaining patients, measles virus (21%), influenza A virus ment were still observed 3 weeks postinfection (p.i.), and
(16%), respiratory syncytial virus (7%), influenza B virus (6%), bronchial lymphoid infiltration was still detected 42 days p.i.
enteroviruses (5%), adenoviruses (4%), herpes simplex viruses Viral antigens appeared in the respiratory tract cells during the
(2%), and cytomegalovirus (1%) were found. De Arruda et al. first day after infection, and an increase in the level of
(19) carried out an extensive study among an impoverished virus-specific proteins was seen during the first 4 days; how-
urban population in Brazil; in this study 175 children less than ever, by 7 days p.i. the findings were negative. Infectious virus
5 years of age in 63 families were monitored for 29 months. was recovered from the upper respiratory tract and lungs
Viruses were identified in 35% of the upper respiratory tract during the first 8 days, and virus antibodies appeared9inmonths
serum
samples collected during symptomatic periods, and PIVs were 7 days p.i. In 5 of 12 rats, antibodies were present
found in 16% of the specimens. Other viruses detected were later. Cell-mediated immune responses were also observed 7
rhinoviruses (46%), enteroviruses (16%), adenoviruses (10%), days p.i., and maximal values were measured 2 weeks later; 3noh
herpes simplex viruses (7%), and influenza viruses (6%). response was found 6 months p.i. Interferon appeared by
270 VAINIONPAA AND HYYPIA CLIN. MICROBIOL. REV.

30-

25 PIV 1
20-

15 -

10-

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
20-
ux
c PIV 2
E
._ 15-

0
10
c)
uz

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
40-

Ply 3
30-

10 L-L1 .LL
20-

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

Year
FIG. 3. Occurrence of PIVI, PIV2, and PIV3 infections from 1981 through 1992 as determined by virus antigen detection from nasopharyngeal
specimens at the Department of Virology, University of Turku.

p.i. in lungs and serum, and the highest values were observed neutralizing activity in individual specimens because many
6 h p.i. secretions containing IgA were not neutralizing and not all
Most reports on immunity to PIVs in humans concern local secretions with neutralizing activity contained IgA. It has been
and circulating antibodies; the importance of cell-mediated shown that in Sendai virus infection of mice, local administra-
immunity in clinical PIV infections is largely unknown. These tion of monoclonal IgA or IgG antibodies is effective in
viruses enter the body via the respiratory route, and so the first protecting the airways from viral infection (67). On the other
specific defense mechanism is the barrier of local antibodies. hand, it has been proposed that passively acquired circulating
Smith et al. (94) studied the protective effect of antibodies in maternal antibodies might participate in the pathogenesis of
the serum and in nasal secretions in adult volunteers who were PIV infections. This mechanism was not supported in experi-
challenged intranasally with PIV1. They found that the pres- ments involving PIV3 infection in hamsters, since partial
ence of antibodies in nasal secretions is a better marker of host passive immunization did not enhance the severity of the
resistance to infection than the level of antibodies in serum. disease (40).
Yanagihara and McIntosh (115) have shown that in infants Reinfections with PIVs are common, as reported for infec-
infected with PIV1 or PIV2, a significant increase in the level tions caused by other respiratory viruses. To study the mech-
of virus-specific nasopharyngeal secretory IgA is detected. anism of this phenomenon and the specificity of the immune
However, there was a discordance between the IgA level and response in more detail, van Wyke Coelingh et al. (107)
VOL. 7, 1994 BIOLOGY OF PARAINFLUENZA VIRUSES 271

