Immunity to poliovirus after infection and
Immunity to poliovirus after infection and
Afweer tegen poliovirus na infectie en
(met een samenvatting in het Nederlands)
ter verkrijging van de graad van doctor aan de Universiteit Utrecht
op gezag van de Rector Magnificus, Prof. Dr. H.O. Voorma,
ingevolge het beluit van het College voor Promoties
in het openbaar te verdedigen
op dinsdag 27 april 1999 des ochtends te 10:30 uur
Martina, Maria, Petronella, Theresia Herremans
Geboren op 21 januari 1970, te ‘s-Hertogenbosch
Promotor: Prof. Dr. J. Verhoef
Copromotores:Dr. M.P.G. Koopmans
Dr. A.M. van Loon
The research described in this thesis was conducted at the Research Laboratory for
Infectious Diseases, National Institute of Public Health and the Environment (RIVM),
Bilthoven, The Netherlands, and was made possible by financial support from the
Foundation for the Advancement of Public Health and Environment (SVM).
Printing of this thesis was sponsored by the RIVM
Cover picture: WHO, used by permission
Printed by Addix, Wijk bij Duurstede, The Netherlands
Voor Tante Riet
Table of Contents
Evaluation of a Poliovirus-Binding Inhibition
assay as an alternative for the virus neutralisation
Differences in the antibody responses to antigenic
sites 1 and 3 of serotype 3 poliovirus after OPV or
IPV vaccination and after natural exposure45
Poliovirus-specific IgA in persons vaccinated with
inactivated poliovirus vaccine (IPV) in The
Induction of mucosal immunity by inactivated
poliovirus vaccine is dependent on previous
mucosal contact with live virus71
poliomyelitis patients and their direct contacts for
the present surveillance of acute flaccid paralysis
from diagnostic investigations of
List of Publications
The early days of poliomyelitis: some historical notes
Sporadic cases of paralytic poliomyelitis have been reported for at least as long as
recorded history . Despite its long history, however, poliomyelitis has had its most
notable effect on humanity within the past one hundred years. While the cumulative
number of poliomyelitis patients world-wide had reached an estimated 10 million
cases by the beginning of this century , no effective vaccine for this disease
existed before the 1950s and ignorance about the route of transmission often
hampered attempts to control its spread.
Poliomyelitis was, therefore, greatly feared, and the paralysis produced by this disease
(especially in the young) was a familiar sight during previous decades . As
societies have improved their methods of sanitation (thereby eliminating a number of
diseases in the process) the likelihood has increased that individuals will be exposed
to poliovirus later rather than earlier in life, if at all. These patterns of exposure have
resulted in a situation whereby this paralytic disease is no longer endemic in the
western world, but occurs instead in sporadic epidemics .
Today, thanks to increased levels of hygiene and vaccination, poliomyelitis is rare in
western countries. The cases that do occur are caused mainly by vaccine-associated
disease within countries that use the live attenuated vaccine, or by wild-type virus
infections within groups that refuse vaccination for religious reasons [5,61,74].
Poliomyelitis is also rapidly decreasing in most developing countries due to the World
Health Organisation’s vaccination campaigns .
Poliomyelitis presented a challenge to the scientific world for many years, as
scientists and epidemiologists struggled to understand the cause of this disease. The
first breakthrough occurred in 1909, when Dr. Karl Landsteiner discovered that
poliomyelitis was caused by a viral infection of unknown origin . Dr. John F.
Enders, along with his colleagues Dr. Thomas Weller and Dr. Frederick Robbins, laid
the foundation for the development of poliovirus vaccines in 1949 when they
demonstrated the growth of poliovirus in cultures of non-neural cells . Work by
other researchers using in vitro viral culture subsequently followed from this
important discovery, and Enders and his colleagues were rewarded for their work with
the Nobel Prize in 1954.
A second important discovery occurred in 1949, when investigators were able to
differentiate between the three different serotypes of poliovirus . In 1952, Dr. Jonas
Salk succeeded in developing a formalin-inactivated poliovirus vaccine (IPV) and in
1955 this inactivated vaccine was approved for the vaccination of children against
poliomyelitis [82,83]. Vaccination campaigns in the USA and Europe soon followed,
with great success. In 1960, the live attenuated vaccine strains developed by Dr.
Albert Sabin were incorporated into a live attenuated oral vaccine (OPV) . This
OPV vaccine, because of its low cost, ease of use, safe administration and
effectiveness against infection is now the vaccine of choice in the world-wide
vaccination campaigns run by the World Health Organisation (WHO) [20,103,104].
The structure of poliovirus
Polioviruses belong to the genus enterovirus within the family Picornaviridae. With a
diameter of 27-30 nm they are among the smallest viruses known, and contain a
single-stranded RNA molecule of positive polarity linked to a small protein at the 5’
region of the genome designated as the genomic virion protein (VPg). The entire
nucleotide sequence has been determined and the total genome consists of 7440, 7440
and 7435 nucleotides for serotypes 1, 2 and 3 respectively . The viral capsid
consists of 20 copies of each of the four structural virion proteins (VP1, VP2, VP3
and VP4) . These viral capsid proteins protect the encapsidated nucleic acids from
degradation and interact with a specific cellular receptor on susceptible host cells: the
CD155 molecule .
Polioviruses can be classified into three distinct
serotypes based on their reaction to reference
panels of neutralising antisera . Virus
neutralising antibodies against one of the three
serotypes do not protect against the other types,
although some cross-reactivity
described between the serotypes 1 and 2 [36,83].
The epitopes responsible for inducing poliovirus
neutralising antibodies are located at the end of
the loops on the three structural proteins: VP1,
VP2, and VP3 . Because VP4 is located
entirely on the inside of the viral capsid, it plays
no known role in the induction of poliovirus-
neutralising antibodies. VP1 is the most exposed
surface protein and plays a major role in the
induction of neutralising antibodies for all three
poliovirus serotypes . Three antigenic sites
(epitopes) involved in virus neutralisation have been identified on polioviruses based
on studies with Sabin-derived mutant viruses resistant to neutralisation by monoclonal
antibodies [40,63] [Figure 1].
Figure 1. Identified B and T cell epitopes on the structural proteins of poliovirus
1302 1 271 1 238 169
= trypsin cleaving site
89 100 220 222 169 170 58 59 6 35
89 100 14 28 189 210 6 35
89 100 286 290 169 17058 59 6 35
= B cell epitopes= T cell epitopes
Dr. Jonas Salk
Monoclonal antibodies induced in mice have determined that antigenic site 1,
composed of amino acids 89 to 100 of VP1, is the major immunogenic site for
serotype 2 and 3 polioviruses . This site is usually immunorecessive in serotype 1
. Site 2 is a complex site combining residues 220 to 222 from VP1 (site 2a) with
residues including 169 and 170 and others on VP2 (site 2b) . Both site 2a and 2b
have been detected in serotype 1 poliovirus, while only site 2b has been detected in
serotype 3 poliovirus. Site 3 is also a complex site and includes the residues 286 to
290 from VP1 (site 3a), and residues 58 and 59 from VP3 (site 3b). Both sites 3a and
3b have been detected in serotype 3 poliovirus, while as yet only site 3b has been
detected in serotype 1 poliovirus . The immunological relevance of these three
antigenic sites in humans is not clear [Figure 1].
It is reported that trypsin, present in the intestinal fluids, can cleave both serotype 1
and serotype 3 polioviruses at antigenic site 1 at residue arginine-98 [26,45,80]. While
the virus in both cases retains its infectivity, the antigenic properties of the poliovirus
are drastically altered, and trypsin-cleaved viruses are not neutralised or
immunoprecipitated by monoclonal antibodies to site 1 of non-treated virions .
Pathogenesis of poliovirus infection
Polioviruses have a restricted host range and humans are the only reservoir of
naturally circulating poliovirus. Poliovirus can infect and cause flaccid paralysis in
chimpanzees and cynomolgus monkeys, but the remaining (wild) populations of these
animals are not large enough to sustain poliovirus circulation in the absence of human
infections . In monkeys, paralysis is initiated most readily by direct inoculation
into the brain or spinal cord, and infection by the oral route is usually asymptomatic
[7,62]. Poliovirus is, however, excreted in the throat and stool after oral infection .
The incubation period of poliovirus is usually between seven and 14 days (range two
to 35 days) . Poliovirus can be detected in the stool for five to six weeks following
infection and is present in the pharynx for one to two weeks after infection [1,41,62].
Transmission occurs mainly via the faecal-oral route and the virus can spread to other
people through contaminated water or food . Survival of poliovirus in the
environment is highly variable, but viral inactivation is usually complete within
months  Following oral ingestion, poliovirus first multiplies in the pharynx and
the small intestine. After initial and continuing replication, probably in lymphoid
tissue of the pharynx and gut (Peyer’s patches), the virus is able to spread to other
lymph nodes until it is eventually detected in the bloodstream (viremia) [8,99].
Electron microscopy has demonstrated that poliovirus particles specifically adhere to
and are endocytosed by intestinal M-cells . These data suggest that M-cells are the
site of poliovirus penetration of the intestinal epithelial barrier in humans.
Viremia can be detected as early as two to three days after infection. Once the virus
has reached the bloodstream, the anterior horn cells of the spinal cord are at risk for
infection unless sufficiently high levels of neutralising antibodies are present in the
circulation . Not much is known about the manner in which polioviruses are able to
cross the blood brain barrier. Data from transgenic mouse experiments show that
polioviruses permeate through the blood brain barrier at a high rate, independently of
the poliovirus receptor . It has also been proposed that polioviruses may cross the
blood brain barrier into the CNS via infected monocytes [39,25]. Meningitis or
paralysis can occur when neutralising antibodies are not able to block infection of the
central nervous system. Inflammation occurs secondary to the infection of the nerve
cells; the infiltrations are mainly lymphocytes, with some polymorph nuclear cells and
plasma cells . In addition to pathological changes in the nervous system,
hyperplasia and inflammatory lesions of lymph nodes, Peyer’s patches and other
lymph follicles in the intestinal tract are frequently observed . Some degree of
recovery of motor functions may occur over the subsequent six months, but paralysis
still present at the end of this time remains permanent.
Most poliovirus infections (90-95%), however, do not result in clinical symptoms.
The infections that do become clinical can be divided into minor illness (4-8%) and
more severe illness (0.1-1%) such as paralysis or meningitis . Minor illness (or
abortive poliomyelitis) is characterised by fever, malaise, sore throat, headache and
vomiting—all symptoms that can easily be mistaken as flu-like. Paralysis and
meningitis are relatively infrequent complications of poliomyelitis. The ratio of sub-
clinical to clinical infection in primary poliovirus infections does not in itself affect
the spread of wild type poliovirus but is important for an accurate assessment of the
extent to which the poliovirus has spread. This ratio varies according to serotype, and
is highest for serotype 3 (estimated at between 4000:1 and 500:1 [13,60,83]). The
lowest ratio is detected for serotype 1 (between 60:1 and 175:1 ) and intermediate
values are found for serotype 2 (1000:1 or higher ).
A sudden increase in muscle atrophy has been observed in 22% to 87% of persons
who have suffered from poliomyelitis, long after their apparent recovery [12,11,53].
Remaining motor neurons take over the function of the lost neurons during the
recovery phases after acute poliomyelitis. Depending on the severity of the damage,
these remaining motor neurons have to innervate more than the usual amounts of
muscle fibres. This compensation mechanism can become exhausted during
subsequent years, leading to new symptoms of muscle weakness now known as post-
polio syndrome [11,12,53]. An alternative explanation suggested by some
investigators is that the reported presence of poliovirus-specific RNA in former
poliomyelitis patients indicates chronic viral infection [50,65].
Laboratory diagnosis of poliomyelitis
Cell culture isolation of poliovirus from the stool or pharynx early in the course of the
disease is diagnostic for poliomyelitis . As the disease progresses, the detection
of the virus in the blood or cerebrospinal fluid (CSF) is also considered to be
diagnostic . The WHO recommends that at least two stool samples should be
obtained from patients suspected of having poliomyelitis in order to increase the
probability of poliovirus isolation. These samples should be taken 24 hours apart as
early as possible in the course of the disease (ideally within the first 15 days after the
onset of disease) [104,105].
