Journal of Zoo and Wildlife Medicine 40(4): 731–743, 2009 Copyright 2009 by American Association of Zoo Veterinarians

SEROLOGIC RESPONSE AND SAFETY TO AGAINST USING INACTIVATED H5N2 IN ZOO BIRDS

Alexis Le´cu, D.V.M., Christophe De Langhe, D.V.M., Thierry Petit, D.V.M., Fre´de´ric Bernard, D.V.M., and Hanny Swam, D.V.M.

Abstract: Due to the spread of the H5N1 highly pathogenic strain of avian influenza virus across Europe, a preventive vaccination occurred in early 2006 among 135 French zoologic institutions. Approximately 25,000 birds were vaccinated with a H5N2 . Among them, 4,369 birds were monitored by members of Association Francophone des Ve´te´rinaires de Parc Zoologique regarding safety issues of the vaccination protocol. A total of 1,686 blood samples were collected before the first injection (n 5 255), at the time of booster (n 5 463), 60 day after the booster (n 5 514), and 180 day (n 5 229) and 330 day (n 5 217) after the initial injection. Thus, sera of 126 species representing 15 different avian orders were tested using the hemagglutinin inhibition assay to evaluate seroconversion and the long-term serologic profile of selected anti-H5 antibody. Safety was considered satisfactory in all orders, and there were no deleterious effects on large-volume injection/body weight ratio. After the second injection, 71% of the birds developed a titer $32, with a mean titer of 558. Titers then decreased in all birds, with 42% of the remaining birds having a titer $32 at day 180 and only 26% at day 330. Results demonstrated that a booster 42 days after initial vaccination was mandatory to raise the titer above 32, considered to be the protective level in poultry, and to increase the number of seroconverted birds. Differences in the serologic responses among the orders and species of birds were detected and could be linked with the variation of vaccine dose injected per body weight or with species-specific immune response. The protocol for additional campaigns will be adjusted for some bird orders through the increase of injected dose or a half yearly booster to sustain better titers over the year. Vaccination is a useful tool, together with biosecurity, that should always be used as a primary method of preventing and controlling avian influenza outbreaks. Key words: avian influenza, birds, H5N1, H5N2, vaccine.