analyzed the appearance of antibodies against individual anti- challenge in experimental animals (45). Belshe et al. (3)
genic sites of HN and F glycoproteins of PIV3 after primary evaluated passage level 18 of this cold-adapted vaccine candi-
and secondary infection in humans. They found that during date in a double-blind, randomized, placebo-controlled study
primary infection, seronegative infants and children developed in infants and young children. Of the seropositive children,
strong antibody responses with neutralizing activity whereas none in the older age group became infected whereas some of
responses of young infants with maternal antibodies were the younger ones shed the virus. All four seronegative young
weaker. Reinfected children exhibited antibody responses children became infected with the vaccine strain, and two of
higher than those observed during primary infection. In adults, them developed an illness characterized by rhinorrhea and
the antibody levels were lower, suggesting that they decay with wheezing. Furthermore, in one case, the virus spread to a
time. In 90% of the children, the antibody response in primary sibling control (although it did not cause illness), indicating
infection was directed to four of the six antigenic sites studied that this vaccine needs further development before it can be
in the HN glycoprotein, including three of the four neutraliza- used safely.
tion sites. Response against the antigenic sites in the F In Sendai virus infection in mice, mutant viruses deficient in
glycoprotein was weaker than against HN sites and varied proteolytic processing of the F glycoprotein cause restricted
considerably from person to person. Reinfection broadened infection but still are able to induce resistance to challenge by
the site-specific responses, although the response against the F the wild-type virus (103). Therefore one possible approach for
glycoprotein still remained restricted. vaccine development could be selection of corresponding
Rydbeck et al. (87) analyzed the protective effect of mono-
clonal antibodies against the glycoproteins of PIV3 in intra- mutants of the human PIV strains. It is important to keep in
cerebrally infected newborn hamsters. They observed that a mind, however, that for all vaccines containing live viruses,
significant degree of protection was obtained with antibodies attenuation must be stable.
against both HN and F glycoproteins but that none of the It has also become possible to construct recombinant viruses
individual antibodies could completely protect the animals (by using, e.g., vaccinia virus as a vector) which express other
against disease. viral polypeptides on the surface of infected cells. In monkeys
Julkunen et al. (57) studied the nature of the antibody immunized intradermally with a single dose of a vaccinia virus
response in PIVI infection in 10 patients. They observed construct expressing the HN or F glycoprotein of PIV3,
diagnostic increases in the levels of IgGI (nine patients), IgG2 replication of challenge virus was significantly restricted (100,
(one patient), IgG3 (three patients), IgG4 (four patients), 101). Such new vaccines require further careful evaluation, and
IgAl (three patients), and IgM subclass antibodies (three studies with other viruses (e.g., adenovirus) may show that they
patients) between acute- and convalescent-phase sera. The are more optimal vectors in this approach to immunization.
results were very similar to those obtained from patients with Subunit vaccines have also been developed for PIVs by using
influenza A virus infection. different strategies. Ambrose et al. (1) tested a PIV3 subunit
Local interferon production has been observed in 30% of vaccine consisting of detergent-solubilized, affinity-purified
PIV-infected children, whereas the corresponding percentages HN and F glycoproteins in cotton rats. Antibody levels ob-
for respiratory syncytial and influenza virus infections were 4 served after intramuscular immunization were similar to those
and 55%, respectively (43). observed in control animals infected intranasally with PIV3.
After intranasal challenge, virus titers in the immunized ani-
PREVENTION AND TREATMENT mals were significantly reduced and inversely correlated with
antibody levels in serum. In another study, cotton rats were
As outlined above, PIVs are a significant cause of morbidity, immunized with the PIV3 F glycoprotein expressed in insect
especially in infants and young children, and therefore several cells by using a baculovirus vector. Low levels of specific
strategies have been used for the development of effective antibodies were detected, and the animals became moderately
vaccines. These have included live attenuated viruses, nonin- protected against subsequent PIV3 challenge (44). Brideau et
fectious viral protein components, and vaccinia virus vectors al. (7) expressed a chimeric polypeptide containing the extra-
expressing PIV glycoproteins. cellular regions of the F and HN glycoproteins of PIV3 in
Live vaccines have the advantages that they induce more insect cells by using a baculovirus vector. Immunization of
effective local mucosal immunity and that the duration of the cotton rats induced neutralizing antibodies, and the animals
protection is longer when compared with inactivated and became protected against an intranasal challenge with PIV3.
subunit vaccines, although the inactivated and subunit vaccines One risk of using inactivated virus or viral protein subunits as
may have fewer adverse effects. The first inactivated vaccines vaccines is the development of atypical and more severe forms
for PIV1, PIV2, and PIV3 were developed in the late 1960s of disease, as has been observed after immunization with
(29). However, the antibody responses were variable and no formalin-inactivated respiratory syncytial virus preparations
protection against disease was observed (48). (59). This adverse phenomenon can now be evaluated by using
Because bovine PIV3 (bPIV3) is antigenically related to the cotton rat model, thus simplifying the development of safe
human PIV3, attempts have been made to use bPIV3 as a vaccines (80).
vaccine against human disease (the Jennerian approach). It has At present, the treatment of PIV infections is symptomatic.
been shown in a cotton rat model that bPIV3 is able to induce Some candidate antiviral drugs, including ribavirin, interfer-
a high level of resistance to human PIV3 (106). Clinical ons, and protease inhibitors, have been tested, but more
evaluation of this vaccine candidate is under way (13). studies are needed to elucidate their efficacy.
Another approach is to use selected variants of human Rapidly accumulating knowledge of PIV structure, replica-
viruses with reduced virulence. Cold-adapted, temperature- tion, and immunopathogenesis offers a good opportunity to
sensitive mutants of PIV3 have been used for this purpose (74). further develop new and more effective vaccines and antiviral
Cold-adapted mutants were first produced by serial passages drugs to prevent and treat PIV infections. Moreover, improved
under suboptimal conditions in cell cultures. These viruses rapid diagnostic methods for PIVs permit specific diagnosis at
seem to have a stable phenotype, are attenuated when evalu- an early stage of the illness and detection of single cases before
ated in animal models, and induce resistance to wild-type PIV3 the onset of an epidemic. This progress should facilitate
272 VAINIONPAA AND HYYPIA CLIN. MICROBIOL. REV.

