Swine flu
What is it about influenza that makes this disease continue to plague humanity? Each year, a very large vaccination campaign is waged on a very significant portion of the population to protect us from the new influenza. How is it that influenza mutates so quickly that we need a new vaccine each year? And why do we continue to live under the specter of the 1918 strain, which was particularly lethal?
Read the attached article regarding the development of swine flu vaccines. How does the situation appear? Are we protected? Relative to standard influenza, how big of a concern is swine flu?
APA format, in-text citation, reference include, 1 page ( 2 paragraphs that cover all the main point).
The Evolving Threat of Influenza Viruses of Animal Origin
and the Challenges in Developing Appropriate Diagnostics
Polly W.Y. Mak,1 Shanthi Jayawardena,1 and Leo L.M. Poon1*
BACKGROUND: An H1N1 subtype of swine origin caused
the first influenza pandemic in this century. This pan-
demic strain was a reassortant of avian, swine, and hu-
man influenza viruses. Many diagnostic laboratories
were overwhelmed by the testing demands related to
this pandemic. Nevertheless, there remains the threat
of other animal influenza viruses, such as highly patho-
genic H5N1. As a part of pandemic preparedness, it is
essential to identify the diagnostic challenges that will
accompany the next pandemic.
CONTENT: We discuss the natural reservoir of influenza
viruses and the possible role of livestock in the emer-
gence of pandemic strains. The current commonly
used molecular tests for influenza diagnosis or surveil-
lance are also briefly reviewed. Some of these ap-
proaches are also used to detect animal viruses. Unfor-
tunately, owing to a lack of systematic surveillance of
animal influenza viruses, established tests may not be
able to detect pandemic strains that have yet to emerge
from the animal reservoir. Thus, multiple strategies
need to be developed for better identification of influ-
enza viruses. In addition, molecular assays for detec-
tion of mutations associated with antiviral resistance
and for viral segment reassortments should also be
encouraged.
SUMMARY: Influenza viruses are highly dynamic vi-
ruses. Regular and systematic influenza surveillance in
both humans and animals is essential to provide a more
comprehensive picture of the prevalent influenza vi-
ruses. To better prepare for the next pandemic, we
should develop some simple and easy-to-use tests for
characterizing newly emerging influenza viruses.
2012 American Association for Clinical Chemistry
Influenza is a contagious respiratory disease responsi-
ble for pandemics and seasonal epidemics, and
it thereby imposes a considerable economic burden
through loss of productivity and the costs of associated
medical treatments. Because of the relatively frequent
zoonotic transmissions of influenza viruses from avian
poultry to humans, attention had been primarily fo-
cused before 2009 on avian influenza viruses (e.g., H5,
H7, and H9 subtypes, which have the potential to cause
pandemics) (1 ). Nature, however, caught us by sur-
prise in that the first pandemic of this century was
caused by a swine virus that is of the same subtype as
human seasonal H1N1 virus (2 ). This occurrence also
reflects the highly unpredictable nature of the influenza
A virus. In this review, we discuss the nature of influ-
enza viruses and the challenges of developing molecu-
lar diagnostic tests for detecting new pandemic strains.
Influenza Viruses
The influenza viruses belong to the family Orthomyxo-
viridae (3 ). There are 3 types of influenza viruses: A, B,
and C. These 3 types exhibit different degrees of anti-
genic variations, host specificity, and pathogenicity.
Type A virus undergoes much faster evolution and
shows more antigenic variation than the other 2 types.
Influenza B and C viruses usually infect humans and
they are not discussed further in this article. By con-
trast, type A virus infects a wide range of avian and
mammalian species. The influenza A viral genome
contains 8 RNA segments encoding at least 10 viral
proteins. These segments have been numbered accord-
ing to length, in descending order. The fourth and sixth
viral segments encode hemagglutinin (HA)2 and neur-
aminidase (NA) surface viral glycoproteins, respec-
tively. Influenza A viruses are classified into different
subtypes depending on the nature of the HA and NA
proteins that they carry. Aquatic fowls are believed to
be the natural reservoirs for all influenza A viruses.
