Purification of Influenza B virus Hemagglutinin by Isoelectric vaccine jab

Document Sample
Purification of Influenza B virus Hemagglutinin by Isoelectric  vaccine jab Powered By Docstoc
					                                                                     JOURNAL OF PREVENTIVE MEDICINE AND HYGIENE 2004; 45: 17-20

          Purification of Influenza B virus Hemagglutinin
                       by Isoelectric Focusing
                                            G. ICARDI*, F. ANSALDI*, P. CROVARI*
     Department of Health Sciences, University of Genoa; ** Department of Physiopathology Experimental Medicine and Public
        Health, University of Siena; *** Department of Internal Medicine, Endocrine-Metabolic Science and Biochemistry,
                                                        University of Siena

                                                             Key words

                                              Influenza • Purification • Hemagglutinin

Influenza viruses cause recurring illnesses in individuals and       means of this technique, we separated Hemagglutinin of the
recurring epidemics in the general population, affecting hun-        strain B/Panama/45/90 from other proteins present in the stud-
dreds of millions of people every year. Today, annual vaccina-       ied bulk, which is the material generally, used to prepare the
tion is the most effective means of controlling influenza. To fur-   subunit influenza vaccine. In conclusion, the isoelectric focus-
ther optimise influenza vaccines and to improve the perfor-          ing technique appears to be useful in order to improve the safety
mance of tests used in the evaluation of vaccine response for        and tolerability of Influenza vaccines. Moreover, purified
characterisation of isolates, a new method for purification of       influenza virus antigens can improve the performances of the
the surface antigens of influenza viruses based on isoelectric       laboratory tests in order to study immunogenicity elicited by the
focusing method has been standardised and evaluated. By              vaccines.

Introduction                                                         However, results of the HI test show a very wide vari-
                                                                     ability and difficulties are encountered in comparing
Influenza, with recurrent worldwide epidemics, pre-                  the results of HI with those obtained with SRH. Be-
sents a major health problem. It affects hundred mil-                sides the neutralisation assay has several limitations
lions of people every year, with high morbidity in                   (for example: the standardisation of the assay is not-
people of all ages, particularly in children, and high               immediate/easy, the fulfilment is quite time-consuming
mortality in high-risk populations 1. In the United                  and it not allows the handling of a large number of iso-
States alone, infections with influenza virus result in              lates by a single run, etc.).
approximately 20,000 deaths and 100,000 hospitalisa-                 To further optimise influenza vaccines and improve the
tions each year 2. Annual vaccination remains the most               performance of laboratory tests, the present investiga-
effective means of controlling influenza, although the               tion was aimed at studying the purification of the main
safety profile, the efficacy and the effectiveness are               surface antigens of influenza viruses.
sub-optimal. Several strategies are currently being
pursued to increase the quality and quantity of in-
fluenza vaccines that are used, on an annual basis,                  Material and methods
with the aim of improving safety, tolerability, im-
munogenicity and effectiveness of the licensed inacti-
vated vaccines. The safety and tolerability of influen-
za vaccines depend on several factors, albeit the puri-              Chiron Vaccines, Siena, Italy, provided purified mono-
ty of the product plays a major role.                                valent B/Panama/45/90 subunit influenza antigens.
The efficacy of vaccines is first evaluated by means of              The bulk was prepared by growing up seed virus in fer-
immunogenicity tests. In Europe, producers must car-                 tilised chicken eggs. The virus is then chemically inac-
ry out clinical trials, every year, for the commerciali-             tivated and, after digestion by detergents, the product
sation of influenza vaccines. Only Hemagglutination                  was extensively purified and concentrated by zonal
Inhibition (HI) and Single Radial Haemolysis (SRH)                   centrifugation.
are authorised by the European Agency for Evaluation                 The final bulk contained almost exclusively: Hemag-
of Medical Products (EMEA), to evaluate the im-                      glutinin (HA), Neuraminidase (NA), small amounts of
munogenicity. Recently the World Health Organiza-                    chicken proteins, thiomerosal and cetylmethyltriethy-
tion (WHO) emphasised the advantages of the neu-                     lammonium chloride.
tralisation test for detecting antibody to influenza                 The total amount of proteins was about 700 micro-
virus.                                                               gramms/ml (Kjeldahl method).