development of methods for prevention of respiratory infec- 17. Colman, P.M., P. A. Hoyne, and M. C. Lawrence. 1993. Sequence
tions caused by PIVs. and structure alignment of paramyxovirus hemagglutinin-neur-
aminidase with influenza virus neuraminidase. J. Virol. 67:2972-
2980.
ACKNOWLEDGMENT 18. Cote, M.-J., D. G. Storey, C. Y. Kang, and K. Dimock. 1987.
We thank Matti Waris for providing unpublished PIV epidemiolog- Nucleotide sequence of the coding and flanking regions of the
ical data and for comments during the preparation of the manuscript. human parainfluenza virus type 3 fusion glycoprotein gene. J.
Gen. Virol. 68:1003-1010.
REFERENCES 19. de Arruda, E., F. G. Hayden, J. F. McAuliffe, M. A. de Sota, S. B.
1. Ambrose, M. W., P. R. Wyde, M. Ewasyshyn, A.-M. Bonneau, B. Mota, M. I. McAuliffe, F. C. Geist, E. P. Carvalho, M. C.
Caplan, H. L. Meyer, and M. Klein. 1991. Evaluation of the Fernandes, R. L. Guerrant, and J. M. Gwaltney, Jr. 1991. Acute
immunogenicity and protective efficacy of a candidate parainflu- respiratory viral infections in ambulatory children of urban
enza virus type 3 subunit vaccine in cotton rats. Vaccine 9:505- northeast Brazil. J. Infect. Dis. 164:252-258.
511. 20. Delage, G., P. Brochu, M. Pelletier, G. Jasmin, and N. Lapointe.
2. Arguedas, A., H. R. Stutman, and J. G. Blanding. 1990. Parain- 1979. Giant-cell pneumonia caused by parainfluenza virus. J.
fluenza type 3 meningitis. Report of two cases and review of the Pediatr. 94:426-429.
literature. Clin. Pediatr. 29:175-178. 21. de Silva, L. M., and M. J. Cloonan. 1991. Brief report. Parain-
3. Belshe, R. B., R. A. Karron, F. K. Newman, E. L. Anderson, S. L. fluenza virus type 3 infections: findings in Sydney and some
Nugent, M. Steinhoff, M. L. Clements, M. H. Wilson, S. L. Hall, observations on variations in seasonality world-wide. J. Med.
E. L. Tierney, and B. R. Murphy. 1992. Evaluation of a live Virol. 35:19-21.
attenuated, cold-adapted parainfluenza virus type 3 vaccine in 22. Dhib-Jalbut, S., D. E. McFarlin, and H. F. McFarland. 1989.
children. J. Clin. Microbiol. 30:2064-2070. Measles virus-polypeptide specificity of the cytotoxic T-lympho-
4. Blumberg, B., C. Giorgi, L. Roux, R. Raju, P. Dowling, A. Chollet, cyte response in multiple sclerosis. J. Neuroimmunol. 21:205-
and D. Kolakofsky. 1985. Sequence determination of the Sendai 212.
virus HN gene and its comparison to the influenza virus glyco- 23. Dimock, K., E. W. Rud, and C. Y. Kang. 1986. 3' Terminal
proteins. Cell 41:269-278. sequence of human parainfluenza virus 3 genomic RNA. Nucleic
5. Blumberg, B. M., C. Giorgi, K. Rose, and D. Kolakofsky. 1985. Acids Res. 14:4694.
Sequence determination of the Sendai virus fusion protein gene. 24. Easton, A. J., and R. P. Eglin. 1989. Epidemiology of parainflu-
J. Gen. Virol. 66:317-331. enza virus type 3 in England and Wales over a ten-year period.
6. Bohn, W., G. Rutter, H. Hohenberg, K. Mannweiler, and P. Epidemiol. Infect. 102:531-535.
Nobis. 1986. Involvement of actin filaments in budding of measles 25. Ebata, S. N., M. J. Cote, C. Y. Kang, and K. Dimock. 1991. The
virus: studies of cytoskeleton of infected cells. Virology 149:91- fusion and hemagglutinin-neuraminidase glycoproteins of human
106. parainfluenza virus 3 are both required for fusion. Virology
7. Brideau, R. J., N. L. Oien, D. J. Lehman, F. L. Homa, and M. W. 183:437-441.
Wathen. 1993. Protection of cotton rats against human parain- 26. Elango, N., J. E. Coligan, R. C. Jambou, and S. Venkatesan. 1986.
fluenza virus type 3 by vaccination with a chimeric FHN subunit Human parainfluenza type 3 virus hemagglutinin-neuraminidase
glycoprotein. J. Gen. Virol. 74:471-477. glycoprotein: nucleotide sequence of mRNA and limited amino
8. Castells, E., V. G. George, and J. C. Hierholzer. 1990. NCI-H292 acid sequence of the purified protein. J. Virol. 57:481-489.
