- Email: [email protected]
Identification of amino acid positions associated with neuraminidase activity of the hemagglutinin-neuraminidase glycoprotein of sendai virus
VIROLOGY
180, 803-806
(199 1)
Identification of Amino Acid Positions of the Hemagglutinin-Neuraminidase WENDY
L.
GORMAN,’
Department
TATSUFUMI
Associated with Neuraminidase Activity Glycoprotein of Sendai Virus
TAKAHASHI,
RUTH
ANN
SCROGGS,
of Virology and Molecular Biology. St. Jude Children’s 332 N. Lauderdale, P.O. Box 3 18, Memphis, Tennessee Received
September
5, 1990; accepted
October
26,
AND Research 38101
ALLEN
PORTNER’
Hospital,
1990
Identification of amino acid positions associated with neuraminidase activity on the hemagglutinin-neuraminidase (HN) glycoprotein of paramyxoviruses has been difficult because neuraminidase-inhibiting antibodies are not neutralizing and thus, escape mutants have not been isolated. Instead, many investigators have correlated an altered neuraminidase (NA) activity of natural virus variants, such as plaque-size variants, with sequence changes in the HN protein. To identify regions on the HN glycoprotein of Sendai virus (SV) that are associated with NA activity, we investigated NA activity of three plaque-size variants which potentially differed from the standard SV (SV/std). NA activity was measured by the ability of virus to elute from chicken etythrocytes as a result of cleaving sialic acid receptors, and by the ability of virus to cleave sialic acid from the small trisaccharide neuraminlactose and the larger substrate fetuin in an in vitro assay. Virions purified from each of the isolated plaques had a HN content and hemagglutinating activity similar to that of SV/std, yet each variant eluted much more rapidly from chicken erythrocytes than SV/std. ln vitro NA activity of the plaque-size variants was 1.6 to 3.8 times greater than that of SV/std, providing supporting evidence for the elution data. Although all plaque-size variants showed elevated NA activity, there was no correlation of activity with plaque size. Sequence analysis showed that one of the variants had an amino acid change from glutamic acid to valine at position 165 and from lysine to glutamic acid at position 461, while a second variant had only the change at position 461. A third variant had a nearby change at position 468, from threonine to lysine. Taken together, these data support the conclusion that the amino acid residues at positions 461-468 and 165 are involved in neuraminidase activity of SV. 0 1991 Academic Press, Inc.
the structure-function relationship and would be useful in predicting which functions might be affected by specific changes in the molecule. Escape mutants of SV selected with anti-HN neutralizing monoclonal antibodies have permitted identification of amino acid positions associated with HA activity alone or in combination with NA activity, and have provided valuable structural information about the HN protein (6). However, antibodies which inhibit only NA activity are not neutralizing, and thus positions associated exclusively with neuraminidase have generally not been obtained with escape mutants. Plaque-size variants of Newcastle disease virus (NDV) with altered NA activity but normal HA activity have been identified, and positions of amino acid changes could therefore be associated with the neuraminidase function (8). Here we report amino acid positions associated with neuraminidase activity of the HN of three SV plaquesize variants. Virus from plaques of different sizes than our standard Enders strain of SV (SV/std) were isolated because of their potential for altered biological activity. Virus from plaque isolates was amplified in 1O-day-old embtyonated hens’ eggs and replaqued to confirm
The hemagglutinin-neuraminidase (HN) surface glycoprotein of Sendai virus (SV) is a multifunctional molecule with two opposing functions: the binding of the virus to sialic acid-containing receptors on the host cell and the enzymatic cleavage of those receptors by the neuraminidase. Cell binding is necessary for initiation of infection, bringing the virion in close proximity to the host cell so that the fusion (F) protein can act on the host membrane and allow viral entry (7). The role of neuraminidase in the infection process is less clear. It does not appear to influence cell binding and viral entry, but may involve release of the progeny virus from sialic acid-containing moities of host cells (2). Data suggest that the sites for the receptor binding and neuraminidase functions. are independent but proximal on the HN molecule because some monoclonal antibodies inhibit only hemagglutinin (HA) or neuraminidase (NA) activities, while others inhibit both (3-7). The precise location of these functional sites on the linear sequence would further our understanding of
’ Present address: Veterans Administration Medical Jefferson Ave, Memphis, TN 38104. ’ To whom correspondence should be addressed.