Polioviruses grow rapidly in cell culture, and cell destruction (cytopathic effect or
CPE) is usually complete within a few days . The serotype of the isolate is
identified by virus neutralisation tests . Intratypic strain differentiation is
necessary to determine whether the poliovirus isolate is wild or vaccine-related
[93,96]. An important benefit of virus isolation is that the molecular analysis of the
isolated viral genome can help to reveal the origin of the isolated poliovirus. The use
of molecular epidemiological methods has enhanced the precision and reliability of
poliovirus surveillance . Poliovirus genomes evolve rapidly (~10-2 nt
substitutions/genome/year) during replication in humans because of the lack of viral
RNA polymerase proof-reading during viral replication . Studies based upon
nucleotide sequence comparisons have revealed the existence of numerous genotypes
endemic in different regions of the world, enabling the study of transmission routes
The determination of the level of poliovirus-neutralising antibodies in serum is
considered to be the most specific assay for the estimation of protection against
poliovirus-induced disease. It is assumed that an antibody titer of 8 or higher is
sufficient for protection against the induction of paralysis, but it is not clear if persons
with lower titers are also protected . Serological tests may be helpful in
supporting or ruling out a diagnosis of poliomyelitis if serum samples are obtained
early in the course of disease. However, the measurement of neutralising antibodies to
poliovirus for diagnostic purposes is not recommended by the WHO , due to the
fact that a) results are often difficult to interpret because antibody titers are similar in
vaccinated and infected persons, and b) neutralising antibodies appear early in the
course of infection, and seroconversion will have already taken place in many cases at
the time of the first clinical symptoms . Determination of neutralising antibody
titers can, however, be helpful in assessing the level of protection against poliovirus
within a population.
Immunity to poliovirus after natural infection
Upon infection, poliovirus replicates in the epithelium of the pharyngeal and intestinal
mucosa and initiates a process that eventually results in mucosal immunity to
poliovirus . Poliovirus-specific secretory antibodies are produced by plasma cells
originating in the gut-associated lymphoid tissues, mainly from the Peyer’s patches
. The predominant class of immunoglobulin in the secretions of the alimentary
tract is secretory IgA, which engages in neutralising activity against poliovirus
[3,46,66,68]. The association between the presence of poliovirus-specific secretory
antibodies and protection against re-infection with poliovirus has been clearly
demonstrated [46,66-70]. Mucosal immunity provides a local barrier to poliovirus
infection, and therefore forms a first line of defense, preventing the pathogen from
entering the host [62,70,71]. Local immunity, however, is not absolute and can be
overcome by a sufficiently large dose of challenge virus . The persistence of
poliovirus-specific secretory IgA has not been studied extensively. However,
poliovirus-specific secretory antibodies have been detected in nasopharyngeal
secretions 10-15 years after natural infection with wild serotype 1 poliovirus .
Following natural exposure, poliovirus-specific IgM and IgG appear in the serum
about 7-10 days after infection, and sufficiently high levels of these antibodies can
block poliovirus entry into the central nervous system . The IgM response precedes
the IgG response, and peaks at about two weeks after the onset of disease,
disappearing from the serum within 60 days . IgG levels increase steadily until
approximately eight weeks after infection. IgA antibodies appear in the serum two to
six weeks after exposure and remain at low levels .
It is generally believed that once a person has been naturally exposed to wild type
poliovirus they are protected for life from further disease induced by that specific
serotype . The acquired immunity against re-infection, however, incomplete, as
was illustrated by Gelfand et al  in their investigation of households containing
poliomyelitis patients in 1953-1955 (the beginning of the vaccination era). They
demonstrated that 20% of the persons with naturally acquired immunity were
undergoing a poliovirus infection . Verlinde and Wilterdink  discovered in
1959 that excretion of poliovirus following challenge occurred in 31% of naturally-
immune children when OPV serotype 1 was used as a challenge virus, 37% when
serotype 2 was used, and 53% when serotype 3 was used.
Cell-mediated immunity to polioviruses in humans has been incompletely
investigated. Poliovirus-specific cell-mediated immunity as determined with
lymphocyte proliferation assays is seen in the early stages of acute poliomyelitis, but
disappears in most patients after three months . Poliovirus-specific CD4-positive
cells (T-helper cells) could be found in the peripheral blood of participants in studies
where persons were immunised with live attenuated poliovirus vaccine [35,88].
Immune lymphocytes proliferate to polyacrylamide gel purified-capsid proteins VP1,
VP2 and VP3 and in some individuals, to synthetic VP4, indicating the presence of T
cell epitopes in each of these proteins [9,88] [Figure 1]. T cell epitopes adjacent to
each of the B cell antigenic sites in VP1 of poliovirus serotype 3 were identified. T
cell lines generated in response to poliovirus infection were cross-reactive between
the three serotypes. The response to the region adjacent to B cell antigenic site 1
(residues 97 to 114) was found to be immunodominant .
T cell clones induced in mice were found to be either serotype-specific or cross-
reactive between two or all three serotypes . As in experiments conducted with
humans, the T cell clones recognised determinants on the surface capsid proteins VP1,
VP2, and VP3 and the internal capsid VP4 [Figure 1]. One serotype 3-specific T cell
clone recognised an epitope within amino acids 257 and 264 of VP1. Three T cell
epitopes corresponding to residues 14 to 28, 189 to 203, and 196 to 210 were
identified on VP3 of type 2 poliovirus. Four T cell epitopes were mapped to an
immunodominant region of VP4, encompassed within residues 6 and 35. The VP4
epitopes were conserved between serotypes. In contrast, T cell clones that recognised
epitopes on VP1 or VP3 were largely serotype specific.
The exact importance of T cell-mediated immunity to poliovirus and its role in
recovery, protection from re-infection and the destruction of nerve tissue is not
known. Neutralising antibodies are thought to be important for clearing poliovirus
infections because children with agammaglobulinemia get persistent infections .
However, it is not possible to cure these patients or clear the poliovirus from the CNS
even with the infusion of high titered antibody into the cerebrospinal fluid [59,78]. In
addition, poliovirus persistence has occurred in persons with pure T cell deficiencies
and normal immunoglobulin levels and antibody responses . It would appear,
therefore, that T cells or other cellular immune mechanisms play at least a partial role
in the clearance of poliovirus from the CNS.
Vaccination against poliomyelitis
It goes without saying that both the development and the successful use of poliovirus
vaccines have exerted a major influence on the spread of poliovirus. The principal aim
of vaccination was originally to protect the vaccinated individual from disease. Within
the context of vaccination programmes for larger populations, vaccination is able to
enhance herd immunity to such an extent that the chain of transmission can be
inhibited or even interrupted within a given community or country . Two different
vaccines are currently used in vaccination campaigns against poliomyelitis: the
inactivated poliovirus vaccine (IPV) and the live attenuated poliovirus vaccine (OPV).
Dr Jonas Salk developed the first inactivated polio vaccine (IPV) using polioviruses
grown on monkey kidney cells that were subsequently formalin-inactivated .
After extensive field testing, IPV was licensed in the United States in 1955, using the
Mahoney (serotype 1), MEF-I (serotype 2) and Saukett (serotype 3) poliovirus strains
. Dutch polio vaccination with IPV started in 1957, and IPV has been produced at
the RIVM in The Netherlands since 1959. Today, The Netherlands, Finland, Sweden,
France, Iceland, Norway and parts of Canada use this inactivated polio vaccine in
their national vaccination programs. The same strains are still used by all
manufacturers of IPV today with the exception of Sweden, where the Brunenders
strain is used for serotype 1 .
The original IPV, given intramuscularly, contained 20, 2, and 4 D antigen units of
poliovirus serotype 1, 2 and 3 respectively. In 1978, the RIVM introduced a new
culture technique using cells on microcarriers to produce a more potent IPV [95,97].
The IPV used today in The Netherlands contains 40, 4 and 7.5 D antigen units per
dose. A total of six IPV vaccinations are given at 3, 4, 5 and 12 months and 4 and 9
years of age. The enhanced IPV (eIPV) used in other countries contains 40, 8 and 32
D-antigenic units per dose of serotype 1, 2, and 3 respectively .
The antigenic site 1 of serotype 3 poliovirus is immunodominant  and
intramuscular vaccination with the complete inactivated virion will, in theory, mainly
induce antibodies to this antigenic site. Upon mucosal infection with wild-type
poliovirus, trypsin (which is present in the gastrointestinal tract) cleaves the antigenic
site 1 of serotype 3 poliovirus, leaving neutralising antibodies to this site useless
[26,45,80]. For this reason it has been suggested that IPV might be supplemented
with trypsin-cleaved serotype 3 poliovirus antigen in order to achieve a vaccine with
sufficiently broad immunogenicity .
The attenuated polioviruses used in the live vaccines are no longer neurovirulent and
rarely cause poliomyelitis (vaccine-associated poliomyelitis) . This vaccine is
applied orally with the advantage that it replicates in the host, inducing local
immunity in the gut and at other mucosal sites. Further, the Sabin vaccine may
contribute to the immunisation of subsequent contacts because is spread faecally .
OPV-induced mucosal immunity is important because it is able to reduce the spread
of wild type polioviruses upon its (re)introduction into a population, thereby assisting
in the creation of sufficient herd immunity . In 1991, the success of controlling
poliomyelitis in the Americas led the EPI Global advisory group to recommend, on a
global basis, the formulation of trivalent OPV with 106, 105 and 105.8 TCID50 per
dose of serotypes 1, 2 and 3, respectively . The schedule recommended today is
one dose at birth and 3 doses of OPV at 6, 10 and 14 weeks of age.
There are some recognised problems with the distribution of OPV. Breaks in the cold-
chain in developing countries, for example, can lead to loss of vaccine effectiveness in
tropical countries . Further, malnutrition and infection with other enteroviruses
are known to interfere with the effect of OPV vaccination . The greatest
disadvantage of OPV, however, is the risk of back-mutation to neurovirulent strains
and the introduction of a wide variety of live mutant viruses into the population
[20,90,108]. However, the risk of vaccine-associated paralytic poliomyelitis (VAPP)
is low, with only one case reported for every 3.3 million doses of trivalent OPV that
are distributed [20,90,108]. Serotype 3 is most commonly associated with paralysis in
vaccine recipients, while serotype 2 has primarily been associated with paralysis
among contacts of OPV recipients . The risk for VAPP is highest following the
first dose of OPV (1:570,000) and among persons with primary B cell
immunodeficiencies [90,108]. Despite the risk of (OPV) vaccine-associated paralysis
this vaccine was chosen by the WHO because of its low cost, ease of use, safe
administration and effectiveness against infection.
Induction of systemic immunity by vaccination with IPV and OPV
Both the (e)IPV and the OPV vaccines are capable of inducing high levels of
circulating neutralising antibodies. However, vaccination with IPV has been described
as inducing high percentages of seroconversions with only one or two doses [4,64,76].
Vaccination with eIPV induces more than 90% seroconversion (titer ≥8) against all
three types of poliovirus after only one dose and 100% seroconversion after two doses
. In contrast, vaccination with OPV requires three or more doses to reach similar
levels of neutralising antibodies in the serum.
The responses to trivalent OPV in tropical countries are generally lower than those
observed in the western world [16,56]. Accumulated data from 15 studies that have
examined the response to three doses of trivalent OPV in developing countries reveal
a wide variation in the number of children seroconverting, with rates ranging from
36% to 99% for serotype 1, 71% to 100% for serotype 2, and 40% to 99% for
serotype 3 . Serological surveys carried out 15 years or more after the beginning
of national OPV coverage indicate at least 95% neutralising antibody seroprevalence
against all three serotypes of poliovirus in persons two years of age and older
[32,58,84]. The average number of seropositives and/or seroconversion is lowest for
serotype 3, followed by serotype 1 for both the OPV and IPV vaccine .
Induction of mucosal immunity by vaccination with IPV and OPV
Several studies have investigated mucosal immunity after vaccination with OPV and
IPV [see Table 1]. Most of these studies are limited in scope because they focus
mainly on serotype 1 poliovirus in young children shortly after immunisation [31,49].