as an alternative cell line for the isolation and propagation of the 27. Ferris, J. A. J., W. A. Aherne, W. S. Locke, J. McQuillin, and P. S.
human paramyxoviruses. Arch. Virol. 115:277-288. Gardner. 1973. Sudden and unexpected deaths in infants: histol-
9. Cattaneo, R., K. Kaelin, K. Baczko, and M. A. Billeter. 1989. ogy and virology. Br. Med. J. 2:439-442.
Measles virus editing provides an additional cysteine-rich pro- 28. Frank, A. L., R. B. Couch, C. A. Grilfis, and B. D. Baxter. 1979.
tein. Cell 56:759-794. Comparison of different tissue cultures for isolation and quanti-
10. Chambers, P., N. S. Millar, and P. T. Emmerson. 1986. Nucle- tation of influenza and parainfluenza viruses. J. Clin. Microbiol.
otide sequence of the gene encoding the fusion glycoprotein of 10:32-36.
Newcastle disease virus. J. Gen. Virol. 67:2685-2694. 29. Fulginiti, V. A., J. Amer, J. J. Eller, J. W. Joyner, Jr., and P.
11. Chanock, R. M., K. M. Johnson, M. K. Cook, D. C. Wong, and A. Askin. 1967. Parainfluenza virus immunization. IV. Simultaneous
Vargosko. 1961. The hemadsorption technique, with special immunization with parainfluenza types 1, 2, and 3 aqueous
reference to the problem of naturally occurring simian para- vaccines. Am. J. Dis. Child. 114:26-28.
influenza virus. Am. Rev. Respir. Dis. 83:125-129. 30. Galinski, M. S. 1991. Paramyxoviridae: transcription and repli-
12. Chanock, R. M., and K. McIntosh. 1990. Parainfluenza viruses, p. cation. Adv. Virus Res. 39:129-162.
963-988. In B. N. Fields, D. M. Knipe, R. M. Chanock, J. 31. Galinski, M. S. 1991. Annotated nucleotide and protein se-
Melnick, B. Roizman, and R. Shope (ed.), Virology. Raven Press, quences for selected paramyxoviridae, p. 537-568. In D. W.
New York. Kingsbury (ed.), The paramyxoviruses. Plenum Press, New
13. Clements, M. L., R. B. Belshe, J. King, F. Newman, T. U. York.
Westblom, E. L. Tierney, W. T. London, and B. R. Murphy. 1991. 32. Galinski, M. S., M. A. Mink, D. M. Lambert, S. L. Wechsler, and
Evaluation of bovine, cold-adapted human, and wild-type human M. W. Pons. 1986. Molecular cloning and sequence analysis of
parainfluenza type 3 viruses in adult volunteers and in chimpan- the human parainfluenza 3 virus RNA encoding the nucleocapsid
zees. J. Clin. Microbiol. 29:1175-1182. protein. Virology 149:139-151.
14. Coelingh, K. J., C. C. Winter, and B. R. Murphy. 1985. Antigenic 33. Galinski, M. S., M. A. Mink, D. M. Lambert, S. L. Wechsler, and
variation in the hemagglutinin-neuraminidase protein of human M. W. Pons. 1986. Molecular cloning and sequence analysis of
parainfluenza type 3 virus. Virology 143:569-582. the human parainfluenza 3 virus mRNA encoding the P and C
15. Coelingh, K. J., C. C. Winter, B. R. Murphy, J. M. Rice, P. C. proteins. Virology 155:46-60.
Kimball, R. A. Olmsted, and P. L. Collins. 1986. Conserved 34. Galinski, M. S., M. A. Mink, D. M. Lambert, S. L. Wechsler, and
epitopes on the hemagglutinin-neuraminidase proteins of human M. W. Pons. 1987. Molecular cloning and sequence analysis of
and bovine parainfluenza type 3 viruses: nucleotide sequence the human parainfluenza 3 virus gene encoding the matrix
analysis of variants selected with monoclonal antibodies. J. Virol. protein. Virology 157:24-30.
60:90-96. 35. Galinski, M. S., M. A. Mink, and M. W. Pons. 1988. Molecular
16. Collins, P. L., and G. Mottet. 1991. Homooligomerization of the cloning and sequence analysis of the human parainfluenza 3 virus
hemagglutinin-neuraminidase glycoprotein of human parainflu- gene encoding the L protein. Virology 165:499-510.
enza virus type 3 occurs before the acquisition of correct intramo- 36. Galinski, M. S., R. M. Troy, and A. K. Banerjee. 1992. RNA
lecular disulfide bonds and mature immunoreactivity. J. Virol. editing in the phosphoprotein gene of the human parainfluenza
65:2362-2371. virus type 3. Virology 186:543-550.
VOL. 7, 1994 BIOLOGY OF PARAINFLUENZA VIRUSES 273