Center,
1030
803
0042.6822/91
$3.00
Copynght Q 1001 by Acadcmlc Press, Inc. All rights of reproduction I” any form resewed.
804
SHORT
Std
SPI
LPl
COMMUNICATIONS
LP2
HN FO NP h
FOG. 1. Proteins of purified virus (10 rg viral protein per lane) were electrophoresed on a 9010 polyacrylamide gel under denaturing conditions, and stained with Coomassie blue. Std is the Sendai virus standard. SPl , LPl , and LP2 are the small (SP) and large (LP) plaquesize variants of Sendai virus.
phenotype (6). Plaque sizes of three variants were significantly different from that of SV/std (P < O.OOl), with average diameters of 1.61 -+ 0.46 mm for a large plaque variant LPI, 1.73 t 0.40 mm for LP2, and 0.92 t 0.19 mm for a small-plaque variant SPl , compared with 1 .16 -t 0.66 mm for SV/std. To directly compare cell binding or NA activity of these plaque variants, we first quantified the amount of HN protein of each variant relative to total viral protein. Proteins of purified virus were separated by SDSPAGE and stained with Coomassie blue. All variants appeared from the protein gel to have amounts of HN per microgram of viral protein equivalent to that of SV/ std (Fig. 1). This finding was confirmed by a densitometry scan of the protein gel, in which the amount of HN of each variant was quantified relative to the total protein for that variant. Compared with SV/std, all variants had a similar HN content (Table 1) and differed from SV/std by no more than 10%. In addition, all plaque variants had an HA activity per microgram of total protein similar to that of SV/std (Table 1). In contrast, the plaque-size variants showed significant differences in their NA activity, eluting much more rapidly from cRBC than SV/std when agglutinated chicken erythrocytes (cRBC) were warmed from 25 to 37” (Table 1). Because elution from cRBC is a measure of neuramini-
dase activity, our results indicated that the variants had significantly increased NA activity despite similar HN content. To corroborate the elution data, NA activity was determined by an in vitro assay (9) using both the small trisaccharide Iv-acetylneuraminlactose and the larger fetuin as substrates. Table 1 shows that the NA activity of the plaque variants was 1.6 to 3.8 times greater than that of SV/std. The results of these neuraminidase assays indicated that the variants were therefore valuable candidates for determining amino acid changes associated with NA activity. To identify amino changes in the primary sequence responsible for the altered NA activity, the HN gene of each variant was sequenced by the dideoxy chain termination method (10). The amino acid substitutions found in each are summarized in Table 1. Two changes at positions 165 and 461 were found in one of the large plaque (LP) variants, while the other LP variant had only the 461 change. The small plaque (SP) variant had a single change nearby at position 468. Amino acid substitutions at all three positions involved a change in charge or hydrophobicity, which could have an impact directly on the active site or indirectly on molecular conformation. In this regard, the mobility of the HN protein was slightly increased in the SP variant and slightly decreased in the LP variants (Fig. 1). From the sequencing results we can conclude that the 461-468 region and the 165 position are involved in the neuraminidase function of SV. Precisely how the 461-468 changes enhance NA activity is not known, but enhancement may be due to a conformational change which makes the enzymatic active site more accessible to the substrate. NA activity appears easily affected by amino acid substitutions, as exemplified by NDV, in which mutants can have enhanced or diminished NA activity relative to the wild type depending on the amino acid substitution at a single position (8). Indeed, the finding that three plaque variants from our SV/std stock differed from SV/std in NA activity but not HA activity suggests a tolerance in this virus for varied NA activity. The greatest enhancement in in vitro NA activity occurred in the variant with the single 461 change, LP2 (Table 1). Although the change at position 468 was only 7 amino acids from 461, the neuraminidase of SPl was clearly not as effective in cleaving sialic acid, either in the elution assay or the in vitro assay (Table 1). The 165 change combined with the 461 substitution (LPI) diminished the enhancement of NA activity of 461 alone by approximately 25% (Table l), suggesting an interaction of these positions near the neuraminidase site. Amino acids at positions 165,461, and 468 are not conserved among six paramyxoviruses (4, 13), although they may