1 Centre of Influenza Research and School of Public Health, Li Ka Shing Faculty of
Medicine, the University of Hong Kong, Hong Kong Special Administrative
Region, Peoples Republic of China.
* Address correspondence to this author at: Centre of Influenza Research, School
of Public Health, LKS Faculty of Medicine, the University of Hong Kong, 21
Sassoon Rd., Hong Kong, Peoples Republic of China. Fax 852-28551241;
e-mail [emailprotected]
Received July 4, 2012; accepted August 24, 2012.
Previously published online at DOI: 10.1373/clinchem.2012.182626
2Nonstandard abbreviations: HA, hemagglutinin; NA, neuraminidase; PB1, poly-
merase 1; NP, nucleocapsid protein; NS, nonstructural protein; PB2, polymerase
2; PA, polymerase PA; SA, sialic acid; OIE, World Organization for Animal Health;
FAO, Food and Agriculture Organization of the United Nations; GISRS, Global
Influenza Surveillance and Response System; RT, reverse transcription; LAMP,
loop-mediated isothermal amplification.
Clinical Chemistry 58:11
15271533 (2012) Mini-Review
1527
There are 16 HA and 9 NA subtypes found in avian
influenza viruses. Recently, an influenza virus with
novel HA and NA subtypes was detected in bats, sug-
gesting that the diversity of animal influenza virus is
larger than previously thought (4 ). Of these HA and
NA subtypes, only H1N1, H2N2, and H3N2 viral sub-
types are known to have established stable lineages in
humans since the last century (5 ).
Influenza pandemics occur at unpredictable inter-
vals and are associated with high infection attack-rates
of variable severity. There have been 4 influenza pan-
demics over the last 100 years, namely, the Spanish flu
in 1918, the Asian flu in 1957, the Hong Kong flu in
1968, and the recent pandemic H1N1 in 2009. All of
these pandemics were initiated by the spread of an
antigenically novel HA from animal sources to hu-
mans. The most catastrophic among these 4 pandemics
was the 1918 pandemic, which cost more than 40 mil-
lion lives and had a fatality rate of about 2.5%. The
other 2 pandemics in the 20th century caused about 1
million deaths each. The recent 2009 pandemic was
3 4 orders of magnitude milder, with case fatality in
Hong Kong ranging from 0.03% in older adults to as
low as 0.0004% in children 514 years old (6 ). Al-
though the 2009 pandemic is considered to have been a
mild one, more than 50% of Hong Kong school-going
children were found to have been infected by the pan-
demic H1N1 during the first wave of attackjust 5
months after the pandemic alert was issued and long
before vaccines were available in adequate quantities. A
pandemic of higher severity (e.g., H5N1) may cause
globally catastrophic impacts to human and economic
health. The World Bank has estimated a pandemic with
virulence comparable to that of the 1918 Spanish flu
will negatively impact the global gross domestic prod-
uct by 4.8% and lead to the loss of over 70 million lives
in the pandemic year alone (7 ).
Animal Reservoirs as a Source for Influenza
Pandemics
Except for the bat influenza virus, all mammalian in-
fluenza viruses, such as equine, swine, and human in-
fluenza viruses, are believed to have evolved from avian
influenza viruses. Owing to the nature of their genome,
influenza viruses can exchange their gene segments in
coinfected cells, thereby generating progeny viruses
with new genotypes. These gene reassortment events
play a key role in the evolution of influenza A virus and
have direct impacts on human health (8 ). The huge
viral gene pool in the avian population is believed to be
responsible for the genesis of pandemic viruses. Se-
quence analysis of the 1918 pandemic H1N1 virus sug-
gested that all 8 gene segments of this virus might be of
avian origin (9 ). However, other investigators have
suggested that this virus might be a reassortant between
avian and mammalian influenza viruses (10 ). By con-
trast, it is widely accepted that the pandemic human
H2N2 and H3N2 viruses were reassortants between
avian and human seasonal influenza A viruses. For the
1957 pandemic, a previously circulating H1N1 human
strain incorporated the avian influenza A virus genes
hemagglutinin (HA) (subtype H2), neuraminidase
(NA) (subtype N2), and polymerase 1 (PB1). Similarly,
avian HA (subtype H3) and PB1 genes were introduced
into a H2N2 human virus and caused the H3N2 pan-
demic in 1968. The H1N1/2009 virus is a product of
multiple reassortments between avian, swine and hu-
man influenza viruses (11 ). The HA, nucleocapsid pro-
tein (NP), and nonstructural protein (NS) genes are in
the classical swine lineage and the NA and M genes are
in the avian-like Eurasian swine H1N1 lineage. The
polymerase 2 (PB2), PB1, and polymerase PA (PA)
genes of pandemic H1N1/2009 are from the North
American swine triple H3N2 reassortant virus in which
the PB1 gene originated from human seasonal H3N2.