37. Garlinghouse, L. E., Jr., G. L. Van Hoosier, Jr., and W. E. tive viruses. J. Gen. Virol. 67:2543-2548.
Giddens, Jr. 1987. Experimental Sendai virus infection in labo- 56. Josephs, S., H. W. Kim, C. D. Brandt, and R. H. Parrott. 1988.
ratory rats. I. Virus replication and immune response. Lab. Anim. Parainfluenza 3 virus and other common respiratory pathogens in
Sci. 37:437-441. children with human immunodeficiency virus infection. Pediatr.
38. Giddens, W. E., Jr., G. L. Van Hoosier, Jr., and L. E. Garling- Infect. Dis. J. 7:207-209.
house. 1987. Experimental Sendai virus infection in laboratory 57. Julkunen, I., T. Hovi, I. Seppala, and 0. Maikela. 1985. Immu-
rats. II. Pathology and immunohistochemistry. Lab. Anim. Sci. noglobulin G subclass antibody responses in influenza A and
37:442-448. parainfluenza type 1 virus infections. Clin. Exp. Immunol. 60:
39. Giuffre, R. M., D. R. Tovell, C. M. Kay, and D. L. J. Tyrrell. 1982. 130-138.
Evidence for an interaction between the membrane protein of a 58. Karron, R. A., K. L. O'Brien, J. L. Froehlich, and V. A. Brown.
paramyxovirus actin. J. Virol. 42:963-968. 1993. Molecular epidemiology of a parainfluenza type 3 virus
40. Glezen, W. P., and G. W. Fernald. 1976. Effect of passive antibody outbreak on a pediatric ward. J. Infect. Dis. 167:1441-1445.
on parainfluenza virus type 3 pneumonia in hamsters. Infect. 59. Kim, H. W., J. G. Canchola, C. D. Brandt, G. Pyles, R. M.
Immun. 14:212-216. Chanock, K. Jensen, and R. H. Parrott. 1969. Respiratory
41. Glezen, W. P., A. L. Frank, L. H. Taber, and J. A. Kasel. 1984. syncytial virus disease in infants despite prior administration of
Parainfluenza virus type 3: seasonality and risk of infection and antigenic inactivated vaccine. Am. J. Epidemiol. 89:422-434.
reinfection in young children. J. Infect. Dis. 150:851-857. 60. Kingsbury, D. W. 1990. Paramyxoviridae and their replication, p.
42. Gupta, K. C., and D. W. Kingsbury. 1984. Complete sequences of 945-962. In B. N. Fields, D. M. Knipe, R. M. Chanock, J.
the intergenic and mRNA start signals in the Sendai virus Melnick, B. Roizman, and R. Shope (ed.), Virology. Raven Press,
genome: homologies with the genome of vesicular stomatitis New York.
virus. Nucleic Acids Res. 12:3829-3841. 61. Kondo, K., H. Bando, M. Tsurudome, M. Kawano, M. Nishio,
43. Hall, C. B., R. G. Douglas, R. L. Simons, and J. M. Geiman. 1978. and Y. Ito. 1990. Sequence analysis of the phosphoprotein (P)
Interferon production in children with respiratory syncytial, genes of human parainfluenza type 4A and 4B viruses and RNA
influenza, and parainfluenza virus infections. J. Pediatr. 93:28-32. editing at transcript of the P genes: the number of G residues
44. Hall, S. L., B. R. Murphy, and K. L. van Wyke Coelingh. 1991. added is imprecise. Virology 178:321-326.
Protection of cotton rats by immunization with the human 62. Korppi, M., M. Leinonen, P. H. Makela, and K. Launiala. 1990.
parainfluenza virus type 3 fusion (F) glycoprotein expressed on Bacterial involvement in parainfluenza virus infection in children.
the surface of insect cells infected with a recombinant baculovi- Scand. J. Infect. Dis. 22:307-312.
rus. Vaccine 9:659-667. 63. Kurilla, M. G., H. 0. Stone, and J. D. Kenne. 1985. RNA
45. Hall, S. L., A. Stokes, E. L. Tierney, W. T. London, R. B. Beishe, sequence and transcriptional properties of the 3' end of the
F. C. Newman, and B. R. Murphy. 1992. Cold-passaged human Newcastle disease virus genome. Virology 145:203-212.
parainfluenza type 3 viruses contain ts and non-ts mutations 64. Luk, D., P. S. Masters, A. Sanchez, and A. Banerjee. 1987.
leading to attenuation in rhesus monkeys. Virus Res. 22:173-184. Complete nucleotide sequence of the matrix protein mRNA and
46. Halonen, P., S. Nikkari, M. Waris, H. Siitari, C. Orwell, J. three intergenic junctions of human parainfluenza virus type 3.
Hierholzer, A. Kendal, and G. Obert. 1989. One-step time- Virology 156:189-192.
resolved fluoroimmunoassays based on monoclonal antibodies in 65. Luk, D., A. Sanchez, and A. K. Banerjee. 1986. Messenger RNA