SHORT
COMMUNICATIONS TABLE
HN CONTENT,
805
1
HEMAGGLUTINATION AND NEUROAMINIDASE ACTIVITIES, AND AMINO ACID CHANGES IN THE HN OF STANDARD SENDAI VIRUS AND PLAQUE-SIZE VARIANTS Amino
Virus std f LPI LP2 SPl
HNitotal”
HAb
23.6 21.2 21.0 24.4
240 240 240 240
% HA remaining” (20 min at 37”) 75 0 0 45
NAd NAL
(% of std) Fetuin
100 288 381 198
100 228 316 163
acid change”
165
461
468
E V
K E E
T
K
a Proteins of purified virus (10 pg total protean) from infected LLC-MK, cells (grown In MEM media without trypsin) were electrophoresed on a 9% polyacrylamide gel under denaturing conditions and stained with Coomassie blue (shown in Fig. 1). The protein content in the gel was determined by whole band analysis with a Biolmage Visage 110 densitometer (Millipore Co.). Integrated intensity values of each variant are expressed as the percentage HN of total virus protein. * Hemagglutinating activity of purified virus (1 pg total protein) from Infected LLC-MK, cells. Virus was titrated in 96.well microtiter plates with 0.5% chicken erythrocytes (cRBC) at 25” for 25 min. c Once the full hemagglutinating pattern had formed, the microtiter plates were incubated at 37” for 20 min. A decrease in HA titer with time was due to the elution of virus from the cRBC by neuraminidase activity. d In vitro neuraminidase activity. Replicate samples of 0.4 pg (total protein) of purified virus were incubated at 37” for 30 min with 100 fig of Iv-acetylneuraminlactose or 625 pg of fetuin, following the procedure of Aymard-Henry (9). e Determined by dideoxy sequencing (10) using genomic RNA. Virions from allanroic fluid were concentrated by differential centrifugation and further purified by sedimentation through sucrose gradients (3). Genomic RNA was then extracted by proteinase K-SDS digestion followed by phenol-chloroform extraction and ethanol precipitation (15). Synthetic primers used for sequencing were complementary to the SV HN sequence of Blumberg (16) and were provided by the Molecular Resource Center, St. Jude Children’s Research Hospital. Nucleotide changes corresponding to amino acid positions 165, 461, and 468 are GAA to GTA, AAA to GAA, and ACG to ACA, respectively. f Std virus consisted of a mixture of genotypes. Of 15 randomly selected plaques, 73% were of the std genotype as determined by HN sequencing. The remaining 27% consisted of genotypic variants, including the plaque-size variants.
be near conserved regions. Therefore, amino acids at these positions may be allowed to vary without adversely affecting gross conformation. What is evident from our study is that NA activity does not correlate with plaque-size phenotype of SV because both large and small plaque-size variants had enhanced NA activity. In support of this, lorio (8) found a large-plaque variant of NDV had diminished NA relative to the wild type. Thus, plaque size variation in SV must be due to factors other than, or in addition to, NA activity, such as viral replication efficiency or fusion activities. Amino acid changes associated with NA in othervariants of Sendai virus as well as other paramyxoviruses have been located in one or the other of the general regions around 165 or 461-468, but not in both. Thompson and Portner (6) described an antibody escape mutant (selected with an HA monoclonal antibody) with an amino acid substitution at position 184, and which had diminished NA activity but average HA activity. Similarly, neuraminidase-associated mutations in the regions of 175 to 20 1 have been found in NDV (8), mumps virus (I I), and bovine parainfluenza 3 (12), leading Morrison and Portner (13) to propose that the conserved region 163-210 forms part of the neuraminidase site for SV. Finally, amino acid substitutions
in the region 454-460 have been associated with NA activity in NDV mutants selected with monoclonal antibodies which inhibit both HA and NA activities (14). The amino terminal region of 165 and the carboxy region around positions 461-468 appear to be involved in the neuraminidase function of the HN of SV. Our study is the first to associate both regions with a change in NA activity. Therefore, we propose that in the three-dimensional folding of the HN molecule, the 165 position may be in close proximity to the 461-468 region.
ACKNOWLEDGMENTS We thank Dr. Carroll Pridgen for laboratory assistance. This research was funded by Public Health Service Research Grant All 1949 from the National Institute of Allergy and Infectious Diseases, by Cancer Center Support Grant CA 21765 from the National Cancer Institute, and by American Lebanese Syrian Associated Charities of St. Jude Children’s Research Hospital.