In addition, they use different vaccine compositions (such as IPV, eIPV and different
OPV vaccines) and many different vaccination schedules. These factors make a
comparison of the various studies difficult. It should also be noted that some of these
studies were carried out shortly after the start of the vaccination era [15,46,81,100], in
areas where wild type poliovirus still circulated, or where OPV vaccination was
widely used [49,73]. It is impossible to exclude the effect of additional infection with
live poliovirus (vaccine or wild-type) under these circumstances.
The mucosal immune response to OPV closely parallels that of natural infection.
After the administration of OPV, the virus is expected to multiply in the same
alimentary tract sites and related lymphoid tissues and is shed in the stool for several
weeks . Secretory IgA is induced in the nasopharynx and intestine approximately
one to three weeks after immunisation . IgA can still be detected in the
nasopharynx at 60 to 100 days after vaccination . The secretory IgA response
rapidly returns after revaccination, but usually for a shorter duration, although some
studies describe the persistence of secretory antibody activity for as long as five to six
years [71,89]. The poliovirus-specific IgA induced after OPV vaccination is also
found to be protective against poliovirus infection as demonstrated by several
challenge experiments with OPV, and after natural exposure [31,38,72,73] [Table 1].
Protection from re-infection is incomplete and re-infection remains possible in about
one third of recipients .
One early study demonstrated that IPV vaccination was unable to induce a secretory
IgA response in the nasopharynx or in stool samples . Other researchers later
demonstrated some induction of poliovirus specific IgA in the nasopharynx as well as
reduced viral shedding after vaccination with the new eIPV [15,22,28,38,81,89,109].
However, the induction of sIgA described in these studies was at lower levels (9%-
88%) than was the case after vaccination with OPV (26%-100%) [22,109].
Most studies show decreased pharyngeal shedding of poliovirus in IPV recipients
compared to non-immunised children [15,28,38,81]. The effect on reduced pharyngeal
shedding was highest in IPV vaccinated children with neutralisation titers of 8 or
higher [44,57]. Only 38% of these children excreted poliovirus from the pharynx. In
contrast, 75% of the group of children with titers below 8 shed poliovirus from the
pharynx . Further analysis of the prevalence of wild-type poliovirus in household
contacts of patients revealed that poliovirus was less frequently isolated from the
nasopharynx in IPV recipients than in non-vaccinated children. [57,101]. In contrast,
these studies have not consistently demonstrated decreased faecal shedding of
poliovirus. A number of these studies [15,49] demonstrated a decrease in the duration
of excretion and the amount of poliovirus present in the stool in IPV recipients
compared with non-immunised children. Other studies, however, report no difference
Several researchers investigated poliovirus infection in IPV vaccinated household
contacts of patients with paralytic poliomyelitis during the first decade after the start
of IPV vaccination in the USA. All studies showed that there was no significant
difference between IPV-vaccinated and non-vaccinated children within the
households in the duration, amount and proportion of persons excreting wild type
Gelfand et al  studied naturally-occurring poliovirus infection in IPV-vaccinated
families, and reported no significant difference in the pattern of transmission between
IPV-vaccinated and non-immunised family members within infected households.
However, among IPV-vaccinated persons with neutralising antibody titers of 128 or
higher, the duration of faecal shedding was shorter (although the proportion shedding
virus did not differ) than in persons with lower neutralising antibody titers in the
serum . Several studies report a correlation between serum NT titers and a
reduced shedding of poliovirus from the stool and nasopharynx [15,44,51,57]. In most
of these cases, however, the possibility that part of the NT titers may have been
induced by additional infection with OPV or with wild-type poliovirus cannot be
excluded. Unfortunately, the NT test does not discriminate between antibodies that are
induced by inactivated versus live poliovirus.
In summary, studies that investigate mucosal immunity after IPV vaccination show
partly conflicting results with respect to the induction of IgA and differences in the
reduction of viral shedding after an OPV challenge [Table 1]. Nevertheless, it is clear
that all studies comparing responses of IPV- and OPV-immunised children show a far
greater decrease in the excretion of challenge virus among those immunised with
Table 1. Summary of studies investigating mucosal immunity after IPV or OPV
Mucosal IgA induction :
% with poliovirus-specific IgA
Stool Saliva Vaccine/Country
IPVOPV IPV OPV
Keller 19680 nd0nd trivalent/Switzerland
Smith  1986ndnd9* 26trivalent/USA/Sweden
Zhaori  1988ndnd 27*70 type 1/USA
Faden 1990nd nd 41-88* 75-100 trivalent/USA
Viral shedding after OPV challenge :
% shedding challenge virus
Stool Pharynx Challenge Virus/
Ghendon 1961 74 37ndndtype 1/USA
Dick 196163nd0 ndtype 1/UK
Henry 19668332ndndtype 1/UK
Onorato  199163*25 1*4type 1/USA
Kok 19927.1*3.30*0 type 1/Kenya
* = eIPV, nd = not determined
The problem remains that methodological differences between these studies often
result in data that are difficult to compare [15,46,81,100]. It is important to bear in
mind that additional mucosal priming can not be excluded in most of these cases.
Whether or not IPV alone is able to exert an effect on mucosal immunity is still
unclear from the results of these studies, and many different opinions currently
surround this topic.
Prospects for eradication of poliovirus
In 1988, the World Health Assembly decided to strive for the eradication of
poliomyelitis by the year 2000 . The WHO estimates that current levels of
immunisation coverage prevent some 550,000 cases of paralytic poliomyelitis each
year. The eradication of poliovirus will be considered complete when a) no more
cases of poliomyelitis caused by wild-type poliovirus occur, and b) poliovirus is no
longer circulating in humans (vaccinated or non-vaccinated) or in the environment
Figure 2. Wild Poliovirus in 1988
Figure 3. Wild Poliovirus in 1998
Methods of reaching this eradication goal include the implementation of a world-wide
vaccine coverage of 80% or more of all new-born children, along with high quality
clinical and environmental surveillance. Clinical surveillance recommended by the
WHO includes the virological investigation of all patients with acute flaccid paralysis
Known or probable wild
Known or probable wild
(possible poliomyelitis cases) and their contacts . Environmental surveillance
includes the detection of poliovirus in the environment (for example, sewage water).
The last cases of wild-type induced poliomyelitis in the United States were reported in
1979. Since then, apart from rare cases of imported poliomyelitis, all reported cases of
paralytic poliomyelitis in the United States have been vaccine-associated. The last
case of poliomyelitis caused by wild-type poliovirus in the rest of the Americas was
reported in August of 1991 in Peru. Despite improved surveillance, no other cases of
poliomyelitis caused by poliovirus have been detected in this region, and in
September 1994 the international Poliomyelitis Eradication Certification Committee
certified the Americas to be free of wild type poliovirus .
Many other regions are now on their way to becoming certified [106,107]. The main
problem areas for poliovirus circulation today are India and large parts of Africa—
developing countries in which routine immunisation alone may not be sufficient for
the interruption wild-type poliovirus transmission [106,107]. A set of national
immunisation days, at which time two doses of OPV one month apart are given to all
children under 5 years of age regardless of their immunisation status, may be more
effective in these regions. The use of OPV during national immunisation days in these
developing countries is either soon to be or has already been applied. This strategy has
proved to be very successful in China and in the Pacific region. Figures 2 and 3 show
the areas of known or probable wild poliovirus transmission in 1988 and 1998
Outbreaks of poliomyelitis in The Netherlands and Finland
For the eradication program to succeed, transmission of wild-type poliovirus must
cease completely . Important information pertaining to the influence of IPV and
OPV vaccination on the transmission of wild-type poliovirus can be obtained through
analysis of recent poliomyelitis outbreaks. Outbreaks of poliomyelitis in countries
using IPV or OPV have shown that epidemics can occur in areas that have been free
of poliomyelitis for several years [43,48,68,74,85,91]. Despite the good clinical
efficacy of both OPV and IPV and a high level of coverage, neither of these vaccines
has been able to completely break the transmission of poliovirus. This can be
concluded from the substantial proportion (between 21% and 39%) of fully vaccinated
persons (both OPV and IPV) that appear to be involved in the chain of wild-type
poliovirus transmission [43,48,68,74,85].
As previously noted, mucosal immunity is considered to be of particular importance
for protection against (re)infection with poliovirus, thereby interrupting the chain of
transmission of wild-type poliovirus [62,70,71]. IPV vaccination is thought to induce
little or no mucosal immunity against poliovirus. For this reason, outbreaks in
countries that use IPV exclusively in their vaccination programs are of special interest
for the study of the transmission of poliovirus. Recent outbreaks in The Netherlands
and in Finland are described below.
Between 1970 and 1980, the immunisation schedule in The Netherlands consisted of
five doses of IPV during childhood. Coverage with three or more doses was higher
than 95%. A type 1 epidemic with 110 notified cases (80 with paralysis) occurred in
1978 among non-immunised members of a religious group rejecting vaccination .
During this outbreak, 21% of fully immunised school children excreted wild type
poliovirus, while 46% of non-immunised children excreted the virus .
Monovalent serotype 1 OPV vaccine was distributed among the risk groups to control
the outbreak, which eventually spread to non-immunised overseas contacts among
members of the religious group in Ontario (Canada) and the USA . The
Netherlands changed to a schedule of 6 IPV doses following the 1978 outbreak,
attaining a level of coverage that exceeded 97%. The members of the religious group,
however, continued to refuse vaccination. In 1992, after 14 years without endemic
cases, a serotype 3 poliovirus outbreak occurred within this non-vaccinated group,
with a total number of 71 patients . This outbreak was investigated at schools, in
the environment, at virus diagnostic laboratories and in the general population. No
spread to other parts of the country was observed during the outbreak .
Six doses of IPV are used in the Finnish vaccination schedule and the coverage has
been more than 90% for many years. An outbreak due to a serotype 3 poliovirus
occurred in 1984 . Nine cases of poliomyelitis were identified, two of which had
received five doses of IPV in the past. Investigation of healthy contacts and other
healthy persons showed the serotype 3 poliovirus to be widespread, and at least
100,000 persons were estimated to have been infected. A significant factor
contributing to this outbreak was impaired herd immunity to the epidemic strain,
which differed from the serotype 3 vaccine strain. Wild-type isolates had alanine 99
substituted by valine, and arginine 98 replaced by either serine or asparagine in the
immunodominant region on VP1 . In addition, the geometric mean titers of serum
samples were lower against the epidemic strain when compared to the titers against
the strains used to manufacture the inactivated vaccine, which also contributed to
lower vaccine efficacy .
Outline of this thesis
The last outbreak in The Netherlands (in 1992/1993) raised a number of questions,
such as whether the IPV vaccine provided sufficient protection against wild-type
serotype 3 poliovirus, and whether the IPV-vaccinated population in The Netherlands
was contributing to poliovirus circulation . In this light, the induction of mucosal
immunity by IPV vaccination in The Netherlands is of special interest, since mucosal
immunity is considered to be of great importance for the interruption of transmission
and a reduction in spread of poliovirus within the population. The experiments
described in this thesis have been conducted to study the contribution of IPV
vaccination to immunity from and protection against infection with poliovirus. We
have developed new immunological tools for the rapid detection of poliovirus-specific
antibodies and for the investigation of the induction of mucosal immunity after IPV
An enzyme-linked immunosorbent assay-based Poliovirus-Binding Inhibition test to
detect and quantify antibodies to polioviruses has been optimised and evaluated for
use in population studies as an alternative to the virus neutralisation test in tissue
culture (Chapter 2).
Chapter 3 presents research examining differences in response to the antigenic sites 1
and 3 of serotype 3 poliovirus between previously OPV- and IPV-vaccinated persons,
to investigate possible gaps in the immune response to trypsin cleaved serotype 3
poliovirus induced by IPV vaccination.
An IgA ELISA was developed and evaluated in order to answer the question of
whether vaccination with IPV induced IgA. The seroprevalence of poliovirus-specific
IgA in the circulation of poliomyelitis patients and in the general IPV-vaccinated
population of The Netherlands was investigated, as well as the IgA response to
poliovirus in naturally-exposed persons after a single dose of IPV. The results are
presented in Chapter 4.
