 Proc. Natl. Acad. Sci. USA
Vol. 87, pp. 4154-4158, June 1990
Biochemistry
Sequence of an influenza virus hemagglutinin determined directly
from a clinical sample
(Hi subtype/virus from nasopharynx/polymerase chain reaction/comparison with egg-grown viruses)
AUGUSTINE RAJAKUMAR*, ELLA M. SWIERKOSZt, AND IRENE T. SCHULZE*1
*Department of Microbiology, Saint Louis University School of Medicine, 1402 South Grand Boulevard, Saint Louis, MO 63104; and tDiagnostic Virology
Laboratory, Cardinal Glennon Children's Hospital, and Department of Pediatrics/Adolescent Medicine and Pathology, Saint Louis University Medical
Center, 1465 South Grand Boulevard, Saint Louis, MO 63104
Communicated by Edwin D. Kilbourne, March 21, 1990 (receivedfor review December 8, 1989)
ABSTRACT The sequence of the HAl region of the he-
magglutinin gene of an influenza virus has been determined
without growing the virus in eggs or in cultured cells. The virus
used was an HI strain of influenza A from a clinical specimen
taken from a patient in 1987. RNA was extracted directly from
virus that had been sedimented out ofthe transport medium in
which the sample had been stored. DNA copies of the hemag-
glutinin gene, obtained by reverse transcription, were then
amplified by thepolymerasechain reaction andweresequenced
by the dideoxy termination method. The deduced amino acid
sequence is highly similar to that of other Hi viruses that had
been isolated at about the same time and cultured for a limited
number ofpassages in eggs. Furthermore, the HAl sequence of
progeny virus from this isolate obtained after one passage in
chicken embryos is identical to that of the virus obtained
directly from the nasopharynx. The results suggest that HI
isolates that have been grown for a limited number ofpassages
in embryonated eggs have HAl subunits that falthfuily repre-
sent the virus population in the clinical samples from which
they were derived.
The hemagglutinin (HA) on the surface ofthe influenza virion
is a glycoprotein that has three biological activities that play
major roles in the infectious process. The HA attaches the
virus to host cells, it is the antigenic target against which
neutralizing antibodies are made, and it fuses the viral
envelope with the membranes ofendocytotic vesicles so that
the transcription complex of the virus can enter the cyto-
plasm of the cell. These three activities are associated with
specific regions of the HA (for reviews see refs. 1-3), and
amino acid substitutions in these regions can change the
ability of the virus to grow in cells of different species.
Substitutions on the rim of the receptor binding pocket can
change the ability of H3 viruses to bind to specific sialic
acid-galactose linkages on host cells (4-7). In addition, the
presence of oligosaccharides at specific sites on the HA can
determine how well the virus binds to certain cells (8, 9) and
how susceptible the HA is to the proteolytic cleavage that is
required for the expression of its fusion activity (10-12).
Thus, changes in the amino acid sequence ofthe HA can alter
virus host range at the cellular level, and it has been clearly
shown that minority forms within a virus population will
rapidly predominate if they have a growth advantage during
cultivation in the laboratory (2, 8, 13-17).
These observations have led to the concern that minor
heterogeneity within clinical samples or mutations that occur
during isolation and growth could lead to laboratory-grown
influenza virus populations that are significantly different
from those in the original clinical samples. Were such selec-
tion to occur during growth of the virus in embryonated
chicken eggs or in certain cultured cells, virus from those
sources would be poorly suited for the study ofHA function
and for the preparation of vaccines.
The studies presented here were undertaken to determine
the sequence ofthe HA gene directly from the virus within a
clinical sample so that nucleotide changes that might be
associated with growth in laboratory hosts could be avoided.
We have used the polymerase chain reaction (PCR; ref. 18)
to amplify DNA made from virus recovered from a patient
with influenza and have determined the nucleotide sequence
ofthe HA gene from its 3' end into the HA2 region. We have
used the same procedure to determine the sequence of this
region of the HA gene after growth ofthis isolate in chicken
embryos. The deduced amino acid sequences are reported
here§ and are compared to those of viruses of the same
subtype isolated since 1977 by growth in embryonated eggs.