INTRODUCTION H5N1 strain appeared in Europe in the late summer of 2005 and affected countries surround- Three types of influenza viruses belong to the ing France in the fall of 2005.2 HPAI is classified Orthomyxoviridae family: A, B, and C. Only as a notifiable disease by the World Organization influenza A viruses infect birds, and all known for Animal Health. The Economic European subtypes of influenza A viruses can infect birds. Community (EEC) opted for a nonvaccination Avian influenza (AI) is characterized as low policy, with long-term confinement of noninfected pathogenic (LPAI) or high pathogenic (HPAI), birds and culling of infected ones as a standard based upon its ability to cause disease in poultry. measure to control HPAI spread. In response to Only influenza A viruses are further classified by the threat of HPAI to highly endangered species subtype on the basis of the two main surface housed in zoologic institutions,24 the EEC allowed glycoproteins: hemagglutinin (HA) and neur- European zoos to use vaccination as a preventive aminidase. To the authors’ knowledge, only tool, based on decision 2005/744/EC. France and LPAI of subtypes H5 and H7 have become 12 other member countries of the EEC applied for HPAI. Wild birds, both free ranging and captive permission to vaccinate birds in 135 zoologic in zoos, are known to be susceptible to AI,9,10,11,12,26 institutions. The EEC required that vaccination including the recent H5N1 HPAI strain.7,9,31 programs include close monitoring procedures However, morbidity and mortality varies greatly and the use of an inactivated vaccine using the between avian species. The highly pathogenic ‘‘Differentiating Infected from Vaccinated Ani- mals’’ technology.4 This was achieved by use of a From Parc Zoologique de Paris, Muse´um National (different N subtype from d’Histoire Naturelle, 53 av de St Maurice, 75012 Paris, the field strains). The implementation of vaccina- France (Le´cu); Intervet SA, 49071 Beaucouze cedex, tion in zoos was discussed with the Association of France (De Langhe, Bernard); Zoo de la Palmyre, 17570 Les Mathes, France (Petit); and Intervet Inter- French Zoo Veterinarians (Association Franco- national BV Boxmeer 5830 AA, The Netherlands phone des Ve´te´rinaires de Parc Zoologique). (Swam). Correspondence should be directed to Dr. The aim of this study was to evaluate the safety Le´cu ([email protected]). and serologic response to an inactivated H5N2 731 732 JOURNAL OF ZOO AND WILDLIFE MEDICINE vaccine administered to different avian species in Serology zoologic institutions and to discuss the result of Birds were individually monitored veterinari- the vaccination protocol to determine how ans of the Association of French Zoo Veterinar- vaccination boosters should be implemented in ians. A total of 1,686 blood samples were the vaccination protocol for HPAI. collected before the first injection (n 5 255), at MATERIALS AND METHODS the time of booster (n 5 463), 4 wk after the booster (n 5 514), and 180 day (n 5 229) and 330 Vaccination day (n 5 217) after the initial injection. The 1,678 An inactivated, water-in-oil adjuvanted H5N2 collected sera represent a total of 15 taxonomic (A/duck/Pottsdam/1402/86) vaccine (Nobilis In- orders for 126 species from 22 zoos. Blood was fluenza H5N2, Intervet SA, 49071 Beaucouze´, collected from ulnar, jugular, or tarsal veins. France) able to induce at least 6 log2 hemagglu- Serum was then harvested and stored between tination inhibition (HI) units/ml antibody rise in 218u and 225uC until shipment to lab. poultry was used for this study. This is corre- To evaluate humoral response to vaccination, sponding to an antigenic mass of 0.125 mgHA H5 antibody titer is determined in a derived HI 5 protein per 0.5 ml as a maximum.21 All the birds test. Influenza virus particles contain HA proteins were vaccinated twice with a 6-wk interval on their surface. These proteins have the capacity between . A subcutaneous route or to agglutinate chicken red blood cells spontane- intramuscular route was used depending on ously. This agglutination can be inhibited by restraint means, bird size, and the veterinarians. antibodies directed against these HA proteins. Birds with a body weight , 1.5 kg received Briefly, 50 ml serum is diluted twofold in phos- 0.25 ml and birds $1.5 kg received 0.5 ml, except phate-buffered saline in V-bottomed microwell Anseriformes and Struthioniformes, which were plastic plates (651101, Greiner, Les Ulis, 91741 given 1 ml. All birds were manually restrained for Courtaboeuf Cedex, France). Thereafter, eight vaccine administration. HA/50 ml of the inactivated homologous antigen (A/duck/Potsdam 2243/84) is added to each well. Safety monitoring After incubation at room temperature for 15 min, 25 ml1% (v/v) chicken red blood cells are added to A questionnaire was sent to all veterinary each well. The plates are incubated at room members of the Association of French Zoo temperature for 1 hr. Thereafter, the plates are Veterinarians after administration of the second read and HI titers determined as the value of the dose to collect information on adverse effects highest dilution of serum causing complete inhibi- associated with the vaccination. Twenty-five tion of the eight HA units of virus. Both positive member institutions responded with reports on andnegativecontrolseraareincludedineachtest. 4,369 different birds. Information collected from All results are given in log2 expression of titer. this questionnaire included body weight, volume H5 antibody titers in nestlings of 64 birds of vaccine administered, and presence of local and/ among six different species were monitored. HI or systemic vaccine reaction. More specifically, a assay was performed using the same protocol general score ranging from one to four based on than in adults. Rheas and aras were hatched from the safety of vaccine was proposed to the vaccinated parents (150–170 day after initial veterinarians for each bird (one 5 poor: mortality vaccination of parents). induced by vaccination itself; two 5 intermediate: bird with transient life threatening signs related to RESULTS vaccination; three 5 good: bird with transient but Safety not life-threatening signs related to vaccination; and four 5 high: birds without any signs related to Results of the questionnaire are reported in vaccination). All institutions enhanced their influ- Table 1. The tolerance score was reported to be enza biosecurity measures, and every vaccinated ‘‘good’’ or ‘‘high’’ for all birds. Local reactions to bird that died was submitted for a complete the vaccine administration were reported in four postmortem examination, including pathology, if birds: three on Anseriformes (species not report- any gross lesion was noticed during necropsy. Chi- ed; two via s.c. and one via i.m. route) and one square analysis was used to compare the frequency from a Ciconiiformes (Threskiornis spinicolis,i.m. of systemic reaction between the birds vaccinated route). Systemic reactions occurred in 17 birds: via i.m. and s.c. route. A P value of , 0.05 was two of the order Anseriformes (Cygnus atratus, considered significant. Branta sandvicensis); two of the order Ciconii- LE´ CU—INFLUENZA H5N2 VACCINE IN ZOO SPECIES 733