The kinetics of the IgA response in serum and saliva after IPV booster vaccination in
previously OPV- and IPV- vaccinated subjects was studied to test the hypothesis that
IPV vaccination is able to induce mucosal poliovirus-specific IgA in persons who
have been previously primed with live poliovirus at mucosal sites. ELISA and
ELISPOT-assays were used for the detection of virus-specific IgA responses (Chapter
New assays that enable the detection of poliovirus-specific antibodies or viral RNA
have been developed during the last decade. The 1992/1993 poliovirus type 3
outbreak in The Netherlands has provided an opportunity to examine the potential of
various methods for poliomyelitis diagnosis and their value in the eradication
program. The results of using these new methods and their implications for the future
diagnosis and clinical surveillance of poliomyelitis are described in Chapter 6.
The results described in Chapters 2 to 6 will be discussed in Chapter 7.
1. Alexander JP, Gary HE, Pallansch MA. Duration of poliovirus excretion and its implication for
acute flaccid paralysis surveillance: a review of the literature. J Infect Dis 1997;175:S176-182.
2. Benyesh-Melnick M, Melnick JL, Rawls WE. Studies of the immunogenicity, communicability
and genetic stability of oral poliovaccine administered during the winter. Am J Epid 1967;86:112-
3. Berger R, Ainbender E, Hodes HL, Zepp HD, Hevizy MM. Demonstration of IgA polio
antibody in saliva, duodenal fluid and urine. Nature 1967;214:420-422.
4. Bernier RH. Improved inactivated poliovirus vaccine: an update. Pediatr Infect Dis 1986;5:289-
5. Bijkerk H. Poliomyelitis epidemic in The Netherlands, 1978. Dev Biol Stand 1979;43:195-206.
6. Bodian D. Differentiation of types of poliomyelitis viruses: the grouping of fourteen strains into
three basic immunological types. Am J Hyg 1949;49:234-245.
7. Bodian D. Poliomyelitis: pathogenesis and histopathology. In: Rivers TM, Horsfall FL jr, eds.
Viral and rickettsial infections of man. 3th ed. Philadelphia: JB Lippincott; 1959:479-518.
8. Bodian D, Horstmann DM. Polioviruses In: Horstfall FL Jr, Tamm I, eds. Viral and rickettsial
infections of man. 4th ed. Philadelphia: JB Lippincot 1965:430-473.
9. Cello J, Strannegrad O, Svennerholm B. A study of the cellular immune response to
enteroviruses in humans: identification of cross-reactive T cell epitopes on the structural proteins
of enteroviruses. J Gen Virol 1996;77:2097-2108.
10. Conyn-van Spaendonck MAE, Oostvogel PM, van Loon AM, van Wijngaarden JK,
Kromhout D. Circulation of poliovirus during the poliomyelitis outbreak in The Netherlands in
1992-1993. Am J Epidem 1996;143:929-934.
11. Dalakas MC, Sever JL, Madden DL, Papadopoulos MN, Shekarchi IC, Albrecht P,
Krezlewicz A. Late postpoliomyelitis muscular atrophy: Clinical, virologic, and immunologic
studies. Rev Infect Dis 1984;6:S562-567.
12. Dalakas MC, Elder G, Hallett M, Ravits J, Baker M, Papadopoulos NM, Albrecht P, Sever J.
A long term follow up study of patients with post poliomyelitis neuromuscular symptoms. N Engl
J Med 1986;314:959-963.
13. Dave K. Annual report 1987-1988. Bombay: Enterovirus Research Center 1989.
14. Davis DC, Lipson MJ, Carver DH. The degree of duration of poliomyelitis virus excretion
among vaccinated household contacts of clinical cases of poliomyelitis. Pediatrics 1958;22:33-40.
15. Dick GWA, Dane DS, McAlister J, Briggs M, Nelson R, Fields CMB. Vaccination against
poliomyelitis with live virus vaccines; effect of previous Salk vaccination on virus excretion. Brit
Med J 1961;2:266-269.
16. Domok I, Balayan MS, Fayinka OA, Skrtic N, Soneji AD, Harland PSEG. Factors affecting
the efficacy of live poliovirus vaccines in warm climates. Bull WHO 1974;51:333-347.
17. Dowdle WR, Birmingham ME. The biological principles of poliovirus eradication. J Infect Dis
18. Emini EA, Jameson BA, Lewis AJ, Larsen GR, Wimmer E. Poliovirus neutralizing epitopes:
analysis and localization with neutralizing monoclonal antibodies. J Virol 1982;43:997-1005.
19. Enders JF, Weller TH. Cultivation of the Lansing strain of poliomyelitis virus in cultures of
various human embryonic cells. Science 1949;109:85-87.
20. Esteves K. Safety on oral poliomyelitis vaccine: results of a WHO enquiry. Bull WHO
21. Expended Programme on Immunization. Global Advisory Group (1990). WER 1991;66:9-12.
22. Faden H, Modlin JF, Thoms ML, Marshall McBean A, Ferdon MB, Ogra PL. Comparative
evaluation of immunization with live attenuated enhanced-potency inactivated trivalent poliovirus
vaccines in childhood: systemic and local immune responses. J Infect Dis 1990;162:1291-1297.
23. Fox J, Gelfand H, Bhat P, LeBlanc D, Conwell D. Immunizing infections with poliomyelitis
viruses and seroimmune patterns in southern Louisiana. Ann New York Acad SC 1955;61:968-
24. Fox JP. Modes of action of poliovirus vaccines and relation to resulting immunity. Rev Infect Dis
25. Freistadt MS, Eberle KE. Correlation between poliovirus type 1 Mahoney replication in blood
cells and neurovirulence. J Virol 1991:70;6486-6260.
26. Fricks CE, Icenogle JP, Hogle JM. Trypsin sensitivity of the Sabin strain of type 1 poliovirus:
cleavage sites in virions and related particles. J Virol 1985;54:856-859.
27. Furesz J. Poliomyelitis outbreak in The Netherlands and Canada. Can Med Assoc J 1979;21:905-
28. Gelezen WP, Lamb J, Belden EA, Chin TDY. Quantitative relationship of pre-existing
homotypic antibodies to the excretion of attenuated poliovirus type 1. Am J Epidem 1966;83:224-
29. Gelfand HM, LeBlanc DR, Fox J, Conwell D. Studies on the development of natural immunity
to poliomyelitis in Louisiana II. Description and analysis of episodes of infection observed in study
group households. Am J Hyg 1959;65:367-385.
30. Gezondheidsraad rapport. Immunity to polioviruses. No. 1992/22, The Hague, December 16,
31. Ghendon YZ, Sanakoyeva II. Comparison of the resistance of the intestinal tract to poliomyelitis
virus (Sabin’s strains) in persons after naturally and experimentally acquired immunity. Acta Virol
32. Goh KT, Yamazaki S. Immune status of the population to poliovirus infection in Singapore. Bull
33. Graham S, Wang ECY, Jenkins O, Borysiewicz LK. Analysis of the human T cell responses to
picornaviruses: identification of T cell epitopes close to B cell epitopes in poliovirus. J Virol
34. Hadfield MG, Seilin M, Houff SA, Adair CF, Markowitz SM, Strauss EJ. Echovirus
meningomyeloencephalitis with administration of intrathecal immunoglobulin. J Neuropathol Exp
35. Hafler DA, Fox DA, Benjamin D, Blue AL, Weiner HL. Secondary immune amplification
following live poliovirus immunization in humans. Clin Immunol Immunopathol 1987;44:321-
36. Hammon WM, Ludwig EH. Possible protective effect of previous type 2 infection against
paralytic poliomyelitis due to type 1 virus. Am J Hyg 1957;66:274-280.
37. Hellen CU, Wimmer E. Maturation of poliovirus capsid proteins. Virology 1992;187:391-397.
38. Henry JL, Jaikaran ES, Davies JR, Tomlinson AJH, Mason PJ, Barnes JM, Beale AJ. A
study of poliovaccination in infancy: excretion following challenge with live virus by children
given killed or living poliovaccine. J Hyg 1966;64:105-120.
39. Hickey WF, Hsu BL, H Kimura. T-lymphocyte entry into the central nervous system. J Neurosci
40. Hogle JM, Chow M, Filman DJ. The three dimensional structure of poliovirus at 2.9 A
resolution. Science 1985;229:1358-1365.
41. Horstmann DM, Ward R, Melnick J. The isolation of poliomyelitis virus from human non-
neural tissues, III- Persistence of virus in stool after acute infection. J Clin Invest 1946;25:278-283.
42. Horstmann DM. The Sabin live poliovirus vaccination trials in the USSR. Yale J Biol Med
43. Hovi T, Huovilainen A, Kuronen T, Kinnunen E, Kuronen T, Lapinleimu K, Poyry T,
Roivainen M, Salama N, Stenvik M. Outbreak of paralytic poliomyelitis in Finland: widespread
circulation of antigenically altered poliovirus type 3 in a vaccinated population. Lancet
44. Howe HA. The quantitation of poliomyelitis virus in the human alimentary tract with reference to
coexisting levels of homologous serum neutralizing antibody. Am J Hyg 1962;75:1-17.
45. Icenogle JP, Minor PD, Ferguson M, Hogle JM. Modulation of humoral responses to a 12-
amino acid site on the poliovirus. J Virol 1986;60:297-301.
46. Keller R, Dwyer JE. Neutralization of poliovirus by IgA antibodies. J Immunol 1968;101:192-
47. Kew OM, Mulders MN, Lipskaya GY, da Silva EE, Pallansch MA. Molecular epidemiology of
polioviruses. Sem in Virol 1995;6:401-414.
48. Kim-Farley RJ, Rutherford G, Lichfield P, Hsu ST, Orenstein WA, Sconberger LB, Bart KJ,
Lui KJ, Lin CC. Outbreak of paralytic poliomyelitis, Taiwan. Lancet 1984;2:1322-1324.
49. Kok PW, Leeuwenberg J, Tukei P, van Wezel AL, Kapsenberg JG, van Steenis G, Galazka
A, Robertson SE, Robinson D, Serological and virological assessment of oral and inactivated
poliovirus vaccines in a rural population in Kenya. Bull WHO 1992;70:93-103.
50. Leon-Morzon ME, Dalakas MC. Detection of poliovirus antibodies and poliovirus genome in
patients with the post-polio syndrome. Ann NY Acad Sci 1995;753:208-218.
51. Lepow ML, Warren RJ, Gray N. Effect of Sabin type 1 poliomyelitis vaccine administered by
mouth to newborn infants. New Engl J Med 1961;264:1071-1078.
52. Livni R, Lasch EE, Englander T, El Massri M, Danon YL. Cell mediated immunity to
polioantigens in the early stage of acute poliomyelitis. Allergol Immunopathol (Madr)
53. Lonnberg F. Late onset polio sequelae in Denmark. Presentation and results of a nation-wide
survey of 3.607 polio survivors. Scand J Rehab Med Suppl 1993;28:7-15.
54. Magrath DI, Evans DMA, Ferguson M, Schild GC, Minor PD, Horaud F, Crainic R, Stevik
H, Hovi T. Anitigenic and molecular properties of type 3 poliovirus responsible for an outbreak of
poliomyelitis in a vaccinated population. J Gen Virol 1986;67:899-905.
55. Mahon BP, Katrak K, Mills KH. Antigenic sequences of poliovirus recognized by T cells:
serotype-specific epitopes on VP1 and VP3 and cross-reactive epitopes on VP4 defined by using
CD4+ T cell clones. J Virol 1992;66:7012-7020.
56. Maldonato YA, Pena-cruz V, de la Luz Sanchez M, Logan L, Blandon S, Cantwell MF,
Matsui SM, Millan-Velasco F, Valdespino JL, Sepulveda J. Host and viral factors affecting the
decreased immunogenicity of Sabin type 3 vaccine after administration of trivalent oral polio
vaccine to rural Mayan Children. J Infect Dis 1997;175:545-553.
57. Marine W, Chin TDY, Gravelle CR. Limitation of fecal and pharyngeal poliovirus excretion in
Salk-vaccinated children: a family study during a type 1 poliomyelitis epidemic. Am J Hyg
58. Mayor TR. Duration of vaccine-induced poliomyelitis immunity. J Fam Pract 1984;19:335-338.
59. McKinney RE, Katz SL, Wilfert CM. Chronic enteroviral meningoencephalitis in
agammaglobulinemic patients Rev Infect Dis 1987;9:334-356.