MATERIALS AND METHODS
Virus Isolation and Growth. Nasopharyngeal swabs taken
forthe diagnosis offebrile respiratory illness were used in this
study. Samples were taken in January of 1987 at Cardinal
Glennon Children's Hospital (Saint Louis, MO). Virus iso-
lation and identification were carried out using tube cultures
of primary rhesus monkey kidney cells obtained from Wh-
itaker M.A. Bioproducts. Swabs that contained influenza
virus were stored at -700C in -2 ml of transport medium
(veal infusion broth containing 0.5% gelatin, penicillin, gen-
tamicin, and amphotericin B).
To obtain egg-grown virus for this study, virus from one of
the clinical samples was diluted 1:9 in sterile phosphate-
buffered saline and injected into the allantoic cavity of two
10-day-old chicken embryos (0.2 ml per embryo). Allantoic
fluid containing 21280 hemagglutinating units per ml was
harvested after 72 hr ofincubation at 34WC and was stored at
-700C.
Plaque-purified influenza A (HlNi), strain WSN, was also
used in these studies. This virus has been grown in the
laboratory for many generations and for these studies was
obtained from Madin-Darby bovine kidney cells.
RNA Extraction. Samples oftransport medium or allantoic
fluid containing virus were thawed rapidly and diluted with
NTE buffer (10 mM Tris HCl, pH 7.5/100 mM NaCl/1 mM
EDTA). Virus from transport medium or allantoic fluid was
concentrated by centrifugation at 105,000 x g for 2 hr,
resuspended in 300 pl of NTE, and quantitated by hemag-
glutination as described (19). Based on the titer of the
Abbreviations: HA, hemagglutinin; PCR, polymerase chain reac-
tion.
tTo whom reprint requests should be addressed.
§The sequence reported in this paper has been deposited in the
GenBank data base (accession no. M33748).
4154
The publication costs ofthis article were defrayed in part by page charge
payment. This article must therefore be hereby marked "advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Proc. Natl. Acad. Sci. USA 87 (1990) 4155
resuspended virus, the clinical samples used in these studies
each contained :1200 hemagglutinating units of virus.
Twenty microliters of virus suspension containing 50-100
hemagglutinating units was mixed with an equal volume of
buffer containing 200 mM Tris HCl (pH 7.4), 300 mM NaCl,
25 mM EDTA, and 2% SDS and incubated with 100 ,g of
proteinase K (Boehringer Mannheim) at 37°C for 1 hr. After
addition of 10 ,g of tRNA, the RNA was extracted with
phenol/chloroform and precipitated with ethanol. The RNA
was then incubated at 37°C for 30 min with 40 units of
RNase-free DNase (Pharmacia), extracted with phenol/
chloroform, reprecipitated, and suspended in 10 ,u of TE
buffer (10 mM Tris HCl, pH 8.1/1 mM EDTA).
cDNA Synthesis. Reverse transcription of the virion RNA
was carried out as described (20) using avian myeloblastosis
virus reverse transcriptase obtained from Life Sciences and
a primer consisting of a 12-nucleotide sequence that is
complementary to the 3' ends of all of the virion segments.
This primer and all others used in this study, unless otherwise
noted, were prepared by Paul Wollenzein (Department of
Biochemistry, Saint Louis University) using an Applied
Biosystems DNA synthesizer and were purified by protocols
described by Applied Biosystems. The cDNA RNA product
obtained by reverse transcription was extracted once with
phenol/chloroform, precipitated with ethanol, and dissolved
in 10 ,ul of TE buffer.
HA Gene Amplification by the PCR. The entire HA1 region
ofthe HA gene and part ofthe HA2 region were amplified by
using Gene-Amp kits obtained from Cetus/Perkin-Elmer.