Table 1. Vaccination tolerance scores of different homologous antigen. Three birds were positive avian orders from a questionnaire representing 4,369 for H5 antibody: Eurasian griffon vulture (Gyps birds in zoologic institutions. The tolerance scores fulvus)(titer5 8), greater rhea (Rhea americana) included four levels: poor, intermediate, high, and very high. (titer 5 16), and black swan (Cygnus atratus) (titer 5 64). Tolerance score Titers were obtained after the initial vaccina- tion for 463 birds. The mean titer was 273. Forty- Avian order High Good Total five percent of the birds had a titer $32 and 57% Anseriforme 1,105 172 1,277 a titer $8 (Table 3) with discrepancies between Galliforme 652 151 803 avian orders (Fig. 1). One month after the Phoenicopteriforme 647 52 699 vaccine booster administration, titers were ob- Falconiforme 162 216 378 Psittaciforme 333 26 359 tained from 514 birds and the mean titers reached Ciconiiforme 188 115 303 558, with 71% of the birds with titers $32 and Struthioniforme 104 29 133 79% with titers $8. No differences were noticed Ciconiiforme 84 0 84 in the mean titers between birds injected either Columbiforme 59 3 62 i.m. or s.c. (P . 0.05), but some birds were Strigiforme 19 34 53 missing administration route data. Gruiforme 31 0 31 In four avian orders, .80% of birds had HI Sphe´nisciformes 24 0 24 titers $32: Phoenicopteriformes (n 5 40), Sphe- Pelecaniforme 9 10 19 nisciformes (n 5 16), Ciconiiformes (n 5 67), and Pelicaniforme 15 0 15 Anseriformes (n 5 147) (Fig. 2). In three avian Passe´riformes 4 0 4 Charadriiforme 2 0 2 orders, ,40% of birds had HI titers $32: Cuculiformes 1 0 1 Columbiformes (n 5 10), Strigiformes (n 5 14), Musophagiforme 0 1 1 and Struthioniformes (n 5 26). Five months after Divers 57 64 121 initial vaccination (day 180) and 10 months after Total 3,496 873 4,369 administration of the booster (day 330), 229 and 217 birds were sampled and the mean titers were 88 and 25, respectively (Table 3). formes (Threskiornis spinicolis, Platalea leucoro- Approximately 80% of birds in the order dia); two of the order Falconiformes (species not Ciconiiformes (n 5 28) had a HI titer $32 at reported); 10 of the order Galliformes (Acryllium day 180. Columbiformes and Charadriiformes vulturinum); and one of the order Psittaciformes were not sampled. Strigiformes, Struthioniformes, (Psittacus erithacus). Duration of time that birds Sphenisciformes, and Gruiformes had the lowest exhibited symptoms was 4.1 day. Signs reported titers. Among all results reported in Table 4, it were transient weakness, apathy, and anorexia. was possible to select only birds individually Some of the safety observations are provided followed (n 5 123) with a HI titer $32 from day with information on volume of vaccine injected 60–330 (Table 5). (Table 2). Volume of vaccine injected per body Two nestling species had positive maternal HI weight ratio (V/BW) of 121 observed birds was titers: Rhea americana (n 5 7) and Ara macao calculated and compared with the standard ratio (n 5 5). The titers were 0 for the four remaining based on poultry, in which 0.5 ml injected in a species: Nyctea scandiaca, Parabuteo unicinctus, 1.5 kg bird would result in a V/BW of 0.33. The Bubo bubo, and Falco tinnunculus (Fig. 3). smallest vaccinated birds (50 g) received 0.25 ml, which is approximately 15 times the V/BW ratio DISCUSSION used for poultry. Safety of vaccination is Data issued from the survey suggested that the qualified as ‘‘high’’ for these birds. No lesions overall safety of this vaccination protocol is thought to be associated with vaccination were good. The greatest risk of this protocol to the noted in vaccinated birds on postmortem exam- birds originates from the physical restraint, with ination. There was no significant differences approximately 0.61% of the mortality associated between safety results of birds injected i.m. or with stress and trauma induced by restraint.6 s.c. (chi-square 5 2.893; P . 0.05). There were few vaccine reactions observed in this study, despite a V/BW ratio 15 times greater Serology than the recommended V/BW ration for some Of 255 birds sampled during initial vaccina- smaller birds. Several of the veterinarians who tion, 252 were negative for H5 by HI using the participated in this study were advised not to 734 JOURNAL OF ZOO AND WILDLIFE MEDICINE

Table 2. Volumes of vaccine administered and veterinary safety considerations according to the birds’ body weights.

Maximum injected volume in ml and respective tolerance

Weight of bird (in kg) High Good Total of birds Volume/weight (V/W ratio) 0.05 0.25 13 5.00 0.083 0.125 4 3.01 0.1 0.25 0.25 25 2.50 0.11 0.125 16 2.27 0.12 0.25 1 2.08 0.138 0.125 3 1.81 0.15 0.25 4 1.67 0.155 0.125 2 1.61 0.2 1 0.25 30 1.25 0.25 0.25 15 1.00 0.5 1 0.25 249 0.50 1 1 1 203 0.25 1.5a 0.5 0.5 55 0.17 2 1 1 714 0.13 3 1 1 559 0.08 20 1 3 0.01 23 0.5 6 0.01 40 1 51 0.01 Maximum injected volume 1.5 1 Number of birds 2174 605 2779

a represents the regular dosage used in poultry. inject terrestrial birds in the pelvic limb muscles purposes of this study that a HI titer $32 should to avoid difficulty in locomotion. be considered protective. A titer $8 suggests that Presence of a protective immune response in the immune system has been triggered, but in humans and poultry is important to prevent chickens, it was not enough to be protective. influenza and subsequent disease.16 However, it should be emphasized that the Antibody production against the HA surface protective titers have not been validated by protein is the principal determinant for protec- challenge experiments for the different zoo tion against infection. Postvaccination titers $32 species. Weak serologic responses in some species in chickens prevented shedding after viral chal- or orders may not correspond to protection failure. lenge16 and was protective against an outbreak of Influenza viruses elicit both humoral and cellular HPAI H5N1.28 Thus, it is assumed for the immunity. Clinical study results from live influenza

Table 3. Number of birds sampled and distribution of mean hemagglutination inhibition titer on each sampling times.

Initial vaccination day 0 Booster vaccination day 30 Day 60 Day 180 Day 330 No. of birds sampled 255 463 514 229 217 Mean titer 1 273 558 88 25 No. of birds with titer .32 1 209 365 96 56 Birds with titer .32 (%) 0 45 715 42 26 Mean titer among those birds with titers .32 64 599 784 207 88 No. of birds with titer .8 3 263 406 107 77 Birds with titer .8(%) 1 57 79 47 35 Mean titer among those birds with titers .8 29 479 707 187 67 Table 4. Vaccinated species in zoos, number of sampled birds, percentage of birds f rom each species with a hemagglutination inhibition (HI) titer $32, and mean titers on day 30, 60, 180, and 330. Dash ‘‘–’’ defined as not performed.