REFERENCES 1. SCHEID, A., and CHOPPIN, 2. SCHEID, A., and CHOPPIN,
P. W., virology 57, 475-490 P. W., J. Viral. 11, 263-270
(1974). (1974).
806
SHORT
COMMUNICATIONS
3. IORIO. R. M., and BRAN, M. A., J. lmmunol. 133, 221552219 (1984). 4. GORMAN, W. L., GILL, D. S., SCROGGS, R. A., and PORTNER, A., vko/ogy 175, 211-221 (1990). 5. PORTNER, A., viiology 115, 375-384 (198 1). 6. PORTNER, A., SCROGGS, R. A., and METZGER, D. W., Virology 158, 61-68 (1987). 7. THOMPSON, S. D., and PORTNER, A., Vifoiogy 160, l-8 (1987). 8. IORIO. R. M., SYDDALL, R. J., GLICKMAN, R. L., RIEL, A. M., SHEEHAN, 1. P., and BRAT, M. A., Virology 173, 196-201 (1989). 9. AYMARD-HENRY, M., COLEMAN, P. T., DOWDLE, W. R., LAVER, W. G., SCHILD, G. C., and WEBSTER, R. G., Buli. WHO 48, 199202 (1973).
70. SANGER, F., NICKLEN, S., and COULSON, A. R., Proc. Nat/. Aced. Sci. USA 84, 5463-5467 (1977). 11. WAXHAM, M. N., and ARONOWSKI, I., Virology 167, 226-232 (1988). 12. SHIODA, T., WAKAO, S., Suzu. S., and SHIBUTA, H., &o/ogy 162, 388-396 (1988). 13. MORRISON, T., and PORTNER, A., In “The Paramyxoviruses” (D. W. Kingsbury, Ed.), Academic Press, in press. 74. YUSOFF, K., NESBIT, M., MCCARTNEY, H., EMMERSON, P. T., and SAMSON, A. C. R., Virus Res. 11, 319-333 (1988). 15. BEAN, W. J., SRIRAM, G., and WEBSTER, R. G.,Ana/. Biochem. 102, 228-232 (1980). 16. BLUMBERG, B., GIORGI, C., Roux, L., RPJU, R., DOWLING, P.. CHOLLET, A., and KOLAKOFSKY, D., Cell 41, 269-278 (1985).
180, 803-806
(199 1)
Identification of Amino Acid Positions of the Hemagglutinin-Neuraminidase WENDY
L.
GORMAN,’
Department
TATSUFUMI
Associated with Neuraminidase Activity Glycoprotein of Sendai Virus
TAKAHASHI,
RUTH
ANN
SCROGGS,
of Virology and Molecular Biology. St. Jude Children’s 332 N. Lauderdale, P.O. Box 3 18, Memphis, Tennessee Received
September
5, 1990; accepted
October
26,
AND Research 38101
ALLEN
PORTNER’
Hospital,
1990
Identification of amino acid positions associated with neuraminidase activity on the hemagglutinin-neuraminidase (HN) glycoprotein of paramyxoviruses has been difficult because neuraminidase-inhibiting antibodies are not neutralizing and thus, escape mutants have not been isolated. Instead, many investigators have correlated an altered neuraminidase (NA) activity of natural virus variants, such as plaque-size variants, with sequence changes in the HN protein. To identify regions on the HN glycoprotein of Sendai virus (SV) that are associated with NA activity, we investigated NA activity of three plaque-size variants which potentially differed from the standard SV (SV/std). NA activity was measured by the ability of virus to elute from chicken etythrocytes as a result of cleaving sialic acid receptors, and by the ability of virus to cleave sialic acid from the small trisaccharide neuraminlactose and the larger substrate fetuin in an in vitro assay. Virions purified from each of the isolated plaques had a HN content and hemagglutinating activity similar to that of SV/std, yet each variant eluted much more rapidly from chicken erythrocytes than SV/std. ln vitro NA activity of the plaque-size variants was 1.6 to 3.8 times greater than that of SV/std, providing supporting evidence for the elution data. Although all plaque-size variants showed elevated NA activity, there was no correlation of activity with plaque size. Sequence analysis showed that one of the variants had an amino acid change from glutamic acid to valine at position 165 and from lysine to glutamic acid at position 461, while a second variant had only the change at position 461. A third variant had a nearby change at position 468, from threonine to lysine. Taken together, these data support the conclusion that the amino acid residues at positions 461-468 and 165 are involved in neuraminidase activity of SV. 0 1991 Academic Press, Inc.