The reaction mixture (100 ,ul) contained the cDNA'RNA
complex described above, 10 mM Tris HCI (pH 8.3), 50 mM
KCl, 1.5 mM MgCI2, 0.001% gelatin, 2.5 units of Thermus
aquaticus (Taq) DNA polymerase, 200 uM dNTP, and 20
pmol of each of two primers as indicated in Results. The
reaction mixture was overlaid with 100 ,ul oflight mineral oil
(Sigma) and amplification was carried out for 25 cycles each
consisting of 1 min ofdenaturation at 94°C, 2 min at 35°C for
reannealing, and 2 min at 72°C for extension. Following
extraction with chloroform to remove the mineral oil, an
aliquot of the amplified DNA was analyzed by electropho-
resis in 1% agarose gel with a Tris/borate/EDTA buffer
system (21). To ensure that there was a sufficient amount of
DNA for multiple sequencing reactions, the DNA from the
first amplification was purified from low-melting agarose gels
and was reamplified as described above.
Nucleotide Sequence Analysis. Double-stranded DNA cop-
ies of the HA gene were sequenced directly from the ampli-
HA 1
fled pool by the dideoxy termination method (22). Primers,
end-labeled with [y-32P]ATP (ICN Radiochemicals) by using
T4 polynucleotide kinase (Pharmacia), were annealed to
300-400 ng of amplified DNA by boiling for 10 min and
allowing the reaction mixture to come to room temperature.
The DNA-primer complex was ethanol-precipitated and was
sequenced using Sequenase kits (United States Biochemi-
cal).
The sequence of both DNA strands was determined from
the virus taken directly from the nasopharynx. The six
primers used for sequencing the coding strand were based on
the published sequences of the WSN and PR8 strains of
influenza A (HlNl) (23, 24). They consisted of bases 11-35,
219-233, 384-404, 549-568, 796-813, and 918-932 of plus-
strand RNA. The five primers used to sequence the noncod-
ing strand were complementary to nucleotides 1005-986,
692-669, 496-478, 238-219, and 116-99 of the A/SL/2/87
plus strand. The same primers were used to determine the
HA sequence of egg-grown virus.
RESULTS
These studies were undertaken to determine the sequence of
the HA gene without there having been opportunities for
selection during growth ofthe virus in a laboratory host. The
HA1 region and the cleavage region that links HA1 to HA2
have been investigated. The part of the HA gene that was
amplified extended from its 3' end to nucleotide 1095, which
is located 31 bases into the HA2 region of this genome
segment.
The primers used for preparation ofthe amplified DNA and
their relationship to the HA gene are shown in Fig. 1. Primers
I and IV (kindly provided by Andrew Caton, Wistar Institute,
Philadelphia) were the first to be used in this study, because
they were expected to produce DNA copies that would
contain most of the HA1 region of the HA gene (see Fig. 1).
The amplified DNA obtained from two clinical samples by
using primers I and IV is shown in Fig. 2A. DNA that
appeared to be heterogeneous in size but was about 1000 base
pairs long was obtained from each. A/SL/2/87 was chosen
for use in all subsequent studies.
As shown in Fig. 2B, the amplified DNA obtained from this
virus could be resolved into three bands by longer periods of
electrophoresis. When the individual bands were eluted from
a gel like that shown in lane 1 and reapplied to a second gel,
each retained its original mobility (lanes 2-4), indicating that
they represent discrete populations ofmolecules. The largest
HA2
CLEAVAGE SITE
VIRION 3,
RNA
1045 bp
5'
/-{- $1 IV
-1773 bp
II W 1095 bp
-u II
~*+Z~1 V
FIG. 1. Primers used in amplification and the DNA products obtained. The primers and their relationship to the HA gene are shown. Primer
I, d(AGCAAAAGCAGG), is complementary to the first 12 nucleotides at the 3' end of all of the virion RNAs. It consists of plus-strand DNA
(see text). Primer II, d(AGTAGAAACAAGG), consists of the first 13 nucleotides at the 5' end of all of the virion RNAs and its minus-strand
DNA. Primer III, d(AGCAAAAGCAGGGGAAAATAAAAACAACCAAAATG), is plus-strand DNA and is complementary to the first 35 bases
at the 3' end of A/SL/2/87 HA RNA. Primer IV, d(ATGTTCCTTAGTCCTGTAACCAT), is complementary to nucleotides 1045-1023 of the
plus-strand HA DNA. Primer V, d(CAATGAAACCGGCAATGGCTCC), is complementary to nucleotides 1095-1074 of the plus-strand of
USSR HA DNA. The regions of the virion RNA represented in amplified DNA synthesized from primers I and IV, I and II, and III and V are
also shown. The DNA used for sequencing directly from the nasopharynx was obtained by using primers I and IV in both the first and second
amplifications or primers I and II followed by primers III and V (see text). DNA was synthesized from egg-grown A/SL/2/87 by using primers
III and V in both amplifications. bp, Base pairs.