                                           R. GASPARINI, ET AL.

                                                                   Fig. 1 - pH distribution in Rotofor fractions
The samples (20 ml, containing approximately 14,000
micrograms of total proteins) were dialyzed against                    12
bidistilled water, then the specimens were concentrated                10
using the Sped-vac sublimation system and finally re-                  8
suspended in 1,400 µ l of bidistilled water. Protein con-              6
tent and HA and NA activities were measured in the fi-                 4
nal samples.
We then analysed the sample protein contents by 12%
SDS-polyacrylamide gel electrophoresis (SDS-PAGE)                              1   3   5       7   9    11        13   15   17    19
(Biorad) and compared them with the Silver Stain
SDS-Page Biorad Standards. The SDS polyacrylamide
gel system ran according to Laemmli’s method (1970).
To estimate the isoelectric point of the B/Panama/45/90
HA, we calculated a mean of the 13 known isoelectric              With the 20 fractions we are able to recover about 30
points. To this end, we used the values, obtained con-            micrograms/ml of proteins, and so because of 2 ml for
sulting ExPASy Molecular Biology Server, of the fol-              each fraction was obtained, 1,200 out 2,000 micro-
lowing viruses: B/Lee/40, B/Maryland/59, B/Singa-                 grams of proteins of the final sample (60%) were re-
pore/222/79, B/Oregon/5/80, B/URSS/100/83, B/Victo-               covered.
ria/3/85, B/Thailand/24/85, B/Ann Arbor/1/86,                     Most proteins were in two peaks (Fig. 2) corresponding
B/Memphis/6/86, B/Victoria/2/87, B/Beijing/1/87,                  to 3-4 and 9-11 fractions. The first one (total protein
B/England/222/82 and B/Bonn/43.                                   amount = 3.6 micrograms/ml) had no Hemagglutinin
                                                                  and Neuraminidase activity; however, this peak was,
ISOELECTRIC FOCUSING, RUNNING CONDITIONS                          probably, the Neuraminidase. The NA activity was ab-
The treated suspension (200 µ l; containing 2,000 mi-             sent, probably because of inactivation by the previous
crograms of proteins) was loaded onto the Rotofor cell.           treatments. The second one (total protein amount =
Focusing was carried out at 12 W constant power for 4             17.1 micrograms/ml) contained 507 HA units/ml.
hours at 4°C the initial conditions were generally 600            Therefore, we recovered the 15% of the HA activity in
V and 20 mA. At equilibrium, the conditions were ap-              9-11 fractions. In these fractions evaluation of biologi-
proximately 1,200 V and 10 mA. Twenty fractions were              cal NA activity was negative.
collected (about 2 ml each) and the pH measured. The              The protein amount in the fraction 9, 10 and 11 was 3.4,
protein content (micrograms/ml) in each fraction was              11.8 and 1.9 micrograms/ml, respectively.
then assessed (DC, Protein Assay, Biorad).                        Distribution of the proteins obtained with SDS-Page on
                                                                  fractions 9 (lane 2), 10 (lane 3) and 11 (lane 4) are
VISUALISATION OF PROTEINS USING SDS-PAGE                          shown in Figure 3. Lane 5 corresponds to the bulk and
Aliquots from each Rotofor fraction were separated by             shows head and foot (HA1 and HA2) of HA (molecu-
SDS-PAGE and the polyacrylamide gels were stained                 lar weight 47 and 29 kD, respectively), NA (molecular
with silver (Pharmacia).                                          weight 60 kD) and negligible quantity of nucleoprotein
                                                                  and matrix protein. Only the head and foot of HA were
                                                                  recovered in fractions 9-11.
HA activity was tested with chicken Red Blood Cells
(RBC), according to the standard virological technique 3.
NA activity was measured according to the method de-               Fig. 2 - Distribution of proteins in Rotofor fractions
scribed by Lambré et al. 4.

Results                                                                10
In the bulk there were 16,384 HA units/ml.                              6
The mean of the isoelectric points of the mentioned 13                  4
B viruses was 8.4.                                                      2
The values of pH of 20 eluted fractions ranged from 5.9                    0
to 9 (Fig. 1).                                                                 1   4       7       10        13        16        19
The theoretical forecast pH of 8.4 corresponded to the
fraction 9-11.