the structure-function relationship and would be useful in predicting which functions might be affected by specific changes in the molecule. Escape mutants of SV selected with anti-HN neutralizing monoclonal antibodies have permitted identification of amino acid positions associated with HA activity alone or in combination with NA activity, and have provided valuable structural information about the HN protein (6). However, antibodies which inhibit only NA activity are not neutralizing, and thus positions associated exclusively with neuraminidase have generally not been obtained with escape mutants. Plaque-size variants of Newcastle disease virus (NDV) with altered NA activity but normal HA activity have been identified, and positions of amino acid changes could therefore be associated with the neuraminidase function (8). Here we report amino acid positions associated with neuraminidase activity of the HN of three SV plaquesize variants. Virus from plaques of different sizes than our standard Enders strain of SV (SV/std) were isolated because of their potential for altered biological activity. Virus from plaque isolates was amplified in 1O-day-old embtyonated hens’ eggs and replaqued to confirm
The hemagglutinin-neuraminidase (HN) surface glycoprotein of Sendai virus (SV) is a multifunctional molecule with two opposing functions: the binding of the virus to sialic acid-containing receptors on the host cell and the enzymatic cleavage of those receptors by the neuraminidase. Cell binding is necessary for initiation of infection, bringing the virion in close proximity to the host cell so that the fusion (F) protein can act on the host membrane and allow viral entry (7). The role of neuraminidase in the infection process is less clear. It does not appear to influence cell binding and viral entry, but may involve release of the progeny virus from sialic acid-containing moities of host cells (2). Data suggest that the sites for the receptor binding and neuraminidase functions. are independent but proximal on the HN molecule because some monoclonal antibodies inhibit only hemagglutinin (HA) or neuraminidase (NA) activities, while others inhibit both (3-7). The precise location of these functional sites on the linear sequence would further our understanding of
’ Present address: Veterans Administration Medical Jefferson Ave, Memphis, TN 38104. ’ To whom correspondence should be addressed.
Center,
1030
803
0042.6822/91
$3.00
Copynght Q 1001 by Acadcmlc Press, Inc. All rights of reproduction I” any form resewed.
804
SHORT
Std
SPI
LPl
COMMUNICATIONS
LP2
HN FO NP h
FOG. 1. Proteins of purified virus (10 rg viral protein per lane) were electrophoresed on a 9010 polyacrylamide gel under denaturing conditions, and stained with Coomassie blue. Std is the Sendai virus standard. SPl , LPl , and LP2 are the small (SP) and large (LP) plaquesize variants of Sendai virus.
phenotype (6). Plaque sizes of three variants were significantly different from that of SV/std (P < O.OOl), with average diameters of 1.61 -+ 0.46 mm for a large plaque variant LPI, 1.73 t 0.40 mm for LP2, and 0.92 t 0.19 mm for a small-plaque variant SPl , compared with 1 .16 -t 0.66 mm for SV/std. To directly compare cell binding or NA activity of these plaque variants, we first quantified the amount of HN protein of each variant relative to total viral protein. Proteins of purified virus were separated by SDSPAGE and stained with Coomassie blue. All variants appeared from the protein gel to have amounts of HN per microgram of viral protein equivalent to that of SV/ std (Fig. 1). This finding was confirmed by a densitometry scan of the protein gel, in which the amount of HN of each variant was quantified relative to the total protein for that variant. Compared with SV/std, all variants had a similar HN content (Table 1) and differed from SV/std by no more than 10%. In addition, all plaque variants had an HA activity per microgram of total protein similar to that of SV/std (Table 1). In contrast, the plaque-size variants showed significant differences in their NA activity, eluting much more rapidly from cRBC than SV/std when agglutinated chicken erythrocytes (cRBC) were warmed from 25 to 37” (Table 1). Because elution from cRBC is a measure of neuramini-