Biochemistry: Rajakumar et al.
4156 Biochemistry: Rajakumar et al.
A B
3 6
Kb
2 .0 -
0 5-
IB1i
B2-
A{
FIG. 2. Agarose gel electrophoresis of the amplified DNA ob-
tained by using primers I and IV. (A) Lane 1, HindIII-digested A
DNA (size markers); lanes 2 and 3, amplified DNA copies of HA
RNA from two clinical samples, A/SL/1/87 and A/SL/2/87, re-
spectively. (B) Lane 1, amplified A/SL/2/87 HA DNA separated
into three bands by electrophoresis for 8 hr at 100 V; lanes 2-4,
purified bands B1, B2, and B3, respectively; lane 5, amplified WSN
HA DNA; lane 6, marker DNA. kb, Kilobase(s).
of these DNAs, designated B1, had the same mobility as the
DNA obtained from the WSN strain of influenza A (HlNl)
(lane 5). Amplified DNA obtained from this virus by the same
procedure as that used with the clinical samples consistently
gave only one band, which was of the size expected from
primers I and IV (1045 base pairs). B1 DNA was therefore
purified by gel electrophoresis and used for sequence anal-
ysis.
To obtain information about the sequence at the cleavage
site between HA1 and HA2, we needed DNA copies that
extended into the HA2 region. We therefore amplified DNA
by using primers I and II (see Fig. 1). Since these primers are
complementary to the conserved regions at the two ends of
all genome segments (25-27), they were expected to produce
complete copies of all eight virion genes, albeit in small
amounts. To obtain a sufficient amount of the HA DNA for
sequencing, we used these complete DNA copies as tem-
plates for reampliflication using two new HA-specific prim-
ers, III and V. Since we had already determined the sequence
ofmost ofthe plus-strand HA DNA, one ofthese new primers
(III) was designed to be complementary to the first 35
nucleotides at the 3' end of the A/SL/2/87 HA gene. This
primer is identical to primer I for the first 12 nucleotides but
contains 23 additional bases that are specific for the sequence
found in A/SL/2/87. Primer III, along with primer V, which
is complementary to nucleotides 1095-1074 in the HA2
region of the USSR plus strand (see Fig. 1), produced DNA
ofone size class. It migrated slightly slowerthan B1 DNA and
had the same Ava I and Ava II cleavage sites as those in B1
DNA (results not shown). By sequencing both strands ofthis
DNA, we have determined the sequence of the first 31 bases
of the HA2 region and have confirmed the sequence that we
had previously obtained from B1 DNA. The use of this
second method ofobtaining amplified DNA has increased the
likelihood that our sequence represents that of the majority
of the viruses in the clinical sample rather than the sequence
ofa minority population that was selectively amplified by the
primers used.
The nucleic acid sequence of A/SL/2/87 after a single
passage in eggs was also determined from DNA obtained by
PCR using primers III and V. This DNA, like that obtained
directly from the clinical sample when these two primers
were used, was homogeneous in size and was slightly larger
than B1 DNA. The nucleotide sequence of this DNA was
identical to that obtained from virus taken directly from the
throat.
The nucleotide sequence of A/SL/2/87 and that of em-
bryonated egg-grown A/TW/1/86 (kindly provided by
Nancy Cox, Centers for Disease Control, Atlanta) are highly
similar. Differences were observed at seven positions (cy-
tosine in A/TW/1/86 versus uracil in A/SL/2/87 at nucle-
otides 251, 371, 848, and 1001; guanine versus adenine at
nucleotides 748 and 793; and adenine versus guanine at
nucleotide 1021). Four of these differences (at residues 251,
371, 848, and 1001) are in the third nucleotide of the codon
and do not change the amino acid specified by the sequence.