This complex genotype suggests that the ancestors of
this virus might undergo multiple gene reassortments
in pigs. These sequencing results strongly suggest that a
swine H1N2, closely related to the North America
H3N2 triple reassortant, might be one of the precursor
viruses. However, due to insufficient surveillance of
swine influenza viruses, exact reassortment events are
yet to be investigated (11 ).
There is a host barrier to prevent avian influenza
from infecting humans. Many avian viruses grow
poorly or even are noninfectious to humans (12 ). With
a few exceptions like highly pathogenic avian H5N1
viruses, direct spread of influenza from birds to hu-
mans is uncommon. The control of the host restriction
is a polygenic trait, but the receptor-binding specificity
of influenza is one of the major determinants. In gen-
eral, avian influenza HA prefers to bind to -2,3-linked
sialic acid (SA), which is prevalent in duck intestines,
whereas human influenza HA prefers to bind to -2,6-
linked SA, which is highly expressed in the upper respi-
ratory tract of humans. Hence, the SA binding prefer-
ence of the virus and the SA expression pattern of the
host have great influence on viral tropism and host
specificity. Interestingly, the swine respiratory tract
possesses receptors for both avian and human influ-
enza A viruses. Experiments involving infections of
many avian and human influenza viruses have also
demonstrated that pigs are susceptible to influenza vi-
ruses of zoonotic origins (13 ). Pigs are, therefore, pro-
posed to be the mixing vessels that can support coin-
fection, replication, and reassortment among human,
avian, and swine viruses. Apart from their role in the
pandemic H1N1/2009, pigs are believed to have func-
tioned as an intermediate host for H2N2 and H3N2
Mini-Review
1528 Clinical Chemistry 58:11 (2012)
pandemics. In addition to these outbreaks, zoonotic
transmissions of swine influenza viruses to humans
have also been reported sporadically (14 ). Of current
particular concern are the recent human cases caused
by a novel reassortant of pandemic H1N1/2009 and
swine H3N2 viruses (15 ). Hence, reassortments of in-
fluenza virus in pigs might have direct implications for
human health.
Although the last pandemic has reiterated the im-
portance of pigs for the genesis of pandemic strains,
one should not overlook the role of poultry. Direct
zoonotic transmissions of influenza viruses from poul-
try to humans (e.g., H5N1, H7N7, and H9N2) have
been detected. In some of these events, limited human-
to-human H5N1 transmissions were also detected.
Furthermore, some of these avian viruses might re-
quire only a few mutations to adapt to mammalian
hosts in experimental settings. Thus, it is possible that a
new pandemic strain may directly emerge from an
avian host.
Influenza research over the last 3 decades has iden-
tified several factors to help assess whether a particular
animal virus might have a pandemic potential (16 ).
These risk factors include: the prevalence of the virus in
livestock; its ability to bind to receptors in the human
airway; its ability to grow in human cells/organs; its
airborne transmissibility between ferrets (the best ani-
mal models for studying human transmission because
they can transmit influenza viruses and develop clinical
signs similar to those seen in humans including fever,
rhinitis, and sneezing); and its ability to reassort with
human influenza viruses. Functional hemagglutinin-
neuraminidase balance has been suggested to facilitate
the transmission of viruses by droplets (17 ). Viruses
with a history of frequent zoonotic transmissions to
humans and with surface glycoproteins that can escape
from herd immunity in humans are also classified into
the high-risk category. In addition, viruses that either
carry known human-adapted mutations or associate
with severe disease outcomes are also considered to be
capable of causing a pandemic threat.