HI titer when n 5 1 or mean titer No. of sampled birds at day Birds with titer $ 32 ( %) when n .1 Order Species 30 60 180 330 30 60 180 330 30 60 180 330 Falconiformes Cathartes aura 0 1 0 1 0 100 0 100 – 256 – 32 Gyps fulvus 55722040710 1223305 Aegypius monachus 54552050200 32737 2

Necrosyrtes monachus 11010000 1 1– 1 LE ´

Gypohierax angolensis 00010000– – – 1 SPECIES ZOO IN VACCINE H5N2 CU—INFLUENZA Vultur gryphus 2 2 2 2 50 50 0 0 65 129 1 1 Neophron percnopterus 3 5 5 5 0 60 40 40 1 263 154 39 Terathopius ecaudatus 1000100000128–– – Lophaetus occipitalis 010 0 1000 0 – 128– – Aquila heliaca 1 1 0 0 100 100 0 0 64 32 – – Parabuteo unicinctus 1 4 3 2 0 50 33 0 16 54 11 1 Polyborus plancus 4 4 0 0 100 100 0 0 112 384 – – Falco cherrug 1 0 4 4 0 0 75 50 1 – 64 25 Milvus milvus 530060670 0 6543– – Haliaeetus leucocephalus 00010000– – – 1 Sagittarius serpentarius 2 2 2 2 0 100 100 0 1 32 96 1 Total 3133282641635019431324914 Anseriformes Ancer indicus 440 1 751000 0 169688– 1 Branta ruficolis 8 9 0 0 38 78 0 0 45 242 – – Branta canadensis 7 7 0 0 71 71 0 0 62 361 – – Aix sponsa 4 3 0 0 50 100 0 0 81 715 – – Anas luzonica 11000100001128–– Anas platyrhynchos 770057710 0 106101– – Aix galericulata 4 4 0 0 50 50 0 0 41 148 – – Cygnus cygnus 25200000 1 11– Coscoroba coscoroba 2 2 2 1 100 100 50 100 544 320 33 64 Cygnus atratus 13 15 17 10 69 93 35 30 467 956 65 73 Cygnus olor 2 2 0 0 100 100 0 0 256 384 – – Dendrocygna automnalis 2200050001 33–– Dendrocygna bicolor 16 15 0 0 50 93 0 0 32 122 – – Dendrocygne viduata 2200050009 65–– Aythya collaris 1 1 0 0 100 100 0 0 256 256 – – Aythya ferina 330067670 0 192768– – Aythya fuligula 11000100001 64–– 735 736 Table 4. Continued.

HI titer when n 5 1 or mean titer No. of sampled birds at day Birds with titer $ 32 ( %) when n .1 Order Species 30 60 180 330 30 60 180 330 30 60 180 330 Aythya nyroca 330 0 01000 0 1 75– – Netta peposaca 16 15 7 0 88 80 57 0 152 502 64 – Netta rufina 7 7 4 0 86 100 50 0 704 704 100 – Anser anser 2 2 0 0 50 100 0 0 17 96 – – Cereops novaehollandiae 120 0 01000 0 8 96– – Chloephaga picta 1 1 0 0 0 100 0 0 16 32 – – Alopochen aegyptiacus 1 1 0 0 0 100 0 0 1 128 – – MEDICINE WILDLIFE AND ZOO OF JOURNAL Anas erythroryncha 1 1 1 0 100 100 100 0 128 256 32 – Anas bahamensis 650 0 671000 0 62205– – Anas acuta 5 5 2 0 0 80 0 0 5 106 1 – Calonetta leucophrys 7 6 0 0 71 100 0 0 188 304 – – Anas cyanoptera 2 1 0 0 100 100 0 0 1,280 128 – – Amazonetta braziliensis 220 0 501000 0 17256– – Marmaronetta angustirostris 2 0 0 0 50 0 0 33 – – – Anas discors 2 1 0 0 100 100 0 0 96 1,024 – – Anas versicolor 440 0 251000 0 65960– – Tadorna ferruginea 1 1 0 0 0 100 0 0 1 128 – – Tadorna tadorna 7 7 4 0 86 86 75 0 357 713 112 – Total 1491473912588644331784058266 Charadriiformes Recuvirostra avocetta 010 0 1000 0 – 32– – Haematopus ostralegus 2 2 0 0 100 100 0 0 80 320 – – Burhinus oedicnemus 45000000 5 10–– Vanellus armatus 1 2 0 0 100 100 0 0 64 160 – – Total 7 10 0 0 43 50 0 0 35 104 – – Ciconiiformes Ciconia ciconia 9 11 7 9 67 100 100 44 112 492 123 30 Threskiornis melanocephalus 8 6 11 8 88 100 100 75 3,008 4,779 431 106 Threskiornis spinicollis 10 10 10 10 100 100 90 70 3,866 4,941 160 91 Eudocimus ruber 10 14 0 0 40 86 0 0 28 166 – – Threskiornis aethiopicus 20 18 0 7 45 83 0 14 424 642 – 14 Leptoptilos crumeniferus 680 0 331000 0 25152– – Total 63 67 28 34 58 93 97 50 1,118 1,430 246 59 Columbiformes Ducula bicolor 330 0 0 330 0 6 86– – Geotrygon versicolor 110 0 02000 0 161 – – Columba flavirostris 33000000 6101–– Table 4. Continued.