The remaining three (at nucleotides 748, 793, and 1021)
change the amino acids at positions 225, 240, and 315 as
indicated in Fig. 3. However, two independent reports ofthe
amino acid sequence ofA/TW/1/86 have been published (30,
31); they differ at amino acids 138, 240, and 315. As indicated
in Fig. 3, the sequence ofA/SL/2/87 differs from both ofthe
published sequences ofA/TW/1/86 by only one amino acid;
aspartic acid (D) is found at position 225 in both nasopha-
ryngeal and egg-grown A/SL/2/87, whereas glycine (G) is
found in egg-grown A/TW/1/86. Both of these recent H1
isolates differ significantly from A/USSR/77.
DISCUSSION
Although interest in influenza virus epidemiology and genet-
ics has led to numerous reports concerning the amino acid
sequence ofthe HA gene, all previously reported sequences
1 10 20
A/USSR/77 MKAKLLVLLCALSATDADTICI
yHC A
ISTYA
A/TW/1/86 ------- FT-------------
AISLI2/87 -----------FT------- -
--- __
30 40 50 *
DTVDTVLEK NTTHSVNLLEDSHN6KLCRLKGIAP
60 70 * 80 90
L(LGKCN IAGWI LGNPECESLFSKKSWSYIAETPNS
IN - S -
* 100 110 120 ***
E NGTCYP GY FADYEELR EQLS S VS S F ER FE I FPKER
-_-__ S
_-
--
-_
_ _-------
S
130 * 140 150 160
SWPK~jNVT7RGVTASCSHKGKSSFYRNLLWLTE 6
- - -SI --K -- - -S/A- -- -- -- -- -- -- - - - - ---
170 180 190
YPNL
FiKSYVNNKEKEVLVLWGVHHPSNIEDOKTIYR
__
___i------------G--------RA--H
200 210 220 230
KENAYVSVVSSNYNRRFTPEIAERPKVRGUAGRINY
T- ---------H ---------K- -- -- --E-----
T------H------K---D-E---
240 250 260 *
YWTLLEPGDTIIFEANGNLIAPUHAFALNRGFGSGI
-GI--6E--------Y----S----
----E--------Y----S----
270 280 290 300
IT S MD ECDTKCaTP(GAIr7ZILP F(NIHP VT IG
___ __ ---- - A -________--- - - -- -- V- -- -- -
-_
-___- -- -- -A ----------
____ -I- -- -- V- -- ---
V
310 320 330
ECPKYVRSTKLRMVTGLRNIPSIQSRGLFGAIAGFI
-----------K/R--------------
-----------R-------_________________
FIG. 3. Amino acid sequence of the HA1 subunit and the HA1/
HA2 cleavage region of nasopharyngeal and egg-grown A/SL/2/87
and oftwo egg-grown H1 strains, A/USSR/77 and A/TW/1/86. The
amino acids are numbered to correspond to those in the H3 HA as
previously described (28). Amino acids present in the H1 subtype but
not in H3 are marked by stars. The glycosylation sites are indicated
by boxes. Dashes indicate no difference in sequence from that found
in A/USSR/77 (29). For A/TW/1/86, two amino acids are given at
three positions. The amino acid above the bar is that found by Cox
etal. (30) and the one below the bar is that found by Robertson (31).
Although both preparations of A/TW/86 that were sequenced con-
tained glycine at residue 225, aspartic acid has been found at that
position in numerous egg-grown isolates (see text). Amino acids
330-339 of A/TW/1/86 have not been reported. Arrowhead indi-
cates the proteolytic cleavage site between HA1 and HA2. A single
sequence is shown for A/SL/2/87, since nasopharyngeal and egg-
grown virus gave identical sequences.
Proc. Natl. Acad. Sci. USA 87 (1990)
Proc. Natl. Acad. Sci. USA 87 (1990) 4157
have been derived from virus strains that had been grown
either in eggs or in cultured cells. Our approach using the
PCR to amplify copies of the HA gene has enabled us to
determine the sequence of this viral gene starting with the
limited amount of virus that can be recovered from a na-
sopharyngeal swab taken during the acute stage ofillness. In
so doing we have been able to determine the sequence ofthis
viral surface glycoprotein without there having been oppor-
tunities for selection ofvariant forms during virus cultivation
in eggs or in cell culture.