It can never be overemphasized that a great major-
ity of human infections caused by animal influenza vi-
ruses are associated with viruses circulating in live-
stock. It is very likely that yet another pandemic strain
will emerge from these populations. For better influ-
enza pandemic preparedness, it is extremely vital to
have a comprehensive picture of the animal influenza
viruses circulating in both animal and human popula-
tions. For this reason, the World Organization for An-
imal Health (OIE) and the Food and Agriculture
Organization of the United Nations (FAO) jointly es-
tablished an international network, the OFFLU (OIE/
FAO Network of Expertise on Animal Influenza)
(http://www.offlu.net/) to support and coordinate
global efforts to prevent, detect, and control influenza
viruses in animals. In addition, the WHO has a Global
Influenza Surveillance and Response System (GISRS)
(http://www.who.int/influenza/gisrs_laboratory/) for
monitoring both human and animal influenza viruses
in humans, and one of the collaborating centers in this
network is particularly focusing on the ecology of ani-
mal influenza viruses. In 2010, the FAO, OIE, and
WHO jointly issued a tripartite concept note to
strengthen the concept of One World, One Health
(http://www.who.int/influenza/resources/documents/
tripartite_concept_note_hanoi_042011_en.pdf).
Detection of Influenza Viruses
Influenza surveillance requires the monitoring of cir-
culating virus strains using diagnostic methods. It is the
first stage of controlling an outbreak because efficient
diagnosis can minimize spread. In a clinical or field
setting, rapid detection of influenza allows physicians
and scientists to initiate prompt treatment, implement
infection control strategies, decrease health costs, and
reduce risks to healthcare workers and the wider com-
munity. Diagnostic tests are frequently evaluated and
developed for improved sensitivity and reliability, be-
cause the influenza virus is constantly evolving. For a
more comprehensive review on the molecular diagnos-
tic tests for influenza virus, readers are encouraged to
read recent the excellent reviews contributed by others
(18 21 ).
There are several methods for diagnosis of influ-
enza and each is used depending on the available re-
sources and context. Clinical signs are commonly used
to establish a tentative diagnosis. However, this diag-
nosis must be confirmed with specific tests, because the
manifestations of disease are variable and could be a
result of other viral or bacterial infections. Clinical on-
set is characterized by high fever, cough, headache,
malaise, and inflammation of the upper and lower re-
spiratory tract. These symptoms can last for several
days. Complications in more susceptible demographic
groups such as the elderly include pneumonia, hemor-
rhagic bronchitis, and death. Viral replication peaks at
24 h, and after 6 days there is little viral shedding. This
poses one of the most important challenges for diag-
nostics, because individuals rarely present themselves
to a doctor in the early stages of illness.
Diagnostic tests include serological tests, virus iso-
lation, rapid antigen tests, and molecular tests (18 21 ).
Serological methods rely on the development of an an-
tibody response that can take several days to develop,
and in some cases the disease is so rampant that an
antibody response is not even formed. Influenza virus
is known to replicate in Madin-Darby canine kidney
cells, and a cytopathic effect can be identified after 23
Diagnostic Challenges for Pandemic Preparedness Mini-Review
Clinical Chemistry 58:11 (2012) 1529
http://www.offlu.net/
http://www.who.int/influenza/gisrslaboratory/
http://www.who.int/influenza/resources/documents/tripartiteconceptnotehanoi042011en.pdf
http://www.who.int/influenza/resources/documents/tripartiteconceptnotehanoi042011en.pdf
days. This virus isolation can take several days and
must be performed in a biosafety level 2 (seasonal in-
fluenza) or above (highly pathogenic or pandemic in-
fluenza viruses) facility. This process also allows prop-
agation of the virus for further investigation, such as
sequencing and biological characterization.
Other nonnucleic acid methods detect a compo-
nent of the virus. Rapid antigen assays include com-
mercially available point-of-care tests. These tests uti-
lize commercial antibodies against influenza antigens.