60. Melnick JL, Ledinko N. Development of neutralizing antibodies against the three types of
poliomyelitis during an epidemic period. Am J Hyg 1953;58:207-222.
61. Melnick JL. Advantages and disadvantages of killed and live poliomyelitis vaccines. Bull WHO
62. Melnick JL. Enteroviruses: Polioviruses, Coxsackieviruses, Echoviruses, and Newer
Enteroviruses. In: Fields Virology, Third Edition 1996 Lippincott-Raven Publishers, Philadelphia,
63. Minor PD, Ferguson M, Evans DMA, Almond JW, Icenogle JP. Antigenic structure of
polioviruses of serotypes 1, 2, and 3. J Gen Virol. 1986;67:1283-1291.
64. Modlin JF, Onorato IM, Marshall McBean AM, Albrecht P, Thoms ML, Nerhood L, Bernier
R. The humoral response to type 1 oral poliovirus vaccine in children previously immunized with
enhanced potency inactivated poliovirus vaccine or live oral poliovirus vaccine. Am J Dis Child
65. Muir P, Nicholson F, Sharief MK, Thompson EJ, Cairns NJ, Lantos P, Spencer GT,
Kaminski HJ, Banatvala JE. Evidence for persistent enterovirus infection of the central nervous
system in patients with previous paralytic poliomyelitis. Ann NY Acad Sci 1995;753:219-232.
66. Ogra PL, Karzon DT, Righthand F, MacGillyvrag M. Immunoglobulin response in serum and
secretions after immunization with live inactivated poliovaccine and natural infection. N Engl J
67. Ogra PL, Karzon DT. Poliovirus antibody response in serum and nasal secretions following
intranasal inoculation with inactivated poliovaccine. J Immunol 1969;102:15-23.
68. Ogra PL, Karzon DT. Distribution of poliovirus antibody in serum, nasopharynx, and alimentary
tract following segmental immunization of lower alimentary tract with polio vaccine. J Immunol
69. Ogra P. Role of nasopharyngeal tonsils and adenoids in the mechanism of mucosal immunity to
poliovirus. American Pediatric Society, 80th Annual meeting, Program and abstracts. 1970.
70. Ogra PL, Karzon DT. Formation and function of poliovirus antibody in different tissues. Prog
Med Virol 1971;13:156-193.
71. Ogra PL. Mucosal immune response to poliovirus vaccines in childhood. Rev Infect Dis
72. Ogra PL, Faden HS, Abraham R, Duffy LC, Sun M, Minor P. Effect of prior immunity on
shedding of virulent revertant virus in feces after oral immunization with live attenuated poliovirus
vaccines. J Infect Dis 1991;164;191-194.
73. Onorato IM, Modlin J, McBean AM, Thoms ML, Losonsky GA, Bernier RH. Mucosal
immunity induced by enhanced-potency inactivated and oral polio vaccines. J Infect Dis
74. Oostvogel PM, Wijngaarden JK, van der Avoort HGAM, Mulders MN, Conyn-van
Spaendonck MAE, Rumke HC, van Steenis G, van Loon AM. Poliomyelitis outbreaks in an
unvaccinated community in The Netherlands, 1992-93. Lancet 1994;344:665-670.
75. Patel V, Ferguson M, Minor PD. Antigenic sites on type 2 poliovirus. Virology 1993;192:361-
76. Patriarca PA, Wright PF, John TJ. Factors affecting the immunogenicity of oral polio vaccine
in developing countries: a review. Rev Inf Dis 1991;13:926-939.
77. Paul JR, Riordan JT, Melnick JL. Antibodies to three different antigenic types of poliomyelitis
virus in sera from North Alaskan Eskimos. Am J Hyg 1951;54:275-285.
78. Riker JB, Brandt CD, Chandra R, Arrobio JD, Nukuno JH. Vaccine-associated poliomyelitis
in a child with thymic abnormality. Pediatrics 1971;48:923-929.
79. Robertson S. Poliomyelitis. In: Immunological basis for immunization. 1993 WHO
80. Roivainen M, Hovi T. Intestinal trypsin can significantly modify antigenic properties of
polioviruses: implications for the use of inactivated poliovirus vaccine. J Virol 1987;61:3749-
81. Sabin AB. Present position of immunization against poliomyelitis with live virus vaccines. Brit
Med J 1959;1:663-680.
82. Salk J. Requirements for persistent immunity to poliomyelitis. Trans Assoc Am Physicians
83. Salk J, Drucker J. Noninfectious poliovirus vaccine. In: Vaccines. SA Plotkin, Mortimer EA Jr,
eds. Saunders Company, Philadelphia, 1988;158-181.
84. Santoro R. Serum antibodies to poliomyelitis in Italy. Bull. WHO 1984;62:591-595.
85. Schaap GJP, Bijkerk G, Coutinho RA, Kapsenberg JG, van Wezel AL. The spread of wild
poliovirus in the well-vaccinated Netherlands in connection with the 1978 epidemic. Prog Med
86. Selvakumar R, John TJ. Intestinal immunity induced by inactivated poliovirus vaccine 1986.
87. Sicinski P, Rowinski J, Warchol JB, Jarzabek Z, Gut W, Szczygiel B, Bielecki K, Kock G.
Poliovirus type 1 enters the human host through intestinal M cells. Gastroenterol. 1990;98:56-58.
88. Simons JU, Kutubuddin M, Chow M. Characterization of poliovirus-specific T lymphocytes in
the peripheral blood of Sabin-vaccinated humans. J Virol 1993;67:1262-1268.
89. Smith JWG, Lee JA, Fletcher WB, Morris CA, Parker DA, Yetts R, Magrath DI, Perkins
FT. The response to oral poliovaccine in persons aged 16-18 years. J Hyg 1976;76:235-247.
90. Strebel PM, Sutter RW, Cochi SL, Biellik RJ, Brink EW, Kew OM, Pallansch MA,
Orenstein WA, Hinman AR. Epidemiology of poliomyelitis in the United States one decade after
the last reported case of indigenous wild virus-associated disease. Clin Infect Dis 1992;14:568-
91. Sutter RW, Patriarca PA, Brogan S, Malankar PG, Pallansch PA, Kew OM, Bass AG, Cochi
SL, Alexander JP, Hall DB, Suleiman AJM, Al-Ghassani AAK, El-Bualy MS. Outbreak of
paralytic poliomyelitis in Oman: evidence for widespread transmission among fully vaccinated
children. Lancet 1991;338:715-720.
92. Turner TB, Hollander DH, Buckley S. Age incidence and seasonal development of neutralizing
antibodies to Lansing poliomyelitis virus. Am J Hygiene 1950;52:323-347.
93. Van der Avoort HGAM, Hull BP, Hovi T, Pallansch MA, Kew OM, Crainic R, Wood DJ,
Mulders MN, van Loon AM. Comparative study of five methods for intratypic differentiation of
polioviruses. J Clin Microbiol 1995;33:2562-2566.
94. Van der Marel P, Hazendonk TG, Henneke MA, van Wezel AL. Induction of neutralizing
antibodies by poliovirus capsid polypeptides VP1, VP2 and VP3. Vaccine 1983;1:17-22.
95. Van Wezel AL. Growth of cell-strains and primary cells on micro carriers in homogenous culture.
Nature 1967; :216-264.
96. Van Wezel AL, Hazendonk AG. Intratypic serodifferentiation of poliomyelitis virus strains by
strain-specific antisera. Intervirology 1979;11:2-8.
97. Van Wezel AL, van Steenis G, van der Marel P, Osterhaus AD. Inactivated poliovirus vaccine:
current production method and new developments. Rev Infect Dis 1984;6:S335-S340.
98. Verlinde J, Wilterdink J. A small-scale trial on vaccination and revaccination with live
attenuated poliovirus in The Netherlands. In: First International Conference on live attenuated
poliovirus vaccine, Washington DC. 1959:355-366.
99. Walker WA, Isselbacher KJ. Intestinal antibodies. New Eng J Med 1977;297:767-773.
100. Wehrle PF, Reichert R, Carbonaro O. Influence of prior active immunization on the presence of
poliomyelitis virus in the pharynx and stools of family contacts of patients with paralytic
poliomeyelitis. Pediatrics 1958;21:353-361.
101. Wehrle PF, Carbonaro O, Day PA. Transmission of polioviruses. Prevalence of polioviruses in
pharyngeal secretions of infected household contacts of patients with clinical disease. Pediatrics
102. White DO, Fenner FJ. Medical Virology 4th edition Academic Press 1994.
103. World Health Organization. Global eradication of poliomyelitis by the year 2000. In: Forthy-
first World Health Assembly, Geneva, 2-13 May 1988: resolutions and decisions annexes Geneva:
World Health Organization 1988. 26. (resolution WHA41.28).
104. World Health Organization. Manual for the virological investigation of poliomyelitis. Geneva:
World Health Organization, 1990. (WHO document EPI/POLIO/90.1).
105. World Health Organization. Acute onset flaccid paralysis. Geneva: World Health Organization,
1993 (WHO document WHO/MNH/EPI/93.3).
106. World Health Organization. Polio: the beginning of the end. Geneva: World Health organization,
107. World Health Organization. Expanded program on immunization poliomyelitis eradication: the
WHO Global Laboratory Network. Weekly Epidemiological Record, 1997:245.
108. Wyatt HV. Poliomyelitis in hypogammaglobulinemics. J Infect Dis 1973;128:802-806.
109. Zhaori G, Sun M, Ogra PL. Characteristics of the immune response to poliovirus virion
polypeptides after immunization with live or inactivated polio vaccines. J Infect Dis 1988;158:160-
Evaluation of a Poliovirus-Binding
Inhibition assay as an alternative for the
virus neutralisation test
Tineke Herremans, Johan Reimerink, Albert Ras,
Harrie van der Avoort, Tjeerd Kimman, Anton Van Loon,
Marina Conyn-Van Spaendonck and Marion Koopmans
Clinical and Diagnostic Laboratory Immunology 1997; 4: 659-664
An enzyme-linked immunosorbent assay (ELISA)-based Poliovirus-Binding
Inhibition test (PoBI) to detect and quantify antibodies to polioviruses was optimised
and evaluated for use in population studies as an alternative to the virus neutralisation
test (NT) in tissue culture. The sensitivity of the inhibition ELISA as compared with
the NT in an inactivated poliovirus vaccine (IPV)-vaccinated population was 98.6%,
97.4% and 92.1% for serotypes 1, 2 and 3 respectively. The specificity of the PoBI
test, as determined with sera from non-vaccinated persons, was also high for all three
serotypes (99.0%, 95.8% and 100%). Antibodies to other enteroviruses did not cross-
react in the serotype 1 or the serotype 3 PoBI, and only low levels of cross-reactivity
were found for serotype 2. We found high correlations between the PoBI and NT
titers for serotypes 1 and 2 in IPV-vaccinated blood donors (0.97 and 0.95), in oral
poliovirus vaccine (OPV)-vaccinated blood donors (0.91 and 0.95) and in naturally
immune persons (0.90 and 0.87). The correlation coefficient for serotype 3, however,
was significantly lower in OPV-vaccinated blood donors (0.73) and in naturally
immune persons (0.76) than in IPV-vaccinated persons (0.94; p<0.01). These results
indicate that the antibody response to serotype 3 poliovirus in IPV-recipients is
different from that in OPV-recipients and naturally infected persons. We conclude
that the PoBI test is a suitable alternative for the NT to estimate the seroprevalence of
neutralising antibodies to poliovirus, especially in large scale population studies.
Poliovirus neutralising antibodies in serum are sufficient for protection against
paralytic disease [2,4,12,16]. These neutralising antibodies, thought to be
predominantly of the IgG isotype, prevent poliovirus from reaching the central
nervous system [5,11]. The neutralisation test (NT) is used as the standard test for the
measurement of immunity to the three serotypes of poliovirus after vaccination or
natural exposure. Advantages of the NT are the high sensitivity, specificity and
acceptance of this assay. The NT has been chosen by the WHO as the reference
method for determining immunity against poliovirus . However, the need to use
cell culture and the long duration of the test (up to six days) make the NT expensive
and less suitable for large scale screening of populations for protection against
poliomyelitis. In addition, in view of the probable eradication of poliovirus in the near
future, the use of live (wild-type) polioviruses in laboratory research and diagnostic
assays (such as NT) will be discouraged or prohibited, and alternative methods for the
immuno-surveillance of populations will be needed.