detection of respiratory viruses, p. 365-370. In A. Balows, R. C. encoding the phosphoprotein (P) gene of human parainfluenza
Tilton, and A. Turano (ed.), Rapid methods and automation in virus 3 is bicistronic. Virology 153:318-325.
microbiology and immunology. Academic Press, Inc., San Diego, 66. Matsuoka, Y., J. Curran, T. Pelet, D. Kolakofsky, R. Ray, and
Calif. R. W. Compans. 1991. The P gene of human parainfluenza virus
47. Hamaguchi, M., T. Yoshida, K. Nishikawa, H. Naruse, and Y. type 1 encodes P and C proteins but not a cysteine-rich V protein.
Nagai. 1983. Transcriptive complex of Newcastle disease virus. I. J. Virol. 65:3406-3410.
Both L and P proteins are required to constitute an active 67. Mazanec, M. B., M. E. Lamm, D. Lyn, A. Portner, and J. G.
complex. Virology 128:105-117. Nedrud. 1992. Comparison of IgA versus IgG monoclonal anti-
48. Heilman, C. A. 1990. Respiratory syncytial and parainfluenza bodies for passive immunization of the murine respiratory tract.
viruses. J. Infect. Dis. 161:402-406. Virus Res. 23:1-12.
49. Henrickson, K. J. 1991. Monoclonal antibodies to human para- 68. McGinnes, L. W., and T. G. Morrison. 1987. The nucleotide
influenza virus type 1 detect major antigenic changes in clinical sequence of the gene encoding the Newcastle disease virus
isolates. J. Infect. Dis. 164:1128-1134. membrane protein and comparisons of membrane protein se-
50. Henrickson, K. J., and L. L. Savatski. 1992. Genetic variation and quences. Virology 156:221-228.
evolution of human parainfluenza virus type 1 hemagglutinin 69. McIntosh, K. 1991. Pathogenesis of severe acute respiratory
neuraminidase: analysis of 12 clinical isolates. J. Infect. Dis. infections in the developing world: respiratory syncytial virus
166:995-1005. and parainfluenza viruses. Rev. Infect. Dis. 13(Suppl. 6):S492-
51. Hiebert, S. W., R. G. Paterson, and R. A. Lamb. 1985. Hemag- SSOO.
glutinin-neuraminidase protein of the paramyxovirus simian virus 70. Millar, N. S., P. Chambers, and P. T. Emmerson. 1986. Nucle-
5: nucleotide sequence of the mRNA predicts an N-terminal otide sequence analysis of the hemagglutinin-neuraminidase
membrane anchor. J. Virol. 54:1-6. gene of Newcastle disease virus. J. Gen. Virol. 67:1917-1927.
52. Iorio, R. M., and M. A. Bratt. 1984. Neutralization of Newcastle 71. Monto, A. S., J. S. Koopman, and E. R. Bryan. 1986. The
disease virus by monoclonal antibodies to the hemagglutinin- Tecumseh study of illness. XIV. Occurrence of respiratory vi-
neuraminidase glycoprotein: requirement for antibodies to four ruses, 1976-1981. Am. J. Epidemiol. 124:359-367.
sites for complete neutralization. J. Virol. 51:445-451. 72. Morgan, E. M., and K. M. Rakestraw. 1986. Sequence of the
53. Iorio, R. M., and M. A. Bratt. 1984. Monoclonal antibodies as Sendai virus L gene: open reading frames upstream of the main
functional probes of the HN glycoprotein of Newcastle disease coding region suggest that the gene may be polycistronic. Virol-
virus: antigenic separation of the hemagglutinating and neur- ogy 154:31-40.
aminidase sites. J. Immunol. 133:2215-2219. 73. Morrison, T., and A. Portner. 1991. Structure, function and
54. Ishida, N., H. Taira, T. Omata, K. Mizumoto, S. Hattori, K. intracellular processing of the glycoprotein of Paramyxoviridae,
Iwasaki, and K. Kawakita. 1986. Sequence of 2617 nucleotides p. 347-382. In D. W. Kingsbury (ed.), The Paramyxoviruses.
from the 3' end of Newcastle disease virus genome RNA and the Plenum Press, New York.
predicted amino acid sequence of viral NP protein. Nucleic Acids 74. Murphy, B. R., G. A. Prince, P. L. Collins, K. van Wyke Coelingh,
Res. 14:6551-6554. R. A. Olmsted, M. K. Spriggs, R. H. Parrott, H.-W. Kim, C. D.
55. Jambou, R. C., N. Elango, S. Venkatesan, and P. L. Collins. 1986. Brandt, and R. M. Chanock. 1988. Current approaches to the
Complete sequence of the major nucleocapsid protein gene of development of vaccines effective against parainfluenza and
human parainfluenza type 3 virus: comparison with other nega- respiratory syncytial viruses. Virus Res. 11:1-15.
274 VAINIONPAA AND HYYPIA CLIN. MICROBIOL. REV.