HI titer when n 5 1 or mean titer No. of sampled birds at day Birds with titer $ 32 ( %) when n .1 Order Species 30 60 180 330 30 60 180 330 30 60 180 330 Caloenas nicobarica 11000000 161–– Columba arquatrix 22000000 1 5–– Total 10 10 0 0 0 30 0 0 7 57 – – Coraciiforme Bucorvus leadbeateri 11110000 1 81 1

Galliforme Gallus gallus 14 14 12 14 85 100 92 37 57 299 82 26 LE ´

Gallus lafayetti 2 2 0 0 100 100 0 0 64 128 – – SPECIES ZOO IN VACCINE H5N2 CU—INFLUENZA Gallus sonneratii 1 1 0 0 0 100 0 0 1 128 – – Polyplectron bicalcaratum 1 1 0 0 100 100 0 0 32 512 – – Lophura leucomelana 22000000 1 5–– Lophura nycthemera 440 0 0 250 0 2 9 – – Syrmaticus elliotii 1 1 0 0 100 0 0 0 32 1 – – Chrysolophus amherstiae 1 1 0 0 0 100 0 0 16 128 – – Syrmaticus mikado 11000000 1 8–– Lophura swinhoii 11000000 4 8–– Catreus wallichi 11000000 1 1–– Lophura hatinhenis 11000000 8 16–– Phasianus colchicus 670017000 6 1–– Crax rubra 2 2 0 0 0 50 0 0 1 132 – – Pavo cristatus 14 13 10 9 79 100 40 56 258 3,820 71 47 Tragopan satyre 1 1 0 0 0 100 0 0 16 32 – – Total 5353222353666848891,0437734 Gruiforme Anthropoı¨des virgo 7 11 5 5 0 82 0 0 9 55 4 2 Grus grus 3 3 5 0 67 67 60 0 219 97 70 – Balearica regulorum 15 15 9 11 0 33 0 9 3 21 5 4 Anthropoides paradisea 990 0 0 440 0 4 23– – Total 343819166 53166 2338223 Pelicaniforme Pelecanus onocrotalus 8 12 2 2 0 33 0 0 1 59 1 1 Pelecanus crispus 2 17 2 2 0 53 0 0 1 53 1 1 Total 10 29 4 4 0 45 0 0 1 56 1 1 Phoenicopteriforme Phoenicopterus ruber ruber 1 1 1 1 100 100 100 0 512 1,024 128 1 Phoenicopterus chiliensis 13 17 14 13 85 100 57 38 447 1,235 87 25 Phoeniconaias minor 10 12 0 19 100 100 0 37 550 704 – 22 Phoenicopterus ruber roseus 11 10 0 0 82 100 0 0 425 1,562 – – Total 35 40 15 33 89 100 60 36 471 1,152 90 22 737 738

Table 4. Continued.

HI titer when n 5 1 or mean titer No. of sampled birds at day Birds with titer $ 32 ( %) when n .1 Order Species 30 60 180 330 30 60 180 330 30 60 180 330 Psittaciforme Amazona amazonica 10 11 11 11 0 45 9 9 3 47 4 4 Amazona aestiva 1000100000 1– – – Amazona ocrocephala 1 1 1 1 0 100 0 0 32 512 32 32 MEDICINE WILDLIFE AND ZOO OF JOURNAL Amazona leucocephala 120 2 0 500 0 1665– 1 Amazona xantholora 12020000 1129– 1 Ara ararauna 6692 05044504 88463129 Ara chloroptera 2 2 4 2 100 100 100 0 144 512 48 5 Ara severa 222 0 0 500 0 1 179 – Cacatua sulphurea 11000000 1 1–– Cacatua leadbeateri 00000000– – – – Cacatua mollucensis 220 0 01000 0 1 48– – Cacatua alba 01000000– 1–– Total 27302720115337151410316517 Sphenisciforme Spheniscus humboldtii 1 1 0 0 0 100 0 0 16 32 – – Spheniscus demersus 71514130 930 0 6 2141 1 Total 8 16 14 13 0 94 0 0 8 203 1 1 Strigiforme Nyctea sandiaca 12220000 1 11 1 Strix alucoa 3 3 5 4 33 100 20 0 176 1,109 52 1 Bubo bubo 77810000101 11 7 Bubo virginianus 221 0 0 500 0 1 331 – Total 13 14 16 17 8 29 6 6 41 243 17 5 Struthioniforme Struthio camelus 182216146500 5 51 1 Rhea americana 4 4 0 6 75 75 0 17 560 2,184 – 7 Total 2226162018150 5 1063401 3 Grand total 463 514 229 218 45 71 42 26 273 558 88 24 LE´ CU—INFLUENZA H5N2 VACCINE IN ZOO SPECIES 739

Figure 1. Percentage of birds with HI titers $ 32 (grey bars) or $ 8 (black bars) in birds of different families sampled 30 days after the initial vaccination.