We have found that the HA gene sequence encompassing
the HA1 region and the HA1/HA2 cleavage region ofa 1987
isolate is identical when obtained directly from virus in the
nasopharynx and from virus grown in chicken embryos.
Furthermore, this sequence is strikingly similar to those of
other H1 viruses isolated in 1986. In fact, the only difference
between A/SL/2/87 and A/TW/1/86 is the presence of
aspartic acid instead ofglycine at amino acid 225. By analogy
to the H3 HA (32), this amino acid is located on the rim ofthe
receptor binding pocket and might be expected to influence
the specificity or the affinity ofthe HA for receptors on host
cells from different sources. It has, in fact, been proposed
that an aspartic acid to glycine substitution occurs during
adaptation ofH1 viruses to growthin eggs (33). This is not the
case with the isolate described here, since aspartic acid
remains at position 225 when A/SL/2/87 is grown for one
passage in eggs. In addition, of some 19 H1 viruses isolated
since 1977 and sequenced after a limited number ofpassages
in eggs, 8 have aspartic acid, 8 have glycine, and 3 have
asparagine at position 225 (30, 31, 33, 34), indicating that all
three of these amino acids are compatible with growth ofH1
strains in eggs. Thus, the sequence data presented here from
DNA synthesized by PCR are consistent with those obtained
previously from egg-grown H1 isolates. The accumulated
evidence indicates that embryonated eggs do not provide a
strong selective pressure for an aspartic acid to glycine
substitution at amino acid 225. Whether this change will
occur during long-term passage of A/SL/2/87 in eggs or in
cultured mammalian cells and whether other amino acid
substitutions will accompany it can now be determined.
As indicated in Fig. 3, glutamine has been found at position
226 in A/SL/2/87 and in the two reference strains. This
amino acid is also on the rim of the receptor binding pocket
and has been implicated in determining the receptor speci-
ficity of H3 strains (4). With H3 strains, glutamine at amino
acid 226 has been found in virus obtained from avian sources
whereas leucine has been found at that position in virus of
human and equine origin (35). This is not the case with H1
strains; glutamine has been found at position 226 in all human
isolates grown in eggs or cultured cells and, as shown here,
in virus obtained directly from the human nasopharynx.
The glycosylation sites on the H1 viruses isolated since
1977 have been highly conserved. As shown in Fig. 3,
A/USSR/77 has eight sites whereas A/TW/86 has nine.
Except for the additional glycosylation sites at amino acid 63
of the HA1 subunit and the shift of the site from 131 to 129
in the 1986 isolates (30, 31), the H1 strains isolated since 1977
have identical glycosylation sites. Since these same glyco-
sylation sites are found in virus taken directly from the
nasopharynx, neither addition nor deletion of glycosylation
sites can be attributed to growth of these viruses in the
laboratory. It remains to be determined, however, whether
all ofthese sites will be maintained through extensive passage
of these viruses in eggs or in cell culture. It is interesting in
this regard that the strains of influenza virus that have been
grown in the laboratory for many years have functional HA1
subunits with as few as four glycosylation sites, only two of
which (those at amino acid 20/21 and 271) are conserved in
all strains. In the case of the WSN strain, deletion of a
glycosylation site from the tip of the HA actually enhances
growth ofthe virus in some mammalian cells (8, 9), and with
an influenza B strain, virus isolated in eggs lacks a glycosyl-
ation site that is present on virus isolated from the same
source in Madin-Darby canine kidney cells (16). Since de-
letion ofglycosylation sites is clearly tolerated during growth
of these viruses in cell cultures, their conservation among
new isolates suggests that the oligosaccharides at these sites
are critical to the survival of these strains in nature, rather
than to the receptor-binding or fusion activity of the HA.
These considerations strengthen the previously made pro-
posal that the oligosaccharides may mask antigenic sites (36)
that would otherwise contribute to immune surveillance and
reduce virus transmission within the human population.