They can produce a result within 20 min but are rela-
tively expensive and have variable performance. An-
other drawback is that many of these tests indicate only
the presence of influenza and do not provide informa-
tion about subtype, although some H5N1-specific
rapid diagnostic tests are becoming available. Recently,
rapid antigen tests for subtyping human H1 and H3
and avian H5 have been reported. The performance of
these tests is yet to be improved (22 ). Nonetheless,
rapid antigen tests are useful in a clinical setting and
can expedite the decision to administer antiviral
treatment.
Among diagnostic tools available for detecting in-
fluenza, molecular tests are the most sensitive and
rapid. These tests can be performed with high through-
put at a moderate cost. These tests amplify target nu-
cleic acids to allow identification. Molecular tests in-
clude reverse-transcription PCR (RT-PCR), real-time
RT-PCR, nucleic acid sequence based amplification,
loop-mediated isothermal amplification (LAMP), mi-
croarray, and pyrosequencing. As a prelude to a pan-
demic, molecular diagnosis of influenza, in particular,
should be harmonized globally. In a pandemic scenario
there will be a sudden increase of samples to be tested,
causing pressure for faster turnaround and the need for
accurate and cost-effective results. Molecular diagnos-
tic techniques show the most potential to meet these
challenges.
RT-PCR is a technique widely used in laboratories
to amplify RNA, and the necessary primer sequences
are often publicly available. An extension of RT-PCR is
real-time RT-PCR, for which quantitative results can
be generated in real time. Real-time RT-PCR is the
most sensitive, informative technique yielding rapid
results, with the only drawback being that the reagents
and start-up cost are high. Recent studies have indi-
cated that viral loads might associate with disease out-
comes, which suggests that the quantitative results
might be a useful prognostic indicator in some clinical
cases. The RT-PCR method uses fluorescent dyes or
probes to detect the amplicon; several platforms are
available, such as SYBR green, Taqman, and molecular
beacons. The systems for RT-PCR and real-time RT-
PCR mainly detect the M gene, which is highly con-
served, for influenza A virus detection. Primers and
probes designed to detect influenza subtypes target
conserved regions of the HA gene. However, these sub-
typing assays are normally designed for viruses with
known significance to human health.
Multiplex PCR is a modification of PCR for which
several primer sets are used concurrently to detect the
presence of several gene targets. This method is useful
in a diagnostic laboratory where multiple pathogens in
a single sample can be investigated. For example, a
multiplex PCR assay system to detect H5N1 and other
human respiratory pathogens has been developed (23 ).
PCR-EIA is an alternative method that identifies
PCR amplicons by use of a biotinylated RNA probe
through an enzyme immunoassay (24 ). This method
has higher sensitivity than the traditional RT-PCR
methods, but it was developed only for detection of the
M gene and limited HA subtypes. In addition, there is
an added cost of the enzymatic reagents.
Nucleic acid sequence based amplification also
employs reverse transcriptase to generate a cDNA (25 ).
The cDNA is synthesized by a primer containing a pro-
moter for the T7 polymerase. The second primer at-
taches to the single-stranded cDNA and the reverse
transcriptase synthesizes a second strand of DNA. The
reaction also contains RNase H to destroy the original
RNA template. The T7 polymerase binds to the double-
stranded DNA to generate a complementary RNA that
acts as a template as the reaction continues in a cyclic
fashion. This isothermal method does not require ex-
pensive PCR instrumentation, but this approach might
be limited by RNA secondary structure, which some-
times makes primer design and multiplexing difficult.
The LAMP assay is also a nucleic acidamplification
method that uses reverse transcriptase and DNA poly-
merase in a single-step isothermal reaction (26, 27 ).
The approach requires more extensive effort for primer
designs and assay optimizations, but the method has
been explored for several reasons. Specifically, it is a
technique that is compliant with the challenges of in-
fluenza surveillance in places with limited resources.
The LAMP assay is an alternative method that is rapid
with high specificity to a target and can be used in a
front-line clinical laboratory without the need for ex-
pensive, complex equipment and specialized staff.
Recently, several studies have demonstrated the
feasibility of using microarrays for the detection and
subtyping of influenza A viruses (28 ). In addition,
some of these arrays could be used to differentiate
highly pathogenic H5N1 from less pathogenic H5N1
viruses (29 ). Pyrosequencing is another new technique
that provides high-throughput screening of influenza
viruses (30 ). Assays for antiviral resistance marker de-
tection have been developed. These approaches would
be useful for conducting influenza surveillance.