Figure 2. Percentage of birds with HI titers $ 32 (grey bars) or $ 8 (black bars) in birds of different families sampled 60 days after the initial vaccination. 740 JOURNAL OF ZOO AND WILDLIFE MEDICINE

Table 5. Percentage of birds with remaining institutions (98.4%).22 It is likely that some zoo hemagglutination inhibition titer $ 32 at day 60, 180, birds, especially waterfowl and other aquatic birds, and 330 only among birds that can be individually are seropositive as the result of previous infection followed from day 60–330. of LPAI viruses of the H5 subtype through direct 20 Day (%) or indirect contact with wild birds. HA titers after the initial vaccination (45% of Order No. 60 180 330 titers $32) and administration of the booster Ciconiiformes 11 100 100 73 vaccination (71% of titers $32) demonstrated Galliformes 13 100 85 38 that the administration of the booster vaccination Phoenicopteriformes 14 100 64 36 is necessary to increase mean titer to .30 or 40, Anseriformes 9 100 56 44 which is believed to be a protective titer, based on Sphenisciformes 12 92 0 0 previous research with poultry. However, because Falconiformes 5 60 40 0 Psittaciformes 15 53 27 20 the number of tested birds at day 30 and day 60 Gruiformes 16 50 0 6 was much higher than at day 0, the presence of Pelicaniformes 4 50 0 0 previously seropositive birds, which could in- Strigiformes 6 33 17 0 crease mean titer, could not be ruled out. Coraciiformes 1 0 0 0 The titers found here are not within the same Struthioniformes 14 0 0 0 range of the results obtained with the same vaccine in Dutch zoos in 2006,22 with HI titers between 500–700 at day 60 in the present study versus virus challenge in vaccinated birds were not between 200–300 for the Dutch study.22 In reported with most species housed in zoos, except addition, the number of avian orders monitored for a few Anseriformes and Galliformes.3,4,17,29,30 in this present study was more extensive. Addi- Moreover, as noticed in a Dutch trial,22 tionally, the protocol used on the largest birds was humoral response obtained with a derived HI different. Struthioniformes and large birds from test using H5 glycoprotein from the actual other orders received up to 2.5 ml in Dutch circulating H5N1 field strain may be different zoologic institutions, whereas the same species in (lower) than the response obtained through a test French zoologic institutions received a maximum using the vaccine antigen.25 of 1 ml. There may be a significant dose/weight The prevalence of anti-H5 influenza antibodies effect beyond 1.4 kg.23 One hundred percent of prior to vaccination was very low, because only rheas had a titer $32, whereas ostriches did not 98.8% of the birds were seronegative. This result is produce a detectable titer. Thus, a new vaccination similar to a previous 2006 study in Dutch zoologic protocol was suggested for this order of birds. A

Figure 3. HI titers in chicks of selected ages for rheas (Rhea americana, #) and Scarlet Macaw (Ara macao, ¤). LE´ CU—INFLUENZA H5N2 VACCINE IN ZOO SPECIES 741 vaccine volume between 2.5–5 ml was recom- that protection and shedding prevention were low mended for Struthioniformes. Early results at day 330. The highest numbers of birds with an showed that 100% of ostriches had protective HI titer $32 at day 330 occurred in orders that titers using these volumes (Intervet, unpub. data). had 100% positive birds at day 60. As noted in previous studies,1,22 other avian The main objective of vaccination is to protect orders had noticeably lower seroconversion rates, birds during the most critical period for an AI ,50% at day 60: Strigiformes (29%, n 5 14), outbreak (winter and spring in Europe) by Columbiformes (30%, n 5 10), and Pelecani- reducing viral shedding in vaccinated individu- formes (45%, n 5 29). The amount of vaccine for als.8,28 This goal may be best achieved if vaccina- these orders was lower than that of other studies, tion is initiated so that day 60 is within this critical which may explain the difference in level of season. According to these results, a booster at antibody production observed at day 60. Peleca- day 180 would be useful for many orders to niformes were an exception, in that titers are sustain a HI titer above a protective level over a reported to be low in all studies and do not longer period. This booster was recommended for appear to be dose dependent.1,19,22 For the 2007 Struthioniformes, Strigiformes, and Falconi- campaign, some recommended vaccine dosages formes in 2007. This second booster would be were increased based on the weight of the bird helpful for all orders of birds, but this implies that (e.g., ratites). However, V/BW discrepancy is not birds are restrained twice a year, which would lead the only factor to affect interspecific serologic to a deleterious effect on breeding season. differences. For instance, in birds ,1.4 kg, it Rhea chicks hatched from vaccinated parents maybe noticed that Columbiformes show very had decreasing titers up to 4 wk of age (Fig. 3). poor titers compared with Ciconiiformes. For This is likely to suppress or decrease vaccination example, at day 30, the mean titer for Columbi- response within these first weeks of life.18 Hence, formes was 5.4 (n 5 10), whereas the mean titer a vaccination protocol for chicks would include for Ciconiiformes was 1,845 (n 5 49). In addition, two booster injections, one at 3–4 wk of age and the V/BW ratio for Ciconiiformes is much lower. one at 9–10 wk of age, to avoid maternal Therefore, other species-specific reactions could antibody interference. interfere with serologic response. Route of injection does not appear to affect CONCLUSION seroconversion based on the results of this study. In this study, the use of the H5N2 inactivated However, the route was not known for all birds. vaccine in French zoo birds was satisfactory in In addition, other parameters, such as the age of terms of safety and often induced adequate serolo- birds at vaccination, can have an effect on gic responses when a booster injection was pro- 15 serology results, but such information was not vided. Avian orders react differently to the vaccine, recorded in this study. regarding their relative dosage/body weight ratio At 180 days after the initial vaccination, .50% or specific immune reactions. Antibody levels of the birds had a titer $32, with differences decrease over time, with a low proportion of birds between orders. Ciconiiformes had the highest still seropositive at day 330. A booster every 6 mo number of seropositive birds for this period. For is likely to enhance long-term titers in some species. practical reasons, including lack of biosecurity, Vaccination should be considered as an ancillary outdoor housing in pools, and ponds with tool to other biosecurity measures to reduce exposure to wild birds, the potential for infection shedding in outbreak situations. Yearly booster cannot be ruled out. Although no birds in the vaccination should be discussed regarding both order Ciconiiformes were reported as seropositive local and global HPAI updated epidemiologic at day 0, in the individually monitored group profiles.21 Passive prophylaxis, including with- (Table 5), six of 11 orders did not have protective drawing food contact with feral birds and hygienic titers remaining at day 180, which is consistent measures, should be also promoted. with a previous study in Singapore.19 In all orders, HI titers did not increase between day Acknowledgments: We thank Intervet for pro- 60–180, with the exception of three birds. viding the vaccine to the zoos and for technical At day 330, none of the orders had a high support. We are also grateful to the veterinarians percentage of seropositive birds. Only 26% of the from the Association Francophone des Ve´te´ri- birds still had an antibody titer within the naires de Parc Zoologique who performed the protective range. Because it was already reported hard work of collecting, archiving, and sending in chickens at day 224,5 it can be assumed here blood samples. 742 JOURNAL OF ZOO AND WILDLIFE MEDICINE