Recently Edevag et al.  described an inhibition-ELISA for the detection of
neutralising antibodies using inactivated polioviruses as antigen. With this inhibition-
ELISA (Poliovirus-Binding Inhibition assay; PoBI), a high correlation with the
standard neutralisation test was found with a small set of sera in a pilot study. The
specificity of the assay was not fully evaluated. Because the PoBI test is a promising
alternative for the NT, we have optimised and evaluated the assay for use as a
replacement of the NT in large scale population studies. In order to do so, we tested
sets of sera from 1] persons vaccinated with inactivated poliovirus vaccine (IPV) or
live attenuated oral poliovirus vaccine (OPV); 2] persons with documented infection
with poliovirus; 3] a known seronegative population; 4] rabbits, each immunised with
one of 41 different enteroviruses and 5] a cross-sectional epidemiological survey
aimed at the evaluation of the national vaccination program in the Dutch population.
Evaluation of a Poliovirus-Binding Inhibition-assay
Materials and Methods
The PoBI test was performed according to the method described by Edevag et al. 
with modifications, and was conducted in two steps. For the first step (pre-incubation
of serum and virus), microtiter cell culture plates (Greiner, Alphen aan den Rijn, The
Netherlands) were blocked with 150 µl of 0.5% bovine serum albumin (BSA) in
phosphate-buffered saline (PBS) for 1 h at 37°C. Two-fold dilutions of serum samples
were made directly into the wells (75 µl/well, 1:2 to 1:4,096) in dilution buffer (PBS
with 0.5% Tween 20, 0.5% BSA, 0.5 M NaCl). Monovalent, inactivated vaccine virus
produced at the National Institute of Public Health and the Environment (RIVM) was
used as antigen. The formalin-inactivated poliovirus was added to each well at
concentrations of 20, 4 and 16 D-antigen units/ml for serotypes 1, 2 and 3,
respectively, in a volume of 75 µl per well. The concentration of D-antigen was
quantified with a direct ELISA . Poliovirus serotype 1 was Mahoney, serotype 2
was MEF, and serotype 3 was a Saukett strain. Serum-virus mixtures were incubated
for 2 h at 37°C.
For the second step, Maxisorp ELISA plates (Nunc, Roskilde, Denmark) were coated
with the IgG fraction of bovine antipoliovirus hyperimmune serum in dilutions of
1:500 for serotypes 1 and 2 and 1:250 for serotype 3 in 0.04 M carbonate-bicarbonate
buffer pH 9.6 (overnight at 4°C). Plates were blocked with 100 µl of 0.5% BSA in
PBS for 1 h at 37°C. After blocking, 100 µl of the pre-incubated serum-virus mixture
was transferred to the ELISA plate and incubated for 2 h at 37°C. For detection of
bound antigen, serotype-specific monoclonal antibodies—type 1 (14D2E9, 1:3,000),
type 2 (6-15C6, 1:10,000), and type 3 (2-13D9, 1:10,000)—in dilution buffer were
added for 1 h at 37°C . The monoclonals used for serotype 1 were directed against
antigenic site 2a, while the monoclonals used for serotype 2 and 3 were directed
against antigenic site 1 of the corresponding poliovirus capsid. Goat anti-mouse IgG
alkaline phosphatase conjugate (Sigma, Zwijndrecht, The Netherlands) was
subsequently added in a dilution of 1:500 and incubated for 1 h at 37°C. The substrate
p-nitrophenylphosphate (Sigma) at a concentration of 1 mg/ml in 0.1 M glycine buffer
(pH 10.4) was incubated at room temperature for 30 min. Plates were read at 405 nm
by use of a Microwell System 510 spectrophotometer (Organon Teknica, Eindhoven,
For the evaluation of the PoBI test, serum samples were considered positive if a
reduction in extinction of ≥50% was reached. The reciprocal of the first serum
dilution that was positive in the inhibition test was taken as the titer of the test sample.
A standard in-house reference serum with known titers was included in each assay.
Optimal dilutions of the coat, detector monoclonal antibodies, and conjugate were
established by checkerboard titrations.
Poliovirus neutralising antibody titers of sera were determined in the standard NT as
recommended by the World Health Organisation  by using the Mahoney (serotype
1), MEF (serotype 2), and Saukettt (serotype 3) virus strains as challenge viruses. In
brief, serial two-fold dilutions of sera to be tested and 100 50% cell culture infective
doses of virus were incubated in 96-microwell plates at 37°C for 3 h. After the
incubation, 1.75 x 104 Hep-2C cells were added per well. The plates were read after
six days of incubation at 37°C. The titers are expressed as the reciprocal of the highest
dilution showing complete neutralisation of the cytopathic effect of 100 50% cell
culture infective doses. Samples were considered positive if NT titers were ≥8 (log2
For evaluation of the PoBI test, the following groups of sera were examined: 1] Sera
were obtained from IPV-vaccinated blood donors in The Netherlands (n=26) and from
OPV-vaccinated blood donors in Belgium (n=42). 2] Sera were obtained from a group
of non-vaccinated elderly persons (age 52 to 85 years) in The Netherlands that had
been given a single dose of IPV (n=47). This group was presumably naturally exposed
at a young age, when poliovirus was endemic in The Netherlands . Sera were
collected at the time of IPV vaccination, and at one and four weeks thereafter. 3]
Negative-control sera were obtained from non-vaccinated children (n=96) from a
population that refuses vaccination for religious reasons. None of these serum samples
had detectable neutralising antibodies (titer, <2) against poliovirus. In addition, all
sera were also negative for antibodies to other components of the vaccine cocktail
(diphtheria and tetanus toxoid) that is used in the routine immunisation of children in
The Netherlands. 4] Serum samples were obtained from rabbits hyperimmunised with
either poliovirus serotype 1 (Brunhilde; NT titer: type 1, 20,480; type 2, 10; type 3,
<10), serotype 2 (MEF-1; NT titer: type 1, 10; type 2, 81,920; type 3, <10), or
serotype 3 (Saukett; NT titer: type 1, <10; type 2, <10; type 3, 40,960) or with other
enteroviruses (coxsackievirus B serotypes 1 to 6, echovirus serotypes 1 to 9, 11 to 27,
and 29 to 33, and enterovirus serotypes 68 to 71). 5] A total of 785 serum samples
were obtained from a cross-sectional epidemiological survey in the province of
Utrecht, The Netherlands, aimed at the evaluation of the national vaccine program.
Regression analysis was used to determine the coefficients of correlation between
results obtained by the PoBI test and neutralisation titers. P values of <0.01 were
Optimisation and properties of the PoBI test
We compared different incubation times in order to be able to reduce the duration of
the PoBI test. Pre-incubation of the serum-virus mixture could be reduced to 2 h at
37°C (instead of an overnight incubation at 37°C) without loss of sensitivity or
changes in PoBI titers. Incubation times of the detecting monoclonal antibody and
conjugate were reduced to 1 h at 37°C without influencing the outcome of the assay.
The PoBI configuration (serum-virus pre-incubation) was compared to the direct
binding of the antigen to the IgG in the ELISA plates. The PoBI titers were found to
be 4- to 16-fold higher when the pre-incubation step was performed, and the
correlation with NT was higher (result not shown).
Evaluation of a Poliovirus-Binding Inhibition-assay
Specificity of the PoBI test
When the previously described protocol  was used, 5.2%, 9.4% and 6.3% of the
negative control samples tested positive in the PoBI test for serotypes 1, 2 and 3,
respectively. The specificity of the assay was increased with the addition of 0.5 M
NaCl in the ELISA dilution buffer to reduce aspecific binding, and this modification
was used throughout the rest of the study. Using this assay format, the specificity was
improved to 99.0%, 95.8% and 100% for serotypes 1, 2 and 3, respectively. Different
blocking agents (BSA, fetal calf serum and milk powder) had no influence on false-
positive results with sera from known seronegative donors. False-positive signals
were strongly reduced or disappeared after (NH4)2SO4 precipitation of the sample,
indicating that the signal was not caused by cross-reacting IgG in the serum sample
(data not shown).
Sensitivity of the PoBI test
The sensitivity of the PoBI test was determined with sera from IPV- and OPV-
vaccinated blood donors and from naturally immune older persons (before and after
IPV vaccination). Overall, the sensitivity was 99.5%, 100% and 96.2% for the
serotype 1, 2 and 3 assays, respectively. The sensitivity of the PoBI test is dependent
on the NT titer and was lowest for sera with low levels (titer, ≤16) of neutralising
antibodies (95%, 100% and 75% for serotypes 1, 2 and 3 respectively). The sensitivity
increased to 100% at neutralisation titers of 16 and 64 for serotypes 1 and 3
respectively (data not shown).
Cross-reactivity of the PoBI test.
Hyperimmune rabbit sera were used to check for possible cross-reactivity between
antibodies to poliovirus and other enteroviruses in the PoBI test (Table 1). Sera to
other enteroviruses (coxsackievirus B serotype 1 to 6; echovirus serotypes 1 to 9, 11
to 27, and 29 to 33; and enterovirus serotypes 68 to 71) did not react in the PoBI test
for serotypes 1 and 3. In the serotype 2 test, low levels of cross-reactivity were
observed with 13 of the 41 antisera that were tested. In addition, moderate levels of
cross-reactivity were detectable between the polioviruses in the PoBI test and the NT
Correlation between the PoBI titer and the NT
We found a high correlation between the NT and the PoBI test similar to that
described by Edevag et al.  with serum samples from IPV-vaccinated persons
(0.97, 0.95 and 0.94 for serotypes 1, 2 and 3, respectively). The coefficient of
correlation between the NT and the PoBI test for serotypes 1 and 2 was high in OPV-
vaccinated blood donors and naturally immune persons (Table 2). However, the
correlation coefficient was significantly lower for serotype 3 in OPV-vaccinated
blood donors (0.73) and naturally immune persons (0.76) than in the IPV-vaccinated
blood donors (P <0.01) (Table 2). The correlation between PoBI and NT titers was
similar for sera collected at different points in time after the vaccination of naturally
immune persons vaccinated with IPV (Table 2).
Table 1. Cross-reactivity of sera from rabbits immunised with different enteroviruses
in the Poliovirus-Binding Inhibition assay and neutralisation test.
Rabbit reference NT titer to:
PoBI titer to:
serum to: Serotype 1Serotype 2 Serotype 3 Serotype 1Serotype 2 Serotype 3
Serotype 120,48010 <10 4,096 164
Serotype 2 1081,920 <1064 16,3844
Serotype 3 <10 <10 40,96088 2,048
Serotype 1-6<2<2<2 <2 <2<2
24-27, 29-33<2 <2<2 <2<2 <2
11-14, 21-23<2 <2<2<22 <2
Serotype 20<2<2 <2 <28 <2
Serotype 68, 71<2 <2<2<24 <2
Serotype 69, 70 <2<2 <2<2 <2 <2
Table 2. Correlation of Poliovirus-Binding Inhibition assay and neutralisation test in
groups of persons with vaccine-induced or natural immunity.
Serotype 1Serotype 2 Serotype 3
Blood donors :
General population :
Naturally immune :
Weeks after IPV vaccination :
NOTE * significantly different from IPV vaccinated blood donors at the 0.01 level. IPV= inactivated
poliovirus vaccine, OPV= oral poliovirus vaccine.
Evaluation of a Poliovirus-Binding Inhibition-assay
Evaluation of immunity in the general (IPV-vaccinated) population
A total of 785, 763 and 771 sera were examined in PoBI and NT for serotypes 1, 2
and 3 respectively. In the general population, the correlation between NT and PoBI
was 0.89, 0.89 and 0.84 for serotypes 1, 2 and 3, which was significantly lower than
the correlation between NT and PoBI using sera from IPV vaccinees for serotype 1
(Table 2). No differences in the correlation between PoBI and NT were found
between the different age groups (data not shown). From the regression line, the PoBI
titer that corresponds with a titer in the NT of 8 was calculated to be 4 [Figure 1].