75. Nishikawa, K., T. Morishima, T. Toyoda, T. Miyadai, T. Yokochi, 1966. Protective effect of antibody to parainfluenza type 1 virus.
T. Yoshida, and Y. Nagai. 1986. Topological and operational N. Engl. J. Med. 275:1145-1152.
delineation of antigenic sites on the HN glycoprotein of New- 95. Southern, J. A., B. Precious, and R. E. Randall. 1990. Two
castle disease virus and their structural requirements. J. Virol. nontemplated nucleotide additions are required to generate the
60:987-993. P mRNA of parainfluenza virus type 2 since the RNA genome
76. Ohigimoto, S., H. Bando, M. Kawano, K. Okamoto, K. Kondo, M. encodes protein V. Virology 177:388-390.
Tsurudome, M. Nishio, and Y. Ito. 1990. Sequence analysis of P 96. Spriggs, M. K., and P. L. Collins. 1986. Sequence analysis of the
gene of human parainfluenza type 2 virus: P and cysteine-rich P and C protein genes of human parainfluenza virus type 3:
proteins are translated by two mRNAs that differ by two non- patterns of amino acid sequence homology among paramyxovirus
templated G residues. Virology 177:116-123. proteins. J. Gen. Virol. 67:2705-2719.
77. Park, K. H., and M. Krystal. 1992. In vivo model for pseudo- 97. Spriggs, M. K., and P. L. Collins. 1986. Human parainfluenza
templated transcription in Sendai virus. J. Virol. 66:7033-7039. virus type 3: messenger RNAs, polypeptide coding assignments,
78. Paterson, R. G., S. M. Thomas, and R. A. Lamb. 1989. Specific intergenic sequences, and genetic map. J. Virol. 59:646-654.
nontemplated nucleotide addition to a simian virus 5 mRNA: 98. Spriggs, M. K., P. L. Collins, E. Tierney, W. T. London, and B. R.
prediction of a common mechanism by which unrecognized Murphy. 1988. Immunization with vaccinia virus recombinants
hybrid P-cysteine-rich proteins are encoded by paramyxovirus 'P' that express the surface glycoproteins of human parainfluenza
genes, p. 232-245. In D. Kolakofsky and B. Mahy (ed.), Genetics virus type 3 (PIV3) protects patas monkeys against PIV3 infec-
and pathogenicity of negative strand viruses. Elsevier Science tion. J. Virol. 62:1293-1296.
Publishing, Inc., New York. 99. Spriggs, M. K., P. R. Johnson, and P. L. Collins. 1987. Sequence
79. Portner, A., R. A. Scroggs, and D. W. Metzger. 1987. Distinct analysis of the matrix protein gene of human parainfluenza virus
functions of antigenic sites of the HN glycoprotein of Sendai type 3: extensive sequence homology among paramyxoviruses. J.
virus. Virology 158:61-68. Gen. Virol. 68:1491-1497.
80. Prince, G. A., A. B. Jenson, V. G. Hemming, B. R. Murphy, E. E. 100. Spriggs, M. K., B. R. Murphy, G. A. Prince, R. A. Olmsted, and
Walsh, R. L. Horswood, and R. M. Chanock. 1986. Enhancement P. L. Collins. 1987. Expression of the F and HN glycoproteins of
of respiratory syncytial virus pulmonary pathology in cotton rats human parainfluenza virus type 3 by recombinant vaccinia vi-
by prior intramuscular inoculation of formalin-inactivated virus. ruses: contributions of the individual proteins to host immunity.
J. Virol. 57:721-728. J. Virol. 61:3416-3423.
81. Prinoski, K., M.-J. Cote, C. Y. Kang, and K. Dimock. 1987. 101. Spriggs, M. K., R. A. Olmsted, S. Venkatesan, J. E. Coligan, and
Nucleotide sequence of the human parainfluenza virus 3 matrix P. L. Collins. 1986. Fusion glycoprotein of human parainfluenza
protein gene. Nucleic Acids Res. 15:3182. virus type 3: nucleotide sequence of the gene, direct identification
82. Prinoski, K., M. J. Cote, C. Y. Kang, and K. Dimock. 1991. of the cleavage-activation site, and comparison with other
Evolution of the fusion protein gene of human parainfluenza paramyxoviruses. Virology 152:241-251.
virus 3. Virus Res. 22:55-69. 102. Stallcup, K. C., C. S. Raine, and B. N. Fields. 1983. Cytochalasin
83. Putto, A., 0. Ruuskanen, and 0. Meurman. 1986. Fever in B inhibits the maturation of measles virus. Virology 124:59-74.
respiratory virus infections. Am. J. Dis. Child. 140:1159-1163. 103. Tashiro, M., and M. Homma. 1985. Protection of mice from
84. Rocchi, G., G. Arangio-Ruiz, V. Giannini, A. M. Jemolo, G. wild-type Sendai virus infection by a trypsin-resistant mutant,
Andreoni, and I. Archetti. 1970. Detection of viraemia in acute TR-2. J. Virol. 53:228-234.
respiratory disease of man. Acta Virol. 14:405-407. 104. van Wyke Coelingh, K., and E. L. Tierney. 1989. Antigenic and
85. Ruuskanen, O., A. Putto, H. Sarkkinen, 0. Meurman, and K. functional organization of human parainfluenza virus type 3
Irjala. 1985. C-reactive protein in respiratory virus infections. J. fusion glycoprotein. J. Virol. 63:375-382.
Pediatr. 107:97-100. 105. van Wyke Coelingh, K., and C. C. Winter. 1990. Naturally