LITERATURE CITED 2006. Avian influenza A virus subtype H5N2 in a red lored Amazon. J. Am. Vet. Assoc. 228: 236–241. 1. Bertelsen, M. F., J. Klausen, E. Holm, C. 13. Humberd, J., K. Boyd, and R. G. Webster. 2007. Grø´ndahl, and P. H. Jø´rgensen. 2007. Serological Emergence of influenza A virus variants after prolonged response to vaccination against avian influenza in zoo shedding from pheasants. J. Virol. 81: 4044–4051. birds using an inactivated H5N9 vaccine. Vaccine 25: 14. Kodihalli, S., V. Sivanandan, K. V. Nagaraja, 4245–4349. D. Shaw, and D. A. Halvorson. 1994. A type-specific 2. Brown, I. H., B. Z. Londt, W. Shell, R. J. avian influenza virus for turkeys: Manvell, J. Banks, R. Gardner, I. Outtrim, S. C. Essen, induction of protective immunity to challenge infection. M. Sabirovic, M. Slomka, and D. J. Alexander. 2006. Vaccine 15: 1467–1472. First incursion of H5N1 highly pathogenic avian 15. Lavoie, E. T., E. M. Sorrell, D. R. Perez, and M. influenza viruses of the Asian lineage into Europe. A Ottinger. 2007. Immunosenescence and age-related Proc. 6th Int. Symp. Avian Influenza, Abstract, 2006: susceptibility to influenza virus in Japanese quail. Dev. 37. St John College, Cambridge, United Kingdom. Comp. Immunol. 31: 407–414. Brown, I., and I. Capu (eds). Veterinary Laboraory 16. Lee, C. W., D. A. Senne, and D. L. Suarez. 2004. Agency, Weybridge, United Kingdom. Effect of vaccine use in the evolution of Mexican lineage 3. Bublot, M., F. X. Le Gros, D. Nieddu, N. H5N2 avian influenza virus. J. Virol. 78: 8372–8381. Pritchard, T. R. Mickle, and D. E. Swayne. 2007. 17. Liert, M., M. Hafez, R. Klopfeisch, D. Lu¨schow, Efficacy of two H5N9-inactivated against C. Prusas, J. P. Teifke, M. Rudolf, C. Grund, D. challenge with a recent H5N1 highly pathogenic avian Kalthoff, T. Mettenleiter, M. Beer, and T. Harder. influenza isolate from a chicken in Thailand. Avian Dis. 2007. Protection and virus shedding of faclons vacci- 51: 332–337. nated against highly pathogenic avian influenza A virus 4. Capua, I., C. Terregino, G. Cattoli, F. Mutinelli, (H5N1). Emerg. Infect. Dis. 13: 1667–1674. and J. F. Rodriguez. 2003. Development of a DIVA 18. LiJun, J., Y. M. Zhang, P. DaXin, H. Q. Liu, J. (differentiating infected from vaccinated animals) strat- Cheng, Z. RuKuan, and X. F. Liu. 2004. Influence of egy using a vaccine containing a heterologous neur- dosage and maternal antibody on responses of chickens aminidase for the control of avian influenza. Avian to recombinant fowl pox virus vaccine against H5 Pathol. 32: 47–55. subtype avian influenza. Chin. J. Vet. Sci. 24: 150–152. 5. Cardona, C. J., B. R. Charlton, and P. R. 19. Oh, S., P. Martelli, O. S. Hock, S. Luz, C. Woolcock. 2006. Persistence of immunity in commercial Furley, E. J. Chiek, L. C. Wee, and N. M. Keun. 2002. egg-laying hens following vaccination with a killed Field study on the use of inactivated H5N2 vaccine in H6N2 avian . Avian Dis. 50: 374–379. avian species. Vet. Rec. 157: 299–300. 6. De Jong, J. C., A. M. Palache, W. E. Beyer, G. F. 20. Oliveira, Jr., J. G., M. S. P. Belluci, J. S. M. Rimmelzwaan, A. C. Boon, and A. D. Osterhaus. 2003. Vianna, C. Mazur, C. M. Andrade, L. P. L. Fedullo, C. Haemagglutination-inhibiting antibody to influenza Portz, and B. O. Loureiro. 2001. Serological survey on virus. Dev. Biol. 115: 63–73. influenza virus in domestic and wild birds from Rio de 7. Dudley, J. P. 2006. Bird flu outbreak in United Janeiro State, Brazil. Arq. Bras. Med. Vet. Zootec. 53: Kingdom reveals global vulnerabilities. Bioscience 56: 299–302. 182–183. 21. Peyre, M., G. Fusheng, S. Desvaux, and F. 8. Ellis, T. M., C. Y. Leung, and M. K. Chow. 2004. Roger. 2009. Avian influenza vaccines: a practical review Vaccination of chicken against H5N1 avian influenza in in relation to their application in the field with a focus on the face of an outbreak interrupts virus transmission. the Asian experience. Epidemiol. Infect. 137: 1–21. Avian Pathol. 33: 405–412. 22. Philippa, J. D. W., C, Baas, W. Beyer, T. 9. Fouchier, R. A. M., J. V. Munster, J. Keawchar- Bestebroer, R. Fouchier, D. Smith, W. Schaftenaar, oen, A. D. M. Osterhaus, and T. Kuiken. 2007. and A. Osterhaus. 2007. Vaccination against highly Virology of avian influenza in relation to wild birds. pathogenic avian influenza H5N1 virus in zoos using an J. Wild. Dis. 43(Suppl. 3): 7–14. adjuvanted inactivated H5N2 vaccine. Vaccine 25: 10. Gaidet, N., T. Dodman, A. Caron, G. Balanc¸a, 3800–3808. S. Desvaux, F. Goutard, G. Cattoli, V. Martin, A. 23. Philippa, J. D. W., V. J. Munster, H. Van Tripodi, F. Lamarque, W. Hagemeijer, and F. Monicat. Bolhuis, T. M. Bestebroer, W. Schaftenaar, W. E. P. 2007. Influenza surveillance in wild birds in Eastern Beyer, R. A. M. Fouchier, T. Kuiken, and A. Osterhaus. Europe, the Middle East and Africa: preliminary results 2005. Highly pathogenic avian influenza (H7N7): from an ongoing FAO-led survey. J. Wild. Dis. vaccination of zoo birds and transmission to non-poultry 43(Suppl. 3): 22–28. species. Vaccine 23: 5743–5750. 11. Gerlach, H. 1994. Viruses/avian influenza A. In: 24. Redrobe, S. P. 2007. Avian influenza H5N1: a Ritchie, B. W., G. J. Harrison, and L. R. Harrison review of the current situation and relevance to zoos. (eds.). Avian Medicine. Wingers Publishing Inc., Lake Int. Zoo. Yb. 41: 96–109. Worth, Florida. Pp. 929–931. 25. Schmitz, A., C. Guillemoto, I. Pierre, J. La- 12. Hawkins, M. G., B. M. Crossly, A. Osofsky, R. mande´, C. Alle´e, D. Gomis, A. Le´cu, Y. Roman, E. J. Webby, C. W. Lee, D. L. Suarez, and S. K. Hietala. Bureau, T. Petit, J. Francart, and V. Jestin. 2007. LE´ CU—INFLUENZA H5N2 VACCINE IN ZOO SPECIES 743