Therefore, sera with a PoBI titer < 4 were considered negative. The sensitivity of the
PoBI test was high for all three serotypes: 98.6%, 97.4% and 92.1% for serotypes 1, 2
and 3 respectively. The positive predictive value of the PoBI test in the general
population was 0.98, 0.97 and 0.97 for serotypes 1, 2 and 3 respectively. Specificity in
this group was 80.3% for serotype 1, 82.0% for serotype 2 and 79.8% for serotype 3.
The negative predictive value was 0.83 and 0.82 for type 1 and 2 respectively, and
only 0.61 for type 3.
Both assays provided a normal distribution of titers in addition to a group of
seronegatives (PoBI and NT titers of <4 and <8 respectively) for all three serotypes
[Figure 2]. PoBI titers were generally two-fold lower than the standard NT titers.
The total number of seronegatives in the general population was estimated to be 7.5%,
12.4% and 17.4% with the PoBI and 7.8%, 12.5% and 13.3% with the NT test. These
percentages were not significantly different (p-level <0.01). The PoBI was able to
provide a good estimate of the total number of seronegatives for all three serotypes
within the different age groups in the general population [Figure 3], and showed
patterns of seroprevalence in the different age groups similar to the NT for all three
Previously described ELISAs used to measure protective antibodies to poliovirus
could not compete with the very sensitive NT [6,7,10,17]. With a direct ELISA
format for serotype 1, Simhon et al.  reached positive predictive values between
0.82 and 0.91, but found high numbers of false-negative results and low negative
predictive values (between 0.29 and 0.55). Hagenaars et al.  described an inhibition
ELISA in which serum antibodies and labelled bovine anti-poliovirus serotype 1
competed for binding places on the bound antigen. The assay correlated well with NT
but the standard NT assay was more sensitive then the inhibition ELISA.
The present format of the PoBI test, in which inhibition of the signal depends on both
reduction of virus-antigen binding to the capture antibody and reduction of binding of
the indicator monoclonal antibody, was found to be a suitable replacement for the NT
in large scale population studies. Although PoBI titers were generally (two-fold)
lower than NT titers, neutralisation positive samples could easily be identified.
Sensitivity and positive predictive values were high in both IPV- and OPV-vaccinated
persons as well as in naturally exposed people. The specificity of the PoBI test was
determined with selected sera from non-vaccinated subjects and was high (95-100%).
In the general population, persons with low NT titers (2 and 4) are considered to be at
risk for poliovirus-induced disease. Therefore, for the purpose of this evaluation, sera
with NT titers < 8 as well as sera with a corresponding cut-off titer in the PoBI of <4
Figure 1. Correlation between the NT and PoBI titers (expressed as 2log values) for the three
poliovirus serotypes in a cross-section of different age groups in the general population.
Evaluation of a Poliovirus-Binding Inhibition-assay
<1123456789 10 11 12 >12
<1123456789 1011 12 >12
<1123456789 10 11 12 >12
Figure 2. Frequency distribution of NT and PoBI titers (expressed as 2log values) in a cross-
section of different age groups in the general population. Results are presented as
percentage of persons positive per titer. The highest serum dilution tested in the PoBI was
1:256 (2log 8) and 1:4096 (2log 12) in the NT.
1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80
1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80
age groups in years
1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80
Figure 3. Seroprevalence of poliovirus antibodies in a cross-section of different age groups in
the general population as determined by PoBI and NT assay. Results are expressed as
percentage seropositives per age group. NT titers were considered positive if the titer was ≥
8. PoBI titers were considered positive if the titer was ≥ 4.
Evaluation of a Poliovirus-Binding Inhibition-assay
were considered seronegative. The detection of these low NT titers in the PoBI is
responsible for the lower specificity of the PoBI calculated from the comparative
serology in the population (79.8-82.0%) as well as for the low negative predictive
value for the serotype 3 assay.
The question remains whether NT titers of 2 and 4 are truly negative or whether they
should be considered low but specific antibody levels. It is most likely that this group
consists of both true negatives and low positive samples. Therefore, PoBI-negative
sera should be re-tested by NT. The PoBI results with rabbit hyperimmune sera
showed that cross-reactivity with other enteroviruses did not occur for serotype 1 and
3, and only at low levels for type 2. Since titers to enterovirus antigens as high as the
levels found in the hyperimmune rabbit serum samples are not likely to be detected in
the general population or in patients, this low level cross-reactivity is not a problem
for serosurveys in which the PoBI is used.
The purpose of this seroprevalence study was to estimate the total number of
seronegatives (as determined by NT and by PoBI) within the Dutch population. A
high percentage of seronegatives in the population indicates a potential risk for
poliomyelitis outbreaks and requires active re-vaccination of the (age) groups at risk.
The PoBI test proved to be an excellent indicator of seroprevalence in all age groups
from an IPV-vaccinated population. Both assays yielded similar estimates of the total
number of seronegatives within the general population for all three serotypes.
We conclude that the PoBI is a suitable test for seroprevalence studies of poliovirus. It
is a less labour intensive assay than the NT, it is easier to perform, it can be further
automated and it is not dependent on a visual screening of the cytopathic effect as is
the case in the standard NT. This assay could be used for population screening when
combined with confirmatory testing of PoBI negatives by the NT. Given the current
prevalence of poliovirus antibodies, this approach could reduce the total number of
serum samples examined in the NT by 87.5%.
We reached a high correlation between PoBI and NT similar to that described by
Edevag et al.  for IPV-vaccinated subjects. However, OPV vaccine recipients and
naturally immune persons provided correlations between PoBI and NT titers against
serotype 3 that were significantly lower. These correlations tended to be lower for the
other serotypes as well. This may be explained in part by the use of IPV vaccine (and
not OPV) as antigen in the PoBI. The greater difference observed in the serotype 3
PoBI test may be explained by a narrower immune response against serotype 3
poliovirus as compared with the other serotypes. In animals, site 1 of serotype 3 is
extremely immunodominant [3,9]. Therefore, the polyclonal coat for the serotype 3
PoBI may consist mainly of (neutralising) antibodies to site 1, and as a result may be
very sensitive to changes in immune response directed to site 1.
In this context, it is intriguing that differences in immune responses have been
observed between infection by wild-type 3 poliovirus or OPV at the one end and IPV-
induced immunity at the other. During infection with live viruses (wild-type or OPV),
site 1 of serotype 3 is cleaved by trypsine during the passage through the gut lumen,
thereby exposing other immunogenic sites on the viral capsid [14,15]. This trypsine
effect will not occur in IPV-recipients because the vaccine is given by intramuscular
injection. The trypsin-dependent immunogenic sites will therefore be less well
exposed to the immune system. In contrast to serotype 3, the PoBI assays for
serotypes 1 and 2 are probably reactive with antibodies to more than one antigenic site
so that differences between OPV- and IPV-vaccinees are not detected. Future work
will focus on the site-specificity of the human antibody response to poliovirus.
In conclusion, the newly developed PoBI test can replace NT in large scale population
studies for determining protective levels of antibodies to polioviruses. PoBI negative
sera should be re-tested by NT for confirmation of seronegativity. One of the major
advantages of the PoBI over the NT is that inactivated virus is used. In view of the
ongoing eradication of poliovirus, the use of live wild-type poliovirus in diagnostic
assays should be discouraged and eventually cease altogether in the near future.
This work was supported by a grant from The Foundation for the Advancement of
Public Health and Environment (SVM), Bilthoven, The Netherlands. We greatly
acknowledge the help of Dr H. Rümke for providing us with sera from naturally
immune persons. We would also like to thank the Utrecht blood donor centre for
providing us with serum samples from IPV-vaccinated adults. We gratefully
acknowledge the assistance of Annemarie Buisman, Jan Sonsma, Hafida Bentala,
Cees Verwey and Sandy Altena in performing the serum neutralisation assays.
Evaluation of a Poliovirus-Binding Inhibition-assay
1. Edevag G, Wahren B, Osterhaus ADME, Sundqvist VA, Granstrom M. Enzyme-Linked
immunosorbent assay-based inhibition test for neutralizing antibodies to polioviruses as an
alternative to the neutralization test in tissue culture. J Clin Microbiol 1995;33:2927-2930.
2. Emini EA, Jameson BA, Lewis AJ, Larsen GR, Wimmer E. Poliovirus neutralizing epitopes:
Analysis and localization with neutralizing monoclonal antibodies. J Virol 1982;43:997-1005.
3. Emini AE, Jameson BA, Wimmer E. Priming for and induction of antipoliovirus neutralizing
antibodies by synthetic peptides. Nature 1983;304:699-703.
4. Faden H, Modlin JF, Thoms ML, Marshall McBean A, Ferdon MB, Ogra PL. Comparative
evaluation of immunization with live attenuated and enhanced-potency inactivated trivalent
poliovirus vaccines in childhood: systemic and local immune responses. J Infect Dis
5. Ferguson M, Minor PD, Magrath DI, Yi-Hua Q, Spitz M, Schild GC. Neutralizing epitopes on
poliovirus type 3 particles: an analysis using monoclonal antibodies. J Gen Virol 1984;65:197-201.
6. Gershy-Damet GM, Koffi KJ. Utilisation of an ELISA technique for the quantification of
antipoliovirus antibodies in human sera. Bull Soc Pathol Exot Filiales 1987;80:289-294.
7. Hagenaars AM, van Delft RW, Nagel J, van Steenis G, van Wezel AL. A modified ELISA
technique for titration of antibodies to poliovirus as an alternative to a virus neutralization test. J
Virol Methods 1983;6:233-239.
8. Herremans MMPT, van Loon AM, Reimerink JHJ, Rumke HC, van der Avoort HGAM,
Kimman TG, Koopmans MPG. Poliovirus-specific IgA in persons vaccinated with inactivated
poliovirus vaccine (IPV) in The Netherlands. Clin Diagn Lab Immunol 1997;4:499-503.
9. Hogle JM, Filman DJ. Poliovirus : three-dimensional structure of a viral antigen. Adv. Vet. Sci.
Comp. Med. 1989;33:65-91.
10. Murthy N, Nair KM, Bhaskaram P. A direct ELISA technique to detect antibodies against
polioviruses. Indian. J. Biochem. Biophys. 1995;32:249-253.
11. Ogra PL, Karzon DT, Righthand F, MacGillyvrag M. Immunoglobulin response in serum and
secretions after immunization with live and inactivated poliovaccine and natural infections. N Engl
J Med 1968;279:893-900.
12. Ogra PL, Karzon DT. Formation and function of poliovirus antibody in different tissues. Progr
Med Virol 1971;13:156-193.
13. Osterhaus ADME, van Wezel AL, Hazendonk TG, Uytdehaag FGCM, van Asten JAAM, van
Steenis B. Monoclonal antibodies to polioviruses: comparison of intratypic strain differentiation of
poliovirus type 1 using monoclonals versus cross-absorbed antisera. Intervirology 1983;20:129-
14. Roivainen M, T. Hovi. Cleavage of VP1 and Modification of antigenic site 1 of type 2 poliovirus
by intestinal trypsine. J Virol 1988;62:3536-3539.
15. Roivanen M, Montagnon B, Chalumeau H, Murray M, Wimmer E, Hovi T. Improved
distribution of antigenic site specificity of poliovirus-neutralizing antibodies induced by a
protease-cleaved immunogen in mice. J Virol 1990;64:559-562.
16. Sabin AB, Michaels RH, Ziring P. Effect of oral poliovaccine in newborn children fed type 1
vaccine at birth. Pediatrics. 1963;31:641-650.
17. Simhon A, Lifshitz A, Abed Y, Lasch EE, Schoub B, Morag A. How to predict the immune
status of poliovirus vaccinees? A comparison of virus neutralization at a very low serumdilution
versus ELISA in a cohort of infants. Int J Epidemiol 1990;19:164-168.
18. Souvras M, Montagnon B, Fanget B, van Wezel AL, Hazendonk AG. Direct enzyme linked
immunosorbent assay (ELISA) for quantitation of poliomyelitis D-antigen.Dev Biol Stand
19. World Health Organization. Manual for the virological investigation of poliomyelitis. Geneva:
World Health Organization, 1990. (WHO document EPI/POLIO/90.1).