86. Ruutu, P., P. Halonen, 0. Meurman, C. Torres, F. Paladin, K. occurring human parainfluenza type 3 viruses exhibit divergence
Yamaoka, and T. E. Tupasi. 1990. Viral lower respiratory tract in amino acid sequence of their fusion protein neutralization
infections in Filipino children. J. Infect. Dis. 161:175-179. epitopes and cleavage sites. J. Virol. 64:1329-1344.
87. Rydbeck, R., A. Love, and E. Norrby. 1988. Protective effects of 106. van Wyke Coelingh, K. L., C. C. Winter, E. D. Jorgensen, and
monoclonal antibodies against parainfluenza virus type 3-induced B. R. Murphy. 1987. Antigenic and structural properties of the
brain infection in hamsters. J. Gen. Virol. 69:1019-1024. hemagglutinin-neuraminidase glycoprotein of human parainflu-
88. Rydbeck, R., A. Love, C. Orvell, and E. Norrby. 1987. Antigenic enza virus type 3: sequence analysis of variants selected with
analysis of human and bovine parainfluenza virus type 3 strains monoclonal antibodies which inhibit infectivity, hemagglutina-
with monoclonal antibodies. J. Gen. Virol. 68:2153-2160. tion, and neuraminidase activities. J. Virol. 61:1473-1477.
89. Sakaguchi, T., S. Takao, K. Kiyotani, Y. Fujii, T. Nakayama, and 107. van Wyke Coelingh, K. L., C. C. Winter, E. L. Tierney, S. L. Hall,
T. Yoshida. 1993. Expression of the HN, F, NP and M proteins of W. T. London, H.-W. Kim, R. M. Chanock, and B. R. Murphy.
Sendai virus by recombinant vaccinia viruses and their contribu- 1990. Antibody responses of humans and nonhuman primates to
tion to protective immunity against Sendai virus infections in individual antigenic sites of the hemagglutinin-neuraminidase
mice. J. Gen. Virol. 74:479-484. and fusion glycoproteins after primary infection or reinfection
90. Sanchez, A., A. K. Banerjee, Y. Furuichi, and M. A. Richardson. with parainfluenza type 3 virus. J. Virol. 64:3833-3843.
1986. Conserved structures among the nucleocapsid proteins of 108. Vidal, S., J. Curran, and D. Kolakofsky. 1990. Editing of the
the Paramyxoviridae: complete nucleotide sequence of human Sendai virus P/C mRNA by G insertion occurs during mRNA
parainfluenza virus type 3 NP mRNA. Virology 152:171-180. synthesis via a virus-encoded activity. J. Virol. 64:239-246.
91. Sarkkinen, H. K., P. E. Halonen, and A. A. Salmi. 1981. Type- 109. Vidal, S., and D. Kolakofsky. 1989. Modified model for the switch
specific detection of parainfluenza viruses by enzyme-immunoas- from Sendai virus transcription to replication. J. Virol. 63:1951-
say and radioimmunoassay in nasopharyngeal specimens of pa- 1958.
tients with acute respiratory disease. J. Gen. Virol. 56:49-57. 110. Vuorinen, T., and 0. Meurman. 1989. Enzyme immunoassays for
92. Shioda, T., Y. Hidaka, T. Kanda, H. Shibuta, A. Nomoto, and K. detection of IgG and IgM antibodies to parainfluenza types 1, 2
Iwasaki. 1983. Sequence of 3687 nucleotides from the 3' end of and 3. J. Virol. Methods 23:63-70.
Sendai virus genome RNA and the predicted amino acid se- 111. Wang, E., B. A. Wolf, R. A. Lamb, P. W. Choppin, and R. A.
quences of viral NP, P, and C proteins. Nucleic Acids Res. Goldberg. 1976. The presence of actin in enveloped viruses, p.
11:7317-7330. 589-599. In R. Goldman, T. Pollard, and J. Rosenbaum (ed.) Cell
93. Shioda, T., K. Iwasaki, and H. Shibuta. 1986. Determination of motility. Cold Spring Harbor Laboratory, Cold Spring Harbor,
the complete nucleotide sequence of the Sendai virus genome N.Y.
RNA and the predicted amino acid sequences of the F, HN and 112. Waris, M., M. Kivivirta, M. Rautiainen, and T. Ziegler. Unpub-
L protein. Nucleic Acids Res. 14:1545-1563. lished data.
94. Smith, C. B., R. H. Purcell, J. A. Bellanti, and R. M. Chanock. 113. Waris, M., T. Ziegler, M. Kivivirta, and 0. Ruuskanen. 1990.
VOL. 7, 1994 BIOLOGY OF PARAINFLUENZA VIRUSES 275

Rapid detection of respiratory syncytial virus and influenza A and 2. Infect. Immun. 30:23-28.
virus in cell cultures by immunoperoxidase staining with mono- 116. Yusoff, K., N. S. Millar, P. Chambers, and P. T. Emmerson. 1987.
clonal antibodies. J. Clin. Microbiol. 28:1159-1162. Nucleotide sequence analysis of the L gene of Newcastle disease
114. Wendt, C. H., D. J. Weisdorf, M. C. Jordan, H. H. Balfour, Jr., virus: homologies with Sendai and vesicular stomatitis viruses.
and M. I. Hertz. 1992. Parainfluenza virus respiratory infection Nucleic Acids Res. 15:3961-3976.
after bone marrow transplantation. N. Engl. J. Med. 326:921-926. 117. Zinserling, A. 1972. Peculiarities of lesions in viral and myco-
115. Yanagihara, R., and K. McIntosh. 1980. Secretory immunological plasma infections of the respiratory tract. Virchows Arch. Abt. A
response in infants and children to parainfluenza virus types 1 Pathol. Anat. 356:259-273.