                                                                            PURIFICATION OF INFLUENZA B VIRUS HEMAGGLUTININ BY ISOELECTRIC FOCUSING

                                                                            ing to the safety and tolerability of the influenza vac-
 Fig. 3 - Silver-stained polyacrilamide gel of Rotofor fractions.
                                                                            However, it is also true that small quantities of impuri-
                                                                            ty, particularly proteins, can improve the immuno-
                                                                            genicity of vaccines. In an empirical perspective that
                                                                            would be acceptable, it is not from a rigorously scien-
                                                                            tific point of view. In fact, it is necessary to put into
                                                                            vaccines only substances of which the biological ef-
                                                                            fects are known.
                                                                            The 3 types of inactivated influenza vaccine, at present
                                                                            used, differ in terms of reactogenicity depending upon
                                                                            to the different purification levels. Thus, in 15-20% of
                                                                            vaccines, the whole vaccines cause local reactions last-
 Lane 1: Silver stain SDS-Page Biorad Standard. Lane 2-4: fraction 9, 10
                                                                            ing for 1-2 days. These reactions appear to be more
 and 11, respectively. Lane 5: untreated sample (bulk)                      common in young children than in adults. Transient
                                                                            systemic reactions, such as fever, malaise and myal-
                                                                            gias, may occur in a minority of vaccine recipients.
Discussion                                                                  Split and subunit vaccines show lower systemic reacto-
                                                                            genicity both in children and adults as compared to the
HA and NA are spike-like structures inserted in the en-                     whole virus preparations. As far as concerns the severe
velope of the influenza virion and are the main antigens                    adverse events, WHO estimates an excess of 1-2 cases
of the virus.                                                               of Guillain-Barré syndrome (GBS) per million subjects
The HA molecule consists of a globular head on a stalk.                     vaccinated 11. As far as the possible aetiology of GBS,
The head is made exclusively of HA1 (47 kD) and con-                        Chen maintains that one possible cause of the associa-
tains the receptor-binding cavity as well as most of the                    tion of GBS and influenza vaccination is that the flu
antigenic sites of the molecule. The stalk (29 kD) con-                     vaccine could contain Campylobacter. In fact, 40-50%
sists of all HA2 and part of HA1. NA is also an integral                    of chickens are infected with this pathogen, which is
membrane glycoprotein and the second major surface                          difficult to eradicate 12. Furthermore, Geier et al. 13
antigen of the influenza virion 5. Since both surface                       found that influenza vaccines contained from a 125- to
antigens are membrane-associated glycoproteins with                         a 1,250-fold increase in endotoxin concentration in
similar biochemical properties, it is difficult to separate                 comparison to an adult tetanus-diphtheria vaccine 13.
the two components 6.                                                       The availability of purer HA could eliminate the possi-
Several approaches have been used to purify the HA                          ble risks related to this endotoxin. Finally, Weise and
and NA antigens. Gallangher et al. 6 and Johansson and                      Carnegie (1988) stressed the similarity in the human
Kilburne 7 adopted the DEAE-Sephadex technique.                             basic proteins from myelin, particularly P2, and NS2
Bucher in 1977 and Gerentes et al. in 1996 used the im-                     (non-structural protein) of influenza virus 14. This pro-
munochromatography method 8 9. These techniques are                         tein shows a molecular weight of 12 kD and our 9-11
difficult to standardize and are both time-consuming                        fractions (Fig. 3), probably, do not contain this.
and expensive. By means of isoelectric focusing, we                         Therefore, those subjects presenting allergic syn-
were able to separate HA from other proteins contained                      dromes would better accept an influenza vaccine with
in the studied bulk, which is the material usually used                     greater purity 15.
to prepare the subunit influenza vaccine. Whereas this                      Moreover, the isoelectric focusing technique allows pu-
technique seems not to damage the HA activity, likely                       rification of influenza virus antigens obtained with
it changes the three-dimensional structure of the NA.                       molecular biology methods. For example, Johansson
In 1999, the World Health Organization’s Department                         obtained influenza A HA and NA expressed by recom-
of Vaccines and Biologicals (WHO/V&B), following a                          binant baculovirus-infected insect cell lines 16.
broad consultation process, established three time-lim-                     A purer antigen would also be more advantageous to
ited Priority Projects to address key issues: the areas of                  prepare influenza vaccines with more than three viral
innovation in vaccines and vaccine delivery systems;                        components. This work hypothesis may be considered
strengthening of immunization systems and accelerat-                        as different variants of A/IH3N2 or different lineages
ing disease control. Immunization safety was identified                     could co-circulate during the same season 17 18.
as one of the main elements in the strengthening of im-                     Purer antigens may represent an important laboratory
munization systems. A Priority Project on Immuniza-                         tool for the characterisation of isolates by means of bet-
tion Safety was, therefore, established 10. Results                         ter and purer monoclonal antibodies and for the evalu-
emerging from the present investigation could con-                          ation of the immunity exerted by vaccine administra-
tribute to achieve the goals of this program, contribut-                    tion.

                                                       R. GASPARINI, ET AL.

In fact, HI is not easily reproducible and SRH does not                       tinin could reduce the aggregation phenomenon, typi-
discriminate well between closely related virus strains                       cal of the whole virus or the subunit particles.
and some concern exists regarding the correlation be-                         Finally, purer HA and NA could be important improv-
tween SRH antibody and protection 19. The purified                            ing for standardisation of recent-developed laboratory
Hemagglutinin would avoid cross-reactions of antibod-                         techniques for evaluation of immune-response after
ies against different genotypes of the same virus strain,                     natural influenza infection or vaccination. Particularly,
increasing specificity of HI and SRH. The purification                        antigens obtained by isoelectrofocusing could be em-
of surface antigens would also reduce interference be-                        ployed to quantify the humoral response by enzyme-
tween antibodies against NA and HA, particularly for                          linked immunosorbent assay (ELISA test) and to study
the SRH technique. Furthermore, purified Hemagglu-                            cell-mediated one 20.