Mini-Review
1530 Clinical Chemistry 58:11 (2012)
All molecular assays are limited by primer and
probe design and RNA quality. Target sites are subject
to mutations that reduce primer specificity. Therefore,
molecular assays must always be evaluated and, if nec-
essary, revised. Sample quality is also an important fac-
tor, because tests show differential detection limits de-
pending on the origin and quality of the sample.
Despite an optimized methodology for molecular as-
say, there are several points at which errors can occur.
Sample handling and storage, nucleic acid extraction,
enzyme transcription fidelity, enzyme inhibition, and
contamination are all critical points for optimal detec-
tion of influenza virus.
Presently, there is no standardized universally ac-
ceptable influenza detection technique (either in poul-
try or in humans) that can allow easy comparison of
surveillance studies in different countries. Because the
resources and technical skills available in each situation
are highly variable, it is unlikely that a consensus detec-
tion method will emerge. For the time being, the WHO
has recommended techniques that are available on its
website.
Apart from providing guidelines and technical
support for influenza diagnosis, GISRS monitors hu-
man influenza activities around the world throughout
the year. Representative and/or important viruses iso-
lated from these surveillance activities will be further
characterized in various WHO collaborating centers
under this network. In particular, it is recommended
that novel and/or unsubtypable influenza viruses be
immediately referred to appropriate reference labora-
tories under GISRS for further characterization (31 ).
Information from these activities is critical for labora-
tory diagnostics, vaccines, antiviral susceptibility, and
risk assessment.
Challenges for Pandemic Preparedness and the
Next Pandemic
The 2009 pandemic might have helped further
strengthen the capacity for diagnosing influenza in
many diagnostic laboratories. Standard assays for in-
fluenza detection and human influenza subtyping are
normally well established for use in clinical diagnostic
laboratories. Although several viral subtypes are recog-
nized as viruses of high pandemic potential, our exist-
ing knowledge is not sufficient to provide an accurate
prediction of the next pandemic strain. Clinical diag-
nostic assays specific for HA subtypes that have a his-
tory of infecting humans have been established. These
assays, however, are entirely based on known HA se-
quences. Some of these assays are viral-lineage specific
and their usefulness is limited to regional applications.
Hence, the question arises, are we really ready for the
next pandemic that may emerge unpredictably? Do re-
gional or routine diagnostic laboratories have the ca-
pacity to analyze unusual influenza viruses in humans?
Is it possible for us to achieve early detection of novel
human influenza viruses once they have started to
emerge in humans?
PCR-based detection assays specific for each HA
and NA subtype have been established to detect viral
sequences in animal specimens (32, 33 ). These assays
would be useful to analyze unsubtypable human influ-
enza A viruses. The amplicons generated from these
assays are usually large, and the products would be ex-
tremely useful for downstream characterization (e.g.,
sequencing). However, the ability to amplify long frag-
ments in these assays is often compromised by the assay
sensitivity. Although these assays would be useful to
detect isolated virus cultures, they might not be highly
suitable for clinical practice. Recently, Tsukamoto et al.
(34 ) reported the development of SYBR green based
real-time RT-PCR for subtyping of all HA and NA
genes of avian influenza viruses. These assays were eval-
uated with an extensive number of avian viruses of dif-
ferent subtypes. In addition, a great proportion of
avian fecal specimens positive for influenza A viruses
were correctly subtyped by these assays. However, it is
not known how well these assays work with human
clinical samples. In addition, there is uncertainty about
whether these assays are able to detect viruses circulat-
ing in nonavian species. Nonetheless, the development
of a panel of diagnostic tests that can detect all HA and
NA subtypes should be encouraged. Ideally, these pan-
els should cross-react with all animal, but not human,
influenza viruses. These assays should be robust and
easy to establish in basic clinical laboratories.