dase activity, our results indicated that the variants had significantly increased NA activity despite similar HN content. To corroborate the elution data, NA activity was determined by an in vitro assay (9) using both the small trisaccharide Iv-acetylneuraminlactose and the larger fetuin as substrates. Table 1 shows that the NA activity of the plaque variants was 1.6 to 3.8 times greater than that of SV/std. The results of these neuraminidase assays indicated that the variants were therefore valuable candidates for determining amino acid changes associated with NA activity. To identify amino changes in the primary sequence responsible for the altered NA activity, the HN gene of each variant was sequenced by the dideoxy chain termination method (10). The amino acid substitutions found in each are summarized in Table 1. Two changes at positions 165 and 461 were found in one of the large plaque (LP) variants, while the other LP variant had only the 461 change. The small plaque (SP) variant had a single change nearby at position 468. Amino acid substitutions at all three positions involved a change in charge or hydrophobicity, which could have an impact directly on the active site or indirectly on molecular conformation. In this regard, the mobility of the HN protein was slightly increased in the SP variant and slightly decreased in the LP variants (Fig. 1). From the sequencing results we can conclude that the 461-468 region and the 165 position are involved in the neuraminidase function of SV. Precisely how the 461-468 changes enhance NA activity is not known, but enhancement may be due to a conformational change which makes the enzymatic active site more accessible to the substrate. NA activity appears easily affected by amino acid substitutions, as exemplified by NDV, in which mutants can have enhanced or diminished NA activity relative to the wild type depending on the amino acid substitution at a single position (8). Indeed, the finding that three plaque variants from our SV/std stock differed from SV/std in NA activity but not HA activity suggests a tolerance in this virus for varied NA activity. The greatest enhancement in in vitro NA activity occurred in the variant with the single 461 change, LP2 (Table 1). Although the change at position 468 was only 7 amino acids from 461, the neuraminidase of SPl was clearly not as effective in cleaving sialic acid, either in the elution assay or the in vitro assay (Table 1). The 165 change combined with the 461 substitution (LPI) diminished the enhancement of NA activity of 461 alone by approximately 25% (Table l), suggesting an interaction of these positions near the neuraminidase site. Amino acids at positions 165,461, and 468 are not conserved among six paramyxoviruses (4, 13), although they may
SHORT
COMMUNICATIONS TABLE
HN CONTENT,
805
1
HEMAGGLUTINATION AND NEUROAMINIDASE ACTIVITIES, AND AMINO ACID CHANGES IN THE HN OF STANDARD SENDAI VIRUS AND PLAQUE-SIZE VARIANTS Amino
Virus std f LPI LP2 SPl
HNitotal”
HAb
23.6 21.2 21.0 24.4
240 240 240 240
% HA remaining” (20 min at 37”) 75 0 0 45
NAd NAL
(% of std) Fetuin
100 288 381 198
100 228 316 163
acid change”
165
461
468
E V
K E E
T
K
a Proteins of purified virus (10 pg total protean) from infected LLC-MK, cells (grown In MEM media without trypsin) were electrophoresed on a 9% polyacrylamide gel under denaturing conditions and stained with Coomassie blue (shown in Fig. 1). The protein content in the gel was determined by whole band analysis with a Biolmage Visage 110 densitometer (Millipore Co.). Integrated intensity values of each variant are expressed as the percentage HN of total virus protein. * Hemagglutinating activity of purified virus (1 pg total protein) from Infected LLC-MK, cells. Virus was titrated in 96.well microtiter plates with 0.5% chicken erythrocytes (cRBC) at 25” for 25 min. c Once the full hemagglutinating pattern had formed, the microtiter plates were incubated at 37” for 20 min. A decrease in HA titer with time was due to the elution of virus from the cRBC by neuraminidase activity. d In vitro neuraminidase activity. Replicate samples of 0.4 pg (total protein) of purified virus were incubated at 37” for 30 min with 100 fig of Iv-acetylneuraminlactose or 625 pg of fetuin, following the procedure of Aymard-Henry (9). e Determined by dideoxy sequencing (10) using genomic RNA. Virions from allanroic fluid were concentrated by differential centrifugation and further purified by sedimentation through sucrose gradients (3). Genomic RNA was then extracted by proteinase K-SDS digestion followed by phenol-chloroform extraction and ethanol precipitation (15). Synthetic primers used for sequencing were complementary to the SV HN sequence of Blumberg (16) and were provided by the Molecular Resource Center, St. Jude Children’s Research Hospital. Nucleotide changes corresponding to amino acid positions 165, 461, and 468 are GAA to GTA, AAA to GAA, and ACG to ACA, respectively. f Std virus consisted of a mixture of genotypes. Of 15 randomly selected plaques, 73% were of the std genotype as determined by HN sequencing. The remaining 27% consisted of genotypic variants, including the plaque-size variants.