References                                                                         Chen R. Cited in: Possible link between flu jab and Guillain-Bar-
                                                                                   ré syndrome under investigation. BMJ 2003;326:620.
     Glezen WP. Serious morbidity and mortality associated with in-           12
                                                                                   Marwick C. Possible link between flu jab and Guillain-Barré syn-
     fluenza epidemics. Epidemiol Rev 1982;4:25-44.                                drome under investigation. BMJ 2003;326:620.
     Centers for Disease Control and Prevention. Prevention and con-          13
                                                                                   Geier MR, Geier DA, Zahalsky AC. Influenza vaccination and
     trol of influenza: recommendations of the Advisory Committee on               Guillain-Barré syndrome small star, filled. Clin Immunol
     Immunization Practices (ACIP). Morb Mortal Wkly Rep                           2003;107:116-21.
     2000;49:619-22.                                                          14
                                                                                   Weise MJ, Carnegie PR. An approach to searching protein se-
     Couch RB, Kasel JA. Influenza. In: Lennette EH, Lennette DA,                  quences for superfamily relationships or chance similarities rele-
     Lennette ET, eds. Diagnostic Procedures for Viral, Rickettsial,               vant to the molecular mimicry hypothesis: application to the ba-
     and Chlamydial Infections. Washington: APHA 1995:431-46.                      sic proteins of myelin. J Neurochem 1988;51:1267-73.
     Lambré CR, Terzidis H, Greffard A, Webster RG. Measurement               15
                                                                                   Bierman CW, Shapiro GG, Pierson WE, Taylor JW, Foy HM, Fox
     of anti-influenza neuraminidase antibody using peroxidase-                    JP. Safety of influenza vaccination in allergic children. J Infect
     linked lectin and microtitre plates coated with natural substrates.           Dis 1997;136:652-5.
     J Immunol Methods 1990;135:49-57.                                        16
                                                                                   Johansson BE. Immunization with influenza A virus hemagglu-
     Webster RG, Bean WJ, Gorman OT, Chambers TM, Kawaoka Y.                       tinin and neuraminidase produced in recombinant baculovirus
     Evolution and ecology of Influenza A Viruses. Microbiol Rev                   results in a balanced and broadened immune response superior to
     1992;56:152-79.                                                               conventional vaccine. Vaccine 1999;17:2073-80.
     Gallangher M, Bucher DJ, Dourmashkin R, Davis JF, Rosenn G,              17
                                                                                   Ansaldi F, D’Agaro P, De Florentiis D, Puzelli S, Lin YP, Grego-
     Kilburne ED. Isolation of neuraminidase of human influenza                    ry V, et al. Molecular characterization of influenza B viruses cir-
     viruses by combination of genetic and biochemical procedures. J               culating in Northern Italy during the 2001-2002 epidemic season.
     Clin Microbiol 1984;20:89-93.                                                 J Med Virol 2003;70:463-9.
     Johansson BE, Kilburne ED. Programmed antigenic stimulation:             18
                                                                                   O’Donnel FT, Munoz FM, Atmar RL, Hwang LY, Demmler GJ,
     kinetics of the immune response to challenge infections of mice               Glezen WP. Epidemiology and molecular characterization of co-
     primed with influenza inactivated whole virus or neuraminadase                circulating influenza A/H3N2 virus variants in children: Houston,
     vaccine. Vaccine 1991;9:330-3.                                                Texas, 1997-8. Epidemiol Infect 2003;130:521-31.
     Bucher DJ. Purification of neuraminidase from influenza viruses          19
                                                                                   Wood JM, Gaines-Das RE, Taylor J, Chakraverty P. Comparison
     by affinity chromatography. Biochim Biophis 1977;82:393-9.                    of influenza serological techniques by international collaborative
     Gerentes L, Kessler N, Thomas G, Aymard M. Simultaneous pu-                   study. Vaccine 1994;12:167-74.
     rification of influenza hemagglutinin and neuraminidase proteins         20
                                                                                   Gasparini R, Pozzi T, Caniffi A. Enzyme immunoassay for the de-
     by immunochromatography. J Virol Methods 1996;58:155-65.                      tection of antibodies towards the surface antigens of influenza
     Scholtz M, Duclos P. Immunization safety: a global priority. Bull             viruses. J Preventive Med Hygiene 1990;31:14-7.
     World Health Organ. 2000;78:153-4.

s Correspondence: Prof. Roberto Gasparini, Department of Health
  Sciences, University of Genoa, via Pastore 1, 16132 Genova,
  Italy - Tel. +39-010-3538527 - Fax +39-010-3538407 E-mail: ga-


Shared By:
Description: Purification of Influenza B virus Hemagglutinin by Isoelectric vaccine jab