Because the above HA- and NA-subtypespecific
primers are not guaranteed to cross-react with the cor-
responding HA and NA subtypes, alternative strategies
that may be helpful for characterization of unsub-
typable human influenza viruses should also be consid-
ered. The M and NP genes of influenza viruses are
highly conserved; several universal RT-PCR assays for
these 2 genes have been established. Sequencing infor-
mation deduced from these amplicons can sometimes
help to partly reveal the origins of some unsubtypable
viruses. Although sequence data generated from these
universal assays might allow the development of more
specific primers for virus detection, these assays are un-
able to specify the HA and NA subtypes.
In practical terms, universal primers that can
cross-react with all HA subtypes would provide addi-
tional advantages for developing tests that are more
specific for newly identified human influenza viruses
(35, 36 ). The identity of these amplicons can be deter-
mined by performing sequencing or microarray analy-
sis in a timely manner. In addition, such studies also
Diagnostic Challenges for Pandemic Preparedness Mini-Review
Clinical Chemistry 58:11 (2012) 1531
might provide useful information for developing rele-
vant serological assays for virus detection.
Apart from the above classical PCR-based tests,
many other alternative approaches might be used to
deduce additional sequencing information of unsub-
typable human influenza viruses. Universal primers
for full-genome amplifications have been reported
(37, 38 ). The deep-sequencing approach can also pro-
vide very comprehensive information. However, be-
cause these assays are technically more demanding or
require additional diagnostic infrastructure, it would
not be easy to implement these tests in general diagnos-
tic laboratories. In addition, one should also consider
the cost, time, and testing materials needed for these
assays.
Determining the antiviral susceptibility of these
novel human influenza viruses would be critical for
prompt antiviral treatments. Currently, there are 2
classes of antiviral agents for influenza, M2 ion channel
inhibitors and NA inhibitors. These drugs are experi-
mentally effective against most animal influenza vi-
ruses. Molecular assays (e.g., real-time RT-PCR assay
and pyrosequencing) for specific mutations known to
confer antiviral resistance (e.g., H275Y in N1, E119V in
N2, and S31N in M2) have been established (39, 40 ),
but these assays are unlikely to be useful for detecting
such mutations in a pandemic strain that is yet to
emerge. Furthermore, viruses containing resistant mu-
tations located at other positions/segments would not
be picked up by these assays. Thus, molecular tests for
antiviral resistance should be interpreted with great
caution. Currently, phenotypic tests, such as neur-
aminidase inhibition assays, are still the preferred op-
tion for testing viruses that are suspected to have anti-
viral resistance mutations. This is particularly true for
viruses that emerge from animals, because our under-
standing about animal influenza viruses is not as com-
prehensive as that for human influenza viruses. In ad-
dition, these tests can be done only on virus cultures,
where some virus mutants might be selectively en-
riched through virus propagation in Madin-Darby ca-
nine kidney cells. Therefore, data obtained from these
assays may not always represent the true nature of the
original specimen.
Molecular tests that allow rapid genotyping of in-
fluenza viruses would also be useful to detect reassor-
tants between human and animal viruses (41, 42 ). Re-
assortment events should be closely monitored by
real-time surveillance, because novel reassortants may
have altered virulence and/or transmissibility. Geno-
typing assays for influenza viruses in humans and poul-
try should be encouraged.
Although in this review we have focused only on
the challenge of molecular testing of influenza viruses,
serological tests are equally important for clinical diag-
nosis and public health surveillance. In addition, the
role of virus isolation is also extremely important for
the control of influenza viruses. Much of our under-
standing of human and animal viruses still relies heav-
ily on basic virus culture techniques. On the other
hand, characterization of live infectious virus isolates
still plays a key role in our current production of influ-
enza vaccine. Although there are many advantages of
using molecular tests for characterizing and detecting
influenza viruses, key regional laboratories are encour-
aged to maintain the basic virological techniques and
expertise to strengthen global pandemic preparedness.
Author Contributions: All authors confirmed they have contributed to
the intellectual content of this paper and have met the following 3 re-
quirements: (a) significant contributions to the conception and design,
acquisition of data, or analysis and interpretation of data; (b) drafting
or revising the article for intellectual content; and (c) final approval of
the published article.
Authors Disclosures or Potential Conflicts of Interest: Upon man-
uscript submission, all authors completed the author disclosure form.
Disclosures and/or