Vaccination pre´ventive contre l’influenza aviaire des inactivated vaccine developed by reverse genetics. oiseaux captifs des zoos franc¸ais. Proceeding of 7e`me Virology 34: 153–162. Journe´e de la Recherche Avicole, Tours, France. 28–29 30. Toro, H., D. C. Tang, D. L. Suarez, M. J. Sylte, Mars 2007. Pp. 334–338. J. Pfeiffer, and K. R. Van Kampe. 2007. Protective 26. Slemons, R. D., R. S. Cooper, and J. S. Orsborn. avian influenza in ovo vaccination with non-replicating 1973. Isolation of type-A influenza viruses from human adenovirus vector. Vaccine. 25: 2886–2891. imported exotic birds. Avian Dis. 17: 746–751. 31. Van Borm, S., I. Thomas, G. Hanquet, B. 27. Song, H., G. R. Nieto, and D. R. Perez. 2007. A Lambrecht, M. Boschlans, G. Dupont, M. Decaes- new generation of modified live-attenuated avian tecker, R. Snacken, and T. Van den Berg. 2005. Highly influenza viruses using a two-strategy combination as pathogenic H5N1 influenza virus in smuggled Thai potential vaccine candidates. J. Virol. 81: 9238–9248. eagles, Belgium. Emerg. Infect. Dis. 11: 702–705. 28. Swayne, D. E., C. W. Lee, and E. Spackman. 32. Van der Goot. J. A., M. van Boven, M. C. M de 2006. Inactivated North American and European Jong, and G. Koch. 2007. Effect of vaccination on H5N2 avian influenza virus vaccines protect chickens transmission of HPAI H5N1: the effect of a single from Asian H5N1 high pathogenicity avian influenza vaccination dose on transmission of highly pathogenic virus. Avian Pathol. 35: 141–146. avian influenza H5N1 in Peking ducks. Avian Dis. 51: 29. Tian, G., S. Zhang, Y. Li, Z. Bu, P. Liu, J. Zhou, 323–324. C. Li, J. Shi, K. Yu, and H. Chen. 2005. Protective efficacy in chicken, geese and ducks of an H5N1- Received for publication 4 March 2008