It is important to point out that the information presented
here should not be taken to indicate that the virus population
within the nasopharynx is homogeneous with respect to its
HA sequence. Minor heterogeneity in the HA genes would
not be detected by the techniques used here. Since we have
sequenced the HA gene directly from amplified DNA pools
obtained by two different procedures, we have reduced the
chances of seeing random substitutions that could be intro-
duced into the sequence by the Taq polymerase. However,
this approach also reduces the possibility of detecting minor
components within a heterogeneous population. Thus, the
deduced amino acid sequence we have obtained is that ofthe
majority component within the clinical samples, and other
approaches are needed to determine whether these popula-
tions are heterogeneous in HA sequence.
Since only low amounts ofvirus are needed for sequencing
by the technique used here, various questions relating to the
virus population within a clinical sample can now be inves-
tigated. For example, will virus from other individuals in-
fected within the same epidemic show exactly the same
sequence as that reported here, or will the major component
found within a clinical sample depend in part on the immune
state of the patient? Will there be any changes in HA
sequence during the progression of the disease? Will pro-
longed passage in different laboratory hosts produce different
populations from this isolate and, ifso, is this due to mutation
and selection during cultivation in the laboratory or to
selection of different minor components within the clinical
sample? With respect to these two alternatives, previous
work indicates that both processes occur. On the one hand,
mutants with single base substitutions in the HA gene have
beenfound to arise in homogeneous populations derived from
single plaques and to replace the original virus strain when
the mutant has a selective advantage over the parent in the
cell culture used for virus propagation (8, 9). On the other,
virus populations obtained directly from patients with influ-
enza have been resolved into subpopulations that differ from
one another by as little as a single amino acid (17, 33, 35-37).
The information presented here indicates that growth of
new isolates for a limited number ofpassages in embryonated
eggs can be expected to provide H1 stocks with HAl se-
quences that faithfully represent those found on virions
within the nasopharynx. Whether this is the case with other
influenza A subtypes and influenza B viruses can now be
determined using the approach employed here. This ap-
proach should also be useful in vaccine development and
evaluation; it should help to ensure that the virus strains that
are incorporated into influenza vaccines have HAs that
mirror those found in the human population.
Lastly, the incomplete DNAs in the amplified pools ob-
tained from the clinical samples need to be investigated. As
indicated above, these smaller DNAs have been obtained
from both ofthe clinical isolates that we have examined. With
A/SL/2/87 they were seen with virus obtained directly from
the nasopharynx and from chicken embryo-grown virus when
primers I and IV were used during amplification (data not
shown), but not when primers III and V were used. They
Biochemistry: Rajakumar et al.
4158 Biochemistry: Rajakumar et al.
were not observed when plaque-isolated virus of the WSN
strain was amplified using primers I and IV. These DNAs
may be generated during amplification as a consequence of
some as yet unidentified variation or imperfection in the
PCR. However, the conditions under which they are found
are consistent with their being authentic DNA copies of
defective HA genome segments.
We wish to thank Drs. H. Peter Zassenhaus, Shahika Aytay, and
Michelle Inkster for advice and support throughout this work, Dr.
Andrew Caton for providing us with primers and advice, Terrie
Konsky for skilled secretarial assistance, and Drs. J. Skehel and R.
Daniels for review of the manuscript. This work was supported in
part by Grants AI10097 and AI23520 from the National Institutes of
Health.
1. Webster, R. G., Laver, W. G., Air, G. M., & Schild, G. C.
(1982) Nature (London) 296, 115-121.
2. Schulze, I. T. (1987) Adv. Cell Culture 5, 59-96.
3. Wiley, D. C. & Skehel, J. J. (1987) Annu. Rev. Biochem. 56,
365-394.
4. Rogers, G. N., Paulson, J. C., Daniels, R. S., Skehel, J. J.,
Wilson, I. A. & Wiley, D. C. (1983) Nature (London) 304,
76-78.
5. Daniels, R. S., Jeffries, S., Yates, P., Schild, G. C., Rogers,
G. N., Paulson, J. C., Wharton, S. A., Douglas, A. R., Skehel,
J. J. & Wiley, D. C. (1987) EMBO J. 6, 1459-1465.
6. Weis, W., Brown, J. H., Cusack, S., Paulson, J. C., Skehel,
J. J. & Wiley, D. C. (1988) Nature (London) 333, 426-431.
7. Suzuki, Y., Kato, H., Naeve, C. W. & Webster, R. G. (1989)
J. Virol. 63, 4298-4302.
8. Crecelius, D. M., Deom, C. M. & Schulze, I. T. (1984) Virol-
ogy 139, 164-177.