be near conserved regions. Therefore, amino acids at these positions may be allowed to vary without adversely affecting gross conformation. What is evident from our study is that NA activity does not correlate with plaque-size phenotype of SV because both large and small plaque-size variants had enhanced NA activity. In support of this, lorio (8) found a large-plaque variant of NDV had diminished NA relative to the wild type. Thus, plaque size variation in SV must be due to factors other than, or in addition to, NA activity, such as viral replication efficiency or fusion activities. Amino acid changes associated with NA in othervariants of Sendai virus as well as other paramyxoviruses have been located in one or the other of the general regions around 165 or 461-468, but not in both. Thompson and Portner (6) described an antibody escape mutant (selected with an HA monoclonal antibody) with an amino acid substitution at position 184, and which had diminished NA activity but average HA activity. Similarly, neuraminidase-associated mutations in the regions of 175 to 20 1 have been found in NDV (8), mumps virus (I I), and bovine parainfluenza 3 (12), leading Morrison and Portner (13) to propose that the conserved region 163-210 forms part of the neuraminidase site for SV. Finally, amino acid substitutions
in the region 454-460 have been associated with NA activity in NDV mutants selected with monoclonal antibodies which inhibit both HA and NA activities (14). The amino terminal region of 165 and the carboxy region around positions 461-468 appear to be involved in the neuraminidase function of the HN of SV. Our study is the first to associate both regions with a change in NA activity. Therefore, we propose that in the three-dimensional folding of the HN molecule, the 165 position may be in close proximity to the 461-468 region.
ACKNOWLEDGMENTS We thank Dr. Carroll Pridgen for laboratory assistance. This research was funded by Public Health Service Research Grant All 1949 from the National Institute of Allergy and Infectious Diseases, by Cancer Center Support Grant CA 21765 from the National Cancer Institute, and by American Lebanese Syrian Associated Charities of St. Jude Children’s Research Hospital.
REFERENCES 1. SCHEID, A., and CHOPPIN, 2. SCHEID, A., and CHOPPIN,
P. W., virology 57, 475-490 P. W., J. Viral. 11, 263-270
(1974). (1974).
806
SHORT
COMMUNICATIONS
3. IORIO. R. M., and BRAN, M. A., J. lmmunol. 133, 221552219 (1984). 4. GORMAN, W. L., GILL, D. S., SCROGGS, R. A., and PORTNER, A., vko/ogy 175, 211-221 (1990). 5. PORTNER, A., viiology 115, 375-384 (198 1). 6. PORTNER, A., SCROGGS, R. A., and METZGER, D. W., Virology 158, 61-68 (1987). 7. THOMPSON, S. D., and PORTNER, A., Vifoiogy 160, l-8 (1987). 8. IORIO. R. M., SYDDALL, R. J., GLICKMAN, R. L., RIEL, A. M., SHEEHAN, 1. P., and BRAT, M. A., Virology 173, 196-201 (1989). 9. AYMARD-HENRY, M., COLEMAN, P. T., DOWDLE, W. R., LAVER, W. G., SCHILD, G. C., and WEBSTER, R. G., Buli. WHO 48, 199202 (1973).
70. SANGER, F., NICKLEN, S., and COULSON, A. R., Proc. Nat/. Aced. Sci. USA 84, 5463-5467 (1977). 11. WAXHAM, M. N., and ARONOWSKI, I., Virology 167, 226-232 (1988). 12. SHIODA, T., WAKAO, S., Suzu. S., and SHIBUTA, H., &o/ogy 162, 388-396 (1988). 13. MORRISON, T., and PORTNER, A., In “The Paramyxoviruses” (D. W. Kingsbury, Ed.), Academic Press, in press. 74. YUSOFF, K., NESBIT, M., MCCARTNEY, H., EMMERSON, P. T., and SAMSON, A. C. R., Virus Res. 11, 319-333 (1988). 15. BEAN, W. J., SRIRAM, G., and WEBSTER, R. G.,Ana/. Biochem. 102, 228-232 (1980). 16. BLUMBERG, B., GIORGI, C., Roux, L., RPJU, R., DOWLING, P.. CHOLLET, A., and KOLAKOFSKY, D., Cell 41, 269-278 (1985).