Differences in the antibody responses to
antigenic sites 1 and 3 of serotype 3
poliovirus after OPV or IPV vaccination
and after natural exposure
Tineke Herremans, Johan Reimerink, Tjeerd Kimman,
Harrie van der Avoort, Marion Koopmans
Submitted for publication
Three important antigenic sites involved in virus neutralisation have been identified
on polioviruses in mouse experiments. These sites are located at the surface of the
virion and have been designated antigenic sites 1, 2 and 3. The antibody response to
antigenic site 1 of serotype 3 poliovirus is considered to be immunodominant in mice,
but little is known about the immunogenicity of these sites in humans. It has been
reported that trypsin, present in the intestinal fluids, can cleave serotype 3
polioviruses at precisely this immunologically important antigenic site, thereby
altering its antigenic properties. In theory, therefore, the site-specific antibody
response induced by intramuscularly-applied inactivated poliovaccine will differ from
that induced by orally-administered live poliovirus that has been exposed to trypsin in
the gut lumen. In the present study, we developed inhibition ELISA assays specific
for (mouse) antigenic sites 1 and 3 in order to measure antibody responses to these
sites in fully-vaccinated IPV (n=63) and OPV-recipients (n=63), and in naturally
infected persons (n=25). We found that both sites are strongly immunogenic in
humans. Similar levels of site-specific antibodies were found for sites 1 and 3 in
naturally infected persons. Similar levels of antibodies to site 1 were detected in IPV
and OPV vaccinees. However, significantly more OPV recipients (88.7%) had
detectable antibodies to antigenic site 3 (p<0.01) when compared to IPV-vaccinated
persons (63.1%). Both previously IPV- and OPV-vaccinated persons responded with a
significant increase in antibodies to sites 1 and 3 after an IPV booster vaccination
(p<0.01). We conclude that the immune response following natural infection with
serotype 3 poliovirus in humans consists of both site 1- and site 3-specific antibodies,
and that these responses can be induced by OPV or by recent IPV vaccination.
The poliovirus capsid consists of 20 copies of each of the four structural virion
proteins (VP1, VP2, VP3 and VP4) . The epitopes responsible for inducing
poliovirus-neutralising antibodies are located on surface-exposed loops in the three
structural proteins: VP1, VP2 and VP3 . VP4 is completely located on the inside of
the viral capsid and plays no known role in the induction of poliovirus-neutralizing
antibodies. VP1 is the most exposed surface protein and plays a major role in the
induction of neutralizing antibodies for all three the poliovirus serotypes .
Three important antigenic sites (epitopes) involved in virus neutralisation have been
identified on polioviruses and have been designated as site 1, 2 and 3 [9,15]. These
sites have been identified through the isolation and characterisation of Sabin mutant
strains resistant to neutralisation by poliovirus-specific antibodies and by epitope
mapping using neutralizing monoclonal antibodies .
Antigenic site 1, composed of amino acids 89 to 100 of VP1, is a major immunogenic
site for serotype 2 and 3 polioviruses as determined by neutralizing monoclonal
antibodies induced in mice . This site is usually immunorecessive in serotype 1
poliovirus . Antigenic site 2 is a complex site including residues 220 to 222 of
VP1 (site 2a) as well as residues 169 and 170 on VP2 (site 2b) . Sites 2a and 2b
have both been detected in serotype 1 poliovirus, while only site 2b has been detected
in serotype 3 poliovirus. Site 3 is also a complex site and includes the residues 286 to
290 from VP1 (site 3a) as well as the residues 58 and 59 and others from VP3 (site
3b). Both sites 3a and 3b have been detected in serotype 3 poliovirus, while
Differences in antibody responses to antigenic sites 1 and 3
neutralising antibodies to site 3b have been detected in serotype 1 poliovirus only,
suggesting that site 3a is not immunogenic in serotype 1 poliovirus .
The location of the amino residues within the three dimensional structure of the virion
indicates that the majority of these amino acids residues are highly exposed and
located within prominent structural features of the viral surface . A deep canyon
or pit on the surface of poliovirus has been identified as the receptor binding site .
The neutralising epitopes themselves are not involved in receptor binding, but binding
of antibodies to these spots probably causes steric hindrance with the actual receptor
binding site within the canyon . Whether all these sites are also antigenic for
humans is not clear.
It has been reported that trypsin treatment can cleave both serotype 1 and serotype 3
polioviruses at antigenic site 1 at residue 98 (arginine) [4,11,23,25]. While the
poliovirus retains its infectivity in both cases, its antigenic properties are drastically
altered, and the trypsin-cleaved viruses are not neutralised or immunoprecipitated by
monoclonal antibodies to site 1 of non-treated virions . This trypsin effect will not
occur in IPV recipients who have received the vaccine through intramuscular
injection. If antigenic site 1 of poliovirus serotype 3 is also immunodominant in
humans, vaccination with IPV could theoretically induce predominantly neutralizing
antibodies to site 1, leaving a possible gap in the immune response to trypsin cleaved
serotype 3 poliovirus [9,15].
This study compared the site-specific humoral immune responses of naturally infected
and IPV- or OPV-vaccinated persons for poliovirus serotype 3. The effect of an IPV
booster vaccination on the site-specific antibody titers was also examined.
Materials and methods
Negative control serum samples were used to test the specificity of the antigenic site
1- and 3-specific assays. These samples were obtained from non-vaccinated children
(n=20) from a religious group in The Netherlands that refuses vaccination. The sera
had been pre-screened by neutralisation test for absence of neutralising antibodies to
poliovirus (titer, <2) . The seroprevalence of antigenic site 1- and 3-specific
antibodies after vaccination was determined for sera from IPV-vaccinated healthy
blood donors (n=63) and from age-matched OPV-vaccinated blood donors from
Belgium (n=63) who had received a complete series of vaccination as children .
The IPV-vaccinated (n=11) and OPV-vaccinated subjects (n=10) were given an IPV
booster vaccination to determine the influence of IPV vaccination on the induction of
antigenic site 1- and 3-specific responses for the different vaccine backgrounds. Blood
samples were collected before booster vaccination and at 3, 7 and 28 days post-
vaccination. Full details of this study have been described elsewhere . Serum
samples from poliomyelitis patients (n=25) from the 1992/1993 serotype 3 epidemic
in The Netherlands were tested to determine antigenic site 1- and 3-specific responses
after natural infection .
A panel of serotype 3-specific monoclonals with specificity for antigenic site 1, 2 or 3
was used to test the newly developed assays for specificity. The monoclonals 204 (site
1), 877 (site 2) and 889 (site 3) were kindly donated by Dr. Ferguson (NIBSC, UK)
[14,15,22]. Monoclonals 2-13D9 (site 1), 2-15E4 (site 1) and 4E5E9 (site 2/3) were
produced at the RIVM (Bilthoven, The Netherlands) [12,20].
Sabin mutant virus strains 335 (site 1) and 4021 (site 3) (kindly provided by Dr
Ferguson) were grown on Hep-2C cells in Eagle MEM supplemented with 10% FCS,
until full cytopathic effect developed . The culture supernatant was collected and
centrifuged for 30 min at 3000 rpm to remove cell debris. The supernatant was
extracted with 10% v/v arklone at 4°C for 45 min during constant shaking followed
by centrifugation (30 min, 3000 rpm). The supernatant was concentrated by direct
ultrafiltration using membranes with a molecular weight cut-off of 10kD (type PM10;
Amicon) . The optimal working dilution of the concentrated virus in the site-
specific assay was determined by checkerboard titration.
Antigenic site 1-specific Poliovirus-Binding Inhibition assay (PoBI)
The site-specific PoBI assays were a modification of the previously described PoBI
assay , except that monoclonal antibodies to specific sites were used as capture
antibodies in the ELISA. The inhibition of the ELISA signal depends on both
reduction of virus-antigen binding to the capture antibody due to the presence of
competing (blocking) antibodies as well as reduction of binding to the indicator
monoclonal antibody by the same mechanism. Briefly, two-fold dilutions of serum
samples were made directly into the wells of microtiter cell culture plates (Greiner,
Alphen aan de Rijn, The Netherlands) and poliovirus antigen was added to each well.
The antigenic site 3 Sabin mutant strain 4021 (kindly provided by Dr. Ferguson) was
used as antigen to prevent cross reactivity with this antigenic site . Serum-virus
mixtures were incubated for 2 hours at 37°C. After washing of the plates, the pre-
incubated serum-virus mixture was transferred to ELISA plates, that had been coated
overnight with the antigenic site 1-specific monoclonal 2-13D9 (IgG isotype) in a
dilution of 1/8000 . After the incubation, the homologous monoclonal antibody
(2-13D9), labelled with horseradish peroxidase, was used to detect bound poliovirus
. TMB was used as a substrate, and colour development was stopped after 15
minutes by the addition of 2M H2SO4. The plates were read at 450nm. Serum samples
were considered positive if a reduction in extinction of ≥50% was reached. The
reciprocal of the first serum dilution that was positive in the inhibition test was taken
as the titer of the test sample. For specificity testing, serum was replaced by serial
dilution of monoclonals to site 1, 2 and 3. Optimal dilutions of reagents were
determined by checkerboard titration.
Antigenic site 3-specific PoBI
The antigenic site 3-specific assay was conducted in a manner similar to the site 1
assay described above, using site 3-specific IgM monoclonal 889 as a capture
antibody in the PoBI. These IgM molecules were first degraded into F(ab’)2
Differences in antibody responses to antigenic sites 1 and 3 Download full-text
fragments by pepsin according to the manufacturer’s instructions [Pierce, Rockford,
USA], since IgM antibodies cannot be coated efficiently . The Sabin mutant strain
335 (kindly provided by Dr. Ferguson) was the antigen used to reduce cross-reactivity
within the antigenic site 1 . Biotin-labelled antigenic site-3 specific monoclonal
889 was used to detect bound antigen. Avidin conjugated with alkaline phosphatase
(Sigma, Zwijndrecht, The Netherlands) was added and incubated for 1 hour at 37°C.
The plates were washed, and 100 µl of p-nitrophenylphosphate at a concentration of
1mg/ml in 0.1 M glycine buffer was added to each well. The plates were read at 405
nm after incubation at room temperature for 30 minutes. Results were analysed as
described for the site 1-specific assay.
Poliovirus-Binding Inhibition assay (PoBI)
A Poliovirus-Binding Inhibition assay (PoBI) was used to determine the total number
of poliovirus-specific antibodies as an indicator of neutralizing antibodies, and was
performed as described . Serum samples were considered positive if a reduction in
extinction of ≥50% was reached. The reciprocal of the first serum dilution that was
positive in the inhibition test was taken as the titer of the test sample.
An unpaired Student’s t-test was used to evaluate the differences between the titers of
site-specific antibodies to poliovirus between two groups. P values of <0.01 were
Specificity of the antigenic site 1- and site 3-specific PoBI serotype 3 assays
None of the negative control serum samples tested (n=20) showed any inhibition of
the antigenic site 1 and site 3 PoBI signal (data not shown). A panel of monoclonals
was used to test the site specificity in the site 1 and site 3 PoBI assay. The antigenic
site 1-specific monoclonals 2-13D9 (titer, >12800; homologous), 2-15E4 (titer,
12800) and 204 (titer, >12800) were able to inhibit the antigenic site 1 PoBI signals,
whereas none of the monoclonals (4E5E9, 877, 889) to antigenic sites 2 and 3 of
serotype 3 poliovirus did so (titer, <100). In the site 3-specific assay, the homologous
monoclonal was able to inhibit the PoBI signal at a high level (titer 6400), and only
low level cross-reactivity was detected with site 1-specific monoclonals 204 (titer,
200), 2-15E4 (titer, 400) and 2-13D9 (titer, 400).
PoBI and antigenic site 1- and site 3-specific immune responses in poliomyelitis
Twenty-five patients from the 1992/93 serotype 3 outbreak in The Netherlands were
tested using the site-specific assays . A median antibody titer of 2log 6 to both
antigenic site 1 and 3 was detected in all patients [Figure 1].