9. Deom, C. M., Caton, A. J. & Schulze, I. T. (1986) Proc. Natl.
Acad. Sci. USA 83, 3771-3775.
10. Bosch, F. X., Garten, W., Klenk, H.-D. & Rott, R. (1981)
Virology 113, 725-735.
11. Rott, R., Orlich, M., Klenk, H.-D., Wang, M. L., Skehel, J. J.
& Wiley, D. C. (1984) EMBO J. 3, 3329-3332.
12. Kawaoka, Y. & Webster, R. G. (1989) J. Virol. 64, 3296-3300.
13. Kilbourne, E. D. (1978) Proc. Natl. Acad. Sci. USA 75, 6258-
6262.
14. Both, G. W., Cheng, H. S. & Kilbourne, E. D. (1983) Proc.
Natl. Acad. Sci. USA 80, 6996-7000.
15. Schild, G. C., Oxford, J. S., DeJong, J. C. & Webster, R. G.
(1983) Nature (London) 303, 706-709.
16. Robertson, J. S., Naeve, C. W., Webster, R. G., Bootman,
J. S., Neuman, R. & Schild, G. C. (1985) Virology 143, 166-
174.
17. Katz, J. M., Naeve, C. W. & Webster, R. G. (1987) Virology
156, 386-395.
18. Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi,
R., Horn, G. T., Mullis, K. B. & Erlich, H. A. (1988) Science
239, 487-491.
19. Noronha-Blob, L. & Schulze, I. T. (1976) Virology 69, 314-
322.
20. Winter, G., Fields, S. & Gait, M. J. (1981) Nucleic Acids Res.
9, 237-245.
21. Maniatis, T., Fritsch, E. & Sambrook, J. (1982) Molecular
Cloning:A Laboratory Manual(Cold Spring Harbor Lab., Cold
Spring Harbor, NY), p. 191.
22. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl.
Acad. Sci. USA 74, 5463-5467.
23. Hiti, A. L., Davis, A. R. & Nayak, D. P. (1981) Virology 111,
113-124.
24. Caton, A., Brownlee, G. G., Yewdell, J. W. & Gerhard, W.
(1982) Cell 31, 417-427.
25. Skehel, J. J. & Hay, A. J. (1978) Nucleic Acids Res. 5, 1207-
1219.
26. Robertson, J. S. (1979) Nucleic Acids Res. 6, 3745-3757.
27. Desselberger, V., Racaniello, V. R., Zazra, J. J. & Palese, P.
(1980) Gene 8, 315-328.
28. Winter, G., Fields, S. & Brownlee, G. G. (1981) Nature (Lon-
don) 292, 72-75.
29. Concannon, P., Cummings, I. W. & Salser, W. A. (1984) J.
Virol. 49, 276-278.
30. Cox, N. J., Black, R. A. & Kendal, A. P. (1989) J. Gen. Virol.
70, 299-313.
31. Robertson, J. S. (1987) J. Gen. Virol. 68, 1205-1208.
32. Wilson, I. A., Skehel, J. J. & Wiley, D. C. (1981) Nature
(London) 289, 366-373.
33. Robertson, J. S., Bootman, J. S., Newman, R., Oxford, J. S.,
Daniels, R. S., Webster, R. G. & Schild, G. C. (1987) Virology
160, 31-37.
34. Raymond, F. L., Caton, A. J., Cox, N. J., Kendal, A. P. &
Brownlee, G. G. (1986) Virology 148, 275-287.
35. Rogers, G. N. & Paulson, J. C. (1983) Virology 127, 361-373.
36. Skehel, J. J., Stevens, D. J., Daniels, R. S., Douglas, A. R.,
Knossow, M., Wilson, I. A. & Wiley, D. C. (1984) Proc. Natl.
Acad. Sci. USA 81, 1779-1783.
37. Katz, J. M. & Webster, R. G. (1988) Virology 165, 446-456.
Proc. Natl. Acad. Sci. USA 87 (1990)
